• Author / Uploaded
• prasad616

• Categories
• Apoplejía
• Medicina basada en la evidencia
• Imagen de resonancia magnética
• Libros
• Sistema circulatorio

Citation preview

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page i

Stroke Practical Management THIRD EDITION

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page ii

Dedication To our many colleagues, young and old, with whom we have shared the care of so many stroke patients, and with whom we have discussed so many interesting ideas.

Acknowledgement The authors of this book are particularly grateful to Joanna Warldlaw, who has drawn much of the line artwork throughout the three editions.

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page iii

Stroke

Practical Management THIRD EDITION

C. Warlow J. van Gijn M. Dennis J. Wardlaw J. Bamford G. Hankey P. Sandercock G. Rinkel P. Langhorne C. Sudlow P. Rothwell

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page iv

© 2007 C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow, P. Rothwell Published by Blackwell Publishing Blackwell Publishing, Inc., 350 Main Street, Malden, Massachusetts 02148-5020, USA Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK Blackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. First edition published 1996 Second edition published 2001 Third edition published 2008 1

2008

Library of Congress Cataloging-in-Publication Data Stroke : practical management / C. Warlow . . . [et al.]. – 3rd ed. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4051-2766-0 (hardcover : alk. paper) 1. Cerebrovascular disease. 2. Cerebrovascular disease – Treatment. I. Warlow, Charles, 1934[DNLM: 1. Cerebrovascular Accident – therapy. 2. Intracranial Hemorrhages – therapy. 3. Ischemic Attack, Transient – therapy. WL 355 S9208 2007] RC388.5.S847 2007 616.8’1 – dc22 2007022955 ISBN: 978-1-4051-2766-0 A catalogue record for this title is available from the British Library Set in 9/12pt Stone Serif by Graphicraft Limited, Hong Kong Printed and bound in Singapore by Fabulous Printers Pte Ltd Commissioning Editor: Martin Sugden Editorial Assistant: Jennifer Seward Development Editor: Lauren Brindley Production Controller: Debbie Wyer For further information on Blackwell Publishing, visit our website: http://www.blackwellpublishing.com The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting a specific method, diagnosis, or treatment by physicians for any particular patient. The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. Readers should consult with a specialist where appropriate. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom.

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page v

Contents

Contributors, vi Acknowledgements, vii Abbreviations, viii 1 Introduction, 1 2 Development of knowledge about cerebrovascular disease, 7 3 Is it a vascular event and where is the lesion?, 35 4 Which arterial territory is involved?, 131 5 What pathological type of stroke is it, cerebral ischaemic or haemorrhage?, 181 6 What caused this transient or persisting ischaemic event?, 259 7 Unusual causes of ischaemic stroke and transient ischaemic attack, 353 8 What caused this intracerebral haemorrhage?, 411 9 What caused this subarachnoid haemorrhage?, 457 10 A practical approach to the management of stroke and transient ischaemic attack patients, 503 11 What are this person’s problems? A problem-based approach to the general management of stroke, 537 12 Specific treatments for acute ischaemic stroke, 635 13 Specific treatment of intracerebral haemorrhage, 703 14 Specific treatment of aneurismal subarachnoid haemorrhage, 719 15 Specific interventions to prevent intracranial haemorrhage, 767 16 Preventing recurrent stroke and other serious vascular events, 789 17 The organization of stroke services, 903 18 Reducing the impact of stroke and improving the public health, 953 Index, 980 Colour plates are found facing p.550

v

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page vi

Contributors

Charles Warlow University of Edinburgh, Western General Hospital, Edinburgh, UK

Jan van Gijn Utrecht University, Utrecht, the Netherlands

Martin Dennis University of Edinburgh, Western General Hospital, Edinburgh, UK

Joanna Wardlaw University of Edinburgh, Western General Hospital, Edinburgh, UK

John Bamford St James’ University Hospital, Leeds, West Yorkshire, UK

Graeme Hankey Royal Perth Hospital, Stroke Unit, Perth WA, Australia

Peter Sandercock University of Edinburgh, Western General Hospital, Edinburgh, UK

Gabriel Rinkel Utrecht University, Utrecht, the Netherlands

Peter Langhorne Academic Section of Geriatric Medicine, Royal Infirmary, Glasgow, UK

Cathie Sudlow University of Edinburgh, Western General Hospital, Edinburgh, UK

Peter Rothwell Department of Clinical Neurology, Radcliffe Infirmary, Oxford, UK

vi

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page vii

Acknowledgements

We have had invaluable help and advice from many people in the preparation of this third edition. So thank you all, including:

Also, thank you to our teachers and colleagues from whom we have learned so many worthwhile things over the years:

Sheena Borthwick Judi Clarke Carl Counsell Ann Deary Alice Emmott Hazel Fraser Paut Greebe Gord Gubitz Ingrid Kane Sarah Keir Alistair Lammie Lynn Legg Richard Lindley Mike McDowell Michael Mackie Ian Marshall Nick Morgan Ross Naylor Sarah Pendlebury David Perry Rustam Al-Shahi Salman Cameron Sellers Mark Smith Ian Starkey Stuart Taylor Brenda Thomas Theo vanVroonhoven Nic Weir

Henry Barnett Lou Caplan David Chadwick Iain Chalmers Rory Collins Hans van Crevel Richard Doll Geoff Donnan Stuart Douglas Shah Ebrahim Rob Edis Barbara Farrell C. Miller Fisher Chris Foote John Fry Mike Gent Sonny Gubbay Michael Harrison Jim Heron Steff Lewis Bryan Matthews Richard Peto Alex Pollock Geoffrey Rose David Sackett Robin Sellar David Shepherd Jim Slattery Rien Vermeulen Ted Stewart-Wynne Derick Wade Eelco Wijdicks

vii

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page viii

Abbreviations

We don’t care much for abbreviations. They are not literate (Oliver Twist was not abbreviated to OT each time Dickens mentioned his name!), they don’t look good on the printed page, and they make things more difficult to read and understand, particularly for non-experts. But they do save space and so we have to use them a bit. However, we will avoid them as far as we can in tables, figures and the practice points. We will try to define any abbreviations the first time they are used in each chapter, or even in each section if they are not very familiar. But, if we fail to be comprehensible, then here is a rather long list to refer to. ACA ACE AChA ACoA ACST ADC ADH ADL ADP ADPKD AF AFx AH AICA AIDS AMI ANCA ANF APS APT APTT ARAS ARD ASA ASD ATIII

Anterior cerebral artery Angiotensin converting enzyme Anterior choroidal artery Anterior communicating artery Asymptomatic Carotid Surgery Trial Apparent diffusion coefficient Antidiuretic hormone Activities of daily living Adenosine diphosphate Autosomal dominant polycystic kidney disease Atrial fibrillation Amaurosis fugax Ataxic hemiparesis Anterior inferior cerebellar artery Acquired immune deficiency syndrome Acute myocardial infarction Antineutrophil cytoplasmic antibody Antinuclear factor Antiphospholipid syndrome Antiplatelet Trialists’ Collaboration Activated partial thromboplastin time Ascending reticular activating system Absolute risk difference Atrial septal aneurysm Atrial septal defect Antithrombin III

ATP ATT AVF AVM BA BIH BMI BP C CAA CADASIL

Adenosine triphosphate Antithrombotic Trialists’ Collaboration Arteriovenous fistula Arteriovenous malformation Basilar artery Benign intracranial hypertension Body mass index Blood pressure Celsius Cerebral amyloid angiopathy Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy CAST Chinese Acute Stroke Trial CAVATAS Carotid and Vertebral Artery Transluminal Angioplasty Study CBF Cerebral blood flow CBFV Cerebral blood flow velocity CBV Cerebral blood volume CCA Common carotid artery CEA Carotid endarterectomy CHD Coronary heart disease CI Confidence interval CK Creatine kinase CMRO2 Cerebral metabolic rate of oxygen CMRglu Cerebral metabolic rate of glucose CNS Central nervous system CPP Cerebral perfusion pressure CPSP Central post-stroke pain CSF Cerebrospinal fluid CT Computed tomography CTA Computed tomographic angiography CVR Cerebrovascular resistance DBP Diastolic blood pressure DCHS Dysarthria clumsy-hand syndrome DIC Disseminated intravascular coagulation DNA Deoxyribose nucleic acid DSA Digital subtraction angiography DSM Diagnostic and statistical manual of mental disorders

viii

..

..

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page ix

Abbreviations

DVT DWI EACA EADL EAFT ECA ECASS ECG EC-IC ECST EEG EMG ESR FDA FIM FLAIR FMD fMRI FMZ GCS GEF GKI HACP Hg HITS HIV HMPAO HTI HU IAA IAA ICA ICH ICIDH ICP ICVT IADSA INR IST IVDSA IVM kPa L LACI LACS LGN LP LSA M MAC

..

..

Deep venous thrombosis (in the legs or pelvis) Diffusion weighted (MR) imaging Epsilon-aminocaproic acid Extended activities of daily living European Atrial Fibrillation Trial External carotid artery European Cooperative Acute Stroke Study Electrocardiogram Extracranial-intracranial European Carotid Surgery Trial Electroencephalogram Electromyography Erythrocyte sedimentation rate Food and Drug Administration Functional independence measure Fluid attenuated inversion recovery Fibromuscular dysplasia Functional magnetic resonance imaging Flumazenil Glasgow Coma Scale Glucose extraction fraction Glucose, potassium and insulin Homolateral ataxia and crural paresis Mercury High intensity transient signals Human immunodeficiency virus Hexamethylpropyleneamine oxime Haemorrhagic transformation of an infarct Hounsfield units Internal auditory artery Intra arterial angiography Internal carotid artery Intracerebral haemorrhage International classification of impairments, disabilities and handicaps Intracranial pressure Intracranial venous thrombosis Intra-arterial digital subtraction angiography International normalized ratio International Stroke Trial Intravenous digital subtraction angiography Intracranial vascular malformation Kilopascals Litre Lacunar infarction Lacunar syndrome Lateral geniculate nucleus Lumbar puncture Lenticulostriate artery Molar Mitral annulus calcification

MAOI MAST-I MCA MCTT MES MI MLF MLP MMSE MR MRA MRC MRI MRS MRV MTT NAA NASCET NELH NG NIHSS NINDS NNT NO OCSP OEF OHS OR PACI PaCO2 PaO2 PACS PCA PChA PCoA PCV PE PEG PET PFO PICA PMS PNH POCI POCS PD PSE PSS PT PTA PVD

Monoamine oxidase inhibitor Multicentre Acute Stroke Trial – Italy Middle cerebral artery Mean cerebral transit time Microembolic signals Myocardial infarction Medial longitudinal fasciculus Mitral leaflet prolapse Mini mental state examination Magnetic resonance Magnetic resonance angiography Medical Research Council Magnetic resonance imaging Magnetic resonance spectroscopy Magnetic resonance venogram Mean transit time N-acetylaspartate North American Symptomatic Carotid Endarterectomy Trial National Electronic Library for Health Nasogastric National Institute of Health Stroke Score National Institute of Neurological Disorders and Stroke Number-needed-to-treat Nitric oxide Oxfordshire Community Stroke Project Oxygen extraction fraction Oxford Handicap Scale Odds ratio Partial anterior circulation infarction Arterial partial pressure of carbon dioxide Arterial partial pressure of oxygen Partial anterior circulation syndrome Posterior cerebral artery Posterior choroidal artery Posterior communicating artery Packed cell volume Pulmonary embolism Percutaneous endoscopic gastrostomy Positron emission tomography Patent foramen ovale Posterior inferior cerebellar artery Pure motor stroke Paroxysmal nocturnal haemoglobinuria Posterior circulation infarction Posterior circulation syndrome Proton density Present state examination Pure sensory stroke Prothrombin time Percutaneous transluminal angioplasty Peripheral vascular disease

ix

9781405127660_1_pre.qxd 10/13/07 11:15 AM Page x

x

Abbreviations

PWI QALYs RAH RCT RIND RNA ROR RR RRR rt-PA SADS SAH SBP SCA SD SEPIVAC SF36 SIADH SK

Perfusion weighted (MR) imaging Quality adjusted life years Recurrent artery of Heubner Randomized controlled trial Reversible ischaemic neurological deficit Ribonucleic acid Relative odds reduction Relative risk Relative risk reduction Recombinant tissue plasminogen activator Schedule for affective disorders and schizophrenia Subarachnoid haemorrhage Systolic blood pressure Superior cerebellar artery Standard deviation Studio epidemiologico sulla incidenza delle vasculopathie acute cerebrali Short form 36 Syndrome of inappropriate secretion of antidiuretic hormone Streptokinase

SLE SMS SPAF SPECT SVD TACI TACS TCD TEA TENS TGA TIA TMB TOAST TTP TTP US VA VB WHO WFNS

Systemic lupus erythematosus Sensorimotor stroke Stroke prevention in atrial fibrillation (trial) Single photon emission computed tomography Small vessel disease Total anterior circulation infarction Total anterior circulation syndrome Transcranial Doppler Tranexamic acid Transcutaneous electrical nerve stimulation Transient global amnesia Transient ischaemic attack Transient monocular blindness Trial of ORG 10172 in Acute Stroke Therapy Thrombotic thrombocytopenic purpura Time to peak Ultrasound Vertebral artery Vertebrobasilar World Health Organization World Federation of Neurological Surgeons

..

9781405127660_4_001.qxd 10/13/07 11:15 AM Page 1

1

Introduction 1.1 Introduction to the first edition

1

1.2 Introduction to the second edition 1.3 Introduction to the third edition

1.1 Introduction to the first edition

1.1.1 Aims and scope of the book We, the authors of this book, regard ourselves as practising – and practical – doctors who look after stroke patients in very routine day-to-day practice. The book is for people like us: neurologists, geriatricians, stroke physicians, radiologists and general internal physicians. But it is not just for doctors. It is also for nurses, therapists, managers and anyone else who wants practical guidance about all and any of the problems to do with stroke – from aetiology to organization of services, from prevention to occupational therapy, and from any facet of cure to any facet of care. In other words, it is for anyone who has to deal with stroke in clinical practice. It is not a book for armchair theoreticians, who usually have no sense of proportion as well as difficulty in seeing the wood from the trees. Or, maybe, it is particularly for them so that they can be led back into the real world. The book takes what is known as a problem-orientated approach. The problems posed by stroke patients are discussed in the sort of order that they are likely to present themselves. Is it a stroke? What sort of stroke is it? What caused it? What can be done about it? How can the patient and carer be supported in the short term and long term? How can any recurrence be prevented? How can stroke services be better organized? Unlike traditional textbooks, which linger on dusty shelves, there are no ‘-ology’ chapters. Aetiology, epidemiology, pathology

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

3

4

and the rest represent just the tools to solve the problems – so they are used when they are needed, and not discussed in isolation. For example, to prevent strokes one needs to know how frequent they are (epidemiology), what types of stroke there are (pathology), what causes them (aetiology) and what evidence there is to support therapeutic intervention (randomized controlled trials). Clinicians mostly operate on a need-to-know basis, and so when a problem arises they need the information to solve it at that moment, from inside their head, from a colleague – and we hope from a book like this.

1.1.2 General principles To solve a problem one obviously needs relevant information. Clinicians, and others, should not be making decisions based on whim, dogma or the last case, although most do, at least some of the time – ourselves included. It is better to search out the reliable information based on some reasonable criterion for what is meant by reliable, get it into a sensible order, review it and make a summary that can be used at the bedside. If one does not have the time to do this – and who does for every problem? – then one has to search out someone else’s systematic review. Or find the answer in this book. Good clinicians have always done all this intuitively, although recently the process has been blessed with the title of ‘evidencebased medicine’, and now even ‘evidence-based patientfocused medicine’! In this book we have used the evidence-based approach, at least where it is possible to do so. Therefore, where a systematic review of a risk factor or a treatment is available we have cited it, and not just emphasized single studies done by us or our friends and with results to suit our prejudices. But so often there is no good evidence or even any evidence at all available, and certainly no systematic reviews. What to do then? Certainly not what most doctors are trained to do: ‘Never be wrong, and if you are, never admit it!’ If we do 1

..

9781405127660_4_001.qxd 10/13/07 11:15 AM Page 2

2

Chapter 1 Introduction

not know something, we will say so. But, like other clinicians, we may have to make decisions even when we do not know what to do, and when nobody else does either. One cannot always adopt the policy of ‘if you don’t know what to do, don’t do it’. Throughout the book we will try to indicate where there is no evidence, or feeble evidence, and describe what we do and will continue to do until better evidence becomes available; after all, it is these murky areas of practice that need to be flagged up as requiring further research. Moreover, in clinical practice, all of us ask respected colleagues for advice, not because they may know something that we do not but because we want to know what they would do in a difficult situation.

1.1.3 Methods We were all taught to look at the ‘methods’ section of a scientific paper before anything else. If the methods are no good, then there is no point in wasting time and reading further. In passing, we do regard it as most peculiar that some medical journals still print the methods section in smaller letters than the rest of the paper. Therefore, before anyone reads further, perhaps we should describe the methods we have adopted. It is now impossible for any single person to write a comprehensive book about stroke that has the feel of having been written by someone with hands-on experience of the whole subject. The range of problems is far too wide. Therefore, the sort of stroke book that we as practitioners want – and we hope others do too – has to be written by a group of people. Rather than putting together a huge multiauthor book, we thought it would be better and more informative, for ourselves as well as readers, to write a book together that would take a particular approach (evidence-based, if you will) and end up with a coherent message. After all, we have all worked together over many years, our views on stroke are more convergent than divergent, and so it should not be too terribly difficult to write a book together. Like many things in medicine, and in life, this book started over a few drinks to provide the initial momentum to get going, on the occasion of a stroke conference in Geneva in 1993. At that time, we decided that the book was to be comprehensive (but not to the extent of citing every known reference), that all areas of stroke must be covered, and who was going to start writing which section. A few months later, the first drafts were then commented on in writing and in detail by all the authors before we got back together for a general discussion – again over a few drinks, but on this occasion at the Stockholm stroke conference in 1994. Momentum restored, we went home to improve what we had written,

and the second draft was sent round to everyone for comments in an attempt to improve the clarity, remove duplication, fill in gaps and expunge as much remaining neurodogma, neurofantasy and neuroastrology as possible. Our final discussion was held at the Bordeaux stroke meeting in 1995, and the drinks that time were more in relief and celebration that the end was in sight. Home we all went to update the manuscript and make final improvements before handing over the whole lot to the publisher in January 1996. This process may well have taken longer than a conventional multiauthor book in which all the sections are written in isolation. But it was surely more fun, and hopefully the result will provide a uniform and coherent view of the subject. It is, we hope, a ‘how to do it’ book, or at least a ‘how we do it’ book.

1.1.4 Using the book This is not a stroke encyclopaedia. Many very much more comprehensive books and monographs are available now, or soon will be. Nor is this really a book to be read from cover to cover. Rather, it is a book that we would like to be used on stroke units and in clinics to help illuminate stroke management at various different stages, both at the level of the individual patient and for patients in general. So we would like it to be kept handy and referred to when a problem crops up: how should swallowing difficulties be identified and managed? Should an angiogram be done? Is raised plasma fibrinogen a cause of stroke? How many beds should a stroke unit have? And so on. If a question is not addressed at all, then we would like to know about it so that it can be dealt with in the next edition, if there is to be one, which will clearly depend on sales, the publisher, and enough congenial European stroke conferences to keep us going. It should be fairly easy to find one’s way around the book from the chapter headings and the contents list at the beginning of each chapter. If that fails, then the index will do instead. We have used a lot of crossreferencing to guide the reader from any starting point and so avoid constant reference to the index. As mentioned earlier, we have tried to be as selective as possible with the referencing. On the one hand, we want to allow readers access to the relevant literature, but on the other hand we do not want the text to be overwhelmed by references – particularly by references to unsound work. To be selective, we have tried to cite recent evidence-based systematic reviews and classic papers describing important work. Other references can probably mostly be found by those who want to dig deeper in the reference lists of the references we have cited.

..

..

9781405127660_4_001.qxd 10/13/07 11:15 AM Page 3

1.2 Introduction to the second edition

Finally, we have liberally scattered what some would call practice points and other maxims throughout the book. These we are all prepared to sign up to, at least in early 1996. Of course, as more evidence becomes available, some of these practice points will become out of date.

1.1.5 Why a stroke book now? Stroke has been somewhat of a Cinderella area of medicine, at least with respect to the other two of the three most common fatal disorders in the developed world – coronary heart disease and cancer. But times are gradually changing, particularly in the last decade when stroke has been moving up the political agenda, when research has been expanding perhaps in the slipstream of coronary heart disease research, when treatments to prevent, if not treat, stroke have become available and when the pharmaceutical industry has taken more notice. It seems that there is so much information about stroke that many practitioners are beginning to be overwhelmed. Therefore, now is a good time to try to capture all this information, digest it and then write down a practical approach to stroke management based on the best available evidence and research. This is our excuse for putting together what we know and what we do not know, what we do and why we do it.

1.2 Introduction to the second edition

Whether we enjoyed our annual ‘stroke book’ dinners at the European stroke conferences too much to abandon them, or whether we thought there really was a lot of updating to do, we found ourselves working on this second edition four short years after the first. It has certainly helped to have been so much encouraged by the many people who seemed to like the book, and find it useful. We have kept to the same format, authors, and principles outlined above in the introduction to the first edition. The first step was for all of us to read the whole book again and collect together any new comments and criticisms for each of the other authors. We then rewrote our respective sections and circulated them to all the other authors for their further comments (and they were not shy in giving them). We prepared our final words in early 2000. A huge technical advance since writing the first edition has been the widespread availability of e-mail and the use of the Internet. Even more than before, we have

..

..

genuinely been able to write material together; one author does a first draft, sends it as an attachment across the world in seconds, the other author appends ideas and e-mails the whole attachment back to the first author, copying to other authors for comments perhaps, and so on until it is perfect. Of course, we still do not all agree about absolutely everything all of the time. After all, we want readers to have a feel for the rough and ragged growing edge of stroke research, where there is bound to be disagreement. If we all knew what to do for stroke patients there would be no need for randomized controlled trials to help us do better – an unrealistic scenario if ever there was one. So where there is uncertainty, and where we disagree, we have tried to make that plain. But, on the whole, we are all still prepared to sign up to the practice points. In this second edition, we have been able to correct the surprising number of minor typographical errors and hope not to have introduced any more, get all the X-rays the right way up, improve on some of the figures, remove some duplication, reorder a few sections, put in some more subheadings to guide the readers, make the section on acute ischaemic stroke more directive, improve the index, and generally tidy the whole thing up. It should now be easier to keep track of intracranial venous thrombosis and, in response to criticism, we have extended the section on leukoaraiosis, even though it is not strictly either a cause or a consequence of stroke. We have also introduced citations to what we have called ‘floating references’ – in other words, published work that is constantly being changed and updated as new information becomes available. An obvious example is the Cochrane Library, which is updated every 3 months and available on CD-ROM and through the Internet. There are no page numbers, and the year of publication is always the present one. We have therefore cited such ‘floating references’ as being in the present year, 2000. But we know that this book will not be read much until the year 2001 and subsequent years, when readers will have to look at the contemporary Cochrane Library, not the one published in 2000. The same applies to the new British Medical Journal series called ‘Clinical Evidence’ which is being updated every 6 months, and to any websites that may be updated at varying intervals and are still very much worth directing readers towards. Rather to our surprise, there is a lot of new information to get across on stroke. Compared with 4 years ago, the concept of organized stroke services staffed by experts in stroke care has taken root and has allowed the increasingly rapid assessment of patients with ‘brain attacks’. It is no longer good enough to sit around waiting 24 h or more to see if a patient is going to have a transient ischaemic attack or a stroke, and then another 24 h for a ·

·

3

9781405127660_4_001.qxd 10/13/07 11:15 AM Page 4

4

Chapter 1 Introduction

computed tomography brain scan to exclude intracerebral haemorrhage. These days we have to assess and scan stroke patients as soon as they arrive in hospital, perhaps give thrombolysis to a few, and enter many more into clinical trials, start aspirin in ischaemic stroke, and get the multidisciplinary team involved – and all of this well within 24 h of symptom onset. Through the Cochrane Library, which was in its infancy when the first edition was published, there is now easy, regularly updated electronic access to systematic reviews of most of the acute interventions and secondary prevention strategies for stroke, although the evidence base for rehabilitation techniques is lagging behind. Catheter angiography is giving way to non-invasive imaging. Magnetic resonance techniques are racing ahead of the evidence as to how they should be used in routine clinical practice. For better or worse, coiling cerebral aneurysms is replacing clipping. The pharmaceutical industry is still tenaciously hanging on to the hope of ‘neuroprotection’ in acute ischaemic stroke, despite numerous disappointments. Hyperhomocysteinaemia and infections are the presently fashionable risk factors for ischaemic stroke, and they may or may not stand the test of time. So, in this second edition, we have tried to capture all these advances – and retreats – and set them in the context of an up-todate understanding of the pathophysiology of stroke and the best available evidence of how to manage it. Of course, it is an impossible task, because something new is always just around the corner. But then ‘breakthroughs’ in medicine take time to mature – maybe years until the evidence becomes unassailable and is gradually accepted by front-line clinicians. And then we can all sit back doing what we believe to be ‘the right thing’ for a few more years until the next ‘breakthrough’ changes our view of the world yet again. We hope that the ideas and recommendations in this book will be sufficient 99% of the time – at least for the next 4 years, when we will have to see about a third edition. ·

1.3 Introduction to the third edition

Six years have gone quickly by since the second edition, much has happened in stroke research and practice in the meantime, and two of the authors are on the edge of retirement – so it is time for this third edition of what we fondly refer to as ‘the book’. Maybe because the original authors were feeling tired, or increasingly unable to cover in depth all we wanted to, or perhaps because we

wanted to ensure our succession, we have recruited four new and younger authors, all of whom have worked closely with us over many years, and whose help we acknowledged in the earlier editions – Gabriel Rinkel, Peter Langhorne, Cathie Sudlow and Peter Rothwell. But, even with their help, the rewriting has had to compete with all the far less interesting things which we have to do these days to satisfy managers, regulatory authorities and others keen to track and measure our every move. And maybe there is less imperative to write books like this which are out of date in at least some ways even before they are published. But then searching the Internet for ‘stroke’ does not come up with a coherent account of the whole subject of managing stroke patients using the best available evidence, which is what this book is all about. So, with the help and encouragement of Blackwell Publishing, here is the third edition of ‘the book’ at last. We have written the book as before with most of the authors commenting on most of the chapters before all the chapters were finally written in the form you can read them in now. Again, you will have to guess who wrote what because we can all lay claim to most of the book in some sense or another. There has been a slight change in the arrangement of the chapters, but loyal readers of the earlier editions will not find this too upsetting – they will still find what they want in more or less its familiar place, and as ever we hope the index has been improved. The practice points we all sign up to and our day-to-day practice should reflect them. The uncertainties we all share – they will be gradually resolved as more research is done, and more uncertainties will then be revealed. The biggest change in this edition is succumbing to the space saving offered by a numbered reference system, and a change in the colour scheme from a pastel green to various shades of purple. As with the second edition, much has changed and there has been more updating than we originally anticipated – what we know about stroke has moved on. Neuroprotection is even less likely to be an effective treatment for ischaemic stroke than it was in the 1990s, we still argue about thrombolysis, clopidogrel cannot very often be recommended, carotid stenting has still to prove its worth, routine evacuation of intracerebral haemorrhage is definitely not a good idea, and hormone replacement therapy far from protecting against vascular disease actually seems to increase the risk. But on the positive side, much has improved in brain and vessel imaging, it is now clear how much blood pressure lowering has to offer in secondary stroke prevention, and cholesterol lowering too. Carotid surgery can now be targeted on the few who really need it, not recommended for the greater number who may or may not need it.

..

..

9781405127660_4_001.qxd 10/13/07 11:15 AM Page 5

1.3 Introduction to the third edition

Coiling has more or less replaced clipping of intracranial aneurysms, an astonishing change in practice brought about by a large trial energetically led by an interventional neuroradiologist and neurosurgeon. And it is not just acute stroke that needs urgent attention nowadays, transient ischaemic attacks must be assessed and managed very quickly to minimize the early high risk of stroke. Stroke services continue to improve all over the world, stroke has moved up the political agenda as we have managed to wrench it out of the rubric of ‘cardiovascular’ disease which always emphasized the cardiac rather than the cerebral, and more and more people are involved in stroke research, which is now a much more

..

..

crowded and competitive field than it was when some of us started in the 1970s. Will there be a fourth edition? We don’t know; this will be in the hands of the remaining authors as Charles Warlow and Jan van Gijn dwindle into retirement of a sort, or at least a life that will not require the relentless battle to keep up with all the stroke literature, critique it, absorb anything that is worthwhile, and then put it into the context of active clinical practice. No one can write well about stroke unless they can connect research with their own active clinical practice – we are not, we hope, armchair theoreticians; we try to practise what we preach.

5

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 7

2

Development of knowledge about cerebrovascular disease 2.1

Ideas change slowly

2.2

The anatomy of the brain and its blood supply

7

2.3

What happens in ‘apoplexy’?

2.4

Cerebral infarction (ischaemic stroke)

2.5

Thrombosis and embolism

16

2.6

Transient ischaemic attacks

17

2.7

Intracerebral haemorrhage

20

2.8

Subarachnoid haemorrhage

2.9

Treatment and its pitfalls

8

10 14

21

25

2.10 Epilogue 28

‘Our knowledge of disorders of the cerebral circulation and its manifestations is deficient in all aspects’ was the opening sentence of the chapter on cerebrovascular diseases in Oppenheim’s textbook of neurology at the beginning of the 20th century.1 More than 90 years later this still holds true, despite the considerable advances that have been made. In fact, the main reason for Oppenheim’s lament, the limitations of pathological anatomy, is to some extent still valid. True, our methods of observation nowadays are no longer confined to the dead, as they were then. They have been greatly expanded, first by angiography, then by brain imaging and measurement of cerebral blood flow and metabolism, and most recently by non-invasive methods of vascular imaging such as ultrasound and magnetic resonance angiography. Yet, our observations are still mostly anatomical, and after the event. It is only in rare instances that are we able to reconstruct the dynamics of a stroke. At least in haemorrhagic stroke, brain computed tomography (CT) or magnetic resonance imaging (MRI) in the acute phase gives an approximate indication of where a blood vessel has ruptured (though not why exactly there and then) and how far the extravasated blood has invaded the brain parenchyma or the subarachnoid space. With ischaemic stroke, the growth of our understanding has been slower. The ubiquity of

the term ‘cerebral thrombosis’ up to the 1970s exemplifies how deficient our understanding was even at that time.2 Embolic occlusion, now known to result more often from arterial lesions than from the heart, can be detected in an early phase by non-invasive angiographic techniques or inferred by means of perfusion imaging, but so often the source of the clot is still elusive. We have also learned to distinguish many causes of cerebral infarction other than atherothrombosis, such as arterial dissection, mitochondrial cytopathies and moyamoya syndrome, but the precise pathogenesis of these conditions is still poorly understood. So it is with humility, rather than in triumph, that we look back on the past. In each era the problems of stroke have been approached by the best minds, with the best tools available. Of course many ideas in the past were wrong, and so presumably are many of our own. Even though we are firm believers in evidence-based medicine, some – perhaps many or even most – of our own notions will not survive the test of time. Our knowledge may have vastly increased in the recent past but it is still a mere island in an ocean of ignorance.

2.1 Ideas change slowly Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

The history of medicine, like that of kings and queens in world history, is usually described by a string of dates 7

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 8

8

Chapter 2 Development of knowledge about cerebrovascular disease

and names, by which we leapfrog from one discovery to another. The interval between such identifiable advances is measured in centuries when we describe the art of medicine at the beginning of civilization, but in mere years where our present times are chronicled. This leads to the impression that we are witnessing a dazzling explosion of knowledge. Some qualification of this view is needed, however. First of all, any generation of mankind takes a myopic view of history in that the importance of recent developments is overestimated. The Swedish Academy of Sciences therefore often waits for years, sometimes even decades, before awarding Nobel prizes, until scientific discoveries have withstood the test of time. When exceptions were made for the prize in medicine, the early accolades were not always borne out: WagnerJauregg’s malaria treatment for neurosyphilis (1927) is no longer regarded as a landmark, while Moniz’s prize (1949) for prefrontal leucotomy no longer seems justified; at least he also introduced contrast angiography of the brain, although this procedure may again not survive beyond the end of this century. We can only hope that the introduction of X-ray CT by Hounsfield (Nobel prize for medicine in 1979) will be judged equally momentous by future generations as by ourselves. Another important caveat if one looks back on progress in medicine is that most discoveries gain ground only slowly. Even if new insights were quickly accepted by peer scientists, which was often not the case, it could still be decades before these had trickled down to the rank and file of medical practitioners. The mention of a certain date for a discovery may create the false impression that this change in medical thinking occurred almost overnight, like the introduction of the single European currency. In most instances, this was far from the truth. An apt example is the extremely slow rate at which the concept of lacunar infarction became accepted by the medical community, despite its potentially profound implications in terms of pathophysiology, treatment and prognosis. The first pathological descriptions date from around 1840,3,4 but it took the clinicopathological correlations of C. Miller Fisher (Fig. 2.7) in the 1960s before the neurological community and its textbooks started to take any notice.5–7 And it was not until new techniques for brain imaging in the 1980s provided instantaneous clinicoanatomical correlations that no practising neurologist could avoid knowing about lacunar infarcts – some 150 years after the first description! It is best to become reconciled to the idea that a slow rate of diffusion of new knowledge is unavoidable. The problem is one of all times. Franciscus Biumi, one of the early pathologists, lamented in 1765: ‘Sed difficile est adultis novas opiniones inserere, evellere insitas’ (But it is difficult to insert new opinions in adults and to remove rooted ones).8 How slowly new ideas were accepted and acted upon, against

the background of contemporary knowledge, can often be inferred from textbooks, particularly if written by fulltime clinicians rather than by research-minded neurologists. Therefore we shall occasionally quote old textbooks to illustrate the development of thinking about stroke. A reverse problem is that a new discovery or even a new fashion may be interpreted beyond its proper limits and linger on as a distorted idea for decades. Take the discovery of vitamin B1 deficiency as the cause of a tropical polyneuropathy almost a century ago; the notion that a neurological condition, considered untreatable almost by definition, could be cured by a simple nutritional supplement made such an impact on the medical community that even in some industrialized countries vitamin B1 is still widely used as a panacea for almost any neurological symptom. So broadly speaking there are two kinds of medical history, that of the cutting edge of research and that of the medical profession as a whole. The landmarks are easy to identify only with the hindsight of present knowledge. In reality, new ideas often only gradually dawned on consecutive scientists, instead of the popular notion of a blinding flash of inspiration occurring in a single individual. For this reason, accounts of the history of stroke are not always identical.9,10 Also many important primary sources are not easy to interpret – not only because they were written in Latin, but also because ‘new observations’ have sometimes been identified only by later historians, in retrospect, while the authors at the time attached no importance to them.11

2.2 The anatomy of the brain and its blood supply

From at least the time of Hippocrates (460–370 BC), the brain was credited with intelligence and thought, and also with movements of the opposite side of the body, through observation of unilateral convulsions after head wounds on the contralateral side.12 Yet, stroke, or ‘apoplexy’ (Greek for ‘being struck down’), was defined as a sudden but mostly general, rather than focal, disorder of the brain. The pathogenesis was explained according to the humoral theory, which assumed a delicate balance between the four humours: blood, phlegm, black bile and yellow bile. Anatomy played almost no part in these explanations. Apoplexy was often attributed to accumulation of black bile in the arteries of the brain, obstructing the passage of animated spirits from the ventricles.13 Galenus of Pergamon (131– 201), a prolific writer and animal experimenter whose

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 9

2.2 The anatomy of the brain and its blood supply

views dominated medicine up to the 17th century,14 distinguished ‘karos’ from ‘apoplexy’, in that respiration was unaffected in the former condition.15 Leading Islamic physicians like Avicenna (980–1037) tried to reconcile Galenic tenets with the Aristotelian view of the heart as the seat of the mind.16 In Western Europe, mostly deprived of Greek learning until the fall of Constantinople in 1453 prompted the Renaissance,17 these Arabic texts were translated into Latin before those of Galen and Hippocrates.18 All these theories had no anatomical counterpart; dissection of the human body was precluded by its divine connotations. Any illustrations of the human brain that are known before the 16th century are crude and schematic representations of Galenic theories, rather than attempts at copying

Fig. 2.1 Plate depicting the blood vessels, from Vesalius’s Tabulae Anatomicae Sex, of 1538.21 This shows the carotid arteries ending up in a network (b) at the base of the brain; the structures marked (a) represent the choroid plexus in the lateral ventricles. The network of blood vessels (rete mirabile) is found in oxen; Galen had assumed it was found also in the human brain, a belief perpetuated throughout the Dark and Middle Ages, up to the early Renaissance. Leonardo da Vinci had also drawn a (human?) brain with a ‘rete mirabile’ at its base.22 Vesalius retracted the existence of a network in his atlas of 1543.

..

..

the forms of nature. As a consequence, many nonneurological disease conditions with sudden onset must have been misclassified as ‘apoplexy’. In 1543 Andries van Wesele (1514–1564), the great Renaissance anatomist who Latinized his name to Andreas Vesalius, produced the first accurate drawings of the brain in his famous book De humani corporis fabrica libri septem, with the help of the draughtsman Johan Stephaan van Calcar and the printer Oporinus in Basle.19 It was the same year in which Copernicus published De revolutionibus, proclaiming the sun and not the earth as the centre of the universe.20 Vesalius largely ignored the blood vessels of the brain, although he retracted an earlier drawing (Fig. 2.1) depicting a ‘rete mirabile’, a network of blood vessels at the base of the brain that

9

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 10

10

Chapter 2 Development of knowledge about cerebrovascular disease

Galen had found in pigs and oxen and that had been extrapolated to the human brain ever since.21,22 Before him, Berengario da Carpi had also denied the existence of the rete.23 Vesalius was vehemently attacked by traditionally minded contemporaries as an iconoclast of Galenic dogmas. Nevertheless, initially, he did not go as far as outright opposition to the central Galenic tenet that blood could pass through the septum between the right and left ventricle of the heart, allowing the mixture of blood and air and the elimination of ‘soot’. Instead, he praised the creator for having made the openings so small that nobody could detect them, another striking example of how the power of theory may mislead even the most inquisitive minds. Only later, in the 1555 edition of his De humani corporis fabrica, did he firmly state that the interventricular septum was tightly closed. The decisive blow to the humoral theory came in 1628, through the description of the circulation by William Harvey (1578–1657);24 it need no longer surprise us that it took many decades before these views were widely accepted. Harvey’s work formed the foundation for the recognition of the role of blood vessels in the pathogenesis of stroke. Thomas Willis (1641–1675) is remembered not so much for having coined the term ‘neurology’, or for his iatrochemical theories, a modernized version of humoral medicine, or for his part in the successful resuscitation of Ann Green after judicial hanging,25 as he is for his work on the anatomy of the brain, first published in 1664,26 especially for his description of the vascular interconnections at the base of the brain (Fig. 2.2).27 Before him, Fallopius, Casserio, Vesling and Wepfer had all observed at least part of the circle,28–31 in the case of Casserio and Vesling even with an illustration.32 But undisputedly, it was Willis who grasped the functional implications of these anastomoses in a passage illustrating his proficiency in performing necropsies as well as postmortem experiments (from a posthumous translation):33 We have elsewhere shewed, that the Cephalick Arteries, viz. the Carotides, and the Vertebrals, do so communicate with one another, and all of them in different places, are so ingraffed one in another mutually, that if it happen, that many of them should be stopped or pressed together at once, yet the blood being admitted to the Head, by the passage of one Artery only, either the Carotid or the Vertebral, it would presently pass thorow all those parts exterior and interior: which indeed we have sufficiently proved by an experiment, for that Ink being squirted in the trunk of one Vessel, quickly filled all the sanguiferous passages, and every where stained the Brain it self. I once opened the dead Carcase of one wasted away, in which the right

Fig. 2.2 Illustration of the base of the brain from Willis’s Cerebri Anatome (1664),26 showing the interconnections between the right and left carotid systems, and also between these two and the posterior circulation (drawing by Christopher Wren).

Arteries, both the Carotid and the Vertebral, within the Skull, were become bony and impervious, and did shut forth the blood from that side, notwithstanding the sick person was not troubled with the astonishing Disease. It seems that the idea of infusing coloured liquids into blood vessels, practised from 1659 onwards and later perfected by Frederik Ruysch (1638–1731) and in the next century by John Hunter (1728–1793),34,35 had come from Christopher Wren (1632–1723).25 Wren also made the etchings for Willis’s book (he is now mainly remembered as the architect of St Paul’s Cathedral and many other churches built after the great fire of London in 1666).

2.3 What happens in ‘apoplexy’?

Willis’s ‘astonishing Disease’, apoplexy, had of old intuitively been attributed to some ill-defined obstruction, whether from want of ‘animal spirits’ via the nerves in

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 11

2.3 What happens in ‘apoplexy’?

the tradition of Greek medicine, or, after Harvey’s time, by deprivation of blood flow. Yet, it should be remembered that the notion of an intrinsic ‘nervous energy’ only slowly lost ground. Even the great 18th-century physician Boerhaave, though clearly recognizing the role of blood vessels and the heart in the development of apoplexy, invoked obstruction of the cerebrospinal fluid.36 In Table 2.1 we have provided a schematic representation of the development of ideas about apoplexy through the ages, together with its relationship to arterial lesions. That Willis had found ‘bony’ and ‘impervious’ arteries in patients who actually had not died from a stroke was probably the reason that he was not outspoken on the pathogenesis of apoplexy. His contemporaries, Wepfer (1620–1695) in Schaffhausen, and Bayle (1622–1709) in Toulouse, only tentatively associated apoplexy with ‘corpora fibrosa’,31 or with calcification of cerebral arteries.37 Wepfer (Fig. 2.3) not only recognized arterial lesions, but he also prompted one of the great advances in the knowledge about stroke by distinguishing between, on the one hand, arterial obstruction preventing the influx

Fig. 2.3 Johann Jakob Wepfer (1620–1695).

..

..

of blood and, on the other, extravasation of blood into the substance of the brain or the ventricular cavities. His interpretation was, however, that blockage of arteries as well as extravasation of blood impeded the transmission of ‘spiritus animalis’ to the brain.11 Accordingly, he regarded apoplexy as a process of global stunning of the brain, while the focal nature of the disease largely escaped him. The four cases of haemorrhage Wepfer described were massive, at the base of the brain or deep in the parenchyma. In cases with obvious hemiplegia, incidentally a term dating back to the Byzantine physician Paulus Aegineta (625–690),38 Wepfer suspected dysfunction of the ipsilateral rather than the contralateral side. He also observed patients who had recovered from apoplectic attacks, and noted that those most liable to apoplexy were ‘the obese, those whose face and hands are livid, and those whose pulse is constantly unequal’. That the paralysis was on the opposite side of the apoplectic lesion was clearly predicted by Domenico Mistichelli (1675–1715) from Pisa on the basis of his observation of the decussation of the pyramids (Fig. 2.4).39 A landmark in the recognition of the anatomical substrate of stroke – and of many other diseases – was the work of Morgagni (1682–1771), professor of medicine and subsequently of pathological anatomy in Padua. In 1761 Morgagni published an impressive series of clinicopathological observations collected over a lifetime (he

Fig. 2.4 Illustration from Mistichelli’s book on apoplexy (1709) in which he shows the decussation of the pyramids and also the outward rotation of the leg on the paralysed side.39

11

Non-haemorrhagic

Rostan (Paris) (1790–1866)57

‘Ramollissement’ (1820): – softening more frequent than haemorrhage – condition not inflammatory?

Rostan

Baillie (London) (1761–1823)42

Morgagni (Padua) (1682–1771)40 ‘Sanguineous apoplexy’ (1761)

Boerhaave

Willis (Oxford) (1621–1675)33

Boerhaave (Leiden) ‘Stoppage of the spirits’ (1668–1738)36 ‘Serous apoplexy’, extravasation of serum? (1761)

Paralysis is unilateral, and crossed with respect to lesion (1709)

Ossification of cerebral arteries (1820)

Hardening of arteries associated with haemorrhage? (1795)

Narrowing due to cartilaginous change (1735)

‘Bony, impervious arteries’ (1684)

1815 Battle of Waterloo; Schubert writes Erlkönig

1776 US Declaration of Independence

1729 Bach writes St Matthew Passion

1707 Union between England and Scotland

1682 Peter I ascends Russian throne

Mistichelli (Pisa) (1675–1715)39

Calcifications (1677)

Bayle (Toulouse) (1622–1709)37

1642 Rembrandt paints Night Watch

‘Corpora fibrosa’

Wepfer (1658)

Extravasation of blood in brain tissue (1658)

Historical events

Wepfer (Schaffhausen) (1620–1695)31

Medical scientist Observations on arterial lesions

0 Birth of Jesus Christ

Sudden loss of consciousness, as a result of brain disease

Haemorrhagic

Ideas about ‘apoplexy’

Galenus Sudden loss of consciousness, as a result of brain disease (Pergamum and Rome) (131–201)15

Hippocrates (Kos) (460–370 BC)13

Medical scientist

12

Table 2.1 Development of ideas about ‘apoplexy’ and its relationship with arterial lesions.

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 12

Chapter 2 Development of knowledge about cerebrovascular disease

..

..

..

..

Cerebral softening caused by obliteration of arteries? (one of possible causes; 1838)

Carswell (London) (1793–1857)62

1837 Queen Victoria ascends the throne of the British Empire

Due to ossification of arteries?

1869 Opening of the Suez Canal 1871 Stanley meets Livingstone at Ujiji 1877 Bell invents telephone, Edison the phonograph 1895 Röntgen discovers X-rays in Würzburg 1907 Ehrlich introduces arsphenamine as treatment for syphilis

Arteriosclerosis leads to thrombosis; thrombi may be torn off and lodge distally (‘embolism’) (1856) End-arteries most vulnerable; paradoxical embolism Thrombosis at the carotid bifurcation may cause secondary embolization to brain (1905)

Cerebral softening caused by capillary Virchow congestion, secondary to ‘irritation’ (1842)

Cohnheim

Chiari (Prague) (1851–1916)70

‘Yellow softening’ of the brain is secondary to arterial obliteration; any inflammation is secondary (1856) ‘Infarction’ (stuffing) is haemorrhagic by definition, as opposed to ischaemic necrosis (1872)

Virchow (Berlin) (1821–1902)66

Cohnheim (Berlin) (1839–1884)68

1859 Darwin publishes The Origin of Species 1863 Manet paints Le Déjeuner sur l’herbe

1848 Year of revolutions; Louis Napoleon elected president of France

1829 Stephenson builds the railway engine called ‘The Rocket’

‘Arteriosclerosis’ (1829)

Cruveilhier (Paris) (1791–1874)63

‘Encephalomalacia’ (1844): – white, or serous (congestion) – red (inflammatory) – yellow (frequent; unexplained)

Abercrombie

Cerebral softening analogous to gangrene of limb? (1836)

Abercrombie (Edinburgh) (1780–1844)60

Rokitansky (Vienna) (1804–1878)219

Lobstein (Strasbourg) (1777–1835)67

Cerebral softening is definitely inflammatory in nature (1824)

Lallemand (Montpelier) (1790–1853)58

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 13

2.3 What happens in ‘apoplexy’? 13

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 14

14

Chapter 2 Development of knowledge about cerebrovascular disease

was 79 at the time of publication), in which he firmly put an end to the era of systemic (humoral) theories of disease and replaced them by an organ-based approach, though he did not include even a single illustration; characteristically, the title of the book was ‘De sedibus et causis morborum . . .’ (about the sites and causes of disease).40 Morgagni not only confirmed the notion of crossed paralysis but also firmly divided apoplexy into ‘sanguineous apoplexy’ and ‘serous apoplexy’ (and a third form which was neither serous nor sanguineous). A decade later, Portal (1742–1832) rightly emphasized that it was impossible to distinguish between these two forms during life.41 However, it would be a mistake to assume that ‘serous’ (non-haemorrhagic) apoplexy was recognized at that time as being the result of impaired blood flow, let alone of mechanical obstruction of blood vessels. Some even linked the arterial hardening with brain haemorrhages and not with the serous apoplexies.42 Although we quoted 17th-century scientists such as Bayle and Wepfer in that they associated some non-haemorrhagic cases of apoplexy with obstruction of blood flow, in the 18th century medical opinion swayed towards ‘vascular congestion’, a kind of pre-haemorrhagic state. That explanation was propounded not only by Morgagni40 but also by many of his contemporaries and followers.41,43,44 John Cheyne (1777–1836) pointed out that autopsy in patients who had survived a ‘stroke of apoplexy’ for a considerable time might show a cavity filled with rusty serum that stained the adjacent brain tissue, but he may have been describing a residual lesion after cerebral haemorrhage rather than infarction.45 The anatomical, organ-based approach exemplified by Morgagni reflected the Italian practice, in which the separation between physicians and surgeons was much less strict than in northern Europe with its more theoretical framework of medicine. The protagonists of the Northern school of thinking were Herman Boerhaave (1668–1738) in Leiden and later William Cullen (1710– 1790) in Edinburgh, the most influential clinical teachers of their time. They established a nosological classification that was based much more on holistic theory, in terms of a disturbed system, than on actual observations at the level of the organ, at least with 20th-century hindsight.46 Probably our own time will be branded as the era of exaggerated reductionism! In the intellectual tradition of the Dutch-Scottish school, purely clinical classifications of apoplexy were proposed in the early 19th century by Serres (with and without paralysis),47 by Abercrombie (primary apoplexy, with deprivation of sense and motion, and sometimes with convulsions, a second type beginning with headache, and a third type with loss of power on one side of the body and of speech, often with recovery),48 and by Hope and Bennett (transient apoplexy, primary apoplexy with death or slow

recovery, ingravescent apoplexy with partial recovery and relapse, and paraplexic apoplexy with paralysis).49 There are several reasons why the brain lesion in what we now call cerebral infarction was not identified until the middle of the 19th century. First, it was impossible to recognize ischaemic softening in patients who had usually died not long after their stroke. Fixation methods were not available until the end of the 18th century; Vicq d’Azyr, Marie Antoinette’s physician, was the first to use alcohol as a tissue fixative,50 while formaldehyde fixation was not employed until a century later.51 Second, it is probable that many patients diagnosed as having died from apoplexy in fact had suffered from other conditions. If in our time the diagnosis is wrong in 20–30% of patients referred with a presumed stroke,52–54 the diagnostic accuracy was presumably no better in centuries past.

2.4 Cerebral infarction (ischaemic stroke)

After Morgagni’s seminal book the organ-based approach to medicine quickly spread from Italy to other countries. In France, the first proponents were surgeons. After the French revolution the strict distinction between medicine and surgery disappeared, driven by the reorganization of hospital care (no longer managed by the church but by the state) and by the need to train a large number of new doctors, for military as well as civilian duties (‘peu lire, beaucoup voir, beaucoup faire’).55,56 It was Leon Rostan (1790–1866; Fig. 2.5), a physician at the Salpêtrière in Paris, who clearly recognized softening of the brain as a separate lesion, distinct from haemorrhage, although the pathogenesis still escaped him. He published his findings in an unillustrated monograph, the first edition of which appeared in 1820.57 The lesions were most commonly found in the corpus striatum, thalamus or centrum semiovale, but they also occurred in the cerebral cortex, brainstem and cerebellum. Old cases showed a yellowish-green discoloration, whereas if the patients had died soon after the event the colour of the lesion was chestnut or reddish. The softening might be so extreme as to lead to the formation of a cyst. In other patients it was difficult to detect any change in firmness or in colour. Rostan distinguished softening of the brain from ‘apoplexy’, a term he no longer used for stroke in general, but which he regarded as being synonymous with haemorrhagic stroke. He supposed that softening of the brain was more common than brain haemorrhage, although some haemorrhages were secondary to softening. The clinical manifestations were thought

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 15

2.4 Cerebral infarction (ischaemic stroke)

seemed for a long time the most logical ‘paradigm’ to fall back on to explain liquefaction of brain tissue.59 The first inkling of a relationship between arterial disease and ‘ramollissement’, as many English writers continued to call brain softening in deference to Rostan, was voiced by Abercrombie, in a later edition of his textbook.60 He drew an analogy with gangrene, caused by ‘failure of circulation’, this in turn being secondary to ‘ossification of arteries’. The role of arterial obstruction as a primary cause of softening of the brain was confirmed by others,61,62 but the theory of inflammation continued to be defended by a few adherents.63,64 Some were aware that apoplexy could be caused by ‘cerebral anaemia’ (as opposed to congestion), not only through loss of blood but also by a reduced vascular pressure, particularly in the case of heart disease.44 Other missing links in the understanding of cerebral infarction were clarified by Rokitansky (1804–1878) in Vienna and by Virchow (1821–1902) in Berlin. Rokitansky divided cerebral softening (which he termed encephalomalacia) into three varieties: red (haemorrhagic) softening, inflammatory in nature; white softening (synonymous with ‘serous apoplexy’) caused by congestion and oedema; and, the most common variety, yellow softening, of which the pathogenesis was unknown.219 Virchow (Fig. 2.6) revolutionized medical thinking about vascular disease by firmly putting the

Fig. 2.5 Léon Rostan (1790–1866).

to occur in two stages: first ‘fugitive’ disturbances in the use of a limb, in speech or in visual or auditory perception, sooner or later followed by hemiplegia and coma, in a slowly progressive fashion. Although Rostan recognized ‘ossification’ of the cerebral arteries, he did not associate these lesions with cerebral softening via obstruction of the arterial system. At any rate he doubted the prevailing opinion that the primary lesion was some kind of inflammatory response. After all, there was redness and swelling (rubor, tumor), if not warmth and pain (calor, dolor), to complete the cardinal signs of inflammation delineated by Celsus in the first century AD. Rostan’s contemporary Lallemand (1790– 1853) was much more outspoken and had little doubt that inflammation was at the root of cerebral softening.58 Readers trained in the 21st century may find this difficult to understand but they should be aware that ‘inflammation’ was a rather common explanation for disease from the middle of the 18th century until some hundred years later.46 Just as in our time some poorly understood medical conditions are often interpreted in terms of autoimmune disease, perhaps erroneously, inflammation

..

..

Fig. 2.6 Rudolph Virchow (1821–1902) teaching at a postmortem in the Charité Hospital in Berlin.

15

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 16

16

Chapter 2 Development of knowledge about cerebrovascular disease

emphasis on changes in the vessel wall rather than in blood; Schiller called it the victory of ‘solidism’ over ‘humoralism’.9 Virchow also firmly established that thrombosis of arteries was caused not by inflammation but by fatty metamorphosis of the vessel wall, even if he had to found his own journal before his papers could be published.65,66 To describe the changes in the arterial wall Virchow revived the term ‘arteriosclerosis’, first used by Lobstein.67 Virchow’s disciple Julius Cohnheim coined the culinary term ‘infarction’ (from the Latin verb infarcire, ‘to stuff into’), but strictly reserved it for haemorrhagic necrosis (‘stuffing’, by seeping of blood into ischaemic tissue, through damaged walls of capillaries) as opposed to ischaemic necrosis.68

2.5 Thrombosis and embolism

Virchow observed thrombosis as the result of atherosclerosis, and also embolism, in patients with gangrene of the lower limbs caused by clots from the heart. The term ‘embolism’ was newly coined by him, at least in medical parlance. He extrapolated these events to the cause of cerebral softening: Here there is either no essential change in the vessel wall and its surroundings, or this is ostensibly secondary. I feel perfectly justified in claiming that these clots never originated in the local circulation but that they are torn off at a distance and carried along in the blood stream as far as they can go.65 The relationship between vegetations on the heart valves and stroke had in fact been suggested a century earlier by Boerhaave’s pupil Gerard van Swieten, personal physician to the Austrian empress Maria Theresa and founder of the Viennese school of medicine: It has been established by many observations that these polyps occasionally attach themselves as excrescences to the columnae carneae of the heart, and perhaps then separate from it and are propelled, along with the blood, into the pulmonary artery or the aorta, and its branches . . . were they thrown into the carotid or vertebral arteries, could disturb – or if they completely blocked all approach of arterial blood to the brain – utterly abolish the functions of the brain.69 For more than a century after Virchow’s accurate pathological descriptions of arterial occlusions, the term ‘cerebral embolism’ was almost synonymous with embolism from the heart (parenthetically, it still is, in many contemporary textbooks and papers – another

illustration of how slowly ideas change). Sources of embolism in the extracranial arteries were hardly considered until the 1960s, at least in teaching. By the same token, the term ‘cerebral thrombosis’ remained firmly entrenched in clinical thinking as being more or less synonymous with cerebral infarction without associated heart disease, the implication being that in these cases the site of the atheromatous occlusion was in the intracranial vessels. For example, this is what the sixth edition of Brain’s Diseases of the Nervous System says on the subject in 1968: Progressive occlusion of cerebral blood vessels impairs the circulation in the regions they supply. The effects of this depend upon the size and situation of the vessel, and the rate of onset of the occlusion particularly in relation to the collateral circulation. Actual obstruction of an artery by atheroma, with or without subsequent thrombosis, causes softening of the region of the brain supplied by the vessel.2 That the notion of ‘local thrombosis’ persisted for such a long time must have been because of its appealing simplicity, not by lack of observations to the contrary. As long ago as 1905, Chiari drew attention to the frequency of atherosclerosis in the region of the carotid bifurcation and suggested that embolization of atheromatous material might be a cause of cerebral softening,70 and not much later Hunt described the relationship between carotid occlusion and stroke.71 The much later general acceptance of extracranial atherosclerosis as an important cause of cerebral ischaemia was prompted by two further developments. The first was the attention generated by Fisher’s studies, in which he re-emphasized the role of atherosclerosis at the carotid bifurcation, at least in white patients.72 He clinically correlated these lesions not only with contralateral hemiplegia but also with attacks of monocular blindness in the ipsilateral eye.73 The second development was imaging. Cerebral angiography by direct puncture of the carotid artery had been introduced by Moniz in 1927,74,75 but imaging of the carotid bifurcation in patients with stroke became common only after the advent of catheter angiography,76 and later of ultrasound techniques. Now it is modern CT or MR angiography that often shows abnormalities of the internal carotid artery near its origin, at least in patients with transient or permanent deficits in the territory of the main trunk of the middle cerebral artery or one of its branches. The earlier patients are investigated, the greater the chance of detecting the site where the embolus has become impacted in the cerebral arterial tree. The therapeutic implications of identifying lesions in the carotid artery in symptomatic patients became clear through the two large randomized controlled trials of carotid endarterectomy in the 1980s and 1990s,

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 17

2.6 Transient ischaemic attacks

which showed overall benefit from the operation for severe degrees of stenosis.77 In 25–30% of patients with temporary or permanent occlusion of large intracranial vessels no source of embolism can be found in the neck or in the heart.78,79 Pathological observations suggesting that the aorta may harbour atherosclerotic lesions80 have been confirmed in a large autopsy series and by transoesophageal echocardiography during life.81,82 Of course, there is more to ischaemic stroke than embolism from large vessels, but the history of small vessel disease and non-atheromatous causes of ischaemia is rather more recent. Before concluding the sections on cerebral infarction, thrombosis and embolism, we should like briefly to draw attention to the term ‘cerebrovascular accident’, which enjoyed some undeserved popularity in the middle half of the last century. The problem was that sometimes the term was used as a synonym for cerebral infarction, at other times to denote stroke in general. In this day and age the term is a highly specific sign of woolly thinking. We can do no better than quote Schiller:9 That rather blurry and pompous piece of nomenclature must have issued from the well-meant tendency to soften the blow to patients and their relatives, also from a desire to replace ‘stroke’, a pithy term that may sound unscientific and lacking gentility. ‘Cerebrovascular accident (CVA)’ can be traced to the early 1930s – between 1932, to be exact, when it was still absent from the 15th edition of Dorland’s Medical Dictionary, and the following edition of 1936 where it first appeared. The occasional medical student or junior doctor who still takes refuge in the term ‘CVA’ in an attempt to cover up ignorance about the precise type of cerebrovascular event in a given patient (while avoiding sharing the term ‘stroke’ with the laity) should either find out or come clean about not knowing.

2.6 Transient ischaemic attacks

It is difficult to trace the first descriptions of what we now call transient ischaemic attacks (TIAs) of the brain or eye, because symptoms representing focal deficits were not clearly distinguished from non-specific symptoms of a more global nature such as fainting or headache.83 Wepfer recorded that he had seen patients who recovered from hemiplegia in one day or less.31 An 18th-century account has been retrieved in the patient’s own words, not muddled by medical interpretation, which makes it as lucid as it would have been today.

..

..

The subject is Jean Paul Grandjean de Fouchy, writing in 1783, at the age of 76 years:84 Toward the end of dinner, I felt a little increase of pain above the left eye and in that very instant I became unable to pronounce the words that I wanted. I heard what was said, and I thought of what I ought to reply, but I spoke other words than those which would express my thoughts, or if I began them I did not complete them, and I substituted other words for them. I had nevertheless all movements as freely as usual . . . I saw all objects clearly, I heard distinctly what was being said; and the organs of thought were, it seemed to me, in a natural state. This sort of paroxysm lasted almost a minute. Once it had become established, in the middle of the 19th century, that cerebral softening was not caused by an inflammatory process but by occlusion of cerebral arteries, temporary episodes of ischaemia were recognized increasingly often in the next few decades.1,85–89 In the course of time, three main theories have been invoked to explain the pathophysiology of TIAs, at least in relation to atherosclerosis: the vasospasm theory, the haemodynamic theory and the thromboembolic theory.83

2.6.1 The vasospasm theory Arterial spasm as a cause of gangrene of the extremities was described by Raynaud (1834–1881) in his doctoral thesis of 1862.90 Others extrapolated his theory of vasospasm to the cerebral circulation.91,92 Russel, writing in 1909 about a 50-year-old farmer who had suffered three attacks of tingling and numbness in the right arm and the right side of the face, dismissed thrombosis (‘Thrombus, once formed, does not break up and disappear in some mysterious way’) and instead invoked a phenomenon of ‘local syncope’, analogous to Raynaud’s disease or some cases of migraine: ‘There must be some vessel constriction, local in site, varying in degree and in extent, coming and going, intermittent’.92 Even the great Osler mounted the bandwagon of the vasospastic theory to explain transient attacks of aphasia and paralysis: ‘We have plenty of evidence that arteries may pass into a state of spasm with obliteration of the lumen and loss of function in the parts supplied’.89 Vasospasm remained the most popular theory to explain TIAs in the first half of the 20th century and provided the rationale for so-called cerebral vasodilators. Up to the 1980s these useless drugs were still widely prescribed in some European countries, not only for TIAs but for ‘senility’ in general; in France they were the third most commonly prescribed medication in 1982.93 In the front line of medicine, however, the vasospastic theory went into decline soon after World War II, firstly

17

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 18

18

Chapter 2 Development of knowledge about cerebrovascular disease

because the cerebral arteries are among the least reactive in the body,94,95 and secondly because more plausible theories emerged (see below). Only under strictly defined conditions can vasospasm be a causal factor in the pathogenesis of cerebral ischaemia, that is, after subarachnoid haemorrhage or in association with migraine, and even in these conditions its role is contentious. Nevertheless, vasospasm has resurfaced as a possible cause of episodes of transient monocular blindness that are frequent and stereotyped and have no altitudinal distribution,96 or even of transient motor or sensory deficits not related to migraine.97 Such events must be extremely rare.

2.6.2 The haemodynamic theory The notion of ‘low flow’ without acute vessel obstruction as a cause of cerebral ischaemia should perhaps be attributed to Ramsay Hunt, who drew an analogy between the symptoms of carotid stenosis or occlusion and the symptoms of intermittent claudication in patients with severe peripheral arterial disease.71 But, it was especially after 1951, when Denny-Brown suggested that TIAs might be caused by ‘episodic insufficiency in the circle of Willis’,95 that interest in the haemodynamic aspects of TIAs was fully aroused. Indeed, it was mainly the surgical community for which the concept of ‘cerebral intermittent claudication’ continued to have great appeal, despite the incongruity of the relatively constant blood flow to the brain and the large fluctuations in flow that occur in the legs, depending on their level of activity. Clinical studies failed to support the notion of haemodynamic failure. After artificial lowering of the blood pressure by means of hexamethonium and postural tilting, in 35 patients who had either experienced TIAs or who had known carotid artery disease, only one of the patients developed symptoms of focal cerebral ischaemia before a syncopal attack which signified global rather than focal ischaemia of the brain.98 Similarly, cerebral ischaemia with naturally occurring attacks of hypotension, such as cardiac arrhythmias, is almost always syncopal and not focal in nature,99 and cardiac arrhythmias do not occur more often in patients with TIAs than in controls.100 Once the first successful carotid reconstruction had been reported,101 the intuitive belief in the haemodynamic theory led to an ever-increasing number of carotid endarterectomies being performed (indeed, often called ‘carotid disobstruction’) in patients with and even without TIAs, despite the absence of any formal proof of efficacy. These developments caused understandable concern in the neurological community.102,103 Fortunately the controversy prompted well-designed clinical trials, which have served to define to a large extent the role of this operation.77

That the haemodynamic theory does not apply to most patients with TIAs is not to say that the exceptional patient cannot suffer from ‘misery perfusion’.104 In the presence of multiple occlusions or stenoses of the extracranial arteries, the haemodynamic reserve may be so poor that minor changes in systolic blood pressure cannot be compensated for. Such triggering events include a change from a sitting to a standing position, turning the head, heating of the face or looking into bright light.105–107 Perhaps for this small group of patients extracranial-intracranial bypass surgery has something to offer after all, despite the negative results of the randomized trial in a large but relatively unselected group of patients with occlusion of the internal carotid or middle cerebral artery.108

2.6.3 The thromboembolic theory In the 1950s C. Miller Fisher (Fig. 2.7) not only gave new impetus to some older observations about the relationship between stroke and atheromatous lesions of

Fig. 2.7 C. Miller Fisher (1913–).

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 19

2.6 Transient ischaemic attacks

F

D

F C B A D

..

F

H G B A D

E

G

C B A D

(F) 9:37 E

•

• •

•

•

E

F

H G

C B A D

(E) 9:35

M F

H

H

L

J F

(G) 9:55 E

E

G

C

(D) 9:30

(C) 9:20

B A D

E

H G

C

(B) 9:19

B A D

(A) 9:15

E

F

H G

C

B A

the carotid bifurcation, but he also provided evidence that the pathogenesis was more complex than could be explained by fixed arterial narrowing. First, he saw a patient in whom hemiplegia had been preceded by attacks of transient monocular blindness in the contralateral eye, that is, ‘the wrong eye’.73 Second, through assiduous ophthalmoscopic observations, he saw white bodies passing slowly through the retinal arteries during an attack of transient monocular blindness (Fig. 2.8), the whitish appearance and friability of the moving material suggesting that these were emboli, largely made up of platelets.109 These findings were confirmed by Ross Russell,110 whilst others saw atheromatous emboli in the retinal vessels, which did not move but had become impacted.111,112 After these direct observations of the ocular fundus, additional – but more indirect – arguments corroborated the notion of artery-to-artery embolism as an important cause of TIAs. • In many patients with attacks involving the cortical territory of the middle cerebral artery there is an associated lesion of the internal carotid artery, but in only very few of them is the stenosis severe enough, with a residual lumen of 1–2 mm, for blood flow to be impaired below critical levels, even assuming there is no collateral circulation.113 In addition, the stenosis is constant but the episodes of ischaemia transient,

..

F

G Cattle trucking in veins

C Fig. 2.8 Diagrams of observations in a patient with an attack of transient monocular blindness in the left eye (except the upper temporal quadrant); the attack had started at 8.55 am, 20 minutes before the beginning of the observations. The column of blood in the retinal arteries was in some places interrupted by white segments, initially at the stems of the superior and inferior retinal arteries (A); also the column of blood in at least six venous branches of the superior half of the retina was broken into transverse bands (so-called cattle trucking). The white segments in the retinal arteries slowly passed through the superior temporal artery (B–H). At (C) the vision in the upper half of the visual field had returned. At (D) a fine trickle of erythrocytes moved slowly along one side of white segment AB to the superior nasal artery, and at (E) vision had also returned in the inferior temporal quadrant. After (H), when the column of blood had been completely restored, vision returned to normal. (From Fisher, 1959;109 by kind permission of the author and Neurology.)

H

S

C

(H) 10:00

B D

H

E

without evidence for cardiac arrhythmias as an additional factor. During carotid endarterectomy, fresh and friable thrombi are seen adherent to atheromatous plaques in the carotid bifurcation, especially in patients with recent attacks.114 In patients with ocular as well as cerebral attacks, the two kinds of attack almost never occur at the same time.114 Manual compression of the carotid artery may lead to dislodgement of atheromatous emboli to the cerebral circulation.115 If patients continue to have TIAs after occlusion of the ipsilateral internal carotid artery, there is often an additional atheromatous lesion in the common carotid or external carotid artery, these vessels being important collateral channels, supplying the hemisphere via retrograde flow through the ophthalmic artery.106 Asymptomatic emboli have been seen to flash up during angiography,116 while fibrin thrombi have been seen to pass through a cortical artery during craniotomy for a bypass procedure.117 Transcranial Doppler monitoring has uncovered an ongoing stream of highintensity transient signals (HITS), probably small emboli, in patients with symptomatic carotid lesions.118 The HITS disappear after carotid endarterectomy,119 the rate depending on the interval since operation.120

19

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 20

20

Chapter 2 Development of knowledge about cerebrovascular disease

Whilst artery-to-artery thromboembolism from atheromatous plaques may seem the most important factor in explaining TIAs and ischaemic strokes, it is not necessarily the only one, not even in a single individual patient. For example, it is probable that emboli have especially damaging effects in vascular beds that are chronically underperfused.

2.7 Intracerebral haemorrhage

Extravasation of blood into the brain parenchyma was recognized as early as 1658 by Wepfer,31 although we commented above that he saw the clot as an obstruction of ‘vital spirits’ rather than as the disease in itself, and subsequently by Morgagni.40 The cause remained obscure, and to a large extent it still is. In 1855, before blood pressure could be measured, Kirkes observed hypertrophy of the heart in 17 of 22 patients with fatal brain haemorrhage.121 Charcot and Bouchard in 1868 examined the brains of patients who had died from intracerebral haemorrhage and immersed these in running water; they found multiple, minute outpouchings of small blood vessels, so-called miliary aneurysms.122 The irony of these two names being joined is that Bouchard, once Charcot’s pupil, in later years generated much hostility between himself and his former chief, because he wanted to found a school of his own and to be considered the most influential man in the faculty of

medicine.123 It was in this adversarial atmosphere that in 1892 Bouchard, as president of the jury that had to judge the competition for the rank of professeur agrégé, did not admit Charcot’s pupil Babinski.124 Babinski subsequently left academic medicine by becoming chief of the Pitié hospital, where he devoted much time to the study of clinical signs, including the now famous ‘toe sign’.125 The aneurysms described by Charcot and Bouchard were white or brownish-coloured nodules about 0.5–2.0 mm in diameter, attached to a small arteriole, most often in the basal ganglia. At the beginning of the 20th century, Charcot and Bouchard’s theory came under attack and some proposed that the primary lesion in intracerebral vessels was atherosclerosis, that most of these dilatations were not true aneurysms at all but false aneurysms caused by intramural dissection, while rupture could also occur by weakening of the vessel wall without previous aneurysm formation;126 also some ‘miliary aneurysms’ may in fact have been clots in perivascular (Virchow-Robin) spaces. Alternative explanations for the pathogenesis of primary intracerebral haemorrhage included primary necrosis of brain tissue or its vessels. Some assumed that arteries dilate and rupture only when a previous infarct had occurred, thus depriving the feeding vessel of its normal support.127,128 The frequent coexistence of hypertension led to several theories other than plain rupture. Rosenblath postulated that a renal toxin caused necrosis of vessel walls,129 Westphal that arterial spasm was an intermediate factor130 and Schwartz that a multitude of terminal arterial branches became permeable.224 In the 1960s injection techniques revived the notion

Fig. 2.9 Godfrey N. Hounsfield (1919– 2004), the British engineer who received the Nobel prize in medicine in 1979 for the development of computed tomography (together with the American physicist A.M. Cormack).

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 21

2.8 Subarachnoid haemorrhage

2.8.1 Diagnosis The first unequivocal description of an aneurysm, though unruptured, was by Franciscus Biumi in 1765, who saw it not on the circle of Willis but in the cavernous sinus (at the time called Vieussens’ receptacle).8 Morgagni had also mentioned dilatations of arteries that may have been aneurysms.40 In 1812 John Cheyne provided the first illustration of lethal subarachnoid haemorrhage at the base of the brain as a result of ‘rupture of the anterior artery of the cerebrum’ (Fig. 2.11), but the aneurysm that must have been the source of the haemorrhage was not recognized at the time.45 One year later Blackall reported a postmortem observation in which the haemorrhage as well as the offending aneurysm (of the basilar artery) were identified in a 20-year-old woman.142 The observation was coincidental, because Blackall was primarily interested in her ‘anasarca’ (generalized oedema, or ‘dropsy’). The brain was also examined by Hodgson, who in his book on diseases of blood vessels Fig. 2.10 CT scan of an intracerebral haemorrhage, from the early 1970s.221

of microaneurysms,131,132 although some still suspect that the injection pressures can artifactually distend or rupture vessel walls.133 Amyloid angiopathy was first recognized as a cause of primary intracerebral haemorrhage in the first half of the 20th century.134–136 This type of haemorrhage occurs especially at the border of white and grey matter and not in the deep regions of the brain that are the most common sites of haemorrhages associated with microaneurysms. The first series of such patients appeared in the 1970s.137,138 The invention of computed tomography by Hounsfield (1919–2004; Fig. 2.9) in the 1970s made it possible to distinguish intracerebral haemorrhage quickly and reliably from cerebral infarction (Fig. 2.10).139,140

2.8 Subarachnoid haemorrhage

The history of ‘meningeal apoplexy’ is relatively short. The disorder was not recognized until three years before the battle of Waterloo; in the following 125 years numerous accounts appeared that combined a few personal cases with attempts to review the entire world literature up to that time, the last being a heroic overview of 1125 patients.141

..

..

Fig. 2.11 The first anatomical illustration of subarachnoid haemorrhage, from Cheyne (1812).45 A probe has been passed into the proximal end of the internal carotid artery and emerges at the presumed site of rupture; the offending aneurysm was not recognized at the time but presumably it was at the origin of the posterior communicating artery from the carotid artery, or at the anterior communicating artery complex.

21

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 22

22

Chapter 2 Development of knowledge about cerebrovascular disease

Fig. 2.12 Pea-sized aneurysmal dilatation of one of the branches of the middle cerebral artery (opened), containing a clot; the abnormality was surrounded by a large, fresh haemorrhage, which had caused the death of the 19-year-old patient, 8 days after a first episode with sudden headache (from Bright, 1831).61

made the point that the extravasated blood was contained under the arachnoid membrane.143 Serres, not aware of these books, published two similar observations in a French periodical.144 Parenthetically, it should be pointed out that medical journals, with articles about a variety of observations, did not emerge until the beginning of the 19th century, whereas the first scientific journals in general date from the middle of the 17th century.145 In England, Richard Bright, one of the champions of the movement of ‘organ-based medicine’ that had started in Italy and France,146 included an illustration of a pea-sized aneurysm on a branch of the middle cerebral artery in his richly illustrated book that appeared in 1831 (Fig. 2.12).61 Series of other fatal cases were reported in the next few decades.147–150 The erroneous notion that aneurysms are congenital malformations, caused by a defect in the muscular layer of the arterial wall, was first put forward in 1887,151 and subsequently adopted by other writers,152,153 to be perpetuated into contemporaneous textbooks and students’ minds. Turnbull also pointed out, correctly, that syphilis was an extremely rare cause of cerebral aneurysms.153 Series of aneurysms diagnosed post mortem were often biased towards those measuring several cm,154 or included septic aneurysms, associated with endocarditis.155 It took a long time before the clinical features were sorted out. In 1852 Brinton observed that fatal rupture was not the only possible presentation of aneurysms, and that other manifestations were local pressure,

convulsive attacks or ‘inflammation’ (a rather fuzzy notion at the time).147 The sudden onset of the headache led Lebert and Bartholow to suppose, in the 1860s and 70s, that the diagnosis might be made during life.149,150 Lebert also observed the characteristic paralysis of the oculomotor nerve in patients before they died from rupture of an aneurysm at the origin of the posterior communicating artery from the internal carotid artery.149 Indeed the diagnosis of ruptured aneurysm was made on two occasions in a patient with sudden headache and oculomotor palsy, by Hutchinson in England and by Bull in Norway,156,157 but apparently these two observations had little impact. The introduction of lumbar puncture, in 1891 by Quincke,158 initially only for therapeutic purposes in hydrocephalic patients, led to the diagnosis of ‘meningeal apoplexy’ in patients who survived a subarachnoid haemorrhage.159–161 It took another three decades before the connection with rupture of a cerebral aneurysm was made. After all, the condition was supposed to be invariably fatal since it had been recognized only after death. The ‘selection bias’ seems obvious in retrospect, but in our own time the same error was made with intraventricular haemorrhage – until CT scanning showed it in those who survived. That sudden headache and meningeal haemorrhage as diagnosed by lumbar puncture could be caused by a ruptured aneurysm without fatal outcome received widespread attention only after Charles Symonds (1890– 1978; Fig. 2.13) had published two landmark articles about the subject in 1923 and 1924.162,163 It had all started in 1920, when Symonds spent some time abroad as a temporary resident in the service of the neurosurgeon Harvey Cushing (1869–1939), who had moved not long before from Baltimore to Harvard University and the Peter Bent Brigham hospital in Boston. A 52-year-old woman had been admitted with repeated episodes of headache and unconsciousness; on examination she had a right oculomotor palsy and blurring of the optic discs. A right subtemporal decompression for a suspected tumour showed recently clotted blood extending over the entire hemisphere, apparently coming from the base of the skull.162 Apparently Symonds suggested a ruptured aneurysm as the cause.164 When the diagnosis was confirmed at autopsy (the patient had died the day after the operation) Cushing ordered Symonds to spend his remaining time in the library to review everything on the subject: ‘Either this was a fluke or there was reason in it’.165 The next advances were neuroradiological. The first angiographic visualization of a cerebral aneurysm during life was reported by Egas Moniz in 1933,166 six years after the technique had been first applied.74 In those days angiography was a hazardous procedure (involving

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 23

2.8 Subarachnoid haemorrhage

Fig. 2.14 Norman Dott (1897–1973).

Fig. 2.13 Sir Charles Symonds (1890–1978).

surgical dissection of the carotid artery), to such an extent that someone like Cushing only rarely had his patients undergo it before neurosurgical exploration. Even today, in the era of selective catheterization, the risks are far from negligible. Fortunately, minimally or non-invasive techniques of angiography by means of computed tomography or magnetic resonance have largely replaced catheter angiography, at least for diagnostic purposes. The greatest leap forward in our times was the advent of computed tomography;139 this technique made it possible to localize the extent of the haemorrhage in a precise fashion, to separate aneurysmal haemorrhage from nonaneurysmal haemorrhage and, by serial investigations, to detect and distinguish the most important complications: rebleeding, delayed ischaemia and hydrocephalus.

2.8.2 Surgical treatment Ligation of the carotid artery has been practised since the times of Ambroise Paré (1510–1590) as a method to stop arterial bleeding in patients with neck wounds. Once

..

..

aneurysms were recognized as the cause of subarachnoid haemorrhage it was a logical step to consider this procedure as a method to decrease the risk of rebleeding.157 Hutchinson would actually have carried out the operation in 1864 had the patient not declined at the last moment, going on to survive for another 11 years.156 Around 1886 Horsley was one of the first who actually ligated the (common) carotid artery in the neck, for a tumorous aneurysm.154 For decades carotid ligation remained the only surgical intervention possible, but most patients were managed conservatively because the complications of surgery were considerable.167 In 1931 the Edinburgh neurosurgeon Norman Dott (1897–1973; Fig. 2.14), at that time only 33 years old, carried out the first intracranial operation for a ruptured aneurysm.168 It was a more or less desperate attempt because the aneurysm had already rebled twice, leaving the patient comatose for some hours after the last episode, also with some degree of right-sided hemiparesis and aphasia. To complicate matters further, the patient was a well-known Edinburgh solicitor, 53 years old and chairman of the board of governors of the Royal Hospital for Sick Children. But, both the patient and the young neurosurgeon were prepared to take the risk.169 About the operation Dott wrote:170 A left frontal approach was employed and it was a difficult matter to elevate the tense and oedematous

23

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 24

24

Chapter 2 Development of knowledge about cerebrovascular disease

twelve minutes. As the retaining instrument was then cautiously withdrawn, no further bleeding occurred. The vessel was further cleared and thin strips of muscle were prepared and wound around it until a thick collar of muscle embedded the aneurysm and adjacent arterial trunks (Fig. 2.15). The patient recovered well and a few weeks later Dott wrote, his sense of triumph carefully hidden: ‘Mr Colin Black’s tibialis anticus seems to have stuck well to his internal carotid – he has gone for a holiday’.169 In later years Dott and his patient went fishing together on a number of occasions and Mr Black’s neurological condition remained good until he died from myocardial infarction 11 years after the momentous operation. Unfortunately, on later occasions the outcome with a direct approach to the aneurysm was often disappointing, if not fatal, and Dott reverted to ligating the internal carotid artery in the neck or the proximal anterior cerebral artery intracranially. In 1937 Dandy was the first to use a clip to occlude the neck of the aneurysm that had bled (Fig. 2.16).171 Yet in some patients a clip could not be secured, and in those

Fig. 2.15 Norman Dott’s drawing of the first intracranial operation for aneurysm. The proximal middle cerebral artery aneurysm was exposed and wrapped with muscle through a left frontal flap. (From Todd et al., 1990;168 by kind permission of the authors and the Journal of Neurology, Neurosurgery and Psychiatry.)

brain and identify the basal structures, which were bloodstained and largely embedded in clot. The left optic nerve was found and the internal carotid artery was defined at its outer side. This vessel was closely followed upwards, outwards and backwards to its bifurcation into the middle and anterior cerebral arteries. As this point was being cleared of tenacious clot a formidable arterial haemorrhage filled the wound. With the aid of suction apparatus, held closely to the bleeding point, we were able to see the aneurysm. It sprang from the upper aspect of the bifurcation junction; it was about 3 mm in diameter; blood spurted freely from its semidetached fundus. Meanwhile a colleague was obtaining fresh muscle from the patient’s leg. A small fragment of muscle was accurately applied to the bleeding point and held firmly in place so that it checked the bleeding and compressed the thin walled aneurysmal sac. Thus it was steadily maintained for

Fig. 2.16 Illustration from Dandy’s 1944 monograph on intracerebral arterial aneurysms.222 The legend is: ‘Typical aneurysm of the intracranial internal carotid artery, showing the narrow neck of the sac and the bulging aneurysm; also the point of rupture. The inset shows the clip placed on the neck of the aneurysm, and the aneurysm itself shrivelled with the electric cautery.’

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 25

2.9 Treatment and its pitfalls

cases he often had to have recourse to so-called trapping, by clipping the parent vessel on either side of the aneurysm. Decades later, Drake devised a technique for approaching basilar artery aneurysms, notoriously difficult until then, and managed to apply clips to them.172 In the 1960s, spring clips, which could be removed when placement was less than optimal, came into use and replaced the silver clips used by Dandy. Nevertheless, the direct operation of aneurysms remained dangerous and controlled trials of the efficacy of aneurysm operations were equivocal. Attempts to increase the safety of the operation included temporary cardiac arrest, hypotension and deep hypothermia, all without much success, although no formal trials were done. In the 1980s, a consensus developed amongst neurosurgeons that direct operation of the aneurysm should best be delayed until 12–14 days after the initial haemorrhage. This regimen meant, of course, that a proportion of patients rebled or suffered other complications in the meantime. The gradual introduction of the operating microscope for aneurysm surgery in the 1970s made early operation (within 3 days) not only feasible but also fashionable, despite the dearth of evidence from controlled clinical trials. The medical management of patients with ruptured aneurysms has also improved in recent years, especially the prevention of delayed ischaemia. In the 1980s the Italian neuroradiologist Guglielmi developed an endovascular method for occluding aneurysms by means of detachable platinum coils, initially only for aneurysms for which a surgical approach was hazardous or impossible.173,174 In the last few years ‘coiling’ has largely replaced the surgical approach, provided the method is feasible for a particular aneurysm.

2.9 Treatment and its pitfalls

Doctoring has always implied treatment. In the past, medical management was almost invariably based on what later turned out to be erroneous pathophysiological concepts, and the treatments were almost invariably ineffective, if not actually harmful. Such pitiful situations are often repeated in present times, much more often than physicians and surgeons care to realize. Anyone who finds it amusing to read about 19th-century regimens, including measures such as bleeding, mustard poultices, castor oil and turpentine enemas as treatments for apoplexy, should read post-1950 treatises about the efficacy of vasodilator drugs or about transplantation

..

..

of omentum to the intracranial cavity, as a chastening experience.

2.9.1 The numerical method Before different treatments could ever be compared, it was necessary to find methods for grouping patients together and also somehow to convert disease outcomes into numbers. The Paris physician Pierre Charles Alexandre Louis (1787–1872; he survives eponymously in the angulus Ludovici of the sternum) is generally credited with the introduction of the numerical method in medicine. In fact his contribution was more a credo than a practical method.175 True enough, there is the famous example of his empirical criticism of bloodletting: of 47 patients with pneumonia treated with bloodletting 18 died, against only nine of the 36 patients in the untreated group.176 But Louis did not have the mathematical training to estimate the likelihood that a difference of this magnitude might arise by chance. It was a mathematician, Jules Gavarret (1809–1890), who criticized the analysis and conclusions of Louis’s studies, although he agreed with the design.177 Even more purely mathematical was the notion of the ‘average human’, an approach proposed by Adolphe Quetelet (1796–1874). Groups and averages, these were notions that evoked not merely resistance but outright revulsion in the ranks of the established medical professionals. How on earth could one ever ignore the unique characteristics of each single individual by forcing these together into an artificial ‘mean’? And how could one ever believe in a standard treatment, any more than in a standard shoe? The advent of experimental physiology intensified the opposition. The famous Claude Bernard (1813–1878) warned that one will never encounter an ‘average’ in nature, and that grouping of observations will obscure the true relationships between natural phenomena.178 And the equally legendary Lord Lister (1827–1912) relied more on the theoretical basis of his antiseptic method than on the actual death rates.179 Until the 20th century, counting disease events was limited to population studies.180 The beginnings of epidemiology can be traced to Sir William Petty (1623– 1687), one of the founders of the Royal Society, and John Graunt (1620–1674). They worked together in collecting numerical data to describe patterns of mortality. A century and a half later, E. Blackmore reported not only on deaths but also on incident cases of disease in Plymouth.181,182 Victorian counterparts took this further. William Farr (1807–1883), who had trained under Louis in Paris, linked self-devised classifications of diseases and occupations to population statistics at the General Registry Office. John Snow (1813–1858) mapped the

25

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 26

26

Chapter 2 Development of knowledge about cerebrovascular disease

occurrence of cholera cases in the streets of London and related these to the positions of the local water pumps; these studies culminated in the famous act of Snow removing the handle from the pump in Broad Street, during the 1854 cholera epidemic. Incidentally, he later specialized in chloroform anaesthesia.183 The first epidemiological studies of stroke were not performed until after World War II. An early study of stroke incidence in the community was done in the UK.184 Population-based studies addressing risk factors specifically for stroke were subsequently reported from the US (the Framingham cohort), Japan and Finland.185–187

2.9.2 Clinical trials The introduction of the randomized controlled clinical trial heralded the era of ‘evidence-based medicine’ or rather ‘organized empiricism’, since medicine is not and probably will never be a positivist science like physics or chemistry.188 Randomization in a therapeutic experiment slowly gained acceptance after the landmark UK Medical Research Council (MRC) trial of streptomycin in pulmonary tuberculosis with random assignment to treatment groups.189 Some forerunners had already used parallel control groups. Louis (1787–1872) had been preceded by James Lind (1716–1794) in 1753 (lemons and oranges to prevent scurvy in sailors). A further step was the introduction of chance to obtain an equal balance between the experimental group and the control group. In 1898 Fibiger (1867–1928) used assignment on alternate days (injection of serum for diphtheria),190,191 and in 1931 Amberson et al. flipped a coin to divide patients with pulmonary tuberculosis into those who received gold treatment and controls.192 Blinding (or masking, as ophthalmologists prefer to say) of patients was also practised by Amberson’s group, as had been done four years earlier by Ferguson et al. in a test of vaccines for the common cold.193 Masking of those who were to assess outcome was advocated in 1944 by the pulmonary physicians Hinshaw and Feldman,194 and eventually carried out in the MRC streptomycin trial of 1948. Allocation in that historic trial took place by means of randomization. An important advantage of random allocation, applied by R.A. Fisher in agriculture in the 1920s, is that it ensures equal and unbiased balancing between the two groups.195 But the main reason why Sir Austin Bradford Hill (1897–1991), the trial’s principal investigator, chose randomization is that at the same time it ensured concealment of the allocation schedule from those involved in entering patients in the trial.196,197 Clinical trials in cerebrovascular disease were no exception to the rule that most methodological errors have to

be committed before they are recognized, as the correct solutions are often counter-intuitive. In the 1950s anticoagulant drugs seemed a rational form of treatment to prevent further strokes in survivors of (presumed) brain infarction. The same Bradford Hill who had pioneered the tuberculosis trial took the initiative for two such trials, the first in 142 and the second, with exclusion of hypertensives, in 131 patients.198,199 There was no significant difference in the rate of non-fatal stroke between the treatment groups and controls, while there was some excess of fatal strokes, possibly haemorrhages, in patients on anticoagulants. From that time onwards anticoagulants were largely abandoned for the prevention of stroke, unless for specific indications such as a source of embolism in the heart. However, it took at least two decades before it dawned on the neurological community that the trials of anticoagulants in brain ischaemia had been too small, separately as well as collectively, to detect even large protective effects – apart from other shortcomings.200 The same applied to an early secondary prevention trial with dipyridamole.201 The first intervention trial in acute stroke was with corticosteroids, by Dyken and White in 1956. They did not use randomization but stratified patients according to their clinical characteristics, and found a trend towards a higher death rate in the treated group (13/17 against 10/19 in controls), and ended up by identifying many of the methodological problems in this type of trial.202 The first trial of carotid endarterectomy excluded surgical mishaps from the analysis;203 subsequently the operation boomed to worrying levels, until checked by methodologically sound trials. The first large trial of aspirin in stroke prevention evoked much controversy,204 for one thing because its initiators had chosen ‘stroke or death’ as the outcome event instead of stroke alone;205 it took time for neurologists to realize that they treat whole patients rather than only their brains! Also, the initial conclusion that aspirin was ineffective for women is now a classical example of the dangers of subgroup analysis.

2.9.3 Measuring outcome: the ghost of Gall One of the stumbling blocks in trials of acute stroke used to be the babel of tongues with regard to the measurement of outcome. Initially, so-called ‘stroke scales’ were applied for this aim, analogous to scales for other specific neurological conditions, such as Parkinson’s disease or multiple sclerosis. Although the stated purpose of ‘stroke scales’ is to measure outcome, these scales are nothing but codifications of the neurological examination, while of course that examination has no other purpose than localizing lesions within the nervous system. With such a diagnostic approach, different functions of the nervous

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 27

2.9 Treatment and its pitfalls

system are separately assessed: power of limbs, speech, visual fields, etc. This reductionist, mechanistic notion of brain function reflects the localizationists’ position in the scientific battle that raged in the second half of the 19th century, the opposing party believing in so-called equipotentiality. The ‘equipotentialists’ believed that the brain worked as a unitary system, brain tissue being omnipotent and flexible in its function. Consequently, brain damage would result in a decrease in the overall level of performance, but not in loss of specific functions. The champion of this camp was the French physiologist Flourens, who supported his views with experiments on dogs and pigeons.206 The alternative notion, that of localization of specific functions, was propounded in a somewhat bizarre fashion by the anatomists Gall and Spurzheim.207 They believed that every intellectual and moral property had its own position on the surface of the brain (Fig. 2.17) and that the degree of development of these dispositions could be identified by locating overlying protuberances on the skull (Fig. 2.18). However, the theory of localization gained respectability after the stimulation experiments of Fritsch and Hitzig on anaesthetized dogs, in which they found that weak electrical currents applied through platinum electrodes to the anterior regions of

Fig. 2.18 The pseudo-science of phrenology lived on well beyond the 19th century. The Lavery Electric Phrenometer of 1907 was intended to lend modern accuracy to the measurements of bumps on the skull.223 (Reproduced by kind permission of Cambridge University Press.)

Fig. 2.17 Phrenology head (Fowler); each region of the skull is supposed to represent a mental faculty, such as ‘mirthfulness’, ‘perception of form’ or ‘ideality’.

..

..

the brain surface produced muscle contractions in the opposite half of the body.208,209 The clash between the two opposing factions culminated in 1881 at the Third International Congress of Medicine, held in London.210 The equipotentialists were represented by the German physiologist Goltz, who showed the audience a dog in which a substantial portion of the brain had been removed by means of a hose, but who could still move all four limbs, trunk and tail, and who had retained all his senses. Later, it would turn out that the lesions were less extensive than had been claimed. On the same afternoon Ferrier showed two chimpanzees, one deaf after removal of the auditory cortex, the other limping with a hemiplegic gait after extirpation of the contralateral motor area (the sight of which led Charcot to jump up and exclaim: ‘Mais c’est un patient!’). The localizationists had won the day, but they won too completely. The greater part of the brain has no ‘primary’ motor, sensory or cognitive tasks, and serves to

27

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 28

28

Chapter 2 Development of knowledge about cerebrovascular disease

Table 2.2 The state of an individual cannot be constructed from separate components. Imagine that you met ‘the boy next door’ from your childhood days after an interval of 30 years, and that your question ‘How are you?’ was answered with a list of details, instead of by a general statement (‘fine’ for example). Profession Civil state Bank account Car Holidays Sport

dentist married positive Volvo Tuscany golf

for 15 years wife 1.68 m, 59 kg a9634.92 240S 3 weeks handicap 8

Total ?

connect and integrate the separate ‘functions’. Similarly, everyday life consists of a multitude of tasks that are integrated and difficult to separate. Mood, initiative and speed of thinking are some of the essential features of human life that can be severely affected by stroke but are sadly ignored in ‘stroke scales’. It is therefore naive to try and rebuild an entire human being from separate ‘building blocks’ (Table 2.2; Fig. 2.19). Patients are more than the sum of their signs. A higher, more integrated level of measurement is needed; that is, scales should measure function not at the level of the organ but at the level of the person (disability scales), or even at the level of social interaction (handicap scales). What really counts for patients is what they can do in life, compared with what they want to do or were once able to do.

2.9.4 Meta-analysis and systematic reviews In the last quarter of the 20th century, Richard Peto and his colleagues Tom Chalmers and Iain Chalmers developed a method to overcome the problem that single studies may or may not show a significant difference in treated patients compared with controls, but that the magnitude of the difference can only be expressed as a confidence interval, which is usually wide. They collated all related trials in a given field by which the differences between the treatment group and the control group in each trial could be combined.211 The key assumption is that, if a given treatment has any material effect on the incidence or outcome of disease, then the direction, although not necessarily the size, of this effect tends to be similar in different circumstances. If all available studies are combined, the confidence interval can be narrowed considerably and reviewer bias is avoided. There clearly was a pressing need for up-to-date systematic reviews of all the available evidence regarding the various aspects of care of stroke patients – indeed, of all medical interventions. This need led to the Cochrane Collaboration, which includes a stroke review group.212

Fig. 2.19 The librarian (1566), by Giuseppe Arcimboldo (1530–1593). Oil on canvas, 97 × 71 cm. (By kind permission of Skokloster Castle, Sweden.)

The graphic representation of systematic reviews started in 1978, with simple lines to depict 95% confidence intervals.213 In 1982 Lewis and Clarke had the idea to combine the separate estimates into an overall estimate, at the bottom of the figure.214,215 Subsequently Richard Peto’s group solved the paradox that small trials were most conspicuous because of their large confidence intervals, by putting a square at the site of the point estimate, the size of the square being proportional to the power of the trial (Fig. 2.20).216 These graphs have since become known as ‘forest plots’, probably because the many lines might be seen as trees.215

2.10 Epilogue

Despite the many advances in the knowledge about stroke that we have highlighted, our story could not but

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 29

References

Ratio of crude death rates (99% Cl) Beta-blocker: control

remain anachronistic and fragmented. It is extremely difficult to try and stand in the shoes of one’s forebears, because to achieve this the mind should be cleared from all knowledge obtained since their time.217 For those of us who can think back as small a time span as three decades, what diagnosis did we make, in those times, in patients we now know to have survived carotid dissection or intracranial venous thrombosis, to name but two examples? Heaven only knows. In the same way, not so much longer ago, it was impossible to distinguish haemorrhage from infarction; or haemorrhage from some mysterious other condition that mimicked haemorrhage but in which the brain looked practically normal; or stroke from other brain diseases; or even stroke from heart disease. Necessarily our account has been anecdotal. In reality the progress of science is slow and continuous, not a succession of breakthroughs. This also applies to the few decades we have witnessed during our own careers. We do not expect a sensational novelty when we walk into hospital tomorrow, but a lot has changed since we were medical students. This refers not only to the body of medical knowledge but also to the methods of medical research. Empirical testing has gained ascendancy over pathophysiological theory, for the treatment as well as the prevention of disease. The rate of change is a bit like the shifting position of the sun across the sky: one cannot see it move, but there is a dramatic sweep between dawn and sunset. We expect to see many more dawns in stroke research.

References 0

0.5

1.0

1.5

2.0

Beta-blocker better Beta-blocker worse Treatment effect P < 0.0001 Fig. 2.20 ‘Forest plot’. Figure redrawn after Lewis and Ellis’s original plot from 1982, which for the first time combined 99% confidence intervals of different placebo-controlled clinical trials of beta blockers after myocardial infarction.214 This modern variant shows the results of each component study as a square centred on the point estimate of the result of each study; the size of each square is proportional to the amount of information provided in that trial. A horizontal line runs through the square to show its confidence interval (CI). The overall estimate from the meta-analysis and its confidence interval are put at the bottom, represented as a diamond.215 (Reproduced by kind permission of the BMJ Publishing Group.)

..

..

1 Oppenheim H. Lehrbuch der Nervenkrankheiten für Ärtzte und Studierende. 6th ed. Berlin: S. Karger, 1913. 2 Brain WR. Diseases of the Nervous System. 6th ed. Oxford: Oxford University Press, 1968. 3 Dechambre A. Mémoire sur la curabilité du ramollissement cérébral. Gaz Med Paris 1838; 6:305–14. 4 Durand-Fardel CLM. Mémoire sur une altération particulière de la substance cérébrale. Gaz Med Paris 1842; 10:23–38. 5 Fisher CM. Lacunes: small, deep cerebral infarcts. Neurology 1965; 15:774–84. 6 Fisher CM, Curry HB. Pure motor hemiplegia of vascular origin. Arch Neurol 1965; 13:30–44. 7 Fisher CM. The arterial lesions underlying lacunes. Acta Neuropathol (Berl) 1969; 12:1–15. 8 Biumi F. Observatio V: Carotis ad receptaculum Vieusenii aneurysmatica etc. In: Sandifort E, editor. Observationes anatomicae, scholiis illustratae (thesaurus dissertationum). Leiden: S. & J. Lichtmans, 1765, pp. 373–9.

29

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 30

30

Chapter 2 Development of knowledge about cerebrovascular disease

9 Schiller F. Concepts of stroke before and after Virchow. Medical History 1970; 14:115–31. 10 McHenry LC. A history of stroke. Int J Neurol 1981; 15:314–26. 11 Karenberg A. Johann Jakob Wepfers Buch uber die Apoplexie (1658). Kritische Anmerkungen zu einem Klassiker der Neurologie. Nervenarzt 1998; 69:93–8. 12 McHenry LC. Garrison’s History of Neurology. Springfield: Charles C. Thomas, 1969. 13 Clarke E. Apoplexy in the Hippocratic writings. Bull Hist Med 1963; 37:301–14. 14 Karenberg A, Hort I. Medieval descriptions and doctrines of stroke: preliminary analysis of select sources. Part I: The struggle for terms and theories – late antiquity and early middle ages (300–800). J Hist Neurosc 1998; 7:162–73. 15 Galenus. Opera Omnia. Leipzig: Cnobloch, 1824. 16 Karenberg A, Hort I. Medieval descriptions and doctrines of stroke: preliminary analysis of select sources. Part II: Between Galenism and Aristotelism – Islamic theories of apoplexy (800–1200). J Hist Neurosc 1998; 7:174–85. 17 Jardine L. Worldly Goods: a new history of the Renaissance. London: Macmillan, 1996. 18 Karenberg A, Hort I. Medieval descriptions and doctrines of stroke: preliminary analysis of select sources. Part III: Multiplying speculations – the high and late middle ages (1000–1450). J Hist Neurosc 1998; 7:186–200. 19 Vesalius A. De humani corporis fabrica. Basle: J. Oporini, 1543. 20 Copernicus N. De revolutionibus orbium coelestium. Nuremberg: J. Petreius, 1543. 21 Vesalius A. Tabulae anatomicae. Venice: D. Bernardini, 1538. 22 Clarke E, Dewhurst K. An Illustrated History of Brain Function: imaging the brain from antiquity to the present. 2nd ed. San Francisco: Norman, 1996. 23 Berengario da Carpi J. Isagogae breves, perlucide ac uberrime, in anatomiam humani corporis etc. Venice: Benedictum Hectoris, 1535. 24 Harvey W. Exercitatio anatomica de motu cordis et sanguinis in animalibus. Frankfurt: G. Fitzer, 1628. 25 Dewhurst K. Thomas Willis’s Oxford Lectures. Oxford: Sandford Publications, 1980. 26 Willis T. Cerebri Anatome. London: Martyn & Allestry, 1664. 27 Meyer A, Hierons R. Observations on the history of the ‘Circle of Willis’. Medical History 1962; 6:119–30. 28 Fallopius G. Observationes anatomicae. Venice: Marcus Antonius Ulmus, 1561. 29 Casserio G. Tabulae Anatomicae (edited by D. Bucretius). Venice: E. Deuchinum, 1627. 30 Vesling J. Syntagma anatomicum, locis pluribus actum, emendatum, novisque iconibus diligenter exornatum. Patavii: Pauli Frombotti Bibliopolae, 1647. 31 Wepfer JJ. Observationes anatomicae, ex cadaveribus eorum, quos sustulit apoplexia, cum exercitatione de ejus loco affecto. Schaffhausen: J.C. Suteri, 1658. 32 Tatu L, Moulin T, Monnier G. The discovery of encephalic arteries. From Johann Jacob Wepfer to Charles Foix. Cerebrovasc Dis 2005; 20:427–32. 33 Willis T. Dr. Willis’s Practice of Physick. London: Dring, Harper & Leigh, 1684.

34 Kidd M, Modlin IM. Frederik Ruysch: master anatomist and depictor of the surreality of death. J Med Biogr 1999; 7:69–77. 35 Moore W. The Knife Man: the extraordinary life and times of John Hunter, father of modern surgery. London: Bantam Press, 2005. 36 van Eems J. Hermanni Boerhaave Praelectiones Academicae de Morbis Nervorum. Leiden: Petrus van der Eijk and Cornelius de Pecker, 1761. 37 Bayle F. Tractatus de apoplexia. Toulouse: B. Guillemette, 1677. 38 Aegineta P. The Seven Books (translated by Francis Adams). London: The Sydenham Society, 1844. 39 Mistichelli D. Trattato dell’apoplessia. Roma: A. de Rossi, 1709. 40 Morgagni GB. De sedibus et causis morborum per anatomen indigatis libri quinque. Venice: ex typographica Remondiana, 1761. 41 Portal A. Observations sur l’apoplexie. Histoire de l’Académie des Sciences 1781; 83:623–30. 42 Baillie M. The Morbid Anatomy of Some of the Most Important Parts of the Human Body. London: J. Johnson & G. Nicol, 1793. 43 Hall M. Lectures on the Nervous System and its Diseases. London: Sherwood, Gilbert & Piper, 1836. 44 Burrows G. On Disorders of Cerebral Circulation and on the Connection between Affections of the Brain and Diseases of the Heart. London: Longman, Brown, Green & Longmans, 1846. 45 Cheyne J. Cases of Apoplexy and Lethargy with Observations on Comatose Patients. London: Underwood, 1812. 46 King LS. Transformations in American Medicine: from Benjamin Rush to William Osler. Baltimore: Johns Hopkins University Press, 1991. 47 Serres ERA. Nouvelle division des apoplexies. Ann Med Chir 1819; 1:246–363. 48 Abercrombie J. Pathological and Clinical Researches on Diseases of the Brain and Spinal Cord. Edinburgh: Waugh & Innes, 1828. 49 Hope J, Bennett JH, Pritchard JC, Taylor RH, Thomson T. Dissertations on nervous diseases. In: Tweedie A, editor. Library of Practical Medicine. Philadelphia: Lea & Blanchard, 1840. 50 Vicq d’Azyr F. Traité d’anatomie et de physiologie. Paris: F.A. Didot, 1786. 51 Blum F. Der Formaldehyd als Härtungsmittel: vorläufige Mitteilung. Z wiss Mikr mikr Technik 1893; 10:314–5. 52 Harbison J, Hossain O, Jenkinson D, Davis J, Louw SJ, Ford GA. Diagnostic accuracy of stroke referrals from primary care, emergency room physicians, and ambulance staff using the face arm speech test. Stroke 2003; 34:71–6. 53 Heckmann JG, Stadter M, Dutsch M, Handschu R, Rauch C, Neundorfer B. Einweisung von NichtSchlaganfallpatienten auf eine Stroke Unit [Hospitalization of non-stroke patients in a Stroke Unit]. Dtsch Med Wochenschr 2004; 129:731–5. 54 Ronning OM, Thommessen B. Nar hjerneslagdiagnosen er feil [Stroke: when the diagnosis is wrong]. Tidsskr Nor Laegeforen 2005; 125:1655–7.

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 31

References

55 Foucault M. Naissance de la clinique. Paris: Presses Universitaires de France, 1963. 56 Bynum WF. Science and the Practice of Medicine in the Nineteenth Century. Cambridge: Cambridge University Press, 1994. 57 Rostan L. Recherches sur le ramollissement du cerveau. Ouvrage dans lequel on s’efforce de distinguer les diverses affections de ce viscère par des signes caractéristiques. 1st ed. Paris: Béchet, 1820. 58 Lallemand F. Recherches anatomo-pathologiques sur l’encéphale et ses dépendances. Paris: Béchet, 1824. 59 Kuhn TS. The Structure of Scientific Revolutions. Chicago: Chicago University Press, 1962. 60 Abercrombie J. Pathological and Practical Researches on Diseases of the Brain and Spinal Cord. 2nd (from 3rd Brit.) ed. Philadelphia: Carey, Lea & Blanchard, 1836. 61 Bright R. Reports of Medical Cases, selected with a view of illustrating the symptoms and cure of diseases by a reference to morbid anatomy. London: Longman, Rees, Orme, Brown & Green, 1831. 62 Carswell R. Pathological Anatomy: illustrations of the elementary forms of disease. London: Longman & Co., 1838. 63 Cruveilhier J. Anatomie pathologique du corps humain; descriptions avec figures lithographiés et coloriés; des diverses altérations morbides dont le corps humain est susceptible. Paris: J.B. Baillière, 1842. 64 Durand-Fardel CLM. Traité du ramollissement du cerveau. Paris: J.-B. Baillière, 1843. 65 Virchow RLK. Ueber die akute Entzündung der Arterien. Archiv Pathol Anat 1847; 1:272–378. 66 Virchow R. Thrombose und Embolie: Gefässentzündung und septische Infection. In: Virchow R, editor. Gesammelte Abhandlungen zur wissenschaftlichen Medizin. Frankfurt: Meidinger, 1856, pp. 219–732. 67 Lobstein JFM. Traité d’anatomie pathologique. Paris: Levrault, 1829. 68 Cohnheim J. Untersuchungen ueber die embolischen Processe. Berlin: Hirschwald, 1872. 69 van Swieten GLB. Commentaria in Hermanni Boerhaave Aphorismos De Cognoscendis et Curandis Morbis. Leiden: J. & H. Verbeek, 1755. 70 Chiari H. Über das Verhalten des Teilungswinkels des Carotis Communis bei der Endarteritis chronica deformans. Verh Ddtsch path Ges 1905; 9:326–30. 71 Hunt JR. The role of the carotid arteries, in the causation of vascular lesions of the brain, with remarks on special features of the symptomatology. Am J Med Sci 1914; 147:704–13. 72 Fisher CM. Occlusion of the internal carotid artery. Arch Neurol Psych 1951; 65:346–77. 73 Fisher CM. Transient monocular blindness associated with hemiplegia. Arch Ophthalmol 1952; 47:167–203. 74 Moniz E. L’encéphalographie artérielle, son importance dans la localisation des tumeurs cérébrales. Rev Neurol (Paris) 1927; 48:72–90. 75 Moniz E. Die cerebrale Arteriographie und Phlebographie. Berlin: Julius Springer, 1940.

..

..

76 Seldinger SI. Catheter replacement of the needle in percutaneous arteriography. Acta Radiol 1953; 39:368–78. 77 Rothwell PM, Eliasziw M, Gutnikov SA, Fox AJ, Taylor DW, Mayberg MR et al. Analysis of pooled data from the randomised controlled trials of endarterectomy for symptomatic carotid stenosis. Lancet 2003; 361:107–16. 78 Sacco RL, Prabhakaran S, Thompson JL, Murphy A, Sciacca RR, Levin B et al. Comparison of warfarin versus aspirin for the prevention of recurrent stroke or death: subgroup analyses from the Warfarin-Aspirin Recurrent Stroke Study. Cerebrovasc Dis 2006; 22:4–12. 79 Jood K, Ladenvall C, Rosengren A, Blomstrand C, Jern C. Family history in ischemic stroke before 70 years of age: the Sahlgrenska Academy Study on Ischemic Stroke. Stroke 2005; 36:1383–7. 80 Soloway HB, Aronson SM. Atheromatous emboli to central nervous system. Arch Neurol 1964; 11:657–67. 81 Amarenco P, Duyckaerts C, Tzourio C, Henin D, Bousser MG, Hauw JJ. The prevalence of ulcerated plaques in the aortic arch in patients with stroke. N Engl J Med 1992; 326:221–5. 82 Amarenco P, Cohen A, Tzourio C, Bertrand B, Hommel M, Besson G et al. Atherosclerotic disease of the aortic arch and the risk of ischemic stroke. N Engl J Med 1994; 331:1474–9. 83 Hachinski VM. Transient cerebral ischemia: a historical sketch. In: Clifford Rose F, Bynum WF, editors. Historical Aspects of the Neurosciences (Festschrift for M. Critchley). New York: Raven Press, 1982, pp. 185–93. 84 Benton AL, Joynt RJ. Early descriptions of aphasia. Arch Neurol 1960; 3:205–22. 85 Wood GB. Treatise on the Practice of Medicine. Philadelphia: Lippincott, 1852. 86 Jackson JH. A lecture on softening of the brain. Lancet 1875; ii:335–8. 87 Hammond WA. Diseases of the Nervous System. New York: D. Appleton, 1881. 88 Gowers WR. A Manual of Diseases of the Nervous System. 2 ed. London: J&A Churchill, 1893. 89 Osler W. Transient attacks of aphasia and paralysis in states of high blood pressure and arteriosclerosis. Can Med Assoc J 1911; 1:919–26. 90 Raynaud M. De l’asphyxie locale et de la gangrène symmétrique des extrémités. Paris: L. Leclerc, 1862. 91 Peabody GL. Relation between arterial disease and visceral changes. Trans Assoc Am Physicians 1891; 6:154–78. 92 Russel W. A post-graduate lecture on intermittent closing of the cerebral arteries: its relation to temporary and permanent paralysis. Br Med J 1909; 2:1109–10. 93 Payer L. Medicine and Culture: notions of health and sickness in Britain, the US, France and West Germany. London: V. Gollancz, 1989. 94 Pickering GW. Transient cerebral paralysis in hypertension and in cerebral embolism with special reference to the pathogenesis of chronic hypertensive encephalopathy. J Am Med Assoc 1948; 137:423–30. 95 Denny-Brown D. The treatment of recurrent cerebrovascular symptoms and the question of ‘vasospasm’. Med Clin North Am 1951; 35:1457–74.

31

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 32

32

Chapter 2 Development of knowledge about cerebrovascular disease

96 Burger SK, Saul RF, Selhorst JB, Thurston SE. Transient monocular blindness caused by vasospasm. N Engl J Med 1991; 325:870–3. 97 Call GK, Fleming MC, Sealfon S, Levine H, Kistler JP, Fisher CM. Reversible cerebral segmental vasoconstriction. Stroke 1988; 19:1159–70. 98 Kendell RE, Marshall J. Role of hypotension in the genesis of transient focal cerebral ischaemic attacks. Br Med J 1963; 2:344–8. 99 Reed RL, Siekert RG, Merideth J. Rarity of transient focal cerebral ischemia in cardiac dysrhythmia. J Am Med Assoc 1973; 223:893–5. 100 De Bono DP, Warlow CP. Potential sources of emboli in patients with presumed transient cerebral or retinal ischaemia. Lancet 1981; i:343–6. 101 Eastcott HHG, Pickering GW, Robb CG. Reconstruction of internal carotid artery in a patient with intermittent attacks of hemiplegia. Lancet 1954; ii:994–6. 102 Warlow C. Carotid endarterectomy: does it work? Stroke 1984; 15:1068–76. 103 Barnett HJM, Plum F, Walton JN. Carotid endarterectomy: an expression of concern. Stroke 1984; 15:941–3. 104 Klijn CJM, Kappelle LJ, Tulleken CAF, van Gijn J. Symptomatic carotid artery occlusion: A reappraisal of hemodynamic factors. Stroke 1997; 28:2084–93. 105 Caplan LR, Sergay S. Positional cerebral ischaemia. J Neurol Neurosurg Psychiatry 1976; 39:385–91. 106 Bogousslavsky J, Regli F. Delayed TIAs distal to bilateral occlusion of carotid arteries: evidence for embolic and hemodynamic mechanisms. Stroke 1983; 14:58–61. 107 Ross Russell RW, Page NGR. Critical perfusion of brain and retina. Brain 1983; 106:419–34. 108 The EC/IC Bypass Study Group. Failure of extracranialintracranial arterial bypass to reduce the risk of ischemic stroke. Results of an international randomized trial. N Engl J Med 1985; 313:1191–200. 109 Fisher CM. Observations on the fundus oculi in transient monocular blindness. Neurology 1959; 9:333–47. 110 Ross Russell RW. Observations on the retinal blood-vessels in monocular blindness. Lancet 1961; 11:1422–8. 111 Witmer R, Schmid A. Cholesterinkristall als retinaler arterieller Embolus. Ophthalmologica 1958; 135:432–3. 112 Hollenhorst RW. Significance of bright plaques in the retinal arterioles. J Am Med Assoc 1961; 178:23–9. 113 Archie JP, Feldtman JP. Critical stenosis of the internal carotid artery. Surgery 1981; 89:67–70. 114 Gunning AJ, Pickering GW, Robb-Smith AHT, Ross Russell RW. Mural thrombosis of the internal carotid artery and subsequent embolism. Q J Med 1964; 33:155–95. 115 Beal MF, Park TS, Fisher CM. Cerebral atheromatous embolism following carotid sinus pressure. Arch Neurol 1981; 38:310–12. 116 Watts C. External carotid artery embolus from the internal carotid artery ‘stump’ during angiography: case report. Stroke 1982; 13:515–17. 117 Barnett HJM. The pathophysiology of transient cerebral ischemic attacks: therapy with antiplatelet antiaggregants. Med Clin North Am 1979; 63:649–80.

118 Markus H. Transcranial Doppler detection of circulating cerebral emboli. A review. Stroke 1993; 24:1246–50. 119 Siebler M, Sitzer M, Rose G, Bendfeldt D, Steinmetz H. Silent cerebral embolism caused by neurologically symptomatic high-grade carotid stenosis. Event rates before and after carotid endarterectomy. Brain 1993; 116:1005–15. 120 van Zuilen EV, Moll FL, Vermeulen FE, Mauser HW, van Gijn J, Ackerstaff RG. Detection of cerebral microemboli by means of transcranial Doppler monitoring before and after carotid endarterectomy. Stroke 1995; 26:210–13. 121 Kirkes WS. On apoplexy in relation to chronic renal disease. Med Times Gaz 1855; 11:515–16. 122 Charcot JM, Bouchard C. Nouvelles recherches sur la pathogénie de l’hémorrhagie cérébrale. Arch Physiol norm pathol 1868; 1:110–27, 643–65, 725–34. 123 Iragui VJ. The Charcot-Bouchard controversy. Arch Neurol 1986; 43:290–5. 124 Satran R. Joseph Babinski in the competitive examination (agrégation) of 1892. Bull N Y Acad Med 1974; 50:626–35. 125 van Gijn J. The Babinski sign: the first hundred years. J Neurol 1996; 243:675–83. 126 Ellis AG. The pathogenesis of spontaneous intracerebral hemorrhage. Proc Pathol Soc Philadelphia 1909; 12:197–235. 127 Hiller F. Zirkulationsstörungen im Gehirn, eine klinische und pathologisch-anatomische Studie. Arch Psychiat Nervenkr 1935; 103:1–53. 128 Globus JH, Epstein JA, Green MA, Marks M. Focal cerebral hemorrhage experimentally induced. J Neuropathol Exp Neurol 1949; 8:113–16. 129 Rosenblath L. Über die Entstehung der Hirnblutung bei dem Schlaganfall. Dtsch Z Nervenkr 1918; 61:10–143. 130 Westphal K. Über die Entstehung und Behandlung der Apoplexia sanguinea. Dtsch med Wschr 1932; 58:685–90. 131 Ross Russell RW. Observations on intracerebral aneurysms. Brain 1963; 86:425–42. 132 Cole FM, Yates PO. The occurrence and significance of intracerebral micro-aneurysms. J Pathol Bacteriol 1967; 93:393–411. 133 Challa VL, Moody DM, Bell MA. The Charcot-Bouchard aneurysm controversy: impact of a new histologic technique. J Neuropathol Exp Neurol 1992; 51:264–71. 134 Fischer O. Die presbyophrene Demenz, deren anatomische Grundlage und klinische Abgrenzung. Z gesamte Neurol Psychiatr 1910; 3:371–471. 135 Scholz W. Studien zur Pathologie der Hirngefässe. II. Die drusige Entartung der Hirnarterien und -capillaren. Z Gesamte Neurol Psychiatr 1938; 162:694–715. 136 Pantelakis S. Un type particulier d’angiopathie sénile du système nerveux central: l’angiopathie congophile: topographie et fréquence. Monatsschr Psychiatr Neurol 1954; 128:219–56. 137 Torack RM. Congophilic angiopathy complicated by surgery and massive hemorrhage: a light and electron microscopic study. Am J Pathol 1975; 81:349–65. 138 Jellinger K. Cerebrovascular amyloidosis with cerebral hemorrhage. J Neurol 1977; 214:195–206.

..

..

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 33

References

139 Hounsfield GN. Computerised transverse axial scanning (tomography): I. Description of system. Br J Radiol 1973; 46:1016–22. 140 Hayward RD, O’Reilly GV. Intracerebral haemorrhage. Accuracy of computerised transverse axial scanning in predicting the underlying aetiology. Lancet 1976; 1:1–4. 141 McDonald CA, Korb M. Intracranial aneurysms. Arch Neurol Psych 1939; 42:298–328. 142 Blackall J. Observations on the Nature and Cure of Dropsies. 5th ed. London: Longman & Co., 1813. 143 Hodgson J. A Treatise on the Diseases of Arteries and Veins, containing the pathology and treatment of aneurisms and wounded arteries. London: T. Underwood, 1815. 144 Serres ERA. Observations sur la rupture des anévrysmes des artères du cerveau. Arch gén Méd 1826; 10:419–31. 145 Pyenson L, Sheets-Pyenson S. Reading: Books and the Spread of Ideas. Servants of nature: a history of scientific institutions, enterprises, and sensibilities. New York: W.W. Norton & Company, 1999: 211–35. 146 Berry D, Mackenzie C. Richard Bright (1789–1858): physician in an age of revolution and reform. London: Royal Society of Medicine Services Ltd., 1992. 147 Brinton W. Report on cases of cerebral aneurism. Trans Pathol Soc London 1852; 3:47–9. 148 Gull W. Cases of aneurism of the cerebral vessels. Guy’s Hosp Rep 1859; 5:281–304. 149 Lebert H. Über die Aneurysmen der Hirnarterien. Eine Abhandlung in Briefen an Herrn Geheimrat Professor Dr. Frerichs. Berl klin Wochenschr 1866; 3:209–405 (8 instalments). 150 Bartholow R. Aneurisms of the arteries at the base of the brain: their symptomatology, diagnosis and treatment. Am J Med Sci 1872; 44:373–86. 151 Eppinger H. Pathogenesis (Histogenesis und Aetiologie) der Aneurysmen einschliesslich des Aneurysma equi verminosum. Arch Klin Chir 1887; 35 (suppl. 1):1–563. 152 Wichern H. Klinische Beiträge zur Kenntnis der Hirnaneurysmen. Dtsch Zschr Nervenheilk 1912; 44:220–63. 153 Turnbull HM. Alterations in arterial structure, and their relation to syphilis. Q J Med 1914; 8:201–54. 154 Beadles CF. Aneurisms of the larger cerebral arteries. Brain 1907; 30:285–336. 155 Fearnsides EG. Intracranial aneurysms. Brain 1916; 39:224–96. 156 Hutchinson J. Aneurism of the internal carotid artery within the skull diagnosed eleven years before the patient’s death: spontaneous cure. Trans Clin Soc London 1875; 8:127–31. 157 Bull E. Akut Hjerneaneurisma-OkulomotoriusparalyseMeningealapoplexi. Norsk Magasin for Laegevidenskapen 1877; 7:890–5. 158 Quincke H. Die Lumbalpunktion des Hydrocephalus. Berl klin Wochenschr 1891; 28:929–33 and 965–8. 159 Froin G. Les hémorrhagies sous-arachnoidiennes et le méchanisme de l’hématolyse en général. Paris: G. Steinheil, 1904. 160 Guillain G. L’albuminurie massive dans le diagnostic des hémorrhagies méningées. Presse Méd 1915; 54:441–2.

..

..

161 Goldflam S. Beiträge zur Aetiologie und Symptomatologie der spontanen subarachnoidalen Blutungen. Dtsch Zschr Nervenheilk 1923; 76:158–82. 162 Symonds CP. Contributions to the clinical study of intracranial aneurysms. Guy’s Hosp Rep 1923; 73:139–58. 163 Symonds CP. Spontaneous subarachnoid haemorrhage. Quart J Med 1924; 18:93–122. 164 Cushing H. Contributions to the clinical study of cerebral aneurysms. Guy’s Hosp Rep 1923; 73:159–63. 165 Symonds CP. Autobiographical introduction. In: Symonds CP, editor. Studies in Neurology. London: Oxford University Press, 1970, pp. 1–23. 166 Moniz E. Anévrysme intra-cranien de la carotide interne droite rendu visible par l’artériographie cérébrale. Rev Oto-Neuro-Ophthal 1933; 11:198–203. 167 Schorstein J. Carotid ligation in saccular intracranial aneurysms. Br J Surg 1940; 28:50–70. 168 Todd NV, Howie JE, Miller JD. Norman Dott’s contribution to aneurysm surgery. J Neurol Neurosurg Psychiatry 1990; 53:455–8. 169 Rush C, Shaw JF. With Sharp Compassion: Norman Dott – freeman surgeon of Edinburgh. Aberdeen: Aberdeen University Press, 1990. 170 Dott N. Intracranial aneurysms: cerebral arterioradiography: surgical treatment. Trans Med Chir Soc Edinb 1932; 47:219–40. 171 Dandy WE. Intracranial aneurysm of internal carotid artery, cured by operation. Ann Surg 1938; 107:654–7. 172 Drake CG. Bleeding aneurysms of the basilar artery; direct surgical management in four cases. J Neurosurg 1961; 18:230–8. 173 Guglielmi G, Vinuela F, Sepetka I, Macellari V. Electrothrombosis of saccular aneurysms via endovascular approach. Part 1: Electrochemical basis, technique, and experimental results. J Neurosurg 1991; 75:1–7. 174 Guglielmi G, Vinuela F, Dion J, Duckwiler G. Electrothrombosis of saccular aneurysms via endovascular approach. Part 2: Preliminary clinical experience. J Neurosurg 1991; 75:8–14. 175 Matthews JR. Quantification and the Quest for Medical Certainty. Princeton: Princeton University Press, 1995. 176 Louis PCA. Recherches sur les effets de la saignée. Paris: de Mignaret, 1835. 177 Gavarret J. Principes généraux de statistique médicale. Paris: Librairies de la Faculté de Médecine de Paris, 1840. 178 Bernard C. Introduction à l’étude de la médicine expérimentale. Paris: J.-B. Baillière, 1865. 179 Lister J. Effect of the antiseptic system of treatment on the salubrity of a surgical hospital. Lancet 1870; i:4–6; 40–2. 180 Stolley PD, Lasky T. Investigating Disease Patterns: the science of epidemiology. New York: W.H. Freeman & Company, 1995. 181 Blackmore E. Reports on the diseases of Plymouth I. Edinburgh Medical and Surgical Journal 1829; 31:266–87. 182 Blackmore E. Reports on the diseases of Plymouth II. Edinburgh Medical and Surgical Journal 1829; 32:1–20. 183 Snow SJ. Operations Without Pain: the practice and science of anaesthesia in Victorian Britain. Houndmills, Basingstoke: Palgrave Macmillan, 2006.

33

9781405127660_4_002.qxd 10/13/07 11:14 AM Page 34

34

Chapter 2 Development of knowledge about cerebrovascular disease

184 Acheson J, Acheson HW, Tellwright JM. The incidence and pattern of cerebrovascular disease in general practice. J R Coll Gen Pract 1968; 16:428–36. 185 Kannel WB, Dawber TR, Cohen ME, McNamara PM. Vascular disease of the brain – epidemiological aspects: the Framingham study. Am J Public Health 1965; 55:1355–66. 186 Hirota Y, Katsuki S, Asano C. A multivariate analysis of risk factors for cerebrovascular disease in Hisayama, Kyushu Island, Japan. Behaviormetrika 1975; 2:1–11. 187 Salonen JT, Puska P, Mustaniemi H. Changes in morbidity and mortality during comprehensive community programme to control cardiovascular diseases during 1972–7 in North Karelia. Br Med J 1979; 2:1178–83. 188 Montgomery M. How Doctors Think: clinical judgment and the practice of medicine. New York: Oxford University Press, 2005. 189 Medical Research Council. Streptomycin treatment of pulmonary tuberculosis. Br Med J 1948; ii:769–82. 190 Fibiger J. Om Serumbehandling af Difteri. Hospitalstidende 1898; 6:309–25, 337–50. 191 Hróbjartsson A, Gøtzsche PC, Gluud C. The controlled clinical trial turns 100 years: Fibiger’s trial of serum treatment of diphtheria. Br Med J 1998; 317:1243–5. 192 Amberson JB, McMahon BT, Pinner M. A clinical trial of sanocrysin in pulmonary tuberculosis. Am Rev Tuberc 1931; 24:401–35. 193 Ferguson FR, Davey AFC, Topley WWC. The value of mixed vaccines in the prevention of the common cold. J Hyg 1927; 26:98–109. 194 Hinshaw HC, Feldman WH. Evaluation of chemotherapeutic agents in clinical tuberculosis. Am Rev Tuberc 1944; 50:202–13. 195 Fisher RA. The arrangement of field experiments. Journal of the Ministry of Agriculture 1926; 33:503–13. 196 Doll R. Controlled trials: the 1948 watershed. Br Med J 1998; 317:1217–20. 197 Chalmers I. Why transition from alternation to randomisation in clinical trials was made. Br Med J 1999; 319:1372. 198 Hill AB, Marshall J, Shaw DA. A controlled clinical trial of long-term anticoagulant therapy in cerebrovascular disease. Quart J Med 1960; 29:597–609. 199 Hill AB, Marshall J, Shaw DA. Cerebrovascular disease: a trial of long-term anticoagulant tharapy. Br Med J 1962; ii:1003–6. 200 Jonas S. Anticoagulant therapy in cerebrovascular disease: a review and meta-analysis. Stroke 1988; 19:1043–8. 201 Acheson J, Danta G, Hutchinson EC. Controlled trial of dipyridamole in cerebral vascular disease. Br Med J 1969; 1:614–15. 202 Dyken ML, White PT. Evaluation of cortisone in treatment of cerebral infarction. J Am Med Assoc 1956; 162:1531–4. 203 Fields WS, Maslenikov V, Meyer JS, Hass WK, Remington RD, Macdonald M. Joint study of extracranial arterial occlusion. V. Progress report of prognosis following surgery or nonsurgical treatment for transient ischemic attacks and cervical carotid artery lesions. J Am Med Assoc 1970; 211:1993–2003.

204 Canadian Cooperative Study Group. A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med 1978; 299:53–9. 205 Kurtzke JF. Controversy in neurology: the Canadian study on TIA and aspirin – a critique of the Canadian TIA study. Ann Neurol 1979; 5:597–9. 206 Flourens MJP. Recherches expérimentales sur les propriétés et les fonctions du système nerveux, dans les animaux vertébrés. Paris: Crevot, 1824. 207 Gall FJ, Spurzheim JC. Anatomie et physiologie du système nerveux en général, et du cerveau en particulier, avec des observations sur la possibilité de reconnaître plusieurs dispositions intellectuelles et morales de l’homme et des animaux, par la configurations de leurs têtes. Paris: Schoell, 1819. 208 Fritsch GT, Hitzig E. Ueber die elektrische Erregbarkeit des Grosshirns. Arch Anat Physiol wiss Med 1870; 37:300–32. 209 Hitzig E. Untersuchungen über das Gehirn. Berlin: A. Hirschwald, 1874. 210 Thorwald J. Das Weltreich der Chirurgen. Stuttgart: Steingrüben, 1957. 211 Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta blockade during and after myocardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis 1985; 27:335–71. 212 Counsell C, Warlow C, Sandercock P, Fraser H, van Gijn J. The Cochrane Collaboration Stroke Review Group. Meeting the need for systematic reviews in stroke care. Stroke 1995; 26:498–502. 213 Freiman JA, Chalmers TC, Smith H, Jr., Kuebler RR. The importance of beta, the type II error and sample size in the design and interpretation of the randomized control trial. Survey of 71 ‘negative’ trials. N Engl J Med 1978; 299:690–4. 214 Lewis JA, Ellis SH. A statistical appraisal of postinfarction betablocker trials. Prim Cardiol 1982; suppl. 1:317. 215 Lewis S, Clarke M. Forest plots: trying to see the wood and the trees. Br Med J 2001; 322:1479–80. 216 Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J 1988; 296:320–31. 217 Temkin O. The historiography of ideas in medicine. In: Clarke E, editor. Modern methods in the history of medicine. London: The Athlone Press, 1971, pp. 1–21. 218 Hippocrates. The Genuine Works of Hippocrates. Baltimore: Williams & Wilkins, 1939. 219 Rokitansky C. Handbuch der pathologischen Anatomie. Wien: Braumüller und Seidel, 1842. 220 Todd EM. The Neuroanatomy of Leonardo da Vinci. Park Ridge: American Association of Neurological Surgeons, 1991. 221 New PJF, Scott WR. Computed Tomography of the Brain and Orbit (EMI scanning). Baltimore: Williams & Wilkins, 1975. 222 Dandy WE. Intracranial Arterial Aneurysms. Ithaca, New York: Comstock Publishing Company, 1944. 223 Blakemore C. Mechanics of the Mind. Cambridge: Cambridge University Press, 1977. 224 Schwartz P. Arten der Schlaganfälle des Gehirns. Berlin: Julius Springer, 1930.

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 35

3

Is it a vascular event and where is the lesion? Identifying and interpreting the symptoms and signs of cerebrovascular disease

3.1 Introduction

35

3.2 Definitions of transient ischaemic attack, stroke and acute stroke syndrome (‘brain attack’ or ‘unstable brain ischaemia’) 35 3.3 The diagnosis of a cerebrovascular event

41

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset 3.5 Differential diagnosis of transient monocular blindness 3.6 Improving the reliability of the clinical diagnosis 3.7 Is it a subarachnoid haemorrhage?

3.1 Introduction

This chapter deals with the first two of several questions that need to be answered in the assessment of patients

81

103

108

111

presenting with a suspected transient ischaemic attack of the brain or eye (TIA) or stroke – is it a vascular event, and where is the lesion? (Table 3.1). We will begin by defining what is meant by a TIA and stroke, and follow by describing our approach to distinguishing TIA and stroke from other differential diagnoses.

Table 3.1 The assessment process in the diagnosis and subsequent management of a vascular event of the brain or eye. Is it a stroke, a transient ischaemic attack, or a brain attack? Which part of the brain has been affected? Which arterial territory has been affected? Is there a recognizable clinical syndrome? What pathological type of cerebrovascular event is it? What disease process caused the cerebrovascular event? What, if any, are the functional consequences? What treatment will improve survival free of handicap?

Chapter 3 Chapter 3 Chapter 4 Chapter 4 Chapter 5 Chapters 6–9 Chapters 10,11 Chapters 12–16

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

3.2 Definitions of transient ischaemic attack, stroke and acute stroke syndrome (‘brain attack’ or ‘unstable brain ischaemia’)

3.2.1 The definition of transient ischaemic attack A standard definition of a TIA is ‘a clinical syndrome characterized by an acute loss of focal cerebral or monocular function with symptoms lasting less than 24 h and which is thought to be due to inadequate cerebral or ocular blood supply as a result of low blood flow, thrombosis or embolism associated with disease of the arteries, heart or blood’.1,2 By definition therefore, TIAs are caused by transient reduction in blood flow to a part of the brain as a result of thromboembolic disease of the arteries, heart and blood. However, it is conceivable that TIAs may also arise from disease of the veins, causing transient reduction 35

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 36

36

Chapter 3 Is it a vascular event and where is the lesion?

in venous return from the brain or eye, but we are not aware of any patients with cerebral venous thrombosis who have presented with symptoms of TIA.3,4 The main features of TIAs are: • sudden onset; • symptoms of loss of focal neurological function (Table 3.2); • symptoms maximal at onset (more or less); • symptoms resolve within 24 h; • brain imaging (CT, MRI and DWI) may or may not show a relevant focal ischaemic lesion in the brain. Focal neurological symptoms Focal neurological symptoms are those that arise from a disturbance in an identifiable and localized area of the brain – for example, unilateral weakness from a lesion of the corticospinal tract (Table 3.2). However, there are some focal neurological symptoms which, when they occur in isolation, should probably not be considered as TIAs, because they occur more commonly in nonvascular conditions. Moreover, when present in isolation, Table 3.2 Focal neurological and ocular symptoms. Motor symptoms Weakness or clumsiness of one side of the body, in whole or in part (hemiparesis, monoparesis and sometimes only just the hand) Simultaneous bilateral weakness* Difficulty in swallowing* Imbalance* Speech/language disturbances Difficulty in understanding or expressing spoken language Difficulty in reading (dyslexia) or writing Difficulty in calculating Slurred speech* Sensory symptoms Altered feeling on one side of the body, in whole or in part Visual symptoms Loss of vision in one eye, in whole or in part Loss of vision in half or quarter of the visual field Bilateral blindness Double vision* Vestibular symptoms A sensation of movement* Behavioural/cognitive symptoms Difficulty in dressing, combing hair, cleaning teeth, geographical disorientation (visuospatial–perceptual dysfunction) Forgetfulness* *As an isolated symptom, this does not necessarily indicate focal cerebral ischaemia unless there is an appropriately sited acute infarct or haemorrhage, or there are additional definite focal symptoms.

they have not been shown to be associated with the high risk of future serious vascular events that characterizes TIAs. Such symptoms include rotational vertigo (caused by benign paroxysmal positional vertigo, vestibular neuritis, etc.), transient amnesia (transient global amnesia, psychogenic amnesia, etc.), and diplopia (myasthenia gravis, superior oblique myokymia, ocular myotonia, near-reflex accommodation spasm, etc.). Further research, based on imaging and long-term prognostic studies, of patients presenting with isolated focal neurological symptoms such as vertigo and dysarthria should help ascertain if, and how often, these symptoms are vascular in origin, and if so, what may help distinguish them from non-vascular causes (possibly older age and presence of other vascular risk factors). Some focal neurological symptoms should probably not be considered as transient ischaemic attacks if they occur in isolation (e.g. rotational vertigo, transient amnesia, and diplopia). Transient ischaemic attacks (and ischaemic stroke) are diagnosed with confidence only if the symptoms can be attributed to ischaemia of a focal brain area, and if there is no better explanation. Non-focal neurological symptoms Non-focal neurological symptoms (Table 3.3) are not usually caused by focal cerebral ischaemia; there are many other more common non-vascular neurological and non-neurological causes: • Generalized weakness caused by depression, acute inflammatory demyelinating polyradiculoneuropathy, myasthenia gravis, botulism, periodic paralysis, tick paralysis, acute toxic motor neuropathy, porphyric polyneuropathy, Miller Fisher syndrome and acute myelopathy; • Light-headedness caused by cardiac arrhythmias, aortic stenosis, cardiac tamponade, myocardial infarction, vasovagal syncope, reflex syncope and carotid sinus syncope; Table 3.3 Non-focal neurological symptoms. Generalized weakness and/or sensory disturbance Light-headedness Faintness ‘Blackouts’ with altered or loss of consciousness or fainting, with or without impaired vision in both eyes Incontinence of urine or faeces Confusion Ringing in the ears (tinnitus)

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 37

3.2 Definitions of transient ischaemic attack, stroke and acute stroke syndrome

(b) 60

Percentage of transient ischaemic attacks

Percentage of transient ischaemic attacks

(a)

50 40 30 20 10 0

720

45 40 35 30 25 20 15 10 5 0

720

Amaurosis fugax (n = 158) Transient ischaemic attacks of the brain (n = 311) Project;6 (b) 469 TIA patients in a hospital-referred series (from Hankey & Warlow, 19942 by kind permission of the authors and W.B. Saunders Co Ltd).

• Postural hypotension secondary to dehydration, blood pressure-lowering drugs (e.g. antihypertensives, dopamine agonists) and autonomic neuropathy; • Drop attacks (see below); • Confusion/delirium caused by metabolic/toxic encephalopathy. Therefore, non-focal symptoms should not be interpreted as being caused by a TIA or stroke unless accompanied by focal neurological symptoms. This is because – in isolation – they have other causes and do not seem to predict an increased risk of future serious vascular events.

patients with focal ischaemic neurological symptoms lasting longer than 24 h (ischaemic stroke) have similar causes and prognosis for future vascular events as patients with focal ischaemic neurological symptoms lasting less than 24 h (TIA).5 The duration of symptoms of TIA and ischaemic stroke are a continuum (Fig. 3.1);2,6 the only ‘relevance’ of longer duration TIAs and minor ischaemic stroke is that a relevant ischaemic lesion is more likely to be demonstrated by brain imaging (Fig. 3.2)7 and the risk of (recurrent) stroke increases.8–11 Indeed, increasing duration of focal neurological symptoms is one of the five main predictors of early recurrent stroke, along with Age, Blood pressure, Clinical features and Diabetes (and Duration of symptoms) in the ABCD2 prognostic model of early stroke risk after TIA10 (section 16.2.1). It has been suggested recently that the lower limit for the duration of symptoms should be 10 min.8,12 We would agree that TIAs lasting less than 10 min are associated with a lower risk of stroke than longer duration TIAs,10 and that focal neurological symptoms of the brain lasting seconds, and perhaps even a few minutes, are seldom likely to be TIAs of the brain (but TIAs of the eye – amaurosis fugax – may last only seconds). However, we remain uncertain just how short focal episodes can be and still be a TIA.

Neurological signs The standard definition of TIA allows abnormal but functionally unimportant focal neurological signs such as reflex asymmetry or an extensor plantar response to persist for longer than 24 h, provided the symptoms have resolved within 24 hours; this occurs in about 5% of patients.1,2

..

6–30

Fig. 3.1 Histogram showing the duration of the longest transient ischaemic attack (TIA) before presentation amongst (a) 184 TIA patients in the Oxfordshire Community Stroke

Non-focal symptoms such as faintness, dizziness or generalized weakness are seldom, if ever, likely to be due to focal cerebral ischaemia (i.e. seldom a transient ischaemic attack or stroke), but may be due to generalized brain ischaemia (e.g. syncope) as well as non-vascular neurological and non-neurological causes (e.g. hyperventilation or other manifestations of anxiety).

..

2–5

Duration of longest transient ischaemic attack (minutes)

It is not known how short a TIA can be (and still be a TIA).

Duration of symptoms

Brain imaging

The upper limit for the duration of symptoms (24 h) has more to do with the earth’s rotation than biology;

A substantial proportion of patients with cerebral TIA have evidence on brain imaging by CT or MRI of focal

37

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 38

Chapter 3 Is it a vascular event and where is the lesion?

Percentage of patients with infarct on CT

38

60

Transient ischaemic attack

49%

50 39%

40 30% 30

23%

20 11% 10 0

1–30 minutes

13% 8%

31–60 minutes

1–4 hours

5–24 hours

1–7 days

Duration of attack

areas of altered signal intensity, suggestive of ischaemia, in an area of the brain relevant to the transient symptoms.13–15 Diffusion-weighted MR imaging (DWI) is more sensitive than CT or MRI T2-weighted images, and identifies a relevant abnormality in up to two-thirds of patients with TIAs (Fig. 3.3).14,16–20 As a result some have proposed a revised definition of TIA which is restricted to patients who experience a brief episode of focal neurological dysfunction, presumptively caused by focal brain ischaemia, but without neuroimaging evidence of acute ischaemia or infarction. It has been further proposed that all other episodes with transient focal neurological symptoms with relevant lesion(s) (presumed infarction) on brain imaging are called ischaemic stroke.12,21–24 However, as stated in previous editions of this book, we do not accept such a definition, based on brain imaging, for the following reasons: • A CT or MR (including DWI) scan then becomes essential for the diagnosis of TIA, precluding up to one-fifth of patients who cannot undergo brain imaging (e.g. because they have an intraocular or intracerebral metallic foreign body or pacemaker, or are claustrophobic). Further, as the time to scanning decreases, and technology advances, the definition of a TIA would be constantly changing. • A positive CT, MRI or MRI-DWI may be a false positive (i.e. the ‘relevant’ abnormality on brain imaging may not represent recent infarction or ischaemia occurring at the time of the TIA (Table 3.4). Further, the limitations of DWI in diagnosing TIA cannot all be overcome (at present) by performing a complete stroke MRI examination, including T2-weighted imaging, angiography and perfusion imaging.25 • There is a gradual increase in the proportion of patients with a ‘relevant’ abnormality on brain imaging as the duration of symptoms increases, with no distinct change at 24 h7,12,18,23,24 (Fig. 3.2). This argues against there being any significantly different underlying pathophysiology between TIA and minor ischaemic stroke. There are also no significant

1–6 weeks

Persisting

Fig. 3.2 Histogram showing the relationship between the duration of focal neurological symptoms due to TIA and ischaemic stroke and the percentage of patients with an appropriately sited abnormality on brain imaging with CT (from Koudstaal, van Gijn, Frenken et al., 19927 by kind permission of the authors and Journal of Neurology, Neurosurgery and Psychiatry).

Table 3.4 Causes of an inaccurate diffusion-weighted MRI (DWI) diagnosis of TIA. False positive DWI diagnosis of TIA (at least 10%) Non-ischaemic cause of DWI hyperintensity Neurologically asymptomatic cause of DWI hyperintensity Chronic, persistent cause of DWI hyperintensity (i.e. old lesion) False negative DWI diagnosis of TIA (at least 10%) Very early imaging after onset of ischaemia (i.e. too early) Short-duration TIA (insufficient time to cause early ischaemic changes on DWI) Penumbral ischaemia (ischaemia is sufficient to cause neurological symptoms but insufficient to cause failure of the sodium-potassium ATPase pump in the neuronal cell membranes) Ischaemic region too small to image Ischaemic region too difficult to image (e.g. in brainstem) Confounding background lesion(s) Late imaging after onset of ischaemia (e.g. > 2 weeks after resolution of symptoms)

differences in the clinical features and prevalence of vascular risk factors among TIA patients with and without a relevant infarct on brain imaging.5 However, there may be a difference in prognosis if the results of initial studies showing the presence of such a ‘TIA scar’ on brain imaging is associated with an increased risk of future stroke are confirmed.11,15,26 • A new diagnostic category would have to be made for patients with a clinically definite stroke (i.e. symptoms for more than 24 h) but who have normal imaging, which could hardly be called a ‘TIA’. • There would be substantial implications of a tissuebased definition of TIA, in which patients with a TIA and a positive DWI would be reclassified as having a stroke, for epidemiological research and insurance policies.27 For example, studies of secular trends in stroke incidence and outcome would be confounded by the inclusion of patients with ‘TIA and brain imaging evidence of infarction’ as a ‘stroke’, whereas they

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 39

3.2 Definitions of transient ischaemic attack, stroke and acute stroke syndrome

were previously considered a TIA and not included as stroke in earlier studies. Further, patients with TIA and brain imaging evidence of infarction may now not be able to obtain life insurance, or their premiums may increase substantially.27 Critical illness insurance companies would even have to pay out for TIAs with DWI evidence of infarction that they would previously not have had to pay out for because the event would now be called a stroke; premiums would invariably rise to cover the new costs. At present there is no diagnostic test based on imaging or blood chemistry that is sufficiently sensitive, specific and widely available to reliably diagnose and exclude transient ischaemic attack and stroke.

(a)

We believe that patients with clinically definite TIAs (i.e. focal neurological symptoms lasting less than 24 hours with no other explanation other than a vascular origin), who have an appropriately sited and presumably ischaemic lesion on brain imaging, should have the fact noted (and considered to possibly be at increased risk of stroke).11,15,26 However, they should still be classified as having had a TIA, or maybe a new term for short duration symptoms of focal brain or ocular ischaemia (less than 24 h) with a relevant lesion on MRI scan (e.g. transient symptoms associated with infarction [TSI]),24 but definitely not a stroke. Our diagnostic criteria for TIA are summarized in Table 3.5. The presence of a presumed ischaemic lesion in the relevant part of the brain on CT or MR scan in a patient who presents with transient symptoms lasting less than 24 h should not change the diagnosis of a transient ischaemic attack (TIA) to a stroke.

(b) Fig. 3.3 (a) MRI (T2 weighted) of the brainstem (pons) in a patient with a recent episode of transient hemiparesis (right face, arm and leg) and conjugate gaze palsy to the left showing a very subtle region of altered signal intensity in the left paramedian pons. (b) MRI (DWI) in the same patient showing a very obvious area of high signal intensity (so-called ‘light bulb’ sign), consistent with early ischaemic change, in the left paramedian pons which was relevant to the patient’s focal neurological symptoms.

..

..

Although bright, high signal-intensity lesions (‘light bulbs’) on DWI and low apparent diffusion coefficient values have a high sensitivity (about 90%) and specificity (about 90%) for the diagnosis of acute focal cerebral ischaemia, the same findings have been reported in other diverse conditions such as focal brain haemorrhage, abscess and tumour (i.e. they are not 100% specific).

3.2.2 The definition of stroke The most widely accepted definition of a stroke is ‘a syndrome characterized by rapidly developing clinical symptoms and/or signs of focal, and at times global (applied to patients in deep coma and those with subarachnoid haemorrhage), loss of cerebral function, with symptoms lasting more than 24 h or leading to death,

39

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 40

40

Chapter 3 Is it a vascular event and where is the lesion? Table 3.5 Diagnostic criteria for transient ischaemic attack (TIA). Nature of symptoms Focal neurological or monocular symptoms Quality of symptoms ‘Negative’ symptoms, representing loss of focal neurological or monocular function (e.g. weakness, numbness, aphasia, loss of vision); rarely, ‘positive’ symptoms occur (e.g. pins and needles, limb shaking, scintillating visual field abnormality) Time course of symptoms Abrupt onset, starting in different parts of the body (e.g. face, arm, leg) at more or less the same time, without intensification or spread (‘march’); deficit maximal usually within a few seconds; symptoms resolve more gradually but completely, usually within an hour and, by definition, always within 24 h. Very brief attacks, lasting only seconds, are unusual except for transient monocular blindness (TMB) (note: we do not know how brief an attack of TMB can be and still be classified as TMB due to transient ischaemia; perhaps 10 s or so?) Associated symptoms TIAs usually occur without warning Antecedent symptoms are rare, but may reflect the cause (e.g. neck and face pain due to carotid dissection, headache due to giant-cell arteritis); otherwise, antecedent symptoms (e.g. headache, nausea or epigastric discomfort) usually suggest migraine or epilepsy Headache may occur during and after a TIA; it is to be distinguished from migraine headache Loss of consciousness is almost never due to a TIA; it usually suggests syncope or epilepsy Neurological signs Following symptomatic recovery, a few physical signs, such as reflex asymmetry or an extensor plantar response, which are not functionally significant, may be found Brain CT or MR scan The scan may show small areas of altered density, consistent with brain ischaemia or infarction, in a relevant part of the brain, or may have areas of hypodensity (on CT) or increased signal (on T2 weighted MR) remote from the symptomatic area Frequency of attacks TIAs often recur, but very frequent stereotyped attacks raise the possibility of partial epileptic seizures (sometimes due to an underlying structural abnormality such as an arteriovenous malformation, chronic subdural haematoma or cerebral tumour) or hypoglycaemia

with no apparent cause other than that of vascular origin’.28 This definition embraces stroke due to cerebral infarction (ischaemic stroke), non-traumatic intracerebral haemorrhage, intraventricular haemorrhage and some cases of subarachnoid haemorrhage (SAH). By convention, this definition does not include retinal

infarction, subdural haemorrhage, epidural haemorrhage, traumatic intracerebral haemorrhage or infarction, infection or tumour; nor does it embrace patients with intracranial venous thrombosis and SAH who are conscious and have a headache but no abnormal neurological signs. It is therefore important to be clear about what is, and what is not, included as ‘stroke’ in conversation, and in the literature. Since the clinical features, aetiology, prognosis and treatment of SAH are, for the most part, quite distinct from those of other forms of stroke,29,30 we think it is better to have a definition of stroke that does not include SAH, and so a completely separate definition for SAH. This is despite the fact that such a change in the definition of stroke would have implications for epidemiological studies of trends in incidence and outcome of stroke, just like changing the definition of TIA (see above), but in this instance would not be so dependent on the ever shifting sands of technology as changing the diagnosis of TIA to a tissue-based definition. Subarachnoid haemorrhage is therefore discussed separately below (section 3.7). The revised definition of stroke which we propose is ‘a clinical syndrome characterized by an acute loss of focal cerebral function with symptoms lasting more than 24 h or leading to death, and which is thought to be due to either spontaneous haemorrhage into the brain substance (haemorrhagic stroke) or inadequate cerebral blood supply to a part of the brain (ischaemic stroke) as a result of low blood flow, thrombosis or embolism associated with diseases of the blood vessels (arteries or veins), heart or blood’. Patients who are being assessed within 24 h of symptom onset and who still have focal neurological symptoms are temporarily classified as having a ‘brain attack’ (or something similar, such as an ‘acute stroke syndrome’ or ‘unstable brain ischaemia’). In practice, the absence of an obvious focal neurological deficit in a patient with a suspected stroke does not necessarily exclude the diagnosis. It may simply be a consequence of a delay in presentation, so that the signs have resolved, or it may be that the signs are rather subtle (but nevertheless functionally important to the patient) and have been missed. Examples include: • isolated dysphasia misinterpreted as confusion; • isolated visuospatial or perceptual disorder (e.g. dressing apraxia, geographical disorientation) which may not be noticed at the bedside examination, at least to begin with;31 • isolated amnesia or other subtle form of cognitive dysfunction;32

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 41

3.3 The diagnosis of a cerebrovascular event

• truncal and gait ataxia due to a cerebellar stroke which is only apparent (or elicitable) if the patient is asked to sit up, get out of bed and walk.33 These types of deficit may be missed (but should not be) in the emergency room, on busy ward rounds, during a hurried post-carotid endarterectomy or coronary artery bypass surgery assessment, and indeed any time.

3.2.3 The overlap between transient ischaemic attack and stroke, and the concept of an acute stroke syndrome (‘brain attack’ or ‘unstable brain ischaemia’) There is a continuum from TIA to ischaemic stroke in terms of duration of symptoms (see above).7 Moreover, patients with TIA and mild ischaemic stroke share a similar age and sex distribution, prevalence of vascular risk factors (and probably therefore pathogenesis) and long-term prognosis for serious vascular events although the short-term prognosis may differ.5 Thus, from the point of view of pathogenesis and treatment (secondary prevention), there seems no pressing need to distinguish TIA from ischaemic stroke, and indeed many trials of secondary prevention have included patients with TIA and non-disabling, mild ischaemic stroke because they are essentially the same condition. The problem in the era of increasingly rapid assessment and treatment of patients with acute cerebrovascular disease is how to use this time-based definition of TIA and stroke in patients who are being seen, and in some cases treated with potentially dangerous drugs (e.g. thrombolytics), within a few hours of the onset of symptoms. For example, if a hemiparetic patient is assessed 2·h after the onset of symptoms, an important question is whether this attack will recover and turn out to be a TIA, or not recover and become a stroke? There is no certain way of knowing, unless the patient is already recovering, but the longer the duration of symptoms of focal neurological dysfunction the more likely the deficit will persist,24 and the greater the risk of subsequent early stroke.9,10,34 Of course, for patients whose symptoms have resolved within 24 h of onset, they can be diagnosed retrospectively as having had a TIA. However, for those who still have symptoms within 24 h of onset, with or without relevant physical signs, it is appropriate to describe the acute presentation of focal cerebral ischaemia by a term such as a ‘brain attack’ or ‘acute stroke syndrome’ or ‘unstable brain ischaemia’.35 This emphasizes the need to rapidly exclude other differential diagnoses of TIA and stroke (e.g. hypoglycaemia, brain tumour), establish the pathological and aetiological subtype of the stroke and risk of recurrent stroke, and intervene with appropriate treatments that may

..

..

include reperfusion therapy for focal brain ischaemia, maintenance of physiological homeostasis, prevention of complications of stroke, early prevention of recurrent stroke and other serious vascular events, and early rehabilitation.36 ‘Time is brain’ in this setting. The anachronistic term ‘cerebrovascular accident (CVA)’ should be abandoned because it misleadingly implies that stroke is a chance event and that little can be done. Reasons to distinguish TIA and minor ischaemic stroke The are at least four reasons to distinguish TIA from minor ischaemic stroke. First, when formulating a differential diagnosis in clinical practice the differential diagnosis of focal neurological symptoms lasting minutes (e.g. epileptic seizures, migraine) is somewhat different from that of attacks lasting several hours to days (e.g. intracranial tumour, intracerebral haemorrhage), and in any event the reliability of the clinical diagnosis of stroke is much better than for TIA (section 3.6). Second, when conducting epidemiological studies of cerebrovascular disease consistency of diagnostic criteria is absolutely essential for comparing results over time and in different regions. Further, complete case ascertainment in incidence and prevalence studies is much less likely for TIA than stroke since patients who experience brief attacks are more likely to ignore or forget them, and are less likely to report them to a doctor than patients who suffer more prolonged or disabling events. Third, for case–control studies, there is less change in ‘acute phase’ blood factors related to thrombosis and tissue infarction, and there is, by definition, no survival bias amongst TIA patients compared with stroke patients. Fourth, distinguishing TIA from minor stroke can also aid assessment of case-mix in individual units and audits of management.

3.3 The diagnosis of a cerebrovascular event

The diagnosis of TIA and stroke is based on a constellation of clinical features that are thought to have a similar pathophysiology (i.e. caused by focal cerebral or ocular ischaemia or haemorrhage) (Table 3.2) and to be associated with similar outcomes (i.e. an increased risk of stroke and other major vascular events).

41

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 42

42

Chapter 3 Is it a vascular event and where is the lesion?

TIA and stroke are clinical diagnoses, based on a history of rapidly developing symptoms and signs of focal, and at times global, loss of focal brain function lasting 24 h respectively with no apparent cause other than that of vascular origin. The absence of a persistent neurological deficit does not exclude the diagnosis of TIA or stroke; such ‘negative’ findings may represent a delay in presentation, signs that have resolved, or subtle signs that have been missed (e.g. visual-spatialperceptual dysfunction). Similarly, the absence of a relevant acute lesion on brain imaging by CT, MRI or DWI does not exclude the diagnosis of TIA or stroke.

The assessment of patients with a suspected TIA or stroke depends on the time that has elapsed since the onset of symptoms. If the patient is assessed within 3– 6 h of stroke onset, the main focus is to establish the diagnosis of stroke, the pathological type and severity, and whether early reperfusion, or antiplatelet therapy and/or carotid endarterectomy may be indicated. If the patient is assessed (or reassessed) after this time, the focus is not toward reperfusion therapy but to ascertaining and minimizing the risk of recurrent stroke and the adverse sequelae and complications of the stroke. The timing of the assessment may also influence the reliability of the clinical assessment and accuracy of the diagnosis;37 focal neurological signs may resolve with time, which can make the diagnosis particularly difficult if the only signs were, for example, visual-spatialperceptual dysfunction. But it can also make the diagnosis easier if neurological signs such as drowsiness and dysphasia recover, so more history can be obtained from the patient. The first contact between clinician and patient is a crucial opportunity to conduct an appropriate history

Assessment History items Age > 45 years Absent history of seizure or epilepsy At baseline, not wheelchair bound or bedridden Blood glucose concentration between 2.8 and 22.2 mmol/L Physical assessment items Facial droop Arm drift Hand grip Speech Criteria for identifying stroke Presence of any physical assessment item All history items answered yes

FAST

LAPSS

and physical examination, and obtain relevant information from any observers, family, friends, the patient’s medical records and paramedical ambulance personnel (particularly when the patient is unable to communicate clearly due to dysphasia, depressed consciousness or knowledge only of a foreign language).38 The prior probability of a stroke among unscreened patients with neurologically relevant symptoms transported to an emergency department is about 10%.39 However, in some countries (e.g. Australia), paramedical ambulance personnel attend most patients admitted to hospital with stroke and they correctly identify about threequarters of stroke patients.40 However, because such personnel tend to overdiagnose stroke (not being aware of other conditions that mimic the symptoms of stroke) several prehospital screening tools have been developed, based on a few core clinical features (Tables 3.6, 3.7), to minimize the false positive diagnoses. These include: • the Face Arm Speech Test (FAST);41,42 • the Cincinnati Prehospital Stroke Scale (CPSS);43 • the Los Angeles Paramedic Stroke Scale (LAPSS);39 • the Melbourne Ambulance Stroke Screen (MASS).40 These tools have proved particularly helpful in rapid assessment of stroke patients ‘in the field’, and in communicating to regional stroke centres the imminent arrival of a patient with probable acute stroke.

History When a patient presents with a suspected transient ischaemic attack, ‘brain attack’ or stroke, the first question to answer is whether it really is a vascular event or not. This begins with, and depends on, a sound, carefully taken clinical history.

CPSS

X X X X

X X

X X X

X X

X X X X

X X X

X X

MASS

X

Table 3.6 Pre-hospital screening tools for the diagnosis of stroke: comparison of the Face Arm Speech Test (FAST), Los Angeles Paramedic Stroke Scale (LAPSS), Cincinnati Prehospital Stroke Scale (CPSS) and Melbourne Ambulance Stroke Screen (MASS).

X X X X X X

..

..

9781405127660_4_003.qxd 10/15/07 3:33 PM Page 43

3.3 The diagnosis of a cerebrovascular event Table 3.7 Motor and speech items assessed in several pre-hospital screening tools used for diagnosis of stroke.39–42 Assessment item Facial droop Patient smiles or shows teeth Arm drift Patient closes eyes and extends both arms for 10 s Hand grip Place a hand in each hand of the patient and ask him/her to squeeze hands Speech Patient repeats a sentence

Normal response

Abnormal response

Both sides move equally

One side does not move

Both arms move equally

One arm does not move or one arm drifts down, compared to the other

Both grip equally

Unilaterally weak or no grip

Normal language and articulation

Slurred or incorrect words, or unable to speak

When the patient is first assessed, take the patient and/ or eyewitness back to the onset of symptoms, recording their own words and not just your interpretation of them. This can usually be achieved by asking the three questions: • ‘When did it happen?’ • ‘Where were you when it happened?’ • ‘What were you doing when it happened?’ For clarification, it is always worth asking patients to describe their symptoms in an alternative way, particularly if the terms they use are rather vague, e.g. ‘dizziness’ or ‘heaviness’. Also it can sometimes be useful to ask patients whether they would have been able to do a specific task at the time of symptom onset; for example, if the patient describes an arm as being ‘dead’, asking whether they could lift the arm above their head would at least give a pointer as to whether the use of the word ‘dead’ was referring to a motor or just a sensory deficit. The use of certain terms is often culturally determined, and it must not be assumed that your interpretation of the term is the same as the patient’s. The most appropriate response if you are unsure is ‘what do you mean by that?’ or ‘try and describe what you mean in another way’. The history should obtain information about the following. • The nature of the symptoms and signs (sections 3.3.1– 3.3.7): – which modalities were/are involved (e.g. motor, sensory, visual)? – which anatomical areas were/are involved (e.g. face, arm, leg, and was it the whole or part of the limb; one or both eyes)? – were/are the symptoms focal or non-focal (Tables 3.2 and 3.3)? – what was/is their quality (i.e. ‘negative’, causing loss of sensory, motor or visual function; or ‘positive’, causing limb jerking, tingling, hallucinations)?

..

..

•

•

•

•

•

– what were/are the functional consequences (e.g. unable to stand, unable to lift arm)? The speed of onset and temporal course of the neurological symptoms (section 3.3.8): – what time of day did they begin? – was the onset sudden? – were the symptoms more or less maximal at onset; did they spread or progress in a stepwise, remitting, or progressive fashion over minutes/hours/days; or were there fluctuations between normal and abnormal function? Were there any possible precipitants (section 3.3.9)? – what was the patient doing at the time and immediately before the onset? Were there any accompanying symptoms (section 3.3.10), such as: – headache, epileptic seizures, panic and anxiety, vomiting, chest pain? Is there any relevant past or family history (section 3.3.11)? – have there been any previous TIAs or strokes? – is there a history of hypertension, hypercholesterolaemia, diabetes mellitus, angina, myocardial infarction, intermittent claudication, or arteritis? – is there a family history of vascular or thrombotic disorders? Are there any relevant lifestyle habits/behaviours (section 3.3.12)? – cigarette smoking, alcohol consumption, diet, physical activity, medications (especially the oral contraceptive pill, antithrombotic drugs, anticoagulants and recreational drugs such as amphetamines).

A record in the notes such as ‘no history available’ probably reflects laziness on the part of a doctor who has not tried fully to obtain it, rather than the real lack of any information.

43

9781405127660_4_003.qxd 10/15/07 3:33 PM Page 44

44

Chapter 3 Is it a vascular event and where is the lesion?

Examination The examination aims to: • confirm the presence of focal neurological signs, if any, anticipated from the history; • discover possible aetiological explanations for the event (e.g. atrial fibrillation, carotid bruits, cardiac murmurs, etc.), some of which may not be anticipated (e.g. malignant hypertension); • identify contraindications to investigation (e.g. a pacemaker – MR examination);

Symptoms

Neurological symptoms Acute onset Weakness Face Arm Leg Incoordination Limb Speech Visual disturbance Paraesthesia Face Arm Leg Vertigo Dizziness Nausea Vomiting Headache Confusion Loss of consciousness Convulsive seizures Neurological signs Weakness Face Arm Leg Visual field defect Eye movement abnormality* Dysphasia/dysarthria Visuospatial neglect Limb ataxia Hemiparetic/ataxic gait Sensory deficits Face Arm Leg

% of patients

• anticipate nursing and rehabilitation needs (e.g. impaired swallowing, urinary incontinence, immobility, pre-existing reduced visual or auditory acuity). A sophisticated knowledge of neurology is not needed to elicit and recognize the clinical features of a cerebrovascular event, as highlighted by the accurate paramedic and emergency room identification of stroke using simple stroke assessment tools (see above). However, a systematic approach is required as well as awareness of the discriminatory potential of symptoms and signs, in isolation and in groups (syndromes) (Table 3.8).44,45

Odds ratio

Stroke or TIA (n = 176)

Non-stroke (n = 167)

96

47

27.6

23 63 54

6 24 22

4.8 5.3 4.1

5 53 11

2 22 7

2.2 4.0 1.7

9 20 17 6 13 10 8 14 5 6 1

7 16 11 5 33 17 13 17 25 41 10

1.3 1.4 1.6 1.2 0.3 0.5 0.6 0.8 0.2 0.1 0.1

45 69 61 24 27 57 23 4 53

3 12 11 2 1 8 5 2 7

27.0 16.6 13.1 12.8 62.2 15.6 5.8 2.3 14.5

3 23 21

1 4 2

2.4 7.9 10.8

Table 3.8 Neurological symptoms and signs amongst patients presenting with suspected transient ischaemic attack and stroke in the Recognition of Stroke in the Emergency Room (ROSIER) study, subdivided into those occurring in patients with stroke or TIA and those with non-stroke.45

*Gaze palsy or opthalmoplegia. TIA, transient ischaemic attack.

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 45

3.3 The diagnosis of a cerebrovascular event

The National Institutes of Health Stroke Scale (NIHSS) is a graded neurological examination which can be used as a template upon which to base a systematic neurological examination. It assesses neurological impairments such as level of consciousness, ocular gaze, visual fields, speech and language function, inattention, motor and sensory impairments, and ataxia, and thereby helps ensure that a reasonably thorough neurological examination is undertaken in the acute phase. It is quick to perform (taking less than a few minutes), is valid, and is reliable among neurologists, non-neurologist physicians, and non-physician coordinators in clinical trials for measuring neurological impairment and stroke severity.46 However, the scale was originally developed for use in trials of treatments for patients who had already been diagnosed with stroke (to measure impairments and grade stroke severity); it was not designed for, and has limited application in, diagnosing stroke and differentiating stroke from its mimics. A sophisticated knowledge of neurology is not needed to elicit and recognize the clinical features of a cerebrovascular event, but physicians must continually make efforts to refine their clinical abilities if the symptoms and signs are to be documented accurately and the diagnosis of stroke and its localization are to be optimized.

3.3.1 The nature of the symptoms and signs The symptoms and signs of TIA and stroke reflect the areas of the brain that are affected by the focal ischaemia or haemorrhage.45 For short duration events, such as TIA, the symptoms also reflect the activities in which the patient was engaged during the attack. For example, if the patient was not speaking or did not try to speak or read during a brief ischaemic event, it is impossible to know whether aphasia or alexia were present or not. Similarly, a weak leg may well not be noticed if the patient was sitting down. As many hours of wakefulness are spent in an alert state with eyes open, an upright posture and often speaking or reading, it is not surprising that most of the symptoms that TIA patients experience are of motor, somatosensory, visual or language function (Table 3.9).6,47,48 Other more transient activities such as swallowing and calculation are, not surprisingly, less frequently reported. Presumably TIAs, like strokes, can start during sleep, but the patient will be unaware of them if they have resolved before waking. Among patients admitted to hospital with symptoms of brain attack (defined as ‘apparently focal brain dysfunction of apparently abrupt onset’), only about twothirds are subsequently diagnosed with stroke; one-third

..

..

Table 3.9 Neurological symptoms during transient ischaemic attacks. % Unilateral weakness, heaviness or clumsiness Unilateral sensory symptoms Slurred speech (dysarthria) Transient monocular blindness Difficulty in speaking (aphasia) Unsteadiness (ataxia) Dizziness (vertigo) Homonymous hemianopia Double vision (diplopia) Bilateral limb weakness Difficulty in swallowing (dysphagia) Crossed motor and sensory loss

50 35 23 18 18 12 5 5 5 4 1 1

From a series of 184 patients with a definite transient ischaemic attack (TIA) in the Oxfordshire Community Stroke Project.6 Many patients had more than one symptom (e.g. weakness as well as sensory loss) and no patients had isolated dysarthria, ataxia, vertigo, diplopia or dysphagia. Lone bilateral blindness was excluded from this analysis but later considered to be a TIA.48

are a stroke mimic.44 Clinical features that increase the odds of a final diagnosis of stroke include a definite history of focal neurological symptoms (odds ratio, OR = 7.2) and being able to determine the exact time of onset of symptoms (OR = 2.6) (Table 3.10).44 This is consistent with the common clinical criteria for a cerebrovascular event (i.e. abrupt onset of focal neurological symptoms or signs of a presumed vascular aetiology). A logistic regression model (Table 3.10) based on eight independent and significant predictors of the diagnosis of stroke (vs non-stroke) resulted in 83% correct classifications in the data set from which it was derived (i.e. it was internally valid).44 Other studies, using modern neuroimaging, have shown that the presence of acute facial weakness, arm drift and/or abnormal speech increased the likelihood of stroke, whereas the absence of all three decreased the odds.49 The Recognition of Stroke in the Emergency Room (ROSIER) scale has been developed and validated as an effective instrument to differentiate stroke from its mimics in the emergency room.45 It consists of seven items (total score from –2 to +5) comprising discriminating elements of the clinical history (loss of consciousness [score –1], convulsive seizures [score –1]) and neurological signs (face, arm, or leg weakness, speech disturbance, and visual field defect [each score +1]). A cut-off score >0 was associated with a sensitivity of 92%, specificity of 86%, positive predictive value of 88% and negative predictive value of 91% in the

45

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 46

46

Chapter 3 Is it a vascular event and where is the lesion?

Variable

OR

95% CI

Known cognitive impairment An exact onset could be determined Definite history of focal neurological symptoms Any abnormal vascular findings* Abnormal findings in any other system** NIHSS = 0*** NIHSS 1–4 NIHSS 5–10 NIHSS > 10 The signs could be lateralized to the left or right side of the brain OCSP classification possible

0.33 2.59 7.21 2.54 0.44

(0.1–0.8) (1.3–5.1) (2.5–20.9) (1.3–5.1) (0.2–0.8)

1.92 3.14 7.23 2.03

(0.7–5.2) (1.03–9.6) (2.2–24.0) (0.9–4.5)

5.09

(2.4–10.7)

Table 3.10 Logistic regression model for predicting the diagnosis of stroke.44

The model gives a predicted probability of stroke (ranging from 0 to 1). *Systolic blood pressure > 150 mm Hg, atrial fibrillation, valvular heart disease, or absent peripheral pulses. **Respiratory, abdominal or other abnormal signs. ***National Institute of Health Stroke Scale (NIHSS) = 0 was entered as the reference group (therefore it does not have a coefficient). OCSP, Oxfordshire Community Stroke Project; OR, odds ratio; CI, confidence interval.

derivation data set (Table 3.8), and similar values in the test data set.45 No one symptom or sign can rule in or rule out the diagnosis of stroke and transient ischaemic attack.

3.3.2 Disturbance of conscious level Consciousness may be defined as ‘the state of awareness of the self and the environment’. Coma is the total absence of such awareness. Vascular diseases are probably the second most common cause of non-traumatic coma after metabolic/toxic disorders; up to 20% of patients with stroke – but not TIAs – may have some impairment of consciousness.44,45,47 (See also section 11.3.) Clinical anatomy Consciousness depends on the proper functioning of the ascending reticular activating system (ARAS). This is a complex functional, rather than anatomical, grouping of neural structures in the paramedian tegmentum of the upper brainstem, the subthalamic region and the thalamus (mainly the intralaminar nuclei). Focal lesions that impair consciousness tend to either disrupt the ARAS directly (i.e. mainly infratentorial lesions), or are large supratentorial lesions, which cause secondary brainstem compression or distortion (Table 3.11; Fig. 3.4).

Table 3.11 Causes of impaired consciousness after stroke. Primary damage to subcortical structures (e.g. thalamus) or to the reticular activating system in the brainstem (e.g. brainstem haemorrhage) Secondary damage to the reticular activating system in the brainstem (e.g. large supratentorial haemorrhage or infarct with transtentorial herniation and midline shift due to oedema) Coexisting metabolic derangement (e.g. hypoglycaemia, hypoxia, renal or hepatic failure) Drugs (e.g. sedatives) To be distinguished from normal consciousness but with impaired responsiveness due to: Locked-in syndrome Akinetic mutism Abulia Severe extrapyramidal bradykinesia Severe depression Catatonia Hysterical conversion syndrome Paralysis from neuromuscular disorders

Clinical assessment Conscious state is assessed by observing the patient’s spontaneous activity and their response to verbal, painful and other stimuli. The Glasgow Coma Scale (GCS) provides a structured way of describing conscious level, and is usually part of the standard ambulance and nursing observation forms (Table 3.12). Because it was developed for patients with head injury and so for more

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 47

3.3 The diagnosis of a cerebrovascular event

Septum pellucidum Crus of fornix Thalamus Corpus callosum Aqueduct Interventricular foramen

Cerebellum

Hypothalamus Midbrain Pons Fig. 3.4 Diagrammatic representation in the sagittal plane of the brainstem areas involved in consciousness (especially the ascending reticular activating system). Table 3.12 The Glasgow Coma Scale. Eye opening E1 None E2 To painful stimuli E3 To command/voices E4 Spontaneously with blinking Motor response (best response in unaffected limb) M1 None M2 Arm extension to painful stimulus M3 Arm flexion to painful stimulus M4 Arm withdraws from painful stimulus M5 Hand localizes painful stimulus in the face (reaches at least chin level). M6 Obeys commands Verbal response V1 None V2 Sounds but no recognizable words V3 Inappropriate words/expletives V4 Confused speech V5 Normal The score should be reported as: Ex, My, Vz, total score =· x·+·y·+·z·/·15.

diffuse rather than focal neurological deficits, care is needed when applying it to patients with stroke. The motor deficit must be assessed on the ‘normal’ side and not on the side with the motor deficit, and in the arm, not the leg, where the motor responses may be largely of spinal origin. The subscore of each item is probably more important than the total, since specific focal deficits, and particularly global aphasia, depress the overall score disproportionately to the level of alertness. The GCS has value as an initial prognostic indicator in acute stroke.50,51 It may also have value in monitoring

..

..

Ascending reticular activating system

IVth ventricle Medulla

the patient’s neurological status over time. Any deterioration in the GCS is a prompt to consider whether it is because of the progression of the neurological deficit or because of non-vascular factors, such as infection, metabolic disturbance, or the effect of drugs (section 11.5). However, it is important not only to document the GCS scores (and over time), but to describe the patient’s neurological impairments qualitatively and quantitatively (and over time) because the GCS measures only three of many important functions of the brain and frequently there is an obvious change in the patient on conventional neurological examination but not in the GCS. The Glasgow Coma Scale is an insensitive measure of neurological function and should not be the only measure to monitor the neurological status of the patient.

Clinical practice Almost instantaneous loss of consciousness caused by a stroke suggests either a subarachnoid haemorrhage (section 3.7) or an intrinsic brainstem haemorrhage. Loss of consciousness within a few hours of onset is usually due to brainstem compression by a large intracerebral haematoma, or cerebellar haematoma or infarct. Early impairment of consciousness after supratentorial infarction is unusual. This is because the associated cerebral oedema responsible for the mass effect, and so the midline shift and brain herniation usually takes 1–3 days to develop, although some evidence of transtentorial herniation may be present within 24·h. Postmortem studies initially showed that infarction of the complete middle

47

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 48

48

Chapter 3 Is it a vascular event and where is the lesion?

cerebral artery territory was needed before significant lateral and caudal (inferior) displacement of midline structures occurs. These findings have been confirmed in several imaging studies during life. Not surprisingly, level of consciousness is one of the best predictors of survival after stroke (section 10.2.7). Loss of consciousness during a TIA is extremely unusual and should prompt a search for alternative explanations such as hypotension (e.g. vasovagal and reflex syncope, cardiac arrhythmia), systemic disorders (e.g. hypoglycaemia) and generalized epileptic seizures.52,53 Even when transient loss of consciousness is followed by focal neurological signs, such as a hemiparesis, it is more commonly caused by an epileptic seizure resulting in a Todd’s paresis52 (section 3.4.2). When loss of consciousness does occur during a TIA, it seems to be associated with brainstem or bihemispheric ischaemia caused by either vertebrobasilar or bilateral carotid occlusive disease, respectively.53,54 A few cases may be due to ischaemia in the territory of small, perforating arteries which supply the upper brainstem including components of the reticular activating system. Impairment of consciousness must be distinguished from impaired responsiveness due to the following. The ‘locked-in’ syndrome is a state of motor deefferentation, where there is usually severe paralysis not only of the limbs but also of the neck, jaw and face. Indeed, the only muscles remaining under voluntary control may be those concerned with vertical eye movements and blinking. All this occurs with clear, and often extremely distressing, retention of awareness. The patient is unable to communicate by word or movement other than by blinking or moving their eyes up and down, but is fully aware of the surroundings and attempts to respond to them. Hearing, vision and often sensation are retained. There is usually an extensive, bilateral lesion in the ventral pons, which interrupts the descending motor tracts as well as the centre for horizontal eye movements in the pons, but the oculomotor nuclei and descending pathways for vertical eye movements are spared together with the ascending reticular activating system (Fig. 3.5). Cognitive functions are normal and so the patients must be given a full explanation of their predicament. Staff may need to be reminded to take appropriate account of the patient’s normal cognition and sensation, since prolonged survival in this state is possible.55,56

The relatives and staff caring for patients with the locked-in syndrome need reminding regularly that sensation, cognitive functions and awareness are all normal.

Fig. 3.5 A T1-weighted magnetic resonance (MR) image, sagittal plane, showing a ventral pontine infarct (arrows) in a patient with the locked-in syndrome.

Akinetic mutism and abulia are states where there is limited responsiveness to the environment, although the patients appear alert (or at least wakeful) in that their eyes are open and they follow objects. However, in contrast to the locked-in syndrome, the physical examination does not reveal evidence of a major lesion of the descending motor pathways. At its most extreme, patients with akinetic mutism may lie with open eyes, follow objects and become agitated or even say the occasional appropriate word following noxious stimuli (thus distinguishing this state from that of coma or the persistent vegetative state); but otherwise they do not respond to their environment. Occasionally, catatonic posturing may occur. If patients recover from this state, they have no recollection of it. ‘Abulia’ describes a less severe presentation of reduced spontaneous movement and speech. Such patients often appear to have a marked flatness of affect, but with adequate stimulation they can be shown to be conscious and have relatively preserved cognition. Both akinetic mutism and abulia occur with bilateral damage to the cingulate gyri, caudate nuclei and anterior limb of the internal capsules, but they can also occur with unilateral lesions of the caudate nucleus. Although these states are most commonly seen after head injury, following anterior communicating artery aneurysmal subarachnoid haemorrhage, or in multi-infarct states, they can occur with unilateral occlusion of the recurrent artery of Heubner.

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 49

3.3 The diagnosis of a cerebrovascular event

3.3.3 Disturbance of higher cerebral function Higher cerebral functions can be divided into those which are ‘distributed’ – i.e. involve several areas of the cortex, such as attention, concentration, memory and higher-order social behaviour – and those which are more ‘localized’, such as speech and language, visuospatial function and praxis.32 Few tests are absolutely specific, however, for a single aspect of higher cerebral function. The nature of the cognitive assessment means that it is often appropriate to blend aspects of history taking with immediate confirmation by means of specific examination. Skilful examiners often weave their assessment into a relaxed conversation with the patients, making it more enjoyable for both. Many of the features of a brief cognitive assessment, an example of which is shown in Table 3.13,32 can be modified to suit this style of assessment. It is important to determine the handedness of an individual patient to guide which hemisphere is dominant for language. The descriptions below assume left-hemisphere dominance. Attention and concentration Attention and concentration are the ability to maintain a coherent stream of thought or action. They are not synonymous with wakefulness. clinical anatomy Attention and concentration are ‘distributed functions’ which depend on the integrated activity of the neocortex (predominantly the prefrontal, posterior parietal and ventral temporal lobes), the thalamus and brainstem. The reticular formation and other brainstem nuclei receive input from both ascending and descending pathways and there are then major ascending tracts to the thalamus, particularly its intralaminar nuclei. clinical assessment Failure of attention results in patients being unable to sustain concentration, and they are often reported to lack interest in things around them or to be tired or distractable. Another common complaint is that they have problems with memory. This may or may not be true, but from a practical point of view, if there is a significant disorder of attention, then extreme care is required when interpreting the results of tests of other higher-order functions such as memory. Attention and concentration can be assessed at the bedside (Table 3.14). clinical practice Acute stroke patients who appear inattentive should be assessed carefully to exclude an underlying focal

..

..

Table 3.13 Features of the 12-minute cognitive assessment.32 Orientation Time (day, date, month, season, year) Place Attention Serial 7s, or Months of the year backwards Language Engage in conversation and assess articulation, fluency, phonemic errors (e.g. ‘the grass is greed’ – Broca’s area lesion) and semantic errors (e.g. ‘the grass is blue’ – posterior perisylvian lesion) Naming of some low frequency items (e.g. stethoscope, nib, cufflinks, watch winder) Comprehension (of both single words and sentences) Repetition (e.g. emerald, aubergine, perimeter, hippopotamus; no ifs, ands or buts) Reading Writing Memory Anterograde: test recall of a name and address after 5 min Retrograde: ask about recent sporting or personal events Executive function Letter (F) and category fluency (animals): e.g. name as many words beginning with the letter F or A or S (>15 words per minute is normal) and animals (15 is low average, 10 is definitely impaired) Praxis Meaningful gestures (e.g. wave, salute) Luria three-step sequencing test (fist-edge-palm) Visuospatial Clock drawing, and overlapping pentagons General neurological assessment with particular attention to: Frontal lobe signs (grasp, pout, palmomental) Eye movements Presence of a movement disorder Pyramidal signs General impression Slowness of thought Inappropriateness Mood

Table 3.14 Bedside testing of attention and concentration. Digit span forwards and backwards* Recite months of the year, or days of the week, backwards Serial subtraction of 7s (although note that calculation ability needs to be intact) *The normal range is forwards: 6·± 1; backwards: 5 ± 1.

49

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 50

50

Chapter 3 Is it a vascular event and where is the lesion?

neurological disturbance such as aphasia, visuospatial or perceptual disturbance, hemianopia, or amnesia. If they really are inattentive they may have a metabolic/ toxic encephalopathy, and an underlying treatable cause (e.g. hyponatraemia, hypoglycaemia, hypoxia, uraemia, dehydration, sepsis). Memory Memory function can be divided into ‘explicit’ memory (available to conscious access) and ‘implicit’ memory (relates to learned responses and conditioned reflexes). Explicit memory may be ‘episodic’ (dealing with specific events and episodes that have been personally experienced) or ‘semantic’ (dealing with knowledge of facts, concepts and the meaning of words, e.g. ‘stroke is a clinical syndrome’). Episodic memory (personally experienced events) comprises anterograde (newly encountered information) and retrograde (past events) components.32 Working memory refers to the very limited capacity which allows us to retain information for a few seconds.

Table 3.15 Bedside testing of memory. First check that patient is attentive (Table 3.14) and that language function is adequate (see below, and Table 3.16). Anterograde verbal memory Ask the patient to name three distinct objects (e.g. ‘Ball, Flag, Tree’ or ‘Boston, Car, Daisy’) Ensure that the patient has registered the information (repeat up to three times if necessary) If the patient can immediately name the objects, ask the patient to repeat the three objects three minutes later Anterograde visual memory Show the patient faces in a magazine Ensure they have recognized them Retest after 5 min Retrograde memory Ask the patient to describe recent events on the ward, or visits from relatives Ask about important historical events and major events in the patient’s life, e.g. date of marriage

The terms ‘short-term memory’ and ‘long-term memory’ are used loosely by clinicians and often rather differently by neuropsychologists. In patients with stroke, in whom the time of onset is known, it may be easier to distinguish anterograde amnesia (failure to acquire new memories) from retrograde amnesia (failure to recall previously learnt material). clinical anatomy Episodic memory depends on the hippocampaldiencephalic system, semantic memory on the anterior temporal lobe, and working memory on the dorsolateral prefrontal cortex.32 clinical assessment Patients with stroke are generally of an age when there is a natural decline in memory anyway (or have coexistent Alzheimer’s disease or vascular cognitive impairment). Therefore, many complain of memory problems before the stroke, and furthermore it is important to determine whether there really is a disturbance of memory (and not aphasia or disturbance of attention and concentration resulting in failure of registration of new information) and, if so, to identify what is a direct result of the stroke. A suggested method for assessing memory is set out in Table 3.15. clinical practice Perhaps the most frequent stroke lesion causing amnesia is infarction of the medial temporal lobe (Fig. 3.6).

Fig. 3.6 MRI DWI showing area of high signal intensity, consistent with infarction, in the left medial temporal lobe (arrow) and cerebral peduncle (arrowhead), due to occlusion of the left posterior cerebral artery at its origin.

Because the cause is usually occlusion of the posterior cerebral artery or one of its branches, the patient may have a coexistent visual disorder (e.g. hemianopia or upper quadrantanopia, colour anomia, visual agnosia). A pure amnesic syndrome may be caused by a vascular lesion involving the mammillothalamic tract or anteriomedian territory of the thalamus (supplied by the polar

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 51

3.3 The diagnosis of a cerebrovascular event

Fig. 3.7 MRI DWI showing area of altered signal intensity, consistent with haemorrhagic infarction, in the left thalamus (arrow).

and paramedian arteries) such as the anterior parts of the dorsomedian nucleus57,58 (Fig. 3.7). In general, there is relative sparing of verbal memory with right thalamic lesions and visuospatial memory with left thalamic lesions, although the amnesia may be global with unilateral lesions. Particularly severe amnesia is more likely with paramedian thalamic infarction, which is frequently bilateral because the left and right paramedian arteries arise from one stem in many people (Fig. 3.8).58 In most cases of thalamic amnesia there are also signs of upper midbrain dysfunction, such as somnolence, vertical gaze palsies and corticospinal and spinothalamic tract signs. The syndrome of transient global amnesia is described in detail in section 3.4.3. Speech and language Language is difficult to define, but may be considered as a system for the expression of thoughts and feelings by the use of sounds and/or conventional symbols. It involves the production (or expression) and comprehension (or reception) of speech, as well as reading and writing. • Aphasia/dysphasia is an acquired disorder of the production and/or comprehension of spoken and/or written language. Subtypes of aphasia include speech (verbal) apraxia, alexia, agraphia and anomia.

..

..

Fig. 3.8 A T2-weighted MR image, axial plane, showing bilateral thalamic infarction (arrows) in a patient with severe global amnesia.

• Speech apraxia/dyspraxia is a syndrome in which there is variable misarticulation of single sounds, in the absence of dysarthria (in which there is constant misarticulation). This is because of impairment of planning of movements required for speech sounds and articulation, despite the ability to articulate speech being intact (e.g. similar to gait apraxia). • Alexia/dyslexia is an inability to name or interpret previously learned printed symbols. The patient can see individual letters, but cannot decode a series of letters into a recognizable word. When it occurs in isolation, this is sometimes referred to as ‘word blindness’. • Agraphia/dysgraphia is an acquired disorder of writing, a subtype of aphasia. • Anomia/dysnomia is an inability to generate a specific name. In the context of a vascular event, it is usually a manifestation of aphasia, but can be an amnesic disorder. • Anarthria/dysarthria is a disorder of articulation of single sounds. • Dysphonia is defined as a disorder of phonation of sounds.

51

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 52

52

Chapter 3 Is it a vascular event and where is the lesion?

Disorders of speech and language are not synonymous; speech involves language and articulation, whereas language involves reading and writing as well as speech.

clinical anatomy In general, non-fluent (expressive, Broca’s) aphasia is a syndrome with elements that range from initial mutism to speech apraxia and the classical pattern of agrammatism (inability to produce grammatical or intelligible speech, usually with simplified sentence structure – telegraphic speech – and errors in tense, number and gender). It is likely to be due to a lesion involving the posterior, inferior, dominant frontal lobe cortex and subcortex. The articulatory component (apraxia) is most heavily represented in the left insular region. Lesions of the dominant thalamus may also result in predominantly non-fluent aphasia. Fluent (receptive, Wernicke’s) aphasia is commonly caused by a more posterior lesion involving the temporal lobe cortex and subcortex. Most patients with stroke have a combination referred to as ‘mixed aphasia’ (or, if severe, ‘global aphasia’), due to more extensive lesions within the dominant hemisphere. Consequently, there is often an associated right hemiparesis and hemianopia. Occasionally, patients with non-fluent aphasia have preserved repetition. This is termed transcortical motor aphasia, and it is usually caused by lesions restricted to the anterior cerebral artery territory and which spare the arcuate fasciculus, between Broca’s area and Wernicke’s area (section 4.2.2). Fluent aphasia with normal repetition (transcortical sensory aphasia) occurs with strokes in the left temporo-occipital region. Dysarthria may be a result of cerebellar (ataxic), pyramidal (spastic), extrapyramidal (hypokinetic), or facial nerve (flaccid) dysfunction. Anarthria may occur as part of a pseudobulbar palsy caused by bilateral lesions of the internal capsule (not necessarily at the same time) or with a single lesion involving both sides of the brainstem. Alexia, with or without agraphia, may result from strokes that involve the medial aspect of the left occipital lobe and the splenium of the corpus callosum. There is usually a right visual field defect but no hemiparesis, and it is thought that the lesion in the splenium interrupts the transfer of visual information from the normal left visual field (right occipital lobe) to the damaged left hemisphere language areas. Gerstmann syndrome is the combination of aphasia, agraphia, right-left disorientation, and acalculia, due to a lesion in the region of the angular gyrus of the dominant hemisphere.59

clinical assessment The first distinction to make is between aphasia/ dysphasia (a disorder of language), anarthria/dysarthria (a disorder of articulation) and dysphonia (a disorder of phonation). If the patient’s speech sounds ‘like a drunk’, and if the ability to understand and express spoken and written language is preserved, then the problem is dysarthria (or dysphonia). If the main difficulty is understanding or expressing spoken or written language – such as difficulty in reading (the patient can see the letters but cannot make sense of them); difficulty in writing, even though the use of the hand is otherwise normal (often not the case); or difficulty in producing sentences, with words not being in their proper place or even non-words being used – then the problem is aphasia. Table 3.16 sets out a scheme of bedside testing that will detect most speech and language problems. There is a general tendency in everyday clinical practice to underestimate the receptive component of aphasia, particularly if the examiner does not go beyond questions requiring a yes/no answer, or simple social conversation. clinical practice Beware labelling a patient as dysphasic when a lack of other symptoms and signs suggest isolated non-dominant hemisphere dysfunction. ‘Crossed’ aphasia is a disturbance of language which occurs from a right hemisphere lesion in a right-hand dominant patient and is seen in about 4% of such patients. It is presumed that some right-handed patients have mixed cerebral dominance for language, but other causes include bilateral strokes (including the thalamus) and previous strokes. It is worth noting that many righthanded patients with right hemisphere strokes show subtle alterations in the affective aspects of speech, such as intonation (aprosody). Isolated dysarthria may be the only manifestation of a lacune at the genu of the internal capsule or in the corona radiata. In such cases, there is specific impairment of corticolingual fibres.60 Foreign accent syndrome is a rare, acquired disorder of speech in which native speakers listening to a patient speaking their language describe hearing a foreignsounding accent – yet the patient may never have been exposed to any other language or dialect before the stroke. It is probably due to an inability to make the normal phonetic and phonemic contrasts of the native language. The syndrome has most often been associated with small, subcortical infarcts in the left cerebral hemisphere.

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 53

3.3 The diagnosis of a cerebrovascular event Table 3.16 Bedside testing of language function. First ensure any hearing aid has a battery, is switched on and that appropriate, clean spectacles are worn Also check that you are using the patient’s native language – if not, use an interpreter Spontaneous speech Consider output (whether fluent or non-fluent), articulation and content: during history-taking and for a structured task (e.g. ‘describe your surroundings’) Auditory comprehension Simple yes/no questions (e.g. Is Russia the capital of Moscow? Can dogs fly? Do you put your shoes on before your socks?) Give commands (being careful not to use non-verbal cues) of one, two and three steps using common objects, such as the manipulation of three different-coloured pens (care not to require the use of limbs with significant weakness or apraxia) Naming Ask the patient to name objects, parts of objects, colours, body parts, famous faces (certain groups, particularly the naming of people, may be more severely affected) If visual agnosia is present, use auditory/tactile presentation, e.g. bunch of keys Repetition ‘West Register Street’ (difficult if dysarthric) ‘No ifs, ands or buts’ (difficult if aphasic) Reading Aloud, e.g. from a book or newspaper Comprehension of the same piece Writing Spontaneous (‘why have you come into hospital?’) Dictation (‘the quick brown fox jumped over the lazy black dog’) Copying Articulation Ask the patient to say: p/p/p/p/p/p (labial sounds, which test the orbicularis oris) t/t/t/t/t/t (lingual sounds, which test the anterior tongue) k/k/k/k/k/k (palatal sounds, which test the posterior tongue and palate) p/t/k/p/t/k (tests the overall coordination of sounds)

Visuospatial dysfunction Many patients with stroke fail to respond to stimulation of, or to report information from, the side contralateral to the cerebral lesion. There are two broad categories of neglect: intrapersonal (i.e. with respect to the patient’s own body) and extrapersonal or topographical (i.e. with respect to the surrounding environment). There are a number of different types and/or degrees of severity of neglect in patients with stroke, and in many, a combination of somatic sensory deficits and disturbed visual perception contribute to the clinically apparent ‘neglect’,

..

..

Table 3.17 Glossary of terms describing disorders of visuospatial function. Hemi-inattention: where the patient’s behaviour during examination suggests an inability to respond appropriately to environmental stimuli on one side, e.g. people approaching, noise, or activity in the ward Sensory or tactile extinction: where the patient fails to register a tactile stimulus (light touch) of adequate intensity on one side of their body when both sides are stimulated simultaneously and adequately (i.e. double simultaneous stimulation) but where the stimulus has been registered when each side was stimulated separately Visual inattention or extinction: where the patient fails to register a visual stimulus (e.g. finger movement) in one homonymous visual field (half field or quadrant) when the same stimulus is presented to both fields simultaneously, but where the patient had no field defect on normal testing Allaesthesia: where the patient consistently attributes sensory stimulation on one side to stimulation of the other; this is related to right/left confusion, where the patient consistently moves the limbs on one side when requested to move the limbs on the other Anosognosia: denial of a sensorimotor hemisyndrome Anosodiaphoria: indifference to/unconcern about a sensorimotor hemisyndrome Asomatognosia: lack of awareness of a body part Somatoparaphrenia (non-belonging): lack of ownership of a paralysed limb Experience of supernumary phantom limbs: reduplication of limbs on the affected side of the body Personification: nicknaming a limb and giving it an identity of its own Misoplegia: the morbid dislike or hatred of paralysed limbs in patients with hemiparesis61 Related phenomena seen in parietal lobe dysfunction: Astereognosis: unable to recognize objects placed in the affected hand yet cutaneous sensation is preserved Agraphaesthesia: unable to identify a number drawn on the palm of the affected hand yet cutaneous sensation is preserved Geographical disorientation: where the patient becomes lost in familiar surroundings despite being able to see Dressing apraxia: unable to dress, or dresses inappropriately, despite having no apparent weakness, sensory loss, visual or neglect problems; this is occasionally seen in an isolated form and probably occurs because of a combination of disordered body image, sensory and visual inattention rather than being a true apraxia

hence the use of the broader term ‘visuospatial dysfunction’. Table 3.17 provides a glossary of the terms that are used,61 Fig. 3.9 shows an example of anosognosia; this patient denies the presence of weakness of the left leg62

53

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 54

54

Chapter 3 Is it a vascular event and where is the lesion?

Fig. 3.9 Neglect and anosognosia: the photograph on the left shows how someone may normally view their hands and legs (before a stroke). The photograph on the right shows how the

same person may view their hands and legs after a major right hemisphere stroke, causing neglect and anosognosia for the left hand and left leg.

Fig. 3.10 Somatoparaphrenia (non-belonging): the photograph on the left shows how someone may normally view their hands and legs (before a stroke). The photograph on the right shows how the same person may view their hands and legs after a major right hemisphere stroke causing

somatoparaphrenia (i.e. the man denies ownership of the paralysed arm and leg on the left side of the body, and even thinks the left arm and leg belong to another person, such as his wife – note the different left hand, wedding ring on the left hand, and the different left leg).

and Fig. 3.10 shows an example of somatoparaphrenia (non-belonging); this patient denies ownership of the paralysed leg on the left side of the body, and even attributes the left leg to another person.

artery occlusion.63 Although it can occur with dominant hemisphere lesions, when it does so detection is often hindered by coexistent language disturbance and inability to use the dominant hand.

clinical anatomy Visuospatial dysfunction is most severe with posterior parietal lesions of the non-dominant hemisphere, particularly those that extend to the visual association areas. Among patients with ischaemic stroke, the aetiology may be middle cerebral or posterior cerebral

clinical assessment Relatives may report little more than ‘confusion’ or ‘difficulty in dressing’. If visuospatial problems are suspected from the history, they should be carefully sought in the examination. Simply observing how patients respond to their environment and carry out tasks, such

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 55

3.3 The diagnosis of a cerebrovascular event

Fig. 3.11 A letter from a patient with left visual neglect, showing neglect of the left side of the page.

as writing (Fig. 3.11), copying flowers (Fig. 3.12) and drawing a clock face (Fig. 3.13) can be revealing. An obvious example would be if a patient (without a hemianopia) does not register the doctor’s presence when approached from one side, even when spoken to. Or they might be unable to find their way back to their hospital bed after being taken to the toilet, suggesting geographical disorientation. The nurses and therapists are often better placed than the doctor to identify visuospatial problems, so it is important that staff are trained to recognize and report them to other members of the team. Table 3.18 sets out a bedside examination that should detect significant visuospatial dysfunction. Of the many cancellation tasks available, the star cancellation test is easy to use and probably the most sensitive (Fig. 3.14). Using two or three different tests increases the sensitivity of detecting visual neglect, but this may not always be practical in the setting of acute stroke.64 Many aspects of these assessments of visuospatial function require subjective judgements to be made by the physician – which probably accounts for the relatively poor inter-observer reliability. Although many other tests to identify and quantify visuospatial dysfunction have been described, the ‘gold standard’ against which the tests are evaluated is often regarded as the functional assessment by an occupational therapist.

..

..

Fig. 3.12 Abnormalities of copying flowers. Five patients with right hemisphere lesions were asked to copy pictures (a) and (b) in panel 1. The figures illustrate the variation seen in copying tasks. Panel 2: this patient has mainly neglected the information on the left side of the page. Panel 3: this patient has omitted the left-hand components of the objects, but has shifted attention to the right-hand side of another object placed to the left of the neglected space. Panel 4: this patient has drawn the right side of both flowers in the pot, but has completely neglected the left of the two separate flowers. Panel 5: this patient has transposed objects in the left field to the right field, i.e. both flowers are drawn on a single stem. Panel 6: this patient has produced a ‘hallucinatory’ rabbit in the left field when copying the separate flowers; this has been termed ‘metamorphopsia’. Table 3.18 Bedside tests of visuospatial function. Is the patient aware, and reacting appropriately to their deficit? Observe the patient’s response to the environment Observe the patient’s ability to carry out a specific task Check for sensory and visual extinction Copy a simple picture, e.g. a flower (Fig. 3.12) Draw a clock face and put the numbers in (Fig. 3.13); this may not be specific for visual neglect but rather reflect other cognitive problems, such as dementia Perform the star cancellation test (Fig. 3.14)

55

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 56

56

Chapter 3 Is it a vascular event and where is the lesion?

Fig. 3.13 (a) A drawing of a clock face by a patient with left visuospatial disturbance after a stroke, showing crowding of the digits on the right and neglect of the left side of the clock face. (b) A drawing of a clock face by a confused, elderly patient who has not had a stroke. The crowding of digits on the right occurs

G

TEN GET

DAY

E J

C

N

E

HER

L

K

MAN

E

READ

N O

U

STAR M

LEG

LEG

ARE Fig. 3.14 Star cancellation test. The chart is placed in front of the patient, who is asked to cross out all the small stars while ignoring the large stars and letters (with permission from Thames Valley Test Company, 7–9 The Green, Flempton, Bury St Edmunds, Suffolk IP28 6EL, UK).

An elderly man who lived alone reported that he woke one morning and thought that there was ‘something in bed with me’. He said it felt warm, and was pressing against the left side of his body. He thought his cat had got into bed with him, but when he touched it with his right hand he realized that it was his left arm. He had had a right parietal infarct during the night. One can often deduce the presence of visuospatial problems by simply observing how the patient responds to the environment and carries out tasks around the ward.

because of failure of planning. Many normal people insert 12, 3, 6 and 9 before other numbers. (c) Rather bizarre drawing of a clock face by a patient with strokes affecting both cerebral hemispheres.

clinical practice When there is an isolated problem with visuospatial function – i.e. when it is not accompanied by a more easily recognized ‘stroke’ deficit such as weakness – the patient’s behaviour may seem extremely bizarre, and even be interpreted as psychiatric disease. An elderly man was escorted to hospital by his concerned passenger after he drove his car along a street unaware that he was scraping along a whole row of other cars on the left-hand side. Examination revealed normal visual fields (when each eye was tested in turn) but visual inattention to the left (when the visual fields of each eye were tested simultaneously). There was also somatosensory extinction on the left when the arms were touched simultaneously. Brain CT scan showed a small right parietal haemorrhage. A middle-aged single lady was flying home from holiday when she became ‘confused’. On disembarking from the aircraft, she was staggering to the left, appeared unable to follow the signs to the customs point, and could not find her passport in her left-hand jacket pocket. She was held initially by the police on suspicion of alcohol or drug intoxication, but then admitted to a psychiatric hospital. It was only a week later, when she had a transient ischaemic attack affecting power in her left hand, that a right parietal infarct and severe stenosis of the right internal carotid artery were discovered. Perhaps not surprisingly, there is some evidence that patients who have varying degrees of indifference to their stroke are more likely to delay seeking medical

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 57

3.3 The diagnosis of a cerebrovascular event

attention.31,33 Visuospatial problems are a major cause of disability and handicap, and impede the patient’s functional recovery (section 11.28).

d b

Disorders of praxis

c

Apraxia is defined as inability to perform learned movements that cannot be explained by weakness, sensory loss, incoordination, inattention and other perceptual disorders, or by failure to understand the command. Dressing and constructional apraxias are best considered as disorders of visuospatial function rather than true apraxias (see above). Although the terms ‘verbal apraxia’ or ‘speech apraxia’ may be used by speech and language therapists when there are repeated phonemic substitutions, in practice such patients usually also have evidence of aphasia and/or dysarthria. clinical anatomy It is thought that the programmes of learned movements (engrams) are maintained predominantly in the left temporoparietal cortex. Messages then pass to the left premotor frontal cortex and finally, via the anterior corpus callosum, to the right premotor frontal cortex (Fig. 3.15). Lesions in the posterior left hemisphere may result in bilateral apraxia (because the message is not transmitted); those in the premotor areas are usually associated with a hemiparesis and therefore apraxia may only be apparent in the non-paralysed limbs; and finally, a lesion of the corpus callosum may cause isolated apraxia of the left limbs, with normal function of the right limbs. clinical assessment Apraxia should always be considered as a potential explanation for disparity between the degree of deficit as tested at the bedside (when one often gives the patient relatively simple commands) and much more severely impaired functional abilities when the patient is observed around the ward (e.g. dressing, swallowing, speaking). Patients have difficulty with miming actions, imitating how an object is used, and even making symbolic gestures. However, at other times they may be observed making the individual movements that would be needed to perform the action. In general, they will have most difficulty in miming the action, less difficulty in imitating the examiner, and least difficulty when actually given the object to use. The more sequences there are to the action, the more difficult it is and so the more sensitive the test. These relatively common problems are sometimes referred to as ‘ideomotor apraxias’, and can be distinguished from ideational apraxias when the patient has difficulty in performing a sequence of

..

..

a

Right

Left

Fig. 3.15 Diagrammatic representation of the lesions that can result in apraxia. (a) The dominant temporoparietal cortex is probably the site of the programmes of learned movements (engrams). Lesions here result in bilateral apraxias, due to failure to transmit the information to both frontal lobes. The clinical signs may be difficult to identify because the interpretation of the command may be affected by receptive aphasia. (b) Although lesions of the dominant frontal lobe are often associated with expressive dysphasia, comprehension is usually relatively spared. The apraxia may only be apparent in the left limbs, since there will usually be a right hemiparesis. (c) Lesions in the anterior corpus callosum may result in isolated apraxia of the left limbs, because of failure of transmission of the motor information to the right frontal lobe, while the right arm and leg move normally. (d) Lesions of the non-dominant frontal lobe are not normally associated with clinically apparent apraxias, because there is usually a left hemiparesis.

movements even though the individual movements can be performed normally. However, the latter probably occurs very rarely in a pure form, and the distinction between ideomotor and ideational apraxias is of little value to clinicians. Table 3.19 suggests ways of screening for apraxia. clinical practice In patients with stroke, the main problem is being sure that the patient has understood the command, because lesions of the relevant areas will often result in aphasia. Nevertheless, 80% of patients with aphasia also have evidence of apraxia with an imitation test (i.e. no verbal command). Because apraxic patients may perform actions reflexly that they are unable to do when asked, this should not be misinterpreted as a sign of a hysterical conversion disorder.

57

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 58

Chapter 3 Is it a vascular event and where is the lesion? Table 3.19 Bedside testing of praxis. Limbs Ask the patient to: Mime the use of a pen, comb and toothbrush Imitate the examiner’s use of the same objects Use the actual objects Orofacial Ask the patient to: Whistle Put tongue out Blow out cheeks Cough Serial actions Ask the patient to: Mime putting the address on a letter Then seal it Then put a stamp on it

3.3.4 Disturbance of the motor system Clinical anatomy Particular areas of the motor cortex, when stimulated, result in movement of a particular body part. This localization of function is traditionally portrayed by the homunculus (or manikin).65 Although the absolute neuroanatomical relationships may be incorrect, it

Ha nd Jaw T Sw ongue allo wi ng

Vocalization

Lips

Sal ivat ion

s kle An

e dl x id M Inde b um Th eck N w Bro ll a yeb e and e d i l Fac Eye

ca tio n

Wrist

Elbow

Shoulder Trunk Hip ee Kn

To es

is still useful as an aide-memoire (Fig. 3.16). The corticospinal tract (Fig. 3.17) descends from the primary and more anterior supplementary motor cortex, the fibres converging in the corona radiata. The fibres then pass through the internal capsule. The traditional view is that those relating to the head pass through the anterior limb; those relating to the mouth, larynx and pharynx are in the genu; those relating to the arm are in the anterior part of the posterior limb; while those relating to the leg lie more posteriorly. In fact, the fibres almost certainly follow an oblique course through the capsule, becoming progressively more posteriorly placed in the caudal (inferior) segments of the capsule. The fibres then pass into the brainstem. Here, the fibres that originate in the precentral gyrus lie in the cerebral peduncles of the midbrain and the base of the pons before entering the medullary pyramids (i.e. the pyramidal tracts). The facial nerve nucleus in the pons has a rostral portion from which fibres innervate the muscles of the upper face, while the more caudal portion of the nucleus supplies fibres to the muscles of the lower face. The caudal loop of the fibres to the facial nerve descends as far as the medulla and explains why lesions of the medullary pyramid or medial medulla can be associated with contralateral upper motor neurone-type facial weakness.66,67 Most fibres in the corticospinal tract decussate in the lower medulla and come to lie in an anterolateral

Lit t Ri le ng

58

ti as M

Fig. 3.16 Topographic organization of the motor cortex in the cerebral hemisphere, coronal view (after Penfield & Rasmussen, 195065).

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 59

3.3 The diagnosis of a cerebrovascular event

Hand

Patients use terms such as ‘heaviness’ and ‘numbness’ interchangeably; further questioning is needed to distinguish between motor and sensory deficits.

Leg Face

Internal capsule

Corticospinal tract Corticobulbar tract Midbrain Cerebral peduncle

Pons

Pyramid

Spinal cord

Medulla oblongata

Lateral corticospinal tract (crossed) Anterior corticospinal tract (uncrossed)

Fig. 3.17 Diagrammatic representation of the corticospinal and corticobulbar tracts.

position in the spinal cord, although a variable proportion remain uncrossed. These uncrossed fibres project to motor neurones in the medial part of the ventral horns, subserving axial and proximal muscles, corresponding with movements of the trunk, or of two limbs together. The uncrossed corticospinal fibres cannot be invoked to explain residual function in distal parts of an otherwise plegic limb, nor deficits in contralateral limbs. Clinical assessment Motor symptoms are usually described as ‘weakness’, ‘heaviness’ and ‘clumsiness’. They are often accompanied by sensory symptoms of some sort, which can lead to diagnostic confusion because a purely weak limb may be described by the patient as ‘numb’ or ‘dead’. Descriptions such as ‘heaviness’ and ‘numbness’ should not simply be accepted as evidence of motor and sensory disturbance, respectively. In our experience, the terms are used interchangeably (and are often culturally determined), and a little more questioning is often required.

..

..

In patients with suspected TIA, unilateral facial weakness is probably under-reported, because they do not realize they have had facial weakness unless they have seen themselves in the mirror, or were seen by someone else. If there is a clear history of slurred speech, but no symptoms of cerebellar or bulbar dysfunction, it is reasonable to suspect facial weakness, because this may cause dysarthria. However, care should be taken before accepting a patient’s or relative’s description of the side of a facial weakness. They should be asked, ‘Which side dropped?’ and ‘Did saliva trickle from one side of the mouth?’ Upper motor neurone facial weakness affects the lower half of the face, while function of the forehead muscles is relatively preserved because of bilateral innervation of the forehead muscles. Mild weakness may only be apparent by observing asymmetry of the nasolabial folds. If the examiner is uncertain whether a facial weakness is present, or whether it is simply the normal sideto-side asymmetry, it may be useful to ask the patient to attempt to whistle, an action that requires fine control of the facial muscles. A mild upper motor neurone facial weakness can be overcome during emotionally generated movements, e.g. a smile. Patients sometimes complain of being ‘generally weak’. This should be viewed as a non-focal neurological symptom, as it is rarely described when they strictly mean motor weakness. It is sometimes used as a term for fatigue, tiredness, lethargy and just occasionally loss of balance. For the clinician, the difficulties with the physical examination lie not with the densely hemiplegic patient with increased tone, brisk deep tendon reflexes and an extensor plantar response, but rather with the patient with a mild neurological deficit. Subtle abnormalities of motor function may be detectable in the hand at a time when there is no objective weakness. Impairment of fine finger movements (or rapid alternating hand movements) is a sensitive clinical test of corticospinal function. This equates with functional problems reported by patients who, in the presence of normal power, often have difficulty with delicate motor tasks such as doing up buttons or controlling a pen and so may describe the problem as ‘clumsiness’. Of course, they are much more likely to notice this in their normally dominant hand. Impairment of fine finger or rapid hand movements is probably the most sensitive clinical test of corticospinal function.

59

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 60

60

Chapter 3 Is it a vascular event and where is the lesion?

Drift from the horizontal of the outstretched arm with the eyes closed, and rapid tapping of the foot against the examiner’s hand, are both good screening tests of motor function in the arm and leg, respectively, but neither test is very specific.

Fig. 3.18 Downward drift of the outstretched left arm in a patient with a right corticospinal tract motor deficit causing a mild left hemiparesis.

Drift of the outstretched pronated arm with the eyes closed is a good screening test of motor function (Fig. 3.18). However, there are several other potential causes including loss of proprioception, when the fingers tend to move independently – so-called ‘piano-playing’ or ‘pseudoathetosis’; neglect, when there tend to be much larger-amplitude movements, including upwards; or cerebellar dysfunction, when there tend to be largeramplitude oscillations, particularly if sharp downward pressure is applied to the arm. Thus, as a screening test for motor dysfunction, it is quite sensitive but not very specific and should be used in conjunction with the examination of fine finger movements. Some clinicians favour performing the test with the patient’s palms facing upwards, and if asymmetrical internal rotation occurs without downward drift this is taken as a sign of very subtle motor dysfunction. Minor motor deficits affecting the leg are probably best detected by drift of the leg (when flexed at the hip against gravity in a supine patient) and rapid tapping of the foot against the examiner’s hand. Additionally, the patient’s gait should be carefully observed.

The pattern of weakness within an individual limb is traditionally taught to be of localizing value. In particular, when the antigravity muscles (i.e. shoulder abductors; elbow, wrist, and finger extensors; hip and knee flexors; and ankle and toe dorsiflexors and ankle everters) are weaker than their counterparts (i.e. shoulder adductors; elbow, wrist, and finger flexors; hip and knee extensors; and ankle and toe flexors and ankle inverters), this is often described as a ‘pyramidal distribution’ of weakness. It has been suggested that this pattern is simply a function of the intrinsically greater strength in antigravity muscles together with the effects of hypertonia – the actual pattern of weakness being equally common in patients with central or peripheral lesions causing muscle weakness.68 The deep tendon reflexes were considered of more value as a localizing feature,68 although the inter-observer reliability of two standard scales for grading tendon reflexes is probably no better than ‘fair’ (kappa < 0.35).69 While the anatomical extent of the motor deficit is important for clinicoanatomical correlation, the severity is helpful in the acute phase for determining prognosis, the potential risks and benefits of interventions (such as thrombolysis) and the functional management and rehabilitation of the patient. There are several methods of quantifying the severity of motor weakness, such as the Medical Research Council (MRC) scale, NIHSS, and Scandinavian Neurological Stroke Scale which have an operational definition of the grades of weakness, and moderately good inter-observer reliability.46 In addition a description of some actions the patient can and cannot perform (e.g. holding a cup of water, combing hair) is helpful when trying to understand the problems that the patient is having and for appropriate goal setting during rehabilitation. And, for assessing change, it can be helpful to ask the patient, or note, what they can just do (e.g. can just extend the fingers against gravity, walk 10 m in 15 s); and worsening or improvement can then be easily observed, even if the weakness appears to have the same MRC grade. Attention should focus on the anatomical extent of the weakness and the functional consequences, rather than solely trying to grade the severity with a motor scale. An extensor plantar response is only one part of a nociceptive spinal flexion reflex, which in its complete

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 61

3.3 The diagnosis of a cerebrovascular event

(a)

Fig. 3.19 The Babinski sign: (a,b) evoked in this case by stroking the lateral part of the dorsum, rather than the sole, of the foot, in order to avoid voluntary withdrawal. The Babinski sign (c) involves contraction of the extensor hallucis longus simultaneously with other muscles that shorten the leg: tibialis anterior, the hamstrings (arrow), and the tensor fasciae latae (from van Gijn, 199570, by kind permission of the author and Postgraduate Medical Journal).

(c)

form (the sign of Babinski) involves flexion at the hip, knee and ankle as well as extension of the great toe70 (Fig. 3.19). Failure to appreciate this perhaps explains, in part, the rather poor reliability of the sign. While the presence of a Babinski response signifies a lesion of the corticospinal tract, it is not invariable, particularly if there is no weakness of the foot. Although swallowing involves both the motor and sensory systems, mention will be made of it here. The

..

..

(b)

gag reflex alone is not an adequate examination of the ninth and tenth cranial nerves, nor is it a good indicator of swallowing ability71 (section 11.17). Sensation on the two sides of the soft palate should be tested separately with an orange-stick, elevation of the palate should be observed, and the patient should be asked to cough. Failure to oppose the vocal cords adequately will result in some air escaping, which should alert the physician that the patient may have swallowing difficulties.

61

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 62

62

Chapter 3 Is it a vascular event and where is the lesion?

Clinical practice Most patients with stroke have motor symptoms or signs. Weakness usually affects one side of the body: the face, arm, or leg in isolation (monoparesis or monoplegia), each limb as a whole or in part, or a combination of these (hemiparesis or hemiplegia). Lesions in the internal capsule and ventral pons tend to result in a hemiparesis/ hemiplegia that is equally severe (proportional) in the arm and leg and is seldom accompanied by other neurological symptoms and signs – a pure motor stroke (section 4.3.2). When there is a brachial monoparesis (isolated weakness of the arm), or the weakness affects predominantly the face, hand or fingers even, it is more likely to be due to a cortical (where function is distributed anatomically) rather than a subcortical (where function is concentrated) lesion.72 When weakness involves the hand only, it is often referred to as a ‘cortical hand’. This can be mistaken for a peripheral nerve lesion, but closer analysis reveals that this would require simultaneous involvement of the median, ulnar and radial nerves, a most unlikely occurrence. It is generally considered to occur because of the large cortical representation of the hand. Isolated upper motor neurone facial weakness, however, seems to be of less localizing value, and it can certainly occur with very small infarcts in the genu of the internal capsule and in the pons. When weakness is confined to, or predominates in, the leg, the lesion is most likely to, but not invariably, involve the territory of the anterior cerebral artery or the sagittal venous sinus73 (section 4.2.2). Crossed weakness (i.e. weakness of one side of the face and the contralateral limbs) indicates a brainstem lesion or bilateral lesions (e.g. both hemispheres, or one hemisphere and contralateral brainstem). Paraplegia, triplegia and tetraplegia all occur more commonly from spinal than brain disorders (although see ‘locked-in syndrome’, section 3.3.2). When bilateral motor signs develop simultaneously – particularly if a cranial nerve palsy or crossed sensory disturbance (pointing to a brainstem lesion) are not present – and there is no sensory or reflex level to suggest a spinal cord lesion, cardiogenic embolism (i.e. causing two or more lesions), some kind of multifocal arteriopathy (such as vasculitis), abnormalities of the circle of Willis (section 4.2.2), or systemic hypotension (resulting in bilateral boundary-zone infarcts; section 4.2.4) must be considered. Very rarely hypotension results in paralysis predominantly of both arms (the ‘man in the barrel’ syndrome), with bilateral infarction in the boundary between the anterior and middle cerebral arteries. Although most strokes causing facial weakness result in a typical upper motor neurone pattern, there are some exceptions, which can lead to the erroneous diagnosis

of Bell’s palsy if there is minimal associated limb weakness. The most obvious exception is that of a brainstem stroke affecting the facial nerve nucleus. Occasionally, patients with very severe lower facial weakness from a supranuclear lesion also have some weakness of the upper face as well, particularly in the first few days after a stroke. This may reflect individual variation in the bilateral innervation of the upper facial (forehead) muscles. Lesions of the seventh cranial nerve nucleus involve both the upper and lower parts of the face; this pattern is not always due to Bell’s palsy. Sometimes patients seem at one moment to have (or are reported to have) a dense hemiplegia (usually leftsided) and yet, very soon afterwards, move the ‘paralysed’ limbs. This may be misinterpreted as a hysterical conversion disorder.74 However, this pattern can also be seen with the so-called capsular warning syndrome, or crescendo small vessel TIAs,75 although in such cases the episodes seem much more discrete (section 6.7.3). It can also occur in patients with a haemodynamically significant internal carotid artery stenosis, presumably due to subtle changes in distal perfusion pressure. However, the majority are patients in whom this seems to be a manifestation of inattention/neglect, or even apraxia. In patients who are recovering from what appears to be an extensive non-dominant hemisphere stroke, what seems to be a dense hemiplegia may improve very rapidly as the inattention/neglect begins to resolve – a fact that needs to be borne in mind when predicting the eventual functional outcome. It is always important to see whether a patient can sit up, get off the bed and walk, provided there is no risk to the patient or physician, whatever the motor deficit when tested on the bed. A severe deficit may be due to neglect and not weakness, and profound ataxia of gait may be associated with no motor deficit at all. Dysphagia is a common feature of acute stroke and an important cause of complications if not appropriately managed (section 11.17), but it is of limited neuroanatomical localizing value. Movement disorders such as hemiballismus, unilateral asterixis, hemichorea and focal dystonia occur in about 1% of patients with TIA and acute stroke due to contralateral, and rarely ipsilateral, small deep vascular lesions of the subthalamic nucleus, striatum or thalamus. Transient cerebral ischaemia may also masquerade as paroxysmal dyskinesia, and involuntary tonic limb spasms may arise contralateral to ventral pontine brainstem infarction. The abnormal movements usually regress spontaneously.

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 63

3.3 The diagnosis of a cerebrovascular event

(a) Fig. 3.20 (a) T2-weighted MR scan showing an extensive infarct in an area usually supplied by the middle cerebral artery (arrows) in a patient who initially presented with several episodes of jerking of the left arm (which were initially thought

to be epileptic) and who awoke three days later with a left hemiparesis. (b) Catheter angiography showing occlusion of the ipsilateral internal carotid artery (arrow).

Occasionally, patients describe jerking movements of the limbs just before the onset of a stroke, or during a TIA. The distinction from focal motor epilepsy may be difficult (section 3.4.2).52,76 However, in contrast to epileptic seizures, these attacks may be provoked by postural change (from lying to sitting or standing up), hyperextension of the neck, walking, coughing, or starting or increasing antihypertensive therapy, and they may be alleviated promptly by sitting or lying down, all of which suggest they are due to ‘low flow’ rather than embolism (section 6.7.5). There is an association with severe internal carotid artery stenosis or occlusion (Fig. 3.20),76 and the attacks usually stop after carotid endarterectomy. This pattern has also been reported with internal boundary-zone infarcts (section 4.2.4). Other disorders that sometimes need to be considered in the differential diagnosis of stroke are the GuillainBarré syndrome, mononeuropathies, drop attacks, cataplexy and motor neurone disease (section 3.4.11).

(also known as cutaneous or exteroceptive) includes light touch, pain and temperature modalities. Deep and proprioceptive sensation refers to deep pressure and joint position sense, respectively. Synthesis and appreciation of these sensory inputs occurs at a cortical level. Discriminative sensation refers to stereognosis, two-point discrimination and graphaesthesia. Paraesthesiae are positive sensory phenomena (e.g. pins and needles) that are presumed to occur because of partial damage to the sensory tracts or posterior horn cells, which become hyperexcitable (perhaps akin to brisk reflexes), such that ectopic impulses are generated either spontaneously or after a normal stimulus-evoked volley of impulses.

3.3.5 Disturbance of the somatic sensory system There are broadly two types of sensory message passing from the periphery to the brain. Superficial sensation

..

..

(b)

Clinical anatomy The main sensory pathways are shown in Fig. 3.21. Impulses for superficial sensation are conveyed in the spinothalamic tracts, which synapse in the dorsal horn, cross the midline at about the same spinal level and then ascend through the lateral spinal cord and brainstem. Fibres carrying similar sensory impulses from the face enter the ipsilateral, descending (or spinal) trigeminal

63

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 64

64

Chapter 3 Is it a vascular event and where is the lesion?

Hand Leg

Face

Thalamus

Medial geniculate nucleus Red nucleus Substantia nigra Spinothalamic tract (light touch, pain, temperature) Vth nerve

postero-medial nuclei (face, tongue, fingers). Lesions of the thalamus often involve all sensory modalities, although deep sensation may be more affected than superficial sensation. From the thalamus there are probably two main projections. The first is to the postcentral or primary somatic cortex where there is somatotopic representation, with the leg uppermost and the face lowermost. The areas of sensory representation in the cortex match those of the motor homunculus (Fig. 3.16). A second projection is to the area adjacent to the upper part of the sylvian fissure and insula. Here there is less discrete localization, but in general, the face is rostral and the leg caudal. Interestingly, stimulation of this area may result in bilateral symptoms. Lesions confined to the parietal lobe generally affect higher-level ‘discriminatory’ functions (i.e. proprioception, two-point discrimination, astereognosis) rather than primary modalities of sensation (e.g. pain, temperature), but sometimes the primary modalities are affected, giving rise to a ‘pseudothalamic’ syndrome.

Pons

Clinical assessment

Spinal tract of V Dorsal root ganglion

Medial lemniscus (deep sensation, joint position sense) Olive Medulla

Spinal cord

Fig. 3.21 Diagrammatic representation of the main sensory pathways between the entry of the dorsal root to the spinal cord and the sensory cortex.

nucleus and cross the midline in the upper cervical spinal cord. They then ascend through the medulla medially, close to the medial lemniscus, and separate to join the medial part of the spinothalamic tract in the pons. The relevant fibres for deep sensation are primarily in the ipsilateral posterior columns of the spinal cord. Decussation (crossing) occurs in the caudal medulla, after which the fibres ascend through the brainstem in the medial lemniscus. Fibres carrying similar sensory impulses from the face enter the primary trigeminal nucleus in the pons and cross the midline at this level to form the trigeminal lemniscus, which lies adjacent to the medial lemniscus. All these ascending fibres converge towards the midbrain and project, in the main, to the posterior group of thalamic nuclei and in particular to the ventropostero-lateral nuclei (trunk and legs) and ventro-

Somatosensory symptoms due to stroke are usually described by the patient as numbness (‘like the numbness I have after going to the dentist’), tingling, or a dead sensation; occasionally as loss of temperature sensation when in the bath or shower; and very rarely as pain (at least at onset). Often, patients find it difficult to describe unusual sensations in a manner that allows accurate classification, and the descriptions seem to vary between cultures. Nevertheless, it is the distribution of the sensory disturbance, usually involving the face and arm, or hand and leg, or face and arm and leg, that stamp the disturbance as involving the central nervous rather than the peripheral nervous system. The difficulty lies with isolated sensory disturbances that are limited to a part of a limb or a part of the face, which can be central or peripheral in origin. It is widely recognized that formal testing of the sensory system is one of the most unreliable parts of the neurological examination (see below). Quite frequently there is no detectable sensory loss. In general, therefore, sensory symptoms even in the absence of a deficit on examination should be noted. Indeed, on the assumption that the patients are able to communicate and do not have neglect, the only situation in which they will have sensory loss without sensory symptoms is when there is a restricted problem of discriminatory rather than primary sensory function (due to a parietal lesion), something that is uncommon in clinical practice. Conversely, care needs to be taken over the interpretation of very transient sensory symptoms, which can

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 65

3.3 The diagnosis of a cerebrovascular event

be within the range of normal experience, although the clinician should not accept uncritically the patients’ common interpretation of sensory symptoms as being due to a ‘trapped nerve’ or ‘lying in a draught’. Formal testing of the sensory system is the most unreliable part of the neurological examination. However, the physician should always take due note of sensory symptoms, even when there is no deficit on examination. Superficial sensation should ideally be tested in the standard manner, using a wisp of cotton wool (light touch), the side of a tuning fork (cold temperature), and an appropriate pin (i.e. not a hat pin or hypodermic needle) or other sharp object that can be disposed of after it has been used (pain). Proprioception may be assessed with the patient’s arms outstretched, the fingers spread and the eyes closed. Look for drift of the patient’s arm in a ‘pseudoathetoid’ or ‘piano-playing’ manner. This can be amplified by asking the patient to touch the tip of his or her nose with the forefinger while the eyes remain closed; those with disturbed proprioception will repeatedly miss the target. This screening test will assess proprioception around proximal as well as distal joints, but drift can also be due to a motor deficit or neglect/ inattention. Therefore, if possible, one should always attempt the traditional method of testing joint position sense of the distal interphalangeal joints. If there is expressive aphasia, it is sometimes worth asking the patient to indicate with gesture the direction of movement. Romberg’s test can be useful to assess position sense in the legs if there are no other deficits affecting them (e.g. cerebellar ataxia). Clinical practice The lacunar syndrome of pure sensory stroke is typically caused by a lateral thalamic infarct or haemorrhage.77 The deficit may involve all modalities, or may spare pain and temperature sensation. If there is extension to the internal capsule, a sensorimotor stroke may occur (section 4.3.2). The Déjerine–Roussy syndrome is caused by more extensive lateral thalamic infarction, and consists of a mild contralateral hemiparesis, marked hemianaesthesia, hemiataxia, astereognosis and frequently paroxysmal pain/hyperaesthesia and choreoathetotic movements. The original cases had extension of the infarcts into the internal capsule and towards the putamen, although most of the features of this syndrome result from involvement of the ventroposterior nuclei of the thalamus57 (section 4.2.3).

..

..

Central pain alone may be caused by lesions at any level of the somatosensory pathways. However, in our experience, most patients with a diffuse pain in the body at the onset of the event often turn out to have a nonorganic/functional disorder, and patients with severe localized limb pain tend to have other disorders, such as nerve root compression in the neck or low back, or even myocardial infarction if the pain is down the arm or in the hand. Nevertheless, we have seen the occasional patient with thalamic infarction or haemorrhage develop unpleasant sensation in the contralateral limb within 2–3 days of onset. Some patients have restricted sensory syndromes that affect unusual combinations of body parts. The most frequent is the cheiro-oral syndrome, where there is a sensory abnormality over the perioral area (sometimes bilaterally) and ipsilateral palm. In some patients, the foot may also be involved (the cheiro-oral-pedal syndrome) and, in both the hand and foot, certain digits may be affected while others are spared (a pseudoradicular distribution). Although lesions at most levels of the sensory pathways can give these patterns, lesions in the ventro-posterior nuclei of the thalamus are the most likely. It is thought that sensory projections from the face (with a particularly large representation for the lips), hand and foot are somatotopically arranged in the ventral portion of this nucleus, and the fingertips have particularly large representation areas, with that for the thumb more medial and the little finger more laterally. The projection areas for the trunk and proximal limbs are relatively small and sited more dorsally. However, a similar proximity of projection areas through the corona radiata and in the sensory cortex may also occur, and consequently lesions in these areas may also result in the cheiro-oral syndrome. An isolated deficit in a pseudoradicular distribution (most often involving the thumb and forefinger) is probably more often caused by a cortical lesion, because a stroke of any given size would there affect fibres from a more restricted anatomical area than in the thalamus. Bilateral symptoms may occur from midpontine lesions, because the sensory fibres from the mouth, arm and leg are once again arranged somatotopically in the medial lemnisci. A defect of pain and temperature sensation below a certain level on the trunk on one side is usually a sign of spinal cord disease, but this may occur from ischaemic lesions in the lateral medulla on the contralateral side. This is thought to be due to the orientation of fibres from different body parts within the lateral spinothalamic tract. Although patients usually have contralateral facial sensory loss as well, during the recovery phase this can disappear, leaving only the pseudospinal sensory loss. It is also worth noting that infarction of the cervical

65

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 66

66

Chapter 3 Is it a vascular event and where is the lesion?

spinal cord causing a partial Brown–Séquard syndrome can be due to bilateral vertebral artery dissection.78

3.3.6–Disturbance of the visual system When trying to assess the visual disturbances that occur in patients with cerebrovascular disease, one needs to consider: the reception of visual stimuli by the eyes; the transmission of the visual information from the eyes to the occipital cortex and visual association areas; and the interpretation of the visual information in the occipital cortex. Additionally, we have included in this section information about pupillary reactions and eye movements. Vision clinical anatomy Lesions at different sites in the visual pathway give highly characteristic abnormalities (Fig. 3.22). Amaurosis fugax (meaning literally ‘fleeting blindness’) and transient monocular blindness (TMB) are terms used interchangeably to describe temporary loss of vision in one eye. TMB may be caused by transient ischaemia in the distribution of the ophthalmic, posterior ciliary, or central retinal artery. Vascular lesions affecting the optic chiasm symmetrically (when one might detect a bitemporal hemianopia) are rare. Indeed, the only vascular

Retina

lesion of note would be a large aneurysm of the circle of Willis. Homonymous visual field deficits (i.e. loss of vision in the corresponding part of the visual fields in both eyes) signify a retrochiasmal lesion (Fig. 3.23). Lesions of the lateral geniculate nucleus may result in a homonymous horizontal sectoranopia (i.e. a segmental defect that respects the vertical but not the horizontal meridian) due to topographical organization within the lateral geniculate nucleus, but these rarely occur in isolation. The optic radiations pass from the lateral geniculate nucleus as the most posterior structures of the internal capsule. Involvement at this level is probably one of the causes of hemianopia in extensive middle cerebral artery territory infarction. The radiations do not seem to be affected by occlusion of a single perforating artery. Restricted lesions of the inferior optic radiation between the lateral geniculate nucleus and the calcarine cortex, where the fibres swing over the temporal horn of the lateral ventricle and deep into the temporal lobe (Meyer’s loop), result in a homonymous superior quadrantanopia, while lesions of the superior optic radiation in the parietal lobe result in a homonymous inferior quadrantanopia. On purely anatomical grounds, one might anticipate encountering an inferior quadrantanopia fairly frequently, because the middle cerebral artery is traditionally considered to supply the area through which the superior but not the inferior optic Visual field deficit Optic nerve

Anterior cerebral artery Middle cerebral artery Internal carotid artery Posterior cerebral artery

Ophthalmic artery Optic chiasm

Optic tract

Optic nerve

Lateral geniculate nucleus

Hypothalamus Optic tract Tectal plate

Superior optic radiation

Lateral geniculate nucleus

Optic radiation

Inferior optic radiation

Combined optic radiation

Visual field projection

Right

Calcarine cortex Visual cortex

Left

Fig. 3.22 Diagram of the visual pathways and their vascular supply, and the visual field defects that may result from vascular lesions at various sites along the visual pathway. The visual fields represented by the arrows on the retina correspond with the arrowheads and tails superimposed on the visual cortex. The dark purple represents visual pathways carrying vision from the left homonymous half fields and the light purple those from the right homonymous half fields. Arteries are shown in solid black.

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 67

3.3 The diagnosis of a cerebrovascular event

Homonymous hemianopia is a loss of corresponding field of vision in both eyes

When it affects the left visual field it looks like this:

f corresponding field of vision in both eyes

When it affects the right visual field it looks like this:

Homonymous hemianopia is a loss o

Fig. 3.23 Homonymous hemianopia.

radiation passes. However, in some patients, the area of supply of the middle cerebral artery extends much more posteriorly than is apparent on standard ‘maps’ (section 4.2.2), and an middle cerebral artery lesion thus produces a homonymous hemianopia by interrupting the optic radiations as they converge from the temporal and parietal lobe on their way to the occipital lobe.79 Also, perforating arteries from the carotid system supply the optic tracts and lateral geniculate nucleus (section 4.2.2). In other patients, the posterior cerebral artery may derive its blood supply via the posterior communicating artery, and therefore, like the middle cerebral artery, be affected by embolism from the internal carotid artery. Perhaps the most likely explanation, however, is that in many patients there is a mixture of a true visual field loss and visual inattention. If the entire calcarine cortex or optic tract on one side is damaged, there will be a complete homonymous hemianiopia, including macular vision. When this is an isolated feature, it is most likely due to a lesion in the occipital lobe, and when this is an infarct it will most often be due to occlusion of the posterior cerebral artery. Embolism is probably the most frequent cause, but giant-cell arteritis and migraine need to be considered. Sparing of macular vision with an otherwise complete hemianopia does occur in posterior cerebral artery territory infarction. The conventional explanation is that the infarction is confined to the lateral cortical

..

..

surface and that the macular cortex is spared because it receives sufficient collateral supply from the terminal branches of the middle cerebral artery (section 4.2.2). In theory, if the deficit in the two eyes is incongruous, it is most likely to be due to a lesion of the optic tracts, whereas if the deficits are entirely congruous, the lesion is likely to be in the calcarine cortex. However, it can be difficult to make this distinction in routine clinical practice. Visual agnosia is when primary visual perception is intact but the patient is unable to identify an object without resorting to the use of other sensory modalities such as touch. Prosopagnosia refers to the inability of patients to recognize familiar faces, even though they can describe them. In its pure form it is very rare, but it can result in enormous distress if patients deny recognizing their close family. Most lesions causing visual agnosia are in the anterior part of the dominant occipital lobe (the so-called visual association areas) and the angular gyrus, although most cases of prosopagnosia have occurred with bilateral lesions. clinical assessment Many patients have great difficulty in describing visual symptoms, particularly when they have resolved. It is therefore very important to be clear what a patient means. For example, the term ‘blackout’ may be used to mean bilateral blindness and also loss of consciousness,

67

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 68

68

Chapter 3 Is it a vascular event and where is the lesion?

and some patients who report ‘blurred vision’ in both eyes are actually trying to describe double vision. We find it useful to establish the functional severity of the visual disturbance – e.g. were they unable to find their way around, or to recognize faces (neither of which would occur with either TMB or an homonymous hemianopia). In vascular TMB the symptoms usually occur without provocation, but they may occasionally be precipitated by bright or white light, a change in posture, exercise, a hot bath or a heavy meal, particularly in patients with severe ipsilateral carotid disease. Generally, the visual loss during vascular TMB is painless (although some patients do complain of a dull ache or numbness in or above the eye) and very rapid. It can be described as if a blind or shutter had come down from above or, less often, up from below. The visual loss may be restricted to either the upper or lower half of the visual field and, less frequently, to the peripheral nasal and/or temporal field (in which case, be suspicious that the visual loss is or was binocular; i.e. a homonymous hemianopia). A pattern of diffuse, constricting or patchy loss may also occur. TMB may recur, usually in a stereotyped fashion, but the area of visual impairment may vary from one episode to the next, depending on which part of the retina is ischaemic. A frequent and, from a vascular anatomical point of view, very important clinical problem is trying to differentiate a transient homonymous hemianopia from TMB. The single most important question is to ask whether the patient covered each eye in turn during the episode. For patients without any residual visual disturbance, it can be useful to cover one eye and ask whether that reproduces the previously experienced effect. After covering the ‘good eye’, patients with TMB (of the ‘bad eye’) will have seen nothing, whereas patients with a homonymous hemianopia will have still seen something in the remaining half of the visual field. On the other hand, when patients cover the affected eye during TMB, they tend not to notice any visual disturbance because of normal vision from the other eye. In our experience, asking patients whether they saw half of everything often seems to cause confusion, and it may be better to ask them to look at your face and describe what they might have seen if they had been having an attack

When a patient complains of transient loss of vision in one eye, do not assume that the visual loss was monocular; it may have been a homonymous hemianopia. Ask the patient if they covered each eye in turn during the episode of loss of vision, and if so if the visual disturbance was present in both eyes or one (and if the latter, which eye).

Patients may not recognize an isolated homonymous hemianopia, or they may simply describe it as ‘blurred vision’ or ‘a shadow’. Even if they covered each eye in turn during the symptoms, it still may not be possible to be really confident, because homonymous hemianopia does not necessarily split macular vision and may be interpreted by the patient as loss of vision in one eye only. The presence of other symptoms may be helpful – for example, if there are ipsilateral visual and brain symptoms the visual problem is likely to be a hemianopia, whereas if they are contralateral it points to TMB. It may be difficult to test visual acuity formally (e.g. with a Snellen chart) in patients with acute stroke, because of drowsiness, aphasia, or the fact that they are bed-bound, but at least testing with a hand-held acuity chart or simply using everyday written material should be attempted. Many stroke patients are elderly, and concomitant eye diseases such as glaucoma, senile macular degeneration, cataracts and diabetic retinopathy are therefore common. It is important to identify these conditions at an early stage, because they may well make rehabilitation significantly more difficult. Indeed, the improved visual acuity that follows a cataract extraction may make the difference between being able to live independently (and safely) or not for a patient who has residual disability from a stroke. Because of the very large representation of the macula area in the occipital cortex, even patients who have a complete homonymous hemianopia do not have significantly reduced visual acuity per se. Elderly patients rarely attempt, let alone accomplish, tests of visual acuity without their own clean spectacles. If the visual loss is persistent (i.e. beyond several hours) and ophthalmoscopy reveals pallor of all or a section of the retina (due to cloudy swelling of the retinal ganglion cells), then the diagnosis is retinal infarction (Fig. 3.24). Additional findings may include an afferent pupillary defect, embolic material in the retinal arteries or arterioles, and a cherry-red spot over the fovea (due to accentuation of the normal fovea, which is devoid of ganglion cells, against the abnormally pale retina) in cases of central retinal artery occlusion (Fig. 3.25). If the eye is red and painful with a fixed, semidilated, oval pupil and cloudy/steamy cornea, then acute glaucoma is the likely diagnosis (section 3.5.1) (Fig. 3.26). Episcleral vascular congestion, a cloudy cornea, neovascularization of the iris (rubeosis iridis) and a sluggishly reactive mid-dilated pupil indicate chronic anterior segment ocular ischaemia which may be due to carotid occlusive disease, or small vessel disease particularly in a diabetic. This is so-called ischaemic oculopathy

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 69

3.3 The diagnosis of a cerebrovascular event

Fig. 3.24 An ocular fundus photograph of a patient with inferior temporal branch retinal artery occlusion, showing pallor of the inferior half of the retina due to cloudy swelling of the retinal ganglion cells caused by retinal infarction. The inferior temporal branch arteriole is attenuated and contains embolic material (arrow). Also reproduced in colour in the plate section.

(Fig. 3.27).2 Because ischaemic oculopathy is a gradual, chronic process it is commonly also called chronic ischaemic oculopathy. If there is a visual field defect of acute onset, such as an absolute or relative inferior altitudinal hemianopia, inferior nasal segmental loss, or central scotoma, and if ophthalmoscopy reveals swelling of a segment or all of the optic disc (which may be indistinguishable from that seen with raised intracranial pressure), flame-shaped haemorrhages near the disc and distended veins

Fig. 3.26 A photograph of the eye of a patient with acute glaucoma, showing congested sclera, cloudy cornea and oval pupil. Also reproduced in colour in the plate section.

..

..

Fig. 3.25 An ocular fundus photograph of a patient with central retinal artery occlusion, showing a cherry-red spot over the fovea (arrow). The cherry red spot is the normal fovea (devoid of ganglion cells), which seems more obvious because the surrounding infarcted retina has lost its red colour and appears pale. Also reproduced in colour in the plate section.

(Fig. 3.28), then the diagnosis is likely to be anterior ischaemic optic neuropathy (section 3.5.2). Later the optic disc becomes pale. Ophthalmoscopy may reveal retinal emboli but these are not necessarily symptomatic (up to 1–2% of the population over the age of 50 may have asymptomatic retinal emboli). The most common type are the bright orange or yellow crystals of cholesterol that originate from ulcerated atheroma in proximal arteries. Although cholesterol crystals are actually white, they appear orange

69

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 70

70

Chapter 3 Is it a vascular event and where is the lesion?

(a)

Fig. 3.28 An ocular fundus photograph of a patient with anterior ischaemic optic neuropathy due to occlusion of the posterior ciliary artery as a result of giant cell arteritis. Note the oedema of the optic disc and flame-shaped haemorrhages (arrow). Also reproduced in colour in the plate section.

(b) Fig. 3.27 (a) and (b) Ischaemic oculopathy of the right eye; note episcleral vascular congestion, cloudy cornea, neovascularization of the iris (rubeosis of the iris) and middilated pupil on external examination of the eye, which indicate chronic anterior segment ischaemia due to carotid occlusive disease (from Hankey & Warlow, 19942 by kind permission of the authors and W.B. Saunders Co. Ltd). Also reproduced in colour in the plate section.

or golden because their thin, fish-scale contour permits blood to pass above and below them and thus produce their characteristic refractile appearance (Fig. 3.29). Most of the crystals, because of their small size, thin, flat structure and lack of adhesiveness, pass through the retinal arterioles rapidly and rarely occlude the larger vessels, although it is probable that large clumps of crystals briefly occlude the central artery of the retina, producing TMB, before breaking up and being flushed away. White plugs of fibrin, platelets, or fatty material are less common. They occur in all sizes and are more likely to be symptomatic. Calcium emboli are chalky white angular crystals that tend to arise from calcific aortic stenosis, and they may permanently occlude the central retinal artery (behind the cribriform plate), or one of

Fig. 3.29 An ocular fundus photograph showing golden orange cholesterol crystals (Hollenshorst plaques) in the cilioretinal artery (arrows). The cilioretinal artery is present in only about one-third of the population. It originates from a branch of the short posterior ciliary artery and supplies the macula. Also reproduced in colour in the plate section.

the branch retinal arterioles near the optic disc. Other less common types of emboli include microorganisms (septic), fat and tumour cells. Roth spots, which are very small white infarcts encircled by haemorrhage, were

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 71

3.3 The diagnosis of a cerebrovascular event

Fig. 3.30 An ocular fundus photograph showing narrowing and tortuosity of retinal arterioles, arteriovenous nipping, retinal haemorrhages and papilloedema. These are the features of hypertensive retinopathy seen in malignant hypertension. Also reproduced in colour in the plate section.

thought to be caused by septic emboli, but it now seems more likely that they are due to rupture of retinal capillaries and the extrusion of blood cells. Whilst the interobserver and intra-observer agreement for the detection of retinal emboli is quite high (kappa = 0.73 and 0.63, respectively), agreement on a range of qualitative assessments of emboli type is much poorer. Narrowing, focal irregularity/constriction and tortuosity of retinal arterioles, arteriovenous nipping and fluffy white patches of transudate (‘cotton-wool patches’ which are thought to represent small focal infarcts in the inner layers of the retina), indicate long-standing hypertension. If papilloedema and retinal haemorrhages are also present, this indicates malignant hypertension, but is now uncommon (Fig. 3.30). Retinal haemorrhages without the other changes of hypertensive or diabetic retinopathy in a patient with a non-traumatic, acute neurological event is strong evidence of a haemorrhagic stroke. They are usually caused by a very sudden increase in intracranial pressure transmitted to the distal optic nerve sheath, where it causes a temporary obstruction of retinal venous outflow. The subsequent rise in retinal venous pressure leads to secondary bleeding from retinal veins and capillaries. The appearance of the haemorrhage depends on its site. Small dot and blot haemorrhages lie in the deep retinal layers; linear haemorrhages in the superficial (nerve fibre) layer; ‘thumbprint’ haemorrhages with frayed borders are preretinal or superficial retinal; and large subhyaloid haemorrhages (large round haemorrhages

..

