Prelims
8/4/05
2:31 pm
Page i
BIPOLAR DISORDER: THE UPSWING IN RESEARCH AND TREATMENT
Prelims
8/4/05
2:31 pm
Page ii
Prelims
8/4/05
2:31 pm
Page iii
BIPOLAR DISORDER: THE UPSWING IN RESEARCH AND TREATMENT EDITED BY COLM MCDONALD Research Training Fellow Division of Psychological Medicine Institute of Psychiatry London, UK
KATJA SCHULZE Research Worker Division of Psychological Medicine Institute of Psychiatry London, UK
ROBIN M MURRAY Professor of Psychiatry Division of Psychological Medicine Institute of Psychiatry London, UK
MAURICIO TOHEN Professor of Psychiatry Department of Psychiatry Harvard Medical School Belmont, Massachusetts, USA
European Foundation for Psychiatry at The Maudsley LONDON AND NEW YORK
Prelims
8/4/05
2:31 pm
Page iv
© 2005 Taylor & Francis, an imprint of the Taylor & Francis Group First published in the United Kingdom in 2005 by Taylor & Francis, an imprint of the Taylor & Francis Group, 2 Park Square, Milton Park Abingdon, Oxon OX14 4RN, UK Tel.: +44 (0) 20 7017 6000 Fax.: +44 (0) 20 7017 6699 Website: www.tandf.co.uk All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Although every effort has been made to ensure that all owners of copyright material have been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our attention. British Library Cataloguing in Publication Data Data available on application Library of Congress Cataloging-in-Publication Data Data available on application ISBN 1-84184-501-9 (Hardback) Distributed in North and South America by Taylor & Francis 2000 NW Corporate Blvd Boca Raton, FL 33431, USA Within Continental USA Tel.: 800 272 7737; Fax.: 800 374 3401 Outside Continental USA Tel.: 561 994 0555; Fax.: 561 361 6018 E-mail:
[email protected] Distributed in the rest of the world by Thomson Publishing Services Cheriton House North Way Andover, Hampshire SP10 5BE, UK Tel.: +44 (0) 1264 332424 E-mail:
[email protected] Composition by Parthenon Publishing Printed and bound by T. G. Hostench S.A., Spain
Prelims
8/4/05
2:31 pm
Page v
Contents
Preface Foreword Contributors
ix xiii xv
Section 1. Do we know the clinical course and epidemiology? 1
The clinical epidemiology of bipolar disorder: a 35-year incidence study in south-east London Noel Kennedy and Robin M Murray
1
2
The functional outcome of bipolar disorder Mauricio Tohen and Julie M Niswander
9
Section 2. Is bipolar disorder a brain disease? 3
Brain abnormalities in bipolar disorder: do they exist and do they change? E Serap Monkul and Jair C Soares
21
4
Structural magnetic resonance imaging studies in bipolar disorder: a meta-analysis Colm McDonald, Jolanta Zanelli, Robin M Murray and Noel Kennedy
27
5
Are subcortical regions too expansive in bipolar disorder? An examination of the nature of prefrontal corticolimbic abnormalities in individuals with bipolar disorder Mary L Phillips
37
6
The Maudsley Bipolar Disorder Project: insights into pathophysiology Sophia Frangou
51
Prelims
8/4/05
vi
2:31 pm
Page vi
Bipolar disorder: the upswing in research & treatment
7
Is any of this real? The word from the grave Paul J Harrison
59
Section 3. Happy genes, blue genes, any genes? 8
How can bipolar disorder be genetically related to both schizophrenia and unipolar depression? Peter McGuffin
69
9
Recent advances in genetics of bipolar disorder Daniel J Müller and James L Kennedy
77
10
Is there a genetic basis to the brain abnormalities of bipolar disorder? Colm McDonald
93
11
Transgenic mouse models for affective disorders based on the neurotrophin hypothesis Peter Gass
103
Section 4. Cortisol: hero or villain? 12
Is the hypothalamic–pituitary–adrenal axis at last paying dividends? David A Cousins and Allan H Young
115
13
Stress on the brain: neuropathology and cortisol dysregulation in bipolar disorder David Cotter
123
14
Cortisol in Chicago (from crime of passion to celebrity headline) Carmine M Pariante
129
15
Biological factors sustaining hypothalamic–pituitary– adrenal axis overactivation in affective disorder: focus on vasopressin Timothy G Dinan, Sinead O’Brien and Lucinda Scott
135
Section 5. What is the role of psychology? 16
Cognitive dysfunction: cause or consequence of bipolar disorder? Samuel R Chamberlain and Barbara J Sahakian
145
Prelims
8/4/05
2:31 pm
Page vii
Contents
vii
17
The neural basis of cognitive function in bipolar disorder Vivienne Curtis
157
18
Psychological treatments: does the evidence stack up? Jan Scott
165
Section 6. Improving the patient’s lot 19
Lithium, the forgotten drug Mario Maj
175
20
Advantages and disadvantages of atypical antipsychotics or valproate in bipolar disorder John Cookson
181
21
Is electroconvulsive therapy still given in bipolar disorder and does repetitive transcranial magnetic stimulation offer more? Andrew Mogg, Savitha Eranti, Graham Pluck and Declan M McLoughlin
193
22
Improving outcome by selecting effective long-term treatment Paul Grof
201
23
Is what we offer to patients half acceptable? Rachel Perkins
211
Index
219
Prelims
8/4/05
2:31 pm
Page viii
Preface
7/4/05
3:33 pm
Page ix
Preface Bipolar disorder is a recurrent illness which, without appropriate treatment, can have a devastating impact on the lives of those affected and their families. The World Health Organization has estimated that bipolar disorder is the sixth leading cause of years lived with disability (schizophrenia came ninth) and yet historically it has been relatively under-researched. This situation has greatly changed in recent years as bipolar disorder has become a focus for research, with consequent advances in our understanding of the aetiology and management of the disorder. This ‘upswing’ in the research and treatment of the illness was discussed at a recent European Foundation for Psychiatry at the Maudsley (EFPM) meeting held at the Institute of Psychiatry. The theme of the conference, and of this resultant volume, is the new research and clinical developments in bipolar disorder research across multiple disciplines. The book comprises six sections. In the first, studies are presented that employ the powerful tools of epidemiology to identify how the incidence and age of onset of bipolar disorder are influenced by temporal variation and demographic variables. The long-term course of the disorder is also considered with a particular emphasis on the relative failure of treatments to affect important functional outcomes for patients, as distinct from the symptomatic improvements which clinicians tend to emphasize. The second section is devoted to investigations attempting to locate the brain regions that are structurally and functionally impaired in bipolar disorder at a macroscopic and microscopic level. Although Kraepelin considered the syndrome of ‘manic depressive insanity’ that he described as a brain disease, for much of the past century the neurobiological basis of the illness has been denied or ignored. However, the application of recent neuroimaging techniques has reinvigorated this field, and bipolar disorder has been associated with subtle deviations of brain structure, and with evidence of impaired functioning of critical brain regions implicated in the processing of cognitive and affective stimuli. In the final chapter of this section, these neuroimaging findings are integrated with the emerging neuropathological literature, most of which emanates from the invaluable resource of postmortem tissue provided by the Stanley Medical Research Institute, to
Preface
7/4/05
x
3:33 pm
Page x
Bipolar disorder: the upswing in research & treatment
develop an understanding of the structural and functional abnormalities at a cellular level. It has long been known that bipolar disorder is highly heritable but, as with many other psychiatric syndromes, progress in identifying the susceptibility genes has been slow. Section 3 includes chapters which review the genetics of the disorder. The topics covered include the genetic overlap between bipolar disorder and its related syndromes, schizophrenia and unipolar depression, how abnormalities of brain structure reflect the impact of susceptibility genes, the depressive-like behaviour of transgenic mice, and the associations of allelic variation in candidate genes within linked regions and the clinical syndrome. A central strand of research into affective disorders has been dysfunction of the hypothalamic–pituitary–adrenal axis. There is clear evidence for impaired regulation of cortisol secretion from the adrenal glands in both unipolar and bipolar disorder, and the precise mechanisms whereby this emerges and its interaction with other components of the axis are exciting areas of current research. Section 4 deals with excessive cortisol secretion in bipolar disorder, addressing its origins, how it is sustained, and its neurotoxic effects. Whether it causes or compensates for the clinical syndrome, as well as its potential for manipulation to therapeutic benefit, is also reviewed. Section 5 addresses the psychology of bipolar disorder, including the manner in which impaired cognitive processing of emotionally coloured stimuli can contribute to the symptoms of affective disorder. There is abundant evidence for cognitive dysfunction during episodes of bipolar illness, but even in remission neurocognitive abnormalities frequently persist. This reflects trait-like dysfunction in the neural networks subserving these functions and these are explored using functional imaging studies. The final chapter of this section considers the role of psychological treatments in the management of bipolar disorder, and in particular the evidence that cognitive behavioural psychotherapy can help to prevent patients relapsing into episodes of illness. The last section is devoted to management of the illness. Bipolar disorder is clinically heterogeneous and optimal treatment strategies must pay heed to the individual patient’s characteristics. Although the number and type of medications available is expanding, important roles remain for lithium and for monotherapy in maintenance treatment. Other issues discussed include the use of antipsychotics and valproate as monotherapy or in combination for the treatment of mania. Novel strategies such as repetitive transcranial magnetic stimulation for bipolar depression and the use of
Preface
7/4/05
3:33 pm
Page xi
Preface: bipolar disorder: the upswing in research and treatment
xi
long-term efficacy studies to tailor treatment on the basis of patients’ clinical characteristics are also reviewed. The final word from Rachel Perkins communicates a unique perspective since she suffers herself from bipolar disorder and is also employed as a senior manager of mental health services within the UK National Health Service. She discusses the practical problems of living with manic depression which are really of importance to patients, most of which are ignored by standard services, with a focus on the critical issue of protecting employment. The aim of the EFPM is to provide high-quality postgraduate education in psychiatry. It is an independent body aiming to provide up-to-date knowledge to Europe’s foremost clinical and academic professionals in mental health. Eli Lilly and Company provided a non-restrictive educational grant for the conference. This book will be of interest to clinicians and academics within psychiatry, psychology and neuroscience as well as other mental health professionals interested in bipolar disorder. The editors believe that the expert contributions capture the considerable progress that has been made in our understanding of this devastating condition in recent years. We are very grateful to all the contributors and hope that the reader will also be stimulated and informed by the wide range of research approaches that this book encompasses. Colm McDonald Katja Schulze Robin M Murray Mauricio Tohen
Preface
7/4/05
3:33 pm
Page xii
Foreword
7/4/05
3:34 pm
Page xiii
Foreword Research on the determinants, treatments and consequences of bipolar disorder has shown a very welcome increase over the last 15 years. In 1990 the seminal Handbook of Bipolar Disorders by Goodwin and Jamieson was published which documented all the studies that had been done on bipolar disorder at that time. Although this book became the bible for those researching bipolar disorder and was both of a high intellectual standing and comprehensive in its scope, re-reading it now reveals the uncertainty we had about the phenotype and the rather poor quality of a lot of studies that were done in bipolar disorder. Many studies were on heterogeneous samples and there was a lack of clarity on outcomes. Over the succeeding 15 years there have been considerable advances in firming up the classification and typology of the disorder. The increased number of published studies has led to much firmer conclusions about the nature of this disorder and its treatment. Another change that has occurred in these 15 years is a much greater understanding of the complexity of bipolar disorder. In the past, the Kraepelinian view that the patient was either manic or depressed or well, held sway, and the fact that lithium worked gave psychiatrists a false sense of security. The reality is much more complex with considerable subsyndromal difficulties in many patients and failure to recover fully in the euthymic phase evident for many. It is also apparent that while the pharmacological and psychological treatment of bipolar disorder have improved considerably, major problems with partial or complete treatment resistance and problems with side-effects and/or compliance remain. This book is an excellent summary of progress in all the important areas of bipolar disorder sometimes using state of the art technology and its title ‘Upswing’ mirrors this. The meeting was an exciting event with new science and good debate and this volume reflects that excellence. The volume also produces interesting pointers to where we may be in 15 years time. There is cause for guarded optimism. Professor Nicol Ferrier Department of Psychiatry, School of Neurology, Neurobiology and Psychiatry, Royal Victoria Infirmary, Newcastle upon Tyne, NE1 4LP, UK
Foreword
7/4/05
3:34 pm
Page xiv
Contributors
7/4/05
3:34 pm
Page xv
Contributors
Samuel R Chamberlain Department of Psychiatry, University of Cambridge School of Clinical Medicine, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2QQ, UK John Cookson BM DPhil FRCPsych FRCP, The Royal London Hospital, St Clement’s, 2A Bow Road, London E3 4LL, UK David Cotter MB BCh MRCPsych PhD, Department of Psychiatry, Royal College of Surgeons in Ireland, Education and Research Centre, Beaumont, Dublin 9, Ireland David A Cousins BMedSci MB BS MRCP MRCPsych, Stanley Research Centre, School of Neurology, Neurobiology and Psychiatry, University of Newcastle upon Tyne, Leazes Wing, Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP, UK Vivienne Curtis Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Timothy G Dinan MD PhD DSc, Department of Psychiatry, University College, Cork, Ireland Savitha Eranti MRCPsych, Section of Old Age Psychiatry, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Sophia Frangou Section of Neurobiology of Psychosis, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Peter Gass MD, Central Institute for Mental Health, Universität Heidelberg J 5, D-68159 Mannheim, Germany
Contributors
7/4/05
xvi
3:34 pm
Page xvi
Bipolar disorder: the upswing in research & treatment
Paul Grof MD PhD FRCPC, Bipolar Research Unit, University of Ottawa, Royal Ottawa Hospital, 1145 Carling Avenue, Ottawa, Ontario, Canada K17 7K4 Paul J Harrison Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK James L Kennedy MD MSc FRCP(C), Neurogenetics Section, Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, 250 College Street, Toronto, Ontario, Canada M5T 1R8 Noel Kennedy MB MD MSc(Med) MRCPsych, Section of General Psychiatry, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK and St. Patrick’s Hospital, Dublin, Ireland Colm McDonald MRCPsych PhD, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Peter McGuffin MB, PhD, FRCP, FRCPsych, FMedSci, Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Declan M McLoughlin PhD MRCPI MRCPsych, Section of Old Age Psychiatry, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Mario Maj MD PhD, Department of Psychiatry, University of Naples, Italy Andrew Mogg MRCPsych, Section of Old Age Psychiatry, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK E Serap Monkul Division of Mood and Anxiety Disorders, Department of Psychiatry, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA
Contributors
7/4/05
3:34 pm
Page xvii
Contributors
xvii
Daniel J Müller MD, Neurogenetics Section, Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, 250 College Street, Toronto, Ontario, Canada M5T 1R8 and Department of Psychiatry, Charité University Medicine Berlin, Campus Charité Mitte, Berlin, Germany Robin M Murray MB MD DSc FRCPsych, Section of General Psychiatry, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Julie M Niswander PhD, Lilly Research Laboratories, Indianapolis, IN, USA Sinead O’Brien MRCPsych, Department of Psychiatry, University College, Cork, Ireland Carmine M Pariante MD, MRCPsych, PhD, Stress, Psychiatry and Immunology Laboratoy (SPI-LAB), Institute of Psychiatry, King’s College London, 1 Windsor Walk, London SE5 8AF, UK Rachel Perkins BA MPhil PhD, South West London and St George’s Mental Health NHS Trust, Springfield University Hospital, Tooting, London SW17 7DJ, UK Mary L Phillips Section of Neuroscience and Emotion, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Graham Pluck PhD, Section of Old Age Psychiatry, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Barbara J Sahakian FMedSci, Department of Psychiatry, University of Cambridge School of Clinical Medicine, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2QQ, UK Katja Schulze Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK
Contributors
7/4/05
3:34 pm
Page xviii
Jan Scott Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK Lucinda Scott PhD MRCPsych, Department of Psychiatry, University College, Cork, Ireland Jair C Soares MD, Division of Mood and Anxiety Disorders, Department of Psychiatry, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA Mauricio Tohen MD DrPH, Department of Psychiatry, Harvard Medical School, McClean Hospital, 115 Mill Street, Belmont, MA 02478, USA Allan H Young MB ChB MPhil PhD FRCPsych, Stanley Research Centre, School of Neurology, Neurobiology and Psychiatry, University of Newcastle upon Tyne, Leazes Wing, Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP, UK Jolanta Zanelli MSc, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, de Crespigny Park, Denmark Hill, London SE5 8AF, UK
Ch 01
7/4/05
3:34 pm
Page 1
chapter 1
The clinical epidemiology of bipolar disorder: a 35-year incidence study in south-east London Noel Kennedy and Robin M Murray
Introduction Numerous studies have established the basic clinical epidemiology of schizophrenia.1 For example, our research group has carried out a number of epidemiological studies of schizophrenia in Camberwell, south-east London, and have found that the incidence of narrowly defined schizophrenia was higher in males than females and that the age at onset of schizophrenia was earlier in men than women, echoing findings from other centres.1,2 We have also shown that the incidence of schizophrenia has doubled in Camberwell since 1964, although this increase may be idiosyncratic to south London.3 It may result at least in part from the influx of migrants to this area, as the incidence of schizophrenia is approximately six times higher among the Black African and African–Caribbean population in south London compared with the White population.4 These elevated incidence rates have not been reported in Caribbean countries and may involve complex biological, psychological and social factors.5 In contrast to schizophrenia, very little is currently known about the basic clinical epidemiology of bipolar disorder. The few published incidence studies, based on small sample sizes, have reported a wide variation in incidence rates of first-episode mania from 1.7 to 4.5 per 100 000 population per year.6 Furthermore, studies to date may have underestimated true incidence.6 Data from the Epidemiological Catchment Area Survey (ECA) have suggested that the risk of developing mania may be increasing in recent generations, although these data may be subject to recall bias.7 Similarly, whether there exist gender differences in incidence or age at onset of bipolar
Ch 01
7/4/05
2
3:34 pm
Page 2
Bipolar disorder: the upswing in research & treatment
disorder is uncertain; earlier studies showed little difference in age at onset between men and women, but more recent studies, using strict operational criteria, have tended to show a later onset in women.8 It also remains uncertain whether migrants have a higher incidence of bipolar disorder as well as schizophrenia, although earlier incidence studies have suggested that this may be so.9,10 Given that our research group have already conducted a number of incidence studies of schizophrenia in Camberwell, we similarly undertook to describe the basic clinical epidemiology of bipolar disorder in this defined catchment area11,12 (and N. Kennedy et al, unpublished) and specifically to address the following questions: 1. 2. 3.
What is the overall incidence of bipolar disorder in Camberwell and has it increased since 1965? (N. Kennedy et al, unpublished) What is the peak incidence of bipolar disorder by age and are there differences in incidence or age at onset by gender?11 Are incidence rates of bipolar disorder higher among African and African–Caribbean migrants than Whites in south London?12
Method Camberwell is an inner-city area that rates highly on deprivation indices. The total population has declined over the years, from 171 000 in 1960 to approximately 120 000 currently. The population has also become increasingly ethnically diverse with currently over a fifth of the population being of African–Caribbean or Black African origin. To describe the clinical epidemiology of bipolar disorder in Camberwell, all adults living in this defined catchment area who presented to psychiatric services between 1965 and 1999 with mania, hypomania, bipolar disorder or any possible psychosis were identified from the Cumulative Camberwell Psychiatric Case Register.3 Patients admitted to hospitals outside the area would normally have been transferred to local hospitals or services for continuing care and these records were also identified in this search. Patients were excluded if they were not resident in the catchment area, had presented previously with a psychotic or manic episode, had a clear organic cause for their symptoms, or had an onset before 16 years of age. All case records were examined and the Operational Checklist for Psychotic Disorders, version 3.4 (OPCRIT)13 completed for the year following presentation. The OPCRIT checklist, based on the Present State Examination,14 was then used to generate DSM-IV diagnoses for cases using the accompanying computer
Ch 01
7/4/05
3:34 pm
Page 3
The clinical epidemiology of bipolar disorder
3
program. Those who met DSM-IV criteria for bipolar I disorder (BPI) or mania became our cases. Bipolar disorder was defined as fulfilling DSM-IV criteria for mania with or without a previous treated depressive episode in primary or secondary care, and mania was defined as fulfilling criteria for mania without a previously treated depressive episode. Data concerning the general population of Camberwell were ascertained from the 1961–91 censuses with intermediate years interpolated. All population data were stratified by age and gender and corrected for under-enumeration.
Results Over the 35-year period, 246 cases fulfilled criteria for DSM-IV BPI from 1443 possible cases of mania, hypomania or psychosis identified. Of these 246 cases, 78% had their first psychiatric presentation with mania, whereas 22% had had a previous treated depressive episode before the onset of first mania. Another 12% described a probable previous depression that was not treated; 141 (57%) cases were female and 106 (43%) were male (female : male rate ratio 1.21). Almost a fifth (16%) were not admitted during index mania and almost three-quarters (72%) were psychotic during their first manic episode. Mean and median ages at onset for mania were 33 and 28 years, respectively, with mean and median ages at onset for bipolar disorder, also including prior treated depressive episodes, being 31 and 26 years, respectively. Peak age at onset was in the 21–25 and 26–30 age groups followed by a second much smaller peak in incidence by age in midlife. The standardized incidence rates of bipolar disorder and mania were 6.5 and 5.2 per 100 000 population, respectively, with females having a somewhat higher incidence of bipolar disorder (female : male rate ratio 1.21). Table 1.1 shows the number of cases of bipolar disorder and mania over each of seven 5-year time bands over the course of this study. The number of cases increased modestly though significantly over the course of the study, particularly during the earlier time periods. This was reflected by a significant rate ratio linear trend, which summarized the increase in risk during each time period for both bipolar disorder (p = 0.034) and mania (p = 0.013). We found that women had a later onset of mania and bipolar disorder by 5 and 4.5 years, respectively, and that this difference remained significant even after adjusting for a number of potentially confounding pre-morbid variables such as ethnicity, family history, developmental abnormalities,
Ch 01
7/4/05
4
3:34 pm
Page 4
Bipolar disorder: the upswing in research & treatment
pre-morbid functioning and employment. There also appeared to be significant differences between men and women in age at onset, as for schizophrenia, with onset being an average of 4–5 years later in women.1 However, age of onset distributions appeared to be different in bipolar disorder and schizophrenia. In schizophrenia men predominated in those with an early onset and women were much more likely than men to have an onset in late life with little difference being found in mid-life.4 By contrast, age at onset differences in bipolar disorder in this study were largely accounted for by women having a higher incidence in mid-life, with men and women having similar incidences in early and late life. We could not directly estimate incidence of bipolar disorder by ethnicity in this 35-year study, as, prior to the 1981 census, population data by ethnicity were based on head of household unadjusted for age or gender, whereas subsequent data are much more accurate. Therefore, to investigate incidences of bipolar disorder in different ethnic groups, our group conducted a 2-year (1997–99) prospective incidence study of first-episode psychosis and mania, the AESOP study, which included the entire Camberwell
Table 1.1 Number of cases of DSM-IV bipolar I disorder and mania during each of seven 5-year time periods over the course of the study Time period
Bipolar I disorder
Mania (no previous depression)
1965–69
23
18
1970–74
33
22
1975–79
37
28
1980–84
43
36
1985–89
36
28
1990–94
38
34
1995–99
36
28
Table 1.2 Incidence ratios of first-episode bipolar disorder by ethnicity in south London (AESOP Study 1997–99) Ethnicity South London
Adjusted rate ratio irr (95% CI)
White
1
Overall non-White
5.5 (2.8–10.6)
Black Caribbean
7.6 (3.7–15.8)
African
4.4 (1.7–11.1)
Ch 01
7/4/05
3:34 pm
Page 5
The clinical epidemiology of bipolar disorder
5
catchment area. As seen from Table 1.2, rates of bipolar disorder among the Black Caribbean and Black African population in London were 7.6 and 4.4 times that of the White population over this 2-year time period, confirming earlier data that, as for schizophrenia, rates of mania were elevated in migrant groups in south London.
Discussion In this 35-year incidence study, overall incidence rates for mania were much higher than in previous studies (Table 1.3).15–19 Why were our rates of mania so high? First, previous studies had a number of methodological limitations, which the current study was designed to overcome. For example, some previous studies excluded non-admitted15,16 or non-psychotic cases,18,19 which would have excluded almost a fifth and over a quarter of our sample, respectively. Similarly, previous studies had an upper age limit of either 60 or 65 years, which would have excluded almost a tenth of our sample.15,16,18 Second, almost 40% of our sample over the course of the study and over 20% of the current population of Camberwell are of Black Caribbean or Black African origin. The influx of ethnic minority groups may partially have accounted for the high incidence of mania in this study, as incidence of bipolar disorder appears to be higher in these ethnic groups in south London. Finally, other factors such as the effects of inner-city living or higher risk of abuse of drugs such as cannabis have been suggested as contributing to the high rates of schizophrenia in urban areas1 and potentially could also contribute to high incidences of mania. Similarly, our finding from the AESOP study, that rates of bipolar disorder are higher among
Table 1.3 Incidence studies of first-episode mania* Incidence per 105 per year Incidence studies
Number of cases
Male
Female
All
Leff et al (1976)15
63
3.1
2.2
2.6
Daly et al (1995)16
30
4.1
4.9
4.5
Veijola et al (1996)17
2
—
—
1.7
Brewin et al (1997)18
22
1.5
4.1
2.8
Scully et al (2002)19
8
3.7
0.6
2.2
Current 35-year incidence study
194
5.1
5.2
5.2
*Data from two incidence studies omitted as some cases may overlap with current study.
Ch 01
7/4/05
6
3:34 pm
Page 6
Bipolar disorder: the upswing in research & treatment
those of Black African or Black Caribbean origin in south London, may reflect social disadvantage, more adverse life events or higher rates of illicit drug abuse in such groups.5 Age at onset of mania and bipolar disorder in this study, with peak onset in the late twenties, was similar to that reported in previous incidence studies and later than reported in prevalence studies such as the ECA.20 This may reflect delays between onset of this disorder (as reported by prevalence studies) and initial presentation to psychiatric services (as reported in incidence studies).20 A significant increase in the incidence of mania and bipolar disorder was seen over the three decades of this study. Results from the few previous studies describing change in incidence of mania over shorter time periods have been equivocal.21–23 However, cohort data from the ECA have suggested that the risk of developing mania has been increasing in recent generations.7 The increase in incidence observed in this study was much more modest than the increased incidence of schizophrenia seen in the same catchment area over a similar time period. Interestingly, the increase in incidence was mainly seen over the first 20 years of this study, which was also the time of greatest population change and migration. Therefore, the increase in incidence over time, similar to schizophrenia, may reflect environmental factors such as migration, socioeconomic change or increasing abuse of alcohol or illicit drugs. In contrast to the gender differences in age at onset found in the current study, early studies of bipolar disorder have generally found little difference in age of onset between men and women.24 However, early studies did not generally use strict operational criteria and were generally based on consecutive admissions or outpatient attendances rather than on epidemiological samples, and therefore they may have been less representative of typical patients with bipolar disorder than the current sample. Furthermore, a number of more recent cross-sectional studies using strict operational criteria have found that women have a later onset of both bipolar disorder and mania.8 Similar age at onset differences, with men having a significantly earlier onset than women, have been described in schizophrenia. However, in schizophrenia men predominate in those with an onset in early life and women in those with an onset in late life. Gender differences in age at onset of schizophrenia have been explained by (1) abnormal neurodevelopment more commonly affecting men; or (2) the protective effects of higher oestrogen levels in women, with a decline during the menopause leading to a surge in incidence in late life among women.1,2 By contrast, the major gender differences in age at onset in bipolar disorder observed in this study were in mid-life, so the explanations for gender differences in age at onset of bipolar
Ch 01
7/4/05
3:34 pm
Page 7
The clinical epidemiology of bipolar disorder
7
disorder are likely to be different. Women may delay seeking treatment compared with men or social factors such as urban living, deprivation or substance misuse may particularly affect young men and therefore influence age at onset differences between men and women.
Acknowledgements We wish to acknowledge the contributions of Dr Jane Boydell, Professor David Castle, Professor Peter Jones, Professor Nori Takei, Professor Jim van Os and Professor Simon Wessely, for their advice and assistance in conducting this study and Professor Peter McGuffin for use of the OPCRIT program. The study was supported by a grant from the Stanley Institute for Medical Research to Professor Murray, and a grant from the Psychiatry Research Trust to Dr Kennedy.
References 1. 2. 3. 4.
5.
6.
7. 8. 9. 10.
Murray RM, Jones PB, Susser E et al, The Epidemiology of Schizophrenia. Cambridge University Press: Cambridge, 2002. Häfner H, Gender differences in schizophrenia. Psychoneuroendocrinology 2003; 28:17–54. Boydell J, van Os J, Lambri M et al, Incidence of schizophrenia in south-east London between 1965 and 1997. Br J Psychiatry 2003; 182:45–49. Castle DJ, Wessely S, van Os J, Murray RM, The effect of gender on age at onset of psychosis. In: Goldberg D, ed. Psychosis in the Inner City: The Camberwell First Episode Study. Maudsley Monographs no. 40. Psychology Press: Hove, 1998:27–36. Sharpley M, Hutchinson G, McKenzie K, Murray RM, Understanding the excess of psychosis among the African–Caribbean population in England. Review of current hypotheses. Br J Psychiatry 2001; 178(Suppl 40):s60–68. Lloyd T, Jones PB, The epidemiology of first-onset mania. In: Tsuang MT, Tohen M, eds. Textbook in Psychiatric Epidemiology. 2nd edn. Wiley-Liss: New York, 2002:445–458. Lasch K, Weissman M, Wickramaratne P et al, Birth-cohort changes in the rates of mania. Psychiatr Res 1990; 33:31–37. Arnold LM, Gender differences in bipolar disorder. Psychiatr Clin North Am 2003; 26:595–620. Der G, Bebbington PE, Depression in inner London: a register study. Soc Psychiatry 1987; 22:73–84. Van Os J, Takei N, Castle DJ et al, The incidence of mania: time trends in relation to gender and ethnicity. Soc Psychiatry Psychiatr Epidemiol 1996; 31:129–136.
Ch 01
7/4/05
8 11.
12.
13.
14.
15. 16. 17. 18.
19.
20. 21. 22.
23.
24.
3:34 pm
Page 8
Bipolar disorder: the upswing in research & treatment Kennedy N, Boydell J, Kalidindi S et al, Gender differences in incidence and age at onset of mania and bipolar disorder over a 35-year period in Camberwell, England. Am J Psychiatry 2005; 162:257–262. Lloyd T, Kennedy N, Fearon P et al, Incidence of bipolar affective disorder in three UK cities: results from the AESOP study. Br J Psychiatry 2005; 186:126–131. McGuffin P, Farmer A, Harvey I, A polydiagnostic application of operational criteria in studies of psychotic illness. Development and reliability of the OPCRIT system. Arch Gen Psychiatry 1991; 48:764–770. Wing JK, Cooper JE, Sartorius N, Measurement and Classification of Psychiatric Symptoms. An Instruction Manual for the PSE and Catego Program. Cambridge University Press: New York, 1974. Leff JP, Fischer M, Bertelsen A, A cross-national epidemiological study of mania. Br J Psychiatry 1976; 129:428–437. Daly I, Webb M, Kaliszer M, First admission incidence study of mania, 1975–1981. Br J Psychiatry 1995; 167:463–468. Veijola J, Rasanen P, Isohanni M et al, Low incidence of mania in northern Finland. Br J Psychiatry 1996; 168:520–521. Brewin J, Cantwell R, Dalkin T et al, Incidence of schizophrenia in Nottingham. A comparison of two cohorts, 1978–80 and 1992–94. Br J Psychiatry 1997; 171:140–144. Scully PJ, Quinn JF, Morgan MG et al, First-episode schizophrenia, bipolar disorder and other psychosis in a rural Irish catchment area: incidence and gender in the Cavan–Monaghan study at 5 years. Br J Psychiatry 2002; 43(Suppl):S3–S9. Bebbington P, Ramana R, The epidemiology of bipolar affective disorder. Soc Psychiatry Psychiatr Epidemiol 1995; 30:279–292. Mander AJ, Diagnosis change, lithium use and admissions for mania in Edinburgh. Acta Psychiatr Scand 1989; 80:434–436. Eagles JM, Whalley LJ, Ageing and affective disorders: the age at first onset of affective disorders in Scotland, 1969–1978. Br J Psychiatry 1985; 147:180–187. Parker G, O’Donnell M, Walter S, Changes in the diagnosis of the functional psychosis associated with the introduction of lithium. Br J Psychiatry 1985; 146:377–382. Goodwin FK, Jamison KR, Course and outcome. In: Goodwin FK, Jamison KR, eds. Manic Depressive Illness. Oxford University Press: New York, 1990: 127–156.
