Landmark papers in neurosurgery 2 edition

Landmark Papers in Neurosurgery

Landmark Papers in... series Titles in the series Landmark Papers in Neurosurgery Edited by Reuben D. Johnson and Alexander L. Green Landmark Papers in Allergy Edited by Aziz Sheikh, Thomas Platts-Mills, and Allison Worth Advisory Editor Stephen Holgate Landmark Papers in Anaesthesia Edited by Nigel R. Webster and Helen F. Galley Landmark Papers in Cardiovascular Medicine Edited by Aung Myat and Tony Gershlick Landmark Papers in General Surgery Edited by Graham MacKay, Richard Molloy, and Patrick Oâ&#x20AC;&#x2122;Dwyer Landmark Papers in Nephrology Edited by John Feehally, Christopher McIntyre, and J. Stewart Cameron

Landmark Papers in Neurosurgery SECOND EDITION Edited by

Reuben D. Johnson, LLB, DPhil, FRCS (Neuro.Surg), MRSNZ

Alexander L. Green, MD, FRCS (SN)

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries

© Oxford University Press 2014 The moral rights of the authors have been asserted First Edition published in 2010 Second Edition published in 2014 Impression: 1 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, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2014930470 ISBN 978–0–19–967402–2 Oxford University Press makes no representation, express or implied, that the drug dosages in this book are correct. Readers must therefore always check the product information and clinical procedures with the most up-to-date published product information and data sheets provided by the manufacturers and the most recent codes of conduct and safety regulations. The authors and the publishers do not accept responsibility or legal liability for any errors in the text or for the misuse or misapplication of material in this work. Except where otherwise stated, drug dosages and recommendations are for the non-pregnant adult who is not breast-feeding Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

This book is dedicated to our families: To my wife Willow, my two daughters Bay and Wyn, and my parents Neil and Susan. Reuben To my wife Caroline, and children Isaac, Benedict, and Dominic. Alex

Foreword to First Edition Joseph C. Maroon While reviewing this volume on Landmark Papers in Neurosurgery, I thought of another book presented to me at the completion of my training at the Radcliffe Infirmary in Oxford—too many years ago. Mr Brian Cummins, then Senior Registrar, regularly thrashed me on the squash courts (although not that skilled!) next to the Radcliffe, when not on call. He shared his winning secret with the presentation of the classic—The Theory and Practice of Gamesmanship or The Art of Winning Games without Actually Cheating by Stephen Potter, another Oxonian. This volume by Reuben Johnson and Alexander Green is the ultimate neurosurgical ‘gamesmanship’ book for whoever reads it. For the resident, seminal large complex investigative studies—‘game changing’ themselves—are analysed, summarized, and then superbly critiqued. The essence of major papers such as the optimal timing for aneurysm surgery, the use of nimodipine, and the indications for decompressive surgery for the management of malignant cerebral infarction can be easily ‘dropped’ on rounds or in conferences with the appearance of having spent hours doing the same meticulous analysis as the authors—without really cheating! For the neurosurgical/neurological faculty, a few quick minutes are all it takes to refresh one’s own database and present a learned discussion or lecture on the history of the use of steroids for cerebral oedema, the rationale for the extent of resection for malignant gliomas, the use of barbiturates in head injury, or how magnesium is neuroprotective in brain injuries. For all in the neurosciences, the authors have distilled thousands of hours of literature review into concise, easily read, and critiqued summaries of papers on head injury, the treatment of spine and spinal cord diseases, and functional neurosurgery that are delightful to read and immediately practical in everyday patient care. Having participated in journal clubs and literature reviews for 25 years, this book, a landmark itself, has the obviousness of so many great innovations that makes the person reading it ask the question—why didn’t I think of this myself? The scientific base and perspective gained from such a critical review of past neurosurgical classics prods the reader to look into the future of our specialty and ask, what’s next? In his letter to Robert Hook on 5 February 1675 Isaac Newton wrote, ‘If I have seen farther, it is by standing on the shoulders of giants’. With these landmark studies the authors have hoisted us high onto the shoulders of neurosurgical giants and the view of the past (as herein beautifully presented), the present, and the future of neurosurgery is breathtaking. Joseph C. Maroon Professor and Heindl Scholar in Neuroscience Department of Neurological Surgery University of Pittsburgh School of Medicine; Team Neurosurgeon, Pittsburgh Steelers Pittsburgh, Pennsylvania, USA

Foreword to First Edition Angelo Franzini Neurosurgery is a lover who becomes ever younger while we grow older. Reading this book, I fell in love again with this young lady. I appreciated the enthusiasm of the authors who have solid experience in the field in reporting new perspectives and leading the reader through the chapters. Clear and understandable guidelines for each of the topics addressed in this book are presented for approaching the problems under discussion. The lectures demonstrate that it takes much longer to understand the indications for a specific surgical procedure than to learn how to perform the hands-on surgical procedure itself. Moreover, the authors remind us that the management of CNS diseases in many instances may be controversial. The choice between more conservative and more aggressive ‘radical’ treatments may be difficult as in the surgical treatment of trigeminal neuralgia or even that of intracranial aneurysms. Reports from up-to-date cooperative studies in the chapter dedicated to neurovascular diseases make this book particularly rich and realistic at the same time. A full range of theories and approaches are clearly examined to ‘educate’ both young and old neurosurgeons. Many of them too often focus their attention on a narrow sub-speciality topic such as cranial base surgery, functional neurosurgery, or spinal surgery. A complete knowledge of all the fields of interest in neurosurgery is clearly not possible, but this book reminds us that many concepts are common and many surgical techniques must be shared among different neurosurgical applications. Endoscopy, neuronavigation, microsurgery, and stereotaxis are common instruments that all neurosurgeons must know in order to offer their patients the best therapeutic options. Travelling through this book, neurosurgeons become aware of the fact that the knowledge of many other disciplines is mandatory in accomplishing that task. Anatomy and physiology are not enough. We need to know functional neuroimaging, physics, genetics, statistics, informatics, and computer technology...read this book. Angelo Franzini Neurosurgeon in Milan

Acknowledgements We would like to extend our thanks to all the contributors to this book and also to the team at Oxford University Press for all their hard work in making this happen. In particular, we would like to thank Peter Stevenson and Eloise MoirFord from Oxford University Press and Papitha Ramesh from Newgen Knowledge Works Pvt. Ltd for helping us complete this project.

Contents Abbreviations Contributors Introduction RD Johnson, AL Green 1 Neurovascular neurosurgery AL Green, RD Johnson, JA Hyam, RSC Kerr 2 Neuro-oncology RD Johnson, AJ Gogos, KJ Drummond, A Taha, KJO Khu, M Bernstein 3 Head injury RD Johnson, F Zhou, T Santarius, JE Wilberger, JV Rosenfeld 4 Spinal surgery RD Johnson, WA Liebenberg, N Maartens, G Barbagallo, M Balsano 5 Functional and epilepsy neurosurgery EAC Pereira, AL Green, RD Johnson, KJ Bulluss, A Astradsson, JA Hyam, TZ Aziz 6 Paediatric neurosurgery RD Johnson, P Richards, J Jayamohan, S Sinha 7 Pituitary surgery RD Johnson, PJ Weir, N Maartens, SA Cudlip, AH Kaye, ER Laws Jr Index

Abbreviations 5-ALA

5-aminolevulinic acid

A.comm

Anterior communicating artery

ACA

Anterior cerebral artery

ACAS

Asymptomatic Carotid Atherosclerosis Study

ALT

Alanine aminotransferase

APACHE

Acute Physiology and Chronic Health Evaluation

ARR

Absolute risk reduction

ASCI

Acute spinal cord injury

ASDH

Acute subdural haematoma

ASIA

American Spinal Injury Association

AVM

Arteriovenous malformation

BCNU

1,3-bis (2-chloroethyl)-1-nitrosourea or carmustine

BCT

Best conventional therapy

BFMDRS

Burkeâ&#x20AC;&#x201C;Fahnâ&#x20AC;&#x201C;Marsden Dystonia Rating Scale

Carcinoma

CAVATAS

Carotid And Vertebral Artery Transluminal Angioplasty Study

CDM

Conventional dose mannitol

CES

Cauda equina syndrome

Cg25

Cingulate gyrus area 25

Cg25WM

Subgenual cingulate white matter

Cluster headache

CCH

Chronic cluster headache

CLEARIVH

Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage

CMM

Conventional medical management

CNS

Central nervous system

CPP

Cerebral perfusion pressure

CRASH

Corticosteroids Randomization After Significant Head Injury

CREST

Carotid Revascularization Endarterectomy versus Stenting Trial

CSDH

Chronic subdural haematoma

CSF

Cerebrospinal fluid

Computed tomography

CTA

Computed tomography angiography

CVP

Central venous pressure

DBI

Diffuse brain injury

DBS

Deep brain stimulation

DECIMAL

Decompressive craniectomy in malignant middle cerebral artery infarction

DECRA

DECompressive CRAniectomy Trial

DESTINY

Decompressive surgery for the treatment of malignant infarction of the middle cerebral artery

DID

Delayed ischaemic deficit

DNT

Dysembryoplastic neuroepithelial tumour

DSM-IV

Diagnostic and Statistical Manual of Mental Disorders, 4th edition

DXT

Deep X-ray therapy

EBRT

External beam radiation therapy

ECOG

Eastern Cooperative Oncology Group

ECST

European Carotid Surgery Trial

ECT

Electroconvulsive therapy

EDH

Extradural haematoma

EEG

Electroencephalogram

EOR

Extent of resection

EORTC

European Organisation for Research and Treatment of Cancer Brain Tumour and Radiotherapy Group

EVA-3S

Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis trial

EVD

Extraventricular drain

FBSS

Failed back surgery syndrome

FIS

Functional Independent Survival

fMRI

Functional magnetic resonance imaging

GBM

Glioblastoma multiforme

GCS

Glasgow Coma Scale

Gastrointestinal

Grey matter

GOS

Glasgow Outcome Scale

GOSE

Glasgow Outcome Scale Extended

GPi

Globus pallidus internus

GSG

Gliadel Study Group

GTR

Gross total resection

HAMLET

Hemicraniectomy after middle cerebral artery infarction with life-threatening oedema trial

HDM

High-dose mannitol

HDRS

Hamilton Depression Rating Scale

HeaDDFIRST

Hemicraniectomy and durotomy on deterioration from infarction-related swelling trial

HHH

Hypertension, hypervolaemia, and haemodilution

HSS

Hypertonic saline solution

HSU

Health State Utility

ICA

Internal carotid artery

ICH

Intracerebral haemorrhage

ICP

Intracranial pressure

ICSS

International Carotid Stenting Study

IPH

Intraparenchymal haematoma

ISAT

International Subarachnoid Aneurysm trial

ISUIA

International Study of Unruptured Intracranial Aneurysms

Intravenous

IVH

Intraventricular haemorrhage

JROSG

Japanese Radiation Oncology Study Group 12

LGG

Low-grade glioma

Lumbar puncture

MCA

Middle cerebral artery

MCS

Motor cortex stimulation

MePred

Methyl prednisolone

MESCC

Metastatic spinal cord compression

MFS

Malignant-free survival

MGMT

O6-methylguanine DNA methyltransferase

MISTIE

Minimally Invasive Surgery plus rt-PA for Intracerebral Hemorrhage Evaluation

mJOA

Modified Japanese Orthopaedic Association scale

MMI

Malignant MCA infarction

MRC

Medical Research Council

MRI

Magnetic resonance imaging

mRS

Modified Rankin Scale

MVD

Microvascular decompression

NASCET

North American Symptomatic Carotid Endarterectomy Trial

NASCIS

National Acute Spinal Cord Injury Study

NCCTG

North Center Cancer Treatment Group

NCICCTG

National Cancer Institute of Canada Clinical Trials Group

NHS

National Health Service

NSAID

Non-steroidal anti-inflammatory drug

NTR

Near total resection

NYIAVMS

New York Islands AVM Study

OCD

Obsessive–compulsive disorder

ODI

Oswestry Disability Index

P.comm

Posterior communicating artery

PAWP

Pulmonary artery wedge pressure

PBTTG

Polymer Brain Tumour Treatment Group

PCA

Posterior cerebral artery

PCO2

Partial pressure of carbon dioxide

PCPC

Paediatric cerebral performance category scale

Parkinson’s disease

PDQ-39

Parkinson’s Disease Questionnaire

PET

Positron emission tomography

PFS

Progression-free survival

PHVD

Post-haemorrhagic ventricular dilatation

Per os

PRCT

Placebo-controlled randomized trial

PRL

Pre-operative prolactin

PROCESS

Prospective Randomized Controlled Multi-centre Trial of the Effectiveness of Spinal Cord Stimulation

PTS

Post-traumatic seizures

QOL

Quality of life 13

QOLIE

Quality of life inventory (Epilepsy)

RCT

Randomized controlled trial

RESCUEicp

Randomized Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intra-Cranial Pressure

REZ

Root entry zone

RPA

Recursive partitioning analysis

Relative risk

RSI

Rapid sequence induction

RTOG

Radiation Therapy Oncology Group

SAH

Subarachnoid haemorrhage

SANTE

Stimulation of the Anterior Nucleus of the Thalamus in Epilepsy

SAPPHIRE

Stenting and Angioplasty with Protection in Patients of High Risk for Endarterectomy

SBI

Sciatica Botherness Index

SCS

Spinal cord stimulation

Standard deviation

SDH

Subdural haematoma

SF-36

Medical Outcomes Study 36-Item Short Form General Health Survey

SIVMS

Scottish Intracranial Vascular Malformation Study

SPACE

Stent-Protected Angioplasty versus Carotid Endarterectomy study

SPECT

Single-photon emission computed tomography

SPORT

Spine Patient Outcomes Research Trial

SRS

Stereotactic radiosurgery

SSRIs

Selective serotonin reuptake inhibitors

STASH

SimvaSTatin in Aneurysmal Subarachnoid Haemorrhage

STICH

International Surgical Trial in Intracerebral Haemorrhage

STIMEP

Assessment of Subthalamic Nucleus Stimulation in Drug Resistant Epilepsy

STN

Subthalamic nucleus

STR

Sub-total resection

SWT

Shuttle-walking test

TBI

Traumatic brain injury

TCD

Transcranial Doppler

Tyrosine hydroxylase

THAM

Tromomethamine

TIA

Transient ischaemic attack

Trigeminal neuralgia

TRISS

Trauma and Injury Severity Score

TSA

Transsphenoidal approach

TTP

Time to progression

UPDRS

Unified Parkinsonâ&#x20AC;&#x2122;s Disease Rating Scale

VAS

Visual analogue scale

Ventral midbrain

VNS

Vagal nerve stimulation

WBRT

Whole brain radiotherapy

WFNS

World Federation of Neurological Surgeons 14

WHO

World Health Organization

White matter

Contributors Arnar Astradsson Copenhagen University Clinic of Neurosurgery Copenhagen, Denmark Tipu Z. Aziz Department of Neurosurgery John Radcliffe Hospital Oxford, UK Massimo Balsano Regional Spine Surgery Department Orthopaedic Department of Thiene and Schio, Vicenza, Italy and University of Verona Verona, Italy Giuseppe Barbagallo Neurosurgical Unit Policlinico University Hospital Catania, Italy Mark Bernstein Department of Neurosurgery Toronto Western Hospital Toronto, Ontario, Canada Kristian J. Bulluss Department of Neurosurgery Austin Health & St Vincentâ&#x20AC;&#x2122;s Hospital Melbourne, Victoria, Australia Simon A. Cudlip Department of Neurosurgery John Radcliffe Hospital Oxford, UK Katharine J. Drummond Department of Neurosurgery The Royal Melbourne Hospital and Department of Surgery University of Melbourne Parkville, Victoria, Australia Andrew J. Gogos Department of Neurosurgery University of Melbourne Parkville, Victoria, Australia Alexander L. Green Department of Neurosurgery John Radcliffe Hospital Oxford, UK Jonathan A. Hyam Department of Neurosurgery National Hospital for Neurology and Neurosurgery Queenâ&#x20AC;&#x2122;s Square London, UK Jay Jayamohan Department of Neurosurgery 16

John Radcliffe Hospital Oxford, UK Reuben D. Johnson Department of Neurosurgery University of Otago Dunedin New Zealand Andrew H. Kaye Department of Neurosurgery Royal Melbourne Hospital Parkville, Victoria, Australia Richard S. C. Kerr Department of Neurosurgery John Radcliffe Hospital Oxford, UK Kathleen Joy O. Khu Section of Neurosurgery Department of Neurosciences Philippine General Hospital Ermita, Manila, Philippines Edward R. Laws Jr Department of Neurosurgery Brigham and Womenâ&#x20AC;&#x2122;s Hospital Harvard Medical School Boston, MA, USA Willem Adriaan Liebenberg Melomed Bellville Hospital Bellville, Western Province, South Africa and Medi Clinic Paarl Paarl, Western Province, South Africa Nicholas Maartens Department of Neurosurgery Alfred Hospital Melbourne, Victoria, Australia Erlick A. C. Pereira Department of Neurosurgery John Radcliffe Hospital Oxford, UK and University of Porto Porto, Portugal Peter Richards Department of Neurosurgery John Radcliffe Hospital Oxford, UK Jeffrey V. Rosenfeld Department of Neurosurgery Alfred Hospital and Department of Surgery Monash University Melbourne, Victoria, Australia Thomas Santarius Department of Neurosurgery Addenbrookeâ&#x20AC;&#x2122;s Hospital Cambridge, UK 17

Saurabh Sinha Sheffield Childrenâ&#x20AC;&#x2122;s Hospital Sheffield, UK Ahmad Taha Department of Neurosurgery Dunedin Public Hospital Otago, New Zealand Philip J. Weir Department of Neurosurgery Belfast Health and Social Care Trust Belfast, UK Jack E. Wilberger Department of Neurosurgery Allegheny General Hospital Drexel University School of Medicine Pittsburgh, Pennsylvania, USA Fei Zhou Department of Neurosurgery Xijing Hospital The Fourth Military Medical University Xiâ&#x20AC;&#x2122;an, China

Introduction The absence of proof does not constitute the proof of absence. Virchow (1880) The correct treatment of a potentially lethal lesion depends upon an accurate knowledge of its natural history. Wylie McKissock (1965)

What constitutes a landmark paper in neurosurgery is highly subjective. In a field as extensive as neurosurgery how can one really pick out a group of studies and say that they are more important than all the others? Any study that has significantly changed the neurosurgical specialty is certainly a landmark study and this will include those studies that have led to a paradigm shift in accepted methods of treating neurosurgical pathology. Studies that affect management decisions for common neurosurgical conditions are also landmark studies. This includes not only clinical trials of treatment modalities but also large epidemiological studies that elucidate the natural history of a disorder. In this volume, we have attempted to include those studies that we feel primarily fall into these two groups. Although the neurovascular, neuro-oncology, and head injury chapters formed the larger chapters in the first addition, this second edition includes a much extended chapter on functional neurosurgery reflecting the rapid expansion of this subspecialty. Furthermore, we have included a new chapter on pituitary surgery which we felt was missing from the first edition. We have included studies that have made a significant attempt to answer an important neurosurgical question even if they have not yet provided a satisfactory answer. In our view at least, these studies should be classified as landmarks, as they will highlight the difficulties of trying to design and carry out such studies in neurosurgical patients. However, in this volume we have not endeavoured to produce a â&#x20AC;&#x2DC;roll of honourâ&#x20AC;&#x2122; of classic studies in neurosurgery, although many such studies have been referenced in the introduction to chapters and in the critiques of individual studies. In this way, there will, at the very least, be a doffing of our caps to such classic studies even if they have not been overtly included in this volume. Furthermore, although this volume is a collection of critiques of landmark studies rather than an evidence-based review of neurosurgery per se, we feel it is useful, where appropriate to stratify studies into the evidence class they represent. There are numerous methods to stratify clinical evidence and in this volume we have adopted the three-tier classification of evidence for therapeutic effectiveness based on that endorsed by the American Association of Neurological Surgeons and the Congress of Neurological Surgeons. This consists of three tiers as follows: Class I

Well-designed, randomized, controlled trial

Class II Well-designed comparative clinical study, e.g. non-randomized cohort study or a case-control study Class III Case series, comparative study with historical controls, or case reports This classification system gives an assessment of the degree of certainty regarding the results of a study. The certainty is proportionate to the class of study with Class I, II, and III reflecting high, moderate, and unclear certainties respectively. It should be noted that we are applying this scheme to individual studies in order to make our assessment as to what class of evidence the study represents. We would also emphasize that the classification of each study is based on our own assessment. We would, for example, classify a randomized trial that did not include a power calculation as a Class II study rather than a Class I study. There are some studies that, although arguably landmark studies in the field of neurosurgery, have been omitted, which will need some brief explanation. We have avoided, for example, as much as possible case series that describe important advances in surgical technique. The reason for this is because it is specifically not our aim to chart the development of surgical techniques. In addition, we feel that surgical technique is also a matter of apprenticeship and schooling. The surgical methods a surgeon uses will be a product of his or her individual training and those aspects of their mastersâ&#x20AC;&#x2122; craft they found most effective in their hands. This is part of the surgical art that is still a matter for the individual, and rightly so. The diversity in neurosurgical techniques is part of the rich tapestry of our specialty. However, we have included case series that have influenced neurosurgical practice. A case series may illustrate an important issue about the timing of surgery or provide valuable information about long-term outcomes following surgical intervention. This volume is intended to be an informative tool to neurosurgical trainees and a useful review for practising senior neurosurgeons. Our overall aims are twofold. Firstly, to provide a succinct review and critique of the published studies. In many ways, this volume could be said to represent a minimum corpus sapiendi of the larger and more influential studies in neurosurgery. This we hope will act as an introduction to the literature to trainees and in particular those coming up to exams. We have, therefore, included here a brief discussion on the issues of trial design 19

and also a brief explanation of some of the more common terms used in clinical trials. We also hope that this volume may be of interest to established neurosurgeons who have lived through the development of our specialty into what it is today. We have primarily focused on the main results of studies that we feel are the most important and relevant. This usually equates to the primary outcome data in most studies. However, we have included secondary outcomes where these address key questions or dilemmas that need to be highlighted for completion. Similarly, we have been fairly unforgiving of post hoc analyses, but have endeavoured to include these where they provide important insights or may form the basis for further studies or trial designs. In this way, we may be criticized for giving an incomplete view of the findings of some studies. This is not our intention, but rather to highlight the take-home messages of the study so that they can be more easily remembered by the reader. In addition, we would emphasize that these critiques are our own interpretation of the studies and a review of published criticisms and we hope that these prove a valuable starting point for further reading. Reuben D. Johnson and Alexander L. Green

Chapter 1

Neurovascular neurosurgery AL Green, RD Johnson, JA Hyam, RSC Kerr 1.0 Introduction 1.1 Timing of aneurysm surgery 1.2 Radiological predictors of vasospasm 1.3 Endovascular coiling versus aneurysm clipping in ruptured aneurysms 1.4 Long-term natural history of unruptured aneurysms 1.5 Nimodipine for prophylaxis of cerebral vasospasm in aneurysmal subarachnoid haemorrhage 1.6 ‘HHH’ therapy for vasospasm 1.7 Statins in the prevention of vasospasm in aneurysmal subarachnoid haemorrhage 1.8 Treatment of cerebral arteriovenous malformations 1.9 Surgery for spontaneous intracerebral haematomas 1.10 Decompressive surgery for malignant cerebral artery infarction 1.11 Carotid endarterectomy for carotid stenosis 1.12 Carotid endarterectomy versus carotid stenting for carotid stenosis

1.0 Introduction The demonstration of a low mortality-rate by a new technique is of no value until an acceptable statistical method of assessment of the natural prognosis and of the proposed treatment is available. Wylie McKissock (1965)

The first clinical trial performed in the field of neurovascular surgery was carried out by Wylie McKissock, Alan Richardson, and Lawrence Walsh at Atkinson Morley’s Hospital, St George’s, in London between 1958 and 1965, when they compared conservative and surgical treatment of anterior communicating aneurysms (McKissock et al., 1965). The results of the trial did not show any difference in mortality between the two groups. The surgical methods used varied throughout the trial to include aneurysmal clipping, ligation of the proximal anterior cerebral artery, wrapping the aneurysm with muslin, and ligation of the common carotid artery. Although the results did not show any benefit from surgical intervention, this single-centre study stands out as a landmark in neurosurgery for several reasons. Firstly, it is the first attempt at a randomized controlled trial (RCT) of surgical management of a common neurosurgical disorder. Secondly, the author’s rationale for carrying out the trial was not just to evaluate surgery but to further elucidate the natural history of the lesion being treated. This point is of particular importance when considering current dilemmas that face the neurosurgeon, such as the management of unruptured intracranial aneurysms and spontaneous non-aneurysmal haematomas. Thirdly, the authors emphasize the limitations of carrying out singlecentre studies in neurosurgery and indicate the need for large multi-centre studies. We have been highly selective in the studies included in this chapter and have included those studies which we feel remain true to the founding principles of the McKissock trial. The first sections of this chapter deal with aneurysmal subarachnoid haemorrhage (SAH) and we have included studies which examine the natural history of aneurysms and their treatment. We have included, therefore, the International Cooperative Study by Kassell which addresses the timing of aneurysm surgery and prognostic factors associated with good and poor outcomes. We have also included the large study of unruptured aneurysms by the International Study of Unruptured Intracranial Aneurysms Investigators (ISUIA) which is an ongoing study addressing one of the most important and controversial problems facing neurovascular surgeons. The International Subarachnoid Aneurysm Trial (ISAT) has been included because of the widespread and profound influence this study has exerted on the management of ruptured intracranial aneurysms. As vasospasm is the greatest cause of neurological morbidity in patients who survive their primary aneurysmal bleed we have also included studies of strategies for the prediction, prevention, and treatment of vasospasm associated with aneurysmal SAH. We have included strategies for which there is an accepted evidence base, such as nimodipine, and which have become accepted practice, such as ‘HHH’ (hypertension, hypervolaemia, and haemodilution) therapy. In addition, we have included some studies which evaluate strategies that have been promising and for which the results of ongoing trials are keenly awaited, e.g. the use of statins. In a subsequent section of the chapter, the treatment of arteriovenous malformations (AVMs) is considered. The decision to treat AVMs depends on balancing the risk of treatment versus the natural history of AVMs. Unfortunately, the natural history of AVMs has been extremely difficult to elucidate, although there are two ongoing population-based studies in Scotland and New York. Nonetheless, the Spetzler–Martin grading system is a landmark in neurosurgery as it has produced an objective system by which 21

outcome and risks of treatment can be applied to individuals with AVMs and for this reason it has been considered (Spetzler and Martin, 1986; Hamilton and Spetzler, 1994). Further sections of this chapter deal with the surgical management of spontaneous non-aneurysmal intraparenchymal haemorrhages (STICH I + II trials) and decompressive surgery for the management of malignant middle cerebral artery (MCA) infarction (DESTINY, DECIMAL, HAMLET, and HeaDFIRST trials). The final sections deal with the role of surgery in the management of carotid artery stenosis (NASCET and SPACE trials). There are, of course, some studies which many would consider to be conspicuous by their absence in this chapter. For example, there are some early classic studies demonstrating angiographic vasospasm, and determining the clinical manifestations and time course of vasospasm (Ecker and Riemenschneider, 1951; Stornelli and French, 1964; Fisher et al., 1977; Weir et al., 1978). There are numerous other examples that could have been included. However, we have endeavoured to keep each chapter concise and to include a selection of the largest published studies that cover most of the areas of neurovascular surgery which are relevant to the everyday practice of all neurosurgeons.

References Clinical Effectiveness Unit. National Study of Subarachnoid Haemorrhage. London: The Royal College of Surgeons of England, 2006. Ecker A, Riemenschneider PA. Arteriographic demonstration of spasm of the intracranial arteries, with reference to saccular arterial aneurysms. J Neurosurg 1951; 8: 660–667. Fisher CM, Roberson GH, Ojemann RG. Cerebral vasospasm with ruptured saccular aneurysm—the clinical manifestations. Neurosurgery 1977; 1: 245–248. Hamilton MG, Spetzler RF. The prospective application of a grading system for arteriovenous malformations. Neurosurgery 1994; 34: 2–6. McKissock W, Richardson A, Walsh L. Anterior communicating aneurysms: a trial of conservative and surgical treatment. Lancet 1965; 1: 873–876. Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg 1986; 65: 476–483. Stornelli SA, French JD. Subarachnoid hemorrhage-factors in prognosis and management. J Neurosurg 1964; 21: 769–780. Weir B, Grace M, Hansen J, Rothberg C. Time course of vasospasm in man. J Neurosurg 1978; 48: 173–178.

1.1 Timing of Aneurysm Surgery Details of Study The International Cooperative Study on the Timing of Aneurysm Surgery was the first large-scale study to look at this issue. Between December 1980 to July 1983 a total of 3521 patients were recruited out of 8879 patients with SAH. In addition to looking at the timing aspect of surgery, many other factors that influence outcome were addressed.

Study References Main Study There are two main references for the study: part 1 (overall management results) and part 2 (surgical results). Both are reviewed here. Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongable GL. The International Cooperative Study on the Timing of Aneurysm Surgery. Part 1: overall management results. J Neurosurg 1990; 73: 18–36. Kassell NF, Torner JC, Jane JA, Haley EC Jr, Adams HP. The International Cooperative Study on the Timing of Aneurysm Surgery. Part 2: surgical results. J Neurosurg 1990; 73: 37–47.

Related References Graff-Radford NR, Torner J, Adams HP Jr, Kassell NF. Factors associated with hydrocephalus after subarachnoid hemorrhage. A report of the Cooperative Aneurysm Study. Arch Neurol 1989; 46: 744–752. Kassell NF, Torner JC. The International Cooperative Study on Timing of Aneurysm Surgery—an update. Stroke 1984; 15: 566–570.

Study Design Class of evidence

Randomization

None (see following text)

Number of patients

3521 (2922 had aneurysm surgery)

Length of follow-up

6 months

Number of centres

68 in 14 countries (24 in USA)

Stratification

Age Sex Presence of hypertension 22

Site and size of aneurysm Aim of the study was twofold; firstly to define the relationship between timing of aneurysm surgery and outcome, secondly to document current medical and surgical management in a number of centres around the world. It was a prospective, observational, epidemiological survey. Assessments were performed by a neurologist and blinded to the timing of surgery. All patients admitted to each participant centre were enrolled, with four ‘logs’ completed for each—‘SAH log’, ‘registration form’, ‘treatment form’, and ‘follow-up form’. Inclusion criteria: admission ≤3 days since first SAH from a saccular aneurysm (computed tomography scan/lumbar puncture (CT/LP) confirmation of bleed, angio/surgical confirmation of aneurysm). Exclusion criteria: delayed admission >3 days since bleed; multiple bleeds; no confirmation of aneurysm. A large number of patients were excluded for other reasons such as evacuation of haematoma, non-participating surgeon, lack of patient/carer consent, etc. but these are not listed as exclusion criteria per se.

Outcome Measures Primary Endpoints ‘Good result’ or death as defined by the Glasgow Outcome Scale (GOS). Neurological examination.

Secondary Endpoints Pre-, intra-, and post-operative complications.

Results Many demographic results including the age, sex, site and size of the aneurysm(s), and the presence of pre-existing hypertension are reported in the results. In 51% of patients, surgery was performed on day 0–3. About 75–80% of patients were considered in ‘good condition’ at the time of admission but at 6 months, only 58% had recovered to their premorbid state without neurological deficit. Nine per cent were moderately disabled, 5% severely disabled, 2% vegetative, and 26% died. Leading causes of death or disability, in descending order, were vasospasm (13.5%), direct effect of the bleed on brain parenchyma (10.6%), re-bleeding (7.5%), operative complications (4%), intracerebral haemorrhage (2%), hydrocephalus (1.7%), and other less common causes. The most important results are probably the prognostic factors, as determined by a univariate analysis. These included: Level of consciousness (p <0.001): 75% who were alert on admission had a good recovery, compared to 11% who were comatose. Age inversely related to outcome (26% between 70 and 87 years had a good outcome). No significant sex differences. Smaller aneurysms (<12 mm) had more favourable results. Outcome better if middle cerebral or internal carotid aneurysm (compared to vertebrobasilar or anterior circulation). Other good prognostic indicators included lower admission blood pressure, clot distribution on CT, absence of pre-existing medical conditions, absence of vasospasm, admission motor response, and orientation. In addition to these results, a number of medical conditions such as pneumonia, cardiac disturbances, gastrointestinal (GI) bleeding, etc. were identified to commonly occur after admission. There was a considerable difference in outcome and mortality (death ranged from 0% to 66%) between centres—use of the Chi squared test determined that this was not related to activity of the individual centres.

Surgical Results At 6 months, 69% who had surgery had a good result, versus 14% dead. Compare this to the 58% good recovery overall. This effect was strongly related to age (90% good result in the 18–29 years age group versus 56% in the 60–69 years age group). Factors associated with good surgical outcome were similar to the overall prognostic factors. Patients who were alert pre-operatively had a more favourable prognosis (overall) if their operation occurred between days 0–3 or after day 10. Operatively mortality, however, only reduced after day 10. Patients who were drowsy pre-operatively had better outcomes when operated after day 10.

Conclusions The main conclusions are that 75% of those admitted within 3 days are in good condition, with a 58% good recovery at 6 months, and 25% death rate. Vasospasm and re-bleeding were the major causes of death or disability, aside from the initial effects of the bleed. There were a number of prognostic factors including admission Glasgow Coma Scale (GCS) and age. The study concluded that there is considerable room for improvement.

Critique This study was performed at a time when most neurosurgeons opted to wait several days after a SAH before operating. At this point, there was little doubt that operative results were better—the patient was medically stable, and the brain was less swollen and friable. The study was really a response to the question of whether the overall management results were better, i.e. by delaying surgery, some patients would suffer re-bleeds and others may suffer vasospasm that could not be adequately treated in the presence of an unsecured aneurysm. The study also sought to look at the epidemiological and prognostic factors in these patients, and was the largest study of its time. In this sense, the study was a well-designed epidemiological study and served as a preliminary to RCTs (although these came over 20 years later). Perhaps one of the criticisms of the study is the length of follow-up which was limited to 6 months. Patients with neurological deficits can still show improvement after this time, although the differences from a longer follow-up would probably be small. This study had a very large impact on the management of patients with suspected aneurysmal SAH. It confirmed that patients with poor grade and older age, with pre-existing medical conditions have a very poor prognosis, and that these patients should not be operated on before day 10. It also had a profound effect on the timing of surgery of good grade patients, prompting an international change in practice to early surgery by day 3. The main difference today is that we now have the option of endovascular treatment. This is often performed on poorer grade patients within the vasospastic period, largely because it is less risky to do so. However, as a large number of aneurysms are still treated surgically, this study still has great relevance. The timing of intervention for aneurysmal SAH remains controversial. Of particular note is a series of 391 patients from the Alfred Hospital in Melbourne who underwent surgery within 24 h following their initial bleed (Laidlaw and Siu, 2002). In this series, 83% of patients with good clinical grades had good outcomes with early surgery. In addition, in this case series, only 15% of patients with poor clinical grades had a poor outcome.

Reference Laidlaw JD, Siu KH. Ultra-early surgery for aneurismal subarachnoid hemorrhage: outcomes for a consecutive series of 391 patients not selected by grade or age. J Neurosurg 2002; 97: 250–258.

1.2 Radiological Predictors of Vasospasm Details of Study The Fisher Scale is commonly quoted in cases of aneurysmal SAH. The original paper is a landmark as it was an attempt to identify which patients would be higher risk for the development of vasospasm and delayed neurological deficits based purely on radiological factors. This was an observational study of 47 patients using presenting CT scan features to predict vasospasm measured by cerebral angiography.

Study References Main Study Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid haemorrhage visualized by computerized tomographic scanning. Neurosurgery 1980; 6(1): 1–8; discussion 8–9.

Related References Frontera JA, Claassen J, Schmidt JM, Wartenberg KE, Temes R, Sander Connolly E, Loch Macdonald R, Mayer SA. Prediction of symptomatic vasospasm after subarachnoid haemorrhage: the modified Fisher scale. Neurosurgery 2006; 58(7): 21–27. Reilly C, Amidei C, Tolentino J, Jahromi BS, MacDonald RL. Clot volume and clearance rate as independent predictors of vasospasm after aneurysmal subarachnoid hemorrhage. J Neurosurgery 2004; 101: 255–261. Smith ML, Abrahams JM, Chandela S, Smith MJ, Hurst RW, Le Roux PD. Subarachnoid hemorrhage on computed tomography scanning and the development of cerebral vasospasm: the Fisher grade revisited. Surg Neurol 2005; 63: 229–234; discussion 234–225.

Study Design Class of evidence

III

Randomization

None. Observational 24

Number of patients

Outcomes

Primary outcome: Angiographic evidence of vasospasm Secondary outcome: Clinical signs of vasospasm

Number of centres

47 cases were analysed retrospectively at Massachusetts General Hospital, United States. Inclusion criteria: • SAH with at least one aneurysm demonstrated on angiography. • CT head performed within first 5 days after SAH. • Angiography had been performed between days 7 to 17 after SAH.

Outcome Measures Primary Endpoint Incidence of vasospasm on cerebral angiography (none, slight–moderate, severe, particular to each vessel, e.g. <2 mm for intradural internal carotid artery).

Secondary Endpoint Incidence of clinical signs of vasospasm, correlated to the relevant arterial territory, e.g. hemiparesis, aphasia with middle cerebral artery (MCA); abulia, incontinence, drowsiness with anterior cerebral artery (ACA).

Results

Although two patients with no SAH (Group I) on CT did develop angiographic severe vasospasm, none developed clinical vasospasm. Patients in Groups II and IV had no incidence of angiographic or clinical vasospasm; however, their numbers were small (n = 7 and n = 5, respectively). Twenty-three of 24 patients with localized clot and/or vertical thick layers of subarachnoid blood >1mm thickness (Group III) had angiographic and clinical signs of vasospasm. Correspondence between location of thick subarachnoid blood and site of severe vasospasm was almost exact. No statistical analysis was performed between groups.

Conclusions Localized clot and/or vertical thick layers of subarachnoid blood >1 mm thickness (Group III) was highly predictive of development of angiographic and clinical vasospasm after aneurysmal SAH.

Critique This is a small, retrospective, observational, single-centre study without statistical analyses, and therefore is limited by multiple potential sources of bias. However, it provided important progress at the time in the understanding of what determines outcomes in SAH patients and correlation of radiological and clinical features. Quantification of objective data to predict outcomes in neurosurgical disease was advancing around the time of this paper and followed closely behind the development of the Glasgow Coma Scale. Further, CT was a relatively new diagnostic tool in 1980 so this paper represents the period of advancement in the modern science and practice of mainstream neurosurgery. In Lyndsay Simon’s critique of the paper, he describes as ‘remarkable’ their ‘indication of a high degree of association between vasospasm and opacification of the basal cisterns’ (Fisher et al., 1980, Discussion). This is now something that clinicians treating these patients take for granted. The radiological evaluation was thorough. Radiological vasospasm was judged on cerebral angiography, still the gold-standard test of vascular arterial architecture. Further, they also considered clinical symptoms of vasospasm 25

which is ultimately the most important outcome for patients. The paper itself is also instructive with careful listing and descriptions of the cisternal anatomy and rigorous disclosure of the luminal measurements they considered as mild, moderate, or severe radiological vasospasm. The authors suggested that the subjective differences between each group made the results of this paper merely ‘preliminary’. Accordingly, the Fisher grading has been revisited since in a much larger number of patients. Two studies found that the Fisher scale did not significantly correlate with development of symptomatic vasospasm (Reilly et al., 2004; Smith et al., 2005). Frontera et al. developed a Modified Fisher Scale using 1355 SAH patients from the placebo arm of the tirilazad trial, 451 (33%) of which developed clinical vasospasm (Frontera et al., 2006). They reassigned the groups according to whether the SAH was thin or thick and whether it was associated with intraventricular haemorrhage (IVH): (I) focal or diffuse thin SAH, no IVH; (II) focal or diffuse thin SAH, with IVH; (III) thick SAH, no IVH; (IV) thick SAH, with IVH. These groups had increasing odds ratios (ORs) for clinical vasospasm from 1.6 in Grade II, 1.6 for Grade III, to 2.2 for Grade IV. Other risk factors they identified were history of hypertension, early angiographic vasospasm, neurological grade, and elevated admission mean arterial blood pressure. Understandably, with larger studies and more common CT usage it has been possible to refine the radiological predictors of vasospasm. To its credit, the grades describing thick SAH still conferred a higher OR, as predicted by the original Fisher study. Fisher’s paper is a landmark from which our understanding of the relationship between radiological and clinical features in aneurysmal SAH has guided modern practice.

References Frontera JA, Claassen J, Schmidt JM, Wartenberg KE, Temes R, Sander Connolly E, Loch Macdonald R, Mayer SA. Prediction of symptomatic vasospasm after subarachnoid haemorrhage: the modified Fisher scale. Neurosurgery 2006; 58(7): 21–27. Reilly C, Amidei C, Tolentino J, Jahromi BS, MacDonald RL. Clot volume and clearance rate as independent predictors of vasospasm after aneurysmal subarachnoid hemorrhage. J Neurosurgery 2004; 101: 255–261. Smith ML, Abrahams JM, Chandela S, Smith MJ, Hurst RW, Le Roux PD. Subarachnoid hemorrhage on computed tomography scanning and the development of cerebral vasospasm: the Fisher grade revisited. Surg Neurol 2005; 63: 229–234; discussion 234–225.

1.3 Endovascular Coiling versus Aneurysm Clipping in Ruptured Aneurysms Details of Study The International Subarachnoid Aneurysm Trial (ISAT) is the most comprehensive study comparing endovascular to surgical treatment in ruptured aneurysms. It has had a greater impact on treatment of ruptured aneurysms than any other study to date.

Study References There is an initial study with 1-year follow-up looking at primary endpoints and several ‘spin-offs’ looking at secondary outcome measures.

Main Study Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, Holman R, for the International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group. 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– 1274.

Related References Clinical Effectiveness Unit. National Study of Subarachnoid Haemorrhage. London: The Royal College of Surgeons of England, 2006. Molyneux A, Kerr R; International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group, Stratton I, Sandercock P, Clarke M, Shrimpton J, Holman R. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomized trial. J Stroke Cerebrovasc Dis 2002; 11: 304–314. Molyneux AJ, Kerr RS, Yu LM, Clarke M, Sneade M, Yarnold JA, Sandercock P, for the International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group. 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: 809–817. Molyneux AJ, Kerr RSC, Birks J, Ramzi N, Yarnold J, Sneade M, Rischmiller J, for the ISAT collaborators. Risk of recurrent subarachnoid hemorrhage, death, or dependence and standardized mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurol 2009; 8(5): 427–433. Scott RB, Eccles F, Molyneux AJ, Kerr RSC, Rothwell PM, Carpenter K. Improved cognitive outcomes with endovascular coiling of ruptured intracranial aneurysms: Neuropsychological outcomes from the International Subarachnoid Aneurysm Trial

(ISAT). Stroke 2010; 41: 1743–1747. Wolstenholme J, Rivero-Arias O, Gray A, Molyneux AJ, Kerr RS, Yarnold JA, Sneade M, for the International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group. Treatment pathways, resource use, and costs of endovascular coiling versus surgical clipping after aSAH. Stroke 2008; 39: 111–119.

Study Design Placebo-controlled randomized trial (PRCT). Class of evidence

Randomization

Non-blinded coiling versus clipping

Number of patients

2143 (1073 coiled, 1070 clipped)

Length of follow-up

Primary outcomes: 1 year Secondary outcomes: Ongoing

Number of centres

43 (centres treating 60–200 cases per year)

Stratification

Age Sex World Federation of Neurosurgeons (WFNS) grade Aneurysm size and location Extent of blood on CT scan

Patients were randomized after admission to a neurosurgical unit and after initial angiography. Out of 9559 patients assessed for eligibility, 2143 were deemed suitable for randomization. Inclusion criteria: SAH within 28 days (CT or LP proven); presence of aneurysm (proven by computed tomography angiogram (CTA) or formal angiogram); good enough clinical state to justify treatment; aneurysms judged to be suitable for either technique (opinion of both surgeon and neuroradiologist) with equipoise regarding which method would be best; consent. Exclusion criteria: >28 days since SAH; clinical condition considered unsuitable for either or both treatments; lack of consent; participation in another SAH trial.

Outcome Measures Primary Endpoints Incidence of death or dependency; modified Rankin Scale (mRS).

Secondary Endpoints Subgroups—WFNS grade at randomization, age, Fischer grade, lumen size of aneurysm, aneurysm site. Incidence of re-bleeding from the treated aneurysm. Quality of life at 1 year. Frequency of epilepsy. Cost-effectiveness. Neuropsychological outcomes. Results of follow-up angiography.

Results Of patients who underwent endovascular coiling, 23.7% were dependent or dead at 1 year compared to 30.6% who had their aneurysm surgically clipped (p <0.002). This led to a relative/absolute risk reduction of dependency or death at 1 year of 22.6%/6.9% respectively. Re-bleeding risk at 1 year was 2 per 1276 patient-years in the endovascular group versus 0 per 1081 patient-years in the surgical group (not significant).

Conclusions The outcome, in terms of survival-free disability, is significantly better with endovascular treatment than with surgical clipping of a ruptured aneurysm.

Critique There is no doubt that the ISAT trial has generated a large amount of controversy amongst neurosurgeons and interventional radiologists alike. Whatever the criticism, it is one of the few multi-centred, randomized trials in neurosurgery and most would admit that it is of considerable importance. However, some have argued that the results have been over-interpreted. One of the major criticisms is that it compares good interventional neuroradiologists to ‘average’ neurosurgeons rather than those who ‘concentrate’ on neurovascular surgery. In other words, there is an inherent bias in the recruiting centres as being those that have a strong interventional radiology interest. The ISAT group have countered this by stating that the trial is a ‘pragmatic’ trial. That is, it tries to determine the best outcome for a patient, in a real-life situation who would be transferred to their regional unit for diagnosis and treatment. It is not a trial of ‘the best possible surgery versus the best possible endovascular treatment’ but a trial of what is the best option for an ‘average patient’. A second criticism of the trial is the randomization process. The trial is biased towards small anterior circulation aneurysms (97.5%). To be fair, the ISAT investigators have never claimed that the trial indicates that all ruptured aneurysms should be coiled in preference to clipping. But some people have perhaps interpreted it thus. Also concerned with the randomization process is that the average time to randomization was slightly longer in the surgical group (1.7 versus 1.1 days) and this may have led to slightly worse outcomes in this group. Since the numbers of rebleeds takes into account those that happen after randomization but before treatment, this may have led to a worsening of results in the surgical group. Since the analysis is based on an intention-to-treat paradigm, some patients allocated endovascular treatment received clipping (for a variety of reasons including patient choice) and 38 allocated clipping crossed over to the endovascular group. However, the analysis is based on the original randomization choice and this has received some criticism in the literature. Probably the most important criticism of the ISAT trial is that the primary endpoints were measured at 1 year and not subsequently. Therefore, the trial shows that in the initial phase, endovascular coiling may be better than surgical treatment. But does this necessarily translate into the long term? There is some evidence from early analysis of the secondary data that, in fact, the surgical group may just be slower to recover and that there is some improvement in mRS with time. Also, there is the issue of long-term re-bleed rates. The late re-bleed rate in the endovascular group is 0.21% per patient-year compared to 0.063% in the surgical group. This, coupled with the fact that the poor outcome at 1 year is much less than in the surgical group of patients <40 years of age, has led some ISAT investigators to suggest that surgery may be better in this age group. This issue has been revisited by the original investigators using longer follow-up (range 6–14 years, mean 9 years) of the ISAT cohort (Molyneux et al., 2009). Thirteen re-bleeds occurred overall after the first year from treated aneurysms: ten after coiling and three after clipping (p = 0.06 in the intention-to-treat analysis but significant when analysed according to actual treatment received, p = 0.02). This is an important result for several reasons. Firstly, it does confirm suspicions that aneurysms are more likely to re-bleed after coiling compared to clipping. However, the overall risk of death 5 years after treatment was still significantly lower in the coiling group (11% versus 14%, p = 0.03). Secondly, this re-bleed result also shows that clipped aneurysms do not confer a subsequent re-bleed rate the same as the general population, which has been assumed by many surgeons. Whether this statistic improves with the increased specialization among neurovascular surgeons can be speculated upon. The neuropsychological outcomes from ISAT were published in 2010 (Scott et al., 2010). 612 patients from the eight participating UK centres were followed up at 12 months with neuropsychological testing if there was no major physical disability (mRS 0–2). Cognitive impairment was lower in the endovascular coiling group compared to 28

clipping (27% versus 39%, p = 0.0055), as was incidence of epilepsy. Again, many of the limitations of the trial are relevant to the interpretation of this result. In particular, as the vast majority of aneurysms were at anterior cerebral/anterior communicating arteries, clipping would entail some frontal lobe retraction and potentially gyrus rectus dissection, increasing the likelihood of cognitive deficits due to the aneurysm location-specific factors. Whether this is the case for aneurysms on other arteries is not answered by these results. The authors have acknowledged the caution that should be taken in extrapolating the ISAT findings beyond the lesions studied. There is concern that the positive ISAT results have resulted in endovascular treatment being given the right to first refusal for all aneurysms and all patients (Darsaut et al., 2013). Although advances in endovascular techniques and technology have occurred since the interruption of ISAT in 2002, whether this translates into better outcomes for all ruptured aneurysm patients has not been proven. Accordingly, the ISAT investigators have initiated ISAT Part II. This is another pragmatic RCT involving at least 50 international centres over 10–12 years aiming to address these questions. Whatever the criticisms, ISAT has provided a wealth of useful data that will continue to be analysed for years to come. The trial has led to a huge shift from surgery to endovascular treatment in some centres, particularly in the United Kingdom and France. Time will tell whether this shift has been appropriate.

References Darsaut TE, Jack AS, Kerr RS, Raymond J. International subarachnoid aneurysm trial—ISAT Part II: study protocol for a randomized controlled trial. Trials 2013; 14: 156. Molyneux AJ, Kerr RSC, Birks J, Ramzi N, Yarnold J, Sneade M, Rischmiller J, for the ISAT collaborators. Risk of recurrent subarachnoid hemorrhage, death, or dependence and standardized mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurol 2009; 8(5): 427–433. Scott RB, Eccles F, Molyneux AJ, Kerr RSC, Rothwell PM, Carpenter K. Improved cognitive outcomes with endovascular coiling of ruptured intracranial aneurysms: Neuropsychological outcomes from the International Subarachnoid Aneurysm Trial (ISAT). Stroke 2010; 41: 1743–1747.

1.4 Long-Term Natural History of Unruptured Aneurysms Details of Study The International Study of Unruptured Intracranial Aneurysms Investigators (ISUIA) was the first large-scale, prospective study looking at the natural history of unruptured aneurysms as well as the risks of treatment of unruptured aneurysms. Factors related to prognosis are elucidated.

Study References Main Study Wiebers DO, Whisnant JP, Huston J 3rd, Meissner I, Brown RD Jr, Piepgras DG, Forbes GS, Thielen K, Nichols D, O’Fallon WM, Peacock J, Jaeger L, Kassell NF, Kongable-Beckman GL, Torner JC, for the International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362: 103–110.

Related Reference ‘ISUIA 1’ (a retrospective study): International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms—risk of rupture and risks of surgical intervention. N Engl J Med 1998; 339: 1725–1733.

Study Design This is a natural history study including 4060 patients from 61 centres worldwide. Eligible patients were prospectively identified by study investigators. Patients were assigned to either an unoperated or operated cohort, depending on whether operative treatment (open or endovascular) was intended. Unoperated patients were then divided into two groups. The study looked at unruptured aneurysms that were either associated with no previous SAH from a separate aneurysm (Group 1) or previous SAH from a separate aneurysm (Group 2). A second objective was to look at factors that increased risk. In the operated cohort, the objective was to look at treatment morbidity and mortality. All patients underwent angiography. Inclusion criteria: one or more unruptured intracranial saccular aneurysms (regardless of symptoms other than rupture, e.g. cranial nerve palsy); Rankin 1 or 2 (self-caring) after previous rupture (patients may not have had previous rupture). Exclusion criteria: fusiform, mycotic, or traumatic aneurysms; aneurysm <2 mm; SAH from single ruptured aneurysm or unknown source; unruptured aneurysm that was manipulated prior to study; previous intracranial haemorrhage of unknown cause or untreated structural abnormality; malignant brain tumour; bedridden or 29

unable to communicate when aneurysm identified.

Outcome Measures SAH and size of aneurysm were the most important factors in the unoperated cohort. Death and disability were the most important factors in the operated cohort.

Results Five-year cumulative rupture rates (5YRR) < 7 mm aneurysms: Cumulative 5YRR Group 1

Group 2

Cavernous carotid artery aneurysms

A.comm or ACA aneurysms

1–5%

Vertebrobasilar, PCA, or P.comm aneurysms

2–5%

3–4%

7–12 mm aneurysms: Cumulative 5YRR Cavernous carotid artery aneurysms

A.comm or ACA aneurysms

2–6%

Vertebrobasilar, PCA, or P.comm aneurysms

14.5%

>12 mm aneurysms: Cumulative 5YRR 13–24 mm

>25 mm

Cavernous carotid artery aneurysms

3.0%

6.4%

A.comm or ACA aneurysms

15.5%

40%

Vertebrobasilar, PCA, or P.comm aneurysms

18.4%

50%

Conditions that led to diagnosis did not differ significantly between cohorts. Fifty-one patients (3%) in the unoperated cohort had a rupture, of which 49 occurred within 5 years. For patients with aneurysms <7 mm, groups 1 and 2 were significantly different (p <0.0001). In summary, Group 1 (no previous SAH from a separate aneurysm) had a lower risk of rupture than Group 2. Rupture rate was dependent on location; the greatest risk being associated with posterior circulation aneurysms. For larger aneurysms, there was no significant difference between groups, but rupture was also related to size and location and reached 50% over 5 years for vertebrobasilar, PCA, or P.comm aneurysms >25mm. Multivariate analysis showed that age was not a factor. In the craniotomy part of the surgical cohort, risks associated with treatment included age >50 years, diameter >12mm, location in posterior circulation, previous ischaemic cerebrovascular disease, and aneurysmal symptoms other than rupture. In the endovascular group, diameter >12 mm and posterior circulation only were associated with poor outcome. Overall mortality and morbidity was between 7.1% and 12.6% depending on group and cohort.

Conclusions Aneurysmal rupture rate is related to size and location of aneurysm and for aneurysms <7 mm, risk is increased with previous SAH from a separate aneurysm. These factors, coupled with the morbidity/mortality data allow neurosurgeons to make an informed choice on whether to operate or not. In general, the risk of rupture for a particular aneurysm over the patient’s remaining lifetime can be compared to the mortality/morbidity risk.

Critique The ISUIA study was a prospective study of >4000 patients and as such is the best natural history study to date. The main limitations (as cited in the study) include the non-randomized nature of the surgical versus non-surgical cohorts, the variable follow-up that was <5 years in >50% of the patients, and the relatively low numbers in the endovascular 30

cohort. Furthermore, the low numbers recruited from each centre have led some critics to suggest that the total numbers represent <10% of patients (on average) from each centre. This implies an inherent selection bias. Another criticism is that there were substantial differences between the patients in the untreated versus the treated groups. For example, the untreated group had a higher incidence of prior SAH, cerebrovascular disease, intracranial haemorrhage, transient ischaemic attack, hypertension and its treatment, myocardial infarction, and alcohol and tobacco abuse. The same group had lower rates of cranial nerve deficit, mass effect, seizures, headaches, CT or MRI diagnosis, family history, and use of oral contraceptives or stimulants. In the no treatment group, 36 cases of SAH due to another (undetermined) potential source of haemorrhage were excluded. No analysis of aneurysm shape or the presence of daughter sacs was included in the study. Despite criticisms regarding selection bias, ISUIA has had a profound impact on the decision to treat unruptured aneurysms. Whilst some have relied on its interpretation more than others, it has provided an invaluable additional tool to neurosurgeons and interventionalists alike.

1.5 Nimodipine for Prophylaxis of Cerebral Vasospasm in Aneurysmal Subarachnoid Haemorrhage Details of Study The ‘British Aneurysm Nimodipine Trial’ was one of the first properly randomized trials involving neurosurgical patients. It showed reduced cerebral infarction and better outcome in patients given nimodipine.

Study Reference Main Study Pickard JD, Murray GD, Illingworth R, Shaw MDM, Teasdale GM, Foy PM, Humphrey PRD, Lang DA, Nelson R, Richards P, Sinar J, Bailey S, Skene A. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ 1989; 298: 636–642.

Related References Allen GS, Ahn HS, Preziosi TJ, Battye R, Boone SC, Boone SC, Chou SN, Kelly DL, Weir BK, Crabbe RA, Lavik PJ, Rosenbloom SB, Dorsey FC, Ingram CR, Mellits DE, Bertsch LA, Boisvert DP, Hundley MB, Johnson RK, Strom JA, Transou CR. Cerebral arterial spasm—a controlled trial of nimodipine in patients with subarachnoid hemorrhage. N Engl J Med 1983; 308: 619–624. Barker FG, Ogilvy CS. Efficacy of prophylactic nimodipine for delayed ischaemic deficit after subarachnoid hemorrhage: a meta-analysis. J Neurosurg 1996; 84: 405–414. Dorhout Mees SM, Rinkel GJE, Feigin VL, Algra A, van den Bergh WM, Vermeulen M, van Gijn N. Calcium antagonists for subarachnoid haemorrhage. Cochrane Database Syst Rev 2007; 3: CD000277.

Study Design Double blind, placebo RCT. Class of evidence

Randomization Nimodipine versus placebo Number of patients

554 total

Length of follow-up

3 months

Number of centres

Stratification A subgroup stratification compared prognostic factors in the two groups, including age, sex, loss of consciousness at ictus, time from haemorrhage to entry, GCS, limb weakness, neck stiffness, hypertension, angiographic and CT findings Aim was to determine if 60 mg oral nimodipine 4-hourly reduces the incidence of cerebral ischaemia and infarction arising de novo after spontaneous aneurysmal SAH. The main distinction in outcome was between moderate or good outcome and poor outcome, i.e. death or severe disability. Treatment started within 96 h of haemorrhage and continued for 21 days. 31

Demographic data and clinical data including past medical history and status on admission (including WFNS grade) were also recorded. Inclusion criteria: ≤96 h since bleed (SAH proven by CT or LP); >18 years of age. Exclusion criteria: ≥96 h since bleed; major comorbidities (renal, hepatic, pulmonary, cardiac disease); coma due to SAH within the week prior to latest SAH; lack of consent.

Outcome Measures Primary Endpoints Rate of cerebral infarction, re-bleed, or poor outcome between the two groups.

Secondary Endpoints Glasgow Outcome Score (scale of 1 to 5) at least 3 months after haemorrhage. Other secondary outcomes were causes of disability or death, including initial bleed, ischaemia, re-bleed, intracranial haematoma, hydrocephalus, operative complications, and other/unknown causes.

Results Follow-up at 3 months showed that 21 days of nimodipine treatment was effective in reducing the incidence of cerebral infarction by one-third (22% with nimodipine compared to 33% with placebo). This constitutes a reduction of 34% or 37% of definite infarcts (p = 0.014). Poor outcomes reduced significantly by 40% with nimodipine compared to placebo (20% versus 33% respectively, p <0.001). Certain factors were individually, but not independently, associated with better outcome but there was no evidence that benefit from treatment was confined to any particular subgroup. There was no significant effect on mortality between the groups.

Conclusions Oral nimodipine 60 mg 4-hourly is well tolerated and reduces cerebral infarction and improves outcome after SAH.

Critique The first randomized, double-blind, placebo-controlled trial of nimodipine was reported in 1983 (Allen et al., 1983). The authors evaluated the prophylactic use of nimodipine for 21 days following aneurysmal SAH in 125 patients of good grade and found that nimodipine was effective in reducing neurological deficits. This study by Pickard et al. reported in 1989 is the largest RCT that evaluated a calcium antagonist and included 554 SAH patients. The demographic and clinical data between the treated and placebo groups are not significantly different. However, there was an increased prevalence of hypertension, neck stiffness, and non-reactive pupils in the nimodipine group. These factors are generally associated with poor prognosis and therefore the fact that the nimodipine group did better would suggest that, if anything, the results are under-representative of any difference. In contrast, the placebo group had a larger number of patients with pre-existing cardiovascular disease and smokers. A second criticism relates to the age of the study. In the late 1980s, the usual treatment for aneurysm rupture was surgical clipping. In this trial, 187 and 181 patients in each group had proven aneurysms on angiography. Therefore, this trial is not confined to treated aneurysmal SAH but nor does it include endovascular treatment. It does, however, suggest that nimodipine is important in the early stages after SAH though the 21-day limit is somewhat arbitrary. This landmark trial has led to virtually every SAH patient being given nimodipine, worldwide. At least five other RCTs of prophylactic nimodipine have been carried out and a meta-analysis concluded that the effectiveness of nimodipine had been well demonstrated and supported routine prophylactic nimodipine administration (Barker et al., 1996). Although other calcium antagonists, such as nicardipine, have been investigated, a systematic review of 27 RCTs concluded that there was only evidence to support the prophylactic use of nimodipine (Dorhout Mees et al., 2007).

References Barker FG, Ogilvy CS. Efficacy of prophylactic nimodipine for delayed ischaemic deficit after subarachnoid hemorrhage: a meta-analysis. J Neurosurg 1996; 84: 405–414. Dorhout Mees SM, Rinkel GJE, Feigin VL, Algra A, van den Bergh WM, Vermeulen M, van Gijn N. Calcium antagonists for subarachnoid haemorrhage. Cochrane Database Syst Rev 2007; 3: CD000277.

1.6 ‘HHH’ Therapy for Vasospasm Details of Study Kassell et al. carried out a large study in 1982 to evaluate the effects of hypertensive therapy and intravascular volume expansion in a series of 58 patients with delayed ischaemic neurological deficits following aneurysmal SAH. This was the first large series to examine the role of hypertension and hypervolaemia to treat neurological deficits due to vasospasm following SAH. 32

Study Reference Main Study Kassell NF, Peerless SJ, Durwood QJ, Beck DW, Drake CG, Adams HP. Treatment of ischaemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 1982; 11: 337–343.

Related Reference Kosnik EJ, Hunt WE. Postoperative hypertension in the management of patients with intracranial arterial aneurysms. J Neurosurg 1976; 45: 148–154.

Study Design This study was a retrospective series analysis of 58 patients with neurological deficits and proven angiographic vasospasm in whom arterial hypertension was induced in an attempt to reverse the neurological deficits.

Treatment Protocol The protocol for inducing hypertension and volume expansion evolved throughout the series but included the use of transfusion of blood and blood products, the use of colloids, and administration of vasopressors if necessary. Fludrocortisone was also administered to help maintain hypervolaemia. Target parameters included a central venous pressure (CVP) of 10 mmHg and pulmonary artery wedge pressure (PAWP) of 18–20 mmHg. Estimates were made on an individual patient basis as to the systolic pressure that would likely be required to reverse the neurological deficits. Once deficit reversal was achieved the systolic pressure was kept at the minimum level required to sustain deficit reversal. Maximal systolic parameters were set at 240 mmHg systolic in patients with secured aneurysms and at 160 mmHg in unsecured aneurysms.

Results Mean age of patients was 51 years. Twenty-two patients had unsecured aneurysms at the time of induced hypertension.

Clinical Outcomes Kassell et al. reported that reversal of deficits was seen within 1 h in 81% and that the period of inducedhypertension varied between 12 h and 8 days. Other complications included dilutional hyponatraemia, coagulopathy, haemothorax, and myocardial infarction. Permanent improvement in neurological deficit

74%

No change in deficits

16%

Deterioration

10%

Most common complications

Aneurysmal re-bleed (19%) Pulmonary oedema (17%)

Most common cause of treatment failure

Established infarction (17%)

Conclusions Hypertension and hypervolaemia is relatively safe and effective in reversing neurological deficits due to vasospasm in patients with SAH. Hypertension/hypervolaemia is most effective for patients with mild deficits.

Critique Delayed ischaemic neurological deficit (DIND) is a major cause of morbidity and mortality following aneurysmal SAH. Kosnik and Hunt were the first to report the effects of raising arterial pressure in cerebral vasospasm (Kosnik and Hunt, 1974). They reported a series of seven patients in whom the neurological deficit was reversed promptly by the elevation of systemic blood pressure and found that infarction was prevented in some of these patients. Kassell et al. carried out the larger study considered here which showed that hypertensive therapy and intravascular volume expansion resulted in sustained neurological benefits in the majority of patients. One of the most significant observations of Kassell et al.’s study was that hypertension and hypervolaemia were most effective in patients with mild deficits and that these could be anticipated in patients between days 5 and 12 post-bleed who had large blood loads on their CT scans. Although they recommended angiography to confirm the presence of spasm they also maintained that if angiography was not accessible then a diagnosis of vasospasm could be made presumptively by 33

excluding other causes of neurological deterioration such as re-bleed or hydrocephalus. Since these early studies, hypertension, hypervolaemia, and haemodilution (HHH) therapy has evolved to include haemodilution to augment rheological properties of blood flow. Although HHH therapy has not been examined with a RCT, it has become the mainstay of medical therapy for the treatment of vasospasm. Recent investigations using cerebral monitoring support its continued use on the basis of improved physiological parameters at least (Raabe et al., 2005). There is no consensus as to how HHH therapy should be achieved although high-dependency care with invasive cardiovascular monitoring is commonly employed. HHH therapy is not without risks and is associated with significant complications including pulmonary oedema, myocardial ischaemia, and electrolyte abnormalities including dilutional hyponatraemia. The prophylactic use of HHH therapy has not been widely supported and preliminary trials have not shown any benefits (Solenski et al., 1995; Solomon et al., 1998). It is to be expected that in the coming years functional imaging modalities and invasive cerebral tissue monitoring may lead to refinements in the optimization of cerebral perfusion augmentation therapy.

References Kosnik EJ, Hunt WE. Postoperative hypertension in the management of patients with intracranial arterial aneurysms. J Neurosurg 1976; 45: 148–154. Raabe A, Beck J, Keller M, Vatter H, Zimmermann M, Seifert V. Relative importance of hypertension compared with hypervolemia for increasing cerebral oxygenation in patients with cerebral vasospasm after subarachnoid hemorrhage. J Neurosurg 2005; 103: 974–981. Solenski NJ, Haley EC Jr, Kassell NF, Kongable G, Germanson T, Truslowski L et al. Medical complications of subarachnoid haemorrhage: a report of the multicentre, cooperative aneurysm study. Participants of the Multicentre Cooperative Aneurysm Study. Crit Care Med 1995; 23: 1007–1017. Solomon RA, Fink ME, Lennihan L. Prophylactic volume expansion therapy for the prevention of delayed cerebral ischaemia after early aneurysm surgery. Results of a preliminary trial. Arch Neurol 1988; 45: 325–332.

1.7 Statins in the Prevention of Vasospasm in Aneurysmal Subarachnoid Haemorrhage Details of Study Whilst there has been evidence of the beneficial effects of statins in SAH prior to 2005, this study is the first randomized, phase II trial and looks specifically at the effect of pravastatin on patient outcome in SAH.

Study Reference There is an initial study with immediate follow-up looking at primary endpoints, followed by a number of other papers looking at extended (6-month) follow-up and other effects of pravastatin. The outcome of this trial has led to a much larger phase III trial of simvastatin that is underway at the time of writing (the ‘STASH’ trial).

Main Study 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: 1627–1632.

Related References Lynch JR, Wang H, McGirt MJ, Floyd J, Friedman AH, Coon AL, Blessing R, Alexander MJ, Graffagnino C, Warner DS, Laskowitz DT. Simvastatin reduces vasospasm after aneurysmal subarachnoid hemorrhage: results of a pilot randomized clinical trial. Stroke 2005; 36: 2024–2026. Tseng MY, Hutchinson PJ, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Effects of acute pravastatin treatment on intensity of rescue therapy, length of inpatient stay, and 6-month outcome in patients after aneurysmal subarachnoid hemorrhage. Stroke 2007; 38: 1545–1550. Tseng MY, Hutchinson PJ, Turner CL, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Biological effects of acute pravastatin treatment in patients after aneurysmal subarachnoid hemorrhage: a double-blind, placebo-controlled trial. J Neurosurg 2007; 107: 1092–1100.

Study Design Double-blind placebo RCT. Class of evidence

Randomization Pravastatin versus placebo Number of

80 (equally distributed) 34

patients Length of follow-up

Not relevant as this is a phase II trial looking at safety and immediate clinical and physiological effects (however, 6-month data is the subject of a related publication)

Number of centres

Stratification None, but baseline characteristics similar in the two groups This is a phase II trial looking at safety and therefore only has small numbers of patients from a single centre. Its main purpose is as a pilot study to assess whether a larger phase III trial is needed to evaluate whether pravastatin reduces vasospasm or its consequences including delayed ischaemic deficit. As well as primary and secondary endpoints, baseline clinical data were measured. These included age, gender, medical history, WFNS grade, and also radiological characteristics including Fisher grade on CT, presence of hydrocephalus or intraventricular blood, location of aneurysm on angiography. Trial medication (pravastatin sodium 40mg once daily) was started within 72 h of ictus and continued up to 14 days or until discharge. The main aim was to see if pravastatin reduces vasospasm or its consequences including delayed ischaemic deficit. Inclusion criteria: aneurysmal SAH; age 18â&#x20AC;&#x201C;84 years. Exclusion criteria: non-aneurysmal SAH; pregnancy; pre-ictal statin therapy; contraindication to statins (liver or renal dysfunction, or alanine aminotransferase (ALT) >50 U/L).

Outcome Measures Primary Endpoints Incidence, severity, and duration of vasospasm on transcranial Doppler (TCD) indices. Duration of impaired cerebral autoregulation.

Secondary Endpoints Evidence of vasospasm-related delayed ischaemic deficits (DIDs). Disability at discharge.

Results

In the placebo group, there was a reduction in both primary endpoints with a 32% reduction in vasospasm on TCD in the pravastatin group (p = 0.006) and a 42% reduction in severe vasospasm (p = 0.044). The duration of vasospasm was shortened by 0.8 days in the pravastatin group (p = 0.068). The period of impaired cerebral autoregulation was shorter with pravastatin (ipsilateral side by 2.4 days, p = 0.011). With regard to the secondary endpoints, the incidence of vasospasm- related DIDs was 83% reduced in the pravastatin group (p <0.001). Pravastatin was associated with a reduction in disability at discharge and a reduction in mortality of 75% (p = 0.037). Subsequent follow-up show that these beneficial effects were still present at 6 months (Tseng et al., 2007).

Conclusions Immediate statin therapy reduces potentially adverse physiological and clinical events after an acute cerebrovascular illness.

Critique Previous studies evaluating the role of statins in the prevention of cerebral vasospasm were retrospective, observational studies. Earlier studies had reported differences in outcome between patients already taking statins and 35

those not on anti-cholesterol medications. This small randomized study by Tseng et al. with pravastatin reported in 2005 demonstrated a reduced incidence of vasospasm in patients treated with pravastatin. The pravastatin incidence of TCD-detected vasospasm was reduced by 32% with a reduced incidence of DIND and mortality. At 6 months, beneficial effects on physical and psychological aspects of functioning have subsequently been reported. Lynch et al. also reported the results of a smaller trial randomized clinical trial of simvastatin (Lynch et al., 2005). Their study included simvastatin versus placebo randomly allocated to 39 patients within 48 h of the SAH ictus and they reported a reduction in vasospasm in those receiving simvastatin (p <0.05). Vasospasm was defined by clinical impression in the presence of at least one confirmatory radiological test (TCD or angiography). The study was not powered to answer the definitive question as to whether simvastatin was effective in reducing vasospasm, but the authors were able to conclude that simvastatin was safe and well tolerated in subarachnoid patients. Both the simvastatin study by Lynch et al. and the pravastatin study by Tseng et al. were randomized, placebo-controlled trials making them valuable contributors to the question of whether statins are beneficial in aneurysmal SAH. They are examples of how even small, simple studies can provide essential evidence that can lead to larger, multi-centre trials to answer a simple clinical question. The authors do not purport to show improved clinical outcome but rather look at short-term delayed ischaemic deficits and, by measuring TCD blood flows between the two groups, attempts to provide a pathophysiological reason for the improvement. In this sense, both studies make the assumption that TCDs are a good indicator of cerebral blood flow and that improvements in TCDs are the reasons for the improved outcome. Like any good ‘pilot’ studies, certain questions are left unanswered. Another restriction of the pravastatin trial is the relatively small number of aneurysms treated with coil embolization, as compared to clipping. Although pravastatin had the same effect on both groups, the coil embolization group was small and this is essentially a trial of pravastatin in aneurysms that were surgically clipped. This will be resolved by the larger multi-centre study. Another question raised in the literature is the increased rate of sepsis, including ventriculitis, in the pravastatin group. Although this did not reach significance, and did not appear to contribute to a worse mortality, it will need to be addressed in a larger trial. The pravastatin trial has led to a multicentre RCT looking at the potential benefit of simvastatin (40 mg for 21 days) in aneurysmal SAH (STASH) that is underway. Some centres have already routinely started using statins in SAH.

Reference Lynch JR, Wang H, McGirt MJ, Floyd J, Friedman AH, Coon AL, Blessing R, Alexanderander MJ, Graffagnino C, Warner DS, Laskowitz DT. Simvastatin reduces vasospasm after aneurysmal subarachnoid hemorrhage: results of a pilot randomized clinical trial. Stroke 2005; 36: 2024–2026.

1.8 Treatment of Cerebral Arteriovenous Malformations Details of Studies Predicting the risks of treatment is difficult due to the heterogeneity of AVMs which vary from ‘simple’ to ‘complex’. Spetzler and Martin reported a six-tier grading system which they retrospectively applied to a series of 100 AVMs (Spetzler and Martin, 1986). Hamilton and Spetzler subsequently validated this grading system prospectively in a series of 120 consecutive patients (Hamilton and Spetzler, 1994). The main findings of these landmarks studies have been summarized and considered here.

References Main Studies Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg 1986; 65: 476–483. Hamilton MG, Spetzler RF. The prospective application of a grading system for arteriovenous malformations. Neurosurgery 1994; 34: 2–6.

Related Reference Luessenhop AJ. Gennarelli TA. Anatomical grading of supratentorial arteriovenous malformations for determining operability. Neurosurgery 1977; 1: 30–35.

Study Design Spetzler and Martin, 1986: retrospective analysis of 100 patients with AVMs. Hamilton and Spetzler, 1994: prospective analysis of 120 patients with AVMs. The grading system proposed by Spetzler and Martin is a six-tier system awarding points as follows: Feature of AVM

Points

Size of AVM

>6 cm (large)

Eloquence of adjacent brain Pattern of venous drainage

3–6 cm (medium)

<3 cm (small)

Eloquent

Non-eloquent

Deep

Superficial only

Grade I lesions: small, superficial, and located in non-eloquent cortex. Grade V lesions: large, deep, and located in critical neurological areas. Grade VI lesions are considered inoperable lesions.

Results Retrospective Series (Spetzler and Martin, 1986) Grading of AVM correlated with results of surgery (Figure 1.1).

Fig 1.1 Correlation of surgical results with Spetzler–Martin AVM grading.

Prospective Series (Hamilton and Spetzler, 1994) Outcomes were assessed using the Glasgow Outcome Score. Follow-up was 77.4% at 1 year. Grading of AVMs was correlated with the development of both temporary (p <0.0001) and permanent (p = 0.008) neurological deficits. Correlation was greatest when all three components of the grading system were applied. Grading of AVMs was correlated with outcome (Figure 1.2)

Fig 1.2 One-year outcomes for patients with Spetzler–Martin Grades I–VI AVMs.

Conclusions The Spetzler, Martin, and Hamilton grading system is a robust and accurate method of predicting risk of intervention in patients with AVMs.

Critique 37

Determination of the natural history of AVMs is essential in order to assess whether the risks of surgical intervention are less than the long-term risks of conservative management. Epidemiological studies have been difficult due to heterogeneous patient populations and variation in treatment practices. Numerous studies have been carried out to address this question. Perhaps the longest prospective study to date was begun in Finland in 1965 with a 24-year follow-up of 160 patients being reported in 1990 (Troupp et al., 1970; Ondra et al., 1990). Two large, prospective, population-based studies are currently ongoing. One is the New York Islands AVM Study (NYIAVMS) (Stapf et al., 2003). The other is the Scottish Intracranial Vascular Malformation Study (SIVMS) (Al-Shahi et al., 2003). However, from the literature to date it appears that the risk of intracranial haemorrhage from AVMs is 1–4% per year with an annual mortality of 1–1.5%, and that the risk of bleeding of unruptured AVMs appears to be approximately 2% per year with re-bleed risk of 18% within the first year (Al-Shahi and Warlow, 2001). Surgery for AVMs is indicated only when the risk of operation is less than the risk of a conservative course of management as determined by the natural history of the AVM. The results of the NYIAVMS and SIVMS studies will be invaluable in helping determine the natural history of AVMs. The grading system of Spetzler and Martin aimed to provide a simplified objective method of predicting the risks of surgical intervention in individual cases of AVM. Although previous grading systems had been reported they were primarily based on the AVM anatomy (Luessenhop and Gennarelli, 1977). The grading system of Spetzler and Martin takes into account the variables of vascular steal, eloquence of adjacent brain, and venous drainage patterns. Their retrospective and prospective data strongly support the predictive validity of their grading scale as a robust mechanism to objectively predict outcome in individual AVMs. The greatest weakness of the study is that it was a single-centre study. However, the authors cited two other published series that had used their grading system as independent validation (Steinmeier et al., 1989; Heros et al., 1990). In addition, this grading system has been criticized for being oversimplified. Samson and Batjer proposed that Grade IV and V AVMs might be considered as a separate entity as it is these grades that are associated with a post-operative morbidity and that a narrative description of the complexities of these AVMs might be more useful to those considering surgical intervention (Samson and Batjer, 1994). Nonetheless, the grading system developed by Spetzler, Martin, and Hamilton remains a landmark in neurovascular neurosurgery and is a useful tool for the comparison of different treatment modalities and regimens. Spetzler’s group has continued to use this classification system to establish the risks and benefits of treatment of AVMs (Han et al., 2003). In addition to surgical resection, several other modalities are now available for the treatment of AVMs including endovascular techniques and stereotactic radiosurgery (Friedman et al., 1995; Martin et al., 2000). However a Cochrane Review by Al-Shahi and Warlow revealed that there are no randomized trials with clear outcomes comparing modalities (Al-Shahi and Warlow, 2006).