..

with a fluid level) lie between the retina and the internal limiting membrane. Subhyaloid haemorrhages and other types of retinal haemorrhage can be seen in about 20% of patients with subarachnoid haemorrhage (section 3.7). Assessment of the visual fields must be tailored to the patient’s overall condition. It is important that the physician has a repertoire of methods and does not give up simply because ‘formal’ testing by confrontation is impossible in a patient who is drowsy, aphasic, cognitively impaired, or just cannot sit up. Kinetic testing (i.e. using moving objects, waggling fingers, etc.) is a less sensitive way of detecting deficits than static methods such as counting fingers or comparing colours in each hemifield. If the patients can understand and communicate, one should first ask them to describe what they see in front of them, perhaps using the same text as for testing visual acuity, and ideally testing each eye individually. After that, hold up fingers sequentially in each quadrant of vision in each eye and ask the patient to count them or, for greater sensitivity, use a pin with a red ball on its end. This will detect a hemianopia or quadrantanopia. Following this, perform bilateral, simultaneous finger movement to detect evidence of visual inattention. Testing of the visual fields with automated perimetry is extremely tiring and of little value in the acute situation. Indeed, it may produce bizarre deficits that are normally associated with non-organic disorders. However, it may be necessary later when there is doubt about eligibility for driving. Although testing the visual fields using conventional confrontation methods is often impossible because the patient is drowsy, dysphasic, cognitively impaired, or cannot sit up, one can usually use other methods to determine whether or not there is an abnormality. If, for whatever reason, the patient cannot follow commands, one will need to use quite gross stimuli to be sure of eliciting and identifying a response if the visual fields are intact. Examples include observing if there is any response to moving a brightly coloured object in one hemifield, getting a colleague to approach the patient from one side, or seeing whether the patient blinks when a threatening stimulus (e.g. a quickly moving finger) is brought towards the eye (one needs to be careful that the associated air current is not simply stimulating a corneal reflex). A hemianopia is almost always associated with ipsilateral loss of this blink reflex, although the reverse is not always true. If the patient is not aphasic and there are members of the team around the bed, ask the patient to point to each one of the team in turn. An asymmetrical response to any of these tests suggests a field defect, inattention, neglect or a combination of these.

71

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 72

72

Chapter 3 Is it a vascular event and where is the lesion?

It is not surprising, given these difficulties, that the inter-observer reliability of the assessment of visual fields is relatively poor (Table 3.31). clinical practice Some patients with severe stenosis or occlusion of one or both internal carotid arteries may experience a visual disturbance in one or both eyes on exposure to bright sunlight or white light. Blurring, dimming, or constriction of the visual field from the periphery to the centre of vision of the involved eye develops over minutes rather than seconds. Objects appear bleached like a photographic negative, or there may be a scotoma or complete visual loss; an altitudinal defect is most unusual. Such transient episodes of monocular or binocular blindness are presumed to be due to low flow in the choroidal circulation. They are typically less rapid in onset than the brief transient attacks of embolic origin, and sight returns more gradually. Sunglasses may be an effective symptomatic treatment. Cortical blindness is a syndrome in which the patient has no vision despite normally functioning eyes and anterior visual pathways. Cases of sudden, spontaneous and simultaneous dimming or loss of vision in all of the visual field of both eyes are presumed to be due to bilateral occipital lobe (visual cortices and optic radiations) ischaemia/infarction due to occlusion of the top of the basilar artery or both posterior cerebral arteries. If the visual symptoms occur in isolation (without associated symptoms of focal cerebral ischaemia, seizures, or reduction in consciousness) in an elderly patient and if they resolve within 24·h, they are probably due to a TIA of the occipital lobes.48 However, when the same symptom occurs in adolescents and young adults, investigations are unlikely to reveal a cause, and the long-term prognosis appears benign. Occasionally, when the deficit persists, true cortical blindness has to be distinguished from non-organic visual loss. This is best detected with an optokinetic drum or a long piece of material with vertical stripes (e.g. a scarf or tape measure) because it is impossible to suppress nystagmus voluntarily if there is visual function. Sometimes, genuine bilateral blindness is denied by the patient (Anton syndrome) which signifies involvement of the association areas adjacent to the primary visual cortex, but otherwise the pathogenesis is as unclear as that of the denial of left hemiplegia or anosognosia in general. Perhaps the right hemisphere component is crucial. Visual hallucinations can occur in patients with stroke involving the occipital, temporal and parietal cortices as well as the eye, optic pathways and cerebral peduncle (section 11.27.3). Those secondary to occipital lesions most commonly consist of elementary

(unformed) visual perceptions, sensations of light and colours, simple geometric figures, and movements. Posterior temporal lesions, involving the association cortex, result in more complex (formed) visual hallucinations, consisting of faces and scenes that may include objects, pictures and people. Lesions in the high midbrain, particularly the pars reticulata of the substantia nigra, may give rise to the so-called ‘peduncular hallucinosis’ of Lhermitte, in which the hallucinations are purely visual, appear natural in form and colour, move about like an animated cartoon, and are usually considered to be unreal, abnormal phenomena (i.e. insight is preserved). More commonly, however, visual hallucinations are due to non-vascular disorders such as migraine or partial seizures (in which case, the hallucinations are usually unformed), psychosis, or an adverse effect of a drug such as levodopa. Micropsia, which is the illusion of objects appearing smaller than normal, and palinopsia, which is the persistence or recurrence of visual images after the stimulus has been removed, can occur with parietal lobe lesions. Flashing lights, shooting stars, scintillations, or other positive phenomena in the area of impaired vision occasionally arise during retinal or optic nerve ischaemia, but are far more commonly features of migraine or glaucoma. Pupils clinical anatomy The size of the pupil is determined by the balance of tonic impulses from the pupillodilator fibres, which receive input from the sympathetic nervous system, and the pupilloconstrictor fibres, which receive input from the parasympathetic nervous system. The sympathetic fibres descend ipsilaterally from the hypothalamus, through the lateral brainstem adjacent to the spinothalamic tract. They occupy a more central position in the lateral grey column of the cervical spinal cord and exit via the first thoracic root. The fibres then pass across the apex of the lung to enter the sympathetic chain, which ascends through the neck in association with the carotid artery. The fibres associated with sweating separate in the superior cervical ganglion and then travel in association with branches of the external carotid artery. The other fibres enter the cranial cavity on the surface of the internal carotid artery. The fibres innervate the pupil via the long ciliary nerves, whilst those supplying the tarsal muscles are carried in the third cranial nerve. Following reception of light by the retina, impulses are conveyed in the optic nerve. After the optic chiasm, they are conducted in both optic tracts to both Edinger– Westphal nuclei (a distinct part of the third nerve nuclear complex). The parasympathetic nerves exit

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 73

3.3 The diagnosis of a cerebrovascular event

Lateral geniculate nucleus From hypothalamus, via Optic tract posterior commissure Optic nerve Midbrain

Retina Fig. 3.31 The neurogenic control of pupil size: an outline of the parasympathetic and sympathetic pathways involved in pupil constriction (–) and pupil dilatation (–). The third cranial nerve nuclear complex consists of: the Edinger-Westphal nuclei concerned with parasympathetic innervation of the pupils; the midline nucleus of Perlia, concerned with convergence and accommodation; and the lateral nuclei, which innervate the levator palpebrae, superior recti, inferior oblique, medial recti and inferior recti muscles. It is possible for vascular lesions to cause ischaemia of the lateral nuclei (resulting in extraocular palsy) but spare the pupilloconstrictor fibres from the Edinger-Westphal nuclei.

Pretectal nucleus Pupil

Optic chiasm

Internal carotid artery Parasympathetic pupilloconstrictor reflex Superior Sympathetic pupillocervical dilator reflex ganglion

Ciliary ganglion

alongside the third nerve and travel with it to the orbit (Fig. 3.31). There, they synapse in the ciliary ganglion, which gives rise to the short ciliary nerves that innervate the sphincter pupillae and the ciliary muscle. Lesions anterior to the lateral geniculate body result in loss of the pupillary light reflex. clinical assessment It is very uncommon for a patient to be aware of their own pupillary abnormalities, but others may notice. Just occasionally, patients notice that their pupils are unequal – they will usually think one is dilated, rather than the more common abnormality of one being constricted. With a truly dilated pupil, the patient may be distressed by abnormal brightness and difficulty focusing. The response of the pupils to light – both direct and consensual – should be tested, as well as accommodation, if possible. Interruption of the descending sympathetic pathway in the brainstem and at other sites before the carotid bifurcation results in a complete ipsilateral Horner syndrome, i.e. miosis, ptosis and loss of sweating on the side of the face. Lesions of the internal carotid artery (e.g. carotid dissection; section 7.2.1) generally spare facial sweating. Sometimes a transient Horner syndrome is the only clue to a carotid dissection. clinical practice There are many causes of anisocoria (unequal pupils) in the elderly, most are not vascular. Perhaps the commonest is the use of drops to treat glaucoma, but any local inflammatory condition (e.g. iritis) can be responsible. Furthermore, physiological anisocoria may occur in

..

..

Edinger-Westphal nucleus

IIIrd nerve

Spinal cord (C8-T1) Ascending sympathetic chain

up to 20% of the normal population. Therefore, anisocoria very much needs to be assessed in the context of any other signs (such as ptosis). Elderly patients with stroke may be using pupilloconstrictor drops for glaucoma, and this may cause unequal pupils. Because of the functional separation of fibres within the third nerve nuclear complex, it is possible for third nerve palsies from midbrain vascular lesions to spare the pupillary reaction, which remains normal to light. On the other hand, in an unconscious patient with extensive damage to the midbrain (either due to intrinsic disease or secondary to pressure from above), the pupils will both be fixed and either dilated or in mid-position (4–5·mm), depending on whether the sympathetic as well as the parasympathetic fibres are involved. Bilateral ‘pinpoint’ pupils in an unconscious patient suggests an extensive lesion in the pons if there is no evidence of drug overdose (e.g. opiates). This is thought to be due to a combination of damage to the sympathetic fibres and irritation of the parasympathetic fibres (lesions solely of sympathetic fibres do not usually result in such intense pupilloconstriction). Despite this, the pupils react to a bright light, although this may be difficult to observe. External ocular movements and eyelids clinical anatomy The external ocular muscles maintain fusion of the images from each retina. The oculomotor (third nerve) complex

73

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 74

74

Chapter 3 Is it a vascular event and where is the lesion?

in the midbrain innervates the medial, superior and inferior recti and inferior oblique muscles. The trochlear nerve (fourth nerve), also originating in the midbrain, innervates the superior oblique muscle, and the abducens nerve (sixth nerve) in the pons innervates the lateral rectus muscle. Other structures of importance include the medial longitudinal fasciculus (MLF) which effectively links the nuclei; the paramedian pontine reticular formation, sometimes known as the pontine lateral gaze centre; and the rostral interstitial nucleus of the MLF in the midbrain which generates the vertical and torsional components of eye movement (Fig. 3.32). The cerebellum and vestibular nuclei are also important for the control of eye movements. The supranuclear control of conjugate eye movement is of relevance to patients with stroke. Voluntary eye movements are initiated in the frontal eye field, which is anterior to the precentral gyrus, whilst the reflex visual pursuit movements involve the occipital cortex and visual association areas. The fibres from these areas do not directly innervate the oculomotor nuclei, but rather their input is integrated by the MLF. The levator palpebrae superioris is innervated by the superior division of the oculomotor nuclei in the midline, as are the superior rectus muscles. Because of this, movement of the eyelids is closely linked to vertical eye movements. The superior and inferior tarsal muscles receive sympathetic innervation via the third cranial nerve. They assist eye opening, and when they are paralysed (as in Horner syndrome), the palpebral fissure is narrowed (but there is not complete ptosis, as is seen with a third nerve palsy, or central nuclear third nerve palsy). Nystagmus is an involuntary, biphasic ocular oscillation that occurs with lesions of the vestibular pathway and cerebellum. In patients with vascular disease, there may be either ischaemia of the labyrinth or of the vestibular nuclei in the brainstem, and the pattern of any associated signs may be of more use in localizing the lesions than attempts to analyse the nystagmus itself – i.e. with nuclear lesions, there are likely to be other signs of brainstem disturbance. clinical assessment Hemispheric stroke involving the frontal cortex results in a failure of voluntary conjugate gaze to the contralateral side. Because of the unopposed action of the intact frontal eye field of the hemisphere not affected by the stroke, there is conjugate deviation of gaze towards the side of the stroke lesion. This normally settles over a period of 1–2 weeks. The patients are usually drowsy so that reflex pursuit movements cannot be assessed, although if the lesion was confined to the frontal lobe one might expect such movements to be retained. If

Frontal lobe Occipital lobe

III

Tectal field for vertical eye movements

III

Midbrain

Superior colliculus

IV

IV Medial longitudinal fasiculus

Inferior colliculus

x y Pons

y VI

VI

Pontine field for horizontal eye movement

Midline Fig. 3.32 The role of the medial longitudinal fasciculus (MLF) in the control of conjugate gaze. Conjugate gaze requires coordinated action of the third, fourth and sixth cranial nerve nuclei and nerves in the brainstem. The MLF links these nuclei and also provides a pathway for inputs from the frontal eye field (for voluntary eye movements – saccades) and the occipital lobe (for reflex movements), as well as the tectal field for vertical eye movements and the pontine field for horizontal eye movements. ‘x’ indicates the site of the lesion that results in the classical bilateral internuclear ophthalmoplegia often associated with multiple sclerosis. Vascular lesions (‘y’) more often result in unilateral internuclear ophthalmoplegia, presumably because they are more likely to respect the vascular territory of a paramedian pontine perforating artery, and thus the midline, than a plaque of demyelination. Conversely, a lesion of the lateral pons may prevent ipsilateral lateral gaze, and there may be a conjugate deviation of the eyes away from the side of the lesion.

patients have an acute pontine lesion, conjugate deviation away from the side of the lesion may occur, but this is less likely to recover. When a patient complains of double vision, the following questions help to identify the site of the problem: • is the double vision present when one eye is closed (monocular diplopia), or only when both eyes are open (binocular diplopia)?

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 75

3.3 The diagnosis of a cerebrovascular event

• are the images separated side by side (horizontal), one above the other (vertical), or at an angle to one another (oblique)? • and in which direction of gaze are the images separated maximally? If the patient is conscious, communicative and cooperative, the eye movements can be tested in the normal way. However, it is difficult to do this if patients cannot follow commands. In such cases, spontaneous movements in each direction should be observed to confirm the absence of a gaze palsy. One can also stimulate patients to look in each direction by doing something ‘interesting’ in different fields of vision – so that they follow the examiner’s face, for example, rather than a finger or a pen. This also has the advantage that the examiner can continually reinforce the command ‘watch my nose’. The patient may be stimulated to look in each direction by doing something ‘interesting’ in different fields of vision – so that he or she follows the examiner’s face, for example, rather than a finger or a pen. Nystagmus of brainstem or cerebellar origin is probably best appreciated by asking the patient to fixate on, and then follow, a moving target. A few irregular ‘jerks’ of the eyes are often seen in normal people when they move their eyes, particularly at the extremes of lateral gaze. Acquired pendular nystagmus (i.e. where there is a sinusoidal waveform) is associated with lesions of the tegmentum of the pons and medulla. Upbeat nystagmus is associated with pontine or cerebellar lesions, and downbeat nystagmus with lesions in the medulla or at the craniocervical junction. A torsional or rotatory component can occur from both central and peripheral lesions. Convergence retraction nystagmus (rhythmic oscillation in which a slow abduction of the eyes in respect to each other is followed by a quick movement of adduction, and is usually accompanied by a quick rhythmic retraction of the eyes into the orbits) is considered indicative of midbrain disease; this is actually a disorder of horizontal eye movement rather than true nystagmus. A number of related disorders, which are probably due to a disturbance of saccadic (voluntary) eye movements, are associated with cerebellar disease. These include ocular dysmetria (where there is overshoot of the eyes on attempted fixation), ocular flutter (where there are occasional bursts of rapid horizontal oscillations without an intersaccadic interval), and so-called square wave jerks. clinical practice Transient diplopia in isolation may be an indication of a brainstem ischaemic event but can also be due to

..

..

myasthenia gravis, for example. However, transient diplopia in association with other symptoms of brainstem or cerebellar dysfunction – such as unilateral or bilateral motor or sensory disturbances, vertigo, ataxia or dysarthria – usually signifies a TIA in the vertebrobasilar circulation. Monocular diplopia is usually due to intraocular disease causing light rays to be dispersed onto the retina (e.g. corneal disease, cataract, vitreous haemorrhage) or to functional (non-neurological) disturbance. It has been reported rarely after occipital lesions but it is not due to paralysis of extraocular muscles. Even though it is sometimes claimed to be a pathognomonic sign of multiple sclerosis, vascular disease can cause an internuclear ophthalmoplegia (failure of adduction in the adducting eye, with nystagmus in the abducting eye), due to involvement of the MLF on the side of the adducting eye (Fig. 3.32). A failure of conjugate horizontal gaze to one side can occur with ischaemia of the ipsilateral (the side to which the patient cannot look) paramedian pontine reticular formation. Additional involvement of the ipsilateral MLF (with failure of adduction of the ipsilateral eye on attempted gaze to the other side) may result in the so-called ‘one-and-a-half syndrome’, where the only remaining horizontal eye movement is abduction of the contralateral eye. An oculomotor (third cranial) nerve palsy at the onset of subarachnoid haemorrhage (SAH) frequently indicates a ruptured aneurysm at the origin of the posterior communicating artery from the internal carotid artery, less frequently an aneurysm of the carotid bifurcation, the posterior cerebral artery, the basilar bifurcation or the superior cerebellar artery. Third nerve palsy may also occur with unruptured aneurysms (presumably by expansion) or several days after SAH as a result of swelling of the ipsilateral cerebral hemisphere because of delayed cerebral ischaemia. Most often the pupil is dilated and unreactive, but in some patients it is spared. Abducens (sixth cranial) nerve palsies, frequently bilateral in the acute stage, may develop after SAH as a false localizing sign of raised intracranial pressure, because of traction on the nerves against the petrous temporal bone, or by downward transtentorial herniation of the diencephalon. Occasionally, posterior circulation aneurysms may cause a sixth nerve palsy as a result of direct compression.

If a patient presents following a sudden, severe headache and is found to have a third cranial nerve palsy, rupture of a posterior communicating artery aneurysm is highly likely.

75

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 76

76

Chapter 3 Is it a vascular event and where is the lesion?

Occasionally, one encounters patients who are unable to open their eyes. If bilateral ptosis is associated with a vertical gaze palsy and there is no suggestion of myasthenia gravis, then it is probably due to a nuclear third nerve palsy (dorsal part, in the midline) or a massive right hemisphere lesion (‘cerebral ptosis’).80 ‘Cerebral ptosis’ must be distinguished from blepharospasm, and from apraxia of eyelid opening in which patients are unable to open their eyes on command but can do so spontaneously (i.e. the eyelid dysfunction is episodic and may be precipitated by eyelid closure, often in patients with extrapyramidal disorders). Ocular bobbing is usually only present with an impaired conscious level and extensive pontine disease. The spontaneous rapid downward movement of the eyes is followed by a slow drift back to the original position. It is thought to occur because of the tendency of such patients to have roving eye movements but, without any horizontal gaze, the only possible movements are in the vertical plane. Oscillopsia is an illusion of movement, or oscillation of the environment. The patient may complain that static objects are oscillating either from side to side or up and down. This can occur with nystagmus or any of the other tonic abnormalities of eye movement, but these can sometimes be difficult to demonstrate. This symptom, although uncommon, can be extremely distressing and disabling. Tortopia is the illusion of transient tilting or inversion of the environment. This can occur with cerebellar ischaemia.

3.3.7 Disturbance of hearing, balance and coordination Clinical anatomy ‘Dizziness’ may refer to light-headedness, imbalance, a feeling of faintness, a lack of mental clarity, or frank vertigo – the patient must be asked to explain exactly what he or she means by dizziness. Vertigo is a subjective or objective illusion of motion (usually rotation) or position. It is a symptom of dysfunction of the peripheral or central vestibulocerebellar system. Dysequilibrium is a sensation of imbalance when standing or walking due to impairment of vestibular, sensory, cerebellar, visual, or motor function, and consequently it may be due to lesions in many parts of the nervous system. Ataxia (derived from the Greek meaning ‘lack of order’) is disordered coordination of the extremities (limb ataxia), and imbalance of sitting (truncal ataxia)

or gait (gait ataxia). It is typically associated with disorders of the cerebellum or the cerebellar connections in the brainstem. However, lesions of the thalamus, particularly within the posterolateral territory supplied by the thalamogeniculate artery, may present with isolated contralateral ataxia – although more often there is an additional motor and/or sensory deficit.57 The most likely explanation is that the ventrolateral nucleus receives input from the cerebellar, vestibular and spinothalamic systems (e.g. dentato-rubro-thalamic tract). A related condition in which patients are unable to stand or even sit unsupported, in the absence of marked motor deficit, has been termed astasia and is associated with lesions in the posterolateral thalamus. Sudden unilateral hearing impairment, with or without ipsilateral tinnitus, is a symptom of dysfunction of the cochlea, vestibulocochlear nerve, or cochlear nucleus. The relationship between the ascending and descending fibre tracts and the cranial nerve nuclei in the brainstem is shown in Fig. 3.33. Clinical assessment When patients complain of an illusory sense of movement, the first step is to distinguish rotatory vertigo or tilting of the visual axis from less specific symptoms such as faintness, and then to localize the disturbance to the brainstem (central) or to the vestibulocochlear nerve or labyrinth (peripheral).81,82 To define as closely as possible the patient’s actual sensations, direct questions often have to be asked; for example, ‘is it a spinning feeling or just light-headedness?’ Descriptions that include a subjective or objective illusion of motion, such as spinning or whirling, which is usually so unpleasant that it makes the patient feel nauseated and also unable to stand, denote what is meant by vertigo. Feelings of lightheadedness, swaying, a swimming feeling, walking on air, queer head or faintness (often with accompanying feelings of panic, palpitations or breathlessness), without a feeling of motion, are non-specific and may be caused by a wide variety of systemic disturbances (usually hypotension, panic or overbreathing). Precipitating factors and premonitory symptoms may be of diagnostic value, as also may be the mode of onset (whether sudden or gradual), the duration, and the presence of any associated symptoms such as deafness, tinnitus and ear pain or fullness. Indeed, it is not so much the character of the vertigo that helps to localize the disorder as the associated features of the attacks. For example, vertigo accompanied by features of brainstem dysfunction such as diplopia and face and limb sensory disturbance, with normal hearing, points to a central cause; whereas vertigo triggered by sudden movements and positional

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 77

3.3 The diagnosis of a cerebrovascular event

Midbrain Aqueduct III nerve nucleus

Superior cerebellar peduncle

Pons Medial longitudinal fasciculus Lateral spinothalamic tract Medial lemniscus

Fourth ventricle

Medial longitudinal fasciculus

Vestibular nuclei

Cerebellar peduncle

VIIIn nucleus

Central sympathetic pathway

VIn nucleus Vn nucleus

Lateral spinothalamic tract

VIIn nucleus Corticospinal and corticobulbar tracts

(a)

Corticospinal tract

Medial lemniscus

(b) Medulla

Vestibular nuclei

XII nerve nucleus

Medial lemniscus

Central sympathetic pathway

Inferior cerebellar peduncle IXn + Xn nuclei

Fig. 3.33 Diagrammatic representation of the main anatomical structures within the brainstem: (a) midbrain; (b) pons; (c) medulla.

changes and associated with auditory or ear symptoms points to a peripheral cause.53 Unsteadiness is a fairly common symptom in stroke/ TIA patients, but unless it is associated with clearly focal symptoms or residual neurological signs of weakness or ataxia, it can be difficult to decide whether the patient means weakness, incoordination, vertigo, presyncope or anxiety, or indeed a combination of these. Sometimes it is helpful to ask patients whether they felt unsteady in the head or in the legs, or whether it was a visual problem. Patients often complain of being unsteady when they have rotational vertigo, but it is useful to ask quite specifically if there was persisting unsteadiness after the vertigo had stopped – a positive response suggesting a central rather than a peripheral problem. The most common manifestation of cerebrovascular disease involving the cerebellum or its connections is truncal ataxia. Signs in the limbs are traditionally considered to need involvement of the ipsilateral cerebellar hemisphere rather than the midline structures, and thus there are often no cerebellar signs when the limbs are

..

..

Olivary nucleus

Trigeminal spinal tract

Lateral spinothalamic tract Corticospinal tract

(c)

examined on the bed. Consequently, the problem may be overlooked if the patient’s gait is not examined. If the disorder is mild, asking the patient to turn round quickly may be the most sensitive way of detecting an abnormality. It is also worth asking about the impact of loss of visual fixation, e.g. in the dark, in a shower, or performing Romberg’s test. Most ataxic syndromes are somewhat worse when visual fixation is lost, but a marked loss of balance should make one think of disordered proprioception (i.e. sensory ataxia). Cerebellar disorders may be missed if the patient’s gait is not examined. Care must be taken when faced with a resolving hemiparesis, since at this time cerebellar-like signs can be elicited that are probably just a manifestation of impaired corticospinal control. This is most relevant when trying to distinguish the lacunar syndromes of pure motor stroke and ataxic hemiparesis from each other (section 4.3.2).

77

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 78

78

Chapter 3 Is it a vascular event and where is the lesion?

Clinical practice Many patients complain of ‘dizziness’, either around the onset of a stroke or at other times, but this term alone is too imprecise to be of localizing value, even between carotid and vertebrobasilar territories, although it is probably more frequent in the latter. Other sensory experiences, such as giddiness, light-headedness, swimminess, or a feeling of passing out may be caused by diffuse cerebral ischaemia due to postural hypotension, vasoactive medication, cardiac dysrhythmias (all common in the elderly), and also with anxiety states, panic attacks and hyperventilation. These symptoms are of no localizing value and should not be considered to be indicative of a stroke/TIA. The tendency to label any episode of ‘dizziness’, especially in the elderly, as vertebrobasilar ischaemia or – worse – insufficiency, should be strongly resisted.53 All too frequently one sees patients who have complained of ‘dizziness’ when turning their neck, have then had a cervical spine X-ray that shows some degenerative changes (extremely common in the elderly population), and are then told that they are ‘trapping the blood supply to the brain’. What little work there is to support this theory is mainly based on postmortem studies or dynamic arteriography in highly selected patients (section 7.1.5). Furthermore, many of the patients initially described under this banner had clear focal disturbances of brainstem and occipital lobe function. There is thought to be a considerable proprioceptive input from structures in and around the cervical spine to the vestibular system, and therefore for the vast majority of patients with these non-focal symptoms, the terms simply engender undue anxiety about impending strokes and divert attention from more likely and potentially treatable explanations of their dizziness. When vertigo is an isolated phenomenon, and particularly when it is induced by head movement, severe, and short-lived, it is almost always a symptom of peripheral rather than central dysfunction. The relatively common condition of benign paroxysmal positional vertigo is described in section 3.4.7. Rarely, and most commonly in patients with diabetes, occlusion of a branch of the anterior cerebellar artery or the basilar artery supplying the inner ear can cause vertigo, unilateral hearing loss or both.83 Some cerebellar strokes and lesions of more central vestibulocerebellar pathways may also mimic ‘peripheral’ symptoms. Multidirectional nystagmus that is not suppressed by visual fixation, an inability to stand without support and a negative ‘head impulse test’ in a patient with a first-ever attack of acute, spontaneous vertigo points to a central rather than a peripheral process.81,82 The head impulse test is a simple clinical

test comprising high acceleration head rotation;84 with severe unilateral vestibular weakness the normal vestibuloocular reflex is replaced by a misalignment of the eyes followed by a series of corrective saccades which are evident to the examiner.84 Sudden hearing loss may be caused by trauma, tympanic membrane rupture, viral and other infections, toxic and metabolic disorders, and ischaemia.85 How often sudden unilateral hearing loss in isolation (without vertigo or other brainstem dysfunction) is due to vascular disease is unknown, however, because histopathological examination of the temporal bone and labyrinth at postmortem is not performed routinely and, if it is, it is usually carried out so long after the onset of any hearing loss that the clinical details are unclear. Nevertheless, there is some histopathological evidence of labyrinthine infarction due to vascular disease. Even greater uncertainty prevails as to whether transient deafness in isolation is ever due to a TIA in the internal auditory artery territory, but we do not see why this should not sometimes be the case. Acute cerebellar syndromes that can mimic strokes may be caused by drug toxicity, Wernicke’s encephalopathy and Creutzfeldt–Jakob disease (sections 3.4.5 and 3.4.6).

3.3.8–The speed of onset of the symptoms and temporal course Patients with a TIA or stroke usually describe their neurological symptoms as coming on abruptly, without warning, and as being more or less maximal at onset. However, patients are often frightened by their symptoms and are probably not very accurate about the exact time (seconds vs minutes) they take to develop. If several parts of the body (e.g. face, arm and leg) are affected, the symptoms usually start in each part almost simultaneously rather than intensifying or spreading (‘marching’) from one part to another – a pattern more typical of focal epilepsy52 or a migrainous aura.86 However, the simultaneous onset of a disturbance of several body parts may not be recognized, and the event misinterpreted as ‘spread’. For example, a patient who is playing the piano at the time of a stroke will notice weakness of the hand but not weakness of the leg until he or she tries to stand up, or dysphasia until they try to speak. Because it is the suddenness of onset that stamps the event as vascular, it is useful to ask patients what they were doing at the time; if they were awake and do not remember, the onset was probably not all that sudden. Less commonly, other patterns are encountered: the symptoms may evolve steadily over minutes or hours,

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 79

3.3 The diagnosis of a cerebrovascular event

they may develop in a stuttering/stepwise fashion over several hours, and occasionally they may continue to increase over a few days. In the last case, it can be virtually impossible to decide whether this is due to the primary condition continuing to evolve, or to other factors such as the development of cerebral oedema, infection, or metabolic upset. Not surprisingly, there is no agreed definition of the widely used term ‘stroke-inprogression’ (section 11.5).87,88 Of course, in the face of a progressing deficit, it may not be possible to be absolutely certain whether the diagnosis is actually that of stroke/TIA at all, and the degree of diagnostic uncertainty will influence the extent of the investigations needed to firm up the clinical diagnosis.

3.3.9 Possible precipitants In addition to giving an impression of the suddenness of onset of the symptoms, asking the patients (or witnesses) what they were doing at the time of symptom onset can also support a diagnosis of a vascular event. • Vigorous physical activity and coitus have been associated with haemorrhagic stroke, particularly subarachnoid haemorrhage. However, apart from isolated case reports, there is no evidence that such activities precipitate TIAs and ischaemic stroke. • A change in posture, neck turning (section 7.1.5), exposure to bright or white light, bending, straining, or sneezing, exercise, a hot bath, or a heavy meal, may provoke cerebral and ocular ischaemic symptoms in people with severe carotid and vertebrobasilar occlusive disease and a compromised collateral cerebral and ocular circulation. Of course, some of these stimuli may also provoke non-vascular symptoms, such as those due to hypoglycaemia (after a large carbohydrate meal) and seizures (after exposure to bright flashing lights). • Certain circumstances may predispose to arterial dissection, such as neck manipulation, road traffic accidents and head injuries. There can often be a delay of days or weeks between the trauma and the first neurological symptoms (sections 7.2.1 and 8.2.13). • Drug abuse is increasingly common, it is not confined to the very youngest age groups (section 7.15), and one should therefore have a relatively low threshold for specific questioning and toxicology screening. • Symptoms beginning shortly after starting or increasing the dose of hypotensive drugs should raise the possibility of a so-called ‘low-flow’ stroke/TIA due to the combination of systemic hypotension, focal arterial stenosis/occlusion/compression and poor collateral circulation (section 6.7.5). A similar argument might also apply to symptoms present on

..

..

waking from a general anaesthetic, although other factors such as intravascular thrombosis may be present, particularly if the symptoms are referable to the posterior circulation. Obviously, patients undergoing cardiac or carotid surgery are at special risk, and this should be explained to them prior to surgery (section 7.18). • Pregnancy and the puerperium are times when otherwise healthy young women may be predisposed to stroke as a result of paradoxical embolism from the venous system of the legs or pelvis, intracranial haemorrhage due to eclampsia, or intracranial venous sinus thrombosis (section 7.14). • There is a complex relationship between migraine and cerebrovascular disease, and this is discussed in detail in sections 3.4.1 and 7.8. • There is some evidence that significant ‘life events’ in the preceding year may increase the risk of a stroke (section 6.6.18).

3.3.10 Accompanying symptoms Any accompanying symptoms may be useful in sorting out whether the pathogenesis is vascular. Headache Headache occurs in about one-sixth of patients at the onset of a transient ischaemic attack of the brain or eye, about one-quarter of patients with acute ischaemic stroke, about one-half of patients with non-traumatic intracerebral haemorrhage, and nearly all patients with subarachnoid haemorrhage.89–93 Cortical ischaemia causes headache more often than small, deep, lacunar infarcts (section 6.7.6).89,91,93 Given that ischaemic strokes are four times as common as haemorrhagic strokes, the predictive value of headache for the presence of haemorrhagic stroke is about 33%, while the predictive value of ‘no headache’ for ischaemic stroke is about 86%. The cause of the headache associated with ischaemic events is unknown. It has been suggested that it is due to the release of vasoactive substances such as serotonin and prostaglandins from activated platelets during cerebral ischaemia. Other possibilities include distortion or dilatation of collateral blood vessels, and mechanical stimulation of intracranial nociceptive afferents. Very occasionally, external carotid artery territory emboli can result in ischaemia of the scalp, and so pain. And, some headaches are probably due to anxiety and muscle tension. Headache may not only be a consequence of the vascular event, but also a marker of the underlying cause of the event. Headache in anyone over the age of 50 or so

79

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 80

80

Chapter 3 Is it a vascular event and where is the lesion?

presenting with ischaemia of the brain or eye demands immediate consideration of giant-cell arteritis (section 7.3.1). Similarly, severe pain on one side of the head, face, eye or neck before or around the time of onset is highly suggestive of carotid or vertebral arterial dissection,94 and may even be the only symptom of carotid dissection95 (section 7.2.1). Other causes of headache with focal neurological signs include carotidynia (secondary to carotid occlusion), migrainous stroke, meningitis and intracranial venous thrombosis, but usually there are other clues to these diagnoses.96

Neck stiffness Meningism means painful resistance to passive or voluntary neck flexion because of irritation of the cervical meninges by subarachnoid blood, or by inflammation. Neck stiffness caused by meningism is a common symptom and sign of blood in the subarachnoid space, but it does not occur immediately; it takes some 3–12·h, and may not develop at all in deeply unconscious patients, or in patients with minor SAH (section 3.7.1). Photophobia

Epileptic seizures About 2% of stroke patients have an epileptic seizure (sections 3.4.2, 6.7.6 and 11.8) at the onset; about half are generalized and half are partial seizures.97 These are more common with intracerebral or subarachnoid haemorrhage than with arterial ischaemic stroke.97 They are, however, a characteristic feature of intracranial venous infarction (section 7.21).3 Onset seizures probably indicate irritation or damage to the cerebral cortex by the stroke, and are associated with an increased risk of further seizures. Epileptic seizures may not only complicate stroke but also be a marker of increased risk of stroke. Looking forward, from the time of diagnosis of epilepsy, older people (>60 years of age) diagnosed with new-onset idiopathic epilepsy have an almost three-fold risk of subsequent stroke compared with others of the same age with no history of seizures.98 Looking back, since the time of diagnosis of stroke, about 3% of patients with acute stroke have a past history of epileptic seizures, one-third of these occurring for the first time in the previous year.97 Older people (>60 years of age) diagnosed with new-onset idiopathic epilepsy should therefore be screened for risk factors for stroke and treated appropriately. Vomiting Vomiting is very rare during TIA and is uncommon even in patients with stroke, at least near the onset. When it does occur, it suggests subarachnoid haemorrhage (see section 3.7.1), posterior fossa stroke (because of vertigo in some cases, and presumably because of direct involvement of the ‘vomiting centre’ in the area postrema in the floor of the fourth ventricle in others), or large supratentorial stroke causing raised intracranial pressure. Vomiting within 2·h of stroke onset is highly predictive of intracranial haemorrhage. Occasionally, patients with brainstem ischaemia can have profuse vomiting with little or no vertigo and few other clinical signs.

Patients are often photophobic and irritable for several days after SAH, presumably as a result of meningeal irritation by the blood (section 3.7.1). Hiccups and abdominal pain Hiccups are brief bursts of intense inspiratory activity, involving the diaphragm and inspiratory intercostal muscles, with reciprocal inhibition of the expiratory intercostal muscles. Glottic closure occurs almost immediately after the onset of diaphragmatic contraction, generating the characteristic sound and sense of discomfort. Hiccups usually resolve spontaneously after a few minutes. If they persist for days, they may indicate underlying structural or functional disturbances of the medulla (in the region of the vagal nuclei and tractus solitarius), or of afferent or efferent nerves to the respiratory muscles, or diaphragmatic irritation. Hiccups are well recognized in patients with lateral medullary infarction, but may also occur with a lesion of any part of the medullary region that is associated with respiratory control. Neurogenic hiccup rarely occurs in isolation; usually there are associated brainstem or long tract signs (section 11.10). If patients who have presented with a stroke also complain of abdominal pain, they may have ischaemia of the bowel or viscera – particularly if there are pointers to cardiogenic embolism being the underlying cause of the stroke (such as atrial fibrillation). Chest pain, palpitations and shortness of breath Chest pain and/or palpitations at the time of onset of the stroke/TIA suggest the possible coexistence of an acute coronary syndrome (due, among other things to simultaneous inflammation and rupture of cerebral and coronary atheromatous plaque) and cardiac dysrhythmias (as a consequence of cardiac ischaemia and as a possible source of embolism to the brain) (section 6.5). Other relevant causes of chest pain and/or shortness of

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 81

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

breath include aortic dissection, pulmonary embolism and pneumonia. Panic and anxiety Sudden loss of limb power, speech or vision is a frightening experience which often evokes considerable anxiety and panic in patients and carers. Patients may consequently hyperventilate and in turn develop presyncopal or sensory symptoms including perioral and distal limb paraesthesia bilaterally, and even unilateral sensory symptoms. Under these circumstances, it is important to try to distinguish between the primary (stroke/ TIA) and secondary (panic/anxiety) symptoms. It is also important to be clear about the timing of the symptoms; for example, palpitations occurring immediately before or concurrently with stroke/TIA are less likely to be a consequence of panic and anxiety than similar symptoms that clearly follow the onset of the neurological symptoms.

3.3.11 Past medical history It is important to ask more than once about previous neurological symptoms. Many patients have told us several days after their transient ischaemic attack (TIA) or stroke about prior TIAs, having not recalled them for a number of reasons at the time of their first consultation. Sometimes this is because of anxiety, and of course a few patients are affected by altered awareness or amnesia during the event. Others may not wish to disclose such information for fear of potential repercussions for their employment or driving status. It is the doctor’s responsibility to explore these possibilities with insight, sensitivity and confidentiality, using all available sources of information. Sometimes a detailed history of previous episodes helps confirm an alternative diagnosis such as focal epilepsy or migraine, while a detailed drug history may identify previous use of aspirin, or warfarin given after an otherwise forgotten event. We would recommend working through a checklist of focal neurological symptoms such as those in Table 3.20. It is also important to ask specifically about: • common vascular risk factors (e.g. hypertension, hyperlipidaemia, diabetes, smoking); • other manifestations of vascular disease (e.g. angina, intermittent claudication of the legs); • heart disease (e.g. valvular heart disease or cardiomyopathy – many lay people think a stroke is a form of heart attack); • pointers to a thrombotic tendency (e.g. prior unexplained deep venous thrombosis);

..

..

Table 3.20 Checklist of symptoms of cerebrovascular disease. Have you ever been told that you have had a stroke, ministroke, transient ischaemic attack or brain attack? If so, when did this occur and can you describe what happened? Have you ever suddenly: Lost vision or gone blind in one eye? Had double vision for more than a few seconds? Had jumbled speech, slurred speech or difficulty in talking? Had weakness or loss of feeling in the face, arm or leg? Had clumsiness of an arm or leg? Had unsteadiness walking? Had a spinning (dizzy) sensation? Lost consciousness? How long did the symptoms last? Do you still have these symptoms? Did you see a doctor about the episode and, if so, who was it and what were you told, and were you admitted to hospital? What medications/drugs/tablets are you taking? (particularly aspirin, clopidogrel, dipyridamole, warfarin)

• and clues to vasculitis (e.g. arthralgia, skin rashes, renal problems (sections 6.7.6, 6.7.7 and 7.3).

3.3.12 Lifestyle, behaviour and family history Relevant and potentially modifiable lifestyle factors include tobacco, alcohol, saturated fat, salt, recreational drug consumption and physical activity. A family history of stroke or myocardial infarction is a risk factor for ischaemic stroke, which is attributable, at least partly, to a familial predisposition to hypertension (section 6.6.3).99,100

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

There is a wide differential diagnosis of focal cerebral symptoms of rapid onset of short duration (Table 3.21) and long duration (Table 3.23). Tables 3.22 and 3.24 show the frequency of these possibilities in practice. Because there is a continuum of the duration of neurological symptoms caused by focal neurological ischaemia,7 patients who are being assessed within 24 hours of symptom onset and who still have focal neurological symptoms are temporarily classified as having a ‘brain attack’ (‘acute stroke syndrome’ or ‘unstable brain ischaemia’; section 3.2.3). Another way of considering

81

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 82

82

Chapter 3 Is it a vascular event and where is the lesion? Table 3.21 Differential diagnosis of transient ischaemic attack (in approximate order of frequency, depending on referral patterns). Syncope Migraine aura (with or without headache) Labyrinthine disorders Partial (focal) epileptic seizures Hyperventilation, anxiety or panic attacks, somatization disorder Intracranial structural lesion: meningioma tumour giant aneurysm arteriovenous malformation chronic subdural haematoma Transient global amnesia Acute demyelination (multiple sclerosis) Drop attacks Metabolic disorders: hypoglycaemia, hyperglycaemia, hypercalcaemia, hyponatraemia Mononeuropathy/radiculopathy Myasthenia gravis Cataplexy

Table 3.22 Final diagnosis of all 512 patients with suspected transient ischaemic attacks notified to the Oxfordshire Community Stroke Project, modified from reference 6. Confirmed transient ischaemic attacks: Incident (i.e. first-ever) Prevalent (i.e. had previous attacks) Lone bilateral blindness* Not transient ischaemic attacks: Migraine Syncope Possible TIA† ‘Funny turn’‡ Isolated vertigo Epilepsy Transient global amnesia Isolated diplopia Drop attack Intracranial meningioma Miscellaneous

209 184 11 14 303 52 48 46 45 33 29 17 4 3 2 24

*Lone bilateral blindness was later classified as a transient ischaemic attack after following up these patients and noting their similar prognosis to patients with definite transient ischaemic attack.48 †Possible transient ischaemic attack was diagnosed in patients with transient focal neurological symptoms in whom the clinical features were not sufficiently clear to make a diagnosis of definite transient ischaemic attack or of anything else. ‡‘Funny turn’ was used to describe transient episodes of only non-focal symptoms not due to any identifiable condition (e.g. isolated and transient confusion).

Table 3.23 Differential diagnosis of stroke (in approximate order of frequency, depending on referral patterns). Systemic illness, or seizure, causing apparent deterioration of previous stroke Epileptic seizure (postictal Todd’s paresis) or non-convulsive seizures Structural intracranial lesion: subdural haematoma brain tumour arteriovenous malformation Metabolic/toxic encephalopathy: hypoglycaemia non-ketotic hyperglycaemia hyponatraemia Wernicke–Korsakoff syndrome hepatic encephalopathy alcohol and drug intoxication septicaemia Functional/non-neurological (e.g. hysteria) Hemiplegic migraine Encephalitis (e.g. herpes simplex virus)/brain abscess Head injury Peripheral nerve lesion(s) Hypertensive encephalopathy Multiple sclerosis Creutzfeldt–Jakob disease Wilson’s disease

the differential diagnoses according to duration of symptoms is outlined in Table 3.25. The top five differential diagnoses of brain attack (TIA and stroke) are the ‘5 Ss’: seizures, syncope, sepsis, subdural haematoma and somatization. The clinical diagnosis of stroke (differentiating ‘stroke’ from ‘not stroke’) is accurate more than 95% of the time if there is a clear history (from the patient or carer) of focal brain dysfunction of sudden onset (or first noticed on waking), with symptoms persisting for more than 24·h. This is particularly true if the patient is elderly or has other vascular diseases or risk factors, because the risk – i.e. the prior (or pretest) probability of stroke – is greater in elderly people with vascular disease than in younger people who have no evidence of vascular disease. However, the accuracy of the diagnosis of stroke is also influenced by the timing of the assessment, the experience and confidence of the examiner, and the likelihood of other differential diagnoses in the community37,41,101–103 (Table 3.24). For example, the diagnosis of stroke (vs not stroke) can be quite difficult within the first few hours of symptom onset if it is not possible to obtain a clear history of the onset, or even the nature of the symptoms, because the patient is unconscious, confused, forgetful or dysphasic. In these patients in

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 83

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset Table 3.24 Final diagnosis in two recent studies of patients presenting with suspected TIA and stroke.44,45

Nor et al.,45 (n = 59)

Stroke ‘mimic’

Seizure Sepsis Metabolic/toxic encephalopathy Space-occupying brain lesion Tumour Subdural haematoma Syncope/presyncope Acute confusional state Vestibular dysfunction (labyrinthitis) Acute mononeuropathy Functional/somatization Dementia Migraine Spinal cord lesion Other

particular, a persistent focal neurological deficit may not be due to stroke but to epileptic seizures (postictal), sepsis, encephalitis, brain abscess, brain tumour, head trauma, or a chronic subdural haematoma.102 Symptoms or signs that are unusual in uncomplicated stroke, such as papilloedema and unexplained fever, should also call into question the diagnosis. Further, a persistent focal neurological deficit may be due to a previous stroke, and the new clinical presentation may be caused by a non-vascular problem such as pneumonia (section 11.12). In the absence of information about the rate of onset and progression of symptoms, it is helpful to search for indirect clues to their cause in the past history and physical examination, and to continue to assess the patient over time for the development of new signs such as fever, and for the improvement that characterizes most non-fatal strokes. The diagnosis of a cerebrovascular event is usually made at the bedside, not in the laboratory or in the radiology department. It depends on the history of the sudden onset of focal neurological symptoms in the appropriate clinical setting (usually an older patient with vascular risk factors) and the exclusion of other conditions that can present in a similar way.

3.4.1–Migraine Transient ischaemic attack Migraines are not considered to be TIAs, because the clinical features (Table 3.26) and prognosis for serious vascular events are very different.104 Classical migraine (migraine with aura) usually starts in younger patients who may have a family history of migraine. The attack begins with an aura that commonly consists of positive

..

..

8 (14%) 8 (14%) 7 (12%) 4 (7%) 3 (5%) 13 (22%) 3 (5%) 7 (12%)

13 (22%)

Hand et al.,44 (n = 109) 23 (21%) 14 (13%) 12 (11%) 10 (9%)

10 7 7 6 6 4 3 3 3

(9%) (6%) (6%) (5%) (5%) (4%) (3%) (3%) (3%)

symptoms of focal cerebral dysfunction that develop gradually over 5–20·min and last less than 60 min.86,105 The most common aura consists of homonymous, unilateral or central visual symptoms, such as flashes of light, zigzag lines, crescents, scintillations, or fortification spectra, which gradually ‘build up’, expand and migrate across the visual field. Somatosensory or motor disturbances, such as paraesthesiae or heaviness in one or more limbs, may also occur, evolving and spreading over a period of minutes in a ‘marching’ fashion (e.g. spread of tingling from hand to arm to face to tongue over several minutes). This serial progression from one accompaniment to another without delay, such as from visual symptoms to paraesthesiae and then to aphasia, is quite characteristic of migraine but can occur with cerebral ischaemia.106 Sometimes, however, the symptoms are negative and consist of ‘blind patches’ (often a homonymous hemianopia) and, rarely, even loss of colour vision. Headache and nausea usually follow the focal neurological symptoms immediately or in less than an hour, but in some patients they precede the neurological symptoms and in others they occur simultaneously. Often there is associated photophobia and phonophobia, which clearly help distinguish migraine attacks from TIAs. The headache usually lasts 4–72 h.86 It is not unusual for patients who have experienced classical migraine to suffer identical auras, but without headache (acephalgic migraine aura), particularly as they get older (otherwise referred to as ‘migraine aura without headache’ or ‘late-life migraine accompaniments’).107 This should not cause confusion with TIA. However, diagnostic difficulty arises when an older patient (over 40 years) with no previous history of classical migraine presents following a first-ever episode of transient symptoms of focal neurological dysfunction that are typical of a migraine aura, but without any associated headache.

83

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 84

84

Chapter 3 Is it a vascular event and where is the lesion? Table 3.25 Differential diagnosis of sudden focal neurological symptoms by duration of symptoms.

Migraine Epilepsy Partial seizures Todd’s paresis Transient global amnesia Structural intracranial lesions Subdural haematoma Tumour Arteriovenous malformation/aneurysm Metabolic/toxic disorders Hypoglycaemia Hyperglycaemia Hyponatraemia Hypercalcaemia Hepatic encephalopathy Wernicke’s encephalopathy Hypertensive encephalopathy Posterior reversible encephalopathy syndrome CNS infections Encephalitis Brain abscess Subdural empyema Creutzfeldt–Jakob disease Progressive multifocal leukoencephalopathy Labyrinthine disorders Vestibular neuronitis Ménière’s disease Benign paroxysmal positional vertigo Benign recurrent vertigo Psychological disorders Hyperventilation Panic attacks Somatization/conversion disorder Head injury Multiple sclerosis Neuromuscular disorders Mononeuropathy Radiculopathy Myasthenia gravis Motor neurone disease

Transient ischaemic attack

Brain attack

Stroke

+++

++

+

+++ +++ +++

++ ++ ++

+ + –

++ + ++

++ ++ ++

++ ++ +

+++ +++ ++ ++ + + + +

++ ++ ++ ++ ++ ++ ++ ++

+ + + + ++ ++ ++ ++

+ + + – –

++ ++ ++ – –

++ ++ ++ + +

+ ++ ++ ++

++ – – +

++ – – –

+++ +++ +++ + +

– – ++ + +

– ++ ++ ++

++ ++ ++ –

++ ++ ++ –

++ ++ + +

+++ Common or frequent; ++ encountered regularly in a busy practice or important, treatable condition; + infrequent; – rare.

Patients with these ‘late-life migraine accompaniments’ have a favourable prognosis with a lower risk of stroke and other serious vascular events compared with TIA patients.104 A 69-year-old former general practitioner was referred by a cardiologist for suspected ocular TIAs, a diagnosis that caused him some alarm. He had unexpectedly suffered a

myocardial infarction at the age of 59, and had been receiving beta-blockers and aspirin since then. The attacks in question, of which he had now had nine, had first occurred more than 2 years previously, but had increased in frequency. Each attack began with a small, bright spot on the left or right side of the centre of vision. He verified that it occurred in both eyes by covering each eye in turn. In the course of 1–3·min, the bright spot would gradually enlarge and change

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 85

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset Table 3.26 Features of migraine aura that help distinguish it from TIA. Gradual onset over minutes (usually) Positive neurological symptoms such as visual scintillations, not blindness Intensification and ’spread’ of symptoms from point to point, or from one symptom to another symptom, over minutes Gradual resolution of symptoms over 20–60 min Headache (often unilateral and pulsatile) and nausea usually follow (but not always) Onset in young and middle-aged adults Family history of migraine is common Recurrent attacks are usually stereotyped

Fig. 3.34 A drawing by a physician who experienced ‘late-life migraine accompaniments’. He wrote the following caption: ‘These zigzag lines appear during an attack in both eyes, more often to the left than to the right of centre of vision, sometimes on both sides. The lines flicker on and off, about six times per second; they are colourless. Sometimes a dazzling white spot will appear within a line, which also turns on and off.’

into a cloudy area, until the centre of vision became obscured; the border of the opaque zone consisted of intensely bright and flickering shapes: zigzag lines, stars and sparks (Fig. 3.34). This phenomenon would remain for about 15·min; if he tried to read, he could not make out the words in the centre. Finally, the visual disturbance would disappear over a few minutes. He never had headache or other symptoms with the attacks, and was able to resume his normal activities afterwards. He could not recall that he had suffered similar attacks, or migraine, earlier in his life. This was migraine, not TIA. A detailed history eliciting the slow onset, and spread and intensification of neurological symptoms – particularly if positive and visual or referable to more

..

..

than one vascular territory during the same attack – points towards the diagnosis of migraine.86,104 Careful questioning about childhood or menstruation-related headaches, and a family history of migraine, may suggest a predisposition to migraine. If the patient is young (i.e. less than about 40 years old), with a normal heart and no other clinical manifestations of arterial disease, it is most unlikely that the symptoms are those of a TIA caused by atherothrombosis. Perhaps this explains why young female patients with suspected TIA, with or without a family history of migraine, do not seem to have any higher risk of stroke than other women of the same age. In this group, there is a remote possibility of thrombophilic and vasculitic disorders but, for the most part, these patients can be reassured that they have not ‘had a stroke’. Because the risks of stroke are very low in absolute terms in people with migraine, they do not require any potentially hazardous or costly investigations or prophylactic treatments108,109 (section 7.8). There is no convincing evidence that arteriovenous malformations or aneurysms occur more frequently than can be accounted for by chance in patients with migraine aura. There can, however, be particular concern about migraine occurring for the first time during pregnancy, because of the recognized propensity for meningiomas and possibly arteriovenous malformations to become symptomatic at this time.110 In this group of patients, we would suggest careful clinical examination looking for residual focal signs following a single attack, and a low threshold for MR scanning after the baby has been born if the attacks are multiple and stereotyped, particularly if the same side of the brain is always involved. In older patients, particularly those with their first attack of acephalgic migraine and if the neurological symptoms are other than positive visual ones, the distinction from TIAs can be much more difficult. From a purely pragmatic point of view, if there is a suspicion that the attack was a TIA, it seems prudent to modify any vascular risk factors and recommend aspirin. Although drugs used to lower blood pressure and cholesterol, and antiplatelet drugs such as aspirin, are not free of risk or expense, some of them (e.g. propranolol and aspirin) may also prevent migraines. Stroke Occasionally, the aura of a previously experienced and otherwise unremarkable attack of migraine persists for days or longer – a so-called migrainous stroke (section 7.8.1). The simultaneous occurrence of a stroke and a migraine attack may be: • coincidental because, after all, both conditions are common, the prevalence of migraine in the general

85

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 86

86

Chapter 3 Is it a vascular event and where is the lesion?

adult white population is about 10% and ischaemic cerebrovascular disease about 0.8%; • causal, either migraine may predispose to cerebral ischaemia by leading to platelet activation, arteriolar constriction and dehydration, or cerebral ischaemia may trigger a migraine attack; • a misdiagnosis (e.g. arterial dissection may cause headache and a neurological deficit due to thromboembolism that is misinterpreted as migraine); or • both are a manifestation of a disorder such as a patent foramen ovale, an arteriovenous malformation, or a mitochondrial disorder111–113 (sections 6.5.12, 7.19 and 8.2.4). The 69-year-old wife of a general practitioner had suffered from classical migraine since she was a teenager. Her attacks always began with visual fortification spectra, progressing to a hemianopia (either left or right) over the next 10·min. The visual disturbance would resolve after about 30·min, and she would then get a severe throbbing headache, nausea and vomiting. She had always recognized bright light as a potential trigger factor. Because of a family history of glaucoma, an optician recommended that she have her visual fields tested using a Humphrey perimeter. She found the flashes of white light uncomfortable, and by the end of the examination she was aware that she had her typical migrainous visual disturbance in her right visual field. The headache was unusually severe and she went straight to bed. When she woke 6·h later, the headache had resolved but the right hemianopia was still present. A brain CT scan later that day showed a recent left occipital infarct. No other explanation for the stroke was found despite detailed investigation. The hemianopia persisted, and it was presumed that this had been a migrainous stroke.

3.4.2 Epilepsy Transient ischaemic attack Partial (focal) seizures can be distinguished from TIAs because they usually cause sudden positive sensory or motor symptoms that spread or ‘march’ fairly quickly to adjacent body parts (Table 3.27).52 Although positive symptoms such as tingling and limb jerking can occur in TIAs,76 they tend to arise in all affected body parts at the same time (i.e. in the face, arm and leg together), whereas the symptoms of partial seizures spread from one body part to another over a minute or so (cf. migraine aura, where any spread is usually over several minutes; section 3.4.1). Negative motor symptoms, such as postictal or Todd’s paresis, are well recognized after a partial motor seizure or a generalized seizure with partial onset, but this should be obvious from the history unless

Table 3.27 Features of partial epileptic seizures that help distinguish them from TIAs. Young and middle-aged adults commonly, but also patients with previous cortical stroke Antecedent symptoms may occur (e.g. epigastric discomfort) which are the beginning of the seizure Onset over several seconds or 1–2 min Positive neurological symptoms usually March or ‘spread’ of symptoms over seconds Rapid resolution of symptoms over a few minutes Amnesia for the event Family history of epilepsy may be present Recurrent attacks are usually stereotyped and reduced with antiepileptic treatment (if necessary)

the patient was asleep, or is aphasic and there is no witness (section 3.3.10). Difficulty arises in the very rare patient in whom epileptic seizures actually cause transient ‘negative’ symptoms during the electrical discharge. One then has to rely on other factors, such as the patient’s age, any past history of seizures and the nature of the symptoms. For example, transient speech arrest (as opposed to aphasia with muddled language output), which is characterized by the sudden onset of cessation of speech, often accompanied by aimless staring and subsequent amnesia for the details of the episode, is usually epileptic rather than ischaemic. Transient aphasia and rarely bilateral blindness or amnesia can also sometimes be epileptic. Distinguishing between seizures and TIAs is occasionally very difficult.76 Sometimes it requires prolonged and careful observation of, and interaction with, the patient (and witnesses) over several visits. The diagnosis should not be rushed, as patients may have several types of attack with different causes. Initially, it is important to explain the diagnostic uncertainty to the patient and why it is necessary to establish the precise diagnosis; to exclude a structural intracranial lesion (with brain imaging); and to advise the patient not to drive or put themselves in a position in which they would be a danger to themselves or others if they were to have another attack. Clues to the diagnosis of epilepsy can be obtained by taking a careful, targeted history, particularly with respect to previous epileptic seizures and any symptoms that immediately preceded the onset of the neurological deficit (e.g. epigastric discomfort and olfactory or gustatory hallucinations may be due to an initial partial sensory seizure involving the temporal or frontal lobes). The interictal electroencephalogram (EEG) can be normal but not usually in patients having several seizures a day. Ambulatory or telemetered EEG requires skilled and careful analysis because of the potential for a false

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 87

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

positive diagnosis of epilepsy. Nevertheless, concomitant video recording of an attack can be very useful. If recurrent partial seizures are suspected, brain MR should be performed to seek a focal structural lesion too small to be seen on CT, such as mesial temporal sclerosis, a hamartoma, or a tumour, usually in the frontal or temporal lobe. However, transient abnormalities that might be mistaken for areas of ischaemia on brain MR have been reported in patients with partial status epilepticus, although the abnormalities tend not to respect normal vascular distributions and to resolve with time. A 64-year-old woman described about 20 attacks of pins and needles in her right arm and leg over a period of 6 weeks. Each attack lasted for about 5·min, and there were no associated symptoms. On closer interrogation, she said that the sensation started in the right foot and then over a period of about 1·min spread ‘like water running up my leg’ to involve the whole leg and arm. Each attack was identical. A brain CT scan showed a glioma in the left parietal lobe. A diagnosis of partial sensory seizures secondary to the glioma was made. Stroke An epileptic seizure is one of the most common reasons for the misdiagnosis of stroke.41,44,45,103 The usual scenario is a patient with postictal confusion, stupor, coma or hemiparesis (Todd’s paresis) where the preceding seizure was unwitnessed or unrecognized.103 Todd’s paresis refers to a focal deficit that may follow a focal motor seizure. In general, the deficit usually resolves very quickly (within minutes), but in patients with prior cerebral damage, e.g. an earlier stroke, the deficit can last for several days, or there may be permanent deterioration, making the distinction from a recurrent stroke very difficult. It is noteworthy that involuntary convulsive-like movements sometimes occur in patients with brainstem strokes.114 These movements vary in nature, frequency and triggers, including fasciculation-like, shivering, jerky, tonic-clonic and intermittent shaking movements. The episodes usually consist of brief clonic contractions of the proximal and distal upper extremities and occur in paroxysms lasting for 3–5 s.114 Their pathogenesis is uncertain but may be related to ischaemia or haemorrhage in the corticospinal tracts, They should not be misinterpreted as decerebrate postures or seizures.

3.4.3–Transient global amnesia Transient global amnesia (TGA) is not considered to be a TIA because its prognosis is so much better. It is a very characteristic clinical syndrome that typically occurs in a middle-aged or elderly person.115 There is a sudden dis-

..

..

Table 3.28 Features of transient global amnesia that help distinguish it from TIAs. Older adults Sudden onset Anterograde and retrograde memory disturbance Repetitive questioning Able to recognize familiar individuals and places with no loss of personal identity Headache can occur No focal neurological symptoms or signs All but complete recovery within a few hours to a day or so, leaving a dense amnesic gap for the attack itself Recurrent attacks are unusual, and if they occur may be due to migraine or partial seizures Diagnosis is very difficult without a witness

order of memory, which is often reported as confusion.116 For some hours, the patient cannot remember anything new (anterograde amnesia) and often cannot recall more distant events over the past weeks or years (retrograde amnesia). There is no loss of personal identity, personality, problem-solving, language or visuospatial function, and the patient can perform complex activities such as driving a car (Table 3.28). The patient seems healthy, but repetitively asks the same questions and has to be reminded continually of what he or she has just asked or done. There are no other symptoms, apart from perhaps headache (migraine is more common in TGA patients than in controls, and there are some theoretical reasons to implicate migraine in its pathogenesis).117 In women, episodes are mainly associated with an emotional precipitating event, a history of anxiety and a pathological personality. In men, they occur more frequently after a physical precipitating event, e.g. diving into a swimming pool.115 After the attack, anterograde memory returns to normal, but the patient cannot remember anything that happened during the amnesic period. The retrograde amnesia tends to diminish with recovery, but leaves a short retrograde gap for the period of the TGA. Recurrences are not very common, about 3% per year. The early reports tended to view TGA as a type of TIA, and of course one does encounter patients with pure amnestic strokes (section 3.3.3). However, careful case–control studies have shown that both the prevalence of vascular risk factors and the rates of subsequent stroke or myocardial infarction are much lower in the TGA group and are indeed approximately similar to those in the age-matched general population.115,118,119 There is no overall difference in the prevalence of epilepsy between TGA patients and control groups, but an important minority (7%) of TGA patients go on to

87

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 88

88

Chapter 3 Is it a vascular event and where is the lesion?

develop epilepsy, usually within a year of presentation. These patients tend to have had shorter attacks of ‘TGA’, lasting less than 1·h, and to have already experienced more than one attack at the time of presentation. It must be presumed that in this minority of cases with ‘TGA’, the cause was temporal lobe epileptic seizures from the beginning. Because of this, driving is generally allowed after a single episode of TGA but not after multiple attacks. Functional imaging, such as perfusion CT, has shown transient bilateral hypoperfusion of the medial temporal lobes or basal ganglia, and unilateral left opercular and insular hypoperfusion, during the TGA attack.120 However, it is not known whether this perfusion defect is a primary event or secondary to a reduction in cerebral metabolic activity in these areas, or due to other factors. Diffusion-weighted MRI during episodes of TGA shows transient signal abnormalities, consistent with oedema, in the uncal-hippocampal region unilaterally or bilaterally in a variable proportion of patients.119,121,122 Ischaemia, due to arterial thromboemboli or venous stasis, has been proposed as a potential mechanism.119,123,124 It has also been hypothesized that diverse stressful precipitants may trigger a release of an excitotoxic neurotransmitter (such as glutamate), which then temporarily shuts down normal memory function in the medial temporal regions via spreading depression, leading in turn to a fall in cerebral perfusion.

Table 3.29 Possible explanations for transient neurological symptoms in patients with a structural intracranial lesion. Partial epileptic seizure Spreading depression of Leao (as some suppose occurs in migraine) Vascular ‘steal’, leading to focal brain ischaemia adjacent to the tumour Vessel encasement or direct compression by the tumour mass Indirect compression of vessels by herniating tissue or coning (usually a preterminal event)

ischaemic stroke, if CT brain imaging is not done early, and certainly within 10–14 days after the onset of symptoms by which time the characteristic changes will have resolved (section 5.4.1).13 Fortunately, intracerebral haemorrhage is visible indefinitely on gradient echo MRI sequences as a low signal (black) ring or dot (section 5.5.1). A 77-year-old man, whilst standing in his garden, suddenly developed weakness of the right arm and unsteadiness on walking. He had no headache or vomiting. He sat down and within 3·h was ‘better’, although on closer questioning it emerged that his arm had not returned to normal for about 3 days. A CT brain scan 8 days later showed a small resolving haemorrhage in the left putamen. This was not a TIA in either its duration or pathology.

3.4.4 Structural intracranial lesions Although structural intracranial lesions often cause focal brain dysfunction, the symptoms and signs generally progress over several days or weeks, or even months, they are not abrupt like strokes or TIAs. About 5–10% of patients who are initially diagnosed as having an acute stroke on the basis of the history and examination turn out to have a structural intracranial lesion on brain imaging, such as a subdural haematoma, tumour or arteriovenous malformation.44,45 Possible explanations for transient neurological symptoms are listed in Table 3.29. Non-traumatic intracerebral haemorrhage Non-traumatic intracerebral haemorrhage almost invariably causes prolonged or permanent focal neurological dysfunction, although there are isolated reports of neurological deficits that have resolved within a few days, and even 24·h. However, one sometimes needs to probe quite hard to be sure about the duration of symptoms from the history. For example, ‘I got better in 24·h’ does not necessarily mean ‘I got completely back to normal’, which is what we are referring to when defining a TIA. Intracerebral haemorrhage can be minsterpreted as TIA, or

Subdural haematoma Transient focal neurological symptoms such as hemiparesis, hemisensory loss, aphasia and speech arrest may occur as a consequence of a chronic subdural haematoma (SDH)125 (section 8.10). And SDH may rarely present with the abrupt onset of focal neurological signs, which then persist and so mimic a stroke.45 Usually the patient has a confusional state and a history of minor neurological symptoms starting a few days to a few weeks previously. SDH should be suspected if there is any evidence of subacute onset of focal neurological symptoms and signs; persistent headache; more confusion and drowsiness than expected from the neurological deficit; or a progressive or fluctuating clinical course. Although SDH occurs in all age groups, it is more frequent among the elderly, alcoholics and patients receiving anticoagulants or who have a bleeding disorder. About 50% of patients recall sustaining a head injury, which may have been mild. SDH is a rare complication of lumbar puncture, and spontaneous or traumatic intracranial hypotension. In the acute phase, the brain CT scan usually shows unilateral hyperdensity in the subdural space, ipsilateral

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 89

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

effacement of sulci, and a mass effect causing shift of the midline and distortion of the ventricular system (Fig. 3.35). There is a transitional stage between 7 and 21 days, during which clotted blood evolves on the CT scan from a region of hyperintensity to one of isointensity. This can easily be missed, particularly if the subdural haematomas are bilateral, when there is little if any midline shift or asymmetrical ventricular compression. Thereafter, the haematoma becomes hypodense and thus more easily visible. MRI is more sensitive than CT for detecting subdural haematomas. In a few cases, a subdural haematoma may prove to be an incidental finding, and the TIA can be attributed to more commonly recognized factors such as carotid artery disease. Tumour Occasionally – perhaps if there is bleeding into a tumour – there can be a sudden focal neurological deficit caused by a brain tumour, although this usually lasts longer than 24·h.126,127 Greater diagnostic difficulty arises when a structural lesion causes a partial and non-convulsive seizure, with or without postictal (Todd’s) paresis, or intermittent focal neurological symptoms (so-called ‘tumour attacks’) that do not seem to be epileptic in origin.44,45,49 The clinical features associated with these ‘tumour attacks’ are usually focal jerking or shaking, pure sensory phenomena, loss of consciousness and isolated aphasia or speech arrest.126 With time, a more obvious epileptic syndrome often declares itself. A 78-year-old woman complained of many attacks of weakness and clumsiness of the left arm over a period of 4 months. The weakness came on suddenly, lasted for between 10 and 45·min, and was not associated with any other symptoms. In between attacks, she had no symptoms. A diagnosis of transient ischaemic attack was made by both her general practitioner and the neurologist, and she was started on aspirin. A brain CT scan was performed later because the attacks continued, and this showed a meningioma involving the right frontal lobe (Fig. 3.36). A final diagnosis of ‘tumour attacks’ was made, but in view of the patient’s age, the neurosurgeon thought an operation was not advisable. Of course it is always important to consider intracranial tumours in the differential diagnosis of a patient with a progressing neurological deficit, particularly if the rate of progression is relatively slow (over days, weeks or months) and there is a history of recent headache or epileptic seizures, papilloedema, or any evidence of a primary extracranial source of malignancy. In practice, most patients with suspected stroke who turn out to have a cerebral tumour have actually had a focal

..

..

neurological deficit (e.g. hemiparesis) gradually evolving over several weeks to several months. Papilloedema is very uncommon in acute stroke. The brain tumours that tend to bleed are glioblastoma, choroid plexus papilloma, meningioma, neuroblastoma, melanoma, hypernephroma, lymphoma, endometrial carcincoma, bronchial carcinoma and choriocarcinoma (section 8.5.1). The CT scan may then disclose intracerebral haemorrhage in an unusual location or associated with considerable surrounding oedema, or multiple haemorrhages, or it may show other metastatic deposits (Fig. 3.37). If there has been no haemorrhage into the tumour, the CT scan usually shows a region of low attenuation (due to cerebral oedema) with imprecise boundaries and some mass effect, causing effacement of sulci or ventricular compression. If there is a breakdown of the blood–brain barrier, as commonly occurs in patients with cerebral tumours, then intravenous injection of iodinated contrast material leaks into the tumour and is seen on CT as an area of diffuse or peripheral enhancement (Fig. 3.38). A similar, but nevertheless distinctive, appearance of enhancement of the gyri, which may be seen within 1–2 weeks of a recent cerebral infarct, is also due to breakdown of the blood–brain barrier (Fig. 3.39). This can make interpretation difficult, particularly if only a contrast-enhanced scan is performed. If the clinical examination and CT are ambiguous, patients should be followed up clinically (because with a tumour they usually deteriorate), and a follow-up CT or MRI performed within a few weeks to a few months (depending on the patient’s progress) to see if the lesion has resolved or, if it is a tumour, continued to grow. Aneurysms and arteriovenous malformations Intracranial aneurysms and cerebral arteriovenous malformations (sections 8.2.3, 8.2.4 and 9.1.1), and particularly cavernous malformations (section 8.2.5), can cause transient focal neurological deficits mimicking a TIA or ischaemic stroke, and even transient monocular blindness.128,129 In the Scottish Intravascular Vascular Malformation Study, focal neurological deficits accounted for about 7% of presentations of symptomatic arteriovenous malformations (Rustam Al-Shahi Salman, personal communication). Possible explanations are embolization of thrombus from within an aneurysm (section 7.6), a partial seizure, a small intraparenchymal haemorrhage (e.g. from a cavernous malformation) and venous hypertension. Vascular steal around an arteriovenous malformation is not a certain mechanism. In many cavernoma cases it is impossible to identify the cause of some

89

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 90

90

Chapter 3 Is it a vascular event and where is the lesion?

(b) (a)

(c)

Fig. 3.35 (a) A plain CT brain scan showing acute right-sided subdural haematoma as an area of high density in the subural space (arrows). (b) A plain CT brain scan showing a subacute left frontal subdural haematoma (arrows) which is isodense, illustrating the difficulty of identifying subdural haematomas on CT at this stage of their evolution. (c) A plain CT brain scan showing chronic bilateral subdural haematomas as areas of low density in the subural space (arrows).

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 91

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

Fig. 3.36 A post-contrast CT brain scan showing a meningioma involving the posterior part of the right frontal lobe (arrow) in a patient who presented with a history suggesting transient ischaemic attacks; over 3 months, this elderly lady had eight attacks of sudden onset of weakness affecting the left arm, each lasting a few minutes. Between attacks the neurological examination was normal.

(a)

(b) Fig. 3.37 A plain CT brain scan showing multiple areas of high density (arrows) due to spontaneous intracerebral haemorrhage in what turned out at postmortem to be metastases from choriocarcinoma.

..

..

Fig. 3.38 A plain CT brain scan of a brain tumour showing (a) a region of low attenuation (due to cerebral oedema) with imprecise boundaries (arrows) and some mass effect, causing effacement of sulci. (b) After intravenous contrast injection there is enhancement (arrow) of the tumour.

91

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 92

92

Chapter 3 Is it a vascular event and where is the lesion?

haemorrhagic telangiectasia, a past history of subarachnoid haemorrhage, or, reputedly, a cranial bruit.128

3.4.5–Metabolic and toxic disorders Encephalopathies due to metabolic or toxic disturbances generally present with epileptic seizures or a subacute alteration in consciousness, and few if any focal neurological signs (perhaps only generalized hyperreflexia, with or without extensor plantar responses). However, occasionally the presentation may be acute, with focal neurological symptoms and signs, which can therefore mimic a stroke or TIA.44,45 The causes of the metabolic encephalopathy include hypoglycaemia, hyperglycaemia, hyponatraemia, hypercapnoea and hypoxia. Hypoglycaemia

Fig. 3.39 A post-contrast CT brain scan 8 days after onset of ischaemic stroke showing a region of low attenuation consistent with infarction (white arrowheads) in the territory of the left middle cerebral artery, and a serpiginous gyral pattern of high attenuation due to breakdown of the blood–brain barrier (black arrows). There is considerable mass effect, with displacement of midline structures (open arrows), effacement of the lateral ventricle and obliteration of sulci.

transient focal neurological attacks which do not sound epileptic, and where early imaging shows no fresh blood or is done too late (or is not done at all in known cases); a psychogenic cause would not be surprising given the stress of knowing there is an intracranial lesion which might bleed. Although aneurysms and arteriovenous malformations are one of the more common causes of intracerebral haemorrhage (sections 8.2.3 and 8.2.4), they may also cause focal neurological symptoms and signs as a direct consequence of their mass effect and compression of surrounding structures, e.g. a third cranial nerve palsy from a posterior communicating artery aneurysm. Brain CT, with and without contrast, MRI or intraarterial cerebral angiography, may be required to make the diagnosis. Just occasionally, arteriovenous malformations can be suspected clinically, e.g. if there is cutaneous evidence of the Sturge–Weber syndrome or hereditary

Hypoglycaemia may cause both transient and permanent focal neurological symptoms and signs, which can occur without any of the characteristic adrenergic symptoms41,103,130,131 (section 7.16). The patient is almost always being treated with hypoglycaemic agents, but other possibilities include factitious hypoglycaemia, insulinoma, Addison’s disease, hypopituitarism, hypothyroidism, sepsis, terminal malignancy, liver failure, starvation, or drugs whose adverse effects include hypoglycaemia. The symptoms tend to be stereotyped in an individual, and are most likely to occur before meals (i.e. before breakfast or during the night, after fasting for some time), after exercise, or 2–3·h after the ingestion of sugars and starch; they are relieved by glucose administration. The blood glucose is usually less than 2.5·mmol/L at the onset of an attack, but it may have normalized spontaneously or with glucose administration by the time the patient is seen. Particularly if a diabetic patient presents with a suspected stroke early in the morning, it is imperative that hypoglycaemia be considered and appropriate treatment given rapidly, although a blood glucose estimation should be mandatory for all patients with suspected stroke. A 63-year-old man was admitted to a surgical ward because of abdominal pain, diarrhoea and signs of intestinal obstruction, 2 months after a partial gastrectomy for a gastric ulcer. He was treated with intravenous fluids and received nothing by mouth, whilst awaiting investigations. Five days after admission, he became restless and confused. Two hours later, left-sided weakness was noted. The neurology resident found a left hemiplegia and hypoaesthesia, including the face, with completely normal consciousness. The provisional diagnosis was ischaemic stroke, but the next morning the nurses found him unresponsive. On repeat examination,

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 93

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

there was no eye opening to pain, extension of the arm was the best motor response, and there was no verbal response. The pupils were normal, but no oculocephalic responses could be elicited. The glucose level was 1.5·mmol/L. After administration of 100·mL of 50% glucose, he quickly recovered. The blood glucose concentration should be assessed in all patients with suspected stroke. Hyperglycaemia

Wernicke’s encephalopathy

The hyperosmolarity of hyperglycaemia can itself cause regional reduction of cerebral blood flow, focal neurological deficits, focal epilepsy, stroke-like syndromes and cerebral infarction132 (section 7.16). These deficits usually resolve as the blood glucose returns to normal.

Wernicke’s encephalopathy can sometimes be mistaken for a stroke because of an unusually sudden onset of diplopia (due to abducens and conjugate gaze palsies and nystagmus), ataxia and mental confusion, either singly or, more often, in various combinations.134 It is due to thiamine deficiency and is seen mainly, although not exclusively, in alcoholics and the malnourished elderly. The diagnosis can be difficult because the history of symptom onset may be unclear due to recent alcohol intoxication or Korsakoff’s psychosis, in which retentive memory is impaired out of all proportion to other cognitive functions in an otherwise alert and responsive patient. Pointers to the diagnosis on examination are signs of peripheral neuropathy (present in more than 80% of patients), postural hypotension (autonomic neuropathy), disordered cardiovascular function (tachycardia, exertional dyspnoea, minor ECG abnormalities), and impaired capacity to discriminate between odours (in the chronic stage of the disease due to a lesion of the medial dorsal nucleus of the thalamus). The diagnosis is supported by a marked reduction in red cell transketolase activity (one of the enzymes of the hexose monophosphate shunt that requires thiamine pyrophosphate as a cofactor) and striking improvement in the oculomotor disorder (but not other disorders such as amnesia, polyneuropathy and blindness) within hours of the administration of thiamine; completely normal values of transketolase are usually attained within 24·h. Failure of the ocular palsies to respond to thiamine within a few days should raise doubts about the diagnosis. The medial thalamic and periaqueductal lesions may be demonstrated on brain MRI.

Hyponatraemia The most frequent neurological manifestation of hyponatraemia is reduced level of attention or consciousness. Long tract signs (6%), tremor (1%), hallucinations (0.5%), myoclonus and seizures (3%) may also occur. The reason for the appearance of focal symptoms, such as long tract signs, is unclear. The deficits usually respond to correction of the hyponatraemia, but this should be done cautiously to minimize the risk of developing central pontine myelinolysis. Of course, a stroke and in particular subarachnoid haemorrhage, may be the cause of hyponatraemia (section 11.18.2) and many cases are iatrogenic (e.g. use of intravenous fluids, diuretics and carbamazepine). Given that many patients with stroke have been smokers, a bronchial carcinoma should be considered if the pattern of serum and urine osmolalities suggests the syndrome of inappropriate ADH secretion. Hypercalcaemia The usual neurological manifestations of hypercalcaemia are either psychiatric symptoms or an encephalopathy, often accompanied by headache and sometimes seizures. Occasionally, cerebral infarction occurs. A proposed mechanism is vasospasm but this has not been established. Hepatic encephalopathy It is most unlikely that an acute or subacute portosystemic encephalopathy would be mistaken for a stroke or TIA. One case report in the literature describes a woman who developed a left hemiparesis following a general anaesthetic for manipulation of her left shoulder and at the time of transfer to the rehabilitation hospital

..

..

was found to have a mild hepatic encephalopathy.133 Acquired (non-Wilsonian) hepatocerebral degeneration can present with dysarthria, ataxia and intention tremor as well as upper motor neurone signs in the limbs, although there is usually a history of progressive deterioration. The catch is that routine liver function tests may be normal but the serum ammonia is raised and other tests of liver function may be abnormal (e.g. prothrombin time).

If Wernicke’s encephalopathy is a diagnostic possibility, take blood (for red cell transketolase, thiamine and glucose levels) and then treat immediately with thiamine (100 mg IV or IM daily for 3–5 days, followed by 100 mg orally, daily) and a glucose infusion (since hypoglycaemia can precipitate Wernicke’s disease); do not waste time waiting for the blood results to come back.

93

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 94

94

Chapter 3 Is it a vascular event and where is the lesion?

Hypertensive encephalopathy Although encephalopathy caused by malignant or accelerated phase hypertension is now rare, it occurs most commonly in patients with an established history of hypertension, although only when the diastolic blood pressure exceeds about 150·mmHg. However, it is also seen in previously normotensive patients whose cerebral autoregulation is normal and so easily exceeded by a rapid rise in blood pressure, sometimes with a diastolic blood pressure of no more than 100·mmHg (i.e. associated with conditions such as pre-eclampsia, acute nephritis, phaeochromocytoma, renin-secreting tumour, ingestion of sympathomimetic drugs and tricyclic antidepressants, ingestion of tyramine in conjunction with a monoamine oxidase inhibitor, head injury, autonomic hyperactivity in patients with the Guillain–Barré syndrome or spinal cord disorders, and with baroreceptor reflex failure after bilateral carotid endarterectomy). The cause is thought to be widespread cerebral oedema resulting from a breakdown of cerebral blood flow autoregulation. The pathological hallmark is fibrinoid necrosis in resistance vessels in the retina and kidneys. The clinical picture is usually dominated by the subacute onset of headache, nausea, vomiting, confusion, declining conscious state, blurred vision, seizures and focal or generalized weakness. There may be focal neurological signs and hypertensive retinopathy (including papilloedema) (Fig. 3.30). Hypertensive encephalopathy can sometimes be mistaken for intracranial venous thrombosis (particularly in pregnancy) or an arterial ischaemic stroke, especially if there is doubt about the onset of symptoms (i.e. if the patient is confused or obtunded) and if the blood pressure is only moderately elevated. Even a severely elevated blood pressure can be a consequence as well as a cause of stroke, but in these cases there is unlikely to be any evidence of end-organ damage such as retinopathy (section 11.7.1). The aim of treatment should be a smooth reduction in blood pressure over hours rather than minutes; indeed precipitous falls in blood pressure can cause ischaemic stroke (section 4.2.4). Posterior reversible encephalopathy syndrome Posterior reversible encephalopathy syndrome (PRES), also called the reversible posterior leukoencephalopathy syndrome, is a fairly recently described cliniconeuroradiological syndrome that has many clinical and imaging similarities to hypertensive encephalopathy, and eclampsia, except there is usually no evidence of hypertensive end-organ damage, and the prognosis is

generally very good.135 Patients present with cortical blindness, headache, altered mental function (drowsiness, coma) and seizures, and brain MR usually shows very extensive white matter abnormalities, typical of oedema, in the posterior regions of both hemispheres (Fig. 3.40). This syndrome has been associated with a number of immunosuppressive therapies (e.g. cyclosporin, cisplatinum, tacrolimus, and intravenous immunoglobulin therapy), vasculitis and drug withdrawal (e.g. clonidine). The pathophysiology is uncertain, but endothelial dysfunction (sometimes induced by cytotoxic therapies), breakdown of the blood–brain barrier and vasogenic oedema are thought to play a role, as in hypertensive encephalopathy. Indeed, PRES and hypertensive encephalopathy are probably the same syndrome, but with different causes. There is usually, but not always, fairly rapid improvement of both the clinical and MR abnormalities with treatment that removes the cause (e.g. stopping cytotoxic drugs and lowering blood pressure) and controls the symptoms (antiepileptic drugs). However, the condition is not always reversible, confined to the posterior regions of the brain, or the white matter, and we agree with others that PRES is a misnomer.135 Wilson’s disease Wilson’s disease may very rarely present as an acute stroke-like problem.136

3.4.6 Central nervous system infections Encephalitis, brain abscess and subdural empyema If a patient with a focal neurological deficit has an altered conscious state and a fever, then focal or multifocal infection of the brain (meningoencephalitis, brain abscess, progressive multifocal leucoencephalopathy, or subdural empyema) or elsewhere in the body (e.g. pneumonia, urinary tract infection, venous thrombophlebitis, septicaemia) needs to be considered (and perhaps even treated empirically), particularly if the prevalence of infection (e.g. HIV) in the community is high.101,137 Indeed, among patients with a focal neurological deficit from a previous stroke, a subsequent systemic infection may lead to neurological decompensation and apparent worsening of the neurological deficit, prompting an incorrect diagnosis of recurrent stroke or ‘extension’ of the initial stroke. There may also be a history of subacute evolution of systemic upset (fever, malaise, lethargy) and focal neurological symptoms, as well as seizures, meningism or a predisposing

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 95

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

..

..

(a)

(b)

Fig. 3.40 (a) A T2-weighted MR scan of a patient with posterior leucoencephalopathy syndrome, which occurred following rapid blood transfusion after placental abruption. On the day of onset (a), the patient suddenly became cortically blind in association with a severe headache. She also had several seizures. On T2 MR, there were areas of increased signal in

both occipital regions (long arrows). Note also the areas of increased cortical signal in the parietal regions (short arrows). All her symptoms resolved in 7 days. (b) Three months later, there was no residual visual field defect and the brain scan was normal.

condition such as sinusitis, mastoiditis, otitis, pneumonia, or congenital heart disease. Of course, it is conceivable that the diagnosis of stroke is correct, and the cause of the stroke is an infection (section 7.11). The EEG, brain CT, MR scan and cerebrospinal fluid (CSF) are usually characteristically abnormal. For example, the parasitic infection cysticercosis may present like a stroke, but the patient will probably have lived in an endemic area and the CT scan often shows an area of calcification within a cyst, as well as scattered foci of parenchymal calcification. In unilateral subdural empyema, the EEG shows extensive unilateral depression of cortical activity and focal delta waves lasting up to 2·s, and the CT shows a non-homogeneous lenticular or semilunar extracerebral lesion with mass effect. However, CT can be normal early on, and MRI may be required to identify the subdural empyema. Likewise, brain CT or MRI of a cerebral abscess usually shows a low-density lesion that is not in a specific vascular territory and has peripheral ring enhancement following intravenous contrast; but sometimes the presentation

is acute and the typical CT findings of an abscess can be delayed for several weeks. It can also be difficult to distinguish radiologically between herpes simplex encephalitis of the frontotemporal lobes and a middle cerebral artery territory infarct (Fig. 3.41). If the patient has focal neurological signs, is systemically unwell with fever and has what looks like a normal CT scan, then an abscess, subdural empyema, and meningoencephalitis need to be excluded by MR scan, EEG and – if safe to do so – CSF examination. Creutzfeldt–Jakob disease Sporadic Creutzfeldt–Jakob disease (CJD) typically presents with a combination of rapidly progressive (over weeks) dementia and myoclonus, which may be accompanied by symptoms and signs of visual, pyramidal and cerebellar dysfunction. However, CJD may occasionally present acutely with a stroke-like syndrome.138,139

95

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 96

96

Chapter 3 Is it a vascular event and where is the lesion?

cerebellar artery territory, followed by the anterior inferior cerebellar artery territory. Patients with infarcts in the territory of the superior cerebellar artery or multiple cerebellar arteries do not tend to have isolated spontaneous prolonged vertigo.144 Ménière’s disease Ménière’s disease is characterized by repeated crises of quite severe rotatory or whirling vertigo, sometimes causing falling to one side, which can be sudden in onset and last from several minutes to a few days.143 Varying degrees of nausea and vomiting, unilateral (initially) low-pitched tinnitus, sensorineural deafness and a feeling of fullness or pressure in the ear are almost always present as well. It usually begins in middle age. The diagnostic difficulty may come with the first attack, when the auditory symptoms may be mild or non-existent and caloric tests normal. In these cases, cerebellar infarction is a possible differential diagnosis (section 3.3.7). Benign paroxysmal positional vertigo

Fig. 3.41 A plain CT brain scan showing a region of low intensity in the right temporal lobe (arrow) due to herpes simplex encephalitis.

If a ‘stroke’ patient deteriorates and develops myoclonus or dementia, think of sporadic Creutzfeldt–Jakob disease and order one or more EEGs (and MRIs).

3.4.7–Labyrinthine disorders Vestibular neuronitis (labyrinthitis) This is probably the commonest cause of severe acute vertigo.140–143 There is associated nausea, vomiting, nystagmus and ataxia, but no deafness or tinnitus. The acute symptoms usually last for several days and may be followed by positional vertigo for some weeks or months. There is scant evidence for viral involvement of the superior part of the trunk of the vestibular nerve. A very similar, if not identical, clinical syndrome can clearly be caused by minor cerebellar strokes and multiple sclerosis, and the diagnosis is only revealed by MR imaging144 (section 3.3.7). The arterial territory most commonly involved in cerebellar infarction causing vertigo is the medial branch of the posterior inferior

Benign paroxysmal positional vertigo is characterized by recurrent episodes of vertigo and nystagmus that occur only after suddenly changing the position of the head – for example, when looking up, rolling over in bed and turning the head toward the affected ear, lying down, bending over and straightening up.141 The vertigo is usually severe but very brief in duration, certainly less than 1·min, and usually less than 15 s. Hearing is normal. There may be a history of recent head trauma, viral illness, stapes surgery or chronic middle ear disease, but many cases are idiopathic. The cause in most cases is canalolithiasis – particles shed by the otolith membrane float freely in the endolymph of the posterior semicircular canal. A much less common cause is cupulolithiasis in which detritus gets caught on the cupula and causes positional vertigo due to excessive loading of the cupula. The diagnosis is established from the history and using the Dix–Hallpike manoeuvre. Benign recurrent vertigo Attacks of spontaneous vertigo not accompanied by cochlear or neurological symptoms in young or middleaged adults, lasting from 20·min to a few hours, have been called ‘benign recurrent vertigo’. The demographics and precipitants of this condition are very similar to those of migraine, and the attacks may respond to standard migraine prophylactic drugs, suggesting that it may be a migraine variant – but the exact pathophysiology remains uncertain.145

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 97

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

3.4.8–Psychological disorders Psychological factors may give rise to subjective experiences (i.e. symptoms), such as hemiparaesthesiae during a panic attack, that commonly need to be differentiated from symptoms that arise from disease (e.g. TIA or stroke), physiological factors (e.g. physiological tremor), behaviours (e.g. excessive rest), and cultural or external factors (e.g. compensation and the welfare state). When there is no disease pathology it becomes tempting to suggest that the symptom must be ‘not real’ and that it is psychogenic and due to a conversion disorder. Conversion disorder is a psychoanalytic concept that describes the occurrence of motor or sensory neurological symptoms other than pain and fatigue that cause distress, are not explained by disease, not malingered (simulated for clear financial or material gain) and are thought to relate to psychological factors.74 Whatever the cause of the symptoms, those unexplained by disease may be as or even more distressing than those caused by disease. The most common psychogenic symptoms mistaken for TIA and stroke are functional weakness and sensory loss.103 Functional weakness Patients with functional weakness are likely to show inconsistency on observation of their behaviour. For example, their gait on entering the consulting room or at the beginning of the examination may differ from their gait when they leave the consulting room, or at the end of the examination. Further, their weakness when they have to take their clothes on or off may be inconsistent with their weakness when they do another functional task, such as when they have to get something from their bag. If the patient complains of unilateral leg weakness, Hoover’s sign of functional weakness is the only test that has been found in controlled studies to have good sensitivity and specificity.146 It relies on the principle that we extend our hip when flexing our contralateral hip against resistance. It can be performed in two ways: Hip extension: look for a discrepancy between voluntary hip extension (which is often weak) and involuntary hip extension (which should be normal) when the opposite hip is being flexed against resistance. It is important when testing involuntary hip extension to ask the patient to concentrate hard on their good leg. Hip flexion: test hip flexion in the ‘weak’ leg while keeping your hand under the good heel. Feel for the absence of downward pressure in the good leg.

..

..

A similar principle can be used to examine weakness of hip abduction which may initially be weak but then come back to normal if tested simultaneously with the ‘good side’. These tests, although useful, should be interpreted cautiously because: • pain in the affected hip may produce greater weakness on direct, compared with indirect, testing as a result of attentional phenomena (related to pain rather than weakness); • cortical neglect can cause a positive Hoover’s sign; • the test may be mildly positive in normal individuals because of a splinting effect; • none of the studies testing its utility were blinded and none mention the problem of neglect. ‘Collapsing’ and ‘give-way’ weakness are commonly present in patients with functional weakness: one moment the limb is strong, the next it is not. This should be not be described as ‘intermittency of effort’ since it is not possible to directly assess someone’s effort. In this situation, normal power can often be achieved transiently with encouragement, for example by saying to the patient, ‘At the count of three, stop me from pushing down . . . ’ Alternatively, gradually increase the force applied to the limb, starting gently and building imperceptibly up to normal force. Inability to understand the instruction, pain in the relevant joint, being generally unwell, and a misguided eagerness of some patients to ‘convince the doctor’ may be problematic. These concerns have been vindicated in the small number of validity studies which have found that this sign is a rather poor discriminator between functional and disease-related symptoms.74 However, it is our impression in daily practice that it is quite a good discriminating test, and more reliable than Hoover’s sign. Functional sensory disturbance Functional sensory disturbance may be reported as a symptom, or detected first by the examiner. While a number of functional sensory signs have been described, none appear to be specific and they should not therefore be used carelessly to make a diagnosis. • Patients may describe sensory loss that ends where the leg or arm ends, at the shoulder or groin. • The hemisensory syndrome is a disturbance which is usually considered by the patient as ‘something is not right down one side’ or that they feel ‘cut in half’. It is usually more patchy than complete, and of variable intensity. There are often accompanying symptoms of intermittent blurring of vision in the ipsilateral eye (asthenopia) and sometimes ipsilateral hearing problems.

97

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 98

98

Chapter 3 Is it a vascular event and where is the lesion?

• Exact splitting of sensation in the midline is said to be a functional sign because cutaneous branches of the intercostal nerves overlap from the contralateral side, so organic sensory loss should be 1 or 2 cm from the midline. However, midline splitting can also occur in thalamic stroke. Similarly, patients with disease should not report a difference in the sensation of a tuning fork placed over the left compared to the right side of the sternum or frontal bone, as these bones are a single unit and must vibrate as one. But these signs seem to be as common in patients with disease and so cannot be recommended.74,147 • Asking patients to ‘Say yes when they feel touch and no when they don’t’, in order to see whether they say ‘no’ when the affected area is touched, is difficult to interpret because the patient may be saying ‘no’ to mean ‘not as much’. • Although often considered left-sided, a systematic review found only a slight left-sided preponderance for functional motor and sensory symptoms.148 • ‘La belle indifférence’, an apparent lack of concern about the nature or implications of symptoms or disability, has no value in discriminating whether the patient is making an effort to appear cheerful in a conscious attempt to avoid being labelled as depressed, or factitious because he or she is deliberately making up the symptom and is not concerned about it. The diagnosis of functional motor and sensory symptoms depends on demonstrating positive functional signs as well as the absence of signs of disease. Most of these signs relate to inconsistency, either internal (for example, Hoover’s sign reveals discrepancies in leg power) or external (for example, tubular field defect is inconsistent with the laws of optics). However, although inconsistency may be evidence that the signs are functional, it does not indicate whether they are consciously or unconsciously produced, and a positive functional sign does not exclude the possibility that the patient also has disease. Dizziness Panic attacks can present somatically with dizziness, a fear of embarrassment, and an inability to escape from situations in which they are likely to occur, such as supermarkets. However, anxiety and phobic avoidance of situations, or head positions, that bring on dizziness does not necessarily indicate a psychogenic aetiology. For example, physiological vestibular sensitivity to certain visual stimuli such as patterned lines or bright lights (sometimes called visual vertigo) may cause symptoms

that also come on in crowded places. Asking the patient to hyperventilate to see if that reproduces the symptoms might appear straightforward, but this is often falsely positive in patients with dizziness caused by disease. Depersonalization and derealization also may be described by the patient as ‘dizziness’. If this sensation is there all the time, the patient may have depersonalization disorder (a chronic form of dissociation). Speech and swallowing symptoms Dysphonia is the most common functional speech complaint. Often the clinical presentation is of whispering or hoarse speech that is initially thought to be laryngitis by the patient, but then persists for months or years. The possibility of spasmodic adductor or abductor dysphonia must always be considered. Functional dysarthria typically resembles a stutter or is extremely slow with long hesitations that are hard to interrupt. The speech may be telegrammatic, consisting only of the main verbs and nouns in a sentence. In its extreme form the patient may become mute. However, these types of speech disturbance can also be seen in patients with disease. Word-finding difficulty is a common symptom in anyone with significant fatigue or concentration problems and may compound any functional dysarthria. True dysphasia as a more severe functional symptom, however, is rare. Globus pharyngis or functional dysphagia is also common. The patient normally complains of a sensation of a ‘ball in the throat’ and investigations do not reveal a cause. Visual symptoms Intermittent blurring of vision that returns to normal if the patient screws up their eyes tight and then relaxes them again is commonly reported. Some of these patients have convergence or accommodation spasm, with a tendency for the convergence reflex to be transiently overactive, either unilaterally or bilaterally. In this situation lateral gaze restriction can sometimes mimic a sixth nerve palsy, but miosis reveals the diagnosis. Voluntary nystagmus appears to be a ‘talent’ possessed by around 10% of the population. Tests for functional visual acuity problems are described in detail elsewhere.149 Simple bedside tests for a patient complaining of complete blindness are to ask them to sign their name or bring their fingers together in front of their eyes (which they should be able to do). They may have a normal response to menace and optokinetic nystagmus with a rotating drum. Decreased acuity in one eye can be assessed with a ‘fogging test’ in which plus lenses of increasing power are placed in front

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 99

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

of the ‘good’ eye until the patient can only be using their ‘bad’ eye to see. Spiral or tubular fields are common, and are often asymptomatic. Remember to test the fields at two distances when looking for a tubular field. Patients with functional hemianopia have been described who have homonymous hemianopia with both eyes open and then, inconsistent with this, a monocular hemianopia in one eye with a full field in the other eye. Monocular diplopia or polyopia may be functional but can be caused by ocular pathology.

3.4.9 Head injury Head injury may cause a stroke, and a stroke may cause head injury. Without a detailed history of the onset from relevant people as well as the patient (witnesses, family members, ambulance officers, family doctor) it can be difficult to be sure what has happened. For example, head injury may cause intracranial haemorrhage, which can be mistaken for a primary stroke if the patient is amnesic for the injury and has no external scalp evidence of injury. Head injury can also cause ischaemic stroke as a result of arterial dissection (section 7.2). On the other hand, stroke can precipitate a fall, causing head injury and, if the CT scan shows intracranial haemorrhage, the primary stroke event may be missed. Brain imaging may shed some light on the cause; intracranial haemorrhage from head injury tends to be more common in the frontal and anterior temporal regions, superficial and multiple, and there may be accompanying extension into the subarachnoid space and associated skull fractures (Fig. 3.42) (section 9.1.4). One can often see soft tissue swelling of the scalp and brain contusions on the CT scan.

3.4.10 Multiple sclerosis Patients with multiple sclerosis (MS) usually develop focal neurological symptoms in their third or fourth decade (as opposed to the seventh and eighth decades for stroke). The onset is typically subacute, with the symptoms coming on over days or even weeks. The diagnosis is seldom difficult because these patients are usually young without any vascular diseases or risk factors, the symptoms are as often positive as negative, there are usually more neurological signs than symptoms (compared with TIA and stroke, in which there are often more symptoms than signs), and some may have evidence of disease in other parts of the central nervous system that is asymptomatic but readily detected by clinical examination or brain/spinal cord MRI. Also, there may be a history of previous episodes that are typical of multiple

..

..

Fig. 3.42 A plain CT brain scan showing blood in the subarachnoid space in a patient who presented with confusion following a head injury. There is blood in the anterior interhemispheric fissure (straight arrow) in a pattern suggesting spontaneous subarachnoid haemorrhage. However, the presentation apparently followed a head injury and the suggestion of frontal haemorrhagic contusion (curved arrows) clouded the interpretation of the CT scan and delayed recognition of the true pathology (anterior communicating artery aneurysmal haemorrhage).

sclerosis, such as optic neuritis or transverse myelitis. However, sometimes symptoms do seem to start abruptly and so can mimic a TIA or stroke.150 The location and shape of the lesions on brain imaging are usually fairly characteristic (Fig. 3.43) (i.e. discrete round or oval lesions in the white matter of the cerebral or cerebellar hemispheres and brainstem, in the corpus callosum, and adjacent to the temporal horns of the lateral ventricles, not corresponding to territories supplied by specific cerebral arteries). In addition, the cerebrospinal fluid usually shows raised immunoglobulin (IgG) and oligoclonal bands, which are not present in the serum. However, none of these features is specific; for example, oligoclonal bands may be found in patients with acute stroke, particularly if due to vasculopathies associated with Behçet’s disease, systemic lupus erythematosus and sarcoidosis, although in many of these cases oligoclonal bands are also present in the serum (section 7.3).

99

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 100

100

Chapter 3 Is it a vascular event and where is the lesion?

Fig. 3.43 A fluid-attenuated inversion recovery (FLAIR) MR brain scan showing typical ‘flame-shaped’ periventricular white matter lesions (white areas) in a patient with multiple sclerosis.

3.4.11 Neuromuscular disorders Mononeuropathy and radiculopathy It is not uncommon for TIA and stroke affecting just the hand or arm to have to be distinguished from a median neuropathy at the wrist (carpal tunnel syndrome), an ulnar neuropathy at the elbow, a radial nerve palsy, or cervical radiculopathy.44,151 This is because peripheral nerve and nerve root lesions may occasionally cause sudden onset (or sudden awareness, perhaps waking up from sleep) of persisting focal sensory or motor symptoms, which can be confused with ‘pseudoradicular’ strokes (those due to small lesions in the contralateral precentral or postcentral gyrus, corona radiata, or thalamus) (section 3.3.5). Similarly, acute tetraparesis and cranial polyneuropathy associated with acute polyradiculoneuropathy (e.g. Guillain-Barré syndrome and its Miller Fisher variant) can sometimes be mistaken for a brainstem stroke.152–154 However, the physical signs (i.e. lower motor neurone signs and/or sensory loss to pain in a dermatomal or nerve distribution) are different from those of an intracranial cortical/subcortical lesion which tend to be associated with upper motor neurone signs and/or loss of discriminative/‘cortical’ sensations, such as joint position sense and twopoint discrimination ability. Nevertheless, there are patients with the ‘cortical hand syndrome’ (section 3.3.4) in whom such distinguishing features are not present. Nerve conduction studies will usually identify a peripheral mononeuropathy unless the problem is very acute, but they are less helpful for more proximal radicular problems unless there are clearly absent F waves.

Myasthenia gravis Although the symptoms of myasthenia gravis usually come on gradually, there are instances of fairly rapid development, sometimes precipitated by infection (usually respiratory), drugs, or emotional upset.155,156 If the muscles of the eyes (levator palpebrae and extraocular muscles) and less often the face, jaw, throat and neck are the first to be affected (causing ptosis, diplopia, dysarthria or dysphagia), this can be mistaken for a brainstem vascular event, but the weakness tends to persist (if left untreated) and to fluctuate, increasing as the day wears on. Excess fatigability can be demonstrated by asking the patient to sustain the activity of the symptomatically involved muscles. Conversely, muscle power improves after a brief rest, or in response to 10 mg of edrophonium (tensilon) intravenously, or neostigmine (1.5·mg) intramuscularly. A preceding dose of atropine 0.5 mg IV not only counteracts parasympathetic overstimulation but also serves as a placebo control for the motor effects. However, because false positive (e.g. motor neurone disease) and false negative (about 10% of cases) results are fairly common, and adverse reactions can occur rarely, the tensilon test is not used unless neurophysiological tests are unavailable or a rapid diagnosis is required. The presence of anti-acetylcholine receptor antibodies confirms the diagnosis in most patients but in those presenting acutely with a suspected ‘stroke’ the test result takes too long to obtain to be useful. Motor neurone disease It is surprising how often patients with bulbar motor neurone disease (MND) are said to have had a brainstem

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 101

3.4 Differential diagnosis of focal cerebral symptoms of sudden onset

stroke. This is probably because MND symptoms sometimes seem to start surprisingly quickly and, early on, there may be little more than subtle dysarthria without any typical features of MND. What then tends to happen is that a brain CT or MRI is performed, and shows leukoaraiosis which is linked with ‘stroke’, despite this finding being quite common in the MND age group. The more useful investigation is electromyography which may reveal subclinical evidence of denervation which could not be caused by a stroke. Also, the MND patient continues to deteriorate, quite unlike a stroke patient.

3.4.12 Important non-focal disorders Syncope Syncope is probably the most important non-focal syndrome of impaired consciousness to distinguish from TIA and stroke, accounting for about one-fifth of patients presenting to the emergency room with a suspected brain attack in a recent study.45 In another study, in which at least two neurologists aimed to validate the diagnosis of TIA made by general practitioners (GPs) and by hospital emergency physicians, the most frequent conditions misdiagnosed as TIAs were transient disturbances of consciousness, mental status and balance.157 Compared with neurologists, GPs considered ‘confusion’ and ‘unexplained fall’ more often compatible with TIA and ‘lower facial palsy’ and ‘monocular blindness’ less often compatible with TIA. Syncope is defined as a loss of consciousness and postural tone due to a sudden fall in blood flow to the brain.158–161 Sometimes this may occur abruptly, without any warning (e.g. due to aortic stenosis, complete heart block), but more commonly there is a preceding feeling of light-headedness, faintness or ‘dizziness’ (not rotational vertigo), bilateral dimming or loss of vision (not to be confused with lone bilateral blindness), sounds seeming to be distant, generalized weakness, and symptoms of adrenergic activity such as nausea, hot and cold feelings and sweating.52 During the attack, the patient is pale, sweaty, clammy and floppy, rather than cyanosed and rigid as in an epileptic seizure. The pulse is absent or difficult to feel (but this cannot be relied on) and the patient may be incontinent of urine. If the patient is lying flat and not held upright (by someone or by an obstacle), then consciousness is regained within seconds, and there is very little mental confusion or difficulty in recalling the warning symptoms (unless there has been head trauma). Additional features during unconsciousness may include multifocal, arrhythmic, myoclonic jerks (particularly in patients who are held upright), head turns, oral automatisms, righting movements (sustained

..

..

head-raising or sitting up), eye movements (upward or lateral deviation of the eyes) and visual and auditory hallucinations, all of which are likely to lead to an erroneous diagnosis of epilepsy. There are usually no focal neurological symptoms unless the drop in blood pressure occurs in the presence of severe occlusive arterial disease in the neck, or impaired cerebral autoregulation due to a previous stroke. The key to the diagnosis is a sound history from a witness as well as the patient – use the telephone if necessary.159–161 Making a correct diagnosis is important, because some causes are quite serious (e.g. Stokes–Adams attacks) and if misdiagnosed as a transient ischaemic attack may lead to the patient being denied an effective and possibly life-saving treatment (e.g. a pacemaker). Drop attacks ‘Drop attacks’ are episodes of sudden loss of postural tone which cause the patient – usually a middle aged or elderly woman – to fall to the ground without apparent loss of consciousness, vertigo or other sensation. The attack occurs without warning and is not induced by a change of posture or movement of the head. The patient may be unable to rise immediately after the fall, despite being uninjured, perhaps because of the surprise. The most common differential diagnosis is vertebrobasilar ischaemia, but here there is usually some warning that the patient is going to fall, there are brainstem symptoms, such as vertigo or diplopia, and the limb weakness is persistent.53,157 Not a single patient in the New England Medical Center Posterior Circulation Registry had a drop attack as the only symptom of posterior circulation ischaemia.54 Drop attacks have also been attributed to syncope, epilepsy, tumours in the region of the foramen magnum or third ventricle, vestibular disease, myxoedema, old age and even subconscious guilt. In the vast majority of cases no cause is found, and although the attacks may persist, the patients do not seem to have a high risk of stroke or other vascular events and so the episodes are called ‘cryptogenic drop attacks’. Rarely the patient may have parasagittal motor cortex/subcortex ischaemia in the territory of both anterior cerebral arteries.162 Cataplexy Cataplexy is characterized by transient episodes of sudden bilateral loss of muscle tone, weakness and areflexia provoked by emotions, and is highly specific to the narcolepsy syndrome.163–165 Although excessive daytime sleepiness usually begins several months before the onset of cataplexy, up to 10% of patients have

101

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 102

102

Chapter 3 Is it a vascular event and where is the lesion?

cataplexy first. Laughter is the most common precipitant, but other forms of emotion and athletic activities can induce cataplexy. The three typical situations that trigger cataplexy, and differentiate it from other types of muscle weakness, are ‘when hearing and telling a joke’, ‘while laughing’, and ‘when angry’. Severe attacks cause complete paralysis except for the respiratory muscles, whereas the more common partial episodes cause patients to drop objects or to sit down or stop walking. Less than 1% of attacks cause unilateral weakness and so might be confused with a TIA. Momentary attacks are the usual pattern, and they generally last less than a minute. Prolonged episodes may be associated with hallucinations. Rarely, cataplexy can be almost continuous (‘status cataplecticus’). Despite the recent discovery of hypocretin deficiency in the lateral hypothalamus as a cause of narcolepsy, the cause of cataplexy remains unclear. Serotonin syndrome The serotonin syndrome is a dose-related range of toxic symptoms caused by excessive stimulation of the central and peripheral nervous system serotonergic (5-HT1A and 5-HT2) receptors. It is a result of increased serotonin concentrations associated with serotinergic drug use (e.g. serotonin precursors, serotonin agonists, serotonin releasers, serotonin reuptake inhibitors, monoamine oxidase inhibitors, some herbal medicines, and antimigraine medications such as sumatriptan and dihydroergotamine). The clinical picture includes mental-status changes, neuromuscular abnormalities and autonomic hyperactivity166 which may be mistaken for stroke (with or without a complication such as infection) because the onset is quite rapid and patients have hyperreflexia and clonus (often greater in the legs than arms).

3.4.13 Neuroimaging in the diagnosis of focal neurological symptoms of sudden onset Transient ischaemic attacks The main purpose of brain imaging in patients with a suspected transient ischaemic attack (TIA) is to identify a relevant focal ischaemic lesion on MR DWI in cases in which the diagnosis of TIA is uncertain, and to exclude any underlying structural intracranial lesion and, rarely, an intracerebral haemorrhage which may present like a TIA.13,167–169 Because the yield of CT for detecting structural lesions is only about 1% in patients with suspected TIA, the routine examination of every patient who has had a single TIA with CT is controversial (section 6.8.3).

We do not image the brain of patients with transient monocular blindness but do image patients with (potentially transient) ‘brain attacks’ with MR DWI. This is because DWI may not only help confirm the diagnosis,169 but also helps localize the ischaemia to one or more arterial territories and so differentiate carotid from vertebrobasilar ischaemia, and sometimes cardiac embolism in clinically uncertain cases. This is particularly important when carotid surgery is being considered. However, there are limited data on the cost-effectiveness of brain imaging in TIA patients.167,170 Stroke The diagnosis of stroke remains primarily clinical, and the main role of brain imaging is to exclude non-vascular structural pathology as the cause of the symptoms, and establish the underlying vascular pathology (Chapter 5) and aetiology (Chapters 6, 7 and 8). The choice of CT or MRI will depend on the question being asked of the test; local availability; how ill, confused or restless the patient is; and cost and effectiveness. If CT is available, the imaging strategy producing the highest number of quality adjusted life years (QALYs) at lowest cost is to ‘scan all patients immediately’ with CT, because CT is practical, quick (a few minutes to scan the brain), widely available, easy to use in ill patients, affordable (£44–130), it accurately identifies intracranial haemorrhage as soon as it occurs (section 5.4.1) and it is essential to image suspected SAH (section 3.7.3).13,167 Strategies that delay CT scanning reduce QALYs and increase cost. However, CT has limitations. Intracerebral haemorrhage will be misinterpreted as ischaemic stroke if CT is not done within 10–14 days after stroke;167 delays in seeking medical attention or poor access to CT for stroke will result in failure to identify up to three-quarters of cases of intracerebral haemorrhage, and may lead to inappropriate treatment (e.g. antiplatelet drugs or carotid revascularization) for many. Although CT shows positive features of ischaemic stroke in many patients with moderate and severe stroke scanned 2–7 days after the event, early signs of ischaemia, within 3–6 h, are difficult to recognize on CT (section 5.4.2). Furthermore, many patients with mild stroke never develop a visible infarct on CT, irrespective of when they are scanned.167 The Acute Cerebral CT Evaluation of Stroke Study (ACCESS), in which as many doctors and radiologists as possible worldwide interpret typical CT scans of stroke over the Internet, aims to improve recognition of early signs of infarction (http:// www.neuroimage.co.uk/access/).171 The advantage of MRI DWI is that it shows ischaemic changes as early as a few minutes after stroke as a bright

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 103

3.5 Differential diagnosis of transient monocular blindness

white lesion (‘light bulb’), and so shows more ischaemic strokes than CT or conventional MRI.13,14,169,172,173 It also shows more ischaemic strokes several weeks later, and therefore is particularly useful for positive identification of an ischaemic stroke in patients presenting up to 8 weeks after stroke.18 For patients presenting too late for CT to show intracerebral haemorrhage, gradient echo MR sequences reveal previous intracerebral haemorrhage indefinitely as a low signal (black) ring or dot, and can therefore distinguish previous haemorrhage from infarction. However, a limitation of MRI is that it may not identify hyperacute intracerebral haemorrhage correctly, and it can fail to detect subarachnoid haemorrhage. Further, it is difficult to use routinely in acute, particularly severe, stroke; it is less often available than CT; it requires more cooperation by the patient for a longer time; it is very noisy and it upsets confused patients. About one-fifth of patients cannot undergo MRI because they are too ill or confused, or have an intraocular or intracerebral metallic foreign body or pacemaker.

3.4.14 Electroencephalography in the diagnosis of focal neurological symptoms of sudden onset With the increasing availability of neuroimaging, the indications for an EEG in patients with suspected TIA or stroke are now very limited. An EEG may be helpful when the clinical diagnosis of TIA is in doubt and partial (focal or localization-related) seizures are a possibility. However, while about one-third of all patients with clinically definite epilepsy consistently have epileptiform discharges on the waking interictal EEG, and about one-half do so on some occasions with repeated sleep-deprived recordings,174 we know very little about the sensitivity and specificity of the test when the diagnosis is in doubt. Further difficulty arises as a result of the poor specificity of EEG abnormalities (all too frequently, patients with non-epileptic events such as TIAs are reported to have an abnormal EEG). The role of the EEG in patients with persisting neurological signs is to help exclude a number of conditions that may mimic stroke, particularly when brain imaging is normal. For example, non-convulsive status epilepticus can present with a sudden confusional state, and in both Creutzfeldt–Jakob disease and herpes simplex encephalitis, where there can be clinical deterioration with new focal neurological signs, the EEG can have characteristic abnormalities (although not in all the patients all the time). The usual abnormal EEG findings in acute stroke, at least in fairly large hemispherical cortical and subcortical stroke, are a localized reduction of normal cortical rhythms and presence of localized slow-wave abnormality.

..

..

Focal EEG slowing, however, is not specific, and only indicates the presence and side of the lesion. Although it has been suggested that the EEG may help distinguish between small deep (lacunar) and cortical infarction, the clinical features and brain CT or MRI scan are far more effective tools. A normal EEG may help to confirm the diagnosis of ‘locked-in syndrome’ caused by a lesion in the ventral pons (section 3.3.2), and can also lend some support to a diagnosis of a somatization disorder, at least if the clinical deficit is extensive.

3.5 Differential diagnosis of transient monocular blindness

Transient monocular blindness (TMB) and amaurosis fugax are two different terms for describing exactly the same symptom – blurring or loss of vision in the whole or part of the visual field of one eye. It is commonly caused by ischaemia of the retina (usually as a result of embolism from atherothrombosis of an artery between the heart and the eye, or from embolism form the heart), and sometimes by ischaemia of the anterior optic nerve (usually due to disease of the posterior ciliary artery, section 3.5.2).175,176 However, there are other causes of TMB which are important to differentiate because the prognosis and treatments differ (Table 3.30). Many are uncommon and frequently go unrecognized if the patient is not examined during the acute episode (which of course is very difficult to achieve). Patients with TMB should be considered for a competent ophthalmological examination to exclude primary disorders of the eye before concluding that the explanation is necessarily vascular disease.

3.5.1 Retinal disorders Retinal migraine or ‘vasospasm’ Migraine with aura (classical migraine) is usually ushered in by ‘positive’ binocular visual symptoms. However, transient monocular visual symptoms followed by pulsatile headache have occasionally been described in patients with known migraine, and classified as ‘retinal migraine’.177–183 TMB as a result of thromboembolism is distinguished from retinal migraine on the basis of the symptoms; the former is characterized by the abrupt onset of ‘negative’ monocular visual phenomena (blindness), which is painless and usually lasts only a few minutes, while

103

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 104

104

Chapter 3 Is it a vascular event and where is the lesion? Table 3.30 Causes of transient monocular blindness. Retinal disorders (section 3.5.1) Vascular Atherothromboembolism or other arterial disorders (e.g. dissection affecting the proximal internal carotid artery, and giant-cell arteritis affecting the posterior ciliary artery) Embolism from the heart (section 6.5) Low retinal artery perfusion pressure Retinal migraine High resistance to retinal perfusion Intracranial vascular malformation Central or branch retinal vein thrombosis Raised intraocular pressure (glaucoma) Raised intracranial pressure Increased blood viscosity (section 7.9) Malignant arterial hypertension (section 3.4.5) Retinal haemorrhage Retinal detachment Paraneoplastic retinopathy Phosphenes Lightning streaks of Moore Chorioretinitis Optic nerve disorders (section 3.5.2) Anterior ischaemic optic neuropathy Malignant arterial hypertension (section 3.4.5) Papilloedema Optic neuritis and Uhthoff’s symptom Dysplastic coloboma Eye/orbital disorders (section 3.5.3) Vitreous haemorrhage Reversible diabetic cataract Lens subluxation Orbital tumour (e.g. optic nerve-sheath meningioma)

the latter is characterized by the gradual build-up of transient monocular visual impairment (i.e. scotoma or blindness), which is usually incomplete and may be associated with ‘positive’ visual symptoms (e.g. scintillations) lasting for up to an hour, as well as a pulsatile headache or orbital pain. Sometimes, however, it can be very difficult to distinguish retinal ischaemia from retinal migraine, particularly in older patients without any headache. It has been suggested that non-invasive investigations, such as carotid ultrasound, may help distinguish between the two – i.e. tight stenosis of the internal carotid artery on the symptomatic side suggests retinal ischaemia (due to artery-to-artery embolism), and the absence of carotid disease favours retinal migraine. However, this is only circumstantial evidence. Patients have been described with frequent (1– 30 episodes per day), stereotyped episodes of brief (lasting less than 3·min) unilateral visual loss caused by presumed vasospasm. Fundus examination during the episodes shows constriction of the retinal arteries and segmentation in a thin and slowly moving column of

blood. The calibre of the retinal vessels is restored with the return of vision. Occasional patients have responded to a calcium channel blocker. However, as there is really no proof of vasospasm, we have to be very cautious in the interpretation of these observations in what appear to be very rare patients. Patients with TMB have a better prognosis than patients with TIA of the brain, despite similar degrees of carotid stenosis184 (section 16.11.8). The reason remains uncertain, but is unlikely to be that some of the cases of TMB were really cases of retinal migraine. The question therefore arises, how do we define retinal migraine? Is it a diagnosis based on the clinical symptoms and signs, the results of investigations, the response to treatment, or the prognosis? It has been traditionally diagnosed on the basis of the clinical history, but this may be non-specific, and it is seldom possible to examine the patient during the episode to see if ‘vasospasm’ is present. Any carotid disease may be coincidental, and of course the response to treatment and the prognosis are hardly helpful at the time the diagnosis has to be made. We cannot, therefore, provide any definitive answer. Arteriovenous malformation Anterior and middle fossa dural arteriovenous malformations very rarely cause TMB. When they do it is probably because of transient lowering of retinal arterial pressure associated with shunting of blood away from the ophthalmic artery. Central or branch retinal vein thrombosis Thrombosis of the central retinal vein, or branch retinal vein, sometimes presents with attacks of TMB.175 The visual loss tends to be patchy rather than complete. The fundoscopic appearance is characteristic: engorged retinal veins and multiple retinal haemorrhages (Fig. 3.44). Angle-closure glaucoma This typically presents in people over the age of 50 years who are hypermetropic (long-sighted). Apposition of the peripheral iris to the trabecular meshwork decreases the outflow of the aqueous humour which in turn increases the intraocular pressure and reduces perfusion pressure to the choroid, retina and disc. Transient monocular visual disturbance may occur, particularly in poor light when the pupil is dilated. The onset is usually subacute. Vision may be decreased, blurred, foggy, or smoke-like, and the patient may see haloes around lights. Some patients complain of light sensitivity during the attack and most, but not all, have eye pain, which may radiate

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 105

3.5 Differential diagnosis of transient monocular blindness

attenuated retinal arterioles, electroretinography demonstrates abnormal cone- and rod-mediated responses, and antiretinal antibodies may be identified in the serum. Over the subsequent months, progressive visual loss occurs, during which time a small-cell carcinoma of the lung often declares itself. Phosphenes

Fig. 3.44 An ocular fundus photograph showing engorged retinal veins and multiple retinal haemorrhages due to central retinal vein thrombosis. Also reproduced in colour in the plate section.

to the side of the head. The symptoms may last from a few minutes to hours. Eye pain and the stereotyped recurrence of attacks under certain lighting conditions are useful clues to the diagnosis, as are red eye, cloudy cornea and oval pupil (Fig. 3.27). The intraocular pressure should be checked in most, if not all, patients with TMB, but in those who have glaucoma it is not always raised between attacks. Retinal and other intraocular haemorrhages A small retinal haemorrhage may cause sudden reduced vision in one eye that resolves within hours. The diagnosis should be evident on ophthalmoscopy, particularly if the pupil is dilated. Similarly, vitreous and anterior chamber haemorrhage may cause TMB. The cause may be suspected from the history; for example, preretinal haemorrhage may occur during physical exertion, sexual activity, or Valsalva manoeuvre.

Phosphenes are flashes of light and coloured spots that are induced by eye movement in a dark environment and occur in the absence of luminous stimuli. They may occur with disease of the visual system at many different sites, such as optic neuritis in the recovery phase, perhaps as a result of the mechanical effects of movement of the optic nerve. However, mechanical pressure from a sharp tap on the normal eyeball may also induce a phosphene, as every child discovers, by stimulating the retina. Phosphenes may also occur in a healthy darkadapted closed eye after a saccade (flick phosphene). Lightning streaks of Moore In a dark environment, elderly people frequently experience recurrent, brief, stereotypic, vertical flashes of light in the temporal visual field of one eye, which are elicited by eye movement. These are known as Moore’s lightning streaks and are benign. It is believed that, with advanced age, the posterior vitreous may collapse and detach from the retina, leading to persistent vitreoretinal adhesions. The mechanical forces associated with eye movement exert traction on the macula and retina, and induce the photopsias (subjective sensations of sparks or flashes of light). Chorioretinitis Macular disease due to chorioretinitis or retinal pigmentary degeneration can sometimes lead to loss of vision in bright light.

Paraneoplastic retinopathy Transient episodes, lasting seconds to minutes, of painless monocular dimming of the central field of vision and overwhelming visual glare and photosensitivity when exposed to bright light suggests not only photoreceptor dysfunction due to transient retinal ischaemia, but also paraneoplastic retinopathy. Patients may also experience transient bizarre entoptic symptoms (alterations in normal light perception resulting from intraoptic phenomena, akin to the subjective perception of light resulting from mechanical compression of the eyeball – phosphenes). Ophthalmoscopy usually reveals

..

..

Central serous retinopathy This typically affects young men. Over a period of hours to days, central vision becomes blurred associated with various degrees of metamorphopsia (defective, distorted vision), micropsia (objects appear smaller than their actual size due to segregation of the photoreceptors), chromatopsia (objects appear unnaturally coloured), central scotoma, and increasing hyperopia (far-sightedness). It lasts for days to weeks, usually resolving within 4–8 weeks. Visual acuity in the acute stage ranges from 6/6 to 6/60 and averages 6/9. On fundoscopy, there is an accumulation

105

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 106

106

Chapter 3 Is it a vascular event and where is the lesion?

of transparent fluid between the retinal pigment epithelium and the outer segments in the macular area, elevating the retina and causing a circumscribed area of retinal detachment at the posterior pole. The diagnosis is confirmed by fluorescein angiography. Typically the fluorescein enters into the blister and stains its contents, identifying one or more leakage points.

3.5.2 Optic nerve disorders Anterior ischaemic optic neuropathy Anterior ischaemic optic neuropathy (AION) is due to ischaemia in the territory of supply of the posterior ciliary arteries, which are branches of the ophthalmic artery, and which supply the anterior part of the optic nerve, the choroid and outer retina. Because the optic disc is close to an arterial border zone between the territories of the two major posterior ciliary arteries, it is particularly liable to ischaemia when the systemic blood pressure falls, when the intraocular pressure rises, or when there is occlusive disease of local small arteries. Less often, AION is caused by embolism from the heart or proximal arteries to, or in situ occlusion of, the posterior ciliary arteries or the arterioles feeding the anterior part of the optic nerve, e.g. by giant-cell arteritis (Fig. 3.45) and other types of vasculitis, such as polyarteritis nodosa (section 7.3). Atherosclerosis of the posterior ciliary artery is another presumed cause, given that patients with AION have an increased prevalence of hypertension and diabetes and an increased risk of subsequent cerebrovascular and cardiovascular events, but we are not aware of any histological proof of atheromatous occlusion of these arteries. Ischaemia of the optic disc is characterized clinically by the sudden onset of painless visual loss in one eye.185 This may involve the whole field, but it tends to be more severe in the lower half because the upper segment of the optic disc is more vulnerable to ischaemia. In the early stages, the circulation may be so precariously balanced that minor postural change may have profound effects on the degree of visual loss. The visual loss tends to be severe, non-progressive and prolonged, but it can be brief (and present as transient monocular blindness), or it can progress over several hours or days. In AION due to giantcell arteritis, the visual loss may develop sequentially in both eyes within a few days and, rarely, almost simultaneously. Normal visual acuity does not exclude AION, and almost any part of the visual field can be affected. The common patterns of visual loss are an altitudinal hemifield defect (loss of either the upper or, more frequently, the lower half of the field in one eye), an inferior nasal segmental loss, and a central scotoma. The disc may appear normal at first, but within a few days

Fig. 3.45 A photomicrograph of anterior ischaemic optic neuropathy caused by giant-cell arteritis. The arrow indicates the infarcted optic nerve head (courtesy of Dr J.F. Cullen, Western General Hospital, Edinburgh). Also reproduced in colour in the plate section.

it becomes pale and swollen, often with small flameshaped haemorrhages radiating from the disc margin (ischaemic papillopathy), distended veins and, occasionally, cotton-wool spots due to ischaemic change in the surrounding retina. The swelling may involve only one segment of the disc, or may be more marked in one segment than in another. The disc swelling is attributed partly to leakage of plasma from damaged blood vessels and partly to arrest of axoplasmic transport along damaged nerve fibres. The disc swelling itself may be indistinguishable from that seen with raised intracranial pressure, but the vision is usually normal in the latter. In about half of the eyes with AION due to giantcell arteritis, the disc swelling has a chalky white appearance. Later, the swelling subsides, to be succeeded by

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 107

3.5 Differential diagnosis of transient monocular blindness

optic atrophy with attenuation of the small blood vessels on the disc surface. The prognosis for recovery of vision is variable. Many patients recover good central vision, but are left with arcuate and sectorial visual field defects corresponding to loss of bundles of nerve fibres. In other patients, visual loss may be complete and permanent. Malignant arterial hypertension Some patients with malignant hypertension experience transient monocular blindness due to ischaemia of the optic nerve head. Associated headache, seizures, encephalopathy, renal impairment, high blood pressure and the characteristic ophthalmoscopic features of hypertension point to the diagnosis (Fig. 3.30). Papilloedema Patients with papilloedema (Fig. 3.46) from any cause may experience transient visual blurring or obscurations, with or without photopsias. The visual loss in chronic papilloedema is often postural, occurring as patients get up from a chair (or bend over), and may involve either eye alone, or both eyes together. The visual loss is typically ‘grey’ rather than ‘black’ and lasts for seconds rather than minutes. The explanation may be transient optic nerve ischaemia secondary to a relative decrease in orbital blood flow, as a result of raised cerebrospinal fluid pressure in the subarachnoid space around the optic nerve, and so increased pressure in the veins draining the optic nerve head. Episodes of visual blurring or blindness in

someone with papilloedema should lead to urgent investigation and appropriate action because permanent visual loss will eventually follow, gradually or suddenly. Optic neuritis and Uhthoff’s symptom Patients with acute and chronic optic nerve demyelination due to multiple sclerosis may experience transiently decreased vision in one or both eyes during exercise (Uhthoff’s symptom), or in association with other causes of increased temperature, emotional stress, increased illumination, eating, drinking, smoking and menstruation.185,186 The pathophysiology is unknown, although reversible conduction block in demyelinated nerve fibres secondary to an increase in body temperature or to changes in blood electrolyte levels or pH are believed to play a role. Ophthalmoscopy may be normal, but if the optic nerve head is inflamed, the optic disc will be swollen and will look similar to papilloedema. Optic disc anomalies Transient monocular visual obscurations are occasionally associated with an elevated optic disc without increased intracranial pressure. Examples include congenital anomalies of the disc, such as drusen or posterior staphyloma.

3.5.3 Orbital disorders Transient changes in the ocular media or intraocular pressure, such as vitreous floaters, vitreous haemorrhage, anterior chamber haemorrhage, lens subluxation, reversible cataract (in a diabetic) and glaucoma may cause transient monocular visual disturbance. Most of these conditions can be excluded by a competent ophthalmological examination. An intraorbital mass, such as an optic nerve sheath meningioma, may produce gaze-evoked transient monocular blindness; the blindness is limited to the duration of gaze in the affected direction (usually abduction of the affected eye) and the visual acuity usually returns to normal about 30·s after the eye moves back to the primary position. The loss of vision is possibly caused by a reduction in flow to the blood vessels surrounding the optic nerve itself.

3.5.4 Corneal and eyelid disorders Fuch’s corneal endothelial dysfunction (corneal guttatae) Fig. 3.46 An ocular fundus photograph showing papilloedema. Note the congested, swollen disc, with loss of the physiological cup, a blurred disc margin and congested retinal veins. Also reproduced in colour in the plate section.

..

..

In the seventh and eighth decade, the corneal endothelial cells may become deficient and malfunction due to abnormal excrescences of the basement (Descemet’s)

107

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 108

108

Chapter 3 Is it a vascular event and where is the lesion?

membrane, called corneal guttatae. The corneal epithelium may then fail to pump fluid out of the cornea and patients complain of blurred vision (due to hydration of the cornea). It tends to occur in the morning and gradually clears as the day evolves, and as the cornea clears due to evaporation of tears. Similar symptoms can occur due to endothelial cell loss as a consequence of intraocular surgery. Ptosis Intermittent ptosis (e.g. myasthenia gravis) can cause transient monocular visual loss.

3.6 Improving the reliability of the clinical diagnosis

For every patient who presents to their family doctor with a definite transient ischaemic attack (TIA), there are many more who present with transient neurological symptoms due to other disorders. For example, in the Oxfordshire Community Stroke Project (OCSP), 512 patients were referred by their family doctor or a hospital doctor with a diagnosis of ‘possible TIA’, of whom 317 (62%) were considered by the OCSP neurologists not to have had a TIA (Table 3.22).6 This problem is not unique to family doctors and junior hospital doctors. The interobserver agreement for the diagnosis of TIA of the brain amongst eight senior and interested neurologists from the same department who interviewed 56 patients in alternating pairs revealed that both neurologists agreed that 36 patients had a TIA and 12 had not, but they disagreed about eight (kappa =·0.65; for perfect agreement kappa would be 1.0).187 Even experienced neurologists with an interest in cerebrovascular disease show considerable interobserver variability in the diagnosis of transient ischaemic attack. This does not imply lack of skill, but rather it is inherent in the clinical assessment of symptoms and signs. Sometimes the available information does not allow one to come to a ‘right answer’ as to whether the event was a transient ischaemic attack or not – in which case one ends up working on the basis of probability. The inter-observer agreement for the diagnosis of stroke (vs not stroke) is made with moderate to good reliability (kappa = 0.77).37

The clinical features associated with high interobserver agreement for the diagnosis of stroke or TIA versus no vascular event are a sudden change in speech, visual loss, diplopia, numbness or tingling, paralysis or weakness and nonorthostatic dizziness (kappa = 0.60).49 Several studies have shown that clinicians can differ in the interpretation of even isolated elements of the history, such as loss of power, loss of sensation, ‘blurred or foggy vision’, and headache.37,188 When assessing patients with suspected stroke, the interobserver agreement for most items of the clinical history is moderate to good (kappa statistics for vascular risk factors range from 0.44 to 0.69) but there is much less inter-observer agreement for various features of the neurological examination (Table 3.31).37,188,189

3.6.1 Reasons for clinical disagreement There are several factors that can increase the likelihood of clinical disagreement. For patients with suspected TIA, the main problem is eliciting and interpreting the history of an event that has resolved by the time the history is taken. For example, since the symptoms of most TIAs resolve within about 15–60·min, the diagnosis is almost always based entirely on the clinical history, which for a number of reasons may not be very clear: the patient may have forgotten the symptoms (because of poor memory or delay in presenting to medical attention); the symptoms may have been remembered, but are difficult to describe (e.g. transient homonymous hemianopia); or the patient may have been so frightened by the attack that he or she was more preoccupied with the immediate outcome than the exact nature of the deficit. All of these problems are more common in the elderly. For patients with suspected stroke, the inter-observer reliability of the clinical assessment is affected by a variety of factors, such as the time since the onset of symptoms; there is a trend for worse inter-observer reliability among patients assessed very early and very late after symptom onset.37 Another problem is that the generally accepted definitions of stroke and TIA lack specific detail about which ‘focal’ symptoms are not acceptable (e.g. ·isolated vertigo or not?). In the context of TIAs, what is an acceptable lower time limit for the duration of symptoms? Is a sudden focal neurological deficit, particularly a sensory deficit, of less than 5·s duration a TIA? Diagnostic conformity can only be achieved if precise criteria are available that are valid, reliable and generally accepted. In the particular case of TIA, it is difficult to determine the validity of any single diagnostic criterion, because there is no ‘gold standard’ against which to

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 109

3.6 Improving the reliability of the clinical diagnosis Table 3.31 Inter-observer agreement for neurological signs in stroke patients.

Kappa value Signs examined

Lindley et al.189

Shinar et al.188

Hand et al.37

Conscious level Confusion Dementia Weakness of arm Weakness of hand Weakness of leg Weakness of face Sensory loss on hand Sensory loss on arm Aphasia/language problem Dysarthria Visuospatial dysfunction Hemianopia Cerebellar signs/ataxia Cranial nerve palsy Extraocular movement disorder

0.60 0.21 NS 0.77 0.68 0.64 0.63 0.19 0.15 0.70 0.51 0.44 0.39 0.46 0.34 0.30

0.38 NS 0.34 NS 0.58 (R) 0.49 (L) NS 0.51 (R) 0.66 (L) 0.50 (R) 0.32 (L) NS 0.54 0.53 NS 0.40 0.45 NS 0.77

0.70 0.45 NS 0.65 0.72 0.57 0.50 0.49 0.49 0.66 0.41 0.41 0.46 NS NS NS

NS, not stated.

judge it, other than perhaps the prognosis. For example, patients with lone bilateral blindness were followed up and found to have a similar prognosis to patients with TIA, implying that lone bilateral blindness is also a TIA.48 Whether the prognosis of patients with isolated dysarthria or vertigo is similar is an important research question. More widespread and consistent consideration, discussion and application of the diagnostic criteria could enhance diagnostic accuracy and inter-observer agreement, and might improve patient management and care.188 However, as the diagnostic criteria become more specific, sensitivity is sacrificed and so an increasing number of genuine TIAs may be discarded and left untreated. Conversely, if the criteria become less specific, there may be a tendency to overdiagnose TIA, but this too can have adverse consequences such as the loss of a job, driver’s or pilot’s licence, money and self-esteem, as well as resulting in inappropriate investigation and the prescription of numerous unnecessary drugs. The inter-observer reliability of the clinical assessment of stroke is also improved by the experience and confidence of the assessor.37

3.6.2 Strategies to reduce inter-observer variation The clinical skills required to obtain an accurate and useful history are listed in Table 3.32, and strategies

..

..

Table 3.32 Clinical skills required to obtain an accurate and useful history. The ability to: Establish understanding Establish information Interview logically Listen Interrupt only when necessary Observe non-verbal cues Establish a good relationship Interpret the interview Tell the story in plain language Tell the story in chronological order Make the story ‘human’ Like this: This 85-year-old widow was standing at the kitchen table peeling potatoes at 7 p.m. on 26 November 2006, with her daughter, when she suddenly stopped talking, dropped the potato peeler she was holding in her right hand and fell to the floor. She was unable to get up and has not been able to speak or move her right arm or leg since. Not like this: This lady developed sudden dysphasia and right hemiparesis.

for preventing or minimizing clinical disagreement in Table 3.33. If these principles are applied with a knowledge of the diagnostic criteria (Table 3.5), then diagnostic inconsistency should be minimized. If there is still uncertainty about the diagnosis at the end of the

109

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 110

110

Chapter 3 Is it a vascular event and where is the lesion? Table 3.33 Strategies for preventing or minimizing clinical disagreement. Assess the patient in a suitable consulting environment, i.e. quiet room, minimal interruptions if possible Necessary equipment available: ophthalmoscope, sphygmomanometer, etc., telephone on desk to call witnesses (to clarify history) or colleagues (to obtain advice) Clarify and confirm key points Repeat key elements of the history or examination Corroborate important findings with witnesses, documents and, if necessary, appropriate tests Ask ‘blinded’ colleagues to see the patient also (e.g. at ward teaching sessions) Record evidence as well as inference, making a clear distinction between the two, reporting exactly what the patient said and then your interpretation (e.g. ‘the patient complained of heaviness of the right arm and leg’ and not ‘the patient complained of right-sided weakness’ or ‘the patient complained of right hemiparesis’) Apply the art and social sciences of medicine, as well as the biological sciences of medicine Make sure you have enough time for the entire consultation

history, then the general examination and special investigations aimed at detecting vascular diseases and risk factors may provide useful circumstantial evidence (Table 3.34).44,45,49,190 The odds of an event being due to cerebrovascular disease are likely to be substantially less if the patient is young and has no vascular risk factors, compared with an elderly patient who has several vascular risk factors, clinical evidence of established vascular disease (e.g. carotid or femoral bruits, absent peripheral pulses) or symptomatic vascular disease elsewhere (e.g. angina, intermittent claudication)190 – i.e. make sure that you use all the clinical evidence available to you. When coming to a view about whether an event was a transient ischaemic attack or stroke, make use of all the clinical evidence, both general and neurological, that is available after a detailed history and examination. If the history is elicited independently by a second physician, a subsequent comparison of the symptoms as well as their interpretation may lead to a greatly increased reliability of the diagnosis. Teaching hospitals in particular are in a privileged position to apply this powerful but expensive diagnostic ‘instrument’, but even so, it is still unrealistic except for the occasional very difficult case. What is often available to those in

Table 3.34 Clinical features (i.e. the milieu) influencing the probability that the event was a stroke or transient ischaemic attack. Very likely to be vascular, almost definite Atrial fibrillation and rheumatic heart disease Frequent carotid-distribution transient ischaemic attacks and focal, long, loud bruit over the carotid bifurcation on the symptomatic side History and physical signs suggestive of infective endocarditis (i.e. fever, splinter haemorrhages, cardiac murmur) Recent myocardial infarction (in last 3–4 weeks) Likely to be vascular, but less definite Atrial fibrillation and non-rheumatic valvular heart disease (but a few fibrillating stroke patients have primary intracerebral haemorrhage as the cause of the stroke) Arterial bruits anywhere (e.g. carotid, orbital, aortic, femoral) Prosthetic heart valve, taking anticoagulants (but some strokes are haemorrhagic, and some transient ischaemic attacks are related to coexistent carotid artery disease) Unlikely to be vascular (particularly if neurological symptoms are transient) Less than 40 years of age, no symptomatic vascular disease, no vascular risk factors, no family history of thrombosis or premature vascular disease, normal heart

hospitals, however, is an account by the referring physician who saw the patient at an earlier time after the onset of symptoms – be sure to read it carefully and go over any symptoms described that were not reported to yourself by the patient. Implicit within this is the responsibility for all clinicians who take a history to document it in the patient’s or witness’s own words and not simply record their own interpretation. The use of checklists written in simple language improves inter-observer reliability, and is likely to be useful for computer-aided diagnosis and further research studies.191 However, although checklists may encourage more thorough history taking, the symptoms still have to be interpreted correctly. Very short checklists have been used successfully by paramedical personnel for diagnosing cerebrovascular events hyper-acutely (section 3.3, table 3.6).39,40,42,43,49 The relatively poor inter-observer agreement for physical signs (Table 3.31) is not peculiar to patients with stroke. The advent of MRI technologies such as DWI, which is becoming more widespread and can show what were hitherto unseen and yet recent relevant ischaemic lesions, has also improved the sensitivity, specificity and reliability of the diagnosis of TIA and minor ischaemic stroke169 (section 3.2.1).

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 111

3.7 Is it a subarachnoid haemorrhage?

3.7 Is it a subarachnoid haemorrhage?

Subarachnoid haemorrhage (SAH) refers to the spontaneous extravasation of blood into the subarachnoid space when a blood vessel near the surface of the brain ruptures. It is a condition, not a disease, which has many causes (section 9.1).29,30 Although, as we will see (section 5.3.1), the clinical distinction between ischaemic stroke and intracerebral haemorrhage is unreliable and we have to rely on imaging, the clinical features of SAH are reasonably distinct; at least this type of stroke can be diagnosed clinically with reasonable confidence. However, confirmatory investigations are needed in almost all cases.

3.7.1 Clinical features Blood in the subarachnoid space is a meningeal irritant and incites a typical clinical response, regardless of the source. Patients usually complain of headache, photophobia, stiff neck and nausea, and they may also vomit. Confusion, restlessness and impaired consciousness are also frequent (Table 3.35). Precipitating factors More often than not, there is no obvious precipitating factor.192 Among the 33 patients with SAH in the OxfordTable 3.35 Diagnosis of subarachnoid haemorrhage. Principal symptoms headache – usually sudden, maximal in seconds, severe, and occipital or retro-orbital – duration: hours (possibly minutes, we do not know) to weeks nausea vomiting neck stiffness photophobia loss of consciousness Neurological signs none (very often) meningism (after several hours) focal neurological signs: third nerve palsy (mostly posterior communicating artery aneurysm), dysphasia, hemiparesis (arteriovenous malformation, intracerebral haemorrhage) subhyaloid haemorrhages in optic fundi fever raised blood pressure altered consciousness

..

..

shire Community Stroke Project, six (18%) occurred while resting (none occurred while asleep), 13 (39%) during moderate activity, and six (18%) during strenuous activity, such as weight-lifting and sexual intercourse; the activity at onset was not known in the other eight patients.193 In a more recent series, higher proportions (up to 50%) of aneurysmal SAHs occurred during physical exertion.194 Sexual activity may precipitate not only SAH,195 but also – and more often – relatively harmless headaches, migrainous or not (section 3.7.2).196 Headache Headache is the cardinal clinical feature of SAH. It is the only symptom in about one-third of patients,194,197 but a symptom at some stage in almost every patient.198,199 Conversely, in a prospective series of patients with sudden headache in general practice, SAH was the cause in one out of four patients, and only in one out of ten patients in whom sudden headache was the only symptom.197 One out of every four patients with sudden, severe headache has a ruptured cerebral aneurysm, or one out of ten if sudden headache is the only symptom. In SAH, the headache is generally diffuse and poorly localized but tends to spread over minutes to hours to the back of the head, neck and down the back as blood tracks into the spinal subarachnoid space. Sometimes the headache is maximal behind the eyes. The headache is often described by patients as the most severe headache they have ever had, but it can occasionally be milder. It is the suddenness of onset which is most characteristic. Speed of onset and disappearance The headache arises suddenly, classically in a split second, ‘like a blow on the head’ or ‘an explosion inside the head’, reaching a maximum within seconds. A potential pitfall is that patients may sometimes use the word ‘sudden’ to describe an episode of headache that came on over half an hour or longer, perhaps depending on the interval after which the history is taken. And even if the headache really comes on within seconds or minutes, this is not specific for ruptured aneurysms, or even for SAH in general. The reason is that in general practice, exceptional forms of common headaches outnumber common forms of a rare disease, in this case a ruptured aneurysm. This is one of the many examples of the risk paradox (section 18.5).200 Another, more striking example is that most children with Down

111

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 112

112

Chapter 3 Is it a vascular event and where is the lesion?

syndrome are born of mothers under 30 years old, despite older mothers being at higher risk. In the same way, most patients with ischaemic stroke are not severely hypertensive, etc. Given that the incidence of aneurysmal haemorrhage is about 10 per 100 000 population per year, a family physician with a practice of 2000 people will, on average, see only one such patient every five years. Patients with rare manifestations of common headache syndromes (i.e. sudden onset of migraine) probably outnumber those with common manifestations of rare headache syndromes (i.e. sudden headache in subarachnoid haemorrhage). Patients with headache may present not only to general practitioners but also to an accident and emergency department where they make up around 1% of all attendances. The proportion with a serious neurological condition ranges from 16% for all patients with any headache, to 75% for those with sudden headache specifically referred to a neurologist.201 The exact speed of onset of ‘sudden’ headache, seconds or minutes, in patients without any other deficits, is of little help for the hospital physician in distinguishing aneurysmal haemorrhage from innocuous headaches, or from non-aneurysmal perimesencephalic haemorrhage (section 9.1.2). The predictive value of the speed of onset (seconds vs 1–5 min) can be calculated from two data sets. First, SAH from ruptured aneurysms is nine times as common as non-aneurysmal perimesencephalic haemorrhage203 and in hospital series these two forms of SAH together are twice as common as innocuous headaches.202 Second, headache develops almost instantaneously in 50% of patients with aneurysmal SAH, in 35% of patients with non-aneurysmal perimesencephalic haemorrhage, and in 68% of patients with benign ‘thunderclap headaches’; for an onset within 1–5 min these proportions are 19%, 35% and 19%, respectively.194 If, for the sake of simplicity, we ignore patients with sudden headache from other serious (nonhaemorrhagic) brain disease, such as intracranial venous thrombosis,204 a simple calculation leads to the disappointing conclusion that an onset within seconds correctly predicts aneurysmal haemorrhage in only 55%, and that headache onset in 5–10 min correctly predicts innocuous headache in only 30%. The headache of SAH usually lasts 1–2 weeks, sometimes longer. We do not know exactly how short in duration a headache may be and still be due to SAH. However, we have not come across anyone with a headache due to proven SAH that resolved within 1 h. However, since there is no prospective study that has

addressed this question it is still conceivable that this may occur, so it is perhaps best to consider SAH in anyone with a sudden unusually severe headache that lasts longer than – shall we say – 1 h? In brief, there are no single or combined features of the headache that distinguish reliably, and at an early stage, between SAH and innocuous types of sudden headache. The discomfort and cost of referring most patients for a brief consultation in hospital (which should include CT scanning and a delayed lumbar puncture if the scan is negative) is probably outweighed by avoidance of the potential disaster of missing a ruptured aneurysm and the patient later being admitted with rebleeding, or another secondary complication. There is no feature of the headache that distinguishes reliably, and at an early stage, between subarachnoid haemorrhage and innocuous types of sudden headache. Therefore, although most people with a sudden severe headache have not had a subarachnoid haemorrhage, they must all be investigated to exclude this diagnosis. A biphasic headache may occur in patients whose SAH is due to dissection of a vertebral artery (section 9.1.3): first, a severe occipital headache radiating from the back of the neck, followed after an interval of hours or days by sudden exacerbation of the headache but of a more diffuse type. A history of one or more previous episodes of suddenonset headache (‘sentinel headaches’) is generally believed to be common in patients with aneurysmal SAH and these are often attributed to a ‘warning leak’ of the finally rupturing aneurysm. However, the notion of ‘minor leaks’ does not really hold up (section 9.2.4). It is doubtful if patients with aneurysmal subarachnoid haemorrhage often have preceding and unrecognized ‘warning leaks’. Whatever the case, doctors must be educated to consider subarachnoid haemorrhage in any patient who reports a sudden severe headache. Vomiting Vomiting (and nausea) is common at the outset, in contrast to other differential diagnoses, such as migraine, in which vomiting more often comes after the headache starts. Neck stiffness Meningism refers to painful resistance to passive or voluntary neck flexion, mainly because of irritation of

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 113

3.7 Is it a subarachnoid haemorrhage?

the cervical meninges by subarachnoid blood, or by inflammation. This sign can be elicited in the supine patient by placing both hands behind the patient’s head and, as one attempts to lift the head up off the pillow the patient does not allow the neck to be flexed and so the examiner lifts the patient’s head, neck and shoulders off the bed, as if the patient were like a board. In contrast, passive rotation of the neck is achieved with ease. Neck stiffness is a common symptom and sign of blood in the subarachnoid space, but it does not occur immediately; it takes some 3–12 hours to appear and may not develop at all in deeply unconscious patients, or in patients with minor SAH.205 Therefore, its absence cannot exclude the diagnosis of SAH in a patient with sudden headache. Brudzinski’s sign (flexion at the hip and knee in response to forward flexion of the neck) is also associated with blood in the subarachnoid space but its sensitivity and specificity in this context are unknown. Pain and stiffness in the back and legs may follow SAH after some hours or days because blood irritates the lumbosacral nerve roots. We know of one patient who was admitted during a weekend because of a sudden headache; on the following Monday, when seen by a new resident, he complained of pain in the back of his legs, but his notes could not be found (they were discovered only later, buried under a pile of X-rays). He was worked up for a back problem, but he suddenly died from what turned out to be rebleeding from a ruptured aneurysm. Photophobia Patients are often photophobic and irritable for several days after SAH. Loss of consciousness Loss of consciousness occurred in 50% of a large group of patients with presumed aneurysmal SAH who were well enough to be entered into a clinical trial of medical treatment.206 As these figures did not include the 10% or so of patients with SAH who die at home or during transportation to hospital,207 or the 20% who reach hospital and die within the first 24 h,208 it is likely that at least 60% of all patients with SAH lose consciousness at, or soon after, onset. The patient may regain alertness and orientation or may remain with various degrees of lethargy, confusion, agitation or obtundation. An acute confusional state can occur and be misinterpreted as psychological in origin (grimacing, spitting, making sucking or kissing sounds, spluttering, singing, whistling, yelling and screaming).209–211

..

..

The impaired consciousness may be caused by the large amount of blood in the subarachnoid space, the reduced cerebral blood flow caused by the sudden increase in cerebrospinal fluid pressure, or by a complication of SAH such as brain displacement by haematoma or hydrocephalus, or a fall in systemic blood pressure or arterial oxygen concentration. Epileptic seizures Epileptic seizures (partial or generalized) may occasionally occur at onset or subsequently, as a result of irritation or damage to the cerebral cortex by the subarachnoid and any intracerebral blood. In the Oxfordshire Community Stroke Project, two of the 33 patients (6%) with SAH had an epileptic seizure at onset, but neither had later seizures.97 Data from other series indicate that about 10% of patients with SAH develop epileptic seizures, most occurring on the first day of the SAH, but one-third not having their first seizure until 6 months later and one-third even more than a year later.212–215 The only independent predictors of epilepsy after SAH are a large amount of cisternal blood on brain CT, and rebleeding.213 Subhyaloid haemorrhage Intraocular haemorrhage develops in approximately 20% of patients with a ruptured aneurysm and may also complicate non-aneurysmal SAH, or intracranial haemorrhage in general. The haemorrhage is caused by a sustained increase in the pressure of the cerebrospinal fluid, with obstruction of the central retinal vein as it traverses the optic nerve sheath, in turn leading to congestion of the retinal veins.216 Mostly, the haemorrhages appear at the time of aneurysmal rupture, but exceptionally later without evidence of aneurysmal rebleeding. Linear streaks of blood or flame-shaped haemorrhages appear in the pre-retinal layer (subhyaloid), usually near the optic disc (Fig. 3.47); one-third lie at the periphery.217 If large, the pre-retinal haemorrhage may extend into the vitreous body (Terson syndrome; section 14.10.1). Patients may complain of large brown blobs obscuring their vision. Focal neurological signs Focal neurological signs may occur when an aneurysm has compressed a cranial nerve or has bled into the brain substance, causing an intracerebral haematoma (section 9.3.6). Sometimes, therefore, the clinical manifestations of a ruptured aneurysm may be indistinguishable from a stroke syndrome due to intracerebral haemorrhage or

113

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 114

114

Chapter 3 Is it a vascular event and where is the lesion?

returns to normal within a few days. The blood pressure changes probably serve to counteract the decrease in cerebral perfusion resulting from increased cerebrospinal fluid pressure and later ischaemia (section 14.3.1). Cardiac arrhythmias and ECG abnormalities are common after SAH (section 14.9.3).221 The mechanism is unexplained but is thought to be sustained sympathetic stimulation, perhaps caused by dysfunction of the insular cortex, which results in reversible structural neurogenic damage to the myocardium, such as contraction bands, focal myocardial necrosis and subendocardial ischaemia.

3.7.2 Differential diagnosis

Fig. 3.47 Ocular fundus of a patient with subhyaloid haemorrhage, appearing as sharply demarcated linear streaks of brick red-coloured blood or flame-shaped haemorrhage in the preretinal layer, adjacent to the optic disc and spreading out from the optic disc. Also reproduced in colour in the plate section.

cerebral infarction, particularly if little or no blood has entered the subarachnoid space. Systemic features Fever, hypertension, albuminuria, glycosuria and electrocardiographic (ECG) changes may be present in the acute phase. Pyrexia rarely exceeds 38.5°C during the first 2–3 days, but thereafter it may rise to over 39°C, presumably due to accumulation of breakdown products of blood in the subarachnoid space; the pulse rate is not raised concomitantly.218 The important distinguishing feature between pyrexia due to blood in the subarachnoid space and pyrexia due to intercurrent infection is the pulse rate; it remains disproportionately low in the former and rises with the latter. Hypertension is a well-established risk factor for SAH;219 a quarter to one-third of patients have a history of hypertension.206,220 On admission, about 50% of patients with aneurysmal SAH have a markedly raised blood pressure. Unfortunately, high blood pressure is found in many patients presenting to an accident and emergency department for any reason, which makes the predictive value of blood pressure readings in that situation abysmally low. In many patients this raised blood pressure is a reactive phenomenon, and not a marker of long-standing hypertension; often the blood pressure

The abrupt onset of a severe headache may not only be caused by subarachnoid haemorrhage (SAH), but also by several other conditions, such as meningitis or encephalitis, intracerebral haemorrhage (particularly posterior fossa haemorrhage), obstruction of the cerebral ventricles, intracranial venous thrombosis, a rapid rise in blood pressure and, finally, a variety of alarming but innocuous conditions.222 Associated features such as female sex, the presence of seizures, loss of consciousness or focal symptoms, vomiting, or exertion at onset increases the probability of aneurysmal SAH,194 but the predictive value of these characteristics is far from helpful. Only a sudden headache during sleep should suggest hypnic headache (see below). Acute painful neck conditions to be distinguished from meningism Meningism may be a feature of SAH, meningitis, a posterior fossa mass and cerebellar tonsillar coning. However, it characteristically disappears as coma deepens. Other causes of a painful or stiff neck include bony lesions (i.e. trauma or arthritis) and ligamentous strain in the neck, extrapyramidal rigidity, systemic infections such as pneumonia, cervical lymphadenitis, parotitis, tonsillitis and upper lobe pneumonia. However, it is usually quite easy to distinguish meningism from these other acute painful neck conditions. For example, pain arising from the cervical spine may not only be felt in the neck and back of the head but also in the shoulder and arm, it is often evoked or exacerbated by certain movements or positions of the neck other than flexion, and there is usually tenderness to palpation over segments of the cervical spine. Meningitis Meningitis is an acute febrile illness that usually presents subacutely over 1 or 2 days with generalized

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 115

3.7 Is it a subarachnoid haemorrhage?

headache, meningism, photophobia and fever. However, it can be difficult to distinguish from SAH if the patient is found confused or comatose, with marked neck stiffness and no available history. Clues to the diagnosis of meningitis include a high fever, tachycardia and a purpuric skin rash (meningococcal meningitis). If the patient is fully conscious and has no focal neurological signs and if meningitis is suspected, a lumbar puncture should be done immediately. But if the patient is very ill, antibiotics and steroids should be given at once, i.e. after venepuncture for blood cultures but before proceeding to brain CT and then, if no blood or intracranial mass is seen, a cerebrospinal fluid examination (section 3.7.3). Cerebellar stroke Cerebellar stroke often gives rise to sudden severe headache, particularly if it is haemorrhagic, and also to nausea and vomiting, but is usually accompanied by neurological symptoms and signs, such as vertigo, dysarthria and unsteadiness, which help distinguish it from SAH. However, if the lesion is large, the patient may present in coma due to direct brainstem compression or obstruction of cerebrospinal fluid flow from the fourth ventricle, causing hydrocephalus and raised intracranial pressure, or there may be meningism without signs of definite brainstem dysfunction. In a consecutive series of 100 patients with an initial diagnosis of subarachnoid haemorrhage, eight had a cerebellar haematoma (and another seven had supratentorial intracerebral haemorrhage).223 Urgent brain CT is required to confirm the diagnosis of cerebellar haematoma, and lumbar puncture should certainly not be done; indeed, lumbar puncture should almost always be preceded by brain CT in unconscious patients, even if there are no focal signs of a mass lesion and no clinical evidence of raised intracranial pressure (such as papilloedema). Intracerebral haemorrhage More than 50% of patients with spontaneous intracerebral haemorrhage have headache at onset, particularly those with superficial lobar haemorrhages, but the headache is generally not as strikingly sudden in onset as SAH.89 Furthermore, focal neurological deficits are almost always present, but they can also occur in about 20–30% of aneurysmal SAHs where there is intraparenchymal extension of the haemorrhage.224 Conversely, some intracerebral haemorrhages, particularly those which are deep, have less prominent focal neurological signs and can easily be mistaken for SAH. CT brain scanning is always required.

..

..

Intraventricular haemorrhage Intraventricular haemorrhage, which may arise within the ventricles or from immediately beneath the ependymal lining, or by extension from an intracerebral haemorrhage (i.e. from a caudate haemorrhage or ruptured subependymal vascular malformation into the intraventricular system), may mimic SAH (section 9.1.5). Also, intraventricular haemorrhage may occur together with SAH, usually from a ruptured aneurysm, most frequently at the anterior communicating artery complex. Patients present with sudden severe headache, confusion, vomiting or collapse with loss of consciousness.225 Again, brain CT or MR scan is required for diagnosis during life. Carotid or vertebral artery dissection Dissection of the wall of the internal carotid artery may cause a fairly distinctive headache syndrome, which is ipsilateral, involving the forehead, periorbital region, face, teeth or neck, and has a burning or throbbing quality (section 7.2.1.). The headache may be associated with an ipsilateral Horner’s syndrome or monocular blindness, and with contralateral focal neurological symptoms or signs. Dissection of the wall of the vertebral artery generally causes pain in the upper posterior neck and occiput, usually on one side, and may be associated with symptoms and signs of posterior circulation ischaemia, such as the lateral medullary syndrome. Transmural dissection of an intracranial artery can cause SAH (section 9.1.3). Acute obstructive hydrocephalus Any acute obstruction of the flow of cerebrospinal fluid causes headache through a rapid increase in intracranial pressure. The headache is commonly bilateral and exacerbated by coughing, sneezing, straining or head movement. Intermittent obstructive hydrocephalus may therefore cause severe paroxysmal headaches. A colloid cyst of the third ventricle is the classical cause of this syndrome; sometimes more than one family member is affected.226 A decreased level of consciousness may follow the headache.227 The outcome can be lethal if a diagnosis is not made.228 Brain CT or MRI usually identifies the offending lesion (Fig. 3.48). Migraine Migraine headache can sometimes arise suddenly (‘crash’ migraine), be severe and prostrating, unilateral or generalized, and associated with photophobia, irritability,

115

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 116

116

Chapter 3 Is it a vascular event and where is the lesion?

‘Exploding head’ syndrome Clusters of attacks characterized by a sensation of sudden noise in the head and terror, rather than pain, can strike individuals over the age of 50 years, particularly during the twilight of sleep.232,233 The cause is uncertain. Cough headache Idiopathic cough headache is defined as head pain brought on by coughing or other Valsalva manoeuvres, but not by prolonged physical exercise, in the absence of any intracranial disorder.234,235 It is probably caused by distension of venous structures in the brain. It is a sudden-onset headache and lasts from 1 s to 30 min, tends to be bilateral and posterior, does not begin earlier than the fifth decade of life, is more frequent in men, is not accompanied by other neurological manifestations and often responds to indomethacin. These clinical characteristics allow its differential diagnosis from posterior fossa lesions, especially herniation of the cerebellar tonsils (Chiari-I malformation), even though a craniocervical magnetic resonance imaging study is useful to rule this out. Fig. 3.48 CT brain scan showing a tumour in the third ventricle (arrow) at the foramen of Monro, causing obstructive hydrocephalus.

mild confusion, anorexia, mild fever, extraocular muscle palsy (ophthalmoplegic migraine) or symptoms of brainstem disturbance (basilar migraine) and thus be mistaken for SAH. However, migraineurs generally have a past or, less diagnostically helpful, a family history of migraine and the headache is commonly unilateral and throbbing, not so rapid in onset and of shorter duration than the headache of SAH.86 Vomiting tends to start well into the migraine attack, in contrast to SAH, in which it commonly occurs at or soon after onset of the headache.

Hypnic headache This type of headache has only been recognized recently.236,237 It wakes patients up in the middle of the night, 2–6 h after sleep onset; it is usually diffuse and bilateral, or in the neck. It occurs almost exclusively in middle age, twice as often in women as in men. Patients usually get out of bed and walk around until the pain subsides, within 10 min to 3 h. Typically the attacks are recurrent, from once a week to 6 times per night. There are no known precipitating factors, there is no photophobia, phonophobia or nausea and there are no concomitant autonomic features, such as in cluster headache or chronic paroxysmal hemicrania. Post-traumatic headache

Idiopathic stabbing headache Three specific varieties of sudden sharp stabbing headache have been described: ice-pick-like pains, ‘jabs and jolts syndrome’ and ophthalmodynia.229,230 The pains are mostly at the temples or orbits but on occasion are elsewhere in the head.231 Migraineurs are particularly susceptible.231 Precipitants – which are seldom present – may be postural change, physical exercise, or head motion. As these pains are transient and lancinating, they are unlikely to be confused with the headache of SAH. The mechanism is unknown.

Immediately after a head injury there is often headache; this may be pulsating and made worse by head movement, jolting, coughing, sneezing and straining.229 Normally the headache gradually disappears as the soft and any bony tissue damage resolves. In a series of 200 patients admitted to hospital with head injury only 83 still had a headache by a day or so afterwards; only 22 of those 83 (11% of all patients) complained spontaneously, and only three required an analgesic.238 The diagnosis of post-traumatic headache should not be confused with SAH if there is a history of head injury,

..

..

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 117

3.7 Is it a subarachnoid haemorrhage?

but the patient may be amnesic and there may be no witness, in which case acute head injury with secondary bleeding into the subarachnoid space can be confused with spontaneous SAH. If the circumstances of a traumatic head injury are unclear, and there is a reasonable chance that spontaneous intracranial haemorrhage was the cause of the accident and so the head injury, then brain CT should be done as soon as the patient’s condition allows, regardless of the severity of the head injury. Benign orgasmic cephalalgia and benign exertional headache Acute, severe, explosive occipital or generalized headache, usually occurring at the moment of sexual orgasm or during strenuous exercise (benign orgasmic cephalalgia and benign exertional headache, respectively) may mimic SAH.196,239,240 The history of onset during sexual intercourse (or masturbation) may not be forthcoming without specific and sensitive enquiry. Points in favour of the diagnosis are a history of similar previous sexual or exertional headaches, no alteration in consciousness, short duration of the headache (minutes to hours) and no signs of meningeal irritation such as neck stiffness, or low back pain, and nor sciatica in the ambulant patient. These headaches can occur at any time in life and do not necessarily occur every time the patient experiences orgasm, or exercises strenuously. If patients present soon after their first-ever sudden orgasmic headache, it is not possible to exclude SAH without brain CT and lumbar puncture. If a patient presents after recurrent attacks, and the history is characteristic, investigation is seldom necessary. Reaction to monoamine oxidase inhibitor drugs People taking classic monoamine oxidase inhibitors (MAOIs), of which phenelzine and tranylcypromine are the most commonly used, may experience sudden severe headache after ingesting sympathomimetic agents, red wine or foods with a high tyramine content, such as mature cheese, pickled herrings, game and yeast extract. This is because MAOIs irreversibly block the ability of both MAO isoforms (A and B) to metabolize dietary tyramine in the liver (A) and gut wall (B). The combination of a classic MAOI and oral tyramine can provoke dangerous hypertension. The headache tends to be over the occipital region of the head and is associated with a rapid rise in blood pressure. It can be relieved by the alpha noradrenergic blocking agent, phentolamine.

..

..

Phaeochromocytoma Patients with a phaeochromocytoma experience acute pressor reactions; in about 80% of attacks they complain of headache241,242 often in combination with palpitations or sweating.243 The headache is usually of sudden onset, bilateral, severe and throbbing. It appears to be related to a rapid increase in blood pressure and lasts less than an hour in about 75% of patients, but it may last from a few minutes to a few hours. Some patients may collapse with loss of consciousness or develop focal neurological signs during the episode, as a result of cerebral oedema or sometimes haemorrhage. Attacks may be provoked by exertion, straining, emotional upset, worry or excitement.242 The diagnosis depends on clinical suspicion being aroused when the history is first taken (which can be difficult because the condition is so rare) and is confirmed by finding increased excretion of catecholamines (metanephrine and vanillylmandelic acid) in three 24-h specimens of urine, or raised plasma free metanephrine or normetanephrine levels during the attack.244 The blood sugar is usually raised at the time of the attack, a useful distinction from hypoglycaemic attacks, which may simulate phaeochromocytoma because of secondary release of adrenaline in response to low blood sugar. The tumour may arise at any point along the line of development of the sympathetic chain from the neck to the pelvis and scrotum. Headache may also reflect idiopathic surges of hypertension, not associated with phaechromocytoma or any other identifiable condition.245 Occipital neuralgia Occipital neuralgia is characterized by an aching or paroxysmal jabbing pain in the posterior neck and occipital region in the distribution of the greater or lesser occipital nerves (Fig. 3.49). It may rarely present quite dramatically, like SAH,246 but there is usually diminished sensation or dysaesthesiae of the affected area (C2 distribution), focal tenderness over the point where the greater occipital nerve trunk crosses the superior nuchal line, and a therapeutic response to infiltration of local anaesthetic near the tender area on the nerve trunk. Benign ‘thunderclap headache’ ‘Thunderclap headache’ is not a true disease entity, but a convenient term to describe unclassifiable varieties of sudden-onset, severe, generalized pain in the head, sometimes with vomiting.247 It may last up to a day or so. Clinically, the syndrome cannot be reliably distinguished

117

9781405127660_4_003.qxd 10/13/07 11:11 AM Page 118

118

Chapter 3 Is it a vascular event and where is the lesion?

Greater occipital nerve

Lesser occipital nerve

from SAH, but the chances of aneurysmal SAH are increased in the presence of female gender, epileptic seizures, loss of consciousness at onset, focal neurological symptoms (e.g. diplopia), vomiting or exertion preceding the onset of headache.194 The diagnosis is therefore made by exclusion, mainly of SAH. About 50% of these patients have a history of typical migraine or tension-type headache (with gradual onset). The prognosis is benign; a 3-year follow-up of 71 patients seen in hospital found identical recurrences in 12, again without evidence of SAH, whereas nearly 50% developed episodes of more obvious migraine or tension headache.248 Of 93 such patients identified in general practice and followed up for a median of 5 years, again none suffered SAH; recurrent attacks of ‘thunderclap headache’ occurred in eight patients, and 13 developed new tension headache or migraine.249 Headache is common in clinical practice, but ‘thunderclap’ headache is not. No physical sign can definitely exclude subarachnoid haemorrhage if a sudden onset headache persists for a few hours.

3.7.3 Investigations to confirm the diagnosis of subarachnoid haemorrhage Investigations are essential in making the diagnosis of subarachnoid haemorrhage (SAH), given the clinical features are relatively non-specific.250 Brain CT scan All patients presenting with a suspected recent SAH (i.e. within the last few days) should initially have an urgent brain CT to determine: • whether there is blood in the subarachnoid space;

Fig. 3.49 Diagram showing the anatomical distribution of the sensory innervation of the greater and lesser occipital nerves.

• the site of any SAH or intracerebral or intraventricular haemorrhage and therefore the likely cause (section 9.4.1 and Fig. 3.50); • the presence of any complications, such as hydrocephalus; • any contraindications to lumbar puncture such as cerebral oedema or haematoma with brain shift, or a large cerebellar infarct, when there is no blood evident on CT; • and whether there is any other intracranial abnormality that may account for the symptoms and signs. The sensitivity of CT in SAH depends on the amount of subarachnoid blood, the resolution of the scanner, the skills of the radiologist and the timing of the CT after symptom onset. The sensitivity is greatest in the first few days and falls thereafter, as blood in the subarachnoid space is resorbed (Fig. 9.19). In fact, the term ‘resorption’ may not always be appropriate to describe this process, diffusion and sedimentation being other explanations. CT evidence of subarachnoid blood can disappear very rapidly. If brain CT is done within 1–2 days after SAH onset, extravasated blood will be demonstrated in more than 95% of patients.202 But the chance of finding subarachnoid blood on brain CT then decreases sharply, to 50% on day 7, 20% on day 9, and almost nil after 10 days.251,252 If brain CT is done within 1–2 days after subarachnoid haemorrhage onset, extravasated blood will be demonstrated in more than 95% of patients. Of course, minute amounts of subarachnoid blood may be overlooked by the uninitiated (Fig. 3.51). Depending on the amount of blood in the cisterns and the delay before scanning, an ‘absent’ or ‘missing’ (isodense)

..

..

9781405127660_4_003.qxd 10/13/07 11:12 AM Page 119

3.7 Is it a subarachnoid haemorrhage?

Fig. 3.50 Brain CT scan (a) showing a 2- or 3-week-old intracerebral haemorrhage in the frontal lobes (arrow) of a patient who presented with an acute behavioural disorder and was misdiagnosed as being a hysterical alcoholic. Subsequent

catheter angiography (b) revealed an anterior communicating artery aneurysm (arrow) which had ruptured and bled into the frontal lobes.

cistern, or absent cortical sulci, may be the only clue to the presence of subarachnoid blood (Fig. 3.52). If subarachnoid haemorrhage is suspected and yet the brain CT scan appears normal, look carefully at the interpeduncular cistern, ambient cisterns, quadrigeminal cistern, the region of the anterior communicating artery and posterior inferior cerebellar artery, the posterior horns of the lateral ventricles and the cortical sulci. If blood is present in these sites, it may be isodense or slightly hyperdense, and hence the normally hypodense cisterns and sulci may be difficult to see and seem ‘absent’.

Fig. 3.51 CT brain scan showing a subtle amount of subarachnoid blood in the anterior part of the interhemispheric fissure (arrow), from a ruptured aneurysm of the anterior communicating artery.

..

..

Haemorrhage from an intracranial aneurysm can result not only in SAH but also in an intracerebral haemorrhage, which is easily seen on plain CT and generally persists for longer than subarachnoid blood, because the resorption of intraparenchymal blood, as seen on CT, occurs over several days to weeks rather than a few days.251 However, small intracerebral haemorrhages can also resolve very quickly, within days (section 5.4.1). About one-quarter of unselected patients who reach hospital alive after aneurysmal haemorrhage have an intracerebral haemorrhage on CT.224,251

119

9781405127660_4_003.qxd 10/13/07 11:12 AM Page 120

120

Chapter 3 Is it a vascular event and where is the lesion?

Fig. 3.52 Brain CT scan of a patient with subarachnoid haemorrhage showing isodense blood in the cortical sulci giving the appearance of ‘absent’ sulci. In particular, the sylvian fissures are not seen because they are filled with just enough blood to raise the density of the cerebrospinal fluid to that of brain parenchyma.

False positive evidence of SAH on CT scans A false positive diagnosis of SAH may be made on the CT brain scan, for example in patients who are comatose and brain dead (i.e. have no cerebral blood flow at the time of the scan). The CT scan not only shows cerebral oedema, but hyperdense material in the subarachnoid space which represents blood in congested subarachnoid blood vessels253–255(Fig. 3.53). Increased density of the tentorium and basal cisterns as a false positive sign of SAH on an unenhanced CT scan has also been described in polycythaemia,256 purulent meningitis,257 subdural haematoma,258,259 gliomatosis cerebri260 and spontaneous intracranial hypotension.261 Lumbar puncture According to a consecutive series from the Netherlands, the negative predictive value of CT scanning performed within 12 h of onset of sudden headache is 97%.202 In other words, there is an important small minority (of about 3%) with sudden headache and normal CT within 12 h who do have xanthochromia in the cerebrospinal fluid, and in whom angiography subsequently confirms a ruptured aneurysm. Therefore a lumbar puncture is necessary in any patient with sudden headache and a normal CT scan, even though the results will be normal in most patients.

Fig. 3.53 Brain CT scan of a patient who is brain dead from general hypoxia. There is generalized oedema, with effacement of the cisternal cerebrospinal fluid spaces. These spaces appear hyperdense, through venous stasis, which may falsely suggest subarachnoid haemorrhage.

Always do a lumbar puncture if the history is suggestive of subarachnoid haemorrhage and the CT scan (performed early, within a few days) is normal. Frequently, the cerebrospinal fluid will be normal too, but occasionally in this setting an abnormal cerebrospinal fluid will provide the only evidence of subarachnoid haemorrhage. Lumbar puncture without prior brain CT is potentially dangerous in patients with an intracerebral haemorrhage.262 Brain herniation may occur even in patients without focal signs or a decreased level of consciousness.263 Once the decision has been taken to do a lumbar puncture, the next requirement is to do it well. This is more difficult than it seems. Before drawing cerebrospinal fluid, the first rule is to wait until at least 6 and preferably 12 h have elapsed after the headache onset. This delay is absolutely essential because if cerebrospinal fluid obtained earlier turns out to be blood-stained, it is absolutely and irrevocably impossible to distinguish between

..

..

9781405127660_4_003.qxd 10/13/07 11:12 AM Page 121

3.7 Is it a subarachnoid haemorrhage?

blood that was there before (genuine SAH) and blood that was introduced by the needle (a bloody tap). With pre-existing blood in the cerebrospinal fluid, bilirubin will have been formed in the interval from SAH onset, from the breakdown of erythrocytes in the cerebrospinal fluid (see below), but of course not with a traumatic tap. A false positive diagnosis of SAH can be almost as damaging as a missed one, because insurance companies are bound to remain wary, despite negative investigations for an aneurysm. Never believe a colleague, however senior, who tells you that ‘the tap went so smoothly, it is impossible that the blood was traumatic’. Even the smoothest puncture can hit a vein. Also the ‘three-tube test’ (a decrease in red blood cells in consecutive tubes) is notoriously unreliable.264 Immediately proceeding with CT or MR angiography in all patients with bloodstained cerebrospinal fluid is not a good idea either, despite some people advocating this way of circumventing the ‘bloody tap’ problem: a small ( 1.5 cm diameter on CT/MR Tests At least one cardiac source of embolism should be identified (from high-risk group for ‘probable’ or medium-risk group for ‘possible’ cardiac embolism) Potential large artery atherosclerosis sources of thrombosis or embolism should be excluded Small artery disease Clinical Patient should have one of the traditional clinical lacunar syndromes and no evidence of cerebral cortical dysfunction, supported by history of diabetes or hypertension Imaging Normal CT/MRI or relevant brainstem or subcortical hemispheric lesion < 1.5 cm diameter Tests Potential large artery atherosclerosis sources of thrombosis or embolism, and cardiac sources of embolism should be excluded Other determined aetiologies Clinical Any symptoms compatible with an acute stroke Imaging CT/MRI findings of acute ischaemic stroke regardless of the size or location Tests Includes patients with non-atherosclerotic vasculopathies, hypercoaguable states or haematological disorders Potential large artery atherosclerosis sources of thrombosis or embolism, and cardiac sources of embolism should be excluded Undetermined aetiology Includes No aetiology determined despite extensive evaluation No aetiology determined but cursory evaluation Patients with two or more potential causes of stroke

As with the OCSP syndromes, a further problem is that of trying to classify the mechanism of stroke when patients are assessed in the hyperacute phase. In a study using the detailed TOAST protocol, the initial ( 1.2 for the hyperattenuated MCA : normal MCA, improved identification of true hyperattenuated MCA signs74 but needs to be confirmed in other studies. A hyperattenuated MCA mimic has been reported in herpes simplex encephalitis, where the low attenuation of the temporal lobe adjacent to the MCA made the MCA look whiter than it actually was.75 Increased haematocrit can also give the appearance of a hyperattenuated MCA, although usually bilateral, not unilateral.76 This sign was always related to occlusion of the MCA in patients who had angiography, giving a specificity of 100%.68 But the hyperattenuated MCA sign was present in only about one-third of patients in the Prolyse in Acute Cerebral Thromboembolism (PROACT) trial on CT scanning within 6 h of stroke – as all had MCA main stem or major branch occlusion on angiography within 6 h of stroke as a trial inclusion criterion, providing a sensitivity of around 33%.77 The hyperattenuated artery sign is more frequent in patients with more severe strokes who are scanned early, e.g. in 50% of 36 patients scanned within 4 h of MCA territory stroke;70 18% of 620 patients scanned within 6 h of stroke;78 and 41% of 620 patients scanned within 12 h of MCA territory stroke;68 but in only 5% of a more general population of acute stroke patients.79 In the first 500 patients randomized in the Third International Stroke Trial (IST3) of tissue plasminogen activator vs control, 41% had a hyperattenuated artery sign on the baseline scan within 6 h of stroke and this was associated with having a total anterior cerebral infarction, i.e. severe stroke. Of those with a hyperattenuated MCA sign on admission, the hyperattenuated artery was still visible on 24-h follow-up CT in 60% (personal data). Patients with the sign on baseline CT more frequently had more extensive tissue hypoattenuation and swelling, and a significantly worse ASPECTS score (5.6 vs 8.3). The hyperattenuated artery sign on CT is a reasonably reliable indicator of an occluded cerebral artery when the hyperattenuation is visible at a distance from the carotid siphon. An absent sign is certainly not a reliable indicator of a patent artery.

Evolution of the CT appearances of infarction Evolution of the CT appearances of infarction is illustrated in Fig. 5.16. Initially, the lesion has ill-defined margins and slight swelling and is slightly hypoattenuated (dark) compared with normal brain. The infarct becomes more clearly demarcated and hypoattenuated during the first few days.80,81 The swelling is usually maximal around the third to fifth days and gradually subsides during the second and third week.82 Infarct swelling can occur very rapidly – within the first 24 h – to cause brain herniation, but generally only with large infarcts. In general, swelling is most apparent in large infarcts but presumably also occurs in small ones, although it is difficult to see and is probably clinically less important. Extensive infarct swelling can compress adjacent normal brain and cause brain herniation (sections 11.5 and 12.1.5). It is an impression that younger patients are more prone to severe infarct swelling than older patients for a similar size of infarct, but perhaps older people – because of brain atrophy – simply have more space inside the cranial cavity to accommodate the swollen brain. The presence of recent haemorrhage in the infarct produces areas of increased density relative to both normal brain and the infarcted tissue, and further contributes to any swelling. The amount of infarct swelling, and the rate at which it appears, varies between patients for reasons which are still not well understood. Limited data from observational studies83 and thrombolysis trials84 indicate that swelling is associated with large infarcts and is more severe in patients whose occluded artery fails to recanalize (spontaneously or pharmacologically) (Fig. 5.20). This notion is supported by very limited data that suggest that treatment with thrombolysis up to 6 h after stroke may reduce the risk of massive infarct oedema (Fig. 5.21). During the second week, the infarct gradually increases in attenuation, sometimes becoming isointense and so indistinguishable from normal brain. This is the socalled ‘fogging effect’ and without close inspection even quite sizeable infarcts may be overlooked (Figs 5.16 and 5.22). Indeed, the ‘fogging effect’ may make the infarct impossible to see on CT scans performed at this time. This may lead to serious underestimation of infarct size. It does not occur in all infarcts and varies from 54% of patients scanned 10 days after onset85 to all patients at some time examined with six consecutive CT scans within 42 days of stroke.86 The ‘fogging effect’ may last up to 2 weeks and then the infarct becomes progressively more hypoattenuated (black). Eventually, a sharply demarcated, atrophic, hypoattenuated (similar to cerebrospinal fluid) defect remains (Fig. 5.16). Although old infarcts usually have sharply

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 201

5.4 CT scanning

Fig. 5.20 An unenhanced CT brain scan from a patient at 4 h after sudden right hemiparesis showing (a) a hyperattenuated left middle cerebral artery (MCA) (arrow) and (b) early hypoattenuation of the left basal ganglia insular and temporal cortex. Twenty-four hours later, (c) the left MCA is still hyperattenuated (arrow) and (d) there is massive swelling of the infarct with midline shift to the right, obstruction and dilatation of the right lateral ventricle.

(a)

(b)

(c)

(d)

Study or subcategory

OR (fixed) 95% Cl

OR (fixed) 95% Cl

ECASS 1995 NINDS 1995 ECASS II 1998 ATLANTIS B 1999

1.54 0.68 0.44 0.82

Total (95% Cl) Total events: 208 (Treatment), 239 (Control) Test for heterogeneity: Chi2 = 6.38, df = 3 (P = 0.09), F = 53.0% Test for overall effect: Z = 2.14 (P = 0.03) 0.1

0.2 0.5 Favours treatment

Fig. 5.21 Forest plot showing the limited data from the randomized controlled trials in acute ischaemic stroke of the effect of thrombolysis on major infarct swelling. The vertical line shows an odds ratio (OR) of 1.0 (no treatment effect), the horizontal lines represent the 95% confidence intervals around the point estimates of each trial, the boxes are the point estimates for each trial, their size being in proportion to the

..

..

1

5 2 Favours control

[0.79, 3.02] [0.49, 0.93] [0.19, 1.03] [0.07, 9.56]

0.74 [0.57, 0.98]

10

amount of information each contains, and the diamond represents the overall OR with its 95% confidence interval to the left of the vertical line of no effect, suggesting that thrombolysis reduces infarct swelling. Note the data for ATLANTIS B are only for the subgroup of patients randomized within 3 h of stroke onset (data for the whole trial are not available).

201

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 202

202

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

(a)

(a)

(b)

Fig. 5.22 CT brain scans to demonstrate ‘fogging’. (a) Three days after onset of a right hemiparesis there is an obvious infarct (hypoattenuated area) in the left parietal cortex and adjacent white matter (thin arrows). (b) At 14 days after onset the recent infarct in the anterior left parietal region is almost invisible – it is now mainly iso-attenuated with normal brain and there is no mass effect – as a result of ‘fogging’.

(b)

Fig. 5.23 CT brain scan (a) without and (b) with intravenous contrast at 10 days after a right basal ganglia infarct in a 10year-old girl. There is marked serpiginous enhancement of the infarct (arrows) attributed to breakdown of the blood–brain barrier (sometimes also referred to previously as ‘luxury perfusion’).

demarcated borders, so making it possible to ‘age’ infarcts approximately, it is not always possible to tell with absolute certainty how old an infarct is, something that may be overlooked when ascribing particular clinical symptoms to lesions seen on CT. It is not always possible to tell the age of an infarct on CT, particularly in the case of small deep infarcts. Therefore, do not assume that a particular hypodense area on CT is necessarily relevant to a recent stroke – it could be old and irrelevant. Effect of intravenous contrast In the first week after stroke, intravenous X-ray contrast usually has little effect on the appearance of an infarct,

although some enhancement of the gyri may be seen.87,88 But, from the end of the first to the third week, more striking contrast enhancement occurs.53,89 The likely cause is a combination of blood–brain barrier breakdown, neovascularization and impaired autoregulation, and the resulting appearance on CT was previously referred to as ‘luxury perfusion’ (because we seldom use contrast at this stage after stroke, this feature is rarely seen).90 The enhancement can be very marked in children and young adults (Fig. 5.23). The tendency to enhance with contrast gradually resolves over the following few weeks. It was suggested some years ago that stroke patients might deteriorate as a result of intravenous contrast.91 It is certainly possible that extravasation of neurotoxic contrast agents could be harmful, but most patients

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 203

5.4 CT scanning

described had large infarcts with a poor prognosis in any case. In recent years, contrast has been used much more frequently in patients with acute ischaemic stroke, although usually in the first few hours, for perfusion CT, CT angiography (section 5.4.3) or during catheter angiography prior to and during intra-arterial thrombolysis. However, the contrast used now is ‘non-ionic’ and considered to be much less toxic (including neurotoxic) than contrast agents available in the 1970s and 1980s which were ‘ionic’. Contrast extravasation into subdermal tissues at the time of intravenous injection certainly causes a nasty local tissue reaction and so contrast extravasating locally into the brain through the ischaemic arteriolar endothelium might also cause a nasty local tissue reaction and worsen outcome, but we have no information on whether this is likely to be relevant in practice. Another potential problem is that all non-ionic contrast agents so far tested significantly slow both spontaneous and thrombolysis-induced thrombus lysis,92 prolonging the time to clot lysis threefold in a canine experimental model.93,94 With the increasing use of CT perfusion prior to intravenous thrombolysis, and of intra-arterial thrombolysis, this delay in lysis time may increase the risk of a poor outcome (in addition to the delay to starting treatment caused by the time taken to do the extra procedure) which may inadvertently be causing more harm than good. There is virtually no information on whether acute stroke patients given contrast (intravenously or intraarterially) have a worse prognosis than those not given contrast. Many of these patients have severe strokes and without a randomized trial comparing patients who received contrast with those who did not (all other factors being equal) it would be difficult to detect whether

contrast is harmful or not. Thus, until there is better evidence that the information provided by perfusion CT improves clinical outcome, or that intra-arterial thrombolysis is better than intravenous, the use of X-ray contrast in acute stroke should probably be avoided unless absolutely necessary. Intravenous contrast is rarely required to clarify the CT diagnosis in acute stroke, and should in general be avoided unless required for CT angiography or perfusion imaging. X-ray contrast agents interact with all thrombolytic drugs tested so far to significantly prolong the time to thrombus lysis, so should probably be avoided. Thus neither perfusion CT nor intra-arterial thrombolysis should be used routinely until there is good evidence to support this practice. How often does the appropriate infarct become visible on CT? In general, the more severe the stroke, the greater the volume of infarcted tissue, and so the more often the infarct is visible on CT scanning. Among 993 stroke patients scanned up to 99 days after stroke, 60% had a visible relevant infarct on CT within the first 24 h, rising to 70% by 72 h.95 A greater proportion of the total anterior circulation infarction (TACI) patients had a visible infarct than those with small cortical, lacunar and posterior circulation infracts, no matter how early or late they were scanned after the stroke. And among 13 000 patients randomized after CT scanning in the International Stroke Trial the highest proportion of patients with visible infarction was in the TACI group, the proportion increasing with time from onset to randomization in the first 48 h (Fig. 5.24).96 Small infarcts

80

Time (days) 0–3 4–6 7 – 12 13 – 24 25 – 48

Fig. 5.24 Data from the First International Stroke Trial showing the effect of time and stroke clinical syndrome on the visibility of infarcts on CT brain scanning. The visibility of increases with time over the first 2 days from the onset of the stroke, and the larger the infarct the more often it is visible (i.e. total anterior circulation infarcts, TACIs, show up more than lacunar infarcts, LACIs, etc.). PACI, partial anterior circulation infarct; POCI, posterior circulation infarct.

..

..

% Patients with visible infarction

70 60 50 40 30 20 10 0

TACI

PACI

POCI

LACI

203

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 204

204

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

appear later than large ones, because there is less tissue to alter their density. Therefore, lacunar infarcts are less likely to show up in the first 24 h and sometimes do not do so at all (Fig. 5.24).95–100 Small infarcts in the brainstem and cerebellum are particularly difficult to visualize with CT because of artifacts arising from the petrous bones. In any series, the proportion of patients with a visible infarct also depends on local factors, such as case mix, time to first scan, whether or not the scan is repeated, generation of CT scanner, etc. Not surprisingly, therefore, the proportion of TACI patients with a visible infarct is lower in some studies than others.101–104 It is uncertain just how many patients with a clinically definite stroke never have an appropriate infarct visible on brain CT, but the proportion is probably quite high (up to 50%) depending on the timing of the scan, the age of the scanner, the thickness of the slices, the cooperation of the patient, the size and age of the infarct, the location (e.g. the brainstem is a difficult area to visualize infarcts on CT), the vigilance of the radiologist, and possibly some pathophysiological characteristic of the lesion itself. However, the main clinical reason for performing the scan is to exclude haemorrhage (or tumour or infection) as the cause of the symptoms and, if the scan is normal, the presumptive diagnosis is of an ischaemic event if the clinical picture is compatible with a stroke. If, in a particular patient’s case, it is absolutely critical to see positive evidence of an ischaemic stroke, then either the CT scan could be repeated at 3–4 days when the infarct is more likely to be visible, and certainly before 7–21 days when fogging is likely to obscure the infarct again. Brain CT only shows the appropriate infarct in about half of patients with an ischaemic stroke overall – those with symptoms of a more extensive stroke are more likely to have a visible relevant infarct than those with symptoms of a minor stroke. Therefore, absence of a visible infarct does not mean that the patient has not had a stroke; a patient with a clinical diagnosis of stroke, and an early CT scan which is normal, has still had an ischaemic stroke. How well do the clinical and CT diagnosis of the site of the stroke lesion correspond? The middle cerebral artery (MCA) territory is the most frequently affected on CT (60%), followed by the posterior cerebral artery (14%), the anterior cerebral artery (5%), the posterior fossa (5%) and the major artery territories combined or boundary zones (14%).105 There is reasonable agreement between the stroke syndrome

defined clinically using the Oxfordshire Community Stroke Project (OCSP) classification,106 with the site of the infarct as demonstrated by cross-sectional imaging (sections 4.3.2, 4.3.3, 4.3.4 and 4.3.5). Of course, in all the studies which assessed this agreement, a proportion of the patients had ‘normal’ imaging – that is, the imaging did not show an appropriate recent infarct – but it is unlikely that any imaging technique, no matter how sophisticated, will ever be able to demonstrate all infarcts. One relatively common discrepancy between the clinical and radiological diagnosis of the ischaemic stroke subtype is that about 20% of patients thought to have a mild cortical infarct clinically actually had a recent and lacunar infarct on imaging in a relevant site, and about 15% of patients thought to have a lacunar stroke clinically actually had a small and relevant cortical infarct radiologically.107,108 These patients behave epidemiologically more like the stroke type suggested by the imaging lesion than by the clinical syndrome, and so for them it is best to reclassify the stroke subtype from cortical to lacunar, and from lacunar to cortical on the basis of the imaging.107 The clinical diagnosis of lacunar stroke is particularly difficult in the first 12–24 h after onset.109,110 This implies that about one-fifth of all lacunar or mild cortical strokes are misclassified clinically and that studies of pathogenesis, risk factors and prognosis for lacunar or mild cortical infarction have probably been ‘clouded’ by inadvertent inclusion of some lacunar infarcts in the mild cortical group and vice versa. Because MR diffusionweighted imaging (DWI) is the most sensitive method of detecting infarcts underlying mild stroke syndromes (section 5.5.2), in future studies where it is necessary to distinguish lacunar from cortical ischaemic stroke this will be needed to establish the diagnosis with the best degree of reliability.111 In studies of stroke lesion site diagnosed clinically vs radiologically, which have used other more complex clinical classifications, the level of agreement between the clinician equipped with the traditional neurological tools of pin and tendon hammer, and the radiologist armed with a scanner, has been poorer.112 Even the National Institute of Neurological Diseases and Stroke (NINDS) classification,113 the Trial of ORG 10172 in Acute Stroke Therapy (TOAST) Trial classification,114 the Stroke Data Bank classification,115 all of which use risk-factor based classification and many investigations to apportion a cause of stroke, still find up to 40% of patients are unclassifiable. A further point to note is that classifications which use features other than simply the neurological symptoms and signs can introduce bias into studies of risk factors for stroke.116

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 205

5.4 CT scanning

Does infarct visibility on CT (early or established) have prognostic significance? Among 3468 patients from 15 studies examining the relationship between infarct signs present on CT scanning within the first 6 h after stroke and functional clinical outcome (assessed at least 1 month or more after stroke), the presence of any early infarct sign increased the risk of poor functional outcome on univariate analyses (odds ratio for any early infarct sign 3.11, 95% CI 2.77–3.49).59 This increased risk of poor functional outcome was present for all early infarct signs for which it was possible to extract data: any hypoattenuation vs no hypoattenuation (OR 2.73), more than 1/3 middle cerebral artery (MCA) territory affected vs less than 1/3 or normal (OR 6.44), less than 1/3 MCA vs normal CT (OR 2.66), and hyperattenuated artery vs no hyperattenuated artery (OR 2.09).59 Studies comparing the ASPECTS CT rating score and outcome indicate that an ASPECTS score < 7 is associated with poor outcome.61,117 There is less information on whether thrombolysis alters the relationship between CT infarct signs and poor outcome. Two studies which sought an ‘early infarct sign–thrombolysis’ interaction found no evidence that thrombolysis given up to 3 h118 or 6 h after stroke119 in the presence of early infarct signs worsened the outcome over that due to early signs alone. The ASPECTS score applied to the PROACT II trial data showed improved functional outcome for those treated with thrombolysis compared with control in patients with ASPECTS > 7 (OR 5) but not in those with ASPECTS of 7 or less (OR 1).120 When ASPECTS was applied to the NINDS, ECASS 2 and ATLANTIS B trials, there was also no association between an ASPECT score < 7 and worse outcome with thrombolysis,121–123 although ECASS 2 patients with a low ASPECTS were at increased risk of haemorrhagic transformation after thrombolysis.123 Early infarct signs are subtle. They are therefore difficult to recognize and may have poor inter-observer reliability59 (see below). Also CT technology has improved since the thrombolysis trials from which these data came (in the first half of the 1990s). Therefore, further data from current trials (e.g. the Third International Stroke Trial, www.ist3.com) are needed to determine whether thrombolysis does or does not modify the early infarct sign–poor prognosis relationship, and if so whether the treatment effect is different in patients with increasingly obvious early infarct signs. At present there is no evidence that patients with early CT signs of ischaemia are at increased risk of poor outcome if given thrombolysis.

..

..

Established visible infarction on CT has a clearer adverse independent relationship with prognosis than early infarct signs, it is easier to recognize, and there are more studies with larger sample sizes so the data are likely to be more robust95,96 (note the the odds ratios for early infarct signs are mostly from univariate analyses which may be why they appear larger than those for established infarct signs). In the International Stroke Trial (IST), visible infarction on the CT scan within 48 h of the stroke onset was independently associated with increased death within 14 days (odds ratio (OR) 1.17), and of death or dependency at 6 months (OR 1.42), an absolute increase of 13%, or 130 per 1000 more dead or dependent patients with visible infarction than without it.96 In our hospital stroke registry study, of 993 patients scanned mostly within the first week of stroke, the presence of a visible infarct was associated with an increased risk of poor functional outcome (OR 2.5; 95% CI 1.9–3.3) and death (OR 4.5; 95% CI 2.7–7.5) at 6 months, even after adjusting for time from stroke to scan, and important clinical prognostic variables.95 Others have also found that established visible infarction conferred an adverse prognosis.124–126 Therefore, it seems that a visible established infarct indicates a profound depth of ischaemia, which reflects a marked drop in blood flow to a largish area of brain, which suggests that the outcome will be poor. Visible infarction on CT, whether within the first 6 h or several days after the stroke, is independently associated with a poor functional outcome compared with no visible infarction. For patients scanned within the first 6 h, there is as yet no definite evidence that thrombolysis of those with early visible infarct signs either worsens their prognosis or is a clear contraindication to treatment. Inter-observer reliability in the analysis of CT scans of stroke patients very early after the stroke, and in established infarction Our systematic review of the inter-observer reliability of early infarct signs on CT scans up to 6 h after stroke (complete up to mid-2003) found 15 studies in which a median of 30 CT scans were rated by six observers.59 Few studies gave any information on the demographics of the patients from whom the scans came. Between one and five early CT signs were analysed including: hypoattenuation, swelling, hyperattenuated middle cerebral artery sign, loss of grey and white matter definition, loss of basal ganglia outline, loss of the insular ribbon, infarction of more or less than one-third of the middle cerebral artery (MCA) territory, and cortical sulcal effacement. The reference standard was mostly a neuroradiologist

205

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 206

206

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

with access to a follow-up CT scan. In the 15 studies, the mean prevalence of any early infarct sign was 61%. For any early CT sign, the inter-observer agreement ranged from 0.14 to 0.78 (kappa). The mean sensitivity and specificity (and range) for infarct sign detection were 66% (20–87%) and 87% (56–100%) respectively. Inter-observer agreement was worst for hypoattenuation and best for the hyperattenuated artery sign. The one study of inter-observer reliability of the ASPECTS scoring system available at the time of the study, which suggested very good inter-rater reliability of the scoring system, has since been replicated in other datasets with similar results.127 The information, where given, suggested that experience improved detection, but knowledge of symptoms did not.59 These rather limited data suggest that some early infarct signs are detected better than others (e.g. mass effect better than hypoattenuation) so there may be ways in which the inter-observer reliability for early infarct signs could be improved by focusing the observer’s attention on specific features. Finding out specifically what makes more experienced observers more reliable would also help. Our ACCESS Study (Acute Cerebral CT Scanning for Stroke, www.neuroimage.co.uk) was a very large interobserver reliability study of early infarct signs on CT to see whether some signs were more reliably detected than others, whether scoring systems improved inter-rater reliability, whether ‘distracters’ on the scan (e.g. old infarcts) affected relability, and whether observer experience had any effect. We used a novel web-based method of showing digital scans to observers, recording some basic demographic information about the observers, and then their interpretation of each scan on a structured questionnaire via the same web page. The results to date, on 207 raters from 36 countries who read all 63 scans, and taking neuroradiologists as the reference standard, indicate that: • hypoattenuation and swelling are recognized to an equal degree; • hyperattenuated arteries are seen better than parenchymal hypoattenuation or swelling; • any scoring system (whether 1/3 MCA,63 ASPECTS61 or our own method62) improves inter-rater reliability; • distracters (e.g. old infarcts) worsen inter-rater reliability, particularly among less-experienced individuals. On CT scans up to 6 h after stroke, signs of hypoattenuation and swelling are detected equally reliably, using any scoring system helps improve interrater reliability for detection of early infarct signs, and any distracter (e.g. old infarcts, leukoaraiosis and atrophy) reduces early infarct detection. The emphasis on detecting subtle early signs of ischaemia may have deflected attention away from

looking for other important causes of the stroke which themselves may have only subtle CT signs. Worryingly, in a study of observer reliability of early infarct and haemorrhage detection on CT, only 17% of admitting physicians and 40% of neurologists were able to recognize intracranial haemorrhage with complete reliability.49 In established infarction, high levels of agreement (substantial to perfect) have been demonstrated.128,129 However, where observers had free access to relevant clinical details, their interpretation of the scans may have been biased,129 although there is little information about what effect this has on scan interpretation. When neurologists and radiologists reviewed the same two CT brain scans from patients with a lacunar stroke (camouflaged by an assortment of other scans), accompanied by misleading clinical information, the diagnosis of lacunar infarction did not appear to be biased by informing the observer that the patient was thought clinically to have had a stroke.100 This lack of bias, even with knowledge of the clinical details, may have been because the study was small and may not have been a true reflection of the difficulties encountered in routine practice when faced with a CT scan showing multiple ‘holes in the brain’ and generalized atrophy. The former makes the diagnosis of recent lacunar infarction difficult, because it is impossible to decide which ‘hole’ is the relevant one unless serial scans show a new ‘hole’ developing, and the latter makes the diagnosis of a small cortical infarct difficult (when is a large sulcus actually an infarct?). In established infarction, the agreement as to the site of infarction, the amount of swelling, and haemorrhagic transformation was excellent between experienced neuroradiologists and good between trainee radiologists when using a simple stroke CT classification system.62 Online training in CT intepretation As, in the interests of speed, most stroke CT scans in the very acute phase are likely to be read by the admitting neurologist or stroke physician (who are both probably less experienced than a dedicated neuroradiologist), methods of improving interpretation of CT scans from patients presenting with acute stroke are required. At the time of writing, there are two sources of scan examples for trainees available over the web, and more in development. At www.neuroimage.co.uk the reader can register to read CT scans from acute stroke patients over the web in their own time, and then obtain feedback comparing their performance with experts and observers from other specialties, as well as seeing what the follow-up scan showed. At www.ist3.com click on ‘see BASP CT Training Series’ for a demonstration of a variety of conditions presenting as acute stroke.

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 207

5.4 CT scanning

Appearances of the infarct: can we infer anything about its cause? A typical established large artery infarct is wedge-shaped, of decreased attenuation compared with normal brain, sharply demarcated, and occupies a recognizable vascular territory (Fig. 6.20c).105,130 It is usually due to embolism from a proximal arterial source or the heart. Lacunar infarcts, thought to arise from occlusion of a single perforating artery (section 6.4), are arbitrarily less than 1.5 or 2 cm in diameter (section 4.3.2), usually rounded in shape, and sited in the deep white matter, basal ganglia and pons (Fig. 6.20d).97,98 Boundary zone infarcts lie in areas of brain at the edge of the large artery vascular territories, i.e. in the parietooccipital region for the middle–posterior cerebral arteries and over the vertex for the anterior–middle cerebral arteries boundary zone, or in the internal boundary zone in the centrum semiovale at the junction of the deep and superficial arterial supply areas (section 4.2.4).130–132 However, the boundary zone areas are potentially quite extensive, and vary between and within individuals (section 4.2.4 and Fig. 6.21),133 so that in any one individual patient it can be very difficult to decide on the

100

basis of brain imaging whether an infarct has arisen from occlusion of a cortical branch of the middle cerebral artery (MCA), which might be embolic, or from poor perfusion in the boundary zone as a result of established internal carotid artery occlusion (section 6.7.5).132,134 Striatocapsular infarcts are larger than lacunes and occur in the deep white matter and basal ganglia, with preservation of the overlying cortex (section 4.2.2). They are thought to arise from transient occlusion of the MCA mainstem, prolonged occlusion of the MCA mainstem with good cortical collaterals, or occlusion of multiple lenticulostriate artery origins from atheroma of the MCA.135–137 This pattern is illustrated in Fig. 6.20e. In our hospital-based stroke registry, we attempted to assign a cause of stroke in all patients admitted during one sample year.138 Among 479 patients, those with large cortical infarcts had the greatest proportion of arterial (59%) or cardiac (29%) embolic sources, or both (15%), followed by small cortical infarcts (45%, 18% and 5%, respectively); lacunar infarcts (33%, 8% and 4%, respectively); and posterior circulation infarcts (32%, 9% and 4%, respectively) (Fig. 5.25). Thus, while patients with a cardiac source of embolism are at greater risk of having

n = 34

n = 100

n = 72

n = 56

n = 119

n = 32

n = 20

Total anterior circulation infarction

Partial anterior circulation infarction

Lacunar infarction

Posterior circulation infarction

Cerebral transient ischaemic attack

Amaurosis fugax

Retinal infarction

% of patients

80

60

40

20

0

Fig. 5.25 The frequency of various sources of emboli in patients with different ischaemic stroke subtypes, ocular and hemispheric transient ischaemic attacks (prepared by Dr Stephanie Lewis, Edinburgh).138

..

..

Source of embolus Cardiac Both cardiac and arterial Arterial Neither

207

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 208

208

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

a large cortical than any other type of infarct (if they have a stroke), 74% of the patients who do have a large cortical infarct have an arterial source of embolism. Patients with several recent infarcts in more than one vascular territory are more likely to have a proximal embolic source (e.g. myocardial mural thrombus, aortic atheroma), but in other circumstances, the use of a complex classification or multiple imaging modalities does not solve the problem of deciding what caused the stroke in patients with more than one possible cause. It is not valid to assume that a particular size, shape and position of an infarct on imaging was definitely caused by a particular embolic source or in situ thrombosis (and of course finding a potential source of embolism does not necessarily mean that it was the cause of the stroke). In up to 15% of patients (probably a greater proportion the harder one looks) there is more than one potential cause of the stroke. As indicated above, about one-fifth of patients presenting with a cortical stroke syndrome (i.e. partial anterior circulation infarction) are found on brain imaging to have a recent lacunar infarct in the correct hemisphere without any sign of the expected cortical infarct. Equally, between one-sixth and one-fifth of patients with a lacunar infarct syndrome may have a recent cortical infarct on brain imaging with no hint of any subcortical lesion. This suggests that the syndrome (and so anatomical location of the brain lesion) attributed to an ischaemic stroke following clinical examination could and indeed should be modified by the position of any recent, and likely to be relevant, infarct on brain imaging, as this better reflects the underlying vascular lesion and so improves management and resource use.

(a)

(b)

The anatomical location attributed to an ischaemic stroke following clinical examination may need to be modified if brain imaging shows a recent and probably relevant infarct in a different territory to that expected clinically. For example, a patient with a clinical lacunar syndrome, but whose CT scan shows a recent cortical infarct in the relevant hemisphere, should be regarded as being similar to a patient with a cortical syndrome (i.e. at high risk of early recurrent stroke, high probability of ipsilateral carotid stenosis or cardiac source of embolism).

Distinction of arterial infarcts from other conditions on CT While many infarcts as a result of arterial occlusion are easy to diagnose on imaging from their site, shape, density and appropriate clinical features, other lesions occasionally produce very similar appearances, which can be confusing. Viral encephalitis, if relatively focal in distribution, can appear exactly like an infarct, although this is rare. The typical appearance of herpes simplex encephalitis with involvement of the medial temporal lobes should not cause confusion, but we have seen patients with lowdensity areas in the temporoparietal region associated with rising viral titres, which resolved following treatment with aciclovir (Fig. 5.26). Encephalitis can also give the impression of a hyperattenuated middle cerebral artery,75 possibly because the hypoattenuated temporal cortex (due to encephalitis) makes the adjacent middle cerebral artery look more hyperattenuated than it really is. As ever, it is vital that the imaging is reviewed with all the clinical information and, if there is any doubt (e.g. fever, subacute onset), other diagnostic tests must be

Fig. 5.26 Unenhanced CT brain scan (a) in a 70-year-old woman with a right hemiparesis, confusion and drowsiness for 6 h, showing an extensive area of low attenuation in the left temporoparietal region including the basal ganglia. There is loss of the outline of the normal basal ganglia (thick arrows) and loss of the overlying cortical sulci indicating slight swelling (compare with the easily seen right parietal sulci; thin arrows). The initial clinical diagnosis was of a stroke (infarct), but 1 week later (b) the patient deteriorated and the CT scan showed extensive haemorrhage in the left temporoparietal region. Subsequent postmortem confirmed the diagnosis of herpes simplex encephalitis.

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 209

5.4 CT scanning

Fig. 5.27 CT brain scan showing a tumour which initially looked like an infarct in a 60-year-old man who had presented with ‘a stroke’. (a) The unenhanced scan shows a left temporal low attenuation area (the ‘infarct’) with a high attenuation dot in the sylvian fissure (the ‘hyperattenuated artery’; arrow). This was misinterpreted as an acute left middle cerebral artery infarct. (b) Several weeks later a CT scan with intravenous contrast shows that the lesion had grown, with enhancement, consistent with a tumour. It was a secondary deposit from a colon primary.

(a)

carried out, including an electroencephalogram, MR scan and cerebrospinal fluid examination. Purulent cerebritis can look like an infarct, although the lesion is usually not wedge-shaped, it involves more white matter than cortex, and the clinical picture should allow the distinction to be made. Tumours: Occasionally, a peripheral metastasis with a large amount of white matter oedema can mimic an infarct on CT, although the density is usually lower than expected for an infarct, and administration of contrast may show up a cortical nodule (Fig. 5.27). If there is still doubt, either a repeat CT a few weeks later (because infarcts and tumours evolve differently, Fig. 5.4) or an MR examination (pinpoint metastases or a central tumour nodule may be detected) will usually determine the cause of the abnormality. Venous infarction is not uncommonly mistaken for an arterial infarct. The difference between them is discussed in section 5.8. Time is a useful diagnostic tool, ‘a chronogram’. If in doubt whether a lesion is a tumour or infarct on CT, repeat the scan in a few weeks. Infarcts get smaller (usually) due to ex vacuo effect, whereas tumours stay the same or get bigger.

5.4.3 CT perfusion imaging and CT angiography Cerebral perfusion imaging with CT is a technique which has come to prominence in the last few years as the technology has improved, enabling rapid ‘cine’ scanning. At the time of writing, spiral and multislice scanners were able to provide sufficiently rapid data acquisition at four slices in the brain to enable the data to be processed to produce cerebral blood flow, cerebral blood volume and mean transit time maps.139 This technique requires an intravenous bolus of about 50 mL of iodinated contrast

..

..

(b)

by pump injector and so exposes the patient to the lysis-retarding effects of the contrast agent described in section 5.4.2. The raw data then need to be processed, sometimes on a separate workstation, to produce the actual perfusion maps. This may take several minutes. CT perfusion imaging is relatively more accessible than MR perfusion imaging, and can be performed during the same examination as the plain CT brain scan. Hence it provides limited data on cerebral perfusion to many who do not have access to MR perfusion imaging or other techniques.139,140 There is considerable interest in whether any lesion seen on the different forms of CT perfusion imaging indicates ‘tissue at risk of infarction’ and so helps guide thrombolytic treatment. However, so far, relatively few observational case series of patients undergoing CT perfusion have been published. All suggest that CT perfusion may improve outcome prediction over plain CT alone,141 identify penumbral tissue142,143 or thresholds of tissue viability,144 but all used very early time points to assess ‘final infarct extent’ (2–7 days) and included a mixture of some thrombolysis-treated and some not thrombolysis treated patients. Larger studies141–143 suggest that relative mean transit time (rMTT) maps predict final infarct extent in the absence of recanalization and that cerebral blood volume maps predict final infarct extent if there is early recanalization (i.e. infarct core),143 although a combination of CBF and CBV may have highest sensitivity and specificity for predicting infarct core.142 These results are preliminary and have yet to be tested in a prospective trial where patients are randomly allocated to thrombolysis vs control. Thus, more work is required to determine whether or not CT perfusion identifies a subset of patients who are more likely to benefit from thrombolysis than can be identified from plain CT alone, and if so which perfusion lesion is the best predictor, or whether the additional time taken

209

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 210

210

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

to perform the CT perfusion and the thrombolysisretarding properties of CT contrast agents negate the beneficial effects of the additional information gained. CT angiography (CTA) has also become more practical and accurate with improvements in CT technology. It requires rapid scanning with a spiral or multislice scanner through the skull base and circle of Willis during an intravenous injection of about 50 mL of iodinated contrast, followed by reconstruction of the data into axial, coronal, sagittal or 3D ‘angiographic’ images of the arteries. This technique can identify major intracranial and branch artery occlusions, and so might improve selection of patients for thrombolysis if, for example, only patients with a visible arterial occlusion were found to benefit.145 However, as with CT perfusion, it takes extra time to acquire and process the CT angiographic images, the contrast agent may delay thrombolysis, and it is unclear whether the additional information really does improve patient selection because there have been no randomized studies of thrombolysis vs control based on CTA. Indeed, the available evidence (n = 151 patients) suggests that CTA probably does not improve outcome prediction beyond that of clinical stroke severity alone and therefore may not aid decision making about thrombolysis treatment either.146 It is useful to note that the NIHSS stroke severity score is closely related to the probability of having an arterial occlusion on angiography within the first 6 h of stroke – patients with proximal intracranial arterial occlusion have worse NIHSS scores.147 Therefore, if it turns out that only patients with a large intracranial artery occlusion should be treated with thrombolysis, it may make sense to use the NIHSS score within 6 h of stroke as a surrogate for angiography-detected intracranial arterial occlusion. This would enable faster treatment delivery, avoid the cost of angiography (whether MRA, CTA or catheter) and the risk of catheter angiography, and minimize exposure to X-ray contrast agents. Until further reliable data are available, CTA is a useful diagnostic tool in occasional patients (but not for routine stroke investigation) but further research is needed to sort out its clinical application.

5.5 Magnetic resonance imaging

Historical note and practical points regarding use of MR Magnetic resonance imaging (MRI) was first used as a clinical tool in the early 1980s. The equipment is more

expensive than for CT scanning, both to purchase and to run, and MR was initially little used in acute stroke. It is not a practical technique for many acutely ill patients for the following reasons. • The patient must be placed inside a tube-like structure, which creates difficulties for being able to see and communicate with the patient, and for access for monitoring and administration of anaesthetics. • The patient must lie still, usually for at least 5 min at a time, although recent fast-scanning techniques mean that diagnostic images may be acquired in seconds. • Because of the need to perform several sequences, the whole scan takes longer than a CT brain scan (i.e. longer to obtain a diagnosis), even with ultrafast scanners which are now becoming available.148 • Fifteen to twenty per cent of patients with acute stroke may not be suitable for MR on medical grounds because they are too ill, vomiting or unable to protect their airway, or because of a contraindication to MRI.149–151 • Many stroke patients are confused, restless and frightened by the noise and vibration of the scanner, and so move during the scan, which impairs the images. • It is inadvisable for patients who cannot protect their airway to be placed supine for any length of time after stroke,152 regardless of the value of the information so obtained. • About 20% of patients with acute stroke become hypoxic while in the MR scanner.150 • In the only study which actually recorded it, about one-third of acute stroke patients required some form of physical intervention while in the scanner (reassurance, assistance while vomiting, administering oxygen, etc.), and 20% that started scanning did not complete all the sequences.150 In view of these constraints, patients with mild stroke (lacunar or small cortical) are more able to cooperate with scanning and a much larger proportion can be scanned (>95%).153 Given all these very real practical issues with MR, and until it is clear that the information gained from MR is so much better than from CT that it is worth the extra time, cost, risk to the patient from hypoxia, etc. and organizational difficulties, the first-line investigation for most stroke patients will probably continue to be CT in the immediate future,154,155 reserving MR for ‘difficult cases’ with specific questions (see below), and for research.156 CT has enjoyed somewhat of a renaissance in the last few years with the advent of widely available multislice perfusion imaging140 and rapid CT angiography with very good image reconstruction.145 However, use of MR scanners may expand as their design evolves, the sequences become faster148,157 and access to patients within the scanner becomes easier (e.g. wider bore), and

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 211

5.5 Magnetic resonance imaging

therefore more user-friendly. The advantages of MR are not only its superior pathoanatomical cross-sectional imaging properties, but that the same machine can image vessels and assess cerebral blood flow non-invasively; provide diffusion imaging and spectroscopy to elucidate the pathogenesis of brain damage in ischaemia and how experimental treatments might modify it; and, by functional imaging, elucidate mechanisms of brain recovery. Many of the general principles of recognizing an infarct by its shape and site, the correlations between clinical findings and infarct site, the time course of infarct swelling, fogging and haemorrhagic transformation as have been described for CT are equally relevant to MR and will not be repeated here. Rather, this section highlights where MR imaging provides different information to CT. Routine MR brain imaging for suspected stroke A routine MR brain scan for stroke should include a midline sagittal localizing view of the brain, usually T1-weighted, followed by axial diffusion-weighted images (DWI), a gradient echo (or T2*) sequence, axial T2-weighted images and/or a fluid attenuated inversion recovery (FLAIR) sequence – all covering the whole brain (Fig. 5.28). MR dynamic susceptibility perfusion imaging following an intravenous injection of a bolus of contrast, MR angiography and MR spectroscopy are also commonly available on modern MR scanners. In simple terms, it is useful to think of: • DWI showing as bright white (hyperintense) areas where water movement is restricted (i.e. intracellular oedema); • FLAIR and T2-weighted images as showing brain water content (so areas of oedema will show up as hyperintense, i.e. white, with CSF also white on T2 and dark on FLAIR; • gradient echo as being particularly sensitive to anything with paramagnetic effects, such as haemoglobin breakdown products, therefore being very good for showing new and old haemorrhage; • T1-weighted images as showing brain structure. Most scanners now use a ‘fast spin echo’ sequence for T2 which takes less time to acquire than the older ‘spin echo’ sequences. However, it is important to recognize that spin echo and fast spin echo do not produce the same images, the main difference being that fast spin echo is less sensitive to haemosiderin, a haemoglobin breakdown product.158 A routine MR examination for stroke should include DWI, gradient echo and either T2 or FLAIR sequences.

..

..

Contraindications to MR All patients should be screened for MR compatibility before they enter the magnetic field, but the key contraindications are mentioned here to help avoid inappropriate referrals: pacemakers, intracranial aneurysm clips (virtually all types), definite metallic intraocular foreign bodies, and cochlear implants (see www.mrisafety.com for details on specific manufacturers’ devices and MRI compatibility). Intracranial aneurysm coils are not contraindications. Metallic prostheses, foreign bodies other than in the eyes or brain, some ventriculoperitoneal shunts, some artificial heart valves and first trimester of pregnancy are all relative contraindications which are useful for the MR department to know about well in advance, so that the individual patient’s circumstances can be checked (to see whether MR scanning will be possible without undue risk) and to tailor the scan to the patient. Patients who are dysphasic may need to have their orbits and chest X-rayed prior to MR to exclude any intraocular metallic foreign body, and pacemakers (feeling the chest wall is probably not a sufficiently reliable substitute for chest X-ray for excluding a pacemaker).

5.5.1 The appearance of haemorrhage on MRI The appearance of haemorrhage on MR is governed by the paramagnetic properties of haemoglobin breakdown products, and so it changes with time (Fig. 5.29).46 Freshly extravasated red blood cells contain oxyhaemoglobin which does not have any paramagnetic properties. As soon as the blood becomes deoxygenated, the signal properties change, and this can be detected on sequences that are particularly sensitive to the oxygen state of haemoglobin, such as the echoplanar sequences used for diffusion-weighted imaging, or T2*-weighted gradient echo sequences. Thus, depending on the state of the oxygen in the intracerebral haemorrhage (ICH), there may be little immediate signal change and, although a lesion may be visible, the differentiation from infarction or some other mass lesion may be difficult within the first few hours, even when specific blood-sensitive sequences are used (Fig. 5.30). In some patients with ICHs scanned within the first few hours, there may be serpiginous low signal (dark) bands at the margins of the lesion on gradient echo (T2*) sequences and sometimes on DWI. However we have seen cases where low signal bands were minimal, only visible on the T2*-weighted gradient echo sequence, and which could easily have been overlooked if only the DWI had been done (Fig. 5.30). Thus, while several publications are optimistic about the ability of MR to detect parenchymal haemorrhage reliably within the first few hours of acute stroke,159–162 we would

211

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 212

212

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

(a) (b)

(d)

(c)

(e)

Fig. 5.28 Appearance of normal brain to compare various MR sequences. (a) T1-weighted midline sagittal – note the black CSF and little difference between grey and white matter. (b) T2-weighed fast spin echo axial – note CSF is very white (hyperintense), and grey matter is whiter (more intense) than white matter. (c) Fluid attenuated inversion recovery (FLAIR) axial – CSF is dark, grey and white matter are of similar intensity. (d) T2-weighted gradient echo axial – CSF is white, grey and white matter can be easily distinguished although the anatomical definition is not great – very sensitive to

haemosiderin and calcium if present (see also Figs 5.30–5.33). (e) Diffusion-weighted axial imaging – poor anatomical detail, dark CSF – acute ischaemic lesions are bright white. Note that the CSF appears white on T2 and gradient echo imaging and dark on the other sequences. T2 and T1 provide the best anatomical detail, FLAIR is useful for seeing abnormal tissue near CSF interfaces (see Fig. 5.38), gradient echo is very sensitive to blood, metal and calcium, and diffusion imaging is very sensitive to acute ischaemia.

continue to urge caution – acute haemorrhage on MR in general is just not as easy to diagnose as CT acute haemorrhage, especially for the inexperienced or unwary. However, increasing use of MR in hyperacute stroke has raised confidence in distinguishing lesions which are primarily haemorrhagic from those which are ischaemic. The time course of the change in appearance of an ICH on different sequences as it ages is now known to be much more variable than was originally thought.163 In general, the pattern is as follows, but it will vary with the

sequences used, and probably factors related to the patient and the ICH: • Once enough deoxyhaemoglobin has formed the ICH shows a centrally hypointense (dark) area on T1-weighted imaging and a markedly hypointense (dark) area on T2-weighted imaging. • As methaemoglobin is formed in the red cells, the lesion on the T1-weighted image becomes hyperintense (bright) while on the T2-weighted image it initially remains dark.

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 213

5.5 Magnetic resonance imaging

T2

Oxy-Hb

Oedema Deoxy-Hb Deoxy-Hb Methaemoglobin

Oedema

Haemosiderin

Haemosiderin

Hemichromes + serous fluid

Methaemoglobin

T1

= Normal brain signal intensity

1

2

3

4

5

6

7

8

Fig. 5.29 Diagram of the change in the appearance of parenchymal brain haemorrhage on MRI. The top row shows the typical appearance on T2-, and the lower row on T1-weighted imaging. The normal brain is represented by the stippled background. To a certain extent, the exact appearance and timing of the changes depend on the field strength of the

• As the methaemoglobin becomes extracellular and the ICH liquefies, the lesion on the T2-weighted image becomes as hyperintense (bright) as on the T1-weighted image. • After several weeks, lesions on T2-weighted images become bright in the centre with a very dark rim, and on T1-weighted images they are also bright centrally and moderately dark around the rim. • Eventually the dark rim (haemosiderin) around a bright ‘hole’ on T2 is the only remaining feature. The exact timing and degree of these signal changes depend on the strength of the magnet, which compartment of the brain the haemorrhage lies in, abnormal clotting, haematocrit, and the exact scan sequence used.46 Some sequences are very sensitive to the presence of haemoglobin breakdown products, e.g. T2* (gradient echo T2)164 and ‘susceptibility-weighted imaging’165 which show microhaemorrhages and petechial haemorrhage in infarcts, previously visible only to the pathologist (Fig. 5.31),166 while other sequences are much less blood-sensitive (see below). The clinical utility of being able to see tiny amounts of petechial haemorrhage or chronic microhaemorrhages remains uncertain (see below), but it is certainly important to make sure that the routine MR imaging protocol for stroke (acute or chronic) includes sequences which can reliably detect blood of any age.

..

..

9

10

15

20

Time (days)

magnet. The most important things to remember are that intracellular deoxyhaemoglobin (deoxy-Hb) is dark on T1 and T2, methaemoglobin is bright on T1 and T2, and haemosiderin is dark on T1 and T2. Haemosiderin persists in the margins of a haematoma for years after the original haemorrhage. Oxy-Hb, oxyhaemoglobin.

Does haemorrhage remain visible indefinitely on MR? The MR features of haemosiderin may persist for life (Fig. 5.32) because, in general, old ICHs are visible pathologically as haemosiderin deposits at postmortem.167 Haemosiderin is one of two major haemoglobin-derived pigments identifiable in tissue sections from around ICHs and represents aggregates of ferritin micelles within lysosomes of phagocytic cells. The other pigment is haematoidin, which is chemically identical to bilirubin, and forms locally as a result of haemoglobin breakdown in a milieu of reduced oxygen tension.167 Haemosiderin is ferromagnetic and therefore visible on MR, whereas haematoidin has no particular magnetic properties and is not visible on MR. We found that among 116 survivors of moderate to severe head injury of whom 78 had one or more parenchymal haemorrhages (total 106) at the time of injury, using a 1.0 Tesla MR scanner between 1 and 5 years after injury, 96 of the 106 (90%) haemorrhages were visible as haemosiderin on spin echo T2weighted imaging. Of the ten haemorrhages without haemosiderin on MR (11% of patients), seven were in patients where another haemorrhage with haemosiderin was still visible elsewhere in the brain. There were no features of the haemorrhage (site, size, density, sharpness of edges) in the acute stage which correlated

213

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 214

214

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

(a)

(c)

(a)

(b)

(d)

Fig. 5.30 MR brain scan at 4 h after sudden right hemiparesis: (a) diffusionweighted; (b) T2-weighted; (c,d) gradient echo axial images. There is a large mass of increased signal on all sequences which could easily be mistaken for a tumour on T2 and an infarct on diffusion imaging. The clue that it was a haemorrhage is the hypointense (dark) wavy lines around the edges of the mass which represent blood products (thick arrows). Note that this patient has had an ICH previously (d) in the left frontal lobe (arrowhead) and has microhaemorrhages (c,d) (thin arrows). The CT scan from this patient is shown in Fig. 5.6.

(b)

Fig. 5.31 MR scan of a 5-day-old left striatocapsular infarct to show the extreme sensitivity of appropriate MR sequences to tiny areas of petechial haemorrhage. (a) T2-weighted spin-echo sequence shows the infarct (arrows), but no haemorrhage. (b) T2-weighted gradient-echo ‘haem’ sequence shows tiny low-signal areas within the infarct caused by tiny amounts of petechial haemorrhage (arrows) which would previously have only been visible to the pathologist. The clinical relevance of this is uncertain.

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 215

5.5 Magnetic resonance imaging

Fig. 5.32 MR scan showing an old ICH in the right basal ganglia. (a) T2 spin echo: the lesion has an obvious low signal (dark) area surrounding the high signal centre (arrow). (b) T2 fast spin echo: the lesion is more difficult to identify as an old haemorrhage as the low signal (dark) area is less obvious. (c) FLAIR: several small deep lesions shown (arrows), but no features to identify any of them as haemorrhage. (d) Gradient echo T2: the haemorrhage is obvious. Note the enlarged perivascular spaces (pinpoint high signal areas) best seen on T2 (b).

(a)

(b)

(c)

(d)

with the likelihood of haemosiderin being present at 1 year. Some quite large ICHs showed no trace of haemosiderin.168 Gradient echo (T2*-weighted) sequences are highly sensitive to deoxyhaemoglobin and haemosiderin (see above) but are still not routinely performed (in acute or chronic stroke) unless specifically requested, as they add an extra 2–5 min to the scanning time.158,163,164,166 Furthermore, areas of calcification give a similar appearance to haemosiderin on gradient echo sequences, and so may cause confusion.164 Fast spin echo T2-weighted and FLAIR sequences, particularly the latter, are too insensitive to deoxyhaemoglobin or haemosiderin to be reliable for identifying acute or old ICHs, or microhaemorrhages.158

..

..

In general, the characteristic signal changes of parenchymal haemorrhage on MRI persist indefinitely, but only if the appropriate imaging sequences are used; therefore, intracerebral haemorrhages can be identified even years after they have occurred. However, not all haemorrhages form haemosiderin during their resolution, therefore not all will be discernible as such on MR performed late after the event, even if the correct sequences are used. Microhaemorrhages More widespread use of MR in patients with stroke, dementia and normal older subjects has revealed that some patients, as well as occasional normal older people,

215

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 216

216

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

(a)

(b)

have solitary or multiple small ( 1.2) and in the PWI lesion used (although most tested a mean transit time derivative). The available data showed that mismatch (vs no mismatch) without thrombolysis was

associated with a non-significant twofold increase in the odds of infarct expansion (OR 2.2), which did not change with thrombolysis (OR 2.0). Half the patients without mismatch also had infarct growth, and thrombolysis did not change the odds ratio (Fig. 5.51).310 These data suggest that mismatch may be associated with an increased risk of infarct growth but are very unreliable, particularly on the question of whether thrombolysis changes infarct growth. A trial testing the mismatch hypothesis is ongoing (EPITHET).327 About half of our own 46-patient cohort had mismatch on admission, and neither DWI/CBF nor DWI/MTT mismatch (both PWI parameters were semi-quantitative) predicted lesion growth; lesion growth was equally common in those with and without mismatch.311 Thus although ‘DWI/PWI mismatch’ is being used to identify patients for inclusion in trials of stroke treatments,326 our results imply that patients without ‘mismatch’ may also benefit and should not be denied that possibility by being excluded from acute stroke treatment trials. It is conceivable that the mismatch approach could be used to extend the time window for treatment. There is increasing evidence of viable but at risk tissue up to 24 h after stroke, which might be amenable to thrombolytic or other treatment.328 For example, thrombolysis is currently licensed for use within 3 h of stroke, but

233

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 234

234

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

Fig. 5.51 An example of MR diffusion-perfusion mismatch. Top row: 4 h after onset of right hemiparesis shows T2, diffusion, ADC, relative CBF and mean transit time (relative MTT) maps. There is early ischaemic change on diffusion and ADC but not on T2; a perfusion deficit is visible on CBF and MTT, largest on MTT, but larger than the diffusion lesion on both CBF and MTT. The ‘DWI/CBF mismatch’ would be smaller

than the ‘DWI/MTT mismatch’. Bottom row: 7 days after stroke, the infarct is now visible on T2, diffusion and ADC and there is no longer any lesion on CBF or MTT which indicates reperfusion. The diffusion lesion does not seem to have grown particularly. This patient did not receive thrombolysis, so these perfusion changes were spontaneous. Also reproduced in colour in the plate section.

some clinicians use mismatch to decide whether to treat patients with thrombolysis between 3 and 6 h of stroke, and trials using DWI/PWI mismatch to select patients up to 9 h after stroke has recently finished (DIAS,326 DEDAS and DIAS 2).

circumstances such as suspected arterial dissection or venous disease (sections 5.8 and 7.21.3).

MR angiography Magnetic resonance angiography allows the acquisition of images of blood vessels, without injection of contrast, by using the signal characteristics of flowing blood.329–331 This technique can be used in acute ischaemic stroke, but the patient must keep very still for several minutes so it is not always of practical help (Fig. 5.52). It is used for the assessment of carotid stenosis in patients being considered for carotid endarterectomy (section 6.8.5)332 and is also being evaluated for the detection of intracranial aneurysms (section 9.4.3). We do not use it routinely in acute stroke as there is no good evidence that routinely identifying the presence of arterial occlusion influences management in the majority of patients. We use it in specific

5.5.3 Advanced magnetic resonance techniques Magnetic resonance (MR) permits other useful methods of examining the brain, including MR spectroscopy to show the effect of ischaemia on brain metabolites, and functional MR to identify areas of the brain which control particular body and mind functions. The details of these techniques are beyond the scope of this chapter, but there are excellent reviews describing each technique, as detailed below. Magnetic resonance spectroscopy Magnetic resonance spectroscopy (MRS) can demonstrate metabolic changes in ischaemic tissue in vivo, particularly hydrogen, phosphate, carbon, fluorine and sodium metabolism.329,333–335 N-acetyl aspartate (considered a marker of ‘normal neurones’), creatine and phosphocreatine, choline-containing compounds, lactate

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 235

5.5 Magnetic resonance imaging

(a)

Fig. 5.52 MR imaging with (a) T2, (b,c) proton density and (d) MR angiography of a left middle cerebral artery (MCA) branch occlusion. Note the high signal of recent thrombus is visible on proton density (c, arrow) and the infarct in the insular cortex (a,b, arrows). (a) On T2 only dark signal mimicking a flow void is visible in the MCA branch so the thrombus might be overlooked. (d) The MR angiogram confirms the MCA branch occlusion (arrow). See also Fig. 5.36c.

(d)

(c)

and pH have all been measured in stroke patients, but it is still too early to be certain of the significance of the changes.334,336–338 MRS can be performed by either the single voxel, in which a small cubic volume of brain – typically 8 cm3 – is sampled, or by chemical shift imaging techniques in which spectra from a whole slice of brain are obtained simultaneously (Fig. 5.53). Both scan techniques require the patient to keep still for up to 10 min and are not in routine use. Functional MR imaging (fMRI) When a part of the cerebral cortex becomes metabolically active its blood flow increases, thereby changing the MR signal obtained from this part of the brain. For example, if the subject is asked to undertake a task, such as wiggling a finger, while in the MR scanner, and the

..

..

(b)

MR prior to the movement is compared with the image during the task, the difference between the two demonstrates the area of cortex activated by the function. It is thought that areas of the brain which may perhaps be ‘taking over’ – or at least influencing in some way – the function of damaged areas, can be identified by their activation on fMRI, and the pattern of brain compensation to injury studied during recovery from ischaemic stroke or head injury, or in patients with brain tumours. However, the technique is difficult and great care is required to obtain meaningful results. The subject must keep his or her head very still and only perform (and think about) the requested task, because inadvertent movement of other parts of the body will appear on MR and be misleading. Nonetheless, the technique shows promise by allowing insights into normal brain function as well as into recovery from injury.339

235

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 236

236

Chapter 5 What pathological type of stroke is it, cerebral ischaemia or haemorrhage?

(a)

(b)

5.5.4 Comparative studies of CT and MRI in acute stroke MR is more sensitive for detecting minor degrees of haemorrhagic transformation of infarcts during followup, but CT is better than MR (even with gradient echo imaging) for detecting acute intracerebral haemorrhage. Although MR may identify low signal areas indicating haemoglobin breakdown products, these can be quite subtle in the acute stage and CT is simply much easier to interpret.340 Most studies of intracerebral haemorrhage on MR in hyperacute stroke report apparently high sensitivity for haemorrhage,159 but in our experience acute haemorrhage can have few distinguishing features from an ischaemic lesion or other pathology and in some cases we have ended up doing a CT scan to make the diagnosis. Most studies comparing the visibility of ischaemic lesions on brain CT and MRI in acute stroke performed MR after CT, making it likely that MR would be more often abnormal than CT. Some did not take account of the case mix of mild to severe strokes. Only one study apparently randomized the order of scanning,341 but others found this to be impractical.151 Several early studies suggested that in patients imaged within the first 6 h of stroke, DWI had a higher sensitivity and specificity and inter-rater reliability than CT.342–345 All found excellent agreement between CT and DWI. DWI was performed on average 2–3 h after CT in all studies. None mentioned the proportion of patients who could not be scanned with either technique or in whom the images were not diagnostic. A more recent and complete comparison of CT and MR in consecutive patients with moderate and severe

Fig. 5.53 MR spectroscopic chemical shift imaging at 1 day after stroke in a patient with a right middle cerebral artery infarct. The T2-weighted image is in the background and the spectroscopic colour maps of the distribution of (a) the normal neuronal metabolite N-acetyl aspartate (NAA) and (b) the marker of anaerobic metabolism lactate, are superimposed. Red indicates large amounts and blues and greens small amounts of the metabolites. You will need to see the colour plate section to appreciate the following. The area of increased T2 signal of the infarct corresponds with the area of blue on the NAA map (in keeping with neuronal loss) and red on the lactate map (indicating anaerobic metabolism taking place in the infarct). Also reproduced in colour in the plate section.

strokes admitted hyperacutely and considered for thrombolysis found that MR could not be obtained in about one-third (patient too unwell, uncooperative, scanner unavailable despite ‘24-hour’ availability) and when both modalities were interpreted using ASPECTS (a structured scoring system for scan interpretation) there was no difference between CT and DWI in the number of infarcts detected or in the inter-rater reliability.151 Thus for patients admitted and scanned within 6 h of symptom onset, who are likely to have moderate to severe strokes, CT with a structured scoring system can detect as many early infarcts as DWI, detects haemorrhage better than MR, and can image all patients whereas the practical limitations of MR prevent imaging in up to one-third of patients. The situation is different for patients with mild strokes, including lacunar syndromes, and who present later, as discussed above. In this case, MR with DWI and gradient echo imaging is much more likely to show the acute ischaemic lesion (or old haemorrhage) than CT and this patient group is easier to scan with a much higher success rate. In patients with moderate to severe strokes, CT with a structured scoring system can detect as many early infarcts within 6 h as DWI, detects haemorrhage better than MR and can image all patients whereas the practical limitations of MR prevent imaging in up to one-third of patients. On the other hand, MRI with DWI is superior for lacunar, posterior fossa and small cortical and subcortical infarcts. Whether MRI is ‘better’ than CT depends on the question being asked of imaging. MRI is certainly less practical and takes longer.

..

..

9781405127660_4_005.qxd 10/13/07 11:05 AM Page 237

5.6 Other ‘sophisticated’ methods of imaging cerebral ischaemia

5.6 Other ‘sophisticated’ methods of imaging cerebral ischaemia

areas in terms of symptoms and signs not fully explained by loss of the area of brain directly involved in the infarct.358,359 Positron emission tomography

Single-photon emission CT Single-photon emission CT (SPECT) is a technique in which cross-sectional images of the brain are obtained by using either a rotating gamma-camera, or a dedicated SPECT scanner, and computerized reconstruction of the data from the emitted radiation following intravenous administration of a radioactive isotope. Various isotopes are available, such as hexamethyl propylene amine oxime (HMPAO) (Ceretec) which images regional perfusion and therefore can identify areas of abnormal perfusion early, or xenon-133, which can measure regional cerebral blood flow quantitatively.346 However, the equipment for SPECT is not widely available, the scanning times are long (20 min or more) and so problems caused by movement artifacts can arise from restless stroke patients, and the isotopes are expensive. A potential source of error, which is particularly relevant to ischaemic stroke, is that blood–brain barrier breakdown alters the behaviour of the isotope and hence the accuracy and interpretation of the images.347 SPECT can demonstrate appropriate areas of low flow within a few hours of stroke onset, and even up to 60% of patients with a transient ischaemic attack have an appropriate abnormality within 24 h of onset.348 In acute ischaemic stroke, SPECT demonstrates an area of relative hypoperfusion corresponding with the ischaemic tissue in most patients studied, although very few if any lacunar stroke patients were included in the studies.348–353 Several studies have indicated that the size of the initial perfusion deficit on SPECT predicts the likely clinical outcome: the larger the deficit, the worse the outcome.84,350–353 Also, evidence of reperfusion on SPECT is associated with improved clinical outcome both spontaneously352,353 and in response to thrombolytic therapy.84,354,355 In the first week after the stroke, the relative blood flow in the infarct may be increased or decreased, corresponding with luxury perfusion or persistent hypoperfusion, respectively.356 Abnormalities may be visible not only in the infarct itself, but also in remote parts of the brain, such as the opposite cerebral hemisphere or the contralateral cerebellum in patients with cerebral cortical infarcts.357,358 This has given rise to the concept of ‘diaschisis’ or ‘shutdown’ of areas of the brain not directly involved in the infarct but whose function is in some way influenced, possibly through association fibres.359 There is some evidence of clinical correlates of ‘shutdown’ of these

..

..

Positron emission tomography (PET) can display a variety of physiological processes, from glucose metabolism to neuroreceptor density, thus allowing study of the normal workings of the brain, as well as the consequences of pathology.360–362 The data, derived from collection and analysis of complicated signals from positronemitting isotopes, may be shown as colour-coded tomographic pictures. PET has been available since the 1970s, but the imaging equipment and isotopes are extremely expensive, the scan times are prolonged and the patient must keep very still. Therefore PET is likely to remain a research tool with limited clinical applications, at least in stroke for the foreseeable future.363,364 PET studies in ischaemic stroke have provided evidence that the ischaemic penumbra really does exist in humans and that neural tissue in the penumbra may survive for up to 48 h.328,361,365–368 They have also suggested that neurones in areas of the brain with cerebral blood flow below the threshold traditionally regarded as producing irreversible damage ( 160/90 mmHg, at least twice pre-stroke) Angina and/or past myocardial infarction Current smoker Claudication and/or absent foot pulses Major cardiac embolic source Transient ischaemic attack Cervical arterial bruit Diabetes mellitus Any of the above

126

52

(a)

Endothelium Fatty streak

92

38

66 60

27 25

50 35 33 24 196

20 14 14 10 80

Lipid core (c)

extensive or severe atheroma when exposed to causal risk factors, such as hypertension, while the arterial anatomy determines where the atheroma occurs. There appear to be important racial differences in the distribution of atheroma, and race is an independent predictor of lesion location.42 White males tend to develop atheroma in the extracranial cerebral vessels, the aorta and coronary arteries while intracranial large vessel disease appears to be relatively more common in black, Hispanic and Asian populations43–45 and tends to affect younger patients43,45 and those with type 1 diabetes mellitus.43 Some sources report that women have more intracranial disease compared to men but this is disputed by others.44 However, carotid bifurcation stenosis does appear to be becoming more frequent in Oriental populations either because the pattern of atheroma is evolving as a result of lifestyle changes, or because previous studies were confounded by selection and other biases.44,46,47 There are very few data on the aetiology of stroke in Africa, but reports from stroke registers in South Africa show that atheroma is a less common cause of stroke in blacks than in whites.48 Individuals with atheroma affecting one artery almost always have atheroma affecting many other arteries, with or without clinical manifestations.

6.3.2 The nature, progression and clinical consequences of atheroma Atheroma begins as intimal fatty streaks in children, it is thought in response to endothelial injury18,19 (Fig. 6.3).

Fibrous cap

(b)

Platelet–fibrin thrombus

Haemorrhage

Necrosis

Calcification Inflammatory cell infiltrate

(d)

Embolus

Platelet–fibrin thrombus

Fibrosis Ulceration

Calcification

Fig. 6.3 The growth, progression and complications of atheromatous plaques: (a) early deposition of lipid in the artery wall as a fatty streak; (b) further build-up of fibrous and lipid material; (c) necrosis, inflammatory cell infiltrate, calcification and new vessel formation, leading to (d) plaque instability, ulceration and platelet–fibrin thrombus formation on the plaque surface.

Over many years, circulating monocyte-derived macrophages adhere to and invade the arterial wall. As a result, there is an inflammatory response with cytokine production and T-lymphocyte activation. Intra- and later extracellular cholesterol and other lipids are deposited, particularly in macrophages, which are then described as foam cells. Arterial smooth-muscle cells migrate into the lesion and proliferate, fibrosis occurs and so fibrolipid plaques are formed. These plaques, with their lipid core

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 265

6.3 Atheroma and large vessel disease

and fibrous cap, encroach upon the media and spread around and along the arterial wall. Some become necrotic, ulcerated and calcified with neovascularization and haemorrhage – so-called complicated plaques. The arterial wall thickens, the vessel dilates or the lumen narrows and the artery becomes stiffer and tortuous. Atherosclerotic lesions may cause ischaemic stroke through acute vessel occlusion, or by reducing vessel diameter and thus regional cerebral blood flow. Whether acute occlusion of a cerebral vessel, or reduction in cerebral blood flow, leads to infarction depends on not only for how long the blood flow is impaired, but also the availability and functional capability of the collateral circulation49–53 (section 4.2). Arterial occlusion secondary to atheroma occurs by three mechanisms. First, thrombi may form on lesions and cause local occlusion. Second, embolization of plaque debris or thrombus may block a more distal vessel; emboli are usually the cause of obstruction of the anterior circulation intracranial vessels29,30 at least in white males in whom intracranial atheroma is relatively rare. Third, small vessel origins may be occluded by growth of plaque in the parent vessel. This is seen particularly in the basilar artery and the subclavian artery around the vertebral origin. Low regional cerebral blood flow secondary to atheroma occurs when plaque growth causes severe reduction in the diameter of the vessel lumen and hypoperfusion of distal brain regions, particularly in the border zones where blood supply is poorest. This may lead to ‘borderzone infarction’ (section 6.7.5) of these regions following severe hypotension or hypoxia. The mechanisms of atherosclerosis-related stroke are discussed further below.