Ch 02
7/4/05
3:35 pm
Page 9
chapter 2
The functional outcome of bipolar disorder Mauricio Tohen and Julie M Niswander
Introduction The longitudinal course of bipolar disorder is defined by recurrent manic and depressive mood episodes. Clinicians treating patients with bipolar disorder are observant of the severe impact these mood episodes have on the lives of patients and their families, including job performance and personal relationships and responsibilities. Despite the magnitude of the impact of mood episodes on day-to-day function, most bipolar disorder studies conducted to date have examined measures of symptoms and syndromal outcome as opposed to more patient-relevant functional improvement. Nonetheless, the study of functional outcome is critical: a recent study has shown that, 12 months after hospitalization for bipolar disorder, syndromal recovery measured 61.0%, whereas functional recovery was reported at only 36.0%.1 Restoration of pre-episode quality of life and level of functioning is of primary importance to the bipolar patient and should therefore serve as the guiding principle in the study and treatment of bipolar disorder. This chapter discusses prospective studies conducted at McLean Hospital in Belmont, Massachusetts that examined functional recovery and sought to determine predictors of functional outcome in patients with bipolar disorder.
Outcomes and identification of predictors A 4-year prospective follow-up Following patients in a naturalistic setting provides insight on the longitudinal course of bipolar disorder. Although there are no definitive predictors of
Ch 02
7/4/05
10
3:35 pm
Page 10
Bipolar disorder: the upswing in research & treatment
future course in bipolar disorder, the outcome after recovery of an index manic episode may identify factors that predict continued remission, interepisode symptoms and functional outcomes. This section presents findings from a naturalistic study conducted in the mid-to-late 1980s at the McLean Hospital (Belmont, Massachusetts, USA), the largest psychiatric teaching facility at Harvard Medical School.2 In this study, a cohort of 75 patients who met the DSM-III criteria for bipolar disorder and had recovered from an index manic episode at time of discharge were followed for 4 years with assessments at 6 and 48 months after discharge. The patients in this study were ≥ 17 years of age, 97% (n = 73) of them being Caucasian, and they did not include patients with mixed or rapid-cycling symptoms; 32% (n = 24) of patients were experiencing their first affective episode. Syndromic recovery was defined as the presence of no more than two DSM-III ‘B’ criteria for an affective episode of mild intensity (< 3) and absence of ‘A’ criteria. Relapse was indicated by meeting the DSM-III criteria for an episode of mania or depression after having achieved remission (recovery of at least 6 consecutive weeks). A key outcome to this study was that no patients were lost to follow-up. In this study, the probability of remaining in remission was 51% at the end of the first year and 44% at 24 months, 33% at 36 months, and 28% at 48 months (Figure 2.1). Interestingly, most relapses occurred in the first year, with only 23% (17 of 75) more patients relapsing by 48 months; the probability of remaining in remission increased the longer the patient had maintained recovery. After an index manic episode, cumulative probabilities suggested that the risk of relapsing into depression was highest during the initial 9 months after recovery, but the risk of relapse into mania remained relatively constant during the 4 years of follow-up. Several risk factors were significantly associated with a shorter time in remission. The presence of psychotic features during the index manic episode predicted time to relapse. At 6 months, the probability of remaining in remission was similar for patients stratified by the absence or presence of psychotic features (67% vs. 63%, respectively; Figure 2.2). However, over time, the probability of remaining in remission differed, based on the presence of psychosis during the index episode; at 3 years, the probability of remaining in remission was 52% for patients without psychotic features and 26% for patients with psychotic features. Differentiation of mood congruence of psychotic features had additional prognostic relevance as moodincongruent psychotic features may predict a shorter time in remission3. The median time in remission for patients with mood-incongruent psychotic mania was 8 months, contrasting with 33 months for patients having
7/4/05
3:35 pm
Page 11
The functional outcome of bipolar disorder
11
1.0 0.9 0.8 0.7 0.6 P (t)
Ch 02
0.5 0.4 0.3 0.2
Major depressive episode Manic episode Manic or major depressive episode
0.1 0.0 0
Number of patients at risk
6
12
18 24 30 Months at risk (t)
36
42
48
75------------ 53------------ 52------------ 51------------ 51 -------------------------- 50-------------------------- 47 75------------ 66------------ 58------------ 55------------ 54 -------------------------- 44-------------------------- 42 75------------ 48------------ 38------------ 34------------ 33 -------------------------- 25-------------------------- 21
Figure 2.1 The cumulative probability of remaining in remission for the number of patients with bipolar disorder who have not relapsed after recovery from an index manic episode up to a given time (dark purple). The probabilities of not relapsing into a manic (grey) or major depressive (light purple) episode are also represented. (Derived from reference 2.)
experienced mood-congruent psychotic features during the index episode (Figure 2.3). In addition to the presence of psychotic features, other predictors of poor outcome included depressive symptoms during the index episode, a history of mood episodes, and a history of alcoholism (Table 2.1). These risk factors were significantly associated with a shorter time in remission. Furthermore, depressive symptoms during the index episode predicted time to relapse into a depressive episode, whereas a history of one or more previous episodes (mania or depression) significantly reduced time to relapse into a manic episode. As both occupational and residential status are indicators of the ability to live productively and independently, these markers serve as measures of functional outcome. Table 2.2 lists the functional outcomes of patients at both 6 and 48 months as assessed by the Modified Vocational Status Index (MVSI) and Modified Location Code Index (MLCI). Of 72 patients, 60% (n = 43) were able to work or study, and 64% (n = 46) had independent
7/4/05
12
3:35 pm
Page 12
Bipolar disorder: the upswing in research & treatment 1.0 0.9 0.8 0.7 P (t)
Ch 02
0.6 0.5 0.4 0.3 Without psychotic features With psychotic features
0.2 0.1 0 Number of patients at risk
6
12
18 24 30 Months at risk (t)
54----------- 34------------ 26----------- 22 ----------- 21 ------------------------21----------- 14------------ 12----------- 12 ----------- 12 -------------------------
36
42
14--------------------------11---------------------------
48
11 10
Figure 2.2 The cumulative probability of not relapsing into any mood episode up to a particular time for patients with bipolar disorder stratified by the presence (purple) and absence (grey) of psychotic features during an index manic episode. (Derived from reference 2.)
residential status 6 months after discharge. However, at 48 months of follow-up, 28% (n = 20) were unable to work or study, and 19% (n = 14) of patients were not able to live independently. Further analyses identified several variables to be significant predictors of an unfavourable outcome (Table 2.2). Risk factors associated with a poor occupational status 6 months after discharge included a history of one or more previous episodes and a history of alcoholism. These factors continued significantly to predict a poor occupational status at 48 months. In addition, psychotic features during the index episode also predicted poor occupational status at 48 months. Predicting an unfavourable residential status at 6 months was a history of alcoholism. Other factors associated with unfavourable living status at 48 months included a history of one or more previous episodes and male sex. The results of this study suggest that patients with bipolar disorder improve between 6 months and 4 years after an index manic episode; however, functional outcome remains less than ideal. A key finding of this study was that predictors of functional outcome varied depending not only on the type of outcome measured but also on the time at which the outcome was
7/4/05
3:35 pm
Page 13
The functional outcome of bipolar disorder
13
1.0 0.9
Patients with mood-congruent psychotic mania Patients with mood-incongruent psychotic mania
0.8 0.7 0.6 P (t)
Ch 02
0.5 0.4 0.3 0.2 0.1 0.0 0
6
12
18 24 30 Months at risk (t)
36
42
48
Number of 24-----------17----------15-----------14----------14 -----------------------11-------------------------8 patients at 30-----------17----------11------------8------------7 -------------------------3-------------------------3 risk
Figure 2.3 The cumulative probability of not relapsing into any mood episode up to a particular time for patients with bipolar disorder experiencing moodcongruent (purple) and mood-incongruent (grey) psychotic features during an index manic episode. (Derived from reference 3.)
Table 2.1 Risk factors associated with a shorter time in remission after an index manic episode Predictors of time to relapse
HRa*
Psychotic features during index episode
2.2
p Value 0.05
Depressive symptoms during index episode
2.0
0.04
History of alcoholism
3.9
0.02
*Hazard ratio was adjusted (HRa) simultaneously for all variables with Cox regression. Derived from reference 2.
assessed. However, as in all naturalistic studies, a significant uncontrolled variable in this study was treatment. Not controlling for treatment has the advantage of obtaining information on treatments received by patients under non-controlled circumstances. In addition, findings of naturalistic studies may be more readily generalizable. At discharge, 97% (73 of 75) of patients were treated with at least one psychotropic drug and 92% (n = 69)
Ch 02
7/4/05
14
3:35 pm
Page 14
Bipolar disorder: the upswing in research & treatment Table 2.2 Functional outcomes and predictors of outcome after an index manic episode Functional outcomes
n (%)
ORa
p Value
6 months Occupational status Able to work/study (< 2 MVSI)
43 (60)
—
—
Unable to work/study (> 3 MVSI)
29 (40)
—
—
Predictors of poor outcome One or more previous mood episodes
—
5.6
0.001
History of alcoholism
—
10.3
0.03
Poor occupational status at baseline
—
15.0
0.05
Residential status Able to live independently (< 3 MLCI)
46 (64)
—
—
Unable to live independently (> 4 MLCI)
26 (36)
—
—
—
14.7
0.02
Predictors of poor outcome History of alcoholism 48 months Occupational status Able to work/study (< 2 MVSI)
52 (72)
—
—
Unable to work/study (> 3 MVSI)
20 (28)
—
—
Predictors of poor outcome One or more previous mood episodes
—
5.4
0.05
History of alcoholism
—
8.2
0.04
Psychotic features during index episode
—
9.0
0.04
Residential status Able to live independently (< 3 MLCI)
58 (81)
—
—
Unable to live independently (> 4 MLCI)
14 (19)
—
—
One or more previous mood episodes
—
4.9
0.01
Male sex
—
6.0
0.02
Predictors of poor outcome
ORa, adjusted odds ratio; MVSI, Modified Vocational Status Index; MLCI, Modified Location Code Index. Derived from reference 2.
of patients received lithium. Fifty-five per cent of patients (n = 41) received an antipsychotic; 15% (n = 11), an antidepressant; and 11% (n = 8), an anticonvulsant. At 48 months, 79% (57 of 72) of patients were receiving at least one medication, with 67% (n = 48) taking lithium; 46% (n = 33), an antipsychotic; 19% (n = 14), an antidepressant; and 21% (n = 15), an anticonvulsant.
Ch 02
7/4/05
3:35 pm
Page 15
The functional outcome of bipolar disorder
15
The McLean–Harvard first-episode study To study the evolution of bipolar disorder comprehensively, the longitudinal follow-up of first-episode patients is critical. A prospective study of bipolar disorder patients that commences near illness onset provides data less confounded by prolonged illness and pre-defined poor outcomes. In a naturalistic setting, the quantification of recovery and identification of predictors of outcome are especially valuable. The McLean–Harvard First-Episode Mania Study4 analysed 166 hospitalized patients experiencing their first manic (75.3%, n = 125) or mixed (24.7%, n = 41) episode, with 88.6% (n = 147) experiencing psychotic features. Patients were recruited for this study between 1989 and 1996. Syndromic recovery was defined as a severity rating of ≤ 3 for the DSMIV ‘A’ criterion for mania, with no ‘B’ criterion rated > 3 and no two ‘B’ criteria rated at 3; Clinical Global Impressions scores were required to be ≤ 2. Patients presenting with an initial mixed episode fulfilled recovery criteria for a manic and a depressive episode. Syndromal remission was achieved by maintaining recovery for at least 8 weeks. Furthermore, symptomatic recovery, reflecting minimal symptom severity, was defined by a total Young Mania Rating Scale score of ≤ 5 and the Hamilton Depression Rating scale score of ≤ 8. The functional recovery from a first lifetime mood episode required that both occupational level and residential status return to or exceed their highest levels during the pre-intake year. As defined, 59.7% (92 of 154) of patients who attained syndromal recovery after initial hospitalization achieved remission and remained in remission by the end of 2 years of follow-up. Conversely, 5.8% (n = 9) experienced an early relapse, and 34.4% experienced a new episode (mania: n = 24; mixed: n = 5; depression: n = 24). Overall, the median latency to 50% risk of any new episode was 26.3 weeks. The risks of new manic or depressive episode were equal in incidence (20.1%); however, the time to new depression was shorter at 17.7 weeks versus 31.6 weeks to new mania. In this cohort, survival-computed proportions reflected 85.5% (142 of 166) of patients recovered syndromically at 6 months (Table 2.3). Nevertheless, only 39.5% (60 of 152) of patients achieved functional recovery at this time point. At the 2-year follow-up, nearly all patients achieved syndromal recovery (97.6%, n = 162), with 71.7% (66 of 92) of patients achieving symptomatic recovery. However, only 43.1% (59 of 137) of patients functionally recovered, as measured by residential and occupational status. Several factors were associated with syndromal and functional recovery: a shorter initial hospitalization, below-median baseline depression ratings and being female predicted a shorter time to syndromal
142/166
60/152
Syndromal
Functional
39.5
85.5
%
59/137
162/166
n/N
43.1
97.6
%
2.82
Shorter initial hospitalization
1.72
Female sex
3.28
1.65
> 30 years
1.99
Below-median baseline depression ratings
Ratio*
Shorter initial hospitalization
Predictors of recovery
Predictors of functional recovery are described as odds ratios (logistic regression). Derived from reference 4.
*Predictors of syndromal recovery are described as hazard ratios (Cox regression).
n/N
Recovery
2 years
0.006
0.001
0.008
0.008
0.001
p Value
16
3:35 pm
6 months
7/4/05
Table 2.3 Recovery and predictors of recovery following a first lifetime manic/mixed mood episode
Ch 02 Page 16
Bipolar disorder: the upswing in research & treatment
7/4/05
3:35 pm
Page 17
The functional outcome of bipolar disorder
17
recovery (Table 2.3). The likelihood of achieving functional recovery was significantly associated with being older than 30 years and having a shorter initial hospitalization. Interestingly, different factors predicted relapse into a new manic versus a new depressive episode. The presence of mood-congruent psychotic features during the index episode predicted mania, but not depression. Whereas a lower pre-morbid occupational status (MVSI < 3 in pre-intake year) predicted mania, a relatively high occupational status (MVSI ≥ 3) predicted depression (Figure 2.4). The presence of co-morbidity did not predict relapse into mania, but was associated with relapse into depression. Not surprisingly, if the index episode was a mixed mood episode, the risk of relapsing into depression was significant, whereas an initial manic state predicted new episodes of mania (Figure 2.5). In summary, a shorter time to onset of a new manic episode was predicted in patients whose initial episode was mania and whose pre-morbid occupational status was low. However, a shorter time to a new depressive episode was predicted in patients experiencing an initial mixed episode and in patients with a higher occupational status. This study reported baseline co-morbidity at 53.0% (88 of 166), with 8.4% of patient co-morbidity categorized as Axis I; 18.7%, substance abuse; and 31.3%, medical. It is noteworthy that approximately 30% of firstepisode bipolar patients presented with a pre-morbid condition (defined as
New mania
100
% Remaining well before a new episode
Ch 02
95
95
90
90
85
85
80
80
75
75
70
70 High occupational status
65
Low occupational status
60 0
24 48 72 96 Weeks of recovery from index episode
New depression
100
High occupational status
65
Low occupational status
60 0
24 48 72 96 Weeks of recovery from index episode
Figure 2.4 Level of pre-morbid occupational status predicts new episodes of mania and depression after recovery from a first lifetime manic or mixed episode. (Derived from reference 4.)
7/4/05
3:35 pm
18
Page 18
Bipolar disorder: the upswing in research & treatment
New mania
100
% Remaining well before a new episode
Ch 02
95
95
90
90
85
85
80
80
75
75
70
70
65
65
60
Index: manic
55
Index: mixed
50
New depression
100
60
Index: manic
55
Index: mixed
50 0
24 48 72 96 Weeks of recovery from index episode
0
24 48 72 96 Weeks of recovery from index episode
Figure 2.5 Time to onset of new episode of mania or depression predicted by the index episode. (Derived from reference 4.)
a medical condition that required medical treatment or medical pharmacological treatment). Furthermore, in a separate study, patients with the first lifetime manic episode aged 65 years and over (n = 14) were significantly more likely to have a pre-morbid neurological condition as compared with bipolar patients of similar age (n = 36).5 At discharge, almost all patients were prescribed a psychotropic medication (95.2%, n = 158). Although treatments varied widely, 75.3% (n = 125) of patients received an antipsychotic, and lithium was prescribed to 68.7% (n = 114) of patients. At 2-year follow-up, lithium was the most frequent treatment, although 35.6% (48 of 135) of patients were taking no medication. The use of antidepressants increased during the 2 years of follow-up. No specific treatment was associated with time to syndromal recovery; similarly, no treatment was significantly associated with a shorter latency to a new mood episode. Overall, patients experiencing a first lifetime manic episode improved during a 2-year follow-up, with 97.6% achieving syndromal recovery. Nonetheless, only 72% reported symptomatic recovery, and functional recovery was attained by fewer than 50% of patients. These findings suggest that patients with bipolar disorder recover syndromically before they recover functionally; symptom severity improves initially followed by a return to pre-episode functioning that requires additional time to achieve. Furthermore, in this cohort of patients, substance abuse or dependence occurred with a co-morbidity of 18%, relatively low compared with a 60%
Ch 02
7/4/05
3:35 pm
Page 19
The functional outcome of bipolar disorder
19
co-morbidity in multiple-episode patients.6 This finding suggests that, in some patients with bipolar disorder, mania may develop first and is followed by a substance use disorder.7
Conclusions Prospective, naturalistic study designs produce a non-biased description of patient outcomes in a clinical setting whereby treatment is not determined by the investigator. Results of such studies advance our understanding of bipolar disorder and help identify illness characteristics and risk factors that may predict outcomes and aid in developing optimal treatment intervention. The studies presented here suggest a disparity in recovery of patients with bipolar disorder. Although achieving syndromic recovery, many patients continue to experience symptoms; subsyndromal morbidity encroaches on work and personal life, accounting for reduced occupational and residential status and poor functional outcome. The McLean–Harvard First-Episode Mania Study found that, after a first lifetime manic/mixed episode, only 43% of patients achieved functional recovery at 2 years.4 Previous studies on effective functioning in bipolar patients have reported similar poor outcomes. Accordingly, only 36% of patients with one or no previous hospitalizations return to pre-morbid function at 12 months;1 a second study described 27% of patients with good overall functioning at the 2-year follow-up.8 Keck et al reported that 24% of patients with a history of previous manic or mixed episodes achieved functional recovery at some time between discharge and 12 months of follow-up.9 Moreover, 26% of patients with bipolar disorder have been described as having a good overall outcome at 2 years, increasing to 47% at 4.5 years.10 Although the majority of these studies were conducted before the availability of today’s newer drugs, treatments for bipolar disorder, even in first-episode mania, have been less than ideal. Treatments are needed that are both effective at improving functional outcome and safe in a patient population with significant medical and substance abuse co-morbidities. The analysis of functional outcome predictors identifies patients at risk for poor psychosocial recovery. According to the studies presented here, a poor functional outcome was predicted by one or more previous episodes, a history of alcoholism and the presence of psychotic features during the index episode. In contrast, a functional recovery was predicted by having a shorter initial hospitalization and being older than 30. Additionally, other
Ch 02
7/4/05
20
3:35 pm
Page 20
Bipolar disorder: the upswing in research & treatment
related studies have identified higher pre-morbid function and higher socioeconomic class with favourable functional outcomes.1,9 To summarize, in the current treatment of bipolar disorder, symptomatic improvement is not correlative with functional improvement. This finding necessitates a greater clinical emphasis on functional outcomes. Functional recovery must be the hallmark of drug development, clinical study design and treatment intervention, whereby patients with bipolar disorder may enjoyably and responsibly return to personal, family and work life.
References 1.
2.
3.
4.
5. 6.
7. 8.
9.
10.
Strakowski SM, Keck PE Jr, McElroy SL et al, Twelve-month outcome after a first hospitalization for affective psychosis. Arch Gen Psychiatry 1998; 55:49–55. Tohen M, Waternaux CM, Tsuang MT, Outcome in mania. A 4-year prospective follow-up of 75 patients utilizing survival analysis. Arch Gen Psychiatry 1990; 47:1106–1111. Tohen M, Tsuang MT, Goodwin DC, Prediction of outcome in mania by moodcongruent or mood-incongruent psychotic features. Am J Psychiatry 1992; 149:1580–1584. Tohen M, Zarate CA Jr, Hennen J et al, The McLean–Harvard First-Episode Mania Study: prediction of recovery and first recurrence. Am J Psychiatry 2003; 160:2099–2107. Tohen M, Shulman KI, Satlin A, First-episode mania in late life. Am J Psychiatry 1994; 151:130–132. Regier DA, Farmer ME, Rae DS et al, Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area (ECA) Study. JAMA 1990; 264:2511–2518. Strakowski SM, DelBello MP, The co-occurrence of bipolar and substance use disorders. Clin Psychol Rev 2000; 20:191–206. Goldberg JF, Harrow M, Grossman LS, Course and outcome in bipolar affective disorder: a longitudinal follow-up study. Am J Psychiatry 1995; 152:379–384. Keck PE Jr, McElroy SL, Strakowski SM et al, 12-month outcome of patients with bipolar disorder following hospitalization for a manic or mixed episode. Am J Psychiatry 1998; 155:646–652. Goldberg JF, Harrow M, Consistency of remission and outcome in bipolar and unipolar mood disorders: a 10-year prospective follow-up. J Affect Disord 2004; 81:123–131.
Ch 03
7/4/05
3:36 pm
Page 21
chapter 3
Brain abnormalities in bipolar disorder: do they exist and do they change? E Serap Monkul and Jair C Soares
This chapter focuses on recent work that has utilized in vivo brain imaging to understand the mechanisms involved in bipolar disorder. Structural magnetic resonance imaging (MRI) and neurochemical studies with magnetic resonance spectroscopy (MRS) have identified changes in prefrontal cortex regions, which are highly interconnected with limbic portions of the brain, including medial temporal lobe regions and the striatum.1 Some of these regions or the connections between them may be impaired and possibly result in the mood dysregulation that we see in patients who have mood disorders.2,3 One of the regions of interest is the anterior cingulate, which is thought to be involved in the pathophysiology of mood disorders. In some of our prior work we measured the cingulate gyrus, subdivided into specific regions, and found a reduction in the grey matter content in the left anterior cingulate in untreated bipolar patients compared with healthy controls.4 Reduction in anterior cingulate grey matter volumes4,5 and density6,7 is a consistently reported finding in recent studies. Cingulate findings are also present in children and adolescents with bipolar disorder,5 and there is evidence that lithium may protect or reverse grey matter changes in this particular brain region.4 Drevets and colleagues found a pronounced reduction in grey matter primarily on the left side in a specific part of the anterior cingulate gyrus, lying ventral to the genu of the corpus callosum, called the ‘subgenual prefrontal cortex’, in subjects with familial mood disorder compared with healthy controls.8 We have not been able to replicate this finding in a recent study, with familial and non-familial bipolar patients, although our sample involved patients who were generally less severely ill9 than those of Drevets et al.8
Ch 03
7/4/05
22
3:36 pm
Page 22
Bipolar disorder: the upswing in research & treatment
Another region that has attracted much attention for research on the pathophysiology of mood disorders is the amygdala, which is involved in regulation of emotions such as fear and anxiety. There are intriguing reports by different groups who have found enlargement of the amygdala in bipolar disorder.10–12 The neuropathology underlying such enlargement is still unclear. The other topic to discuss briefly relates to the hyperintense lesions, which are non-specific markers, and generally reflect changes in the water content in the brain. They are present in normal aging, dementia, epilepsy, schizophrenia, and several other neuropsychiatric disorders.13 In the literature on mood disorders, there are several studies suggesting that patients with bipolar disorder have these lesions at higher rates compared with wellmatched healthy controls,2,13 although controversy exists as to whether individuals with bipolar disorder are more likely than other psychiatric patients to have these lesions.14,15 These lesions seem to be more directly involved in late life depression,16 and their importance may perhaps be related to disrupting pathways that interconnect those brain circuits that modulate mood.2 The other technique to discuss is MRS, which utilizes the same hardware as MRI. MRS results are in the form of graphs from which one can quantitate certain chemicals of interest, such as myoinositol, N-acetylaspartate (NAA), choline-containing compounds and many others.17 NAA is a known specific marker of neuronal viability and functioning, and any type of brain insult will result in decreased levels of NAA in the brain. There are also data showing that, if those insults are removed, levels of NAA go back up. There have been two published studies in the bipolar literature suggesting a decrease in NAA levels in the dorsolateral prefrontal cortex.18,19 Most of the anatomical studies did not find detectable anatomical changes in the dorsolateral prefrontal cortex, so decreased NAA levels could be an early marker of neuronal impairment in this particular brain region, before anatomical changes take place. The question we asked in our most recent studies had to do with whether young bipolar patients (children and adolescents who have bipolar disorder) had such a change already early on, or whether it was something that developed over time (J.C. Soares, unpublished work). This was a cross-sectional study, and bipolar children/adolescents (mean age 13 years, well matched with healthy controls) had a reduction of NAA levels in the dorsolateral prefrontal cortex, which is consistent with what has been reported previously in adults18 and children.19 In the same
Ch 03
7/4/05
3:36 pm
Page 23
Brain abnormalities in bipolar disorder
23
sample, there was a pronounced reduction in grey matter not only on the left side, but also on the right side, so the reduction in grey matter volumes seems to be even more extensive in this sample with early onset of disease. It is also of interest that these children primarily had familial bipolar disorder, as almost all of them had a first-degree relative with bipolar disorder, most often their parent. These findings suggest that some changes might be already present in the prefrontal cortex at the onset of the symptoms or early in the disease process. Recently we published a cross-sectional study where we reported that, over time, patients with bipolar disorder may lose more grey matter compared with healthy controls.20 This study revealed an inverse relationship between age and total grey matter volume in bipolar disorder patients, significantly more pronounced than would be expected in healthy controls, suggesting that bipolar patients are losing grey matter at faster rates than healthy individuals. Both age and length of illness appear to be important factors in some of the structural imaging changes that we find in bipolar patients. Further work needs to be done to examine their relationship with regional brain changes, with the hypothesis that, perhaps in some of those regions involved in mood regulation, one would see more striking relationships with age and length of illness. Another interesting finding has emerged from our recent structural MRI studies in children and adolescents with bipolar disorder. Although amygdala volumes appear to be reduced in young bipolar patients21–23 in contrast to adult patients,10–12 we24 found a direct correlation between age and amygdala volumes in a patient group with mean age of 16, suggestive of neurodegenerative and/or compensatory mechanisms as the disease progressed. This was not a follow-up study, but it is an intriguing finding that suggests that perhaps during adolescence there are abnormal neurodevelopmental processes affecting the medial temporal lobe structures in bipolar patients. In conclusion, bipolar disorder, in both the paediatric and the adult age groups, seems to involve frontolimbic brain abnormalities, both structural and functional.
The relationship of these with specific illness domains,
course and treatment response has not yet been characterized. Further studies will be needed to characterize the role of such abnormalities in the pathophysiology of the illness and their origin, which could be neurodevelopmental and/or neurodegenerative.
Ch 03
7/4/05
24
3:36 pm
Page 24
Bipolar disorder: the upswing in research & treatment
Acknowledgements This work was partly supported by NIH grants MH 01736, MH 068766 and M01-RR-01346 (UTHSCSA GCRC), NARSAD, the Veterans Administration (VA Merit Review), and the Krus Endowed Chair in Psychiatry (UTHSCSA).
References 1. 2. 3. 4.
5. 6. 7. 8. 9.
10.
11.
12. 13.
14.
15.
Tekin S, Cummings JL, Frontal-subcortical neuronal circuits and clinical neuropsychiatry: an update. J Psychosom Res 2002; 53:647–654. Soares JC, Mann JJ, The anatomy of mood disorders – review of structural neuroimaging studies. Biol Psychiatry 1997; 41:86–106. Strakowski SM, DelBello MP, Adler C et al, Neuroimaging in bipolar disorder. Bipolar Disord 2000; 2:148–164. Soares JC, Sassi RB, Brambilla P et al, Decreased left anterior cingulate volumes in untreated bipolar disorder patients. Presented at the Society for Neuroscience Meeting, Orlando, FL: 2–7 November, 2002. Kaur S, Sassi R, Axelson D et al, Anatomical MRI study of cingulate cortex in adolescent bipolar patients. Biol Psychiatry 2003; 53:72S. Doris A, Belton E, Ebmeier KP et al, Reduction of cingulate gray matter density in poor outcome bipolar illness. Psychiatry Res 2004; 130:153–159. Lyoo IK, Kim MJ, Stoll AL et al, Frontal lobe gray matter density decreases in bipolar I disorder. Biol Psychiatry 2004; 55:648–651. Drevets WC, Price JL, Simpson JR Jr et al, Subgenual prefrontal cortex abnormalities in mood disorders. Nature 1997; 386:824–827. Brambilla P, Nicoletti MA, Harenski K et al, Anatomical MRI study of subgenual prefrontal cortex in bipolar and unipolar subjects. Neuropsychopharmacology 2002; 27:792–799. Altshuler LL, Bartzokis G, Grieder T et al, An MRI study of temporal lobe structures in men with bipolar disorder or schizophrenia. Biol Psychiatry 2000; 48:147–162. Strakowski SM, DelBello MP, Sax KW et al, Brain magnetic resonance imaging of structural abnormalities in bipolar disorder. Arch Gen Psychiatry 1999; 56:254–260. Brambilla P, Harenski K, Nicoletti M et al, MRI investigation of temporal lobe structures in bipolar patients. J Psychiatr Res 2003; 37:287–295. Altshuler LL, Curran JG, Hauser P et al, T2 hyperintensities in bipolar disorder: magnetic resonance imaging comparison and literature meta-analysis. Am J Psychiatry 1995; 152:1139–1144. Moore PB, Shepherd DJ, Eccleston D et al, Cerebral white matter lesions in bipolar affective disorder: relationship to outcome. Br J Psychiatry 2001; 178:172–176. Breeze JL, Hesdorffer DC, Hong X et al, Clinical significance of brain white matter hyperintensities in young adults with psychiatric illness. Harvard Rev Psychiatry 2003; 11:269–283.
Ch 03
7/4/05
3:36 pm
Page 25
Brain abnormalities in bipolar disorder 16. 17. 18. 19. 20.
21. 22.
23.
24.
25
Videbech P, MRI findings in patients with affective disorder: a meta-analysis. Acta Psychiatr Scand 1997; 96:157–168. Stanley JA, In vivo magnetic resonance spectroscopy and its application to neuropsychiatric disorders. Can J Psychiatry 2002; 47:315–326. Winsberg ME, Sachs N, Tate DL et al, Decreased dorsolateral prefrontal Nacetyl aspartate in bipolar disorder. Biol Psychiatry 2000; 47:475–481. Chang KD, Adleman N, Dienes K et al, Decreased N-acetyl aspartate in children with familial bipolar disorder. Biol Psychiatry 2003; 53:1059–1065. Brambilla P, Harenski K, Nicoletti M et al, Differential effects of age on brain gray matter in bipolar patients and healthy individuals. Neuropsychobiology 2001; 43:242–247. Caetano SC, Olvera R, Hunter K et al, Abnormal amygdala volumes in pediatric bipolar disorder. Biol Psychiatry 2004; 55:111S. Blumberg HP, Kaufman J, Martin A et al, Amygdala and hippocampal volumes in adolescents and adults with bipolar disorder. Arch Gen Psychiatry 2003; 60:1201–1208. DelBello MP, Zimmerman ME, Mills NP et al, Magnetic resonance imaging analysis of amygdala and other subcortical brain regions in adolescents with bipolar disorder. Bipolar Disord 2004; 6:43–52. Caetano SC, Nicoletti MA, Hatch JP et al, Associations of age and length of illness with hippocampus and amygdala volumes in mood disorder patients. Biol Psychiatry 2004; 55:109S.