References Al-Shahi R, Warlow C. A systematic review of the frequency and prognosis of arteriovenous malformations of the brain in adults. Brain 2001; 124: 1900–1926. Al-Shahi R, Bhattacharya JJ, Currie DG, Papanastassiou V, Ritchie V, Roberts RC, Sellar RJ, Warlow CP, for the Scottish Intracranial Vascular Malformation Study Collaborators. Prospective, population-based detection of intracranial vascular malformations in adults: the Scottish Intracranial Vascular Malformation Study (SIVMS). Stroke 2003; 34: 1163– 1169. Al-Shahi R, Warlow CP. Interventions for treating brain arteriovenous malformations in adults Cochrane Database Syst Rev 2006; 1: CD003436. Friedman WA, Bova FJ, Mendenhall WM. Linear accelerator radiosurgery for arteriovenous malformations: the relationship of size to outcome. J Neurosurg 1995; 82: 180–189. Han PP, Ponce FA, Spetzler RF. Intention-to-treat analysis of Spetzler-Martin grades IV and V arteriovenous malformations: natural history and treatment paradigm. J Neurosurg 2003; 98: 3–7. Heros RC, Korosue K, Diebold PM. Surgical excision of cerebral arteriovenous malformations: late results. Neurosurgery 1990; 26: 570–577 Luessenhop AJ, Gennarelli TA. Anatomical grading of supratentorial arteriovenous malformations for determining operability. Neurosurgery 1977; 1: 30–35. Martin NA, Khanna R, Doberstein C, Bentson J. Therapeutic embolisation of arteriovenous malformations: the case for and against. Clin Neurosurg 2000; 46: 295–318. Ondra S, Troupp H, George ED, Schwab K. The natural history of symptomatic arteriovenous malformations if the brain: a 24-year follow-up assessment. J Neurosurg 1990; 73: 387–391. Samson DS, Batjer HH. Grading systems for AVMs: comments. Neurosurgery 1994; 34: 6–7. Stapf C, Mast H, Sciacca RR, Berenstein A, Nelson PK, Gobin YP, Pile-Spellman J, Mohr JP, for the New York Islands AVM Study Collaborators. The New York Islands AVM Study: design, study progress, and initial results. Stroke 2003; 34: 29–33. Steinmeier R, Schramm J, Müller HG, Fahlbusch R. Evaluation of prognostic factors in cerebral arteriovenous malformations. Neurosurgery 1989; 24:193–200. Troupp J, Martilla I, Halonen V. Arteriovenous malformations of the brain. Prognosis without operation. Acta Neurochir 1970; 22: 125–128.

1.9 Surgery for Spontaneous Intracerebral Haematomas Details of Studies The International Surgical Trial in Intracerebral Haematoma (STICH) and its successor, STICH II, are the largest RCTs to date looking at the role of early surgery in the management of intracerebral haematomas. These studies were headed by Professor Mendelow from Newcastle General Hospital and were funded by the Medical Research Council (UK). The subsequent trial, STICH II, was designed to prospectively study the subset of patients in the first trial who appeared to benefit from early surgery.

The STICH Trial Study Reference Main Study Mendelow AD, Gregson BA, Fernandes HN, Murray GD, Teasdale GM, Hope DH, Karimi A, Shaw M, Barer DH, for the STICH investigators. 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: 387–397.

Related References Broderick JP. The STICH trial: what does it tell us and where do we go from here? Stroke 2005; 36: 1619–1620. Mendelow AD, Unterberg A. Surgical treatment of intracerebral haemorrhage. Curr Opin Crit Care 2007; 13: 169–174.

Study Design International multi-centre PRCT. Class of evidence

Randomization

Early surgery versus best medical management

Number of patients 1033 patients Outcomes

Primary outcomes: GOS at 6 months Secondary outcomes: Mortality at 6 months Prognosis at 6 months

Follow-up

93% completed follow-up

Number of centres 83 centres in 27 countries Stratification

According to predicted ‘good’ or ‘poor’ prognosis using a prognostic score at randomization

Patients randomized and analysed on an intention-to-treat basis. Surgery was carried out within 24 h of randomization. Surgery included craniotomy and CT-guided aspiration of the clot and the method chosen was left to the discretion of the operating surgeon. The option of delayed surgery remained open to those who were randomized to the medical arm. 503 patients underwent early surgery and 530 patients received best medical therapy. Primary outcome was assessed using the 8-point GOS at 6 months. Inclusion criteria: CT evidence of intracerebral haematoma within 72 h; clinical uncertainty regarding the benefits of either treatment arm of the trial; haematoma diameter >2 cm; GCS ≥5. Exclusion criteria: aneurysmal bleed; infratentorial bleed; extension of bleed into brainstem; any co-morbid factor that might interfere with outcome assessment.

Outcome Measures Primary Endpoints GOS, Barthel Index, and Rankin Scale assessed by postal questionnaire at 6 months. Patients with a poor prognosis at randomization were deemed to have a favourable outcome if there was a severe disability or better on the GOS. Patients with good prognosis were deemed to have a favourable outcome if they had a moderate disability or better on the GOS.

Secondary Endpoint Mortality at 6 months.

Results

Prognosis-based analyses did not reveal any statistically significant differences between the two arms of the trial. Subgroup analysis showed that a favourable outcome was more likely with early surgery for superficially based lesions (â&#x2030;¤1 cm from cortical surface) with a 29% relative benefit but this difference was not statistically significant.

Conclusions There is no overall benefit from early surgery versus initial conservative treatment for spontaneous supratentorial haematomas.

Critique Patients suffering from spontaneous intracerebral haematomas represent a significant proportion of neurosurgical emergencies. Nine previous clinical trials looking at the role of surgery produced conflicting results regarding the role of surgery in the management of patients with non-aneurysmal spontaneous supratentorial haematomas. The STICH trial is the largest trial to date addressing this question. Unfortunately, STICH has been overinterpreted by some to mean that there is no benefit from surgery for all supratentorial haematomas. However, STICH only looked at haematomas for which the responsible surgeon was uncertain regarding the benefits of surgery versus conservative management. The results of STICH confirm that surgeons are correct to be uncertain for these patients but the results cannot be extrapolated to all intracerebral haematomas. In STICH, the timing of surgery was relatively long after presentation (median time to surgery 30 h). There may, therefore, be a role for much earlier surgery, e.g. within 12 h of the initial bleed. In addition, most patients underwent craniotomy (77%) and so the question remains as to whether there is a role for more minimally invasive methods for evacuating haematomas. The subgroup analysis suggesting benefit from early surgery in patients with superficial lobar haemorrhage was the basis for the subsequent STICH II trial, described in the following section.

Reference Mendelow AD, Unterberg A. Surgical treatment of intracerebral haemorrhage. Curr Opin Crit Care 2007; 13: 169â&#x20AC;&#x201C;174.

The STICH II Trial Study Reference Main Study Mendelow AD, Gregson BA, Rowan EN, Murray GD, Gholkar A, Mitchell PM, for the STICH investigators. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): a randomised trial. Lancet 2013; 382: 397â&#x20AC;&#x201C;408.

Study Design International multi-centre PRCT. Class of evidence

Randomization

Early surgery (12 h from randomization) versus best medical management

Number of patients 601 patients randomized Outcomes

Primary outcomes: Extended GOS at 6 months Secondary outcomes: 40

Mortality, time to death, Rankin and EuroQoL score at 6 months 21% cross-over from medical group Follow-up

98%

Number of centres 78 centres in 27 countries Stratification

According to predicted ‘good’ or ‘poor’ prognosis using a prognostic score at randomization

Patients randomized and analysed on an intention-to-treat basis. No blinding of patients/relatives or clinical staff. Surgery was carried out within 12 h of randomization (in contrast to 24 h in STICH). Surgery included craniotomy and CT-guided aspiration of the clot and the method chosen was left to the discretion of the operating surgeon. The option of delayed surgery remained open to those who were randomized to the medical arm. 297 patients underwent early surgery and 286 patients received best medical therapy. Primary outcome was assessed using the 8-point extended Glasgow Outcome Scale (GOSE) at 6 months. Inclusion criteria: • CT evidence of intracerebral haematoma within 48 h. • Clinical uncertainty regarding the benefits of either treatment arm of the trial. • Lobar haematoma. • Superficial haematoma (up to 1 cm from surface of cortex, volume 10–100 mL). • Motor score 5 or 6; eye opening score 2 or more (i.e. conscious at randomization). Exclusion criteria: aneurysmal/tumour/traumatic bleed; AVM on angiography; basal ganglia, thalamic, cerebellar, brainstem location; intraventricular bleed; any co-morbid factor that might interfere with outcome assessment.

Outcome Measures Primary Endpoints GOSE, Barthel Index, and Rankin Scale assessed by postal questionnaire at 6 months. Patients with a poor prognosis at randomization were deemed to have a favourable outcome if there was a severe disability or better on the GOS. Patients with good prognosis were deemed to have a favourable outcome if they had a moderate disability or better on the GOS. Secondary Endpoints 6 months: mortality, time to death, Rankin and EuroQoL.

Results Prognosis-based analyses did not reveal any statistically significant differences between the two arms of the trial. Survival advantage with no vegetative survivors at 6 months in the surgery group by this was not statistically significant.

Conclusions There is no overall benefit from early surgery versus initial conservative treatment for spontaneous superficial lobar supratentorial haematomas.

Critique STICH II was a well-designed study and executed study with excellent follow-up which offers Level I evidence informing the management of lobar superficial spontaneous supratentorial intracerebral haemorrhage (ICH). Crossover from the medical arm to surgery was 21%. Therefore the trial was effectively a comparison of early 41

surgery versus initial conservative management followed by delayed surgery if there was clinical deterioration. STICH II suffers from the same limitation as the first STICH trial in that patients were only eligible if the responsible surgeon was uncertain regarding the benefits of surgery versus initial conservative management. Half of the trial population were either fully alert or confused. Early surgery for these patients may indeed have not been superior to initial conservative management and this may have contributed to the lack of significant difference overall in primary outcome variables. Further, the trial was powered to detect a 12% difference in primary outcome variables so perhaps a larger study population may have shown a significant difference between the two groups. There is a view held by some that there is a fundamental problem that needs to be addressed regarding the role of surgery in the evacuation of intracerebral haematomas—the concept of a penumbra around the clot. The rationale of clot evacuation is to control intracranial pressure (ICP) and to prevent further damage to surrounding brain. However, there is no firm evidence that there is a surrounding penumbra of brain that is at risk from the clot and more basic science needs to be done to elucidate this problem further. Certainly, the toxic effects of the clot to the surrounding brain remain to be ascertained. An alternative view to this is that preclinical animal studies may support clot evacuation and so it is reasonable to hypothesize that similar evacuation in people may have benefit. It would appear that further studies, perhaps with imaging modalities such as functional magnetic resonance imaging (fMRI) or singlephoton emission computed tomography (SPECT) in patients with ICH may shed more light on the natural history of intracerebral haematomas. Further trials to evaluate surgical intracerebral haemorrhage management issues are ongoing. CLEAR IVH is looking at whether there is a benefit for surgery to remove intraventricular clots <30 mL. MISTIE is looking at minimal invasive surgery to remove deep ICH.

1.10 Decompressive Surgery for Malignant Cerebral Artery Infarction Details of Studies Malignant MCA infarction (MMI) is associated with a mortality rate of 80%. Since 2000, three European trials have addressed the role of decompressive surgery in these patients: the DECIMAL trial (decompressive craniectomy in malignant MCA infraction) performed in France; the DESTINY trial (decompressive surgery for the treatment of malignant Infarction of the MCA); performed in Germany; and the HAMLET trial (hemicraniectomy after MCA infarction with life-treatening (o)edema trial) performed in the Netherlands. Although HAMLET was still ongoing, a pooled analysis of these three trials was published in 2007. The final results of HAMLET have recently been published in 2009. In addition, a North American trial, the HeADDFIRST trial (hemicraniectomy and durotomy on deterioration from infarction-related swelling trial), was carried out between 2000 and 2003, although this was only ever published in abstract form.

Study References Main Studies DECIMAL Trial Vahedi K, Vicaut E, Mateo J, Kurtz A, Orabi M, Guichard JP, Boutron C, Couvreur G, Rouanet F, Touzé E, Guillon B, Carpentier A, Yelnik A, George B, Payen D, Bousser MG, for the DECIMAL Investigators. Sequential-design, multicenter, randomised, controlled trial of early decompressive craniectomy in malignant middle cerebral artery infarction (DECIMAL trial). Stroke 2007; 38: 2506–2517.

DESTINY Trial Decompressive surgery for the treatment of malignant infarction of the middle cerebral artery (DESTINY): a randomised controlled trial. Stroke 2007; 38: 2518–2525.

HAMLET Trial Hofmeijer J, Amelink GJ, Algra A, van Gijn J, Macleod MR, Kappelle LJ, van der Worp HB; HAMLET investigators. 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.

Pooled Analysis of DECIMAL, DESTINY, and HAMLET Trials Vahedi, K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink GJ, Schmiedeck P, Schwab S, Rothwell PM, Bousser MG, van der Worp HB, Hacke W, for the DECIMAL, DESTINY and HAMLET investigators. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised trials. Lancet Neurol 2007; 6: 215–222.

HeADDFIRST Trial Frank JI. Hemicraniectomy and durotomy upon deterioration from infarction related swelling trial (HeADDFIRST): first public presentation of the primary study findings. Neurology 2003; 60(Suppl 1): A426.

Related References 42

Carandang RA, Krieger DW. Decompressive hemicraniectomy and durotomy for malignant middle cerebral artery infarction. Neurocrit Care 2008; 8: 286–289. Gupta R, Conolly ES, Mayer S, Elkind MS. Hemicraniectomy for massive middle cerebral artery territory infarction: a systematic review. Stroke 2004; 35: 539–543. Hacke W, Schwab S, Horn M, Spranger M, DeGeorgia M, von Kummer R. ‘Malignant’ middle cerebral artery infarction: clinical course and prognostic signs. Arch Neurol 1996; 53: 309–315.

Study Designs European Trials PRCTs.

Inclusion criteria very similar in all trials apart from age and time allowed from onset of symptoms: • Age: decimal 18–55 years; DESTINY 18–60 years; pooled analysis of DESTINY/DECIMAL/HAMLET 18–60 years. • Time from onset of symptoms: DECIMAL <24 h; DESTINY <36 h; HAMLET <45 h. Exclusion criteria also similar: significant pre-stroke disability; significant haemorrhagic infarction; coagulopathy; poor neurological state (e.g. fixed pupils). MMI criteria varied slightly between trials but was defined on the basis of the following criteria: clinical signs of infarction on National Institutes of Health Stroke Score (NIHSS) including a score of ≥1 for the level of consciousness; radiological (CT or MRI) documentation of unilateral MCA infarction of a predetermined percentage. Functional outcome defined using the modified Rankin scores and dichotomized into ‘favourable’ or ‘good’ (mRS ≤3) or ‘unfavourable’ or ‘poor’ (mRS ≥4). Analyses on an intention-to-treat basis in DESTINY and HAMLET. A sequential method of analysis applied in DECIMAL.

North American Trial (HeADDFIRST) Class of evidence

Randomization

Surgery versus medical care

Number of patients

Follow-up

180 days Primary endpoint: Mortality Secondary endpoint: Functional outcome

Number of centres

Stratification

None 43

Inclusion criteria: ages 18–75; NIHSS >18; premorbid mRS <2 with complete MCA ± ACA or PCA infarction; infarct volume >50% MCA territory or >90 cm3 on early CT, or > 180 cm3 on late CT. Randomization to surgery or medical care was triggered by development of midline shift (≥7 mm septal or >4 mm pineal gland displacement).

Pooled Analysis of DECIMAL, DESTINY, and HAMLET Primary endpoint: mRS ≤4.

Results DECIMAL and DESTINY DECIMAL was discontinued early because of recruitment problems and an interim analysis indicating a significant benefit of surgery on mortality. Recruitment to DESTINY was discontinued early because a predetermined analysis at 6 months showed a significant benefit of surgery on mortality.

Absolute reduction in death of 52.8% with surgery in DECIMAL trial. There were no bedridden patients in the DECIMAL trial at the end of 12 months (mRS 5) who had undergone surgery.

Pooled Analysis of DECIMAL, DESTINY, and HAMLET

There were no significant differences in the outcome measures between the three trials at the time of the pooled analysis. The absolute risk reduction (ARR) for mortality at 12 months was 51.2%. Seventy-five per cent of survivors receiving medical care had a ‘favourable’ outcome (mRS <4) versus 55% of survivors who received surgery. Forty per cent of survivors who had surgery had a moderately severe disability and were unable to walk without assistance or attend to their own bodily needs without assistance (mRS of 4) versus 8% of those who received medical care.

HAMLET ARR in case fatality was 38%.

HeADDFIRST A non-significant reduction in mortality was reported. Functional outcomes not reported.

Conclusions DECIMAL: decompressive surgery improves survival in young patients with MMI but with an increased number of patients with moderately severe disability. DESTINY: pooled analysis—early decompressive surgery for MMI reduces mortality and increases the number of patients with a favourable functional outcome. HAMLET: surgical decompression within 48 h of onset of symptomatic MCA infarction did not improve functional outcomes compared to medical treatment. HeADDFIRST: there was no difference in mortality at 180 days between surgical or medical management.

Critique The pooled analysis of three ongoing trials is almost unique in the literature and the results of this pooled analysis are in keeping with reported findings from uncontrolled case series. However, one of the most fundamental dilemmas facing neurosurgeons is highlighted by the results of these studies and that is the question of what constitutes a ‘favourable’ outcome. A mRS ≤3 is generally accepted as ‘favourable’ but the pooled analysis used mRS ≤4, thereby including patients who were left with moderately severe disability. In fact, although surgery reduced mortality, a greater number of survivors (tenfold) are left with moderately severe disability. The authors of the pooled analysis have been careful to emphasize, therefore, that patients and clinicians need to be willing to accept the possibility of this survival outcome. From one perspective, therefore, hemicraniectomy for MMI is a life-saving procedure. An alternative view is that hemicraniectomy saves lives at the cost of producing unacceptable levels of disability in the survivors. Indeed, the validity of trial designs that dichotomize outcomes into ‘favourable’ and ‘unfavourable’ has been widely criticized in the literature and it has been pointed out by numerous people that ‘favourable’ is not necessarily synonymous with ‘acceptable’ or ‘desirable’ outcomes. Various criticisms have been raised against these trials, including the issue of whether non-blinding of treatment arms had any effect on patient management, and in particular the use of intensive care resources in the two groups. For example, in the DECIMAL trial all patients undergoing surgery received mechanical ventilation as compared to just over only two-thirds of patients managed medically. Whether this was an effect of non-blinding remains open to question (Mayer, 2007). Various other concerns regarding the results have been raised and the way in which physicians and surgeons will use the information. For example, there is a tendency to avoid hemicraniectomy in patients with dominant hemisphere MMI due to the perception that global aphasia is a cruel outcome that should be avoided at all costs. Mayer has pointed out that the benefit of hemicraniectomy in the pooled analysis was independent of the presence or absence of aphasia and that dominant hemisphere involvement may not necessarily be an acceptable reason for withholding hemicraniectomy (Mayer, 2007). One of the greatest criticisms of these trials is whether the criteria for patient selection can really reflect any degree of understanding of the natural history of MMI. The processes which determine which patients develop fatal brain oedema are not understood and there is clearly a need for larger imaging studies to evaluate the natural history of these lesions before we can fully elucidate the role of surgical or other interventions. The decision to perform hemicraniectomy for MMI is still a matter for consideration on an individual case-by-case basis. Issues regarding the optimal timing of surgery still need to be resolved. The authors of the HAMLET trial updated the pooled analysis of the DESTINY/DECIMAL/HAMLET trials and reported a benefit of surgery for those operated on with 48 h of onset of stroke symptoms. However, no conclusions can be drawn about those patients operated on after this time period. Age is certainly an important factor with regard to outcome as it appears that the mortality even with surgery for patients with MMI aged >50 years is more than twice that of patients aged <50 (Gupta et al., 2004). Surgical decompression for MMI remains a complex dilemma for both physicians, surgeons, and their patients.

References Gupta R, Conolly ES, Mayer S, Elkind MS. Hemicraniectomy for massive middle cerebral artery territory infarction: a systematic review. Stroke 2004; 35: 539–543.

Mayer SA. Hemicraniectomy: a second chance on life for patients with space-occupying MCA infarction. Stroke 2007; 38: 2410–2412.

1.11 Carotid Endarterectomy for Carotid Stenosis Details of Study Carotid surgery for patients with symptomatic carotid stenosis has been performed for >50 years but until 1991, no large, comprehensive trial had been conducted. The North American Symptomatic Carotid Endarterectomy Trial (NASCET) started recruiting in 1987 and initial reports were published in 1991. NASCET was a pivotal trial as it demonstrated that symptomatic patients with 70–99% stenosis benefited from surgery as compared to conservative management. Indeed, a clinical alert in 1991 stated that for 70–99% stenosis, surgery was clearly beneficial and recruitment of this group was stopped. Other trials in the 1990s include the European Carotid Surgery Trial (ECST) and the Veterans Affairs Cooperative Symptomatic Carotid Stenosis Trial. These were followed by the Asymptomatic Carotid Atherosclerosis Study (ACAS) in 1995 that suggested that asymptomatic patients with >60% stenosis would benefit from surgery. More recently, with the advent of carotid stenting, trials have compared surgery to interventional radiological procedures. These are discussed in a separate section. In this section, we focus on the NASCET trial.

Study References Main Studies Barnett HJ, Taylor DW, Eliasziw M, Fox AJ, Ferguson GG, Haynes RB, Rankin RN, Clagett GP, Hachinski VC, Sackett DL, Thorpe KE, Meldrum HE, Spence JD, for the North American Symptomatic Carotid Endarterectomy Trial Collaborators. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. N Engl J Med 1998; 339: 1415–1425. National Institute of Neurological Disorders and Stroke, Stroke and Trauma Division. North American Symptomatic Carotid Endarterectomy Trial (NASCET) investigators. Clinical alert: benefit of carotid endarterectomy for patients with high-grade stenosis of the internal carotid artery. Stroke 1991; 22: 816–817. The 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: 445–453.

Related References European Carotid Surgery Trialists’ Group. Randomized trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998; 351: 1379–1387. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995; 273: 1421–1428. Mayberg MR, Wilson SE, Yatsu F, Weiss DG, Messina L, Hershey LA, Colling C, Eskridge J, Deykin D, Winn HR. Veterans Affairs Cooperative Studies Program 309 Trialist Group. Carotid endarterectomy and prevention of cerebral ischaemia in symptomatic carotid stenosis. JAMA 1991; 266: 3289–3294. Paciaroni M, Eliasziw M, Sharpe BL, Kappelle J, Chaturvedi S, Meldrum H, Barnett HJM. Long-term clinical and angiographic outcomes in symptomatic patients with 70% to 90% carotid artery stenosis. Stroke 2000; 31: 2037–2042.

Study Design Non-blinded multi-centre RCT. Class of evidence

Randomization Carotid endarterectomy versus medical management Number of patients

659 (After 1991, 131, i.e. not 70–99% stenosis)

Follow-up

Mean = 3.6 years for medical group Mean = 7.0 years for surgical group Primary endpoint: Ipsilateral stroke Secondary endpoints: Any stroke Death

Number of centres

Stratification

By degree of stenosis, i.e. <70% or 70–99%. A subsequent analysis divided the latter into 70– 46

84% and 85–99% Patients were randomized within 120 days of diagnosis of severe carotid bifurcation stenosis and ipsilateral transient ischaemic attack (TIA) or stroke. This was the first randomized study to show that surgery is beneficial for 70–99% stenosis. Multiplanar anteroposterior, oblique, and lateral selective ICA angiogram was required. Before 1991, 4-monthly clinic visits. After 1991, annual visits, bi-annual telephone assessments. If a stroke occurred, the patient had an extra clinic visit. All patients underwent annual carotid ultrasound scan. All patients had enteric-coated aspirin as well as antihypertensive, antilipidaemic and antidiabetic therapies, if required. Several later studies including those in Europe showed similar results. Inclusion criteria: TIA or disabling stroke + severe carotid bifurcation stenosis within 120 days; age ≤80 years; consent. Exclusion criteria: cardiac source of potential embolism; angiographic evidence of intracranial lesion > extracranial lesion; other life-threatening or disabling conditions.

Outcome Measures Primary Endpoint Fatal or non-fatal stroke ipsilateral to the randomized carotid artery.

Secondary Endpoints Strokes in any territory. Death.

Results Baseline characteristics were similar in the two groups. In February 1991, an interim report (cited earlier) demonstrated that, in 659 patients with severe stenosis (70–99%), endarterectomy was associated with an absolute reduction of 17% in the risk of ipsilateral stroke at 2 years. Therefore, patients with severe stenosis were not enrolled after 1991 but were continued to be followed up until 1997. Enrolment of patients with <70% stenosis continued until 1996. The overall results showed that in the 50–69% stenosis group, the 5-year ipsilateral stroke rate was 15.7% in the surgical group compared to 22.2% in the medical group (p = 0.045). In the <50% stenosis group there was no significant difference between surgical and medical treatments (p = 0.16). In the severe stenosis group (70–99%), the 30-day rate of disabling ipsilateral stroke or death was 2.1% and this increased to 6.7% at 8 years. Benefits were greatest for men, those with hemispheric symptoms, and recent stroke.

Conclusions Patients with severe carotid bifurcation stenosis (70–99%) have a clear benefit from endarterectomy that is longlasting. Those with 50–69% have a moderate benefit, and other risk factors as well as surgeon skill should be taken into account. Surgery for stenosis <50% does not yield any benefit.

Critique The results of the NASCET trial were further corroborated by the European Carotid Surgery Trial (ECST) that showed a 11.6% risk reduction of stroke at 3 years following surgery, in patients with >60% stenosis. Also similar to NASCET, the trial demonstrated that symptomatic patients with <40% stenosis had a worse outcome if treated surgically. The benefit of endarterectomy for asymptomatic patients is somewhat more controversial. The Asymptomatic Carotid Atherosclerosis Study (ACAS) demonstrated a significant 5.9% reduction in perioperative stroke or death and stroke at 5 years following surgery. However, in this trial, the investigators had an extremely low 30-day complication rate of only 2.3% (stroke or death). Similarly, the surgical complication rate was only 6% in the NASCET trial. These results imply that in order to get a 17% absolute reduction in stroke or death, a surgeon needs to have a complication rate of 6% or less. For the symptomatic, severe stenosis group, there is clear benefit from endarterectomy but in the moderate group, a surgeon has to be near the NASCET investigators’ rate or better to see a benefit. Timing of surgery is an important factor. The cumulative risk results in this study show that the greatest risk of stroke is in the first 6 months of symptoms with a decrease continuing until approximately 2 years. After this, the difference between surgery and conservative treatment is less clear. Therefore, when interpreting the trial results, surgeons should consider whether the patients still have symptoms as well as the degree of stenosis. Another important factor to consider is the how the degree of stenosis has been measured. In the NASCET trial, all patients underwent angiography. The trial investigators state that the narrowest portion of the stenosis should not be 47

compared to the area of post-stenotic dilatation as this would yield false results. As the trial used angiography, it says nothing about the degree of stenosis as measured by ultrasonography, a technique commonly used by some centres. This question has been the subject of other subsequent studies and is beyond the scope of this section. This large, randomized, multi-centre trial unequivocally demonstrated that surgery for 70–99% carotid stenosis reduces the rate of stroke or death and that this is a long-lasting effect. The publication of the 1991 Clinical Alert as well as the 1998 paper had a profound effect on the rates of carotid endarterectomy. Prior to this, rates of carotid endarterectomy fell as studies questioned the use of the procedure. After the NASCET trial, indications for surgery were clearly defined, and provided operations were performed by skilled surgeons in high-volume centres, benefit was clear. Therefore, rates increased in the 1990s but in selected patients. These results were subsequently confirmed.

1.12 Carotid Endarterectomy versus Carotid Stenting for Carotid Stenosis Details of Study Following the results of the NASCET trial (see section 1.11), treatment protocols for patients with symptomatic, severe carotid stenosis became better defined. With the advancement of endovascular stenting techniques, it became inevitable that stenting of carotid arteries would become a viable alternative to surgery. But which technique is better and which has the lowest complication rate? There are a number of studies in the 10-year period from 1998 to 2008 that have looked at this. These include two large, multi-centre European trials (SPACE and EVA-3S) that recruited 1200 and 527 patients respectively, and CAVATAS that recruited 504 patients (Coward et al., 2007; Eckstein et al., 2008; Mas et al., 2008). There have also been a number of smaller randomized trials including SAPPHIRE, WALLSTENT, Leicester, Kentucky A and Kentucky B. Two other large multi-centre trials (ICSS and CREST) are ongoing. On the basis of these studies, there are only minor differences between treatments in the immediate (30-day) period after surgery but longer-term follow-up may show differences. Here we concentrate on the SPACE study as it is the largest and one of the earlier trials.

Study References Main Study Eckstein HH, Ringleb P, Allenberg JR, Berger J, Fraedrich G, Hacke W, Hennerici M, Stingele R, Fiehler J, Zeumer H, Jansen O. Results of the Stent-Protected Angioplasty versus Carotid Endarterectomy (SPACE) study to treat symptomatic stenoses at 2 years: a multinational, prospective, randomised trial. Lancet Neurol 2008; 7: 893–902.

Related References Coward LJ, McCabe DJ, Ederle J, Featherstone RL, Clifton A, Brown MM, for the CAVATAS Investigators. Long-term outcome after angioplasty and stenting for symptomatic vertebral artery stenosis compared with medical treatment in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomized trial. Stroke 2007; 38: 1526–1530. Gurm HS, Yadav JS, Fayad P, Katzen BT, Mishkel GJ, Bajwa TK, Ansel G, Strickman NE, Wang H, Cohen SA, Massaro JM, Cutlip DE, for the SAPPHIRE Investigators. Long-term results of carotid stenting versus endarterectomy in high-risk patients. N Engl J Med 2008; 358: 1572–1579. Mas JL, Trinquart L, Leys D, Albucher JF, Rousseau H, Viguier A, Bossavy JP, Denis B, Piquet P, Garnier P, Viader F, Touzé E, Julia P, Giroud M, Krause D, Hosseini H, Becquemin JP, Hinzelin G, Houdart E, Hénon H, Neau JP, Bracard S, Onnient Y, Padovani R, Chatellier G, for the EVA-3S investigators. Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) trial: results up to 4 years from a randomised, multicentre trial. Lancet Neurol 2008; 7: 885–892.

Study Design PRCT. Class of evidence

Randomization

Non-blinded carotid endarterectomy versus carotid stenting

Number of patients 613 assigned to stenting versus 601 assigned to surgery Length of follow-up 2 years Number of centres

Stratification

All patients have at least 70% stenosis according to ECST or 50% according to NASCET

SPACE is one of the largest early trials looking at stent versus surgery for carotid stenosis. It is the equivalent of NASCET except that the two conditions are surgery and stenting rather than surgery and conservative treatment. Essentially, the study showed that there is little difference in outcome between the two groups. However, the 48

degree of re-stenosis is higher in the stent group. The results of this study are similar to other stent versus surgery trials. Each centre had to demonstrate, in advance, their expertise in dealing with carotid artery stenosis, and quality committees were set up to define guidelines. Multidisciplinary teams comprising interventionalists, vascular surgeons, and neurologists decided on each case. Inclusion criteria: symptomatic stenosis (TIA or stroke) of carotid bifurcation or internal carotid artery within 180 days (proven by Duplex sonography or angiography); mRS ≤2; >50 years of age; informed consent. Exclusion criteria: intracranial haemorrhage within last 90 days; uncontrolled arterial hypertension; known intracranial angioma; <2 years life expectancy; aspirin, clopidogrel or ASS contraindicated; contrast media contraindicated; planned simultaneous surgery; stenosis due to other reasons—external compression, dissection, recurrent after surgery, fibromuscular dysplasia, floating thrombus, intracranial stenosis and 100% stenosis.

Outcome Measures Primary Endpoints Ipsilateral stroke (cerebral infarction and/or bleeding with >24 h functional impairment) or death (any cause) from randomization up to 30th day (± 3 days) from treatment.

Secondary Endpoints Ipsilateral stroke or death from vascular causes within 24 months ± 14 days from randomization. Re-stenosis of at least 70% on duplex sonography at 6, 12, 24 months ± 14 days from randomization. Procedural technical failure, including re-stenosis on 6th ± 1 day or 30th ± 3 days from treatment. Ipsilateral stroke with Rankin score of 3 or more from randomization up to 30 ± 3 days. Incidence of any strokes within 30 ± 3 days. Incidence of any strokes within 24 months ± 14 days.

Results There were no differences in baseline characteristics between the two groups; 601 were randomized to endarterectomy and 613 to stenting. The 30-day rates of primary endpoint events in the intention-to-treat analysis was 6.45% in the surgery group versus 6.92% in the angioplasty group (p = 0.09). With adjustment for major protocol violations, this was 5.51% in the surgery versus 6.81% in the angioplasty group. Indeed, carotid endarterectomy patients tended to have better outcomes in most of the 30-day endpoints. More patients in the angioplasty group received double antiplatelet treatment (e.g. aspirin and clopidogrel) than in the surgery group. With regard to 2-year outcomes, there were no significant differences but the trend was for fewer complications in the surgery group. With regard to mortality, there were 32 deaths in the angioplasty group, compared to 28 in the carotid endarterectomy group (Kaplan–Meier estimates of 6.2% and 4.9% respectively (p = 0.63)). The Kaplan–Meier estimates of ipsilateral strokes or death within 2 years plus any periprocedural strokes or death was 9.5% in the stenting group versus 8.8% in the surgery group (p = 0.62). Recurrent stenosis of 70% or more were significantly more common in the stenting group (10.7%) compared to the surgery group (4.6%, p = 0.0009). Subgroup analyses revealed that there was an age-related increase in primary outcome events in patients with carotid angioplasty with stenting compared to no change across age groups in the surgical group. This means that patients <68 years of age had a lower peri-procedural risk in the angioplasty group and carotid endarterectomy had a lower peri-procedural risk in those patients aged >68 years.

Conclusions This study failed to show a non-inferiority of carotid angioplasty and stenting versus carotid endarterectomy for the 30-day complication rates. Indeed, overall, surgery was the favourable option. However, there was no difference between the two treatments and risk of cerebrovascular events at 2 years.

Critique The SPACE trial is a large, randomized trial that failed to show that carotid artery stenting is better than carotid endarterectomy. However, as commented on by Wiesmann et al. the trial was based on a frequency of 5% to reach the primary endpoint in each treatment arm (Wiesmann et al., 2008). As the frequency was much higher in each arm, it was underpowered to establish non-inferiority of stenting versus surgery and would require a further 1200–1800 patients. However, the results agree broadly with most of the other similar trials that are listed earlier in this section. The main difference with the SAPPHIRE trial is that the latter looks at patients at high risk for carotid endarterectomy and as a consequence has higher complication rates for both treatments. One of the main criticisms of these trials is that the follow-up (usually limited to 2 years) is aimed at complications related to the procedures rather than longterm re-stenosis rates. If long-term re-stenosis is due to progression of atherosclerosis, one might expect re-stenosis 49

to take >2 years. There is some evidence that most of the re-stenoses in the trial were related to intimal hyperplasia and this may have overestimated the incidence, though this has not been proven. A further criticism, stated by the SPACE investigators themselves, is that the trial did not look at secondary prevention strategies such as lipidlowering drugs or smoking status. In summary, of all of the studies, carotid angioplasty plus stenting has been shown to have a slightly higher periprocedural (up to 30 days) stroke risk but surgery has a higher rate of cranial nerve palsy or myocardial infarction. There is no difference between the periprocedural disabling stroke and death risks between the two groups. Also, there is no evidence that an embolic protection device influences outcome. Therefore, surgery is still the standard treatment for this condition but two further trials—ICSS and CREST with large patient numbers—are awaited.

Impact on Field This study has shown that there is little advantage of stenting over surgery for carotid stenosis, but at the same time there are only minor differences in complication rates. The main question now is whether the long-term re-stenosis rates will be any different and this will necessitate the outcomes of the CREST and ICSS trials.

Reference Wiesmann M, Schöpf V, Jansen O, Brückmann H. Stent-protected angioplasty versus carotid endarterectomy in patients with carotid artery stenosis: meta-analysis of randomized trial data. Eur Radiol 2008; 18(12): 2956–2966.

Chapter 2

Neuro-oncology RD Johnson, AJ Gogos, KJ Drummond, A Taha, KJO Khu, M Bernstein 2.0 Introduction 2.1 Steroids for the management of cerebral oedema associated with brain tumours 2.2 Surgery for single brain metastases 2.3 Adjuvant radiotherapy for single brain metastases 2.4 Stereotactic radiosurgery for brain metastases 2.5 Extent of resection of malignant glioma 2.6 Early concomitant systemic chemotherapy with radiotherapy for glioblastoma 2.7 Adjuvant localized chemotherapy (carmustine wafers) for malignant gliomas 2.8 Brachytherapy for malignant gliomas 2.9 Extent of resection of low-grade gliomas 2.10 Radiotherapy for low-grade gliomas 2.11 Dysembryoplastic neuroepithelial tumour 2.12 Meningioma resection grading 119

2.0 Introduction Controversies and debate in neuro-oncology are relevant to the practice of every neurosurgeon. There is a wealth of published literature in this area that continues to expand, seemingly exponentially. We have included, therefore, in this chapter a selection of studies that address certain key issues that we feel are directly relevant to the practising neurooncological surgeon today. We open this chapter by considering a randomized trial of the role of steroids in the management of cerebral oedema associated with brain tumours. The use of dexamethasone for this purpose was the result of work by Joseph Galicich, Lyle French, and James Melby from the University of Minnesota in the 1960s (Galicich et al., 1961). Their work has been described as ‘one of the greatest contributions in the history of neurosurgery’ and we would recommend that the interested reader consult a recent legacy review of the original work by McLelland and Long (McLelland and Long, 2008). The study chosen for inclusion in this chapter is the randomized trial carried out by Vecht et al. in the Netherlands to establish the efficacy and optimal dosing regimen for dexamethasone (Vecht et al., 1994). The next two sections address two important areas of neuro-oncology: the role of surgery and adjuvant radiotherapy for single brain metastases. Brain metastases occur in approximately 25% of patients with cancer and constitute the most common type of brain tumour. In those patients without a known primary lesion, neurosurgical intervention may be necessary in order to obtain a tissue diagnosis. In those with multiple cerebral metastases, radiotherapy is the accepted treatment option. However, the situation is more complex in patients with a single brain metastasis and known extracranial disease and we have included the three largest and most widely referenced trials that have addressed the role of surgery in addition to radiotherapy in this situation (Patchell et al., 1990; Vecht et al., 1993; Mintz et al., 1996). We have also included the only randomized trial to consider the role of radiotherapy as an adjunct to surgery for a single brain metastasis (Patchell et al., 1998). There is considerable controversy regarding the role of surgery for multiple cerebral metastases, and a survey of the literature reveals several conflicting views from published case series. Perhaps in a future edition we will be able to critique the results of a randomized controlled trial in this area. Stereotactic radiosurgery for brain metastases has also been evaluated by two prospective randomized trials and we have included these here (Andrews et al., 2004; Aoyama et al., 2006). The next sections of the chapter deal with the management of high-grade gliomas. Numerous studies have attempted to address the role of surgical resection in malignant gliomas and particularly whether ‘total’ resection is really beneficial. We have included four of the larger and well-designed studies addressing this issue (Keles et al., 1999; Lacroix et al., 2001; Vuorinen et al., 2003; Stummer et al., 2006). Following on from this we have looked at key studies evaluating the role of chemotherapy in high-grade glioma. The benefits of chemotherapy in glioblastoma have been controversial and this has been reflected by different approaches to the treatment of glioblastoma in Europe and the United States over the last 25 years. The role of early concomitant systemic temozolomide chemotherapy for glioblastoma was evaluated by the European Organisation for Research and Treatment of Cancer Brain Tumour and Radiotherapy Groups (EORTC) and the National Cancer Institute of Canada Clinical Trials Group (NCIC) and is familiarly known as the ‘temozolomide trial’ (Stupp et al., 2005). This trial showed the greatest improvement in 51

mortality since the Brain Tumour Study Group established the benefits of radiotherapy in 1978 (Walker et al., 1978). This temozolomide trial has led to the routine use of temozolomide chemotherapy in patients with glioblastomas. There is also a growing body of evidence to support the role of localized chemotherapy by way of carmustine wafers and we have included the two largest trials conducted on this issue (Brem et al., 1995; Valtonen et al., 1997; Westphal et al., 2003; Westphal et al., 2006). The next section considers brachytherapy for high-grade gliomas. This modality received quite a bit of attention in the 1980s. Several large randomized controlled trials, however, found that it was of no benefit. These are landmark studies as the findings resulted in the discontinuation of this therapy. We have considered two of these landmark trials in this section (Laperriere et al., 1998; Selker et al., 2002). The following sections deal with low-grade gliomas (LGGs)—two retrospective studies which evaluate the extent of resection on outcome for LGGs (McGirt et al., 2008; Smith et al., 2008). The role of radiotherapy in the management of LGGs presents a difficult dilemma for the neurosurgical oncologist. Several trials have been carried out to address this issue: EORTC I (‘Believers trial’); EORTC II (‘Non-believers trial’); and NCCTG-RTOG-ECOG trial (‘US trial’). We have, therefore, included a section that looks at these trials together (Karim et al., 1996; Karim et al., 2002; Shaw et al., 2002; van den Bent et al., 2005). In this second edition we have included two further sections relevant to neurosurgeons, which we feel were overlooked in the first edition. The first is a clinicopathological study describing dysembryoplastic neuroepithelial tumours (DNTs) (Daumas-Duport et al., 1988). This landmark study is pertinent to all neurosurgeons because of its description of a new surgically curable entity which if properly identified allows this subgroup of young patients to avoid unnecessary adjuvant therapy. Furthermore, the study demonstrates the limitations of our histopathological grading systems for brain tumours. The second study is Donald Simpson’s classic grading of meningioma resection and correlation with prognosis in terms of tumour recurrence (Simpson, 1957).

Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ, Maruyama Y, Kryscio RJ, Markesbery WR, Macdonald JS, Young B. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990; 322: 494–500. Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, Markesbery WR, Foon KA, Young B. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998; 280: 1485–1489. Selker RG, Shapiro WR, Burger P, Blackwood MS, Deutsch M, Arena VA, Van Gilder JC, Wu J, Malkin MG, Mealey J, Neal JH, Olson J, Robertson JT, Barnett GH, Bloomfield S, Albright R, Hochberg FH, Hiesiger E, Green S. The Brain Tumor Cooperative Group NIH Trial 87–01: a randomized comparison of surgery, external radiotherapy, and carmustine versus surgery, interstitial radiotherapy boost, external radiation therapy, and carmustine. Neurosurgery 2002; 51: 343–357. Shaw E, Arusell R, Scheithauer B, O’Fallon J, O’Neill B, Dinapoli R, Nelson D, Earle J, Jones C, Cascino T, Nichols D, Ivnik R, Hellman R, Curran W, Abrams R. Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group Study. J Clin Oncol 2002; 20: 2267–2276. Simpson D. The recurrence of intracranial meningiomas after surgical treatment. J Neurol Neurosurg Psychiatry 1957; 20: 22– 39. Smith JS, Chang EF, Lamborn KR, Chang SM, Prados MD, Cha S, Tihan T, Vandenberg S. Role of extent of resection in the long-term outcome of low-grade hemispheric gliomas. J Clin Oncol 2008; 10: 1338–1345. Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ; ALA-Glioma Study Group. Fluorescenceguided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled mulitcentre phase III trial. Lancet Oncol 2006; 7: 392–401. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB, Belander K, Brande AA, Marosi C, Bogdahn U, Curschmann K, Janzer RC, Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross G, Eisenhauer E, Mirimanoff RO, for the European Organisation for Research and Treatment of Cancer Brain Tumour and Radiotherapy Groups and the National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996. Valtonen S, Timonen U, Toivanen P, Kalimo H, Kivipetto L, Heiskanen O, Unsgaard G, Kuurne T. Interstitial chemotherapy with carmustine with carmustine-loaded polymers for high-grade gliomas: a randomised double-blind study. Neurosurgery 1997; 41: 44–48. Van den Bent MJ, Afra D, de Witte O, Ben Hassel M, Schraub S, Hoang-Xuan K, Malmström PO, Collette L, Piérart M, Mirimanoff R, Karim AB. EORTC Radiotherapy and Brain Tumor Groups and the UK Medical Research Council. Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial. Lancet 2005; 366: 985–990. Vecht CJ, Haaxma-Reiche H, Noordijk EM, Padberg GW, Voormenolen JH, Hoekstra FM, Tans JT, Lambooij N, Metsaars JA, Wattendorf AR, Brand R, Hermans J. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993; 33: 583–590. Vecht CJ, Hovestadt A, Verbiest HB, Verbiest HB, van Vliet JJ, van Putten WL. Dose-effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a randomized controlled study of doses 4, 8 and 16 mg per day. Neurology 1994; 44: 675–680. Vuorinen V, Hinkka S, Farkilla M, Jasskelainen J. Debulking or biopsy of malignant glioma in elderly people—a randomized study. Acta Neurochir (Wein) 2003; 145: 5–10. Walker MD, Alexanderander E Jr, Hunt WE, MacCarty CS, Mahaley MS Jr, Mealey J Jr, Norrell HA, Owens G, Ransohoff J, Wilson CB, Gehan EA, Strike TA. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. A cooperative clinical trial. J Neurosurg 1978; 49: 333–343. Westphal M, Hilt DC, Bortey E, Delavault P, Olivares R, Wamke PC, Whittle IR, Jääskeäinen J, Ram Z. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma. Neuro-Oncology 2003; 5: 79–88. Westphal M, Ram Z, Riddle V, Hilt D, Bortey E. Gliadel wafer in initial surgery for malignant glioma: long-term follow-up of a multi-center controlled trial. Acta Neurochir (Wein) 2006; 148: 269–275.

2.1 Steroids for the Management of Cerebral Oedema Associated with Brain Tumours Details of Study This is the only randomized study addressing the therapeutic efficacy of steroids to manage cerebral oedema in patients with brain tumours. The trial was carried out in the Netherlands in the early 1990s.

Study References Main Study Vecht CJ, Hovestadt A, Verbiest HB, Verbiest HB, van Vliet JJ, van Putten WL. Dose-effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a randomized controlled study of doses 4, 8 and 16 mg per day. Neurology 1994; 44: 675–680.

Related Reference Galicich JH, French LA, Melby JC. Use of dexamethasone in treatment of cerebral oedema associated with brain tumours. J

Lancet 1961; 81: 46â&#x20AC;&#x201C;53.

Study Design Double-blind RCT. Class of evidence

Randomization

Low-dose versus high-dose dexamethasone

Number of patients

Follow-up

8 weeks Primary endpoints: Neurological status Functional status Quality of life Secondary endpoint: Side effects

Number of centres

Stratification

None

Included patients: metastatic brain tumours on CT; Karnofsky score â&#x2030;¤80. Patients randomized between 4, 8, and 16 mg/day of dexamethasone but in two series: Series 1 (8 mg versus 16 mg); Series 2 (4 mg versus 16 mg).

Outcome Measures Primary Endpoints Neurological status. Functional status (Karnofsky score). Quality of life. Side effects: standardized questionnaire and clinical examination. Assessment at 1, 4, and 8 weeks.

Results

Ninety-two per cent follow-up at 4 weeks. Although patients receiving 16 mg/day had approximately 25% improvement on proximal muscle weakness in the first month there was no significant improvement in the subsequent month. Patients receiving 4 mg/day experienced <50% the number of cushingoid facies as those receiving 16 mg/day (p = 0.03). There were no significant differences in the improvement of Karnofsky scores between dosing regimens at 1 week or any other time point.

Conclusions After 1 week, 4 mg is as effective as 16 mg of dexamethasone in patients with no impending signs of brain herniation. Toxic effects of dexamethasone are dose dependent and are much more frequent if 16 mg is administered for prolonged periods (1 month or more).

Critique 54

Brain oedema is one of the greatest factors contributing to neurological decline and impairment of quality of life in patients with brain tumours. The use of steroids in the management of brain tumours was established in the 1950s and 1960s after a number of observations by several clinicians. In 1957, Kofman et al. had noted the relief of neurological symptoms in a patient with metastatic breast cancer who received prednisolone for adrenal suppression (Kofman et al., 1957). They reported benefits from the administration of prednisolone to a series of 20 patients with brain tumours (Kofman et al., 1957). Following this, Joseph Galicich, at the University of Minnesota, noted a circadian periodicity in the permeability of the blood–brain barrier in mice that was directly reciprocal to the endogenous corticosteroid circadian rhythms. This observation led to a trial showing that dexamethasone was beneficial in treating patients with neurological deficits from brain tumours (Galicich et al., 1961). This study by Galicich et al. led to the widespread use of dexamethasone in the treatment of cerebral oedema associated with brain tumours and has been referred to as arguably the ‘greatest translational research contribution in the history of neurosurgery’ (McLelland and Long, 2008). The use of dexamethasone is associated with the risk of adverse effects including cushingoid facies, psychosis, diabetes, and peptic ulceration. The later trial by Vecht et al. addresses the question of how the dose of dexamethasone affects efficacy and incidence of side effects. The follow-up period and outcome assessments used by Vecht and colleagues were appropriate to answer these questions. The trial design included two series because interim analysis revealed that the effect of a dose difference of 8 mg may be too small. The trial established the efficacy and dosing of dexamethasone for cerebral oedema in patients with brain tumours. Prior to this trial the standard dose of dexamethasone was 16 mg/day. On the basis of their results, Vecht and colleagues recommended the following dosing regimens: Neurological status of patient

Dosing regimen

↓ GCS or signs of ↑ ICP/impending herniation

10 mg IV stat + 4 × 4 mg/day orally

GCS 15/15 + no signs of ↑ ICP

4 mg/day orally

References Galicich JH, French LA, Melby JC. Use of dexamethasone in treatment of cerebral oedema associated with brain tumours. J Lancet 1961; 81: 46–53. Kofman S, Garvin JS, Nagamani D, Taylor SG. Treatment of cerebral metastases from breast carcinoma with prednisolone. JAMA 1957; 163: 1473–1476. McLelland S, Long DM. Genesis of the use of corticosteroids in the treatment and prevention of brain oedema. Neurosurgery 2008; 62: 965–968.

2.2 Surgery for Single Brain Metastases Details of Studies There have been three randomized trials evaluating the role of surgical resection in the treatment of solitary brain metastasis. All three trials compared surgical resection plus radiotherapy versus radiotherapy alone. The first study was carried out in the University of Kentucky in the United States between 1985 and 1989 (Patchell et al., 1990). The second study was carried out in the Netherlands between 1985 and 1991 (Vecht et al., 1993), and the third study was carried out in Canada (Mintz et al., 1996).

Study References Main Studies Mintz AH, Kestle J, Rathbone MP, Gaspar L, Hugenholtz H, Fisher B, Duncan G, Skingley P, Foster G, Levine M. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996; 78: 1470–1476. Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ, Maruyama Y, Kryscio RJ, Markesbery WR, Macdonald JS, Young B. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990; 322: 494–500. Vecht CJ, Haaxma-Reiche H, Noordijk EM, Padberg GW, Voormenolen JH, Hoekstra FH, Tans JT, Lambooij N, Metsaars JA, Wattendorf AR, Brand R, Hermans J. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993; 33: 583–590.

Related References Gaspar L, Scott C, Ratman M, Asbell S, Phillips T, Wasserman T, McKenna WG, Byhardt R. Recursive partitioning analysis (RPA) of prognostic factors in three radiation therapy oncology group (RTOG) brain metastases trials. In J Radiat Oncol Biol Phys 1997; 37: 745–751. Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, Markesbery WR, Foon KA, Young B. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998; 280: 1485–1489.

Study Designs 55

All RCTs. Goal of surgery in all three trials was total removal of the brain metastasis. Biopsy was undertaken to confirm the diagnosis in the radiotherapy arms in the studies by Patchell et al. and Mintz et al. but not in the study by Vecht et al. Stratification by location consisted of dividing lesions into supratentorial and infratentorial groups. All trials carried out an intention-to-treat analysis.

Inclusion and Exclusion Criteria

Outcome Measures Primary Endpoints Survival was the primary outcome in all three trials. In addition, the recurrence of intracerebral metastasis was a primary outcome in the Patchell et al. trial.

Secondary Endpoints Patchell et al. (1990)

Recurrence at the site of the original metastasis was confirmed with CT or MRI scan Quality of life defined as KPS of ≥70 (self-caring, but unable to work or maintain normal activity)

Vecht et al. (1993)

FIS defined as ≤1 on the 5-point WHO performance scale (0 = independent; 1 = symptoms but almost completely independent) and ≤1 on a 4-point neurological scale (0 = normal; 1 = minor symptoms)

Mintz et al. (1996)

Functional status: independence defined as proportion of time that patient had KPS of ≥70 Quality of life: measured using the Spitzer QOL 5-domain index

Results Follow-up for primary outcome was 100% in all the three trials.

Median Survival

WBRT = whole brain radiotherapy.

Other Significant Findings Patchell et al. (1990) With death from neurological causes used as an endpoint, median survival was greater in the surgery group compared to the whole brain radiotherapy (WBRT) group (62 weeks versus 26 weeks, p < 0.0009). Rate of recurrence of the original metastasis was much lower in the surgically treated group compared with those treated with radiotherapy alone and this was statistically significant (20% versus 52%, p < 0.02). Median length of time to recurrence was much lower in the surgery group compared to the WBRT group and this was statistically significant (21 weeks versus 59 weeks, p < 0.0001). Patients maintained quality of life (KPS ≥70) for much longer in the surgery group compared to the WBRT group and this was statistically significant (38 weeks versus 8 weeks, p < 0.005). Vecht et al. (1993) Patients with stable extracranial disease had a better FIS with combined treatment (p = 0.01). Mintz et al. (1996) There were no differences in the functional status or quality of life of patients in either treatment arm.

Conclusions Two out of the three trials found that surgery plus radiotherapy is superior to radiotherapy alone for single cerebral metastasis. Patchell et al. (1990) concluded that surgery plus WBRT results in longer life, fewer recurrences of brain metastasis, and a better quality of life for patients with brain metastases. Vecht et al. (1993) concluded that surgery is beneficial for patients with single brain metastases and stable extracranial disease but WBRT is sufficient for patients with progressive extracranial disease within the previous 3 months. Mintz et al. (1996) concluded that the addition of surgery to radiation therapy did not improve the outcome of 57

patients with a single brain metastasis.

Critique Prior to 1990 the only evidence supporting the role of surgery for single cerebral metastases in patients with known systemic disease came from published case series that may have been biased by selecting only patients in good clinical condition for surgery. Uncontrolled case series looking at the effectiveness of surgery in brain metastases provided conflicting results and, therefore, the role of surgery in these patients remained to be determined. Previous studies had revealed that the median survival for untreated brain metastases is approximately 1 month but this could be improved to 3 months in the majority of patients with steroids and WBRT (Cairncross et al., 1980). It is important to note that only half of patients with brain metastases have a single metastasis. Of these, another half will not be eligible for surgery for reasons of extracranial disease. Therefore, only a small portion of patients with solitary brain metastasis are really candidates for surgery. Patchell et al. (1990) analysed deaths according to whether they were from neurological dysfunction or due to systemic disease. They found that the reason for the overall longer survival in the surgery group was a large reduction in deaths from neurological dysfunction, there being no effect on deaths from systemic disease. Overall survival was still <10% at 3 months in the Patchell et al. study. Thus, although there was no difference in long-term survival, their results clearly favour a benefit of surgery in terms of survival and quality of life in patients with single metastases and limited, well-controlled extracranial disease. Patchell et al. found that the only two factors reducing the risk of recurrence of the original metastasis were surgical treatment and the absence of disseminated disease (multivariate analysis). However, surgery had no effect on the development of other brain metastases. In terms of quality of life they reported a significant benefit of surgery. However, older patients and those with disseminated disease were associated with poor quality of life. A potential weakness of Patchell et al.â&#x20AC;&#x2122;s 1990 study is that it is a small study and could, therefore, be affected by inadequate randomization and it reduces the probability of including long-term survivors of surgery (Posner, 1990). Nonetheless, their study represents a landmark in neurosurgery for being the first published RCT evaluating the role of surgery in the management of single brain metastases. Indeed, all three trials considered here are relatively small in size. The study by Vecht et al. (1993) has been criticized for using a non-standard radiation dose (up to 40 Gy/day). In addition, in their study no MRI was used to confirm whether metastases were indeed solitary. Although 50% of patients with extracranial cancer will have a single brain metastasis detected on CT scan, this figure is reduced to 30% with MRI. It is possible, therefore, that in the study by Vecht et al. the patient population contained patients with more than one intracranial metastasis. The study by Mintz et al. (1996) has been criticized by for including a large proportion of patients with active systemic disease and for differences in the distribution of histological diagnoses between the two groups (Wronski and Lederman, 1997). The surgical group containing tumours with poorer prognoses on the basis of histology (more colorectal primaries) compared to the radiation group (more breast tumours). However, the authors have refuted this as a potential source of bias arguing that the most important factor influencing survival was the extent of extracranial disease, not individual histology (Levine, 1997). All three studies have been criticized for containing a variety of pathologies with different radiosensitivities. Ideally, a trial would include only radiosensitive tumours. Nevertheless, the trials reflect realistic clinical practice. Furthermore, there has at least one other trial attempting to address the role of radiotherapy for solitary brain metastases, but this was closed prematurely (Roos et al., 2006). Further controlled studies are required to evaluate the role of surgery in different tumour types. In addition, there is a need for controlled studies evaluating the role of surgery in multiple cerebral metastases. However, these three trials have established the role of surgical resection as a treatment option in the management of patients with a single brain metastasis. It is significant that the trial by Mintz et al. (1996) appears to contradict the findings of the previous two trials. As has been mentioned already, the results of this third trial may have been influenced by the inclusion of patients with factors associated with a poor outcome with surgery. This led to a number of studies to assess the prognostic factors in patients with cerebral metastasis in order to facilitate the selection of good surgical candidates. The most important of these studies was carried out by the Radiation Therapy Oncology Group (RTOG), which developed the recursive partitioning analysis (RPA) classification on the basis of a retrospective study of 1200 patients. The RPA classification is a statistical method of classifying patients on the basis of several factors: KPS score; age; and extent of extracranial disease (Gaspar et al., 1997). The RPA classification is as follows: Class I

KPS â&#x2030;Ľ70 Age <65 years Controlled primary tumour Brain is only site of metastasis

Class II

KPS ≥70 Age >65 years Uncontrolled systemic disease/other symptomatic metastases

Class III

KPS <70

In their study of 1200 patients treated with radiotherapy, the authors found that the median survival for the different classes was as follows: Class I (7.1 months); Class II (4.2 months); and Class III (2.3 months). The RPA classification has also been shown to have prognostic value in patients treated surgically (Agboola et al., 1998). Class I patients are the group most likely to benefit from craniotomy, although it is important to consider the histology of the primary tumour as well, e.g. patients with metastasis from renal cell carcinoma or melanoma may have poorer survival rates than breast cancer metastasis (Sills, 2005). If RPA classification had been used in the three trials reviewed in this section it is possible that all three may have shown a beneficial effect of surgery. It is likely that RPA classification will be used in the design and analysis of future trials.

References Agboola O, Benott B, Cross P, Da Silva V, Esche B, Lesuik H, Gonsalves C. Prognositc factors derived from recursive partition analysis (RPA) of radiation therapy oncology group (RTOG) brain metastasis trials applied to surgically resected and irradiated brain metastatic cases. Int J Radiation Oncology Biol Phys 1998; 42: 155–159. Cairncross JG, Jim JH, Posner JB. Radiation therapy for brain metastases. Ann Neurol 1980; 7: 175–224. Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, McKenna WG, Byhardt. Recursive partitioning analysis (RPA) of prognostic factors in three radiation therapy oncology group (RTOG) brain metastases trials. In J Radiat Oncol Biol Phys 1997; 37: 745–751. Levine M. Correspondence: a randomised trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis—author reply. Cancer 1997; 80: 1003–1004. Posner JB. Surgery for metastases to the brain. N Engl J Med 1990; 322: 544–545. Roos DE, Wirth A, Burmeister BH, Spry NA, Drummond KJ, Beresford JA, McClure BE. Whole brain irradiation following surgery or radiosurgery for solitary brain metastases: mature results of a prematurely closed randomized TransTasman Radiation Oncology Group trial (TROG 98.05). Radiother Oncol 2006; 80: 318–322. Sills AK. Current treatment approaches to surgery for brain metastases. Neurosurgery 2005; 57: S4–S24. Wronski M, Lederman G. Correspondence: a randomised trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1997; 80: 1002–1003.

2.3 Adjuvant Radiotherapy for Single Brain Metastases Details of Study This study is the only RCT looking at WBRT as an adjunct to surgery. The study was carried out in Kentucky, United States, in the 1990s.

Study References Main Study Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, Markesbery WR, Foon KA, Young B. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998; 280: 1485–1489.

Related Reference Patchell RA, Regine WF. The rationale for adjuvant whole brain radiation therapy with radiosurgery in the treatment of single brain metastases. Technol Cancer Res Treat 2003; 2: 111–115.

Study Design Multi-centre, parallel group RCT. Class of evidence

Randomization

Surgery versus surgery + WBRT

Number of patients

Follow-up

Lifespan of patients Primary endpoint: Recurrence of tumour Secondary endpoints: 59

Survival Cause of death Functional independence Number of centres

>1 (unspecified)

Stratification

Extent of disease Primary tumour type

Inclusion criteria: patients >18 years with single metastasis completely resected. Exclusion criteria: incomplete resection; leptomeningeal metastases; previous radiotherapy; other malignancy; KPS <70; need for emergency surgery; highly radiosensitive primary tumours (e.g. small cell lung cancer). MRIs were carried out to establish the presence of single metastases. DXT started within 28 days of surgery. Intention-to-treat analysis.

Outcome Measures Primary Endpoint Recurrence of metastases: 3-monthly MRIs for the first year and then 6-monthly; recurrence divided into original (site of resection) or distant (other site in brain).

Secondary Endpoints Length of survival. Cause of death. Functional independence: defined as period of time with KPS >70.

Results

Conclusions Routine post-operative WBRT reduces recurrence of brain metastases and reduces death from neurological causes.

Critique This trial is unique as WBRT is an established treatment for brain metastases. Indeed, the trial has been criticized for looking at a somewhat unconventional approach for the management of brain metastases. However, the use of WBRT as an adjuvant therapy had been looked at in clinical series but no clinical trial had previously been carried out. Following their study reported in 1990 looking at surgery plus WBRT versus WBRT alone, the question as to whether WBRT after surgery has any benefit remained to be addressed by way of a RCT. The primary endpoint of this trial was recurrence of metastasis in the brain and the trial size was based on answering this question. The trial was not powered to make conclusions regarding survival. The results need to be interpreted, therefore, with this in mind. Certainly, the results showed that the recurrence of the brain metastasis was significantly lower in those receiving post-operative WBRT. There was no difference in overall mortality in patients receiving surgery alone versus those receiving post-operative DXT. However, patients in the surgery plus DXT group were much more likely to die from systemic causes, while those in the DXT group were much likely to die from neurological causes. Although the trial was not large enough to extrapolate these findings to the wider population, the findings are in keeping with the previous trial by Patchell et al. which looked at WBRT with or without surgery in these patients (Patchell et al., 1990). It appears, therefore, that treatment focused on the single brain metastasis may well reduce the incidence of death from brain disease. This finding suggests that the treatment arms alter the mode of, 60

but not the time of, death and begs the question as to whether one cause of death is more acceptable by another to patients and their families. Patchell et al. concluded that their results supported the routine use of WBRT in patients with single brain metastases. However, it has been pointed out that a subset of patients in the surgery group who died of systemic causes had the longest median survival of 88 weeks and this would suggest that DXT may be detrimental in this subgroup of patients (Carol and Rosa, 1999). Nonetheless, this criticism is based on a post hoc analysis of a small number of patients only and Patchell et al. have emphasized that the primary endpoint of their study was recurrence of the original brain metastasis and not survival (Patchell et al., 1999). This study by Patchell et al. is the only randomized prospective study evaluating the role of WBRT as an adjuvant treatment to surgical resection and as such is a landmark study in neuro-oncology. There is another randomized phase III trial underway for melanoma metastases and the results of this are keenly anticipated (Fogarty et al., 2011).

References Carol W, Rosa S. Letters to the editor. JAMA 1999; 281: 1695. Fogarty G, Morton RL, Vardy J, Nowak AK, Mandel C, Forder PM, Hong A, Hruby G, Burmeister B, Shivalingam B, Dhillon H, Thompson JF. Whole brain radiotherapy after local treatment of brain metastases—a randomised phase III trial. BMC Cancer 2013; 11: 142. Patchell RA, Tibbs PA, Regine WF, Mohiuddin M, Kryscia RH, Markesbery WR, Foon KA, Young B, Dempsey RJ. Author reply. JAMA 1999; 281: 1696.

2.4 Stereotactic Radiosurgery for Brain Metastases Details of Studies Two multi-centre randomized trials have been carried out to evaluate the role of stereotactic radiosurgery in the management of cerebral metastases. The first trial, carried out by RTOG in North America between 1996 and 2001, compared WBRT with or without stereotactic radiosurgery (SRS) in patients with one to three metastases. The second trial, carried out by the Japanese Radiation Oncology Study Group (JROSG) in Japan between 1999 and 2004, compared SRS with or without WBRT in patients with one to four metastases.

Study References Main Studies Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ Jr. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004; 363: 1655–1672. Aoyama H, Shirato H, Tago T, Nakagawa K, Toyoda T, Hatano K, Kenjyo M, Oya N, Hirota S, Shioura H, Kunieda E, Inomata T, Hayakawa K, Katoh N, Kobashi G. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for the treatment of brain metastases—a randomised controlled trial. JAMA 2006; 21: 2483–2491.

Related Reference Raizer J. Radiosurgery and whole-brain radiation therapy for brain metastases either both or as the optimal treatment. JAMA 2006; 295: 2535–2536.

Study Design Class of evidence

RTOG trial I

JRSOG trial I

Randomization

WBRT versus WBRT + SRS

SRS versus SRS +WBRT

Number of patients

333

132

Follow-up

100%

Primary endpoint: Survival

Primary endpoint: Overall survival

Secondary endpoints: Tumour response Brain tumour recurrence Functional performance Cause of death

Secondary endpoints: Brain tumour recurrence Salvage brain treatment Functional preservation Toxic effects of radiation Cause of death

Number of centres Stratification

11 Number of brain metastases Extent of extracranial disease

Number of brain metastases Extent of extracranial disease Primary tumour site (lung versus other sites)

Intention-to-treat analyses used in both trials. RTOG eligibility: • Inclusion criteria: age ≥18 years; one to three brain metastases confirmed with contrast MRI; metastasis diameter ≤4 cm; KPS ≥70; consent. • Exclusion criteria: brainstem metastases; active systemic disease; deranged haematology (e.g. low haemoglobin, platelets, or neutrophil count); RPA class III. JRSOG eligibility: • Inclusion criteria: age ≥18 years; one to four brain metastases confirmed with contrast MRI; metastasis diameter ≤3 cm; histological confirmation of systemic disease; KPS ≥70; consent. • Exclusion criteria: small cell lung carcinoma; lymphoma; germinoma; multiple myeloma. Patients in the RTOG study all received WBRT in daily 2.5 Gy fractions to a total of 37.5 Gy with SRS doses varying according to the size of the metastases. Patients in the JRSOG study who received WBRT were given 10 fractions to a total of 30 Gy and SRS doses varied according to size of the metastases.

Results Overall Survival In the RTOG trial, a survival benefit was found with the addition of SRS to WBRT in those patients with a single brain metastasis: mean survival time in the WBRT plus SRS group was 6.5 months compared to 4.9 months in the WBRT-alone group (p < 0.04).

Brain Tumour Recurrence RTOG trial: radiological evaluation at 1 year revealed better control in the WBRT plus SRS group (82%) than in the WBRT-alone group (71%, p = 0.01). However, this evaluation was carried out in < 23% of the enrolled patients. JRSOG trial: recurrence rate at 12 months was much greater in the SRS-alone group (76.4%) than in the SRS plus WBRT group (46.8%; p < 0.001).

Functional Performance RTOG trial: KPS scores were more likely to have improved at 6-month follow-up in the WBRT plus SRS group (43%) than in the WBRT-alone group (27%, p = 0.03). JRSOG trial: although no significant differences in KPS scores were found between groups it is noteworthy that significantly more patients in the SRS-alone group (86%) showed neurological deterioration attributable to progression of brain metastases than those in the SRS plus WBRT group (59%, p = 0.05).

Cause of Death No significant differences in the causes of death (neurological versus non-neurological) were found between the groups in either study.

Conclusions RTOG Trial An SRS boost following WBRT is better than WBRT alone for single brain metastasis and should be a standard treatment for all patients with a single brain metastasis. 62

An SRS boost following WBRT improves performance in all patients with up to three brain metastases and should be considered for all patients with two to three brain metastases.

JRSOG Trial The addition of WBRT to SRS does not improve survival in patients with brain metastases. Intracranial relapse was more likely with SRS alone, but without increased risk of neurological death.

Critique Intracranial metastases occur in approximately one-third of patients with cancer. WBRT has been the mainstay of treatment for brain metastases with surgery. The rationale for WBRT is based on the assumption that haematogenous spread of the primary tumour has seeded the whole brain and that, in addition to macroscopic metastases, there will be microscopic metastases that will remain occult on conventional neuroimaging modalities such as CT and MRI. However, an alternative view is that intracranial disease may be actually locally limited. Certainly, surgery appears to be beneficial in patients with single metastases who are good surgical candidates. However, the majority of patients will have multiple metastases and SRS has several potential advantages that may be beneficial in this situation. SRS allows for the focal administration of high-dose radiation to multiple lesions including some that may be otherwise surgically inaccessible. In addition, SRS may potentially reduce the neurotoxic effects of WBRT if it is effective. The two trials considered here have taken different approaches to evaluate the effects of SRS in the management of cerebral metastases. The RTOG study evaluated the addition of SRS to conventional WBRT, whereas the JRSOG study took the alternative approach of evaluating the addition of WBRT to SRS therapy. In this way, the JRSOG study has some philosophical similarities with the study by Patchell et al. which considered the addition of WBRT to surgery for single brain metastasis (Patchell et al., 1998). However, Patchell and colleagues have published one of the more unforgiving critical appraisals of the JRSOG study in which they raise several points of contention (Patchell et al., 2006). They argue that as it was unlikely that any survival difference would be seen between the two groups, it was incorrect to conduct power calculations on the basis of a potential survival benefit. They propose that the study should have been powered to calculate non-inferiority of SRS alone and that this would have required the recruitment of 17-fold more patients than are actually included in the study, indicating that the JRSOG study is grossly underpowered. The view of Patchell et al. is that the only statistically supported finding in the JRSOG trial is that the addition of WBRT to SRS reduces recurrence of brain metastases. They express the view, therefore, that the JRSOG supports the upfront use of WBRT, a conclusion that would be supported by the finding that fewer patients in the SRS plus WBRT suffered neurological deterioration from tumour progression. The RTOG trial may have been hindered by several factors, not the least of which was the inclusion of patients who had undergone surgical resection of brain metastases without any stratification. In addition, there was limited complete radiological follow-up of patients (less than half the patient population). Notwithstanding the limitations of these studies they are both landmark studies. The RTOG study is the first completed large multi-centre randomized trial evaluating the role of a SRS boost following WBRT for brain metastases. The JRSOG study, although subject to much criticism, has also added significantly to the debate regarding the role of SRS in the management of brain metastases. A phase III trial evaluating microsurgery plus WBRT versus SRS alone for single brain metastases was recently published but unfortunately due to patient accrual the study was discontinued early and so no conclusions could be drawn (Muacevic et al., 2008). The role of SRS will be further elucidated with large, well-designed, multicentre randomized trials.

References Muacevic A, Wowra B, Siefert A, Tonn JG, Steiger HJ, Kreth FW. Microsurgery plus whole brain irradiation versus gamma knife surgery for treatment of single metastases to the brain: a randomised controlled multicentre phase III trial. J Neurooncol 2008; 87: 299–307. Patchell RA, Regine WF, Renschler M, Loeffler JS, Sawaya R, Chin LS, Andrews DW. Editorial: comments about the prospective randomised trial by Aoyama et al. Surg Neurol 2006; 66: 459–460.

2.5 Extent of Resection of Malignant Glioma Details of Studies Numerous clinical series have evaluated the relationship between the extent of resection (EOR) and survival in patients with malignant glioma. Although ‘total’ resection appears to prolong life more than subtotal resection, this may represent a selection bias towards younger, fitter patients with tumours in non-eloquent regions for more aggressive surgery. The advent of CT and MRI has allowed for the more accurate assessment of the extent of tumour resection. Further, surgical adjuncts such as fluorescent compounds and intraoperative MRI have attempted to improve assessment of EOR in theatre. Four studies have been selected for inclusion here. The first two are the better and larger of the earlier retrospective analyses of patients undergoing surgery for malignant glioma. These were 63

carried out at the Washington Medical Center in Seattle, Washington (Keles et al., 1999), and the MD Anderson Cancer Center in Houston, Texas (Lacroix et al., 2001), in the United States. The third is the only RCT that compares biopsy with resection and was carried out in Finland (Vuorinen et al., 2003). The final paper is the first multi-centre RCT comparing two different methods of surgical resection (fluorescence-assisted versus conventional surgery) for malignant glioma and was co-ordinated from Germany (Stummer et al., 2006).

Study References Main Studies Keles GE, Anderson B, Berger MS. The effect of extent of resection on time to tumor progression and survival in patients with glioblastoma mulitforme of the cerebral hemisphere. Cancer 1999; 74: 1784–1791. Lacroix M, Abi-Said D, Fourney DR, Gokaslan ZL, Shi W, DeMonte F, Lang FF, McCutcheon IE, Hassenbusch SJ, Holland E, Hess K, Michael C, Miller D, Sawaya R. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001; 95: 190–198. Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ; ALA-Glioma Study Group. Fluorescenceguided surgery with 5-alaminovulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol 2006; 7: 392–401. Vuorinen V, Hinkka S, Farkilla M, Jasskelainen J. Debulking or biopsy of malignant glioma in elderly people—a randomized study. Acta Neurochir (Wein) 2003; 145: 5–10.

Related Reference Stummer W, Reulen HJ, Meinel T, Pichlmeier U, Schumacher W, Tonn JC, Rohde V, Oppel F, Turowski B, Woiciechowsky C, Franz K, Pietsch T, for the ALA-Glioma Study Group. Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias. Lancet Oncol 2008; 7: 392–401.

Study Designs

* Stummer et al.’s paper may be considered class I evidence for the role of 5-ALA in patients with high-grade gliomas. However, it should only be considered class II evidence for the question under consideration in this chapter, as maximal safe resection was attempted in both treatment arms.

Keles et al. (1999) Inclusion criteria: KPS >70 and a pathological diagnosis of GBM. Radiological extent of resection was assessed by volumetric measurements of pre-operative and post-operative CTs and/or MRI. 64

Lacroix et al. (2001) Included 416 consecutive patients with a histological diagnosis of GBM. EOR calculations were based on pre- and post-surgical enhancing tumour volumes, or high T2 signal for nonenhancing lesions. Multivariate analyses were performed, including other known predictors of survival, such as age and KPS.

Vuorinen et al. (2003) Inclusion criteria: radiological diagnosis of supratentorial malignant glioma; KPS >60 and age >65. All patients received radiotherapy unless their clinical state deteriorated too far. Intention-to-treat analysis.

Stummer et al. (2006) Total resection was defined as the absence of contrast enhancement of MRI at 72 h. Fluorescence was achieved by the pre-operative administration of 5-aminolevulinic acid (5-ALA) and performance of surgery under blue (400â&#x20AC;&#x201C;410 nm wavelength) light. Conventional surgery was performed under white light. PFS at 6 months was assessed with MRI. The study was terminated early and 270 patients were included in the analysis.

Results Keles et al. (1999) EOR also showed a significant correlations with post-operative KPS (p < 0.05).

Lacroix et al. (2001)

Vuorinen et al. (2003)

Stummer et al. (2006)

Conclusions Keles et al. (1999) The extent of tumour resection in glioblastoma multiforme affects overall survival and time to progression.

Lacroix et al. (2001) 65

Greater than 98% tumour debulking is associated with improved survival.

Vuorinen et al. (2003) Craniotomy and debulking offers a modest survival advantage over biopsy in elderly patients with GBM.

Stummer et al. (2006) Fluorescence-assisted surgery with 5-ALA enables more complete resection of malignant glioma with improved PFS.

Critique Although the study by Keles and colleagues is only a retrospective series, it is to be commended because it attempts to reduce the effect of selection bias by using strict inclusion criteria. In addition, the authors used radiological assessments of the extent of tumour resection rather than estimates by the operating surgeon. The study was restricted to glioblastoma multiforme, not all high-grade glioma. The results of this study are informative and may be generally representative of the effect of the extent of resection in high-grade glioma. Lacroix and colleagues’ sample is larger than Keles et al.’s and they performed multivariate analyses, demonstrating that extent of resection was still associated with improved survival when known confounders are included. Both papers are landmark because they used radiological measures of tumour resection and attempted to reduce bias from other prognostic variables. Their findings were supported by Sanai and colleagues’ study of 500 patients (Sanai et al., 2011), validating the retrospective association in the modern era. Further, they showed that lower percentage resections were still associated with improved survival. The study by Vuorinen and colleagues is limited by the small number of patients and the restriction to elderly. However, this inclusion criterion was felt necessary by the authors to provide equipoise in randomizing patients to biopsy only. The results may only be generalized to younger patients with caution. The study did show a statistically significant benefit of resection in patients >65 years. It is landmark because it is the only study to provided randomized evidence for extent of resection. There were, however, significant methodological short-comings in this study, including the small number of patients included and the fact that, on final analysis, not all had malignant glioma. Stummer and colleagues’ RCT of a surgical adjunct provides high-level evidence for extent of resection in glioblastoma multiforme. The study was a well-designed, international, multi-centre trial. In a subsequent analysis, the authors have been able to compare survival rates in patients with and without ‘complete’ resection (‘complete’ according to their radiological definition). They report that survival is greater in patients with ‘complete’ resection (16.7 versus 11.8 months, p < 0.0001). The authors acknowledge that their trial is level II evidence for the effect of extent of resection on survival. It is, however, unlikely that we will ever have level I evidence for this question and papers such as Stummer et al.’s are likely to be the best evidence available.