6.3.3 Atheromatous plaques complicated by thrombosis: atherothrombosis From an early stage, or perhaps even from the very first stage, atheromatous plaques promote platelet adhesion, activation and aggregation, which initiates blood coagulation and thus mural thrombosis.54–56 At first, any thrombus may be lysed by fibrinolytic mechanisms in the vessel wall, or incorporated into the plaque, which re-endothelializes and so ‘heals’. Gradually, the atheroand then atherothrombotic plaque grows, in part because of repeated episodes of mural thrombosis layering one on top of the other, and eventually the lumen may become obstructed. Such occlusive intraluminal thrombus may then propagate proximally or distally in the column of stagnant blood, but usually no further than the next arterial branching point. Thrombosis is opposed by the release of prostacyclin and nitric oxide, both vasodilators, from the vascular

..

..

endothelium and by endothelium-derived plasminogen activator. The balance of these pro- and antithrombotic factors may determine whether a thrombus complicating an atheromatous plaque or an occlusive embolus grows, is lysed or becomes incorporated into the arterial wall and so contributes to the gradually enlarging atherothrombotic plaque. Symptomatic in situ acute atherothrombotic occlusion – rather than artery-to-artery embolism – does not appear to be a very common cause for ischaemic stroke or transient ischaemic attacks (TIAs) in the carotid territory. Perhaps this is because atheroma affects the larger arteries (e.g. internal carotid artery (ICA) rather than middle cerebral artery) and it takes a very large plaque to occlude them, or because the potential for collateral blood flow is better distal to larger arteries.29,57 Indeed, once the ICA has occluded, the risk of ipsilateral ischaemic stroke appears to be less than for severe stenosis.58–60 On the other hand, symptomatic in situ atherothrombotic occlusion may be more common in the posterior circulation (e.g. of the basilar artery) but even here artery-to-artery embolism is well described.61–65

6.3.4 Embolism from atherothrombotic plaques: atherothromboembolism Atheroma and/or thrombus may embolize – in whole or in part – to obstruct a smaller distal artery, usually at a branching point – the same one or different ones on several occasions. Emboli consist of any combination of cholesterol crystals and other debris from the plaque, platelet aggregates, and fibrin which may be recently formed and relatively friable or old and well organized. Depending on their size, composition, consistency and age – and presumably the blood flow conditions at the site of impaction – emboli may be lysed, fragment and then be swept on into the microcirculation. Alternatively, they may permanently occlude the distal artery and promote local antero- and retrograde thrombosis, which is further encouraged by the release of thromboxane A2 from platelets, which is also a vasoconstrictor. Emboli are transmitted to the brain or eye via their normal arterial supply, which itself varies somewhat in distribution between individuals (section 4.2). An embolus from an atherosclerotic carotid bifurcation – usually a plaque at the origin of the internal carotid artery (ICA) but sometimes the distal common or proximal external carotid arteries – normally goes to the eye or the anterior two-thirds of the cerebral hemisphere. But, on occasion, it may go to the occipital cortex if blood is flowing from the ICA via the posterior communicating artery to the posterior cerebral artery, or if there is a fetal

265

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 266

266

Chapter 6 What caused this transient or persisting ischaemic event?

origin (section 4.2.3) of the posterior cerebral artery from the distal ICA (5–10% of individuals). However, if an artery is already occluded, then an embolus may travel via the collateral circulation and impact in an unexpected place. For example, with severe vertebral arterial disease, and therefore poor flow distally into the basilar artery, an embolus from the ICA origin may reach the basilar artery via the circle of Willis. With ICA occlusion, it is still possible to have an ipsilateral middle cerebral artery (MCA) distribution cerebral infarct as a result of: • an embolus travelling from the contralateral ICA origin via the anterior communicating artery; • an embolus from any blind stump of the occluded ICA, or from disease of the ipsilateral external carotid artery (ECA), via the ECA and orbital collaterals to the MCA; • an embolus from the tail of thrombus in the ICA distal to the occlusion; or • low flow distal to the ICA occlusion, perhaps within a boundary zone (section 6.7.5), particularly if the collateral blood supply is poor, cerebrovascular reactivity is impaired and the oxygen extraction ratio high.652,66–69 Curiously, emboli from the neck arteries (or from the heart) seldom seem to enter the small perforating arteries of the brain to cause lacunar infarction (section 6.4) perhaps as a consequence of the fact that the perforating vessels arise at a 90° angle from the parent vessel.

(a)

(b)

6.3.5 Atherothrombosis/embolism as an acute-on-chronic disorder: plaque instability Like the coronary arteries, atheromatous plaques in the cerebral circulation – particularly at the carotid bifurcation – become ‘active’ or ‘unstable’ from time to time with fissuring, cracking or rupture of the fibrous cap, or ulceration. The histological features of plaque instability are a thin fibrous cap, large lipid core, reduced smooth muscle content and high macrophage density31,70–72 (Fig. 6.4a). If the thrombogenic centre of the plaque is exposed to flowing blood, then complicating thrombosis occurs. Plaque instability may even be a ‘systemic’ tendency because irregularity on catheter angiography – and so presumed instability and ulceration – of symptomatic carotid stenosis is associated with irregularity of the asymptomatic contralateral carotid artery, and with coronary events assumed to be caused by plaque rupture.73 At other times the plaque is quiescent with a thick fibrous cap, or slowly growing, without causing any clinical symptoms30,31,74–77 (Fig. 6.4b). In other words, atherothromboembolism is an ‘acute-on-chronic disorder’. It is no surprise therefore that the clinical complications of atheroma reflect this: • transient ischaemic attacks (TIA) tend to cluster in time78 (section 3.2.1); • stroke tends to occur early after a TIA and affect the same arterial territory79 (section 16.2.1);

Fig. 6.4 Photomicrographs of transverse sections of the carotid sinus illustrating the pathological features of atherosclerotic plaque stability. (a) An unstable plaque characterized by a large necrotic core (✽), and a thin fibrous cap heavily infiltrated by macrophages (arrow). Elsewhere the plaque is ulcerated and the lumen occluded by thrombus, part of which had embolized to the ipsilateral middle cerebral artery causing a fatal ischaemic stroke. (b) A comparable stenotic but stable plaque, comprising largely fibrous tissue (F) with focal calcification (arrow). There is no significant necrotic core or inflammatory cell infiltrate. L lumen; A tunica adventitia; M tunica media. (Provided by Dr Alistair Lammie.)

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 267

6.4 Intracranial small vessel disease

• the risk of ischaemic stroke ipsilateral to severe carotid stenosis is highest soon after symptomatic presentation and then declines,73,80 even though the stenosis itself seldom regresses (Fig. 16.38); • presumed artery-to-artery embolic strokes tend to recur particularly early81 (section 16.2.3); • emboli are more often detected with transcranial Doppler sonography if carotid stenosis is severe or recently symptomatic;82–87 and • the rate of Doppler-detected emboli in the middle cerebral artery tends to decline with time after stroke.88 Increasing severity of symptomatic atherothrombotic stenosis, at least at the origin of the internal carotid artery, is undoubtedly a powerful predictor and cause of ischaemic stroke ipsilateral to the lesion.89 However, this cannot be the whole explanation because: • the risk of stroke distal to an asymptomatic stenosis is far less than distal to a recently symptomatic stenosis of the same severity (section 16.12.5); • by no means all patients with severe stenosis have a stroke (section 16.11.5); and • relatively mildly stenosed plaques can be complicated by acute carotid occlusion.31 Curiously, there is not such an obvious relationship between increasing coronary artery stenosis and coronary events. This may be because the coronary arteries are harder to image repeatedly and are more anatomically complicated, because coronary events are more often ‘silent’, or because the coronary arteries are smaller and more likely to be blocked if even a small plaque ruptures.90 Independently of the severity of stenosis, plaque irregularity on catheter angiography is associated with increased stroke risk, probably because irregularity represents plaque ulceration and instability with thrombosis and so likely complicating embolism71,85,91 (section 16.11.8). Atherothromboembolism is an acute-on-chronic disorder, both in its pathology and in its clinical manifestations. Although the formation of atherothrombotic plaques must be a long and gradual process over many years, the clinical manifestations usually occur acutely (e.g. an ischaemic stroke) and tend to cluster in time. For example, stroke tends to occur sooner rather than later after a transient ischaemic attack, perhaps as a result of the breakdown and ‘instability’ of an atherothrombotic plaque which later ‘heals’. Exactly why one plaque becomes unstable (and then perhaps ulcerates with complicating thrombosis) and another does not, is unknown. Histological comparisons

..

..

of recently symptomatic with asymptomatic carotid plaques – matched for stenosis severity – are not easy to perform. However, it seems as though intraplaque haemorrhage, calcification and the lipid core are similar in both but, crucially, the thickness of the fibrous cap has not generally been assessed.92–94 Similar problems arise with attempts to compare inflammatory processes and the expression of adhesion molecules,95,96 metalloproteinase expression,97 differences in plaque tissue factor98 and in plaque geometry and motion and so potential stresses on the fibrous cap.99,100 Whether infection causes plaque instability is very unclear (section 6.6.17). In all these cross-sectional studies there is always the possibility of reverse causality – that by becoming symptomatic a plaque is changed in its anatomy, motion, biochemistry, and so on.

6.3.6 Dolichoectasia This somewhat unusual pattern of arterial disease tends to affect the medium-sized arteries at the base of the brain, particularly the basilar artery, mostly in the elderly but occasionally in children101–105 (section 4.2.3). The arteries are widened, tortuous, elongated and are often enlarged enough to be seen as characteristic flow voids on MRI, and as tortuous channels – even without enhancement – or by virtue of calcification in their walls on brain CT106 (Fig. 6.5). When found in an individual, this arterial abnormality is not necessarily the cause of any ischaemic stroke (and very rarely of intracranial haemorrhage, despite the aneurysmal dilatation). However, these enlarged vessels can contain thrombus which embolizes, or occludes the origin of small branch arteries of the ectatic vessel, or even occludes the ectatic vessel itself. Cranial nerve and brainstem dysfunction due to direct compression or small vessel ischaemia, or hydrocephalus caused by cerebrospinal fluid pathway compression, are other occasional complications of basilar (or vertebral) ectasia.104,105,107–109 Atheroma is the most common cause.110 Other causes include various types of congenital defect in the vessel wall, Marfan syndrome, pseudoxanthoma elasticum and Fabry’s disease111 – both in homozygotic males and heterozygotic females.

6.4 Intracranial small vessel disease

There are a number of pathologies which affect the small (40–400 µm diameter) arteries, arterioles and veins

267

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 268

268

Chapter 6 What caused this transient or persisting ischaemic event?

(a)

(b)

(c)

Fig. 6.5 Vertibrobasilar dolichoectasia. (a) On the brain CT scan there is a 1.5–2 cm rounded mass (arrows) which is of slightly higher density than adjacent brain and sits in the left cerebellopontine angle. The mass is indenting the brainstem. On adjacent sections the mass was obviously longitudinal and contiguous with the vertebral and top end of the basilar arteries. (b) MR examination of the same patient shows the mixed signal mass (arrows). The presence of increased signal indicates either very slowly flowing or partially clotted blood. (c) Catheter angiography shows a lateral projection of the vertebrobasilar circulation. Although the main area of expansion of the arterial system is in the lower basilar artery (arrows), the dolichoectasia actually extends from the upper right vertebral artery almost to the very tip of the basilar artery.

of the meninges and brain – for example, cerebral amyloid angiopathy (section 8.2.2); vasculitis (section 7.3); atheroma near the origin of the small perforating arteries (see below); and the angiopathy underlying cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoenceophalopathy (CADASIL)

(section 7.20.1). Much more common, however, is what has been termed hyaline arteriosclerosis. This is an almost universal change in the small arteries and arterioles of the aged brain, particularly it is said in the presence of hypertension or diabetes, but sometimes in quite young patients without any of the classical

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 269

6.4 Intracranial small vessel disease

(a)

(b) Fig. 6.6 Photomicrograph of perforating lenticulostriate artery branches in the putamen, illustrating two distinctive patterns of vessel pathology. (a) Concentric hyaline wall thickening with a few remaining vascular smooth-muscle cell nuclei (arrow). The lumen remains patent. Such ‘simple’ small vessel disease is an almost invariable feature of elderly brains, most prominent in hypertensive and diabetic patients. (b) A complex, disorganized vessel segment showing an asymmetrical destructive process with focal fibrinoid material (✽) and mural foam cells (arrow). The lumen is visible cut in two planes of section. In this case the vascular lesion was adjacent to and, presumably, the cause of a right striatocapsular lacunar infarct. This ‘complex’ vessel lesion corresponds with what Miller Fisher termed ‘lipohyalinosis’. (Provided by Dr Alistair Lammie.)

vascular risk factors112 (Fig. 6.6). Important factors in its development appear to include not only hypertension and diabetes, but also breakdown of the blood–brain barrier with incorporation of plasma proteins into the vessel wall, possibly due to endothelial dysfunction.113 The smooth muscle cells in the wall are eventually replaced by collagen, which reduces vascular distensibility, and presumably reactivity, but not necessarily the size of the lumen.

..

..

Unfortunately, over the years, the pathological nomenclature for small vessel disease has been very confusing and, as a result, the ‘simple’ small vessel disease (SVD) described above has frequently been confused with a more aggressive-looking disorder of small arteries with disorganization of their walls and foam cell infiltration. It is this ‘complex’ SVD that Fisher first called ‘segmental arterial disorganization’ and then ‘lipohyalinosis’. The term fibrinoid vessel wall necrosis refers to a different, characteristic, but more acute change in the vessel wall seen, for example, as a consequence of accelerated hypertension and as a reactive phenomenon around acute intracerebral haematomas, spontaneous or traumatic. When healed, this probably takes on the appearance of ‘complex’ SVD114,115 (Fig. 6.6). However, it is by no means certain that ‘complex’ SVD and acute fibrinoid necrosis are more advanced forms of ‘simple’ SVD, and there is no evidence that ‘simple’ SVD is the ‘healed’ version of complex SVD. Nonetheless, both ‘simple’ and ‘complex’ SVD tend to affect the lenticulostriate perforating branches of the middle cerebral artery, the thalamoperforating branches of the proximal posterior cerebral artery, the perforating branches of the basilar artery to the brainstem, and the vessels in the periventricular white matter.116 Most of what is known about small vessel disease and lacunar stroke comes from a very small number of very careful clinicoanatomical observations by Fisher. He did not describe ‘simple’ SVD as the underlying vascular pathology of lacunes, but what we prefer to call ‘complex’ SVD in some cases, and atheroma affecting the origins of the small vessels where they come off the circle of Willis and major cerebral arteries in others.117–119 In theory, one can think of ‘mural atheroma’ affecting the parent arteries, ‘junctional atheroma’ affecting the origin of the small perforating arteries where they leave the parent artery, and ‘microatheroma’ affecting the proximal parts of these small arteries. In practice, these distinctions are more or less impossible to make at postmortem and the contribution of intracranial atheroma to lacunar infarction has still to be established. The current view is that both ‘complex’ SVD and atheroma at or near the origin of the small perforating vessels arising from the major cerebral arteries cause most, but not all, of the small deep infarcts responsible for lacunar ischaemic strokes (and, by implication, lacunar transient ischaemic attacks) which make up about one-quarter of symptomatic cerebral ischaemic events9,12,58,120–124 (sections 4.3.2 and 6.7.3). However, this hypothesis is not universally accepted125,126 since there is remarkably little direct postmortem evidence of occlusion of these vessels leading to lacunar infarcts,

269

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 270

Chapter 6 What caused this transient or persisting ischaemic event?

25 Lacunar (n = 222) Non-lacunar (n = 404)

20 Percentage of patients

270

15

10

5

0

0 –9

10–19 20–29 30–39 40 –49 50–59 60–69 70–79 80–89 90–99

Severity of stenosis (%) of symptomatic carotid artery

largely because the case fatality of lacunar stroke is so low and pathological material so scanty. The main supporting evidence for a specific small vessel lesion leading to lacunar infarction is indirect: • The relative lack of large vessel atheroma or embolic sources in the heart in the vast majority of lacunar patients, compared to those with cortical infarcts12,127–136 (Fig. 6.7). • Emboli are rarely if ever detected in the middle cerebral or common carotid artery by Doppler ultrasound in most studies of patients with lacunar infarction. However, it is not clear at what stage after the onset of stroke it is best to look – too early and an occluded middle cerebral artery might prevent the passage of emboli, too late and any proximal embolic source may have ‘healed’.88,137–140 • The low risk of early recurrence also argues against the concept of an active embolic source, either in the heart or an unstable atheromatous plaque.11,81,135,136 • The ‘capsular warning syndrome’ might suggest that a single perforating vessel is intermittently on the verge of occluding before it finally does so (section 6.7.3). • The impaired cerebrovascular reactivity in lacunar patients might suggest a specific problem, if not pathology, of the small intracerebral resistance vessels,141 such as endothelial dysfunction.113 It is conceivable that ‘complex’ SVD causes vessel rupture, leading to intracerebral haemorrhage as well as arterial occlusion, maybe at sites of so-called Charcot– Bouchard microaneurysms, although these may be artifacts of the pathological specimens. In fact, whether they are real or not is semantic, because what matters is not if the vessel wall bulges, but what weakens it in the first place122,142–146 (section 8.2.1). On the other hand, leukoaraiosis appears to be more associated with ‘simple’

100

Fig. 6.7 The relationship between the severity of symptomatic carotid stenosis and the likelihood of finding a nonlacunar (territorial) or lacunar infarct on the baseline brain CT in 626 patients in the European Carotid Surgery Trial. Those with severe stenosis are less likely to have lacunar infarcts. (With permission from Boiten et al. 1996.150)

SVD, and indeed this may be one underlying vascular cause, although it is clearly insufficient on its own, being so common in elderly brains. Interestingly, ‘cortical’ (presumed atherothrombotic) and ‘lacunar’ ischaemic stroke patients probably have a similar vascular risk factor profile (section 6.6), including hypertension.6,12,135,136,147–151 It is conceivable therefore that the same type of individual (i.e. hypertensive, diabetic, etc.) develops either small vessel disease (complex or atheroma) and so lacunar infarcts, or large vessel atherothromboembolism and so cortical infarction. The difference in the type of ‘degenerative’ vascular disease that occurs may perhaps reflect differing genetic susceptibilities. But, on the other hand, many individual patients have these different ischaemic stroke types at different times.144,152 On balance, we believe it likely that the lacunar hypothesis is correct and so, whatever the exact nature of the underlying small vessel lesion, lacunar infarction is seldom the result of embolization from proximal sites. About one-quarter of all ischaemic strokes and transient ischaemic attacks are ‘lacunar’ and many, if not most, lacunar infarcts are caused by disease of the small intracranial perforating arteries, either ‘complex’ small vessel disease, or atheroma affecting their proximal parts as they arise from their parent cerebral arteries. One question that has never been answered is whether there is a similar small vessel disease that affects the blood supply to the optic nerve and retina: are there ‘lacunar’ ocular syndromes equivalent to lacunar cerebral syndromes? What is clear, however, is that a high proportion of patients with ischaemic amaurosis fugax,

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 271

6.5 Embolism from the heart

retinal infarction and anterior ischaemic optic neuropathy do not have any detectable and likely proximal source of embolism (or evidence of low flow) in the arterial supply to the eye or the heart. Perhaps it is these patients who have small vessel disease like patients with ischaemic lacunar strokes.58,153,154

6.5 Embolism from the heart

That embolic material can pass from the heart to the brain, and from the venous system through the heart to the brain (paradoxical embolism), as well as to other organs, is undisputed. However, not all cardiac sources of embolism pose equal threats. For example, a mechanical prosthetic valve is much more likely to cause thromboembolism than mitral leaflet prolapse. In developed countries, embolism from the heart probably causes about one-fifth of ischaemic stroke and transient ischaemic attacks (TIAs), although a potential embolic source may be present in nearer one-third155–160. However, there are two very real and tiresome problems. As technology advances, more and more potential cardiac sources of embolism are being identified (Table 6.4), and patients may have two or more competing causes of cerebral ischaemia, such as carotid stenosis and atrial fibrillation. Therefore, it may well be unclear whether embolism from the heart is the cause in an individual patient, especially when the cardiac lesion is common in normal people.57,58,154,161–163 Not all emboli are of the same size, of the same age, or made of the same thing (fibrin, platelets, calcium, infected vegetations, tumour, etc.). Some are large and impact permanently in the mainstem of the middle cerebral artery to cause total anterior circulation infarction, others impact in a more distal branch of a cerebral artery to cause a partial anterior circulation infarct, others merely cause a transient ischaemic attack, and still others are asymptomatic.164–167 Emboli may also occlude the basilar artery and its branches, and even the internal carotid artery in the neck.61,165 Embolism from the heart causes about one-fifth of ischaemic strokes and transient ischaemic attacks. The most substantial embolic threats are non-rheumatic and rheumatic atrial fibrillation, infective endocarditis, prosthetic heart valve, recent myocardial infarction, dilated cardiomyopathy, intracardiac tumours and rheumatic mitral stenosis.

..

..

Table 6.4 Cardiac sources of embolism in anatomical sequence. Right-to-left shunt (paradoxical embolism from the venous system or right-atrial thrombus) (section 6.5.12) Patent foramen ovale Atrial septal defect Ventriculoseptal defect Pulmonary arteriovenous fistula Left atrium Thrombus atrial fibrillation* (section 6.5.1) sinoatrial disease (sick sinus syndrome) (section 6.5.14) atrial septal aneurysm (section 6.5.12) Myxoma and other cardiac tumours* (section 6.5.13) Mitral valve Rheumatic endocarditis (stenosis* or regurgitation) (section 6.5.4) Infective endocarditis* (section 6.5.9) Mitral annulus calcification (section 6.5.6) Non-bacterial thrombotic (marantic) endocarditis (section 6.5.10) Systemic lupus erythematosus/antiphospholipid syndrome (section 7.3.3 and 4) Prosthetic heart valve* (section 6.5.3) Papillary fibroelastoma (section 6.5.13) Mitral leaflet prolapse (uncertain) (section 6.5.7) Mitral valve strands (uncertain) (section 6.5.8) Left ventricle Mural thrombus acute myocardial infarction (previous few weeks)* (section 7.10) left-ventricular aneurysm or akinetic segment (section 7.10) dilated or restrictive cardiomyopathy* (section 6.5.11) mechanical ‘artificial’ heart* blunt chest injury (myocardial contusion) Myxoma and other cardiac tumours* (section 6.5.13) Hydatid cyst Primary oxalosis (uncertain, section 7.20.6) Aortic valve Rheumatic endocarditis (stenosis or regurgitation) (section 6.5.4) Infective endocarditis* (section 6.5.9) Syphilis Non-bacterial thrombotic (marantic) endocarditis (section 6.5.10) Systemic lupus erythematosus/antiphospholipid syndrome (section 7.3.3 and 4) Prosthetic heart valve* (section 6.5.3) Calcific stenosis/sclerosis/calcification (section 6.5.6) Aneurysm of the sinus of Valsalva Congenital heart disease (particularly with right-to-left shunt) Cardiac manipulation/surgery/catheterization/ valvuloplasty/angioplasty* (section 7.18.1) *Substantial risk of embolism.

271

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 272

272

Chapter 6 What caused this transient or persisting ischaemic event? Table 6.5 Prevalence of potential cardiac sources of embolism in 244 patients with a first-ever-in-a-lifetime ischaemic stroke in the Oxfordshire Community Stroke Project (Sandercock et al. 1989).10 Source

Number

Percentage

Any atrial fibrillation without rheumatic heart disease with rheumatic heart disease Mitral regurgitation Recent ( 85

Fig. 6.15 The incidence of first-ever-in-a-lifetime ischaemic stroke and transient ischaemic attack (TIA) in the Oxfordshire Community Stroke Project. The flattening of TIA and, to some extent, stroke incidence in old age may be because cases did not come to medical attention or, when they did, they were not correctly diagnosed in the elderly.

281

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 282

Chapter 6 What caused this transient or persisting ischaemic event?

First-ever-in-a-lifetime ischaemic stroke Transient ischaemic attack

3

4.00

2.00 Relative risk of stroke

Male/female ratio of incidence

282

2

1

1.00

0.50

0.25

0

15– 44

45–54

55– 64

65–74

75 – 84

> 85

Age (years) Fig. 6.16 The ratio of male : female incidence of first-ever-in-alifetime ischaemic stroke and transient ischaemic attack in the Oxfordshire Community Stroke Project.

explained by the natural menopause, although bilateral oophorectomy without oestrogen replacement about doubles the risk of vascular events.317,318 Use of hormone replacement therapy in women after the menopause does not confer a benefit in terms of reduced cardiovascular morbidity, in fact there is evidence that such treatment is associated with an increased risk of acute coronary sydrome, stroke and venous thromboembolism319–321 (section 7.13.2).

6.6.3 Blood pressure In healthy populations of both sexes and independently of age, increasing blood pressure is strongly associated with overall stroke risk, and of all the main pathological types, including ischaemic stroke.322–324 The relationship between usual diastolic blood pressure and subsequent stroke is log–linear with no threshold below which stroke risk becomes stable, at least not within the ‘normal’ range of 70–110 mmHg (Fig. 6.17). The proportional increase in stroke risk associated with a given increase in blood pressure is similar at all levels of blood pressure. This risk almost doubles with each 7.5 mmHg increase in usual diastolic blood pressure in Western populations, and with each 5.0 mmHg in Japanese and Chinese populations.325–327 Because moderately raised blood pressure is so common in the middle-aged and elderly in developed countries, high blood pressure probably accounts for more ischaemic strokes than any other risk factor. The strength of the association between blood pressure and stroke is attenuated with increasing age, although

1

2

3

4

5

76

84

91

98

105

Approximate mean usual diastolic blood pressure (mmHg) in each of five categories, 1–5 Fig. 6.17 The risk of stroke related to the usual diastolic blood pressure in five categories, defined by usual baseline blood pressure, from a pooled analysis of seven prospective observational studies. Solid squares represent stroke risks relative to risk in the whole study population; their size is proportional to the number of strokes in each blood pressure category. The vertical lines represent 95% confidence intervals. Note the doubling scale of the y-axis. (With permission from MacMahon et al. 1990.325)

the absolute risk of stroke in the elderly is far higher than in the young.327 Nevertheless, hypertension is still a risk factor in the very elderly, although it is weaker because stroke may be associated with low blood pressure secondary to cardiac failure and other comorbid conditions.328 Moreover, in patients with bilateral severe carotid stenosis, stroke risk is higher at low blood pressures suggesting that aggressive blood-pressure lowering may be harmful in this group (section 16.3.1).329 The relationship between stroke and systolic blood pressure is possibly stronger than with diastolic pressure, and even ‘isolated’ systolic hypertension with a ‘normal’ diastolic blood pressure is associated with increased stroke risk.327,330–333 There is no doubt, as confirmed by the results of randomized controlled trials, that the relationship between increasing blood pressure and stroke risk is causal (section 16.3). However, it is not clear whether all types of stroke are prevented by reducing blood pressure, largely because in the clinical trials they have all been lumped together, haemorrhagic with ischaemic strokes of various types182,334–337 (section 16.3.2). Progression of carotid stenosis, at least as assessed by ultrasound, is slowed by treating hypertension.338

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 283

6.6 Risk factors for ischaemic stroke

numbers.311,344,371 So far, randomized trials have not shown that diabetic treatment definitely reduces the risk of stroke372 (section 16.9), although improved glycaemic control does lead to a reduction in overall rates of vascular complications of diabetes and a reduction in premature death. In the UKPDS study, overweight patients with type 2 diabetes who were treated with metformin had a 32% reduction in diabetes-related adverse outcomes and a 42% reduction in diabetes-related deaths.372

Hypertension seems to increase the risk of ischaemic stroke by increasing the extent and severity of atheroma (section 6.3) as well as the prevalence of ‘simple’ and ‘complex’ intracranial small vessel disease339–344 (section 6.4).

6.6.4 Smoking Cigarette-smoking is associated with approximately double the risk of ischaemic stroke in males and females, but less obviously in the elderly, and there is a dose–response relationship.345–350 There are not as many data on passive smoking and stroke as there are for coronary events where the association is surprisingly strong.351–354 As one would expect, most of the ultrasound and angiogram studies link carotid disease with smoking.344,355–357 Although cigar-smoking increases the risk of coronary events by about one-quarter, there are insufficient data to link either pipe- or cigar-smoking with stroke, perhaps because there are fewer people who still indulge in these habits.358 The risk of stroke gradually declines after stopping smoking so supporting a causal relationship, but a satisfactory randomized controlled trial proved impossible.345,350,359–361

6.6.6 Blood lipids Increasing plasma total cholesterol, increasing lowdensity lipoprotein-cholesterol and decreasing levels of high-density lipoprotein-cholesterol are all strong risk factors for coronary heart disease, whereas triglyceride levels are not. A long-term reduction of plasma cholesterol by 1 mmol/L should and does reduce the relative risk of coronary events by at least one-third,373 and also reduces coronary risk in the elderly.374–379 On the other hand, the relationship with ischaemic stroke is less clear. Very large systematic reviews of cohort studies have not revealed any association between all stroke types combined and increasing plasma total cholesterol at baseline, except perhaps under the age of 45 years326,380 (Fig. 6.18). Case–control studies provide less reliable measures of association because of their biases, particularly the changes in plasma lipids following stroke381 (section 6.8.1). Relating carotid intima-media thickness or stenosis with blood lipids is perhaps too far from the clinical consequences of atheroma to be relevant, or the studies have been too small to be reliable.310,312,342,344,382,383 Surprisingly therefore, the Heart Protection Study of cholesterol lowering in patients with known vascular disease or diabetes showed that simvastatin definitely reduced the risk of stroke on follow-up, although it did

6.6.5 Diabetes mellitus Any studies linking diabetes with fatal stroke will exaggerate the association, because diabetics who have a stroke are more likely to die of it than non-diabetics362 (sections 10.2.7 and 11.18.3). Indeed, the the risk of fatal stroke is higher in those with a higher HbA1C at diagnosis363 In fact, diabetes about doubles the risk of ischaemic stroke over and above confounding with hypertension and other risk factors.364–370 Diabetics also have thicker carotid arterial walls but the relationship with carotid stenosis is less clear, probably because of lack of patient

Fig. 6.18 Adjusted proportional risk of stroke (with 95% confidence intervals) by age and usual plasma cholesterol from a systematic review of 45 prospective observational studies. (With permission from the Prospective Studies Collaboration 1995.380)

..

..

Proportional stroke rate

1.2 1.0

1.2

1.2 1.1

1.1

1.0

1.0 1.1 1.0 0.9

0.8

0.8

0.7 0.7 0.6

4.5

5.0

5.5

Age at screening < 45 yr 45–64 65 yr+ 6.0

Usual total plasma cholesterol (mmol/L)

1.0

283

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 284

Chapter 6 What caused this transient or persisting ischaemic event?

not reduce the risk of recurrent stroke384,385 (section 16.4.2). This may have been because incident strokes occurred on average 4.6 years before the study onset and hence, at the time of the study, patients would have been at low risk of recurrent stroke but at high risk of coronary vascular disease. More recently, the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial showed that atorvastatin in patients who had had a stroke or TIA within 1– 6 months before study entry did reduce stroke risk.386 However, statin treatment was associated with a small but significant increase in risk of haemorrhagic stroke. Interestingly, the same trend had been found in the Heart Protection Study (HPS) in the 3280 patients with previous stroke or TIA,384 in whom simvastatin 40 mg increased the risk of haemorrhagic stroke. Thus, the randomized evidence does suggest that there is a causal association between plasma LDL cholesterol and risk of ischaemic stroke – but more work is required to determine whether the increase in risk of haemorrhagic stroke is genuine. The marked contrast between coronary disease and ischaemic stroke in their association with plasma cholesterol is even more curious now it is clear that cholesterol lowering reduces the risk of myocardial infarction and stroke.384,386,387 It appears that the observational epidemiology has missed an ischaemic stroke–lipid connection, possibly because: • the seemingly negative association between increasing plasma cholesterol and intracranial haemorrhage has obscured a positive association with ischaemic stroke in studies where the pathological type of stroke was not accounted for (Fig. 6.19), which would tie in with the possibility that reducing plasma cholesterol with statins increases the risk of haemorrhagic stroke; • the over-representation of fatal, and therefore more likely to be haemorrhagic, strokes in some studies; • the narrow range of cholesterol levels examined in many studies; • the loss of stroke-susceptible individuals from the study population by prior death from coronary disease; • uncertainties about the effect of stroke itself on lipid levels in case–control studies; and • not differentiating ischaemic strokes likely to be caused by intracranial small vessel disease from those caused by large vessel atherothrombosis. Alternatively, the lack of a strong plasma cholesterol association with ischaemic stroke may be correct and perhaps the statins reduce stroke risk by some mechanism other than by cholesterol lowering.388,389 Both acute plaque-stabilizing effects of statins390 and neuroprotective effects391,392 have been postulated. Some support for potentially non-lipid effects comes from the

4.00 Non-haemorrhagic stroke (11 studies, 60 750 participants, 494 events) 2.00

1.00

0.50

Relative risk of stroke

284

0.25 4.0

4.00

4.5

5.0

5.5

6.0

Haemorrhagic stroke (11 studies, 60 750 participants, 404 events)

2.00

1.00

0.50

0.25 4.0

4.5

5.0

5.5

6.0

Approximate mean usual total plasma cholesterol concentration (mmol/L) Fig. 6.19 Overall adjusted relative risk (95% confidence intervals) of non-haemorrhagic and haemorrhagic stroke by usual plasma cholesterol from a systematic review of prospective observational studies in China and Japan. The size of the solid squares is proportional to the number of strokes in each cholesterol category. Note the doubling scale of the y-axis. (With permission from the Eastern Stroke & Coronary Heart Disease Collaborative Research Group, 1998.326)

apparent short-term effect of statins on risk of stroke after acute coronary syndromes.393 The inconsistency between the association between total or LDL cholesterol and ischaemic stroke and the effectiveness of statins in preventing ischaemic stroke also raises the possibility that other lipid subfractions that are affected by statins might be better predictors of ischaemic stroke.394

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 285

6.6 Risk factors for ischaemic stroke

Any association between plasma lipoprotein (a) and apolipoprotein E genotype with ischaemic stroke is still rather uncertain,395–397 but apolipoproteins (apo A1 and apo B), which make up the protein moiety of lipoproteins, do appear to be predictive of ischaemic stroke.398,399 Apo A1 is mainly found in HDL and apo B in LDL. Differences in prognostic value of apolipoproteins and LDL cholesterol are possible for several reasons. First, measurement of LDL cholesterol is an estimate of the mass of cholesterol in the LDL fraction of plasma, whereas measurements of apo B and apo A1 provide information on the total number of atherogenic (apo B) or anti-atherogenic (apo A1) particles. Second, the ratio of apo B : apo A1 best reflects the status of cholesterol transport to and from peripheral tissues. Third, unlike the relationships between total and LDL cholesterol and CHD, which weaken with age, apo B retains its predictive power in the elderly.400 Apo B is an established risk factor for coronary vascular events,400,401 and statin-mediated coronary event risk reduction has been attributed to apo B reduction.402 Data on apolipoproteins and ischaemic stroke risk are somewhat limited and conflicting, although there appears to be a relationship between raised apo B and an increased risk of ischaemic stroke, and an inverse correlation with apo A1 levels.398,399

6.6.7 Haemostatic variables Despite much effort, very few consistent associations have been found between coagulation parameters, fibrinolytic activity, platelet behaviour and vascular disease.403,404 Often these variables are altered by acute stroke so that most case–control studies are invalid, and it has not been practical to carry out very many long-term cohort studies. Plasma fibrinogen has a strong and consistent positive association with stroke and coronary events, including recurrent stroke.405,406 Cigarette-smoking is a confounding variable so the effect of cigarette-smoking on stroke may be mediated, at least in part, by increasing the fibrinogen level, thus accelerating thrombosis. Less important but still confusing confounding factors include age, hypertension, diabetes, hyperlipidaemia, lack of exercise, social class, social activity, season of the year, alcohol consumption and stress. It is not yet certain therefore whether increasing plasma fibrinogen really is a causal factor and, if so, whether it acts by increasing plasma viscosity or through promoting thrombosis. The confusion is compounded because there is no standard method for measuring plasma fibrinogen, it tends to rise after acute events including infections, it is not easy to lower plasma fibrinogen, and no satisfactory randomized controlled trials have been reported.344,381,405–412

..

..

Raised haematocrit is an uncertain risk factor for stroke and other acute vascular events, although it does seem to be associated with an increased case fatality in ischaemic stroke.413 The association is confounded by the fact that cigarette-smoking, blood pressure and plasma fibrinogen are all positively associated with haematocrit.414–416 Both raised haematocrit and raised plasma fibrinogen increase whole blood viscosity, another potentially causal risk factor.411 No randomized trials are available of lowering haematocrit, viscosity or fibrinogen. Raised plasma factor VII coagulant activity may be a risk factor for coronary events, and also polymorphisms of the factor VII gene, but there are very few data for stroke.417–419 Raised von Willebrand factor antigen is another possible risk factor for ischaemic stroke.417,420 Low blood fibrinolytic activity and high plasma plasminogen-activator type I are coronary risk factors and raised tissue plasminogen activator antigen may be associated with both coronary and stroke risk, perhaps because it is a marker of endogenous fibrinolytic activity.417,421–426 Abnormal platelet behaviour has not been convincingly linked with subsequent stroke in cohort studies, and any case–control studies after stroke have great potential for bias as stroke alters platelet function.427 Polymorphisms of platelet membrane glycoprotein IIIa were at first associated with stroke but, as seems to happen so often with genetic studies, then rejected on the basis of larger and more methodologically sound work.428

6.6.8 Plasma homocysteine Because the rare inborn recessive condition of homocystinuria is complicated by arterial and venous thrombosis (section 7.20.2), it is natural to imagine that mildly raised levels of plasma homocysteine could be a risk factor for or even a cause of vascular disease in general. Raised plasma homocysteine is positively associated with increasing age, abnormal lipids, smoking, diabetes, chronic renal failure and hypertension, and it is also raised in nutritional deficiencies and probably after stroke and myocardial infarction. The observational data linking coronary events, stroke, venous thromboembolism and carotid intima-media thickness and stenosis with increasing plasma homocysteine (hyperhomocysteinaemia) are now fairly robust.429–435 In view of the strength of the epidemiological association between homocysteine and risk of stroke, several trials of homocysteine-lowering with folic acid and pyridoxine are being performed, some of which have now been reported, but so far the results are not promising and hyperhomocystinaemia cannot yet be regarded as a causal risk factor (section 16.8.1).

285

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 286

286

Chapter 6 What caused this transient or persisting ischaemic event?

6.6.9 Physical exercise Physical exercise somewhat reduces blood pressure, weight, plasma cholesterol, plasma fibrinogen and the risk of non-insulin-dependent diabetes mellitus and is associated with less cigarette-smoking.436,437 Therefore, not surprisingly, lack of exercise is associated with coronary events and there is also strong evidence of a similar association with stroke.438–443 So far there is insufficient evidence from randomized trials to be sure that deliberately increasing exercise levels, in sporting activities or as part of a generally healthy lifestyle, reduces the risk of vascular events in the general population444 (section 16.7.4). However, there is evidence (see below) that exercise can prevent progression to diabetes in patients with the metabolic syndrome as part of an overall lifestyle modification programme.445,446

6.6.10 Obesity Any relationship between obesity and stroke is likely to be confounded by the positive association of obesity with hypertension, diabetes, hypercholesterolaemia and lack of exercise, and the negative association with smoking and concurrent illness. Nevertheless, stroke is more common in the obese and so is overall mortality. How to measure obesity is itself somewhat controversial; a raised waist : hip ratio as a measure of central obesity, and perhaps change in body weight, may be stronger risk factors than the traditional measure of weight compared with height (see below).447–451

exercise;445,446 and angiotensin-converting enzyme (ACE) inhibitors or A2 blockers (ARB).455–457

6.6.12 Diet It is technically difficult to measure what people eat and how they cook it, relate any particular diet to events such as stroke that occur years later, and to be sure that any association is not better explained by a confounding variable.458 For example, people who eat a lot of fruit tend to have a generally ‘healthy’ lifestyle, not to smoke, and they may use less salt.459 Salt, by increasing systemic blood pressure, is probably associated with increased stroke risk but this issue is still controversial.460–463 In observational studies, diets low in the following may be associated with a raised risk of coronary disease and stroke:464 • potassium465–467 • calcium466,468 • fresh fruit and vegetables469,470 • fish469–473 • antioxidants such as vitamin E,474–476 vitamin C,477–479 beta-carotene480–482 and flavonols.483,484 On the whole, randomized trials of dietary interventions and vitamin supplementation have been disappointing (section 16.7.3). Either the theory is wrong, or the interventions were too little, too late, for insufficient time or the trials were too small.385,455,485–489 The effect of a vegetarian diet is unclear.490 Despite earlier enthusiasm, it now appears that there is no association between coffee consumption and vascular disease.491–494

6.6.11 The metabolic syndrome Obesity, specifically visceral obesity, plays a key role in the development of the metabolic syndrome and type 2 diabetes. Visceral adipocytes are insulin-resistant and highly metabolically active and promote dyslipidaemia, hypertension and reduced systemic thrombolysis, as well as leading to a relatively pro-inflammatory state.452 The constellation of metabolic abnormalities including central obesity, decreased HDL, raised triglycerides and blood pressure, and hyperglycaemia is known as the metabolic syndrome, and this is associated with a threefold increased risk of type 2 diabetes and a twofold increase in cardiovascular risk, including stroke.453,454 It is thought to be the main driver for the modern-day epidemic of diabetes and vascular disease. As well as the prevention of acute vascular events in patients with the metabolic syndrome, primary prevention of which is mostly the same as for other groups,453,454 one additional aim should be to prevent progression to frank diabetes. Two strategies have been shown to be effective in this regard – lifestyle modification with diet and

6.6.13 Alcohol The relationship between alcohol, ischaemic stroke and carotid atheroma is complex and may be U-shaped.495 While heavy alcohol consumption may be an independent and perhaps causal risk factor, it seems that modest consumers are protected to some extent compared with abstainers.496–504 Whether ‘binge’ drinking is associated with stroke is uncertain505,506 but it is possible that irregular drinking carries a higher risk.507 Much of the confusion in relating alcohol consumption with stroke is because:508 • people are not always truthful about their alcohol consumption; • it is difficult to measure alcohol consumption accurately, particularly over time; • there are varied ways of expressing alcohol consumption (per day, per week, grams, units, number of drinks, regular, binge, etc.); • different types of alcoholic drinks may have different effects;

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 287

6.6 Risk factors for ischaemic stroke

• pattern of drinking behaviour may change over time; • combining ex-drinkers, some of whom may have given up drinking because of symptoms of vascular or other diseases, with lifetime non-drinkers in the analyses; • the biases inherent in case–control studies; • publication bias; • small numbers and so imprecise estimates; • confounding with cigarette-smoking, hypertension and deprivation which are positively related, and with exercise which is negatively related with alcohol consumption; • confounding with unknown or unmeasurable factors which might link no drinking or heavy drinking with an excess risk of vascular disease (e.g. healthy vs risky lifestyle); • lumping ischaemic and haemorrhagic strokes together. It is difficult to think of the same biological reason for ischaemic stroke in non-drinkers and heavy drinkers. Possible causal explanations, at least in heavy drinkers, are that alcohol almost certainly increases the blood pressure; traumatic arterial dissection in the neck during an alcohol-related injury; dehydration, hyperviscosity and platelet activation perhaps; sleep apnoea and hypoxia; and that alcohol can cause both atrial fibrillation and cardiomyopathy with embolism to the brain.509–512 Perhaps it would be more productive to ask why modest drinking is protective, if it really is. Increased plasma high-density lipoprotein-cholesterol and lower fibrinogen levels are possibilities.513 It is most unlikely that a randomized trial of modest alcohol consumption, in the hope of reducing the risk of vascular disorders without increasing the risk of other disorders, will ever be feasible. Therefore, we are left trying to interpret, with difficulty, the available observational data (section 16.7.2).

6.6.14 Ethnicity There are no good population-based data on stroke incidence in developing countries, with yawning gaps particularly in South and South-East Asia and subSaharan Africa48,514,515 (section 18.2.2). Nonetheless, and notwithstanding the difficulty in defining ethnicity, stroke incidence does seem greater in black than white people living in Western countries, probably both ischaemic stroke and intracerebral haemorrhage.516–520 The high prevalence of hypertension, diabetes, sickle cell trait and social deprivation in black people may be part of the explanation.521–523 South Asian populations in the UK have a high stroke mortality as well as a high prevalence of coronary disease, central obesity, insulin resistance, diabetes mellitus and other risk factors.524,525 This is in part because they are genetically more at risk

..

..

than white populations by virtue of higher serum lipoprotein (a) concentrations.526–529 Also, Maori and Pacific people living in New Zealand have a higher stroke risk than white New Zealanders.530 Moreover, the fall in stroke incidence that has occurred in white New Zealanders has not been mirrored by any similar fall in the Maori and Pacific population.531

6.6.15 Specific genes for common types of ischaemic stroke Based on the assumption that at least some of the risk for stroke is genetic, large numbers of studies have now been done or are being done in an attempt to identify the genes involved.532 However, on the whole, methodological rigour – and so the reliability of the results of these studies – has been limited. It seems likely that the genetic component of stroke risk is modest and that many (probably hundreds) of genes are involved, each one contributing only a small increased risk. But, studies so far have generally not been large enough to detect reliably the sort of relative risks (i.e. the risk ratio comparing one genotype with another at a particular genetic locus) that might realistically be expected (probably about 1.2 to 1.5). Other methodological limitations include: poor choice of controls in case–control studies and/or failure to use methods designed to detect and control for selection bias; inadequate distinction between the different pathological types and subtypes of stroke for which genetic influences may differ; failure to replicate positive results in an independent and adequately sized study; and testing of multiple genetic or subgroup hypotheses with no adjustment of p-values for declaring statistical significance.533,534 Candidate gene studies Most genetic studies so far have been candidate gene studies, in which the frequency of different genotypes at a specific locus or loci within a gene or genes thought likely to be in some way connected with stroke risk are compared between stroke cases and stroke-free controls. Candidate genes have generally been selected on the basis of their known or presumed involvement in the control of factors or pathways likely to influence stroke risk: blood pressure, lipid metabolism, inflammation, coagulation, homocysteine metabolism and so on.535,536 Rigorous meta-analyses of candidate gene studies, both in stroke and other vascular diseases such as coronary heart disease, have highlighted the various methodological issues discussed above. In particular, they have drawn attention to the problems of small study size (most studies have included a few hundred or less, rather than the required 1000 or more cases and controls),

287

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 288

288

Chapter 6 What caused this transient or persisting ischaemic event?

leading to the potential for publication bias (i.e. small positive studies are far more likely to be published and available for inclusion in meta-analyses than small negative ones), with very large studies tending to detect either much smaller relative risks or no evidence of altered risk at all.534,537,538 Large numbers of candidate gene studies have together identified a handful of genes (those encoding factor V Leiden, methylenetatrahydrofolate reductase, prothrombin and angiotensin-converting enzyme) which, on the basis of results from metaanalyses, seem likely to influence risk of ischaemic stroke modestly.536 However, since none of the contributing studies was large enough to be independently reliable, occult publication bias might still account for the apparent effects of these few genes, and confirmation in further very large studies is needed. Linkage studies As yet there have been far fewer stroke genetics studies that use more traditional genetic study designs, based on collecting information and DNA from related individuals with and without the disease of interest. This is at least partly because family members of stroke patients (who of course tend to be elderly) are often no longer still alive, and so obtaining information and samples for DNA extraction from large enough numbers of relatives is challenging.539 There are some exceptions, though. For example, the Icelandic population has excellent genealogical records allowing the construction of complex pedigrees, along with stored DNA and medical histories on large numbers of people in the population. Family or pedigree-based genetic study designs allow for linkage analysis, which has the attraction of not being driven by hypotheses about specific candidates, but instead sets out to identify which regions of the genome may influence disease risk, through analysing the degree of sharing of genetic information (at markers across the whole genome) between related individuals with and without disease. The Icelandic deCODE group has carried out a number of linkage studies to identify regions of the genome likely to influence risk of various common diseases, including stroke (see http://www.decode.com/ Population-Approach.php). Through this process, two candidate genes for ischaemic stroke, encoding the enzymes phosphodiesterase-4D and arachidonate 5-lipoxygenase-activating protein ALOX5AP, have been identified.540 There is still some debate, however, about their influence in non-Icelandic populations, since their effect on ischaemic stroke risk has been confirmed in some, but not all, subsequent studies, and the confirmatory studies have not been free of the methodological problems discussed above.541,542

Whole genome association studies The combination of technological developments allowing rapid throughput genotyping at multiple loci, the attraction of non-hypothesis-driven genetic studies, and the recognized limitations of traditional linkage approaches have together led to a new wave of genomewide association studies, where multiple (up to 500 000 or more) polymorphisms across the genome are genotyped and compared in cases and controls, looking for loci where significant differences may suggest genetic influences on disease risk. The statistical considerations required for such studies are considerable, since huge numbers of potential genetic risk factors, albeit not completely independent of each other (because segments of the genome tend to be transmitted together during meiosis – linkage dysequilibrium), are tested simultaneously.543,544 Thus, there is great potential for missing real genetic effects or for finding spurious ones. Nonetheless, this approach has had a few (perhaps lucky) successes in common diseases; one small, preliminary study in stroke has now been published, and no doubt others will follow.545 Genetic studies of other vascular diseases, known vascular risk factors and intermediate phenotypes The results of genetic studies of other vascular diseases (in particular coronary heart disease) may also prove useful in identifying likely genetic candidates for ischaemic stroke. In addition, since much of the genetic risk of stroke seems likely to be mediated through alreadyknown risk factors such as cholesterol and blood pressure (sections 6.6.3 and 6.6.6), studies of these may also provide useful information about likely stroke risk genes. Finally, genetic studies have started to emerge of so-called ‘intermediate phenotypes’, markers of predisposition to stroke (or other vascular diseases), which can be measured in large numbers of subjects both with and without vascular risk factors or disease. These intermediate phenotypes include carotid intima media thickness and leukoaraiosis (measured or graded on CT or MR brain scans), and both linkage and candidate gene approaches have been used.533,546 As with studies of clinically manifest disease, methodological rigour and adequate study size are crucial, and so systematic review and meta-analysis approaches will be helpful in interpreting the large numbers of studies and quantities of data emerging.547 We are still far from being able to quantify the genetic influences on stroke, or to identify reliably the genes involved. One serious challenge is that the available laboratory technology is currently outstripping our ability to properly analyse and interpret the results. In

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 289

6.6 Risk factors for ischaemic stroke

the future, only careful integration of information from all of the different approaches described above, together with the results of laboratory studies of gene expression and function, from the molecular to the animal model level, will shed useful light on the genetic influences on stroke, hopefully allowing us to understand better the way in which already-known risk factors influence stroke, and reliably to identify new ones.

6.6.16 Social deprivation Social deprivation, low socioeconomic status and unemployment are all inextricably linked and associated with increased stroke risk.548,549 This association is partly caused by a higher prevalence of vascular risk factors, stress and adverse health behaviours, such as smoking, poor diet and lack of exercise in deprived populations.550 For example, these factors appear to account for most of the north–south gradient in stroke mortality in the UK.551 Also, it has been suggested that poor health and nutrition in utero or infancy are associated with the development of vascular risk factors such as hypertension and adult vascular disease, including stroke.552–555 It is extremely difficult to disentangle early from later life influences on health and so, not surprisingly, the ‘early life’ or ‘Barker’ hypothesis has been refuted by some.556–559 The possibility of reducing stroke risk by tackling social deprivation is yet another reason to add to the many others to support what has to be political rather than medical action.

6.6.17 Infection There has been interest in the notion that infection contributes to the development of atheroma, the progression of the atherothrombotic plaque, and perhaps plaque instability. However, the evidence from observational epidemiological studies of infections in general,560 and of chronic dental infection,561 together with serological evidence of specific infectious agents, such as Chlamydia pneumoniae, Helicobacter pylori and cytomegalovirus, is not very convincing, even for coronary disease where there are – as usual – more data than for stroke.562–569 The generalized inflammatory response in patients with coronary artery disease may be too non-specific to be convincing evidence of infection, and there may well be the familiar problems of confounding and publication bias.412,563,564,570 Nonetheless, the fact that evidence of chlamydial infection can be found in atherosclerotic plaques is certainly interesting although reverse causality is a possibility, i.e. atheroma becomes infected more often than normal arterial wall.571 An early randomized trial to eliminate chlamydia was far

..

..

too small to be reliable despite being published in a high-impact journal.572 Further antibiotic intervention trials are ongoing.573,574

6.6.18 ‘Stress’ It is part of folklore that stress causes strokes: ‘I was so upset, I nearly had a stroke’. Indeed, there are some striking anecdotes. This 82-year-old lady was admitted for investigation of anaemia and hepatomegaly. She was found to have multiple liver metastases. When she was told the news, she stopped speaking and never spoke again. It took a day or so before the medical staff realized that her mute state was not ‘psychological’ but that she was aphasic. A brain CT scan showed a left cortical infarct. She died some days later from her malignancy. However, showing that long-term psychological stressors are associated with stroke, or that shorter-term ones precipitate stroke, is not easy. There is some evidence that severely threatening life events,575,576 anxiety and depression,577 high levels of anger expression578,579 and psychological stress577,580,581 may trigger the onset of stroke, perhaps in people already at risk of stroke.

6.6.19 Non-stroke vascular disease Because atheroma in one artery is likely to be accompanied by atheroma in others (section 6.3.1), and because embolism from the heart is a common cause of ischaemic stroke (section 6.5), non-stroke vascular disorders are associated with (i.e. are risk factors for) ischaemic stroke and transient ischaemic attack (Table 6.8). Coronary heart disease (e.g. angina or myocardial infarction) has been repeatedly associated with an increased risk of stroke in postmortem,582,583 twin,584 case– control190,585–587 and cohort studies.170,194,330 Therefore, it is not surprising that electrocardiogram abnormalities and cardiac failure are also associated with increased risk of stroke because they both so often reflect coronary Table 6.8 Degenerative vascular disorders outside the head associated with an increased risk of ischaemic stroke and transient ischaemic attack. Myocardial infarction/angina Cardiac failure Left ventricular hypertrophy Atrial fibrillation Cervical arterial bruit/stenosis Peripheral arterial disease

289

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 290

290

Chapter 6 What caused this transient or persisting ischaemic event?

disease and hypertension, as is left ventricular hypertrophy.330,588–592 Atrial fibrillation is considered in section 6.5.1. Cervical bruits (carotid or supraclavicular) are generally caused by stenosis of the underlying arteries (section 6.7.7), mostly as a result of atheroma, and so become more common with age; about 5% of asymptomatic people over the age of 75 years have bruits.593–595 Carotid bruits are clearly a risk factor for ischaemic stroke, but not necessarily in the same arterial territory as the bruit, and also for coronary events, because atheroma of one artery is likely to be accompanied by atheroma of other arteries in the same predisposed individual.596,597 The risk of these various vascular events increases with carotid stenosis severity (section 16.11.5). Atheroma affecting the leg arteries is associated with cerebrovascular and coronary disease in the same individuals so often that it is not surprising that claudicants have an increased risk of stroke and other serious vascular events.194,598–602 Little seems to be known about the prevalence of abdominal aortic aneurysms in ischaemic stroke/TIA patients. It is said to be quite high, about 10–20%, depending on the selection of patients and the size criterion for what constitutes an aneurysm.602–605 The incidence of aneurysm rupture in the general population is 5–10 times higher in men than in women,316 and so there might be an argument for screening men presenting with transient ischaemic attack or stroke, particularly perhaps if the cause is thought to be atherothromboembolism and bearing in mind the advantages of screening even in ‘normal’ elderly men from the Multicentre Aneurysm Screening Study (MASS).606

6.6.20 Other risk factors Innumerable other risk factors for coronary heart disease, and/or ischaemic stroke, have been suggested and supported by varying degrees of evidence. However, it may take many years, even decades, before a particular risk factor is accepted as causal, as has been the case for hypertension and smoking. So often initial discoveries of ‘new’ risk factors are found subsequently to be spurious.607 Even by 1981, 246 risk factors had been counted and by now there must be many more.608 Some may or may not be important in the causal pathway to stroke but, even if not, they may still conceivably be helpful in predicting future stroke in individuals and populations. This distinction is important: for example, claudication is clearly not on the causal pathway to stroke but is so strongly associated with stroke risk that it is included in many statistical models which predict later stroke (section 16.2).

6.7 From symptoms, signs and clinical syndrome to cause

Once the diagnosis of ischaemic stroke or transient ischaemic attack (TIA) has been made, the cause of the ischaemic event needs to be established since this has implications for treatment, prognosis and risk of recurrence. Four clinical syndromes, based on the history and examination, can be identified: total anterior circulation infarction, partial anterior circulation infarction, lacunar infarction and posterior circulation infarction. The clinical syndrome reasonably predicts the site and size of the brain lesion, which takes one a long way towards the likely cause of the ischaemic event (Fig. 6.20). Clinical localization is easier if the patient has had an established stroke with stable physical signs rather than being in the very early stages when the signs are still evolving (brain attack), or has had a TIA and any signs have disappeared. However, stroke localization based on the history and examination is by no means infallible: in about onequarter of cases where a recent lesion is visible on brain imaging, it is not in the expected place to explain the clinical syndrome.133 For example, although most pure motor strokes are caused by a lacunar infarct as a result of small vessel disease, in a few cases the CT or MR scan shows striatocapsular infarction which is likely to be caused by middle cerebral artery occlusion with good cortical collaterals (sections 4.2.2 and 6.7.2; fig. 6.20e). And in a recent study of 150 patients with minor stroke and a single acute lesion on DWI, agreement on localization to anterior or posterior vascular territory between DWI and the clinical judgements of three independent neurologists was only moderate (kappa = 0.5).609 Thus, brain imaging, in particular new techniques such as MR DWI, can and if possible should be used to refine lesion localization and thus aid in the search for the cause of the ischaemic stroke/TIA in an individual patient. In acute stroke, computed tomography or magnetic resonance imaging should be neither the first nor only way to classify patients on the basis of the size and site of any ischaemic brain lesion. Imaging is used to confirm and refine where the symptoms and neurological signs – in other words, the clinical syndrome – suggest the lesion is. From there, it is possible to narrow down the potential causes of the infarct.

6.7.1 Total anterior circulation infarction The acute ischaemic stroke clinical syndrome of a total anterior circulation infarction (TACI) comprises

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 291

6.7 From symptoms, signs and clinical syndrome to cause

a hemiparesis, with or without hemisensory loss, homonymous hemianopia, and a new cortical deficit such as aphasia or neglect. It is a good predictor of infarction of most of the middle cerebral artery (MCA) territory on brain CT, as a consequence of occlusion of either the MCA mainstem (or proximal large branch) or the internal carotid artery (ICA) in the neck (section 4.3.4) (Fig. 6.20c). Occasionally, a TACI can be caused by occlusion of the posterior cerebral artery, but the hemiparesis is usually rather mild. The cause of the arterial occlusion therefore is usually in the heart (e.g. embolism as a consequence of atrial fibrillation, recent myocardial infarction, etc.), or it is atherothrombosis complicated by embolism or occasionally propagating thrombosis of the ICA, or embolism from the aortic arch. Therefore, if the heart is clinically normal (history, examination, chest X-ray and electrocardiogram) and if there is no evidence of arterial disease in the neck (bruits, palpation perhaps, but mainly duplex sonography), then it is important to consider rarities, for example, infective endocarditis (echocardiogram, blood cultures) and carotid dissection (MR angiogram, and certainly recheck for past history of neck trauma, or associated neck or face pain). Although transcranial Doppler may confirm an MCA mainstem occlusion (but not if it has already recanalized), it will not help much in the search for a cause.134 A catheter angiogram might be diagnostically helpful if it could be justified on the basis of changing the patient’s management (e.g. traumatic carotid dissection could lead to later litigation, fibromuscular dysplasia could stop the search for other explanations, a giant aneurysm with contained thrombus might be surgically treatable, and so on). MRI and MR angiography, and CT angiography, are now widely and more appropriately used to show lesions, such as dissection and aneurysms, but they are still not always easily available in patients in the acute stage of stroke (sections 6.8.3 and 6.8.4).

6.7.2 Partial anterior circulation infarction Partial anterior circulation infarction (PACI) is a more restricted clinical syndrome with only two out of the three components of a TACI; or a new isolated cortical deficit, such as aphasia; or a predominantly proprioceptive deficit in one limb; or a motor/sensory deficit restricted to one body area or part of one body area (e.g. one leg, one hand, etc.) (section 4.3.5). This syndrome is reasonably predictive of a restricted cortical infarct caused by occlusion of a branch of the middle cerebral artery (MCA) or, much less commonly, of the anterior cerebral artery, as a result of embolism from the heart or from proximal sites of atherothrombosis (usually the

..

..

carotid bifurcation), or to any other cause of a TACI (Fig. 6.20b). Investigation is therefore similar to that for the patient with a TACI, except it is usually easier because the patient is fully conscious and less neurologically impaired. However, investigation must be quicker because of the higher risk of early recurrence (section 16.2.3)79,80 and because the patient has more to lose from a recurrence which might, next time, be a TACI. The potential for secondary prevention must be considered, particularly eligibility for carotid endarterectomy, and this requires early duplex sonography to find any severe carotid stenosis (section 6.8.5), and eligibility for anticoagulation if the patient is in atrial fibrillation. Transcranial Doppler (TCD) is unlikely to demonstrate the blocked cerebral artery because this is almost always distal to the MCA mainstem, at a point where TCD is not particularly sensitive.134 Occasionally, however, patients with a large PACI, but falling short of the full definition of a TACI, do have MCA mainstem occlusion, presumably because good collateral flow to the margins of the central infarcted area of brain restricts the clinical syndrome. This is particularly likely with striatocapsular infarction, which usually presents as a PACI (Figs 4.14, 6.20e). Some other PACI syndromes are caused by infarction in the centrum semiovale (section 4.2.2) and in boundary zones (section 6.7.5). Anterior choroidal artery infarcts may also present as a PACI (or a lacunar) syndrome and they seem to be caused by either embolism from proximal sites or intracranial small vessel disease (section 4.2.2). Total and partial anterior circulation infarction/transient ischaemic attacks are usually caused by occlusion of the mainstem or a branch of the middle cerebral artery, by occlusion of the anterior cerebral artery, or by occlusion of the internal carotid artery. Such occlusions are usually caused by embolism from the heart, embolism from proximal arterial sites of atherothombosis (the internal carotid artery origin, the aortic arch, etc.), and sometimes by thrombotic occlusion of severe internal carotid artery stenosis.

6.7.3 Lacunar infarction Lacunar syndromes, the vast majority of which are ischaemic rather than due to intracerebral haemorrhage, are almost always caused by small, deep, infarcts more likely to be seen on MRI than brain CT (section 4.3.2) (Fig. 6.20d). These small, deep, infarcts are mostly caused by a vasculopathy affecting the small perforating arteries of the brain, and not by embolism from proximal

291

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 292

292

Chapter 6 What caused this transient or persisting ischaemic event?

(a)

D C

D

C B

A

(b)

(c)

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 293

6.7 From symptoms, signs and clinical syndrome to cause

(d)

(e)

Fig. 6.20 (opposite & above) Various patterns of arterial occlusion cause different types of ischaemic stroke. Left-hand columns: axial CT brain scan at the level of the basal ganglia; middle columns, diagram to correspond with the CT brain scan with the area of infarction shaded; right-hand columns: diagram of the middle cerebral artery (MCA) and anterior cerebral arteries on a coronal brain section with the area of infarction shaded. A, main trunk of MCA; B, lenticulostriate perforating branches of the MCA; C, cortical branches of the MCA; D, cortical branches of the anterior cerebral arteries. (a) Normal arterial anatomy and CT scan. (b) Occlusion – usually embolic (straight arrow) from heart, aorta or internal carotid artery – of a

..

..

cortical branch of the MCA and restricted cortical infarct on CT (curved arrows); partial anterior circulation infarction (PACI). (c) Occlusion – usually embolic (straight arrow) as in (b) above – of MCA mainstem to cause infarction of entire MCA territory (curved arrows); total anterior circulation infarction (TACI). (d) Occlusion of one lenticulostriate artery (straight arrow) to cause a lacunar infarct (curved arrow); lacunar infarction (LACI). Note that the patient has an old lacunar infarct in the opposite hemisphere. (e) Occlusion of the MCA mainstem (straight arrow) but with good cortical collaterals from the anterior and posterior cerebral arteries to cause a striatocapsular infarct (curved arrows).

293

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 294

294

Chapter 6 What caused this transient or persisting ischaemic event?

arterial sources or the heart (section 6.4) and thus have a low risk of early recurrence71 There is not therefore the same urgency to rule out a cardiac source of embolism or severe carotid stenosis as there is for a partial anterior circulation infarction (PACI) (section 6.7.2). The vast majority of lacunar stroke syndromes are caused by ischaemia rather than haemorrhage. Most ischaemic lacunar strokes are the result of a small, deep, not a cortical, infarct. These small, deep infarcts are usually within the distribution of a small, perforating artery. The underlying vascular pathology is probably ‘complex’ small vessel disease, which differs from atheroma, but sometimes atheroma of the parent artery may occlude the mouth of the perforating artery. Lacunar infarcts are seldom caused by embolism from the heart or from proximal arterial sources. The capsular warning syndrome has a rather characteristic pattern. Over hours or days, there is cluster of transient ischaemic attacks (TIAs), consisting typically of weakness down the whole of one side of the body without any cognitive or language deficit (i.e. pure motor lacunar TIAs). These may be followed within hours or days by a lacunar infarct in the internal capsule. This syndrome is presumably caused by intermittent closure of a single lenticulostriate or other perforating artery, followed by complete occlusion, and one is unlikely to find a proximal arterial or cardiac cause, as in any other type of lacunar ischaemic stroke.346

6.7.4 Posterior circulation infarction Ischaemia and infarction in the brainstem and/or occipital region is aetiologically more heterogeneous than in the other three main clinical syndromes61,62 (section 4.3.3). Emboli from the heart may reach a small artery supplying the brainstem (e.g. superior cerebellar artery) to cause a fairly restricted deficit, block the basilar artery to produce a major brainstem stroke, travel on to block one or both posterior cerebral arteries to cause a homonymous hemianopia or cortical blindness, or any combination of these deficits. Similarly, embolism from the vertebral artery (as a result of atherothrombosis usually, but sometimes another disorder such as dissection) or from atherothrombosis of the basilar artery, aortic arch or innominate or subclavian arteries produces exactly the same neurological features as embolism from the heart.63,165,610 Even embolism from the carotid territory can, in some individuals with a dominant posterior communicating artery or a persistent trigeminal artery, cause occlusion of the posterior cerebral artery and even

brainstem infarction.611,612 Basilar occlusion, usually as a result of severe atherothrombotic stenosis, is likely to produce massive brainstem infarction. Obstruction, usually by atherothrombosis, of the origin of the small arteries arising from the basilar artery, can produce restricted brainstem syndromes, as can ‘complex’ small vessel disease within the brainstem; certainly some patients with a lacunar syndrome have a small infarct in the brainstem. A posterior circulation infarction (POCI) does not therefore provide much of a clue to the cause of the ischaemic event. An exception is the patient with simultaneous brainstem signs and a homonymous hemianopia, where embolism from the heart or a proximal artery must be the likely cause and not small vessel disease. Posterior circulation infarction/transient ischaemic attack can be due to almost any cause of cerebral ischaemia, which makes it very difficult to be certain of the exact cause in an individual patient if one knows no more than just the lesion localization. Cerebellar ischaemic strokes (section 4.2.3) are mostly caused by embolism from the heart, vertebral and basilar arteries, or by atherothrombotic occlusion at the origin of the cerebellar arteries; some are said to result from low blood flow alone.613–616 Thalamic infarcts (section 4.2.3) can be caused by: ‘complex’ small vessel disease affecting one of the small perforating arteries; atheromatous occlusion of these same arteries where they arise from the posterior cerebral and other medium-sized arteries; and occlusion of these latter arteries by embolism from the heart, basilar, vertebral and other proximal arterial sites.617–619

6.7.5 Ischaemic strokes and transient ischaemic attacks caused by low cerebral blood flow The pressure gradient across, and blood flow through, large arteries is not affected until their diameter is reduced by more than 50%, often not until by much more.620–622 Not surprisingly, therefore, even if there is severe disease of the carotid or vertebral arteries, cerebral perfusion pressure is usually normal. However, in some patients, as the stenosis becomes more severe, flow does fall, and eventually cerebral vasodilatation (autoregulation) cannot compensate for the low cerebral perfusion pressure. Regional cerebral blood flow (CBF) then falls, particularly if the collateral circulation is compromised because the circle of Willis, for example, is incomplete or diseased53,66,623–625 (section 4.2.2). At this stage of exhausted cerebral perfusion reserve, the ratio of cerebral blood flow : cerebral blood volume falls below about 6.0,

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 295

6.7 From symptoms, signs and clinical syndrome to cause

oxygen extraction fraction starts to rise and stroke risk probably also rises53,626–629 (section 12.1.2). Using transcranial Doppler (TCD) to demonstrate impaired cerebrovascular reactivity to a chemical rather than perfusion challenge is an indirect but more practical alternative to positron emission tomography (PET), but the correlation is not perfect624 (section 6.8.9). Isotopic measurement of the mean cerebral transit time,630 gradient echo and perfusion-weighted MRI,625,631 MR angiography,632 CT perfusion633,634 and near-infrared spectroscopy635 are other possibilities. Therefore, although the notion that ischaemic strokes may be caused by ‘hypotension’ goes back many years, low regional CBF alone is not particularly common and cannot easily explain more than a small fraction of strokes. Severe arterial stenosis or occlusion, or good evidence of a fall in systemic blood pressure just before onset, are simply not present in most ischaemic stroke cases.636 Most ischaemic strokes and transient ischaemic attacks (TIAs) must be caused, we believe, by embolic or in situ acute (usually thrombotic) occlusion of an artery to the brain causing blood flow to be suddenly cut off, so causing ischaemia in its territory of supply. Naturally, at times, focal ischaemia could also be caused just by low flow without acute vessel occlusion but usually only distal to a severely stenosed or occluded internal carotid (ICA) or other artery. This is where the vascular bed is likely to be maximally dilated and therefore where the brain is particularly vulnerable to any fall in perfusion pressure (even more so if arteries carrying collateral blood flow are also diseased). Under these circumstances, a small drop in systemic blood pressure might cause transient or permanent focal ischaemia without any acute occlusive event. Under normal circumstances quite a large fall in blood pressure does not cause cerebral symptoms, provided it is transient. This is because of autoregulation of CBF (section 12.1.2). If it does, the symptoms are much more likely to be nonfocal (faintness, bilateral blurring of vision, etc.) than focal637 (sections 3.2.1 and 3.4.12). Boundary-zone ischaemia and infarction Sometimes, ischaemia occurs not within but between major arterial territories in their boundary zones (section 4.2.4). Because this is where perfusion pressure is likely to be most attenuated, it is conceivable that ‘low flow’ as a result of low perfusion pressure, as well as the more common acute arterial occlusion caused by embolism, can cause ischaemia in these areas.638 The alternative term of watershed infarction is a misnomer based on geographical ignorance. A watershed is the line separating the water flowing into different river basins (i.e. the

..

..

high ground between two drainage areas), which is quite different from the pattern of arterial supply where flow is from larger to smaller vessels. The evidence that at least some boundary-zone infarcts are caused by low flow rather than acute arterial occlusion is that sudden, profound and relatively prolonged hypotension (e.g. as a result of cardiac arrest or cardiac surgery) sometimes causes infarction bilaterally in the posterior boundary zones, between the supply territories of the middle cerebral artery (MCA) and the posterior cerebral artery in the parieto-occipital region. The clinical features include cortical blindness, visual disorientation and agnosia, and amnesia. Unilateral posterior boundary-zone infarction causes contralateral hemianopia, cortical sensory loss and, if in the dominant hemisphere, aphasia. Also, distal to severe carotid stenosis or occlusion, unilateral infarction is well recognized in the anterior boundary zone between the supply territories of the MCA and anterior cerebral artery in the frontoparasagittal region, but this does not necessarily mean that the cause was low flow rather than embolism. The clinical features are contralateral weakness of the leg more than the arm and sparing the face, some impaired sensation in the same distribution, and aphasia if in the dominant hemisphere57,162,639 There is an internal or subcortical boundary zone in the corona radiata and centrum semiovale, lateral and/or above the lateral ventricle. This lies between the supply of the lenticulostriate perforating branches from the MCA trunk, and the medullary perforating arteries which arise from the cortical branches of the MCA and the anterior and, perhaps, posterior cerebral arteries. Infarction can occur within this internal boundary zone, usually causing a lacunar or partial anterior circulation syndrome, in association with severe carotid disease and sometimes an obvious haemodynamic precipitating cause.640 The diagnosis of ‘low flow’ as the cause of ischaemic strokes and TIAs It would be simplistic to presume that all ischaemic strokes are caused by acute arterial occlusion. However, the definitive separation of stroke resulting from ‘low flow’ from acute arterial occlusion is far from easy. It is probably best inferred from the circumstances surrounding the onset of the symptoms, and to some extent by their nature, and not very much from either the neurological signs or the site of any visible infarct on brain imaging. Unfortunately, any clinical guideance to ‘low flow’ ischaemic episodes cannot be validated against a ‘gold standard’ because at present there is none. Most ischaemic strokes and TIAs occur ‘out of the blue’ with no precipitating activity. However, on the basis of a

295

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 296

296

Chapter 6 What caused this transient or persisting ischaemic event?

number of convincing case reports, a fall in cerebral perfusion pressure as the cause – mostly resulting from a fall in systemic blood pressure – should be suspected if the symptoms start under certain circumstances: • on standing or sitting up quickly, even if postural hypotension cannot be demonstrated in the clinic; • immediately after a heavy meal; • in very hot weather; • after a hot bath or warming the face; • with exercise, coughing or hyperventilation; • during a Valsalva manoeuvre, but paradoxical embolism is another possibility; • during a clinically obvious episode of cardiac dysrhythmia (chest pain, palpitations, etc.), but embolism from the heart is also possible; • during operative hypotension (section 7.18); or • if the patient has recently been started on or increased the dose of any drug likely to cause hypotension, such as calcium blockers or vasodilators. In addition, there is usually very obvious evidence of severe arterial disease in the neck, i.e. bruits and/or absent pulsations.641–653 Transient ischaemic attacks (TIAs) caused by low flow may be atypical ‘limb shaking TIAs’ and tend to develop over minutes rather than seconds. These consist of stereotyped jerking and shaking of one arm and/or leg contralateral to the cerebral ischaemia and so are easily confused with focal motor seizures (section 3.3.4), or there is monocular or binocular visual blurring, dimming,

.. . .................................... Variable vascular supply ........................... or ‘boundary-zones’

fragmentation or bleaching, often only in bright light (section 3.3.6). Movement disorders including chorea have also been reported but involvement of the face is thought not to occur, perhaps because the blood flow to the facial motor cortex is not significantly compromised. ‘Limb shaking TIAs’ typically occur on standing up and may be abolished by a reduction in any hypertensive therapy, or carotid surgery, since they are frequently associated with severe occlusive carotid disease. There may be additional non-focal features, such as faintness, mental vagueness or even loss of consciousness.654–659 Low-flow ischaemic oculopathy is discussed in section 3.3.6. It is very important to note that boundary-zone infarction on brain imaging (or at postmortem) is not necessarily caused by low cerebral blood flow (without acute arterial occlusion) but this assumption has bedevilled much of the literature. The brain CT/MRI-defined site and size of any visible recent infarction is not an accurate way to diagnose a low-flow ischaemic stroke. This is, first, because some boundary-zone infarcts result from embolism.638,660,661 Second, there is much variation between individuals in where the boundary zones are, and they may even change with time in the same individual in response to changes in peripheral vascular resistance (Fig. 6.21). Third, however boundary-zone infarcts are defined on imaging, there is little difference between them and territorial presumed-embolic infarcts in patient demographic characteristics, vascular risk

Boundary of deep and superficial arteries

Fig. 6.21 The anterior and posterior boundary zones between the territories of the middle, anterior and posterior cerebral arteries. The maximum extent of these variable zones is shown on CT templates (see also Figs 4.11, 4.17, 4.23 and 4.25).

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 297

6.7 From symptoms, signs and clinical syndrome to cause

factors and even in the prevalence of arterial disease in the neck severe enough to cause low flow, which is far more often assumed than measured.662–665 However, there have not been many comparative studies, definitions of boundary zones vary and the numbers of patients have been small. On balance, although some boundary-zone infarcts may have a haemodynamic (i.e. low flow) cause, many others could be caused by embolism or acute occlusive thrombosis. After all, any arterial territory has a terminal zone which forms a boundary with adjacent arterial territories and which is probably particularly vulnerable to ischaemia. There is no reason to suppose that this zone is more susceptible to ischaemia caused by low flow without acute arterial occlusion, than as a result of acute arterial occlusion. Another possibility is that boundaryzone infarcts are caused by a combination of embolization to the margins of the territorial supply of a cerebral artery, as well as insufficient perfusion pressure to clear the emboli because of severe arterial disease in the neck, or operative hypotension.661 Clearly, in view of the diagnostic difficulties, it is quite conceivable that low flow is a more frequent cause of cerebral ischaemia, or less frequent, than currently believed. The exact sites of the boundary zones between the territories of supply of the major cerebral arteries are so variable between, and even within, individuals that the diagnosis of infarction in a boundary zone based on CT/MRI alone is all but impossible. Boundary-zone infarction can be caused by acute arterial obstruction, while low blood flow does not necessarily cause infarction only within boundary zones. Implications for treatment Being certain that an ischaemic episode is caused either by low flow alone, or by acute arterial obstruction, seldom really matters. It makes very little difference if ischaemic stroke or TIA caused by low flow is recognized as such because unless the precipitating factor(s) can be avoided or reversed (particularly over-treatment of hypertension), the management is exactly the same as for presumed embolic causes of ischaemia, i.e. antithrombotic drugs, statins, careful blood pressure lowering, management of any other causal vascular risk factors and surgical relief of any obstruction to blood flow if it is practical and safe to do so (Chapter 16). However, there is certainly a case for less aggressive treatment of hypertension if there is good evidence of low flow symptoms. And one must acknowledge that in a patient with or without known severe arterial disease in the neck, an ischaemic episode may occasionally be caused by low

..

..

flow rather than embolism or some other cause of acute arterial obstruction, and even that different episodes at different times in the same patient can be caused by different mechanisms.

6.7.6 Clues from the history The vast majority of transient ischaemic attacks (TIAs) and ischaemic strokes start suddenly, without any obvious provocation, and there are few if any symptoms other than those of a focal neurological or ocular deficit. Sometimes there can be clues to the cause in the history (Table 6.9), as well as to whether the patient has had a stroke or TIA in the first place (Chapter 3). These clues may require some tenacity to recognize, or perhaps just an ability to take a history instead of rushing to order lots of tests. There will be no clues from the history if no one bothers to take a history in the rush to organize a brain scan. Gradual onset Gradual onset of ischaemic stroke or TIA over hours or days, rather than seconds or minutes, is unusual but is becoming more recognized now that strokes are being seen much earlier (section 3.3.8). If the onset is gradual, and ischaemic stroke or TIA is not likely to be caused by low flow (section 6.7.5) or migraine (section 7.8), then the diagnosis should be considered particularly carefully and a structural intracranial (or ocular) lesion looked for again, or for the first time if brain imaging has not already been carried out (e.g. intracranial tumour, chronic subdural haematoma, cerebral abscess, section 3.4.4). Under the age of 50 years, multiple sclerosis should also be considered (section 3.4.10). However, a priori, focal neurological deficits which develop over hours, and even over 1 or 2 days, in elderly patients are still more likely to have a vascular than a non-vascular cause because in them vascular causes are so much more common than conditions such as brain tumours. It is only when progression occurs over a longer period that the likelihood of a non-vascular cause (such as chronic subdural haematoma) starts to rise. Precipitating factors The exact activity and time of onset may both be important (section 3.3.9). Anything to suggest a drop in cerebral perfusion or blood pressure may be relevant (section 6.7.5), as is any operative procedure (section 7.18).

297

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 298

298

Chapter 6 What caused this transient or persisting ischaemic event? Table 6.9 Important clues from the history which may suggest the cause of an ischaemic stroke or transient ischaemic attack, or that the diagnosis of cerebrovascular disease should be reconsidered. Gradual onset Low cerebral blood flow without acute occlusion (section 6.7.5) Migraine (section 7.8) Structural intracranial lesion (section 3.4.4) Multiple sclerosis (section 3.4.10) Precipitating factors Suspected systemic hypotension or low cerebral perfusion pressure (standing up or sitting up quickly, heavy meal, hot weather, hot bath, warming the face, exercise, coughing, hyperventilation, chest pain or palpitations, starting or changing blood pressure-lowering drugs) (section 6.7.5) Pregnancy/puerperium (section 7.14) Surgery (section 7.18) Head-turning (section 7.1.5) Hypoglycaemia (section 7.16) Valsalva manoeuvre (paradoxical embolism, section 6.5.12; or low flow, section 6.7.5) Recent headache Carotid/vertebral dissection (section 7.2.1) Migrainous stroke/transient ischaemic attack (section 7.8.1) Intracranial venous thrombosis (section 7.21.2) Giant-cell arteritis (or other inflammatory vascular disorders) (section 7.3.1) Structural intracranial lesion (section 3.4.4) Epileptic seizures Intracranial venous thrombosis (section 7.21.2) Mitochondrial diseases (section 7.19) Non-vascular intracranial lesion (section 3.4.4) Malaise Inflammatory arterial disorders (section 7.3) Infective endocarditis (section 6.5.9) Cardiac myxoma (section 6.5.13) Cancer (section 7.12) Thrombotic thrombocytopenic purpura (section 7.9.3) Sarcoidosis (section 7.3.16) Chest pain Myocardial infarction (section 7.10) Aortic dissection (section 7.2.3) Paradoxical embolism (sections 6.5.12)

Head-turning is an occasional cause (section 7.1.5). Recurrent attacks first thing in the morning or during exercise suggest hypoglycaemia, which is easy to think of in a diabetic patient on hypoglycaemic drugs but more difficult if there is a less obvious cause of hypoglycaemia, such as the very rare insulinoma or drugs such as pentamidine (sections 3.4.5 and 7.16). Onset during a Valsalva manoeuvre (e.g. lifting a heavy object) suggests a low flow ischaemic stroke (section 6.7.5) or paradoxical embolism (section 6.5.12), and so sets off a search for

Non-stroke vascular disease or vascular risk factors Heart disease (section 6.6.19) Claudication (section 6.6.19) Hypertension (section 6.6.3) Smoking (section 6.6.4) Drugs Oral contraceptives (section 7.13.1) Oestrogens in men (section 7.13) Blood pressure-lowering/vasodilators (section 6.7.5) Hypoglycaemic drugs (section 7.16) Cocaine (section 7.15.1) Amphetamines (section 7.15.1) Ephedrine (section 7.15.1) Phenylpropanolamine (section 7.15.1) ‘Ecstasy’ (section 7.15.1) Anti-inflammatory drugs (section 7.15.2) Antipsychotic drugs (section 7.15.3) Deoxycoformycin (section 7.3.21) Allopurinol (section 7.3.21) l-asparaginase (section 7.12) Injury Chronic subdural haematoma (section 3.4.4) Vertebral/carotid artery dissection (section 7.2.1) Cerebral air embolism (section 7.1.7) Fat embolism (section 7.1.8) Self-audible bruits Internal carotid artery stenosis (distal) (section 6.7.6) Dural arteriovenous fistula (section 8.2.8) Glomus tumour Caroticocavernous fistula (section 8.2.14) Raised intracranial pressure Intracranial venous thrombosis (section 7.21.2) Past medical history Inflammatory bowel disease (section 7.17) Coeliac disease (section 7.17) Homocystinuria (section 7.20.2) Cancer (section 7.12) Irradiation of the head or neck (section 7.12) Recurrent deep venous thrombosis (sections 7.3.4 and 7.9.11) Recurrent miscarriages (section 7.3.4) Recent surgery/long distance travel (section 6.5.12) Family history (Table 6.10)

deep venous thrombosis if there is evidence on echocardiography of a patent foramen ovale (section 6.9.3). Headache Headache at around the onset of ischaemic stroke or TIA occurs in about 25% of patients, is usually mild and, if localized at all, tends to be related to the position of the brain/eye lesion (sections 3.3.10 and 11.9). It is more common with vertebrobasilar than carotid

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 299

6.7 From symptoms, signs and clinical syndrome to cause

Table 6.10 Causes of familial stroke (including intracranial haemorrhage) and transient ischaemic attack. Connective tissue disorders Ehlers–Danlos syndrome (section 7.20.7) Pseudoxanthoma elasticum (section 7.20.7) Marfan syndrome (section 7.20.7) Fibromuscular dysplasia (section 7.4.1) Familial mitral leaflet prolapse (section 6.5.7) Haematological disorders Sickle cell disease/trait (section 7.9.8) Antithrombin III deficiency (section 7.9.11) Protein C deficiency (section 7.9.11) Protein S deficiency (section 7.9.11) Plasminogen abnormality/deficiency (section 7.9.11) Haemophilia and other inherited coagulation factor deficiencies (section 8.4.4) Others Familial hypercholesterolaemia Neurofibromatosis (section 7.20.5) Homocystinuria (section 7.20.2) Fabry’s disease (section 7.20.3) Tuberous sclerosis (section 7.20.4) Dutch and Icelandic cerebral amyloid angiopathy (section 8.2.2) Migraine (section 7.8) Familial cardiac myxoma (section 6.5.13) Familial cardiomyopathies (section 6.5.11) Mitochondrial diseases (section 7.19) CADASIL (section 7.20.1) Sneddon syndrome (section 7.3.5) Arteriovenous malformations (section 8.2.4) Cavernous malformations (section 8.2.5) Intracranial saccular aneurysms (section 8.2.3) CADASIL, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy.

distribution ischaemia, and less common with lacunar ischaemia.666–671 Severe pain unilaterally in the head, face, neck or eye at around or before the time of stroke onset is highly suggestive of carotid dissection, while vertebral dissection tends to cause unilateral or sometimes bilateral occipital pain (section 7.2.1). Migrainous stroke may be accompanied by headache (section 7.8.1) and patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoencephalo-pathy (CADASIL) usually have a history of migraine (section 7.20.1). In the context of the differential diagnosis of TIAs, migraine should be fairly obvious, unless there is no headache (section 3.4.1). It is important to note that vertebrobasilar ischaemia may cause similar symptoms to migraine with headache, gradual onset of focal neurological symptoms and visual disturbance. Although intracranial venous thrombosis usually causes either intracranial hypertension alone or a subacute

..

..

encephalopathy, a focal onset does occur and headache (occurring in around 75% of patients) can be a clue (section 7.21.2). Stroke or TIA in the context of a patient who has had a headache for days or weeks previously must raise the possibility of giant-cell arteritis and other inflammatory vascular disorders (section 7.3.1). Pain in the jaw muscles with chewing, which resolves with rest, strongly suggests claudication, which is caused by external carotid artery disease as a result of giant-cell arteritis far more often than atherothrombosis. Epileptic seizures Epileptic seizures, partial or generalized, within hours of stroke onset are distinctly unusual in adults (about 5%) and should lead to a reconsideration of non-stroke brain pathologies (section 3.4.4). They are rather more common in childhood stroke. They are more likely with haemorrhagic than ischaemic strokes and if the infarct is extensive and involves the cerebral cortex (section 11.8).672–676 They are also likely with venous infarction (up to 40%) (section 7.21.12) and mitochondrial disorders (section 7.19). Partial motor seizures can be confused with limb-shaking TIAs, but the former are more clonic and the jerking spreads in a typical Jacksonian way from one body part to another and the latter are supposed never to involve the face (sections 3.3.4 and 6.7.5). Very rarely, transient focal ischaemia seems to cause just partial epileptic seizures, but proving a causal relationship is seldom possible.677 Interestingly, onset of idiopathic seizures late in life is a powerful independent predictor of subsequent stroke.678 Because the diagnosis of stroke may be wrong if a tumour on CT is misinterpreted as an infarct (contrast enhancement can look very similar in both, section 5.4.2), partial seizures after a ‘stroke’ should always be an indication to re-examine the diagnosis and reassess the imaging. Also, seizures in the presence of a history of a few days of malaise, headache and fever should suggest encephalitis and the need for an electroencephalogram to show bilateral diffuse rather than unilateral focal slow waves, and cerebrospinal fluid examination (raised white cell count, but this can also occur in stroke, section 6.8.11). Malaise Stroke in the context of a patient who has been generally unwell for days, weeks or months suggests an inflammatory arterial disorder, particularly giant-cell arteritis (section 7.3.1), infective endocarditis (section 6.5.9), cardiac myxoma (section 6.5.13), cancer (section 7.12), thrombotic thrombocytopenic purpura (section 7.9.3) or even sarcoidosis (section 7.3.16).

299

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 300

300

Chapter 6 What caused this transient or persisting ischaemic event?

Chest pain Chest pain may suggest a recent myocardial infarction with complicating stroke (section 7.10); aortic dissection, particularly if the pain is also interscapular (section 7.2.3); and pleuritic pain suggests pulmonary embolism and the possibility of paradoxical embolism (section 6.5.12). Vascular risk factors Vascular risk factors (section 6.6) and diseases should be sought. It is most unusual for an ischaemic stroke or TIA to occur in someone with no vascular risk factors, unless they are very old, or are young with some unusual cause of stroke (Table 6.3). Heart disease of any sort may be relevant (source of embolism to the brain, dysrhythmias causing low-flow ischaemia, etc.) and cardiac symptoms should be specifically elicited in the history: angina, shortness of breath, palpitations, and so on. Drugs and drug users Drugs may well be relevant: oral contraceptives and hormone replacement therapy in women and oestrogens in men (section 7.13); anything which lowers the blood pressure (section 6.7.5); hypoglycaemic agents (section 7.16); and illicit drugs (section 7.15.1). Injury Any injury in the days and weeks before ischaemic stroke or TIA onset is crucial information. A head injury might have caused a chronic subdural haematoma (highly unlikely if more than 3 months previously) although this should have been considered earlier at the stroke vs non-stroke stage (section 3.4.4). Of possible relevance is an injury to the neck in the hours, days or even few weeks before onset, because this may cause carotid or vertebral dissection (section 7.2.1). After long bone fracture, fat embolism may cause a generalized encephalopathy, but occasionally there are additional focal features (section 7.1.8).679–682 It is therefore essential to ask about any injury, strangulation, car crash, unusual yoga exercises, neck manipulation and so forth, in any unexplained stroke (Table 6.2). Self-audible bruits Pulsatile self-audible bruits are rare. They can be differentiated from tinnitus because they are in time with the pulse. They may be audible to the examiner on auscultation of the neck, orbit or cranium. They are unlikely

to be caused by carotid bifurcation atherothrombosis because the source of the sound is too far from the ear. They are much more likely to indicate distal internal carotid artery stenosis (due to dissection or, rarely, atherothrombosis), dural arteriovenous fistula near the petrous temporal bone, glomus tumour, caroticocavernous fistula, intracranial venous thrombosis, symptomatic and idiopathic intracranial hypertension, a loop in the internal carotid artery, or just heightened awareness of one’s own pulse.683,684 Past medical history Past medical history of inflammatory bowel disease (section 7.17), coeliac disease (section 7.17), irradiation of the head and neck (section 7.12), cancer (section 7.12) or even homocystinuria (section 7.20.2) may be important. Recurrent deep venous thrombosis (DVT) suggests thrombophilia (section 7.9.11), particularly if there is a family history, or the antiphospholipid syndrome (section 7.3.4). Recurrent miscarriage is another feature of the antiphospholipid syndrome. Any reason for a recent DVT (e.g. a long cramped journey, admission to hospital for an acute medical disorder, or surgery) should raise the question of paradoxical embolism (sections 6.5.12 and 6.9.3). If a patient has, or has had, deep venous thrombosis in the legs, then consider paradoxical embolism to the brain, a familial clotting factor problem or the antiphospholipid syndrome. Previous strokes and/or transient ischaemic attacks Previous strokes and/or TIAs in different vascular territories are more likely with a proximal embolic source in the heart, or arch of the aorta, than with a single arterial lesion in the neck or head. Attacks going back months or more make some causes unlikely (e.g. infective endocarditis, arterial dissection). Family history There are several rare familial conditions that may be complicated by ischaemic stroke and TIAs (Table 6.10). There is also increasing interest in complex genetic disorders thought to be caused by multiple gene interactions, presumably influenced by environmental factors (section 6.6.15).535,685–687 However, family history of stroke is only a modest risk factor for ischaemic stroke.687 Moreover, much of the association appears to be secondary to hereditability of risk factors for stroke such as hypertension and diabetes.687,688 On the other hand, these

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 301

6.7 From symptoms, signs and clinical syndrome to cause

same risk factors are clearly influenced by the environment (for example, diets rich in fat and in salt tend to raise the plasma cholesterol and blood pressure, respectively). Just how easy it will be to separate out shared genes from shared environment in a disease as common as stroke remains to be seen. Disentangling the interactions and working out the pathway from genotype to phenotype will be a monumental task (section 6.6.15). Whatever the mechanism(s), one can at least reassure patients with TIA or stroke that a family history of stroke is associated with little or no increase in the risk of a future stroke.689,690

• remember that comorbidity, such as pneumonia, sedative drugs, infection and hypoglycaemia, may all make the neurological deficit seem worse than it really is (section 11.4). If the neurological deficit is mild and yet the patient is drowsy, then consider chronic subdural haematoma, cerebral vasculitis, non-bacterial thrombotic endocarditis, intracranial venous thrombosis, mitochondrial disorders, thrombotic thrombocytopenic purpura, sedative drugs, hypoglycaemia, familial hemiplegic migraine and comorbidity, such as pneumonia or other infections.

6.7.7 Clues from the examination Eyes Neurological examination Neurological examination is primarily to localize the brain lesion; of course, in patients with transient ischaemic attacks (TIAs), or those seen some days after a minor stroke, there will probably be no signs at all (section 3.3.1). Occasionally, however, there may be a clue to the cause. A Horner syndrome ipsilateral to a carotid distribution infarct (i.e. not as the result of a brainstem stroke, where it might be expected) suggests dissection of the internal carotid artery (ICA) or sometimes acute atherothrombotic carotid occlusion (section 7.2.1). Lower cranial nerve lesions ipsilateral to a hemispheric cerebral infarct can also occur in carotid dissection and, like Horner syndrome, are caused by stretching and bulging of the arterial wall in relation to the affected nerves, or ischaemia. Ocular ischaemia, as well as third, fourth and sixth cranial nerve palsies – sometimes with orbital pain – has been described ipsilateral to acute ICA occlusion and stenosis, presumably caused by ischaemia of the nerve trunks.691,692 In a total anterior circulation infarct or brainstem stroke some drowsiness is expected, but with more restricted infarcts, consciousness is normal. Therefore if consciousness is impaired and yet the ‘stroke’ itself seems mild, it is important to: • reconsider the differential diagnosis (particularly chronic subdural haematoma) (section 3.4.4); • consider the diffuse encephalopathic disorders which have focal features and which may masquerade as stroke, e.g. cerebral vasculitis of some sort (section 7.3), endocarditis (sections 6.5.9 and 6.5.10), intracranial venous thrombosis (section 7.21), mitochondrial disorders (section 7.19), thrombotic thrombocytopenic purpura (section 7.9.3), familial hemiplegic migraine693 and Hashimoto’s encephalitis;694

..

..

The eyes may provide general clues to the cause of a stroke (e.g. diabetic or hypertensive retinopathy), or may reveal papilloedema which would make the diagnosis of ischaemic stroke, or even intracerebral haemorrhage, most unlikely. In addition, it is worth searching thoroughly for evidence of emboli which are very often completely asymptomatic695,696 (section 3.3.6): • Fibrin–platelet emboli are dull greyish-white amorphous plugs but are rarely observed, perhaps because they move through the retinal circulation and disperse; they suggest embolism from the heart or proximal sources of atherothrombosis. • Cholesterol emboli quite often stick at arteriolar branching points, usually without obstructing the blood flow, and appear as glittering orange or yellow bodies reflecting the ophthalmoscope light; obviously these strongly suggest embolization from proximal atheromatous plaques, but they are often asymptomatic. • ‘Calcific’ retinal emboli appear as solid, white and non-reflective bodies and tend to lodge near the edge of the optic disc; they suggest embolism from aortic or mitral valve calcification. Localized areas of periarteriolar sheathing, seen as opaque white obliteration of segments of the retinal arterioles, suggest embolism, usually cholesterol, in the past. Roth spots in the retina are very suggestive of infective endocarditis (section 6.5.9). Dislocated lenses should suggest Marfan syndrome (section 7.20.7) or homocystinuria (section 7.20.2); angioid streaks in the retina suggest pseudoxanthoma elasticum (section 7.20.7); and in hyperviscosity syndromes there is a characteristic retinopathy (section 7.9.10). Dilated episcleral vessels are a clue to abnormal anastomoses between branches of the external carotid artery (ECA) and orbital branches of the internal carotid artery (ICA), distal to severe ICA disease.697 With very

301

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 302

302

Chapter 6 What caused this transient or persisting ischaemic event?

severe ICA disease, usually accompanied by severe disease of the ipsilateral ECA, the eye may occasionally become so ischaemic that venous stasis retinopathy develops, although arterial disease does not invariably underlie this condition698 (section 3.3.6). Haemorrhages are scattered around the retina with microaneurysms, and the retinal veins are dilated and irregular. The retinal blood flow is extremely impaired, as demonstrated by lightly compressing the eye with one finger while observing the fundus and noting collapse of the central retinal artery. With more extreme ischaemia, ischaemic oculopathy may develop with impaired visual acuity, eye pain, rubeosis of the iris (dilated blood vessels), fixed dilated pupil, ‘low pressure’ glaucoma, cataract and corneal oedema.699,700 Pre-existing raised intraocular pressure, i.e. glaucoma, makes the eye more susceptible to low blood flow and ischaemia as a result. Arterial pulses It is always worth feeling both radial pulses simultaneously. Any inequality in timing or volume suggests subclavian or innominate stenosis or occlusion, and this is further supported if there is an ipsilateral supraclavicular bruit or lower blood pressure in the arm with the weak or delayed pulse. An elderly patient presented with a sudden left-sided hemiparesis and no other symptoms. She had right carotid and supraclavicular bruits. Brain CT was normal. Three days later the duplex examination showed narrowing of the right common carotid artery which appeared to be caused by dissection of the aortic arch. Only then did she admit to some mild chest pain before the stroke. Unequal pulses and blood pressures were found in her upper limbs, and chest CT confirmed the aortic dissection. The lessons are that any pain in or around the chest may be relevant and should have been more thoroughly sought, and in all stroke patients both radial pulses must be felt routinely before, not after, arterial imaging. Normally, the internal carotid artery pulse is too deep and rostral to be felt in the neck. Therefore, any loss of the ‘carotid’ pulsation reflects common carotid artery (CCA) or innominate occlusion or severe stenosis, both rather rare situations or, perhaps more likely, the artery is too deep to be felt or the neck too thick. The arterial pulse felt in the neck comes from the common, not the internal, carotid artery. The superficial temporal pulses should be easily felt and symmetrical. If there is unilateral absence or delay,

this suggests external carotid artery (ECA) or CCA disease. Tenderness of any of the branches of the ECA (occipital, facial, superficial temporal) points towards giant-cell arteritis. Tenderness of the carotid artery in the neck (i.e. the CCA) can occur in acute carotid occlusion but is more likely to be a sign of dissection, or possibly arteritis. Absence of several neck and arm pulses in a young person suggests Takayasu’s arteritis (section 7.3.2). Other causes of widespread disease of the aortic arch are atheroma, giant-cell arteritis, syphilis, subintimal fibrosis, arterial dissection and trauma.264,701,702 Delayed or absent leg pulses suggest coarctation of the aorta or, much more commonly, peripheral vascular disease (PVD) which is very common in patients with TIA and ischaemic stroke (Table 6.3) and may need treating in its own right. Furthermore, PVD is an important predictor of future serious vascular events (Table 16.4). The state of the femoral artery is important to assess before cerebral angiography via the femoral route (section 6.8.5) and, if after angiography the leg pulses disappear, then it was a complication of the angiography. Obviously, any evidence of systemic embolism would direct the search towards a source of emboli in the heart (section 6.5). Finally, while the hand is on the abdomen, aortic aneurysm should be considered while searching for any masses or hepatosplenomegaly. Although the prevalence of aortic aneurysm in these stroke/TIA patients is unknown, it could well be quite high, particularly in men and if the patient has carotid stenosis (section 6.6.19). Diagnosis is important because of the benefits of surgery in patients with larger aneurysms.606 Cervical bruits Listening to the neck is a favourite occupation for inquisitive physicians, and acquisitive surgeons, and can lead to some useful information (Fig. 6.22). A localized bruit, occasionally palpable, over the carotid bifurcation (i.e. high up under the jaw) is predictive of some degree of carotid stenosis, but very tight stenosis (or occlusion) may not cause a bruit at all (Fig. 6.23) (Table 6.11). External carotid stenosis can also cause a bruit in the same place. An innocent carotid bruit is more common in women,703 probably due to sex differences in carotid bifurcation anatomy.704 Bruits transmitted from the heart become attenuated as one listens further up the neck towards the angle of the jaw, thyroid bruits are bilateral and more obviously over the gland, a hyperdynamic circulation tends to cause a diffuse bruit, and venous hums are more continuous and roaring, and are obliterated by light pressure over

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 303

6.7 From symptoms, signs and clinical syndrome to cause Table 6.11 The source of neck bruits.

Bruit from carotid bifurcation

Bruit from subclavian artery, or vertebral artery origin

Transmitted bruit from the heart

Fig. 6.22 The sites of various cervical bruits. Note that a bruit arising from the carotid bifurcation is high up under the angle of the jaw. Localized supraclavicular bruits are caused either by subclavian or vertebral origin artery stenosis.

Carotid bifurcation arterial bruit Internal carotid artery origin stenosis External carotid artery origin stenosis Supraclavicular arterial bruit Subclavian artery stenosis Vertebral artery origin stenosis Can be normal in young adults Diffuse neck bruit Thyrotoxicosis Hyperdynamic circulation (pregnancy, anaemia, fever, haemodialysis) Transmitted bruit from the heart and great vessels Aortic stenosis/regurgitation Mitral regurgitation Patent ductus arteriosus Coarctation of the aorta Venous hum

Carotid bruits are neither sufficiently specific nor sensitive to diagnose carotid stenosis severe enough to consider surgery. Arterial imaging is needed. Physicians and surgeons desperate to use their stethoscopes would do better to measure the blood pressure and listen to the heart. Cardiac examination

No bruit

Carotid bruit

Carotid bruit (%)

100 80 60 40 20 0

Normal

1–24

25–49

50–74

75–99 Occluded

Diameter stenosis of symptomatic carotid artery (%) Fig. 6.23 The percentage of patients with a localized bruit over the symptomatic carotid bifurcation for various degrees of stenosis as estimated (using the European Carotid Surgery Trial method) from 298 carotid angiograms. (Adapted with permission from Hankey and Warlow, 1990.790)

the ipsilateral jugular vein.593 An arterial bruit in the supraclavicular fossa suggests either subclavian or proximal vertebral arterial disease, but a transmitted bruit from aortic stenosis must also be considered. Normal young adults quite often have a short supraclavicular bruit; the reason is unknown.

..

..

Cardiac examination is important, particularly to look for any cardiac source of embolism (section 6.5). If physicians feel under-confident about their cardiological abilities, then they should get properly trained or a cardiologist will have to be consulted. Atrial fibrillation (AF) will already have been suspected from the radial pulse; left ventricular hypertrophy suggests hypertension or aortic stenosis, and most major cardiac sources of embolism are fairly obvious clinically (e.g. AF, mitral stenosis, prosthetic heart valve). Fever Fever is distinctly unusual in the first few hours after stroke onset. Any raised temperature at this time must therefore be taken seriously and endocarditis or other infections, inflammatory vascular disorders, deep venous thrombosis or cardiac myxoma considered. Later on, fever is quite common and usually reflects some complication of the stroke (section 11.12). Skin and nails The skin and nails occasionally provide clues to the cause of ischaemic stroke or transient ischaemic attack (Table 6.12).

303

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 304

304

Chapter 6 What caused this transient or persisting ischaemic event?

Finger-clubbing

Right-to-left intracardiac shunt (section 6.5.12) Cancer (section 7.12) Pulmonary arteriovenous malformation (section 6.5.12) Infective endocarditis (section 6.5.9) Inflammatory bowel disease (section 7.17) Splinter haemorrhages Infective endocarditis (section 6.5.9) Cholesterol embolization syndrome (section 7.7) Vasculitis (section 7.3) Scleroderma Systemic sclerosis (section 7.3.13) Livedo reticularis Sneddon syndrome (section 7.3.5) Systemic lupus erythematosus (section 7.3.4) Polyarteritis nodosa (section 7.3.6) Cholesterol embolization syndrome (section 7.7) Lax skin Ehlers–Danlos syndrome (section 7.20.7) Pseudoxanthoma elasticum (section 7.20.7) Skin colour Anaemia (section 7.9.7) Polycythaemia (section 7.9.1) Cyanosis (right-to-left intracardiac shunt, pulmonary arteriovenous malformation) (section 6.5.12) Porcelain-white papules/scars Kohlmeier–Degos disease (section 7.3.15) Skin scars Ehlers–Danlos syndrome (section 7.20.7) Petechiae/purpura/bruising Thrombotic thrombocytopenic purpura (section 7.9.3) Fat embolism (section 7.1.8) Cholesterol embolization syndrome (section 7.7) Ehlers–Danlos syndrome (section 7.20.7) Orogenital ulceration Behçet’s disease (section 7.3.11) Rash Fabry’s disease (section 7.20.3) Systemic lupus erythematosus (section 7.3.3) Tuberous sclerosis (section 7.20.4) Epidermal naevi Epidermal naevus syndrome Café-au-lait patches Neurofibromatosis (section 7.20.5) Thrombosed superficial veins, Intravenous drug users (section 7.15.1) needle marks

Getting to the bottom of the cause of an ischaemic stroke or transient ischaemic attack requires much more than just neurological skills. Stroke medicine, like neurology, is part of general internal medicine. It is important therefore that doctors looking after stroke patients have a good general internal medical training.

6.8 Investigation

Investigations are mainly to help unravel the pathological type of stroke (ischaemic stroke vs intracerebral haemorrhage vs subarachnoid haemorrhage, as discussed in Chapter 5) and then to determine the cause of the cerebral ischaemia (or intracranial haemorrhage; Chapters 8 and 9), particularly a cause that will influence

Table 6.12 Clues to the cause of ischaemic stroke/transient ischaemic attack from examination of the skin and nails.

immediate treatment or long-term management. Investigation may also provide important prognostic information, e.g. severe carotid stenosis on ultrasound (section 16.11.8). In addition, the many patients who also have angina or other cardiac symptoms, claudication or suspected aortic aneurysm may well need specific investigations directed at these problems with a view to appropriate treatment. Ideally, any investigation should be practical, feasible, accurate, safe, non-invasive, inexpensive and, most importantly, informative in the sense that the result (positive or negative, high or low, etc.) will influence patient management and outcome. Idle curiosity or financial gain are not good reasons to order any test.

6.8.1 Routine investigations Although there are no absolute rules, all ischaemic stroke/transient ischaemic attack (TIA) patients, unless

..

..

9781405127660_4_006.qxd 10/13/07 2:11 PM Page 305

6.8 Investigation

they are already heavily dependent or institutionalized, or have already been recently investigated for a previous event or some other problem, should have basic noninvasive first-line investigations within a few hours of presentation. None of these necessarily require hospital admission, although brain imaging does require attendance at hospital (Table 6.13). The chance of picking up a relevant abnormality (yield) may be very low for some tests – e.g. full blood count and erythrocyte sedimentation rate (ESR) – but these are cheap and the consequences of missing a treatable disorder, such as giant-cell arteritis, are serious. There is a higher chance of picking up a treatable abnormality with the blood glucose, urine analysis and electrocardiogram (ECG). Depending on the definition, many or even most patients are hypercholesterolaemic, and how this should be acted on is discussed in section 16.4; immediately after stroke, but probably not TIA, there is a transient fall in plasma cholesterol, which will underestimate the usual level.705,706 Brain imaging is discussed in sections 5.4, 5.5 and 6.8.3. If patients have the basic tests and if the results are read, written in the records and acted upon, Table 6.13 First-line investigations for ischaemic stroke/transient ischaemic attack.

Investigation Full blood count

this may well do more good than the inappropriate ordering of a huge range of further tests while missing some crucial clue from one of the routine investigations (such as an ESR of 100 mm in the first hour). All patients should have a full blood count, erythrocyte sedimentation rate, plasma glucose, urea, electrolytes and cholesterol, urine analysis, and electrocardiogram. Most should also have a CT and/or MR brain scan.

6.8.2 Second-line investigations for selected patients Second-line investigations (Table 6.14) are usually more costly, invasive and/or dangerous, so they must be targeted on patients most likely to gain from a useful change in management as a consequence of the test result. The likelihood of a relevant result depends on the selection of patients for the investigation – a balance must be struck between over-investigation (inconvenience, high cost, possibly high-risk, low yield, false-positive results

Disorders suggested

Anaemia, polycythaemia, leukaemia, thrombocythaemia, heparin-induced thrombocytopenia with thrombosis, infections Erythrocyte Vasculitis, infections, cardiac myxoma, sedimentation rate hyperviscosity, cholesterol embolization syndrome, non-bacterial thrombotic endocarditis Plasma glucose Diabetes mellitus, hypoglycaemia Urea and electrolytes Diuretic-induced hypokalaemia, renal failure, hyponatraemia Plasma cholesterol Hypercholesterolaemia Syphilis serology Syphilis, anticardiolipin syndrome Urinalysis Diabetes, renal disease, infective endocarditis, vasculitis, Fabry’s disease ECG Dysrhythmia, left-ventricular hypertrophy, silent myocardial infarction Unenhanced CT Intracerebral haemorrhage, non-vascular brain scan intracranial mimic of stroke-syndrome

Yield* (%) 1

2

5 3 45 10% >20%

in the completeness of the assessment of stroke patients following the introduction of a stroke clerking or admission form.55,56 The use of the form also makes it much easier to access the information subsequently and to identify relevant items that are missing. However, one has to acknowledge the practical difficulties of introducing disease-specific assessment forms in situations – e.g. accident and emergency departments – where patients with a wide range of medical problems are being assessed. For example, it may not be clear until the end of the assessment what the most likely diagnosis is and therefore which form should have been used. One solution might be a ‘core form’ for all patients, with supplementary sheets for specific common conditions. There is reasonable evidence that patient-specific reminders (e.g. ‘if ischaemic stroke, prescribe aspirin’), which can be included in such forms (perhaps as part of an integrated care pathway), improve adherence to management guidelines although randomized trials have not yet demonstrated improvements in patient outcomes.57,58 A clerking or admission form will improve the completeness and relevance of the initial assessment and will facilitate communication and audit, but the introduction of diagnosis-specific forms is not without its problems. Although a physician is often the first person to assess the patient, it is important to emphasize that assessment should often involve other members of the multidisciplinary team (section 10.3.5). It is often very valuable, even on the day of the stroke, to involve the nurse, the physiotherapist and a speech and language therapist. Advice on the patient’s risk of pressure sores, lifting, handling and positioning and the patient’s ability to

Fig. 10.14 ABCD scale to estimate the risk of stroke early after a transient ischaemic attack. DBP, diastolic blood pressure; SBP, systolic blood pressure.

swallow safely are all relevant to their care from the moment of hospital admission, or indeed from the moment of assessment at home. The components of the initial assessment and their use are summarized in Table 10.8. It is important to emphasize again that assessment may well be staged; it is not a ‘once only’ process, but should continue throughout the course of the patient’s recovery from stroke. Assessment is not a ‘once only’ process, but should continue throughout the course of the patient’s recovery from stroke. Health professionals involved in the initial assessment of stroke patients learn that, although the patient may be a useful source of information, other people often provide more information that is essential to planning treatment. This is particularly important when the patient, for a variety of reasons, cannot communicate. It is usually valuable to spend a little time interviewing the family, neighbours, general practitioner, ambulance technicians or nursing staff, using a telephone if necessary (Fig. 10.15). To complete an assessment it is usually valuable to talk to family, neighbours, or family doctor; essential information can often be collected by a telephone call to the appropriate person. Social environment The patient’s social environment is a very important factor in determining the overall effect that a stroke will have on an individual and his or her family. Accurate knowledge of social networks is therefore critical when

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 521

10.3 Delivering an integrated management plan Table 10.8 Information that should be collected during the initial assessment of a stroke patient, and the potential value. Type of information

Diagnosis

Cause

Prognosis

Hyper-acute treatment

Demographics History of onset Risk factors Coexisting disease Medication Social details Pre-stroke function General examination Neurological examination Investigations

+ + + +

+ + + + + +

+ + + +

+ +

+ + +

+ +

+ + + + +

+ +

+ + +

Problems

Secondary prevention +

+ + + + + + + +

+ + + + + + +

because one of the major aims of rehabilitation is to minimize handicap. For example, it may be more appropriate to put greater energy and resources into occupational therapy for a craftsman than for a schoolteacher, who might require relatively more speech therapy. The patient should, if possible, have a major role in deciding such priorities. It is so often difficult to see the real person behind the facial weakness, severe aphasia, hemiplegia and incontinence.

Fig. 10.15 The telephone is a valuable diagnostic tool.

setting longer-term goals for rehabilitation and in planning discharge from hospital. Moreover, this allows one to build up a picture of the patient as a person rather than as ‘just another stroke’ or ‘that fibrillating hemiplegic in bed six’. It is often difficult to see the real person behind the facial weakness, severe aphasia, hemiplegia and incontinence. Finding out about the patient’s prestroke life might encourage members of the team who are caring for the patient to treat him or her with more understanding and sympathy. Also, it is this background information that allows one to judge the likely effect that the individual’s disabilities will have on their role in society, i.e. the likely handicap. This is important,

..

..

Although a lot of this background information can be collected over a longer period while the patient is recovering, it can be useful early on and may be more easily collected at the initial assessment. So often, the family – who may be the only source of this sort of information – disappear within a few hours of the admission and may then not reappear to be asked the relevant questions. It is therefore vital to seize the opportunity when a patient is first admitted, or at least during the next day or two, to obtain as much information from the family and friends as possible. Clearly, this may not be regarded as a medical priority, but professionals other than doctors, in particular the nursing staff, are often well placed to collect it. However, often this is not done very well. A complete picture of the patient’s pre-stroke life will be useful when deciding how aggressively to manage a patient (e.g. neurosurgery for obstructive hydrocephalus). Because this background information is so important, but may not be available at the initial assessment, one needs to have some method of identifying which items of data are missing so they can be sought later on. The clerking or admission form, or a patient record that is shared by the different professions involved (the so-called ‘combined’ or ‘single-patient’ record), has the potential if properly

521

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 522

522

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

used to fulfil this role. Many stroke units have introduced ‘integrated care pathways’, which usually include an admission form, multidisciplinary records and guidance on how to manage common problems (section 17.14). It is just as important to know the home and social circumstances of a stroke patient for early decision making (such as the desirability of emergency operation) as for later rehabilitation and discharge from hospital. By the end of the initial assessment, which may be punctuated by giving a hyperacute treatment, one should have collected enough information to produce a diagnostic formulation, including certainty of stroke diagnosis, site and size of the brain lesion, and the likely causes. This leads on to choices about the amount of further investigation required (e.g. echocardiography or not?), and enables the physician to talk to the patient and/or the family about the likely diagnosis, prognosis and management. At this stage, it is valuable to consider, and even list, the particular problems the patient has, to ensure they have all been identified and addressed. It may also be useful to use a checklist of the most common ones that occur at this early stage (Table 10.9).

10.3.3 Identifying problems and setting goals The patient’s assessment, both initially and subsequently, should identify any major problems, and it is these that Table 10.9 Important things to think about during the first day after a patient has had a severe stroke. Maintenance of a clear airway Treatment of co-existing or underlying disease Need to review the patient’s usual medication(s) Investigation and avoidance of fever Adequacy of oxygenation Swallowing ability Hydration Nutritional status Exclusion of urinary retention Management of urinary incontinence Exclusion of fractures if the patient may have fallen Prevention of: Deep venous thrombosis Pressure ulcers Aspiration Trauma Faecal impaction Protection of a flaccid shoulder Obtaining information from and giving information to the patient and family

determine the patient’s management. Problems can occur at every level of the patient’s illness – i.e. pathology, impairment, disability and handicap (section 10.1). For example, a problem at the pathology level might be ‘diagnostic uncertainty’ or ‘raised erythrocyte sedimentation rate’, which should lead on to further investigation if appropriate; while a problem at the level of handicap might be that the patient provides the only income for a large family, who now have no money to feed themselves. Having identified a problem, one can then formulate a plan to solve or at least alleviate that problem. Thus, a problem list can be turned into an action plan for the individual patient. Some problems can be dealt with very simply (e.g. antibiotics for a urinary tract infection). These are the ‘short-loop problems’ (section 10.1.2). The goal here is simply to remove the problem. Other problems such as immobility – which are more complex, respond more slowly to therapy and may require several different types of intervention – are the ‘long-loop problems’ (section 10.1.2); here it is useful to set a long-term goal of removing or alleviating the problem, but it is also helpful to set intermediate goals that allow one to judge whether progress is being made towards the long-term goal. Why a goal-orientated approach has several advantages Setting goals allows forward planning and provides a useful focus for multidisciplinary team meetings (section 10.3.7). Intermediate goals allow members of the team to coordinate their work, assuming that goals are achieved on time, and so improve efficiency. For example, if patients are to dress the lower half of their body, they must be able to stand. If the physiotherapist can estimate when the patient will be able to stand independently, the occupational therapist can plan when to start working on dressing the lower half. Setting longer-term goals can, for example, allow advanced planning of a pre-discharge home visit and final discharge to the community, which can reduce the patient’s length of stay in hospital by the number of days or weeks needed to plan a home visit or to make any necessary adaptations to the patient’s home before discharge (e.g. stair rails). If realistic goals are set, they can then be used to help motivate patients, especially if they have been involved in choosing or setting the goal. Recovery from a stroke may be very slow, so slow that the patient and even the therapists are unable to discern any progress being made towards the long-term goal (e.g. to achieve independent mobility). If one sets and achieves intermediate goals (e.g. sitting balance), progress is more easily perceived and morale maintained. The management of patients

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 523

10.3 Delivering an integrated management plan

Fig. 10.16 The ‘road’ to recovery after a hemiplegic stroke, showing some mobility ‘milestones’: 1, sitting balance for 1 min; 2, standing for 10 s; 3, 10 steps unaided; 4, timed 10-m walk (Smith & Baer 1999).59

with stroke in hospital is often allowed to drift without direction or leadership. The clinician who is responsible for the patient waves to them on a weekly ward round, in the belief that the therapists are actively rehabilitating the patient. If the head of the multidisciplinary team maintains discipline, and encourages the setting of both intermediate and long-term goals, drift can be avoided. This discipline benefits the patients, the team members and the service as a whole. The greater efficiency reduces length of stay in hospital and allows a greater proportion of stroke patients to be treated in the stroke unit without the need for extra staff or beds. Describing goals Where the goal is simply the removal of a problem, such as a urinary tract infection, it is fairly easy to describe it and then measure progress, e.g. relief of symptoms and sterile urine on a repeat culture. Similarly, longterm goals are often easy to describe, but should take into account the patient’s need for accommodation, physical and emotional support, how they might fill their time and what role they play in society. Judgements about whether these long-term goals have been achieved are relatively straightforward. For example, if the goal is to get a patient home to live with their family, or to return to work, one does not need any complex measures of outcome. In some areas, it is even fairly easy to set intermediate goals. For example, many patients with stroke

..

..

have problems with mobility, but usually the patient will achieve certain physical milestones on the road to recovery59 (Fig. 10.16). Further improvement in mobility can be measured by recording the time it takes the patient to walk 10 m.60 Thus, it is a fairly simple process to set an intermediate goal, which the patient or carer can understand, in terms of the level of function and the date by which it should be achieved. If goals are to be reproducible, and therefore useful, a standard method for assessing attainment or not needs to be established and used. It is more complex to set goals for problems such as language or activities of daily living (ADL) than for mobility. For example, a goal for independent dressing a patient’s top half would need to take account of important factors including the clothes worn, fasteners involved and the amount of prompting allowed (section 11.32.5). One could use a score on any one of the huge number of measures of language function or ADL as intermediate goals, but these scores are not easily understood by patients and carers, or even by the professionals using them. For example, it is unlikely to mean much to a patient, or even the team members, to aim for a Barthel index score of 12 (out of a maximum of 20) within the next 2 weeks. Despite these difficulties, the team should attempt to identify problems, specify intermediate and long-term goals, and introduce some meaningful measures to determine whether progress is being made towards achieving each of them.

523

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 524

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

10

In every stroke patient, intermediate and long-term goals should be described so that progress towards them can be measured. Moreover, everyone will feel a sense of achievement when the goals are met.

10.3.4 Goal setting: a diagnostic tool Goals may also be useful in identifying new or previously unrecognized problems. If a patient is not achieving the goals that have been set, it may be due to a number of reasons, which can be divided into team factors, patient factors and carer factors.

8 Function

524

Complication

6 4 Failing to achieve goals

2

0

1

2

3

4

5

6

7

8

9

10

Time (weeks) Expected Observed

Team factors If the goals are too ambitious because of inaccurate diagnosis, inadequate assessment or uncertainty about the prognosis, then patients may fail to achieve them and this will have a detrimental effect on the patient’s and the team’s morale. If goals are too easy, then progress may be slower than is, in fact, possible. Also, goal setting must be realistic if one is to use it to coordinate care (section 10.3.3). To set realistic goals in such a way that they are more often achieved than not requires an understanding of the prognosis and of the likely effectiveness of potential interventions. We have already seen that accurate predictions of progress and outcome are difficult in individual patients (section 10.2.7), but informal judgements made by the team may be more accurate, because they are based on past experience and observation of the patient’s progress over a period of time. Another reason why a patient may not be achieving a goal might be lack of appropriate treatment. Thus, progress may be hampered by too little therapy, or by the wrong sort of therapy. However, since there is currently so little information about the optimum amount or the relative effectiveness of most interventions, it is difficult to sort this out. Team members will much more often have to modify their therapy based on their own experience rather than on evidence from properly conducted randomized controlled trials. Goals should be meaningful and challenging, but achievable. Patient and carer factors New, unrecognized medical or psychological problems (e.g. infection, recurrent stroke or depression) may not present overtly – especially in their early stages – but may develop in a less specific way, causing a patient’s progress to slow, stop, or even reverse. One can draw a parallel

Fig. 10.17 ‘Failure to thrive’ after a stroke. Deviation from the expected recovery pattern might be due to any number of factors, including recurrent stroke, infections, depression, etc.

with the concept of ‘failure to thrive’ in paediatric practice (Fig. 10.17). If a patient is failing to achieve his or her goals (or milestones), then one needs to identify the likely reason or reasons (Tables 11.5 and 11.7). When a patient is failing to achieve his or her goals (or milestones) then one must identify the reason, and if possible do something about it. Sometimes the patient or the carer may have different goals from those of the team looking after the patient. This is ‘goal mismatch’. For example, a patient who does not want to live alone but would prefer to live with his or her daughter may not achieve the level of independence expected. Therefore, when setting goals ideally one should agree them with the patient and carers – although they may be reticent about discussing such matters openly. It is also important to involve the patient, and perhaps the carer, in setting goals, to ensure that the goals are really relevant to them. Most stroke patients are retired, so that leisure activities may be particularly important to their quality of life (section 11.33.3). The patient may be less interested in a goal aiming at achieving self-care in dressing than in being able to read or do the gardening. In hospital practice, where activities of daily living (ADL) abilities often determine the length of stay, too much emphasis can be placed on ADL-related goals because of the pressures on the team to make beds available for new patients and to minimize costs by discharging the patient as early as possible. There are however important practical difficulties in involving patients in goal setting, including their cognitive,

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 525

10.3 Delivering an integrated management plan

communication and visuuospatial problems and the clinician’s lack of experience or time.61,62 Goal setting may sometimes involve just an individual professional, but ideally it should involve the rest of the team, the patient and sometimes the patient’s family.

10.3.5 The stroke team Although a physician usually has overall responsibility for the management of the patient, other members of the multidisciplinary team play an essential part. Stroke unit care reduces the case fatality, physical dependency and need for institutionalization compared with care on a general medical ward (section 17.6). The main difference between the two models of care is that stroke unit care is coordinated with a multidisciplinary team. Because stroke patients have such a broad range of problems, their care demands input from several professions. For coordination of the professionals’ input, it is important that at least some of them should work as a core team, with regular meetings to discuss patients’ progress and problems (Table 10.10). Other professionals (Table 10.11), who may not be regular members of the team, should be available for consultation about individual patients. Although well established in rehabilitation settings, the team has an important role in all phases of treatment, even on the day of the stroke. Of course, the type and intensity of input from different members of the team will vary at different stages of the patient’s illness. By working closely together and sharing information and skills, some blurring of the boundaries between the roles of the professions becomes possible, and this can provide greater flexibility and efficiency. For example, if the nursing staff are trained by the speech and language therapist to screen for dysphagia, this will provide every patient with early screening (even at weekends), enabling the speech and language therapist to focus on patients with definite swallowing or communication problems (sections 11.17 and 11.30). Some have suggested that we should develop a hybrid therapist who could take on several roles, but this interesting idea

Table 10.10 The core stroke team. Physician Nurse Physiotherapist Occupational therapist Speech and language therapist Social worker

..

..

Table 10.11 Other professionals who may be helpful in the management of particular stroke patients. Others who may be consulted

Example of problem

Clinical psychologist Psychiatrist Neurosurgeon Vascular surgeon Radiologist

Antisocial behaviour Severe depression Obstructive hydrocephalus Peripheral artery embolus Unusual CT scan appearance Painful shoulder Fractured neck of femur Refractive problems Persistent diplopia Shortened leg, foot drop Weight loss Formulations for dysphagia Ill-fitting dentures

Rheumatologist Orthopaedic surgeon Optometrist Ophthalmologist Orthotist Dietician Pharmacist Dentist CT, computerized tomography.

Physiotherapist Occupational therapist

Nurse

Patient and carer

Social worker

Physician Speech therapist

Fig. 10.18 The traditional model of care or rehabilitation, in which each member of the multidisciplinary team interacts independently with the patient and/or carer.

has, not surprisingly, met with considerable opposition from the existing professions. This concept may have particular merits in situations in which it is difficult to coordinate the activities of a multidisciplinary team, for example in a rural community rehabilitation setting. Figures 10.18 and 10.19 illustrate two models of how members of the stroke team can provide input to patients and their carers. In the first, each professional predominantly works directly with the patient and/or carer, while in the second each professional has less direct patient contact, but influences the care given by a primary nurse. The two models represent the extremes, and it is likely that in the real world care is provided in an intermediate way. Although the model represented in Fig. 10.19 realistically means that patients must be treated in a stroke unit, it has other important advantages.

525

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 526

526

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

Speech therapist

Occupational therapist

Physician Physiotherapist

Nurse Physician

Social worker

Patient and carer Fig. 10.19 The model of care or rehabilitation that can be adopted on a geographically defined stroke unit, in which each member of the multidisciplinary team influences the nursing input to the patients and carers, as well as having direct interaction with them.

Stroke patients, even on a stroke unit, are directly involved in therapeutic activity with therapists for only a small proportion of their time.63 Training the nurses to practise with the patients the activities initiated by the therapists should encourage consistency and increase the total amount of therapy received by patients. After all, the nurses are caring for patients on a ward throughout the 24-h period. If, for example, the physiotherapists are teaching a patient how to transfer from bed to wheelchair, it is important that the patient should continue to do this in the same way between physiotherapy sessions. Otherwise, many of the skills the patient acquires during physiotherapy sessions may not be used during activities on the ward or, more importantly, at home. The same principles apply to input from other therapists. Whether therapy delivered by a non-specialist is as effective as therapy delivered by a highly trained specialist needs to be evaluated in randomized trials. One of the important functions of the team meeting is to harmonize the activity of the therapists and the nurses who provide much of the daily input to the patients. This may be at least one of the reasons why stroke units seem to achieve better outcomes (section 17.6).

10.3.6 The roles of the team members This section outlines the main functions of the core members of the team. In practice, and certainly where a team is functioning well, there will be some blurring of the roles of the team members with each performing the less specialized functions of the others. The roles described and the professional labels used reflect those in our own clinical practice but they inevitably vary in different healthcare settings.

Few doctors really understand the important role that they can play in the care of stroke patients beyond making the initial diagnosis, searching for a treatable cause, giving acute medical therapies and initiating secondary preventive measures. The physician needs to be much more involved in the activities of the team for several reasons: • Team leader: Usually, rightly or wrongly, political power in health services lies principally with doctors, and therefore if stroke patients are to have access to adequate facilities, the doctor must be involved. • Source of knowledge: The physician should be knowledgeable about the disease processes – both stroke and non-stroke – that underlie the patients’ functional problems, and should thus understand the prognosis and likely effect of interventions. This is essential for predicting outcome and setting appropriate goals. • Medical care underpinning rehabilitation: Stroke patients commonly have coexisting and complicating medical problems (e.g. diabetes, heart failure, deep venous thrombosis), which need to be identified and treated.4 • Ability to chair meetings effectively: The physician will often have the broadest knowledge about stroke and is therefore in a good – although not necessarily unique – position to coordinate the team and to chair team meetings. This role is important, because team meetings can become very time-consuming and may lose direction without a strong chairperson (section 10.3.7). Overly long meetings are boring, demoralizing and inefficient. • Legal responsibility: The legal responsibility for the patient is the doctor’s, and therefore the doctor must be involved at all stages of the patient’s care. In the UK, to encourage specialist physician involvement in stroke care and management, there has been a move towards establishing stroke medicine as a subspecialty with its own training schemes (http://www. basp.ac.uk/2004specialtraining.pdf). Stroke physicians are trained to deal with the whole patient journey, from the hyperacute phase through to rehabilitation and longterm follow-up, and to manage most post-stroke problems (Chapter 11). In other countries these roles are often distributed among other specialists including neurologists, internal medicine specialists, rehabilitation specialists and geriatricians. Nursing staff The nursing staff probably have the broadest role in the management of patients with stroke, which includes at least four major components:

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 527

10.3 Delivering an integrated management plan

• A daily assessment of the patient’s problems, both existing and new, and of their abilities and disabilities. Every week, our nurses assess the patients using the Barthel index, which focuses their attention on important functional issues. Based on this assessment, they evolve a care plan that aims to meet the individual’s needs. • Provision of all the basic needs of dependent patients after a stroke (e.g. feeding, washing, dressing, toileting, turning and transferring). • Provision of skilled nursing to prevent the development of complications such as pressure ulcers (section 11.16), painful shoulders (section 11.24), other injuries (section 11.26) and aspiration pneumonia (sections 11.12 and 11.17). This involves correctly positioning and handling the immobile patient (section 11.11). If one adopts a model of teamworking approximating to that shown in Fig. 10.19, then the nurse’s role will include many of the functions of other members of the team. • Supporting the patient and family. With increasing emphasis by managers and politicians on improving easily measured clinical outcomes such as case fatality and disability, other aspects of caring for and about patients are easily forgotten. But these are very important to patients and close family members64 – informing, reassuring, encouraging, advising, supporting and sympathizing. The nurses usually have greatest contact with the patients and family and therefore have an important role in this area. In some healthcare systems trained nurses are not readily available, and much of the patient’s day-to-day needs are met by relatives. In many countries specific training for stroke unit nurses is not yet available (http:// www.nationalstrokenursingforum.com/). To fully realize the potential of stroke unit nurses requires considerable investment in specialist education and training. Physiotherapist The physiotherapist has several important roles in caring for stroke patients, depending on their individual needs and the stage of the illness (http://www.csp.org.uk/). Soon after a severe stroke, the physiotherapist may be involved in at least seven functions: • Providing a detailed assessment of the motor and sensory problems of patients to help estimate their prognosis. • Assessing and treating chest problems, including pneumonia and retention of secretions. Speech and language therapists often find it useful to have a physiotherapist present at swallowing assessments, to help position the patient and to deal promptly with any aspiration (section 11.17).

..

..

• Advising nurses and other carers on the best way to position patients to prevent unhelpful changes in muscle tone that may lead ultimately to contractures and further limitation of function (section 11.20). • Teaching the nurses and informal carers the best way to handle the patient to avoid pain or injury to the patient or carer. This will often involve teaching proper methods of transferring, lifting, standing and walking the patient. • Providing therapy to relieve the symptoms associated with painful shoulders or swollen limbs (sections 11.24 and 11.25). • Providing therapy to improve the patient’s mobility and arm function (section 11.21). • Advising on walking aids and splints (section 11.32.1) that may sometimes improve a patient’s function. In some countries physiotherapists have a more limited role. They may not be so involved in assessment and formulation of treatment but simply deliver the interventions prescribed by a rehabilitation specialist. Occupational therapist Occupational therapists fulfil several roles in the management of stroke patients (http://www.cot.org.uk/ specialist/nanot/forums/stroke.php). These are usually fairly limited in the very early period after a severe stroke, but become more important as the patient recovers and as self-care becomes more relevant. These roles include the following five components: • An early assessment of patients to find out how each impairment is likely to restrict their function. This requires an assessment of what the patients were able to do before the stroke, what they can do now and what their home circumstances are, e.g. ease of access to the front door, bedroom, toilet, or bathroom; or the circumstances into which the patients will be discharged, e.g. a relative’s home or nursing home. • An assessment of the patient’s visuospatial functioning. It is important to remember that many evaluations of the objective tests of visuospatial functioning use an occupational therapist’s assessment as the ‘gold standard’65 (section 11.28). • Training the patient and the carer to carry out everyday activities, despite the patient’s impairments, is a crucial role (section 11.32). This involves finding the best way of achieving a particular activity for that individual patient. Most input is spent on the activities of daily living (Table 11.45), although in specialized units or those dealing with younger patients, therapy aimed at return to an occupation (section 11.33.2) or leisure activity (section 11.33.3) may be available.

527

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 528

528

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

• Provision of aids and adaptations to allow patients to function better. In the UK, this includes wheelchair provision, feeding and kitchen aids, and bathroom aids, among many others (section 11.32). • Assessment of patients’ ability to function in their own home, e.g. home visit which is often done before discharge with patients who have been admitted to hospital, and is an important element of discharge planning. It is important to identify problems that are specific to the patient’s own home environment and which may have to be solved by further training or by provision of aids or adaptations. Speech and language therapist Speech and language therapists have several roles in the care of stroke patients (http://www.rcslt.org/). They include: • Assessment of swallowing safety both initially and as the patients improve, so that their diet and fluid intake matches their swallowing abilities (section 11.17). • Teaching the patient, nurses or family who are involved with feeding, techniques that help overcome a swallowing impairment and avoid aspiration. • Teaching the patients exercises that may increase the rate of recovery of swallowing problems (section 11.17). • Diagnosis and assessment of the patient’s communication problems (section 11.30). • Informing both formal and informal carers of the nature of the patient’s communication problems. Patients, their families and even the professionals caring for the patient often have great difficulty in understanding what is wrong with a dysphasic patient. They may think the patient is ‘confused’, ‘demented’, or simply ‘mad’. Because problems with communication so often lead to emotional distress, most speech and language therapists also take on a counselling role. • Teaching patients, carers and even volunteers, strategies to allow the patients to communicate effectively using language (spoken or written), gesture, or communication aids where appropriate. • Providing therapy that may enhance the recovery of communication difficulties (section 11.30). Social worker The role of the social worker is bound to vary in different societies. In the UK, Australia and the Netherlands (which are the countries where we have direct experience), the social worker is likely to: • Provide patients and their families with practical advice and help at all stages of the patients’ illness. For example, arranging subsidized transport for the family to

visit the hospital, or extra home care for a dependent relative if the stroke patient was the main carer before admission. Social workers often help with any financial problems that have arisen because the main breadwinner has had a stroke, e.g. by applying for allowances or grants. • Be involved helping to plan the patient’s discharge from hospital, or change of accommodation. Social workers often spend a lot of time identifying the wishes and needs of the patient and family and then, with the rest of the team, trying to meet them. Where patients are unable to make decisions for themselves and have no close family, the social worker may need to act as the patient’s advocate and make arrangements for financial affairs to be taken care of, perhaps even arranging any change of accommodation (e.g. transfer to a nursing home). • Follow up patients and families after the patient’s discharge from hospital, to identify their changing need for support and to make adjustments to any care package. • Provide counselling that may be helpful in allowing patients and families to come to terms with the change in circumstances brought about by the stroke. Some organize groups of patients, carers, or both to help solve problems. One function that should be shared by all members of the team is that of monitoring the patient’s condition. Nurses and therapists, in particular, spend a lot of their time handling patients and observing the patient’s performance, so they are often in an ideal position to notice the relatively minor changes that may be an early sign of a complication, which might benefit from early treatment. All members of the team should be alert to changes in the patient’s condition that may indicate the development of a complication. Are the interventions of the team members effective? Although organized stroke care improves outcomes, some question the effectiveness of physiotherapy, occupational therapy, speech and language therapy and social work. They point to the lack of research evidence to support the effectiveness of these professionals (but interestingly, without questioning their own effectiveness), or they emphasize particular studies that appear to demonstrate the ineffectiveness of other professions. Unfortunately, these ‘negative’ studies have, in general, asked the wrong questions, measured the wrong outcomes, been too small, and have not evaluated the intervention in the context of a well-organized stroke team. This has possibly resulted in ‘false-negative’ studies,

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 529

10.3 Delivering an integrated management plan

MULTIDISCIPLINARY TEAM MEETINGS Addressograph

WARD: Date:

/

/

For CPR: Yes:

Week No:

No:

Barthel:

Milestones:

None, Sit, Stand, Steps, 10m

Falls Risk? YES/NO

Nursing: present /absent

Weight:

Medical: Consultant present /absent

Nutrition: Dietician present /absent

Physiotherapy: present /absent

Speech & Language therapy: present /absent

Occupational therapy: present /absent

Social Work: present /absent

Action Plan: 1. 2. 3. Discharge within one week? Y/N Signed:

Designation:

Fig. 10.20 A proforma for recording the results of a multidisciplinary team meeting.

resulting in rejection of valuable team members. But, as everyone must realize, failure to demonstrate a definite effect cannot be taken as proof of a lack of effect.66 There may be uncertainty about the effectiveness, optimum ‘dose’ and duration of some of the specific therapeutic

..

..

manoeuvres that therapists use (in common with many medical practices) but, as we have shown, this type of input forms only a small part of their contribution to the care of stroke patients. In fact, there are an increasing number of well-conducted randomized trials that have

529

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 530

530

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

established that interventions by therapists do improve patient outcome, as discussed in Chapter 11.67–69

10.3.7 Multidisciplinary team meetings The team must meet regularly if it is to work effectively. All the stroke units included in the recent systematic review of the randomized controlled trials held at least weekly meetings of the multidisciplinary team, separate from conventional ward rounds.70 These meetings have several important functions: • The entire team can be introduced to new patients and their problems. • Existing patients can be reviewed, and if individual team members have noted a change in their condition or a new problem, this can be communicated to the other members. • After reviewing each patient’s progress and any developing problems, realistic goals can be set jointly, and an appropriate course of action to meet those goals can be agreed (section 10.3.3). • Verbal reports of individual therapists’ assessments, and in particular the results of pre-discharge home visits, can be discussed and detailed plans for discharge can be made. • Meetings have an educational role for the team members, students and visitors. For example, we often show the patient’s brain CT and discuss the likely pathogenesis of the stroke and any rationale for treatment or therapy. So that important details are not overlooked, it may be useful to agree a formal structure to the discussions which can be reinforced by using a standard form on which team discussions are recorded (Fig. 10.20). The structure we have adopted is discussed below. This is not rigidly adhered to, but provides a framework for discussions about individual patients. Structure of team discussion The medical details of any new patients are presented by the physician, including a brief account of the patient’s symptoms and any relevant past history, including risk factors and the presumed cause of the stroke. The patient’s social background is also presented briefly by the physician, but other members of the team are often able to contribute extra details. The patient’s neurological impairments are discussed, and the therapists have often identified some that may not have been obvious to the physician. By the time of the meeting, the nursing staff can report on the functional consequences of these impairments (i.e. disability), and using the information about pre-stroke activities, some estimate of the stroke’s

likely effect on the patient’s life after the stroke can be made (i.e. predicted handicap). At each subsequent meeting, we introduce the patient briefly with a résumé of the date of stroke, clinical type and presumed cause. We then summarize the problems, goals and actions that were agreed at the last meeting. Each member of the team is then invited to update the rest of the team on the patient’s progress, what problems and goals have been solved or achieved, what new goals they plan to set, and how they plan to achieve them. Usually we do this in a set order that tends to reflect the way we think about patients, i.e. pathology, impairment, disability and handicap. The physician starts by updating the team on any changes in the patient’s medical condition or the results of any important investigations that may be relevant to other team members. The nurses then give an overall report on the patient’s progress, including current performance on the Barthel index. This usefully highlights many of the patient’s functional problems, as well as giving an objective indication of progress. The therapists follow: first the physiotherapist, then the speech therapist and lastly the occupational therapist. The first two tend to focus more on impairments and the patient’s basic functions such as mobility, swallowing and communication, while the occupational therapist usually focuses on the broader range of disabilities and how these are likely to affect the patient’s everyday life. Lastly, the social worker reports on any problems which close contact with the family may have revealed and progress regarding discharge planning. This sequence also has the advantage that the occupational therapist and social worker can make use of the information from the others in formulating their own goals and actions. The discussion is then opened up, and longer-term goals such as timing of home visits, discharges and case (family) conferences are set, and we decide who will do what by the next meeting including who will communicate with the patient and family to ensure that they are involved with the goal-setting process and that their views are being taken into account. Multidisciplinary teams may lack effectiveness if each member is not given an equal chance to contribute. Inevitably, teams will include some more assertive individuals and other less outgoing members, the former tending to dominate discussions. It is important that the team leader – often the physician – ensures that individual members do not monopolize the discussions to the extent that others are not heard. The framework we have adopted lessens the likelihood of this arising, although of course not every member will necessarily have something useful to contribute in every case.

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 531

10.4 Some difficult ethical issues

It is important that the person chairing the team meetings ensures that no individual member, including the chairperson, monopolizes the discussion to the extent that others are not heard.

We find it useful to have separate team meetings where patients are not discussed when we discuss general management issues, how the team is working, what the problems are, and any changes that might be made in the way we work.

Involving patients and carers in team meetings Neither the patient nor the family have been mentioned yet as having a direct input into the team meetings, although their input is obviously crucial in setting goals and planning discharge from hospital. We do not invite them to attend each meeting, but we do seek their views on certain issues in advance of the meeting. We also try to ensure that the nature and conclusions of the discussion at the meeting are communicated back to the patient and/or family by the most appropriate team member, most often the nurse. Although it is sometimes vital to involve the patient and families in discussions, we feel that their attendance at each weekly meeting would inhibit the discussion and might be distressing for them. It would also make meetings longer and more unwieldy. We hold separate team meetings (case conferences or family meetings) to which the patient, family and any other people who are involved – such as the district (community) nurse and home care organizer – are invited. These meetings are useful in planning hospital discharge in complex cases and in resolving differences of opinion between the team members and the patient or family. The timing of these meetings is tailored to the individual patient. Team building It is useful to have separate team meetings at which patients are not discussed. On these occasions, we discuss general management issues, how the team is working, what the problems are, and any changes that might be made in the way we work. We discuss critical incidents to ensure that we learn from these. These sessions also allow individuals to tell the rest of the team about certain specialized aspects of their jobs. This, we believe, not only encourages some blurring of the distinctions between their functions but also aids team building. Each member can more fully appreciate the contribution made by other members. We also have weekly meetings, involving the consultants and trainees, to discuss patients entering our stroke register where we review their clinical and radiological details. These provide an additional opportunity for education, training, quality control and discussion of difficult cases with consultant colleagues.

..

..

Recording the work of the team We have not yet found a totally satisfactory method of recording the work of the team. Traditionally, members have each kept their own records so that they may refer to them, but this inevitably leads to much unnecessary duplication. However, a unified record has disadvantages, the main one being that different members of the team require information of varying detail. Thus, the physiotherapist’s record contains very detailed notes on the patient’s motor functioning that are irrelevant to the social worker. Also, the records may be needed simultaneously by the physiotherapist in the gymnasium and by the pharmacist on the stroke unit. In our units, each profession has therefore kept its own records, but they do all make entries in a combined medical and nursing record. The nurses base their care plans on the discussions at the team meetings. This has been helped by their adoption of a problem-orientated approach, so that the nurses regularly record the patients’ problems, the goals and actions aimed at attaining the goals. A compromise in which core details are collected once and shared, while more detailed records are kept by each team member, might provide the best and most efficient solution. We also increasingly use structured records in which the results of the most commonly used assessments (e.g. swallowing, continence) and actions taken require only a tick, date and signature. In the future, increased use of information technology may facilitate a combined patient record that avoids some of these problems but even now we do ensure that changes in the patient’s condition, key decisions and proposed actions (by whom and by when) are recorded in the medical notes. This is important to avoid unnecessary and time-wasting repetition of discussions at future meetings and to ensure that decisions lead on to actions.

10.4 Some difficult ethical issues

When patients are judged unlikely to achieve a reasonable quality of life if they were to survive, many

531

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 532

532

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

clinicians decide not to strive to ensure survival. However, this is an extremely difficult decision since, as we have already discussed, there are considerable difficulties in predicting outcome reliably at an early stage (section 10.2.7). Also the value that the patient would place on survival in a given state is impossible to judge. Patients who have had a major stroke are often not in a position to make decisions about their treatment because, due to a depressed conscious level, language or visuospatial problems, they are unable to understand or communicate. Under these circumstances, the family or potential carers should be included in the decision-making process. The family can often provide information allowing judgements to be made about what the patient’s likely wishes are. However, some relatives may have their own interests and not just those of the patient to consider. Younger relatives may gain financially from the death of an older relative, and death removes the obligation to care for a severely disabled person. Patients may even have written down what they would like to happen in the event of a life-threatening illness (i.e. a living will). The legal status of these so-called ‘living wills’ varies from country to country, however. Also, it is not at all clear whether patients’ judgements made while in good health about the value they might place on their life if they were in a dependent state remain true when they have become disabled. We have encountered many people who have become severely disabled and yet have accepted their disability and felt that their quality of life is acceptable or even very good. It is difficult to know how to react to patients’ previously expressed view that they would never, ever want to live in a dependent state or go to a nursing home, when they then seem to cling to life after a major stroke. One has to be particularly careful in assessing the likely prognosis of an individual patient with an apparently severe stroke. A relatively minor stroke that is complicated by another medical problem (e.g. pneumonia) may be impossible to differentiate from a severe stroke (section 11.4). If the complicating medical problem can be identified and treated then the patient may achieve a very satisfactory clinical outcome despite the apparently gloomy initial prognosis. It is not always easy to distinguish between a patient who has had a catastrophic stroke and one who has had a less severe stroke that is complicated and worsened by infection, epileptic seizures or metabolic problems. The prognosis is much worse for the former patient than the latter one.

Perhaps the most common dilemma is whether or not to give fluids or nutritional support to a patient with an ‘apparently severe stroke’ who is unable to swallow. Early nutritional support is likely to improve the proportion of patients surviving although not perhaps the functional status of survivors.71 Few would argue against giving parenteral fluids to a conscious patient to prevent dehydration, thirst and discomfort. However, if the patient is unconscious and thus probably unaware of any discomfort, then there is more uncertainty about what to do for the best. Also, if the patient develops an infection very soon after the onset of the stroke, apart from the difficulties mentioned above in assessing the severity of stroke, the question arises as to how aggressively one should treat the patient. Are intravenous antibiotics, physiotherapy or even artificial ventilation and inotropic support appropriate in this situation? There are some facts worth considering before making these difficult decisions. Most patients who are unconscious soon after a stroke die in the first few days from the direct neurological effects of the stroke on brainstem function, and not from dehydration or infection (section 10.2.3). Therefore, by simply giving fluids, it is unlikely that the lives of many patients will be greatly prolonged, but this may give more time to make an accurate assessment of prognosis. If one elects to give a patient fluids, antibiotics or even nutritional support via a feeding tube, this clearly does not have to be continued on a long-term basis. However, in some countries there are laws that prevent clinicians from withdrawing treatment of this kind, and many clinicians feel more uncomfortable about stopping a treatment that they have started than about not starting it in the first place. Apart from worrying about medicolegal issues, one has to consider the reaction of the family. However, in our experience, if the physician has spent enough time informing the family of the patient’s state and likely prognosis, and has involved them in the decisions, then withdrawal of fluids, feeding or antibiotics is seldom a major problem. Unfortunately, in some countries there is a trend for such decisions – especially in severely brain-damaged children and patients in a persistent vegetative state – to be delegated to the legal profession. We cannot hope to have the answers to all these difficult problems, but perhaps it is worth suggesting a general approach that we have found practicable: • First, collect accurate and unbiased information about the patient’s life before the stroke. It is not good enough to base one’s decisions on the severity of just the stroke, since this may be difficult to estimate accurately, and other factors will influence the prognosis (section 11.4). The patient’s pre-stroke functional level and social activity are crucial in making these

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 533

References

decisions. Unfortunately, this information is often not collected in sufficient detail during the acute phase, because its relevance to acute care is not recognized (section 10.3.2). One might be less aggressive in treating patient A, who was previously handicapped, miserable and in pain from severe arthritis and living in an institution, than patient B, who has an equally severe stroke but was previously living at home independently with his or her family. Usually, the patient’s pre-stroke handicap sets the ceiling on the potential post-stroke outcome, so that these two patients have very different outlooks. It is still a difficult – some would say an impossible – judgement to say whether patient A’s life is worth prolonging. • Next, it is important to make a detailed assessment of the patient’s condition in order to make the best possible judgement about the prognosis. • Third, formulate the various alternative management plans that might be adopted. These should include those from the most aggressive to the most conservative. • Finally, all options must be discussed with other members of the team and any close family members. The family needs to be given accurate information about the diagnosis, problems, likely outcomes and treatment possibilities. It is probably unfair to suggest that the decisions are all theirs, but any decision the clinician makes should ideally be compatible with their views. If one does not take the family’s wishes into account one is inviting trouble, complaints – and nowadays litigation. It is worth making it clear to them that few decisions are irreversible (except perhaps turning off a ventilator), since circumstances change. One is often in a much better position to make an accurate prognosis having observed the patient’s progress over a few days than on the day of the stroke, so what appeared to be an appropriate intervention on day 1 may appear less appropriate after a few days. It can also be useful to involve an independent and experienced clinician to provide a second opinion. The development of better tools to help predict prognosis, and studies that give a better understanding of the effect of our supportive interventions, should make these difficult decisions somewhat easier in the future. Professional bodies often provide specific advice which is applicable to particular countries (e.g. http://www. gmc-uk.org/standards/default.htm).

Involve other members of the team and the patient’s family in any decisions about how aggressively one should strive to keep a patient alive. Emphasize that most decisions can be reviewed – and reversed – in the light of the patient’s progress.

..

..

References

1 WHO. International Classification of Impairments, Disabilities and Handicaps. Geneva: World Health Organization; 1980. 2 WHO. International Classification of Functioning, Disability and Health (ICF). Geneva: World Health Organization; 2001. 3 Bornman J. The World Health Organization’s terminology and classification: application to severe disability. Disabil Rehabil 2004; 26(3):182–8. 4 Langhorne P, Stott DJ, Robertson L, MacDonald J, Jones L, McAlpine C et al. Medical complications after stroke: a multicenter study. Stroke 2000; 31(6):1223–9. 5 Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. A prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project 1981–86. 2. Incidence, case fatality rates and overall outcome at one year of cerebral infarction, primary intracerebral and subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1990; 53(1):16–22. 6 Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM et al. Change in stroke incidence, mortality, case-fatality, severity, and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet 2004; 363(9425):1925–33. 7 Bonita R, Broad JB, Beaglehole R. Changes in stroke incidence and case-fatality in Auckland, New Zealand, 1981–91. Lancet 1993; 342(8885):1470–3. 8 Peltonnen M, Stegmayr B, Asplund K. Marked improvement since 1985 in short-term and long-term survival after stroke. Cerebrovasc Dis 1999; 9(Suppl. 1): 62. 9 Gresham GE, Kelly-Hayes M, Wolf PA, Beiser AS, Kase CS, D’Agostino RB. Survival and functional status 20 or more years after first stroke: the Framingham Study. Stroke 1998; 29(4):793–7. 10 Bamford J, Dennis M, Sandercock P, Burn J, Warlow C. The frequency, causes and timing of death within 30 days of a first stroke: the Oxfordshire Community Stroke Project. J Neurol Neurosurg Psychiatry 1990; 53(10):824–9. 11 Burn J, Dennis M, Bamford J, Sandercock P, Wade D, Warlow C. Long-term risk of recurrent stroke after a firstever stroke. The Oxfordshire Community Stroke Project. Stroke 1994; 25(2):333–7. 12 Van GJ. Subarachnoid haemorrhage. Lancet 1992; 339(8794):653–5. 13 Dombovy ML, Basford JR, Whisnant JP, Bergstralh EJ. Disability and use of rehabilitation services following stroke in Rochester, Minnesota, 1975–1979. Stroke 1987; 18(5):830–6. 14 Kwakkel G, Kollen B, Lindeman E. Understanding the pattern of functional recovery after stroke: facts and theories. Restor Neurol Neurosci 2004; 22(3–5):281–99. 15 Gray CS, French JM, Bates D, Cartlidge NE, James OF, Venables G. Motor recovery following acute stroke. Age Ageing 1990; 19(3):179–84.

533

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 534

534

Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients

16 Duncan PW, Goldstein LB, Matchar D, Divine GW, Feussner J. Measurement of motor recovery after stroke. Outcome assessment and sample size requirements. Stroke 1992; 23(8):1084–9. 17 Ashburn A. Physical recovery following stroke. Physiotherapy 1997; 83:480–90. 18 Kwon S, Hartzema AG, Duncan PW, Min-Lai S. Disability measures in stroke: relationship among the Barthel Index, the Functional Independence Measure, and the Modified Rankin Scale. Stroke 2004; 35(4):918–23. 19 Dennis MS. Outcome after brain haemorrhage. Cerebrovasc Dis 2003; 16 Suppl 1:9–13. 20 Barber M, Roditi G, Stott DJ, Langhorne P. Poor outcome in primary intracerebral haemorrhage: results of a matched comparison. Postgrad Med J 2004; 80(940):89–92. 21 Lipson DM, Sangha H, Foley NC, Bhogal S, Pohani G, Teasell RW. Recovery from stroke: differences between subtypes. Int J Rehabil Res 2005; 28(4):303–8. 22 Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003; 2(1):43–53. 23 Tanaka H, Ueda Y, Date C, Baba T, Yamashita H, Hayashi M et al. Incidence of stroke in Shibata, Japan: 1976–1978. Stroke 1981; 12(4):460–6. 24 Asplund K, Bonita R, Kuulasmaa K, Rajakangas AM, Schaedlich H, Suzuki K et al. Multinational comparisons of stroke epidemiology. Evaluation of case ascertainment in the WHO MONICA Stroke Study. World Health Organization Monitoring Trends and Determinants in Cardiovascular Disease. Stroke 1995; 26(3):355–60. 25 Feigin VL, Nikitin I, Kholodov VA, Shishkin SV, Novokhatskaia MV, Belenko AI, Khatsenko VN. [The epidemiology of cerebral stroke in Siberia based on registry data]. Zh Nevropatol Psikhiatr Im S S Korsakova 1996; 96(6):59–65. 26 Petty GW, Brown RD, Jr., Whisnant JP, Sicks JD, O’Fallon WM, Wiebers DO. Ischemic stroke: outcomes, patient mix, and practice variation for neurologists and generalists in a community. Neurology 1998; 50(6):1669–78. 27 Wijdicks EF, Rabinstein AA. Absolutely no hope? Some ambiguity of futility of care in devastating acute stroke. Crit Care Med 2004; 32(11):2332–42. 28 Counsell C, Dennis M. Systematic review of prognostic models in patients with acute stroke. Cerebrovasc Dis 2001; 12(3):159–70. 29 Meijer R, Ihnenfeldt DS, van Limbeek J, Vermeulen M, de Haan RJ. Prognostic factors in the subacute phase after stroke for the future residence after six months to one year. A systematic review of the literature. Clin Rehabil 2003; 17(5):512–20. 30 Meijer R, Ihnenfeldt DS, de Groot IJ, van Limbeek J, Vermeulen M, de Haan RJ. Prognostic factors for ambulation and activities of daily living in the subacute phase after stroke. A systematic review of the literature. Clin Rehabil 2003; 17(2):119–29. 31 Gladman JR, Harwood DM, Barer DH. Predicting the outcome of acute stroke: prospective evaluation of five

32

33

34

35

36

37

38

39

40

41

42

43

44

45

multivariate models and comparison with simple methods. J Neurol Neurosurg Psychiatry 1992; 55(5):347–51. Taub NA, Wolfe CD, Richardson E, Burney PG. Predicting the disability of first-time stroke sufferers at 1 year. 12month follow-up of a population-based cohort in southeast England. Stroke 1994; 25(2):352–7. Wardlaw JM, Lewis SC, Dennis MS, Counsell C, McDowall M. Is visible infarction on computed tomography associated with an adverse prognosis in acute ischemic stroke? Stroke 1998; 29(7):1315 –9. Nuutinen J, Liu Y, Laakso MP, Karonen JO, Roivainen R, Vanninen RL et al. Assessing the outcome of stroke: a comparison between MRI and clinical stroke scales. Acta Neurol Scand 2006; 113(2):100–7. Hendricks HT, Zwarts MJ, Plat EF, van Limbeek J. Systematic review for the early prediction of motor and functional outcome after stroke by using motor-evoked potentials. Arch Phys Med Rehabil 2002; 83(9):1303–8. Kwakkel G, Wagenaar RC, Kollen BJ, Lankhorst GJ. Predicting disability in stroke: a critical review of the literature. Age Ageing 1996; 25(6):479–89. Counsell C, Dennis M, McDowall M, Warlow C. Predicting outcome after acute and subacute stroke: development and validation of new prognostic models. Stroke 2002; 33(4):1041–7. Weingarten S, Bolus R, Riedinger MS, Maldonado L, Stein S, Ellrodt AG. The principle of parsimony: Glasgow Coma Scale score predicts mortality as well as the APACHE II score for stroke patients. Stroke 1990; 21(9):1280–2. Wright CJ, Swinton LC, Green TL, Hill MD. Predicting final disposition after stroke using the Orpington Prognostic Score. Can J Neurol Sci 2004; 31(4):494–8. Horgan NF, Cunningham CJ, Coakley D, Walsh JB, O’Neill D, O’Regan M et al. Validating the Orpington Prognostic Score in an Irish in-patient stroke population. Ir Med J 2005; 98(6):172, 174–5. Weir NU, Counsell CE, McDowall M, Gunkel A, Dennis MS. Reliability of the variables in a new set of models that predict outcome after stroke. J Neurol Neurosurg Psychiatry 2003; 74(4):447–51. Dennis MS, Lewis SC, Warlow C. Routine oral nutritional supplementation for stroke patients in hospital (FOOD): a multicentre randomised controlled trial. Lancet 2005; 365(9461):755–63. Counsell C, Dennis MS, Lewis S, Warlow C. Performance of a statistical model to predict stroke outcome in the context of a large, simple, randomized, controlled trial of feeding. Stroke 2003; 34(1):127–33. Frankel MR, Morgenstern LB, Kwiatkowski T, Lu M, Tilley BC, Broderick JP et al. Predicting prognosis after stroke: a placebo group analysis from the National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial. Neurology 2000; 55(7):952–9. Weimar C, Konig IR, Kraywinkel K, Ziegler A, Diener HC. Age and National Institutes of Health Stroke Scale Score within 6 hours after onset are accurate predictors of outcome after cerebral ischemia: development and external validation of prognostic models. Stroke 2004; 35(1):158–62.

..

..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 535

References

46 Muir KW, Weir CJ, Murray GD, Povey C, Lees KR. Comparison of neurological scales and scoring systems for acute stroke prognosis. Stroke 1996; 27(10):1817–20. 47 Lai SM, Duncan PW, Keighley J. Prediction of functional outcome after stroke: comparison of the Orpington Prognostic Scale and the NIH Stroke Scale. Stroke 1998; 29(9):1838– 42. 48 Tilling K, Sterne JA, Rudd AG, Glass TA, Wityk RJ, Wolfe CD. A new method for predicting recovery after stroke. Stroke 2001; 32(12):2867–73. 49 Tilling K, Sterne JA, Wolfe CD. Multilevel growth curve models with covariate effects: application to recovery after stroke. Stat Med 2001; 20(5):685–704. 50 Counsell C, Dennis M, McDowall M. Predicting functional outcome in acute stroke: comparison of a simple six variable model with other predictive systems and informal clinical prediction. J Neurol Neurosurg Psychiatry 2004; 75(3):401–5. 51 Harbison J, Hossain O, Jenkinson D, Davis J, Louw SJ, Ford GA. Diagnostic accuracy of stroke referrals from primary care, emergency room physicians, and ambulance staff using the face arm speech test. Stroke 2003; 34(1):71–6. 52 Nor AM, McAllister C, Louw SJ, Dyker AG, Davis M, Jenkinson D, Ford GA. Agreement between ambulance paramedic- and physician-recorded neurological signs with Face Arm Speech Test (FAST) in acute stroke patients. Stroke 2004; 35(6):1355–9. 53 Rothwell PM, Giles MF, Flossmann E, Lovelock CE, Redgrave JN, Warlow CP, Mehta Z. A simple score (ABCD) to identify individuals at high early risk of stroke after transient ischaemic attack. Lancet 2005; 366(9479):29–36. 54 Hill MD, Weir NU. Is the ABCD score truly useful? Stroke 2006; 37(7):1636. 55 Davenport RJ, Dennis MS, Warlow CP. Improving the recording of the clinical assessment of stroke patients using a clerking pro forma. Age Ageing 1995; 24(1):43–8. 56 Hancock RJ, Oddy M, Saweirs WM, Court B. The RCP stroke audit package in practice. J R Coll Physicians Lond 1997; 31(1):74 –8. 57 Bero LA, Grilli R, Grimshaw JM, Harvey E, Oxman AD, Thomson MA. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. The Cochrane Effective Practice and Organization of Care Review Group. Br Med J 1998; 317(7156):465–8. 58 Kwan J, Sandercock P. In-hospital care pathways for stroke. [update of Cochrane Database Syst Rev 2002; (2):CD002924; PMID: 12076460]. Cochrane Database Syst Rev 2004; (4):CD002924. 59 Smith MT, Baer GD. Achievement of simple mobility milestones after stroke. Arch Phys Med Rehabil 1999; 80(4):442–7. 60 Wade DT, Wood VA, Heller A, Maggs J, Langton HR. Walking after stroke. Measurement and recovery over the first 3 months. Scand J Rehabil Med 1987; 19(1):25–30.

..

..

61 Parry RH. Communication during goal-setting in physiotherapy treatment sessions. Clin Rehabil 2004; 18(6):668–82. 62 Wressle E, Eeg-Olofsson AM, Marcusson J, Henriksson C. Improved client participation in the rehabilitation process using a client-centred goal formulation structure. J Rehabil Med 2002; 34(1):5–11. 63 Bernhardt J, Dewey H, Thrift A, Donnan G. Inactive and alone: physical activity within the first 14 days of acute stroke unit care. Stroke 2004; 35(4):1005–9. 64 Pound P, Bury M, Gompertz P, Ebrahim S. Stroke patients’ views on their admission to hospital. Br Med J 1995; 311(6996):18–22. 65 Stone SP, Wilson B, Wroot A, Halligan PW, Lange LS, Marshall JC, Greenwood RJ. The assessment of visuo-spatial neglect after acute stroke. J Neurol Neurosurg Psychiatry 1991; 54(4):345–50. 66 Altman DG, Bland JM. Absence of evidence is not evidence of absence. Br Med J 1995; 311(7003):485. 67 Greener J, Enderby P, Whurr R. Speech and language therapy for aphasia following stroke. Cochrane Database Syst Rev 2000; (2):CD000425. 68 Walker MF, Leonardi-Bee J, Bath P, Langhorne P, Dewey M, Corr S et al. Individual patient data meta-analysis of randomized controlled trials of community occupational therapy for stroke patients. Stroke 2004; 35(9):2226–32. 69 Steultjens EM, Dekker J, Bouter LM, Van de Nes JC, Cup EH, Van den Ende CH. Occupational therapy for stroke patients: a systematic review. Stroke 2003; 34(3):676–87. 70 Langhorne P, Pollock A. What are the components of effective stroke unit care? Age Ageing 2002; 31(5):365–71. 71 Dennis MS, Lewis SC, Warlow C. Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomised controlled trial. Lancet 2005; 365(9461):764 –72. 72 Sackett DL, Haynes RB, Guyatt GH, Tugwell MD. Clinical Epidemiology: A Basic Science for Clinical Medicine. 2nd ed. Boston: Little, Brown, 1991. 73 Al-Shahi R, Vousden C, Warlow C. Bias from requiring explicit consent from all participants in observational research: prospective, population based study. Br Med J 2005; 331(7522):942. 74 Counsell C. The Prediction of Outcome in Patients with Acute Stroke. Cambridge: University of Cambridge; 1998. 75 Dennis MS, Burn JP, Sandercock PA, Bamford JM, Wade DT, Warlow CP. Long-term survival after first-ever stroke: the Oxfordshire Community Stroke Project. Stroke 1993; 24(6):796–800. 76 Langhorne P, Murray GD, Stott DJ, McAlpine C, MacDonald J. Profiles of Recovery after Stroke: Their Value in Monitoring Recovery and Predicting Outcome. Chief Scientists Office report. K/RED/4/C284, in press.

535

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 537

11

What are this patient’s problems? A problem-based approach to the general management of stroke

11.1

Introduction

11.2

Airway, breathing and circulation

11.3

Reduced level of consciousness

11.4

Severe stroke vs apparently severe stroke

11.5

Worsening after a stroke

11.6

Coexisting medical problems

11.7

High and low blood pressure after stroke

11.8

Epileptic seizures

11.9

Headache, nausea and vomiting

11.10 Hiccups

538 538

543 544

545 548 549

551 553

553

11.11 Immobility and poor positioning 11.12 Fever and infection

553

555

11.13 Venous thromboembolism

557

11.14 Urinary incontinence and retention

560

11.15 Faecal incontinence and constipation 11.16 Pressure ulcers

564

564

11.17 Swallowing problems

567

11.18 Metabolic disturbances 11.19 Nutritional problems

572

575

11.20 Spasticity and contractures

579

11.21 Limb weakness, poor truncal control and unsteady gait 11.22 Sensory impairments

11.23 Pain (excluding headache) 11.24 Painful shoulder

587

588

11.25 Swollen and cold limbs 11.26 Falls and fractures 11.27 Visual problems

590

591 592

11.28 Visuospatial dysfunction 11.29 Cognitive dysfunction

593 595

11.30 Communication difficulties 11.31 Psychological problems

598

599

11.32 Dependency in activities of daily living 11.33 Social difficulties 11.34 Carer problems

583

586

603

614 617

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

537

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 538

538

Chapter 11 What are this patient’s problems?

11.1 Introduction

The patient’s general management, as distinct from the specific treatment of their stroke (Chapters 12–14), is primarily aimed at identifying and solving existing problems, as well as anticipating and preventing potential problems, at different stages of their illness. This chapter will cover the common problems which occur in patients who have had a stroke. Each section is loosely structured as follows: • General description of each problem which includes a definition, its frequency, causes and clinical significance, and prognosis. • Assessment, including methods of detection, simple clinical assessments and measures which may be appropriate for goal setting, audit or research. • Prevention and treatment, including interventions which may reduce the risk of a problem developing or hasten recovery. Post-stroke problems have seldom been systematically identified in community-based incidence studies (section 17.11.1) so their frequency in unselected populations is often unknown. Moreover, compared with other aspects of treatment there are few large randomized controlled trials (RCTs) or systematic reviews addressing these problems. This is partly because there are major

methodological difficulties in performing RCTs and systematic reviews to evaluate non-pharmacological interventions (Table 11.1). However, the evidence base is improving. Existing trials are being systematically appraised (www.effectivestrokecare.org) and more trials are being performed. Where there is no high-quality evidence the content of this chapter reflects our own clinical experience. Many of the topics are not specific to stroke medicine but might be found in any textbook of internal medicine or surgery. Therefore, we have not attempted to provide comprehensive reviews, but instead have concentrated on those aspects of assessment and management which are particularly relevant to stroke patients.

11.2 Airway, breathing and circulation

Inadequate airway, breathing or circulation are lifethreatening. Urgent resuscitative measures must be taken. Even if not an immediate threat to survival, it seems sensible to optimize the delivery of oxygen and glucose to the brain to minimize brain damage and so achieve the best possible outcome for the patient (section 12.1).

Table 11.1 Problems in performing randomized controlled trials and systematic reviews of physiotherapy interventions. Lack of theoretical model or rationale founded on basic science for many interventions Ethical problems of performing randomized controlled trials due to therapists’ certainty about the effectiveness of their own therapy in individual patients Difficulty in getting patients and therapists, to accept the possibility of ‘no treatment’ Strong patient preferences based on their belief in the benefits and acceptability of therapy Difficulty in blinding patients to treatment allocation Difficulty in designing convincing placebo treatments, or establishing appropriate ‘controls’ Difficulties in defining a therapy (in terms of type, dose, frequency, timing and duration) which has to be tailored to the individual patient; this leads to difficulties in applying the results in practice Failure to identify key components or interactions between components of an intervention Moderate treatment effects mean that large numbers of patients have to be randomized The need to randomize large numbers of patients may necessitate multicentre trials but these raise difficulties in standardizing and monitoring treatment Therapy is very labour-intensive and therefore expensive, so that bodies which fund research may not be willing to fund the therapy Difficulties agreeing suitable measures of outcome which are sensitive to both what therapists expect to influence, and what is relevant to the patient and family In unblinded trials, patient loyalty to a therapist may bias their responses to subjective outcome scales Various different outcome measures which hinders systematic review of trials Problems due to complex interactions with other therapies given simultaneously

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 539

11.2 Airway, breathing and circulation

11.2.1 Obstructed airway Patients with a decreased level of consciousness, impaired bulbar function or who have aspirated may have an obstructed or partially obstructed airway. Central cyanosis, noisy airflow with grunting, snoring or gurgling, an irregular breathing pattern and indrawing of the suprasternal area and intercostal muscles may indicate an obstruction. Transient obstruction is common in the acute phase of stroke during sleep (section 11.2.2) and it is, therefore, important that apnoeic spells due to an obstructed airway are not mistakenly attributed to

Fig. 11.1 It is essential to maintain a patent airway in a patient with a decreased level of consciousness. (a) In comatose patients with their head in a resting or flexed position the tongue falls backwards to obstruct the hypopharnyx (arrow) and the epiglottis obstructs the larynx. (b) Tilting the head back by lifting the chin stretches the anterior neck structures, which opens the airway.

(a)

(a)

Fig. 11.2 Airways. (a) Oropharyngeal (Guedel) airway. To check that the airway is the correct size for the patient hold the airway across the cheek from corner of mouth to tip of ear lobe (the angle of the jaw bone may also be used). If this matches then the airway is the correct size. If too long or too short select another size. Incorrect sizing can result in the airway being blocked rather than kept open. Inserted upside down, they are moved along the roof of the mouth and rotated into position

..

..

periodic respiration and so ignored (e.g. Cheyne–Stokes; section 11.2.2). If an obstructed airway is suspected, the oropharynx should be cleared of any foreign matter with a sweep of a gloved finger, the patient’s jaw pulled forward, and the neck extended to stop the tongue falling back to obstruct the airway (Fig. 11.1). Positioning the patient in the coma (i.e. recovery) position may be enough to keep the airway clear, although in some situations an oropharyngeal or nasopharyngeal airway, or even endotracheal intubation, may be required (Fig. 11.2).

(b)

(b)

sitting behind the tongue and lifting it forward slightly. (b) Nasopharyngeal airway. To measure an appropriately sized nasopharyngeal airway, it is traditionally taught that the correct size is related to the patient’s little finger or nostril. However these measures may not relate to the internal anatomy of their nose so it may more appropriate to size the airway by the patient’s sex and height.457,458

539

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 540

540

Chapter 11 What are this patient’s problems?

Thorax

≥10s Abdomen

Nasal Flow

≥10s

Fig. 11.3 Diagram showing Cheyne– Stokes breathing (courtesy of A Rowat); upper two traces show muscle activity, lower trace air flow.

Do not mistakenly attribute apnoeic spells due to intermittent airway obstruction to periodic or Cheyne–Stokes respiration.

simply reflects co-existing severe cardiorespiratory disease occurring in severe strokes.3 Abnormal patterns of breathing

11.2.2 Inadequate breathing Strokes may weaken the intercostal muscles and the diaphragm, leading to reduced ventilation, poor cough and an increased risk of pneumonia.1 Furthermore, the brain lesion itself may directly, or more often indirectly, impair the function of the respiratory centre in the medulla which results in various disordered patterns of breathing (Fig. 11.3) while the patient is awake, but much more commonly during sleep.2 Hypoxia Hypoxic episodes (defined as saturations of < 90% for > 10% of monitoring period) have been described in about one-fifth of stroke patients within the first few hours of admission.3 They occur more commonly during transfers between hospital departments and are associated with greater stroke severity and pre-existing cardiorespiratory disease. In patients with cardiorespiratory disease, hypoxia may be more marked in the supine position and reduced by sitting.4 Hypoxia is associated with poorer survival but it is unclear whether this is an effect of hypoxia exacerbating the brain damage, or

The abnormal patterns of breathing associated with stroke include obstructive and central sleep apnoea, periodic respiration (Cheyne–Stokes), hyperventilation (‘forced respiration’), irregular (ataxic) breathing, apneustic (held in expiration) breathing and, ultimately, complete apnoea.5 Sleep apnoea is the commonest abnormality and has been identified in up to two-thirds of hospital-admitted patients, depending on the definitions and detection methods used. It is usually ‘obstructive’ although may occasionally be ‘central’. Obstructive sleep apnoea has been associated with greater age, obesity, large neck circumference, and in some studies greater stroke severity.2 It is associated with poorer survival and functional outcome but it is unclear whether this is due to incomplete adjustment for stroke severity, the associated hypoxia or fluctuations in blood pressure and cerebral perfusion.6,7 Periodic respiration (Cheyne–Stokes), where there are regular alternating phases of hyperventilation and hypoventilation, occurs in about a quarter of patients although it is seldom recognized by healthcare staff.8 Other neurological functions, e.g. wakefulness, often vary in phase with breathing. The mechanisms of periodic respiration are still debated but it probably reflects a change in the sensitivity of the brainstem respiratory

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 541

11.2 Airway, breathing and circulation

centre to the arterial pressure of carbon dioxide and/or slowed central circulation, with a resultant delay in the feedback loop controlling respiration.9,10 Changes in oxygenation, pH and cerebral blood flow occur in periodic respiration but their importance is unclear. Periodic respiration is associated with greater stroke severity and poorer outcomes although it may be present in patients who are alert and who subsequently make a reasonable recovery.8 Periodic respiration (Cheyne–Stokes) does not necessarily imply a hopeless prognosis. Co-existing cardiopulmonary disease (e.g. chronic obstructive pulmonary disease, pneumonia) is probably a more frequent cause of inadequate ventilation than the abnormal breathing patterns due to the stroke itself. Its detection depends on a thorough initial assessment including an adequate history, physical examination and some simple investigations, e.g. chest X-ray and electrocardiogram (section 10.3.2). Assessment The adequacy of ventilation should be assessed clinically by checking for central cyanosis and examining the chest. We increasingly use pulse oximetry to assess and monitor our patients to detect hypoxaemia. Finger probes are more accurate than those on the ear, and it probably does not matter whether the probe is placed on the paretic or non-paretic hand.11,12 Arterial blood gas analysis is useful in monitoring therapy in patients with type II respiratory failure, those requiring artificial ventilation and those with severe metabolic disturbances such as acidosis. It is easy to miss abnormalities in breathing pattern. Encourage all members of the team to observe the breathing pattern while they are assessing the patient.

Treatment There are currently no data to support the routine administration of increased concentrations of oxygen to patients who are not hypoxic so do not automatically give oxygen to every admitted stroke patient.13 However, if the patient has an oxygen saturation of < 95% in the first couple of days after stroke onset it seems sensible (although not supported by strong evidence) to increase incrementally the inspired oxygen concentration to achieve a saturation of > 95% and so reduce the the likelihood that any ischaemic but still viable brain tissue is

..

..

adversely affected. Patients prefer oxygen delivered via nasal prongs rather than with a face mask.14 In patients with chronic lung disease and a tendency to hypercapnoea, 24% oxygen should be used initially with careful monitoring of the patient’s respiration, neurological function (section 11.5) and arterial blood gases. Although continuous positive airways pressure (CPAP) has been shown to reduce symptoms of obstructive sleep apnoea in general, in patients with recent stroke compliance with treatment tends to be very poor and there is no reliable evidence of benefit.15–17 Treatment of any coexisting lung or cardiac disease should of course be given. Sitting the patient up may help to improve oxygenation.4 If patients are sat out of bed, it is important that they are well supported in an upright position and not slumped in the chair, which worsens ventilation. Sitting may also reduce the intracranial pressure but may cause other problems (section 11.11). Sedative drugs, given to promote sleep, to facilitate imaging or to control seizures, may precipitate periodic respiration or respiratory failure and should generally be avoided if at all possible. Tracheal intubation and mechanical ventilation may be used to maintain ventilation, or to reduce intracranial pressure (section 12.7.4). Indications vary between centres but generally include: deteriorating conscious level, severe hypoxia or hypercapnoea, and inability to maintain an airway. Depending on the indications and case mix, reported frequency of in-hospital case fatality among patients undergoing mechanical ventilation varies between 50 and 90%. Older age, evidence of brainstem dysfunction and comorbidities are predictably associated with worse outcomes; however, a small proportion of survivors make a reasonable recovery.18–20 Often, such aggressive management is considered inappropriate because of the patient’s low probability of regaining a good quality of life (section 10.4). If patients are nursed in a sitting position, it is best if they are well supported and upright in a chair rather than slumped in bed or a chair, which makes breathing more difficult.

11.2.3 Poor circulation Both severe brainstem dysfunction and massive subarachnoid haemorrhage can cause major circulatory problems such as neurogenic pulmonary oedema, cardiac dysrhythmias and erratic blood pressure with severe hypertension or hypotension (section 14.9.3 and 14.9.4). More frequently, circulatory failure, with or without hypotension or hypoxia related to pulmonary oedema, is due to

541

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 542

542

Chapter 11 What are this patient’s problems?

coexistent heart disease (e.g. congestive cardiac failure, myocardial infarction, atrial fibrillation), hypovolaemia (e.g. dehydration, bleeding) or severe infection. Cardiac dysrhythmias Cardiac dysrhythmias are quite common after an acute stroke, although they rarely cause major problems.21 The most clinically important, in terms of frequency and effect on management, is atrial fibrillation which occurs in about one-fifth of patients with ischaemic stroke and one in ten of those with intracerebral haemorrhage.22,23 In most cases, atrial fibrillation precedes the stroke and in some it is the most likely cause of the stroke (section 6.5.1). Of course, a small proportion of ischaemic strokes, perhaps 5%, occur in the context of a recent (within 6 weeks) myocardial infarction (section 7.10).24 Routine monitoring of the patient’s cardiac rhythm, either at the bedside or preferably with ambulatory systems which do not inhibit early mobilization, will identify abnormalities, most commonly paroxysmal atrial fibrillation. The yield will depend on the duration of monitoring. A 24-h ECG monitor might detect previously undected atrial fibrillation in up to 5% and 7-day monitoring in a further 5% of hospitalized patients with ischaemic stroke.25,26 The detection of atrial fibrillation will influence decisions about anticoagulation for secondary prevention (section 16.6). The feasibility and cost of routinely monitoring rhythm inevitably varies between centres and it is unclear what the yield is in terms of altering management, perhaps by starting anticoagulation and preventing recurrences. Certainly, as a minimum we would advise monitoring those with palpitations, syncope, unexplained breathlessness or recent myocardial infarction. Cardiac dysrhythmias, other than atrial fibrillation, are quite common after an acute stroke, but seldom seem to be a problem unless they are due not to the stroke itself but to a recent myocardial infarction.

Myocardial injury It is important to identify the presence and likely cause of any circulatory failure, by clinical assessment of the cardiovascular system backed up with relevant investigations (e.g. cardiac enzymes, electrocardiogram (ECG) and echocardiography). Repolarization abnormalities and ischaemic-like changes on the ECG occur in about three-quarters of patients with subarachnoid haemorrhage but are far less common with other types of stroke.

In SAH the changes are most often secondary to the cerebral insult while in other stroke types they usually reflect associated cardiac disease.27 The specificity of ECG changes for acute myocardial infarction in the context of an acute stroke is low.28 Levels of plasma creatine kinase (CK) and of its more cardiac-specific isoenzyme CK muscle–brain (MB) are quite often raised but the interpretation may be difficult where the patient has fallen, been injured, lain on a hard floor or had a generalized seizure which may all increase the CK level. Indeed even when CK-MB is raised, troponins, which are more specific markers of myocardial injury, are often normal indicating that even CK-MB may not be of cardiac origin.29 Raised troponins have been reported in up to a third of patients with stroke and may be markers of a concurrent myocardial infarction or small areas of myocardial necrosis, which have been identified at autopsy.30 Raised troponins are associated with poorer survival even having adjusted for age and stroke severity.27,31,32 Patients with circulatory failure, atrial fibrillation and recent myocardial infarction have a poor prognosis for survival and functional recovery after stroke.22,33 Active treatment of these problems is likely to improve the outcome but a detailed account is beyond the scope of this book. Gastrointestinal haemorrhage About 3% of stroke patients have an upper gastrointestinal bleed in the first few weeks.34 Bleeding most commonly occurs from petechiae, erosions and ulcers.35 Bleeds are more common in the elderly, in more severe strokes and those with intracerebral haemorrhage; not surprisingly, therefore, gastrointestinal haemorrhage has been associated with high case fatality. Bleeding appears to be more common in patients fed via an enteral tube and may be more common with nasogastric, rather than percutaneous endoscopic, gastrostomy.36 Clearly, resulting hypotension and anaemia might both exacerbate cerebral ischaemia and worsen the neurological outcome. One should obviously be aware of this bleeding potential when prescribing antithrombotic and antiinflammatory medication.37 We try to avoid using nonsteroidal anti-inflammatory drugs (NSAIDs) unless there is a strong indication. Proton pump inhibitors (PPI), double dose H2 receptor antagonists and misoprostol have all been shown to reduce gastric and duodenal ulcers in patients (not stroke) taking NSAIDs chronically.38 A meta-analysis of 20 trials with a total of 2624 mainly Chinese patients suggested an 80% odds reduction in stress-related gastrointestinal haemorrhage among stroke patients and even a reduction in mortality associated with PPI and H2 receptor antagonists.39 However, the rate of bleeding in the control groups of

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 543

11.3 Reduced level of consciousness Table 11.2 The frequency of various clinical features among 675 patients with a first-ever-in-a-lifetime stroke at their initial assessment by a study neurologist (from the Oxfordshire Community Stroke Project, unpublished data). Clinical feature

Number (%)

Not assessable (%)

Glasgow Coma Scale (motor < 6) Glasgow Coma Scale (eyes < 4) Glasgow Coma Scale (eyes + motor < 10) High blood pressure (systolic > 160 mmHg) Very high blood pressure (systolic > 200 mmHg) High blood pressure (diastolic > 100 mmHg) Very high blood pressure (diastolic > 120 mmHg) Epileptic seizures within 24 h Facial weakness Arm or hand weakness Leg weakness Unilateral weakness of at least two of face, arm and leg Sensory loss (proprioception) Sensory loss (spinothalamic) Homonymous visual field defect Gaze palsy Mental test score < 8/10433 Visuospatial dysfunction Dysphasia Dysarthria

86 (13) 98 (15) 111 (16) 311 (46) 90 (13) 131 (19) 23 (3) 14 (2) 256 (38) 344 (51) 307 (45) 331 (49) 101 (15) 196 (29) 113 (18) 50 (7) 85 (13) 81 (12) 122 (19) 135 (20)

15 (2) 16 (2) 16 (2) 19 (3) 19 (3) 22 (3) 22 (3) 0 (0) 40 (6) 32 (5) 34 (5) 0 (0) 168 (25) 140 (21) 134 (20) 36 (5) 135 (20) 179 (27) 111 (16) 127 (19)

these trials was far higher than that reported in observational studies in white populations. Management of active gastrointestinal haemorrhage is the same as in other situations.

been argued that differences in pressure between brain compartments may be more important than the overall pressure. However, treatable complications of the stroke may also depress conscious level (Table 11.3). Level of consciousness is an important indicator of the severity of the stroke and a valuable prognostic variable44 (section 10.2.7).

11.3 Reduced level of consciousness

Table 11.3 Remediable causes of a reduced level of consciousness after stroke.

About 15% of stroke patients have a reduced level of consciousness during the first few days after the stroke (Table 11.2). The causes are not entirely clear (Table 11.3). It may arise from direct damage (haemorrhage or infarction) to the brainstem or more commonly indirect damage from supratentorial lesions associated with brain swelling and midline shift.40,41 However, we not infrequently see patients with large ischaemic, apparently unilateral, hemispheric strokes who become drowsy within hours of onset, before one would expect enough oedema to have developed to raise intracranial pressure markedly, and with little or no apparent mass effect on brain imaging.42 Indeed, in one study, only a minority of patients whose conscious level deteriorated after a large hemispheric ischaemic stroke had globally raised intracranial pressure on invasive monitoring.43 It has

..

..

Causes General Hypoxia Hypotension Severe infection Electrolyte imbalance Hypoglycaemia Drugs: e.g. benzodiazepines, opiate analgesics Neurological Epileptic seizures Raised intracranial pressure Obstructive hydrocephalus (e.g. due to cerebellar haematoma) Cerebral ischaemia after subarachnoid haemorrhage

Section

11.2.2 11.2.3 11.12 11.18.1 and 11.18.2 11.18.4

11.8 12.7, 13.3.2 13.3.6, 14.8 14.5, 14.7

543

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 544

544

Chapter 11 What are this patient’s problems?

Assessment The most commonly used measure of level of consciousness is the Glasgow Coma Scale (GCS) (Table 3.12), which reliably documents the patient’s spontaneous actions as well as those in response to verbal and painful stimuli.45 Although useful, the GCS, which was originally designed for patients with head injury, has a number of problems when applied to stroke. The most important is that many patients with stroke are aphasic and therefore have a reduced score on the verbal response scale. Thus, a patient may have a reduced total score on the GCS but a normal level of consciousness. Another problem is that inexperienced users may apply the motor scale to the affected arm in a patient with a hemiparesis and obtain a reduced motor response rather than the ‘best’ response. The GCS communicates most information if the component scores are recorded separately. Alternative measures include the reaction level scale and components of so-called ‘stroke scales’ (sections 10.2.7 and 17.12.5).46 Patients with a reduced level of consciousness are at greater risk of developing various complications (Table 11.4) so expert nursing care is needed to minimize these risks. Also, a falling level of consciousness may alert one to an important and potentially reversible condition (Table 11.3). Therefore, patients with severe strokes, in whom active treatment would be considered, should be monitored with the GCS for the first few days, especially where staff change frequently, to avoid delays in initiating investigation and treatment of any worsening in their condition. We usually reduce the frequency of monitoring, e.g. hourly, 2, 3 or 6-hourly, over the first few days after the stroke. The speed at which we do this will depend on the team’s judgement about the likelihood of deterioration and the appropriateness and urgency of any intervention aimed at reversing the deterioration. Also, the availability of nursing staff to perform the measurements and the patient’s need for Table 11.4 Common complications in patients with a depressed level of consciousness after stroke. Complication

Section

Airway obstruction Aspiration Fever and infection Dehydration Malnutrition Urinary incontinence Faecal incontinence or constipation Pressure ulcers

11.2.1 11.17 11.12 11.18.1 11.19 11.14 11.15 11.16

unbroken sleep have to be taken into account. Therefore, one might monitor a patient’s GCS hourly during the first 2 days after a cerebellar haematoma (in whom decompressive surgery would be considered) but only 6-hourly in a patient who is stable 2 days after a mild ischaemic stroke. The Glasgow Coma Scale is a useful tool for monitoring patients only if it is applied properly. Staff need to be trained in its proper use. Our patients are frequently said by medical and nursing staff to be confused or aphasic with a GCS of 15/15. Clearly this is nonsense! The investigation and treatment of patients with a falling level of consciousness are similar to those with other indicators of worsening and are discussed in section 11.5. The specific treatment of patients with raised intracranial pressure is discussed in sections 12.7 and 13.3.2.

11.4 Severe stroke vs apparently severe stroke

It is important to be aware that severe concurrent illness, or stroke complications, may make a stroke appear much worse than it really is (Table 11.5). The clinical features which might indicate a severe stroke (e.g. reduced level of consciousness), and so a poor prognosis, may in fact be due to the complicating illness rather than the stroke itself. For example, a patient with a pure motor stroke (lacunar syndrome) (section 4.3.2) and a severe chest infection may be drowsy or confused. It is then difficult to distinguish this clinical picture from that due to a major middle cerebral artery territory infarction (section 4.3.4). Of course, an infection is more easily treated than a large volume of necrotic brain, so with appropriate therapy the prognosis of the two types of patient will be quite different. A thorough general examination will identify signs such as fever, confusion, increased respiratory rate and reduced oxygen saturation, and usually reveals any relevant coexisting disorder. Simple investigations such as a white blood cell count, C-reactive protein, urea and electrolytes, urine microscopy and culture, chest X-ray, ECG and blood cultures are useful, not only to identify the cause of the stroke (sections 6.8.1 and 6.8.2), but also to alert one to serious coexisting non-stroke disease. Seizures that have occurred since stroke onset make the

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 545

11.5 Worsening after a stroke Table 11.5 Causes of apparently severe strokes. Section Non-neurological Infection Respiratory Urinary Septicaemia Metabolic Dehydration Electrolyte disturbance Hypoglycaemia Drugs Major and minor tranquillizers Baclofen Lithium toxicity Anti-epileptic drug toxicity Anti-emetics Hypoxia Pneumonia/chest infection Pulmonary embolism Chronic pulmonary disease Pulmonary oedema Hypercapnoea Chronic pulmonary disease Limb or bowel ischaemia in patients with a cardiac or aortic arch source of embolism Neurological Obstructive hydrocephalus in patients with stroke in the posterior fossa, or subarachnoid haemorrhage Epileptic seizures, including complex partial seizures

11.12.1 11.12.2

11.18.1 11.18.2, 11.18.3 11.18.4 11.29 11.20 11.8 11.9 11.12.1 11.13 11.6 11.2.3 11.2.2

13.3.6, 12.7.3 and 14.8 11.8

assessment of stroke severity particularly difficult (section 11.8). It is also important not to overlook the possibility that any decreased conscious level is due to sedative drugs. The treatment of apparently severe strokes will depend on the specific problem identified or suspected. A 70-year-old man was found unconscious at home and admitted to hospital. In the receiving unit he had a partial seizure affecting his right side with secondary generalization. Subsequent neurological examination revealed a reduced conscious level, deviated gaze and a flaccid paralysis of the right side. Brain CT scan on the day of admission was normal. A diagnosis of a severe ischaemic stroke was made and the decision not to resuscitate was recorded in the notes. He was given intravenous fluids but no anti-epileptic drugs. He was later reviewed by the stroke physician who, having found occasional twitching movements of the right hand,

..

..

reversed the decision not to resuscitate and arranged an EEG which showed status epilepticus. Following intravenous phenytoin the patient improved and 4 days later was discharged from hospital without residual neurological deficits. Therefore, do not ‘write off’ people with apparently severe strokes and recent seizures.

11.5 Worsening after a stroke

Although stroke onset is usually abrupt, the patient’s neurological condition may worsen hours, days or, rarely, weeks after the initial assessment (as well as improve). Patients may have a reducing level of consciousness, worsening of existing neurological deficits, or new deficits indicating dysfunction in another part of the brain. A large number of terms including ‘strokein-evolution’, ‘stroke-in-progression’, ‘early neurological deterioration’ and ‘progressing stroke’ have in the past been applied to this situation, which probably reflects the considerable level of interest in the problem and uncertainty as to its causes.47 After all, here is a situation where one might be able to intervene to prevent a major stroke. One well-validated definition of ‘progressing stroke’ is based on a deterioration in the Scandinavian Stroke Scale48 (Table 11.6) of at least two points on the conscious level, arm, leg or eye movement subscales and/or at least three points on the speech subscale between baseline and day 3.49,50 The same change between consecutive measurements within the first 3 days was referred to as an ‘early deterioration episode’. Clearly, if we accept that the neurological deficit commonly increases over minutes or hours, then the earlier in the course of the stroke we first see the patient, the more likely we are to observe subsequent worsening. Indeed, it is likely that, as we attempt to introduce acute treatments for stroke, we will become much more aware of the ‘normal’ worsening over the first few hours after stroke onset. We will probably see patients earlier in the development of their stroke and monitor them more closely because they will be in clinical trials, or given a specific treatment for acute stroke. Not surprisingly therefore, estimates of the frequency of worsening in the first day or two have varied considerably but have been reported in up to 40% of hospital-admitted patients.51 Early worsening is more likely to be due to a neurological mechanism than a systemic complication of the stroke.52 The factors underlying the progression of neurological deficits in the first day or two after stroke

545

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 546

546

Chapter 11 What are this patient’s problems? Table 11.6 Scandinavian Stroke Scale.48 Consciousness: – fully conscious – somnolent, can be awaked to full consciousness – reacts to verbal command, but is not fully conscious – no reaction to verbal command Eye movement: – no gaze palsy – gaze palsy present – conjugate eye deviation Arm, motor power:* – raises arm with normal strength – raises arm with reduced strength – raises arm with flexion in elbow – can move, but not against gravity – paralysis Hand, motor power *: – normal strength – reduced strength in full range – some movement, fingertips do not reach palm – paralysis Leg, motor power *: – normal strength – raises straight leg with reduced strength – raises leg with flexion of knee – can move, but not against gravity – paralysis Orientation: – correct for time, place and person – two of these – one of these – completely disorientated Speech: – no aphasia – limited vocabulary or incoherent speech – more than yes/no, but not longer sentences – only yes/no or less Facial palsy: – none/dubious – present Gait: – walks 5 m without aids – walks with aids – walks with help of another person – sits without support – bedridden/wheelchair TOTAL


6 4 2 0 4 2 0 6 5 4 2 0










6 4 2 0




6 5 4 2 0




onset are unclear.53 For example, what causes the progression of deficits in some patients with lacunar infarction?54 Early worsening is likely to reflect complex interactions between biochemical and haemodynamic factors which are known to be important in the development of ischaemic stroke (sections 12.1.3 and 12.1.4). A history of diabetes or coronary heart disease, low and high arterial blood pressure, early computed tomography (CT) signs of infarction, evidence of siphon or middle cerebral artery occlusion, various biochemical (e.g. glucose and glutamate) and haematological (e.g. D-dimers) parameters have all been associated with a greater risk of early worsening.47,55–57 Greater age, initial stroke severity and the presence of cerebral oedema on scanning have been associated with late deterioration.47 Worsening has a number of potential causes, some reversible, so it is important to detect and treat them early. The literature has tended to emphasize the neurological causes (Table 11.7) but it is important not to miss the non-neurological ones which tend to occur after the first couple of days and which are more treatable. There is, not surprisingly, considerable overlap with the causes of ‘apparently severe stroke’ (Table 11.5). The outcome of patients whose neurological condition worsens after initial presentation is predictably worse than that of patients who remain stable or improve rapidly.

Table 11.7 Causes of worsening after stroke.

6 4 2 0

Section




10 6 3 0 2 0





12 9 6 3 0

*Motor power is assessed only on the affected side.




Neurological Progression/completion of the stroke Extension/early recurrence Haemorrhagic transformation of an infarct Development of oedema around the infarct or haemorrhage* Obstructive hydrocephalus in patients with stroke in the posterior fossa, or after subarachnoid haemorrhage* Epileptic seizures* Delayed ischaemia* (in subarachnoid haemorrhage) Incorrect diagnosis Intracranial tumour* Cerebral abscess* Encephalitis Chronic subdural haematoma* Subdural empyema* Non-neurological* See Table 11.5

16.2 5.7 12.1.5 13.2.6, 12.7.3 and 14.8 11.8 14.5, 14.7 3.4

*Remediable causes of worsening.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 547

11.5 Worsening after a stroke Table 11.8 Parameters to monitor and which can detect worsening. Section Conscious level, i.e. Glasgow Coma Scale Pupillary responses Eye movements Limb movements Stroke scale (e.g. Scandinavian Stroke Scale) Temperature Pulse rate Blood pressure Respiratory rate Pulse oximetry Fluid balance

3.3.2, 11.3 3.3.6 3.3.6 3.3.4 Table 11.6 11.12 11.2.3 11.7 11.2.2, 11.12 11.2.2 11.18.1

Assessment To ensure that clinical worsening is identified as early as possible, i.e. when the potential for therapeutic reversal is usually greatest, it is important to monitor the patient’s condition. Regular measurements such as those in Table 11.8 will usually alert one to any problem. However, experienced nursing staff who have regular close contact with the patient may detect a problem at an early stage before it becomes obvious to other members of the team. Family members, who often spend long periods with ill patients, may also detect subtle but important changes in the patient’s condition and they should be listened to. It is important that the physician caring for the patient encourages free communication so this sort of information is made known to those directing the patient’s care. Although there is a long list of potential causes of worsening, it is important to consider first those that are most readily reversible (Table 11.7). Most can be diagnosed by a clinical assessment supplemented by simple laboratory investigations. An urgent repeat CT brain scan is advisable in some situations, for example deterioration in conscious level in a patient with a cerebellar stroke or subarachnoid haemorrhage who may be developing obstructive hydrocephalus amenable to neurosurgical intervention. In a patient treated with thrombolytic drugs, an abrupt change in blood pressure, new onset of headache and/or vomiting or neurological deterioration should prompt an urgent CT scan to detect intracranial bleeding due to the treatment. The detection of haemorrhagic transformation of infarction in a patient receiving antithrombotic drugs may influence the decision to continue the medication or not, although there is no reliable evidence for the best policy

..

..

in this situation (sections 5.7, 12.4.2 and 12.5.7). A CT or MRI brain scan that shows that the deterioration was due to an early recurrent ischaemic stroke, especially one in a different arterial territory to the initial stroke, may encourage one to investigate further for a proximal arterial or cardiac source of embolism. Treatment Clearly, any treatment will depend on the reason for the worsening (see relevant sections) but it should be emphasized that at present there is little evidence from RCTs to support the use of treatments such as anticoagulants, thrombolysis, neuroprotection, haemodilution and manipulation of blood pressure for the treatment of worsening (Chapter 12). A small trial which randomized 98 patients to either routine care or intensive physiological monitoring showed that the more intensively monitored patients received more intensive treatment (e.g. supplementary oxygen) and deteriorated less frequently.58 This preliminary finding, which needs to be confirmed in larger studies, could have important implications for the management of patients with acute stroke (section 17.6.2). A 65-year-old woman had an acute ischaemic stroke affecting her cerebellum and brainstem. This occurred on a background of a previous occipital ischaemic stroke, impaired renal function and hypertension. After a stormy early course requiring drainage of obstructive hydrocephalus and ventilation on the intensive care unit, she made good progress. She could sit independently, help with washing herself and dressing, and could take a soft diet and fluids safely. She had diplopia, poor balance and complained of vertigo and vomiting which was exacerbated by movement. She was started on regular oral metoclopramide with some relief of these symptoms. She was transferred to a separate rehabilitation ward on a Friday afternoon. Over the weekend she deteriorated. Her speech became very unclear, her swallowing unsafe and she was unable to sit. Infection, metabolic abnormalities and recurrent hydrocephalus were excluded and a repeat MR scan showed no new stroke lesion although this seemed the most likely explanation for her deterioration. The family were advised that the prognosis was poor. The speech therapist, assessing the patient prior to insertion of a percutaneous endoscopic gastrostomy, remarked that the patient’s tongue movements were like those of a patient with Parkinson’s disease – the penny dropped! The drug chart was reviewed, the metoclopramide was withdrawn and the patient returned to her previous functional state over the next week. She was eventually discharged home, walking and requiring minimal help with everyday activities.

547

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 548

548

Chapter 11 What are this patient’s problems? Table 11.9 Frequency of coexisting medical problems among 675 patients with a first-ever-in-a-lifetime stroke (from the Oxfordshire Community Stroke Project, unpublished data).

Previous angina Previous myocardial infarction Cardiac failure Intermittent claudication Diabetes mellitus Previous epileptic seizures Previous malignancy Dependent before stroke (Rankin > 2)

n

(%)

106 112 52 112 63 19 74 103

(16) (17) (8) (17) (9) (3) (11) (15)

No data were available on respiratory or musculoskeletal problems.

which contribute to disability, progress. Thus, a patient may reach their optimal functional recovery some months after a stroke and then deteriorate due to progression of a comorbid condition. If this kind of deterioration can be anticipated, it may allow for a more flexible package of care to cope with such fluctuations. There can be few things more dispiriting than to strive to discharge a patient into one form of accommodation and then hear that within a few months their condition has deteriorated to such an extent that the accommodation is no longer suitable. This can shatter the morale of the patient and their carer. Functional deterioration months after a stroke is unlikely to be due to the initial stroke and much more likely to be caused by a recurrent stroke or the progression of some comorbid condition such as angina, arthritis or intermittent claudication.

11.6 Coexisting medical problems Assessment Coexisting medical problems are common in stroke patients because they are usually elderly and have associated vascular disease (Table 11.9). Although we have already mentioned the importance of cardiorespiratory diseases in the immediate management of the patient (section 11.2), these and other conditions can be important for many other reasons, not least that they may require treatment in their own right. Severe non-stroke illness can make a mild stroke appear severe and thus lead to an inaccurate prognosis and possibly inappropriate treatment (section 11.4). Coexistent cardiorespiratory (e.g. angina, cardiac failure, chronic obstructive pulmonary disease), musculoskeletal (e.g. arthritis, back pain, amputation) and psychiatric conditions (e.g. depression, anxiety) often compound stroke-related disability. So, for example, after several months of rehabilitation, one might have taught a patient with a severe hemiparesis to walk again, but if he or she also has chronic obstructive pulmonary disease the added effort of walking with a hemiplegic gait may mean that the patient cannot walk any useful distance. It is important to be aware of the limitations on rehabilitation imposed by any pre-existing disease before spending months trying to make a patient walk. It may be more realistic to teach the patient to be independent in a wheelchair. Even if one cannot estimate the impact of non-stroke disease on recovery, knowledge of its existence may explain why patients are not achieving their rehabilitation goals (section 10.3.4). Although most patients who survive a stroke improve over weeks or months, many of the coexisting problems,

A thorough history and examination at the time of admission, with reassessment when the patient is more active, should identify the main coexisting medical problems – the patient’s medication will often provide a clue. An assessment of the patient’s pre-stroke functional status (i.e. what could they do and not do?) is invaluable, not only in predicting outcome (section 10.2.7) but also for identifying comorbidities. Unfortunately, this is often not recorded in the medical records unless specifically prompted.59,60 One approach might be to estimate routinely a pre-stroke Barthel Index (section 17.12.5) since this covers most of the important activities of daily living (ADL). This sort of ADL checklist is also useful in making a prognosis and in setting rehabilitation goals since, depending on the cause and duration, the pre-stroke functional impairment will determine the best achievable post-stroke functional status. If a patient has, as a result of arthritis, been immobile for 10 years before the stroke, it is ridiculous to try to get the patient to walk after the stroke. Unfortunately, this is often attempted simply because nobody has obtained an accurate picture of the patient’s pre-stroke function. This is more likely to happen where the patient has difficulties with communication and no carer is available. An assessment of the patient’s pre-stroke function may also be very important in making decisions even within the first few hours after stroke. Although it may be impossible for anyone to judge a person’s quality of life, other than that person themselves, one may deduce something from the patient’s function. This may be important where, for example, one is considering antibiotics

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 549

11.7 High and low blood pressure after stroke

Clearly, one should aim to minimize the effect of any comorbidity by giving as effective treatments as possible. Where it is not possible to influence the disease directly, it is still important to take account of comorbidity in one’s overall approach to the patient. One’s intermediate and long-term goals also have to take this into account (section 10.3.3).

11.7 High and low blood pressure after stroke

Dead within 14 days (%)

Treatment

30

68 66

25

64 20

62

15

60 58

10

56 54

5 160/90 mmHg) before the onset. They will tend to have higher blood pressures than those without previous hypertension and are more likely to have evidence of end-organ damage, e.g. hypertensive retinopathy, impaired renal function and left-ventricular hypertrophy. High blood pressure in the acute phase of both haemorrhagic and ischaemic strokes is associated with a poor outcome, an effect which appears to be independent of age and stroke severity (Fig. 11.4).61,63–68 Assessment There is considerable variation in the methods used and the frequency of monitoring blood pressure after acute stroke. Traditionally, the blood pressure has been measured in the standard way with a sphygmomanometer and an appropriately sized cuff kept at the level of the

..

..

patient’s heart. However, increasingly, semi-automated non-invasive systems are used which allow more frequent monitoring (even when there are few nursing staff) and thus earlier intervention. These systems must be calibrated to avoid misleading readings.69 The presence of atrial fibrillation reduces the reproducibility of both manual and semi-automated blood pressure recordings.70,71 In the intensive care unit, intra-arterial monitoring is frequently used although this can very occasionally cause peripheral ischaemia. Perhaps not surprisingly, more intensive monitoring often leads to more manipulation of blood pressure, the value of which is currently unclear (see below). Semi-automatic blood pressure monitors should be properly calibrated to avoid misleading results. In patients with very low or high blood pressure, where the reading is likely to lead to a significant change in management, we would normally check the reading with a traditional sphygmomanometer. There is no consistent difference between the blood pressure measured in the weak vs the unaffected arm, although there is often a difference between arms which is unrelated to the side of the stroke and probably reflects occlusive arterial disease affecting one arm.72 Therefore, ideally the blood pressure should be checked in both arms on at least one occasion, and if the blood

549

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 550

550

Chapter 11 What are this patient’s problems?

pressures are different, it should be monitored consistently in the arm giving the highest reading to avoid a spurious label of labile hypertension. If the blood pressure is raised on admission it should be monitored to establish whether or not it falls spontaneously and the patient examined to identify evidence of end-organ damage. Treatment There is considerable uncertainty about the risks and benefits of lowering the blood pressure in the acute phase of stroke. Treatment may theoretically reduce the likelihood of rebleeding in intracerebral and subarachnoid haemorrhage, and of brain oedema and haemorrhagic transformation in cerebral infarction. However, lowering the blood pressure may reduce cerebral perfusion where cerebral autoregulation is impaired and thus further increase ischaemic damage73 (section 12.1.2). Intravenous calcium channel blockers which, apart from their potential neuroprotective action, also lower arterial blood pressure have been associated with worse outcomes in several randomized controlled trials (RCTs)74 (section 12.8.3). Small trials have not established how blood pressure should be manipulated in the acute phase of stroke and larger trials are ongoing to determine how and when we should do so61 (http://www.strokecenter.org/ trials/). Until RCTs are available, we offer the following advice: • If the blood pressure remains raised (i.e. > 140/90 mmHg) for more than 1 week, or where there is evidence of end-organ damage, we would give the patient general advice (i.e. salt restriction, weight loss and moderation of alcohol intake) and start an antihypertensive drug, accepting that this timing is arbitrary. The issues of when to start antihypertensive treatment, the blood pressure level and the choice of agent, for the purposes of secondary prevention, are discussed in section 16.3. • Our management is similar whatever the pathological type of stroke although we would be less inclined to lower blood pressure if the patient has severe stenoses or occlusions of the carotid and/or vertebral arteries. • Where the patient is already on antihypertensive treatment, pending the results of ongoing RCTs, it seems reasonable to continue it as long as the patient can swallow the tablets safely and has not become hypovolaemic, which may increase the drug effect and lead to marked and potentially damaging hypotension. • If the blood pressure is very high (i.e. > 220/ > 120 mmHg) there may be evidence of organ damage which would prompt earlier initiation of bloodpressure-lowering drugs (Table 11.10).

Table 11.10 Circumstances in which we would consider lowering the blood pressure immediately after an acute stroke. Papilloedema or retinal haemorrhages and exudates Marked renal failure with microscopic haematuria and proteinuria Left ventricular failure diagnosed on clinical features and supported by evidence from the chest X-ray and/or echocardiogram Features of hypertensive encephalopathy, e.g. seizures, reduced conscious level Aortic dissection Note: even these features may be misleading in acute stroke because left heart failure may frequently be due to coexistent ischaemic heart disease, and seizures and drowsiness may be due to the stroke itself.

• The aim of therapy should be a moderate reduction in blood pressure over a day or so, not minutes. An angiotensin-converting enzyme inhibitor (in the absence of contraindications) is a reasonable first choice. Thiazides probably have little effect in the acute phase although they remain a very reasonable choice later.75,76 Transcutaneous glyceryl trinitrate is effective in lowering blood pressure, and the patches can easily be applied in dysphagic patients and withdrawn relatively quickly if blood pressure drops too far.77 Intravenous labetalol, with very careful intraarterial monitoring of blood pressure, may be useful in resistant cases. Do not lower the blood pressure in the first few days after a stroke unless there is evidence of accelerated hypertension or end-organ damage, or the patient is in of one of the ongoing multicentre randomized controlled trials (http://www.strokecenter.org/trials/).

11.7.2 Low blood pressure It is difficult to define low blood pressure since, although one could set an arbitrary lower value for the systolic or mean blood pressure, clinically significant hypotension is that which leads to dysfunction of one or more organs. The level at which this occurs will depend on the patient’s age, their usual blood pressure, the state of their arterial tree and whether cerebral autoregulation is intact or impaired. There is good evidence that autoregulation in the brain is impaired after stroke so that even if a patient has the same blood pressure after stroke as before, cerebral perfusion might still be reduced.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 551

11.8 Epileptic seizures

Assessment The monitoring of blood pressure after stroke has been discussed in section 11.7.1. The assessment of the clinical importance of hypotension should comprise a clinical examination and some simple investigations (e.g. blood urea, arterial blood gases) to identify the features of ‘shock’ mentioned already. Having established that the patient has low blood pressure which is associated with under-perfusion of tissues, it is then important to establish the cause. Is the patient hypovolaemic, due to dehydration (section 11.18.1) or blood loss (section 11.2.3)? Has the patient had a pulmonary embolus (section 11.13), or are they in heart failure (section 11.2.3) or septic (section 11.12)? Is the patient on drugs which could lower blood pressure excessively? These questions can normally be answered following a thorough clinical examination including assessment of the jugular venous pressure, review of the drug and fluid balance charts, and some simple investigations including haemoglobin and haematocrit, neutrophil count, C-reactive protein, urine and blood cultures, cardiac enzymes, an ECG and chest X-ray. Occasionally, further investigation with, for example, a CT pulmonary angiogram, an echocardiogram or measurement of right atrial or pulmonary wedge pressures is required to sort out the cause.

blood pressure in acute stroke.78 Trials of sympathomimetics to raise blood pressure are in progress (http://www.strokecenter.org/trials/).

11.8 Epileptic seizures

11.8.1 Early seizures About 5% of patients have an epileptic seizure within the first week or two of their stroke (so-called onset seizures), most occurring within 24 h. Inevitably, estimates of the frequency of onset seizures vary because of differences in case selection, diagnostic criteria, lack of witnesses and methods of follow-up. Most onset seizures are partial (focal) although often with secondary generalization (section 3.3.10) (Table 11.2). Onset seizures are more common in severe strokes, in haemorrhagic stroke and stroke involving the cerebral cortex.79,80

11.8.2 Later seizures In population-based cohorts, which are relatively unaffected by hospital referral bias, the risk of having a first seizure, excluding onset seizures, is between 3% and 5% in the first year after a stroke and about 1–2% per year thereafter (Fig. 11.5). About 3% will have more than one seizure and could be regarded as having developed

100

This will obviously depend on the cause of the low blood pressure. In our experience, hypovolaemia is the most frequent problem and patients usually improve with intravenous fluids. Obviously, it is important to exclude cardiac failure before giving fluids in this way. There is little evidence on which to base a decision to increase the

..

..

Intracerebral haemorrhage

80 70

Subarachnoid haemorrhage

60 50 40 30 20 10 0

Treatment

Ischaemic stroke

90 % still seizure-free

Unfortunately, judgements about the optimal level of blood pressure after acute stroke are very difficult in routine practice since we have no easy and reliable techniques for assessing organ perfusion and more importantly function, other than the clinical state of the patient. Of course, when blood pressure is very low, patients may show signs of ‘shock’ (e.g. cold extremities, low urine output, worsening renal function, mental confusion, lactic acidosis) and actions to improve organ perfusion are required. In the International Stroke Trial, low blood pressure after acute stroke was associated with poor outcome, even after adjusting for stroke severity (Fig. 11.4).66 However, the low blood pressure might not be the cause of the poor outcome but rather the consequence of important comorbidity (e.g. heart failure, atrial fibrillation) or complications (e.g. dehydration, pulmonary embolism).

1

2

3

4

Time (years) Fig. 11.5 A Kaplan–Meier plot showing the proportion of patients remaining seizure-free at increasing intervals after a first-ever-in-a-lifetime stroke in the Oxfordshire Community Stroke Project. Separate plots are shown for patients with ischaemic stroke (n = 545), non-truamatic intracerebral haemorrhage (n = 66) and subarachnoid haemorrhage (n = 33). Adapted from Burn et al., 1997.80

5

551

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 552

552

Chapter 11 What are this patient’s problems?

epilepsy following a stroke.79 This represents a greatly increased relative risk of seizure (perhaps 20-fold) compared with stroke-free individuals of similar age. Patients with onset seizures, haemorrhagic strokes and infarcts involving the cerebral cortex have the highest overall risk of later seizures. Seizures may recur in about 50% of the patients, but are rarely troublesome if accurately diagnosed and appropriately treated. Patients who become functionally independent and who have not yet had a seizure are at very low risk of post-stroke seizures. The theoretical future risk of seizures is not great enough to prevent the patient driving an ordinary car in the UK where the acceptable risk of a seizure is < 20% per annum. The risk of having an epileptic seizure after a firstever-in-a-lifetime stroke is, on average, about 5% in the first year and 1–2% per year thereafter. However, the risk is higher in patients with haemorrhagic stroke or with large ischaemic strokes involving the cortex, and lower in patients with lacunar and posterior circulation strokes.

Assessment The diagnosis of seizures should, as usual, be based on a detailed description of the attack from the patient, and if possible a witness, and may very occasionally be confirmed by electroencephalography (EEG) during a seizure. Video-EEG monitoring is occasionally useful where patients are having frequent attacks of uncertain nature. If patients have seizures in the first few days after the stroke, and especially if they are partial (focal), they should be investigated. Non-stroke brain lesions complicated by post-seizure impairments (e.g. Todd’s paresis) may mimic strokes (section 3.4.2). Also, because the neurological deficits and conscious level may be temporarily much worse immediately after a seizure, ‘onset seizures’ can make the assessment of stroke severity unreliable (section 11.4). We have also seen occasional patients who have had non-convulsive status epilepticus, a diagnosis which can only be confirmed by EEG, who may, for example, be severely aphasic, and who have improved dramatically with anti-epileptic treatment. Furthermore, seizures should not automatically be attributed to the stroke since many other causes may be present coincidentally, or as a consequence of the stroke (Table 11.11); appropriate investigations should be performed to exclude these. Finally, seizures may mimic recurrent stroke if associated with worsening of the original focal deficit; this is particularly confusing if any seizures have been unwitnessed and the patient presents with worsening of their earlier stroke (section 11.5).

Table 11.11 Causes of epileptic seizures after ‘stroke’. Section General Alcohol withdrawal Anti-epileptic drug withdrawal Hypoglycaemia Hyperglycaemia, especially non-ketotic hyperglycaemia Hyper/hyponatraemia Hypocalcaemia or hypomagnesaemia Drugs: Baclofen given for spasticity Antibiotics for infections, e.g. ciprofloxacin Antidepressants given for emotionalism or depression Phenothiazines given for agitation or hiccups Anti-arrhythmics for associated atrial fibrillation Neurological Due to the primary stroke lesion Haemorrhagic transformation of infarction Underlying pathology: Arteriovenous malformation Intracranial venous thrombosis Mitochondrial cytopathy Hypertensive encephalopathy Wrong diagnosis Herpes simplex encephalitis Cerebral abscess Intracranial tumour Subdural empyema

11.18.4 11.18.3 11.18.1 & 11.18.2

11.20 11.12

11.31 11.10 & 11.29.1 11.2.3

5.7

8.2.4 7.21 7.19 3.4.5 3.4

Treatment Any precipitating cause should be treated. Status epilepticus, although rare, should be managed in the normal way.81 There is no evidence to support the routine use of prophylactic anti-epileptic drugs in stroke patients (including those with subarachnoid haemorrhage, section 14.3.3) who have not yet had a seizure but who are thought to be at high risk. Usually, an isolated seizure does not require treatment since the risk of further seizures is only about 50% over the next few years. However, if seizures recur, or if the patient wishes to minimize the risk of recurrence because of the implications for driving, employment or leisure activities, treatment after the first seizure may be warranted. We are not aware of any studies which have compared the efficacy of different anti-epileptic drugs in preventing seizures

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 553

11.9 Headache, nausea and vomiting

after stroke. Indeed, until recently there has been little evidence to suggest that any one of the most commonly used first-line drugs (i.e. phenytoin, sodium valproate, carbamazepine and lamotrigine) is more effective than any other in preventing partial (focal) or generalized seizures in adults.82,83,461,462 The choice will depend on availability, efficacy the risk of adverse effects and cost. Recent evidence suggests that sodium valproate is the most effective agent in generalized seizures and lamotrigine in partial seizures.461,462 Patients who have had a seizure should be advised not to drive (for a period which varies depending on national regulations), and to inform the necessary authorities (www.dvla.gov.uk/at_a_glance/ ch1_neurological.htm, www.epilepsy.com/epilepsy/ social_driving.html, http://www.ltsa.govt.nz/factsheets/ 17.html). However, many stroke patients who have seizures are too disabled to drive anyway (section 11.32.2). The risk of seizures after acute stroke, including subarachnoid haemorrhage, is not high enough to justify routine prophylactic anti-epileptic drugs. These may have significant adverse effects and are expensive.

vertebrobasilar strokes. It may not respond to antiemetics, indeed some believe that these may prevent tolerance developing. The place of so-called vestibular rehabilitation in such cases is unclear.

11.10 Hiccups

Hiccups are due to involuntary diaphragmatic contractions against a closed glottis. The precise cause is unclear. Although uncommon after stroke they may be persistent and troublesome in patients with strokes affecting the medulla.89 If they persist, other causes should be considered (e.g. uraemia, diaphragmatic irritation). Numerous folk cures (e.g. sudden frights), acupuncture and drugs (Table 11.12) have been suggested as effective treatments.90,91 It may be worth trying a short trial of each of the drugs in Table 11.12 when hiccups are persistent and distressing to the patient since any response is likely to be rapid and may not require ongoing drug treatment. Chlorpromazine, baclofen and gabapentin are probably the most frequently used. The choice should take into account the likely adverse effects.

11.9 Headache, nausea and vomiting

Headache is quite a common symptom at the time of stroke and may provide clues to the pathological type (e.g. haemorrhagic) and cause (e.g. giant-cell arteritis) (section 6.7.6). It is often associated with nausea or vomiting, most commonly in haemorrhagic and cerebellar strokes and less often in lacunar infarcts.84–88 It is more common in those with prior migraine. Headache which occurs after the onset of stroke may be caused by treatment, for example due to intracranial bleeding secondary to thrombolysis or more commonly dipyridamole started for secondary prevention (section 16.5.5). Nausea or vomiting without headache is generally secondary to vertigo. These symptoms, which can be severe initially, usually improve within a few days. Having established that there is no important underlying cause (e.g. venous sinus thrombosis, giantcell arteritis, arterial dissection) most patients simply require reassurance that the symptoms will improve, adequate analgesia and/or anti-emetics. However, these drugs may have adverse effects (see case history in section 11.5). Persistent vertigo which may be positional and associated with nausea or vomiting can be a particularly troublesome symptom, most commonly after

..

..

11.11 Immobility and poor positioning

Immobility is a major consequence of impaired conscious level; severe motor deficits including weakness, ataxia and apraxia; and less commonly of sensory (i.e. proprioceptive) and visuospatial deficits. Immobility makes the patient vulnerable to a number of complications such as infections (section 11.12), deep venous Table 11.12 Some of the drugs used to treat hiccups (most have important adverse effects). Chlorpromazine Haloperidol Gabapentin Baclofen Metoclopramide Sodium valproate Phenytoin Carbamazepine Nifedipine Amitriptyline

553

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 554

Section Maintenance of a clear airway Cerebral perfusion Cerebral oedema Muscle tone Swallowing Limb oedema Stimulation

11.2.1 11.7.2 11.3 11.20 11.19 11.25 11.31.3

28.0 19.7

P = 0.0001

20.8

11.4 14.1

CBP (mmHg)

Table 11.13 Physiological factors to take into account when deciding in which position a patient should be nursed.

ICP (mmHg)

Chapter 11 What are this patient’s problems?

104.1 7.4 89.5

P = 0.006 98.8

74.9

84.3 69.8

CPP (mmHg)

thrombosis and pulmonary embolism (section 11.13), pressure ulcers (section 11.16), contractures (section 11.20) and falls and resulting injuries (section 11.26). Immobile patients are unable to position themselves to maintain comfort, to facilitate activities such as drinking and passing urine, and to relieve pressure over bony prominences. They are in the ignominious situation of always having to ask others to position them. There has been little formal research of positioning after stroke.92,93 For example, should the patient be nursed sitting or lying, and if lying, on which side? There is some evidence for the optimum positioning of mechanically ventilated patients: nursing semirecumbent is associated with a lower risk of pneumonia.94 In the self-ventilating stroke patient the optimal position(s) are unclear but decisions should take into account the following physiological factors (Table 11.13): • Maintenance of a clear airway: In unconscious patients correct positioning is vital to maintain a clear airway and so reduce the risk of aspiration (section 11.2.1). • Oxygenation: Position may influence the patient’s ability to breathe, ventilate their lungs and oxygenate their blood (section 11.2.2). Oxygenation is usually optimal in the sitting position.92,95 • Cerebral perfusion: Lying flat may increase cerebral perfusion.96 Also, patients who are hypovolaemic may have reduced systemic blood pressure when sitting, which may reduce cerebral perfusion and possibly increase cerebral ischaemia.92,97 However, if intracranial pressure falls further on sitting, cerebral perfusion pressure may increase. • Cerebral oedema: Intracranial pressure is highly dependent on posture, being higher in supine patients and lower when sitting.92,98 The relationships between oxygenation, cerebral perfusion, intracranial pressure and cerebral blood flow are so complex that it is difficult to predict the optimum position for nursing an individual patient (Fig. 11.6). Whether our own anecdotal observation that some patients with severe strokes are more alert when sitting up than when lying

CBF (mL/100 g brain per min)

554

88.4

88.3 P = 0.8

69.7

70.2

51.0

52.1

69.3 48.9

64.7 P = 0.66

28.5

47.8 30.9

0° Position

30° head up

Fig. 11.6 Graph showing the interrelationships between intracranial pressure (ICP), carotid blood pressure (CBP), cerebral perfusion pressure (CPP) and cerebral blood flow (CBF) with changes in posture in patients with head injury. Data from Feldman et al., 1992.98

is explained by reduced intracranial pressure, reduced cerebral oedema, improved oxygenation, and/or increased sensory or social stimulation is not at all clear. • Tone: The position of a patient influences the tone in their trunk and limbs.99 Positioning of the patient is used by physiotherapists and nurses to promote higher or lower tone, whichever seems most appropriate for that particular patient.100 Spasticity and the tendency to develop contractures may be reduced by careful positioning to reduce tone (section 11.20), while patients who have low tone in their truncal muscles may benefit from positioning which promotes increased tone and leads to better truncal control. The positioning charts which are so often strategically placed by the patient’s bed to guide nursing care give the

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 555

11.12 Fever and infection

..

..

(Fig. 11.23) are, in our experience, valuable tools in positioning immobile stroke patients and they potentially allow some patients to reposition themselves to maintain comfort.

impression that we know what we are doing (see Fig. 11.22). However, although there is some agreement about the optimum positioning of patients with hemiplegia (e.g. fingers extended, spine straight), there are many areas of uncertainty (e.g. optimum position of the head, foot and unaffected limbs).100 There is clearly a need for further research into this area although this will be hampered by difficulties in prescribing specific positioning regimens, in ensuring compliance, and in reliably assessing outcomes (section 11.20). • Swallowing: Swallowing is easier and safer if the patient is sitting up with their neck flexed (section 11.17). • Pressure area care: For immobile patients who cannot shift their own weight, their position must be changed frequently enough to avoid developing ischaemia of the skin and subcutaneous tissues over bony prominences at weight-bearing points (section 11.16). • Limb oedema: The ankles and the paralysed arm of immobile hemiparetic patients frequently become oedematous and painful. This may increase muscle tone and further reduce function (section 11.25). • Stimulation: It is difficult for patients to see what is going on around them when lying flat. This lack of sensory and social stimulation and contact with daily events will encourage too much sleep during the day and may lead to boredom, reduced morale and sometimes confusion (section 11.31.3). The team should assess (and regularly reassess) each patient and decide which of these potential problems are most important. For example, is the patient hypoxic or dysphagic or at particular risk of pressure ulcers? Depending on this assessment, a positioning regimen which sets out what are thought to be the best positions, and the frequency of repositioning, should be prescribed and re-evaluated. However, educational programmes for nursing staff are probably only moderately effective in increasing the proportion of time that patients are positioned according to the prescribed regime.101 Equipment such as electrically operated multi-sectional profiling beds (Fig. 11.7) and specialist rehabilitation chairs

Fever is quite common after stroke although its frequency obviously depends on the population of stroke patients studied, the definitions used, and the method, timing and duration of monitoring. All published studies have been of hospital-referred patients, most have defined fever as an axillary or rectal temperature of > 37.5°C, and have monitored temperature for 2–7 days. Patients with fever during the first few days after stroke have a worse outcome than those without.102,103 It is unclear whether this adverse prognosis is simply because fever is a marker of a severe stroke (i.e. due to loss of central temperature control or resorption of subarachnoid blood), an indicator of an infective complication (e.g. pneumonia or urinary infection), or increases cerebral damage. The last is an attractive, although unproven concept, since it is consistent with research in animal models which has shown that hyperthermia increases and hypothermia decreases ischaemic cerebral damage.104 Table 11.14 lists some of the potential causes of fever after stroke, of which infection is probably the

Fig. 11.7 A multi-sectional profiling bed – a useful tool for positioning stroke patients. It even helps assess patient

hydration by facilitating examination of the jugular venous pressure, and testing for postural hypotension.

Electrically operated multi-sectional profiling beds are helpful in positioning patients and may provide some immobile patients with the opportunity to reposition themselves. We also find them helpful for examining the jugular venous pressure and testing for postural hypotension, both essential to assessing patient fluid status.

11.12 Fever and infection

555

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 556

556

Chapter 11 What are this patient’s problems? Table 11.14 Causes of fever after stroke. Causes Infective/inflammatory complications of the stroke Urinary infection Pneumonia Deep venous thrombosis Pulmonary embolism Pressure ulcers Infected intravenous access site Vascular problems, e.g. infarction of myocardium, bowel or limb Inflammatory causes of the stroke itself Infective endocarditis Arteritis Coincidental conditions Upper respiratory tract infection Drug allergy

accumulation of secretions, and care to avoid aspiration seem sensible measures to take (section 11.17). Section

11.12.2 Urinary infections 11.14 11.12.1 11.13 11.13 11.16

6.5.9 7.3

most common. Fever and infection may pre-date the stroke onset, and occasionally may actually cause or at least precipitate the stroke (sections 6.6.17 and 7.11). Of course, immobile stroke patients are prone to infections, the most common sites being the chest and urinary tract.105 There is some evidence that patients with stroke and other types of brain damage acquire an immune deficiency, mediated via the hypothalamic pituitary axis, which may contribute to their risk of infection.106,107 However, the clinical relevance of these findings has yet to be established. Infections are (in general) associated with worse outcomes, even having adjusted for other important prognostic factors as far as it is possible to do so, and they cause many deaths, interrupt rehabilitation and slow recovery.108

11.12.1 Chest infections Chest infections are much more common in the acute stage than later, occurring in about 20% of patients within the first month or two. Chest infections may be due to aspiration, failure to clear secretions, the patient’s immobility, reduced chest wall or diaphragmatic movement on the hemiparetic side, or comorbidities including chronic airway disease.109 However, there is no clear relationship between the side of any hemiparesis and side of pneumonia. Alterations in the bacterial flora in the mouth which have been observed after stroke110 and in tube-fed individuals111,112 may be associated with an increased risk of chest infection. Chest infections may be minimized by efforts to improve oral hygiene or alter the mouth flora113 but further randomized trials are required to evaluate this approach. In the meantime careful positioning, physiotherapy and suction to avoid the

About a quarter of hospitalized stroke patients develop a urinary infection within the first 2 months and this remains common over the subsequent months.105 Urinary infections can be avoided by maintaining adequate hydration and thus urine output, and by avoiding unnecessary bladder catheterization (section 11.14). Given that incomplete bladder emptying is associated with an increased risk of infection it seems sensible to avoid constipation and drugs with anticholinergic effects if possible.114 It is uncertain whether intermittent or continuous catheterization and/or prophylactic antibiotics are of benefit for patients with a persisting increased post-void residual volume.115–117 Assessment The patient’s temperature should be monitored at least 6-hourly during the first few days after the stroke and thereafter if there are any signs of infection or functional deterioration. However, fever may not accompany infection, especially in elderly and immunocompromised patients. Therefore, any functional deterioration or failure to attain a rehabilitation goal should prompt a search for occult infection, even if there is no fever. Obviously, the cause of fever should be identified using clinical assessment supplemented with appropriate investigations (e.g. blood neutrophil count, C-reactive protein, cultures of urine and blood, chest X-ray). Dipstick testing of a midstream urine to detect leucocytes and nitrates is insufficiently accurate to be relied on to exclude or confirm a urinary infection (section 11.14).118 Prevention and treatment There is currently little evidence to support the use of prophylactic antipyretics to reduce fever after stroke (section 11.19). In fact, the routine use of paracetamol to lower body temperature and attempts to induce more profound hypothermia after stroke are being evaluated in ongoing RCTs119 (http://www.strokecenter.org/ trials/index.htm). A single-centre RCT of levofloxacin (a fluoroquinolone antibiotic) in acute stroke failed to demonstrate a reduction in infections but further trials will be required to establish whether alternative antibiotic regimes might be more effective.120,121 The treatment of fever will depend on the cause (Table 11.14). We quite often start a broad-spectrum antibiotic, once specimens for microbiological testing have been taken,

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 557

11.13 Venous thromboembolism

(a)

(b) Fig. 11.8 (a) An MR scan of the legs using direct thrombus imaging to demonstrate extensive deep venous thrombosis (arrows). (b) A colour flow Doppler image of the femoral vein demonstrating thrombus within the lumen (arrow); this is

usually associated with an inability to compress the vein and if the vein is occluded a loss of augmentation of flow with respiration or squeezing the calf. (Also reproduced in colour in the plate section.)

since delays in treating infections may impede patients’ progress. However, with an increasing incidence of Clostridium difficile toxin-associated diarrhoea in many hospitals, the risks of early use of broad-spectrum antibiotics must be carefully weighed against the potential benefits.122 Appropriate antibiotics and supportive treatment (e.g. physiotherapy, oxygen) should be given in established infection. Also, it seems reasonable, whatever the cause of the fever, to use a fan and prescribe paracetamol (acetaminophen) since fever may worsen outcome (see above). Even if such interventions appear to be without risk, one must remember that they take up a nurse’s time which might be spent to greater effect in some other activity.

of the frequency depend on the types of patients included, the use of preventive measures, and the timing and method of detection. The most sensitive techniques, such as MRI direct thrombus imaging, detect DVTs in about 40% of hospitalized patients treated with aspirin and graduated compression stockings and about half of these are above-knee123 (Fig. 11.8a). Less sensitive, but more widely available, tests such as compression Doppler ultrasound (Fig. 11.8b) and plethysmography identify DVTs in about 20% of immobile hospitalized patients, above-knee in about half of these. Although DVT is said to be less common among Chinese a recent study has shown a similar incidence to that seen in white patients.124 Clinically apparent DVT confirmed on investigation is less common, occurring in under 5%. DVTs are most often asymptomatic, or unrecognized, but may still lead on to important complications.123 Pulmonary embolism (PE) occurs in about 10% of hospitalized patients but like DVT is frequently not recognized.123 It is an important cause of preventable death after stroke and is a frequent finding at autopsy.125

Any functional deterioration or failure to attain a rehabilitation goal should prompt a search for occult infection.

11.13 Venous thromboembolism

Deep venous thrombosis (DVT) of the legs is common in patients with a recent stroke, particularly older patients with a severe hemiplegia who are immobile. Estimates

..

..

Assessment DVT should be suspected if a patient’s leg becomes swollen, hot or painful or if the patient develops a fever. Unfortunately, the clinical diagnosis can be difficult because many paretic legs become swollen, mostly

557

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 558

558

Chapter 11 What are this patient’s problems?

because of the effects of gravity and lack of movement. If a paretic leg swells while a patient is still being nursed in bed, DVT is a likely cause, but where a patient is sitting out or mobilizing the clinical diagnosis is much less certain. Stroke patients who have communication difficulties, sensory loss or neglect may well not complain of discomfort or swelling, so that clinical detection will depend on the vigilance of members of the multidisciplinary team. Stroke patients who have communication difficulties, sensory loss or neglect may well not complain of discomfort or swelling associated with deep venous thrombosis, so that clinical detection will depend on the vigilance of members of the multidisciplinary team. If a patient develops a swollen leg on a stroke unit, deep venous thrombosis has to be actively excluded. Where the patient develops clinical evidence of a DVT or pulmonary embolism, confirmatory investigations must be carried out if treatment with anticoagulants is being considered. We would normally use compression Doppler ultrasound in the first instance to confirm the diagnosis since this is non-invasive, widely available and reasonably sensitive (>90%) and specific (>90%) in detecting at least above-knee DVT in symptomatic patients.126 However, it is operator dependent and if there is doubt about the result, or if one wishes to exclude thrombosis in the calf veins, contrast X-ray venography should be performed. The clinical diagnosis of deep venous thrombosis in stroke patients is particularly difficult because, on the one hand, a swollen leg may be due to paralysis and dependency while, on the other hand, a patient may not complain about pain and swelling because of language and perceptual problems. The value of screening asymptomatic stroke patients for DVT has not been established. In considering such a policy one has to remember that the sensitivity and specificity of non-invasive tests, such as D-dimers and compression Doppler ultrasound, are lower in patients who do not have symptoms of DVT than in those with symptoms, and more ‘positives’ will be ‘falsepositives’.127–129 Given the high frequency of PE in hospitalized stroke patients, if a patient has clinical features compatible with PE – breathlessness and/or tachypnoea, with or without pleuritic chest pain and/or haemoptysis for which there is no other reasonable clinical explanation – the probability of PE is high. In such patients a CT

pulmonary angiogram can be used to confirm or exclude a PE130 because it is difficult to distinguish clinically between breathlessness due to pulmonary embolism and other causes without further investigation. Prevention Manoeuvres which may reduce the risk of DVT and pulmonary embolism include: • Early mobilization of the patient and avoidance of prolonged bed rest, although the effectiveness of this regimen after stroke has never been tested in RCTs. • Hydration/fluids may influence the risk of venous thromboembolism. A raised urea, probably indicating dehydration, is associated with a higher risk of DVT.131 Also a systematic review of RCTs testing haemodilution in stroke indicated that this probably reduces the risk of DVT and PE (odds ratio 0.54, 95% confidence interval 0.30–0.99).132 It is unclear whether this is a specific effect of haemodilution or a non-specific effect of improved hydration. We give intravenous saline to most of our patients with acute stroke and immobility, in part because they are often unable to take adequate fluids orally (sections 11.17 and 11.18.1). • Full-length graduated compression stockings. Systematic reviews of RCTs in patients undergoing surgery have shown about a 60% reduction in the odds of developing DVT.133,134 Most of these RCTs tested full-length stockings or did not specify the length. There is very little evidence that short (below-knee) stockings effectively prevent DVT.134 But, in stroke, unlike surgery, stockings cannot be applied before the onset of the insult (i.e. the surgery itself), so DVTs may develop before stockings can be applied, and patients may be immobile for weeks and have prolonged leg paralysis. Moreover, compression stockings have potential hazards: they occasionally cause pressure ulcers (Fig. 11.9a) or acute limb ischaemia, the latter particularly in those with diabetes, peripheral neuropathy or peripheral vascular disease (Fig. 11.9b), they are uncomfortable and unpopular with patients and nurses; and considerable nursing resources are consumed in their application and monitoring. If used, stockings should be fitted in accordance with the manufacturer’s instructions and removed daily to check for skin problems. The evidence for their effectiveness in stroke patients is inconclusive but a large randomized trial is in progress (www.clotstrial.com;135 http:// www.strokecenter.org/trials). • Aspirin, started within 48 h of a presumed ischaemic stroke, reduces the relative risk of PE by about 30% and improves the patients’ overall long-term outcome136 (section 12.3). We start aspirin routinely (first dose

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 559

11.13 Venous thromboembolism

(a)

300 mg and 75 mg per day thereafter) as soon as we have excluded intracranial haemorrhage or other contraindications. • Heparin has been shown to reduce the risk of DVT in patients with ischaemic stroke, but this benefit is offset by the risk of haemorrhagic complications so that routine use of heparin does not improve overall outcome137 (section 12.4). We reserve unfractionated (5000 u twice per day) or low-molecular-weight heparin for patients we judge to be at particularly high risk of DVT and PE and low risk of haemorrhagic complications, accepting that these judgements are based on inadequate evidence. Such patients might include those with an ischaemic stroke with severe leg weakness and immobility, and cancer, thrombophilia or previous venous thromboembolism (section 12.4.3). • Other methods of prophylaxis, e.g. external pneumatic compression and functional electrical stimulation, have been suggested but not evaluated adequately in stroke.135,138 Treatment If a patient with a confirmed ischaemic stroke has a proven above-knee DVT or pulmonary embolism, we use subcutaneous low-molecular-weight heparin (LMWH) despite the lack of direct evidence that this is more effective than intravenous unfractionated heparin in stroke patients. We prefer LMWH because it is easier to use and does not need monitoring, and there is evidence in patients other than those with stroke that it is just as, or more, effective.139–141 If patients are symptomatic we also usually treat DVT restricted to the calf veins although this depends on the presence of any relative contraindications to treatment.142 In patients with DVT restricted to calf veins who have no symptoms we would also anticoagulate if a repeat Doppler ultrasound showed propagation of the thrombus in the popliteal or femoral veins. We normally continue heparin for a few days while starting oral anticoagulants which we continue for about 6 months, depending on our judgement of the patient’s risk of recurrence, taking into account their

(b)

..

..

Fig. 11.9 (a) A pressure ulcer behind the knee due to graduated compression stockings and inadequate monitoring by nursing staff. (b) A patient with peripheral vascular disease and diabetes mellitus who was fitted with graduated compression stockings. Note the necrotic skin over the anterior border of the tibia (white arrow) and where the stockings were creased at the ankle (black arrow). There were also necrotic areas over both heels which failed to heal and led to a right above-knee amputation following an unsuccessful revascularization procedure.

559

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 560

560

Chapter 11 What are this patient’s problems?

immobility and risk of bleeding.143 We do not believe that intracranial haemorrhage is an absolute contraindication to anticoagulation since, if a patient has had a life-threatening pulmonary embolus, the risk of anticoagulation may be worth taking. Alternatively one might consider insertion of a caval filter but there are no reliable studies which help us decide between these two treatments.144 In a small number of patients, thrombolysis may need to be considered for massive PE or DVT but this has not been evaluated in stroke patients and would inevitably be associated with a risk of intracranial bleeding.

11.14 Urinary incontinence and retention

Between one- and two-thirds of acute stroke patients admitted to hospital are incontinent of urine in the first few days.145,146 Urinary incontinence is more common in older patients, those with severe strokes, other disabling conditions and diabetes.145,147 Urinary incontinence may be caused by the stroke itself, but perhaps onefifth of patients have been incontinent before the stroke. Although detrusor instability is the most common single cause of urinary incontinence after the first 4 weeks, many other factors may contribute in the acute stage (Table 11.15). Urinary incontinence is an important cause of distress to patients and carers, increases the risk of pressure ulcers (section 11.16), often interferes with rehabilitation (e.g. by interrupting physiotherapy sessions or increasing spasticity; section 11.20) and influences the patient’s requirements for ongoing nursing care.147,148 Urinary incontinence may also cause patients to become dehydrated because patients tend to restrict their own fluid intake to limit urinary incontinence without telling the nurses or doctors. Urinary incontinence may cause patients to become dehydrated. This is because many patients restrict their own fluid intake to limit urinary incontinence – without telling the nurses or doctors. Assessment To identify patients with urinary incontinence one simply has to ask the carer or nursing staff. These are the people most aware of urinary problems since they have to deal with the consequences. It is important to ask, since many people consider incontinence an inevitable consequence of stroke and thus not worthy of mention.

Table 11.15 Factors which may contribute to urinary incontinence. Section Reduced conscious level Immobility (cannot get to the toilet in time) Communication problems (cannot ask to go to the toilet) Impaired upper limb function (cannot manipulate clothes or the urinal) Dyspraxia Loss of inhibition of bladder contraction (detrusor instability so cannot wait to go) Urinary infection (often without any other symptoms) Urinary overflow due to outflow obstruction (e.g. prostatism) Faecal impaction Excess urinary flow due to high fluid intake, diuretics and poorly controlled diabetes Too few carers/nurses (cannot attend to patients in time) Importance of maintaining continence underestimated by carers/nurses

11.3 11.11 11.30 11.21 11.28

11.12.2

11.15 11.18.3

Routine use of a measure such as the Barthel Index to monitor patient progress on the stroke unit should identify all patients with urinary incontinence. It is often useful, but frequently overlooked, to ask the patients themselves what they think is causing their incontinence. This may, for instance, help distinguish true incontinence from accidental spillage of urine from a urinal because of the patient’s poor manual dexterity (Fig. 11.10). Many of our patients with severe stroke, often with some cognitive dysfunction, seem unaware of either voiding or being wet. They usually deny incontinence when asked. More detailed information, including urinary volumes, frequency and times of voiding, which can be collated with a micturition chart, may be useful in identifying the causes of incontinence (e.g. diuretics, communication difficulties) and in formulating a management plan. Many of our patients with severe stroke, often with some cognitive dysfunction, seem unaware of either voiding or being wet. They usually deny incontinence when asked. Although we exclude specific causes and try both behavioural and pharmacological interventions we usually resort to incontinence pads or an indwelling catheter. Where the cause of urinary incontinence is unclear, and if it persists for more than a few days, the patient

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 561

11.14 Urinary incontinence and retention

(a)

(b) Fig. 11.10 Some patients have difficulty manipulating a urinal which leads to spills and, effectively, incontinence. (a) This urinal can be inverted without leaking; (b) a simple

one-way valve inserted in the neck of the urinal prevents the contents spilling.

should be investigated. Urine dipstick testing for leucocytes and nitrates may help to exclude an infection but, given the high frequency of urinary infection in stroke patients (section 11.12.2), it is insufficiently accurate to be used as a single test; microscopy and culture should also be performed to exclude or confirm the diagnosis.118 Measurement of postmicturition bladder volumes (by bladder ultrasound or catheterization) may be useful in assessing bladder sensation, contractility and outflow. We reserve formal urodynamic studies, which can better identify detrusor hyper- and hyporeflexia and bladder outflow problems, for the few patients with unexplained, troublesome incontinence which persists for weeks after the stroke and where the incontinence is out of proportion to the stroke severity.

among survivors, incontinence resolves in the majority, unless it pre-dated the stroke. Younger age, less severe stroke and lacunar stroke have all been associated with better rates of resolution.147,149 Persisting urinary incontinence is associated with worse functional outcome and higher rates of institutionalization.147,149,150

A negative urine dipstick test for leucocytes and nitrates does not reliably exclude a urinary infection; a midstream urine collection should be obtained if possible and sent for microscopy and culture.