Ch 03
7/4/05
3:36 pm
Page 26
Ch 04
7/4/05
3:36 pm
Page 27
chapter 4
Structural magnetic resonance imaging studies in bipolar disorder: a meta-analysis Colm McDonald, Jolanta Zanelli, Robin M Murray and Noel Kennedy
Introduction Bipolar disorder is frequently described in reviews as being associated with subtle structural brain abnormalities, but it is also acknowledged that the existing literature is sparse.1,2 The most consistently reported brain abnormalities are increased rates of white matter hyperintensities and mild lateral ventricular enlargement, with studies often disagreeing on other regional volume deviations. This disagreement is contributed to by the small numbers of subjects studied, sample heterogeneity and a dependence on tests of significance with inconsistent ‘positive’ and ‘negative’ findings. Meta-analysis is a powerful technique for integrating quantitative data from several studies; the resultant increase in sample size provides more statistical power to detect subtle volume deviations and provide a more accurate estimate of the effect size. Previous meta-analyses have found that bipolar disorder is associated with preservation of cerebral size3 and increased rates of white matter hyperintensities.4,5 However, these prior meta-analyses were mostly based on computed tomography (CT) studies or qualitative ratings of magnetic resonance imaging (MRI) studies. In this chapter we present results of a meta-analysis of regional brain volume abnormalities of subjects with bipolar disorder based on studies using high-resolution MRI and complete coverage of each brain region.
Methodology A systematic search of electronic databases supplemented by manual searches was used to identify relevant studies. Studies were included if they
Ch 04
7/4/05
28
3:36 pm
Page 28
Bipolar disorder: the upswing in research & treatment
(1) were published in full in a peer-reviewed journal by December 2003; (2) compared a group of subjects with bipolar disorder and a normal comparison group; (3) reported means and standard deviations of regional volumetric brain measurements; and (4) reported on a brain structure that was evaluated by at least two other studies, and where data were available for at least 50 bipolar patients and controls in total. A rigorous standard of methodology was applied to ensure high precision, and studies were not included if they reported only area measurements, or if volumetric measurements were based on non-contiguous slices or incomplete coverage of the brain structure. As a readily interpretable measure of effect size, we used the ratio of the mean volume of the bipolar group divided by the mean volume of the control group. The meta-analysis was based on the logarithm of the ratio of the group means, which is more likely to be normally distributed. A metaanalysis of the logarithm of the volume ratios of all the measured structures was carried out using a random effects model. This assumes the ‘population’ of studies to have variable true effects that are normally distributed, and aims to estimate the overall mean of this distribution of effect sizes. The delta method was used to obtain variance of the logarithm of this effect size.6 For each brain region we also calculated I2 as a test of heterogeneity.7 This measure is an estimate of the percentage of total variation across studies that are due to true heterogeneity rather than chance, where larger values imply increasing heterogeneity. When five or more studies were included, the possibility of publication bias was investigated using the Egger test8 to test for the presence of a surplus of low-precision studies providing effect sizes of magnitude greater than the average. We adopted a significance level of p < 0.05 for all analyses. Further details on the methodology are described by McDonald et al.9
Results Twenty-six studies met all the criteria for inclusion in the meta-analysis and included 404 patients with bipolar disorder. The majority (77%) of studies had been published since 1998. Most studies used DSM-III-R or DSM-IV criteria for diagnosis, and most were of patients with recurrent episodes of illness rather than first-episode mania. In all of the studies that provided information on medication, the majority or all of the patients were medicated with mood stabilizers and/or other psychotropic medication at the time of scanning.
Ch 04
7/4/05
3:36 pm
Page 29
Structural magnetic resonance imaging studies in bipolar disorder
29
Table 4.1 demonstrates the comparison of regional brain volumes of bipolar disorder patients and controls for the structures included in the meta-analysis. Patients with bipolar disorder had significantly larger total lateral ventricular volume than controls. Ventricular volume was divided into right and left in all but one of these studies, and right lateral ventricle volume was found to be significantly greater in patients with bipolar disorder than in comparison subjects, whereas left lateral ventricular volume difference did not reach significance. A forest plot demonstrating the individual study results for right lateral ventricular volume is shown in Figure 4.1. There was no significant difference between bipolar patients and comparison subjects in third ventricular volume. No significant differences were seen in total brain volume or in the volume of any regional cortical, subcortical or limbic brain structures. No significant publication bias was detected in the studies reporting lateral ventricular volume enlargement (Figure 4.2). High heterogeneity was detected for several of the structures studied, especially the third ventricle, amygdala, left subgenual prefrontal cortex and left hippocampus–amygdala.
Discussion This meta-analysis of methodologically robust regional morphometric MRI studies demonstrates that bipolar disorder is indeed associated with lateral ventricular enlargement, especially on the right, but with no significant differences in the other structures examined. Elkis and colleagues10 previously performed a meta-analysis on studies using a broad category of affective disorders and reported an association with increased ventricular size which was less prominent than that found in schizophrenia. Most studies included in that meta-analysis were based upon CT data, and the majority of patients had unipolar depression. The present study is consistent with this previous meta-analysis but builds upon the findings by identifying increased ventricular volume in a more homogeneous sample of patients with bipolar disorder. The finding of more prominent right-sided lateral ventricular enlargement echoes evidence that bipolar disorder is more likely to be associated with right-sided cerebral pathology, for example when mania occurs in association with cerebral trauma, temporal lobe epilepsy or stroke.11 This meta-analysis also confirms that global cerebral volume is preserved in bipolar disorder, in contrast to schizophrenia, which is
Ch 04
7/4/05
30
3:36 pm
Page 30
Bipolar disorder: the upswing in research & treatment Table 4.1 Comparison of regional brain volumes of bipolar disorder subjects and controls from 26 studies (adapted from reference 9) Bipolar disorder/
%
p
Structure
control
Volume*
Value I2 (p value)
Heterogeneity Egger test
Whole brain (11 studies)
239/292
99 (96–101)
0.26
(p value)
44% (0.05) 0.06 (0.97)
Total lateral ventricles 160/168 (6 studies)
113 0.03 (101–127)
5% (0.39)
–0.37 (0.84)
Right lateral ventricle 143/152 (5 studies)
114 0.03 (101–128)
0% (0.52)
–2.10 (0.26)
Left lateral ventricle
108
0.32
26% (0.24)
0.07 (0.97)
0.16
67% (0.006) 0.38 (0.87)
143/152
(5 studies) Third ventricle
(93–124) 139/161
(6 studies)
113 (95–135)
Hippocampus (8 studies)
245/273
99 (96–101)
0.24
2% (0.43)
Amygdala (5 studies)
115/175
101 (91–112)
0.90
83% (0.001) –4.56 (0.54)
Left hippocampus–
158/247
102
0.76
91% (0.001) –0.37 (0.91)
0.85
58% (0.01) 1.86 (0.21)
amygdala (8 studies) Right hippocampus–
0.12 (0.90)
(92–112) 158/247
100
amygdala (8 studies)
(96–104)
Left subgenual 95/76 prefrontal (4 studies)
80 (52–123)
0.31
98% (0.001) —
Right subgenual 95/76 prefrontal (4 studies)
94 (82–108)
0.36
51% (0.10)
Left temporal lobe (6 studies)
178/215
101 (97–105)
0.62
52% (0.03) –1.93 (0.51)
Right temporal lobe (6 studies)
178/215
99 (96–102)
0.55
46% (0.06)
Thalamus (5 studies)
129/124
103 (98–109)
0.26
58% (0.05) 1.82 (0.60)
Caudate (7 studies)
172/187
102 (97–107)
0.42
35% (0.14)
0.41 (0.73)
Putamen (6 studies)
165/154
102 (99–106)
0.20
0% (0.95)
–1.15 (0.02)
Globus pallidus (3 studies)
76/74
105 (95–116)
0.38
61% (0.05)
—
—
–3.93 (0.20)
*Volume for subjects with bipolar disorder in relation to volume for control subjects; the brain volume of control subjects is assumed to be 100%. Significant p values in bold.
7/4/05
3:36 pm
Page 31
Structural magnetic resonance imaging studies in bipolar disorder
31
2 1.8
Effect size
1.6 1.4 1.2 1 0.8 0.6 0.4
Overall
Swayze 90
Swayze 90
Strak 99 Study
McInt
Bramb
01
01
Strak 02
Figure 4.1 Forest plot of meta-analysis results for right lateral ventricular volume.
0.5
Log effect
Ch 04
0
–0.5 00
0.1 0.2 Standard error of log effect size
0.3
Figure 4.2 Funnel plot of seven studies reporting total lateral ventricular volume results, demonstrating no publication bias.
characterized by a small decrease in brain size.12 This is consistent with a previous meta-analysis of seven studies of brain volume in bipolar disorder by Hoge et al,3 three of which were included in the present meta-analysis (the other four were CT studies or MRI studies from which complete volumetric measurements of the cerebrum could not be derived).
Ch 04
7/4/05
32
3:36 pm
Page 32
Bipolar disorder: the upswing in research & treatment
One of the most striking findings is the volume preservation in bipolar disorder of most regions assessed. This is in sharp contrast to schizophrenia, where a recent meta-analysis using similar methodology detected widespread volume deficits of up to 9% in some brain regions.12 For example, hippocampal volume was specifically measured in nearly 250 patients with bipolar disorder included in the current meta-analysis but did not differ at all from controls. Recent meta-analyses have demonstrated volume deficit of the hippocampus in both schizophrenia12 and major depressive disorder,13 and this study therefore provides powerful support for the existence of morphometric distinctions between the major diagnoses within the spectra of psychotic and affective disorders. The pathophysiology of hippocampal volume deficit in depression has been linked to hypercortisolaemia,13 and in schizophrenia to the impact of susceptibility genes,14 obstetric complications15 and glutamatergic excitotoxic damage.16 Either such factors do not impact upon hippocampal volume in bipolar disorder or other features associated with the illness or its treatment have a protective effect. Conventional antipsychotic medications are known to be associated with basal ganglia enlargement but recent evidence suggests that their use in patients experiencing their first episode of psychosis may be linked to volume deficit of other grey matter regions including the frontal and temporal cortex.17 Since these medications are more likely to be used chronically in schizophrenia but intermittently in bipolar disorder during manic relapse, it is possible that the consistent presence of grey matter deficit in schizophrenia and its relative absence in bipolar disorder is contributed to by differential treatment with conventional antipsychotics. Alternatively, other medications used to treat bipolar disorder may reverse or ameliorate primary structural changes associated with the illness, as indicated by evidence that lithium is neurotrophic and can increase grey matter volume in vivo.18,19 Given the apparent volume preservation of subcortical grey matter structures surrounding the ventricles in bipolar disorder, what is the likely corresponding tissue loss underlying the ventricular enlargement? One possibility is that white matter pathology with associated volume deficit is responsible for this. Increased rates of white matter hyperintensities are a consistent feature of bipolar disorder5 and emerging evidence using computational morphometry analysis of structural MRI scans points to extensive white matter volume deficits.20 Furthermore, gene expression profiling studies of frontal cortical tissue have identified specific down-regulation of genes related to myelination and oligodendrocyte function in bipolar disorder.21 These data suggest that a potentially critical component in the
Ch 04
7/4/05
3:36 pm
Page 33
Structural magnetic resonance imaging studies in bipolar disorder
33
pathophysiology of the illness is disturbed structural connectivity of distributed neural networks.
Heterogeneity Another principal finding is the very high levels of heterogeneity which were found for several structures, including the amygdala, third ventricle (which is largely bordered by the thalamus) and left subgenual prefrontal cortex. These structures have been a focus of interest in bipolar disorder research, since they form part of a limbic–thalamic–cortical circuit which, together with a limbic–striatal–pallidal–thalamic circuit, is thought to play a role in the pathophysiology of mood disorders.1 Functional imaging studies indicate that bipolar disorder is characterized by overactivity of limbic and subcortical structures and underactivity of the prefrontal cortex in response to emotionally salient stimuli, consistent with a hypothesis whereby impaired prefrontal regulation linked to excessive subcortical activity underlies mood dysregulation in bipolar disorder.22 A pattern of reduced subgenual prefrontal cortex volume and increased amygdala/thalamic volume would therefore be consistent with the functional imaging findings, if one presumes that functional activity is likely to be correlated with regional volume (which remains largely uninvestigated). Although some studies support this pattern of volume deviation, this meta-analysis demonstrates that the structural imaging literature is wholly inconsistent regarding these critical structures. Indeed, studies of the same structures included in the meta-analysis often demonstrate effects in opposing directions. Five studies were included which examined amygdala volume; two studies reported significant volume deficit and three studies reported significant volume enlargement of this structure, with no reports of volume equivalence. Quantitatively combining these studies as in the present meta-analysis unsurprisingly demonstrates that no conclusive effect has emerged and that very high heterogeneity exists. Probable sources of heterogeneity include methodological differences and clinical sample variation. Although patients fulfilled criteria for operationally defined bipolar disorder, there were considerable clinical variations within both the individual studies and the meta-analysis as a whole in chronicity, severity, phase of illness, types of treatment and co-morbid conditions such as substance abuse, all of which could potentially impact upon regional volume as measured by MRI. The meta-analysis therefore underlines the particular need in bipolar disorder research for future structural morphometry studies to capture more homogeneous clinical samples. Study designs likely to be more informative include those focusing on
Ch 04
7/4/05
34
3:36 pm
Page 34
Bipolar disorder: the upswing in research & treatment
(1) first-episode medication-naive patients; (2) patients during the euthymic phase of illness; (3) patients who experience psychotic symptoms during illness exacerbation; (4) subjects at high genetic risk of illness, such as firstdegree relatives and offspring; and (5) longitudinal studies which can examine the effects of age, duration of illness, differing mood-stabilizing medications and number of illness episodes on brain morphometry.
Conclusion Quantitatively combining the results of high-quality structural MRI studies of regional brain volume demonstrates that bipolar disorder is associated with lateral ventricular enlargement that is more prominent on the right, but with preservation of most of the other brain regions commonly chosen for measurement. Considerable heterogeneity exists in the results of studies to date and it remains unclear whether volumetric deviations exist in regions such as the amygdala, thalamus and subgenual prefrontal cortex in bipolar disorder, either as a whole or in subsamples defined by symptom profile, medication status, chronicity or severity. Research into the structural morphometry of bipolar disorder remains at an early stage of development. Considerable further research is warranted in this field before we can conclusively describe the brain structural deviations of this disorder.
Acknowledgement Colm McDonald is supported by the Wellcome Trust.
References 1. 2.
3. 4.
Soares JC, Mann JJ, The anatomy of mood disorders – review of structural neuroimaging studies. Biol Psychiatry 1997; 41:86–106. Strakowski SM, Adler CM, DelBello MP, Volumetric MRI studies of mood disorders: do they distinguish unipolar and bipolar disorder? Bipolar Disord 2002; 4:80–88. Hoge EA, Friedman L, Schulz SC, Meta-analysis of brain size in bipolar disorder. Schizophr Res 1999; 37:177–181. Altshuler LL, Curran JG, Hauser P et al, T2 hyperintensities in bipolar disorder: magnetic resonance imaging comparison and literature meta-analysis. Am J Psychiatry 1995; 152:1139–1144.
Ch 04
7/4/05
3:36 pm
Page 35
Structural magnetic resonance imaging studies in bipolar disorder 5. 6. 7. 8. 9.
10.
11. 12. 13.
14.
15.
16. 17.
18. 19. 20.
21. 22.
35
Videbech P, MRI findings in patients with affective disorder: a meta-analysis. Acta Psychiatr Scand 1997; 96:157–168. Dunn G, Design and Analysis of Reliability Studies: The Statistical Evaluation of Measurement Errors. Edward Arnold: London, 1989. Higgins JP, Thompson SG, Deeks JJ, Altman DG, Measuring inconsistency in meta-analyses. BMJ 2003; 327:557–560. Egger M, Davey Smith G, Schneider M, Minder C, Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315:629–634. McDonald C, Zanelli J, Rabe-Hesketh S et al, Meta-analysis of magnetic resonance imaging brain morphometry studies in bipolar disorder. Biol Psychiatry 2004; 56:411–417. Elkis H, Friedman L, Wise A, Meltzer HY, Meta-analyses of studies of ventricular enlargement and cortical sulcal prominence in mood disorders. Comparisons with controls or patients with schizophrenia. Arch Gen Psychiatry 1995; 52:735–746. Flor-Henry P, Lateralized temporal–limbic dysfunction and psychopathology. Ann NY Acad Sci 1976; 280:777–797. Wright IC, Rabe-Hesketh S, Woodruff PWR et al, Meta-analysis of regional brain volumes in schizophrenia. Am J Psychiatry 2000; 157:16–25. Campbell S, Marriott M, Nahmias C, MacQueen GM, Lower hippocampal volume in patients suffering from depression: a meta-analysis. Am J Psychiatry 2004; 161:598–607. Seidman LJ, Faraone SV, Goldstein JM et al, Left hippocampal volume as a vulnerability indicator for schizophrenia: a magnetic resonance imaging morphometric study of nonpsychotic first-degree relatives. Arch Gen Psychiatry 2002; 59:839–849. Schulze K, McDonald C, Frangou S et al, Hippocampal volume in familial and nonfamilial schizophrenic probands and their unaffected relatives. Biol Psychiatry 2003; 53:562–570. McCarley RW, Wible CG, Frumin M et al, MRI anatomy of schizophrenia. Biol Psychiatry 1999; 45:1099–1119. Dazzan P, Morgan K, Orr KG et al, Different effects of typical and atypical antipsychotics on grey matter in first episode psychosis: the AESOP study. Neuropsychopharmacology in press. Moore GJ, Bebchuk JM, Wilds IB et al, Lithium-induced increase in human brain grey matter. Lancet 2000; 356:1241–1242. Sassi RB, Nicoletti M, Brambilla P et al, Increased gray matter volume in lithium-treated bipolar disorder patients. Neurosci Lett 2002; 329:243–245. McDonald C, Bullmore E, Sham P et al, Regional volume deviations of brain structure in schizophrenia and psychotic bipolar disorder: a computational morphometry study. Br J Psychiatry in press. Tkachev D, Mimmack ML, Ryan MM et al, Oligodendrocyte dysfunction in schizophrenia and bipolar disorder. Lancet 2003; 362:798–805. Phillips ML, Drevets WC, Rauch SL, Lane R, Neurobiology of emotion perception II: Implications for major psychiatric disorders. Biol Psychiatry 2003; 54:515–528.
Ch 04
7/4/05
3:36 pm
Page 36
Ch 05
7/4/05
3:36 pm
Page 37
chapter 5
Are subcortical regions too expansive in bipolar disorder? An examination of the nature of prefrontal corticolimbic abnormalities in individuals with bipolar disorder Mary L Phillips
Introduction Bipolar disorder affects up to 1.5% of the population of the USA,1 with illness relapse rates estimated at between 37% and 44% per year,2,3 a total mortality elevated by 58% (predominantly from suicide and cardiovascular disease4), and syndromic recovery after 1 year following manic or mixed episodes only at 48%.5 Whilst it is clear clinically that mood dysregulation, or affective instability, is a key symptom of the disorder, the nature of the neural mechanism underlying this abnormality remains poorly understood. Clarification of this mechanism will be crucial for the future development of effective therapeutic interventions for this common but poorly treated disorder. In this chapter, the focus is therefore upon a discussion of findings from studies that have employed neuroimaging techniques to measure neural responses to emotionally salient stimuli in individuals with the disorder. It will conclude with a postulated neural mechanism for the affective instability in the disorder based on these findings.
Neural responses to emotionally salient stimuli in individuals with bipolar disorder The accurate recognition of facial expressions is crucial for successful interpersonal function in the social environment.6 Facial expressions represent
Ch 05
7/4/05
38
3:36 pm
Page 38
Bipolar disorder: the upswing in research & treatment
ecologically valid, emotionally salient stimuli. In healthy individuals, findings from neuroimaging studies have implicated a network of predominantly anterior limbic regions in the response to and appraisal of emotional stimuli. These regions include the amygdala, but also other areas: the ventral striatum, hippocampus and anterior insula.7,8–12 Dorsal and ventromedial prefrontal cortical regions have been implicated in the regulation of these responses, although further studies are required to clarify the neural mechanisms associated with the regulation of responses to facial expressions and other emotionally salient material.13 Previous studies of euthymic and remitted individuals with bipolar disorder indicate impaired fear14 and enhanced disgust recognition15 in facial expressions of adolescent individuals with bipolar disorder, a tendency to misinterpret the faces of peers as being angry,16 and of manic individuals with the disorder, both specific impairments in the recognition of fear and disgust of unfamiliar others,17 and generalized deficits in the recognition of all emotional expressions.18 Other studies employing tasks examining indirect biases in the identification of material as emotional or neutral, including the emotional Stroop task19 and the affective go/no-go task,20 have indicated negative attentional biases in depressed individuals with bipolar disorder,19 and both negative and positive attentional biases in manic individuals.19,20 These findings suggest deficits in the recognition of emotive stimuli, including negative facial expressions, in euthymic individuals with bipolar disorder and those in a mood episode. There has been limited examination in individuals with bipolar disorder of the neural mechanism underlying this abnormality in processing emotional stimuli. Instead, studies in individuals with the disorder have focused upon the examination of neural responses during rest and performance of executive and memory tasks. Findings from these studies suggest dysfunctional prefrontal cortical–subcortical interactions in euthymic, in addition to symptomatic, individuals with the disorder at rest and during performance of such tasks. Reports include predominant reductions in activity in dorsal and ventral prefrontal cortical regions,21–26 but increases in activity within the dorsal anterior cingulate gyrus27–29 and subcortical regions,28,30,31, which have been positively correlated with mania severity.26,28,29 Earlier studies provided conflicting findings regarding subcortical–temporal cortical activity in manic individuals at rest.32–34 Of the few studies directly comparing euthymic with mood episode individuals, some reports have indicated an amelioration of abnormal neural responses in euthymic individuals during executive task performance,22,24,25,28 although others have suggested greater impairments in prefrontal cortical
7/4/05
3:36 pm
Page 39
Are subcortical regions too expansive in bipolar disorder?
39
activity in euthymic compared with depressed individuals with bipolar disorder.26 Regarding neural responses to emotional stimuli, our recent findings in remitted individuals with bipolar disorder, using a facial expression paradigm, indicate increased activity within limbic and subcortical regions, predominantly to expressions of fear and happiness, in the absence of any deficits in facial expression recognition35 (Figure 5.1). These findings support earlier reports of increased subcortical (amygdala) activity to fearful expressions14 in remitted individuals. Our findings also indicate subsyndromal depression-related abnormalities, namely a positive correlation
B
L amygdala/ putamen
0.04
0
Neural response
A
Neural response
Ch 05
L amygdala/ ventrolateral PFC
0.04
0 CON
0.04
BD
MDD
ventromedial PFC
0
CON
BD
MDD
Figure 5.1 The figure depicts brain slices in healthy individuals (CON), individuals with bipolar disorder (BD) and individuals with major depressive disorder (MDD) in responses to facial expressions of happiness (A) and fear (B) contrasted with neutral faces. The graphs demonstrate that BD compared with CON and MDD demonstrated increased activity within the left amygdala (blue), ventral striatum (putamen) and ventromedial prefrontal cortex (PFC; pink) in response to happy faces, and increased activity within the left amygdala in response to fearful faces (blue).
Ch 05
7/4/05
40
3:36 pm
Page 40
Bipolar disorder: the upswing in research & treatment
between depression severity and hippocampal response to sad expressions, in remitted individuals with bipolar disorder. We also demonstrated increased ventromedial prefrontal cortical responses in these individuals, particularly in response to expressions of mild happiness.
Neural responses during mood induction in individuals with bipolar disorder Positive and negative mood states can be induced with specific moodinduction paradigms, including the use of facial expressions and/or autobiographical memory scripts, associated with activity within the ventral striatum and ventromedial prefrontal cortex in healthy individuals,36,37 and emotive scenes from a standardized series (e.g. the International Affective Picture Series, IAPS38), associated with subcortical limbic responses in healthy and anxiety-disordered populations.7,39 Few studies to date, however, have examined neural responses during mood induction paradigms in individuals with bipolar disorder. We have developed a mood induction paradigm involving autobiographical scripts to induce happy or sad mood, followed by presentation of emotion-congruent facial expressions.40 Using this paradigm, we have demonstrated in individuals with major depressive disorder during happy mood induction an absence of the normal increase in autonomic response (as measured by skin conductance recordings, SCR), and increased activity within the dorsomedial and ventromedial prefrontal cortex, regions associated with regulation of emotional responses.13 Other studies using similar mood induction paradigms have demonstrated relative decreases in activity within these regions during sad mood induction in euthymic and depressed individuals with bipolar disorder.41 One study, employing emotive scenes, has demonstrated in depressed individuals with bipolar disorder increased subcortical responses to positive and negative scenes during affect generation.42
Neural responses during performance of other emotion-processing tasks in individuals with bipolar disorder Interestingly, other studies employing an affective go/no-go paradigm, in which individuals respond to emotional target words (either happy or sad) and inhibit responses to emotional distractors (either happy or sad) have demonstrated in manic individuals with bipolar disorder decreased ventromedial prefrontal cortical responses during semantic task versus orthographic go/no-go task performance, but increased ventrolateral prefrontal cortical responses to emotional versus neutral targets, and elevated ventral
Ch 05
7/4/05
3:36 pm
Page 41
Are subcortical regions too expansive in bipolar disorder?
41
and medial prefrontal cortical responses to emotional distractors.43 Employing this paradigm, a similar pattern of increased response within the ventral anterior cingulate gyrus to sad targets, and increased ventrolateral prefrontal cortical response to sad distractors, was demonstrated in individuals with major depressive disorder.44
Summary Together, findings from studies examining neural responses during mood induction and to non-facial, emotionally salient stimuli (emotional scenes and words), indicate decreased activity in a dorsomedial prefrontal cortical region during sad mood induction, increased subcortical activity to emotional scenes, and increases in predominantly ventral regions of the prefrontal cortex to emotional word targets and distractors in individuals with bipolar disorder. This complex pattern of decreases and increases in prefrontal cortical and subcortical responses requires further study, but does include relative decreases in activity in regions associated wth mood regulation (e.g. ventromedial–dorsomedial prefrontal cortices), but relative increases in activity in regions assoicated with decision-making about and affective responses to emotional material (subcortical regions, ventral anterior cingulate gyrus, ventromedial–ventrolateral prefrontal cortices).
Structural neural abnormalities in individuals with bipolar disorder Findings to date in individuals with bipolar disorder regarding regions important for emotion processing, including the amygdala and hippocampus, have been variable, with studies reporting volume increases, decreases, or no abnormality within the amygdalae, and volume decreases or no abnormalities in the hippocampi.45–48 Other studies have reported increased ventral striatal (caudate nucleus and putamen) volumes,46,49 and decreased middle, superior and inferior, including subgenual cingulate, prefrontal cortical volumes,50–52 although yet others have reported no significant differences in prefrontal cortical volumes between individuals with bipolar disorder and healthy volunteers.53 Interestingly, a recent study has provided further evidence for increased grey matter volumes in the bilateral thalamus, insulae and cortical regions involved in the response to emotional stimuli and mood generation in individuals with bipolar disorder.54
Ch 05
7/4/05
42
3:36 pm
Page 42
Bipolar disorder: the upswing in research & treatment
The effect of previous illness history and medication on persistent abnormalities in these functional neural responses To date, previous studies have suggested that the magnitude of executive dysfunction within remitted individuals may be associated with longer illness duration and number of illness episode,55,56 particularly manic episodes,57,58 suggestive of a positive correlation between prefrontal cortical dysfunction and these clincal variables, whilst a history of psychosis has been associated with greater verbal memory impairment.57 Structural neuroimaging studies suggest that enlarged ventricular volumes and decreased putamen size59–61 are associated with an increased number of previous episodes of illness. There are discrepant findings regarding the effect of psychotropic medication upon neurocognitive function in individuals with bipolar disorder, however. Neuroleptic medication has been associated with attentional impairments in healthy volunteers,62 but also with no impairment in attention in individuals with psychiatric disorders.63 There are conflicting findings regarding the effect of lithium on cognitive function,64–68 but little effect of other mood stabilizers,69 or antidepressants,70 on cognitive function, whilst citalopram has been associated with a ‘normalization’ of the otherwise increased recognition of fear in euthymic individuals with a previous history of major depressive disorder.71 The effect of psychotropic medication upon structural and functional neuroanatomy in individuals with bipolar disorder is largely unknown. Long-term use of lithium has been associated with an increase in volume of the subgenual cingulate gyru.72,73 Neuroleptic medication levels have been positively correlated with mean regional cerebral blood flow at rest74 and prefrontal cortical activation during decision-making in manic individuals with bipolar disorder,27 whilst a relative reduction in subcortical activity has been demonstrated in manic and depressed individuals with bipolar disorder taking neuroleptic and mood-stabilizing medications compared with unmedicated individuals.30 Whilst findings in individuals with major depressive disorder, predominantly at rest, have indicated increased prefrontal cortical (dorsolateral and ventromedial prefrontal cortex) and decreased limbic, hipppocampal and subgenual cingulate gyral responses after successful treatment with medication,75–78 although a reversed pattern after successful cognitive behavioural79 and interpersonal therapy,80,81 it remains to be determined whether similar changes in neural response to emotionally salient stimuli occur in individuals with bipolar disorder after treatment.
Ch 05
7/4/05
3:36 pm
Page 43
Are subcortical regions too expansive in bipolar disorder?
43
The predictive value of these functional neural abnormalities for the clinical course Few studies have examined the extent to which neurocognitive function predicts clinical outcome in individuals with bipolar disorder. Previous studies have indicated that greater cognitive dysfunction per se,82,83 a history of psychotic symptoms associated with greater cognitive dysfunction84 and an increased number of white matter hyperintensities,85 may be associated with poorer clinical outcome in the disorder.
Conclusion Recent functional neuroimaging data indicate that a pattern of increased limbic (amygdalar and ventral striatal) responses to ecologically valid ‘probes’ of activity within neural systems important for emotion processing (facial expressions in others) is indeed present in remitted individuals with bipolar disorder, and therefore may exist as a persistent marker of the disorder.35 The relationship between this abnormal pattern of neural response to facial expressions and any deficits in facial expression identification per se remains unclear, however. Other data suggest increased subcortical activity during mood generation with positively and negatively valenced scenes in depressed individuals,42 and decreased medial prefrontal cortical activity during sad mood generation with autobiographical retrieval in euthymic and depressed individuals with bipolar disorder.41 The increased subcortical and limbic response to facial expressions and during positive and negative mood generation with emotive scenes, together with the decreased medial prefrontal cortical response during sad mood induction with autobiographical memory retrieval, suggests in individuals with bipolar disorder a reduced regulation by prefrontal cortical regions of subcortical responses to emotive stimuli, and during mood generation procedures. Findings also indicate structural volume increases in subcortical, and volume decreases in prefrontal cortical, regions. Taken together, these findings suggest a potential neural mechanism underlying the affective instability in bipolar disorder86 (Figure 5.2). Here, it has been hypothesized that the affective instability in the disorder may result from a combination of increased activity within subcortical and limbic regions implicated in the initial appraisal of emotive stimuli (amygdala ventral striatum, anterior insula), resulting in increased activity in regions
Ch 05
7/4/05
44
3:36 pm
Page 44
Bipolar disorder: the upswing in research & treatment
associated with mood generation and decisions about emotional material (ventromedial and ventrolateral prefrontal cortices, ventral anterior cingulate gyrus), and reduced activity within regions implicated in the regulation of these responses (dorsal and ventromedial prefrontal cortices). Future research employing emotion processing paradigms and neuroimaging techniques will help to clarify further the nature of the dysfunction in neural systems underlying mood regulation in the disorder.