References Sanai N, Polley MY, McDermott MW, Parsa AT, Berger MS. An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg 2011; 115: 3–8. Stummer W, Reulen HJ, Meinel T, Pichlmeier U, Schumacher W, Tonn JC, Rohde V, Oppel F, Turowski B, Woiciechowsky C, Franz K, Pietsch T, for the ALA-Glioma Study Group. Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias. Lancet Oncol 2006; 7: 392–401.

2.6 Early Concomitant Systemic Chemotherapy with Radiotherapy for Glioblastoma Details of study The role of systemic temozolomide chemotherapy for glioblastoma was evaluated by the EORTC and the NCIC and is familiarly known as the ‘temozolomide trial’ or the ‘Stupp protocol’ (Stupp et al., 2005). This trial was carried out in Europe and North America between 2000 and 2002 and has led to the routine use of temozolomide chemotherapy in patients with glioblastomas.

Study References Main Study Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB, Belander K, Brande AA, Marosi C, Bogdahn U, Curschmann K, janzer RC, Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross G, Eisenhauer E, Mirimanoff RO, for the European Organisation for Research and Treatment of Cancer Brain Tumour and Radiotherapy Groups and the National Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996.

Related Reference Hegi R, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, Kros JM, Hainfellner JA, Mason W, Mariani L,

Bromberg JEC, Hau P, Mirimanoff RO, Cairncross G, Janzer RC, Stupp R. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352: 997–1003.

Study Design A prospective multi-institutional RCT. Class of evidence

Randomization

DXT alone versus DXT + temozolomide

Number of patients

573

Follow-up

Lifespan of patients Primary endpoint: Overall survival Secondary endpoints: Quality of life Safety PFS

Number of centres

85 centres in 15 countries

Stratification

WHO performance status

Inclusion criteria: newly diagnosed glioblastoma multiforme (GBM); age 18–70; ‘good clinical state’; consent. A total of 60 Gy radiotherapy was administered in daily 2 Gy fractions over 6 weeks (5 days per week). Patients in the temozolomide arm were also given continuous daily temozolomide (75 mg/m2 of body surface area per day) for the duration of the radiotherapy and this was followed by six adjuvant cycles of temozolomide (150–200 mg/m2 for 5 days during every 28-day cycle).

Outcome Measures Primary Endpoint Overall survival was analysed using the Kaplan–Meier method.

Secondary Endpoints Safety: patients were evaluated regularly for adverse events including haematological monitoring. Quality of life questionnaires were also used during regular follow-up. PFS was also analysed using the Kaplan–Meier method.

Results The median age of patients was 56 years and 86% underwent debulking surgery. Ninety-three per cent of patients had a histological confirmation of glioblastoma.

The relative reduction in risk of death in patients receiving temozolomide was 37% (hazard ratio of 0.63 compared to the DXT-alone group). The hazard ratio for death or disease progression in the temozolomide group compared to the DXT-alone group was 0.54. Two-year survival rates were greater in the temozolomide group (26.5%) compared to the DXT-alone group (10.4%). A translational study found that O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation is associated with a survival benefit for temozolomide + DXT (21 months versus 15 months, p < 0.05). There were no significant differences in the safety profiles of each treatment arm. There was no adverse effect on quality of life by the addition of temozolomide therapy. Subgroup analysis did not reveal any difference in survival advantage between subgroups.

Conclusions The early addition of temozolomide chemotherapy to radiotherapy has a significant survival advantage in patients with glioblastoma.

Critique Chemotherapy given before, or as an adjuvant to radiotherapy has been evaluated in multiple clinical trials and found to have minimal impact on survival in patients with glioblastoma (Fine et al., 1993; Stewart, 2002). This temozolomide trial addressed the question as to whether early addition of temozolomide chemotherapy to radiotherapy would have any advantage over radiotherapy alone. The concomitant use of a chemotherapeutic agent with radiotherapy was an unconventional approach and this is a landmark study as it reports the greatest improvement in survival (>50%) in patients with glioblastoma since the trial of radiotherapy versus chemotherapy with an alkylating agent over a quarter of a century earlier (Walker et al., 1978). This may reflect the selection of patients with good prognostic indicators (tumour resection, young age, good functional status) and the less toxic effects of temozolomide compared to other nitrosourea-based chemotherapeutic agents. The beneficial effects of temozolomide have more recently been reported to last up to 5 years (Stupp et al., 2009). The translational study by Hegi et al. found that methylation of the MGMT promoter was associated with an even greater survival advantage (Hegi et al., 2005). However, the clear survival advantage seen for all patients receiving temozolomide would appear to negate the value of reserving temozolomide for patients in this subgroup alone and so this study does not necessarily have the same impact as molecular genetic classification of oligodendrogliomas has had (Cairncross et al., 1998). The temozolomide trial demonstrates a survival benefit with low toxicity and has led to a new standard of care in patients with malignant glioma (Mason and Cairncross, 2005). The question remains as to whether the concomitant or the adjuvant phase of temozolomide is the more important contributor to survival, or if, indeed both are necessary. This question is being addressed in the CATNON phase III multi-centre trial for anaplastic astrocytoma, which has four treatment arms, examining each of the components of the ‘Stupp protocol’.

References Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein DM, Hammond RR, Silver JS, Stark PC, Macdonald DR, Ino Y, Ramsay DA, Louis DN. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998; 90: 1473–1479. Fine HA, Dear KB, Loeffer JS, Black PM, Canellos GP. Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults. Cancer 1993; 71: 2585–2597. Hegi R, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, Kros JM, Hainfellner JA, Mason W, Mariani L, Bromberg JEC, Hau P, Mirimanoff RO, Cairncross G, Janzer RC, Stupp R. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352: 997–1003. Mason WP, Cairncross JG. Drug insight: temozolomide as a treatment for malignant glioma—impact of a recent trial. Nat Clin Pract Neurol 2005; 1: 88–95. Stewart LA. Chemotherapy in adult high-grade glioma: a systematic review and meta-analysis of individual patient data from 12 randomised trials. Lancet 2002; 359: 1011–1018. Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Janzer RC, Ludwin SK, Allgeier A, Fisher B, Belanger K, Hau P, Brandes AA, Gijtenbeek J, Marosi C, Vecht CJ, Mokhtari K, Wesseling P, Villa S, Eisenhauer E, Gorila T, Weller M, Lacombe D, Cairncross JG, Mirimanoff RO, for the European Organisation for Research and Treatment of Cancer Brain Tumour and Radiation Oncology Groups; National Cancer Institute of Canada Clinical Trials Group. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trail. Lancet Oncol 2009; 10: 459–466. Walker MD, Alexander E Jr, Hunt WE, MacCarty CS, Mahaley MS Jr, Mealey J Jr, Norrell HA, Owens G, Ransohoff J, Wilson CB, Gehan EA, Strike TA. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. A cooperative clinical trial. J Neurosurg 1978; 49: 333–343.

2.7 Adjuvant Localized Chemotherapy (Carmustine Wafers) for Malignant Gliomas Details of Studies Three randomized controlled trials have been carried out to evaluate the effect of local administration of carmustine wafers to the tumour site in patients with malignant gliomas. The first was performed in the United States between 1989 and 1993 by the Polymer Brain Tumor Treatment Group (PBTTG) and evaluated their efficacy against placebo in the treatment of recurrent glioblastomas requiring reoperation (Brem et al., 1995). The second trial, which was carried out in Finland between 1992 and 1993, was designed to evaluate the efficacy of carmustine wafers applied at the time of the first operation (Valtonen et al., 1997). Unfortunately, this trial included only a small number of patients and was discontinued early because of difficulties in availability of carmustine wafers. However, a third much larger trial, carried out by the Gliadel Study Group (GSG) in both North America and Europe between 1997 and 2000, also 68

evaluated the efficacy of carmustine wafers applied to the resection cavity (Westphal et al., 2003). A further longterm follow-up of this trial has also been published (Westphal et al., 2006). The PBTTG and GSG trials are summarized here.

Study References Main Studies Brem H, Piantadosi S, Burger PC, Walker M, Selker R, Vick NA, Black K, Sisti M, Brem S, Mohr G, Muller P, Morawetz R, Schold SC, for the Polymer-Brain Tumor Treatment Group. Placebo-controlled trial of safety and efficacy of intraoperative controlled delivery by biodegradable polymers of chemotherapy for recurrent gliomas. Lancet 1995; 345: 1008–1012. Westphal M, Hilt DC, Bortey E, Delavault P, Olivaras R, Warnke PC, Whittle IR, Jääskeläinen J, Ram Z. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma. Neuro-Oncology 2003; 5: 79–88. Westphal M, Ram Z, Riddle V, Hilt D, Bortey E. Gliadel wafer in initial surgery for malignant glioma: long-term follow-up of a multi-center controlled trial. Acta Neurochir (Wein) 2006; 148: 269–275.

Related Reference Valtonen S, Timonen U, Toivanen P, Kalimo H, Kivipelto L, Heiskanen O, Unsgaard G, Kuume T. Interstitial chemotherapy with carmustine with carmustine-loaded polymers for high-grade gliomas: a randomised double-blind study. Neurosurgery 1997; 41: 44–48.

Study Designs Both multi-centre, double-blind prospective RCTs. PBTTG trial (recurrent high-grade gliomas) Class of evidence I

GSG trial (primary high-grade gliomas) I

Randomization

Carmustine versus placebo

Number of patients

222

240

Follow-up

30 months (extended to 56 months on long-term follow- At least 6 months and lifetime of up) patients Primary endpoint: Survival

Primary endpoint: Overall survival at 12 months

Secondary endpoints: Complications Toxicity Quality of life

Secondary endpoints: Disease progression Quality of life Safety

Number of centres

38 (in 14 countries)

Stratification

Institution

Study centre Country

PBTTG Trial Inclusion criteria: unilateral tumour (size ≥1 cm on contrast-CT); KPS ≥ 60; completion of DXT; no nitrosoureas in previous 6 weeks; independent surgical decision on need for recurrent surgery. A proportional hazards regression model was used for statistical analysis in order to control for chance imbalances and differences in strong prognostic factors between groups. Intention-to-treat analysis.

GSG Trial Inclusion criteria: age 18–65; intra-operative frozen section confirming diagnosis of malignant glioma; KPS ≥ 60; radiological evidence of a single unilateral supratentorial tumour. The sample size of the trial was calculated to detect an 18% difference in 1-year survival between Gliadel® 69

and placebo (Îą level 0.05, power 0.90). A multiple regression analysis using the Cox proportional hazards model was used to account for effects of prognostic factors on survival. A secondary analysis was carried out for the glioblastoma subgroup although the trial was not designed to detect differences between histological subgroups.

Outcome Measures Kaplanâ&#x20AC;&#x201C;Meier curves were employed in both studies for the primary endpoints of survival.

Results

* After adjusting for prognostic factors.

PBTTG Trial No statistically significant effect of carmustine wafers was found in the primary analysis. A treatment effect was seen for carmustine wafers only once adjustment had been made for the effects of prognostic factors: >75% resection; KPS >70; young age. Subgroup analysis of patients with GBM showed an apparent 6-month survival advantage for patients with carmustine over placebo (carmustine group 56%, placebo 36%, p = 0.02). There was no significant difference in adverse events between the two groups.

GSG Trial Survival benefits remained significant at 3-year follow-up (carmustine group 9.2%, placebo group 1.7%, p = 0.01). Hazard ratio was 0.73 (p = 0.018) representing a 27% risk reduction. A subgroup analysis of patients with GBM did not reveal a statistically significant median survival benefit for carmustine (carmustine group 13.1 months, placebo group 11.4 months, p = 0.08). There was no significant difference in adverse events between the two groups.

Conclusions PBTTG trial: biodegradable polymers can assist delivery of carmustine directly to the brain. GSG trial: newly diagnosed malignant glioma patients benefit from carmustine wafers applied to the resection cavity at first operation.

Critique The rationale for local delivery of a chemotherapeutic agent to glioblastomas is to combat local recurrence, avoid systemic side effects, and to circumvent the bloodâ&#x20AC;&#x201C;brain barrier. Primary analysis in the PBTG trial by Brem et al. did not reveal a benefit for the carmustine wafers over placebo in patients with recurrent malignant gliomas. However, a statistically significant survival advantage was apparent in a subgroup analysis of patients with GBM and in the overall patient population once strong prognostic indicators were taken into account. However, conclusions from these post hoc analyses have been interpreted with caution as the study was not stratified for prognostic indicators or histological subgroups. Nevertheless, the work by Brem et al. stands out as a landmark by demonstrating the potential benefits of a new treatment strategy without significantly more adverse effects than conventional treatment regimens. The GSG trial by Westphal et al. showed a 2-month survival advantage for patients with newly diagnosed malignant gliomas receiving carmustine wafers to the resection cavity at primary resection. No statistically significant survival benefit was seen in a subgroup analysis for patients with GBM. However, as with similar analyses in the 70

trial by Brem et al. this was a post hoc analysis and the trial had not been stratified for histological subgroups. The GSG trial has defined the potential benefits of carmustine wafers in patients with primary malignant glioma (Perry et al., 2007). Further experience and future studies will help determine the patients most likely to benefit from this treatment and whether it is effective when combined with concomitant temozolomide and radiotherapy.

Reference Perry J, Chambers A, Spithoff K, Laperriere N. Gliadel wafers in the treatment of malignant glioma: a systematic review. Curr Oncol 2007; 14: 189–194.

2.8 Brachytherapy for Malignant Gliomas Details of Studies There have been a number of series reporting the efficacy of brachytherapy or interstitial radiotherapy in the treatment of newly diagnosed and recurrent malignant gliomas. To date, however, there have been only two randomized trials. The first study was by the University of Toronto group (Laperriere et al., 1998) while the second one was by the Brain Tumor Cooperative Group in the United States (Selker et al., 2002).

Study References Main Studies Laperriere NJ, Leung PM, McKenzie S, Milosevic M, Wong S, Glen J, Pintilie M, Bernstein M. Randomized study of brachytherapy in the initial management of patients with malignant astrocytoma. Int J Radiat Oncol Biol Phys 1998; 41: 1005–1011. Selker RG, Shapiro WR, Burger P, Blackwood MS, Deutsch M, Arena VA, Van Gilder JC, Wu J, Malkin MG, Mealey J, Neal JH, Olson J, Robertson JT, Barnett GH, Bloomfield S, Albright R, Hochberg FH, Hiesiger E, Green S. The Brain Tumor Cooperative Group NIH Trial 87-01: a randomized comparison of surgery, external radiotherapy, and carmustine versus surgery, interstitial radiotherapy boost, external radiation therapy, and carmustine. Neurosurgery 2002; 51: 343–357.

Study Designs All RCTs. Study

Laperriere et al. (1998)

Selker et al. (2002)

Class of evidence

Randomization EBRT* versus EBRT + brachytherapy with I125 implants

Brachytherapy with I125 implants + EBRT + BCNU versus EBRT + BCNU

Number of patients

140

270

Follow-up

Lifespan of patient

3 years

Primary endpoint: Primary endpoint: Overall survival from the date of initial Survival time from date of randomization surgery

Number of centres

Secondary endpoints: Quality of life KPS Steroid usage Tumour recurrence

Secondary endpoints: KPS Tumour recurrence

Stratification Age KPS

Institution Age Extent of surgery KPS

* EBRT = external beam radiation therapy.

Histopathological diagnosis was obtained through surgery, either biopsy or subtotal resection of the tumour. Both studies utilized stereotactically implanted high-activity I125 seeds to deliver brachytherapy. The total 71

radiation dose delivered was 60 Gy to the tumour perimeter. Laperriere’s group utilized conventional external beam radiation therapy (EBRT) on all enhancing tumour plus a 2.5 cm margin of surrounding brain. The prescribed dose was 50 Gy to the midplane in 25 fractions in 5 weeks. Selker’s group utilized WBRT 43 Gy in 25 fractions plus a coned-down boost to the tumour volume for the first 64 patients. Subsequent patients received 60.2 Gy in 35 fractions to the tumour border plus a 3 cm margin. The chemotherapeutic agent used by Selker’s group was 1, 3-bis (2-chloroethy 1)-1-nitrosourea (BCNU). It was administered intravenously at 200 mg/m2 and given every 8 weeks until the course was completed or if the patient developed adverse effects. Extent of resection and tumour recurrence were determined by contrast-enhanced CT scan.

Inclusion and Exclusion Criteria Laperriere et al. (1998)

Selker et al. (2002)

Inclusion criteria

• Biopsy-proven supratentorial malignant astrocytoma of the brain • Maximum tumour diameter ≤6 cm • No involvement of the corpus callosum • Age 18–70 • KPS ≥70

• Age >15 • Good KPS • Biopsy-proven supratentorial malignant glioma

Exclusion criteria

• Tumour diameter >6 cm • Corpus callosum involved • Poor KPS

• Midline cross-over in the corpus callosum or other midline structures • Multi-centric tumours • Absence of a contrast-enhancing ‘target’ after surgery • Age <15 • Other known primary malignancies • KPS <50 in the immediate 3-week postoperative period • Inability to effect a surgical resection, if indicated

Outcome Measures Primary Endpoint Survival.

Secondary Endpoints Laperriere et al. (1998) Quality of life determined by linear analogue self-report scales. KPS. Steroid usage. Tumour recurrence. Selker et al. (2002) KPS. Tumour recurrence.

Results Median Survival

Other Significant Findings Laperriere et al. (1998) 72

There is a statistically significant increase in median dexamethasone dosage for patients on the implant arm of the study, although the KPS does not differ between the two arms of the study. Recurrence patterns vary between the two treatment groups. The non-implant arm shows a higher recurrence rate at the original site of the tumour, whereas the implant arm exhibits a higher incidence of multifocal recurrence. Prognostic factors that are associated with improved survival include age, performance status, chemotherapy at recurrence, and reoperation at recurrence. Selker et al. (2002) Prognostic factors associated with improved survival include KPS, pathology (GBM versus non-GBM), sex, and age.

Conclusions Both trials concluded that stereotactic radiation implants do not confer a survival advantage in patients with newly diagnosed malignant gliomas.

Critique Brachytherapy is a form of radiotherapy delivered by implanting radioactive sources directly into the tumour. It delivers high doses of radiation to the tumour while sparing normal surrounding brain. Pathologic studies confirm that I125 brachytherapy decreases the proliferative capacity of the tumour (Siddiqi et al., 1997). However, this does not necessarily translate into clinical outcome such as improved survival. Numerous series have been reported on the efficacy of brachytherapy (Davis, 1987; Kumar et al., 1989; Gutin et al., 1991; Hitchon et al., 1992; Prados et al., 1992), but these were all non-randomized studies with a high degree of patient selection. Brachytherapy has also been used for recurrent malignant gliomas (Bernstein et al., 1994) and recurrent brain metastases (Bernstein et al., 1995), but because of the small study populations, the results were inconclusive. The main argument challenging the encouraging outcomes in the earlier brachytherapy studies is the significant selection bias with patients who were eligible for brachytherapy being younger and with a better performance status. In addition, their tumours were smaller, more peripherally located, and more circumscribed. The inherent characteristics of this subgroup may already be associated with a better prognosis despite the type of treatment given. The second argument is that patients who undergo brachytherapy may develop more significant radiation necrosis and thus, have a higher rate of reoperation to reduce the mass effect of the necrotic lesion and/or the tumour load. The additional survival may have been partly related to the beneficial effects of reoperation. Both arguments were put to rest by the RCTs discussed in this section. Both studies eliminated the effect of favourable patient and tumour characteristics by randomization, and both studies showed an equivalent number of reoperations in each group, which was not statistically significant. Laperriere et al. can be criticized for using a dosage of 50 Gy during EBRT. At the time of the study, the optimal dose of EBRT for malignant gliomas had not been established, but the present recommendation is 60 Gy (Bleehen et al., 1991). However, this did not affect the design of the study because both treatment arms received the same dose of radiation. Selker et al. implemented their study across 14 centres in the United States. This implies a wide variability in surgeon technical expertise and experience, and could affect the results of the study. The group also had a change in protocol with regard to EBRT: they utilized WBRT for the first 64 patients then modified their protocol afterwards to EBRT to the tumour plus a 3 cm margin. Although the group took into account this protocol change, with statistical analyses revealing that there is no difference in outcome between the patients who received the original versus the revised protocols, the actual effect in terms of amount of radiation necrosis and performance status may be unaccounted for. Both studies utilized contrast-enhanced CT scans to plan dosimetry, measure the extent of resection, and detect tumour recurrence. During the active years of the study, MRI was still widely unavailable. It is not known whether using a better imaging modality such as MRI would affect the results of the study, but certainly it would give a more accurate estimate of the extent of resection and detect recurrent disease earlier. Despite minor flaws, both studies are well designed and well executed. From the results, it is quite conclusive that brachytherapy has no role in the initial management of malignant glioma patients.

References Bernstein M, Laperriere N, Glen J, Leung P, Thomason C, Landon AE. Brachytherapy for recurrent malignant astrocytoma. Int J Radiat Oncol Biol Phys 1994; 30; 1213â&#x20AC;&#x201C;1217. Bernstein M, Cabantog A, Laperriere N, Leung P, Thomason C. Brachytherapy for recurrent single brain metastasis. Can J Neurol Sci 1995; 22: 13â&#x20AC;&#x201C;16.

Davis RL. Recurrent malignant gliomas: survival following interstitial brachytherapy with high-activity iodine-125 sources. J Neurosurg 1987; 67: 864–873. Gutin PH, Prados MD, Phillips TL, Wara WM, Larson DA, Leibel SA, Sneed PK, Levine VA, Weaver KA, Silver P, Lamborn K, Lamb S, Ham RN. External irradiation followed by an interstitial high activity iodine-125 implant ‘boost’ in the initial treatment of malignant gliomas: NCOG Study 6G-82-2. Int J Radiat Oncol Biol Phys 1991; 21; 601–606. Hitchon PW, VanGilger JC, Wen BC, Jani S. Brachytherapy for malignant recurrent and untreated gliomas. Stereo Funct Neurosurg 1992; 59: 174–178. Kumar PP, Good RR, Jones EO, Patil AA, Leibrock LG, McComb RD. Survival of patients with glioblastoma multiforme treated by intraoperative high-activity cobalt 60 endocurietherapy. Cancer 1989; 64: 1409–1413. Prados MD, Gutin PH, Phillips TL, Wara WM, Sneed PK, Larson DA, Lamb SA, Ham B, Malec MK, Wilson CB. Interstitial brachytherapy for newly diagnosed patients with malignant gliomas: The UCSF experience. Int J Radiat Oncol Biol Phys 1992; 24: 593–597. Siddiqi SN, Provias J, Laperriere N, Bernstein M. Effects of iodine-125 brachytherapy on the proliferative capacity and histopathological features of glioblastoma recurring after initial therapy. Neurosurgery 1997; 40: 910–917.

2.9 Extent of Resection of Low-Grade Gliomas Details of studies Two retrospective studies are included here which address the question of whether the extent of resection (EOR) of low-grade gliomas (LGGs) affects outcome. The first study was carried out in San Francisco, California, USA, with the analysis of patients who underwent resection of LGG at the University of California between 1989 and 2005 (Smith et al., 2008). The second was carried out in Baltimore, Maryland, USA with the analysis of patients who underwent resection LGG at the Johns Hopkins Department of Neurosurgery between 1996 and 2007 (McGirt et al., 2008).

Study References Main Studies McGirt MJ, Chaichana KL, Attenello FJ, Weingart JD, Than K, Burger P, Olivi AO, Brem H, Quinones-Hinojosa A. Extent of surgical resection is independently associated with survival in patients with hemispheric low-grade glioma. Neurosurgery 2008; 63: 700–708. Smith JS, Chang EF, Lamborn KR, Chang SM, Prados MD, Cha S, Tihan T, Vandenberg S. Role of extent of resection in the long-term outcome of low-grade hemispheric gliomas. J Clin Oncol 2008; 10: 1338–1345.

Related References Englot DJ, Berger MS, Barbaro NM, Chang EF. Predictors of seizure freedom after resection of supratentorial low-grade gliomas. A review. J Neurosurg 2011; 115: 240–244. Laws ER Jr, Taylor WF, Clifton MB, Okazaki H. Neurosurgical management of low-grade astrocytoma of the cerebral hemispheres. J Neurosurg 1984; 61: 665–673.

Study Designs Both retrospective cohort studies. Baltimore group (McGirt et al.)

San Francisco group (Smith et al.)

Class of evidence Number of patients

II 170

II 216

Randomization

None

Follow-up

Lifetime of patients

Primary endpoint:

Survival

Age KPS Tumour histological subtype Tumour size Primary versus revision surgery

Age KPS Tumour histological subtype Tumour size Tumour location

Stratification/adjustments

Both studies used MRI to assess the extent of resection. The Baltimore group classified extent of resection according to the extent of fluid-attenuated inversion recovery (FLAIR) signal abnormality on MRI carried out <48 h post-operatively and classified EOR as follow: gross 74

total resection (GTR); near total resection (NTR); and sub-total resection (STR). The San Francisco group calculated the percentage EOR using volumetric analysis of immediate post-operative MRI FLAIR signal on axial slices.

Outcome Measures Survival was assessed by both groups and both groups used overall survival (OS) and progression-free survival (PFS) as outcome measures. Both studies also analysed survival without malignant progression of the tumour, with progression being defined as either radiological evidence (gadolinium enhancement) or histological evidence of higher-grade tumour. Although slightly different terms were used by the two groups for malignant progression, for the sake of clarity this outcome will be defined here as malignant-free survival (MFS).

Results Baltimore Group (McGirt et al., 2008)

GTR versus STR was associated with OS and PFS as shown in the table. There was no significant difference in survival rates between STR and NTR.

San Francisco Group (Smith et al., 2008)

Conclusions Baltimore Group (McGirt et al., 2008) Greater EOR improves outcome for patients with LGG and should be safely attempted when not limited by eloquent cortex.

San Francisco Group (Smith et al., 2008) Greater EOR improves outcome for adult patients with LGG.

Critique LGGs pose a particularly difficult management dilemma for neurosurgeons. The prognosis of LGGs can be up to tenfold better than for high-grade gliomas. Although little is known about prognostic factors in these patients, the EOR has long been suspected to affect survival and early retrospective surgical series have been extremely influential in supporting this view (Laws et al., 1984). There are no clinical trials specifically examining this question and metaanalysis of the literature would simply be a pooling of retrospective studies with widely varying statistical methodologies applied. Also, many earlier case series include histological grades that have now been better differentiated with more known about the influence of histological subtype on survival. The question of whether extent of resection affects survival in LGG patients remains of utmost importance. LGGs are often in or near eloquent cortical regions and the risk of incurring neurological defects needs to be balanced against the need for total resection. With the more widespread application of neuronavigation and intra-operative MRI there is the possibility to perform more aggressive resections safely. 75

The two studies carried out here have their limitations, which are acknowledged by the authors. Both studies are retrospective and do not include any comparisons with patients undergoing biopsy. Furthermore, there is the possibility that patients who underwent less than total resection may reflect those in whom the tumour had infiltrated so as to prevent safe resection and, therefore, were already likely to have a worse prognosis. Furthermore, there are potential methodological difficulties in using contrast enhancement as a measure of resection and of tumour progression. Notwithstanding these criticisms, these studies are perhaps the best studies to date that address the role of the extent of resection of LGGs on patient outcome. A clinical trial addressing the role of EOR for LGG is unlikely to be forthcoming because of a lack of equipoise amongst neurosurgeons and potential difficulties recruiting enough patients due to the unwillingness of many patients to be randomized to the ‘no resection’ or ‘biopsy only’ arms (Smith et al., 2008). The two retrospective series included here are landmark in that they are, and will continue to be, influential in guiding neurosurgical decisionmaking regarding the EOR for LGG. It is likely that prospective studies evaluating the role of modalities such as intraoperative MRI will also be influential in guiding decision-making regarding the EOR. Indeed, there are already studies that suggest that using intra-operative MRI to help achieve total resection of LGG may improve survival when compared to known survival rates from national databases (Claus et al., 2005). Englot et al. published a systematic review looking specifically at the effect of the EOR of LGG on seizure control (Englot et al., 2011). They included 773 patients from 18 studies and addressed seizure freedom being the primary endpoint. They found that the odds ratio for seizure freedom after gross total resection was 3.41 (p < 0.001) and that this was the single strongest predictor of seizure outcome.

References Claus EB, Horlacher A, Hsu L, Schwartz RB, Dello-lacono D, Talos F, Jolesz FA, Black PM. Survival rate in patients with low-grade glioma after intraoperative magnetic resonance image guidance. Cancer 2005; 103: 1227–1233. Englot DJ, Berger MS, Barbaro NM, Chang EF. Predictors of seizure freedom after resection of supratentorial low-grade gliomas. A review. J Neurosurg. 2011; 115: 240–244. Laws ER Jr, Taylor WF, Clifton MB, Okazaki H. Neurosurgical management of low-grade astrocytoma of the cerebral hemispheres. J Neurosurg 1984; 61: 66–673. Smith JS, Chang EF, Lamborn KR, Chang SM, Prados MD, Cha S, Tihan T, Vandenberg S. Role of extent of resection in the long-term outcome of low-grade hemispheric gliomas. J Clin Oncol 2008; 10: 1338–1345.

2.10 Radiotherapy for Low-Grade Gliomas Details of Studies There have been several trials evaluating the role of radiotherapy in the management of LGGs. These trials have been carried out in Europe by the European Organisation for Research and Treatment of Cancer (EORTC) and in North America by the North Central Cancer Treatment Group (NCCTG) in conjunction with the RTOG and the Eastern Cooperative Oncology Group (ECOG). The question of a dose–response relation for LGGs treated with radiotherapy was addressed by the first European Trial (EORTC I, or the ‘Believers trial’) and the NCCTG-RTOG-ECOG trial (the ‘US trial’). The second European trial addressed the question of early versus delayed radiotherapy (at the time of disease progression) in LGG. These studies are, therefore, outlined here according to these two questions: low-dose versus high-dose radiotherapy; early versus delayed radiotherapy.

Study References Main Studies EORTC I: ‘Believers trial’ Karim AB, Maat B, Hatlevoll R, Menten J, Rutten EH, Thomas DG, Mascarenhas F, Horiot JC, Parvinen LM, van Reijn M, Jager JJ, Fabrini MG, van Alphen AM, Hamers HP, Gaspar L, Noordman E, Pierart M, van Glabbeke M. A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma. European Organization for Research and Treatment of Cancer (EORTC) Study 22844. Int J Rad Oncol Biol Phys 1996; 36: 549–556.

EORTC II: ‘Non-believers trial’ Karim AB, Afra D, Cornu P, Bleehen N, Schraub S, De Witte O, Darcel F, Stenning S, Pierart M, Van Glabbeke M. Randomised trial on the efficacy of radiotherapy for cerebral low-grade glioma in the adult: European Organization for Research and Treatment of Cancer Study 22844 with the Medical Research Council study BRO4: an interim analysis. Int J Rad Oncol Biol Phys 2002; 52: 316–324. Van den Bent MJ, Afra D, de Witte O, Ben Hassel M, Schraub S, Hoang-Xuan K, Malmström PO, Collette L, Piérart M, Mirimanoff R, Karim AB. EORTC Radiotherapy and Brain Tumor Groups and the UK Medical Research Council. Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial. Lancet 2005; 366: 985–990.

NCCTG-RTOG-ECOG: ‘US trial’ Shaw E, Arusell R, Scheithauer B, O’Fallon J, O’Neill B, Dinapoli R, Nelson D, Earle J, Jones C, Cascino T, Nichols D,

Ivnik R, Hellman R, Curran W, Abrams R. Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group Study. J Clin Oncol 2002; 20: 2267–2276.

Related Reference Brown PD, Buckner JC, O’Fallon JR, Iturria NL, Brown CA, O’Neill BP, Scheithauer BW, Dinapoli RP, Arussel RM, Abrams RA, Curran WJ, Shaw EG. Adult patients with supratentorial pilocytic astrocytomas: a prospective multicenter clinical trial. Int J Radiat Oncol Biol Phys 2004; 58: 1153–1160.

Study Designs All multi-centre PRCT.

EORTC I (‘Believers trial’) Inclusion criteria: adults (16–65 years); histological diagnosis of GI or GII supratentorial glioma (astrocytoma, oligodendroglioma, mixed oligodendroglioma). Exclusion criteria: poor neurological status; totally excised GI astrocytomas; pregnant patients; other ‘incurable’ malignancy. DXT doses: low (45 Gy); high (59.4 Gy).

NCCTG-RTOG-ECOG: ‘US trial’ Inclusion criteria: Adults (>18 years); histological diagnosis of supratentorial glioma (astrocytoma, oligodendroglioma, mixed oligoastrocytoma). DXT doses: Low (50.4 Gy); high (64.8 Gy).

EORTC II: ‘Non-believers trial’ Inclusion criteria: histological diagnosis of supratentorial LGG (GI/II astrocytoma; oligodendroglioma); KPS ≥60; adults (16–65 years). Exclusion criteria: completely resected GI pilocytic astrocytoma; significant other malignancy; pregnancy. Dose of DXT: 54 Gy. Timing of DXT: early (<8 weeks from surgery); delayed (at time of progression). Intention-to-treat analysis. 77

Outcome Measures EORTC I ‘Believers trial’

NCCTG I ‘US trial’

EORTC II ‘Non-believers trial’ Overall survival (OS) OS OS Progression-free survival (PFS) at Time to progression (TTP) TTP 5 years Cognitive function: Mini-Mental State Examination Cognitive function: MMSE (MMSE)

Results Low-Dose versus High-Dose DXT

In the ‘Believers trial’, the following factors were significantly associated with better overall survival: histology and grade (Grade I astrocytoma, oligodendroglioma, mixed oligodendroglioma); younger age. In the ‘US trial’, the following three factors were significantly associated with better overall survival: histology (oligodendroglioma or oligo-dominant mixed tumours); young patients (<40 years); and smaller tumours (<5 cm). In the ‘US trial’, the combination of histology and age together was the most powerful prognostic indicator of 5year survival: patients with oligodendroglioma and <40 years (82%) versus patients with astrocytoma ≥40 years (32%).

Early versus Delayed DXT There was no significant difference in the median survival for the early-DXT group (7.2 years) compared to the delayed-DXT group (7.4 years).

However, patients in the delayed-DXT group survived significantly longer after progression than those in the early-DXT group (3.4 years versus 1.0 years, p < 0.0001). Post hoc analysis revealed that significantly less patients in the early-DXT group suffered seizures in the first year (25% versus 41%).

Conclusions There is no improvement in survival with higher-dose radiotherapy (‘Believers trial’ and ‘US trial’). Early radiotherapy does not improve overall survival, but it does lengthen progression-free survival (‘Nonbelievers trial’).

Critique LGGs constitute approximately 10% of primary CNS malignancy, affect predominantly young adults, and are associated with a survival rates of <35% at 10 years. These trials have addressed two important questions regarding the role of radiotherapy to control this disease: 1 Is the effect of radiotherapy dose-dependent? 2 Is it better to administer radiotherapy early on or to delay administration until the disease progresses?

The EORTC I and NCCTG-RTOG-ECOG trials appear to support the use of lower-dose radiotherapy regimens. However, some feel that these trials cannot be interpreted clearly due to the histological heterogeneity of the gliomas included (Wessels et al., 2003). One of the major weaknesses of the EORTC II study (‘Non-believers’) is that the cohort includes both low- and high-grade gliomas as 26% of the tumours were reclassified as high-grade astrocytomas on histological review. In addition, over one-third of patients in the delayed-DXT group did not receive radiotherapy. It is felt by some that the fact that median survival was much greater following progression in the delayed-DXT group (2.4 years longer) could influence the decision when to give delayed-DXT (Knisely, 2006). The authors of the EORTC II study acknowledged that there was a lack of quality of life data, which would have allowed more considered discussion of their results. The issues of dose–response and timing of radiotherapy in the management of LGGs remain controversial despite the results of these trials. Indeed, Papagikos et al. have warned against an overly dogmatic approach to therapeutic options of LGGs in light of the conclusions from these trials (Papagikos et al., 2005). There is considerable interest regarding the role of radiosurgery and adjuvant chemotherapy. A trial of radiotherapy versus radiotherapy plus chemotherapy was completed in the early 1990s, which did not show any benefit of adjuvant chemotherapy (Eyre et al., 1993). However, new chemotherapeutic regimens are now being explored, which are tailored to specific histological types (Stege et al., 2005; Wen and DeAngelis, 2007). The management of LGGs remains still predominantly in the domain of the specialist neuro-oncological surgeon and neuro-oncologist.

References Eyre HJ, Crowley JJ, Townsend JJ, Eltringham JR, Morantz RA, Schulman SF, Quagliana JM, al-Samaf M. A randomised trial of radiotherapy versus radiotherapy plus CCNU for incompletely resected low-grade gliomas: a Southwest Oncology Group study. J Neurosurg 1993; 78: 909–914. Knisely J. Early or delayed radiotherapy for low-grade glioma? Lancet Oncol 2006; 6: 921. Papagikos MA, Shaw EG, Stieber VW. Lessons learned from randomised clinical trials in adult low-grade glioma. Lancet Oncol 2005; 6: 240–244. Stege EM, Kros JM, de Bruin HG, Enting RH, van Heuvel I, Looijenga LH, van der Rijt CD, Smitt PA, van den Bent MJ. Successful treatment of low-grade oligodendroglial tumors with a chemotherapy regimen of procarbazine, lomustine, and vincristine. Cancer 2005; 103: 802–809. Wen PY, DeAngelis LM. Chemotherapy for low-grade gliomas: emerging consensus on its benefits. Neurology 2007; 68: 1762– 1733. Wessels PH, Weber WEJ, Raven G, Ramaekers FCS, Hopman AHN, Twinjnstra A. Supratentorial grade II astroctyoma: biological features and clinical course. Lancet Neurol 2003; 2: 395–402.

2.11 Dysembryoplastic Neuroepithelial Tumour Details of Study The limitations of histopathological classification for intrinsic brain tumours are well recognized. Although classification systems such as that of the World Health Organization (WHO) provide a common language for clinical practice and help guide treatment regimens, there are tumours, particularly in the paediatric population, which do not seem to correlate with current research findings. Furthermore, predominantly morphological classification systems do not take into consideration our evolving understanding of tumour genetics and biology. This landmark paper described a subset of patients with tumours which, whilst superficially resembling oligodendromas, demonstrated a distinct clinical and radiological presentation and were associated with a much better prognosis (Daumas-Duport et al., 1988). The group was initially defined at the Hôpital Sainte Anne, Paris, France (n = 20) followed by a further 19 patients at the Mayo Clinic, Rochester, Minnesota, USA, after re-examination of previously collected LGG specimens. The tumours all occurred in children or young adults who presented with complex partial seizures.

Study Reference Main Study Daumas-Duport C, Scheithauer BW, Chodkiewicz JP, Laws ER Jr, Vedrenne C. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patients with intractable partial seizures. Report of thirty-nine cases. Neurosurgery 23; 5: 545–556.

Study Design Retrospective clinicopathological review of patient series. Class of evidence

III

Number of patients 39 Length of follow-up Mean 9 years (range 1–18 years) Number of centres 2, Sainte Anne Hospital, Paris, France (n = 20) and the Mayo Clinic, Rochester, USA (n = 19) 79

Stratification

N/A

In this two-centre study, all tumours were intracortical and supratentorial and most were located within the temporal lobe. Microscopically, the tumours exhibited multi-nodular architecture and were composed of astrocytes, oligodendrocytes, and neurons. Cell lines demonstrated a high degree of heterogeneity. Sixty per cent of cases were reported to exhibit a ‘specific glioneuronal element’ with columns of axons and oligodendroglia-like cells orientated perpendicular to the cortical surface and interposed by normal appearing neurons. A nodular component and adjacent areas of cortical dysplasia were present in the majority of tumours.

Outcome Measures Primary Endpoints Seizure freedom. Clinical or radiological progression.

Results Gross total resection was achieved in 56% of patients. Despite this, there were no clinical or radiological recurrences. Thirty out of 39 patients were rendered seizure free and all patients had a reduction in seizure frequency post-operatively.

Conclusions DNETs are a clinically and pathologically distinct form of primary brain neoplasm. They are benign and potentially surgically curable with no demonstrated risk of progression or recurrence.

Critique Although similar pathological findings had been described previously, these cases were identified in epilepsy specimens and were not associated with a radiological or operative mass lesion (Cavanagh, 1958). Daumas-Duport et al. identified a distinct pathological subgroup of epileptogenic tumours with an excellent prognosis with surgery alone. Since Daumas-Duport et al.’s initial series, >700 cases have been reported in the literature. Although the basic pathological description has not altered (Louis et al., 2007), several histological subtypes have been identified. However, these are not of clinical or prognostic significance (Louis et al., 2007). The lineage, pathogenesis, and genetics of DNETs are yet to be elucidated. Although a retrospective series, most of the clinical and prognostic findings of this landmark study have been borne out in subsequent literature. Extracortical and infratentorial cases have been now described, as have recurrent and more aggressive tumours. These, however, are rare. Grade progression has only been described twice, both times in patients with atypical presentations (Rushing et al., 2003). However, for most patients with DNETs this paper changed neurosurgeons’ understanding of the underlying pathology and spared them from unnecessary re-operations or adjuvant treatments. Furthermore, this landmark paper was one of the first studies to demonstrate the inadequacies of accepted histopathological classification systems. Subsequent papers have further subcategorized brain tumours, more recently based on genetic and molecular markers. Better classification allows for more accurate prediction of prognosis and greater likelihood of generating targeted therapies.

References Cavanagh J. On certain small tumours encountered in the temporal lobe. Brain 1958; 81: 389–405. Louis DN, Ohgaki H, Wiestler OD, Cavenne WK (eds). WHO Classification of Tumours of the Central Nervous System, 4th edition. Lyon: International Agency for Research on Cancer, 2007. Rushing EJ, Thompson LD, Mena H. Malignant transformation of a dysembryoplastic neuroepithelial tumor after radiation and chemotherapy. Ann Diagn Pathol 2003; 7: 240–244.

2.12 Meningioma Resection Grading Details of Study In the 1950s, improved surgical technique and longevity meant that meningioma recurrence became increasingly problematic. Previous attempts to grade the risk of recurrence had been based on tumour histology, location, or ‘completeness of resection’ as a binary, ill-defined variable. In 1957, Donald Simpson, an Australian surgeon studying in Oxford, United Kingdom, published a single-author case series of meningiomas treated between 1928 and 1954 in London or the Radcliffe Infirmary, Oxford (Simpson, 1957).

Study References 80

Main Study Simpson D. The recurrence of intracranial meningiomas after surgical treatment. J Neurol Neurosurg Psychiatry 1957; 20: 22– 39.

Related Reference Heros RC. Simpson grades. J Neurosurg 2012; 117: 997–998.

Study Design Retrospective review of patient records Class of evidence

III

Randomization

N/A

Number of patients

339

Length of follow-up

6 months to >20 years.

Number of centres

2, Radcliffe Infirmary, Oxford (n = 235), London (n = 97)

Stratification

Outcome Measures Primary Endpoint Clinical recurrence: symptoms directly referable to the tumour or death, either confirmed to be from meningioma at autopsy or thought likely to be so.

Simpson’s Meningioma Resection Grading Scale Grade

Intradural tumour

Dural attachment

Macroscopically complete resection

Excised, abnormal bone removed

Macroscopically complete resection

In situ, diathermied

III

Macroscopically complete resection

In situ

Partial removal

Decompression, with or without biopsy

Results Overall, 55/265 (21%) of tumours recurred.

Conclusions The extent of tumour and dural resection can be used to stratify the risk of recurrence.

Critique Simpson’s paper provided a clearly defined grading scale for extent of meningioma resection. He designed the grading system relying on operative notes. The duration of follow-up was excellent and relatively few patients were lost. Although Simpson did not perform any statistical analyses, the Chi-squared test can be applied to the results table giving a p value of <0.001. Aside from demonstrating an association between extent of resection and the probability of recurrence, the scale provided a standardized language for discussion of the degree of resection and redefined what constitutes a ‘total resection’. Simpson’s study was performed in the pre CT/MRI era, thus recurrence is refined clinically rather than radiologically—a more relevant outcome measure. Most subsequent studies have replicated Simpson’s findings, although with lower overall recurrence rates. However, in 2010 Sughrue et al. from UCSF in California, USA published their series of 879 patients (Sughrue et al., 2010). Although they found a trend towards lower recurrence rates with Grade I resections, their results were not 81

statically significant. They have, therefore, questioned the need for such aggressive dural and bone resection. However, it must be noted that their sample included 189 (22%) base of skull tumours, a much larger proportion than most other series. Additionally, these included radiological progression as their definition of recurrence, rather than symptomatic recurrence as in Simpson’s paper. A subsequent paper by Hasseleid et al. from Oslo, Sweden, recapitulated Simpson’s findings in 391 patients with convexity meningiomas (Hasseleid et al., 2012). Importantly, they defined recurrence as the need for retreatment and had almost double the duration of follow-up as Sughrue et al.’s sample. These findings suggest a Simpson Grade I resection should remain the goal of meningioma surgery.

References Hasseleid BF, Meling TR, Rønning P, Scheie D, Helseth E. Surgery for convexity meningioma: Simpson Grade I resection as the goal. J Neuosurg 2012; 117: 999–1006. Heros RC. Editorial: Simpson grades. J Neurosurg 2012; 117: 997–998. Sughrue ME, Rutkowski MJ, Chen CJ, Shangari G, Kane AJ, Parsa AT, Berger MS, McDermott MW. The relevance of Simpson Grade I and II resection in modern neurosurgical treatment of World Health Organization Grade I meningiomas. J Neurosurg 2010; 113: 1029–1035.

Chapter 3

Head injury RD Johnson, F Zhou, T Santarius, JE Wilberger, JV Rosenfeld 3.0 Introduction 3.1 Glasgow coma scale and glasgow outcome scale 3.2 Timing of surgery for acute traumatic extra-axial haematomas 3.3 Surgery for chronic subdural haematomas 3.4 Decompressive craniectomy for severe traumatic brain injury 3.5 Intracranial pressure monitoring in head injury 3.6 Steroids in head injury 3.7 Barbiturates in head injury 3.8 Hyperosmolar therapy for control of raised intracranial pressure in head injury 3.9 Hypothermia in head injury 3.10 Hyperventilation in head injury 3.11 Magnesium for neuroprotection in head injury 3.12 Epidemiology of post-traumatic seizures 3.13 Phenytoin for prevention of post-traumatic seizures 3.14 Pre-hospital intubation for traumatic brain injury 189

3.0 Introduction Head injuries are common, and moderate to severe traumatic brain injury (TBI) affects young and old alike with potentially high morbidity and mortality. Surgical treatment of head injury may represent the oldest neurosurgical procedures. Indeed, surgical options for head injuries are mentioned in the Edwin Smith papyrus, which refers to a period before 2500 bc (Breasted, 1930). In the modern era, neurosurgeons and neuro-intensivists are the main carers for this group of patients. Several advances have been influential in shaping current methods of management of headinjured patients. The pathological nature of head injury has been further elucidated (Graham and Adams, 1971). As it appears that there is nothing that can be done to reverse the primary brain injury, the focus of our efforts is primarily targeted at the cascade of events leading to secondary brain injury. Efforts are directed at controlling intracranial pressure (ICP) and maintaining adequate brain oxygenation and perfusion. This has required an increased cooperation between neurosurgeons and neuro-intensivists. Another advance that has had a seismic impact in the early diagnosis of traumatic pathology in head-injured patients is the advent of cross-sectional computed tomography (CT) scanning (Hounsfield, 1973; French and Dublin, 1977): an advance that was recognized by the award of the 1979 Nobel Prize in medicine to Cormack and Hounsfield. Mortality from civilian head injuries has been significantly reduced over the last 35 years (Seelig et al., 1981). Nonetheless, many patients are left with extreme disabilities if they survive. The organization of head injury pharmacological trials to improve outcome has been extremely difficult for logistical reasons and because of technical difficulties with their design (Narayan et al., 2002). Dickinson et al. carried out a broad-range analysis of head injury trials, and in their analysis they reported several significant and dramatic findings (Dickinson et al., 2000). First, they found that in the period up until 1998 there were a total of 208 discrete published randomized trials of interventions in head injury, but only 4% of these trials were of sufficient power to detect a 10% difference in outcome among the cohorts (power level set at 80%, Î˛ = 0.2). Second, they found that there were problems with blinding, not only in terms of patients and surgeons/doctors, but also of those evaluating outcome, with just over 20% of published trials with any blinding of outcome assessment. Dickinson et al. highlighted the small size of head injury trials, and calculated that the total number of patients enrolled in all head injury trials together up until 1998 was less than some of the larger individual trials in stroke and heart disease. A further consideration is that head injury is a heterogeneous group of entities and current classification systems, including those based on the Glasgow Coma Scale (GCS), do not necessarily reflect this. It may be difficult, therefore, to derive useful information from head injury studies unless a large number of subjects have been included, and negative results from small studies should be treated with caution (Saatman et al., 2008). Several topics have been chosen for inclusion in this chapter that we feel are particularly relevant to the practising neurosurgeon and have addressed areas where there was significant equipoise. The opening section considers the landmark papers describing the GCS and Glasgow Outcome Scale (GOS) that have had such a profound impact on the development and practice of neurosurgery (Teasdale and Jennett, 1974; Jennett and Bond, 1975). The next section 83

deals with three seminal case studies regarding the timing of operative management of acute extradural and subdural haematomas (Mendelow et al., 1979; Seelig et al., 1981; Wilberger et al., 1991). This is followed by a section that contains two studies addressing surgical technique for the management of chronic subdural haematomas (Svien and Gelety, 1964; Santarius et al., 2009). We are pleased to be able to include a section on decompressive craniectomy by including the results of the Early Decompressive Craniectomy in Patients with Severe Traumatic Brain Injury (DECRA) study (Cooper et al., 2011). Studies addressing the role of pharmacological interventions including steroids, barbiturates, hyperosmolar therapies, and magnesium have been included. In the next section we consider two multi-centre randomized trials addressing the efficacy of hypothermia for severe traumatic head injury and consider a post hoc subgroup meta-analysis (Clifton et al., 2001; Clifton et al., 2011; Clifton et al., 2012). The next section looks at a seminal study that addressed the role of hyperventilation on outcome in head injury (Muizelaar et al., 1991). The next section reviews the role of magnesium in severe traumatic brain injury (Temkin et al., 2007). The following two sections of this chapter address landmark studies in post-traumatic seizures (PTS) following head injury. The first of these sections considers a population-based study of PTS by Annegers et al. (1998), which is the largest of its kind taking in 4541 patients over a period of 50 years and has produced the most accurate data regarding the epidemiology of PTS. The study emphasizes the importance of this problem and provided valuable data regarding the risk of PTS. The other study included here is a Danisk population based study of a different design including over 1.6 million people (Christensen et al., 2009). The penultimate section addresses a randomized, double-blind trial of phenytoin for the prophylaxis of PTS (Temkin et al., 1990). Although many studies had already been published on seizure prophylaxis for post-traumatic epilepsy, Temkin’s study stands out as being the largest study with sufficient power to be able to analyse the efficacy of phenytoin. The results of Temkin’s study have been confirmed by several meta-analyses of all published trials. The final section summarizes the first prospective randomized trial on the role of pre-hospital intubation in severe TBI (Bernard et al., 2012). This chapter refers almost exclusively to the adult population. Two further studies looking at the issues of decompressive craniectomy and hypothermia in paediatric head injury have been included in the paediatric neurosurgery chapter. Furthermore, the head injury guidelines drawn up by the United States Brain Trauma Foundation are based on their Task Force review of the available evidence (Bullock et al., 2006; Brain Trauma Foundation, 2007). These guidelines are a landmark in neurosurgery and are recommended reading for all those with an interest in head injury. Similarly, the 2012 consensus statement on concussion in sport is a landmark in neurosurgery which we would recommend reviewing (McCrory et al., 2013). Since the first edition of this volume there have been developments in the field of TBI research which are likely to lead to more landmark studies. There is an increasing interest in the role of biomarkers and MRI for prognostication in (Chew et al., 2012; Mondello et al., 2013). Data from battlefield and sports concussion studies are also revealing new insights into the pathophysiology of TBI (Petraglia et al., 2012; DeKosky et al., 2013; Duckworth et al., 2013; Kontos et al., 2013). New avenues of research and areas in which advances are likely to be seen are eloquently discussed in a recent review (Rosenfeld et al., 2012).

References Annegers JF, Hauser WA, Coan SP, Rocca WA. A population-based study of seizures after traumatic brain injuries. N Engl J Med 1998; 338: 20–24. Bernard S, Nguyen V, Cameron P, Masci K, Fitzgerald M, Cooper DJ, Walker T, Myles P, Murray L, Taylor D, Smith K, Patrick I, Edington J, Bacon A, Rosenfeld JV, Judson R. Prehospital rapid sequence induction improves functional outcome for patients with severe traumatic brain injury. A randomized controlled trial. Ann Surg 2012; 252: 959–965. Brain Trauma Foundation. Guidelines for the management of severe traumatic brain injury. J Neurotrauma 2007; 24: S1– S106. Breasted JH. The Edwin Smith Surgical Papyrus. Chicago, IL: University of Chicago Press, 1930. Bullock MR, Chestnut R, Ghajar J, Gordon D, Hartl R, Newell DW, Servadei F, Walters BC, Wilberger JE. Surgical management of TBI author group. Neurosurgery 2006; 58: S2-1–S2-3. Chew BG, Spearman CM, Quigley MR, Wilber JE. The prognostic significance of traumatic brainstem injury detected on T2weighted MRI. J Neurosurg 2012; 117: 722–728. Clifton GL, Miller ER, Choi SC, Levin HS, McCauley S, Smith K, Muizelaar JP, Wagner FC, Marion DW, Luerssen TG, Chestnut RM, Schwartz M. Lack of effect of induction of hypothermia after acute brain injury. N Engl J Med 2001; 344: 556–563. Clifton GL, Valadka, Zygun D, Coffey CS, Drever P, Fourwinds S, Janis LS, Wilde E, Taylor P, Harshman K, Conley A, Puccio A, Levin HS, McCauley SR, Bucholz RD, Smith KR, Schmidy JH, Scott JN, Yonas H, Okonkwo DO. Very early hypothermia induction in patients with severe brain injury (the National Acute Brain Injury Study: Hypothermia II): a randomised trial. Lancet 2011; 10: 131–139. Clifton GL, Coffey CS, Fourwinds S, Zygun D, Valadka A, Smith KR Jr, Frisby ML, Bucholz RD, Wilde EA, Levin HS, Okonkwo DO. Early induction of hypothermia for evacuated intracranial hematomas: a post hoc analysis of two clinical trials. J Neurosurg 2012; 117: 714–720. Cooper DJ, Rosenfeld JV, Murray L, Arabi YM, Davies AR, D’Urso P, Kossmann T, Ponsford J, Seppelt I, Reilly P, Wolfe R. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 364:1493–1502. DeKosky ST, Blennow K, Ikonomovic MD, Gandy S. Acute and chronic traumatic encephalopathies: pathogenesis and

biomarkers. Nat Rev Neurol 2013; 9: 192–200 Dickinson K, Bunn F, Wentz R, Edwards P, Roberts I. Size and quality of randomised controlled trials in head injury: review of published studies. BMJ 2000; 320: 1308–1311. Duckworth JL, Grimes J, Ling GS. Pathophysiology of battlefield associated traumatic brain injury. Pathophysiology 2013; 20(1): 23–30. French BN, Dublin AB. The value of computerized tomography in the management of 1000 consecutive head injuries. Surg Neurol 1977; 7: 171–183. Hounsfield GN. Computerized transverse axial scanning (tomography). 1. Description of system. Br J Radiol 1973; 46: 1016– 1022. Hutchinson PJ, Corteen E, Czosnyka M, Mendelow AD, Menon DK, Mitchell P, Murray G, Pickard JD, Rickels E, Sahuquillo J, Servadei F, Teasdale GM, Timofeev I, Unterberg A, Kirkpatrick PJ. Decompressive craniectomy in traumatic brain injury: the randomized multicenter RESCUEicp study (www.RESCUEicp.com). Acta Neurochir Suppl 2006; 96: 17–20. Jennett B, Bond M. Assessment of outcome after severe brain damage: a practical scale. Lancet 1975; 1: 480–484. Kontos AP, Kotwai RS, Elbin RJ, Lutz RH, Forsten RD, Benson PJ, Guskiewicz KM. Residual effects of combat-related mild traumatic brain injury. J Neuortrauma 2013; 30: 680–686. McCrory P, Meeuwisse WH, Aubry M, Cantu B, Dvorak J, Echemendia RJ, Engebretsen L, Johnston K, Kutcher JS, Raftery M, Sills A, Benson BW, Davis GA, Ellenbogen RG, Guskiewicz K, Herring SA, Iversen GL, Jordan BD, Kissick J, McCrew M, McINtosh AS, Maddocks D, Makdissi M, Purcell L, Putukian M, Schneider K, Tator CH, Turner M. Consensus statement on concussion in sport: the 4th international conference on concussion in sport held in Zurich, November 2012. Br J Sports Med 2013; 47: 250–258. Mendelow AD, Karmi MZ, Paul KS, Fuller GAG, Gillingham FJ. Extradural haematomas: effect of delayed treatment. BMJ 1979; 1: 1240–1249. Mondello S, Buki A, Italiano D, Jeromin A. α-synuclein in CSF of patients with severe traumatic brain injury. Neurology 2013; 80(18): 1662–1668. Muizelaar JP, Marmarou A, Ward JD, Kontos HA, Choi SC, Becker DP, Gruemer H, Young HF. Adverse effects of prolonged hyperventilation in patients with severe traumatic brain injury. J Neurosurg 1991; 75: 731–739. Narayan RK, Michel ME, Ansell B, Baethmann A, Biegon A, Bracken MB, Bullock MR, Choi SC, Clifton GL, Contant CF, Coplin WM, Dietrich WD, Ghajar J, Grady SM, Grossman RG, Hall ED, Heetderks W, Hovda DA, Jallo J, Katz RL, Knoller N, Kochanek PM, Maas AI, Majde J, Marion DW, Marmarou A, Marshall LF, McIntosh TK, Miller E, Mohberg N, Muizelaar JP, Pitts LH, Quinn P, Riesenfeld G, Robertson CS, Strauss KI, Teasdale G, Temkin N, Tuma R, Wade C, Walker MD, Weinrich M, Whyte J, Wilberger JE, Young AB, Yurkewicz L. Clinical trials in head injury. J Neurotrauma 2002; 19: 503–557. Petraglia AL, Maroon JC, Bailes JE. From the field of play to the field of combat: a review of the pharmacological management of concussion. Neurosurgery 2012; 70: 1520–1533. Rosenfeld JV, Maas AI, Bragge P, Morgant-Kossman MC, Manley GT, Gruen RL. Early management of severe traumatic brain injury. Lancet 2012; 380: 1088–1098. Saatman KE, Duhaime AC, Bullock R, Maas AI, Valadka A, Manley GT; Workshop Scientific Team and Advisory Panel Members. Classification of traumatic brain injury for targeted therapies. J Neurotrauma 2008; 25: 719–738. Santarius T, Kirkpatrick PJ, Ganesan D, Chia HL, Jallah I, Smielewski P, Richards HK, Marcus H, Parker RA, Price SH, Kirollos RW, Pickard JD, Hutchinson PJ. Use of drains versus no drains after burr-hole evacuation of chronic subdural haematoma: a randomised controlled trial. Lancet 2009; 374: 1067–1073. Seelig JM, Becker DP, Miller JD, Greenberg RP, Ward JD, Choi SC. Traumatic acute subdural hematoma: major mortality reduction in comatose patients treated within four hours. JAMA 1981; 304: 1511–1518. Svien HJ, Gelety JE. On the surgical management of encapsulated subdural hematoma. J Neurosurgery 1964; 21: 172–177. Teasdale G, Jennett B. Assessment of coma and impaired consciousness: a practical scale. Lancet 1974; 2: 81–84. Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn HR. A randomised, double-blind study of phenytoin for the prevention of post-traumatic seizures. N Engl J Med 1990; 323: 497–502. Wilberger JE, Harris M, Diamond DL. Acute sudbural haematoma: morbidity, mortality and operative timing. J Neurosurg 1991; 74: 212–218.

3.1 Glasgow Coma Scale and Glasgow Outcome Scale Details of Studies Teasdale and Jennett first described the Glasgow Coma Scale (GCS) in 1974 (Teasdale and Jennett, 1974). The GCS has become the most widely used scoring system for describing the neurological status in head-injured patients worldwide. The Glasgow Outcome Scale (GOS) was first reported by Jennett and Bond in 1975 (Jennett and Bond, 1975). The GCS and GOS have become a widely accepted method of reporting outcomes not only in studies of head injuries, but also in other neurosurgical conditions.

Study References Main Studies Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975; 1: 480–484. Jennett B, Snoek J, Bond MR, Brooks N. Disability after severe head injury: observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry 1981; 44: 285–293.

Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974; 2: 81–84. Teasdale G, Jennett B. Assessment and prognosis of coma after head injury. Acta Neurochir 1976; 34: 45–55.

Related References Jennett B, Teasdale G. Aspects of coma after severe head injury. Lancet 1977; 878–881. Jennett B, Teasdale G, Galbraith S, Pickard J, Grant H, Braakman R, Avezaat C, Mass A, Minderhoud J, Vecht C, Heiden J, Small R, Caten W, Kurze T. Severe head injuries in three countries. J Neurol Neurosurg Psychiatry 1977; 40: 291–298.

Glasgow Coma Scale The original GCS was described as follows: A clinical scale has been evolved for assessing the depth and duration of impaired consciousness and coma. Three aspects of behaviour are independently measured—motor responsiveness, verbal performance, and eye opening. (Jennett and Teasdale, 1974, p 81)

The first description in 1974 consisted of a 14-point scale as follows: Eye opening Spontaneous To speech To pain None

Best verbal response Orientated Confused Inappropriate Incomprehensible None

Best motor response Obeying Localizing Flexing Extending None

However, numerical values were not added until it was modified to a 15-point scale in 1976:

The assessment of motor response was described in detail with ‘obeying commands’ being defined as the best response possible. Jennett and Teasdale emphasized that this response should be interpreted carefully: The observer must take care not to interpret a grasp reflex or a postural adjustment as a response to command. The terms ‘purposeful’ and ‘voluntary’ are avoided as we believe they cannot be judged objectively. (Jennett and Teasdale, 1974, p 82)

The authors also described in detail the method by which noxious stimuli were to be applied, starting with pressure on the nail bed with a pencil to test for flexion and then to the head, neck, or trunk to test for localization. Motor responses are described in detail: localizing is defined as a limb moving to remove a noxious stimulus at more than one site (e.g. the opposite limb moving to site of nail bed pressure); withdrawing is normal flexion of the elbow or knee to local painful stimulus; abnormal flexion is slow withdrawal with pronation of the wrist and adduction of the shoulder; extensor response is defined as adduction and internal rotation of the shoulder with pronation of the forearm. They also indicate the importance of recording the best motor response if there is a discrepancy between limbs as the lesser response may reflect localized brain damage rather than overall conscious level. They indicate that eye opening in response to pain should be assessed from stimulus to the limbs to avoid the grimacing reflex causing eye closure. Verbal responses are described as follows: orientated (in time, place, person); confused conversation (attends and responds to questions but with incorrect or muddled answers); inappropriate words (intelligible words but random and unrelated to questions asked); incomprehensible speech (moans and groans only).

Glasgow Outcome Scale Death Vegetative state

No evidence of meaningful response 86

Severe disability

Conscious but needs the assistance of another person for either physical or mental reasons

Moderate disability Independent but disabled Good recovery

Resumption of normal occupation and social activities

Patients in a vegetative state were said to be able to breathe spontaneously and may even eye open, swallow, or show reflex responses in the limbs. However, a vegetative state was said to indicate a lack of function of the cerebral cortex. The three categories of disability described a range of physical and mental dependencies. However, as it was felt that these categories were so compressed as to be insensitive to degrees of improvement within each category the scale was extended in 1981 to further subdivide each disability category into two subcategories (upper and lower) so that, including death as an outcome, an 8-point scale was produced.

Critique The GCS is the most widely utilized scale for the measurement of level of consciousness worldwide. The GCS has been incorporated into other scores of ill health and trauma including the Acute Physiology and Chronic Health Evaluation (APACHE) II score and the Trauma and Injury Severity Score (TRISS). Indeed, the only country that has ever rejected the GCS in favour of another scoring system is Sweden (Starmark et al., 1988; Marion and Carlier, 1994). Despite its widespread acceptance and use, the GCS has not been without its critics. Although the GCS was developed primarily as a research tool in order to predict outcome and to allow for bedside assessments of fluctuations of conscious level, it has been widely utilized as an indicator for clinical decision-making. This is demonstrated by the adage ‘If the GCS is less than 8, then you need to intubate’. The GCS score has become utilized to alert practitioners to the need to protect the airway with formal intubation (Chestnut, 1997). Likewise, a GCS of 8 is a threshold for consideration of ICP monitoring. However, using the GCS in these ways implies that it is summarized as a single numerical score. Jennett and Teasdale assigned a numerical score to the GCS and in their assessment of 700 head-injured patients concluded that all combinations that resulted in a GCS of <8 were definitive of coma (Jennett and Teasdale, 1977). However, they have always emphasized that the conscious level of a patient should always be described fully in terms of the three separate responses (Teasdale et al., 1983). Indeed, the summation of the score to a single figure has been criticized for weighting the motor score over the other components (Bhatty and Kapoor, 1993). However, this weighting towards the motor score may be desirable as it is possible that the motor response is the best predictor of neurological outcome (Jagger et al., 1983). The other main criticism of the GCS is the fact that it does not incorporate any measure of brainstem reflexes (Segatore and Way, 1992). However, the original aim of the GCS was to produce a scale that provided interobserver reliability so that frequent and repeated assessments could be made by the bedside. Scoring systems that have incorporated brainstem reflexes have been more complicated and, therefore, less inter-rater reliability. Jennett and Teasdale made it clear in their original description that the GCS was not developed to replace a full neurological examination: It is no part of our case to deny the value of a detailed appraisal of the patient as a whole, and of neurological function in particular (Jennett and Teasdale, 1974, p 83)

Another area of criticism of the GCS is its application to intubated patients: a problem that is generally circumvented by recording the verbal score with a ‘T’ (Meredith et al., 1988). Furthermore, there have been concerns regarding the use of the GCS in the pre-hospital setting as it appears that such scores may not correlate with either severity of head injury or serve as prognostic indicators of outcome (Marion and Carlier, 1994). Notwithstanding these weaknesses and criticisms, the description of the GGS was a true landmark in neurosurgery as it brought consistency to scoring conscious levels in head-injured patients and all the evidence indicates that it is likely to remain in widespread use for a long time to come. The aim of the GOS was to describe overall social outcome in patients in terms of a limited set of outcomes, which were ‘sufficiently clearly defined to be used reliably by observers in several centres, some of whom were in different countries’ (Jennett et al., 1981). Almost immediately after the GOS was described, it was proposed by others that it should be adopted worldwide for the follow-up of series of head-injured patients (Langfitt, 1978). The original GOS was, however, relatively insensitive to subtle changes and for weighting physical disability relative to cognitive dysfunction (Kaye and Andrews, 2000). It appears that most patients will reach their final point on the 5-point scale by 6 months: an interval at which it is also usually possible to successfully follow-up patients (Jennett, 2005). Indeed, it appears that early GOS at 3 months correlates well with long-term GOS (King et al., 2005). Modification of the GOS to the Glasgow Outcome Scale—Extended (GOSE) has addressed this and the GOSE has been found to be more sensitive for detecting changes in mild to moderate head-injured patients, particularly when assessed on the basis of a structured interview, and thus more useful for longer-term follow-up (Wilson et al., 1998; Levin et al., 2001). In many research papers, outcomes are dichotomized into ‘poor’ (severe disability or vegetative state) or ‘good’ (moderate disability or good recovery). The GCS and GOS have been widely adopted for the management and study of traumatic and non-traumatic 87

patients. Jennett has emphasized that training in their proper use is necessary to ensure that they are not misleading (Jennett, 2005).

References Bhatty GB, Kapoor N. The Glasgow Coma Scale: a mathematical critique. Acta Neurochir 1993; 120: 132–135. Chestnut RM. The management of severe traumatic brain injury. Emerg Med Clin North Am 1997; 15: 581–604. Jagger J, Jane JA, Rimet R. The Glasgow Coma Scale: to sum or not to sum? [letter] Lancet 1983; 2: 97. Jennett B. Development of the Glasgow Coma and Outcome Scales. Nepal J Neurosci 2005; 2: 24–28. Kaye AH, Andrews D. Glasgow Outcome Scale: research scale or blunt instrument? Lancet 2000; 356: 1540–1541. King JT, Jr., Carlier PM, Marion DW. Early Glasgow Outcome Scale scores predict long-term functional outcome in patients with severe traumatic brain injury. J Neurotrauma 2005; 22: 947–954. Langfitt TW. Measuring the outcome from head injuries. J Neurosurg 1978; 48: 673–678. Levin HS, Boake C, Song J, McCauley S, Contant C, Diaz-Marchan P, Brundage S, Goodman H, Kotrla KJ. Validity and sensitivity to change of the extended Glasgow Outcome Scale in mild to moderate traumatic brain injury. J Neurotrauma 2001; 18: 575–584. Marion DW, Carlier PM. Problems with initial Glasgow Coma Scale assessment caused by prehospital treatment of patients with head injuries: results of a national survey. J Trauma 1994; 36: 89–95. Meredith W, Rutledge R, Fakhry SM, Emery S, Kromhout-Schiro S. The conundrum of the Glasgow Coma Scale in intubated patients: a linear regression prediction of the Glasgow verbal score from the Glasgow eye and motor scores. J Trauma 1988; 44: 839–845. Segatore M, Way C. The Glasgow Coma Scale: time for a change. Heart Lung 1992; 21: 548–557. Starmark JE, Stålhammar D, Holmgren E, Rosander B. A comparison of the Glasgow Coma Scale and the Reaction Level Scale (RLS85). J Neurosurg 1988; 69: 699–706. Teasdale G, Jennett B, Murray L, Murray G. Glasgow Coma Scale: to sum or not to sum? [letter] Lancet 1983; 2: 678. Wilson JT, Pettigrew LE, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and the Extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma 1998; 15: 573–585.

3.2 Timing of Surgery for Acute Traumatic Extra-axial Haematomas Details of Studies Three seminal case series concerned with the timing of surgery to evacuate extra-axial haematomas are included in this section. The first is a series of 83 patients treated for traumatic extradural haematomas (EDHs) in the Edinburgh area in Scotland in the periods 1951–1960 and 1968–1977 (Mendelow et al., 1979). This series compared the effects on outcome of delayed treatment following neurological deterioration and emphasized the need for immediate operation in patients deteriorating from extradural haematomas. The second study from Richmond, Virginia, USA, is a series of 82 patients with traumatic acute subdural haematomas (ASDH) admitted to the Division of Neurological Surgery, Medical College of Virginia, between 1972 and 1980 (Seelig et al., 1981). This paper established the principle of time to definitive surgical treatment as the most significant factor in the management of traumatic ASDH. The third study is a series of 101 patients with traumatic ASDH admitted to Allegheny General Hospital, Pittsburgh, Pennsylvania, USA, between 1982 and 1987. This paper, while confirming the general trend regarding the effect of timing of surgery on outcome, highlighted the importance of control of ICP and the effect of primary brain injury.

Study References Main Studies Mendelow AD, Karmi MZ, Paul KS, Fuller GAG, Gillingham FJ. Extradural haematomas: effect of delayed treatment. BMJ 1979; 1: 1240–1249. Seelig JM, Becker DP, Miller JD, Greenberg RP, Ward JD, Choi SC. Traumatic acute subdural hematoma: major mortality reduction in comatose patients treated within four hours. JAMA 1981; 304: 1511–1518. Wilberger JE, Harris M, Diamond DL. Acute subdural haematoma: morbidity, mortality and operative timing. J Neurosurg 1991; 74: 212–218.

Related Reference Bullock MR, Chestnut R, Ghajar J, Gordon D, Hartl R, Newell DW, Servadi F, Walters BC, Wilberger JE, for the Surgical Management of Traumatic Brain Injury Author Group. Surgical management of acute subdural haematomas. Neurosurgery 2006; 58: S16–S24.

Study Designs All retrospective case series.

In the study by Mendelow et al. delay was defined as the time between neurological deterioration and surgery. In the studies by Seelig et al. and Wilberger et al. delay was defined as time from injury to surgery.

Results Mendelow et al. (1979) In the study by Mendelow et al. the two time periods compared reflected a change in the management of head-injured patients: in the second time period, patients were routinely admitted directly to the neurosurgical unit within 24 h. This resulted in a fourfold reduction in delay to surgery (p = 0.06) (see Figure 3.1). Delay to surgery in survivors 1.9 h

Delay to surgery in non-survivors 15.7 h

Seelig et al. (1981)

Statistical significance p <0.05

Fig. 3.1 Relationship between mortality, functional outcome and timing of surgery post injury in the study by Mendelow et al.

Post-operative control of ICP <20 mmHg was associated with a functional recovery in 79% of patients (p < 0.001). Poor pre-operative neurological status was associated with an increased mortality (p < 0.05).

Wilberger et al. (1991) Although there was a trend for earlier surgery to improve mortality rates and functional recovery this was only statistically significant in patients who underwent surgery >12 h following injury (p < 0.05) (see Figure 3.2).

Fig. 3.2 Relationship between mortality, functional recovery and timing of surgery post injury in the study by Wilberger et al.

Poor pre-operative neurological status (GCS <5) was significantly associated with high mortality (>75%, p < 0.05). Control of ICP was significantly associated with mortality and functional recovery: mortality with ICP <20 mmHg was 40% compared to a mortality of >95% with ICP > 45 mmHg (p < 0.05); no patients with ICP > 45 mmHg had a functional recovery (p < 0.05).

Conclusions Mendelow et al. (1979) Delay in surgical evacuation of EDH leads to increased morbidity and mortality, with delays of >2 h being unacceptable. 90

Seelig et al. (1981) The diagnosis and surgical evacuation of ASDH within 4 h of injury considerably reduces mortality.

Wilberger et al. (1991) Although timely evacuation of ASDH is warranted, effective control of intracranial pressure appears to be more critical to outcome.

Critique Extra-axial haematomas have long been considered surgically curable pathological entities and it has long been recognized that evacuation at the earliest opportunity may be beneficial (Putnam and Cushing, 1925; Chambers, 1951). The study by Mendelow et al. supported this view in the case of extradural haematomas with their finding that the mean delay to surgery was more than eight times longer in non-survivors (15.7 h) than survivors (1.9 h). The aim of the study by Mendelow and collegues was primarily to evaluate the effect of a change in head injury services in Edinburgh between two time periods. In the second time period, head-injured patients were admitted to a neurosurgical unit for 24 h routinely. This practice certainly reduced delays to surgery to a degree that almost reached statistical significance (p = 0.06). There have been other studies that have supported the immediate surgical intervention for EDH (Bricolo and Pasut, 1984). However, the series by Mendelow et al. is often cited as providing clear evidence that delay in evacuation following clinical deterioration is associated with worsening of outcome. The series by Seelig et al. is a landmark in the field of head injury as it established the earliest possible evacuation of ASDH, and preferably within 4 h, as a benchmark of neurosurgical care. The later series by Wilberger et al. although broadly supporting this finding, also revealed that the extent of the underlying primary brain injury and appropriate strategies to prevent secondary brain injury may be more important than the timing of surgery in determining outcome. This study is also a landmark, particularly because it has led to an increased focus on how to manage ICP during the interval between injury and surgery. These landmark studies that have been highly influential in establishing the general acceptance of timely intervention for traumatic extra-axial haematomas are the benchmark. Nonetheless, it should be recognized that there are studies that suggest there may be a role for a conservative approach in selected patients: in particular those with small-volume haematomas that do not cause compression or midline shift (Hamilton and Wallace, 1992; Croce et al., 1994; Servadi et al., 1998). Nonetheless, there is little contention regarding the role of emergency surgery in patients deteriorating from traumatic extra-axial lesions. Indeed, discussions regarding the evidence to support surgery in this situation have been likened to discussions about the evidence for the benefits for parachutes in determining outcome for sky-divers. However, there is an area of traumatic head injury that remains more controversial and that is the role of surgery in patients with traumatic contusions or traumatic intraparenchymal haematomas. In this setting, it appears that neurosurgeons are much more conservative in their approach and there appears to be enough equipoise to facilitate a large multi-centre randomized trial (Compagne et al., 2005). Indeed, the STITCH (trauma) trial started recruiting patients in 2009 (<http://research.ncl.ac.uk/trauma.STITCH/>). Further developments have been made with regard to optimizing other variables such as pre-hospital resuscitation and control of cerebral perfusion and intracranial pressure. However, traumatic EDH and ASDH represent a diverse population of patients and comprehensive guidelines have been developed by leading expert groups on how to best manage these lesions (Bullock et al., 2006a; Bullock et al., 2006b).

References Bricolo AP, Pasut LM. Extradural hematoma: toward zero mortality. A prospective study. Neurosurgery 1984; 14: 8–12. Bullock MR, Chestnut R, Ghajar J, Gordon D, Hartl R, Newell DW, Servadi F, Walters BC, Wilberger JE, for the Surgical Management of Traumatic Brain Injury Author Group. Surgical management of acute subdural haematomas. Neurosurgery 2006a; 58: S7–S15. Bullock MR, Chestnut R, Ghajar J, Gordon D, Hartl R, Newell DW, Servadi F, Walters BC, Wilberger JE, for the Surgical Management of Traumatic Brain Injury Author Group. Surgical management of acute subdural haematomas. Neurosurgery 2006b; 58: S16–S24. Chambers JW. Acute subdural hematoma. J Neurosurg 1951; 8: 263–268. Compagne C, Murray GD, Teasdale GM, Maas AIR, Esposito D, Princi P, D’Avella D, Servadi F. The management of patients with intradural post-traumatic mass lesions: a multi-centre survey of current approaches to surgical management in 729 patients coordinated by the European Brain Injury Consortium. Neurosurgery 2005; 57: 1183–1191. Croce MA, Dent DL, Menke PG, Robertson JT, Hinson MS, Young BH, Donovan TB, Pritchard FE, Minard G, Kudsk KA. Acute subdural haematoma: nonsurgical management of selected patients. J Trauma 1994; 36: 820–826. Hamilton M, Wallace C. Non-operative management of acute epidural hematoma diagnosed by CT: the neuroradiologist’s role. AJNR 1992; 13: 853–859. Putnam TJ, Cushing H. Chronic subdural hematoma: its pathology, its relation to pachymeningitis hemorrhagia and its surgical treatment. Arch Surg 1925; 11: 329–393. Servadi F, Nasi MT, Cremoni AM, Giuliani G, Cenni P, Nanni A. Importance of a reliable admission Glasgow Coma Scale score for determining the need for evacuation of posttraumatic subdural hematomas: a prospective study of 65 patients. J

Trauma 1998; 44: 868–873.