Prognosis Urinary incontinence is an important predictor of poor survival and functional outcome after acute stroke (section 10.2.7): between 30% and 60% of incontinent patients die within the next few months.145,149 However,

..

..

Treatment There is very little reliable evidence about the treatment of post-stroke incontinence.151 If a patient is incontinent, but able to understand, then a careful explanation of the cause and likely prognosis should be given. Carers often benefit from such information too. Because urgency of micturition is such a common cause of incontinence in stroke patients, simple steps such as regular toileting, offering aphasic patients some means to alert the nurses to their needs, improving their mobility, or providing a commode by the bed, can all be effective. Provision of suitable clothing, such as trousers with a flap fastened with Velcro, may allow patients to use a urinal or commode independently and thus promote continence. Obviously, one should strive to treat the underlying cause (e.g. infection, outflow obstruction) and where possible remove exacerbating factors (e.g. excessive fluids, uncontrolled hyperglycaemia or diuretics). Having excluded easily treatable causes, we first employ ‘bladder retraining’ where patients are prompted to void

561

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 562

562

Chapter 11 What are this patient’s problems? Table 11.16 Drugs used to inhibit bladder contractility, and their adverse effects. Anticholinergic drugs Flavoxate hydrochloride Oxybutynin hydrochloride Tolterodine tartrate Propiverine hydrochloride Propantheline bromide Tricyclic antidepressants Imipramine Amitriptyline Nortriptyline Common adverse effects Dry mouth Blurred vision Nausea/vomiting Constipation/diarrhoea Confusion in the elderly Retention where there is bladder neck obstruction Precipitation of acute glaucoma

regularly. If this does not achieve continence, we use a bedside ultrasound machine to exclude a postmicturition residual of 100 mL or more and then introduce an anticholinergic drug (e.g. oxybutinin or tolteridine) assuming there are no contraindications, e.g. closed angle glaucoma.152 One must obviously be alert to the possible adverse effects of these drugs (Table 11.16). This approach achieves reasonable results with relatively few patients requiring formal urodynamic studies.153 An indwelling catheter should be avoided if at all possible because it makes resolution of urinary incontinence impossible to detect, and may lead to a number of complications (Fig. 11.11, Table 11.17). Intermittent catheterization facilitates detection of the resolution of incontinence and probably reduces the rate of bacturia but it is certainly more labour intensive and therefore may not be practical.114 Other aids and appliances may be useful (Table 11.18) in avoiding unnecessary catheterization. However, when patients are at high risk of pressure ulcers (section 11.16), and other means have failed to keep them dry, or if accurate monitoring of fluid balance is required for some reason, an indwelling catheter is the best option. Catheterization may also be required to relieve urinary retention (see below) until any cause or precipitating factor can be removed, e.g. enlarged prostate, urinary infection, severe constipation, anticholinergic drugs. There is limited evidence that prophylactic antibiotics reduce the rate of bacturia in patients with indwelling catheters and this policy may increase bacterial resistance and rates of Clostridium difficile-associated diarrhoea.112,114

Fig 11.11 Traumatic hypospadias caused by an indwelling catheter. This is most likely to occur in a man with cognitive or communication problems who may not indicate he is uncomfortable, and who is being looked after by inadequately trained nursing staff.

Occasionally, if urinary incontinence is a bar to discharge into the community, long-term catheterization with a silastic catheter may be the preferred option. This issue should be discussed with the patient and, where appropriate, their carer. Continence advisers, nurses with specialist training in the management of incontinence and access to information, aids and appliances, can often help other professionals, patients and their carers.151 The choice of continence aids and catheters is huge but there are few rigorous studies to establish which are the most cost-effective, and most acceptable to patients and carers.154,155 In many parts of the UK, laundry services run by health authorities or social services provide invaluable assistance to families having to cope with incontinence. An indwelling catheter should be avoided if possible because this makes resolution of urinary incontinence impossible to detect and may lead to a number of complications.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 563

11.14 Urinary incontinence and retention Table 11.17 Problems (and solutions) of indwelling urinary catheters.434 Problem

Solution

Pain on insertion

Explain to the patient what is going to happen Use plenty of anaesthetic gel and allow time for it to work Ensure foreskin is not left retracted after insertion Avoid traction and monitor carefully; if it occurs replace with suprapubic catheter Explain why catheter is needed, how it works and how long it will be in place Provide a discreet drainage bag Use well-supported leg bag for mobile patients Use appropriate size of catheter Inhibit any involuntary bladder contraction which causes bypassing with an anticholinergic drug (Table 11.16) Change catheter if blocked Ensure adequate urine flow Remove encrusted catheter Avoid unnecessary catheterization since no proven method of preventing infection Ensure balloon inflated to correct volume or use larger volume balloon

Paraphimosis Traumatic hypospadius (Fig. 11.11) Poor self-esteem Immobility because of drainage bag Leakage

Blockage Infection Catheter falls out due to urethral dilatation or pelvic floor laxity Catheter rejection due to bladder contraction Catheter pulled out by patient Pain on catheter removal

Failure of balloon deflation

Avoid large volume balloon Inhibit with anticholinergic drug (Table 11.16) Manage without a catheter to avoid further trauma Avoid routine changes Explain procedure to patient Allow adequate time for balloon deflation Introduce ureteric catheter stylet along inflation channel

Urinary retention Table 11.18 Aids and appliances which may be useful in patients with urinary incontinence.435 Absorbent pads and pants These vary in the volume of urine they can absorb, their shape, and the method of holding them in position. Urinals Useful for men who are immobile or have urgency which gives them insufficient time to reach the toilet. They can be fitted with a non-spill valve for patients who have poor manual dexterity (Fig. 11.10), or fluid absorbing granules, to reduce spillage. Bedside commode Useful where urgency is associated with poor mobility so the patient has insufficient time to get to the toilet (Fig. 11.33b). Penile sheath Often viewed as an alternative to an indwelling catheter in men without bladder outflow obstruction, but they easily fall off and are therefore unsuitable for agitated or confused patients. Other problems include skin erosions due to urinary stasis or the adhesive strip, and twisting of the sheath and penile retraction during voiding which causes leakage.

..

..

Urinary retention, which may be acute or chronic, is common in stroke patients, more so in men. The main cause is pre-existing bladder outflow obstruction which may be exacerbated by constipation, immobility and drugs such as tricyclic antidepressants which have antimuscarinic effects.156 Urinary retention may present with dribbling incontinence, agitation or confusion and is easily missed in patients with a reduced conscious level, communication difficulties or other cognitive problems. It is important to palpate the patient’s abdomen or if in doubt perform a bladder ultrasound, on admission and later, if urinary problems or agitation develop, to exclude a distended bladder. Chronic retention with a postmicturition bladder volume of greater than 150 mL increases the risk of infection. A urethral catheter provides prompt relief. Removal of any exacerbating factors may avoid the need for a catheter or at least allow it to be removed quickly. In men with benign prostatic hypertrophy, alpha-blocking drugs (e.g. prazosin) or finasteride (which inhibits the metabolism of testosterone to dihydrotestosterone in the prostate) may enable one to remove the catheter without recurrence of

563

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 564

564

Chapter 11 What are this patient’s problems?

retention. Surgeons and anaesthetists are often unwilling to consider transurethral resection of an enlarged prostate until several months have passed following a stroke. We are not aware of any evidence to indicate how long we should delay. It is important to palpate the patient’s abdomen on admission and, later, if urinary problems or agitation develop, to exclude a distended bladder.

11.15 Faecal incontinence and constipation

Constipation is very common after stroke and may lead to abdominal discomfort, anxiety and faecal smearing or incontinence. Immobility, poor fluid and food intake, constipating analgesics and anticholinergic drugs are common causes. Faecal incontinence affects about onethird of patients after stroke and has been associated with increasing age, diabetes, other disabling conditions, stroke severity, immobility and size of brain lesion.145,157

Faecal incontinence which is not associated with severe cognitive problems is almost always remediable by dealing with constipation or diarrhoea. The rectum is often full of soft faeces which the patient cannot evacuate effectively. Treatment There is little reliable research to guide treatment.158 Avoidance of constipation by ensuring an adequate intake of fluid and fibre is the best approach, but laxatives, suppositories and, occasionally, enemas are sometimes required. We find that stimulating laxatives (e.g. senna) are generally more effective than osmotic ones (e.g. lactulose) in elderly patients although the choice will be influenced by whether the stool is hard or soft. It is important to remember that laxatives may cause incontinence in immobile patients. Where patients are unable to toilet themselves, and a carer is not constantly available, it may be necessary to induce constipation, with for example codeine phosphate, and then relieve this with regular enemas to coincide with visits from a carer. Simple interventions including advice on diet, fluids and use of laxatives are helpful in reducing problems in the longer term.

Assessment The frequency of bowel movements should be monitored. Simple monitoring will detect constipation and diarrhoea and may help establish the pattern of any faecal incontinence. Review of the drug chart will often identify possible causes. Abdominal and rectal examination will usually identify faecal impaction and indicate whether the constipated stool is hard or soft. Occasionally, if the patient has faecal incontinence associated with diarrhoea, it may be useful to culture the stool, and test for Clostridium difficile toxin to exclude infection or to X-ray the abdomen to exclude high faecal impaction. More detailed investigation is not required unless there are persistent unexplained problems. Prognosis Faecal incontinence early after stroke onset often resolves in survivors but may develop among those who were initially continent.145,157 Later development of faecal incontinence is usually due to preventable causes such as drugs and inadequate staffing levels. Achieving continence can be a crucial step in discharging a patient home since faecal incontinence is so practically and socially difficult to cope with, and is invariably a considerable strain for the carer.

11.16 Pressure ulcers

Pressure ulcers occur when local pressure on skin and subcutaneous tissues exceeds the capillary opening pressure for long enough to cause ischaemia. In addition, friction may cause blistering and tears in the skin. Pressure ulcers usually occur over weight-bearing bony prominences (Fig. 11.12). Ulcers occur in patients who are immobile and unable to redistribute their own weight when lying or sitting. The frequency of pressure ulcers in hospitalized stroke patients is bound to vary depending on the population studied, the prophylactic methods used and the diagnostic criteria. Pressure ulcers are more common in patients who are malnourished, infected, incontinent or have serious underlying disease.159 They cause pain, increase spasticity, slow the recovery process, and may be fatal if complicated by severe sepsis. They prolong length of stay in hospital, often require intensive treatment and can therefore be extremely expensive to health services.160 They can and should be prevented, although they may develop in the interval between the onset of the stroke and admission to hospital.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 565

11.16 Pressure ulcers

(a)

Assessment Immobile patients should be examined regularly (sometimes several times in a day) to identify early signs of pressure damage, i.e. skin redness. It is important that patients who are at particular risk of developing pressure ulcers are identified as early as possible so that preventive measures can be taken. The Waterlow Scale (Table 11.19) is one of many clinical scoring systems developed to indicate an individual patient’s risk. Most scales include some measure of mobility, continence, cognitive function and nutritional status and none, based on less than ideal evaluations, is clearly superior to the others.161

Skin Subcutaneous tissue Bony prominence

(b)

Shoulders (3%) Spine (2%)

Sacrum (36%) Trochanter (3%)

Natal cleft (3%)

Heels (25%)

Buttocks (21%)

Malleoli (2%) Other (4%)

Fig. 11.12 (a) The distortions of tissues over a bony prominence due to compression or shear which may lead to pressure sores and (b) the anatomical distribution of established pressure sores based on data from a cross-sectional UK survey of pressure sores in patients being nursed on the Pegasus Airwave System. Although patients with cerebrovascular disease were the largest group in this sample, they made up only 14% of the whole group. (From St Clair, 1992.459)

..

..

Each patient’s risk of pressure ulcers should be assessed and documented; actions, appropriate to the level of risk should be taken to prevent pressure ulcers developing. Pressure ulcers can be prevented by good nursing. The clinician should be alert to behaviour which, in patients with communication and cognitive problems, may indicate a painful pressure area. Patients may repeatedly move themselves out of a desired position. For example, patients, colloquially known as ‘thrusters’, with a painful sacrum may force themselves out of a chair by extending their bodies at the waist. This may become a major problem for nursing staff. If patients develop pressure ulcers it is useful to have some objective measure of their severity so that healing or lack of healing can be monitored. Photographs incorporating a centimetre scale are a convenient and reliable method to demonstrate change, but where this facility is not available, tracing the limits of the ulcer or simply measuring it in several planes is useful.162 Patients who are at risk, or who have established pressure ulcers, should be investigated to exclude malnutrition, hypoalbuminaemia, anaemia and infection (in the pressure ulcer or elsewhere), all of which can slow healing.162 Prevention The most important way to prevent ulcers is to relieve the pressure on the tissues for long enough, and at frequent enough intervals, to allow the tissues to receive an adequate blood supply. This can usually be achieved by regular turning of patients (2- to 4-hourly depending on the assessment of risk) but this takes up a lot of skilled nursing resources. Although the introduction of a variety of special mattresses and cushions (Table 11.20) may reduce the need for regular turning, most patients do still need to be turned. Some beds are designed to turn the patient automatically (e.g. the net suspension bed).

565

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 566

566

Chapter 11 What are this patient’s problems? Table 11.19 The Waterlow Scale. Build/weight

Visual skin

Average 0 Above average 2 Below average 3

Healthy Tissue paper Dry Oedematous Clammy Discoloured Broken

Continence 0 1 1 1 1 2 3

Mobility

Complete Occasionally incontinent Catheter/incontinent of faeces Doubly incontinent

0 1 2 3

Fully mobile Restricted/difficult Restless/fidgety Apathetic Inert/in traction

Sex/age 0 1 2 3 4

Male Female 14–49 50–64 65–74 75–80 81+

Appetite 1 2 1 2 3 4 5

Average 0 Poor 1 Anorexic 2

A total score of 10 indicates a patient is at risk of pressure ulcers, one of 15 indicates a high risk and a score of 20 a very high risk. In addition to the basic scale in which the scores for each of the six domains (i.e. weight, skin, continence, mobility, age/sex and appetite) are summed, additional points are added for special risk factors: poor nutrition (8 points); sensory deprivation including stroke (5 points); high-dose anti-inflammatory drugs, steroids (3 points); smoking > 10/day (1 point); orthopaedic surgery or fracture below waist (3 points).

Table 11.20 Specialized mattresses, cushions and beds. Passive systems Sheepskin fleeces and bootees which reduce skin shear and moisture; natural fleeces are better than man-made ones but they are rendered ineffective by poor cleaning and being covered by sheets. Padded mattresses containing polyester fibres, e.g. Spenco. Polystyrene bead system. Foam mattresses (e.g. Vaperm) vary in their pressure-relieving properties. Gel pads can be used under heels and sacrum. Roho cushions are effective but very expensive (Fig. 11.13a). Active systems Ripple mattresses and airwave systems provide alternating pressure; the larger the cells the better but they tend to break down and leak. Low air loss systems (e.g. Mediscus) providing constant low pressure; although effective, tend to be noisy, expensive and complex, needing regular maintenance and training (Fig. 11.13b). Flotation beds and deep water beds are difficult to nurse patients on, are very heavy and some patients get motion sickness. Dry flotation providing constant low pressure produced by glass microspheres blown by air. Air can be turned off when the patient needs to be repositioned. Effective but bulky and expensive. Mechanical beds turn the patient, e.g. net suspension bed. Effectiveness uncertain, patients may not like being suspended (in full view of people around them). Useful for turning patients who are in pain.

Pressure-relieving mattresses and cushions (Fig. 11.13a) are divided into ‘passive’ and ‘active’ systems (Table 11.20). The ‘passive’ systems distribute the patient’s weight through a larger area and make it easier for them to reposition themselves. High-specification foam mattresses are more effective than standard foam mattresses.160 ‘Active’ systems (Fig. 11.13b) usually work by inflating and deflating air cells to relieve pressure on each point at regular intervals, and are more effective than the ‘passive’ systems in intensive care patients.160 However, they are expensive and can make certain nursing tasks more difficult, e.g. positioning a patient to reduce the risk of contractures, to help breathing and to facilitate swallowing, because they offer a less firm base.

Other interventions such as staff education, nutritional support and local treatments (e.g. creams, lotions) applied to unbroken areas of skin may have a role in prevention but their effectiveness has not yet been demonstrated.163,164 The ultimate choice of preventive method will depend on an assessment of the individual patient’s risk of pressure ulcer, the availability of nurses, the patient’s other needs, e.g. positioning, and available resources. Further research is required to identify the most cost-effective strategy for preventing pressure ulcers.160 Studies will have to take into account patients’ absolute risk of developing ulcers, the reduction in risk with each intervention and the cost of the intervention, as well as the cost

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 567

11.17 Swallowing problems

(a)

(b)

Fig. 11.13 Pressure-relieving cushion and bed: (a) Roho cushion; (b) low air-loss bed.

of treating any pressure ulcers which develop. However, we believe that, whatever technology is employed, adequate numbers of skilled nursing staff will still be essential if pressure ulcers are to be prevented. Finally, it is important to recognize that certain management strategies (e.g. positioning in a chair, use of graduated compression stockings) may actually cause pressure ulcers unless they are properly applied and monitored (Fig. 11.9). Treatment For patients with established pressure ulcers, the relief of pressure probably remains the most important factor in promoting healing. ‘Active’ pressure-relieving systems are more effective than ‘passive’ ones.165 In addition, it is important to optimize the patient’s general condition by providing a good diet with adequate protein intake and by treating concurrent illness aggressively (e.g. infections, cardiac failure). This may need intensive nursing input in frail, elderly, anorexic patients who are drowsy or have swallowing problems. The pain associated with pressure ulcers may increase tone and lead to contractures (section 11.20) which may well hinder rehabilitation, while the discomfort may also affect the patient’s morale and even further worsen their outcome. Adequate analgesia, with opiates if necessary, should be given to patients, especially before renewal of dressings. Antibiotics may be required if there is local or systemic infection (spreading cellulitis, osteomyelitis). Debridement to remove necrotic tissue, and skin grafting to achieve skin coverage, may sometimes be necessary. A bewildering variety of local dressings and treatments (e.g. vitamin C, zinc, ultrasound, electrical stimulation, ultraviolet light) which aim to promote healing and

..

..

reduce infection are available. Some small RCTs evaluating these interventions have been reported, but need to be systematically reviewed.166,167

11.17 Swallowing problems

Up to one-half of conscious patients admitted to hospital with an acute stroke cannot swallow safely on bedside testing.168 However, estimates of the frequency of swallowing difficulties vary because of differences in definitions, timing and methods of detecting dysphagia and in selecting patients for study. Swallowing difficulties have been associated with a high case fatality and poor functional outcome and certainly put patients at risk of aspiration, pneumonia, dehydration and malnutrition.168 However, much of the excess mortality and morbidity is probably due to the severity of the stroke itself rather than to swallowing difficulties.

11.17.1 Mechanisms of dysphagia The following patterns of dysphagia after stroke have been identified with videofluoroscopy:169,170 • Poor oral control (oral preparatory phase) and delayed triggering of the swallow leads to aspiration before the swallow, i.e. liquid trickles over the back of the tongue before the swallow starts. In addition, patients with weakness or incoordination of the face or tongue often have difficulty keeping fluids in their mouth, and in chewing and manipulation of food to produce a wellformed food bolus.

567

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 568

568

Chapter 11 What are this patient’s problems?

• Failure of laryngeal adduction leads to aspiration during the swallow itself. • Reduced pharyngeal ‘peristalsis’, or cricopharyngeal dysfunction, may allow food to collect in the pharynx and spill over, past the vocal cords and into the trachea. Thus, aspiration occurs after the swallow. Poor oral control and delayed triggering of the swallow are the most common mechanisms causing dysphagia after stroke but more than one abnormality can usually be identified in any individual patient.

11.17.2 Detection of dysphagia Despite their frequency, and the serious consequences of failing to detect them, swallowing problems are often not sought systematically in patients admitted to hospital with an acute stroke.60 The current recommended approach is for a suitably trained nurse or other healthcare worker to perform a simple bedside swallow screen (Fig. 11.14), to distinguish patients able to take oral food and fluids safely from those who require a more detailed

Addressograph, or Name

SWALLOWING SCREENING TEST – an interdisciplinary tool Pre-assessment criteria: Site: If the patient is drowsy and unable to sit upright, than it is NOT SAFE to complete this assessment. They should remain Nil By Mouth (NBM). Monitor conscious level: attempt to screen daily until completed.

DoB Unit number

NB: Nutritional screening required 6.7 mmol/L, or 120 mg/dL) occurs in about one-third of non-diabetic acute stroke patients and two-thirds of those with diabetes.194 About one-quarter of those with hyperglycaemia are known to have diabetes mellitus already and another quarter have a raised HbA1C which suggests that their blood glucose has been high for some time before the stroke, referred to as ‘latent diabetes’.195 Whether the hyperglycaemia in non-diabetics is due to release of catecholamines and corticosteroids as part of the stress response is controversial.196 Hyperglycaemia, after a stroke, at least in non-diabetic patients, is associated with increased case fatality and poor functional outcome.194,197 This could be explained by more severe strokes producing a greater stress response and, thus, hyperglycaemia so that hyperglycaemia is simply a marker of severe stroke. However, some studies have demonstrated that hyperglycaemia is associated with a poor outcome having adjusted for stroke severity and other baseline prognostic factors.194 This finding, along with some (but not all) animal work showing that hyperglycaemia can exacerbate ischaemic neuronal damage, has led many to believe there is a causal relationship between hyperglycaemia and poor outcome.197,198

stroke. This will keep the patient free from thirst and avoid excessive diuresis which may cause dehydration (section 11.18.1). Whether more aggressive control of blood sugar, which has at least theoretical benefits for the ischaemic penumbra, is sensible will depend on any benefits, and the risks of hypoglycaemia, which will almost certainly depend on the intensity of monitoring available.198 One small and one moderate sized, but underpowered RCT have shown that glucose-potassiuminsulin (GKI) infusions can be used with reasonable safety to control hyperglycaemia in acute stroke but have not demonstrated any improvement in survival or functional outcomes.199,200

11.18.4 Hypoglycaemia Hypoglycaemia may occasionally mimic stroke or transient ischaemic attack (sections 3.4.5 and 7.16). Ideally, it should be excluded on first contact with medical services (e.g. paramedics, primary care physician) by measuring the glucose on a capillary sample in all patients taking hypoglycaemic medication.201 Hypoglycaemia also occurs after stroke because of efforts to normalize blood sugar in those with hyperglycaemia (section 11.18.3) and in diabetic patients receiving hypoglycaemic medication but a reduced food intake because of dysphagia (section 11.17) or other post-stroke problems (section 11.19.1). Since hypoglycaemia may, if severe or undetected, cause worsening of the neurological deficit, the blood sugar should be monitored particularly carefully in diabetic patients on hypoglycaemic medication.

Assessment A random blood glucose should be measured in all patients with stroke. In those with hyperglycaemia, a fasting blood glucose and an HbA1C will help distinguish latent diabetes from hyperglycaemia due to the stroke itself. Blood sugar levels are likely to fall spontaneously in the first few days after stroke onset.199 If necessary, a glucose tolerance test after the acute stage of the illness (i.e. when the patient is medically stable) may be helpful in sorting out which patients have diabetes or impaired glucose tolerance, and which simply have hyperglycaemia related to the acute stroke. Patients with established diabetes and latent diabetes should be assessed to exclude vascular (both micro and macro) and neurological complications. Treatment We currently aim to keep the blood sugar less than 11 mmol/L (200 mg/dL) in the first few days after an acute

..

..

11.19 Nutritional problems

11.19.1 Malnutrition Malnutrition is a common and often unrecognized problem in patients admitted to hospital, especially the elderly.202 Inevitably, the reported frequency of malnutrition after stroke has varied depending on patient selection, the definitions of malnutrition and the method and timing of assessments. Estimates vary from 8% to 34%.203,204 It is not clear which factors are associated with malnutrition on admission, but the non-stroke literature suggests that malnutrition is more frequent in older patients, those living in institutions and poor social circumstances, and those with prior cognitive impairment, physical disability or gastrointestinal

575

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 576

576

Chapter 11 What are this patient’s problems?

disease. In a proportion of stroke patients, nutritional status may worsen during hospital admission.205,206 Like any acute illness, stroke may cause a negative energy balance and greater nutritional demands but, at the same time, stroke patients may be less able to adapt to these.207,208 They often have swallowing difficulties (section 11.17) and even those who can swallow may have a poor appetite because of lack of taste associated with the stroke itself209 or medication, intercurrent illness, depression, or apathy; and they may eat slowly because of facial weakness, lack of dentures or poor arm function. Poor nutrition has been associated with reduced muscle strength, reduced resistance to infection and impaired wound healing (although not specifically in stroke patients). Among patients with stroke, muscle weakness, infections and pressure ulcers are common and account for many deaths and much morbidity.105 Malnutrition on admission is associated with increased case fatality and poorer functional outcomes even having adjusted for other factors, but it is unclear to what extent this association is causal.210,211

11.19.2 Obesity Obesity is common among hospitalized stroke patients.36,212 As well as being a risk factor for stroke (section 6.6.10), and presumably recurrent stroke obesity can also be a problem during recovery from stroke. Where patients have restricted mobility, especially where they rely on others for help with transfers, obesity can be a crucial factor in how long they remain in hospital and how much support they require. It is also a problem in the long term in achieving adequate control of vascular risk factors such as hypertension (section 16.7.3) and diabetes. Patients quite often gain weight after stroke, presumably because of decreased energy expenditure and excessive calorie intake.

11.19.3 Assessment In routine clinical practice there are practical difficulties in assessing stroke patients’ nutritional status. A dietary and weight history may not be available because of patients’ communication problems and an alternative source of this information may not be available if, as is common, the patient lives alone. Simple assessments of weight and height to estimate the body mass index (BMI) pose problems in immobile stroke patients. Specialized equipment, of limited availability, such as weighing beds or hoists and scales which accommodate wheelchairs, may be required and height may have to be estimated from the patient’s demi-span or heel–knee

Fig. 11.19 A hoist which incorporates an electronic scale (inset) for weighing immobile stroke patients who are unable to stand or sit unaided.

length (Fig. 11.19). Laboratory parameters such as haemoglobin, serum protein, albumin and transferrin may not necessarily reflect nutritional status. More complex anthropometric measures, vitamin estimations, antigen skin testing and bioelectric impedance are all used in research but are not widely available. Awareness of the possibility of malnutrition is a key factor in identifying malnourished patients. It is probably worthwhile using a standardized screening tool for nutritional problems in the stroke unit which can highlight those patients who would benefit from a dietician’s input, but none have been adequately evaluated for stroke patients.213,214 A simple and informal end-of-thebed assessment (i.e. is the patient undernourished, normal or overweight?) is better than no assessment at all since it reliably identifies most stroke patients with a low BMI and abnormal anthropometry, as well as predicting outcome.210,215 Estimation of the BMI, serial weights to identify weight loss, and monitoring of dietary intake should be used to screen patients on admission and monitor them while on the stroke unit. Simple laboratory tests including a serum albumin may be worth monitoring where there is clinical evidence of poor or worsening nutrition. All patients admitted to hospital with stroke should have an early assessment of their nutritional status. In the absence of a formal assessment tool the team’s judgement of whether the patient is undernourished, normal or overweight is better than nothing. At least it focuses the team on this important aspect of care.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 577

11.19 Nutrition

11.19.4 Prevention and treatment of malnutrition General approach It is useful to involve the speech and language therapist and dietician in the assessment and care of patients who have swallowing problems, or when nutritional intake may be inadequate for other reasons, e.g. confusion. Patients often eat very slowly after stroke and need supervision to ensure safe swallowing. Simple measures such as providing appetizing food of an appropriate consistency (section 11.17), placing the patient’s meal in their intact visual field, and ensuring that the patient has well-fitting dentures should not be overlooked. Labelling a patient’s dentures with their name while in hospital helps prevent losses and unfortunate mix-ups. Staff shortages may mean that patients receive insufficient food, which will eventually cause malnutrition, or have food forced into them hurriedly by unthinking staff, which is very demeaning and adversely affects morale. Where trained staff are unable to cope, the patient’s family or even volunteers can be easily trained to help with feeding. Dentures are often lost during hospital admission. Labelling a patient’s dentures with their name helps prevent losses and unfortunate mix-ups. Most people do not like to be fed; they prefer to feed themselves. By providing patients with suitable feeding utensils or foods which they can pick up may allow them to feed themselves (section 11.32.6). If not, adequate staff or family members, who have been instructed in how to do it, need to be employed. Oral nutritional supplements Routine oral nutritional supplementation, which provide both proteins and calories, in hospitalized stroke patients is probably not worthwhile since they have not been shown to have a clinically useful impact on outcome. The FOOD trial randomly allocated 4023 patients with an acute stroke to routine oral nutritional supplements or not, for the duration of their hospital admission. There was no significant difference in survival or functional outcomes overall, but there were insufficient patients enrolled who were undernourished to determine if this subgroup might benefit.212 Therefore, the results of the FOOD trial, taken along with a large number of much smaller single-centre RCTs of nutritional support in elderly patients, suggest that oral supplements are probably useful, but just in those who are identified as malnourished or who have inadequate food intake.212

..

..

Enteral tube feeding Where swallowing is impaired, early introduction of nasogastric tube feeding probably reduces the risk of death although maybe at the expense of keeping more patients with severe disability alive (Fig. 11.20).36 This also facilitates administration of important medications. Early tube feeding does not appear to influence the risk of chest infections but does seem to increase the risk of gastrointestinal haemorrhage (section 11.2.3) although it is unclear why this occurs and whether any preventive treatment such as a proton pump inhibitor would be effective (Table 11.26). Inserting nasogastric tubes is often difficult and ensuring that they are in the stomach, rather than in the lungs, at the beginning of each feed is not straightforward.216 Also, patients frequently find the nasogastric tube uncomfortable, and feeding is often interrupted by the patient repeatedly removing the tube. By restraining the patient (Fig. 11.21a,b) one can probably improve the continuity of nasogastric feeding but such restraints are not acceptable to some patients, their families and healthcare staff. Although enteral feeding via a percutaneous endoscopic gastrostomy (PEG) is perceived as more comfortable and reliable in providing the prescribed feed, its routine early use is not associated with better survival or functional outcomes.36 Indeed, the FOOD trial which included 321 patients allocated to initial tube feeding via nasogastric or PEG showed that patients allocated to PEG feeding tended to have worse outcomes (7.8% increased risk of death or severe disability (95% CI –0.08% to 15.5%). However, where a patient is likely to need prolonged tube feeding, or where nasogastric feeding is not practical, a PEG is usually the best option (Table 11.26). But one does have to take account of the associated complications. About one-fifth will develop aspiration pneumonia shortly after insertion, about 10% will develop a wound infection, and potentially life-threatening complications such as peritonitis and major bleeding occur in about 1%.217–222 Patients, or more frequently their relatives, need to be counselled about these risks and the real possibility that their frail relative may die in the days following the procedure.

The early introduction of PEG tube feeding, rather than persisting with a nasogastric tube, in dysphagic patients is unlikely to be associated with better outcomes. However, if nasogastric tube feeding becomes impractical there may be no alternative to PEG insertion.

577

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 578

578

Chapter 11 What are this patient’s problems?

Point estimate from FOOD trial Feed 100 patients early – prevent about six deaths

There was a 5.8% absolute (95% CI − 0.8 –12.5) reduction in the proportion of patients who had died at 6months follow-up in the early enteral tube feeding group compared with the avoid early feed group. The reduction in death or dependency was only 1.2% (95% CI − 4.2– 6.6).

Lower 95% confidence interval: feed 100 patients early and cause about one additional death

Upper 95% confidence interval: feed 100 patients early and prevent about 13 deaths

Fig. 11.20 Results from the FOOD trial comparing outcomes (deaths by 6 months per 100 patients treated) following enteral tube feeding (mostly via nasogastric tube) started early after hospital admission versus avoiding enteral tube feeding for at least 1 week.

11.19.5 Treatment of obesity

11.19.6 Other considerations

Patients who are obese, in particular if this causes problems with mobility or control of diabetes or blood pressure, should be encouraged to lose weight and be offered dietary advice (section 16.7).

It is important to remember that eating plays an important role in our social lives and is a source of pleasure. Eating with other people during the recovery phase of stroke encourages communication and social

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 579

11.20 Spasticity and contractures Table 11.26 Advantages and disadvantages of nasogastric and percutaneous endoscopic gastrostomy feeding.

Advantages

Disadvantages

Nasogastric tube

Widely available Cheap

Percutaneous endoscopic gastrostomy

Tube rarely displaced Cosmetically acceptable Long-term use practical

Easily inserted into lungs Often pulled out Unsightly Uncomfortable May lead to aspiration Nasal irritation/ulcer Gastric bleeding Interference with swallow Limited availability Expensive Aspiration when sedated for insertion Wound infection Bleeding Peritonitis

(a)

(b) Fig. 11.21 (a) A mitten used to prevent a patient from removing a nasogastric tube. (b) An American football helmet to prevent a patient from pulling out a nasogastric tube.

interaction. Patients who are concerned by their appearance after a stroke (e.g. due to facial weakness) may gain confidence from this opportunity to socialize. On the other hand, dribbling or ending up with the food down their shirt, or on the table or floor, may distress patients and inhibit social interaction. Advice from the occupational therapist regarding equipment to allow the patient to eat independently should be sought to maximize the positive, and minimize the negative, aspects of eating in a communal setting (section 11.32.6).

..

..

(Reproduced from Levine and Morris, 1995460 with permission.)

11.20 Spasticity and contractures

Spasticity has been defined as a motor disorder characterized by velocity-dependent increase in muscle tone with exaggerated tendon reflexes. It develops in between one-fifth and one-third of hospital-admitted stroke patients over the first year and is more common in those with hemiparesis and after severe strokes, and is more

579

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 580

580

Chapter 11 What are this patient’s problems?

common in the arm than the leg.223,224 However, spasticity contributes to disability in only a minority of patients with hemiplegia.224,225 Spasticity is usually accompanied by muscle weakness and clumsiness and sometimes by flexor or extensor spasms. Immediately after a stroke the muscular tone in the limbs and trunk may be lower, the same as, or higher than normal. The reasons for this variation are unclear. Although tone may be lower than normal in the acute phase, in most patients who do not recover it tends to increase over the first few weeks. In patients with a hemiparesis, the tone in the arm is usually greater in the flexors than extensors, while in the leg it is greater in the extensors than the flexors. This explains the typical hemiplegic posture (i.e. elbow, wrist and fingers flexed and arm adducted and internally rotated with the leg extended at the hip and knee and the foot plantar flexed and inverted). Tone in the truncal muscles may also be abnormally high or low. Tone may increase in any muscle group so much that it restricts the active movement which the residual muscle strength can produce. Imbalance in muscle tone can eventually result in shortening of muscles and permanent deformity and so restrict the full range of movement, i.e. contractures. Associated reactions are involuntary movements of the affected side (most typically flexion of the arm) elicited by a variety of stimuli including the use of unaffected limbs (e.g. self-propelling a wheelchair, yawning or coughing) and the upright posture.99,226 Associated reactions become more obvious with increases in tone. Associated reactions may be misinterpreted as voluntary movements by family and ill-informed staff who should be educated about their significance to prevent them becoming unduly optimistic about the patient’s motor recovery. Spasticity and contractures may cause pain, deformity, disability and, if severe, secondary complications such as pressure ulcers at points of contact between soft tissues (e.g. on the inner aspect of the knees in a patient with contractures of the thigh adductors). Associated reactions may be misinterpreted as voluntary movements by family and ill-informed staff who should be educated about their significance to prevent them becoming unduly optimistic about the patient’s motor recovery. Assessment Tone and spasticity: Physicians are trained to assess the tone in a limb by asking the patient to relax (which is almost guaranteed to have the opposite effect) and then moving the limb through its range of movement at each

Table 11.27 The modified Ashworth scale for the clinical assessment of muscle tone.438 0 1

No increase in muscle tone Slight increase in muscle tone, manifested by a catch and release, or by a minimal resistance at the end of the range of motion when the affected part(s) is moved in flexion or extension 1+ Slight increase in muscle tone, manifested by a catch, followed by minimal resistance throughout the remainder (less than half) of the range of movement 2 More marked increase in muscle tone through most of the range of movement but affected parts easily moved 3 Considerable increase in muscle tone, passive movement difficult 4 Affected parts rigid in flexion or extension

joint at different speeds and noting any resistance to these movements. Unfortunately, many physicians do not appreciate how much tone is influenced by factors such as the patient’s position, anxiety, fatigue, pain and medication. Tone may change from minute to minute. This makes it difficult to assess objectively with good inter-observer reliability. Physiotherapists, who spend far more time than physicians handling patients, are more aware of changes in tone and try to take advantage of them in their therapy. Formal measurement of tone can be attempted using clinical scales (e.g. the modified Ashworth Scale, Table 11.27) or techniques such as electrogoniometry or quantitative neurophysiology. Unfortunately, the last two are not widely available or practical in routine clinical practice. The inter-observer reliability of the modified Ashworth Scale is generally good for assessing tone in the arm and around the knee but poor for assessing tone around the ankle.227 When assessing the effectiveness of treatment in an individual patient, or in a group of patients in RCTs, it may be better to measure function (e.g. walking speed or dressing), or the achievement of a specific goal (to allow the palm of the hand to be accessed to maintain hygiene) than relying on measurement of tone itself. Indeed, the management of abnormal tone, like all other aspects of rehabilitation, should be directed at achieving realistic, relevant and measurable goals (section 10.3.3). Contractures are easier to assess because, by definition, the deformity is fixed. Thus, one can objectively measure the range of movement around a joint and repeat the measure to determine whether or not an intervention has improved the range. However, occasionally an apparent contracture responds to injection of botulinum toxin (see below) indicating that the shortening is due to

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 581

11.20 Spasticity and contractures

muscle contraction rather than permanent shortening of muscles or tendons. Prevention and treatment In rehabilitation our aim is to modulate changes in muscle tone to the patient’s advantage. For instance, to increase tone in a flaccid leg and so provide the patient with a more secure base on which to walk, or reduce tone in the arm to facilitate more active movement. We use several complementary approaches to prevent the development of unwanted patterns of tone and to alleviate existing problems due to spasticity and contractures. Some are applicable to any patient while others are only required to deal with exceptional problems: • Avoidance of exacerbating factors: Pain (section 11.23), urinary retention (section 11.14), severe constipation (section 11.15), skin irritation, pressure ulcers (section 11.16), anxiety (11.31.1) and any other unpleasant stimulus may cause an unwanted increase in tone. These factors must be avoided or alleviated. • Positioning and seating: Poor positioning, especially in immobile patients, can lead to detrimental changes in tone. For example, long periods spent lying supine will increase extensor spasms, presumably by facilitation of basic reflexes. Regular positioning, such as external rotation of the shoulder, appears to reduce unwanted tonal changes.228 Unfortunately, there seems to be quite a lot of uncertainty about the optimum positions for patients with hemiplegia and there are also practical difficulties in keeping them in the required position, i.e. patients tend to move.229 The positioning charts which can be found on most stroke units (Fig. 11.22) can only be used as a general guide. Finding the optimum position for a patient is often a matter of trial and error and relies on the experience of the therapists and nurses. Appropriate seating which can be tailored to the individual patient is essential (Fig. 11.23). Ideally, patients should be seated in a balanced, symmetrical and stable position which they find comfortable and which enables them to function, e.g. eat, drink, etc. • Passive movements and physiotherapy: It is important that muscles are not allowed to remain in a relaxed and shortened position for too long. For example, if the arm is left in a flexed posture at the elbow, or the foot plantar flexed in bed, this can lead to permanent shortening and contractures. Patients’ limbs should be moved passively to stretch the muscles and maintain the range of movement, even in the very acute stages. However, it is important that carers are taught to do this without damaging vulnerable structures such as the shoulder (section 11.24). Also, handling methods, during transfers for example, can influence tone, at

..

..

least in the short term.230 Physiotherapists have an important role in teaching nurses and informal carers the correct positioning and handling techniques to minimize risk of injury (to patient and handler) and unwanted spasticity. Patients with a hemiparesis often attempt to ‘overuse’ their sound side to achieve mobility but, as a consequence, the tone may increase in the affected side. Using the unaffected leg to self-propel a wheelchair is said to lead to increased spasticity in the affected arm and leg.231 The impact of any changes in tone associated with such activities on functional recovery has not been established.232 The aim of many of the facilitation and inhibition techniques used by therapists is to use basic reflexes and postural changes in tone to promote function. However, these techniques, although widely accepted and practised, have not been evaluated adequately in RCTs (section 11.21). Various physical techniques including the application of cold, of heat, splinting and electrical stimulation can, at least for a short time, reduce spasticity.230,233,234 This may relieve discomfort, allow improved hygiene, and plaster casts or splints to be applied. Whether these physical techniques have direct longer-term benefits is unclear:233 • Splinting and casting: Occasionally these may be necessary to prevent or treat contractures. Progressive splinting and application of casts can improve range of movement but the optimum duration and best methods are unclear. Also, badly fitted casts and splints can cause pain, pressure ulcers and tendon damage which may exacerbate rather than relieve spasticity and contractures. • Oral antispastic drugs: Where spasticity is not adequately controlled by physical techniques, or where the patient is suffering from painful muscle spasms, certain drugs have been advocated: baclofen, dantrolene, tizanidine and diazepam which all reduce tone by altering neurotransmitter function or ion flux in the spinal cord, or in muscles. But there is little reliable evidence from RCTs to indicate that they usefully reduce spasticity or improve function in stroke patients.235 Indeed, in our experience they rarely make much difference and they have a number of adverse effects (Table 11.28). Adverse effects are much more common in older patients but can be minimized by starting with low doses and increasing the dose slowly until the desired effect is achieved, or adverse effects necessitate withdrawal. We normally only use baclofen or tizanidine for a trial period in patients who our physiotherapists feel might benefit. • Local and regional treatments: Injection of botulinum toxin directly into muscles can reduce troublesome spasticity. Small RCTs have confirmed that this does

581

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 582

582

Chapter 11 What are this patient’s problems?

Lying on affected side No backrest. One or two pillows for head. Affected shoulder pulled well forward. Place good leg forward on a pillow. Pillow placed behind back.

Sitting up Sitting well back and in centre of chair. Affected arm placed well forward on table or pillow. Feet flat on floor. Knees directly above the feet.

Lying on unaffected side No backrest. One or two pillows for head. Affected shoulder forward with arm on pillow. Place affected leg backward on a pillow. Pillow placed behind back.

Sitting in bed Sitting in bed is not desirable. Affected arm placed on pillow. Legs are straight. Sitting upright and well supported.

indeed reduce tone, which may reduce clawing of the fingers and allow better skin hygiene.236–238 This treatment appears to be safe, at least in the short term, but because its effects wear off over several months, repeated injections may be required and treatment costs can be very high. Sometimes, a single injection can allow simpler measures to be introduced which avoids the need for further injections. Further large trials are in progress to establish whether function really is improved (www.strokecentre.org). Very occasionally, and usually where simple measures (see above) have been inadequate, spasticity can be so troublesome as to warrant more invasive procedures: • local nerve blocks with ethanol or phenol although occasionally useful to solve specifc problems, can

Fig. 11.22 The typical chart used to guide the positioning of stroke patients with hemiplegia (shown in black) whilst in bed.

lead to unwanted muscle weakness and painful dysaesthesias;239 • intrathecal infusion of baclofen using implantable pumps;240 • surgical procedures such as anterior and posterior rhizotomy and more recently lesioning of the dorsal root entry zone (so-called drez-otomy);233 and • tendon lengthening and transfers which can, for example, help to reduce equinovarus deformity.233 Anecdotally, the prevalence of severe contractures appears to have declined dramatically over the last 30 years which is probably a result of improvements in the standards of general care. We now virtually never have to resort to the more invasive procedures described here.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 583

11.21 Limb weakness, poor truncal control and unsteady gait

changes in tone, is associated with the development of a painful shoulder (section 11.24) and swelling of the hand (section 11.25). Poor hand and arm function is a major cause of dependency in activities of daily living. Weakness of the leg, which affects about 45% of patients (Table 11.2), may be severe enough to immobilize the patient and thus predispose to the complications of immobility (section 11.11). Leg weakness, making it difficult to stand, transfer or walk independently, is one of the most important factors prolonging hospital stay in stroke patients. In patients with hemiparesis, which affects about 55% of patients, the arm is usually weaker than the leg (Table 11.2). Assessment

Fig. 11.23 A chair which can be tailored to the individual needs of the patient. This model can be raised, or lowered, the back rest and seat can be adjusted, arms can be altered and extra supports can be inserted.

11.21 Limb weakness, poor truncal control and unsteady gait

These three aspects of the patient’s condition are impossible to separate and will therefore be discussed together. Weakness of an arm, leg or both, sometimes with unilateral facial weakness, is probably the most common and widely recognized impairment caused by stroke. However, there are often associated but less obvious problems with the axial muscles which impair truncal control and walking. Facial weakness, which affects about 40% of patients (Table 11.2), apart from its cosmetic effects, may contribute to dysarthria (section 11.30.2) and cause problems with the oral preparatory phase of swallowing (section 11.17). Weakness of the upper limb, which affects about 50% of patients (Table 11.2), along with

..

..

In assessing function in a patient with stroke, unlike making the diagnosis, it is more useful to observe the range and control of voluntary movements of the limbs than assessing power in individual muscle groups. For instance, do the patients have only coarse control of movement around the hip or shoulder, or have they retained movement at the more distal joints? In stroke patients, distal movements are usually more severely impaired than proximal ones. The assessment of motor function and truncal control has already been described (section 3.3.4). Recovery from a hemiplegic stroke has been likened to early infant development, in that the recovery of truncal control follows the same general pattern as that of a growing child. Head control returns first, followed by rolling over, sitting balance and then standing balance, and lastly the patient can walk with increasing steadiness and speed (Fig. 10.16). After a stroke it is useful to know where the patient is on this ‘developmental ladder’ when assessing prognosis and setting goals for rehabilitation.241 It is also important to assess truncal control and gait since truncal ataxia can occur without limb incoordination in patients with midline cerebellar lesions. Indeed, it is not unknown for patients to undergo full gastrointestinal and metabolic investigation to elucidate the cause of vomiting before their truncal ataxia is noted and a cerebellar stroke diagnosed. It also seems absurd that, although immobility is the main reason for a stroke patient needing to stay in hospital for rehabilitation, mobility and balance are very often not assessed properly by doctors admitting stroke patients.59 Having stressed the importance of testing truncal control and gait, it is important that in doing so neither the patient nor the doctor are put at risk of injury. Poor handling and lifting technique may dislocate a patient’s flaccid shoulder (section 11.24), result in a fracture from a fall (section 11.26) and may even injure the doctor’s

583

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 584

584

Chapter 11 What are this patient’s problems? Table 11.28 Adverse effects of antispasticity drugs. Adverse effect

Baclofen

Diazepam

Dantrolene

Tizanidine

Sedation/central nervous system depression Confusion Hypotonia/weakness Unsteadiness/ataxia Exacerbation of epileptic seizures Psychosis/hallucinations Insomnia Headache Urinary retention Dry mouth Hypotension Nausea/vomiting Diarrhoea/constipation Abnormal liver function Hyperglycaemia Visual disturbance Skin rashes Pericarditis/pleural effusions Blood dyscrasia Withdrawal symptoms Drug interactions

++ + + + ++ ++ + + +

+++ + + ++

+

+

++ +

+ +

+ + + + + + +

+ + + +

+ ++

+ + + +

+ + + ++ +

+ ++

+ +++

+ + + +

+ +

+

+ reported occasionally; ++ quite frequent; +++ potentially fatal.

back. Physiotherapists should ideally provide appropriate training to all staff, and informal carers, who are involved in handling patients. The severity of weakness of individual muscle groups is often graded with the Medical Research Council (MRC) Scale242 (Table 11.29). This was originally designed to assess motor weakness arising from injuries to single peripheral nerves, not stroke. Unfortunately, although the MRC Scale has good inter-observer reliability if applied rigorously,227 it is often misused and the optional expansion to include extra grades (4 plus and 5 minus) makes it even less reliable in the routine recording of motor weakness. The motricity index, a modification of the MRC Scale (Table 11.29), for use in patients with stroke, allows the observer to grade the severity of the hemiparesis rather than each separate muscle group.243 This can be useful in charting patients’ progress for research purposes but is of limited value in routine clinical practice because it is difficult to remember the weights applied to individual movements and it requires a small block to assess grip strength. Several other tools are available for objectively measuring and recording motor function (Table 11.30).

Treatment The amount of spontaneous recovery of motor function is highly variable. The more severe the initial impairment, the less likely is full recovery. One study suggested that motor and sensory function 5 days after stroke onset explained 74% of the variance in motor function at 6 months with the Fugl-Meyer Scale.244 The pattern of recovery of motor function parallels that of other strokerelated deficits, with the most rapid recovery occurring in the first few weeks and then the pace of improvement slows over subsequent months (section 10.2.5). In patients with hemiparesis it is generally thought that motor function in the leg improves more than that in the arm, although this has been questioned.245 Also, unless the patient has some return of grip within 1 month of the stroke, useful return of function is unlikely, although not impossible.246 Physiotherapy is the main therapeutic option in hemiparesis although techniques vary. The two broad approaches most commonly employed are the ‘facilitation and inhibition’ technique and the ‘functional’ approach.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 585

11.21 Limb weakness, poor truncal control and unsteady gait Table 11.29 The MRC Scale of Weakness, and the motricity index which was developed from the MRC Scale for use in stroke patients.242,243 MRC Scale 0 No contraction 1 Flicker or trace of contraction 2 Active movement with gravity eliminated 3 Active movement against gravity 4 Active movement against resistance 5 Full strength The motricity index Arm Pinch grip Elbow flexion (from 90°) Shoulder abduction (from chest) Total arm score Leg Ankle dorsiflexion (from plantar flexed) Knee extension (from 90°) Hip flexion Total leg score Scoring system for pinch grip 0 Pinch grip, no movement 11 Beginnings of prehension 19 Grips block (not against gravity) 22 Grips block against gravity 26 Against pull but weak 33 Normal Scoring system for movements other than pinch grip 0 No movement 9 Palpable contraction only 14 Movement but not against gravity/limited range 19 Movement against gravity/full range 25 Weaker than other side 33 Normal

— — — — — — — —

Table 11.30 Measurements of motor function.439 Impairments MRC Scale (Table 11.29) Motricity index (Table 11.29) Trunk control test Motor club assessment Rivermead motor assessment Dynamometry Disability Arm Nine-hole peg test Frenchay arm test/battery Action research arm test Truncal control/mobility Standing balance Functional ambulation category Timed 10-m walk Truncal control Rivermead mobility index Subsection of Office of Population Censuses and Surveys (OPCS) scale440

..

..

Table 11.31 Common features of facilitation/inhibition-based therapy.249 Recognition of the intimate relationship between sensation and movement Recognition of the importance of basic reflex activity Use of sensory input and different postures to facilitate or inhibit reflex activity and movement Motor relearning based on repetition of activity and frequency of stimulation Treatment of the body as an integrated unit rather than focusing on one part Close personal interaction between the therapist and patient

• The facilitation and inhibition technique is based on the premise that posture and sensory stimuli can modify basic reflex patterns which emerge after cerebral damage. Several workers have developed different treatments based on this concept, the best-known being those of Bobath247 and Brunnstrom.248 Although these techniques differ, certain features are common to all249 (Table 11.31): to achieve as normal a posture and pattern of movement on the affected side as possible. • On the other hand, the functional approach simply aims, through training and strengthening of the unaffected side, to compensate for the impairment to achieve maximum function. For example, patients may be encouraged to transfer and walk as soon as possible after the stroke. Supporters of the facilitation and inhibition approach claim that, although the functional approach might achieve earlier independence, it results in more abnormal patterns of tone and movement which in the long term may lead to contractures and loss of function. Vigorous activity involving the unaffected side may increase the tone in the affected limbs during the activity. This does not seem to occur with all activities (e.g. pedalling) and any long-term effects on tone are unclear.250 We believe there is a place for both approaches in clinical practice. Quite often, where a patient has not achieved useful independent mobility despite a prolonged period of physiotherapy (with the facilitation and inhibition approach), we switch to a functional approach to maximize the patient’s autonomy. For example, we will train patients to transfer independently and self-propel a wheelchair even though this may be associated with, at least in the short term, unwanted changes in tone. There has been very little formal evaluation of the physiotherapy techniques. Although several small RCTs

585

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 586

586

Chapter 11 What are this patient’s problems?

have been reported, no definite conclusions about their relative merits can be drawn.251 In any case, comparisons of different techniques may have limited relevance to current clinical practice as many therapists adopt an eclectic approach, using selected aspects of each technique where appropriate for individual patients. There are, therefore, some important questions about physiotherapy after stroke which need to be answered in properly designed RCTs: • When should physiotherapy start ? • How long should it continue? • What is the optimum intensity of physiotherapy?252 • Which specific therapeutic interventions are the most effective? • Is therapy provided by relatively unskilled therapists as effective as that provided by skilled therapists? • Which patients gain most from physiotherapy and can we prospectively identify them? The results of small RCTs support the hypothesis that physiotherapy, especially focused on achieving particular tasks, improves function even when started late after stroke.253 The trials generally indicate that therapy has a greater impact on specific motor impairments than the resulting disability. This may be because the resulting disabilities are the consequence of sensory and cognitive as well as motor problems. The size of any treatment effect is probably influenced by the intensity of treatment.252 However, many older, sicker patients may not be able to tolerate intensive regimes, which emphasizes the need for research to identify the optimum physiotherapy regime for particular subgroups of patients.254 Given the methodological difficulties in systematically reviewing and performing RCTs of physiotherapy techniques (and those of other therapists), this is a daunting challenge (Table 11.1). Other interventions A large number of physical techniques have been developed with the aim of improving motor function or gait. Some have been evaluated in small RCTs which have included highly selected patients and focused more on impairment than disability as outcome measures. None are supported by enough evidence to recommend their routine use.253 These techniques include: • electromyographic, visual and auditory feedback;255–257 • functional electrical stimulation, which is effective as an orthosis; when applied it can reduce foot drop to facilitate gait, and probably increases muscle strength, but it is unclear whether it improves functional outcome and many patients find it uncomfortable;253,258 • acupuncture259 and transcutaneous electrical nerve stimulation (TENS);260

• treadmill gait retraining with or without bodyweight support;261 • ‘forced use’ or ‘constraint induced therapy’ where the unaffected arm is immobilized for a major part of the day and during physiotherapy sessions;262 • drugs to enhance motor recovery.263,264 Other approaches In this section we have dealt with interventions that aim to decrease disability by improving impairments. The complementary strategy of providing appropriate mobility aids, e.g. walking sticks, splints and wheelchairs is dealt with in section 11.32.1.

11.22 Sensory impairments

In about one-fifth of patients with acute stroke it is impossible to assess sensation adequately because of reduced conscious level, confusion or communication problems, but about one-third of the remainder have impairment of at least one sensory modality (Table 11.2). Sensory problems are more easily identified in patients with right rather than left hemisphere stroke, probably because they have fewer communication difficulties. Severe sensory loss may be as disabling as paralysis, especially when it affects proprioception. Furthermore, loss of pain and temperature sensation in a limb, or sensory loss with neglect, may put a patient at risk of injury from hot water, etc. And disordered sensation with numbness or paraesthesia, even without functional difficulties may, if persistent, be as distressing to some patients as central post-stroke pain (section 11.23). We have discussed some of the difficulties in assessing sensory function earlier (section 3.3.5). Patients often complain bitterly about what appears to the doctor to be a minor change in sensation. Do not underestimate the effect which facial numbness, or tingling in a hand, can have on the morale of a patient. Little is known specifically about the recovery of sensation after stroke, although it probably follows a similar pattern seen in most other impairments (section 10.2.5). However, sensory symptoms may evolve and even become more distressing with time and they may worsen during intercurrent illness (e.g. infections) leading to concerns about recurrent stroke. Under

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 587

11.23 Pain (excluding headache)

these circumstances it is important to give appropriate explanation and reassurance to the patient and any carer. Although patients may be given sensory stimulation as part of their therapy, the effect this or any other intervention has on sensation is unknown.265 Where patients have lost temperature or pain sensation in a limb, especially if there is associated neglect, it is important to counsel them about commonsense strategies to avoid injury to the limb.

11.23 Pain (excluding headache)

Pain is a common complaint among stroke patients. Perhaps one-third require analgesia for pain (excluding shoulder pain) during hospital admission after an acute stroke and, although it becomes less of a problem with time, persisting severe pain may affect about one-fifth of patients.105,266 There are many potential causes, some of which are coincidental and others which are in some way due to the stroke (Table 11.32). Usually, the cause becomes obvious after one has asked the patient about the distribution, nature and onset of the pain and has examined the relevant area. However, some pains (e.g. due to spasticity, axial arthritis and central post-stroke pain) may be difficult for patients to describe and localize. Diagnosis and assessment of analgesic requirements are particularly difficult in patients with communication and cognitive problems.

Table 11.32 Causes of pain after stroke. Section Headache due to vascular pathology Painful shoulder Deep venous thrombosis and pulmonary embolism Pressure ulcers Limb spasticity Fractures Arterial occlusion with ischaemia of limb, bowel or myocardium Coexisting arthritis exacerbated by immobility or therapy Instrumentation, e.g. catheter, intravenous cannulae, nasogastric tube Central post-stroke pain (thalamic pain/ Dejerine–Roussy syndrome)

..

..

3.7.1, 11.9 11.24 11.13 11.16 11.20 11.26

11.23

The treatment of pain depends on the cause. Simple analgesics (e.g. paracetamol) along with some reassurance that the pain does not indicate any serious problem may be all that is needed. Other interventions such as local application of heat or cold, and transcutaneous electrical nerve stimulation (TENS) or acupuncture, may relieve symptoms with a low risk of adverse effects. Pain due to spasticity should initially be treated by alleviating exacerbating factors and by carefully positioning the patient (section 11.20). Antispasticity drugs are occasionally required, seldom work and have significant adverse effects (Table 11.28). Musculoskeletal pain can be treated with simple analgesics and, if these are ineffective and there are no contraindications, a non-steroidal anti-inflammatory drug. If a joint becomes acutely painful it may be necessary to rest it and investigate to exclude more serious causes, e.g. septic arthritis, fracture, or gout exacerbated by diuretics or aspirin. We find it useful to discuss the patient’s pain at the multidisciplinary team meeting where one can establish the pattern, severity and control of pain in different settings, e.g. on the ward, at night, during therapy sessions. This provides important clues to its cause and the best approach to treatment. Central post-stroke pain This is variably described as a superficial burning, lacerating or pricking sensation often exacerbated by factors including touch, movement, cold and anxiety.267 It usually affects one-half of the body but may be more localized, affecting a quadrant, one limb or just the face. It affects up to 10% of hospitalized patients surviving at 6 months in total, and 5% severely268 though this estimate is certainly higher than we would expect in our clinical practice. A far less common, but related, problem is post-stroke pruritus.269 Central post-stroke pain is usually associated with some abnormality of pinprick or temperature sensation and may be accompanied by autonomic changes, for example sweating or cold.270 It may start immediately after the stroke but more frequently after a delay of weeks or months.267 Although the term ‘thalamic pain’ is commonly used synonymously with central poststroke pain, this is misleading since the pain occurs in patients with stroke lesions affecting any part of the sensory pathways.271 Central post-stroke pain is often resistant to therapy. Avoidance of those factors which exacerbate the symptoms is an important first step. Tricyclic antidepressants (e.g. amitryptiline) may alleviate the symptoms and also lift any associated depression.272 A small dose should be used initially, usually at bedtime, and be increased

587

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 588

588

Chapter 11 What are this patient’s problems?

slowly until adequate symptom control is achieved or adverse effects become troublesome. If tricyclic antidepressants are ineffective, addition of lamotrigine or gabapentin are reasonable second-line treatments. A range of alternative drugs including other anti-epileptics (e.g. carbamazepine, phenytoin, valproate, clonazepam), anti-arrhythmics (e.g. mexiletine), anaesthetics (e.g. ketamine), opiates and intrathecal baclofen have all been advocated but none adequately evaluated in placebo-controlled randomized trials.272–274 Physical methods such as acupuncture and TENS are worth trying since they may occasionally provide relief and are at least fairly safe without any lasting adverse effects. Psychological interventions may help but have not been formally evaluated. More invasive and destructive techniques such as stereotactic mesencephalic tractotomy, or deep-brain stimulation, are occasionally used in severe cases which are resistant to other treatment modalities, but are not necessarily effective.

11.24 Painful shoulder

Shoulder pain is reported by at least one-fifth of patients in the first 6 months after stroke. Its frequency increases over the first few months. It is more common in those with severe sensorimotor deficits, and thus hospitaladmitted patients, and usually affects the shoulder on the hemiparetic side.275–277 Although many factors (Table 11.33) have been associated with painful shoulder, their role in its development is unclear.277 A small proportion of patients who complain of a painful shoulder after a stroke have the other clinical features which comprise the syndrome of reflex sympathetic dystrophy or shoulder–hand syndrome (Table 11.34). It is unclear whether this represents a distinct entity or simply the severe end of a spectrum. Our ignorance of the causes and prognosis of shoulder pain is in part due to major problems of definition and the lack of well-validated and reliable assessment tools.278 But, although our understanding of the epidemiology and causes of painful shoulder is incomplete, no one involved in stroke rehabilitation can doubt its importance. It causes patients great discomfort, it may seriously affect morale, and can inhibit recovery. In some patients it persists for months, even years.

Table 11.33 Factors associated with painful shoulder after stroke. Section Associated features Shoulder pain before the stroke Low tone allowing glenohumeral subluxation/malalignment Spasticity Severe weakness of the arm Sensory loss Neglect Visual field deficits Neurological mechanisms Reflex sympathetic dystrophy (shoulder–hand syndrome) Central post-stroke pain Brachial plexus injury Orthopaedic problems Adhesive capsulitis (frozen shoulder) Rotator cuff tears due to improper handling or positioning Acromioclavicular arthritis Glenohumeral arthritis Subdeltoid tendinitis

Fig. 11.25 11.20 11.21 11.22 11.28 11.27.1 11.24 11.23

Table 11.34 Features of reflex sympathetic dystrophy. Pain and tenderness on abduction, flexion and external rotation of the arm at the shoulder Pain and swelling over the carpal bones Swelling of metacarpophalangeal and proximal interphalangeal joints Changes in temperature, colour and dryness of the skin of the hand Loss of dorsal hand skin creases, and nail changes Osteoporosis Note: there is probably considerable overlap between reflex sympathetic dystrophy and the cold arms described by Wanklyn286,441 (section 11.25).

Prevention and treatment Treatment of an established painful shoulder is often ineffective so that any measures to prevent its development are important. It is probably useful to introduce policies on the stroke unit which have been identified as being associated with a lower frequency of the problem in non-randomized studies (Table 11.35). It may also be useful to identify individuals who are at particularly high risk, based on the factors in Table 11.33, and make all staff aware of the potential problem. More specific

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 589

11.24 Painful shoulder

(a)

(b)

Fig. 11.24 By supporting the weight of the arm, glenohumeral subluxation can be reduced. This may be achieved when the patient is sitting using an arm support (a) which attaches to the chair or wheelchair or, alternatively, a perspex tray (b) which,

because it is transparent, allows the patient to check on the position of their feet. Both are better than pillows which invariably end up on the floor. Several designs of sling (c) are available to reduce subluxation when patients are upright.

Table 11.35 General measures which may reduce the frequency of painful shoulder after stroke.

unproven.279–281 Many therapists worry that the use of slings and cuffs will inhibit recovery of the arm because it is held in a position which promotes spasticity. When a patient complains of shoulder pain it is important to exclude glenohumeral dislocation (Fig. 11.25), fracture or specific shoulder syndromes. For example, painful arc (supraspinatus tendinitis) may respond better to specific measures (e.g. local steroid injection) although even the evidence supporting treatments for these specific syndromes is poor.282,283 In established painful shoulder many treatments have been suggested, some have been evaluated in small RCTs, but further studies are needed to define the best treatments (Table 11.36). Some interventions are probably harmless (e.g. application of cold, heat, taping, TENS) and if they produce even short-term relief are worth trying. Others have potentially important adverse effects and costs (e.g. oral corticosteroids) and therefore need to be evaluated further before being adopted into routine clinical practice.277

Instruct all staff and carers to: Support the flaccid arm to reduce subluxation Teach patients not to allow the affected arm to hang unsupported when sitting or standing. While sitting they might use one of several arm supports which attach to the chair or wheelchair. All are more effective than a pillow which spends most of the time on the floor. Shoulder/arm orthoses may, depending on their design, prevent subluxation but have not been shown to reduce the frequency of painful shoulder (Fig. 11.24c). Avoid pulling on the affected arm when handling the patient Staff and carers should be trained in methods of handling and lifting patients to avoid traction injuries. Avoid any activity which causes shoulder discomfort Therapy sessions sometimes do more harm than good. Maintain range of passive shoulder movements

interventions including slings, cuffs and taping to support the flaccid arm (Fig. 11.24), and functional electrical stimulation may reduce subluxation but their effect on the frequency of painful shoulder and arm function is

..

..

(c)

Shoulder pain after stroke is common, ill understood, difficult to prevent and none of the suggested treatments are supported by reliable evidence.

589

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 590

590

Chapter 11 What are this patient’s problems? Table 11.36 Treatments used for painful shoulders.

Fig. 11.25 X-ray of shoulder showing glenohumeral subluxation, a common finding in stroke patients, but does it matter? Note the widened joint space (double-headed arrow) and the increased distance between the lower border of the glenoid cavity (short arrow) and the lower border of the humeral head (long arrow). Photograph provided by Dr Allan Stephenson.

11.25 Swollen and cold limbs

Swelling with pitting oedema, and sometimes pain, quite often occurs in the paralysed or neglected hand, arm or leg, usually within the first few weeks.284,285 The swelling may limit the movement of the affected part and the pain not only further restricts movement but also exacerbates spasticity and associated reactions (section 11.20). Some patients complain of coldness of the limb, more often of the arm than the leg.286 Swelling more often occurs in the legs of patients who sit for prolonged periods (section 11.13). Gravity and lack of muscle contraction, which reduce venous and lymphatic return, presumably play a part but autonomic changes and control of regional blood flow may be relevant.287 An isolated painful, swollen hand may be a mild form of the shoulder–hand syndrome. There are a number of other causes which need to be considered (Table 11.37). People with fractures or acute ischaemia of the limb usually complain of severe pain, but stroke patients with sensory loss, visuospatial dysfunction or communication

Physiotherapy Positioning and mobilization Exercises Heat or cold Support Strapping Shoulder/arm orthoses (Fig. 11.24c) Bobath sling Rood shoulder support Arm supports for bed or chairs Lapboard (Fig. 11.24b) Forearm support Wheelchair outrigger (Fig. 11.24a) Medication Systemic Analgesics Non-steroidal anti-inflammatory drugs Corticosteroids442 Antispastic drugs (Table 11.28) Phenoxybenzamine Antidepressants Local Corticosteroid injection of shoulder Local anaesthetic Botulinum toxin to periarticular muscles Stellate ganglion block Other physical Ultrasound Acupuncture Biofeedback Transcutaneous electrical nerve stimulation (TENS) Surgery Sympathectomy Humeral head suspension Relief of contractures

Table 11.37 Causes of swollen limb after stroke. Section Gravity in a dependent limb Lack of muscle contraction Deep venous thrombosis Compression of veins or lymphatics by tumour, etc. Cardiac failure Hypoalbuminaemia Occult injury Acute ischaemia Gout Reflex sympathetic dystrophy

11.13

11.19 11.26

11.24 (Table 11.34)

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 591

11.26 Falls and fractures

difficulties may not, which can lead to these diagnoses being overlooked.

Table 11.38 Factors likely to contribute to falls after stroke. Section

Assessment The other causes of a swollen limb should be excluded by clinical examination and appropriate investigation before attributing the swelling simply to immobility or dependency. Investigation is not usually necessary but Doppler ultrasound or venography to exclude deep venous thrombosis of the leg (section 11.13), a simple X-ray of a swollen wrist or ankle to exclude a fracture, and a serum albumin may be useful. One can monitor the effect of any intervention by simply measuring the circumference of the limb although plethysmography has been used in research studies.288 Treatment The treatment obviously depends on the cause. Where the swelling appears to be due to immobility we try the following: • elevation of the affected limb when at rest; • encourage active movement (this may be impossible with severe weakness but is important if neglect is the main cause); • graduated compression stocking on the leg (full length or below-knee depending on extent of swelling) or a bandage on the arm, although the latter may worsen swelling of the hand; and • intermittent compression of the limb although this was shown to be ineffective in reducing arm swelling in a randomized trial.289 In addition, where the limb is painful, simple analgesia may ameliorate the secondary effects of pain on tone which leads to spasticity and contractures. Diuretics should be avoided unless there is evidence of heart failure, since immobile stroke patients have a tendency to become dehydrated (section 11.18.1) and incontinent (section 11.14).

11.26 Falls and fractures

Falls are common after stroke. About one-third of patients are expected to fall during their in-hospital care.105,290–292 Falls are also frequent following discharge from hospital.293,294 Many factors contribute to this tendency; some are listed in Table 11.38. Combinations of such factors are associated with higher risks, though

..

..

The patient Muscle weakness (especially of 11.21 quadriceps) Sensory loss (especially visual 11.22 & 11.27 impairments) Impaired balance, righting reflexes 11.21 and ataxia Confusion 11.29 Visuospatial neglect, e.g. denial of 11.28 hemiparesis Deformity, e.g. plantarflexion causing 11.20 toe catching Epileptic seizures 11.8 Postural hypotension due to drugs or 11.2.3 & 11.18.1 dehydration The environment Inappropriate footwear, e.g. slippers Slippery floors, deep pile carpets and loose rugs Excess furniture Poorly positioned rails and inappropriate aids Lack of supervision Fire doors (these may close automatically and hit slow-moving stroke patients) Drugs: Sedatives and hypnotics Hypotensive drugs Antispastic drugs Table 11.28 Anti-epileptic drugs

patients with severe post-stroke impairments are often at lower risk because they are unable to mobilize at all.294 Several ‘falls risk scores’ have been developed which incorporate these (and other) factors but none have yet been shown to be practical and sufficiently accurate to gain widespread use.291,295,296 A small proportion of falls occurring in hospital ( 70 years 16.6 s.399 Treatment Where the patient has residual problems, a physiotherapist or occupational therapist can improve mobility.400 Training aimed at improving cardiorespiratory function probably increases walking speed.401 Patients’ mobility

quite often deteriorates, after a period of stability, in the months or years after a stroke. This may be attributed to further strokes, progression of coexisting pathology (e.g. arthritis) or attenuation of the benefits of regular physiotherapy. It is important that whoever is responsible for monitoring the patient’s progress, most often the general (family) practitioner, is alert to this and refers the patient back to the physiotherapist. Although not backed up by RCTs, we have little doubt of the value of so-called ‘top-up’ physiotherapy in patients such as this, although its effectiveness will inevitably depend on the cause of any deterioration. Therapy in the patient’s own home may be more helpful than in a hospital outpatient department since the therapist can then deal with the real, everyday problems that the patient is experiencing. For example, patients can be taught to climb their own stairs and to overcome the particular problems associated with the layout

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 605

11.32 Dependency in activities of daily living Table 11.47 Walking aids.451 Shoes Should be comfortable, supportive and have non-slip soles Occasionally, a smooth sole facilitates a smooth swing phase in patients with foot drop when walking on carpets Walking sticks May improve standing stability and walking speed but must be tailored to the patient (i.e. length, handle shape, weight and appearance)452,453 but not evidence based;454 a long stick can be used to encourage patients to put weight through their affected leg (Fig. 11.26d) Replace worn rubber ferrule (i.e. rubber bit on end of stick) Should be held in the unaffected hand in patients with hemiparesis but this means the patient’s functioning hand is no longer available for other tasks Other uses include extending the patient’s reach, and ‘self-defence’ Tripods and quadrapods Provide a broader base of support than a walking stick Their weight may cause associated reactions and may worsen the quality of gait452 Not useful on uneven ground Walking frames (Fig. 11.26) Available in many sizes and shapes with or without wheels Useful where balance is poor, especially if the patient tends to lean backwards Often give patients confidence Needs good upper limb function and, of course, prevents patient from using hands or carrying things; many patients attach a basket or use a wheeled trolley to overcome this problem Patients often try to pull up on their frame to rise from a chair, which can cause them to fall Frames may trip up patients and be difficult to manoeuvre with lots of furniture and thick carpets Cannot be used on stairs, therefore patients may require several frames, one on each level Beware bent frames, protruding screws, loose hand-grips and worn ferrules Other uses, drying underclothes Ankle–foot orthosis or brace (Fig. 11.27) Usually of thermoplastic construction and tailored to the individual Occasionally useful in spastic foot drop to improve gait pattern,455 but does not improve spasticity456 and may actually exacerbate the problem Beware in patients with oedema and tendency to leg ulcers Knee brace Occasionally required to prevent hyperextension of the knee Functional electrical stimulation An experimental treatment which has been used to correct foot drop after stroke (section 11.21) Hand rails Short grab-rails can be useful to provide support, especially at thresholds and doorways Full-length rails on staircases are invaluable to those with poor balance, confidence or vision Advice on positioning hand rails should be obtained from an occupational therapist; poorly placed hand rails can cause accidents or may not be used

of their own home. Two RCTs which compared the effectiveness of home vs hospital-based physiotherapy showed that improvement in ADL function was slightly greater in those treated at home, but other outcomes were no different.402 Although the therapist may be able to reduce the patient’s degree of impairment, even when intervention starts very late after a stroke, the therapist’s main tool at this point is the provision of mobility aids (Tables 11.47 and 11.48). When one prescribes an aid for patients, it is important to provide them and any carers with training in its use, and to ensure that it is maintained in good

..

..

order (Figs 11.26–11.36). Many patients do not use the aids provided, which is both a waste of resources and an indication of the lack of value of that aid. It is important to follow up patients who have been given aids to ensure they fit the patient’s needs and are actually being used, and to retrieve any unused equipment. If one prescribes an aid, one should teach the patient or carer how to use it, ensure that it solves the problem, check that it is maintained in good working order, and provide follow-up to ensure its continued use or appropriate removal.

605

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 606

606

Chapter 11 What are this patient’s problems? Table 11.48 Other aids to mobility. Chairs (Fig. 11.28) Higher seats (but not so high as to cause the feet to dangle) without too much backward angle, and firm arm rests (of the correct height, shape and length) facilitate transfers It is easier to stand from a chair if the patient can place his/her heels under it when preparing to stand Chairs should be tailored to the individual’s requirements, but existing chairs may be raised or lowered Chairs on castors can be dangerous Ejector chairs (Fig. 11.29) The seat or chair may be sprung to provide an initial lift to allow the patient to rise These must be tailored to the individual’s weight to avoid the patient being catapulted across the room Electric chairs which slowly bring the patient to a semi-erect position are occasionally useful but are expensive, bulky and require training to use properly Beds A firm mattress at the correct height will facilitate transfers from a bed Blocks or leg extensions can be used to modify an existing bed A grab-rail attached to the bed frame or floor can facilitate independent transfers (Fig. 11.30a) It may be impossible to attach aids to a divan-style bed Mechanical devices which help the patient to sit in bed can occasionally be helpful (Fig. 11.30b) Wheelchairs An invaluable means to improve mobility Many hundreds of designs including self-propelled, carer-propelled and electric chairs Need to be tailored to individual’s requirements Must be kept in good working order (Fig. 11.31) Ramps Useful to those in wheelchairs and their carers Many designs available for different situations Lifts Many types, including: Domestic stair-lifts (Fig. 11.32) for those who cannot manage stairs Through-the-floor lifts for accessing upper floors in wheelchairs Short-rise lifts for accessing vehicles in wheelchairs or where there is insufficient space for a ramp Lifts are expensive and may require structural alterations to accommodate them

Patients who are unable to walk outside, or who can walk only short distances, or who have difficulty using their own or public transport, may be helped by putting them in contact with any local transport schemes. For example, in the UK special financial allowances are available to help with the added cost of being immobile, specially adapted taxis can be provided and some shopping centres provide electric wheelchairs. In the last few years there has been a major campaign to improve access to public places for patients with disabilities.

11.32.2 Driving Many people who have had a stroke never return to driving.403,404 There are several reasons: severe residual motor, sensory, visual or cognitive impairments are the most common. In many countries the authorities place restrictions on driving after stroke, and particularly if the stroke is complicated by epileptic seizures (section 11.8). The variation in regulations probably reflects the dearth

of relevant research in this area. Special, and usually more stringent, regulations apply to those who drive commercial vehicles, taxis and heavy goods vehicles. The main issue for driving authorities is not so much risk of recurrence – well under the 20% per annum for normal drivers in the UK – but physical disability and cognitive difficulties which can be so difficult to judge. Patients are often not asked by their doctors whether they drive, thus many continue to drive and put lives at risk in contravention of their local regulations.404–406 It is the doctor’s responsibility to ask patients whether or not they drive so they can be given relevant information. Failure to do so might have serious consequences, although it is unclear whether patients who have had a stroke and return to driving are at increased risk of road traffic accidents.407 There appears to be little agreement about the optimal method of assessing a disabled patient’s fitness to drive. Informal judgements are often made by patients or their family doctors.404 Bedside testing of neurological

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 607

11.32 Dependency in activities of daily living

Fig. 11.26 Walking aids. (a) A typical light-weight walking frame of adjustable height which may be useful if a patient has poor balance but reasonable arm and hand function. (b) A walking frame with wheels which may be better than (a) in patients who tend to fall backwards or who have Parkinson’s disease. (c) A tea trolley which will provide patients with support as well as allow them to take things from place to place. (d) A long walking stick held in the unaffected hand can be used to encourage the patient to put more of their weight through their affected leg.

(a)

(b)

(c)

(d)

impairments including cognitive function, assessments in driving simulators and road tests have all been advocated.408 One might start with a bedside assessment to demonstrate any physical, and probably more importantly visual, visuospatial and cognitive impairments, which would make safe driving very unlikely.409,410

..

..

Patients who pass on the bedside testing could then be assessed on a driving simulator or an ‘off-road’ test which would identify those who are clearly unsafe to drive.411 The remainder could then be given an on-road test in a dual-control car with an appropriately trained instructor,412,413 which can be performed with good inter-rater

607

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 608

608

Chapter 11 What are this patient’s problems?

(a)

(b)

(a)

Fig. 11.27 Ankle supports. (a) A thermoplastic ankle–foot orthosis: the famous AFO splint. Although this is very useful in patients with foot drop secondary to lower motor neurone lesions, in stroke patients it can sometimes increase the tendency to plantarflexion by stimulating the sole of the foot. (b) A lateral ankle support may be useful in patients who have a tendency to invert their foot when walking.

(b) Fig. 11.28 Choosing an appropriate chair for a stroke patient who has difficulties getting from sitting to standing. (a) A bad chair to ensure the user never escapes! The seat is too low and slopes backwards, the arms are soft and offer little resistance to allow the person to push up on them, the base is solid so the patient cannot tuck their feet under them, and this chair is on castors so that if the patient eventually manages to stand the chair slides away and causes them to fall backwards. Note the Velcro pads on the arms to allow a tray to be fixed across the user’s escape route. (b) A good chair to facilitate easy transfers. This chair is of reasonable height, is upright, has firm but padded arms and allows the user to tuck their feet underneath them which makes it easier to get their weight over their feet. This chair does not have wings, which may provide support for the user’s head when sleeping but can stop interaction with other people sitting to either side. Fig. 11.29 An ejector chair. The number of springs (arrow) can be altered depending on the weight of the user. Some patients find that this type of chair pushes them off balance.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 609

11.32 Dependency in activities of daily living

(a) Fig. 11.30 Aids to getting out of bed. Although many people can get out of a bed independently during a therapy session when they are ‘warmed up’, they may need aids to get out in the middle of the night or first thing in the morning. (a) A grabrail attached to the bed can facilitate transfers from lying to

(b) standing and may therefore prevent incontinence at night. The height of this model can be adjusted to suit the individual. (b) A pneumatic mattress elevator can be helpful in getting patients from lying to sitting.

Fig. 11.31 (opposite) Things to look for when examining a wheelchair. This wheelchair has many hazardous and uncomfortable features. Note the white sticky tape covering the name of the hospital on the side of this wheelchair, which wished, very reasonably, to remain anonymous. 1: Check that the tyres are properly inflated (arrow). Flat tyres make the wheelchair hard work to propel and the brakes inoperative. 2: Check the brakes work (arrow). 3: Check that the sides and arms (arrow) are removable to facilitate transfers onto toilets, into cars, etc. 4: Check that the foot plates are not fixed (arrow) or flapping about. Either problem may cause injury. The foot plates should fold neatly and securely out of the way when not in use. 5: Check that the arms (arrow) are properly padded to avoid discomfort and compression neuropathy of the ulnar nerve. 6: Check that the back (arrow) and seat (arrow) are not sagging and that the appropriate cushion is being used to reduce the risk of pressure ulcers. 7: Check that the grips on the handles are in good order (arrow).

reliability. This stepwise approach should minimize the risk of injury to the patient, instructor and other road users. Many countries provide specialist centres for the assessment of disabled drivers. These centres not only assess driving skills, but also provide advice on vehicle adaptations which allow patients with severe physical disability to drive. Training on a driving simulator has been evaluated in a small RCT, and while definite

..

..

conclusions are limited by methodological issues, this approach is promising.414

11.32.3 Toileting The problems of urinary and faecal incontinence have been discussed in sections 11.14 and 11.15, respectively. About 10% of surviving patients are still dependent in

609

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 610

610

Chapter 11 What are this patient’s problems? Table 11.49 Toileting problems and solutions. Cannot get to toilet because of mobility or access problems Solutions: Urinal (Fig. 11.10) Commode, many designs to suit individual needs (Fig. 11.33) Chemical toilet, useful where regular emptying of a commode is not possible Cannot transfer on and off the toilet, or poor balance making manipulation of clothes difficult Solutions: Grab-rails Toilet frame (Fig. 11.33) Raised or adjustable toilet seat (Fig. 11.33) Cannot clean myself Solutions: Toilet paper holder on the unaffected side which can be used with one hand Large aperture toilet seat Self-cleaning toilet, like a bidet and toilet in one

11.32.4 Washing and bathing

Fig. 11.32 A typical stair-lift for domestic use. The arm-rests fold out of the way to facilitate transfers. The foot-rest and seat fold up when not in use. The controls are on the arm-rests.

toileting 1 year after their first stroke due to inability to transfer independently, walk, or dress and undress (Table 11.45). For the cognitively intact patient, this is an embarrassing disability which severely damages their self-esteem. An assessment by an occupation therapist should define the severity and cause of the problem. The patient’s ability to toilet themselves is very dependent on the environment so that a home assessment may be particularly useful. Simple factors such as the width of the door to the toilet or bathroom, the position and height of the toilet, and the position of the toilet roll holder can make a crucial difference to whether or not the patient can use the toilet independently. Therapy aimed at improving performance in mobility, transfers and dressing will all facilitate independence in toileting. Table 11.49 lists some common problems and simple solutions, and Fig. 11.33 shows some simple toilet aids.

About 10% and 30% are dependent in washing and bathing respectively 1 year after a first-ever-in-a-lifetime stroke (Table 11.45). Motor, sensory, visuospatial and cognitive impairments all contribute to these disabilities. Although poor arm function makes washing and grooming more difficult, most patients can perform these tasks with their unaffected arm, but those with visuospatial and cognitive deficits, even when arm function seems quite good, may still be unable to wash and groom themselves independently. Independence in bathing obviously requires some independence in mobility and transfers. Other disabling conditions in the elderly (e.g. painful arthritis) often contribute. A skilled assessment by an occupational therapist, preferably in the patient’s own home, will delineate the problems. This assessment ensures there is adequate access to the bathroom (e.g. for a wheelchair). The size and layout of the bathroom will determine which aids can be used. Therapy aimed at improving those impairments, especially motor impairments, which are contributing to the disability in bathing and washing may be of some help but, again, at this stage the provision of the wide range of available aids to allow the patient to compensate for their impairments is more effective (Fig. 11.34 and Table 11.50). Specific training in the use of bathing aids, both in hospital and in the patient’s home, does appear to increase use of the bath.415 Of course, in many countries patients prefer to take showers rather than a bath, which generally reduces the difficulty. One general point is that a patient who has trouble bathing should ensure that somebody else is in the house and that the bathroom door is left unlocked, so that if they fall or are unable to get out of the bath somebody can easily reach them.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 611

11.32 Dependency in activities of daily living

(a)

(b)

Fig. 11.33 Simple toilet aids. (a) A raised toilet seat with a frame. (b) A commode; this one attaches securely to the bed (arrow) for use at night. Table 11.50 Problems with grooming or bathing/showering and solutions. Cannot brush dentures with one hand Solutions: Half fill sink with water, attach brush with a suction pad to the sink and brush dentures with functioning hand Cannot get into, and even more important, out of the bath Solutions: Grab-rails appropriately mounted around bath Non-slip bath mat Bath stool Bath board Inflatable or hydraulic bath seats which can lower patients into or raise them out of the bath (Fig. 11.34) Hoists Replace bath with shower Fig. 11.34 A pneumatic bath aid to lower patients into and to lift them out of a bath. However, the humble bath rail (arrow) with some other simple aids is very commonly all that is needed to allow the person to use a bath independently.

Patients who have difficulties bathing independently should not be left in their house alone while bathing. Some form of alarm or even a cordless or mobile telephone will increase safety.

..

..

Some bathroom adaptations are expensive and cause considerable disruption to the patient’s home. It is important that a proper assessment is done to avoid unnecessary building work and to ensure that any modifications really are likely to help the patient and/or carer. For some patients who live alone, or where aids and adaptations are not appropriate, it may be necessary to recommend either a simple strip wash or to introduce a nurse, care assistant or family member to help with bathing and so allow the patient to remain in their own home.

611

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 612

612

Chapter 11 What are this patient’s problems?

(a)

(b)

Fig. 11.35 Simple dressing aids. (a) A reaching aid. (b) Elastic laces (short black arrow), spiral laces which one simply pulls tight (long black arrow) and Velcro fasteners (white arrow) are

invaluable dressing aids for those with only one functioning hand or poor manual dexterity.

Table 11.51 Problems with dressing and solutions (Fig. 11.35).

the disability is due simply to a motor deficit, then ‘dressing practice’, where the patient is taught to compensate for their impairments, is often successful.416 If the patient has cognitive or visuospatial problems or other non-stroke problems (e.g. arthritis), the benefit of this type of therapy is less obvious. There has been little formal evaluation of dressing practice. Advice on the best type of clothes can allow patients to become independent, i.e. avoidance of tight clothes with small buttons or zips, and of shoes with laces. Oval buttons are easier to manipulate than round ones.417 Velcro fastenings have revolutionized dressing for patients with only one functioning hand, being easier to use and preferred by patients than other fasteners.417 Some of the common problems and possible solutions are shown in Table 11.51 and Fig. 11.35. If patients still cannot dress themselves, then care assistants or a member of the family will be needed to help.

Cannot manage zips or buttons Solutions: Button hook or toggle button hook allows patient to fasten buttons with one hand Replace round buttons with oval ones Hook to pull up zip Replace zips and buttons with Velcro Cannot get shoes on or tie shoe laces Solutions: Long-handled shoe horn Replace shoe laces with elastic ones or Velcro straps Shoe ties Cannot reach or pull up clothes Solutions: Dressing hook or even a walking stick Reaching aid

11.32.5 Dressing About 20% of surviving patients need help from another person to dress 1 year after their first-ever-in-a-lifetime stroke (Table 11.45). Dependency in dressing is usually due to a combination of arm weakness or incoordination, inability to stand independently to pull up lower garments, and cognitive and visuospatial problems.416 A painful shoulder makes dressing even more difficult (section 11.24). A detailed assessment by an occupational therapist should elucidate the causes and define the degree of disability. Occupational therapists spend a considerable proportion of their time training patients to dress themselves. If

11.32.6 Feeding One year after a first-ever-in-a-lifetime stroke, about 20% of surviving patients need some help from another person to feed (Table 11.45).418 Dependency in feeding is one of the most demeaning problems because of its associations with infancy. Impaired function of the arm and hand is the most common cause, although problems with sitting balance, facial weakness and visuospatial and sensory function may all be important as well. A small number of patients may have residual problems swallowing (section 11.17). Most problems can be

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 613

11.32 Dependency in activities of daily living

(a) Fig. 11.36 Simple feeding aids. (a) A combined fork and knife, and a plate with a high side or plate guard facilitates eating onehanded. The non-slip tray helps by fixing the plate and also reduces the risk of spills. (b) This drinking straw has a one-way

valve so that when the user stops sucking the fluid does not fall back into the beaker (arrow). This may be useful in patients with facial weakness and poor lip closure but who have a normal swallow.

Table 11.52 Feeding problems and solutions (Fig. 11.36).

chair of appropriate height, and that the knife is sharp. Furthermore, the type of food may make the difference between patients being able to feed themselves or not. A hemiplegic patient may be independent eating sandwiches but dependent for eating a steak. Some of the more common feeding problems, and the aids which can help, are shown in Table 11.52 and Fig. 11.36.

Cannot cut up food because of poor arm function Solutions: If moderate hand weakness, a large-handled sharp knife may be enough Alternatively, a combined knife and spoon or fork may help, but warn the patient to be careful not to cut their mouth Cannot hold bowl or plate still when cutting or spooning Solution: A non-slip mat or tray, or containers with suction pads Cannot push peas onto fork or spoon Solution: Provide a dish with a raised side or rim Dribbling due to facial weakness Solutions: A mug with a spout, or a straw may help. Straws are available with a non-return valve if the patient cannot suck using a conventional one because of facial weakness, but has an intact swallow

identified simply by talking to the patient and/or carer, but often it is useful for a trained observer, usually an occupational therapist, to assess the patient. Intensive therapy may improve arm function (section 11.21) but, late after stroke, the provision and training in the use of feeding aids is probably more effective. Sometimes very simple problems, which require very simple solutions, are overlooked. For instance, it is important that any dentures are in good working order and in the mouth rather than in a glass of water on the bedside table, that the patient is sitting at a table on a

..

..

(b)

11.32.7 Preparing food Many patients are unable to prepare food for themselves or their family because of a wide variety of post-stroke impairments. Cognitive and visuospatial problems may put patients, and others living in the same building, in danger from lacerations, burns, fires, gas poisoning, explosions and flooding. Again, problems are easily identified by talking to, or observing, the patient in the kitchen. An occupational therapist may be able to teach patients alternative methods of performing certain tasks and for many with physical rather than cognitive problems there is a large variety of kitchen aids available to help with specific difficulties (Table 11.53). Patients with cognitive problems may be helped by things such as electric kettles which turn themselves off and gas detection systems which interrupt the supply automatically, but for those with severe cognitive problems it may be necessary to ban them from the kitchen or to remove cookers. Where patients cannot prepare food for themselves, it is obviously important to ensure delivery of prepared meals to their homes, or to have a care assistant who

613

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 614

614

Chapter 11 What are this patient’s problems? Table 11.53 Problems with food preparation and solutions. Cannot open cans with one functional hand Solution: Wall-mounted (electric) tin opener Cannot open bottles with one functional hand Solution: A bottle or jar stabilizer to fix the bottles to work surface Cannot cut up or peel ingredients with one functional hand Solution: A plate or board with spikes on which to impale the item to be cut

can prepare meals. Often, relatives and social services provide these services.

11.33 Social difficulties

We have already discussed how stroke-related impairments result in disability or dependency in activities of daily living (ADL). However, there is more to life than just ADL. Environmental factors become extremely important in determining the effect of the stroke on a person’s role in society and their handicap. This section will address some of the social factors which are often of greatest concern to patients and their carers.

11.33.1 Accommodation About 50% of the patients who survive a stroke are dependent in ADL (Table 11.45). Some patients with complex or worsening disability may not, if admitted to hospital, be able to return to their own home. A patient’s ability to return home depends on the answers to the following questions, among others: • Is the disability likely to improve or worsen with time? • Will aids and adaptations reduce dependency? • What is the patient’s accommodation like? • What level of informal support is available, e.g. family? • What level of community support is available, e.g. community nursing, home helps? Any assessment has to identify the needs of the patient to determine whether those needs can be practically provided in the patient’s current accommodation, or whether alternative or modified accommodation will be needed. This may involve one or more visits home with the patient and any carer to establish exactly what the practical problems are. If a short visit is not enough, then home visits (with the potential for ‘round the clock’ supervision) for a few days and nights can be useful to

predict how the patient and carer will cope in the longer term. The layout of the home is obviously important. Are there steps up to the front door which make access difficult? Are the toilet and bathroom upstairs? Is the house cluttered with furniture? Are the carpets deep pile which may cause difficulty for a patient with a foot drop? Some stroke units have predischarge apartments in the hospital which allow the team and the patient to assess under supervision how well the patient copes. This may not be the same as discharge into the community, however. When to do an accommodation assessment is difficult to predict because it often takes a considerable time to find alternative accommodation, or to make major structural changes to the patient’s current home. Thus, if one is going to avoid unnecessary delays in hospital, one has to make a decision before the patient has achieved their optimal functional status. Considerable judgement is required to ‘best guess’ the patient’s final functional level and to identify their accommodation needs well in advance of hospital discharge. After a thorough assessment, the patient and their family, with help from the team, need to decide where they wish to live, taking into account all the practical issues – including any financial constraints. After all, given unlimited funds, it is almost always possible to maintain even the most severely disabled patient at home. The final decision is often a result of negotiation between the patient, carer and team members. For instance, patients with visuospatial or cognitive problems may not appreciate the likely problems which will face their carers after discharge. For these sorts of reasons it is not always possible to fulfil the wishes of both patient and carer. Options for alternative accommodation, and the availability of support in the community, vary from place to place, country to country, and from time to time. However, in general, one needs a range of accommodation offering a variety of levels of support and supervision to meet the needs of individual patients.

11.33.2 Employment About one-third of patients who have a stroke are of employment age. Estmates of the proportion who return to work have varied widely due mainly to methodological factors.419 Obviously, the nature of previous employment, residual impairments and disabilities, the patient’s own wishes, the attitudes and policies of employers, health and safety legislation and insurance issues will determine whether return to employment is feasible.420 It is also likely that the local arrangements for paying sickness benefit will influence whether patients

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 615

11.33 Social difficulties

return to work, and the timing of their return. Of course, many patients who are approaching retirement age may not want to return to work anyway. Assessment of the patient’s ability to return to work is often left to the patient and family, but involvement of members of the team (i.e. occupational therapist, speech and language therapist, physiotherapist and social worker) can be very useful to help explore the possible work options. Some employers have occupational health departments which can provide further specialist advice on their own specific regulations covering return to work, and on any alterations to the working environment or the job itself which would facilitate this. It is important that patients and their families are counselled about the patient’s limitations. They are often under the misconception that patients should rest after a stroke and that physical activity will bring on another stroke.421 This misconception may have made them rule out, inappropriately, the possibility of return to work. Many patients and their carers are under the misconception that exercise, hard work or stress will bring on another stroke. They should be counselled to dispel these myths. Patients may require specific occupational therapy to improve the physical skills required for particular jobs, and/or retraining to change employment. Patients often complain of marked fatigue (section 11.31.4) so that a return to part-time work, at least initially, is often more successful than struggling to cope with full-time work. And they may find that although they can perform the physical aspects of their work, their concentration may be impaired. In many countries, special schemes are available to provide employment for disabled people.

Counselling the patient and their carers about the value of maintaining leisure activities and social contacts is useful, and can have dramatic consequences (Fig. 11.37), along with practical help in achieving them. Therapy specifically directed at improving participation in leisure activities, or using leisure goals to improve activities of daily living function, was not effective in a large multicentre trial.425

11.33.4 Air travel We are frequently asked whether a patient can travel on a commercial flight after a stroke. We are not aware of any research into the safety, or otherwise, of doing so. Prolonged immobility on longhaul flights with the attendant risk of venous thromboembolism, and the effect of altitude (most commercial aeroplanes are pressurized to a level equivalent to an altitude of about 7000 feet) on the brain, are two issues which need to be considered. In all patients, but especially those who have had an ischaemic stroke associated with a patent foramen ovale, it may be wise to emphasize simple measures to reduce the risk of venous thromboembolism such as adequate hydration and avoiding prolonged immobility.426 In giving advice one also needs to take into account: • the importance of the flight (e.g. is it to get home or simply part of a holiday?); • the interval from the stroke which will in part determine the risk of recurrent stroke; • the ability of the patient to manage everyday activities, such as toileting and feeding, on an aeroplane; • the availability of an accompanying carer; • the attitude of the airline and the patient’s travel insurer. Many patients who have a stroke while on holiday are repatriated by air within a week or two without apparent ill effects.

11.33.3 Leisure Two-thirds of stroke patients are retired from employment. For them, resumption of a leisure activity is more relevant than work. Restriction in leisure activities may be the result of physical or cognitive impairments but may also be caused by psychological factors and even fear that an activity may bring on a further stroke. Many disabled stroke patients are unable to continue with their normal leisure activities and do not take up new ones which are within their abilities.422 Reduction in leisure activities will exacerbate social isolation, lower mood and adversely affect relationships with carers. The level of social activities, including leisure, can be measured using the Frenchay activities index; however, the Nottingham leisure questionnaire has been developed specifically for this purpose.423,424

..

..

11.33.5 Sex Although many stroke patients are elderly, they were often sexually active before their stroke. But after stroke, libido, coital frequency and satisfaction are reduced among both patients and their partners.427 This reduction in activity is due to both physical and psychosocial factors, the latter probably being of greater importance (Table 11.54). Reduced sexual activity may contribute to a worsening emotional relationship with their partner. A patient and their partner may believe that reduced satisfaction with their sexual relationship is an inevitable consequence of stroke. Although we often feel embarrassed about talking to people about these aspects of their lives, it is important that sexual problems are discussed. Physical difficulties due to the patient’s

615

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 616

616

Chapter 11 What are this patient’s problems?

(a)

(b)

(c)

impairments can often be overcome with a little commonsense and a realization that intercourse, with vaginal penetration, is not an absolute requirement. The psychological problems contributing to sexual dysfunction are often the most difficult to sort out. Sometimes patients and their partners simply need to be reassured that sexual activity, like any other physical activity, will not precipitate another stroke. Verbal information may usefully be supplemented by leaflets, supplied by charities and patient organizations, which include advice for patients and carers.

Fig. 11.37 Leisure activities for patients with mobility problems after stroke: (a) abseiling in a wheelchair; (b) preparing to go canoeing; (c) horse riding. (Photographs by Renzo Mazzolini for the Chest Heart and Stroke Association, Scotland.)

We encourage patients to resume sexual activity as soon as they wish after a stroke. One exception might be a patient with a recent rupture of an aneurysm which has not for some reason been coiled or clipped. One might imagine that the increase in blood pressure associated with orgasm could cause a rebleed. Unfortunately, strokes can put tremendous strains on a relationship which are more difficult to manage. Where sexual dysfunction is due to medication (Table 11.54), withdrawal and substitution with an alternative drug may be effective. In men with impotence, sildenafil

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 617

11.34 Carers Table 11.54 Causes of reduced sexual activity and satisfaction after stroke. Psychosocial factors Loss of interest Fear of impotence Fear of bringing on another stroke Emotional changes may adversely affect the relationship Inability to discuss problems Physical factors Physical disability may make sexual intercourse difficult or impossible, e.g. indwelling catheter, contractures Impotence and reduced libido due to: Drugs (e.g. thiazide diuretics, beta-blockers, tricyclic antidepressants) Comorbidity (e.g. diabetes, peripheral vascular disease)

(Viagra) may be useful although experience of its use after stroke is limited and the manufacturers recommend caution in this situation. Many patients with stroke will be taking nitrates for ischaemic heart disease which precludes the concurrent use of sildenafil. Referral to a sexual dysfunction clinic, or for marital guidance, may be useful if problems persist after attending to the simple and obvious.

11.33.6 Finance Stroke may place a considerable financial burden on patients and their families. Employment, and therefore income, may be affected.428 Disability, and the aids and adaptations needed to overcome it, may also be costly. Even when the patient is still in hospital, carers may have difficulty meeting the costs of transport to visit the patient and, if driving, parking when they get there. This is important because the patient’s morale may suffer if regular visiting is not possible and this may adversely affect outcome. Of course, in some unenlightened health systems, the patient or family may have to pay directly for their healthcare and financial constraints may even prevent patients from receiving the necessary treatment. The professionals involved must be alert to any financial problems which affect patients and be ready to offer help as required. In the UK, and other countries, where the level of government support for home care depends on the patient’s personal finances, a financial assessment may be necessary for planning care. In the UK, social workers are responsible for these assessments although other agencies may become involved. Depending on assessment of their needs, and sometimes a financial assessment, patients or carers may be eligible for financial benefits from government, charitable

..

..

and other sources (e.g. superannuation schemes, insurance policies, etc.).

11.34 Carer problems

Looking after a patient with disability places considerable physical and emotional strains on the carer. Carers worry about the patient’s needs and their ability to fulfil them. Caring may well limit the carer’s employment and leisure activities and lead to social isolation and loneliness. Carers of disabled stroke patients are often anxious and depressed and have poor physical health.429 The extent of these difficulties appears to be related to the patient’s level of dependency, mood, behaviour and cognitive function. However, the magnitude of the burden of caregiving, as perceived by the carer, may relate to their own physical and emotional state as much as that of the patient. The impact that a stroke has on the family changes over time. Periods which are likely to cause carers particular stress and when extra support may be needed are: • immediately after the stroke when the carer has to come to terms with what is potentially a life-threatening event and one which may have a major effect on the patient’s and potential carer’s future life together; • during a prolonged period of in-patient care; visiting may be difficult because of travelling and also because the patient’s behaviour may put emotional pressure on the carer; • around the period of hospital discharge; suddenly the patient who has been looked after by a team of skilled professionals appears, at least to the carer, to be only the carer’s sole responsibility; and • during the weeks and months following hospital discharge when professional support dwindles (sometimes inappropriately and abruptly), friends stop calling, and the carer becomes physically and emotionally exhausted. Several stages of adjustment which families typically go through have been identified (Table 11.55). These do not apply to all carers, but it is useful to be aware of them in managing patients and their families. Although the physical aspects of caring for a person with a disabling stroke are hard, it is often the patient’s psychological and resulting behavioural problems which cause most distress to carers. Carers may note a change in personality, the patient may become short-tempered and irritable, depressed or apathetic. Such changes may lead to a deterioration in their relationship which may

617

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 618

618

Chapter 11 What are this patient’s problems?

Stage 1:

Crisis

Stage 2:

Treatment stage

Stage 3:

Realization of disability

Stage 4:

Adjustment

Shock Confusion High anxiety High expectation of recovery Denial that disability is permanent Periods of grieving Fears for the future Job Mobility Lifestyle About coping Anger Feelings of rejection Despair Frustration Depression

be compounded by a cessation or disturbance of their sexual relationship (section 11.33.5). Carers often have feelings of guilt which add to their distress. They worry that they contributed to the stroke, perhaps by giving the patient the wrong diet or because of some petty incident which the carer feels should have been avoided. They feel guilty about not visiting enough or for not having the patient home soon enough. After hospital discharge they feel guilty about wanting to carry on with their own lives. They often worry that the patient will fall, have another stroke or even die unless they are in constant attendance. These fears, apart from adding to the distress of carers, may also cause the carers to become overprotective towards the patient, which may prejudice the patient’s outcome. Assessment It is important that all those involved in managing a stroke patient are aware of the burden that caring for a disabled patient places on the family and other carers. The carer should be invited to discuss their problems. This is usually best done when the patient is absent since carers often feel uncomfortable or guilty when talking about their own problems if the patient is present. Indeed, carers often need a lot of encouragement to discuss their problems at all. However, they should be encouraged to do so not only for their own sake, but also for the patient’s. If they are not coping this will adversely affect the patient. It is important to assess the carer’s physical and mental ability to go on providing the necessary care. If the patient is in hospital, it is often useful to get the carer to help with the patient’s nursing care and attend their

Table 11.55 Stages of adjustment amongst families of patients with stroke.428

therapy sessions. This gives the carer an indication of what caring may involve, it can help the carer and team members identify and hopefully resolve problems before discharge, and it provides a valuable opportunity to ‘train’ the carer. Pre-discharge visits home for a day or weekend fulfil similar functions. Carers may need physical or simply psychological support. A wide range of assessment tools have been developed to measure the amount of caregiving provided and the subjective burden this places on the caregiver (e.g. caregiver strain index).430 Prevention and treatment Physical support is usually limited while the patient is in hospital, although some carers may need help with finances (section 11.33.6) and visiting. However, physical support is likely to become more important after the patient is discharged home. Examples include: • providing help with housework to give the carer more time for providing personal care; • providing a care assistant or district nurse to help with the patient’s personal care; • providing training in practical aspects of caring e.g. transfers, dressing;431 • providing a laundry service if the patient has persisting urinary or faecal incontinence; • arranging for the patient to attend a day hospital or day centre, or arranging ‘patient sitting’ services to allow the carer to go shopping, have their hair done, or attend some social function; and • arranging regular respite admissions to a hospital or nursing home to allow the carer to go on holiday, or simply to have a well-earned rest.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 619

References Table 11.56 Common questions asked by carers. Acutely What is a stroke? Will they die? Will they be disabled? Why did this happen? Was it my fault? Will it happen again? After hospital discharge How long will they keep improving for? Will their speech get better? Why are they not the person I knew before? Can I leave them alone to go out? Can they exercise, will it bring on another stroke? Will I always feel so tired? Where can I get help with money? Where can I get help with bathing? Can I have a rest or holiday, i.e. respite?

Such services may be expensive but they probably prevent or delay the need for long-term institutional care which is even more expensive. Psychological support: Carers often need help coming to terms with the changes in the person who has had a stroke. They have many questions and sources of concern (Table 11.56). Support is needed while the patient is in hospital and thereafter. It may take a variety of forms: • an informal talk with the consultant, nurse, therapist or social worker; these are valuable opportunities for carers to ask questions; • a carers’ group where they can ask team members questions, share experiences and provide mutual support; • formal sessions with a counsellor which may help them come to terms with their problems.432 The setting in which support is given needs to be tailored to the individual since, for example, not everybody wants to attend a group. The need for these sorts of services may not be apparent while patients are in hospital, or when they attend an outpatient department. Also, for patients who are not admitted after their stroke, or who remain in hospital for just a few days, the opportunities for the patient and their carer to ask questions and obtain advice are often limited. One approach is to provide a dedicated stroke family support worker who can identify the physical and emotional needs of patients and their families and try to meet them using all available resources. In many places this role is already, at least partially, carried out by social workers and other members of the team, but there are often difficulties in bridging the gap between hospital and community care. Although such support is valued by its recipients, its impact on patient and carer outcomes is unclear (section 17.8.2).

..

..

Giving information (section 17.10.1): Carers may know little about stroke, its causes and consequences and have often received misleading information from families and friends.421 Carers, like patients, vary in the amount, type and format of information they want about stroke. Information-giving therefore needs to be tailored to the individual. Leaflets, audio or video tapes may usefully reinforce verbal transfer of information but more formal evaluation of their relative effectiveness is required. Education programmes appear to increase carers’ knowledge and satisfaction with information received but not necessarily their emotional outcome.432 It is important that patients and carers are given consistent information and advice to avoid confusion. Good communication between potential providers of information is therefore vital.

References

1 Houston JG, Morris AD, Grosset DG, Lees KR, McMillan N, Bone I. Ultrasonic evaluation of movement of the diaphragm after acute cerebral infarction. J Neurol Neurosurg Psychiatry 1995; 58(6):738–41. 2 Turkington PM, Bamford J, Wanklyn P, Elliott MW. Prevalence and predictors of upper airway obstruction in the first 24 hours after acute stroke. Stroke 2002; 33(8):2037–42. 3 Rowat AM, Dennis MS, Wardlaw JM. Hypoxaemia in acute stroke is frequent and worsens outcome. Cerebrovascular Diseases 2006; 21(3):166–72. 4 Tyson SF, Nightingale P. The effects of position on oxygen saturation in acute stroke: a systematic review. Clin Rehabil 2004; 18(8):863–71. 5 North JB, Jennett S. Abnormal breathing patterns associated with acute brain damage. Arch Neurol 1974; 31(5):338–44. 6 Turkington PM, Bamford J, Wanklyn P, Elliott MW. Effect of upper airway obstruction on blood pressure variability after stroke. Clin Sci (Lond) 2004; 107(1):75–9. 7 Bassetti CL, Milanova M, Gugger M. Sleep-disordered breathing and acute ischemic stroke: diagnosis, risk factors, treatment, evolution, and long-term clinical outcome. Stroke 2006; 37(4):967–72. 8 Rowat AM, Dennis MS, Wardlaw JM. Central periodic breathing observed on hospital admission is associated with an adverse prognosis in conscious acute stroke patients. Cerebrovasc Dis 2006; 21(5–6):340–7. 9 Naughton MT. Pathophysiology and treatment of Cheyne-Stokes respiration. Thorax 1998; 53(6):514–18. 10 Cherniack NS, Longobardo G, Evangelista CJ. Causes of Cheyne-Stokes respiration. Neurocrit Care 2005; 3(3):271–9.

619

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 620

620

Chapter 11 What are this patient’s problems?

11 Jensen LA, Onyskiw JE, Prasad NG. Meta-analysis of arterial oxygen saturation monitoring by pulse oximetry in adults. Heart Lung 1998; 27(6):387–408. 12 Roffe C, Sills S, Wilde K, Crome P. Effect of hemiparetic stroke on pulse oximetry readings on the affected side. Stroke 2001; 32(8):1808–10. 13 Ronning OM, Guldvog B. Should stroke victims routinely receive supplemental oxygen? A quasi-randomized controlled trial. Stroke 1999; 30(10):2033–7. 14 Ali K, Sills S, Roffe C. The effect of different doses of oxygen administration on oxygen saturation in patients with stroke. Neurocrit Care 2005; 3(1):24–6. 15 Wessendorf TE, Wang YM, Thilmann AF, Sorgenfrei U, Konietzko N, Teschler H. Treatment of obstructive sleep apnoea with nasal continuous positive airway pressure in stroke. Eur Respir J 2001; 18(4):623–9. 16 White J, Cates C, Wright J. Continuous positive airways pressure for obstructive sleep apnoea. Cochrane Database Syst Rev 2002; (2):CD001106. 17 Hsu CY, Vennelle M, Li HY, Engleman HM, Dennis MS, Douglas NJ. Sleep disordered breathing after stroke. A randomized controlled trial of continuous positive airway pressure. J Neurol Neurosurg Psychiatry 2006; 77(10): 1143–9. 18 Foerch C, Kessler KR, Steckel DA, Steinmetz H, Sitzer M. Survival and quality of life outcome after mechanical ventilation in elderly stroke patients. J Neurol Neurosurg Psychiatry 2004; 75(7):988–93. 19 Rabinstein AA, Wijdicks EF. Outcome of survivors of acute stroke who require prolonged ventilatory assistance and tracheostomy. Cerebrovasc Dis 2004; 18(4):325–31. 20 Schielke E, Busch MA, Hildenhagen T, Holtkamp M, Kuchler I, Harms L, Masuhr F. Functional, cognitive and emotional long-term outcome of patients with ischemic stroke requiring mechanical ventilation. J Neurol 2005; 252(6):648–54. 21 Rem JA, Hachinski VC, Boughner DR, Barnett HJ. Value of cardiac monitoring and echocardiography in TIA and stroke patients. Stroke 1985; 16(6):950–6. 22 Sandercock P, Bamford J, Dennis M, Burn J, Slattery J, Jones L et al. Atrial fibrillation and stroke: prevalence in different types of stroke and influence on early and long term prognosis (Oxfordshire community stroke project). Br Med J 1992; 305(6867):1460–5. 23 Marini C, De Santis F, Sacco S, Russo T, Olivieri L, Totaro R, Carolei A. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a populationbased study. Stroke 2005; 36(6):1115–19. 24 Sandercock PA, Warlow CP, Jones LN, Starkey IR. Predisposing factors for cerebral infarction: the Oxfordshire Community Stroke Project. Br Med J 1989; 298(6666):75–80. 25 Shafqat S, Kelly PJ, Furie KL. Holter monitoring in the diagnosis of stroke mechanism. Int Med J 2004; 34(6):305–9. 26 Jabaudon D, Sztajzel J, Sievert K, Landis T, Sztajzel R. Usefulness of ambulatory 7-day ECG monitoring for the detection of atrial fibrillation and flutter after acute

27

28

29

30

31

32

33

34 35

36

37

38

39

40

41 42

stroke and transient ischemic attack. Stroke 2004; 35(7):1647–51. Fure B, Bruun WT, Thommessen B. Electrocardiographic and troponin T changes in acute ischaemic stroke. J Int Med 2006; 259(6):592–7. Khechinashvili G, Asplund K. Electrocardiographic changes in patients with acute stroke: a systematic review. Cerebrovasc Dis 2002; 14(2):67–76. Ay H, Arsava EM, Saribas O. Creatine kinase-MB elevation after stroke is not cardiac in origin: comparison with troponin T levels. Stroke 2002; 33(1):286–9. Dimant J, Grob D. Electrocardiographic changes and myocardial damage in patients with acute cerebrovascular accidents. Stroke 1977; 8(4):448–55. James P, Ellis CJ, Whitlock RM, McNeil AR, Henley J, Anderson NE. Relation between troponin T concentration and mortality in patients presenting with an acute stroke: observational study. Br Med J 2000; 320(7248):1502–4. Di Angelantonio E, Fiorelli M, Toni D, Sacchetti ML, Lorenzano S, Falcou A et al. Prognostic significance of admission levels of troponin I in patients with acute ischaemic stroke. J Neurol Neurosurg Psychiatry 2005; 76(1):76–81. Mooe T, Olofsson BO, Stegmayr B, Eriksson P. Ischemic stroke. Impact of a recent myocardial infarction. Stroke 1999; 30(5):997–1001. Davenport RJ, Dennis MS, Warlow CP. Gastrointestinal hemorrhage after acute stroke. Stroke 1996; 27(3):421–4. Kitamura T, Ito K. Acute gastric changes in patients with acute stroke. Part 1: with reference to gastroendoscopic findings. Stroke 1976; 7(5):460–3. Dennis MS, Lewis SC, Warlow C. Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomised controlled trial. Lancet 2005; 365(9461):764–72. Koch M, Dezi A, Ferrario F, Capurso I. Prevention of nonsteroidal anti-inflammatory drug-induced gastrointestinal mucosal injury: a meta-analysis of randomized controlled clinical trials. Arch Intern Med 1996; 156(20):2321–32. Rostom A, Dube C, Wells G, Tugwell P, Welch V, Jolicoeur E, McGowan J. Prevention of NSAID-induced gastroduodenal ulcers. Cochrane Database Syst Rev 2002; (4):CD002296. Wang Y, Zeng C, Wu Z. Proton pump inhibitor and H2RA pharmacological prevention of stress ulcer bleeding in stroke patients: a systematic review of randomized controlled trials. Chinese J Evidence-based Med 2006; 6(2):107–16. Plum F, Posner JB. Supratentorial lesions causing coma. In: Plum F, Posner JB, eds. The Diagnosis of Stupor and Coma. Philadelphia: FA Davis Company, 1980, pp. 134–6. Parvizi J, Damasio AR. Neuroanatomical correlates of brainstem coma. Brain 2003; 126(Pt 7):1524–36. De MA, Melo TP, Crespo M, Ferro JM. Comas in the emergency room of a central hospital. Acta Med Port 1992; 5(8):429–32.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 621

References

43 Frank JI. Large hemispheric infarction, deterioration, and intracranial pressure. Neurology 1995; 45(7):1286–90. 44 Weir CJ, Bradford AP, Lees KR. The prognostic value of the components of the Glasgow Coma Scale following acute stroke. Q J Med 2003; 96(1):67–74. 45 Wijdicks EF, Kokmen E, O’Brien PC. Measurement of impaired consciousness in the neurological intensive care unit: a new test. J Neurol Neurosurg Psychiatry 1998; 64(1):117–19. 46 Walther SM, Jonasson U, Gill H. Comparison of the Glasgow Coma Scale and the Reaction Level Scale for assessment of cerebral responsiveness in the critically ill. Intensive Care Med 2003; 29(6):933–8. 47 Castillo J. Deteriorating stroke: diagnostic criteria, predictors, mechanisms and treatment. Cerebrovasc Dis 1999; 9 Suppl 3:1–8. 48 Scandinavian Stroke Study Group. Multicenter trial of hemodilution in ischemic stroke: background and study protocol. Scandinavian Stroke Study Group. Stroke 1985; 16(5):885–90. 49 Birschel P, Ellul J, Barer D. Progressing stroke: towards an internationally agreed definition. Cerebrovasc Dis 2004; 17(2–3):242–52. 50 Barber M, Stott DJ, Langhorne P. An internationally agreed definition of progressing stroke. Cerebrovasc Dis 2004; 18(3):255–6. 51 Davalos A, Toni D, Iweins F, Lesaffre E, Bastianello S, Castillo J. Neurological deterioration in acute ischemic stroke: potential predictors and associated factors in the European cooperative acute stroke study (ECASS) I. Stroke 1999; 30(12):2631–6. 52 Karepov VG, Gur AY, Bova I, Aronovich BD, Bornstein NM. Stroke-in-evolution: infarct-inherent mechanisms versus systemic causes. Cerebrovasc Dis 2006; 21(1–2):42–6. 53 Barber M, Wright F, Stott DJ, Langhorne P. Predictors of early neurological deterioration after ischaemic stroke: a case-control study. Gerontology 2004; 50(2):102–9. 54 Nakamura K, Saku Y, Ibayashi S, Fujishima M. Progressive motor deficits in lacunar infarction. Neurology 1999; 52(1):29–33. 55 Castillo J, Noya M. [Mechanisms of progression of cerebral infarction]. Neurologia 1999; 14(Suppl 2):2–12. 56 Barber M, Langhorne P, Rumley A, Lowe GD, Stott DJ. Hemostatic function and progressing ischemic stroke: D-dimer predicts early clinical progression. Stroke 2004; 35(6):1421–5. 57 Barber M, Langhorne P, Rumley A, Lowe GD, Stott DJ. D-dimer predicts early clinical progression in ischemic stroke: confirmation using routine clinical assays. Stroke 2006; 37(4):1113–15. 58 Davis M, Hollyman C, McGiven M, Chambers I, Egbuji J, Barer D. Physiological monitoring in acute stroke. Age Ageing 1999; 28(Suppl 1):45. 59 Davenport RJ, Dennis MS, Warlow CP. Improving the recording of the clinical assessment of stroke patients using a clerking pro forma. Age Ageing 1995; 24(1):43–8. 60 Rudd AG, Irwin P, Rutledge Z, Lowe D, Wade D, Morris R, Pearson MG. The national sentinel audit for stroke: a tool

..

..

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

for raising standards of care. J R Coll Physicians Lond 1999; 33(5):460–4. Blood pressure in Acute Stroke Collaboration (BASC). Interventions for deliberately altering blood pressure in acute stroke [update of Cochrane Database Syst Rev 2000; (2):CD000039; PMID: 10796286]. Cochrane Database Syst Rev 2001; (3):CD000039. Harper G, Castleden CM, Potter JF. Factors affecting changes in blood pressure after acute stroke. Stroke 1994; 25(9):1726–9. Abboud H, Labreuche J, Plouin F, Amarenco P, GENIC I. High blood pressure in early acute stroke: a sign of a poor outcome? J Hypertens 2006; 24(2):381–6. Aslanyan S, Weir CJ, Lees KR. Elevated pulse pressure during the acute period of ischemic stroke is associated with poor stroke outcome. Stroke 2004; 35(6):e153–e155. Jensen MB, Yoo B, Clarke WR, Davis PH, Adams HR, Jr. Blood pressure as an independent prognostic factor in acute ischemic stroke. Can J Neurol Sci 2006; 33(1):34–8. Leonardi-Bee J, Bath PM, Phillips SJ, Sandercock PA. Blood pressure and clinical outcomes in the International Stroke Trial. Stroke 2002; 33(5):1315–20. Rodriguez-Garcia JL, Botia E, de La Sierra A, Villanueva MA, Gonzalez-Spinola J. Significance of elevated blood pressure and its management on the short-term outcome of patients with acute ischemic stroke. Am J Hypertens 2005; 18(3):379–84. Willmot M, Leonardi-Bee J, Bath PM. High blood pressure in acute stroke and subsequent outcome: a systematic review. Hypertension 2004; 43(1):18–24. White WB, Anwar YA. Evaluation of the overall efficacy of the Omron office digital blood pressure HEM-907 monitor in adults. Blood Press Monit 2001; 6(2):107–10. Sykes D, Dewar R, Mohanaruban K, Donovan K, Nicklason F, Thomas DM, Fisher D. Measuring blood pressure in the elderly: does atrial fibrillation increase observer variability? Br Med J 1990; 300(6718):162–3. Stewart MJ, Gough K, Padfield PL. The accuracy of automated blood pressure measuring devices in patients with controlled atrial fibrillation. J Hypertens 1995; 13(3):297–300. Panayiotou BN, Harper GD, Fotherby MD, Potter JF, Castleden CM. Interarm blood pressure difference in acute hemiplegia. J Am Geriatr Soc 1993; 41(4):422–3. Bath P, Chalmers J, Powers W, Beilin L, Davis S, Lenfant C et al. International Society of Hypertension (ISH): statement on the management of blood pressure in acute stroke. J Hypertens 2003; 21(4):665–72. Horn J, Limburg M. Calcium antagonists for acute ischemic stroke. Cochrane Database Syst Rev 2000; (2):CD001928. Eames PJ, Robinson TG, Panerai RB, Potter JF. The systemic haemodynamic and cerebral autoregulatory effects of bendrofluazide in the subacute post-stroke period. J Hypertens 2004; 22(10):2017–24. Eames PJ, Robinson TG, Panerai RB, Potter JF. Bendrofluazide fails to reduce elevated blood pressure

621

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 622

622

Chapter 11 What are this patient’s problems?

77

78

79 80

81 82

83

84 85

86

87

88

89

90 91

92

93

94

levels in the immediate post-stroke period. Cerebrovascular Diseases 2005; 19(4):253–9. Willmot M, Ghadami A, Whysall B, Clarke W, Wardlaw J, Bath PM. Transdermal glyceryl trinitrate lowers blood pressure and maintains cerebral blood flow in recent stroke [see comment]. Hypertension 2006; 47(6):1209–15. Mistri AK, Robinson TG, Potter JF. Pressor therapy in acute ischemic stroke: systematic review. Stroke 2006; 37(6):1565–71. Camilo O, Goldstein LB. Seizures and epilepsy after ischemic stroke. Stroke 2004; 35(7):1769–75. Burn J, Dennis M, Bamford J, Sandercock P, Wade D, Warlow C. Epileptic seizures after a first stroke: the Oxfordshire Community Stroke Project. Br Med J 1997; 315(7122):1582–7. Kelso ARC, Cock HR. Status epilepticus. Pract Neurol 2006; 5(6):322–33. Marson AG, Williamson PR, Hutton JL, Clough HE, Chadwick DW. Carbamazepine versus valproate monotherapy for epilepsy. Cochrane Database Syst Rev 2000; (3):CD001030. Tudur SC, Marson AG, Clough HE, Williamson PR. Carbamazepine versus phenytoin monotherapy for epilepsy. Cochrane Database Syst Rev 2002; (2):CD001911. Canhao P, Melo TP, Salgado AV. Nausea and vomiting in acute ischemic stroke. Cerebrovasc Dis 1997; 7:220–5. Ferro JM, Melo TP, Oliveira V, Salgado AV, Crespo M, Canhao P, Pinto AN. A multivariate study of headache associated with ischemic stroke. Headache 1995; 35(6):315–19. Arboix A, Garcia-Trallero O, Garcia-Eroles L, Massons J, Comes E, Targa C. Stroke-related headache: a clinical study in lacunar infarction. Headache 2005; 45(10):1345–52. Mitsias PD, Ramadan NM, Levine SR, Schultz L, Welch KM. Factors determining headache at onset of acute ischemic stroke. Cephalalgia 2006; 26(2):150–7. Tentschert S, Wimmer R, Greisenegger S, Lang W, Lalouschek W. Headache at stroke onset in 2196 patients with ischemic stroke or transient ischemic attack. Stroke 2005; 36(2):e1–3. Kumar A, Dromerick AW. Intractable hiccups during stroke rehabilitation. Arch Phys Med Rehabil 1998; 79(6):697–9. Schiff E, River Y, Oliven A, Odeh M. Acupuncture therapy for persistent hiccups. Am J Med Sci 2002; 323(3):166–8. Moretti R, Torre P, Antonello RM, Ukmar M, Cazzato G, Bava A. Gabapentin as a drug therapy of intractable hiccup because of vascular lesion: a three-year follow-up. Neurologist 2004; 10(2):102–6. Bhalla A, Tallis RC, Pomeroy VM. The effects of positioning after stroke on physiological homeostasis: a review. Age Ageing 2005; 34(4):401–6. Barr J, Stocks J, Wagstaff S, Dey P. Positional interventions for acute stroke patients. (Protocol). Cochrane Database System Rev 2004; (4). Drakulovic MB, Torres A, Bauer TT, Nicolas JM, Nogue S, Ferrer M. Supine body position as a risk factor for

95

96

97

98

99

100

101

102

103

104

105

106

107

108

nosocomial pneumonia in mechanically ventilated patients: a randomised trial. Lancet 1999; 354(9193):1851–8. Rowat AM, Wardlaw JM, Dennis MS, Warlow CP. Patient positioning influences oxygen saturation in the acute phase of stroke. Cerebrovasc Dis 2001; 12(1):66–72. Wojner-Alexander AW, Garami Z, Chernyshev OY, Alexandrov AV. Heads down: flat positioning improves blood flow velocity in acute ischemic stroke. Neurology 2005; 64(8):1354–7. Wojner AW, El-Mitwalli A, Alexandrov AV. Effect of head positioning on intracranial blood flow velocities in acute ischemic stroke: a pilot study. Crit Care Nurs Q 2002; 24(4):57–66. Feldman Z, Kanter MJ, Robertson CS, Contant CF, Hayes C, Sheinberg MA et al. Effect of head elevation on intracranial pressure, cerebral perfusion pressure, and cerebral blood flow in head-injured patients. J Neurosurg 1992; 76(2):207–11. Fleuren JF, Nederhand MJ, Hermens HJ. Influence of posture and muscle length on stretch reflex activity in poststroke patients with spasticity. Arch Phys Med Rehabil 2006; 87(7):981–8. Chatterton HJ, Pomeroy VM, Gratton J. Positioning for stroke patients: a survey of physiotherapists’ aims and practices. Disabil Rehabil 2001; 23(10):413–21. Jones A, Tilling K, Wilson-Barnett J, Newham DJ, Wolfe CD. Effect of recommended positioning on stroke outcome at six months: a randomized controlled trial. Clin Rehabil 2005; 19(2):138–45. Kammersgaard LP, Jorgensen HS, Rungby JA, Reith J, Nakayama H, Weber UJ et al. Admission body temperature predicts long-term mortality after acute stroke: the Copenhagen Stroke Study. Stroke 2002; 33(7):1759–62. Hajat C, Hajat S, Sharma P. Effects of poststroke pyrexia on stroke outcome: a meta-analysis of studies in patients. Stroke 2000; 31(2):410–14. Corbett D, Thornhill J. Temperature modulation (hypothermic and hyperthermic conditions) and its influence on histological and behavioral outcomes following cerebral ischemia. Brain Pathol 2000; 10(1):145–52. Langhorne P, Stott DJ, Robertson L, MacDonald J, Jones L, McAlpine C et al. Medical complications after stroke: a multicenter study. Stroke 2000; 31(6):1223–9. Meisel C, Schwab JM, Prass K, Meisel A, Dirnagl U. Central nervous system injury-induced immune deficiency syndrome. Nature Reviews Neuroscience 2005; 6(10):775–86. Schwab JM, Prass K, Meisel A. Secondary immune deficiency after CNS injury: characteristics, pathophysiology and clinical implications. Neuroform 2005; 11(1):5–13. Aslanyan S, Weir CJ, Diener HC, Kaste M, Lees KR. Pneumonia and urinary tract infection after acute ischaemic stroke: a tertiary analysis of the GAIN International trial. Eur J Neurol 2004; 11(1):49–53.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 623

References

109 Khedr EM, El SO, Khedr T, bdel aziz aY, Awad EM. Assessment of corticodiaphragmatic pathway and pulmonary function in acute ischemic stroke patients. Eur J Neurol 2000; 7(5):509–16. 110 Millns B, Gosney M, Jack CI, Martin MV, Wright AE. Acute stroke predisposes to oral Gram-negative bacilli: a cause of aspiration pneumonia? Gerontology 2003; 49(3):173–6. 111 Leibovitz A, Plotnikov G, Habot B, Rosenberg M, Wolf A, Nagler R et al. Saliva secretion and oral flora in prolonged nasogastric tube-fed elderly patients. Isr Med Assoc J 2003; 5(5):329–32. 112 Leibovitz A, Plotnikov G, Habot B, Rosenberg M, Segal R. Pathogenic colonization of oral flora in frail elderly patients fed by nasogastric tube or percutaneous enterogastric tube. J Gerontol A Biol Sci Med Sci 2003; 58(1):52–5. 113 Gosney M, Martin MV, Wright AE. The role of selective decontamination of the digestive tract in acute stroke. Age Ageing 2006; 35(1):42–7. 114 Dromerick AW, Edwards DF. Relation of postvoid residual to urinary tract infection during stroke rehabilitation. Arch Phys Med Rehab 2003; 84(9):1369–72. 115 Niel-Weise BS, van den Broek PJ. Antibiotic policies for short-term catheter bladder drainage in adults. Cochrane Database Syst Rev 2005; (3):CD005428. 116 Niel-Weise BS, van den Broek PJ. Urinary catheter policies for short-term bladder drainage in adults. Cochrane Database Syst Rev 2005; (3):CD004203. 117 Niel-Weise BS, van den Broek PJ. Urinary catheter policies for long-term bladder drainage. Cochrane Database Syst Rev 2005; (1):CD004201. 118 Berger RE. The urine dipstick test useful to rule out infections: a meta-analysis of the accuracy. J Urol 2005; 174(3):941–2. 119 Correia M, Silva M, Veloso M. Cooling therapy for acute stroke. Cochrane Database Syst Rev 2000; (2):CD001247. 120 Chamorro A, Horcajada JP, Obach V, Vargas M, Revilla M, Torres F et al. The early systemic prophylaxis of infection after stroke study: A randomized clinical trial. Stroke 2005; 36(7):1495–500. 121 Vargas M, Horcajada JP, Obach V, Revilla M, Cervera A, Torres F et al. Clinical consequences of infection in patients with acute stroke: is it prime time for further antibiotic trials? Stroke 2006; 37(2):461–5. 122 Davey P, Brown E, Fenelon L, Finch R, Gould I, Holmes A et al. Systematic review of antimicrobial drug prescribing in hospitals. Emerg Infect Dis 2006; 12(2):211–16. 123 Kelly J, Rudd A, Lewis RR, Coshall C, Moody A, Hunt BJ. Venous thromboembolism after acute ischemic stroke: a prospective study using magnetic resonance direct thrombus imaging. Stroke 2004; 35(10):2320–5. 124 De Silva DA, Pey HB, Wong MC, Chang HM, Chen CP. Deep vein thrombosis following ischemic stroke among Asians. Cerebrovasc Dis 2006; 22(4):245–50.

..

..

125 Warlow C, Ogston D, Douglas AS. Deep venous thrombosis of the legs after strokes. Part I: incidence and predisposing factors. Br Med J 1976; 1(6019):1178–81. 126 Zierler BK. Screening for acute DVT: optimal utilization of the vascular diagnostic laboratory. Semin Vasc Surg 2001; 14(3):206–14. 127 Sachdev U, Teodorescu VJ, Shao M, Russo T, Jacobs TS, Silverberg D et al. Incidence and distribution of lower extremity deep vein thrombosis in rehabilitation patients: implications for screening. Vasc Endovascular Surg 2006; 40(3):205–11. 128 Kassai B, Boissel JP, Cucherat M, Sonie S, Shah NR, Leizorovicz A. A systematic review of the accuracy of ultrasound in the diagnosis of deep venous thrombosis in asymptomatic patients. Thromb Haemost 2004; 91(4):655–66. 129 Kelly J, Rudd A, Lewis RR, Coshall C, Parmar K, Moody A, Hunt BJ. Screening for proximal deep vein thrombosis after acute ischemic stroke: a prospective study using clinical factors and plasma D-dimers. J Thromb Haemost 2004; 2(8):1321–6. 130 British Thoracic Society guidelines for the management of suspected acute pulmonary embolism. Thorax 2003; 58(6):470–83. 131 Kelly J, Hunt BJ, Lewis RR, Swaminathan R, Moody A, Seed PT, Rudd A. Dehydration and venous thromboembolism after acute stroke. Q J Med 2004; 97(5):293–6. 132 Asplund K, Israelsson K, Schampi I. Haemodilution for acute ischaemic stroke. Cochrane Database Syst Rev 2000; (2):CD000103. 133 Amaragiri SV, Lees TA. Elastic compression stockings for prevention of deep vein thrombosis. Cochrane Database Syst Rev 2000; (3):CD001484. 134 Roderick P, Ferris G, Wilson K, Halls H, Jackson D, Collins R, Baigent C. Towards evidence-based guidelines for the prevention of venous thromboembolism: systematic reviews of mechanical methods, oral anticoagulation, dextran and regional anaesthesia as thromboprophylaxis. Health Technol Assess 2005; 9(49):iii–x, 1. 135 Mazzone C, Chiodo GF, Sandercock P, Miccio M, Salvi R. Physical methods for preventing deep vein thrombosis in stroke. Cochrane Database Syst Rev 2004; (4):CD001922. 136 Sandercock P, Gubitz G, Foley P, Counsell C. Antiplatelet therapy for acute ischaemic stroke. Cochrane Database Syst Rev 2003; (2):CD000029. 137 Gubitz G, Sandercock P, Counsell C. Anticoagulants for acute ischaemic stroke. Cochrane Database Syst Rev 2004; (3):CD000024. 138 Lacut K, Bressollette L, Le Gal G, Etienne E, De Tinteniac A, Renault A et al. VICTORIAh (Venous Intermittent Compression and Thrombosis Occurrence Related to Intracerebral Acute hemorrhage) Investigators. Prevention of venous thrombosis in patients with acute intracerebral hemorrhage. Neurology 2005; 65(6):865–9. 139 Van Dongen CJ, van den Belt AG, Prins MH, Lensing AW. Fixed dose subcutaneous low molecular weight heparins versus adjusted dose unfractionated heparin for venous

623

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 624

624

Chapter 11 What are this patient’s problems?

140

141

142

143

144

145

146 147

148

149

150

151

152

153

154

thromboembolism. Cochrane Database Syst Rev 2004; (4):CD001100. Sandercock P, Counsell C, Stobbs SL. Low-molecularweight heparins or heparinoids versus standard unfractionated heparin for acute ischaemic stroke. Cochrane Database Syst Rev 2005; (2):CD000119. Van den Belt AG, Prins MH, Lensing AW, Castro AA, Clark OA, Atallah AN, Burihan E. Fixed dose subcutaneous low molecular weight heparins versus adjusted dose unfractionated heparin for venous thromboembolism. Cochrane Database Syst Rev 2000; (2):CD001100. Righini M, Paris S, Le Gal G, Laroche JP, Perrier A, Bounameaux H. Clinical relevance of distal deep vein thrombosis. Review of literature data. Thromb Haemost 2006; 95(1):56–64. Ost D, Tepper J, Mihara H, Lander O, Heinzer R, Fein A. Duration of anticoagulation following venous thromboembolism: a meta-analysis. J Am Med Assoc 2005; 294(6):706–15. Kelly J, Hunt BJ, Lewis RR, Rudd A. Anticoagulation or inferior vena cava filter placement for patients with primary intracerebral hemorrhage developing venous thromboembolism? Stroke 2003; 34(12):2999–3005. Nakayama H, Jorgensen HS, Pedersen PM, Raaschou HO, Olsen TS. Prevalence and risk factors of incontinence after stroke. The Copenhagen Stroke Study. Stroke 1997; 28(1):58–62. Brittain KR, Peet SM, Castleden CM. Stroke and incontinence. Stroke 1998; 29(2):524–8. Kolominsky-Rabas PL, Hilz MJ, Neundoerfer B, Heuschmann PU. Impact of urinary incontinence after stroke: results from a prospective population-based stroke register; [see comment]. Neurourol Urodyn 2003; 22(4):322–7. Brittain KR, Castleden CM. Suicide in patients with stroke. Depression may be caused by symptoms affecting lower urinary tract. Br Med J 1998; 317(7164):1016–17. Patel M, Coshall C, Lawrence E, Rudd AG, Wolfe CD. Recovery from poststroke urinary incontinence: associated factors and impact on outcome. J Am Geriatr Soc 2001; 49(9):1229–33. Patel M, Coshall C, Rudd AG, Wolfe CD. Natural history and effects on 2-year outcomes of urinary incontinence after stroke. Stroke 2001; 32(1):122–7. Thomas LH, Barrett J, Cross S, French B, Leathley M, Sutton C, Watkins C. Prevention and treatment of urinary incontinence after stroke in adults. Cochrane Database Syst Rev 2005; (3):CD004462. Hay-Smith J, Herbison P, Ellis G, Moore K. Anticholinergic drugs versus placebo for overactive bladder syndrome in adults. Cochrane Database Syst Rev 2002; (3):CD003781. Khorsandi M, Ginsberg PC, Harkaway RC. Reassessing the role of urodynamics after cerebrovascular accident. Males versus females. Urol Int 1998; 61(3):142–6. Shirran E, Brazzelli M. Absorbent products for the containment of urinary and/or faecal incontinence in adults. Cochrane Database Syst Rev 2000; (2):CD001406.

155 Brosnahan J, Jull A, Tracy C. Types of urethral catheters for management of short-term voiding problems in hospitalised adults. Cochrane Database Syst Rev 2004; (1):CD004013. 156 Wu J, Baguley IJ. Urinary retention in a general rehabilitation unit: prevalence, clinical outcome, and the role of screening. Arch Phys Med Rehabil 2005; 86(9):1772–7. 157 Harari D, Coshall C, Rudd AG, Wolfe CD. New-onset fecal incontinence after stroke: prevalence, natural history, risk factors, and impact. Stroke 2003; 34(1):144–50. 158 Coggrave M, Wiesel PH, Norton C. Management of faecal incontinence and constipation in adults with central neurological diseases. Cochrane Database Syst Rev 2006; (2):CD002115. 159 Berlowitz DR, Wilking SV. Risk factors for pressure sores. A comparison of cross-sectional and cohort-derived data. J Am Geriatr Soc 1989; 37(11):1043–50. 160 Cullum N, McInnes E, Bell-Syer SE, Legood R. Support surfaces for pressure ulcer prevention. Cochrane Database Syst Rev 2004; (3):CD001735. 161 Pancorbo-Hidalgo PL, Garcia-Fernandez FP, Lopez-Medina IM, Varez-Nieto C. Risk assessment scales for pressure ulcer prevention: a systematic review. J Adv Nurs 2006; 54(1):94–110. 162 Allman RM, Walker JM, Hart MK, Laprade CA, Noel LB, Smith CR. Air-fluidized beds or conventional therapy for pressure sores. A randomized trial. Ann Intern Med 1987; 107(5):641–8. 163 McQueen J, MacLennan K, McDiarmid M, Gold S. Educational interventions for healthcare professionals to prevent pressure ulcers. (Protocol). Cochrane Database of Syst Rev 2005; (2). 164 Langer G, Schloemer G, Knerr A, Kuss O, Behrens J. Nutritional interventions for preventing and treating pressure ulcers. Cochrane Database Syst Rev 2003; (4):CD003216. 165 Cullum N, Deeks J, Sheldon TA, Song F, Fletcher AW. Beds, mattresses and cushions for pressure sore prevention and treatment. Cochrane Database Syst Rev 2000; (2):CD001735. 166 Flemming K, Cullum N. Therapeutic ultrasound for pressure sores. Cochrane Database Syst Rev 2000; (4):CD001275. 167 Flemming K, Cullum N. Electromagnetic therapy for the treatment of pressure sores. Cochrane Database Syst Rev 2001; (1):CD002930. 168 Singh S, Hamdy S. Dysphagia in stroke patients. Postgrad Med J 2006; 82(968):383–91. 169 Mann G, Hankey GJ, Cameron D. Swallowing function after stroke: prognosis and prognostic factors at 6 months. Stroke 1999; 30(4):744–8. 170 Veis SL, Logemann JA. Swallowing disorders in persons with cerebrovascular accident. Arch Phys Med Rehabil 1985; 66(6):372–5. 171 Scottish Intercollegiate Guidelines Network. Management of Patients with Stroke: Identification and Management of Dysphagia. SIGN Guideline 78. Scottish Intercollegiate Guidelines Network, 2004.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 625

References

172 Hinchey JA, Shephard T, Furie K, Smith D, Wang D, Tonn S. Formal dysphagia screening protocols prevent pneumonia. Stroke 2005; 36(9):1972–6. 173 Huang JY, Zhang DY, Yao Y, Xia QX, Fan QQ. Training in swallowing prevents aspiration pneumonia in stroke patients with dysphagia. J Int Med Res 2006; 34(3):303–6. 174 Ramsey D, Smithard D, Donaldson N, Kalra L. Is the gag reflex useful in the management of swallowing problems in acute stroke? Dysphagia 2005; 20(2):105–7. 175 Ramsey D, Smithard D, Kalra L. Silent aspiration: what do we know? Dysphagia 2005; 20(3):218–25. 176 Ramsey DJ, Smithard DG, Kalra L. Early assessments of dysphagia and aspiration risk in acute stroke patients. Stroke 2003; 34(5):1252–7. 177 Addington WR, Stephens RE, Gilliland KA. Assessing the laryngeal cough reflex and the risk of developing pneumonia after stroke: an interhospital comparison. Stroke 1999; 30(6):1203–7. 178 Addington WR, Stephens RE, Gilliland K, Rodriguez M. Assessing the laryngeal cough reflex and the risk of developing pneumonia after stroke. Arch Phys Med Rehabil 1999; 80(2):150–4. 179 Smithard DG, O’Neill PA, Park C, England R, Renwick DS, Wyatt R et al. Can bedside assessment reliably exclude aspiration following acute stroke? Age Ageing 1998; 27(2):99–106. 180 Carnaby G, Hankey GJ, Pizzi J. Behavioural intervention for dysphagia in acute stroke: a randomised controlled trial. Lancet Neurol 2006; 5(1):31–7. 181 Dennis M. Dysphagia in acute stroke: a long-awaited trial. Lancet Neurol 2006; 5(1):16–17. 182 Power ML, Fraser CH, Hobson A, Singh S, Tyrrell P, Nicholson DA et al. Evaluating oral stimulation as a treatment for dysphagia after stroke. Dysphagia 2006; 21(1):49–55. 183 Berkovic SF, Bladin PF, Darby DG. Metabolic disorders presenting as stroke. Med J Aust 1984; 140(7):421–4. 184 O’Neill PA, McLean KA. Water homeostasis and ageing. Med Lab Sci 1992; 49(4):291–8. 185 Churchill M, Grimm S, Reding M. Risks of diuretic usage following stroke. Neurorehabil Neural Repair 2004; 18(3):161–5. 186 Whelan K. Inadequate fluid intakes in dysphagic acute stroke. Clin Nutr 2001; 20(5):423–8. 187 Bhalla A, Sankaralingam S, Dundas R, Swaminathan R, Wolfe CD, Rudd AG. Influence of raised plasma osmolality on clinical outcome after acute stroke. Stroke 2000; 31(9):2043–8. 188 Gross CR, Lindquist RD, Woolley AC, Granieri R, Allard K, Webster B. Clinical indicators of dehydration severity in elderly patients. J Emerg Med 1992; 10(3):267–74. 189 Kavouras SA. Assessing hydration status. Curr Opin Clin Nutr Metab Care 2002; 5(5):519–24. 190 Challiner YC, Jarrett D, Hayward MJ, al-Jubouri MA, Julious SA. A comparison of intravenous and subcutaneous hydration in elderly acute stroke patients. Postgrad Med J 1994; 70(821):195–7.

..

..

191 O’Neill PA, Davies I, Fullerton KJ, Bennett D. Fluid balance in elderly patients following acute stroke. Age Ageing 1992; 21(4):280–5. 192 Joynt RJ, Feibel JH, Sladek CM. Antidiuretic hormone levels in stroke patients. Ann Neurol 1981; 9(2):182–4. 193 Kumar S, Fowler M, Gonzalez-Toledo E, Jaffe SL. Central pontine myelinolysis, an update. Neurol Res 2006; 28(3):360–6. 194 Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke 2001; 32(10):2426–32. 195 Van Kooten F, Hoogerbrugge N, Naarding P, Koudstaal PJ. Hyperglycemia in the acute phase of stroke is not caused by stress. Stroke 1993; 24(8):1129–32. 196 Scott JF, Gray CS, O’Connell JE, Alberti KG. Glucose and insulin therapy in acute stroke: why delay further? Q J Med 1998; 91(7):511–15. 197 Kagansky N, Levy S, Knobler H. The role of hyperglycemia in acute stroke. Arch Neurol 2001; 58(8):1209–12. 198 Garg R, Chaudhuri A, Munschauer F, Dandona P. Hyperglycemia, insulin, and acute ischemic stroke: a mechanistic justification for a trial of insulin infusion therapy. Stroke 2006; 37(1):267–73. 199 Gray CS, Hildreth AJ, Alberti GK, O’Connell JE, Collaboration GIST. Poststroke hyperglycemia: natural history and immediate management. [see comment] [erratum appears in Stroke 2004; 35(5):1229]. Stroke 2004; 35(1):122–6. 200 Gray CS, Hildrett AJ, Sandercoat PA, O’Connell JE, Johnston DE, Cartlidge NEF et al., For the GIST Trialists collaboration. Glucose–potassium–insulin infusions in the management of post-stroke hyperglycaemia: the UK Glucose Insulin in Stroke Trial (GIST–UK). Lancet Neurol 2007; 6(5): 397–406. 201 Abarbanell NR. Is prehospital blood glucose measurement necessary in suspected cerebrovascular accident patients? Am J Emerg Med 2005; 23(7):823–7. 202 Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospitalized patients: a prospective study. J Am Med Assoc 1999; 281(21):2013–19. 203 Unosson M, Ek AC, Bjurulf P, von Schenck H, Larsson J. Feeding dependence and nutritional status after acute stroke. Stroke 1994; 25(2):366–71. 204 Choi-Kwon S, Yang YH, Kim EK, Jeon MY, Kim JS. Nutritional status in acute stroke: undernutrition versus overnutrition in different stroke subtypes. Acta Neurol Scand 1998; 98(3):187–92. 205 Smithard DG, O’Neill PA, Parks C, Morris J. Complications and outcome after acute stroke. Does dysphagia matter? Stroke 1996; 27(7):1200–4. 206 Gariballa SE, Parker SG, Taub N, Castleden CM. Influence of nutritional status on clinical outcome after acute stroke. Am J Clin Nutr 1998; 68(2):275–81. 207 Bardutzky J, Georgiadis D, Kollmar R, Schwarz S, Schwab S. Energy demand in patients with stroke who are sedated and receiving mechanical ventilation. J Neurosurg 2004; 100(2):266–71.

625

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 626

626

Chapter 11 What are this patient’s problems?

208 Chalela JA, Haymore J, Schellinger PD, Kang DW, Warach S. Acute stroke patients are being underfed: a nitrogen balance study. [see comment.] Neurocrit Care 2004; 1(3):331–4. 209 Heckmann JG, Stossel C, Lang CJ, Neundorfer B, Tomandl B, Hummel T. Taste disorders in acute stroke: a prospective observational study on taste disorders in 102 stroke patients. Stroke 2005; 36(8):1690–4. 210 Poor nutritional status on admission predicts poor outcomes after stroke: observational data from the FOOD trial. Stroke 2003; 34(6):1450–6. 211 Martineau J, Bauer JD, Isenring E, Cohen S. Malnutrition determined by the patient-generated subjective global assessment is associated with poor outcomes in acute stroke patients. Clin Nutr 2005; 24(6):1073–7. 212 Dennis MS, Lewis SC, Warlow C. Routine oral nutritional supplementation for stroke patients in hospital (FOOD): a multicentre randomised controlled trial. Lancet 2005; 365(9461):755–63. 213 Thorsdottir I, Jonsson PV, Asgeirsdottir AE, Hjaltadottir I, Bjornsson S, Ramel A. Fast and simple screening for nutritional status in hospitalized, elderly people. J Hum Nutr Diet 2005; 18(1):53–60. 214 Arrowsmith H. A critical evaluation of the use of nutrition screening tools by nurses. Br J Nurs 1999; 8(22):1483–90. 215 Mead GE, Donaldson L, North P, Dennis MS. An informal assessment of nutritional status in acute stroke for use in an international multicentre trial of feeding regimens. Int J Clin Pract 1998; 52(5):316–18. 216 Coombes R. NHS safety agency issues guidance on nasogastric tubes. Br Med J 2005; 330:438. 217 Wanklyn P, Cox N, Belfield P. Outcome in patients who require a gastrostomy after stroke. Age Ageing 1995; 24(6):510–14. 218 Norton B, Homer-Ward M, Donnelly MT, Long RG, Holmes GK. A randomised prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding after acute dysphagic stroke. Br Med J 1996; 312(7022):13–16. 219 James A, Kapur K, Hawthorne AB. Long-term outcome of percutaneous endoscopic gastrostomy feeding in patients with dysphagic stroke. Age Ageing 1998; 27(6):671–6. 220 Skelly R, Terry H, Millar E, Cohen D. Outcomes of percutaneous endoscopic gastrostomy feeding. Age Ageing 1999; 28(4):416. 221 Wijdicks EF, McMahon MM. Percutaneous endoscopic gastrostomy after acute stroke: complications and outcome. Cerebrovasc Dis 1999; 9(2):109–11. 222 Teasell R, Foley N, McRae M, Finestone H. Use of percutaneous gastrojejunostomy feeding tubes in the rehabilitation of stroke patients. Arch Phys Med Rehabil 2001; 82(10):1412–15. 223 Leathley MJ, Gregson JM, Moore AP, Smith TL, Sharma AK, Watkins CL. Predicting spasticity after stroke in those surviving to 12 months. Clin Rehabil 2004; 18(4):438–43. 224 Sommerfeld DK, Eek EU, Svensson AK, Holmqvist LW, Von Arbin MH. Spasticity after stroke: its occurrence and

225

226

227

228

229

230

231

232

233 234 235 236

237 238

239

240

association with motor impairments and activity limitations. Stroke 2004; 35(1):134–9. Welmer AK, von Arbin M, Widen HL, Sommerfeld DK. Spasticity and its association with functioning and healthrelated quality of life 18 months after stroke. Cerebrovasc Dis 2006; 21(4):247–53. Bhakta BB, Cozens JA, Chamberlain MA, Bamford JM. Quantifying associated reactions in the paretic arm in stroke and their relationship to spasticity. Clin Rehabil 2001; 15(2):195–206. Gregson JM, Leathley MJ, Moore AP, Smith TL, Sharma AK, Watkins CL. Reliability of measurements of muscle tone and muscle power in stroke patients. Age Ageing 2000; 29(3):223–8. Ada L, Goddard E, McCully J, Stavrinos T, Bampton J. Thirty minutes of positioning reduces the development of shoulder external rotation contracture after stroke: a randomized controlled trial. Arch Phys Med Rehabil 2005; 86(2):230–4. Carr EK, Kenney FD. Positioning of the stroke patient: a review of the literature. Int J Nurs Stud 1992; 29(4):355–69. Brown RA, Holdsworth L, Leslie GC, Mutch WJ, Part NJ. The effects of time after stroke and selected therapeutic techniques on quadriceps muscle tone in stroke patients. Physiother Theory Pract 1993; 9:131–42. Cornall C. Self propelling wheelchairs: the effect on spasticity in hemiplegic patients. Physiother Theory Pract 1991; 7:13–21. Barrett JA, Watkins C, Plant R, Dickenson H, Clayton L, Sharma AK et al. The COSTAR wheelchair study: a twocentre pilot study of self-propulsion in a wheelchair in early stroke rehabilitation. Collaborative Stroke Audit and Research. Clin Rehabil 2001; 15(1):32–41. Barnes MP. Management of spasticity. Age Ageing 1998; 27(2):239–45. Walton K. Management of patients with spasticity: a practical approach. Pract Neurol 2006; 3:342–53. Gallichio JE. Pharmacologic management of spasticity following stroke. Phys Ther 2004; 84(10):973–81. Bhakta BB, Cozens JA, Chamberlain MA, Bamford JM. Impact of botulinum toxin type A on disability and carer burden due to arm spasticity after stroke: a randomised double blind placebo controlled trial. J Neurol Neurosurg Psychiatry 2000; 69(2):217–21. Botulinum toxin A injections improved wrist and finger spasticity after stroke. ACP Journal Club 2003; 138(1):22. Centre for Reviews and Dissemination. Botulinum toxin type A for the treatment of the upper limb spasticity after stroke: a meta-analysis (Provisional record). Database Abstr Rev Effects 2006; 63:20. Centre for Reviews and Dissemination. Treatment of upper extremity spasticity in stroke patients by focal neuronal or neuromuscular blockade: a systematic review of the literature. Database Abstr Rev Effects 2006; (3):26. Centre for Reviews and Dissemination. Intrathecal baclofen for severe spasticity: a meta-analysis. Database Abstr Rev Effects 2006; (3):22.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 627

References

241 Smith MT, Baer GD. Achievement of simple mobility milestones after stroke. Arch Phys Med Rehabil 1999; 80(4):442–7. 242 Medical Research Council. Aids to the Examination of the Peripheral Nervous System. Grimsby: Castle Press, 1982. 243 Demeurisse G, Demol O, Robaye E. Motor evaluation in vascular hemiplegia. Eur Neurol 1980; 19(6):382–9. 244 Duncan PW, Goldstein LB, Matchar D, Divine GW, Feussner J. Measurement of motor recovery after stroke. Outcome assessment and sample size requirements. Stroke 1992; 23(8):1084–9. 245 Duncan PW, Goldstein LB, Horner RD, Landsman PB, Samsa GP, Matchar DB. Similar motor recovery of upper and lower extremities after stroke. Stroke 1994; 25(6):1181–8. 246 Sunderland A, Tinson D, Bradley L, Hewer RL. Arm function after stroke. An evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry 1989; 52(11):1267–72. 247 Bobath B. Adult Hemiplegia Evaluation and Treatment. 2nd ed. London: Heinemann; 1978. 248 Brunnstrom S. Movement Therapy in Hemiplegia. New York: Harper & Row, 1970. 249 Flanagan EM. Methods for facilitation and inhibition of motor activity. Am J Phys Med 1967; 46(1):1006–11. 250 Brown DA, Kautz SA. Increased workload enhances force output during pedaling exercise in persons with poststroke hemiplegia. Stroke 1998; 29(3):598–606. 251 Pollock A, Baer G, Pomeroy V, Langhorne P. Physiotherapy treatment approaches for the recovery of postural control and lower limb function following stroke. Cochrane Database of Systematic Reviews 2006; (3). 252 Kwakkel G, van Peppen R, Wagenaar RC, Wood DS, Richards C, Ashburn A et al. Effects of augmented exercise therapy time after stroke: a meta-analysis. Stroke 2004; 35(11):2529–39. 253 Van Peppen RP, Kwakkel G, Wood-Dauphinee S, Hendriks HJ, Van der Wees PJ, Dekker J. The impact of physical therapy on functional outcomes after stroke: what’s the evidence? Clin Rehabil 2004; 18(8):833–62. 254 Lincoln NB, Parry RH, Vass CD. Randomized, controlled trial to evaluate increased intensity of physiotherapy treatment of arm function after stroke. Stroke 1999; 30(3):573–9. 255 Barclay-Goddard R, Stevenson T, Poluha W, Moffatt ME, Taback SP. Force platform feedback for standing balance training after stroke. Cochrane Database Syst Rev 2004; (4):CD004129. 256 Heller F. [Postural biofeedback and locomotion reeducation in stroke patients]. Ann Readapt Med Phys 48(4):187–95. 257 Van Peppen RP, Kortsmit M, Lindeman E, Kwakkel G. Effects of visual feedback therapy on postural control in bilateral standing after stroke: a systematic review. J Rehabil Med 2006; 38(1):3–9. 258 Pomeroy VM, King L, Pollock A, Baily-Hallam A, Langhorne P. Electrostimulation for promoting recovery

..

..

259

260

261

262

263

264

265

266

267

268

269 270

271

272

273

274

275

of movement or functional ability after stroke. Cochrane Database Syst Rev 2006; (2):CD003241. Wu HM, Tang JL, Lin XP, Lau J, Leung PC, Woo J, Li YP. Acupuncture for stroke rehabilitation. Cochrane Database of Systematic Reviews 2006; (3). Robbins SM, Houghton PE, Woodbury MG, Brown JL. The therapeutic effect of functional and transcutaneous electric stimulation on improving gait speed in stroke patients: a meta-analysis. Arch Phys Med Rehabil 2006; 87(6):853–9. Moseley AM, Stark A, Cameron ID, Pollock A. Treadmill training and body weight support for walking after stroke. Cochrane Database Syst Rev 2006; (3). Hakkennes S, Keating JL. Constraint-induced movement therapy following stroke: a systematic review of randomised controlled trials. Aust J Physiother 2005; 51(4):221–31. Gladstone DJ, Black SE. Enhancing recovery after stroke with noradrenergic pharmacotherapy: a new frontier? Can J Neurol Sci 2000; 27(2):97–105. Martinsson L, Hardemark HG, Wahlgren NG. Amphetamines for improving stroke recovery: a systematic cochrane review. Stroke 2003; 34(11):2766. Carey LM, Matyas TA. Training of somatosensory discrimination after stroke: facilitation of stimulus generalization. Am J Phys Med Rehabil 2005; 84(6):428–42. Jonsson AC, Lindgren I, Hallstrom B, Norrving B, Lindgren A. Prevalence and intensity of pain after stroke: a population based study focusing on patients’ perspectives. J Neurol Neurosurg Psychiatry 2006; 77(5):590–5. Bowsher D. Central pain: clinical and physiological characteristics. J Neurol Neurosurg Psychiatry 1996; 61(1):62–9. Andersen G, Vestergaard K, Ingeman-Nielsen M, Jensen TS. Incidence of central post-stroke pain. Pain 1995; 61(2):187–93. Kimyai-Asadi A, Nousari HC, Kimyai-Asadi T, Milani F. Poststroke pruritus. Stroke 1999; 30(3):692–3. Vestergaard K, Nielsen J, Andersen G, Ingeman-Nielsen M, Arendt-Nielsen L, Jensen TS. Sensory abnormalities in consecutive, unselected patients with central post-stroke pain. Pain 1995; 61(2):177–86. Bowsher D, Leijon G, Thuomas KA. Central poststroke pain: correlation of MRI with clinical pain characteristics and sensory abnormalities. Neurology 1998; 51(5):1352–8. Frese A, Husstedt IW, Ringelstein EB, Evers S. Pharmacologic treatment of central post-stroke pain. Clin J Pain 2006; 22(3):252–60. Wiffen P, Collins S, McQuay H, Carroll D, Jadad A, Moore A. Anticonvulsant drugs for acute and chronic pain [Systematic Review]. Cochrane Database Syst Rev 2006; (3). Wiffen PJ, McQuay HJ, Moore RA. Carbamazepine for acute and chronic pain [Systematic Review]. Cochrane Database Syst Rev 2006; (3). Ratnasabapathy Y, Broad J, Baskett J, Pledger M, Marshall J, Bonita R. Shoulder pain in people with a

627

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 628

628

Chapter 11 What are this patient’s problems?

276

277

278

279

280

281

282

283

284

285

286

287

288

289

stroke: a population-based study. Clin Rehabil 2003; 17(3):304–11. Gamble GE, Barberan E, Laasch HU, Bowsher D, Tyrrell PJ, Jones AK. Poststroke shoulder pain: a prospective study of the association and risk factors in 152 patients from a consecutive cohort of 205 patients presenting with stroke. Eur J Pain 2002; 6(6):467–74. Turner-Stokes L, Jackson D. Shoulder pain after stroke: a review of the evidence base to inform the development of an integrated care pathway. Clin Rehabil 2002; 16(3):276–98. Pomeroy VM, Frames C, Faragher EB, Hesketh A, Hill E, Watson P, Main CJ. Reliability of a measure of post-stroke shoulder pain in patients with and without aphasia and/or unilateral spatial neglect. Clin Rehabil 2000; 14(6):584–91. Ada L, Foongchomcheay A, Canning C. Supportive devices for preventing and treating subluxation of the shoulder after stroke. Cochrane Database Syst Rev 2005; (1):CD003863. Ada L, Foongchomcheay A. Efficacy of electrical stimulation in preventing or reducing subluxation of the shoulder after stroke: a meta-analysis. Aust J Physiother 2002; 48(4):257–67. Price CIM, Pandyan AD. Electrical stimulation for preventing and treating post-stroke shoulder pain. Cochrane Database of Systematic Reviews 2006; (3). Green S, Buchbinder R, Glazier R, Forbes A. Interventions for shoulder pain. Cochrane Database Syst Rev 2000; (2):CD001156. Green S, Buchbinder R, Glazier R, Forbes A. Systematic review of randomised controlled trials of interventions for painful shoulder: selection criteria, outcome assessment, and efficacy. Br Med J 1998; 316(7128):354–60. Geurts AC, Visschers BA, van Limbeek J, Ribbers GM. Systematic review of aetiology and treatment of poststroke hand oedema and shoulder-hand syndrome. Scand J Rehabil Med 2000; 32(1):4–10. Boomkamp-Koppen HG, Visser-Meily JM, Post MW, Prevo AJ. Poststroke hand swelling and oedema: prevalence and relationship with impairment and disability. Clin Rehabil 2005; 19(5):552–9. Wanklyn P, Forster A, Young J, Mulley G. Prevalence and associated features of the cold hemiplegic arm. Stroke 1995; 26(10):1867–70. Wang JS, Yang CF, Liaw MY, Wong MK. Suppressed cutaneous endothelial vascular control and hemodynamic changes in paretic extremities with edema in the extremities of patients with hemiplegia. Arch Phys Med Rehabil 2002; 83(7):1017–23. Post MW, Visser-Meily JM, Boomkamp-Koppen HG, Prevo AJ. Assessment of oedema in stroke patients: comparison of visual inspection by therapists and volumetric assessment. Disabil Rehabil 2003; 25(22):1265–70. Roper TA, Redford S, Tallis RC. Intermittent compression for the treatment of the oedematous hand in hemiplegic stroke: a randomized controlled trial. Age Ageing 1999; 28(1):9–13.

290 Nyberg L, Gustafson Y. Patient falls in stroke rehabilitation. A challenge to rehabilitation strategies. Stroke 1995; 26(5):838–42. 291 Teasell R, McRae M, Foley N, Bhardwaj A. The incidence and consequences of falls in stroke patients during inpatient rehabilitation: factors associated with high risk. Arch Phys Med Rehabil 2002; 83(3):329–33. 292 Perennou D, El FA, Masmoudi M, Benaim C, Loigerot M, Didier JP, Pelissier J. [Incidence, circumstances and consequences of falls in patients undergoing rehabilitation after a first stroke]. Ann Readapt Med Phys 2005; 48(3):138–45. 293 Forster A, Young J. Incidence and consequences of falls due to stroke: a systematic inquiry. Br Med J 1995; 311(6997):83–6. 294 Yates JS, Lai SM, Duncan PW, Studenski S. Falls in community-dwelling stroke survivors: an accumulated impairments model. J Rehabil Res Dev 2002; 39(3):385–94. 295 Nyberg L, Gustafson Y. Fall prediction index for patients in stroke rehabilitation. Stroke 1997; 28(4):716–21. 296 Hyndman D, Ashburn A. Stops walking when talking as a predictor of falls in people with stroke living in the community. J Neurol Neurosurg Psychiatry 2004; 75(7):994–7. 297 Watanabe Y. Fear of falling among stroke survivors after discharge from inpatient rehabilitation. Int J Rehabil Res 2005; 28(2):149–52. 298 Dennis MS, Lo KM, McDowall M, West T. Fractures after stroke: frequency, types, and associations. Stroke 2002; 33(3):728–34. 299 Ramnemark A, Nyberg L, Borssen B, Olsson T, Gustafson Y. Fractures after stroke. Osteoporos Int 1998; 8(1):92–5. 300 Melton LJ, III, Brown RD, Jr., Achenbach SJ, O’Fallon WM, Whisnant JP. Long-term fracture risk following ischemic stroke: a population-based study. Osteoporos Int 2001; 12(11):980–6. 301 Whitson HE, Pieper CF, Sanders L, Horner RD, Duncan PW, Lyles KW. Adding injury to insult: fracture risk after stroke in veterans. J Am Geriatr Soc 2006; 54(7):1082–8. 302 Kanis J, Oden A, Johnell O. Acute and long-term increase in fracture risk after hospitalization for stroke. Stroke 2001; 32(3):702–6. 303 Ramnemark A, Nyberg L, Lorentzon R, Olsson T, Gustafson Y. Hemiosteoporosis after severe stroke, independent of changes in body composition and weight. Stroke 1999; 30(4):755–60. 304 Poole KE, Loveridge N, Barker PJ, Halsall DJ, Rose C, Reeve J, Warburton EA. Reduced vitamin D in acute stroke. Stroke 2006; 37(1):243–5. 305 Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev 2003; (4):CD000340. 306 Lyons RA, Sander LV, Weightman AL, Patterson J, Jones SA, Rolfe B, Kemp A, Johansen A. Modification of the home environment for the reduction of injuries. Cochrane Database Syst Rev 2003; (4):CD003600.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 629

References

307 Parker MJ, Gillespie WJ, Gillespie LD. Hip protectors for preventing hip fractures in older people. Cochrane Database Syst Rev 2005; (3):CD001255. 308 Poole KE, Reeve J, Warburton EA. Falls, fractures, and osteoporosis after stroke: time to think about protection? Stroke 2002; 33(5):1432–6. 309 Farquhar CM, Marjoribanks J, Lethaby A, Lamberts Q, Suckling JA. Long term hormone therapy for perimenopausal and postmenopausal women. Cochrane Database Syst Rev 2005; (3):CD004143. 310 Lotery AJ, Wiggam MI, Jackson AJ, Refson K, Fullerton KJ, Gilmore DH, Beringer TR. Correctable visual impairment in stroke rehabilitation patients. Age Ageing 2000; 29(3):221–2. 311 Gray CS, French JM, Bates D, Cartlidge NE, Venables GS, James OF. Recovery of visual fields in acute stroke: homonymous hemianopia associated with adverse prognosis. Age Ageing 1989; 18(6):419–21. 312 Nelles G, Esser J, Eckstein A, Tiede A, Gerhard H, Diener HC. Compensatory visual field training for patients with hemianopia after stroke. Neurosci Lett 2001 June 29; 306(3):189–92. 313 Loverro J, Reding M. Bed orientation and rehabilitation outcome for patients with stroke and hemianopsia or visual neglect. J Neurol Rehabil 1988; 147:150. 314 Rossi PW, Kheyfets S, Reding MJ. Fresnel prisms improve visual perception in stroke patients with homonymous hemianopia or unilateral visual neglect. Neurology 1990; 40(10):1597–9. 315 Rousseaux M, Bernati T, Saj A, Kozlowski O. Ineffectiveness of prism adaptation on spatial neglect signs. Stroke 2006; 37(2):542–3. 316 Rathore SS, Hinn AR, Cooper LS, Tyroler HA, Rosamond WD. Characterization of incident stroke signs and symptoms: findings from the atherosclerosis risk in communities study. Stroke 2002; 33(11):2718–21. 317 Straube A, Leigh RJ, Bronstein A, Heide W, Riordan-Eva P, Tijssen CC et al. EFNS task force: therapy of nystagmus and oscillopsia. Eur J Neurol 2004; 11(2):83–9. 318 Bowen A, McKenna K, Tallis RC. Reasons for variability in the reported rate of occurrence of unilateral spatial neglect after stroke. Stroke 1999; 30(6):1196–202. 319 Stone SP, Halligan PW, Greenwood RJ. The incidence of neglect phenomena and related disorders in patients with an acute right or left hemisphere stroke. Age Ageing 1993; 22(1):46–52. 320 Appelros P, Nydevik I, Karlsson GM, Thorwalls A, Seiger A. Assessing unilateral neglect: shortcomings of standard test methods. Disabil Rehabil 2003; 25(9):473–9. 321 Qiang W, Sonoda S, Suzuki M, Okamoto S, Saitoh E. Reliability and validity of a wheelchair collision test for screening behavioral assessment of unilateral neglect after stroke. Am J Phys Med Rehabil 2005; 84(3):161–6. 322 Menon A, Korner-Bitensky N. Evaluating unilateral spatial neglect post stroke: working your way through the maze of assessment choices. Top Stroke Rehabil 2004; 11(3):41–66.

..

..

323 Gillen R, Tennen H, McKee T. Unilateral spatial neglect: relation to rehabilitation outcomes in patients with right hemisphere stroke. Arch Phys Med Rehabil 2005; 86(4):763–7. 324 Appelros P, Karlsson GM, Seiger A, Nydevik I. Prognosis for patients with neglect and anosognosia with special reference to cognitive impairment. J Rehabil Med 2003; 35(6):254–8. 325 Appelros P, Nydevik I, Karlsson GM, Thorwalls A, Seiger A. Recovery from unilateral neglect after right-hemisphere stroke. Disabil Rehabil 2004; 26(8):471–7. 326 Stone SP, Patel P, Greenwood RJ, Halligan PW. Measuring visual neglect in acute stroke and predicting its recovery: the visual neglect recovery index. J Neurol Neurosurg Psychiatry 1992; 55(6):431–6. 327 Bowen A, Lincoln NB, Dewey M. Cognitive rehabilitation for spatial neglect following stroke. Cochrane Database Syst Rev 2006; (3). 328 Appelros P, Viitanen M. What causes increased stroke mortality in patients with prestroke dementia? Cerebrovasc Dis 2005; 19(5):323–7. 329 Jorm AF. The Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE): a review. Int Psychogeriatr 2004; 16(3):275–93. 330 Caeiro L, Ferro JM, Albuquerque R, Figueira ML. Delirium in the first days of acute stroke. J Neurol 2004; 251(2):171–8. 331 Santos CO, Caeiro L, Ferro JM, Albuquerque R, Luisa FM. Anger, hostility and aggression in the first days of acute stroke. Eur J Neurol 2006; 13(4):351–8. 332 Arden M, Mayou R, Feldman E, Hawton K. Cognitive impairment in the elderly medically ill: how often is it missed? Int J Geriatr Psychiatry 1993; 8:929–37. 333 Dudas RB, Berrios GE, Hodges JR. The Addenbrooke’s Cognitive Examination (ACE) in the differential diagnosis of early dementias versus affective disorder. Am J Geriatr Psychiatry 2005; 13(3):218–26. 334 Larner AJ. An audit of the Addenbrooke’s Cognitive Examination (ACE) in clinical practice. Int J Geriatr Psychiatry 2005; 20(6):593–4. 335 Lonergan E, Luxenberg J, Colford J. Haloperidol for agitation in dementia. Cochrane Database Syst Rev 2002; (2):CD002852. 336 Ivan CS, Seshadri S, Beiser A, Au R, Kase CS, Kelly-Hayes M, Wolf PA. Dementia after stroke: the Framingham Study. Stroke 2004; 35(6):1264–8. 337 Ballard C, Rowan E, Stephens S, Kalaria R, Kenny RA. Prospective follow-up study between 3 and 15 months after stroke: improvements and decline in cognitive function among dementia-free stroke survivors >75 years of age. Stroke 2003; 34(10):2440–4. 338 Rasquin SM, Lodder J, Ponds RW, Winkens I, Jolles J, Verhey FR. Cognitive functioning after stroke: a one-year follow-up study. Dement Geriatr Cog Dis 2004; 18(2):138–44. 339 Leys D, Henon H, Kowiak-Cordoliani MA, Pasquier F. Poststroke dementia. Lancet Neurol 2005; 4(11):752–9.

629

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 630

630

Chapter 11 What are this patient’s problems?

340 Barba R, Morin MD, Cemillan C, Delgado C, Domingo J, Del ST. Previous and incident dementia as risk factors for mortality in stroke patients. Stroke 2002; 33(8):1993–8. 341 De Konig I, van Kooten F, Koudstaal PJ, Dippel DW. Diagnostic value of the Rotterdam-CAMCOG in poststroke dementia. J Neurol Neurosurg Psychiatry 2005; 76(2):263–5. 342 Raven JC. Guide to Using the Coloured Progressive Matrices. London: HK Lewis & Co., 1965. 343 Malouf R, Birks J. Donepezil for vascular cognitive impairment. Cochrane Database Syst Rev 2004; (1):CD004395. 344 Kimura M, Robinson RG, Kosier JT. Treatment of cognitive impairment after poststroke depression: a double-blind treatment trial. Stroke 2000; 31(7):1482–6. 345 Lincoln NB, Majid MJ, Weyman N. Cognitive rehabilitation for attention deficits following stroke. Cochrane Database Syst Rev 2000; (4):CD002842. 346 Majid MJ, Lincoln NB, Weyman N. Cognitive rehabilitation for memory deficits following stroke. Cochrane Database Syst Rev 2000; (3):CD002293. 347 Engelter ST, Gostynski M, Papa S, Frei M, Born C, JdacicGross V et al. Epidemiology of aphasia attributable to first ischemic stroke: incidence, severity, fluency, etiology, and thrombolysis. Stroke 2006; 37(6):1379–84. 348 Pedersen PM, Vinter K, Olsen TS. Aphasia after stroke: type, severity and prognosis. The Copenhagen aphasia study. Cerebrovasc Dis 2004; 17(1):35–43. 349 Berthier ML. Poststroke aphasia: epidemiology, pathophysiology and treatment. Drugs Aging 2005; 22(2):163–82. 350 Enderby PM, Wood VA, Wade DT, Hewer RL. The Frenchay Aphasia Screening Test: a short, simple test for aphasia appropriate for non-specialists. Int Rehabil Med 1987; 8(4):166–70. 351 Enderby P, Crow E. Frenchay Aphasia Screening Test: validity and comparability. Disabil Rehabil 1996; 18(5):238–40. 352 Greener J, Enderby P, Whurr R. Speech and language therapy for aphasia following stroke. Cochrane Database Syst Rev 2006; (3). 353 Roby RR. A meta-analysis of clinical outcomes in the treatment of aphasia. J Speech Hearing Res 1998; 41:172–87. 354 Bakheit AM. Drug treatment of poststroke aphasia. Exp Rev Neurotherapeutics 2004; 4(2):211–17. 355 Greener J, Enderby P, Whurr R. Pharmacological treatment for aphasia following stroke. Cochrane Database Syst Rev 2006; (3). 356 Sellars C, Hughes T, Langhorne P. Speech and language therapy for dysarthria due to non-progressive brain damage. Cochrane Database Syst Rev 2006; (3). 357 Beekman AT, Penninx BW, Deeg DJ, Ormel J, Smit JH, Braam AW, van Tilburg W. Depression in survivor of stroke: a community-based study of prevalence, risk factors and consequences. Soc Psychiatry Psychiatr Epidemiol 1998; 33(10):463–70.

358 Whyte EM, Mulsant BH, Vanderbilt J, Dodge HH, Ganguli M. Depression after stroke: a prospective epidemiological study. J Am Geriatr Soc 2004; 52(5):774–8. 359 House A. Mood disorders after stroke: a review of the evidence. Int J Geriatr Psychiatry 1987; 2:211–21. 360 Hackett ML, Yapa C, Parag V, Anderson CS. Frequency of depression after stroke: a systematic review of observational studies. Stroke 2005; 36(6):1330–40. 361 Hackett ML, Anderson CS, Auckland Regional Community Stroke (ARCOS) Study Group. Frequency, management, and predictors of abnormal mood after stroke: the Auckland Regional Community Stroke (ARCOS) study, 2002 to 2003. Stroke 2006; 37(8):2123–8. 362 Leppavuori A, Pohjasvaara T, Vataja R, Kaste M, Erkinjuntti T. Generalized anxiety disorders three to four months after ischemic stroke. Cerebrovasc Dis 2003; 16(3):257–64. 363 Dennis M, O’Rourke S, Lewis S, Sharpe M, Warlow C. Emotional outcomes after stroke: factors associated with poor outcome. J Neurol Neurosurg Psychiatry 2000; 68(1):47–52. 364 Dennis M, O’Rourke S, Lewis S, Sharpe M, Warlow C. Emotional outcomes after stroke: factors associated with poor outcome. J Neurol Neurosurg Psychiatry 2000; 68(1):47–52. 365 Hackett ML, Anderson CS. Predictors of depression after stroke: a systematic review of observational studies. Stroke 2005; 36(10):2296–301. 366 Singh A, Black SE, Herrmann N, Leibovitch FS, Ebert PL, Lawrence J, Szalai JP. Functional and neuroanatomic correlations in poststroke depression: the Sunnybrook Stroke Study. Stroke 2000; 31(3):637–44. 367 Yu L, Liu CK, Chen JW, Wang SY, Wu YH, Yu SH. Relationship between post-stroke depression and lesion location: a meta-analysis. Kaohsiung J Med Sci 2004; 20(8):372–80. 368 Bhogal SK, Teasell R, Foley N, Speechley M. Lesion location and poststroke depression: systematic review of the methodological limitations in the literature. Stroke 2004; 35(3):794–802. 369 Singh A, Herrmann N, Black SE. The importance of lesion location in poststroke depression: a critical review. Can J Psychiatry 1998; 43(9):921–7. 370 Carson AJ, MacHale S, Allen K, Lawrie SM, Dennis M, House A, Sharpe M. Depression after stroke and lesion location: a systematic review. Lancet 2000; 356(9224):122–6. 371 Morrison V, Pollard B, Johnston M, MacWalter R. Anxiety and depression 3 years following stroke: demographic, clinical, and psychological predictors. J Psychosom Res 2005; 59(4):209–13. 372 Bruggimann L, Annoni JM, Staub F, von Steinbuchel N, Van der Linden M, Bogousslavsky J. Chronic posttraumatic stress symptoms after nonsevere stroke. Neurology 2006; 66(4):513–16. 373 Ramasubbu R, Patten SB. Effect of depression on stroke morbidity and mortality. Can J Psychiatry 2003; 48(4):250–7.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 631

References

374 Spalletta G, Ripa A, Caltagirone C. Symptom profile of DSM-IV major and minor depressive disorders in first-ever stroke patients. Am J Geriatr Psychiatry 2005; 13(2):108–15. 375 Watkins C, Daniels L, Jack C, Dickinson H, van Den BM. Accuracy of a single question in screening for depression in a cohort of patients after stroke: comparative study. Br Med J 2001; 323(7322):1159. 376 Aben I, Verhey F, Lousberg R, Lodder J, Honig A. Validity of the Beck depression inventory, hospital anxiety and depression scale, SCL-90, and Hamilton depression rating scale as screening instruments for depression in stroke patients. Psychosomatics 2002; 43(5):386–93. 377 Benaim C, Cailly B, Perennou D, Pelissier J. Validation of the aphasic depression rating scale. Stroke 2004; 35(7):1692–6. 378 Leeds L, Meara RJ, Hobson JP. The utility of the Stroke Aphasia Depression Questionnaire (SADQ) in a stroke rehabilitation unit. Clin Rehabil 2004; 18(2):228–31. 379 Forster A, Smith J, Young J, Knapp P, House A, Wright J. Information provision for stroke patients and their caregivers. Cochrane Database Syst Rev 2006; (3). 380 Anderson CS, Hackett ML, House AO. Interventions for preventing depression after stroke. Cochrane Database Syst Rev 2006; (3). 381 Hackett ML, Anderson CS, House AO. Management of depression after stroke: a systematic review of pharmacological therapies. Stroke 2005; 36(5):1098–103. 382 Mottram P, Wilson K, Strobl J. Antidepressants for depressed elderly. Cochrane Database Syst Rev 2006; (1):CD003491. 383 Cipriani A, Brambilla P, Furukawa T, Geddes J, Gregis M, Hotopf M et al. Fluoxetine versus other types of pharmacotherapy for depression. Cochrane Database Syst Rev 2005; (4):CD004185. 384 Hackett ML, Anderson CS, House AO. Interventions for treating depression after stroke. Cochrane Database Syst Rev 2006; (3). 385 Cole MG, Elie LM, McCusker J, Bellavance F, Mansour A. Feasibility and effectiveness of treatments for post-stroke depression in elderly inpatients: systematic review. J Geriatr Psychiatry Neurol 2001; 14(1):37–41. 386 Murray GB, Shea V, Conn DK. Electroconvulsive therapy for poststroke depression. J Clin Psychiatry 1986; 47(5):258–60. 387 House AO, Hackett ML, Anderson CS, Horrocks JA. Pharmaceutical interventions for emotionalism after stroke. Cochrane Database Syst Rev 2006; (3). 388 Calvert T, Knapp P, House A. Psychological associations with emotionalism after stroke. J Neurol Neurosurg Psychiatry 1998; 65(6):928–9. 389 Lincoln NB, Gamlen R, Thomason H. Behavioural mapping of patients on a stroke unit. Int Disabil Stud 1989; 11(4):149–54. 390 Bernhardt J, Dewey H, Thrift A, Donnan G. Inactive and alone: physical activity within the first 14 days of acute stroke unit care. Stroke 2004; 35(4):1005–9.

..

..

391 Lincoln NB, Willis D, Philips SA, Juby LC, Berman P. Comparison of rehabilitation practice on hospital wards for stroke patients. Stroke 1996; 27(1):18–23. 392 Choi-Kwon S, Han SW, Kwon SU, Kim JS. Poststroke fatigue: characteristics and related factors. Cerebrovasc Dis 2005; 19(2):84–90. 393 De Groot MH, Phillips SJ, Eskes GA. Fatigue associated with stroke and other neurologic conditions: Implications for stroke rehabilitation. Arch Phys Med Rehabil 2003; 84(11):1714–20. 394 Glader EL, Stegmayr B, Asplund K. Poststroke fatigue: a 2-year follow-up study of stroke patients in Sweden. Stroke 2002; 33(5):1327–33. 395 Outpatient Service Trialists. Therapy-based rehabilitation services for stroke patients at home. Cochrane Database Syst Rev 2003; 1. 396 Logan PA, Gladman JR, Drummond AE, Radford KA. A study of interventions and related outcomes in a randomized controlled trial of occupational therapy and leisure therapy for community stroke patients. Clin Rehabil 2003; 17(3):249–55. 397 Collen FM, Wade DT. Residual mobility problems after stroke. Int Disabil Stud 1991; 13(1):12–5. 398 Collen FM, Wade DT, Robb GF, Bradshaw CM. The Rivermead Mobility Index: a further development of the Rivermead Motor Assessment. Int Disabil Stud 1991; 13(2):50–4. 399 Wade DT, Wood VA, Heller A, Maggs J, Langton HR. Walking after stroke. Measurement and recovery over the first 3 months. Scand J Rehabil Med 1987; 19(1):25–30. 400 Green J, Young J, Forster A, Collen F, Wade D. Combined analysis of two randomized trials of community physiotherapy for patients more than one year post stroke. Clin Rehabil 2004; 18(3):249–52. 401 Saunders DH, Greig CA, Young A, Mead GE. Physical fitness training for stroke patients. Cochrane Database Syst Rev 2006; (3). 402 Gladman J, Forster A, Young J. Hospital- and home-based rehabilitation after discharge from hospital for stroke patients: analysis of two trials. Age Ageing 1995; 24(1):49–53. 403 Legh-Smith J, Wade DT, Hewer RL. Driving after a stroke. J R Soc Med 1986; 79(4):200–3. 404 Fisk GD, Owsley C, Pulley LV. Driving after stroke: driving exposure, advice, and evaluations. Arch Phys Med Rehabil 1997; 78(12):1338–45. 405 Gommans J, Sye D, MacDonald A. Guideline recommendations for the management of patients admitted with acute stroke: implications of a local audit. N Z Med J 2005; 118(1214):U1435. 406 Samuelsson SM. Physicians’ control of driving after stroke attacks. Tidsskr Den Norske Laegef 2005; 125(19):2610–12. 407 Haselkorn JK, Mueller BA, Rivara FA. Characteristics of drivers and driving record after traumatic and nontraumatic brain injury. Arch Phys Med Rehabil 1998; 79(7):738–42.

631

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 632

632

Chapter 11 What are this patient’s problems?

408 Akinwuntan AE, Feys H, De Weerdt W, Baten G, Arno P, Kiekens C. Prediction of driving after stroke: A prospective study. Neurorehabilitation & Neural Repair 2006; 20(3):417–23. 409 Mazer BL, Korner-Bitensky NA, Sofer S. Predicting ability to drive after stroke. Arch Phys Med Rehabil 1998; 79(7):743–50. 410 Radford KA, Lincoln NB. Concurrent validity of the stroke drivers screening assessment. Arch Phys Med Rehabil 2004; 85(2):324–8. 411 Fox GK, Bowden SC, Smith DS. On-road assessment of driving competence after brain impairment: review of current practice and recommendations for a standardized examination. Arch Phys Med Rehabil 1998; 79(10):1288–96. 412 Akinwuntan AE, DeWeerdt W, Feys H, Baten G, Arno P, Kiekens C. Reliability of a Road Test after Stroke. Arch Phys Med Rehabil 2003; 84(12):1792–6. 413 Akinwuntan AE, De Weerdt W, Feys H, Baten G, Arno P, Kiekens C. The validity of a road test after stroke. Arch Phys Med Rehabil 2005; 86(3):421–6. 414 Akinwuntan AE, De Weerdt W, Feys H, Pauwels J, Baten G, Arno P, Kiekens C. Effect of simulator training on driving after stroke: A randomized controlled trial. Neurology 2005; 65(6):843–50. 415 Chiu CWY, Man DWK. The effect of training older adults with stroke to use home-based assistive devices. OTJR: Occupation, Participation, & Health 2006; 24(3):113–20. 416 Walker CM, Sunderland A, Sharma J, Walker MF. The impact of cognitive impairment on upper body dressing difficulties after stroke: a video analysis of patterns of recovery. J Neurol Neurosurg Psychiatry 2004; 75(1):43–8. 417 Huck J, Bonhotal BH. Fastener systems on apparel for hemiplegic stroke victims. Appl Ergon 1997; 28(4):277–82. 418 Perry L, McLaren S. An exploration of nutrition and eating disabilities in relation to quality of life at 6 months poststroke. Health Soc Care Community 2004; 12(4):288–97. 419 Wozniak MA, Kittner SJ. Return to work after ischemic stroke: a methodological review. Neuroepidemiology 2002; 21(4):159–66. 420 Vestling M, Tufvesson B, Iwarsson S. Indicators for return to work after stroke and the importance of work for subjective well-being and life satisfaction. J Rehabil Med 2003; 35(3):127–31. 421 Wellwood I, Dennis MS, Warlow CP. Perceptions and knowledge of stroke among surviving patients with stroke and their carers. Age Ageing 1994; 23(4):293–8. 422 Drummond A. Leisure activity after stroke. Int Disabil Stud 1990; 12(4):157–60. 423 Green J, Forster A, Young J. A test-retest reliability study of the Barthel Index, the Rivermead Mobility Index, the Nottingham Extended Activities of Daily Living Scale and the Frenchay Activities Index in stroke patients. Disabil Rehabil 2001; 23(15):670–6. 424 Drummond AE, Parker CJ, Gladman JR, Logan PA. Development and validation of the Nottingham Leisure Questionnaire (NLQ). Clin Rehabil 2001; 15(6):647–56. 425 Parker CJ, Gladman JR, Drummond AE, Dewey ME, Lincoln NB, Barer D et al. A multicentre randomized

426

427

428 429

430

431

432

433

434

435

436

437

438 439 440

441

442

443

controlled trial of leisure therapy and conventional occupational therapy after stroke. TOTAL Study Group. Trial of Occupational Therapy and Leisure. Clin Rehabil 2001; 15(1):42–52. Kakkos SK, Geroulakos G. Economy class stroke syndrome: case report and review of the literature. Eur J Vasc Endovasc Surg 2004; 27(3):239–43. Cheung RT. Sexual functioning in Chinese stroke patients with mild or no disability. Cerebrovasc Dis 2002; 14(2):122–8. Holbrook M. Stroke: social and emotional outcome. J R Coll Physicians Lond 1982; 16(2):100–4. Van Exel NJ, Koopmanschap MA, van den BB, Brouwer WB, van den Bos GA. Burden of informal caregiving for stroke patients. Identification of caregivers at risk of adverse health effects. Cerebrovasc Dis 2005; 19(1):11–17. Visser-Meily JM, Post MW, Riphagen II, Lindeman E. Measures used to assess burden among caregivers of stroke patients: a review. Clin Rehabil 2004; 18(6):601–23. Kalra L, Evans A, Perez I, Melbourn A, Patel A, Knapp M, Donaldson N. Training carers of stroke patients: randomised controlled trial. Br Med J 2004; 328(7448):1099. Visser-Meily A, van Heugten C, Post M, Schepers V, Lindeman E. Intervention studies for caregivers of stroke survivors: a critical review. Patient Educ Couns 2005; 56(3):257–67. Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing 1972; 1(4):233–8. Belfield PW. Urinary catheters. In: GP Mulley, ed. Everyday Aids and Appliances. London: BMJ Publishers, 1989, pp. 55–9. Smith N. Aids for urinary incontinence. In: GP Mulley, ed. Everyday Aids and Appliances. London: BMJ Publishers, 1989, pp. 50–4. Kidd D, Lawson J, Nesbitt R, MacMahon J. Aspiration in acute stroke: a clinical study with videofluoroscopy. Q J Med 1993; 86(12):825–9. Scottish Intercollegiate Guidelines Network. Management of Patients with Stroke: Identification and Management of Dysphagia. Edinburgh: Scottish Intercollegiate Guidelines Network, 2004. Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner 1964; 192:540–2. Wade DT. Measurement in neurological rehabilitation. Curr Opin Neurol Neurosurg 1992; 5(5):682–6. Martin J, Meltzer H, Elliot D. The Prevalence of Disability among Adults. Office of Population Censuses and Surveys. London: HMSO, 1988. Wanklyn P, Forster A, Young J. Hemiplegic shoulder pain (HSP): natural history and investigation of associated features. Disabil Rehabil 1996; 18(10):497–501. Braus DF, Krauss JK, Strobel J. The shoulder-hand syndrome after stroke: a prospective clinical trial. Ann Neurol 1994; 36(5):728–33. Wilson B, Cockburn J, Halligan P. Behavioural Inattention Test. Titchfield, Hants: Thames Valley Test Company, 1987.

..

..

9781405127660_4_011.qxd 10/13/07 10:55 AM Page 633

References

444 Folstein MF, Folstein SE, McHugh PR. ‘Mini-mental state’: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12(3):189–98. 445 O’Rourke S, MacHale S, Signorini D, Dennis M. Detecting psychiatric morbidity after stroke: comparison of the GHQ and the HAD Scale. Stroke 1998; 29(5):980–5. 446 Johnson G, Burvill PW, Anderson CS, Jamrozik K, StewartWynne EG, Chakera TM. Screening instruments for depression and anxiety following stroke: experience in the Perth community stroke study. Acta Psychiatr Scand 1995; 91(4):252–7. 447 House A, Dennis M, Hawton K, Warlow C. Methods of identifying mood disorders in stroke patients: experience in the Oxfordshire Community Stroke Project. Age Ageing 1989; 18(6):371–9. 448 Shinar D, Gross CR, Price TR, Banko M, Bolduc PL, Robinson RG. Screening for depression in stroke patients: the reliability and validity of the Center for Epidemiologic Studies Depression Scale. Stroke 1986; 17(2):241–5. 449 Agrell B, Dehlin O. Comparison of six depression rating scales in geriatric stroke patients. Stroke 1989; 20(9):1190–4. 450 Williams LS, Brizendine EJ, Plue L, Bakas T, Tu W, Hendrie H, Kroenke K. Performance of the PHQ-9 as a screening tool for depression after stroke. Stroke 2005; 36(3):635–8. 451 Mulley G. Walking frames. In: G Mulley, ed. More Everyday Aids and Appliances. London: BMJ Publishing, 1991, pp. 174–81. 452 Tyson SF, Ashburn A. The influence of walking aids on hemiplegic gait. Physiother Theory Pract 1994; 10:77–86.

..

..

453 Lu CL, Yu B, Basford JR, Johnson ME, An KN. Influences of cane length on the stability of stroke patients. J Rehabil Res Dev 1997; 34(1):91–100. 454 Tyson SF. The support taken through walking aids during hemiplegic gait. Clin Rehabil 1998; 12(5):395–401. 455 Tyson S, Thornton H, Downes A. The effect of a hinged ankle-foot orthosis on hemiplegic gait: four single case studies. Physiother Theory Pract 1998; 14:75–85. 456 Beckerman H, Becher J, Lankhorst GJ, Verbeek AL. Walking ability of stroke patients: efficacy of tibial nerve blocking and a polypropylene ankle-foot orthosis. Arch Phys Med Rehabil 1996; 77(11):1144–51. 457 Roberts K, Porter K. How do you size a nasopharyngeal airway. Resuscitation 2003; 56(1):19–23. 458 Roberts K, Whalley H, Bleetman A. The nasopharyngeal airway: dispelling myths and establishing the facts. Emerg Med J 2005; 22(6):394–6. 459 St Clair M. Survey of the uses of the Pegasus Airwave System in the United Kingdom. J Tiss Viabil 1992; 2:9–16. 460 Levine JA, Morris JC. The use of a football helmet to secure a nasogastric tube. Nutrition 1995; 11(3):285. 461 Marson AG, Al-Kharusi AM, Alwaidh M, Appleton R, Baker GA, Chadwick DW, et al. The SANAD study of effectiveness of carbamazepine, gabapentin, lamotrigine, oxcarbazepine, or topiramate for treatment of partial epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369:1000–15. 462 Marson AG, Al-Kharusi AM, Alwaidh M, Appleton R, Baker GA, Chadwick DW, et al. The SANAD study of effectiveness of valproate, lamotrigine or topiramate for generalised and unclassifiable epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369:1016–26.

633

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 635

12

Specific treatments for acute ischaemic stroke 12.1 Pathophysiology of acute ischaemic stroke 12.2 General treatment considerations 12.3 Routine use: aspirin

635

654

660

12.4 Selective use: anticoagulants

665

12.5 Selective use: reperfusion with thrombolytic drugs and other methods 12.6 Unproven value: other reperfusion strategies

675

687

12.7 Unproven value: treatment of cerebral oedema and raised intracranial pressure 12.8 Unproven value: other interventions

Focal cerebral ischaemia, caused by acute occlusion of a cerebral blood vessel or sometimes just by low blood flow, initiates a series of events which can lead to irreversible neuronal damage and cell death (i.e. infarction) in the part of the brain supplied or drained by that vessel. Several pathophysiological cascades run in sequence (and in parallel).1,2 In the vascular system, rapid changes in platelet and coagulation factors, the vessel wall (particularly the endothelium) and in the thrombus itself interact to produce a very dynamic state, not only at the site of vessel occlusion, but also more remotely, in both the macro- and microcirculation. In brain tissue, changes occur in neurones, glial cells and other structural components in differing degrees and at different times after the onset of ischaemia, which means that cerebral infarction is a dynamic and highly unstable process, not a discrete ‘one-off’ event. In other words, infarction is not an ‘all-or-nothing’ episode which is instantaneous in onset, maximal in severity at the moment of onset and irreversibly complete within 6 h.3 This chapter begins with a review of pathophysiology, particularly aspects which relate to the specific treatments that are described later in the chapter. As far as possible, we base the decision whether or not to use a particular treatment, not merely on its putative physiological effects, but on the evidence from randomized trials in patients which demonstrate the balance of

689

691

clinical risk and benefit associated with its use. In view of this, the section on each of the specific treatments will assess the strength of evidence available and, where possible, base any recommendations on the results of a systematic review (and meta-analysis) of all the relevant randomized trials of that particular treatment.

12.1 Pathophysiology of acute ischaemic stroke

The pathophysiology of acute ischaemic stroke encompasses two sequential processes: the vascular, haematological or cardiac events that cause the initial reduction in local cerebral blood flow (Chapters 6 and 7) and then the alterations of cellular chemistry that are caused by ischaemia and which lead to necrosis of neurones, glia and other brain cells. This section will discuss cerebral metabolism, regulation of cerebral blood flow, molecular consequences of cerebral ischaemia and how understanding these processes leads to the development of various treatments for acute ischaemic stroke.

12.1.1 Cerebral metabolism Energy demand and cerebral blood flow

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

The human brain has a high metabolic demand for energy and, unlike other organs, uses glucose (about 75–100 mg/min, or 125 g/day) as its main substrate for 635

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 636

636

Chapter 12 Specific treatments for acute ischaemic stroke

Glucose

ATP ADP

Glucose-6-phosphate Fructose-6-phosphate

ATP ADP

Glycolysis Glucose + 2Pi + 2ADP 2lactate + 2ATP + 2H2O for glycolysis

Fructose-1,6-phosphate Glyceraldehyde-3-phosphate (2)

2NAD+ +2Pi

3-phosphoglyceroyl phosphate (2)

2ADP 2ATP

3-phosphoglycerate (2) 2-phosphoglycerate (2) Phosphoenolpyruvate (2)

2ADP 2ATP

Pyruvate (2) NAD+ NADH

2NADH 2NAD+

CoA CO2

Acetyl-CoA

Lactate (2)

Oxaloacetate The tricarboxylic acid cycle 2Pyruvate + 5O2 + 30ADP + 30Pi 6CO2 + 30ATP + 34H2O

Citrate Malate Isocitrate

CO2

Fumarate -ketoglutarate Succinate

CO2

Succinyl-CoA 2H

2H

2H

2H

NAD+ Flavoprotein

ADP + Pi

Coenzyme Q

ATP

Cytochrome b

Electron transport and oxidative phosphorylation

ADP + Pi

Cytochrome c

ATP

Cytochrome a3

ADP + Pi ATP

2H+ + 1/2O2

Overall: glucose + 6O2 + 36Pi + 36ADP

H2O

6CO2 + 36ATP + 42H2O

energy metabolism. Glucose is metabolized in the brain entirely via the glycolytic sequence and the tricarboxylic acid cycle (Fig. 12.1). Each molecule of glucose is broken down in a series of enzymatic steps (glycolysis) into two molecules (2 m) of pyruvate. During these reactions, the oxidized form of

Fig. 12.1 Biochemical pathway of the aerobic metabolism of glucose.

nicotinamide – adenine dinucleotide (NAD+) – is reduced (to NADH) and 2 m each of adenosine diphosphate (ADP) and intracellular phosphorus are converted to 2 m of adenosine triphosphate (ATP). In the presence of oxygen, pyruvate is metabolized, first by pyruvate dehydrogenase and then by a series of mitochondrial

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 637

12.1 Pathophysiology of acute ischaemic stroke

reactions, to carbon dioxide (CO2) and water (H2O) with the formation of 36 m of ATP. This is the maximum ATP yield. In the absence of oxygen, this sequence of events is blocked or retarded at the stage of pyruvate oxidation, leading to the reduction of pyruvate to lactate by NADH and lactic dehydrogenase. Anaerobic glycolysis therefore still leads to the formation of ATP, as well as lactate, but the energy yield is relatively small (2 m rather than 36 m of ATP from 1 m of glucose). In addition, lactic acid accumulates within and outside cells (hence, the cell is acidified) and mitochondria lose their ability to sequester calcium, so any calcium entering or released within the cell will raise the intracellular calcium level.4–6 Although classical neuroenergetics states that glucose is the exclusive energy substrate of brain cells, and its full oxidation provides all the necessary energy to support brain function, recent data have revealed a more intricate picture in which astrocytes play a central role in metabolic coupling and supplying lactate as an additional energy substrate in register with glutamatergic activity.7 The basic mechanism involves glutamate-stimulated aerobic glycolysis; the sodium-coupled reuptake of glutamate by astrocytes and the ensuing activation of the Na-K-ATPase triggers glucose uptake and processing via glycolysis, resulting in the release of lactate from astrocytes. Lactate can then contribute to the activitydependent fuelling of the neuronal energy demands associated with synaptic transmission. An operational model, the ‘astrocyte-neurone lactate shuttle’, is supported experimentally by a large body of evidence, which provides a molecular and cellular basis for interpreting data obtained from functional brain imaging studies.8 When neurones are in the presence of both glucose and lactate, they preferentially use lactate as their main oxidative substrate.9 The brain uses glucose as its main source of energy. During aerobic metabolism each molecule of glucose produces 36 molecules of adenosine triphosphate (ATP), but during anaerobic metabolism only two molecules of ATP are produced along with lactic acid. ATP is the universal currency for energy. Neurones in the brain require a constant supply of ATP to maintain their integrity and to keep the major intracellular cation, potassium ions (K+), within the cell, and the major extracellular cations, sodium (Na+) and calcium ions (Ca++), outside the cell. Because the brain is unable to store energy, it requires a constant supply of oxygenated blood containing an adequate glucose concentration to maintain its function and structural integrity.

..

..

The resting brain consumes energy at the same rate as a 20-watt light bulb. Global cerebral blood flow (CBF), reflecting both grey and white matter compartments, per unit of brain in a healthy young adult, is about 50–55 mL/100 g of brain per minute, with significantly higher values in those below 20 years of age and lower values in those over 60 years.10 For a brain of average weight (1300–1400 g in a 60–65 kg adult), which is only 2% of total adult body weight, the total CBF at rest is disproportionately large at about 800 mL/min, which is 15–20% of the total cardiac output. At this level of blood flow, whole brain oxygen consumption, usually measured as the cerebral metabolic rate of oxygen (CMRo2), is about 3.3–3.5 mL/ 100 g of brain per minute, or 45 mL of oxygen per minute, which is 20% of the total oxygen consumption of the body at rest.

12.1.2 Cerebral blood flow regulation The fraction of oxygen extracted from the blood, the oxygen extraction fraction (OEF), is fairly constant throughout the brain because CBF, cerebral blood volume (CBV) and CMRo2 as well as the cerebral metabolic rate of glucose (CMRglu) are all coupled.10 In the normal resting brain, measurements of CBF are therefore a reliable reflection of cerebral metabolism (CMRo2). If CBF falls, however (down to 20–25 mL/100 g brain per minute), the OEF increases to maintain the CMRo2 (see below). Over the past 60 years, the methods for measuring CBF have become more accurate and reliable, and have had a major impact on our understanding of the regulation of CBF and the pathogenesis of cerebral ischaemia.3,11–17 Positron emission tomography (PET) now enables CBF, CMRo2, OEF and CMRglu all to be measured in various regions of interest in the brain, both in normal people and after stroke.11,14,18 Cerebral perfusion pressure (CPP) Under normal conditions, blood flow through the brain is determined by the CPP and by the cerebrovascular resistance (CVR) imposed by blood viscosity and the size of the intracranial vessels (i.e. flow = pressure/resistance). The CPP represents the difference between arterial pressure forcing blood into the cerebral circulation and the venous pressure. The mean CPP is the mean systemic arterial pressure at the base of the brain in the recumbent position, which approximates to the diastolic blood pressure (about 80 mmHg), plus one-third of the pulse pressure (one-third of about 40 mmHg) minus the

637

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 638

638

Chapter 12 Specific treatments for acute ischaemic stroke

intracranial venous pressure (about 10 mmHg), i.e. 80–85 mmHg. Cerebrovascular resistance Under normal conditions, when resting CPP is constant, any change in CBF must be caused by a change in CVR, usually as a result of alteration in the diameter of small intracranial arteries or arterioles. Under these circumstances, there is a direct correlation between CBF and the intravascular CBV. CBF and CBV will both increase as vessels dilate and both decrease as vessels constrict. The CBV : CBF ratio remains relatively constant over a wide range of CBF at normal CPP. When an artery narrows causing CVR to increase, or when CBF increases, the blood flow velocity in that segment of artery increases. Although it may seem paradoxical that a reduction in luminal diameter causes an increase in blood flow velocity, think of using a hose: putting one’s finger over the nozzle generates a highpressure jet of water. The narrower the lumen at the nozzle, the greater the pressure (and velocity of flow) in the stream of water until the lumen is nearly occluded, at which point velocity becomes substantially reduced and the water dribbles out of the hose. This is one of the principles governing the interpretation of blood flow velocities in the major basal arteries by transcranial Doppler ultrasound (section 6.8.9). Mean blood flow velocities within the intracranial arteries vary from 40 to 70 cm/s. As blood flow velocity is proportional to the second power of the vessel radius, it cannot be equated linearly with volume blood flow (mL/s), which is proportional to the fourth power of the vessel radius. If vessel calibre were constant, some assumptions could be made about volume flow from velocity measures, but the calibre of large cerebral arteries varies with changes in blood pressure, partial pressure of arterial carbon dioxide (PaCO2), intracranial pressure and age.

fairly stable under steady state conditions. For example, during voluntary hand movements, the metabolic activity of the contralateral motor cortex increases over a few seconds and is accompanied by rapid vasodilatation of the local cerebral resistance vessels, leading to an increase in CBF and CBV, rather than any increase in the OEF or glucose extraction fraction. Conversely, low regional metabolic activity (as may occur in a cerebral infarct) is associated with reduced metabolic demand and so low CBF. Therefore, low flow does not necessarily mean vessel occlusion but, as in this case, it can mean non-functioning brain. Although this coupling of flow with metabolism and function has been suspected for over a century, the mechanism is unknown; it may be that the metabolically active areas of brain produce vasodilatory metabolites, or the resistance vessels may be under neural regulation, or a combination of both. This is the principle underpinning functional magnetic resonance imaging (fMRI).19 In normal brain, blood flow is closely coupled with metabolic demand. However, if the brain is damaged, blood flow and metabolism become uncoupled and so normal flow no longer necessarily implies normal metabolism and function. Arterial carbon dioxide tension Arterial carbon dioxide tension (PaCO2) has a potent effect on CBF;20 a 1 mmHg rise in PaCO2 within the range of 20–60 mmHg (2.7–8.0 kPa) in normal individuals causes an immediate 3–5% increase in CBF due to dilatation of cerebral resistance vessels. In chronic respiratory failure however, causing CO2 retention, CBF is normal. Changes in arterial oxygen tension (PaO2) have a modest inverse effect on CBF, unless the PaO2 falls below about 50 mmHg (6.7 kPa), when the resultant decline in the oxygen saturation of the blood leads to a fall in CVR and an increase in CBF. Increasing the PaO2 above the normal level has little effect on CBF.

Metabolic rate of cerebral tissue In the resting brain with normal CPP, CBF is closely matched to the metabolic demands of the tissue. Therefore, grey matter (which has a high metabolic rate) has higher regional CBF than white matter, which has a relatively low metabolic rate. The ratio between CBF and metabolism is fairly uniform in all areas of the brain and, consequently, the OEF and functional extraction of glucose from the blood are much the same in different areas. Normally, regional OEF is about one-third and the regional glucose extraction fraction is about 10%. Similarly, local flow varies directly with local brain function by 10–20%, even though global CBF tends to be

Whole-blood viscosity Normally, CBF is inversely related to whole-blood viscosity. As the main determinant of whole-blood viscosity (at normal shear rates) is the haematocrit, it follows that CBF and haematocrit are inversely related. However, this relationship is not because the high haematocrit raises viscosity and thereby slows flow (at least not in normal vessels); rather, the higher oxygen content of high haematocrit blood allows CBF to be lower and yet to maintain normal oxygen delivery to the tissues in accordance with metabolic demands. A practical example is encountered in patients with leukaemia or paraproteinaemia who

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 639

12.1 Pathophysiology of acute ischaemic stroke

have very high blood viscosity but normal CBF (or even high CBF if anaemia coexists), because CBF depends more on the oxygen content of the blood (which is normal or low) than the viscosity (which is high). However, at very low shear rates, which might be found in ischaemic brain, for example, because of local vasodilatation, whole-blood viscosity depends more on plasma fibrinogen than haematocrit. In addition, other local factors such as red cell aggregation, platelet aggregation and perhaps increasing red cell fragility as a result of anoxia, all of which increase blood viscosity, may come into play to reduce flow. Autoregulation Under normal conditions (i.e. mean systemic arterial blood pressure within 60–160 mmHg), CBF is maintained at a relatively constant level despite changes in the cerebral perfusion pressure (Fig. 12.2). This capacity to maintain a constant CBF is due to the phenomenon of autoregulation.21–23 Autoregulation is the ability of cerebral blood flow to remain constant in the face of changes in mean arterial pressure and cerebral perfusion pressure.

Cerebral blood flow (mL/100 g brain per minute)

100

75

50

25 0 20

60

100

140

180

Mean systemic arterial pressure (mmHg) Normal brain tissue Normal brain tissue, chronic hypertension Mildly ischaemic brain tissue Moderately ischaemic brain tissue Severely ischaemic brain tissue Fig. 12.2 Autoregulation: relationship between mean systemic arterial blood pressure and cerebral blood flow (CBF) in normal brain tissue and in brain ischaemia. Under normal conditions, CBF is maintained at a relatively constant level, independent of the systemic arterial blood pressure, as long as the mean pressure remains between about 60 and 160 mmHg. This capacity to maintain a constant CBF is due to the phenomenon of autoregulation. In chronic hypertension the curve is shifted to the right.

..

..

Autoregulation is achieved primarily by varying precapillary resistance; compensatory vasodilatation of pial and intracerebral arterioles occurs when the blood pressure falls, and compensatory vasoconstriction when the blood pressure rises. Whether myogenic, metabolic or neurogenic processes are responsible for this response is unknown. If mean arterial pressure falls below about 40–50 mmHg, compensatory vasodilatation and therefore cerebral perfusion reserve is exhausted, and CBF parallels the blood pressure (section 6.7.5). Because oxygen delivery to the brain normally far exceeds demand, metabolic activity can still be maintained at a mean blood pressure of around 40–50 mmHg by increasing the oxygen extraction from the blood (Fig. 12.3). This state of increased OEF has been termed ‘misery perfusion’.14 However, when the OEF is maximal, a state of ‘ischaemia’ exists; flow is inadequate ( 0.70 CMRo2 OEF Tracer binding Relative binding > 3.4** and CBV < 14 mL/ 100 g brain/min Tracer uptake Uptake ratio > 1.3*

SPECT [99mTc]-labelled CBF HMPAO or ECD

Relative CBF < 65%*

Not validated, poor resolution

Quantitative, validated

Complex, time-consuming, not widely available, expensive

Based on physiological neuronal integrity

As above and only suitable for cortex, basis not validated

Produces a direct positive As above and validation incomplete, image of viable long imaging time tissue hypoxia Cheap and relatively available

Provides perfusion only so thresholds uncertain, limited spatial resolution, slow brain kinetics

*relative to mean contralateral hemisphere; **relative to contralateral healthy white matter; ADC, apparent diffusion coefficient; CBF, cerebral blood flow; CBV, cerebral blood volume; DWI, diffusion-weighted imaging; ECD, ethyl-cysteinate dimer; FMISO, fluoromisonidazole; FMZ, flumazenil; HMPAO, hexamethylpropyleneamine oxime; MTT, mean transit time; NAA, N-acetylaspartate; OEF, oxygen extraction fraction; PWI, perfusion-weighted imaging; SPECT, single photon emission computed tomography; TTP, time to peak.

equipment and the complex logistics of a multidisciplinary team. PET is able to differentiate between normal, penumbral, and infarcted tissue in the acute stage of ischaemic stroke. However, because of its logistic and practical limitations, its clinical role in acute stroke is limited.

MRI DWI-PWI A more widely applicable method is the magnetic resonance imaging (MRI) technique of combined diffusionweighted imaging (DWI) and perfusion-weighted imaging (PWI) as a measure of tissue integrity and cerebral perfusion respectively (section 5.5.2).12,13,17,19,61–63

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 647

12.1 Pathophysiology of acute ischaemic stroke

The apparent diffusion coefficient (ADC) on DWI is an ‘apparent’ or ‘relative’ measure of the diffusion of free water in one area compared with other areas during the DWI sequence.61 It is decreased in areas of ischaemic brain which have a critically low CBF.64,65 A decrease in the ADC is seen as an area of hyperdensity on diffusionweighted images (due to a restriction in the diffusional movement of water), as soon as 11 min after stroke.66 Because DWI abnormalities typically evolve into infarction without reperfusion or neuroprotection.62 It has been suggested that the DWI abnormality corresponds to the ischaemic core, but this is not always the case.12,63,67 PWI, on the other hand, provides information on the haemodynamic status of the tissue by using paramagnetic contrast agents, for example, gadolinium-based chelates. On the basis of magnetic resonance perfusion imaging data, maps of relative CBF can be calculated, and these demonstrate impaired perfusion of both the ischaemic core and the surrounding brain regions, thereby complementing the information derived from DWI.68 During the first few hours of stroke evolution, PWI typically demonstrates regions with abnormal perfusion that are larger than the DWI abnormalities. It has been postulated that this mismatch reflects the ischaemic penumbra (i.e. functionally impaired ‘tissue at risk’ surrounding the irreversibly damaged ischaemic

Table 12.3 Validity of the DWI-PWI mismatch hypothesis to identify the ischaemic penumbra.19

core).17,19,69 Typically, a PWI > DWI lesion is associated with subsequent infarct enlargement but, because PWI is very sensitive in detecting perfusion defects, the PWI/ DWI mismatch region may comprise not only tissue at risk but also hypoperfused tissue with CBF values above the critical viability thresholds. Further, the reliability of assessing the degree of diffusion-perfusion mismatch is not established.70 The diffusion-perfusion mismatch concept of the ischaemic penumbra (hypoperfusion volume > DWI lesion) has now been modified with the recognition that a portion of hypoperfused brain reflects benign oligaemia, and that the so-called DWI core may in fact be partially salvageable. The validity and reliability of DWI and PWI evidence of mismatch as a marker of the ischaemic penumbra, as compared with PET, has been supported in recent small studies.63 However, its validity in predicting lesion expansion and response to therapy is uncertain and is being evaluated in ongoing studies.12,17,19,71–73 Table 12.3 summarizes the evidence base for the DWI-PWI mismatch hypothesis.19 T2-based blood oxygen level-dependent (BOLD) MR imaging can visualize deoxyhaemoglobin and may serve as a valid non-invasive indicator of the oxygen extraction fraction and thus the metabolic state of threatened brain tissue.39

Operational criteria (see Table 12.1)

DWI-PWI mismatch

Hypoperfusion < 20 mL/ 100 g brain per min

Absolute CBF values unreliable with PWI; but TTP > 4 s delay shown to correspond to PET-derived CBF < 20 mL/100 g per min Volume of DWI lesion and of mismatch significantly correlate with acute-stage stroke severity In humans, acutely the ADC starts to decline for CBF consistent with cellular dysfunction but not death, values corresponding to the penumbra threshold. ADC declines represent cytotoxic oedema and consistently correspond to reduced CMRO2 in humans. In the rat, ADC declines shown to correspond with impaired aerobic glycoslysis and ATP production Without early reperfusion, tissue-at-risk progresses to full infarction Salvage of tissue-at-risk transfers to better outcome and correlates with clinical recovery

Abnormal neuronal function documented by a correlation with acute clinical deficit Physiological and/or biochemical characteristics

Uncertain fate Salvage of this tissue is correlated with better clinical recovery

ADC, apparent diffusion coefficient; CBF, cerebral blood flow; CMRo2, cerebral metabolic rate of oxygen; NIHSS, National Institute of Health Stroke Scale; TTP, time to peak.

..

..

647

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 648

648

Chapter 12 Specific treatments for acute ischaemic stroke

Meanwhile, other incremental refinements in the definition of the ischaemic penumbra are likely, including measurement of biochemical thresholds with spectroscopy, and imaging of a ‘molecular penumbra’ by means of measuring gene markers of protein expression.1 Although early studies suggested that various MRI techniques such as perfusion- and diffusion-weighted imaging could differentiate between penumbra and infarct, more recent studies have shown that this may be an oversimplification. CT perfusion Measures of CBF and CBV obtained from CT perfusion can be sensitive and specific for infarction, and may have utility for differentiating between infarct and penumbra.34 Penumbral regions are characterized by a mismatch between CBF (which is reduced) and CBV (which is maintained), whereas infarcted areas show a matched decrease in both parameters. Contrast-enhanced CT may be able to delineate infarcted and penumbral tissue in acute stroke. Reperfusion and brain damage Reperfusion within the revival times of ischaemic tissues may salvage cells and aid recovery (by restoring oxygen and nutrient delivery to ischaemic brain tissue). However, it may also be detrimental, causing so-called ‘reperfusion injury’ due to the resupply of glucose (which may increase lactic acidosis), oxygen (which may trigger production of injurious free radicals), water and osmotic equivalents (which may exacerbate vasogenic oedema), and blood-borne cells (such as neutrophils) which may exacerbate ischaemic damage.74,75 Rescue of the penumbra, either by restoration of blood supply or by interruption of the adverse metabolic or neurochemical cascade, is the basis of acute stroke therapy. As discussed below, leucocytes migrate into the injured region within hours of reperfusion and may cause tissue injury by occluding the microvasculature, generating oxygen-free radicals, releasing cytotoxic enzymes, altering vasomotor reactivity and increasing cytokine and chemo-attractant release.76 Although, as ever, for every potential benefit of treatment there are potential harms, in humans the empirical evidence suggests that the benefits of reperfusion are greater than the hazards (section 12.5).

12.1.4 Phases and mediators of cell death With sudden, severe and prolonged reductions in cerebral blood flow (CBF) below about 10 mL/100 g brain per minute, ischaemic necrosis (infarction/cell death) progresses within minutes in the core of the affected region because of low ATP levels and energy stores, ionic disruption and metabolic failure.77 However, not all brain cells die immediately after stroke. In the ischaemic penumbra that surrounds the anoxic core, cells can be rescued by rapid reperfusion. Ischaemic brain cells die by the convergence of three main mechanisms: excitoxicity and ionic imbalance, oxidative stress and apoptotic-like cell death (Fig. 12.7).77–82 Excitotoxicity When the blood supply to a part of the brain is interrupted, neurones deprived of their supply of oxygen and glucose rapidly (within minutes) fail to generate sufficient ATP and energy to maintain ionic gradients across membranes because of failure of the Na+, K+-ATPase pump. This causes an increase in extracellular K+ as well as an influx of Na+, Cl– and Ca++ into the cells. The initial increase in extracellular K+ may spread, triggering depolarizations of neurones and reversal of the amino acid transporters. In these conditions, both voltage-operated and receptor-operated calcium channels are recruited, thus provoking an elevation of free cytosolic Ca++. The depolarized presynaptic neurones release the neurotransmitter glutamate into the synaptic cleft. Reuptake of glutamate is impaired, leading to an increase in extracellular concentration of the excitatory amino acid glutamate. Glutamate activates postsynaptic receptors, including the ionotropic N-methyl-D-aspartate (NMDA), 2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl) proprionate (AMPA) and kainate receptors.83 Upon their activation, these open their associated ion channel to allow the influx of Ca++ and Na+ ions.2,84 Glutamateevoked Na+ increase in astrocytes evokes the transmission of intercellular Na+ and Ca++ metabolic waves and an increase in uptake of glucose.85 Under physiological conditions, calcium ions (Ca++) govern a multitude of cellular processes, including cell growth, differentiation and synaptic activity. Consequently, there are energy-dependent homeostatic mechanisms to maintain a low intracellular Ca++ concentration (about 100 times lower than its extracellular concentration) so that Ca++ signals remain spatially and temporally localized. This permits multiple independent Ca-mediated signalling pathways to occur in the same cell. In excitotoxicity, the excessive synaptic release of glutamate which leads to excessive influx of Ca++,

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 649

12.1 Pathophysiology of acute ischaemic stroke

Reduced CBF

Reperfusion Ischaemia

Cytokine release

Hypoglycaemia Reduced ATP formation Acidosis

Failure of Na-K-ATPase pump

Adhesion molecule expression

Reduced tissue pH, Increased extracellular K+ and increased Na+ ions

Leucocyte adhesion to endothelium

Depolarization of presynaptic neurons

Na+ ions into the cell

Release of excitatory amino acid (glutamate) Opening of voltage-dependent calcium channels and influx of calcium

Activation of NMDA and AMPA receptors

Phospholipase activation

Increase in intracellular calcium

NO production

Increase NO synthase

Free radical formation

DNA damage

Mitochondrial injury

Energy depletion

Protein malfolding

Activation of proteases, lipases, and endonucleases

Membrane damage

Activation of proapoptotic proteins

Fig. 12.7 Schematic diagram of some potential mechanisms of ischaemic brain damage and reperfusion injury. CBF, cerebral blood flow; ATP, adenosine triphosphate; AMPA, D-amino-3hydroxyl-5-ethyl-4-isoxazole propionate; NMDA, N-methyl-D-aspartate; NO, nitric oxide.

Cytochrome C Caspase activation DNA fragmentation Apoptotic cell death

together with any Ca++ release from intracellular compartments, can overwhelm Ca++ regulatory mechanisms and this leads to disregulation of Ca++ homeostasis, and ultimately to cell damage and death. The cell damage associated with a non-physiological unregulated rise in intracellular cytoplasmic Ca++ is caused mainly by changes in total calcium influx.6 The initial calcium load is sequestered, at least in part, by the mitochondria. Mitochondrial injury is associated with failure of mitochondrial functions such as oxidative phosphorylation, and release of reactive oxygen

..

..

Necrotic cell death

species (see below).6,84 Mitochondrial calcium accumulation and oxidative stress can then trigger the assembly (opening) of a high-conductance pore in the inner mitochondrial membrane. The mitochondrial permeability transition (MPT) pore leads to a collapse of the electrochemical potential for H+, thereby arresting ATP production and triggering further production of reactive oxygen species. An increase in total calcium influx also contributes to cell death by activation of Ca++-ATPase (which results in further consumption of cellular ATP); activation of

649

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 650

650

Chapter 12 Specific treatments for acute ischaemic stroke

Ca++-dependent phospholipases, synthases, proteases and endonucleases (which degrade key cytoskeletal and enzymatic proteins such as phospholipids, proteins and nucleic acids); and alteration of protein phosphorylation, which secondarily affect protein synthesis and genome expression.2,84,86 In addition, the augmented intracellular Ca++ enhances the increase in extracellular glutamate, thus propagating excitotoxicity.84 Although Ca++ disregulation is paramount to neurodegeneration, the exact mechanism by which Ca++ ions actually mediate excitotoxicity is less clear.2 One hypothesis suggests that Ca++-dependent neurotoxicity occurs following the activation of distinct signalling cascades downstream from key points of Ca++ entry at synapses, and that triggers of these cascades are physically colocalized with specific glutamate receptors.2 Calcium influx is mediated directly and predominantly by NMDA receptors, but it is also triggered secondarily by Na+ influx through AMPA-, kainate- and NMDA-receptor-gated channels. The resultant excessive Ca++ influx leads to elevated intracellular and intramitochondial Ca++ concentrations and lethal metabolic derangements, which include calcium-dependent activation of intracellular enzyme systems, failure of oxidative phosphorylation and the generation of free radicals that further compromise cells by attacking proteins, lipids and nucleic acids. Oxidative stress (production of free radicals) A free radical is any atom, group of atoms or molecule with an unpaired electron in its outermost orbital.80 Free radicals are produced in small quantities by normal cellular processes in all aerobic cells; for example, ‘leaks’ in mitochondrial electron transport allow oxygen to accept single electrons, forming superoxide (O 2– ). However, they are inherently toxic – they can react with and damage proteins, nucleic acids, lipids and other classes of molecules such as the extracellular matrix glycosaminoglycans (e.g. hyaluronic acid). The sulphur-containing amino acids and the polyunsaturated fatty acids (found in high concentrations in the brain) are particularly vulnerable. Fortunately, cells possess appropriate defence mechanisms in the form of free radical scavengers and enzymes which metabolize free radicals or their precursors. During severe brain ischaemia, insufficient oxygen is available to accept electrons passed along the mitochondrial electron transport chain, leading to eventual reduction (‘electron saturation’) of the components of

this system. The free radicals which are elevated in cerebral ischaemia include O2–, hydroxyl (OH) and nitric oxide (NO). Nitric oxide is generated primarily by neuronal and inducible NO synthases (NOS).84 Like other free radicals, the free radicals which are elevated in cerebral ischaemia react with and damage proteins, nucleic acids and lipids, particularly the fatty acid component of membrane phospholipids, producing changes in the fluidity and permeability of the cellular membranes (lipid peroxidation).80,84 The free radicals also cause microvascular dysfunction and disrupt the blood–brain barrier, leading to brain oedema. In addition, the conversion of xanthine dehydrogenase to xanthine oxidase promotes the cellular formation of toxic oxygen free radicals such as the superoxide anion which further breaks down membrane, cytoskeletal and nuclear structures. An important source of oxidative stress-mediated brain damage is the oxidant reactions due to the formation of peroxynitrite, a powerful oxidant that results from the interaction between NO and superoxide. This anion has been shown to cause cell damage by several mechanisms that include lipid peroxidation, tryrosine nitration, sulphydryl oxidation and nitrosylation, and DNA breakage.84 Neurones undergoing oxidative stressrelated injuries typically display a biphasic or sustained pattern of extracellular signal regulated protein kinase (ERK1/2) activation.87 With reperfusion, reactive oxygen radicals may be generated as by-products of the reactions of free arachidonic acid (released from membrane phospholipids during ischaemia) to produce prostaglandins and leukotrienes, which perhaps lead to reperfusion injury to the brain and its microvessels. Apoptosis Apoptosis is a mode of programmed cell death in which the cell synthesizes proteins and plays an active role in its own demise. It occurs both physiologically and pathologically. For example, during normal human embryonic development, apoptosis results in the loss of the interdigital webs required for normal formation of the fingers and toes. Likewise, tadpoles lose their tails as they develop into frogs. The normal turnover of cells in the intestinal villi is also apoptotic, as is the turnover of normal lymphocytes. Indeed, inhibition of the apoptotic death of lymphocytes may lead to B-cell lymphoma. Although necrosis is the predominant mechanism of cell death after stroke (mediated by excitotoxicity, as described above), growing evidence indicates that hypoxic/ischaemic cell death also continues to some extent hours to days after the onset of ischaemia,

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 651

12.1 Pathophysiology of acute ischaemic stroke

particularly within the peri-infarct zone or ischaemic penumbra, by apoptosis as a consequence of a genetically regulated programme that allows cells to die with minimal inflammation or release of genetic material. Families of executioner enzymes (e.g. caspases, apoptosis inducing factor88) are expressed in neurones and become activated during and after ischaemic stroke. Active caspases kill cells by cleaving critical cell repair and homeostatic proteins as well as cytoskeletal proteins.82 In other words, they dismantle multiple cell processes in the cytoplasm and nucleus to promote cell death by suicidal mechanisms resembling apoptosis.81 The c-Jun N-terminal protein kinase (JNK) signalling pathway is implicated in ischaemia-induced neuronal apoptosis.89 Apoptosis is characterized morphologically by coarse, regularly shaped chromatin condensation, loss of cell volume and extrusion of membrane-bound cytoplasmic fragments (apoptotic bodies), and biochemically by DNA fragmentation in neurones and glia after ischaemic injury. Necrosis may proceed by analogous programmed pathways recently revealed in the nematode Caenorhabditis elegans.90 Other mediators of ischaemic cell death Although the original model of excitotoxicity emphasizes calcium influx through glutamate receptor-coupled ion channels, ionic imbalance (and cell death) may also proceed via other routes.81 a c ido sis Acidosis arises as a result of energy deprivation by sustained tissue ischaemia. It may contribute to tissue damage and prevent or retard recovery during reoxygenation by several mechanisms: activation of novel classes of acid-sensing ion channels that further perturb sodium and calcium homeostasis,91 and inhibition of mitochondrial respiration and lactate oxidation.4,5 Cellular acidosis may also promote intracellular oedema formation by inducing Na+ and Cl– accumulation in the cell via coupled Na+/H+ and Cl–/hydrogen carbonate (HCO 3– ) exchange. peri-infarct depolarization As stated above, ischaemia-induced failure of the Na+, K+-ATPase pump causes an increase in extracellular K+ which may spread, triggering a propagating wave of neuronal and glial depolarization (cortical spreading depression) associated with depressed neuronal bioelectrical activity, transient loss of membrane ionic gradients, and massive surges in extracellular potassium, neurotransmitters and intracellular calcium (150 times the basal level).92 After an initial transient hyperaemia (up to

..

..

200% of baseline), cortical spreading depression exacerbates cellular injury and augments tissue damage by causing profound oligaemia that persists for up to 3 days, thereby imposing an energy burden of re-establishing ionic equilibrium in hypoxic tissue.93 Cortical spreading depression also alters blood–brain permeability by activating brain matrix metalloproteinases,92 and is associated with changes in immediate early genes, growth factors and inflammatory mediators such as interleukin1β and tumour necrosis factor-α.92 Proteolysis of the neurovascular matrix by plasminogen activator and metalloproteinases is believed to be linked with haemorrhagic transformation of the brain infarction and brain oedema. However, treatment with matrix metalloproteinase inhibitors has not been shown to protect the brain. Indeed, a recent study suggested that such inhibitors might contribute to brain damage.94,95 i nfla mma t i on Cerebral ischaemia triggers an inflammatory reaction, by means of an active gene expression, within hours to days that may last up to several months.76,96 Important transcription factors are activated and/or synthesized, including NF-κB, hypoxia-inducible factor 1, interferon regulatory factor 1 and STAT3.97 Several mediators are released or activated, such as adhesion molecules, proteolytic enzymes belonging to the family of metalloproteinases (including matrix metalloproteinases), and inflammatory cytokines.98,99 Cytokines might activate the expression of inflammation-related genes, such as iNOS and cycloxygenase-2.98 The result is infiltration of leucocytes into the brain parenchyma, and the activation of resident microglial and astroglial cells.98,99 The effects of inflammation are both detrimental (e.g. exacerbation of oedema) and potentially beneficial. altered gene expression Ischaemia not only decreases protein and mRNA synthesis, but it also induces at least 100 genes (and thus protein synthesis) which include immediate early genes, stress proteins, growth factors, adhesion proteins, cytokines, kinases and genes directly regulating apoptosis.87,89

Cell death occurs by a necrotic pathway characterized by either ischaemic cell change or oedematous cell change. Death also occurs via an apoptotic-like pathway that is characterized by DNA laddering, a dependence on caspase activity, characteristic protein and phospholipid changes, and morphological attributes of apoptosis. Death may also occur by autophagocytosis.

651

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 652

652

Chapter 12 Specific treatments for acute ischaemic stroke

The four major stages of cell death The cell death process has four major stages:78 • The induction stage includes several changes initiated by ischaemia and reperfusion that are very likely to play major roles in cell death. These include inhibition (and subsequent reactivation) of electron transport, decreased ATP, decreased pH, increased cell Ca++, release of glutamate, increased arachidonic acid, and also gene activation leading to cytokine synthesis, synthesis of enzymes involved in free radical production, and accumulation of leucocytes. These changes lead to the activation of five damaging events, termed perpetrators: the damaging actions of free radicals and their product peroxynitrite; the actions of the Ca++dependent protease calpain; the activity of phospholipases; the activity of poly-ADPribose polymerase (PARP); and the activation of the apoptotic pathway. • The second stage of cell death involves the long-term changes in macromolecules or key metabolites that are caused by the perpetrators. • The third stage involves long-term damaging effects of these macromolecular and metabolite changes, and of some of the induction processes, on critical cell functions and structures that lead to the defined end stages of cell damage. These targeted functions and structures include the plasmalemma, the mitochondria, the cytoskeleton, protein synthesis and kinase activities. • The fourth stage is the progression to the morphological and biochemical end stages of cell death. Of these four stages, the last two are the least well understood. Rather little is known of how the perpetrators affect the structures and functions and whether and how each of these changes contribute to cell death. The key step in ischaemic cell death is adequate activation of the perpetrators.78 Implications for future neuroprotective therapies The neurovascular unit – a conceptual model comprised of cerebral endothelial cells, astrocytes, neurones and the extracellular matrix – provides a framework for understanding the pathophysiology of ischaemic stroke and developing treatments based on an integrative and dynamic view of evolving tissue damage. Drugs which aim to protect the neurovascular unit (neuroprotectants) have been evaluated preclinically and clinically in a large number of trials (section 12.8.1). However, a systematic review of 1026 experimental drug treatments in acute stroke (clinical studies of 114 drugs, animal models of 912 drugs) found that there was no evidence that drugs used clinically were more effective experimentally than

those tested only in animal models.100 The quality of the studies was highly variable: • poor trial study design; • inadequate sample size; • unreliable outcome evaluation; • anatomical and aetiological hetereogeneity in the study populations; • extrapolating results from young rats to old humans with comorbidities; • dose ceilings imposed by adverse effects of the drug; • inadequate drug penetration to the site of the lesion; • possible differences in responsiveness between ischaemic grey and white matter; • drug delivery after the therapeutic time window had closed; • publication bias in the basic science literature; • misleading hypotheses from the basic scientific researchers linking NMDA-receptor-mediated excitotoxicity to hypoxic-ischaemic neuronal death. All this questions whether the most efficacious drugs have been selected for stroke clinical trials, and suggests that greater rigour in the conduct, reporting and analysis of animal data is required to improve the translation of scientific advances from bench to bedside.100 Moreover, new approaches and targets for cytoprotection are required.84,101–105 There seem to be so many pathways leading from ischaemia to neuronal cell death that blocking just one of them is probably fruitless; perhaps this is like ligating a feeding artery to an arteriovenous malformation (there are always others that will take over to keep up its blood supply), or trying to extinguish a large blazing fire with just a small garden hose.

12.1.5 Ischaemic cerebral oedema Cerebral ischaemia not only causes loss of neuronal function but also cerebral oedema. Within hours of onset, cytotoxic cerebral oedema arises as a result of cell membrane damage allowing intracellular accumulation of water. The grey matter is affected more than the white matter. The CT and MRI scan appearances are of wellcircumscribed altered signal density involving the cortex and subcortex (sections 5.4.2 and 5.5.2). Cytotoxic oedema reaches a maximum volume at 2–4 days and then subsides over 1–2 weeks. Initially, endothelial tight junctions are maintained and the blood–brain barrier remains intact. However, after several days of ischaemia, breakdown of the blood– brain barrier occurs, leading to vasogenic cerebral oedema as plasma constituents enter the brain extracellular space, affecting the white matter more than the grey matter. Blood–brain barrier breakdown may be initiated and regulated by several proinflammatory mediators

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 653

12.1 Pathophysiology of acute ischaemic stroke

(oxidative mediators, adhesion molecules, cytokines, chemokines). These mediators not only regulate the magnitude of leucocyte extravasation into brain parenchyma, but also act directly on brain endothelial cells causing loosening of junction complexes between endothelial cells, increasing brain endothelial barrier permeability, and causing vasogenic oedema.106 The brain scan appearance of vasogenic oedema includes the characteristic finger-like projections of altered signal density in the white matter, which characteristically accompany cerebral tumours. Imaging studies indicate that fluid volume after infarction peaks in 7–10 days and oedema remains detectable for about 1 month. The effects of cerebral oedema are to compromise blood flow even further by increasing pressure in the extravascular space, thus causing vascular congestion and sometimes haemorrhagic transformation, and to cause mass effect, brain shift and eventually brain herniation (Figs 12.8 and 12.9). The main danger of herniation is that it can initiate vascular and obstructive complications

C

P U

Fig. 12.8 Uncal and cingulate herniation: a mass such as a large cerebral haemorrhage or oedematous infarct displaces the diencephalon and mesencephalon horizontally and caudally. The cingulate gyrus (C) on the side of the lesion herniates under the falx cerebri. The uncus (U) of the ipsilateral temporal lobe herniates under the tentorium cerebelli and becomes grooved and swollen and may compress the ipsilateral oculomotor (third cranial) nerve causing pupillary dilatation (Hutchinson’s sign). The cerebral peduncle (P) opposite the supratentorial notch becomes compressed against the edge of the tentorium, leading to grooving (Kernohan’s notch)107 and causes a paresis ipsilateral to the cerebral mass lesion. Central downward displacement also occurs but is less marked than in Fig. 12.9. Modified from Plum F, Posner JB. The Diagnosis of Stupor and Coma. Philadelphia, PA: F.A. Davis & Co., 1985.319

..

..

Fig. 12.9 Central transtentorial herniation. Diffuse or multifocal swelling of the cerebral hemispheres (or bilateral subdural or extradural haematomas) compresses and elongates the diencephalon from above. The mammillary bodies are displaced caudally. The cingulate gyrus is not herniated. Modified from Plum F, Posner JB. The Diagnosis of Stupor and Coma. Philadelphia, PA: F.A. Davis & Co., 1985.319

which aggravate the original expanding lesion by compressing important vessels and tissues and causing even more cerebral ischaemia, congestion and oedema which, in turn, enhance the expanding process. In addition, the herniating brain can compress the aqueduct and subarachnoid spaces and so interfere with cerebrospinal fluid circulation, leading to hydrocephalus and increased cerebrospinal fluid pressure. There are three anatomical patterns of supratentorial brain shift which can be identified by their end stages: cingulate herniation, central transtentorial herniation and uncal herniation. Cingulate herniation Cingulate herniation occurs when the expanding cerebral hemisphere shifts across the intracranial cavity, forcing the ipsilateral cingulate gyrus under the falx cerebri, compressing and displacing the internal cerebral vein and the ipsilateral anterior cerebral artery. This may lead to additional infarction in the territory of the anterior cerebral artery (Fig. 12.8). Central transtentorial herniation Central transtentorial herniation of the diencephalon is the end result of displacement of the cerebral

653

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 654

654

Chapter 12 Specific treatments for acute ischaemic stroke

hemispheres and basal nuclei, compressing and eventually displacing the diencephalon and the adjoining midbrain rostro-caudally through the tentorial notch (Fig. 12.9). The great cerebral vein is compressed, which raises the hydrostatic pressure of the entire deep territory it drains. In addition, downward displacement of the midbrain and pons stretches the medial perforating branches of the basilar artery (the artery cannot shift downward because it is tethered to the circle of Willis), leading to paramedian brainstem ischaemia (and haemorrhage if perfusion continues). Uncal herniation Uncal herniation characteristically occurs when expanding lesions in the temporal fossa or temporal lobe shift the inner, basal edge of the uncus and hippocampal gyrus toward the midline so that they bulge over the incisural edge of the tentorium, and push the adjacent midbrain against the opposite incisural edge (Fig. 12.8).107 The third cranial nerve and the posterior cerebral artery on the side of the expanding temporal lobe are often caught between the overhanging swollen uncus and the free edge of the tentorium or the petroclinoid ligament, leading to a third nerve palsy and occipital and medial temporal lobe infarction and swelling, which further compounds the problem. Transtentorial herniation is the most common cause of death during the first week after acute stroke, accounting for about 80% of deaths in cerebral infarction and 90% in intracerebral haemorrhage. The risk of death peaks within 24 h for intracerebral haemorrhage, but later at 4–5 days for cerebral infarction as ischaemic cerebral oedema develops. Brainstem compression, with subsequent haemorrhage and infarction within it, accounts for the serious morbidity and mortality associated with herniation. Transtentorial herniation is the most common cause of death within the first week of onset of both ischaemic stroke and intracerebral haemorrhage.

12.2 General treatment considerations

The primary aims of the specific treatments considered in this chapter are, broadly speaking, to minimize the volume of brain damaged by ischaemia. The assumption is that, if this can be achieved without excess hazard, neurological impairment, disability and handicap should

all correspondingly be reduced in survivors. For patients with a large volume of ischaemic brain, minimizing the volume which infarcts should also reduce the risk of early death, particularly from ischaemic cerebral oedema and transtentorial herniation (section 12.1.5). The pathophysiology of acute cerebral ischaemia is complex (section 12.1.3), and many potential treatments may have more than one mechanism of action; for some, the precise mechanism is unknown. However, the main targets of acute treatment are to restore and then maintain blood flow, and simultaneously keep alive as much ischaemic brain tissue as possible while the blood supply is restored, either spontaneously or therapeutically.

12.2.1 How to evaluate evidence about different treatments Evidence-based medicine We try to base our decisions on whether or not to use a particular treatment in clinical practice on the evidence from randomized trials and systematic reviews of randomized trials. This approach is part of what is now known as ‘evidence-based medicine’, a style of medical practice which Sackett described as ‘a life-long process of self-directed learning in which caring for our patients creates the need for clinically important information about diagnosis, prognosis and therapy’.108 The book by Sackett and co-authors on the subject is short and clear and provides succinct advice on how to find the best evidence, appraise it critically and then apply it in everyday clinical practice (Table 12.4).108 This section summarizes what the best forms of evidence are (and why) and where up-to-date evidence can be found. Why are randomized trials the best way to evaluate treatments? There is now little disagreement that the randomized controlled trial (RCT) is the best way to evaluate most treatments109 (Table 12.4). Weaker research designs, i.e. case series without any controls, series with retrospective controls, and series with concurrent controls but nonrandom treatment allocation, may introduce far too many biases into the assessment of treatment effect.108–110 The Medical Research Council trial of streptomycin for pulmonary tuberculosis, designed by the statistician Sir Austin Bradford Hill, is probably the strongest contender for the title ‘first clinical trial to use strictly random allocation’, although there is some debate about this claim.111,112 The trial not only established the benefits of the treatment beyond all reasonable doubt, but also served as the most ethical and equitable way of utilizing

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 655

12.2 General treatment considerations Table 12.4 Methodological questions in the critical assessment of an article evaluating treatment (modified from Sackett).108,109 Was the assignment of patients to treatment really randomized?* Was the similarity between groups documented? Was prognostic stratification used in treatment allocation? Was foreknowledge of the randomly allocated treatment possible? Were all clinically relevant outcomes reported? Was case fatality as well as morbidity reported? Were deaths from all causes reported? Were quality-of-life assessments conducted? Was outcome assessment blind to treatment allocation? Were the study patients recognizably similar to your own? Were reproducibly defined exclusion criteria stated? Was the setting primary, secondary or tertiary care? Was the type of patient included in the study clearly described? Were both statistical and clinical significance considered? If statistically significant, was the difference clinically important? If not statistically significant, was the study big enough to show a clinically important difference if one really exists? Is the therapeutic manoeuvre feasible in your practice? Available, affordable, sensible? Were contamination† and co-intervention avoided? Was the manoeuvre administered blind? Was compliance measured? Were all patients who entered the study accounted for at its conclusion? (i.e. were drop-outs, withdrawals, non-compliers and those who crossed over handled appropriately in the analysis?) *Methods of allocation such as alternation, use of hospital number or date of birth all provide foreknowledge of the next treatment allocation, and so allow the trial allocation process to be subverted which may lead to selection bias between the two treatment groups. †In this context, contamination implies that some patients allocated active treatment either did not receive the treatment or were given the treatment which the other treatment group should have received, or some non-trial treatment with similar properties to one of the trial interventions under test. Note: reports of RCTs should conform to the requirements of the CONSORT statement as far as possible.127

the very limited supply of the drug that was available in Britain at the time. It also prevented the treatment ‘creeping’ into routine clinical practice without being properly evaluated. We could answer many of the therapeutic questions posed in this book a great deal faster if a larger proportion of stroke patients were entered in appropriate randomized controlled trials.

..

..