A
B Regulation Dorsomedial PFC Dorsal ACG Ventromedial PFC
Mood Ventrolateral PFC Ventromedial PFC Ventral ACG
Appraisal Amygdala Insula Thalamus Ventral striatum
Stimulus presentation
Regulation Dorsomedial PFC Dorsal ACG Ventromedial PFC
Mood Ventrolateral PFC Ventromedial PFC Ventral ACG Appraisal Amygdala Insula Thalamus Ventral striatum Stimulus presentation
Figure 5.2 (A) Neural structures important for appraisal of emotionally salient information, mood generation and the regulation of emotional behaviour. A predominantly ventral system is important for the identification of emotional information and the generation of affect state (green), whilst a predominantly dorsal system is important for selective attention and regulation of behaviour responses to emotional stimuli (orange). The arrows (red) represent the reciprocal functional relationship that exists between these two distinct but parallel neural systems. PFC, prefrontal cortex; ACG, anterior cingulate gyrus. (B) Model for the neural basis of the affective instability in individuals with bipolar disorder. Enlarged rather than decreased amygdalar volumes, and increased amygdalar activity during performance of attentional tasks and in response to emotional stimuli, together with reduced prefrontal cortical volumes and reduced prefrontal metabolism during task performance and at rest, would be consistent with increased but dysfunctional identification of emotional stimuli and mood generation of emotional states, and an impaired regulation of the subsequent emotional behaviour (represented by the reduction in size of the red arrows).
Ch 05
7/4/05
3:36 pm
Page 45
Are subcortical regions too expansive in bipolar disorder?
45
References 1.
2.
3. 4. 5.
6. 7.
8.
9. 10.
11. 12. 13.
14. 15. 16.
17. 18.
Kessler RC, McGonagle KA, Zhao S et al, Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry 1994; 51:8–19. O’Connell RA, Mayo JA, Flatow L, Cuthbertson B, O’Brien BE, Outcome of bipolar disorder on long-term treatment with lithium. Br J Psychiatry 1991; 159:123–129. Gitlin MJ, Swendsen J, Heller TL, Hammen C, Relapse and impairment in bipolar disorder. Am J Psychiatry 1995; 152:1635–1640. Angst F, Stassen HH, Clayton PJ, Angst J, Mortality of patients with mood disorders: follow-up over 34–38 years. J Affect Disord 2002; 68:167–181. Keck PE Jr, McElroy SL, Strakowski SM et al, 12-month outcome of patients with bipolar disorder following hospitalization for a manic or mixed episode. Am J Psychiatry 1998; 155:646–652. Darwin C, The Expression of the Emotions in Man and Animals, 3rd edn. Harper Collins: London, 1872/1998. Mataix-Cols D, Cullen S, Lange K et al, Neural correlates of anxiety associated with obsessive–compulsive symptom dimensions in normal volunteers. Biol Psychiatry 2003; 53:482–493. Morris JS, Frith CD, Perrett DI et al, A differential neural response in the human amygdala to fearful and happy facial expressions. Nature 1996; 383:812–815. Phillips ML, Young AW, Senior C et al, A specific neural substrate for perceiving facial expressions of disgust. Nature 1997; 389:495–498. Sprengelmeyer R, Rausch M, Eysel UT, Przuntek H, Neural structures associated with recognition of facial expressions of basic emotions. Proc R Soc Lond B Biol Sci 1998; 265:1927–1931. Calder AJ, Lawrence AD, Young AW, Neuropsychology of fear and loathing. Nat Rev Neurosci 2001; 2:352–363. Surguladze SA, Brammer MJ, Young AW et al, A preferential increase in the extrastriate response to signals of danger. Neuroimage 2003; 19:1317–1328. Phillips ML, Drevets WC, Rauch SL, Lane R, Neurobiology of emotion perception I: The neural basis of normal emotion perception. Biol Psychiatry 2003; 54:504–514. Yurgelun-Todd DA, Gruber SA, Kanayama G et al, fMRI during affect discrimination in bipolar affective disorder. Bipolar Disord 2000; 2:237–248. Harmer CJ, Grayson L, Goodwin GM, Enhanced recognition of disgust in bipolar illness. Biol Psychiatry 2002; 51:298–304. McClure EB, Pope K, Hoberman AJ et al, Facial expression recognition in adolescents with mood and anxiety disorders. Am J Psychiatry 2003; 160:1172–1174. Lembke A, Ketter TA, Impaired recognition of facial emotion in mania. Am J Psychiatry 2002; 159:302–304. Getz GE, Shear PK, Strakowski SM, Facial affect recognition deficits in bipolar disorder. J Int Neuropsychol Soc 2003; 9:623–632.
Ch 05
7/4/05
46 19.
20. 21.
22.
23.
24. 25.
26.
27. 28. 29.
30.
31. 32.
33.
34.
35.
3:36 pm
Page 46
Bipolar disorder: the upswing in research & treatment Lyon HM, Startup M, Bentall RP, Social cognition and the manic defense: attributions, selective attention, and self-schema in bipolar affective disorder. J Abnorm Psychol 1999; 108:273–282. Murphy FC, Sahakian BJ, Rubinsztein JS et al, Emotional bias and inhibitory control processes in mania and depression. Psychol Med 1999; 29:1307–1321. Baxter LR Jr, Phelps ME, Mazziotta JC et al, Cerebral metabolic rates for glucose in mood disorders. Studies with positron emission tomography and fluorodeoxyglucose F 18. Arch Gen Psychiatry 1985; 42:441–447. Baxter LR Jr, Schwartz JM, Phelps ME et al, Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psychiatry 1989; 46:243–250. Ketter TA, Kimbrell TA, George MS et al, Effects of mood and subtype on cerebral glucose metabolism in treatment-resistant bipolar disorder. Biol Psychiatry 2001; 49:97–109. Martinot JL, Hardy P, Feline A et al, Left prefrontal glucose hypometabolism in the depressed state: a confirmation. Am J Psychiatry 1990; 147:1313–1317. Blumberg HP, Stern E, Ricketts S et al, Rostral and orbital prefrontal cortex dysfunction in the manic state of bipolar disorder. Am J Psychiatry 1999; 156:1986–1988. Blumberg HP, Leung HC, Skudlarski P et al, A functional magnetic resonance imaging study of bipolar disorder: state- and trait-related dysfunction in ventral prefrontal cortices. Arch Gen Psychiatry 2003; 60:601–609. Rubinsztein JS, Fletcher PC, Rogers RD et al, Decision-making in mania: a PET study. Brain 2001; 124:2550–2563. Blumberg HP, Stern E, Martinez D et al, Increased anterior cingulate and caudate activity in bipolar mania. Biol Psychiatry 2000; 48:1045–1052. Goodwin GM, Cavanagh JT, Glabus MF et al, Uptake of 99mTc-exametazime shown by single photon emission computed tomography before and after lithium withdrawal in bipolar patients: associations with mania. Br J Psychiatry 1997; 170:426–430. Caligiuri MP, Brown GG, Meloy MJ et al, An fMRI study of affective state and medication on cortical and subcortical brain regions during motor performance in bipolar disorder. Psychiatry Res 2003; 123:171–182. Berns GS, Martin M, Proper SM, Limbic hyperreactivity in bipolar II disorder. Am J Psychiatry 2002; 159:304–306. al Mousawi AH, Evans N, Ebmeier KP et al, Limbic dysfunction in schizophrenia and mania. A study using 18F-labelled fluorodeoxyglucose and positron emission tomography. Br J Psychiatry 1996; 169:509–516. O’Connell RA, Van Heertum RL, Luck D et al, Single-photon emission computed tomography of the brain in acute mania and schizophrenia. J Neuroimaging 1995; 5:101–104. Gyulai L, Alavi A, Broich K et al, I-123 iofetamine single-photon computed emission tomography in rapid cycling bipolar disorder: a clinical study. Biol Psychiatry 1997; 41:152–161. Lawrence NS, Williams AM, Surguladze S et al, Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry 2004; 55:578–587.
Ch 05
7/4/05
3:36 pm
Page 47
Are subcortical regions too expansive in bipolar disorder? 36.
37. 38.
39.
40. 41.
42. 43.
44.
45. 46.
47.
48.
49.
50.
51.
52.
47
Mayberg HS, Liotti M, Brannan SK et al, Reciprocal limbic–cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999; 156:675–682. Breiter HC, Gollub RL, Weisskoff RM et al, Acute effects of cocaine on human brain activity and emotion. Neuron 1997; 19:591–611. Lang PJ, Bradley MM, Cuthbert BN, International Affective Picture System (IAPS). NIMH Center for the Study of Emotion and Attention: University of Florida, Gainesville, 1997. Phillips ML, Senior C, David AS, Perception of threat in schizophrenics with persecutory delusions: an investigation using visual scan paths. Psychol Med 2000; 30:157–167. Keedwell PA, Andrew C, Williams SCR et al, The neural correlates of depression. Biol Psychiatry 2003; 53:1S–217S. Kruger S, Seminowicz D, Goldapple K et al, State and trait influences on mood regulation in bipolar disorder: blood flow differences with an acute mood challenge. Biol Psychiatry 2003; 54:1274–1283. Mahli GS, Lagopoulos J, Ward PB et al, Cognitive generation of affect in bipolar depression: an fMRI study. Eur J Neurosci 2004; 19:741–754. Elliott R, Ogilvie A, Rubinsztein JS et al, Abnormal ventral frontal response during performance of an affective go/no go task in patients with mania. Biol Psychiatry 2004; 55:1163–1170. Elliott R, Rubinsztein JS, Sahakian BJ, Dolan RJ, The neural basis of moodcongruent processing biases in depression. Arch Gen Psychiatry 2002; 59:597–604. Brambilla P, Harenski K, Nicoletti M et al, MRI investigation of temporal lobe structures in bipolar patients. J Psychiatr Res 2003; 37:287–295. Strakowski SM, DelBello MP, Sax KW et al, Brain magnetic resonance imaging of structural abnormalities in bipolar disorder. Arch Gen Psychiatry 1999; 56:254–260. Altshuler LL, Bartzokis G, Grieder T et al, Amygdala enlargement in bipolar disorder and hippocampal reduction in schizophrenia: an MRI study demonstrating neuroanatomic specificity. Arch Gen Psychiatry 1998; 55:663–664. Blumberg HP, Martin A, Kaufman J et al, Frontostriatal abnormalities in adolescents with bipolar disorder: preliminary observations from functional MRI. Am J Psychiatry 2003; 160:1345–1347. Aylward EH, Roberts-Twillie JV, Barta PE et al, Basal ganglia volumes and white matter hyperintensities in patients with bipolar disorder. Am J Psychiatry 1994; 151:687–693. Sax KW, Strakowski SM, Zimmerman ME et al, Frontosubcortical neuroanatomy and the continuous performance test in mania. Am J Psychiatry 1999; 156:139–141. Lopez-Larson MP, DelBello MP, Zimmerman ME et al, Regional prefrontal gray and white matter abnormalities in bipolar disorder. Biol Psychiatry 2002; 52:93–100. Sharma V, Menon R, Carr TJ et al, An MRI study of subgenual prefrontal cortex in patients with familial and non-familial bipolar I disorder. J Affect Disord 2003; 77:167–171.
Ch 05
7/4/05
48 53.
54.
55. 56.
57.
58. 59.
60.
61.
62.
63. 64. 65. 66. 67. 68. 69. 70. 71.
3:36 pm
Page 48
Bipolar disorder: the upswing in research & treatment Brambilla P, Nicoletti MA, Harenski K et al, Anatomical MRI study of subgenual prefrontal cortex in bipolar and unipolar subjects. Neuropsychopharmacology 2002; 27:792–799. Lochhead RA, Parsey RV, Oquendo MA, Mann JJ, Regional brain gray matter volume differences in patients with bipolar disorder as assessed by optimized voxel-based morphometry. Biol Psychiatry 2004; 55:1154–1162. Clark L, Iversen SD, Goodwin GM, Sustained attention deficit in bipolar disorder. Br J Psychiatry 2002; 180:313–319. Cavanagh JT, Van Beck M, Muir W, Blackwood DH, Case–control study of neurocognitive function in euthymic patients with bipolar disorder: an association with mania. Br J Psychiatry 2002; 180:320–326. Martinez-Aran A, Vieta E, Reinares M et al, Cognitive function across manic or hypomanic, depressed, and euthymic states in bipolar disorder. Am J Psychiatry 2004; 161:262–270. Zubieta JK, Huguelet P, O’Neil RL, Giordani BJ, Cognitive function in euthymic bipolar I disorder. Psychiatry Res 2001; 102:9–20. Ali SO, Denicoff KD, Altshuler LL et al, Relationship between prior course of illness and neuroanatomic structures in bipolar disorder: a preliminary study. Neuropsychiatry Neuropsychol Behav Neurol 2001; 14:227–232. Brambilla P, Harenski K, Nicoletti M et al, Differential effects of age on brain gray matter in bipolar patients and healthy individuals. Neuropsychobiology 2001; 43:242–247. Strakowski SM, Adler CM, DelBello MP, Volumetric MRI studies of mood disorders: do they distinguish unipolar and bipolar disorder? Bipolar Disord 2002; 4:80–88. Mehta MA, Sahakian BJ, McKenna PJ, Robbins TW, Systemic sulpiride in young adult volunteers simulates the profile of cognitive deficits in Parkinson’s disease. Psychopharmacology (Berl) 1999; 146:162–174. King DJ, Psychomotor impairment and cognitive disturbances induced by neuroleptics. Acta Psychiatr Scand Suppl 1994; 380:53–58. Kocsis JH, Shaw ED, Stokes PE et al, Neuropsychologic effects of lithium discontinuation. J Clin Psychopharmacol 1993; 13:268–275. Ananth J, Gold J, Ghadirian AM, Long-term effects of lithium carbonate on cognitive functions. J Psychiatr Eval Treat 1981; 3:551–555. Engelsmann F, Katz J, Ghadirian AM, Schachter D, Lithium and memory: a long-term follow-up study. J Clin Psychopharmacol 1988; 8:207–212. Ferrier IN, Stanton BR, Kelly TP, Scott J, Neuropsychological function in euthymic patients with bipolar disorder. Br J Psychiatry 1999; 175:246–251. Kessing LV, Cognitive impairment in the euthymic phase of affective disorder. Psychol Med 1998; 28:1027–1038. Devinsky O, Cognitive and behavioral effects of antiepileptic drugs. Epilepsia 1995; 36(Suppl 2):S46–S65. Thompson PJ, Annual Review of Human Psychopharmacology 1991; 6:79–90 Bhagwagar Z, Cowen PJ, Goodwin GM, Harmer CJ, Normalization of enhanced fear recognition by acute SSRI treatment in subjects with a previous history of depression. Am J Psychiatry 2004; 161:166–168.
Ch 05
7/4/05
3:36 pm
Page 49
Are subcortical regions too expansive in bipolar disorder? 72.
73. 74. 75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
49
Manji HK, Moore GJ, Chen G, Clinical and preclinical evidence for the neurotrophic effects of mood stabilizers: implications for the pathophysiology and treatment of manic-depressive illness. Biol Psychiatry 2000; 48:740–754. Harrison PJ, The neuropathology of primary mood disorder. Brain 2002; 125:1428–1449. Silfverskiold P, Risberg J, Regional cerebral blood flow in depression and mania. Arch Gen Psychiatry 1989; 46:253–259. Mayberg HS, Brannan SK, Tekell JL et al, Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry 2000; 48:830–843. Bench CJ, Friston KJ, Brown RG et al, Regional cerebral blood flow in depression measured by positron emission tomography: the relationship with clinical dimensions. Psychol Med 1993; 23:579–590. Kennedy SH, Evans KR, Kruger S et al, Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. Am J Psychiatry 2001; 158:899–905. Goodwin GM, Austin MP, Dougall N et al, State changes in brain activity shown by the uptake of 99mTc-exametazime with single photon emission tomography in major depression before and after treatment. J Affect Disord 1993; 29:243–253. Goldapple K, Segal Z, Garson C et al, Modulation of cortical–limbic pathways in major depression: treatment-specific effects of cognitive behavior therapy. Arch Gen Psychiatry 2004; 61:34–41. Brody AL, Saxena S, Stoessel P et al, Regional brain metabolic changes in patients with major depression treated with either paroxetine or interpersonal therapy: preliminary findings. Arch Gen Psychiatry 2001; 58:631–640. Martin SD, Martin E, Rai SS et al, Brain blood flow changes in depressed patients treated with interpersonal psychotherapy or venlafaxine hydrochloride: preliminary findings. Arch Gen Psychiatry 2001; 58:641–648. Martinez-Aran A, Penades R, Vieta E et al, Executive function in patients with remitted bipolar disorder and schizophrenia and its relationship with functional outcome. Psychother Psychosom 2002; 71:39–46. Martinez-Aran A, Vieta E, Colom F et al, Cognitive impairment in euthymic bipolar patients: implications for clinical and functional outcome. Bipolar Disord 2004; 6:224–232. Tohen M, Hennen J, Zarate CM Jr et al, Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry 2000; 157:220–228. Moore PB, Shepherd DJ, Eccleston D et al, Cerebral white matter lesions in bipolar affective disorder: relationship to outcome. Br J Psychiatry 2001; 178:172–176. Phillips ML, Drevets WC, Rauch SL, Lane R, Neurobiology of emotion perception II: Implications for major psychiatric disorders. Biol Psychiatry 2003; 54:515–528.
Ch 05
7/4/05
3:36 pm
Page 50
Ch 06
7/4/05
3:37 pm
Page 51
chapter 6
The Maudsley Bipolar Disorder Project: insights into pathophysiology Sophia Frangou
The structural neuroanatomy of bipolar disorder Neuroanatomical changes in bipolar disorder are mostly regional and involve the prefrontal cortex (PFC), the anterior cingulate and the amygdala. Reductions in grey matter volume (or density) have been found in the dorsal PFC and they appear to be more pronounced on the left.1–3 Similarly, reduction has been reported in the ventral PFC particularly in Brodmann areas 44 and 47.1,3 Volume (or density) decreases in the left cingulate gyrus have been reported in some3,4 but not all studies.1,5 In contrast, the volume of the amygdala has appeared to be enlarged bilaterally or only on the left in recent studies.1,5–7
The functional neuroanatomy of bipolar disorder Resting-state functional imaging studies, despite differences in methodological approaches, have found reduced activity in the dorsal PFC, mostly on the left8 and increased activity in the amygdala in depressive states.9 Manic states have been associated with decreased activity in the ventral PFC and increased activity in the anterior cingulate cortex (ACC).10,11 Trait-related decreases in brain activation have also been reported within the left ventral PFC (Brodmann areas 47 and 10).12 However, the relative contributions of dorsal and ventral prefrontal functioning in bipolar disorder remain unclear.
Ch 06
7/4/05
52
3:37 pm
Page 52
Bipolar disorder: the upswing in research & treatment
The Maudsley Bipolar Disorder Project The Maudsley Bipolar Disorder Project started in 1999 and is ongoing. It has a case–control design and comprises interconnected modules focusing on cognition, and structural and functional neuroanatomy.13–15 Diagnosis in patients and its absence in controls was established using the Structured Clinical Interview for DSM-IV Axis I Disorders; symptomatology was assessed using the 31-item Hamilton Depression Rating Scale (HAMD), and the Mania Rating Scale (MRS) from the Schedule for Affective Disorders and Schizophrenia – Change Version. All participants had detailed cognitive testing that assessed the domains of general intelligence (IQ), memory and executive function. Here we present data from the functional imaging component of the study. For this component, 14 participants were selected from the pool of BDI patients and matched controls form the Maudsley Bipolar Disorder Project. Patients were selected on the following criteria: (1) remitted clinical status; (2) monotherapy with mood stabilizer; and (3) test performance less than 0.5 standard deviation below the control mean for all tests. Only eight patients fulfilled these requirements. Seven agreed to participate and were individually matched to an equal number of controls on age, gender, years of education and IQ. The mean age of the seven participating patients (five women and two men) was 37 ± 5.88 years and mean years of education were 12.43 ± 2.64. Their mean age of onset was 21.6 ± 6.5 years and they had experienced an average of 9 ± 2.6 episodes. On the day of their functional magnetic resonance imaging (fMRI) assessment, patients’ mean scores on the HAMD and MRS were 5.14 ± 0.29 and 0.14 ± 3.37, respectively. They were on treatment with lithium (n = 4; dose range 600–1000 mg/day) and sodium valproate (n = 3; dose range 750–1000 mg/day). Their mean full scale IQ was 102.57 ± 16.21. The mean age of controls was 39 ± 5.88; years of education were 13.43 ± 3.26, and mean full scale IQ was 104.71 ± 15.88. All participants were right handed. All participants underwent fMRI whilst performing cognitive paradigms selected on their relative specificity in engaging dorsal and ventral PFC. The N-back letter-sequencing task, a verbal working memory task, has been demonstrated to engage a wide network of brain regions with dorsal PFC activation being the most consistent finding across different experimental manipulations.16 The Iowa Gambling Task, initially developed to test decision-making in individuals with PFC lesions,17 has been shown to engage the bilateral ventral prefrontal regions.18 Gradient-echo echoplanar magnetic resonance images were acquired by using a 1.5T GE
Ch 06
7/4/05
3:37 pm
Page 53
The Maudsley Bipolar Disorder Project
53
Neurovascular Signa MR system (General Electric, Milwaukee, WI, USA) fitted with 40 mT/m high-speed gradients and were analysed on a SPARC Ultra 10 workstation (Sun Microsystems, Palo Alto, CA, USA) using MATLAB (version 5.3, The Mathworks Inc, Natick, MA, USA) and SPM99 software (Statistical Parametric Mapping, The Wellcome Department of Cognitive Neurology, London; http://www.fil.ion.ucl.ac.uk/spm).
N-back task During the N-back task, in the ‘control’ or 0-back condition, participants responded by button press when a designated target letter appeared (letter ‘X’). In the three active conditions, the 1-back, 2-back and 3-back, the target letter was defined as any letter that was identical to the one presented in the preceding 1, 2, or 3 trials, respectively. All stimuli were visually presented to subjects by means of a prismatic mirror as they lay in the scanner and responses were monitored. A series of 13 letters was presented with an interstimulus interval of 2.3 seconds during each epoch. There were 18 epochs in all, each lasting 30 seconds with the total experiment time of 9 minutes. All conditions were matched for number of target letters presented. The order of the tasks was pseudorandomized to avoid any systematic order effects. Reaction time to target letters and accuracy were recorded. Patients and controls did not differ in reaction time or accuracy (Figure 6.1). During the active conditions (1-, 2-, 3-back tasks minus the 0-back task) no significant differences were seen in patients and controls in the pattern and degree of brain regions activated. When we examined the effect of memory load, in the controls increased activation was seen bilaterally to the superior (BA 6) and middle frontal gyrus (BA 9,46) and the anterior cingulate gyrus (BA 32) as well as the right superior parietal lobule (BA 7). In bipolar disorder patients the effect of increasing memory load was localized to the left superior parietal lobule (BA 7) and the right middle frontal gyrus (BA 10). Therefore, it seems that, in bipolar disorder, dorsal PFC dysfunction is subtle and only becomes apparent with increasing mental load.
Gambling task During the Decision-Making Task, subjects were instructed to select from 96 cards arranged in four piles (decks) in order to win ‘pretend’ money. Unknown to the subjects, deck A had frequent, small magnitude punishments, deck B had infrequent, but higher punishments, deck C had
Ch 06
7/4/05
3:37 pm
54
Page 54
Bipolar disorder: the upswing in research & treatment
frequent, small rewards, and deck D had infrequent, higher rewards. Subjects were asked to select a card from a deck of their choice, once every 5 seconds, and the win or loss associated with their choice appeared visually on the screen. Each ‘active’ condition lasted 1 minute and was alternated eight times with the ‘control’ condition, when subjects were required to select cards as before, but were informed that no money would be won or lost during this epoch; total experiment time was 16 minutes. Performance measures were the net global outcome score (net score) calculated by subtracting the total number of cards selected from the disadvantage decks (A+B) from the total number of cards selected from the advantage decks (C+D). In controls activation associated with incentive decision making was evident in the left superior frontal gyrus (BA 6), the bilateral middle (BA 8,9,46) and inferior (BA 44,45) frontal gyri, and the right superior parietal lobule (BA 7). Patients demonstrated activation of the left superior (BA 7) and inferior (BA 40) parietal lobules and the right superior frontal
A
B
Figure 6.1 Cerebral activation in controls (A) and patients (B) during the N-back task.
Ch 06
7/4/05
3:37 pm
Page 55
The Maudsley Bipolar Disorder Project
55
gyrus (BA 10). Compared with controls, patients showed significantly less activation in the frontal cortices (Figure 6.2). Human lesion studies suggest that the interface between the dorsal and ventral PFC is crucial for incentive decision making.19 Studies in primates have indicated that reward-related dorsal PFC modulation is probably driven by brain areas within the ventral PFC which are primarily involved with processing and representing incentive information. This hypothesis is supported by electrophysiological measurements in primates. Wallis and Miller20 recorded neuronal activity from the dorsal and ventral PFC of rhesus monkeys while choosing between pictures associated with different amounts of a juice reward. They found that, whilst both dorsal and ventral PFC were involved in encoding the incentive value (reward) of the different stimuli (pictures), neural activation peaked earlier in the ventral PFC, suggesting that this region generated incentive information that enters the dorsal PFC where it is used to influence behavioural response. According to this model, deficits in the ventral PFC should attenuate reward information
A
B
Figure 6.2 Cerebral activation in controls (A) and patients (B) during the Gambling Task.
Ch 06
7/4/05
56
3:37 pm
Page 56
Bipolar disorder: the upswing in research & treatment
in the dorsal PFC. The findings of the Gambling Task in the bipolar disorder sample in our study are consistent with this model. In summary, this study provides evidence of PFC dysfunction as a trait deficit in bipolar disorder that is not predicted by task performance. It also suggests that dorsal PFC dysfunction may be more subtle than ventral, and that the greatest deficit may be seen in tasks crucially dependent on the interaction between dorsal and ventral PFC.
References 1.
2.
3. 4. 5.
6.
7.
8.
9.
10.
11. 12.
Frangou S, Hadjulis M, Chitnis X et al, The Maudsley Bipolar Disorder Project: brain structural changes in bipolar 1 disorder. Bipolar Disord 2002; 4:123–124. Lopez-Larson MP, DelBello MP, Zimmerman ME et al, Regional prefrontal gray and white matter abnormalities in bipolar disorder. Biol Psychiatry 2002; 52:93–100. Lyoo IK, Kim MJ, Stoll AL et al, Frontal lobe gray matter density decreases in bipolar I disorder. Biol Psychiatry 2004; 55:648–51. Hirayasu Y, Shenton ME, Salisbury DF et al, Subgenual cingulate cortex volume in first-episode psychosis. Am J Psychiatry 1999; 156:1091–1093. Brambilla P, Nicoletti MA, Harenski K et al, Anatomical MRI study of subgenual prefrontal cortex in bipolar and unipolar subjects. Neuropsychopharmacology 2002; 27:792–799. Strakowski SM, DelBello MP, Sax KW et al, Brain magnetic resonance imaging of structural abnormalities in bipolar disorder. Arch Gen Psychiatry 1999; 56:254–260. Altshuler LL, Bartzokis G, Grieder T et al, An MRI study of temporal lobe structures in men with bipolar disorder or schizophrenia. Biol Psychiatry 2000; 48:147–162. Baxter LR, Schwartz JM, Phelps ME et al, Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psychiatry 1989; 46:243–250. Drevets WC, Price JL, Bardgett ME et al, Glucose metabolism in the amygdala in depression: relationship to diagnostic subtype and plasma cortisol levels. Pharmacol Biochem Behav 2002; 71:431–47. Blumberg HP, Stern E, Ricketts S et al, Rostral and orbital prefrontal cortex dysfunction in the manic state of bipolar disorder. Am J Psychiatry 1999; 156: 1986–1988. Blumberg HP, Stern E, Martinez D et al, Increased anterior cingulate and caudate activity in bipolar mania. Biol Psychiatry 2000; 48:1045–52. Blumberg HP, Leung HC, Skudlarski P et al, A functional magnetic resonance imaging study of bipolar disorder: state- and trait-related dysfunction in ventral prefrontal cortices. Arch Gen Psychiatry 2003; 60:601–9.
Ch 06
7/4/05
3:37 pm
Page 57
The Maudsley Bipolar Disorder Project 13.
14.
15.
16. 17.
18. 19.
20.
57
Frangou S, Raymont V, Bettany D, The Maudsley Bipolar Disorder Project. A pharmaco-epidemiological survey of prescribing patterns in bipolar I disorder. Bipolar Disord 2002; 4:378–385. Donaldson S, Goldstein LH, Landau S et al, The Maudsley Bipolar Disorder Project: The effect of medication, family history, and duration of illness on IQ and memory in bipolar 1 disorder. J Clin Psychiatr 2003; 64:86–93. Raymont V, Bettany D, Frangou S, The Maudsley Bipolar Disorder Project. Clinical characteristics of bipolar disorder I in a catchment area derived treatment sample. Eur Psychiatry 2003; 18:13–17. Fletcher PC, Henson RN, Frontal lobes and human memory: insights from functional neuroimaging. Brain 2001; 124:849–881. Bechara A, Damasio AR, Damasio H, Anderson SW, Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 1994; 50:7–15. Ernst M, Bolla K, Mouratidis M et al, Decision-making in a risk-taking task: A PET study. Neuropsychopharmacology 2002; 26:682–691. Bechara A, Damasio H, Tranel D, Anderson SW, Dissociation of working memory from decision making within the human prefrontal cortex. J Neurosci 1998; 18:428–37. Wallis JD, Miller EK, Neuronal activity in primate dorsolateral and orbital prefrontal cortex during performance of a reward preference task. Eur J Neurosci 2003; 18:2069–2081.
Ch 06
7/4/05
3:37 pm
Page 58
Ch 07
7/4/05
3:43 pm
Page 59
chapter 7
Is any of this real? The word from the grave Paul J Harrison
Introduction As with all messages from the grave, we have to be sceptical when discussing the neuropathology of bipolar disorder. So far, there have been some intriguing whispers, but they remain difficult to hear, and we still cannot be sure how best to decipher them. In this chapter, I cover the main findings from postmortem studies of bipolar disorder, and try to link them with some other themes. I have not attempted to provide a systematic overview of the subject; for this see two recent reviews.1,2 At the start it is important to point out that neuropathological studies of bipolar disorder only began a few years ago. It is no exaggeration to say there were no data worth discussing until a study was published in 1998.3 The reason why the field is so recent is largely a practical one: people simply did not collect enough brains from patients who had suffered from bipolar disorder, with adequate characterization and appropriate controls,4 to carry out any kind of quantitative, or molecular, studies. It is thanks to the Stanley Foundation (now the Stanley Medical Research Institute) that the situation changed in the mid-1990s, because of their funding of an autopsy series of brains from patients (15 in each group) with bipolar disorder as well as schizophrenia, major depression and control subjects.5 This series was made available to investigators across the world. Part of the deal was that one had to study all 60 subjects in all experiments (the material was coded), and so people began to study bipolar disorder. A significant proportion of the world literature on the neuropathology of bipolar disorder now comes from this series. The gene expression component of this work has been meta-analysed on a region-by-region basis.6,7
Ch 07
7/4/05
60
3:43 pm
Page 60
Bipolar disorder: the upswing in research & treatment
A few other practical issues are also worth noting. First, the concepts, methods, targets and interpretation of the research have been much influenced by the experience with schizophrenia which had emerged over the preceding 10–15 years, and by the fact that it is the latter disorder that is often of primary interest. The majority of papers therefore discuss bipolar disorder findings in terms of similarities (of which there are many) and differences (of which there are relatively few) with those of schizophrenia. At the same time, however, comparison with other mood disorders cannot be neglected, and this is emphasized in some publications. Thus, for example, some papers concentrate on bipolar versus unipolar mood disorder, or emphasize the importance of familiality not polarity.3 These different ways of cutting the cake (or comparing two cakes) makes the literature somewhat confusing. The second point is that neuropathologists have to pick a particular part of the brain to study: the brain is too big (and too structurally complex even within a single area) to examine the whole thing. Because one has to start somewhere, research has naturally focused on where researchers think the positive findings are most likely to be. (As the man said when asked why he robbed banks: ‘they’re where the money is’.) The most relevant clues have been those from brain imaging, and hence many of the neuropathological data are in the anterior cingulate cortex and orbitofrontal cortex, because of the many data implicating these regions (Figure 7.1). The best example of this kind is the combined study of magnetic resonance imaging (MRI) and positron emission tomography (PET) showing smaller volumes and reduced metabolic activity in mood disorder in the ‘subgenual’ part of the cingulate cortex,8 which led the same group to carry out the landmark postmortem study mentioned.3 Whilst this ‘candidate region’ approach may have been inevitable, it does of course lead to a circularity in the argument – it means we do not know where the pathology of bipolar disorder might be centred because people have not looked, to anything like the same extent, in many other areas of the brain – compare the cerebellum in schizophrenia: hardly an area which many people implicated, nor were interested in, a decade ago, yet where there is now good evidence for structural and functional involvement.