3.3 Surgery for Chronic Subdural Haematomas Details of Studies Two studies are included in this section. First is a series of 69 patients with chronic subdural haematoma (CSDH) treated between 1955 and 1960 either with craniotomy or burr hole evacuation (Svien and Gelety, 1964). This was the first study to demonstrate that burr hole evacuation is an overall better operation than craniotomy for the treatment of primary CSDH. Second is a RCT addressing one of the key questions regarding the surgical management of CSDH, namely the role of drains following burr hole evacuation (Santarius et al., 2009). The trial was carried out at Addenbrooke’s Hospital in Cambridge, United Kingdom, between 2004 and 2007.

Study References Main Studies Santarius T, Kirkpatrick PJ, Ganesan D, Chia HL, Jallah I, Smielewski P, Richards HK, Marcus H, Parker RA, Price SH, Kirollos RW, Pickard JD, Hutchinson PJ. Use of drains versus no drains after burr-hole evacuation of chronic subdural haematoma: a randomised controlled trial. Lancet 2009; 374: 1067–1073. Svien HJ, Gelety JE. On the surgical management of encapsulated subdural hematoma. J Neurosurgery 1964; 21: 172–177.

Related References Lega BC, Danish SF, Malhotra NR, Sonnad SS, Stein SC. Choosing the best operation for chronic subdural hematoma: a decision analysis. J Neurosurg 2010; 113(3): 615–621. Weigel R, Schmiedek P, Krauss JK. Outcome of contemporary surgery for chronic subdural haematoma: evidence based review. J Neurol Neurosurg Psychiatry 2003; 74: 937–943.

Study Design Svien and Gelety (1964) Case series. Class of evidence

III

Comparing

Craniotomy versus burr hole evacuation

Number of patients

Outcomes

Recurrence requiring re-drainage Functional outcome at discharge Functional outcome at follow-up (3 months to 8 years)

Number of centres

Inclusion criteria: age >18 years; symptomatic CSDH confirmed on CT scan, presence of CSDH membranes confirmed pre-operatively. Exclusion criteria: patients with radiological evidence of skull fracture, severe associated injuries or history of unconsciousness of >5 min.

Santarius et al. (2009) RCT. Class of evidence

Randomization

Drain versus no drain

Number of patients

215

Follow-up

Primary outcome: Recurrence requiring re-drainage Secondary outcomes: Mortality at 30 days and 6 months 100% follow-up for primary outcome and 98% for secondary outcomes

Number of centres

Inclusion criteria: age >18 years; symptomatic CSDH confirmed on CT scan. Exclusion criteria: indication for surgery other than burr hole evacuation; insertion of CSF shunt ipsilateral to CSDH within preceding 6 months; use of drain deemed unsafe by surgeon. Analysis was carried out on an intention-to-treat basis.

Results Svien and Gelety (1964) Santarius et al. (2009)

* Not done in the paper. â&#x20AC; Data on the

functional status at follow-up are too complex for presentation in a table.

The trial was stopped early because of the significant benefit in reduction of recurrence with the use of a drain. At discharge, patients with drains were reported to have fewer neurological deficits, a better functional status, and more favourable modified Rankin scores. There were no significant differences in complication rates between the two groups.

Conclusions Svien and Gelety (1964) Burr hole evacuation is a preferred technique to craniotomy in the treatment of CSDH.

Santarius et al. (2009) Use of a drain following burr hole drainage of chronic subdural haematoma is safe and associated with reduced recurrence and mortality at 6 months.

Critique CSDH is a common clinical problem associated with considerable morbidity and mortality. There are numerous surgical options available to address this problem and an equally large number of opinions in the neurosurgical field as to which is the best technique. In this section we discuss two papers. The first has initiated a change in practice from that of craniotomy to burr hole drainage being the treatment of choice for primary CSDH. The second has provided randomized trial level of evidence for the use of drains with burr hole drainage and is likely to change practice from not using to using drains. Until the paper by Svien and Gelety, the prevailing method of treatment of CSDH, including primary CSDH, was craniotomy (Markwalder, 1981; Markwalder and Seiler, 1985; Hamilton et al., 1993). The authors showed, for the first time, that it is not necessary to perform craniotomy and membranectomy for CSDH. In fact, the outcome was better and the recurrence rate lower, although the latter not reaching statistical significance. The standard of the evidence presented in this paper is barely of Class III, and a lot of criticism can be made about the rigorousness of the definition of clinical status of patents at discharge and follow-up. However, following this study, numerous prospective and retrospective case series as well as two meta-analyses have confirmed the findings of the paper by Svien and Gelety (Weigel et al., 2003; Mondorf et al., 2009; Lega et al., 2010). As many surgeons prefer minicraniotomies as the method of choice for evacuation of CSDH, believing that these are associated with a lower recurrence rate, it would be beneficial to compare these two techniques in a clinical trial (Hamilton et al., 1993; Lee et al., 2009; Mondorf et al., 2009). In addition to the discussion of craniotomy versus burr hole craniostomy, there has been an ongoing debate about the merits of drains with burr hole craniostomy. Therefore, in their trial, Santarius et al. have chosen to examine a single question: Does the use of a drain when carrying out burr hole craniostomy for CSDH affect outcome? The authors concede that their trial has the weakness of being a single-centre study. However, the trial has certainly been 93

successful in answering the question it addressed and the results are consistent with previous prospective studies (Wakai et al., 1990; Tsutsumi et al., 1997). There has also been reluctance among some neurosurgeons to use drains because of concerns about potential complications of drain insertion. However, this trial by Santarius et al., similarly to other studies, did not find any significant difference in complication rates with the use of a drain (Mori and Maeda, 2001; Lind et al., 2003). This trial stands out as a landmark study by showing that a simple additional intervention can have a significant beneficial effect on patient outcome. This finding was in marked contrast to the widely held beliefs regarding the use of drains (Santarius et al., 2008). This trial has also opened the way for further trials to evaluate important questions in the management of chronic subdural haematomas. Is there a role for steroids to prevent recurrence, for example? Undoubtedly, there will be further trials comparing operative techniques and perhaps even larger multi-centre trials.

References Hamilton MG, Frizzell JB, Tranmer BI. Chronic subdural hematoma: the role for craniotomy reevaluated. Neurosurgery 1993; 33: 67–72. Lee JK, Choi JH, Kim CH, Lee HK, Moon JG. Chronic subdural hematomas: a comparative study of three types of operative procedures. J Korean Neurosurg Soc 2009; 46: 210–214. Lega BC, Danish SF, Malhotra NR, Sonnad SS, Stein SC. Choosing the best operation for chronic subdural hematoma: a decision analysis. J Neurosurg 2010; 113(3): 615–621. Lind CR, Lind CJ, Mee EW. Reduction in the number of repeated operations for the treatment of subacute and chronic subdural hematomas by placement of subdural drains. J Neurosurg 2003; 99: 44–46. Markwalder TM. Chronic subdural hematomas: a review. J Neurosurg 1981; 54: 637–645. Markwalder TM, Seiler RW. Chronic subdural hematomas: to drain or not to drain? Neurosurgery 1985; 16: 185–188. Mondorf Y, Abu-Owaimer M, Gaab MR, Oertel JM. Chronic subdural hematoma—craniotomy versus burr hole trepanation. Br J Neurosurg 2009; 23: 612–616. Mori K, Maeda M. Surgical treatment of chronic subdural hematoma in 500 consecutive cases: clinical characteristics, surgical outcome, complications, and recurrence rate. Neurologia Medico-Chirurgica 2001; 41: 371–381. Santarius T, Lawton R, Kirkpatrick PJ, Hutchinson PJ. The management of primary chronic subdural haematoma: a questionnaire survey of practice in the United Kingdom and the Republic of Ireland. Br J Neurosurg 2008; 22: 529–534. Tsutsumi K, Maeda K, Iijima A, Usui M, Okada Y, Kirino T. The relationship of preoperative magnetic resonance imaging findings and closed system drainage in the recurrence of chronic subdural hematoma. J Neurosurgery 1997; 87: 870–875. Wakai S, Hashimoto K, Watanabe N, Inoh S, Ochiai C, Nagai M. Efficacy of closed-system drainage in treating chronic subdural hematoma: a prospective comparative study. Neurosurgery 1990; 26: 771–773. Weigel R, Schmiedek P, Krauss JK. Outcome of contemporary surgery for chronic subdural haematoma: evidence based review. J Neurol Neurosurg Psychiatry 2003; 74: 937–943.

3.4 Decompressive Craniectomy for Severe Traumatic Brain Injury Details of Study Decompressive craniectomy is widely used as a surgical method of treating medically refractory intracranial hypertension following severe TBI. Despite a suggestion from non-randomized trials and controlled trials with historical controls that such surgery may be beneficial if maximal medical treatment fails to control ICP, there has been a paucity of well-designed studies to evaluate its efficacy. The Decompressive Craniectomy (DECRA) trial is, therefore, a landmark study in being the first well-designed, multi-centre RCT published to address this issue. The DECRA trial was carried out between 2002 and 2010 in Australia, New Zealand, and Saudi Arabia.

Study References Main Study Cooper DJ, Rosenfeld JV, Murray L, Arabi YM, Davies AR, D’Urso P, Kossmann T, Ponsford J, Seppelt I, Reilly P, Wolfe R. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 364: 1493–1502.

Related Studies Hutchinson PJ, Corteen E, Czosnyka M, Mendelow AD, Menon DK, Mitchell P, Murray G, Pickard JD, Rickels E, Sahuquillo J, Servadei F, Teasdale GM, Timofeev I, Unterberg A, Kirkpatrick PJ. Decompressive craniectomy in traumatic brain injury: the randomized multicenter RESCUEicp study (www.RESCUEicp.com). Acta Neurochir Suppl 2006; 96: 17–20. Sahuquillo J, Arikan F. Decompressive craniectomy for the treatment of refractory high intracranial pressure in traumatic brain injury. Cochrane Database Syst Rev 2006; 1: CD003983.

Study Design Multi-centre RCT. Class of evidence

I 94

Randomization

Craniectomy plus standard care versus standard care alone

Number of patients

155

Follow-up

Primary outcome: GOSE at 6 months Secondary outcomes: ICP measurements Intracranial hypertension index Unfavourable GOSE score ICU stay Mortality

Number of centres

Stratification

Centre ICP monitoring technique (EVD versus intraparenchymal monitor)

Inclusion criteria: age 15â&#x20AC;&#x201C;59 years; severe, non-penetrating head injury (GCS 3â&#x20AC;&#x201C;8); informed consent from next of kin. Exclusion criteria: fixed dilated pupils; spinal cord injury; pre-hospital cardiac arrest. All patients were cared for in advance neuro-intensive facilities with ICP monitoring. Refractory ICP was defined as >20 mmHg for >15 min within a 1 h period. Patients were randomized within 72 h of injury to decompressive craniectomy plus standard care versus standard care alone. Standard care consisted of all accepted methods recommended by the Brain Trauma Foundation of controlling ICP: sedation, normalization of arterial PCO2, hyper-osmolar therapy, neuromuscular paralysis, and external ventricular drainage. Decompressive craniectomy consisted of a bifrontal craniectomy with bilateral dural opening but preservation of the falx and sagittal sinus. Assessors of GOSE were blinded to the treatment arm. Analysis was carried out on an intention-to-treat basis.

Results Primary Outcome The original primary outcome was an unfavourable outcome on the GOSE and power calculations required a sample size of 210 patients. However, at interim analysis it was determined that by changing the primary outcome to the GOSE score that the trial could be terminated at 150 participants.

The craniectomy group had a worse GOSE score than those receiving standard care alone (odds ratio1.84, p = 0.03) and a greater risk of unfavourable outcome (odds ratio 2.21, p = 0.02).

Secondary Outcomes

Conclusions Early decompressive craniectomy was effective in reducing ICP and length of ICU stay but it was associated with more unfavourable outcomes for adult patients with severe TBI.

Critique This is a significant landmark study as it is the first well-designed, large, multi-centre trial of a neurosurgical procedure versus standard care in the management of severe TBI. Although decompressive craniectomy is widely used to manage refractory increased ICP, the results of the DECRA trial highlight the need to conduct clinical trials to evaluate these methods. The finding that craniectomy was found to be associated with more unfavourable outcomes at 6 months calls into question the validity of decompressive craniectomy in severe TBI. The authors acknowledge that the change in the primary outcome during the course of the trial is not optimal. Nonetheless, the same results were observed for both the original and revised primary outcomes and no beneficial effect of craniectomy was found, contrary to the study hypothesis. The authors speculated that the harmful effect of craniectomy may have been due to axonal stretch injury when the brain expands outside the skull vault. Various reservations have been expressed about the DECRA trial. It has been suggested that bicoronal craniectomy may be too aggressive, and that results would be perhaps differ in unilateral craniectomy for asymmetric intracranial hypertension (Chi, 2011). The DECRA trial results were discussed in correspondence to the New England Journal of Medicine (Cramer and Slooter, 2011; Hautefeuille et al., 2011; Hutchinson and Kirkpatrick, 2011; Romero, 2011; Simard et al., 2011; Timmons et al., 2011). Timmons et al. expressed several concerns: the criteria for refractory ICP were not stringent enough (>20 mmHg for >15 min in any single hour) and did not allow for an escalated optimization of ICP; the inclusion criteria of patients was too narrow and no patients with mass lesions were included; the decision not to divide the falx and sagittal sinus limits the efficacy of decompression; there were more patients with bilaterally unreactive pupils in the surgical group which may skew results; there may have been a change in practice due to the long accrual time of the study (Timmons et al., 2011). These concerns have been firmly rejected by the authors of DECRA (Cooper et al., 2011). The definition of a refractory ICP used in the DECRA trial was in accordance with accepted guidelines. Mass lesions were excluded because outcomes are known to be different in these patients and further heterogeneity of the study population would be misleading. There are various surgical techniques that can be employed to perform a decompressive craniectomy and the method used in the DECRA trial achieved the goal of lowering ICP. The baseline characteristics of the two groups in the trial were balanced and even adjusting for fixed pupils the outcome data were not affected. The authors also emphasized that there was no reason to suspect a change in practice throughout the course of the study as all centres adhered to practice guidelines. One of the key issues following the DECRA trial is whether the result should lead to a practice change. At present the results of the Randomised Evaluation of Surgery of Intracranial Pressure (RESCUEicp) is still underway. RECSCUEicp results are eagerly awaited as they will complement the results of DECRA. RESCUEicp differs in two main ways from DECRA: the threshold for refractory ICP is higher (25 mmHg) and it does not exclude patients with previously evacuated haematomas. Indeed, the point has been emphasized that unilateral decompressive surgery at the time of mass lesion evacuation needs to be evaluated in a separate trial (Servadei, 2011). Nonetheless, there has been support for the view that the DECRA trial should be used to implement a practice change and prevent a harmful procedure in a specific subgroup of patients (Ray, 2011).

References Chi JH. Craniectomy for traumatic brain injury: results from the DECRA trial. Neurosurgery 2011; 68: N19–N20. Cooper DJ, Rosenfeld JV, Davies AR. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 376. Cramer OL, Slooter AJ. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 375. Hautefeuille S, Francony G, Payen JF. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 374–375. Hutchinson PH, Kirkpatrick PJ. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 375. Randomised Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intra-Cranial Pressure (RESCUEicp) home page: <http://www.rescueicp.com>. Ray K. Traumatic brain injury: poor outcome after decompressive craniectomy. Nat Rev Neurol 2011; 7: 242. Romero CM. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 373–374. Servadei F. Clinical value of decompressive craniectomy. N Engl J Med 2011; 364: 1558–1559. Simard JM, Kahle KY, Walcott BP. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 374. Timmons SD, Ullman JS, Eisenberg HM. Craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 365: 373.

3.5 Intracranial Pressure Monitoring in Head Injury Details of Studies Intracranial pressure (ICP) monitoring was introduced into neurosurgical practice in the second half of the twentieth century (Guillaume and Hanny, 1951; Lundberg, 1960). However, two studies carried out in the 1970s and early 1980s at the Medical College of Virginia in Richmond, Virginia, USA, firmly established the role of ICP monitoring 96

in head-injured patients (Miller et al., 1977; Miller et al., 1981). The influence of these studies has meant that over the years ICP monitoring has become the standard of care for severe TBI even though trial data was lacking. However, in 2012, a landmark trial of ICP monitoring in TBI was reported from centres in Bolivia and Ecuador (Chestnut et al., 2012). The trial was called Benchmark Evidence from South American Trials: Treatment of Intracranial Pressure (BEST:TRIP). Although it was widely believed that such a trial would not be possible due to the accepted value of ICP monitoring, the investigators overcame this problem by instituting a trial in intensive care units where the ICP monitoring was not routinely used due to local doubts about its efficacy. Notwithstanding that this landmark study is unlikely to change current ICP practice, it is included here as it is the first such randomized study to evaluate the efficacy of ICP monitoring.

Study References Main Studies Chestnut RM, Temkin N, Carney N, Dikmen S, Rondina C, Videtta W, Petroni G, Pridgeon J, Barber J, Machamer J, Chaddock K, Celix K, Cherner M, Hendrix T. A trial of intracranial-pressure monitoring in traumatic brain injury. N Engl J Med 2012; 367: 2471–2481. Miller JD, Becker DP, Ward JD, Sullivan HG, Adams WE, Rosner MJ. Significance of intracranial hypertension in severe head injury. J Neurosurg 1977; 47: 503–516. Miller JD, Butterworth JF, Gudeman SK, Faulkner JE, Choi SC, Selhorst JB, Harbison JW, Lutz HA, Young HF, Becker DP. Further experience in the management of severe head injury. J Neurosurg 1981; 54: 289–299.

Related References Farahvar A, Gerber LM, Chiu YL, Carney N, Hartl R, Ghajar J. Increased mortality in patients with severe traumatic brain injury treated without intracranial pressure monitoring. J Neurosurg 2012; 117: 729–734. Guillaume J, Hanny P. Manométrie intracranienne continué; intérêt de la méthode et prémiers résultats. [Continuous intracranial manometry; importance of the method and first results]. Rev Neurol 1951; 84: 131–142. Lundberg N. Continuous recording and control of in neurosurgical practice. Acta Psychiatr Neurol Scand 1960; 3: 190–193. Narayan RK, Kishore PRS, Becker DP, Ward JD, Enas GG, Greenberg RP, Da Silva AD, Lipper MH, Choi SC, Mayhall CG, Lutz HA, Young HF. Intracranial pressure: to monitor or not to monitor? A review of our experience with severe head injury. J Neurosurg 1982; 56: 650–659.

Study Designs Miller et al. (1977) and were retrospective case series analyses. Miller et al. (1981) was a prospective consecutive series analysis. Chestnut et al. (2012) was a multicentre RCT.

ICP recordings in the studies by Miller et al. were made by transducing from a ventricular catheter placed in the frontal horn of one of the lateral ventricles. In the first study by Miller et al. the initial threshold for treating raised ICP was a sustained rise over 40 mmHg. However, by the end of the study this was reduced to 30 mmHg and in the later studies the threshold was set at an ICP of 25 mmHg for over 15 min. Outcomes in the studies by Miller et al. were assessed using the 5-point GOS with a good outcome being defined as moderate disability or good recovery and a poor outcome being defined as a significant disability, vegetative state, or death. In the BEST:TRIP trial by Chestnut et al. an intraparenchymal ICP monitor was placed and ICP was maintained <20 mmHg in accordance with the Brain Trauma Foundation/American Association of Neurological Surgeons guidelines. The composite primary outcome in the BEST:TRIP trial was a composite of 21 components including measures of survival time, impaired consciousness, and functional and neuropsychological measures. The primary outcome was a composite score of 0â&#x20AC;&#x201C;100.

Results Miller et al. (1977) ICP >40 mmHg on admission was associated with a poor outcome: 69% mortality and only 25% good outcome (p < 0.01). Low ICP on admission (<10 mmHg) was associated with much better outcomes: 14% mortality and 78% survived to a good outcome (p < 0.05). These findings of early ICP recording were even more significant when applied to patients with diffuse brain injury. In patients with mass lesions, only very high ICP (>40 mmHg) was associated with a poor outcome. Patients with diffuse injuries who experienced delayed elevations in ICP >20 mmHg had a greater proportion of poor outcomes (46%) compared to those whose ICP remained <20 mmHg (21%, p < 0.02). 50% of fatalities were associated with uncontrolled intracranial hypertension.

Miller et al. (1981) The authors reported a significant correlation between ICP control and outcome (p < 0.001: more patients with a well-controlled ICP (<20 mmHg) throughout had a much better outcome (74% good outcome, 18% mortality) 98

compared to patients with raised but reducible ICPs (55% good outcome, 26% mortality) and those with uncontrolled ICP rises (only 3% good outcome, and 92% mortality).

Chestnut et al. (2012) The study reached the planned samples size of 324 determined by a power calculation and analysis was done on an intention-to-treat basis. There was no significant difference between the two groups in the primary outcome: composite score of 56/100 in the ICP monitored group compared to 53 in the imaging-clinical examination group (p = 0.49). There were no differences in 6-month mortality between the two groups: 41% in the ICP monitored group compared to 39% in the imaging-clinical examination group (p = 0.60).

Conclusions Miller et al. (1977) Elevated ICP is related to poor outcome in severely head-injured patients.

Miller et al. (1981) Even moderate intracranial hypertension is associated with a poor outcome in patients with severe head injuries.

Chestnut et al. (2012) Care focused on ICP monitoring is no better than care based on imaging and clinical examination in patients with severe TBI.

Critique Lundberg et al. reported a preliminary assessment of continuous ICP monitoring in a series of 30 patients in 1965 (Lundberg et al., 1965). In addition, Johnston and Jennett emphasized that ICP monitoring in head-injured patients might be used to aid diagnosis, guide management, and predict outcomes (Johnston and Jennett, 1973). However, until the studies from Virginia there was a poor correlation between ICP and outcome in severely head-injured patients as reported in the literature. These two studies are landmark studies because they led to the widespread use of ICP monitoring in head-injured patients. However, the value of managing patients with ICP monitoring has been challenged, particularly as there appeared to be similar outcomes in patients managed without (Stuart et al., 1983). The previous absence of a randomized trial was often cited to suggest that there is no evidence for the value of ICP monitoring and that its use may inappropriately prolong ventilation and intensive therapy in severely injured patients (Cremer et al., 2005). In a survey of 67 centres from 12 European countries, patients who were ICP monitored appeared to have more interventions and poorer outcomes than those who were not (Stocchetti et al., 2001). In a review of the available evidence, Stocchetti et al. outlined three potential conclusions regarding ICP monitoring (Stocchetti et al., 2001). The pessimistic view that it is not of proven value and the decision to monitor is a matter of personal opinion; the nihilistic view that it is an invasive procedure without proven benefit and should not be used; and the optimistic view that benefits may be demonstrated in large trials. The trial by results of the BEST:TRIP trial are a landmark addition to the literature on ICP monitoring in severe TBI. In their discussion, the authors of the trial acknowledge that the fact that the trial was conducted in Bolivia and Ecuador might limit the generalization of their findings to other patient populations. Furthermore, the fact that pre-hospital care in the study countries may be inferior to that in more affluent countries may have resulted in less severe brain injuries being included. Nonetheless, both the ICU care and the criteria for severe TBI were consistent with those in wealthier countries. Chestnut et al. were keen to emphasize, that whilst they do not challenge the value of recording ICP, they feel their data support a reassessment of treatments to manipulate ICP recordings. The Cambridge head injury group have published their evaluation of the BEST:TRIP trial and argue that there should be no fundamental change in the management of severe TBI (Hutchinson et al., 2013). Perhaps the most powerful objection by the Cambridge group was the non-conventional use of a composite outcome measure and they observe that if the more conventional GOS was used then mortality and favourable outcome favoured the ICP monitoring arm, although the difference was small (5%) and non-significant. Furthermore, the Cambridge group raise concerns regarding the power calculation as the study had only a 40% power to detect a 10% difference in favourable outcome on the GOS, this raising the risk of a type II error. They also highlight the inherent problems of focusing on the absolute value of ICP and argue that waveform analysis is also an essential component. Furthermore, a recent large prospective study of 1446 patients treated with ICP lowering measures in 20 trauma centres in New York State between 2000 and 2009 revealed that use of an ICP monitor was significantly lower mortality compared to those treated without a monitor (Farahvar et al., 2012). Although this was not a randomized trial, it is the largest prospective series to date and will add further support for the continued use of ICP monitoring in the management of these patients.

References Cremer OL, van Dijk GW, van Wensen E, Brekelmans GJ, Moons KG, Leenen LP, Kalkman CJ. Effect of intracranial pressure monitoring and targeted intensive care on functional outcome after severe head injury. Crit Care Med 2005; 33: 2207–2213. Farahvar A, Gerber LM, Chiu YL, Carney N, Hartl R, Ghajar J. Increased mortality in patients with severe traumatic brain injury treated without intracranial pressure monitoring. J Neurosurg 2012; 117: 729–734. Hutchinson PJ, Kolias AG, Czosnyka M, Kirkpatrick PJ, Pickard JD, Menon K. Intracranial pressure monitoring in severe traumatic brain injury. BMJ 2013; 346: f1000. Johnston IH, Jennett B. The place of intracranial pressure monitoring in neurosurgical practice. Acta Neurochir 1973; 29: 53– 63. Lundberg NL, Troupp H, Lorin H. Continuous recording of the ventricular-fluid pressure in patients with severe acute traumatic brain injury. A preliminary report. J Neurosurg 1965; 22: 581–590. Stocchetti N, Penny KI, Dearden M, Braakman R, Cohadon F, Iannotti F, Lapierre F, Karimi A, Maas A Jr, Murray GD, Ohman J, Persson L, Servadei F, Teasdale GM, Trojanowski T, Unterberg A, for the European Brain Injury Consortium. Intensive care management of head-injured patients in Europe: a survey from the European brain injury consortium. Intensive Care Med 2001; 27: 400–406. Stocchetti N, Lonhgi L, Zanier ER. Intracranial pressure monitoring for traumatic brain injury: available evidence and clinical implications. Minerva Anestesiol 2008; 74: 197–203. Stuart GG, Merry GS, Smith JA, Yelland JD. Severe head injury managed without intracranial pressure monitoring. J Neurosurg 1983; 59(4): 601–605.

3.6 Steroids in Head Injury Details of Studies The Corticosteroids Randomization After Significant Head Injury (CRASH) trial is the largest multi-centre, international RCT looking at the effect of methyl prednisolone on the risk of death and disability after head injury.

Study References Main Studies CRASH trial collaborators. Final results of MRC CRASH, a randomised placebo-controlled trial of intravenous corticosteroid in adults with head injury—outcomes at 6 months. Lancet 2005; 365: 1957–1959. CRASH trial collaborators. Lancet. Effects of intravenous corticosteroids on death within 14 days in 10008 adults with clinically significant head injury (MRC CRASH trial). Lancet 2004; 364: 1321–1328.

Related References Alderson P, Roberts I. Corticosteroids for acute traumatic brain injury. Cochrane Database Syst Rev 2005; 1: CD000196. Bracken MB. CRASH (Corticosteroids Randomisation after Significant Head Injury): landmark and storm warning. Neurosurgery 2005; 57: 1300–1302. Peto R. Possible explanations of the CRASH result. Lancet 2005; 364: 213.

Study Design Placebo-controlled randomized trial (PRCT). Class of evidence

Randomization

Methyl prednisolone infusion versus placebo

Number of patients

10,008

Follow-up

Primary outcomes: Death within 2 weeks of injury Death and disability at 6 months Secondary outcomes: None Percentage of patients followed up at each stage? 100% at 2 weeks 96.7% at 6 months

Number of centres

239 centres in 49 countries worldwide

Stratification

Presence or absence of generalized motor seizures

CRASH set out to recruit 20,000 patients but was stopped after 10,008 were randomized. There were more patients in CRASH than all previous trials looking at methyl prednisolone in head injury 100

combined. Methyl prednisolone was administered within 8 h of injury with a loading dose of 2 g (in 100 mL) over 1 h, followed by a 48 h maintenance dose.

Outcome Measures Primary Endpoints Death within 2 weeks. Death or disability (assessed with a questionnaire version of GOS) at 6 months.

Results

Conclusions Steroids should not be routinely used in the treatment of head injury.

Critique The CRASH trial is the first large-scale RCT in severe head-injured patients. Although many other previous trials had been carried out (>15), meta-analysis of previous trials had shown that there might be a potential benefit in the use steroids to treat head injury. However, once the CRASH trial is included in this meta-analysis the data support the conclusions of the CRASH trial. The CRASH trial conclusion could perhaps be more specific in that the results do not support the routine use of methyl prednisolone in the treatment of head injury rather than all steroids per se. The CRASH trial was criticized for testing only a small effect thus requiring such large numbers to be recruited. The trial was planned to detect a 2% reduction in mortality from 15% to 13% and was planned for the randomization of 20,000 patients. The CRASH trial is the largest trial in head injury to date and included more patients than all other trials combined that had previously addressed the role of steroids in head injury. In addition, CRASH is the largest trial of severe head injury as it included 3944 such patients. CRASH has conclusively answered the question regarding methyl prednisolone in the treatment of head injury. Furthermore, the massive population of head-injured patients included in the CRASH trial has allowed for the development of prognostic models for predicting outcome. The MRC CRASH Trial Collaborators developed two web-based prognostic models that can be applied to individual patients in order to predict outcome (MRC CRASH Trial Collaborators, 2008). This allows for the application of models based on population-based data to individual practice. In addition, this may be useful for prognostic stratification in future trials. The CRASH trial has allowed investigators to move on to research in other therapeutic modalities for head injury. A CRASH2 trial is now underway looking into the effects of tranexamic acid on survival following head injury. The ProTECT study is now underway to look at the effects of progesterone in head injury. The Phase II results of ProTECT have been published and no significant adverse effects were seen with progesterone (Wright et al., 2007). The ProTECT study has laid the foundations for a much larger multi-centre study into the effect of progesterone in the treatment of head injury. There has also been a trial investigating the neuroprotective effects of magnesium sulphate infusions after TBI.

Reference Wright DW, Kellermann AL, Hertzberg VS, Clark PL, Frankel M, Goldstein FC, Salomone JP, Dent LL, Harris OA, Ander DS, Lowery DW, Patel MM, Denson DD, Gordon AB, Wald MM, Gupta S, Hoffman SW, Stein DG. ProTECT: a randomised clinical trial of progesterone for acute traumatic brain injury. Ann Emerg Med 2007; 49: 391â&#x20AC;&#x201C;402.

3.7 Barbiturates in Head Injury Details of Study The study by Eisenberg et al. is the first RCT to assess the efficacy of pentobarbital to treat elevated ICP in severely head-injured patients. Although two previous trials had been carried out to assess pentobarbital to prevent rises in ICP in these patients, neither had revealed a benefit (Schwartz et al., 1984; Ward et al., 1985). The trial by Eisenberg et al. was carried out over a 5-year period between 1982 and 1987 in the United States. 101

Study References Main Study Eisenberg HM, Frankowski RF, Contant CF, Marshall LF, Walker MD. High-dose barbiturates control elevated intracranial pressure in patients with severe head injury. J Neurosurg 1988; 69: 15–23.

Related References Roberts I. Barbiturates for acute traumatic brain injury. Cochrane Database Syst Rev 2000; 2: CD000033. Schwartz ML, Tator CH, Rowed DW, Reif SR, Meguro K, Andrews DF. The University of Toronto Head Injury Treatment Study: a prospective, randomised comparison of pentobarbital and mannitol. Can J Neurol Sci 1984; 11: 434–440. Ward JD, Becker DP, Miller JD, Choi SC, Marmarou A, Wood C, Newlon PG, Keenan R. Failure of prophylactic barbiturate coma in the treatment of severe head injury. J Neurosurg 1985; 62: 383–388.

Study Design PRCT. Class of evidence

Randomization

Best conventional therapy versus pentobarbital

Number of patients

Follow-up

6 months Primary endpoint: Response to treatment Other endpoints: Survival; GOS

Number of centres

Stratification

Medical complications Time to randomization Initial GCS

Inclusion criteria: GCS 4–7 post-resuscitation; age 15–50 years; serum osmolality/≥315 mOsm/kg; mannitol given within 1 h prior to randomization. Exclusion criteria: GCS 3; fixed pupils; pregnancy. Conventional therapy standardized across all five centres. Intracranial mass lesions, haematomas, and accessible contusions resected. Patients with uncontrolled ICP despite best conventional therapy (BCT) received pentobarbital (titrated to serum concentration) plus continued BCT. ICP randomization criteria based on ICP levels and length of time ICP raised, e.g. randomization for closed head injury if ICP >25 mmHg (30 min), >30 mmHg (15 min), or 40 mmHg (1 min). Lower values used for open injuries. Response to treatment in patients with closed head injury defined as successful if ICP <20 mmHg for 48 h (lower value for open head injury). Failed reductions in ICP or severe clinical deterioration (e.g. fixed pupil, or death) were defined as unsuccessful treatment. Patients whose ICP remained uncontrolled in the BCT arm were allowed to cross-over to pentobarbital treatment.

Results

Multiple logistic model statistical analysis revealed a significant positive treatment effect of pentobarbital (p = 0.04). 102

Significant effects of timing of randomization were also found: twice as many. Uncontrolled ICP was robustly associated with death in both treatment arms (>90% of patients with controlled ICP survived).

Conclusions High-dose barbiturates are an appropriate adjunct in the control of raised ICP in severely head-injured patients.

Critique Severe head injury is associated with extremely high morbidity and mortality. Although the primary insult is not treatable, prevention of secondary injury due to the resultant cascade of insults may be feasible. The trial by Eisenberg et al. aimed to evaluate the role of barbiturate therapy in this regard. Unfortunately, only 12% of patients considered for randomization met the entry criteria and the total number of patients in the trial is still quite low. Nonetheless, the trial was still a multi-centre trial with standardized treatment regimens. Survival was not the primary outcome measure, but rather control of ICP. This was primarily to avoid the ethical dilemma of not using barbiturates when ICP reached potentially lethal levels. The authors pointed out that there is a possibility that raised ICP and outcome are both predetermined by the pathology of severe head injury. However, ICP >20 mmHg had been consistently found to be significantly associated with poor outcome in previous studies of severely head-injured patients, which provided an acceptable rationale to design a trial with ICP as primary outcome. As stated earlier in this section, two previous studies on prophylactic efficacy of barbiturates to control ICP had been performed that did not suggest a benefit (Schwartz et al., 1984; Ward et al., 1985). The trial by Eisenberg et al., therefore, has provided the first evidence that barbiturates may be effective in controlling ICP in severely headinjured patients. However, the debate regarding the relationship between ICP and outcome is still ongoing. In a Cochrane Review of all published trials involving barbiturates, Roberts concluded that although barbiturates controlled ICP in severely head-injured patients, there was no evidence to support any beneficial effect on outcome (Roberts, 2000).

References Roberts I. Barbiturates for acute traumatic brain injury. Cochrane Database Syst Rev 2000; 2: CD000033. Schwartz ML, Tator CH, Rowed DW, Reif SR, Meguro K, Andrews DF. The University of Toronto Head Injury Treatment Study: a prospective, randomised comparison of pentobarbital and mannitol. Can J Neurol Sci 1984; 11: 434–440. Ward JD, Becker DP, Miller JD, Choi SC, Marmarou A, Wood C, Newlon PG, Keenan R. Failure of prophylactic barbiturate coma in the treatment of severe head injury. J Neurosurg 1985; 62: 383–388.

3.8 Hyperosmolar Therapy for Control of Raised Intracranial Pressure in Head Injury Details of Studies Mannitol and saline are the two most widely used hyperosmolar therapies available to control raised ICP in head injury. The efficacy of mannitol has been studied by a series of reasonably sized randomized controlled clinical trials. The first of these was carried out in Toronto, Canada, and compared mannitol with pentobarbital (Schwartz et al., 1984). The second compared mannitol to saline, as opposed to hypertonic saline (Sayre et al., 1996). The third trial, carried out in North Carolina, USA, included two regimens both of which used mannitol but compared two different sets of parameters (physiological measurements versus ICP) to guide administration (Smith et al., 1986). A further three trials have been carried out by Cruz et al. in San Paolo, Brazil (Cruz et al., 2001; Cruz et al., 2002; Cruz et al., 2004). These three trials assessed the effects of early high-dose mannitol in three different groups of severely headinjured patients: patients with surgically treated acute subdural haematomas (Cruz et al., 2001); patients with acute temporal intraparenchymal haemorrhages (Cruz et al., 2002); and patients with severe diffuse brain injury (Cruz et al., 2004). These trials show a large variation in their design and in the primary question they set out to evaluate. Notwithstanding this we have included them all here as they are all randomized studies using mannitol in head-injured patients. The efficacy of hypertonic saline has also been evaluated in comparison with Ringer’s lactate in a large trial carried out in Melbourne, Australia (Cooper et al., 2004). In addition, one small study from Marseille, France, has compared hypertonic saline with mannitol (Vialet et al., 2003).

Study References Main Studies Mannitol Cruz J, Minoja G, Okuchi K. Improving clinical outcomes from acute subdural hematomas with the emergency preoperative administration of high doses of mannitol: a randomized trial. Neurosurgery 2001; 49: 864–871. Cruz J, Minoja G, Okuchi K. Major clinical and physiological benefits of early high doses of mannitol for intraparenchymal

103

temporal lobe hemorrhages with abnormal pupillary widening: a randomised trial. Neurosurgery 2002; 51: 628–637. Cruz J, Minoja G, Okuchi K, Facco E. Successful use of the new high-dose mannitol treatment in patients with Glasgow Coma Scale scores of 3 and bilateral abnormal pupillary widening: a randomized trial. Neurosurgery 2004; 100: 376–383. Sayre MR, Daily SW, Stern SA, Storer DL, van Loveren HR, Hurst JM. Out-of-hospital administration of mannitol does not change systolic blood pressure. Acad Emerg Med 1996; 3: 840–848. Schwartz M, Tator C, Rowed DW, Reid SR, Meguro K, Andrews DF. The University of Toronto Head Injury Treatment Study: a prospective randomised comparison of pento-barbital and mannitol. Can J Neurol Sci 1984; 11: 434–440. Smith HP, Kelly DL Jr, McWhorter JM, Armstrong D, Johnson RD, Transou C, Howard G. Comparison of mannitol regimens in patients with severe head injury undergoing intracranial pressure monitoring. J Neurosurg 1986; 65: 820–824.

Hypertonic Saline and Mannitol versus Hypertonic Saline Cooper DJ, Myles PS, McDermott FT, Murray LJ, Laidlaw J, Cooper G, Tremayne AB, Bernard SS, Ponsford J, for the HTS Study Investigators. Prehospital hypertonic saline resuscitation of patients with hypotension and severe traumatic brain injury: a randomized controlled trial. JAMA 2004 17; 291: 1350–1357. Vialet R, Albanese J, Thomachot L, Antonini R, Bourgouin A, Alliez B, Martin C. Isovolume hypertonic solutes (sodium chloride or mannitol) in the treatment of refractory posttraumatic intracranial hypertension: 2 mgL/kg 7.5% saline is more effective than 2 ml/kg 20% mannitol. Crit Care Med 2003; 31: 1683–1687.

Related Reference Wakai A, Roberts I, Schierhout G. Mannitol for acute traumatic brain injury. Cochrane Database Syst Rev 2007; 24: CD001049.

Study Designs All PRCTs but blinding varied between trials.

Schwartz et al. (1984)

Hyperosmolar treatment regimen 1 g/kg of 20% mannitol given initially (followed by incremental titration to achieve ICP control) and pentobarbital as an IV bolus of up to 10 mL/kg followed by a continuous infusion at 0.5–3 mg/kg/h

Smith et al. (1986)

Mannitol administration guided according to neurological signs and physiological parameters or gave a 250 mL bolus of 20% mannitol for ICP >25 mmHg followed by incremental boluses as required

Sayre et al. (1996)

Pre-hospital administration of 5 ml/kg of 20% mannitol compared with administration of 5 mL/kg of 0.9% saline

Cruz et al. (2001, 2002, 2004)

Early high-dose mannitol (HDM) at 1.4 g/kg compared with conventional-dose mannitol (CDM) at 0.7 g/kg

Cooper et al. A 250 mL infusion of 7.5% hypertonic saline was compared with 250 mL of Ringer’s lactate (both through a peripheral line) (2004) Vialet et al. (2003)

2 mL/kg body weight of 20% mannitol compared with 2 mL/kg body weight of 7.5% hypertonic saline solution (HSS) given when either ICP >25 mmHg, or CPP <70 mmHg for >5 min

The trial by Schwartz et al. contained two cohorts of patients: those without intracranial haematomas and those with elevated ICP following haematoma evacuation. The three trials by Cruz et al. looked at three different cohorts of patients respectively: those with subdural haematomas (ASDH); those with temporal intraparenchymal haematomas (temporal IPH); and those with diffuse brain injury (DBI). Vialet et al. defined treatment failure as an inability to reduce ICP below 25 mmHg or to increase cerebral perfusion pressure (CPP) to >70 mmHg after two sequential boluses of hyperosmolar fluid.

Results Schwartz et al. (1984)

104

Smith et al. (1986) Mean ICP was 5.5 mmHg lower in the empirically treated group compared to the ICP-guided group (p <0.05).

Sayre et al. (1996) After 2 h, those patients receiving mannitol had a lower systolic blood pressure compared to placebo (116 Âą 24 mmHg versus 142 Âą 25 mmHg, p < 0.003). However, there was no overall difference in blood pressure between mannitol and placebo groups over the whole 2 h observation period.

Cruz et al. (2001, 2002, 2004) 6-month clinical outcomes.

In their 2001 study of patients with ASDH, Cruz et al. reported a significant improvement in abnormal preoperative pupillary widening in the HDM group (p < 0.0001). In their 2002 study of patients with traumatic temporal IPH, Cruz et al. reported that HDM resulted in significant improvements in pre-operative abnormal pupillary widening both bilaterally (p < 0.03) and unilaterally (p < 0.01). In their 2004 study of patients with DBI, Cruz et al. reported a significant early improvement of bilateral abnormal pupillary widening in the HDM group (p < 0.02).

Cooper et al. (2004) 6-month clinical outcomes.

Vialet et al. (2003) 105

Mean daily outcomes. There was no significant difference in mortality or GOS between the two treatment arms.

Conclusions Schwartz et al. (1984) There is no difference between mannitol and pentobarbital in the treatment of intracranial hypertension following head injury. Pentobarbital may be harmful in head-injured patients without intracranial haematomas.

Smith et al. (1986) Regular, frequent administration of mannitol may result in better overall control of ICP than waiting until ICP rises above 25 mmHg.

Sayre et al. (1996) Out-of-hospital administration of 1.0 g/kg of mannitol to multiple-trauma head-injured patients is not associated with significant hypotension.

Cruz et al. (2001, 2002, 2004) Early administration of HDM leads to significant improvements and better clinical outcomes in patients with ASDH, traumatic temporal IPH, and DBI.

Cooper et al. (2004) Hypertonic saline in the pre-hospital setting for severely head-injured patients is no better than conventional fluids alone for resuscitating hypotensive patients or improving neurological outcomes at 6 months.

Vialet et al. (2003) Hypertonic saline was more effective in controlling refractory intracranial hypertension than mannitol in patients with severe head injury.

Critique The utility of mannitol in controlling ICP has been well established. Its effect on outcome has been more variable. Early data on hypertonic saline are similar to that for mannitol. The study by Schwartz et al. focused predominantly on the effect of mannitol on ICP control. This study has been criticized because it allowed cross-over between groups if there was subsequent raised ICP. Nonetheless, the finding that pentobarbital may have been harmful in the patients in this study has contributed considerably to the widespread use of mannitol to control ICP in head-injured patients. The trial by Smith et al. assessed whether there was any benefit in the use of ICP monitoring to guide the use of mannitol in severely head-injured patients. They found that there was no statistically significant difference in either mortality or neurological outcome using ICP monitoring. However, the authors did suggest that their finding that mean ICP was lower in the empirically treated group might indicate that better ICP control could be achieved with small regular doses of mannitol rather than waiting until ICP rises above 25 mmHg. In the study by Sayre et al., death was not the primary outcome as they powered their study to detect a drop in systolic BP to <90 mmHg (83% power). The study was not, therefore, powered to detect a difference in survival at 2 h. However, as hypotension is associated with a doubling in mortality in patients with severe head injury their chosen endpoint is a reasonable one (Gentleman and Jennett, 1981). Cooper et al.â&#x20AC;&#x2122;s study was powered to detect a single grade change in GOS, which represents a meaningful clinical difference in outcome. This trial is also, therefore, a landmark trial for being the first resuscitation fluid trial to measure neurological outcome as a primary outcome. One of the weaknesses of Cooper et al.â&#x20AC;&#x2122;s study is that it included multiple-injured patients, which raised concerns regarding whether conclusions could be extrapolated to patients, which isolated head injuries. Although this is a concern, Cooper et al. reported that the outcomes of patients were better than would have been predicted from trauma severity scores. The trials by Sayre et al. and Cooper et al. are particularly noteworthy in that they are examples of randomized 106

blinded trials conducted for interventions in the pre-hospital setting. Lewis has emphasized that trials conducted for therapies in this setting are particularly difficult because of difficulties with availability of personnel, equipment, space, and lighting (Lewis, 2004). Lewis also highlights that such trials face ethical issues regarding consent because of the need to show that patients are in a life-threatening situation and that available treatment options are unproven or are believed to be ineffective. Only in these circumstances can prospective written informed consent be bypassed to enrol patients into a trial. The patients in the first HDM trial by Cruz et al. had acute traumatic subdural haematomas requiring emergency evacuation. In such circumstances, common practice in most centres is to give mannitol to patients with pupillary abnormalities, and nothing to patients with normal pupillary responses, yet all these patients initially received lowdose (0.6–0.7 g/kg) mannitol regardless. A recent Cochrane Review questioned the integrity of Cruz’s data, in particular its randomization (Wakai et al., 2007). Wilberger found it particularly noteworthy that the mortality rate reported for this study is the lowest ever observed (Wilberger, 2001). Marion indicated that the management by Cruz pays close attention to the standardization of care and the close monitoring of physiological parameters and control of cerebral perfusion and oxygenation (Marion, 2001). However, Marion also raised concerns that statistical analysis did not take into account confounding variables such as the initial GCS. Concerns have been raised regarding the unusual lack of hypotensive episodes in patients included in the trials reported by Cruz (Valadka, 2002). Perhaps the most intriguing question resulting from Cruz’s work, however, is how a single early HDM bolus might lead to such long-term beneficial effects. Notwithstanding this, Zygun has criticized the lack of any blinding that would have been feasible seeing as the HDM was a single one-off bolus given early in the management of the patient (Zygun, 2004). More serious concerns regarding the integrity of Cruz’s work have been expressed in the literature (Roberts et al., 2007). However, these studies have not been formally retracted from the literature and so they have been included here. We leave it to the reader to determine for themselves whether the data should be considered reliable. Cooper’s study comparing mannitol with hypertonic saline has been criticized for being powered to detect a 20% improvement in GOSE. At the lower end of the GOSE, this could represent a change from death to persistent vegetative state (Zygun, 2004a; Zygun, 2004b). Zygun has expressed the view that the attainment of a functional neurological status would be deemed a more meaningful outcome by most clinicians. It is possible that a higher than normal serum sodium on arrival in hospital in the hypertonic saline group could have affected blinding in this study. Vialet’s trial had small numbers of patients, but was sufficiently powered as repeated measures were taken from each patient. However, Vialet’s small numbers raise concerns about case heterogeneity, and the extremely poor outcomes at 90 days (all patients in the study were dead or severely disabled). This raises questions about whether the study group was appropriately representative or managed optimally otherwise. The evidence for hyperosmolar therapy is limited both in number of studies undertaken and in delineating several aspects of its use. There remains little evidence about whether hyperosmolar therapies should be given as boluses or continuous infusions, whether there is an optimal dose, an optimal rate, whether losses from diuresis should be replaced, whether clinical thresholds should be guided by ICP or according to fixed schedules (and if so when they should be best timed), and whether serum osmolarity alterations alter outcomes. All these studies are landmark studies as they have greatly contributed to our knowledge regarding the use of hypertonic osmotic therapies in head-injured patients. Nonetheless, the mechanism of action of these agents remains to be fully elucidated. Although osmotic tissue dehydration may still play some role in the action of hyperosmolar therapies, they work primarily through immediate rheological effects, diluting the blood and increasing the deformability of erythrocytes, thereby decreasing blood viscosity and promoting cerebral blood flow. Thus, mechanistic studies suggest that bolus administration and replacing urinary losses are best practice. There is limited clinical evidence that successive mannitol boluses accumulate in cerebral tissue and exacerbate ICP and that cumulative hyperosmolar effects can have detrimental neurological sequelae (Wakai et al., 2007). Theoretical concerns with hypertonic saline include the development of central pontine myelinolysis and rapid brain shrinkage leading to tearing of bridging vessels. However, pontine myelinolysis is seen if chronic hyponatraemia is rapidly corrected, and hyponatraemia is usually not an immediate problem in acute severe head injury. There is no evidence that hypertonic saline is superior to mannitol and both are in widespread clinical use.

References Gentleman D, Jennett B. Hazards of inter-hospital transfer of comatose head-injured patients. Lancet 1981; 2: 853–854. Lewis RJ. Prehospital care of the multiply injured patient. The challenge of figuring out what works. JAMA 2004; 291: 1382– 1383. Marion DW. Improved outcomes with high dose mannitol treatment: comments. Neurosurgery 2001; 49: 871. Roberts I, Smith R, Evans S. Doubts over head injury studies. BMJ 2007; 334: 392–394. Wakai A, Roberts I, Schierhout G. Mannitol for acute traumatic brain injury. Cochrane Database Syst Rev 2007; 24: CD001049. Valadka AB. Emergency benefits of high-dose mannitol: comments. Neurosurgery 2002; 51: 637–638. Wilberger JE. Improved outcomes with high dose mannitol treatment: comments. Neurosurgery 2001; 49: 871. Zygun D. High-dose mannitol. J Neurosurg 2004a; 101: 567. Zygun D. Hypertonic saline for prehospital treatment of traumatic brain injury. JAMA 2004b; 24: 2943.

107

3.9 Hypothermia in Head Injury Details of Study There have been numerous small single-centre trials assessing the role of therapeutic hypothermia in severe TBI. However, the first multi-centre randomized trial of treatment with hypothermia for patients with severe TBI was the National Acute Brain Injury Study: Hypothermia (NABIS: H I) carried out by Clifton et al. (2001). NABIS: H I aimed to determine the efficacy of therapeutic hypothermia within 8 h of injury and was carried out in the United States between 1994 and 1998. The second multi-centre randomized trial was the National Acute Brain Injury Study: Hypothermia (NABIS: H II), also by Clifton et al. (2011). NABIS: H II aimed to assess the efficacy of very early therapeutic hypothermia within 2.5 h of injury and was also carried out in the United States and Canada between 2005 and 2009.

Study Reference Main Studies Clifton GL, Miller ER, Choi SC, Levin HS, McCauley S, Smith K, Muizelaar JP, Wagner FC, Marion DW, Luerssen TG, Chestnut RM, Schwartz M. Lack of effect of induction of hypothermia after acute brain injury. N Engl J Med 2001; 344: 556–563. Clifton GL, Valadka, Zygun D, Coffey CS, Drever P, Fourwinds S, Janis LS, Wilde E, Taylor P, Harshman K, Conley A, Puccio A, Levin HS, McCauley SR, Bucholz RD, Smith KR, Schmidy JH, Scott JN, Yonas H, Okonkwo DO. Very early hypothermia induction in patients with severe brain injury (the National Acute Brain Injury Study: Hypothermia II): a randomised trial. Lancet 2011; 10: 131–139.

Related References Clifton GL, Coffey CS, Fourwinds S, Zygun D, Valadka A, Smith KR Jr, Frisby ML, Bucholz RD, Wilde EA, Levin HS, Okonkwo DO. Early induction of hypothermia for evacuated intracranial hematomas: a post hoc analysis of two clinical trials. J Neurosurg 2012; 117: 714–720. Harris OA, Colford JM, Good MC, Matx PG. The role of hypothermia in the management of severe brain injury. A metaanalysis. Arch Neurol 2002; 59: 1077–1083. Henderson WR, Dhingra VK, Chittock DR, Fenwick JC, Ronco JJ. Hypothermia in the management of traumatic brain injury. A systematic review and meta-analysis. Intensive Care Med 2003; 29: 1637–1644. Hutchison JS, Ward RE, Lacroix JL, Hebert PC, Barnes MA, Bohn DJ, Dirks PB, Douchette S, Fergusson D. Gottesman R, Joffe AR, Kirkpalani HM, Meyer PG, Morris KP, Moher D, Singh RN, Skippen PW for the Hypothermia Pediatric Head Injury Trial Investigators and the Canadian Critical Care Trials Group. Hypothermia therapy after traumatic brain injury in children. N Engl J Med 2008; 358: 2447–2456. McIntyre LA, Fergusson DA, Hérbert PC, Moher D, Hutchison JS. Prolonged therapeutic hypothermia after traumatic brain injury in adults. JAMA 2003; 289: 2992–2999. Shiozaki T, Hayakata T, Taneda T Nakajima Y, Hashiguchi N, Fujimi S, Nakamori Y, Tanaka H, Shimazu T, Sugimoto H. A multicenter prospective randomised controlled trial of the efficacy of mild hypothermia for severely head injured patients with low intracranial pressure. J Neurosurg 2001; 94: 50–54. Sydenham E, Roberts I, Alderson P. Hypothermia for traumatic head injury. Cochrane Database Syst Rev 2009; 2: CD001048.

Study Design Both multi-centre RCTs.

Class of evidence

NABIS: H I II

NABIS: H II II

Randomization

Hypothermia versus normothermia

Early hypothermia versus normothermia

Number of patients

392 patients

Follow-up

6 months

Primary outcome: Functional status

Secondary outcome: None

Number of centres

Stratification

GCS score

By centre

NABIS: H I 108

Inclusion criteria: age 16–65 years; non-penetrating head injury; GCS 3–8 post-resuscitation. Patients were excluded in the following circumstances: fixed pupils; poor resuscitation (low systolic BP, or poor oxygenation); non-neurological life-threatening injury; persistent medical condition; bleeding; pregnancy; delay in initiation of cooling. Target cooling temperature was 33°C (bladder) and was achieved with a combination of surface cooling, cold fluids, gastric lavage, and room air ventilation. Target temperature was aimed to be achieved within 8 h following injury. All patients had ICP monitoring. Analysis on an intention-to-treat basis.

NABIS: H II Inclusion criteria: age 16–45; non-penetrating head injury; reduced conscious level with motor score of 5 or less on the GCS. There were two sets stringent of exclusion criteria: acute exclusion criteria in the at randomization field or on arrival in the emergency department—pregnancy, significant cardiovascular compromise (hypotension or tachycardia), or not reachable by study personnel within 2.5 h; and exclusion criteria after resuscitation—fixed pupils, poor resuscitation (low systolic blood pressure (BP), or poor oxygenation); non-neurological lifethreatening injury, confirmed pregnancy, or normal CT brain despite coma. Patients randomized to cooling were kept at 35°C until resuscitation was completed and then the target cooling temperature of 33°C was achieved with a combination of surface cooling, cold fluids, gastric lavage and room air ventilation. Patients assigned to hypothermia were cooled for 48 h.

Outcome Measures Primary Endpoint In both trials the GOS was assessed at 6 months: favourable outcome defined as good recovery or moderate disability; poor outcome defined as severe disability or worse.

Other Endpoints In NABIS: H I neurobehavioural and neuropsychological tests at 6 months. Deaths and complications were also recorded in both trials.

Results NABIS: H I The trial was stopped after 392 patients enrolled (199 to hypothermia arm, 193 to normothermia arm). Interim analysis had shown a probability of <0.01 of detecting a treatment effect if the trial was continued to target numbers of 500 patients.

In patients over the age of 45 there was a slightly higher percentage with poor outcome in the hypothermia group (88%) compared to the normothermia group (69%) but the difference was not statistically significant. No differences found in the two groups on neurobehavioural or neuropsychological testing. Patients who were hypothermic on admission who were randomized to the hypothermic arm had 17% less poor outcomes than hypothermic patients randomized to the normothermia arm (not statistically significant). Although mean ICP was not affected by hypothermia there were statistically fewer patients with ICPs over 30 in the first 4 days.

NABIS: H II Of 232 patients randomized, 135 were excluded after resuscitation. The trial was stopped for futility after an interim analysis of 97 patients (67 hypothermia, 68 normothermia). Patients in the hypothermia group received significantly more interventions for raised ICP (p = 0.002).

109

There were no significant differences in complications between the two groups. Subgroup analyses suggested that there might be differences between the two groups depending on whether there was diffuse brain injury or removal of intracranial haematoma, and this trend reached significance for intracranial haematomas.

Conclusions NABIS: H I Hypothermic treatment within 8 h following severe TBI does not improve functional outcome.

NABIS: H II Early hypothermic treatment within 2.5 h following severe TBI did not improve outcome. However, there role of early hypothermia in patients with evacuated haematomas needs further evaluation.

Critique There are two hypothetical mechanisms by which hypothermia has been postulated to be beneficial in severe TBI: control of ICP rises; and a neuroprotective effect preventing secondary brain injury. This study by Clifton et al. is by far the largest study to date looking at the clinical effects of therapeutic hypothermia in severe TBI and has been called a landmark achievement for proving that it is ineffective (Narajan, 2001). It is of note that the NABIS: H I trial was discontinued early due to lack of treatment effect. The authors also raised the possibility that there may even have been a detrimental effect of hypothermia in patients over the age of 45, although this was not statistically significant. It has been suggested that there is a possibility that the results of the NABIS: H I trial were affected by differences in fluid therapy and medication therapies between centres participating in the trial (Polderman et al., 2001). Indeed, there was a statistically significant difference between the two treatment arms in terms of fluid balance, vasopressor therapy, and use of muscle relaxants. However, four separate meta-analyses of the available published trials support the conclusion that hypothermia is not beneficial in severe TBI (Harris et al., 2002; Henderson et al., 2003; McIntyre et al., 2003, Sydenham et al., 2009). Despite the negative result of the NABIS: H I trial, several points were highlighted by the authors that warranted further investigation. They observed that patients who were hypothermic on admission appeared to have a slightly better outcome (Clifton et al., 2002). This significant observation was the impetus to carry out the NABIS: H II trial (Clifton, 2004). The NABIS: H II trial was limited mainly by its size and the fact that it was terminated early. Nonetheless, the trial did not show any effect of very early hypothermic treatment and is a significant landmark study. The authors emphasized that their subgroup analysis suggested that there may be some benefit for patients undergoing evacuation of intracranial haematomas. In order to assess this further they undertook a post hoc analysis of both trials (Clifton et al., 2012). Patients in NABIS: H I were selected who reached early hypothermia following craniotomy at the same time points as those in NABIS: H II, i.e. 35° C within 1.5 h of craniotomy. A meta-analysis of the two trials was then carried out comparing patients who reached early hypothermia following craniotomy (within 1.5 h) compared to those who did not (late hypothermia), or were treated with normothermia. The authors found a significantly reduced poor outcome in those successfully cooled to 35° C within 1.5 h of craniotomy (41%) compared to the late hypothermia/normothermia group (62%, p = 0.009). The authors acknowledge that this finding has the weakness of being a post hoc analysis, but hypothesized that this effect may be due to a protective effect of hypothermia on cerebral reperfusion injury following the evacuation of haematomas. It is envisaged that these findings will form the basis of further clinical trial evaluation of hypothermia in severe TBI.

References Clifton GL. Is keeping cool still hot? An update on hypothermia in brain injury. Curr Opin Crit Care 2004; 10: 116–119.

110

Clifton GL, Miller E, Choi SC, Levin HS, McCauley S, Smith KRJ, Muizelaar JP, Marion DW, Luerssen TG. Hypothermia on admission in patients with severe brain injury. J Neurotrauma 2002; 19: 293–301. Clifton GL, Coffey CS, Fourwinds S, Zygun D, Valadka A, Smith KR Jr, Frisby ML, Bucholz RD, Wilde EA, Levin HS, Okonkwo DO. Early induction of hypothermia for evacuated intracranial hematomas: a post hoc analysis of two clinical trials. J Neurosurg 2012; 117: 714–720. Harris OA, Colford JM, Good MC, Matx PG. The role of hypothermia in the management of severe brain injury. A metaanalysis. Arch Neurol 2002; 59: 1077–1083. Henderson WR, Dhingra VK, Chittock DR, Fenwick JC, Ronco JJ. Hypothermia in the management of traumatic brain injury. A systematic review and meta-analysis. Intensive Care Med 2003; 29: 1637–1644. McIntire LA, Fergusson DA, Herbert PC et al. Prolonged therapeutic hypothermia after traumatic brain injury in adults. JAMA 2003; 289: 2992–2999. Narajan RK. Hypothermia for traumatic brain injury—a good idea proved ineffective. [editorial] N Engl J Med 2001; 344: 602– 603. Polderman KH, Girbes ARJ, Peerdeman SM, Vandertop WP. Hypothermia. [review/comment] J Neurosurg 2001; 94: 853– 855. Sydenham E, Roberts I, Alderson P. Hypothermia for traumatic head injury. Cochrane Database Syst Rev 2009; 2: CD001048.

3.10 Hyperventilation in Head Injury Details of Study The only randomized trial evaluating the role of hyperventilation on outcome in severe TBI was (Muizelaar et al., 1991) carried out at the Medical College of Virginia and was published in 1991. This trial has been the most influential study to date in indicating that prolonged hyperventilation should be avoided in severe TBI. The trial compared normoventilation, hyperventilation, and hyperventilation plus tromomethamine (THAM). THAM was introduced to examine whether there was any effect of loss of CSF buffer during hyperventilation.

Study References Main Study Muizelaar JP, Marmarou A, Ward JD, Kontos HA, Choi SC, Becker DP, Gruemer H, Young HF. Adverse effects of prolonged hyperventilation in patients with severe traumatic brain injury. J Neurosurg 1991; 75: 731–739.

Related Reference Schierhout G, Roberts I. Hyperventilation therapy for acute traumatic brain injury. Cochrane Database Syst Rev 2000; 2: CD000566.

Study Design RCT. Class of evidence

Randomization

Normoventilation versus hyperventilation versus THAM

Number of patients

113

Follow-up

Primary outcome: GOS at 3, 6, and 12 months Secondary outcome: None 100% follow-up

Number of centres

Stratification

On severity of head injury based on motor score of GCS (1–3 and 4–5)

Inclusion criteria: age >3; GCS <9 following resuscitation and treatment of mass lesion. In the normoventilation group, PaCO2 was kept in the range 30–35 mmHg for a period of 5 days. In the hyperventilation groups, PaCO2 was kept in the 24–28 mmHg range for a period of 5 days. THAM was administered as a bolus followed by a sustained intravenous infusion for 5 days.

Outcomes GOS scores were used and a favourable outcome was defined as a good outcome/moderate disability.

111

Results The authors classified patients as having a favourable outcome if they were good or moderately disabled according to the GOS. Outcome results at 3 and 6 months for patients presenting with a GCS motor score 4–5.

There were no differences at 12-month follow-up. No differences were seen in the group with lower motor scores.

Conclusions Prophylactic hyperventilation is deleterious in head-injured patients who presented with a motor score of 4–5. The authors also concluded that the deleterious effect of sustained hyperventilation could be overcome by THAM.

Critique Hyperventilation results in vasoconstriction induced by hypocarbia and effectively lowers ICP by reducing cerebral blood flow. Hyperventilation to rapidly reduce ICP has, therefore, long been employed in the management of TBI. However, there has always been concern that the reduction of cerebral blood flow itself could be deleterious due to resultant cerebral ischaemia. The authors of this study concluded that prolonged prophylactic hyperventilation is deleterious in head-injured patients. This study has been extremely influential as it has prompted recommendations that prolonged hyperventilation in TBI patients should be avoided because of potentially deleterious effects on outcome. However, this study has several weaknesses. Firstly, there was no blinding of the evaluator to the treatment received. Secondly, the majority of patients (86%) in this study did not have raised ICP on admission to hospital. Hyperventilation was, therefore, used prophylactically in this study, and it remains to be elucidated as to whether such measures to reduce ICP once it is elevated could be effective. Thirdly, there was no power calculation employed to determine the sample size required. Schierhout and Roberts concluded that the data from this trial were insufficient to determine whether hyperventilation is harmful or beneficial in TBI (Schierhout and Roberts, 2000). Nonetheless, this study is the best study available to date examining the effects of hyperventilation. Further studies are required to examine what levels of PaCO2 are optimal for the severe TBI patient and whether there is any role for either prolonged or transient hyperventilation to reduce elevated ICP in these patients.

Reference Schierhout G, Roberts I. Hyperventilation therapy for acute traumatic brain injury. Cochrane Database Syst Rev 2000; 2: CD000566.

3.11 Magnesium for Neuroprotection in Head injury Details of Study The study by Temkin et al. (2007) set out to determine whether there is any benefit from magnesium sulphate infusions in preventing secondary brain injury in patients with TBI. The study was carried out at the Haborview Medical Center in Seattle, Washington, USA, over a 6-year period between 1998 and 2004.

Study References Main Study Temkin NR, Anderson GD, Winn HR, Ellenbagen RG, Schuster J, Lucas T, Newell DW, Mansfield PN, Machamer JE, Barber J, Dikmen SS. Magnesium sulfate for neuroprotection after traumatic brain injury: a randomised controlled trial. Lancet Neurol 2007; 6: 29–38.

Related Reference Muir KW, Lees KR, Ford I, Davis S. Magnesium for acute stroke (Intravenous Magnesium Efficacy in Stroke trial): a randomised controlled trial. Lancet 2004; 363: 439–445.

Study Design Double-blind, parallel group, randomized trial. 112

A phase III trial. Class of evidence

Randomization

Magnesium versus placebo

Number of patients

499

Follow-up

6 months Primary outcomes: Functional status, seizures, neuropsychological tests Secondary outcome: None

Number of centres

Stratification

Severity of injury Age of patient

IV MgSO4 (or saline placebo) given within 8 h of injury and continued for 5 days. Target plasma concentration was initially high 1.25–2.5 mmol/L but this was adjusted later to a lower level of 1.0–1.8 mmol/L. Patients with hypomagnesia had their magnesium levels corrected. Moderate TBI defined as: GCS 9–12 (or motor score 4–5). Severe TBI defined as: 3–8 (or motor score 1–3), or need for intracranial surgery. Patients excluded if <14 years of age, pregnant, or delay in receiving infusion. Analysis on intention-to-treat basis.

Outcome Measures Primary Endpoints Survival time. Functional outcome at 6 months: GOSE. Seizures—time to early (<1 week post injury) or late (>1 week post injury). Neuropsychological tests of attention, information processing, memory, and intellectual function. All endpoints analysed at 6 months although some, e.g. GOSE, assessed at 1 and 3 months also.

Secondary Endpoint None.

Results 93% follow-up at 6 months. Primary outcomes were analysed as a composite and no positive effect for MgSO4 found. Blood pressure was lower in the MgSO4 treated group for the higher serum levels but not for the lower target levels.

GOSE of >5 (moderate disability or better) was 49% for high-dose MgSO4 versus 40% for placebo and 38% for low-dose MgSO4 versus 42% for placebo.

Conclusions Intravenous MgSO4 given within 8 h of moderate to severe TBI does not improve outcome and may even have a detrimental effect. 113

Critique Although the study was carried out at a single centre, the authors have argued that it was a regional level I trauma unit for a whole state and so it is appropriate to generalize from their results. However, the power of the trial would undoubtedly be increased by the inclusion of more than one study centre. The use of a composite primary outcome is an interesting method to increase power of trial and the authors have been commended for this novel approach. It has been standard practice to dichotomize patient outcome into favourable or unfavourable on the basis of the GOSE score. Proponents of composite analysis argue that dichotomizing outcome may be insensitive. However, the authors of this trial have been criticized for using 39 outcome measures to be included in their composite analysis, which may be too many (Maas and Murray, 2007). Freemantle et al. have reviewed the arguments for and against the use of composite outcomes in clinical trials (Freemantle et al., 2005). One of the most crucial points that needs to be emphasized when interpreting the results of this trial is that it was restarted with a lower target MgSO4 concentration and that analysis of the two target levels was carried out separately. It has been questioned as to whether an analysis of the two groups together would have given rise to more drug-specific conclusions (Maas and Murray, 2007). As things stand, conclusions can only be made regarding specific dosing regimens. Nonetheless, the timing and target ranges used in the trial are consistent with those that have been widely used in the management of head injury. It is legitimate for the authors, therefore, to draw the conclusion that there is no evidence for a beneficial effect of standard methods of MgSO4 infusions in head injury. Following this trial there is currently no evidence to support the use of magnesium sulphate infusions as a neuroprotective measure in head injury. In addition, the novel use of a composite analysis may influence the design and outcome assessment of future head-injury trials.

References Freemantle N, Calvert M, Wood J, Eastaugh J, Griffin C. Composite outcomes in randomised trials: greater precision but with greater uncertainty. JAMA 2005; 289: 2554–2559. Maas AIR, Murray GD. Magnesium for neuroprotection after traumatic brain injury. Lancet Neurol 2007; 6: 20–21.

3.12 Epidemiology of Post-traumatic Seizures Details of Studies Two large population-based studies looking at the epidemiology of PTS stand out as landmarks in elucidating the epidemiology of post-traumatic seizures in TBI. The first study included 4541 people who suffered traumatic brain injuries over a 50-year period (1935–1984) in Olmsted County, Minnesota, USA (Annegers et al., 1998). The second study followed 78,572 people with TBI born over a 25-year period (1977–2002) in Denmark (Christensen et al., 2009).

Study References Main Studies Annegers JF, Hauser WA, Coan SP, Rocca WA. A population-based study of seizures after traumatic brain injuries. N Engl J Med 1998; 338: 20–24. Christensen J, Pedersen MG, Pedersen CB, Sidenius P, Olsen J, Vestergaard M. Long-term risk of epilepsy after traumatic brain injury in children and young adults: a population-based cohort study. Lancet 2009; 373: 1105–1110.

Related References Englander J, Bushnik T, Duong TT, Cifu DX, Zafonte R, Wright J, Hughes R, Bergman W. Analyzing risk factors for late posttraumatic seizures: a prospective, multicenter investigation. Arch Phys Med Rehabil 2003; 83: 365–373. Salazar AM, Jabbari B, Vance SC, Grafman J, Amin D, Dillon JD. Epilepsy after penetrating head injury. I. Clinical correlates: a report of the Vietnam Head Injury Study. Neurology 1985; 35: 1406–1414. Temkin NR. Risk factors for posttraumatic seizures in adults. Epilepsia 2003; 44: 18–20.

Study Design Annegers et al. (1998) Mild TBI

No skull fracture Loss of consciousness or post-traumatic amnesia <30 min

Moderate TBI

≥1 of the following: Loss of consciousness or post-traumatic amnesia >30 min but <24 h Skull fracture

Severe TBI

≥1 of the following: 114

Contusion Intracranial haematoma Loss of consciousness or post-traumatic amnesia >24 h

Christensen et al. (2009) This was a population-based study that used the Civil Registration System in Denmark to identify 1,605,216 people born between 1977 and 2002, which amounted to a total of 19,527,337 person-years. Relative risk (RR) of epilepsy was calculated for mild brain injury, severe brain injury, and skull fracture. Time since injury and variables of age, sex, and family history of epilepsy were considered. Definition of mild and severe brain injury were according to the American Congress of Rehabilitation Medicine: mild brain injury is manifest by altered brain function (loss of consciousness <30 min, GCS not <13, amnesia <24 h, confusion); severe brain injury includes contusion and intracranial haemorrhage.

Results Annegers et al. (1998) There was no increased risk of seizures for mild TBI after 5 years. Patients with moderate TBI retain a significantly increased risk of seizures for over 10 years. Patients with severe TBI retained a significantly increased risk of seizures for over 20 years. Strongest risk factors for PTS: brain contusions and subdural haematomas. Other risk factors for PTS: skull fracture and prolonged loss of consciousness. Severity of TBI Mild

Cumulative 5-year probability of seizure 0.7%

Standardized incidence ratio 1.5

Moderate

1.2%

2.9

Severe

10.0%

17.0

Christensen et al. (2009) RR of epilepsy increased with age, and in patients >15 years of age at time of injury RR was 3.51 for mild and 12.24 for severe head injury. RR of epilepsy was higher in women (2.49) compared to men (2.01). RR of epilepsy was also increased in those with a family history of epilepsy (5.75 mild and 10.09 severe head injury). Relative risk of epilepsy after head injury 2.22

Relative risk of epilepsy >10 years after head injury 1.51

Moderate head injury

7.40

4.29

Skull fracture

2.17

2.06

Mild head injury

Conclusions Annegers et al. (1998) The risk of PTS increases with the severity of the TBI and varies according to the time since the injury.

Christensen et al. (2009) Traumatic head injury is a long-lasting risk factor for epilepsy and there is a window for prevention of post-traumatic epilepsy.

Critique These two studies are the largest and best population-based studies concerning the epidemiology of post-traumatic epilepsy available. The study by Annegers et al. gave valuable data regarding the duration of risk of PTS and the relation of severity of injury to this increased risk. The authors also identified several factors, such as subdural haematoma and brain contusions, which increase the risk of PTS. However, the study by Christensen et al. found that the risk of epilepsy following TBI remains high for far longer than 5 years, and in their study was still present at >10 years following injury. However, this difference may reflect the wider inclusion criteria of the Danish study or the 115

smaller sample size of the US study. The Danish study has been criticized for not evaluating the type of seizures or the timing of the seizures, e.g. early or late (Shorvon and Neligan, 2009). Furthermore, there were no details regarding whether the dura was breached nor in those patients with skull fractures. Another criticism is that the data collection only included outpatient follow-up since 1995 and it is possible that many seizures will have been diagnosed in this setting. Nonetheless, both studies included here provide the best epidemiological data available for elucidating the risk of post-traumatic epilepsy.

Reference Shorvon S, Neligan A. Risk of epilepsy after head trauma. Lancet 2009; 373: 1060–1061.

3.13 Phenytoin for Prevention of Post-traumatic Seizures Details of Study This study (Temkin et al., 1990) was the first PRCT deemed to have sufficient power to evaluate the efficacy of phenytoin for the prophylaxis of PTS following serious head injury. The study was carried out in the mid-1980s at the Harborview Medical Center, Seattle, Washington, USA.

Study References Main Study Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn HR. A randomised, double-blind study of phenytoin for the prevention of post-traumatic seizures. N Engl J Med 1990; 323: 497–502.

Related References Annegers JF, Hauser WA, Coan SP, Rocca WA. A population-based study of seizures after traumatic brain injuries. N Engl J Med 1998; 338: 20–24. Schierhout G, Roberts I. Anti-epileptic drugs for preventing seizures following acute traumatic brain injury. Cochrane Database Syst Rev 2001; 4: CD000173.

Study Design Double-blind PRCT. Class of evidence

Randomization

Phenytoin versus placebo

Number of patients

404

Follow-up

Primary outcome: Early seizures (1st week) Secondary outcomes: Late seizures (1 week to 24 months) Percentage of patients followed up at each stage? (> 80%?)

Number of centres

Stratification

None

Patients were included if: presence of a severe head injury defined as one or more of: cortical contusion (visible on CT scan); subdural haematoma (SDH); EDH; ICH; depressed skull fracture; penetrating head injury; seizure <24 h from injury; GCS ≤10). Patients were excluded if they were <16 years old or there was a delay of >24 h before loading of drug. Patients were also excluded if there were other predisposing risk factors for seizures, e.g. previous severe head injury, history of severe alcoholism, or previous neurological conditions with risk of seizures. Initial loading dose of phenytoin was 20 mg/kg with maintenance dose adjusted according to serum levels. Phenytoin/placebo continued for 12 months and was then tapered off fully. Analysis was on an intention-to-treat basis.

Outcome Measures Primary Endpoints The occurrence of seizures: early (<1 week); or late (>1 week). Diagnosis of seizures by experienced clinician with the use of EEG recording if required. 116

Other Endpoint Analysed Possible adverse effects of phenytoin.

Results Phenytoin treatment reduced risk of seizures by 73% in the first week. Additional secondary analysis revealed no difference in the cumulative probability of all seizures (early or late) in both groups.

Conclusions Phenytoin has a beneficial effect in reducing the incidence of post-traumatic seizures only in the first week following severe head injury.

Critique Although previous RCTs had been carried out, this study may be regarded as the first PRCT with sufficient numbers of patients recruited to analyse the efficacy of phenytoin for the prophylaxis of post-traumatic seizures. Meta-analysis of all published trials supports the findings of this study. A Cochrane Review in 2001 identified six published RCTs looking at seizure prophylaxis for post-traumatic epilepsy and concluded that the use of anticonvulsants would keep one in ten patients seizure free in the first week following head injury but that there was no evidence for efficacy against late seizures (Schierhout and Roberts, 2001).

Reference Schierhout G, Roberts I. Anti-epileptic drugs for preventing seizures following acute traumatic brain injury. Cochrane Database Syst Rev 2001; 4: CD000173.

3.14 Pre-hospital Intubation for Traumatic Brain Injury Details of Study Pre-hospital intubation (PHI) of TBI patients is controversial. Although the main objective of PHI is to prevent secondary brain damage there have been concerns that complications of PHI in the field can exacerbate secondary brain injury. This landmark study by Bernard et al. (2012) is the first prospective RCT to evaluate rapid sequence intubation (RSI) by paramedics in the pre-hospital setting of adult patients with severe TBI. The trial was carried out in Victoria, Australia between 2004 and 2008.

Study References Main Study Bernard S, Nguyen V, Cameron P, Masci K, Fitzgerald M, Cooper DJ, Walker T, Myles P, Murray L, Taylor D, Smith K, Patrick I, Edington J, Bacon A, Rosenfeld JV, Judson R. Prehospital rapid sequence induction improves functional outcome for patients with severe traumatic brain injury. A randomized controlled trial. Ann Surg 2012; 252: 959â&#x20AC;&#x201C;965.

Related Reference Bernard S. Paramedic intubation of patients with severe traumatic brain injury: a review of current Australian practice and recommendations for change. Emerg Med Australas 2006; 18: 221â&#x20AC;&#x201C;228.