Strictly random allocation and proper concealment minimize selection bias.110,113 If the randomizing clinician can find out what treatment the next patient will be allocated (e.g. by holding a sealed trial allocation envelope up to a bright light), he may be tempted not to enter the patient in the trial. Allocation concealment reduces or avoids this happening.113 A central telephone randomization system offers complete concealment: the clinician telephones a randomization centre and gives the patient’s details, these are then recorded on the central computer during the call, and the computer generates the next treatment allocation only after all the baseline data are checked and entered; the patient is then irrevocably in the trial and selection bias cannot occur.113 Studies where patients are entered into a trial by selecting a numbered drug pack in the participating hospital, administering the drug/ placebo contained in the pack, and then telephoning a central office to notify the trial administrators that a patient has been entered into the trial, are prone to all sorts of bias. In the worst case, the pack could be opened and given but the patient details then not notified to the trial office if there was an early adverse event – the pack could just be recorded as ‘opened and discarded’. The pack drug might not have been administered when the hospital said it was, but possibly earlier so as to look better on the records. There is no possibility of balancing the treatment allocation on important key prognostic variables resulting in imbalances between treatment and control arms which could skew the results. Unfortunately, with this system of ‘randomization’ any of these scenarios is possible, and probably has happened. It is important to reduce observer bias, if possible, by ‘blinding’ both the patients and the observers who collect the outcome data.110,113 Random error is reduced by recruiting a sufficiently large sample of patients. However, RCTs need to be surprisingly large to provide really reliable evidence on the balance of risk and benefit for a particular treatment.110 If the measure of outcome is an uncommon but serious event, such as death, and the effect of the treatment on that outcome is only moderate, then the trial may need to recruit several tens of thousands of patients.110 To recruit such large numbers often involves clinicians in many countries, and hence the trial design needs to be simple, with minimal data collection and audit (though such simplicity is more difficult to achieve in the current over-regulated and bureaucratic research climate).110,114 The effort of conducting such large RCTs is substantial, and so many of the ‘mega trials’ have used a factorial design to assess two or three treatments at the same time without loss of statistical efficiency.110

655

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 656

656

Chapter 12 Specific treatments for acute ischaemic stroke

Good trial design seeks to reduce bias and random error in the assessment of treatment effect: strict randomization with concealment reduces selection bias; blinding reduces observer bias; and recruitment of large numbers of patients reduces random error. Problems with randomized controlled trials in acute stroke from 1956 to the present Eight years after the streptomycin trial was published in 1948, Dyken and White wrote of the many methodological problems they identified in evaluating treatments for acute stroke; not the least was to know, in an individual patient, whether or not treatment had been effective.115 They wrote this in the report of the first quasirandomized study of a medical treatment for acute stroke.115 Thirty-six patients were alternately (but not randomly) assigned to cortisone or control. Thirteen of the cortisone patients and 10 of the controls died; a trend against treatment, but an inconclusive result. Sadly, a review of the quality of the final reports of RCTs of drug therapies for acute stroke conducted over the next 40 years (i.e. between 1956 and 1996) makes depressing reading.116 Many trials had weaknesses which included: over-complex eligibility criteria; incomplete descriptions of the methods used (particularly the method of randomization); inadequate sample size; inappropriate measures of outcome; poor standards of execution (i.e. poor trial discipline); an unacceptably high proportion of patients ‘lost’ to follow-up; inappropriate statistical methods (often with inappropriate subgroup analysis); and failing to account for all patients randomized.116 On the other hand, trial quality did appear to improve in parallel with an increase in trial size.116 While this is encouraging, a more recent review of reports of the randomized acute stroke trials completed before 2001 showed that most trials were underpowered, i.e. power 0.0001) adverse)

95% confidence interval

0.1 0.25 0.5 1.0 2.0 4.0 10.0 Anticoagulant Anticoagulant better worse

Fig. 12.11 Anticoagulants in acute ischaemic stroke: relative effects within the treatment period. Results of a systematic review of the 21 randomized trials comparing heparin with control groups in patients with acute ischaemic stroke: effects on recurrent ischaemic stroke, symptomatic intracranial haemorrhage and death, all within the scheduled treatment period. The estimate of treatment effect for each heparin regimen is expressed as an odds ratio (solid square) and its 99% confidence interval (horizontal line). The size of the square is proportional to the amount of information available. An odds ratio of 1.0 corresponds to a treatment effect of zero, an odds ratio of less than 1.0 suggests treatment is better than control, and an odds ratio of greater than 1.0 suggests treatment is worse than control. The overall result and its 95% CI is represented by an open diamond. The figures given to the right are relative odds reductions (SD). (Drawn from data from Sandercock, 2003.)156 NS, not significant; SD, standard deviation; s.c., subcutaneous; i.v., intravenous.

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 667

12.4 Selective use: anticoagulants

Events/Patients Anticoagulant Control

Anticoagulant regimen

Odds ratio & CI

Odds reduction

Death at the end of follow-up Unfractionated heparin s.c. Unfractionated heparin i.v. Low-molecular-weight heparin s.c. Heparinoid s.c. Heparinoid i.v. Thrombin inhibitor i.v. Oral anticoagulant

2196/9861 (22.3%) 17/112 (15.2%) 178/723 (24.6%) 4/50 (8.0%) 42/646 (6.5%) not evaluated 8/26 (30.8%)

2129/9879 (21.6%) 8/113 (7.1%) 88/355 (24.8%) 4/25 (16.0%) 38/635 (6.0%) not evaluated 7/25 (28.0%)

–5% (2P > 0.1:NS adverse)

Total Death or dependency at the end of follow-up

Fig. 12.12 Anticoagulants in acute ischaemic stroke: relative effects at the end of follow-up. Results of a systematic review of the five randomized trials comparing anticoagulants with control groups in patients with acute ischaemic stroke reporting long-term outcome: effects on death from all causes, and death or dependency. Same conventions as Fig. 12.11. (From Gubitz et al. ‘Anticoagulants for acute ischaemic stroke’, Cochrane Database Systematic Reviews 2004: (3)172 copyright Cochrane Collaboration, reproduced with permission.)

Unfractionated heparin s.c. Unfractionated heparin i.v. Low-molecular-weight heparin s.c. Heparinoid s.c. Heparinoid i.v. Thrombin inhibitor i.v. Oral anticoagulant Total

99% or

95% confidence interval

Bleeding risks with different regimens In the Cochrane review, each of the regimens tested, when compared with control, appeared to increase the risks of both symptomatic intracranial haemorrhage and extracranial haemorrhage.172 The relative increase in symptomatic intracranial haemorrhage was consistent across the different regimens, although (because of small numbers) it was only statistically significant for subcutaneous unfractionated heparin (Fig. 12.11). Indirect

..

6062/9718 (62.4%) not evaluated 210/355 (59.2%) 15/29 (51.7%) 167/635 (26.3%) not evaluated not evaluated

1% (2P > 0.1:NS)

nine fewer recurrent ischaemic strokes per 1000 patients treated (OR 0.76; 95% CI 0.65–0.88), they were also associated with a similar sized nine per 1000 increase in symptomatic intracranial haemorrhages (OR 2.52; 95% CI 1.92–3.30) (Fig. 12.11). There was no evidence that anticoagulants reduced the odds of death from all causes at the end of follow-up (OR 1.05; 95% CI 0.98–1.12) (Fig. 12.12);172 if anything, there was a non-significant excess of deaths with anticoagulants. Similarly, there was no evidence that anticoagulants reduced the odds of being dead or dependent at the end of follow-up (OR 0.99; 95% CI 0.93–1.04) (Fig. 12.12).

..

6063/9717 (62.4%) not evaluated 400/723 (55.3%) 13/28 (46.4%) 159/641 (24.8%) not evaluated not evaluated

0.0 0.5 1.0 1.5 2.0 Anticoagulant Anticoagulant better worse

comparisons of different dosing regimens showed consistently higher bleeding risks with higher dose regimens. In the IST, patients allocated to subcutaneous unfractionated heparin were randomized to high dose (12 500 IU twice daily) or to low dose (5000 IU twice daily) and the proportions with symptomatic intracranial haemorrhage were 1.8% and 0.7%, respectively – a highly significant 11 per 1000 excess with the higher dose (2P < 0.00001).154 As far as extracranial bleeds are concerned, for every 1000 acute ischaemic stroke patients treated with anticoagulants, about nine have a major extracranial haemorrhage. The indirect comparisons of different agents showed that the bleeding risks were higher with higher dose regimens. In the IST, the risk of major extracranial bleeds was 2% among patients allocated high-dose and 0.6% among those allocated low-dose heparin – a highly significant 14 per 1000 excess with the higher dose (2P < 0.00001). A systematic review of all trials directly comparing high- with low-dose anticoagulants in acute ischaemic stroke supported the finding that the bleeding risks were dose-dependent for both intra-and extracranial bleeds.182

667

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 668

668

Chapter 12 Specific treatments for acute ischaemic stroke

The RCTs published since then comparing different doses of low-molecular-weight heparin in acute ischaemic stroke also showed that intracranial bleeding was dose-dependent.179,183 A recent open RCT comparing fulldose intravenous heparin with low-dose subcutaneous heparin, started within 3 h of stroke onset, also showed a higher risk of intracranial haemorrhages with the higher dose (6.2% vs 1.4%).177 Deep venous thrombosis and pulmonary embolism Data on the effects of anticoagulants on deep venous thrombosis (DVT) are only available for 916 acute stroke patients. There was heterogeneity between the trials, which makes it harder to give a reliable overall estimate of treatment effect. Overall, symptomatic or asymptomatic DVT occurred in 43% of controls and 15% of treated patients, a highly significant 79% reduction in relative odds with anticoagulants (95% CI 61–85%). Thus, for every 1000 patients treated, 280 avoid DVT. Fatal or non-fatal pulmonary embolism (PE) was not systematically sought in the trials. It was reported in only 0.9% of controls and 0.6% of treated patients, a significant 39% reduction in relative odds with anticoagulants (95% CI 17–55%), i.e. for every 1000 patients treated, three avoid PE. It is difficult to judge whether the reductions in DVT or PE are dose-dependent from the indirect comparisons. In the IST, fatal or non-fatal PE occurred in 0.4% of those allocated high dose and 0.7% of those allocated low dose, a non-significant difference.154 A systematic review of all of the direct randomized comparisons confirmed the greater reduction in PE with higher doses, but the absolute benefit was very small.182 If we allow for the likely under-ascertainment of pulmonary embolism (section 12.3.2), and assume that the true risk in the controls was 3%, and then apply the same 39% proportional reduction (i.e. from 3% to 1.85%), for every 1000 patients treated about 12 would avoid PE. However, even if the benefit is that large, it will still be substantially offset, since an extra nine patients will have a major extracranial haemorrhage associated with anticoagulants. Suspected cardioembolic ischaemic stroke All patients with atrial fibrillation (AF) and ischaemic stroke should have a CT brain scan, since about 5–10% turn out to have intracerebral haemorrhage. A detailed analysis of the 3169 patients in IST with acute ischaemic stroke and AF confirmed that such stroke patients have a higher risk of early death than those in sinus rhythm.184 This higher death rate was accounted for by the older age and larger infarcts among those with AF but not

by a higher risk of early recurrent ischaemic stroke. In patients with AF, the absolute risk of early recurrent stroke is low (in the IST, about 5% of patients in AF allocated to ‘avoid heparin’ had a recurrent stroke of ischaemic or unknown type within 14 days), and there was no net advantage to treatment with heparin.184 The Cochrane review showed no statistically significant difference in the effect of treatment on death or dependence at final follow-up if a cardiac source of embolism was suspected (OR 1.00; 95% CI 0.85–1.18) or not suspected (OR 1.00; 95% CI 0.94–1.06) to be the cause of the stroke.172 The small Heparin in Acute Embolic Stroke Trial (HAEST) compared a low-molecular-weight heparin with aspirin in 449 patients with acute ischaemic stroke of suspected cardioembolic origin and did not show any advantage to low-molecular-weight heparin.185 These data do not support the widespread use of either intensive (i.e. full-dose intravenous) unfractionated or lowmolecular-weight heparin regimens in the acute phase of ischaemic stroke associated with AF.160 In patients with suspected acute cardioembolic ischaemic stroke, there is no evidence to support the routine use of immediate anticoagulation with heparin, heparinoid or low-molecular-weight heparin. Aspirin is a safe and effective alternative to anticoagulants, which reduces the risk of early recurrent stroke in patients with acute ischaemic stroke and atrial fibrillation. Progressing hemispheric stroke By far the most important step is to find out why the stroke is ‘progressing’. There are numerous causes besides propagating cerebral arterial thrombosis for a worsening neurological deficit (section 11.5) (Table 11.7). These factors should be identified and – if possible – treated. For patients in whom all other possible causes for progression have been excluded, many textbooks and reviews recommend immediate intravenous heparin. However, there have not been any trials of intravenous heparin specifically in patients with progressing stroke, and neither the trials of intravenous unfractionated heparin nor of intravenous heparinoid in patients with acute ischaemic stroke showed net benefit overall.172 There is therefore no direct or indirect evidence to support the use of heparin (either intravenous or subcutaneous) in this particular type of patient.160 Basilar thrombosis The IST included over 2000 patients with posterior circulation ischaemic infarcts, and there was no evidence

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 669

12.4 Selective use: anticoagulants

that the effects of treatment in this subgroup were any different to those seen in the trial overall.154 However, it is likely that only a small proportion had occlusion of the basilar artery. Further trials of anticoagulants in ‘vertebrobasilar territory infarcts’ are probably not warranted; but a trial focused on patients with proven basilar occlusion might be justified, although trials which seek to recruit a type of patient only rarely encountered in clinical practice are notoriously difficult to do. This lack of evidence led a recent guideline to state that anticoagulants were ‘not recommended’ for this indication.160 Acute stroke in patients with acute myocardial infarction (See also section 7.10.) The first step in management is to determine whether the stroke was due to intracerebral haemorrhage, either spontaneous or iatrogenic (e.g. caused by thrombolytic or anticoagulant treatment). Some patients may have iatrogenic ischaemic strokes due to particulate emboli reaching the brain as a complication of some invasive cardiological procedure such as coronary angiography or angioplasty. The value of anticoagulants in these latter patients is not clear, as the embolic material is often atheromatous debris from the large arteries rather than fresh thrombus or platelet aggregates. Patients with full-thickness anterior myocardial infarction (MI) have a higher than average risk of developing left ventricular (LV) thrombus (section 7.10). Although two small trials186,187 showed that medium-dose subcutaneous heparin (12 500 IU twice daily) reduced the frequency of LV thrombus, an overview of the 26 trials of anticoagulants in acute MI (including 73 000 patients) found little evidence of any significant further net clinical benefit (in terms of major clinical events) from adding either subcutaneous or intravenous heparin to the treatment of patients who are given aspirin.188,189 The value of anticoagulants in a patient with an acute MI complicated by acute ischaemic stroke is therefore unclear. Oral anticoagulants, given to stroke-free MI survivors for several months, probably reduce the risk of having a first stroke in the longer term, but given the cost and inconvenience of oral anticoagulants and the uncertainty about the optimal intensity of treatment, aspirin is probably the long-term antithrombotic agent of choice in most patients.189,190 However, 6 months of oral anticoagulants may be worthwhile in patients with MI complicated by acute ischaemic stroke thought to be due to embolism from the left ventricle, especially if congestive heart failure, AF or extensive LV dysfunction are present.189

..

..

12.4.3 Who to treat Although there is no indication to use anticoagulants routinely in patients with acute ischaemic stroke,160,166,191 there are still some occasions where they may be appropriate; these are listed in Table 12.8 and discussed below. The use of heparin after thrombolysis to prevent re-occlusion of the opened cerebral artery is discussed in section 12.5.2. Heparin, when used for the immediate treatment of patients with acute ischaemic stroke, has risks (haemorrhage in the brain and elsewhere) which exactly cancel out the benefits (fewer recurrent ischaemic strokes and less venous thromboembolism). We therefore do not use this treatment routinely. Selected patients at high risk of deep venous thrombosis Factors which increase the risk of deep venous thrombosis (DVT) and pulmonary embolism (PE) after stroke include: hemiplegia or immobility; dehydration; a history of previous episodes of venous thromboembolism; and a known thrombophilia or active cancer.191 Simple, reliable methods to predict which patients are most likely to develop DVT would be very useful. Guidelines vary in their recommendations about whether or not heparin should be used for DVT prevention. The Scottish Intercollegiate Guidelines Network recommended aspirin for all patients with ischaemic stroke and that low-dose subcutaneous heparin should only be added to aspirin in patients at high risk of DVT and PE.191 The Joint American Stroke Association/American Academy of Neurology (ASA/AAN) guideline states ‘Subcutaneous unfractionated heparin, low molecular weight heparins, and heparinoids may be considered for DVT prophylaxis in at-risk patients with acute ischaemic stroke, recognising that non-pharmacologic treatments for DVT prevention also exist. A benefit in reducing the incidence of PE has not been demonstrated. The relative benefits of these agents must be weighed against the risk of systemic and intracerebral haemorrhage.’160 It is not clear whether low-dose subcutaneous heparin can add useful benefit to aspirin for the prevention of DVT and PE without causing an increase in bleeding complications (Table 12.7 gives an estimate of the increase in risk of extracranial bleeding when subcutaneous heparin is added to aspirin).154 Other strategies for prevention of DVT and PE (e.g. rehydration, early mobilization, graded compression stockings) and the management of stroke patients who develop symptoms and signs suggestive of DVT or PE after admission are discussed in section 11.13.

669

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 670

670

Chapter 12 Specific treatments for acute ischaemic stroke

Anticoagulants worthwhile Transient ischaemic attack or mild ischaemic stroke with complete recovery within 1–2 days and in atrial fibrillation (start as soon as possible) Non-disabling ischaemic stroke, with or without minor (petechial) haemorrhagic transformation and in atrial fibrillation (best time to start unclear) Anticoagulants probably worthwhile Acute myocardial infarction within past few weeks and confirmed ischaemic stroke or transient ischaemic attack Disabling ischaemic stroke and atrial fibrillation Anticoagulants not worthwhile Low risk of recurrent cardioembolic stroke without anticoagulants: Low-risk cardiac lesion (e.g. hypertrophic cardiomyopathy, minor degree of mitral leaflet prolapse) Other, non-cardiac lesion, a more likely cause of the ischaemic stroke Little to gain from long-term anticoagulation: Already severely disabled before the stroke Moribund High risk of cerebral haemorrhage on anticoagulants: Infective endocarditis Large cerebral infarct with midline shift Likely to comply poorly with anticoagulation Severe, uncontrolled hypertension Major haemorrhagic transformation on brain CT Absolute or relative contraindication to anticoagulants (Table 12.9)

The choice of method for deep venous thrombosis prophylaxis largely depends on its perceived risk, the likely duration of immobility, and any predisposing factors. For patients with mild strokes who mobilize within a day or so, and have no other predisposing factors for deep venous thrombosis, aspirin is probably adequate. For immobile patients with predisposing factors (and who do not have bad arterial disease or sensory loss in the legs), graded compression stockings may provide additional benefit. If stockings are not acceptable, then low-dose subcutaneous unfractionated heparin in addition to aspirin is an alternative. Atrial fibrillation Patients with atrial fibrillation (AF) who have had a stroke or transient ischaemic attack are likely to benefit from long-term oral anticoagulants as secondary prevention, provided there are no contraindications (Table 12.8) (see section 16.6 for further details on the selection of patients with AF most likely to benefit). However, as discussed above, both fixed-dose subcutaneous unfractionated heparin, adjusted-dose intravenous unfractionated heparin and low-molecular-weight heparin are

Table 12.8 Patients with acute ischaemic stroke of suspected cardioembolic origin who should be considered for immediate anticoagulation, and those who should have anticoagulants deferred (or avoided completely).

not recommended for the prevention of early recurrent ischaemic stroke in acute ischaemic stroke presumed due to cardioembolic stroke.160 To help decide whether (and when) to use anticoagulants in the acute phase, one should therefore take into account the factors that are likely to increase the risk of haemorrhagic transformation of the infarct (section 12.4.5), the likely risk of recurrent ischaemic stroke within the first week or so, and the severity of the stroke (sections 12.4.5 and 16.6.6). If anticoagulants are indicated, see section 12.4.5 on when to start and section 12.4.6 for the choice of agent.

12.4.4 Who not to treat Relative contraindications and cautions about the use of anticoagulants are given in Table 12.9. Any patient with acute stroke due to established intracranial haemorrhage, or who has not had a CT brain scan to exclude intracranial haemorrhage, should not be given anticoagulants. The evidence on the effects of heparin in patients with haemorrhagic stroke is scant but suggests – as one might expect – it may well be associated with some harm, though the confidence interval is wide.159 Warfarin should not be used during pregnancy (Table 12.9 and section 16.6.3).

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 671

12.4 Selective use: anticoagulants Table 12.9 Contraindications to, and cautions with, full-dose anticoagulants (these depend on individual circumstances and are seldom absolute) (adapted from Scottish Intercollegiate Guidelines Network, 1999).193 Uncorrected major bleeding disorder Thrombocytopenia Haemophilia Liver failure Renal failure Uncontrolled severe hypertension Systolic blood pressure over 200 mmHg Diastolic blood pressure over 120 mmHg Potential bleeding lesions Active peptic ulcer Oesophageal varices Unruptured intracranial aneurysm (especially if large and untreated) Proliferative retinopathy Recent organ biopsy Recent trauma or surgery to head, orbit or spine Confirmed intracranial or intraspinal bleeding Evidence on MR or CT of previous intracranial bleeding Recent stroke, but patient has not had brain CT scan or MRI Recent stroke, with large infarct on brain CT or MRI (i.e. high risk of haemorrhagic transformation) History of heparin-induced thrombocytopenia or thrombosis (if heparin planned) If warfarin planned Pregnancy Homozygous protein C deficiency (risk of skin necrosis) History of warfarin-related skin necrosis Uncooperative/or unreliable patients (long-term therapy) Risk of falling Unable to monitor anticoagulation intensity

would wait at least 1–2 weeks before starting. We would then consider the relative contraindications listed in Table 12.9 and, if necessary, delay the start of anticoagulants further (or abandon the idea altogether). When oral anticoagulants are started in this way (i.e. without aggressive loading doses) some time after the acute event, concomitant heparin (to overcome any transient prothrombotic state associated with the start of warfarin) is probably not needed.193 In patients with transient ischaemic attack or with an ischaemic stroke which resolves within a day or so, the risk of haemorrhagic transformation is probably negligible; if such a patient has a clear indication for the use of long-term anticoagulants (most commonly AF), we would start treatment immediately. In a patient with a disabling acute ischaemic stroke and atrial fibrilation, the risk of stroke recurrence within the first 2 weeks is not high, so we would start aspirin immediately, avoid anticoagulants for the first week or two, or longer if we felt the risk of haemorrhagic transformation of the infarct remained high (e.g. large infarct on CT with mass effect). Patients with transient ischaemic attack or very minor ischaemic stroke and atrial fibrilation can be started on both aspirin and oral anticoagulants immediately. If aspirin is given with anticoagulants, it should be stopped once the international normalized ratio is in the therapeutic range.

12.4.6 Agent/dose/route/adverse effects Agent

12.4.5 When to start The available trial data do not provide reliable evidence on the best time to start anticoagulants. So, in the occasional circumstance where we do feel compelled to use them in acute stroke (section 12.4.3), the decision must be based on the likely risk of events without treatment, and the risk of haemorrhagic transformation with treatment. In the IST, the frequency of recurrent ischaemic or undefined stroke within 14 days was not significantly different in those with and without AF: 3.9% among those with AF vs 3.3% without.154,184 Symptomatic haemorrhagic transformation of the infarct occurs most commonly during the first 2 weeks and the risk is highest in patients with large infarcts192 and in patients given more intensive heparin regimens.154,177,182 Therefore, in a patient with a major ischaemic stroke and AF, we

..

..

Oral anticoagulants do not act sufficiently fast to have a major effect within the first few hours of onset of an ischaemic stroke. Anticoagulants must be given by injection if they are to achieve a rapid effect. Most of the evidence on heparin in acute stroke comes from trials which used unfractionated heparin.172 Though other agents have been tested, systematic reviews of the direct and indirect comparisons of different regimens did not provide evidence that any one agent was better than unfractionated subcutaneous heparin.172,173 In choosing between the different heparin regimens and aspirin, the ASA/AAN guideline states ‘Intravenous unfractionated heparin or high-dose low molecular weight heparin or heparinoids are not recommended for any specific subgroup of patients with acute ischaemic stroke that is based on any presumed stroke mechanism or location (e.g., cardioembolic, large vessel atherosclerotic, vertebrobasilar, or ‘progressing’ stroke) because data are insufficient. Although the low molecular weight heparin,

671

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 672

672

Chapter 12 Specific treatments for acute ischaemic stroke

dalteparin, at high doses may be efficacious in patients with AF, it is not more efficacious than aspirin in this setting. Because aspirin is easier to administer, it, rather than dalteparin, is recommended for the various stroke subgroups (Grade A).’160 The results of the trials published since that guideline176,178,179,320 have not materially altered those conclusions. Dose and route The risk of bleeding with heparin is clearly dosedependent;154,179,182,183 the higher the dose, the higher the risk of intracranial and extracranial haemorrhage and there is no evidence to support the use of full-dose adjusted regimens with intravenous unfractionated heparin or heparinoid.160 The Rapid Anticoagulation Prevents Ischaemic Damage (RAPID) trial in 67 patients with non-lacunar ischaemic stroke within 12 h of onset compared full doses of intravenous unfractionated heparin with aspirin 300 mg daily; showed no evidence of an advantage of full-dose intravenous heparin over aspirin,176 although a more recent RCT of full-dose intravenous heparin vs low-dose subcutaneous heparin within 3 h of stroke onset suggested net benefit from the higher heparin dose.177 Hence, since bleeding is dose-related, if anticoagulants are to be used, low-dose subcutaneous regimens are preferable (e.g. 5000 IU unfractionated heparin two or three times daily) since they are simpler, do not require complex monitoring and are associated with lower bleeding risks. The options for the treatment of established deep venous thrombosis and pulmonary embolism are discussed in section 11.13. Adverse effects Table 12.10 lists the most important adverse effects and cautions to be considered when using heparin. The most life-threatening risks are of massive intra- or extracranial haemorrhage. Management of severe extracranial haemorrhage consists of stopping any heparin administration; estimation of the clotting time; and reversal with intravenous protamine sulphate in a dose of 1 mg for every 100 IU of heparin infused in the previous hour.193 Protamine should be given slowly over 10 min and the patient monitored for hypotension and bradycardia. Not more than 40 mg of protamine should be administered in any one injection. For advice on reversal of low-molecular-weight heparin or heparinoids, consult the manufacturer’s data sheet. Patients on oral anticoagulants who suffer a head injury or develop any of the neurological symptoms listed in Table 12.11, or who require a diagnostic lumbar puncture, should be investigated and treated as described in Tables 12.11–12.13.193

Table 12.10 Adverse effects of heparin. Local minor complications of subcutaneous heparin at injection site Discomfort Bruising Local complications of intravenous heparin at cannula site (or elsewhere) Pain at cannula site Infection at cannula site (sometimes with severe systemic infection) Reduced patient mobility because of infusion lines and pump Systemic complications Intracranial bleeding Haemorrhagic transformation of cerebral infarct (potentially disabling or fatal) Intracerebral haemorrhage Subarachnoid haemorrhage Subdural haematoma Extracranial haemorrhage Subcutaneous (can sometimes be massive) Visceral (haematemesis, melaena, haematuria) Thrombocytopenia Type I: dose and duration related, reversible, mild, usually asymptomatic, not serious and often resolves spontaneously Type II: idiosyncratic, allergic, severe (may be complicated by arterial and venous thrombosis). Affects 3–11% of patients treated with intravenous heparin and less than 1% of patients treated with subcutaneous heparin Osteoporosis Skin necrosis Alopecia

Reversal of warfarin must be done in consultation with the local haematology specialist (Table 12.13).

12.4.7 Acute stroke in a patient already receiving anticoagulants Urgent reversal of anticoagulation in patients with intracranial bleeding on anticoagulants If a patient on oral anticoagulants presents with symptoms suggestive of a stroke or develops any of the neurological symptoms outlined in Table 12.11 (any of which may indicate intracranial bleeding), they must have a CT or MR scan performed immediately.193 If the CT or MR scan shows extradural, subdural, subarachnoid or intracerebral haemorrhage, then anticoagulants should be stopped and urgently reversed. If the scan does not show intracranial bleeding, consider the possibility that the neurological deficit might be due to a spinal haematoma, especially if the neurological deficit spares the face and is associated with neck pain (the diagnosis is best

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 673

12.4 Selective use: anticoagulants Table 12.11 Advice on management of problems in patients on anticoagulants. Indication for immediate CT or MR brain scan if patient hasa Significant head injury (i.e. any loss of consciousness or amnesia or the presence of a scalp laceration) Headache (especially of recent onset, or increasing in severity) Acute neck pain with neurological deficit in limbsb Drowsiness Confusion Onset of focal neurological symptoms or signs If the scan shows intracranial bleeding Reverse anticoagulants (Table 12.12) If scan shows extradural or subdural bleeding or hydrocephalus, consider surgical decompression after reversal If scan shows only subarachnoid bleeding, consider CT or MR angiography to exclude underlying aneurysm Is the patient being considered for a diagnostic or therapeutic procedure involving the spine?c Lumbar puncture Epidural or spinal anaesthesia Contrast myelography aThe

CT or MR scan must be performed as soon after admission as possible. Delay may reduce the potential benefits of reversing the anticoagulants if intracranial bleeding is found.196 bIn patients with acute neck pain with a neurological deficit in the limbs, but no evidence of intracranial bleeding on the brain scan, MR scanning of the spine may be needed to exclude spinal epidural haematoma.194 cLumbar puncture, epidural anaesthesia and myelography in patients on anticoagulants can cause spinal epidural haemorrhage, spinal cord compression and paraplegia. The need for the procedure should be discussed with a neurologist or neurosurgeon before anticoagulants are reversed, since reversal carries a risk of thromboembolism. Alternative investigation (e.g. MR scanning) or alternative methods of anaesthesia may be preferable.

confirmed by MR scanning of the spine); a spinal epidural haematoma may require neurosurgical treatment.194 A narrative review found no reliable evidence from RCTs to select the best method for rapid reversal, but the options are: vitamin K, prothrombin complex concentrates (PCC), recombinant factor VIIa, or fresh frozen plasma (FFP)195 (section 12.2.4) (Tables 12.12–12.13). There is weak evidence from a retrospective nonrandomized observational study in 55 patients with anticoagulant-related intracerebral haemorrhage that haematoma growth within 24 h was significantly lower in patients receiving PCC compared with FFP or vitamin K.196 There was some evidence that achieving a normal INR within 2 h, generally through the use

..

..

Table 12.12 Strategies for reversal of oral anticoagulation.193 Life-threatening haemorrhage (intracranial or major gastrointestinal bleeding) Stop warfarin prescription and Give intravenous vitamin K1 (5 mg, repeated if necessary) and either Intravenous factor IX complex concentrate (also contains factors II and X) (50 IU/kg body weight) plus Factor VII concentrate (50 IU/kg body weight) (if available) or Recombinant factor VIIa or Fresh frozen plasma (15 mL/kg – approximately 1 L for adult) Less severe haemorrhage (e.g. haematuria, epistaxis) Stop warfarin for 1–2 days Give vitamin K1, 0.5–2 mg intravenously or 5–10 mg orally High international normalized ratio (INR) but no haemorrhage Stop warfarin, monitor INR, restart warfarin when INR < 5.0

of PCC, was associated with a better outcome.196 RCTs of the different options are clearly needed.195,196 After successful reversal of the haemostatic deficit, further treatment may be needed, depending on the location of the haematoma and any underlying causative lesion, for example: evacuation of a subdural haematoma, clipping or coiling of a ruptured intracranial aneurysm, or evacuation of an intracerebral or spinal haematoma. Risk of arterial thromboembolism after stopping anticoagulants Patients with a high risk of arterial thromboembolism (usually because they either have a mechanical prosthetic heart valve or AF) who develop intracranial bleeding while on oral anticoagulants create particular management difficulties.197–199 There are competing risks to be balanced: of valve thrombosis and re-embolization if anticoagulants are permanently withdrawn, and of further intracranial bleeding if anticoagulants are reinstituted.197–201 A retrospective non-randomized observational study of 141 patients with a high risk of ischaemic stroke who had an intracranial haemorrhage while taking warfarin examined these competing risks.198 The indication for anticoagulants fell into one of three categories: prosthetic heart valve in 52 patients, AF in 53 and a recurrent TIA or an ischaemic stroke in 36. The median time off warfarin after the bleed was 10 days. Three patients had an ischaemic stroke within 30 days of stopping warfarin. In the three groups, the Kaplan-Meier estimate of the probability of having an ischaemic stroke within 30 days of stopping warfarin was: prosthetic heart valve 2.9% (95% CI 0–8.0%); AF 2.6% (95% CI 0–7.6%); and TIA or an ischaemic stroke 4.8% (95% CI 0–13.6%)

673

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 674

674

Chapter 12 Specific treatments for acute ischaemic stroke Table 12.13 Options for reversal of warfarin therapy.193 Method and dosage

Advantages

Disadvantages

Stop warfarin

Simple

Vitamin K1 0.5–5 mg i.v. 5–10 mg orally Factor IX complex concentrate 50 IU/kg body weight

Safe if given slowly

May take several days for INR to normalize, particularly with liver disease 2–6 h to take effect 12–24 h to take effect

Factor VII concentrate 50 IU/kg body weight Recombinant factor VIIa Fresh frozen plasma 1L for adult

Should be given with factor IX complex concentrate No risk of blood-borne infection Acts immediately

Acts immediately Small volume Can be given quickly

respectively. Another retrospective review reported that none of the 16 patients with a prosthetic heart valve who had intracranial haemorrhage and then had their warfarin discontinued (for a median of 7 days) suffered a recurrent thromboembolic event while off warfarin.201 It therefore seems from these very limited data that the risk of recurrent thromboembolism from stopping oral anticoagulants for a week or two after an intracranial haemorrhage is low. Is it possible to restart anticoagulation after intracranial bleeding, and if so, in whom, and when? After an intracerebral haemorrhage on anticoagulants, one needs to estimate the risk of recurrent thromboembolism (section 16.6.2 for risk prediction in patients with AF; assessing the risk in patients with mechanical prosthetic heart valves requires expert cardiological advice). In the study of 141 patients by Phan et al., of the 35 who had warfarin therapy restarted, none had recurrence of intracranial bleeding during the same hospital admission.198 Butler has reported 2-year followup data on 13 patients with mechanical prosthetic heart valves who restarted anticoagulation after an intracranial haemorrhage. Of the four patients with intracerebral haemorrhage as the initial bleed, none suffered recurrent intracranial bleeding and two suffered thromboembolic events. On reintroduction of oral anticoagulants, the target INR was lowered in 9 of 13 patients. The authors concluded that careful reintroduction of oral anticoagulation is appropriate in these patients.

Exposes patients to blood product which rarely may transmit hepatitis B (but safe from hepatitis C and HIV) Effect is temporary (8–24 h), may need to be repeated Should be combined with vitamin K1 Risk of thromboembolism As for factor IX complex Expense Large intravenous volume load Allergic reactions Not as efficacious as factor IX complex concentrate Difficult to give repeat infusions because of large volume Virally inactivated plasma is preferred

However, there are many factors one has to consider in decision-making for an individual patient. Consider the (fairly typical) case of an elderly patient in AF who has been on anticoagulants for a number of years because of a prior ischaemic stroke who presents with an intracerebral haemorrhage. A number of questions immediately spring to mind: was the bleed due to over-anticoagulation (and if so, is there a level of international normalized ratio (INR) which clearly separates ‘excessive’ from ‘acceptable’)?; if the INR was in the therapeutic range, how far should investigations to identify the cause be pursued?; and, is the patient still at high risk from recurrent thrombembolism (and if so, should the target INR be lowered)? If the scan shows infarction, what are the management options? If the CT or MR is normal or shows infarction with no evidence of any haemorrhage, the reason for the infarction must be sought. Often, the cause is inadequate warfarin dose, but infective endocarditis must be ruled out. If the INR was below target range before – or at the time of onset of – the stroke, more careful supervision to maintain the INR within the target range for that patient should reduce the risk of further ischaemic strokes. If INR control had been poor despite apparent compliance with the prescribed warfarin dose, one must make careful enquiry about recent changes in other prescribed drug treatment, the use of non-prescribed medication (including herbal and complementary therapies) or

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 675

12.5 Selective use: reperfusion with thrombolytic drugs and other methods

changes in diet, since all of these can interact with warfarin. However, if a patient suffers a recurrent ischaemic stroke despite an adequate INR, one can consider adding low-dose aspirin; while this may reduce the risk of recurrent stroke, it also increases the risk of intracranial haemorrhage171,202 (section 16.6.2).

12.5 Selective use: reperfusion with thrombolytic drugs and other methods

onset of confirmed ischaemic stroke. Individual patient data meta-analyses of the rt-PA203 and streptokinase trials204 have also been published. Since then a number of small-scale RCTs have been published: a dose-finding study of intravenous tenecteplase;205 two phase II studies of intravenous desmoteplase;52,53,206 and, of intra-arterial urokinase in acute posterior circulation stroke.207 Intravenous glycoprotein (Gp) IIb/IIIa inhibitors are discussed in section 12.6.1. There have not been any large-scale trials of mechanical clot retrieval or ultrasound-assisted reperfusion; the limited data on these interventions are reviewed in section 12.6.3. Fewer data than for acute myocardial infarction

12.5.1 Rationale Potential benefits Restoration of blood flow, to reperfuse the ischaemic brain as soon as possible after a cerebral vessel has been occluded, should lessen the volume of brain damaged by ischaemia, reduce the likelihood of major cerebral oedema and result in a better clinical outcome (section 12.1.3). Therefore, therapeutic attempts to hasten reperfusion by removing any occluding thrombus with thrombolytic drugs, mechanical methods, ultrasound or a combination of these approaches ought to be helpful. Potential risks Thrombolytic drugs will also lyse haemostatic plugs and thus may increase bleeding into the brain in the area of ischaemia, in areas of the brain remote from the ischaemia or in extracranial sites (into the skin, joints, the gastrointestinal or urinary systems). Mechanical methods require intra-arterial instrumentation, which can cause a number of complications, including arterial dissection and rupture.

12.5.2 Evidence Data available for reperfusion with thrombolytic drugs and mechanical methods Eighteen RCTs including 5727 patients are included in the most up-to-date systematic review available in the Cochrane Library.133 About half the data relate to trials of intravenous recombinant tissue plasminogen activator (rt-PA) and the remainder to intravenous streptokinase (42%), intravenous urokinase (3%) and intra-arterial recombinant pro-urokinase (5%). Ninety per cent of the data are derived from trials of intravenous thrombolysis with either rt-PA or streptokinase, given within 6 h of

..

..

Over 58 000 patients have been randomized in trials of intravenous thrombolysis vs control in patients with acute myocardial infarction.208 Compared with this very large amount of data, there are relatively few on the effects of thrombolysis for ischaemic stroke (especially when one recalls the size of the global burden of stroke and that the incidence of acute stroke and acute MI are similar in Western populations). Furthermore, it required a systematic review of these data to establish that, though early treatment was the most effective, for patients with ST elevation or bundle branch block, the ‘time window’ for worthwhile benefit was up to 12 h (certainly much longer than had originally been thought plausible).208 Intracranial haemorrhage There is a very consistent trend across all the stroke trials for an excess of fatal intracranial haemorrhage with thrombolysis and the proportional excess is similar with all the agents. The absolute excess depends on the risk of haemorrhage in the control group, and this risk was somewhat different between trials. For intravenous rt-PA, the only licensed thrombolytic agent, treatment increases the risk of fatal intracranial haemorrhage about fourfold from 1% in controls to about 4% (P < 0.00001), equivalent to 33 extra fatal intracranial haemorrhages per 1000 patients treated (Fig. 12.13). rt-PA increases fatal or non-fatal symptomatic intracranial haemorrhages threefold (P < 0.00001).133 Many of the haemorrhages occur within the ischaemic area, but others occur at sites remote from the ischaemia. These remote bleeds may represent bleeding from an undetected pathological vascular lesion (small arteriovenous malformation, vessel affected by amyloid angiopathy, or a previous microbleed).209 However, data are also accumulating to suggest that the presence of microbleeds does not materially influence the risk of bleeding with thrombolysis.210

675

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 676

676

Chapter 12 Specific treatments for acute ischaemic stroke

Thrombolysis n /N Intravenous urokinase vs control Atarashi 1985 Chinese UK 2003 Subtotal (95% CI)

Control n /N

1/192 8/317

0/ 94 0/148

Intravenous streptokinase vs control MAST-E 1996 26/156 MAST-I 1995 8 /157 Morris 1995 2 /10 ASK 1996 14 /174 Subtotal (95% CI)

2 /154 0 /156 0/10 2 /166

Intravenous rt-PA vs control Haley 1993 ECASS 1995 NINDS 1995 ECASS II 1998 ATLANTIS B 1999 ATLANTIS A 2000 Subtotal (95% CI)

1/13 7/ 307 1/312 4 / 391 1/ 306 0/ 71

0/14 19/313 9/ 312 18/409 8/307 8/ 71

Intravenous streptokinase + oral aspirin vs oral aspirin MAST-I 1995 13/156 Subtotal (95% CI)

Odds ratio & 95% CI

Odds reduction

2 /153

Intra-arterial pro-urokinase + intravenous heparin vs intravenous heparin PROACT 1998 1/14 1/ 26 Subtotal (95% CI) –343%* (2P 1.4)

History of bleeding disorder (e.g. haemophilia) History of recent bleeding Arterial puncture at non-compressible site within past 14 days Major surgery within past 14 days Previous stroke or serious head injury within past 3 months Neurological deficit trivial and likely to recover within next few hours Evidence of accelerated hypertension Blood glucose is < 3 or > 22 mmol/L Seek expert advice before giving thrombolysis if any of the following are true:c Unruptured intracranial aneurysm Menstruation Pregnancy Decision to give thrombolysis or not must be made by senior member of stroke team If decision is to give thrombolysis, transfer to place where thrombolysis can be given and monitored closely: Establish intravenous (i.v.) infusion if not already done so Estimate weight, draw up infusion with a dose of 0.9 mg/kg rt-PA Infuse rt-PA i.v.: 10% as a bolus over 1–2 min, remainder over 60 min Monitor pulse and BP: Every 15 min during infusion Every 30 min for next 2 h, then Hourly for 5 h Neurological observations hourly Postacute management: Transfer to stroke unit (if not already there) Start aspirin 160–300 mg daily 24 h after infusion: If able to swallow safely, by mouth If not able to swallow safely, per rectum or i.v. (section 12.3.6)

a

See www.ist3.com for useful documents, protocols and training resources for thrombolysis in stroke. This list of contraindications is based on the UK summary of product characteristics, i.e. the European licence for the use of alteplase for stroke. In the IST-3 trial, the protocol states that ‘If the patient has had a stroke within the previous 14 days or has had treatment for acute ischaemic stroke with thrombolytic therapy within the past 14 days’ this is an absolute contraindication to treatment; patients with a history of an ischaemic stroke more than 2 weeks previously (with or without thrombolytic treatment) may – at the treating physician’s discretion – be considered for treatment or inclusion in the trial. c See section 12.5.4 for details. INR, international normalized ratio; rt-PA, tissue plaminogen activator. b

physicians, neurologists and stroke physicians from giving the treatment.149,246 Although clinicians worry that intracranial bleeding due to treatment may lead to litigation, in practice, failure to administer the treatment in an otherwise eligible patient is the more common reason for litigation, in North America at least.246 However, when asked, many people (who have not yet had a stroke) would be prepared to accept the short-term risks (of severe haemorrhage or death) in order to have the chance of surviving free of disability.247,248 A patient’s willingness

..

..

to accept the risks of treatment is often influenced by their own perception of the severity of the stroke. If the patient has neglect or other perceptual difficulties, this can further complicate the consent procedure.249 A patient presented with witnessed abrupt onset of complete left hemiplegia, 1 h before arrival at the hospital, had a CT consistent with early ischaemic change in the right cerebral hemisphere, and met all the criteria for thrombolytic treatment within the current approval. However, because

683

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 684

684

Chapter 12 Specific treatments for acute ischaemic stroke

she also had anosognosia, she was completely unable to recognize that she had had a stroke, and hence saw no reason to be exposed to the risks associated with thrombolysis. The stroke team felt it inappropriate to treat her against her express wishes. The scenario above is drawn from a published case report;249 a comment on the case stated that it would have been reasonable to give the treatment without the patient’s consent.250 This is supported by the fact that the patient in question was left with substantial residual neurological impairment requiring major assistance in all activities of daily living.249 The risks of thrombolysis are definite and not trivial, so fully informed consent is desirable, whether thrombolysis is to be used in routine clinical practice or in an RCT. But, in practice, getting fully informed consent is often a barrier to early treatment:251 the patient’s acute neurological impairment may impair communication and writing; a close relative or partner may not be available at the critical moment; and the pressure of time may make decision-making difficult for all concerned.252–257 However, one must seek to achieve whatever degree of consent is required to meet local clinical governance and research practice ethical standards. Simple materials (adaptable for routine clinical or trial use) have been developed with input from patients and lay people, that can facilitate the process of consent (see www.ist3.com).248,257 However, the rigorous standards for informed consent required in research do not obtain when thrombolysis is used in routine practice. A survey of the documentation of consent for routine clinical use of thrombolysis for stroke in Connecticut hospitals found that a substantial percentage of patients who received rt-PA for stroke had no consent documented.258 Surrogates often provided consent when the patients had the capacity to give consent; conversely, patients with diminished capacity sometimes provided their own consent.258 There should be no double standards for randomized trials and clinical practice, yet patients given treatments in routine practice are often not given any explanation or information. Many patients refused randomization in the first International Stroke Trial when they were told that aspirin and heparin might cause intracranial bleeding, yet how many tens of thousands of patients with acute stroke have been treated with these two drugs (before and since the large-scale trials began) without being asked for their consent after a full explanation of the risks and benefits? Similar double standards seem to apply to thrombolysis in randomized trials and routine clinical practice.

Treatment The aim should be to treat – without delay – patients who meet all of the following criteria: • the time of symptom onset is known with certainty; • the clinical diagnosis is definitely an acute stroke of sufficient severity to merit the risks of treatment; • no clear clinical or radiological contraindications are identified; • consent (or relative’s assent) is documented in accordance with local requirements; • passage through the above system has been fast enough to allow treatment well within 3 h of onset; • there is a vacant bed and sufficient staff available on the acute stroke unit to monitor the patient during the infusion and the immediate post-treatment period. Patients who fulfil the above criteria, but do not exactly meet the strict terms of the current approval, and for whom thrombolysis is considered ‘promising but unproven’ should be considered for inclusion in IST-3 or one of the other relevant RCTs. The urgent priorities are to establish, by means of appropriate large-scale randomized trials, whether the balance between early hazard and long-term benefit is still favourable when thrombolysis is used in a wider variety of patients. These data will help decide which categories of patient are most likely to benefit and whether the benefits really persist for more than just a few months.

12.5.4 Who not to treat The absolute contraindications to thrombolysis are listed in Table 12.14. In addition, there are many relative contraindications reported in the literature, but which are not supported by good evidence. There are case reports of thrombolysis being used for the treatment of acute ischaemic stroke in patients who do have these relative contraindications, including: pregnancy;259 menstruation;260 recent surgery;261 unruptured intracranial aneurysms262 and anticoagulation with an INR > 1.7.263,264 We are aware of cases with other contraindications being treated (remote past history of intracranial haemorrhage but current ischaemic stroke, history of ischaemic stroke within past 3 months, etc.); in our view, it would be more informative to consider including patients who have some of these features of uncertain significance in relevant RCTs in order to determine reliably whether or not they should influence patient selection.

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 685

12.5 Selective use: reperfusion with thrombolytic drugs and other methods

The decision about treatment often requires a synthesis of clinical and radiological information. A 70-year-old man who lived alone was brought by ambulance to the emergency department at 9 am, following an emergency call at 8 am. He had aphasia and profound weakness of his right arm. The CT scan showed an area of ischaemia in the left middle cerebral artery territory appropriate to his symptoms. However, the lesion showed marked hypoattenuation and quite well-defined edges, suggesting it had been present for much longer than the 1 h suggested by the ambulance crew. Thrombolysis was not given. The patient’s niece later confirmed that her uncle’s speech had been ‘muddled’ when she had telephoned him at 10 pm the evening before, suggesting his ischaemic stroke lesion had been present for at least 11 h by the time he presented to hospital. At present, age over 80 is a contraindication to treatment within the European approval, whereas it is not in the USA and Canada. The risk of intracranial bleeding with rt-PA rises with age, but so too does the risk of a poor outcome without treatment.203,265 Hence, given the very limited published RCT data in this age group (just 42 patients), while we accept that such patients can be treated,219 we often prefer to enrol them in IST-3. We do this, because we – and the patients – are usually genuinely uncertain of the balance of risks and benefits.

12.5.5 When to start treatment – as soon as possible! Since the benefit (but perhaps not the risk) of treatment is very time dependent, the sooner it can be given, the better (whether as part of routine clinical practice, or in a randomized trial). A systems approach is needed to ensure every step in the process of care is completed as efficiently as possible; from the moment the patient develops their stroke symptoms to the safe completion of the treatment in hospital. There are many barriers to the delivery of efficient care, but a variety of interventions to overcome these barriers have been evaluated (section 17.5).214,251,266 Many centres have found it useful to create a ‘stroke thrombolysis box’ (Fig. 12.17). The design and the selection of its contents will vary according to local circumstances, but it should be portable, contain everything needed to carry out a rapid assessment of a potential stroke thrombolysis patient and then initiate therapy (i.e. all the information, printed protocols, drugs, specimen bottles, request forms, intravenous infusion pumps, drug prescription, drug administration charts, clinical monitoring records, etc.). If everything is in one place, valuable time can be saved.

..

..

Fig. 12.17 ‘Thrombolysis box’ used on the acute stroke unit at the Western General Hospital, Edinburgh. The top drawer contains the infusion pump and written protocols (on laminated sheets) for nursing management, monitoring, and management of suspected complications. The lower drawers contain the other items required for drug administration. Photograph by Dr I Kane (reproduced with permission).

12.5.6 Agent/dose/route/concomitant treatment Agent Intravenous rt-PA is the only agent that has wide regulatory approval for use in stroke. In Europe, at the time of writing, the approval was provisional on two conditions being met: • Satisfactory evidence of the safety of rt-PA in routine practice. Up to the end of April 2006 clinicians were required to record (in the SITS-MOST registry: http:// www.acutestroke.org/) the use of, and the clinical outcome in, patients treated with rt-PA when treated within the terms of the approval. After that date, registering the data has been voluntary. • Completion of the ECASS-3 RCT.267

685

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 686

686

Chapter 12 Specific treatments for acute ischaemic stroke

Dose and route The approved regimen is 0.9 mg/kg of rt-PA intravenously; 10% of the dose as a bolus over 1 min and the remainder as an infusion over 1 h, and the maximum dose is 90 mg.133,230 Intra-arterial thrombolysis does not have regulatory approval, but a variety of regimens have been used: in the PROACT-II trial, patients were given 9 mg of recombinant pro-urokinase intra-arterially;234 in AUST, intraarterial urokinase was given in increments of 100 000 IU to a maximum dose of 1 000 000 IU; and, in a recent large case-series, up to 1 000 000 IU of urokinase was given, and if recanalization was not achieved at that dose, gentle mechanical disruption of the clot with the catheter tip was attempted.268 Intra-arterial thrombolysis has a number of potential disadvantages: delay in initiation of fibrinolysis while the diagnostic angiogram is performed and the microcatheter positioned (start of the thrombolytic infusion typically occurs 50–90 min later than the start of an intravenous infusion); the procedure is labour- and capital-intensive; and the intervention can only be performed at tertiary and secondary hospitals capable of acute endovascular therapy.231 Concomitant antithrombotic treatment Current US and UK guidelines recommend that antithrombotic therapy with aspirin or heparin should be avoided for at least 24 h after intravenous thrombolysis with rt-PA.166,167 The situation is less clear for patients who have had intra-arterial thrombolysis. Some centres follow one of the anticoagulant regimes employed in the RCTs of intra-arterial thrombolysis,207,233,234 whereas others – in the light of the results of the CAST and IST trials – use aspirin instead of heparin as the post-treatment antithrombotic regimen.268 Blood pressure management for patients having thrombolytic treatment The North American Stroke Thrombolysis Guidelines235 recommended the NINDS trial treatment algorithm,132 but the more recent version on acute stroke did not make such firm recommendations167 (section 11.7 for general guidance on blood pressure monitoring and management).

12.5.7 Adverse effects Intracranial and major extracranial haemorrhage are the most feared adverse effects of thrombolytic treatment.

All treatment protocols should have clear arrangements on how to monitor for, and then treat, these complications if they arise. Examples of the sort of monitoring and management guides (which can be adapted as needed) are available at the IST-3 website (www.ist3.com); they can be laminated and kept in the ‘thrombolysis box’. Tables 12.15 and 12.16 outline advice on the management of suspected intracranial or extracranial haemorrhage.

Table 12.15 Advice on management of suspected intracranial bleeding in patients treated with rt-PA. Suspect intracranial bleeding if Neurological deterioration New headache Fall in conscious level Acute hypertension Seizure Nausea or vomiting Initial actions Stop infusion of rt-PA Arrange an urgent CT scan Check fibrinogen, PT, APTT, full blood count and send blood for ‘group and save’ Support circulation with i.v. fluids if needed If the scan shows intracranial bleeding If scan shows extradural or subdural bleeding or hydrocephalus, consider surgical decompression after reversal of any haemostatic defcit If scan shows only subarachnoid bleeding, consider CT or MR angiography to exclude underlying aneurysm If the scan shows no intracranial bleeding If no intracranial bleeding on CT or MR, consider other causes of deterioration If the patient has a neurological deficit consistent with a spinal location, sparing the face, especially if there is acute neck or back pain, consider MR scan of spine to exclude epidural haematoma The manufacturers of rt-PA suggest that most patients with severe bleeding can be managed with volume replacement.245 It is rarely necessary to replace the clotting factors because of the short half-life of the drug and the minimal effect on the systemic coagulation factors. In those who fail to respond, transfusion of cryoprecipitate, fresh frozen plasma and platelets should be considered; the British Society of Haematology guidelines for the use of thrombolysis suggest for severe life-threatening bleeding after rt-PA, that has not responded to initial measures, giving a fibrinolytic inhibitor such as aprotinin or tranexamic acid and replacement of clotting factors depending on the results of a coagulation screen (see also Table 12.16).318

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 687

12.6 Unproven value: other reperfusion strategies Table 12.16 Advice on management of suspected extracranial bleeding in patients treated with rt-PA. Suspect extracranial bleeding if Drop in blood pressure Clinical shock Evidence of blood loss, e.g. melaena, haematuria Initial actions Stop infusion of rt-PA Use mechanical compression, if possible, to control bleeding Check fibrinogen, PT, APTT, full blood count and send blood for ‘group and save’ or cross-match depending on situation Support circulation with fluids and blood transfusion as appropriate Discuss results with local haematology department For severe life-threatening bleeding not responding to fluid replacement Consider giving a fibrinolytic inhibitor (depending on the results of a coagulation screen), either: 500 000 kallikrein inactivator units of aprotinin i.v. over 10 min followed by 200 000 units over 4 h or tranexamic acid i.v. 1 g over 15 min repeated every 8 h as necessary or fresh frozen plasma and/or cryoprecipitate Delay any surgery until haemostatic deficit corrected

Tissue plasminogen activator can cause anaphylactoid reactions which may require prompt supportive treatment: adrenaline (0.5–1 mL 1 in 1000 i.m. or s.c., but not i.v.) – the dose depends on the severity of the reaction, and it may be necessary to repeat the dose; hydrocortisone 200 mg i.v.; chlorpheniramine 10 mg i.v.; salbutamol nebulizer 5 mg; and fluid resuscitation if shocked. It is important to be aware of an uncommon anaphylactoid reaction, orolingual angioneurotic oedema, which occurs in about 2–5% of patients treated with rt-PA, and is more likely if the patient is on an ACE inhibitor.271,272 It can become life-threatening if the rapid tongue swelling causes upper airway obstruction, which may then require endotracheal intubation.272

12.6 Unproven value: other reperfusion strategies

12.6.1 Gp IIb/IIIa inhibitors Rationale

The manufacturers of rt-PA suggest that most patients with severe bleeding can be managed with volume replacement.245 It is rarely necessary to replace the clotting factors because of the short half-life of the drug and the minimal effect on the systemic coagulation factors. In those who fail to respond, transfusion of cryoprecipitate, fresh frozen plasma and platelets should be considered; the British Society of Haematology guidelines for the use of thrombolysis suggest for severe life-threatening bleeding after rt-PA, that has not responded to initial measures, giving a fibrinolytic inhibitor such as aprotinin or tranexamic acid and replacement of clotting factors depending on the results of a coagulation screen.318 a The options listed are based on expert opinion and not on reliable RCT evidence; it is prudent to discuss the options with your local haematologist.

If a patient develops intracranial bleeding, early guidelines suggested consideration of surgical evacuation of any intracranial clot.235 This may not be wise because the STICH trial and a systematic review have not been able to demonstrate net benefit from evacuation of intracerebral haematoma.269,270 However, if the patient has subdural or extradural bleeding but no major intracerebral haematoma, surgical drainage may be indicated. Surgery for any type of intracranial haemorrhage should be delayed until the haemostatic deficit is fully reversed.

..

..

The platelet glycoprotein (Gp) IIb/IIIa receptor is the final common pathway of platelet aggregation. Agents which block this receptor have antithrombotic and thrombolytic actions; in patients undergoing primary coronary stenting for the treatment of acute myocardial infarction, the Gp IIb/IIIa inhibitor abciximab improved coronary patency before stenting, the success of the stenting procedure, coronary patency at 6 months, left ventricular function and clinical outcomes.273 A meta-analysis showed that, in patients with acute myocardial infarction, adjunctive abciximab significantly reduced 30-day and long-term mortality in patients treated with primary angioplasty but not in those receiving fibrinolysis.274 Abciximab and a number of other similar agents are therefore now being tested in acute ischaemic stroke. Evidence Abciximab is the agent of this class that has been most widely tested as a treatment for acute ischaemic stroke. A 400-patient phase II study showed promising results,275 so an 1800 patient efficacy RCT was initiated (AbESTT-2), but it was halted prematurely after 808 patients had been enrolled, because of an unfavourable risk : benefit ratio (largely a higher than anticipated risk of intracranial haemorrhage).276 In the light of the AbESTT-2 trial results, abciximab is unlikely to become an established treatment

687

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 688

688

Chapter 12 Specific treatments for acute ischaemic stroke

for acute stroke.276 The evidence for the use of tirofiban in acute stroke is more limited.277 Two Gp IIb/IIIa inhibitors, eptifibatide and tirofiban, are still being evaluated, either alone, or in combination with thrombolytic agents, in ongoing small-scale RCTs, and it is possible that one of these will prove effective in acute stroke. There are a large number of uncontrolled case series of intra-arterial Gp IIb/IIIa inhibitors (most relating to abciximab or tirofiban) for ‘rescue’ treatment for the cerebral embolic complications of percutaneous coronary interventions278 and interventional neuroradiology procedures (e.g. coiling of intracranial aneurysms,279 carotid angioplasty and stenting280,281). A retrospective review of 1373 consecutive patients who underwent neuroendovascular procedures identified 29 (2%) where the procedure was complicated by acute cerebral thromboembolic events and were then treated with abciximab.282 Angiographic improvement was achieved in 29 (81%) of 36 arteries. Three intracerebral haemorrhages (10%) occurred with abciximab when administered with concurrent mechanical clot disruption; in two of these haemorrhages, recombinant tissue plasminogen activator was also administered.282 A non-randomized study comparing abciximab with emboli prevention devices suggested the use of emboli prevention devices may be safer.281

12.6.2 Defibrinogenation Rationale There are several defibrinogenating agents. The most widely tested is ancrod, a serine protease derived from the venom of the Malayan pit viper. It causes a fall in the levels of plasma fibrinogen, plasminogen, plasminogen activator inhibitor and antiplasmin.283 Large quantities of circulating fibrinogen and fibrin degradation products are generated and tissue plasminogen activator is released from the vascular endothelium. Ancrod also reduces plasma and whole-blood viscosity.283 Defibrinogenation might therefore improve perfusion in the ischaemic brain and so have a net beneficial effect, provided it is not associated with a substantial excess of major bleeding. Evidence A Cochrane systematic review of defibrinogenating agents in acute ischaemic stroke included five trials involving 2926 patients.284 Four trials tested ancrod and one trial defibrase. The reviewers concluded that, although fibrinogen-depleting agents appeared promising, the data from the ESTAT trial were needed before

more reliable conclusions could be drawn.284 The ESTAT trial has recently been published. It was a randomized placebo-controlled trial of ancrod which included 1222 patients with an acute ischaemic stroke. The primary outcome was functional success at 3 months (survival, Barthel Index of 95 or 100, or return to prestroke level). Functional success at 3 months did not differ between patients given ancrod (42%) and those given placebo (42%) (P = 0.94; OR 0.99, 95% CI 0.76–1.29).285 Conclusion At present, there is not enough evidence to justify the use of defibrinogenating agents in routine clinical practice.

12.6.3 Interventional neuroradiology and ultrasound augmentation of clot lysis Rationale For patients with acute ST-segment elevation myocardial infarction (AMI), primary percutaneous transluminal coronary angioplasty (PTCA) is often used as the primary revascularization method in preference to intravenous thrombolytic therapy. A systematic review of the trials comparing PTCA with thrombolysis in AMI, involving 7739 patients, found that primary PTCA was significantly better than thrombolytic therapy at reducing early death, non-fatal reinfarction, stroke, and the combined outcome cluster of death, non-fatal reinfarction and stroke.286 The advantage to primary PTCA-allocated patients persisted during long-term follow-up, and was independent of both the type of thrombolytic agent used, and whether or not the patient was transferred for primary PTCA.286 In centres that regularly perform neuro-interventional procedures in acute ischaemic stroke, if intravenous or intra-arterial thrombolytic drug therapy does not reopen the occluded artery, a number of techniques can now be tried to restore flow, including: direct mechanical balloon angioplasty of the thrombus; mechanical removal of clot; intravascular stenting of the underlying occlusive atherosclerotic lesion; suction thrombectomy; and laser-assisted thrombolysis of emboli.167,231 Transcranial Doppler ultrasound (with or without supplementary microbubble injection) on its own, or as an adjunct to thrombolytic therapy, may also improve clot lysis.231,287 Evidence The process for approval of the use of devices and procedures for acute stroke treatment is controversial because

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 689

12.7 Unproven value: treatment of cerebral oedema and raised intracranial pressure

devices are not subject to the same stringent approval processes as drugs (i.e. there is no requirement for phase III efficacy trials).288 In acute ischaemic stroke, case series have reported the use of mechanical clot disruption,289 and the use of the MERCI clot retrieval device.290 The FDA has approved the MERCI device for the removal of clot from brain arteries on the basis of a non-randomized study,290 (though not as treatment for acute ischaemic stroke), a decision regarded by some as premature.288,291 The device is, however, being tested in the MR-RESCUE trial.292 Future studies will need to evaluate the overall efficacy when compared with intra-arterial or intravenous treatment, overall safety and the major complication rate (e.g. vessel rupture).288,289,293 The CLOTBUST RCT has shown that, in patients presenting within 3 h of acute ischaemic stroke onset who were treated with intravenous rt-PA, 2 h of continuous transcranial Doppler ultrasound increased the frequency of cerebral arterial recanalization and this was associated with a trend to better clinical outcomes.294 Another phase II ultrasound trial, TRUMBI, is ongoing,287 and CLOTBUST-2, evaluating microbubbleaugmented sono-thrombolysis is planned.231 Conclusion Further RCTs of interventional radiology (preferably on a cardiological scale) will be needed to assess its place in routine clinical practice.288,291,293 As with intra-arterial thrombolysis, the number of centres able to deliver such highly specialized treatments will need to increase dramatically for such trials to become feasible, and enormous investment would be required to minimize the average time to treatment. Ultrasound augmentation of thrombolysis is simpler, and potentially more widely applicable.

12.7 Unproven value: treatment of cerebral oedema and raised intracranial pressure

12.7.1 Corticosteroids Rationale Vasogenic cerebral oedema, which tends to develop about 24–48 h or more after stroke onset, particularly in and around large infarcts, may be reduced by corticosteroids (such as dexamethasone) (section 12.1.5). Cytotoxic oedema, that develops almost immediately after the

..

..

onset of ischaemia, might be reduced by neuroprotective agents which can prevent influx of sodium ions and water into cells (section 12.1.5). Evidence There have been at least 17 apparently randomized trials in acute stroke, but unfortunately only seven (involving 453 people) were of sufficient quality to merit inclusion in a Cochrane systematic review.295 There was no significant difference in the odds of death within 1 year (odds ratio 1.08; 95% CI 0.68–1.72). And nor did treatment appear to improve functional outcome in survivors. The results were inconsistent between individual trials. Conclusion These data do not support the routine or selective use of corticosteroids in acute ischaemic stroke. We do not use corticosteroids as a routine treatment for unselected patients with ischaemic stroke, nor do we use them in patients who have massive cerebral oedema complicated by transtentorial herniation. Our only indication is in patients with inflammatory vascular disorders (e.g. giant-cell arteritis, polyarteritis nodosa, systemic lupus erythematosus; section 7.3).

12.7.2 Glycerol and mannitol Rationale A 10% solution of glycerol is a hyperosmolar agent which is said to reduce cerebral oedema and possibly increase cerebral blood flow, and mannitol is another osmotically active agent. These treatments aim to reduce intracranial pressure by reducing infarct-related oedema and thereby improve perfusion in and around infarcts. Evidence A Cochrane systematic review included 11 randomized trials comparing intravenous glycerol with control (482 glycerol treated patients were compared with 463 control patients).296 Glycerol was associated with a significant reduction in the odds of death during the scheduled treatment period (OR 0.65; 95% CI 0.44–0.97). However, at the end of the scheduled follow-up period, there was no significant difference in the odds of death (OR 0.98; 95% CI 0.73–1.31). Functional outcome was reported in only two studies and there were non-significantly more patients who had a good outcome at the end of scheduled follow-up in the glycerol group (OR 0.73; 95% CI

689

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 690

690

Chapter 12 Specific treatments for acute ischaemic stroke

0.37–1.42). Haemolysis seemed to be the only relevant adverse effect of glycerol treatment. A Cochrane systematic review of mannitol in stroke included three small trials with 226 randomized patients. Neither beneficial nor harmful effects of mannitol could be proved.297 The number of included patients was small, and the follow-up was short. Case fatality, the proportion of dependent patients, and adverse effects were not reported. A similar Cochrane review of mannitol in traumatic head injury found no evidence of benefit in patients with raised intracranial pressure who did not have an operable intracranial haematoma.298 Conclusion These data do not support the routine or selective use of glycerol or mannitol in acute ischaemic stroke, and so we generally do not use them in our own clinical practice (though our colleagues in intensive care may use mannitol in some of our patients with severe elevations of intracranial pressure after ischaemic stroke).

12.7.3 Surgical decompression Posterior fossa decompression or shunting for massive cerebellar infarcts Surgical decompression (or ventricular shunting) of massive cerebellar infarcts may improve cerebral perfusion, relieve any obstructive hydrocephalus and so prevent early death. Decompressive surgery has not been evaluated by adequately designed trials, so it is still controversial as to which patients with cerebellar infarction should have decompressive surgery, shunting or medical therapy.

included in DESTINY,302 DECIMAL303 and HAMLET304 and treated within 48 h after stroke onset were pooled for analysis.321 The protocol for these analyses was designed prospectively when the trials were still recruiting patients and outcomes were defined without knowledge of the results of the individual trials. The primary outcome measure was the score on the modified Rankin scale (mRS) at 1 year dichotomized between favourable (0–4) and unfavourable (5 and death) outcome. A total of 93 patients were included in the pooled analysis. More patients in the decompressive surgery group than in t he control group had an mRS ≤ 4 (75% vs 24%; pooled absolute risk reduction 51% [95% CI 34–69]), an mRS ≤ 3 (43% vs 21%; 23% [95% CI 5–41]), and survived (78% vs 29%; 50% [95% CI 33–67]), indicating numberneeded-to-treat of two for survival with mRS ≤ 4, four for survival with mRS ≤ 3, and two for survival irrespective of functional outcome. The effect of surgery was highly consistent across the three trials. The authors concluded ‘In patients with malignant MCA infarction, decompressive surgery undertaken within 48 h of stroke onset reduces mortality and increases the number of patients with a favourable functional outcome. The decision to perform decompressive surgery should, however, be made on an individual basis in every patient.321 Conclusion: decompressive surgery Even with these results, the decision about whether and when to consider decompressive surgery therefore remains very difficult; leave it too late and you ‘miss the boat’ and irreversible damage occurs. Operate too soon, and one risks worsening the ischaemic brain insult. We do occasionally refer younger patients with no comorbidity and a ‘malignant’ course for surgery, and hope to recruit patients in the HAMLET trial.

Hemicraniotomy for massive supratentorial infarcts Decompressive surgical techniques that attempt to relieve high intracranial pressure due to oedema have been described,299 but their efficacy in reducing case fatality and disability has, until recently, been very uncertain. A Cochrane systematic review did not identify any completed RCTs, five observational studies and some small series and single case reports, so no clear conclusions could be drawn.300 At least five RCTs have since been initiated: HAMLET, HEADDFIRST, DESTINY, HEMMI and DECIMAL, of which three (HEADDFIRST,301 DESTINY302 and DECIMAL303) have been completed (recruitment into HAMLET continues). A prospectively pooled analysis of individual data for patients aged between 18 years and 60 years, with space-occupying MCA infarction,

12.7.4 Hyperventilation Hyperventilation is used to lower intracranial pressure, but there is no randomized evidence on its effects in patients with stroke available on the Cochrane Controlled Trials Register. A Cochrane review of hyperventilation in severe head injury concluded the data available were inadequate to assess any potential benefit or harm that might result from hyperventilation.305 We therefore would not recommend hyperventilation to reduce raised intracranial pressure, though we accept that our colleagues in intensive care may sometimes administer hyperventilation to some of our patients with severe elevations of intracranial pressure after ischaemic stroke.

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 691

12.8 Unproven value: other interventions

12.8 Unproven value: other interventions

12.8.1 Neuroprotective agents Rationale There are many – perhaps too many – points in the pathophysiological cascade between vessel occlusion and irreversible cell death where pharmacological intervention might be beneficial (sections 12.1.3 and 12.1.4), and the pharmaceutical industry has worked through a very large number of compounds for clinical development and testing. There is no generally agreed definition of a neuroprotective drug, but in the context of acute ischaemic stroke, the aim of this class of agents is to limit the volume of brain damaged by ischaemia. There are a number of steps to be taken between first identifying a promising agent during preclinical testing in animals and man before any large-scale clinical trials are mounted.306,307 And systematic reviews of the available animal data offer the prospect of reducing bias in the assessment of preclinical data.308–311 Greater attention to achieving these milestones might increase the chances of a successful clinical development and licensing of a neuroprotective drug than hitherto. Many trials have been completed, but no neuroprotective drug has been licensed for clinical use. The ever-increasing number of compounds withdrawn from clinical development in recent years has been a disappointment. This has certainly tempered the opinions of even the diehard optimists who said, just a few years back, that the neuroprotective drugs would make stroke rehabilitation units redundant. It appeared in early 2006 that an effective agent had been found, when the SAINTI trial of the agent NXY-059 in 1700 patients with an acute ischaemic stroke showed a significant reduction in post-stroke disabilities (modified Rankin Scale) among patients treated with the study drug (P = 0.038).312 Though the full results of a second phase III trial, SAINT2, are awaited, a press release in October 2006 stated that ‘the SAINT-2 trial did not demonstrate a statistically significant reduction on the primary endpoint of strokerelated disability in patients treated with NXY-059, as assessed versus placebo using the modified Rankin Scale (mRS) (P = 0.33, odds ratio 0.94).’313 These disappointing results led the company to discontinue clinical development of the compound. Despite these unpromising results, the situation may change, and it is conceivable that one of the neuroprotective compounds currently undergoing trials may

..

..

gain a licence for use in acute ischaemic stroke in the next few years. It is probable that the extremely large reductions in cerebral infarct volumes achieved with neuroprotective agents in experimental animal models will translate into only moderate reductions, if any, in disability when used in humans; neuroprotective therapy is unlikely to prove a panacea for acute stroke.

12.8.2 Haemodilution Rationale Haemodilution aims to reduce whole-blood viscosity by an amount that increases cerebral blood flow yet does not reduce cerebral oxygen delivery. The hypothesis is that this should reduce ischaemic damage in and around the infarct. Haemodilution is generally performed by giving an infusion of dextran, hydroxyethyl starch or albumin. Such infusions increase blood volume (hypervolaemic haemodilution). In patients with acute stroke in whom an increase in total blood volume may be undesirable, haemodilution can be achieved isovolaemically by simultaneously removing several hundred millilitres of blood. If the optimum haematocrit is achieved, any increase in cerebral oxygen delivery achieved in practice might be neuroprotective and so lessen infarct volume. Evidence A Cochrane systematic review of haemodilution started within 72 h of stroke onset included 18 trials (including 2766 patients). Haemodilution did not significantly reduce deaths within the first 4 weeks (OR 1.09; 95% CI 0.86–1.38). Similarly, haemodilution did not influence deaths within 3–6 months (OR 1.01; 95% CI 0.84– 1.22), or death and dependency or institutionalization (OR 0.98; 95% CI 0.84–1.15). No statistically significant benefits were documented for any particular type of haemodiluting agent. The overall results of this review were compatible both with modest benefit and moderate harm of haemodilution for acute ischaemic stroke, but it has not been proven to improve survival or functional outcome.314 Conclusion Given this information, we can see no reason to use this treatment in routine clinical practice, although it is used

691

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 692

692

Chapter 12 Specific treatments for acute ischaemic stroke

quite widely in Austria and in some parts of Eastern Europe.

12.8.3 Calcium antagonists Rationale The influx of calcium through voltage-sensitive channels has a role in neuronal death from ischaemia; this influx can be inhibited by a number of agents, so-called calcium channel blockers or antagonists (section 12.1.4). Evidence A systematic review included 29 RCTs (7665 patients).315 No effect of calcium antagonists on poor outcome at the end of follow-up (relative risk, 1.04; 95% CI 0.98–1.09) or on death at end of follow-up (relative risk, 1.07; 95% CI 0.98–1.17) was found. Indirect comparisons of the effects of intravenous and oral administration showed an increase in the number of patients with poor outcome with intravenous administration. Comparisons of different doses of nimodipine suggested that the highest doses were associated with poorer outcome. Administration within 12 h of onset was associated with an increase in the proportion of patients with poor outcome, but this effect was largely due to the poor results associated with intravenous administration. Conclusion The trials do not provide any evidence to support the routine use of oral calcium antagonists in patients presenting within 12 h, or intravenous administration. We do not routinely use calcium antagonists in acute ischaemic stroke patients.

References 1 Sharp FR, Lu A, Tang Y, Millhorn DE. Multiple molecular penumbras after focal cerebral ischemia. J Cereb Blood Flow Metab 2000; 20(7):1011–32. 2 Arundine M, Tymianski M. Molecular mechanisms of calcium-dependent neurodegeneration in excitotoxicity. Cell Calcium 2003; 34(4–5):325–37. 3 Baron JC. Mapping the ischaemic penumbra with PET: implications for acute stroke treatment. Cerebrovasc Dis 1999; 9(4):193–201. 4 Siesjo BK. Pathophysiology and treatment of focal cerebral ischemia. Part I: Pathophysiology. J Neurosurg 1992; 77(2):169–84.

5 Siesjo BK. Pathophysiology and treatment of focal cerebral ischemia. Part II: Mechanisms of damage and treatment. J Neurosurg 1992; 77(3):337–54. 6 Kristian T, Siesjo BK. Calcium in ischemic cell death. Stroke 1998; 29(3):705–18. 7 Pellerin L, Magistretti PJ. Neuroenergetics: calling upon astrocytes to satisfy hungry neurons. Neuroscientist 2004; 10(1):53–62. 8 Magistretti PJ. Neuron-glia metabolic coupling and plasticity. J Exp Biol 2006; 209(Pt 12):2304–11. 9 Bouzier-Sore AK, Voisin P, Canioni P, Magistretti PJ, Pellerin L. Lactate is a preferential oxidative energy substrate over glucose for neurons in culture. J Cereb Blood Flow Metab 2003; 23(11):1298–306. 10 Leenders KL, Perani D, Lammertsma AA, Heather JD, Buckingham P, Healy MJ et al. Cerebral blood flow, blood volume and oxygen utilization: normal values and effect of age. Brain 1990; 113 ( Pt 1):27–47. 11 Baron JC. Mapping the ischaemic penumbra with PET: a new approach. Brain 2001; 124(Pt 1):2–4. 12 Guadagno JV, Warburton EA, Aigbirhio FI, Smielewski P, Fryer TD, Harding S et al. Does the acute diffusionweighted imaging lesion represent penumbra as well as core? A combined quantitative PET/MRI voxel-based study. J Cereb Blood Flow Metab 2004; 24(11):1249–54. 13 Guadagno JV, Donnan GA, Markus R, Gillard JH, Baron JC. Imaging the ischaemic penumbra. Curr Opin Neurol 2004; 17(1):61–7. 14 Baron JC. Stroke research in the modern era: images versus dogmas. Cerebrovasc Dis 2005; 20(3):154–63. 15 Gibbs JM, Wise RJ, Leenders KL, Jones T. Evaluation of cerebral perfusion reserve in patients with carotid-artery occlusion. Lancet 1984; 1(8372):310–14. 16 Warach S, Baron JC. Neuroimaging. Stroke 2004; 35(2):351–3. 17 Warach S, Wardlaw J. Advances in imaging 2005. Stroke 2006; 37(2):297–8. 18 Derdeyn CP. Positron emission tomography imaging of cerebral ischemia. Neuroimaging Clin N Am 2005; 15(2):341–xi. 19 Muir KW, Buchan A, von Kummer R, Rother J, Baron JC. Imaging of acute stroke. Lancet Neurol 2006; 5(9):755–68. 20 Johnston AJ, Steiner LA, Balestreri M, Gupta AK, Menon DK. Hyperoxia and the cerebral hemodynamic responses to moderate hyperventilation. Acta Anaesthesiol Scand 2003; 47(4):391–6. 21 Symon L, Branston NM, Strong AJ. Autoregulation in acute focal ischemia: an experimental study. Stroke 1976; 7(6):547–54. 22 Eames PJ, Blake MJ, Dawson SL, Panerai RB, Potter JF. Dynamic cerebral autoregulation and beat to beat blood pressure control are impaired in acute ischaemic stroke. J Neurol Neurosurg Psychiatry 2002; 72(4):467–72. 23 Dawson SL, Panerai RB, Potter JF. Serial changes in static and dynamic cerebral autoregulation after acute ischaemic stroke. Cerebrovasc Dis 2003; 16(1):69–75. 24 Pulsinelli W. Pathophysiology of acute ischaemic stroke. Lancet 1992; 339(8792):533–6.

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 693

References

25 Yam AT, Lang EW, Lagopoulos J, Yip K, Griffith J, Mudaliar Y et al. Cerebral autoregulation and ageing. J Clin Neurosci 2005; 12(6):643–6. 26 Aaslid R, Lindegaard KF, Sorteberg W, Nornes H. Cerebral autoregulation dynamics in humans. Stroke 1989; 20(1):45–52. 27 Panerai RB, Dawson SL, Eames PJ, Potter JF. Cerebral blood flow velocity response to induced and spontaneous sudden changes in arterial blood pressure. Am J Physiol Heart Circ Physiol 2001; 280(5):H2162–H2174. 28 Derdeyn CP, Videen TO, Yundt KD, Fritsch SM, Carpenter DA, Grubb RL et al. Variability of cerebral blood volume and oxygen extraction: stages of cerebral haemodynamic impairment revisited. Brain 2002; 125(Pt 3):595–607. 29 Powers WJ. Cerebral hemodynamics in ischemic cerebrovascular disease. Ann Neurol 1991; 29(3):231–40. 30 Schellinger PD, Latour LL, Wu CS, Chalela JA, Warach S. The association between neurological deficit in acute ischemic stroke and mean transit time: comparison of four different perfusion MRI algorithms. Neuroradiology 2006; 48(2):69–77. 31 Muir KW, Santosh C. Imaging of acute stroke and transient ischaemic attack. J Neurol Neurosurg Psychiatry 2005; 76 Suppl 3:iii19–iii28. 32 Pindzola RR, Balzer JR, Nemoto EM, Goldstein S, Yonas H. Cerebrovascular reserve in patients with carotid occlusive disease assessed by stable xenon-enhanced CT cerebral blood flow and transcranial Doppler. Stroke 2001; 32(8):1811–17. 33 Mahagne MH, David O, Darcourt J, Migneco O, Dunac A, Chatel M et al. Voxel-based mapping of cortical ischemic damage using Tc 99 m L,L-ethyl cysteinate dimer SPECT in acute stroke. J Neuroimaging 2004; 14(1):23–32. 34 Murphy BD, Fox AJ, Lee DH, Sahlas DJ, Black SE, Hogan MJ et al. Identification of penumbra and infarct in acute ischemic stroke using computed tomography perfusionderived blood flow and blood volume measurements. Stroke 2006; 37(7):1771–7. 35 Koroshetz WJ, Lev MH. Contrast computed tomography scan in acute stroke: ‘You can’t always get what you want but . . . you get what you need’. Ann Neurol 2002; 51(4):415–16. 36 Eastwood JD, Alexander MJ, Petrella JR, Provenzale JM. Dynamic CT perfusion imaging with acetazolamide challenge for the preprocedural evaluation of a patient with symptomatic middle cerebral artery occlusive disease. Am J Neuroradiol 2002; 23(2):285–7. 37 Hellier KD, Hampton JL, Guadagno JV, Higgins NP, Antoun NM, Day DJ et al. Perfusion CT helps decision making for thrombolysis when there is no clear time of onset. J Neurol Neurosurg Psychiatry 2006; 77(3):417–19. 38 Ziyeh S, Rick J, Reinhard M, Hetzel A, Mader I, Speck O. Blood oxygen level-dependent MRI of cerebral CO2 reactivity in severe carotid stenosis and occlusion. Stroke 2005; 36(4):751–6. 39 Geisler BS, Brandhoff F, Fiehler J, Saager C, Speck O, Rother J et al. Blood-oxygen-level-dependent MRI allows

..

..

40

41 42

43

44

45

46 47

48

49 50

51

52

53

54

metabolic description of tissue at risk in acute stroke patients. Stroke 2006; 37(7):1778–84. Roc AC, Wang J, Ances BM, Liebeskind DS, Kasner SE, Detre JA. Altered hemodynamics and regional cerebral blood flow in patients with hemodynamically significant stenoses. Stroke 2006; 37(2):382–7. Markus HS. Transcranial Doppler ultrasound. J Neurol Neurosurg Psychiatry 1999; 67(2):135–7. Derdeyn CP. Is the acetazolamide test valid for quantitative assessment of maximal cerebral autoregulatory vasodilation? Stroke 2000; 31(9):2271–2. Yamauchi H, Kudoh T, Sugimoto K, Takahashi M, Kishibe Y, Okazawa H. Pattern of collaterals, type of infarcts, and haemodynamic impairment in carotid artery occlusion. J Neurol Neurosurg Psychiatry 2004; 75(12):1697–701. Markus H, Cullinane M. Severely impaired cerebrovascular reactivity predicts stroke and TIA risk in patients with carotid artery stenosis and occlusion. Brain 2001; 124(Pt 3):457–67. Baron JC. Using PET to identify carotid occlusion patients at high risk of subsequent stroke: further insights. J Neurol Neurosurg Psychiatry 2004; 75(12):1659–60. Derdeyn CP, Grubb RL, Jr., Powers WJ. PET screening of carotid occlusion. Adm Radiol J 2001;20(1):20–5. Derdeyn CP, Gage BF, Grubb RL, Jr., Powers WJ. Costeffectiveness analysis of therapy for symptomatic carotid occlusion: PET screening before selective extracranial-tointracranial bypass versus medical treatment. J Nucl Med 2000; 41(5):800–7. Htun P, Fateh-Moghadam S, Tomandl B, Handschu R, Klinger K, Stellos K et al. Course of platelet activation and platelet-leukocyte interaction in cerebrovascular ischemia. Stroke 2006; 37(9):2283–7. Warach S. Tissue viability thresholds in acute stroke: the 4-factor model. Stroke 2001; 32(11):2460–1. Heiss WD, Rosner G. Functional recovery of cortical neurons as related to degree and duration of ischemia. Ann Neurol 1983; 14(3):294–301. Marchal G, Beaudouin V, Rioux P, de la Sayette V, Le Doze F, Viader F et al. Prolonged persistence of substantial volumes of potentially viable brain tissue after stroke: a correlative PET-CT study with voxel-based data analysis. Stroke 1996; 27(4):599–606. Hacke W, Albers G, Al Rawi Y, Bogousslavsky J, Davalos A, Eliasziw M et al. The Desmoteplase in Acute Ischemic Stroke Trial (DIAS): a phase II MRI-Based 9-hour window acute stroke thrombolysis trial with intravenous desmoteplase. Stroke 2005; 36:66–73. Furlan AJ, Eyding D, Albers GW, Al-Rawi Y, Lees KR, Rowley HA et al. Dose Escalation of Desmoteplase for Acute Ischemic Stroke (DEDAS): evidence of safety and efficacy 3 to 9 hours after stroke onset. Stroke 2006; 37(5):1227–31. Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343:311–22.

693

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 694

694

Chapter 12 Specific treatments for acute ischaemic stroke

55 Baron JC. How healthy is the acutely reperfused ischemic penumbra? Cerebrovasc Dis 2005; 20 (Suppl 2):25–31. 56 Ginsberg MD. Adventures in the pathophysiology of brain ischemia: penumbra, gene expression, neuroprotection: the 2002 Thomas Willis Lecture. Stroke 2003; 34(1):214–23. 57 Markus R, Reutens DC, Kazui S, Read S, Wright P, Pearce DC et al. Hypoxic tissue in ischaemic stroke: persistence and clinical consequences of spontaneous survival. Brain 2004; 127(Pt 6):1427–36. 58 Baron JC, Warach S. Imaging. Stroke 2005; 36(2):196–9. 59 Heiss WD. Best measure of ischemic penumbra: positron emission tomography. Stroke 2003; 34(10):2534–5. 60 Heiss WD, Kracht LW, Thiel A, Grond M, Pawlik G. Penumbral probability thresholds of cortical flumazenil binding and blood flow predicting tissue outcome in patients with cerebral ischaemia. Brain 2001; 124(Pt 1):20–9. 61 Latchaw RE, Yonas H, Hunter GJ, Yuh WT, Ueda T, Sorensen AG et al. Guidelines and recommendations for perfusion imaging in cerebral ischemia: a scientific statement for healthcare professionals by the writing group on perfusion imaging, from the Council on Cardiovascular Radiology of the American Heart Association. Stroke 2003; 34(4):1084–104. 62 Warach S. Measurement of the ischemic penumbra with MRI: it’s about time. Stroke 2003; 34(10):2533–4. 63 Sobesky J, Zaro WO, Lehnhardt FG, Hesselmann V, Neveling M, Jacobs A et al. Does the mismatch match the penumbra? Magnetic resonance imaging and positron emission tomography in early ischemic stroke. Stroke 2005; 36(5):980–5. 64 Guadagno JV, Jones PS, Fryer TD, Barret O, Aigbirhio FI, Carpenter TA et al. Local relationships between restricted water diffusion and oxygen consumption in the ischemic human brain. Stroke 2006; 37(7):1741–8. 65 Rivers CS, Wardlaw JM. What has diffusion imaging in animals told us about diffusion imaging in patients with ischaemic stroke? Cerebrovasc Dis 2005; 19(5):328–36. 66 Hjort N, Christensen S, Solling C, Ashkanian M, Wu O, Rohl L et al. Ischemic injury detected by diffusion imaging 11 minutes after stroke. Ann Neurol 2005; 58(3):462–5. 67 Rivers CS, Wardlaw JM, Armitage PA, Bastin ME, Carpenter TK, Cvoro V et al. Do acute diffusion- and perfusion-weighted MRI lesions identify final infarct volume in ischemic stroke? Stroke 2006; 37(1):98–104. 68 Shih LC, Saver JL, Alger JR, Starkman S, Leary MC, Vinuela F et al. Perfusion-weighted magnetic resonance imaging thresholds identifying core, irreversibly infarcted tissue. Stroke 2003; 34(6):1425–30. 69 Wardlaw JM, Farrall AJ. Diagnosis of stroke on neuroimaging. Br Med J 2004; 328(7441):655–6. 70 Coutts SB, Simon JE, Tomanek AI, Barber PA, Chan J, Hudon ME et al. Reliability of assessing percentage of diffusion-perfusion mismatch. Stroke 2003; 34(7):1681–3. 71 Kidwell CS, Alger JR, Saver JL. Beyond mismatch: evolving paradigms in imaging the ischemic penumbra with

72

73

74

75

76

77

78 79 80

81 82 83

84

85

86

87

88

89

multimodal magnetic resonance imaging. Stroke 2003; 34(11):2729–35. Butcher KS, Parsons M, MacGregor L, Barber PA, Chalk J, Bladin C et al. Refining the perfusion-diffusion mismatch hypothesis. Stroke 2005; 36(6):1153–9. Kucinski T, Naumann D, Knab R, Schoder V, Wegener S, Fiehler J et al. Tissue at risk is overestimated in perfusionweighted imaging: MR imaging in acute stroke patients without vessel recanalization. Am J Neuroradiol 2005; 26(4):815–19. Coutts SB, Hill MD, Hu WY. Hyperperfusion syndrome: toward a stricter definition. Neurosurgery 2003; 53(5):1053–8. Bright R, Raval AP, Dembner JM, Perez-Pinzon MA, Steinberg GK, Yenari MA et al. Protein kinase C delta mediates cerebral reperfusion injury in vivo. J Neurosci 2004; 24(31):6880–8. Price CJ, Warburton EA, Menon DK. Human cellular inflammation in the pathology of acute cerebral ischaemia. J Neurol Neurosurg Psychiatry 2003; 74(11):1476–84. Lo EH, Dalkara T, Moskowitz MA. Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci 2003; 4(5):399–415. Lipton P. Ischemic cell death in brain neurons. Physiol Rev 1999; 79(4):1431–568. Graham SH, Chen J. Programmed cell death in cerebral ischemia. J Cereb Blood Flow Metab 2001; 21(2):99–109. Chan PH. Reactive oxygen radicals in signaling and damage in the ischemic brain. J Cereb Blood Flow Metab 2001; 21(1):2–14. Lo EH, Moskowitz MA, Jacobs TP. Exciting, radical, suicidal: how brain cells die after stroke. Stroke 2005; 36(2):189–92. Moskowitz MA, Lo EH. Neurogenesis and apoptotic cell death. Stroke 2003; 34(2):324–6. Bergersen LH, Magistretti PJ, Pellerin L. Selective postsynaptic co-localization of MCT2 with AMPA receptor GluR2/3 subunits at excitatory synapses exhibiting AMPA receptor trafficking. Cereb Cortex 2005; 15(4):361–70. Lizasoain I, Cardenas A, Hurtado O, Romera C, Mallolas J, Lorenzo P et al. Targets of cytoprotection in acute ischemic stroke: present and future. Cerebrovasc Dis 2006; 21 (Suppl 2):1–8. Bernardinelli Y, Magistretti PJ, Chatton JY. Astrocytes generate Na+-mediated metabolic waves. Proc Natl Acad Sci USA 2004; 101(41):14937–42. Aarts MM, Arundine M, Tymianski M. Novel concepts in excitotoxic neurodegeneration after stroke. Expert Rev Mol Med 2003; 2003:1–22. Chu CT, Levinthal DJ, Kulich SM, Chalovich EM, DeFranco DB. Oxidative neuronal injury. The dark side of ERK1/2. Eur J Biochem 2004; 271(11):2060–6. Lipton SA, Bossy-Wetzel E. Dueling activities of AIF in cell death versus survival: DNA binding and redox activity. Cell 2002; 111(2):147–50. Okuno S, Saito A, Hayashi T, Chan PH. The c-Jun Nterminal protein kinase signaling pathway mediates Bax activation and subsequent neuronal apoptosis through

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 695

References

90

91

92

93

94

95 96

97

98

99

100

101

102

103

104

105

106

..

..

interaction with Bim after transient focal cerebral ischemia. J Neurosci 2004; 24(36):7879–87. Driscoll M, Gerstbrein B. Dying for a cause: invertebrate genetics takes on human neurodegeneration. Nat Rev Genet 2003; 4(3):181–94. Xiong ZG, Zhu XM, Chu XP, Minami M, Hey J, Wei WL et al. Neuroprotection in ischemia: blocking calciumpermeable acid-sensing ion channels. Cell 2004; 118(6):687–98. Gursoy-Ozdemir Y, Qiu J, Matsuoka N, Bolay H, Bermpohl D, Jin H et al. Cortical spreading depression activates and upregulates MMP-9. J Clin Invest 2004; 113(10):1447–55. Otori T, Greenberg JH, Welsh FA. Cortical spreading depression causes a long-lasting decrease in cerebral blood flow and induces tolerance to permanent focal ischemia in rat brain. J Cereb Blood Flow Metab 2003; 23(1):43–50. Zhao BQ, Wang S, Kim HY, Storrie H, Rosen BR, Mooney DJ et al. Role of matrix metalloproteinases in delayed cortical responses after stroke. Nat Med 2006; 12(4):441–5. Zlokovic BV. Remodeling after stroke. Nat Med 2006; 12(4):390–1. Perera MN, Ma HK, Arakawa S, Howells DW, Markus RM, Rowe CC et al. Inflammation following stroke. J Clin Neurosci 2006; 13:1–8. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci 1999; 22(9):391–7. Del Zoppo GJ, Ginis I, Hallenbeck JM, Iadecola C, Wang X, Feuerstein GZ. Inflammation and stroke: putative role for cytokines, adhesion molecules and iNOS in brain response to ischemia. Brain Pathol 2000; 10(1):95–112. Feuerstein GZ, Wang X, Barone FC. Inflammatory gene expression in cerebral ischemia and trauma: potential new therapeutic targets. Ann N Y Acad Sci 1997; 825:179–93. O’Collins VE, Macleod MR, Donnan GA, Horky LL, van der Worp BH, Howells DW. 1026 experimental treatments in acute stroke. Ann Neurol 2006; 59(3):467–77. Wahlgren NG, Ahmed N. Neuroprotection in cerebral ischaemia: facts and fancies: the need for new approaches. Cerebrovasc Dis 2004;17 (Suppl 1):153–66. Fagan SC, Hess DC, Machado LS, Hohnadel EJ, Pollock DM, Ergul A. Tactics for vascular protection after acute ischemic stroke. Pharmacotherapy 2005; 25(3):387–95. Hurtado O, Pradillo JM, Alonso-Escolano D, Lorenzo P, Sobrino T, Castillo J et al. Neurorepair versus neuroprotection in stroke. Cerebrovasc Dis 2006; 21 (Suppl 2):54–63. Singhal AB, Lo EH, Dalkara T, Moskowitz MA. Advances in stroke neuroprotection: hyperoxia and beyond. Neuroimaging Clin N Am 2005; 15(3):697–xiii. Falcao AL, Reutens DC, Markus R, Koga M, Read SJ, Tochon-Danguy H et al. The resistance to ischemia of white and gray matter after stroke. Ann Neurol 2004; 56(5):695–701. Stamatovic SM, Dimitrijevic OB, Keep RF, Andjelkovic AV. Inflammation and brain edema: new insights into the role of chemokines and their receptors. Acta Neurochir Suppl 2006; 96:444–50.

107 Pearce JM. Kernohan’s notch. Eur Neurol 2006; 55(4):230–2. 108 Sackett DL, Strauss S, Richardson WS, Rosenberg W, Haynes RB. Evidence-Based Medicine: How To Practice and Teach EBM. 2nd edn. Edinburgh: Churchill Livingstone, 2000. 109 Sackett DL, Haynes RB, Guyatt G, Tugwell P. Clinical Epidemiology. A Basic Science for Clinical Medicine. Toronto: Little, Brown & Co., 1991. 110 Collins R, MacMahon S. Reliable assessment of the effects of treatment on mortality and major morbidity, I: clinical trials. Lancet 2001; 357(9253):373–80. 111 Medical Research Council. Streptomycin treatment of pulmonary tuberculosis. Br Med J 1948; 2:769–82. 112 Yoshioka A. Use of randomisation in the Medical Research Council’s clinical trial of streptomycin in pulmonary tuberculosis in the 1940s. Br Med J 2005; 317:1220–3. 113 Lewis SC, Warlow CP. How to spot bias and other potential problems in randomised controlled trials. J Neurol Neurosurg Psychiatry 2004; 75(2):181–7. 114 Califf RM. Simple principles of clinical trials remain powerful. J Am Med Assoc 2005; 293(4):489–91. 115 Dyken M, White P. Evaluation of cortisone in the treatment of cerebral infarction. J Am Med Assoc 1956; 162:1531–4. 116 Bath FJ, Owen VE, Bath PM. Quality of full and final publications reporting acute stroke trials: a systematic review. Stroke 1998; 29(10):2203–10. 117 Weaver CS, Leonardi-Bee J, Bath-Hextall FJ, Bath PM. Sample size calculations in acute stroke trials: a systematic review of their reporting, characteristics, and relationship with outcome. Stroke 2004; 35(5):1216–24. 118 Dorman PJ, Counsell C, Sandercock P. Reports of randomized trials in acute stroke, 1955 to 1995. What proportions were commercially sponsored? Stroke 1999; 30(10):1995–8. 119 Kidwell CS, Liebeskind DS, Starkman S, Saver JL. Trends in acute ischemic stroke trials through the 20th century. Stroke 2001; 32(6):1349–59. 120 Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. Br Med J 2003; 326(7400):1167–70. 121 Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. J Am Med Assoc 2003; 289(4):454–65. 122 Ioannidis JP. Effect of the statistical significance of results on the time to completion and publication of randomized efficacy trials. [see comment]. J Am Med Assoc 1998; 279(4):281–6. 123 Krzyzanowska MK, Pintilie M, Tannock IF. Factors associated with failure to publish large randomized trials presented at an oncology meeting. J Am Med Assoc 2003; 290(4):495–501. 124 Donnan G, Davis SM, Kaste M, International Trial Subcommittee of the International Stroke Liaison Committee ASA. Recommendations for the relationship between sponsors and investigators in the design and conduct of clinical stroke trials. Stroke 2003; 34(4):1041–5.

695

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 696

696

Chapter 12 Specific treatments for acute ischaemic stroke

125 Warlow C. Over-regulation of clinical research: a threat to public health. Clin Med 2005; 5(1):33–8. 126 Warlow C. The Willis Lecture 2003: evaluating treatments for stroke patients too slowly: time to get out of second gear. Stroke 2004; 35(9):2211–19. 127 Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet 2001; 357(9263):1191–4. 128 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999; 354(9193):1896–900. 129 Schulz KF, Grimes DA. Multiplicity in randomised trials II: subgroup and interim analyses. Lancet 2005; 365(9471):1657–61. 130 Gent M, Barnett HJ, Sackett DL, Taylor DW. A randomized trial of aspirin and sulfinpyrazone in patients with threatened stroke. Results and methodologic issues. Circulation 1980; 62(6 Pt 2):V97–105. 131 Antiplatelet Therapy Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy. I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Antiplatelet Trialists’ Collaboration. [see comments]. [erratum appears in Br Med J 1994; 308(6943):1540]. Br Med J 1994; 308(6921):81–106. 132 The National Institute of Stroke and Neurological Disorders rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995; 333(24):1581–7. 133 Wardlaw JM, Zoppo G, Yamaguchi T, Berge E. Thrombolysis for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000213; PMID: 10796329]. Cochrane Database Syst Rev 2003; (3):CD000213. 134 Alderson P, Bero LA, Eccles M, Grilli R, Grimshaw JM, Oxman AD et al. Effective Practice and Organisation of Care Group. About the Cochrane Collaboration (Collaborative Review Groups (CRGs)). Cochrane Library 2005;(3). 135 Jamtvedt G, Young JM, Kristoffersen DT, Thomson O’Brien MA, Oxman AD. Audit and feedback: effects on professional practice and health care outcomes. [update of Cochrane Database Syst Rev 2000; (2):CD000259; PMID: 10796520]. Cochrane Database Sys Rev 2003; (3):CD000259. 136 Sandercock P, Fraser H, Thomas B, McInnes A, Dixon S. Cochrane Stroke Group 10 years on: progress to date and future challenges. Stroke 2003; 34(10):2537–9. 137 Jadad AR, Cook DJ, Jones A, Klassen TP, Tugwell P, Moher M et al. Methodology and reports of systematic reviews and meta-analyses: a comparison of Cochrane reviews with articles published in paper-based journals. J Am Med Assoc 1998; 280(3):278–80. 138 Asplund K. Stroke care in the UK: is it good enough? Lancet Neurol 2002; 1:341. 139 Asplund K. Stroke in Europe: widening gap between East and West. Cerebrovasc Dis 1996; 6:3–6.

140 Asplund K, Ashburner S, Cargill K, Hux M, Lees K, Drummond M et al. Health care resource use and stroke outcome. Multinational comparisons within the GAIN International trial. Int J Tech Assess Hlth Care 2003; 19(2):267–77. 141 Stegmayr B, Asplund K, Hulter-Asberg K, Norrving B, Terent A, Wester PO. Large variations in medical antithrombotic treatment of stroke patients in hospitals. Results from Riks-Stroke, the Swedish National Quality Register. Stroke (online) 2004; 35:E199. 142 Moussouttas MM, Lichtman JH, Krumholtz HM, Cerese J, Brass L. The use of heparin anticoagulation in acute ischemic stroke among academic medical centers. Stroke 1999; 30:265. 143 Al Sadat A, Sunbulli M, Chaturvedi S. Use of intravenous heparin by North American neurologists: do the data matter? Stroke 2002; 33(6):1574–7. 144 Lakshminarayan K, Solid CA, Collins AJ, Anderson DC, Herzog CA. Atrial fibrillation and stroke in the general medicare population: a 10-year perspective (1992 to 2002). Stroke 2006; 37(8):1969–74. 145 Reed SD, Cramer SC, Blough DK, Meyer K, Jarvik JJ. Treatment with tissue plasminogen activator and inpatient mortality rates for patients with ischemic stroke treated in community hospitals. Stroke 2001; 32:1832–40. 146 Johnston SC, Fung LH, Gillum LA, Smith WS, Brass LM, Lichtman JH et al. Utilization of intravenous tissue-type plasminogen activator for ischemic stroke at academic medical centers: the influence of ethnicity [see comment]. Stroke 2001; 32(5):1061–8. 147 Heuschmann P, Berger K, Misselwitz B, Hermanek P, Leffmann C, Adelmann M. Frequency of thrombolytic therapy in patients with acute ischemic stroke and the risk of in-hospital mortality. The German Stroke Registers Study Group. Stroke 2003; 34:1106–13. 148 Katzan L, Furlan AJ, Lloyd L, Frank J, Harper D, Hinchey J et al. Use of tissue-type plasminogen activator for acute ischemic stroke. The Cleveland Area Experience. J Am Med Assoc 2000; 283:1151–8. 149 Brown D, Barsan W, Lisabeth L, Gallery M, Morgenstern L. Survey of emergency physicians about recombinant tissue plasminogen activator for acute ischemic stroke. Ann Emerg Med 2005; 46(1):56–60. 150 Bednar MM, Gross CE. Antiplatelet therapy in acute cerebral ischemia. [see comment]. Stroke 1999; 30(4):887–93. 151 van Kooten F, Ciabattoni G, Patrono C, Dippel DW, Koudstaal PJ. Platelet activation and lipid peroxidation in patients with acute ischemic stroke. Stroke 1997; 28(8):1557–63. 152 Antiplatelet Therapy Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy. III: Reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. Antiplatelet Trialists’ Collaboration. Br Med J 1994; 308(6923):235–46. 153 ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 697

References

154

155

156

157

158

159

160

161

162

163

164

165

..

..

streptokinase, oral aspirin, both, or neither among 17 187 cases of suspected acute myocardial infarction: ISIS-2. [see comment]. Lancet 1988; 2(8607):349–60. Sandercock P, Collins R, Counsell C, Farrell B, Peto R, Slattery J et al. The International Stroke Trial (IST): A randomised trial of aspirin, subcutaneous heparin, both, or neither among 19 435 patients with acute ischaemic stroke. Lancet 1997; 349:1569–81. Chen ZM, Collins R, Liu LS, Pan HC, Peto R, Xie JX. CAST: Randomised placebo-controlled trial of early aspirin use in 20 000 patients with acute ischaemic stroke. Lancet 1997; 349:1641–9. Sandercock P, Gubitz G, Foley P, Counsell C. Antiplatelet therapy for acute ischaemic stroke. [see comment] [update of Cochrane Database Syst Rev 2000; (2):CD000029; PMID: 10796284]. Cochrane Database Syst Rev 2003; (2):CD000029. Chen ZM, Sandercock P, Pan HC, Counsell C, Collins R, Liu LS et al. Indications for early aspirin use in acute ischemic stroke : a combined analysis of 40 000 randomized patients from the chinese acute stroke trial and the international stroke trial. Stroke 2000; 31(6):1240–9. Pulmonary Embolism Prevention (PEP) Trial Collaborative Group. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) trial. [see comment]. Lancet 2000; 355(9212):1295–302. Keir SL, Wardlaw JM, Sandercock PA, Chen Z. Antithrombotic therapy in patients with any form of intracranial haemorrhage: a systematic review of the available controlled studies. Cerebrovasc Dis 2002; 14(3–4):197–206. Coull BM, Williams LS, Goldstein LB, Meschia JF, Heitzman D, Chaturvedi S et al. Anticoagulants and antiplatelet agents in acute ischemic stroke: report of the Joint Stroke Guideline Development Committee of the American Academy of Neurology and the American Stroke Association (a division of the American Heart Association). Stroke 2002; 33(7):1934–42. Toyoda K, Okada Y, Minematsu K, Kamouchi M, Fujimoto S, Ibayashi S et al. Antiplatelet therapy contributes to acute deterioration of intracerebral hemorrhage. Neurology 2005; 65(7):1000–4. Viswanathan A, Rakich SM, Engel C, Snider R, Rosand J, Greenberg SM et al. Antiplatelet use after intracerebral hemorrhage. Neurology 2006; 66(2):206–9. Wardlaw JM, Seymour J, Cairns J, Keir S, Lewis S, Sandercock P. Immediate computed tomography scanning of acute stroke is cost-effective and improves quality of life. Stroke 2004; 35(11):2477–83. MAST-I Group. Randomised controlled trial of streptokinase, aspirin, and combination of both in treatment of acute ischaemic stroke. Multicentre Acute Stroke Trial–Italy (MAST-I) Group. Lancet 1995; 346(8989):1509–14. Schmulling S, Rudolf J, Strotmann-Tack T, Grond M, Schneweis S, Sobesky J et al. Acetylsalicylic acid pretreatment, concomitant heparin therapy and the risk

166

167

168 169

170

171

172

173

174

175

176

177

178

179

of early intracranial hemorrhage following systemic thrombolysis for acute ischemic stroke. Cerebrovasc Dis 2003; 16(3):183–90. Intercollegiate Stroke Working Party. National Clinical Guidelines for Stroke. 2nd edn. Royal College of Physicians of London, 2004. Adams HP, Adams RJ, Brott T, del Zoppo GJ, Furlan AJ, Goldstein LB et al. Guidelines for the Early Management of Patients With Ischemic Stroke. A Scientific Statement From the Stroke Council of the American Stroke Association. Stroke 2003; 34, 1056–83. Patrono C. Aspirin as an antiplatelet drug. N Engl J Med 1994; 330(18):1287–94. Patrono C, Ciabattoni G, Davi G. Thromboxane biosynthesis in cardiovascular diseases. Stroke 1990; 21(12 Suppl):IV130–IV133. Hop JW, Rinkel GJ, Algra A, Berkelbach van der Sprenkel JW, van Gijn J. Randomized pilot trial of postoperative aspirin in subarachnoid hemorrhage. Neurology 2000; 54(4):872–8. Hart RG, Tonarelli S, Pearce LA. Avoiding central nervous system bleeding during antithrombotic therapy. Recent data and ideas. Stroke 2005; 36:1588–93. Gubitz G, Sandercock P, Counsell C. Anticoagulants for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000024; PMID: 10796283]. Cochrane Database Syst Rev 2004; (3):CD000024. Sandercock P, Counsell C, Stobbs SL. Low-molecularweight heparins or heparinoids versus standard unfractionated heparin for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2001; (4):CD000119; PMID: 11687069]. Cochrane Database Syst Rev 2005; (2):CD000119. Bath PM, Iddenden R, Bath FJ. Low-molecular-weight heparins and heparinoids in acute ischemic stroke: a meta-analysis of randomized controlled trials. Stroke 2000; 31(7):1770–8. Berge E, Sandercock P. Anticoagulants versus antiplatelet agents for acute ischaemic stroke. [see comment]. Cochrane Database Syst Rev 2002; (4):CD003242. Chamorro A, Busse O, Obach V, Toni D, Sandercock P, Reverter JC et al. The rapid anticoagulation prevents ischemic damage study in acute stroke: final results from the writing committee. Cerebrovasc Dis 2005; 19(6):402–4. Camerlingo M, Salvi P, Belloni G, Gamba T, Cesana BM, Mamoli A. Intravenous heparin started within the first 3 hours after onset of symptoms as a treatment for acute nonlacunar hemispheric cerebral infarctions. Stroke 2005; 36(11):2415–20. Diener HC, Ringelstein EB, von KR, Landgraf H, Koppenhagen K, Harenberg J et al. Prophylaxis of thrombotic and embolic events in acute ischemic stroke with the low-molecular-weight heparin certoparin: results of the PROTECT Trial. Stroke 2006; 37:139–44. Diener HC, Ringelstein EB, von Kummer R, Langohr HD, Bewermeyer H, Landgraf H et al. Treatment of acute ischemic stroke with the low-molecular-weight heparin certoparin: results of the TOPAS trial. Therapy of Patients

697

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 698

698

Chapter 12 Specific treatments for acute ischaemic stroke

180

181

182

183

184

185

186

187

188

189

190

191

With Acute Stroke (TOPAS) Investigators. Stroke 2001; 32(1):22–9. Wong KS, Chen C, Ng PW, Tsoi TH, Li HL, Fong WC et al. Low-molecular-weight heparin compared with aspirin for the treatment of acute ischaemic stroke in Asian patients with large artery occlusive disease: a randomised study. Lancet Neurol 2007; 6:407–13. LaMonte MP, Nash ML, Wang DZ, Woolfenden AR, Schultz J, Hursting MJ et al. Argatroban anticoagulation in patients with acute ischemic stroke (ARGIS-1): a randomized, placebo-controlled safety study. Stroke 2004; 35(7):1677–82. Gubitz G, Sandercock P, Counsell C. Immediate anticoagulant therapy for acute ischaemic stroke: a systematic review of seven randomised trials directly comparing different doses of the same anticoagulant. Stroke 2000; 31:308. Bath PM, Lindenstrom E, Boysen G, De Deyn P, Friis P, Leys D et al. Tinzaparin in acute ischaemic stroke (TAIST): a randomised aspirin-controlled trial. [see comments]. [erratum appears in Lancet 2001; 358(9289):1276]. Lancet 2001; 358(9283):702–10. Saxena R, Lewis S, Berge E, Sandercock P, Koudstaal PJ. Risk of early death and recurrent stroke and effect of heparin in 3169 patients with acute ischemic stroke and atrial fibrillation in the International Stroke Trial. Stroke 2001; 32(10):2333–7. Berge E, Abdelnoor M, Nakstad PH, Sandset PM. Low molecular-weight heparin versus aspirin in patients with acute ischaemic stroke and atrial fibrillation: a doubleblind randomised study. HAEST Study Group. Heparin in Acute Embolic Stroke Trial. [see comments]. Lancet 2000; 355(9211):1205–10. Randomised controlled trial of subcutaneous calciumheparin in acute myocardial infarction. The SCATI (Studio sulla Calciparina nell’Angina e nella Trombosi Ventricolare nell’Infarto) Group. Lancet 1989; 2(8656):182–6. Turpie AG, Robinson JG, Doyle DJ, Mulji AS, Mishkel GJ, Sealey BJ et al. Comparison of high-dose with low-dose subcutaneous heparin to prevent left ventricular mural thrombosis in patients with acute transmural anterior myocardial infarction. N Engl J Med 1989; 320(6):352–7. Collins R, MacMahon S, Flather M, Baigent C, Remvig L, Mortensen S et al. Clinical effects of anticoagulant therapy in suspected acute myocardial infarction: systematic overview of randomised trials. [see comment]. Br Med J 1996; 313(7058):652–9. Cairns JA, Theroux P, Lewis HD, Jr., Ezekowitz M, Meade TW. Antithrombotic agents in coronary artery disease. Chest 2001; 119(Suppl 1):228S–52S. Vaitkus PT, Berlin JA, Schwartz JS, Barnathan ES. Stroke complicating acute myocardial infarction. A meta-analysis of risk modification by anticoagulation and thrombolytic therapy. Arch Intern Med 1992; 152(10):2020–4. Scottish Intercollegiate Guidelines Network. Management of Patients with Stroke. Rehabilitation, Prevention and Management of Complications, and Discharge Planning. Edinburgh: Royal College of Physicians Edinburgh, 2002.

192 Lindley RI, Wardlaw J, Sandercock P, Rimdusid P, Lewis SC, Signorini D et al. Frequency and risk factors for spontaneous hemorrhagic transformation of cerebral infarction. J Stroke Cerebrovasc Dis 2004; 13(6):235–46. 193 Scottish Intercollegiate Guidelines Network. Antithrombotic Therapy. Edinburgh: Royal College of Physicians Edinburgh, 1999. 194 Toyonaga M, Hagiwara N, Irie F, Toyoda K, Fujimoto S, Hitotsumatsu T et al. [Acute cervical spinal epidural hematoma during antithrombotic therapy: dual warnings against antithrombotic therapy]. Rinsho Shinkeigaku 2003; 43(5):287–90. 195 Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: current practices and unresolved questions. Stroke 2006; 37(1):256–62. 196 Huttner HB, Schellinger PD, Hartmann M, Kohrmann M, Juettler E, Wikner J et al. Hematoma growth and outcome in treated neurocritical care patients with intracerebral hemorrhage related to oral anticoagulant therapy: comparison of acute treatment strategies using vitamin k, fresh frozen plasma, and prothrombin complex concentrates. Stroke 2006; 37(6):1465–70. 197 Wijdicks EF, Schievink WI, Brown RD, Mullany CY. Early anticoagulation in patients with prosthetic heart valves and intracerebral hematoma. [comment]. Neurology 1999; 52(3):676–7. 198 Phan TG, Koh M, Wijdicks EFM. Safety of discontinuation of anticoagulation in patients with intracranial hemorrhage at high thromboembolic risk. Arch Neurol 2000; 57(12):1710–13. 199 Phan TG, Wijdicks EF, Phan TG, Wijdicks EF. Management of intracranial bleeding associated with anticoagulation: balancing the risk of further bleeding against thromboembolism from prosthetic heart valves. J Neurol Neurosurg Psychiatry 2001; 70(6):820–1. 200 Eckman MH, Rosand J, Knudsen KA, Singer DE, Greenberg SM. can patients be anticoagulated after intracerebral hemorrhage?: A decision analysis. Stroke 2003; 34(7):1710–16. 201 Butler A, Tait RC, Eckman MH, Singer DE, Rosand J, Knudsen KA et al. Restarting oral anticoagulation after intracranial hemorrhage. Response. Stroke 2004; 35(1):5e–6. 202 Little SH, Massel DR. Antiplatelet and anticoagulation for patients with prosthetic heart valves. Cochrane Database Syst Rev 2003; (4):CD003464. 203 Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Broderick JP et al. Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet 2004; 363(9411):768–74. 204 Cornu C, Boutitie F, Candelise L, Boissel JP, Donnan G, Hommel M et al. Streptokinase in acute ischemic stroke: an individual patient data meta-analysis: The Thrombolysis in Acute Stroke Pooling Project. Stroke 2000; 31(7):1555–60. 205 Haley EC, Lyden P, Johnston KC, Hemmen TM, the TNK in Stroke Investigators. A pilot dose-escalation safety study of

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 699

References

206

207

208

209

210

211

212

213

214

215

216 217 218

219

..

..

tenecteplase in acute ischemic stroke. Stroke 2005; 36:607–12. Hacke W. The results of the joint analysis of two phase II trials on desmoteplase in acute ischemic stroke with treatment 3 to 9 hours after stroke onset. Cerebrovasc Dis 2005; 19 (Suppl 2):69. Macleod MR, Davis SM, Mitchell P, Gerraty R, Hankey G, Fitt G et al. Results of a multicentre, randomised controlled trial of intra-arterial urokinase in the treatment of acute posterior circulation ischaemic stroke. Cerebrovasc Dis 2005; 20:12–17. Fibrinolytic Therapy Trialists’ Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343:311–22. McCarron MO, Nicoll JA. Cerebral amyloid angiopathy and thrombolysis-related intracerebral haemorrhage. Lancet Neurol 2004; 3(8):484–92. Kakuda W, Thijs VN, Lansberg MG, Bammer R, Wechsler L, Kemp S et al. Clinical importance of microbleeds in patients receiving IV thrombolysis. Neurology 2005; 65(8):1175–8. Koudstaal PJ, Stibbe J, Vermeulen M. Fatal ischaemic brain oedema after early thrombolysis with tissue plasminogen activator in acute stroke. Br Med J 1988; 297(6663):1571–4. Rudolf J, Grond M, Stenzel C, Neveling M, Heiss WD. Incidence of space-occupying brain edema following systemic thrombolysis of acute supratentorial ischemia. Cerebrovasc Dis 1998; 8(3):166–71. Kwiatkowski T, Libman RB, Frankel M, Tilley B, Morgenstern L, Lu M et al. Effects of tissue plasminogen activator for acute ischemic stroke at one year. N Engl J Med 1999; 340:1781–7. Sandercock P, Berge E, Dennis M, Hand P, Kwan J, Lewis S et al. A systematic review of the effectiveness, cost-effectiveness and barriers to the implementation of thrombolytic and neuroprotective treatment in the NHS. Health Technology Assessment 2002; 6(26):1–112. Counsell C, Dennis M. Systematic review of prognostic models in patients with acute stroke. Cerebrovasc Dis 2001; 12:159–70. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. Br Med J 2003; 326(7382):219. Lagakos SW. The challenge of subgroup analyses: reporting without distorting. N Engl J Med 2006; 354(16):1667–9. O’Fallon WM, Asplund K, Goldfrank LR, Hertzberg V, Ingall TJ, Louis TA. Report of the t-PA Review Committee. 2004. http://www.ninds.nih.gov/funding/ review_committees/t-pa_review_committee/ t-pa_committee_report.pdf Simon JE, Sandler DL, Pexman JH, Hill MD, Buchan AM, Calgary Stroke Program. Is intravenous recombinant tissue plasminogen activator (rt-PA) safe for use in patients over 80 years old with acute ischaemic stroke? The Calgary experience. [see comment]. Age Ageing 2004; 33(2):143–9.

220 Von Kummer R. Early major ischemic changes on computed tomography should preclude use of tissue plasminogen activator. Stroke 2003; 34:820–1. 221 Lyden P. Early major ischemic changes on computed tomography should not preclude use of tissue plasminogen activator. Stroke 2003; 33:812–22. 222 Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliability and outcome after thrombolytic treatment: systematic review. Radiology 2005; 235(2):444–53. 223 Demchuk AM, Hill MD, Barber PA, Silver B, Patel SC, Levine SR. Importance of early ischemic computed tomography changes using ASPECTS in NINDS rt-PA Stroke Study. Stroke 2005; 36(10):2110–15. 224 Dzialowski I, Hill MD, Coutts SB, Demchuk AM, Kent DM, Wunderlich O et al. Extent of early ischemic changes on computed tomography (CT) before thrombolysis: prognostic value of the Alberta Stroke Program Early CT Score in ECASS II. Stroke 2006; 37(4):973–8. 225 Schellinger PD, Fiebach JB. Perfusion-weighted imaging/diffusion-weighted imaging mismatch on MRI can now be used to select patients for recombinant tissue plasminogen activator beyond 3 hours: pro. [see comment]. Stroke 2005; 36(5):1104–5. 226 Zivin JA. Perfusion-weighted imaging/diffusion-weighted imaging mismatch on MRI can now be used to select patients for recombinant tissue plasminogen activator beyond 3 hours [see comment]. Stroke 2005; 36(5):1105–6. 227 Davis SM. Echoplanar Imaging Thrombolysis Evaluation Trial. EPITHET. http://www strokecenter org/trials/TrialDetail aspx?tid=420, 2005; available from: http://www.astn.org.au/epithet/Epithet_home.htm 228 Dundar Y, Hill R, Dickson R, Walley T. Comparative efficacy of thrombolytics in acute myocardial infarction: a systematic review. Q J Med 2003; 96(2):103–13. 229 Walley T, Dundar Y, Hill R, Dickson R. Superiority and equivalence in thrombolytic drugs: an interpretation. Q J Med 2003; 96(2):155–60. 230 Mielke O, Wardlaw J, Liu M. Thrombolysis (different doses, routes of administration and agents) for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000514; PMID: 10796381]. Cochrane Database Syst Rev 2004; (4):CD000514. 231 Molina CA, Saver JL, Molina CA, Saver JL. Extending reperfusion therapy for acute ischemic stroke: emerging pharmacological, mechanical, and imaging strategies. Stroke 2005; 36(10):2311–20. 232 Ciccone A, Motto C, Aritzu E, Piana A, Candelise L. Risk of aspirin use plus thrombolysis after acute ischaemic stroke: a further MAST-I analysis. MAST-I Collaborative Group. Multicentre Acute Stroke Trial–Italy. Lancet 1998; 352(9131):880. 233 del Zoppo GJ, Higashida R, Furlan AJ, Pessin M, Rowley HA, Gent M et al. PROACT: a phase II randomized trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke. Stroke 1998; 29:4–11.

699

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 700

700

Chapter 12 Specific treatments for acute ischaemic stroke

234 Furlan AJ, Higashida R, Wechsler LR, Gent M, Rowley H, Kase C et al. Intra-arterial prourokinase for acute ischemic stroke. The PROACT II Study: a randomized controlled trial. J Am Med Assoc 1999; 282:2003–11. 235 Adams HP, Brott TG, Furlan AJ. Guidelines for thrombolytic therapy in stroke: a supplement to the guidelines for the management of patients with acute ischaemic stroke. A Statement for Healthcare Professionals from a Special Writing Group of the Stroke Council, American Heart Association. Stroke 1996; 27:1711–18. 236 Ingall TJ, O’Fallon WM, Goldfrank LR, Hertzberg V, Louis TA, Christianson TJH. Findings from the reanalysis of the NINDS Tissue Plasminogen Activator for Acute Ischemic Stroke Treatment Trial. Stroke 2004; 35:2418–24. 237 Mori E, Yoneda Y, Tabuchi M, Yoshida T, Ohkawa S, Ohsumi Y et al. Intravenous recombinant tissue plasminogen activator in acute carotid artery territory stroke. Neurology 1992; 42:976–82. 238 Yamaguchi T, Hayakawa T, Kiuchi H, Japanese Thrombolysis Study Group. Intravenous tissue plasminogen activator ameliorates the outcome of hyperacute embolic stroke. Cerebrovasc Dis 1993; 3:269–72. 239 IMS study Investigators. Combined intravenous and intra-arterial recanalization for acute ischemic stroke: the Interventional Management of Stroke Study. [see comment]. Stroke 2004; 35(4):904–11. 240 Khatri P, Kasner SE. Ischemic strokes after cardiac catheterization: opportune thrombolysis candidates? Arch Neurol 2006; 63(6):817–21. 241 Berenstein A, Siller K, Setton A, Nelson P, Levin D, Kupersmith MJ. Intra arterial urokinase for acute ischaemic stroke during interventional neuroradiological procedures. Neurology 1994; 44 (Suppl 2):A356. 242 Zaidat OO, Slivka AP, Mohammad Y, Graffagnino C, Smith TP, Enterline DS et al. Intra-arterial thrombolytic therapy in peri-coronary angiography ischemic stroke. Stroke 2005; 36(5):1089–90. 243 Hankey G, Warlow C. Treatment and secondary prevention of stroke: evidence, costs and effects on individuals and populations. Lancet 1999; 354:1457–63. 244 Gilligan AK, Thrift AG, Sturm JW, Dewey HM, Macdonell RA, Donnan GA. Stroke units, tissue plasminogen activator, aspirin and neuroprotection: which stroke intervention could provide the greatest community benefit? Cerebrovasc Dis 2005; 20(4):239–44. 245 Anonymous. Summary of product characteristics: Actilyse. Bracknell, UK: Boehringer Ingelheim, 2003. 246 Weintraub MI. Thrombolysis (tissue plasminogen activator) in stroke: a medicolegal quagmire. Stroke 2006; 37(7):1917–22. 247 Ciccone A, Sterzi R, Crespi V, Defanti CA, Pasetti C, Bioethics and Neurology Study Group of the Italian Neurological Society. Thrombolysis for acute ischemic stroke: the patient’s point of view. Cerebrovasc Dis 2001; 12(4):335–40. 248 Koops L, Lindley R. Thrombolysis for acute ischaemic stroke: consumer involvement in the design of new randomised controlled trial. Br Med J 2002; 325:415–17.

249 Katz JM, Segal AZ, Katz JM, Segal AZ. Should thrombolysis be given to a stroke patient refusing therapy due to profound anosognosia? [see comment]. Neurology 2004; 63(12):2440. 250 Worrall BB, Chen DT, Dimberg EL, Worrall BB, Chen DT, Dimberg EL. Should thrombolysis be given to a stroke patient refusing therapy due to profound anosognosia? [comment]. Neurology 2005; 65(3):500. 251 Kwan J, Hand P, Sandercock P. A systematic review of barriers to delivery of thrombolysis for acute stroke. [see comment]. Age Ageing 2004; 33(2):116–21. 252 Ciccone A, Bonito V, Italian Neurological Society’s Study Group for Bioethics and Palliative Care in Neurology. Thrombolysis for acute ischemic stroke: the problem of consent. Neurol Sci 2001; 22(5):339–51. 253 Fleck LM, Hayes OW. Ethics and consent to treat issues in acute stroke therapy. Emerg Med Clin N Am 2002; 20(3):703–15. 254 Ciccone A. Consent to thrombolysis in acute ischaemic stroke: from trial to practice. Lancet Neurol 2003; 2:375–8. 255 Bateman BT, Meyers PM, Schumacher HC, Mangla S, Pile-Spellman J. Conducting stroke research with an exception from the requirement for informed consent. Stroke 2003; 34(5):1317–23. 256 Demarquay G, Derex L, Nighoghossian N, Adeleine P, Philippeau F, Honnorat J et al. Ethical issues of informed consent in acute stroke: analysis of the modalities of consent in 56 patients enrolled in urgent therapeutic trials. Cerebrovasc Dis 2005; 19(2):65–8. 257 Kane I, Lindley R, Lewis S, Sandercock P. Impact of stroke syndrome and stroke severity on the process of consent in the Third International Stroke Trial. Cerebrovasc Dis 2006; 21(5–6):348–52. 258 Rosenbaum JR, Bravata DM, Concato J, Brass LM, Kim N, Fried TR et al. Informed consent for thrombolytic therapy for patients with acute ischemic stroke treated in routine clinical practice. [see comment]. Stroke 2004; 35(9):e353–e355. 259 Johnson DM, Kramer DC, Cohen E, Rochon M, Rosner M, Weinberger J. Thrombolytic therapy for acute stroke in late pregnancy with intra-arterial recombinant tissue plasminogen activator. Stroke 2005; 36(6):e53–e55. 260 Wein TH, Hickenbottom SL, Morgenstern LB, Demchuk AM, Grotta JC. Safety of tissue plasminogen activator for acute stroke in menstruating women. Stroke 2002; 33(10):2506–8. 261 Chalela JA, Katzan I, Liebeskind DS, Rasmussen P, Zaidat O, Suarez JI et al. Safety of intra-arterial thrombolysis in the postoperative period. Stroke 2001; 32(6):1365–9. 262 Kane I, Sandercock P, Thomas B. Can patients with unruptured intracranial aneurysms be treated with thrombolysis? A short systematic review. Cerebrovasc Dis 2005; 20(1):51–2. 263 Talkad A, Mathews M, Honings D, Jahnel J, Wang D. Reversal of warfarin-induced anticoagulation with factor VIIa prior to rt-PA in acute stroke. Neurology 2005; 64(8):1480–1.

..

..

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 701

References

264 Linfante I, Reddy AS, Andreone V, Caplan LR, Selim M, Hirsch JA. Intra-arterial thrombolysis in patients treated with warfarin. Cerebrovasc Dis 2005; 19(2):133–5. 265 Tanne D, Kasner ES, Demchuk AM, Koren-Morag N, Hanson S, Grond M et al. Markers of increased risk of intracerebral hemorrhage after intravenous recombinant tissue plasminogen activator therapy for acute ischemic stroke in clinical practice: the multicenter rt-PA acute stroke survey. Circulation 2002; 105:1679–85. 266 Kwan J, Hand P, Sandercock P. Improving the efficiency of delivery of thrombolysis for acute stroke: a systematic review. Q J Med 2004; 97(5):273–9. 267 Hacke W. Third European Acute Stroke Study (ECASS-III): Placebo controlled trial of alteplase (rt-PA) in acute ischemic hemispheric stroke where thrombolysis is initiated between 3 and 4 hours after stroke onset. http://www strokecenter org/trials/trialDetail aspx?tid=475&search_string=ECASS 2005; available from http://www.strokecenter.org/trials/trialDetail.aspx? tid=475&search_string=ECASS. 268 Arnold M, Schroth G, Nedeltchev K, Loher T, Remonda L, Stepper F et al. Intra-arterial thrombolysis in 100 patients with acute stroke due to middle cerebral artery occlusion. Stroke 2002; 33(7):1828–33. 269 Fernandes HM, Gregson B, Siddique S, Mendelow AD. Surgery in intracerebral hemorrhage. The uncertainty continues. Stroke 2000; 31(10):2511–16. 270 Mendelow AD, Gregson BA, Fernandes HM, Murray GD, Teasdale GM, Hope DT et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. [see comment]. Lancet 2005; 365(9457):387–97. 271 Hill MD, Lye T, Moss H, Barber PA, Demchuk AM, Newcommon NJ et al. Hemi-orolingual angioedema and ACE inhibition after alteplase treatment of stroke. Neurology 2003; 60(9):1525–7. 272 Engelter ST, Fluri F, Buitrago-Tellez C, Marsch S, Steck AJ, Ruegg S et al. Life-threatening orolingual angioedema during thrombolysis in acute ischemic stroke. J Neurol 2005; 252(10):1167–70. 273 Montalescot G, Barragan P, Wittenberg O, Ecollan P, Elhadad S, Villain P et al. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001; 344(25):1895–903. 274 De Luca G, Suryapranata H, Stone GW, Antoniucci D, Tcheng JE, Neumann FJ et al. Abciximab as adjunctive therapy to reperfusion in acute ST-segment elevation myocardial infarction: a meta-analysis of randomized trials. J Am Med Assoc 2005; 293(14):1759–65. 275 Abciximab Emergent Stroke Treatment Trial Investigators. Emergency administration of abciximab for treatment of patients with acute ischemic stroke: results of a randomized phase 2 trial. Stroke 2005; 36(4):880–90. 276 Ringleb PA. Thrombolytics, anticoagulants, and antiplatelet agents. Stroke 2006; 37(2):312–13.

..

..

277 Bukow SC, Daffertshofer M, Hennerici MG, Bukow SC, Daffertshofer M, Hennerici MG. Tirofiban for the treatment of ischaemic stroke. Expert Opin Pharmacother 2006; 7(1):73–9. 278 Velianou JL, Strauss BH, Kreatsoulas C, Pericak D, Natarajan MK. Evaluation of the role of abciximab (Reopro) as a rescue agent during percutaneous coronary interventions: in-hospital and six-month outcomes. [see comments]. Cathet Cardiovasc Interv 2000; 51(2):138–44. 279 Fiorella D, Albuquerque FC, Han P, McDougall CG. Strategies for the management of intraprocedural thromboembolic complications with abciximab (ReoPro). Neurosurgery 2004; 54(5):1089–97. 280 Arab D, Yahia AM, Qureshi AI. Use of intravenous abciximab as adjunctive therapy for carotid angioplasty and stent placement. Int J Cardiovasc Interv 2003; 5(2):61–6. 281 Chan AW, Yadav JS, Bhatt DL, Bajzer CT, Gum PA, Roffi M et al. Comparison of the safety and efficacy of emboli prevention devices versus platelet glycoprotein IIb/IIIa inhibition during carotid stenting. Am J Cardiol 2005; 95(6):791–5. 282 Velat GJ, Burry MV, Eskioglu E, Dettorre RR, Firment CS, Mericle RA et al. The use of abciximab in the treatment of acute cerebral thromboembolic events during neuroendovascular procedures. Surg Neurol 2006; 65(4):352–8. 283 Sherman DG, Atkinson RP, Chippendale T, Levin KA, Ng K, Futrell N et al. Intravenous ancrod for treatment of acute ischemic stroke: the STAT study: a randomized controlled trial. Stroke Treatment with Ancrod Trial. J Am Med Assoc 2000; 283(18):2395–403. 284 Liu M, Counsell C, Zhao XL, Wardlaw J. Fibrinogen depleting agents for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000091; PMID: 10796295]. Cochrane Database Syst Rev 2003; (3):CD000091. 285 Hennerici MG, Kay R, Bogousslavsky J, Lenzi GL, Verstraete M, Orgogozo JM. Intravenous ancrod for acute ischaemic stroke in the European Stroke Treatment with Ancrod Trial: a randomised controlled trial. Lancet 2006; 368(9550):1871–8. 286 Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. [see comment]. Lancet 2003; 361(9351):13–20. 287 Daffertshofer M, Hennerici M. Ultrasound in the treatment of ischaemic stroke. Lancet Neurol 2003; 2(5):283–90. 288 Furlan AJ, Fisher M. Devices, drugs, and the Food and Drug Administration: increasing implications for ischemic stroke. Stroke 2005; 36(2):398–9. 289 Noser EA, Shaltoni HM, Hall CE, Alexandrov AV, Garami Z, Cacayorin ED et al. Aggressive mechanical clot disruption: a safe adjunct to thrombolytic therapy in acute stroke? Stroke 2005; 36(2):292–6. 290 Smith WS, Sung G, Starkman S, Saver JL, Kidwell CS, Gobin YP et al. Safety and efficacy of mechanical embolectomy in acute ischemic stroke: Results of the MERCI trial. Stroke 2005; 36(7):1432–8.

701

9781405127660_4_012.qxd 10/16/07 10:47 AM Page 702

702

Chapter 12 Specific treatments for acute ischaemic stroke

291 Becker KJ, Brott TG. Approval of the MERCI clot retriever: a critical view. Stroke 2005; 36(2):400–3. 292 Kidwell CS, Jahan R, Starkman S, Saver J. MR and recanalization of stroke clots using embolectomy (MR RESCUE). Stroke 2005; International Stroke Conference 2005. Ongoing Clinical Trials Abstracts. 293 Tomsick TA. Mechanical embolus removal: a new day dawning. [comment]. Stroke 2005; 36(7):1439–40. 294 Alexandrov AV, Molina CA, Grotta JC, Garami Z, Ford SR, varez-Sabin J et al. Ultrasound-enhanced systemic thrombolysis for acute ischemic stroke. [see comment]. NEJM 2004; 18; 351:2170–8. 295 Qizilbash N, Lewington SL, Lopez-Arrieta JM, Qizilbash N, Lewington SL, Lopez-Arrieta JM. Corticosteroids for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000064; PMID: 10796290]. Cochrane Database Syst Rev 2002; (2):CD000064. 296 Righetti E, Celani MG, Cantisani T, Sterzi R, Boysen G, Ricci S. Glycerol for acute stroke. [update of Cochrane Database Syst Rev 2000; (4):CD000096; PMID: 11034673]. Cochrane Database Syst Rev 2004; (2):CD000096. 297 Bereczki D, Liu M, Prado GF, Fekete I. Cochrane report: A systematic review of mannitol therapy for acute ischemic stroke and cerebral parenchymal hemorrhage. [see comment]. Stroke 2000; 31(11):2719–22. 298 Roberts I, Schierhout G, Wakai A. Mannitol for acute traumatic brain injury. [update of Cochrane Database Syst Rev 2000; (2):CD001049; PMID: 10796744]. Cochrane Database Syst Rev 2003; (2):CD001049. 299 Georgiadis D, Schwarz S, Aschoff A, Schwab S. Hemicraniectomy and moderate hypothermia in patients with severe ischemic stroke. Stroke 2002; 33(6):1584–8. 300 Morley NC, Berge E, Cruz-Flores S, Whittle IR. Surgical decompression for cerebral oedema in acute ischaemic stroke. Cochrane Database Syst Rev 2002; (3):CD003435. 301 Frank J. HeaDDFIRST. Hemicraniectomy and Durotomy for Deterioration From Infarction Relating Swelling Trial. 2000; available from http://www.strokecenter.org/trials/ TrialDetail.aspx?tid=70. 302 The DESTINY Study Group. DESTINY – DEcompressive Surgery for the Treatment of malignant INfarction of the middle cerebral arterY: Preliminary Results. Cerebrovasc Dis 2006; 21 (Suppl 4):59–60. 303 Vahedi K, Vicaut E, Blanquet A, Payen D, Bousser MG. DECIMAL trial: DEcompressive Craniectomy In MALignant Middle Cerebral Artery Infarcts. 2004; available from http://www.strokecenter.org/trials/ TrialDetail.aspx?tid=564. 304 Hofmeijer J, Amelink GJ, Algra A, van GJ, Macleod MR, Kappelle LJ et al. Hemicraniectomy after middle cerebral artery infarction with life-threatening Edema trial (HAMLET). Protocol for a randomised controlled trial of decompressive surgery in space-occupying hemispheric infarction. Trials 2006; 7:29. 305 Schierhout G, Roberts I. Hyperventilation therapy for acute traumatic brain injury. Cochrane Database Syst Rev 2000; (2):CD000566.

306 Fisher M, Stroke Therapy Academic Industry Roundtable. Recommendations for advancing development of acute stroke therapies: Stroke Therapy Academic Industry Roundtable 3. Stroke 2003; 34(6):1539–46. 307 Fisher M. Enhancing the development and approval of acute stroke therapies Stroke Therapy Academic Industry Roundtable. Stroke 2005; 36:1808–13. 308 Sandercock P, Roberts I. Systematic reviews of animal experiments. Lancet 2002; 360(9333):586. 309 Pound P, Ebrahim S, Sandercock P, Bracken MB, Roberts I, Reviewing Animal Trials Systematically (RATS) Group. Where is the evidence that animal research benefits humans? [see comment]. Br Med J 2004; 328(7438):514–17. 310 Macleod MR, O’Collins T, Howells DW, Donnan G. Pooling of animal experimental data reveals influence of study design and publication bias. Stroke 2004; 35(5):1203–8. 311 Macleod MR, Ebrahim S, Roberts I. Surveying the literature from animal experiments: systematic review and metaanalysis are important contributions. [comment]. Br Med J 2005; 331(7508):110. 312 Lees KR, Zivin JA, Ashwood T, Davalos A, Davis SM, Diener HC et al. NXY-059 for acute ischemic stroke. N Engl J Med 2006; 354(6):588–600. 313 Renovis Inc. NXY-059 does not meet efficacy endpoints in phase III trial for acute ischemic stroke. 2006; available from: http://www.medicalnewstoday.com/ medicalnews.php?newsid=55212. 314 Asplund K. Haemodilution for acute ischaemic stroke. [update of Cochrane Database Syst Rev 2000; (2):CD000103; PMID: 10796299]. Cochrane Database Syst Rev 2002; (4):CD000103. 315 Horn J, Limburg M. Calcium antagonists for ischemic stroke. A systematic review. Stroke 2001; 32:570–6. 316 Cochrane Handbook for Systematic Reviews of Interventions 4.2.5 [updated May 2005]. In: Higgins JPT, Green S, editors. In: The Cochrane Library, Issue 3, 2005. Chichester, UK: John Wiley & Sons, Ltd; 2005. 317 Altman DG, Machin D, Bryant TN, Gardner M. Statistics with Confidence. 2nd ed. London: BMJ Books; 2000. 318 Ludlam C, Bennett B, Fox KA, Lowe G, Reid A. Guidelines for the use of thrombolytic therapy. Haemostasis and Thrombosis Task Force of the British Committee for Standards in Haematology. Blood Coag Fibrin 1995; 6(3):273–85. 319 Plum F, Posner JB. The Diagnosis of Stupor and Coma, Philadelphia, PA: F.A. Davis & Co., 1985. 320 Sherman DG, Albers GW, Bladin C, Fieschi C, Gabbai AA, Kase CS et al. The efficacy and safety of enoxaparin versus unfractionated heparin for the prevention of venous thromboembolism after acute ischaemic stroke (PREVAIL Study): an open-label randomised comparison. Lancet 2007; 369:1347–55. 321 Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A et al. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials. Lancet Neurol 2007; 6:215–22.

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 703

13

Specific treatment of intracerebral haemorrhage 13.1 Pathophysiology

703

13.2 Prognosis and predictive models 13.3 Management

703

704

13.1 Pathophysiology

The events following non-traumatic spontaneous intracerebral haemorrhage (ICH) have been most intensively studied for the most common type, rupture of one or more deep perforating arteries. These events are surprisingly complex, to begin with because bleeding from a single site often leads adjacent perforators to rupture as well (section 8.2.1). Furthermore, the extravasated blood inevitably causes disruption of white matter tracts and irreversible damage to neurones in the deep nuclei or cortex. The protective encasement of the skull may become a disadvantage with any sudden increase in volume within the intracranial cavity, as with other causes of brain swelling and space-occupying lesions. Apart from the brain tissue destroyed by the haemorrhage itself, the attendant increase in intracranial pressure may threaten other parts of the brain, particularly – but not exclusively – when the intracranial pressure reaches levels of the same order of magnitude as the arterial pressure, bringing the cerebral perfusion pressure close to zero.1 Direct mechanical compression of the brain tissue surrounding the haemorrhage and, to some extent, vasoconstrictor and pro-inflammatory substances in extravasated blood also lead to impaired blood supply.2,3 Cellular ischaemia leads to further swelling from oedema,4,5 which is initially cytotoxic and later vasogenic (section 12.1.5). The zone of ischaemia around the haemorrhage may swell through systemic factors such as hypotension or hypoxia. Often there is also loss

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

of cerebral autoregulation in the vasculature supplying the region of the haemorrhage. Some perifocal ischaemic damage occurs at the time of bleeding and cannot be prevented, but the question to be considered here is whether the vicious cycle of ongoing ischaemia causing steadily increasing pressure can be interrupted in its early stages. Hydrocephalus may be an additional space-occupying factor.6 This complication is especially likely to occur with cerebellar haemorrhages, but a large haemorrhage in the region of the basal ganglia may also cause enlargement of the ventricular system, by rupture into the third ventricle, or through dilatation of the opposite lateral ventricle, with midline shift and obstruction of the third ventricle, while the ipsilateral ventricle is compressed.

13.2 Prognosis and predictive models

According to population-based studies the short-term prognosis of patients with ICH is poor, with over 40% dying within the first month.7,8 Several hospital-based studies have attempted to develop models to predict prognosis based on patients’ characteristics on admission, but almost without exception these have not been externally validated (in independent data sets). An accurate prognostic model might theoretically help in treatment decisions, particularly in selecting which patients should benefit from intensive care, and in informing relatives about the chances of recovery. The most recent systematic review selected 18 hospitalbased studies performed before 2004 that met essential criteria (multivariate analysis with logistic regression; no surgical treatment; paper in one of four major European languages); the models proposed in these studies were 703

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 704

704

Chapter 13 Specific treatment of intracerebral haemorrhage

then validated against the outcome in 122 other patients with intracerebral haemorrhage admitted to a single centre.9 Most prediction models proved easy to apply and most could generate a high probability of death or dependency in patients with the combination of factors providing the highest probability of poor outcome (most often the level of consciousness on admission, age, volume of haemorrhage and intraventricular extension). Nevertheless, among the 10 prognostic models that could identity patients with a greater than 90% probability of death or poor outcome, only a small proportion of patients actually had this combination of unfavourable characteristics (5–48%, median 30%). Also the accuracy of the prediction of 30-day case fatality was far from ideal: in these 10 subsets of the validation series the case fatality ranged between 67 and 100% (median 89%);9 similarly inaccurate was a later predictive model, the so-called Essen ICH score.10 Early withdrawal of support may then result in a self-fulfilling prophecy without scientific foundation.11 Clearly characteristics on admission alone, used in the presently available prognostic models, do not allow accurate prediction of outcome in a sufficient proportion of patients to be clinically useful. Single-centre studies published after the systematic review cited above claimed that high levels of D-dimer or troponin, or low levels of cholesterol or triglycerides in serum, are independently associated with poor outcome;12–14 obviously these findings require confirmation. Early characteristics predicting the prognosis for survival of patients with intracerebral haemorrhage are: level of consciousness; age; volume of haemorrhage and intraventricular extension of haemorrhage, but so far predictive models apply to only a small proportion of patients and are not sufficiently accurate for clinical decision making. One way to improve models to predict poor outcome is to restrict the analysis to specific subgroups of ICH, according to location or presumed cause. This has been attempted for lobar haemorrhages,15 haemorrhages in the caudate nucleus16 and pontine haemorrhages,17 but again there has been no external validation. Perhaps a more fruitful approach is to move away from characteristics on first admission and to base predictive models more on the clinical course in the first 24 h,9 if only because expansion of the haemorrhage occurs in a substantial proportion of patients within that period (section 8.2.1); not surprisingly this is an adverse factor.18 Additional determinants in the early clinical course that probably should be included in the equation are fever19 and the development of oedema around the haemorrhage.20

13.3 Management

13.3.1 Initial management Admission to a specialized stroke unit is beneficial, as for stroke patients in general.21 Once the patient’s ventilation and circulation have been secured and the diagnosis of intracerebral haemorrhage confirmed by computed tomography (CT) scanning or magnetic resonance imaging (MRI), the next step is often to identify the cause before any specific treatment is initiated (Chapter 8). For example, if surgical evacuation of a lifethreatening haemorrhage is considered at all, medical measures may have to be taken first (such as substitution of clotting factors in patients with liver failure or on anticoagulants). Also the surgeon should think twice if amyloid angiopathy is a likely cause (section 13.3.7). Repeat CT scanning is indicated if there is clinical deterioration, which may be caused by rebleeding at the same or sometimes at a distant site, or by hydrocephalus; if there are no structural changes, the search for systemic disorders should be intensified (section 11.5). Recombinant factor (rF) VIIa is known to decrease the severity of haemorrhage in certain surgical settings, also in patients without clotting abnormalities.22 A Cochrane review of four phase II randomized clinical trials involving adults aged 18 years or over, within 4 h of ICH found a reduced relative risk of death or dependence on the modified Rankin scale within 90 days of disease onset; 0.79, 95% confidence interval (CI) 0.67–0.93 (Fig. 13.1), but not on the extended Glasgow Outcome Scale; 0.90 (95% CI, 0.81–1.01).23 There was a statistically significant excess of arterial thromboembolism at 160 µg/kg rFVIIa. In conclusion, adults with acute ICH may benefit from haemostatic therapy with rFVIIa, but the evidence on major clinical outcomes is neither robust nor precise. Claims that this treatment is costeffective are based on the point estimate of benefit and are therefore premature.24 The blood pressure is usually increased in patients with ICH, through pre-existing hypertension, a response to a sudden increase in intracranial pressure, or both these factors.25 There are theoretical arguments in favour of decreasing the blood pressure (in the hope of stopping ongoing bleeding from ruptured small arteries),26 as well as in favour of increasing it further (in the hope of salvaging marginally perfused areas of brain that are compressed around the haemorrhage). In the absence of even the flimsiest evidence supporting one or other point of view (not surprisingly, a randomized

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 705

13.3 Management

Study rFVIla EurAsia llA rFVIla llB rFVIla USA llA

Relative risk (fixed) 95% CI

rFVIla n/N

Placebo n/N

16/36

5/11

6.7

0.98 [0.46, 2.06]

160/303

66/96

87.7

0.77 [0.65, 0.91]

15/32

4/8

5.6

0.94 [0.43, 2.06]

100.0

0.79 [0.67, 0.93]

Total (95% CI) 371 115 Total events: 191 (rFVIla). 75 (Placebo) Test for heterogeneity chi-square = 0.61 df = 2 P = 0.74 I 2 = 0.0% Test for overall, effect Z = 2.77 P = 0.006 0.1 0.2 0.5 Favours rFVIla Fig. 13.1 Meta-analysis of recombinant factor VIIa for intracerebral haemorrhage. Each trial is represented by a horizontal line, the length of which describes the 95% confidence interval around the point estimate of the treatment effect represented by a square, the size of which is proportional

trial with 50 patients found no difference in outcome),27 the only rational course is to leave the blood pressure alone in the acute phase unless it is so high that organ damage develops (especially in the retina and kidney). This is essentially the same for patients with ischaemic stroke (section 11.7.1). This is not to say that the patient should be discharged with a high blood pressure; a single observational study supports the notion that the longterm risk not only of vascular events in general but also of rebleeding may be lowered by adequate long-term blood pressure control.28

13.3.2 Reduction of intracranial pressure Apart from the local effects of the haemorrhage itself and the subsequent perifocal oedema, other factors also contribute to raised intracranial pressure. Frequently, these can be treated more directly than the cerebral lesion itself, so it is important to be aware of them. They include fever, hypoxia, hypertension, seizures and elevations of intrathoracic pressure. A vexing and unsolved question is whether it is useful to measure the intracranial pressure by the introduction of an intraventricular catheter or other device, at least in patients with a decreased level of consciousness. Such an approach can certainly be rationalized.29 On the other hand, any invasive procedure may cause complications,30 while the presumed advantages in terms of outcome have not been subjected to controlled studies. For what it is worth, an observational comparison in traumatic brain injury failed to find any advantage of management based on intracranial pressure monitoring.31 We prefer to err on the side of caution and base management on clinical and radiological features rather than the ‘pressure’ as measured directly.

..

..

1

Weight (%)

Relative risk (fixed) 95% CI

2 5 10 Favours placebo to the amount of data in the trial. The diamond represents the overall treatment effect, and its width the 95% confidence interval. The outcome is death or dependency (modified Rankin scale 4–6) at day 90.23

Cerebrospinal fluid drainage Insertion of a catheter in the lateral ventricle may be a definitive treatment in patients with cerebellar haemorrhage and no signs of direct compression of the brainstem (section 13.3.6). But, in patients with supratentorial haemorrhage and hydrocephalus, the benefits of CSF diversion are decidedly uncertain. In a series of 22 patients with supratentorial ICH and hydrocephalus who were treated with ventriculostomy, intracranial pressure was controlled at < 20 mmHg in 20, but only a single patient showed any improvement in not only the degree of hydrocephalus but also in the level of consciousness.32 Just three patients, with small haemorrhage volumes, survived to 3 months. An alternative strategy is to divert the CSF internally, through callosotomy and fenestration of the septum pellucidum.33 Controlled studies for all these procedures are woefully lacking. In patients with extensive intraventricular haemorrhage secondary to deep intraparenchymal or aneurysmal rupture, drainage of CSF is also performed in some centres, in an attempt to improve the dismal prognosis. Systematic reviews of observational studies suggest that this procedure may be helpful, especially when it is combined with instillation of fibrinolytic drugs, but only direct comparisons in a randomized trial can definitively establish the value of these therapeutic measures.34,35 Later observational studies have not solved the problem.36,37 Hyperventilation There is no doubt that hyperventilation decreases intracranial pressure, but at the same time there is no controlled evidence to allay concerns that the cure may

705

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 706

706

Chapter 13 Specific treatment of intracerebral haemorrhage

be worse than the disease. The intracranial pressure goes down because hypocapnia (usually down to values of the order of 4 kPa) causes vasoconstriction; in other words, ischaemia by compression is exchanged for ischaemia by vasoconstriction.38 In head-injured patients, a single and so far the only randomized controlled trial of prolonged hyperventilation not only failed to show any benefit but also raised concerns about potential harm.39,40 For patients with ICH, no controlled trials of hyperventilation have been done, but the experience with head trauma is not encouraging. Provisionally it seems best to reserve this treatment for bridging a few hours in patients for whom surgical evacuation is planned (sections 13.3.5 and 13.3.6). Osmotic agents Osmotic agents include mannitol, glycerol and sorbitol; they are administered by intravenous infusion.41 Iodidecontaining contrast agents used in CT scanning or angiography may inadvertently have the same dehydrating effect.42 Osmotic agents extract water from the extracellular into the intravascular compartment of the brain. Whether they dehydrate mainly normal instead of damaged brain tissue is a matter of debate.41 A disturbed blood–brain barrier may facilitate entry of hyperosmotic agents into brain tissue, thereby decreasing the osmotic gradient needed for effective dehydration, but on the other hand it may facilitate movement of fluid from brain tissue to blood. The interval between the infusion and the decrease of intracranial pressure is usually less than 1 h, but the effect of a single dose lasts no more than 4–6 h because the concentration of the solute becomes equilibrated between the intravascular and extracellular compartments. Potential dangers of this treatment include hypotension, hypokalaemia, renal failure from hyperosmolality, haemolysis and congestive heart failure.43 Intravenous mannitol (20–25% solution) has become the mainstay of osmotic therapy in many centres, despite a long-standing lack of evidence from controlled studies.44 Recently a clinical trial in India randomized 128 patients with any type of supratentorial haemorrhage within 6 days and found no difference in case fatality at 1 month.45 The correct dose cannot be predicted, but a safe regimen is to give 0.75–1 g/kg initially, then 0.25–0.5 g/kg every 3–5 h, depending on clinical response and osmolality (the aim being 295–310 mosm/L).46 A controlled trial of a high dose (1.4 mg/kg) against a standard dose (0.7 mg/kg) of mannitol in 141 preoperative patients with traumatic temporal lobe haemorrhages and ipsilateral pupillary dilatation found better outcome after 6 months in the high-dose group (61% vs 33%),47 but the relevance of this trial is

questionable for patients with spontaneous intracerebral haemorrhages who are managed conservatively. With mannitol infusions, at the very least the central venous pressure should be monitored, and kept between 5 and 12 mmHg, to prevent hypovolaemia. For intravenous glycerol, meta-analysis of one minute trial (eight patients) and one moderately-sized controlled trial in patients with intracerebral haemorrhage failed to find evidence of benefit.43,48 Hypertonic saline is apparently ‘emerging’ in neurosurgical practice as an alternative to other agents,49 but no controlled studies are available. Corticosteroids Corticosteroids reduce peritumoral oedema and at first sight it seems rational to expect a similar effect on oedematous brain tissue around an intracerebral haemorrhage. To date there have been four relevant but small trials, which have been summed up in a Cochrane review.50 At 1 month after randomization 57 of 92 patients (62%) allocated to glucocorticoid treatment had died, compared with 50 of 94 patients (53%) allocated to the control group; relative risk 1.14 (95% CI, 0.91–1.42). Treatment with glucocorticoids did not reduce poor outcome at 1 month after randomization; relative risk, 0.95 (95% CI, 0.83–1.09; Fig. 13.2). The risk of any adverse effects in the treatment group (13%) was higher than that in the control group patients (8.8%) but the confidence interval was very wide; relative risk, RR 1.48 (95% CI, 0.87–2.51). The most common adverse effects were infections (23%), exacerbation of diabetes mellitus (11%) and gastrointestinal bleeding (5%) but none of them reached a statistically significant difference between the treatment and control groups. Thus, any benefit in terms of neurological function is moderate at best, whereas the adverse effects of corticosteroids in these patients are clearly substantial.50 Mannitol is widely used in patients with intracerebral haemorrhage and a depressed level of consciousness, to decrease intracranial pressure and alleviate the space-occupying effect of the haemorrhage in a deteriorating patient, although this custom is not backed up by randomized trials with clinical status rather than pressure as the outcome variable. With corticosteroids the benefits have so far not been proved to outweigh the disadvantages.

13.3.3 Prevention of deep venous thrombosis and pulmonary embolism (See also section 11.13.) Patients with ICH have a substantial risk of deep venous thrombosis (DVT):

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 707

13.3 Management

Study

Treatment n/N

Control n/N

Desai 1998

10/12

Ogun 2001 Poungvarin 1987

Relative risk (fixed) 95% CI

Weight (%)

Relative risk (fixed) 95% CI

12/14

17.3

0.97 [0.70, 1.35]

14/15

11/12

19.1

1.02 [0.82, 1.27]

37/46

41/47

63.5

0.92 [0.77, 1.10]

100.0

0.95 [0.83, 1.09]

At one month

73 73 Subtotal (95% CI) Total events: 61 (Treatment), 64 (Control) Test for heterogeneity chi-square = 0.52 df = 2 P = 0.77 I 2 = 0.0% Test for overall effect Z = 0.76 P = 0.4 At six months Subtotal (95% CI) 0 0 Total events: 0 (Treatment), 0 (Control) Test for heterogeneity: not applicable Test for overall effect: not applicable Total (95% CI) 73 73 Total events: 61 (Treatment), 64 (Control) Test for heterogeneity chi-square = 0.52 df = 2 P = 0.77 I 2 = 0.0% Test for overall effect Z = 0.76 P = 0.4

0.0

100.0

Not estimable

0.95 [0.83, 1.09]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control Fig. 13.2 Glucocorticoid treatment vs controls in intracerebral haemorrhage. The same conventions as in Fig. 13.1. The outcome is ‘poor outcome at the end of follow up’.50

between 30% and 70%, depending on the severity of the stroke and the degree of immobilization, and on how hard one looks for it. The risk of pulmonary embolism has not been studied specifically for patients with haemorrhagic stroke, but extrapolation from what is known for ischaemic stroke or stroke in general the estimate is 1–2% in the first month (section 11.13).51,52 However, physicians are intuitively cautious in using antithrombotic drugs in patients with intracerebral haemorrhages. On the other hand, drug regimens that prevent clotting as a consequence of venous stasis do not necessarily increase the risk of a small artery in the brain rupturing for a second time. The definitive answer should of course come from clinical trials, not theoretical considerations. Subcutaneous heparin Subcutaneous heparin, usually given as 5000 U two or three times a day, reduces the frequency of DVT by 68%, at least after general, orthopaedic and urological surgery.53 In a small, but so far unique, clinical trial in 46 patients with ICH this regimen was applied 4 days after the onset in the active treatment group and 10 days after the haemorrhage in the control group.54 In the control group 40% (9 of 23) developed some evidence of pulmonary embolism (as detected by perfusion scintigraphy of the lungs) against 22% (5 of 22) in the treated group. In a subsequent non-randomized study, in 68 patients with ICH, treatment was started on day 2, 4 or 10 after the haemorrhage.55 For what it is worth, the risk of

..

..

pulmonary embolism was significantly lower in the group with treatment started on day 2, without a concomitant increase in rebleeding. But, as the group with early treatment was studied after the two other dosage groups, the results may well have been influenced by factors other than the interval until the initiation of treatment. Aspirin Aspirin is effective in reducing the risk of DVT (by approximately 39%) as well as pulmonary embolism in an overview of all trials in the perioperative period and in medical patients at increased risk (section 11.13). In a systematic review of all published trials comparing any antithrombotic agent with control among patients with any form of intracranial haemorrhage, there were two randomized trials in which aspirin had been inadvertently given (for variable periods) to 398 patients with acute ICH (and placebo to 375 patients).56 The odds ratio (OR) for these ICH patients treated with aspirin was 0.96 (95% CI, 0.62–1.5) for death, and 1.02 (95% CI, 0.5–1.8) for recurrent haemorrhage. These scant data do not support reliable conclusions about the safety or otherwise of antithrombotic agents in patients with acute intracerebral haemorrhage. Compression stockings and intermittent pneumatic compression Physical methods are of course likely to be the safest method of thrombosis prevention in patients with ICH,

707

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 708

708

Chapter 13 Specific treatment of intracerebral haemorrhage

although occasionally they may cause pressure sores (section 11.13). Essentially there are two methods: intermittent pneumatic compression, and graduated compression stockings which vary the pressure from being highest at the ankle and decreasing proximally. A systematic review of 16 trials (totalling more than 2200 patients) of graduated compression stockings in operated or otherwise immobilized patients for a variety of conditions found an odds ratio for deep venous thrombosis of 0.34 (95% CI, 0.25–0.46) if graduated pressure stockings was the only treatment, and of 0.24 (95% CI, 0.15–0.37) if they were used on a background of other prophylactic measures.57 In stroke patients alone, two small trials (123 patients in all) could not demonstrate any such effect.58 A later trial tested intermittent pneumatic compression, specifically in 151 patients with ICH; all were also fitted with graduated elastic stockings.59 On day 10, DVT (as assessed by compression ultrasonography) was found in 5% of the patients in the group with intermittent compression, against 16% in those with stockings alone, corresponding to a relative risk of 0.29 (95% CI, 0.08–1.00), a marginally significant difference. Pulmonary embolism was not included in these systematic reviews, but there is no reason to assume that the relationship with DVT is different in patients with ICH. Also a significantly protective effect of compression stockings on the risk of pulmonary emboli has been demonstrated in a large trial of patients who underwent cardiothoracic operations.60 Thus, unless the disadvantages of heparin are proved to be merely theoretical, application of graduated compression stockings, preferably supplemented with intermittent pneumatic compression, seems to offer the greatest protection and the smallest risk, although the method is fairly labour intensive and the stockings can be uncomfortable after a time. To resolve the existing uncertainties, a multicentre trial of graduated compression stockings has been initiated in the UK (section 11.13). Subcutaneous heparin and the application of graduated compression stockings both decrease the risk of deep venous thrombosis in bedridden patients by approximately 70%, and aspirin by approximately 39%, as judged from clinical trials in a variety of postoperative conditions. Because the safety of heparin and aspirin is largely unknown in patients with intracerebral haemorrhage, compression stockings are presently the preferred method of prophylaxis, despite being labour intensive, and despite the lack of direct evidence in patients with primary intracerebral haemorrhage.

13.3.4 Treatment of iatrogenic intracerebral haemorrhage Anticoagulants Anticoagulants are associated with an increased risk of intracerebral haemorrhage (section 8.4.1). It seems rational to reverse the deficiency of clotting factors as quickly as possible and to accept the risk of a thrombotic event, especially in view of the frequent progression of clinical deficits that occurs in these patients. Understandably the speed of treatment has never been put to the test in a clinical trial, but observational data confirm that early normalization of clotting status is associated with a relatively low rate of haemorrhage enlargement.61 In current practice the first step is i.v. 10–20 mg vitamin K, at not more than 5 mg/min, rapidly followed by infusion of prothrombin complex concentrate (PCC), which contains the coagulation factors II, VII, IX and X;62 every half hour counts in halting progression of the haemorrhage.63 Infusion of these factors alone restores the coagulation system more rapidly than whole plasma and is safer from the point of view of transmission of virus particles (section 12.4.7).64–66 When to resume anticoagulants in patients with a strong indication for this treatment (e.g. those with mechanical heart valves) is a problem many clinicians will recognize. Only anecdotal experience is available, such as a series from the Mayo Clinic in which only one ischaemic stroke occurred in 52 patients with mechanical heart valves in whom anticoagulants were discontinued for a median period of 10 days,67 in contrast to seven similar patients from Heidelberg in three of whom large brain infarcts occurred within 4 days of normalization of the international normalized ratio (INR).68 The essential issue in deciding when to restart is probably not when blood has disappeared from the CT scan (that may take weeks or months), but when the ruptured vessel can be assumed to have sufficiently healed. Between 1–2 weeks after the haemorrhage – depending on the stringency of the indication – is probably not a bad guess.69 No rebleeds were reported with such a regimen, although the numbers were small.70,71 Thrombolytic therapy Thrombolytic therapy after myocardial infarction is complicated by intracerebral haemorrhage in only a small proportion (section 8.4.3), but the case fatality is high. This grim prognosis warrants attempts at intervention, even without the benefits of controlled clinical trials. These measures include control of any hypertension and infusion of coagulation factors; the use of antifibrinolytic drugs is controversial even from

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 709

13.3 Management

a theoretical point of view. Surgical treatment is hazardous, given that amyloid angiopathy is often a contributing factor (section 13.3.7). Life-threatening arrhythmias may occur as a complication of the myocardial infarct for which the treatment was given, often before the onset of the neurological symptoms, and these should of course be promptly treated.72 In patients with ischaemic stroke who develop major intracerebral haemorrhage after thrombolytic treatment (section 12.5.2) there is no evidence about what to do. In view of the little we know about operative treatment for intracerebral haemorrhage in general (section 13.3.5) it seems best to leave well alone. The utility of prothrombin complex concentrate or recombinant factor VIIa has not been studied so far. Aspirin Aspirin treatment is associated with a small risk of intracerebral haemorrhage (section 8.4.2). It is reasonable to stop the drug once the diagnosis has been made, although no disasters have been documented in instances in which aspirin treatment was continued, or even restarted. At any rate the antiplatelet effect lasts for several days after discontinuation of the drug.73 In the trials of aspirin for secondary prevention after cerebral ischaemia some patients with small intracerebral haemorrhages must have been inadvertently included before CT scanning established the true diagnosis; there were no obviously untoward effects, but the confidence interval is still wide (section 12.3.3).74 In patients on regular treatment with aspirin the outcome after ICH is relatively poor,75 but this is explained by confounding factors; in a large observational study from Germany, the effect disappeared after adjustment for age and premorbid condition.76

13.3.5 Surgical treatment of supratentorial haemorrhage The frequency of surgical intervention for intracerebral haemorrhage varies widely between centres, and even between participants in a clinical trial.77 There are four possible surgical procedures to treat ICH: simple aspiration, craniotomy with open surgery, endoscopic evacuation, and stereotactic aspiration. Unless specifically indicated otherwise, the assumption is that we are mostly dealing with deep haemorrhages, from degenerative small vessel disease. Simple aspiration Aspiration, not accompanied by any other intervention, was attempted mainly in the 1950s but this was

..

..

subsequently abandoned because only small amounts of clot could be obtained, and because the procedure could precipitate ‘blind’ rebleeding.78,79 Craniotomy Open surgery was the method for removing haematomas in five trials, one of them (with 131 patients) dating from the pre-CT era. A Cochrane review (last updated in 1999) of this early study and two subsequent trials (totalling 64 randomized patients) in the 1980s showed a non-significant trend towards increased risk of death and dependency among survivors (odds ratio 1.99, 99% CI, 0.92–4.31). Two subsequent but small single-centre randomized trials (54 patients altogether) showed results that were entirely within the range of the overview.80,81 A pseudo-randomized trial in Houston tested the optimal time for clot evacuation – assuming that the operation was indicated at all; in the experimental group patients were operated within 4 h of symptom onset, patients operated within 12 h forming the control group. The study was stopped early because of rebleeding in the ‘ultra-early’ group.82 The year 2005 saw publication of a landmark trial (STICH) that has considerably added to the modest information from controlled studies so far.83 A total of 1003 patients were randomized, four times as many as in all previous trials taken together; all had supratentorial haemorrhage. In the group allocated to early operation (craniotomy in 75%) the procedure was scheduled within 24 h, whereas in the other group treatment was initially conservative. A second notable feature was the introduction of a ‘sliding dichotomy’ for assessing outcome, to take account of the methodological problem that expectations about the result of any treatment are higher when the initial condition of the patient is better.84 The result was that the cut-off point for ‘favourable outcome’ depended on three prognostic features at baseline: level of consciousness on the Glasgow Coma Scale, age and volume of the haemorrhage. For 516 patients with a good prognostic profile on admission a favourable outcome included only the categories ‘good recovery’ and ‘moderate disability’ on the extended Glasgow outcome scale, whereas for the other half of the study patients with a poorer prognosis ‘severe disability’ was included in the definition of a favourable outcome.85 The analysis of outcome after 6 months failed to show any benefit of craniotomy: of patients randomized to early surgery, 122 (26%) had a favourable outcome compared with 118 (24%) of those randomized to initially conservative treatment; odds ratio 0.89 (95% CI, 0.66– 1.19). Separate analysis of 12 prespecified subgroups showed that operation was perhaps beneficial in patients with superficial haemorrhages (1 cm or less from the

709

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 710

710

Chapter 13 Specific treatment of intracerebral haemorrhage

cortical surface), but of course this may be chance effect. The efficacy of operative treatment did not significantly differ between patients with deep haemorrhages (approximately 60%) and those with lobar haemorrhage. No systematic review has yet been completed that included this major trial with the five small trials performed previously, but of course such an analysis will not change the current conclusion that – as a routine – craniotomy is not the treatment of choice in patients with intracerebral haemorrhage. A question not addressed by the STICH trial is whether operation improves outcome in patients deteriorating because of haemorrhage expansion. A consecutive series of 26 such patients from the Mayo Clinic (of whom 24 had lobar haemorrhage) reported that no patient regained independence – with or without operation – if corneal and oculocephalic reflexes had been lost, but that approximately onequarter of the remaining group (6 of 21) did regain independence.86 Given the present state of knowledge routine craniotomy and evacuation of the haemorrhage is not indicated for patients with supratentorial haemorrhage, deep or lobar. No subgroup of patients has yet been reliably identified for whom operation might be beneficial.

agents, has been reported in several observational studies involving hundreds of patients with supratentorial haemorrhage; some of these concentrated on haemorrhages with intraventricular extension.34 Subsequently two controlled trials of this technique have been performed. A Japanese trial randomized 242 patients with putaminal haemorrhage and a moderately decreased level of consciousness (eyes closed but opened to stimuli) between stereotactic haemorrhage evacuation and conservative treatment. Outcome, assessed by a blinded observer in terms of independence, was better in the surgical group.91 A smaller trial in the Netherlands (70 patients) combined stereotactic aspiration with liquefaction by means of a plasminogen activating substance; no conclusive differences emerged.92

13.3.6 Surgical treatment of infratentorial haemorrhage Cerebellar haemorrhage Cerebellar haemorrhages (Figs 8.29 and 13.3) have fairly characteristic clinical features (section 3.3.7), with the exception of massive haemorrhages, which are clinically

Endoscopic evacuation There are enthusiasts for almost any conceivable type of operation. Endoscopic evacuation of brain haemorrhage by stereotactic methods is no exception.87,88 The only controlled study so far involved two groups of 50 patients.89 The authors’ interpretation was that surgery had no effect in putaminal and thalamic haemorrhage, but that it was beneficial in subcortical haemorrhages, provided patients are aged 60 years or less and are alert or somnolent. The problem with this analysis is not only that it depends on subgroups, but also no overall analysis was reported and some outcome categories were not reported at all. Recalculation of the results for all patients according to the proportion who were dead or dependent showed an odds ratio of 0.45 in favour of endoscopic operation, but this was not statistically significant (95% CI, 0.19–1.04).90 In conclusion, stereotactic endoscopic evacuation is a promising technique, but its benefits and specific indications remain to be confirmed by wellconducted and large clinical trials. Stereotactic aspiration with thrombolysis Since the 1990s, stereotactic aspiration without endoscopy, mostly combined with instillation of fibrinolytic

Fig. 13.3 Unenhanced CT scan showing large cerebellar haematoma.

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 711

13.3 Management

indistinguishable from brainstem strokes, and very small haemorrhages which may simulate a peripheral disorder of the vestibular system.93 For decades there has been a strong impression that surgical evacuation saves lives in patients with cerebellar haemorrhages who have clinical evidence of progressive brainstem compression.94 The need for timely surgical treatment in patients who are in danger of progressive brainstem compression is dictated on the one hand by the potentially fatal outcome and on the other hand by the often surprisingly mild sequelae after operation.95 So strong is this notion, that a clinical trial in this category of patients is as unlikely to be mounted as it would be in acute appendicitis. This is not to say there are no areas of controversy. First, there is no doubt that some patients can be managed conservatively, but there is uncertainty about the selection criteria.96,97 In general, an impaired level of consciousness (that is, a Glasgow Coma Scale Score of 13/15 or less) seems a good indication for operative intervention, provided corneal and oculocephalic reflexes have not been lost, in which case the outcome is invariably fatal.98 Some neurosurgeons advocate surgery with large haemorrhages (>3–4 cm in diameter), even in alert patients, on the basis of experience that delayed deterioration of consciousness may be so rapid that the patient cannot be salvaged at this later stage.99 Others maintain that such rapid deterioration depends not so much on the absolute size of the haemorrhage as on the degree of effacement of the fourth ventricle and that operative evacuation is indicated in all patients in whom the fourth ventricle is completely obliterated.100 Since effacement of the fourth ventricle is associated with hydrocephalus and location of the haemorrhage in the midline (vermis), it is not surprising that these features also predispose to secondary deterioration.101 A rare category of cerebellar haemorrhages are those associated with neurosurgical operations (for any lesion) in the supratentorial compartment, so-called ‘remote cerebellar haemorrhages’.102 The pathogenesis is unclear, while the clinical course is almost invariably self-limiting. A second contentious issue is that in the past some neurosurgeons have argued that it is not direct compression of the brainstem but obstructive hydrocephalus that is the main problem, and that ventriculostomy is a sufficient measure to prevent a fatal outcome.103 However, there are two major objections against ventricular shunting as a panacea for cerebellar haemorrhages. The first is the rapid disappearance of pupillary, corneal and oculocephalic reflexes testifying brainstem compression in some patients for whom operation is too late, something not seen with acute hydrocephalus from other causes.104 The second is the frequent failure of ventricular

..

..

shunting to improve the level of consciousness.105 Nevertheless we do not wish to deny that ventriculostomy can be the only effective treatment in some patients with appropriate clinical features of hydrocephalus, i.e. in whom deterioration of consciousness is gradual, with sustained downward gaze and small unreactive pupils but no signs of pontine compression.100 Surgical evacuation of cerebellar haemorrhages can be life-saving, often with surprisingly few neurological sequelae. Sound indications for evacuation are the combination of a depressed level of consciousness with signs of progressive brainstem compression (unless all brainstem reflexes have been lost for more than a few hours, in which case a fatal outcome is inevitable), or haemorrhage diameter is greater than 3–4 cm. If the patient has a depressed level of consciousness and hydrocephalus, without signs of brainstem compression and with a haemorrhage less than 3 cm, ventriculostomy can be carried out as an initial (and perhaps only) procedure. As with supratentorial haemorrhage, some patients with cerebellar haemorrhage have been successfully treated by stereotactic aspiration, with or without instillation of fibrinolytic drugs.106 Any advantage of these techniques over the conventional approach with suboccipital craniotomy remains to be proved. The same applies to ventriculostomy by endoscopic perforation of the third ventricle towards the basal cisterns,107 rather than by the conventional method of a catheter inserted in the lateral ventricle. Pontine haemorrhage Pontine haemorrhages (Figs 8.31 and 13.4) are not as invariably fatal as when the diagnosis could only be made at postmortem, but the case fatality is still around 50%.17,108 The outcome depends to an important extent on the cause, since haemorrhage in the pons from a cavernous malformation is rarely fatal.109 Cavernous malformations are separately dealt with below (section 13.3.8). The management of patients with ‘hypertensive’ pontine haemorrhage is usually conservative, but some case reports have documented successful stereotactic aspiration.110 The natural history may or may not have been influenced by these interventions, and surgical failures in similar cases are likely to receive rather less publicity. There is an extremely narrow margin of uncertainty between patients with poor prognosis because of absent corneal and oculocephalic reflexes and those with small haemorrhages who will do well with conservative management.

711

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 712

712

Chapter 13 Specific treatment of intracerebral haemorrhage

There is little doubt that (recurrent) haemorrhages from cavernous malformations are less destructive than brainstem haemorrhages from an arterial source.109 Also the lesions may spontaneously regress.117 Successes of surgical treatment make it to a publication,118 but the failure rate can only be guessed at. At any rate incomplete removal can prompt rebleeding.119 Stereotactic cobalt-generated radiation (‘gamma knife’) has been applied to a large number of patients with cavernous malformations; there is no controlled evidence for the efficacy of this treatment and nothing more than a suggestion that seizures or episodes of rebleeding are less likely to occur afterwards.120,121

Fig. 13.4 Unenhanced CT brain scan showing a pontine haematoma (arrows) in a patient who walked, unsteadily, into (and out of) the outpatient clinic.

Cavernous malformations, whether in a cerebral hemisphere or in the brainstem, carry a moderate risk of (re)bleeding, while the functional deficits from such haemorrhages are limited. Surgical treatment or stereotactic radiation is regularly performed but there are no randomized controlled studies.

13.3.9 Treatment of dural arteriovenous fistulae 13.3.7 Treatment of lobar haemorrhage from (presumed) amyloid angiopathy In patients with lobar haemorrhage from presumed amyloid angiopathy and an impaired level of consciousness the prognosis is often poor after surgical intervention.111 Uncontrolled series with a large proportion of survivors after operative treatment fail to answer the question whether the outcome would have been worse without intervention.112,113 Given these uncertainties and the danger that the operation provokes new haemorrhages from brittle vessels at distant sites,114 conservative treatment seems the best option unless a deteriorating level of consciousness justifies the risk of operation.

13.3.8 Treatment of cavernous malformations For cavernous malformations in a cerebral hemisphere or in the cerebellum, any intervention is exceptional, because recurrent haemorrhages from these lesions are usually limited in severity (section 8.2.5). Series with operated patients have been published,115 but these underline rather than answer the question whether surgical treatment really improves on the natural history. The management of cavernous malformations in the brainstem is also controversial, since the natural history often is rather favourable, regardless of location.116 Probably pressure from patients themselves accounts for some of the operations.

These malformations are heterogeneous, in that they may or may not be secondary to occlusion of a major sinus and that the abnormal venous drainage from meningeal arteries may be channelled into a dural sinus or into superficial veins on the surface of the brain, cerebellum, or brainstem (section 8.2.8). Accordingly, there is a large variety of methods used to occlude the fistula. Surgical techniques include selective ligation of leptomeningeal draining veins,122,123 resection of fistulous sinus tracts,122 and a cranial base approach with extradural bone removal.124 Endovascular techniques may consist of an approach from the arterial or venous side,125,126 recanalization and stenting of a venous sinus,127 or venous embolization via a craniotomy.128 Finally stereotactic cobalt-generated radiation (‘gamma knife’) is also used for obliterating arteriovenous fistulae,129,130 sometimes in conjunction with an endovascular approach.131

13.3.10 Treatment of moyamoya syndrome associated with haemorrhage Gradual stenosis or occlusion of the terminal portions of the internal carotid arteries or proximal middle cerebral arteries, most often the result of an idiopathic proliferative process of the endothelial layer, may lead to the formation of an abnormal collateral network of fragile collaterals, which occasionally rupture (sections 8.2.12 and 7.5). To relieve the pressure on the collaterals by

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 713

References

constructing a bypass such as between the superficial temporal artery and the middle cerebral artery is a logical idea,132 but whether this indeed prevents (re)bleeding remains uncertain. In fact an aneurysm may form and rupture at the site of the arterial bypass.133

References

1 Rosand J, Eskey C, Chang Y, Gonzalez RG, Greenberg SM, Koroshetz WJ. Dynamic single-section CT demonstrates reduced cerebral blood flow in acute intracerebral hemorrhage. Cerebrovasc Dis 2002; 14:214–20. 2 Castillo J, Dávalos A, Alvarez-Sabín J et al. Molecular signatures of brain injury after intracerebral hemorrhage. Neurology 2002; 58:624–9. 3 Butcher KS, Baird T, MacGregor L, Desmond P, Tress B, Davis S. Perihematomal edema in primary intracerebral hemorrhage is plasma derived. Stroke 2004; 35:1879–85. 4 Gebel JM, Jr., Jauch EC, Brott TG et al. Natural history of perihematomal edema in patients with hyperacute spontaneous intracerebral hemorrhage. Stroke 2002; 33:2631–5. 5 Siddique MS, Fernandes HM, Wooldridge TD, Fenwick JD, Slomka P, Mendelow AD. Reversible ischemia around intracerebral hemorrhage: a single-photon emission computerized tomography study. J Neurosurg 2002; 96:736–41. 6 Bhattathiri PS, Gregson B, Prasad KS, Mendelow AD. Intraventricular hemorrhage and hydrocephalus after spontaneous intracerebral hemorrhage: results from the STICH trial. Acta Neurochir Suppl 2006; 96:65–8. 7 Dennis MS. Outcome after brain haemorrhage. Cerebrovasc Dis 2003; 16(Suppl 1):9–13. 8 Flaherty ML, Haverbusch M, Sekar P et al. Long-term mortality after intracerebral hemorrhage. Neurology 2006; 66:1182–6. 9 Ariesen MJ, Algra A, Van der Worp HB, Rinkel GJE. Applicability and relevance of models that predict short term outcome after intracerebral haemorrhage. J Neurol Neurosurg Psychiatry 2005; 76:839–44. 10 Weimar C, Benemann J, Diener HC, for the German Stroke Study Collaboration. Development and validation of the Essen Intracerebral Haemorrhage Score. J Neurol Neurosurg Psychiatry 2006; 77:601–5. 11 Becker KJ, Baxter AB, Cohen WA et al. Withdrawal of support in intracerebral hemorrhage may lead to selffulfilling prophecies. Neurology 2001; 56:766–72. 12 Delgado P, Alvarez-Sabin J, Abilleira S et al. Plasma d-dimer predicts poor outcome after acute intracerebral hemorrhage. Neurology 2006; 67:94–8.

..

..

13 Hays A, Diringer MN. Elevated troponin levels are associated with higher mortality following intracerebral hemorrhage. Neurology 2006; 66:1330–4. 14 Roquer J, Rodriguez CA, Gomis M, Ois A, Munteis E, Bohm P. Serum lipid levels and in-hospital mortality in patients with intracerebral hemorrhage. Neurology 2005; 65:1198–202. 15 Flemming KD, Wijdicks EFM, Li HZ. Can we predict poor outcome at presentation in patients with lobar hemorrhage? Cerebrovasc Dis 2001; 11:183–9. 16 Liliang PC, Liang CL, Lu CH et al. Hypertensive caudate hemorrhage prognostic predictor, outcome, and role of external ventricular drainage. Stroke 2001; 32:1195–200. 17 Wijdicks EFM, St. Louis E. Clinical profiles predictive of outcome in pontine hemorrhage. Neurology 1997; 49:1342–6. 18 Davis SM, Broderick J, Hennerici M et al. Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage. Neurology 2006; 66:1175–81. 19 Schwarz S, Häfner K, Aschoff A, Schwab S. Incidence and prognostic significance of fever following intracerebral hemorrhage. Neurology 2000; 54:354–61. 20 Gebel JM, Jr., Jauch EC, Brott TG et al. Relative edema volume is a predictor of outcome in patients with hyperacute spontaneous intracerebral hemorrhage. Stroke 2002; 33:2636–41. 21 Ronning OM, Guldvog B, Stavem K. The benefit of an acute stroke unit in patients with intracranial haemorrhage: a controlled trial. J Neurol Neurosurg Psychiatry 2001; 70:631–4. 22 Friederich PW, Henny CP, Messelink EJ et al. Effect of recombinant activated factor VII on perioperative blood loss in patients undergoing retropubic prostatectomy: a double-blind placebo-controlled randomised trial. Lancet 2003; 361:201–5. 23 You H, Al-Shahi R. Haemostatic drug therapies for acute primary intracerebral haemorrhage. Cochrane Database Syst Rev 2006; CD005951. 24 Earnshaw SR, Joshi AV, Wilson MR, Rosand J. Costeffectiveness of recombinant activated factor VII in the treatment of intracerebral hemorrhage. Stroke 2006; 37:2751–8. 25 Tomson J, Lip GY. Blood pressure changes in acute haemorrhagic stroke. Blood Press Monit 2005; 10:197–9. 26 Ohwaki K, Yano E, Nagashima H, Hirata M, Nakagomi T, Tamura A. Blood pressure management in acute intracerebral hemorrhage: relationship between elevated blood pressure and hematoma enlargement. Stroke 2004; 35:1364–7. 27 Qureshi AI, Mohammad YM, Yahia AM et al. A prospective multicenter study to evaluate the feasibility and safety of aggressive antihypertensive treatment in patients with acute intracerebral hemorrhage. J Intensive Care Med 2005; 20:34–42. 28 Arakawa S, Saku Y, Ibayashi S, Nagao T, Fujishima M. Blood pressure control and recurrence of hypertensive brain hemorrhage. Stroke 1998; 29:1806–9.

713

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 714

714

Chapter 13 Specific treatment of intracerebral haemorrhage

29 Broderick JP, Adams HP, Jr., Barsan W et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke 1999; 30:905–15. 30 Guyot LL, Dowling C, Diaz FG, Michael DB. Cerebral monitoring devices: analysis of complications. Acta Neurochir Suppl 1998; 71:47–9. 31 Cremer OL, van Dijk GW, van Wensen E et al. Effect of intracranial pressure monitoring and targeted intensive care on functional outcome after severe head injury. Crit Care Med 2005; 33:2207–13. 32 Adams RE, Diringer MN. Response to external ventricular drainage in spontaneous intracerebral hemorrhage with hydrocephalus. Neurology 1998; 50:519–23. 33 Holtzman RN, Brust JC, Ainyette IG et al. Acute ventricular hemorrhage in adults with hydrocephalus managed by corpus callosotomy and fenestration of the septum pellucidum. Report of three cases. J Neurosurg 2001; 95:111–15. 34 Nieuwkamp DJ, de Gans K, Rinkel GJE, Algra A. Treatment and outcome of severe intraventricular extension in patients with subarachnoid or intracerebral hemorrhage: a systematic review of the literature. J Neurol 2000; 247:117–21. 35 Lapointe M, Haines S. Fibrinolytic therapy for intraventricular hemorrhage in adults. Cochrane Database Syst Rev 2002; CD003692. 36 Lee MW, Pang KY, Ho WW, Wong CK. Outcome analysis of intraventricular thrombolytic therapy for intraventricular haemorrhage. Hong Kong Med J 2003; 9:335–40. 37 Fountas KN, Kapsalaki EZ, Parish DC et al. Intraventricular administration of rt-PA in patients with intraventricular hemorrhage. South Med J 2005; 98:767–73. 38 Stocchetti N, Maas AI, Chieregato A, van der Plas AA. Hyperventilation in head injury: a review. Chest 2005; 127:1812–27. 39 Muizelaar JP, Marmarou A, Ward JD et al. Adverse effects of prolonged hyperventilation in patients with severe head injury: a randomized clinical trial. J Neurosurg 1991; 75:731–9. 40 Schierhout G, Roberts I. Hyperventilation therapy for acute traumatic brain injury. Cochrane Database Syst Rev 2000; CD000566. 41 Nau R. Osmotherapy for elevated intracranial pressure: a critical reappraisal. Clin Pharmacokinet 2000; 38:23–40. 42 Bettmann MA. Frequently asked questions: iodinated contrast agents. Radiographics 2004; 24(Suppl 1):S3–10. 43 Yu YL, Kumana CR, Lauder IJ et al. Treatment of acute cerebral hemorrhage with intravenous glycerol. A doubleblind, placebo-controlled, randomized trial. Stroke 1992; 23:967–71. 44 Bereczki D, Liu M, do Prado GF, Fekete I. Mannitol for acute stroke. Cochrane Database Syst Rev 2001; CD001153. 45 Misra UK, Kalita J, Ranjan P, Mandal SK. Mannitol in intracerebral hemorrhage: a randomized controlled study. J Neurol Sci 2005; 234:41–5.

46 Dziedzic T, Szczudlik A, Klimkowicz A, Rog TM, Slowik A. Is mannitol safe for patients with intracerebral hemorrhages? Renal considerations. Clin Neurol Neurosurg 2003; 105:87–9. 47 Cruz J, Minoja G, Okuchi K. Major clinical and physiological benefits of early high doses of mannitol for intraparenchymal temporal lobe hemorrhages with abnormal pupillary widening: a randomized trial. Neurosurgery 2002; 51:628–37. 48 Righetti E, Celani MG, Cantisani T, Sterzi R, Boysen G, Ricci S. Glycerol for acute stroke. Cochrane Database Syst Rev 2004; CD000096. 49 Ogden AT, Mayer SA, Connolly ES, Jr. Hyperosmolar agents in neurosurgical practice: the evolving role of hypertonic saline. Neurosurgery 2005; 57:207–15. 50 Feigin VL, Anderson N, Rinkel GJ, Algra A, van Gijn J, Bennett DA. Corticosteroids for aneurysmal subarachnoid haemorrhage and primary intracerebral haemorrhage. Cochrane Database Syst Rev 2005; CD004583. 51 Davenport RJ, Dennis MS, Wellwood I, Warlow CP. Complications after acute stroke. Stroke 1996; 27:415–20. 52 Kelly J, Hunt BJ, Lewis RR, Rudd A. Anticoagulation or inferior vena cava filter placement for patients with primary intracerebral hemorrhage developing venous thromboembolism? Stroke 2003; 34:2999–3005. 53 Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in fatal pulmonary embolism and venous thrombosis by perioperative administration of subcutaneous heparin. Overview of results of randomized trials in general, orthopedic, and urologic surgery. N Engl J Med 1988; 318:1162–73. 54 Dickmann U, Voth E, Schicha H, Henze T, Prange H, Emrich D. Heparin therapy, deep-vein thrombosis and pulmonary embolism after intracerebral hemorrhage. Klin Wochenschr 1988; 66:1182–3. 55 Boeer A, Voth E, Henze T, Prange HW. Early heparin therapy in patients with spontaneous intracerebral haemorrhage. J Neurol Neurosurg Psychiatry 1991; 54:466–7. 56 Keir SL, Lewis SC, Wardlaw JM, Sandercock PA, on behalf of IST and CAST collaborative groups. Effect of aspirin or heparin inadvertently given to patients with hemorrhagic stroke. Stroke 2000; 31:314. 57 Amaragiri SV, Lees TA. Elastic compression stockings for prevention of deep vein thrombosis. Cochrane Database Syst Rev 2000; CD001484. 58 Mazzone C, Chiodo GF, Sandercock P, Miccio M, Salvi R. Physical methods for preventing deep vein thrombosis in stroke. Cochrane Database Syst Rev 2004; CD001922. 59 Lacut K, Bressollette L, Le Gal G et al. Prevention of venous thrombosis in patients with acute intracerebral hemorrhage. Neurology 2005; 65:865–9. 60 Ramos R, Salem BI, De Pawlikowski MP, Coordes C, Eisenberg S, Leidenfrost R. The efficacy of pneumatic compression stockings in the prevention of pulmonary embolism after cardiac surgery. Chest 1996; 109:82–5. 61 Yasaka M, Minematsu K, Naritomi H, Sakata T, Yamaguchi T. Predisposing factors for enlargement

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 715

References

62

63

64

65

66

67

68

69

70

71

72

73 74

75

..

..

of intracerebral hemorrhage in patients treated with warfarin. Thromb Haemost 2003; 89:278–83. Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: current practices and unresolved questions. Stroke 2006; 37:256–62. Goldstein JN, Thomas SH, Frontiero V et al. Timing of fresh frozen plasma administration and rapid correction of coagulopathy in warfarin-related intracerebral hemorrhage. Stroke 2006; 37:151–5. Makris M, Greaves M, Phillips WS, Kitchen S, Rosendaal FR, Preston FE. Emergency oral anticoagulant reversal: the relative efficacy of infusions of fresh frozen plasma and clotting factor concentrate on correction of the coagulopathy. Thromb Haemost 1997; 77:477–80. Butler AC, Tait RC. Management of oral anticoagulantinduced intracranial haemorrhage. Blood Rev 1998; 12:35–44. Huttner HB, Schellinger PD, Hartmann M et al. Hematoma growth and outcome in treated neurocritical care patients with intracerebral hemorrhage related to oral anticoagulant therapy: comparison of acute treatment strategies using vitamin K, fresh frozen plasma, and prothrombin complex concentrates. Stroke 2006; 37:1465–70. Phan TG, Koh M, Wijdicks EFM. Safety of discontinuation of anticoagulation in patients with intracranial hemorrhage at high thromboembolic risk. Arch Neurol 2000; 57:1710–13. Bertram M, Bonsanto M, Hacke W, Schwab S. Managing the therapeutic dilemma: patients with spontaneous intracerebral hemorrhage and urgent need for anticoagulation. J Neurol 2000; 247:209–14. Estol CJ, Kase CS. Need for continued use of anticoagulants after intracerebral hemorrhage. Curr Treat Options Cardiovasc Med 2003; 5:201–9. Wijdicks EFM, Schievink WI, Brown RD, Mullany CJ. The dilemma of discontinuation of anticoagulation therapy for patients with intracranial hemorrhage and mechanical heart valves. Neurosurgery 1998; 42:769–73. Leker RR, Abramsky O. Early anticoagulation in patients with prosthetic heart valves and intracerebral hematoma. Neurology 1998; 50:1489–91. Sloan MA, Price TR, Petito CK et al. Clinical features and pathogenesis of intracerebral hemorrhage after rt-PA and heparin therapy for acute myocardial infarction: The Thrombolysis in Myocardial Infarction (TIMI) II pilot and randomized clinical trial combined experience. Neurology 1995; 45:649–58. Patrono C. Aspirin as an antiplatelet drug. N Engl J Med 1994; 330:1287–94. Keir SL, Wardlaw JM, Sandercock PAG, Chen ZM. Antithrombotic therapy in patients with any form of intracranial haemorrhage: a systematic review of the available controlled studies. Cerebrovasc Dis 2002; 14:197–206. Saloheimo P, Ahonen M, Juvela S, Pyhtinen J, Savolainen ER, Hillbom M. Regular aspirin-use preceding the onset of

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

primary intracerebral hemorrhage is an independent predictor for death. Stroke 2006; 37:129–33. Foerch C, Sitzer M, Steinmetz H, Neumann-Haefelin T. Pretreatment with antiplatelet agents is not independently associated with unfavorable outcome in intracerebral hemorrhage. Stroke 2006; 37:2165–7. Gregson BA, Mendelow AD. International variations in surgical practice for spontaneous intracerebral hemorrhage. Stroke 2003; 34:2593–7. McKissock W, Richardson A, Walsh L. Primary intracerebral haematoma: results of surgical treatment in 244 consecutive cases. Lancet 1959; ii:683–6. Mitsuno T, Kanaya H, Shirakata S, Ohsawa K, Ishikawa Y. Surgical treatment of hypertensive intracerebral hemorrhage. J Neurosurg 1966; 24:70–6. Morgenstern LB, Frankowski RF, Shedden P, Pasteur W, Grotta JC. Surgical treatment for intracerebral hemorrhage (STICH): a single-center, randomized clinical trial. Neurology 1998; 51:1359–63. Zuccarello M, Brott T, Derex L et al. Early surgical treatment for supratentorial intracerebral hemorrhage: a randomized feasibility study. Stroke 1999; 30:1833–9. Morgenstern LB, Demchuk AM, Kim DH, Frankowski RF, Grotta JC. Rebleeding leads to poor outcome in ultra-early craniotomy for intracerebral hemorrhage. Neurology 2001; 56:1294–9. Mendelow AD, Gregson BA, Fernandes HM et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet 2005; 365:387–97. Murray GD, Barer D, Choi S et al. Design and analysis of phase III trials with ordered outcome scales: the concept of the sliding dichotomy. J Neurotrauma 2005; 22:511–17. Mendelow AD, Teasdale GM, Barer D, Fernandes HM, Murray GD, Gregson BA. Outcome assignment in the International Surgical Trial of Intracerebral Haemorrhage. Acta Neurochir (Wien) 2003; 145:679–81. Rabinstein AA, Atkinson JL, Wijdicks EFM. Emergency craniotomy in patients worsening due to expanded cerebral hematoma: to what purpose? Neurology 2002; 58:1367–72. Qiu Y, Lin Y, Tian X, Luo Q. Hypertensive intracranial hematomas: endoscopic-assisted keyhole evacuation and application of patent viewing dissector. Chin Med J (Engl) 2003; 116:195–9. Nishihara T, Nagata K, Tanaka S et al. Newly developed endoscopic instruments for the removal of intracerebral hematoma. Neurocrit Care 2005; 2:67–74. Auer LM, Deinsberger W, Niederkorn K et al. Endoscopic surgery versus medical treatment for spontaneous intracerebral hematoma: a randomized study. J Neurosurg 1989; 70:530–5. Hankey GJ, Hon C. Surgery for primary intracerebral hemorrhage: Is it safe and effective? A systematic review of case series and randomized trials. Stroke 1997; 28:2126–32.

715

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 716

716

Chapter 13 Specific treatment of intracerebral haemorrhage

91 Hattori N, Katayama Y, Maya Y, Gatherer A. Impact of stereotactic hematoma evacuation on activities of daily living during the chronic period following spontaneous putaminal hemorrhage: a randomized study. J Neurosurg 2004; 101:417–20. 92 Teernstra OP, Evers SM, Lodder J, Leffers P, Franke CL, Blaauw G. Stereotactic treatment of intracerebral hematoma by means of a plasminogen activator: a multicenter randomized controlled trial (SICHPA). Stroke 2003; 34:968–74. 93 Jensen MB, St Louis EK. Management of acute cerebellar stroke. Arch Neurol 2005; 62:537–44. 94 Fisher CM, Picard EH, Polak A, Dalal P, Ojemann R. Acute hypertensive cerebellar hemorrhage. J Nerv Ment Dis 1965; 140:38–57. 95 Cohen ZR, Ram Z, Knoller N, Peles E, Hadani M. Management and outcome of non-traumatic cerebellar haemorrhage. Cerebrovasc Dis 2002; 14:207–13. 96 Wijdicks EFM, Louis EKS, Atkinson JD, Li HZ. Clinician’s biases toward surgery in cerebellar hematomas: an analysis of decision-making in 94 patients. Cerebrovasc Dis 2000; 10:93–6. 97 Wessels PH, Ter Berg JWM, Spincemaille GH, Dippel DWJ. Treatment of cerebellar hematoma in the Netherlands: a questionnaire survey. Cerebrovasc Dis 2001; 11:190–4. 98 St Louis EK, Wijdicks EFM, Li HZ, Atkinson JD. Predictors of poor outcome in patients with a spontaneous cerebellar hematoma. Can J Neurol Sci 2000; 27:32–6. 99 Kobayashi S, Sato A, Kageyama Y, Nakamura H, Watanabe Y, Yamaura A. Treatment of hypertensive cerebellar hemorrhage: surgical or conservative management? Neurosurgery 1994; 34:246–50. 100 Kirollos RW, Tyagi AK, Ross SA, Van Hille PT, Marks PV. Management of spontaneous cerebellar hematomas: a prospective treatment protocol. Neurosurgery 2001; 49:1378–86. 101 St Louis EK, Wijdicks EFM, Li HZ. Predicting neurologic deterioration in patients with cerebellar hematomas. Neurology 1998; 51:1364–9. 102 Amini A, Osborn AG, McCall TD, Couldwell WT. Remote cerebellar hemorrhage. Am J Neuroradiol 2006; 27:387–90. 103 Shenkin HA, Zavala H. Cerebellar strokes: mortality, surgical indications, and results of ventricular drainage. Lancet 1982; ii:429–31. 104 Van Gijn J, Hijdra A, Wijdicks EFM, Vermeulen M, van Crevel H. Acute hydrocephalus after aneurysmal subarachnoid hemorrhage. J Neurosurg 1985; 63:355–62. 105 Mathew P, Teasdale G, Bannan A, Oluoch-Olunya D. Neurosurgical management of cerebellar haematoma and infarct. J Neurol Neurosurg Psychiatry 1995; 59:287–92. 106 Mohadjer M, Eggert R, May J, Mayfrank L. CT-guided stereotactic fibrinolysis of spontaneous and hypertensive cerebellar hemorrhage: long-term results. J Neurosurg 1990; 73:217–22. 107 Roux FE, Boetto S, Tremoulet M. Third ventriculocisternostomy in cerebellar haematomas. Acta Neurochir (Wien) 2002; 144:337–42.

108 Balci K, Asil T, Kerimoglu M, Celik Y, Utku U. Clinical and neuroradiological predictors of mortality in patients with primary pontine hemorrhage. Clin Neurol Neurosurg 2005; 108:36–9. 109 Rabinstein AA, Tisch SH, McClelland RL, Wijdicks EFM. Cause is the main predictor of outcome in patients with pontine hemorrhage. Cerebrovasc Dis 2004; 17:66–71. 110 Shitamichi M, Nakamura J, Sasaki T, Suematsu K, Tokuda S. Computed tomography guided stereotactic aspiration of pontine hemorrhages. Stereotact Funct Neurosurg 1990; 54–55:453–6. 111 McCarron MO, Nicoll JA, Love S, Ironside JW. Surgical intervention, biopsy and APOE genotype in cerebral amyloid angiopathy-related haemorrhage. Br J Neurosurg 1999; 13:462–7. 112 Greene GM, Godersky JC, Biller J, Hart MN, Adams HP, Jr. Surgical experience with cerebral amyloid angiopathy. Stroke 1990; 21:1545–9. 113 Izumihara A, Ishihara T, Iwamoto N, Yamashita K, Ito H. Postoperative outcome of 37 patients with lobar intracerebral hemorrhage related to cerebral amyloid angiopathy. Stroke 1999; 30:29–33. 114 Brisman MH, Bederson JB, Sen CN, Germano IM, Moore F, Post KD. Intracerebral hemorrhage occurring remote from the craniotomy site. Neurosurgery 1996; 39:1114–21. 115 Zotta D, Di RA, Scogna A, Ricci A, Ricci G, Galzio RJ. Supratentorial cavernomas in eloquent brain areas: application of neuronavigation and functional MRI in operative planning. J Neurosurg Sci 2005; 49:13–9. 116 Kupersmith MJ, Kalish H, Epstein F et al. Natural history of brainstem cavernous malformations. Neurosurgery 2001; 48:47–53. 117 Yasui T, Komiyama M, Iwai Y et al. A brainstem cavernoma demonstrating a dramatic, spontaneous decrease in size during follow-up: case report and review of the literature. Surg Neurol 2005; 63:170–3. 118 Ferroli P, Sinisi M, Franzini A, Giombini S, Solero CL, Broggi G. Brainstem cavernomas: long-term results of microsurgical resection in 52 patients. Neurosurgery 2005; 56:1203–12. 119 Kikuta K, Nozaki K, Takahashi JA, Miyamoto S, Kikuchi H, Hashimoto N. Postoperative evaluation of microsurgical resection for cavernous malformations of the brainstem. J Neurosurg 2004; 101:607–12. 120 Régis J, Bartolomei F, Kida Y et al. Radiosurgery for epilepsy associated with cavernous malformation: retrospective study in 49 patients. Neurosurgery 2000; 47:1091–7. 121 Liu KD, Chung WY, Wu HM et al. Gamma knife surgery for cavernous hemangiomas: an analysis of 125 patients. J Neurosurg 2005; 102(Suppl):81–6. 122 Collice M, D’Aliberti G, Arena O, Solaini C, Fontana RA, Talamonti C. Surgical treatment of intracranial dural arteriovenous fistulae: role of venous drainage. Neurosurgery 2000; 47:56–66. 123 Van Dijk JM, TerBrugge KG, Willinsky RA, Wallace MC. Selective disconnection of cortical venous reflux as

..

..

9781405127660_4_013.qxd 10/13/07 10:54 AM Page 717

References

124

125

126

127

128

..

..

treatment for cranial dural arteriovenous fistulas. J Neurosurg 2004; 101:31–5. Kattner DA, Roth TC, Giannotta SL. Cranial base approaches for the surgical treatment of aggressive posterior fossa dural arteriovenous fistulae with leptomeningeal drainage: report of four technical cases. Neurosurgery 2002; 50:1156–60. Klisch J, Huppertz HJ, Spetzger U, Hetzel A, Seeger W, Schumacher M. Transvenous treatment of carotid cavernous and dural arteriovenous fistulae: results for 31 patients and review of the literature. Neurosurgery 2003; 53:836–56. Tomak PR, Cloft HJ, Kaga A, Cawley CM, Dion J, Barrow DL. Evolution of the management of tentorial dural arteriovenous malformations. Neurosurgery 2003; 52:750–60. Murphy KJ, Gailloud P, Venbrux A, Deramond H, Hanley D, Rigamonti D. Endovascular treatment of a grade IV transverse sinus dural arteriovenous fistula by sinus recanalization, angioplasty, and stent placement: technical case report. Neurosurgery 2000; 46:497–500. Houdart E, Saint-Maurice JP, Chapot R et al. Transcranial approach for venous embolization of dural arteriovenous fistulas. J Neurosurg 2002; 97:280–6.

129 O’Leary S, Hodgson TJ, Coley SC, Kemeny AA, Radatz MW. Intracranial dural arteriovenous malformations: results of stereotactic radiosurgery in 17 patients. Clin Oncol (R Coll Radiol) 2002; 14:97–102. 130 Pan DH, Chung WY, Guo WY et al. Stereotactic radiosurgery for the treatment of dural arteriovenous fistulas involving the transverse-sigmoid sinus. J Neurosurg 2002; 96:823–9. 131 Friedman JA, Pollock BE, Nichols DA, Gorman DA, Foote RL, Stafford SL. Results of combined stereotactic radiosurgery and transarterial embolization for dural arteriovenous fistulas of the transverse and sigmoid sinuses. J Neurosurg 2001; 94:886–91. 132 Kawaguchi S, Okuno S, Sakaki T. Effect of direct arterial bypass on the prevention of future stroke in patients with the hemorrhagic variety of moyamoya disease. J Neurosurg 2000; 93:397–401. 133 Nishimoto T, Yuki K, Sasaki T, Murakami T, Kodama Y, Kurisu K. A ruptured middle cerebral artery aneurysm originating from the site of anastomosis 20 years after extracranial-intracranial bypass for moyamoya disease: case report. Surg Neurol 2005; 64:261–5.

717

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 719

14

Specific treatment of aneurysmal subarachnoid haemorrhage 14.1

General principles

719

14.2

Treatable causes of poor clinical condition on admission and their management

14.3

General care

14.4

Prevention of rebleeding

14.5

Prevention of delayed cerebral ischaemia

14.6

Management of rebleeding

14.7

Management of delayed cerebral ischaemia

14.8

Management of acute hydrocephalus

14.9

Management of systemic complications

721

725 728 742

14.10 Late sequelae and complications

14.1 General principles

The essence of managing patients with subarachnoid haemorrhage (SAH) from a ruptured aneurysm is deceptively simple: make the diagnosis, locate the aneurysm and occlude it. But, although occlusion of the ruptured aneurysm by an interventional radiologist or neurosurgeon is an important part, it is by no means the only part in the management. Treating patients with aneurysmal SAH requires taking care of acutely ill patients threatened by many more complications than only rebleeding, and should be done by a dedicated, multidisciplinary team. This chapter deals with the treatment of patients with a (suspected) aneurysmal cause of their SAH. Those with perimesencephalic non-aneurysmal haemorrhage are not threatened by complications except for the rare patient who develops a symptomatic hydrocephalus within the first 1 or 2 days after the onset. We usually

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

735 744

746 749

752

discharge patients with a perimesencephalic haemorrhage after 2 or 3 days, and do not give medications other than analgesics. In contrast, patients with an aneurysmal pattern of haemorrhage, even without an aneurysm (or other cause) detected, are admitted for at least 14 days and treated similarly to patients who have had an aneurysm detected. The case fatality after aneurysmal SAH is still as high as 50%,1 if one takes into account the 10–15% who die before receiving medical attention.2 And of the 50% who survive the initial weeks after the haemorrhage, again half are left with permanent deficits in high-level (neuropsychological) functioning, resulting in impairment in their social role.3 To a large extent this poor outcome is caused by the many complications other than rebleeding that beset the course of the disease: delayed cerebral ischaemia, hydrocephalus, a variety of systemic disorders,4 and complications of the aneurysm treatment itself.5 However, despite the considerable proportion of patients who die from the impact of the first haemorrhage and the many complications that can occur, the prognosis for patients with SAH has gradually improved over the last decades.1,6 Patients should always be transferred as quickly as possible to a centre where a multidisciplinary team is available 24 hours a day, 7 days a week, and where enough patients are managed to maintain and improve standards of care.7 The need for referral applies especially to 719

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 720

720

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

(c) (a)

(b)

(d)

(e)

(f) Fig. 14.1 CT scans of a patient with extensive initial ischaemia. This 50-year-old woman suddenly lost consciousness, and remained comatose. (a, b, c) CT scan on admission showing subarachnoid blood in the frontal interhemispheric fissure and sylvian fissures (white arrowheads), and areas with poorly

demarcated white and grey matter (black arrowheads), but no other abnormalities that explain the coma. (d, e, f) CT scan 12 h later showing extensive areas of infarction (black arrowheads).

patients who are designated as ‘poor grade cases’;8 these may be the very patients who need urgent intervention because of, for example, early rebleeding, progressive brain shift from a haematoma, acute hydrocephalus or hypovolaemia. For the first 2 weeks after the haemorrhage, patients should be managed in an intensive or medium care unit. The neurologist can play a central role in the team, which should consist of neurosurgeons, interventional neuroradiologists, neuro-intensivists,

rehabilitation physicians, neuropsychologists, and a dedicated team of nurses, physiotherapists, speech therapists and occupational therapists. The neurologist is well suited and available for assessing the patient’s general medical and neurological condition on a frequent basis, and can act immediately if there is any clinical deterioration. Moreover, the neurologist can use his or her experience from working in a stroke unit in the management of SAH patients.

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 721

14.2 Treatable causes of poor clinical condition on admission and their management

14.2 Treatable causes of poor clinical condition on admission and their management

A decreased level of consciousness, with the initial haemorrhage or after early rebleeding, may be caused by neurological complications such as an intracerebral haematoma, extensive intraventricular haemorrhage, subdural haematoma or hydrocephalus,8 and by cardiopulmonary complications. Only by exclusion should it be assumed that the cause is global brain damage as a result of the high pressure after aneurysmal rupture,

resulting in impaired or absent cerebral blood flow and subsequent ischaemia (Fig. 14.1).

14.2.1 Early rebleeding In the first few hours after the initial haemorrhage, up to 15% of patients have a sudden episode of clinical deterioration that suggests rebleeding (Fig. 14.2).9–11 Because such episodes often occur before the first CT scan, or even before admission to hospital, a definite diagnosis is difficult and the true frequency of rebleeding on the first day is invariably underestimated. If respiratory or cardiac arrest occurs at the time of (suspected) rebleeding during transportation to hospital or early after arrival at the

(c)

(a)

(b)

(d)

(e)

Fig. 14.2 CT scans of a patient with early rebleeding. (a,b) CT scan on admission showing large amounts of subarachnoid blood in the basal cisterns and left sylvian fissure. (c) CT angiogram on admission showing an aneurysm of the left

..

..

middle cerebral artery (arrow). Shortly after the initial scan the patient suddenly lost consciousness. (d, e) Repeated CT scan after the clinical deterioration confirmed rebleeding with more extravasated blood (arrowheads).

721

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 722

722

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

(a)

(b)

(c)

(d)

emergency department, the patient should be resuscitated and ventilated, because survival without brain damage is still possible (section 14.6.2).12,13 After resuscitation, it will usually become clear within a matter of hours whether there is persisting brainstem dysfunction.

14.2.2 Intracerebral extension of the haemorrhage Intracerebral extension of the haemorrhage occurs in about one-third of patients with ruptured aneurysms14 and, not surprisingly, the outcome is worse than in patients with purely subarachnoid blood.15 When the most likely cause of deterioration in the level of consciousness is a large temporal lobe haemorrhage, usually from a ruptured aneurysm of the middle cerebral artery, immediate

Fig. 14.3 CT scans of a patient with a poor clinical condition on admission from intracerebral extension of the haemorrhage. (a) CT scan on admission showing intracerebral extension of the haemorrhage (arrowheads) and (b) shift of the lateral (arrowhead) and third ventricle (arrow) across the midline. (c) CT angiogram showing an aneurysm of the right middle cerebral artery (arrow). The patient was operated upon immediately with clipping of the aneurysm and evacuation of the haematoma. (d) Repeat CT scan 3 months after the haemorrhage showing only a small loss of parenchyma (arrowhead); the patient had regained independence for activities of daily living.

evacuation should be considered (Figs 14.3 and 14.4). The traditional intervention is evacuation of the haematoma with simultaneous clipping of the aneurysm if it can be identified, usually only by MR angiography or CT angiography rather than by catheter angiography; this course of action is backed up by a small randomized study, in which 11 of 15 patients in the operated group survived, against 3 of 15 in the conservatively treated group (relative risk 0.27; 95% CI 0.09–0.74).16 Nowadays of course this approach may be modified to immediate occlusion of the aneurysm by endovascular coiling, followed by evacuation of the haematoma.17 Another option is immediate aneurysm occlusion followed by an extensive hemicraniectomy that allows external expansion of the brain.18

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 723

14.2 Treatable causes of poor clinical condition on admission and their management

Subdural haematomas secondary to a ruptured aneurysm may be life threatening, in which case immediate evacuation seems called for, despite the lack of controlled studies.8,25

14.2.4 Acute hydrocephalus and intraventricular haemorrhage

Fig. 14.4 ‘Enlarging haematoma’. Top: CT scan on day of subarachnoid haemorrhage from a ruptured aneurysm of the right middle cerebral artery, in a 51-year-old man on anticoagulant treatment because of myocardial infarction with mitral valve regurgitation. Haematoma in the right sylvian fissure (thin arrow) and in the subinsular cortex (thick arrow); at the time of the scan anticoagulation had already been reversed by infusion of a concentrate of coagulation factors. Bottom: CT scan 5 days after the haemorrhage, showing a marked increase of the space-occupying effect, with compression of the right lateral ventricle and the third ventricle. The increased mass effect results not only from a newly formed rim of oedema around the haematoma (short arrow), but also from an increase in size of the haematoma itself. There had been no new neurological signs or other clinical features indicating rebleeding. The newly formed collection of blood in the posterior horn of the left lateral ventricle (long arrow) is often seen in the absence of rebleeding (sedimentation effect).

14.2.3 Acute subdural haematoma In about 2% of patients an acute subdural haematoma complicates rupture of an intracranial aneurysm, usually at the origin of the posterior communicating artery (Fig. 14.5),19,20 more often with recurrent aneurysmal rupture than a first bleed.21 Sometimes there is no associated haemorrhage in the subarachnoid space to indicate a ruptured aneurysm as the underlying cause.22–24

..

..

Gradual obtundation within 24 h of haemorrhage, sometimes accompanied by slow pupillary responses to light and downward deviation of the eyes, is fairly characteristic of acute hydrocephalus;26,27 and some patients lapse into coma from hydrocephalus within a few hours of the haemorrhage. If the diagnosis is confirmed by brain CT, then a lumbar puncture or early ventricular drainage may be needed,28,29 although some patients improve spontaneously in the first 24 h. This is an area badly in need of randomized trials. Acute hydrocephalus with a large amount of intraventricular blood is often associated with a poor clinical condition from the outset (Fig. 14.6). In patients with ruptured aneurysms, i.e. an arterial source of bleeding, an intraventricular haemorrhage volume of 20 mL or more is invariably lethal.30 An indirect comparison of observational studies suggests that an external ventricular catheter is not very helpful, but that a strategy where drainage is combined with fibrinolysis through the drain results in a good outcome in half the patients.31 This needs to be confirmed in randomized controlled trials.32 Surgical evacuation of a ‘packed’ intraventricular haemorrhage, especially in the fourth ventricle, is pointless according to some,33 while others disagree.34 At any rate, ventriculostomy is certainly indicated in patients with acute symptomatic hydrocephalus caused by a small clot plugging the proximal aqueduct.35

14.2.5 Cardiopulmonary complications Pulmonary oedema and cardiac dysfunction, which can aggravate the pulmonary oedema,36 may develop within hours of the haemorrhage, even in patients who are admitted in good clinical condition (Fig. 14.7). A rapid decline in consciousness associated with hypoxaemia and hypotension occurs in the emergency department or soon after admission.37 According to a systematic review, pulmonary oedema and echocardiographic abnormalities occur in 4% of patients admitted in good clinical condition,38 but because these complications are related to the clinical condition, in patients admitted in coma pulmonary oedema is found in one-third and echocardiographic abnormalities in one-half.37 The cardiopulmonary dysfunction can last for weeks, but even after prolonged intensive care treatment good recovery can still occur.37

723

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 724

724

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

(a)

(b)

Fig. 14.5 CT scans of a patient with a poor clinical condition on admission from a subdural haematoma. (a, b) Subarachnoid haemorrhage in the frontal interhemispheric fissure and the sylvian fissures with shift of the anterior interhemispheric fissure over the midline (arrow). The intracerebral extension (white open arrowhead) is too small to explain this midline shift. There is a large subdural haematoma (black arrowheads).

(c) (a)

(d)

(b)

(e)

Fig. 14.6 CT scans of a patient with a poor clinical condition on admission from extensive intraventricular extension of the haemorrhage and hydrocephalus. (a) Slice showing the aneurysm (open white arrowheads), intracerebral extension of the haemorrhage (closed white arrowheads) and blood in the fourth ventricle (arrow). (b) Slice showing blood in the lateral (upper arrow) and third (lower arrow) ventricle. (c) CT angiogram showing the aneurysm on the anterior communicating artery (arrowhead). (d, e) Later CT scan showing disappearance of blood from the ventricles (arrows).

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 725

14.3 General care

(b) (a) Fig. 14.7 A 70-year-old man admitted with a subarachnoid haemorrhage. On admission blood pressure was 100/50 and the pulse 50/min. He opened his eyes when spoken to, obeyed simple commands and was disoriented. Within hours he

14.3 General care

14.3.1 Blood pressure Aggressive treatment of high blood pressure carries a definite risk of ischaemia in brain areas with loss of autoregulation. The rational approach is therefore to advise against treating hypertension following aneurysmal rupture. The empirical evidence from clinical trials is sparse, but tends to support the avoidance of antihypertensive drugs. An observational study from the 1980s, in which all events had been documented by serial CT scanning, compared patients who had been treated for SAH-induced hypertension with normotensive controls; the rate of rebleeding was lower but the rate of cerebral infarction was higher than in untreated controls, despite the blood pressures being, on average, higher than in the controls.39 This suggests that hypertension after SAH is a compensatory phenomenon, at least to some extent, and one that

..

..

developed shortness of breath and lapsed into coma. (a) Chest X-ray on admission. (b) Chest X-ray several hours later showing pulmonary oedema (more prominent on the right side).

should not be interfered with. In keeping with this notion, a further observational study from the same centre (Rotterdam) suggested that the combined strategy of avoiding antihypertensive medication and increasing fluid intake decreases the risk of cerebral infarction.40 It seems best to reserve antihypertensive therapy for patients with extreme elevations of blood pressure (mean arterial pressure [(systolic blood pressure + 2 × diastolic blood pressure)/3] of 130 mmHg or over), and for those with signs of rapidly progressive end organ deterioration, such as proteinuria. An effective treatment is esmolol, an ultra-short acting, cardioselective beta-blocking drug (as a 0.5–1 mg/kg intravenous loading dose over 1 min, followed by an infusion starting at 50 µg/kg per min and increasing up to 300 µg/kg per min as necessary), or labetalol hydrochloride, a combined α1- and non-selective β-adrenergic receptor blocker (as an intravenous loading dose of 20 mg, followed by repeated incremental doses of 20–80 mg by intravenous bolus every 10 min until the desired level of blood pressure is achieved).41 It is reasonable to aim for a 25% decrease in mean arterial

725

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 726

726

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

pressure below the initial baseline. For monitoring the blood pressure in these situations an arterial line is almost mandatory. In practice, however, there is rarely any need for antihypertensive therapy. In many patients surges of high blood pressure can be attenuated by adequate pain management. Hypertension in the acute phase of subarachnoid haemorrhage can be left untreated unless the blood pressure is extremely high, or there are signs of end-organ damage.

14.3.2 Fluids and electrolytes Fluid management after subarachnoid haemorrhage (SAH) should aim to prevent a reduction in plasma volume, because this may contribute to the development of cerebral ischaemia (section 14.7). However, the arguments for a liberal (some might say aggressive) regimen of fluid administration are indirect: • Approximately one-third of SAH patients decrease their plasma volume by more than 10% between the second and tenth day after the haemorrhage; this is associated with a negative sodium balance; in other words, there is loss of sodium as well as of water – so-called ‘cerebral salt wasting’.42,43 • Fluid restriction in patients with hyponatraemia, applied in the past because it was erroneously attributed to water retention via inappropriate secretion of antidiuretic hormone, associated with an increased risk of cerebral ischaemia.44 • Two cohort studies with historical controls suggested that a daily intake of at least 3 L of saline (against 1.5–2.0 L in the past) was associated with a lower risk of delayed cerebral ischaemia and a better overall outcome.40,45 The interpretation of these two studies is difficult not only because of their observational nature, but also because the liberal administration of saline in the second period was accompanied by avoidance of antihypertensive drugs as well. Despite the incomplete evidence, it seems reasonable to prevent hypovolaemia. We aim for normovolaemia by giving 2.5–3.5 L/day of isotonic saline, unless contraindicated by signs of impending cardiac failure (see below). The amount of intravenous saline should be reduced if the patient is also receiving nutritional solution via the enteral route. In patients with fever, from whatever cause (section 14.3.3), fluid intake should be gradually increased.46 Fluid requirements may be guided by recording the central venous pressure (the directly measured value should be above 8 mmHg), especially in patients with hyponatraemia or a negative fluid balance who are at risk

of developing heart failure, but frequent calculation of fluid balance (four times a day until approximately day 10) is the main way to estimate how much fluid should be given to keep the patient normovolaemic. Pulmonary artery catheters have been advocated on the basis of historical controls,47 but in a retrospective series of 184 patients with SAH they were associated with complications such as catheter-related sepsis (13%), congestive heart failure (2%), subclavian venous thrombosis (1%) and pneumothorax (1%).48 Moreover, their usefulness in critically ill patients in general has become uncertain.49,50 The lack of evidence for prophylactic hypervolaemia instead of aiming for normovolaemia is discussed in section 14.5.7. We recommend a high fluid intake of 2.5–3.5 L per day of isotonic saline in patients with aneurysmal subarachnoid haemorrhage, to compensate for ‘cerebral salt wasting’ and to prevent hypovolaemia which predisposes to cerebral ischaemia.

14.3.3 Analgesics and general nursing care Headache Headache can sometimes be managed with just mild analgesics such as paracetamol (acetaminophen); salicylates are best avoided because of their antihaemostatic effect which is unwanted in patients who may have to undergo neurosurgical clipping of the aneurysm or external ventricular drainage; also aspirin administered after aneurysm occlusion probably does not protect against delayed cerebral ischaemia (section 14.5.4). In practice, the pain is usually so severe that codeine needs to be added, which will not mask neurological signs. Sometimes, pain can be alleviated with anxiolytic drugs such as midazolam (5 mg via an infusion pump). Often a synthetic opiate such as tramadol is needed to obtain relief (for doses see Table 14.1). As a last resort, severe headache can be treated with piritramide. Constipation is a common problem with all opiates. In patients with a decreased level of consciousness headache may present as restlessness. The ‘analgesic staircase’ consists of frequent doses of paracetamol, then codeine if necessary, tramadol or, as a last resort, piritramide. For as long as the aneurysm has not been occluded the patient should be kept on complete bed rest, traditionally flat and with as few external stimuli as possible, on the assumption that any form of excitement increases the risk of rebleeding.

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 727

14.3 General care Table 14.1 Recommendations for nursing and general management of patients with subarachnoid haemorrhage.

Nursing Continuous observation (Glasgow Coma Scale, temperature, ECG monitoring, pupils, any focal deficits) Nutrition Oral route Only with intact cough and swallowing reflexes Keep stools soft by adequate fluid intake and by restriction of milk content; if necessary add laxatives Nasogastric tube Deflate endotracheal cuff (if present) on insertion Confirm proper placement by X-ray Begin with small test feeds of 5% dextrose Prevent aspiration by feeding in sitting position and by checking gastric residue every hour Tablets should be crushed and flushed down Total parenteral nutrition should only be used as a last resort Blood pressure Do not treat hypertension unless the mean arterial pressure is ≥ 130 mmHg or there is clinical or laboratory evidence of progressive end organ damage Fluids and electrolytes Intravenous line mandatory Give at least 3 L/day (isotonic saline, 0.9%) Insert an indwelling bladder catheter Compensate for a negative fluid balance and for fever Monitor electrolytes (and white blood cell count) at least every other day Pain Start with paracetamol (acetaminophen) 500 mg every 3–4 h; avoid aspirin Midazolam can be used if pain is accompanied by anxiety (5 mg via infusion pump) For severe pain, use codeine (30–60 mg every 4 h, as needed), tramadol (50–100 mg every 4 h, as needed) or, as a last resort, piritramide 0.2–0.3 mg/kg body weight i.m. (maximum 80 mg per 24 h, in four doses). Prevention of deep venous thrombosis and pulmonary embolism Compression stockings or intermittent compression by pneumatic devices, or both before aneurysm occlusion. Low molecular heparin if the aneurysm has been treated.

Monitoring conscious level Continuous assessment of the consciousness level is essential, preferably recorded with the Glasgow Coma Scale (section 3.3.2), although it should be remembered that with inexperienced observers errors of more than one point may occur.51,52 Any change in the level of consciousness may signify early cerebral ischaemia, rebleeding, hydrocephalus or a systemic medical complication. The nurses should be familiar with the peak times of occurrence of rebleeding and cerebral ischaemia, and they should be able to detect the early signs of delirium, which is seldom caused by SAH per se but more often by the combined effects of isolation, sleep deprivation, narcotic drugs and alcohol withdrawal.

Stool softeners should be prescribed routinely. Enemas are contraindicated because they markedly increase intraabdominal pressure and secondarily intracranial pressure. The bladder A catheter is almost always needed for accurate calculation of fluid balance, except in men who are alert and have perfect bladder control. Condom devices leak or slip off too often to be of any use in these critically ill patients. Intermittent catheterization may substantially decrease the risk of urinary infections but the procedure is too stressful for patients with SAH (and requires more time from the nursing staff than most hospital budgets allow). Confusion

Constipation Coughing and straining must be rigorously prevented because of the attendant surges in arterial blood pressure.

..

..

In confused patients specific causes of confusion should be excluded, such as pain, hypoxia, alcohol withdrawal, infection or a blocked urinary catheter.

727

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 728

728

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

Venous access A reliable intravenous route should be maintained for emergency administration of plasma expanders or antiepileptic drugs. Oral administration of antacids is said to be sufficient to protect against stress bleeding; proton pump inhibitors should be considered only in mechanically ventilated patients or in patients with a previous history of ulcers. Fever The body temperature should be measured frequently, up to four times a day, depending on the time since the SAH, and the level of consciousness. After the first few hours mild fever (up to 38.5°C) often occurs, probably as the result of the inflammatory reaction in the subarachnoid space,53 in which case the pulse is characteristically normal.54 Infection should be suspected if the temperature exceeds 38.5°C, if the pulse is elevated as well, or if the patient has vomited. An elevated white cell count is not helpful in distinguishing infection from noninfective causes of pyrexia.46 Venous thromboemolism Deep venous thrombosis (DVT) is not as common after SAH as in patients with ischaemic stroke, presumably because the patients are restless, generally younger and, most important, seldom have a paralysed leg; in a large and prospective series DVT was clinically diagnosed in only 4%.55 DVT can be prevented by subcutaneous lowdose heparin or heparinoids (section 11.13), but the obvious fear is that anticoagulation will not be confined to the venous system. In a placebo-controlled trial of the low-molecular-weight heparinoid enoxaparin given after aneurysm occlusion, intracranial bleeding complications occurred more often in the treatment group, while there was no influence on overall outcome or the risk of delayed ischaemia.56 In the absence of proof of effectiveness, many physicians prefer to err on the side of caution and do not recommend prophylactic heparin or the low-molecular-weight heparinoids. Graduated compression stockings reduce the risk of DVT in patients undergoing general surgical, neurosurgical or orthopaedic procedures (section 11.13).57 Therefore, because low-molecular-weight heparinoids increase the risk of intracranial bleeding,56 graduated compression stockings are the preferred form of DVT prevention in SAH patients – although admittedly this advice lacks support from a randomized trial in this specific situation. However, because compression stockings must be individually fitted to be efficacious, some favour pneumatic devices that apply intermittent venous compression to the

legs; a study specifically performed in patients with SAH but with non-randomized controls suggested DVT can be successfully prevented in this way.58 And in a randomized trial, but in patients with intracerebral haemorrhage, the combination of stockings and intermittent pneumatic compression resulted in a smaller risk of DVT than prevention by stockings alone (relative risk 0.29; 95% CI 0.08–1.00); there is no reason to assume that this effect would be any different in patients with SAH.59 The devices are well tolerated by patients and can be managed easily by the nursing staff but one problem is that their success depends on a repair service that keeps them running. It has been argued that impedance plethysmography or duplex Doppler sonography should be performed to exclude clinically occult thrombosis before any kind of compression is applied to the calves, but no systematic studies have answered the question whether the danger of dislodging emboli to the lungs is imaginary or not. Antiepilpetic drugs The prophylactic use of anti-epileptic drugs is controversial. Seizures in the first few weeks after aneurysmal rupture occur in about 10% (6–16%) of patients, and may be easily confused with rebleeding.60–62 The majority occur soon after the initial haemorrhage.63 Predictive factors for epileptic seizures have not been consistently identified. Intracranial surgery,64 and probably also intracerebral extension of the haemorrhage, increase the risk, but a randomized trial of anti-epileptic drugs after supratentorial craniotomy for benign lesions (not only aneurysms) in 276 patients failed to show benefit in terms of seizure rate or case fatality, although the confidence intervals were wide.65 Currently, we adhere to the principle of ‘in dubio abstine’ (abstain when in doubt) because the possible disadvantage of a serious drug reaction may well outweigh any benefits. Our recommendations for the general management of patients with SAH are summarized in Table 14.1. Before considering the diagnosis and prevention of all the possible complications we will discuss the prevention of the two most common causes of clinical deterioration: rebleeding and cerebral ischaemia.

14.4 Prevention of rebleeding

14.4.1 Risk of rebleeding In the first few hours after admission for the initial haemorrhage, up to 15% of patients experience a sudden

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 729

14.4 Prevention of rebleeding

episode of clinical deterioration that suggests rebleeding (section 14.2.1). In patients who survive the first day, the risk of rebleeding is more or less evenly distributed over the next 4 weeks.66 After the first day, the cumulative risk within the first 3 weeks, and without measures to prevent it, is about 40%.67 Between 4 weeks and 6 months after the haemorrhage the risk of rebleeding gradually declines, from the initial level of 1–2%/day to a constant level of about 3%/year68 (section 14.10.2).

..

14.4.2 Endovascular occlusion

After rebleeding the prognosis is poor; 80% of patients die or remain disabled.4 Indeed, given its frequency and morbidity, rebleeding is still a major cause of poor outcome, even in centres aiming at early occlusion of the aneurysm.45,69,70 Unfortunately, there are few prognostic factors that predict an increased risk of rebleeding. Some studies suggest that it occurs more often in patients with a previous bout of sudden headache,71 with a decreased level of consciousness,72 or with larger aneurysms72–74 but none of these factors reliably identify the patients at highest risk. And even if high-risk groups could be identified, there would still be almost as many rebleeds because most occur in the larger number of patients at moderate risk – the prevention paradox again (section 18.5.1).75

Over the last decade, endovascular occlusion of aneurysms has largely replaced surgical occlusion as the intervention of choice for the prevention of rebleeding, at least in centres where this expertise is available. The technique consists of packing the aneurysm with platinum coils, nowadays with a system for controlled detachment.76 This was initially mainly used for posterior circulation aneurysms (Fig. 14.8), and then later for aneurysms on the anterior circulation (Figs 14.9 and 14.10). The procedure is generally performed under general anaesthesia, though not invariably.77 Remarkably soon after its introduction in the early 1990s coiling was compared with neurosurgical clipping in randomized trials. A systematic review of these trials (one large and two small trials, of which one was still unpublished) included a total of 2272 patients (Fig. 14.11).78 Most of the patients were in good clinical condition and had an aneurysm of the anterior circulation. After 1 year of follow-up, the relative risk of a poor outcome (death or dependency) for coiling vs clipping was 0.76 (95% CI 0.67–0.88). The absolute risk reduction of a poor outcome was 7% (95% CI 4–11%). For anterior circulation aneurysms the relative risk of a poor outcome was 0.78 (95% CI 0.68–0.90) and the absolute risk reduction was 7% (95% CI 3–10%). For posterior circulation

(a)

(b)

Fig. 14.8 (a) Catheter angiogram showing aneurysm at the tip of the basilar artery (arrow). (b) Repeat angiogram after the

aneurysm has been occluded by detachable coils, through an endovascular approach.

The risk of rebleeding after rupture of an intracranial aneurysm, without medical, endovascular or surgical intervention, is about 40% in the first 3 weeks, i.e. 1–2% per day.

..

Rebleeding of a ruptured aneurysm cannot be predicted reliably.

729

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 730

730

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

(a)

(b) Fig. 14.9 Endovascular occlusion of an aneurysm on the anterior communicating artery. Same patient as in Fig. 14.6. (a) Catheter angiogram before occlusion showing the aneurysm (arrow) and branching vessels (arrowheads). (b) Catheter angiogram after occlusion showing the coils mass in the aneurysm (arrow) and the still patent branching vessels (arrowheads).

aneurysms the relative risk was 0.41 (95% CI 0.19–0.92) and the absolute risk reduction 27% (95% CI 6–48%). Also the cognitive impairment after 1 year tends to be less after coiling than after clipping.79

Several qualifications are needed to avoid the impression that these results fully define the relative merits of endovascular and surgical occlusion of aneurysms: • First, aneurysms in some locations are more suitable for occlusion by one rather than the other technique. Basilar artery aneurysms and many other types of posterior circulation aneurysms are relatively easy to coil,80 whereas surgical treatment is often very difficult and complicated. Conversely, aneurysms at the trifurcation of the middle cerebral artery are often difficult to coil without interfering with major arterial branches. These preferences are reflected in the underrepresentation of posterior circulation aneurysms and aneurysms of the middle cerebral artery among the 2143 patients randomized in by far the largest trial, the international ISAT study.81 • Second, patients over 70 years old were underrepresented in the ISAT trial, reflecting a preference for coiling – or perhaps for no treatment – in many aged patients with SAH. However, it is not an unrealistic guess that the relative advantages of coiling over surgical treatment are even greater in patients over the age of 70 than in younger patients, given that the risk of complications from treatment of unruptured aneurysms in older patients is much higher for operation than for coiling.5 SAH patients older than 70 or 75 years of age have in general a poor prognosis, but several observational studies suggest that they can still benefit from endovascular or surgical occlusion of the aneurysm, especially if admitted in a good condition; without treatment the most important cause of a poor outcome is recurrent haemorrhage.82 • A third and last concern is the durability of the occlusion with coils, given that the early rebleeding rate after surgical occlusion is very low (section 14.10.2). Ideally, after coiling, any remaining aneursymal lumen becomes occluded by a process of reactive thrombosis, but early or late rebleeding does nevertheless occur – even after a technically competent procedure. In a consecutive series of 431 patients, six (1.4%) rebled within 30 days after coiling; in four of these patients the aneurysm had been completely occluded during the procedure.83 All six patients died after rebleeding. Risk factors for rebleeding in this series were a haematoma in the vicinity of the aneurysm, anterior communicating aneurysms, and small aneurysm size. In the ISAT trial the rate of rebleeding from the target aneurysm between the procedure and the end of the first year was somewhat higher after coiling than clipping (2.6% vs 1.0%), but the initial 7.4% absolute gain in the avoidance of death or dependence was maintained at 1 year, while the early survival advantage was still seen to 7 years.64 Most episodes of rebleeding occurred in

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 731

14.4 Prevention of rebleeding

(b)

(a)

(c)

(d) (e) Fig. 14.10 Endovascular occlusion of an aneurysm on the left middle cerebral artery. (a) Catheter angiogram before occlusion showing the aneurysm (arrow). (b) Three-dimensional reconstruction showing detailed delineation of the aneurysm with relatively wide neck (arrow). (c) Catheter angiogram after occlusion showing the coils mass in the aneurysm (arrow).

Study or sub-category Brilstra 2000b Koivisto 2000 ISAT

coil n/N

clip n/N

3/8 11/52 250/1063

4/8 14 / 57 326/1055

(d) MRA after 6 months suggests some reopening of the neck of the aneurysm (arrow). (e) Catheter angiogram after 6 months confirming the reopening of the neck (arrow). The fundus of the aneurysm is still occluded by the coils mass (arrowhead). On further follow-up angiograms the reopening remained the same and no further treatment was performed.

RR (fixed) 95% CI

Total (95% CI) 1123 1120 Total events: 264 (coil), 344 (clip) Test for heterogeneity Chi2 = 0.12, df = 2 (P = 0.94), I 2 = 0% Test for overall effect: Z = 3.83 (P = 0.0001) 0.1

Weight (%)

RR (fixed) 95% CI

1.16 3.88 94.96

0.75 [0.24, 2.33] 0.86 [0.43, 1.72] 0.76 [0.66, 0.88]

100.00

0.76 [0.67, 0.88]

0.2 0.5 1 2 5 10 Favours coiling Favours clipping

Fig. 14.11 Forest plot of Cochrane review of endovascular coiling vs surgical clipping in patients with aneurysmal subarachnoid haemorrhage (SAH). Outcome, death or dependency at 12 months after SAH. RR, relative risk.

..

..

731

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 732

732

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

the first 30 days, with either procedure. Rebleeding after the first month occurred in five of a cohort of 381 patients (1.3%) with a mean period of follow-up of 4 years.84 More data on the risk of late rebleeding after coiling will follow in the near future, but in any event this must be weighed against the risk of late rebleeding after surgical treatment, which is around 3% after 10 years.85,86 In a multicentre study that compared rebleeding after the initial year after coiling or clipping, rebleeding rates were low in both cohorts after a relatively short period of follow-up (mean 4.4 years).87 Rebleeding after 1 year occurred in one patient treated with coil embolization during 904 person-years of follow-up (annual rate 0.11%; 95% CI 0–0.63%) and in no patients treated with surgical clipping during 2666 person-years (annual rate 0%; 95% CI 0–0.14%).

3 months of therapy with antiplatelet drugs but this is supported only by comparisons with historical controls,94 not by randomized trials.95 Another complication is that one or more coils may become dislodged and herniate or embolize into the parent vessel, which again may lead to ischaemic brain damage, unless a surgical approach retrieves the fragment.96,97 Finally, rare complications include secondary infection of the thrombosed aneurysm,98 worsening of mass effect,99 intracranial artery dissection, and false aneurysms of the femoral artery. In some patients with large or complex aneurysms coiling is combined with surgical treatment,100,101 or with stenting.102 Stents may also be useful in fusiform aneurysms.103,104 Follow-up

When to coil The optimal timing of coiling has not been studied, but given the risk of rebleeding in the first few days, the general opinion is ‘the sooner the better’.88 If endovascular occlusion is not possible in the first few days, there is probably no need to avoid the period between day 4 and 12 which is regarded as unfavourable for surgical intervention (section 14.4.3), since delayed cerebral ischaemia is relatively unlikely to be precipitated by endovascular occlusion.89 The complications of coiling Bearing in mind that coiling is not a technically feasible option in every patient,90 a problem that can be reduced by the use of three-dimensional angiography,91 and despite its many advantages over surgical treatment, coiling is not without its hazards. In a large consecutive series from a single centre, procedural complications occurred in 40 of 681 patients (5.9%; 95% CI 4.2–7.9%), leading to death in 18 (procedural mortality, 2.6%; 95% CI 1.6–4.2%) and to disability in another 22 patients (procedural morbidity, 3.2%; 95% CI 2.0–4.9%).92 The most frequent complication was thromboembolism, which occurred in 4.7% of patients; intraprocedural re-rupture occurred in 1.2% of patients (Fig. 14.12). If rebleeding occurs during coiling the bleeding is much more difficult to control than during surgical exploration, which explains the high case fatality of 40%.93 Risk factors for procedural complications are small aneurysm size and the use of a temporary occlusion balloon.92,93 To reduce the risk of thromboembolism, most interventional radiologists start heparin during treatment and advise continuing for at least 1–2 days. Some advise even

There is still uncertainty about how long patients need to be followed up with imaging beyond the period of 6 months that is fairly standard, and also about the most suitable radiological technique. Conventional catheter angiography is time-consuming and carries a small but definite risk. MR angiography is feasible and gives good-quality images,105–108 but its test characteristics and effectiveness have not yet been studied in large series of patients. A final area of uncertainty is the need for a second procedure for aneurysm necks that have recanalized by impaction of the coils, by re-coiling109 or surgical occlusion.110 Since rebleeding can occur after incomplete aneurysm occlusion and re-coiling has a low risk of complications,109 currently most teams recommend re-coiling, but the risk : benefit ratio of this strategy has not yet been properly assessed. New developments are radioactive platinum coils and hydrogel-coated coils,111,112 but their usefulness has not yet been properly evaluated.

14.4.3 Surgical occlusion Surgical occlusion of ruptured aneurysms has now become the second choice, at least for most patients and most aneurysms. A statistical modelling exercise that weighed the prevention of rebleeding against the complications of the operation at current standards estimated an absolute reduction in the risk of a poor outcome of almost 10% and a relative risk reduction of 0.81 for surgical occlusion vs no occlusion.67 In fact, no proper randomized trial has ever been done in modern times; the only randomized trial that compared surgical with conservative treatment in SAH patients,

..

..

9781405127660_4_014.qxd 10/13/07 2:14 PM Page 733

14.4 Prevention of rebleeding

(c) (b) (a)

(d)

(e)

Fig. 14.12 CT scans and angiograms of a patient with rupture of the aneurysm during coiling. (a) CT on admission showing blood in the basal cisterns and fissures (arrowheads), and in the fourth ventricle (arrow). (b) Reconstruction of the CT angiogram showing an aneurysm (circle) at the origin of the posterior inferior cerebellar artery (which in this patient branches at the junction of the vertebral and basilar artery). (Also reproduced in colour in the plate section.) (c) Three-dimensional reconstruction of the catheter angiogram showing detailed delineation of the aneurysm. (d) Catheter angiogram before the coiling procedure showing the aneurysm (arrow). (e) Catheter angiogram during the coiling procedure showing the partially coiled aneurysm (arrow) and extravasation of contrast (arrowheads). (f) Catheter angiogram after occlusion showing the coils mass in the aneurysm (arrow). (g) CT scan after the procedure showing contrast material in the basal cisterns (arrowheads) and fourth ventricle (arrow).

..

..

(f)

(g)

733

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 734

734

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

performed in the 1950s in London, tested carotid ligation, not aneurysm clipping, and was restricted to patients who had survived the first 12 days.113,114 Nowadays, aneurysms are usually clipped early, i.e. within 3 days of the initial bleed – if possible within 24 h – and this policy is no longer restricted to patients admitted in good clinical condition.70 The main rationale for early surgery is, of course, optimal prevention of early rebleeding. The only randomized trial of the timing of operation, performed in Finland, allocated 216 patients to one of three groups: operation within 3 days, after 7 days, or in the intermediate period.115 The outcome tended to be better after early than intermediate or late surgery, but as the difference was not statistically significant a disadvantage of early surgery could not be excluded.115,116 The same result – no difference in outcome after early or late operation – also emerged from observational studies.117–119 Some believe that early operation also helps prevent ischaemia, because the clots that surround blood vessels in the subarachnoid space can be washed away and no longer contribute their assumed effect in the development of vasospasm, but there is little evidence to support this notion (section 14.5.8). Another theoretical advantage of early aneurysm clipping is that if cerebral ischaemia does develop hypertensive and hypervolaemic treatment can be given without the danger of rebleeding (section 14.7.2).

Although early aneurysm operation is now the usual practice when surgical clipping is indicated, this policy is vulnerable to challenge and reversal because it has not been supported by any randomized controlled trial. Neurosurgical and anaesthetic techniques have undergone considerable evolution other than through the introduction of the operating microscope. For example, three-dimensional angiography facilitates accurate placing of the occlusive clip.120 Mild intra-operative hypothermia confers no benefit.121 Although this is not a textbook about technical details of operative treatment, it is useful for physicians to know something about these developments and Table 14.2 lists some of them. A most unusual complication of operative treatment of aneurysms is a delayed type hypersensitivity against the metal alloy presenting as intense generalized pruritus, which may necessitate removal of the clip.122 It should not be assumed that surgical treatment is always definitive; rebleeding from the treated aneurysm can occur even in the first year after rupture, and also later during follow-up at which time haemorrhages from newly developed aneurysms at the site of the original aneurysm or at other sites, or from small unruptured incidental aneurysms that were left untreated at time of the subarachnoid haemorrhage, may begin to be seen (section 14.10.2).123

Vessel reconstruction, with fusiform aneurysms or giant aneurysms Permanent occlusion of parent vessel319–321 Temporary occlusion of parent vessel322,323 ‘Trapping’ without revascularization320 Opening the aneurysm for thrombectomy, followed by reconstruction of the artery319,324 Excision of aneurysm, with end-to-end anastomosis of parent artery319,325 Excision of aneurysm, with large-artery bypass, supratentorially,326,327 or infratentorially321,328 Combination with endovascular techniques Intra-operative angiography, to verify clip placement329,330 Incomplete clipping, followed by endovascular coiling331,332 Combination with endoscopy, to verify clip placement 333 Combination with intra-operative MR angiography, to verify clip placement 334 Special clips Clips with different shapes and physical properties335 Titanium clips, which do not cause artifacts on MR imaging336 Approaches to aneurysms in specific locations Distal part of posterior inferior cerebellar artery: skull-base and far-lateral transcondylar approach321 Non-branching site of intracranial internal carotid artery (‘blister-like aneurysms’): clipping or wrapping,337 or suturing followed by an encircling clip,338 or balloon occlusion followed by trapping339 Middle cerebral artery: side-to-side anastomosis after unplanned occlusion of a major branch340

Table 14.2 Special neurosurgical techniques used for occlusion of aneurysms.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 735

14.5 Prevention of delayed cerebral ischaemia

14.4.4 Pharmacological treatment

But this is not necessarily the end of the story if measures could be found to prevent the attendant ischaemia.