Neuropathology of bipolar disorder: findings and interpretations There are three basic kinds of neuropathology reported in bipolar disorder (Table 7.1). The first observation, as reported by Öngür and colleagues,3
Ch 07
7/4/05
3:43 pm
Page 61
Is any of this real? The word from the grave
A
B
32′ 24′ 24
9 10
61
46
33
32
25
12 11
47 C
24c 24b 24a sg24
Figure 7.1 Areas of the cerebral cortex implicated in bipolar disorder by neuropathological studies. (A) Lateral surface of the left hemisphere; (B) medial surface of the right hemisphere; (C) coronal section through the left hemisphere at the level of the dashed line in (A) and (B). The numbers refer to Brodmann areas. The subgenual region of the anterior cingulate cortex is sg24. For detailed discussion see reference 2.
was unexpected: there were fewer glial cells in the subgenual cingulate cortex. The decrease was also seen in major depression subjects with a family history. The finding has been replicated, albeit equivocally, in subsequent studies of the cingulate cortex and other prefrontal areas.2,9,10 The second kind of abnormality affects neurons, in the same areas and in the hippocampus, which may be slightly fewer in number, or smaller.2,10,11 The third aspect is ‘synaptic pathology’, where one uses molecular markers of synaptic terminals as one way of getting at the connectivity between the neurons. Three studies have shown a reduction in the expression of synaptic proteins, two in the anterior cingulate cortex and one in the hippocampus.2,12 So these are the elements of pathology of which we need to consider the significance. Before doing so, I should again emphasize that these are very preliminary studies, often on small numbers of subjects, or the same subjects. The studies face the various difficulties of postmortem brain studies, meaning that one has always to be alert to confounders of different kinds,
Ch 07
7/4/05
62
3:43 pm
Page 62
Bipolar disorder: the upswing in research & treatment Table 7.1 Main themes in bipolar disorder neuropathology Strength of evidence Brain areas involved Anterior cingulate cortex
++++
Orbitofrontal cortex
+++
Dorsolateral prefrontal cortex
+++
Hippocampus
++
Amygdala
+
Main features Reduced number or density of glia
+++
Smaller neurons
++
Decreased synaptic markers
++
Reduced neuronal density
++
Clinical correlations Independent of mood state
++
May be related to duration of illness
+
May be ameliorated by mood stabilizers
+
Overlaps with schizophrenia
++
Overlaps with major depression
++
for example those related to the cause of death or autopsy delay.4 Though surmountable, these can jeopardize the robustness of the results, given the subtlety of the differences being investigated. Nevertheless, for the rest of the chapter I shall assume that the above findings are basically true, and discuss what they may mean. The first question concerns the clinicopathological correlations. It has been suggested that grey matter volumes may change with the duration of the illness. We also know that bipolar disorder runs an episodic course. Therefore, the neuropathology could be the pathology of having had bipolar disorder and its sequelae – most of the subjects died after having many years of illness, so we are not looking at first episode or unmedicated subjects. We also need to bear in mind that patients die at different phases of the illness: some were euthymic when they died, many were depressed, and a few may have been manic. It therefore seems unlikely that we are looking at the pathology of abnormal mood per se. Instead it is more likely that we are looking at the pathology of one of the more enduring aspects of bipolar
Ch 07
7/4/05
3:43 pm
Page 63
Is any of this real? The word from the grave
63
disorder. That could be the neurocognitive abnormalities, at least some of which persist during euthymia; or, it could be the neuropathology of vulnerability to mood disturbance and of the circuits that regulate mood. Alternatively, we may be studying the neuropathology of the genetic predisposition to bipolar disorder (see below) – there have never been any postmortem studies of unaffected relatives or obligate carriers of bipolar disorder. The next point concerns anatomical specificity and precision. For example, even within the anterior cingulate cortex, to which Brodmann assigned a single number (area 24), it is clear that there is anatomical, functional and neurochemical heterogeneity13,14 (see Figure 7.1). Thus, while there is a temptation to simplify things when we talk about this part of the brain as if it were a single area, it is not. To illustrate, a student working with me did a very careful counting study at different sites within area 24.15 As one moves from the back of area 24 round to the front, one goes from having a cortex which is thick with a deep layer 5, and many large neurons and glia, to a thinner cortex with a shallow layer 5, smaller neurons, and fewer glia. The proportion of different neuron types also changes. This heterogeneity has a number of implications. One is practical – you must be very careful where you are sampling from, because if sampling is not from exactly the same area in the different groups, the differences can be larger than the differences being reported between bipolar disorder and controls. The second point, parenthetically, is that the metabolic activity of an area of the brain is determined to a large extent by its cellular and synaptic composition. Thus, some of the differences in activity seen in functional imaging studies may reflect the local cellular and synaptic content. Therefore, however pretty the pseudocolour blobs are, it is necessary to think exactly where the blob is, and to what extent the local cytoarchitecture affects the interpretation. Another issue relates to therapeutics. We have heard already that lithium may ‘grow your brain’, based on MRI findings, and the animal literature suggests that lithium may enhance neurogenesis.16 Obviously, such effects are of potential interest, not least to the pharmaceutical industry and clinicians. If it is true that some of the therapeutic effects of mood stabilizers are due to long-term alterations in synaptic and neuronal plasticity, this has implications not only for therapeutics but also for interpreting the extant neuropathology. That is, perhaps the medication received by the patients reversed or at least reduced some of the neuronal, synaptic and glial deficits. In other words, there would be more rather than less pathology if drug-free patients were studied. One paper of ours produced weak evidence supportive of this notion.17
Ch 07
7/4/05
64
3:43 pm
Page 64
Bipolar disorder: the upswing in research & treatment
Next, I return to the most striking and arguably well-replicated finding, the reduction of glia. Until recently these cells were rather boring; they were not thought to do very much except fill in the spaces between neurons, support them metabolically, and help the brain to monitor and respond to injury, infection, etc. In fact, glia are inextricably linked to neuronal and synaptic structural and functional integrity. Manipulating one affects the other. Glia can even release and uptake ‘neuro’transmitters and express many receptors; about the only thing they cannot do which a neuron can is to discharge an action potential. In this light, a loss of glia can have a range of causes and consequences, any one or more of which may underlie their involvement in bipolar disorder.9 For example, if a glial deficit were a primary event, it is plausible, even likely, that synapses and neurons will appear to be structurally abnormal in some way (i.e. be visible neuropathologically), as well as being functionally abnormal, which is presumably what is being seen in in vivo studies. This hypothetical scenario is shown in Figure 7.2. Equally, there could be a primary change in neuronal organization and activity, leading to secondary glial changes. Complicating matters further, glia are not a single entity, but exist as at least three fundamentally distinct classes of cell (astrocytes, oligodendrocytes and microglia). Each has its own origins, roles and pathological implications. The present evidence in bipolar disorder particularly implicates the oligodendrocytes, which regulate myelination and may relate to the evidence for myelin abnormalities in bipolar disorder.18 The final and most fundamental question concerns the cause(s) of the neuropathology. Prefacing Chapter 8, I am going to suggest that the structural changes in bipolar disorder are genetic in origin. That is, some of the pathology might be directly related to the genetic predisposition to bipolar disorder, just as has been postulated for schizophrenia.19,20 Indeed, the genetics of schizophrenia is (partly) the genetics of bipolar disorder, and individual genes associated with schizophrenia are now being associated with bipolar disorder too, such as G72 and COMT. Interestingly, the genes for schizophrenia all appear (at least with the eye of faith) to have a common effect upon synaptic plasticity and synaptic function, and so by extrapolation this principle may apply, to a greater or lesser extent, to bipolar disorder. If this is true, then it is relatively easy to see how they may impact upon the kinds of abnormality outlined here. However, before getting carried away with this notion, or indeed aligning bipolar disorder too closely with schizophrenia, we must also bear in mind that the morphological abnormalities seen in bipolar disorder are at least as similar to those of major depression as they are to schizophrenia (e.g. the glial loss, smaller
Ch 07
7/4/05
3:44 pm
Page 65
Is any of this real? The word from the grave
65
neurons); this is somewhat harder to explain in terms of shared genetic diathesis. Indeed, the question of the relationships between bipolar disorder, major depression and schizophrenia remain just as unclear neuropathologically as they are in every other respect. We should also bear in mind that MRI has revealed another structural aspect of bipolar disorder, namely, an excess of signal hyperintensities in the subcortical white matter. These have yet to be studied neuropathologically – in elderly unipolar depression they are focal areas of ischaemia and infarction21 – and such lesions do not fit as neatly into a genetic, developmental model.
Genetic factors
Environmental factors
Gliogenesis
Glial toxicity
Glial deficit
Formation or maintenance of synapses
Neuronal size and density
HPA axis dysfunction
Dendrites and dendritic spines
Altered circuitry
Altered neurotransmission
Impaired plasticity
MOOD DISORDER
Figure 7.2 A hypothetical glia-based origin of neuropathology in bipolar disorder.
Ch 07
7/4/05
66
3:44 pm
Page 66
Bipolar disorder: the upswing in research & treatment
Summary Neuropathological studies are providing evidence that there is a structural component to the neurobiology of bipolar disorder, and the first clues as to what its nature might be. The evidence is accumulating, but remains preliminary. To modify the metaphor in the title I was given, there is something in the graveyard, but much more digging will be needed to identify what it is and how it got there. In addition, a separate dig will be needed to identify the white matter neuropathology, and to establish whether it belongs to the same corpse.
References 1. 2. 3. 4.
5. 6.
7.
8. 9. 10.
11.
12.
Vawter MP, Freed WJ, Kleinman JE, Neuropathology of bipolar disorder. Biol Psychiatry 2000; 48:486–504. Harrison PJ, The neuropathology of primary mood disorder. Brain 2002; 125:1428–1449. Öngür D, Drevets WC, Price JL. Glial reduction in the subgenual prefrontal cortex in mood disorders. Proc Natl Acad Sci USA 1998; 95:13290–13295. Lewis DA, The human brain revisited: Opportunities and challenges in postmortem studies of psychiatric disorders. Neuropsychopharmacology 2002; 26:143–154. Torrey EF, Webster M, Knable M et al, The stanley foundation brain collection and neuropathology consortium. Schizophr Res 2000; 44:151–155. Knable MB, Torrey EF, Webster MJ, Bartko JJ, Multivariate analysis of prefrontal cortical data from the Stanley Foundation Neuropathology Consortium. Brain Res Bull 2001; 55:651–659. Knable MB, Barci BM, Bartko JJ et al, Abnormalities of the cingulate gyrus in bipolar disorder and other severe psychiatric illness: postmortem findings from the Stanley Foundation Neuropathology Consortium and literature review. Clin Neurosci Res 2002; 2:171–181. Drevets WC, Price JL, Simpson JR Jr et al, Subgenual prefrontal cortex abnormalities in mood disorders. Nature 1997; 386:824–827. Cotter DR, Pariante CM, Everall IP, Glial cell abnormalities in major psychiatric disorders: The evidence and implications. Brain Res Bull 2001; 55:585–595. Rajkowska G, Halaris A, Selemon LD, Reductions in neuronal and glial density characterize the dorsolateral prefrontal cortex in bipolar disorder. Biol Psychiatry 2001; 49:741–752. Benes FM, Kwok EK, Vincent SL, Todtenkopf MS, Reduction of nonpyramidal cells in sector CA2 of schizophrenics and manic depressives. Biol Psychiatry 1998; 44:88–97. Eastwood SL, Harrison PJ, Hippocampal synaptic pathology in schizophrenia, bipolar disorder and major depression: a study of complexin mRNAs. Mol Psychiatry 2000; 5:425–432.
Ch 07
7/4/05
3:44 pm
Page 67
Is any of this real? The word from the grave 13. 14. 15. 16.
17.
18. 19. 20.
21.
67
Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR, Human cingulate cortex: surface features, flat maps, and cytoarchitecture. J Comp Neurol 1995; 359:490–506. Paus T, Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci 2002; 2:417–424. Gittins RM, Harrison PJ, A quantitative morphometric study of the human anterior cingulate cortex. Brain Res 2004; 1013:212–222. Manji HK, Moore GJ, Chen G. Clinical and preclinical evidence for the neurotrophic effects of mood stabilizers: implications for the pathophysiology and treatment of manic-depressive illness. Biol Psychiatry 2000; 48:740–754. Eastwood SL, Harrison PJ, Synaptic pathology in the anterior cingulate cortex in schizophrenia and mood disorders: a review and a Western blot study of synaptophysin, GAP-43 and the complexins. Brain Res Bull 2001; 55: 569–578. Tkachev D, Mimmack M, Ryan MM et al, Oligodendrocyte dysfunction in schizophrenia and bipolar disorder. Lancet 2003; 362:798–804. Harrison PJ, Owen MJ, Genes for schizophrenia? Recent findings and their pathophysiological implications. Lancet 361:417–419. Harrison PJ, Weinberger DR, Schizophrenia genes, gene expression and neuropathology: on the matter of their convergence. Mol Psychiatry 2005; 10:40–68. Thomas AJ, O’Brien JT, Davis S et al, Ischemic basis for deep white matter hyperintensities in major depression – a neuropathological study. Arch Gen Psychiatry 2002; 59:785–792.
Ch 07
7/4/05
3:44 pm
Page 68
Ch 08
7/4/05
3:44 pm
Page 69
chapter 8
How can bipolar disorder be genetically related to both schizophrenia and unipolar depression? Peter McGuffin
Familial overlap This chapter addresses the seemingly complicated issue of how bipolar disorder (BPD) can be genetically related to both schizophrenia and unipolar depression (UPD). The least controversial aspect of this question concerns the overlap between UPD and BPD which, until the publication of a family study independently in the same year, 1966, by Angst1 and Perris and D’Elia2 were usually lumped together. In Table 8.1 the family study results published over approximately the following 20 years are summarized. These show a fairly consistent pattern which is that, if the starting point is a proband or index case with bipolar disorder, there is an increase in the
Table 8.1 Affective disorder in first-degree relatives of bipolar and unipolar probands (data from studies reviewed by McGuffin and Katz, 19863) Relatives Proband
Number
Age-corrected*
type
of studies
n at risk
Bipolar
12
Unipolar
7
Morbid risk† (range): % Bipolar
Unipolar
3710
7.8 (1.5–17.9)
—
3648
—
11.4 (0.5–22.4)
2319
0.6 (0.3–2.1)
9.1 (5.9–18.4)
*Corrected denominator (Bezugsziffer) to allow for relatives who have not lived through the period of risk. †Weighted means.
Ch 08
7/4/05
70
3:44 pm
Page 70
Bipolar disorder: the upswing in research & treatment
frequency of both UPD and BPD among relatives, whereas in families ascertained via a UPD proband there was an excess only of UPD. This pattern of findings was influential in convincing most researchers and clinicians that UPD and BPD were to some extent aetiologically different. Subsequently, in the 1980s, some groups began to take a more adventurous approach to family studies and began looking at the overlap with other disorders, particularly schizophrenia and schizoaffective disorder. For example, Gershon et al4 found that, in the relatives of patients with schizoaffective disorder, but not in the relatives of patients with schizophrenia, there was an increase in BPD. They also found that the incidence of major depressive disorder was increased in the relatives of both types of proband. Nevertheless, the received wisdom from twin studies was that pairs of identical individuals always show the same phenotype if they are concordant for psychosis.5 It was therefore a surprise to my colleagues and me to come across a set of triplets (the ‘Z.’ triplets) where the proband was listed on the Maudsley twin register with a diagnosis of manic depression or BPD whereas his two identical triplets had been diagnosed elsewhere as suffering from schizophrenia.6 We therefore set out to test the hypothesis that this could be explained by diagnostic error. We did this both by interviewing the triplets using two separate research interviews, the Present State Examination7 and the Schedule for Affective Disorders and Schizophrenia,8 and by obtaining ‘blind’ diagnostic assessments from three distinguished experts. The blind assessments were made from case abstracts of each of the triplets’ histories and description of mental state from which all
Organic Dx
Schizophrenia
Bipolar Dx
Unipolar Dx
Other non-psychotic Dx
Figure 8.1 A diagnostic hierarchy?
Ch 08
7/4/05
3:44 pm
Page 71
Genetic relationship to both schizophrenia and unipolar depression
71
identifying information, including current age, had been removed. The abstracts were presented to the assessors along with case abstracts based on three completely unrelated ‘decoy’ patients. To our surprise, both approaches, the expert panel judgements and the diagnoses from the research interviews, led to the triplets’ hospital diagnoses being confirmed. Furthermore, blood testing looking at several genetic polymorphisms made the probability very low that the triplets were not monozygotic. We concluded that, although genetically identical individuals such as the Z. triplets who had non-identical psychoses may be rare, we had effectively performed a ‘black swan’ observation, refuting the orthodox Kraepelinian idea that BPD and schizophrenia are totally distinct at the genetic level. (The philosopher of science Karl Popper once proposed that the way to test the hypothesis ‘all swans are white’ is to search for a black swan.)
Explanatory hypotheses If we accept that at a familial and perhaps at a genetic level there is a degree of overlap between schizophrenia, BPD and UPD, how can we explain the observed patterns? The simplest hypothesis is that we are dealing not with several but with a single disorder – a ‘unitary psychosis’. A variant on this theme is that there is a continuum of psychosis with schizophrenia at one end and unipolar depression at the other. Neither of these hypotheses works well, either in the sense of explaining the observed data in a way that brings new insights, or of appealing to practical clinical commonsense. I will therefore concentrate on two more interesting hypotheses that are amenable to testing and refutation. These are the notion of a diagnostic hierarchy and the alternative notion of overlapping sets of polygenes. The idea of a diagnostic hierarchy has been highly influential in modern psychiatric classification schemes. Although the authors of current systems such as DSMIV9 have been at pains to produce diagnostic categories that are ‘atheoretical’, all of the diagnostic criteria in DSMIV contain exclusion clauses that are implicitly based on a diagnostic hierarchy, principally that shown in Figure 8.1. Here, disorders resulting from a demonstrable brain lesion or metabolic imbalance take precedence over any ‘functional’ disorders, and functional disorders are then ranked with schizophrenia at the top, followed by bipolar disorder, unipolar disorder and then other forms of illness. Translating such a model into genetic terms is fairly straightforward if we invoke the notion of thresholds on a continuum of liability. The simple single threshold model is illustrated in Figure 8.2.
Ch 08
7/4/05
72
3:45 pm
Page 72
Bipolar disorder: the upswing in research & treatment
Population
Affected
Liability
Relatives
Figure 8.2 The polygenic liability-threshold model.
Here it is assumed that liability to develop a disorder is continuously distributed in the population and contributed to by multiple factors, both genetic and environmental, such that it will tend to have a normal distribution. Only those individuals who at some point exceed the threshold are classed as affected. When the disorder is familial, relatives of affected probands have an increased mean liability such that more of them lie beyond the threshold than occurs in the general population.10 This can be extended to multiple thresholds by considering broader versus narrower forms of illness with probands who have the narrow and more extreme forms tending to have more relatives affected than probands with broader forms.11 This is shown diagrammatically in Figure 8.3. Such a model has been fitted to family data by Gershon et al12 where it was hypothesized that the disorder beyond the most extreme threshold is schizoaffective disorder, with bipolar I, bipolar II and unipolar disorder lying beyond successively broader thresholds. It was found that the model explained the data satisfactorily. However, the main problem with such an approach is that a large family dataset may be required in order to reject the model. Twin data can sometimes provide a more stringent test. McGuffin et al13 applied a two-threshold model, in which bipolar disorder was the narrow form of illness and unipolar disorder the broad form, to
Ch 08
7/4/05
3:45 pm
Page 73
Genetic relationship to both schizophrenia and unipolar depression
73
Relatives affected +
Relatives affected ++ Relatives affected +++
Figure 8.3 A multiple threshold model.
twin data consisting of 67 pairs ascertained through bipolar probands and 177 pairs ascertained through unipolar probands. They were able to reject conclusively a model where the two forms for affective disorder are on the same continuum of liability. They then went on to fit a rather different kind of model, testing the extent to which liability to the two forms of disorder have correlated genetic and environmental bases. They concluded that the genetic component of UPD and BPD is substantially correlated but that most of the genetic variation in liability to the manic syndrome is specific to mania. Fitting such a model requires a departure from conventional nosology, in that subjects who have had both manic and depressive symptoms are classified as having two syndromes, whereas subjects who have had only manic or only depressive symptoms are classified as having a single syndrome. The approach has been applied also to schizophrenia, schizoaffective disorder and bipolar disorder in a twin study14 where the usual hierarchical rules were suspended and subjects were classed as having one, two or all three disorders. The principal finding was that there is a set of genes that contribute to all three syndromes and specific genes that contribute only to bipolar disorder or to schizophrenia. Interestingly, there was no evidence of a specific set of genes contributing to schizoaffective disorder, but there was evidence of a specific environmental effect. These findings have been viewed as somewhat controversial and a commentary in the same issue of the journal suggested that the authors’ interpretation that there are overlapping sets of genes contributing to both sides of the Kraepelinian dichotomy was incorrect.15 However, there is now emerging evidence from molecular studies that at least some of the same genes do indeed contribute to both schizophrenia and bipolar disorder.
Ch 08
7/4/05
74
3:45 pm
Page 74
Bipolar disorder: the upswing in research & treatment
The molecular findings Modern gene finding studies are beginning to yield useful results in both schizophrenia and affective disorders.16 The positional cloning approach has been described in greater detail elsewhere in this volume (Chapter 9). Essentially positional cloning consists of searching through the genome, the 23 pairs of chromosomes, using linkage analysis, narrowing down the region of interest by linkage disequilibrium mapping and then finding potential candidates within the refined region with which further association studies are performed to confirm or reject the hypothesis of a role in the disorder. The starting point then is linkage, and two major sets of metaanalyses have recently been performed in order to try to make sense of the large body of data that now exists. One such meta-analysis17 implicated two regions on chromosomes 13q and 22q in both schizophrenia and bipolar disorder and a third region on chromosome 8 that was specific to schizophrenia. Subsequently there has been much interest in candidate genes within these regions. A search across the 13q region in schizophrenia resulted in the identification of a novel primate-specific gene known as G72, the gene product of which was found to interact in vitro, using a method called yeast 2 hybrid analysis, with D-amino acid oxidase (DAAO).18 There was also statistical evidence of epistasis (gene–gene interaction) between G72 and the DAAO gene. There have now been several studies supporting the association between G72 and schizophrenia19–21 but there have also been several studies that have found an association between G72 and bipolar disorder.20,22,23 So far, a number of other positional candidates have been implicated only in schizophrenia,16 but the prediction that follows on from the twin analysis described earlier is that future molecular studies will uncover sets of genes that contribute either to schizophrenia or to bipolar disorder and gene sets contributing to both syndromes.
Conclusions Modern genetic evidence does not support the idea that schizophrenia and the affective disorders are completely distinct at an aetiological level. Neither does it support the idea that there is a ‘continuum’ of psychosis with schizophrenia at one end and BPD at the other. Rather, there appear to be overlapping gene sets. That is, there are genes that contribute both to schizophrenia and to BPD, and genes that contribute only to one or other
Ch 08
7/4/05
3:45 pm
Page 75
Genetic relationship to both schizophrenia and unipolar depression
75
syndrome. BPD also shows genetic overlap with UPD but a simple hierarchical, or two-threshold model, where BPD is a more extreme and severe form on a single continuum of liability does not explain the available twin data. There does appear to be a sizeable overlap between the genes contributing to unipolar and bipolar disorders, but most of the genetic liability to the manic syndrome is specific to mania. This is a result that, as Bearden et al24 have pointed out in a recent review, has major implications for genefinding studies.
References 1.
2.
3. 4.
5. 6. 7. 8. 9.
10. 11.
12.
13.
Angst J, Zur aetilogie und nosologie endogener depressiver psychosen. Monographen ans der Neurologie und Psychiatrie, N112. Springer: Berlin, 1966. Perris C, D’Elia G, A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses X: mortality, suicide, and life cycles. Acta Psychiatr Scand 1966; 42(194 Suppl):172–183. McGuffin P, Katz R, Nature, nurture and affective disorders. In: Deakin JW (ed), The Biology of Depression. Gaskell Press: London, 1986:26–51. Gershon ES, Delisi LE, Hamovit J et al, A controlled family study of chronic psychoses. Schizophrenia and schizoaffective disorder. Arch Gen Psychiatry 1988; 45:328–336. Gottesman II, Shields J, Schizophrenia – the Epigenetic Puzzle. Cambridge University Press: Cambridge, 1982. McGuffin P, Reveley A, Holland A, Identical triplets: non-identical psychosis? Br J Psychiatry 1982; 140:1–6. Wing JK, Cooper JE, Sartorius N, The Measurement and Classification of Psychiatric Symptoms. Cambridge University Press: Cambridge, 1974. Endicott J, Spitzer RL, A diagnostic interview: the schedule for affective disorders and schizophrenia. Arch Gen Psychiatry 1978; 35:837–844. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSMIV). American Psychiatric Association Press; Washington DC, 1994. Falconer DS, The inheritance of liability to certain diseases, estimated from the incidence among relatives. Ann Hum Genet 1965; 29:51–76. Reich T, James JW, Morris CA, The use of multiple thresholds in determining the mode of transmission of semi-continuous traits. Ann Hum Genet 1972; 36:163–184. Gershon ES, Hamovit J, Guroff JJ et al, A family study of schizoaffective bipolar I, bipolar II, unipolar, and normal control probands. Arch Gen Psychiatry 1982; 39:1157–1167. McGuffin P, Rijsdijk F, Andrew M et al, The heritability of bipolar affective disorder and the genetic relationship to unipolar depression. Arch Gen Psychiatry 2003; 60:497–502.
Ch 08
7/4/05
76 14. 15. 16. 17. 18.
19. 20.
21.
22.
23.
24.
3:45 pm
Page 76
Bipolar disorder: the upswing in research & treatment Cardno AG, Rijsdijk FV, Sham PC et al, A twin study of genetic relationships between psychotic symptoms. Am J Psychiatry 2002; 159:539–545. Kendler KS, Hierarchy and heritability: the role of diagnosis and modeling in psychiatric genetics. Am J Psychiatry 2002; 159:515–518. Elkin A, Kalidindi S, McGuffin P, Have schizophrenia genes been found? Curr Opin Psychiatry 2004; 17:107–113. Badner JA, Gershon ES, Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Mol Psychiatry 2002; 7:405–411. Chumakov I, Blumenfield M, Guerassimenko O et al, Genetic and physiological data implicating the new human gene G72 and the gene for D-amino acid oxidase in schizophrenia. Proc Natl Acad Sci USA 2002; 99:13675–13680. Wang X, He G, Gu N et al, Association of G72/G30 with schizophrenia in the Chinese population. Biochem Biophys Res Commun 2004; 319:1281–1286. Schumacher J, Jamra RA, Freudenberg J et al, Examination of G72 and Damino-acid oxidase as genetic risk factors for schizophrenia and bipolar affective disorder. Mol Psychiatry 2004; 9:203–207. Addington AM, Gornick M, Sporn AL et al, Polymorphisms in the 13q33.2 gene G72/G30 are associated with childhood-onset schizophrenia and psychosis not otherwise specified. Biol Psychiatry 2004; 55:976–980. Chen YS, Akula N, Detera-Wadleigh SD et al, Findings in an independent sample support an association between bipolar affective disorder and the G72/G30 locus on chromosome 13q33. Mol Psychiatry 2004; 9:87–92. Hattori E, Liu C, Badner JA et al, Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. Am J Hum Genet 2003; 72:1131–1140. Bearden CE, Reus VI, Freimer NB, Why genetic investigation of psychiatric disorders is so difficult. Curr Opin Genet Dev 2004; 14:280–286.
Ch 09
7/4/05
3:46 pm
Page 77
chapter 9
Recent advances in genetics of bipolar disorder Daniel J Müller and James L Kennedy
Introduction It has long been recognized that manic depression or bipolar disorder (BD) runs in families and therefore a familial and/or genetic component in the aetiology of BD has been postulated. Epidemiological research involving family, twin and adoption studies led to the observation that genetic factors do confer susceptibility to BD.1 The overall prevalence of BD appears to be about 1% in the general population. In contrast, first-degree relatives have a 5–10-fold increased risk of developing BD, while it appears that the risk for second-degree relatives falls between risks for first-degree relatives and the general population. Concordance for BD has consistently been found to be significantly higher in monozygotic than in dizygotic twins (about 50% versus 10%, which is similar to the rate for first-degree relatives).1,2 These findings point to some important conclusions: the clustering of BD in families is based on the presence of genetic factors; however, since penetrance is incomplete, the presence of specific non-genetic (epigenetic and/or environmental) factors is likely to influence the occurrence of BD.3 Finally, it is generally accepted, based on statistical models, that not one but many (or at least several) genes of moderate or small effect contribute to BD.4
From symptoms to syndromes: the pitfalls of phenotype definition The recognition and distinction of manic symptoms as pathognomonic features of BD (e.g. elevated mood, pressured speech, grandiose delusions) can usually be achieved with high reliability among clinicians. However, as with
Ch 09
7/4/05
78
3:46 pm
Page 78
Bipolar disorder: the upswing in research & treatment
many other complex disorders, the clinical presentation of BD itself is highly heterogeneous. This heterogeneity is likely to be caused by a variety of specific genes that may interact with neuronal circuitries or synaptic processes. It is likely that only through the combination of different genes is a threshold reached and disorder-related symptoms appear. It remains a matter of debate whether affective and psychotic symptoms may represent a continuum that ranges from dysthymia through depression, bipolar II and bipolar I towards schizoaffective disorder and schizophrenia.5 Therefore, and since genes ‘do not read’ DSM-IV or ICD-10, it seems plausible that genetic studies of BD must take into account the heterogeneity of BD and should focus on core symptoms that are known to be ‘familial’ or genetically determined. However, only recently have researchers begun to focus on symptoms, endophenotypes (such as imaging or electrophysiological techniques) or to incorporate covariates in their analyses with some promising preliminary findings. Response pattern to mood stabilizers is another promising clinical measure to refine the phenotype of patients with BD.6
How to find genes involved in bipolar disorder: principles of genetic linkage and association studies It should be emphasized here that genetic mechanisms in the aetiology of BD are poorly understood. Whether or not the susceptibility is associated with DNA sequence variation alone or in combination with epigenetic mechanisms (i.e. gene expression controlled by potentially reversible changes in DNA methylation and/or chromatin structure3) remains unknown at present. Nonetheless, the first step is to analyse genomic DNA sequence variation patterns in subjects affected with BD compared with non-affected individuals. In Mendelian disorders, distinct genetic variations have already been proven to be causative for the occurrence of a given disease. Therefore, it is reasonable to hypothesize that distinct genetic variants will increase the susceptibility to a complex disorder such as BD. However, genes that are involved in the aetiology of BD remain unknown and facts are more complicated when one considers that BD risk conferring sequence variation may occur anywhere in the genomic DNA (e.g. in promoter regions, in intronic and/or exonic regions or in intragenic regions). Finally, sequence variations may be harmless (in terms of increasing disease risk) but may also affect gene expression, protein structure, or mRNA stability.7 The functional relevance of any sequence variation needs to be established in further analyses.