Study Design Prospective RCT. Class of evidence

Randomization

Pre-hospital intubation by paramedics versus in-hospital intubation by physicians

Number of patients

312

Follow-up

6 months Primary outcome: 117

GOSE Secondary outcomes: Favourable versus unfavourable outcome Length of ICU stay Survival to hospital discharge Number of centres

Stratification Inclusion criteria: evidence of severe head trauma at scene; GCS <10; age ≥15 years; intact airway reflexes. Exclusion criteria: trauma hospital accessible <10 min from scene; no intravenous access; allergy to relevant anaesthetic agents. Anaesthetic agents used for RSI included fentanyl, midazolam, and succinylcholine and capnograph waveforms were used to assess endotracheal tube placement. Patients randomized to in-hospital intubation, high-flow supplemental oxygen, and bag/mask ventilation if required. Assessor of the GOSE at 6 months were blinded to the treatment allocation. Favourable outcome was defined as a GOSE of 1–4 and unfavourable outcome was defined as a GOSE of 5–8. Analysis was done on an intention-to-treat basis.

Results One hundred and sixty patients were randomized to the paramedic RSI (98% follow-up at 6 months) and 152 to hospital intubation (93% follow-up at 6 months). There were no differences found between the two treatments in any of the other secondary outcomes.

Conclusions Pre-hospital RSI by paramedics increases the favourable neurological outcome at 6 months.

Critique The prevention of secondary brain injury in patients with severe TBI can result from hypoxia, hypocapnia, and cardiovascular compromise. Although early endotracheal intubation in the hospital setting is standard practice, the role of intubation in the pre-hospital setting is more controversial. This is because of questions regarding the success rate of RSI in the field compared to in-hospital settings. Failure to intubate may exacerbate hypoxia and the use of sedative agents may result in cardiovascular compromise in inadequately resuscitated patients. The study by Bernard et al. is, therefore, a landmark study as it is the first prospective RCT to address this question. The authors acknowledged several limitations of their study including the difference in follow-up rates between the two groups, and even the addition of one more patient to either group may have negated the statistical significance of the findings. Nonetheless, this was a well-designed study which did not show any increase in mortality from pre-hospital intubation. However, because patients within 10 min reach of a trauma hospital were excluded, it has been suggested that the conclusions may not be applicable to urban settings (Yeh and Velmahos, 2012).

Reference Yeh DD, Velmahos GC. Prehospital intubation for traumatic brain injury: do it correctly or not at all. ANZ J Surg 2012; 82: 484–488.

118

Chapter 4

Spinal surgery RD Johnson, WA Liebenberg, N Maartens, G Barbagallo, M Balsano 4.0 Introduction 4.1 Steroid use in acute spinal cord injury 4.2 Steroid use in metastatic spinal cord compression 4.3 Timing of surgery for acute spinal cord injury 4.4 Decompressive surgery for spinal metastasis 4.5 Surgery for lumbar disc herniation 4.6 Microscopic sequestrectomy for lumbar disc herniation 4.7 Surgery for cauda equina syndrome 4.8 Surgery for lumbar stenosis 4.9 Spinal stabilization for chronic back pain 4.10 Surgery for cervical spondylotic myelopathy 4.11 Surgery for cervical radiculopathy

4.0 Introduction The field of spinal surgery is shared by both neurosurgeons and spinal orthopaedic surgeons. Over the last few decades improved spinal imaging modalities have become widely available in the form of computed tomography (CT), magnetic resonance imaging (MRI) and radio-isotope scans (RI). This has allowed for better correlation of clinical, radiological, and surgical findings. The classic dilemmas and controversies that have faced spinal surgery in the past are, therefore, being revisited in this era of neuroimaging. Searching an online database with the terms ‘spinal surgery’ and ‘clinical trial’ will bring up hundreds, if not thousands, of results. This also reflects the advent of a myriad of new spinal adjuncts and prostheses that are making their way into the spinal surgeon’s tool box. In this chapter, we have endeavoured to steer away from studies that compare spinal prosthetics or fusion devices and concentrate rather on the core aspects of spinal practice that is relevant to most neurosurgeons. We have, therefore, included studies that address the role of decompressive surgery in the management of spine or nerve root compression whether this may be from trauma, degenerative disease, or tumours. This approach has allowed us to include trials looking at the role of steroids in the management of spinal cord compression secondary to trauma and metastatic disease (Bracken et al., 1984; Bracken et al., 1985; Vecht et al., 1989; Bracken et al., 1990; Bracken et al., 1997). In addition, we have included studies on the role of surgical decompression in the management of cord compression secondary to trauma and metastatic disease (Vaccaro et al., 1997; Patchell et al., 2005). Several studies have been included that address the role of surgery in lumbar disc prolapse including the results of the recently reported Spine Patient Outcomes Research Trial (SPORT) in the United States (Weber, 1983; Weinstein, et al., 2006; Barth et al., 2008a; Barth et al., 2008b; Peul et al., 2008). Included here is a section on the timing of surgery for cauda equina syndrome (CES) (O’Laoire et al., 1981). This is a highly contentious area and although there is no randomized trial there are two meta-analyses that have provoked considerable discussion and debate in the literature (Ahn et al., 2000; Todd, 2005). The role of surgery in the management of lumbar stenosis was also examined as part of the SPORT trial and so we have included it here for completion (Weinstein et al., 2008). This large trial demonstrates many of the difficulties in conducting trials in spinal patients. The Medical Research Council (MRC) trial on the role of spinal stabilization surgery in chronic back pain has also been included in this chapter (Fairbank et al., 2005). There are numerous case series and evidence-based reviews of different aspects of cervical spine surgery including randomized studies evaluating the efficacy of different spinal implants. However, we have chosen to include here two prospective randomized studies that compare surgery with conservative measures for the treatment of cervical myelopathy and cervical radiculopathy. Global assessment of sagittal balance is becoming an integral part of the assessment of patients undergoing surgery for degenerate spinal conditions. The introduction of the concept of pelvic parameters of sagittal balance has been a paradigm shift in the field of spinal surgery (Johnson et al., 2013). It has not been possible to select one paper, or indeed only a few papers, for inclusion in this volume as the concept of sagittal balance has developed over a number of years. However, for the neurosurgeon interested in the spine, these papers form a series of landmark papers that are essential reading. We would, therefore, point the interested reader to two papers by Gelb and Van Royen, respectively, describing studies of the normal spine that indicate that sagittal balance is achieved when the C7 plumb line (a line passing vertically from the centre of the C7 vertebral body) lies posterior to the gravity line, i.e. posterior 119

to the centre of the femoral heads (Gelb et al., 1995; Van Royen et al., 1998). The landmark studies that have been so influential in establishing pelvic parameters as a concept in spinal surgery span a period of more than 20 years (During et al., 1985; Jackson et al., 1998; Legaye and Duval-Beaupère, 2005). There is now an increasing body of evidence to suggest pelvic indices may affect the natural history and outcomes from surgical intervention. It would appear, therefore, that planning surgery to maintain pelvic parameters within as normal limits as possible is likely to improve outcomes.

References Ahn UM, Ahn NU, Buchowski MS, Garrett ES, Sieber AN, Kostuik JP. Cauda equina syndrome secondary to lumbar disc herniation. A meta-analysis of surgical outcomes. Spine 2000; 25: 1515–1522. Barth M, Diepers M, Weiss C, Thomé C. Two-year outcome after lumbar microdiscectomy versus microscopic sequestrectomy: part1: evaluation of clinical outcome. Spine 2008a; 33: 265–272. Barth M, Diepers M, Weiss C, Thomé C. Two-year outcome after lumbar microdiscectomy versus microscopic sequestrectomy: part 2: radiographic evaluation and correlation with clinical outcome. Spine 2008b; 33: 273–279. Bracken MB, Collings WF, Freeman DF, Shepard MJ, Wagner FW, Silten RM, Hellenbrand KG, Ransohoff J, Hunt WE, Perot PL Jr, Grossman RG, Green BA, Eisenberg HM, Rifkinson N, Goodman JH, Meagher JN, Fischer B, Clifton GL, Flamm ES, Rawe SE. Efficacy of methylprednisolone in acute spinal cord injury. JAMA 1984; 251: 45–52. Bracken MB, Shepard MJ, Hellenbrand KG, Collins WF, Leo LS, Freeman DF, Wagner FC, Flamm ES, Eisenberg HM, Goodman JH, Perot PL Jr, Green BA, Grossman RG, Meagher JN, Young W, Fischer B, Clifton GL, Hunt WE, Rifkinson N. Methylprednisolone and neurological function 1 year after spinal cord injury. Results of the National Acute Spinal Cord Injury Study. J Neurosurg 1985; 63: 704–713. Bracken MB, Shepard MJ, Collins WF, Holford TR, Young W, Baskin DS, Eisenberg HM, Flamm E, Leo-Summers L, Maroon J, Marshall LF, Perot PL Jr, Piepmeier J, Sonntag VKH, Wagner FC, Wilberger JE, Winn HR. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med 1990; 322: 1405–1411. Bracken MB, Shepard MJ, Holford TR, Leo-Summers L, Aldrich EF, Fazl M, Fehlings M, Herr DL, Hitchon PW, Marshall LF, Nockels RP, Pascale V, Perot PL Jr, Piepmeier J, Sonntag VK, Wagner F, Wilberger JE, Winn HR, Young W. Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 1997; 277: 1597–1604. During J, Goudfrooij H, Keessen W, Beeker TW, Crowe A. Toward standards for posture. Postural characteristics of the lower back system in normal and pathologic conditions. Spine (Phila Pa 1976) 1985; 10: 83–87. Fairbank J, Frost H, Wilson-MacDonald J, Yu LM, Barker K, Collins R for the Spine Stabilisation Trial Group. Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial. BMJ 2005; 330: 1233–1240. Gelb DE, Lenke LG, Bridwell KH, Blanke K, McEnery KW. An analysis of sagittal spinal alignment in 100 asymptomatic middle and older aged volunteers. Spine (Phila Pa 1976) 1995; 20:1351–1358. Jackson RP, Peterson MD, McManus AC, Hales C. Compensatory spinopelvic balance over the hip axis and better reliability in measuring lordosis to the pelvic radius on standing lateral radiographs of adult volunteers and patients. Spine (Phila Pa 1976) 1998; 23:1750–1767. Johnson RD, Valore A, Villaminar A, Comisso M, Balsano M. Sagittal balance and pelvic parameters – a paradigm shift in spinal surgery. J Clin Neurosci 2013; 20: 191–196. Legaye J, Duval-Beaupère G. Sagittal plane alignment of the spine and gravity: a radiological and clinical evaluation. Acta Orthop Belg 2005; 71: 213–220. O’Laoire SA, Crockard HA, Thomas DG. Prognosis for sphincter recovery after operation for cauda equina compression owing to lumbar disc prolapse. BMJ 1981; 282: 1852–1854. Patchell RA, Tibbs PA, Regine WF, Saris S, Kryscio RJ, Mohiuddin M, Young B. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer. Lancet 2005; 366: 643–648. Peul WC, van den Hout WB, Brand R, Thomeer RTWM, Koes BW, for the Leiden–The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised controlled trial. BMJ 2008; 336: 1355–1358.

SPORT trial—several things looked at but reported separately, so covered here in its separate entities. Todd NV. Cauda equina syndrome: the timing of surgery probably does influence outcome. Br J Neurosurg 2005; 19: 301–306. Vaccaro AR, Daugherty RJ, Sheehan J, Sheehan TP, Dante SJ, Cotle JM, Baderston RA, Herbison GJ, Northup BE. Neurologic outcome of early versus later surgery for cervical cord injury. Spine 1997; 22: 2609–2613. Van Royen BJ, Toussaint HM, Kingma I, Bot SD, Caspers M, Harlaar J, Wuisman PI. Accuracy of the sagittal vertical axis in a standing lateral radiograph as a measurement of balance in spinal deformities. Eur Spine J 1998; 7: 408–412. Vecht CJ, Haaxma-Reiche H, van Putten WL, de Visser M, Vries EP, Twiijnstra A. Initial bolus of conventional versus high-dose dexamethasone in metastatic spinal cord compression. Neurology 1989; 39: 1255–1257. Weber H. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine 1983; 8: 131–140. Weinstein JN, Tosteson TD, Lurie JD, Tosteson ANA, Hnascom B, Skinner JS, Abdu WA, Hilibrand AS, Boden SD, Deyo RA. Surgical vs nonoperative treatment of lumbar disk herniation: the spine patient outcomes research trial (SPORT): a randomised trial. JAMA 2006; 296: 2441–2450.

120

Weinstein JN, Lurie JD, Tosteson TD, Hanscom B, Tosteson ANA, Blood EA, Birkmeyer NJO, Hilibrand AS, Herkowitz H, Cammisa FP, Todd JA, Emery SE, Lenke LG, Abdu WA, Longley M, Errico TJ, Hu SS. Surgery versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 2007; 356: 2257–2270. Weinstein JN, Tosteson TD, Lurie JD, Tosteson ANA, Blood E, Hanscom B, Herkwoitz H, Cammisa F, Albert T, Boden SD, Hilibrand A, Goldberg H, Berven S, An H for the SPORT Investigators. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008; 358: 794–810.

4.1 Steroid Use in Acute Spinal Cord Injury Details of Study The National Acute Spinal Cord Injury Study (NASCIS) is the largest study investigating the effects of the steroid methyl prednisolone (MePred) in acute spinal cord injury. There have been three parts to the study that are referred to as NASCIS I, NASCIS II, and NASCIS III. These studies were carried out in the United States in the 1980s and 1990s.

Study References Main Study There are four main references for the NASCIS study: Bracken MB, Collings WF, Freeman DF, Shepard MJ, Wagner FW, Silten RM, Hellenbrand KG, Ransohoff J, Hunt WE, Perot PL Jr, Grossman RG, Green BA, Eisenberg HM, Rifkinson N, Goodman JH, Meagher JN, Fischer B, Clifton GL, Flamm ES, Rawe SE. Efficacy of methylprednisolone in acute spinal cord injury. JAMA 1984; 251: 45–52. Bracken MB, Shepard MJ, Hellenbrand KG, Collins WF, Leo LS, Freeman DF, Wagner FC, Flamm ES, Eisenberg HM, Goodman JH, Perot PL Jr, Green BA, Grossman RG, Meagher JN, Young W, Fischer B, Clifton GL, Hunt WE, Rifkinson N. Methylprednisolone and neurological function 1 year after spinal cord injury. Results of the National Acute Spinal Cord Injury Study. J Neurosurg 1985; 63: 704–713. Bracken MB, Shepard MJ, Collins WF, Holford TR, Young W, Baskin DS, Eisenberg HM, Flamm E, Leo-Summers L, Maroon J, Marshall LF, Perot PL Jr, Piepmeier J, Sonntag VKH, Wagner FC, Wilberger JE, Winn HR. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med 1990; 322: 1405–1411. Bracken MB, Shepard MJ, Holford TR, Leo-Summers L, Aldrich EF, Fazl M, Fehlings M, Herr DL, Hitchon PW, Marshall LF, Nockels RP, Pascale V, Perot PL Jr, Piepmeier J, Sonntag VK, Wagner F, Wilberger JE, Winn HR, Young W. Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 1997; 277: 1597–1604.

Related References Bracken MB. Steroids for acute spinal cord injury. Cochrane Database Syst Rev 2002; 3: CD001046. Otani K, Abe H, Kadoya K, Nagakawa H, Ikata T, Tominagu S. Beneficial effect of methylprednisolone sodium citrate in the treatment of acute spinal cord injury. Sekitui Skeizui J 1994; 7: 633–647. Petijean ME, Pontillart V, Dixmarias F, Wiart K, Sztark F, Lassie P, Thicoipe M, Dabadie P. Traitement medicamenteux de la lesion medullaire traumatique au stade aigu. Ann Fr Aneth Reanim 1998; 17: 115–122.

Study Design Multi-centre, blinded, PRCTs.

121

Dosing NASCIS I: low-dose regimen—loading dose of MePred was 100 mg followed by 25 mg every 6 h for 10 days; moderate-dose regimen—1000 mg bolus followed by 250 mg every 6 h for 10 days. NASCIS II: MePred was given as an intravenous bolus of 30 mg/kg followed by 5.4 mg/kg for 23 h. NASCIS III: MePred bolus and maintenance infusions given as per NASCIS II except continued for further 23 or 47 h.

Timing of Bolus NASCIS II looked at early (<8 h from injury) versus late (>8 h) administration of MePred. NASCIS III also looked at ultra-early (<3 h) versus early (3–8 h) administration of MePred.

Outcome Measures Primary Endpoints Neurological status was measured on continuous numerical scales in a blinded manner, e.g. muscle function was measured in 14 muscle segments on a 6-point scale between 0 and 5 (total score 70).

Results NASCIS I There was no difference between moderate-dose and low-dose MePred. There was a trend towards better outcome for moderate-dose MePred if given within 8 h.

NASCIS II Patients who received MePred within 8 h of injury had a statistically significant improvement in motor and sensory function. There was no effect of naloxone.

NASCIS III No statistically significant benefit was seen for continuing MePred treatment for 48 h. No statistically significant benefit was seen for ultra-early administration of MePred.

Conclusions MePred improves outcome of acute spinal cord injury (ASCI) if given within 8 h of injury.

Critique The positive results of the NASCIS trials pertain only to post hoc subgroup analyses. For example, NASCIS II only showed a benefit with MePred for the subgroup of patients who received it within 8 h of injury. It is imperative, 122

therefore, to re-emphasize that the conclusions from this subgroup analysis cannot be extended to all patients within the trial. The large range of the neurological scores used has meant that it is questionable whether small improvements are clinically relevant (Spencer and Bazarian, 2003). Spinal cord injury is a devastating disorder with approximately 10% mortality and high rates of severe disability. NASCIS I is probably the first clinical trial of a therapeutic intervention in ASCI. Together the three NASCIS trials form the largest study to date looking at the effects of steroids in ASCI. However, the results remain controversial and there is as yet no guideline or recommendations regarding the use of steroids in ASCI (Coleman et al., 2000; Hubert, 2000). The use of MePred is, therefore, still a treatment choice available to the managing surgeon.

References Coleman WP, Benzel D, Cahill DW, Ducker T, Geisler F, Green B, Gropper MR, Goffin J, Madsen PW 3rd, Maiman DJ, Ondra SL, Rosner M, Sasso RC, Trost GR, Zeidman S. A critical appraisal of the reporting of the National Acute Spinal Cord Injury Studies (II and III) of methylprednisolone in acute spinal cord injury. J Spinal Disord 2000; 13: 185–199. Hubert RJ. Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. J Neurosurg 2000; 93: S1–S7. Spencer MT, Bazarian JJ. Are corticosteroids effective in traumatic spinal cord injury? Ann Emerg Med 2003; 41: 410–413.

4.2 Steroid Use in Metastatic Spinal Cord Compression Details of Study A blinded randomized controlled trial of high-dose dexamethasone as an adjunct to radiotherapy in patients with metastatic spinal cord compression (MESCC) from solid tumours was carried out between 1987 and 1989 in Rigshospitalet, Copenhagen, Denmark (Sorensen et al., 1994).

Study References Main Study Sorensen PS, Helwig-Larson S, Mouridesen H, Hansen HH. Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: a randomized trial. Eur J Cancer 1994; 30A: 22–27.

Related Reference Sciubba DM, Gokaslan ZL. Diagnosis and management of metastatic spine disease. Surgical Oncology 2006; 15: 141–151.

Study Design A single-blind PRCT. Class of evidence

Randomization

High-dose dexamethasone versus no steroids

Number of patients

Follow-up

6 months Primary outcomes: Preservation or return of gait function Secondary outcomes: Side effects Survival

Number of centres

Stratification

Primary tumour Gait function

Dexamethasone was administered as a bolus of 96 mg IV followed by 96 mg PO for 3 days. Inclusion criteria: clinical and radiological evidence of MESCC. Exclusion criteria: lymphoma patients; surgical decompression; previous epidural metastases; meningeal carcinomatosis; peptic ulcers. Analysis was done on an intention-to-treat basis.

Outcomes Primary Outcomes Gait function. 123

Successful treatment was defined as walking ability retained (ambulatory patients) or walking ability regained (non-ambulatory patients). The same neurologist assessed all patients at 3 weeks after treatment and then 3-monthly until 2 years, or until patient deceased.

Results Follow-up 2 years (or until death).

There was no difference in survival between the two groups. Eleven per cent of those receiving steroids experienced side effects. A subgroup analysis was carried out in patients with breast cancer, which showed that 94% of patients receiving dexamethasone achieved a successful result compared to 69% receiving no steroids. This result was not statistically significant.

Conclusion Steroids should be administered routinely to all patients with MESCC.

Critique MESCC is one of the most devastating and dreaded complications of cancer. Rapid neurological deterioration can result in paralysis and loss of sphincter function. The early diagnosis and treatment of this condition involves close cooperation between oncologists and neurosurgeons. Steroids are part of the treatment armamentarium available to try and reduce the incidence of severe neurological deficits in this condition. This study by Sorensen et al. was the first randomized trial to examine the efficacy of steroids for MESCC. Prior to 1993 there was only a plethora of anecdotal reports regarding the benefits of steroids in this condition. The first report published in the late 1960s was of two patients with disseminated pelvic malignancies whose paraparesis improved following the administration of MePred (Cantu, 1968). Clarke and Saunders reported improvement in limb neurology in two children who were given steroids for a presumptive diagnosis of Guillain–Barré syndrome: both patients were subsequently found to have malignant cord compression (Clarke and Saunders, 1975). These observations of a beneficial effect of steroids were confirmed in animal models of malignant cord compression (Ushio et al., 1977). This led to the development of dosing regimens and protocols for the use in patients (Gilbert et al., 1978; Greenberg et al., 1980). The study by Sorenson et al. followed on from this work. Although the benefit reached only borderline statistical significance the authors concluded that steroids should be used as an adjunct in malignant cord compression. This study established the role of steroids in this devastating condition.

References Cantu RC. Corticosteroids for spinal metastases. Lancet 1968; 2: 912. Clarke PR, Saunders M. Steroid-induced remission in spinal canal reticulum cell sarcoma. Report of two cases. J Neurosurg 1975; 42: 346–413. Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978; 3: 40–51. Greenberg HS, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: results with a new treatment protocol. Ann Neurol 1980; 8: 361–366. Ushio Y, Posner R, Posner JB, Shapiro WR. Experimental spinal cord compression by epidural neoplasm. Neurology 1977; 27: 422–429.

4.3 Timing of Surgery for Acute Spinal Cord Injury Details of Study Most studies on the timing of surgery in ASCI are either retrospective or prospective case series. This study undertaken at the Regional Spinal Cord Injury Center of Delaware Valley between 1992 and 1995 is the only attempt at a RCT of surgery for ASCI.

Study References Main Study 124

Vaccaro AR, Daugherty RJ, Sheehan J, Sheehan TP, Dante SJ, Cotle JM, Balsderston, RA, Herbison GJ, Northup BE. Neurologic outcome of early versus later surgery for cervical cord injury. Spine 1997; 22: 2609–2613.

Related References Bagnall AM, Jones L, Duffy S, Riemsma RP. Spinal fixation surgery for acute traumatic spinal cord injury. Cochrane Database Syst Rev 2008; 1: CD004725. Fehlings MG, Perrin RG. The role and timing of early decompression for cervical spinal cord injury: update with a review of recent clinical evidence. Injury 2005; 36: SB13–SB26. Fehlings MG, Perrin RG. The timing of surgical intervention in the treatment of spinal cord injury: a systematic review of recent clinical evidence. Spine 2006; 31: 528–535. Tator CH, Fehlings MG, Thorpe K, Taylor W. Current use and timing of spinal surgery for management of acute spinal cord injury in North America: results of a retrospective multicenter study. J Neurosurg 1999; 91: 12–18.

Study Design PRCT. Class of evidence

Randomization

Early surgery versus late surgery for cervical spinal cord trauma

Number of patients

64 randomized

Follow-up

<1 year Primary outcomes: Neurological outcome Functional outcome Secondary outcomes: Length of hospital stay

Number of centres

Stratification

Age and sex

Inclusion criteria: age 15–75 years; neurological impairment A–D on American Spinal Injury Association (ASIA) scale; neurological level C3–T1; admission within 48 h of injury; radiological evidence of cord compression. Exclusion criteria: other injuries preventing neurological evaluation or surgery; coexisting spinal cord disease; worsening neurology due to blood, disc, or bony fragments within the canal. Early surgery was <72 h from injury. Late surgery was >5 days from injury. Surgery included decompression ± stabilization procedures. Neurological outcome was assessed by comparing standard neurological examination before (on admission) and after surgery (mean 300 days).

Results Mean time to surgery was 1.8 days in the early group and 16.8 days in the late group. There were no significant differences in the neurological or functional outcomes between the two groups, or in the length of hospital stay.

Conclusions There is no benefit between surgery within 72 h of injury and delayed surgery in cervical spinal cord injury.

Critique The study by Vaccaro et al. includes only cases of cervical cord injury and the length of time to early surgery (mean 1.8 days) may not be early enough. It is still possible that there may be a benefit of earlier surgery within 8 or 12 h of injury. Tator et al. reported one of the largest case series in the literature looking at the effect of timing of surgery on outcome in ASCI at all spinal levels (Tator et al., 1999). They conducted a retrospective analysis of over 500 cases of ASCI admitted to 36 centres in North America over a period of 9 months. The results suggested that there is no agreement on the timing of surgery for ASCI and that further RCTs are needed. To date, the study by Vaccaro et al. remains the only attempt at a randomized trial. Fehlings and Perrin have published several comprehensive reviews of the literature on the timing of surgery in ASCI (Fehlings and Perrin, 2005, 2006). On the basis of the published data, 125

they have made recommendations regarding the timing of surgery in ASCI. However, they emphasize that with the lack of definitive evidence urgent decompression remains only a reasonable practice option that can be carried out safely.

References Fehlings MG, Perrin RG. The role and timing of early decompression for cervical spinal cord injury: update with a review of recent clinical evidence. Injury 2005; 36: SB13–SB26. Fehlings MG, Perrin RG. The timing of surgical intervention in the treatment of spinal cord injury: a systematic review of recent clinical evidence. Spine 2006; 31: 528–535. Tator CH, Fehlings MG, Thorpe K, Taylor W. Current use and timing of spinal surgery for management of acute spinal cord injury in North America: results of a retrospective multicenter study. J Neurosurg 1999; 91: 12–18.

4.4 Decompressive Surgery for Spinal Metastasis Details of Study This multi-institutional study is the largest randomized study looking at the role of decompressive surgery in the management of metastatic spinal cord compression (MESCC). The study was carried out by the Bluegrass NeuroOncology Consortium in the United States.

Study References Main Study Patchell RA, Tibbs PA, Regine WF, Payne R, Saris S, Kryscio RJ, Mohiuddin M, Young B. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet 2005; 366: 643– 648.

Related Reference Ibrahim A, Crockard A, Antonietti P, Boriani S, Bunger C, Gasbarrini A, Grejes A, Harms J, Kawahara N, Mazel C, Melcher R, Tomita K. Does spinal surgery improve quality of life for those with extradural (spinal) osseus metastases? An international multicenter prospective observational study. J Neurosurg Spine 2008; 8: 271–278.

Study Design Multi-institutional, randomized trial. Class of evidence I Randomization

Surgery plus radiotherapy versus radiotherapy alone

Number of patients

101 randomized

Follow-up

Period of time not specified, but monthly assessment of all patients Primary outcomes: Ability to walk Secondary outcomes: Urinary continence, muscle strength and functional status, need for steroids and opiates, survival time

Number of centres

Stratification

Tumour type Spinal stability Ambulatory status Institution

MESCC was defined radiologically as displacement of the spinal cord by an epidural mass. Inclusion criteria: age >18 years; at least one neurological sign; tissue diagnosis of non-CNS tumour; prognosis >3 months. Exclusion criteria: paraplegia >48 h; radiosensitive tumour (lymphomas, leukaemia, multiple myeloma, germ cell tumour); previous MESCC. Patients received treatment within 24 h of randomization. Radiation dose was 3.0 Gy × 10 fractions. Surgery was to decompress the spinal cord but no constraints were placed on technique or methods of fixation. 126

Surgical patients received radiotherapy within 14 days of surgery.

Outcome Measures Primary Endpoint Ambulatory status: ‘ambulant’ was defined as taking two steps with each foot unassisted (cane or walker allowed).

Secondary Endpoints Urinary continence. Functional status: Frankel functional scale score. Muscle strength: ASIA motor score. Steroid use: calculation of mean daily doses.

Results The trial was stopped early as interim analysis suggested that surgical therapy was superior.

Patients in the surgery group also did significantly better in all secondary outcomes (continence, functional scores, muscle strength, and less steroid use).

Conclusions Surgical decompression with radiotherapy is superior to radiotherapy alone in MESCC.

Critique MESCC is a significant problem and approximately 30% of cancer patients will develop symptomatic MESCC (Sciubba and Gokaslan, 2006). Although this disorder does not alter life expectancy, resultant neurological deficit significantly affects quality of life. This study does appear to show a benefit for surgery, although no direct comparison of radiotherapy alone. Several criticisms have been made regarding this study (Ibrahim et al., 2008). Firstly, there may have been selection bias in this study as the inclusion criteria were quite narrow. Patients with a prognosis of <3 months or who had been paraplegic for >48 h were excluded. Secondly, recruitment rates to the study were also low (approximately 1 patient per year over 10 years). Thirdly, the definition of ‘ambulant’ as ‘2 steps with each foot’ is highly questionable. Early studies looking at the role of laminectomy in MESCC had suggested that surgery may be associated with a poor outcome (Findlay, 1984). Radiotherapy alone has, therefore, become an accepted treatment regimen for MESCC. However, methods of spinal fixation have significantly improved over recent years and it is now possible to undertake decompressive surgery in cases that were previously not deemed to be surgical candidates. This study by Patchell and colleagues has been highly significant in re-establishing the role of surgery in MESCC. Further studies are already beginning to follow to look at this complex clinical problem. One particularly good multi-centre observational study suggests that surgical decompression for MESCC is associated with an improved quality of life (Ibrahim et al., 2008).

References Findlay GF. Adverse effects of the management of malignant spinal cord compression. J Neurol Neurosurg Psychiatry 1984; 47: 761–768. Ibrahim A, Crockard A, Antonietti P, Boriani S, Bunger C, Gasbarrini A, Grejes A, Harms J, Kawahara N, Mazel C, Melcher R, Tomita K. Does spinal surgery improve quality of life for those with extradural (spinal) osseus metastases? An international mutlicenter prospective observational study. J Neurosurg Spine 2008; 8: 271–278. Sciubba DM, Gokaslan ZL. Diagnosis and management of metastatic spine disease. Surgical Oncology 2006; 15: 141–151.

127

4.5 Surgery for Lumbar Disc Herniation Details of Studies There are three landmark trials that address the issue of surgical intervention for lumbar disc herniation. The first trial influenced practice for over 30 years (Weber, 1983). This was a single-centre study that followed up a randomized cohort of patients in which there was equipoise regarding surgical intervention and an observational cohort of patients in which it was felt there was no equipoise. The study took place in the Ullevall Hospital, Oslo, Norway, over a 10year period between the 1970s and 1980s. The second study was the SPORT (Weinstein et al., 2006). SPORT was conducted in the United States between 2000 and 2004 and examined surgery versus conservative management for lumbar disc prolapse. The third was carried out by the Leiden–The Hague Spine Intervention Prognostic Study Group (Peul et al., 2007; Peul et al., 2008). This study looked at early surgery versus prolonged conservative management in the management of sciatica due to lumbar disc herniation and was carried out in the Netherlands between 2002 and 2005.

Study References Main Studies Norwegian Study (Weber, 1983) Weber H. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine 1983; 8: 131–140.

US Study: SPORT (Weinstein et al., 2006) Weinstein JN, Tosteson TD, Lurie JD, Tosteson ANA, Hnascom B, Skinner JS, Abdu WA, Hilibrand AS, Boden SD, Deyo RA. Surgical vs nonoperative treatment of lumbar disk herniation: the spine patient outcomes research trial (SPORT): a randomised trial. JAMA 2006; 296: 2441–2450.

Netherlands Study (Peul et al., 2007; Peul et al., 2008) Peul WC, van Houwelingen HC, van den Hout WB, Brand R, Eekof JA, Tans JT, Thomeer RTWM, Koes BW for the Leiden—The Hague Spine Intervention Prognostic Study Group. Surgery versus prolonged conservative treatment for sciatica. N Engl J Med 2007; 356: 2245–2256. Peul WC, van den Hout WB, Brand R, Thomeer RTWM, Koes BW, for the Leiden–The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised controlled trial. BMJ 2008; 336: 1355–1358.

Related References Fairbank J. Prolapsed intervertebral disc. BMJ 2008; 336: 1317–1318. Gibson J, Grant I, Waddell G. The Cochrane review of surgery for lumbar disc prolapse and degenerative lumbar spondylosis. Spine 1999; 24: 1820–1832. Gibson JN, Waddell G. Surgical interventions for lumbar disc prolapse: updated Cochrane Review. Spine 2007; 32: 1735–1747. Van den Hout WB, Peul WC, Koes BVW, Brand R, Klevit J, Thomeer RTWM for the Leiden–The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica from lumbar disc herniation: cost utility analysis alongside a randomised controlled trial. BMJ 2008; 336: 1351–1354. Weinstein JN, Lurie JD, Tosteson TD, Hanscom B, Tosteson ANA, Blood EA, Birkmeyer NJO, Hilibrand AS, Herkowitz H, Cammisa FP, Todd JA, Emery SE, Lenke LG, Abdu WA, Longley M, Errico TJ, Hu SS. Surgery versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 2007; 356: 2257–2270. Weinstein JN, Tosteson TD, Lurie JD, Tosteson ANA, Blood E, Hanscom B, Herkwoitz H, Cammisa F, Albert T, Boden SD, Hilibrand A, Goldberg H, Berven S, An H for the SPORT Investigators. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008; 358: 794–810.

Study Designs All three studies involved randomization, but only the US and Netherlands studies were PRCT.

128

Inclusion and Exclusion Criteria

In the Norwegian study, uncertainty regarding the indication for surgery triggered randomization. Surgery was standard open discectomy in all three studies. Surgery was performed within 2 weeks of randomization in the Netherlands study and was therefore early surgery. Analysis on an intention-to-treat basis for the US and Netherlands studies.

Outcome Measures In the Norwegian study the categorization of outcomes into four groups (good, fair, poor, bad) was based on an appraisal by the examining doctors at follow-up.

129

Results Norwegian Study

There was a statistically significant benefit of surgery at 1 year (p = 0.0015). Twenty-six per cent of patients in the conservatively managed group were operated on during the first year. The results of surgery were statistically better whether or not these 26% of patients were included in the analysis. Beyond 4 years, although there was a trend towards benefit, the difference was not statistically significant.

US Study (SPORT)

Ninety-four per cent follow-up at 1 year, but follow-up at 2 years was <80%. There was a statistically significant improvement of Sciatica Botherness Index (SBI) in the surgery group at 1 year compared to the conservatively managed group (p = 0.003). Self-rated progress favoured surgery (p = 0.4). There was significant cross-over between groups with: 40% of the surgical group; 45% of the conservatively managed group. An as-treated analysis indicated a benefit of surgery.

Netherlands Study

Median time to recovery

Surgery

Conservative management

4.0 weeks (95% CI, 3.7â&#x20AC;&#x201C;4.3)

12.11 weeks (95% CI, 9.5â&#x20AC;&#x201C;14.8)

Follow-up was 99% at 1 year and 77% at 2 years. Early surgery produced a 17.7% better relief of leg pain compared to conservative treatment at 8 weeks. Inverse Kaplanâ&#x20AC;&#x201C;Meier curves were used to estimate the cumulative incidence of recovery: the hazard ratio was 130

1.97% in favour of early surgery (95% CI, 1.72–2.22). The short-term benefit of early surgery ceased to be statistically significant by 6 months.

Conclusions Norwegian Study Surgical treatment was better than conservative management at 1-year follow-up, but this difference became less pronounced over a 10-year period.

US Study (SPORT) Netherlands Study Early surgery for sciatica due to lumbar disc prolapse leads to faster recovery and relief of leg pain. However, there are no long-term benefits.

Critique The natural history of sciatica due to lumbar disc herniation is such that the majority of patients will improve significantly within 8 weeks. Surgery is generally reserved for those patients who do not experience improvement within this time period. The Norwegian study by Weber appeared to demonstrate a benefit of surgery 1 year following surgery. Although this benefit was not seen at 4-year follow-up, this study influenced practice over the next two decades. This study would perhaps be criticized today for the methodology by which outcomes were assessed, which, on the whole, appears to be largely subjective assessments. The US study (SPORT trial) did not show any differences between surgery and conservative management in the intention-to-treat analysis. However, this study is hampered by a number of weaknesses including the large number of cross-overs between treatment groups. Nonetheless, supporters of the trial argue that this reflects the reality of spinal practice and is the only way in which a trial for this condition can be carried out. The same arguments are made to support the use of an as-treated analysis. However, this methodology does not allow the exclusion of the placebo effect and the possibility of false-positive outcomes. Fairbank has made the point that the fact that 44% of the conservative arm switched to surgery reflects the impact of this condition on the patient (Fairbank, 2008). The Netherlands study by Peul et al. differs from previous studies in that it evaluated the role of early surgery (within 2 weeks of randomization). In addition, Peul et al. specifically looked at the speed of recovery between conservative and surgical groups. Their study shows that although surgery results in faster recovery compared to conservative management, there is no overall difference in the longer-term outcomes. The findings of this study are upheld in an updated Cochrane review of published studies on lumbar disc surgery, which concluded that, for carefully selected patients with sciatica due to lumbar disc herniation, surgery provides a faster relief from the acute attack than conservative management (Gibson and Waddell, 2007). The lack of any long-term benefit following surgery means that the risks of surgery need to be balanced against the risks of conservative management. Surgical risks include a 1% risk of neurological damage. However, the risks of conservative management have not been quantified in sciatica and may include further neurological deterioration and the development of cauda equina syndrome (Fairbank, 2008). There are, therefore, insufficient data to justify surgical intervention in lumbar disc herniation on the balancing of risks. However, there may be a rationale for early surgical intervention based on a cost–benefit analysis. The cost of surgical intervention can be weighed against the cost of lost productivity for the longer period of recovery in patients managed conservatively. The authors of the Netherlands study went on to examine this cost–benefit analysis and found that there appears to be a strong economic argument supporting continued surgery for lumbar disc herniations producing sciatica (van den Hout et al., 2008). In summary, sciatica due to lumbar disc herniation is a common problem. However, there is still controversy regarding the natural history of this disorder. The trials reviewed here are landmarks in the field of neurosurgery as they have provided valuable information regarding the natural history of this disorder and its treatment specifically the treatment in the surgical group by discectomy without fusion. This in fact supports the notion that frequently in patients with disc herniation, the motion segment is damaged in totality. Sequestrectomy and discectomy does not address the pathology of the motion segment, but purely mechanical nerve compression. It is therefore usually effective for the treatment of radiculopathy more so than the treatment of mechanical lumbar back pain.

References Fairbank J. Prolapsed intervertebral disc. BMJ 2008; 336: 1317–1318. Gibson JNA, Waddell G. Surgical interventions for lumbar disc prolapse: updated Cochrane Review. Spine 2007; 32: 1735– 1747. Van den Hout WB, Peul WC, Koes BVW, Brand R, Klevit J, Thomeer RTWM for the Leiden–The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica from lumbar disc herniation: cost utility analysis alongside a randomised controlled trial. BMJ 2008; 336: 1351–1354.

131

4.6 Microscopic Sequestrectomy for Lumbar Disc Herniation Details of Study This prospective randomized study aimed to evaluate whether there was any difference in outcomes in patients undergoing standard microdiscectomy or microscopic sequestration of disc fragments. The study was carried out in a single centre in Heidelberg in Germany.

Study References Main Study Barth M, Diepers M, Weiss C, Thomé C. Two-year outcome after lumbar microdiscectomy versus microscopic sequestrectomy: part 1: evaluation of clinical outcome. Spine 2008; 33: 265–272.

Related References Barth M, Diepers M, Weiss C, Thomé C. Two-year outcome after lumbar microdiscectomy versus microscopic sequestrectomy: part 2: radiographic evaluation and correlation with clinical outcome. Spine 2008; 33: 273–279. Caspar W. A new surgical procedure for lumbar disc herniation causing less tissue damage through a microsurgical approach. Adv Neurosurg 1977; 4: 74–77. Thomé C, Barth M, Schard J, Schmiedek P. Outcome after lumbar sequestrectomy compared with microdiscectomy: a prospective randomised study. J Neurosurg Spine 2005; 2: 271–278. Williams RW. Microlumbar discectomy: a conservative surgical approach to the virgin herniated lumbar disc. Spine 1978; 3: 175–182. Yasargil MG. Microsurgical operation of herniated disc. Adv Neurosurg 1977; 4: 81–82.

Study Design Single-centre PRCT. Class of evidence

Randomization

Standard microdiscectomy versus sequestrectomy

Number of patients

Follow-up

2 years Primary outcomes: Neurological status Clinical symptoms Quality of life Secondary outcomes: Reherniation rates Functional and economic status

Number of centres

Inclusion criteria: age 18–60 years; no previous lumbar surgery; MRI confirmation of lumbar disc prolapse; no concomitant spinal disease. Standard discectomy involved removal of herniated disc matter and clearance of the disc space. Sequestrectomy involved removal of loose disc material and the disc space was not entered.

Outcome Measures Primary Endpoints Neurological status was established by clinical neurological examination. Quality of life was assessed using the SF-36 questionnaire.

Secondary Endpoints Functional and economic status were assessed using Prolo scores from questionnaires.

Results Follow-up was 93% at 2 years. There was no statistical difference in the reherniation rates: 12.5% sequestrectomy; 10% discectomy. There was no difference in neurological status or clinical symptomatology between the two groups. An analysis of overall outcome suggested that there was a statistically significant better overall outcome with 132

sequestrectomy at 2 years (p = 0.004) and a significantly greater improvement of overall outcome over time with sequestrectomy (p = 0.029). There appeared to be a statistically significant benefit of surgery in terms of parameters assessed by questionnaires including quality of life.

Conclusions Sequestrectomy may be advantageous compared to standard microdiscectomy and reherniation rates are similar with both techniques.

Critique In their earlier publication, the same authors had shown that there was no difference in outcome between the two techniques at 6-month follow-up (Thomé et al., 2005). This study extends these findings from that study to 2-year follow-up. However, the study includes two operating surgeons only at a single institution and is hindered by problems of blinding. The authors of the study attack what they refer to as ‘discectomy dogma’ which holds that unless the disc space is cleared there remains a risk of reherniation, subsequent nerve root compression, and clinical deterioration. In order to support their view that there is no scientific basis for this, the authors also carried out a radiological follow-up study by way of MRI on their cohort of patients (Barth et al., 2008). They report that sequestrectomy was associated with less post-operative disc degeneration and end-plate changes. Although this trial is a single-centre, unblinded study, it provokes substantial questions regarding the accepted rationale behind standard microdiscectomy and certainly paves the way for larger multi-centre trials.

References Barth M, Diepers M, Weiss C, Thomé C. Two-year outcome after lumbar microdiscectomy versus microscopic sequestrectomy: part 2: radiographic evaluation and correlation with clinical outcome. Spine 2008; 33: 273–279. Thomé C, Barth M, Schard J, Schmiedek P. Outcome after lumbar sequestrectomy compared with microdiscectomy: a prospective randomised study. J Neurosurg Spine 2005; 2: 271–278.

4.7 Surgery for Cauda Equina Syndrome Details of Studies There is no prospective randomized trial evaluating the timing of surgery for CES. However, case series reported in the literature have strongly supported the view that CES is a diagnostic and surgical emergency and that early surgery results in a better outcome than delayed surgery (O’Laoire et al., 1981; Hellström et al., 1986; Dinning and Schaefer, 1993; Shapiro, 1993; Kennedy et al., 1999; Chang et al., 2000; Shapiro, 2000). In this section, we have summarized the findings from these case series regarding the timing of surgery and sphincter function. Two meta-analyses of published case series have been carried out, one by Ahn et al. and one by Todd (Ahn et al., 2000; Todd, 2005; Jerwood and Todd, 2006). The meta-analysis carried out by Todd included all the case series listed here.

Study References Main Studies Case Series Chang HS, Nakaagawa H, Mizuno J. Lumbar herniated disc presenting with cauda equine syndrome. Long-term follow-up of four cases. Surg Neurol 2000; 53: 100–105. Dinning TAR, Schaefer HR. Discogenic compression of the cauda equina: a surgical emergency. Aust NZ J Surg 1993; 63: 927–934. Hellström P, Kortelainen P, Kontturi M. Late urodynamic findings after surgery for cauda equine syndrome caused by a prolapsed lumbar intervertebral disk. J Urol 1986; 135: 308–312. Kennedy JG, Soffe KE, McGrath A, Stephens MM, Walsh MG, McManus F. Predictors of outcome in cauda equina syndrome. Eur J Spine 1999; 8: 317–322. O’Laoire SA, Crockard HA, Thomas DG. Prognosis for sphincter recovery after operation for cauda equina compression owing to lumbar disc prolapse. BMJ 1981; 282: 1852–1854. Shapiro S. Cauda equina syndrome secondary to lumbar disc herniation. Neurosurgery 1993; 32: 743–747. Shapiro S. Medical realities of cauda equina syndrome secondary to lumbar disc herniation. Spine 2000; 25: 348–351.

Meta-analyses Ahn UM, Ahn NU, Buchowski MS, Garrett ES, Sieber AN, Kostuik JP. Cauda equina syndrome secondary to lumbar disc herniation. A meta-analysis of surgical outcomes. Spine 2000; 25: 1515–1522. Jerwood D, Todd NV. Reanalysis of the timing of cauda equina surgery. Br J Neurosurg 2006; 20: 178–179. Todd NV. Cauda equina syndrome: the timing of surgery probably does influence outcome. Br J Neurosurg 2005; 19: 301–306.

133

Related References Findlay G. Meta-analysis and the timing of cauda equina surgery. Br J Neurosurg 2008; 22: 137–138. Gleave JR, Macfarlane R. Cauda equina syndrome: what is the relationship between timing of surgery and outcome? Br J Neurosurg 2002; 16: 325–328. Macfarlane R. Meta-analysis and the timing of cauda equina surgery. Br J Neurosurg 2007; 21: 635.

Case Series Study Designs All retrospective analysis of case series.

Results

Meta-analyses Study Designs Ahn et al. (2000) Logistic regression analysis was carried out to determine correlation between timing of surgery and clinical outcomes. The authors compartmentalized timing of surgery into five groups: <24 h; 24–48 h; 2–10 days; 11 days to 1 month; >1 month. Todd (2005) Todd looked at patients in the literature who had been operated on within 24 or 48 h from onset of CES and developed two null hypotheses: 1 There is no benefit to early decompression within 24 h. 2 There is no benefit to early decompression within 48 h. Todd, therefore, analysed the literature with only one input variable—the timing of surgery. The only output variable analysed was recovery of sphincter function defined as ‘socially normal bladder function’. Statistical significance of odds ratios were calculated for the benefits of decompression on sphincter function in early versus late surgery.

Results

134

Analysis Ahn et al. (2000)

Findings Conclusions There was no significant benefit of surgery within 24 There is a benefit to surgery performed within 48 h or after 48 h h There was a significant beneficial effect of surgery between 24 and 48 h

Todd (2005) The probability of benefit from surgery within 24 h is The timing of surgery following CES probably p = 0.03 influences outcome The probability of benefit from surgery within 48 h is p = 0.005

Critique The first description of neurological compromise from a ruptured lumbar intervertebral disc was published in 1934 (Mixter and Barr, 1934). In 1959, Shephard published a review of CES cases presenting to Maida Vale Hospital, London, and concluded that early surgery was necessary to minimize permanent neurological damage (Shephard, 1959). O’Laoire et al. followed on from Shepard in their report of a series of patients with CES presenting to the National Hospitals for Nervous Diseases at Queen Square and Maida Vale, and University College London, between 1960 and 1980 (O’Laoire et al., 1981). O’Laoire et al. expressed their opinion regarding the management of this condition in no uncertain terms: The urgency of the diagnosis and treatment may be compared to that for extradural haematoma in head injury. O’Laoire et al. (1981, p 1852)

This is representative of the prevailing view regarding the timing of surgery for CES. The summary of conclusions from the case series included here reflects this view also. The meta-analysis carried out by Ahn et al. has received strong criticism for methodology and design. Indeed, it has been maintained that the logistic regression analysis performed by Ahn et al. is not a meta-analysis at all as it includes widely diverse patient populations with incomparable input and output variables (Kohles et al., 2004; Todd, 2005). The meta-analysis performed by Todd is certainly more rigorous in the application of the rules of meta-analysis and they demonstrated a statistically significant disproval of both their null hypotheses. Furthermore, in a re-analysis of Todd’s meta-analysis, Jerwood and Todd concluded that there is ‘overwhelming statistical evidence’ for the benefit of surgery to be performed as soon as is practically possible (Jerwood and Todd, 2006). Macfarlane has criticized the conclusions drawn from the metaanalysis of on several grounds (Macfarlane, 2007). Firstly, Macfarlane argues that it does not make physiological sense that patients with a complete CES should show signs of recovery up to 24 h after compression: larger peripheral nerves suffer irreversible injury after only a few hours. Secondly, Macfarlane expresses the view that emergency surgery can be associated with increased morbidity as it may be carried out in suboptimal conditions out of hours. Furthermore, Gleave and Macfarlane have indicated that in many of the reported series catheter placement may have been erroneously equated with loss of sphincter function (Gleave and Macfarlane, 2002). Findlay has also objected that re-analysis by Jerwood and Todd places too much emphasis on the later series from Shapiro et al. which may have included smaller discs than are typical of CES (Findlay, 2008). Findlay’s critique concluded that whilst the benefit of surgery for CES within 24 h remains unanswered it appears clear that an evolving case of CES will likely prevent further neurological deterioration. The answers to these questions have implications for resource management, consent, and medicolegal causality of neurological disability. It is possible that further studies will have to include some form of dynamic bladder function measurements in order to determine more accurately the extent of the CES perioperatively.

References Kohles SS, Kohles DA, Kar AP, Erlich VM, Polissar NL. Time-dependent surgical outcomes following cauda equina syndrome diagnosis: comments on a meta-analysis. Spine 2004; 29: 1281–1287. Mixter JM, Barr JS. Rupture of the intervertebral disc with involvement of the spinal canal. N Eng J Med 1934; 211: 210–215. Shephard RH. Diagnosis and prognosis of cauda equine syndrome produced by protrusion of lumbar disc. BMJ 1959; ii: 1434– 1439.

4.8 Surgery for Lumbar Stenosis Details of Study As part of the SPORT trial carried out in the United States, the investigators assessed the role of decompressive laminectomy versus conservative management in the treatment of lumbar stenosis.

Study References 135

Main Study Weinstein JN, Tosteson TD, Lurie JD, Tosteson ANA, Blood E, Hanscom B, Herkwoitz H, Cammisa F, Albert T, Boden SD, Hilibrand A, Goldberg H, Berven S, An H for the SPORT Investigators. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008; 358: 794–810.

Related Reference Weinstein JN, Lurie JD, Tosteson TD, Hanscom B, Tosteson ANA, Blood EA, Birkmeyer NJO, Hilibrand AS, Herkowitz H, Cammisa FP, Todd JA, Emery SE, Lenke LG, Abdu WA, Longley M, Errico TJ, Hu SS. Surgery versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 2007; 356: 2257–2270.

Study Design RCT plus concomitant observational cohort. Class of evidence

Randomization

Decompressive laminectomy versus conservative care

Number of patients

289

Follow-up

2 years Primary outcomes: Bodily pain Physical function Secondary outcomes: Patient-reported improvement

Number of centres

Stratification

None

There was also a concurrent observational cohort of 365 patients who refused randomization. Inclusion criteria: ≥12 weeks of neurogenic claudication and radiological evidence of lumbar stenosis. Exclusion criteria: lumbar spondylolisthesis; lumbar instability (defined radiologically on lateral lumbar films).

Outcome Measures Primary Endpoints Pain measured using the SF-36 which scores 0–100 with reducing severity. Physical function also measured with the SF-36 and also the Oswestry Disability Index (ODI) which scores 0– 100 with increasing severity. Outcomes were measured as changes from baseline scores.

Secondary Endpoint Patient-reported improvement.

Results Randomized cohort: 85% follow-up at 1 year; 76% follow-up at 2 years. Observational cohort: 92% follow-up at 1 year; 88% follow-up at 2 years. Cross-over: 42% of those randomized to conservative management had undergone surgery at 1 year compared to 63% of those randomized to surgery. Primary outcomes of randomized cohort at 1 and 2 years (intention-to-treat analysis):

136

In an as-treated analysis there appeared to be a benefit of surgery on primary outcomes in both randomized and observational cohorts.

Conclusions Surgery is better than conservative care in the management of lumbar stenosis.

Critique Follow-up in the randomized cohort was <80% at 2 years and so there is doubt regarding the validity of the observed benefit of surgery on bodily pain at 2 years. At 1 year where there was >80% follow-up there were no differences in the primary outcomes in the intention-to-treat analysis. Differences were reported using an as-treated analysis that combined both randomized and observational cohorts. There were difficulties with high cross-over rates in the randomized cohort and it appeared that patients crossing over to surgery had different demographics. The SPORT trial also included a similarly designed study that looked at a randomized cohort and a non-randomized cohort of patients undergoing surgery or non-operative care for lumbar spine degenerative spondylolisthesis (Weinstein et al., 2007). This study also had the same problems due to the high rates of cross-over between treatment groups and only an as-treated analysis showed a favourable effect of surgery. The SPORT studies are important because they show the inherent difficulties in carrying out RCTs in spinal patients. However, the studies may reflect the realities of spinal practice and any criticism of their inherent weakness needs to be weighed with this in mind.

Reference Weinstein JN, Lurie JD, Tosteson TD, Hanscom B, Tosteson ANA, Blood EA, Birkmeyer NJO, Hillibrand AS, Herkowitz H, Cammisa FP, Albert TJ, Emery SE, Lenke LG, Abdu WA, Longley M, Errico TJ, Hu SS. Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 2007; 356: 2257–2270.

4.9 Spinal Stabilization for Chronic Back Pain Details of Study The MRC spine stabilization trial was carried out in response to the NHS standing group on health technology in 1994 concluding that there was weak evidence for surgery in chronic low back pain. The trial was carried out in 15 secondary care orthopaedic and rehabilitation centres across the United Kingdom.

Study References Main Study Fairbank J, Frost H, Wilson-MacDonald J, Yu LM, Barker K, Collins R for the Spine Stabilisation Trial Group. Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial. BMJ 2005; 330: 1233–1240.

Related Reference Fritzell P, Hagg O, Wessburg P, Nordwall A, Group SLSS. Chronic back pain and fusion: a comparison of three surgical techniques: a prospective randomised controlled trial from the Swedish Lumbar Spine Study Group. Spine 2002; 27: 1131– 1141.

Study Design PRCT. Class of evidence

137

Randomization

Lumbar spine fusion versus intensive rehabilitation

Number of patients

349

Follow-up

2 years Primary outcomes: Back pain Mobility Secondary outcomes: General health assessment Psychological assessment Complications

Number of centres

Inclusion criteria: clinician and patient uncertainty regarding which treatment option is best; ≥12 months of chronic lower back pain; age 18–55 years. Exclusion criteria: previous stabilization surgery; significant co-morbidities; pregnancy; psychiatric disease. Choice of surgical method of stabilization was left to the discretion of the operating surgeon. Rehabilitation programmes were outpatient based with similar intensity regimens employed between study centres. Analysis was on an intention-to-treat basis.

Outcome Measures Primary Endpoints Back pain: ODI; 0 = no disability, 100 = severe disability. Mobility: shuttle-walking test (SWT) which measures maximal walking distance in metres.

Secondary Endpoints General health assessment: the SF-36. Psychological assessment: distress and risk assessment method (modified Zung depression index and sensory perception questionnaire)

Results Eighty-one per cent follow-up at 2 years. Cross-over: 28% of patients randomized to rehabilitation had undergone surgery by 2 years; 7% of patients randomized to surgery had rehabilitation instead of surgery. There was a small but statistically significant effect of surgery in improving ODI scores: –4.1 (p = 0.045). There was no difference in any of the other outcome measures.

Conclusions There is no clear evidence for the benefit of surgery over rehabilitation in the treatment of chronic low back pain patients.

Critique

4.10 Surgery for Cervical Spondylotic Myelopathy Details of Study The largest trial comparing surgery with conservative measures for the treatment of cervical spondylotic myelopathy was a 3-year prospective randomized study carried out between 1993 and 2000 in the Czech Republic.

Study References Main Study Kadaňka Z, Mareš M, Bednaník J, Smrcka V, Krbec M, Stejskal L, Chaloupka R, Surelová D, Novotný O, Urbánek I, Dušek L. Approaches to spondylotic cervical myelopathy: conservative versus surgical results in a 3-year follow-up study. Spine 2002; 27: 2205–2211.

Related References 138

Kadaňka Z, Bednaník J, Vohanka S, Vlach O, Stejskal L, Chaloupka R, Filipovicova D, Surelová D, Adamova B, Novotný O, Nemex M, Smrcka V, Urbánek I. Conservative treatment versus surgery in spondylotic cervical myelopathy: a prospective randomized study. Eur Spine J 2000; 9: 538–544. Kadaňka Z, Mareš M, Bednaník J, Smrcka V, Krbec M, Chaloupka R, Dušek L. Predictive factors for spondylotic cervical myelopathy treated conservatively or surgically. Eur J Neurol 2005; 12: 55–63.

Study Design Class of evidence

Randomization

Surgery versus conservative management

Number of patients

Follow-up

3 years Primary endpoint: Clinical improvement

Number of centres

Stratification

Age

Inclusion criteria: clinical cervical cord dysfunction; radiological evidence of cord compression on MRI; age <75 years; a modified Japanese Orthopaedic Association scale (mJOA) score of ≥12. Exclusion criteria: contraindications to surgery; pervious surgery; other significant neurological disease. The majority of patients undergoing surgery underwent anterior decompression. Analysis was on an intention-to-treat basis.

Outcomes Assessment Primary Endpoints Cervical cord function was graded clinically using the mJOA, which gives a total score out of 18. Mobility was assessed using a timed 10 m walk. Video-monitoring and patient self-evaluation of daily activities were also employed.

Results No significant difference was detected in the mJOA scores over the 3-year period. There was a small but significant improvement in the 10 m walk favouring those treated conservatively. There were no significant differences found between the two groups in evaluation of daily activities.

Conclusion The authors concluded that, on average, their study did not show that surgery is superior to conservative therapy for the treatment of cervical spondylotic myelopathy.

Critique One of the greatest weaknesses of this study was the small sample size. There is a possibility of a Type 2 error as power calculations indicate that a 42% difference in mJOA score would need to be seen in order to detect a significant difference with the number of patients included in this study (Matz et al., 2009). Nonetheless, the results of this study are in keeping with a similar study that also included electrophysiological data (Bednarík et al., 1999). In a further analysis of their data, the authors found that older patients appeared to do better with conservative treatment (Kadaňka et al., 2005).

References Bednarik J, Kadaňka Z, Vohanka S, Stejskal L, Vlach O, Schroder R. The value of somatosensory- and motor-evoked potentials in predicting and monitoring the effect of therapy in spondylotic cervical myelopathy. Prospective randomized study. Spine 1999; 24: 1593–1598. Kadaňka Z, Mareš M, Bednarík J, Smrcka V, Krbec M, Chaloupka R, Dušek L. Predictive factors for spondylotic cervical myelopathy treated conservatively or surgically. Eur J Neurol 2005; 12: 55–63. Matz PG, Holly LT, Mummaneni PV, Anderson PA, Groff MW, Heary RF, Kaiser MG, Ryken TC, Choudhri TF, Vresilovic EJ, Resnick DK. Anterior cervical surgery for the treatment of cervical degenerative myelopathy. J Neurosurg Spine 2009; 11: 170–173.

139

4.11 Surgery for Cervical Radiculopathy Details of Study There is only one prospective randomized trial that has compared surgery with conservative management for cervical radiculopathy. This study was carried out in Lund, Sweden, and compared surgery, physiotherapy, and immobilization with a cervical collar.

Study References Main Study Persson LCG, Moritz U, Brandt L, Carlsson CA. Cervical radiculopathy: pain, muscle weakness and sensory loss in patients with cervical radiculopathy treated with surgery, physiotherapy or cervical collar. A prospective controlled study. Eur Spine J 1997; 6: 256â&#x20AC;&#x201C;266.

Related Reference Fouyas IP, Statham FX, Sandercock PAG. Cochrane review on the role of surgery in cervical spondylotic radiculomyelopathy. Spine 2002; 27: 736â&#x20AC;&#x201C;747.

Study Design Single-centre PRT. Class of evidence

Randomization

Surgery versus physiotherapy versus immobilization with rigid cervical collar

Number of patients

Follow-up

16 months Primary endpoints: Relief of radicular pain Relief of sensory loss/paraesthesia Muscle strength

Number of centres

Stratification

None

Inclusion criteria: clinical and radiological evidence of cervical radiculopathy without spinal cord compression. Exclusion criteria: cervical cord compression; whiplash; psychiatric co-morbidities. Surgery was primarily anterior cervical discectomy and fusion with a Cloward technique. Analysis was on an intention-to-treat basis. Follow-up was 98% at 16 months.

Outcome Measures Primary Endpoints Pain was measured with a visual analogue scale: current pain and worst pain in the preceding week were scored. Sensory loss and paraesthesias were assessed by clinical examination by a physiotherapist. Muscle strength was measured using several dynamic devices.

Results At 3 months, patients undergoing surgery had statistically significantly less pain compared to those who received physiotherapy and those who underwent treatment in a rigid collar.

At 1 year, there was no difference in the relief of pain between any of the groups. 140

Although there was a significant relief of sensory loss/paraesthesia in the surgical group at 4 months, there was no difference between any of the three groups at 16-month follow-up. Muscle strength was slightly better in the surgery group at 4 months but there were no differences at 16 months.

Conclusion Surgery results in a more rapid relief of radicular pain, sensory loss, and muscle weakness compared to conservative measures although the longer-term outcomes appear to be similar.

Critique References De Palma AF, Subin DK. Study of the cervical syndrome. Clin Orthop 1965; 38: 135â&#x20AC;&#x201C;141. Fouyas IP, Statham FX, Sandercock PAG. Cochrane review on the role of surgery in cervical spondylotic radiculomyelopathy. Spine 2002; 27: 736â&#x20AC;&#x201C;747.

141

Chapter 5

Functional and epilepsy neurosurgery EAC Pereira, AL Green, RD Johnson, KJ Bulluss, A Astradsson, JA Hyam, TZ Aziz 5.0 5.1 5.2 5.3 5.4

Introduction Surgery for temporal lobe epilepsy Neuromodulation for epilepsy Deep brain stimulation for Parkinson’s disease: efficacy of deep brain stimulation in Parkinson’s disease Deep brain stimulation for Parkinson’s disease: efficacy of subthalamic versus pallidal deep brain stimulation in Parkinson’s disease 5.5 Deep brain stimulation for Parkinson’s disease: early subthalamic stimulation in Parkinson’s disease 5.6 Pallidal stimulation for dystonia: generalized and segmental dystonia 5.7 Spinal cord stimulation for failed back surgery syndrome 5.8 Neuromodulation for cluster headache and migraine: occipital nerve stimulation for cluster headache 5.9 Neuromodulation for cluster headache and migraine: deep brain stimulation for cluster headache 5.10 Neuromodulation for cluster headache and migraine: occipital nerve stimulation for migraine 5.11 Neurosurgical treatment of trigeminal neuralgia: microvascular decompression for trigeminal neuralgia 5.12 Neurosurgical treatment of trigeminal neuralgia: ablative techniques for trigeminal neuralgia 5.13 Deep brain stimulation for treatment-resistant depression 5.14 Molecular and cellular therapies for Parkinson’s disease

5.0 Introduction For this chapter we have selected some landmark studies in the field of epilepsy and functional neurosurgery. These two subspecialty areas have become closer over recent years due to the wider use of neuromodulation procedures to control epilepsy, including vagal nerve stimulation (VNS) and, more recently, deep brain stimulation (DBS). The first epilepsy surgery study discussed is the trial by Wiebe et al. evaluating the efficacy of surgery in temporal lobe epilepsy refractory to medical management, which was published in the New England Journal of Medicine in 2001 (Wiebe et al., 2001). This study was the first RCT assessing the efficacy of surgery for temporal lobe epilepsy. This trial has received widespread acclaim for confirming the efficacy of surgical intervention in epilepsy. Engel has pointed out that the results of this trial have laid open the question as to whether surgery should be considered earlier rather than as a last resort in medically refractory temporal lobe epilepsy (Engel, 2001). It is worth noting the circumstances in which it was possible to conduct this trial as surgical intervention is such a well-established method of treating temporal lobe epilepsy. The waiting list for temporal lobe surgery was 12 months and so this allowed Wiebe et al. to fast track patients randomized to surgery to almost immediate pre-operative evaluation. This allowed for the ethical randomization of patients with a 12-month window in which to compare the two treatment arms. In the second section we continue with the theme of epilepsy surgery and consider the two largest double-blind randomized trials evaluating VNS for the treatment of epilepsy (Ben-Menachem et al., 1994; Handforth et al., 1998) and the SANTE trial of DBS (Fisher et al., 2010). The next sections of this chapter deal with functional neurosurgery by way of DBS and its indications. Functional neurosurgery encompasses brain lesioning and DBS in order to alter brain function. DBS was made possible by combining stereotactic methods with advances in implanted electrical stimulators and an increased understanding of the functional anatomy of the deep nuclei of the brain. Stereotactic apparatus were introduced by Horsley and Clarke and further developed for surgical use by Spiegel and Wycis (Horsley and Clarke, 1908; Spiegal et al., 1947). The advent of stereotactic methods has facilitated the development of two related fields of neurosurgery: functional neurosurgery and stereotactic radiosurgery. DBS was first used clinically in the management of cancer pain and was then applied to a myriad of pain syndromes over the next half century. DBS for pain has predominantly been evaluated by case series rather than larger multi-centre controlled trials. However, a meta-analysis of DBS for pain relief concluded that it is effective in wellselected patients (Bittar et al., 2005). Many of the developments in DBS for movement disorders have been the result of pre-clinical, primate-based research and these in many ways are the real landmark studies in the history of this field of DBS. For example, the identification of the subthalamic nucleus (STN) as a target for DBS in the treatment of Parkinson’s disease (PD) was the direct result of findings in primate models (Bergman et al., 1990; Aziz et al., 1991). However, as the aim of this volume is to highlight landmark clinical studies we have chosen to include studies of DBS for movement disorders which we feel meet this criteria. We have considered five studies evaluating DBS for 142

PD. The first of these is a long-term follow-up study evaluating the efficacy of STN stimulation (Krack et al., 2003). The second is a randomized trial comparing STN DBS to medical management (Handforth et al., 1998). The third is the first randomized trial of best medical therapy versus STN DBS for advanced PD (Weaver et al., 2009). We then critique a trial comparing STN to GPi DBS for PD (Follett et al., 2010) and finally review a trial of STN DBS for early PD (Schuepbach et al., 2013). We then consider a clinical trial of neurostimulation versus sham-stimulation of the globus pallidus internus (GPi) for generalized and cervical dystonia (Kupsch et al., 2006) and a multi-centre trial focusing upon cervical dystonia (Kiss et al., 2007). DBS for essential tremor is reasonably well established as an effective therapy that is not reviewed further here as no significant large clinical trials have been required to establish its efficacy. Chronic pain historically forms an important aspect of neurosurgery that has become more the realm of anaesthetists in recent times. Nevertheless, percutaneous neuromodulation interventions have led to suboptimal placement and lead migration causing neurosurgeons to become increasingly involved in spinal cord stimulation, which we discuss (Kumar et al., 2007). New devices have generated a resurgence of interest in peripheral nerve stimulation. Alongside these therapies, DBS and motor cortex stimulation (MCS) remain for refractory syndromes in selected patients. While recent case series of DBS and MCS exist, some using on-off randomized paradigms (Boccard et al., 2013; Pereira et al., 2013; Lefaucheur et al., 2009), these treatments have been applied for several decades and new RCT evidence versus best medical therapies remains lacking, and the cohorts of patients described are usually quite heterogeneous in aetiology, therefore landmark studies have not been included here. DBS and occipital nerve stimulation for cluster headache and migraine are then discussed because of their novel neurosurgical indications despite a lack of randomized trial evidence (Burns et al., 2007; Leone et al., 2006; Saper et al., 2011). Trigeminal neuralgia is given its own section as it remains very much a part of neurosurgery with a variety of interventions from microvascular decompression to ablative techniques reviewed (Barker et al., 1996; Lopez et al., 2004). Another area in which DBS is gaining an interested following is functional neurosurgery for psychiatric disorders. This area was formerly referred to as ‘psychosurgery’ and has, regrettably, a rather notorious history which has held back development in this field. It is worth, therefore, considering this history briefly here. It is likely that ‘psychosurgery’ may have an ancient pedigree with literature on trephination for psychosis dating back to 1500 bc (Mashour et al., 2005). The first modern psychosurgical procedure was performed by Gottlieb Burckhardt when he carried out ‘topectomy’ (the removal of multiple foci of cerebral cortex) for psychosis in the late nineteenth century (Burckhardt, 1891). However, it was in the first half of the twentieth century that the Europeans Egas Moniz and Almeida Lima developed the prefrontal leucotomy for psychosis (Moniz, 1937). Although Moniz coined the term ‘psychosurgery’ and was awarded the Nobel Prize in Physiology or Medicine in 1949, the field took a downturn following the work of Walter Freeman who developed the transorbital frontal leucotomy with fellow American James Watts (Freeman, 1948). Although developed with the best of intentions, it is unfortunate that this procedure entered widespread, and often indiscriminate, use in the hands of non-surgically trained psychiatrists and physicians with such widespread complications that it was eventually rejected by most neurosurgeons, and indeed made illegal in some countries. This, combined with the introduction of an effective antipsychotic, chlorpromazine, into the pharmacopoeia, resulted in the effective death of ‘psychosurgery’. There is, perhaps, some irony in the fact that the use of chlorpromazine as an antipsychotic stemmed from the observations of the French neurosurgeon Henri Laborit. Despite the decline of frontal lobectomy, other lesioning procedures have survived the test of time including cingulotomy for obsessive–compulsive disorder (OCD) which was made popular by Ballantyne (Ballantyne, 1967). With the advent of DBS the field of functional neurosurgery is making inroads again into the realms of psychiatric disorders. We have included in this chapter the study by Mayberg et al. in Toronto that evaluates subgenual cingulated white matter as a target for DBS in treatment-resistant depression (Mayberg et al., 2005). Although this is a pilot study it represents a landmark in functional neurosurgery for extrapolating observations from functional imaging studies to develop a testable hypothesis regarding new efficacious targets for DBS in a major debilitating disorder. Subsequent efficacy has proved less impressive (Lozano et al., 2012), but intriguingly the research has spawned novel psychiatric indications for DBS such as anorexia nervosa (Lipsman et al., 2013). Stereotactic radiosurgery was conceived by Lars Leksell of the Karolinska Institute in Stockholm (Leksell, 1951). Leksell combined stereotactic localization with radiation physics and together with Börge Larsson developed the Leksell gamma knife (Leksell, 1983). Stereotactic radiosurgery, therefore, allows the delivery of focal ablative lesions to a closed cranium and has been deployed successfully in the management of vascular malformations (AVMs), benign tumours (vestibular schwannomas and meningiomas), and pain syndromes (trigeminal neuralgia). Pollock, in a review of the clinical evidence for the efficacy of stereotactic radiosurgery, found that the majority of studies provided only Level III evidence (Pollock, 2006). This in part reflects the widespread establishment of this technique over the last 30 years. However, there are two RCTs evaluating the role of stereotactic radiosurgery in the management of brain metastases and these have been included in the neuro-oncology chapter (Andrews et al., 2004; Aoyama et al., 2006). Finally, we look to the future reviewing important studies in the development of molecular and cellular therapies for PD. Such therapies may one day replace DBS as the current neurosurgical therapy of choice and their translation to the clinic in terms of safety and efficacy is discussed (Kordower et al., 2008; Li et al., 2008; Mendez et al., 2008). 143

References Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ Jr. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004; 363: 1655–1672. Aoyama H, Shirato H, Tago T, Nakagawa K, Toyoda T, Hatano K, Kenjyo M, Oya N, Hirota S, Shioura H, Kunieda E, Inomata T, Hayakawa K, Katoh N, Kobashi G. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for the treatment of brain metastases—a randomised controlled trial. JAMA 2006; 21: 2483–2491. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the primate. Mov Disord 1991; 6: 288–292. Ballantyne HT Jr, Cassidy WL, Flanagan NB, Marino R Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967; 26: 488–495. Ben-Menachem E, Mañon-Espaillat R, Ristanovic R, Wilder BJ, Stefan H, Mirza W, Tarver WB, Wernicke JF. Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group. Epilepsia 1994; 35: 616–626. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesion of the subthalamic nucleus. Science 1990; 249: 1436–1438. Bittar RG, Kar-Purkayastha I, Owen SL, Bear RE, Green A, Wang SY, Aziz TZ. Deep brain stimulation for pain relief: a meta-analysis. J Clin Neurosci 2005; 12: 515–519. Boccard SG, Pereira EA, Moir L, Aziz TZ, Green AL. Long-term outcomes of deep brain stimulation for neuropathic pain. Neurosurgery 2013; 72(2): 221–230. Burckhardt G. Über Rindenexcisionen, als Beitrag zur operativen Therapie der Psychosen. Allg Z Psychiatr Med 1891; 47: 463–548. Burns B, Watkins L, Goadsby PJ. Successful treatment of medically intractable cluster headache using occipital nerve stimulation (ONS). Lancet 2007; 369: 1099–1106. Engel J. Finally, a randomised, controlled trial of epilepsy surgery. N Engl J Med 2001; 345: 365–366. Fisher R, Salanova V, Witt T, Worth R, Henry T, Gross R, Oommen K, Osorio I, Nazzaro J, Labar D, Kaplitt M, Sperling M, Sandok E, Neal J, Handforth A, Stern J, DeSalles A, Chung S, Shetter A, Bergen D, Bakay R, Henderson J, French J, Baltuch G, Rosenfeld W, Youkilis A, Marks W, Garcia P, Barbaro N, Fountain N, Bazil C, Goodman R, McKhann G, Babu Krishnamurthy K, Papavassiliou S, Epstein C, Pollard J, Tonder L, Grebin J, Coffey R, Graves N, for the SANTE Study Group. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia 2010; 51(5): 899–908. Follett KA, Weaver FM, Stern M, Hur K, Harris CL, Luo P, Marks WJ Jr, Rothlind J, Sagher O, Moy C, Pahwa R, Burchiel K, Hogarth P, Lai EC, Duda JE, Holloway K, Samii A, Horn S, Bronstein JM, Stoner G, Starr PA, Simpson R, Baltuch G, De Salles A, Huang GD, Reda DJ, for the CSP (Veteran Affairs Cooperative Studies Program) 468 Study Group. Pallidal versus subthalamic deep-brain stimulation for Parkinson’s disease. N Engl J Med 2010; 362: 2077– 2091. Freeman W. Transorbital leucotomy. Lancet 1948; 2: 371–373. Handforth A, DeGiorgio CM, Schachter SC, Uthman BM, Naritoku DK, Tecoma ES, Henry TR, Collins SD, Vaughn BV, Gilmartin RC, Labar DR, Morris GL 3rd, Salinsky MC, Osorio I, Ristanovic RK, Labiner DM, Jones JC, Murphy JV, Ney GC, Wheless JW. Vagus nerve stimulation therapy for partial-onset seizures: a randomized active-control trial. Neurology 1998; 51: 48–55. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908; 31: 45–124. Kiss ZH, Doig-Beyaert K, Eliasziw M, Tsui J, Haffenden A, Suchowersky O; Functional and Stereotactic Section of the Canadian Neurosurgical Society; Canadian Movement Disorders Group. The Canadian multicentre study of deep brain stimulation for cervical dystonia. Brain 2007; 130(Pt 11): 2879–2886. Kordower JH, Chu Y, Hauser RA, Freeman TB, Olanow CW. Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson’s disease. Nat Med 2008; 14: 504–506. Kupsch A, Benecke R, Müller J, Trottenberg T, Schneider GH, Poewe W, Eisner W, Wolters A, Müller JU, Deuschl G, Pinsker MO, Skogseid IM, Roeste GK, Vollmer-Haase J, Brentrup A, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Naumann M, Volkmann J, for the Deep-Brain Stimulation for Dystonia Study Group. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006; 355: 1978–1990. Lefaucheur JP, Drouot X, Cunin P, Bruckert R, Lepetit H, Créange A, Wolkenstein P, Maison P, Keravel Y, Nguyen JP. Motor cortex stimulation for the treatment of refractory peripheral neuropathic pain. Brain 2009; 132(Pt 6): 1463–1471. Leksell L. The stereotactic method and radiosurgery of the brain. Acta Chir Scand 1951; 107: 316–319. Leksell L. Stereotactic radiosurgery. J Neurol Neurosurg Psychiatry 1983; 46: 797–803. Leone M, Franzini A, Broggi G, Bussone G. Hypothalamic stimulation for intractable cluster headache: long-term experience. Neurology 2006; 67: 150–152. Li JY, Englund E, Holton JL, Soulet D, Hagell P, Lees AJ, Lashley T, Quinn NP, Rehncrona S, Bjorklund A, Widner H, Revesz T, Lindvall O, Brundin P. Lewy bodies in grafted neurons in subjects with Parkinson’s disease suggest host-tograft disease propagation. Nat Med 2008; 14: 501–503. Lipsman N, Woodside DB, Giacobbe P, Hamani C, Carter JC, Norwood SJ, Sutandar K, Staab R, Elias G, Lyman CH, Smith GS, Lozano AM. Subcallosal cingulate deep brain stimulation for treatment-refractory anorexia nervosa: a phase 1 pilot trial. Lancet 2013; 381(9875): 1361–1370. Lozano AM, Giacobbe P, Hamani C, Rizvi SJ, Kennedy SH, Kolivakis TT, Debonnel G, Sadikot AF, Lam RW, Howard

144

AK, Ilcewicz-Klimek M, Honey CR, Mayberg HS. A multicenter pilot study of subcallosal cingulate area deep brain stimulation for treatment-resistant depression. J Neurosurg 2012; 116(2): 315–322. Mashour GA, Walker EE, Martuza RL. Psychosurgery: past, present, and future. Brain Res Rev 2005; 48: 409–419. Mendez I, Vinuela A, Astradsson A, Mukhida K, Hallett P, Robertson H, Tierney T, Holness R, Dagher A, Trojanowski JQ, Isacson O. Dopamine neurons implanted into people with Parkinson’s disease survive without pathology for 14 years. Nat Med 2008; 14: 507–509. Moniz E. Essai d’in traitement chirurgical de certaine psychoses. Bull Acad Med 1937; 93: 1379–1385. Pereira EA, Boccard SG, Linhares P, Chamadoira C, Rosas MJ, Abreu P, Rebelo V, Vaz R, Aziz TZ. Thalamic deep brain stimulation for neuropathic pain after amputation or brachial plexus avulsion. Neurosurg Focus 2013; 35(3): E7. Pollock BE. An evidence-based medicine review of stereotactic radiosurgery. Prog Neurol Surg 2006; 19: 152–170. Saper JR, Dodick DW, Silberstein SD, McCarville D, Sun M, Goadsby PJ. Occipital nerve stimulation for the treatment of intractable chronic migraine headache: ONSTIM feasibility study. Cephalalgia 2011; 31(3): 271–285. Schuepbach WMM, Rau J, Knudsen K, Volkmann J, Krack P, Timmerman L, Hälbig TD, Hesekamp H, Navarro SM, Meier N, Falk D, Mehdorn M, et al. for the EARLYSTIM Study Group. Neurostimulation for Parkinson’s disease with early motor complications. N Engl J Med 2013; 368: 610–622. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on human brain. Science 1947; 106: 349–350. Weaver FM, Follett K, Stern M, Hur K, Harris C, Marks WJ Jr, Rothlind J, Sagher O, Reda D, Moy CS, Pahwa R, Burchiel K, Hogarth P, Lai EC, Duda JE, Holloway K, Samii A, Horn S, Bronstein J, Stoner G, Heemskerk J, Huang GD; CSP 468 Study Group. Bilateral deep brain stimulation vs best medical therapy for patients with advanced Parkinson disease: a randomized controlled trial. JAMA 2009; 301: 63–73. Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001; 345: 311–318.