Antifibrinolytic drugs Because fibrinolytic activity in the cerebrospinal fluid is thought to break down clots sealing the tear in the aneurysm sac, it is logical to suppose that rebleeding might be prevented by antifibrinolytic drugs which cross the blood–brain barrier rapidly after SAH.124 The two most commonly used are tranexamic acid (TEA; 1 g intravenously or 1.5 g orally, four to six times daily), or epsilonaminocaproic acid (EACA; 3–4.5 g 3-hourly intravenously or orally). Both agents are structurally similar to lysine and so block the lysine binding sites by which the plasminogen molecules bind to complementary sites on fibrin. It is assumed that it takes 36 h to achieve complete inhibition of fibrinolysis in the cerebrospinal fluid. Ten randomized trials of antifibrinolytic drugs in SAH have so far been performed. The most recent Cochrane review included nine of them, involving 1399 patients. Treatment did not significantly influence death from all causes (OR 0.99; 95% CI 0.79–1.24) or poor outcome (death, vegetative state or severe disability; OR 1.12; 95% CI 0.88–1.43). On the other hand, there were striking differences in specific event rates: antifibrinolytic treatment reduced the risk of rebleeding at the end of followup, with some heterogeneity between the trials (OR 0.55; 95% CI 0.42–0.71), but increased the risk of cerebral ischaemia (OR 1.39; 95% CI 1.07–1.82) with considerable heterogeneity between the most recent study, in which specific treatments to prevent cerebral ischemia were used,125 and the four older studies. However, even in the trial where antifibrinolytic agents were tested in the presence of nimodipine and a strategy avoiding hypovolaemia, the drug still did not improve overall outcome.125 The tenth and most recent trial was performed in Sweden and included 505 patients; but again the overall outcome did not appreciably improve in patients treated with TEA, despite an impressive reduction in rebleeding.126 Antifibrinolytic drugs prevent rebleeding after aneurysmal rupture, but because they increase the risk of cerebral ischaemia they have no useful effect on overall outcome. How antifibrinolytic treatment precipitates cerebral ischaemia is still unclear. Possible explanations include increased blood viscosity, the development of microthrombi, delayed clearance of blood clots around the arteries at the base of the brain, and the development of hydrocephalus through delayed resorption of blood. In brief, antifibrinolytic drugs work, but they do not help.

..

..

Recombinant factor VIIa Theoretically, this activated coagulation factor might prevent rebleeding. Unfortunately, although an openlabel, dose escalation safety study showed no evidence of ischaemic complications in the first nine patients, it was suspended when the tenth patient developed middle cerebral artery branch occlusion contralateral to the aneurysm.127

14.5 Prevention of delayed cerebral ischaemia

14.5.1 Pathogenesis and risk factors Unlike ischaemic stroke as a result of disease of the extraor intracranial arteries, cerebral ischaemia or infarction after subarachnoid haemorrhage (SAH) is not usually confined to the territory of a single cerebral artery or one of its branches.128 Vasospasm is often implicated as the cause because its peak frequency from day 5 to 14 coincides with that of delayed ischaemia, and also because it is often generalized, in keeping with the multifocal or diffuse nature of the clinical manifestations and of the ischaemic lesions found on brain CT and at autopsy. Nevertheless, we have strong objections to the widespread use of the term ‘vasospasm’ as a synonym for delayed cerebral ischaemia, not so much out of fastidiousness on our part, but because the exchange of one term for the other impedes understanding and, even worse, progress. The eventual goal of treatment in this area is preventing cerebral ischaemia, not preventing arterial narrowing. There are at least five reasons why vasospasm is not synonymous with delayed cerebral ischaemia: • First, arterial narrowing does not necessarily signify contraction of smooth muscle in the arterial wall, but may represent necrosis and secondary oedema,129–131 or intimal proliferation.132,133 • Second, arterial narrowing can be asymptomatic even if severe.134–136 • Third, interpretation of the severity or even the presence of arterial narrowing on an angiogram is subject to considerable variation between observers.137 • Fourth, strategies that reduce vasospasm may not be beneficial (endothelin receptor antagonists; section

735

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 736

736

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

14.5.6) and strategies that reduce the risk of cerebral ischaemia may do so without influencing vasospasm (calcium antagonists; section 14.5.2). • Last, arterial narrowing is only one of the many factors involved in the pathogenesis of cerebral ischaemia in patients with ruptured aneurysms. One-third of patients with vasospasm do not develop cerebral ischaemia and one-third of patients with cerebral ischaemia have no vasospasm (Fig. 14.13).138 Many of the predictive factors for delayed cerebral ischaemia (Table 14.3) are interrelated or depend on

intermediate factors, and these interactions are complex and poorly understood. Also one should bear in mind that a sizeable proportion of patients who eventually develop delayed ischaemia have none of these risk factors.138 An often quoted study postulated a close relationship between the location of subarachnoid blood and the ‘thickness’ of the clot on the one hand with the occurrence of vasospasm and delayed cerebral ischaemia on the other, but these observations were based on only 41 patients.139 Moreover, the method of assessing local amounts of subarachnoid blood (called the Fisher scale,

(c)

(a)

(b)

(f)

(d)

(e)

Fig. 14.13 CT scans and CT angiograms of a patient who deteriorated 5 days after the onset of subarachnoid haemorrhage. The cause of the deterioration was secondary ischaemia; CT angiograms show unchanged diameter of the intracranial vessels of the circle of Willis. (a, b) CT scan on admission showing diffuse subarachnoid blood in the basal cisterns and fissures (white arrowheads). (c) CT angiogram on

admission with normal size of intracranial vessels and an aneurysm of the anterior communicating artery (arrow). (d, e) Repeat CT scan after the clinical deterioration showing areas of infarction in the territory of the right anterior cerebral artery (black arrowheads). (f) CT angiogram after the deterioration shows unchanged diameter of the intracranial vessels. An intraventricular drain has been inserted (arrow).

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 737

14.5 Prevention of delayed cerebral ischaemia Table 14.3 Factors at baseline and during the clinical course that have been associated with delayed cerebral ischaemia after subarachnoid haemorrhage. At baseline: Cerebral arterial narrowing on initial angiogram139,341 but also: Previous cigarette smoking?342 Previous cocaine use343 Loss of consciousness > 1 h at onset142,148 Amount of subarachnoid blood on early CT scanning142,344–347 Location of ruptured aneurysm on the anterior communicating artery or internal carotid artery,346 or on the anterior choroidal artery348 Acute hydrocephalus26,143,193,349 Perfusion deficit on initial CT scan147 4G-allele in the 4G/5G-promotor polymorphism in the plasminogen activator inhibitor-1 (PAI-1) gene350 During the clinical course: Hyponatraemia and hypovolaemia42–44 Treatment with antihypertensive drugs39 Hypotension during anaesthesia351 Treatment with antifibrinolytic drugs352 Emboli on transcranial Doppler monitoring after surgical clipping,353 but not confirmed in another study354 Early ischaemic lesions on MR scanning355 Impaired cerebral autoregulation136,356

after the last author), has wide inter-observer variation.140 Not surprisingly, therefore, in a comparison with a method of assessing the amount of blood in each of the cisterns and fissures separately in a semi-quantitative manner, this Fisher scale performed worse: it did not predict delayed cerebral ischaemia and poor clinical outcome, whereas the semi-quantitative scale did.141 Although there is little doubt that the overall amount of subarachnoid blood as determined on CT scan is a powerful predictor of delayed cerebral ischaemia, larger series have failed to show any relationship between the anatomical distribution or even the side of infarction and the site of the extravasated blood in the subarachnoid space,128,142 or at best in half the patients.143 Another awkward fact – often conveniently overlooked – is that delayed cerebral ischaemia is rare after SAH from arteriovenous malformations despite arterial narrowing in some cases,144 and also after bleeding from traumatic brain injury,145 while it does not occur at all in perimesencephalic non-aneurysmal SAH, even after matching for the amount of extravasated blood.146 It seems that subarachnoid blood is not a sufficient factor for the development of delayed cerebral ischaemia – the source has to be a ruptured artery. This suggestion is further supported by the finding that a generalized

..

..

perfusion deficit at the time of bleeding is an independent risk factor for the development of delayed cerebral ischaemia.147 Cerebral ischaemia after subarachnoid haemorrhage occurs only if the source is a ruptured aneurysm; whether or not it develops is strongly related to the total amount of subarachnoid blood, much less to the distribution of the extravasated blood. The search for chemicals that might be mediators for arterial narrowing and the development of delayed cerebral ischaemia has been intensified as several new candidate molecules have been recognized (Table 14.4). However, the multitude of these isolated and often conflicting findings attests to the paltriness of our current insights into the chain of molecular events between aneurysmal rupture and delayed cerebral ischaemia. At any rate, the relationship between the amount of extravasated blood and the development of delayed cerebral ischaemia is not sufficiently explained by the simplistic notion that toxic substances are released from clots around the large arteries at the base of the brain. Because loss of consciousness at the time of the haemorrhage is an important and independent predictive factor for

Table 14.4 Intermediate substances implicated in the pathogenesis of delayed cerebral ischaemia. For many, the relation with delayed cerebral ischaemia could not be confirmed in other studies. nitric oxide357 endothelin358–362 protein C kinase358 lipid peroxides363 free fatty acids364 soluble intercellular adhesion molecule-1 (ICAM-1)365,366 P-selectin and L-selectin367 sP-selectin368 E-selectin369 von Willebrand factor370,371 D-dimer349,372 prothrombin fragments I and II, thrombin-antithrombin III complex372 plasminogen activator inhibitor-1372,373 vascular endothelial growth factor370 inflammatory cytokines374,375 matrix metalloproteinase-9370 brain natriuretic peptide376 transferrin377 adrenomedullin (protective)378 tissue oxygen pressure379,380 lactate in CSF381,382 excitatory amino acids382

737

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 738

738

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

(c) (a)

(b)

(d)

(e)

(g)

(f)

Fig. 14.14 CT scans of a patient with subarachnoid haemorrhage from an anterior communicating artery aneurysm with long-lasting loss of consciousness after the haemorrhage. During the following days his clinical condition improved, but on the sixth day his level of consciousness decreased again. (a, b, c) CT on admission showing intracerebral extension of the haemorrhage (arrows) and blood in the left sylvian fissure (arrowhead). There are already some areas of hypodensity (black arrowheads). (d) CT angiogram on admission showing an aneurysm on the anterior communicating artery (white open arrow). The left A1 part of the anterior communicating artery is present (black arrows), but on the right side this artery is missing (arrow). (e, f, g) CT scans after the clinical deterioration showing large areas of infarction in the territories of both anterior cerebral arteries (black arrowheads).

delayed cerebral ischaemia,142,148 it is conceivable that global ischaemia during this brief period, along with a massive increase in intracranial pressure, may sensitize neurones to marginal perfusion associated with later

complications, such as arterial narrowing and hypovolaemia (Fig. 14.14). Practical measures that may help to prevent ischaemia are first of all avoidance of antihypertensive drugs

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 739

14.5 Prevention of delayed cerebral ischaemia

(section 14.3.1) and an adequate intake of fluid and sodium, although the evidence for these recommendations is rather slender (section 14.3.2). Other interventions will be discussed below.

14.5.2 Calcium antagonists A Cochrane review has assessed the trials of calcium antagonists in patients with aneurysmal SAH:149 nimodipine (eight trials, 1574 patients), nicardipine (two trials, 954 patients), AT877 (one trial, 276 patients) and magnesium (one trial, 40 patients). Overall, calcium antagonists reduced the risk of poor outcome: relative risk (RR) 0.82 (95% CI 0.72–0.93); the absolute risk reduction was 5.1%, and the corresponding number of patients needed to treat to prevent a single poor outcome event was 20. For oral nimodipine alone the RR was 0.70 (0.58–0.84). The RR of death on treatment with calcium antagonists was 0.90 (95% CI 0.76–1.07), of clinical signs of delayed cerebral ischaemia 0.67 (95% CI 0.60–0.76), and of CT or MR confirmed infarction 0.80 (95% CI 0.71–0.89). In brief, the risk reduction for ‘poor outcome’ is statistically robust, but depends mainly on trials with oral nimodipine, and especially on a single large trial;150 the evidence for nicardipine and AT877 was inconclusive (Fig. 14.15). The intermediate factors through which nimodipine exerts its beneficial effect after aneurysmal SAH remain uncertain. Calcium entry-blocking drugs were tested for this indication because they inhibit the contractile properties of smooth muscle cells, particularly in cerebral arteries. Paradoxically, but not inconsistent with their effect on delayed cerebral ischaemia, several studies of nimodipine found there was no difference between treated patients and controls in the frequency of arterial narrowing on a repeat angiogram.150–152 Therefore, at least with nimodipine, the protective effect on neurones seems to be more important than any reduction in vasospasm (section 14.5.1). The practical implication is that the regimen in the dominant nimodipine trial (60 mg orally every 4 h, for 3 weeks) is currently regarded as the standard treatment in patients with aneurysmal SAH. But, given the uncertain effect on case fatality and the possibility that the results of the meta-analysis might be affected by unpublished negative trials, the benefits of nimodipine cannot be regarded as being beyond all reasonable doubt.

Oral nimodipine probably reduces the risk of a poor outcome after subarachnoid haemorrhage by about one-third.

..

..

If the patient is unable to swallow, the nimodipine tablets should be crushed and washed down a nasogastric tube with normal saline. Intravenous administration is advocated by the manufacturer and is more expensive, but there is no evidence to support this.149 Moreover, intravenous administration of nicardipine does not improve outcome.149 The lack of effectiveness of calcium antagonists via the intravenous route is probably explained by the resulting hypotension,153 which may even be a problem if nimodipine is given orally. Therefore, if no other cause for hypotension is found, the dose of nimodipine should be at first halved (to 30 mg every 4 h) and subsequently discontinued if the blood pressure continues to fall or the patient remains hypotensive.

14.5.3 Magnesium sulphate Hypomagnesaemia occurs in more than 50% of patients with SAH and is associated with delayed cerebral ischaemia and poor outcome.154 Magnesium reduces infarct volume after experimental SAH in rats155 and its putative modes of action are inhibition of the release of excitatory amino acids and blockade of the N-methyl-Daspartate-glutamate receptor. It is also a non-competitive antagonist of voltage-dependent calcium channels and dilates cerebral arteries. Two randomized controlled trials have studied intravenous magnesium sulphate in addition to nimodipine. The smallest included only 60 patients and was inevitably inconclusive.156 The larger trial included 283 patients but was still intended as a preliminary (phase II) study with delayed cerebral ischaemia and not overall clinical outcome as the primary measure of effectiveness.157 Magnesium sulphate reduced delayed cerebral ischaemia by 34% (hazard ratio 0.66; 95% CI 0.38–1.14). After 3 months, the risk reduction for poor outcome was 23% (risk ratio 0.77; 95% CI 0.54–1.09). At that time, 18 patients in the treatment group and six in the placebo group had an excellent outcome (risk ratio 3.4; 95% CI 1.3–8.9). A phase III trial has subsequently been launched with poor outcome as the primary measure of outcome (http://www.controlled-trials.com/mrct/trial/ MAGNESIUM/2061/132459.html). Just one trial has compared intravenous nimodipine with magnesium sulphate, and found no difference in delayed cerebral ischaemia, but the trial was too small (104 patients) to draw sound conclusions.158

14.5.4 Aspirin and other antithrombotic agents Several studies have found that platelets are activated from day 3 after subarachnoid haemorrhage, mostly

739

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 740

740

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

Study or sub-category

Treatment n/N

01 nimodipine, intravenously only Subtotal (95% CI) Total events: 0 (treatment), 0 (control) Test for heterogeneity: not applicable Test for overall effect: not applicable

0

Control n/N

03 nimodipine, orally only Neil-Dwyer 1987 17/37 9/38 Petruk 1988 54/82 44 /72 Pickard 1989 91/276 55/278 Subtotal (95% Cl) 395 388 Total events: 108 (treatment), 162 (control) Test for heterogeneity: χ2 = 7.26, df = 2 (P = 0.03), I 2 = 72.5% Test for overall effect: Z = 3.77 (P = 0.0002)

05 nicardipine, orally Subtotal (95% Cl) Total events: 0 (treatment), 0 (control) Test for heterogeneity: not applicable Test for overall effect: not applicable

0

06 AT877 Shibuya 1992 33/131 Subtotal (95% Cl) 131 Total events: 33 (treatment), 41 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.90 (P = 0.37) 07 magnesium Veyna 2002 7/20 Subtotal (95% Cl) 20 Total events: 7 (treatment), 8 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.90 (P = 0.37)

Weight (%)

125/448 448

RR (fixed) 95% CI Not estimable

0

02 nimodipine, intravenously followed by orally Ohman 1991 23/109 17/104 Han 1993 23/180 17/142 Subtotal (95%Cl) 246 246 Total events: 34 (treatment), 46 (control) Test for heterogeneity: χ2 = 0.21, df = 1 (P = 0.65), I 2 = 0% Test for overall effect: Z = 0.77 (P = 0.44)

04 nicardipine, intravenously Haley 1993 118/438 Subtotal (95% Cl) 438 Total events: 118 (treatment), 125 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.32 (P = 0.75)

RR (fixed) 95% CI

6.00 5.42 11.41

0.77 [0.44, 1.36] 0.94 [0.52, 1.69] 0.85 [0.57, 1.28]

4.60 13.48 24.39 42.47

0.52 [0.26, 1.01] 0.93 [0.73, 1.18] 0.60 [0.45, 0.80] 0.70 [0.58, 0.84]

33.00 33.00

0.97 [0.78, 1.20] 0.97 [0.78, 1.20]

0

Not estimable

41/136 136

10.74 10.74

0.84 [0.57, 1.23] 0.84 [0.57, 1.23]

8 /16 16

2.37 2.37

0.70 [0.32, 1.52] 0.70 [0.32, 1.52]

100.00

0.82 [0.72, 0.93]

Total (95% CI) 1223 Total events: 300 (treatment), 382 (control) Test for heterogeneity: χ2 = 9.97, df = 7 (P = 0.19), I 2 = 29.8% Test for overall effect: Z = 3.15 (P = 0.002)

10 0.1 0.2 0.5 1 2 5 Favours treatment Favours control Fig. 14.15 Forest plot of the Cochrane review on the effectiveness of calcium antagonists on death and dependency, 3–6 months after subarachnoid haemorrhage.149 The results are

shown separately for each drug and route of administration. RR, relative risk; CI, confidence interval.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 741

14.5 Prevention of delayed cerebral ischaemia

inferred from increased levels of thromboxane B2, the stable metabolite of thromboxane A2, which promotes platelet aggregation and vasoconstriction (section 16.5.3).159,160 The question then is whether interventions aimed at counteracting platelet activation are therapeutically useful. A retrospective analysis of 242 patients who had survived the first 4 days after SAH showed that patients who had used salicylates before their haemorrhage had a significantly decreased risk of delayed cerebral ischaemia, with or without permanent deficits (relative risk 0.40; 95% CI 0.18–0.93).161 The first randomized clinical trial was reported as early as 1982 but it failed to show any benefit from aspirin;162 however, the number of patients was small,53 unoperated patients were also included, and all were treated with tranexamic acid which increases the risk of ischaemia (section 14.4.4). A further study of aspirin after early operation in 50 patients confirmed that this treatment was feasible and probably safe.163 The efficacy, however, could not be confirmed in a randomized trial of aspirin given after occlusion of the aneurysm that aimed to include 200 patients; this was prematurely stopped after the second interim analysis because by then the chances of a positive effect were negligible. At the final analysis, aspirin did not reduce the risk of delayed ischaemia (hazard ratio (HR) 1.83; 95% CI 0.8–3.9) nor that of poor outcome (HR 0.79; 95% CI 0.4–1.6).95 Three other antiplatelet drugs have been tested in patients with SAH: dipyridamole;164 the thromboxane A2 synthetase inhibitor, cataclot;165 and the experimental antiplatelet agent OKY-46.166 Ischaemic neurological deficits were reduced in one166 but not in the other trial that assessed this outcome event.165 In a systematic overview of these three trials and the three published aspirin trials, the risk of delayed cerebral ischaemia (reported in only four of the five trials) was decreased, but not to a statistically significant level (relative risk 0.81; 95% CI 0.6–1.1). For the three aspirin studies alone, the relative risk for delayed cerebral ischaemia was 1.34 (95% CI 0.8–2.3). Poor clinical outcome was not significantly different between patients treated with antiplatelet drugs and controls (relative risk 0.86; 95% CI 0.6–1.3) (Dorhout Mees et al., unpublished data). A low-molecular-weight heparinoid, enoxoparin has been tested in a trial of 170 patients but it did not improve outcome and was associated with intracranial haemorrhage in 4 of 85 patients in the experimental group.56

14.5.5 Statins HMG-CoA reductase inhibitors, or statins, are primarily used to lower LDL-cholesterol levels but they also have anti-inflammatory, immunomodulatory, antithrombotic

..

..

and vascular effects. It has often been claimed that these ‘pleiotropic effects’ contribute to the reduction in vascular events such as myocardial infarction beyond that expected from LDL-cholesterol reduction alone, although this has not been confirmed by a meta-regression analysis of the randomized trials.167 In patients with SAH, two randomized trials have been performed so far. One included only 39 patients and found that 80 mg simvastatin given within 48 h of the onset reduced ‘vasospasm’ (undefined),168 and the other enrolled 80 patients and found that 40 mg pravastatin given within 72 h reduced angiographic vasospasm and impairment of autoregulatory responses as well as vasospasmrelated ischaemic complications.169 On the other hand, an observational study found that previous use of statins increased the risk of angiographic vasospasm though not that of associated ischaemic complications.170 In conclusion, the evidence for a beneficial effect of statins after SAH is still very meagre.

14.5.6 Other drugs Tirilazad mesylate, a 21-aminosteroid free radical scavenger, has not consistently improved outcome in four randomized controlled trials, with a total of more than 3500 patients.171–174 Although, in one single subgroup (the one with the highest dosage) a beneficial effect on overall outcome was seen,171 this could not be reproduced in the corresponding subgroup from a parallel trial,172 nor in two subsequent trials.173,174 In these last two trials, only women were included and they were treated with even higher dosages than in the preceding trials, because in the first two trials women had appeared to respond less than men. A paper reporting that patients on the study drug in a small subgroup from New Zealand had less fatigue and better neuropsychological performance than controls suggests desperation rather than conviction among the investigators and the sponsoring company.175 A single trial with another hydroxyl radical scavenger, N′-propylenedinicotinamide (nicaraven), in 162 patients, showed a decrease in delayed cerebral ischaemia but not in poor clinical outcome at 3 months after SAH.176 Curiously enough, the reverse was found in a trial of 286 patients with ebselen, a seleno-organic compound with antioxidant activity through a glutathione peroxidaselike action.177 A formal overview of all the evidence for free radical scavengers is not yet available, but the case for these drugs seems weak. Nizofenone, an anionic channel blocker believed to inhibit glutamate release, was studied in a randomized trial of 100 patients, of whom only 90 were included in the analysis;178 angiographic vasospasm and poor

741

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 742

742

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

outcome were not affected, and an effect on poor outcome was found only in an on-treatment analysis in the subgroup of patients with vasospasm. The endothelinA/B receptor antagonist TAK-044 was tested in a multicentre phase II trial in 420 patients;179 there was a non-significant risk reduction of 0.8 in delayed ischaemic deficits (95% CI 0.61–1.06).

14.5.7 Increasing the plasma volume The usefulness of circulatory volume expansion to prevent delayed cerebral ischaemia after SAH has been assessed in a recent Cochrane review.180 Only one truly randomized and one quasi-randomized trial with comparable baseline characteristics for both groups could be included in the analyses. Volume expansion did not improve outcome (relative risk [RR] 1.0; 95% CI 0.5–2.2), nor delayed cerebral ischaemia (RR 1.1; 95% CI 0.5–2.2), but tended to increase the complications (RR 1.8; 95% CI 0.9–3.7). In another quasi-randomized trial, not included in the analyses, outcome was assessed only on the day of operation (7–10 days after SAH). In the period before operation, treatment resulted in a reduction of secondary ischaemia (RR 0.33; 95% CI 0.11–0.99) and case fatality (RR 0.20; 95% CI 0.07–1.2). In conclusion, volume expansion has been studied properly in only two trials of patients with aneurysmal SAH, with very small numbers, and so far there is no sound evidence to support this therapy. Because of its mineralocorticoid activity (reabsorption of sodium in the distal tubules of the kidney) fludrocortisone might, in theory, prevent a negative sodium balance, hypovolaemia and so ischaemic complications.181 A randomized study in 91 patients with SAH showed that although fludrocortisone acetate did reduce natriuresis in the first 6 days after the haemorrhage, there was no definite effect on plasma volume depletion or on ischaemic complications, although any beneficial effects may have been masked because patients in the control group were often treated with plasma expanders after they had developed clinical signs of ischaemia.182 These results were confirmed by a smaller trial in 30 patients.183 Finally, hydrocortisone was also shown in a small trial of 28 patients to prevent hyponatraemia and a drop in central venous pressure.184 The evidence from these studies is insufficiently conclusive to warrant routine fludrocortisone for all patients with SAH. Although it is reasonable to prevent a decrease in plasma volume, there is no good evidence to support prevention of cerebral ischaemia after SAH through increasing plasma volume by infusion of albumin or colloids, or by the administration of fludrocortisone.

14.5.8 Cisternal drainage and intracisternal fibrinolysis On the assumption that vasospasm increases the risk of delayed cerebral ischaemia and that extravasated blood induces vasospasm, removal of the subarachnoid blood by drainage or fibrinolysis has been suggested and then studied in several trials. For example, in a nonrandomized comparison, patients treated with lumbar drainage of CSF had cerebral infarction less often and more often returned home than patients with no lumbar drainage.185 Because no randomized trials have been performed yet, this strategy cannot be recommended for routine clinical practice. An even more aggressive way to remove subarachnoid blood is intracisternal fibrinolysis. A meta-analysis of this treatment strategy included nine trials of which only one was randomized.186 Pooled results demonstrated beneficial effects of treatment but the conclusions for practice are obviously very limited by the predominance of non-randomized studies. An open, randomized, controlled trial not yet included in the meta-analysis tested fibrinolysis in 110 patients treated with endovascular coiling.187 Urokinase was administered into the cisterna magna through a microcatheter inserted via a lumbar puncture. There was a statistically significant improvement in the primary outcome of clinical vasospasm, defined as clinical deterioration combined with evidence of vasospasm on angiography, but case fatality was not reduced even though patients in the treated group more often had a good clinical outcome. Larger studies with overall clinical outcome as the primary measurement of outcome are needed before this treatment can be implemented in clinical practice.

14.6 Management of rebleeding

14.6.1 Diagnosis Loss of consciousness was the cardinal feature of rebleeding in a prospective series of 39 patients with CT-proven rebleeding: of the 36 patients who were awake at the time, 35 lost consciousness, with preceding headache in one-third; in the remaining patient a sudden increase in headache was the only symptom.188 Serial CT scanning uncovered rebleeding in two of the three remaining patients, in whom the level of consciousness was already minimal. No ‘silent rebleeding’ was found in awake patients. In patients with a first episode of subarachnoid

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 743

14.6 Management of rebleeding Table 14.5 Specific causes of sudden deterioration in patients with aneurysmal subarachnoid haemorrhage.383 Rebleeding (two-thirds) (section 14.6.1) Epileptic seizure (section 14.3.3) Delayed cerebral ischaemia, with atypical, sudden onset (section 14.7) Ventricular fibrillation (section 14.9.3)

haemorrhage, loss of consciousness is less common, occurring in slightly less than half.55 That unconsciousness occurs more often with rebleeding than with a first rupture confirms that rebleeding has a more severe impact on the brain, and so a higher risk of a poor outcome (80%).4 In fatal episodes of rebleeding most patients immediately lose their brainstem reflexes at the same time as consciousness, but without initial apnoea. However, rebleeding is not the only cause of a sudden deterioration of consciousness in a patient with a ruptured aneurysm; other complications underlie about one-third of such episodes (Table 14.5). Therefore, once the respiratory and circulatory state of the patient has sufficiently stabilized, it is mandatory to confirm the diagnosis of rebleeding by repeat brain CT.

14.6.2 To resuscitate or not? Whether patients with rebleeding should be resuscitated and ventilated if respiratory arrest occurs is not academic; in the series of 39 patients with CT-confirmed rebleeding, 14 had initial respiratory abnormalities that called for assisted ventilation. Spontaneous respiration returned within 1 h in eight of these 14 patients, and in three more between 1 and 24 h.188 In another study of episodes of respiratory arrest but in which first bleeds were also included, whether the patient would or would not regain spontaneous respiration could not be predicted from the anatomical site of haemorrhage on CT, the initial presence or absence of brainstem reflexes, or the type of respiratory disorder.12 Many patients with initial apnoea who were successfully resuscitated later died from subsequent complications, but survival without brain damage is clearly possible – even after respiratory arrest. Rebleeding can also cause cardiac arrest, but again in this situation resuscitation is worthwhile. Cardiac function usually recovers within minutes and some patients can have a good overall outcome. Out of 11 patients with 14 episodes of well-documented cardiac arrest, six survived and three of them became independent for activities of daily living.13 After resuscitation, it will

..

..

usually become clear within a matter of hours whether the patient will indeed survive the episode or whether dysfunction of the brainstem will persist. There are no grounds to fear that resuscitation will only result in prolongation of a vegetative state and, in patients who do progress to a state of brain death, the resuscitation procedure at least allows organ donation to be considered, with some benefits to others. Sudden apnoea or cardiac arrest in a patient with subarachnoid haemorrhage usually signifies rebleeding. With full resuscitative measures cardiac function usually recovers within minutes and spontaneous respiration will rapidly return in approximately 50% of the cases, followed within hours by return of consciousness and of brainstem reflexes if these had been lost at the same time. If spontaneous respiration does not return, then organ donation should be considered.

14.6.3 Emergency occlusion of the aneurysm A large haematoma that causes brain shift without gross intraventricular haemorrhage is infrequent after rebleeding, occurring in around 10% of patients.189 In these rare cases, emergency evacuation of the haematoma after aneurysm occlusion may be indicated, as after first rupture (section 14.2.2). A more common reason for urgent aneurysm occlusion after rebleeding is the concern that, among the survivors, 50–75% will have further episodes of rebleeding if the aneurysm is left untreated.9,188 This implies that emergency clipping or coiling of the aneurysm should be seriously considered in patients who regain consciousness after rebleeding. Of course, the risk of the operation is increased after rebleeding but the risks of a wait-and-see policy at that stage seem even more intimidating. The management of rebleeding is summarized in Table 14.6.

Table 14.6 Management of rebleeding. If cardiac or respiratory arrest, resuscitate and ventilate; cardiac resuscitation is usually successful and within hours either spontaneous respiration will return or all brainstem functions will be lost Repeat CT brain scan to confirm diagnosis Consider emergency clipping or coiling of aneurysm after recovery, because many patients will have further episodes of rebleeding, with a high case fatality Large intracerebral haematomas can be evacuated at the same time

743

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 744

744

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Table 14.7 Causes of gradual deterioration after subarachnoid haemorrhage. Oedema surrounding intracerebral haematoma Delayed cerebral ischaemia (section 14.7.1) Hydrocephalus (section 14.8.1) Unsuspected rebleeding (section 14.6.1) Systemic complication (section 14.9) Hyponatraemia Disturbance of heart rhythm Hypoxia from pulmonary oedema or chest infection Hypotension Infection

14.7 Management of delayed cerebral ischaemia

14.7.1 Diagnosis The diagnosis of delayed cerebral ischaemia after SAH is far less straightforward than diagnosing ischaemic stroke, not only in daily practice, but also in research. To compound the problem, in many papers delayed cerebral ischaemia is poorly defined or not even defined at all.190 This difficulty in recognising reliably delayed cerebral ischaemia is disturbing because it is only one of several other causes of gradual deterioration in patients after SAH (Table 14.7). The clinical manifestations of delayed cerebral ischaemia evolve gradually, over several hours, and usually between day 4 and 12 after the haemorrhage. In 25% of the patients ischaemia causes hemispheric focal deficits, in another 25% a decrease in the level of consciousness, and in the remaining 50% these two features develop at the same time.128 The diagnosis should be based not only on this clinical picture, but also on the results of laboratory investigations and a repeat brain CT to exclude other intracranial causes of subacute deterioration, such as oedema surrounding an intracerebral haematoma, dilatation of the ventricular system or unsuspected rebleeding. Positive evidence of ischaemia is shown by CT in about 80% of patients after exclusion of these other causes, at least after a few days from the onset of the clinical signs of delayed cerebral ischaemia.128 The disadvantage of CT scanning is that it confirms cerebral ischaemia only some time after the clinical deterioration, which is too late for interventions that may be applied early after the onset of the ischaemia. However, currently CT scanning is the only method to confirm delayed cerebral ischaemia. One advantage of CT scanning is that it provides information about the

extent and location of the ischaemia. A recent study distinguished two patterns of infarction: single cortical infarcts, typically near the ruptured aneurysm, and multiple widespread lesions including subcortical locations and often unrelated to the site of aneurysm rupture.143 MRI is more sensitive in detecting early changes in the water content of the brain, especially with diffusionweighted imaging,191 but the acquisition time needed to obtain good images is often too long for ill and restless patients. Single-photon emission CT (SPECT) scanning may be helpful but hypoperfusion may also result from oedema around a resolving haematoma or, in the basal parts of the brain, from hydrocephalus.192,193 Transcranial Doppler sonography (TCD) may suggest impending cerebral ischaemia by the increased blood flow velocity from arterial narrowing, but there is considerable overlap with patients who do not develop ischaemia. A systematic review of studies in which TCD findings were compared with vascular imaging found that only for the middle cerebral artery was there good specificity (99%; 95% CI 98–100%) and moderate sensitivity (67%; 95% CI 48–87%).194 For other arteries, TCD was neither accurate nor useful. However, most of these data are of low methodological quality, bias cannot be ruled out and the data reporting was often uncritical. Furthermore, demonstration of arterial narrowing does not prove in itself that clinical deterioration has been caused by ischaemia. Factors such as intracranial pressure, arterial blood pressure and respiration may all influence the form of the velocity wave. Multi-slice CT angiography provides results that are in good agreement with those of catheter angiography,195,196 but the question still remains how useful is it to obtain information about arterial narrowing, given the imperfect correlation with the occurrence of brain ischaemia. In a study where vasospasm was diagnosed by TCD and catheter angiography, its presence was associated with delayed cerebral ischaemia in only two-thirds of patients (positive predictive value 0.67) and its absence with no delayed cerebral ischaemia in a similar proportion (negative predictive value 0.72).138 Because delayed cerebral ischaemia is difficult to diagnose from the outset, many physicians rely on TCD or angiographic studies to rule in or out the presence of vasospasm. However, since the negative and positive predictive values of vasospasm for cerebral ischaemia are moderate at best, knowing whether there is or is not vasospasm is not very helpful. In fact, ordering a transcranial Doppler (or even worse catheter angiography) in a patient suspected of delayed cerebral ischaemia is comparable to ordering a duplex of the carotid artery to confirm by the presence or absence of carotid artery

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 745

14.7 Management of delayed cerebral ischaemia

stenosis a transient ischaemic attack as the cause of an episode of unclear neurological symptoms.

14.7.2 Induced hypertension and volume expansion Since the 1960s, induced hypertension has been used to combat ischaemic deficits in patients with SAH. Particularly remarkable was the report of a patient in whom postoperative aphasia and right hemiparesis fluctuated with the level of the blood pressure.197 Later, induced hypertension was often combined with volume expansion, but again only in uncontrolled case series, of which just one had an acceptable definition of outcome events.198 Another case series argued that hypertensive and hypervolaemic therapy was unlikely to be successful in patients with a Glasgow Coma Score of 11 or less, as well as with hydrocephalus.199 However, on the basis of these and other case reports and series many physicians have used induced hypertension and hypervolaemia and seen their patients improve, but randomized controlled trials are sadly missing. If raising the blood pressure and increasing plasma volume can indeed reverse ischaemic deficits – which remains to be proven beyond doubt – the most plausible explanation is a defect of cerebral autoregulation that makes the perfusion of the injured brain passively dependent on the systemic blood pressure. To add haemodilution to hypertensive and hypervolaemic treatment (the so-called ‘triple-H’ regimen) is not only of uncertain benefit, but even controversial.200 The risks of deliberately increasing the arterial pressure and plasma volume include rebleeding of untreated aneurysms additional to the ruptured aneurysm, although this procedure did not in fact cause haemorrhages in a series of 40 patients with 73 unsecured additional aneurysms.201 Other complications are increased cerebral oedema, haemorrhagic transformation in areas of infarction,202 reversible leucencephalopathy,203 myocardial infarction and congestive heart failure. Certainly the circulatory system should be closely monitored, though arterial lines and pulmonary artery catheters carry their own risks: infection, pneumothorax, haemothorax, ventricular arrhythmia and pulmonary infarction.48–50 A currently recommended regimen (Table 14.8), although not supported by rigorously controlled trials, is to start with plasma volume expansion with Hetastarch or another colloid solution (5% albumin does not increase plasma volume).204 The aim is to raise the central venous filling pressure to approximately 8–12 mmHg. If there is no clinical improvement, one might consider raising the blood pressure with dopamine or dobutamine by 20–40 mmHg above pre-treatment

..

..

Table 14.8 Management of cerebral ischaemia. Immediately give 500 mL of colloid solution i.v. Consider subclavian vein catheter and maintain central venous pressure between 8 and 12 mmHg Consider increasing mean arterial pressure by 20–40 mmHg above baseline values Correct any hypoxaemia Maintain fluid intake with at least 3 L of normal saline (0.9%) per 24 h Correct hyponatraemia if severe (Table 14.10)

values. These procedures should be carried out only in an intensive care unit with facilities for specialized care and close monitoring.

14.7.3 Transluminal angioplasty and vasodilating drugs Only a few centres have reported on endovascular treatment of ‘symptomatic vasospasm’ after SAH, claiming sustained improvement in more than half the cases, but these series were uncontrolled and there must be publication bias.205 Some of these studies reported results only for arteries, and not – more relevantly – for patients. Vessel rupture is precipitated by this procedure in about 1%, even after the aneurysm has been occluded, and other complications such as hyperperfusion injury in 4%.205 In another uncontrolled series of patients treated with transluminal angioplasty, papaverine injection or both, overall clinical outcome was poor despite successful arterial dilatation. Half the patients died or remained disabled, and half the survivors had permanent deficits from cerebral infarction.206 In view of the risks, the high costs and the lack of controlled trials, transluminal angioplasty should still be regarded as a strictly experimental procedure.207 The same caution applies to uncontrolled reports of improvement of ischaemic deficits after intra-arterial infusion of drugs through super-selective catheterization. Papaverine has gained undeserved popularity,208 the more so because not all impressions are positive.209,210 Intra-arterial milrinone, verapamil or nicardipine dilate vessels but whether they improve clinical outcome is very uncertain.211–213 Calcitonin gene-related peptide (CGRP) is a potent vasodilator in the carotid vascular bed, but a randomized, multicentre, single-blind clinical trial in 62 patients with ischaemic complications after SAH failed to show any benefit in terms of overall clinical outcome: the relative risk of a poor outcome in CGRP-treated patients was 0.88 (95% CI 0.60–1.28).214

745

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 746

746

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

14.8

Management of acute hydrocephalus

14.8.1 Diagnosis About 20% of unselected patients admitted within 3 days of SAH develop acute hydrocephalus, defined as a bicaudate index above the 95th percentile for age (Fig. 14.16).26,215,216 Predisposing factors are frank intraventricular haemorrhage (Fig. 14.17),26,217,218 extensive haemorrhage in the perimesencephalic cisterns in the tentorial hiatus (Fig. 14.18)217,219 and – in only a single study – increasing age.218 Fenestration of the lamina terminalis as part of any surgical treatment may reduce the risk of hydrocephalus according to observational studies,220,221 but the increasing application of

Fig. 14.17 CT scan of a patient with hydrocephalus from intraventricular extension of the haemorrhage. The lateral (upper black arrow) and temporal horns (lower black arrow) are enlarged; there is blood in the lateral (upper white arrow) and third (lower white arrow) ventricles.

Fig. 14.16 CT brain scan of a patient with hydrocephalus (same patient as in Fig. 14.12) showing the bicaudate index (A/B), a simple and linear method for measuring the size of the ventricular system. A is the width of the frontal horns between the parallel walls of the caudate nuclei, at the level of the foramina of Monro, B the diameter of the brain at the same level. The 95th percentile for the bicaudate index is 0.16 at age 30 years or under, 0.18 at 50 years, 0.19 at 60 years, 0.21 at 80 years, and 0.25 at 100 years.

endovascular techniques for aneurysm occlusion will make this procedure less feasible. The typical presentation of acute hydrocephalus is that of a patient who is alert immediately after the initial haemorrhage, but who in the next few hours becomes increasingly drowsy, to the point that he or she only moans and localizes to pain. Nonetheless only half of all patients with acute hydrocephalus present in this way.26 In the others consciousness is impaired from the onset, or the course is unknown because the patient was alone at the time of haemorrhage. If the patient is admitted very early and secondary deterioration occurs because of hydrocephalus, serial investigation by CT may show that the level of consciousness correlates more or less inversely with the width of the lateral ventricles.27 But when different patients are compared within one series, the relationship between the level of consciousness and the degree of ventricular dilatation is rather erratic.26,215 Ocular signs do not always accompany the obtundation as a result of acute hydrocephalus; they help to corroborate but not to exclude the diagnosis. In a prospective study in which 30 of 34 patients with acute

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 747

14.8 Management of acute hydrocephalus

(a)

(b)

Fig. 14.18 CT scans of a patient with hydrocephalus from diffuse subarachnoid blood without intraventricular extension of the haemorrhage. The hydrocephalus was treated with lumbar punctures. (a, b) CT scan on admission showing diffuse

subarachnoid blood (white arrowheads) and no blood in the fourth ventricle (black arrow). The lateral and third ventricles are enlarged (white arrows). (c) Later CT scan showing disappearance of the hydrocephalus.

hydrocephalus had an impaired level of consciousness, nine of these 30 had small, non-reactive pupils, and four of these nine also showed persistent downward deviation of the eyes, with otherwise intact brainstem reflexes.26 These eye signs reflect dilatation of the proximal part of the aqueduct, which causes dysfunction of the pretectal area.222 All nine patients with non-reactive pupils had a relative ventricular size of more than 1.20 and were in coma, i.e. they did not open their eyes, obey commands or utter words.

Table 14.9 Management of acute hydrocephalus

Repeat CT scanning is required to diagnose or exclude hydrocephalus in a patient with subarachnoid haemorrhage who deteriorates within hours or days of the initial event, with or without eye signs of hydrocephalus (small, unreactive pupils and downward deviation of gaze). Patients with intraventricular blood or with extensive haemorrhage in the perimesencephalic cisterns are particularly likley to develop acute hydrocephalus.

14.8.2 Possible interventions Possible interventions are listed in Table 14.9. Wait-and-see A policy of wait-and-see for 24 h is eminently justified in patients with dilated ventricles who are alert because only about one-third of them will become symptomatic in the next few days.215 Postponing interventions for a day can also be rewarding if the level of consciousness

..

..

(c)

Consider diagnosis if level of consciousness gradually deteriorates, particularly on the first day after the bleed Repeat the CT brain scan and compare the bicaudate index with that on any previous scan Spontaneous improvement occurs within 24 h in 50% of patients (except those with massive intraventricular haemorrhage); take action if patient further deteriorates or fails to improve within 24 h Lumbar punctures are reasonably safe if there is no brain shift, and effective in about 50% of the patients who have no intraventricular obstruction External drainage of the ventricles is very effective in restoring the level of consciousness, but may increase the risk of rebleeding (consider emergency clipping or coiling at the same time), and does increase the risk of infection (this may to some degree be prevented by prophylactic antibiotics, subcutaneous tunnelling, or both)

is decreased because spontaneous improvement within this period has been documented in approximately half of the patients (7 of 13) with acute hydrocephalus who were only drowsy, and in almost half (19 of 43) who had a Glasgow Coma Score of 12/14 or worse but no massive intraventricular haemorrhage.215 It is not always easy to make a definitive decision on the need for surgical measures even after 1 day has elapsed because patients may temporarily improve to some extent but then reach a plateau or again deteriorate; such fluctuations are encountered in about one-third of patients with symptomatic hydrocephalus.215 Any further

747

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 748

748

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

deterioration in the level of consciousness warrants active intervention. Lumbar puncture Lumbar puncture was suggested as a therapeutic measure a long time ago,223 but formal studies are scarce. In a prospective but uncontrolled study, 17 patients were treated in this way because they had acute hydrocephalus with neither a haematoma nor gross intraventricular haemorrhage.28 Between one and seven lumbar punctures per patient were performed in the first 10 days, the number depending on the rate of improvement; each time a maximum of 20 mL of cerebrospinal fluid was removed, the aim being a closing pressure of 15 cmH2O. Of the 17 patients, 12 showed initial improvement: of these 12, six fully recovered, two showed incomplete improvement but fully recovered after insertion of an internal shunt, and four patients died of other complications several days after the lumbar punctures had been started. Of the five remaining patients in whom lumbar puncture had no effect, two recovered after an internal shunt and three died of other complications. Whether the risk of rebleeding is increased by lumbar punctures or drainage is uncertain. In a study of the risk of rebleeding after lumbar puncture, a series of patients with SAH and hydrocephalus treated with one or more lumbar punctures within 4 days after the haemorrhage and before aneurysm occlusion was compared with control patients with untreated hydrocephalus and with control patients without ventricular enlargement. Patients and controls were matched for interval since SAH, use of tranexamic acid, clinical condition on admission, and age. In the group treated with one or more lumbar punctures, rebleeding occurred in one of 21 patients (5%), in three of 21 controls (14%) with untreated hydrocephalus and in none of the 21 controls without hydrocephalus. Thus, this study did not confirm an increased risk of rebleeding after lumbar puncture, but because of the small number of patients it could not rule this out.224 Until randomized controlled trials are available, and we think these are still needed and ethically justifiable, the tentative conclusion is that lumbar puncture seems a safe and reasonably effective way of treating those forms of acute hydrocephalus that are not obviously caused by intraventricular obstruction. In patients deteriorating from acute hydrocephalus after subarachnoid haemorrhage it is worth trying lumbar punctures if spontaneous improvement does not occur within 24 h, and if the probable site of obstruction is in the subarachnoid space, not in the ventricular system.

External ventricular drainage External drainage of the cerebral ventricles by a catheter inserted through a burr hole is, in many centres, the most common method of treating acute hydrocephalus; the improvement is usually rapid and sometimes dramatic.26,215 Internal drainage, to the right atrium or peritoneal cavity, is rarely considered in the first few days because the blood in the cerebrospinal fluid will almost inevitably block the shunt system. However, a major concern is that the abrupt lowering of the intracranial pressure could precipitate rebleeding due to decreased transmural pressure, or removal of clot sealing the ruptured aneurysm. Therefore, the pressure of the cerebrospinal fluid after external ventricular drainage should be kept between 15 and 25 mmHg,225 although this does not abolish the risk of rebleeding.215 Aneurysm pulsatilty and size increase after ventricular drainage, which further supports the notion of an increased risk of rebleeding.226 Indeed, several studies have suggested a significantly increased risk of rebleeding in patients with external ventricular drainage, compared with patients without it.227 However, many variables that could affect the rebleeding rate, such as the timing of surgery, the timing and duration of drainage, the size of the aneurysm, as well as the severity of the initial haemorrhage, do not seem to have been adequately explored in most of these studies.227 In a recent study, the risk of rebleeding after external ventricular drainage within 4 days after the haemorrhage and before aneurysm occlusion was compared with control patients with untreated hydrocephalus, and with control patients without ventricular enlargement. Patients and controls were matched for interval since SAH, duration of exposure, use of tranexamic acid, clinical condition on admission, and age. Rebleeding occurred in seven of 34 patients treated with external ventricular drainage (21%), in seven of 34 controls (21%) with untreated hydrocephalus, and in six of 34 controls (18%) without hydrocephalus.224 Thus, this study did not confirm an increased risk, but because of the small number of patients it could not rule it out.224 Ventriculitis is a frequent complication of external drainage, especially if it is continued for more than a few days.215 But regular exchange of the intraventricular catheter, in a randomized controlled study, did not decrease the risk of infection.228 Likewise, a review of studies, all retrospective, before that trial was published found prophylactic catheter exchange did not modify the risk of developing later infection.229 Some advocate very rigid antiseptic techniques and prophylactic antibiotics,230 but without proper evidence. Implementation of a protocol for insertion and for handling the drain in

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 749

14.9 Management of systemic complications

the intensive care unit with strict surveillance of adhering to the protocol may be helpful; in one single-centre study it reduced the rate of infections compared with the period before the protocol.231 Long subcutaneous tunnelling has also been recommended, but infection has only been compared with risks reported in the literature,232 or with the risk of percutaneous drainage via Rickham reservoirs in a hospital where neurosurgeons could choose between these two methods.233 External lumbar drainage probably carries a lower risk of infection than ventricular drainage,234 although not all studies agree;235 obviously lumbar drainage cannot be used in patients with large intracerebral haematomas or extensive intraventricular haemorrhage. As with lumbar punctures and external ventricular drainage, external lumbar drainage may increase the risk of rebleeding,236 but there are no data from randomized trials. To shorten the period for which ventricular catheterization is necessary, test occlusion is often applied. But gradual weaning by sequential increases in the pressure of the external ventricular drainage system over 4 days preceding drain closure conferred no advantage, according to a randomized study of 81 patients.237

14.9 Management of systemic complications

Neurologists and neurosurgeons are regularly confronted by non-neurological complications in patients with aneurysmal subarachnoid haemorrhage: fever, anaemia, hypertension and hypotension, hyperglycaemia, hypernatraemia and hyponatraemia, hypomagnesaemia, cardiac failure and arrhythmias, and pulmonary oedema and pneumonia. More than half the patients have one or more of these complications and they are an important contributor to poor outcome.238–240 The importance of medical complications is further underlined by the finding that grading scales in which they are taken into account are more accurate predictors of outcome than scales made up only of neurological characteristics.241

14.9.1 Hyponatraemia and other electrolyte disturbances Both hyper- and, more often, hyponatraemia can occur after SAH,242 although not all studies agree.240 Hypernatraemia has been independently associated with poor outcome in some studies,242 but not in others.240 Hyponatraemia is associated with hypovolaemia and

..

..

thereby with the development of delayed cerebral ischaemia. After the syndrome of inappropriate secretion of antidiuretic hormone (SIADH) had been initially described in the 1950s,243 hyponatraemia in SAH was incorrectly attributed to this syndrome. In SIADH there is a continuing secretion of ADH, inappropriate to changes of plasma volume and osmolality. The extracellular volume increases and the expansion of the intravascular component of this volume causes a dilutional hyponatraemia. Natriuresis takes place because the volume expansion increases the glomerular filtration rate and inhibits the secretion of aldosterone. Balance studies have shown that the degree of natriuresis is relatively small and approximately equals intake. The high concentration of urinary sodium in SIADH can be simply explained by the fact that the sodium intake must be excreted in a smaller volume of urine. By contrast, hyponatraemia after SAH results from excessive natriuresis, or cerebral salt wasting.244 Serial measurement of plasma volume shows that this is decreased in most patients who developed hyponatraemia, and is preceded by a negative sodium balance in all instances.42 Plasma volume considerably decreases, even in some patients with normal sodium levels, usually as a result of excessive natriuresis. Serum ADH levels are increased or normal on admission, but decrease by the time hyponatraemia occurs. Predisposing factors for the development of hyponatraemia are hydrocephalus, particularly enlargement of the third ventricle,245 and ruptured aneurysms of the anterior communicating artery.246 Mechanical pressure on the hypothalamus can perhaps disturb sodium and water homeostasis. Four substances have been identified that are related to natriuresis and may act as intermediary factors: a digoxin-like substance,247 atrial natriuretic factor,248–252 brain natriuretic peptide252–255 and dendroaspis natriuretic peptide.256 The frequency of hyponatraemia depends on the cut-off point; if defined as a sodium level of 134 mmol or less on at least two consecutive days, it occurs in about one-third of patients.182 It develops most commonly between the second and tenth day. Severe hyponatraemia (120–124 mmol/L) occurs in 4%.42 Correction of hyponatraemia in SAH is in practice a problem of correcting volume depletion (Table 14.10). Acute symptomatic hyponatraemia is rare and requires urgent treatment with hypertonic saline (1.8% or even 3%). However, over-rapid infusion of sodium may precipitate myelinolysis in the pons and the white matter of the cerebral hemispheres.257 If possible, correction should not be faster than 8 mmol/L/day.258,259 A mild degree of hyponatraemia (125–134 mmol/L) is usually well tolerated, self-limiting and need not be treated

749

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 750

750

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Table 14.10 Management of hyponatraemia. Almost invariably caused by sodium depletion, not by sodium dilution Associated hypovolaemia increases the risk of delayed cerebral ischaemia Give isotonic saline (with or without plasma expander) or a mixture of glucose and saline; no free water If necessary, add fludrocortisone acetate, 400 mg/day in two doses, orally or intravenously Keep central venous pressure between 8 and 12 mmHg, or pulmonary capillary wedge pressure between 14 and 18 mmHg

in itself. Hyponatraemia in patients with evidence of a negative fluid balance or excessive natriuresis is corrected with saline (0.9%; sodium concentration 150 mmol/L). Hyponatraemia after subarachnoid haemorrhage usually reflects cerebral salt wasting (sodium depletion) and not secretion of antidiuretic hormone (sodium dilution). Because hyponatraemia may lead to hypovolaemia, it should not be treated with fluid restriction. Hypomagnesaemia is found in half the patients on admission, or during the clinical course.154 On admission it is related to poor clinical condition and large amounts of extravasated blood, but not with the later occurrence of delayed cerebral ischaemia or poor outcome, at least not independently. By contrast, hypomagnesaemia developing between day 2 and 12 after SAH is independently associated with the development of delayed cerebral ischaemia.154 Why hypomagnesemia occurs after SAH is unclear; renal excretion of magnesium is an unlikely explanation because hypomagnesaemia occurs within hours of the haemorrhage. Hypomagnesaemia has been linked to cardiac arrhythmias,260 and in cases of acute arrhythmia 4–8 mmol as an intravenous load in 5–10 min, followed by 25 mmol per day is recommended.260 The aim is to keep the magnesium concentration above 0.4 mmol/L. Infusion of 64 mmol per day results in concentrations between 1.0 and 2.0 mmol/L,261 and is associated with a reduced risk of developing delayed cerebral ischaemia,157 but whether this regimen results in improved overall outcome is not yet clear (section 14.5.3). Low serum potassium occurs in approximately half the patients with SAH, and is associated with an increased risk of severe Q–Tc prolongation.262 But whether hypokalaemia is associated with overall outcome is

unsettled. The influence of serum potassium on delayed cerebral ischemia and outcome has been investigated in only a single retrospective study without consecutive data collection in operated patients between 1971 and 1987; potassium levels were not related to delayed cerebral ischemia or death.263

14.9.2 Hyperglycaemia Hyperglycaemia defined as a plasma concentration of > 11.1 mmol/L is found in one-third of the patients during their clinical course, and is associated with a poor clinical condition on admission.240 From admission through day 10, glucose levels remain higher in patients with a poor outcome compared with those with a good clinical outcome.264 Hyperglycaemia is independently associated with poor outcome.264,265 Whether correction of hyperglycaemia results in improved outcome is an unresolved issue.

14.9.3 Disorders of cardiac rhythm and function Aneurysmal rupture is commonly associated with cardiac arrhythmias, ischaemia-like electrocardiographic (ECG) abnormalities, and sometimes with cardiac arrest.13 It is therefore not surprising that patients are sometimes initially misdiagnosed as having acute myocardial infarction and admitted to coronary care units. Cardiogenic shock may also occur, usually in combination with pulmonary oedema (section 14.9.4). The probable explanation is sustained sympathetic stimulation, associated with massive excretion of catecholamines,266,267 and also of cortisol and ADH.267 This may result in structural damage to the myocardium, according to echocardiographic evidence,268 raised troponin I levels,269–272 and histological features at postmortem – contraction bands, focal myocardial necrosis and subendocardial ischaemia.273 Secretion of B-type natriuretic peptide is probably a consequence of myocardial damage.274 Low magnesium levels may play an intermediary role.275 The most common ECG abnormalities in SAH (Fig. 14.19) are ST depression and elevation, T-wave changes, pathological Q waves and bundle branch block.276 Life-threatening arrhythmias such as ventricular fibrillation and ‘torsade de pointe’ may occur,277,278 but in under 5% of patients.12,13,279 A striking finding in a large series of patients investigated by serial ECGs was that every patient had at least one abnormal ECG.280 Virtually all ECG abnormalities changed to other abnormalities, in no consistent order, and then disappeared, in an observation period of 10 days.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 751

14.9 Management of systemic complications

V1

V4

V1

V4

V1

V4

V2

V5

V2

V5

V2

V5

V3

V6

V3

V6

V3

V6

Day 1

Day 0

(a)

(b)

Fig. 14.19 ECG abnormalities in a 74-year-old woman with subarachnoid haemorrhage. (a) Day 0: ischaemic ST segment, prominent U wave and prolonged Q–Tc interval. (b) Day 1: sinus bradycardia, ischaemic ST segment, ischaemic T wave,

A prolonged Q–T interval often represents delayed ventricular repolarization and predisposes to ventricular arrhythmias, but in patients with SAH it is more important as an indicator of severe intracranial disease than of potentially serious cardiac complications.280,281 Rhythm disturbances have some association with lesions of the insular cortex,282 more often with right- than with left-sided haemorrhages,3,283 an association that is not limited to ruptured aneurysms.284 Abnormalities of heart rhythm after subarachnoid haemorrhage mostly represent ‘smoke’ rather than ‘fire’ and rarely need to be treated. Generally, severe ventricular arrhythmias are brief. Betablockade has been proposed as preventive treatment aimed at lowering the sympathetic tone. In patients with head injury a double-blind, randomized study found that beta-blockers reduced catecholamine-induced cardiac necrosis, but not in-hospital case fatality.285 In patients with SAH, routine administration of beta-blockers is not warranted until there is evidence of improved overall outcome; the net benefits may be disappointing because beta-blockers lower blood pressure. Left ventricular dysfunction also occurs, most often in the initial days after the SAH.286 In a review of four studies of consecutive series of patients examined by echocardiography, irrespective of clinical condition or ECG findings, left ventricular dysfunction was found in 12% of patients.287 These patients had an increased risk of delayed cerebral ischaemia and poor outcome.

..

..

Day 9 (c)

prolonged Q–Tc interval and signs of left ventricular hypertrophy. (c) Day 9: sinus tachycardia, transient pathological Q wave and ischaemic ST segment.

14.9.4 Pulmonary oedema Pulmonary oedema occurs to some degree in approximately one-third of patients with SAH36,288 but the fulminant variety is much rarer (2% in the largest series to date).289 The onset is usually rapid, within hours of the SAH. What triggers pulmonary oedema is unclear. Raised intracranial pressure can lead to a massive sympathetic discharge, mediated by the anterior hypothalamus – the same location that is held responsible for electrolyte disturbances (section 14.9.1). This would lead to increased permeability of endothelial cells in the pulmonary capillary bed. Any cardiac failure may aggravate the pulmonary oedema.36 The typical clinical picture consists of unexpected dyspnoea, cyanosis and the production of pink and frothy sputum. Many patients are pale, sweat excessively and are hypertensive. A chest X-ray usually demonstrates impressive pulmonary oedema (Fig. 14.7) which may disappear in a matter of hours following positive end-expiratory pressure ventilation. A problem is that liberal administration of fluids is beneficial for brain perfusion but may delay recovery of pulmonary oedema and hence impair brain oxygenation, and that positive end-expiratory pressure ventilation increases intracranial pressure. If there is associated left ventricular failure, clinically manifested by sudden hypotension following initially elevated blood pressure, transient lactic acidosis and mild elevation of the creatine kinase MB fraction,37,290 inotropic agents are indicated (intravenous dobutamine, 5–15 µg/kg/min).291

751

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 752

752

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

14.10 Late sequelae and complications

In general, functional outcome improves significantly between 4 months and 18 months after SAH with improved quality of life.292 Nevertheless, specific late complications require attention. Therefore, we favour a multidisciplinary clinic for patients who have had an SAH where neurologists and rehabilitation physicians are at hand.

14.10.1 Vitreous haemorrhages Preretinal haemorrhages associated with SAH (section 3.7.1) may break into the vitreous cavity (Terson syndrome). This has been documented in 13% of patients in prospective studies;293 the rates were three times as high in patients who had lost consciousness at some stage.294 These haemorrhages occur in one or both eyes, usually at the time of the initial haemorrhage, but sometimes several days later, and then mostly in association with rebleeding. It may, however, take days or weeks before the patient is sufficiently alert to complain about blurred vision. In most cases, the vitreous haemorrhage clears spontaneously in a matter of weeks to months. If not, vitrectomy may be indicated to improve vision.295 Operation is unnecessary if preretinal haemorrhages have not involved the vitreous body.296

14.10.2 New episodes of subarachnoid haemorrhage When the ruptured aneurysm is left untreated, the rate of rebleeding after the first 6 months have passed is 3.5% per year during the first decade, but of course this situation is now increasingly unusual.297 However, even in patients with successfully occluded ruptured aneurysms new episodes of bleeding do occur – from de novo aneurysms, incidental aneurysms that were left untreated at the time of the initial haemorrhage, or from regrowth of the aneurysm that caused the first bleed. From the patient’s perspective the source of a new episode is not very relevant, but from a management perspective it is. In a study of 30 patients with late rebleeding (after an average period of more than 7 years) the ruptured aneurysm had developed at a new location in somewhat more than half the cases, it had regrown at the previous aneurysm site in about a quarter, and in the rest the source of bleeding was an additional aneurysm that had been overlooked at the time of the first episode.298 In the surgical arm of the ISAT trial rebleeding from the clipped aneurysm occurred in 1% of the operated

patients within the first year.64 A long-term study in Japan estimated the rebleeding rate after initially successful clipping was 2.2% after 10 years and 9% after 20 years,85 while in a Dutch cohort the rate of new episodes of SAH (from newly developed aneurysms, incidental aneurysms that were left untreated and from regrowth aneurysms combined) was 3.2% after 10 years.86 Despite this relatively high risk of new episodes of SAH after successful surgical treatment of the initial episode, a modelling study estimated that repeated screening and treatment of newly detected aneurysms could not be recommended, because the prevented episodes of new episodes of SAH would be outweighed by the complications of the diagnostic and therapeutic procedures. For patients with an increased risk of both aneurysm development and aneurysm rupture, screening was beneficial in this model, but at present we cannot identify who these patients are (section 15.2.5). After endovascular occlusion of the aneurysm, there is still little information about the rate of rebleeding in the long term (section 14.4.2).

14.10.3 Late hydrocephalus Permanent shunting for hydrocephalus persisting for 2–4 weeks after the initial haemorrhage is needed in approximately 20% of patients, according to two neurosurgical series.299,300 Aneurysm site (anterior communicating artery complex or posterior circulation) is an established risk factor for late hydrocephalus.301,302 Putative risk factors, identified only in single studies, are age,303 intraventricular haemorrhage301 and the clearance rate of subarachnoid blood.304 The risk of chronic hydrocephalus is no lower after endovascular treatment than after surgical clipping.305–307 If fenestration of the lamina terminalis is performed, as part of the microsurgical procedure for aneurysm repair, late hydrocephalus is a rare event.308

14.10.4 Epilepsy New-onset epilepsy (two or more unprovoked seizures) in the first year after discharge from hospital was reported in 17 of 242 patients (7%) from Columbia University, New York, while an additional 10 patients (4%) had a single seizure.309 In a British cohort late epilepsy occurred in 4.9% in the first year after the haemorrhage (23 of 872 patients).310 Unfortunately neither cohort recorded a possible relationship with seizures at the time of the haemorrhage or during hospital stay; this is important because in a small cohort of 95 patients from Detroit all eight with epileptic seizures after discharge fell into this category.311 Putative risk factors (identified

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 753

References

in single studies only) include focal lesions such as subdural haematoma and cerebral infarction,309,312 as well as disability on discharge,310 which of course is largely a consequence of structural brain damage. De novo epileptic seizures are relatively rare after coiling.64,313

14.10.5 Cognitive dysfunction and psychosocial impact It is important to appreciate just who should be asked if a patient has ‘recovered’. The patients rate their own outcome worse than the treating physician, but better than their spouses.314 Even patients who seem to make a good functional recovery and who are independent in activities of daily living often can have psychosocial and cognitive deficits in the first year after the SAH.3,315,316 Although improvement still occurs between 4 and 18 months after the SAH, many patients and their partners still experience reduced quality of life 1–2 years after the haemorrhage.317 The psychosocial impact at even longer follow-up is considerable. In a survey of 610 patients who were interviewed about 9 years after their SAH, a quarter of the employed patients had stopped working and another quarter worked shorter hours or had a position with less responsibility.318 On average, patients returned to work about 9 months after discharge (range 0–96). A total of 60% of the patients reported changes in personality, most commonly increased irritability (37%) or emotionality (29%). The patients had a statistically significant higher depression score than the reference population, and only 25% reported a complete recovery without any psychosocial or neurological problems.

5

6

7

8

9

10

11

12

13

14

References

1 Hop JW, Rinkel GJE, Algra A, van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke 1997; 28(3):660–4. 2 Huang J, Van Gelder JM. The probability of sudden death from rupture of intracranial aneurysms: a meta-analysis. Neurosurgery 2002; 51:1101–5. 3 Hackett ML, Anderson CS, for the ACROSS Group. Health outcomes 1 year after subarachnoid hemorrhage: an international population-based study. Neurology 2000; 55:658–62. 4 Roos YBWEM, de Haan RJ, Beenen LFM, Groen RJM, Albrecht KW, Vermeulen M. Complications and outcome in patients with aneurysmal subarachnoid haemorrhage: a prospective hospital based cohort study in the

..

..

15

16

17

18

Netherlands. J Neurol Neurosurg Psychiatry 2000; 68:337–41. Wiebers DO, Whisnant JP, Huston J, III, Meissner I, Brown RD, Jr., Piepgras DG et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362:103–10. Stegmayr B, Eriksson M, Asplund K. Declining mortality from subarachnoid hemorrhage: changes in incidence and case fatality from 1985 through 2000. Stroke 2004; 35:2059–63. Bardach NS, Zhao SJ, Gress DR, Lawton MT, Johnston SC. Association between subarachnoid hemorrhage outcomes and number of cases treated at California hospitals. Stroke 2002; 33:1851–6. Sasaki T, Sato M, Oinuma M, Sakuma J, Suzuki K, Matsumoto M et al. Management of poor-grade patients with aneurysmal subarachnoid hemorrhage in the acute stage: importance of close monitoring for neurological grade changes. Surg Neurol 2004; 62:531–5. Inagawa T, Kamiya K, Ogasawara H, Yano T. Rebleeding of ruptured intracranial aneurysms in the acute stage. Surg Neurol 1987; 28(2):93–9. Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Koike T, Tanaka R. Ultra-early rebleeding in spontaneous subarachnoid hemorrhage. J Neurosurg 1996; 84:35–42. Ohkuma H, Tsurutani H, Suzuki S. Incidence and significance of early aneurysmal rebleeding before neurosurgical or neurological management. Stroke 2001; 32:1176–80. Hijdra A, Vermeulen M, van Gijn J, van Crevel H. Respiratory arrest in subarachnoid hemorrhage. Neurology 1984; 34(11):1501–3. Toussaint LG, III, Friedman JA, Wijdicks EFM, Piepgras DG, Pichelmann MA, McIver JI et al. Survival of cardiac arrest after aneurysmal subarachnoid hemorrhage. Neurosurgery 2005; 57:25–31. Van Gijn J, van Dongen KJ. The time course of aneurysmal haemorrhage on computed tomograms. Neuroradiology 1982; 23(3):153–6. Hauerberg J, Eskesen V, Rosenorn J. The prognostic significance of intracerebral haematoma as shown on CT scanning after aneurysmal subarachnoid haemorrhage. Br J Neurosurg 1994; 8(3):333–9. Heiskanen O, Poranen A, Kuurne T, Valtonen S, Kaste M. Acute surgery for intracerebral haematomas caused by rupture of an intracranial arterial aneurysm: a prospective randomized study. Acta Neurochir (Wien) 1988; 90(3–4):81–3. Niemann DB, Wills AD, Maartens NF, Kerr RS, Byrne JV, Molyneux AJ. Treatment of intracerebral hematomas caused by aneurysm rupture: coil placement followed by clot evacuation. J Neurosurg 2003; 99:843–7. Smith ER, Carter BS, Ogilvy CS. Proposed use of prophylactic decompressive craniectomy in poor-grade aneurysmal subarachnoid hemorrhage patients presenting with associated large sylvian hematomas. Neurosurgery 2002; 51:117–24.

753

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 754

754

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

19 Inamasu J, Saito R, Nakamura Y, Ichikizaki K, Suga S, Kawase T et al. Acute subdural hematoma caused by ruptured cerebral aneurysms: diagnostic and therapeutic pitfalls. Resuscitation 2002; 52:71–6. 20 Gelabert-Gonzalez M, Iglesias-Pais M, Fernandez-Villa J. Acute subdural haematoma due to ruptured intracranial aneurysms. Neurosurg Rev 2004; 27:259–62. 21 Linn FHH, Rinkel GJE, Algra A, van Gijn J. The notion of ‘warning leaks’ in subarachnoid haemorrhage: are such patients in fact admitted with a rebleed? J Neurol Neurosurg Psychiatry 2000; 68:332–6. 22 Nonaka Y, Kusumoto M, Mori K, Maeda M. Pure acute subdural haematoma without subarachnoid haemorrhage caused by rupture of internal carotid artery aneurysm. Acta Neurochir (Wien) 2000; 142:941–4. 23 Krishnaney AA, Rasmussen PA, Masaryk T. Bilateral tentorial subdural hematoma without subarachnoid hemorrhage secondary to anterior communicating artery aneurysm rupture: a case report and review of the literature. Am J Neuroradiol 2004; 25:1006–7. 24 Koerbel A, Ernemann U, Freudenstein D. Acute subdural haematoma without subarachnoid haemorrhage caused by rupture of an internal carotid artery bifurcation aneurysm: case report and review of literature. Br J Radiol 2005; 78:646–50. 25 O’Sullivan MG, Whyman M, Steers JW, Whittle IR, Miller JD. Acute subdural haematoma secondary to ruptured intracranial aneurysm: diagnosis and management. Br J Neurosurg 1994; 8(4):439–45. 26 Van Gijn J, Hijdra A, Wijdicks EFM, Vermeulen M, van Crevel H. Acute hydrocephalus after aneurysmal subarachnoid hemorrhage. J Neurosurg 1985; 63(3):355–62. 27 Rinkel GJE, Wijdicks EFM, Ramos LMP, van Gijn J. Progression of acute hydrocephalus in subarachnoid haemorrhage: a case report documented by serial CT scanning. J Neurol Neurosurg Psychiatry 1990; 53(4):354–5. 28 Hasan D, Lindsay KW, Vermeulen M. Treatment of acute hydrocephalus after subarachnoid hemorrhage with serial lumbar puncture. Stroke 1991; 22(2):190–4. 29 Suzuki M, Otawara Y, Doi M, Ogasawara K, Ogawa A. Neurological grades of patients with poor-grade subarachnoid hemorrhage improve after short-term pretreatment. Neurosurgery 2000; 47:1098–104. 30 Roos YBWEM, Hasan D, Vermeulen M. Outcome in patients with large intraventricular haemorrhages: a volumetric study. J Neurol Neurosurg Psychiatry 1995; 58:622–4. 31 Nieuwkamp DJ, de Gans K, Rinkel GJE, Algra A. Treatment and outcome of severe intraventricular extension in patients with subarachnoid or intracerebral hemorrhage: a systematic review of the literature. J Neurol 2000; 247:117–21. 32 Naff NJ, Carhuapoma JR, Williams MA, Bhardwaj A, Ulatowski JA, Bederson J et al. Treatment of intraventricular hemorrhage with urokinase: effects on 30-day survival. Stroke 2000; 31:841–7. 33 Shimoda M, Oda S, Shibata M, Tominaga J, Kittaka M, Tsugane R. Results of early surgical evacuation of packed

34

35

36

37

38

39

40

41 42

43

44

45

46

47

48

intraventricular hemorrhage from aneurysm rupture in patients with poor-grade subarachnoid hemorrhage. J Neurosurg 1999; 91:408–14. Lagares A, Putman CM, Ogilvy CS. Posterior fossa decompression and clot evacuation for fourth ventricle hemorrhage after aneurysmal rupture: case report. Neurosurgery 2001; 49:208–11. Yoshimoto Y, Ochiai C, Kawamata K, Endo M, Nagai M. Aqueductal blood clot as a cause of acute hydrocephalus in subarachnoid hemorrhage. Am J Neuroradiol 1996; 17:1183–6. McLaughlin N, Bojanowski MW, Girard F, Denault A. Pulmonary edema and cardiac dysfunction following subarachnoid hemorrhage. Can J Neurol Sci 2005; 32:178–85. Parr MJ, Finfer SR, Morgan MK. Reversible cardiogenic shock complicating subarachnoid haemorrhage. Br Med J 1996; 313(7058):681–3. Macrea LM, Tramer MR, Walder B. Spontaneous subarachnoid hemorrhage and serious cardiopulmonary dysfunction: a systematic review. Resuscitation 2005; 65:139–48. Wijdicks EFM, Vermeulen M, Murray GD, Hijdra A, van Gijn J. The effects of treating hypertension following aneurysmal subarachnoid hemorrhage. Clin Neurol Neurosurg 1990; 92(2):111–17. Hasan D, Vermeulen M, Wijdicks EFM, Hijdra A, van Gijn J. Effect of fluid intake and antihypertensive treatment on cerebral ischemia after subarachnoid hemorrhage. Stroke 1989; 20(11):1511–15. Varon J, Marik PE. Clinical review: the management of hypertensive crises. Crit Care 2003; 7:374–84. Wijdicks EFM, Vermeulen M, ten Haaf JA, Hijdra A, Bakker WH, van Gijn J. Volume depletion and natriuresis in patients with a ruptured intracranial aneurysm. Ann Neurol 1985; 18(2):211–16. Hasan D, Wijdicks EFM, Vermeulen M. Hyponatremia is associated with cerebral ischemia in patients with aneurysmal subarachnoid hemorrhage. Ann Neurol 1990; 27(1):106–8. Wijdicks EFM, Vermeulen M, Hijdra A, van Gijn J. Hyponatremia and cerebral infarction in patients with ruptured intracranial aneurysms: is fluid restriction harmful? Ann Neurol 1985; 17(2):137–40. Vermeij FH, Hasan D, Bijvoet HWC, Avezaat CJJ. Impact of medical treatment on the outcome of patients after aneurysmal subarachnoid hemorrhage. Stroke 1998; 29:924–30. Oliveira-Filho J, Ezzeddine MA, Segal AZ, Buonanno FS, Chang Y, Ogilvy CS et al. Fever in subarachnoid hemorrhage: relationship to vasospasm and outcome. Neurology 2001; 56:1299–304. Kim DH, Haney CL, Van Ginhoven G. Reduction of pulmonary edema after SAH with a pulmonary artery catheter-guided hemodynamic management protocol. Neurocrit Care 2005; 3:11–15. Rosenwasser RH, Jallo JI, Getch CC, Liebman KE. Complications of Swan-Ganz catheterization for

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 755

References

49

50

51

52

53

54

55

56

57

58

59

60

61

62 63

..

..

hemodynamic monitoring in patients with subarachnoid hemorrhage. Neurosurgery 1995; 37:872–5. Harvey S, Harrison DA, Singer M, Ashcroft J, Jones CM, Elbourne D et al. Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet 2005; 366:472–7. Shah MR, Hasselblad V, Stevenson LW, Binanay C, O’Connor CM, Sopko G et al. Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. J Am Med Assoc 2005; 294:1664–70. Rowley G, Fielding K. Reliability and accuracy of the Glasgow Coma Scale with experienced and inexperienced users. Lancet 1991; 337(8740):535–8. Gill MR, Reiley DG, Green SM. Interrater reliability of Glasgow Coma Scale scores in the emergency department. Ann Emerg Med 2004; 43:215–23. Takagi K, Tsuchiya Y, Okinaga K, Hirata M, Nakagomi T, Tamura A. Natural hypothermia immediately after transient global cerebral ischemia induced by spontaneous subarachnoid hemorrhage. J Neurosurg 2003; 98:50–56. Rousseaux P, Scherpereel R, Bernard MH, Graftieaux JP, Guyot JF. Fever and cerebral vasospasm in intracranial aneurysms. Surg Neurol 1980; 14:459–65. Vermeulen M, Lindsay KW, Murray GD, Cheah F, Hijdra A, Muizelaar JP et al. Antifibrinolytic treatment in subarachnoid hemorrhage. N Engl J Med 1984; 311(7):432–7. Siironen J, Juvela S, Varis J, Porras M, Poussa K, Ilveskero S et al. No effect of enoxaparin on outcome of aneurysmal subarachnoid hemorrhage: a randomized, double-blind, placebo-controlled clinical trial. J Neurosurg 2003; 99:953–9. Amaragiri SV, Lees TA. Elastic compression stockings for prevention of deep vein thrombosis. Cochrane Database Syst Rev 2000; CD001484. Black PM, Crowell RM, Abbott WM. External pneumatic calf compression reduces deep venous thrombosis in patients with ruptured intracranial aneurysms. Neurosurgery 1986; 18(1):25–8. Lacut K, Bressollette L, Le Gal G, Etienne E, De Tinteniac A, Renault A et al. Prevention of venous thrombosis in patients with acute intracerebral hemorrhage. Neurology 2005; 65:865–9. Hart RG, Byer JA, Slaughter JR, Hewett JE, Easton JD. Occurrence and implications of seizures in subarachnoid hemorrhage due to ruptured intracranial aneurysms. Neurosurgery 1981; 8(4):417–21. Hasan D, Schonck RS, Avezaat CJ, Tanghe HL, van Gijn J, van der Lugt PJ. Epileptic seizures after subarachnoid hemorrhage. Ann Neurol 1993; 33(3):286–91. Pinto AN, Canhao P, Ferro JM. Seizures at the onset of subarachnoid haemorrhage. J Neurol 1996; 243(2):161–4. Butzkueven H, Evans AH, Pitman A, Leopold C, Jolley DJ, Kaye AH et al. Onset seizures independently predict poor outcome after subarachnoid hemorrhage. Neurology 2000; 55:1315–20.

64 Molyneux AJ, Kerr RS, Yu LM, Clarke M, Sneade M, Yarnold JA et al. International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet 2005; 366:809–17. 65 Foy PM, Chadwick DW, Rajgopalan N, Johnson AL, Shaw MD. Do prophylactic anticonvulsant drugs alter the pattern of seizures after craniotomy? J Neurol Neurosurg Psychiatry 1992; 55(9):753–7. 66 Brilstra EH, Rinkel GJE, Algra A, van Gijn J. Rebleeding, secondary ischemia, and timing of operation in patients with subarachnoid hemorrhage. Neurology 2000; 55:1656–60. 67 Brilstra EH, Algra A, Rinkel GJE, Tulleken CAF, van Gijn J. Effectiveness of neurosurgical clip application in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 97:1036–41. 68 Jane JA, Kassell NF, Torner JC, Winn HR. The natural history of aneurysms and arteriovenous malformations. J Neurosurg 1985; 62(3):321–3. 69 Roos YBWEM, Beenen LF, Groen RJ, Albrecht KW, Vermeulen M. Timing of surgery in patients with aneurysmal subarachnoid haemorrhage: rebleeding is still the major cause of poor outcome in neurosurgical units that aim at early surgery. J Neurol Neurosurg Psychiatry 1997; 63:490–3. 70 Laidlaw JD, Siu KH. Ultra-early surgery for aneurysmal subarachnoid hemorrhage: outcomes for a consecutive series of 391 patients not selected by grade or age. J Neurosurg 2002; 97:250–8. 71 Beck J, Raabe A, Szelenyi A, Berkefeld J, Gerlach R, Setzer M et al. Sentinel headache and the risk of rebleeding after aneurysmal subarachnoid hemorrhage. Stroke 2006; 37:2733–2737. 72 Naidech AM, Janjua N, Kreiter KT, Ostapkovich ND, Fitzsimmons BF, Parra A et al. Predictors and impact of aneurysm rebleeding after subarachnoid hemorrhage. Arch Neurol 2005; 62:410–16. 73 Inagawa T, Hirano A. Ruptured intracranial aneurysms: an autopsy study of 133 patients. Surg Neurol 1990; 33(2):117–23. 74 Pleizier CM, Ruigrok YM, Rinkel GJE. Relation between age and number of aneurysms in patients with subarachnoid haemorrhage. Cerebrovasc Dis 2002; 14:51–3. 75 Rose G. The Strategy of Preventive Medicine. Oxford: Oxford Medical Publications, 1992. 76 Guglielmi G, Vinuela F, Duckwiler G, Dion J, Lylyk P, Berenstein A et al. Endovascular treatment of posterior circulation aneurysms by electrothrombosis using electrically detachable coils. J Neurosurg 1992; 77(4):515–24. 77 Qureshi AI, Suri MFK, Khan J, Kim SH, Fessler RD, Ringer AJ et al. Endovascular treatment of intracranial aneurysms by using Guglielmi detachable coils in awake patients: safety and feasibility. J Neurosurg 2001; 94:880–5.

755

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 756

756

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

78 Van der Schaaf IC, Algra A, Wermer MJH, Molyneux A, Clarke M, van Gijn J et al. Endovascular coiling versus neurosurgical clipping for patients with aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2005; (4) CD003085. 79 Hadjivassiliou M, Tooth CL, Romanowski CAJ, Byrne J, Battersby RDE, Oxbury S et al. Aneurysmal SAH: cognitive outcome and structural damage after clipping or coiling. Neurology 2001; 56:1672–7. 80 Lozier AP, Connolly ES, Jr., Lavine SD, Solomon RA. Guglielmi detachable coil embolization of posterior circulation aneurysms: a systematic review of the literature. Stroke 2002; 33:2509–18. 81 Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002; 360:1267–74. 82 Nieuwkamp DJ, Rinkel GJE, Silva R, Greebe P, Schokking DA, Ferro JM. Subarachnoid haemorrhage in patients older than 75 years: clinical course, treatment and outcome. J Neurol Neurosurg Psychiatry 2006; 77:933–7. 83 Sluzewski M, Van Rooij WJ. Early rebleeding after coiling of ruptured cerebral aneurysms: incidence, morbidity, and risk factors. Am J Neuroradiol 2005; 26:1739–43. 84 Sluzewski M, Van Rooij WJ, Beute GN, Nijssen PC. Late rebleeding of ruptured intracranial aneurysms treated with detachable coils. Am J Neuroradiol 2005; 26:2542–9. 85 Tsutsumi K, Ueki K, Usui M, Kwak S, Kirino T. Risk of recurrent subarachnoid hemorrhage after complete obliteration of cerebral aneurysms. Stroke 1998; 29:2511–13. 86 Wermer MJH, Greebe P, Algra A, Rinkel GJE. Incidence of recurrent subarachnoid hemorrhage after clipping for ruptured intracranial aneurysms. Stroke 2005; 36:2394–9. 87 CARAT Investigators. Rates of delayed rebleeding from intracranial aneurysms are low after surgical and endovascular treatment. Stroke 2006; 37:1437–42. 88 Baltsavias GS, Byrne JV, Halsey J, Coley SC, Sohn MJ, Molyneux AJ. Effects of timing of coil embolization after aneurysmal subarachnoid hemorrhage on procedural morbidity and outcomes. Neurosurgery 2000; 47:1320–9. 89 Rabinstein AA, Pichelmann MA, Friedman JA, Piepgras DG, Nichols DA, McIver JI et al. Symptomatic vasospasm and outcomes following aneurysmal subarachnoid hemorrhage: a comparison between surgical repair and endovascular coil occlusion. J Neurosurg 2003; 98:319–25. 90 Shanno GB, Armonda RA, Benitez RP, Rosenwasser RH. Assessment of acutely unsuccessful attempts at detachable coiling in intracranial aneurysms. Neurosurgery 2001; 48:1066–72. 91 Albuquerque FC, Spetzler RF, Zabramski JM, McDougall CG. Effects of three-dimensional angiography on the coiling of cerebral aneurysms. Neurosurgery 2002; 51:597–605. 92 Van Rooij WJ, Sluzewski M, Beute GN, Nijssen PC. Procedural complications of coiling of ruptured

93

94

95

96

97

98

99

100

101

102

103

104

intracranial aneurysms: incidence and risk factors in a consecutive series of 681 patients. Am J Neuroradiol 2006; 27:1498–501. Sluzewski M, Bosch JA, Van Rooij WJ, Nijssen PCG, Wijnalda D. Rupture of intracranial aneurysms during treatment with Guglielmi detachable coils: incidence, outcome, and risk factors. J Neurosurg 2001; 94:238–40. Ries T, Buhk JH, Kucinski T, Goebell E, Gryska U, Zeumer H et al. Intravenous administration of acetylsalicylic acid during endovascular treatment of cerebral aneurysms reduces the rate of thromboembolic events. Stroke 2006; 37:1816–21. Van den Bergh WM, Algra A, Dorhout Mees SM, van Kooten F, Dirven CM, van Gijn J et al. Randomized controlled trial of acetylsalicylic acid in aneurysmal subarachnoid hemorrhage. The MASH Study. Stroke 2006; 37:2326–30. Spetzger U, Reul J, Thron A, Warnke JP, Gilsbach JM. Microsurgical embolectomy and removal of a migrated coil from the middle cerebral artery. Cerebrovasc Dis 1997; 7:226–31. Raftopoulos C, Goffette P, Billa RF, Mathurin P. Transvascular coil hooking procedure to retrieve an unraveled Guglielmi detachable coil: Technical note. Neurosurgery 2002; 50:912–14. Kirollos RW, Bosma JJD, Radhakrishnan J, Pigott TDJ. Endovascularly treated cerebral aneurysm using Guglielmi detachable coils acting as a nidus for brain abscess formation secondary to Salmonella bacteremia: case report. Neurosurgery 2002; 51:234–7. Russell SM, Nelson PK, Jafar JJ. Neurological deterioration after coil embolization of a giant basilar apex aneurysm with resolution following parent artery clip ligation: case report and review of the literature. J Neurosurg 2002; 97:705–8. Hacein-Bey L, Connolly ES, Jr., Mayer SA, Young WL, PileSpellman J, Solomon RA. Complex intracranial aneurysms: Combined operative and endovascular approaches. Neurosurgery 1998; 43:1304–12. Lawton MT, Quinone-Hinojosa A, Sanai N, Malek JY, Dowd CF. Combined microsurgical and endovascular management of complex intracranial aneurysms. Neurosurgery 2003; 52:263–74. Lanzino G, Wakhloo AK, Fessler RD, Hartney ML, Guterman LR, Hopkins LN. Efficacy and current limitations of intravascular stents for intracranial internal carotid, vertebral, and basilar artery aneurysms. J Neurosurg 1999; 91:538–46. Higashida RT, Smith W, Gress D, Urwin R, Dowd CF, Balousek PA et al. Intravascular stent and endovascular coil placement for a ruptured fusiform aneurysm of the basilar artery: case report and review of the literature. J Neurosurg 1997; 87:944–9. Hoh BL, Putman CM, Budzik RF, Carter BS, Ogilvy CS. Combined surgical and endovascular techniques of flow alteration to treat fusiform and complex wide-necked intracranial aneurysms that are unsuitable for clipping or coil embolization. J Neurosurg 2001; 95:24–35.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 757

References

105 Boulin A, Pierot L. Follow-up of intracranial aneurysms treated with detachable coils: Comparison of gadoliniumenhanced 3D time-of-flight MR angiography and digital subtraction angiography. Radiology 2001; 219:108–13. 106 Majoie CB, Sprengers ME, Van Rooij WJ, Lavini C, Sluzewski M, van Rijn JC et al. MR angiography at 3T versus digital subtraction angiography in the follow-up of intracranial aneurysms treated with detachable coils. Am J Neuroradiol 2005; 26:1349–56. 107 Gauvrit JY, Leclerc X, Caron S, Taschner CA, Lejeune JP, Pruvo JP. Intracranial aneurysms treated with Guglielmi detachable coils: imaging follow-up with contrastenhanced MR angiography. Stroke 2006; 37:1033–7. 108 Pierot L, Delcour C, Bouquigny F, Breidt D, Feuillet B, Lanoix O et al. Follow-up of intracranial aneurysms selectively treated with coils: prospective evaluation of contrast-enhanced MR angiography. Am J Neuroradiol 2006; 27:744–9. 109 Slob MJ, Sluzewski M, Van Rooij WJ, Roks G, Rinkel GJE. Additional coiling of previously coiled cerebral aneurysms: clinical and angiographic results. Am J Neuroradiol 2004; 25:1373–6. 110 Zhang YJ, Barrow DL, Cawley CM, Dion JE. Neurosurgical management of intracranial aneurysms previously treated with endovascular therapy. Neurosurgery 2003; 52:283–93. 111 Raymond J, Roy D, Leblanc P, Roorda S, Janicki C, Normandeau L et al. Endovascular treatment of intracranial aneurysms with radioactive coils: initial clinical experience. Stroke 2003; 34:2801–6. 112 Gaba RC, Ansari SA, Roy SS, Marden FA, Viana MA, Malisch TW. Embolization of intracranial aneurysms with hydrogel-coated coils versus inert platinum coils: effects on packing density, coil length and quantity, procedure performance, cost, length of hospital stay, and durability of therapy. Stroke 2006; 37:1443–50. 113 McKissock W, Richardson A, Walsh L. ‘Posteriorcommunicating aneurysms’: a controlled trial of conservative and surgical treatment of ruptured aneurysms of the internal carotid artery at or near the point of origin of the posterior communicating artery. Lancet 1960; i:1203–6. 114 McKissock W, Richardson A, Walsh L. Anterior communicating aneurysms: a trial of conservative and surgical treatment. Lancet 1965; i:873–6. 115 Öhman J, Heiskanen O. Timing of operation for ruptured supratentorial aneurysms: a prospective randomized study. J Neurosurg 1989; 70(1):55–60. 116 Whitfield PC, Kirkpatrick PJ. Timing of surgery for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2001; CD001697. 117 De Gans K, Nieuwkamp DJ, Rinkel GJE, Algra A. Timing of aneurysm surgery in subarachnoid hemorrhage: A systematic review of the literature. Neurosurgery 2002; 50:336–40. 118 Ross N, Hutchinson PJ, Seeley H, Kirkpatrick PJ. Timing of surgery for supratentorial aneurysmal subarachnoid

..

..

119

120

121

122

123

124

125

126

127

128

129

130

131

132

haemorrhage: report of a prospective study. J Neurol Neurosurg Psychiatry 2002; 72:480–4. Nieuwkamp DJ, de GK, Algra A, Albrecht KW, Boomstra S, Brouwers PJ et al. Timing of aneurysm surgery in subarachnoid haemorrhage: an observational study in The Netherlands. Acta Neurochir (Wien) 2005; 147:815–21. Tanoue S, Kiyosue H, Kenai H, Nakamura T, Yamashita M, Mori H. Three-dimensional reconstructed images after rotational angiography in the evaluation of intracranial aneurysms: Surgical correlation. Neurosurgery 2000; 47:866–71. Todd MM, Hindman BJ, Clarke WR, Torner JC. Mild intraoperative hypothermia during surgery for intracranial aneurysm. N Engl J Med 2005; 352:135–45. Ross IB, Warrington RJ, Halliday WC. Cell-mediated allergy to a cerebral aneurysm clip: case report. Neurosurgery 1998; 43:1209–11. Wermer MJH, Van der Schaaf IC, Velthuis BK, Algra A, Buskens E, Rinkel GJE. Follow-up screening after subarachnoid haemorrhage: frequency and determinants of new aneurysms and enlargement of existing aneurysms. Brain 2005; 128:2421–9. Fodstad H, Pilbrant A, Schannong M, Stromberg S. Determination of tranexamic acid (AMCA) and fibrin/fibrinogen degradation products in cerebrospinal fluid after aneurysmal subarachnoid haemorrhage. Acta Neurochir (Wien) 1981; 58(1–2):1–13. Roos YBWEM, for the STAR Study Group. Antifibrinolytic treatment in subarachnoid hemorrhage: a randomized placebo-controlled trial. Neurology 2000; 54:77–82. Hillman J, Fridriksson S, Nilsson O, Yu Z, Saveland H, Jakobsson KE. Immediate administration of tranexamic acid and reduced incidence of early rebleeding after aneurysmal subarachnoid hemorrhage: a prospective randomized study. J Neurosurg 2002; 97:771–8. Pickard JD, Kirkpatrick PJ, Melsen T, Andreasen RB, Gelling L, Fryer T et al. Potential role of NovoSeven(R) in the prevention of rebleeding following aneurysmal subarachnoid haemorrhage. Blood Coagul Fibrinolysis 2000; 11:S117–S120. Hijdra A, van Gijn J, Stefanko S, van Dongen KJ, Vermeulen M, van Crevel H. Delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage: clinicoanatomic correlations. Neurology 1986; 36(3):329–33. Conway LW, McDonald LW. Structural changes of the intradural arteries following subarachnoid hemorrhage. J Neurosurg 1972; 37:715–23. Hughes JT, Schianchi PM. Cerebral artery spasm: a histological study at necropsy of the blood vessels in cases of subarachnoid hemorrhage. J Neurosurg 1978; 48:515–25. Smith RR, Clower BR, Peeler DF, Jr., Yoshioka J. The angiopathy of subarachnoid hemorrhage: angiographic and morphologic correlates. Stroke 1983; 14(2):240–5. Findlay JM, Weir BK, Kanamaru K, Espinosa F. Arterial wall changes in cerebral vasospasm. Neurosurgery 1989; 25(5):736–45.

757

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 758

758

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

133 Borel CO, McKee A, Parra A, Haglund MA, Solan A, Prabhakar V et al. Possible role for vascular cell proliferation in cerebral vasospasm after subarachnoid hemorrhage. Stroke 2003; 34:427–32. 134 Millikan CH. Cerebral vasospasm and ruptured intracranial aneurysm. Arch Neurol 1975; 32(7):433–49. 135 Fisher CM, Roberson GH, Ojemann RG. Cerebral vasospasm with ruptured saccular aneurysm: the clinical manifestations. Neurosurgery 1977; 1(3):245–8. 136 Lam JMK, Smielewski P, Czosnyka M, Pickard JD, Kirkpatrick PJ. Predicting delayed ischemic deficits after aneurysmal subarachnoid hemorrhage using a transient hyperemic response test of cerebral autoregulation. Neurosurgery 2000; 47:819–25. 137 Eskesen V, Karle A, Kruse A, Kruse Larsen C, Praestholm J, Schmidt K. Observer variability in assessment of angiographic vasospasm after aneurysmal subarachnoid haemorrhage. Acta Neurochir (Wien) 1987; 87(1–2):54–57. 138 Rabinstein AA, Friedman JA, Weigand SD, McClelland RL, Fulgham JR, Manno EM et al. Predictors of cerebral infarction in aneurysmal subarachnoid hemorrhage. Stroke 2004; 35:1862–6. 139 Kistler JP, Crowell RM, Davis KR, Heros R, Ojemann RG, Zervas T et al. The relation of cerebral vasospasm to the extent and location of subarachnoid blood visualized by CT scan: a prospective study. Neurology 1983; 33(4):424–36. 140 Svensson E, Starmark JE, Ekholm S, Von Essen C, Johansson A. Analysis of interobserver disagreement in the assessment of subarachnoid blood and acute hydrocephalus on CT scans. Neurol Res 1996; 18:487–94. 141 Van Norden AG, van Dijk GW, van Huizen MD, Algra A, Rinkel GJE. Interobserver agreement and predictive value for outcome of two rating scales for the amount of extravasated blood after aneurysmal subarachnoid haemorrhage. J Neurol 2006; 253:1217–20. 142 Brouwers PJAM, Wijdicks EFM, van Gijn J. Infarction after aneurysm rupture does not depend on distribution or clearance rate of blood. Stroke 1992; 23(3):374–9. 143 Rabinstein AA, Weigand S, Atkinson JL, Wijdicks EFM. Patterns of cerebral infarction in aneurysmal subarachnoid hemorrhage. Stroke 2005; 36:992–7. 144 Maeda K, Kurita H, Nakamura T, Usui M, Tsutsumi K, Morimoto T et al. Occurrence of severe vasospasm following intraventricular hemorrhage from an arteriovenous malformation: report of two cases. J Neurosurg 1997; 87:436–9. 145 Servadei F, Murray GD, Teasdale GM, Dearden M, Iannotti F, Lapierre F et al. Traumatic subarachnoid hemorrhage: demographic and clinical study of 750 patients from the European Brain Injury Consortium survey of head injuries. Neurosurgery 2002; 50:261–7. 146 Rinkel GJE, Wijdicks EFM, Vermeulen M, Hasan D, Brouwers PJAM, van Gijn J. The clinical course of perimesencephalic nonaneurysmal subarachnoid hemorrhage. Ann Neurol 1991; 29(5):463–8. 147 Van der Schaaf I, Wermer MJ, van der GY, Velthuis BK, van de Kraats CIB, Rinkel GJE. Prognostic value of cerebral perfusion-computed tomography in the acute stage after

148

149

150

151

152

153

154

155

156

157

158

159

160

161

subarachnoid hemorrhage for the development of delayed cerebral ischemia. Stroke 2006; 37:409–13. Hop JW, Rinkel GJE, Algra A, van Gijn J. Initial loss of consciousness and risk of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. Stroke 1999; 30:2268–71. Rinkel GJE, Feigin VL, Algra A, Van den Bergh WM, Vermeulen M, van Gijn J. Calcium antagonists for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2005; CD000277. Pickard JD, Murray GD, Illingworth R, Shaw MD, Teasdale GM, Foy PM et al. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. Br Med J 1989; 298(6674):636–42. Philippon J, Grob R, Dagreou F, Guggiari M, Rivierez M, Viars P. Prevention of vasospasm in subarachnoid haemorrhage: a controlled study with nimodipine. Acta Neurochir (Wien) 1986; 82(3–4):110–14. Petruk KC, West M, Mohr G, Weir BK, Benoit BG, Gentili F et al. Nimodipine treatment in poor-grade aneurysm patients: results of a multicenter double-blind placebocontrolled trial. J Neurosurg 1988; 68(4):505–17. Ahmed N, Nasman P, Wahlgren NG. Effect of intravenous nimodipine on blood pressure and outcome after acute stroke. Stroke 2000; 31:1250–5. Van den Bergh WM, Algra A, Van der Sprenkel JWB, Tulleken CAF, Rinkel GJE. Hypomagnesemia after aneurysmal subarachnoid hemorrhage. Neurosurgery 2003; 52:276–81. Van den Bergh WM, Zuur JK, Kamerling NA, Van Asseldonk JHT, Rinkel GJE, Tulleken CAF et al. Role of magnesium in the reduction of ischemic depolarization and lesion volume after experimental subarachnoid hemorrhage. J Neurosurg 2002; 97:416–22. Wong GK, Chan MT, Boet R, Poon WS, Gin T. Intravenous magnesium sulfate after aneurysmal subarachnoid hemorrhage: a prospective randomized pilot study. J Neurosurg Anesthesiol 2006; 18:142–8. Van den Bergh WM, on behalf of the MASH study group. Magnesium sulfate in aneurysmal subarachnoid hemorrhage: a randomized controlled trial. Stroke 2005; 36:1011–15. Schmid-Elsässer R, Kunz M, Zausinger S, Prueckner S, Briegel J, Steiger HJ. Intravenous magnesium versus nimodipine in the treatment of patients with aneurysmal subarachnoid hemorrhage: a randomized study. Neurosurgery 2006; 58:1054–65. Juvela S, Kaste M, Hillbom M. Platelet thromboxane release after subarachnoid hemorrhage and surgery. Stroke 1990; 21(4):566–71. Ohkuma H, Suzuki S, Kimura M, Sobata E. Role of platelet function in symptomatic cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke 1991; 22:854–9. Juvela S. Aspirin and delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. J Neurosurg 1995; 82:945–52.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 759

References

162 Mendelow AD, Stockdill G, Steers AJ, Hayes J, Gillingham FJ. Double-blind trial of aspirin in patient receiving tranexamic acid for subarachnoid hemorrhage. Acta Neurochir (Wien) 1982; 62(3–4):195–202. 163 Hop JW, Rinkel GJE, Algra A, Berkelbach van der Sprenkel JW, van Gijn J. Randomized pilot trial of postoperative aspirin in subarachnoid hemorrhage. Neurology 2000; 54:872–8. 164 Shaw MD, Foy PM, Conway M, Pickard JD, Maloney P, Spillane JA et al. Dipyridamole and postoperative ischemic deficits in aneurysmal subarachnoid hemorrhage. J Neurosurg 1985; 63:699–703. 165 Tokiyoshi K, Ohnishi T, Nii Y. Efficacy and toxicity of thromboxane synthetase inhibitor for cerebral vasospasm after subarachnoid hemorrhage. Surg Neurol 1991; 36:112–18. 166 Suzuki S, Sano K, Handa H, Asano T, Tamura A, Yonekawa Y et al. Clinical study of OKY-046, a thromboxane synthetase inhibitor, in prevention of cerebral vasospasms and delayed cerebral ischaemic symptoms after subarachnoid haemorrhage due to aneurysmal rupture: a randomized double-blind study. Neurol Res 1989; 11:79–88. 167 Robinson JG, Smith B, Maheshwari N, Schrott H. Pleiotropic effects of statins: benefit beyond cholesterol reduction? A meta-regression analysis. J Am Coll Cardiol 2005; 46:1855–62. 168 Lynch JR, Wang H, McGirt MJ, Floyd J, Friedman AH, Coon AL et al. Simvastatin reduces vasospasm after aneurysmal subarachnoid hemorrhage: results of a pilot randomized clinical trial. Stroke 2005; 36:2024–6. 169 Tseng MY, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Effects of acute treatment with pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized placebo-controlled trial. Stroke 2005; 36:1627–32. 170 Singhal AB, Topcuoglu MA, Dorer DJ, Ogilvy CS, Carter BS, Koroshetz WJ. SSRI and statin use increases the risk for vasospasm after subarachnoid hemorrhage. Neurology 2005; 64:1008–13. 171 Kassell NF, Haley EC, Jr., Apperson-Hansen C, Stat M, Alves WM, Dorsch NW et al. Randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in Europe, Australia, and New Zealand. J Neurosurg 1996; 84:221–8. 172 Haley EC, Jr., Kassell NF, Apperson-Hansen C, Maile MH, Alves WM. A randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in North America. J Neurosurg 1997; 86(3):467–74. 173 Lanzino G, Kassell NF, Dorsch NWC, Pasqualin A, Brandt L, Schmiedek P et al. Double-blind, randomized, vehiclecontrolled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part I. A cooperative study in Europe, Australia, New Zealand, and South Africa. J Neurosurg 1999; 90:1011–17.

..

..

174 Lanzino G, Kassell NF. Double-blind, randomized, vehiclecontrolled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part II. A cooperative study in North America. J Neurosurg 1999; 90:1018–24. 175 Ogden JA, Mee EW, Utley T. Too little, too late: Does tirilazad mesylate reduce fatigue after subarachnoid hemorrhage? Neurosurgery 1998; 43:782–7. 176 Asano T, Takakura K, Sano K, Kikuchi H, Nagai H, Saito I et al. Effects of a hydroxyl radical scavenger on delayed ischemic neurological deficits following aneurysmal subarachnoid hemorrhage: results of a multicenter, placebo-controlled double-blind trial. J Neurosurg 1996; 84:792–803. 177 Saito I, Asano T, Sano K, Takakura K, Abe H, Yoshimoto T et al. Neuroprotective effect of an antioxidant, ebselen, in patients with delayed neurological deficits after aneurysmal subarachnoid hemorrhage. Neurosurgery 1998; 42:269–77. 178 Saito I, Asano T, Ochiai C, Takakura K, Tamura A, Sano K. A double-blind clinical evaluation of the effect of Nizofenone (Y-9179) on delayed ischemic neurological deficits following aneurysmal rupture. Neurol Res 1983; 5:29–47. 179 Shaw MDM, Vermeulen M, Murray GD, Pickard JD, Bell BA, Teasdale GM. Efficacy and safety of the endothelinA/B receptor antagonist TAK-044 in treating subarachnoid hemorrhage: a report by the Steering Committee on behalf of the UK/Netherlands/Eire TAK-044 Subarachnoid Haemorrhage Study Group. J Neurosurg 2000; 93:992–7. 180 Rinkel GJE, Feigin VL, Algra A, van Gijn J. Circulatory volume expansion therapy for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2004; CD000483. 181 Wijdicks EFM, Vermeulen M, van Brummelen P, van Gijn J. The effect of fludrocortisone acetate on plasma volume and natriuresis in patients with aneurysmal subarachnoid hemorrhage. Clin Neurol Neurosurg 1988; 90(3):209–14. 182 Hasan D, Lindsay KW, Wijdicks EFM, Murray GD, Brouwers PJAM, Bakker WH et al. Effect of fludrocortisone acetate in patients with subarachnoid hemorrhage. Stroke 1989; 20(9):1156–61. 183 Mori T, Katayama Y, Kawamata T, Hirayama T. Improved efficiency of hypervolemic therapy with inhibition of natriuresis by fludrocortisone in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 1999; 91:947–52. 184 Moro N, Katayama Y, Kojima J, Mori T, Kawamata T. Prophylactic management of excessive natriuresis with hydrocortisone for efficient hypervolemic therapy after subarachnoid hemorrhage. Stroke 2003; 34:2807–11. 185 Klimo P, Jr., Kestle JR, MacDonald JD, Schmidt RH. Marked reduction of cerebral vasospasm with lumbar drainage of cerebrospinal fluid after subarachnoid hemorrhage. J Neurosurg 2004; 100:215–24. 186 Amin-Hanjani S, Ogilvy CS, Barker FG. Does intracisternal thrombolysis prevent vasospasm after aneurysmal

759

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 760

760

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

187

188

189

190

191

192

193

194

195

196

197

198

199

subarachnoid hemorrhage? A meta-analysis. Neurosurgery 2004; 54:326–34. Hamada J, Kai Y, Morioka M, Yano S, Mizuno T, Hirano T et al. Effect on cerebral vasospasm of coil embolization followed by microcatheter intrathecal urokinase infusion into the cisterna magna: a prospective randomized study. Stroke 2003; 34:2549–54. Hijdra A, Vermeulen M, van Gijn J, van Crevel H. Rerupture of intracranial aneurysms: a clinicoanatomic study. J Neurosurg 1987; 67(1):29–33. Hijdra A, Braakman R, van Gijn J, Vermeulen M, van Crevel H. Aneurysmal subarachnoid hemorrhage: complications and outcome in a hospital population. Stroke 1987; 18(6):1061–7. Van der Schaaf IC, Ruigrok YM, Rinkel GJE, Algra A, van Gijn J. Study design and outcome measures in studies on aneurysmal subarachnoid hemorrhage. Stroke 2002; 33:2043–6. Phan TG, Huston J, III, Campeau NG, Wijdicks EFM, Atkinson JL, Fulgham JR. Value of diffusion-weighted imaging in patients with a nonlocalizing examination and vasospasm from subarachnoid hemorrhage. Cerebrovasc Dis 2003; 15:177–81. Davis S, Andrews J, Lichtenstein M, Kaye A, Tress B, Rossiter S et al. A single-photon emission computed tomography study of hypoperfusion after subarachnoid hemorrhage. Stroke 1990; 21(2):252–9. Hasan D, van Peski J, Loeve I, Krenning EP, Vermeulen M. Single photon emission computed tomography in patients with acute hydrocephalus or with cerebral ischaemia after subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1991; 54(6):490–3. Lysakowski C, Walder B, Costanza MC, Tramér MR. Transcranial Doppler versus angiography in patients with vasospasm due to a ruptured cerebral aneurysm: a systematic review. Stroke 2001; 32:2292–8. Otawara Y, Ogasawara K, Ogawa A, Sasaki M, Takahashi K. Evaluation of vasospasm after subarachnoid hemorrhage by use of multislice computed tomographic angiography. Neurosurgery 2002; 51:939–42. Yoon DY, Choi CS, Kim KH, Cho BM. Multidetector-row CT angiography of cerebral vasospasm after aneurysmal subarachnoid hemorrhage: comparison of volumerendered images and digital subtraction angiography. Am J Neuroradiol 2006; 27:370–7. Kosnik EJ, Hunt WE. Postoperative hypertension in the management of patients with intracranial arterial aneurysms. J Neurosurg 1976; 45(2):148–54. Kassell NF, Peerless SJ, Durward QJ, Beck DW, Drake CG, Adams HP, Jr. Treatment of ischemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 1982; 11(3):337–43. Qureshi AI, Suarez JI, Bhardwaj A, Yahia AM, Tamargo RJ, Ulatowski JA. Early predictors of outcome in patients receiving hypervolemic and hypertensive therapy for symptomatic vasospasm after subarachnoid hemorrhage. Crit Care Med 2000; 28:824–9.

200 Sen J, Belli A, Albon H, Morgan L, Petzold A, Kitchen N. Triple-H therapy in the management of aneurysmal subarachnoid haemorrhage. Lancet Neurol 2003; 2:614–21. 201 Hoh BL, Carter BS, Ogilvy CS. Risk of hemorrhage from unsecured, unruptured aneurysms during and after hypertensive hypervolemic therapy. Neurosurgery 2002; 50:1207–11. 202 Amin-Hanjani S, Schwartz RB, Sathi S, Stieg PE. Hypertensive encephalopathy as a complication of hyperdynamic therapy for vasospasm: report of two cases. Neurosurgery 1999; 44:1113–16. 203 Wartenberg KE, Parra A. CT and CT-perfusion findings of reversible leukoencephalopathy during triple-H therapy for symptomatic subarachnoid hemorrhage-related vasospasm. J Neuroimaging 2006; 16:170–5. 204 Mayer SA, Solomon RA, Fink ME, Lennihan L, Stern L, Beckford A et al. Effect of 5% albumin solution on sodium balance and blood volume after subarachnoid hemorrhage. Neurosurgery 1998; 42:759–67. 205 Hoh BL, Ogilvy CS. Endovascular treatment of cerebral vasospasm: transluminal balloon angioplasty, intra-arterial papaverine, and intra-arterial nicardipine. Neurosurg Clin N Am 2005; 16:501–16, vi. 206 Rabinstein AA, Friedman JA, Nichols DA, Pichelmann MA, McClelland RL, Manno EM et al. Predictors of outcome after endovascular treatment of cerebral vasospasm. Am J Neuroradiol 2004; 25:1778–82. 207 Polin RS, Coenen VA, Hansen CA, Shin P, Baskaya MK, Nanda A et al. Efficacy of transluminal angioplasty for the management of symptomatic cerebral vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg 2000; 92:284–90. 208 Fandino J, Kaku Y, Schuknecht B, Valavanis A, Yonekawa Y. Improvement of cerebral oxygenation patterns and metabolic validation of super-selective intraarterial infusion of papaverine for the treatment of cerebral vasospasm. J Neurosurg 1998; 89:93–100. 209 Polin RS, Hansen CA, German P, Chadduck JB, Kassell NF. Intra-arterially administered papaverine for the treatment of symptomatic cerebral vasospasm. Neurosurgery 1998; 42:1256–64. 210 Smith WS, Dowd CF, Johnston SC, Ko NU, DeArmond SJ, Dillon WP et al. Neurotoxicity of intra-arterial papaverine preserved with chlorobutanol used for the treatment of cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke 2004; 35:2518–22. 211 Arakawa Y, Kikuta K, Hojo M, Goto Y, Ishii A, Yamagata S. Milrinone for the treatment of cerebral vasospasm after subarachnoid hemorrhage: report of seven cases. Neurosurgery 2001; 48:723–8. 212 Feng L, Fitzsimmons BF, Young WL, Berman MF, Lin E, Aagaard BD et al. Intraarterially administered verapamil as adjunct therapy for cerebral vasospasm: safety and 2-year experience. Am J Neuroradiol 2002; 23:1284–10. 213 Badjatia N, Topcuoglu MA, Pryor JC, Rabinov JD, Ogilvy CS, Carter BS et al. Preliminary experience with intraarterial nicardipine as a treatment for cerebral vasospasm. Am J Neuroradiol 2004; 25:819–26.

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 761

References

214 European CGRP in Subarachnoid Haemorrhage Study Group. Effect of calcitonin-gene-related peptide in patients with delayed postoperative cerebral ischaemia after aneurysmal subarachnoid haemorrhage. Lancet 1992; 339(8797):831–4. 215 Hasan D, Vermeulen M, Wijdicks EFM, Hijdra A, van Gijn J. Management problems in acute hydrocephalus after subarachnoid hemorrhage. Stroke 1989; 20(6):747–53. 216 Milhorat TH. Acute hydrocephalus after aneurysmal subarachnoid hemorrhage. Neurosurgery 1987; 20(1):15–20. 217 Hasan D, Tanghe HL. Distribution of cisternal blood in patients with acute hydrocephalus after subarachnoid hemorrhage. Ann Neurol 1992; 31(4):374–8. 218 Dorai Z, Hynan LS, Kopitnik TA, Samson D. Factors related to hydrocephalus after aneurysmal subarachnoid hemorrhage. Neurosurgery 2003; 52:763–9. 219 Rinkel GJE, Wijdicks EFM, Vermeulen M, Tans JTJ, Hasan D, van Gijn J. Acute hydrocephalus in nonaneurysmal perimesencephalic hemorrhage: evidence of CSF block at the tentorial hiatus. Neurology 1992; 42(9):1805–7. 220 Komotar RJ, Olivi A, Rigamonti D, Tamargo RJ. Microsurgical fenestration of the lamina terminalis reduces the incidence of shuntdependent hydrocephalus after aneurysmal subarachnoid hemorrhage. Neurosurgery 2002; 51:1403–12. 221 Andaluz N, Zuccarello M. Fenestration of the lamina terminalis as a valuable adjunct in aneurysm surgery. Neurosurgery 2004; 55:1050–9. 222 Swash M. Periaqueductal dysfunction (the sylvian aqueduct syndrome): a sign of hydrocephalus? J Neurol Neurosurg Psychiatry 1974; 37(1):21–6. 223 Kolluri VR, Sengupta RP. Symptomatic hydrocephalus following aneurysmal subarachnoid hemorrhage. Surg Neurol 1984; 21(4):402–4. 224 Hellingman CA, Van den Bergh WM, Beijer I, van Dijk GW, Algra A, van Gijn J et al. Risk of rebleeding after treatment of acute hydrocephalus in patients with aneurysmal subarachnoid hemorrhage. Stroke 2007; 38(1):96–9. 225 Pickard JD. Early posthaemorrhagic hydrocephalus. Br Med J 1984; 289(6445):569–70. 226 Wardlaw JM, Cannon J, Statham PFX, Price R. Does the size of intracranial aneurysms change with intracranial pressure? Observations based on color ‘power’ transcranial Doppler ultrasound. J Neurosurg 1998; 88:846–50. 227 Fountas KN, Kapsalaki EZ, Machinis T, Karampelas I, Smisson HF, Robinson JS. Review of the literature regarding the relationship of rebleeding and external ventricular drainage in patients with subarachnoid hemorrhage of aneurysmal origin. Neurosurg Rev 2005; 29:14–18. 228 Wong GKC, Poon WS, Wai S, Yu LM, Lyon D, Lam JMK. Failure of regular external ventricular drain exchange to reduce cerebrospinal fluid infection: result of a randomised controlled trial. J Neurol Neurosurg Psychiatry 2002; 73:759–61.

..

..

229 Lozier AP, Sciacca RR, Romagnoli MF, Connolly ES, Jr. Ventriculostomy-related infections: a critical review of the literature. Neurosurgery 2002; 51:170–81. 230 Choksey MS, Malik IA. Zero tolerance to shunt infections: can it be achieved? J Neurol Neurosurg Psychiatry 2004; 75:87–91. 231 Korinek AM, Reina M, Boch AL, Rivera AO, De BD, Puybasset L. Prevention of external ventricular drainrelated ventriculitis. Acta Neurochir (Wien) 2005; 147:39–45. 232 Khanna RK, Rosenblum ML, Rock JP, Malik GM. Prolonged external ventricular drainage with percutaneous long-tunnel ventriculostomies. J Neurosurg 1995; 83:791–4. 233 Kim DK, Uttley D, Bell BA, Marsh HT, Moore AJ. Comparison of rates of infection of two methods of emergency ventricular drainage. J Neurol Neurosurg Psychiatry 1995; 58:444–6. 234 Coplin WM, Avellino AM, Kim DK, Winn HR, Grady MS. Bacterial meningitis associated with lumbar drains: a retrospective cohort study. J Neurol Neurosurg Psychiatry 1999; 67:468–73. 235 Schade RP, Schinkel J, Visser LG, van Dijk JM, Voormolen JH, Kuijper EJ. Bacterial meningitis caused by the use of ventricular or lumbar cerebrospinal fluid catheters. J Neurosurg 2005; 102:229–34. 236 Ruijs AC, Dirven CM, Algra A, Beijer I, Vandertop WP, Rinkel G. The risk of rebleeding after external lumbar drainage in patients with untreated ruptured cerebral aneurysms. Acta Neurochir (Wien) 2005; 147:1157–62. 237 Klopfenstein JD, Kim LJ, Feiz-Erfan I, Hott JS, Goslar P, Zabramski JM et al. Comparison of rapid and gradual weaning from external ventricular drainage in patients with aneurysmal subarachnoid hemorrhage: a prospective randomized trial. J Neurosurg 2004; 100:225–9. 238 Solenski NJ, Haley EC, Jr., Kassell NF, Kongable G, Germanson T, Truskowski L et al. Medical complications of aneurysmal subarachnoid hemorrhage: a report of the multicenter, cooperative aneurysm study. Crit Care Med 1995; 23:1007–17. 239 Gruber A, Reinprecht A, Görzer H, Fridrich P, Czech T, Illievich UM et al. Pulmonary function and radiographic abnormalities related to neurological outcome after aneurysmal subarachnoid hemorrhage. J Neurosurg 1998; 88:28–37. 240 Wartenberg KE, Schmidt JM, Claassen J, Temes RE, Frontera JA, Ostapkovich N et al. Impact of medical complications on outcome after subarachnoid hemorrhage. Crit Care Med 2006; 34:617–23. 241 Schuiling WJ, de Weerd AW, Dennesen PJ, Algra A, Rinkel GJE. The simplified acute physiology score to predict outcome in patients with subarachnoid hemorrhage. Neurosurgery 2005; 57:230–6. 242 Qureshi AI, Suri MFK, Sung GY, Straw RN, Yahia AM, Saad M et al. Prognostic significance of hypernatremia and hyponatremia among patients with aneurysmal subarachnoid hemorrhage. Neurosurgery 2002; 50:749–55.

761

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 762

762

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

243 Bartter FC, Schwartz WB. The syndrome of inappropriate secretion of antdiuretic hormone. Am J Med 1967; 42:790–806. 244 Harrigan MR. Cerebral salt wasting syndrome: a review. Neurosurgery 1996; 38:152–60. 245 Wijdicks EFM, van Dongen KJ, van Gijn J, Hijdra A, Vermeulen M. Enlargement of the third ventricle and hyponatraemia in aneurysmal subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1988; 51(4):516–20. 246 Sayama T, Inamura T, Matsushima T, Inoha S, Inoue T, Fukui M. High incidence of hyponatremia in patients with ruptured anterior communicating artery aneurysms. Neurol Res 2000; 22:151–5. 247 Wijdicks EFM, Vermeulen M, van Brummelen P, den Boer NC, van Gijn J. Digoxin-like immunoreactive substance in patients with aneurysmal subarachnoid haemorrhage. Br Med J 1987; 294(6574):729–32. 248 Diringer M, Ladenson PW, Stern BJ, Schleimer J, Hanley DF. Plasma atrial natriuretic factor and subarachnoid hemorrhage. Stroke 1988; 19(9):1119–24. 249 Rosenfeld JV, Barnett GH, Sila CA, Little JR, Bravo EL, Beck GJ. The effect of subarachnoid hemorrhage on blood and CSF atrial natriuretic factor. J Neurosurg 1989; 71(1):32–7. 250 Wijdicks EFM, Ropper AH, Hunnicutt EJ, Richardson GS, Nathanson JA. Atrial natriuretic factor and salt wasting after aneurysmal subarachnoid hemorrhage. Stroke 1991; 22(12):1519–24. 251 Isotani E, Suzuki R, Tomita K, Hokari M, Monma S, Marumo F et al. Alterations in plasma concentrations of natriuretic peptides and antidiuretic hormone after subarachnoid hemorrhage. Stroke 1994; 25:2198–203. 252 Wijdicks EFM, Schievink WI, Burnett JC, Jr. Natriuretic peptide system and endothelin in aneurysmal subarachnoid hemorrhage. J Neurosurg 1997; 87:275–80. 253 Berendes E, Walter M, Cullen P, Prien T, Van Aken H, Horsthemke J et al. Secretion of brain natriuretic peptide in patients with aneurysmal subarachnoid haemorrhage. Lancet 1997; 349:245–9. 254 Tomida M, Muraki M, Uemura K, Yamasaki K. Plasma concentrations of brain natriuretic peptide in patients with subarachnoid hemorrhage. Stroke 1998; 29:1584–7. 255 McGirt MJ, Blessing R, Nimjee SM, Friedman AH, Alexander MJ, Laskowitz DT et al. Correlation of serum brain natriuretic peptide with hyponatremia and delayed ischemic neurological deficits after subarachnoid hemorrhage. Neurosurgery 2004; 54:1369–73. 256 Khurana VG, Wijdicks EFM, Heublein DM, McClelland RL, Meyer FB, Piepgras DG et al. A pilot study of dendroaspis natriuretic peptide in aneurysmal subarachnoid hemorrhage. Neurosurgery 2004; 55:69–75. 257 Martin RJ. Central pontine and extrapontine myelinolysis: the osmotic demyelination syndromes. J Neurol Neurosurg Psychiatry 2004; 75 (Suppl 3):iii22–iii28. 258 Adrogué HJ, Madias NE. Hyponatremia. N Engl J Med 2000; 342:1581–9. 259 Adrogué HJ. Consequences of inadequate management of hyponatremia. Am J Nephrol 2005; 25:240–9.

260 Weisinger JR, Bellorin-Font E. Magnesium and phosphorus. Lancet 1998; 352:391–396. 261 van Norden AG, Van den Bergh WM, Rinkel GJE. Dose evaluation for long-term magnesium treatment in aneurysmal subarachnoid haemorrhage. J Clin Pharm Ther 2005; 30:439–42. 262 Fukui S, Katoh H, Tsuzuki N, Ishihara S, Otani N, Ooigawa H et al. Multivariate analysis of risk factors for QT prolongation following subarachnoid hemorrhage. Crit Care 2003; 7:R7–R12. 263 Disney L, Weir B, Grace M, Roberts P. Trends in blood pressure, osmolality and electrolytes after subarachnoid hemorrhage from aneurysms. Can J Neurol Sci 1989; 16:299–304. 264 Dorhout Mees SM, van Dijk GW, Algra A, Kempink DR, Rinkel GJE. Glucose levels and outcome after subarachnoid hemorrhage. Neurology 2003; 61:1132–3. 265 Frontera JA, Fernandez A, Claassen J, Schmidt M, Schumacher HC, Wartenberg K et al. Hyperglycemia after SAH: predictors, associated complications, and impact on outcome. Stroke 2006; 37:199–203. 266 Naredi S, Lambert G, Edén E, Zäll S, Runnerstam M, Rydenhag B et al. Increased sympathetic nervous activity in patients with nontraumatic subarachnoid hemorrhage. Stroke 2000; 31:901–6. 267 Espiner EA, Leikis R, Ferch RD, MacFarlane MR, Bonkowski JA, Frampton CM et al. The neuro-cardioendocrine response to acute subarachnoid haemorrhage. Clin Endocrinol (Oxf) 2002; 56:629–35. 268 Mayer SA, Lin J, Homma S, Solomon RA, Lennihan L, Sherman D et al. Myocardial injury and left ventricular performance after subarachnoid hemorrhage. Stroke 1999; 30:780–6. 269 Parekh N, Venkatesh B, Cross D, Leditschke A, Atherton J, Miles W et al. Cardiac troponin I predicts myocardial dysfunction in aneurysmal subarachnoid hemorrhage. J Am Coll Cardiol 2000; 36:1328–35. 270 Tung P, Kopelnik A, Banki N, Ong K, Ko N, Lawton MT et al. Predictors of neurocardiogenic injury after subarachnoid hemorrhage. Stroke 2004; 35:548–51. 271 Schuiling WJ, Dennesen PJ, Tans JT, Kingma LM, Algra A, Rinkel GJE. Troponin I in predicting cardiac or pulmonary complications and outcome in subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2005; 76:1565–9. 272 Naidech AM, Kreiter KT, Janjua N, Ostapkovich ND, Parra A, Commichau C et al. Cardiac troponin elevation, cardiovascular morbidity, and outcome after subarachnoid hemorrhage. Circulation 2005; 112:2851–6. 273 Greenhoot JH, Reichenbach DD. Cardiac injury and subarachnoid hemorrhage: a clinical, pathological and physiological correlation. J Neurosurg 1969; 30:521–31. 274 Tung PP, Olmsted E, Kopelnik A, Banki NM, Drew BJ, Ko N et al. Plasma B-type natriuretic peptide levels are associated with early cardiac dysfunction after subarachnoid hemorrhage. Stroke 2005; 36:1567–9. 275 Van den Bergh WM, Algra A, Rinkel GJE. Electrocardiographic abnormalities and serum

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 763

References

276

277

278

279

280

281

282

283

284 285

286

287

288

289

290

..

..

magnesium in patients with subarachnoid hemorrhage. Stroke 2004; 35:644–8. Khechinashvili G, Asplund K. Electrocardiographic changes in patients with acute stroke: a systematic review. Cerebrovasc Dis 2002; 14:67–76. Carruth JE, Silverman ME. Torsade de pointe atypical ventricular tachycardia complicating subarachnoid hemorrhage. Chest 1980; 78(6):886–8. Hess EP, Boie ET, White RD. Survival of a neurologically intact patient with subarachnoid hemorrhage and cardiopulmonary arrest. Mayo Clin Proc 2005; 80:1073–6. Andreoli A, di Pasquale G, Pinelli G, Grazi P, Tognetti F, Testa C. Subarachnoid hemorrhage: frequency and severity of cardiac arrhythmias: a survey of 70 cases studied in the acute phase. Stroke 1987; 18(3):558–64. Brouwers PJAM, Wijdicks EFM, Hasan D, Vermeulen M, Wever EF, Frericks H et al. Serial electrocardiographic recording in aneurysmal subarachnoid hemorrhage. Stroke 1989; 20(9):1162–7. Manninen PH, Ayra B, Gelb AW, Pelz D. Association between electrocardiographic abnormalities and intracranial blood in patients following acute subarachnoid hemorrhage. J Neurosurg Anesthesiol 1995; 7(1):12–16. Svigelj V, Grad A, Tekavcic I, Kiauta T. Cardiac arrhythmia associated with reversible damage to insula in a patient with subarachnoid hemorrhage. Stroke 1994; 25(5):1053–5. Hirashima Y, Takashima S, Matsumura N, Kurimoto M, Origasa H, Endo S. Right sylvian fissure subarachnoid hemorrhage has electrocardiographic consequences. Stroke 2001; 32:2278–81. Cheung RT, Hachinski V. The insula and cerebrogenic sudden death. Arch Neurol 2000; 57:1685–8. Cruickshank JM, Neil-Dwyer G, Degaute JP, Hayes Y, Kuurne T, Kytta J et al. Reduction of stress/catecholamineinduced cardiac necrosis by beta 1-selective blockade. Lancet 1987; 2(8559):585–9. Banki N, Kopelnik A, Tung P, Lawton MT, Gress D, Drew B et al. Prospective analysis of prevalence, distribution, and rate of recovery of left ventricular systolic dysfunction in patients with subarachnoid hemorrhage. J Neurosurg 2006; 105:15–20. Schuiling WJ, Dennesen PJ, Rinkel GJE. Extracerebral organ dysfunction in the acute stage after aneurysmal subarachnoid hemorrhage. Neurocrit Care 2005; 3:1–10. Vespa PM, Bleck TP. Neurogenic pulmonary edema and other mechanisms of impaired oxygenation after aneurysmal subarachnoid hemorrhage. Neurocrit Care 2004; 1:157–70. Friedman JA, Pichelmann MA, Piepgras DG, McIver JI, Toussaint LG, III, McClelland RL et al. Pulmonary complications of aneurysmal subarachnoid hemorrhage. Neurosurgery 2003; 52:1025–31. Mayer SA, Fink ME, Homma S, Sherman D, LiMandri G, Lennihan L et al. Cardiac injury associated with neurogenic pulmonary edema following subarachnoid hemorrhage. Neurology 1994; 44(5):815–20.

291 Deehan SC, Grant IS. Haemodynamic changes in neurogenic pulmonary oedema: effect of dobutamine. Intensive Care Med 1996; 22:672–6. 292 Hop JW, Rinkel GJE, Algra A, van Gijn J. Changes in functional outcome and quality of life in patients and caregivers after aneurysmal subarachnoid hemorrhage. J Neurosurg 2001; 95:957–63. 293 McCarron MO, Alberts MJ, McCarron P. A systematic review of Terson’s syndrome: frequency and prognosis after subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2004; 75:491–3. 294 Frizzell RT, Kuhn F, Morris R, Quinn C, Fisher WS, III. Screening for ocular hemorrhages in patients with ruptured cerebral aneurysms: a prospective study of 99 patients. Neurosurgery 1997; 41:529–33. 295 Kuhn F, Morris R, Witherspoon CD, Mester V. Terson syndrome: results of vitrectomy and the significance of vitreous hemorrhage in patients with subarachnoid hemorrhage. Ophthalmology 1998; 105:472–7. 296 Stiebel-Kalish H, Turtel LS, Kupersmith MJ. The natural history of nontraumatic subarachnoid hemorrhage-related intraocular hemorrhages. Retina 2004; 24:36–40. 297 Winn HR, Richardson AE, Jane JA. The long-term prognosis in untreated cerebral aneurysms: I. The incidence of late hemorrhage in cerebral aneurysm: a 10-year evaluation of 364 patients. Ann Neurol 1977; 1(4):358–70. 298 Wermer MJH, Rinkel GJE, Greebe P, Albrecht KW, Dirven CM, Tulleken CAF. Late recurrence of subarachnoid hemorrhage after treatment for ruptured aneurysms: patient characteristics and outcomes. Neurosurgery 2005; 56:197–204. 299 Vale FL, Bradley EL, Fisher WS, III. The relationship of subarachnoid hemorrhage and the need for postoperative shunting. J Neurosurg 1997; 86(3):462–6. 300 Gruber A, Reinprecht A, Bavinzski G, Czech T, Richling B. Chronic shunt-dependent hydrocephalus after early surgical and early endovascular treatment of ruptured intracranial aneurysms. Neurosurgery 1999; 44:503–9. 301 Tapaninaho A, Hernesniemi J, Vapalahti M, Niskanen M, Kari A, Luukkonen M et al. Shunt-dependent hydrocephalus after subarachnoid haemorrhage and aneurysm surgery: timing of surgery is not a risk factor. Acta Neurochir (Wien) 1993; 123(3–4):118–24. 302 Pietila TA, Heimberger KC, Palleske H, Brock M. Influence of aneurysm location on the development of chronic hydrocephalus following SAH. Acta Neurochir (Wien) 1995; 137:70–3. 303 Yoshioka H, Inagawa T, Tokuda Y, Inokuchi F. Chronic hydrocephalus in elderly patients following subarachnoid hemorrhage. Surg Neurol 2000; 53:119–24. 304 Hirashima Y, Hamada H, Hayashi N, Kuwayama N, Origasa H, Endo S. Independent predictors of late hydrocephalus in patients with aneurysmal subarachnoid hemorrhage: analysis by multivariate logistic regression model. Cerebrovasc Dis 2003; 16:205–10. 305 Bavinzski G, Killer M, Gruber A, Reinprecht A, Gross CE, Richling B. Treatment of basilar artery bifurcation

763

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 764

764

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

306

307

308

309

310

311

312

313

314

315

316

317

318

319

aneurysms by using Guglielmi detachable coils: a 6-year experience. J Neurosurg 1999; 90:843–52. Sethi H, Moore A, Dervin J, Clifton A, MacSweeney JE. Hydrocephalus: comparison of clipping and embolization in aneurysm treatment. J Neurosurg 2000; 92:991–4. Dehdashti AR, Rilliet B, Rufenacht DA, de Tribolet N. Shunt-dependent hydrocephalus after rupture of intracranial aneurysms: a prospective study of the influence of treatment modality. J Neurosurg 2004; 101:402–7. Tomasello F, D’Avella D, De Divitiis O. Does lamina terminalis fenestration reduce the incidence of chronic hydrocephalus after subarachnoid hemorrhage? Neurosurgery 1999; 45:827–31. Claassen J, Peery S, Kreiter KT, Hirsch LJ, Du EY, Connolly ES et al. Predictors and clinical impact of epilepsy after subarachnoid hemorrhage. Neurology 2003; 60:208–14. Buczacki SJ, Kirkpatrick PJ, Seeley HM, Hutchinson PJ. Late epilepsy following open surgery for aneurysmal subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2004; 75:1620–2. Rhoney DH, Tipps LB, Murry KR, Basham MC, Michael DB, Coplin WM. Anticonvulsant prophylaxis and timing of seizures after aneurysmal subarachnoid hemorrhage. Neurology 2000; 55:258–65. Olafsson E, Gudmundsson G, Hauser WA. Risk of epilepsy in long-term survivors of surgery for aneurysmal subarachnoid hemorrhage: a population-based study in Iceland. Epilepsia 2000; 41:1201–5. Byrne JV, Boardman P, Ioannidis I, Adcock J, Traill Z. Seizures after aneurysmal subarachnoid hemorrhage treated with coil embolization. Neurosurgery 2003; 52:545–50. Buchanan KM, Elias LJ, Goplen GB. Differing perspectives on outcome after subarachnoid hemorrhage: the patient, the relative, the neurosurgeon. Neurosurgery 2000; 46:831–8. Mayer SA, Kreiter KT, Copeland D, Bernardini GL, Bates JE, Peery S et al. Global and domain-specific cognitive impairment and outcome after subarachnoid hemorrhage. Neurology 2002; 59:1750–8. Powell J, Kitchen N, Heslin J, Greenwood R. Psychosocial outcomes at three and nine months after good neurological recovery from aneurysmal subarachnoid haemorrhage: predictors and prognosis. J Neurol Neurosurg Psychiatry 2002; 72:772–81. Powell J, Kitchen N, Heslin J, Greenwood R. Psychosocial outcomes at 18 months after good neurological recovery from aneurysmal subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2004; 75:1119–24. Wermer MJ, Kool H, Albrecht KW, Rinkel GJE, for the ASTRA study group. Long-term effects of subarachnoid hemorrhage on employment, relationships, personality and mood. Neurosurgery 2007; 60(1):91–7, discussion 97–8. Anson JA, Lawton MT, Spetzler RF. Characteristics and surgical treatment of dolichoectatic and fusiform aneurysms. J Neurosurg 1996; 84:185–93.

320 Drake CG, Peerless SJ. Giant fusiform intracranial aneurysms: review of 120 patients treated surgically from 1965 to 1992. J Neurosurg 1997; 87:141–62. 321 Lewis SB, Chang DWJ, Peace DA, LaFrentz PJ, Day AL. Distal posterior inferior cerebellar artery aneurysms: clinical features and management. J Neurosurg 2002; 97:756–66. 322 Ogilvy CS, Carter BS, Kaplan S, Rich C, Crowell RM. Temporary vessel occlusion for aneurysm surgery: Risk factors for stroke in patients protected by induced hypothermia and hypertension and intravenous mannitol administration. J Neurosurg 1996; 84:785–91. 323 Piepgras DG, Khurana VG, Whisnant JP. Ruptured giant intracranial aneurysms. Part II. A retrospective analysis of timing and outcome of surgical treatment. J Neurosurg 1998; 88:430–5. 324 Sinson G, Philips MF, Flamm ES. Intraoperative endovascular surgery for cerebral aneurysms. J Neurosurg 1996; 84:63–70. 325 Chang HS, Fukushima T, Miyazaki S, Tamagawa T. Fusiform posterior cerebral artery aneurysm treated with excision and end-to-end anastomosis: case report. J Neurosurg 1986; 64(3):501–4. 326 Brilstra EH, Rinkel GJE, Klijn CJM, van der Zwan A, Algra A, Lo RTH et al. Excimer laser-assisted bypass in aneurysm treatment: short-term outcomes. J Neurosurg 2002; 97:1029–15. 327 Sekhar LN, Duff JM, Kalavakonda C, Olding M. Cerebral revascularization using radial artery grafts for the treatment of complex intracranial aneurysms: techniques and outcomes for 17 patients. Neurosurgery 2001; 49:646–58. 328 Hamada J, Todaka T, Yano S, Kai Y, Morioka M, Ushio Y. Vertebral artery-posterior inferior cerebellar artery bypass with a superficial temporal artery graft to treat aneurysms involving the posterior inferior cerebellar artery. J Neurosurg 2002; 96:867–71. 329 Derdeyn CP, Moran CJ, Cross DT, III, Sherburn EW, Dacey RG, Jr. Intracranial aneurysm: anatomic factors that predict the usefulness of intraoperative angiography. Radiology 1997; 205:335–9. 330 Tang G, Cawley CM, Dion JE, Barrow DL. Intraoperative angiography during aneurysm surgery: a prospective evaluation of efficacy. J Neurosurg 2002; 96:993–9. 331 Forsting M, Albert FK, Jansen O, Von Kummer R, Aschoff A, Kunze S et al. Coil placement after clipping: Endovascular treatment of incompletely clipped cerebral aneurysms: report of two cases. J Neurosurg 1996; 85:966–9. 332 Rabinstein AA, Nichols DA. Endovascular coil embolization of cerebral aneurysm remnants after incomplete surgical obliteration. Stroke 2002; 33:1809–15. 333 Kalavakonda C, Sekhar LN, Ramachandran P, Hechl P. Endoscope-assisted microsurgery for intracranial aneurysms. Neurosurgery 2002; 51:1119–26. 334 Sutherland GR, Kaibara T, Wallace C, Tomanek B, Richter M. Intraoperative assessment of aneurysm clipping using magnetic resonance angiography and diffusion-weighted

..

..

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 765

References

335

336

337

338

339

340

341

342

343

344

345

346

347

348

..

..

imaging: technical case report. Neurosurgery 2002; 50:893–7. Ooka K, Shibuya M, Suzuki Y. A comparative study of intracranial aneurysm clips: Closing and opening forces and physical endurance. Neurosurgery 1997; 40:318–23. Lawton MT, Heiserman JE, Prendergast VC, Zabramski JM, Spetzler RF. Titanium aneurysm clips. 3. Clinical application in 16 patients with subarachnoid hemorrhage. Neurosurgery 1996; 38:1170–5. Ogawa A, Suzuki M, Ogasawara K. Aneurysms at nonbranching sites in the supraclinoid portion of the internal carotid artery: Internal carotid artery trunk aneurysms. Neurosurgery 2000; 47:578–83. Yanaka K, Meguro K, Nose T. Repair of a tear at the base of a blister-like aneurysm with suturing and an encircling clip: technical note. Neurosurgery 2002; 50:218–21. Pelz DM, Ferguson GG, Lownie SP, Kachur E. Combined endovascular/neurosurgical therapy of blister-like distal internal carotid aneurysms. Can J Neurol Sci 2003; 30:49–53. Steiger HJ, Ito S, Schmid-Elsässer R, Uhl E. M2/M2 side-toside rescue anastomosis for accidental M2 trunk occlusion during middle cerebral artery aneurysm clipping: technical note. Neurosurgery 2001; 49:743–7. Baldwin ME, Macdonald RL, Huo D, Novakovia RL, Goldenberg FD, Frank JI et al. Early vasospasm on admission angiography in patients with aneurysmal subarachnoid hemorrhage is a predictor for in-hospital complications and poor outcome. Stroke 2004; 35:2506–11. Lasner TM, Weil RJ, Riina HA, King JT, Zager EL, Raps EC et al. Cigarette smoking-induced increase in the risk of symptomatic vasospasm after aneurysmal subarachnoid hemorrhage. J Neurosurg 1997; 87:381–4. Conway JE, Tamargo RJ. Cocaine use is an independent risk factor for cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke 2001; 32:2338–43. Hijdra A, van Gijn J, Nagelkerke NJ, Vermeulen M, van Crevel H. Prediction of delayed cerebral ischemia, rebleeding, and outcome after aneurysmal subarachnoid hemorrhage. Stroke 1988; 19(10):1250–6. Öhman J, Servo A, Heiskanen O. Risks factors for cerebral infarction in good-grade patients after aneurysmal subarachnoid hemorrhage and surgery: a prospective study. J Neurosurg 1991; 74(1):14–20. Qureshi AI, Sung GY, Razumovsky AY, Lane K, Straw RN, Ulatowski JA. Early identification of patients at risk for symptomatic vasospasm after aneurysmal subarachnoid hemorrhage. Crit Care Med 2000; 28:984–90. Friedman JA, Goerss SJ, Meyer FB, Piepgras DG, Pichelmann MA, McIver JI et al. Volumetric quantification of Fisher Grade 3 aneurysmal subarachnoid hemorrhage: a novel method to predict symptomatic vasospasm on admission computerized tomography scans. J Neurosurg 2002; 97:401–7. Friedman JA, Pichelmann MA, Piepgras DG, Atkinson JLD, Maher CO, Meyer FB et al. Ischemic complications of surgery for anterior choroidal artery aneurysms. J Neurosurg 2001; 94:565–72.

349 Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Koike T, Tanaka R. Serial changes of hemostasis in aneurysmal subarachnoid hemorrhage with special reference to delayed ischemic neurological deficits. J Neurosurg 1997; 86(4):594–602. 350 Vergouwen MD, Frijns CJ, Roos YB, Rinkel GJE, Baas F, Vermeulen M. Plasminogen activator inhibitor-1 4G allele in the 4G/5G promoter polymorphism increases the occurrence of cerebral ischemia after aneurysmal subarachnoid hemorrhage. Stroke 2004; 35:1280–3. 351 Chang HS, Hongo K, Nakagawa H. Adverse effects of limited hypotensive anesthesia on the outcome of patients with subarachnoid hemorrhage. J Neurosurg 2000; 92:971–5. 352 Roos YBWEM, Rinkel GJE, Vermeulen M, Algra A, van Gijn J. Antifibrinolytic treatment in aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2003; (2):CD001245. 353 Giller CA, Giller AM, Landreneau F. Detection of emboli after surgery for intracerebral aneurysms. Neurosurgery 1998; 42:490–3. 354 Romano JG, Forteza AM, Concha M, Koch S, Heros RC, Morcos JJ et al. Detection of microemboli by transcranial Doppler ultrasonography in aneurysmal subarachnoid hemorrhage. Neurosurgery 2002; 50:1026–30. 355 Dreier JP, Sakowitz OW, Harder A, Zimmer C, Dirnagl U, Valdueza JM et al. Focal laminar cortical MR signal abnormalities after subarachnoid hemorrhage. Ann Neurol 2002; 52:825–9. 356 Lang EW, Diehl RR, Mehdorn HM. Cerebral autoregulation testing after aneurysmal subarachnoid hemorrhage: the phase relationship between arterial blood pressure and cerebral blood flow velocity. Crit Care Med 2001; 29:158–63. 357 Khaldi A, Zauner A, Reinert M, Woodward JJ, Bullock MR. Measurement of nitric oxide and brain tissue oxygen tension in patients after severe subarachnoid hemorrhage. Neurosurgery 2001; 49:33–8. 358 Sobey CG, Faraci FM. Subarachnoid haemorrhage: What happens to the cerebral arteries? Clin Exp Pharmacol Physiol 1998; 25:867–76. 359 Fassbender K, Hodapp B, Rossol S, Bertsch T, Schmeck J, Schütt S et al. Endothelin-1 in subarachnoid hemorrhage: an acute-phase reactant produced by cerebrospinal fluid leukocytes. Stroke 2000; 31:2971–5. 360 Juvela S. Plasma endothelin concentrations after aneurysmal subarachnoid hemorrhage. J Neurosurg 2000; 92:390–400. 361 Mascia L, Fedorko L, Stewart DJ, Mohamed F, Terbrugge K, Ranieri VM et al. Temporal relationship between endothelin-1 concentrations and cerebral vasospasm in patients with aneurysmal subarachnoid hemorrhage. Stroke 2001; 32:1185–9. 362 Juvela S. Plasma endothelin and big endothelin concentrations and serum endothelin-converting enzyme activity following aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 97:1287–93.

765

9781405127660_4_014.qxd 10/13/07 2:15 PM Page 766

766

Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage

363 Kamezaki T, Yanaka K, Nagase S, Fujita K, Kato N, Nose T. Increased levels of lipid peroxides as predictive of symptomatic vasospasm and poor outcome after aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 97:1302–5. 364 Pilitsis JG, Coplin WM, O’Regan MH, Wellwood JM, Diaz FG, Fairfax MR et al. Free fatty acids in human cerebrospinal fluid following subarachnoid hemorrhage and their potential role in vasospasm: a preliminary observation. J Neurosurg 2002; 97:272–9. 365 Mack WJ, Mocco J, Hoh DJ, Huang J, Choudhri TF, Kreiter KT et al. Outcome prediction with serum intercellular adhesion molecule-1 levels after aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 96:71–5. 366 Mocco J, Mack WJ, Kim GH, Lozier AP, Laufer I, Kreiter KT et al. Rise in serum soluble intercellular adhesion molecule-1 levels with vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 97:537–41. 367 Nissen JJ, Mantle D, Gregson B, Mendelow AD. Serum concentration of adhesion molecules in patients with delayed ischaemic neurological deficit after aneurysmal subarachnoid haemorrhage: the immunoglobulin and selectin superfamilies. J Neurol Neurosurg Psychiatry 2001; 71:329–33. 368 Frijns CJM, Kasius KM, Algra A, Fijnheer R, Rinkel GJE. Endothelial cell activation markers and delayed cerebral ischaemia in subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2006; 77:863–7. 369 Polin RS, Bavbek M, Shaffrey ME, Billups K, Bogaev CA, Kassell NF et al. Detection of soluble E-selectin, ICAM-1, VCAM-1, and L-selectin in the cerebrospinal fluid of patients after subarachnoid hemorrhage. J Neurosurg 1998; 89:559–67. 370 McGirt MJ, Lynch JR, Blessing R, Warner DS, Friedman AH, Laskowitz DT. Serum von Willebrand factor, matrix metalloproteinase-9, and vascular endothelial growth factor levels predict the onset of cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Neurosurgery 2002; 51:1128–34. 371 Frijns CJ, Fijnheer R, Algra A, van Mourik JA, van Gijn J, Rinkel GJ. Early circulating levels of endothelial cell activation markers in aneurysmal subarachnoid haemorrhage: associations with cerebral ischaemic events and outcome. J Neurol Neurosurg Psychiatry 2006; 77:77–83.

372 Peltonen S, Juvela S, Kaste M, Lassila R. Hemostasis and fibrinolysis activation after subarachnoid hemorrhage. J Neurosurg 1997; 87:207–14. 373 Ikeda K, Asakura H, Futami K, Yamashita J. Coagulative and fibrinolytic activation in cerebrospinal fluid and plasma after subarachnoid hemorrhage. Neurosurgery 1997; 41:344–9. 374 Gaetani P, Tartara F, Pignatti P, Tancioni F, Baena RRY, De Benedetti F. Cisternal CSF levels of cytokines after subarachnoid hemorrhage. Neurol Res 1998; 20:337–42. 375 Fassbender K, Hodapp B, Rossol S, Bertsch T, Schmeck J, Schütt S et al. Inflammatory cytokines in subarachnoid haemorrhage: association with abnormal blood flow velocities in basal cerebral arteries. J Neurol Neurosurg Psychiatry 2001; 70:534–7. 376 Sviri GE, Feinsod M, Soustiel JF. Brain natriuretic peptide and cerebral vasospasm in subarachnoid hemorrhage: clinical and TCD correlations. Stroke 2000; 31:118–22. 377 Takenaka KV, Sakai N, Murase S, Kuroda T, Okumura A, Sawada M. Elevated transferrin concentration in cerebral spinal fluid after subarachnoid hemorrhage. Neurol Res 2000; 22:797–801. 378 Wijdicks EFM, Heublein DM, Burnett JC, Jr. Increase and uncoupling of adrenomedullin from the natriuretic peptide system in aneurysmal subarachnoid hemorrhage. J Neurosurg 2001; 94:252–6. 379 Hoffman WE, Wheeler P, Edelman G, Charbel FT, Torres NJ, Ausman JI. Hypoxic brain tissue following subarachnoid hemorrhage. Anesthesiology 2000; 92:442–6. 380 Kett-White R, Hutchinson PJ, Al Rawi PG, Gupta AK, Pickard JD, Kirkpatrick PJ. Adverse cerebral events detected after subarachnoid hemorrhage using brain oxygen and microdialysis probes. Neurosurgery 2002; 50:1213–21. 381 Schulz MK, Wang LP, Tange M, Bjerre P. Cerebral microdialysis monitoring: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2000; 93:808–14. 382 Staub F, Graf R, Gabel P, Köchling M, Klug N, Heiss WD. Multiple interstitial substances measured by microdialysis in patients with subarachnoid hemorrhage. Neurosurgery 2000; 47:1106–14. 383 Vermeulen M, van Gijn J, Hijdra A, van Crevel H. Causes of acute deterioration in patients with a ruptured intracranial aneurysm: a prospective study with serial CT scanning. J Neurosurg 1984; 60(5):935–9.

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 767

15

Specific interventions to prevent intracranial haemorrhage 15.1 Introduction

767

15.2 Saccular intracranial aneurysms

767

15.3 Arteriovenous malformations of the brain

15.1 Introduction

This chapter will deal with patients who have not yet had a stroke but who do have a vascular lesion that might give rise to a stroke in the future, mostly unruptured intracranial aneurysms or arteriovenous malformations (AVMs). Sometimes these lesions warrant preventive intervention, but in many situations uncertainty abounds,1 and in many other situations intervention will probably do more harm than good.2 A general point to be made about the management of these lesions is that there is no hurry. Most of these patients are in perfect health and it is important to carefully balance the risks and benefits of all treatment options and to take time to counsel the ‘patient’. Primum non nocere is a pivotal guideline when a lesion is found by chance, or as the result of an ‘excludogram’ in patients with innocuous types of headache or dizziness, or – even worse – as the result of whole-body MR scans for personal health checks. Doing no harm also includes refraining from descriptions of these lesions as ‘a time bomb in your head’ before referring the unfortunate patient to a neuro-interventional centre. For many of these patients no intervention is the best option, but having to live with an untreated aneurysm or AVM imposes a threat on quality of life, of which anxiety is an important component and may need managing in its own right.3 Careful counselling and weighing the pros and cons is even

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

780

more important when screening for aneurysms or AVMs is considered. Screenees often have unrealistic risk perceptions,4 and screening for intracranial aneurysms is associated with considerable psychosocial effects, both positive and negative.5 Incidentally found aneurysms or arteriovenous malformations can give rise to subarachnoid or intracerebral haemorrhage, but in many situations preventive treatment will either do more harm than good, or is of uncertain benefit. Other intracranial vascular malformations such as cavernous malformations (section 8.2.5), cranial dural arteriovenous fistulae (section 8.2.8.), and venous angiomas (section 8.2.6) are not discussed in this chapter. Their management is seldom guided by their risk of causing a stroke (section 13.3.8). And abnormalities such as venous angiomas are better left untreated, in part because they have an important function in venous drainage.

15.2 Saccular intracranial aneurysms

15.2.1 Prevalence of aneurysms Intracranial saccular aneurysms are almost never found in neonates, and they are rare in children (section 9.1.1). Their frequency in the general adult population depends on the definition of their size and the diligence with which the search for unruptured aneurysms has been performed. In our systematic overview of studies reporting the prevalence of intracranial aneurysms in patients 767

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 768

768

Chapter 15 Specific interventions to prevent intracranial haemorrhage

Frequency Retrospective autopsy studies Prospective autopsy studies Retrospective angiography studies Prospective angiography studies Age < 20 20–39 40–59 60–80 > 80 Adult without risk factors Risk factors Women Atherosclerotic diseases One or more affected relatives Autosomal dominant polycystic kidney disease Sizes of aneurysms < 6 mm 6–10 mm 11–20 mm > 20 mm Sites of aneurysms Anterior communicating artery Middle cerebral artery Internal carotid artery Posterior circulation

% 0.4 3.6 3.7 6.0 % 0.01 1.3 1.8 2.3 2.1 2.3 Relative risk 1.3 2.3 4.0 4.4 % 72 21 6 2 % 24 30 42 10

studied for reasons other than subarachnoid haemorrhage (SAH), 23 studies were identified, totalling 56 304 patients.6 The prevalence was lowest in retrospective autopsy studies and highest in prospective angiography studies (Table 15.1). The angiography studies were somewhat biased towards overestimation, because in around 15% of the patients the indication was a planned operation for a suspected pituitary tumour, polycystic kidney disease or a family history of SAH, conditions in which the frequency of aneurysms is higher than average. For adults without specific risk factors for aneurysms, the prevalence was 2.3% (95% CI 1.7–3.1%); most of the aneurysms were small and located in the anterior circulation. In the general population, the prevalence of known aneurysms in Olmsted County, Minnesota, was estimated at 0.83%,7 which implies that even in that highly investigated region only about one out of four aneurysms is detected. The prevalence of aneurysms is higher in women than in men and tends to increase with increasing age; other risk factors are the presence of carotid artery stenosis or ischaemic heart disease (not unexpected because these conditions share common risk factors with aneurysms such as smoking, hypertension and excessive use of alcohol (section 9.1.1), having one or more affected

(95% CI) (0.4–0.5) (3.1–4.1) (3.0–4.4) (5.3–6.8)

Table 15.1 Prevalence of, risk factors for, and site of unruptured intracranial aneurysms.6

(0.00–0.03) (0.8–2.1) (1.4–2.2) (1.9–2.6) (1.5–3.0) (1.7–3.1) (95% CI) (0.9–2.0) (1.7–3.1) (2.7–6.0) (2.7–7.3) (68–77) (17–25) (4–10) (0.1–3) (95% CI) (20–27) (26–34) (37–46) (7.7–13)

relatives with SAH, autosomal dominant polycystic kidney disease (ADPKD), and a previous episode of SAH. Around 2% of the adult population have one or more intracranial aneurysms. Most are very small and located in the anterior circulation. The risk of having an intracranial aneurysm is higher in women, in people with a family history of intracranial aneurysms, and in patients with autosomal dominant polycystic kidney disease or a history of aneurysmal subarachnoid haemorrhage. Family history The chance of having an aneurysm and the lifetime risk of subarachnoid haemorrhage depend on the number of affected first-degree relatives, defined as parents, siblings and children (Table 15.2).6,8 The lifetime risk of SAH in individuals with two or more affected first-degree relatives is unknown because there are so few families with this history, and because members of these families tend to seek screening for aneurysms and treatment of those detected. The chance of an aneurysm depends not only on the number of affected relatives but also on the nature of the relationship. If the affected relatives are

..

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 769

15.2 Saccular intracranial aneurysms Table 15.2 Risk of having an aneurysm, and lifetime risk of subarachnoid haemorrhage, according to the number of affected first-degree relatives.

Number affected firstdegree relatives

Risk of aneurysm % (95% CI)

Risk of subarachnoid haemorrhage % (95% CI)

None 1 2 or more

2.3 (1.7–3.1) 4.0 (2.6–5.8) 8.0 (3.9–14)

0.6 (0.6–0.7) 3.3 (1.0–11) No data

siblings, the risk of having an aneurysm is higher than if they are parents or children.6,8,9 No other characteristics of the patient who presents for screening or their relatives have been identified that affect the risk of an aneurysm. After negative screening, new aneurysms may develop. People with two or more affected relatives and with normal magnetic resonance angiography have a 7% risk of developing an aneurysm within 5 years of screening.10 Aneurysms are more commonly large and multiple in familial than in sporadic SAH.11 However, because only 10% of cases of SAH are associated with a family history, large and multiple aneurysms are actually more commonly seen in sporadic than in familial SAH. Patients with familial SAH tend to be younger than sporadic cases, and in families with two generations affected the age at onset is earlier in the later than the previous generation.12–14 Autosomal dominant polycystic kidney disease SAH from a ruptured aneurysm is common in patients with autosomal dominant polycystic kidney disease (ADPKD),15 and causes a large proportion of deaths in these patients (section 9.1.1). In a series of 101 autopsies, of which 89 included assessment of the brain, 17 had had a recent SAH from a ruptured aneurysm.16 Although selection bias may have occurred, the proportion of deaths from ruptured aneurysms was much higher than in the general population.6 There are some differences in the features of aneurysms between patients with and without ADPKD: • The most common site of aneurysm in ADPKD patients is the middle cerebral artery, whereas it is the internal carotid artery in the general population.6,17 • Aneurysms in ADPKD patients are more often larger than 10 mm (in 25%) than in patients without ADPKD (in 10%).17 • A positive family history, defined as a first-degree or second-degree relative with SAH or intracranial aneurysm, is found in 40% of ADPKD compared with 10% of those without ADPKD.8 In ADPKD patients, the large proportion of aneurysms of the middle cerebral artery, the large proportion of large aneurysms and the younger age at onset of SAH17 more closely matches patients with familial than sporadic SAH.11,12

..

..

Intracranial aneurysms are found in about 10% of patients with ADPKD.6 But apart from a family history for SAH, no clinical characteristics have been identified that increase the risk of aneurysm in these patients.18 Although the position of the mutation in PKD1 is predictive of development of intracranial aneurysms,19 currently this mutation detection is not yet available in clinical practice. Aneurysms commonly develop in these patients in the absence of hypertension,17,20 which indicates that blood pressure is not an essential factor in the development of their aneurysms. Patients with ADPKD are at risk of new episodes of SAH from newly developed aneurysms,21 but at present there are no precise estimates of this risk. Only three studies, each with no more than 20 patients and with various periods of follow-up, have been published thus far. The risk of aneurysm development ranged from 1 in 17 patients (6%) after 7 years to 5 of 20 patients (25%) after 15 years.22–24 The size of the newly developed aneurysms ranged from 2 to 7 mm. Development of new aneurysms The continuing development of aneurysms during life and the presence of multiple aneurysms in up to 30% of patients with SAH suggest that patients who have been successfully treated for one aneurysm are at risk of developing another one. A few series have reported on the risk of developing recurrent (at the same site as the previously clipped aneurysm) or de novo (at another site) aneurysm. This risk of finding a new aneurysm increases, not unexpectedly, with the increasing interval after the initial SAH from around 2% after 4 years, to 10% after 10 years, to more than 15% after 15 years.25–29 Around 20% of the newly found aneurysms are located at the site of the original aneurysm, and the other 80% at other sites (Figs 15.1 and 15.2). Approximately one-third of the detected aneurysms are truly de novo, the remaining two-thirds were in fact already visible in retrospect.29 Risk factors for de novo aneurysm formation are multiple aneurysms at the time of the SAH (Hazard ratio (HR) 3.2; 95% CI 1.2–8.6), current smoking (HR 3.8; 95% CI 1.5–9.4) and hypertension (HR 2.3; 95% CI 1.1–4.9).27,28 All this suggests that having one or more intracranial aneurysms should not be considered as a single event but

769

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 770

770

Chapter 15 Specific interventions to prevent intracranial haemorrhage

(a)

(b)

Fig. 15.1 Regrowth of an aneurysm at the site of the initial aneurysm. (a) Preoperative catheter angiogram from 1992 of a patient with a subarachnoid haemorrhage, presumably from the anterior communicating artery aneurysm (arrow). There is an additional aneurysm distal on the posterior communicating

(b)

(a)

rather as a continuous process – aneurysms can and do develop de novo, and they grow. In addition to recurrent or de novo aneurysms, patients who have survived an SAH may have had additional aneurysms that were small and therefore missed during angiography, or were thought to be too small to be harmful and therefore left untreated. These aneurysms can grow, which increases the risk of rupture. The enlargement increases with longer periods of follow-up from 25% after 9 years to 45% after 19 years.28,29 These new and enlarged aneurysms clearly put patients who have survived an SAH at some increased risk

(c) artery (arrowhead). (b) Postoperative angiogram from 1992 showing complete obliteration of both aneurysms by clips. (c) Angiogram from 2003 showing regrowth of the aneurysm on the anterior communicating artery (arrow). (Adapted from Werner et al.29 with permission from Oxford Journals.)

Fig. 15.2 Development of a new aneurysm at a different site from the initial aneurysm. (a) Preoperative catheter angiogram from 1993 of the left carotid artery of a patient with a subarachnoid haemorrhage from a ruptured aneurysm of the anterior communicating artery (not shown). There is no aneurysm on the carotid artery (arrow). (b) Angiogram from 2001 showing a new aneurysm on the left carotid artery (arrow). (Reproduced with permission from Werner et al.10 © 2003 Stroke.)

of another SAH. The risk estimates vary between 3% in 10 years to 9% in 20 years after successful clipping of a ruptured aneurysm.30,31 In the first 10 years after SAH the incidence of SAH is about 20 times higher than the risk in an age- and sex-matched cohort in the general population (section 14.10.2).31 These estimates include all causes of SAH, including re-rupture of the initially treated aneurysm. Since endovascular-treated aneurysms have a higher risk of re-rupture at least in the initial year after the SAH,32 patients with endovascular-treated aneurysms probably have an even higher risk of recurrent SAH than the estimates given here, but precise estimates are not available yet.

..

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 771

15.2 Saccular intracranial aneurysms

Intracranial aneurysms are not congenital lesions, but develop during life. Patients who have survived an episode of subarachnoid haemorrhage are at risk of having a new episode from new aneurysms, from already existing but enlarging aneurysms, and from recurrence of the treated aneurysm.

15.2.2 Risk of rupture of unruptured intracranial aneurysms In our systematic review of all the follow-up studies on the risk of SAH from unruptured aneurysms between 1965 and 1995, the overall risk of rupture was 1.9% per annum.6 The absolute risk for aneurysms equal to or less than 10 mm diameter (i.e. the most common type) was 0.7% per annum. Subsequent to this systematic overview, the first report from the International Study of Unruptured Intracranial Aneurysms (ISUIA) study group was published, in which a retrospective cohort of 1449 patients with 1937 intracranial aneurysms greater than 2 mm diameter had been followed up for an average of 8.3 years.33 Half the patients had one or more incidental aneurysms (no history of SAH from a different aneurysm), the other half had an aneurysm additional to a ruptured aneurysm that had been successfully repaired. In the former group, the cumulative rate of rupture of aneurysms that were less than 10 mm in diameter at diagnosis was less than 0.05% per year; in the latter group, it was approximately 10 times higher (0.5% per year). The size and location of the aneurysm were independent predictors of rupture: the relative risk was 11.6 for aneurysms greater than 10 mm, 13.7 for posterior circulation aneurysms and 8.0 for posterior communicating artery aneurysms. Of course, the retrospective design used for case finding in this study may have led to important sources of bias. For example, the risk of rupture may have been underestimated because one inclusion criterion was a complete set of angiograms being available, and yet it is conceivable that the films of patients who

Table 15.3 Five-year cumulative risk of rupture of an unruptured intracranial aneurysm according to the prospective ISUIA report.2

had died in the meantime, e.g. from aneurysmal rupture, may have been removed from the hospital records. The later report from the ISUIA study group describes the largest ever prospectively studied cohort of patients with unruptured aneurysms, 1692 patients of whom 1077 had incidental aneurysms and 615 additional aneurysms.2 The period of follow-up was relatively short, a mean of 4.1 years, during which 534 patients (32%) in the intention-not-to-treat arm nevertheless had their aneurysm occluded, and another 193 (11%) died for reasons other than aneurysmal rupture. This case fatality is rather high, which suggests that in many patients the decision to withhold treatment was based on patient characteristics (e.g. reduced life expectancy) rather than aneurysm characteristics. The risk of rupture for the remaining patients was presented as 5 years cumulative risks (Table 15.3), and not as mean rupture rates per year, which is a strength of the paper. However, despite the merits of being the largest prospective study on risk of rupture of unruptured intracranial aneurysms, the ISUIA report has several important weaknesses: • First, the dichotomy for small vs large aneurysms was set at 7 mm and not at 10 mm as in previous reports from this group, but this dichotomy was not prespecified in the study protocol. • Second, aneurysms of the posterior communicating artery were included with posterior circulation aneurysms and not with anterior circulation aneurysms. This classification of posterior communicating aneurysms (which arise from the internal carotid artery and are therefore technically on the anterior circulation) is different from all other studies, and again was not pre-specified in the study protocol. Subsequently this approach has been criticized by a review of 11 papers (including the two ISUIA reports) that contained sufficient data to calculate rupture rates for anatomical subgroups;34 bleeding rates for aneurysms of the posterior communicating artery (0.46% per year) were

Size

ACA/MCA/ICA

Post communicating/ post circulation

< 7 mm no previous SAH < 7 mm but previous SAH 7–12 mm 13–24 mm > 24 mm

0% (95% CI unknown) 1.5% (95% CI unknown) 2.6% (95% CI unknown) 14.5% (95% CI unknown) 40% (95% CI unknown)

2.5% (95% CI unknown) 3.4% (95% CI unknown) 14.5% (95% CI unknown) 18.4% (95% CI unknown) 50% (95% CI unknown)

SAH, subarachnoid haemorrhage; ACA, anterior cerebral artery (includes anterior communicating artery); MCA, middle cerebral artery; ICA, internal carotid artery.

..

..

771

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 772

772

Chapter 15 Specific interventions to prevent intracranial haemorrhage

similar to those of aneurysms on the remaining sites on the anterior circulation (0.49% per year) but much lower than those for aneurysms of the posterior circulation (1.8% per year). • A third criticism is that only 51 aneurysms ruptured during follow-up which precludes a sound analysis of risk factors for rupture. • Fourth, the confidence intervals are not given, and cannot be calculated from the data presented in the paper, probably they are very wide for lack of numbers. For example for patients with aneurysms < 7 mm on the anterior circulation the number at risk decreased from 535 at 1 year of follow-up to only 35 at 6 years. The uncertainty around the estimates is one reason why longer follow-up of these patients is crucial. The second reason is that real data on 5 or 10 years and preferably even longer periods of follow-up are likely to be more accurate than extrapolations of risks from short-term follow-up into yearly risks of rupture that are considered to be constant during the remaining life expectancy of patients. Existing decision models on aneurysm management (implicitly) incorporate a constant, time-independent risk of rupture. It is however likely that aneurysms exhibit fast growth in short periods of time,35,36 and that the risk of rupture is confined to these episodes. It is therefore over-simplistic to calculate a mean rupture rate per annum from the observed but short period of follow-up and extrapolate that rupture rate to the remaining life expectancy. Very long-term follow-up in patients with unruptured aneurysms is Variable

Age < 20 years 20–29 years 40–59 years 60–79 years > 80 years Gender Men Women Hypertension No Yes Smoking No Yes Population Non-Japanese/Finnish Japanese or Finnish

Patient years of follow-up

No. with subarachnoid haemorrhage

essential before firm conclusions can be made. This longterm follow-up can also take into account the risk of SAH from newly developed aneurysms.37 Patients of course are more interested in the risk of rupture in the following 20 or 30 years than in the next 5 years alone. Happily the ISUIA investigators are now continuing their follow-up. In the 2007 update of our systematic review of follow-up studies on the risk of SAH from unruptured aneurysms we included 19 studies with a total of 4705 patients.38 The overall risk of rupture of untreated aneurysms in the studies with a mean follow-up < 5 years was 1.2% (95% CI 1.0–1.5), in those with a mean follow-up between 5 and 10 years 0.6% (0.5–0.7%) and in the studies with a mean follow-up of greater than 10 years 1.3% (0.9–1.8%) per patient-year. The lack of an increase in risk of rupture with increasing duration of follow-up might be explained by the different study populations have inherently different risks of rupture. Pooled analysis of individual data, taking also into account risk factors other than duration of follow-up, are needed to better estimate the relation between duration of followup and risk of rupture. Patient characteristics that had in univariable analysis a statistically significant association with an increased risk of rupture were age > 60, female gender and Japanese or Finnish descent (Table 15.4). In addition, although smoking increased the risk of rupture, this was not statistically significant. There were not enough data to evaluate the effects of excessive alcohol use or a family history of SAH. Circumstantial evidence, however, suggests that aneurysms in patients with a Relative risk (95% CI)

— 848 1830 709 12

— 12 24 19 0

— (0.5–2.2) 1.0 2.0 (1.1–3.7) —

2255 2885

32 65

1.0 1.6 (1.1–2.4)

2357 572

35 9

1.0 1.1 (0.5–2.2)

1404 1304

13 20

1.0 1.7 (0.8–3.3)

20422 6093

111 113

1.0 3.4 (2.6–4.4)

Table 15.4 Relative risk of rupture of an unruptured intracranial aneurysm according to patient characteristics.38

Relative risk of 1.0 indicates the reference category.

..

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 773

15.2 Saccular intracranial aneurysms Table 15.5 Relative risk of rupture of an unruptured intracranial aneurysm according to aneurysm characteristics.38

Variable

Site of aneurysm ACA ICA* MCA Posterior circulation** Cavernous sinus Size of aneurysm < 5 mm 5–10 mm > 10 mm > 12 mm giant (>15 mm) Type of aneurysm Incidental Additional Symptomatic

Patient years of follow-up

No. with subarachnoid haemorrhage

Relative risk (95% CI)

1083 3558 2734 791 2159

19 46 33 26 2

1.4 (0.8–2.3) 1.0 0.9 (0.6–1.5) 2.5 (1.6–4.1) 0.1 (0–0.3)

1939 1187 3670 1089 293

10 14 55 42 18

1.0 2.3 (1.0–5.2) 2.9 (1.5–5.7) 7.5 (3.8–14.9) 11.9 (5.5–25.8)

3315 3158 472

50 46 31

1.0 1.0 (0.7–1.4) 4.4 (2.8–6.8)

*includes posterior communicating artery; **posterior circulation = vertebral artery, basilar artery and posterior cerebral artery; ACA, anterior cerebral artery (includes anterior communicating artery); ICA, internal carotid artery; MCA, middle cerebral artery. Relative risk of 1.0 indicates the reference category.

family history of SAH are at higher risk of rupture; individuals with one affected relative carry a 5.5-fold greater lifetime risk for SAH than the general population (Table 15.2). Nevertheless, the chance of finding an aneurysm by screening an individual with one affected relative is only 1.7 times higher than the general population. A possible explanation is that familial intracranial aneurysms have a higher risk of rupture. An alternative explanation is that individuals with a family history of SAH are more likely to develop a new aneurysm than those without. Aneurysm characteristics that were related to an increased risk of rupture were site at the posterior circulation, size > 5 mm and symptoms caused by the aneurysm other than SAH (Table 15.5). We could not confirm a higher risk of rupture in additional compared with incidental aneurysms but this should not be considered as proof of absence of a difference. A potential explanation is that the additional aneurysms were smaller than the incidental aneurysms; a higher risk for additional aneurysms may therefore have been masked by their smaller size with inherently lower rupture risk. The published data from the parent papers did not allow any multivariable analysis. Thus, we could not assess the independent contribution of patient and aneurysm characteristics to the risk of aneurysm rupture. After all, it is likely that some of the characteristics

..

..

are not independent, for example although we found high rate ratios for both large and for symptomatic aneurysms, it is unlikely that these factors are independent because larger aneurysms have a greater tendency to cause compression problems than smaller ones. Due to this lack of multivariable analyses, there is still considerable uncertainty for the calculation of aneurysmal rupture risk in an individual. Pooled analysis of individual data are needed to identify independent risk factors for aneurysm rupture, and follow-up time should be taken into account since the growth of aneurysms is probably not constant over time. Only then will we have more reliable risk estimates for individual patients. Small aneurysms are less likely to rupture than large aneurysms. Other risk factors for rupture are age, female gender and the site of the aneurysm. However, there are still many uncertainties about the absolute risks of rupture, especially for periods longer than 5 years after detection of the aneurysm. The importance of aneurysm size for the risk of rupture puzzles many physicians and surgeons, because most ruptured aneurysms they encounter in clinical practice are smaller than 10 mm, or even smaller than 5 mm.39,40 The most likely explanation for this paradox is that the vast majority of aneurysms are small, and that

773

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 774

774

Chapter 15 Specific interventions to prevent intracranial haemorrhage

even a minute fraction of those that do rupture will still outnumber a large proportion of the much smaller group of patients with larger aneurysms that rupture (in the same way as most children with Down syndrome are born to young mothers, even though the risk increases with age). This ‘prevention paradox’ means that screening the population for people at highest risk (large aneurysms) will do little to change the burden of the disease (i.e. ruptured aneurysms) because most patients who actually have an SAH have ruptured a small aneurysm.41

15.2.3 Treatment options Neurosurgical clipping and endovascular coiling are not without risks. Those of neurosurgical clipping have been studied most intensively, because this treatment has been available the longest. Neurosurgical clipping A systematic review of studies on complications of neurosurgical clipping for unruptured aneurysms published between 1966 and 1996 included 61 studies, on 2460 patients, with at least 2568 aneurysms.42 Of these aneurysms one-quarter were larger than 25 mm and one-third located on the posterior circulation, which implies a bias towards ‘dangerous’ and ‘difficult-to-treat’ aneurysms. Overall case fatality was 2.6% (95% CI 2.0–3.3%); permanent morbidity occurred in another 10.9% (95% CI 9.6–12.2%). Postoperative mortality and morbidity were significantly lower in more recent years, for non-giant aneurysms and for aneurysms on the anterior circulation. The most recent report of the ISUIA investigators included 1917 prospectively identified patients who underwent neurosurgical clipping of an unruptured aneurysm.2 Poor outcome from surgical treatment was defined as death or a Rankin score of 3, 4 or 5 (indicating dependence in the sense that help was required for activities of daily living), a score less than 24 on the mini-mental state examination, or a score less than 27 on a telephone interview for cognitive status (both indicating a serious cognitive abnormality). The proportion of patients with a poor outcome ranged between 5% and 40% according to their age and site and size of the aneurysm (Table 15.6). Of note is that both the review and the ISUIA report describe the data from many centres in many countries, with unspecified but probably varying degrees of treatment team expertise. Reports from single centres with a lot of experience describe complication rates as low as 2% in patients with aneurysms smaller than 10 mm,43–45 but assessment

of outcome in these studies was less rigorous than in the ISUIA study. Nevertheless, the experience of the neurosurgeon seems to be an important factor in the risk of complications.44,46 Endovascular coiling A systematic review of success and complications of occlusion by means of coils included 30 studies with 1379 patients published before 2003.47 The overall case fatality was 0.6% (95% CI 0.2–1%) and the permanent morbidity was 7% (95% CI 5.3–8.7%). The bleeding risk of treated aneurysms was 0.9% per year (95% CI 0.4– 1.4%), but rebleeding occurred only in incompletely coiled aneurysms larger than 10 mm. For the subset of patients with aneurysms larger than 10 mm the bleeding risk was 3.5% per year (95% CI 0.8–6.2%). Unfortunately, most of the studies included in the review were weak, so weak that the pre-specified subgroup analysis of good quality studies could not be performed because none of the included studies satisfied the predefined criteria for high quality. An important and promising observation was that the complication rate was significantly lower in recent studies compared with the older studies. Data from the ISUIA investigators were published after the inclusion period of the systematic review, but are comparable (Table 15.6).2 An advantage of the ISUIA report over the systematic review is that the ISUIA investigators were able to assess complication rates according to site and size of the aneurysms, and the age of the patients. As was the case with neurosurgical clipping, the risk of complications increased with increasing size of aneurysm. A difference compared with clipping is that while the risk of complications rises considerably after the age of 50 for clipping it remains stable for coiling. Only after the age of 70 does the complication rate increase for coiling.2 Similar to neurosurgical clipping, the systematic review and the ISUIA report describe data from many centres in many countries, with varying degrees of expertise in treatment, whereas single-centre reports describe lower complication rates. For example, an observational study from the Netherlands reported on 48 patients with 58 incidental aneurysms and no previous SAH; the case fatality was zero and morbidity was 2%, even though half the aneurysms were larger than 10 mm.48 However, such single-centre reports are subject to several sources of bias, and are likely to give an underestimate of the actual complication risks.49,50 Interestingly, as well as clinically obvious complications, almost half the patients develop new lesions on MR imaging during endovascular treatment of aneurysms.51

..

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 775

15.2 Saccular intracranial aneurysms Table 15.6 Risk of complications (%) of treatment of unruptured intracranial aneurysms according to site and size of the aneurysm and age of the patient in the prospective ISUIA 2003 report.2

Site, size, age

Anterior / < 50 yrs < 13 mm 13–24 mm > 24 mm Anterior / > 50 yrs < 13 mm 13–24 mm > 24 mm Posterior / < 50 yrs < 13 mm 13–24 mm > 24 mm Posterior / > 50 yrs < 13 mm 13–24 mm > 24 mm

Surgical clipping (n = 1917)

Endovascular coiling (n = 451)

%

(95% CI)

%

(95% CI)

5.7 4.3 23

(4.3–7.1) (0.1–8.6) (2–42)

7.1 4.6 15

(0.1–15) (0.1–15) (0.1–44)

11 24 31

(7.9–14) (18–30) (19–43)

6.3 8.3 13

(1.3–11) (2.1–15) (1.3–25)

11 29 44

(1.6–20) (4.7–53) (unknown)*

11 17 no data

(0.1–28) (0.1–57)

12 43 no data

(3.2–21)

7.9 23 40

(0.1–19) (5.5–41) (unknown)*

(unknown)*

*The confidence intervals are recalculated from bars in a figure; for this confidence interval the bar extended to the upper limit of the figure.

There are no randomized comparisons between clipping and coiling for unruptured aneurysms. An indirect prospective comparison between only 19 patients treated by coiling and 32 treated by clipping revealed that in the short term, operation of patients with an unruptured aneurysm had a considerable impact on their functional health and quality of life. After 1 year, improvement had occurred but recovery was incomplete.52 Coiling did not affect functional health and quality of life, but one of the patients treated by coiling had a haemorrhage from the coiled aneurysm during follow-up. In a retrospective cohort study from the USA using inpatient data from 1996 to 2000, no difference between clipping and coiling was found for case fatality and discharge to long-term care.53 On the basis of a fourlevel discharge status outcome scale, coiled patients had a significantly better discharge disposition. Moreover, length of stay was longer and charges were higher for clipped than for coiled patients. In another study from the US of patients who were in retrospect eligible for both clipping and coiling, those who had actually been treated by clipping had a longer stay in hospital, more often were dependent on help for activities of daily living, and more often reported persistent new symptoms or disability after treatment.54 Thus, for patients with unruptured aneurysms, coiling seems to be associated with fewer complications than clipping. Because longterm follow-up data are not yet available, it is not known if these two treatments prevent SAH to the same extent.

..

..

Neurosurgical clipping and endovascular coiling of unruptured aneurysms both carry a definite risk of death or permanent disability, even for small aneurysms. The risk increases for larger aneurysms and with increasing age, although the risk starts to increase at an earlier age for clipping than for coiling. Indirect comparisons suggest that the impact on health is less severe for coiling than for clipping, especially in the short term.

15.2.4 Management of patients with unruptured aneurysms Every time a cerebral aneurysm is a surprise finding on an imaging study performed for another purpose, there is a practical dilemma: is the risk of preventive clipping or coiling of the aneurysm at that time outweighed by the risk of death or disability from rupture of the untreated aneurysm at some time later in life, if it ruptures at all? Two pivotal factors in this balance of risks are the size and site of the aneurysm, but these factors are not necessarily helpful because both the risk of rupture and the risk of treatment complications are greater for large aneurysms and for aneurysms on the posterior circulation. Age is the most discriminating factor, because at a young age the benefit of treatment – which is long term – is large, provided life expectancy is not reduced for other reasons, and the risk of treatment relatively small. With increasing age, the benefit decreases (because

775

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 776

776

Chapter 15 Specific interventions to prevent intracranial haemorrhage

life expectancy decreases) while the risk of complications of preventive treatment increases. Further factors that should be taken into account are the type of aneurysm (incidental, additional or symptomatic; section 15.2.2), family history and comorbidity. Patients with a positive family history may have an increased risk of rupture, but definitely have a different risk perception than someone with no experience of an SAH in their family. If someone has already lost one or more relatives from SAH (in most instances from rupture of a small aneurysm) it will be very difficult for him or her to accept that their own aneurysm has a negligible risk of rupture, even if it is small. Comorbidity reduces life expectancy (and thereby the benefit of preventive treatment) and often also increases the risk of complications. For example, for a 64-year-old patient with a minor ischaemic stroke and 50% stenosis of the symptomatic carotid artery, the chance of dying from ischaemic cardio- or cerebral vascular disease in the next 10 years is probably much higher than the risk of dying from SAH from the incidental 9-mm aneurysm on the left MCA detected during work-up after the stoke. Moreover, the carotid stenosis increases the risk of complications from preventive treatment. For balancing the risks and benefits, decision analysis can be helpful, but most of the decision models developed thus far are over-simplistic. An important drawback is that these models (implicitly) presume a constant, time-independent risk of rupture, whereas a ‘chaotic’ or ‘periodic’ growth process with inherent short-lasting episodes of high risk of rupture is much more likely.36 This means that the rupture risks found during a certain period of time cannot be extrapolated to the patient’s remaining lifetime. A further problem is that the risk of preventive treatment for the aneurysm in question and for the neurosurgeon or radiologist in particular should be known. In many instances the overall risk of the procedure will be no more than an educated guess (probably underestimated by the surgeon or radiologist and overestimated by the neurologist). Also, the remaining life expectancy often will be another educated guess, especially because for many patients the aneurysm will have been detected during the work-up for another disease, which may have its own impact on life expectancy. Thus, before decision models can be used in clinical practice, they have to be refined. Despite all these uncertainties, in some situations giving balanced advice is not that difficult. In a young person with an aneurysm larger than 10 mm at the top of the basilar artery that is easily accessible for the radiologist (who has ample experience and a good track record for the procedure), most physicians would advise preventive treatment. Another ‘easy’ example is the 75-year-old man

with diabetes, a recent myocardial infarction, a more than 70% (asymptomatic) carotid artery stenosis and a 4mm ipsilateral middle cerebral artery aneurysm. Most physicians would advise to leave the aneurysm untreated, although one of us saw such a patient die from SAH the very day after the advice ‘not to treat’. A third example, which starts easy, is the young (say 35 years old) patient with a 3-mm aneurysm on the anterior communicating artery and no further risk factors. Most of us would advise not treating the aneurysm. The difficulty in this example is whether or not to perform follow-up imaging to assess aneurysmal growth. The yield of annual followup is small. In a prospective study of 93 patients with 125 small aneurysms CTA or MRA was performed after an interval of 1 year;55 in three patients (3.2%) an aneurysm enlarged slightly (0.5 to 1.5 mm), but for none did the enlargement lead to a decision to treat. In the meantime, two of the 93 patients (2.2%) had an SAH: one from an aneurysm at the clip site of a previous operation that ruptured without enlargement, and the other from a newly developed dissecting aneurysm. The number of enlarged aneurysms was too small for sound analysis of risk factors, but patients with enlarged aneurysm more often had a positive family history of SAH or a previous episode of stroke, were younger, more often had multiple aneurysms and more often were current smokers than patients without enlargement.55 Since one of every four aneurysms increases in size over 10 years and one of every two aneurysms over 20 years (section 15.2.1) further studies are clearly needed to assess the optimal screening interval. In some other circumstances the decision whether or not to treat is made simple by the patient, for example the patient with a 6-mm aneurysm of the posterior communicating artery who has already lost three relatives from SAH. Most such patients will ask for treatment regardless of the actual balance of risks and benefits. In all other circumstances the best way to deal with the problem is to spend a lot of time in the outpatient clinic discussing the risks, benefits and most importantly the uncertainties of each option. In this way the patient can come to an informed decision. Small unruptured aneurysms in patients with no other risk factors should probably be left alone; large but easily accessible aneurysms in young patients should probably be offered treatment. In most other situations the physician should discuss in depth the risks, benefits and uncertainties of each strategy. The eventual decision is left to the patient and thus influenced by the patient’s own perception of risk. Whatever the final decision regarding treatment, it is crucial to stress that patients should stop smoking (or

..

..

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 777

15.2 Saccular intracranial aneurysms

should not start smoking) and have their blood pressure checked regularly and hypertension treated, since active smoking and hypertension are strong risk factors for growth and rupture of aneurysms.28,29,56 All patients should be offered advice on risk factor modification regardless of what decision is taken about occlusion of the aneurysm.

15.2.5 Screening for aneurysms The eventual goal of screening is not to detect or to treat an aneurysm, but to increase the number of quality years of life. Therefore, before intracranial vessels are imaged, the risks and benefits of screening should be weighed up. This process includes calculation of the risks of the diagnostic procedures and any treatment, and the gain in life expectancy by preventive treatment of aneurysms detected by screening. But the assessment of risks should also include the amount of anxiety before screening, the reassurance that can be given with a negative result, and the anxiety that can be caused by finding an aneurysm – e.g. if a 3-mm aneurysm is found and is left untreated but followed up regularly,3 or if an unrelated abnormality is found. Invitation for screening in itself does not lead to increased anxiety or depression,4 but actual screening for familial aneurysms is associated with considerable psychosocial effects, both positive and negative.5 Despite these effects, only a small minority of screenees regret being screened,5 and most are motivated for follow-up screening.10 Finally, we should like to emphasize that even screening, repeated screening and preventive treatment cannot prevent all episodes of SAH. In rare instances, aneurysms can develop and rupture within the regular screening interval of 5 years.57 No clinical trials have been done on screening for aneurysms in patients at increased risk, and presumably such trials will not be done in the near future because the follow-up needs to be 20 years or longer in a large sample of patients. As for decisions on the management of unruptured aneurysms, decisions on screening must be made from calculations and assumptions, including assumptions on perceived quality of life. This requires extensive counselling in the outpatient clinic. During this counselling the physician should provide clear and detailed information because relatives substantially underestimate the risk of harbouring an aneurysm and of aneurysmal rupture.4 Imaging techniques for screening Screening has become possible by the advent of non- or minimally invasive imaging techniques for intracranial

..

..

vessels. The test characteristics of CT angiography (CTA) and magnetic resonance angiography (MRA) are similar.58 Both have a substantial risk of missing very small aneurysms,59 but these are the lesions that are typically not treated if detected. MRA can be used to screen for familial aneurysms,60 and has the advantages over CTA that contrast injection is not needed and there is no radiation exposure. In patients who have previously had an aneurysm clipped, the clips cause extensive artifact with MRA,61 and older intracranial aneurysm clips are usually a contraindication – for them CTA is the preferred assessment technique (Fig. 15.3).62 MRA is the best option in patients who have been coiled, because coils produce little artifact with this technique (Fig. 15.4).63 Indications for screening Screening should be considered in individuals with two or more affected first-degree relatives, and in patients with autosomal dominant polycystic kidney disease (ADPKD). Because aneurysms are very rare before the age of 20, screening is typically started after this age. If a first screen is negative, repeated screening should be discussed. In a study of individuals with a positive family history and negative initial screening, we found a new aneurysm at repeat screening 5 years later in 7%.10 If the life expectancy of the patient is short, because of advanced age or comorbidity, screening should not be advised. The maximum age for screening is 60–70 years of age, depending on the individual’s health status. Screening should also be considered in identical twins if SAH has occurred in one of the pair, although no systematic studies in twins have been done. Thus far, at least 13 pairs of identical twins with aneurysms in both have been reported.64 In many twin pairs, the aneurysms are at the same site and SAH tends to occur at around the same age,65 but in some twin pairs SAH occurs decades apart.66 At least two twin pairs have been reported with SAH in one, but no aneurysm in the other at the time of screening;67 such discordant identical twins may have been under-reported. In individuals with only one affected first-degree relative, screening is not very efficient or effective. To prevent one episode of fatal SAH, 300 at-risk people must be screened. The general advice is, therefore, not to screen individuals with only one affected relative; exceptions can be made for siblings with one affected relative who are younger than 40 years and who are highly anxious about SAH, and therefore already have impaired quality of life. Patients with Ehlers-Danlos syndrome type IV are typically advised against screening, because the fragility of the vessel wall substantially increases the risks of

777

9781405127660_4_015.qxd 10/13/07 10:51 AM Page 778

778

Chapter 15 Specific interventions to prevent intracranial haemorrhage

(a)

(b) Fig. 15.3 Follow-up CT angiogram in a patient with a ruptured posterior communicating artery aneurysm treated by neurosurgical clipping. (a) CT angiography shows clip artifacts at the site of the left posterior communicating artery aneurysm, and an aneurysm at the basilar artery tip (arrow). (b) Reconstruction of the CT angiogram confirms the basilar artery tip aneurysm (arrow) and provides better detail of the anatomy. (Note that this reconstruction is a view from above, thus the right side of the patient is on the right.) (Also reproduced in colour in the plate section.)

treatment.68 Because an increased risk of SAH has not been confirmed for patients with neurofibromatosis or Marfan syndrome, there is no indication for screening these patients.

Although patients with no family history and no ADPKD who have survived an episode of SAH are at increased risk for a new episode from a new aneurysm or from recurrence of the treated aneurysm,31 according to a Markov decision model, screening of all such patients cannot be recommended (Wermer et al./ASTRA study group, unpublished data). Although screening prevented almost half of the recurrent SAHs and slightly increased life expectancy, it reduced quality of life and increased costs. In patients with a more than a twofold risk of both aneurysm formation and aneurysm rupture, screening was cost-saving but did not increase quality of life. Only if the risks of aneurysm formation and aneurysm rupture were at least 4.5 times increased, did screening both reduce costs and increase quality of life. Thus, screening seems beneficial only in individuals with previous SAH who have a considerably increased risk of aneurysm formation and of aneurysm rupture. Unfortunately we cannot yet identify who these patients are. Risk factors for recurrent SAH are smoking, multiple aneurysms at the time of the SAH, and young age at time of the initial SAH.31 However, a recurrent SAH is the result of both the formation of a new aneurysm and its rupture. If only one of these risks is increased, screening 5 years after the SAH is not beneficial because even if the risk of aneurysm formation is strongly increased, if the risk of rupture is low, many ‘safe’ aneurysms will be detected and treated at the costs of complications and resources. Also, if the risk of rupture is increased but the risk of formation is low, aneurysms will be present for only a short period of time, and therefore difficult to detect. Because we have insufficient data on these risk factors for formation and rupture separately, we are for the time being reluctant to screen patients after an SAH, and consider it only in patients (especially women) with an initial episode at a very young age and with multiple aneurysms. Advice on counselling Counselling when considering screening is timeconsuming, but commonly two or more siblings or other relatives can be counselled at one consultation, and many relatives prefer to come together. When seeing people with a possible family history of SAH, we draw a family tree to start off with. If the individuals report that relatives have had a stroke or SAH, we try to get as much information as possible. Often medical records of relatives who have died are difficult to retrieve, because they have been destroyed or because – perversely – the hospital will only provide copies of records with written informed consent of the patient, even after death. However, in a study in the Netherlands the information recalled by a relative has a positive predictive value for

..

..

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 779

15.2 Saccular intracranial aneurysms

(a)

(b)

(c)

(d) Fig. 15.4 Follow-up after coiling of an anterior communicating artery aneurysm. (a) CT angiogram reconstruction showing an anterior communicating artery aneurysm (arrow). (Also reproduced in colour in the plate section.) (b) Catheter angiography before coiling confirms the aneurysm (arrow). (c) After coiling, the aneurysm is occluded. (d) MR angiography 6 months after coiling shows a new aneurysm next to the initial aneurysm (arrow). (e) Catheter angiography before repeat coiling confirms the new aneurysm (arrow) adjacent to the coiled lumen of the initial aneurysm (arrowheads). (f) After repeat coiling, the new aneurysm is also occluded (arrowheads).

(f)

(e)

SAH of about 70%, which makes it a sufficiently useful tool for clinical practice.69 During the next part of the consultation, we explain the procedure and the risks of screening – the chance of having an aneurysm, of an aneurysm rupturing, of treatment, and of finding a very small aneurysm that will not be treated but followed up. We also discuss the implications for driving and flying licences and life insurance (which differ by country), and the yield of repeated screening if the initial screening is

..

..

negative. In general our advice is to repeat screening every 5 years, and every 2 or 3 years in the rare families in whom rupture of newly developed aneurysms has occurred within 5 years of screening. Individuals with two or more affected first-degree relatives are mostly very willing to undergo repeated screening. In one study, more than 80% returned 5 years after the initial screening, having been given the advice to undergo repeated screening.10 At the end of the initial consultation, we

779

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 780

780

Chapter 15 Specific interventions to prevent intracranial haemorrhage

leave the decision about screening to the patient: to go for screening; to make a second appointment for further discussion or information (sometimes siblings come together, but without their spouses, and eventually want to make a final decision with their spouses); or to make no further appointment. Audio recording the consultation and giving the tape to the patient might be helpful, but we do not have personal experience with this method of informing individuals. If the screening candidate wishes to be screened, another important decision that has to be made beforehand is what to do about unrelated findings. Some people do not wish to be informed about arachnoid cysts, non-specific white matter lesions of unknown relevance, or other incidental findings. Of course, all of these individuals, including those who do not want to be screened, should be advised against smoking and to have their blood pressure checked regularly.

15.3 Arteriovenous malformations of the brain

The section on incidental intracranial aneurysms described many uncertainties and unanswered questions. For incidental brain arteriovenous malformations (AVMs) the situation is even worse. Uncertainty about the risk of haemorrhage and of treatment is fuelled by the presence of three treatment modalities (and combinations of these modalities) as well as the option of no intervention. But there is some light in the darkness: these uncertainties have prompted the launch of an international trial on the management of unruptured brain AVMs (www.arubastudy.org). This trial will compare the strategy ‘no intervention’ with intervention (if the brain AVM is potentially treatable), the type of intervention being left to the treating team. Clearly, this trial will not answer all the questions, but at least will serve as a starting point for evidence-based treatment advice and for further studies, depending on the outcome.

15.3.1 Frequency and presentation of brain arteriovenous malformations Brain AVMs are much rarer than intracranial aneurysms (section 8.2.4). Their prevalence ranges between 15 and 19 per 100 000 adults,70 the overall detection rate is about 1 per 100 000 adults per year,71–73 and the incidence of haemorrhage from a brain AVM around 0.5 per 100 000 adults per year.71,73 Compared with intracranial

aneurysms, brain AVMs are seldom multiple (around 4% of patients),74,75 and there are no known risk factors. There are no data that modifiable risk factors such as smoking and hypertension increase the risk of brain AVMs, and also there are no sound data to support a familial preponderance of brain AVMs. In a recent systematic review of the literature, only 53 patients from 25 families were identified (van Beijnum, unpublished data). This number of families is so small that it may be chance rather than a common familial factor that is important in the development of brain AVMs. The characteristics of the patients with familial brain AVMs were in general similar to those of patients with sporadic brain AVMs. The only difference was that in the reported familial cases the age at presentation (27 years) was younger than in series of patients with sporadic brain AVMs (mean difference 8 years) (van Beijnum, unpublished data). Although brain AVMs are thought to be congenital lesions,76 rare instances of development during life have been reported.77,78 The good news for patients is that in rare instances AVMs disappear spontaneously.79 In around half of all patients in whom a brain AVM is detected, the reason for detection is an intracranial haemorrhage,80 although recent data suggest that this proportion is declining, at least where access to noninvasive brain imaging is so easy that non-haemorrhagic presentations are more often detected.1 Other reasons for detection are epilepsy, focal neurological deficits and as an incidental finding in patients investigated for other reasons or non-specific complaints. In applicants for military flying duties, brain AVMs are found by chance in around 0.1%.81 Initial presentation with haemorrhage is considered the main determinant of future haemorrhage.82 In contrast to previously held beliefs, recurrent bleeding episodes tend to occur with highest frequency in the first few months after the initial bleed. The rate of subsequent haemorrhage has been reported as high as 33% in the first year after presentation with haemorrhage, decreasing to 11% per annum in subsequent years.83 Over the years, the risk of subsequent haemorrhage tends to become similar to that of patients who presented in other ways.84 The high rate of subsequent bleeding, at least in the initial years after presentation with haemorrhage, favours obliteration of the AVM, if technically feasible. Moreover, although there is not the immediate hurry as there is in patients with a ruptured aneurysm to treat the lesions within the first 1 or 2 days, the aim should be to treat the lesion within the first few weeks after the haemorrhage, if the patient’s condition allows it. The greatest uncertainty arises when an unruptured brain AVM is detected. The remainder of this section will focus on this issue.

..

..

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 781

15.3 Arteriovenous malformations of the brain

15.3.2 Natural history of brain arteriovenous malformations

15.3.3 Treatment options for brain arteriovenous malformations

Little is known of the natural history of brain AVMs. The knowledge we have is biased, and based on wrong outcome measures and erroneous assumptions. First, bias is introduced because natural history has been studied in patients who received no treatment, and the main reason for not treating these patients was that the AVM was (very) difficult to treat. The natural history of brain AVMs that are difficult to treat may well differ from that of patients with ‘easily’ treatable AVMs. Second, most studies on natural history have taken ‘haemorrhage’ as the measure of outcome, and yet haemorrhage may not always lead to disability; indeed, several studies suggest that haemorrhage from brain AVMs is less detrimental than intracerebral haemorrhage in general.85 Also, recurrent haemorrhage from an AVM often does not result in increased morbidity.84,85 Third, follow-up in most studies has been relatively short,86 and the haemorrhage rates found within this short period of follow-up have been recalculated as ‘yearly rates of rupture’, even though haemorrhage rates decrease over time after an initial haemorrhage.84 Studies comparing brain AVM characteristics between patients who presented with haemorrhage and patients who presented with other symptoms or no symptoms at all, concur in finding deep venous drainage, a single draining vein, venous stenosis and high pressure in the feeding artery more often in patients with haemorrhage than in other patients.86 The proof that these same factors determine the occurrence of haemorrhage in the future in patients with unruptured brain AVMs has to come from prospective follow-up studies. In the prospectively followed series of untreated patients from the Columbia AVM Databank exclusive deep venous drainage was indeed associated with increased risk of haemorrhage, other risk factors being increasing age and deep AVM location.82 On the basis of these data, for patients without previous haemorrhage, annual haemorrhage rates have been estimated as 0.9% for patients with no risk factors, 2.4% for those with deep drainage and 3.1% for those with deep location of the haemorrhage.1 However, as pointed out above, extrapolating observed haemorrhage rates to constant future haemorrhage rates is probably unrealistic.

The available options for treatment are: medical management only (e.g. of epilepsy or headaches), surgical excision, endovascular embolization, stereotactic radiotherapy (often described as radiosurgery) and combinations of these treatment modalities. There are no randomized trials or non-randomized comparisons for these strategies, even though they have been used for more than a decade.78,79 We are left with several observational studies of these strategies, and not even a systematic overview. Methodological weaknesses abound. Many of the studies are retrospective, and most of them used occurrence of haemorrhage and not measures of morbidity, disability or dependence as the outcome measurement.78 Such an outcome measurement excludes complications from treatment other than haemorrhage. For stereotactic radiotherapy of brain AVMs in the brainstem, nonhaemorrhagic complications have been reported in up to 7% of patients.80 Even recent studies published in highranking journals have used reduction in haemorrhage rate after the start of the therapy as the outcome measure,81 but such a reduction might very well reflect the natural course; without a randomized comparison with an untreated group, no conclusions can be drawn from these studies. Comparing results between treatment strategies from observational studies is further hampered by heterogeneity of patients and by different treatment goals. Patients so often cannot be compared because they were selected for the particular intervention because of certain characteristics of the AVM (which often makes these patients ineligible for other interventions). Also, surgical intervention usually aims at total excision of the AVM, whereas endovascular treatment may be offered to diminish the size of an AVM so that it becomes treatable by surgery or stereotactic radiotherapy.

The long-term risk of haemorrhage and more importantly of death or disability from brain arteriovenous malformations is unknown. For patients without previous rupture and no risk factors (deep venous drainage) these risks are probably very low.

..

..

Neurosurgical excision The most important determinants for neurosurgical management are the size and location of the malformation, along with the type of venous drainage. For example, a small and superficial AVM in the occipital lobe is much more easily accessible than an anomaly involving almost an entire hemisphere, with feeders from all the major arteries and with extensive drainage into the deep venous system. For these reasons, Spetzler and Martin have proposed a grading system for AVMs that takes account of these three key features (Table 15.7).87 Of course, it is more useful to describe the three features separately than to rely on the sum score alone; in the process of summation important information gets lost.

781

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 782

782

Chapter 15 Specific interventions to prevent intracranial haemorrhage Table 15.7 Grading system for arteriovenous malformations, according to Spetzler and Martin (1986).87 Graded feature Size of arteriovenous malformation small (6 cm) Location non-eloquent area eloquent area Pattern of venous drainage superficial only deep (any part)

Points assigned

1 2 3 0 1 0 1

Grade = [size] + [eloquence] + [venous drainage], i.e. [1, 2 or 3] + [0 or 1] + [0 or 1].

A cautionary note is appropriate about the dangers of making too facile a distinction between ‘eloquent’ and so-called ‘silent’ areas of the brain. Operation in specialized areas of the cerebral cortex may result in easily recognizable deficits such as hemiparesis, aphasia or a visual field defect, but this does not mean that less specialized areas of the brain are functionally unimportant; they play an important role in complex mental processes, which are at least as important as elementary functions. Surgical excision of, say, the right temporal lobe will not result in impairments that are obvious when the patient is making a brief visit to a hospital clinic. But a conversation with the patient’s partner will drastically cure any previous belief on the part of the surgeon that ‘silent’ areas of the brain can be excised without any consequences for the patient’s mood or personality. To illustrate this point with a comparison: if the function of the intact brain is symbolized by a beautiful painting, lesions in ‘eloquent’ brain areas can be compared with holes or blots in the picture, whereas lesions in ‘silent’ areas correspond with darkening of the varnish and fading of the colours. The term ‘silent area of the brain’ should be interpreted as an area with a general rather than a specific function (such as language or movements), not as an area that is redundant. Neurosurgical management is the treatment that has been available for longest and many series have reported on the success rate, management morbidity and case fatality. In a systematic review (published in French), data on 2452 patients were included.88 It is important to

keep in mind that is impossible to segregate the results of the studies according to whether the brain AVM was ruptured or not – another criticism of the treatment literature, in addition to its uncontrolled, non-randomized nature. Morbidity was found in all studies, and varied between 1.5% and 19%; overall postoperative mortality was 3.3%. The risk of complications was related to increasing Spetzler-Martin grade, and to the size and site of the AVM. For AVMs smaller than 3 cm a 4.6% morbidity (from 1.5% to 9.7%) and a zero mortality were reported. Total excision of the AVM was reported in 97% of patients on average, with a range from 91% to 100%. Thus, neurosurgical intervention is a successful strategy, at least for patients selected for this treatment, and has a reasonably low risk of complications for patients with small AVMs. Series published after this review reported similar results.89 Endovascular embolization The aim of endovascular therapy is to obliterate the feeding arteries and the vessels at the site of the nidus by use of embolic agents, such as N-butylcyanoacrylate, polyvinyl alcohol particles, detachable coils, or Onyx liquid polymer. The results from the different agents in terms of occlusion of the AVM do not seem to differ significantly.90 Endovascular embolization is most often used to reduce the size of the AVM to make it amenable for neurosurgical treatment, or for stereotactic radiotherapy. Often, multiple procedures are necessary to reach this goal. In those series where surgical excision is preceded by endovascular treatment, the morbidity related to this pre-surgical embolization ranges from 4% to 8.9%.88,91 In a study on combined endovascular and neurosurgical intervention with prospective assessment of complications by an independent physician, disabling treatment-related complications occurred in 5% (95% CI 1–9%) of the patients, and non-disabling new deficits in another 42% (95% CI 33–51%), while none of the patients died.92 Risk factors for complications were non-haemorrhagic presentation, deep venous drainage, AVM location in an ‘eloquent’ brain region and large AVM size. As is the case for the series describing the neurosurgical treatment of AVMs, it is difficult to disentangle the results of endovascular treatment according to whether the brain AVM had ruptured or not. Endovascular embolization as the only therapy results in complete obliteration of the AVM in only 10–40% of the patients, depending on their selection.93–95 In large series from experienced centres reporting on complications from endovascular therapy in general (including embolization before neurosurgery or stereotactic

..

..

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 783

15.3 Arteriovenous malformations of the brain

radiotherapy) case fatality varies from 1% to 3%, and morbidity from 2% to 6%.94,95

safer and is probably less effective than surgical resection, certainly in the short term.

Stereotactic radiotherapy (radiosurgery)

15.3.4 Management of patients with unruptured arteriovenous malformations

Focused radiation is an increasingly important option in the management of brain AVMs that are considered inaccessible to surgical treatment, i.e. lesions in the deep regions of a cerebral hemisphere, in the brainstem or in important primary projection areas such as for language or for motor control of the dominant hand. But lesions of more than 3.0–3.5 cm in size are not suitable for stereotactic radiotherapy, because the dose to normal tissues would be excessive, so for larger brain AVMs embolization is often used to reduce their size first. The rationale for radiation treatment is that it damages and eventually occludes the blood vessels constituting the AVM.96 This takes place in the long term, after months or years, which is an important drawback, because in the meantime there is still a risk of haemorrhage.97 This risk of haemorrhage may even be increased within this ‘obliteration period’,98 although this suggestion is not confirmed in all relevant studies.99 In a prospective study, presentation of the AVM with haemorrhage was a risk factor for haemorrhage after stereotactic radiotherapy.100 In this study haemorrhage occurred in 3.8% (95% CI 2.0–6.7%) of patients with no haemorrhage and in 8.6% (95% CI 6.1–11.6%) of patients with a haemorrhage before starting the therapy. Haemorrhage resulted in death in 7% and morbidity in almost 50% of patients. Patients with angiographic disappearance of the AVM after stereotactic radiotherapy are considered to be ‘cured’, but even in them new haemorrhages can sometimes occur. In a series of 236 patients with angiographic evidence of obliteration of the AVM after stereotactic radiotherapy, the cumulative risk of haemorrhage for the initial 10 years after treatment was 2.2%.101 Other complications from stereotactic radiotherapy are radiation injury of the brain parenchyma (6.5%), radiation-induced cranial nerve palsies (1%) and new or worsened seizures (1%).102 In one out of four patients these complications were disabling and 2% of patients died from the complications. Unfortunately, the study did not describe risk factors for developing these complications. In half the patients, the symptoms resolved over time.102 Radiation injury of the brain parenchyma may result in the formation of cysts that can appear as long as 2–7 years later.103 These cysts may need to be treated with a permanent shunt or with total excision.103 In summary, if any intervention at all is needed for an AVM, the technique of focused radiation is not much

..

..

Ideally, patients with unruptured brain AVMs for whom treatment is considered should be entered into a randomized controlled trial. Currently, one such trial is running (www.arubastudy.org) and, if possible, patients should be referred to a centre that is participating in this trial (or future trials). If this is not possible, the pros and cons of all treatment options should be discussed with the patient, and based upon these uncertainties a decision has to be made. An important factor in this decision is the patient’s psychological attitude towards living with an unoperated lesion in the brain that may unexpectedly bleed. In general, knowledge of having such a lesion results in a decreased quality of life,3 but some patients show admirable sangfroid in coping with this knowledge. For others, however, their life may turn out to be so dominated by the perceived danger that they insist on intervention even when the balance of risks would seem to argue against such a course of action. If treatment is opted for, it seems that open surgery is best for superficial and small AVMs. For larger lesions, endovascular embolization preceding neurosurgery or stereotactic radiotherapy seems a reasonable approach. Stereotactic radiotherapy is probably a good option for AVMs that are inaccessible to surgery such as small and deep lesions, fed by small blood vessels. Given the absence of a clear familial preponderance of brain AVMs, there is no indication for screening relatives of patients with an AVM, probably not even if several relatives are affected. In some relatives brain AVMs have been detected by means of screening,104–106 but in other families screening did not yield any new AVM.107–109 Screening of close relatives of all patients with an AVM is bound to have a low yield. Furthermore, this would be very expensive. Another reason to withhold screening is that the clinical consequences of finding an asymptomatic AVM are still uncertain. Migraine is sometimes suggested as an indication for screening for an AVM, but in patients with migraine the chance of finding one is well below 1%.110 Moreover, in view of the dilemmas with regard to invasive treatment, it is far from certain that a patient with migraine benefits from the knowledge that there is an underlying AVM. Another disadvantage of ordering imaging studies in all migraineurs with stereotyped attacks is the cost, discomfort and risk of the negative investigations in the vast majority.

783

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 784

784

Chapter 15 Specific interventions to prevent intracranial haemorrhage

References 16 1 Stapf C, Mohr JP, Choi JH, Hartmann A, Mast H. Invasive treatment of unruptured brain arteriovenous malformations is experimental therapy. Curr Opin Neurol 2006; 19(1):63–8. 2 Wiebers DO, Whisnant JP, Huston J, III, Meissner I, Brown RDJ, Piepgras DG et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362(9378):103–10. 3 Van der Schaaf IC, Brilstra EH, Rinkel GJE, Bossuyt PMM, van Gijn J. Quality of life, anxiety, and depression in patients with an untreated intracranial aneurysm or arteriovenous malformation. Stroke 2002; 33(2):440–3. 4 Bossuyt PM, Raaymakers TWM, Bonsel GJ, Rinkel GJE. Screening families for intracranial aneurysms: Anxiety, perceived risk, and informed choice. Prev Med 2005; 41:795–9. 5 Wermer MJ, van der Schaaf IC, Van Nunen P, Bossuyt PM, Anderson CS, Rinkel GJE. Psychosocial impact of screening for intracranial aneurysms in relatives with familial subarachnoid hemorrhage. Stroke 2005; 36(4):836–40. 6 Rinkel GJE, Djibuti M, Algra A, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms: a systematic review. Stroke 1998; 29(1):251–6. 7 Menghini VV, Brown RD, Jr., Sicks JD, O’Fallon WM, Wiebers DO. Incidence and prevalence of intracranial aneurysms and hemorrhage in Olmsted County, Minnesota, 1965 to 1995. Neurology 1998; 51(2):405–11. 8 Bromberg JEC, Rinkel GJE, Algra A, Greebe P, van Duyn CM, Hasan D et al. Subarachnoid haemorrhage in first and second degree relatives of patients with subarachnoid haemorrhage. Br Med J 1995; 311:288–9. 9 Raaymakers TWM, MARS study group. Aneurysms in relatives of patients with subarachnoid hemorrhage. Frequency and risk factors. Neurology 1999; 53(5):982–8. 10 Wermer MJ, Rinkel GJE, van Gijn J. Repeated screening for intracranial aneurysms in familial subarachnoid hemorrhage. Stroke 2003; 34:2788–91. 11 Ruigrok YM, Rinkel GJE, Algra A, Raaymakers TWM, van Gijn J. Characteristics of intracranial aneurysms in patients with familial subarachnoid hemorrhage. Neurology 2004; 62:891–4. 12 Bromberg JEC, Rinkel GJE, Algra A, van Duyn CM, Greebe P, Ramos LMP et al. Familial subarachnoid hemorrhage: distinctive features and patterns of inheritance. Ann Neurol 1995; 38(6):929–34. 13 Ruigrok YM, Rinkel GJE, Wijmenga C, van Gijn J. Anticipation and phenotype in familial intracranial aneurysms. J Neurol Neurosurg Psychiatry 2004; 75:1436–42. 14 Struycken PM, Pals G, Limburg M, Pronk JC, Wijmenga C, Pearson PL et al. Anticipation in familial intracranial aneurysms in consecutive generations. Eur J Hum Genet 2003; 11:737–43. 15 Belz MM, Hughes RL, Kaehny WD, Johnson AM, FickBrosnahan GM, Earnest MP et al. Familial clustering of

17

18

19

20

21

22

23

24

25

26

27

28

29

ruptured intracranial aneurysms in autosomal dominant polycystic kidney disease. Am J Kidney Dis 2001; 38(4):770–6. Schievink WI, Torres VE, Piepgras DG, Wiebers DO. Saccular intracranial aneurysms in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 1992; 3(1):88–95. Gieteling EW, Rinkel GJE. Characteristics of intracranial aneurysms and subarachnoid haemorrhage in patients with polycystic kidney disease. J Neurol 2003; 250(4):418–23. Huston J, III, Torres VE, Sulivan PP, Offord KP, Wiebers DO. Value of magnetic resonance angiography for the detection of intracranial aneurysms in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 1993; 3:1871–7. Rossetti S, Chauveau D, Kubly V, Slezak JM, Saggar-Malik AK, Pei Y et al. Association of mutation position in polycystic kidney disease 1 (PKD1) gene and development of a vascular phenotype. Lancet 2003; 361:2196–201. Chauveau D, Pirson Y, Verellen-Dumoulin C, Macnicol A, Gonzalo A, Grunfeld JP. Intracranial aneurysms in autosomal dominant polycystic kidney disease. Kidney Int 1994; 45(4):1140–6. Chauveau D, Sirieix ME, Schillinger F, Legendre C, Grünfeld JP. Recurrent rupture of intracranial aneurysms in autosomal dominant polycystic kidney disease. Br Med J 1990; 301:966–7. Nakajima F, Shibahara N, Arai M, Gohji K, Ueda H, Katsuoka Y. Intracranial aneurysms and autosomal dominant polycystic kidney disease: followup study by magnetic resonance angiography. J Urol 2000; 164(2):311–13. Belz MM, Fick-Brosnahan GM, Hughes RL, Rubinstein D, Chapman AB, Johnson AM et al. Recurrence of intracranial aneurysms in autosomal-dominant polycystic kidney disease. Kidney Int 2003; 63:1824–30. Gibbs GF, Huston J, III, Qian Q, Kubly V, Harris PC, Brown RD, Jr. et al. Follow-up of intracranial aneurysms in autosomal-dominant polycystic kidney disease. Kidney Int 2004; 65:1621–7. Akyuz M, Tuncer R, Yilmaz S, Sindel T. Angiographic follow-up after surgical treatment of intracranial aneurysms. Acta Neurochir (Wien) 2004; 146:245–50. David CA, Vishteh AG, Spetzler RF, Lemole M, Lawton MT, Partovi S. Late angiographic follow-up review of surgically treated aneurysms. J Neurosurg 1999; 91(3):396–401. Tsutsumi K, Ueki K, Morita A, Usui M, Kirino T. Risk of aneurysm recurrence in patients with clipped cerebral aneurysms. Results of long-term follow-up angiography. Stroke 2001; 32(5):1191–4. Juvela S, Poussa K, Porras M. Factors affecting formation and growth of intracranial aneurysms: a long-term followup study. Stroke 2001; 32(2):485–91. Wermer MJ, van der Schaaf IC, Velthuis BK, Algra A, Buskens E, Rinkel GJE. Follow-up screening after subarachnoid haemorrhage: frequency and determinants

..

..

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 785

References

30

31

32

33

34

35

36

37

38

39

40

41 42

43

44

45

..

..

of new aneurysms and enlargement of existing aneurysms. Brain 2005; 128:2421–9. Tsutsumi K, Ueki K, Usui M, Kwak S, Kirino T. Risk of recurrent subarachnoid hemorrhage after complete obliteration of cerebral aneurysms. Stroke 1998; 29:2511–13. Wermer MJ, Greebe P, Algra A, Rinkel GJE. Incidence of recurrent subarachnoid hemorrhage after clipping for ruptured intracranial aneurysms. Stroke 2005; 36(11):2394–9. Molyneux AJ, Kerr RS, Yu LM, Clarke M, Sneade M, Yarnold JA et al. International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet 2005; 366(9488):809–17. The International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms: risk of rupture and risks of surgical intervention. N Engl J Med 1998; 339:1725–33. Clarke G, Mendelow AD, Mitchell P. Predicting the risk of rupture of intracranial aneurysms based on anatomical location. Acta Neurochir (Wien) 2005; 147:259–63. Kailasnath P, Chaloupka JC, Dickey PS. A multiplicative statistical model predicts the size distribution of unruptured intracranial aneurysms. Neurol Res 1998; 20(5):421–6. Mitchell P, Jakubowski J. Estimate of the maximum time interval between formation of cerebral aneurysm and rupture. J Neurol Neurosurg Psychiatry 2000; 69(6):760–7. Tsutsumi K, Ueki K, Usui M, Kwak S, Kirino T. Risk of subarachnoid hemorrhage after surgical treatment of unruptured cerebral aneurysms. Stroke 1999; 30(6):1181–4. Wermer MJ, van der Schaaf IC, Algra A, Rinkel GJE. Risk of rupture of unruptured intracranial aneurysms in relation to patient and aneurysm characteristics: an updated metaanalysis. Stroke. 2007 Apr; 38(4):1404–10. Jane JA, Kassell NF, Torner JC, Winn HR. The natural history of aneurysms and arteriovenous malformations. J Neurosurg 1985; 62:321–3. Juvela S, Porras M, Heiskanen O. Natural history of unruptured intracranial aneurysms: a long-term follow-up study. J Neurosurg 1993; 79(2):174–82. Rose G. The Strategy of Preventive Medicine. Oxford: Oxford Medical Publications, 1992. Raaymakers TWM, Rinkel GJE, Limburg M, Algra A. Mortality and morbidity of surgery for unruptured intracranial aneurysms: a meta-analysis. Stroke 1998; 29(8):1531–8. Solomon RA, Fink ME, Pile-Spellman J. Surgical management of unruptured intracranial aneurysms. J Neurosurg 1994; 80:440–6. Chyatte D, Porterfield R. Functional outcome after repair of unruptured intracranial aneurysms. J Neurosurg 2001; 94(3):417–21. Matsumoto K, Akagi K, Abekura M, Nakajima Y, Yoshiminie T. Investigation of the surgically treated and

46

47

48

49

50

51

52

53

54

55

56

57

58

59

untreated unruptured cerebral aneurysms of the anterior circulation. Surg Neurol 2003; 60:516–22. Solomon RA, Mayer SA, Tarmey JJ. Relationship between the volume of craniotomies for cerebral aneurysm performed at New York state hospitals and in-hospital mortality. Stroke 1996; 27(1):13–17. Lanterna LA, Tredici G, Dimitrov BD, Biroli F. Treatment of unruptured cerebral aneurysms by embolization with guglielmi detachable coils: case-fatality, morbidity, and effectiveness in preventing bleeding – a systematic review of the literature. Neurosurgery 2004; 55(4):767–75. Van Rooij WJ, de Gast A, Sluzewski M, Nijssen PC, Beute GN. Coiling of truly incidental intracranial aneurysms. Am J Neuroradiol 2006; 27(2):293–6. Rothwell PM, Slattery J, Warlow CP. A systematic review of the risks of stroke and death due to endarterectomy for symptomatic carotid stenosis. Stroke 1996; 27(2):260–5. Lee T, Baytion M, Sciacca R, Mohr JP, Pile-Spellman J. Aggregate analysis of the literature for unruptured intracranial aneurysm treatment. Am J Neuroradiol 2005; 26(8):1902–8. Grunwald IQ, Papanagiotou P, Politi M, Struffert T, Roth C, Reith W. Endovascular treatment of unruptured intracranial aneurysms: occurrence of thromboembolic events. Neurosurgery 2006; 58(4):612–18. Brilstra EH, Rinkel GJE, van der Graaf Y, Sluzewski M, Groen RJ, Lo RT et al. Quality of life after treatment of unruptured intracranial aneurysms by neurosurgical clipping or by embolisation with coils. a prospective, observational study. Cerebrovasc Dis 2004; 17:44–52. Barker FG. Age-dependent differences in short-term outcome after surgical or endovascular treatment of unruptured intracranial aneurysms in the United States, 1996–2000. Neurosurg 2004; 54:18–30. Johnston SC, Wilson CB, Halbach VV, Higashida RT, Dowd CF, McDermott MW et al. Endovascular and surgical treatment of unruptured cerebral aneurysms: comparison of risks. Ann Neurol 2000; 48:11–19. Wermer MJ, van der Schaaf IC, Velthuis BK, Majoie CB, Albrecht KW, Rinkel GJE. Yield of short-term follow-up CT/MR angiography for small aneurysms detected at screening. Stroke 2006; 37(2):414–18. Juvela S, Porras M, Poussa K. Natural history of unruptured intracranial aneurysms: probability of and risk factors for aneurysm rupture. J Neurosurg 2000; 93:379–87. Schievink WI, Limburg M, Dreissen JJ, Peeters FL, ter Berg HW. Screening for unruptured familial intracranial aneurysms: subarachnoid hemorrhage 2 years after angiography negative for aneurysms. Neurosurgery 1991; 29:434–7. White PM, Teasdale EM, Wardlaw JM, Easton V. Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort. Radiology 2001; 219(3):739–49. Wardlaw JM, White PM. The detection and management of unruptured intracranial aneurysms. Brain 2000; 123(Pt 2):205–21.

785

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 786

786

Chapter 15 Specific interventions to prevent intracranial haemorrhage

60 Raaymakers TWM, Buys PC, Verbeeten B, Jr., Ramos LMP, Witkamp TD, Hulsmans FJ et al. MR angiography as a screening tool for intracranial aneurysms: feasibility, test characteristics, and interobserver agreement. Am J Roentgenol 1999; 173(6):1469–75. 61 Gonner F, Lovblad KO, Heid O, Remonda L, Guzman R, Barth A et al. Magnetic resonance angiography with ultrashort echo times reduces the artefact of aneurysm clips. Neuroradiology 2002; 44:755–8. 62 Van der Schaaf IC, Velthuis BK, Wermer MJ, Frenkel NJ, Majoie CB, Witkamp TD et al. Multislice computed tomography angiography screening for new aneurysms in patients with previously clip-treated intracranial aneurysms: Feasibility, positive predictive value, and interobserver agreement. J Neurosurg 2006; 105(5):682–8. 63 Majoie CB, Sprengers ME, Van Rooij WJ, Lavini C, Sluzewski M, van Rijn JC et al. MR angiography at 3T versus digital subtraction angiography in the follow-up of intracranial aneurysms treated with detachable coils. Am J Neuroradiol 2005; 26(6):1349–56. 64 Ohno S, Ikeda Y, Onitsuka T, Nakajima S, Uchino H, Haraoka J et al. [Cerebral aneurysms in identical twins]. No Shinkei Geka 2004; 32:875–9. 65 Puchner MJ, Lohmann F, Valdueza JM, Siepmann G, Freckmann N. Monozygotic twins not identical with respect to the existence of intracranial aneurysms: a case report. Surg Neurol 1994; 41:284–9. 66 Hager P, Steiger HJ. Identical cerebral aneurysms in siblings: report of two families. J Clin Neurosci 2004; 11:80–4. 67 Astradsson A, Astrup J. An intracranial aneurysm in one identical twin, but no aneurysm in the other. Br J Neurosurg 2001; 15:168–71. 68 Schievink WI, Link MJ, Piepgras DG, Spetzler RF. Intracranial aneurysm surgery in Ehlers-Danlos syndrome type IV. Neurosurg 2002; 51(3):607–13. 69 Bromberg JEC, Rinkel GJE, Algra A, Greebe P, Beldman T, van Gijn J. Validation of family history in subarachnoid hemorrhage. Stroke 1996; 27(4):630–2. 70 Al Shahi R, Fang JS, Lewis SC, Warlow CP. Prevalence of adults with brain arteriovenous malformations: a community based study in Scotland using capturerecapture analysis. J Neurol Neurosurg Psychiatry 2002; 73(5):547–51. 71 Stapf C, Labovitz DL, Sciacca RR, Mast H, Mohr JP, Sacco RL. Incidence of adult brain arteriovenous malformation hemorrhage in a prospective populationbased stroke survey. Cerebrovasc Dis 2002; 13(1):43–6. 72 Al Shahi R, Bhattacharya JJ, Currie DG, Papanastassiou V, Ritchie V, Roberts RC et al. Prospective, population-based detection of intracranial vascular malformations in adults: the Scottish Intracranial Vascular Malformation Study (SIVMS). Stroke 2003; 34(5):1163–9. 73 Stapf C, Mast H, Sciacca RR, Berenstein A, Nelson PK, Gobin YP et al. The New York Islands AVM Study: design, study progress, and initial results. Stroke 2003; 34(5):e29–e33.

74 Salcman M, Scholtz H, Numaguchi Y. Multiple intracerebral arteriovenous malformations: report of three cases and review of the literature. Surg Neurol 1992; 38(2):121–8. 75 Willinsky RA, Lasjaunias P, TerBrugge K, Burrows P. Multiple cerebral arteriovenous malformations (AVMs): review of our experience from 203 patients with cerebral vascular lesions. Neuroradiology 1990; 32(3):207–10. 76 Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet 2002; 359(9309):863–73. 77 Bulsara KR, Alexander MJ, Villavicencio AT, Graffagnino C. De novo cerebral arteriovenous malformation: case report. Neurosurg 2002; 50(5):1137–41. 78 Gonzalez LF, Bristol RE, Porter RW, Spetzler RF. De novo presentation of an arteriovenous malformation. Case report and review of the literature. J Neurosurg 2005; 102(4):726–9. 79 Buis DR, van den BR, Lycklama G, van der Worp HB, Dirven CM, Vandertop WP. Spontaneous regression of brain arteriovenous malformations: a clinical study and a systematic review of the literature. J Neurol 2004; 251(11):1375–82. 80 Brown RD, Wiebers DO, Torner JC, O’Fallon WM. Incidence and prevalence of intracranial vascular malformation in Omsted County, Minnesota, 1965 to 1992. Neurology 1996; 46:949–52. 81 Weber F, Knopf H. Incidental findings in magnetic resonance imaging of the brains of healthy young men. J Neurol Sci 2006; 240(1–2):81–4. 82 Stapf C, Mast H, Sciacca RR, Choi JH, Khaw AV, Connolly ES et al. Predictors of hemorrhage in patients with untreated brain arteriovenous malformation. Neurology 2006; 66(9):1350–5. 83 Mast H, Young WL, Koennecke HC, Sciacca RR, Osipov A, Pile-Spellman J et al. Risk of spontaneous haemorrhage after diagnosis of cerebral arteriovenous malformation. Lancet 1997; 350(9084):1065–8. 84 Halim AX, Johnston SC, Singh V, McCulloch CE, Bennett JP, Achrol AS et al. Longitudinal risk of intracranial hemorrhage in patients with arteriovenous malformation of the brain within a defined population. Stroke 2004; 35:1697–702. 85 Choi JH, Mast H, Sciacca RR, Hartmann A, Khaw AV, Mohr JP et al. Clinical outcome after first and recurrent hemorrhage in patients with untreated brain arteriovenous malformation. Stroke 2006; 37:1243–7. 86 Al Shahi R, Warlow C. A systematic review of the frequency and prognosis of arteriovenous malformations of the brain in adults. Brain 2001; 124(10):1900–26. 87 Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg 1986; 65(4):476–83. 88 Castel JP, Kantor G. [Postoperative morbidity and mortality after microsurgical exclusion of cerebral arteriovenous malformations. Current data and analysis of recent literature]. Neurochirurgie 2001; 47(2–3 Pt 2):369–83. 89 Morgan MK, Rochford AM, Tsahtsarlis A, Little N, Faulder KC. Surgical risks associated with the management of

..

..

9781405127660_4_015.qxd 10/13/07 10:52 AM Page 787

References

90

91

92

93

94

95

96

97

98 99

..

..

Grade I and II brain arteriovenous malformations. Neurosurgery 2004; 54(4):832–7. Al Shahi R, Warlow C. Arteriovenous malformations of the brain: ready to randomise? J Neurol Neurosurg Psychiatry 2005; 76(10):1327–9. Taylor CL, Dutton K, Rappard G, Pride GL, Replogle R, Purdy PD et al. Complications of preoperative embolization of cerebral arteriovenous malformations. J Neurosurg 2004; 100(5):810–12. Hartmann A, Mast H, Mohr JP, Pile-Spellman J, Connolly ES, Sciacca RR et al. Determinants of staged endovascular and surgical treatment outcome of brain arteriovenous malformations. Stroke 2005; 36(11):2431–5. Wikholm G, Lundqvist C, Svendsen P. The Goteborg cohort of embolized cerebral arteriovenous malformations: a 6-year follow-up. Neurosurgery 2001; 49(4):799–805. Hartmann A, Pile-Spellman J, Stapf C, Sciacca RR, Faulstich A, Mohr JP et al. Risk of endovascular treatment of brain arteriovenous malformations. Stroke 2002; 33(7):1816–20. Haw CS, TerBrugge K, Willinsky R, Tomlinson G. Complications of embolization of arteriovenous malformations of the brain. J Neurosurg 2006; 104(2):226–32. Ogilvy CS. Radiation therapy for arteriovenous malformations: a review. Neurosurgery 1990; 26(5):725–35. Friedman WA, Bova FJ, Bollampally S, Bradshaw P. Analysis of factors predictive of success or complications in arteriovenous malformation radiosurgery. Neurosurgery 2003; 52(2):296–307. Pollock BE. Stereotactic radiosurgery for arteriovenous malformations. Neurosurg Clin N Am 1999; 10(2):281–90. Maruyama K, Kawahara N, Shin M, Tago M, Kishimoto J, Kurita H et al. The risk of hemorrhage after radiosurgery for cerebral arteriovenous malformations. N Engl J Med 2005; 352(2):146–53.

100 Nataf F, Ghossoub M, Schlienger M, Moussa R, Meder JF, Roux FX. Bleeding after radiosurgery for cerebral arteriovenous malformations. Neurosurgery 2004; 55(2):298–305. 101 Shin M, Kawahara N, Maruyama K, Tago M, Ueki K, Kirino T. Risk of hemorrhage from an arteriovenous malformation confirmed to have been obliterated on angiography after stereotactic radiosurgery. J Neurosurg 2005; 102(5):842–6. 102 Flickinger JC, Kondziolka D, Lunsford LD, Pollock BE, Yamamoto M, Gorman DA et al. A multi-institutional analysis of complication outcomes after arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 1999; 44(1):67–74. 103 Pollock BE, Brown RD, Jr. Management of cysts arising after radiosurgery to treat intracranial arteriovenous malformations. Neurosurgery 2001; 49(2):259–64. 104 Boyd MC, Steinbok P, Paty DW. Familial arteriovenous malformations. Report of four cases in one family. J Neurosurg 1985; 62(4):597–9. 105 Larsen PD, Hellbusch LC, Lefkowitz DM, Schaefer GB. Cerebral arteriovenous malformation in three successive generations. Pediatr Neurol 1997; 17(1):74–6. 106 Snead OC, III, Acker JD, Morawetz R. Familial arteriovenous malformation. Ann Neurol 1979; 5(6):585–7. 107 Herzig R, Burval S, Vladyka V, Janouskova L, Krivanek P, Krupka B et al. Familial occurrence of cerebral arteriovenous malformation in sisters: case report and review of the literature. Eur J Neurol 2000; 7(1):95–100. 108 Kamiryo T, Nelson PK, Bose A, Zalzal P, Jafar JJ. Familial arteriovenous malformations in siblings. Surg Neurol 2000; 53(3):255–9. 109 Goto S, Abe M, Tsuji T, Tabuchi K. Familial arteriovenous malformations of the brain: two case reports. Neurol Med Chir (Tokyo) 1994; 34(4):221–4. 110 Frishberg BM. Neuroimaging in presumed primary headache disorders. Semin Neurol 1997; 17(4):373–82.

787

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 789

16

Preventing recurrent stroke and other serious vascular events 16.1

General approach to preventing recurrent stroke and other serious vascular events

16.2

Prognosis and prediction of future vascular events

16.3

Pharmacological blood pressure reduction

16.4

Pharmacological cholesterol reduction

16.5

Antiplatelet drugs

16.6

Anticoagulants

16.7

Lifestyle modification

16.8

Dietary supplements: B vitamins and antioxidants

16.9

Management of diabetes mellitus and glucose intolerance

789

790

801

809

816

830 841

16.10 Treatment of specific underlying causes

847

847

16.11 Endarterectomy for symptomatic carotid stenosis 16.12 Endarterectomy for asymptomatic carotid stenosis 16.13 Carotid angioplasty and stenting

845

849 870

873

16.14 Carotid endarterectomy before, during or after coronary artery surgery? 16.15 Extra-to-intracranial bypass surgery

16.16 Surgery and angioplasty for vertebrobasilar ischaemia 16.17 Other surgical procedures

16.1.1 What are we trying to prevent? This chapter deals with the long-term prevention of recurrent vascular events among patients with a prior ischaemic stroke, transient ischaemic attack (TIA) or intracerebral haemorrhage. Specific treatment of acute ischaemic stroke, intracerebral haemorrhage and subarachnoid haemorhage, treatment of the rare causes of

Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

877

878

16.18 Putting secondary prevention into practice

16.1 General approach to preventing recurrent stroke and other serious vascular events

876

877

878

stroke, and specific strategies for the prevention of intracranial haemorrhage are covered in Chapters 7, 12–15. Patients with stroke and TIA are not just at high risk of stroke, but also of myocardial infarction, lower limb ischaemia, and cognitive impairment due to vascular disease of the brain. Doctors caring for patients with stroke therefore need to pay attention to the heart and peripheral vascular tree not just in their clinical examination but also when deciding on treatments for secondary prevention. Indeed, many trials and systematic reviews of drug and lifestyle interventions for prevention after stroke or TIA have focused primarily on the prevention of ‘serious vascular events’, usually defined as the combination of stroke, myocardial infarction and death from vascular causes. We too will focus largely on this outcome cluster, since it captures most serious recurrent vascular events from which patients with stroke or TIA are at risk. However, it omits the serious consequences of lower limb ischaemia (such as gangrene or amputation) and the cognitive impairment and 789

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 790

790

Chapter 16 Preventing recurrent stroke and other serious vascular events

consequent disability resulting from vascular dementia (section 11.29.2). Reducing the risk of these additional outcomes is clearly also very important, and a number of the more recent trials of interventions that primarily aim to prevent recurrent stroke and other serious vascular events (such as blood pressure reduction and cholesterol reduction) have sought or are seeking to determine whether the intervention also reduces the frequency and severity of cognitive impairment due to cerebrovascular disease (see relevant section on each of the specific interventions). Although primary prevention of stroke is not the main subject of this chapter, we have included here a discussion of the evidence for using various interventions for people with no prior history of stroke or TIA, and sometimes for those with no history of vascular disease at all, since it makes sense to discuss this alongside the evidence for these same interventions in secondary prevention.

16.1.2 Evidence from randomized trials and systematic reviews As in the other chapters of this book, we will try to base our treatment recommendations on the best available evidence, using information from randomized controlled trials (RCTs) and systematic reviews of these wherever possible (section 12.2.1). Sometimes, however, no relevant RCTs are available and we have to make treatment decisions and recommendations by extrapolating the evidence from RCTs conducted in other patient groups (but not those with a stroke or TIA), or by relying on other sources such as observational studies, clinical reasoning, pathophysiology, common sense or anecdotal experience.

16.1.3 Relative and absolute effects of treatment It is important to emphasize that the decision whether or not to use a particular preventive treatment should be based on the absolute rather than relative benefit of that treatment (and, likewise, on the absolute rather than relative increase in any serious risks) (section 12.2.1 and Table 12.5). For example, for patients whose absolute risk of stroke in the next year is 2%, a treatment which reduces the relative risk of stroke by 50% will reduce the absolute risk to 1%, yielding an absolute benefit of 1%, or 10 strokes avoided per 1000 patients treated, so 100 (1000/10) patients have to be treated for 1 year to prevent one stroke. On the other hand, if the same treatment is given to patients whose untreated risk of stroke in the next year is five times higher, 10%, the same relative risk reduction will reduce the absolute risk to 5%, an absolute reduction of 5%, or 50 strokes avoided per 1000

patients treated. Only 20 (1000/50) of these high-risk patients have to be treated for 1 year to prevent one stroke. The relative effect of a particular treatment on a particular outcome is usually consistent across subgoups of patients at different levels of untreated absolute risk. This is not invariably the case, particularly when the outcome under consideration is a composite one, made up of several different component outcomes, and especially if the risk of some of these is increased and of others is decreased by treatment. It usually holds true, however, if the benefits and risks of treatment are considered separately.1 Thus, if we know the relative effect of a treatment on an outcome, we can calculate the absolute benefit for any particular patient, so long as we know (or can estimate) what their untreated risk of the outcome is. Similarly, if we consider adverse effects of treatment, if we know both the relative increased risk from treatment and our patient’s untreated risk, we can calculate the absolute excess risk of any important adverse effect. It is the balance of absolute benefit on the one hand, and absolute risk of adverse effects on the other, that will determine the absolute net benefit of treatment in a particular patient or group of patients. The size of this net benefit will then help to determine whether or not taking the treatment is worthwhile. To make sensible treatment decisions, it is therefore essential to know something of the likely prognosis for our patients (section 16.2). Other factors, such as the costs of treatment, whether or not the treatment is available, and the patient’s attitudes and preferences will also contribute to the decision-making process.

16.2 Prognosis and prediction of future vascular events

The prognosis for early death and disability after stroke has been discussed in section 10.2. This section describes the early and long-term prognosis of patients with transient ischaemic attacks (TIAs) and stroke for important future vascular events that might be prevented by appropriate treatment strategies, particularly recurrent stroke but also myocardial infarction and death. As highlighted in section 10.2, much of the information we have about the prognosis of these patients comes from following large cohorts of patients over time. The most informative studies of prognosis are those adhering to the criteria set out in Table 10.2. But, because the prognosis of an individual patient may vary considerably from the average prognosis of all patients (because of their different

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 791

16.2 Prognosis and prediction of future vascular events

25 Minor stroke TIA 20 Risk of stroke (%)

underlying causes of the TIA/stroke and different comorbidities, etc. – see below), we also discuss ways in which the risk of particular individual patients might be more accurately predicted, allowing us to answer the question ‘What is the risk of another important vascular event for this patient?’ and thereby to be able to identify the appropriate goals and treatments for an individual patient.

15 10 5

16.2.1 Prognosis early after transient ischaemic attack and mild ischaemic stroke High early risk of stroke after a transient ischaemic attack or mild ischaemic stroke About 20% of patients with stroke have a preceding transient ischaemic attack (TIA), and a similar proportion of major ischaemic strokes are preceded by minor ones. These ‘warning’ events provide an opportunity for prevention, and evidence is accumulating for the beneficial effects of various interventions very early after a TIA or mild (i.e. non-disabling) ischaemic stroke (sections 16.3 to 16.6 and chapter 12). Patients with TIA and mild ischaemic stroke are not only similar in terms of age, sex and prevalence of coexistent vascular diseases and risk factors, but they also have a similar (albeit not identical) early and long-term prognosis for recurrent stroke and other vascular events and will therefore be considered together.2–6 Recent studies have shown that the very early risk of stroke after a TIA or mild ischaemic stroke is much higher than previously thought.7,8 Two emergencydepartment-based studies of TIA in California and Alberta reported cumulative risks of stroke at 30 days of about 7% and at 90 days of about 10%.9,10 In one of the studies, almost half of the 90-day strokes occurred within the first 2 days.10 Two prospective, communitybased cohort studies in Oxfordshire, UK, one in the 1980s and the other in 2002–2004, found even higher cumulative stroke risks from the date of onset of firstever TIA of about 8% at 7 days, 12% at 30 days and 17% at 90 days. The cumulative risks of recurrent stroke early after a mild ischaemic stroke in the 2002–2004 Oxfordshire study were similar: 12% at 7 days, 15% at 30 days and 18% at 90 days (Fig. 16.1, Table 16.1). These risk estimates were hardly affected by excluding patients who had a TIA or mild ischaemic stroke during the study period but did not seek medical attention until they had a subsequent stroke.2 A further population-based study in patients with a TIA in the USA found very similar risks to those reported in Oxfordshire.11 Four studies which reported the timing of preceding TIAs in a total of about 2500 patients with stroke found that 23% gave a history

..

..

Log rank P = 0.8 0

0

30

60

90

Days Fig. 16.1 Cumulative early risk of stroke after a transient ischaemic attack (TIA) or mild ischaemic stroke in the Oxford Vascular Study. (Reprinted from Coull et al.2 © 2004 with permission of BMJ Publishing Group.)

of prior TIA, of which 17% occurred on the day of the stroke and 43% within the previous week (Fig. 16.2).12 The implication of all these studies is that the risk of stroke early after a TIA or mild ischaemic stroke is very high, suggesting that assessment and preventive interventions should start very quickly indeed.7 The early risk of stroke after a transient ischaemic attack or mild ischaemic stroke is very high. Cumulative risks are up to about 10% at 1 week, 14% at 1 month, and 18% at 3 months, so assessment and preventive interventions must start very quickly. Predicting the early risk of stroke after a TIA or mild ischaemic stroke in individual patients Two of the above studies have shown that the clinical features of a TIA provide substantial prognostic information, and the authors have produced simple risk scores to allow doctors more accurately to assess the likely risk of early stroke for individual patients. In the California emergency department cohort, increasing age, longer symptom duration, motor weakness, speech impairment and diabetes independently increased the risk of stroke at 90 days.10 In the Oxfordshire community-based cohorts, the independent predictors of stroke risk at 7 days were the same, with the addition of increased blood pressure.13 Both groups developed simple scoring systems suitable for everyday clinical practice, and both scores have subsequently been validated and shown to perform well in several independent population-based, clinicbased and emergency-department-based data sets.14 A further score, based on the variables included in the California and Oxfordshire ABCD scores, the ABCD2

791

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 792

Chapter 16 Preventing recurrent stroke and other serious vascular events

After TIA or mild ischaemic stroke Death Average annual risk Stroke In 1st week In 1st month In 1st 3 months In 1st year In each subsequent year† Myocardial infarction Average annual risk After any ischaemic stroke Death In 1st month In 1st year In each subsequent year Stroke In 1st year In each subsequent year Myocardial infarction Average annual risk After intracerebral haemorrhage Death In 1st month Average annual risk thereafter Stroke Average annual risk Myocardial infarction

Hospital-referred

Community-based

5%

8%

— 7% 10% 15% 3%

10% 14% 18% at least 20% 6%

2%

2%

— — —

10–20% 25% 7%

— —

at least 10–15% 4–5%

2%

2%

— —

42% 8%

4% —

6% —

Table 16.1 Approximate absolute average risks of death, stroke, and myocardial infarction after transient ischaemic attack (TIA) or mild ischaemic stroke, any ischaemic stroke, and intracerebral haemorrhage.*

*Data from: Johnston et al., 2000;10 Hill et al., 2004;9 Coull et al., 2004;2 Hankey and Warlow, 1994;19 Touzé et al., 2005;20 van Wijk et al., 2005;5 Feigin et al., 2003;26 Dennis et al., 1993;27 Petty et al., 1998;29 Hardie et al., 2003;28 Dennis et al., 2003;55 Counsell et al., 1995;57 Bailey et al., 2001.60 †Risks from Hankey and Warlow, 1994,19 adjusted downwards slightly to allow for overestimation caused by inclusion of higher-risk period in first year.

35 30 Percentage of patients

792

Time from first TIA Time from most recent TIA

25 20 15 10 5 0 –14 –13 –12 –11 –10

–9

–8

–7 – 6 Days

–5

–4

–3

–2

–1

0

Fig. 16.2 Distribution of time from preceding transient ischaemic attack to stroke for patients who reported a transient ischaemic attack (TIA) within the previous 14 days of a stroke. (From Rothwell and Warlow, 2005.12) (Figs 16.2, 16.3, 16.6–16.8 & 16.12 reprinted with permission from Lippincott, Williams & Wilkins.)