Ch 09
7/4/05
3:46 pm
Page 79
Recent advances in genetics of bipolar disorder
79
Two complementary approaches are widely applied in medical genetics to identify genes involved in the aetiology of a given disorder: linkage analysis and genetic association studies. Linkage analyses are based on the principle that two loci are linked if they are close to each other and are co-inherited, meaning that they are not transmitted independently. Thus, in linkage analyses two chromosomal loci are considered: one that may harbour the unknown locus (or putative gene) and one that is a marker trait for which the site is known. Typically, large family pedigrees are chosen for linkage studies using trait markers spanning the whole genome. Subsequently, analyses seek to determine those markers that are typically inherited (and are ‘linked’) to the disorder. Finally, conclusions can be drawn that indicate the broader location of susceptibility genes. Statistical analyses are divided into parametric and nonparametric linkage analyses, depending upon whether parameters (such as mode of inheritance, level of penetrance) are known or included alternatively based on a priori hypotheses. Furthermore, analyses may be performed considering two markers or multiple markers (i.e. two point or multipoint analyses). Estimates of significance levels are generally given in log of the odds ratio (LOD) scores (for parametric analyses) or NPL scores (for non-parametric analyses).8,9 The LOD score is defined by the probability that the locus is linked to disease divided by the probability that the locus is not linked to disease given the observed data. Specific guidelines have been proposed on how to interpret or classify LOD scores.10 Based on these criteria, LOD scores above 2.2 are suggestive of significant findings, whereas a LOD score above 3.6 will establish significance of findings obtained in linkage analyses. However, interpretation is more complex, as individual results need to be put into the context of findings obtained from independent studies. Therefore, LOD scores above 1.0 are generally reported in linkage studies, as such scores may receive importance in the presence of similar (or higher) LOD scores from independent studies. Genetic association studies typically compare DNA sequence variations between unrelated cases and controls. Controls can either be recruited from a healthy population (i.e. not being affected by the disorder; this is also called case–control design) or from healthy family members (so-called family-based design). The latter approach has the advantage that family member controls are not prone to population stratification, meaning that false-positive findings are less likely to be due to allelic heterogeneity adherent in specific populations. Association studies can either be performed with candidate genes that have been proved (or have been supposed) to play an aetiological role in a given disorder (and are called functional candidate
Ch 09
7/4/05
80
3:46 pm
Page 80
Bipolar disorder: the upswing in research & treatment
genes) or may be chosen on the basis of their specific location that has emerged from linkage studies (and are called positional candidate genes).11 Problems with linkage and association studies derive from the commonly accepted assumption that genetic susceptibility to BD is heterogeneous in terms of loci and alleles. Locus heterogeneity means a disorder will develop if a combination of genes in locus A and/or B (and/or C, etc.) confers susceptibility to BD. Thus, locus heterogeneity will cause problems (in terms of difficulties of replicating initial findings) in linkage studies. Allelic heterogeneity means that different polymorphisms in one gene will confer susceptibility to BD, comparable to the scenario wherein many different variants of one gene all cause cystic fibrosis. Hence, allelic heterogeneity will hamper replication findings in allelic association studies. It should be noted that linkage and association studies are not juxtaposed strategies but are to be seen as complementary approaches. Once a chromosomal locus has been identified, this region will ideally be narrowed down with more markers towards the highest signals. Once that ‘hot spot’ site with highest linkage signals has been identified, subsequent case– control studies should be carried out to identify those specific genetic variations that confer susceptibility to BD. In the presence of only few compelling candidate genes, these positional approaches have been suggested to represent the most powerful strategies in genetic studies of BD.12 Other genetic mechanisms that could be involved in the aetiology of BD include some types of chromosomal aberrations, dynamic mutations (expanded trinucleotide repeat sequences), or mitochondrial mutations. The latter are not likely to be involved in the aetiology of BD, although this cannot be excluded at this point.13,14 Finally, the phenomenon of epistasis (gene–gene interactions) currently receives much attention. However, there are no widely accepted statistical models for epistasis, although this field is rapidly progressing.
Linkage studies in bipolar disorder Linkage analyses from numerous genome scans have yielded several regions that may harbour susceptibility genes for bipolar disorder. Not surprisingly, results arising out of these studies are hampered by inconsistent findings, as comparisons among study findings remain complicated, owing to multiple study heterogeneities (e.g. sample sizes, ascertainment strategies, molecular genetic techniques and so on). The most interesting findings, indicated either by highly significant findings in single studies or through
Ch 09
7/4/05
3:46 pm
Page 81
Recent advances in genetics of bipolar disorder
81
replication of initial findings, include chromosomal regions 1q31-q32, 4p15-16, 6pter-p24, 8q24, 10p13, 10p14, 10q21-26, 12q23-24.1, 13q31-32, 16p13, 18p11.2, 18q12, 18q21-23, 20p11.2-q11.2, 21q22, 22q11-q13 and Xq24-28.5,8,14,15 It is interesting to note that linkage analysis on chromosome 18 indicated a parent-of-origin effect, implying complex mechanisms such as mitochondrial mutations or genomic imprinting to be involved in the aetiology of BD.16–19 Most of the genome scans that have been performed were included in two recent large meta-analyses.20,21 Badner and Gershon21 found the strongest evidence for linkage for chromosomes 13q and 22q. Depending on their disease models (i.e. including patients with bipolar II and major depression), Segurado et al20 found most significant results on chromosomes 8q, 9p22.3-21.1; 10q11.21-22.1, 14q24.1-32.12 and 18p-q. However, none of these regions achieved genomewide statistical significance. Reasons for inconsistent findings among the analyses may be due to methodological factors that varied considerably in terms of dataset selection, disease modelling, statistical analyses, etc. Furthermore, as pointed out by Segurado et al,20 negative findings do not disprove linkage and therefore more efforts are needed to identify BD susceptibility loci in the future. Moreover, it should be noted that these chromosomal regions represent millions of base-pairs. On average, a new genetic polymorphism occurs once every thousand base-pairs. Thus, hundreds of DNA markers are typically needed to narrow down the region of interest to the true risk-conferring gene variant. Such investigations are usually laborious, costly and time consuming. However, these efforts are currently underway and are likely to unearth new genes associated with BD.
Candidate gene association studies in bipolar disorder Studies focusing on particular genes using case–control association studies have yielded some positive findings, but also negative (non-replication) findings. Thus, the significance of specific results needs to be weighed in light of several important parameters in order to estimate their true relevance. It remains difficult to depict algorithms (or guidelines) as to how to interpret findings arising from association studies. Nonetheless, each study should at least address some important questions:
Ch 09
7/4/05
82 1. 2.
3.
4.
5.
6.
7.
3:46 pm
Page 82
Bipolar disorder: the upswing in research & treatment
Sample size (of cases and controls): Are samples large enough in terms of statistical power? (Discussed in more detail by reference 22). Ascertainment: Were samples selected in ethnically homogeneous populations? Were samples ascertained randomly or systematically? Were diagnoses, symptoms and course of disorder all assessed by interviewers trained to conduct clinical (semi-) structured interviews and chart assessments reliably? Selection of the candidate gene: How has the gene been selected (functional or positional candidate gene)? Are there any (reasonable) assumptions or findings emerging from neurobiological research providing at least some evidence that the gene in question is involved in the pathophysiology of BD? Are there any supportive findings deriving from linkage analyses to pinpoint the particular gene as a positional candidate gene? Gene structure/selection of genetic polymorphisms: Has the gene already been investigated in terms of its structure (e.g. promoter, intron/exon boundaries, etc.)? Is there any evidence that the polymorphism itself bears functional relevance (e.g. modulation of gene expression)? Are there polymorphisms with minor allele frequencies high enough for statistical analyses? Genotyping: High-quality assurance needs to be used in terms of technological equipment and data management (i.e. supervision, crosschecking, etc.) Statistical analyses: Depending on the number of gene variants that were chosen for the analyses, appropriate strategies and tests have to be chosen carefully (e.g. haplotype analyses, problems of multiple testing, etc.) Initial (positive) finding – or replication? (Rule of thumb: An initial positive finding is interesting, but only further replications will determine the true ‘significance’.)
The observer of molecular genetic studies in BD will quickly realize that most studies suffer from at least some methodological flaws, although numerous candidate gene studies have been performed. Most of our current knowledge on genetics of BD is hypothesis-based and some important aspects remain yet to be determined. For example, the pathophysiology of BD remains largely unknown. Hence, a functional candidate cannot be determined with certainty, unless each of the perhaps 30 000 human genes are to be considered as candidate genes. Moreover, the functional relevance of a genetic polymorphism is often impossible to determine; if a polymorphism is found to be functionally relevant in vitro, this is not necessarily
Ch 09
7/4/05
3:46 pm
Page 83
Recent advances in genetics of bipolar disorder
83
true in vivo. Conversely, if a test indicates that a gene polymorphism is not functionally relevant in vitro, this may not be so in vivo – or may be due to lack of sensitivity of the given test.
Neurotransmitter systems Based on current knowledge in psychopharmacology, most association studies focused on a priori functional candidate genes that are related to the mechanism of action of monoamine neurotransmitters such as serotonin, noradrenaline (norepinephrine) or dopamine. Polymorphisms of the catechol-o-methyl transferase (COMT) gene, monoamine oxidase A (MAOA) gene and the serotonin transporter (5-HTT) gene revealed most interesting results even though these findings could not be replicated in all studies (e.g. references 11, 23–32). These findings are most consistent with the notion that these genes are likely to play a minor role (i.e. with odds ratios < 2) and may perhaps be associated with yet undefined subgroups of patients affected with BD. In the dopamine system, the genetic mechanism of imprinting was suggested in the aetiology of BD by Muglia et al,33 who found evidence of a risk effect of functional variable number of tandem repeats (VNTRs) in exon III of the DRD4 receptor gene. In that report, the risk was transmitted through maternal, but not paternal, transmission.
G72/G30 locus (and DAAO) and brain-derived neurotrophic factor A large variety of functional and positional candidate genes have been analysed for association with BD over the years. Here we focus on two candidate genes that we believe preliminary studies suggest consistently interesting findings.
G72/G30 locus and DAAO Chumakov et al34 identified the gene G72 on chromosome 13q33.2, a region that has previously been found to be linked with schizophrenia and BD in some studies. G72 overlaps with another gene, G30 and therefore this locus was named the G72/G30 locus. Chumakov et al34 demonstrated that the G72 protein interacts as activator on the D-amino acid oxidase (DAAO) enzyme. In turn, DAAO oxidizes D-serine, an activator of N-methyl-Daspartate (NMDA)-type glutamate receptors. Modulation of NMDA receptors may be involved in the aetiology of schizophrenia.35,36 Thus, the G72/G30 locus may be regarded as a positional and functional candidate gene for
Ch 09
7/4/05
84
3:46 pm
Page 84
Bipolar disorder: the upswing in research & treatment
schizophrenia and bipolar disorder. A shared genetic susceptibility between BD and schizophrenia has been suggested for some time.37 In their initial study, Chumakov et al34 identified polymorphisms of the G72 gene associated with schizophrenia without including patients with BD. In contrast, Hattori et al38 included patients with BD and found significant associations with markers at and around the G72/G30 locus. Importantly, similar significant findings were found independently at the G72/G30 locus in patients with BD in subsequent studies by Chen et al39 and Schumacher et al.40 Even though findings varied with respect to specific DNA markers and statistical results among these studies, the results relating to the G72/G30 locus should be regarded as an important development in the search for susceptibility genes for BD. Nonetheless, further studies are clearly needed.
Brain-derived neurotrophic factor The brain-derived neurotrophic factor (BDNF) gene is located on chromosome 11p14.1.41 One study by Detera-Wadleigh et al42 identified moderate pairwise parametric LOD scores of 1.84 at marker D11S915 (11p15-p14) and of 1.62 at marker D11S904 (11p14-p13) under the assumption of a dominant model of inheritance with 50% penetrance. The affection status model in this study was restricted to bipolar I, bipolar II (with major depression) and schizoaffective disorder. However, non-parametric multipoint linkage analyses revealed no comparable findings in the same study. Besides that study, one other recent study by Zandi et al43 detected a signal of highest linkage on chromosome 11p15.5. Nonetheless, there is compelling evidence that BDNF is involved in a variety of complex mechanisms that are thought to be relevant in BD. The mechanism of action of BDNF is multifunctional and complex44 and this paragraph will only provide a short summary. BDNF is a neuronal growth factor that was found to be critical for the normal development of central 5HT neurons in mice. For example, BDNF-deficient mice display an aggressive hyperphagic phenotype;45 symptoms that are observed in mood disorders and have long been hypothesized to be associated with serotonin dysfunction. Serotonin dysfunction is likely to play a major role in depression, anxiety and suicidal behaviour, which are common symptoms in BD. An increased BDNF expression has been observed in the frontal cortex and hippocampus after treatment with antidepressants, lithium, or valproic acid in rats.46 In human postmortem studies, increased hippocampal BDNF immunoreactivity was found in patients treated with antidepressant medications.47 In contrast, decreased levels of BDNF mRNA and protein in postmortem frontal cortex and hippocampus were detected in suicide completers versus controls.48 Serum levels of
Ch 09
7/4/05
3:46 pm
Page 85
Recent advances in genetics of bipolar disorder
85
BDNF were found to increase significantly after antidepressant medication was administered in patients with major depression compared with untreated patients with major depression or controls. Moreover, a significant negative correlation between serum BDNF and Hamilton Rating Scale for Depression (HAM-D) scores was noted in patients.49 In summary, these findings clearly suggest that modulation of BDNF is associated with symptoms pertaining to depression and/or BD. Moreover, symptom-relieving drugs might act by modulating (i.e. increasing) BDNF expression or BDNF distribution. The BDNF gene is relatively large (approximately 67 kb) with seven probable promoters and fifteen alternatively spliced exons producing 11 different transcripts creating five variations of the BDNF protein (NCBI Aceview). The BDNF gene structure is thus complex and the gene-map has been modified over the course of time.44,50,51 The map we present here (Figure 9.1) was constructed from the primer sequences used in our ongoing studies and exon sequence given on NCBI Aceview. The most prominent polymorphism that has been studied extensively is a G to A substitution that results in an amino acid substitution (valine to methionine) in codon 66 (named Val66Met polymorphism). The Val66Met polymorphism is found in alternatively spliced exons 13 through 15 (previously called exon 11,44 exon 6,50 or exon 551) one of which is contained in
BDNF located on (–) strand
Exon:
1 23 4 5
6 7 8 9 10
11 12 13 14 15
val66met (rs6265) A(Met)/G(val)
(GT)n repeat
(+) strand direction
Figure 9.1 Map of brain-derived neurotrophic factor (BDNF) gene with location of Val66Met and GT(n) repeat polymorphisms.
Ch 09
7/4/05
86
3:46 pm
Page 86
Bipolar disorder: the upswing in research & treatment
all BDNF transcript variants. This polymorphism has been proven to be of functional relevance since in cultured hippocampal neurons the Val allele (vBDNF) has been shown to increase intraneuronal BDNF peptide secretion and distribution compared to the Met allele (mBDNF).52 Using fluorescent microscopy techniques, neurons expressing vBDNF were shown to express BDNF in the cell body and distal processes (dendrites). In contrast, mBDNF was mainly localized in cell bodies. Furthermore, it appeared that neurons expressing mBNDF displayed an impaired (i.e. decreased) secretion of BDNF (see Figure 4 A–C in reference 52). The Val allele was found to be associated with BD in two independent studies simultaneously.53,54 In the study of Neves-Perreira et al,53 an intronic polymorphism, a GT(n) repeat upstream of the Val66Met polymorphism was found to be associated with BD. Although these studies varied in methodology55 they both included large samples of similar ethnic groups and used a family-based association approach. In another family-based association study, by Geller et al,56 a significant association was found with the Val allele and BD in children with a prepubertal and early adolescent BD phenotype. Negative findings used case–control designs and samples derived from other ethnic backgrounds.50,57–60 However, in a recent study using a case–control design and larger samples of patients with major depression and BD, significant associations were detected with three marker haplotypes including the Val66Met polymorphism.61 One study, by Strauss et al,62 reported a negative finding with the Val66Met polymorphism in adults with childhood-onset mood disorders, but a significant association with the GT(n) repeat and haplotype containing the Val allele. Inconsistent findings among these studies indicate that BDNF apparently plays a role but not in all patients affected with BD; more likely to be in specific – yet undefined – subgroups of patients. It has long been recognized that BD represents a heterogeneous disorder in terms of clinical and/or epidemiological characteristics (e.g. sex, age at onset, psychotic symptoms, course of disorder, response pattern to medication, etc.). Preliminary findings in our follow-up studies indicate that our significant findings with the BDNF gene in BD are mainly driven by those patients who displayed ‘rapid cycling’ (i.e. the occurrence of four or more mood episodes (major depressive, manic, mixed or hypomanic) within a time frame of 12 months).63 Furthermore, significant findings between the BDNF gene and psychiatric disorders other than BD (e.g. schizophrenia,60,64 obsessive–compulsive disorder,51 adult attention deficit hyperactivity disorder65) indicate that
Ch 09
7/4/05
3:46 pm
Page 87
Recent advances in genetics of bipolar disorder
87
BDNF is likely to be associated with a broader range of symptoms that are encountered in a variety of neuropsychiatric disorders. In conclusion, the meaningfulness of the BDNF gene in BD needs to be further elucidated by focusing on specific characteristics adherent to subgroups of BD patients, and also symptom sharing with other known disorders in neuropsychiatry.
Summary and outlook There is unequivocal evidence that genetic factors play a major role in the aetiology of BD. More precisely, we expect that a yet unknown number of gene variants confer susceptibility to BD that will lead to exacerbation of manic or depressive episodes in conjunction with environmental factors such as stress or major life events. The search for vulnerability genes over the past 20 years has not been as successful as was initially hoped. This time period has left behind many inconsistent findings and some false leads. However, rather than looking back in dismay it is perhaps more important to learn lessons from the past studies in order to achieve better interpretation of current findings and address the correct questions in the future. Therefore, we emphasized on the interpretation of study findings rather than presenting each single locus or candidate gene that has been studied; understanding the strengths and weaknesses of genetic studies, as well as integrating genetic results with neurobiology and clinical sciences is more valuable than simply tabulating each and every LOD score and p value. The findings in BDNF tell us an important and demonstrative story. The BDNF gene is most likely not to be a risk factor for BD per se but a risk factor for BD in certain groups of patients. Sources of inconsistent findings may be that – aside from the problem of large sample sizes that are needed to identify genes with small effects (and most samples were relatively small) – patients ascertained for family-based studies may differ from patients ascertained for case–control samples.66 The group of patients may be defined by certain characteristics (e.g. earlier age at onset) or specific course of disorder (e.g. rapid cycling) or distinct symptoms (e.g. depressive symptomatology), or a combination of such features related to BD. Definite answers are likely to be obtained in the near future, provided that the field is prepared to take into consideration the lessons learned from the past. We are confident that the detection of genetic factors underlying BD will become a reality over the coming years.
Ch 09
7/4/05
88
3:46 pm
Page 88
Bipolar disorder: the upswing in research & treatment
Acknowledgements This work was supported by a Canadian Institutes of Health Research (CIHR) operating grant to JLK, and a CIHR postdoctoral fellowship award to DJM. Thanks to Matthew Lanktree who assisted in this work.
References 1. 2. 3. 4. 5. 6. 7.
8. 9.
10. 11. 12. 13.
14. 15.
16.
Smoller JW, Finn CT, Family, twin, and adoption studies of bipolar disorder. Am J Med Genet C Semin Med Genet 2003; 123:48–58. Belmaker RH, Bipolar disorder. N Engl J Med 2004; 351:476–486. Petronis A, Epigenetics and bipolar disorder: new opportunities and challenges. Am J Med Genet C Semin Med Genet 2003; 123:65–75. Kelsoe JR, Arguments for the genetic basis of the bipolar spectrum. J Affect Disord 2003; 73:183–197. Tsuang MT, Taylor L, Faraone SV, An overview of the genetics of psychotic mood disorders. J Psychiatr Res 2004; 38:3–15. Alda M, Pharmacogenetic aspects of bipolar disorder. Pharmacogenomics 2003; 4:35–40. Duan J, Wainwright MS, Comeron JM et al, Synonymous mutations in the human dopamine receptor D2 (DRD2) affect mRNA stability and synthesis of the receptor. Hum Mol Genet 2003; 12:205–216. Sklar P, Linkage analysis in psychiatric disorders: the emerging picture. Annu Rev Genomics Hum Genet 2002; 3:371–413. Schulze TG, McMahon FJ, Genetic linkage and association studies in bipolar affective disorder: a time for optimism. Am J Med Genet C Semin Med Genet 2003; 123:36–47. Lander ES, Kruglyak L, Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995; 11:241–247. Craddock N, Dave S, Greening J, Association studies of bipolar disorder. Bipolar Disord 2001; 3:284–298. Gershon ES, Kelsoe JR, Kendler KS, Watson JD, A scientific opportunity. Science 2001; 294:957. Fortune MT, Kennedy JL, Vincent JB, Anticipation and CAG*CTG repeat expansion in schizophrenia and bipolar affective disorder. Curr Psychiatry Rep 2003; 5:145–154. Baron M, Manic-depression genes and the new millennium: poised for discovery. Mol Psychiatry 2002; 7:342–358. Cichon S, Schumacher J, Muller DJ et al, A genome screen for genes predisposing to bipolar affective disorder detects a new susceptibility locus on 8q. Hum Mol Genet 2001; 10:2933–2944. McMahon FJ, Stine OC, Meyers DA et al, Patterns of maternal transmission in bipolar affective disorder. Am J Hum Genet 1995; 56:1277–1286.
Ch 09
7/4/05
3:46 pm
Page 89
Recent advances in genetics of bipolar disorder 17.
18.
19.
20.
21. 22. 23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
89
Gershon ES, Badner JA, Detera-Wadleigh SD et al, Maternal inheritance and chromosome 18 allele sharing in unilineal bipolar illness pedigrees. Am J Med Genet 1996; 67:202–207. Nothen MM, Cichon S, Rohleder H et al, Evaluation of linkage of bipolar affective disorder to chromosome 18 in a sample of 57 German families. Mol Psychiatry 1999; 4:76–84. McInnis MG, Lan TH, Willour VL et al, Genome-wide scan of bipolar disorder in 65 pedigrees: supportive evidence for linkage at 8q24, 18q22, 4q32, 2p12, and 13q12. Mol Psychiatry 2003; 8:288–298. Segurado R, Detera-Wadleigh SD, Levinson DF et al, Genome scan metaanalysis of schizophrenia and bipolar disorder, part III: Bipolar disorder. Am J Hum Genet 2003; 73:49–62. Badner JA, Gershon ES, Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Mol Psychiatry 2002; 7:405–411. Risch N, Merikangas K, The future of genetic studies of complex human diseases. Science 1996; 273:1516–1517. Kirov G, Norton N, Jones I et al, A functional polymorphism in the promoter of monoamine oxidase A gene and bipolar affective disorder. Int J Neuropsychopharmacol 1999; 2:293–298. Kunugi H, Ishida S, Kato T et al, A functional polymorphism in the promoter region of monoamine oxidase-A gene and mood disorders. Mol Psychiatry 1999; 4:393–395. Preisig M, Bellivier F, Fenton BT et al, Association between bipolar disorder and monoamine oxidase A gene polymorphisms: results of a multicenter study. Am J Psychiatry 2000; 157:948–955. Ho LW, Furlong RA, Rubinsztein JS et al, Genetic associations with clinical characteristics in bipolar affective disorder and recurrent unipolar depressive disorder. Am J Med Genet 2000; 96:36–42. Mundo E, Walker M, Tims H et al, Lack of linkage disequilibrium between serotonin transporter protein gene (SLC6A4) and bipolar disorder. Am J Med Genet 2000; 96:379–383. Jones I, Craddock N, Candidate gene studies of bipolar disorder. Ann Med 2001; 33:248–256. Rotondo A, Mazzanti C, Dell’Osso L et al, Catechol o-methyltransferase, serotonin transporter, and tryptophan hydroxylase gene polymorphisms in bipolar disorder patients with and without comorbid panic disorder. Am J Psychiatry 2002; 159:23–29. Gutierrez B, Arias B, Gasto C et al, Association analysis between a functional polymorphism in the monoamine oxidase A gene promoter and severe mood disorders. Psychiatr Genet 2004; 14:203–208. Mendlewicz J, Massat I, Souery D et al, Serotonin transporter 5HTTLPR polymorphism and affective disorders: no evidence of association in a large European multicenter study. Eur J Hum Genet 2004; 12:377–382. Lasky-Su JA, Faraone SV, Glatt SJ, Tsuang MT, Meta-analysis of the association between two polymorphisms in the serotonin transporter gene and affective disorders. Am J Med Genet B Neuropsyhchiatr Genet 2005; 133:110–115.
Ch 09
7/4/05
90 33.
34.
35.
36. 37. 38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
3:46 pm
Page 90
Bipolar disorder: the upswing in research & treatment Muglia P, Petronis A, Mundo E et al, Dopamine D4 receptor and tyrosine hydroxylase genes in bipolar disorder: evidence for a role of DRD4. Mol Psychiatry 2002; 7:860–866. Chumakov I, Blumenfeld M, Guerassimenko O et al, Genetic and physiological data implicating the new human gene G72 and the gene for D-amino acid oxidase in schizophrenia. Proc Natl Acad Sci USA 2002; 99:13675–13680. Mundo E, Tharmalingham S, Neves-Pereira M et al, Evidence that the N-methylD-aspartate subunit 1 receptor gene (GRIN1) confers susceptibility to bipolar disorder. Mol Psychiatry 2003; 8:241–245. Sivagnansundaram S, Müller DJ, Gubanov A et al, Genetics of schizophrenia: Current strategies. Clin Neurosci Res 2003; 3:5–16. Berrettini W, Evidence for shared susceptibility in bipolar disorder and schizophrenia. Am J Med Genet C Semin Med Genet 2003; 123:59–64. Hattori E, Liu C, Badner JA et al, Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. Am J Hum Genet 2003; 72:1131–1140. Chen YS, Akula N, Detera-Wadleigh SD et al, Findings in an independent sample support an association between bipolar affective disorder and the G72/G30 locus on chromosome 13q33. Mol Psychiatry 2004; 9:87–92; image 5. Schumacher J, Jamra RA, Freudenberg J et al, Examination of G72 and Damino-acid oxidase as genetic risk factors for schizophrenia and bipolar affective disorder. Mol Psychiatry 2004; 9:203–207. Hanson IM, Seawright A, van Heyningen V, The human BDNF gene maps between FSHB and HVBS1 at the boundary of 11p13-p14. Genomics 1992; 13:1331–1333. Detera-Wadleigh SD, Badner JA, Berrettini W et al, A high-density genome scan detects evidence for a bipolar-disorder susceptibility locus on 13q32 and other potential loci on 1q32 and 18p11.2. Proc Natl Acad Sci USA 1999; 96:5604–5609. Zandi PP, Willour VL, Huo Y et al, Genome scan of a second wave of NIMH genetics initiative bipolar pedigrees: chromosomes 2, 11, 13, 14, and X. Am J Med Genet B Neuropsychiatr Genet 2003; 119:69–76. Green E, Craddock N, Brain-derived neurotrophic factor as a potential risk locus for bipolar disorder: evidence, limitations, and implications. Curr Psychiatry Rep 2003; 5:469–476. Lyons WE, Mamounas LA, Ricaurte GA et al, Brain-derived neurotrophic factor-deficient mice develop aggressiveness and hyperphagia in conjunction with brain serotonergic abnormalities. Proc Natl Acad Sci USA 1999; 96:15239–15244. Fukumoto T, Morinobu S, Okamoto Y et al, Chronic lithium treatment increases the expression of brain-derived neurotrophic factor in the rat brain. Psychopharmacology (Berl) 2001; 158:100–106. Chen B, Dowlatshahi D, MacQueen GM et al, Increased hippocampal BDNF immunoreactivity in subjects treated with antidepressant medication. Biol Psychiatry 2001; 50:260–265. Dwivedi Y, Rizavi HS, Conley RR et al, Altered gene expression of brain-derived neurotrophic factor and receptor tyrosine kinase B in postmortem brain of suicide subjects. Arch Gen Psychiatry 2003; 60:804–815.
Ch 09
7/4/05
3:46 pm
Page 91
Recent advances in genetics of bipolar disorder 49.
50.
51.
52.
53.
54.
55. 56.
57.
58.
59.
60. 61.
62.
63.
64.
91
Shimizu E, Hashimoto K, Okamura N et al, Alterations of serum levels of brainderived neurotrophic factor (BDNF) in depressed patients with or without antidepressants. Biol Psychiatry 2003; 54:70–75. Nakata K, Ujike H, Sakai A et al, Association study of the brain-derived neurotrophic factor (BDNF) gene with bipolar disorder. Neurosci Lett 2003; 337:17–20. Hall D, Dhilla A, Charalambous A et al, Sequence variants of the brain-derived neurotrophic factor (BDNF) gene are strongly associated with obsessive– compulsive disorder. Am J Hum Genet 2003; 73:370–376. Egan MF, Kojima M, Callicott JH et al, The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 2003; 112:257–269. Neves-Pereira M, Mundo E, Muglia P et al, The brain-derived neurotrophic factor gene confers susceptibility to bipolar disorder: evidence from a familybased association study. Am J Hum Genet 2002; 71:651–655. Sklar P, Gabriel SB, McInnis MG et al, Family-based association study of 76 candidate genes in bipolar disorder: BDNF is a potential risk locus. Brainderived neutrophic factor. Mol Psychiatry 2002; 7:579–593. Schulze TG, Hardy J, McMahon FJ, Inconsistent designs of association studies: a missed opportunity. Mol Psychiatry 2003; 8:770–772. Geller B, Badner JA, Tillman R et al, Linkage disequilibrium of the brainderived neurotrophic factor Val66Met polymorphism in children with a prepubertal and early adolescent bipolar disorder phenotype. Am J Psychiatry 2004; 161:1698–1700. Hong CJ, Huo SJ, Yen FC et al, Association study of a brain-derived neurotrophic-factor genetic polymorphism and mood disorders, age of onset and suicidal behavior. Neuropsychobiology 2003; 48:186–189. Kunugi H, Iijima Y, Tatsumi M et al, No association between the Val66Met polymorphism of the brain-derived neurotrophic factor gene and bipolar disorder in a Japanese population: a multicenter study. Biol Psychiatry 2004; 56:376–378. Skibinska M, Hauser J, Czerski PM et al, Association analysis of brain-derived neurotrophic factor (BDNF) gene Val66Met polymorphism in schizophrenia and bipolar affective disorder. World J Biol Psychiatry 2004; 5:215–220. Neves-Pereira M, Cheung JK, Pasdar A et al, BDNF gene is a risk factor for schizophrenia in a Scottish population. Mol Psychiatry 2005; 10:208–212. Cichon S, Schumacher J, Abou Jamra R et al, Supportive evidence for a relationship between genetic variations at the brain-derived-neurotrophic factor (BDNF) locus and depressive symptoms in affective disorder and schizophrenia. Am J Hum Genet (Neuropsych Genet) 2004; 130B:27. Strauss J, Barr CL, George CJ et al, Association study of brain-derived neurotrophic factor in adults with a history of childhood onset mood disorder. Am J Med Genet 2004; 131B:16–19. Müller DJ, De Luca V, Sicard T et al, ‘Rapid cycling’ mainly determines significant findings between the BDNF gene and bipolar disorder. Am J Hum Genet (Neuropsych Genet) 2004; 130B:45. Muglia P, Vicente AM, Verga M et al, Association between the BDNF gene and schizophrenia. Mol Psychiatry 2003; 8:146–147.
Ch 09
7/4/05
92 65.
66.
3:46 pm
Page 92
Bipolar disorder: the upswing in research & treatment Lanktree M, Muglia P, Squassina A et al, A potential role for brain derived neurotrophic factor (BDNF) in adult ADHD. Am J Hum Genet (Neuropsych Genet) 2004; 130B:96. Schulze TG, Muller DJ, Krauss H et al, Caught in the trio trap? Potential selection bias inherent to association studies using parent–offspring trios. Am J Med Genet 2001; 105:351–353.