5.1 Surgery for Temporal Lobe Epilepsy Details of Study This study by the Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Group is the first PRCT to assess the efficacy and safety of surgical treatment for temporal lobe epilepsy. The study was carried out at the London Health Sciences Center, University of Western Ontario, Canada.

Study References Main Study Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001; 345: 311–318.

Related Reference Engel J Jr. The timing of surgical intervention for mesial temporal lobe epilepsy: a plan for a randomised controlled clinical trial. Arch Neurol 1999; 56: 1338–1341.

Study Design PRCT. Class of evidence

Randomization

Medical treatment versus surgery

Number of patients

Follow-up

Primary outcome: 1 year Secondary outcome: None

Number of centres

Stratification

Presence or absence of generalized motor seizures

Patients randomized to medical treatment were put on a 1-year waiting list for surgery (the standard practice in the study centre). Patients randomized to surgery were admitted for pre-operative evaluation within 48 h of randomization. Of the 86 eligible patients, 80 agreed to participate with 40 randomized to each arm of the trial. Analysis was performed on an intention-to-treat basis.

Outcome Measures 145

Primary Endpoint Freedom from seizures impairing awareness (i.e. disabling seizures) at 1 year.

Other Endpoints Analysed Freedom from all seizures, including auras. Frequency of seizures in those that were not seizure free was analysed by calculating percentage change in monthly average. Mean severity of all seizures assessed at 3-monthly intervals using the Liverpool Seizure Severity Scale which scores the severity from 10â&#x20AC;&#x201C;48 with higher scores reflecting increased severity. Quality of life was assessed using a standard epilepsy quality of life inventory (QOLIE-89) which score quality of life from 0â&#x20AC;&#x201C;100 with higher score reflecting superior quality of life. Number of patients employed or attending school was recorded at 3-monthly intervals.

Results There was 100% follow-up with no crossovers.

Four patients in surgical group did not undergo surgery; of those that did undergo surgery 64% were free from disabling seizures and 42% were free from all seizures. One patient in the medical group died from sudden unexplained causes. Four patients had adverse effects from surgery: one patient with wound infection; one with sensory disturbance in one lower limb secondary to a focal thalamic infarct; and two with verbal memory disturbance affecting occupation. Depression occurred in 18% of patients in the surgical group and in 20% of patients in the medical group.

Conclusions Surgery for temporal lobe epilepsy is not only safe but superior to prolonged medical treatment.

Critique The study by Weibe et al. showed that RCTs of surgery for epilepsy are feasible. In addition, the trial showed that surgery for temporal lobe epilepsy is safe and effective. The follow-up time of the trial had to be limited to 1 year as all patients entered into the trial would ultimately undergo surgery, although those randomized to the medical arm would go on the 1-year waiting list. Indeed, it was this 1-year waiting list that allowed the ethical randomization of patients and allowed the trial to be undertaken. Even with this short period of follow-up the authors were able to show a statistically significant difference between the two treatment arms. Several criticisms have been levied against the trial including the outcome assessments used. For example, the questionnaires used to assess outcome may not have picked up on subtle personality changes following the resection of amygdala and hippocampus. However, the authors of the trial have indicated that patients undergoing surgery rated themselves better than the medically treated patients on functions that would be affected by the loss of the amygdala and hippocampus, including memory and emotional well-being. Although questions have been raised regarding the antiepileptic medication used to try and control seizures in patients in the medical arm of the trial, Wiebe et al. have emphasized that all patients were given optimal doses and medication combinations by experienced epilepsy specialists. This trial has confirmed that there is a role of surgery for medically refractory epilepsy. It has estimated that, at least in the United States, that only approximately 1.5% of eligible patients undergo surgery for epilepsy control (Engel and Shewmon, 1993). Interest has now moved to investigate whether seizure relief is influenced by the 146

degree of temporal lobe resection. A recent review has noted that standard anterior temporal lobectomy when compared with selective amygdalohippocampectomy, has an improved chance of seizure freedom from disabling seizures in patients with temporal lobe epilepsy (Josephson et al., 2013).

References Engel J Jr, Shewmon DA. Overview: who should be considered a surgical candidate?, in Engel J Jr (ed), Surgical Treatment of the Epilepsies, 2nd ed. New York: Raven Press, 1993, pp 23–24. Josephson CD, Dykeman J, Fiest KM, Liu X, Salder RM, Jette N, Weibe S. Systematic review and meta-analysis of standard vs selective temporal lobe epilepsy surgery. Neurology 2013; 80(18): 1669–1676.

5.2 Neuromodulation for Epilepsy Vagal Nerve Stimulation Studies Details of Studies VNS is now a well-established treatment for refractory epilepsy and a number of efficacy trials have been performed. The trials described in this section are landmarks because they represent the early prospective, randomized studies that changed efficacy studies from a number of case series to Level I evidence and led to the much more widespread use of VNS.

Study References Main Studies Ben-Menachem E, Mañon-Espaillat R, Ristanovic R, Wilder BJ, Stefan H, Mirza W, Tarver WB, Wernicke JF. Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group. Epilepsia 1994; 35: 616–626 Handforth A, DeGiorgio CM, Schachter SC, Uthman BM, Naritoku DK, Tecoma ES, Henry TR, Collins SD, Vaughn BV, Gilmartin RC, Labar DR, Morris GL 3rd, Salinsky MC, Osorio I, Ristanovic RK, Labiner DM, Jones JC, Murphy JV, Ney GC, Wheless JW. Vagus nerve stimulation therapy for partial-onset seizures: a randomized active-control trial. Neurology 1998; 51: 48–55.

Related References George R, Salinsky M, Kuzniecky R, Rosenfeld W, Bergen D, Tarver WB, Wernicke JF. Vagus nerve stimulation for treatment of partial seizures: 3. Long-term follow-up on first 67 patients exiting a controlled study. First International Vagus Nerve Stimulation Study Group. Epilepsia 1994; 35: 637–643. Ramsay RE, Uthman BM, Augustinsson LE, Upton AR, Naritoku D, Willis J, Treig T, Barolat G, Wernicke JF. Vagus nerve stimulation for treatment of partial seizures: 2. Safety, side effects, and tolerability. First International Vagus Nerve Stimulation Study Group. Epilepsia 1994; 35: 627–633.

Study Designs

Class of evidence

Ben-Menachem et al. (1994)

Handforth et al. (1998)

Randomization ‘High’ versus ‘low’ stimulation parameters (see following list)

High versus low stimulation parameters

Number of patients

83 (67 included in analysis)

254 (196 included in analysis)

Follow-up

14-week period of stimulation (last 12 weeks included in efficacy analysis)

3 months after 2-week ‘ramp-up’ period where stimulation maximized

Number of centres

17 (12 in USA)

20 (all in USA)

Stratification None

None

Inclusion criteria: • Ben-Menachem et al. (1994): medically refractory partial seizures (at least 6 per month during a 12-week baseline period). • Handforth et al. (1998): six or more partial-onset seizures involving alteration in consciousness (complex partial or secondary generalized) over 30 days with no more than 21 days between seizures; other seizure types; be able to submit accurate seizure counts (or by a carer); 12–65 years; use contraception if female and 147

fertile; be on one to three marketed antiepileptic drugs on a stable regimen. Exclusion criteria: • Ben-Menachem at al. (1994): a concomitant unstable medical condition; seizure aetiology best treated by another means, e.g. resective surgery; pregnancy. • Handforth et al. (1998): deteriorating neurological or medical conditions; pregnancy; cardiac or pulmonary disease; active peptic ulcer; history of non-epileptic seizures; 1 one or more episodes of status epilepticus in past 12 months; prior cervical vagotomy; inability to consent; prior VNS; prior DBS; resective epilepsy surgery; inability to perform pulmonary function tests or to attend clinic. Demographics in the two trials are very similar with mean age around 32–35 years with a range of 13–60 years. The Ben-Menachem et al. (1994) study was a smaller, preliminary RCT that showed VNS is safe and potentially effective in seizure reduction. Handforth et al. (1998) went on to confirm these results in a larger cohort with a high (99%) completion rate.

Outcome Measures Primary Endpoints Ben-Menachem et al. (1994) Overall change in seizure frequency for high- and low-frequency groups. Handforth et al. (1998) Percentage change in total seizure frequency during treatment period compared to baseline. Secondary Endpoints Ben-Menachem et al. (1994) Changes in seizure intensity and duration. Patient reported ability to ‘abort’ or ‘decrease’ a seizure as a result of magnet use. Global ratings by patients, companions, and investigators (reported by Ramsey et al., 1994). Handforth et al. (1998) Between-group comparisons of seizures involving alteration of awareness. Within-group changes in seizure frequency during treatment compared to baseline. Number of patients with 50% or 75% seizure frequency reductions. Global evaluation scores. Adverse events.

Results Reduction in seizure frequency was similar in both trials and both trials showed that high-frequency VNS significantly reduces seizures compared to baseline, whereas low-frequency VNS does not. Seizure reduction occurred for both total seizures as well as partial-onset seizures with alteration of awareness. These reductions occurred whether or not the patient had auras. Seizure intensity or duration was not significantly changed. However, regarding all of these results, there was considerable individual variation in response, and although there was a mean reduction in frequency, some patients suffered an increase in seizure frequency with VNS. Global ratings of change indicated significant improvement reported by interviewers but not by the patients or companions. Interestingly, global ratings improved with both high and low stimulation but were significantly higher with the former. Side effects of stimulation included voice change, throat paraesthesiae, dyspnoea, and cough. Infection occurred in approximately 11–12% of patients and there were no reported device-related deaths.

Conclusions High-frequency VNS significantly reduces the frequency of seizures in patients with refractory epilepsy.

Deep Brain Stimulation Study Details of Study 148

This study, Stimulation of the Anterior Nucleus of the Thalamus for Epilepsy (SANTE trial) is a multi-centre, doubleblinded randomized trial to assess the efficacy and safety of basal ganglia stimulation to control medical refractory partial seizures. The study was carried out at 17 neurosurgical centres within the United States.

Study Reference Main Study Fisher R, Salanova V, Witt T, Worth R, Henry T, Gross R, Oommen K, Osorio I, Nazzaro J, Labar D, Kaplitt M, Sperling M, Sandok E, Neal J, Handforth A, Stern J, DeSalles A, Chung S, Shetter A, Bergen D, Bakay R, Henderson J, French J, Baltuch G, Rosenfeld W, Youkilis A, Marks W, Garcia P, Barbaro N, Fountain N, Bazil C, Goodman R, McKhann G, Babu Krishnamurthy K, Papavassiliou S, Epstein C, Pollard J, Tonder L, Grebin J, Coffey R, Graves N; SANTE Study Group. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia 2010; 51(5): 899â&#x20AC;&#x201C;908.

Study Design PRCT. Class of evidence

Randomization

Stimulation versus no stimulation within the first 3 months

Number of patients

110

Follow-up

Primary outcome: 3 months Secondary outcome: 2 years

Number of centres

Stratification

Percentage difference in frequency partial seizures

Of the 157 eligible patients, 110 underwent bilateral electrode implantation; 54 patients were randomized to stimulation and 55 to control group. Analysis was performed on an intention-to-treat basis. Generalized estimating equations (GEE) model-adjusted mean per cent difference in seizure frequency

Outcome Measures Primary Endpoint Comparison of seizure reductions in blinded phase (first 3 months). Other Endpoints Analysed Mean severity of all seizures assessed using the Liverpool Seizure Severity Scale. Quality of life was assessed using a standard epilepsy quality of life inventory (QOLIE-31).

Results

Fourteen patients (13%) were seizure free for at least 6 months. There was a reduction in epilepsy-related injuries (7% active versus 25% control).

Other Findings 12.7% of patients became seizure free for at least 6 months. Among 110 implanted patients with mean follow-up of 3 years there were five deaths. One patient died in the baseline phase before surgery from sudden unexpected death in epilepsy (SUDEP). In long-term phase one 149

patient drowned and another committed suicide. One patient in the unblinded phase and long-term follow-up phase died from SUDEP. There were five asymptomatic haemorrhages (4.5%). Fourteen patients (12.7%) developed infections in either the stimulator pocket (7.3%), extension lead (5.5%), or the burr hole (1.8%). While neuropsychological scores for cognition and mood did not differ between the two groups, stimulationrelated complications compared to control group included depression (eight versus one) and memory impairment (seven versus one). The depression in 50% of the patient and all of the memory impairments resolved over the length of the trial.

Conclusions Bilateral stimulation of the anterior nuclei of the thalamus reduces seizure frequency.

Critique Cooper et al. were the first to trial brain stimulation for the treatment of epilepsy (Cooper et al., 1973; Cooper et al., 1977a; Cooper et al., 1977b; Cooper et al., 1978). Unfortunately, controlled trials failed to confirm these results (Van Buren et al., 1978, Wright et al., 1984). Several DBS targets have been used to treat epilepsy including the hippocampus (Velasco et al., 2000), the caudate nucleus (Chkhenkeli and Chkhenkeli, 1997), the centromedian thalamic nucleus (Fischer et al., 1992; Velasco et al., 1995), and the posterior hypothalamus (Mirski and Fisher, 1994). More recently the subthalamic nucleus has been selected as a target for DBS in epilepsy (Benabid et al., 2002). The recent publication of the SANTE study has provided further evidence that DBS can be used to control medical refractory epilepsy. It may be expected that DBS will be more efficacious in seizure reduction than VNS as the stimulation is directly affecting the brain rather than relying on some method of anterograde stimulation. However, there is no doubt that VNS is safer as it does not have a risk of stroke. The landmark VNS studies described here are impressive in that they applied the principles of prospective, randomized drug trials to a surgical treatment and were well executed. They are, however, far from perfect. One of the main problems is the use of a low-frequency stimulation group as a ‘placebo’. This was a necessary compromise in order to ensure adequate ‘blinding’ as those with no stimulation would be aware that they were not receiving it. But it is not as good as a ‘true’ placebo group. The fact that there was a seizure reduction in this ‘placebo’ group illustrates this point. A second flaw is that seizure reduction in the first 3 months may underestimate the long-term improvement, as evidence suggests that efficacy improves even up to 18 months after the onset of therapy (Uthman et al., 1993; Salinsky et al., 1996). Despite these criticisms concerning efficacy, both trials established VNS as a safe therapy with beneficial effects, in some patients more than others. The study by Fisher et al. demonstrates that anterior nucleus stimulation for medically refractory focal epilepsy does reduce seizure frequency. One patient outlier had a significant impact on the results. In this patient, stimulation induces >200 seizures in 3 days that ceased when the stimulation was switched off and this phenomenon did not recur when a lower voltage was used. For this reason, the results of this trial are given with this outlier included as well as excluded from the analysis. It is important to highlight that within the patient population a significant proportion have undergone previous surgical intervention with 45% having received previous vagal nerve stimulation and 25% another form of epilepsy surgery. During the blinded phase there was a significant increase in depression events (mild or moderate with half resolving spontaneously) and memory impairment (all resolving spontaneously). As with DBS for movement disorders the mechanism of action is unclear but concurrent thalamic and scalp EEG after implantation suggest a recruiting rhythm within the limbic system that is elicited with low stimulation that leads to the reduction in seizure frequency (Zumsteg et al., 2006). This trial has established that DBS is efficacious in a selected group of patients with medically refractory epilepsy.

References Benabid AL, Minotti L, Koudsié A, de Saint Martin A, Hirsch E. Antiepileptic effect of high-frequency stimulation of the subthalamic nucleus (corpus luysi) in a case of medically intractable epilepsy caused by focal dysplasia: a 30-month follow-up: technical case report. Neurosurgery 2002; 50: 1385–1391. Chkhenkeli SA, Chkhenkeli IS. Effects of therapeutic stimulation of nucleus caudatus on epileptic electrical activity of brain patients with intractable epilepsy. Stereotact Funct Neurosurg 1997; 69: 221–224. Cooper IS, Amin I, Gilman S. The effect of chronic cerebellar stimulation upon epilepsy in man. Trans Am Neurol Assoc 1973; 98: 192–196. Cooper IS, Amin I, Riklan M, Waltz JM, Poon TP. Chronic cerebellar stimulation in epilepsy. Clinical and anatomical studies. Arch Neurol 1976; 33: 559–570. Cooper IS, Amin I, Upton A, Riklan M, Watkins S, McLellan L. Safety and efficacy of chronic cerebellar stimulation. Appl Neurophysiol 1977; 40: 124–134. Cooper IS, Upton AR. Use of chronic cerebellar stimulation for disorders of inhibition. Lancet 1978; 1: 595–560. Fischer RS, Uematsu S, Krauss GL, Cysyk BJ, McPherson R, Lesser RP, Gordon B, Schwerdt P, Rise M. Placebocontrolled pilot study of centromedian thalamic stimulation in treatment of intractable seizures. Epilepsia 1992; 33: 841–851.

150

Mirski MA, Fisher RS. Electrical stimulation of the mammillary nuclei increases seizure threshold to pentylenetetrazol in rats. Epilepsia 1994; 35: 1309–1316. Ramsay RE, Uthman BM, Augustinsson LE, Upton AR, Naritoku D, Willis J, Treig T, Barolat G, Wernicke JF. Vagus nerve stimulation for treatment of partial seizures: 2. Safety, side effects, and tolerability. First International Vagus Nerve Stimulation Study Group. Epilepsia 1994; 35: 627–636. Salinsky MC, Uthman BM, Ristanovic RK, Wernicke JF, Tarver WB. Vagus nerve stimulation for the treatment of medically intractable seizures. Results of a 1-year open-extension trial. Vagus Nerve Stimulation Study Group. Arch Neurol 1996; 53: 1176–1180. Uthman BM, Wilder BJ, Penry JK, Dean C, Ramsay RE, Reid SA, Hammond EJ, Tarver WB, Wernicke JF. Treatment of epilepsy by stimulation of the vagus nerve. Neurology 1993; 43: 1338–1345. Van Buren JM, Wood JH, Oakley J, Hambrecht F. Preliminary evaluation of cerebellar stimulation by double-blind stimulation and biological criteria in the treatment of epilepsy. J Neurosurg 1978; 48: 407–16. Velasco M, Velasco F, Velasco AL, Boleaga B, Jimenez F, Brito F, Marquez I. Subacute electrical stimulation of the hippocampus blocks intractable temporal lobe seizures and paroxysmal EEG activities. Epilepsia 2000; 41: 158–169. Velasco F, Velasco M, Velasco AL, Jiminez F, Marquez J, Rise M. Electrical stimulation of the centromedian thalamic nucleus in the control of the seizures: long-term studies. Epilepsia 1995; 36: 63–71. Wright GD, McLellan DL, Brice JG. A double-blind trial of chronic cerebellar stimulation in twelve patients with severe epilepsy. J Neurol Neurosurg Psychiatry 1984; 47: 769–74. Zumsteg D, Lozano AM, Wennberg RA. Rhythmic cortical EEG synchronization with low frequency stimulation of the anterior and medial thalamus for epilepsy. Clin Neurophysiol 2006; 117: 2272–2278.

5.3 Deep Brain Stimulation for Parkinson’s Disease: Efficacy of Deep Brain Stimulation in Parkinson’s Disease Details of Studies The STN was identified as a surgical target for the control of symptoms of PD as the result of findings in pre-clinical primate research. Three studies are considered here. The first is a landmark as it was the first large follow-up study evaluating the efficacy of STN DBS in PD after 5 years and was carried out between 1993 and 1997 in Grenoble, France (Krack et al., 2003). The second is a randomized-pairs trial carried out between 2001 and 2004 in Germany and Austria comparing STN DBS plus medical management with best medical management (Deuschl et al., 2006). The third is the first published RCT of DBS versus best medical therapy for advanced PD (Weaver et al., 2009).

Study References Main Studies Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Schäfer H, Bötzel K, Daniels C, Deutschländer A, Dillmann U, Eisner W, Gruber D, Hamel W, Herzog J, Hilker R, Klebe S, Kloss M, Koy J, Krause M, Kupsch A, Lorenz D, Lorenzl S, Mehdorn HM, Moringlane JR, Oertel W, Pinsker MO, Reichmann H, Reuss A, Schneider GH, Schnitzler A, Steude U, Sturm V, Timmermann L, Tronnier V, Trottenberg T, Wojtecki L, Wolf E, Poewe W, Voges J, for the German Parkinson Study Group, Neurostimulation Section. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006; 355: 896–908. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, Koudsie A, Limousin PD, Benazzouz A, LeBas JF, Benabid AL, Pollak P. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003; 349: 1925–1934. Weaver FM, Follett K, Stern M, Hur K, Harris C, Marks WJ Jr, Rothlind J, Sagher O, Reda D, Moy CS, Pahwa R, Burchiel K, Hogarth P, Lai EC, Duda JE, Holloway K, Samii A, Horn S, Bronstein J, Stoner G, Heemskerk J, Huang GD; CSP 468 Study Group. Bilateral deep brain stimulation vs best medical therapy for patients with advanced Parkinson disease: a randomized controlled trial. JAMA 2009; 301: 63–73.

Related References Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the primate. Mov Disord 1991; 6: 288–292. Ballantyne HT Jr, Cassidy WL, Flanagan NB, Marino R Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967; 26: 488–495. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesion of the subthalamic nucleus. Science 1990; 249: 1436–1438. Witt K, Daniels C, Reiff J, Krack P, Volkmann J, Pinsker MO, Krause M, Tronnier V, Kloss M, Schnitzler A, Wojtecki L, Bötzel K, Danek A, Hilker R, Sturm V, Kupsch A, Karner E, Deuschl G. Neuropsychological and psychiatric changes after deep brain stimulation for Parkinson’s disease: a randomised, multicentre study. Lancet Neurol 2008; 7: 605–614.

Grenoble 5-Year Follow-up Study Study Design Follow-up study of a surgical series.

151

Class of evidence

III

Randomization

None

Number of patients

Length of follow-up

5 years Primary endpoints: Activities of daily living (ADL) Motor examination Secondary endpoints: Tremor, rigidity, limb akinesia, speech, postural stability, gait, dyskinesia Neuropsychological testing Depression and dementia assessments Dopaminergic medication dosage Requirements Stimulation settings

Number of centres

Stratification

None

This was the first major paper looking at outcome of STN stimulation in PD with long-term follow-up. Forty-nine consecutive patients were assessed (there is no control group). It is important to look at results both ‘on’ and ‘off’ medication. Good results ‘off’ medication can lead to reduction in levodopa and, therefore improvements, in levodopa side effects such as dyskinesias. This study documents the adverse effects of STN stimulation, in particular the adverse neuropsychological effects that have led some clinicians to prefer globus pallidus DBS. Inclusion criteria: clinical diagnosis of PD; severe levodopa-related complications despite optimal medication; age <70 years; bilateral STN stimulation intended. Exclusion criteria: surgical contraindications to DBS (e.g. pacemaker); dementia or psychiatric illness.

Outcome Measures Primary Endpoints Scores on part II (ADL) and part III (motor scores) on the Unified Parkinson’s Disease Rating Scale (UPDRS) at 1, 3, and 5 years. Secondary Endpoints Subscores on part III (limb tremor, limb rigidity and limb akinesia, speech, postural stability and gait) and part IV (dyskinesias) on the UPDRS. Schwab and England Activities of Daily Living Scale. Neuropsychological tests. Mattis dementia rating scale. Beck Depression Inventory. Dose of dopaminergic treatment. Stimulation settings.

Results On average, patients had a disease duration of 14.6 years. With stimulation in the off medication state, the total part III (motor) UPDRS score improved by 66% at 1 year, 59% at 3 years, and 54% at 5 years. At 5 years, tremor improved by 75%, rigidity by 71%, and akinesia by 49%. Speech initially improved but returned to baseline at 5 years. Postural stability and gait also improved. ADL (part II UPDRS) improved by 66% at 1 year, 51% and 49% at 3 and 5 years respectively (this reduction at 5 years being significant). Five years after surgery, the Schwab and England score showed that most patients were independent (73%) compared to most being dependent on a carer pre-operatively (33% independent). The incidence of painful dystonia off medication also dramatically reduced after surgery. There was generally worsening in the ‘on’ medication scores with stimulation, as expected. There was no change in the depression scores, but dementia scores worsened over the 5 years. The authors put this down to deterioration due to the degenerative nature of the disease. Probably the most significant results are that levodopa requirement reduced from a mean of 1409 mg at baseline to 518 mg at 5 years (p <0.001). Eleven patients were able to stop levodopa altogether. Adverse events included three deaths (one intracranial haemorrhage, one myocardial infarction 11 months after surgery, and one suicide 6 months after surgery). Other adverse events included dementia and weight gain.

Conclusions 152

Bilateral STN DBS in PD results in marked improvement in motor function whilst patients are off medication and in dyskinesia whilst on medication.

Germany/Austria randomized-pairs trial (Deuschl et al., 2006) Study Design Multi-centre unblinded randomized-pairs trial Class of evidence

Randomization

Bilateral STN-DBS + medical therapy versus best medical treatment

Number of patients

156

Length of follow-up

6 months Primary endpoints: Quality of life Severity of symptoms Secondary endpoints: Dyskinesia ADL Neuropsychiatric function Health-related quality of life Adverse events

Number of centres

Stratification

None

Inclusion criteria: PD for at least 5 years; <75 years of age; ADL impaired by motor symptoms or dyskinesias despite optimal medical therapy; informed consent. Exclusion criteria: dementia; psychiatric symptoms; contraindications to surgery. Patients enrolled in pairs with randomization to each arm of the trial within 6 weeks. Analysis was on an intention-to-treat basis.

Outcomes Primary Endpoints All primary endpoints were calculated as changes from baseline to 6 months. Quality of life assessed by Parkinson’s Disease Questionnaire (PDQ-39). Severity in symptoms assessed by Unified Parkinson’s Disease Rating Scale Part III (UPDRS-III) whilst the patients were not on medication. Secondary Endpoints Dyskinesia and ADL assessed using the Unified Parkinson’s Disease Rating Scale Part II (UPDRS-II). Neuropsychiatric function: Montgomery and Asberg Depression Rating Scale and the Brief Psychiatric Rating Scale. Health-related quality of life assessed using Medical Outcomes Study 36-item Short-Form General Health Survey (SF-36).

Results

An additional analysis on a per-protocol basis gave a favoured outcome in terms of quality of life to 75% of the STN DBS patients versus 25% for the best medical management group (p <0.001). STN DBS was associated with a 25% improvement in the PDQ-39 summary index and a 22% improvement in the SF-36 score. ADL were improved by 39% in the STN DBS group but only by 5% in the best medical management group. Dyskinesia improved in the STN DBS group but not in the best medical management group. Adverse events were significantly greater in the DBS STN group (12.8%) compared to the best medical 153

management group (3.8%, p = 0.04). There were three deaths in the DBS STN group: one due to an intracerebral haematoma; one from pneumonia; and one suicide. There was one death in the best medical group from a motor vehicle accident whilst driving during a psychotic episode.

Conclusion DBS STN resulted in a significant and meaningful improvement in quality of life compared to best medical management.

Veterans Affairs and University Hospitals US Trial (Weaver et al., 2009) Study Design PRCT of DBS versus best medical therapy Class of evidence

Randomization

Non-blinded DBS versus medical. DBS further randomized into GPi or STN (patients blinded to stimulation site)

Number of patients

255 (134 medical, 60 STN, 61 GPi)

Length of follow-up

6 months Primary endpoint: Time spent in ‘on’ state without disabling dyskinesias Secondary endpoints: Motor function Quality of life Neurocognitive function Adverse events

Number of centres

Stratification

Study site Patient age (<70 versus ≥70 years)

This was the first large, multi-centre, randomized controlled, blinded trial comparing risks and benefits of DBS versus best medical therapy. Comparison of GPi versus STN targets expected after 2-year follow-up. DBS significantly improved duration of ‘on’ time without troubling dyskinesias (4.6 h/day) although there were a significant number of adverse events, greater in the DBS group. Inclusion criteria: idiopathic PD; Hoehn and Yahr ≥stage 2 off medication; responsive to levodopa; persistent disabling symptoms despite medication (e.g. motor fluctuations, dyskinesia); ≥3/24 h poor symptom control; stable medical therapy ≥1 month; ≥ 21 years of age. Exclusion criteria: atypical syndromes; previous surgery for PD; contraindications to surgery; active alcohol or drug abuse; dementia; pregnancy.

Outcome Measures Primary Endpoint Baseline to 6-month change in time spent in the ‘on’ state without troubling dyskinesia. Secondary Endpoints Hoehn and Yahr and Schwab and England scales. Stand-walk-sit test. UPDRS. PDQ-39. Medication usage. Neurocognitive battery including: Mattis Dementia Rating Scale, standardized tests of attention, memory, 154

verbal, executive functioning, language. ‘On’ and ‘off’ time assessed by self-report motor diaries. Adverse events.

Results The baseline characteristics did not differ between groups except the best medical therapy patients were treated with PD medications for longer (12.6 versus 10.8 years, p = 0.01) and had a lower working memory index (97.3 versus 101.2, p = 0.02). DBS patients gained 4.6 h/day of ‘on’ time without troubling dyskinesia compared to 0 h/day in the medical arm (p <0.001). ‘Off’ time decreased by 2.4 h/day in the DBS group compared to 0 h in the medical arm (p <0.001). Similar changes were experienced by patients over 70 years of age. Motor function improved by 12.3 points in the DBS group (in the off medication state) compared to 1.7 in the medical arm (p <0.001). ADL and complications of therapy were also significantly better in the DBS group. Stand–walk–sit test improved by 9.0 s in the DBS group compared to a 0.2 s worsening at 6 months in the medical arm (p = 0.046). Medications decreased by 296 mg levodopa equivalent in the DBS group and increased in the medical arm. Quality of life improved in seven of eight sections of the PDQ-39 subscales in the DBS group whereas there was little change in the medical arm. Some of the neurocognitive measures showed slight but significant worsening in the DBS group (working memory, processing speed, phonemic fluency, and delayed recall on the Brief Visuospatial Memory Test). However, there were no significant differences on the scales of depression, dementia, and most of the measures assessing language, executive functioning and learning, and memory functioning. Regarding adverse events, the DBS group received a significantly greater number of falls, gait disturbance, depression, and dystonia. In the DBS group, surgical site infection was 9.9% and surgical site pain 9.0%. One DBS patient died as a result of cerebral haemorrhage related to the procedure. The overall incidence risk of a serious adverse event was 3.8 times higher in the DBS group and these included psychiatric disorders. However, 99% of these were resolved at 6 months.

Conclusions DBS is more effective than best medical therapy in alleviating disability in moderate to severe PD patients with levodopa-responsive motor complications and no significant cognitive impairment.

Critique PD is a degenerative condition and available medications are aimed at symptomatic control with none being available to reverse the on-going neuronal loss. DBS is intended to improve the symptoms of the disease and therefore improve quality of life. The identification of the subthalamic nucleus (STN) as a target for DBS in the treatment of PD was the direct result of findings in primate models (Aziz et al., 1990; Bergman et al., 1990). Pollak et al. were the first to report STN DBS for the treatment of a patient with PD (Pollak et al., 1993). Although there are many studies looking at STN stimulation in PD, the study by Krack et al. is the longest proof-of-principle follow-up study carried out. The trial by Deuschl et al. was the first study comparing DBS + medication to medical therapy alone, and the Weaver et al. study is the largest to date and the first to randomize to patients to DBS versus best medical therapy. This latter study also stands out amongst the literature because the clinical outcome was measured by blinded neurologists and because of the extensive quality of life and neuropsychological data measured. The UK ‘PDSurg’ trial is currently underway with patients being randomized over the first 12 months as to whether the neurostimulators are activated or not. The results of this trial are also eagerly awaited. Krack et al. showed that bilateral STN stimulation is an effective treatment for all of the motor parts of the UPDRS score off medication, except speech. It allows patients to significantly reduce their dopaminergic medication and therefore reduces the incidence and severity of dyskinesias. There are few adverse effects, but significantly, dementia appears to be increased, irreversibly in two patients. There is also an increase in other neuropsychiatric symptoms with stimulation, including hypomania and depression. In the Weaver et al. study, there was an increase in depression, confusion, and anxiety in the DBS group but it should be noted that these neuropsychiatric complications also occur in the medical group and that the increases were small. STN stimulation is generally used in the situation where the patient already responds to levodopa (as it has a similar effect) and is not effective if the patient is levodopa unresponsive. Therefore the aim of stimulation is to allow a reduction in medication and an improvement in the side effects—mainly the disabling dyskinesias that occur in the ‘on’ state. The study by Krack et al. showed that STN stimulation significantly improves the motor symptoms of PD, leads to medication reduction, and improvements in ADL. The main limitation of the study is the lack of any randomization and so it is not possible to say whether STN stimulation is better than medication, as there is no official control arm. The rationale for proposing that STN stimulation represents an improvement over medication is that the investigators are able to study the patients off all treatment (drugs and medication). Another criticism is that the assessments were unblinded. In a single group, this could lead to recorder bias. Nevertheless, this study represents the best long-term outcome data available to date. Another important factor that should be borne in mind when considering the results of this study is that when it was started there was a tendency to operate on younger patients with severe disease. 155

The trial by Deuschl et al. stands out as a landmark in neurostimulation studies for being the first ‘large’ randomized trial in this area. Although the trial may be criticized for being unblinded, the authors indicate that the need to adjust medication regimens in DBS patients makes it impossible to carry out a blinded study with placebo stimulation. The use of sham surgery was deemed unethical because of the potential complications. The use of a paired-analysis allowed relatively small numbers of patients in order to obtain adequate power and it is nonetheless a well-designed and well-executed study. It is of note that the authors chose not to use motor function as a primary outcome and instead used quality of life. The efficacy of DBS needs to account for not only motor function, but neurobehavioural effects, adverse effects, and surgical complications, all of which affect quality of life. The authors of the trial went on to evaluate the neurobehavioural effects at 6 months in 60 patients receiving STN DBS and 63 patients receiving best medical management (Witt et al., 2008). A wide range of neuropsychiatric tests were used and the primary outcome chosen was cognitive functioning. Secondary outcomes included effects on neurobehavioural variables (executive functioning, depression, anxiety, manic symptoms) and quality of life. Although there appeared to be a selective decrease in frontal executive functioning and verbal fluency, the authors concluded that these did not affect quality of life. The study by Weaver et al. stands out for the reasons given earlier—i.e. the size of the study, rigorous randomization, blinded assessments, and detailed neuropsychological outcomes. This study has conclusively demonstrated that DBS improves motor outcome without disabling dyskinesias at a small cost of minor dysexecutive functions but as these are limited to certain categories such as working memory and phonemic fluency, they do not affect quality of life. Probably the most useful aspect of this study is the reporting of adverse events which were much higher in the DBS group. Although the majority of these were resolved at 6 months, there was one death related to cerebral haemorrhage. Despite this, it is difficult to comment on death rates in series <1000 as the reported death rates in DBS are around 0.5–1% in previous studies. An important aspect of this trial is the inclusion of GPi patients in a randomized fashion and the results of GPi versus STN as a target in PD will be very interesting. The potential neuropsychological and psychiatric sequelae of DBS of the STN have been a source of controversy and concern. However, early series of patients undergoing STN DBS revealed inconsistent effects on cognition and neuropsychological sequelae. A comprehensive review of the literature by Woods et al. found that the most common findings were increased verbal fluency and improvements in self-reported symptoms of depression: approximately 69% of studies including scores of verbal fluency reported a significant post-operative reduction (Woods et al., 2002). There was no consistency in reports on post-operative changes in frontal/executkve function, cognition, memory, or attention and it appears that changes in these modalities occur in 1–2% of patients only (Woods et al., 2002). However, cognitive deficits and mood disturbance appear to be more frequently reported in patients receiving DBS of the STN than in patients undergoing pallidal DBS, a trend which has also been reflected in direct comparisons of these two targets (Volkmann et al., 2001; Walter and Vitek, 2004; Rodriguez-Oroz et al., 2005). Deuschl’s group have reported the results of a randomized trial of STN DBS versus best medical management in advanced PD to assess the neuropsychological and psychiatric effects of DBS for PD (Witt et al., 2008). They found that there was no decline in cognition but there was a selective decrease in frontal cognitive functions. As DBS has become more widespread, more and older patients are receiving the treatment. The initial anxieties regarding the ‘limbic’ side effects of stimulation including cognitive decline, behavioural changes, and psychosis (as suggested in the Krack et al. study with the increased incidence of dementia) are becoming more pronounced as later studies have longer follow-up and include older patients. The pendulum that had largely swung towards STN stimulation in favour of globus pallidus stimulation is perhaps starting to swing back, particularly in older patients or those with a hint of psychotic or dementia-related issues. Nonetheless there are still many unanswered questions regarding STN DBS. For example, although DBS has been used when medical therapy has failed, it is possible that the earlier use of neurosurgery may prevent deterioration. Schüpbach et al. carried out a pilot RCT of 20 patients in which ten were randomized to early bilateral STN DBS and ten to optimal medical management (Schüpbach et al., 2006). The authors found a statistically significant benefit for DBS in terms of quality of life (p <0.05) and severity of symptoms (p <0.001) over a period of 18 months, thus supporting the idea that there may be potential benefits of STN stimulation as a therapeutic option earlier in PD. STN stimulation has become a popular treatment with >40,000 implants worldwide. These studies have helped to confirm that it is a useful treatment with lasting effects, but importantly have also identified the negative effects of stimulation. Thus, there are a number of ongoing studies comparing STN to GPi stimulation (including Weaver et al. part two) and other large, randomized trials of DBS such as the ‘PDSurg’ trial.

References Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the primate. Mov Disord 1991; 6: 288–292. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesion of the subthalamic nucleus. Science 1990; 249: 1436–1438. Pollack P, Benabid AL, Gross C, Gao DM, Laurent A, Benazzouz A, Hoffman D, Gentil M, Perret J. Effects of the stimulation of the subthalamic nucleus in Parkinson disease. Revue Neurologique 1993; 149: 175–176. Rodriguez-Oroz MC, Obeso JA, Lang AE, Houeto JL, Pollak P, Rehncrona S, Kulisevsky J, Albanese A, Volkmann J,

156

Hariz MI, Quinn NP, Speelman JD, Guridi J, Zamarbide I, Gironell A, Molet J, Pascual-Sedano B, Pidoux B, Bonnet AM, Agid Y, Xie J, Benabid AL, Lozano AM, Saint-Cyr J, Romito L, Contarino MF, Scerrati M, Fraix V, Van Blercom N. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005; 128: 2240–2249. Schüpbach WMM, Maltête, D, Houeto JL, Tezenas du Montcel S, Mallet L, Welter ML, Gargiulo M, Béhar C, Bonnet AM, Czernecki V, Pidoux B, Navarro S, Dormont D, Cornu P, Agid Y. Neurosurgery at an earlier stage of Parkinson disease. Neurology 2007; 68: 267–271. Volkmann J, Allert N, Voges J, Weiss PH, Freund HJ, Sturm V. Safety and efficacy of pallidal or subthalamic nucleus stimulation in advance PD. Neurology 2001; 56: 548–551. Walter BL, Vitek JL. Surgical treatment for Parkinson’s disease. Lancet Neurol 2004; 3: 719–728. Witt K, Daniels C, Reiff J, Krack P, Volkmann J, Pinsker MO, Krause M, Tronnier V, Kloss M, Schnitzler A, Wojtecki L, Bötzel K, Danek A, Hilker R, Sturm V, Kupsch A, Karner E, Deuschl G. Neuropsychological and psychiatric changes after deep brain stimulation for Parkinson’s disease: a randomised, multicentre study. Lancet Neurol 2008; 7: 605–614. Woods SP, Fields JA, Tröster AI. Neuropsychological sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a critical review. Neuropsychology Rev 2002; 12: 111–126.

5.4 Deep Brain Stimulation for Parkinson’s Disease: Efficacy of Subthalamic versus Pallidal Deep Brain Stimulation in Parkinson’s Disease Details of Study Deep brain stimulation of the Globus pallidus interna (GPi) has always been a viable alternative to subthalamic nucleus stimulation in the treatment of PD. Indeed, some proponents of GPi stimulation have claimed (and still do) that it is a better target than STN for a variety of reasons, but particularly because it is not associated with the limbic or neuropsychiatric complications that can occur with the latter. Whilst many centres have tended to prefer STN as a target, the debate over which is better has now re-emerged. In a randomized, prospective study of 23 PD patients by Anderson et al. (2005), itself an extension of an earlier study by the same group (Burchiel et al., 1999), outcomes were similar after subthalamic and pallidal stimulation. We herein discuss the landmark study by Follett et al., for the Veteran Affairs Cooperative Studies Program, who undertook a double-blinded, randomized, prospective, multicentre trial in the US of 299 PD patients who received either bilateral GPi or STN stimulation (Follett et al., 2010) to address this question.

Study References Main Study Follett KA, Weaver FM, Stern M, Hur K, Harris CL, Luo P, Marks WJ Jr, Rothlind J, Sagher O, Moy C, Pahwa R, Burchiel K, Hogarth P, Lai EC, Duda JE, Holloway K, Samii A, Horn S, Bronstein JM, Stoner G, Starr PA, Simpson R, Baltuch G, De Salles A, Huang GD, Reda DJ, for the CSP (Veteran Affairs Cooperative Studies Program) 468 Study Group. Pallidal versus subthalamic deep-brain stimulation for Parkinson’s disease. N Engl J Med 2010; 362: 2077– 2091.

Related References Anderson VC, Burchiel KJ, Hogarth P, Favre J, Hammerstad JP. Pallidal vs subthalamic nucleus deep brain stimulation in Parkinson disease. Arch Neurol 2005; 62: 554–560 Burchiel K, Anderson VC, Favre J, Hammerstad JP. Comparison of pallidal and subthalamic nucleus deep brain stimulation for advanced Parkinson’s disease: results of a randomized, blinded pilot study. Neurosurgery 1999; 45(6): 1375–1382

Study Design Double-blind (assessing neurologist and patient), prospective, randomized trial. Class of evidence

Randomization

Bilateral GPi or STN stimulation

Number of patients

299

Length of follow-up

24 months

Number of centres

Inclusion criteria: age >21 years old; severity > Stage 2 (Hoehn & Yahr scale) off medication; responsive to levodopa; persistent disabling motor symptoms; >3 h per day of poor motor function or symptom control. Major exclusion criteria: atypical syndromes; active substance abuse; dementia; pregnancy; surgical contraindications.

Outcome Measures 157

Primary Endpoint Change in motor function (UPDRS Pt III motor score) off medications at 24 months compared to baseline

Secondary Endpoints Neurocognitive outcomes including depression. Levodopa requirement. ADL (UPDRS-II). Quality of life—PDQ-39 scores.

Results STN and pallidal stimulation produced equivalent motor improvements. Levodopa requirement was reduced most by STN stimulation compared to the GPi group (165 mg difference, p = 0.02). There was no significant difference in PDQ-39 scores between groups. Adverse events occurred in 56% of STN cases versus 51% of pallidal stimulation cases.

Conclusions There are minimal differences to suggest superiority between GPi and STN as targets for DBS in PD, in terms of motor function. These results do suggest a greater reduction in levodopa requirements with STN stimulation although this was offset by an increase in depression scores and decrease in an index of cognitive functioning.

Critique Achieving double blinding is notoriously difficult in surgical trials. The authors have achieved a highly satisfactory blinding in this study of both patients and assessors. Although it is very obvious to patients or assessors whether a surgical intervention has been performed, in this case all patients underwent deep brain stimulation surgery. Only intracranial location of the electrodes differed which would not be evident externally. This is an important strength of this paper. Another strength is the recognition of the non-motor outcomes of stimulation. Although there was no significant difference in motor outcome between stimulation locations, the trial provides important data on neurocognitive and quality of life measures. PDQ-39 scores did not differ significantly between groups. The study did show, however, that STN stimulation was associated with a worsening in depression score. It also found a worsening in a single index of cognition, the processing speed index, after STN stimulation. Inadvertent effects on the limbic and cognitive divisions of the STN have blamed previously for mood and neurocognitive changes after subthalamic stimulation and this study helps to confirm that this is a likely property of this target. There is concern over the long-term efficacy of pallidal stimulation in PD. Pallidal stimulation efficacy has been reported to decrease with time, with successful salvage by changing to STN stimulation. The follow-up period of two years in this study is insufficient to fully reassure clinicians over this issue. Further follow-up of this cohort may yet answer this issue in the long term.

5.5 Deep Brain Stimulation for Parkinson’s Disease: Early Subthalamic Stimulation in Parkinson’s Disease Details of Study DBS is employed late in PD, typically after more than a decade of disease duration, when there are significant limitations in quality of life, social function, and/or professional function. At this late stage, other symptoms emerge that are unresponsive to medical therapy. PD confers a significant cost on patients and society, in terms of healthcare costs, reduced productivity, and quality of life. There is an argument for institution of DBS earlier in the disease natural history when patients are optimally responsive to dopamine to maximize the benefits in motor symptoms and quality of life from neurostimulation, rather than later when cognitive or other symptoms counteract motor improvements. Judging how early to initiate neurostimulation is critical as the surgery itself has small but important risks, and follow-up is more intense and costly. The EARLYSTIM Study Group undertook a single-blinded, 158

prospective, multi-centre, RCT of early STN stimulation compared to medical therapy alone in France and Germany in adults with PD <60 years old. This study follows a pilot study by Schuepbach et al. in which a 24% improvement in PDQ-39 questionnaire score was found in patients with early and mild motor complications (Schuepbach et al., 2007).

Study References Main Study Schuepbach WMM, Rau J, Knudsen K, Volkmann J, Krack P, Timmerman L, Hälbig TD, Hesekamp H, Navarro SM, Meier N, Falk D, Mehdorn M, Paschen S, Maarouf M, Barbe MT, Fink GR, Kupsch A, Gruber D, Schneider GH, Seigneuret E, Kistner A, Chaynes P, Ory-Magne F, Brefel Courbon C, Vesper J, Schnitzler A, Wojtecki L, Houeto JL, Bataille B, Maltête D, Damier P, Raoul S, Sixel-Doering F, Hellwig D, Gharabaghi A, Krüger R, Pinsker MO, Amtage F, Régis JM, Witjas T, Thobois S, Mertens P, Kloss M, Hartmann A, Oertel WH, Post B, Speelman H, Agid Y, Schade-Brittinger C, Deuschl G, for the EARLYSTIM Study Group. Neurostimulation for Parkinson’s disease with early motor complications. N Engl J Med 2013; 368: 610–622.

Related Reference Schuepbach WM, Maltete D, Houeto JL, du Montcel ST, Mallet L, Welter ML, Gargiulo M, Béhar C, Bonnet AM, Czernecki V, Pidoux B, Navarro S, Dormont D, Cornu P, Agid Y. Neurosurgery at an earlier stage of Parkinson’s disease: a randomized, controlled trial. Neurology 2007; 68: 267–271.

Study Design Single-blinded (outcome-rating neurologist, not patient), prospective, multi-centre, randomized trial. Class of evidence

Randomization

Bilateral STN stimulation plus medical therapy versus medical therapy alone

Number of patients

251

Length of follow-up 24 months Centres

Multi-centre (17) within France and Germany

Blinding

Assessing neurologist blinded (not patients)

Outcome assessment Patient-reported questionnaire; neurologist review of standardized patient video recordings

Inclusion Criteria Age 18–60 years. Disease duration 4 years or more. Disease severity <Stage 3 (Hoehn and Yahr scale) on medication. Dopamine response of motor signs >50% (as per UPDRS-III). Motor fluctuations or dyskinesia duration <3 years. UPDRS-II ADL score >6 at worst despite medication Social and occupational mild-to-moderate function (51–80% on Social & Occupational Functioning Assessment Scale)

Major Exclusion Criteria Dementia. Major depression with suicidal ideation. Acute psychosis.

Outcome Measures Primary Endpoint PDQ-39 summary index score at 0 versus 24 months.

Secondary Endpoints Motor disability. ADLs. UPDRS-II, -III, -IV levodopa-induced motor complications. Time with good mobility. Time with no dyskinesia.

159

Results

Mean disease duration was 7.5 years before surgery. Results of stimulation with medication versus medication alone were superior. Primary endpoint: PDQ-39 between group difference was 8 points in favour of stimulation. Secondary endpoints: motor disability, ADLs, levodopa-induced motor complications, time with good mobility and no dyskinesia all improved in favour of stimulation.

Conclusions STN stimulation with medical therapy is superior to medical therapy alone at an earlier stage of PD, with respect to motor signs and quality of life.

Critique This is an important trial which is needed to address the difficult but critical question of when to subject patients to neurosurgery during a progressive neurodegenerative disease. This is especially pertinent when the period in question is whilst medical therapy is still effective. The trial has several strengths and weaknesses. The primary outcome measure was a patient-reported questionnaire; however, the patients were not blinded to the intervention. This is an important potential source of bias. Other studies have overcome this limitation using implantation in both experimental and control groups with sham stimulation used in the latter group until the end of the trial period. The improvements in PDQ-39 were, however, corroborated with assessment of standardized videos by blinded neurologists, whereby significant improvements in secondary outcome measures of motor function were verified in the stimulation group. Two patients in the stimulation and one in the medical group committed suicide. Given the relatively small numbers in the trial compared to number of suicides, it is not possible to draw conclusions regarding suicide risk between the two groups. Clinical application of the results of this trial may be limited. PD patients fitting the criteria for the study are in the minority with only 11% of PD patients being diagnosed before 60 years of age (Kleiner-Fisman et al., 2006) and 30% having dementia (Deuschl et al., 2006). Balancing the costs and availability of surgery, hardware, and medical follow-up versus years of lost productivity plus healthcare costs of motor symptoms is another factor. For patients with high function, no dementia, and good response to levodopa, this study provides important data on which to base decisions regarding early STN stimulation.

References Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Schäfer H, Bötzel K, Daniels C, Deutschländer A, Dillmann U, Eisner W, Gruber D, Hamel W, Herzog J, Hilker R, Klebe S, Kloss M, Koy J, Krause M, Kupsch A, Lorenz D, Lorenzl S, Mehdorn HM, Moringlane JR, Oertel W, Pinsker MO, Reichmann H, Reuss A, Schneider GH, Schnitzler A, Steude U, Sturm V, Timmermann L, Tronnier V, Trottenberg T, Wojtecki L, Wolf E, Poewe W, Voges J, for the German Parkinson Study Group, Neurostimulation Section. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006; 355: 896–908. Kleiner-Fisman G, Herzog J, Fisman DN, Tamma F, Lyons KE, Pahwa R, Lang AE, Deuschl G. Subthalamic nucleus deep brain stimulation: summary and meta-analysis of outcomes. Mov Disord 2006; 21(Suppl 14): S290–S304.

5.6 Pallidal Stimulation for Dystonia: Generalized and Segmental Dystonia Details of Study The first and most important study of DBS for dystonia was carried out by the Deep-Brain Stimulation for Dystonia Study Group in Germany, Austria, and Norway between 2002 and 2004. This study evaluated patients with primary generalized or segmental dystonia with bilateral stimulation of the ventral border of the globus pallidus internus (GPi) being compared to sham stimulation for the first 3 months after surgery. The study concludes that stimulation leads to a significant clinical improvement. 160

Study References Main Study Kupsch A, Benecke R, Müller J, Trottenberg T, Schneider GH, Poewe W, Eisner W, Wolters A, Müller JU, Deuschl G, Pinsker MO, Skogseid IM, Roeste GK, Vollmer-Haase J, Brentrup A, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Naumann M, Volkmann J; Deep-Brain Stimulation for Dystonia Study Group. Pallidal deepbrain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006; 355: 1978–1990.

Related References Mueller J, Skogseid IM, Benecke R, Kupsch A, Trottenberg T, Poewe W, Schneider GH, Eisner W, Wolters A, Müller JU, Deuschl G, Pinsker MO, Roeste GK, Vollmer-Haase J, Brentrup A, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Naumann M, Volkmann J, for the Deep-Brain Stimulation for Dystonia Study Group. Pallidal deep brain stimulation improves quality of life in segmental and generalized dystonia: results from a prospective, randomized sham-controlled trial. Mov Disord 2008; 23: 131–134. Volkmann J, Wolters A, Kupsch A, Müller J, Kühn AA, Schneider GH, Poewe W, Hering S, Eisner W, Müller JU, Deuschl G, Pinsker MO, Skogseid IM, Roeste GK, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Classen J, Naumann M, Benecke R, for the DBS study group for dystonia. Pallidal deep brain stimulation in patients with primary generalised or segmental dystonia: 5-year follow-up of a randomised trial. Lancet Neurol 2012; 11(12): 1029– 1038.

Study Design This study looks at patients with primary generalized or segmental dystonia. Class of evidence

Randomization Double-blind, sham controlled Followed by 6 months of open-label treatment leading to a total of 6 months of stimulation Number of patients

Length of follow-up

3 months Comparisons were also made after 6 months of stimulation, i.e. at 6 months in the stimulation group, 9 months in the sham group

Number of centres

10 (all European)

Stratification

None

Previous studies are limited to case reports or case series but they universally show improvement in dystonia with bilateral stimulation. The Burke–Fahn–Marsden Dystonia Rating Scale (BFMDRS) was significantly lower in the stimulation group. Drug use and depression also decreased in the stimulation group. The most important adverse event was implant infection. Inclusion criteria ‘Marked disability’ owing to primary generalized or segmental dystonia

Exclusion criteria Surgical contraindications

Age 14–75 years

Previous brain surgery Cognitive impairment (<120 on the Mattis Dementia Rating Scale) Moderate to severe depression (>25 on the Beck Depression Inventory) Marked brain atrophy on CT or MRI

Outcome Measures Primary Endpoints Motor score of the BFMDRS at 3 months and after 6 months of continuous stimulation. The primary endpoint of the 5-year follow-up study was the change in dystonia severity at 3 and 5 years on the BFMDRS motor score compared to baseline and 6-month follow-up, as analysed on an intention-to-treat basis. 161

Secondary Endpoints Quality of life assessed with the SF-36 questionnaire. ADL and Disability Score on the BFMDRS. Described as ‘exploratory endpoints’; Severity of dystonia and pain using a visual analogue score. Chronometric measurements of walking and finger tapping. Cognitive and mental status using the Mattis Dementia Rating Scale, the Brief Psychiatric Rating Scale, the Beck Depression Inventory, and the Beck Anxiety Inventory.

Results Sixteen patients had segmental dystonia and 24 had primary generalized dystonia. At 3 months, the severity scores were significantly lower in the neurostimulated group compared to the sham group (p <0.001). Specifically, the movement score reduced by 39.3% in the neurostimulated group, compared to 4.9% in the sham group. Fifteen patients in the stimulated group fulfilled the authors’ criterion for a ‘positive response to stimulation’ (>25% improvement in movement score) compared to three in the sham group. Similarly, disability scores in the stimulated group improved by 37.5% compared to 8.3% in the sham group. Neurostimulation was significantly better in all symptom subscores of the BFMDRS and most of the disability scores. Quality of life, as assessed with the SF-36 questionnaire, improved by 29.8% in the stimulated group compared to 11.4% in the sham group. With the open-label extension and stimulation in all patients, further improvements occurred, although the extra 3 months in the stimulated group did not produce significant changes. Overall, the severity of the dystonia, as assessed by the movement score, improved by 75% in five patients, >50% in 18 patients, and >25% in 30 patients. The presence of the DYT1 mutation in the Torsin A gene did not influence the degree of improvement in the movement score with stimulation. There was no significant difference between the segmental and generalized dystonia patients. Other results include a reduction in medication of 32.1% at 6 months, a significant reduction in depression but no changes on the other cognitive/psychiatric scores. There were six adverse events in the stimulated group compared to three in the sham group during the randomized phase of the study. Most were related to hardware complication such as infection. All adverse events resolved without permanent sequelae. There were 13 adverse events in the open-label phase, mostly related to stimulation. Some of these (such as dysarthria) were accepted as tolerable side effects. In the extension study, the 47.9% 6-month improvement seen in the first study was increased to a 61.1% improvement at 3 years and 57.8% at 5 years (n = 32 out of the original 40). These were significant improvements and the 3- and 5-year improvements were significantly better than the 6-month improvements. There were, however, 21 serious adverse events in the follow-up study and many of these were related to device breakage, malfunction, or infection.

Conclusions Neurostimulation of the GPi is an effective treatment for both primary generalized and segmental dystonia sustained and increased efficacy at 5 years. It is a safe procedure and should be considered as a first-line treatment.

Critique Sham-controlled surgical trials are relatively rare but neurostimulation lends itself to these types of trials as the sham group can still crossover to stimulation after the comparison period is over. This was one of the first trials of this nature in functional neurosurgery. Dystonia is a rare condition that is generally not very responsive to medical treatment. Initial case series of neurostimulation for dystonia looked encouraging. There are two factors that make a randomized trial more difficult. Firstly, the numbers are small. Therefore, the benefit has to be great to show that the treatment is statistically significantly different to controls. This trial was powered on the basis that 40 patients would be needed to show a 25% difference (90% power) with a 10% drop out rate and a 5% error. The other problem is that the effects of neurostimulation on the GPi are slow to act. It is very rare to get an instantaneous response (other than side effects) and in some patients, it can take months to get full benefit from stimulation. One criticism of the original study therefore, was that the study period may have been too short and an even greater difference may have been demonstrated with a longer study. However, this question was very well answered by the open-label follow-up study. Another criticism is that there were positive responses in the sham (control) group (three patients). If the improvement is due to stimulation, why should this be the case? One possibility is the ‘stun’ or lesional effect of electrode implantation that is certainly seen in subthalamic stimulation for PD. However, placebo probably plays a part also. A final criticism is that this trial involved both segmental and generalized dystonia rather than a single entity. The pathophysiology of dystonia is poorly understood and therefore this may or may not be a valid criticism.

Impact on Field This was the first major randomized trial in functional neurosurgery. Although many centres practised neurostimulation for dystonia before this study, it has provided adequate justification for the continuing use of this 162

technique.

Cervical Dystonia Details of Study Prior to 2007 there were numerous case reports of DBS for cervical dystonia and studies of pallidotomy and thalamotomy. This study was the first prospective multi-centre and single-blinded DBS study in cervical dystonia and therefore represents an improvement on previous reports.

Study References Main Study Kiss ZH, Doig-Beyaert K, Eliasziw M, Tsui J, Haffenden A, Suchowersky O; Functional and Stereotactic Section of the Canadian Neurosurgical Society; Canadian Movement Disorders Group. The Canadian multicentre study of deep brain stimulation for cervical dystonia. Brain 2007; 130(Pt 11): 2879–2886.

Related References Krauss JK, Pohle T, Weber S, Ozdoba C, Burgunder JM. Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet 1999; 354: 837–838. Yianni J, Bain PG, Giladi N, Auca M, Gregory R, Joint C, Nandi D, Stein J, Scott R, Aziz T. Globus pallidus internus deep brain stimulation for dystonic conditions: a prospective audit. Mov Disord 2003; 18: 436–42

Study Design Class of evidence

III

Randomization

Single blinded, non-randomized.

Number of patients

Length of follow-up

12 months Outcome measures were made at baseline, 6 months and 12 months

Number of centres

5 (all Canadian)

Stratification

None

This study looks at patients with primary medication-resistant, chronic cervical dystonia. All patients had DBS of the GPi and there were five males, five females. Previous studies are limited to case reports or case series but as with those in generalized dystonia, they universally show improvement in dystonia with bilateral stimulation. The Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) was significantly lower following DBS. Quality of life measures and depression scores also improved. Adverse events included stimulation-related dysphagia and dysarthria in two patients which resolved with stimulator adjustment. Significant decline in phonemic fluency and verbal memory were seen in two patients but these did not impact on daily living. Inclusion criteria Cervical dystonia of at least 5 years’ duration

Exclusion criteria Secondary causes

Initial response to botulinum A and/or B injection with subsequent failure

Psychiatric disturbances

Normal neurological examination (except dystonia)

Generalized dystonia

Normal MRI Normal cognitive function

Outcome Measures Primary Endpoint The severity subscale of the TWSTRS (assessed by two blinded neurologists).

Secondary Endpoints Disability and pain subscores of the TWSTRS. Beck Depression Inventory. SF-36. 163

Global assessment of change. Adverse events.

Results Median age was 57.5 years (range 47–64). Median disease duration was 16.5 years (range 5–28). Eight patients had associated tremor and one had a history of head injury (see ‘Critique’). The primary outcome measure (severity on the TWSTRS scale) improved from a mean (SD) of 14.7 (4.2) pre-operatively to 10.6 (4.8) at 6 months and 8.4 (4.4) at 12 months (p = 0.003). Similar improvements were demonstrated in the secondary outcome measures; disability scores improved from 14.9 (3.8) to 5.4 (7.0) and pain from 26.6 (3.6) to 9.2 (13.1). Both were significant (p <0.001). Beck depression scores improved from 14.2 (7.2) to 6 (3.5) (p <0.001). Quality of life (SF-36) improved from 90.9 (11.9) to 112.9 (18) (p = 0.003). Nine out of ten patients and neurologists scored the global assessment of change as ‘good’ or ‘very good’ but one patient suffered significant depression after surgery. In this patient, the dystonia worsened, despite stimulation after her depression was treated. The time course of improvements varied from almost immediate to weeks or up to 6 months in one patient. Overall, medication usage was reduced post-operatively. Adverse events included dysphagia (one patient) that subsided with reduction of stimulation but was detectable on barium swallow at 1 year; dysarthria (one patient) that resolved with stimulator adjustment; transient facial weakness (resolved spontaneously by 3 months); shingles in the V1 distribution of the trigeminal nerve (one patient) that delayed the second side. In this patient, a small subdural collection was seen during the second surgery and drained during that procedure (this caused a mild hemiparesis that resolved by 12 months). Two patients had mild swallowing difficulties. Neuropsychology outcomes showed that two patients had significant declines (>2 SD); phonemic fluency in one and verbal memory in the other. Neither of these were thought to negatively impact on daily life or working ability.

Conclusions Bilateral GPi DBS led to a sustained and significant improvement in head and neck postures over a 1-year period. Other secondary measures such as pain, disability, quality of life, and depression also improved. There were a few side effects including dysphagia and reduction on some cognitive tests.

Critique In terms of the modern standards of evidence-based medicine, this trial does not compare with the large, randomized, blinded trials that we have come to expect, and in fact it represents only Class III evidence. However, it is considered a landmark here because it is the first attempt to assess DBS for cervical dystonia in a prospective series with blinded assessments. The main limitations of the study are that there is no randomization and the numbers are extremely small. The lack of randomization may be related to the small numbers in that one can only randomize to two groups if there are enough subjects to compare. Therefore, this study shows a promising trend to improvement but does not ‘clinch the deal’. We have to remember that cervical dystonia requiring DBS is even rarer than generalized dystonia and so a prospective RCT would have to involve many centres and would take several years. Another criticism of this study is the purity of the dystonia that the subjects were suffering. Five had prominent associated tremor and three had ‘minimal’ tremor. One had writer’s cramp and one had had a head injury that the authors reported had ‘initiated’ the dystonia but was not the cause of it. How do they know this? The presence of tremor would imply that these patients may have an element of ‘dystonic tremor’ and therefore a dystonic condition that goes beyond a torticollis and is more like a segmental dystonia. Some experts would say that it is almost impossible to find a ‘pure’ cervical dystonia but the prominence of tremor in this group is striking. Again, this may represent the problem with recruitment. A further criticism is that the patients in this study were aged between 47 and 64 and therefore this represents a single age group. What about younger patients? There is some evidence that younger patients or those with a shorter disease duration may do better (possibly due to less fixed postures) and therefore this study may underplay the improvements that might be achieved in a younger age group. There were a significant number of adverse events—these were mainly, but not all, transient. Depression was not mentioned as an adverse event but described in association with the poor response in one patient. Could this have been a reaction to the poor result or a direct result of stimulation? This is unknown—greater numbers would be needed to demonstrate an association. This complication was not listed in the adverse events but should have been.

Impact on Field

5.7 Spinal Cord Stimulation for Failed Back Surgery Syndrome Details of Study Between 10% and 40% of patients who undergo lumbosacral surgery end up with persistent or recurrent chronic 164

neuropathic pain, usually in the legs, known as ‘failed back surgery syndrome’ (FBSS). Spinal cord stimulation (SCS) has been used to treat FBSS for a number of years and around 35,000 systems are implanted worldwide each year. This is the reason, therefore, that the PROCESS study reported here is a landmark in the field of neuromodulation. It was the first study to report greater pain relief, quality of life, and functional capacity in this patient group. Since the PROCESS study, a 24-month follow-up has also been published that demonstrates long-lasting pain relief.

Study References Main Study Kumar K, Taylor RS, Jacques L, Eldabe S, Meglio M, Molet J, Thomson S, O’Callaghan J, Eisenberg E, Milbouw G, Buchser E, Fortini G, Richardson J, North RB. Spinal cord stimulation versus conventional medical management for neuropathic pain: a multicentre randomised controlled trial in patients with failed back surgery syndrome. Pain 2007; 132(1– 2): 179–188.

Related Reference Kumar K, Taylor RS, Jacques L, Eldabe S, Meglio M, Molet J, Thomson S, O’Callaghan J, Eisenberg E, Milbouw G, Buchser E, Fortini G, Richardson J, North RB. The effects of spinal cord stimulation in neuropathic pain are sustained: a 24-month follow-up of the prospective randomized controlled multicenter trial of the effectiveness of spinal cord stimulation. Neurosurgery 2008; 63(4): 762–770.

Study Design Non-blinded, prospective, randomized, multi-centre trial. Class of evidence

Randomization

Spinal cord stimulator versus conventional medical management

Number of patients

100 (88 completed 12-month follow-up)

Length of follow-up

12 months

Number of centres

Stratification

None

This was the first prospective randomized trial comparing SCS + conventional medical therapy (CMM) to CMM alone. The study conclusively demonstrated that SCS improves pain relief and quality of life for FBSS. The trial also highlighted the large number of device-related problems that occur in these patients. Twelve-month follow-up study shows that the improvements persist at 2 years. Inclusion criteria: ≥18 years of age; neuropathic radicular pain (L4 and/or L5 and/or S1) predominantly in legs; intensity ≥50/100 mm on VAS; at least 6 months’ duration; at least one anatomically successful surgery for a herniated disc. Exclusion criteria: another significant or disabling chronic pain condition; perceived inability to receive or operate SCS system; history of coagulation disorder, lupus erythematosus, diabetic neuropathy, rheumatoid arthritis, ankylosing spondylitis, active psychiatric disorder or condition known to affect pain perception; inability to evaluate treatment outcome; life expectancy <1 year; actual or planned pregnancy.

Outcome Measures Primary Endpoint Proportion of patients achieving at least 50% leg pain relief at 6 months.

Secondary Endpoints Four-day pain diary leg pain (VAS). Four-day pain diary back pain (VAS). SF-36 Quality of Life Questionnaire. Oswestry Disability Index. Morphine and other drug use. Use of non-drug therapies (e.g. acupuncture or massage). Patient satisfaction (satisfied with pain relief, would agree to operation again and return to work). Adverse events.

Results 165

Primary outcome: 48% of patients in the SCS arm achieved at least 50% pain relief compared to 9% in the medical arm (p <0.001). Regarding secondary outcomes, in general, SCS patients improved significantly on most pain and quality of life scores. Opioid use was not significantly different between the two groups at 6 months but tended to be less in the SCS group.

Crossover: 28/44 medical patients crossed to SCS at 6 months (+ a further four who failed screening) compared to only five SCS patients who agreed to cross to medical management. Complications; 27/84 patients (32%) who received an electrode (either trial or implanted system) experienced 40 device-related complications over 12 months; 20/27, i.e. 24% of the total, required surgery to rectify the problem. Events included electrode migration (10%), infection or wound complications (8%), loss of paraesthesiae (7%). Non-device complications were higher in the medical arm (52% versus 35%) and consisted largely of drug-related events.

Conclusions Compared with conventional medical therapy alone, SCS improves pain relief, quality of life, functional capacity, and patient satisfaction in selected patients with neuropathic pain related to FBSS.

Critique The PROCESS study was a well-designed and well-executed trial that represents a landmark because, despite the tens of thousands of SCS systems implanted prior to the trial, efficacy of SCS had not been demonstrated. The main limitation of the trial is that it demonstrates efficacy at 6 months but not in the long term. This can be likened to some of the studies that show lumbar discectomy to be efficacious for short-term (i.e. 6â&#x20AC;&#x201C;12 months) pain relief but not over longer periods. However, as a result of the trial there is no doubt that pain is alleviated and that patients are satisfied with the result. Furthermore, the follow-up study looks at 2-year outcomes and goes a long way in answering this question. The trial stands out from many other pain trials in that these measures of patient satisfaction and questions such as whether the patient would have the procedure again are practical questions rather than abstract measures of pain severity. There have been other RCTs in SCS. For example, Kemler et al. compared SCS plus physical therapy to physical therapy alone and showed that SCS was efficacious (Kemler et al., 2000; Kemler et al., 2004; Kemler et al., 2006). However, it is notable that all of the patients had previously failed physical therapy and therefore SCS was being compared to a failed treatment. Also, the long-term outcomes showed that the effects of SCS diminished with time. 166

This may be important when analysing the results of the PROCESS study, as outcomes have only been published to 2 years. This ‘tolerance’ to stimulation has been seen before in studies of DBS for pain (Romanelli et al., 2004). A second RCT and one that is directly relevant to PROCESS was performed by North et al. comparing reoperation versus SCS in patients with FBSS (North et al., 2005). This trial had a few problems. For example, there was a high rate of unintended crossover in both groups (54% re-operation to SCS and 21% the opposite way) and patients receiving worker’s compensation were less likely to enter the trial due to lack of insurance authorization. In summary, the PROCESS study is the first to show that short-term analgesia is better with SCS than medical management in FBSS, although as pointed out by Turner et al. in their editorial (Turner et al., 2007), as there was no sham stimulation, whether the effects were due to ‘active’ effects of SCS or placebo is difficult to determine.

References Kemler MA, Barendse GA, van Kleef M, de Vet HC, Rijks CP, Furnée CA, van den Wildenberg FA. Spinal cord stimulation in patients with chronic reflex sympathetic dystrophy. N Engl J Med 2000; 343: 618–624. Kemler MA, De Vet HC, Barendse GA, Van Den Wildenberg FA, Van Kleef M. The effect of spinal cord stimulation in patients with chronic reflex sympathetic dystrophy: two years’ follow-up of the randomized controlled trial. Ann Neurol 2004; 55: 13–18. Kemler MA, de Vet HC, Barendse GA, van den Wildenberg FA, van Kleef M. Spinal cord stimulation for chronic reflex sympathetic dystrophy—five-year follow-up. N Engl J Med 2006; 354: 2394–2396. North RB, Kidd DH, Farrokhi F, Piantadosi SA. Spinal cord stimulation versus repeated lumbosacral spine surgery for chronic pain: a randomized, controlled trial. Neurosurgery 2005; 56: 98–106; discussion 106–107. Romanelli P, Heit G. Patient-controlled deep brain stimulation can overcome analgesic tolerance. Stereotact Funct Neurosurg 2004; 82: 77–79. Turner JA, Deyo RA, Loeser JD. Spinal cord stimulation: stimulating questions. Pain 2007; 132: 10–11.

5.8 Neuromodulation for Cluster Headache and Migraine: Occipital Nerve Stimulation for Cluster Headache Details of Study Cluster headache (CH) is a disabling headache syndrome that is characterized by bouts of severe headaches that are extremely severe and has been labelled the ‘suicide headache’. Attacks are unilateral and associated with autonomic symptoms and occur in ‘clusters’. In the episodic form, there may be many weeks or months between clusters, but when they occur, patients may experience several attacks per day. Chronic cluster headache (CCH) is said to occur when a patient has no more than 1 month headache free in 12 months and may occur de novo (primary CCH) or after episodic CH (secondary). There have been a number of small case series of occipital nerve stimulation (ONS) for CH but numbers are too few for a RCT.

Study References Main Study Burns B, Watkins L and Goadsby PJ. Successful treatment of medically intractable cluster headache using occipital nerve stimulation (ONS). Lancet 2007; 369: 1099–1106.

Related References Burns B, Watkins L, Goadsby PJ. Treatment of intractable chronic cluster headache by occipital nerve stimulation in 14 patients. Neurology 2009; 72(4): 341–345. Magis D, Allena M, Bolla M, De Pasqua V, Remacle JM, Schoenen J. Occipital nerve stimulation for drug-resistant chronic cluster headache: a prospective pilot study. Lancet Neurol 2007; 6(4): 314–321. Schwedt TJ, Dodick DW, Hentz J, Trentman TL, Zimmerman RS. Occipital nerve stimulation for chronic headache—longterm safety and efficacy. Cephalalgia 2007; 27(2): 153–157.

Study Design This is a case series of eight patients who chose ONS over DBS or an ablative trigeminal procedure. It is an audit of outcome rather than a trial. Inclusion criterion was a diagnosis of cch, as defined by the Headache Classification Committee of the International Headache Society. Trial stimulation or successful occipital nerve block was not used as a selection criterion. Class of evidence

III

Randomization

None

Number of patients

167

Length of follow-up

Median 20 months Range 6–27 months

Number of centres

Stratification

None

Outcome Measures Patients were asked: ‘Would you recommend the procedure to a fellow cluster headache sufferer?’ Use of triptans. Frequency, duration, and severity of headaches (retrospective case note review).

Results Eight patients (seven male, one female) were implanted and had a median duration of disease of 6 years. Five had secondary CCH, i.e. they had progressed from the episodic form to the chronic form and three had primary CCH. At a median of 20 months (range 8–27), six of the eight patients reported an improvement in their condition and said that they would recommend an ONS to other patients with CCH in similar circumstances. Two were described as ‘marked improvement’ (≥90% reduction in attacks), three as ‘moderate’ (≥40%), and one as ‘mild’ (≥25%). ‘Improvement’ mainly pertained to frequency or severity rather than duration of attacks. Regarding medication, one patient stopped using triptans, three reduced their use, and in four patients triptan use remained unchanged. The benefit from the ONS was not immediate and was found to take weeks or months. In contrast, when the ONS became faulty (in one patient with >90% relief), the attacks recurred almost immediately. Interestingly, the first patient had a unilateral electrode because 95% of the attacks were left-sided. Gradually, a greater proportion started on the right side and therefore a second electrode (on the other side) was implanted. After implantation of the second electrode, 70% were left-sided and 30% right-sided (with an overall 40% reduction). Subsequently, all patients were implanted bilaterally. Regarding adverse events, there were eight serious adverse events requiring surgical intervention. These included three due to electrode migration (all in the same patient), one due to electrode failure, and five relating to battery malfunction or depletion. Minor complications included neck stiffness and pain related to the extension leads. All patients experienced paraesthesiae in the occipital region but generally considered this a reassuring sensation.

Conclusions ONS shows promise for CCH and persists for a long time. The effect is not predicted by occipital nerve block. The procedure is straightforward and safe, although not without complications.

Critique If this study was looking at a common disease such as migraine, it would be heavily criticized for its wide range of follow-up times, miniscule numbers, different disease subtypes (primary and secondary CCH), single centre, and high rate of adverse events. However, in the context of CCH, we are dealing with a very rare problem and in this subgroup, an extremely refractory group of patients. In addition, since the treatment is expensive, each case required individual funding. For these reasons, the numbers available for study are extremely small and this study is considered a landmark paper for the fact that it was one of the first such series to attempt to document the efficacy of ONS in CCH in a quasi-controlled manner, and was published in a high-impact journal. The outcome measures chosen are interesting. One could be positive or negative about ‘whether a patient would recommend the procedure to a fellow sufferer’. Most commonly, such studies would tend to use validated outcomes such as headache severity/number of attacks or quality of life measures such as SF-36 or Euroqol 5D as the primary outcome and this study may be criticized for not using a validated outcome measure. However, one could also look at this outcome positively in that this single question asks about the most important aspect of treatment to the patient. The fact that the answers were positive shows that the study has had the desired result. Another criticism is the vagueness of the patient ‘estimate’ of percentage change in their CHs. In a prospective trial, one would expect this to be appropriately validated with the use of patient diaries. Such an approach, relying on a patient’s memory, may be subject to bias and there is much evidence in the scientific literature of biased outcome reporting in pain patients, due to psychological factors etc. Despite the criticisms and the fact that it is difficult to understand the significance of the results of this study, it was one of a few small studies that provided important early information as to the efficacy of ONS in CCH and paved the way for future studies. It leaves many questions unanswered such as ‘What are the effective stimulation parameters?’ and ‘What is the long-term outcome?’. It also showed that complications related to the device are common, mainly comprising migration of leads. However, overall, this study provided a good initial exploration.

5.9 Neuromodulation for Cluster Headache and Migraine: Deep Brain 168

Stimulation for Cluster Headache Details of Study DBS for CCH gained prominence amongst neurosurgeons just before ONS and was largely pioneered by the Milan group led by Leone. One of the important aspects of this study is that the idea and targeting of the posterior hypothalamus was based mainly on PET imaging studies by May and colleagues (May et al., 1998). Most DBS targets have been based on either animal studies (e.g. the STN or PPN in PD) or historical lesions in humans. Interestingly, the example of CCH was the first of many, including subgenual cingulate stimulation for depression.