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 793

16.2 Prognosis and prediction of future vascular events Table 16.2 Independent risk factors and scoring systems for early stroke risk after transient ischaemic attack (TIA) derived from an emergency department cohort of patients with TIA in the USA (California score), a population-based study of TIA in the UK (Oxfordshire ABCD score), and a combination of both cohorts (ABCD2 score).* Risk factor

Points for score

California score (stroke risk at 90 days) Age ≥ 60 years Clinical features Unilateral weakness Speech disturbance Duration of symptoms ≥ 10 minutes Diabetes Oxfordshire ‘ABCD’ score (stroke risk at 7 days) Age ≥ 60 years Blood pressure Systolic > 140 and/or diastolic ≥ 90 mmHg Clinical features Unilateral weakness Speech disturbance without weakness Other Duration of symptoms ≥ 60 min 10–59 min < 10 min Combined ABCD2 score (stroke risk at 2 days) Age ≥ 60 years Blood pressure Systolic > 140 and/or diastolic ≥ 90 mmHg Clinical features Unilateral weakness Speech disturbance without weakness Other Duration of symptoms ≥ 60 min 10–59 min < 10 min Diabetes

1 1 1 1 1 1 1 2 1 0 2 1 0 1 1 2 1 0 2 1 0 1

*Data from Johnston et al., 2000;10 Rothwell et al., 2005;13 Johnston et al., 2007.14

score, was found to be slightly better at predicting risk at 2 days and also performed well at 7 and 90 days (Tables 16.2 and 16.3).14 In fact, these scores may predict subsequent stroke risk simply because patients with higher scores are more likely to have actually had a TIA than those with lower scores. However they work, any one of them is a potentially very useful tool for identifying groups of patients with a diagnosis of TIA who are at particularly high risk, and who merit immediate evaluation, further investigation, possibly admission to hospital, and urgent intervention with preventive strategies.14 Other clinical features may also help in the prediction of early stroke risk, allowing these scoring systems to be further refined. For example, the early risk of stroke after transient or permanent monocular blindness due to ischaemia is lower than after cerebral TIAs, and a systematic review and meta-analysis of cohort studies demonstrated that the early risk after vertebrobasilar territory TIAs is probably higher than after carotid territory events.15 The early risk also depends on the vascular pathology. A recent meta-analysis of 1700 patients with a first-ever stroke in four population-based studies showed that the risks of recurrent stroke were highest in those with large artery atherothrombotic strokes and lowest in those with lacunar ischaemic stroke due to small vessel disease (Fig. 16.3).16 However, these ischaemic stroke subtype differences may well be less marked in patients with TIA, partly because it is more difficult to categorize TIA patients than strokes and so bias is introduced, and also because some patients with lacunar TIAs have a very high risk of subsequent stroke, the so-called ‘capsular warning syndrome’.17 The presence of infarction on a CT or diffusion-weighted MR brain scan increases the risk of stroke after a TIA or mild ischaemic stroke, while detection of cerebral microemboli might predict risk in patients with large artery atherothrombotic disease.7,17,18 Novel biomarkers may also help in risk prediction. Further research is needed to assess whether these various characteristics add independent prognostic value to the existing – and remarkably good – predictive models.7

Table 16.3 Stroke risks at 2, 7 and 90 days after transient ischaemic attack in patients at low, moderate and high risk according to the ABCD2 score. Data are from a total of 4799 patients in the California and Oxfordshire derivation cohorts and four California and Oxfordshire validation cohorts.* Risk level

ABCD2 score

Patients (n)

Risk at 2 days (no. strokes)

Risk at 7 days (no. strokes)

Risk at 90 days (no. strokes)

Low Moderate High

0–3 4–5 6–7

1628 2159 1012

1.0% (17) 4.1% (89) 8.1% (82)

1.2% (19) 5.9% (128) 11.7% (118)

3.1% (50) 9.8% (211) 17.8% (180)

*Data from Johnston et al., 2007.14

..

..

793

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 794

Chapter 16 Preventing recurrent stroke and other serious vascular events

Survival free of recurrent stroke

794

1 0.9 0.8 CE LAA SVS UND

0.7 0.6

Log rank P < 0.001

0.5 0

30

60

90

Days Actuarial recurrent stroke risks (95% CI) At 7 days SVS

0%

At 1 month 2.0% (0–4.2)

At 3 months 3.4% (0.5–6.3)

CE

2.5% (0.1–4.9)

4.6% (1.3–7.9)

11.9% (6.4–17.4)

UND

2.3% (0.5–4.1)

6.5% (3.6–9.6)

9.3% (5.6–13.0)

LAA

4.0% (0.2–7.8)

12.6% (5.9–19.3)

19.2% (11.2–27.2)

Fig. 16.3 Three-month survival free of recurrent stroke by Trial of ORG 10172 in Acute Stroke Treatment (TOAST) sub type for patients in the Oxford Vascular Study and the Oxfordshire Community Stroke Project. CE, cardioembolic; LAA, large-artery atherosclerosis; SVS, small-vessel stroke; UND, undetermined. The corresponding actuarial risks of recurrent stroke in each subtype are given below the figure with 95% confidence intervals. (From Lovett et al., 2004.16)

Simple clinical features of a transient ischaemic attack (such as patient age, blood pressure, neurological deficit and duration of symptoms) can help predict which patients are at highest risk of stroke in the first few days, weeks and months after onset.

16.2.2 Longer-term prognosis after transient ischaemic attack or mild ischaemic stroke While many studies of prognosis after a transient ischaemic attack (TIA) or mild ischaemic stroke underestimated the very early risk of stroke, they still provide very useful information on the longer-term prognosis. Data from studies that more or less fulfil the criteria for an ideal prognostic study (Table 10.2) have shown approximate average annual risks over the first 5 years as follows (Table 16.1):19 • Death: about 8% per year in community-based studies (about 1.4 times greater than individuals of the same age and sex without a TIA), and about 5% per year in hospital-referred cohorts. About 40% of deaths after

TIA are due to ischaemic heart disease, 25% to stroke, 5% to other vascular events such as ruptured aortic aneurysm and 30% to non-vascular disorders. • Stroke: about 7% per year in community-based studies (about 7 times greater than individuals of the same age and sex without a TIA), and about 4% per year in hospital-referred cohorts. Most of these strokes are ischaemic and about two-thirds are major disabling strokes. • Myocardial infarction: about 2% per year in community and hospital-referred cohort studies.20 • Vascular events (stroke, myocardial infarction or vascular death): about 10% per year in community-based studies and about 8% per year in hospital-referred cohorts. For stroke, we know that the early risk is particularly high (section 16.2.1). Of the studies that have provided reliable data on early risk (section 16.2.1), only one hospital-based TIA cohort study reported results beyond the first few months, and found a risk of stroke of about 15% at 1 year. This may have been an underestimate, since strokes in the first 24 h were excluded and follow-up depended on administrative data alone.9 The first-year stroke risk in community-based studies is bound to be higher than this and probably in excess of 20%, since by 3 months the risk is already 17–18% (section 16.2.1).2 Therefore, the average annual risks given above substantially underestimate the risks of stroke (and so of vascular events) in the first year and, by incorporating some of the early high risk, slightly overestimate the risks of these outcomes from year 2 onwards. Prospective follow-up data for about 10 years have recently been published for a cohort of around 2500 patients recruited within 3 months of a mild ischaemic stroke or TIA into a factorial randomized trial of different aspirin doses and of atenolol vs placebo (the Dutch TIA trial). The average annual risks of a death, stroke and other vascular events in the first few years were slightly lower than those reported above for hospital-referred cohorts.5 This is consistent with previous observations, and reflects the slightly lower age and generally rather better prognosis of patients recruited into trials.21 The longer-term follow-up in this study showed that the relatively high early annual stroke risk fell over the first few years to around 1–2% after the fourth year, annual mortality increased gradually with time from less than 4% to 7–8% by the tenth year of follow-up, and the annual risk of a vascular event fell from about 7% to about 3.5% over the first 3 years and then rose steadily thereafter to around 7% by year 10 (Fig. 16.4).5 A long-term follow-up study of 290 patients with a previous TIA, who had originally been recruited into either a community-based or hospital-based cohort in Oxfordshire and had survived free of stroke for a median of 3.8 years from their most

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 795

50

10 9 8 7 6 5 4 3 2 1 0

40

0

2

4 6 8 10 Time (years after index event)

12

Vascular event (with 95% CIs) Mortality Stroke Fig. 16.4 Annual risks of death, stroke and vascular events over time in a cohort of transient ischaemic attack and mild ischaemic stroke patients recruited into a randomized trial. (Reprinted from van Wijk et al. Lancet5 © 2005 with permission from Elsevier)

recent TIA and were therefore selected to be at low stroke risk, found constant risks up to 10 years of about 2% per year for stroke, 3% per year for myocardial infarction or death attributable to coronary heart disease and 4% per year for vascular events (Fig. 16.5).22 The message from these studies is that, although the very high early risk of stroke after TIA or mild ischaemic stroke falls to a constant, lower rate after the first few months, the annual risks of myocardial infarction and of all vascular events remain constant, or even increase, with long-term follow-up. The implication is that, after preventive efforts are started early, they need to be sustained in the long term. After a transient ischaemic attack or mild ischaemic stroke, the very high early risk of stroke falls to a constant lower rate after the first few months, while the annual risks of myocardial infarction and of all vascular events remain constant, or even increase, with long-term follow-up. Preventive efforts must therefore be started early and sustained long term. Predicting long-term prognosis after transient ischaemic attack or mild ischaemic stroke in individual patients Several large cohort studies have identified independent predictors of the long-term risks of vascular outcomes after transient ischaemic attack (TIA) or mild ischaemic stroke, and three predictive models have been developed and validated in independent data sets to allow more refined estimates of the risks in individual patients.1,23–25 The predictive variables used in these three models are

..

..

Stroke, MI or vascular death MI or CHD Stroke

45

Event rate (%)

Yearly risk (%)

16.2 Prognosis and prediction of future vascular events

35 30 25 20 15 10 5 0 0

2

No. of patients remaining Stroke: 290 264 MI or CHD: 290 265 Stroke, MI, 290 260 or vasc death

4 6 Time (years) 230 230 225

196 200 189

8

164 173 157

10

136 137 130

Fig. 16.5 Kaplan–Meier event rates for any stroke, any myocardial infarction (MI) or death from coronary heart disease (CHD), and any stroke, myocardial infarction, or vascular death among patients with a prior transient ischaemic attack (from the Oxfordshire Community Stroke Project and a contemporaneous hospital-referred series of TIA patients) and a median of 3.8 years between their most recent transient ischaemic attack and the start of follow-up. (From Clark et al.22 © 2005 BMJ Publishing Group.)

shown in Table 16.4. The predicted risks can be calculated using information in the original publications; online versions of two of the models are available (Table 16.4) allowing rapid calculation of predicted risk; and for the model specifically developed to predict the risk of ipsilateral ischaemic stroke in patients with recently symptomatic carotid stenosis being considered for surgery, a simplified version using a set of coloured charts is available (section 16.11.8 and Fig. 16.46), similar to those used for prediction of cardiovascular risk in primary care (section 18.5.1). Although the variables included in these predictive models vary, because of differences between studies in the variables available for modelling, the types of patients included, the outcomes predicted and the precise statistical methods used, it is striking that the variables are largely those available from the clinical history, with just a few from clinical examination and investigations. Modelling of long-term risk of stroke, death and vascular events in the Dutch TIA trial also showed the predictive power of the history alone. Simply documenting the patient’s age, sex, history of myocardial infarction, diabetes, hypertension and peripheral artery disease provided almost as much information about the risk of a

795

1653 TIA patients in the UK-TIA trial 107 TIA patients in the Oxfordshire Community Stroke Project

2449 patients in the UK-TIA trial 6431 prior ischaemic stroke patients in the CAPRIE trial 340 ischaemic stroke and TIA patients in the population-based NoMaSS study 759 medically treated patients with recent TIA or mild ischaemic stroke and 50–99% ipsilateral carotid stenosis in the NASCET trial†

469 patients with hospital-referred TIAs

525 female patients with TIA or mild ischaemic stroke in the WEST trial

5-year risk of ipsilateral carotid territory ischaemic stroke

Increasing ipsilateral carotid stenosis or near occlusion‡ Male sex Increasing age Time since last event (risk decreases with time) Nature of presenting event (major stroke vs minor stroke vs multiple TIAs vs single TIAs vs ocular attacks only) Diabetes mellitus Previous myocardial infarction Peripheral vascular disease Treated hypertension Irregular/ulcerated plaque (on carotid angiography)

Increasing age Male sex Cerebral vs only ocular TIAs Carotid and vertebrobasilar TIAs Increasing number of TIAs in preceding 3 months Peripheral vascular disease Left ventricular hypertrophy (ECG) Residual neurological signs (Ischaemic heart disease – only in coronary events model) Congestive heart failure Diabetes mellitus Prior stroke Age > 70 years Stroke (not TIA) Severe hypertension Coronary artery disease

5-year risks of: – stroke – coronary events – vascular events (stroke, MI or vascular death)

2-year risk of stroke or death

Predictive variables increasing risk in model

Outcomes predicted

www.stroke.ox.ac.uk

—

www.dcn.ed.ac.uk/model/models.asp

Website for online version

*Data from: Hankey et al., 199223; Hankey et al., 199324; Kernan et al., 200025; Rothwell et al., 2005.1 †% stenosis measured using the NASCET criteria. ‡Near occlusion reduces risk compared with increasing stenosis above 50%. TIA, transient ischaemic attack; MI, myocardial infarction; ECG, electrocardiogram; WEST, Women’s Estrogen to prevent Stroke Trial; UK-TIA, United Kingdom Transient Ischaemic Attack trial; CAPRIE, Clopidogrel versus Aspirin in Patients at Risk of Ischaemic Events trial; NoMaSS, Northern Manhattan Stroke Study; ECST, European carotid Surgery Trial; NASCET, North American Symptomatic Carotid Endarterectomy Trial.

1211 medically treated patients with recent TIA or mild ischaemic stroke and 0–99% ipsilateral carotid stenosis in the ESCT trial†

Test cohorts

Derivation cohort

796

Table 16.4 Variables used in prediction models for long-term outcome after transient ischaemic attack (TIA) or mild ischaemic stroke.*

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 796

Chapter 16 Preventing recurrent stroke and other serious vascular events

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 797

16.2 Prognosis and prediction of future vascular events

Simple clinical features can help predict an individual patient’s long-term risk of stroke and other vascular events after a transient ischaemic attack or mild ischaemic stroke.

100

IS (n = 545) ICH (n = 66)

80 Per cent surviving

future event as adding the results of the examination findings and diagnostic procedures such as CT scan and ECG.5 Thus, as discussed for very early risk prediction (section 16.2.1), to further improve existing models of risk prediction, we need to know whether extra variables, such as cerebral microemboli, brain imaging findings, and so on, have additional predictive value.

60 40 20 0 0

The prognosis for death and disability in the first few months after acute ischaemic stroke has been discussed in section 10.2. Older cohort studies of the long-term prognosis after stroke generally reported results for all types of stroke combined, often because brain imaging was not performed on a large proportion of the patients so that the pathological type of stroke (ischaemic or haemorrhagic) could not be assigned accurately. Because most strokes are ischaemic in type, the results of these studies mainly reflect the prognosis of ischaemic stroke. However, several studies in which a much larger proportion of patients had brain imaging early enough after their stroke to allow a pathological diagnosis have reported outcome up to several years after stroke, and some of these have also reported data on ischaemic stroke subtypes. Death after any ischaemic stroke Community-based studies of first-ever strokes from 1990 onwards have reported a risk of dying in the first month after ischaemic stroke of 10–20%26 (Table 16.1). Predictors and causes of early case fatality have been discussed in section 10.2. About one-quarter of ischaemic stroke patients in community-based studies in white patients were dead within 1 year of their stroke, about half by 5 years, and over three-quarters by 10 years. The risk of dying is highest in the first year, and thereafter falls to about 6–7% per year, similar to the long-term annual risk of death after a TIA (Fig. 16.6 and Table 16.1) (section 16.2.2).27–29 Although deaths related to the index stroke are commonest in the first month, in the longer term deaths from recurrent stroke, cardiac disease and non-vascular causes become increasingly important (Fig. 16.7).27,28,30–32 Early after ischaemic stroke, the risk of death is highest in those with cardioembolic stroke and lowest in those

..

..

1

2

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

4

5

IS (n = 173) ICH (n = 32)

0 (b)

3 Years

(a)

Probability

16.2.3 Prognosis after all severities of ischaemic stroke

1

2

3

4 5 6 Survival (years)

7

8

9

10

Fig. 16.6 Kaplan-Meier curves showing probability of survival over long-term follow-up stratified by the pathology of firstever stroke in (a) the Oxfordshire Community Stroke Project, UK and (b) the Perth Community Stroke Study, Australia; IS, ischaematic stroke; ICH, intracerebral haemorrhage. (From Dennis et al., 199327 and Hardie et al., 2003.28)

with lacunar stroke. Most of this difference between ischaemic subtypes occurs in the first month, probably as a result of larger infarct size and higher stroke recurrence risk among the non-lacunar subtypes. Beyond the first month, the difference in risk of dying between ischaemic subtypes attenuates, but the early difference means that the risk of being dead by 5 years is about 80% after cardioembolic stroke, about 50% after ischaemic stroke of uncertain cause, and about one-third after large artery (atherothrombotic) or lacunar stroke, although these risks are generally lower in hospital-based series (Figs 16.8 and 16.9).30,33–35 Increasing age, initial infarct severity (of which ischaemic subtype is one important indicator), stroke recurrence, cardiovascular and respiratory comorbidities, and seizures appear to be independent predictors of death at any time after ischaemic stroke.31

797

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 798

Chapter 16 Preventing recurrent stroke and other serious vascular events

100

100% 90%

80 Survival (%)

80% 70% Per cent

798

60% 50%

60 40 20

40% 0 30%

0

20%

2 3 Years after stroke

Lacune Atherosclerosis

10% 0%

1

ICH

Isch

SAH

Und

Total

Pathology of first-ever stroke Alive Undetermined Non-vascular

4

5

Uncertain cause Cardioembolic P < 0.0001

Fig. 16.8 Observed percentage surviving (Kaplan-Meier estimates) after incident ischaemic stroke among 442 residents of Rochester, Minnesota, 1985–1989, by ischaemic stroke subtype. (From Petty et al., 2000.35)

Cardiovascular Recurrent stroke First stroke

Fig. 16.7 Histogram showing cause of death and survival patterns over 10 years stratified by the pathology of first-ever stroke. Data are from the Perth Community Stroke Study, Australia. Isch, ischaemic stroke; ICH, intraceberal haemorrhage; SAH, subarachnoid haemorrhage; Und, unknown pathology. (From Hardie et al., 2003.28)

year after any ischaemic stroke, but all studies still show a much higher risk during the first year (mainly confined to the first month) and an average annual risk of about 4–5% per year thereafter, similar to the longer-term risk after a TIA or mild ischaemic stroke (Table 16.1). As we saw above (section 16.2.2), the early recurrence risk is substantially higher after large artery atherothrombotic ischaemic stroke than lacunar stroke, with cardioembolic stroke and other types of ischaemic stroke intermediate between these two. In the longer term, however, this difference attenuates (Figs 16.3 and 16.9). The

Recurrent stroke after any ischaemic stroke As discussed for TIA and mild ischaemic stroke, varying definitions of recurrent stroke in different studies mean that many underestimate the recurrence risk in the first

No. studies

No. patients

Peto OR (95% CI)

0–1 month

7

2638

3.8 (2.8 –5.2)

1–12 months

7

2334

2.3 (1.7–3.1)

1–5 years

4

971

1.8 (1.3 –2.5)

0–1 month

6

2137

2.1 (1.2–3.7)

1–12 months

6

1665

1.2 (0.9 –1.8)

1–5 years

3

575

1.6 (1.0 –2.7)

0.2

0.5

Death

Recurrent stroke

0.1

Non-lacunar infarction better outcome

1

2

5

10

Lacunar infarction better outcome

Fig. 16.9 Pooled odds ratios (OR) (nonlacunar vs lacunar ischaemic stroke) from a meta-analysis of community and hospital-based inception cohort studies with available data on death and recurrent stroke at 1 month, 1–12 months, and 1–5 years after ischaemic stroke. The pooled ORs are represented by diamonds, with 95% confidence intervals (CI) represented by the width of the diamonds. (Data from Jackson and Sudlow, 2005.33)

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 799

16.2 Prognosis and prediction of future vascular events

100

Recurrent strokes %

80

60

28% 68%

40

20

47% 15%

0 Lacunar index ischaemic stroke Other ischaemic Intracerebral haemorrhage

Non-lacunar index ischaemic stroke Non-lacunar ischaemic stroke Lacunar ischaemic stroke

Fig. 16.10 Pooled data from prospective studies of recurrent stroke subtypes following lacunar and non-lacunar ischaemic stroke at baseline. Total number of recurrent strokes among lacunar patients = 279, and among non-lacunar patients = 117. Other ischaemic = unclassified ischaemic recurrences; ICH, intracerebral haemorrhage. (Reprinted from Jackson and Sudlow33 © 2005 with permission of Oxford University Press.)

particularly high risk early after larger artery atherothrombotic stroke may be due to the presence of unstable atherosclerotic plaque causing artery-to-artery embolism, with the risk falling when the plaque endothelializes or ‘stabilizes’ in some other way.36 At least three-quarters of recurrent strokes after an ischaemic stroke are ischaemic again. However, the true figure may be as high as 90% or more, since in most studies the proportion of recurrent strokes of uncertain pathological type was rather high, and most of these were probably ischaemic.35,37–40 There is also some evidence that recurrent ischaemic strokes tend to be of the same subtype as the index stroke (Fig. 16.10).33,41,42 However, potential biases mean that the data should be interpreted cautiously. Distinguishing between genuine stroke recurrence and other causes of neurological worsening, especially when the symptoms and or signs are similar to those from the index stroke, can be very difficult, definitions of stroke recurrence vary between studies, and assigning a precise pathological type and subtype is also very challenging.8 Of course, modern brain imaging techniques, in particular diffusion-weighted magnetic resonance imaging (DW-MRI) (section 5.5.2), are very helpful, but very few studies of recurrent stroke

..

..

so far have used MRI, and none DWI. Furthermore, even when state-of-the-art imaging is readily available, patients with recurrent stroke may not be admitted to hospital, or may be to too unwell for – or have other contraindications to – MR scanning.33 Relatively few studies have assessed independent predictors of recurrent stroke in the long term after any ischaemic stroke, but several have assessed these after any stroke (i.e. of all pathological types). The most consistent predictor is age, with previous stroke, initial stroke severity, stroke subtype, ischaemic heart disease, atrial fibrillation, increased blood pressure, diabetes and smoking also emerging in several studies.30,37,38,40,43–47 We are not aware of any validated models for predicting recurrent stroke or other vascular outcomes in individual patients after any ischaemic stroke, although earlier we have discussed prediction models for patients with TIA or mild ischaemic stroke (section 16.2.2). There have been a number of studies specifically assessing risk of recurrent stroke after ischaemic stroke in young patients (under 50 years). The numbers of recurrences have generally been very small, making their estimates imprecise, but all have found much lower risks of recurrence than in older patients, particularly in those with no identified cause or risk factors for ischaemic stroke.48–54 Therefore, secondary preventive interventions may not be appropriate for this group of very lowrisk young patients, but the identification of precisely which young patients in this low-risk group should or should not receive preventive treatments needs more precise information than is currently available. Probably the only way to achieve this would be by pooled analysis of individual patient data sets from many different centres, since single-centre studies will never be able to accrue sufficient numbers of patients and outcome events for precise estimates. Myocardial infarction after any ischaemic stroke The risk of myocardial infarction after any ischaemic stroke is about 2% per year, the same as after a TIA or mild ischaemic stroke (Table 16.1).20 There is no clear evidence of an initially high early risk as there is for recurrent stroke. Within the first month after ischaemic stroke, the risk of recurrent stroke is at least 2.5 times that of myocardial infarction, but thereafter the ratio falls to about two times. However, although recurrent strokes occur more commonly than cardiac events over the long term after an ischaemic stroke, cardiac mortality is about two times higher than stroke mortality from about 1 month onwards.32 There is very little information on whether the risk of myocardial infarction after ischaemic stroke varies by ischaemic stroke subtype. What little

799

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 800

800

Chapter 16 Preventing recurrent stroke and other serious vascular events

information there is suggests that cardiac deaths occur with approximately equal frequency among the different ischaemic stroke subtypes, but even in a meta-analysis of data from all the available studies, numbers of events were limited, making the results imprecise.33 After transient ischaemic attacks and ischaemic stroke it is important to recognize the risk of non-stroke vascular events such as myocardial infarction; indeed, coronary heart disease is a more likely cause of death than stroke in the long term. Like patients with a TIA or mild ischaemic stroke, patients with any ischaemic stroke remain at elevated risk of recurrent stroke and other serious vascular events for at least 10 years after their initial event, and so prevention efforts must also be sustained long term.

16.2.4 Prognosis after intracerebral haemorrhage The prognosis for death and disability in the first few months after intracerebral haemorrhage (ICH) has been discussed in section 10.2. Here, we consider what is known about vascular events and death, both early on and in the long term after ICH. Death after intracerebral haemorrhage Community-based studies have shown a much higher early case fatality after ICH than after ischaemic stroke, with a pooled estimate of 42% dying within 1 month (Table 16.1). The relatively low incidence of ICH, which comprises between 10 and 20% of all first-ever strokes, at least in white populations where the incidence has been best studied, combined with the high early case fatality, means that the number of patients available for long-term follow-up in single centres is relatively small, making estimates from individual studies imprecise.55 In addition, increasing use of brain imaging earlier after stroke is probably leading to an increased proportion of mild haemorrhages being diagnosed (section 5.2), so changing the nature of the cohorts being followed up.56 Factors associated with a high early case fatality are older age, clinical severity of the initial stroke, the size of the haematoma, the presence of intraventricular blood and anticoagulant treatment. Some of these, such as haematoma size, intraventricular blood and clinical severity, are correlated with each other and may not be independent outcome predictors.55 After the first month, the average annual risk of death is about 8% per year, similar to the long-term risk after

an ischaemic stroke (Fig. 16.6). Over the long term, the survival curves for ischaemic stroke and ICH appear to converge. This may be due to a survival effect, with the longer-term survivors of ICH being younger and fitter than those with an index ischaemic stroke.27,28,55,57,58 The main causes of death in 1-month survivors are the initial or a recurrent stroke (about one-third) and cardiac or other vascular causes (about one-quarter).59 The risk of death in the first month after intracerebral haemorrhage, as currently diagnosed, is at least twice that for ischaemic stroke. Thereafter, the long-term risk of death appears similar to ischaemic stroke. Recurrent stroke and myocardial infarction after intracerebral haemorrhage A systematic review of studies reporting recurrent stroke after ICH analysed data from three community-based populations (including a total of 146 patients) and seven hospital-based cohorts (including a total of 1734 patients). Pooled recurrent stroke rates were 6.2% per year in the community-based studies and 4% per year in the hospital-based studies, similar to the recurrence rates after an ischaemic stroke (section 16.2.3). However, while most recurrences after ischaemic stroke are ischaemic again, over half (59%) of recurrences after ICH were haemorrhagic again, 26% were ischaemic and 15% were of uncertain pathology.60 The long-term risk of recurrent stroke after intracerebral haemorrhage seems similar to that for ischaemic stroke (excluding the very high early risk after ischaemic stroke). Unsurprisingly, the risk of recurrence seems to depend on the underlying cause of the initial haemorrhage. In younger patients, the commonest cause of ICH is an intracranial vascular malformation, and the specific recurrence risk is discussed in section 15.3. In studies which distinguished between deep and lobar haemorrhage, recurrence was about twice as frequent in the former group.60 This may reflect the different predominant underlying causes in these different locations, with a preponderance of cerebral amyloid angiopathy thought to underly many lobar haemorrhages in older patients, and intracranial small vessel disease thought to be the vascular pathology causing most deep haemorrhages. Haemorrhages due to cerebral amyloid angiopathy are said to be particularly likely to recur. Possession of an apolipoprotein E ε4 or ε2 allele may predispose to cerebral amyloid angiopathy and lobar ICH, and one study has demonstrated an increased risk

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 801

16.3 Pharmacological blood pressure reduction

of recurrent ICH in patients with lobar haemorrhage who are carriers of either the ε4 or ε2 allele.61 Several observational studies have also shown that uncontrolled hypertension is a risk factor for recurrence after ICH, and randomized trials of blood pressure reduction among patients with a history of stroke, including those with ICH, support this (section 16.3.2).62–67 There are really no reliable data about the risk of myocardial infarction long term specifically after ICH, and no validated prognostic models for predicting stroke, myocardial infarction or vascular death. Since the numbers of individuals available for study in singlecentre cohorts is small, pooling of data from cohort studies of ICH collecting similar baseline data and using similar definitions is probably the only feasible way of obtaining more precise data on independent predictors of prognosis after ICH. Up to 90% of recurrent strokes after an ischaemic stroke are ischaemic again, while around 60% of recurrent strokes after intracerebral haemorrhage are haemorrhagic again.

16.3 Pharmacological blood pressure reduction

16.3.1 Relationship between blood pressure and stroke in observational studies Raised blood pressure is the most important causal and treatable risk factor for stroke (section 6.6.3). Systematic reviews of prospective observational studies in healthy, middle-aged adults, with appropriate adjustment for regression dilution bias, have demonstrated a continuous log linear relationship between usual systolic and diastolic blood pressure and stroke risk. For every 10 mmHg reduction in usual systolic and 5 mmHg reduction in usual diastolic blood pressure, the relative risk of stroke decreases by about one-third.68–71 This relationship is not so steep in the elderly, but their higher absolute risk of stroke means that their absolute reduction in stroke risk for a given absolute decrease in blood pressure is still substantially greater than in younger people. The relationship is similar in Asian and white populations, and for ischaemic and haemorrhagic stroke. Unfortunately, there were neither sufficient numbers of events nor enough detail on stroke subtypes in these large overviews of prospective observational studies to allow comparisons between blood pressure and different

..

..

subtypes of ischaemic or haemorrhagic stroke.68–71 However, methodologically rigorous studies comparing the frequency of prior hypertension in patients with different subtypes of ischaemic stroke do not support the common assertion that lacunar ischaemic stroke is more strongly associated with raised blood pressure than other ischaemic stroke subtypes.72,73 Similarly, data from the most methodologically sound studies comparing the frequency of prior hypertension in patients with deep vs lobar locations of intracerebral haemorrhage provide little – if any – evidence to support the often stated view that hypertension is associated particularly with deep intracerebral haemorrhage (section 8.2.1).74 Some studies of the association between blood pressure after a stroke and subsequent stroke risk have suggested that the risk of stroke recurrence may be higher at both high and low blood pressures, giving a U- or J-shaped relationship.75,76 However, the higher recurrence risk at the two ends of the blood pressure distribution may not be because of a causal effect of blood pressure but because of so-called ‘reverse causality’, i.e. severe strokes may be independently associated both with a very high or low post-stroke blood pressure as well as with a higher risk of recurrence. In fact, in patients with a previous mild ischaemic stroke or TIA included in a randomized trial of antiplatelet treatment, the relationship between the risk of stroke and both systolic and diastolic blood pressure was similar to that observed in healthy populations, with lower blood pressures conferring a lower risk of stroke throughout the observed range of blood pressures. Each 10 mmHg decrease in usual systolic pressure and 5 mmHg decrease in usual diastolic pressure was associated with about a one-third decrease in stroke risk (Fig. 16.11).77 However, the relationship between blood pressure and stroke risk may be different in patients with severe disease of the extracranial arteries in the neck, since cerebral perfusion (and autoregulation of cerebral blood flow) may be so severely impaired that it is directly dependent on systemic blood pressure. Data from patients with a history of recent mild ischaemic stroke or TIA and stenosis of at least one carotid artery who were randomized to medical treatment in two large carotid surgery trials showed that, although the risk of stroke increased with blood pressure in most patients with symptomatic carotid artery disease, it was not so steep as in other patients with prior stroke or TIA, despite appropriate correction for regression dilution bias (Fig. 16.12). And, importantly, the relationship was reversed in patients with bilateral severe carotid stenosis, so that in this small group of patients lower blood pressure was associated with an increasing risk of stroke (Fig. 16.12c).78

801

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 802

Chapter 16 Preventing recurrent stroke and other serious vascular events

Relative risk of stroke

2.0

1.0

0.5

0.25 95 120 170 75 80 85 90 130 140 150 160 Usual diastolic blood pressure (mmHg) Usual systolic blood pressure (mmHg)

Fig. 16.11 Relative risk of stroke by approximate usual diastolic and systolic blood pressure among 2435 patients with a history of transient ischaemic attack or mild ischaemic stroke in the UK-TIA aspirin trial. Solid squares represent stroke risk in each category relative to the risk in the whole study population. The sizes of the squares are proportional to the number of events in each category. The vertical lines indicate the 95% confidence intervals. (From Rodgers et al.77 © 1996 BMJ Publishing Group.)

(c) Both stenoses ≥70%

(a) Both stenoses 18 years with stroke or TIA within the last 1–6 months, and no known ischaemic heart disease. Those with haemorrhagic stroke only included if considered at high risk of ischaemic stroke or ischaemic heart disease Baseline LDL cholesterol 2.6–4.9 mmol/L Atorvastatin 80 mg daily vs placebo 4731 4731

Adults aged 40–80 years with history of ischaemic stroke or TIA, ischaemic heart disease, peripheral arterial disease, diabetes, or hypertension Baseline total cholesterol > 3.5 mmol/L

Treatments compared Simvastatin 40 mg daily vs placebo Total number of participants 20 536 Number of participants with prior 3280 stroke or TIA Baseline characteristics* Mean time from stroke/TIA 3 months 4 years to randomization Mean age (years) 63 66 Male : female (%) 60 : 40 75 : 25 Mean baseline total/LDL 5.5/3.4 5.9/3.4 cholesterol (mmol/L) Mean follow-up (years)* 4.9 4.8 Mean difference in LDL cholesterol 1.4 1 between treatment groups (placebo-statin, mmol/L)† Effects on various outcomes† (number and % with each outcome and relative risk for statin vs placebo) Atorvastatin Placebo RR Simvastatin Placebo RR (n = 2365) (n = 2366) (95% CI) (n = 1645) (n = 1635) (95% CI) Any stroke 265 311 0.85 169 170 0.99 (11%) (13%) (0.73 to 0.99) (10%) (10%) (0.81 to 1.21) Ischaemic stroke 218 274 0.80 100 122 0.81 (9%) (12%) (0.67 to 0.94) (6%) (7%) (0.63 to 1.05) Haemorrhagic stroke 55 33 1.7 21 11 1.9 (2%) (1%) (1.1 to 2.6) (1%) (0.7%) (0.9 to 3.9) Major coronary events‡ 81 120 0.68 171 218 0.78 (3%) (5%) (0.51 to 0.89) (10%) (13%) (0.65 to 0.94) Major vascular events§ 530 687 0.77 406 488 0.83 (22%) (29%) (0.74 to 0.86) (25%) (30%) (0.74 to 0.93) SPARCL Stroke Prevention by Aggressive Reduction of Cholesterol Levels study. HPS Heart Protection Study. RR relative risk. *Of patients with prior stroke or TIA. †In patients with prior stroke or TIA. ‡Non-fatal myocardial infarction or coronary death. §Stroke, major coronary event, revascularization, emergency hospitalization for coronary ischaemia, or clinically significant peripheral arterial disease in the SPARCL trial; stroke, major coronary event or revascularization in the HPS.

P450 (simvastatin and atorvastatin) are combined with inhibitors of this enzyme (such as ‘azole’ antifungal agents, HIV protease inhibitors, amiodarone, verapamil, diltiazem and grapefruit juice), and it is important always to consider this possibility when prescribing them.126

..

..

Liver problems The US Food and Drug Administration reported a rate of liver failure attributable to statins of one per million person-years of use, but there were no cases of liver failure reported in the RCTs, and a very small excess (0.2%)

813

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 814

814

Chapter 16 Preventing recurrent stroke and other serious vascular events

of raised liver enzymes in those allocated a statin compared with placebo.110,118,120 Peripheral neuropathy Case–control studies have found users of statins to have an increased risk of developing polyneuropathy of at least 2.5 times the background population risk, with a substantial increase in this risk after prolonged treatment. Assuming the background population risk of idiopathic polyneuropathy to be about one per 10 000 person-years, this implies an excess absolute risk of at least 2.5 per 10 000 person-years. However, no significant excess of peripheral neuropathy was reported in the largest of the statin trials.127,128 These data are generally very reassuring. With the exception of the SPARCL trial, the large trials assessed statin doses equivalent to simvastatin 40 mg or less, but safety data from trials comparing intensive vs moderate cholesterol-lowering statin regimens are also generally reassuring about higher statin doses as well (section 16.4.6). However, it is important to remember that the RCTs generally screened for and excluded individuals with raised liver enzymes, raised creatinine kinase or impaired renal function, and the largest trial had a several-week pre-randomization run-in period that included a period of active treatment, designed to identify and exclude those unlikely to continue to take allocated treatment for several years. Their results on adverse effects should be interpreted in this context, since patients developing early adverse effects or considered to be at risk of adverse effects were not randomized.118 Cholesterol-lowering treatments do not increase the risks of cancer or suicide, and serious adverse effects of statin monotherapy (such as rhabdomyolysis or liver failure) are extremely rare. However, the combination of a statin with a fibrate substantially increases the risk of rhabdomyolysis and should be avoided or used only with extreme caution.

16.4.5 Who to treat and who not to treat The specific evidence for patients with a prior stroke or TIA from the HPS and the SPARCL trial, and the systematic reviews of RCTs of a statin vs placebo in a wide range of individuals at increased risk of major vascular events, suggests that cholesterol lowering with a statin should be considered in everyone with a history of an ischaemic cerebrovascular event. Although subgroup analyses suggest that the effects of lowering cholesterol are similar regardless of the baseline cholesterol or LDL cholesterol level, participants in

HPS had a non-fasting baseline cholesterol level of > 3.5 mmol/L, and those in SPARCL an LDL-cholesterol of > 2.6 mmol/L.118,120 Since these are the trials from which most of the direct evidence comes, we do not recommend statins for patients whose untreated cholesterol or LDL cholesterol levels are below these levels, but in our experience of clinical practice in European populations this is rare. Neither do we recommend routine prescription of a statin for patients with a history of intracerebral haemorrhage (ICH) but no ischaemic vascular events, since only very few such patients were included in the RCTs. Furthermore, these patients are at higher risk of a subsequent ICH than those with a prior ischaemic cerebrovascular event (section 16.2.4), and there may be an increased risk of haemorrhagic stroke with statin treatment (section 16.4.2). For patients with a history of ICH who are also considered to be at particularly high risk of future ischaemic stroke or coronary events, it is probably reasonable to prescribe a statin. There must have been patients like this in the RCTs of statins, although they have not been separately enumerated, and direct evidence in this group of patients is not available. The mean age of the participants in the statin trials was about 60, and there were relatively few very elderly (over 80 years). However, subgroup analyses found no evidence that the relative effects of treatment were influenced by age.117 Furthermore, an observational study of statin use in routine clinical practice in Scotland found that, although the patients treated with statins in the population under study were older on average than the trial participants, their overall effects were similar to those found in the trials.129 Thus, old age per se should not be a barrier to treatment with a statin, but the possible problems of polypharmacy should always be considered, especially in older patients (section 16.18.2).

16.4.6 How low to go and which statin regimen to use The indirect evidence evidence from the RCTs suggests that greater benefits are conferred by larger reductions in LDL cholesterol (section 16.4.2 and Fig. 16.20). Furthermore, recent RCTs directly comparing intensive with moderate lipid-lowering statin regimens suggest that more intensive regimens achieve significantly greater reductions in LDL cholesterol, reduce progression of coronary artery atherosclerosis, and reduce the risks of major vascular events including stroke.130–133 The two largest of these trials between them included almost 20 000 patients and found no significant difference between atorvastatin 80 mg daily and atorvastatin 10 mg or simvastatin 20 mg daily in the risks of death,

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 815

16.4 Pharmacological cholesterol reduction

cancer, violent or traumatic death, rhabdomyolysis or myopathy. In both trials there was a small (no more than 1%) absolute increase in the risk of persistently raised transaminases with the higher dose of atorvastatin but no cases of liver failure were reported.131,133 One of the trials found a slight excess of myalgia and gastrointestinal symptoms with the higher dose of atorvastatin.133 The ongoing Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine (SEARCH) trial of 12 000 patients with ischaemic heart disease randomized to 80 mg vs 20 mg daily simvastatin for 5 years will provide additional evidence on this issue (see http://www.ctsu.ox.ac.uk/projects/search/index_html). At present our recommended statin regimen is either the HPS regimen of simvastatin 40 mg daily or the SPARCL regimen of atorvastatin 80 mg daily. In keeping with the indirect and direct comparisons between different statin regimens, we would expect atorvastatin 80 mg to produce a larger relative reduction in cholesterol, LDL cholesterol and clinical outcomes than simvastatin 40 mg. However, since more intensive lipid lowering may be associated with a slightly increased risk of adverse effects (see above), one might prefer to use atorvastatin 80 mg daily in patients at particularly high risk of further ischaemic vascular events, in whom the absolute benefit of increasing the intensity of cholesterollowering treatment will be greatest. Cost may also influence the decision since, at least in the UK, atorvastatin is currently substantially more expensive than simvastatin. Sometimes, patients who are unable to tolerate one particular statin can tolerate another, and it is reasonable to use alternative statins to simvastatin or atorvastatin if necessary. Approximate equivalent doses to simvastatin 40 mg (in terms of LDL cholesterol lowering effect) of other statins are shown in Table 16.6. 110

16.4.7 When to start There is no specific evidence from RCTs for the use of statins in the very early period after an ischaemic stroke or TIA. However, impressive reductions in ischaemic vascular events, including stroke, have been reported with the use of statins within the first few days of an acute coronary syndrome, suggesting that early use after cerebrovascular events may also be beneficial.134 Furthermore, there is no reason to suppose that the potential adverse effects of statin treatment should be any greater early after an ischaemic stroke or TIA, while the high absolute risk of recurrence at this time should make the absolute benefits of treatment greater. An ongoing RCT of simvastatin 40 mg daily in patients who have had a TIA or minor ischaemic stroke within the previous

..

..

Table 16.6 Approximate equivalent doses to simvastatin 40 mg daily (in terms of LDL cholesterol lowering effect) of various other statins. (Data from Law et al., 2003.110) Statin regimen

Absolute (%) reduction in LDL cholesterol in mmol/L*

Simvastatin 40 mg daily Rosuvastatin 5 mg daily Atorvastatin 10 mg daily Lovastatin 40 mg daily Fluvastatin 80 mg daily Pravastatin 80 mg daily

1.8 (37%) 1.8 (38%) 1.8 (37%) 1.8 (37%) 1.6 (33%) 1.6 (33%)

*Standardized to pre-treatment serum LDL cholesterol concentration of 4.8 mmol/L. Summary estimates from 164 randomized placebo-controlled trials of statin vs placebo in 38 000 people.

24 h should provide some evidence on this issue (see http://www.strokecenter.org/trials/TrialDetail.aspx?tid= 483).7 In the meantime, we tend to start treatment with a statin as soon as the diagnosis is made of an ischaemic stroke or TIA with a baseline total cholesterol of > 3.5 mmol/L or LDL cholesterol > 2.6 mmol/L. All patients with a prior ischaemic stroke or transient ischaemic attack and a pre-treatment total cholesterol level of > 3.5 mmol/L or LDL cholesterol level of > 2.6 mmol/L should be considered for treatment with a statin. Both simvastatin 40 mg daily and atorvastin 80 mg daily have been shown to be beneficial in these patients. We generally start treatment as soon as the diagnosis has been confidently made.

16.4.8 The role of non-statin lipid-lowering drugs Because the evidence for statins is more extensive and compelling than for other lipid-lowering drugs, these should generally be the first choice of lipid-lowering agent. For patients with very abnormal lipid profiles (very high cholesterol, triglycerides, or both) which respond insufficiently to maximum statin dose, other lipidlowering drugs such as anion-exchange resins (which prevent reabsorption of bile acids, so increasing clearance of LDL cholesterol), ezetimibe (which inhibits the intestinal absorption of cholesterol), or – with great caution because of the increased risk of rhabdomyolysis – fibrates (which act mainly by decreasing triglyceride concentration) may be used in addition. This is generally best done under the supervision of a lipid specialist. For patients unable to tolerate a statin, and despite the lack

815

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 816

816

Chapter 16 Preventing recurrent stroke and other serious vascular events

of evidence from RCTs for a reduction in stroke, and the absence of large RCTs spefically in patients with a prior stroke or TIA, it seems reasonable to use an alternative drug to try to lower the LDL cholesterol concentration to some extent, since this seems very likely to reduce the risk of future vascular events.

16.5 Antiplatelet drugs

Platelets are involved in the pathophysiology not just of acute thrombosis in arteries and veins, but probably also in the process of atherogenesis itself (section 6.3). Antiplatelet drugs should, in theory, therefore reduce the risk not only of ischaemic stroke but also of myocardial infarction and other serious thrombotic events in both arteries and veins, such as intracranial venous thrombosis and pulmonary emboli. Their most important potential adverse effect is bleeding, most importantly intracranial haemorrhage because it is so frequently fatal or disabling, and serious extracranial haemorrhage (mainly from the gastrointestinal tract) leading to hospital admission and sometimes requiring blood transfusion or surgery. Therefore, probably the best outcome to assess the overall benefits of antiplatelet treatment is the composite outcome of ‘stroke, myocardial infarction and vascular death’ (whichever occurs first), which we will refer to as ‘serious vascular events’. This outcome includes not only the serious ischaemic events which antiplatelet treatment should prevent, but also the most important potential adverse effects, all intracranial as well as fatal extracranial haemorrhages. Because the number of events is greater, this composite outcome also provides more statistical power, and therefore precision, in estimating treatment effect than any of the individual components. However, if the number of outcome events is sufficient, it is not unreasonable to explore treatment effects on individual outcomes, such as stroke or myocardial infarction separately.135 There is also increasing interest in whether aspirin might reduce the risk of progressive cognitive impairment and vascular dementia in high-risk subjects.136 The most comprehensive evidence for the role of antiplatelet treatment in the prevention of serious vascular events in patients at high risk of vascular disease (including those with a previous ischaemic stroke or TIA) is the most recent systematic review of RCTs by the Antithrombotic Trialists’ Collaboration (ATT).137 Several large, relevant randomized trials have been completed since then, and these will also be considered below.

16.5.1 Antiplatelet treatment for long-term secondary prevention of serious vascular events Evidence from RCTs of immediate antiplatelet treatment (mainly aspirin 160–300 mg daily) in over 40 000 patients with acute ischaemic stroke, mainly randomized within 48 h of onset into one of two large trials, has shown a clear net benefit.137–139 This has been discussed in detail in section 12.3. Benefits in a wide range of high-risk patients The ATT overview included more than 140 000 patients at high risk of vascular disease in 195 RCTs comparing an antiplatelet regimen (mostly aspirin) with placebo or open control; antiplatelet treatment produced a highly significant relative reduction of about one-quarter in the risk of a serious vascular event among all types of high-risk patient, excluding those with acute ischaemic stroke for whom the relative reduction was somewhat smaller. The relative effects of were similar, regardless of whether the patients had a prior ischaemic stroke or TIA, a prior or acute myocardial infarction, stable or unstable angina, peripheral arterial disease, atrial fibrillation or some other high-risk condition (Fig. 16.21).137 In addition, previous analyses found that the relative reduction in vascular events was similar in males and females, diabetics and non-diabetics, older patients and younger patients, and in patients with and without hypertension.140 Benefits in patients with a previous ischaemic stroke or TIA Among the 20 000 or so patients with a prior ischaemic stroke or TIA, included in 21 RCTs, antiplatelet treatment produced a relative reduction of 22% in the odds of a serious vascular event (Fig. 16.21). This corresponded to the avoidance or delay of 36 serious vascular events (25 non-fatal strokes, six non-fatal myocardial infarctions and seven vascular deaths) per 1000 patients treated for about 3 years. Antiplatelet treatment was also associated with a reduction in all-cause mortality of about 15 per 1000 patients (Fig. 16.22).137 Risks in patients with a previous ischaemic stroke or TIA In the ATT overview, antiplatelet treatment produced about a one-fifth proportional increase in the risk of intracranial haemorrhage, and about half of these events were fatal. However, the absolute excess risk among patients with a prior ischaemic stroke or TIA was less than one per 1000 patients over about 3 years of treatment. This excess risk is accounted for in the estimate of effect on all serious vascular events (see above).

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 817

16.5 Antiplatelet drugs

No. (%) of vascular events Category of trial

No. of trials Allocated with data antiplatelet

Odds ratio (CI)

Adjusted Observed– control expected Variance

Previous myocardial infarction

12

1345/9984 (13.5)

1708/10 022 (17.0)

–159.8

567.6

25 (4)

Acute myocardial infarction

15

1007/9658 (10.4)

1370/9644 (14.2)

–181.5

519.2

30 (4)

Previous stroke/transient 21 ischaemic attack

2045/11 493 (17.8)

2464/11 527 (21.4)

–152.1

625.8

22 (4)

7

1670/20 418 (8.2)

1858/20 403 (9.1)

–94.6

795.3

11 (3)

Other high risk

140

1638/20 359 (8.0)

2102/20 543 (10.2)

–222.3

737.0

26 (3)

Subtotal: all except acute stroke

188

6035/51 494 (11.7)

7644/51 736 (14.8)

–715.7

2449.6

25 (2)

All trials

195

7705/71 912 (10.7)

9502/72 139 (13.2)

–810.3

3244.9

22 (2)

Acute stroke

0

Heterogeneity of odds reductions between: Five categories of trial: c2 = 21.4, df = 4; P = 0.0003 Acute stroke v other: c2 = 18.0, df = 1; P = 0.00002

2.0

1.5 Antiplatelet worse

results for all trials (and 95% confidence interval) is represented by open diamonds. Adjusted control totals have been calculated after converting any unevenly randomized trials to even ones by counting control groups more than once, but other statistical calculations are based on actual numbers from individual trials. SE, standard error. (From Antithrombotic Trialists’ Collaboration137 © 2002 BMJ Publishing Group.)

Adjusted % of patients (+1 SE)

P value:

6 (2) 0.0009

25 (5) 75 mg and < 75 mg in 3750 patients in the

..

..

ATT overview found no significant difference in the effects on vascular events (Fig. 16.25a);137 most of these patients were included in an RCT comparing 283 mg vs 30 mg daily in patients with a recent TIA or minor ischaemic stroke.163 However, because the confidence intervals were very wide, the possibility of a small (yet clinically important) difference in effectiveness between these doses of > 75 mg and < 75 mg cannot be excluded on the basis of these direct comparisons. Indirect comparisons of the effects of different doses compared with control are less reliable but the number of outcome events was much higher. Trials comparing daily aspirin doses of 500–1500 mg, 160–325 mg, 75– 150 mg and < 75 mg with control in high-risk patients (excluding those with acute ischaemic stroke) in the ATT overview revealed similar relative reductions in serious vascular events for the three higher daily dose ranges, but a somewhat smaller effect with < 75 mg daily (Fig. 16.25b).137 Another meta-analysis based on indirect comparisons between the effects of different daily aspirin doses on stroke found no definite evidence of differences between doses.164 Considering all the available evidence from direct and indirect comparisons in high-risk patients, it seems reasonable to conclude that aspirin at a dose of 75–150 mg daily is as effective as higher doses. Doses below 75 mg daily may be as effective, but this remains somewhat uncertain. Risks of different doses of aspirin A meta-analysis of trials in which participants were randomized to aspirin or control treatment found that aspirin produced a small increased risk of intracranial haemorrhage of about one per 1000 patients treated for 3 years (i.e. similar to the risk of antiplatelet treatment in general; section 16.5.1). There was no clear variation in risk with the dose.165 In RCTs directly comparing different daily doses, there were no significant differences in the risk of intracranial haemorrhage but the numbers of events were very small and the confidence intervals wide.137 In contrast, observational studies have suggested that the risk may be dose-related, but their methodological limitations prevent firm conclusions from being drawn.166,167 In the ATT overview, aspirin produced a small increased risk of major extracranial haemorrhage, similar to the risk for antiplatelet treatment in general (section 16.5.1). Both indirect and direct comparisons found similar risks of major extracranial haemorrhage with different daily doses.137 Another meta-analysis found the relative excess risk of gastrointestinal bleeding with aspirin to be about 70%, with no definite variation in

821

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 822

822

Chapter 16 Preventing recurrent stroke and other serious vascular events

(a) Category of trial

No. (%) of vascular events

Odds ratio (CI)

No. of trials Observed– with data Higher dose Lower dose expected Variance

Higher v lower aspirin doses: 500–1500 mg v 75–325 mg 7

Higher dose : Lower dose

% Odds reduction (SE)

227/1608 (14.1)

231/1589 (14.5)

–3.1

93.0

3 (10)

≥75 mg v 0.1). CI, confidence interval.

symptoms compared with medium dose (75–325 mg daily), and that medium dose (283 mg daily) is associated with a trend towards an increased odds of these symptoms compared with low dose (30 mg daily)137 (Table 16.7). Aspirin 75–150 mg daily is probably the most appropriate dose for long-term secondary prevention of serious vascular events to maximize benefits and to minimize adverse effects.

16.5.4 What to do with aspirin treatment ‘failures’ Because aspirin only prevents about one-quarter of vascular events that would otherwise occur, inevitably some patients will have events despite taking aspirin with or without other secondary preventive treatments. This may not necessarily represent a ‘failure’ of treatment, since patients who do have further events while on

..

..

aspirin might have had more or earlier events without aspirin. Nonetheless, there is increasing interest in the possibility of inter-individual variation in the response to aspirin, and a large number of laboratory measures of so-called ‘aspirin resistance’ are now available. These tests are not currently useful in clinical practice, since they have not been standardized and, although several studies have suggested an association between aspirin resistance and risk of subsequent vascular events, methodological problems limit the reliability of their results.171 Non-platelet sources of thromboxane A2 (particularly in inflammatory states) or altered thromboxane metabolism (e.g. in smokers) can alter thromboxane measurements in laboratory tests of aspirin resistance. Furthermore, although these laboratory measures may predict an increased risk of vascular events, it is not clear whether or not aspirin can still reduce this risk: There are a number of potential reasons why patients may continue to have events while taking aspirin.171 • Poor adherence. Up to 40% of patients with vascular disease do not adhere to their aspirin treatment, making

823

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 824

824

Chapter 16 Preventing recurrent stroke and other serious vascular events

this a common and often neglected reason for aspirin ‘failure’.171 Poor adherence may be as much of an issue with other secondary preventive medication, and this also needs to be explored.172 • Incorrect clinical diagnosis. In patients with repeated events despite the implementation of all reasonable secondary preventive measures, including regular aspirin, it is always worth revisiting the diagnosis. Sometimes the actual cause of recurrent events (e.g. focal seizures, migraine) only comes to light with follow-up, the occurrence of further episodes, and the emergence of additional pieces of key information from the patient or others (sections 3.4 and 3.5). • Non-atherothromboembolic pathology, such as arterial dissection, arteritis, or infective endocarditis. Not all vascular events are due to atherothromboembolism (section 6.2 and chapter 7),173 and furthermore the complex pathophysiology of atherothrombotic and embolic processes means that we would not expect aspirin to prevent all ischaemic events. • Inadequate aspirin dose. The evidence (section 16.5.3) from RCTs suggests that a daily dose of 75 mg daily is as effective as higher doses. However, this is based on meta-analyses of data from large numbers of patients and does not necessarily apply to each individual. It is uncertain whether individual patients’ dose requirements vary. Once other possible causes of treatment failure have been excluded, one reasonable, although non-evidence-based, strategy is to increase the dose of aspirin above 75–150 mg daily. • Interference from other drugs. There is some evidence that gastric acid suppression with a proton-pump inhibitor may impair aspirin absorption, and that concurrent intake of non-steroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen may inhibit binding of aspirin to platelet cyclo-oxygenase-1. However, data are conflicting and further studies are needed to assess these possibilities further. • Genetically determined reduced effect of aspirin. Polymorphisms in the genes encoding cyclo-oxygenase and other proteins that are involved in platelet pathways can modify the antiplatelet effects of aspirin, and may account for inter-individual differences in laboratory measures of aspirin resistance, and possibly in clinically evident aspirin treatment ‘failures’. • Increased turnover of platelets, for example in infection, inflammation or following surgery, may reduce the effectiveness of aspirin. Apart from increasing the dose of aspirin, other strategies used by some clinicians are to add another antiplatelet drug, such as modified release dipyridamole, to aspirin (if the patient is not already taking it), or even to replace aspirin with an alternative such as clopidogrel

(the evidence for alternative regimens is discussed in section 16.5.5). However, there have been no RCTs of alternative antiplatelet strategies among the specific group of patients who have events while taking aspirin. Neither is there any evidence that, for patients in sinus rhythm, oral anticoagulants alone, or in combination with aspirin, offer any definite net advantage over aspirin alone, and they can do harm (section 16.6.1). We would, however, consider the addition of clopidogrel to aspirin, or the substitution of antiplatelet treatment with heparin, sometimes followed by warfarin, in patients who are having very frequent TIAs on aspirin alone (sections 16.5.5 and 16.6.2). In patients who have further events while taking aspirin, it is important to check that the patient really is having atherothromboembolic episodes and, if so, that they are actually taking their aspirin and sticking to other secondary preventive strategies. It is reasonable to consider the addition of modified dipyridamole 200 mg twice daily if the patient is not already taking this, but there is no randomized evidence to support increasing the aspirin dose.

16.5.5 Alternative antiplatelet regimens to aspirin Aspirin acts on only one of a number of pathways leading to platelet activation and so thrombosis. Antiplatelet drugs acting through different pathways might therefore be more effective than aspirin if given as alternatives to, or combined with, aspirin (Fig. 16.24). However, any differences in effects on clinical outcomes between two antiplatelet regimens are likely to be small and so to detect such differences reliably, randomized comparisons would need to include very large numbers (i.e. tens of thousands) of patients. In the ATT overview, indirect comparisons provided no clear evidence of any differences in effects on serious vascular events,137 and in the direct comparisons the numbers of patients were generally too small to exclude a difference. However, several recent large trials have provided further information about alternative antiplatelet regimens. Thienopyridines vs aspirin The thienopyridine antiplatelet drugs (ticlopidine and clopidogrel) prevent adenosine diphosphate (ADP) binding to its receptor on platelets, so reducing ADPdependent activation of the glycoprotein IIb-IIIa complex which is the major receptor for fibrinogen on the platelet surface (Fig. 16.24).161 A systematic review of RCTs of a thienopyridine vs aspirin in high-risk patients identified 10 relevant trials

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 825

16.5 Antiplatelet drugs Fig. 16.26 Meta-analysis of randomized trials of a thienopyridine antiplatelet drug (ticlopidine or clopidogrel) vs aspirin in high vascular risk patients. Outcome is vascular events. Odds ratios (OR) for thienopyridine vs aspirin for each trial shown as squares, with area proportional to statistical weight, and 95% confidence intervals (CI) are represented by horizontal lines. Pooled odds ratio (and 95% CI) represented by diamond. AAASPS, African American Antiplatelet Stroke Prevention Study; CAPRIE, Clopidogrel versus Aspirin to Prevent Recurrent Ischemic Events; STAMI, Study of Ticlopidine versus Aspirin after Myocardial Infarction; TASS, Ticlopidine Aspirin Stroke Study. (Data of from Sudlow et al., in press.174)

Trial Ticlopidine vs aspirin AAASPS Japanese-B STAMI Sadowski Schoop TASS

Clopidogrel vs aspirin CAPRIE Total

in 26 865 patients (Fig. 16.26). Aspirin was compared with clopidogrel in one trial of 19 185 patients with ischaemic stroke, myocardial infarction or peripheral arterial disease, and with ticlopidine in the remaining nine trials in a total of 7633 patients, most of whom had a recent TIA or mild ischaemic stroke. The thienopyridines modestly and just significantly reduced the odds of a serious vascular event compared with aspirin (OR 0.92; 95% CI 0.85–0.99; 10 events avoided per 1000 patients treated, 95% CI 0–20 events per 1000), but the wide confidence interval included the possibility of no additional benefit. The result was similar, although not statistically significant, when only trials that randomized patients on the basis of their history of ischaemic stroke or TIA were considered.174 There have been several subsequent attempts to identify subgroups of patients likely to gain particular benefit from a thienopyridine (particularly clopidogrel) instead of aspirin, but in the face of an overall result of such marginal statistical significance, apparent differences in treatment effects between subgroups are highly likely to occur through the play of chance alone, and such analyses should be regarded with extreme caution.175,176 In terms of adverse effects, the systematic review found no significant difference between the thienopyridines and aspirin in intracranial or major extracranial haemorrhage. On the other hand, compared with aspirin, the thienopyridines were associated with a lower risk of gastrointestinal upset and haemorrhage. However, ticlopidine approximately doubled and clopidogrel increased by about one-third the risks of both skin rash and diarrhoea compared with aspirin. Ticlopidine, but not clopidogrel, was associated with an excess of neutropenia compared with aspirin, but there was no

..

..

Thienopyridine n/N

Aspirin n/N

128/902 8/170 18/734 19/154 0/31 370/1529

110/907 16/170 26/736 29/166 2/31 395/1540

976/9599

1063/9586

1519/13119 1641/13136

Peto OR (95% CI)

Heterogeneity between seven trials: c26df = 9.7, P = 0.1, I 2 = 38%

0.92 (0.85 to 0.99)

0.1 0.2 0.5 1 2 5 10 Favours thienopyridine Favours aspirin

significant difference between the thienopyridines and aspirin in the risk of thrombocytopenia.174 Ticlopidine has been associated with thrombocytopenia in observational studies, and both ticlopidine and clopidogrel have also been associated in post-marketing surveillance studies with thrombotic thrombocytopenic purpura, although this is very rare.177–182 The few trials which have compared the two thienopyridines directly have suggested better safety and tolerability with clopidogrel vs ticlopidine,183–185 which makes clopidogrel the thienopyridine of choice on safety grounds. Although, in the systematic review, the thienopyridines had a significantly lower rate of gastrointestinal upset and haemorrhage compared with aspirin, the most commonly used dose of aspirin in the trials of 325 mg was higher than the currently recommended daily dose of 75–150 mg, which is associated with less gastrointestinal adverse effects, and may compare more favourably with the thienopyridines.174 In summary, clopidogrel appears to be as effective as aspirin (and possibly somewhat more so), and is probably as safe, although it has a different adverse effect profile. The high cost of clopidogrel and the uncertainty of any additional benefit compared to aspirin make it unreasonable to suggest that it should replace aspirin as the first-choice antiplatelet drug for all patients at high vascular risk. It has been suggested that clopidogrel should be an alternative to aspirin in patients who are intolerant of or are allergic to aspirin. However we have no direct evidence of the relative effectiveness of thienopyridines compared with aspirin in this particular subgroup of patients as they were excluded from the RCTs. Furthermore, in an RCT in patients who developed peptic ulcer bleeding while taking aspirin to reduce

825

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 826

826

Chapter 16 Preventing recurrent stroke and other serious vascular events

vascular events, those who were assigned aspirin plus esomeprazole (a proton-pump inhibitor) had a statistically significant reduction in the cumulative incidence of recurrent ulcer bleeding in comparison to those treated with clopidogrel only. The combination of clopidogrel with a proton-pump inhibitor was not assessed.186 Thus, while clopidogrel still seems a reasonable alternative antiplatelet drug for patients who are genuinely allergic to aspirin, in patients who experience gastrointestinal problems with aspirin, current evidence provides more support for adding a proton-pump inhibitor than for replacing aspirin with clopidogrel. Clopidogrel appears to be as effective and as safe as aspirin, but is much more expensive. It is a reasonable alternative antiplatelet drug for patients with a history of ischaemic stroke or transient ischaemic attack, or for others at high risk of vascular events, who are genuinely allergic to aspirin. For patients unable to tolerate aspirin because of upper gastrointestinal symptoms, the addition of a proton-pump inhibitor is probably safer than changing to clopidogrel. Thienopyridines plus aspirin vs aspirin Because the thienopyridines act through a different mechanism to that of aspirin, their effects might be expected to be additive, making the combination of aspirin with a thienopyridine a theoretically attractive one, so long as any excess bleeding risks do not outweigh additional benefit. When the ATT antiplatelet trials database was completed in 1997, there was just one small RCT of aspirin plus a thienopyridine (ticlopidine) vs aspirin alone in about 1000 patients undergoing coronary artery stenting, and the results were inconclusive.137,187 Since then, several RCTs have assessed adding clopidogrel to aspirin in over 60 000 patients with acute coronary syndromes (with or without ST segment elevation) and/or those undergoing percutaneous coronary intervention. In this high-risk setting, the combination has shown definite reductions in serious vascular events compared with aspirin alone, although this is at the expense of a small increased risk of major (but not intracranial or life-threatening) haemorrhage.188–191 However, the same benefits of the combination of clopidogrel and aspirin have not been seen in long-term secondary prevention. The Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management and Avoidance (CHARISMA) randomized trial recruited over 15 000 patients with symptoms of ischaemic heart disease, ischaemic stroke, TIA, or peripheral arterial disease within the previous 5 years, or multiple atherothrombotic risk factors. There was no statistically

significant difference between treatment groups in the rate of serious vascular events, but the risk of moderate to severe bleeding was increased with combination treatment. There was a suggestion that patients with previous symptoms of vascular disease might achieve net benefit with combination treatment while asymptomatic patients with multiple risk factors might be harmed.192 However, this subgroup effect was one of many tested and could well have arisen purely by chance, and so cannot be regarded as anything more than hypothesis generating. Another RCT, the Management of AtheroThrombosis with Clopidogrel in High-risk patients (MATCH) trial, compared clopidogrel plus aspirin with clopidogrel alone in over 7000 high-risk patients with a recent (within 3 months) ischaemic stroke or TIA and found no significant effect on the combined outcome of ischaemic stroke, myocardial infarction, vascular death, or rehospitalization for acute ischaemia, but there was an increased risk of life-threatening bleeding with the combination compared with clopidogrel alone.193 Thus, there is currently no clear evidence to support the use of clopidogrel plus aspirin (rather than single antiplatelet therapy, for which the first choice is aspirin alone – see above) in patients with a prior ischaemic stroke or TIA, or other patients with stable previously symptomatic vascular disease. There are, however, some indications that the combination of clopidogrel and aspirin may ultimately prove to be of net benefit very early after ischaemic stroke or TIA, when the risk of recurrence is highest. In the MATCH trial, subgroup analyses suggested the possibility of net benefit from combination treatment in patients randomized within a week of symptom onset, and the overall negative result of MATCH may reflect the inclusion of a large proportion of patients with lacunar ischaemic events, who may benefit less from intensive antiplatelet treatment but be at higher risk of intracerebral bleeding compared to those with large artery disease.193 In addition, an RCT of clopidogrel plus aspirin vs aspirin alone in patients with recently symptomatic carotid stenosis and ongoing asymptomatic emboli detected by transcranial Doppler ultrasound found that the combination was more effective than aspirin alone in reducing asymptomatic emboli.18,194 Further RCTs assessing clopidogrel plus aspirin vs aspirin alone very early after a minor ischaemic stroke or TIA are either ongoing or planned (see http://www. strokecenter.org/trials/TrialDetail.aspx?tid=483).7 There is currently no clear evidence from randomized trials to support the use of the combination of clopidogrel plus aspirin to prevent vascular events in patients with ischaemic stroke or transient ischaemic attacks.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 827

16.5 Antiplatelet drugs

Dipyridamole vs aspirin Dipyridamole is a pyrimidopyridine derivative with both antiplatelet and vasodilator properties, whose mechanism of action on platelets remains a subject of controversy. Several possible antiplatelet actions have been observed in vitro, including inhibition of platelet phosphodiesterase, direct stimulation of the release of prostacyclin from endothelial cells, and inhibition of adenosine uptake by platelets. All of these putative mechanisms increase intraplatelet cyclic AMP, which inhibits the mobilization of free calcium, central to platelet activation (Fig. 16.24). However, although dipyridamole is widely accepted to be an antiplatelet drug, none of these actions has been demonstrated in vivo at the doses used in clinical practice.161 It is also a vasodilator, which is why it is used in diagnostic stress echocardiography and thallium imaging, and during rapid intravenous administration in these procedures it tends to cause hypotension.195 However, in a randomized comparison of aspirin vs aspirin plus 400 mg of oral dipyridamole daily in about 600 patients with recent cerebral ischaemia it did not appear to affect the blood pressure over the long term.196 The ATT meta-analysis of direct randomized comparisons between dipyridamole alone and aspirin alone in a total of about 3500 high-risk patients found no significant difference in the effect on serious vascular events (OR dipyridamole vs aspirin 1.02; 95% CI 0.85– 1.21).137 Since the largest body of evidence for the use of any single antiplatelet drug is for aspirin (Fig. 16.25b), and the wide confidence interval includes the possibility that dipyridamole is less effective than aspirin, this implies that dipyridamole alone should not generally be considered as an alternative to aspirin. Dipyridamole alone should not generally be considered as an alternative to aspirin for secondary prevention of serious vascular events in high-risk patients, including those with a history of ischaemic stroke or transient ischaemic attack.

Dipyridamole plus aspirin vs aspirin Dipyridamole plus aspirin was compared with aspirin alone in 25 RCTs in high-risk patients in the ATT overview. Overall, the combination produced a nonsignificant reduction in serious vascular events. When the separate components of the composite outcome were assessed, the combination appeared to be particularly effective in reducing non-fatal stroke, but not non-fatal myocardial infarction or vascular death.137

..

..

This non-fatal stroke result was derived mainly from one large study, the second European Stroke Prevention Study (ESPS-2), in which about 6000 patients with a previous ischaemic stroke or TIA were randomly assigned in a factorial design to aspirin 50 mg daily, modified-release dipyridamole 400 mg daily, both or neither.197 Dipyridamole plus aspirin significantly reduced the relative risk of vascular events by about one-fifth compared with aspirin alone, due to a significant reduction in stroke, with no detectable effect on either myocardial infarction or vascular death (although numbers of non-stroke outcomes were small (Fig. 16.27).137,198 Until further evidence emerged, it remained unclear whether the favourable results for stroke were explained by chance (since the number of patients and relevant outcome events was relatively small), the dose of aspirin (since 50 mg might be less effective than 75 mg or more daily – see above) or the particular dose and preparation of dipyridamole used.195 Analyses that included only trials with ischaemic stroke and TIA patients, or that considered only ESPS-2 (the only study to use the modified-release preparation of dipyridamole), suggested that the combination reduced vascular events compared with aspirin alone,197,199 but since these results were dominated by the stroke outcomes in ESPS-2, they too were subject to the possible effects of chance and the very low dose of aspirin. A further RCT, the European/Australasian Stroke Prevention in Reversible Ischaemia Trial (ESPRIT) has now compared aspirin (median daily dose 75 mg) plus dipyridamole (mainly modified release, 200 mg twice daily) vs aspirin alone in around 3000 patients with a prior ischaemic stroke or TIA of presumed arterial origin (patients with atrial fibrillation were excluded).198 Like ESPS-2, the combination produced a statistically significant one-fifth relative reduction in vascular events, arising from reductions in each of ischaemic stroke, cardiac events and vascular death (although none of these separate outcomes was independently statistically significant) (Fig. 16.27). ESPRIT also showed that benefits continued to accrue for up to 5 years, and there were no significant differences in the relative effects of treatment according to several characteristics, including age, gender, history of ischaemic heart disease and aspirin dose.198 An up-to-date meta-analysis of all RCTs comparing the combination of aspirin plus dipyridamole with aspirin alone in patients with a prior ischaemic stroke or TIA largely reflects the results of ESPS-2 and ESPRIT, shows a significant one-fifth relative reduction in vascular events (Fig. 16.27). In ESPRIT, the annual risk of vascular events in the aspirin-only arm was 5%, falling to 4% in the aspirin plus dipyridamole arm, and so adding modifiedrelease dipyridamole 200 mg twice daily to aspirin at this

827

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 828

828

Chapter 16 Preventing recurrent stroke and other serious vascular events

Weight (%)

Risk ratio (fixed) 95% CI

11/147 31/198 85/442 3/95 882 130

1.67 4.93 13.46 0.45 20.51

1.17 (0.53–2.56) 0.95 (0.60–1.51) 0.92 (0.70–1.21) 2.16 (0.56–8.37) 0.97 (0.78–1.22)

246/1650 1650 246

314/1649 1649 314

49.42 49.42

0.78 (0.67–0.91) 0.78 (0.67–0.91)

149/1363 1363 149

192/1376 1376 192

30.07 30.07

0.78 (0.64–0.96) 0.78 (0.64–0.96)

3907 636

100.00

0.82 (0.74–0.91)

Aspirin+dipyridamole (n/N)

Aspirin alone (n/N)

Pre-ESPS 2 Toulouse transient ischaemic attack12 12/137 AICLA11 30/202 ACCS10 79/448 Kaye13 6/88 Subtotal 875 Total events 127 Test for heterogeneity: c2 = 1.73, df = 3, P = 0.63, I2 = 0% Test for overall effect: Z = 0.24, P = 0.81 ESPS 2 ESPS 2 Subtotal Total events Test for heterogeneity: not applicable Test for overall effect: Z = 3.15, P = 0.002 ESPRIT ESPRIT Subtotal Total events Test for heterogeneity: not applicable Test for overall effect: Z = 2.39, P = 0.02

Total 3888 Total events 522 Test for heterogeneity: c2 = 4.31, df = 5, P = 0.51, I 2 = 0% Test for overall effect: Z = 3.61, P = 0.0003

Risk ratio (95% CI)

0.5 0.7 1 1.5 2 Favours aspirin+ Favours aspirin alone dipyridamole

Fig. 16.27 Meta-analysis of randomized trials of aspirin + dipyridamole vs aspirin alone in patients with a prior ischaemic stroke or TIA. Outcome = vascular events (stroke, myocardial infarction or vascular death). Purple squares represent trialspecific risk ratios with the area of each square proportional to the statistical weight of that study, and horizontal lines

denoting 95% confidence intervals (CI). Pooled risk ratios are shown as diamonds whose width denotes the 95% CI. ESPS 2, Second European Stroke Prevention Study; ESPRIT, European/ Australian Stroke Prevention in Reversible Ischaemia Trial. (Reprinted from The ESPRIT Study Group, Lancet 198 © 2006 with permission from Elsevier.)

level of risk should prevent about 10 vascular events per 1000 patients treated per year (i.e. 100 patients would have to be treated for 1 year to prevent one event).198

daily of modified-release dipyridamole separately (section 16.5.3). Neither ESPS-2 nor ESPRIT found any excess major bleeding in patients allocated the combination compared with aspirin alone. However, both reported an excess of premature cessations of treatment in the combination treatment arm, mainly due to adverse effects, particularly dipyridamole-induced headache, which may occur in up to one-third of people receiving dipyridamole but this usually settles in 1–2 weeks and might be reduced by dipyridamole dose titration.197,198,200 Since rapid intravenous injection of dipyridamole reduces blood pressure (see above), there have been anxieties about its use in patients with ischaemic heart disease.195 However, in ESPRIT, long-term oral dipyridamole did not affect blood pressure, and the benefits of adding dipyridamole to aspirin were similar in those with and without ischaemic heart disease.196,198

The addition of dipyridamole (modified-release formulation, 200 mg twice daily) to aspirin produces about a one-fifth reduction in the relative risk of vascular events in patients with a prior ischaemic stroke or transient ischaemic attack, corresponding to an average absolute reduction of about one event per 100 treated per year (in patients already receiving other proven secondary prevention treatments). Although a combined preparation of 200 mg modifiedrelease dipyridamole and 25 mg aspirin, designed to be taken twice daily, is available, most of the authors of this book are more comfortable with continuing a daily aspirin dose of 75–150 mg and adding 200 mg twice

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 829

16.5 Antiplatelet drugs

Patients should be warned of the risk of dipyridamolerelated headache, which may occur in up to one-third of people receiving dipyridamole, but it usually settles in 1–2 weeks. Triflusal vs aspirin Triflusal is structurally related to aspirin and exerts its antitplatelet effect mainly through inhibition of platelet cyclo-oxygenase (Fig. 16.24), but it is also thought to act on several other targets involved in platelet aggregation, and to increase nitric oxide synthesis, so giving it vasodilatory potential.201 A systematic review of RCTs of aspirin vs triflusal in high-risk patients identified five trials including a total of 5219 patients, more than half of whom had a history of ischaemic stroke or TIA. For serious vascular events in aggregate, there was no significant difference between the two drugs (OR 1.04; 95% CI 0.87–1.23), although the wide 95% confidence interval could not exclude a clinically important difference in favour of one or other drug. Aspirin was associated with a higher risk of major and minor haemorrhages and a lower risk of non-haemorrhagic gastrointestinal adverse events compared with triflusal.201 Although the lower risk of bleeding makes triflusal a potentially attractive alternative, it remains unclear whether it is definitely as effective as aspirin, and this will only be established by further, very large trials. Glycoprotein IIb/IIIa receptor antagonists plus aspirin vs aspirin While all the antiplatelet drugs discussed so far inhibit one of several possible pathways of platelet activation, the glycoprotein (GP) IIb/IIIa receptor antagonists prevent the binding of fibrinogen to the GP IIb/IIIa receptor, and so inhibit the final common pathway of platelet aggregation (Fig. 16.24). The GP IIb/IIIa receptor antagonists include monoclonal antibodies against the receptor, naturally occurring peptides isolated from snake venoms, synthetic peptides, and peptidomimetic and non-peptide molecules that compete with fibrinogen and other ligands for occupancy of the platelet receptor.161 RCTs in over 20 000 patients have demonstrated that the addition to standard antithrombotic therapy of a short intravenous infusion of a GP IIb/IIIa receptor antagonist immediately prior to percutaneous coronary intervention reduces the risk of subsequent ischaemic complications, albeit at the cost of a higher risk of major extracranial bleeding. RCTs have also assessed intravenous IIb/IIIa GP antagonists in over 30 000 patients with acute coronary syndromes without persistent ECG ST-segment elevation not undergoing percutaneous

..

..

intervention, and in smaller numbers of patients with ST-segment acute myocardial infarction receiving thrombolytic therapy. The balance of risks and benefits in these two groups of patients remains unclear.161,202,203 And pilot studies suggest that the intravenous GP IIb/IIIa receptor antagonists may have a role in the early treatment of acute ischaemic stroke, but far more data will be needed before the balance between benefits and risks can be determined reliably (section 12.6.1).204 Not surprisingly, the success of short-term high-grade blockade of the GP IIb/IIIa receptor with intravenous agents led to the development of oral GP IIb/IIIa receptor antagonists, in the hope that these may be beneficial in the long-term prevention of ischaemic vascular events in high-risk patients. Five large-scale RCTs involving over 40 000 patients, mainly with ischaemic heart disease but also including several hundred patients with prior ischaemic stroke, followed up for between 3 and 24 months, have now assessed these drugs. The results were very disappointing. There was no reduction in risk of recurrent ischaemic vascular events, but there was an increased risk of major haemorrhage and of death, which could not be entirely attributed to haemorrhage. The explanation for these results remains elusive, but may be related to the lower level of inhibition of the GP IIb/IIa receptor compared with the > 80% inhibition achieved with short-term intravenous administration. This may lead to paradoxical prothrombotic and proinflammatory effects of the orally administered drugs.161,205,206

16.5.6 How long should antiplatelet treatment continue? It is difficult to determine from the available RCTs how long treatment should continue, because there are no large directly randomized comparisons between different durations of treatment, few trials have assessed the effects of antiplatelet treatment for longer than 2 or 3 years, and none have assessed the effects of stopping treatment after a period of freedom from vascular events. In their earlier analyses, the ATT (then the Antiplatelet Trialists’ Collaboration) made indirect comparisons between the effects of treatment during sequential years of follow-up. Although there was an apparent pattern of greater effect during the earlier years, this may be explained by decreasing adherence to treatment over time, as well as bias in the comparisons in later years produced by the longer delay of vascular events in high-risk patients in the active treatment group. Therefore, without direct evidence to the contrary, it seems sensible to continue treatment indefinitely in patients who remain at high risk of a serious vascular event, unless some contraindication develops.140

829

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 830

Chapter 16 Preventing recurrent stroke and other serious vascular events

8%

Fatal intracranial haemorrhage

16.6 Anticoagulants

16.6.1 Anticoagulants for long-term prevention of vascular events in patients with a prior noncardioembolic ischaemic stroke or TIA Anticoagulants vs control A systematic review of all randomized controlled trials (RCTs) of all types of anticoagulants given early after acute ischaemic stroke found that, although they reduced the risk of early recurrent ischaemic stroke and pulmonary embolism, they increased the risk of intracranial and major extracranial haemorrhage, with no net benefit in terms of death or death and dependency at the end of follow-up.207 This evidence has been discussed in detail in section 12.4. Of more relevance here are the 11 RCTs comparing long-term treatment with an anticoagulant vs control in around 2500 patients in sinus rhythm with a previous presumed non-cardioembolic ischaemic stroke or TIA, followed up for about 2 years. These trials generally assessed an oral anticoagulant, either a coumarin, such as warfarin, or phenindione. Their methodological quality was generally poor; in particular, nine were completed in the 1960s and 1970s and so pre-dated routine CT scanning and the international normalized ratio (INR) to monitor anticoagulation. A meta-analysis of these trials found no significant difference between anticoagulants and control in the outcomes of serious

Major extracranial haemorrhage

6%

4% 40/604 (6.6%) 2%

20/618 (3.2%)

10/579 (1.7%)

7/596 (1.2%)

ol

A

nt ic

Co nt r

oa gu la nt s

ol Co nt r

oa gu la nt s

0%

nt ic

Both platelets and fibrin are important components of thrombi. Thus, inhibitors of the coagulation cascade, whose final step is the conversion of fibrinogen to fibrin, should reduce thrombotic and thromboembolic events. Anticoagulants first became available for clinical use in the 1950s and have been extensively used in patients with ischaemic stroke and TIA ever since, despite limited evidence from RCTs. As for antiplatelet drugs, the main risk of anticoagulation is major haemorrhage. Thus, as for antiplatelet treatment, the most appropriate outcome for assessing the balance of benefits and risks of anticoagulants is usually the combination of serious vascular events: stroke, myocardial infacrtion or vascular death. Given that the net benefits of antiplatelet treatment have been reliably demonstrated in a wide range of highrisk patients, including those with a prior ischaemic stroke or transient ischaemic attack (TIA) (section 16.5.1), the main question of clinical relevance today is: are there any patients for whom anticoagulants given alone or combined with antiplatelet treatment confer greater net benefit than antiplatelet treatment alone?

A

830

Fig. 16.28 Randomized trials of anticoagulants vs control in patients with a prior presumed non-cardioembolic ischaemic stroke or transient ischaemic attack. Percentage of patients randomized experiencing fatal intracranial haemorrhage and major extracranial haemorrhage during about 2 years of followup. (Data from Sandercock et al., 2003.208)

vascular events, death and dependency, death, or recurrent ischaemic stroke. However, anticoagulants statistically significantly increased fatal intracranial haemorrhages more than twofold, and fatal extracranial haemorrhages more than threefold, equivalent to anticoagulants causing about 11 additional fatal intracranial haemorrhages and 25 additional major extracranial haemorrhages per year for every 1000 patients treated (Fig. 16.28).208 Anticoagulants vs antiplatelet treatment Six RCTs have compared various different intensities of oral anticoagulation (mainly warfarin) with antiplatelet treatment (mainly aspirin) in over 5000 patients with non-cardioembolic ischaemic stroke or TIA followed for 1–2 years.209–211 Over 90% of these patients were included in one of the four largest trials, and the results for outcomes that were recorded in all (or at least three) of these are shown in Fig. 16.29. The target INRs in the warfarin arms of the three trials were low (INR 1.4–2.8), medium (INR 2.0–3.0) and high (INR 3.0–4.5).210–213 In none of the trials was there any evidence of net benefit from anticoagulants, but there was a statistically significantly increased risk of major haemorrhage and of death with both moderate and high-intensity anticoagulants. The excess risk of major haemorrhage appeared to be

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 831

16.6 Anticoagulants

Warfarin n/N

Aspirin n/N

Peto OR (95% CI)

Recurrent ischaemic stroke or death WARSS (INR 1.4 –2.8)

196/1103

176/1103

1.14 (0.91–1.42)

WASID (INR 2.0 –3.0)

72/289

64/280

1.12 (0.76–1.65)

SPIRIT (INR 3.0 – 4.5)

49/651

31/665

1.65 (1.05–2.59)

Recurrent stroke, myocardial infarction or vascular death WASID (INR 2.0 –3.0)

71/289

67/280

1.04 (0.71–1.52)

ESPRIT (INR 2.0 –3.0)

79/536

92/532

0.83 (0.60–1.15)

SPIRIT (INR 3.0 – 4.5)

55/651

33/665

1.75 (1.13–2.69)

WARSS (INR 1.4 –2.8)

38/1103

30/1103

1.27 (0.79–2.07)

WASID (INR 2.0 – 3.0)

24/289

9/280

2.53 (1.26–5.12)

ESPRIT (INR 2.0 – 3.0)

45/536

18/532

2.46 (1.48–4.10)

SPIRIT (INR 3.0 – 4.5)

53/651

6/665

5.42 (3.21–9.13)

WARSS (INR 1.4 –2.8)

7/1103

5/1103

1.40 (0.45–4.35)

WASID (INR 2.0 – 3.0)

2/289

1/280

1.89 (0.20–18.26)

ESPRIT (INR 2.0 – 3.0)

45/536

18/532

2.46 (1.48–4.10)

SPIRIT (INR 3.0 – 4.5)

17/651

1/665

6.19 (2.44–15.69)

WARSS (INR 1.4 –2.8)

47/1103

53/1103

0.88 (0.59–1.32)

WASID (INR 2.0 – 3.0)

28/289

12/280

2.28 (1.20–4.34)

ESPRIT (INR 2.0-3.0)

59/536

44/532

1.37 (0.91–2.05)

SPIRIT (INR 3.0 – 4.5)

35/651

15/665

2.35 (1.33–4.13)

Major haemorrhage

Fatal haemorrhage

Death

0.2 0.5 1 2 5 Favours warfarin Favours aspirin Fig. 16.29 Meta-analysis showing various outcomes from large randomized trials of warfarin vs aspirin in patients with a prior presumed non-cardioembolic ischaemic stroke or transient ischaemic attack. Odds ratios (OR) for each outcome are shown for the individual trials as squares, whose size is proportional to the statistical weight of the trial for the outcome, with horizontal lines denoting 95% confidence intervals (CI). INR, international normalized ratio. WARSS, Warfarin

..

..

10

Aspirin Recurrent Stroke Study; WASID, Warfarin-Aspirin Symptomatic Intracranial Disease trial; ESPRIT, European/ Australasian Stroke Prevention in Reversible Ischaemia Trial; SPIRIT, Stroke Prevention In Reversible Ischemia Trial. (Data from The Stroke Prevention In Reversible Ischemia Trial [SPIRIT] Study Group, 1997;212 Mohr et al., 2001;213 Chimowitz et al., 2005;210 The ESPRIT Study Group, 2007.211)

831

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 832

832

Chapter 16 Preventing recurrent stroke and other serious vascular events

dose-related (Fig. 16.29). Although the Warfarin Aspirin Recurrent Stroke Study (WARSS) trial of low-intensity anticoagulation vs aspirin found no significant difference between treatments in major haemorrhage, there was a significant 60% increase in the relative odds of minor haemorrhage.213 Further analyses of data from patients in the Stroke Prevention in Reversible Ischemia Trial (SPIRIT) and patients with non-disabling cerebral ischaemia with atrial fibrillation in the European Atrial Fibrillation Trial (EAFT) found that intensity of anticoagulation, cerebral ischaemia of arterial rather than cardioembolic origin, presence of leukoaraiosis on the pre-treatment brain scan, and age over 65 years were independent predictors of increased risk of anticoagulantrelated major haemorrhage.214 There is therefore no intensity of anticoagulation at which the beneficial effect of reducing ischaemic events suggested by some of the trials exceeds the increased risk of major haemorrhage in patients with non-cardioembolic ischaemic stroke or TIA.211,213 Anticoagulants should not be used for the long-term prevention of recurrent stroke and other vascular events after a presumed non-cardioembolic ischaemic stroke or transient ischaemic attack, because they increase the risk of haemorrhage compared with either control or aspirin without any clear evidence of net benefit.

16.6.2 Anticoagulants for long-term prevention of vascular events in patients with a prior ischaemic stroke or TIA in atrial fibrillation Atrial fibrillation (AF) is an important risk factor for stroke, mainly ischaemic cardioembolic stroke, increasing risk about fivefold (section 6.5.1).215,216 However, the absolute risk of stroke varies 20-fold among AF patients depending on their age, risk factors and history of vascular disease. A history of stroke or TIA is the most consistent and powerful independent predictor of stroke, increasing the absolute risk to around 10% per year in patients receiving aspirin but no anticoagulants. Other consistent independent predictors of stroke risk are age (increasing risk by about 1.5 times per decade) and raised blood pressure. Diabetes and female gender may also independently increase the risk of stroke. Many other potential predictors have been studied but the results have not been consistent. Various risk stratification schemes have now been developed to allow patients with AF to be grouped into those with low, moderate, high or very high risk of subsequent stroke (Table 16.8). Along with the evidence from RCTs (see below and section 16.5.1), and an assessment of the patient’s likely

Table 16.8 Risk stratification scheme for patients in atrial fibrillation. (Adapted from Hart and Halperin, 1999218 and Gage et al., 2004.217) Very high risk (~12% per year untreated risk of stroke) Previous ischaemic stroke or transient ischaemic attack High risk (~5 to 8% per year untreated risk of stroke) Age > 65 years and hypertension, diabetes, heart failure, left ventricular dysfunction High-risk cardiac disease, e.g. valvular heart disease or recent myocardial infarction Moderate risk (~3 to 5% per year untreated risk of stroke) Age > 65 years with no other risk factors Age < 65 years and hypertension or diabetes Low risk (~1% per year untreated risk of stroke) Age < 65 years and no other risk factors

risk of adverse effects of treatment, these help to inform the best choice of antithrombotic treatment for individual patients.217,218 Our recommendations (below) for treatment of patients with AF apply also to patients with paroxysmal atrial fibrillation or atrial flutter, because the available evidence suggests that they have similarly increased risks of stroke.219 Oral anticoagulants vs control Systematic reviews of RCTs of adjusted-dose warfarin (with an average achieved INR of around 2.5) vs control in almost 3000 patients with AF have shown that oral anticoagulation produces a clear and consistent reduction in the risk of stroke of around two-thirds, and of all serious vascular events of around one-half (Fig. 16.30). The relative reductions were similar in those with and without a previous history of ischaemic stroke or TIA. However, control group patients with a history of ischaemic stroke or TIA had an absolute stroke risk of 12% per year, compared with < 5% per year in those without such a history, so the absolute reduction in stroke risk in the secondary prevention trials was much higher than in the primary prevention trials (>8% vs about 3% per year). Excess risks of major haemorrhage were small, about 2 per thousand per year for intracranial haemorrhage and about 3 per thousand per year for major extracranial haemorrhage.144,220,221 Because the RCTs excluded large numbers of patients with a variety of contraindications to anticoagulants (e.g. those with a history of gastrointestinal bleeding, at risk of falls, likely poor adherence to treatment, alcoholism, etc.), concerns have persisted about the generalizability of the results to everyday clinical practice. Reassuringly, observational studies of AF patients in non-trial clinical settings have confirmed their relative

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 833

(a)

Warfarin n/N

Control n/N

Peto OR (95% CI)

AFASAK

18/315

30/315

BAATAF

12/205

24/201

CAFA

12/181

13/184

SPAF1

9/193

20/194

SPINAF-primary prevention

18/260

31/265

Subtotal

69/1154

118/1159

8/21

11/25

EAFT

43/225

67/214

Subtotal

51/246

78/239

0.55 (0.37–0.82)

120/1400

196/1398

0.56 (0.44–0.71)

No prior ischaemic stroke or TIA

0.57 (0.42–0.77)

Prior ischaemic stroke or TIA SPINAF-secondary prevention

Total (serious vascular events)

Heterogeneity between trials: c26df = 2.6, P = 0.9, I 2 = 0%

0.1

0.2

0.5

1

Favours warfarin

2

5

10

Favours control

Warfarin n/N

Control n/N

AFASAK

6/315

17/315

BAATAF

3/205

11/201

CAFA

6/181

9/184

SPAF1

7/193

15/194

SPINAF-primary prevention

5/260

19/265

27/1154

71/1159

2/21

4/25

EAFT

20/225

50/214

Subtotal

22/246

54/239

0.36 (0.22–0.58)

Total (all stroke)

49/1400

125/1398

0.38 (0.28–0.52)

(b)

Peto OR (95% CI)

No prior ischaemic stroke or TIA

Subtotal

0.39 (0.26–0.59)

Prior ischaemic stroke or TIA SPINAF-secondary prevention

Heterogeneity between trials: c26df = 2.3, P = 0.9, I 2 = 0%

0.1

0.2

0.5

Favours warfarin Fig. 16.30 Meta-analysis showing effects on (a) serious vascular events and (b) all stroke (ischaemic and haemorrhagic) in randomized trials of adjusted-dose warfarin vs control in patients with non-rheumatic atrial fibrillation, with and without a history of ischaemic stroke or transient ischaemic attack (TIA). Odds ratios (OR) for each trial are shown as squares, whose size is proportional to the statistical weight of the trial for the relevant outcome, with horizontal lines denoting 95% confidence intervals (CI). Pooled results for

..

..

1

2

5

10

Favours control

subgroups and overall are shown as diamonds (with 95% CI). AFASAK, Copenhagen Atrial Fibrillation, Aspirin and Anticoagulation Study; BAATAF, Boston Area Anticoagulation Trial for Atrial Fibrillation; CAFA, Canadian Atrial Fibrillation Anticoagulation Study; SPAF, Stroke Prevention in Atrial Fibrillation Study; SPINAF, Stroke Prevention in Nonrheumatic Atrial Fibrillation; EAFT, European Atrial Fibrillation Trial. (Data from Hart et al., 1999;144 Aguilar and Hart, 2005;220 Saxena and Koudstaal, 2004.221)

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 834

834

Chapter 16 Preventing recurrent stroke and other serious vascular events

Warfarin n/N

Aspirin n/N

AFASAK1

9/335

16/336

AFASAK2

11/170

9/169

PATAF

3/131

4/141

SPAF2

39/555

42/545

Subtotal

62/1191

71/1191

EAFT

20/225

52/230

Subtotal

20/225

52/230

0.36 (0.22–0.59)

Total

82/1416

123/1421

0.64 (0.48–0.86)

Peto OR (95% CI)

No prior ischaemic stroke or TIA

0.86 (0.60–1.22)

Prior ischaemic stroke or TIA

Heterogeneity between five trials: c24df = 9.6, P = 0.05, I 2 = 58%

0.1

0.2

0.5

Favours warfarin

1

2

5

10

Favours aspirin

Fig. 16.31 Meta-analysis showing effects on stroke in randomized trials of warfarin vs aspirin in patients with nonrheumatic atrial fibrillation, with and without a history of ischaemic stroke or transient ischaemic attack (TIA). Odds ratios (OR) for each trial are shown as squares, whose size is proportional to the statistical weight of the trial for the relevant outcome, with horizontal lines denoting 95% confidence intervals (CI). Pooled results for subgroups and overall are

shown as diamonds (with 95% CI). AFASAK 1 and 2, first and second Copenhagen Atrial Fibrillation, Aspirin and Anticoagulation Studies; PATAF, Primary Prevention of Arterial Thromboembolism in Nonrheumatic Atrial Fibrillation Study; SPAF 2, second Stroke Prevention in Atrial Fibrillation Study; EAFT, European Atrial Fibrillation Trial. (Data from Hart et al., 1999;144 van Walraven et al., 2002;225 Saxena and Koudstaal, 2004.221)

effectiveness, and have generally shown rates of major haemorrhage similar to the RCTs.222,223 However, highquality monitoring and management of anticoagulation are required to maintain the net benefit and low rate of adverse effects. Of note is the fact that the very elderly, who face the highest risk of both stroke and haemorrhagic complications, were not well represented in the RCTs, and so further studies in such patients are required to better assess the balance of risks and benefits of oral anticoagulants.219,224

The pooled results (whether or not the indobufen study is included) suggest that oral anticoagulants are more effective at preventing stroke than aspirin or other antiplatelet treatment, with a relative risk reduction of about one-third, overall (Fig. 16.31).144,221,225 There was statistically significant heterogeneity between the results of the individual trials, and the relative benefits of treatment appeared somewhat greater in those with a prior ischaemic stroke or TIA. Because of their substantially higher risk of stroke on aspirin (see above), the absolute reduction in stroke risk with adjusted-dose warfarin vs aspirin was much higher for patients with a history of prior ischaemic stroke or TIA vs those without such a history (7% per year vs 0.4% per year). The risk of any major haemorrhage (extracranial or intracranial) was increased by about twofold, an absolute increase of about 1% per year, and appeared not to vary substantially between groups with different levels of baseline risk of stroke. Mortality was similar in the warfarin and aspirin or antiplatelet treatment groups.144,218,221,225

Oral anticoagulants vs antiplatelet treatment Adjusted-dose warfarin (with an average achieved INR of around 2.5) has been directly compared with aspirin in five trials among about 3000 patients with AF, one of them among patients with a prior ischaemic stroke or TIA. One further RCT compared adjusted-dose warfarin with the antiplatelet drug, indobufen, a reversible inhibitor of cyclo-oxygenase.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 835

16.6 Anticoagulants

The superiority of adjusted-dose warfarin over antiplatelet treatment is strengthened by the results of the Atrial Fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events-W (ACTIVE-W trial). This trial randomized 6700 patients with AF and one other risk factor to either warfarin with a target INR of 2–3 or the combination of clopidogrel and aspirin. About 1000 participants had a history of stroke or TIA. The trial was stopped early when an interim analysis showed clear superiority of adjusted-dose warfarin, with highly significant relative reductions of one-third in stroke and in the primary outcome of stroke, systemic embolism, myocardial infarction or vascular death.226 In the types of patients who were randomized in these trials and in the trials of anticoagulants vs control (see above), it is clear that the benefits of adjusted-dose warfarin with a target INR of around 2.5 far outweigh the risks in patients with a history of an ischaemic stroke or TIA, so long as there are no contraindications to anticoagulants. Unfortunately, many patients do have such contraindications; factors that may make clinicians reluctant to prescribe long-term anticoagulants include uncontrolled hypertension, recent gastrointestinal bleeding, alcoholic liver disease, confusion or dementia, tendency to falls, difficulties with access to an anticoagulant clinic, and extensive leukoaraiosis on the brain scan.227 These factors increase the likelihood of over-anticoagulation and major bleeding.224 For these patients, aspirin is a reasonable alternative

antithrombotic treatment, albeit less effective (section 16.5.3). Patients with a prior ischaemic stroke or transient ischaemic attack and atrial fibrillation are at very high risk of subsequent stroke. They should generally be treated with oral anticoagulants or, if anticoagulation is contraindicated, with aspirin (or some other antiplatelet regimen if aspirin cannot be used for some reason). For patients without a history of ischaemic stroke or TIA, the balance of benefits of warfarin on the one hand, and the inconvenience of monitoring and risks of major bleeding on the other, need to be considered on an individual patient basis. There are several risk stratification schemes to help with this process. They vary somewhat, but all use a combination of clinical variables (age, blood pressure, diabetes, and sometimes heart failure, other cardiac disease, and echocardiographic findings) to assign the patient to low, moderate, high or very high-risk groups (Table 16.8). Table 16.9 shows the estimated annual risks of stroke in untreated patients in these groups along with the estimated absolute reductions in risk of stroke and increases in risk of major extracranial bleeding with adjusted-dose warfarin or aspirin for patients in the various baseline risk categories: • In low-risk patients (under 65 years without any risk factors other than AF) the absolute benefits and risks of

Table 16.9 Approximate size of main treatment benefits and harms according to risk of stroke in patients with atrial fibrillation. (Adapted from Hart et al., 1999144 and Hart and Halperin, 1999.218) Stroke in patients without a prior ischaemic stroke or transient ischaemic attack

Untreated risk (% per year)

Low risk Moderate risk High risk 1% 3 to 5% 5 to 8% Events avoided (↓) or produced (↑) per 1000 patients treated per year with: Adjusted dose warfarin ↓6 ↓18 to 30 ↓30 to 48 vs no treatment† (NNT 167) (NNT 33 to 55) (NNT 21 to 33) Aspirin vs no treatment‡ ↓2 ↓6 to 10 ↓10 to 16 (NNT 500) (NNT 100 to 167) (NNT 62 to 100) Adjusted dose warfarin ↓4 ↓12 to 20 ↓20 to 32 vs aspirin¶ (NNT 250) (NNT 50 to 83) (NNT 31 to 50)

Stroke in patients with a prior ischaemic stroke or transient ischaemic attack

Major extracranial haemorrhage§

Very high risk 12%

0.6%

↓72 (NNT 14) ↓24 (NNT 42) ↓48 (NNT 21)

↑3 (NNH 333) ↑1 (NNH 1000) ↑2 (NNH 500)

†Relative risk reduction for stroke = 60%. ‡Relative risk reduction for stroke = 20%. ¶Relative risk reduction for stroke = 40%. §Based on estimates from relevant systematic reviews of randomized trials. Bleeding risks may be higher in patients outside clinical trials, particularly the elderly. NNT, number-needed-to-treat (number treated to avoid one stroke). NNH, number-needed-to-harm (number treated to produce one major extracranial bleed).

..

..

835

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 836

836

Chapter 16 Preventing recurrent stroke and other serious vascular events

warfarin are very finely balanced, and aspirin seems a better option because of its lower risk and ease of use. • For patients without a history of ischaemic stroke or TIA but at high risk (over the age of 65 years and with at least one other risk factor) or for those with a history of prior ischaemic stroke or TIA, the absolute reduction in stroke with warfarin (compared with either control or with aspirin) is substantial, and outweighs the absolute bleeding risks, making anticoagulants the appropriate choice of antithrombotic treatment in most cases, provided there are no contraindications. • For patients at moderate risk (over the age of 65 years, or under 65 with risk factors other than AF), either adjusted-dose warfarin or aspirin may be justified, depending on the patient’s preferences, the inconvenience of anticoagulant intensity monitoring, and likely risk of bleeding complications, which may exceed the average risks derived from clinical trials shown in Table 16.9, particularly in the elderly. An ongoing RCT should provide further information on the balance of risks and benefits of anticoagulation in the elderly.228 The risks of stroke and of anticoagulant-related bleeding should be assessed for individuals with atrial fibrillation. In the absence of contraindications, anticoagulants are appropriate for those at high risk of stroke, while aspirin is appropriate for those at low risk. Either anticoagulants or aspirin may be appropriate for moderate-risk individuals, depending on the balance of risks and benefits, and patient preference. Different anticoagulant intensities Several RCTs have assessed very low intensities and/or fixed doses of anticoagulants in an attempt to reduce the risk of bleeding and the inconvenience of regular monitoring for adjusted-dose anticoagulation. Two have compared adjusted-dose warfarin (with an average achieved INR of about 2.5) with aspirin plus low, fixed doses of warfarin (1–3 mg, mean achieved INR about 1.3) in about 1400 patients with AF, over one-third of whom had a history of stroke or TIA. The first and largest of these trials recruited patients at high risk of stroke, and was stopped early after an interim analysis found large reductions in relative and absolute risk for adjusted-dose warfarin, without a significant difference in the risk of major bleeding. The second trial was also stopped early in response to the results of the first, and found no difference between treatment arms. A further three RCTs have compared adjusted-dose warfarin (target INRs of 2–3.5) with low or fixed doses of warfarin alone (mean achieved INR 1.1–1.4) in almost 900 patients without a prior

stroke or TIA. No significant difference was found between treatments but there was a non-significant reduction in stroke with adjusted-dose warfarin, in keeping with the results of the trials comparing adjusted-dose warfarin with aspirin plus low-intensity warfarin.144,219 Data from observational studies support those from the RCTs, and suggest that a target INR of about 2.5 is optimal. Below about 2.0, the risk of ischaemic stroke increases and strokes that do occur are more likely to be severe or fatal, while above about 3.0 there is an increased risk of major bleeding. Patients over 75 years old are at greater risk of major bleeding on oral anticoagulants, but are also at increased risk of stroke, and there is currently no evidence to suggest that the target INR should be lower in them.219,224,229–231 The addition of antiplatelet treatment to anticoagulants Two RCTs have assessed adding antiplatelet treatment to adjusted-dose warfarin with a slightly lower than standard INR range compared with standard adjusted-dose warfarin. One small trial in 157 AF patients at high risk of stroke compared the oral anticoagulant fluindione (target INR 2.0–2.6) alone or combined with aspirin 100 mg daily. It was stopped early because of an increased rate of major haemorrhage in the combination group, and the number of events accrued was too small to assess the effect on stroke or other vascular events.232 The second study comprised two parallel RCTs, one comparing oral anticoagulants (mean achieved INR 2) plus the antiplatelet drug triflusal 600 mg daily vs oral anticoagulants alone (mean achieved INR 2.5) in around 500 high-risk patients with AF and rheumatic mitral stenosis or a history of embolism, and the other comparing triflusal alone vs oral anticoagulants (mean achieved INR 2.5) alone vs triflusal plus oral anticoagulants (mean achieved INR 2) in around 700 intermediate-risk patients with AF and aged over 60 years, with hypertension or heart failure. The primary outcome of vascular death and non-fatal stroke or systemic embolism was significantly reduced in the combination treatment arm compared with anticoagulants alone in both high- and intermediate-risk groups. Rates of severe bleeding were slightly lower in the combination treatment group compared with anticoagulants alone but numbers of events were small and the difference was not significant.233 This result raises the possibility that the combination of oral anticoagulation with an INR of about 2 with antiplatelet treatment may be more effective and safer than oral anticoagulants alone with an INR of about 2.5.219 Further trials are needed to explore this possibility further.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 837

16.6 Anticoagulants

A systematic review of six RCTs comparing aspirin combined with warfarin vs the same intensity of warfarin alone in almost 4000 patients with ischaemic heart disease or prosthetic heart valves found that the addition of aspirin approximately doubled the risk of intracranial haemorrhage.234 For most categories of patients, and especially those already at high risk of bleeding with anticoagulants, the combination should be avoided. However, in some categories of patients at particularly high risk of thromboembolism (e.g. those with mechanical prosthetic valves), the greater antithrombotic benefit of the combination may outweigh the extra bleeding risk.137,147,187 The combination of oral anticoagulation (target INR 2.5) with aspirin or another antiplatelet drug should generally be avoided because of the increased risk of major bleeding. However, the greater benefit of the combination may outweigh the bleeding risk in some patients at particularly high risk of thromboembolism (e.g. those with mechanical heart valves). Alternative anticoagulants to warfarin Warfarin and other vitamin K antagonists are associated with an increased risk of major bleeding, and require regular, careful monitoring of anticoagulation intensity (through measurement of the INR) with appropriate adjustment of the dose as required. This represents a significant burden of cost and time, both for health services and for patients. Indeed, several studies have shown that warfarin is taken by only one- to two-thirds of appropriate patients with AF, at least in part because physicians tend to be influenced more by major adverse bleeding associated with warfarin in their patients than by ischaemic strokes in those with AF not taking warfarin.235 The development of an oral anticoagulant that is as effective as warfarin, and as safe or safer, and without the requirement for regular monitoring, would be a significant advance. The orally administered anticoagulant, ximelagatran, showed considerable promise in this regard. It is a prodrug form of the direct thrombin inhibitor, melagatran, which is rapidly absorbed in the small intestine and then bioconverted to its active form. Importantly, its anticoagulant effect is predictable at a fixed dose, and there have been no reported drug or food interactions, making monitoring of anticoagulation intensity unnecessary. There have been a number of RCTs comparing it with either low-molecular-weight heparin, adjusted-dose warfarin, or both in a total of around 14 000 patients in several different settings: prevention of venous

..

..

thromboembolism after orthopaedic surgery, treatment of patients with venous thromboembolism, acute myocardial infarction, and atrial fibrillation. These generally found either increased or similar effectiveness and similar or reduced bleeding risks with ximelagatran.236 Pooled results of the two large RCTs comparing ximelagatran with adjusted-dose warfarin (target INR 2–3) in 7300 patients with non-valvular AF and at least one additional risk factor for stroke found no significant difference between treatments in the risk of the primary outcome of stroke or systemic embolism (although the confidence interval included the possibility of a clinically important difference in either direction), and significantly less major bleeding in the ximelagatran group.237 Unfortunately, there have been concerns about the potential hepatotoxicity of ximelagatran, with the RCTs showing about a 6% absolute excess of asymptomatic elevations in transaminases after 1–2 months, and two deaths in the ximelagatran group in one of the AF trials, one from hepatitis with necrosis and the other from gastrointestinal haemorrhage in a patient with markedly elevated transaminases.236,237 Despite these concerns, ximelagatran received marketing approval from France in December 2003, and then from the European Medicines Agency in May 2004 for short-term use in the prevention of venous thromboembolic events after orthopaedic surgery, but the United States Food and Drug Administration withheld marketing approval in October 2004. In February 2006, AstraZeneca withdrew ximelagatran from the market and terminated its development, because of further concerns about hepatotoxicity arising during a trial of ximelagatran in extended venous thromboembolism prophylaxis up to 35 days after orthopaedic surgery (http://www.astrazeneca.com/ archivelist/). Thus, the search is still on for an alternative to warfarin and other vitamin K antagonists, which for the time being are the only available oral anticoagulants of proven effectiveness. Other anticoagulant molecules are in development and there are planned or ongoing trials of long-acting subcutaneous heparinoids, and of oral factor Xa antagonists and direct thrombin inhibitors (see http://www.clinicaltrials.gov/ct/show/NCT00070655, http://www.clinicaltrials.gov/ct/show/NCT00412984, http://www.clinicaltrials.gov/ct/show/NCT00262600).219 Anticoagulants for patients with repeated transient ischaemic attacks or ischaemic strokes despite optimum antiplatelet drug therapy In our experience, this situation is relatively uncommon. The first step should be to verify again that the attacks are truly ischaemic in nature and are not due to

837

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 838

838

Chapter 16 Preventing recurrent stroke and other serious vascular events

structural brain disease of some sort, low-flow ischaemic events, epilepsy, migraine, psychogenic disorder, etc. (sections 3.4, 3.5 and 6.7.5). If the diagnosis is correct, these admittedly rare patients are often very worried, as are their doctors. Of course, those with carotid ischaemic events and severe carotid stenosis should be considered for endarterectomy, but sometimes this is not possible or practical (section 16.11.6), and very occasionally the attacks continue after what appears to be a technically adequate operation. Two or three months of moderateintensity oral anticoagulants (INR 2–3) may perhaps reduce the frequency and severity of the TIAs, which is comforting for the patient and reduces anxiety all round. Sometimes, if the attacks are very frequent, it may be necessary to admit the patients to hospital for unfractionated or low-molecular-weight heparin in the first instance. We do this in the full knowledge that this treatment is not based on evidence from RCTs and may not reduce the risk of stroke or vascular death; it is purely given for symptom control. It is often possible, after a few weeks, to withdraw the anticoagulants gradually and replace them with aspirin or some other antiplatelet regimen. In such cases, the unstable atheromatous plaque that was causing the frequent attacks (we presume) has perhaps ‘healed’ and become covered with endothelium (section 6.3.5).

16.6.3 Anticoagulants for patients with mechanical prosthetic heart valves Patients with mechanical prosthetic heart valves are at very high risk of embolism, and require anticoagulation whether or not they have a history of ischaemic stroke or other embolic problems. For some types of mechanical prosthetic valve, and in patients who have additional risk factors, such as a history of embolism, atrial fibrillation or cardiac failure, a higher target INR of 3 (range 2.5–3.5) combined with low-dose aspirin is warranted.147 The management of acute stroke in patients with prosthetic mechanical valves already on anticoagulants has been discussed in sections 12.4.7 and 13.3.4. The management of ischaemic stroke and other thromboembolic events among pregnant women with mechanical prosthetic heart valves is controversial because evidence is limited, and this is discussed in detail elsewhere.238 In brief, since oral anticoagulants increase the risk of fetal abnormalities, they should be avoided during the first trimester. Also, because they increase the risk of fetal bleeding during delivery, they should also be avoided in the weeks before delivery. Options for anticoagulation in these patients during pregnancy include low-molecular-weight heparin throughout pregnancy,

very carefully monitored adjusted-dose unfractionated heparin throughout pregnancy, or unfractionated or low-molecular-weight heparin until the 13th week and then change to warfarin until the middle of the third trimester before restarting heparin for the later weeks of pregnancy until after delivery. In very high-risk cases, the addition of low-dose aspirin may be warranted.238

16.6.4 Risks of bleeding with oral anticoagulants The main determinants of the risk of bleeding with warfarin and other vitamin K antagonists are:224 • Intensity of anticoagulation: the risk is higher at INR above 3 compared with INR of 2–3. Variation in the intensity of anticoagulation (i.e. poor control of the INR) also increases the risk of bleeding independent of the mean INR. • Underlying patient characteristics: most consistently increasing age, history of gastrointestinal bleeding and (for intracranial haemorrhage) history of stroke, especially if brain imaging shows leukoaraiosis. Other characteristics less consistently associated with increased risk are a history of hypertension, alcoholism, falls, dementia, liver disease and malignancy. • Length of treatment: the risk of bleeding is higher early in the course of treatment and falls with time. • Concomitant medications: there is a long list of drugs which should either be avoided or used with caution, with increased monitoring of the INR in patients taking warfarin or another vitamin K antagonist because they may increase the risk of bleeding or embolic events through a variety of different types of interaction (Table 16.10).239 The possibility that any new drug might interact with warfarin should always be considered and checked in a reliable source (e.g. the British National Formulary) before it is started or stopped. It is also worth bearing in mind that herbal medicines and some food and drink items (for example cranberry juice) may interact with warfarin.240 A number of models to aid with prediction of risk of anticoagulant-related bleeding have been developed. These should not be used in isolation but in conjunction with assessments of each individual patient’s functional status, risk of ischaemic events and personal preferences. Other factors, such as microhaemorrhages on an MR brain scan and genotype (e.g. apolipoprotein E genotype may modify the risk of cerebral amyloid angiopathy and so of intracerebral haemorrhage, and polymorphisms in the genes encoding cytochrome P450 enzymes and epoxide reductase complex 1 may identify patients with altered anticoagulant metabolism at increased risk of bleeding during the initiation of anticoagulants), may prove useful in the future to refine clinical risk prediction,

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 839

16.6 Anticoagulants Table 16.10 Drugs which interact with oral anticoagulants (adapted from the Scottish Intercollegiate Guidelines Network, 1999, see http://www. sign.ac.uk/pdf/sign36.pdf).

Avoid Aspirin Analgesics Antifungals Diabetes Non-steroidal anti-inflammatory Others Adjust dose Ulcer healing Anti-arrhythmics Lipid-lowering Anti-epileptics

Alcohol dependency Antibiotics/antifungals

Thyroid

Non-steroidal anti-inflammatory Gout Others

Except where combination specifically indicated Co-proxamol Ketorolac (postoperative) Miconazole Glucagon Azapropazone Phenylbutazone Enteral feeds containing vitamin K Cimetidine Omeprazole Amiodarone Propafenone Fibrates Carbamazepine Phenobarbitone Phenytoin Primidone Disulfiram Aztreonam Cephamandole Chloramphenicol Ciprofloxacin Co-trimoxazole Erythromycin Griseofulvin Metronidazole Ofloxacin Rifampicin Sulphonamides Carbimazole Thiouracils Thyroxine Diflunisal Allopurinol Sulphinpyrazone Aminoglutethimide Barbiturates Cyclosporin Mercaptopurine Oral contraceptives

Monitor international normalized ratio Gastrointestinal motility Cisapride Anti-arrhythmics Quinidine Lipid-lowering Cholestyramine Statins Antidepressants Serotonin uptake antagonists Antibiotics/antifungals Check product data sheet, if not listed under ‘adjust dose’ above Diabetes Tolbutamide Non-steroidal anti-inflammatory Check product data sheet, if not listed under ‘avoid’ or ‘adjust dose’ above Others Anabolic steroids Corticosteroids Hormone antagonists Ifosfamide Influenza vaccine Rawachol Sucralfate

..

..

839

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 840

840

Chapter 16 Preventing recurrent stroke and other serious vascular events

but further studies are needed to establish their value in this context.224,241–243 The risk of anticoagulant-induced bleeding depends on the intensity of anticoagulation, age, various other patient characteristics, time since anticoagulation started, and concomitant medications. The optimum balance between benefits and risks for most patients in atrial fibrillation is probably achieved with an INR of 2.5 (target range 2–3).

16.6.5 Who not to treat In general, the main contraindication to oral anticoagulation is a high risk of bleeding (section 16.6.4). However, many of the characteristics (e.g. age) which increase the risk of bleeding also increase the risk of thromboembolism and there has to be a trade-off between risks and benefits in individual patients. So for patients in atrial fibrillation at low risk of stroke the risks of anticoagulation are probably too high and they are more appropriately treated with aspirin (section 16.5.1). Patients with active infective endocarditis have a high risk of intracerebral haemorrhage, and should not be treated with anticoagulants even though they have a serious embolic risk.244 If possible, we always obtain a brain CT (or MR) scan to exclude intracranial haemorrhage before starting anticoagulants, because anticoagulation of patients with intracranial bleeding is presumably unwise, at least in the short term, if not in the long term. Brain CT needs to be performed within the first few days of stroke before any evidence of intracranial haemorrhage has disappeared, but for patients presenting later, MR scanning with appropriate sequences is necessary to detect previous intracerebral haemorrhage (section 5.5.1). As was discussed above for antiplatelet treatment (section 16.5.2), there are sometimes compelling reasons for long-term anticoagulants in patients with a history of intracranial haemorrhage, and in such cases the potential benefits and hazards have to be considered on an individual basis. For example, patients with atrial fibrillation, particularly those with rheumatic valve disease or a mechanical heart valve at high risk of subsequent ischaemic events, may warrant long-term anticoagulants despite a prior history of intracerebral haemorrhage.156 It is not clear whether or not the presence of haemorrhagic transformation of a cerebral infarct is an absolute contraindication to the use of anticoagulants. If we felt the patient had a clear indication for immediate anticoagulation, we would not delay if the scan showed a minor degree of haemorrhagic transformation (e.g. a few petechiae). On the other hand, if there were confluent

haemorrhage or a parenchymatous haematoma, we would prefer to wait a week or two at least.

16.6.6 Practicalities of long-term anticoagulation Starting anticoagulants If oral anticoagulants are to be used for long-term secondary prevention, a baseline full blood count, platelet count, INR and liver function tests should be done – and the results known – before treatment is started. On the rare occasions when there is an urgent need for immediate anticoagulation, intravenous unfractionated heparin or low-molecular-weight heparin can be given, since oral anticoagulants take several days to become effective. Most hospitals have their own local guidelines for the administration (and monitoring in the case of intravenous unfractionated heparin) of these drugs, and these should be followed. After taking the first few doses of oral anticoagulants without concomitant heparin, some patients may develop a transient pro-coagulant state. Therefore, in non-urgent situations, for patients with protein S or protein C deficiency or another thrombophilic condition, it is prudent to begin heparin therapy before or at the same time as warfarin to protect against this and then to withdraw the heparin once a therapeutic INR has been reached.239 There is room for flexibility in selecting a starting dose of warfarin. A dose of 10 mg is often preferred for many patients, but several studies have suggested a starting dose of 5 mg or less in the elderly, or patients with liver disease, congestive heart failure or at increased risk of bleeding. The starting dose generally needs to be given daily for 1–2 days until the INR is in the therapeutic range, with subsequent dosing based on the INR response. Suitable dosing schedules are available for guiding this process.239 Monitoring anticoagulants In hospitalized patients, daily monitoring is appropriate until a therapeutic INR has been achieved and maintained for at least two consecutive days, followed by two to three times weekly for 1–2 weeks, and then less often, depending on the stability of the INR. Once the INR is stable, the frequency of monitoring can be reduced to once every few weeks. The American College of Chest Physicians suggests no more than 4 weeks between tests, while the British Society for Haematology’s guidelines suggest 12 weeks, although more frequent monitoring may be required in patients with an unstable INR, for whatever reason.239,245

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 841

16.7 Lifestyle modification

Different models of anticoagulation management Since the risk of bleeding and of embolic complications is increased by instability in the anticoagulation intensity, careful management of anticoagulation is crucial to maintain the benefits and reduce the risks. The key principles are that management of anticoagulation should be well organized and coordinated, incorporating patient education, systematic INR testing, tracking, follow-up, and effective communication of results and dosing decisions to the patient. There are a number of different ways of achieving this, some of which may be effectively combined, including anticoagulation clinics based in hospitals or in the community, INR testing in general practice with dosing advice from specialist departments in the hospital, patient self-testing with dosing advice from the hospital or patient self-management of dosing decisions, and computer software programs to aid in dose adjustment.239 There is evidence from a systematic review of RCTs that, compared with standard monitoring, patient self-monitoring or self-management improves outcomes, with fewer thromboembolic events, lower mortality and a higher proportion of INR values in range, although no clear reduction in haemorrhage risk. However, self-management is only feasible for patients who are capable of self-monitoring and self-adjusting therapy, and so requires their identification and education.246 Otherwise there is little clear evidence from RCTs to suggest the superiority of one method of managing anticoagulation over another.239 The best choice is likely to depend on the patient, their doctor and local health services.

16.6.7 How long should treatment continue? As with antiplatelet treatment (section 16.5.6), most of the trials have tested only a few years of anticoagulants and none have directly compared different treatment durations. Thus, it is not known whether the balance of risk and benefit alters with prolonged therapy, particularly as the patient becomes older and so at higher risk of bleeding. If a patient has a serious haemorrhage, not due to over-anticoagulation, then the reason for giving anticoagulants should be reviewed, and the risk of thromboembolism reassessed. Since occult pathological lesions in the gastrointestinal tract and genitourinary system often present with bleeding, investigation of such patients (even if apparently over-anticoagulated) is usually justified.224 Further research is clearly required to determine the optimum duration of treatment. In the meantime, if we consider that a patient remains at high risk (whatever their age), no bleeding events have occurred, and no factors that might increase the risk of

..

..

bleeding have developed (e.g. tendency to falls or dementia), most of us would keep the patient on anticoagulants indefinitely.

16.7 Lifestyle modification

16.7.1 Smoking cessation Risks of smoking In 2000, smoking was estimated to have caused almost 5 million premature deaths worldwide, about half in developed and half in developing countries, and about 80% of these in men. The leading causes of smoking-related death were vascular disease (including ischaemic heart disease and stroke), chronic obstructive pulmonary disease, and lung cancer.247 Results from 40 years of prospective observations on over 30 000 male British doctors demonstrated that almost half of all regular cigarette smokers will eventually be killed by their habit.248 Smoking is clearly a risk factor for ischaemic stroke (section 6.6.4) and for subarachnoid haemorrhage (section 9.1.1), although its effect on the risk of intracerebral haemorrhage is less clear.249–252 Benefits of stopping Because of the clearly established adverse effects of all forms of tobacco consumption (cigarettes, pipes, cigars and chewing tobacco) and the overwhelming indications for stopping, there have, unsurprisingly, been no successful RCTs of ‘continue smoking’ vs ‘stop smoking’. However, observational studies show substantial early and long-term health benefits of stopping smoking for smokers of all ages. The excess mortality falls soon after stopping and continues to do so for at least 10–15 years. Stopping smoking before the age of about 35 years leads to the avoidance of almost all the subsequent excess mortality, but stopping later in life still confers substantial benefit.248 Stopping smoking halves the risk of mortality due to ischaemic heart disease within a year, and the risk of stroke declines substantially within 2–5 years.253–256 Best methods to aid smoking cessation Systematic reviews have demonstrated the effectiveness of a variety of interventions to help people stop

841

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 842

842

Chapter 16 Preventing recurrent stroke and other serious vascular events

Intervention Brief advice from a doctor vs no advice/usual care Intensive advice from a doctor vs no advice/usual care

Trials

Participants

OR (95% CI)

17

12499

1.7 (1.5–2.1)

8

6923

2.0 (1.7–2.4)

15

9775

1.4 (1.2–1.7)

Nursing intervention (advice/counselling) versus control/usual care 20

10289

1.5 (1.3–1.7)

11

13733

1.2 (1.1–1.5)

3

7790

1.4 (1.1–1.7)

Telephone counselling vs less intensive intervention

13

16462

1.6 (1.4–1.8)

Individual behavioural counselling vs minimal contact control

17

6384

1.6 (1.3–1.8)

7

815

2.2 (1.4–3.5)

Intensive vs minimal advice from a doctor

Self help materials vs no materials Individually tailored self help materials vs no materials

Group behavioural therapy vs no intervention Group behavioural therapy vs self help materials

16

4395

2.0 (1.6–2.6)

Nicotine replacement therapy vs control/placebo

105

39503

1.8 (1.7–1.9)

Gum

52

17783

1.7 (1.5–1.8)

Patches

38

16691

1.8 (1.6–2.0)

Nasal spray

4

887

2.4 (1.6–3.4)

Inhaled

4

976

2.1 (1.4–3.2)

Sublingual

5

2739

2.1 (1.6–2.6)

19

6443

2.1 (1.8–2.4)

4

703

2.8 (1.7–4.6)

Bupoprion vs placebo Nortriptyline vs placebo

1

10

Fig. 16.32 Meta-analyses of randomized trials of various interventions for smoking cessation showing effects on abstinence from smoking. Squares represent summary odds ratios (OR) for the various interventions shown, with the size of each square proportional to the amount of statistical

information for that intervention. Horizontal lines are 95% confidence intervals (CI). (Data from Lancaster and Stead, 2004;257 Rice and Stead, 2004;259 Lancaster and Stead, 2005;258;261 Stead and Lancaster, 2005;260 Stead et al., 2003;262 Silagy et al., 2004;264 Hughes et al., 2004.265)

smoking, and in general these increase quit rates by about 1.5 to 2-fold (Fig. 16.32). Advice from doctors, structured interventions from nurses, and individual and group counselling are all effective.257–260 Generic selfhelp materials and telephone counselling are no better than brief advice but are more effective than doing nothing, and personalized self-help materials are better than standard materials.261,262 Simple advice to give up smoking is one of the most cost-effective interventions in the whole of medicine.263 All forms of nicotine replacement therapy are similarly effective, and remain effective when given in combination with various types of support, such as advice or counselling.264 There are no known serious adverse effects, but all can cause local irritation, of the nose, mouth or skin, depending on the route of administration. The antidepressants bupropion and nortriptyline also increase quit rates. Bupropion can cause dry mouth and insomnia, but the only serious adverse effect is seizures, for which the risk is about one in 1000. The drug is therefore contraindicated in those with a history of current

or previous epilepsy, and should be used with caution in people who may have a lowered seizure threshold, which includes patients with a history of cerebrovascular disease (section 11.8). It should not be used with a monoamine oxidase inhibitor.265 All smokers, including those with a history of stroke or transient ischaemic attack, should be advised to stop, and interventions such as counselling, nicotine replacement or bupropion should be used if needed to help them achieve this.

16.7.2 Avoiding excess alcohol Observational studies suggest that, compared with no alcohol consumption, light to moderate alcohol consumption (about 1–3 units per day) reduces the risk of death, ischaemic heart disease and stroke, while excessive alcohol consumption increases the risk of these outcomes, giving rise to J-shaped associations (sections 6.6.13 and

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 843

16.7 Lifestyle modification

Natural logarithm relative risk

3

Men

Women

2 1 0 –1 –2 –3 0

20

40

60 80 100 Alcohol intake, g/day

120

140

20

30 40 50 Alcohol intake, g/day

60

70

80

curve. (Reprinted from Reynolds et al.269 © 2003 with permission from the American Medical Association.)

8.5.2) (Fig. 16.33).266–269 Several studies have compared the effects of beer, wine and spirits, and, in general, the effects of different types of alcohol are very similar.270–273 On the basis of the observational evidence, it is therefore sensible to advise avoiding heavy alcohol intake for everyone, including patients who have had a stroke or TIA, but it is conceivable that a couple of drinks a day may provide net protection against further vascular events. However, since it is difficult to completely exclude residual confounding in the observational studies, and since there is an association between the mean level of alcohol consumption and the prevalence of heavy drinking at the population level, it does not seem sensible to encourage the use of alcohol for vascular prevention as a public health measure.274–276

found that these modest reductions for at least 4 weeks among hypertensive people reduced blood pressure by about 5 mmHg systolic/3 mmHg diastolic, while among normotensive individuals they reduced blood pressure by about 2 mmHg systolic/1 mmHg diastolic; there was a correlation between the reduction in urinary sodium excretion and the reduction in blood pressure.277 Older people with higher blood pressures are likely to achieve greater absolute reductions in blood pressure with dietary modification. In another systematic review of RCTs assessing the longer-term effects of dietary salt restriction on blood pressure in about 3500 people, followed up for between 6 months and 7 years, very few vascular outcomes were recorded with no significant differences detected. Salt intake was reduced by an average of only about 2 g/day, and so blood pressure fell by only 1.1 mmHg systolic/0.5 mmHg diastolic. This shows that long-term maintenance of low salt intake is difficult for individuals, even with a great deal of support, advice and encouragement. Nevertheless, this small reduction in salt intake did cause a small but significant fall in blood pressure, which, on a population basis, should lead to a worthwhile reduction in vascular events, including strokes. The review also found that people on antihypertensive medication were able to stop their medication more often on a reduced sodium diet as compared with controls, while maintaining similar blood-pressure control.278 Subsequently, a long-term follow-up study of outcomes 10–15 years after two of the trials included found no significant reduction in all-cause mortality, but did find, among the 2400 people for whom morbidity data were available (77% of those originally randomized), a just statistically significant one-quarter relative reduction in myocardial infarction, stroke, coronary

16.7.3 Dietary modification Dietary advice to reduce blood pressure by modifying salt intake, and potassium Observational data suggest an association between dietary salt (sodium) consumption and blood pressure, and public health recommendations in most developed countries are to reduce salt intake by about half, from approximately 10 g down to 5 g/day. Modest reductions of this size can be achieved with relatively simple dietary modifications (not adding salt at the table or in cooking, and avoiding salty foods). A systematic review of RCTs

..

10

Fig. 16.33 Log relative risk of stroke versus alcohol consumption for men and women. Scatterplot of results of individual studies in a meta-analysis, and meta-regression

Everyone, including those who have had a stroke or transient ischaemic attack, should avoid heavy drinking, but a couple of units of alcohol per day may protect against future vascular events.

..

0

843

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 844

844

Chapter 16 Preventing recurrent stroke and other serious vascular events

revascularization or cardiovascular death.279 Given the marginal statistical significance and rather large proportion of participants lost to follow-up, this is not a conclusive result, but nonetheless considerably strengthens the evidence that a lower dietary salt intake reduces the risk of vascular events. Reducing dietary salt intake reduces blood pressure, particularly in older people with high blood pressure, perhaps resulting in long-term reduction in vascular events. It may also help those on antihypertensive medication to stop their treatment without a rise in blood pressure. However, it is difficult for people to stick to a low salt diet. Observational epidemiological studies have shown an association between increased potassium and reductions in blood pressure, and in stroke risk.280 A systematic review of RCTs of increased dietary potassium found that short-term increases equivalent to about five bananas per day reduced blood pressure by about 3 mmHg systolic/ 2 mmHg diastolic, and appeared more effective in those with higher sodium intake, suggesting that potassium supplementation may be helpful for blood pressure reduction in people who are unable to reduce their intake of sodium.281 The best way to increase dietary potassium is to increase consumption of fresh fruit and vegetables.280 Dietary advice to reduce cholesterol A meta-analysis of metabolic ward studies of diets designed to reduce cholesterol levels in healthy volunteers found that dietary change could reduce blood cholesterol concentrations by about 0.8 mmol/L (i.e. by 10–15%). Achieving this involved replacing 60% of saturated fats with other fats and avoiding about 60% of dietary cholesterol in a typical British diet.282 In the real world, rather than under carefully controlled conditions, reductions of this size are not achievable. Several systematic reviews of RCTs in free-living subjects found that dietary advice produced total cholesterol reductions of only around 5% (about 0.3 mmol/L with a baseline cholesterol of 6.3 mmol/L), and more intensive diets were somewhat more effective than less intensive ones. Difficulties in complying with the prescribed dietary change explain the difference between the real world and the ward-based studies.283,284 Thus, although some useful reduction in cholesterol levels can be achieved with dietary advice, it is important to be realistic about the reductions in vascular events that are likely to result. A systematic review of the effects on vascular outcomes found 27 trials in which the dietary interventions included provision of the majority

of food for trial participants over several years, advice about diets with dietary fat restriction or modification, and provision of a combination of dietary advice and supplementation. Meta-analysis showed a trend towards a reduction in vascular mortality, and a marginally significant reduction in fatal and non-fatal vascular events (including myocardial infarction, angina, stroke, heart failure, peripheral vascular events and unplanned cardiovascular interventions) (rate ratio 0.84; 95% CI 0.72–0.99).285 No significant harms are likely to result from dietary advice, however, so despite small effects at the individual level, it seems very reasonable to recommend cholesterol-lowering dietary advice for all stroke and TIA patients. It seems sensible to advise all patients with a prior stroke or transient ischaemic attack to reduce their dietary intake of saturated fat, since this produces moderate reductions in cholesterol levels, which are probably associated with small reductions in subsequent vascular events. Weight reduction In observational studies increasing body mass index (BMI) above 21–22 is associated with an increased risk of ischaemic heart disease, hypertension, stroke, diabetes, cancer and osteoarthritis (section 6.6.10). The effects on ischaemic heart disease and stroke are mediated mainly through effects on blood pressure, diabetes and changes in lipid profile rather than raised BMI per se.286,287 A systematic review of RCTs of various nonsurgical interventions for weight loss in overweight and obese people found that both dietary and drug interventions led to modest reductions in weight and, as anticipated, favourable effects on vascular risk factors. The most widely studied dietary intervention was a low-fat or reduced calorie diet, producing reductions at 1 year of about 5 kg in weight, along with small reductions in total cholesterol, diastolic and systolic blood pressure, and fasting plasma glucose. The drugs orlistat and sibutramine both led to weight reductions of 3–4 kg at 1 year. Exercise programmes and behaviour therapy produced additional weight loss when combined with dietary interventions, but there was no clear evidence to suggest that individual therapy was more effective than group or family therapy. However, limited sample sizes and follow-up duration mean that there is no reliable evidence about the effects of any of these interventions on vascular outcomes or mortality.288 Another systematic review of RCTs in morbidly obese people (BMI > 40) found that surgical interventions, compared with conventional treatment, led to long-term weight loss of 23–37 kg,

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 845

16.8 Dietary supplements: B vitamins and antioxidants

as well as improved quality of life and reduced comorbidity, and that gastric bypass was the most effective surgical procedure.289 The Mediterranean diet Observational epidemiological studies have shown that increased consumption of fruit and vegetables, increased fish consumption (particularly oily fish, which is rich in omega 3 fatty acids), and increased dietary fibre are all associated with a reduced risk of stroke and ischaemic heart disease.290–296 These apparently protective dietary components are combined with low saturated fat intake in the so-called ‘Mediterranean diet’, which is characterized by a high intake of fruit, vegetables and cereals, a moderate to high intake of fish, a low intake of saturated lipids but high intake of unsaturated lipids (particularly olive oil), a low to moderate intake of dairy products, a low intake of meat and a modest intake of alcohol, mainly as wine. This Mediterranean diet is associated in observational studies with reduced mortality from all causes, vascular disease and cancer.297,298 However, there is no evidence from either observational studies or from RCTs that dietary or supplemental omega 3 fats alter total mortality, combined vascular events, stroke or cancer in people with, or at high risk of, vascular disease or in the general population.299,300 One small RCT of the Mediterranean diet in 605 ischaemic heart disease patients did find that all-cause mortality, myocardial infarction and cancer risks were more than halved but the small numbers of outcome events and early stopping of the study mean that these results may well be overoptimistic.301 And the very promising results of another RCT of a similar diet, the Indo-Mediterranean diet, have been called into question.302–304 A systematic review of RCTs of dietary advice (encouraging reduced consumption of at least one of fat, saturated fatty acids, cholesterol, or sodium, or increased consumption of at least one of fruit, vegetables, polyunsaturated fatty acids, monounsaturated fatty acids, fish, fibre or potassium) found modest benefits on dietary intake and several vascular risk factors, but longer-term effects on vascular outcomes are unknown.305 Obese individuals should be encouraged to lose weight using dietary – or if necessary pharmacological or surgical – interventions, and all patients should receive general advice about a healthy diet, low in saturated fat, with plenty of fish, fibre, fruit and vegetables. These interventions have beneficial effects on vascular risk factors, and seem likely to produce small reductions in vascular outcomes, although there is no clear evidence that they do.

..

..

16.7.4 Exercise Observational studies have shown that physical activity is associated with a lower risk of myocardial infarction and stroke, but most were done in middle-aged people without a history of vascular disease, limiting their applicability to other groups (section 6.6.9).306,307 A systematic review of RCTs found that aerobic exercise reduces blood pressure, but any effects on vascular outcomes are uncertain.307,308 Cardiac rehabilitation programmes (of which exercise is just a part) reduce the frequency of death and major cardiac events after myocardial infarction.309 However, the effects of exercise programmes alone after myocardial infarction or stroke on the occurrence of further vascular events are unknown. It therefore seems reasonable to encourage patients to return to normal physical activities as much as possible after a stroke, but there is no evidence that exercise programmes reduce the risk of recurrent stroke or other vascular events. However, exercise may bring other benefits; for example, a programme of muscle strengthening and balance retraining reduces the risk of falls in elderly people.310

16.8 Dietary supplements: B vitamins and antioxidants

16.8.1 Folic acid and other B vitamins Systematic reviews of observational studies have suggested an association between raised plasma homocysteine concentrations and both ischaemic heart disease and stroke.311,312 While this does not necessarily imply a cause-and-effect relationship, these findings are supported by the association with ischaemic heart disease and stroke of the common functional polymorphism C677T in the gene encoding methyltetrahydrofolate reductase (MTHFR), an enzyme involved in homocysteine metabolism, showing that the observed increases in risk of both stroke and ischaemic heart disease in individuals with the TT genotype is close to that predicted from the increase in homocysteine concentration conferred by this variant.312–315 However, the apparent association could be due to another nearby gene in linkage disequilibrium with the MTHFR gene, to confounding by different ethnic composition of cases and controls (population stratification), or to publication bias (positive studies are more likely to be published and so included in systematic reviews than negative ones) (section 6.6.8).316,317

845

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 846

846

Chapter 16 Preventing recurrent stroke and other serious vascular events

If the association between homocysteine concentrations and vascular disease is causal, then lowering homocysteine concentrations should reduce the risk of stroke and ischaemic heart disease. Such lowering is relatively easy to achieve, and RCTs of 0.5–5 mg folic acid daily (with about 0.5 mg vitamin B12 to minimize the potential for neurological complications due to B12 deficiency) do lower homocysteine concentrations by about onequarter to one-third (e.g. from about 12 to 8–9 µmol/L).318 Four large RCTs of folic acid supplements in the prevention of vascular events in high-risk patients have now been completed: • The Vitamin Intervention for Stroke Prevention (VISP) trial randomly allocated patients with a previous ischaemic stroke to high vs low doses of folate, vitamin B6 and vitamin B12 and followed them for a mean of 20 months. Compared with low dose, high-dose vitamins reduced homocysteine concentrations by 2 µmol/L, but did not affect recurrent ischaemic stroke, ischaemic heart disease or death.319 • In the second Cambridge Heart Antioxidant Study (CHAOS-2) (which has so far only been published in abstract form), patients with ischaemic heart disease received folic acid or placebo for almost 2 years. There was a 13% reduction in homocysteine concentration in the folic acid group, but no reduction in vascular outcomes.320 • The Norwegian Vitamin Study (NORVIT) randomized patients who had survived an acute myocardial infarction in a factorial design to folic acid and vitamin B12, folic acid, vitamin B12 and vitamin B6, vitamin B6, or placebo, and followed them for 3.5 years. Treatment with folic acid and vitamin B12 reduced homocysteine levels by almost one-third (over 4 µmol/L), but again had no significant effect on recurrent myocardial infarction, stroke or death.321 • Finally, in the Heart Outcomes Prevention Evaluation (HOPE) 2 trial, patients with vascular disease or diabetes were randomized to daily treatment with folic acid, vitamin B6 and vitamin B12 or placebo for an average of 5 years. The difference in mean plasma homocysteine levels between active treatment and placebo groups was 3.2 µmol/L. Active treatment significantly reduced the relative risk of stroke by about one-quarter, but did not reduce vascular death, myocardial infarction, or major cardiovascular events combined.322 A meta-analysis of RCTs of folic acid supplementation, including these four large trials along with several smaller ones among a total of 17 000 participants with pre-existing vascular disease, found no evidence of a reduction in all-cause mortality or vascular outcomes.323 Although these results appear disappointing, they are

still compatible with a beneficial effect of folate supplementation, since only a relatively small number of outcome events occurred in these four trials, and in CHAOS-2 and VISP both the treatment duration and the difference in homocysteine concentrations achieved between groups were limited. At least eight further large trials in a variety of secondary prevention settings are still under way, between them including over 37 000 participants and expected to generate almost 8000 vascular events. In combination, these trials should have adequate power to detect moderate effects of homocysteine lowering on vascular outcomes, and to assess whether the effects vary among different populations.324 Two of the ongoing studies are among or include patients with a history of ischaemic stroke or TIA, and together will include a total of around 9000 such patients.325,326 Thus, the currently available evidence does not rule out a beneficial role for folate supplementation in prevention of serious vascular outcomes. However, since there are possible hazards (e.g. of breast and prostate cancer),327 even of something as benign-sounding as vitamin supplements, for now we must await the evidence that will emerge from the ongoing RCTs. At present there is no evidence from randomized trials to support the routine use of homocysteine lowering with folate supplements to prevent vascular events, either in primary prevention or after a stroke or transient ischaemic attack.

16.8.2 Antioxidants Laboratory studies suggest that antioxidants protect against atherosclerosis by limiting the oxidation of lowdensity lipoprotein in the arterial wall.328,329 In keeping with this, observational epidemiological studies have found an association between low intake and low levels of antioxidants (mainly beta-carotene, vitamin C and vitamin E) and risk of stroke and ischaemic heart disease (section 6.6.12).296,330 This suggests that dietary antioxidant supplements might reduce the risk of ischaemic stroke and other vascular outcomes. A number of dietary constituents have antioxidant properties, including certain minerals (e.g. selenium, copper, zinc, manganese), vitamins (C, E), pro-vitamins (beta-carotene) and flavonoids. However, tens of thousands of individuals have now been randomized in several trials assessing the effects of antioxidant supplements (mainly vitamins E and C and beta-carotene) on vascular outcomes, both in primary and secondary prevention settings. Systematic reviews have found no evidence of benefit, even with follow-up for over 7 years.331–333 In fact, beta-carotene

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 847

16.9 Management of diabetes mellitus and glucose intolerance

appeared to be associated with increased vascular mortality.333 Thus, there is no indication to use antioxidant supplements for secondary prevention after stroke. Randomized trials do not support the use of antioxidant supplements after a stroke or transient ischaemic attack.

16.9 Management of diabetes mellitus and glucose intolerance

Diabetes mellitus and glucose intolerance are important risk factors for ischaemic stroke and other vascular diseases; patients with type 2 diabetes have an annual risk of stroke, MI or vascular death of about 6%334–337 (section 6.6.5). There is good evidence that bloodpressure-lowering drugs, statins to reduce cholesterol, and antiplatelet treatment produce similar relative benefits in people with and without diabetes, so it is clearly appropriate to use these treatments after stroke, regardless of whether or not the patient has diabetes.117,140,338 But does tight diabetes control affect the risk of stroke or myocardial infarction? Randomized trials (the United Kingdom Prospective Diabetes Study [UKPDS] and another smaller trial) found that intensive blood-glucose control (aiming for a fasting plasma glucose level of less than 6 mmol/L) with either a sulphonylurea or insulin in patients with type 2 diabetes substantially decreased the risk of microvascular complications (retinopathy and renal damage), but increased the risk of hypoglycaemic episodes and weight gain, and did not appear to reduce macrovascular disease (i.e. stroke and myocardial infarction).339,340 The UKPDS also found that blood glucose control with metformin in overweight patients reduced microvascular complications, as well as reducing risk of death, myocardial infarction and stroke (the latter non-significantly), with less excess risk of weight gain and hypoglycaemia than with sulphonylureas or insulin.341 More recently, a randomized placebo-controlled trial of the new hypoglycaemic drug, pioglitazone, including over 5000 patients with type 2 diabetes and a history of stroke, ischaemic heart disease or peripheral arterial disease, found that treatment reduced the risk of death, myocardial infarction or stroke, such that treatment of 1000 subjects like those in the trial for 3 years would result in the avoidance or delay of 21 myocardial infarctions, strokes or deaths. Active treatment also reduced the need to start insulin. However, this benefit was at the expense of an in-

..

..

creased risk of hypoglycaemic episodes, weight gain and heart failure, although there was no overall difference in the risk of serious adverse events between treatment groups.342 Thus, since careful control of glycaemia in diabetes improves outcome in terms of microvascular – and perhaps macrovascular – complications, detecting and treating undiagnosed diabetes and ensuring adequate treatment for known diabetes in patients who have had a stroke or TIA is appropriate. In overweight patients, metformin is probably the most appropriate first choice, but otherwise there remain some uncertainties about the best way of achieving glycaemic control. Management of diabetes is probably best coordinated in a dedicated diabetes clinic, where one is available. The management of hyperglycaemia and diabetes in the acute phase of stroke is discussed in section 11.18.3. Good glycaemic control in patients with diabetes reduces the risk of microvascular – and perhaps macrovascular – complications.

16.10 Treatment of specific underlying causes

Specific treatment of rare causes of stroke has been discussed in Chapter 7. Here, we outline the specific treatment of some of the commoner causes of stroke.

16.10.1 Atrial fibrillation: rate control vs attempted cardioversion to sinus rhythm Patients in atrial fibrillation (AF) can in theory be managed either by attempted electrical or pharmacological cardioversion to sinus rhythm or by rate control along with long-term antithrombotic (anticoagulant or antiplatelet) treatment to reduce the risk of embolism. The advantages of cardioversion are that successful restoration of sinus rhythm should improve left ventricular ejection fraction and maximal exercise capacity, and it may reduce the risk of embolic complications so obviating the need for long-term anticoagulation with its attendant inconvenience and risks. However, it is not certain that restoration of sinus rhythm does reduce embolic risk. In addition, about three-quarters of patients spontaneously revert to AF within 1 year of electrical cardioversion, and this occurs in one-quarter to one-half of those initially restored to sinus rhythm using drugs. Finally, neither the procedure of electrical cardioversion nor the drugs used for pharmacological cardioversion are free of risk.343,344

847

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 848

848

Chapter 16 Preventing recurrent stroke and other serious vascular events

A systematic review of electrical cardioversion followed by anti-arrhythmic drugs to maintain sinus rhythm, generally with oral anticoagulants for a few weeks before and after the electrical cardioversion, vs rate control with digoxin, a beta-blocker or a calciumchannel antagonist, plus oral long-term anti-coagulants identified three RCTs. There was no significant difference in death, stroke and TIA, or peripheral embolism. Physical aspects of quality of life were significantly better in the rhythm control group, perhaps because of the larger proportion in sinus rhythm, but since quality of life was only reported for a little over half of the total number of patients, this result may not be reliable.344 Another systematic review of pharmacological cardioversion vs rate control identified two RCTs that had been published in full, between them including about 4300 subjects with a mean age of 70 years, with almost 95% of the subjects included in just one trial.343 In this trial, the rhythm control group received the treating physician’s drug of choice (most often amiodarone or sotalol, with over half receiving digoxin, a betablocker or a calcium-channel antagonist as initial therapy), electrical cardioversion as necessary (this occurred in about one-third of subjects), and anticoagulation was encouraged (and received by 70% throughout the course of the trial). The rate control group received digoxin, beta-blockers, or calcium-channel antagonists, and anticoagulation was mandatory (and received by 85% throughout the course of the trial). Follow-up was for a mean of 3.5 years. Again there was no significant difference between treatment groups in death, ischaemic stroke, systemic embolism, intracranial haemorrhage or extracranial haemorrhage. More patients needed hospitalization, and a slight excess of malignant dysrhythmias occurred during follow-up in the rhythm control compared with the rate control group.345 Therefore, the available evidence does not suggest any definite advantage of either electrical or pharmacological cardioversion over rate control with long-term anticoagulation, and suggests some excess risks of pharmacological rhythm control. However, these results only apply to patients eligible for anticoagulants, and do not guide us in choosing the most appropriate treatment strategy for patients in whom anticoagulants are clearly contraindicated. At present, we tend to opt for a policy of rate control plus antiplatelet treatment (usually aspirin) in these patients. Furthermore, the results cannot be generalized to younger patients with new-onset atrial fibrillation. In older patients with atrial fibrillation, a strategy of rate control (plus anticoagulation if possible, and antiplatelet treatment if not) is a highly acceptable primary strategy.

16.10.2 Patent foramen ovale and atrial septal aneurysm A patent foramen ovale (PFO) is present in about onethird of the general population and is associated with an atrial septal aneurysm (ASA) in about 20% of cases.346 A meta-analysis of case–control studies found that, among younger ( 12 weeks. These observations were consistent across the 50–69% and ≥ 70% stenosis groups and similar trends were present in both ECST and NASCET. The fall-off in the absolute benefit of surgery with time since last event (Fig. 16.43) is particularly important in terms of auditing the performance of stroke prevention services in routine clinical practice. Women had a lower risk of ipsilateral ischaemic stroke on medical treatment and a higher operative risk in comparison to men. For recently symptomatic carotid stenosis, surgery is very clearly beneficial in women with ≥ 70% stenosis group

50–69% stenosis group Events/Patients Surgical Medical

Events/Patients

ARR (%) 95% CI

Surgical Medical

Sex Male Female

65/546 36/262

89/489 21/205

8.0 –2.7

3.4 –12.5 – 8.8 –3.5

47/467 27/199

95/384 39/166

15.0 9.9

9.8–20.2 1.8–18.0

Age < 65 years 65 –74 years >75 years

44/375 43/329 14/104

33/270 55/324 22/100

1.3 5.4 10.7

– 4.0 – 6.7 – 0.4 –11.2 – 0.2–21.6

42/356 31/272 1/38

60/280 54/218 20/52

9.8 13.5 37.2

3.8–15.7 6.5–20.5 22.9–51.5

Time since last event < 2 weeks 17/158 2 – 4 weeks 21/135 4 –12 weeks 36/312 > 12 weeks 27/203

34/150 20/110 40/280 16/154

6.2–23.4 14.8 3.3 – 6.3 –13.0 4.0 –1.7–9.7 –2.9 –10.2– 4.3

23/167 10/133 27/248 14/118

54/149 24/105 41/218 15/78

23.0 15.9 7.9 7.4

13.6–32.4 6.6–25.2 1.3–14.4 –3.3–18.1

Primary symptomatic event 17/170 17/155 Ocular only 40/295 38/238 TIA 44/343 55/301 Stroke

1.5 3.8 7.5

–5.5 – 8.4 –2.6 –10.1 1.6 –13.5

18/214 25/207 31/245

22/156 53/199 59/195

5.5 15.4 17.7

–1.2–12.1 7.7–23.1 9.9–25.5

Diabetes Yes No

32/136 78/558

6.2 4.3

– 4.1–16.5 0.4 – 8.2

8/92 66/574

24/96 110/454

16.7 12.9

6.0–27.4 8.1–17.7

Symptomatic plaque surface Smooth 40/296 35/266 Irreg/ulcerated 71/512 75/428

3.3 5.7

–2.2– 8.9 0.7–10.6

34/257 40/409

44/207 90/343

7.8 17.1

0.8–14.8 11.6–22.6

5/32 69/634

14/37 120/513

23.6 12.7

3.2–44.0 8.3–17.1

74/666

134/550

13.5

9.1–17.9

29/143 72/665

Contralateral carotid occlusion Yes 9/31 5/45 No 92/ 777 105/649

TOTAL

101/808 110/694

–16.0 –34.8 – 2.8 1.9 –9.5 5.7

4.7

1.0 – 8.4 –10

0

10

20

30

% Absolute risk reduction (95% CI)

Fig. 16.42 Absolute reduction with surgery in the 5-year risk of ipsilateral carotid territory ischaemic stroke and any stroke or death within 30 days after surgery according to predefined subgroup variables in: patients with 50–69% stenosis; patients with ≥ 70% stenosis. The size of the purple boxes is

..

..

ARR (%) 95% CI

–10

0

10

20

30

40

50

% Absolute risk reduction (95% CI)

proportional to the amount of data represented by each box and the horizontal lines represent 95% confidence intervals. (Derived from Rothwell et al., 2004.466) CI, confidence interval; ARR, absolute risk reduction.

865

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 866

Chapter 16 Preventing recurrent stroke and other serious vascular events 30.2 17.6

30.0 14.8

8.9

11.4

20.0 3.3

4.0

10.0 0.0

–2.9 –10.0

12+ 0–2 2– 4 4–12 Weeks from event to randomization

Fig. 16.43 Absolute risk reduction (ARR) with surgery in the 5-year risk of ipsilateral carotid territory ischaemic stroke and any stroke or death within 30 days after surgery in patients with 50–69% stenosis (light purple bars) and ≥ 70% stenosis (purple bars) without near-occlusion, stratified by the time from last symptomatic event to randomization (Rothwell et al., 2004466). The numbers above the bars indicate the actual absolute risk reduction. CI, confidence interval.

≥ 70% stenosis, but not in women with 50–69% stenosis (Fig. 16.42). In contrast, surgery reduced the 5-year absolute risk of stroke by 8.0% in men with 50–69% stenosis. These same patterns were also shown in both of the large published trials of endarterectomy for asymptomatic carotid stenosis.494 Benefit from surgery increased with age in the pooled analysis, particularly in patients aged over 75 years (Fig. 16.42). Although patients randomized in trials generally have a good prognosis and there is some evidence of an increased operative mortality in elderly patients in routine clinical practice, as discussed above there is no increase in the operative risk of stroke and death in older age groups. There is therefore no justification for withholding surgery in patients aged over 75 years who are deemed to be medically fit. The evidence suggests that benefit is likely to be greatest in this group because of their high risk of stroke without surgery. Finally, benefit from surgery is probably greatest in patients with stroke, intermediate in those with cerebral TIA and lowest in those with ocular events (Fig. 16.42). There was also a trend in the trials towards greater benefit in patients with irregular plaque than a smooth plaque. Which individuals benefit most? Although there are some clinically useful subgroup observations in the pooled analysis of the trials, univariate subgroup analysis is often of only limited use in clinical practice. Individual patients frequently have several important risk factors, each of which interacts in

a way that cannot be described using univariate subgroup analysis, and all of which should be taken into account to determine the likely balance of risk and benefit from surgery.495 For example, what would be the likely benefit from surgery in a 78-year-old (increased benefit) female (reduced benefit) with 70% stenosis who presented within 2 weeks (increased benefit) of an ocular ischaemic event (reduced benefit) and who was found to have an ulcerated carotid plaque (increased benefit)? One way to weigh the often-conflicting effects of the important characteristics of an individual patient on the likely benefit from treatment is to base decisions on the predicted absolute risks of a poor outcome with each treatment option, using prognostic models. One such model for prediction of the risk of stroke on medical treatment in patients with recently symptomatic carotid stenosis has been derived from the ECST (Table 16.14).1,495 The model was validated using data from the NASCET and showed very good agreement between predicted and observed medical risk, reliably distinguishing between individuals with a 10% risk of ipsilateral ischaemic stroke after 5 years of follow-up and those with a risk of over 40% (Fig. 16.44). Importantly, Fig. 16.44 also shows that the operative risk of stroke or death in patients who were randomized to surgery in NASCET was unrelated to the medical risk. Thus, 50

40 Observed risk (%)

40.0

ARR (%), 95% CI

866

30

20

10

0 0

10

20

30

40

50

Predicted medical risk (%) Fig. 16.44 Validation of the ECST model (Table 16.14) for the 5-year risk of stroke on medical treatment in patients with 50–99% stenosis in NASCET (Rothwell et al., 20051). Predicted medical risk is plotted against observed risk of stroke in patients randomized to medical treatment in NASCET (squares) and against the observed operative risk of stroke or death in patients randomized to surgical treatment (diamonds). Groups are quintiles of predicted risk. Error bars represent 95% confidence intervals.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 867

16.11 Endarterectomy for symptomatic carotid stenosis Table 16.14 A predictive model for the 5-year risk of ipsilateral ischaemic stroke on medical treatment in patients with recently symptomatic carotid stenosis (Rothwell et al., 20051). Hazard ratios derived from the model are used for the scoring system. The score for the 5-year risk of stroke is the product of the individual scores for each of the risk factors present. The score is converted into a risk using the graphic in Fig. 16.45. An example is shown. MODEL

SCORING SYSTEM

Risk factor

Hazard ratio (95%CI)

P-value

Risk factor

Stenosis (per 10%)

1.18 (1.10–1.25)

T polymorphism and coronary heart disease: does totality of evidence support causal role for homocysteine and preventive potential of folate? Br Med J 2005; 331(7524):1053. 318 Lowering blood homocysteine with folic acid based supplements: meta-analysis of randomised trials. Homocysteine Lowering Trialists’ Collaboration. Br Med J 1998; 316(7135):894–8. 319 Toole JF, Malinow MR, Chambless LE, Spence JD, Pettigrew LC, Howard VJ et al. Lowering homocysteine in patients with ischemic stroke to prevent recurrent stroke, myocardial infarction, and death: the Vitamin Intervention for Stroke Prevention (VISP) randomized controlled trial. J Am Med Assoc 2004; 291(5):565–75. 320 Baker F, Picton D, Blackwood S. Blinded comparison of folic acid and placebo in patients with ischemic heart disease: an outcome trial. Circulation 2002; 106(Suppl 11):741. 321 Bonaa KH, Njolstad I, Ueland PM, Schirmer H, Tverdal A, Steigen T et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med 2006; 354(15):1578–88.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 893

References

322 Lonn E, Yusuf S, Arnold MJ, Sheridan P, Pogue J, Micks M et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med 2006; 354(15):1567–77. 323 Bazzano LA, Reynolds K, Holder KN, He J. Effect of folic acid supplementation on risk of cardiovascular diseases: a meta-analysis of randomized controlled trials. J Am Med Assoc 2006; 296(22):2720–6. 324 B-Vitamin Treatment Trialists’ Collaboration. Homocysteine-lowering trials for prevention of cardiovascular events: a review of the design and power of the large randomized trials. Am Heart J 2006; 151(2):282–7. 325 The VITATOPS Trial Study Group. The VITATOPS (Vitamins to Prevent Stroke) Trial: rationale and design of an international, large, simple, randomised trial of homocysteine-lowering multivitamin therapy in patients with recent transient ischaemic attack or stroke. Cerebrovasc Dis 2002; 13(2):120–6. 326 Galan P, de Bree A, Mennen L, Potier de Courcy G, Preziozi P, Bertrais S et al. Background and rationale of the SU.FOL.OM3 study: double-blind randomized placebocontrolled secondary prevention trial to test the impact of supplementation with folate, vitamin B6 and B12 and/or omega-3 fatty acids on the prevention of recurrent ischemic events in subjects with atherosclerosis in the coronary or cerebral arteries. J Nutr Health Aging 2003; 7(6):428–35. 327 Davey SG, Ebrahim S. Folate supplementation and cardiovascular disease. Lancet 2005; 366(9498):1679–81. 328 Brown BG, Crowley J. Is there any hope for vitamin E? J Am Med Assoc 2005; 293(11):1387–90. 329 Witztum JL. The oxidation hypothesis of atherosclerosis. Lancet 1994; 344(8925):793–5. 330 Jha P, Flather M, Lonn E, Farkouh M, Yusuf S. The antioxidant vitamins and cardiovascular disease. A critical review of epidemiologic and clinical trial data. Ann Intern Med 1995; 123(11):860–72. 331 Eidelman RS, Hollar D, Hebert PR, Lamas GA, Hennekens CH. Randomized trials of vitamin E in the treatment and prevention of cardiovascular disease. Arch Intern Med 2004; 164(14):1552–6. 332 Lonn E, Bosch J, Yusuf S, Sheridan P, Pogue J, Arnold JM et al. Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial. J Am Med Assoc 2005; 293(11):1338–47. 333 Vivekananthan DP, Penn MS, Sapp SK, Hsu A, Topol EJ. Use of antioxidant vitamins for the prevention of cardiovascular disease: meta-analysis of randomised trials. Lancet 2003; 361(9374):2017–23. 334 Kanters SD, Banga JD, Stolk RP, Algra A. Incidence and determinants of mortality and cardiovascular events in diabetes mellitus: a meta-analysis. Vasc Med 1999; 4(2):67–75. 335 Lawes CM, Parag V, Bennett DA, Suh I, Lam TH, Whitlock G et al. Blood glucose and risk of cardiovascular disease in the Asia Pacific region. Diabetes Care 2004; 27(12):2836–42.

..

..

336 Tanne D, Koren-Morag N, Goldbourt U. Fasting plasma glucose and risk of incident ischemic stroke or transient ischemic attacks: a prospective cohort study. Stroke 2004; 35(10):2351–5. 337 Tuomilehto J, Rastenyte D. Diabetes and glucose intolerance as risk factors for stroke. J Cardiovasc Risk 1999; 6(4):241–9. 338 Sigal R, Malcolm J, Arnaout A. Prevention of cardiovascular events in diabetes. Clin Evid 2006; 15:623–45. 339 UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998; 352(9131):837–53. 340 Abraira C, Colwell J, Nuttall F, Sawin CT, Henderson W, Comstock JP et al. Cardiovascular events and correlates in the Veterans Affairs Diabetes Feasibility Trial. Veterans Affairs Cooperative Study on Glycemic Control and Complications in Type II Diabetes. Arch Intern Med 1997; 157(2):181–8. 341 UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998; 352(9131):854–65. 342 Dormandy JA, Charbonnel B, Eckland DJ, Erdmann E, Massi-Benedetti M, Moules IK et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 2005; 366(9493):1279–89. 343 Cordina J, Mead G. Pharmacological cardioversion for atrial fibrillation and flutter. Cochrane Database Syst Rev 2005; (2):CD003713. 344 Mead GE, Elder AT, Flapan AD, Kelman A. Electrical cardioversion for atrial fibrillation and flutter. Cochrane Database Syst Rev 2005; (3):CD002903. 345 Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002; 347(23):1825–33. 346 Mas J-L. Patent forman ovale and stroke. Pract Neurol 2003; 3:4–11. 347 Overell JR, Bone I, Lees KR. Interatrial septal abnormalities and stroke: a meta-analysis of case–control studies. Neurology 2000; 55(8):1172–9. 348 Amarenco P. Patent foramen ovale and the risk of stroke: smoking gun guilty by association? Heart 2005; 91(4):441–3. 349 Mas JL, Arquizan C, Lamy C, Zuber M, Cabanes L, Derumeaux G et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med 2001; 345(24):1740–6. 350 Homma S, Sacco RL, Di Tullio MR, Sciacca RR, Mohr JP. Effect of medical treatment in stroke patients with patent foramen ovale: patent foramen ovale in Cryptogenic Stroke Study. Circulation 2002; 105(22):2625–31. 351 Meier B. Closure of patent foramen ovale: technique, pitfalls, complications, and follow up. Heart 2005; 91(4):444–8.

893

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 894

894

Chapter 16 Preventing recurrent stroke and other serious vascular events

352 Martin F, Sanchez PL, Doherty E, Colon-Hernandez PJ, Delgado G, Inglessis I et al. Percutaneous transcatheter closure of patent foramen ovale in patients with paradoxical embolism. Circulation 2002; 106(9):1121–6. 353 Khairy P, O’Donnell CP, Landzberg MJ. Transcatheter closure versus medical therapy of patent foramen ovale and presumed paradoxical thromboemboli: a systematic review. Ann Intern Med 2003; 139(9):753–60. 354 Flachskampf FA, Daniel WG. Closure of patent foramen ovale: is the case really closed as well? Heart 2005; 91(4):449–50. 355 Wu LA, Malouf JF, Dearani JA, Hagler DJ, Reeder GS, Petty GW et al. Patent foramen ovale in cryptogenic stroke: current understanding and management options. Arch Intern Med 2004; 164(9):950–6. 356 Thompson JE. The evolution of surgery for the treatment and prevention of stroke. The Willis Lecture. Stroke 1996; 27(8):1427–34. 357 Chao WH, Kwan ST, Lyman RS, Loucks HH. Thrombosis of the left internal carotid artery. Arch Surg 1938; 37:100–11. 358 Carrea R, Molins M, Murphy G. Surgical treatment of spontaneous thrombosis of the internal carotid artery in the neck. Carotid-carotideal anastomosis. Report of a case. Acta Neurologica Latin America 1955; 1:71–8. 359 DeBakey ME. Successful carotid endarterectomy for cerebrovascular insufficiency. Nineteen-year follow-up. J Am Med Assoc 1975; 233(10):1083–5. 360 Eastcott HH, Pickering GW, Rob CG. Reconstruction of internal carotid artery in a patient with intermittent attacks of hemiplegia. Lancet 1954; 267(6846):994–6. 361 Pokras R, Dyken ML. Dramatic changes in the performance of endarterectomy for diseases of the extracranial arteries of the head. Stroke 1988; 19(10):1289–90. 362 Tu JV, Hannan EL, Anderson GM, Iron K, Wu K, Vranizan K et al. The fall and rise of carotid endarterectomy in the United States and Canada. N Engl J Med 1998; 339(20):1441–7. 363 Fields WS, Maslenikov V, Meyer JS, Hass WK, Remington RD, Macdonald M. Joint study of extracranial arterial occlusion. V. Progress report of prognosis following surgery or nonsurgical treatment for transient cerebral ischemic attacks and cervical carotid artery lesions. J Am Med Assoc 1970; 211(12):1993–2003. 364 Shaw DA, Venables GS, Cartlidge NE, Bates D, Dickinson PH. Carotid endarterectomy in patients with transient cerebral ischaemia. J Neurol Sci 1984; 64(1):45–53. 365 Warlow C. Carotid endarterectomy: does it work? Stroke 1984; 15(6):1068–76. 366 UK-TIA Study Group. Variation in the use of angiography and carotid endarterectomy by neurologists in the UK-TIA aspirin trial. Br Med J (Clin Res Ed) 1983; 286(6364):514–17. 367 Barnett H, Meldrum H. Status of carotid endarterectomy. Curr Opin Neurol 1994; 7(1):54–9. 368 Winslow CM, Solomon DH, Chassin MR, Kosecoff J, Merrick NJ, Brook RH. The appropriateness of carotid endarterectomy. N Engl J Med 1988; 318(12):721–7. 369 European Carotid Surgery Trialists’ Collaborative Group. MRC European Carotid Surgery Trial: interim results for

370

371

372

373

374

375

376

377 378

379

380

381

382 383 384

385

symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis. Lancet 1991; 337(8752):1235–43. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991; 325(7):445–53. Mayberg MR, Wilson SE, Yatsu F, Weiss DG, Messina L, Hershey LA et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. Veterans Affairs Cooperative Studies Program 309 Trialist Group. J Am Med Assoc 1991; 266(23):3289–94. Cebul RD, Snow RJ, Pine R, Hertzer NR, Norris DG. Indications, outcomes, and provider volumes for carotid endarterectomy. J Am Med Assoc 1998; 279(16):1282–7. Matsen SL, Chang DC, Perler BA, Roseborough GS, Williams GM. Trends in the in-hospital stroke rate following carotid endarterectomy in California and Maryland. J Vasc Surg 2006; 44(3):488–95. Ferris G, Roderick P, Smithies A, George S, Gabbay J, Couper N et al. An epidemiological needs assessment of carotid endarterectomy in an English health region. Is the need being met? Br Med J 1998; 317(7156):447–51. Fairhead JF, Rothwell PM. Underinvestigation and undertreatment of carotid disease in elderly patients with transient ischaemic attack and stroke: comparative population based study. Br Med J 2006; 333(7567):525–7. Eliasziw M, Spence JD, Barnett HJ. Carotid endarterectomy does not affect long-term blood pressure: observations from the NASCET. North American Symptomatic Carotid Endarterectomy Trial. Cerebrovasc Dis 1998; 8(1):20–4. Loftus CM, Quest DO. Technical controversies in carotid artery surgery. Neurosurgery 1987; 20(3):490–5. Darling RC, III, Paty PS, Shah DM, Chang BB, Leather RP. Eversion endarterectomy of the internal carotid artery: technique and results in 449 procedures. Surgery 1996; 120(4):635–9. Cao P, Giordano G, De Rango P, Zannetti S, Chiesa R, Coppi G et al. A randomized study on eversion versus standard carotid endarterectomy: study design and preliminary results: the Everest Trial. J Vasc Surg 1998; 27(4):595–605. Cao PG, De Rango P, Zannetti S, Giordano G, Ricci S, Celani MG. Eversion versus conventional carotid endarterectomy for preventing stroke. Cochrane Database Syst Rev 2000;(4):CD001921. Brothers TE. Initial experience with eversion carotid endarterectomy: absence of a learning curve for the first 100 patients. J Vasc Surg 2005; 42:429–34. Ferguson GG. Carotid endarterectomy. To shunt or not to shunt? Arch Neurol 1986; 43(6):615–17. Ojemann RG, Heros RC. Carotid endarterectomy. To shunt or not to shunt? Arch Neurol 1986; 43(6):617–18. Naylor AR, Bell PR, Ruckley CV. Monitoring and cerebral protection during carotid endarterectomy. Br J Surg 1992; 79(8):735–41. Belardi P, Lucertini G, Ermirio D. Stump pressure and transcranial Doppler for predicting shunting in carotid

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 895

References

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

..

..

endarterectomy. Eur J Vasc Endovasc Surg 2003; 25(2):164–7. Bond R, Warlow CP, Naylor AR, Rothwell PM. Variation in surgical and anaesthetic technique and associations with operative risk in the European carotid surgery trial: implications for trials of ancillary techniques. Eur J Vasc Endovasc Surg 2002; 23(2):117–26. Schneider PA, Ringelstein EB, Rossman ME, Dilley RB, Sobel DF, Otis SM et al. Importance of cerebral collateral pathways during carotid endarterectomy. Stroke 1988; 19(11):1328–34. Bond R, Rerkasem K, Counsell C, Salinas R, Naylor R, Warlow CP et al. Routine or selective carotid artery shunting for carotid endarterectomy (and different methods of monitoring in selective shunting). Cochrane Database Syst Rev 2002; (2):CD000190. Cunningham EJ, Bond R, Mehta Z, Mayberg MR, Warlow CP, Rothwell PM. Long-term durability of carotid endarterectomy for symptomatic stenosis and risk factors for late postoperative stroke. Stroke 2002; 33(11):2658–63. Bond R, Rerkasem K, Rothwell PM. Systematic review of the risks of carotid endarterectomy in relation to the clinical indication for and timing of surgery. Stroke 2003; 34(9):2290–301. Yadav JS, Roubin GS, King P, Iyer S, Vitek J. Angioplasty and stenting for restenosis after carotid endarterectomy. Initial experience. Stroke 1996; 27(11):2075–9. Frericks H, Kievit J, van Baalen JM, van Bockel JH. Carotid recurrent stenosis and risk of ipsilateral stroke: a systematic review of the literature. Stroke 1998; 29(1):244–50. Hunter GC, Palmaz JC, Hayashi HH, Raviola CA, Vogt PJ, Guernsey JM. The etiology of symptoms in patients with recurrent carotid stenosis. Arch Surg 1987; 122(3):311–15. Bond R, Rerkasem K, Aburahma AF, Naylor AR, Rothwell PM. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2004; (2):CD000160. Bond R, Rerkasem K, Naylor R, Rothwell PM. Patches of different types for carotid patch angioplasty. Cochrane Database Syst Rev 2004; (2):CD000071. Rerkasem K, Bond R, Rothwell PM. Local versus general anaesthesia for carotid endarterectomy. Cochrane Database Syst Rev 2004; (2):CD000126. Naylor AR, Ruckley CV. Complications after carotid surgery. In: Campbell B, ed. Complications in Arterial Surgery. Oxford: Butterworth-Heinemann, 1996, pp. 73–88. Ferguson GG, Eliasziw M, Barr HW, Clagett GP, Barnes RW, Wallace MC et al. The North American Symptomatic Carotid Endarterectomy Trial: surgical results in 1415 patients. Stroke 1999; 30(9):1751–8. Bond R, Narayan SK, Rothwell PM, Warlow CP. Clinical and radiographic risk factors for operative stroke and death in the European carotid surgery trial. Eur J Vasc Endovasc Surg 2002; 23(2):108–16. Bond R, Rerkasem K, Shearman CP, Rothwell PM. Time trends in the published risks of stroke and death due to endarterectomy for symptomatic carotid stenosis. Cerebrovasc Dis 2004; 18(1):37–46.

401 Rothwell PM, Slattery J, Warlow CP. A systematic review of the risks of stroke and death due to endarterectomy for symptomatic carotid stenosis. Stroke 1996; 27(2):260–5. 402 Campbell WB. Can reported carotid surgical results be misleading? In: Greenhalgh RM, Hollier LH, eds. Surgery for Stroke. London: W B Saunders, 1993, pp. 331–7. 403 Rothwell P, Warlow C. Is self-audit reliable? Lancet 1995; 346(8990):1623. 404 Rothwell PM, Warlow CP. Interpretation of operative risks of individual surgeons. European Carotid Surgery Trialists’ Collaborative Group. Lancet 1999; 353(9161):1325. 405 Kucey DS, Bowyer B, Iron K, Austin P, Anderson G, Tu JV. Determinants of outcome after carotid endarterectomy. J Vasc Surg 1998; 28(6):1051–8. 406 Killeen SD, Andrews EJ, Redmond HP, Fulton GJ. Provider volume and outcomes for abdominal aortic aneurysm repair, carotid endarterectomy, and lower extremity revascularization procedures. J Vasc Surg 2007; 45(3):615–26. 407 Steed DL, Peitzman AB, Grundy BL, Webster MW. Causes of stroke in carotid endarterectomy. Surgery 1982; 92(4):634–41. 408 Krul JM, van Gijn J, Ackerstaff RG, Eikelboom BC, Theodorides T, Vermeulen FE. Site and pathogenesis of infarcts associated with carotid endarterectomy. Stroke 1989; 20(3):324–8. 409 Riles TS, Imparato AM, Jacobowitz GR, Lamparello PJ, Giangola G, Adelman MA et al. The cause of perioperative stroke after carotid endarterectomy. J Vasc Surg 1994; 19(2):206–14. 410 Spencer MP. Transcranial Doppler monitoring and causes of stroke from carotid endarterectomy. Stroke 1997; 28(4):685–91. 411 Gaunt ME, Naylor AR, Sayers RD, Ratliff DA, Bell PR. Sources of air embolization during carotid surgery: the role of transcranial Doppler ultrasonography. Br J Surg 1993; 80(9):1121. 412 Jansen C, Ramos LM, van Heesewijk JP, Moll FL, van Gijn J, Ackerstaff RG. Impact of microembolism and hemodynamic changes in the brain during carotid endarterectomy. Stroke 1994; 25(5):992–7. 413 Levi CR, O’Malley HM, Fell G, Roberts AK, Hoare MC, Royle JP et al. Transcranial Doppler detected cerebral microembolism following carotid endarterectomy: high microembolic signal loads predict postoperative cerebral ischaemia. Brain 1997; 120(Pt 4):621–9. 414 Wilson PV, Ammar AD. The incidence of ischemic stroke versus intracerebral hemorrhage after carotid endarterectomy: a review of 2452 cases. Ann Vasc Surg 2005; 19(1):1–4. 415 Ouriel K, Shortell CK, Illig KA, Greenberg RK, Green RM. Intracerebral hemorrhage after carotid endarterectomy: incidence, contribution to neurologic morbidity, and predictive factors. J Vasc Surg 1999; 29(1):82–7. 416 Solomon RA, Loftus CM, Quest DO, Correll JW. Incidence and etiology of intracerebral hemorrhage following carotid endarterectomy. J Neurosurg 1986; 64(1):29–34.

895

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 896

896

Chapter 16 Preventing recurrent stroke and other serious vascular events

417 Hafner DH, Smith RB, III, King OW, Perdue GD, Stewart MT, Rosenthal D et al. Massive intracerebral hemorrhage following carotid endarterectomy. Arch Surg 1987; 122(3):305–7. 418 Piepgras DG, Morgan MK, Sundt TM, Jr., Yanagihara T, Mussman LM. Intracerebral hemorrhage after carotid endarterectomy. J Neurosurg 1988; 68(4):532–6. 419 Jansen C, Sprengers AM, Moll FL, Vermeulen FE, Hamerlijnck RP, van Gijn J et al. Prediction of intracerebral haemorrhage after carotid endarterectomy by clinical criteria and intraoperative transcranial Doppler monitoring. Eur J Vasc Surg 1994; 8(3):303–8. 420 Adhiyaman V, Alexander S. Cerebral hyperperfusion syndrome following carotid endarterectomy. Q J Med 2007; 100(4):239–44. 421 Andrews BT, Levy ML, Dillon W, Weinstein PR. Unilateral normal perfusion pressure breakthrough after carotid endarterectomy: case report. Neurosurgery 1987; 21(4):568–71. 422 Schroeder T, Sillesen H, Sorensen O, Engell HC. Cerebral hyperperfusion following carotid endarterectomy. J Neurosurg 1987; 66(6):824–9. 423 Naylor AR, Whyman MR, Wildsmith JA, McClure JH, Jenkins AM, Merrick MV et al. Factors influencing the hyperaemic response after carotid endarterectomy. Br J Surg 1993; 80(12):1523–7. 424 Chambers BR, Smidt V, Koh P. Hyperfusion postendarterectomy. Cerebrovasc Dis 1994; 4:32–7. 425 Breen JC, Caplan LR, Dewitt LD, Belkin M, Mackey WC, O’Donnell TP. Brain edema after carotid surgery. Neurology 1996; 46(1):175–81. 426 Van Mook WN, Rennenberg RJ, Schurink GW, van Oostenbrugge RJ, Mess WH, Hofman PA et al. Cerebral hyperperfusion syndrome. Lancet Neurol 2005; 4(12):877–88. 427 Lopez-Valdes E, Chang HM, Pessin MS, Caplan LR. Cerebral vasoconstriction after carotid surgery. Neurology 1997; 49(1):303–4. 428 Rothwell PM, Slattery J, Warlow CP. A systematic comparison of the risks of stroke and death due to carotid endarterectomy for symptomatic and asymptomatic stenosis. Stroke 1996; 27(2):266–9. 429 Wennberg DE, Lucas FL, Birkmeyer JD, Bredenberg CE, Fisher ES. Variation in carotid endarterectomy mortality in the Medicare population: trial hospitals, volume, and patient characteristics. J Am Med Assoc 1998; 279(16):1278–81. 430 Riles TS, Kopelman I, Imparato AM. Myocardial infarction following carotid endarterectomy: a review of 683 operations. Surgery 1979; 85(3):249–52. 431 Urbinati S, Di PG, Andreoli A, Lusa AM, Carini G, Grazi P et al. Preoperative noninvasive coronary risk stratification in candidates for carotid endarterectomy. Stroke 1994; 25(10):2022–7. 432 Paciaroni M, Eliasziw M, Kappelle LJ, Finan JW, Ferguson GG, Barnett HJ. Medical complications associated with carotid endarterectomy. North American Symptomatic Carotid Endarterectomy Trial (NASCET). Stroke 1999; 30(9):1759–63.

433 Cunningham EJ, Bond R, Mayberg MR, Warlow CP, Rothwell PM. Risk of persistent cranial nerve injury after carotid endarterectomy. J Neurosurg 2004; 101(3):445–8. 434 Gutrecht JA, Jones HR, Jr. Bilateral hypoglossal nerve injury after bilateral carotid endarterectomy. Stroke 1988; 19(2):261–2. 435 Maniglia AJ, Han DP. Cranial nerve injuries following carotid endarterectomy: an analysis of 336 procedures. Head Neck 1991; 13(2):121–4. 436 Sweeney PJ, Wilbourn AJ. Spinal accessory (11th) nerve palsy following carotid endarterectomy. Neurology 1992; 42(3 Pt 1):674–5. 437 Graver LM, Mulcare RJ. Pseudoaneurysm after carotid endarterectomy. J Cardiovasc Surg (Torino) 1986; 27(3):294–7. 438 Martin-Negrier ML, Belleannee G, Vital C, Orgogozo JM. Primitive malignant fibrous histiocytoma of the neck with carotid occlusion and multiple cerebral ischemic lesions. Stroke 1996; 27(3):536–7. 439 Lindblad B, Persson NH, Takolander R, Bergqvist D. Does low-dose acetylsalicylic acid prevent stroke after carotid surgery? A double-blind, placebo-controlled randomized trial. Stroke 1993; 24(8):1125–8. 440 De Marinis M, Zaccaria A, Faraglia V, Fiorani P, Maira G, Agnoli A. Post-endarterectomy headache and the role of the oculosympathetic system. J Neurol Neurosurg Psychiatry 1991; 54(4):314–17. 441 Ille O, Woimant F, Pruna A, Corabianu O, Idatte JM, Haguenau M. Hypertensive encephalopathy after bilateral carotid endarterectomy. Stroke 1995; 26(3):488–91. 442 Youkey JR, Clagett GP, Jaffin JH, Parisi JE, Rich NM. Focal motor seizures complicating carotid endarterectomy. Arch Surg 1984; 119(9):1080–4. 443 Naylor AR, Evans J, Thompson MM, London NJ, Abbott RJ, Cherryman G et al. Seizures after carotid endarterectomy: hyperperfusion, dysautoregulation or hypertensive encephalopathy? Eur J Vasc Endovasc Surg 2003; 26(1):39–44. 444 Truax BT. Gustatory pain: a complication of carotid endarterectomy. Neurology 1989; 39(9):1258–60. 445 European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998; 351(9113):1379–87. 446 Barnett HJ, Taylor DW, Eliasziw M, Fox AJ, Ferguson GG, Haynes RB et al. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med 1998; 339(20):1415–25. 447 Rothwell PM, Eliasziw M, Gutnikov SA, Fox AJ, Taylor DW, Mayberg MR et al. Analysis of pooled data from the randomised controlled trials of endarterectomy for symptomatic carotid stenosis. Lancet 2003; 361(9352):107–16. 448 Markus HS, Thomson ND, Brown MM. Asymptomatic cerebral embolic signals in symptomatic and

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 897

References

449

450

451

452 453

454

455

456

457

458

459

460

461

462

463

..

..

asymptomatic carotid artery disease. Brain 1995; 118(Pt 4):1005–11. Van Zuilen EV, Moll FL, Vermeulen FE, Mauser HW, van Gijn J, Ackerstaff RG. Detection of cerebral microemboli by means of transcranial Doppler monitoring before and after carotid endarterectomy. Stroke 1995; 26(2):210–13. Lloyd AJ, Hayes PD, London NJ, Bell PR, Naylor AR. Does carotid endarterectomy lead to a decline in cognitive function or health related quality of life? J Clin Exp Neuropsychol 2004; 26(6):817–25. Bossema ER, Brand N, Moll FL, Ackerstaff RG, van Doornen LJ. Perioperative microembolism is not associated with cognitive outcome three months after carotid endarterectomy. Eur J Vasc Endovasc Surg 2005; 29(3):262–8. Lal BK. Cognitive function after carotid artery revascularization. Vasc Endovascular Surg 2007; 41(1):5–13. Bo M, Massaia M, Speme S, Cappa G, Strumia K, Cerrato P et al. Risk of cognitive decline in older patients after carotid endarterectomy: an observational study. J Am Geriatr Soc 2006; 54(6):932–6. Lunn S, Crawley F, Harrison MJ, Brown MM, Newman SP. Impact of carotid endarterectomy upon cognitive functioning. A systematic review of the literature. Cerebrovasc Dis 1999; 9(2):74–81. Schroeder T. Hemodynamic significance of internal carotid artery disease. Acta Neurol Scand 1988; 77(5):353–72. Naylor AR, Merrick MV, Sandercock PA, Gillespie I, Allen P, Griffin TM et al. Serial imaging of the carotid bifurcation and cerebrovascular reserve after carotid endarterectomy. Br J Surg 1993; 80(10):1278–82. Yonas H, Smith HA, Durham SR, Pentheny SL, Johnson DW. Increased stroke risk predicted by compromised cerebral blood flow reactivity. J Neurosurg 1993; 79(4):483–9. Hartl WH, Janssen I, Furst H. Effect of carotid endarterectomy on patterns of cerebrovascular reactivity in patients with unilateral carotid artery stenosis. Stroke 1994; 25(10):1952–7. Yamauchi H, Fukuyama H, Nagahama Y, Nabatame H, Nakamura K, Yamamoto Y et al. Evidence of misery perfusion and risk for recurrent stroke in major cerebral arterial occlusive diseases from PET. J Neurol Neurosurg Psychiatry 1996; 61(1):18–25. Visser GH, van Huffelen AC, Wieneke GH, Eikelboom BC. Bilateral increase in CO2 reactivity after unilateral carotid endarterectomy. Stroke 1997; 28(5):899–905. Silvestrini M, Vernieri F, Pasqualetti P, Matteis M, Passarelli F, Troisi E et al. Impaired cerebral vasoreactivity and risk of stroke in patients with asymptomatic carotid artery stenosis. J Am Med Assoc 2000; 283(16):2122–7. Markus H, Cullinane M. Severely impaired cerebrovascular reactivity predicts stroke and TIA risk in patients with carotid artery stenosis and occlusion. Brain 2001; 124(Pt 3):457–67. Benade MM, Warlow CP. Cost of identifying patients for carotid endarterectomy. Stroke 2002; 33(2):435–9.

464 Benade MM, Warlow CP. Costs and benefits of carotid endarterectomy and associated preoperative arterial imaging: a systematic review of health economic literature. Stroke 2002; 33(2):629–38. 465 Rothwell PM, Eliasziw M, Gutnikov SA, Warlow CP, Barnett HJ. Sex difference in the effect of time from symptoms to surgery on benefit from carotid endarterectomy for transient ischemic attack and nondisabling stroke. Stroke 2004; 35(12):2855–61. 466 Rothwell PM, Eliasziw M, Gutnikov SA, Warlow CP, Barnett HJ. Endarterectomy for symptomatic carotid stenosis in relation to clinical subgroups and timing of surgery. Lancet 2004; 363(9413):915–24. 467 Bond R, Rerkasem K, Cuffe R, Rothwell PM. A systematic review of the associations between age and sex and the operative risks of carotid endarterectomy. Cerebrovasc Dis 2005; 20(2):69–77. 468 Sundt TM, Sandok BA, Whisnant JP. Carotid endarterectomy. Complications and preoperative assessment of risk. Mayo Clin Proc 1975; 50(6):301–6. 469 McCrory DC, Goldstein LB, Samsa GP, Oddone EZ, Landsman PB, Moore WS et al. Predicting complications of carotid endarterectomy. Stroke 1993; 24(9):1285–91. 470 Goldstein LB, McCrory DC, Landsman PB, Samsa GP, Ancukiewicz M, Oddone EZ et al. Multicenter review of preoperative risk factors for carotid endarterectomy in patients with ipsilateral symptoms. Stroke 1994; 25(6):1116–21. 471 Golledge J, Cuming R, Beattie DK, Davies AH, Greenhalgh RM. Influence of patient-related variables on the outcome of carotid endarterectomy. J Vasc Surg 1996; 24(1):120–6. 472 Rothwell PM, Slattery J, Warlow CP. Clinical and angiographic predictors of stroke and death from carotid endarterectomy: systematic review. Br Med J 1997; 315(7122):1571–7. 473 Pritz MB. Timing of carotid endarterectomy after stroke. Stroke 1997; 28(12):2563–67. 474 Eckstein HH, Schumacher H, Klemm K, Laubach H, Kraus T, Ringleb P et al. Emergency carotid endarterectomy. Cerebrovasc Dis 1999; 9(5):270–81. 475 Fairhead JF, Rothwell PM. The need for urgency in identification and treatment of symptomatic carotid stenosis is already established. Cerebrovasc Dis 2005; 19(6):355–8. 476 Blaser T, Hofmann K, Buerger T, Effenberger O, Wallesch CW, Goertler M. Risk of stroke, transient ischemic attack, and vessel occlusion before endarterectomy in patients with symptomatic severe carotid stenosis. Stroke 2002; 33(4):1057–62. 477 Fairhead JF, Mehta Z, Rothwell PM. Population-based study of delays in carotid imaging and surgery and the risk of recurrent stroke. Neurology 2005; 65(3):371–5. 478 Welsh S, Mead G, Chant H, Picton A, O’Neill PA, McCollum CN. Early carotid surgery in acute stroke: a multicentre randomised pilot study. Cerebrovasc Dis 2004; 18(3):200–5.

897

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 898

898

Chapter 16 Preventing recurrent stroke and other serious vascular events

479 Rantner B, Pavelka M, Posch L, Schmidauer C, Fraedrich G. Carotid endarterectomy after ischemic stroke: is there a justification for delayed surgery? Eur J Vasc Endovasc Surg 2005; 30(1):36–40. 480 Rodgers H, Oliver SE, Dobson R, Thomson RG. A regional collaborative audit of the practice and outcome of carotid endarterectomy in the United Kingdom. Northern Regional Carotid Endarterectomy Audit Group. Eur J Vasc Endovasc Surg 2000; 19(4):362–9. 481 Turnbull RG, Taylor DC, Hsiang YN, Salvian AJ, Nanji S, O’Hanley G et al. Assessment of patient waiting times for vascular surgery. Can J Surg 2000; 43(2):105–11. 482 Pell JP, Slack R, Dennis M, Welch G. Improvements in carotid endarterectomy in Scotland: results of a national prospective survey. Scott Med J 2004; 49(2):53–6. 483 Rothwell PM, Warlow CP. Low risk of ischemic stroke in patients with reduced internal carotid artery lumen diameter distal to severe symptomatic carotid stenosis: cerebral protection due to low poststenotic flow? On behalf of the European Carotid Surgery Trialists’ Collaborative Group. Stroke 2000; 31(3):622–30. 484 Morgenstern LB, Fox AJ, Sharpe BL, Eliasziw M, Barnett HJ, Grotta JC. The risks and benefits of carotid endarterectomy in patients with near occlusion of the carotid artery. North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Neurology 1997; 48(4):911–15. 485 Eliasziw M, Streifler JY, Fox AJ, Hachinski VC, Ferguson GG, Barnett HJ. Significance of plaque ulceration in symptomatic patients with high-grade carotid stenosis. North American Symptomatic Carotid Endarterectomy Trial. Stroke 1994; 25(2):304–8. 486 Rothwell PM, Gibson R, Warlow CP. Interrelation between plaque surface morphology and degree of stenosis on carotid angiograms and the risk of ischemic stroke in patients with symptomatic carotid stenosis. On behalf of the European Carotid Surgery Trialists’ Collaborative Group. Stroke 2000; 31(3):615–21. 487 Rothwell PM, Gibson RJ, Slattery J, Sellar RJ, Warlow CP. Equivalence of measurements of carotid stenosis: a comparison of three methods on 1001 angiograms. European Carotid Surgery Trialists’ Collaborative Group. Stroke 1994; 25(12):2435–9. 488 Rothwell PM, Gutnikov SA, Warlow CP. Reanalysis of the final results of the European Carotid Surgery Trial. Stroke 2003; 34(2):514–23. 489 Johnston DC, Goldstein LB. Clinical carotid endarterectomy decision making: noninvasive vascular imaging versus angiography. Neurology 2001; 56(8):1009–15. 490 Norris JW, Rothwell PM. Noninvasive carotid imaging to select patients for endarterectomy: is it really safer than conventional angiography? Neurology 2001; 56(8):990–1. 491 Rothwell PM, Pendlebury ST, Wardlaw J, Warlow CP. Critical appraisal of the design and reporting of studies of imaging and measurement of carotid stenosis. Stroke 2000; 31(6):1444–50.

492 Ascher E, Markevich N, Hingorani A, Kallakuri S. Pseudoocclusions of the internal carotid artery: a rationale for treatment on the basis of a modified carotid duplex scan protocol. J Vasc Surg 2002; 35(2):340–5. 493 Berman SS, Devine JJ, Erdoes LS, Hunter GC. Distinguishing carotid artery pseudo-occlusion with color-flow Doppler. Stroke 1995; 26(3):434–8. 494 Rothwell PM. ACST: which subgroups will benefit most from carotid endarterectomy? Lancet 2004; 364(9440):1122–3. 495 Rothwell PM, Warlow CP. Prediction of benefit from carotid endarterectomy in individual patients: a riskmodelling study. European Carotid Surgery Trialists’ Collaborative Group. Lancet 1999; 353(9170):2105–10. 496 Mackinnon AD, Aaslid R, Markus HS. Ambulatory transcranial Doppler cerebral embolic signal detection in symptomatic and asymptomatic carotid stenosis. Stroke 2005; 36(8):1726–30. 497 Molloy J, Markus HS. Asymptomatic embolization predicts stroke and TIA risk in patients with carotid artery stenosis. Stroke 1999; 30(7):1440–3. 498 Mead GE, Lewis SC, Wardlaw JM, Dennis MS, Warlow CP. Severe ipsilateral carotid stenosis and middle cerebral artery disease in lacunar ischaemic stroke: innocent bystanders? J Neurol 2002; 249(3):266–71. 499 Boiten J, Rothwell P, Slattery J, Warlow C, for the European Carotid Surgery Trialists’ Collaborative Group. Lacunar stroke in the European Carotid Surgery Trial: risk factos, distribution of carotid stenosis, effect of surgery and type of recurrent stroke. Cerebrovasc Dis 1996; 6:281–7. 500 Inzitari D, Eliasziw M, Sharpe BL, Fox AJ, Barnett HJ. Risk factors and outcome of patients with carotid artery stenosis presenting with lacunar stroke. North American Symptomatic Carotid Endarterectomy Trial Group. Neurology 2000; 54(3):660–6. 501 Chaturvedi S, Bruno A, Feasby T, Holloway R, Benavente O, Cohen SN et al. Carotid endarterectomy: an evidencebased review. Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 2005; 65(6):794–801. 502 Chambers BR, Donnan GA. Carotid endarterectomy for asymptomatic carotid stenosis. Cochrane Database Syst Rev 2005; (4):CD001923. 503 Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. J Am Med Assoc 1995; 273(18):1421–8. 504 Moore WS, Young B, Baker WH, Robertson JT, Toole JF, Vescera CL et al. Surgical results: a justification of the surgeon selection process for the ACAS trial. The ACAS Investigators. J Vasc Surg 1996; 23(2):323–8. 505 MRC Asymptomatic Carotid Surgery Trial (ACST) Collaborative Group. Halliday A, Mansfield A, Marro J, Peto C, Peto R, Potter J et al. Prevention of disabling and fatal strokes by successful carotid endarterectomy in patients without recent neurological symptoms: randomised controlled trial. Lancet 2004; 363(9420):1491–502.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 899

References

506 Bond R, Rerkasem K, Rothwell P. High morbidity due to endarterectomy for asymptomatic carotid stenosis. Cerebrovasc Dis 2003; 16(Suppl 4):65. 507 Kresowik TF, Bratzler DW, Kresowik RA, Hendel ME, Grund SL, Brown KR et al. Multistate improvement in process and outcomes of carotid endarterectomy. J Vasc Surg 2004; 39(2):372–80. 508 Halliday AW, Thomas D, Mansfield A. The Asymptomatic Carotid Surgery Trial (ACST). Rationale and design. Steering Committee. Eur J Vasc Surg 1994; 8(6):703–10. 509 Cuffe RL, Rothwell PM. Effect of nonoptimal imaging on the relationship between the measured degree of symptomatic carotid stenosis and risk of ischemic stroke. Stroke 2006; 37(7):1785–91. 510 Nicolaides AN, Kakkos SK, Griffin M, Sabetai M, Dhanjil S, Tegos T et al. Severity of asymptomatic carotid stenosis and risk of ipsilateral hemispheric ischaemic events: results from the ACSRS study. Eur J Vasc Endovasc Surg 2005; 30(3):275–84. 511 Baker WH, Howard VJ, Howard G, Toole JF. Effect of contralateral occlusion on long-term efficacy of endarterectomy in the asymptomatic carotid atherosclerosis study (ACAS). ACAS Investigators. Stroke 2000; 31(10):2330–4. 512 Spence JD, Tamayo A, Lownie SP, Ng WP, Ferguson GG. Absence of microemboli on transcranial Doppler identifies low-risk patients with asymptomatic carotid stenosis. Stroke 2005; 36(11):2373–8. 513 Abbott AL, Chambers BR, Stork JL, Levi CR, Bladin CF, Donnan GA. Embolic signals and prediction of ipsilateral stroke or transient ischemic attack in asymptomatic carotid stenosis: a multicenter prospective cohort study. Stroke 2005; 36(6):1128–33. 514 Markus H, Cullinane M. Asymptomatic Carotid Emboli (ACES) Study. Cerebrovasc Dis 2000; 10(Suppl 1):3. 515 Gronholdt ML. Ultrasound and lipoproteins as predictors of lipid-rich, rupture-prone plaques in the carotid artery. Arterioscler Thromb Vasc Biol 1999; 19(1):2–13. 516 Nicolaides AN, Kakkos SK, Griffin M, Sabetai M, Dhanjil S, Thomas DJ et al. Effect of image normalization on carotid plaque classification and the risk of ipsilateral hemispheric ischemic events: results from the asymptomatic carotid stenosis and risk of stroke study. Vascular 2005; 13(4):211–21. 517 Takaya N, Yuan C, Chu B, Saam T, Underhill H, Cai J et al. Association between carotid plaque characteristics and subsequent ischemic cerebrovascular events: a prospective assessment with MRI: initial results. Stroke 2006; 37(3):818–23. 518 Cai J, Hatsukami TS, Ferguson MS, Kerwin WS, Saam T, Chu B et al. In vivo quantitative measurement of intact fibrous cap and lipid-rich necrotic core size in atherosclerotic carotid plaque: comparison of highresolution, contrast-enhanced magnetic resonance imaging and histology. Circulation 2005; 112(22):3437–44. 519 Van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an

..

..

520

521

522

523

524

525

526

527

528

529

530

531

532

533

inflammatory process irrespective of the dominant plaque morphology. Circulation 1994; 89(1):36–44. Redgrave JN, Lovett JK, Gallagher PJ, Rothwell PM. Histological assessment of 526 symptomatic carotid plaques in relation to the nature and timing of ischemic symptoms: the Oxford plaque study. Circulation 2006; 113(19):2320–8. Trivedi RA, King-Im JM, Graves MJ, Cross JJ, Horsley J, Goddard MJ et al. In vivo detection of macrophages in human carotid atheroma: temporal dependence of ultrasmall superparamagnetic particles of iron oxideenhanced MRI. Stroke 2004; 35(7):1631–5. Tang T, Howarth SP, Miller SR, Trivedi R, Graves MJ, King-Im JU et al. Assessment of inflammatory burden contralateral to the symptomatic carotid stenosis using high-resolution ultrasmall, superparamagnetic iron oxideenhanced MRI. Stroke 2006; 37(9):2266–70. Dotter CT, Judkins MP, Rosch J. Nonoperative treatment of arterial occlusive disease: a radiologically facilitated technique. Radiol Clin North Am 1967; 5(3):531–42. Mathur A, Roubin GS, Iyer SS, Piamsonboon C, Liu MW, Gomez CR et al. Predictors of stroke complicating carotid artery stenting. Circulation 1998; 97(13):1239–45. McCabe DJ, Brown MM, Clifton A. Fatal cerebral reperfusion hemorrhage after carotid stenting. Stroke 1999; 30(11):2483–6. Qureshi AI, Luft AR, Sharma M, Janardhan V, Lopes DK, Khan J et al. Frequency and determinants of postprocedural hemodynamic instability after carotid angioplasty and stenting. Stroke 1999; 30(10):2086–93. CAVATAS Investigators. Endovascular versus surgical treatment in patients with carotid stenosis in the Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomised trial. Lancet 2001; 357:1729–37. Naylor AR, Bolia A, Abbott RJ, Pye IF, Smith J, Lennard N et al. Randomized study of carotid angioplasty and stenting versus carotid endarterectomy: a stopped trial. J Vasc Surg 1998; 28(2):326–34. Alberts MJ, for the Publications Committee of the WALLSTENT. Results of a multicentre prospective randomised trial of carotid artery stenting vs carotid endarterectomy. Stroke 2001; 32:325. Brooks WH, McClure RR, Jones MR, Coleman TC, Breathitt L. Carotid angioplasty and stenting versus carotid endarterectomy: randomized trial in a community hospital. J Am Coll Cardiol 2001; 38(6):1589–95. Yadav JS, Wholey MH, Kuntz RE, Fayad P, Katzen BT, Mishkel GJ et al. Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med 2004; 351(15):1493–501. Coward LJ, Featherstone RL, Brown MM. Safety and eficacy of endovascular treatment of carotid artery stenosis compared with carotid endarterectomy: a Cochrane systematic review of the randomized evidence. Stroke 2005; 36(4):905–11. Reimers B, Corvaja N, Moshiri S, Sacca S, Albiero R, Di Mario C et al. Cerebral protection with filter devices during carotid artery stenting. Circulation 2001; 104(1):12–15.

899

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 900

900

Chapter 16 Preventing recurrent stroke and other serious vascular events

534 Mas JL, Chatellier G, Beyssen B, Branchereau A, Moulin T, Becquemin JP et al. Endarterectomy versus stenting in patients with symptomatic severe carotid stenosis. N Engl J Med 2006; 355(16):1660–71. 535 Featherstone RL, Brown MM, Coward LJ. International carotid stenting study: protocol for a randomised clinical trial comparing carotid stenting with endarterectomy in symptomatic carotid artery stenosis. Cerebrovasc Dis 2004; 18(1):69–74. 536 CARESS Steering Committee. Carotid Revascularization Using Endarterectomy or Stenting Systems (CaRESS) phase I clinical trial: 1-year results. J Vasc Surg 2005; 42(2):213–19. 537 Major ongoing stroke trials. Stroke 2006; 37(10):e36–e44. 538 Graor RA, Hetzer NR. Management of coexistent carotid artery and coronary artery disease. Stroke 1988; 19(11):1441–4. 539 Akins CW. The case for concomitant carotid and coronary artery surgery. Br Heart J 1995; 74(2):97–8. 540 Davenport RJ, Dennis MS, Sandercock PA, Warlow CP, Starkey IR, Ruckley CV et al. How should a patient presenting with unstable angina and a recent stroke be managed? Br Med J 1995; 310(6992):1449–52. 541 Hertzer NR, Loop FD, Beven EG, O’Hara PJ, Krajewski LP. Surgical staging for simultaneous coronary and carotid disease: a study including prospective randomization. J Vasc Surg 1989; 9(3):455–63. 542 Borger MA, Fremes SE, Weisel RD, Cohen G, Rao V, Lindsay TF et al. Coronary bypass and carotid endarterectomy: does a combined approach increase risk? a meta-analysis. Ann Thorac Surg 1999; 68(1):14–20. 543 Naylor AR, Mehta Z, Rothwell PM, Bell PR. Carotid artery disease and stroke during coronary artery bypass: a critical review of the literature. Eur J Vasc Endovasc Surg 2002; 23(4):283–94. 544 Naylor AR, Cuffe RL, Rothwell PM, Bell PR. A systematic review of outcomes following staged and synchronous carotid endarterectomy and coronary artery bypass. Eur J Vasc Endovasc Surg 2003; 25(5):380–9. 545 Naylor AR. A critical review of the role of carotid disease and the outcomes of staged and synchronous carotid surgery. Semin Cardiothorac Vasc Anesth 2004; 8(1):37–42. 546 Latchaw RE, Ausman JI, Lee MC. Superficial temporal-middle cerebral artery bypass. A detailed analysis of multiple pre- and postoperative angiograms in 40 consecutive patients. J Neurosurg 1979; 51(4):455–65. 547 Hankey GJ, Warlow C. Prognosis of symptomatic carotid artery occlusion: an overview. Cerebrovasc Dis 1991; 1:245–56. 548 Karnik R, Valentin A, Ammerer HP, Donath P, Slany J. Evaluation of vasomotor reactivity by transcranial Doppler and acetazolamide test before and after extracranialintracranial bypass in patients with internal carotid artery occlusion. Stroke 1992; 23(6):812–17. 549 Klijn CJ, Kappelle LJ, Tulleken CA, van Gijn J. Symptomatic carotid artery occlusion. A reappraisal of hemodynamic factors. Stroke 1997; 28(10):2084–93.

550 Powers WJ, Derdeyn CP, Fritsch SM, Carpenter DA, Yundt KD, Videen TO et al. Benign prognosis of neversymptomatic carotid occlusion. Neurology 2000; 54(4):878–82. 551 EC-IC Bypass Study Group. Failure of extracranialintracranial arterial bypass to reduce the risk of ischaemic stroke: results of an international randomised trial. N Engl J Med 1985; 313:1191–200. 552 Warlow CP. Extracranial to intracranial bypass and the prevention of stroke. J Neurol 1986; 233(3):129–30. 553 Derdeyn CP, Grubb RL, Jr., Powers WJ. Indications for cerebral revascularization for patients with atherosclerotic carotid occlusion. Skull Base 2005; 15(1):7–14. 554 Grubb RL, Jr., Powers WJ, Derdeyn CP, Adams HP, Jr., Clarke WR. The Carotid Occlusion Surgery Study. Neurosurg Focus 2003; 14(3):e9. 555 Eberhardt O, Naegele T, Raygrotzki S, Weller M, Ernemann U. Stenting of vertebrobasilar arteries in symptomatic atherosclerotic disease and acute occlusion: case series and review of the literature. J Vasc Surg 2006; 43(6):1145–54. 556 Cloud GC, Crawley F, Clifton A, McCabe DJ, Brown MM, Markus HS. Vertebral artery origin angioplasty and primary stenting: safety and restenosis rates in a prospective series. J Neurol Neurosurg Psychiatry 2003; 74(5):586–90. 557 Coward LJ, Featherstone RL, Brown MM. Percutaneous transluminal angioplasty and stenting for vertebral artery stenosis. Cochrane Database Syst Rev 2005; (2):CD000516. 558 Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005; 353(5):487–97. 559 Benner JS, Glynn RJ, Mogun H, Neumann PJ, Weinstein MC, Avorn J. Long-term persistence in use of statin therapy in elderly patients. J Am Med Assoc 2002; 288(4):455–61. 560 Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy in elderly patients with and without acute coronary syndromes. J Am Med Assoc 2002; 288(4):462–7. 561 Trewby PN, Reddy AV, Trewby CS, Ashton VJ, Brennan G, Inglis J. Are preventive drugs preventive enough? A study of patients’ expectation of benefit from preventive drugs. Clin Med 2002; 2(6):527–33. 562 Hamman GF, Weimar C, Glahn J, Busse O, Diener HC. Adherence to secondary stroke prevention strategies: results from the German Stroke Data Bank. Cerebrovasc Dis 2003; 15(4):282–8. 563 Haynes RB, Yao X, Degani A, Kripalani S, Garg A, McDonald HP. Interventions to enhance medication adherence. Cochrane Database Syst Rev 2005; (4):CD000011. 564 Gigerenzer G, Edwards A. Simple tools for understanding risks: from innumeracy to insight. Br Med J 2003; 327(7417):741–4. 565 Edwards A, Elwyn G, Mulley A. Explaining risks: turning numerical data into meaningful pictures. Br Med J 2002; 324(7341):827–30. 566 Alaszewski A, Horlick-Jones T. How can doctors communicate information about risk more effectively? Br Med J 2003; 327(7417):728–31.

..

..

9781405127660_4_016.qxd 10/13/07 10:51 AM Page 901

References

567 Fuller R, Dudley N, Blacktop J. Avoidance hierarchies and preferences for anticoagulation: semi-qualitative analysis of older patients’ views about stroke prevention and the use of warfarin. Age Ageing 2004; 33(6):608–11. 568 Zhan C, Correa-de-Araujo R, Bierman AS, Sangl J, Miller MR, Wickizer SW et al. Suboptimal prescribing in elderly outpatients: potentially harmful drug-drug and drug-disease combinations. J Am Geriatr Soc 2005; 53(2):262–7. 569 White C. ‘Polypill’ to fight cardiovascular disease. Summary of rapid responses. Br Med J 2003; 327:809.

..

..

570 Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more than 80%. Br Med J 2003; 326(7404):1419. 571 Zarins CK, Gewertz BL. Atlas of Vascular Surgery. New York: Churchill Livingstone, 1987. 572 Schroeder T, Levi N. What steps can I take to minimise inadvertent cranial nerve injury. In: Naylor AR, Mackey AE, eds. Carotid Artery Surgery: A Problem Based Approach. London: Saunders, 1999. 573 The European Carotid Surgery Trialists Collaborative Group. Risk of stroke in the distribution of an asymptomatic carotid artery. Lancet 1995; 345(8944):209–12.

901

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 903

17

The organization of stroke services 17.1

Introduction

17.2

Planning and developing a stroke service

903

17.3

Comprehensive stroke service

17.4

General principles when discussing a comprehensive stroke service

17.5

Interventions to improve access to early specialist assessment and treatment

17.6

Organized inpatient (stroke unit) care

17.7

Transfer from hospital to community

17.8

Continuing rehabilitation and reintegration back to normal life

17.9

Longer-term follow-up and chronic disease management

906

906 910

914 921

17.10 Generic issues in stroke service delivery

923

925

926

17.11 Planning, developing and maintaining a stroke service 17.12 Evaluating and monitoring stroke services 17.13 Stroke guidelines

906

927

934

943

17.14 Integrated care pathways

943

17.15 Impact of a comprehensive stroke service 17.16 Cost-effectiveness of stroke services

17.1 Introduction

17.1.1 The impact of stroke Governments, and in particular those responsible for providing healthcare, have become increasingly aware of the impact that stroke has on the health of the population and the cost to the community. In the UK, stroke patients account for about 6% of hospital costs and almost 5% of National Health Service costs1 (www.nao.org.uk/stroke). Studies from other countries (e.g. Sweden, US, Canada, Netherlands and Japan) suggest that the financial burden may be even greater than in the UK, possibly because of greater expenditure on all health services.2–11 Moreover, demographic changes are likely to cause increasing mortality and morbidity in the developing world.12 Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0.

944

944

Relatively little attention was paid to stroke in the UK until the publication of the King’s Fund Consensus Conference13 in 1988. This highlighted the many deficiencies in the services provided for stroke patients, concluding that ‘services were often haphazard and poorly tailored to the patient’s needs’. Since then, in the UK, stroke has moved up the political agenda, and was identified as a key chapter in the National Service Framework for Older People in England and Wales (www.dh.gov.uk/PolicyAndGuidance/HealthAndSocial CareTopics/OlderPeoplesServices) and the Coronary Heart Disease and Stroke strategy in Scotland (www. scotland.gov.uk/library5/health/chds). There has been a similar and continuing emphasis on stroke in many other countries. These changes have led to a tremendous surge of interest in stroke in general, and in stroke services in particular. Over the last few years, an increasing amount of research has been done to determine the best and most cost-effective ways of providing care for stroke patients. This chapter will cover the organization of services for people who have had a transient ischaemic attack or stroke. Inevitably, the discussion will tend to reflect the UK – and to some extent other North European and 903

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 904

904

Chapter 17 The organization of stroke services

Australian – models of care, but we hope it will have relevance for services elsewhere.

17.1.2 Aims of stroke services The overall aim of stroke services is to deliver the care required by patients and their families in the most efficient, effective, equitable and humane manner possible. Such services may not necessarily be stroke-specific; they could be part of those for internal medicine, care of the elderly, neurology, rehabilitation or continuing care. Good organization is probably the most important factor in determining service effectiveness. When considering exactly how stroke services should best be organized, it is useful to consider the main objectives when caring for patients with stroke and transient ischaemic attack (TIA) (Table 17.1 and Fig. 17.1). In addition, given the lack of evidence for many of our interventions, services should facilitate research and education. We have not included primary prevention among the components of care, although this is potentially the most effective method of reducing stroke-related death, disability and handicap, at least in the long term (Chapter 18). Primary stroke prevention has so much in common with the prevention of other vascular diseases that it makes more sense to link these preventive services together, especially as their success is likely to depend more on political and social change than on health services.

Table 17.1 Key components for managing stroke and transient ischaemic attack. Public awareness to recognize warning signs and seek help promptly (section 17.5.1) Prompt and accurate assessment and diagnosis (including transportation) (Chapter 3) Specific acute medical and surgical treatment (Chapters 11, 12, 13, 14 and 15) Identification and assessment of patients’ problems (chapter 11) Secondary prevention of further vascular events (Chapter 16) General care, including interventions to resolve problems (includes many aspects of rehabilitation) (Chapter 11) Provision of information and advice (section 17.10.1) Terminal care for patients who are unlikely to survive (section 10.4) Hospital discharge and reintegration into the community (section 17.7) Continuing or long-term care for severely disabled patients (section 17.8) Follow-up to detect and manage late-onset problems (section 17.9)

17.1.3 Establishing the effectiveness of stroke services In an ideal world our decisions about the delivery of stroke services would always be informed by robust evidence from randomized trials and priority given to those aspects of care that have been proven to be effective. However, we must recognize that carrying out randomized trials of complex interventions such as stroke services is challenging14 and these trials are often few in number and difficult to interpret. There is no reason why most of the important methodological principles (www.cochrane.org/resources/handbook) of robust clinical trials (random treatment allocation, concealment of treatment allocation, blinding of follow-up, intention-to-treat analyses) should not apply to those of stroke services. However, randomized trials of complex interventions such as stroke services offer some unique challenges. In particular it is often difficult to develop and describe the intervention adequately, to blind the trial participants to their treatment and to rule out confounding from other aspects of care. Therefore even if a service is shown to work well in one setting, specific local factors may have influenced the results. For these reasons we have approached the evaluation of stroke services in the following way: • Has the service component been shown to work in a specific circumstance (in one randomized trial)? • Has it been shown to work in several settings (systematic review of clinical trials)? • Do we know if the benefits justify the costs (economic analyses)? The Stroke Review Group of the Cochrane Collaboration is an excellent source of this type of information, providing an increasing number of systematic reviews (http://www.dcn.ed.ac.uk/csrg). It is important that those who are responsible for planning services should be aware of all the available evidence concerning the effectiveness of interventions, and the methods of delivering these interventions to the appropriate patients. However, often we will not have any reliable evidence from randomized trials to guide our decisions simply because of the methodological challenges outlined above. Planners should be aware that lack of evidence of benefit is not the same as evidence of lack of benefit. If there are reliable data concerning the cost of interventions, and various other aspects of the service, this may allow healthcare planners to make more informed choices about which services should be provided. At the same time, they should not deny patients services which are generally accepted as being effective just because of lack of randomized trial evidence.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 905

17.1 Introduction

Patient develops symptoms Seeks help Public education programme Assess urgency and appropriateness Specialist assessment / triage Transient symptoms and low risk of recurrence

Persisting symptoms and/or high risk of recurrence

Diagnosis/assessment, secondary prevention

Diagnosis/assessment, emergency treatment

Rapid access neurovascular clinic

Acute stroke assessment area*

Diagnosis/assessment Medical management Nursing management Multidisciplinary rehabilitation Prevention of complications Secondary prevention Discharge planning Comprehensive stroke unit Fig. 17.1 A diagram illustrating the components, functions and interrelationships of a hospital-based stroke service that is integrated with community services. The upper box indicates the objectives of the services and the lower (purple) boxes (in italics) a proposed solution. We believe that elements of the rehabilitation process are important even on the day of the stroke onset. *Could be in stroke unit or stroke centre. **Could continue in comprehensive stroke unit.

Early supported discharge

Rehabilitation stroke unit**

Continuing rehabilitation/community reintegration

17.1.4 Pressures that may shape stroke services Several factors other than evidence of effectiveness or cost-effectiveness may shape stroke service delivery and constrain the options available to clinicians and service planners. These can include: • Local healthcare culture and economy: the traditional approach to providing healthcare, and the way it is funded or reimbursed, will influence the way services

..

Continuing multidisciplinary team rehabilitation

Therapy-based rehabilitation/stroke liaison worker services

Clinical trials of services can be complex and challenging – remember that lack of evidence of benefit is not the same as evidence of lack of benefit.

..

Discharge planning Multidisciplinary rehabilitation

Re-assessment of needs/further rehabilitation Primary care/chronic disease management

are delivered. Developing a stroke service in the US or Germany would present different challenges from doing so in Scandinavia, the Netherlands or the UK. There are likely to be particular challenges in the former Soviet bloc countries and the developing world where many basic assumptions about the availability of healthcare resources may simply not apply. • Needs of different patient groups: stroke presents a complex challenge to service planners in that most patients require a similar general service but a small number (for instance those with subarachnoid haemorrhage or those eligible for intravenous thrombolysis) may require more specialist services. There is frequently a tension

905

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 906

906

Chapter 17 The organization of stroke services

between the need to centralize specialist services for the few and the need to devolve more general services for the majority to local hospitals and communities. • Views of patients and families: one of the common complaints from patients and carers is the discontinuity of services they receive (http://www.chss.org.uk/ pdf/research/Improving_services_patients_and_carers_ views.pdf), resulting in fragmentation of care and dissatisfaction with services. • Resources available: it is relatively easier to organize or reorganize a stroke service if the basic staffing levels and investigation services are already available, albeit distributed around the hospital or community. For this reason much of our discussion will be relevant mainly to well-resourced services in developed countries.

can the needs of the population realistically be met? How should these resources best be organized? • How do we know if we have succeeded (section 17.12)? How can the performance of the stroke services be monitored and maintained? In this chapter we will begin (sections 17.3–17.10) with a review of the effectiveness (and where possible cost-effectiveness) of components of stroke services. In the second part (sections 17.11–17.16) we will discuss practical approaches to planning, developing, monitoring and maintaining a stroke service.

17.3 Comprehensive stroke service Local stroke services must be tailored to local conditions; there is no perfect blueprint that can be applied everywhere.

17.2 Planning and developing a stroke service

The way in which a service is best provided will depend on local history, geography, needs, resources, people and politics. Any stroke service must therefore be tailored to the local conditions to achieve maximum effectiveness. For this reason, it is difficult to be dogmatic about exactly how services should be organized. In this chapter we will attempt to provide general guidance about the principles that should be of use to the clinician, public health physician or health service manager (administrator) in planning a service. When planning or reviewing stroke services, it is useful to start by addressing several questions: • Where do we want to be (sections 17.5–17.9)? What is the evidence for the effectiveness (and costeffectiveness) of the components of both the existing and planned stroke service? • Where are we now (sections 17.11)? What are the needs of the population to be served by the stroke service and what are the current resources, people and facilities committed to the management of patients with TIAs and stroke? • How will we get to where we want to be (section 17.12)? What are the major gaps in the present provision of services (i.e. unmet needs and failure to provide effective interventions)? What resources, people and facilities will be required to meet the needs of the population? Assuming that resources are limited, to what extent

We have used the term ‘comprehensive stroke service’ to mean a stroke service that covers most of the needs of patients with stroke or transient ischaemic attack (TIA) and which is integrated in a way that provides a continuous patient journey – ‘a seamless service’ (Table 17.2 and Fig. 17.1). In doing so it should provide all of the functions outlined in Table 17.1. Figure 17.1 outlines what we believe to be important components of a comprehensive stroke service, and in this chapter we will discuss the evidence and rationale for some of those suggestions. A comprehensive stroke service should provide for most of the needs of most patients and do so within a ‘seamless service’.

17.4 General principles when discussing a comprehensive stroke service

When proposing various approaches to service delivery – and suggesting particular components of a comprehensive stroke service – we have tried to employ some general principles: • Basic needs first – stroke services should first ensure they provide basic care (from medical, nursing and therapy staff) for all stroke patients and their families. • Evidence-based options – priority should be given to those aspects of care that are generally accepted as being, or have been proven to be, effective. The Stroke Review Group of the Cochrane Collaboration is an excellent source of this type of information (http:// www.dcn.ed.ac.uk/csrg).

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 907

17.4 General principles when discussing a comprehensive stroke service Table 17.2 Key objectives and service solutions for managing patients with stroke and transient ischaemic attack. Objectives

Proposed service options

Cross-reference

Public awareness to recognize warning symptoms and seek help promptly Prompt and accurate assessment and diagnosis

Public education campaigns

section 17.5.1

Transport (ambulance and helicopter) protocols Emergency department and hospital protocols Designated stroke centres Telemedicine networks Rapid access TIA clinics Designated stroke centres Telemedicine networks Stroke units Designated stroke centres Telemedicine networks Stroke units Stroke units Designated stroke centres Rapid access TIA clinics Stroke units

sections 17.5.2, 17.5.3, 17.5.4 section 17.5.6, 17.5.7

Stroke units

section 17.6 section 10.4 section 17.7

Specific acute medical and surgical treatment Identification and assessment of patients’ problems Secondary prevention of further vascular events General care, including interventions to resolve problems (includes many aspects of rehabilitation) Terminal care for patients who are unlikely to survive Hospital discharge and reintegration into the community Continuing or long-term care for severely disabled patients Follow-up to detect and manage late-onset problems

Early supported discharge services Discharge planning Therapy-based rehabilitation services Stroke liaison worker services Chronic disease management Outpatient clinics

• Patient and carer views – surveys of the views of patients and carers have frequently highlighted a wish that care is coordinated and provided by expert staff. A survey of user views in Scotland (http://www. chss.org.uk/pdf/research/Improving_services_patients _and_carers_views.pdf) highlighted five main themes: – care should be provided in designated stroke units; – staff should be trained in the physical and emotional needs of stroke patients; – comprehensive information should be available to all patients and carers; – trained liaison staff should help with patients returning home; – there should be access to ‘someone to talk to’ (someone who understands the challenges that patients and their carers face). • Awareness of alternatives – where possible we have tried to acknowledge that there may be valid alternative approaches to achieving the same objective. However, frequently there is no research evidence to help make such decisions about the best service option.

..

..

sections 17.5.5, 17.5.6, 17.6 chapters 12, 13, 14 sections 17.5.5, 17.5.6, 17.6

sections 17.5.5, 17.5.6, 17.6 chapter 16 section 17.6 chapter 10

section 17.8 section 17.9

• Level of development – most of our discussion reflects experience from developed western economies. It is at present difficult to give specific advice for other healthcare settings although many of the general principles above will apply.

17.4.1 Why refer stroke and transient ischaemic attack patients to hospital? Before discussing each service component in turn we also need to consider why we have emphasized a system of hospital-based rather than community-based services (at least during the earlier phase of the illness). The vast majority of patients who have a stroke or TIA have it in the community and not in hospital. Community-based incidence studies have shown wide variations in hospital admission rates for stroke, varying from 55% in Oxfordshire, UK, in the 1980s to over 95% in Sweden and Germany in the 1990s.15–17 The main reason for admission to hospital in the past was for nursing care rather than diagnosis and treatment.15 However, the need for early imaging to establish whether a stroke

907

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 908

908

Chapter 17 The organization of stroke services

is ischaemic or haemorrhagic has now become accepted and immediate imaging appears to be the most costeffective option.18 The potential for early intervention to limit brain damage and prevent stroke recurrence19–21 is increasingly being recognized (Chapters 12, 16). Finally there are a small number of comparisons (in clinical trials) of a conventional policy of admitting most patients to hospital with a policy of avoiding admission by providing a rapid-response service in the community to support patients in their own homes. Although the evidence is limited,22,23 major practical problems were identified with the policy of hospital avoidance23,24 and clinical outcomes were best in those admitted directly to a stroke unit. Transient events, transient ischaemic attacks (TIAs) and episodes that may be confused with them (Table 17.3) Table 17.3 Some of the non-cerebrovascular problems referred to one of our neurovascular clinics by general practitioners over a period of 5 years. There were, of course, countless patients with transient symptoms in whom no definite diagnosis could be made. General medical problems Cardiac syncope (dysrhythmias, aortic stenosis) Vasovagal syncope Cough syncope Postural hypotension Hyperventilation Sleep apnoea Hypoglycaemia Neurological problems Migraine (both with and without headache) Epilepsy Transient global amnesia Glioma Meningioma Cerebral metastases Subdural haematoma Lymphocytic meningitis Peripheral neuropathy Guillain–Barré syndrome Cervical myelopathy Brachial neuritis Mononeuropathies Herpes zoster neuropathy Bell’s palsy Syringobulbia Myasthenia gravis Multiple sclerosis Motor neurone disease Creutzfelt–Jakob disease Psychiatric problems Somatization disorder Ophthalmic problems Retinal vein occlusion Glaucoma

are less reliably diagnosed than strokes and more often require input from a specialist, especially when the diagnosis is crucial to future decision-making – for example, in distinguishing between carotid and posterior circulation TIAs in a patient with severe carotid stenosis (section 4.3.1) (Table 4.3). Thus, referral to hospital for assessment, even if only as an outpatient, is the accepted norm for such patients.

17.4.2 Who may not need hospital admission? In view of the above discussions, can we identify a group of individuals who do not need hospital admission after stroke or TIA? In the absence of direct evidence from randomized trials comparing different policies of care, we can only base our decisions on a logical rationale. Hospital admission appears to be valuable for rapid diagnosis and assessment, prompt acute treatment and secondary prevention plus effective early multidisciplinary rehabilitation.25 We would therefore suggest that the following groups may be able to be managed without hospital admission: • For people with a TIA who are at low risk of early recurrence, early assessment at a one-stop neurovascular outpatient clinic61,62 could be justified providing we can reliably identify them (section 16.2). • People with a stroke who are at low risk of early recurrence or other complications such as neurological deterioration (section 16.2); once again this requires accurate identification of low-risk individuals. In practice, predicting neurological deterioration and other complications is difficult so this group is likely to include only those individuals who have delayed referral for several days and so are past the higher-risk period for developing problems. • People who refuse hospital admission; rapid assessment at a one-stop neurovascular clinic may be very appropriate for this group.

17.4.3 Why is rapid specialist assessment and treatment necessary? Rapid specialist assessment of patients experiencing symptoms of acute stroke or TIA can only really be justified if that assessment improves patient outcomes (including ‘softer’ outcomes such as reassurance and advice). The arguments in favour of rapid assessment have been made for both acute stroke and TIA patients: • Thrombolysis for acute stroke has been licensed for several years in US and has had a restricted licence in Europe since 2003. However, thrombolysis has not had the impact many of its supporters would have expected. First, it is a complex intervention requiring

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 909

17.4 General principles when discussing a comprehensive stroke service

intensive use of resources and personnel with a narrow therapeutic window. Current product licences restrict its use to within 3 h of symptom onset. This requires very rapid specialist assessment ensuring accurate diagnosis of ischaemic stroke prior to the initiation of treatment. In doing so, one needs to demonstrate the absence of haemorrhage on brain scanning, good blood pressure control, National Institute of Health stroke scale (NIHSS) score of greater than 4, and an exclusion of patients at high risk of bleeding with thrombolytic therapy. Treatment must be given within 3 h of symptom onset and the earlier that treatment is given the better the outcome (section 12.5). Initial enthusiasm in the US resulted in a large number of centres commencing acute treatment with recombinant tissue plasminogen activator (rt-PA). However, this demonstrated the very significant hazards of thrombolysis when implementation is poorly controlled. Haemorrhage rates as high as 16% were experienced and this high risk was closely correlated with violations of treatment protocols.26 The implementation of a stroke quality improvement programme was associated with a reduced rate of protocol deviation and symptomatic intracranial haemorrhage. As a result of these experiences, pressure groups such as the Brain Attack Coalition have proposed criteria for designation of stroke centres that are permitted to provide treatment with rt-PA.27 • Patients with intracerebral haemorrhage may benefit from early treatment with recombinant factor VII,28 although, at the time of writing, the treatment is not licensed for this use (section 13.3.1).

• Recent epidemiological studies suggest that similar issues may arise with the rapid diagnosis and treatment of patients with TIA or mild stroke.29,30 The risk of very early recurrence of cerebrovascular events appears to be much greater than was previously thought (approximately 10% in the first week and up to 20% in the first month; section 16.2.1) emphasizing the importance of very early diagnostic assessment and prevention. More recently, systems of identifying very high-risk individuals have been reported based on the presenting clinical features21,31 or MRI findings.32 The underlying assumption is that identifying individuals at high risk of recurrent stroke or TIA will lead to effective prevention of these recurrent events. Although this remains uncertain for drug treatments, there is good evidence that very early carotid endarterectomy (within 2 weeks) is not associated with an increased operative risk and results in improved stroke prevention (section 16.11.8).33,34 Furthermore there is no convincing argument for delaying the implementation of other secondary prevention treatments beyond the first few days post-stroke. Randomized trials of very early secondary prevention are currently under way.

17.4.4 Barriers to rapid assessment and treatment If we accept that rapid specialist assessment and treatment is worthwhile – at least for a proportion of stroke and TIA patients – we then need to consider how best to achieve this (Table 17.4). A recent systematic review of 54 observational studies examining the barriers to

Table 17.4 Potential barriers to rapid specialist assessment and treatment, and interventions aiming to overcome them.

recognition

react

response

reveal

reperfusion

..

..

Potential barriers

Proposed solutions which have been associated with some improvements in observational studies

Failure of the patient or family to recognize symptoms of stroke or to seek urgent help Failure of public and primary care medical services to recognize stroke as an emergency Failure of ambulance and emergency medical services to treat stroke as a medical emergency Delays in diagnostic assessment (particularly brain imaging)

Public education programmes (section 17.5.1)

Delays in delivery of acute drug therapy (where appropriate) including physician’s uncertainty

Education programmes for the public and primary care services (section 17.5.1) Training of paramedics and emergency medical services; rapid transport systems (section 17.5.2) Training of emergency Telemedicine department staff and networks (to redesign of hospital facilitate remote systems (section 17.5.3) specialist Centralization of expertise assessment and (designated stroke centre) management) (section 17.5.5) (section 17.5.6)

Centralized Brain Attack Centre initiatives (designated stroke centres) (section 17.5.5)

909

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 910

910

Chapter 17 The organization of stroke services

delivery of thrombolysis in acute stroke is also relevant when considering the barriers to more comprehensive stroke assessment and treatment, including transient ischaemic attacks.35 Barriers included: • failure of the patient or family to recognize symptoms of stroke or to seek urgent help; • initially seeking help from the general practitioner rather than calling an ambulance; • paramedical and emergency department staff triage stroke as non-urgent; • delays in obtaining brain imaging; • inefficient processes of in-hospital emergency stroke care; • difficulties in obtaining consent for treatment (thrombolysis); • physician uncertainty of administering treatment (thrombolysis). These potential barriers operate right across the initial patient pathway and a range of solutions is likely to be required if acute treatments are to be provided efficiently and equitably.

17.5 Interventions to improve access to early specialist assessment and treatment

When considering the necessary components to overcome the barriers to early assessment and treatment, researchers have developed a checklist (based on the ‘five Rs’) to describe the key stages: • recognition of symptoms; • reaction to the acute illness; • response of emergency services; • revealing the diagnosis; • reperfusion therapy. We can use this scheme to consider the potential solutions to these barriers. When you take this approach (Table 17.4), it is striking that there is often no really reliable evidence for particular components of the early patient journey, and that many of the interventions that have been tested have overlapping roles.

17.5.1 Public education programmes One of the most consistently reported prehospital barriers to acute stroke assessment and treatment is that patients or families have a poor knowledge of stroke. Poor awareness of stroke symptoms36 and a failure to recognize their seriousness37 often delays the request for

urgent medical help.38 Even with adequate knowledge the public has to be convinced that it is appropriate to call emergency services, not primary care services which so often leads to delays in management (see below). A number of suggestions have been made for public education programmes to increase the knowledge of stroke symptoms39,40 in the hope that they would reduce delays and so improve treatment rates with thrombolysis, and any other new emerging acute stroke treatments. However, experience in acute myocardial infarction has raised concerns about the effectiveness of media campaigns.41 Furthermore, even where people have appropriate knowledge this may not translate into action; patients in the Asymptomatic Carotid Atherosclerosis Study42 showed delays in reporting stroke symptoms despite receiving targeted education. Finally, the symptoms transmitted to the public in education campaigns are non-specific and identification and immediate action by the public could result in a large increase in the number of non-stroke patients reporting to emergency departments. Only one controlled clinical trial43 has evaluated a community education programme by comparing two communities. In the ‘intervention community’ the programme targeted ‘at risk’ members of the public, emergency medical services, emergency department physicians, neurologists and primary care providers, a higher priority for the transport of acute stroke patients by the emergency medical services. The control community did not receive these interventions. The proportion of eligible patients treated with rt-PA appeared to be greater in the intervention community and there was also a non-significant increase in the proportion of patients presenting within 2 h of symptom onset (Table 17.5). There were also reductions in within-hospital delays and in the reluctance of physicians to give rt-PA. The observed gains were relatively modest and we still need more robust evidence from properly randomized trials. Although public education programmes are likely to be important, they may have only a relatively modest impact on speedy referral rates.

17.5.2 Transport protocols (prehospital assessment and transport) The next stage in the chain of referral is the rapid response of ambulance and emergency medical services to treat stroke as a medical emergency. There are no randomized trials of this approach but observational studies have described the use of rapid recognition instruments44,45 to allow paramedical staff to triage patients with suspected stroke. These instruments appear to have acceptable diagnostic accuracy but the their impact on

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 911

17.5 Interventions to improve access to early specialist assessment and treatment Table 17.5 Interventions aiming to facilitate rapid access to specialist assessment and treatment in acute stroke. Intervention:

Education programmes for the public and primary care services

Training of paramedics and emergency medical services; rapid transport systems

Training of emergency department staff and redesign of hospital systems

Designated stroke centres (public education, rapid transport, emergency department training, hospital system redesign)

Telemedicine networks (networks between sites including local staff training)

Country:

USA

UK, USA

USA, Germany

USA, Germany

Main observations:

Increase in proportion of people presenting early with symptoms of suspected stroke 5%

Acceptable diagnostic accuracy of ambulance staff in identifying stroke patients

Reduction in hospital delays and door-toneedle times

USA, Canada, Germany Reduced delays both prehospital and within hospital

NA

9%

4–17%

Telemedicine approach feasible and acceptable Modest increases in consultation time (of 6–14 min) 2–20%

6%

21% (in selected helicopter transport patients) 45,183,184

11%

4–22%

2–25%

185

185,186

55,56

Absolute increase in rt-PA use*: Postintervention rt-PA use References:

43

Summary of the findings of a range of studies describing changes in the chain of referral and treatment of acute stroke patients. None of these were randomized controlled trials. *% delivery after intervention minus % delivery before. rt-PA recombinant tissue plasminogen activator; NA, no data available.

patient outcomes is unknown. Furthermore, the implementation and evaluation of ambulance protocols has often been included in an assessment of wider service changes (e.g. implementation of public education programmes and establishment of stroke centres). Helicopter transport appears to be a feasible transport option (Table 17.5) and there are claims that it can be cost-effective.46 However, it is notable that even in reports of helicopter transport of highly selected patients only a minority transported this way actually received rt-PA.

17.5.3 Emergency department protocols Many enthusiasts have advocated hospital acute ‘stroke teams’ to increase the proportion of patients who are eligible to receive thrombolytic therapy. Once again, there are no randomized trials of such strategies, although several observational studies have been described (Table 17.5). In general, these studies indicate that the protocols were associated with reduced time delays within the emergency department, particularly if backed up with educational programmes and rapid access to appropriate expert advice. However, it is also clear that

..

..

these are complex systems and frequently other changes were also made (e.g. moving CT scanning and nearpatient biochemistry testing to within the emergency department) which may have influenced the outcome.

17.5.4 Coordination of services One of the major challenges with any of the initiatives outlined above is how to coordinate them within the broader picture of delivering stroke care, in particular, the need to be able to access expert advice and diagnostic imaging within a very short time. In general, two approaches appear to have been applied to try and tackle this problem: • designated stroke centres, a centralized approach where patients are transported to a single centre (‘hub’ model); • telemedicine networks emphasizing a network of expert advice usually delivered from some form of specialized centre using teleconferencing technology (‘hub and spoke’ model). We shall consider in turn the rationale and evidence for these different approaches.

911

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 912

912

Chapter 17 The organization of stroke services

17.5.5 Designated stroke centres Following reports from the USA that without careful control of thrombolytic management, there is a high risk of protocol violations and treatment complications,26 stroke quality improvement programmes have been developed. These were associated with a fall in protocol deviations from 50% to 19% and in symptomatic haemorrhage from 16% to 6%. As a result of this and other experience, the Brain Attack Coalition (a group with representatives from major professional and advocacy organizations involved in stroke care in the US) suggested criteria for the designation of stroke centres:27,47 • acute stroke teams; • stroke units or teams; • written care protocols; • integrated emergency response systems; • support services (in particular, the availability and interpretation of CT scans and rapid laboratory testing). However, critics48 have commented that the establishment of stroke units was not given emphasis in the Brain Attack Coalition recommendations. Stroke units, for which there exists substantial randomized trial evidence and which are seen as central to the Helsingborg Declaration (with the target that organized stroke unit care should be available for all patients by the year 2005), were not considered an absolute requirement for primary stroke centres. By contrast, acute stroke teams, which lack evidence of effectiveness,49 were included among the key elements. In hospital surveys in the US,50,51 stroke units were established in only a minority of centres. Similarly, limited access to stroke units was also seen in Canada and Australia.52,53 Because stroke unit care is applicable to all patients, the overall impact on outcome is likely to be considerably greater than that of tissue plasminogen activator – there is much more to acute stroke care than thrombolysis. There are no randomized or controlled clinical trials evaluating designated stroke centres. However, four prospective observational studies (Table 17.5) have reported improvements associated with establishing a stroke centre. The establishment of such centres was complex but included: • forming a ‘brain attack’ team; • establishing a care pathway (based on the five ‘Rs’ recognition, reaction, response, reveal and reperfusion; Table 17.4); • establishing a protected bed in a stroke unit; • establishing public education campaigns; • raising the priority of stroke with emergency medical services; • ambulances bypassing other hospitals to get to a stroke centre.

No data are available on whether patient outcomes were improved by such changes but across these studies there was a reduction in symptom onset to treatment time and a modest increase in the delivery of rt-PA (Table 17.5). There do not appear to be any analysis of the impact of such changes on other aspects of stroke care such as access to rehabilitation services and number of patient transfers. In summary, designated stroke centres have gained ground in North America and Germany where the healthcare system appears to be conducive to such systems of care. In particular, stroke centres appear to have become more established where there are more fragmented care pathways (different providers delivering acute care and rehabilitation) with relatively short lengths of stay. There should be concern about the unnecessary transfer of unstable patients to stroke centres,54 since the majority will not receive the intended treatment (thrombolysis) and the stroke centre model (at least in the US) does not emphasize the best evidence for service delivery through stroke units.48 This, among other pressures, has raised interest in alternative ways of delivering expert assessment and diagnosis. The most prevalent of these appears to be telemedicine services. Because stroke unit care is applicable to all patients, the overall impact on outcome is likely to be considerably greater than that of tissue plasminogen activator – there is much more to acute stroke care than thrombolysis.

17.5.6 Telemedicine services (networks) Telemedicine has been broadly defined as ‘the use of telecommunication technology to provide medical information and services’.55,56 In relation to stroke it can be defined as ‘the process by which electronic visual and audio communications are used to provide diagnostic and consultation support to practitioners at distant sites, to assist in or directly deliver medical care to patients, and to enhance the skills and knowledge of distant medical care providers’. Telemedicine consultation has been incorporated into a variety of aspects of modern healthcare but only relatively recently into acute stroke services. Some of the stated advantages57 of such systems over traditional telephone consultations are: • assisting in neurological assessment; • identification of neurological deterioration and provision of immediate feedback on treatment; • identification of patients who do not require rt-PA; • to assess risk and benefit questions directly with the family without time delay; • direct supervision of specialist procedures such as delivery of rt-PA;

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 913

17.5 Interventions to improve access to early specialist assessment and treatment

• to monitor haemodynamic status and neurological examination prior to transport to a stroke centre; • ability of the stroke specialist to direct advice to the transport team on the therapeutic needs of the patient. Proponents of telemedicine for acute stroke have pointed out that over half of hospitals in the US are located in rural areas and there is currently a shortage of neurologists and radiologists to provide expert advice across the whole country.58 In addition, surveys of neurologists in the US have shown less than half felt comfortable about giving intravenous rt-PA.58 Telemedicine also fits well within the concept of establishing primary and comprehensive stroke centres that provide a network of coverage to related sites. Lastly, telemedicine may allow other sites within a network to participate in clinical trials; from consenting to follow-up, all the components for a quality stroke trial can potentially be achieved through telemedicine. Many of the reported barriers are essentially man-made (for example, reimbursement for services and the availability of suitable equipment). There are no completed randomized trials of telemedicine services in acute stroke to provide any reliable evidence about their impact on patient outcomes. However, several key questions have been addressed using other study designs.55,56 These indicate (Table 17.5) that telemedicine services in acute stroke appear to be: • Practical; experience from the US and Germany55,56 has demonstrated the feasibility of establishing telemedicine networks. These systems usually used some form of teleconferencing facility linked to a stroke centre. Technical failures were uncommon (0–4% of consultations). • Reliable; if telemedicine services are to be useful they must offer reliable diagnostic evaluation of both clinical features and brain imaging. The interrater reliability of assessing neurological status (using the NIH Stroke Scale, the Scandinavian Stroke Scale, or the European Stroke Scale) was as good as with face-to-face assessment.55 Two studies59,60 reported complete agreement between telemedicine assessment of CT scan and conventional neuroradiology from the point of view of eligibility for thrombolysis and major exclusions. • Feasible and acceptable; on average telemedicine consultations were about 10 min longer than conventional bedside consultations. Although this appears to be more than offset by reductions in the need for transfers to other hospitals, there were no explicit comparisons. Patient and clinician satisfaction with the telemedicine service was reported to be good.55,56 • Associated with improved delivery of rt-PA; the proportion of patients receiving rt-PA increased in association with telemedicine services.

..

..

Telemedicine (teleconsultation) services appear to offer a promising approach to improve access to acute assessment and treatment, particularly where the healthcare economy does not favour stroke centres, where populations are dispersed, and where local hospital stroke services can provide good post-acute stroke care (but not rapid assessment and treatment). A final point is that telemedicine services may allow workable medical on-call rotas to be developed by allowing a large group of experienced stroke specialists – who may be based in different hospitals – to cover a region.

17.5.7 Transient ischaemic attack services Most of the debate has focused on services for patients with stroke. However, substantial numbers of patients have a transient ischaemic attack (TIA) and have made a full clinical recovery before they seek help. The risk of stroke after a TIA, or of recurrent stroke after a first stroke, is very high in the early period and tails off later (section 16.2.1). Thus, one has most to gain from starting secondary prevention as early as possible. Also, since the accurate diagnosis of TIA depends on a good history, it makes sense to assess patients as soon after the event as possible. Traditionally, services have varied between countries. In the US it is relatively common for TIA patients to be seen in the emergency room. In Scandinavia and Germany they are often admitted to hospital. In the UK the move has been towards early specialist assessment in outpatient clinics, so-called ‘one-stop’ neurovascular, TIA or stroke clinics which are characterized by: • rapid access to a specialist opinion on TIAs, minor strokes and episodes that may mimic them; • streamlined access to the necessary investigations (which requires close liaison with the radiology department); • close links with surgeons who can offer timely carotid endarterectomy. These clinics ought to minimize unnecessary hospital admission and delay in accessing specialist opinion, investigation and treatment. However, they attract patients with a wide range of other neurological conditions (Table 17.3), so the clinician must have neurological training, or at least easy access to sound neurological advice.61,62

Rapid-access ‘one-stop’ neurovascular clinics should provide rapid clinical assessment of patients who may have had a transient ischaemic attack or minor stroke, with streamlined and cost-effective investigations and early intervention to reduce the risk of a serious vascular event.

913

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 914

914

Chapter 17 The organization of stroke services

there are some uncertainties about whether current treatments prevent early recurrence, most would argue that a very high early risk of recurrence (e.g. greater than 20% in the first week) would warrant hospital admission. Indeed, to deny admission to similar highrisk individuals with an acute coronary syndrome would be considered unacceptable. The next few years are likely to see a greatly increased move towards rapid assessment and treatment of TIA and minor stroke patients and the solutions adopted will vary depending on the region. However, it is crucial that we have good evidence about what forms of early assessment and treatment effectively reduce early recurrence.