Ch 10
7/4/05
3:47 pm
Page 93
chapter 10
Is there a genetic basis to the brain abnormalities of bipolar disorder? Colm McDonald ‘…I could demonstrate a hereditary taint (for manic depressive insanity) in about 80 per cent of cases observed in Heidelberg.’ Emil Kraepelin1 Familial liability to bipolar disorder has been recognized since the clinical syndrome was introduced a century ago, and twin studies have further demonstrated that such familiality is related to genetic transmission of the illness.2 The heritability of operationally defined bipolar disorder is estimated to be over 80%, similar to that of other psychotic disorders such as schizophrenia,2,3 with the remaining variance attributable to non-shared environmental risk factors. Clearly, susceptibility genes must exist for this illness, but progress in identifying such genes has been exceedingly slow, owing to the complexity of genetic transmission. There are unlikely to be any genes of major effect for bipolar disorder, expect perhaps in some rare pedigrees, and instead inheritance is probably due to many genes of small effect. Furthermore, the molecular genetics research of bipolar disorder, like many other psychiatric disorders, is hindered by the inaccuracy of phenotypic definition; we lack laboratory diagnostic tests, even at postmortem, and the use of the clinical syndrome alone as the phenotype is problematic because of the effects of genetic heterogeneity and reduced penetrance.
What do susceptibility genes do? Susceptibility genes do not directly produce the symptoms of bipolar disorder. Genes code for proteins and such proteins impact upon brain function and brain structure at some level. The symptoms by which we diagnose the
Ch 10
7/4/05
94
3:47 pm
Page 94
Bipolar disorder: the upswing in research & treatment
illness emerge from the underlying neurobiological abnormalities produced by susceptibility genes. Therefore, an alternative and potentially complementary approach to the use of the qualitative clinical syndrome in molecular genetic studies is to identify those neurobiological markers of brain dysfunction or abnormal structure which are produced by susceptibility genes and to utilize these as alternative phenotypes. This approach is gaining momentum within schizophrenia research but has rarely been applied to bipolar disorder. Such ‘endophenotypes’, which are presumed more proximal to gene action than the clinical syndrome, can be identified by employing study designs that examine how candidate neurobiological markers vary among subjects with increasing genetic risk. The first such design within schizophrenia research examined how biological variables differed in subjects with a strong family history, who were presumably enriched with susceptibility genes, compared with those with no family history of illness. This design is problematic not least since the presence of illness itself, with accompanying disease progression, chronic medication and other health effects, can impact upon the biological variables being measured. A more powerful design is to use unaffected subjects who are at a higher genetic risk for the illness, such as co-twins, first-degree relatives or offspring. Such relatives are free from any medication or disease effects but share 50% of the affected patient’s genes (or 100% in the case of unaffected monozygotic cotwins) and are thus more likely to manifest the effects of susceptibility genes upon neurobiological markers than are control subjects with no family history of illness. Further power to detect genetic effects can be attained by introducing a gradient of genetic liability among the unaffected relatives studied, for example including relatives from both multiply affected and singly affected families, or unaffected co-twins from both monozygotic and dizygotic pairs. In support of this approach, there is accumulating evidence within the schizophrenia literature that unaffected first-degree relatives from multiply affected families have more severe abnormalities of potential cognitive and morphometric endophenotypic measures than unaffected relatives from singly affected families.4,5
Why might brain morphometry represent a useful endophenotype? Potential endophenotypic neurobiological markers should be: (1) heritable themselves; (2) measurable in both affected and unaffected subjects; (3)
Ch 10
7/4/05
3:47 pm
Page 95
Is there a genetic basis to the brain abnormalities of bipolar disorder?
95
manifest whether or not the illness is active; (4) associated with the illness in the general population; and (5) found more frequently in unaffected relatives of patients than controls.6 Brain morphometry meets some of these criteria. Magnetic resonance imaging (MRI) brain scans are easily measurable in unaffected individuals, and several MRI studies of healthy twins have demonstrated high degrees of genetic control over volumetric measurements of the cerebrum, grey and white matter and several subregions of the brain.7,8 The morphometry of bipolar disorder is relatively underresearched compared to schizophrenia; however, the illness is associated with subtle deviation from normal brain anatomy at the group level when patients with bipolar disorder are compared with controls (see Chapter 4 for a meta-analysis of structural brain deviations in bipolar disorder). The most consistent evidence supports mild enlargement of lateral ventricular volume and higher rates of white matter hyperintensities, with some evidence for reduced volume of parts of the prefrontal lobe, such as the anterior cingulate gyrus.9,10 There has been little attempt to date to identify brain abnormalities in the unaffected relatives of bipolar disorder patients in order to elucidate any effect of bipolar disorder susceptibility genes on gross brain structure.
The Maudsley Family Study of Psychosis This large study was initially developed to identify endophenotypic markers of schizophrenia by examining a range of brain structural and functional measurements in patients with schizophrenia and their unaffected relatives at varying genetic risk. Recently, the approach has also been extended to psychotic bipolar disorder. T1-weighted MRI brain scans were successfully obtained in 37 patients who fulfilled DSM-IV criteria for bipolar I disorder and 50 of their unaffected first-degree relatives. The patients had all experienced psychotic symptoms at some stage during episodes of illness exacerbation. All the subjects were from multiply affected families, in that the patients had at least one other relative with psychotic bipolar disorder or another functional psychotic disorder among their first- and/or seconddegree relatives. The patients had a mean age of 41 years and mean duration of illness of 18 years. All were outpatients at the time of assessment and 33 were taking mood-stabilizing medication, mostly lithium. Although all families were multiply affected, the density of illness within the families varied considerably. In order to model variation in genetic risk among subjects, we calculated a continuous measure of likely genetic liabil-
Ch 10
7/4/05
3:47 pm
96
Page 96
Bipolar disorder: the upswing in research & treatment
ity using a quantitative scale.11 This measure is based upon a multifactorial liability threshold model of illness, which assumes that liability within the population is normally distributed, and utilizes information about affection status from all adult members in each pedigree as far as second degree from the index patient, taking into account family size and genetic relatedness to affected individuals. Examples of the genetic liability scores produced for families of differing density are demonstrated in Figure 10.1. This measure of genetic liability was then used to predict regional volume variation of grey and white matter throughout the entire brain derived from computational morphometry analyses of the MRI scans of all patients and relatives. Optimized voxel-based morphometry12 was used to segment MRI images into grey matter, white matter and cerebrospinal fluid and to normalize grey and white matter maps to templates derived from healthy controls matched on demographic variables. Non-parametric analyses controlling for the confounds of age, gender and subject group and utilizing cluster level statistics to account for multiple testing at the voxel level were employed.
A
0.01
0.03
0.03
0.01
0.01
0.01
0.21
0.63
1.75
0.19
1.95
–0.08
0.42
0.42
1.99
0.19
B 0.03
0.01
0.01
0.01
0.09
–0.03
0.01
0.01
0.01
0.03 0.01
0.21
0.32
–0.03 0.01
0.01
0.01
0.01
0.26
0.26
1.68
0.12
0.12
0.12
0.12
0.11
–0.02
1.68
0.10
–0.03
0.01
0.01
0.01
0.01
0.01
Figure 10.1 Genetic liability scores in families with differing density of illness. (A) An unaffected parent (arrowed), who appears to be transmitting genetic susceptibility to her son, has a relatively high genetic liability of 0.63. (B) Although still multiply affected, the large number of unaffected individuals in this pedigree is reflected by the relatively low genetic liability score of 0.10 for an unaffected sibling (arrowed).
Ch 10
7/4/05
3:47 pm
Page 97
Is there a genetic basis to the brain abnormalities of bipolar disorder?
97
Grey matter Increasing genetic liability for bipolar disorder was associated with reduced grey matter volume in a discrete region comprising the right medial frontal lobe, including the pregenual and subgenual sections of the anterior cingulate gyrus, and the ventral striatum (Figure 10.2A). The relationship between increasing genetic liability and reduced volume in these regions was present in both patients and their unaffected relatives, since no interaction existed between subject group and genetic liability score upon regional grey matter volume defined by the cluster in a multiple regression analysis employing multilevel modelling (B, –0.27, 95% CI –1.85–1.30, p = 0.72; Figure 10.2B). This indicates that the association was not determined solely by abnormalities in the patients. The anterior cingulate gyrus and ventral striatum are components of brain circuits which are critical in the regulation of normal emotion13 and abnormalities of this region have been previously detected in familial bipolar disorder using structural and functional neuroimaging.14 The present study demonstrates that volume deficit is not solely a disease-related phenomenon but reflects the impact of susceptibility genes upon this area, since volume deficit is identified also in those unaffected relatives at highest genetic liability for bipolar disorder.
White matter Increasing genetic liability for bipolar disorder was also associated with distributed white matter volume deficits involving the anterior corpus callosum as well as bilateral frontal, left temporoparietal and right parietal regions (Figure 10.3A). The relationship between increasing genetic liability and reduced volume in these regions was present in both patients and their unaffected relatives, since no interaction existed between subject group and genetic liability score upon first principal component factor scores, which summarized (explaining 80.7% of the variance) the highly intercorrelated regional white matter volumes defined by the clusters, in a multiple regression analysis employing multilevel modelling (B, 0.11, 95% CI –1.82–2.03, p = 0.72; Figure 10.3B). This indicates that the association was not determined solely by abnormalities in the patients. These hemispheric white matter regions are characteristically occupied by major intrahemispheric tracts: the superior longitudinal fasciculus, which connects the frontal lobe to the temporal, parietal and occipital lobes; and the inferior longitudinal fasciculus, which connects the temporal pole to the occipital lobe.
7/4/05
98
3:49 pm
Page 98
Bipolar disorder: the upswing in research & treatment
A –3
0
6
12
20
28
B 1.6 Patients Relatives 1.4 Grey matter endophenotype
Ch 10
1.2
1.0
0.8
0.6 0.4
0.6
0.8
1.0
1.2
1.4
1.6
Genetic liability score
Figure 10.2 Grey matter endophenotype of bipolar disorder. (A) Map of grey matter volume deficits (red voxels) associated with increasing genetic liability for bipolar disorder superimposed onto a single brain in standard stereotactic space. Cluster-wise probability of false-positive activation (p = 0.008), producing < 1 expected false-positive test. The brain slices are orientated in the plane of the Talairach atlas; distance (mm) above or below the intercommissural line is inset in the left corner of each slice; the right side of the brain is depicted by the right side of each axial slice. (B) Linear associations between grey matter volume deficit (value under grey matter cluster) and genetic risk for bipolar disorder (genetic liability scores adjusted to the sample mean for age, gender and affection status) estimated separately for bipolar patients and their non-psychotic relatives.
7/4/05
3:54 pm
Page 99
Is there a genetic basis to the brain abnormalities of bipolar disorder?
A
0
6
12
20
28
35
42
50
99
B 3 Patients Relatives
2 White matter endophenotype
Ch 10
1 0 –1 –2 –3 –4 0.4
0.6
0.8
1.0
1.2
1.4
1.6
Genetic liability score
Figure 10.3 White matter endophenotype of bipolar disorder. (A) Map of white matter volume deficits (red voxels) associated with genetic liability for bipolar disorder superimposed onto a single brain in standard stereotactic space. Clusterwise probability of false-positive activation (p = 0.01), producing < 1 expected false-positive test. The brain slices are orientated in the plane of the Talairach atlas; distance (mm) above or below the intercommissural line is inset in the left corner of each slice; the right side of the brain is depicted by the right side of each axial slice. (B) Linear associations between systemic white matter volume deficit (first principal component factor scores, which summarize correlated white matter deficit) and genetic risk for bipolar disorder (genetic liability scores adjusted to the sample mean for age, gender and affection status) estimated separately for bipolar patients and their non-psychotic relatives.
Ch 10
7/4/05
3:54 pm
100
Page 100
Bipolar disorder: the upswing in research & treatment
White matter abnormalities in bipolar disorder are supported by convergent findings from case–control studies using varying research methodologies. Studies have consistently demonstrated increased rates of hyperintensities detectable on T2-weighted MRI images;9 volume deficit of left hemispheric white matter has been reported;15 reduced fractional anisotropy and macromolecular density in the prefrontal white matter has been detected using diffusion tensor imaging16 and magnetization transfer imaging17 respectively; abnormalities of myelination and oligodendrocyte function are indicated by down-regulation of gene expression in postmortem prefrontal tissue.18 The present study indicates that white matter abnormalities are not confined to the impact of the illness or its treatment, but that some susceptibility genes for bipolar disorder are likely to be associated with reduced volume of inter- and intrahemispheric white matter tracts. Interestingly, an overlapping endophenotype comprising left temporoparietal white matter volume reduction was identified for schizophrenia using similar methodology, consistent with the hypothesis that shared susceptibility genes for psychosis in general produce disturbed left frontotemporal anatomical connectivity.11
What are the implications of identifying morphometric endophenotypes of bipolar disorder? These results from the Maudsley Family Study of Psychosis demonstrate that susceptibility genes for bipolar disorder do have an impact upon gross brain structure which is detectable from MRI scans. Volume deficit in cortical and subcortical regions involved in mood regulation and in more widespread inter- and intra-hemispheric white matter tracts is associated with increasing genetic liability, not only in patients but also in unaffected relatives who have never experienced the illness of bipolar disorder or its treatment. The identification of these morphometric endophenotypes is likely to represent more proximal effects of susceptibility genes for bipolar disorder than the clinical syndrome, and we will now use them as alternative phenotypes in molecular genetic studies in two ways: (1) allelic variation in genes which may confer susceptibility to bipolar disorder identified by studies using the clinical syndrome as the phenotype, will now be tested for association with the morphometric endophenotypes to assess whether such genetic variation impacts upon these brain structures; (2) the morphometric endophenotypes identified by this study will themselves be used as
Ch 10
7/4/05
3:54 pm
Page 101
Is there a genetic basis to the brain abnormalities of bipolar disorder?
101
phenotypes in molecular genetic studies informed by the neurobiology of the tissue classes and regions identified – for example, functional polymorphisms in genes coding for myelination and oligodendrocyte function are suitable candidates for association studies using the white matter endophenotype identified by the present study. In conclusion, genetic liability for psychotic bipolar disorder is manifest in volumetric deficit of grey matter in the right anterior cingulate and ventral striatum, and of white matter in bilateral frontal and temporoparietal regions. The identification of these morphometric endophenotypes provides a neuroanatomical substrate for molecular genetic studies of bipolar disorder and should facilitate the identification of the elusive genes that confer susceptibility to this highly heritable illness.
Acknowledgement Colm McDonald is supported by the Wellcome Trust.
References 1. 2.
3.
4.
5.
6. 7. 8.
Kraepelin E, Manic Depressive Insanity and Paranoia. [Translated by R.M. Barclay] E&S Livingstone: Edinburgh, 1921:165. McGuffin P, Rijsdijk F, Andrew M et al, The heritability of bipolar affective disorder and the genetic relationship to unipolar depression. Arch Gen Psychiatry 2003; 60:497–502. Cardno AG, Marshall EJ, Coid B et al, Heritability estimates for psychotic disorders: the Maudsley twin psychosis series. Arch Gen Psychiatry 1999; 56:162–168. Faraone SV, Seidman LJ, Kremen WS et al, Neuropsychological functioning among the nonpsychotic relatives of schizophrenic patients: the effect of genetic loading. Biol Psychiatry 2000; 48:120–126. McDonald C, Grech A, Toulopoulou T et al, Brain volumes in familial and nonfamilial schizophrenic probands and their unaffected relatives. Am J Med Genet 2002; 114:616–625. Leboyer M, Bellivier F, Nosten-Bertrand M et al, Psychiatric genetics: search for phenotypes. Trends Neurosci 1998; 21:102–105. White T, Andreasen NC, Nopoulos P, Brain volumes and surface morphology in monozygotic twins. Cereb Cortex 2002; 12:486–493. Wright IC, Sham P, Murray RM et al, Genetic contributions to regional variability in human brain structure: methods and preliminary results. Neuroimage 2002; 17:256–271.
Ch 10
7/4/05
3:54 pm
102 9. 10.
11.
12. 13. 14. 15. 16.
17.
18.
Page 102
Bipolar disorder: the upswing in research & treatment Bearden CE, Hoffman KM, Cannon TD, The neuropsychology and neuroanatomy of bipolar disorder: a critical review. Bipolar Disord 2001; 3:106–150. McDonald C, Zanelli J, Rabe-Hesketh S et al, Meta-analysis of magnetic resonance imaging brain morphometry studies in bipolar disorder. Biol Psychiatry 2004; 56:411–417. McDonald C, Bullmore ET, Sham PC et al, Association of genetic risks for schizophrenia and bipolar disorder with specific and generic brain structural endophenotypes. Arch Gen Psychiatry 2004; 61:974–984. Good CD, Johnsrude IS, Ashburner J et al, A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001; 14:21–36. Rolls ET, The Brain and Emotion. Oxford University Press: Oxford, 1999. Drevets WC, Price JL, Simpson JR Jr et al, Subgenual prefrontal cortex abnormalities in mood disorders. Nature 1997; 386:824–827. Kieseppa T, van Erp TG, Haukka J et al, Reduced left hemispheric white matter volume in twins with bipolar I disorder. Biol Psychiatry 2003; 54:896–905. Adler CM, Holland SK, Schmithorst V et al, Abnormal frontal white matter tracts in bipolar disorder: a diffusion tensor imaging study. Bipolar Disord 2004; 6:197–203. Bruno SD, Barker GJ, Cercignani M et al, A study of bipolar disorder using magnetization transfer imaging and voxel-based morphometry. Brain 2004; 127:2433–2440. Tkachev D, Mimmack ML, Ryan MM et al, Oligodendrocyte dysfunction in schizophrenia and bipolar disorder. Lancet 2003; 362:798–805.
Ch 11
7/4/05
3:56 pm
Page 103
chapter 11
Transgenic mouse models for affective disorders based on the neurotrophin hypothesis Peter Gass
Introduction According to the monoaminergic hypothesis, depression is related to an impairment of neurotransmission by serotonin, norepinephrine (noradrenaline) and dopamine. These deficiencies can result from several mechanisms: (1) decreased synthesis or increased degradation of neurotransmitters; (2) altered expression or function of the respective neurotransmitter receptors; or (3) impairment of signal transduction systems activated by post-synaptic receptors. Most antidepressant drugs act primarily via the first mechanism, aiming to improve monoaminergic transmission by increasing the presence of neurotransmitters inside the synapse. It is still unclear how the various antidepressants with their different modes of action (e.g. serotonin or norepinephrine re-uptake or degradation inhibitors) finally induce emotional and behavioural improvements. Common to all classes of antidepressants is also that their onset of action takes 2–3 weeks. Therefore, a key biological mechanism must be responsible for the fact that only chronic administration has mood-elevating effects, while enhancement of serotoninergic or noradrenergic neurotransmission occurs within minutes after drug intake. Recent concepts for pathogenesis and therapy claim slowly developing plasticity changes induced by chronic alterations in monoaminergic neurotransmission. Two biological systems have attracted special attention: (1) the stress-responsive hypothalamic–pituitary–adrenal (HPA) system, which is disinhibited in many patients with major depressive episodes;1,2 and (2) the neurotrophin brainderived neurotrophic factor (BDNF), which has been implicated in
Ch 11
7/4/05
3:56 pm
104
Page 104
Bipolar disorder: the upswing in research & treatment
hippocampal maladaptative processes related to depressive episodes.3,4 Interestingly, regulatory mechanisms that control stress hormones and hippocampal BDNF expression are linked to each other. This has led to the socalled ‘neurotrophin hypothesis of depression’.
The neurotrophin hypothesis of depression The CREB–BDNF–TrkB pathway A variety of signalling molecules (e.g. neurotransmitters, growth factors) that act via cAMP- or Ca2+-activated kinases, or by direct activation of tyrosine kinases, lead to the phosphorylation of the transcription factor cAMP response element (CRE)-binding protein (CREB). This activation results in elevations in CRE-mediated gene transcription of downstream target genes such as BDNF. BDNF exerts its effects mainly by activation of tyrosine receptor kinase B (TrkB) (Figure 11.1). Neurotrophins and especially BDNF influence structural plasticity and have trophic effects on neurons. Clinical and experimental observations have led to the concept that a deficiency in BDNF contributes to the pathophysiology of depression.3,4 BDNF has a high concentration in brain regions thought to be involved in the pathogenesis of depression, such as the hippocampus or neocortex.5 Moreover, CREB-mediated BDNF induction is a downstream effect of chronic but not acute antidepressive treatment.3,4
Stress decreases the activity of the CREB–BDNF–TrkB pathway Chronic stress, and a subsequent rise in plasma corticosteroids, is regarded as a major cause in the pathogenesis of depressive disorders.6 Accordingly, physical and psychosocial stresses have been used to induce a depression-like syndrome in rodents.7,8 Stress or corticosteroid injections cause a decrease in BDNF mRNA in the hippocampus and other brain areas thought to be involved in the pathogenesis of depression, most probably by activation of glucocorticoid receptors.9–11 This effect is mitigated by electroconvulsive therapy (ECT) or by chronic administration of antidepressants.10 Similar to the ECT findings in vivo, up-regulation of BDNF triggered by depolarization in cultured hippocampal neurons is blocked by dexamethasone, a synthetic glucocorticoid.12 In a genetic rat model of depression, the Flinders Sensitive Line, several brain regions have lower BDNF levels as control lines.13
Ch 11
7/4/05
3:56 pm
Page 105
Transgenic mouse models for affective disorders
105
At the clinical level, postmortem investigations show that untreated depressive patients have lower levels of CREB in the temporal cortex than healthy subjects.14,15 Moreover, patients with major depression have lower BDNF serum concentrations than controls, and these levels are negatively correlated with scores in clinical depression rating scales.16 Altogether, human and animal studies suggest that the depressive state is positively correlated with CREB and BDNF expression, because reduced levels are caused by stress exposure and reversed by antidepressant treatment.
Antidepressants activate the CREB–BDNF–TrkB pathway Chronic but not acute treatment with antidepressants increases BDNF and CREB mRNA levels in the hippocampus.10,17 Long-term ECT also induces
Stress
Depression
Glucocorticoids
GR
TrkB + – Plasticity & trophic effects
BDNF + CREB + Antidepressants & ECT
Kinases
Monoaminergic neurotransmitters & growth factors
Figure 11.1 The neurotrophin hypothesis of depression. Chronic stress combined with a rise in plasma corticosteroid levels is regarded as a major factor in the pathogenesis of depressive disorders. Corticosteroids can downregulate brainderived neurotrophic factor (BDNF) expression through an activation of glucocorticoid receptors (GR). Decreased BDNF expression leads to reduced tyrosine receptor kinase B (TrkB) activation, which, directly or via other transmitter systems, decreases the activation of the transcription factor cAMP response element binding protein (CREB). By a vicious cycle, the reduced CREB activity further diminishes BDNF expression. Antidepressants block the stress-induced decrease in BDNF expression and are also able directly to enhance CREB, BDNF and TrkB expression and signalling. ECT, electroconvulsive therapy.
Ch 11
7/4/05
3:56 pm
106
Page 106
Bipolar disorder: the upswing in research & treatment
the expression of BDNF and TrkB in limbic brain regions.10,18,19 Moreover, chronic antidepressant treatment induces TrkB phosphorylation, which in turn enhances CREB activation.20 Because the CREB–BDNF–TrkB pathway is a closed circle, it is difficult to determine which molecular changes are the cause and which are the effect (Figure 11.1). Results from animal models suggest that antidepressants do not only prevent a stress-induced decrease in CREB–BDNF–TrkB signalling, but are also able to activate this pathway. At the clinical level, higher concentrations of CREB and BDNF are found in patients under antidepressive medication than in untreated patients.14,15 Patients with a major depressive disorder on antidepressant medication show increased TrkB expression when compared with healthy subjects.21 Successfully treated patients also show a significant increase of CREB phosphorylation in T lymphocytes.22 Thus, current clinical data also suggest that antidepressant effects are mediated by activation of the CREB–BDNF–TrkB pathway.
Neurotrophins have antidepressant-like effects BDNF, when infused near the raphe nucleus, the main source of serotonergic innervation of the hippocampus, has an antidepressant-like effect in the learned-helplessness model of depression.23 This paradigm uses uncontrollable and inescapable stress to evoke physiological and behavioural abnormalities similar to those observed in human depression.24 Bilateral BDNF infusion into the hippocampus also produces antidepressant-like effects in the learned-helplessness paradigm and in the forced swim test.25 In summary, while antidepressive drugs activate the CREB–BDNF–TrkB pathway, exogenous application of neurotrophins can also induce an antidepressant-like effect.
Transgenic mice Several strains of mice with altered or disrupted expression of CREB, BDNF or
TrkB
have
been
used
to
analyse
individual
steps
of
the
CREB–BDNF–TrkB signalling cascade, and to test the predictions made by the neurotrophin hypothesis of depression. Generally, one would expect a predisposition for depression, if this pathway were compromised. A battery of behavioural tests has been developed to define syndromically – similarly to the human classification systems – a depression-like state in mice.26 The
Ch 11
7/4/05
3:56 pm
Page 107
Transgenic mouse models for affective disorders
107
following emotional states or behaviours can be investigated by tests: ‘anhedonia’ (sucrose preference test), ‘despair’ (forced swim test, tail suspension test), diminished interest (novel object approach), anxiety (elevated T-maze, dark–light-box test, openfield test) and general activity (openfield test, activity boxes).
Transgenic mice with compromised CREB expression A complete knockout of all isoforms of the CREB gene (α, ∆ and β) leads to postnatal death due to respiratory failure, preventing analyses of adult animals.27 Therefore, most behavioural experiments have been performed in mice lacking the α and ∆ isoforms (so-called CREBα∆ mutant mice), leading to a more than 90% reduction of CRE-binding activity in the brain.28–30 In contrast to the predictions made by the CREB–BDNF–TrkB hypothesis, CREBα∆ mutant mice demonstrated significantly less despair behaviour than controls.30 Similar results are described in mice with inducible overexpression of a dominant-negative CREB form, leading to decreased CREB activity in the forebrain.31 These animals show reduced depression-like behaviours in the learned-helplessness paradigm. In accordance with these findings, inducible overexpression of CREB evokes increased depressionlike behaviours.31 Thus, opposed to expectations from clinical and experimental data, reduced CREB expression or function does not induce despair behaviour but rather has an antidepressant-like effect. Despite the paradoxical reduction of behavioural despair in CREBα∆ mutant mice, treatment with antidepressants results in a further decrease of the despair reaction, to a similar extent as observed in wild-type mice, indicating that the mechanisms that underlie the behavioural effects of antidepressant drugs are less dependent on CREB activation.30 However, the antidepressant-induced upregulation of BDNF in the hippocampus and cortex is absent in CREBα∆ mutant animals,30 demostrating that CREB is an upstream regulator of BDNF and may be a critical downstream mediator of transcriptional effects of antidepressants. Surprisingly, these data also suggest an uncoupling between transcriptional BDNF upregulation and direct behavioural effects of antidepressants.
Transgenic mice with reduced BDNF expression Conventional BDNF knockout mice do not survive to adulthood, because of severe developmental defects. Behavioural experiments are therefore restricted to heterozygous mice (BDNF+/–) or mice with a conditional
Ch 11
7/4/05
3:56 pm
108
Page 108
Bipolar disorder: the upswing in research & treatment
knockout. BDNF+/– mice with about 50% reduced BDNF levels are behaviourally indistinguishable from control littermates in a test battery for locomotor, exploratory, anxiety- and depression-related behaviours.32,33 According to the theory, one would expect that these mice would be more susceptible to developing depression-like symptoms. However, they do not differ in their hedonic capacity (sucrose consumption) nor in behavioural measures for despair.32,33 The HPA-system as a key marker for major depressive episodes is also not affected in these animals.33 A coping deficit of BDNF+/– mice in the learned-helplessness paradigm has been attributed to an impaired pain sensitivity.32 The fact that all other measures of the stress response are unchanged in BDNF+/– mice is perplexing, considering the proposed role of BDNF in the pathophysiology of depression. A possible explanation may be that a 50% reduction of BDNF is not sufficient to induce strong behavioural effects, and a compensatory mechanism may additionally occur during development. Indeed, mice with a complete forebrain-specific conditional BDNF knockout display increased anxiety-related behaviours.34 Other behavioural tests more related to anhedonia, despair and coping should be performed with this strain. Interestingly, there is an uncoupling between baseline behaviours and antidepressant-induced behavioural alterations in BDNF mutant mice. Acute treatment with imipramine fails to reduce the despair reaction of BDNF+/– mice in the forced swim test that is seen in wild-type mice, suggesting that BDNF directly mediates the behavioural effects of antidepressant drugs.20
Transgenic mice with reduced TrkB expression Mice with a forebrain-specific disruption of TrkB receptors also lack depression-like behaviours and HPA-system alterations.35 Similar results are obtained with mice constitutively overexpressing a dominant negative form of TrkB, leading to a reduced TrkB activation in the brain.20 However, similar to BDNF+/– mice, antidepressants do not reduce the despair reaction in TrkB-deficient mice.20 These findings confirm that antidepressants directly influence despair behaviour in mice via BDNF–TrkB signalling.20 The temporal profile of these experiments – with behavioural alterations observed as early as 30 minutes after drug treatment – strongly indicates that these effects occur at a post-transcriptional level. This interpretation is in line with findings in CREB-mutant mice (see above), which exhibit a normal reduction of despair behaviour in response to antidepressants, but have regular BDNF levels under baseline conditions.
Ch 11
7/4/05
3:56 pm
Page 109
Transgenic mouse models for affective disorders
109
Discussion Based on clinical and experimental observations, the neurotrophin hypothesis of depression as conceptualized originally made the following predictions: (1) reduced activity of the CREB–BDNF–TrkB pathway is implicated in the pathogenesis of depression; and (2) activation of the CREB–BDNF–TrkB pathway is part of the molecular mechanisms of antidepressive therapy.3,4 This concept has been challenged by studies with transgenic mice. According to the neurotrophin hypothesis, mice with genetic disruptions of any part of this pathway are expected to display depression-like behaviours. However, none of several mouse strains investigated has exhibited such behaviours, some mice being even less ‘depressive’ than the controls. This surprising result could be due to factors that may mask or prevent a depressive phenotype. First, a complete knockout of genes coding for any part of the pathway is not compatible with survival to adulthood. Therefore, most studies have been performed with heterozygous mice or strains with incomplete downregulation of the targeted genes. This reduction in gene expression may not have been sufficient to result in behavioural alterations. However, this interpretation seems unlikely regarding the results obtained with many other mouse strains where a 50% under- or overexpression of genes has caused strong behavioural effects. Second, mice carrying a transgene or mutation since early embryogenesis may develop compensatory mechanisms to overcome the effects of the targeted genes. Third, mice with genetic modifications of the CREB–BDNF–TrkB pathway have been mostly investigated under basal conditions. Subjecting them to stress-induced depression models may unmask a predisposition to develop depression-like behaviours. Fourth, depression is a multi-genetic disease and inactivation of one gene may not suffice to induce a depressive state. The fact that mice with mutations impairing the CREB–BDNF–TrkB pathway do not show any features of depression challenges the hypothesis that this signalling pathway plays a major role in the pathogenesis of depression. This critical view is supported by further experimental evidence. Contradictory to the concept that stress causes a decreased activity of the CREB–BDNF–TrkB cascade, the forced swim test induces a longlasting increase in CREB phosphorylation in several rat brain areas including the hippocampus.36 Immobilization stress produces a rapid increase in BDNF mRNA and protein in the hypothalamus.37 Chronic unpredictable and uncontrollable stress also evokes an increase in TrkB expression in hippocampal neurons. These results have been interpreted as
Ch 11
7/4/05
3:56 pm
110
Page 110
Bipolar disorder: the upswing in research & treatment
compensatory adaptation to the stress-induced BDNF decrease that could serve to protect the neurons from damage.38 Furthermore, two genetic rat models of depression, the congenital learned-helplessness rats as well as the depressive Flinders strain, exhibit unaltered levels of BDNF in the hippocampus.13,39 Despite conflicting results on the role of the CREB–BDNF–TrkB pathway in the pathogenesis of depression, this signalling cascade seems to be clearly involved in the mechanisms of antidepressive therapy. Thus, chronic antidepressive treatment evokes an increase in BDNF expression.10 This increase is absent in mice with impaired CREB function,30 indicating that CREB is important in mediating the transcriptional effects of antidepressants. On the other hand, mice with reduced BDNF expression (BDNF+/–) as well as mice with impaired TrkB function do not show the normal reduction of behavioural despair seen in wild-type mice following antidepressant treatment, indicating that BDNF-mediated TrkB activation is necessary for at least some of the behavioural effects of antidepressants.20 These findings in transgenic mice together with the direct antidepressive effects of BDNF in mouse models of depression23,25 and the increased BDNF levels in patients on antidepressive medication15 provide cumulative evidence for a central role of this neurotrophin and its receptor in the molecular mechanisms of antidepressive therapy. There are several possible biological mechanisms that could mediate the antidepressant effects of BDNF. On a molecular level, the mitogen-activated protein (MAP) kinase cascade is activated by BDNF–TrkB signalling and induces the plasticity-related transcription factor c-fos. On a cellular level, BDNF is involved in synaptic remodelling and functioning.40,41 Such cellular adaptations may also improve monoaminergic signalling in the forebrain, which is a major participant in the pathogenesis of depression according to the monoaminergic hypothesis. Mice that are compromised in BDNF–TrkB signalling, represent a valuable tool to study molecular and cellular downstream effects of this signalling pathway. The utility of BDNF itself as an antidepressant in humans is limited, owing to the restrictions by the blood–brain barrier. Therefore, pharmacological strategies have to be developed to generate small-molecule agents that increase the expression and promote the release of BDNF more specifically and more efficiently than currently available antidepressants.