Study References Main Study Leone M, Franzini A, Broggi G, Bussone G. Hypothalamic stimulation for intractable cluster headache: long-term experience. Neurology 2006; 67: 150–152.

Related References Fontaine D, Lazorthes Y, Mertens P, Blond S, Géraud G, Fabre N, Navez M, Lucas C, Dubois F, Gonfrier S, Paquis P, Lantéri-Minet M. Safety and efficacy of deep brain stimulation in refractory cluster headache: a randomized placebocontrolled double-blind trial followed by a 1-year open extension. J Headache Pain 2010; 11(1): 23–31. Leone M, Franzini A, Bussone G. Stereotactic stimulation of posterior hypothalamic gray matter in a patient with intractable cluster headache. N Engl J Med 2001; 345(19): 1428–1429.

Study Design This is a case series of 16 patients treated with DBS. There were no formal inclusion or exclusion criteria as such, as this was not a formal trial. Twelve stimulators (nine patients) were turned off in a blinded fashion after at least 3 months of being pain-free. There does not appear to be a formal protocol as such. Class of evidence

III

Randomization

None

Number of patients

Length of follow-up

Mean 23 months Range 1–52 months

Number of centres

Stratification

None

Outcome Measures ‘Headache response’ (described as ‘pain free’ or ‘sporadic attacks’ etc.); time to response after stimulator inserted; percentage of pain-free days. Use of prophylactic medications. The effect of turning the stimulation off in nine patients (12 electrodes).

Results Major improvements in pain were described for 15/18 implants (13 patients). A persistent pain-free state was achieved in ten patients, almost pain-free but with sporadic attacks in three patients. The remaining three patients still continued to have CH attacks but had markedly reduced pain intensity, frequency, and duration. Four patients continued to require prophylactic medications to reduce the attacks but the rest were free from medication postoperatively. The mean time to ‘stable benefit’, i.e. pain free or reduced attacks, was 42 days. Regarding the uncontrolled, single-blind switching off, this was performed in nine patients (12 electrodes) with a range of 2–8 months of off time (mean 5 months). In one patient, the stimulators were turned off numerous times with a consistent return of attacks each time. In four patients, the attacks returned only sporadically and so the system was left off. The remaining five patients experienced full-blown attacks at a mean of 2 months after switching off. In patients with bilateral attacks and bilateral electrodes, only the ipsilateral electrode controlled the attacks. With respect to side effects and adverse events, a small asymptomatic intraventricular haemorrhage was seen in one patient on post-operative imaging and resolved spontaneously. Diplopia was a common side effect observed when increasing the voltage too rapidly but subsided spontaneously. There were no effects observed relating to hormonal changes, weight (other than small amounts associated with steroids), autonomic changes, sleep–wake cycle, or electrolyte balance. 169

Conclusions DBS is an efficacious and safe treatment for drug-refractory cluster headache. The procedure is well tolerated, needs to be ipsilateral, but is subject to the known risks of DBS of other areas.

Critique This case series can be criticized in the same way as ONS (see section 5.8). It is neither case-controlled, randomized, nor double blinded. The numbers are small, and it is inherently subject to the bias of such series. However, in a similar way to the paper described for ONS, it is a landmark paper simply because it represents a first serious exploration and innovation of DBS for CH. Whilst there was much spontaneity in the design of the study (no fixed protocols etc.), no-one can argue that the results are extremely impressive and indeed more impressive than ONS in terms of the proportion of patients who were headache free and those who had a significant reduction in severity and frequency of attacks. However, for the clinician and patient deciding on treatment options, the small haemorrhage in one patient, whilst asymptomatic, is a reminder that DBS has serious risks and these need to be balanced against benefit and other options. At the time of this study, ONS was not established for CH and therefore, DBS may have been considered as first-line treatment. Now that the sensible pathway may be considered to start with ONS (which has fewer risks) and proceed to DBS if unsuccessful, this study reminds us that here is an option that may work if ONS fails, or for the risk takers, may even be considered first as the results are better. The study, whilst statistically not a test of efficacy, clearly demonstrates it. It is elegant in that many other factors that could potentially be altered with manipulation of the posterior hypothalamus were investigated, such as autonomic parameters, sleep–wake cycle, and hormonal changes. However, these findings may be rather surprising and one wonders whether the investigation of some of these factors were rigorous. For example, the results are not shown in the supplementary data and presumably sleep–wake cycle was not measured formally. Overall, however, this represents a significant first large case series of a very rare condition and has influenced the treatment of CCH to the benefit of many severely affected, medically-refractory patients.

Reference May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. Hypothalamic activation in cluster headache attacks. Lancet 1998; 352(9124): 275–278.

5.10 Neuromodulation for Cluster Headache and Migraine: Occipital Nerve Stimulation for Migraine Details of Study Peripheral nerve stimulation is a rapidly expanding field in the treatment of chronic pain. ONS is one such treatment that involves placement of an electrode in the vicinity of the greater occipital nerve (and the lesser occipital nerve may also be stimulated, depending on electrode length). There have now been a number of studies and this treatment is currently being considered by the National Institute for Health and Care Excellence (NICE) (NICE, 2013).

Study References Main Study Saper JR, Dodick DW, Silberstein SD, McCarville S, Sun M, Goadsby PJ. Occipital nerve stimulation for the treatment of intractable chronic migraine headache: ONSTIM feasibility study. Cephalalgia 2011; 31(3): 271–285.

Related References Dodick DW, Trentman T, Zimmerman R, Eross EJ. Occipital nerve stimulation for intractable chronic primary headache disorders. Cephalalgia 2003; 23: 701. Silberstein SD, Dodick DW, Saper J, Huh B, Slavin KV, Sharan A, Reed K, Narouze S, Mogilner A, Goldstein J, Trentman T, Vaisma J, Ordia J, Weber P, Deer T, Levy R, Diaz RL, Washburn SN, Mekhail N. Safety and efficacy of peripheral nerve stimulation of the occipital nerves for the management of chronic migraine: results from a randomized, multicenter, double-blinded, controlled study. Cephalalgia 2012; 32(16): 1165–1179. Weiner RL, Reed KL. Peripheral neurostimulation for control of intractable occipital neuralgia. Neuromodulation 1999; 2: 217– 222.

Study Design A prospective, randomized, multi-centre study comparing ONS to medical management for intractable chronic migraine. Class of evidence

170

Randomization Subjects randomized to active stimulation (AS), pre-set stimulation (PS), i.e. sham stimulation, and medical management (MM) in a 2:1:1 ratio Number of patients

110 (75 continued in treatment group; 33 AS, 17 (PS), 17 (MM), 8 ‘ancillary’

Length of follow-up

3 months (up to 36 months open label) Primary endpoint:* Safety and adverse events Secondary endpoints:* Number of headache days Pain intensity Headache duration

Number of centres

Stratification None * As this was a feasibility study, these are technically not ‘endpoints’ as such.

The main aim of this study was to collect safety and primary efficacy data. Intractable chronic migraine defined as headache ≥15 days per month. Eligible subjects received an occipital nerve block and responders were randomized to AS, PS, or MM. ‘Responder’ to ONS defined as ≥50% reduction in headache days per month or ≥3-point reduction in intensity compared to baseline. Key inclusion criteria: ≥15 headache days per month for 3 months in the absence of medication overuse; pain involving occipital/suboccipital region; refractory to medication; onset before 50 years; ≥18 years. Key exclusion criteria: other health conditions likely to affect the study; previous nerve ablation procedure likely to affect the C2/3 area; rebound headaches; other electrical stimulation device; MRI may be required.

Outcome Measures Safety (adverse events); changes in headache days; change in overall pain intensity; percentage change in number of days with severe, prolonged headache; change in hours per headache day; quality of life scores; Profile of Mood States (POMS); SF-36; MIDAS; functional disability; and subject satisfaction.

Results Seventy-five subjects were enrolled out of 110 screened. At 3 months, percentage reduction in headache days per month was 27.0 ± 44.8% in the active stimulation (AS) group, 8.8 ± 28.6% in the pre-set stimulation (PS) group, and 4.4 ± 19.1% in the medical management (MM) group. A separate ‘ancillary’ group were a non-randomized group of eight patients who had ‘failed’ occipital nerve block and the response rate was 39.9 ± 51.0%. These results equated to a reduction in headache days of 6.7 ± 10 in the AS group and a reduction of intensity of 1.5 ± 1.6. For the majority of outcome measures (including change in headache days, pain intensity, prolonged severe headache, and headache duration), improvement compared to baseline in the AS group was not statistically different compared to the control groups (PS and MM) although there was a trend towards improvement.

Regarding the disability and quality of life scores, there were again, trends to improvement in the POMS (8.7 171

versus 1.6 in AS versus MM), disability scale (0.3 for AS versus 0.0 in MM) but comparison of the groups did not reach statistical significance. Adverse events were divided into device related and non-device related. In the device-related category, 56 ‘adverse device events’ (ADEs) occurred in 36 of the 51 subjects. Three serious adverse effects requiring hospitalization were implant infection, lead migration, and post-operative nausea. The most frequent ADE was lead migration that occurred in 24%. The main non-device related adverse effect was worsening of migraine that occurred in 9% of the AS group, 41% of the PS group, and 24% of the MM group. Adverse effects related to medication were similar across treatment groups.

Conclusions Further investigation of ONS for medically intractable chronic migraine is justified by this study. Although this study cannot establish efficacy, the results are promising.

Critique This study was the first ‘large’ study comparing ONS to medical treatment in a randomized fashion. As it is a feasibility study, it was not powered to evaluate efficacy but rather to look at safety and the feasibility of conducting a more robust trial within a focused patient population and with longer follow-up times and better blinding and endpoints. The authors in fact make these conclusions themselves. As such, the trial was successful in providing good, randomized pilot data to inform power calculations and to give an indication of safety using one particular stimulator system. There are a number of possible criticisms. Firstly, the follow-up duration of 3 months was short. Given that migraine can have a complex temporal course, a longer duration in a larger trial would be useful. This would help to rule out ‘insertional’ effects of electrode implantation as well as helping to provide justification for what happens to be a very expensive procedure. Secondly, the 39% responder rate is not particularly impressive. However, the authors point out that this group of patients are medically refractory and represent a group of patients who ‘lead lives that are painful and compromised’ and that for them, there is little other hope. On an individual level, the responder rate alone is not as important as the comparison between the responder rate, the risk of the procedure, and the financial cost. If the risk of serious adverse events is very low, a patient with refractory migraine may well be prepared to put up with a 39% chance of being a ‘responder’. Also, as the authors point out, the AS response rate of 39% is comparable to drugs such as topiramate for preventive migraine treatment. Other criticisms might include the degree of blinding which is difficult in trials involving neuromodulation. Whilst the medical arm cannot be blinded, the PS group may have had some sensation and indeed it is possible the settings could have had an effect. However, these hurdles are almost impossible to jump and the trial design was as pragmatic as possible in this regard. Again, the authors make this point. A final criticism might be that the trial was sponsored by the manufacturer of the stimulator system, although there appears to be no bias in even the methodology or the reporting of the results. This study as well as a number of others (mainly smaller studies) were reviewed by NICE (NICE, 2013). The summary concluded that ONS carries a surgical risk and the follow-up from the studies is short, but that ONS holds promise and is worthy of further investigation. In this study, most of the surgical complications are related to the implant itself and it is possible that further advancement of the technology may reduce these problems. In summary, this study has provided good initial data that ONS is a treatment worth considering and needs further investigation. At the time of writing, the PRISM study is pursuing such a course and the results are eagerly awaited.

Reference NICE. Occipital Nerve Stimulation for Intractable Chronic Migraine (IPG 452). <http:www.guidance.nice.org.uk/ipg452>

5.11 Neurosurgical Treatment of Trigeminal Neuralgia: Microvascular Decompression for Trigeminal Neuralgia The long-term outcome of microvascular decompression (MVD) for trigeminal neuralgia (TN) was established by Janetta’s group in 1996 with the publication of results of a series of 1185 patients from the Presbyterian-University Hospital in Pittsburgh, Pennsylvania, USA, undergoing MVD between 1972 and 1991 (Barker et al., 1996).

Study References Main Study Barker FG, Janetta PJ, Bissonette DJ, PAC, Larkins MV, Jho HD. The long-term outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 1996; 334: 1077–1083.

Related Reference Janetta PJ. Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurgery 1976; 26: 159–162.

172

Study Design At surgery compressing arteries were separated from the nerve and compressing veins were cauterized and cut. Class of evidence

Randomization

None (see following text)

Number of patients

1185

Length of follow-up

>1 year Median 6.2 years 91% at 5 years

Number of centres

Stratification

None

Outcome Measures Primary Endpoint Relief of pain: complete relief (excellent outcome) defined as ≥98% pain relief without need for medication; partial relief (good outcome) defined as 75% reduction in pain; failure (poor outcome) defined as recurrence of >25% of pre-operative pain or need for a further surgical procedure.

Secondary Endpoints Operative findings. Complications.

Results When repeat surgery was included excellent outcomes were achieved in 80% of patients at 1 year and 70% of patients at 10 years. The commonest vessel causing compression was the superior cerebellar artery (75%). Venous compression was seen in 68%. Complications were uncommon with CSF leak, hearing loss, and facial numbness being the most frequent. Recurrence rates were <2% at 5 years and <1% at 10 years. Risk factors for recurrence included lack of immediate post-operative relief, female sex, venous compression, and pre-operative symptoms of >8 years’ duration.

Conclusion MVD for TN is safe and has a high rate of long-term success rate.

Critique MVD of the trigeminal nerve for TN was popularized by Gardner and Janetta (Gardner, 1962; Gardner, 1968; Gardner and Miklos, 1959; Janetta, 1967). Janetta’s series considered here is a landmark in the neurosurgical literature as it established the favourable long-term results of MVD. MVD is firmly established as the most successful treatment option for TN refractory to medical therapy. However, there has been substantial debate in the literature regarding the mechanism by which this operation is effective. Walter Dandy is credited in the 1940s with proposing that TN may be caused by compression of the trigeminal nerve by the superior cerebellar artery at its point of entry into the pons. Janetta is credited subsequently with elaborating the concept of pulsatile compression at the root entry zone (REZ), which was defined as a junctional area between central and peripheral myelin (Janetta, 1977, 1980). It was postulated that relief of this compression explains the efficacy of MVD. However, Gardner and Miklos had previously stated that the critical component of MVD is manipulation of the nerve itself (Gardner and Miklos, 1959). This observation led to the hypothesis that MVD is effective because of trauma to the nerve itself, a theory that would appear to be supported by the efficacy of partial section of the trigeminal nerve. The most vociferous opponent of 173

Janetta’s theory of microvascular compression at the REZ was Adams in Oxford who presented anatomical, clinical, neuropathological, and neurophysiological findings that he proposed supported the hypothesis that trauma to the nerve during dissection and subsequent ‘decompensation’ of the nerve were more likely to explain the efficacy of MVD (Adams, 1989). The mechanism of MVD efficacy remains to be elucidated. However, more recently Sindou has reported the results of large series for which clear-cut marked vascular compression at surgery is associated with higher success rates of >90% (Sindou, 2007).

References Adams CBT. Microvascular decompression: an alternative view and hypothesis. J Neurosurg 1989; 57: 1–12. Gardner WJ. Concerning the mechanism of trigeminal neuralgia and hemifacial spasm. J Neurosurg 1962; 19: 947–958. Gardner WJ. Trigeminal neuralgia. Clin Neursurg 1968; 15: 1–56. Gardner WJ, Miklos MV. Response of trigeminal neuralgia and hemifacial spasm. JAMA 1959; 170: 1773–1776. Janetta PJ. Treatment of trigeminal neuralgia by suboccipital and transtentorial cranial operations. Clin Neurosurg 1977; 24: 538–549. Janetta PJ. Neurovascular compression in cranial and systemic disease. Ann Surg 1980; 192: 518–525. Sindou M, Leston J, Decullier E, Chapus F. Microvascular decompression for primary trigeminal neuralgia: long-term effectiveness and prognostic factors in a series of 362 consecutive patients with clear-cut neurovascular conflicts who underwent pure decompression. J Neurosurg 2007; 107: 1144–1153.

5.12 Neurosurgical Treatment of Trigeminal Neuralgia: Ablative Techniques for Trigeminal Neuralgia Details of Study A summary of the best observation analysis of ablative techniques for TN was performed by a meta-analysis of the techniques of ablative therapy for TN including radiofrequency ablation (RFA), glycerol rhizolysis (GR), balloon compression (BC), and stereotactic radiosurgery (SRS) (Lopez et al., 2004).

Study Reference Main Study Lopez BC, Hamlyn PJ, Zakrzewska JM, PAC, Larkins MV, Jho HD. Systematic review of ablative neurosurgical techniques for the treatment of trigeminal neuralgia. Neurosurgery 2004; 54: 973–983.

Study design At least ten or more of data/quality criteria present. Kaplan–Meier actuarial analysis for single procedure. Not >20% of patients lost to follow-up monitoring. In series of patients, patients treated more than once were excluded. Minimum of 12-month median/mean follow-up period. Minimum of 30 patients treated in the whole series. Study not dealing exclusively or mainly with recurrences or secondary TN. For stereotactic radiosurgery studies, separate reporting of rates of complete pain relief with/without medication, with a minimal treatment dose of 70 Gy. Class of evidence

Randomization

None (see following text)

Number of patients

1185

Length of follow-up

>1 year Median 6.2 years 91% at 5 years

Number of studies

Outcome Measures Primary Endpoints Duration of complete pain relief with or without medications. Outcome data beyond mean/median discarded and data only analysed when two or more studies were available.

Results 174

RFA provided the best early and late complete pain relief. Sensory loss was experienced by all patients that underwent RFA or GR if the procedure was done correctly (permanent sensory loss occurred in 30% of RFA patients, 50% GR patients, and 20% SRS patients). Of patients that underwent RFA, 29.2% experienced a complication but most complications were transient. Corneal numbness occurred in 10% of percutaneous patients and 7% of SRS patients. Anaesthesia dolorosa was more likely after RFA or GC. There is insufficient data to compare BC with other techniques.

Conclusion RFA provides the better rates of complete pain control for the treatment of TN by ablative techniques compared with SRS and GR. SRS had the fewest side effects.

Critique As compared there are various surgical strategies that are available to treat TN including MVD, partial rhizotomy of the sensory root, percutaneous compression (balloon compression), glycerol rhizotomy, radiofrequency thermorhizotomy, and stereotactic radiosurgery. Within the meta-analysis of the observational studies, it has been demonstrated that the benefit of long-term effectiveness is proportional to the degree of sensory loss and risk of trigeminal side effects (Lopez et al., 2004). Problems with this paper include the heterogeneity of the patient population and lack of methodological uniformity. Furthermore, this analysis suggests that stereotactic radiosurgery may promise high efficacy with a low complication rate. Although there has been no study comparing the long-term effectiveness between surgical and percutaneous ablative procedures, a review of the literature of those studies which had at least a 5-year mean follow-up found that MVD has the best reported long-term results (Tatli et al., 2008).

References Lopez BC, Hamlyn PJ, Zakrzewska JM, PAC, Larkins MV, Jho HD. Systematic review of ablative neurosurgical techniques for the treatment of trigeminal neuralgia. Neurosurgery 2004; 54: 973–983. Tatli M, Satici O, Kanpolat Y, Sindou M. Various surgical modalities for trigeminal neuralgia: literature study of respective long-term outcomes. Acta Neurochir (Wein) 2008; 150: 234–255.

5.13 Deep Brain Stimulation for Treatment-Resistant Depression Details of Study This study by Mayberg et al. tested the hypothesis that DBS to modulate activity of the grey matter in area 25 of the cingulate gyrus (Cg25) would produce clinical benefits in patients with treatment-resistant depression. The study was carried out in Toronto, Canada using subgenual cingulate white mater (Cg25WM) as a target in six patients.

Study References Main Study Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005; 45: 651–660.

Related References Greenberg BD, Friehs G, Carpenter L, Tyrka A, Malone A, Rezai A, Shapira N, Foote K, Okun M, Goodman W, Rasmussen S, Price L. Deep brain stimulation: clinical findings in intractable depression and OCD. Neuropsychopharmacology 2005; 29: S32 Lozano AM, Giacobbe P, Hamani C, Rizvi SJ, Kennedy SH, Kolivakis TT, Debonnel G, Sadikot AF, Lam RW, Howard AK, Ilcewicz-Klimek M, Honey CR, Mayberg HS. A multicenter pilot study of subcallosal cingulate area deep brain stimulation for treatment-resistant depression. J Neurosurg 2012; 116(2): 315–322. Mayberg HS, Liotti M, Brannan SK, McGunnis S, Mahurin RK, Jerabek PA, Silva JA, Tekell JL, Martin CC, Fox PT. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999; 156: 675–682.

175

Study Design A pilot study to evaluate Cg25WM as a target for DBS in treatment-resistant depression. Class of evidence

III

Randomization

None

Number of patients

Follow-up

6 months Primary endpoints: Clinical depression Regional cerebral blood flow measurements Secondary endpoints: Neuropsychological testing Adverse events

Number of centres

Stratification

None

Inclusion criteria: treatment-resistant depression; DSM-IV criteria for a major depressive episode of >1 year’s duration; minimum score of 20 on the Hamilton Depression Rating Scale (HDRS); age <60 years. Exclusion criteria: previous stroke; significant cerebrovascular risk factors; other Axis I psychiatric disorder (e.g. schizophrenia, OCD); psychotic symptoms; active suicidal intent; recent substance abuse; >60 years of age; inability to comply with follow-up; contraindications to DBS (e.g. cardiac pacemaker). Treatment-resistant depression was defined as failure to respond to at least four different classes of antidepressant medication in maximal doses. In addition, five out of six patients had failed to respond to ECT.

Surgical Technique Microelectrodes were inserted using a Leksell stereotactic frame. Medtronic 3387 quadripolar DBS electrodes were implanted bilaterally. Electrode positioning was achieved by MRI mapping: the GM/WM transition area 25 was located at the midpoint between the genu of corpus callosum and the anterior commissure. Microelectrode recordings were used to guide insertion into the transitional area between Cg25 (neuronal active area) and Cg25WM (cell sparse area). Final electrode positioning was confirmed by post-operative MRI. Stimulation parameters were reassessed and adjusted at weekly intervals with increments of 1.0 V. The mean parameters uses were 4.0 V with a 60 µs pulse width at a frequency of 130 Hz.

Outcomes Primary Endpoints Clinical depression: improvements in depression were monitored using the 17-item HDRS with a clinical response being defined as a reduction of ≥50%. Regional blood flow: measured using PET.

Secondary Endpoints Neuropsychological testing of cognitive, intellectual, and frontal functioning. Monitoring of adverse events.

Results Clinical Response Five out of six patients showed a clinical response (≥50% reduction in HDRS) at 2 months. Clinical response was maintained in four out of these five patients at 6 months.

Regional Blood Flow Measurements Reduced Cg25WM activity was seen in all patients: both responders and non-responders alike. In addition, responders showed an area of hyperactivity in the medial frontal cortex (BA10).

176

Other Findings Improvement in early morning sleep disturbance was the earliest sign of a clinical response and was seen in four out of the six patients. Improvements were seen in neuropsychological tests.

Adverse Effects There were no adverse effects on orbitofrontal functioning (such events would be indicative of local DBS adverse effects). There were no adverse affective or autonomic effects of increments in stimulator settings. All patients experienced psychomotor slowing at higher voltage settings >7 V). Two patients received antibiotics for superficial infections related to the connector cables.

Conclusions Cg25WM is an effective target for DBS in treatment-resistant depression.

Critique Major depression is the most common psychiatric disorder and depression that is resistant to medication and ECT is an extremely severe and debilitating condition. In addition to DBS, vagal nerve stimulation has also been assessed in the treatment of treatment-resistant depression, and although no randomized trial has been carried out, preliminary studies show some promise (Rush et al., 2000). DBS is still in the exploratory phases for the treatment of this disorder. Greenberg et al. noticed improvement in co-morbid depression symptomatology in patients undergoing DBS of the ventral portion of the anterior limb of the internal capsule and the adjacent dorsal ventral striatum for treatment of OCD (Greenberg et al., 2003). The authors then evaluated DBS in the ventral internal capsule in a series of five patients with depression and reported an improvement on the HDRS in all five patients with a mean improvement from 31.4 to 15.8 over 3 months (Greenberg et al., 2005). The study summarized here by Mayberg et al. arose from their previous observations that there was elevated activity of Cg25 in neuroimaging studies of patients with severe treatment-resistant depression (Mayberg et al., 1999). This observation was in concordance with reports of suppression of activity in this same region by antidepressive treatments including selective serotonin reuptake inhibitors (SSRIs) and electroconvulsive therapy. Mayberg et al., therefore, hypothesized that chronic stimulation to modulate Cg25 grey matter may ameliorate symptoms in treatment-resistant depression. The results of their study showed that four out of six patients showed a sustained clinical benefit from DBS of the Cg25WM at 6 months. It should be emphasized that this is a pilot study and Mayberg et al. pointed out that, although encouraging, their results are limited by the small sample size, limited follow-up, and the lack of sham surgery or a placebo-control arm. However, Mayberg et al. did undertake a trial of blinded DBS discontinuation in one patient and their findings appeared to support Cg25WM stimulation to effect symptom relief. A subsequent three-centre trial by the same group in 21 patients showed that 48% of patients at 6 months and only 29% at 1 year had a 50% or greater reduction in HDRS (Lozano et al., 2012). Notwithstanding these difficulties and the reduced efficacy shown by the larger multi-centre case series, this study represents a significant landmark in identifying the Cg25WM as a target for DBS in treatment-resistant depression. Also, it is an example of how findings from functional imaging studies have been translated into a novel surgical intervention for a major debilitating disorder.

References Greenberg BD, Price LH, Rauch SL, Friehs G, Noren G, Malone D, Carpenter LL, Rezai AR, Rasmussen SA. Neurosurgery for intractable obsessive-compulsive disorder and depression: critical issues. Neurosurg Clin N Am 2003; 14: 199–212. Lozano AM, Giacobbe P, Hamani C, Rizvi SJ, Kennedy SH, Kolivakis TT, Debonnel G, Sadikot AF, Lam RW, Howard AK, Ilcewicz-Klimek M, Honey CR, Mayberg HS. A multicenter pilot study of subcallosal cingulate area deep brain stimulation for treatment-resistant depression. J Neurosurg 2012; 116(2): 315–322. Rush AJ, George MS, Sackeim HA, Marangell LB, Husain MM, Giller C, Nahas Z, Haines S, Simpsn RK Jr, Goodman R. Vagus nerve stimulation (VNS) for treatment resistant depression: a multicenter study. Biol Psychiatry 2000; 47: 276–286.

5.14 Molecular and Cellular Therapies for Parkinson’s Disease Details of Studies Neural transplantation for the treatment of neurodegenerative disorders has long been a holy grail for neurosurgeons. The three first studies discussed, by Li et al., Kordower et al., and Mendez et al., all published in 2008, are the first post-mortem demonstrations that fetal neural cells grafted into patients with PD can survive and function for over a decade in the host’s brain. They are also to date the most detailed and most cited landmark studies demonstrating very long-term survival of grafted neural cells. Furthermore, these studies demonstrated for the first time that in long-term 177

surviving grafts, a fraction of grafted cells can develop Lewy body pathology-like inclusions, characteristic for the host parkinsonian pathology. These studies provided for the first time a proof of principle that grafted neural cells can survive long term and at the same time provide clinical benefits in the diseased parkinsonian brain. They thus lay the foundation for further work in this field, including the development of alternative cell sources for transplantation.

Study References Main Studies Kordower JH, Chu Y, Hauser RA, Freeman TB, Olanow CW. Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson’s disease. Nat Med 2008; 14: 504–506. Li JY, Englund E, Holton JL, Soulet D, Hagell P, Lees AJ, Lashley T, Quinn NP, Rehncrona S, Bjorklund A, Widner H, Revesz T, Lindvall O, Brundin P. Lewy bodies in grafted neurons in subjects with Parkinson’s disease suggest host-tograft disease propagation. Nat Med 2008; 14: 501–503. Mendez I, Vinuela A, Astradsson A, Mukhida K, Hallett P, Robertson H, Tierney T, Holness R, Dagher A, Trojanowski JQ, Isacson O. Dopamine neurons implanted into people with Parkinson’s disease survive without pathology for 14 years. Nat Med 2008; 14: 507–509.

Related References Astradsson A, Cooper O, Vinuela A, Isacson O. Recent advances in cell-based therapy for Parkinson disease. Neurosurg Focus 2008; 24(3–4): E6. Cooper O, Astradsson A, Hallett P, Robertson H, Mendez I, Isacson O. Lack of functional relevance of isolated cell damage in transplants of Parkinson’s disease patients. J Neurol 2009; 256(Suppl 3): 310–316. Hargus G, Cooper O, Deleidi M, Levy A, Lee K, Marlow E, Yow A, Soldner F, Hockemeyer D, Hallett PJ, Osborn T, Jaenisch R, Isacson O. Differentiated Parkinson patient-derived induced pluripotent stem cells grow in the adult rodent brain and reduce motor asymmetry in Parkinsonian rats. Proc Natl Acad Sci U S A 2010; 107(36): 15921–15926. Kordower JH, Chu Y, Hauser RA, Olanow CW, Freeman TB. Transplanted dopaminergic neurons develop PD pathologic changes: a second case report. Mov Disord 2008; 23: 2303–2306. Mendez I, Sanchez-Pernaute R, Cooper O, Viñuela A, Ferrari D, Björklund L, Dagher A, Isacson O. Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson’s disease. Brain 2005; 128(Pt 7): 1498–1510.

Study Designs Case studies to evaluate long-term survival and related pathology and function of transplanted fetal dopamine neurons in patients with PD. Class of evidence

Randomization

None

Number of patients

Follow-up

Up to 16 years Primary endpoints: Graft survival at post-mortem Pathology at post-mortem Secondary endpoints: Clinical benefits Adverse events

Number of centres

Stratification

None

Inclusion criteria: idiopathic PD and pre-operative PET imaging consistent with PD. Good response to levodopa from the onset of the disease, but maximum tolerated medication not providing adequate relief of symptoms and causing unacceptable side effects. Exclusion criteria: atypical parkinsonism, pronounced dementia, epilepsy, previous brain surgery, severe depression, cerebrovascular disease and medical contraindications to surgery.

Surgical Technique Fetal ventral midbrain tissue was collected with maternal consent from women undergoing elective abortion between 6 and 9 weeks after conception. Fetal ventral midbrains were dissected under sterile conditions and either single-cell suspensions or solid blocks of tissue were prepared for transplantation. Patients were fitted with a stereotactic head frame and the stereotactic coordinates for targets in the post 178

commissural putamen were calculated using MRI images and a computerized stereotactic planning workstation. Transplantation cannulas were inserted into the different targets in the post-commissural putamen and either a cell suspension (Li et al., 2008; Mendez et al., 2005) or solid graft (Kordower et al., 2008) delivered along the predetermined tracts. Immunosuppression with cyclosporine alone (Mendez et al., 2008; Kordower et al., 2008) or corticosteroids, azathioprine, and cyclosporine (Li et al., 2008) was given post-operatively for at least 6 months.

Post-mortem Analysis Fixed brain blocks of grafted striatum and the substantia nigra were serially cut in 40 μm thick sections on a cryostat. Immunohistochemistry using standard techniques was performed on free-floating sections, including staining for tyrosine hydroxylase, the microglial marker, CD45, alpha-synuclein, phosphorylated alpha-synuclein, and ubiquitin. Immunofluorescence staining was examined using a confocal microscope and design based stereology was performed to assess cell numbers and graft volumes.

Outcome Measures Primary Endpoints Long-term graft survival was demonstrated in all seven patients at post-mortem, ranging from 9–16 years after transplantation. Grafts contained numerous tyrosine hydroxylase (TH)-positive dopamine neurons, in the order of 10,000– 100,000 per graft. Grafts were well integrated, reinnervating the host striatum. Dopamine grafts in four of the seven patients reported were found to have Lewy body-like inclusions, characteristic of PD. However only an estimated 1–5% of grafted neurons in these patients contained Lewy body-like inclusions. Solid grafts elicited a stronger host immune reaction than cell suspension grafts, demonstrated by the increased presence of activated microglia.

Secondary Endpoints Clinical outcome was highly variable, ranging from little if any demonstrable benefits to marked improvements in measures of PD function, including UPDRS motor off medication scores, off time, and dyskinesias, and substantially reduced antiparkinsonian medication requirements, with benefits lasting for over a decade. The variable clinical benefits were probably a reflection of the variation in the number of surviving grafted cells. No adverse events, such as graft-induced dyskinesias, were encountered.

Conclusions Transplanted fetal dopamine neurons in PD patients can survive for over a decade and provide marked and prolonged clinical benefits in a subset of patients. A fraction of grafted neurons can develop Lewy body-like inclusions over time. However, this is unlikely to affect graft function.

Critique One of the first neural transplantation studies in PD utilized dissociated fetal ventral midbrain (VM) tissue transplanted into the striatum of patients that had developed PD after accidental exposure of 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine. These patients exhibited marked motor improvement correlating with an increase of fluorodopa uptake in the striatum on PET scans. Furthermore, fluorodopa uptake and postsynaptic dopamine receptor occupancy PET studies have indicated that fetal grafts in the striatum can survive for up to a decade and provide sustained motor benefits in PD. Two double-blind, randomized controlled, clinical trials of neural transplantation for PD in the United States have been reported. Freed et al. randomized 40 patients to receive either bilateral putaminal transplants of fetal ventral midbrain tissue from two embryos per side or sham surgery (Freed et al., 2001). Solid tissue was transplanted and no immunosuppression was given. The study failed to meet its primary endpoint of clinical improvement on a selfreporting scale. However, a treatment effect was observed in younger patients. Also the trial was concluded after only a year, a time which is insufficient for the growth and integration of human fetal dopamine neurons and the development of functional effects. Indeed several more patients showed clinical improvement after the conclusion of the trial, 2–3 years after transplantation surgery. Unfortunately, off-period dyskinesias were observed in 15% of the patients. Olanow et al. randomized 34 patients to receive bilateral putaminal fetal VM tissue from one or four donors 179

per side or undergo a placebo procedure (Olanow et al., 2003). Solid tissue pieces were transplanted. A 6-month course of immunosuppression was given. The trial failed to meet its primary endpoint of improvement in the motor component of the UPDRS, although a treatment effect was observed in milder disease. Concerning off-period dyskinesias were observed in 56% of the patients. Few post-mortem studies of the survival of grafted cells have been reported. The studies by Li et al., Kordower et al., and Mendez et al. are the first in-depth, long-term post-mortem studies of the fate of grafted fetal dopamine neurons in PD. These studies have demonstrated that fetal cells can survive and integrate for over a decade in the host brain and provide long-term functional benefits. However Li et al. and Kordower et al. also demonstrated that a fraction of grafted dopamine neurons developed Lewy body-like inclusion pathology, the hallmark of PD, in their four long-term surviving patients, while Mendez et al. reported no such pathology in their three long-term surviving patients, although subsequent analysis revealed that in one of their patients, with the youngest graft, one or two grafted neurons appeared to contain neuromelanin and Lewy body-like pathology. Despite the occurrence of Lewy body-like pathology in grafted neurons of four of the seven patients reported, it should be stressed that only 1–5% of grafted neurons of these contained any pathology, whereas the vast majority of them were healthy looking. The occasional appearance of Lewy body-like inclusions is therefore unlikely to affect graft function and indeed long-term benefits were observed beyond a decade in some these patients. Furthermore, the occurrence of Lewy body-like pathology in grafted cells seems to be related to transplantation techniques and host– graft reaction, as patients with the most inclusions had solid tissue grafts, and more activated microglia. In conclusion, transplanted fetal dopamine neurons can survive for up to 16 years despite ongoing neurodegeneration of the host brain. Furthermore, these and several other case studies have demonstrated that grafted fetal dopamine neurons can provide substantial long-term clinical benefits in PD. These findings are encouraging for the future development of fetal and stem cell-derived therapy for PD.

References Astradsson A, Cooper O, Vinuela A, Isacson O. Recent advances in cell-based therapy for Parkinson disease. Neurosurg Focus 2008; 24: E6. Cooper O, Astradsson A, Hallett P, Robertson H, Mendez I, Isacson O. Lack of functional relevance of isolated cell damage in transplants of Parkinson’s disease patients. J Neurol 2009; 256(Suppl 3): 310–316. Freed CR, Greene PE, Breeze RE, Tsai WY, DuMouchel W, Kao R, Dillon S, Winfield H, Culver S, Trojanowski JQ, Eidelberg D, Fahn S. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001; 344: 710–719. Kordower JH, Chu Y, Hauser RA, Freeman TB, Olanow CW. Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson’s disease. Nat Med 2008; 14: 504–506. Kordower JH, Chu Y, Hauser RA, Olanow CW, Freeman TB. Transplanted dopaminergic neurons develop PD pathologic changes: a second case report. Mov Disord 2008; 23: 2303–2306. Li JY, Englund E, Holton JL, Soulet D, Hagell P, Lees AJ, Lashley T, Quinn NP, Rehncrona S, Bjorklund A, Widner H, Revesz T, Lindvall O, Brundin P. Lewy bodies in grafted neurons in subjects with Parkinson’s disease suggest host-tograft disease propagation. Nat Med 2008; 14: 501–503. Mendez I, Sanchez-Pernaute R, Cooper O, Vinuela A, Ferrari D, Bjorklund L, Dagher A, Isacson O. Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson’s disease. Brain 2005; 128: 1498–1510. Mendez I, Vinuela A, Astradsson A, Mukhida K, Hallett P, Robertson H, Tierney T, Holness R, Dagher A, Trojanowski JQ, Isacson O. Dopamine neurons implanted into people with Parkinson’s disease survive without pathology for 14 years. Nat Med 2008; 14: 507–509. Olanow CW, Goetz CG, Kordower JH, Stoessl AJ, Sossi V, Brin MF, Shannon KM, Nauert GM, Perl DP, Godbold J, Freeman TB A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 2003; 54: 403–414.

180

Chapter 6

Paediatric neurosurgery RD Johnson, P Richards, J Jayamohan, S Sinha 6.0 6.1 6.2 6.3 6.4

Introduction Diuretic therapy in post-haemorrhagic ventricular dilatation Shunt design trial Decompressive craniectomy in paediatric head injury Hypothermia in paediatric head injury

6.0 Introduction Paediatric neurosurgery is a complex subspecialty that does not lend itself easily to large clinical trials. Outcomes may be more complex and multidimensional in paediatric neurosurgery than they are in adult neurosurgery (Kan and Kestle, 2007). In addition, there may be fewer areas where true clinical equipoise exists regarding treatment options. Nonetheless, there is likely to be an increasing number of larger clinical studies undertaken in paediatric neurosurgery over the next few years. In this chapter, four studies that highlight different aspects of the kinds of problems faced by the paediatric neurosurgical discipline have been chosen for inclusion. Firstly, a trial of diuretic therapy for post-haemorrhagic ventricular dilatation (PHVD) in premature infants is considered (Kennedy et al., 2001). This trial is important because it addresses the efficacy of a medical treatment that became widespread before any proper assessment of its clinical efficacy. The trial demonstrated that this treatment regimen may not only be ineffective but potentially increases the risk of death or disability. The second study considered is the Shunt Design Trial, which is one of very few randomized trials that examines neurosurgical devices (Drake et al., 1998). The third study is a single-centre randomized trial addressing the role of decompressive craniectomy in paediatric head injury (Taylor et al., 2001). Although this is only a pilot study it remains the only published randomized study to date that suggests there may be a beneficial role for decompressive craniectomy in head injury. The fourth, and final, study included in this chapter is one of the largest multi-centre trials carried out in paediatric neurosurgical patients, which examines the role of hypothermia in managing severe paediatric head injury (Hutchison et al., 2008).

References Drake JM, Kestle JRW, Milner R, Cinalli G, Boop F, Piatt J, Haines S, Schiff S, Cochrane DD, Steinbolc P, MacNeil N for the collaborators. Randomised trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 1998; 43: 294–303. Hutchison JS, Ward RE, Lacroix JL, Hebert PC, Barnes MA, Bohn DJ, Dirks PB, Douchette S, Fergusson D, Gottesman R, Joffe AR, Kirkpalani HM, Meyer PG, Morris KP, Moher D, Singh RN, Skippen PW for the Hypothermia Pediatric Head Injury Trial Investigators and the Canadian Critical Care Trials Group. N Engl J Med 2008; 358: 2447–2456. Kan P, Kestle JRW. Designing randomised clinical trials in pediatric neurosurgery. Childs Nerv Syst 2007; 23: 385–390. Kennedy CR, Ayers S, Campbell MJ, Elbourne D, Hope P, Johnson A, on behalf of the International PHVD drug trial group. Randomised, controlled trial of acetazolamide and furosemide in post-haemorrhagic ventricular dilatation in infancy: follow-up at 1 year. Paediatrics 2001; 108: 597–607. Taylor A, Butt W, Rosenfeld J, Shann F, Ditchfield M, Lewis E, Klug G, Wallace D, Henning R, Tibballs J. A randomised trial of very early decompressive craniectomy in children with traumatic brain injury and sustained intracranial hypertension. Childs Nerv Syst 2001; 17: 154–162.

6.1 Diuretic Therapy in Post-haemorrhagic Ventricular Dilatation Details of Study This multi-centre randomized controlled trial addressed the question of whether the widespread use of drug treatment (furosemide and acetazolamide) in the treatment of PHVD in preterm infants has any effect on reducing the need for surgery by way of cerebrospinal fluid (CSF) diversion. The trial was carried out between 1992 and 1996 in 55 centres worldwide.

Study References Main Study 181

International PHVD drug trial group. International randomised controlled trial of acetazolamide and furosemide in posthaemorrhagic ventricular dilatation in infancy. Lancet 1998; 352: 433–440.

Related References Libenson MH, Kaye EM, Rosman NP, Gilmore HE. Acetazolamide and furosemide for posthemorrhagic hydrocephalus of the newborn. Pediatr Neurol 1999; 20: 185–191. Ventriculomegaly Trial Group. Randomised trial of early tapping in neonatal posthaemorrhagic ventricular dilatation. Arch Dis Child 1990; 65: 3–10.

Study Design A multi-centre PRCT. Class of evidence

Randomization

Drug treatment versus standard treatment

Number of patients

177 (129 with 1-year data)

Follow-up

Primary outcomes: Death or shunt placement or both at 1 year Secondary outcome: Neurodevelopmental status at 1 year

Number of centres

55 centres worldwide

Stratification

Presence of cerebral parenchymal lesions on ultrasonography

Inclusion criteria: age <3 months past term; ultrasound evidence of germinal or intraventricular haemorrhage; progressive ventricular dilatation with ventricular index >4 mm above 97th percentile. Drug treatment consisted of acetazolamide 100 mg/day (initiated at 25 mg/day and increased in 25 mg/day increments over 4 days) and furosemide 1 mg/day. Standard treatment included removal of CSF if head growth was double normal rate over 2 weeks or there were signs of raised intracranial pressure. Shunt insertion was recommended if: head circumference was ≥1.5 cm above 97th percentile; or head growth was ≥1.5 cm/week for 2 weeks; and presence of signs of raised intracranial pressure.

Results Follow-up was 93% at 1 year for the primary outcome. Follow-up was 83% at 1 year for the secondary outcome. Infants in the drug therapy group had a significantly increased risk (p = 0.012) of death, impairment, or disability at 1 year. Risk ratio of 1.40 (1.12–1.76).

Conclusions Drug therapy (acetazolamide and furosemide) in infants with PHVD does not reduce the rate of shunt placement and is associated with an increased risk of death, shunt placement, and neurological disability.

Critique Post-haemorrhagic ventricular dilatation is a condition that affects 17 in 1000 infants born at <32 weeks of gestation. With the increase in survival of neonates born at 28 weeks of gestation, this condition continues to be a significant source of infant morbidity and mortality. At present the best long-term solution is insertion of a ventriculoperitoneal shunt. However, the risks of surgery and the subsequent complications of shunt surgery cannot be discounted. Several methodologies have been tried to reduce the requirement of surgical treatment, including the use of diuretics. Diuretic drugs such as acetazolamide and furosemide can lead to a reduction in CSF production. Studies dating back to 1958 have been reported using such drugs to medically manage hydrocephalus and negate the need for shunt insertion (Birzis, 1958; Chaplin, 1980; Donat, 1980; Shinnar, 1985). Despite the lack of Level I or II evidence to show their efficacy, these drugs have found widespread usage in the treatment of hydrocephalus. Given the potential 182

complications of diuretic therapy, the International PHVD Drug Trial Group undertook this study to address this particular issue. The study was well designed with simple outcomes (death or shunt insertion before 1 year of age) that minimized any potential bias. In conceiving an international multi-centre trial, it was possible to recruit a large number of patients to allow a meaningful conclusion. Randomization was performed to allow a balance between the groups and within the respective participating centres. Although it was not possible to conceal treatment allocation from the parents or doctors (in order to monitor electrolyte and acid–base status), this was not revealed to the paediatricians who carried out the neurodevelopmental assessment. It may be argued that all infants should have had their acid–base and electrolyte status assessed, which would allow blinding of the parents and treating physicians but it seems unlikely that this would have altered the outcome. Infants in either group who proceeded to shunt placement were found to have similar ventricular indices and head size, suggesting that the criteria for surgery were met equally in both groups. The trial has been criticized for using both acetazolamide and furosemide together. However, the trial authors have argued that the risk of adverse effects from using two drugs in combination does not necessarily equate with a lack of efficacy. The mechanism of action for diuretics is the same in all types of hydrocephalus. The usage of either a single drug or differing doses is therefore unlikely to be of additional benefit. There was no evidence in the trial that any of the deaths were related to a specific biological mechanism that would implicate side effects of the diuretics. While there was an expected rise in alveolar partial pressure of carbon dioxide (PCO2) there was no increase in the number of infants requiring ventilation. The results show a significantly increased risk of death and shunt placement as well as neurodevelopmental disability in the diuretic therapy group. There was no delay to the requirement for shunt placement with diuretic therapy. This effect was still significant after multiple logistic regression was used to negate the potential confounding effects of birth weight, gestational age at birth, post-natal age, and head circumference. The fact that the trial was stopped early due to the significant advantage of standard treatment emphasizes the deleterious effects of diuretic therapy in the treatment of PHVD. Overall, the study was well constructed and obtained an important and significant conclusion. The authors reach a valid conclusion in stating that diuretic therapy cannot be recommended in the treatment of PHVD. While the results of this trial were in progress, a further smaller trial was carried out at a single institute (Libenson et al., 1999) in which 16 patients were recruited. The results of this trial did not show a deleterious effect of diuretic therapy. However, the trial does not mention the method of randomization and there are unequal numbers in the study and control groups. Power calculations are not provided and neither the intervention nor the outcome was blinded. Subsequent meta-analysis of the two studies also reaches the conclusion that diuretic therapy cannot be recommended.

References Birzis L, Carter CH, Maren TH. Effects of acetazolamide on CSF pressure and electrolytes in hydrocephalus. Neurology 1958; 8: 522–528. Chaplin ER, Goldstein GW, Myerberg DZ, Hunt JV, Tooley WH. Posthaemorrhagic hydrocephalus in the preterm infant. Pediatrics 1980; 65: 901–909. Donat JF. Acetazolamide-induced improvement in hydrocephalus. Arch Neurol 1980; 37: 376. Libenson MH, Carter CH, Kaye EM, Rosman NP, Gilmore HE. Acetazolamide and furosemide for posthaemorrhagic hydrocephalus of the newborn. Pediatr Neurol 1999; 20: 185–191. Shinnar S, Gammon E, Bergman EW Jr, Epstein M, Freeman JM. Management of hydrocephalus in infancy: use of acetazolamide and furosemide to avoid cerebrospinal fluid shunts. J Pediatr 1985; 107: 31–37.

6.2 Shunt Design Trial Details of Study The Shunt Design Trial assessed whether two new shunt valves could decrease the 1-year shunt failure rate compared to differential pressure valves. The study was carried out at 12 centres in North America and Europe between 1993 and 1996. The two new valves studied were the Delta valve (Medtronic PS Medical) and the Orbis-Sigma valve (NMT Cordis).

Study References Main Study Drake JM, Kestle JRW, Milner R, Cinalli G, Boop F, Piatt J, Haines S, Schiff S, Cochrane DD, Steinbok P, MacNeil N for the collaborators. Randomised trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 1998; 43: 294–303.

Related References Drake JM, Kestle J. Rationale and methodology of the multicenter pediatric cerebrospinal fluid shunt design trial. Childs Nerv Syst 1996; 12: 434–447.

183

Kestle J, Drake J, Milner R, Sainte-Rose C, Cinalli G, Boop F, Piatt J, Haines S, Schiff S, Cochrane D, Steinbok P, MacNeil N for the collaborators. Long-term follow-up data from the shunt design trial. Pediatr Neurosurg 2000; 33: 230– 236.

Study Design Inclusion criteria: age 0–18 years; newly diagnosed hydrocephalus; radiological evidence of ventriculomegaly; requirement for first shunt insertion. Exclusion criteria: previous shunt; active infection; spread of tumour to subarachnoid space; loculations requiring more than one shunt; Dandy–Walker malformation; arachnoid cyst causing hydrocephalus; systemic contraindication to shunting. Shunt failure defined as: obstruction; overdrainage; loculation of ventricles; infection. Sample size calculations were based on establishing a reduction in shunt failure rates from 40% to 20%. Class of evidence

Randomization

Standard differential pressure valve versus Delta valve versus Orbis-Sigma valve

Number of patients

344

Follow-up

Primary outcomes: Shunt failures at 1 year Time to shunt failure Secondary outcomes: Death, surgical complications Type of shunt malfunction Length of hospital stay

Number of centres

12 centres in North America and Europe

Stratification

Age (<6 months, ≥6 months) Study centre

Results Overall shunt failure at 1 year was 39% with all three valves. There were no significant differences in causes of shunt failure between the three valves. There was no significant advantage with any of the three valves. There were no deaths related to shunt failure.

Conclusions There was no benefit of one valve design over another in terms of shunt failure at 1 year.

Critique CSF shunts in the paediatric population are hampered by a shunt failure rate of approximately 40%. Shunt valve design has changed over the decades since the introduction of the first valves, but the only evidence of improved efficacy has been drawn from uncontrolled case series. The Shunt Design Trial is the first randomized trial to address the question of whether novel valve designs would have any effect on shunt failure rates. The three valves involved in the trial were the standard differential pressure, the Orbis-Sigma, and Delta valves. The latter two were designed to limit overdrainage in the upright position. The study was well designed to attempt to answer the hypothesis. Previously published reports have shown a 40% shunt failure rate and the trial organizers felt that a reduction to 20% would show a good clinical effect. As such, the sample size was calculated to ensure the trial had sufficient power to account for such a reduction. The study was multi-centric with randomization and stratified by centre and age (above and below 6 months) to optimize the trial. The primary and secondary outcome measures used are easily defined and are non-biased. A blinded independent analysis of these outcomes was carried out, strengthening the nature of the study. The results of the study clearly show no difference between the three shunt valve designs. The overall shunt failure rate at 1 year was 39%. Log rank analysis and Cox regression model to account for potential causes of bias failed to show any difference between the valves (log rank = 2.90, p = 0.24). While the sample sizes do not produce sufficient power to allow analysis for specific age groups, secondary analysis suggests that the result is valid for older children as well. The study was not sufficiently powered to assess the differences in the valve function in children with tumours in whom third ventriculostomy is the favoured option. Similarly, the nature of hydrocephalus in adults is predominantly different to that of children and it is not therefore appropriate to extrapolate these results to the adult 184

population. The authors rightly conclude that there is no difference between the standard differential pressure valve, the newer Orbis-Sigma valve, and the Delta valve in reducing the rates of shunt failure in children. Overall the study is very well designed and represents a major landmark in being not only the first randomized trial of different shunt valve designs but also one of the very few trials of a neurosurgical device.

6.3 Decompressive Craniectomy in Paediatric Head Injury Details of Study Taylor et al. carried out a single-centre study of early decompressive craniectomy in paediatric head-injured patients admitted to the intensive care unit at the Royal Children’s Hospital, Melbourne, Australia, between 1991 and 1998.

Study References Main Study Taylor A, Butt W, Rosenfeld J, Shann F, Ditchfield M, Lewis E, Klug G, Wallace D, Henning R, Tibballs J. A randomised trial of very early decompressive craniectomy in children with traumatic brain injury and sustained intracranial hypertension. Childs Nerv Syst 2001; 17: 154–162.

Related Reference Sahuquillo J, Arikan F. Decompressive craniectomy for the treatment of refractory high intracranial pressure in traumatic brain injury. Cochrane Database Syst Rev 2006; 1: CD003983.

Study Design Inclusion criteria: age 1–16 years; severe traumatic brain injury (TBI) with functioning intraventricular catheter. Functional outcome was assessed using the Glasgow Outcome Scale (GOS) and the Health State Utility (HSU) Index. Both groups received conventional medical management to control cerebral perfusion pressure (CPP) and intracranial pressure (ICP). Children who had uncontrolled ICP (20–24 mmHg for >30 min, 25–29 mmHg for >10 min, >30 mmHg for >1 min) were randomized. Those randomized to surgery underwent a bi-temporal craniotomy within 6 h of randomization. Favourable outcome was defined as those who were functionally normal or who had mild disability. Unfavourable outcome was defined as moderate disability or worse. Class of evidence

Randomization

Decompressive craniectomy versus medical management

Number of patients

Follow-up

Primary outcome: Functional outcome at 6 months Secondary outcome: Physiological parameters including ICP control

Number of centres

Stratification

Severity of brain injury

Results Follow-up was 100% at 6 months.

There was a non-significant trend towards a shorter time in intensive care in the decompressive craniectomy group. 185

Conclusions There may be an advantage in early decompressive craniectomy in the paediatric patients with severe traumatic head injury in terms of control of ICP and favourable outcomes. However, a larger multi-centre trial is required to assess this further.

Critique Decompressive craniectomy remains a rather controversial area. There is often a strength of opinion among clinicians either for or against it without any significant evidence to back either view. To date, there is no Level I or II evidence in the adult population that addresses this issue. Two studies are in progress (RescueICP and DECRA) that hope to provide evidence in the near future for the adult population. Given this background, the study by Taylor et al. tries to address this aspect in the paediatric population. The study was well performed with attempts to remove any bias. It struggled, however, in a number of ways. The study took 7 years to perform and despite this only 27 patients were recruited. The authors mention that clinical management policies changed during this period but one would expect that these changes would affect both arms in a similar fashion. Patients were selected if they had a traumatic brain injury and had a functioning intraventricular catheter. Diffuse brain injury often leads to cerebral swelling without increase of ventricular size and often significant compression of the ventricular system, contraindicating the placement of such a catheter. The majority of patients with diffuse injury are usually managed with an intracranial pressure monitor and may not have been included in this study. The surgical treatment can also expect some criticism. Craniectomy was performed by removal of 3–4 cm discs of bone bi-temporally with no attempt to open the dura. Studies looking at expansion have showed that a large craniectomy with dural opening is required in order to maximize the benefit of craniectomy. Perhaps with a more definitive decompression more significant results may have been seen. The outcome analysis was described as being carried out by telephone interview, chart review, and/or discussion with the treating physician. While the unit has significant experience of follow-up in this fashion, outcome measures are more accurately achieved by both clinical and neuropsychological review. Functional outcome measures in children were carried out at 6 months only, with no further evaluation. It has been shown that outcome status in children changes over time and it would have been useful to have reassessed outcome at 1, 2, and 5 years to give a more long-term picture. Due to the changes in management protocol over the study period, statistical analysis was performed on two separate occasions. The results showed a favourable outcome in 54% of patients in the decompressive group compared with 14% in the control (medical) group. Using the two-tailed Fisher’s test, p = 0.046, but owing to the repeated analysis the p value for statistical significance is reduced further to p <0.0221, leading to the conclusion that early decompressive craniectomy may have a beneficial effect. Despite some of the earlier-mentioned shortcomings, the study by Taylor et al. was well designed and performed. The results, while not statistically significant, do suggest that there is likely to be a benefit from early decompressive surgery. Given the difficulty of recruitment in a single centre and the limitations of studies with small numbers, the authors rightly conclude that a larger multi-centre trial is required to further address this important question.

6.4 Hypothermia in Paediatric Head Injury Details of Study The efficacy of hypothermia in the management of paediatric head injury was examined in a multi-centre international trial by the Hypothermia Paediatric Head Injury Trial Investigators and the Canadian Critical Care Trials Group between 1999 and 2004 in North America.

Study References Main Study Hutchison JS, Ward RE, Lacroix JL, Hebert PC, Barnes MA, Bohn DJ, Dirks PB, Douchette S, Fergusson D, Gottesman R, Joffe AR, Kirkpalani HM, Meyer PG, Morris KP, Moher D, Singh RN, Skippen PW for the Hypothermia Pediatric Head Injury Trial Investigators and the Canadian Critical Care Trials Group. N Engl J Med 2008; 358: 2447–2456.

Related Reference Clifton GL, Miller ER, Choi SC, Levin HS, McCauley S, Smith K, Muizelaar JP, Wagner FC, Marion DW, Luerssen TG, Chestnut RM, Schwartz M. Lack of effect of induction of hypothermia after acute brain injury. N Engl J Med 2001; 344: 556–563.

Study Design 186

Class of evidence

Randomization

Hypothermia versus normothermia

Number of patients

225

Follow-up

Primary outcomes: Morbidity and mortality at 6 months Secondary outcomes: Overall mortality Co-intervention requirements Physiological variables Functional psychological outcomes Adverse events

Number of centres

17 centres in three countries

Stratification

Age <7 and ≥7 years

Inclusion criteria: age 1–17 years; GCS <8; CT evidence of brain injury; need for ventilation. Exclusion criteria: >8 h post injury; brain death; cervical cord injury; prolonged cardiac arrest; non-accidental injury. Patients in the hypothermia group were cooled to 32.5°C within 8 h of injury and therapy was continued for 24 h. Patients in the normothermia group were kept at 37°C. Paediatric Cerebral Performance Category scale (PCPC) was used to assess outcome. This is a 6-point scale: (1) normal performance; (2) mild disability; (3) moderate disability; (4) severe disability; (5) persistent vegetative state; (6) death. Unfavourable outcome was defined as a PCPC score of ≥4. Favourable outcome was defined as a PCPC score of ≤3. Analysis was performed on an intention-to-treat basis.

Results Follow-up was 91% for primary outcome data.

The degree of hypotension and requirement for vasoactive agents in the hypothermia group during the rewarming period was significantly greater than in the normothermia group. Significantly more interventions were required to control ICP in the normothermia group within the first 24 h compared to the hypothermia group.

Conclusions Hypothermia initiated within 8 h from injury and continued for 24 h is not associated with improved neurological outcome and may increase mortality in the paediatric population.

Critique While animal models suggest that hypothermia improves survival in traumatic brain injury, the evidence for a similar response in children has remained lacking. Hutchinson et al. therefore designed this trial in order to explore the outcome following hypothermia therapy after traumatic brain injury. This was a prospective multi-centre randomized study encompassing 17 centres in three countries. The study was well designed with a simple primary outcome, which was assessed by persons blinded to the original assigned arm of treatment. A secondary outcome to assess neuropsychological changes was carried out by telephone by a trained psychologist, as well as interviews with parents at several time-points following injury. In order to reduce any potential bias, the management protocol for both arms was decided by consensus prior to the study, allowing for comparison across all the centres. Randomization and stratification was carried out according to each centre and by age (grouped into <7 or ≥7 years). The latter was fashioned as children sustaining significant injuries at a younger age have poorer neuropsychological recovery than older children. The criteria for cooling were 187

based on best-published evidence at the time of the trial. Statistical analysis incorporated adequate power calculations and took into account potential losses to follow-up. An interim analysis at two separate points was used to ensure that the trial should not be stopped. Results were analysed on an intention-to-treat basis and logistic regression used to account for factors that could adversely influence the results. The follow-up at 6 months was 91%, which was in keeping with the predicted power calculations. There was no significant benefit from the use of hypothermia when the primary and secondary outcomes were analysed. Sensitivity analysis used to account for the patients lost to follow-up when biased against hypothermic treatment suggest a significant risk of an unfavourable outcome in the hypothermia group (p = 0.001). However, when biased towards normothermic treatment, the results showed no increased risk in the hypothermia group (p = 0.82). Subgroup analysis for primary outcome showed no significant difference between the two treatment arms except in patients in whom the ICP remained <20 mmHg. In this particular subgroup, hypothermia treatment had an increased risk of an unfavourable outcome (RR 2.12, p = 0.03). In a subgroup analysis of secondary outcome, there was a significant trend to poorer visual memory in the hypothermia group when assessed at 12 months (p = 0.05). The mean time taken to achieve hypothermia in these patients was approximately 6 h. It may be argued that perhaps earlier institution of cooling may have a better outcome as is seen in animal models (15 min to cooling); however, the practicalities of assessment and treatment in children with severe head injury make faster treatment highly unlikely. The authors mention that unpublished results looking at a subgroup of children who were able to be treated early showed no benefit either. Hypothermic treatment was only used for 24 h in this study, which may not be long enough to optimize its beneficial aspects. There is some evidence that treatment for >48 h may reduce the risk of death and an unfavourable outcome in adults (systematic reviewâ&#x20AC;&#x201D;McIntyre et al., 2003). Although the study ran over 5 years, the number of children recruited was only 225. Small treatment effects are unlikely to be detectable with such a small sample and leaves room for a larger trial to explore other potential benefits. While there are some aspects that could be improved upon, the trial was well designed. The two treatment arms were well matched for management with no significant co-interventions that would bias the results. It would be of benefit to reassess the children at further intervals, as outcome status in children changes over time. This would provide a more long-term picture. A more recent systematic review and meta-analysis by the Cochrane Collaboration (Alderson et al., 2004) showed no overall benefit from either immediate or deferred hypothermia treatment in adults with severe brain injury. This meta-analysis included the McIntyre review (mentioned earlier) in which the suggested beneficial results were obtained by pooling results from both deferred and immediate hypothermia studies. The outcome of the trial reaches a valid conclusion that hypothermia for 24 h is not beneficial in the management of severe head injury in children but still leaves questions about potential benefit of earlier or a more prolonged treatment in certain subgroups of patients. Further larger trials would be required to assess these more specific aspects.

References Alderson P, Gadkray C, Signorini DF. Therapeutic hypothermia for head injury. Cochrane Database Syst Rev 2004; 4: CD001048. McIntyre LA, Fergusson DA, HĂŠbert PC, Moher D, Hutchison JS. Prolonged therapeutic hypothermia after traumatic brain injury in adults: a systematic review. JAMA 2003; 289: 2992â&#x20AC;&#x201C;2999.

188

Chapter 7

Pituitary surgery RD Johnson, PJ Weir, N Maartens, SA Cudlip, AH Kaye, ER Laws Jr 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7

Introduction Timing of surgical intervention for pituitary apoplexy Classification of cavernous sinus invasion by pituitary adenomas Biological correlates of invasive adenomas Complications of pituitary surgery Surgical treatment of prolactinomas Extended transsphenoidal approaches to anterior cranial base lesions Endoscopic transsphenoidal surgery

7.0 Introduction Pituitary surgery is a unique subspecialty within neurosurgery which has been shaped, and continues to be influenced, by the way in which surgical techniques developed. In many ways, the true landmarks in pituitary surgery were the work of the bold pioneers who developed these techniques and undertook the early operations. The courage of the surgeons in this story is matched only by the bravery of their patients who subjected themselves to the unknown in the hope of a cure. It is not the aim of this chapter to cover in detail the historical development of pituitary surgery. Nonetheless, it is not possible to do this area justice without a brief mention of some of the highlights of this story. The great physician of Marcus Aurelius, Galen of Pergamon, made the first speculation about the function of the pituitary c.150 ad and referred to it as the phlegm gland which drained waste from the ventricles of the brain to the nose. It is perhaps not too difficult to see why such a hypothesis would be formulated, as Galen’s dissections of this area may have resulted in the release of CSF from the pituitary fossa and through the nose. Furthermore, he may well have dissected some Rathke’s cysts which would have been very similar to ‘phlegm’. In 1543, the great anatomist Vesalius included a plate of the hypothalamic pituitary unit in the seventh book of his De Humani Corporis Fabrica in 1543, which is almost certainly the oldest image of the hypothalamic pituitary unit in Western literature. In this depiction, Vesalius depicts four ducts draining the ‘phlegm’ away from the pituitary gland in accordance with the Galenic view. Again, it is conceivable that the ducts depicted by Vesalius were draining veins to or between the cavernous sinuses. It was not until several hundred years later that there was any appreciation of a relationship between pathology in the pituitary region and disease. In the eighteenth century, Antonii De Haan, in Vienna, described a case of amenorrhoea associated with a pituitary tumour. In the late nineteenth century, Pierre Marie, in Paris, described the association of acromegaly with pituitary enlargement. There were other developments in our understanding of the development of the pituitary gland in the nineteenth century, when Rathke undertook his great embryological work in Germany, leading to the discovery of the origins of what we now know as Rathke’s cleft cysts. The developments of endocrine physiology are the territory of a separate volume altogether. However, the pituitary gland is now known to be the ‘master endocrine gland’, coordinating a complex neurohormonal system, and clinical endocrinology is a highly subspecialized academic area of internal medicine. The neurosurgeon, however, is primarily interested in pituitary pathology that can be treated surgically. In essence, this consists of symptomatic space-occupying lesions of the sella turcica which cause either endocrine or neurological compromise (macroadenomas, meningiomas, Rathke’s cysts, and craniopharyngiomas), but also hormone-secreting lesions (prolactinomas, Cushing’s disease, acromegaly) or haemorrhage (pituitary tumour apoplexy). When assessing the success of surgery in treating these lesions it is necessary to define what constitutes a satisfactory outcome following pituitary surgery. If the intent of surgery is to decompress a space-occupying lesion, then success may be measured by preservation of pituitary function, improvement of vision, or prevention of recurrence or progression of the tumour. However, if the intent of surgery is restoration of normal pituitary function, then long-term remission rates may best be defined by biochemical success rather than by extent of resection. The history of the development of surgical approaches to the pituitary region is a story of master and apprentice, of neurosurgical ego, and stubborn defiance, which crosses back and forth across the Atlantic from the old world to the new and back again several times. The first attempt at resection of a pituitary tumour was undertaken by Victor Horsely in the UK in 1889 by a subtemporal transcranial route. Horsley treated ten patients with a 20% mortality rate, and did not report his results until 1906. The first frontal transcranial operations were undertaken by Krause in Berlin in 1904 (in 1902 he exposed the chiasm transfrontally for a bullet wound!) and had a higher mortality rate of 50%. 189

The transnasal route to the pituitary region was well known to the ancient Egyptians who used this avenue to clear the skull of its cerebral contents during the process of mummification. However, it was Schoffler in Innsbruck, in 1907 who carried out the first successful transsphenoidal approach (TSA) in a living patient. Although his first patient died soon after, Schoffler was responsible for bringing the TSA into conventional surgical practice. Various famous surgeons modified Schoffler’s operation: Anton von Eiselsberg and Theodor Kocher both modified the incision; Julius von Hochenegg made more aggressive frontal sinus resections; and Albert Halstead developed infranasal and gingival approaches. Perhaps the most intriguing episode, however, were the simultaneous operations carried out by Oskar Hirsch in Vienna and Harvey Cushing in Baltimore, on opposite sides of the Atlantic on the same day, 4th June 1910. Hirsch introduced the endonasal procedure, and Cushing undertook a modified Schoffler approach. Cushing eventually performed 360 sublabial transsphenoidal procedures for pituitary tumours reporting a 5.6% mortality but abandoned it in in 1927 in favour of a subfrontal approach which, in his hands, had a mortality of 4.5%. Such a difference in mortality is certainly within the margin of error, and this abandonment of the TSA is perhaps more a testament to Cushing’s meticulous nature. Hirsch carried out the endonasal TSA in >600 patients. His mortality was 5.4% initially, but following the introduction of antibiotics the mortality fell to 1.5%. Hirsch emigrated to Boston in 1938 following expulsion from Austria by the Nazis. He was never recognized as an independent surgeon in his own right and was almost certainly over-shadowed by Cushing. Hirsch has been called ‘the obscure voice in the wilderness’ that kept the TSA alive. A similarly intriguing part of the saga was how the TSA became resurrected following Cushing’s abandonment. Norman Dott from Edinburgh undertook a Fellowship with Cushing in 1923–1924 and carried on using the TSA in Scotland on >100 patients without any mortality. It is likely that the antibiotic era and a greater appreciation of the role of perioperative steroids were responsible for this low complication rate. Furthermore, Dott had a low recurrence rate, and this was almost certainly due to the availability and use of adjunctive radiotherapy. Dott never published his clinical results and it is has been hypothesized that this was an example of the apprentice showing deference to the master, Cushing. Gerard Guiot from Paris travelled to Edinburgh to learn the technique from Dott, and back in France he operated on >1000 patients, and introduced X-ray image intensification to facilitate his surgery. Jules Hardy travelled from Canada to learn the TSA from Guiot and returned back across the Atlantic to Montreal; he introduced the operating microscope to the procedure. Further technical advances in pituitary surgery have come with the developments in extended TSAs by Laws in Charlottesville, and in endoscopic techniques by Cappabianca in Naples, and Kassam and Jho in Pittsburgh. The principles provided in the studies included here are of value to the entire neuro-endocrine team. This team must assess the patient, anticipate the effect of each intervention, and closely follow the result that this intervention has on the patient. As treatment modalities continue to improve, the neuro-endocrine team’s knowledge of how to use these modalities effectively in managing patients must expand as well. Only with this knowledge and experience, will an improved remission rate from the treatment of pituitary adenomas be a reality. The first section looks at the first study to evaluate the effect of timing of surgical intervention for pituitary apoplexy (Bills et al., 1993). The second section considers a landmark classification scheme developed to describe the degree of cavernous sinus invasion by pituitary tumours on MRI (Knosp et al., 1993). The next section looks at one of the early papers assessing biological correlates of pituitary tumours in order to aid prognostication by the pituitary multidisciplinary team (Thapar et al., 1996). The following section deals with a large study looking at the complications of pituitary surgery (Ciric et al., 1997). Recently there has been much discussion in the literature, and at conferences, about the need for pituitary centres of excellence. It will be very interesting to see how this discussion influences practice. Although the centralization of pituitary surgery into fewer centres may be beneficial, this is far from certain. From the surgical perspective, a greater case volume might be expected to lead to improved outcomes due to increased surgical expertise. From the endocrinologist’s perspective, it would appear that the best surgeons will be those with the greatest caseloads in specialized centres. Furthermore, better surgical results may save money because of less need for adjuvant therapy. However, there is controversy as to what the ‘magic number’ of surgical cases per year is in order to significantly improve outcomes. This paper by Ciric et al. considers this question with respect to complication rates for pituitary surgery. The next section looks at a study which evaluated the role of surgery in prolactinomas and emphasized the role of surgery as a legitimate option instead of medical therapy alone in these lesions (Tyrrell et al., 1999). The following two sections deal with developments in surgical technique: the extended TSA (Kaptain et al., 2001) and endoscopic transsphenoidal surgery (Tabaee et al., 2009). There are, of course, many other studies which could be included in this chapter. For example, a key question once it is determined that surgical intervention is indicated for a pituitary lesion, is which is the appropriate surgical approach? The surgeon must first decide if a transcranial or transsphenoidal approach is warranted for a particular lesion. Dr Vinko Dolenc analysed his series of 210 patients with tumours extending beyond the sella turcica. In this study, Dr Dolenc compared the results using a classical approach to the sellar and parasellar regions as compared to an approach utilizing the anatomical triangles of the lateral wall of the cavernous sinus. Patients who were operated on using the latter approach were found to have both an increased rate of compete resection (92.5% versus 66.5%) and improved visual function (52% versus 26%). The extent of resection was estimated based on follow-up CT or 190

MR imaging performed 3 months after surgical resection. As intra-operative MRI improves and becomes more widely available as a surgical adjunct it is likely to prove a valuable tool in improving the extent of resection. Nimsky et al. analysed the role of intra-operative MRI in resecting pituitary macroadenomas via the TSA (Nimsky et al., 2006). They reported 85 patients in whom complete tumour removal was intended, and 21 patients in whom partial tumour removal was intended. Intra-operative MRI led to an extensive resection in 34% of the former group and in 38% of the latter group. In addition, complete tumour removal rates increased from 58% to 82%, attributed to MRI in the former group. Further landmark studies with intraoperative MRI are envisaged to continue the development of this neurosurgical subspecialty. Focused radiation therapy has also been a landmark in the treatment of pituitary tumours. In managing patients with pituitary adenomas, the neuro-endocrine team must accurately assess whether the patient would benefit from adjuvant therapy. This assessment requires an evaluation of postoperative hormone levels, MRI for residual tumour, and physical examination findings for improvement or resolution of signs and symptoms secondary to the lesion. For those patients requiring adjuvant therapy, radiosurgery serves as a valuable tool for treating residual tumour. Understanding the radiation tolerance of functioning pituitary tissue is important in maximizing post-treatment pituitary function. Vladyka et al. performed a review of 63 patients who had received gamma knife radiosurgery for pituitary lesions to determine the sensitivity of hypophyseal function to focal radiation (Vladyka et al., 2003). In this study, the authors compared patients who experienced worsening pituitary function after radiosurgery to those patients whose pituitary function was unchanged. The data revealed a safe radiation dose to the hypophysis of 15 Gy for thyrotropic and gonadotropic tumours and 18 Gy for adrenocorticotropic tumours. An association was also found between the level of radiation to the infundibulum and decreased pituitary function. Although we have not included this study in a separate section within this chapter, we would consider it a landmark study.

References Bills DC, Meyer FB, Laws ER Jr, Ebersold MJ, Scheithauer BW, Ilstrup DM, Abboud CF. A retrospective analysis of pituitary apoplexy. Neurosurgery 1993; 33: 608–609. Ciric I, Ragin A, Baumgartner C, Pierce D. Complications of transsphenoidal surgery: results of a national survey, review of the literature, and personal experience. Neurosurgery 1997; 40: 225–237. Dolenc VV. Transcranial epidural approach to pituitary tumours extending beyond the sella. Neurosurgery 1997; 41: 542–552. Kaptain GJ, Vincent DA, Sheehan JP, Laws ER Jr. Transsphenoidal approaches for the extracapsular resection of midline suprasellar and anterior cranial base lesions. Neurosurgery 2001; 49: 94–100. Knosp E, Steiner E, Matula K, Kitz K, Matula C. Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classification compared with surgical findings. Neurosurgery 1993; 33: 610–618. Nimsky C, Keller BV, Ganslandt O, Fahlbusch R. Intraoperative high-filed magnetic resonance imaging in transsphenoidal surgery of hormonally inactive pituitary macroadenomas. Neurosurgery 2006; 58: 105–113. Tabaee A, Anand VK, Barrón Y, Hiltzik DH, Brown SM, Kacker A, Mazumdar M, Schwartz TH. Endoscopic pituitary surgery: a systematic review and meta-analysis. J Neurosurg 2009; 111: 545–554. Thapar K, Kovacs K, Scheithauer BW, Stefanbeanu L, Horvarth E, Pernicone P, Murray D, Laws ER. Proliferation and invasiveness among pituitary adenomas and carcinomas: an analysis using the MIB-1 antibody. Neurosurgery 1996; 38: 99– 107. Tyrrell JB, Lamborn KR, Hannogan LT, Applebury CB, Wilson CB. Transsphenoidal microsurgical therapy of prolactinomas: initial outcomes and long-term results. Neurosurgery 1999; 44: 254–261. Vladyka V, Liscák R, Novotný J Jr, Marek J, Jezková J. Radiation tolerance of functioning pituitary tissue in gamma knife surgery for pituitary adenomas. Neurosurgery 2003; 52(2): 309–316.

7.1 Timing of Surgical Intervention for Pituitary Apoplexy Details of Study Visual deficits as a result of pituitary apoplexy are an uncommon but serious complication of spontaneous pituitary haemorrhage. The accepted management strategy is urgent surgical decompression of the sellar contents. In order to evaluate the improvement of visual defects as a function of timing of surgery, a retrospective analysis of case histories of patients presenting with pituitary apoplexy to the Department of Neurological Surgery at the Mayo Clinic in Rochester, Minnesota, USA, between 1975 and 1991 was performed.

Study Reference Main Study Bills DC, Meyer FB, Laws ER Jr, Ebersold MJ, Scheithauer BW, Ilstrup DM, Abboud CF. A retrospective analysis of pituitary apoplexy. Neurosurgery 1993; 33: 608–609.

Study Design Class of evidence

III 191

Randomization

None (see following list)

Number of patients

Length of follow-up

2.8 years (average)

Number of centres

Stratification

None

The aim of the study was to assess the recovery of visual deficits from pituitary apoplexy as a function of surgical timing. It was a retrospective analysis of patient case histories presenting to a single institution (Mayo Clinic, USA). Only patients who presented with abrupt headache or visual disturbance and evidence of bleeding into a pituitary adenoma were included. The patients were subdivided into those who received surgery within 3 days of presentation, between days 4â&#x20AC;&#x201C;7 post-presentation, and >1 week following presentation.

Outcome Measures Primary Endpoints Improvement in visual deficits: ocular paresis, visual field defects, or reduction in visual acuity. Complete resolution of visual acuity was defined as at least 20/25 vision in both eyes or a return to baseline prior to apoplexy.

Secondary Endpoints Data regarding the incidence of pituitary apoplexy and the clinical features of the syndrome were also collected. The endocrinological outcome of patients was also reviewed.

Results In this analysis, 2034 pituitary adenomas were managed surgically and included 37 patients with apoplexy. The study revealed an annual incidence of pituitary apoplexy of 1.9% in patients with pituitary adenomas undergoing surgery. The most common presenting feature was headache. At presentation the most common visual defects were: ocular paresis (78%); visual field defects (64%); and reduction in visual acuity (52%). The most commonly affected cranial nerve in ocular paresis was the oculomotor nerve (58%), followed by abducens (30%) and trochlear (12%). The improvements in visual defects following surgery are summarized here:

Improvement in visual acuity was significantly better for those undergoing surgery within 1 week (complete resolution in all patients) compared to those undergoing surgery >1 week from onset (complete resolution in less than half of patients). There was no significant difference between those undergoing surgery within 3 days and those undergoing surgery between 4â&#x20AC;&#x201C;7 days. Long-term steroid therapy for pituitary insufficiency was necessary in 82% of patients.

Conclusions In pituitary apoplexy patients with a stable conscious level and stable but impaired visual function, decompressive surgery within 1 week is recommended to optimize visual outcome. More urgent surgery is not necessary but delays beyond 1 week may impair visual recovery.