However, the recognition that TIA patients may have a 20% risk of early stroke occurrence within the first 2 weeks, with the majority in the first day,30 has raised concerns about delays in assessment and diagnosis even where rapid-access clinics are available. In addition, the effectiveness of carotid endarterectomy is greatest if carried out within 2 weeks of symptom onset.34 These developments challenge us to diagnose and treat TIA and minor stroke far more quickly. In general, three approaches appear feasible (although none have been tested in clinical trials). • Outpatient management: with a daily stroke clinic, it is possible to see most TIA and minor stroke patients within 48 h of symptom onset.34 However, this is resource-intensive63 and presents challenges for the organization of investigations. • Assessment within emergency departments: if the emergency departments have rapid access to investigations, including ultrasound and CT scanning, then it may be possible to provide a rapid turnaround of investigation and early secondary prevention. This approach, which appears to be common in the US,64 carries the risk of inappropriate investigation and treatment if non-specialist staff are planning patient care. • Hospital admission: in some countries (e.g. Scandinavia) hospital admission for TIA is an established approach, however it is not common elsewhere (including the UK). Recently a scoring system has been developed which may identify individuals at highest risk of early recurrence.31,65 If a subgroup of TIA and minor stroke patients really are at extremely high risk of early recurrence, then a short period of hospital admission would be justified provided we can provide rapid assessment and effective treatment. Although

17.6 Organized inpatient (stroke unit) care

The types of stroke services that hospitals traditionally provided have varied from place to place reflecting local interests and politics. However, there is now widespread acceptance that hospital-based stroke services should be organized within stroke units.66,67 Much of the evidence for this comes from a systematic review of clinical trials that compared the outcome for stroke patients cared for in a specialist stroke unit with those cared for in general wards. Patients managed in stroke units are more likely to survive, return home and regain independence (Table 17.6). Stroke units may also improve patients’ quality of life, and improvements in outcome may persist for several years.68

Table 17.6 Summary of patient outcomes in the stroke unit trials.

Home (independent) Home (dependent) Institutional care Dead

Stroke unita

Conventional carea

Odds ratiob (95% CI)

Absolute risk differencec (95% CI)

Difference in number of outcomes per 1000 admittedd

45% 17% 14% 23%

40% 15% 16% 28%

1.25 (1.12, 1.40) 1.16 (0.88, 1.53) 0.84 (0.72, 0.98) 0.80 (0.70, 0.91)

0.05 (0.02, 0.07) 0.01 (–0.01, 0.03) –0.02 (0.00, 0.04) –0.04 (–0.02, –0.06)

50 10 –20 –40

a The proportion (%) of patients with various outcomes at the end of scheduled follow-up (median 1 year) in the randomized trials of stroke unit care vs conventional care. b The odds ratio for that outcome (95% confidence interval). c The proportion of outcomes achieved (+) or avoided (–) with stroke unit care. d The number of outcomes achieved (+) or avoided (–) for every 1000 patients cared for in a stroke unit, assuming the absolute risk of an outcome in the population is similar to that in the trials. Figures based on data from 31 trials (6900 patients).68

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 915

17.6 Organized inpatient (stroke unit) care

Stroke patients managed in a stroke unit are more likely to survive, return home and regain independence than those managed in a general medical or even in a neurology ward. These trials were testing much more heterogeneous interventions than is usual in drug trials, where the intervention is precisely defined in terms of the chemical, the dose and the timing. Because of this heterogeneity of the intervention – but not of the results – it is sometimes difficult to generalize from the stroke unit overview, and some important questions remain when applying the results to everyday clinical practice.

17.6.1 What is stroke unit care? Although comprehensive stroke unit care is a complex and multi-faceted intervention, the key components are reasonably well described69 (Table 17.7): Table 17.7 Outline of comprehensive stroke unit care.

..

..

• Ward base: effective stroke units have usually been based in a discrete ward with dedicated nursing staff. Mobile stroke teams do not appear to improve patient outcomes.70 • Specialist staffing: they have been staffed by medical, nursing and therapy staff with a specialist interest and expertise in stroke and/or rehabilitation. • Multidisciplinary team working: they have always included good multidisciplinary communication (defined as a formal meeting of all staff once per week to plan the management of individual patients). • Education and training: they have incorporated programmes of education and training for staff and provision of information for patients and carers. Although many stroke units were not described in detail, several consistent features of the process of care have been described66,69 (Table 17.7). This typically did not depend on high-tech facilities but did implement a systematic approach to care that incorporated:

Structure Geographically discrete ward Medical staff with specialist interest in stroke and rehabilitation Nursing staff with specialist interest in stroke and rehabilitation Multidisciplinary staffing (nursing, medical, physiotherapy, occupational therapy, speech therapy, social work) Coordination of care Regular multidisciplinary team meetings (formal meeting of all staff once weekly, informal meetings 2–3× per week) Close linking of nursing and multidisciplinary team care Educational programmes for staff Assessment and monitoring Rapid admission to stroke unit Medical history and examination Routine investigations (biochemistry, haematology, ECG, CT scanning) Further selective investigations (carotid ultrasound, echocardiogram, MR scanning) Nursing assessments (vital signs, general care needs, swallow test, fluid balance, pressure areas, neurological monitoring) Therapy assessments of impairments and disability Early management Careful management of food and fluids Pyrexia management, paracetamol for pyrexia, antibiotic for suspected infection Hypoxia management, oxygen if hypoxia, drowsiness or cardiorespiratory disease Glycaemic management, insulin for hyperglycaemia Early mobilization, up to sit, stand and walk as soon as possible Careful positioning and handling Pressure area care Avoid urinary catheterization if possible Ongoing multidisciplinary rehabilitation Early goal setting Early involvement of carers in rehabilitation Provision of information to patients and carers Discharge planning Early assessment of discharge needs Discharge plan involving patient and carers

915

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 916

916

Chapter 17 The organization of stroke services

• Careful assessment and monitoring – of medical, nursing and therapy needs. • Early active management – incorporating management of food and fluids, control of pyrexia, hypoxia, hyperglycaemia, early mobilization, careful positioning and handling, and avoidance of urinary catheterization. • Ongoing multidisciplinary rehabilitation – with early goal setting, early involvement of carers in rehabilitation, and provision of information to patients and carers. This also included early planning of discharge needs. Many of these processes of care will come as no surprise to those experienced in stroke care but recent audits support concerns that many are not routinely provided.71 Beyond describing the basic components of stroke unit care, it is difficult to determine whether the effectiveness of the stroke units is due to the total package of care, or to particular components. Some of the individual components can be evaluated in randomized trials (e.g. guidelines for prevention of deep venous thrombosis, early mobilization, intensive physiotherapy), and these trials can be reviewed systematically to provide reliable data, but some of the less well-defined components and any possible synergy between them is much less amenable to such assessment. For example, better communication between health professionals, stroke patients and their carers, which is so often inadequate

and a major source of dissatisfaction in non-specialized wards, may in part explain the success of stroke units, but this would be difficult, if not impossible, to prove.72,73

17.6.2 Types of stroke unit Although the basic principles of stroke unit care are reasonably well described they have been delivered in a variety of ways, and the term ‘stroke unit’ means different things to different people. So it is important to define our terms (Table 17.8). Inevitably, the question arises as to which type of unit is most effective. This is very difficult to answer with confidence and attempts to do so74 are prone to bias. They do indicate (Table 17.9) that effective stroke units have usually provided organized multidisciplinary care for a reasonable period of time. We shall discuss this in more detail. Effective models of organized inpatient (stroke unit) care are able to provide specialist multidisciplinary care for most of the inpatient pathway of care. Acute stroke units ‘Acute’ refers to the policy of rapid admission of the stroke patient to the stroke unit. Admitting all acute

Table 17.8 Classification of organized inpatient (stroke unit) care. Type

Philosophy of care

Patient group

MDT base

Timing of admission

Timing of discharge from a stroke unit

Type of care

Acute (intensive) stroke unit

Acute care; life support

Stroke

Ward

Acute (hours)

Early (3–7 days)

Acute medical & nursing care (with high staffing levels)

Acute (semiintensive) stroke unit

Acute care; close monitoring

Stroke

Ward

Acute (hours)

Early (3–7 days)

Acute medical & nursing care Monitoring & management of physiological variables

Comprehensive stroke unit

Acute care and multidisciplinary rehabilitation

Stroke

Ward

Acute (hours)

Later (days – weeks) some referral to specialist rehabilitation

Acute medical & nursing care Non-intensive management of physiological variables Early active multidisciplinary rehabilitation

Rehabilitation stroke unit

Multidisciplinary rehabilitation

Stroke

Ward

Delayed (days)

Later (weeks)

Multidisciplinary rehabilitation

Mixed rehabilitation unit

Multidisciplinary rehabilitation

Stroke & other disabling illness

Ward

Early (hours– days)

Later (weeks)

Multidisciplinary rehabilitation

Mobile stroke team

Acute care and/or multidisciplinary rehabilitation

Stroke

Mobile (no ward) base

Early (hours– days)

Later (weeks)

Acute medical care and/or multidisciplinary rehabilitation; no specialist nursing input

This table summarizes, in broad terms, the characteristics of different types of stroke unit. MDT multidisciplinary team.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 917

17.6 Organized inpatient (stroke unit) care Table 17.9 Effectiveness of different types of stroke unit: summary of direct and indirect comparisons of different models of stroke unit care. Type of stroke unit evaluated

Number of comparisons (subjects)

Control service comparator in trials

Estimated odds ratio for death or dependency (95% CI) compared with a general ward

Statistical significance

Acute (intensive) stroke unit Acute (semi-intensive) stroke unit

None 3 (530)

No data 0.77 (0.53–1.12)

No data P = 0.17

Comprehensive stroke unit

10 (3010)

No data Mixed rehabilitation unit, comprehensive stroke unit General ward, mobile stroke team General ward, mixed rehabilitation unit General ward, rehabilitation stroke unit General ward, comprehensive stroke unit

0.89 (0.79–0.99)

P = 0.03

0.85 (0.77–0.93)

P = 0.0007

0.87 (0.79–0.96)

P = 0.007

0.98 (0.91–1.10)

P = 0.5

Rehabilitation stroke unit

7 (870)

Mixed rehabilitation unit

9 (1170)

Mobile stroke team

5 (1000)

The results of a series of comparisons of different types of stroke unit care.74 This approach attempts to provide a standard measure of stroke unit effect where trials have used different control groups. Data analysis used methods185 of combining direct comparisons (trials that directly compare a stroke unit vs general ward) with indirect comparisons (e.g. the effect of a stroke unit vs general ward is inferred from trials of stroke unit vs mobile team and mobile team vs general ward). Data are presented as the summary estimate of the odds ratio (95% confidence interval) for death or dependency of stroke unit vs general ward. Please note that these estimates are based in part on indirect comparisons of treatments and are potentially subject to bias (e.g. from different patient groups being recruited into different trials).

stroke patients directly into a unit makes the introduction of assessment protocols easier, allows expertise to be focused, and will certainly facilitate the large randomized trials of acute interventions that are needed to identify effective treatments.75 It also facilitates a policy of active early mobilization (section 11.11), hydration (section 11.18), control of temperature (section 11.12), avoidance of hypoxia (section 11.2) and large changes in blood pressure (section 11.7). Although there are no reliable data from randomized trials demonstrating that any one of these interventions improves outcomes, they are supported by a reasonable theoretical rationale and some observational data.76–79 In some centres, particularly in North America and Germany, there has been a vogue for admitting stroke patients to ward areas with facilities for intensive monitoring of physiological functions (cardiac, respiratory and neurological). Interventions are introduced to correct these abnormalities (e.g. raised intracranial pressure, systemic hypertension), in the belief that this will improve outcome.80 Broadly speaking two approaches have been described: • Intensive care units which can offer all monitoring (including intracranial monitoring) and life support options (e.g. respiratory support). There have been several non-randomized studies of stroke intensive care units,81–84 but there is no good evidence that these

..

..

improve patient outcome. The approach has come under more scrutiny recently, presumably because it inevitably requires extra resources due to the high staffing levels and expensive equipment. Also, there is currently little evidence from randomized trials that the various individual interventions employed are effective. Thus, although stroke intensive care units may help, we need randomized trials to evaluate them. • Semi-intensive units are similar to coronary care units where monitoring and intervention focus on physiological variables but not life support. There have been three small clinical trials of semi-intensive units that have rather inconclusive findings. One85 indicated no benefit over care in a less intensive setting while two others86 have indicated a potential benefit. As mentioned above, there is limited evidence for the individual elements of monitoring and intervention although one trial79 has indicated that intensive monitoring can reduce early neurological deterioration after stroke. Comprehensive stroke units Perhaps the most successfully implemented model has been the comprehensive stroke unit, which admits patients acutely and then provides at least a few weeks

917

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 918

918

Chapter 17 The organization of stroke services Study or sub-category

Odds ratio 95% Cl

Comprehensive stroke ward vs general medical ward Akershus Athens Beijing Edinburgh Goteborg-Sahlgren Joinville Perth Trondheim Umea Subtotal (95% CI) Test for overall effect: Z = 2.39 (P = 0.02)

0.83 (0.71– 0.97)

Rehabilitation stroke ward vs general medical ward Dover (GMVV) Nottingham (GMVV) Orpington-1993 (GMVV) Orpington-1995 Subtotal (95% Cl) Test for overall effect: Z = 0.92 (P = 0.36)

0.83 (0.57–1.23)

Mobile stroke team vs general medical ward Cape Town Manchester Montreal Uppsala Subtotal (95% Cl) Test for overall effect: Z = 0.22 (P = 0.82)

0.96 (0.69 –1.34)

Mixed rehabilitation ward vs general medical ward Birmingham Helsinki IIIinois Kuopio New York Newcastle Subtotal (95% Cl) Test for overall effect: Z = 2.57 (P = 0.01)

0.65 (0.47– 0.90)

Total (95% Cl) Total events: 1266 (Treatment), 1341 (Control) Test for overall effect: Z = 3.22 (P = 0.001) 0.1

0.2 0.5 Favours stroke unit

Odds ratio 95% Cl

0.82 (0.73 – 0.92)

1

2 5 10 Favours general ward

of rehabilitation. Such a model, which is widespread in Norway and Sweden, is supported by the largest group of clinical trials included in the systematic review (Fig. 17.2) and results from a national stroke register in Sweden.87 Although we believe that rehabilitation should start on the day of the stroke, such units do present some practical challenges. For example, very sick stroke patients might require care that would disrupt a rehabilitation unit, at a time when they are unlikely to benefit from a rehabilitation environment. However, models that separate acute assessment and rehabilitation areas may disorientate some patients (and their families) and can compromise continuity of care. The comprehensive unit approach appears to be well suited to

Fig. 17.2 The results of a systematic review of randomized trials testing the effectiveness of stroke unit care compared with a general medical ward. The result of each individual trial, expressed as the odds ratio, is represented by a purple box with a horizontal line indicating the 95% confidence interval. The block size is proportional to the amount of information in the trial. An odds ratio to the left of the vertical line indicates that the odds of the outcome (in this case events refer to death or dependency between 6 and 12 months after randomization) is less with stroke unit care than care in general wards. Estimates based on an overview of all the trials are represented by solid diamonds. Reproduced with permission from the Stroke Unit Trialists’ Collaboration.88

medium-sized hospitals where one team can manage most stroke patients within one unit. In practice, although these units provide most care for most patients, referral of some patients with ongoing complex rehabilitation needs to other rehabilitation services is common. Rehabilitation stroke units Several trials have indicated benefit from rehabilitation units that admit patients a few days after stroke onset and continue rehabilitation for several weeks. These trials have inevitably examined a more selected patient group who are stable enough for that environment and have ongoing rehabilitation needs.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 919

17.6 Organized inpatient (stroke unit) care

Mixed rehabilitation units The meta-analysis of organized inpatient (stroke unit) care predominantly included trials comparing organized care within stroke-specific units with care in general wards (general medicine or general neurology). However, some trials explored the impact of organizing stroke care within generic rehabilitation services (e.g. geriatric medicine or neurological rehabilitation services); patients achieve better outcomes in mixed rehabilitation units than in general wards without multidisciplinary team care.88 Comparisons with stroke-specific units indicate a trend towards better outcomes in stroke-specific units (Table 17.9) but there are insufficient data to determine whether stroke-specific units are clearly more effective. Mixed rehabilitation units may have a role in smaller hospitals or very specialized services (e.g. young adult rehabilitation) where there may be too few stroke patients to make stroke-specific services viable. However, stroke-specific services certainly allow more specialization among the team members, which enhances the educational and research potential of the service. Training of junior doctors and other staff, which might suffer if all the stroke patients were managed by a single team, can be protected and probably improved by organizing rotation of staff through the unit. Moreover, stroke specialists are more likely to be enthusiastic about teaching students and staff about strokes than generalists or those with another specialist interest. Mobile stroke teams Overall the trials in the meta-analysis indicated that a stroke team working across several general wards may improve aspects of the processes of care (e.g. access to specialist assessments) but cannot achieve patient outcomes as good as those of a team based in a stroke unit.24,49,70 Probably the most important advantage of having the patients in one location is that the nursing staff can play a greater role in the rehabilitation process. When patients are scattered, it is more difficult to incorporate the essential role of nurses. Also, stroke patients managed in acute general areas have to compete for nursing time with patients who may be perceived to have more urgent needs (e.g. chest pain). Stroke patients may, for example, need regular toileting to maintain continence and thus dignity. These aspects of care are very important, but can be seen as less urgent. A geographically defined stroke unit removes this competition for nursing time, and allows the nurses to take on a new role – not just as carers, but also as facilitators of patient independence – and to continue therapy (directed by specialist therapists) throughout the 24-h period. The

..

..

main role for stroke teams is probably to provide outreach care for those unable to get into the stroke unit, and to prioritize admission to the stroke unit. The main advantage of caring for stroke patients in one place is that nurses can play a major role in the rehabilitation process.

17.6.3 Who should be admitted to the stroke unit? Stratification of patients within the meta-analysis by stroke severity showed that patients with mild, moderate and severe strokes are all likely to benefit from stroke unit care.88,89 In terms of absolute outcomes more severe patients have the greatest survival advantage from stroke unit care but milder patients gained more in terms of regaining independence. Data from non-randomized, and thus potentially biased, comparisons of outcomes following admission to a stroke unit or general care support the notion that stroke unit care benefits unselected stroke patients.87,90–92 Of course, bed shortages may force staff to make triage decisions. Many of the serious early complications of stroke (including early neurological deterioration) are more common with severe stroke and may respond to stroke unit care. Milder stroke patients (e.g. those who are mobile) probably have less to lose from not being admitted to a stroke unit. There is little evidence that patients of particular ages gain more or less from care in a stroke unit.88,93 There are good reasons to believe that older stroke patients may be at higher risk of some complications, and therefore have more to gain from admission to a stroke unit. Although we think that needs rather than age should dictate where and by whom patients are managed, local conditions will often dictate which service is the best option. For example, where an age-related geriatric service (e.g. one that admits any patient who is over 75 years old, whatever the problem) already provides effective stroke rehabilitation, there may be a case for adding a new rehabilitation service for younger stroke patients rather than dismantling the current service. Professionals and patients’ families are often concerned that younger patients’ morale will suffer if they are treated in a ward with mainly older patients, although this concern is by no means universal.

17.6.4 How long should patients remain in the unit? Some units, particularly those that admit patients acutely, define maximum lengths of stay. It seems to us that the only reason to do this is to allow admission of new cases.

919

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 920

920

Chapter 17 The organization of stroke services

If the unit is of sufficient size for the population’s needs, works flexibly and is efficient in discharging patients, then a defined maximum length of stay should not be needed. If one does insist on a maximum length of stay, one must ensure that there are facilities and staff elsewhere to deliver appropriate continuing care (e.g. ‘slower-stream’ rehabilitation facilities;93 and that patients are not left to languish on an acute medical ward. One might argue that patients who are no longer improving, but are having to wait for placement in the community or an institution, should not be kept in a stroke unit. However, for some individuals the unit may offer the best environment to maintain any functional improvement already gained. Moves under these circumstances should only be considered where beds are limited and patients who are judged likely to gain more from the unit environment are waiting to be admitted.

17.6.5 How large should a stroke unit be? Age-specific and sex-specific stroke incidence data and details of the hospital catchment population, along with hospital activity data, should allow an estimate of the number of patients who are likely to require admission to hospital each year (section 17.11.3). Unfortunately, there may be variations due simply to chance or the season of the year. Although consistent seasonal differences in the incidence of stroke have not been demonstrated in community-based studies, at least in temperate regions, there is an excess of hospital admissions and stroke deaths during the winter.94,95 Of course, this may simply reflect referral bias and a higher case fatality during cold weather. Whatever the explanation for any seasonal variation, it does cause difficulties when planning stroke services. Prior to the development of a stroke unit, a survey in one of our own medical units, which admitted between 200 and 250 stroke patients each year, demonstrated that the number of stroke inpatients on any one day varied between 9 and 35 over a year. Therefore, whatever organization one sets up to manage these patients, it must be flexible enough to cope with large fluctuations in their numbers. The unit should be able to accommodate different proportions of men and women, as the proportions are bound to fluctuate. The inevitable limit on the number of beds may be managed by ensuring that the stroke unit is part of a larger area, into which it can expand with demand and then contract again. Such arrangements also mean that, at times, non-stroke patients are cared for in the stroke unit. Inevitably, there are times when resources are not adequate to meet all the needs of the patients, and it is then important for the team to support its members in the difficult task of prioritizing – in other words rationing – care.

A survey in one of our own medical units, which admits between 200 and 250 stroke patients each year, demonstrated that the number of stroke inpatients on any one day varied between 9 and 35 over a year. Therefore, stroke services must be flexible enough to cope with large fluctuations in the numbers and types of patients referred.

17.6.6 Who should staff and run the stroke unit? The units included in the systematic review were run by geriatricians, neurologists, general (internal) physicians and rehabilitationists. Indirect comparisons of the benefits of units run by different specialist groups did not show any significant differences.66,88 We believe that whoever is responsible should have the necessary knowledge, training and above all enthusiasm to take on the task. The most appropriate professional group will vary from place to place. For example, in the Netherlands, practically all acute stroke patients are managed by neurologists, while in the UK most are managed, at least initially, by general physicians and geriatricians.71,96 British neurologists may have the knowledge and training to diagnose and investigate stroke patients, but unfortunately most do not have the time or interest, support staff, access to beds, or training in rehabilitation to run a comprehensive stroke service without help from other specialists. In the UK, geriatricians are often in the best position to take a leading role, although most need extra training in neurology and the active participation of their local neurologist – who can very usefully contribute to diagnosis and management of patients, especially those with unusual causes of stroke, patients with ‘funny turns’ and the many and varied neurological problems that arise in inpatient and outpatient stroke care (Table 17.3). The advent of thrombolysis requires additional training, expertise and support in assessment of the ‘brain attack’ patient and interpreting acute imaging. In the UK training for stroke specialists is being implemented and an increasing number of hospitals are appointing specialist stroke physicians (from a variety of specialist backgrounds) to coordinate stroke services.

17.6.7 Planning a stroke unit development The good evidence for the effectiveness of stroke units means that in many countries their development is supported by national initiatives. However, the development of such units is often resisted by those who perceive them as a threat. Some points are worth considering in local discussions.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 921

17.7 Transfer from hospital to community

• Stroke units generally make more efficient use of existing staff and beds, and so may eventually even increase the resources available to other specialties. • De-skilling of their junior medical, nursing and paramedical staff can be overcome by rotating staff and students through the unit. • Adopting an evolutionary approach to developing the service may be encounter less resistance – for example, one might introduce a stroke assessment protocol before trying to set up an acute assessment area, or a stroke team working on the general medical wards before trying to set up a geographically defined stroke unit. Stroke units are not just a research intervention, they have been implemented in a range of settings with improvements in outcome.71,87,91,92 • One can try to influence the local organizations that fund healthcare (i.e. health authorities and general or family practitioners in the UK; health insurers in other countries) to exert pressure for change, since they are generally keen to fund services for which there is scientific evidence of efficacy (curiously, and irritatingly for those of us concerned with stroke, neither coronary care units nor regional oncology services have been nearly as well evaluated as stroke units, and yet their utility is said to be ‘obvious’ and they are widely encouraged). Curiously, and irritatingly for those of us concerned with stroke, neither coronary care units nor regional oncology services have been nearly as well evaluated as stroke units, and yet their utility is said to be ‘obvious’ and they are widely encouraged. There are likely to be significant pressures (other than evidence of effectiveness) that will influence local service delivery, since its structure must be tailored to local needs, resources, geography, people and politics. One needs to consider what is the best way to deliver comprehensive stroke care within the constraints of the local circumstances. We suggest three factors come into play: • evidence that a particular model of stroke unit is effective (Table 17.9); • the ability of the stroke unit model to deliver all aspects of care required; • the ability of the stroke unit to provide for the broadest group of stroke patients (i.e. meet the needs of the stroke population). The following list of options outline some common solutions: • Comprehensive stroke unit: where possible, the comprehensive stroke unit, which combines both acute care and rehabilitation for the majority of patients, is probably the preferred option. This could include the delivery of thrombolysis or take over the care

..

..

of patients following thrombolysis in an emergency department setting. This option has a good evidence base to support it and would be most appropriate for small to medium-sized hospitals (e.g. 100–300 stroke patients per year). • Acute stroke unit and rehabilitation stroke unit: in larger hospitals, it may be difficult to provide all stroke unit beds in one place. In this circumstance, a combination of acute stroke unit and rehabilitation stroke unit has often been adopted. In theory this should provide comprehensive stroke unit care but is largely untested in clinical trials. If this model is adopted, we suggest that the components of comprehensive stroke unit care must be provided throughout the patient journey (e.g. early mobilization and early multidisciplinary planning applies in the acute stroke unit, patients in the rehabilitation stroke unit should have access to acute medical care). In addition, the patient and family should experience a continuous process of care rather than discontinuity between components of the service. • Mixed rehabilitation unit: in smaller hospitals with dispersed populations (e.g. less than 100 stroke admissions per year), it may be appropriate to have protocols of delivery of acute care in general wards and to provide rehabilitation services in a mixed rehabilitation setting. Telemedicine services (section 17.5.6) may offer options for accessing expert advice from larger stroke centres. In a stroke unit development it is important to compare the characteristics of your planned service with those in the stroke unit trials to ensure you are delivering care that is, as far as possible, evidence-based (Tables 17.8 and 17.9).

17.7 Transfer from hospital to community

One of the main areas of concern to patients, and more particularly to carers, is the organization (or rather the lack of organization) of hospital discharge.72,97 One can well understand their concern. One day patients are being cared for in hospital by a team of professionals, and the next they are at home and the responsibility of the carers. A number of things can be attempted to try and reduce the stress of the transition from hospital to home: • Provide adequate information (section 17.10.1) and train the carers while the patient is in hospital, for example invite the carers to therapy sessions and involve them in the patient’s care on the unit. Unfortunately, trials of information provision and patient education do not provide clear evidence to guide practice.

921

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 922

922

Chapter 17 The organization of stroke services

• Predischarge home visits with the patient and one or more members of the team to ensure that the home environment is tailored to the patient’s needs. Also, informal visits home, initially for a day and then graduating to overnight and weekend stays, allow the patient and carer to gain confidence, identify potential problems and help maintain morale. Although this is a well-established approach we could not identify any clinical trials of such policies. • Predischarge case conferences allow the patient and carer to meet with the hospital-based team and any professionals who are to be involved in their care in the community. Once again this is a well-established approach for which we could not identify any clinical trials. • Clear guidelines about who to contact in the event of problems. General (family) practitioners, or one of their team, are the ideal points of contact, but they can only fulfill this role if adequately briefed before the patient’s discharge. It follows that detailed records of the patient’s problems, and the plans for support in the community, must be relayed to those expected to monitor the home situation. • A programme of training carers to manage their new role has been tested in one moderately large randomized trial.98 This involved training by stroke unit staff about stroke plus practical caring skills. The brief programme also continued after discharge. This approach was surprisingly effective – trained carers reported less caregiver burden, anxiety or depression and had a higher quality of life, while patients in the carer training group experienced less anxiety and depression and a better quality of life. The intervention was also Hospital admission

cost-effective, reducing the overall cost of caring99 and is the subject of further research. In some circumstances, it may be appropriate to further break down the boundary between hospital and home by organizing for the patient to attend a day hospital or outpatient department regularly to be reviewed medically, or to receive further rehabilitation input.

17.7.1 Early supported discharge services Partly in response to the well-reported limitations of conventional hospital discharge arrangements, a number of services have been developed to try and improve the transition between hospital and community, in particular, early supported discharge services, which aim to accelerate discharge home from hospital but provide more continuity of rehabilitation in the home setting. To date, 12 randomized trials have tested this approach to care in a variety of settings around the world.70,100 Most were based around a small multidisciplinary team of physiotherapy, occupational therapy, nursing and assistant staff – with input from medical, speech and language therapy and social work staff. The teams were either hospital-based (and went out to the patient’s home) or community-based (and came into hospital to recruit patients). All incorporated regular multidisciplinary team meetings to plan patient care. A typical pathway of care is shown in Table 17.10. Typically these services input for up to 3 months but in some cases this might be shorter with handover to other community services.70 Even compared with high-quality care from a hospitalbased stroke unit, an early supported discharge team

Early identification of eligible patients in hospital Early assessment by a ‘key worker’ from the early supported discharge team (one individual who supervised that patient’s care) Assessment of home needs through a home visit (with or without the patient present) Identification of recovery goals with patient and carer Plan discharge home Discharge home Discharge home with very early input (within 24 h) by members of the early supported discharge team Implement rehabilitation plan Continuing rehabilitation in the home setting (up to 4–5 days per week if necessary) Access relevant services Multidisciplinary team review of progress Negotiated withdrawal of the team as recovery goals are achieved Planned discharge from the service with later follow-up and review Discharge from early supported discharge team

Table 17.10 Illustrative pathway of care with early supported discharge service.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 923

17.8 Continuing rehabilitation and reintegration back to normal life Study or sub-category

Fig. 17.3 The results of a systematic review of randomized trials testing the effectiveness of early supported discharge (ESD) services compared with conventional discharge services. The result of each individual trial, expressed as the odds ratio, is represented by a purple box with a horizontal line indicating the 95% confidence interval. The box size is proportional to the amount of information in the trial. An odds ratio to the left of the vertical line indicates that the odds of the outcome (in this case events refer to death or dependency between 6 and 12 months after randomization) is less with the ESD service than with conventional care. The estimates based on an overview of all the trials are represented by diamonds. Reproduced with permission from the Early Supported Discharge Trialists.100

ESD team coordination and delivery Adelaide Belfast London Manchester Montreal Newcastle Stockholm Subtotal (95% Cl) Test for overall effect: Z = 2.40 (P = 0.02) ESD team coordination Oslo Trondheim Subtotal (95% Cl) Test for overall effect: Z = 1.89 (P = 0.06) No ESD team Akershus Bangkok Subtotal (95% Cl) Test for overall effect: Z = 0.92 (P = 0.36) Total (95% Cl) Total events: 359 (Treatment), 398 (Control) Test for overall effect: Z = 2.28 (P = 0.02)

could not only accelerate discharge home (with an average reduction in length of stay of 7 days), but could also result in the patient having a greater chance of remaining at home and regaining independence (Fig. 17.3). Overall, for every 100 patients randomized to early supported discharge services, an extra five (95% CI 1–9) remained at home and an extra five (95% CI 1–10) were independent at 6–12 months after the stroke. Good results were most likely with a well-resourced, coordinated multidisciplinary supported discharge team and if patients were recruited with mild to moderate stroke severity (i.e. those who achieve a Barthel Index of at least 10/20 in the first week – equivalent to rapidly regaining standing balance). There is a suggestion that such services may not work as well in more dispersed rural populations101 but this requires confirmation. Economic analyses100 indicate that the potential saving in hospital costs (reduced hospital bed-days) is greater than the additional costs of community rehabilitation. In practice it is often difficult to release such costs, but at the very least early supported discharge services appear to offer a way of both improving patient care and optimizing the use of a limited number of hospital beds. In addition to the ‘harder’ outcomes above, it is also noteworthy that patient and carers allocated to early supported discharge services were more likely to report satisfaction with their services.70 Early supported

..

..

Odds ratio 95% Cl

Odds ratio 95% Cl

0.71 (0.53–0.94)

0.68 (0.46–1.01)

1.23 (0.79–1.91) 0.79 (0.64–0.97)

0.1 0.2 0.5 1 2 5 10 Favours ESD service Favours control

discharge services appear to be an essential component of a truly comprehensive stroke service and should particularly target patients with mild to moderately severe strokes. Early supported discharge services – based on a specialist multidisciplinary team – can help accelerate discharge home and improve the longer-term recovery of selected stroke patients without incurring excessive costs.

17.8 Continuing rehabilitation and reintegration back to normal life

Even where stroke patients have received good care in hospital and around the discharge period, they may still have difficulty maintaining independence and reintegration back to normal life (Table 17.11). At this stage of the patient journey, services are often very variable and may be completely non-existent. This probably reflects the diversity of approaches in different countries but also a limited evidence base to suggest that effective interventions really can improve recovery. In general terms two broad approaches have been tested at this late stage:

923

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 924

924

Chapter 17 The organization of stroke services Table 17.11 Common problems that arise late after a stroke, often when the patient is no longer in hospital. Patient Deteriorating function due to: – inactivity – progressing comorbidities – lack of continuing rehabilitation input? – depression or anxiety, even agoraphobia (section 11.31) – overprotection from the carer Social isolation Financial difficulties (section 11.33.6) Sexual dysfunction (section 11.33.5) Undetected rise in blood pressure Central post-stroke pain (section 11.23) Carer (section 11.34) Physical ill-health due to the strain of caring Depression or anxiety Poor relationship with patient because of personality change Social isolation because unable to get out to meet people Financial difficulties

• Therapy-based rehabilitation services (provided by physiotherapy, occupational therapy or multidisciplinary staff and primarily aiming to increase activities in daily living). In practice this might include a range of taskrelated interventions aiming to improve mobility, activities of daily living or specific tasks such as dressing. Study or sub-category

Odds ratio 95% Cl

Mixed service Hong Kong London Philadelphia South London Subtotal (95% Cl) Test for overall effect: Z = 1.12 (P = 0.26) Physiotherapy Copenhagen Kansas Subtotal (95% Cl) Test for overall effect: Z = 0.77 (P = 0.44) Occupational therapy Cardiff Glasgow Nottingham 1995 Nottingham 1997 Nottingham 1999 TOTAl Subtotal (95% Cl) Test for overall effect: Z = 2.23 (P = 0.03) Total (95% Cl) Total events: 245 (Treatment), 218 (Control) Test for overall effect: Z = 2.61 (P = 0.009)

• Stroke liaison worker services (provided by stroke nurses, family support workers or specialist social workers and primarily aiming to improve participation in normal living and quality of life). In practice these have included a mix of interventions which could deliver a programme of rehabilitation (e.g.102) or respond to identified problems.103 There is considerable variation within these broad approaches but they do provide a framework for considering the evidence.

17.8.1 Therapy-based rehabilitation services In a systematic review104 of therapy-based rehabilitation services, most of the trials assessed input in the patient’s home by an occupational therapist although the review also included physiotherapy or multidisciplinary team services (such as community teams and day hospital services). Therapy-based rehabilitation services (when compared with no routine intervention) helped prevent stroke patients deteriorating in their ability to carry out activities of daily living (ADL) (Fig. 17.4) and improved ADL scores. Therefore, even relatively late after stroke onset (several months) patients may gain from input from a therapist. What is less clear is the absolute benefit likely to be achieved and the cost-effectiveness of these services. However they do indicate that within a Odds ratio 95% Cl

0.73 (0.42–1.27)

0.66 (0.23 –1.90)

0.73 (0.55 – 0.96) 0.72 (0.57– 0.92)

0.1 0.2 0.5 1 2 5 10 Favours therapy Favours control

Fig 17.4 The results of a systematic review of randomized trials testing the effectiveness of therapy-based rehabilitation (provided to patients living at home) compared with conventional services (usually no routine input). The result of each individual trial, expressed as the odds ratio, is represented by a purple box with a horizontal line indicating the 95% confidence interval. The block size is proportional to the amount of information in the trial. An odds ratio to the left of the vertical line indicates that the odds of the outcome (in this case events refer to poor outcomes – death, deterioration in activities of daily living or dependency between 6 and 12 months after randomization) is less with the therapy service than with conventional care. Estimate based on an overview of all the trials are represented by diamonds at the bottom of the figure. Reproduced with permission from the Outpatient Therapy Trialists.104

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 925

17.9 Longer-term follow-up and chronic disease management

comprehensive stroke service, some system of review and further intervention by therapists are valuable. A comprehensive stroke service should include systems that can review the stroke patient’s progress and provide further intervention by therapists if required. The discussion above does not indicate how such therapy services should be delivered (e.g. in the patient’s home, outpatient clinic, day hospital). The effectiveness of domiciliary physiotherapy and of physiotherapy provided in a day hospital have been compared in randomized trials.105–109 Together they demonstrate only small differences in outcome.110 The relative costs of providing care in these settings varied between the studies, with no clear conclusion.

17.8.2 Stroke liaison worker services The generic title of ‘stroke liaison worker’ can be defined as ‘someone who provides emotional and social support and information to stroke patients and their families and liaises with services with the aim of reducing aspects of handicap and improving quality of life for patients with stroke and/or their carers’.111 These services (which have also been termed ‘specialist nurse support’, ‘stroke family care worker’ and ‘stroke family support organizer’) often involve approaching patients and families during hospital admission, when they can provide information and education about stroke. They are also available for input after discharge home, particularly to identify problems or unmet needs and to develop customized solutions.112 At least 15 randomized trials have tested out this type of service in the UK, Australia, the US and the Netherlands. Their impact on patient outcomes is unclear102,111,113,114 but they do appear to be valued by patients and carers. In summary, in a comprehensive stroke service there should be systems to allow review of stroke patients’ progress and to provide therapy input if required. There may be a number of ways to deliver such services (domiciliary, outpatient clinic, day hospital).

17.9 Longer-term follow-up and chronic disease management

disease management. The rationale is to provide a system for managing the various needs, in particular prevention of disease and management of disability. The best setting in which to provide such chronic disease management is likely to depend on the local healthcare system. In countries such as the UK and the Netherlands with a well-developed primary care service, it may be appropriate for much of this continuing care to be provided by the patient’s own general family practitioner. In countries without developed primary care services this may need to be done through a hospital outpatient clinic. Whatever system is in place, it is important that it provides not only secondary prevention but an opportunity for reassessment, and identification and treatment of problems that may become apparent only after several weeks or months (Table 17.11). A follow-up checklist should ensure that late problems are not overlooked (Table 17.12). Some patients make little progress in the first few months after the stroke (perhaps due to intercurrent illness) and are discharged to a supported environment, but then unexpectedly begin to improve. Ideally, such patients should be identified and re-enter a rehabilitation programme, but few services are sufficiently well organized or adequately resourced to offer this. Secondary prevention, which will be lifelong, can be provided in a hospital-based clinic, but is managed more conveniently (for the patient) in the primary healthcare sector.

Table 17.12 A stroke follow-up checklist. Impairments Disabilities Aids and adaptations

Support services New problems Aspirations Work Driving Carer Prevention

Stroke has now been recognized as one of the diseases which requires ongoing management – so-called chronic

..

..

Ask about weakness, balance, speech, pain Do you need help with any everyday activities? Do you need any aids? Have they been delivered yet? Are you using them? Are they in good working order? Are they in place? Are they appropriate and adequate? Any new problems since last seen? Anything you want to do but cannot do? Are you back to work? Are you driving a car? Do you have a carer? How is your carer coping? Check blood pressure, cholesterol, diabetic control, smoking, diet, exercise, adherence with medication

925

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 926

926

Chapter 17 The organization of stroke services

17.10 Generic issues in stroke service delivery

17.10.1 Information for patients and carers One generic theme in the management of stroke – which is relevant throughout the patient pathway – is communication of information. Studies of the attitudes of patients and their carers to medical services in general, and stroke services in particular, have demonstrated that one of the greatest sources of dissatisfaction is with communication.72,97 Patients and carers may receive very little information about the nature of stroke, its cause, management and likely prognosis.25,115 Even where information is provided, it may be in a form that is difficult to understand or retain. Patients’ and carers’ perceptions of the stroke service are likely to depend not just on the degree of recovery, but also on the quality of communication. Although it is easy to show that many patients receive little information, one must also remember that for some it may be enough. Some patients do not want a lot of information, preferring to trust in the professionals’ judgement.115,116

It is therefore important to tailor the provision of information to the individual’s needs and wishes. These complexities may in part explain the mixed results from clinical trials of information and education interventions.109 There is some evidence that information combined with educational sessions improves knowledge and is more effective than providing information only. Information provision alone seems to have no measurable effect on mood, perceived health status or quality of life for patients or carers. The impact on satisfaction is inconsistent. In the absence of good evidence to guide us we aim to provide information using a number of different media, including: • a notice board on the stroke unit (Fig. 17.5); • an information pack containing appropriate leaflets; • audio and video tapes; • individual interviews with patients and carers by members of the team; • patient and carer groups. However, there is probably no substitute for one of the team sitting down with the patient and family on one or more occasions to explain the situation and answer any specific questions. This can then be backed up with

Fig. 17.5 A notice board at the entrance to a stroke unit. Typically it introduces the members of the stroke team, indicating how they might be contacted. It presents, in simple terms, what a stroke is and how it can affect the patient. We also include useful pamphlets and information relating to patient and carer groups.

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 927

17.11 Planning, developing and maintaining a stroke service Table 17.13 Tips on improving communication with carers of hospitalized stroke patients. Provide a notice board at the entrance to the unit introducing staff and informing relatives about how to contact members of the team Hold ward rounds during visiting times Hold a regular ‘open access’ clinic where carers can meet the consultant Invite carers to participate in patient care (including therapy sessions) Set up a carers’ group Document the content of any discussions with relatives in the notes and report these at team meetings to ensure consistency Arrange predischarge case conferences or family meetings Back up verbal communication with written or audio material

written material. One approach, which has been used in other areas (e.g. oncology), is to record the interview and give the recording to the patient or family so they can review what has been said as and when they wish.117 This might overcome the problem of patients and families only taking in a small proportion of the information given to them. Like all other areas of stroke care, a service needs to establish a system ensuring that input – in this case the provision of information – is tailored to the individual needs of patients and carers (Table 17.13).

occasionally be life-saving and produce a good functional outcome (sections 12.7.3, 13.3.5, 13.3.6). Neuroradiological interventions are now a key component of subarachnoid haemorrhage management (section 14.6). Inevitably, many smaller hospitals do not have these facilities. However, for the 1–2% of stroke patients who are at risk of deteriorating rapidly with a surgically remediable complication (e.g. acute hydrocephalus), a management plan should be available. This should include a policy to ensure the early identification and safe transfer of suitable patients to a neurosurgical centre. • Transfer for continuing rehabilitation; where patients have been admitted to one service for acute care but have to be referred to a separate institution for rehabilitation, needless delays can result. In such circumstances, one will often read in the case notes ‘waiting for rehabilitation’ when, of course, rehabilitation should have started on the day of the stroke (section 11.11). Thus, one needs to organize services so that the patient’s needs are matched by the care provided at all stages of his or her illness. • Transfer from hospital to community: these challenges were considered in section 17.7. Many problems can arise when patients are transferred from one service to another. Remember rehabilitation is a journey, not a destination.

17.10.3 The hazards of transfer 17.10.2 Integration of services Inevitably, most stroke patients require both communitybased and hospital-based resources at some stage in their illness. They may even need more than one hospitalbased service. It is therefore important to consider how these can be integrated to ensure that patients are appropriately placed at each stage of their illness and that transfers between each part of the service are as seamless as possible. Figure 17.1 illustrates how the components of a stroke service might fit together. Many places have the necessary facilities and skills to provide an excellent service to patients with stroke. More often, the problem lies with the organization and integration of these facilities. The key areas where problems may arise are: • Transfer from community to hospital; section 17.4 summarizes many initiatives to improve such transfers. Agreed protocols between primary and secondary care are essential. • Transfer to specialist hospital services (e.g. neurosurgery and neuroradiology); these facilities should be readily available, because urgent neurosurgery can

..

..

Where patients have to be transferred between institutions (or teams) to receive the appropriate care, there is a very real danger that continuity of care will suffer. A consistent approach to patients and their families will not be achieved without excellent communication between the professionals involved. Patients’ medical records should follow them through the system, and ideally at least one health professional should be involved in a patient’s care from admission to discharge, and perhaps even beyond.

17.11 Planning, developing and maintaining a stroke service

In the remaining sections of this chapter, we will focus on the practical aspects of planning, developing, establishing, monitoring and maintaining a stroke service. The exact manner in which the service is best provided will depend on local history, geography, needs, resources,

927

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 928

928

Chapter 17 The organization of stroke services

people and politics. Any stroke service must therefore be tailored to the local conditions to achieve maximum effectiveness. Therefore it is difficult to be dogmatic about exactly how services should be organized.

17.11.1 Determining the needs of the population If the aim of the service is to provide care for all those in the population who require it, not just those who can afford to pay for it, then the first factor to consider is the incidence of stroke and transient ischaemic attack in that population. This provides the basis for a ‘needs assessment’ which is fundamental to determining how much stroke service should be provided. Incidence of acute stroke and transient ischaemic attack Despite the huge burden that cerebrovascular disease places on communities throughout the world, there are less reliable data on its burden than one might expect. Although a large number of ‘incidence’ studies have appeared in the literature over the past 40 years, most have had methodological weaknesses that make their results, at least in part, unreliable. The criteria for an ‘ideal’ study are listed in Table 17.14. Table 17.5 gives the agespecific incidence of stroke provided by some more-orless ‘ideal’ studies, most of which were based on white populations because little reliable information is available elsewhere. There are quite good data on the incidence of stroke and TIA in many white populations, some data on Oriental populations, but little reliable data for other parts of Asia, South America or Africa.

17.11.2 Possible approaches to assessing need Because of the dearth of reliable data on stroke incidence (i.e. the number of first-ever-in-a-lifetime cases of stroke occurring in the population over a defined time period), those planning stroke services may decide to base their estimates of need on routinely collected mortality data (i.e. the number of deaths attributed to stroke in the population over a defined time period). They should, however, be aware of the potential problems of adopting this policy. Mortality statistics These depend on the collation of data from death certificates. They are thus dependent on the accuracy of death certification, which even in countries with quite high postmortem rates is known to be poor (section

Table 17.14 Criteria for an ‘ideal’ stroke and transient ischaemic attack (TIA) incidence study.122,186 A large, stable well-defined study population; the number and sex of the people in the population should be available in at least 10-year age intervals during the study. This usually requires a recent census. Complete ascertainment of all patients with either stroke or TIA occurring in that population, whether referred to hospital or not. This requires multiple overlapping methods to detect cases, including contacting primary health teams, review of hospital admissions, imaging records and death certificates. Accurate assessment of the cross-boundary flows in both directions. First-ever-in-a-lifetime strokes and TIAs should be distinguished from recurrent strokes and TIAs. Prospective assessment of all suspected cases so that standard diagnostic criteria (WHO definition) can be applied rigorously soon after the patient presents to medical attention (so-called ‘hot pursuit’). Studies should register patients with ‘TIAs’ as well as strokes to ensure that mild strokes, which may be misclassified as TIAs by referring doctors in routine clinical practice, are not under-represented. Brain imaging to determine the pathological type of stroke. This should be performed early enough after stroke onset to reliably distinguish ischaemic from haemorrhagic stroke. Case ascertainment over whole years to avoid bias due to any seasonal fluctuations in incidence. Standard methods of data presentation – i.e. not more than 5 years of data averaged together, incidence for men and women presented separately, incidence in those of over 85 years old if possible, incidence presented as mid-decade age bands (e.g. 55–64 years) but 5-year bands available, 95% confidence intervals.

18.2.1). The accuracy of mortality statistics also depends on the accuracy of the population denominators used and thus on the reliability and timing of the most recent census. Furthermore, mortality statistics only include the deaths attributed to stroke (and not the number of stroke episodes), and are not in themselves of much value in estimating the need for health services. Although they are bound to reflect, at least indirectly, the incidence of stroke in the population, the case fatality may differ from place to place, and so there will not be a uniform relationship between stroke mortality and incidence. There is some evidence that case fatality is falling, while incidence is not (section 18.2). Hospital admission or discharge statistics These are an alternative source of information which may reflect the incidence of stroke. Outpatient attendances

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 929

17.11 Planning, developing and maintaining a stroke service Table 17.15 Annual age-specific incidence of stroke per 100 000 population in the 1980s and 1990s. (Adapted from Feigin, 200512 and Sudlow & Warlow, 1997.187) Age (years) Place and mid-year of study Northern Europe Norway, Innherred (1995) Sweden, Soderhamn (1990) Denmark, Frederiksberg (1989) Finland, Espoo-Kaunianen (1990) UK, London (1996) UK, Oxfordshire (1984) UK, Teesside (1996) UK, Scottish Borders (1999) Eastern Europe and Russia Poland, Warsaw‡ (1991) Ukraine, Uzhgorod (2000) Georgia, Tbilisi (2002) Russia, Novosibirsk (1992) Central and Southern Europe France, Dijon (1987) Germany, Erlangen (1995) Greece, Arcadia (1994) Italy, Aosta (1989) Italy, Belluno (1993) Italy, Umbria (1988) Italy, Calabria (1996) Portugal, Porto (1999) Australia and New Zealand Australia, Perth (1989) New Zealand, Auckland (1991) Australia, Melbourne (1996) North America US, Rochester (1988) South America and West Indies French West Indies, Martinique (1999) Barbados, West Indies (2002) Chile, Iquiqie (2000) Eastern Asia Japan, Oyabe (1989)

65–74

75–84

≥85

217 313 306 306 221 291 297 255

741 976 712 618 516 690 611 659

1820 2056 1298

3039 2995 1599

891 1428 1247 1587

1892 2009 2099 2400

76 400 78 246

268 750 360 496

408 1500 721 1060

901§ 2600 1029 1554

1355¶ 750 1030 1513

10 16 14 13 10 5 10

62 105 82 82 114 115 69 162

119 196 218 255 242 280 149 337

410 508 568 707 720 541 570 681

979 1226 1220 1607 1317 1458 1454 1092

1641 2117 2661 3237 3413 2180 2040 1685

17 18*

98 82 105

207 253 213

511 647 535

1679 1267 1290

2369 1967 2900

9

62

269

642

1272

2111

11 8

124 94 100

272 219 308

630 578 462

1196 963 1037

1820 2208 1089

153

308

781

1940

4385

0–44

45–54

55–64

12† 12 4 21 9 11 13

40 67 104 125 87 57 89 131

14 10 12 28

*Age group 15–44 years. †Age group 15–44 years. ‡Subarachnoid haemorrhage excluded. §Age group 75–79 years. ¶Age group ≥ 80 years.

are seldom recorded. Again, inaccuracies in diagnostic codes may limit their usefulness.118,119 Also, data may be distorted by double counting, which frequently occurs when patients are transferred from an acute centre to a rehabilitation or continuing care facility. Stroke incidence estimated from hospital discharge data has been found to be much higher than that derived from a

..

..

population-based stroke register.120 But the biggest problem is that these data only include those patients who are admitted to hospital with a stroke. They are therefore a better measure of the hospital service that is currently provided than of the population’s needs. The relationship between stroke incidence and hospitalization can only be known with certainty where a reliable stroke

929

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 930

930

Chapter 17 The organization of stroke services

incidence study has been performed to determine the proportion of stroke patients who attend hospital. Where hospital admission rates have been determined, they vary considerably between places – 55% to over 95% in developed countries, while there is no information from Asia and Africa.15,16,34,121 There is also relatively little information about changes in admission rates with time.121 The rate of hospital admission, or even the attendance at an outpatient clinic, depends on several factors that are independent of population need and may be difficult to predict. Stroke prevalence is another measure of the frequency of stroke that some suggest is useful in planning services. Prevalence is the number of people who have ever had a stroke living in the population at any one point in time. Prevalence data could be useful in determining the needs for long-term support services in the community. However, we would argue that the greater the time that elapses after an acute stroke or transient ischaemic attack (TIA), the less important disease-specific services become. If one is interested in determining the need for long-term support services, it is more useful to estimate the prevalence of disability due to all causes, rather than just that related to stroke. In addition, stroke prevalence can never reflect the true burden of stroke, because the patients who die soon after a stroke are not represented. There are also a number of important methodological difficulties with measuring the prevalence of stroke and TIA, not least the need to make accurate diagnoses sometimes years after the actual event and to survey thousands of people. Where this has been done the prevalence usually ranges from 5 to 10 per 1000 population.122 As a quicker alternative, one can estimate the prevalence of stroke from its incidence and case fatality using the ‘bath principle’ (Fig. 17.6). Where this has been done for a predominantly white population in New Zealand, the prevalence was estimated to be 8 per 1000 population (9 in men, 7 in women) aged 15 years and older.123 Of the prevalent cases, around 55% had not made a complete recovery from their stroke, and 21% needed help with self-care activities. Stroke prevalence is difficult to measure directly, and of limited use for planning stroke services.

17.11.3 A practical approach to needs assessment The vast majority of health service planners are not fortunate enough to have had a recent, methodologically sound, stroke incidence study in the population they serve. Rather than carrying out their own incidence study, which is time-consuming and expensive, we recommend the following approach for determining local

Incidence Prevalence ∞

incidence case fatality

Case fatality

Fig. 17.6 A bath with water running in (representing incident cases), the water level (representing prevalence) and water going down the plug hole (representing deaths in prevalent cases). The prevalence is directly proportional to the incidence and inversely proportional to the case fatality among prevalent cases, although the mathematical relationship is quite complex.

needs, which uses both local data and information from incidence studies from other areas (Table 17.16). If one looks at the estimates of incidence from the most reliable studies (Table 17.14), it is surprising how little variation there is in the incidence of stroke (Table 17.15). However, these estimates are based mainly on white populations, and the incidence may vary in different ethnic groups. For instance, in both New York and London, the incidence of stroke is significantly higher in blacks than in whites124,125 – although this has been attributed to case-finding artifact.126 Also, the incidence of stroke among people of Pacific origin in New Zealand is higher than in those of European origin.123 However, the incidence of stroke among ethnic groups may not be the same as the incidence in the same ethnic group living in the original population from which they came, and it may be that differences between ethnic groups in the same country may be more to do with socioeconomic circumstances than race. One should start with the age-specific and sex-specific stroke incidence in the population that is closest to the local one in terms of geography, ethnic composition and culture. These rates can then be applied to the local population numbers to obtain an age-specific and sexstandardized incidence. A comparison of routinely collected data from the area in which the stroke incidence study was performed with equivalent data from one’s own vicinity may give further evidence of the relevance of the incidence data. The cause-specific mortality data are likely to be the most reliable and easily obtained. One can then judge whether the incidence locally is likely to be greater or less than that in the available incidence study. An alternative approach might be to use local hospital admission or discharge data and to adjust them to take

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 931

17.11 Planning, developing and maintaining a stroke service Table 17.16 A step-by-step guide to estimating approximately the number of strokes in a local population of interest.

Step 1 Step 2 Step 3

Step 4 Step 5

Step 6 Step 7

Obtain the most accurate census data for the population of interest for each sex and age band. Identify the ‘ideal’ incidence study that is likely to have been done in a similar population to the one of interest (e.g. geography, race). Multiply each age-specific and sex-specific incidence by the number of people of that age and sex in the population of interest – e.g. if the incidence in men 65–74 years old is 690 per 100 000 and there are 11 000 men in this age band in the population of interest, one would expect about 76 men (i.e. 690 × 11 000/100 000) between 65 and 74 years of age to have a stroke each year in the population of interest. Sum the numbers of patients of each sex in each age band expected to have stroke to obtain the total number expected in the population. Consider making an adjustment for any marked differences in, for example, the cause-specific mortality between the population of interest and the population in which the incidence study was done. If one is interested in transient ischaemic attacks, then their incidence is usually about 30% that of stroke (section 18.4). The number of recurrent strokes is of the order of 30% of first-ever-in-alifetime strokes, so that to estimate the total number of strokes likely to occur in the population of interest, the number of incident strokes should be inflated by 30%.

If interested in the total numbers of patients who could be referred to your service with suspected stroke or TIA then add the additional number of referred patients who may be expected to have a non-stroke diagnosis (possibly 50% for TIAs and 30% for acute stroke).44,188

account of the likely proportion of stroke patients admitted. This will be more reliable for stroke than for TIAs. One could estimate this proportion admitted by surveying the local primary health teams and asking them to report the proportion of patients with acute stroke who they refer to hospital. This may be reliable enough if they report that they refer virtually all cases (as in Sweden), but if the proportion is smaller, their estimate may be misleading.16 Also, hospital discharge data may be inaccurate for a number of reasons:118,119 • inaccuracy of routine clinical diagnosis, especially of TIAs; • lack of CT brain scans to confirm the stroke diagnosis and exclude other diagnoses; • use of vague terms, e.g. ‘acute hemiparesis’, ‘cerebrovascular disease’ in medical records and discharge summaries from which routine codes may be derived; • coding errors; • failure to distinguish acute stroke admissions from those due to complications of an earlier stroke; • failure to code strokes occurring in hospital, or in the context of another diagnosis. Table 17.17 gives the likelihood that a patient allocated one of the cerebrovascular codes on discharge from one of five Scottish hospitals had actually had an acute stroke. It was also reported from Norway120 that certain diagnostic codes were more likely to identify true strokes

..

..

than others. Therefore, if one plans to use hospital discharge data, one would be wise to be selective in one’s choice of discharge codes. The crude incidence of stroke tells one how common the problem is, but in itself is not sufficient to determine the health service needs of the population. Before one can plan a service, one needs to have estimates of the following. Age-and sex-specific incidence. Younger patients require different facilities from older ones, e.g. retraining for employment. Older women more often live alone, and may require more formal support in the community. Type-specific incidence (that is transient ischaemic attack, ischaemic stroke, primary intracerebral haemorrhage and subarachnoid haemorrhage). These data may be useful in more detailed planning of the population’s needs. Patients with TIA and minor ischaemic stroke require prompt diagnosis, investigation and initiation and supervision of secondary prevention but not prolonged inpatient care, rehabilitation or community support services. Patients with subarachnoid haemorrhage require emergency hospitalization and neuroradiological and neurosurgical facilities (Chapter 14). Haemorrhagic strokes have a higher early case fatality, so although they may require more care in the very early stages, the longer-term burden of severe disability may be less than for patients with ischaemic stroke. The

931

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 932

932

Chapter 17 The organization of stroke services

Code

All hospital admissions

Emergency admissions only

Table 17.17 Positive predictive values of cerebrovascular disease codes (ICD 9 or 10) for a diagnosis of stroke.

Positive predictive value for acute stroke (%) Subarachnoid haemorrhage Intracranial haemorrhage Non-traumatic intracranial haemorrhage Stenosis/occlusion of a precerebral artery Ischaemic stroke Transient ischaemic attack Stroke unspecified Cerebrovascular disease, unspecified Cerebrovascular disease sequelae

85–88 87–91 58–75 1 83–93 5 78–87 25–28 0

88–93 92–94 63–75 0 86–95 6 82–88 31 0

The proportion of patients who were allocated a cerebrovascular code (ICD 9 or 10) on discharge from five Scottish hospitals and who actually had had an acute stroke. The values are higher for emergency admissions. This study did not attempt to identify patients who had a stroke but received a non-stroke code.

estimates of the relative frequency of the pathological type of stroke are remarkably similar in most of the published incidence studies, which come from predominantly white populations (Fig. 17.7). Although the proportion of patients who are haemorrhagic is probably higher in Oriental populations, it is difficult to be certain, because of limited information. The proportion of strokes due to subarachnoid haemorrhage and primary intracerebral haemorrhage is greater in the young than the old. The prognosis of stroke and its subtypes. From the typespecific incidence and case fatality, one can estimate the likely requirements for assessment and diagnostic services, acute care, rehabilitation and terminal care services, long-term care, community support and secondary prevention. The prognosis of stroke and its subtypes has been discussed elsewhere (section 10.2.3). Information on stroke severity and comorbidity would be invaluable, since these will be major determinants of patients’ health service use. However, such data are unlikely to be routinely available unless a stroke register has been kept. Changes over time. When one is planning a stroke service, one has to take account of any changes that may occur in the future, because one’s service will need to alter to take these into account. Trends in stroke mortality and incidence are discussed later, and there is evidence that the incidence, and maybe even the severity, of stroke is falling in many western populations. Apart from the changing incidence, one also has to take into account changing demographics. In most developed

countries, older people account for an increasing proportion of the population. Therefore, for a disease such as stroke in which the incidence is much higher in older people, the total number of strokes will increase unless offset by a falling incidence. However, one has to remember that case fatality is higher in older patients and in those with prior disability, and therefore fewer of these older stroke patients will survive to require longterm care. Changes in clinical practice are likely to force rapid and quite unpredictable alterations in the shape of clinical services for stroke. The changes in management of stroke have, until now, been small and gradual, but with recent increases in research efforts, major changes are more likely and the needs of the population could alter rapidly and unpredictably. The introduction of coiling rather than clipping of ruptured intracranial aneurysms after the ISAT trial (section 14.4.2) has resulted in large, sudden and difficult-to-manage changes in services; patients once admitted under neurosurgeons and operated on may now be admitted under neurologists and treated by radiologists. People’s expectations of healthcare, often driven by media reports of medical successes and failures, may also force changes in services that go beyond the evidence of effectiveness.

17.11.4 Assuming that resources are limited, to what extent will the needs of the population be met? Having obtained an estimate of one’s local age-specific, sex-specific and type-specific incidence and outcome of

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 933

17.11 Planning, developing and maintaining a stroke service

100%

10 2 11

90% 80%

12 5

5

6 5 11

9

2 5

9 3

10

15

10

9 1 8

15 2 10

70% 60% 50% 40% Fig. 17.7 Histogram showing the proportions of patients aged between 45 and 84 years with a first-ever-in-alifetime stroke due to ischaemia, primary intracerebral haemorrhage, subarachnoid haemorrhage and uncertain, in more or less ‘ideal’ incidence studies. Only studies where a computed tomography (CT) brain scan was done in more than 70%, and including patients with subarachnoid haemorrhage, are covered.

77

..

86

78

83 73

82

30%

72

20% 10% 0%

Dijon, France

Perth, Australia

Umbria, Italy

Rochester, USA

Undetermined Subarachnoid haemorrhage Primary intracerebral haemorrhage Ischaemic stroke

stroke and TIA, one is still a long way from determining the needs of the population. After all, this estimate does not indicate the actual resources these people will require, and of course at some stage somebody has to make a political decision about how completely the population’s needs are to be met. Inevitably, where there are limited resources for health services, choices have to be made about allocating resources between areas (e.g. should one build a new hospital or a new school?). Having been given limited resources, however, it is important that those planning stroke services should use them in as an efficient a manner as possible and should prioritize (i.e. ration) to make sure that sufficient funds are available for whatever are perceived to be the most important aspects of the service. These decisions are usually taken by politicians but, because they depend on information about the effectiveness of components of the stroke services, it is essential for the politicians to receive sound medical advice. The perceived pressures to develop particular components of stroke services and the way in which these services develop are likely to reflect a number of factors apart from the effectiveness of the service offered: • Local healthcare culture and economy. The types of services provided for stroke patients vary greatly between the developed economies. In the US and Germany services often focus around centralized neurological centres or stroke centres where selected patients receive acute care but not ongoing rehabilitation. There is evidence that access to these services127 may be reduced for certain patient groups such as the elderly. In contrast the Scandinavian countries have tended to approach stroke services with a more devolved delivery offering

..

73

Oxfordshire, England

Aosta, Italy

Frederiksberg, Denmark Soderhamn, Sweden

The numbers by the columns indicate the percentage estimates

similar services in all parts of the country. In the UK, stroke services are tending to evolve out of existing rehabilitation facilities, which were traditionally provided by geriatric medicine or general medicine services. • The needs of the patient population. Within stroke care, as within healthcare in general, there is frequently a tension between the need to centralize services for individuals with very specialist needs (for example, subarachnoid haemorrhage) and the aspiration to provide as much high-quality care as close to the individual’s home as possible. • Patients’ wishes. Once again there is a tension between pressures to centralize and fragment care, for example, between acute and rehabilitation services and to provide continuity throughout the patient journey which is usually valued by patients and carers. • Costs and resources. There are increasing pressures to control the costs of healthcare services and this is particularly relevant when considering emergency cover by senior medical staff. Once again centralized services meet these challenges by having large groups of staff, or at least networks of staff, who are available to be on call.

17.11.5 Planning a stroke service Having established the aims of the service, the needs of the population and the degree to which one expects to fulfil these, one can then plan the development of services. Usually, one starts with the existing services, even if these are inadequate and chaotic. It is then useful to consider the following two questions prior to making any changes:

933

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 934

934

Chapter 17 The organization of stroke services

• What are the current resources committed to the management of patients with TIA and stroke? • What are the major gaps (i.e. unmet needs and failure to provide cost-effective interventions) in the present provision of services? One needs to establish the strengths and weaknesses of the current services in order to identify the most important areas for improvement. Priority should be given to providing patients with basic care (e.g. nursing to provide for basic needs) and to delivering those interventions that are of proven effectiveness. Although information about current services may already be available, it is likely that in addition to routinely collected data, it will be useful to carry out a survey to determine: • How many and what sort of patients are currently being managed (i.e. demographic and clinical data)? • Where are they being managed (i.e. in the hospital or community, in accident and emergency departments, neurology, internal medicine or geriatric medicine)? • By whom are they being managed (i.e. general or family practitioner, neurologist, general physician)? • How are they being managed (i.e. the process of care)? • What resources are currently being used? A community-based register that identified all patients in the population who had a stroke or TIA would be an ideal but expensive and impractical way of answering these questions. A hospital-based stroke register is a practical alternative that can help to answer most of the questions, although clearly it cannot provide detailed information about patients who are not referred to hospital. A register is an invaluable tool for monitoring the performance of services as well as planning them. Set up a hospital-based stroke register to get some idea of the current state of the local stroke service – however fragmented and chaotic it is. The simplest, quickest and most practical approach is to carry out a survey of the current services against certain standards. Those working within a service are often fully aware of its deficiencies, and their knowledge should not be ignored (although sadly it often is). This will only work where there is a willingness by the important parties to acknowledge deficiencies and to change practice. This approach can identify areas of strength and weakness, and may determine which area to concentrate on first (e.g. Sentinel Audit128). An honest, objective appraisal of services against some agreed standards by those involved in providing them, perhaps facilitated by an independent observer, may be the best stimulus for service development.

The immediate priorities for improving a service will also depend on how easily or cheaply particular problems can be solved. For instance, poor standards of medical assessment may be improved by the introduction of a protocol and education for junior medical staff with little implication for resources, while the provision of an occupational therapy department where there isn’t one has major resource implications. Planning a stroke unit is likely to be less challenging in a well-resourced hospital setting where reorganization of staff and work patterns is needed. However the obstacles are formidable if the key staff members are not available at all, and even if they are many staff do not like being reorganized without very good reason, much explanation and proper support.

17.12 Evaluating and monitoring stroke services

We believe that the most reliable way of determining the relative effectiveness of interventions is an appropriately designed randomized trial where feasible, or a systematic review if more than one trial is available. This of course is not an option in the evaluation of a local stroke service rather than stroke services in general, so we have to rely on less robust methodologies. Non-randomized comparisons of the process of care, or of patient and carer outcomes achieved by services, are the only practical methods of evaluation. If one is setting up a new service in a hospital, the process, or outcomes, can be compared with those in a nearby hospital without a new service. Alternatively, one could measure the process and outcomes achieved by the existing service and then measure whether these are improved after the new service has been established (i.e. a beforeand-after study). However, such evaluations can be misleading. They may demonstrate improvements (or worsening), or differences in the process or outcome of care, but they cannot provide reliable data to indicate that any changes are actually due to the new service. One cannot rely on such non-randomized evaluations to influence practice elsewhere, but they may fulfil important local functions. One could reasonably argue that, as long as the new service is not much more expensive to run, it does not matter whether any improvements observed can be attributed directly to that service. Obviously, if the new service was very costly, one would want reassurance that the improvements had not occurred spontaneously. But if a non-randomized evaluation

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 935

17.12 Evaluating and monitoring stroke services

demonstrates either no improvement or even a worse outcome, it is difficult to know how to respond. Were the changes due to the new service, in which case it ought to be modified, or were they due to some unforeseen confounding factor? There are several other methodological problems, which can affect randomized as well as non-randomized comparisons, and which need to be considered: • small numbers of patients; so that any change observed may be accounted for by the play of chance or missing a real and worthwhile change because too few patients were studied; • observer bias in assessing the process or outcome; often, the observers have an interest in the result of the evaluation, which may influence their judgements. So what aspects of a service can we measure? The simplest way to monitor the service is to measure the amount and nature of work being carried out. Unfortunately, politicians and those who fund healthcare place too much emphasis on the volume rather than quality of service. A stroke service may include some components (e.g. processes) for which there is little doubt of their effectiveness (e.g. aspirin for cerebral infarct, low-risk carotid surgery). When assessing performance, it is therefore most important to monitor how well the service delivers those components whose effectiveness is established. In assessing the quality of care (or ‘clinical audit’, as it tends to be called) one should consider three aspects of the service:128 the structure, or facilities available; the process of care; and the outcomes for those treated. Structure From all that has been said above, it should be fairly obvious that a stroke service needs certain essential facilities to provide all the components necessary to care for patients with stroke and TIA (Tables 17.1 and 17.2). This makes the setting of standards and the measurement of performance for structure relatively straightforward. Some basic standards (incorporated into Scottish service standards; http://www.nhshealthquality.org/nhsqis/ 1288.html) for structure might include: • an identified individual who is responsible for the organization of stroke services; • early access to a stroke unit; • multidisciplinary staffing of the stroke unit (section 17.6); • early access to outpatient assessment by a specialist for patients who do not need admission to hospital (i.e. a ‘one-stop’ neurovascular outpatient clinic; • prompt access to CT scanning;

..

..

• prompt access to non-invasive vascular imaging, and carotid surgery if necessary (section 16.11); • continuing care facilities, both community-based and institutional. However, care has to be taken in defining these standards. For example, what does ‘prompt access’ really mean, what is a specialist, what does a stroke unit consist of? One could very quickly fulfil a requirement for a stroke unit by simply relabelling a general medical ward, but of course one is unlikely to accrue the benefits of stroke unit care.

7.12.2 Process Some aspects of the process of care are easily monitored, e.g. waiting times for appointments. It may be relatively easy to define a standard (for example, patients with TIAs and ischaemic strokes should be given aspirin to reduce the risk of further vascular events, unless there are contraindications), but for other procedures for which there is less scientific justification, it is more difficult. The lack of scientific justification can, and frequently is, overcome by using the combined views of recognized experts, i.e. a consensus. However, what appears to be a fairly straightforward standard cannot be applied sensibly to every patient because – for example – they cannot comply with the assessment. The UK National Sentinel Audit of Stroke71 addressed this problem by having a ‘no but . . .’ clause attached to each standard (Fig. 17.8). This is essential for the process of care to be compared in different groups of patients. Another difficulty is that, by directly observing the care, one is likely to alter its delivery (the so-called Hawthorne effect). Also, such an approach is likely to be very costly if performed on all stroke admissions. The alternative is to audit the records of care, but this immediately raises the question of the validity of the medical record, i.e. whether the records reflect the actual care provided. However, most people would agree that good records probably do reflect good care given, and that this is a reasonable method of measuring the process of care. The other methodological problems involved in audits of case notes are summarized in Table 17.18. Monitoring the process of care by case note review raises a number of important methodological problems that must be addressed if one’s assessment is to be useful and valid. Although one may be measuring the performance against some ‘ideal’ standard, one is likely to want to compare the performance in the same service over time, or to compare performances between services. To do this

935

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 936

936

Chapter 17 The organization of stroke services

Patient assessment 2.1.0 First 24 hours Neurological assessment in the first 24 hours 2.1.1 (a) Have the following been specifically recorded in the first 24 hours? Yes No No but (i) Conscious level (eg Glasgow Coma Scale, alert / oriented) (ii) Eye movements (eg Cranials 3, 4 and 6 intact, doll's eye response if unconscious) (iii) Limb movements (response to pain if conscious)

(b) If patient is noted to be conscious, are the following recorded? Answer No, but if....... impaired level of consciousness is documented Yes No No but (i) Screening for swallowing disorders (not gag reflex) (ii) Communication (iii) Trunk control or gait

(c) If patient is alert and able to communicate, is there a formal assessment of: Answer No. but if..... impaired level of consciousness/communication is documented. Yes No No but (i) Formal mental test (eg mental test score) (ii) Visual fields (iii) Visual inattention (iv) Sensory testing

Table 17.18 Important methodological issues in the audit of case notes. Patient selection bias The case notes audited should be a representative sample of all those treated (either a consecutive series or a random sample). Beware the missing notes because so often these patients are more ‘interesting’ to somebody, perhaps because they had a rare form of stroke, or died. Case note retrieval bias Poor-quality case notes or those of dead patients may be more difficult to retrieve, which might bias the audit in a favourable direction. A high proportion retrieved is an important step in reducing bias.189 Lack of precision A sufficiently large number of case notes should be audited to provide a precise estimate of performance and to allow precise comparisons to be made with other centres, or with audits performed at different times in the same centre. Observer bias Auditors may have an interest in the result of the audit, which may influence their assessment of performance; blinded, or at least impartial, observers should be employed if possible. Poor inter-observer reliability of measure If the measure of performance is not reliable, then it will be more difficult to demonstrate real differences between centres. Also, if there is a consistent difference in the way an audit measure is applied by different auditors, this may produce invalid comparisons. Differences in case mix A standard that is applicable to one patient may not apply to another. It is important that standards should be adjusted for differences in case mix.

Fig. 17.8 One of the sections from the National Sentinel Audit of Stroke. This one relates to the recording of the neurological examination. The interval (e.g. 24 h) since admission at which these data should be recorded, and the circumstances in which it is acceptable for the information not to be recorded, are given in each section.132

meaningfully, one has to have a valid and reliable measure of performance, one needs to audit enough cases to produce statistically robust results, and one must be able to take into account differences in case mix between services or changes over time. The Stroke Audit Package was originally developed by the Royal College of Physicians to overcome these methodological problems and enable valid comparisons to be made.129 But because this package only addressed a limited number of the more medical aspects of care, the Royal College of Physicians of London therefore also developed the National Sentinel Audit, which includes other important aspects of care delivered by different members of the multidisciplinary team.130

17.12.3 Outcome The term ‘outcome’ is used in different ways and so causes confusion. Clinicians use the term to refer to the clinical outcome of the patient or carer. Outcomes therefore include survival, functional status, complications, or less easily defined concepts such as quality of life. Others, particularly those with a management background, use the term ‘outcome’ to refer to any result of an intervention, e.g. reduced waiting times or readmission rates. In this section, we use ‘outcome’ to refer to the clinical outcome of patients (i.e. physical, functional, cognitive, emotional, etc.). Since the main aim of stroke services is to optimize the outcome for patients and carers, the measurement of outcome is obviously the most relevant criterion by which to judge the performance of a service. This recognition has prompted the development of outcome analyses such as Dr Foster’s notes (drfoster.co.uk). Unfortunately,

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 937

17.12 Evaluating and monitoring stroke services

the use of outcomes to reflect the quality of care is the most challenging area of stroke audit, and there are many well-known difficulties to be overcome.131–136 Until they are, those involved in providing and monitoring health services must be extremely careful not to misinterpret outcome data. The observed outcome in a group of patients treated by a particular service will be determined by four factors: • the quality and effectiveness of the care provided; • the method of measurement of outcome (e.g. who is measuring it, and how?); • chance (or random error); • case mix (or mix of prognostic factors at baseline).

17.12.4 The quality and effectiveness of care This is the aspect we hope outcomes will reflect. However, it is important to remember that most interventions have only small or moderate-sized effects, which may be difficult to detect even in large randomized trials. For example, a wildly implausible 50% relative reduction in the death rate from 30% to 15% after the opening of a stroke unit would require a sample of 200 patients both Table 17.19 Aspects of outcome that may be relevant in assessing stroke services, and some tools for measuring these outcomes.

Many attempts to monitor the quality of service by measuring patient outcomes have relied on mortality data, presumably because they are often routinely available, reasonably objective and may indicate where there are major problems. Unfortunately, mortality is unlikely to be influenced by many components of care (discharge planning being one obvious example). Some outcome measures that may better reflect the quality of care are shown in Table 17.19. They measure outcome at different Promising measurement tools

Survival

Case fatality during a defined time period, e.g. at 30 days or 6 months after stroke onset Proportion of patients developing pressure ulcers or fractures; there are difficulties in defining these and reliably recording them; paradoxically, better services may identify more and record them more often Probably not very useful, and not easily collected after hospital discharge 10-m walking speed Nine-hole peg test Hospital anxiety and depression scale General health questionnaire Many of the most disabled patients will not be capable of responding to measures of psychological outcome Barthel Index Functional Independence Measure (FIM) Three simple questions (Fig. 17.9) Oxford Handicap Scale (also known as the Modified Rankin Scale) – should be measured at a defined point after the stroke, e.g. 6 months; easily collected after hospital discharge London Handicap Scale190 – a difficult area with no well-tested measures; the Oxford Handicap Scale does not really address handicap in isolation Hospsat and Homesat99 – is this an outcome or process measure? Nottingham Health Profile Short Form 36 Euroqol – potentially interesting because it could allow comparisons with other disease states; however, many stroke patients cannot complete the questionnaires because of cognitive problems

Residual impairments Mobility Arm function Psychological outcome

Disability

Handicap

Patient or carer satisfaction General health or health-related quality of life

..

17.12.5 The method of measurement of outcome

Outcome

Complications

..

before and after its introduction to eliminate the effects of chance, and in itself contribute nothing to the elimination of biases of various sorts. The impact of stroke unit care is far less than this (Table 17.6). There are several examples of studies of stroke unit services71,91 where treatment in a hospital with a stroke unit was associated with improved outcomes even after adjusting for case mix. However, they were based on large patient samples of several thousands, and smaller studies have often failed to show a clear link between process and outcome.137

937

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 938

938

Chapter 17 The organization of stroke services

Validity, different types include: • Criterion validity, when the measure is related to an accepted ‘gold standard’. • Construct validity, where the measure is related to existing measures of similar aspects of outcome. • Content (or face) validity, which relies on expert agreement that the measure is a reasonable reflection of what it is supposed to be measuring. There can be considerable difficulties in demonstrating the validity of a particular measure.192 Reliability This is the reproducibility of a measurement, most commonly between observers (inter-observer reliability) and over time (intra-observer or test-retest reliability). Relevance The scale should measure some aspect of outcome that is relevant to the patient or carer as well as to the doctor. Thus, the size of a cerebral infarct on a CT brain scan is of little relevance, while patients’ ability to look after themselves is very important to the patient and carer. Practicality Scales vary in their complexity and the time taken to complete an assessment. Studies of long-term outcome involving hundreds of patients need very simple measures which can be completed by postal or telephone questionnaire, while smaller studies in hospital can afford to use more complex measures. Sensitivity A scale should distinguish between patients who have different outcomes or detect important changes in a particular patient. Usually, more sensitive scales are more complex and unfortunately less reliable. Communicability Ideally, the measure will mean something to other health professionals or even patients. It is more useful to know that a patient feels ‘fine’ than to be told that their score on a particular stroke scale was 23 out of 100, for example.

levels of disease – pathology, impairment, activity, participation and quality of life. It is important that they should have acceptable validity (i.e. they measure what they are intended to measure) and reliability (i.e. they are reproducible in different settings and when used by different people). A number of different types of scale have been developed and used to measure outcome after stroke and there are a number of features that one should look for in choosing an outcome measure (Table 17.20).138 One can loosely categorize the measures of outcome after stroke under the following headings. Stroke scales So-called ‘stroke scales’ (e.g. the Scandinavian Stroke Scale, Canadian Stroke Scale, National Institute of Health Stroke Scale) were largely developed to describe the severity of acute stroke and to monitor changes in the patient’s condition.139,140 Most concentrate on the type and severity of the neurological impairments. They have been criticized for lacking relevance for patients, being complex and therefore impractical, and for summing ‘apples and pears’.138 They rely on a clinical examination,

Table 17.20 Important features of scales for the measurement of outcome after stroke.

although reduced scales can be completed from case notes.141 We do not think they are particularly useful in evaluating stroke services. They might be used as a measure of case mix. ‘Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise’.142 Functional scales (activity) These include measures of disability or dependence in activities of daily living (ADL), such as the Barthel Index, the Nottingham ADL Scale and the Functional Independence Measure (FIM).143,144 Under this heading one could also include the so-called extended ADL (EADL) scales such as the Frenchay Activities Index and the Nottingham Extended ADL Scale145–147 which identify whether patients are participating in more complex activities such as shopping, leisure or work. These scales appear to measure relevant aspects of outcome, although some demonstrate ceiling effects (e.g. the Barthel Index) and may not pick up problems in particular areas, e.g. communication. Indeed, one can score maximum

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 939

17.12 Evaluating and monitoring stroke services

points on the Barthel and yet be blind, deaf and mute. The scales are in general ‘ordinal’, so that care must be taken in choosing the appropriate statistical method to describe or compare groups of patients. Some of these scales are simple enough to incorporate into a postal or telephone questionnaire, and may therefore be used in large studies of long-term outcome.148 Handicap (participation) Handicap, or its more acceptable converse, participation, is difficult to define and therefore difficult to measure, but is undoubtedly of relevance to stroke patients and their carers. The Oxford Handicap Scale (Table 17.19), which is a modification of the Rankin Scale, sounds from its name as if it measures handicap, but it really measures a combination of symptoms, dependency and change in lifestyle.149,150 However, it has been widely used, is relevant and simple enough to be used reliably over the telephone, and is therefore useful in large studies.151 Precision of assessment can be improved by using a structured interview.152

Is the patient alive? No Yes

Dead

‘Do you require help from anybody for everyday activities?’ Yes No

Poor

‘Has the stroke left you with any problems?’ Yes Indifferent No Good

Quality of life Like handicap, quality of life is difficult to define and thus measure. A large number of generic measures (otherwise known as multidimensional measures) have been developed that attempt to measure outcomes in relation to various aspects, including physical function, psychological function, pain and social function. They include the Short Form 36, Nottingham Health Profile, EuroQol, Sickness Impact Profile, Stroke Impact Scale and the Assessment of Quality of Life (AQoL) instrument.153,154 Most provide a profile of outcome rather than an overall measure, and group comparisons are therefore complex. However, the EuroQol and AQoL provide a single measure of ‘utility’, and researchers have made some headway in deriving summary scores for the other measures. Because they are generic (i.e. can be used across many different health states) they offer health economists and others the opportunity to compare the utility of different health outcomes in different diseases. Some are long and complex (e.g. the Sickness Impact Profile) and are not suited to large-scale studies in which face-to-face administration is not practical. Also, they all rely on patients’ views of their health status, which limits their use in patients with severe communication and cognitive difficulties. It is unclear how valid the carers’ responses on behalf of the patient are to these questionnaires.155 Like motherhood, quality of life is much admired, difficult to define and even more difficult to measure.

..

..

Fig. 17.9 ‘Three simple questions’ that can be used to place stroke patients into four different outcome categories.160

Three simple questions We have used ‘three simple questions’ to categorize patients into those with poor, fair and good outcomes after stroke (Fig. 17.9).103,156 This approach appears to be reasonably valid and reliable, and is certainly practical when the outcome of very large numbers of patients needs to be measured. Further work is required to establish the optimal wording of the simple questions and to test them in different languages and settings.157 Patient satisfaction Many healthcare systems are being influenced by market forces and the idea that patients are consumers. This has placed increasing importance on the satisfaction of our ‘clients’ (i.e. patients) with their healthcare. Many health service managers regard patient satisfaction as being an important outcome, although some consider satisfaction to be a measure of process; measures of patient and carer satisfaction with hospital and home care have been developed.97

939

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 940

940

Chapter 17 The organization of stroke services

Patient and carer satisfaction appears to reflect the process of care and the patients’ outcomes. Improvements in satisfaction have been observed in trials of occupational therapy158 and early supported discharge services.70 As one would expect, those with poorer physical outcomes and depression are likely to report less satisfaction with care.159 However, patients – and particularly women and the elderly – appear to have low expectations and are often satisfied with what professionals would regard as poor treatment.72,159 When should we measure outcome? Outcomes some months after the stroke are probably most relevant to patients, but are more difficult and expensive to measure than at an earlier stage. Many services monitor the patient’s functional status at the time of hospital discharge, and this information is easily and cheaply collected. However, because patients usually improve for several months after a stroke, the longer they stay in hospital the better their outcome at discharge. Furthermore, patients tend to be discharged at much the same level of disability. Thus, such measures are easily manipulated and difficult to interpret. It is far more relevant to measure the outcome at a fixed interval after the stroke, but after discharge this will inevitably be more time-consuming and expensive. However, some of the simpler measures can be completed by telephone or postal questionnaire.148,151,160 Some measures (e.g. EuroQol), which seem ideally suited for use as postal questionnaires, include visual analogue scales, but these appear to be particularly unreliable in stroke patients.161

made. Thus, it is important to measure outcomes in a large, representative sample of patients or carers. This has implications for the type of measure of outcome used, since it must be simple and practical to administer to large numbers, exactly as in large randomized trials. Therefore, to obtain precise estimates of outcome may take several years for a hospital to accumulate enough data to provide precise estimates – for example, of its case fatality. Thus, there is likely to be a considerable delay between a change in the quality of care and any statistically significant change in measured outcomes. Figure 17.10 shows the case fatality in five Scottish hospitals with and without adjustment for important case mix variables. This illustrates that even with several hundred patients, the 95% confidence interval around the estimates of case fatality are surprisingly wide. We suggest that power calculations should be performed (as for randomized trials) before instigating any audit to demonstrate changes in the outcomes following modification of a service. 1. Unadjusted A B C D E

2. Adjusted for age, sex and deprivation

How to score dead patients? Because many patients die after stroke, outcome measurements can only be applied to the survivors. If these measures are averaged, and groups of patients compared, then there may be a serious problem of interpretation if there are more survivors in one group than in the other. Some attempt to get round this by giving the worst score to the dead patients and then including them in the analysis but, depending on the scale used, this is not necessarily valid. One solution is to measure the proportion of patients who are ‘dead or disabled/handicapped’, but this may sacrifice sensitivity. However, in studies including large numbers, such dichotomized outcomes may be adequate.162

17.12.6 Chance (or random error) With small numbers, the imprecision of the estimate of performance may prevent useful comparisons being

3. Adjusted for study variables

A B C D E

A B C D E –20

–15

–10

–5

0

5

10

15

20

Fig. 17.10 Number of deaths above or below that predicted per 100 stroke patients admitted (w score) to each of five Scottish hospitals (A through E) at 6 months. Plots are shown for (1) unadjusted case fatality; (2) after adjustment for age, sex and deprivation; and (3) after adjustment for the study prognostic variables (age; whether the patient lived alone before the stroke and was independent in simple activities of daily living; and, on admission, whether the patient could speak and was orientated in time and place, could lift both arms against gravity, and could walk without the help of another person). Used with permission from Weir et al., 2001.136

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 941

17.12 Evaluating and monitoring stroke services

When planning an audit, estimate the likely number of cases that will need to be included in order to identify a difference reliably (i.e. do a power calculation).

Although the evidence that institutions with greater throughput have better outcomes is conflicting (e.g. for carotid endarterectomy), one argument for stipulating a minimum patient volume per year is to ensure that measures of performance can be reasonably precise. If one’s local surgeon performed 50 operations in the previous year, with only two deaths or perioperative strokes, this is a very acceptable 4% complication rate. However, the 95% confidence interval extends up to a very unacceptable 14%. It is therefore very difficult to know with any certainty whether one’s local surgeon has results that make carotid endarterectomy worthwhile. If outcomes are measured in a relatively small number of patients or carers, bad outcomes may reflect bad luck rather than bad care, while conversely, good outcomes may reflect good luck rather than good care.

17.12.7 Case mix The most important determinant of outcome is probably not the quality or even the effectiveness of care, but the type of patient treated. The patient’s age, pre-stroke status, comorbidities and the severity of the stroke are bound to have an overwhelming effect on the outcome, and may well obscure any real effect that our treatments may have. That is why large randomized trials are required to demonstrate modest treatment effects. Case mix can vary considerably in different services and in the same service over time, which means that raw outcome data simply cannot be used to reflect the quality or effectiveness of care; they have to be adjusted for differences in case mix. Unfortunately, this assumes that we know how to adjust for case mix in stroke, which is not the case. Good case mix descriptors include those factors that are highly predictive of outcome. But as we have already seen, our ability to predict outcome after stroke, even in terms of survival, is relatively poor (section 10.2.7). Also, we can only correct for those prognostic factors that we can identify and measure reliably. After all, we rely on randomized controlled trials to provide evidence of the effectiveness of interventions simply because randomization ensures that the different treatment groups are balanced for recognized, unrecognized and unmeasurable prognostic factors and that the treatment allocation is not biased. In the report comparing outcomes after stroke in Scottish hospitals, the only

..

..

prognostic factors that were routinely collected, and could therefore be adjusted for, were age, sex and social deprivation.136 If more powerful predictive factors such as those identified in the Oxfordshire Community Stroke Project (Fig. 17.10) are taken into account, most of the variation between hospitals with respect to case fatality disappears and can be accounted for by chance alone.136,163 Unfortunately, these variables are not routinely available to allow such adjustment. If variations in outcome remain, it is impossible to know whether they are due to failure to completely adjust for case mix, or some aspect of the care given.84 In accounting for random variation, one must also recognize the imprecision of any statistical model used, which will depend on the size of the cohort from which the model was derived. The problems of adjusting for case mix are even greater if one considers other relevant outcomes such as quality of life, where we know almost nothing about the factors that predict this. Before we use outcomes to reflect the effectiveness and quality of care, and to alter services as a consequence, we have to develop reliable methods of interpreting them. It will be interesting to see whether this will ever be possible. Crude measures of patient outcome do not necessarily reflect the quality or effectiveness of the care provided. Even adjusting for case mix may not solve this problem. Rather than attempting to interpret measures of outcome at a particular interval after the stroke, the change in the patient’s condition can be used as an ‘outcome’. For example, some research groups have used the functional independence measure (FIM) to assess patients on admission to and discharge from a treatment programme. Any change in the FIM might be considered, at least in part, to be a measure of the effectiveness of the treatment programme, although most of the improvement may actually be spontaneous. The change in FIM can be divided by some measure of the amount of treatment provided, e.g. length of stay, to give an idea of ‘efficiency’. Unfortunately, differences in case mix such as age, severity and location of the brain lesion and other medical problems, as well as the interval since the stroke, are likely to influence the rate of change in the FIM. To interpret the change in the FIM as a reflection of effectiveness or efficiency would therefore still require a measure of case mix.164 Another problem is that measures such as the FIM are not ‘interval’ scales (Table 17.19). A change of 10 points at one end of the scale is therefore not equivalent to a 10-point change at the other. This makes changes in score difficult to interpret.

941

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 942

942

Chapter 17 The organization of stroke services

17.12.8 A practical approach Given the limitations of using structure, process or outcome measurement, it is probably best not to rely on any one method, but to use a combination. This has the advantage of reflecting most aspects of the service. Also, if one relies on only a small number of performance indicators, it can have a number of adverse effects on health services. Effort and resources may be directed at improving one service or one aspect of a service, to the detriment of other areas (so-called ‘measure fixation’). Clinicians and managers whose income or reputations depend on the indicators may alter their practice (e.g. refuse to treat sicker patients) or manipulate data to enhance their apparent performance (i.e. gaming). For instance, by altering the coding of factors used to adjust for case mix (e.g. stroke severity), one can increase the expected mortality for one’s patients and thus improve one’s apparent performance (i.e. the observed/expected mortality ratio). One of the best examples of such gaming was following the publication of institution-specific and surgeon-specific death rates for cardiac surgery in New York.165 The reported proportion of patients with comorbidities, such as renal failure and chronic obstructive pulmonary disease, which were used to adjust outcomes for case mix, increased several-fold over 2 years. It is highly unlikely that the type of patient had changed. The surgeons were simply reporting more comorbidity. Whatever measures are used it is important that the audit covers all stroke patients, not just those admitted to a stroke unit. We suggest the following approach to assessing the quality of stroke services, although this will inevitably change as our understanding improves: • Perform a structural audit to ensure that services include access to the facilities shown in Table 17.7. • Assess the process of care by performing a regular audit of a representative sample of case notes using a well-tested and reliable audit instrument that will allow comparisons to be made with other units, or in the same unit over time, e.g. the National Sentinel Audit of Stroke.71 The results should identify where the problems lie.134 Also, by setting challenging targets one could use this approach to drive forward improvements in services. • Collect any outcome data that purchasers or commissioners of the stroke service require – but these should be kept simple, to minimize the cost and because they are likely to be easier to interpret. However, any interpretation will require more complex data about the patients who the service treats. The case mix and outcome data together could form the basis for a minimum data set; our suggestions are outlined in Table 17.21 and include those variables that we have found useful in predicting outcomes after stroke (sec-

Table 17.21 Minimum data set for stroke. This reflects known factors that influence the outcome after stroke and also data that may be routinely available to health service workers. Case mix data Age Sex Marital status, or living alone before the stroke Pre-stroke function, i.e. was the patient independent in activities of daily living? Stroke severity: Conscious level (normal or reduced?) Can the patient talk? Is the patient orientated in time, place and person? Severity of weakness – can he or she lift an affected limb against gravity? Can he or she walk independently? Urinary incontinence during first week after the stroke? Process Number of physicians responsible for stroke care–a lower number suggests more specialization Proportion of patients discussed by a multidisciplinary team Proportion of patients managed on a stroke unit Proportion of patients having CT or magnetic resonance scans Proportion of patients with ischaemic stroke on aspirin at discharge Proportion of patients with ischaemic stroke and atrial fibrillation on anticoagulants at discharge Carotid imaging for appropriate patients Delays to accessing the above items Outcomes Survival at 30 days Independence/dependence in activities of daily living at 6 months Complications, e.g. pressure ulcers Place of residence at 6 months

tion 10.2.7). These items include some demographic data, which are usually collected routinely (e.g. age, marital status) and which are likely to relate to outcome. Pre-stroke function, which will also reflect comorbidity, relates closely to functional outcome, and, to a lesser extent, increases the risk of death. Other possible indicators of stroke severity are level of consciousness, urinary incontinence, severity of motor weakness, or the proportion of patients with total anterior circulation syndromes (section 10.2). Poor outcomes may at best reflect poor care, but do not – unlike audits of process – identify the areas in which the service needs to be improved. • Monitor the frequency of complications after stroke. Although this is unlikely to provide any quantitative information about the quality of care, the data might be used to identify problems. For example, a high or rising proportion of patients with pressure sores may indicate inadequate numbers of nurses, poor-quality nursing care or delays in discharging dependent

..

..

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 943

17.13 Stroke guidelines

patients. Unfortunately, there are considerable problems in defining complications and in providing reliable diagnostic criteria to allow monitoring.22,166,167 A system of critical incident recording might focus on these and other problems and provide a simple indicator that the service may be performing poorly. • Provide an environment in which the identification of problems is encouraged and not penalized. One suspects that informal judgements that ‘there may be a problem with quality’ made by those working within a service, or using a service, are as sensitive a method of identifying major problems as any of those discussed above. This approach will only be effective if those working within a service are objective and honest and if the system does not punish healthcare workers, but encourages everyone constantly to provide better services. • Develop a system to provide an external and independent review of services by professionals from another centre. This can usefully identify problems that need to be resolved. It is essential that politicians and health service managers understand the difficulties in interpreting measures of process and outcome, and their limitations. They must not make important decisions about the distribution of resources on the basis of simplistic analyses of crude data. However, large discrepancies in the apparent performance of different stroke services should trigger a detailed enquiry into the possible explanation. Also, it is important not to use the results of a non-randomized evaluation of a local stroke service to guide service development elsewhere.

17.13 Stroke guidelines

In recent years, huge numbers of guidelines have been written about how to manage stroke patients.19–21,168 These are intended to describe best practice in the most common clinical situations, and not necessarily how every individual patient should be treated. They provide useful standards against which at least the process of care can be monitored. In the past, guidelines have been based on incomplete and thus potentially biased and misleading assessments of the evidence for the effectiveness of interventions. However, the methodology for guideline development is becoming increasingly rigorous, with recommendations being based on systematic reviews of the literature. There are standards against which one can assess the quality of a clinical guideline

..

..

Table 17.22 Factors associated with a greater chance of improving practice in accordance with guidelines. Consistently effective interventions Educational outreach visits (for prescribing in North America) Patient-specific reminders (manual or computerized) Multifaceted interventions (a combination that includes two or more of the following: audit and feedback, reminders, local consensus processes, or marketing) Interactive educational meetings (participation of healthcare providers in workshops that include discussion or practice) Interventions of uncertain effectiveness Audit and feedback (or any summary of clinical performance) The use of local opinion leaders (practitioners identified by their colleagues as influential) Local consensus processes (inclusion of participating practitioners in discussions to ensure they agree that the chosen clinical problem is important and the approach to managing the problem is appropriate) Patient-mediated interventions (any intervention aimed at changing the performance of healthcare providers for which specific information was sought from or given to patients) Interventions that have little or no effect Educational materials (distribution of recommendations for clinical care, including clinical practice guidelines, audio-visual materials and electronic publications) Educational meetings (such as lectures) Adapted from Davis et al., 2006.192

(Scottish Intercollegiate Guidelines Network No. 50; http://www.show.scot.nhs.uk/sign/home.htm). If guidelines are to improve clinical practice, it is important that they are effectively implemented. Several factors have been identified (Table 17.22) that may improve practice in accordance with guidelines.169,192 When one has to decide whether to implement clinical guidelines, it is important to assess the methodological rigour with which they were developed. Failure to do so could lead to the adoption of ineffective or even harmful practices.

17.14 Integrated care pathways

Clearly, a multifaceted approach must be taken to ensure that stroke patients are managed in line with guidelines unless there is a valid reason to deviate from them in an individual patient. One approach that is increasingly used in stroke units is the integrated care pathway, which often facilitates local activities that are very likely

943

9781405127660_4_017.qxd 10/13/07 2:15 PM Page 944

944

Chapter 17 The organization of stroke services

to increase adherence to guidelines (for an example, see http://www.dcn.ed.ac.uk/spgm). The development of integrated care pathways usually requires local scrutiny of available guidelines and discussion of how patients should be managed, and introducing them involves local educational sessions. They usually incorporate patient-specific reminders, i.e. guidance on how to manage common problems. They are often a focus of local audits, which measure the degree of compliance with the pathways and the reasons for any deviations, and feed these back to the staff in the unit. There is no robust evidence from randomized trials that integrated care pathways lead to better patient outcomes, although they may at least improve documentation.170,171

17.15 Impact of a comprehensive stroke service

When making the case for establishing and maintaining a comprehensive stroke service, it is worth considering the potential impact that such changes may bring. The following arguments justify establishing and maintaining a comprehensive stroke service: • Clinical standards and guidelines. Many countries now have standards or guidelines that require service

components to be in place (e.g. stroke units) – see Table 17.7. • Quality of care. Many clinical standards cannot be adequately met without having a coordinated service in place (e.g. coordinated multidisciplinary team care). • Patient outcomes. Table 17.23 outlines an estimate of the potential impact of a comprehensive stroke service using data from systematic reviews of randomized trials. The figures are approximate, and assume that your patient population is similar to those in the clinical trials. However, they do reveal the potential impact in clinical outcomes. The stroke unit component of these estimates is also supported by epidemiological observations that implementing stroke unit care improves stroke recovery.71,87,91 • Cost-effectiveness. Research into the cost-effectiveness of stroke services indicates that not only may they be cost-neutral but they may actually save money to society in general. For this reason it is worth considering in more detail the cost-effectiveness of stroke services.

17.16 Cost-effectiveness of stroke services

Because of its high frequency in most populations, the resulting severe disability and the need for prolonged institutional care, stroke places a very considerable

Table 17.23 Potential impact of different components of a comprehensive stroke service (based on a hypothetical population of 1000 cerebrovascular patients; 300 with transient ischaemic attack (TIA) and 700 with stroke). Component

Absolute gain per 1000 patients treated

Survivors

Thrombolysis (rt-PA within 3 h) Very rapid access TIA service Comprehensive stroke unit Early supported discharge service Therapy-based rehabilitation service Total

Independent survivors

Proportion of the population eligible for treatment

Actual health gain in the stroke population (per 1000 patients) Survivors

Independent survivors

0

100

10%

0

7

(1)

(40)

30%

(90%), the distribution of pathological types was also similar to that in white high-income populations. There is a lack of population-based data on the incidence of stroke and its pathological types from large areas of the world, particularly developing countries (Fig. 18.5). Thus, while it has been suggested that the proportion of strokes due to intracerebral haemorrhage is higher in China, India and Africa than it is in mainly white populations in high-income countries, this could be due to methodological artifact, since few incidence studies with adequate brain imaging have been performed in developing countries, and those that have

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 959

18.2 Measuring – and monitoring changes in – the impact of stroke

been done were generally hospital-based and so prone to hospital admission bias.24,45 Where it has been studied reliably (mainly in highincome white populations), stroke incidence does not vary as much between countries as stroke mortality. Incidence appears higher in Japan, Russia, Bulgaria and Ukraine, but reliable stroke incidence studies are simply not available from many parts of the world, including China, India and Africa. Ethnic variation Variation in stroke incidence between countries, if real rather than due to methodological artifact, could be due to variations in environmental factors operating at the level of the population (e.g. climate, air pollution, environmental tobacco smoke) or the individual (e.g. smoking, blood pressure, diet), or differences in stroke risk between different ethnic groups, which may or may not be due to genetic factors. Rather few population-based stroke incidence studies have been able to compare stroke incidence between different ethnic groups within the same population, so reducing the population level environmental variation. Two population-based stroke registers in US cities and one in south London in the UK have found the age- and sex-adjusted incidence of stroke in black people to be about twice that in white people.46–48 One of these – the south London study – fulfilled proposed methodological criteria for an ideal stroke incidence study, and found that the higher stroke incidence in black people persisted after adjusting for socioeconomic group.48 Much of the remaining difference could be due to ethnic differences in vascular risk factors, and population-based studies in south London and in the US have shown that such differences exist.49–51 The south London study also found a slightly higher proportion of intracerebral haemorrhage and of lacunar ischaemic stroke among black compared with white stroke patients. In addition, it found survival after stroke to be better among black than white patients, even after adjusting for known confounders.52 Although both a higher stroke incidence and better survival might be explained by better case-finding (and so identification of more mild cases) among blacks (at least in the UK, where black people may be more likely to use the free and more easily monitored National Health Service), this seems unlikely to explain all of the difference. Further work is needed to explore the reasons for these intriguing ethnic differences in stroke incidence and survival.53 A recent ‘ideal’ population-based study in Auckland, New Zealand, compared the incidence of stroke and its

..

..

main pathological types among those of European origin, Maori/Pacific people and Asian and other ethnicities. The age-adjusted incidence of all types of stroke was almost twice as high in Maori/Pacific people and about one-third higher in Asian and other ethnicities than in European New Zealanders. This was accounted for by a higher incidence both of ischaemic stroke (relative risk 1.7 for Maori/Pacific people and 1.3 for Asian/other ethnicities) and intracerebral haemorrhage (relative risk 2.7 for Maori/Pacific people and 2.3 for Asian/ other ethnicities), with no significant difference for either subarachnoid haemorrhage or strokes of undetermined type. Chance, and ethnic differences in vascular risk factors and socioeconomic status, could be at least partly responsible for the differences observed, and further work is needed to establish the true explanation.54 Few studies have reliably compared stroke incidence between different ethnic groups in the same population. Those that have show a higher incidence in non-white, and generally lower income, ethnic groups than in whites, but it is uncertain whether these differences are due to different socioeconomic status, vascular risk factor patterns, genetic factors, or a combination of these. Time trends Rather few stroke incidence studies have been repeated, or continued for long enough, to provide reliable time trend information. Of the few population-based studies that have assessed stroke incidence trends in a given population, most showed a decline in stroke incidence through to the late 1970s or early 1980s. Thereafter, from the 1980s to the late 1990s, stroke incidence has stabilized or even increased slightly in some studies, while in others the decline in incidence has continued. Once again, the studies of incidence trends were mainly in white high-income countries, with a relative lack of information from other parts of the world.26,28–30,55–60 The most dramatic results came from Oxfordshire in the UK where there was a reduction between 1981 and 2004 in age- and sex-adjusted stroke incidence of almost one-third. Unlike most previous studies of stroke incidence trends, data were available on trends for the main pathological types of stroke. The overall decline was due to a decrease of about one-half in intracerebral haemorrhage, and of about one-quarter in ischaemic stroke, with no change in subarachnoid haemorrhage.30 This latter finding extended the findings of a previous systematic review, which suggested that the incidence of aneurysmal subarachnoid haemorrhage remained stable between 1960 and the early 1990s.61

959

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 960

960

Chapter 18 Reducing the impact of stroke and improving the public health

So, if stroke incidence is falling, at least in affluent areas of the world, why is this? In the Oxfordshire study, there were also substantial reductions in premorbid vascular risk factors, with major increases in premorbid treatment with vascular preventive drugs, suggesting that improvements in primary preventive strategies could be at least partly responsible for the observed decline in stroke incidence.30 In the Auckland study, where the decline in stroke incidence was much less dramatic, changes in vascular risk factors were difficult to interpret, since the frequency of some fell while others increased.60 In the WHO-MONICA study, repeated population surveys of vascular risk factors and continuous monitoring of stroke events were conducted in people up to the age of 64 over a period of 7–13 years in 15 populations in nine countries. Variations in stroke trends between populations could be partly – but not wholly – explained by changes in known vascular risk factors.62 The identification and treatment of transient ischaemic attacks cannot have had much impact on stroke incidence, largely because only a small proportion of strokes are preceded by TIAs30,63 (section 18.4.1). Of course, because most populations are ageing, the number of stroke patients may increase even if the incidence in each age band decreases. Using current estimates of stroke incidence and population projections based on demographic information, calculations show that the number of strokes in all European Union and European Free Trade Association countries will increase from 1.1 million per year in 2000 to 1.5 million per year in 2025 without any change in stroke incidence for each age group. If the age-adjusted incidence increases by 2% per 5 years, then there will be 1.65 million strokes by 2025, and even if incidence falls by 2% per 5 years, the absolute number of stroke events will still increase to 1.35 million.64 However, the number of dependent stroke survivors will not necessarily increase as much because elderly dependent people are more likely to die early as a consequence of stroke than young independent people.65 We need far more dramatic falls in stroke incidence (so far only recorded in the Oxfordshire, UK study) if absolute numbers of strokes are to fall because of the increasing numbers of older people in the population. In general, therefore, as we saw for mortality and incidence variation between different countries, the dramatic decline in mortality over time seen in many parts of the world has not been accompanied by as steep a decline in incidence, and therefore cannot be entirely explained by falling incidence (section 18.2.1). However, the Oxfordshire stroke incidence data are very encouraging, and do suggest that primary prevention efforts may be starting to pay off.30

In the few places where it has been measured reasonably reliably, stroke incidence mainly declined up to the early 1980s. Thereafter, incidence has stayed about the same or even increased slightly in some places, while in others the downward trend has continued, perhaps due to improved primary prevention. We must now consider the four strategies which might reduce stroke burden in the future, whether such strategies could be implemented, and how the effects, if any, should be monitored.

18.3 Likely effect and costs of acute stroke treatment on stroke burden in the population

18.3.1 Effects Enormous efforts are being devoted to developing and testing promising new treatments for acute stroke and at long last there has been some progress, at least for ischaemic stroke (Chapter 12). Also, reorganizing haphazard stroke care into stroke units is not just an effective strategy to reduce case fatality and dependency, but greatly facilitates randomized trials of new treatments (section 17.6.2). But what impact might the present treatment of acute stroke be having on the overall burden of stroke? To assess this, we have used previously described methods combined with the most recent available data.66 Only three interventions have been shown in randomized trials to reduce either death or the combined outcome of death or dependency after acute ischaemic stroke or intracerebral haemorrhage: organized stroke unit care, aspirin given within 48 h of onset of acute ischaemic stroke, and intravenous thrombolysis given with the first few hours of onset of acute ischaemic stroke. Clearly, treating acute stroke will have no effect on stroke incidence, except that some patients otherwise destined to have a stroke might only have a transient ischaemic attack if they are successfully treated with intravenous thrombolysis within a few hours of onset. Acute treatments for subarachnoid haemorrhage are not considered here, since this accounts for only a small proportion of all stroke patients. If we imagine a notional European population of 1 million people then, using incidence estimates from the recently conducted Oxford Vascular Study in the UK, we might expect 1600 people to have a first-ever-in-a-lifetime

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 961

18.4 Likely effects and costs of long-term treatment after transient ischaemic attack Table 18.1 Estimated number of cerebrovascular events occurring each year in a notional European population of 1 million people.* Event type

Incident

Recurrent

Transient ischaemic attack Ischaemic stroke Intracerebral haemorrhage Subarachnoid haemorrhage Any stroke

660 1410 120 70 1600

440 600 30 30 660

*Data are taken from the Oxford Vascular Study, a recent population-based study of the incidence of acute vascular events, including stroke and transient ischaemic attack (Rothwell et al., 200570).

stroke and 660 a recurrent stroke each year (Table 18.1). If we assume that the untreated outcomes in terms of death or death and dependency of these 2260 stroke patients are similar to those observed in the Oxfordshire Community Stroke Project, then by 3 months after stroke we would expect 520 (23%) to have died and 1240 (55%) to be either dead or dependent, while at 1 year we would expect 700 (31%) to have died but the same total number (i.e. 1240) to be dead or dependent (see footnotes to Table 18.2, Table 10.3 and Figs 10.4 and 10.6). If, very optimistically, we assume that 80% of these 2260 stroke patients can be admitted to – and receive organized care in – a stroke unit, then the number of stroke deaths avoided at 1 year will be 108 out of 700, i.e. 15% (Table 18.2a). The effect of giving early aspirin to the 85% of the patients whose stroke is ischaemic and who have no contraindication to or intolerance of aspirin would be to reduce stroke deaths at 3 months by 15 out of 520, i.e. 3% (Table 18.2a). The number of patients who are either dead or dependent from 3 months to 1 year could be reduced by 24 out of 1240 (2%) with optimum use of intravenous thrombolysis (giving i.v. thrombolyis within 3 h of onset to an optimistic 10% of all acute ischaemic stroke patients), by 29 (2%) with optimum use of aspirin for acute ischaemic stroke, and by 72 (6%) with optimal provision of stroke unit care (Table 18.2b). Thus, the impact of treatment for acute stroke on the number of stroke deaths, or on the number of people with stroke-related death or dependency in the population, even making optimistic assumptions about treatment delivery, is only small. This is because some patients die before treatment can be started; all the available treatments are of limited effectiveness, and most have risks as well as benefits; and many treatments are contraindicated in various types of patients.

..

..

The treatment of acute stroke cannot influence stroke incidence, and will have rather little impact on strokerelated morbidity and mortality in the population, either because the effect of treatment is weak (like aspirin) or because treatment is only suitable for a small proportion of patients (like thrombolysis).

18.3.2 Costs Detailed cost-effectiveness analyses can give healthcare providers very valuable information on the relative health gains from various interventions or strategies. However, such detailed analyses are beyond the scope of this chapter. Table 18.3 shows the direct drug costs (for the individual, the population, and per death or death or dependency avoided annually) of the acute stroke drug treatments discussed above, and demonstrates that the current cost per poor outcome avoided is far less for aspirin than it is for intravenous thrombolysis. Costs have been taken from the British National Formulary (see http://www.bnf.org/bnf/bnf/current/index.htm).67 The costs of stroke unit care may be slightly greater or slightly less than the alternative of routine general medical care, and so we have assumed stroke units to be cost-neutral. Since they reduce death and disability, however, they are highly likely to be cost-effective.

18.4 Likely effects and costs of long-term treatment after transient ischaemic attack and stroke on stroke occurrence: secondary stroke prevention

18.4.1 Effects Transient ischaemic attack (TIA) and stroke patients are at high risk of subsequent stroke (section 16.2). Of course, TIAs are essentially ischaemic strokes which, by definition, have recovered within 24 h and so treating TIAs is really secondary stroke prevention. If the very early high-risk period is included, then on average about 20% of patients with a TIA or (mainly ischaemic) stroke will have a recurrent stroke within 1 year, and then about 5% per year thereafter (section 16.2 and Table 16.1). For many of these individual patients, a combination of secondary preventive interventions (including lifestyle changes such as quitting smoking, antiplatelet, cholesterol-lowering and blood-pressure-lowering drugs and, in some cases, carotid endarterectomy) (chapter 16) can and should be used to reduce risk of recurrent stroke

961

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 962

962

Chapter 18 Reducing the impact of stroke and improving the public health Table 18.2 The estimated effect on the population of interventions for acute stroke on (a) death and (b) death or dependency in a notional European population of 1 million people and 2260 patients with an incident or recurrent acute stroke each year. (a) Death Intervention

Stroke unit* Aspirin†

Relative % dead % dead with risk without intervention reduction intervention (%)

Absolute risk reduction (%)

Number of deaths avoided per 1000 treated

Number of patients to be treated to prevent one death

Number in target population (% of all 2260 strokes)

Number (%) of all stroke deaths avoided in population of 1 million§

18 6

31 13

25 12.2

6 0.8

60 8

17 125

1808 (80%) 1910 (85%)

108/700 (15%) 15/520 (3%)

(b) Death or dependency Intervention Relative risk reduction (%)

% dead/ dependent without intervention

% dead/ dependent with intervention

Absolute risk reduction (%)

Number of deaths/ dependents avoided per 1000 treated

Number of patients to be treated to prevent one death/ dependency

Number in target population (% of all 2260 strokes)

Number (%) of all stroke deaths/ dependents avoided in population of 1 million§

Stroke unit* 8 Aspirin† 3 Thrombolysis‡ 17

55 50 60.3

51 48.5 49.7

4 1.5 10.6

40 15 106

25 67 9

1808 (80%) 1910 (85%) 226 (10%)

72/1240 (6%) 29/1240 (2%) 24/1240 (2%)

*Data on the relative effects of stroke units on death and death or dependency at 1 year are taken from the most recent Cochrane systematic review (Stroke Unit Trialists’ Collaboration, 2006, table 17.6). Estimates of 1-year outcomes without stroke unit care are taken from the Oxfordshire Community Stroke Project (OCSP), a population-based study of the incidence of and outcome after stroke in the 1980s (Chapter 10, Table 10.3 and Figs 10.4 and 10.6): % of all 2260 patients dead at 1 year = 31%, and dead/dependent at 1 year = 55%. †Data on the relative effects of aspirin on death and death or dependency at ~3 months are taken from the most recent Cochrane systematic review (Sandercock et al., 2003, section 12.3.2). Estimates of 3-month outcomes without aspirin are taken from the OCSP (Chapter 10, Table 10.3 and Figs 10.4 and 10.6): % of ischaemic stroke patients dead at 3 months = 13%, and dead/dependent at 3 months = 50%. ‡Data on the relative effects of thrombolysis on death or dependency at 3 months are taken from the most recent Cochrane systematic review, using data available for intravenous treatment within 3 h of onset (the time window for which treatment is currently licensed). The effects of treatment are likely to be overestimates, because they are dominated by one trial, in which there were problems with the randomization process (Wardlaw et al., 2003, section 12.5.2). Estimates of 3-month outcomes without thrombolysis are taken from the systematic review, since patients eligible for treatment with thrombolysis are not typical of ischaemic stroke patients in the community in general, making the randomized trial data more likely to be representative than that for unselected ischaemic stroke patients from the OCSP. §Estimates of outcomes for all 2260 stroke patients are taken from the OCSP (Chapter 10, Table 10.3 and Figs 10.4 and 10.6): number (%) of all 2260 stroke patients dead at 3 months = 520 (23%); dead or dependent at 3 months = 1240 (55%); dead at 1 year = 700 (31%); dead or dependent at 1 year = 1240 (55%).

(and of other vascular events). But what impact might these interventions have on stroke occurrence at the population level? We can calculate the approximate number of prevalent TIA and stroke patients in our notional European population of 1 million people by using available information on incidence and survival, as shown in Table 18.4. Our secondary preventive efforts will be targeted

at the 13 000 prevalent stroke and TIA patients in the population, of whom about 1000 (8%) each year will have a stroke that is potentially preventable by secondary preventive efforts. Table 18.4 shows how much we might expect to reduce strokes in the population by the optimal use of the various secondary preventive interventions whose benefits are proven by robust evidence from randomized trials

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 963

18.4 Likely effects and costs of long-term treatment after transient ischaemic attack Table 18.3 Costs of effective drug treatments for acute stroke.

Treatment

Cost per patient with acute stroke*

Aspirin for 2 weeks after acute ischaemic stroke I.V. thrombolysis for very early acute ischaemic stroke (rt-PA)

Annual cost of treating all eligible patients in population of 1 million†

Cost per poor outcome avoided‡ Death

Death or disability

£0.50

£955

£62.50

£33.50

£480

£108 480

–

£4320

*Costs are given in UK £ sterling and taken from the British National Formulary, March 2007.67 These are estimated drug costs only, and so are a minimum since we have not included the costs of administration, adverse effects, or monitoring. †Calculated using data on numbers of eligible patients given in Table 18.2. ‡Calculated using data on number-needed-to-treat given in Table 18.2.

(chapter 16), and again we have drawn on previously established methods and the most recent available data to make the necessary calculations.66 All the patients with an ischaemic stroke (around 90% of the strokes) and all those with a TIA should be considered for long-term antiplatelet treatment (section 16.5.1). If we make the optimistic assumption that, after excluding patients with a contraindication to aspirin and those with atrial fibrillation for whom oral anticoagulants are not contraindicated, we can treat around 85% of patients with an ischaemic stroke or TIA (i.e. about 77% of all prevalent stroke and TIA patients) with aspirin, then at best we could prevent 180 (8%) of the 2260 strokes occurring each year (Table 18.4). Assuming that we can treat around 85% of prevalent stroke and TIA patients with a statin to reduce cholesterol (those with an ischaemic event and with a pretreatment cholesterol of ≥ 3.5 mmol/L or LDL-cholesterol of ≥ 2.6 mmol/L), we could prevent 166 (7%) of the strokes occurring each year (Table 18.4). A total of perhaps around 20% of all the prevalent stroke and TIA patients will be ineligible for routine blood pressure reduction because they have had a subarachnoid haemorrhage (the secondary preventive blood-pressure-lowering randomized trials in patients with cerebrovascular disease included those with TIA, ischaemic stroke and intracerebral – but generally not subarachnoid – haemorrhage – section 16.3.2), already have a blood pressure below around 120–130/70, or cannot tolerate any blood-pressure-lowering medication (section 16.3.4). At the most optimistic, we might therefore treat 80% of the prevalent patients, and would expect, at best, to avoid 146 (6%) of the 2260 strokes occurring each year (Table 18.4). A small proportion of patients who have started aspirin will have a contraindication

..

..

to an additional antiplatelet drug, so we might expect to be able to treat about 65% of all prevalent stroke and TIA patients with modified-release dipyridamole in addition to aspirin, so avoiding about 94 (4%) of all strokes occurring annually (Table 18.4). For oral anticoagulants, potentially eligible patients are those with a prior ischaemic event who have atrial fibrillation or some other high-risk cardiac source of emboli. These probably comprise up to about 20% of prevalent ischaemic stroke/TIA patients, but an optimistic maximum of 50% of these (about 1000 patients) are likely to be free of contraindications to anticoagulation.68,69 Assuming we do treat these 1000 patients with oral anticoagulants, then we could avoid 73 (3%) of all the strokes occurring each year (Table 18.4). Finally, each year only a very small proportion of patients with a new ischaemic stroke or TIA will be eligible for carotid endarterectomy, since they must be fit enough and willing for surgical intervention and have severe recently symptomatic carotid stenosis. In the recent OXVASC community-based study of stroke and TIA in Oxfordshire, 95% of patients with a TIA or non-disabling ischaemic stroke (i.e. the majority who might be eligible for carotid endarterectomy) had carotid imaging. Over a 3-year period, 40 out of a total of 918 patients with an incident or recurrent stroke or TIA (4%) had severe carotid stenosis with relevant symptoms in the last 6 months.70,71 If we make the optimistic assumption that 80% of these are fit enough and willing to have surgery and can be operated on within 2 weeks of their symptom onset, then we could prevent an average of 4 (0.2%) of the 2260 strokes occurring each year in the population (Table 18.4). So, with optimistic assumptions, each of the secondary preventive treatments discussed above might reduce the

963

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 964

964

Chapter 18 Reducing the impact of stroke and improving the public health Table 18.4 The estimated effect on strokes occurring in the population of interventions for prevention of recurrent stroke (and other vascular events) among 13 000 stroke and TIA patients prevalent in a notional European population of 1 million people.* Intervention

Relative risk reduction (%)

% recurrent stroke per year without intervention

% recurrent stroke per year with intervention

Absolute risk reduction (%)

Number of strokes avoided per 1000 treated per year

Number of patients to be treated to prevent one stroke per year

Number in target population (% of all 13 000 prevalent TIA/stroke patients)

Number (%) of all 2260 strokes avoided each year in population of 1 million

Aspirin† Statins to reduce cholesterol‡ Drugs to reduce blood pressure§ Modified release dipyridamole added to aspirin¶ Anticoagulants¥ Carotid endarterectomy$

20 17

9.0 9.0

7.2 7.5

1.8 1.5

18 15

56 67

10 000 (77%) 11 050 (85%)

180/2260 (8%) 166/2260 (7%)

23

6.0

4.6

1.4

14

71

10 400 (80%)

146/2260 (6%)

22

4.8

3.7

1.1

11

91

8500 (65%)

94/2260 (4%)

61 59

12 6.0

4.7 2.5

7.3 3.5

73 35

14 29

1000 (8%) 117 (1%)

73/2260 (3%) 4/2260 (0.2%)

*The prevalence of stroke and TIA is assumed to be given by the incidence multiplied by the time (in years) taken for half of the incident cases to die. Every year there are 1600 incident strokes (Table 18.1), of whom about half will die within 5 years (section 16.2), so that the number of prevalent stroke patients in the population at any one point in time is about 8000. Each year, there are also 660 incident TIAs, of whom about half will die in 7–8 years (based on 5-year follow-up data from the Oxfordshire Community Stroke Project, adjusted for later increases in mortality as documented in long-term follow-up of a cohort of patients with a TIA or mild ischaemic stroke recruited into a randomized trial) (Dennis et al., 1990;104 van Wijk et al., 2005105), so that the number of prevalent TIA patients in the population at any one point in time is about 5000. Three thousand three hundred and sixty of the 13 000 prevalent patients have had an incident or recurrent stroke or TIA within the last year (see Table 18.1 above). The risk of recurrent stroke in the first year after a TIA or stroke is about 20% (section 16.2). We assume that, for all eligible stroke patients, aspirin and a statin are started immediately, while blood pressure reduction, dipyridamole and anticoagulants are started on average 1–2 weeks after the TIA or stroke. For these slightly later treatments, the ‘treatable’ risk in the first year decreases to about 10% (since we miss the very early high risk period). The annual stroke recurrence risk in the remaining 9640 patients is about 5% per year. For aspirin and statin treatment, this means that the overall average annual untreated risk for all 13 000 patients is about 9% (equivalent to 20% of 3360 + 5% of 9640), and for blood pressure reduction the overall average annual untreated risk for all 13 000 patients is about 6% (equivalent to 10% of 3360 + 5% of 9640). Since the average annual untreated risk from 1–2 weeks post event is about 6%, but the proven benefits of dipyridamole are in addition to those of aspirin, we assume that aspirin has already been started, reducing the risk without dipyridamole by 20% to 4.8%. For anticoagulants, mainly for patients with atrial fibrillation, the untreated average annual risk is estimated at about 12% (the control group risk in randomized trials of oral anticoagulants vs control for long-term prevention in patients with atrial fibrillation and a history of ischaemic stroke or TIA – section 16.6.2). †Relative risk reduction based on only the aspirin vs control trials in the Antithrombotic Trialists’ overview of antiplatelet therapy, assuming the effect on stroke among patients with prior ischaemic stroke or TIA to be similar to that found for all patients at high risk of vascular disease (Antithrombotic Trialists’ Collaboration, 2002106). ‡Relative risk reduction taken from systematic review of statins to reduce cholesterol, assuming the effect on stroke of a 1 mmol reduction in LDL cholesterol among patients with prior ischaemic stroke or TIA to be similar to that found overall (Cholesterol Treatment Trialists’ Collaboration, 200587). §Relative risk reduction for reduction in blood pressure of 8 mmHg systolic/4 mmHg diastolic taken from a systematic review of blood pressure lowering after prior stroke or TIA (Rashid et al., 2003107). ¶Relative risk reduction based on a meta-analysis of randomized trials of aspirin + dipyridamole vs aspirin alone in patients with a prior ischaemic stroke or TIA (The ESPRIT Study Group, 2006108). ¥Relative risk reduction based on a meta-analysis of randomized trials of warfarin vs control in patients with atrial fibrillation (section 16.6.2 and Fig. 16.30b). $Relative risk reduction based on meta-analysis of two large randomized trials of carotid endarterectomy for recently symptomatic severe stenosis. This is the reduction in any ipsilateral ischaemic stroke and any stroke or death within 30 days of surgery for patents operated on within 2 weeks of symptom onset. The relative and absolute risk reductions are assumed to remain constant over the 5-year follow-up period in order to obtain a notional average annual risk reduction. The average annual risk without intervention is that of the control group within the subgroup of patients operated on within 2 weeks of symptom onset in the meta-analysis (30% over 5 years, about 6% per year) (Rothwell et al., 2004109).

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 965

18.4 Likely effects and costs of long-term treatment after transient ischaemic attack

number of strokes occurring each year in the population by up to a few per cent. These effects may be additive but the figures shown do not, except in the case of adding dipyridamole to aspirin, account for decreasing increment in absolute benefit, since each new treatment is, in reality, given to patients whose baseline risk is lowered by taking whichever treatment(s) they have already started (section 16.18.2). We saw above that about 1000 (less than half) of the 2260 strokes occurring each year in our population of 1 million people occur in those with a prior stroke or TIA and so are potentially preventable by secondary preventive strategies (Table 18.4). Thus, although treatment for a relatively small number of high-risk patients who have already had a stroke or TIA will be reasonably effective for many (but not all) in terms of reduced stroke risk, it will have little impact on stroke in the population because most strokes occur in lower-risk individuals who have not already had a stroke or TIA (the so-called ‘prevention paradox’ – section 18.5.3). Furthermore, as Table 18.4 shows, treatment is always much less than 100% effective and not without risk, and each treatment is only appropriate for a variable proportion of all prevalent stroke and TIA patients. The optimistic assumptions used in Table 18.4 include

Table 18.5 Costs of interventions for long-term secondary prevention of stroke.

assuming that treatment adherence is as good as in the randomized trials and that the treatments are used for all eligible patients in a timely way. In real clinical practice, effective drugs and surgery are not applied optimally,72,73 and patients prescribed long-term medications only take about 50% of their pills.74 However, it is worth bearing in mind that the secondary preventive treatments discussed will also have a small impact on non-stroke vascular events (such as ischaemic heart disease) in addition to their modest effects on stroke at the population level. Prevention of stroke after a stroke or transient ischaemic attack is an example of the ‘high-risk’ prevention strategy and will have little effect on stroke incidence, even though the individual patient may have much to gain.

18.4.2 Costs Table 18.5 shows the direct costs of the various secondary preventive interventions listed in Table 18.4 and discussed in section 18.4.1 above. As in section 18.3.2, we have estimated drug costs from the British National Formulary,67 and these are a minimum because the costs of monitoring,

Intervention*

Cost per patient (per year)†

Annual cost of treating all eligible patients in population of 1 million‡

Cost per stroke prevented¶

Aspirin Simvastatin 40 mg daily Atorvastatin 10 mg daily Indapamide 2.5 mg daily and perindopril 4 mg daily Bendrofluazide 2.5 mg daily and lisinopril 10 mg daily Modified release dipyridamole added to aspirin Anticoagulants Carotid endarterectomy

£16 £44 £235 £32 + £133 = £164

£160 000 £486 200 £2 396 750 £1 705 600

£896 £2 948 £15 745 £11 644

£15 + £20 = £35

£364 000

£2 485

£118

£1 003 000

£10 738

£54§ £4 350¥

£54 000 £508 950

£756 £126 150

*Effects on blood pressure and cholesterol of the two alternative regimens for blood pressure and cholesterol reduction shown are assumed to be approximately equivalent (section 16.3.3 and Table 16.6). †Drug costs are given in UK £ sterling and taken from the British National Formulary, March 2007.67 These are estimated drug costs only, and so are a minimum since we have not included the costs of administration, adverse effects, or monitoring. ‡Calculated using data on numbers of eligible patients given in Table 18.4. ¶Calculated using data on number-needed-to-treat given in Table 18.4. §Allowing for any combination of 1, 3 and 5 mg tablets each day. ¥From Benade and Warlow, 2002,75 with cost inflation to 2006 (Wardlaw et al., 200676). Only the costs of the endarterectomy procedure are given here, and costs of investigative workup are not included.

..

..

965

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 966

966

Chapter 18 Reducing the impact of stroke and improving the public health

adverse effects and so on are not included. The cost of carotid endarterectomy (the procedure itself only) is taken from a recent UK-based prospective costing study and inflated to 2006.75,76 Thus, these are only approximate, UK-specific estimates, but nonetheless they give a fairly good idea of the relative costs per stroke avoided of the various different interventions. Aspirin and warfarin appear by far the cheapest per stroke avoided, although the non-inclusion of monitoring and adverse effects costs will lead to a substantial exaggeration of the apparent cost-effectiveness of warfarin here. For blood pressure and cholesterol reduction, Table 18.5 also shows the rather dramatic effects of choosing one specific drug regimen rather than another of approximately equivalent efficacy. It is clear that the average cost of carotid endarterectomy per stroke avoided is high. Quite apart from its risks, this is a good reason to consider restricting carotid endarterectomy to those patients at highest risk without surgery (sections 16.11.6, 16.11.7 and 16.11.8).

18.5 Primary stroke prevention: targeted high-risk strategy or population-level ‘mass’ strategy?

A recent assessment of the global burden of disease attributable to 20 selected leading, potentially modifiable

risk factors showed that, above certain theoretical minimum values, high blood pressure, high cholesterol, overweight, low fruit and vegetable intake, physical inactivity, smoking and alcohol consumption together account for over two-thirds of both disease burden (in terms of healthy life expectancy) and mortality due to stroke. Even higher proportions of disease burden and mortality from ischaemic heart disease could be explained by the same seven risk factors (Table 18.6).77 The implication is that, if we were able to reduce the level of these risk factors to the theoretical minimum in the entire population, then we could substantially reduce the population-wide burden of stroke, ischaemic heart disease and several other major diseases. Even though the reduction of all of these risk factors to their theoretical minimum seems unlikely to be possible, there is clearly great potential for reducing the global burden of stroke and other vascular diseases by intervening to reduce known risks. Public health epidemiologists generally agree that the strategies of identifying and treating those at highest risk in the population, as well as adopting population-wide strategies to reduce the average level of risk factors in the whole population (the third and fourth strategies outlined in section 18.1 above) are complementary, and that both will be necessary if we are to effectively reduce the burden of stroke and other vascular diseases through altering the known important risk factors.78–80 However, debate continues about the relative acceptability, feasibility, effectiveness and cost-effectiveness of the two approaches.

Table 18.6 Contributions to global disease burden* and mortality due to stroke and ischaemic heart disease of seven major risk factors (data from Ezzati et al., 200377). Risk factor

High blood pressure High cholesterol High BMI Low fruit and vegetable intake Physical inactivity

Smoking Alcohol

Theoretical minimum

Systolic 115 mmHg 3.8 mmol/L 21 kg/m2 600 g/day for adults 2.5 hours per week of moderateintensity activity None None

Individual risk factor contributions to:†

Joint risk factor contributions to:‡

Stroke disease burden*

IHD disease burden*

Stroke disease burden*

62% 18% 13% 11% 7%

49% 56% 21% 31% 22%

12% 4%

12% 2%

70–76%

Stroke mortality

IHD disease burden*

IHD mortality

65–73%

83–89%

78–85%

*Expressed in disability-adjusted life years (DALYs). BMI, body mass index; IHD, ischaemic heart disease. †Per cent reduction in disease burden that would occur if exposure to risk factor reduced to theoretical minimum. ‡Per cent reduction in disease burden or mortality that would occur if exposure to all seven risk factors reduced to theoretical minimum.

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 967

18.5 Primary stroke prevention: targeted high-risk strategy or population-level ‘mass’ strategy?

18.5.1 The targeted high-risk strategy Opportunistic identification or widespread systematic screening? It seems intuitively appropriate to identify for specific risk-reducing interventions those individuals who have not yet had symptoms of vascular disease, but are at highest risk. However, it is not universally agreed whether such individuals should be identified through opportunistic identification and/or through widespread, systematic screening of apparently healthy people. The latter will depend both on a combination of an individual’s interpretation of any symptoms, background knowledge, and healthcare-seeking behaviour, as well as on the quality, availability and accessibility of healthcare services, and the attitudes and behaviour of healthcare professionals. Both of these in turn will depend on cultural, geographic and socioeconomic factors. Opportunistic identification of people considered to be at increased risk of vascular disease may occur because: they are being assessed medically for some other reason (e.g. identification of high blood pressure or asymptomatic diabetes when registering with a new family doctor, undergoing preoperative assessment prior to surgery, or having an occupational health check); they choose to attend a healthcare professional for a health check; or they develop or have a risk factor which can cause symptoms leading to their presentation to healthcare services (e.g. atrial fibrillation presenting with palpitations or breathlessness, diabetes presenting with thirst, polyuria and weight loss, and so on). To justify a widespread systematic screening programme, there should be evidence that the screening procedure identifies people who genuinely have the risk factor or risk factors in question, that there is a possibility of intervening to reduce risk, that such intervention is of net benefit to individuals, that appropriate resources are available to provide the intervention, and that the whole screening programme (including, importantly, the costs of treating those identified) is cost-effective. The psychological, behavioural and quality-of-life impact of identifying, labelling and treating apparently healthy, previously asymptomatic people as being at high risk needs to be carefully considered and weighed against the potential benefits.81 These considerations must surely

..

..

become increasingly important as the level of risk or of a particular risk factor leading to identification, labelling and treatment is reduced. The targeted high-risk strategy for prevention of stroke (and other vascular diseases) requires the identification and treatment of those at high risk. The potential consequences of ‘medicalizing’ apparently healthy, asymptomatic people need to be weighed against any potential benefits of treatment. Single risk factors vs risk scores to estimate overall absolute risk A further crucial point is that identifying people on the basis of single risk factors will not reliably identify those at highest risk. It is combinations of risk factors that are important in determining overall risk. The concept that a large number of people at low and moderate risk may give rise to more cases of disease than a small number of people at high risk, described by the British epidemiologist Geoffrey Rose and illustrated in Fig. 18.6,82 occurs particularly when single risk factors (or poorly predictive risk scores or models) are used to allocate people to lowor high-risk categories,78,83 and also depends on the level of the risk factor (and so associated average risk at levels above and below it) above which risk is considered to be high. As an example, Table 18.7 shows the expected number of cases of stroke arising in a notional population of 10 000 men aged 40–49 over a 24-year period for different levels of systolic blood pressure.84 Over half of all the strokes occur in the 72% of men with systolic blood pressures between 120 and 159 mmHg. If the level for high blood pressure and high risk is set at 180 mmHg, then only 22% of cases of stroke arise in those at high risk as defined by blood pressure. But, if high blood pressure is considered to be > 140 mmHg, then over two-thirds (68%) of stroke cases arise in those at high risk. Number of events (i.e. patients with the disease of interest)

Seven potentially modifiable risk factors (high blood pressure, high cholesterol, overweight, low fruit and vegetable intake, physical inactivity, smoking and alcohol consumption) together account for over twothirds of morbidity and mortality due to stroke and ischaemic heart disease.

Increasing risk of the disease of interest in the population Fig. 18.6 Most cases of a disease occur in people at moderate risk; rather few (shaded) occur in people at the very highest risk in the right-hand tail of a normal distribution.

967

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 968

968

Chapter 18 Reducing the impact of stroke and improving the public health

Systolic blood pressure (mmHg)

Prevalence (%)

Incidence of stroke per 100 men during 24 years follow-up

Expected number of cases of stroke occurring over 24 years in a population of 10 000 men aged 40–49 at baseline (% of total)*

< 120 120–139 140–159 160–179 180+ Total

16 43 29 8 4 100

1.7 3.6 5.6 12.8 31.8

27 (4.7) 155 (27) 162 (28.3) 102 (17.8) 127 (22.2) 573 (100)

Table 18.7 Systolic blood pressure and the risk of stroke in men aged 40–49 years. The prevalence of various levels of blood pressure and the incidence and expected number of cases of stroke arising in each blood pressure stratum over 24 years, using Framingham Study data (from Table 2.3 in Ebrahim and Harwood, 199984).

*Calculated by multiplying column 2 by column 3.

Specific levels of risk factors (e.g. of blood pressure and cholesterol) are of little clinical relevance when considered in isolation from a person’s overall estimated risk of future vascular disease. For practically any risk factor that can be measured as a continuous variable (blood pressure and cholesterol are good examples), risk of vascular disease increases continuously with increase in the level of the risk factor within the range of values observed in most populations (section 6.6.3), so the notion that some people have or do not have, say, hypertension or hypercholesterolaemia is an artificial one based on arbitrary cut-off values.85 Indeed, since an individual’s absolute risk of a future vascular event is determined by the combined effect of all vascular risk factors present, differences in risk can vary more than 20-fold in patients with the same blood pressure or cholesterol levels.86 Figure 18.7 shows how risk of future vascular disease is increased by successively adding risk factors in individuals with different systolic blood pressure (Fig. 18.7a) or blood cholesterol concentration (Fig. 18.7b), and how someone with a blood pressure or cholesterol concentration at the lower end of the spectrum and several other risk factors may have a higher overall risk than someone with a much higher blood pressure or cholesterol concentration but no other risk factors.86 There is now increasing evidence from observational studies and randomized trials, in particular of blood pressure and cholesterol lowering, both in primary and secondary prevention settings, that we should be thinking less in terms of reducing levels of these risk factors down to specific targets in individuals whose risk factor level is above a particular arbitrary cut-off value (these are still the focus of many treatment guidelines), but more in terms of reducing these risk factors as much as possible (at least down to the lower end of the ranges studied in the trials and observational studies) in those at highest absolute risk of future vascular events.85,87,88 Both observational and randomized trial data suggest that the greater the absolute amount by which either blood pressure or

cholesterol are lowered, the greater the relative benefit on clinically meaningful outcomes (sections 16.3 and 16.4).85 Several risk prediction scores, based on models derived from population-based cohort studies (mainly the Framingham cohort study) in which risk factors were measured and subjects followed up over many years for the subsequent occurrence of vascular events, are now widely available for estimating individual risk, mainly in primary care settings. These scores are presented for use as paper or hand-held calculator-based equations, colour charts, or online calculators, and allow clinicians to integrate information on a number of different recognized risk factors (including age, gender, blood pressure, cholesterol, atrial fibrillation, diabetes, obesity, smoking and family history) to estimate an individual’s risk of various vascular events over the next 5–10 years.86,89 Integrated risk factor scoring systems such as these seem intuitively likely to be useful, certainly better than single risk factor identification methods, and they are increasingly used in risk factor treatment guidelines and in primary care settings, but they are not without their limitations: • The currently available scores tend to be focused on ischaemic heart disease, although a few either include stroke or are specific for stroke86,89 (see http://www. bhsoc.org/Cardiovascular_Risk_Charts_and_Calculato rs.stm, http://www.riskscore.org.uk/). In patients without a history of a vascular event, those that estimate together the combined risk of all potentially important future vascular events (including stroke as well as ischaemic heart disease) are likely to be the most helpful in identifying people at the highest overall vascular risk. • The predictive accuracy of the existing risk scores is suboptimal. Validation studies of Framingham-based risk scores suggest that vascular risk in low-risk populations is likely to be overestimated, potentially resulting in unnecessary treatment of many people, while that in high-risk populations is likely to be underestimated.89

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 969

18.5 Primary stroke prevention: targeted high-risk strategy or population-level ‘mass’ strategy?

(a) 50 Five-year CVD risk per 100 people

45

44%

40 35

33%

30 25

24%

20

18%

15

12%

10 6%

5

3% 180 Total

public health is inevitably to do with the Department of Health. These interventions are no more invasive or prescriptive than providing a clean water supply or mass immunization, both of which are clearly in the domain of public health services. But, one has to accept the ‘prevention paradox’ that a preventive strategy that brings large benefits to the community offers little to each participating individual.82 The mass strategy for prevention of stroke and other vascular diseases requires a small leftward shift in the entire distribution of relevant risk factors such as blood pressure. This may have a surprisingly large impact on the incidence of stroke and other vascular diseases although each individual will gain very little.

18.5.4 Potential impact of the mass population strategy compared with the targeted high-risk strategy The mass population strategy cannot be evaluated by a randomized controlled trial, not because the outcome is unmeasurable or because the sample size would have to be too big or it would take too long to demonstrate an effect, but because the strategy requires everybody to be ‘treated’ and it would simply be impracticable to apply the ‘treatment’ to a random 50% of the population. But the effects and costs can be estimated using more or less sophisticated modelling techniques, and this allows theoretical comparisons to be made with the targeted high-risk strategy. For example, reducing the population mean systolic blood pressure by just 1–2% could reduce the incidence of stroke by about 16% (Table 18.10).84 Such blood pressure

Prevalence (%)

Incidence of stroke per 100 men during 24-year follow-up

Before blood pressure reduction

After blood pressure reduction

16 43 29 8 4 100

23 43 26 6 2

1.7 3.6 5.6 12.8 31.8

Expected number of cases of stroke occurring over 24 years in a population of 10 000 men aged 40–49 at baseline Before blood pressure reduction

After blood pressure reduction

27 155 162 102 127 573

39 155 146 77 64 481

i.e. a proportional reduction of (573–481)/573 = 16% in stroke numbers (and incidence).

..

..

973

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 974

Chapter 18 Reducing the impact of stroke and improving the public health

reductions should be achievable through legislation leading to population-wide reductions in salt intake of around 15%,79 and indeed substantially larger reductions in mean population blood pressure may be possible.80 Modelling exercises to compare the potential effects on the burden of stroke and vascular disease of the targeted high-risk and mass population strategies depend on the assumptions made about the effects of various interventions, the age, sex and risk factor profile of the population, and the thresholds set for intervention in the targeted high-risk strategy. Calculating and comparing cost-effectiveness involves further assumptions about the costs of these interventions. A complex – but elegant and informative – global analysis of the reduction in morbidity and mortality from ischaemic heart disease and stroke (expressed as disability-adjusted life years) for 14 World Health Organization epidemiological subregions of the world assessed and compared the effects and cost-effectiveness of various populationlevel (mass population strategy) and individual (targeted high-risk strategy) health interventions aimed at reducing the risks associated with raised cholesterol and blood pressure.79 The effect sizes of the various interventions were based on systematic reviews and meta-analyses of randomized trials and observational studies. The population-level interventions assessed were: • voluntary agreements or legislation on salt content to ensure labelling and stepwise decreases in salt content of processed foods (assumed to achieve reductions in dietary salt content of 15–30% and in blood pressure of 1–5% depending on age, sex and world subregion); • health education through the mass media (focusing on cholesterol concentration and body mass index and assumed to achieve a 2% reduction in blood cholesterol concentrations); • and a combination of these. The individual health interventions assessed were drugs to reduce blood pressure and cholesterol, and a combination of these with aspirin, similar to those assessed by others and described above (section 18.5.2) and shown in Table 18.8. For the individual health interventions (targeted high-risk strategy), different cholesterol and blood pressure threshold levels as well as various different absolute risk thresholds were evaluated. Combinations of the targeted high-risk and mass population strategies were also assessed.79 The results of this modelling exercise for one of the European subregions are shown graphically in Fig. 18.9. Unsurprisingly, the cost-effectiveness (i.e. the cost per DALY averted) was lowest for the population mass strategy interventions. However, perhaps rather counterintuitively, the potential to reduce the burden of disease was larger for the targeted high-risk approaches, albeit at

6000 Costs, 2000 int$ (millions)

974

P5

4000

P2

P9 C4

P4 P8 P3

2000

P1

C3

P7 P6

C2 C1

0

N1 N3 N4 N2

0

0.5 1 DALYs (millions)

1.5

Fig. 18.9 Yearly costs (2000 int$ = international dollars, base year 2000 – vertical axis) and effectiveness (DALYs, disabilityadjusted life years averted – horizontal axis) of various interventions to reduce the burden of stroke and ischaemic heart disease in the World Health Organization EurA subregion of Europe. (From Murray et al., 2003.79) Interventions N1 to N4 are population-level interventions to reduce salt intake through voluntary agreements with industry (N1) or legislation (N2), health education through the mass media focusing on cholesterol concentrations and body mass index (N3) and a combination of N2 and N3 (N4). Interventions P1 to P9 are individual interventions based on drugs to lower blood pressure at systolic blood pressure thresholds of 160 or 140 mmHg (P1 and P2), drugs to reduce cholesterol concentration thresholds of 6.2 and 5.7 mmol/L (P3 and P4), a combination of P2 and P3 (P5), and a combined drug treatment strategy (statin, diuretic, beta-blocker and aspirin) based on thresholds of absolute 10-year risk of ischaemic heart disease or stroke of 35% (P6), 25% (P7), 15% (P8) and 5% (P9). Interventions C1 to C4 are combined population and individual interventions (combination of N4 with P6 to P9). The slope of the line connecting the origin to each point is the cost per DALY averted. The steeper the slope, the more expensive the intervention is per DALY averted (less costeffective). The solid lines join the least expensive interventions for increasing effectiveness in terms of DALYs averted and show the best sequential choices of interventions to be made with increasing resources. Costs are reported in international dollars, a hypothetical currency where one dollar has the same purchasing power as one US dollar regardless of location.

a higher cost per DALY averted. This may relate – at least in part – to the effect size estimates. The population level effect sizes were assumed to be rather modest (which may be realistic) while the effects of cholesterol and blood pressure lowering were assumed to be the same as those achieved in randomized trials of cholesterol and blood-pressure-lowering drugs (which, given doubts about acceptability and adherence, is probably not realistic). However, moving between different world subregions and varying the range of values used for effects and

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 975

18.6 Goals, strategies and measuring success (or failure)

costs in an uncertainty analysis did not change the general pattern of the results. This modelling exercise suggests that, in parts of the world where resources are scarce, public health policy should focus on investing in the mass population strategies first, with additional resources used for individually targeted combination interventions for high-risk individuals on the basis of estimated absolute risk of vascular disease, shifting the threshold for intervention to lower levels of risk as more resources become available.79 Additional interventions not included in the modelling exercise above could of course enhance the mass population strategy. For example, small reductions in smoking population-wide could have a surprising impact on stroke burden and cost, as well as on other vascular diseases and a host of other common smoking-related diseases.101,102 A further argument in favour of the mass population strategy is that it focuses on the determinants of risk factor distributions rather than simply the treatment of risk factors. Population approaches may be more likely to reduce the development of vascular disease, while a high-risk strategy without priority given to population approaches would ensure a steady supply of middle-aged people requiring drug treatment.80 Common sense as well as modelling exercises suggest that the mass population strategy is almost certainly more cost-effective than the targeted high-risk approach. A combination of both probably achieves the best population health for the available resources. Are there any risks of the mass population strategy? Because moving the distribution of a risk factor to the left may increase the number of individuals at very low risk of the disease in question, one would need to be reasonably confident that these same individuals or society in general would not be systematically disadvantaged in some way, for example by an increased risk of depression or fatigue at very low blood pressures,103 of intracranial haemorrhage at very low cholesterol levels (section 6.6.6) or of hypothermia in the very thin and elderly. For the most part, however, the risk factor levels below which significant problems might possibly arise are beyond the values likely to be reached by risk factor reductions, certainly in societies adopting ‘western’ lifestyles.85 Because being simple goes against the grain of societies which believe in a quick technological or mystical fix for illness, or which may crave short-term results in the marketplace, and in societies which exalt the individual over the community, the mass population prevention strategy may be very difficult to implement. However,

..

..

the argument in favour of this strategy for not just stroke prevention, but for health promotion and the prevention of disease in general, is persuasive, if not compelling.82

18.6 Goals, strategies and measuring success (or failure)

Setting targets or goals in any human endeavour is surely reasonable provided they are not so easily achieved that no effort is required, nor so difficult that demoralization sets in. This certainly applies to stroke rehabilitation (section 10.3.3) and could be applied to reducing the burden of stroke and other vascular diseases in the whole community. The difficulty is deciding which target to select. It has to be at the level of the whole community, or a random sample of it, to avoid selection bias; it has to be simple so that it is measurable in very large numbers of people at reasonable cost; it has to be robust so that it can be measured reliably at various points in time; it must be valid and so measure something meaningful; and it almost certainly has to be measurable in routine practice to ensure complete coverage for the foreseeable future. At present, the only measure which comes anywhere near fulfilling these criteria is stroke mortality, despite its imperfections (section 18.2). However, one can reasonably hope, and even expect, that an intervention which reduces stroke incidence and disability would also have some impact on stroke mortality, although this impact could well be attenuated by the fact that not all types of stroke are fatal, mortality statistics are inaccurate, and so on. At present, mortality statistics are, despite their imperfections, the only routinely available method to measure the success or failure of stroke prevention programmes. In the meantime it is surely important to develop some routine method of measuring stroke incidence, as well as of outcome and survival, at the level of the whole community, not just in hospitalized patients. This must be practicable, and inexpensive, which will be difficult and take time to achieve. At the very least, one might think that routine measurement of stroke outcome is required to inform healthcare decisions, for how else can one system or provider of care be compared with another? But maybe even this is too fraught with difficulties to be useful, and measuring process of care is probably more meaningful (section 17.12.2).

975

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 976

976

Chapter 18 Reducing the impact of stroke and improving the public health

As to which stroke prevention strategy to choose, it seems sensible to pursue all of them while accepting that each one on its own is unlikely to have a dramatic effect. It would be impossible to deny an effective treatment to an individual if the benefits outweighed the risks, and the cost was acceptable, either to the individual or to a thirdparty payer. It would be inconceivable not to continue to search for and evaluate treatments for acute stroke. And it would be unforgivable to ignore the theoretical attractions of the ‘mass’ strategy for stroke prevention. Finally, almost any strategy for stroke prevention must be intimately linked with almost any strategy for the prevention of other vascular diseases (mainly coronary heart disease and peripheral vascular disease), because they are largely caused by the same underlying degenerative vascular disorders. So, stroke prevention should not be considered in isolation, but as part of a general strategy for the prevention of all forms of vascular disease, including stroke. Stroke prevention must not be seen in isolation, but as part of a programme to prevent the clinical manifestations of all forms of degenerative vascular disease.

References

1 Warlow CP. Epidemiology of stroke. Lancet 1998; 352 Suppl 3:SIII1–SIII4. 2 Rothwell PM. The high cost of not funding stroke research: a comparison with heart disease and cancer. Lancet 2001; 357(9268):1612–16. 3 Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet 2006; 367(9524):1747–57. 4 Martin J, Meltzer H, Elliot D. The Prevalence of Disability Among Adults. London: HMSO, 1988. 5 Giroud M, Lemesle M, Quantin C, Vourch M, Becker F, Milan C et al. A hospital-based and a population-based stroke registry yield different results: the experience in Dijon, France. Neuroepidemiology 1997; 16(1):15–21. 6 Di Carlo A, Candelise L, Gandolfo C, Grigoletto F, Volonnino G, Baldereschi M et al. Influence of different screening procedures on the stroke prevalence estimates: the Italian Longitudinal Study on Aging. Cerebrovasc Dis 1999; 9(4):231–7. 7 Levi F, Lucchini F, Negri E, La Vecchia C. Trends in mortality from cardiovascular and cerebrovascular diseases in Europe and other areas of the world. Heart 2002; 88(2):119–24.

8 Sarti C, Rastenyte D, Cepaitis Z, Tuomilehto J. International trends in mortality from stroke, 1968 to 1994. Stroke 2000; 31(7):1588–601. 9 Lawlor DA, Smith GD, Leon DA, Sterne JA, Ebrahim S. Secular trends in mortality by stroke subtype in the 20th century: a retrospective analysis. Lancet 2002; 360(9348):1818–23. 10 Halkes PH, van Gijn J, Kappelle LJ, Koudstaal PJ, Algra A. Classification of cause of death after stroke in clinical research. Stroke 2006; 37(6):1521–4. 11 Corwin LE, Wolf PA, Kannel WB, McNamara PM. Accuracy of death certification of stroke: the Framingham Study. Stroke 1982; 13(6):818–21. 12 Hasuo Y, Ueda K, Kiyohara Y, Wada J, Kawano H, Kato I et al. Accuracy of diagnosis on death certificates for underlying causes of death in a long-term autopsy-based population study in Hisayama, Japan; with special reference to cardiovascular diseases. J Clin Epidemiol 1989; 42(6):577–84. 13 Iso H, Jacobs DR, Jr, Goldman L. Accuracy of death certificate diagnosis of intracranial hemorrhage and nonhemorrhagic stroke. The Minnesota Heart Survey. Am J Epidemiol 1990; 132(5):993–8. 14 Netsky MG, Miyaji T. Prevalence of cerebral hemorrhage and thrombosis in Japan: study of the major causes of death. J Chronic Dis 1976; 29(11):711–21. 15 Phillips LH, Whisnant JP, Reagan TJ. Sudden death from stroke. Stroke 1977; 8(3):392–5. 16 Thomas AC, Knapman PA, Krikler DM, Davies MJ. Community study of the causes of ‘natural’ sudden death. Br Med J 1988; 297(6661):1453–6. 17 Bonita R, Beaglehole R. Explaining stroke mortality trends. Lancet 1993; 341(8859):1510–11. 18 Lanska DJ. Decline in autopsies for deaths attributed to cerebrovascular disease. Stroke 1993; 24(1):71–5. 19 Roulson J, Benbow EW, Hasleton PS. Discrepancies between clinical and autopsy diagnosis and the value of post mortem histology; a meta-analysis and review. Histopathology 2005; 47(6):551–9. 20 Hatano S. Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ 1976; 54(5):541–53. 21 Chandramohan D, Maude GH, Rodrigues LC, Hayes RJ. Verbal autopsies for adult deaths: issues in their development and validation. Int J Epidemiol 1994; 23(2):213–22. 22 Kaufman JS, Asuzu MC, Rotimi CN, Johnson OO, Owoaje EE, Cooper RS. The absence of adult mortality data for sub-Saharan Africa: a practical solution. Bull World Health Organ 1997; 75(5):389–95. 23 Walker RW, McLarty DG, Kitange HM, Whiting D, Masuki G, Mtasiwa DM et al. Stroke mortality in urban and rural Tanzania. Adult Morbidity and Mortality Project. Lancet 2000; 355(9216):1684–7. 24 Connor MD, Walker R, Modi G, Warlow CP. Burden of stroke in black populations in sub-Saharan Africa. Lancet Neurol 2007; 6(3):269–78.

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 977

References

25 Bonneux L, Looman CW, Barendregt JJ, Van der Maas PJ. Regression analysis of recent changes in cardiovascular morbidity and mortality in The Netherlands. Br Med J 1997; 314(7083):789–92. 26 Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003; 2(1):43–53. 27 Sarti C, Stegmayr B, Tolonen H, Mahonen M, Tuomilehto J, Asplund K. Are changes in mortality from stroke caused by changes in stroke event rates or case fatality? Results from the WHO MONICA Project. Stroke 2003; 34(8):1833–40. 28 Terent A. Trends in stroke incidence and 10-year survival in Soderhamn, Sweden, 1975–2001. Stroke 2003; 34(6):1353–8. 29 Pessah-Rasmussen H, Engstrom G, Jerntorp I, Janzon L. Increasing stroke incidence and decreasing case fatality, 1989–1998: a study from the stroke register in Malmo, Sweden. Stroke 2003; 34(4):913–18. 30 Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM et al. Change in stroke incidence, mortality, casefatality, severity, and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet 2004; 363(9425):1925–33. 31 Rudd AG, Hoffman A, Irwin P, Lowe D, Pearson MG. Stroke unit care and outcome: results from the 2001 National Sentinel Audit of Stroke (England, Wales, and Northern Ireland). Stroke 2005; 36(1):103–6. 32 Sudlow CL, Warlow CP. Comparable studies of the incidence of stroke and its pathological types: results from an international collaboration. International Stroke Incidence Collaboration. Stroke 1997; 28(3):491–9. 33 Appelros P, Nydevik I, Seiger A, Terent A. High incidence rates of stroke in Orebro, Sweden: Further support for regional incidence differences within Scandinavia. Cerebrovasc Dis 2002; 14(3–4):161–8. 34 Di Carlo A, Inzitari D, Galati F, Baldereschi M, Giunta V, Grillo G et al. A prospective community-based study of stroke in Southern Italy: the Vibo Valentia incidence of stroke study (VISS). Methodology, incidence and case fatality at 28 days, 3 and 12 months. Cerebrovasc Dis 2003; 16(4):410–17. 35 Correia M, Silva MR, Matos I, Magalhaes R, Lopes JC, Ferro JM et al. Prospective community-based study of stroke in Northern Portugal: incidence and case fatality in rural and urban populations. Stroke 2004; 35(9):2048–53. 36 Corbin DO, Poddar V, Hennis A, Gaskin A, Rambarat C, Wilks R et al. Incidence and case fatality rates of first-ever stroke in a black Caribbean population: the Barbados Register of Strokes. Stroke 2004; 35(6):1254–8. 37 Tsiskaridze A, Djibuti M, van MG, Lomidze G, Apridonidze S, Gauarashvili I et al. Stroke incidence and 30-day casefatality in a suburb of Tbilisi: results of the first prospective population-based study in Georgia. Stroke 2004; 35(11):2523–8. 38 Lavados PM, Sacks C, Prina L, Escobar A, Tossi C, Araya F et al. Incidence, 30-day case-fatality rate, and prognosis of

..

..

39

40

41

42

43

44

45

46

47

48

49

50

51

52

stroke in Iquique, Chile: a 2-year community-based prospective study (PISCIS project). Lancet 2005; 365(9478):2206–15. Syme PD, Byrne AW, Chen R, Devenny R, Forbes JF. Community-based stroke incidence in a Scottish population: the Scottish Borders Stroke Study. Stroke 2005; 36(9):1837–43. Morikawa Y, Nakagawa H, Naruse Y, Nishijo M, Miura K, Tabata M et al. Trends in stroke incidence and acute case fatality in a Japanese rural area : the Oyabe study. Stroke 2000; 31(7):1583–7. Feigin VL, Wiebers DO, Whisnant JP, O’Fallon WM. Stroke incidence and 30-day case-fatality rates in Novosibirsk, Russia, 1982 through 1992. Stroke 1995; 26(6):924–9. Powles J, Kirov P, Feschieva N, Stanoev M, Atanasova V. Stroke in urban and rural populations in north-east Bulgaria: incidence and case fatality findings from a ‘hot pursuit’ study. BMC Public Health 2002; 2:24. Mihalka L, Smolanka V, Bulecza B, Mulesa S, Bereczki D. A population study of stroke in West Ukraine: incidence, stroke services, and 30-day case fatality. Stroke 2001; 32(10):2227–31. Smadja D, Cabre P, May F, Fanon JL, Rene-Corail P, Riocreux C et al. ERMANCIA: Epidemiology of Stroke in Martinique, French West Indies: Part I: methodology, incidence, and 30-day case fatality rate. Stroke 2001; 32(12):2741–7. Liu M, Wu B, Wang WZ, Lee LM, Zhang SH, Kong LZ. Stroke in China: epidemiology, prevention, and management strategies. Lancet Neurol 2007; 6(5):456–64. Sacco RL, Boden-Albala B, Gan R, Chen X, Kargman DE, Shea S et al. Stroke incidence among white, black, and Hispanic residents of an urban community: the Northern Manhattan Stroke Study. Am J Epidemiol 1998; 147(3):259–68. Kissela B, Schneider A, Kleindorfer D, Khoury J, Miller R, Alwell K et al. Stroke in a biracial population: the excess burden of stroke among blacks. Stroke 2004; 35(2):426–31. Wolfe CD, Rudd AG, Howard R, Coshall C, Stewart J, Lawrence E et al. Incidence and case fatality rates of stroke subtypes in a multiethnic population: the South London Stroke Register. J Neurol Neurosurg Psychiatry 2002; 72(2):211–16. Hajat C, Tilling K, Stewart JA, Lemic-Stojcevic N, Wolfe CD. Ethnic differences in risk factors for ischemic stroke: a European case-control study. Stroke 2004; 35(7):1562–7. McGruder HF, Malarcher AM, Antoine TL, Greenlund KJ, Croft JB. Racial and ethnic disparities in cardiovascular risk factors among stroke survivors: United States 1999 to 2001. Stroke 2004; 35(7):1557–61. Feigin VL, Rodgers A. Ethnic disparities in risk factors for stroke: what are the implications? Stroke 2004; 35(7):1568–9. Wolfe CD, Smeeton NC, Coshall C, Tilling K, Rudd AG. Survival differences after stroke in a multiethnic population: follow-up study with the South London stroke register. Br Med J 2005; 331(7514):431.

977

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 978

978

Chapter 18 Reducing the impact of stroke and improving the public health

53 Sudlow C. Survival after stroke in south London. Br Med J 2005; 331(7514):414–15. 54 Feigin V, Carter K, Hackett M, Barber PA, McNaughton H, Dyall L et al. Ethnic disparities in incidence of stroke subtypes: Auckland Regional Community Stroke Study, 2002–2003. Lancet Neurol 2006; 5(2):130–9. 55 Hong Y, Bots ML, Pan X, Hofman A, Grobbee DE, Chen H. Stroke incidence and mortality in rural and urban Shanghai from 1984 through 1991. Findings from a community-based registry. Stroke 1994; 25(6):1165–9. 56 Lemesle M, Milan C, Faivre J, Moreau T, Giroud M, Dumas R. Incidence trends of ischemic stroke and transient ischemic attacks in a well-defined French population from 1985 through 1994. Stroke 1999; 30(2):371–7. 57 Jorgensen HS, Plesner AM, Hubbe P, Larsen K. Marked increase of stroke incidence in men between 1972 and 1990 in Frederiksberg, Denmark. Stroke 1992; 23(12):1701–4. 58 Kubo M, Kiyohara Y, Kato I, Tanizaki Y, Arima H, Tanaka K et al. Trends in the incidence, mortality, and survival rate of cardiovascular disease in a Japanese community: the Hisayama study. Stroke 2003; 34(10):2349–54. 59 Sivenius J, Tuomilehto J, Immonen-Raiha P, Kaarisalo M, Sarti C, Torppa J et al. Continuous 15-year decrease in incidence and mortality of stroke in Finland: the FINSTROKE study. Stroke 2004; 35(2):420–5. 60 Anderson CS, Carter KN, Hackett ML, Feigin V, Barber PA, Broad JB et al. Trends in stroke incidence in Auckland, New Zealand, during 1981 to 2003. Stroke 2005; 36(10):2087–93. 61 Linn FH, Rinkel GJ, Algra A, van Gijn J. Incidence of subarachnoid hemorrhage: role of region, year, and rate of computed tomography: a meta-analysis. Stroke 1996; 27(4):625–9. 62 Tolonen H, Mahonen M, Asplund K, Rastenyte D, Kuulasmaa K, Vanuzzo D et al. Do trends in population levels of blood pressure and other cardiovascular risk factors explain trends in stroke event rates? Comparisons of 15 populations in 9 countries within the WHO MONICA Stroke Project. World Health Organization Monitoring of Trends and Determinants in Cardiovascular Disease. Stroke 2002; 33(10):2367–75. 63 Rothwell PM, Warlow CP. Timing of TIAs preceding stroke: time window for prevention is very short. Neurology 2005; 64(5):817–20. 64 Truelsen T, Piechowski-Jozwiak B, Bonita R, Mathers C, Bogousslavsky J, Boysen G. Stroke incidence and prevalence in Europe: a review of available data. Eur J Neurol 2006; 13(6):581–98. 65 Malmgren R, Bamford J, Warlow C, Sandercock P, Slattery J. Projecting the number of patients with first ever strokes and patients newly handicapped by stroke in England and Wales. Br Med J 1989; 298(6674):656–60. 66 Hankey GJ, Warlow CP. Treatment and secondary prevention of stroke: evidence, costs, and effects on individuals and populations. Lancet 1999; 354(9188):1457–63.

67 Mehta DK. British National Formulary. 53rd edn. London: BMJ Publishing Group Ltd and RPS Publishing, 2007. 68 Hylek EM, D’Antonio J, Evans-Molina C, Shea C, Henault LE, Regan S. Translating the results of randomized trials into clinical practice: the challenge of warfarin candidacy among hospitalized elderly patients with atrial fibrillation. Stroke 2006; 37(4):1075–80. 69 Somerfield J, Barber PA, Anderson NE, Kumar A, Spriggs D, Charleston A et al. Not all patients with atrial fibrillationassociated ischemic stroke can be started on anticoagulant therapy. Stroke 2006; 37(5):1217–20. 70 Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE et al. Population-based study of event-rate, incidence, case fatality, and mortality for all acute vascular events in all arterial territories (Oxford Vascular Study). Lancet 2005; 366(9499):1773–83. 71 Fairhead JF, Rothwell PM. Underinvestigation and undertreatment of carotid disease in elderly patients with transient ischaemic attack and stroke: comparative population based study. Br Med J 2006; 333(7567):525–7. 72 Rudd AG, Lowe D, Hoffman A, Irwin P, Pearson M. Secondary prevention for stroke in the United Kingdom: results from the National Sentinel Audit of Stroke. Age Ageing 2004; 33(3):280–6. 73 Ovbiagele B, Saver JL, Fredieu A, Suzuki S, McNair N, Dandekar A et al. PROTECT: a coordinated stroke treatment program to prevent recurrent thromboembolic events. Neurology 2004; 63(7):1217–22. 74 Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005; 353(5):487–97. 75 Benade MM, Warlow CP. Cost of identifying patients for carotid endarterectomy. Stroke 2002; 33(2):435–9. 76 Wardlaw JM, Chappell FM, Stevenson M, De Nigris E, Thomas S, Gillard J et al. Accurate, practical and costeffective assessment of carotid stenosis in the UK. Health Technology Assessment 2006; 10(30):1–200. 77 Ezzati M, Hoorn SV, Rodgers A, Lopez AD, Mathers CD, Murray CJ. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet 2003; 362(9380):271–80. 78 Jackson R, Lynch J, Harper S. Preventing coronary heart disease. Br Med J 2006; 332(7542):617–18. 79 Murray CJ, Lauer JA, Hutubessy RC, Niessen L, Tomijima N, Rodgers A et al. Effectiveness and costs of interventions to lower systolic blood pressure and cholesterol: a global and regional analysis on reduction of cardiovasculardisease risk. Lancet 2003; 361(9359):717–25. 80 Emberson J, Whincup P, Morris R, Walker M, Ebrahim S. Evaluating the impact of population and high-risk strategies for the primary prevention of cardiovascular disease. Eur Heart J 2004; 25(6):484–91. 81 Marteau TM, Kinmonth AL. Screening for cardiovascular risk: public health imperative or matter for individual informed choice? Br Med J 2002; 325(7355):78–80. 82 Rose G. The Strategy of Preventive Medicine. New York: Oxford University Press, 1992. 83 Rothwell PM, Mehta Z, Howard SC, Gutnikov SA, Warlow CP. Treating individuals 3: from subgroups to

..

..

9781405127660_4_018.qxd 10/13/07 2:16 PM Page 979

References

84

85 86

87

88 89

90

91

92 93

94

95

96

..

..

individuals: general principles and the example of carotid endarterectomy. Lancet 2005; 365(9455):256–65. Ebrahim S, Harwood R. Stroke: Epidemiology, Evidence and Clinical Practice. 2nd edn. New York: Oxford University Press, 1999. Law MR, Wald NJ. Risk factor thresholds: their existence under scrutiny. Br Med J 2002; 324(7353):1570–6. Jackson R, Lawes CM, Bennett DA, Milne RJ, Rodgers A. Treatment with drugs to lower blood pressure and blood cholesterol based on an individual’s absolute cardiovascular risk. Lancet 2005; 365(9457):434–41. Cholesterol Treatment Trialists’ Collaborators. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366(9493):1267–78. MacMahon S, Neal B, Rodgers A. Hypertension: time to move on. Lancet 2005; 365(9464):1108–9. Brindle P, Beswick A, Fahey T, Ebrahim S. Accuracy and impact of risk assessment in the primary prevention of cardiovascular disease: a systematic review. Heart 2006; 92(12):1752–9. Brindle P, May M, Gill P, Cappuccio F, D’Agostino R, Sr., Fischbacher C et al. Primary prevention of cardiovascular disease: a web-based risk score for seven British black and minority ethnic groups. Heart 2006; 92(11):1595–602. D’Agostino RB, Sr., Grundy S, Sullivan LM, Wilson P. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. J Am Med Assoc 2001; 286(2):180–7. Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more than 80%. Br Med J 2003; 326(7404):1419. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and gender of patients with atrial fibrillation: analysis and implications. Arch Intern Med 1995; 155(5):469–73. Hobbs FDR, Fitzmaurice DA, Jowett S, Mant J, Bryan S, Raftery J et al. A randomised controlled trial and costeffectiveness study of systematic screening (targeted and total population screening) versus routine practice for the detection of atrial fibrillation in people aged 65 and over. The SAFE study. Health Technology Assessment 2005; 9(40):1–90. Ricci S, Flamini FO, Celani MG, Marini M, Antonini D, Bartolini S et al. Prevalence of internal carotid-artery stenosis in subjects older than 49 years: a population study. Cerebrovasc Dis 1991; 1:16–19. O’Leary DH, Anderson KM, Wolf PA, Evans JC, Poehlman HW. Cholesterol and carotid atherosclerosis

97

98

99

100

101

102

103

104

105

106

107

108

109

in older persons: the Framingham Study. Ann Epidemiol 1992; 2(1–2):147–53. Willeit J, Kiechl S. Prevalence and risk factors of asymptomatic extracranial carotid artery atherosclerosis: a population-based study. Arterioscler Thrombosis 1993; 13:661–8. Prati P, Vanuzzo D, Casaroli M, Di Chiara A, De Biasi F, Feruglio GA et al. Prevalence and determinants of carotid atherosclerosis in a general population. Stroke 1992; 23(12):1705–11. Lee TT, Solomon NA, Heidenreich PA, Oehlert J, Garber AM. Cost-effectiveness of screening for carotid stenosis in asymptomatic persons. Ann Intern Med 1997; 126(5):337–46. Whitty CJ, Sudlow CL, Warlow CP. Investigating individual subjects and screening populations for asymptomatic carotid stenosis can be harmful. J Neurol Neurosurg Psychiatry 1998; 64(5):619–23. Lightwood JM, Glantz SA. Short-term economic and health benefits of smoking cessation: myocardial infarction and stroke. Circulation 1997; 96(4):1089–96. Ezzati M, Lopez AD. Estimates of global mortality attributable to smoking in 2000. Lancet 2003; 362(9387):847–52. Barrett-Connor E, Palinkas LA. Low blood pressure and depression in older men: a population based study. Br Med J 1994; 308(6926):446–9. Dennis M, Bamford J, Sandercock P, Warlow C. Prognosis of transient ischemic attacks in the Oxfordshire Community Stroke Project. Stroke 1990; 21(6):848–53. Van Wijk I, Kappelle LJ, van Gijn J, Koudstaal PJ, Franke CL, Vermeulen M et al. Long-term survival and vascular event risk after transient ischaemic attack or minor ischaemic stroke: a cohort study. Lancet 2005; 365(9477):2098–104. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. Br Med J 2002; 324(7329):71–86. Rashid P, Leonardi-Bee J, Bath P. Blood pressure reduction and secondary prevention of stroke and other vascular events: a systematic review. Stroke 2003; 34(11):2741–8. The ESPRIT Study Group. Aspirin plus dipyridamole versus aspirin alone after cerebral ischaemia of arterial origin (ESPRIT): randomised controlled trial. Lancet 2006; 367(9523):1665–73. Rothwell PM, Eliasziw M, Gutnikov SA, Warlow CP, Barnett HJ. Endarterectomy for symptomatic carotid stenosis in relation to clinical subgroups and timing of surgery. Lancet 2004; 363(9413):915–24.

979

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 980

Index

Page numbers in italics represent figures, those in bold represent tables.

ABCD scale 520 abdominal aortic aneurysm 290 abdominal pain 80 Abercrombie, J 13 abulia 48 accommodation 614 ACE inhibitors 574 acidosis 651 activated partial thromboplastin time 306 activities of daily living 603–14, 603 dressing 612, 612 driving 606–7, 609 feeding 612–13, 613, 613 food preparation 613–14, 614 mobility 603–6, 604–6 toileting 609–10, 611 washing and bathing 610–11, 611, 611 acute arterial hypertension 430 acute confusional states 595–6 acute posterior multifocal placoid pigment epitheliopathy 367 acute stroke care 659 acute stroke syndrome 40, 41 acute stroke units 916–17 aetiological subclassification of stroke 171–3, 172, 173 age 281, 281 agraphaesthesia 53 agraphia 51 air embolism 355 air travel 615 airway, breathing and circulation 538–43 inadequate breathing 540–1 obstructed airway 539–40, 539, 540 poor circulation 541–3 akinetic mutism 48 alcohol 842–3, 843 alcohol, and stroke risk 286–7, 433 alexia 51, 52 allaesthesia 53, 594 altered gene expression 651–2

Alzheimer’s disease, amyloid angiopathy 420–1 amaurosis fugax see transient monocular blindness Ambrose, James 187 amitryptyline 562 amphetamines 381, 433, 435 amyloid angiopathy 21, 417, 712 and Alzheimer’s disease 420–1 familial forms 421 αβ amyloid protein 420 anaemia 374 anarthria 51 aneurysmal rupture, risk of 771–4, 771–3 aneurysms 89–92, 90–2 see also individual types angiography 242–3, 243 see also carotid imaging cerebral arteries 360 angioplasty see carotid angioplasty angle-closure glaucoma 104–5 anisocoria 73 ankle supports 608 anomia 51 anosodiaphoria 53, 540 anosognosia 53, 54, 594 anterior cerebral artery 139, 140–2, 141, 142, 143 anterior choroidal artery 139, 139 anterior communicating artery aneurysm 48, 99, 119, 194, 195, 475, 477, 483, 770 anterior inferior cerebellar artery 151–2 anterior ischaemic optic neuropathy 106–7, 106 anticoagulants 390, 431, 431, 469, 665–75, 708, 830–41 acute stroke in patients receiving 672–5, 673, 674 adverse effects 672, 672 agent 671–2

benefits 665 cost of interventions 965 dose and route 672 drug interactions 839 effect on stroke incidence 964 evidence for efficacy 665–9, 666, 667 basilar thrombosis 668–9 bleeding risks 667–8 data available 665–6 deep venous thrombosis/pulmonary embolism 668 effects on major outcomes 666–7, 666, 667 myocardial infarction 669 progressing hemispheric stroke 668 suspected cardioembolic ischaemic stroke 668 long-term prevention 830–8, 830, 831, 833, 834, 835 long-term treatment 840–1 mechanical prosthetic heart valves 838 rationale 665 risk of bleeding 838, 840 risks 665 treatment time 841 when to start 671 who not to treat 670–1, 671, 840 who to treat 669–70, 670 antiepileptic drugs 728 antifibrinolytics 735 antihypertensives, choice of 803–7, 804–6 anti-inflammatory drugs 382 antinuclear antibodies 306 antioxidants 846–7 antiphospholipid syndrome 363–4 antiplatelet drugs 431–2, 816–29, 820 aspirin see aspirin dipyridamole 827–9, 828 glycoprotein IIb/IIIa receptor antagonists 829

980

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 981

Index

length of treatment 829 long-term secondary prevention 816–20, 817, 819 thienopyridines 824–6, 825 triflusal 829 who not to treat 820 antispasticity drugs, adverse effects 584 antithrombotic drugs 971, 972 Anton syndrome 72 anxiety 81, 599–601 aorta 16, 17, 135, 264, 293, 300, 302, 315, 322, 355, 361, 362, 368, 370, 376, 383, 442, 472 aortic aneurysm 290 aortic arch atheroma 878 aortic arch dissection 358 aortic valve calcification 274 aphasia 51, 52, 598–9 apo B 285 apoplexy 8, 10–14 apoptosis 650–1 stages of 652 apparent diffusion coefficient (ADC) 227 apraxia 57–8, 57, 58 aprosody 52 arterial carbon dioxide tension 638 arterial dissection 355–8, 356, 430 aortic arch 358 cervical artery 355–8, 357 intracranial artery 358 arterial infarcts 241 arterial pulses 302 ‘arterial spin labelling’ 231 arteries see cerebral arterial supply arteriovenous fistulae see dural arteriovenous fistulae arteriovenous malformations 89–92, 90–2, 104, 193, 422–3, 423, 464, 780–3 frequency and presentation 780 grading system 782 natural history 781 treatment options 781–3 endovascular embolization 782–3 neurosurgical excision 781–2 stereotactic radiotherapy 783 unruptured 783 arteritis 321 artifacts 187, 194, 204, 224, 237, 270, 315, 468, 486, 734, 778 artificial heart valves 838 Ashworth scale 580 asomatognosia 53 ASPECTS 198 aspergillosis 468 aspiration 567–72 aspirin 558–9, 660–5, 707, 709, 820–3, 971 adverse effects 665 benefits 660–1, 821

..

..

combined with clopidogrel 432 costs 963, 965 delayed cerebral ischaemia 739, 741 dose and route 664 effect on stroke incidence 964 evidence for efficacy 661–2, 661 rationale 660–1 risks 661, 821–3, 822, 823 treatment failures 823–4 when to start 664 who not to treat 664 who to treat 662–4, 663 see also antiplatelet drugs assessment 43, 519–22, 520, 521, 521 astereognosis 53 ataxia 76 homolateral 161–2 ataxic hemiparesis 161–2 atheroma 262–7 atherothromboembolism 265–6 atherothrombosis 265 distribution of 262–4, 263, 264 dolichoectasia 267 of leg arteries 290 nature, progression and clinical consequences 264–5, 264 plaque instability 266–7, 266 atheromatous emboli 19 atherosclerosis in region of carotid bifurcation 16 atherothromboembolism 265–6, 322, 323 atherothrombosis 265 atorvastatin 815, 965 atrial fibrillation 272–3 anticoagulants 670 aspirin 663 treatment 847–8 atrial septal aneurysm 276–7, 848–9, 849 attention 49–50, 49 atypical antipsychotics 382 autoimmune disease 321 autoregulation 639–40, 639, 640 autosomal dominant polycystic kidney disease 769 Avicenna 9 AVM see arteriovenous malformations Babinski sign 61 baclofen 584 Barthel index 560 basilar artery 151 basilar branch occlusion 165 Bayle 11, 12, 14 beds 609 behaviour 81 Behçet’s disease 365–6 bendrofluazide 965 benign orgasmic cephalalgia 122

benign paroxysmal positional vertigo 96 benign recurrent vertigo 96 Bennett 14 Berengario da Carpi 10 Bernard, Claude 25 biomarkers 186 Biumi, Franciscus 8, 21 Blackall 21 Blackmore, E. 25 bladder care 727 blinding 26, 281, 486, 538, 655, 904 blood cultures 306 bloodletting 25 blood lipids 283–5, 283, 284 blood pressure 282–3, 282, 704, 968 high 549–50, 549, 550 low 550–1 subarachnoid haemorrhage 474, 725–6 blood pressure reduction 801–9, 964, 969–71, 973 blood pressure and stroke 801–3, 802, 803 choice of antihypertensive 803–7, 804 –6 dietary advice 843–4 how low 807–9, 808 randomized controlled trials 803 when to start 809 who not to treat 807–9, 808 Boerhaave, Herman 12, 13 borderzone infarction 265 boredom 602–3 boundary-zone infarcts 295 Bradford Hill, Austin 26 brain anatomy 8–10, 9, 10 blood supply 8–10, 9, 10 imaging 37–8, 39 brain abscess 94–5, 96 brain attack 40, 41, 260 differential diagnosis 83, 84 brain haemorrhage 14, 20, 39, 213, 426, 432, 435, 439, 710 see also intracerebral haemorrhage brain imaging 307–10 breast carcinoma 196 Bright, Richard 22 Broca’s area 52 Brown–Séquard syndrome 66 bruising 304 Buerger’s disease 367 CADASIL 216, 268, 385–6, 386 café-au-lait patches 304 calcific emboli 301 calcium 306 calcium antagonists 592 delayed cerebral ischaemia 739, 740

981

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 982

982

Index

calcium influx 650 cancer 378–9, 378, 379 candidate gene studies 287–8 cannabis 381 capsular warning syndrome 62, 270, 294 caput medusae 425, 426 carbamazepine, and hyponatraemia 574 cardiac arrhythmias 750–1, 751 cardiac dysrhythmias 542 cardiac examination 303 cardiac sources of embolism 271–8, 271, 322, 323–5, 324 atrial fibrillation 272–3 atrial septal aneurysm 276–7 coronary heart disease 273 infective endocarditis 274–5, 275 intracardiac tumours 277, 278 mitral valve prolapse 274 mitral valve strands 274 non-bacterial thrombotic (marantic) endocarditis 275–6, 275 non-ischaemic ‘primary’ cardiomyopathies 276 non-rheumatic sclerosis 274 paradoxical embolism 276–7, 276 patent foramen ovale 276–7 prevalence 272 prosthetic heart valves 273–4, 273 rheumatic valvular disease 274 sinoatrial disease (sick sinus syndrome) 277–8 valve calcification 274 cardiopulmonary complications 723, 725 cardiopulmonary disease 541 carer problems 617–19, 618, 619 caroticocavernous fistula 430, 430 carotid angioplasty 873–6, 874 –6 carotid artery dissection 115, 357 hypoplastic 369 ligation 23 carotid bifurcation 135–6 carotid body 136 ‘carotid disobstruction’ 18 carotid endarterectomy 849–73, 964, 965, 971–2 asymptomatic carotid stenosis 870–3, 872, 873 benefits 857–9, 857, 858, 862–6, 863–6, 867, 868 causes unrelated to carotid stenosis 868–70, 869 complications 853, 855–6, 856 patient selection 859–60, 859 perioperative stroke 853–5 procedures 850–3, 851 risk factors for surgery 860–2, 860

timing in relation to coronary artery surgery 876–7 carotid imaging 311–17 catheter angiography 311–13, 312, 313 CT angiography 315, 316 duplex sonography 313–15, 314 magnetic resonance angiography 315, 316 carotid occlusion 16, 80, 267, 301, 302, 642, 861, 865, 876 carotid pressure 554 carotid siphon 137, 137 Carswell, R 13 case–control studies 280 case fatality 957 Casserio, G 10 cataplexy 101–2 catheter angiography 311–13, 312, 313, 442–3 subarachnoid haemorrhage 487–9, 489 caudate haemorrhage 115, 438 causes of stroke 259, 260, 261–351, 353–409, 545 acute myocardial infarction 376–7, 376 arterial dissection 355–8, 356 atheroma 262–7 cancer 378–9, 378, 379 cardiac sources of embolism 271–8, 271 cholesterol embolization syndrome 370 congenital arterial anomalies 368–9, 368 drugs of misuse 381 embolism from arterial aneurysms 370 exogenous female sex hormones 379–80 gastrointestinal disorders 382 haematological disorders 371–6, 372 hypoglycaemia 382 identification of 321–5 infections 377–8, 388 inflammatory vasculopathies and connective tissue diseases 358–68 intracranial small vessel disease 267–71 intracranial venous thrombosis 387–90, 388 migraine 371 mitochondrial diseases 384–5, 385 Moyamoya syndrome 369–70, 369, 370 pregnancy and puerperium 380–1 single gene disorders 385–7 trauma 354–5 cavernous malformations 423–5 treatment 712 cavernous sinus 137, 137 celecoxib 382

cell death 648–52 acidosis 651 altered gene expression 651–2 apoptosis 650–1 excitotoxicity 648–50, 649 inflammation 651 oxidative stress 650 peri-infarct depolarization 651 central/branch retinal vein thrombosis 104, 105 central nervous system infections 94–6 Creutzfeldt–Jakob disease 95 encephalitis, brain abscess and subdural empyema 94–5, 96 central pain 65 central post-stroke pain 487–8 central serous retinopathy 105–6 central transtentorial herniation 653–4, 653 centrum semiovale 14, 147, 157, 207, 225, 291, 295, 426 cephalalgia, benign orgasmic 122 cerebellar arterial supply 152–4, 152–4 cerebellar haemorrhage 710–11, 710 cerebellar ischaemic stroke 294 cerebellar stroke 115 cerebral amyloid angiopathy 196, 417–21, 418–19 and Alzheimer’s disease 420–1 familial forms 421 genetics 420 ‘cerebral anaemia’ 15 cerebral arteries 133–58 angiography 360 anterior (carotid) system 135–48 anterior cerebral artery 140–2, 141, 142, 143 anterior choroidal artery 139, 139 carotid bifurcation 135–6 carotid siphon and cavernous sinus 137, 137 circle of Willis 139–40, 140 common carotid artery 135, 135 distal internal carotid artery 139 external carotid artery 136 extracranial internal carotid artery 136–7 middle cerebral artery 143–8 posterior communicating artery 138–9, 138 recurrent artery of Heubner 142–3, 143 supraclinoid internal carotid artery 137–8, 138 boundary zones 157–8, 158 cerebral arterial supply dissection 355–8, 357 functional characteristics 133

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 983

Index

posterior (vertebrobasilar) system 148–57 anterior inferior cerebellar artery 151–2 basilar artery 151 cerebellar arterial supply 152–4, 152– 4 extracranial vertebral system 148–9, 149 intracranial vertebral artery 149–50, 149 posterior cerebral artery 154–6, 155 posterior inferior cerebellar artery 150–1, 150 superior cerebellar artery 152, 152 thalamic arterial supply 156–7, 156 cerebral blood flow 554, 635–7, 636 global 637 regulation 637–41 cerebral blood volume 198, 209, 231, 232, 294, 637, 640, 641, 645, 646 cerebral circulation imaging 310–11 cerebral hyperperfusion 854 cerebral infarction 14–16 cerebral ischaemia 642–8 collateral blood flow 642 critical flow thresholds 642–3, 643 delayed 735–42 aspirin 739, 741 calcium antagonists 739, 740 cisternal drainage and intracisternal fibrinolysis 742 diagnosis 744–5 increasing plasma volume 742 induced hypertension and volume expansion 745, 745 magnesium sulphate 739 pathogenesis and risk factors 735–9, 736, 737, 738 statins 741 transluminal angioplasty and vasodilatation 745 duration of 644, 644 imaging techniques 645–8, 646, 647 ischaemic penumbra 644–5 loss of cellular ion homeostasis 643 loss of neuronal activity 643 oligaemia 643 sites of 642 thrombosis 642 cerebral metabolism 635–7, 636 cerebral oedema 652–4, 653 central transtentorial herniation 653–4, 653 cingulate herniation 653, 653 uncal herniation 654 cerebral perfusion pressure 554, 637–8 cerebral perfusion reserve 640–1, 641

..

..

cerebral salt wasting 573 ‘cerebral thrombosis’ 7, 16 cerebral tissue metabolic rate 638 cerebral tumours 433, 434 cerebral vasodilators cerebrospinal fluid drainage 705 examination 185–6, 186 ‘cerebrovascular accident’ 17, 41 cerebrovascular reactivity 319–20 cerebrovascular resistance 638 cervical arterial dissection 317–18 cervical bruits 290, 302–3, 303 CHADS2 score 273 chairs 608 Chalmers, Ian 28 Chalmers, Tom 28 Charcot 20, 27 Charcot–Bouchard microaneurysms 270 cheiro-oral-pedal syndrome 65, 161 cheiro-oral syndrome 65, 161 chest infection 556 chest pain 80–1, 300 Cheyne, John 13, 21 Cheyne–Stokes breathing 540 Chiari 13, 16 Chlamydia pneumoniae 289 cholesterol emboli 301 cholesterol embolization syndrome 370 cholesterol lowering drugs 809–16, 969–71 adverse effects 812–14 liver problems 813–14 muscle problems 812–13 peripheral neuropathy 814 benefits of 812 choice of drug 814–15 cholesterol and stroke 809–10 how low to go 814 –15 non-statins 815–16 randomized controlled trials 810–12, 811, 812, 813 when to start 815 who to treat 814 see also statins choriocarcinoma 89, 91, 196, 378, 381, 419, 433, 436 chorioretinitis 105 choroid plexus papilloma 89, 196 Cincinnati Prehospital Stroke Scale (CPSS) 42 cingulate herniation 653, 653 circle of Willis 21, 139–40, 140 circulatory problems 541–3 cardiac dysrhythmias 542 gastrointestinal haemorrhage 542–3 myocardial injury 542 cisternal drainage 742

classification of stroke 132, 158–71, 159 Trial of Org 10172 in Acute Stroke (TOAST) 171, 172 clinical diagnosis 108–10 reasons for disagreement 108–9 reduction of inter-observer variation 109–10, 109, 110 clinical scoring methods 185 lacunar syndromes 160–4, 160, 161, 163 Oxfordshire Community Stroke Project classification 168–71, 168–71 partial anterior circulation syndromes 166–7, 167 posterior circulation syndromes 164–5, 164, 165 syndromes of uncertain origin 167–8 total anterior circulation syndrome 165–6, 165, 166 clinical trials 26 clipping of aneurysms 774 clopidogrel see thienopyridines combined with aspirin 432 clotting factor deficiency 432 coagulopathy 194, 378, 379, 390, 433, 439, 469 cocaine 381, 435, 468 cochlear implant 211 Cochrane Central Register of Controlled Trials (CENTRAL) 659 Cochrane Collaboration 28 Cochrane Effective Practice and Organization of Care (EPOC) Group 659 coexisting medical problems 548–9 Cogan syndrome 368 cognitive dysfunction 595–8, 753 acute confusional states 595–6 causes of 596 post-stroke dementia 596–8 Cohnheim, Julius Friedrich 13, 16 cohort studies 280 coiling see endovascular occlusion coin lesions 277, 277 collateral blood flow 642 collaterals 133, 136, 166, 207, 223, 241, 266, 290, 293, 362, 369, 370, 374, 429, 642, 712, 877 colour anomia 50 commodes 611 common carotid artery 135, 135 complicated plaques 265 comprehensive stroke units 917–18 compression stockings 558, 707–8 computed tomography 21, 182, 183, 185, 186–210 brain scan 118–20, 119, 120, 187, 307–8, 309, 439–40, 440, 441

983

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 984

984

Index

intracerebral haemorrhage 188–96 changing appearance with time 189–91, 190 interpretation 193–4, 194 late appearance 191–3, 191–3 recent 188–9, 189 timing of 187–8 underlying cause 194–6, 195, 196 visibility of 194 ischaemic stroke 196–209 cause of infarct 207–8, 207 correspondence with clinical diagnosis 204 differential diagnosis 208–9, 208 early signs 196–200, 197–9 evolution 200–2, 201–2 inter-observer reliability 205–6 intravenous contrast 202–4, 203 online training 206 prognostic significance 205 subarachnoid haemorrhage 476–85, 476–85 blood in subarachnoid cisterns 477, 479–80, 480–3 intracerebral extension 476–7, 477–9 intraventricular haemorrhage 480–1, 482 non-aneurysmal perimesencephalic haemorrhage 482–3, 484, 485 subdural haematoma 481–2 venous thrombosis 243–4, 244 vs magnetic resonance imaging 236 computed tomography angiography 210, 315, 316 subarachnoid haemorrhage 487 computed tomography perfusion 209–10, 648 concentration 49–50, 49 confusion 727 congenital arterial anomalies 368–9, 368 congophilic angiopathy 417 see also amyloid angiopathy connective tissue diseases 358–68 consciousness 46–8 brain areas involved in 47 causes of impairment 46 clinical anatomy 46 clinical assessment 46–7, 47 clinical practice 47–8 loss of 113, 472 reduced level of 543–4, 543 complications 544 remediable causes 543 subarachnoid haemorrhage 473–4, 727 CONSORT group 656 Constantinople, fall of 9

constipation 564 subarachnoid haemorrhage 727 continuous positive airways presure 541 contractures see spasticity and contractures contraindications to magnetic resonance imaging 211 contrast 202–4, 203 controlled clinical trials 654–5, 790 Copernicus 9 corneal and eyelid disorders 107–8 Fuch’s cornea endothelial dysfunction 107–8 ptosis 108 coronary artery bypass surgery 878 coronary circulation imaging 320 coronary heart disease 273, 289–90 corpora fibrosa 11 cortical blindness 72 ‘cortical hand’ 62 cortical vein thrombosis 196, 242 corticosteroids 689, 706, 707 costs acute treatment 961 long-term treatment 965–6, 965 ‘cotton-wool patches’ 71 cough headache 116 craniotomy 709–10 C-reactive protein 186 Creutzfeldt–Jakob disease 95 critical flow thresholds 642–3, 643 ‘crossed’ aphasia 52 crural paresis 161–2 Cruveilhier, Jean 13 CT see computed tomography Cullen, William 14 Cushing, Harvey 22 CVA see cerebrovascular accident da Carpi, Berengario 10 DALYs see disability-adjusted life years Dandy 24–5, 24 dantrolene 584 Database of Abstracts of Reviews of Effectiveness (DARE) 659 da Vinci, Leonardo 9 D-dimer 186 death, prognosis for 507–9, 507, 508, 508 decussation of pyramids 11, 11 deep cerebral vein occlusion 196 deep sensation 63 deep venous thrombosis 557–60, 557, 559 anticoagulants 668 aspirin therapy 662 assessment 558 prevention 558–9, 559 prevention of 706–8 treatment 559–60

defibrinogenation 688 de Fouchy, Jean Paul Grandjean 17 dehydration 572–3, 573 Déjerine–Roussy syndrome 65, 157 dementia, post-stroke 596–8 Denny–Brown 18 ‘dense triangle sign’ 243 dentures 577 deoxyhaemoglobin 212, 213, 215, 426, 647 dependency 509, 509 depression 599–601 deterioration after stroke 545–8, 546–8 assessment 547, 547 causes of 546 Scandinavian Stroke Scale 546 treatment 547–8, 548 detrusor instability 560 ‘developmental venous anomaly’ 425 diabetes mellitus 283, 847 diagnosis, three-step model 173 diapedesis 223 diazepam 584 diet 286, 843–5 dietary supplements 845–7 antioxidants 846–7 folic acid 845–6, 971 differential diagnosis 81–103 diffusion-perfusion mismatch 217, 232–4, 233, 234 diffusion-weighted imaging 216, 221, 226–7, 310 and clinical outcome 229, 229 dilute Russell’s viper time 306 diplopia 74, 593 monocular 75 transient 75 dipyridamole 827–9, 828, 964, 965 disability 504 disability-adjusted life years (DALYs) 974 disability scales 28 discrimination sensation 63 dissection see arterial dissection disseminated intravascular coagulation 376 distal internal carotid artery 139 diuretics, and hyponatraemia 574 dizziness 76, 78, 98 dolichoectasia 151, 267, 268 Dott, Norman 23, 24 Drake 25 dressing 612, 612 dressing apraxia 53 drez-otomy 582 drift of outstretched arm 60 driving 606–7, 609 drop attacks 101 drug abuse 79

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 985

Index

drug screens 306 drugs of misuse 381 duplex sonography 313–15, 314 dural arteriovenous fistulae 427, 427, 466, 468 treatment 712 dural sinuses 122, 244, 387, 388, 389, 427, 442, 712 Dutch type amyloid 421 DWI see diffusion-weighted imaging dysarthria 51, 52, 599 dysarthria-clumsy hand syndrome 161–2 dysequilibrium 76 dysgraphia 51 dyslexia 51 dysnomia 51 dysphagia 62, 567–72 detection 568–71, 568–72, 570, 571 mechanisms of 567–8, 568 prognosis and management 571–2 dysphasia 51, 52 dysphonia 51, 52, 98 dyspnoea 80–1 dyspraxia 51 Eales disease 368 early supported discharge services 922–3, 922, 923 echocardiography 323, 324 echoplanar MRI 211 eclampsia 381 Edinger–Westphal nuclei 72, 73 Ehlers–Danlos syndrome 387, 460 electrocardiography 325 electroencephalography 103, 307, 320 embolism 16–17, 370 emotionalism 601–2 employment 614 –15 ‘empty delta sign’ 243, 243 encephalitis 94–5, 96 encephalomalacia 15 endocarditis infective 274–5, 275 non-bacterial thrombotic (marantic) 275–6, 275 endoscopic evacuation 709, 710 endovascular occlusion 729–32, 729–31, 733 complications of coiling 732 follow-up 732 when to coil 732 energy demand 554, 635–7, 636 enteral tube feeding 577, 578 epidermal naevi 304 epilepsy 752–3 differential diagnosis 86–7 stroke 87, 87 transient ischaemic attacks 86–7, 86

..

..

epileptic seizures 80, 113, 299, 436, 551–3, 551, 552 assessment 552 early 551 late 551–2 subarachnoid haemorrhage 472 treatment 552–3 equipotentiality 27 erythrocyte sedimentation rate 305 essential thrombocythaemia 373 état criblé 144, 146 état lacunaire 144 ethical issues 531–3 ethnicity 287 ethnic variation in stroke incidence 959 etoricoxib 382 evidence-based medicine 26, 654, 655 examination 44–5, 44, 436–9, 437, 438 excitotoxicity 648–50, 649 exercise 286, 845 ‘exploding head’ syndrome 116 external carotid artery 136 external ocular movements 73–6 external ventricular drainage 748–9 extinction 53 extracranial internal carotid artery 136–7 extracranial vertebral system 148–9, 149 extracranial vessels 135 extra-to-intracranial bypass surgery 877 eye examination 301–2 eyelids 73–6 Fabry’s disease 386–7 Face Arm Speech Test (FAST) 42, 519 facial weakness 483 factor VII 285 factor VIIa, recombinant 704, 705, 735 faecal incontinence 564 falls 591–2, 591 falx 189, 653 familial forms 424, 425 amyloid angiopathy 421 familial stroke, causes 299 family history 81, 300–1 Farr, William 25 fat embolism 355 fatigue 603 feeding 612–13, 613, 613 fever 303, 555–7, 556 subarachnoid haemorrhage 727 fibrinogen 285 fibrinoid necrosis 94, 269, 415 fibrin-platelet emboli 301 fibrocartilaginous embolism 355 fibrohyalinosis 216 fibromuscular dysplasia 368–9, 368 finance 617 finger-clubbing 304

Fisher, C. Miller 8, 18 FLAIR 211, 212, 215 flavoxate hydrochloride 562 Flourens 27 flow void 137, 199, 218, 235, 241, 267, 309, 357, 443, 466 fluids and electrolytes 305 subarachnoid haemorrhage 726 fluvastatin 815 ‘fogging effect’ 200, 202, 223 folic acid supplementation 845–6, 971 food preparation 613–14, 614 foreign accent syndrome 52 ‘forest plots’ 28, 29 fractures 591–2, 591 free radicals 650 French Revolution 14 Fritsch, GT 27 Fuch’s cornea endothelial dysfunction 107–8 full blood count 305 functional MRI 235 functional scales 938–9 functional sensory disturbance 97–8 functional weakness 97 fusiform aneurysms 464–6, 466 gag reflex 569 gait ataxia 76 Galen 9 Galenus of Pergamon 8–9, 12 Gall 27 gastrointestinal disorders 382 gastrointestinal haemorrhage 542–3 Gavarret, Jules 25 gaze paresis 540 gender 281–2, 282 gene expression 651–2 genes 425 genetic background in sporadic cases 420 genetic studies 287–9 geographical disorientation 53 geographic variation in stroke incidence 957–9, 958 Gerstmann syndrome 52 giant cell arteritis 70, 361–2, 361 Glasgow Coma Scale 47, 544 glaucoma 69 glioblastoma 89, 183, 195, 196, 433, 434 global aphasia 52 globus pharyngis 98 glucose intolerance 847 glycerol 689–90, 706 glycoprotein IIb/IIIa receptor antagonists 829 goals 975–6 goal setting 524–5, 524 Goltz 27

985

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 986

986

Index

gradient echo 88, 103, 190–2, 211, 212, 213, 214, 215, 216, 218, 226, 236, 295, 309, 385, 416, 417, 420, 424, 425, 435, 442, 485 Graunt, John 25 Guglielmi 25 guidelines 943, 943 haematocrit 285 haematoidin 213 haematological disorders 371–6, 372 anaemia 374 disseminated intravascular coagulation 376 essential thrombocythaemia 373 intravascular lymphoma 373 leukaemia 373 lymphomatoid granulomatosis 373– 4 paraproteinaemias 374 –5 paroxysmal noctural haemoglobinuria 374 –5 polycythaemia 373 sickle cell disease 374 thrombophilias 375–6 thrombotic thrombocytopenic purpura 373 haematoma expansion 722–3, 723 haemodilution 691–2 haemodynamic theory 18 haemoglobin 306 haemorrhage from deep arterioles 416–17 pontine 711, 712 haemorrhagic metastasis 193 haemorrhagic necrosis 16 haemorrhagic transformation of infarct 237–41, 238–40, 427–8 haemosiderin 191, 192, 213 haemostatic variables 285 hallucinations, visual 593 handicap 504, 939 handicap scales 28 Harvey, William 10 headache 79–80, 111, 298–9, 553 benign exertional 117 benign ‘thunderclap’ 117–18 cough 116 hypnic 116 idiopathic stabbing 116 sentinel 112 subarachnoid haemorrhage 726 headache, post-traumatic 116–17 head injury 99, 99, 354, 469 hearing, balance and coordination disturbance 76–8 clinical anatomy 76 clinical assessment 76–7 clinical practice 78

heart see cardiac Helicobacter pylori 289 hemianopia 50 hemiataxia 157 hemi-attention 594 hemiballismus 62 hemicraniotomy 690 hemi-inattention 53 hemiparesis 475 hemiplegia 11, 15–17, 19, 62, 72, 92, 136, 159, 160, 161, 165, 521, 555, 557, 580, 581, 582, 669, 683 Henoch–Schönlein purpura 365 heparin 559 adverse effects 672 subcutaneous 707 see also anticoagulants hepatic encephalopathy 93 herniation 46, 47, 75, 116, 120, 189, 200, 389, 390, 415, 508, 653–4 central transtentorial 653–4, 653 cingulate 653, 653 uncal 654 hiccups 80, 553, 553 high blood pressure 549–50, 549, 550 see hypertension higher cerebral functions 49–58 attention and concentration 49–50, 49 disorders of praxis 57–8, 57, 58 memory 50–1, 50, 50, 51 speech and language 51–3, 53 visuospatial dysfunction 53–7, 53, 54 – 6 high-intensity transient signals 19 Hippocrates 12 history 7–34, 42–3, 43 Hitzig, E 27 HIV serology 306 Hodkinson Abbreviated Mental Test 596 Hollenhorst plaques 70 homocysteine 285 homocystine 306 homocystinuria 386 homolateral ataxia and crural paresis 161–2 homonymous hemianopia 67 homonymous visual field defects 66, 67 Hope, J 14 hormone replacement therapy 380 Horner syndrome 73, 356 hospital referral bias 260 Hounsfield, Godfrey, N. 20 Hounsfield units (HU) 188 humoral theory 8 Hunter, John 10 Hunt, Ramsay 18 hyaline arteriosclerosis 268, 269

hydatid cysts 278 hydrocephalus 703 acute 723, 724 diagnosis 746–7, 746, 747 acute obstructive 115 interventions 747–9, 747 late 752 hyperattenuated artery sign 198, 200 ‘hyperattenuated sinus sign’ 196, 243, 244 hypercalcaemia 93 hypercoagulability 375–6 hyperdense artery sign 218 hyperglycaemia 93, 575, 750 hypertension 412–17, 430, 436 induced 745, 745 microaneurysms 413–16, 413–16 microbleeds 416 see also high blood pressure hypertensive encephalopathy 94 hypertensive retinopathy 71 hyperventilation 690, 705–6 hypnic headache 116 hypoglycaemia 92–3, 382, 575 hyponatraemia 93, 573–5, 574, 749–50, 750 hypoplasia 140, 242, 355, 369, 442 hypospadias, traumatic 562 hypoxia 540 Icelandic type amyloid 421 ICH see intracerebral haemorrhage idiopathic reversible cerebral ‘vasoconstriction’ 367 imaging brain 307–10 carotid 311–17 cerebral circulation 310–11 coronary circulation 320 techniques 646 transcranial Doppler ultrasound 245, 295, 318–20 vascular 318 see also individual techniques imipramine 562 immobility 553–5, 554, 554 impact of stroke 903–4 impairment 504 incidence of stroke 181–3, 182, 183, 957–60, 961 ethnic variation 959 geographic variation 957–9, 958 time trends 959–60 incontinence see faecal incontinence; urinary incontinence indapamide 965 indwelling catheters 562, 563 infarction 16, 241

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 987

Index

infection 289, 377–8, 388, 555–7 chest 556 urinary tract 556–7 infective endocarditis 274–5, 275 inflammation 15, 22, 651 inflammatory bowel disease 321 inflammatory vasculopathies 358–68 information sources 659 infratentorial haemorrhage 710–11, 710 initial management 704–5, 705 innominate artery stenosis 878 integrated care pathways 943–4 integrated management plan 518–31 assessment 519–22, 520, 521, 521 goal setting 524–5, 524 problem identification 522–4, 522, 523 stroke team 525–6, 525, 525, 526 multidisciplinary team meetings 529, 530–1 roles of team members 526–30 intellectual deterioration 417, 436 intermediate phenotypes 288 intermittent claudication 18 intermittent pneumatic compression 707–8 internal boundary zone 63, 147, 148, 158, 207, 295 internal carotid artery 134, 262 loops 369 occlusion 854 internal carotid-posterior communicating artery aneurysm 478, 480 interventional neuroradiology 688–9 intonation 52 intra-arterial thrombolysis 681 intracardiac tumours 277, 278 intracerebral haemorrhage 20–1, 20, 21, 91, 703–17 causes of 411–56 clues from examination 436–9 clues from history 436 haemodynamic factors 430 haemostatic factors 431–3 relative frequency 435–6 structural factors 412–30 computed tomography 188–96 changing appearance with time 189–91, 190 interpretation 193–4, 194 late appearance 191–3, 191–3 recent 188–9, 189 timing of 187–8 underlying cause 194–6, 195, 196 visibility of 194 differential diagnosis 115 iatrogenic 708–9 incidence of 182

..

..

initial management 704 –5, 705 investigations 439–43 mortality 800 non-traumatic 88 pathophysiology 703 prevention of deep venous thrombosis/pulmonary embolism 706–8 prognosis 703–4, 800–2 recurrence 800–1 reduction of intracranial pressure 705–6 surgical treatment cavernous malformations 712 dural arteriovenous fistulae 712 infratentorial haemorrhage 710–11, 710 lobar haemorrhage 712 moyamoya syndrome associated with haemorrhage 712–13 supratentorial haemorrhage 709–10 treatment anticoagulants 708 aspirin 709 thrombolytic therapy 708–9 vs haemorrhagic transformation of infarct 237–41, 238–40 vs ischaemic stroke 183–6, 184 intracisternal fibrinolysis 742 intracranial aneurysm clips 25, 211, 468, 734, 734, 770, 777 intracranial aneurysm coils 774–5 see also coiling intracranial arteries 135, 318 dissection 358, 463–4 intracranial haemorrhage, thrombolytic therapy 675–6, 676 intracranial pressure 554 reduction of 705–6 intracranial venous thrombosis see venous thrombosis intraocular foreign bodies 211 intraocular haemorrhage 105 intraocular metallic foreign body 211 intravascular amyloid 420 intravascular lymphoma 373 intraventricular haemorrhage 115, 471, 480–1, 482, 723, 724 investigation 304 –21 cervical arterial dissection 317–18 cost-effectiveness 320–1 electroencephalogram 320 imaging brain 307–10 carotid 311–17 cerebral circulation 310–11 coronary circulation 320 vascular 318

lumbar puncture 320 routine 304–5, 305 second-line 305–7, 306–7 transcranial Doppler ultrasound 318–20 vertebrobasilar transient ischaemic attacks 317 Iowa mutation 421 ischaemic necrosis 13, 16, 189, 363, 648 ischaemic oculopathy 68–9, 70, 302 ischaemic penumbra 644–5 definition of 645, 645 ischaemic stroke 14–16, 37, 38 causes see causes of stroke computed tomography 196–209 cause of infarct 207–8, 207 correspondence with clinical diagnosis 204 differential diagnosis 208–9, 208 early signs 196–200, 197–9 evolution 200–2, 201–2 inter-observer reliability 205–6 intravenous contrast 202–4, 203 online training 206 prognostic significance 205 costs acute treatment 961 long-term treatment 965–6, 965 effects of long-term treatment 961–5, 962–4 gradual onset 297, 298 magnetic resonance imaging 216–34 apparent diffusion coefficient 227–8, 227, 228 and clinical practice 229, 229 diffusion imaging 224, 225 diffusion-perfusion mismatch 232–4, 233, 234 diffusion weighted imaging 226–7 early changes 218–19, 219, 220 early evolution 219, 221, 222–3, 222 early signs 216–18, 217, 217, 218 late appearance 223 new vs old infarcts 223–4, 224 perfusion-weighted MRI 231–2, 231 prediction of outcome 228 presence of ischaemic lesion 229–31, 230 thrombosed artery 218, 218 visibility of lesions 226 pathophysiology 635–54 cerebral blood flow regulation 637–41 cerebral ischaemia 642–8 cerebral metabolism 635–7, 636 ischaemic cerebral oedema 652–4, 653 phases and mediators of cell death 648–52

987

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 988

988

Index

precipitating factors 297–8 risk factors 278–90 treatment see treatment vs intracerebral haemorrhage 183–6, 184 isodense subdural haematoma 194 isointense 189, 191, 197, 200, 202, 244, 442 isolated dysarthria 52 Italian kindred 421 Kawasaki disease 365 Kohlmeier–Degos disease 366 laboratory studies 439 labyrinthine disorders 96 benign paroxysmal positional vertigo 96 benign recurrent vertigo 96 vestibular neuronitis (labyrinthitis) 96 lacunar infarction 144, 146, 291–4, 292, 293 lacunar syndromes 65, 160–4, 160, 161, 163 Lallemand, Claud François 13, 14 large vessel disease see atheroma lax skin 304 leisure 615, 616 lenticulostriate arteries 135, 143–4, 144 leptomeningeal arteries 166, 223, 366, 368, 369, 385, 421, 468, 641, 712 leucocyte α-galactosidase A 306 leukaemia 321, 373, 432–3 leukoaraiosis 181, 216, 417 lifestyle 81 lifestyle modification 841–5, 971 avoiding excess alcohol 842–3, 843 dietary changes 843–5 exercise 845 smoking cessation 841–2, 842 ligation of carotid artery 23 lightning streaks of Moore 105 limb weakness 583–6, 585 Lind, James 26 linkage studies 288 lipids see blood lipids lipohyalinosis 269 lipoprotein fractionation 306 lisinopril 965 Lister, Joseph 25 livedo reticularis 304, 364, 365 liver function tests 306 lobar haemorrhage 712 locked-in syndrome 48, 62, 151 Los Angeles Paramedic Stroke Scale (LAPSS) 42 Louis, Pierre Charles Alexandre 25, 26 lovastatin 815

low blood pressure 550–1 low cerebral blood flow 294–7 boundary-zone ischaemia and infarction 295 implications for treatment 297 in ischaemic strokes and transient ischaemic attacks 295–7, 296 lumbar puncture 120–2, 320 hydrocephalus 748 subarachnoid haemorrhage 485 lumiracoxib 382 lung carcinoma 196 luxury perfusion 202, 642 lymphoma 373 lymphomatoid granulomatosis 373–4 magnesium sulphate 739 magnetic resonance angiography 234, 235, 315, 316 subarachnoid haemorrhage 487 magnetic resonance diffusion-weighted imaging see diffusion-weighted imaging magnetic resonance imaging 134, 182, 210–36, 308–10, 440, 442, 442 contraindications 211 haemorrhage 211–16, 212–16 history 210–11 ischaemic stroke 216–34 apparent diffusion coefficient 227–8, 227, 228 and clinical practice 229, 229 diffusion imaging 224, 225 diffusion-perfusion mismatch 232–4, 233, 234 diffusion weighted imaging 226–7 early changes 218–19, 219, 220 early evolution 219, 221, 222–3, 222 early signs 216–18, 217, 217, 218 late appearance 223 new vs old infarcts 223–4, 224 perfusion-weighted MRI 231–2, 231 prediction of outcome 228 presence of ischaemic lesion 229–31, 230 thrombosed artery 218, 218 visibility of lesions 226 subarachnoid haemorrhage 485–7, 486 suspected stroke 211 transient ischaemic attacks 38, 39, 224, 226 venous thrombosis 244–5 vs computed tomography 236 magnetic resonance spectroscopy 234–5, 236 malaise 299 malignant arterial hypertension 107 malignant atrophic papulosis 366

malnutrition 575–6 prevention and treatment 577 ‘man in the barrel’ syndrome 62 mannitol 689–90, 706 marantic vegetations 275–6, 275 marche à petits pas 144 Marfan syndrome 387 masking see blinding matrix metalloproteinase 186 1/3 MCA 198 mean transit time 209, 231, 233, 646 measures of success 975–6 mechanical ventilation 541 medial longitudinal fasciculus 74 medical history 81, 81, 300 Mediterranean diet 845 medulla 77 medullary perforating arteries 147–8, 148 melanoma 89, 196, 378, 419, 433, 436, 465 Melbourne Ambulance Stroke Screen (MASS) 42 memory 50–1, 50, 50, 51 ‘meningeal apoplexy’ see subarachnoid haemorrhage meningioma 91 meningism 114 meningitis 114–15 meta-analysis 28, 658 metabolic disturbances 572–5 dehydration 572–3, 573 hyperglycaemia 575 hypoglycaemia 575 hyponatraemia 573–5, 574 metabolic syndrome 286 metabolic and toxic disorders 92–4 hepatic encephalopathy 93 hypercalcaemia 93 hyperglycaemia 93 hypertensive encephalopathy 94 hypoglycaemia 92–3 hyponatraemia 93 posterior reversible encephalopathy syndrome 94 Wernicke’s encephalopathy 93 Wilson’s disease 94 methaemoglobin 212, 213, 213, 357, 426 microaneurysms 413–16, 413–16 microbleeds 416 microhaemorrhage 215–16, 216 micropsia 72 midbrain 77 middle cerebral artery 135, 143–8, 262 aneurysm 24, 195, 474, 477, 487, 776 cortical branches 144, 147 lenticulostriate branches 143–4, 144 medullary perforating arteries 147–8, 148

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 989

Index

migraine 83–6, 321, 371, 430 aura 85 differential diagnosis 115–16 stroke 85–6 transient ischaemic attack 83–6, 84, 85 as risk factor for stroke 371 visual symptoms 85 migrainous stroke 83–6, 321, 371 miliary aneurysms 20 Millard–Gubler syndrome 164 Mini Mental State Examination 597 ‘misery perfusion’ 18 mismatch 217, 232–4, 233, 234 misoplegia 53 Mistichelli, Domenico 11, 12 mitochondrial diseases 384–5, 385 mitral valve calcification 274 mitral valve prolapse 274 mitral valve strands 274 mixed aphasia 52 mixed essential cryoglobulinaemia 366 mobile stroke teams 919 mobility 603–6, 604–6 modified Ashworth scale 580 Moniz, Egas 22 monoamine oxidase inhibitors 117 monocular blindness 19 mononeuropathy 100 monoparesis 62, 475 monoplegia 62 Morgagni, Giovanni Battista 11, 12, 13 mortality 797–8, 797, 798, 954–7, 954, 955 changes in case fatality 957 changes in diagnostic accuracy 956–7 competing causes of death 957 death certification practice 956 errors in age standardization 956 mixture of stroke types 957 motor cortex 58 motor neurone disease 100–1 motor system disturbance 58–63 clinical anatomy 58–9, 59 clinical assessment 59–61, 60, 61 clinical practice 62–3 motricity index 584, 585 moyamoya syndrome 369–70, 369, 370, 429–30 associated with haemorrhage 712–13 MR compatibility 211 MRI see magnetic resonance imaging MRI DWI-PWI 646–8, 647 multidisciplinary team meetings 529, 530–1 involvement of patients and carers 531 recording work of team 531

..

..

structure of discussion 530 team building 531 multiple ICHs 196, 432 multiple sclerosis 99, 100 myasthenia gravis 100 ‘mycotic’ aneurysms 274–5, 275, 428–9, 467–8, 473 myocardial contusion 278 myocardial infarction 376–7, 376, 799–800 myocardial injury 542 myxoma 277, 278 cerebral metastases 468 nasogastric feeding 579 National Institute of Health Stroke Scale (NIHSS) 45 nausea 553 neck bruit see cervical bruits neck injury non-penetrating 354 penetrating 354 neck stiffness 80, 112–13 subarachnoid haemorrhage 474 neck trauma 321 neglect 54 neovascular capillary proliferation 223 ‘nervous energy’ 11 neuroblastoma 89, 196 neurofibromatosis 387 neuroimaging 102–3 neurological examination 301 neurological signs 109 neuromuscular disorders 100–1 mononeuropathy and radiculopathy 100 motor neurone disease 100–1 myasthenia gravis 100 neuroprotective agents 691 nicaraven 741 nizofenone 741–2 non-bacterial thrombotic (marantic) endocarditis 275–6, 275 non-belonging 540 non-ischaemic ’primary’ cardiomyopathies 276 non-penetrating neck injury 354 non-rheumatic sclerosis 274 non-stroke vascular disease 289–90, 289 non-traumatic intracerebral haemorrhage 88 ‘no reflow’ phenomenon 223 nortriptyline 562 number needed to treat 656 numerical method 657 nursing staff 526–7 nutritional problems 575–9 assessment 576, 576

malnutrition 575–6 prevention and treatment 577 obesity 576 treatment 578 nystagmus 74, 75 obesity 286, 576 treatment 578 obstructed airway 539–40, 539, 540 occipital neuralgia 117 occlusion of major venous sinus 427 occupational therapist 527–8 ocular bobbing 76 oedema see cerebral oedema oligaemia 643 oligodendroglioma 195, 378 ‘one-and-a-half syndrome’ 75 ophthalmoplegia, internuclear 75 ophthalmoscopy 69–70 optic disc anomalies 107 optic nerve disorders 106–7 anterior ischaemic optic neuropathy 106–7, 106 malignant arterial hypertension 107 optic disc anomalies 107 optic neuritis 107 papilloedema 107, 107 Uhthoff’s syndrome 107 optic neuritis 107 oral contraceptives 321, 379–80 oral nutritional supplements 577 orbital disorders 107 orogenital ulceration 304 oscillopsia 74, 76, 593 Osler 17 osmotic agents 706 ‘ossification’ of arteries 15 outcome 26–7 oxazepine, and hyponatraemia 574 Oxfordshire Community Stroke Project 108, 166, 168–71, 168–71 oxidative stress 650 oxybutynin hydrochloride 562 oxygen extraction fraction 637, 641 oxygen therapy 541 oxyhaemoglobin 121, 211, 212, 213, 215, 426, 647 pacemakers 38, 44, 101, 103, 211 pain 587–8, 587 palinopsia 72 palpitations 80–1 panic 81 papilloedema 89, 107, 107 paradoxical embolism 276–7, 276 paraesthesiae 63 paraneoplastic retinopathy 105 paraneoplastic vasculitis 367

989

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 990

990

Index

paraparesis 475 paraplegia 62 paraproteinaemias 374–5 parasagittal motor cortex 101, 189, 243, 244, 389, 425, 428, 438, 439 Paré, Ambroise 23 Parinaud syndrome 475 paroxysmal noctural haemoglobinuria 374–5 partial anterior circulation infarction 166–7, 167, 291 patent foramen ovale 276–7, 848–9, 849 pathology 504 patient information 926–7, 926 patient satisfaction 939–40 patterns of recovery 509–11, 510 Paulus Aegineta 11 penetrating neck injury 354 percutaneous endoscopic gastrostomy feeding 579 perfusion-weighted imaging 231–2, 231 peri-infarct depolarization 651 perimesencephalic haemorrhage, nonaneurysmal 461–3, 462, 463, 482–3 periodic respiration 540–1, 540 perioperative stroke 383–4, 383, 384 perivascular inflammation 418 personification 53 petechiae 304 Peto, Richard 28 Petty, William 25 phaeochromocytoma 117 phantom limbs 53 phosphenes 105 photophobia 80, 113 phrenology 27 physiotherapist 527 physiotherapy 538, 581 ‘piano playing’ 60, 65 pituitary apoplexy 467, 473 plaque instability 266–7, 266 plasma cholesterol 305 plasma glucose 305 plasma volume expansion 742 platelet glycoprotein inhibitors 687–8 platelets 285 polycythaemia 373 Polypill 881–2 pons 77 pontine haemorrhage 711, 712 poor positioning 553–5, 554, 554 porcelain-white papules 304 Portal 14 positron emission tomography 237, 645–6 posterior cerebral artery 154–6, 155 posterior circulation infarction 164–5, 164, 165, 294

posterior communicating artery 138–9, 138 posterior fossa decompression 690 posterior fossa haemorrhage 114 posterior inferior cerebellar artery aneurysm 195 posterior leucoencephalopathy syndrome 95 posterior reversible encephalopathy syndrome 94 post-traumatic headache 116–17 pravastatin 815 praxis bedside testing 58 disorders of 57–8, 57, 58 precipitating factors 79 predictive models 228, 516, 518, 678, 703, 704, 793, 795 pregnancy 380–1 pressure ulcers 564–7, 565, 566, 567 prevention 789–901, 966–75, 966 aims of 789–90 anticoagulants 830–41 antiplatelet drugs 816–29 blood pressure reduction 801–9, 964, 969–71, 973 carotid angioplasty and stenting 873–6 carotid endarterectomy 849–73, 964, 965, 971–2 choice of intervention 879–82, 880, 881 cholesterol reduction 809–16 dietary supplements 845–7 evidence 790 extra-to-intracranial bypass surgery 877 involvement of patients and carers 878–9 lifestyle modifications 841–5 management of diabetes/glucose intolerance 847 mass population strategy 972–3, 972 potential impact 973–5, 973, 974 relative and absolute effects of treatment 790 targeted high-risk strategy 967–9, 967, 968, 969 potential impact of 969–72, 970, 972 treatment of specific underlying causes 847–9, 849 treatment of vertebrobasilar ischaemia 877–8 previous head injury 427 primary angiitis 366–7 primary oxalosis 387 primary systemic vasculitis 364–5 prognosis 505–18, 506, 790–801 after intracerebral haemorrhage 800–2 for death 507–9, 507, 508, 508 for dependency 509, 509

early 791–4, 791, 792, 792, 793, 794 individual patients 513–18, 514 early death 513 longer-term outcomes 513 methods of prediction 513–18, 514–16, 517, 518 information gathering 507 likelihood of death 797–8, 797, 798 long-term 794–7, 795, 796 myocardial infarction 799–800 patterns of recovery 509–11, 510 recurrence 798–9, 799 progressive systemic sclerosis 366 PROGRESS trial 804, 805 propantheline bromide 562 propiverine hydrochloride 562 proprioception 60 proprioceptive sensation 63 prosopagnosia 67 prosthetic heart valves 273–4, 273 protein C 306 protein S 306 proximal common carotid artery stenosis 878 pseudoathetosis 60, 65 pseudocalcification 191, 193 psychological disorders 97–9, 599–603, 600 boredom 602–3 dizziness 98 emotionalism 601–2 fatigue 603 functional sensory disturbance 97–8 functional weakness 97 low mood and anxiety 599–601 speech and swallowing symptoms 98 visual symptoms 98–9 ptosis 76, 108 public education 910 public health aspects 953–79 pulmonary embolism 558 anticoagulants 668 aspirin therapy 662 prevention of 706–8 pulmonary oedema 751 pulsatile tinnitus 369, 427 pulses see arterial pulses pupils 72–3, 73 pure motor stroke 160–1, 161 pure sensory stroke 65, 161 purpura 304 purulent cerebritis 209 PWI see perfusion-weighted imaging pyrexia 474 quadrantanopia 50 quality adjusted life years (QALYs) 188 quality of life 503, 939

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 991

Index

Quetelet, Adolphe 25 Quincke 22 radiculopathy 100 ‘ramollissement’ 15 random allocation 655 randomization 655 randomized controlled trials 654–5, 790 blood pressure reduction 803 cholesterol-lowering drugs 810–12, 811, 812, 813 problems with 656 Rankin Handicap Scale 937, 939 rashes 304 Raynaud, Maurice 17 rebleeding 417, 417, 728–35 diagnosis 742–3, 743 early 721–2, 721 emergency occlusion 743–4, 743, 744 endovascular occlusion 729–32, 729–31, 733 pharmacological treatment 735 resuscitation 743 risk of 728–9 surgical occlusion 732, 734, 734 Recognition of Stroke in the Emergency Room (ROSIER) study 44, 45 recurrence of deep intracerebral haemorrhage 417 recurrence of stroke 798–9, 799 recurrent artery of Heubner 142–3, 143 reduction of intracranial pressure 705–6 referral bias 512, 512 reflex sympathetic dystrophy 588 rehabilitation 504, 504, 923–5, 924 therapy-based 924–5, 924 rehabilitation stroke units 918 rehydration 574 relapsing polychondritis 366 relative frequency of causes 435–6 renal carcinoma 196 Rendu–Osler–Weber syndrome 422 reperfusion injury 648 ‘rete mirabile’ 9 retinal disorders 103–6 angle-closure glaucoma 104–5 arteriovenous malformation 104 central/branch retinal vein thrombosis 104, 105 central serous retinopathy 105–6 chorioretinitis 105 intraocular haemorrhage 105 lightning streaks of Moore 105 paraneoplastic retinopathy 105 phosphenes 105 retinal migraine 103–4 retinal infarction 69 retinal migraine 103–4

..

..

rheumatic valvular disease 274 rheumatoid disease 365 risk factors 278–90, 278, 966 age 281, 281 alcohol 286–7 blood lipids 283–5, 283, 284 blood pressure 282–3, 282, 968 diabetes mellitus 283 diet 286 ethnicity 287 genetic studies 287–9 haemostatic variables 285 infection 289 metabolic syndrome 286 non-stroke vascular disease 289–90, 289 obesity 286 physical exercise 286 plasma homocysteine 285 sex 281–2, 282 smoking 283 social deprivation 289 stress 289 risk of haemorrhage 188, 420, 424, 427, 432, 675, 780, 781, 783, 832 Rivermead Mobility Index 604 rofecoxib 382 Roho cushion 566, 567 Rokitansky, Carl Freiherr von 13, 15–16 Rostan, Léon 12, 14, 15 rosuvastatin 815 rotational vertebral artery occlusion 354–5 Russell, Ross 19 Ruysch, Frederik 10 S100-β 186 saccular aneurysms 421–2, 458–61, 459, 767–80 acquired risk factors 461 autosomal dominant polycystic kidney disease 460 conditions associated with 459 Ehlers–Danlos disease type IV 460 familial occurrence 459–60 on feeding arteries genetic studies 460 prevalence 767–71, 768 autosomal dominant polycystic kidney disease 769 development of new aneurysms 769–70, 770 family history 768–9, 769 risk of rupture 771–4, 771–3 screening for 777–80, 778, 779 structural abnormalities 459 treatment options 774–5, 775 endovascular coiling 774–5 neurosurgical clipping 774

triggers for rupture 461 unruptured 775–7 ‘sanguineous apoplexy’ 12, 14 sarcoid vasculitis 366 Scandinavian Stroke Scale 546 Schiller, Friedrich 17 scleroderma 304 secondary brain tumours 196 segmental arterial disorganization see lipohyalinosis selection bias 512 selective serotonin reuptake inhibitors, and hyponatraemia 574 self-audible bruits 300 sensorimotor stroke 162–4, 163 sensory ataxia 76, 77 sensory impairments 586–7 sensory/tactile extinction 53 sentinel headache 112 septic arteritis 428–9 serotonin syndrome 102 ‘serous apoplexy’ 15 Serres 14, 22 severity of stroke 544–5, 545 sex 615–17, 617 shoulder-hand syndrome 590 shoulder pain 588–90, 588, 590, 590, 591 sickle cell disease 374, 468 sick sinus syndrome 277–8 sildenafil 382 ‘silent’ cerebral infarction 310 simvastatin 815, 965 single gene disorders 385–7 single-photon emission computed tomography 237 sinoatrial disease (sick sinus syndrome) 277–8 sinus thrombosis 242–3, 243 see also venous thrombosis sinus of Valsalva aneurysm 278 Siriraj score 185 sixth nerve palsy 75, 475 Sjögren syndrome 365 skin and nails 303, 304 sleep apnoea 540 small vessel disease 267–71, 322, 323 smoking 283 smoking cessation 841–2, 842 Sneddon syndrome 364, 364 Snow, John 25 social deprivation 289 social difficulties 614–17 accommodation 614 air travel 615 employment 614–15 finance 617 leisure 615, 616 sex 615–17, 617

991

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 992

992

Index

social worker 528 softening of the brain 13, 14, 15 somatic sensory disturbance 63–6 clinical anatomy 63–4, 64 clinical assessment 64–5 clinical practice 65–6 somatoparaphrenia 53, 54 spasticity and contractures 579–83, 584 assessment 580–1, 580 prevention and treatment 581–3, 582, 583 speech and language 51–3, 53 language testing 53 speech and language therapist 528 speech symptoms 98 speech (verbal) apraxia 51 spinal arteriovenous malformations 468 spinal artery aneurysm 468 spinal imaging in subarachnoid haemorrhage 489 ‘spiritus animalis’ 11 splinter haemorrhages 304 Spurzheim, JC 27 stair-lifts 610 Staphylococcus aureus 275 statins costs of intervention 965 delayed cerebral ischaemia 741 effect on stroke incidence 964 randomized controlled trials 810–12, 811, 812 see also cholesterol lowering drugs stenting 873–6 stereotactic aspiration with thrombolysis 710 strategies 975–6 stress 289 striatocapsular infarct 143, 145 stroke definition 39–41 diagnosis 46 differential diagnosis 82, 84 neuroimaging 102–3 overlap with transient ischaemic attacks 41 pre-hospital screening tools 42 symptoms and signs 45–6, 46 stroke burden 953–79 costs of treatment 961 effects of treatment 960–1, 961 stroke-in-progression 79, 545 stroke liaison workers 925 stroke prevention see prevention Stroke Prevention in Atrial Fibrillation III risk stratification scheme 273 Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial 284

stroke scales 26–7, 28, 516, 938 stroke services 903–52 aims of 904, 904 comprehensive stroke service 906 cost-effectiveness 944–6, 945, 946 effectiveness of 904–5, 905 evaluation and monitoring 934–43 case mix 941, 942 chance 940–1 measurement of outcome 937–40, 937, 938, 939, 940 outcome 936–7 process 935–6, 936 quality and effectiveness of care 937 generic issues 926–7 hazards of transfer 927 information for patients and carers 926–7, 926 integration of services 927 guidelines 943, 943 impact of 944, 944 improving access to 910–14, 911 coordination of services 911 designated stroke centres 912 emergency department protocols 911 public education programmes 910 telemedicine 912–13 transient ischaemic attack service 913–14 transport protocols 910–11 integrated care pathways 943–4 long-term management 925, 925 maintenance of 927–34 meeting the need 932–3, 933 needs assessment 930–2, 930, 931, 932 needs of population 928–30, 928, 929 organized inpatient (stroke unit) care 914–21 planning and development 906, 933–4 pressures shaping 905–6 principles of 906–10, 907–9 reintegration to normal life 923–5 transfer from hospital to community 921–3, 922, 923 stroke team 525–6, 525, 525, 526 multidisciplinary team meetings 529, 530–1 roles of team members 526–30 stroke units 914–21, 914 length of stay 919–20 outline of care 915–16, 915, 916 planning and development 920–1 size of 920 staffing 920 types of 916–19, 917, 918, 919

structural intracranial lesions 88–92, 88 aneurysms and arteriovenous malformations 89–92, 90–2 electrocephalography 103 head injury 99, 99 neuroimaging 102–3 non-traumatic intracerebral haemorrhage 88 subdural haematoma 88–9 tumour 89 subarachnoid cisterns, blood in 477, 479, 480, 480–3 subarachnoid haemorrhage 21–5, 21–4, 111–22, 457–502, 719–66, 720 antecedent events 472–3 blood pressure 474 causes 457–71, 457, 458, 465 aneurysms of spinal arteries 468 anticoagulants 469 arterial dissection 463–4 arteriovenous malformations 464 cocaine use 468 dural arteriovenous fistulae 466, 468 fusiform aneurysms 464–6 head trauma 469 idiopathic 470–1 intracranial venous thrombosis 467 metastases from cardiac myxoma 468 mycotic aneurysms 467–8 non-aneurysmal perimesencephalic haemorrhage 461–3 pituitary apoplexy 467 saccular aneurysms 458–61 sickle cell disease 468 spinal arteriovenous malformations 468 superficial siderosis of central nervous system 469–70 clinical features 111–14 epileptic seizures 113 focal neurological signs 113–14 headache 111 loss of consciousness 113 neck stiffness 112–13 photophobia 113 precipitating factors 111 speed of onset/disappearance 111–12 subhyaloid haemorrhage 113, 114 systemic features 114 vomiting 112 delayed cerebral ischaemia 735–42 aspirin 739, 741 calcium antagonists 739, 740 cisternal drainage and intracisternal fibrinolysis 742 diagnosis 744–5

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 993

Index

increasing plasma volume 742 induced hypertension and volume expansion 745, 745 magnesium sulphate 739 pathogenesis and risk factors 735–9, 736, 737, 738 statins 741 transluminal angioplasty and vasodilatation 745 diagnosis 21–3, 21–3, 111 differential diagnosis 114–18 acute obstructive hydrocephalus 115 benign exertional headache 117 benign orgasmic cephalalgia 117 benign ‘thunderclap headache’ 117–18 carotid/vertebral artery dissection 115 cerebellar stroke 115 cough headache 116 ‘exploding head’ syndrome 116 hypnic headache 116 idiopathic stabbing headache 116 intracerebral haemorrhage 115 intraventricular haemorrhage 115 meningism 114 meningitis 114–15 migraine 115–16 monoamine oxidase inhibitors 117 occipital neuralgia 117 phaeochromocytoma 117 post-traumatic headache 116–17 epileptic seizures 472 family history 473 focal neurological deficits 474–6, 474 general care 725–8, 727 antiepileptic drugs 728 bladder 727 blood pressure 725–6 confusion 727 conscious level 727 constipation 727 fever 728 fluids and electrolytes 726 headache 726 venous access 728 venous thromboembolism 728 hydrocephalus 746–9 diagnosis 746–7, 746, 747 interventions 747–9, 747 investigations 118–22, 476–89 brain CT scan 118–20, 119, 120, 476–85, 476–85 cerebral catheter angiography 487–9, 489 computed tomographic angiography 487 false positive evidence 120 late presentation 122

..

..

lumbar puncture 120–2, 485 magnetic resonance angiography 487 magnetic resonance imaging 485–7, 486 late sequelae/complications 752–3 cognitive dysfunction 753 epilepsy 752–3 late hydrocephalus 752 psychosocial impact 753 rebleeding 752 vitreous haemorrhage 752 level of consciousness 473–4 loss of consciousness 472 medical history 473 nature of headache at onset 471–2 neck stiffness 474 poor clinical condition acute hydrocephalus and intraventricular haemorrhage 723 acute subdural haematoma 723, 724 cardiopulmonary complications 723, 725 early rebleeding 721–2, 721 intracerebral extension 722–3, 722, 723 pyrexia 474 rebleeding 728–35 diagnosis 742–3, 743 early 721–2, 721 emergency occlusion 743–4, 743, 744 endovascular occlusion 729–32, 729–31, 733 pharmacological treatment 735 resuscitation 743 risk of 728–9 surgical occlusion 732, 734, 734 subhyaloid haemorrhages 474 surgical treatment 23–5, 23, 24 systemic complications 749–51 disorders of cardiac rhythm 750–1, 751 hyperglycaemia 750 hyponatraemia 749–50, 750 pulmonary oedema 751 subclassification aetiological 171–3, 172, 173 clinical 132, 158–71, 159 lacunar syndromes 160–4, 160, 161, 163 Oxfordshire Community Stroke Project classification 168–71, 168–71 partial anterior circulation syndromes 166–7, 167 posterior circulation syndromes 164–5, 164, 165

syndromes of uncertain origin 167–8 total anterior circulation syndrome 165–6, 165, 166 subclavian artery injury 354 subcutaneous heparin 707 subdural empyema 94–5, 96 subdural haematoma 88–9, 90, 443–4, 444, 481–2 acute 723, 724 subgroup analysis 658 subhyaloid haemorrhage 113, 114 superficial sensation 63 superficial siderosis 469–70, 470 superior cerebellar artery 152, 152 supraclinoid internal carotid artery 137–8, 138 supratentorial haemorrhage 709–10 surgical decompression 690 surgical occlusion 732, 734, 734 surgical treatment intratentorial haemorrhage 710–11, 710 supratentorial haemorrhage 709–10 Susac syndrome 367 susceptibility-weighted imaging 213, 218, 442 swallowing problems see dysphagia swallowing symptoms 98 swollen limbs 590–1, 590 Symonds, Charles 22, 23 symptoms and signs 45–6, 46 speed of onset 78–9 syncope 101 syphilis serology 305 systematic reviews 28, 657–8, 658, 790 systemic lupus erythematosus 363 T2*-weighted gradient echo 211, 416 T2*-weighted image 215, 216, 218, 416, 418 tactile extinction 594 TAK-044 742 Takayasu’s arteritis 149, 302, 362–3, 362 telangiectasias 426–7 telemedicine 912–13 temporal arteritis see giant cell arteritis tetraplegia 62 thalamic arterial supply 156–7, 156 thalamic dementia 157 thalamic haemorrhage 186, 216, 438, 710 thalamic infarcts 294 thalamogeniculate arteries 157 thienopyridines 824–6, 825 third nerve palsy 75, 475 three-step model of diagnosis 173 thrombin time 306

993

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 994

994

Index

thromboangiitis obliterans 367 thrombocytopenia 432–3 thromboembolic theory 18–20, 19 thrombolytic therapy 432, 675–87, 708–9 adverse effects 686–7, 686, 687 agent 685 benefits 675 blood pressure management 686 concomitant antithrombotic treatment 686 dose and route 686 evidence for efficacy 675–82 after invasive procedures 681–2 blood pressure management 681 cardioembolic stroke 682 choice of agent 680–1 death 676–8, 677, 678 intra-arterial thrombolysis 681 intracranial haemorrhage 675–6, 676 massive infarct swelling 676 myocardial infarction 675 prognosis vs response 678–9 radiological features 679–80 randomized controlled trials 675 time since stroke onset 679, 679 non-neurological indications rationale 675 risk 675 when to start treatment 685, 685 who not to treat 684–5 who to treat 682–4, 683 thrombophilias 375–6 thrombosis 16–17 thrombotic thrombocytopenic purpura 373 thyroid carcinoma 196, 321 thyroid function tests 306 tibolone 382 ticlopidine 824 tinnitus 76 tirilazad mesylate 741 tizanidine 584 TOAST classification see Trial of Org 10172 in Acute Stroke (TOAST) 171, 172 Todd’s paresis 87, 89 toileting 609–10, 611 tolterodine tartrate 562 ‘top of the basilar syndrome’ 151 tortopia 76 total anterior circulation infarction 165–6, 165, 166, 290–1 tracheal intubation 541 transcortical motor aphasia 51, 52 transcortical sensory aphasia 52 transcranial Doppler ultrasound 245, 295, 318–20

cerebrovascular reactivity 319–20 emboli detection 318–19, 319 intracranial arterial occlusion/stenosis 318 transient episodes 72, 82, 101, 105, 418 transient global amnesia 87–8, 87 transient ischaemic attacks 17–20, 35–41 brain imaging 37–8, 39 costs of long-term treatment 965–6, 965 definition 35–9 diagnosis 82 diagnostic criteria 40 differential diagnosis 82, 84 duration 37, 37, 38 effects of long-term treatment 961–5, 962–4 gradual onset 297, 298 haemodynamic theory 18 MRI diagnosis 38, 39 neuroimaging 102 neurological signs 37 neurological symptoms 45 focal 36 non-focal 36–7, 36 overlap with stroke 41 precipitating factors 297–8 symptoms and signs 45–6, 46 thromboembolic theory 18–20, 19 treatment see treatment vasospasm theory 17–18 vertebrobasilar 317 transient monocular blindness 37, 66, 103, 103–8 causes 104 corneal and eyelid disorders 107–8 Fuch’s cornea endothelial dysfunction 107–8 ptosis 108 optic nerve disorders 106–7 anterior ischaemic optic neuropathy 106–7, 106 malignant arterial hypertension 107 optic disc anomalies 107 optic neuritis 107 papilloedema 107, 107 Uhthoff’s syndrome 107 orbital disorders 107 retinal disorders 103–6 angle-closure glaucoma 104–5 arteriovenous malformation 104 central/branch retinal vein thrombosis 104, 105 central serous retinopathy 105–6 chorioretinitis 105 intraocular haemorrhage 105

lightning streaks of Moore 105 paraneoplastic retinopathy 105 phosphenes 105 retinal migraine 103–4 transient or permanent focal deficit transluminal angioplasty 745 see carotid angioplasty trauma 354–5, 435 cerebral air embolism 355 fat embolism 355 fibrocartilaginous embolism 355 head injury 354 head turning 354–5 non-penetrating neck injury 354 penetrating neck injury 354 subclavian artery injury 354 trauma (‘spät-Apoplexie’) 412, 435 treatment 25–8, 503–35, 635–702 aims of 503–5 ethical issues 531–3 evaluation of evidence 654–9 integrated management plan 518–31 numerical method 25–6 rehabilitation 504 underlying causes 660 variation worldwide 660 which to use 659–60 see also individual drugs Trial of Org 10172 in Acute Stroke Treatment (TOAST) 171, 172 triflusal 829 triplegia 62 tuberous sclerosis 387 tumours 89, 209 Uhthoff’s syndrome 107 ultrasound augmentation of clot lysis 688–9 uncal herniation 654 unstable brain ischaemia 40, 41 unsteady gait 583–6, 585 urea 305 urinalysis 305 urinary incontinence 560–4 assessment 560–1, 561 catheters see indwelling catheters factors contributing to 560 prognosis 561 treatment 561–3, 562, 562, 563 urinary retention 563–4 urinary tract infection 556–7 valdecoxib 382 valve calcification 274 valvular fibroelastoma 277 valvular heart disease 436 van Calcar, Johan Stephaan 9 van Swieten, Gerard 16

..

..

9781405127660_5_ind.qxd 10/13/07 9:44 AM Page 995

Index

van Wesele, Andries 9 vascular imaging 318 vascular risk factors 300 vascular steal 89 vasculitis 359, 360, 435 paraneoplastic 367 primary systemic 364–5 sarcoid 366 vasospasm theory 17–18 venography 325 venous access 728 venous infarcts 209, 241–2, 241 venous malformations 425–6 venous stasis retinopathy 302 venous thrombosis 241–5, 387–90, 388, 467 angiography of veins and sinuses 242–3, 243 CT scanning 243–4, 244 and intracerebral haemorrhage 428 MRI 244–5 subarachnoid haemorrhage 728 transcranial Doppler ultrasound 245 ventriculoperitoneal shunts vertebral artery dissection 115, 463–4, 464 hypoplastic 369 vertebrobasilar insufficiency 149 vertebrobasilar ischaemia 877–8 vertebrobasilar transient ischaemic attacks 317

..

..

vertigo 76, 78 benign paroxysmal positional 96 benign recurrent 96 Vesalius, Andreas 9, 10 Vesling 10 vestibular neuronitis (labyrinthitis) 96 Vicq d’Azyr 14 Vieussens’ receptacle 21 viral encephalitis 208–9 Virchow–Robin spaces 20 Virchow, Rudolf 13, 15 vision 66–72 clinical anatomy 66–7, 66, 67 clinical assessment 67–72 clinical practice 72 visual agnosia 50, 67 visual extinction 594 visual field assessment 71 visual field defects 475 visual hallucinations 72 visual inattention 53 visual neglect 55, 594 visual problems 98–9, 592–3 diplopia and oscillopsia 593 visual field defects 592–3 visual hallucinations 593 visual system disturbance 66–76, 66 external ocular movements 73–6 eyelids 73–6 pupils 72–3, 73 vision 66–72

visuospatial dysfunction 53–7, 593–5, 594 vitreous haemorrhage 752 vomiting 80, 112, 553 von Willebrand factor antigen 285 walking aids 6–7, 605 Wallenberg syndrome 150, 164 warfarin reversal of therapy 674 see also anticoagulants washing and bathing 610–11, 611, 611 Waterlow Scale 566 weakness 59 pyramidal distribution 60 Weber syndrome 151, 164 weight reduction 844–5 Wepfer, Johann Jakob 11, 12 Wernicke’s area 52 Wernicke’s encephalopathy 93 wheelchairs 609 WHO International Classification of Functioning, Disabilities and Health 503, 504 whole-blood viscosity 285, 638–9 whole genome association studies 288 Willis, Thomas 10, 11 Wilson’s disease 94 word-finding difficulty 98 Wren, Christopher 10

995