Critique The pituitary apoplectic syndrome is an uncommon but serious complication of pituitary adenoma. There is, of course, a spectrum of apoplexy from incidental radiological findings to patients who are critically ill with deteriorating visual function. The main differential diagnoses are SAH and bacterial meningitis. The most sensitive imaging modality is MRI although the diagnosis may be made on CT. The immediate management of the apoplectic patient is supportive, with close monitoring of vital signs and visual function. Administration of corticosteroids is critical to avoid cardiovascular collapse due to adrenocorticotropic hormone insufficiency. Expert endocrinology advice should 192

be sought, and intravenous hydrocortisone is, therefore, the mainstay of immediate medical management (Kearney et al., 2006). In addition, correction of significant hyponatraemia is advised prior to surgery. The role of surgery in pituitary apoplexy has been controversial. In performing an appropriate assessment of a pituitary lesion, the surgeon must determine the urgency by which critical neural structures must be decompressed. One of the most important of these structures is the optic apparatus. Of the various pituitary lesions, pituitary apoplexy is the most critical to diagnose, as early decompression is associated with improvement in visual deficits. Bills et al. analysed 37 patients with symptomatic pituitary apoplexy and found ocular paresis, a reduction in visual fields, and a reduction in visual acuity in the majority of patients. Surgery resulted in improvement in 100%, 95%, and 88% of these symptoms, respectively. The degree of visual improvement was found to be significantly improved in those patients who underwent surgery within 1 week of the onset of symptoms. The timing of surgical intervention is especially important in conscious patients with residual vision in each eye.

Reference Kearney T, Dang C. Diabetic and endocrine emergencies. Postgrad Med J. 2007; 83(976): 79–86.

7.2 Classification of Cavernous Sinus Invasion by Pituitary Adenomas Details of Study MRI scanning significantly improved upon CT imaging to delineate pituitary lesions and define the degree of invasiveness. This study, carried out in Mainz, Germany, and Vienna, Austria, established a radiological classification scheme in order to determine when critical invasion of the cavernous sinus can be determined on MRI.

Study Reference Main Study Knosp E, Steiner E, Matula K, Kitz K, Matula C. Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classification compared with surgical findings. Neurosurgery 1993; 33: 610–618.

Study Design Clinical study comparing surgical and radiological findings. Case series of 25 patients. A midsellar coronal MRI scan with gadolinium enhancement was chosen as the reference plain in order to visualize the intracavernous and supracavernous ICA. Three intercarotid lines are used in this plane to determine invasion of the carotid sinus (Figure 7.1).

Fig. 7.1 A coronal MRI scan (T1 + gadolinium) of a pituitary macroadenoma (arrow) showing the three intercarotid lines: the medial tangent (A) which runs between the medial aspects of the intra- and supracavernous ICA, the ‘intercarotid line’ (B)

193

passing through the centre of the intra- and supracavernous ICA, and the lateral tangent (C) which runs between the lateral intraand supracavernous ICA.

Grade 0

Normal findings within the cavernous space The adenoma does not pass beyond the medial intercarotid line

Grade 1

The medial tangent is passed but with no extension beyond the intercarotid line

Grade 2

The tumour extends beyond the intercarotid line but not beyond the lateral tangent

Grade 3

Tumour extends beyond the lateral tangent

Grade 4

The tumour totally encases the intracavernous carotid artery

Grades of invasion were defined depending on how far there was extension beyond the respective lines.

Results All patients underwent surgery and cavernous sinus invasion was seen in at least one side. Radiological grade

Percentage of sides in which cavernous sinus invasion was seen at surgery

Grade 0

Grade 1

Grade 2

88%

Grade 3

86%

Grade 4

100%

Conclusions Pituitary adenomas which are radiologically Grade 2 or more are highly likely to have cavernous sinus invasion confirmed at surgery.

Critique The surgical treatment of pituitary lesions requires the surgeon to understand the degree of regional dural invasion by the adenoma in order to assess the most appropriate treatment options. Pituitary adenomas may invade the cavernous sinus in 6–10% of cases (Ahmadi et al., 1985; Fahlbusch and Buchfelder, 1988). The degree of invasion can impact the surgeon’s ability to achieve gross total resection and may necessitate adjuvant treatment. In an effort to better classify invasion, Knosp et al. compared surgical results to pre-operative MRI in 25 patients with suspected invasive lesions. They defined five grades of invasion, ranging from 0–4, based on the level of encroachment on a tangential line between the intra- and supracavernous internal carotid artery in coronal MR images. The data showed that invasion past this intracarotid line, defined as Grade 2, suggested that the adenoma invaded into the cavernous sinus space. In Grade 2 patients, seven of eight patients showed invasion of the medial wall of the cavernous sinus during surgery. On the other hand, the Grade 0 or Grade 1 patients displayed no definitive cavernous sinus invasion. In addition, a major degree of invasion correlated with a statistically significant increase in tumour size. There have been several classification schemes for pituitary adenomas. The first was by Hardy and this was further modified by Wilson (Hardy, 1976; Hardy, 1983; Wilson, 1984). Other classification schemes include those of Fahlbusch and Buchfelder (Fahlbusch and Buchfelder, 1988). The impetus to classify the tumours radiologically was to validate a system that was relevant both in terms of tumour biology and surgical planning. This classification scheme by Knosp et al. specifically takes into account parasellar extension, and correlated well with surgical findings.

References Ahmadi J, North CM, Segall HD, Zee CS, Weiss MH. Cavernous sinus invasion by pituitary adenomas. AJNR 1985; 6: 893– 898. Fahlbusch R, Buchfelder M. Transsphenoidal surgery of parasellar pituitary adenomas. Acta Neurochir (Wien) 1988; 92: 93– 99. Hardy J, Vezina JL. Transsphenoidal neurosurgery of intracranial neoplasm, in Tompson RA, Green JR (eds), Advances in Neurology. New York: Raven Press, 1976, vol 15, pp 261–275. Hardy J. Transsphenoidal microsurgery of prolactinomas: report on 355 cases, in Tolis G, Stefanis C, Mountokalakis T (eds), Prolactin and Prolactinomas. New York: Raven Press, 1983, pp 431–440. Wilson CB. A decade of pituitary microsurgery. J Neurosurg 1984; 61: 814–833.

194

7.3 Biological Correlates of Invasive Adenomas Details of Study Although pituitary adenomas are histologically benign lesions, they can exhibit aggressive local growth with invasion of the surrounding structures in approximately one-third of cases. This local invasion poses a difficult challenge to neurosurgeons as it impacts adversely on disease-free survival and surgical cure. Previous studies had described the nature of invasion by pituitary adenomas. This clinicopathological study was carried out by four centres in North America: Toronto, Ontario (Canada); Rochester, Minnesota (USA); Orlando, Florida (USA), and Charlottesville, Virginia (USA).

Study References Main Study Thapar K, Kovacs K, Scheithauer BW, Stefabeanu L, Horvath E, Pernicone P, Murray D, Laws ER. Proliferation and invasiveness among pituitary adenomas and carcinomas: an analysis using the MIB-1 antibody. Neurosurgery 1996; 38: 99– 107.

Related References Matsuyama J. Ki-67 expression for predicting progression of postoperative residual pituitary adenomas: correlations with clinical variables. Neurol Med Chir (Tokyo) 2012; 52: 563–569. Scheithauer BW, Kovacs KT, Laws ER Jr, Randall RV. Pathology of invasive pituitary tumours with specific reference to functional classification. J Neurosurg 1986; 65: 733–744.

Study Design Clinicopathological study of 70 patients undergoing surgical resection of a pituitary macroadenoma. The degree of invasiveness of the tumours studied was defined on the basis of intra-operative and MRI findings. The monoclonal antibody MIB-1 was used to determine growth fractions based on expression of the Ki-67 cell cycle-specific nuclear antigen which has been shown to be associated with proliferative activity. Control specimens (positive and negative) were obtained from post-mortem specimens. The study also included seven cases of pituitary carcinoma. The primary aim of the study was to evaluate the MIB-1 antibody as a practical tool to evaluate the proliferative activity of pituitary tumours. The secondary aim of the study was to evaluate the value of this marker in discriminating aggressive and locally invasive tumours from more indolent tumours.

Results Thirty-seven non-invasive adenomas, 33 invasive adenomas, and seven pituitary carcinomas were analysed.

KI-67 labelling index was statistically different between non-invasive and invasive adenomas (p = 0.0029). A threshold labelling index of >3% with MIB-1 was established to distinguish invasive from non-invasive adenomas and determine specificity, sensitivity and predictive values.

Conclusions MIB-1 antibody is an effective method of inferring that a tumour may be biologically aggressive. The KI-67 labelling index is a useful piece of biological information about pituitary tumours and it can guide the clinical team in determining the potential for more rapid symptomatic regrowth following incomplete excision.

Critique Various biological factors are thought to influence the invasive potential of pituitary adenomas. With one-third of 195

pituitary adenomas invading surrounding structures, an understanding of these biological factors can influence postoperative patient management. The MIB-1 antibody is an effective measure of the Ki-67 labelling index, a measure of growth fraction: Ki-67 is a nuclear antigen expressed in late phases of the cell cycle. This landmark study by Thapar et al. analysed this labelling index and found that a Ki-67 >3% correlated with a more biologically aggressive tumour, and consequently, a tumour capable of more invasive growth. Such a tumour would not only be more difficult to resect completely but would also theoretically be subject to faster re-growth. The management team can use such information to determine the role of adjuvant therapy. Since this early study by Thapar et al. there have been several studies confirming the finding that high Ki-67 values predict more aggressive tumour progression. A study from Fukushima in Japan has found that a Ki-67 index of 2.0 or more predicts tumour progression with high specificity (Matsuyama, 2012). The role of biological correlates predictive of tumour aggression in daily clinical practice remains controversial. However, if confirmed in larger studies, then patients with a higher Ki-67 index could be considered for earlier adjuvant therapy, and certainly for shorter follow-up imaging intervals. This study by Thapar et al. is a landmark in establishing the principle.

Reference Matsuyama J. Ki-67 expression for predicting progression of postoperative residual pituitary adenomas: correlations with clinical variables. Neurol Med Chir (Tokyo) 2012; 52: 563â&#x20AC;&#x201C;569.

7.4 Complications of Pituitary Surgery Details of Study This questionnaire study constitutes the largest national survey of the complications of pituitary surgery. It was carried out by a group at the Northwestern University Medical School, Evanston, Illinois, USA. A total of 3172 neurosurgeons were included in the study.

Study Reference Main Study Ciric I, Ragin A, Baumgartner C, Pierce D. Complications of transsphenoidal surgery: results of a national survey, review of the literature, and personal experience. Neurosurgery 1997; 40: 225â&#x20AC;&#x201C;237.

Study Design National questionnaire survey. The primary objective of the study was to determine the incidence of complications of transsphenoidal surgery. Secondary objectives included assessment of how surgical experience with the procedure influenced complications and to review causation, treatment, and prevention of complications. Fourteen specific complications were included in the questionnaire.

Results Neurosurgeons were divided into three groups related to extent of surgical experience: those who had performed < 200 operations; those who had performed 200â&#x20AC;&#x201C;500 operations; and those who had performed > 500 operations. The most common operative complications were anterior pituitary dysfunction (19.4%), diabetes insipidus (17.8%), and sinusitis (8.5%). Neurological complications were seen in <2% of operations: central nervous system injury (1.3%); ophthalmoplegia (1.4%); loss of vision (1.8%). CSF leak was seen in 3.9% of operations and meningitis in 1.5%. Carotid injury was seen in 1.1% and death occurred in 0.9% of operations. The complication rate was inversely correlated with surgical experience and this finding was statistically significant for all complications (p < 0.001 for 13 out of 14 of the complications).

Conclusions Transsphenoidal surgery is a reasonably safe procedure with a mortality rate of <1%. However, the incidence of complications is higher in the hands of less experienced surgeons.

Critique The treatment of pituitary adenoma requires the neuro-endocrine team to accurately assess each individual patient. The complications in the management of the pituitary lesion are a direct reflection of the experience of this team in choosing the appropriate combination of medical, surgical, and adjuvant therapies. Ciric et al. meticulously analysed the complications encountered by 958 neurosurgeons who performed transsphenoidal surgery, using questionnaires regarding 14 specific complications of this operation. Of the respondents, 3% reported having performed >500 196

operations, 9.7% reported having performed between 200 and 500 operations, and 87.3% reported having performed <200 operations. The data indicated that the incidence of complications is significantly lower in surgeons having performed >200 transsphenoidal operations, and even lower in those with >500 operations. This decrease in complications is hypothesized to be a direct result of an improved understanding of the surgical anatomy and, perhaps more importantly, a better understanding of the indications for surgical intervention. The logical conclusion is that pituitary surgery should be undertaken in centres with a high caseload to ensure that experienced surgeons are undertaking the procedures to minimize complications. Indeed, there has been discussion about pituitary centres of excellence being established (McLaughlin et al., 2012). Despite convincing arguments in favour of establishing such centres, this may not be practical in certain parts of the world where case volume is much lower. Furthermore, there may be factors other than caseload that could be altered to reduce complications. The neurosurgical unit in Wellington, New Zealand, published the results of their preliminary experience with endoscopic endonasal transsphenoidal surgery, and found that their complication rates were minimal (Yang et al., 2012). This study by Yang et al. demonstrated that it is possible for smaller units performing <200 cases per year to be able to achieve comparable results to larger units, although it should be recognized that it is difficult to make comparison between studies carried out 15 years apart.

References McLaughlin N, Laws ER, Oyesiku NM, Katznelson L, Kelly DF. Pituitary centres of excellence. Neurosurgery 2012; 71: 916–923. Yang JY, De Ruiter I, Parker A, Wormald PJ, Robinson S, Wickremesekera A. Endoscopic endonasal transsphenoidal surgery: a mentoring surgical model. ANZ J Surg 2012; 82: 452–456.

7.5 Surgical Treatment of Prolactinomas Details of Study This study analysed the efficacy of surgery to control pituitary prolactinomas as an alternative to long-term medical treatment. The study was carried out in the Department of Neurological Surgery at the University of California, San Francisco, California, USA.

Study Reference Main Study Tyrrell JB, Lamborn KR, Hannogan LT, Applebury CB, Wilson CB. Transsphenoidal microsurgical therapy of prolactinomas: initial outcomes and long-term results. Neurosurgery 1999; 44: 254–261.

Study Design A retrospective review of 219 patients treated in a single institution over two time periods: 1976–1979 (Group 1, 121 patients) and 1988–1999 (Group 2, 98 patients). Endocrinological follow-up (84%) was a mean of 15.6 years (Group 1) and 3.2 years (Group 2). Clinical information was available for 91% of patients. Exclusion criteria: age <18 years; previous pituitary surgery or radiation therapy. All patients in the study were female.

Outcome Measures Remission was defined as a prolactin level remaining in the normal range for the laboratory they were tested in preoperatively.

Results Eighty-five per cent of patients experiencing initial endocrinological remission remained in remission at the latest follow-up.

Intrasellar microprolactinomas with pre-operative prolactin levels <100 ng/mL

Initial surgical remission 92%

Intrasellar microadenomas

91%

Larger invasive macroadenomas and patients with pre-operative prolactin levels >200 ng/mL

37–41%

197

Conclusions Surgery is a safe and effective alternative to medical treatment for the long-term control of prolactinomas in selected patients.

Critique Endocrinological evaluation of pituitary lesions by the neuro-endocrine team is critical in fully diagnosing the nature of the lesion. A pre-operative endocrinological assessment is particularly important in managing lesions that have effective medical treatment options. Prolactinomas serve as the prime example of pituitary adenoma subtypes conducive to non-invasive treatment options. Bromocriptine, a long-acting dopamine agonist, was shown in the 1970s to control prolactin hypersecretion, control prolactinoma growth, and normalize prolactin values in >80% of patients. A newer, longer-acting dopamine agonist, cabergoline, has helped to reduce side effects and displays a similar response rate to bromocriptine. Despite the advantages of dopamine agonists, medical treatment commonly requires lifelong therapy. Some patients may develop intolerance or resistance to the medication. In an effort to assess the role of surgery as an alternative treatment option, Tyrrell et al. retrospectively reviewed the records of 219 patients with prolactinomas who had undergone transsphenoidal microsurgery. The authors reported that women with pre-operative prolactin levels of >200 ng/mL and those with larger and more invasive tumours experienced an initial remission rate between 37% and 41%. In comparison, women with pre-operative prolactin levels of <100 ng/mL and smaller and less invasive tumours experienced an initial remission rate between 83% and 92%. Of the women who experienced initial remission, 89% continued to have remission of clinical symptoms and 85% continued to have normal prolactin levels. Consequently, the authors concluded that a high initial post-operative remission rate could be anticipated in patients with microadenomas or non-invasive macroadenomas, as well as in patients with relatively low pre-operative prolactin levels. They stressed the importance of considering surgical resection as a legitimate treatment option in certain patients with prolactinomas and in patients unresponsive to or intolerant of medical management.

7.6 Extended Transsphenoidal Approaches to Anterior Cranial Base Lesions Details of Study This paper describes experience with 14 patients undergoing an extended TSA to resect midline suprasellar and anterior cranial base lesions. The surgery was carried out by Edward Laws Jr at the Department of Neurosurgery, University of Virginia, Charlottesville, Virginia, USA, between 1999 and 2000.

Study Reference Main Study Kaptain GJ, Vincent DA, Sheehan HP, Laws ER Jr. Transsphenoidal approaches for the extracapsular resection of midline suprasellar and anterior cranial base lesions. Neurosurgery 2001; 49: 94â&#x20AC;&#x201C;100.

Study Design A prospective surgical series of patients all undergoing transsphenoidal procedures by Edward Laws Jr.

Outcome Measures Extent of resection was confirmed on follow-up MRI at 3 and 6 months post-operatively. Endocrinological assessment was made during the hospital stay. All complications were recorded.

Results The majority of tumours were resected using a transsphenoidal/transtubercular approach (71%) and the others were resected using a transsellar/transdiaphragmatic approach (29%). In all cases the tumour capsule was dissected away from the optic nerves.

198

In patients with pre-existing visual field defects there was visual improvement in 33%.

Conclusions Gross total extracapsular resection of suprasellar tumours is possible via the TSA.

Critique Over the past 40 years, the TSA to sellar and parasellar lesions has become a mainstay of the neurosurgical practice in treating pituitary lesions. The standard approach to the sella via the sphenoid sinus has been expanded to resect lesions that extend to a suprasellar location as well as along the anterior cranial base. Kaptain et al. analysed patients with various lesions in these locations utilizing either a transsellar/transdiaphragmatic approach or a transsphenoidal/transtubercular approach. An extracapsular tumour resection was performed followed by placement of abdominal fat for repair of the dural defect as well as reconstruction of the sellar floor and planum sphenoidale. Gross total resection was achieved in all patients with pituitary adenomas and the majority of these patients displayed improved visual function. Adjunctive endonasal endoscopy was used to assist with the microscopic resection of the tumour.

7.7 Endoscopic Transsphenoidal Surgery Details of Study Endoscopic transsphenoidal pituitary surgery has developed over the last 15 years based on various theoretical advantages including improved visualization, reduced sinonasal trauma, potentially better resection rates, and reduced complications. In order to describe the complication rates and short-term outcomes of this technique a group at the New York Presbyterian-Weil Medical College of Cornell University in New York, USA, carried out a systematic review and meta-analysis of published endoscopic results prior to 2006.

Study References Main Study Tabaee A, Anand VK, Barrón Y, Hiltzik DH, Brown SM, Kacker A, Mazumdar M, Schwartz TH. Endoscopic pituitary surgery: a systematic review and meta-analysis. J Neurosurg 2009; 111: 545–554.

Related Reference Cappabianca P, de Divitiis E. Endoscopy and transsphenoidal surgery. Neurosurgery 2004; 54: 1043–1048.

Study Design Systematic review and meta-analysis. A random effects model was used to assess heterogeneity.

Outcome Measures Complications of endoscopic surgery. Short-term outcomes including extent of resection and improvement in visual function.

Results Nine studies met the inclusion criteria with a total of 821 patients. Reporting discrepancies between studies made it difficult to assess post-operative outcomes and complication rates. Mean operating time was 102–255 minutes. 199

Average hospital stay was 1.4–4.4 days. Improvement in pre-operative visual defects ranged from 62–100%. Pooled gross total tumour removal rate was 78%. Complication rates were compared to published ‘open’ TSA operations. Endoscopic TSA

‘Open’ TSA

CSF leak

1–4%

Permanent diabetes insipidus

0.5–15%

Epistaxis

<1%

1–4%

Death (vascular injury)

0.24%

<1%

Conclusions Endoscopic pituitary surgery is safe and effective in the short term.

Critique The difficulties of performing a randomized trial of endoscopic techniques versus open surgery mean that the impact of endoscopic techniques remains to be determined. Nonetheless, as with all innovations in surgical techniques key protagonists will continue to push the boundaries until it becomes a necessity of other surgeons to follow suit in order to remain on the same playing field. Although systematic review and meta-analysis are limited as a means of assessing the validity of surgical techniques, this study by Tabaee et al. has been included as it is the best attempt at describing the complications and short-term outcomes of endoscopic pituitary surgery. The authors point out that, inevitably, early publications on endoscopic techniques have focused on technical aspects of the procedure which makes assessment problematic. Furthermore, the lack of control groups in case series means that comparison needs to be made with published historical data. Notwithstanding these limitations, this is the first assessment of this modality that was not a single institution series. A minimally invasive TSA using the endoscope as a primary means of visualization has become a valuable alternative to the microscope over the past decade. Cappabianca and de Divitiis have provided a thorough description of this technique as well as the advantages as compared to the traditional approach (Cappabianca and de Divitiis, 2004). These advantages include an improved view of the anatomy permitted by the endoscope as well as the possibility of shorter hospitalization stays for the patient. Furthermore, the advancement of the endoscope into the tumour cavity at the end of the case may allow for identification and removal of residual tumour and result in more complete resection. As surgeons continue to use the endoscope for a variety of pituitary lesions, the indications and outcomes for the use of the microscope versus the endoscope will become better understood.

Reference Cappabianca P, de Divitiis E. Endoscopy and transsphenoidal surgery. Neurosurgery 2004; 54: 1043–1048.

200

Index A Abboud, CF 345–347 Abdu, WA 211–216 Abi-Said, D 85–89 ablative therapy, trigeminal neuralgia 305–306 Abrams, R 109–113 acetazolamide, post-haemorrhagic ventricular dilatation 321–323 acute spinal cord injury (ASCI) methyl prednisolone (NASCIS) 197–199 timing of surgery 205–206 acute subdural haematomas, timing of surgery 133–137 Adams, CBT 302–303 Adams, HP 5–8, 27–29 Adams, WE 147–151 adjuvant radiotherapy, solitary brain metastases 77–79 Afra, D 109–113 Agid, Y 271–274 Ahn, NU 221–224 Ahn, UM 221–224 Albanese, J 161–167 Albert, T 225–227 Albright, R 99–103 Alderson, P 335 Aldrich, EF 197–199 Algra, A 45–50 Allenberg, JR 55–58 Allgeier, A 91–93 Alliez, B 161–167 Al-Shahi, R 37 Amelink, GJ 45–50 Amtage, F 271–274 An, H 225–227 Anand, VK 363–364 Anderson, B 85–89 Anderson, GD 179–181 Anderson, VC 267 Andrews, DF 161–167 Andrews, DW 81–84 aneurysm clipping, ISAT 13–17 Annegers, JF 183–185 anorexia nervosa, deep brain stimulation 241 anterior communicating aneurysms, comparison of treatments 3 antipsychotics 241 Antonini, R 161–167 Aoyama, H 81–84 Applebury, CB 359–360 Arabi, YM 143–146 Ardouin, C 258–259 Arena, VA 99–103 Armstrong, D 161–167 arteriovenous malformations (AVMs) 4 natural history 37 Spetzler–Martin grading system 35–37 Arusell, R 109–113 Astradsson, A 311–315 astrocytoma, anaplastic CATNON trial 93 see also high-grade gliomas Asymptomatic Carotid Atherosclerosis Study (ACAS) 51, 53 Attenello, FJ 105–108

201

Ayers, S 321–323

B Babu Krishnamurthy, K 251–254 back pain, chronic failed back surgery syndrome, spinal cord stimulation 283–286 spinal stabilization 229–230 Bacon, A 189–190 Bahary, JP 81–84 Bailey, S 23–25 Bakay, R 251–254 Ballantyne, HT 241 balloon compression, trigeminal neuralgia 306 Balsderston, RA 205–206 Baltuch, G pallidal versus subthalamic DBS 267–269 SANTE trial 251–254 Barbaro, N 251–254 Barbe, MT 271–274 Barber, J intracranial pressure monitoring 147–151 magnesium sulphate therapy in head injury 179–181 barbiturates, use in head injury 157–159, 164, 165 Barer, DH 39–41 Barker, FG 301–303 Barker, K 229–230 Barnes, MA 333–336 Barnett, GH 99–103 Barnett, HJ 51–54 Barr, JS 223 Barrón, Y 363–364 Barth, M 217–219 Baskin, DS 197–199 Bataille, B 271–274 Batir, A 258–259 Batjer, HH 37 Baumgartner, C 357–358 Bazil, C 251–254 Beck, DW 27–29 Becker, DP extra-axial haematomas, timing of surgery 133–137 hyperventilation in head injury 175–177 intracranial pressure monitoring 147–151 Bednanik, J 231–232 Belander, K 91–93 ‘Believers trial’, radiotherapy for low-grade gliomas 109–113 Benabid, AL 253, 258–259 Benazzouz, A 258–259 Benchmark Evidence from South American Trials: Treatment of Intracranial Pressure (BEST:TRIP) 147–151 Benecke, R 275–278 Ben Hassel, M 109–113 Ben-Menachem, E 249–251 Bergen, D 251–254 Berger, J 55–58 Berger, MS 85–89 Bernard, SS hyperosmolar therapy for raised ICP 161–167 pre-hospital intubation in head injury 189–190 Bernstein, M 99–103 Berven, S 225–227 Bills, DC 345–347 Bissonette, DJ 301–303 Bittar, RG 239 Bjorklund, A 311–315 Black, K 95–98 Blackwood, MS 99–103

202

Bleehen, N 109–113 Blood, E 225–227 Bloomfield, S 99–103 Blume, WT 245–247 Boden, SD surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Bogdahn, U 91–93 Bohn, DJ 333–336 Bond, M 127–130 Boop, F 325–327 Bortey, E 95–98 Bötzel, K 259–261 Bourgouin, A 161–167 Bousser, MG 45–50 Boutron, C 45–50 brachytherapy, high-grade gliomas 62 Bracken, MB 197–199 brain metastases 61 recursive partitioning analysis (RPA) classification 74 solitary adjuvant radiotherapy 77–79 surgical resection 69–74 stereotactic radiosurgery 81–84 Brain Tumour Cooperative Group, trial of brachytherapy 99–103 brain tumours classification 115 see also brain metastases; dysembryoplastic neuroepithelial tumours; high-grade gliomas; low-grade gliomas; meningiomas Brand, R surgery for lumbar disc herniation 211–216 surgery for solitary brain metastases 69–74 Brande, AA 91–93 Brandt, L 233–234 Brefel Courbon, C 271–274 Brem, H extent of resection of low-grade gliomas 105–108 localized chemotherapy for malignant gliomas 95–98 Brem, S 95–98 Brentrup, A 275–278 Bricolo, AP 136 British Aneurysm Nimodipine Trial 23–25 Broggi, G 293–295 bromocriptine therapy, prolactinoma 360 Bronstein, J, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal DBS 267–269 Brook, N 127–130 Brown, RD Jr 19–21 Brown, SM 363–364 Brundin, P 311–315 Bucholz, RD 169–173 Buchowski, MS 221–224 Buchser, E 283–286 Bullock, MR 137 Burchiel, K 261–263, 267–269 Burckhardt, G. 240 Burger, PC brachytherapy for malignant gliomas 99–103 carmustine wafer therapy 95–98 extent of resection of low-grade gliomas 105–108 Burns, B 289–291 burr hole evacuation, chronic subdural haematomas 139–142 Bussone, G 293–295 Butt, W 329–331 Butterworth, JF 147–151

C 203

cabergoline therapy, prolactinoma 360 Cairncross, JG 72, 91–93 Cameron, P 189–190 Cammisa, F 225–227 Campbell, MJ 321–323 Cantu, RC 202 Cappabianca, P 364 Carlsson, CA 233–234 carmustine wafer therapy 62, 95–98 Carney, N 147–151 Carol, W 79 carotid endarterectomy versus carotid stenting (SPACE study) 55–58 versus conservative treatment (NASCET) 51–54 Carpentier, A 45–50 Cascino, T 109–113 CATNON trial 93 cauda equina syndrome, surgical intervention 221–224 Cavanagh, J 116 cavernous sinus invasion, pituitary adenomas 349–351 Celix, K 147–151 cerebral oedema, tumour-associated, dexamethasone therapy 61, 65–67 cervical dystonia, deep brain stimulation 278–281 cervical radiculopathy, surgical intervention 233–234 cervical spondylotic myelopathy, surgical intervention 231–232 Cha, S 105–108 Chabal, S 187–188 Chabardes, S 258–259 Chaddock, K 147–151 Chaichana, KL 105–108 Chaloupka, R 231–232 Chang, EF 105–108 Chang, HS 221–224 Chang, SM 105–108 Chaynes, P 271–274 chemotherapy for low-grade gliomas 113 temozolomide trial 91–93 Cherner, M 147–151 Chestnut, RM hypothermia in head injury 169–173 intracranial pressure monitoring 147–151 Chia, HL 139–142 Chkhenkeli, IS 253 Chkhenkeli, SA 253 chlorpromazine 241 Chodkeiwicz, JP 115–116 Choi, SC extra-axial haematomas, timing of surgery 133–137 hyperventilation in head injury 175–177 hypothermia in head injury 169–173 intracranial pressure monitoring 147–151 Christensen, J 183–185 chronic back pain, spinal stabilization 229–230 chronic subdural haematomas (CSDHs), surgical management 139–142 Chu, Y 311–315 Chung, S 251–254 Cinalli, G 325–327 cingulate gyrus stimulation, for treatment-resistant depression 307–310 cingulotomy 241 Ciric, I 357–358 Clagett, GP 51–54 Clarke, M 13–17 Clarke, PR 202 Claus, EB 107–108 Clifton, GL

204

hypothermia in head injury 169–173 NASCIS 197–199 cluster headache (CH) deep brain stimulation 293–295 occipital nerve stimulation 289–291 Coan, SP 183–185 Cochrane, DD 325–327 Coffey, CS, hypothermia in head injury 169–173 Coffey, R 251–254 Collette, L 109–113 Collings, WF 197–199 Collins, R 229–230 Collins, SD 249–251 Compagne, C 137 composite analysis 181 computed tomography, predictors of vasospasm 9–11 Conly, A 169–173 conscious level assessment, Glasgow Outcome Scale 127–130 Contant, CF 157–159 Cooper, DJ DECRA trial 143–146 hyperosmolar therapy for raised ICP 161–167 pre-hospital intubation in head injury 189–190 Cooper, G 161–167 Cooper, IS 253 Cornu, P 109–113 corticosteroids see steroids Corticosteroids Randomization After Significant Head Injury (CRASH) trial 153–155 Cotle, JM 205–206 Couvreur, G 45–50 cranial base lesions, extended transsphenoidal approaches 361–362 craniectomy, decompressive in head injury DECRA trial 143–146 in paediatric patients 329–331 in malignant MCA infarction (DECIMAL trial) 45–50 Cremer, OL 150 CREST trial 58 Crockard, HA 221–224 Cruz, J 161–167 Curran, W radiotherapy for low-grade gliomas 109–113 stereotactic radiosurgery for brain metastases 81–84 Curschmann, K 91–93 Cushing, H 340 Czosnyka, M 31–34

D Dagher, A 311–315 Daily, SW 161–167 Damier, P 271–274 Dandy, W. 302 Daniels, C 259–261 Dante, SJ 205–206 Darcel, F 109–113 Daugherty, RJ 205–206 Daumas-Duport, C 115–116 Davies, AR 143–146 Davis, JM 9–11 decompressive craniectomy in head injury DECRA trial 143–146 paediatric patients 329–331 in malignant MCA infarction (DECIMAL trial) 45–50 decompressive spinal surgery for cervical spondylotic myelopathy 231–232

205

for lumbar stenosis 225–227 for metastatic spinal cord compression 207–209 de Divitiis, E 364 deep brain stimulation (DBS) 239–241 for cluster headache 293–295 for dystonia cervical 278–281 generalized and segmental 275–278 for epilepsy 251–253 for obsessive-compulsive disorder 309 for Parkinson’s disease 257–266 early subthalamic stimulation 271–274 subthalamic versus pallidal 267–269 for treatment-resistant depression 307–310 Deep-Brain Stimulation for Dystonia Study Group 275–278 deep X-ray therapy (DXT), solitary brain metastases 79 DeGiorgio, CM 249–251 De Haan, A 339 Delevault, P 95–98 Delta valve, Shunt Design Trial 325–327 Demas, W 81–84 DeMonte, F 85–89 Dempsey, RJ adjuvant radiotherapy for solitary brain metastases 77–79 surgery for solitary brain metastases 69–74 De Palma, AF 234 depression, deep brain stimulation 307–310 DeSalles, A SANTE trial 251–254 subthalamic versus pallidal DBS 267–269 DESTINY (decompressive surgery for the treatment of malignant infarction of the MCA) trial 45–50 Deuschl, G 264 deep brain stimulation for dystonia 275–278 deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Deutsch, M 99–103 Deutschländer, A 259–261 De Witte, O 109–113 dexamethasone in metastatic spinal cord compression 201–203 for tumour-associated cerebral oedema 61, 65–67 Deyo, RA 211–216 Diamond, DL 133–137 Dickinson, K 123 Diepers, M 217–219 Dikmen, S intracranial pressure monitoring 147–151 magnesium sulphate therapy in head injury 179–181 phenytoin prophylaxis of post-traumatic seizures 187–188 Dillmann, U 259–261 Dinapoli, R 109–113 Dinning, TAR 221–224 Dirks, PB 333–336 ‘discectomy dogma’ 218–219 disc herniation, lumbar, surgical intervention 211–216 microscopic sequestrectomy 217–219 Ditchfield, M 329–331 diuretic therapy, in post-haemorrhagic ventricular dilatation 321–323 Dodick, DW 297–300 Doig-Beyaert, K 278–281 Dolenc, V 341–342 dopamine agonists, for prolactinoma 360 Dott, N 340 Douchette, S 333–336 drains, use after evacuation of chronic subdural haematoma 139–142 Drake, CG 27–29

206

Drake, JM 325–327 Drever, P 169–173 Duda, JE, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal DBS 267–269 Duncan, G 69–74 D’Urso, P 143–146 Durwood, QJ 27–29 Dušek, L 231–232 dysembryoplastic neuroepithelial tumours (DNTs) 62, 115–116 dystonia, pallidal deep brain stimulation cervical dystonia 278–281 generalized and segmental dystonia 275–278

E Earle, J 109–113 EARLYSTIM Study Group 271–274 Ebersold, MJ 345–347 Eckstein, HH 55–58 Edington, J 189–190 Eekof, JA 211–216 Eiliasziw, M 51–54 Eisenberg, E 283–286 Eisenberg, HM barbiturates in head injury 157–159 NASCIS 197–199 Eisenhauer, E 91–93 Eisner, W, deep brain stimulation for dystonia 275–278 Parkinson’s disease 259–261 Elbourne, D 321–323 Eldabe, S 283–286 Eliasziw, M deep brain stimulation for cervical dystonia 278–281 surgery for temporal lobe epilepsy 245–247 Ellenbagen, RG 179–181 endoscopic transsphenoidal surgery 363–364 endovascular coiling, ISAT 13–17 Engel, J 239 Englot, DJ 108 Englund, E 311–315 epilepsy 239 neuromodulation deep brain stimulation 251–253 vagal nerve stimulation 249–251, 253–254 post-traumatic seizures epidemiology 183–185 phenytoin prophylaxis 187–188 surgery for temporal lobe epilepsy 245–247 Epstein, C 251–254 European Carotid Surgery Trial (ECST) 53 European Organisation for Research and Treatment of Cancer (EORTC) temozolomide trial 62, 91–93 trials of radiotherapy for gliomas 62, 109–113 evidence, classification of xxi–xxii extent of resection high-grade gliomas 85–89 low-grade gliomas 105–108 extra-axial haematomas, timing of surgery 133–137 extradural haematomas, timing of surgery 133–137 Eyre, HJ 113

F Fabrini, MG 109–113 Facco, E 161–167 failed back surgery syndrome (FBSS), spinal cord stimulation 283–286 Fairbank, J 215, 229–230

207

Falk, D 271–274 Farahvar, A 151 Farkilla, M 85–89 Faulkner, JE 147–151 Fazl, M 197–199 Fehlings, MG 197–199, 206 Ferguson, GG 51–54 Fergusson, D 333–336 Fernandes, HN 39–41 fetal neural transplants 311–315 Fiehler, J 55–58 Findlay, GF 209, 223 Fink, GR 271–274 Fink, ME 29 Fischer, B 197–199 Fischer, RS 253 Fisher, B surgery for solitary brain metastases 69–74 temozolomide trial 91–93 Fisher, CM 9–11 Fisher, R 251–254 Fitzgerald, M 189–190 Flamm, ES 197–199 Flanders, AE 81–84 Fogarty, G 79 Follett, K, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal DBS 267–269 Foon, KA 77–79 Forbes, GS 19–21 Fortini, G 283–286 Foster, G 69–74 Fountain, N 251–254 Fourney, DR 85–89 Fourwinds, S 169–173 Fouyas, IP 234 Fox, AJ 51–54 Foy, PM 23–25 Fraedrich, G 55–58 Fraix, V 258–259 Frank, JI 45–50 Frankowski, RF 157–159 Franzini, A 293–295 Freed, CR 314 Freeman, DF 197–199 Freeman, TB 311–315 Freeman, W. 240–241 Freemantle, N 181 French, J 251–254 Friedman, WA 37 Frost, H 229–230 Fuller, GAG 133–137 functional neurosurgery 239–241 see also deep brain stimulation; occipital nerve stimulation; vagal nerve stimulation furosemide 321–323

G Galen of Pergamon 339 Galicich, J 66–67 Ganesan, D 139–142 Garcia, P 251–254 Gardner, WJ 302 Garrett, ES 221–224 Gaspar, L radiotherapy for low-grade gliomas 109–113 stereotactic radiosurgery for brain metastases 81–84 surgery for solitary brain metastases 69–74

208

Gelb, DE 194 Gelety, JE 139–142 generalized dystonia, deep brain stimulation 275–278 Gennarelli, TA 37 George, B 45–50 German Parkinson Study Group 259–261 Gharabaghi, A 271–274 Gholkar, A 41–43 Gibson, JNA 215 Gilbert, RW 203 Gillingham, FJ 133–137 Gilmartin, RC 249–251 Girvin, JP 245–247 Glasgow Coma Scale (GCS) 127–128, 129–130 Glasgow Outcome Scale (GOS) 128–129, 130 Gleave, JR 223 Glen, J 99–103 Gliadel Study Group (GSG) trial 95–98 glioblastoma multiforme carmustine wafer therapy 95–98 extent of resection 85–89 temozolomide chemotherapy 91–93 gliomas, low grade see low-grade gliomas globus pallidus interna (GPi), deep brain stimulation for dystonia cervical 278–281 generalized and segmental 275–278 for Parkinson’s disease 267–269 glycerol rhizolysis, for trigeminal neuralgia 306 Goadsby, PJ, occipital nerve stimulation for cluster headache 289–291 for migraine 297–300 Gokaslan, ZL 85–89, 208 Goldberg, H 225–227 Goodman, JH 197–199 Goodman, R 251–254 Gorlia, T 91–93 Gottesman, R 333–336 Graves, N 251–254 Grebin, J 251–254 Green, BA 197–199 Green, S 99–103 Greenberg, BD 309 Greenberg, HS 203 Greenberg, RP 133–137 Gregson, BA 39–43 Gross, R 251–254 Grossman, RG 197–199 Gruber, D, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Gruemer, H 175–177 Gudeman, SK 147–151 Guichard, JP 45–50 Guillon, B 45–50 Guiot, G 340–341 Gupta, R 49

H Haaxma-Reiche, H 69–74 Hachinski, VC 51–54 Hacke, W decompressive surgery 45–50 SPACE study 55–58 haematomas acute extra-axial, timing of surgery 133–137 chronic subdural, surgical management 139–142

209

intracerebral penumbra 43 STICH Trial 39–41 STICH II Trial 41–43 haemodilution, HHH therapy 27–29 Haffenden, A 278–281 Hagell, P 311–315 Haines, S 325–327 Hälbig, TD 271–274 Haley, EC 5–8 Hallett, P 311–315 Halstead, A 340 Hamani, C 307–310 Hamel,,W 259–261 Hamers, HP 109–113 Hamilton, MG 35–37 Hamlyn, PJ 305–306 Han, PP 37 Handforth, A deep brain stimulation for epilepsy 251–254 vagal nerve stimulation for epilepsy 249–251 Hannogan, LT 359–360 Hanscom, B surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Hansen, HH 201–203 Harbison, JW 147–151 Hardy, J 341, 350 Harris, C, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Harris, M 133–137 Harshman, K 169–173 Hartmann, A 271–274 Hasseleid, BF 120 Hassenbusch, SJ 85–89 Hatano, K 81–84 Hatlevoll, R 109–113 Hauser, RA 311–315 Hauser, WA 183–185 Hayakawa, K 81–84 Haynes, RB 51–54 headache cluster headache deep brain stimulation 293–295 occipital nerve stimulation 289–291 migraine, occipital nerve stimulation 297–300 HeADDFIRST (hemicraniectomy and durotomy on deterioration from infarction-related swelling trial) 45–50 head injuries 123–125 barbiturate therapy 157–159 decompressive craniectomy 143–146 paediatric patients 329–331 GCS and GOS 127–130 hyperosmolar therapy 161–167 hyperventilation therapy 175–177 intracranial pressure monitoring 147–151 magnesium sulphate therapy 179–181 post-traumatic seizures epidemiology 183–185 phenytoin prophylaxis 187–188 pre-hospital intubation 189–190 steroid therapy 153–155 therapeutic hypothermia 169–173 Hypothermia Paediatric Head Injury Trial 333–336 see also acute subdural haematomas; chronic subdural haematomas; extradural haematomas Heemskerk, J 261–263 Hegi, R 93

210

Heit, G 286 Hellenbrand, KG 197–199 Hellman, R 109–113 Hellström, P 221–224 Hellwig, D 271–274 Helwig-Larson, S 201–203 hemicraniectomy, in malignant MCA infarction 45–50 Henderson, J 251–254 Hendrix, T 147–151 Hennerici, M 55–58 Henning, R 329–331 Henry, TR deep brain stimulation for epilepsy 251–254 vagal nerve stimulation for epilepsy 249–251 Herbison, GJ 205–206 Herkwoitz, H 225–227 Hermans, J 69–74 Heros, RC 37 Herr, DL 197–199 Herzog, J 259–261 Hesekamp, H 271–274 Hess, K 85–89 ‘HHH’ therapy 27–29 Hiesiger, E 99–103 high-grade gliomas 61–62 brachytherapy 99–103 carmustine wafer therapy 95–98 extent of resection 85–89 temozolomide trial 91–93 Hilibrand, AS surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Hilker, R 259–261 Hilt, DC 95–98 Hiltzik, DH 363–364 Hinkka, S 85–89 Hirota, S 81–84 Hirsch, O 340 Hitchon, PW 197–199 Hoang-Xuan, K 109–113 Hochberg, FH 99–103 Hoekstra, FH 69–74 Hofmeijer, J 45–50 Hogarth, P, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Holford, TR 197–199 Holland, E 85–89 Holloway, K, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Holman, R 13–17 Holness, R 311–315 Holton, JL 311–315 Hope, DH 39–41 Hope, P 321–323 Horiot, JC 109–113 Horn, S, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Horsely, V 340 Houeto, JL 271–274 Hovarth, E 353–355 Hovestadt, A 65–67 Howard, G 161–167 Huang, GD, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Hugenholtz, H 69–74 Humphrey, PRD 23–25

211

Hunt, WE 28, 197–199 Hur, K, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Hurst, JM 161–167 Huston, J 3rd 19–21 Hutchinson, PJ chronic subdural haematomas 139–142 intracranial pressure monitoring 150–151 Hutchison, JS 333–336 hydrocephalus post-haemorrhagic ventricular dilatation, diuretic therapy 321–323 Shunt Design Trial 325–327 hyperosmolar therapy, for raised intracranial pressure in head injury 161–167 hypertension, HHH therapy 27–29 hyperventilation therapy, head injuries 175–177 hypervolaemia, HHH therapy 27–29 hypothalamic stimulation, for cluster headache 293–295 hypothermia, therapeutic 169–173 Hypothermia Paediatric Head Injury Trial 333–336

I Ibrahim, A 209 ICSS trial 58 Illingworth, R 23–25 Ilstrup, DM 345–347 Inomata, T 81–84 intercarotid lines, in grading of cavernous sinus invasion 349–350 International Cooperative Study on the Timing of Aneurysm Surgery 3, 5–8 International PHVD drug trial group 321–323 International Study of Unruptured Intracranial Aneurysms (ISUIA) 3, 19–21 International Subarachnoid Aneurysm Trial (ISAT) 3, 13–17 intracerebral haematomas penumbra 43 STICH Trial 39–41 STICH II Trial 41–43 intracranial aneurysms unruptured, natural history (ISUIA) 19–21 see also subarachnoid haemorrhage (SAH) intracranial hypertension hyperventilation therapy 176 traumatic DECRA trial 143–146 hyperosmolar therapy 161–167 pentobarbital therapy 157–159 intracranial pressure monitoring, after head injury 147–151 intubation, pre-hospital, in head injury 189–190 Isacson, O 311–315 Ivnik, R 109–113

J Jääkeläinen, J 95–98 Jacques, L 283–286 Jaeger, L 19–21 Jager, JJ 109–113 Jallah, I 139–142 Jane, JA 5–8 Janetta, PJ 301–303 Janis, LS 169–173 Jansen, O 55–58 Janzer, RC 91–93 Japanese Radiation Oncology Study Group (JROSG), stereotactic radiosurgery for brain metastases 81–84 Jasselainen, J 85–89 Jennett, B 127–130, 150 Jerwood, D 221–224 Jho, HD ablative techniques for trigeminal neuralgia 305–306

212

microvascular decompression for trigeminal neuralgia 301–303 Joffe, AR 333–336 Johnson, A 321–323 Johnson, RD 161–167 Johnston, IHS 150 Jones, C 109–113 Jones, JC 249–251 Josephson, CD 247 Judson, R 189–190 Juettler, E 45–50

K Kacker, A 363–364 Kadañka, Z 231–232 Kaplitt, M 251–254 Kappelle, LJ 45–50 Kaptain, GJ 361–362 Karim, AB 109–113 Karimi, A 39–41 Karmi, MZ 133–137 Kassell, NF 3 HHH therapy 27–29 ISUIA 19–21 timing of aneurysm surgery 5–8 Katoh, N 81–84 Kearney, T 347 Keihm, J 187–188 Keles, GE 85–89 Kelly, DL Jr. 161–167 Kemler, MA 286 Kenjyo, M 81–84 Kennedy, CR 321–323 Kennedy, JG 221–224 Kennedy, SH 307–310 Kerr, R 13–17 Kestle, JRW Shunt Design Trial 325–327 surgery for solitary brain metastases 69–74 Ki-67 labelling index, pituitary adenomas 353–355 Kirkpalani, HM 333–336 Kirkpatrick, PJ chronic subdural haematomas 139–142 vasospasm prophylaxis, statins 31–34 Kirollos, RW 139–142 Kiss, ZH 278–281 Kistler, JP 9–11 Kistner, A 271–274 Kitz, K 349–351 Klebe, S 259–261 Kloss, M, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Klug, G 329–331 Knosp, E 349–351 Knudsen, K 271–274 Kobashi, G 81–84 Kocher, T 340 Koes, BW 211–216 Kofman, S 66 Kongable-Beckman GL ISUIA 19–21 timing of aneurysm surgery 5–8 Kontos, HA 175–177 Kontturi, M 221–224 Kordower, JH 311–315 Kortelainen, P 221–224 Kosnik, EJ 28

213

Kossmann, T 143–146 Kostuik, JP 221–224 Koudsie, A 258–259 Kovacs, K 353–355 Koy, J 259–261 Krack, P, deep brain stimulation for Parkinson’s disease 258–261, 263–264 early subthalamic stimulation 271–274 Krause, F 340 Krause, M, deep brain stimulation for Parkinson’s disease 259–261 for dystonia 275–278 Krbec, M 231–232 Krüger, R 271–274 Kryscio, RJ adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 surgery for solitary brain metastases 69–74 Kumar, K 283–286 Kunieda, E 81–84 Kupsch, A, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Kurtz, A 45–50 Labar, DR deep brain stimulation for epilepsy 251–254 vagal nerve stimulation for epilepsy 249–251

L Labiner, DM 249–251 Laborit, H 241 Lacombe, D 91–93 Lacroix, JL 333–336 Lacroix, M 85–89 Lai, EC, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Laidlaw, JD 8, 161–167 Lambooij, N 69–74 Lamborn, KR extent of resection of low-grade gliomas 105–108 surgical treatment of prolactinomas 359–360 Lang, DA 23–25 Lang, FF 85–89 Laperriere, NJ 99–103 Larkins, MV ablative techniques for trigeminal neuralgia 305–306 microvascular decompression for trigeminal neuralgia 301–303 Larsson, B 241 Lashley, T 311–315 Laws, ER Jr. 107 biological correlates on invasive pituitary adenomas 353–355 dysembryoplastic neuroepithelial tumours (DNTs) 115–116 extended transsphenoidal approaches 361–362 pituitary apoplexy, timing of surgery 345–347 LeBas, JF 258–259 Leden–The Hague Spine Intervention Prognostic Study Group 211–216 Lees, AJ 311–315 Leone, M 293–295 Leo-summers, L 197–199 Leskell, L 241 Leung, PM 99–103 Levin, HS 169–173 Levine, M 69–74 Lewis, E 329–331 Lewis, RJ 165–166 Lewy bodies, development in grafted neurons 311, 313, 314–315

214

Li, JY 311–315 Libenson, MH 323 Lima, A. 240 Limousin, PD 258–259 Lindvall, O 311–315 Lipsman, N 241 Long, DM 67 Lopez, BC 305–306 Lorenz, D 259–261 Lorenzl, S 259–261 Louis, DN 116 low-grade gliomas (LGGs) 62 chemotherapy 113 extent of resection 105–108 radiotherapy 109–113 seizure outcome 108 Lozano, AM 307–310 Lucas, T 179–181 Ludwin, SK 91–93 Luerssen, TG 169–173 Luessenhop, AJ 37 lumbar degenerative spondylolisthesis, surgical intervention 227 lumbar disc herniation, surgical intervention 211–216 microscopic sequestrectomy 217–219 lumbar stenosis, surgical intervention 225–227 Lundberg, NL 150 Luo, P 267–269 Lurie, JD surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Lutz, HA 147–151 Lynch, JR 33

M Maarouf, M 271–274 Maas, AIR 181 Maat, B 109–113 Macdonald, JS 69–74 Machamer, J intracranial pressure monitoring 147–151 magnesium sulphate therapy in head injury 179–181 Macleod, MR 45–50 MacNeil, N 325–327 magnesium sulphate, in head injury 155, 179–181 magnetic resonance imaging (MRI) classification of cavernous sinus invasion 349–351 intra-operative in pituitary microadenoma resection 342 in resection of low-grade gliomas 107–108 malignant glioma brachytherapy 99–103 carmustine wafer therapy 95–98 extent of resection 85–89 temozolomide trial 91–93 malignant MCA infarction (MMI), decompressive surgery 45–50 Malkin, MG 99–103 Malmström, PO 109–113 Maltête, D 271–274 mannitol, hyperosmolar therapy for raised ICP 161–167 Mañon-Espaillat, R 249–251 Mansfield, PN 179–181 Marcus, H 139–142 Mareš, M 231–232 Marie, P 339 Marion, DW 166, 169–173 Markesbery, WR

215

adjuvant radiotherapy for solitary brain metastases 77–79 surgery for solitary brain metastases 69–74 Marks, W 251–254 Marks, WJ Jr, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Marmarou, A 175–177 Maroon, J 197–199 Marosi, C 91–93 Marshall, LF barbiturates in head injury 157–159 NASCIS 197–199 Martin, C 161–167 Martin, NA 35–37 Maruyama, Y 69–74 Mascarenhas, F 109–113 Masci, K 189–190 Mason, WP 91–93 Mateo, J 45–50 Matsuyama, J 354 Matula, C 349–351 Matula, K 349–351 Matz, PG 225–227 May, A 293 Mayberg, HS 307–310 Mayer, SA 49 Mazumdar, M 363–364 McCarville, S 297–300 McCauley, S 169–173 McCutcheon, IE 85–89 McDermott, FT 161–167 McFarlane, R 223 McGirt, MJ 105–108 McGrath, A 221–224 McIntyre, LA 335 McKenzie, S 99–103 McKhann, G 251–254 McKissock, W. 3 McLaughlin, N 358 McLelland, S 67 McManus, F 221–224 McNeely, HE 307–310 McWhorter, JM 161–167 Meagher, JN 197–199 Mealey, J 99–103 Meglio, M 283–286 Meguro, K 161–167 Mehdorn, HM, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Mehta, MP 81–84 Meier, N 271–274 Meinel, T 85–89 Meissner, I 19–21 Meldrum, HE 51–54 Mendelow, AD extradural haematomas, timing of surgery 133–137 intracerebral haematomas, STICH trial 39–43 Mendez, I 311–315 meningiomas, resection grading 119–120 Menten, J 109–113 Mertens, P 271–274 metastases, intracranial see brain metastases metastatic spinal cord compression (MESCC) decompressive surgery 207–209 dexamethasone 201–203 O-methylguanine-DNA methyltransferase (MGMT) promoter methylation, and temozolomide therapy 92, 93 methyl prednisolone

216

in acute spinal cord injury 197–199 in head injury (CRASH trial) 153–155 in metastatic spinal cord compression 202 Metsaars, JA 69–74 Meyer, FB 345–347 Meyer, PG 333–336 MIB-1 antibody, in evaluation of pituitary adenomas 353–355 Michael, C 85–89 microscopic sequestrectomy, for lumbar disc herniation 217–219 microvascular decompression (MVD), for trigeminal neuralgia 301–303 migraine, occipital nerve stimulation 297–300 Miklos, MV 302 Milbouw, G 283–286 Miller, D 85–89 Miller, ER 169–173 Miller, JD extra-axial haematomas, timing of surgery 133–137 intracranial pressure monitoring 147–151 Milner, R 325–327 Milosevic, M 99–103 minicraniotomy, evacuation of chronic subdural haematoma 141 Minoja, G 161–167 Mintz, AH 69–74 Mirimanoff, RO radiotherapy for low-grade gliomas 109–113 temozolomide trial 91–93 Mirski, MA 253 Mirza, W 249–251 Mitchell, PM 41–43 Mixter, JM 223 Mizuno, J 221–224 Moher, D 333–336 Mohiuddin, M adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 Mohr, G 95–98 Molet, J 283–286 Molyneux, A 13–17 Moniz, E. 240 Morawetz, R 95–98 Moringlane, JR 259–261 Moritz, U 233–234 Morris, GL 3rd 249–251 Morris, KP 333–336 motor cortex stimulation (MCS) 240 Mouridesen, H 201–203 Moy, CS, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Muacevic, A 84 Muizelaar, JP hyperventilation in head injury 175–177 hypothermia in head injury 169–173 Mukhida, K 311–315 Müller, JU 275–278 Muller, P 95–98 Murphy, JV 249–251 Murray, D 353–355 Murray, GD 181 nimodipine prophylaxis of cerebral vasospasm 23–25 surgery for spontaneous intracerebral haematomas 39–43 Murray, LJ decompressive craniectomy for head injury 143–146 hyperosmolar therapy for raised ICP 161–167 pre-hospital intubation in head injury 189–190 Myles, PS hyperosmolar therapy for raised ICP 161–167

217

pre-hospital intubation in head injury 189–190

N Nakaagawa, H 221–224 Nakagawa, K 81–84 Naritoku, DK 249–251 National Acute Brain Injury Study: Hypothermia (NABIS: H I, NABIS: H II) 169–173 National Acute Spinal Cord Injury Study (NASCIS) 197–199 Naumann, M 275–278 Navarro, SM 271–274 Nazzaro, J 251–254 Neal, JH brachytherapy for malignant gliomas 99–103 deep brain stimulation for epilepsy 251–254 Neligan, A 183–185 Nelson, D 109–113 Nelson, R 23–25 neural cell transplants, for Parkinson’s disease 311–315 neuromodulation for epilepsy deep brain stimulation 251–253 vagal nerve stimulation 249–251, 253–254 neuropathic pain, failed back surgery syndrome, spinal cord stimulation 283–286 neuroprotection in head injury magnesium sulphate 179–181 therapeutic hypothermia 169–173 Newell, DW 179–181 New York Islands AVM Study (NYIAVMS) 37 Ney, GC 249–251 Nguyen, V 189–190 Nichols, D natural history of unruptured aneurysms 19–21 radiotherapy for low-grade gliomas 109–113 Nikkhah, G 275–278 nimodipine, vasospasm prophylaxis 23–25 Nimsky, C 342 Nockels, RP 197–199 ‘Non-believers trial’, radiotherapy for low-grade gliomas 109–113 Noordijk, EM 69–74 Noordman, E 109–113 North, RB 283–286 North American Symptomatic Carotid Endarterectomy Trial (NASCET) 51–54 Northup, BE 205–206 Novotný, O 231–232

O obsessive–compulsive disorder (OCD), deep brain stimulation 309 O’Callaghan J 283–286 occipital nerve stimulation (ONS) for cluster headache 289–291 for migraine 297–300 Oertel, W, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 O’Fallon, J 109–113 O’Fallon, WM 19–21 Okonkwo, DO 169–173 Okuchi, K 161–167 Olanow, CW 311–315 O’Laoire, SA 221–224 Olivaras, R 95–98 Olivi, AO 105–108 Olsen, J 183–185 Olson, J 99–103 Ondra, S 37 O’Neill, B 109–113 Oommen, K 251–254 Orabi, M 45–50

218

Orbis-Sigma valve, Shunt Design Trial 325–327 Ory-Magne, F 271–274 Osorio, I deep brain stimulation for epilepsy 251–254 vagal nerve stimulation for epilepsy 249–251 Oya, N 81–84

P Padberg, GW 69–74 paediatric neurosurgery 319 decompressive craniectomy in head injury 329–331 Hypothermia Paediatric Head Injury Trial 333–336 post-haemorrhagic ventricular dilatation, diuretic therapy 321–323 Shunt Design Trial 325–327 Pahwa, R, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 pain management, deep brain stimulation 239, 240 pallidal deep brain stimulation for dystonia cervical 278–281 generalized and segmental 275–278 for Parkinson’s disease 267–269 Papagikos, MA 112 Papavassiliou, S 251–254 Parker, RA 139–142 Parkinson’s disease deep brain stimulation 257–266 early subthalamic stimulation 271–274 subthalamic versus pallidal 267–269 neural transplantation 311–315 Parvinen, LM 109–113 Pascale, V 197–199 Paschen, S 271–274 Pasut, LM 136 Patchell, RA 84 adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 surgery for solitary brain metastases 69–74 Patrick, I 189–190 Paul, KS 133–137 Payen, D 45–50 Payne, R 207–209 Peacock, J 19–21 Pedersen, CB 183–185 Pedersen, MG 183–185 Peerless, SJ 27–29 pentobarbital, in head injury 157–159, 164, 165 penumbra, intracerebral haematomas 43 Pernicone, P 353–355 Perot, PL Jr 197–199 Perrin, RG 206 Perry, J 98 Persson, LCG 233–234 Petroni, G 147–151 Peul, WC 211–216 phenytoin, prevention of post-traumatic seizures 187–188 Piantadosi, S 95–98 Piatt, J 325–327 Pichlmeier, U 85–89 Pickard, JD chronic subdural haematomas, surgical management 139–142 vasospasm prophylaxis nimodipine 23–25 statins 31–34 Piepgras, DG 19–21 Piepmeier, J 197–199

219

Piérart, M 109–113 Pierce, D 357–358 Pinsker, MO, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Pintilie, M 99–103 pituitary historical background 339 space-occupying lesions 339 pituitary adenomas biological correlates of invasion 353–355 classification of cavernous sinus invasion 349–351 intra-operative MRI 342 radiosurgery 342 pituitary apoplexy, timing of surgery 345–347 pituitary centres of excellence 358 pituitary surgery 341–342 choice of surgical approach 341–342 complications 357–358 development 340 endoscopic transsphenoidal surgery 363–364 extended transsphenoidal approaches 361–362 outcome evaluation 339–340 for prolactinoma 359–360 Poewe, W, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 Polderman, KH 172 Pollak, P 258–259, 263 Pollard, J 251–254 Pollock, BE 241 Polymer Brain Tumour Treatment Group (PBTTG) trial 95–98 Ponsford, J DECRA trial 143–146 hyperosmolar therapy for raised ICP 161–167 Posner, JB 73 Post, B 271–274 post-haemorrhagic ventricular dilatation (PVHD), diuretic therapy 321–323 post-traumatic seizures epidemiology 183–185 phenytoin prophylaxis 187–188 Prados, MD 105–108 pravastatin, vasospasm prophylaxis 31–34 prefrontal leucotomy 240 pre-hospital intubation, in head injury 189–190 premature infants, post-haemorrhagic ventricular dilatation, diuretic therapy 321–323 Price, SH 139–142 Pridgeon, J 147–151 PROCESS study 283–286 progesterone, in head injury (ProTECT study) 155 prolactinomas, surgical treatment 359–360 ‘psychosurgery’ 240–241 Puccio, A 169–173

Q Quinn, NP 311–315 Quinones-Hinojosa, A 105–108

R Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis (RPA) classification 74 stereotactic radiosurgery for brain metastases 81–84 radiofrequency ablation, for trigeminal neuralgia 306 radiosurgery for pituitary adenoma 342

220

stereotactic 241 for brain metastases 81–84 for trigeminal neuralgia 306 radiotherapy brachytherapy for malignant gliomas 99–103 for brain metastases 61 adjuvant therapy 77–79 RPA classification 74 solitary tumours 72, 73 for low-grade gliomas 109–113 Ragin, A 357–358 Ram, Z 95–98 Randomised Evaluation of Surgery of Intracranial Pressure (ReSCUEicp) 146 Rankin, RN 51–54 Ransohoff, J 197–199 Raoul, S 271–274 rapid sequence intubation (RSI), in head injury 189–190 Rathbone, MP 69–74 Rau, J 271–274 Rawe, SE 197–199 re-bleeding, endovascular coiling versus aneurysm clipping 16 recursive partitioning analysis (RPA) classification of brain metastases 74 Reda, D, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Regine, WF adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 Régis, JM 271–274 Rehncrona, S 311–315 Reichmann, H 259–261 Reid, SR 161–167 Reilly, P 143–146 Reulen, HJ 85–89 Reuss, A 259–261 Revesz, T 311–315 Richards, HK chronic subdural haematomas, surgical management 139–142 vasospasm prevention 31–34 Richards, P 23–25 Richardson, A. 3 Richardson, J 283–286 Riddle, V 95–98 Rifkinson, N 197–199 Ringleb, P 55–58 Ristanovic, R 249–251 Roberts, I 159, 166, 177, 188 Robertson, H 311–315 Robertson, JT 99–103 Rocca, WA 183–185 Roeste, GK 275–278 Romanelli, P 286 Rondina, C 147–151 Roos, DE 73 Rosa, S 79 Rosenfeld, JV decompressive craniectomy in paediatric head injury 329–331 DECRA trial 143–146 pre-hospital intubation in head injury 189–190 Rosenfeld, W 251–254 Rosner, MJ 147–151 Rothlind, J, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Rothwell, PM 45–50 Rotman, M 81–84 Rouanet, F 45–50 Rowan, EN 41–43

221

Rowed, DW 161–167 Rush, AJ 309 Rushing, EJ 116 Rutten, EH 109–113 Ryu, J 81–84

S Saatman, KE 123 Sackett, DI 51–54 Sagher, O, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 sagittal balance 193–194 Salanova, V 251–254 saline, hypertonic, for raised intracranial pressure 161–167 Salinsky, MC 249–251 Samii, A, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Samson, DS 37 Sanai, N 88 Sandercock, P 13–17 Sandok, E 251–254 Santarius, T 139–142 Saper, JR 297–300 SAPPHIRE trial 58 Saris, S 207–209 Saunders, M 202 Sawaya, R 85–89 Sayre, MR 161–167 Schachter, SC 249–251 Schade-Brittinger, C, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Schaefer, HR 221–224 Schäfer, H 259–261 Scheithauer, B dysembryoplastic neuroepithelial tumours (DNTs) 115–116 MIB-1 antibody 353–355 pituitary apoplexy, timing of surgery 345–347 radiotherapy for low-grade gliomas 109–113 Schell, MC 81–84 Schierhout, G 177, 188 Schiff, S 325–327 Schmidy, JH 169–173 Schmiedeck, P 45–50 Schneider, GH, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Schnitzler, A, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Schoffler, H 340 Schold, SC 95–98 Schraub, S 109–113 Schuepbach, WMM 271–274 Schüpbach, WMM 265 Schuster, J 179–181 Schwab, S 45–50 Schwalb, JM 307–310 Schwartz, ML 157, 159 hyperosmolar therapy for raised ICP 161–167 hypothermia in head injury 169–173 Schwartz, TH 363–364 sciatica see lumbar disc herniation Sciubba, DM 208 Scott, CB 81–84

222

Scott, JN 169–173 Scottish Intracranial Vascular Malformations Study (SIVMS) 37 Seelig, JM 133–137 segmental dystonia, deep brain stimulation 275–278 Seigneuret, E 271–274 seizure outcome, low-grade gliomas 108 seizures, post-traumatic epidemiology 183–185 phenytoin prophylaxis 187–188 Selhorst, JB 147–151 Selker, R brachytherapy for malignant gliomas 99–103 carmustine wafer therapy 95–98 Seminiwicz, D 307–310 Seppelt, I 143–146 sequestrectomy, for lumbar disc herniation 217–219 Shann, F 329–331 Shapiro, S 221–224 Shapiro, WR 99–103 Shaw, E 109–113 Shaw, MDM 23–25 STICH Trial 39–41 Sheehan, HP 361–362 Sheehan, J 205–206 Sheehan, TP 205–206 Shepard, MJ 197–199 Shephard, RH 223 Shetter, A 251–254 Shi, W 85–89 Shioura, H 81–84 Shirato, H 81–84 Shorvon, S 183–185 Shrimpton, J 13–17 Shunt Design Trial 325–327 Siddiqi, SN 102 Sidenius, P 183–185 Sieber, AN 221–224 Silberstein, SD 297–300 Simons, L 11 Simpson, D 119–120 Simpson, R 267–269 simvastatin, vasospasm prophylaxis 31, 33, 34 Sinar, J 23–25 Sindou, M 302–303 Singh, RN 333–336 Sisti, M 95–98 Siu, KH 8 Sixel-Doering, F 271–274 Skene, A 23–25 Skingley, P 69–74 Skinner, JS 211–216 Skippen, PW 333–336 Skogseid, IM 275–278 Smielewski, P 139–142 Smith, HP 161–167 Smith, JS 105–108 Smith, K hypothermia in head injury 169–173 pre-hospital intubation in head injury 189–190 Smrcka, V 231–232 Snoek, J 127–130 Soffe, KE 221–224 Solenski, NJ 29 solitary brain metastases adjuvant radiotherapy 77–79 surgical resection 69–74

223

Solomon, RA 29 Sonntag, VKH 197–199 Sorensen, PS 201–203 Souhami, L 81–84 Soulet, D 311–315 Speelman, H 271–274 Spence, JD 51–54 Sperduto, PW 81–84 Sperling, M 251–254 Spetzler, RF 35–37 Spetzler–Martin grading system 4, 35 spinal cord compression, metastatic decompressive surgery 207–209 dexamethasone 201–203 spinal cord injury, acute methyl prednisolone (NASCIS) 197–199 timing of surgery 205–206 spinal cord stimulation (SCS), for failed back surgery syndrome 283–286 spinal stabilization, for chronic back pain 229–230 spinal surgery 193–194 spine patient outcomes research (SPORT) trial 211–216 surgery for lumbar degenerative spondylolisthesis 227 surgery for lumbar stenosis 225–227 Spine Stabilization Trial Group 229–230 spondylolisthesis, surgical intervention 227 Stapf, C 37 Starr, PA 267–269 STASH trial 31, 34 statins, vasospasm prophylaxis 31–34 Stefabeanu, L 353–355 Stefan, H 249–251 Steinbok, P 325–327 Steiner, E 349–351 Steinmeier, R 37 Stejskal, L 231–232 Stenning, S 109–113 Stent-Protected Angioplasty versus Carotid Endarterectomy (SPACE) study 55–58 Stephens, MM 221–224 stereotactic radiosurgery 241 for brain metastases 81–84 for trigeminal neuralgia 306 Stern, J 251–254 Stern, M, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Stern, SA 161–167 steroids in acute spinal cord injury 197–199 in head injury 153–155 in metastatic spinal cord compression 201–203 in pituitary apoplexy 346–347 for tumour-associated cerebral oedema 65–67 Steude, U 259–261 STICH (trauma) trial 137 Stilten, RM 197–199 Stimulation of the Anterior Nucleus of the Thalamus for Epilepsy (SANTE) study 251–254 Stingele, R 55–58 Stochetti, N 150 Stoner, G, deep brain stimulation for Parkinson’s disease 261–263 subthalamic versus pallidal stimulation 267–269 Storer, DL 161–167 Stratton, I 13–17 Stuart, GG 150 Stummer, W 85–89 Sturm, V 259–261 Stuup, R 91–93 subarachnoid haemorrhage (SAH) 3

224

endovascular coiling versus aneurysm clipping, ISAT 13–17 prognostic factors 6–7 timing of aneurysm surgery 5–8 vasospasm HHH therapy 27–29 radiological predictors 9–11 vasospasm prophylaxis nimodipine 23–25 statins 31–34 subdural haematomas see acute subdural haematomas; chronic subdural haematomas Subin, DK 234 subthalamic nucleus (STN), DBS for Parkinson’s disease 257–266 adverse effects 263–264, 265 comparison with pallidal DBS 267–269 early stimulation 271–274 in older patients 265 Suchowersky, O 278–281 Sughrue, ME 120 Sullivan, HG 147–151 Sun, M 297–300 suprasellar lesions, extended transsphenoidal approaches 361–362 Surelová, D 231–232 surgical technique xxii Surgical Trial in Intracerebral Haematoma (STICH) 39–41 STICH II Trial 41–43 Svien, HJ 139–142

T Tabaee, A 363–364 Tago, T 81–84 Tans, JT surgery for lumbar disc herniation 211–216 surgery for solitary brain metastases 69–74 Taphoorn, MJB 91–93 Tarver, WB 249–251 Tatli, M 306 Tator, C 161–167, 206 Taylor, A 329–331 Taylor, D 189–190 Taylor, DW 51–54 Taylor, P 169–173 Taylor, RS 283–286 Teasdale, GM Glasgow Coma Scale 127–130 STICH Trial 39–41 vasospasm, nimodipine prophylaxis 23–25 Tecoma, ES 249–251 Temkin, N intracranial pressure monitoring 147–151 magnesium sulphate therapy in head injury 179–181 phenytoin prophylaxis of post-traumatic seizures 187–188 temozolomide trial 62, 91–93 temporal lobe epilepsy, surgical intervention 245–247 thalamus, anterior nucleus stimulation (SANTE study) 251–254 Than, K 105–108 Thapar, K 353–355 Thielen, K 19–21 Thobois, S 271–274 Thomachot, L 161–167 Thomas, DG radiotherapy for low-grade gliomas 109–113 surgery for cauda equina syndrome 221–224 Thomé, C 217–219 Thomeer, RTWM 211–216 Thomson, S 283–286 Thorpe, KE 51–54

225

Tibballs, J 329–331 Tibbs, PA adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 surgery for solitary brain metastases 69–74 Tierney, T 311–315 Tihan, T 105–108 Timmermann, L, deep brain stimulation for Parkinson’s disease 259–261 early subthalamic stimulation 271–274 Timmons, SD 145 Todd, NV 221–224 Tonder, L 251–254 ‘topectomy’ 240 Torner, JC ISUIA 19–21 timing of aneurysm surgery 5–8 Tosteson, ANA surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Tosteson, TD surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Touzé, E 45–50 Toyoda, T 81–84 tranexamic acid, in head injury (CRASH2 trial) 155 transorbital frontal leucotomy 240–241 Transou, C 161–167 transsphenoidal surgery complications 357–358 endoscopic 363–364 extended approaches 361–362 historical background 340–341 see also pituitary surgery trauma see head injuries; spinal cord injury, acute Tremayne, AB 161–167 trigeminal neuralgia 240 ablative therapy 305–306 microvascular decompression 301–303 Trojanowski, IQ 311–315 tromomethamine (THAM), use in hyperventilation therapy 175–176 Tronnier, V, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 Trottenberg, T, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 Troupp, J 37 Tseng, MY 31–34 Tsui, J 278–281 Turner, JA 286 Tyrrell, JB 359–360

226

U unruptured aneurysms, natural history, ISUIA 19–21 Urbánek, I 231–232 Ushio, Y 202–203 ‘US trial’, radiotherapy for low-grade gliomas 109–113 Uthman, BM 249–251

V Vaccaro, AR 205–206 vagal nerve stimulation (VNS) for epilepsy 249–251, 253–254 for treatment-resistant depression 309 Vahedi, K 45–50 Valadka, AB 166, 169–173 van Alpen, AM 109–113 Van Blercom, N 258–259 van den Bent, MJ radiotherapy for low-grade gliomas 109–113 temozolomide trial 91–93 Vandenberg, S 105–108 van den Hout, WB 211–216 van der Worp, HB 45–50 van Gijn, J 45–50 Van Gilder, JC 99–103 van Glabbeke, M 109–113 van Houwelingen, HC 211–216 van Loveren, HR 161–167 van Putten, WL 65–67 van Reijn, M 109–113 Van Royen, BJ 194 van Vliet, JJ 65–67 vasospasm 3 HHH therapy 27–29 radiological predictors 9–11 vasospasm prophylaxis HHH therapy 29 nimodipine 23–25 statins 31–34 Vaughn, BV 249–251 Vecht, CJ dexamethasone for cerebral oedema 65–67 surgery for single brain metastases 69–74 Vedrenne, C. 115–116 vegetative state 129 Velasco, M 253 Velmahos, GC 190 Verbiest, HB 65–67 Vesalius, Anatomica Fabrica 339 Vesper, J, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 271–274 Vestergaard, M 183–185 Vialet, R 161–167 Vicaut, E 45–50 Vick, NA 95–98 Videtta, W 147–151 Vincent, DA 361–362 Vinuela, A 311–315 Vladyka, V 342 Voges, J, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261 Volkmann, J, deep brain stimulation for dystonia 275–278 for Parkinson’s disease 259–261

227

early subthalamic stimulation 271–274 Vollmer-Haase, J 275–278 Von Eiselberg, A 340 von Hochenegg, J 340 Voon, V 307–310 Voormenolen, JH 69–74 Vuorinen, V 85–89

W Waddell, G 215 Wagner, FC 169–173 Wagner, FW 197–199 Wakai, A 166, 167 Walker, M 95–98 Walker, MD 157–159 Walker, T 189–190 Wallace, D 329–331 Walsh, JW 69–74 Walsh, L. 3 Walsh, MG 221–224 Ward, JD barbiturates in head injury 157, 159 extra-axial haematomas, timing of surgery 133–137 hyperventilation in head injury 175–177 intracranial pressure monitoring 147–151 Ward, RE 333–336 Warlow, C 37 Warnke, PC 95–98 Watkins, L 289–291 Wattendorf, AR 69–74 Watts, J. 240–241 Weaver, FM, deep brain stimulation for Parkinson’s disease 261–263, 264–265 subthalamic versus pallidal stimulation 267–269 Weber, H 211–216 Weingart, JD 105–108 Weinstein, JN surgery for lumbar disc herniation 211–216 surgery for lumbar stenosis 225–227 Weiss, C 217–219 Weller, M 91–93 Werner-Wasik, M 81–84 Wernicke, JF 249–251 Westphal, M 95–98 Wheless, JW 249–251 Whisnant, JP 19–21 Whittle, IR 95–98 whole brain radiotherapy (WBRT), adjuvant therapy for solitary brain metastases 77–79 Widner, H 311–315 Wiebe, S 239, 245–247 Wiebers, DO 19–21 Wiestler, OD 85–89 Wilberger, JE 166 extra-axial haematomas, timing of surgery 133–137 NASCIS 197–199 Wilde, E 169–173 Wilder, BJ 249–251 Wilensky, AJ 187–188 Wilson, CB 350, 359–360 Wilson-McDonald, J 229–230 Winn, HR magnesium sulphate therapy in head injury 179–181 NASCIS 197–199 phenytoin prophylaxis of post-traumatic seizures 187–188 Witjas, T 271–274 Witt, T 251–254 Wojtecki, L, deep brain stimulation for Parkinson’s disease 259–261

228

early subthalamic stimulation 271–274 Wolf, E 259–261 Wolfe, R 143–146 Wolters, A 275–278 Wong, S 99–103 Woods, SP 265 Worth, R 251–254 Wright, DW 155 Wu, J 99–103

Y Yang, JY 358 Yeh, DD 190 Yelnik, A 45–50 Yonas, H 169–173 Youkilis, A 251–254 Young, B adjuvant radiotherapy for solitary brain metastases 77–79 decompressive surgery for spinal metastasis 207–209 surgery for solitary brain metastases 69–74 Young, HF 147–151, 175–177 Young, W 197–199 Yu, LM 229–230

Z Zakrzewska, JM 305–306 Zanella, F 85–89 Zeumer, H 55–58 Zumsteg, D 254 Zygun, D 166, 169–173

229

MáťĽc láťĽc Title Page Copyright Page Dedication Contents Abbreviations Contributors Introduction 1 Neurovascular neurosurgery 2 Neuro-oncology 3 Head injury 4 Spinal surgery 5 Functional and epilepsy neurosurgery 6 Paediatric neurosurgery 7 Pituitary surgery Index

4 5 6 10 11 16 19 21 51 83 119 142 181 189 201

230