Ch 11
7/4/05
3:56 pm
Page 111
Transgenic mouse models for affective disorders
111
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
Holsboer F, Barden N, Antidepressants and hypothalamic–pituitary– adrenocortical regulation. Endocr Rev 1996; 17:187–205. Nemeroff CB, The corticotropin-releasing factor (CRF) hypothesis of depression: new findings and new directions. Mol Psychiatry 1996; 1:336–342. Duman RS, Heninger GR, Nestler EJ, A molecular and cellular theory of depression. Arch Gen Psychiatry 1997; 54:597–606. Altar CA, Neurotrophins and depression. Trends Pharmacol Sci 1999; 20:59–61. Lewin GR, Barde YA, Physiology of the neurotrophins. Annu Rev Neurosci 1996; 19:289–317. Holsboer F, The corticosteroid receptor hypothesis of depression. Neuropsychopharmacology 2000; 23:477–501. Porsolt RD, Animal models of depression: utility for transgenic research. Rev Neurosci 2000; 11:53–58. Willner P, Mitchell PJ, The validity of animal models of predisposition to depression. Behav Pharmacol 2002; 13:169–188. Barbany G, Persson H, Regulation of neurotrophin mRNA expression in the rat brain by glucocorticoids. Eur J Neurosci 1992; 4:396–403. Nibuya M, Morinobu S, Duman RS, Regulation of BDNF and trkB mRNA in rat brain by chronic electroconvulsive seizure and antidepressant drug treatments. J Neurosci 1995; 15:7539–7547. Schaaf MJ, Hoetelmans RW, de Kloet ER, Vreugdenhil E, Corticosterone regulates expression of BDNF and trkB but not NT-3 and trkC mRNA in the rat hippocampus. J Neurosci Res 1997; 48:334–341. Cosi C, Spoerri PE, Comelli MC et al, Glucocorticoids depress activitydependent expression of BDNF mRNA in hippocampal neurones. Neuroreport 1993; 4:527–530. Angelucci F, Aloe L, Vasquez PJ, Mathe AA, Mapping the differences in the brain concentration of brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) in an animal model of depression. Neuroreport 2000; 11:1369–1373. Dowlatshahi D, MacQueen GM, Wang JF, Young LT, Increased temporal cortex CREB concentrations and antidepressant treatment in major depression. Lancet 1998; 352:1754–1755. Chen B, Dowlatshahi D, MacQueen GM et al, Increased hippocampal BDNF immunoreactivity in subjects treated with antidepressant medication. Biol Psychiatry 2001; 50:260–265. Karege F, Perret G, Bondolfi G et al, Decreased serum brain-derived neurotrophic factor levels in major depressed patients. Psychiatry Res 2002; 109:143–148. Nibuya M, Nestler EJ, Duman RS, Chronic antidepressant administration increases the expression of cAMP response element binding protein (CREB) in rat hippocampus. J Neurosci 1996; 16:2365–2372. Duman RS, Vaidya VA, Molecular and cellular actions of chronic electroconvulsive seizures. J ECT 1998; 14:181–193.
Ch 11
7/4/05
3:56 pm
Page 112
112
Bipolar disorder: the upswing in research & treatment
19.
Zetterstrom TS, Pei Q, Grahame-Smith DG, Repeated electroconvulsive shock extends the duration of enhanced gene expression for BDNF in rat brain compared with a single administration. Brain Res Mol Brain Res 1998; 57:106–110. Saarelainen T, Hendolin P, Lucas G et al, Activation of the TrkB neurotrophin receptor is induced by antidepressant drugs and is required for antidepressantinduced behavioural effects. J Neurosci 2003; 23:349–357. Bayer TA, Schramm M, Feldmann N et al, Antidepressant drug exposure is associated with mRNA levels of tyrosine receptor kinase B in major depressive disorder. Prog Neuropsychopharmacol Biol Psychiatry 2000; 24:881–888. Koch JM, Kell S, Hinze-Selch D, Aldenhoff JB, Changes in CREBphosphorylation during recovery from major depression. J Psychiatr Res 2002; 36:369–375. Siuciak JA, Lewis DR, Wiegand SJ, Lindsay RM, Antidepressant-like effect of brain-derived neurotrophic factor (BDNF). Pharmacol Biochem Behav 1997; 56:131–137. Shanks N, Anisman H, Strain-specific effects of antidepressants on escape deficits induced by inescapable shock. Psychopharmacology 1989; 99:122–128. Shirayama Y, Chen AC, Nakagawa S et al, Brain-derived neurotrophic factor produces antidepressant effects in behavioural models of depression. J Neurosci 2002; 22:3251–3261. Gass P, Reichardt HM, Strekalova T et al, Mice with targeted mutations of glucocorticoid and mineralocorticoid receptors: models for depression and anxiety? Physiol Behav 2001; 73:811–825. Rudolph D, Tafuri A, Gass P et al, Impaired fetal T cell development and perinatal lethality in mice lacking the cAMP response element binding protein. Proc Natl Acad Sci USA 1998; 95:4481–4486. Blendy JA, Kaestner KH, Schmid W et al, Targeting of the CREB gene leads to up-regulation of a novel CREB mRNA isoform. EMBO J 1996; 15:1098–1106. Graves L, Dalvi A, Lucki I et al, Behavioural analysis of CREB alphadelta mutation on a B6/129 F1 hybrid background. Hippocampus 2002; 12:18–26. Conti AC, Cryan JF, Dalvi A et al, cAMP response element-binding protein is essential for the upregulation of brain-derived neurotrophic factor transcription, but not the behavioral or endocrine responses to antidepressant drugs. J Neurosci 2002; 22:3262–3268. Newton SS, Thome J, Wallace TL et al, Inhibition of cAMP response elementbinding protein or dynorphin in the nucleus accumbens produces an antidepressant-like effect. J Neurosci 2002; 22:10883–10890. MacQueen GM, Ramakrishnan K, Croll SD et al, Performance of heterozygous brain-derived neurotrophic factor knockout mice on behavioral analogues of anxiety, nociception, and depression. Behav Neurosci 2001; 115:1145–1153. Chourbaji S, Hellweg R, Brandis D et al, Mice with reduced brain-derived neurotrophic factor expression show decreased choline acetyltransferase activity, but regular brain monoamine levels and unaltered emotional behavior. Mol Brain Res 2004; 121:28–36.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29. 30.
31.
32.
33.
Ch 11
7/4/05
3:56 pm
Page 113
Transgenic mouse models for affective disorders 34.
35.
36.
37.
38. 39. 40. 41.
113
Rios M, Fan G, Fekete C et al, Conditional deletion of brain-derived neurotrophic factor in the postnatal brain leads to obesity and hyperactivity. Mol Endocrinol 2001; 15:1748–1757. Zörner B, Wolfer DP, Brandis D et al, Forebrain-specific trkB-receptor knockout mice: behaviorally more hyperactive than ‘depressive’. Biol Psychiatry 2003; 54:972–982. Bilang-Bleuel A, Rech J, De Carli S et al, Forced swimming evokes a biphasic response in CREB phosphorylation in extrahypothalamic limbic and neocortical brain structures in the rat. Eur J Neurosci 2002; 15:1048–1060. Rage F, Givalois L, Marmigere F et al, Immobilization stress rapidly modulates BDNF mRNA expression in the hypothalamus of adult male rats. Neuroscience 2002; 112:309–318. Nibuya M, Takahashi M, Russell DS, Duman RS, Repeated stress increases catalytic TrkB mRNA in rat hippocampus. Neurosci Lett 1999; 267:81–84. Vollmayr B, Henn FA, Learned helplessness in the rat: improvements in validity and reliability. Brain Res Brain Res Protoc 2001; 8:1–7. Poo MM, Neurotrophins as synaptic modulators. Nat Rev Neurosci 2001; 2:24–32. Manji HK, Quiroz JA, Sporn J et al, Enhancing neuronal plasticity and cellular resilience to develop novel, improved therapeutics for difficult-to-treat depression. Biol Psychiatry 2003; 53:707–742.
Ch 11
7/4/05
3:56 pm
Page 114
Ch 12
7/4/05
3:57 pm
Page 115
chapter 12
Is the hypothalamic– pituitary–adrenal axis at last paying dividends? David A Cousins and Allan H Young
Psychiatry is the medical specialty concerned with pathological human psychology. Throughout its history, attempts have been made to explain psychological experiences in terms of neurophysiology and neuropathology. Such practice is probably overly reductionistic, as it is not legitimate to explain all psychological problems in terms of biological processes.1 Even if the mind–brain debate concludes in favour of materialism, the subjective experience of consciousness will remain. What then is the role of biological psychiatry? Biological psychiatry has two main purposes: to understand the pathophysiology of psychiatric illness, and to use that understanding to develop therapeutic strategies. Some of the progress made in these regards will be the subject of this chapter, with specific reference to the hypothalamic–pituitary–adrenal (HPA) axis in bipolar affective disorder.
Pathophysiology of bipolar affective disorder There is an overwhelming amount of information in this field that has yet to fit into a cohesive paradigm. A simple overarching model would incorporate the influence of genes, environment and stress, and the ongoing effects of the illness itself. Stress, in particular its main substrate system the HPA axis, has the potential to integrate many of our findings. We will first review the structure and function of the HPA axis, the evidence for its disturbance in bipolar affective disorder and finally the consequences of that disturbance.
Ch 12
7/4/05
3:57 pm
116
Page 116
Bipolar disorder: the upswing in research & treatment
The HPA axis The structural and functional organization of the HPA axis is well established and is represented in Figure 12.1. In response to stress, neurosecretory cells in the paraventricular nucleus of the hypothalamus secrete corticotrophin releasing hormone (CRH) into the microportal circulatory system
Midbrain
Amygdala
Hippocampus
±ve
±ve
+ve
–ve Hypothalamus –ve CRH +ve
Pituitary –ve ACTH +ve
Adrenals
Cortisol
Figure 12.1 The hypothalamic–pituitary–adrenal axis. Purple triangle, glucocorticoid receptors; CRH, corticotrophin releasing hormone; ACTH, adrenocorticotrophic hormone.
Ch 12
7/4/05
3:57 pm
Page 117
Is the hypothalamic–pituitary–adrenal axis at last paying dividends?
117
of the pituitary stalk. CRH acts on the anterior pituitary, which in turn releases adrenocorticotrophic hormone (ACTH) into the systemic circulation. ACTH governs the release of cortisol from the adrenal cortex. Cortisol is the end product of the HPA axis and has numerous central and peripheral effects, including completion of a feedback loop. The HPA axis is highly regulated by this feedback, and by neuronal inputs to the hypothalamus from a number of brain regions (amygdala, hippocampus and certain midbrain nuclei). It is through these neuronal inputs that the system responds to both physical and psychological stressors. At a cellular level, the effects of cortisol are mediated through intracellular glucocorticoid receptors, of which there are two subtypes: mineralocorticoid receptor (MR) and glucocorticoid receptor (GR). Activated receptors move from the cytosol to the nucleus and interact with transcription factors or bind to specific DNA, thus promoting the expression of various genes. GRs are ubiquitously distributed throughout the body and brain, MRs less so, being found in the kidney and the limbic system. The relative contribution of the receptor subtypes in the regulation of the HPA axis remains unclear. MRs have a high affinity for cortisol and aldosterone; GRs have low affinity for cortisol but avidly bind synthetic steroids. It is proposed that MRs regulate basal cortisol secretion when hormonal levels are low, and that GRs become increasingly important as levels rise and the MRs become saturated. GRs may therefore be pivotal in the response to circadian rhythms and to stress. The HPA axis is vulnerable in a way that mirrors our understanding of patients’ vulnerabilities. There are genetic effects; genes control the system at various different levels including hormone production and receptor expression. The potential interaction from multiple genes of small effect is large and may predispose to, and perpetuate, abnormal stress reactions. There are early environmental effects; it is well established that the HPA axis is open to influence by early life events in utero and the postnatal period. There are precipitating factors; both physical and psychological stress results in increased production of cortisol. This is useful in times of acute stress but is maladaptive in the long term at cellular and system levels. There are maintaining factors; serotonergic and dopaminergic neurotransmitter systems are thought to be modulated by cortisol, and ongoing hypercortisolaemia may have a detrimental effect upon neuronal function. Increased frequency of episodes of illness with ongoing time and stress (eventually entering rapid cycling) could potentially be mediated through the HPA axis.2
Ch 12
7/4/05
3:57 pm
118
Page 118
Bipolar disorder: the upswing in research & treatment
HPA dysfunction in bipolar affective disorder Abnormalities have been demonstrated and replicated at all levels of the HPA axis in patients suffering from affective disorders. Structurally there is enlargement of the pituitary and adrenal glands. Functionally there are increased levels of cortisol (present in plasma, urine and cerebrospinal fluid), an exaggerated cortisol response to ACTH, and hypersecretion of CRH. This pattern of abnormalities is suggestive of an impaired feedback loop, possibly resulting from GR abnormalities such as decreased receptor number and/or function. Such abnormalities have been confirmed in postmortem studies.3 The functional integrity of the receptors can be assessed using the dexamethasone suppression test (DST), with reports of cortisol non-suppression in affective disorders supporting a primary GR abnormality.4 The ability of the DST reliably to distinguish between controls and patients with bipolar disorder may be dramatically increased by being combined with a CRH infusion. This dexamethasone (DEX)/CRH test is proving promising, with significant abnormalities in the HPA axis being demonstrated in patients with bipolar affective disorder.2 Importantly, these abnormalities are present during both illness and recovery.
Cognitive impairment and hypercortisolaemia Patients with affective disorder commonly report impairment of cognitive function, typically in terms of poor attention, concentration and memory. Baddeley proposed a now popular model of working memory in which a central executive (the ‘attentional controller’) co-ordinates two slave systems, the ‘phonological loop’ and the ‘visuospatial scratch pad’. This system is probably served by a wide variety of brain structures including the frontal lobes and hippocampal formations. It has been suggested that dysfunction of the central executive would manifest as impaired set shifting, planning, verbal fluency and response inhibition. Recently, systematic evaluations of cognitive function in patients with affective disorders have identified specific deficits in executive function, learning and memory tasks.5 These findings were thought to be independent of drugs effects. Although these deficits can improve on remission of affective symptoms, the impairment in executive function has been shown to persist in a cohort of patients with bipolar affective disorder, prospectively verified as being euthymic.5,6 This challenges the view that patients with severe mood disorders make a full interepisode recovery. Animal studies have demonstrated that chronic administration of glucocorticoids results in abnormalities of learning and memory, with associated
Ch 12
7/4/05
3:57 pm
Page 119
Is the hypothalamic–pituitary–adrenal axis at last paying dividends?
119
atrophy of neurons in the hippocampal formation.7 Observation of patients with pathologically high cortisol levels, such as in Cushing’s disease, provides a natural experiment. It is now established that such patients have significant cognitive impairments.8 More objective evidence can be drawn from experimental models using normal volunteers. Chronic administration of hydrocortisone to normal male volunteers results in cognitive impairments that appear to be mediated in part via the effects on the frontal lobe. Further, although such subjects have faster reaction times, they make more errors in keeping with observations of behaviour at times of stress. Recently, it has been demonstrated that the frontal lobes are adversely affected by cortisol and suffer a pattern of degeneration similar to that occurring in the hippocampus. It is therefore feasible that the neurodegenerative effects of cortisol may underlie some of the cognitive deficits observed in patients with severe affective disorders.9
Developing therapeutic strategies There is substantial evidence that HPA axis dysfunction is important in the pathophysiology of mood disorders and the associated cognitive deficits. Modulation of this dysfunction and counteraction of the effects of hypercortisolaemia may provide potential treatments for mood disorders. Antidepressant effects have been demonstrated with agents possessing antiglucocorticoid effects,10 steroid synthesis inhibitors11 and pituitary glucocorticoid receptor agonists.12
Glucocorticoid receptor antagonists As previously discussed, one possible explanation for the HPA axis dysfunction seen in bipolar affective disorder is a primary abnormality of GR number and/or function. A reduction in GR number and/or function is thought to underlie the hypercortisolaemia through impaired negative feedback. Naturally, the question arises of how the hypercortisolaemia exerts its detrimental effect when its mediator, the glucocorticoid receptor, is not functioning. There are a number of possible explanations. First, the hypercortisolaemia may be sufficient to override the impaired GR function with the HPA system re-establishing equilibrium at higher cortisol levels. Second, the receptors regulating HPA feedback (hippocampus and hypothalamus) may be dysfunctional, but other receptors (frontal lobes, etc.) may be intact. Thus, the normal GRs in some brain regions would be exposed to the
Ch 12
7/4/05
3:57 pm
120
Page 120
Bipolar disorder: the upswing in research & treatment
detrimental effects of high cortisol levels whilst the receptors governing feedback would be unable to respond. Finally, it is possible that the adverse effects of cortisol are mediated through secondary mechanisms or nonreceptor-mediated processes. A novel therapeutic strategy that is attracting increasing attention is the use of GR antagonists such as mifepristone (Mifeprex, RU-486).13 Administration of GR antagonists is thought to result in an immediate antiglucocorticoid effect followed by a compensatory upregulation of GR numbers. This upregulation could lead to improved negative feedback, effectively ‘resetting’ the HPA axis. Animal studies have confirmed such increases in receptor number following administration of mifepristone.7 GR antagonists have been shown to be beneficial in severe depression, and that only a brief period of treatment may be adequate to restore normal HPA function.13 We have recently conducted an exploratory trial investigating the effects of mifepristone as a treatment of bipolar affective disorder. We hypothesized that the GR antagonist would improve neurocognitive functioning and reduce depressive symptoms in our patients. To test this hypothesis, we undertook a randomized, double-blind, placebo-controlled trial with a cross-over design in which mifepristone was administered as an adjunctive agent. Twenty patients, aged 18–65 years, with a diagnosis of bipolar affective disorder (confirmed using the Structured Clinical Interview for DSMIV14) were recruited. All patients had residual depressive symptoms. Patients’ medication had been unchanged for 6 weeks prior to entry, and remained so throughout the study. Patients were randomly assigned to receive either mifepristone 600 mg/day for 7 days or placebo, administered in a double-blind manner. At day 21, the groups crossed over and received the alternative treatment. Neurocognitive function was assessed at baseline and 21 days after each treatment. Mood ratings were performed at baseline and at weekly intervals thereafter. Fourteen days after the mifepristone administration phase, assessment of neurocognitive function revealed a significant improvement compared to baseline. Patients showed a reduced error rate on the spatial working memory task of the magnitude of 20%. The degree of improvement correlated positively with baseline cortisol output prior to mifepristone administration. Patients also improved in terms of verbal fluency and spatial recognition memory. No changes were observed at any time points following the placebo phase. Similarly, mifepristone was associated with improvements in depression rating scales and the Brief Psychiatric Rating Scale (BPRS) at 14 days. No change was seen in the placebo phase. These results support
Ch 12
7/4/05
3:57 pm
Page 121
Is the hypothalamic–pituitary–adrenal axis at last paying dividends?
121
the hypothesis that mifepristone, a GR antagonist, improves neurocognitive function and has a putative antidepressant action in patients with bipolar affective disorder.
Conclusion Abnormalities in the function of the HPA axis, with resultant hypercortisolaemia, have been consistently demonstrated and replicated in bipolar affective disorder. These abnormalities can persist despite recovery from depression. Neurocognitive impairments, in particular abnormalities on tests of frontal lobe function, are present in patients with bipolar affective disorder. Similarly, these can persist into euthymia. Hypercortisolaemia is associated with cognitive impairment and neurodegeneration, as demonstrated in animal studies and normal human volunteers. It is reasonable to suggest that the disruption in mood and cognition seen in bipolar patients could arise, in part, from HPA axis dysfunction. The HPA axis can be successfully manipulated with good therapeutic effect, such that GR antagonists may become a viable treatment option in the future. Large-scale trials will be required to confirm this.
References 1. 2. 3.
4.
5.
6. 7.
Fish F, Clinical Psychopathology: Signs and Symptoms in Psychiatry. John Wright and Sons: Bristol, 1967. Watson S, Gallagher P, Ritchie JC et al, Hypothalamic–pituitary–adrenal axis function in patients with bipolar disorder. Br J Psychiatry 2004; 18:496–503. Webster MJ, O’Grady J, Orthmann J, Weickert CS, Decreased glucocorticoids receptor mRNA levels in individuals with depression, bipolar disorder and schizophrenia. Schizophr Res 2000; 41:111–112. Zhou DF, Shen YC, Shu LN, Lo HC, Dexamethasone suppression test and urinary MHPG X SO4 determination in depressive disorders. Biol Psychiatry 1987; 22:883–891. Thompson JM, Gray JM, Hughes JH et al, A component process analysis of working memory dysfunction in bipolar affective disorder. Bipolar Disord 2001; 3:60 (abstr). Ferrier IN, Thompson JM, Cognitive impairment in bipolar affective disorder: implications for the bipolar diathesis. Br J Psychiatry 2002; 180:293–295. Lupien SJ, McEwan BS, The acute effects of corticosteriods on cognition: integration of animal and human model studies. Brain Res Bain Res Rev 1997; 24:1–27.
Ch 12
7/4/05
3:57 pm
122 8. 9. 10.
11.
12.
13.
14.
Page 122
Bipolar disorder: the upswing in research & treatment Wolkowitz OM, Reus VI, Weingartner H et al, Cognitive effects of corticosteroids. Am J Psychiatry 1990; 147:1297–1303. Sapolsky RM, Krey LC, McEwan BS, The neuroendocrinology of stress and aging: the glucocorticoid cascade hypothesis. Endocr Rev 1986; 7:284–301. Wolkowitz OM, Reus VI, Keebler R et al, Double-blind treatment of major depression with dehydroepiandrosterone. Am J Psychiatry 1999; 156: 646–649. Ravaris CL, Nelpa I, Huang M et al, Effect of ketoconazole on a hypophysectomized, hypercortisolemic, psychotically depressed woman. Arch Gen Psychiatry 1988; 45:966–967. Arana GW, Santos AB, Laraia MT et al, Dexamethasone for the treatment of depression: a randomized, placebo-controlled, double-blind trial. Am J Psychiatry 1995; 152:265–267. Murphy BE, Filipini D, Ghadirian JL, Possible use of glucocorticoid receptor antagonists in the treatment of major depression: preliminary results using RU486. J Psychiatry Neurosci 1993; 18:209–213. First MB, Spitzer RL, Gibbon M et al, Structured Clinical Interview for DSMIV Axis I Disorders (SCID). New York State Psychiatric Institute, Biometrics Research: New York, 1999.
Ch 13
7/4/05
3:58 pm
Page 123
chapter 13
Stress on the brain: neuropathology and cortisol dysregulation in bipolar disorder David Cotter
Introduction Considering the prevalence and importance of major depressive disorder (MDD) and bipolar disorder (BPD), it is surprising that we have only very recently started to look carefully at the cerebral cellular architecture of these disorders. Structural imaging investigations of living subjects with mood disorders can point us in the direction of pathology. These imaging investigations have shown a variety of important changes. In the case of MDD, the subgenual anterior cingulate cortex, the hippocampus and the frontal cortex have been shown to be reduced in volume. In BPD, the hippocampal volume has also been shown to be reduced, and the amygdala may be increased in size. Functional imaging investigations have also provided clues to the neuronal circuits implicated in mood disorders and point to involvement of the corticolimbic and the corticostriatal networks.
Neuropathology of major depression and bipolar disorder Neuronal size correlates with the extent of a neuron’s efferent and afferent connections, and consequently reduced neuronal size points to functional and/or structural dysconnectivity. Neuronal number is important, as a deficit would have functional implications and may reflect abnormal cell
Ch 13
7/4/05
3:58 pm
124
Page 124
Bipolar disorder: the upswing in research & treatment
death or apoptosis, such as occurs in some neurodegenerative disorders, or may reflect altered neurogenesis. In MDD and BPD, there is evidence for reduced neuronal size in the anterior cingulate cortex, the dorsolateral prefrontal cortex and the orbitofrontal cortex. There may also be a reduced density of larger neurons in MDD and BPD, which probably equates with a reduced median or mean neuronal size. Reduced density of smaller interneurons in BPD has also been described. Deficits in glial cell numbers and density are observed in MDD, and to a lesser extent in BPD. This is exciting, because the finding is relatively new (the main finding and the replications have occurred within the past 4 years) and potentially important in terms of providing new therapeutic strategies. Glial cells have traditionally been viewed as neuronal supporting cells in the central nervous system – so-called ‘mind glue’ with their primary roles in glutamatergic neurotransmission, glucose metabolism and neurotrophic support largely ignored. However, over the past few years several independent studies have shown a cortical glial cell deficit in affective disorders, and these observations have come at a time when the important role of glial cells in normal cortical function has been fundamentally re-evaluated. The orbitofrontal cortex, anterior cingulate cortex and dorsolateral prefrontal cortex are the regions in which these changes have been observed. This is in keeping with the functional and structural imaging studies, and indeed the neuropsychological investigations that have already implicated these brain regions. Glial cells are composed of three different cell types (microglia, oligodendroglia and astrocytes) and it is not yet clear which of these is responsible for the observed deficit in MDD. Support for the possibility that astrocyte deficits are responsible for the glial cell changes have been provided by proteomics investigations based on the Stanley Foundation Brain series, although others have also found evidence for reductions in markers of oligodendroglia. Synaptic changes have also been observed in BPD and MDD. However, because postmortem tissue is not always suitable for quantitation of dendrites and synapses, they have been evaluated through a number of methods including direct assessment of silver-stained dendrites and dendritic spines, immunocytochemical staining of dendrites and quantitation of gene products localized to synaptic and presynaptic compartments. Three synaptic proteins (synaptophysin, complexin I and growth-associated protein-43 (GAP-43)) have been shown to be reduced in the anterior cingulate cortex in BPD, with only complexin II reduced in MDD. In the prefrontal cortex in MDD, GAP-43 and synaptophysin are unaltered and there are no studies yet of BPD. Studies of the hippocampus have shown reduced levels of synaptic
Ch 13
7/4/05
3:58 pm
Page 125
Stress on the brain: neuropathology and cortisol dysregulation
125
associated protein-25 and complexin I and II in BPD, with no changes in MDD. The consensus of these studies is that synaptic pathology is present in mood disorders in the cortical limbic regions and that the changes are more marked in BPD than MDD. Magnetic resonance imaging (MRI) investigations have shown a strong association between subcortical white matter hyperintensities (WMH) on T2-weighted images and both MDD and BPD. These lesions are observed chiefly in the deep white matter, around the basal ganglia and in the periventricular region. In MDD these lesions are more common in the elderly with vascular disease. In both elderly MDD and BPD subjects, WMHs confer a poorer prognosis. Diffusion tensor imaging has shown that WMH reflect damage to white matter tracts, leading to the suggestion that the mood disorder in these subjects is due to interruption of the frontal cortical–subcortical connections. While the neuropathological basis of these lesions has not yet been characterized, it is likely that mood disorder of this type is neuropathologically distinct.
Evidence for a common neuropathology in major depression, bipolar disorder and schizophrenia There are some important similarities in the neuropathology of schizophrenia, MDD and BPD, which suggest that a common process of change is involved in each disorder. Macroscopic neuroanatomical investigations of brain pathology in schizophrenia, BPD and MDD show differences that are generally quantitative rather than qualitative. For example, ventricular dilatation and reduced hippocampal and frontal brain volumes are seen in schizophrenia, but they are also present to a lesser degree in MDD and BPD. The single main departure from this pattern is that the volume of the amygdala may be specifically enlarged in BPD, possibly because of drug treatment. Microscopically, reductions in dendritic spine density, neuronal size, and synaptic proteins have been described in mood disorders as well as in schizophrenia. More recently it has become apparent that glial cell loss may be a feature of MDD, BPD and schizophrenia, depending, possibly, on the presence of co-existing affective symptoms and on which region of the brain is investigated. This similar pattern of changes in cortical cellular architecture in schizophrenia and mood disorders suggests that a common pathophysiology may underlie aspects of these psychiatric diseases. What aspects of illness, common to MDD, BPD and schizophrenia, could cause changes in keeping
Ch 13
7/4/05
3:58 pm
126
Page 126
Bipolar disorder: the upswing in research & treatment
with the known cellular changes described above? Glucocorticoid-related neurotoxicity is a candidate that needs to be considered.
A role for glucocorticoids in the neuronal changes of mood disorders and schizophrenia? There is substantial evidence that hyperactivity of the hypothalamic– pituitary-adrenal (HPA) axis is involved in the pathogenesis of mood disorder. Impaired function of the glucocorticoid receptor (GR) and subsequent altered feedback inhibition by endogenous glucocorticoids probably represents the mechanism by which the HPA axis is activated in depression. In contrast, to date there has been no firm evidence that HPA hyperactivity is part of the pathogenesis of schizophrenia. This may be because few studies have assessed subjects during the acute phase of the schizophrenic illness which is the period of the illness most likely to be associated with a stressrelated elevation of glucocorticoids. Recently, however, reduced GR gene expression has been described in the frontal cortex in schizophrenia and major depression, providing the first firm evidence that HPA axis abnormalities are a feature of these disorders. There are several other lines of investigation that support the view that glucocorticoid-related neurotoxicity may be implicated in depression and schizophrenia. First, in vitro investigations have shown that high levels of glucocorticoid hormones result in reduced neuronal volume and dendritic arborization, and these changes have been observed in both disorders. Second, elevated plasma glucocorticoid levels are associated with hippocampal volume reductions in MDD, post-traumatic stress disorder, Cushing’s disease and normal aging, and such reductions have been observed in the phase of a first psychotic episode. Third, the functional effect of glucocorticoids on reducing hippocampal glial cell activation and proliferation mirrors the glial deficit observed in MDD, BPD and possibly schizophrenia. Consequently, the glial deficit found in these disorders may also relate to glucocorticoid effects.
Conclusion Evidence is accumulating that there are brain changes occurring during and possibly after the period of the first presentations of BPD and schizophrenia, i.e. changes that are not developmental in the traditional sense.
Ch 13
7/4/05
3:58 pm
Page 127
Stress on the brain: neuropathology and cortisol dysregulation
127
Furthermore, these changes are not specific to schizophrenia, in terms of either macroscopic or microscopic brain structure, for they are also present to a generally milder degree in subjects with BPD and MDD, and they are in keeping with glucocorticoid-related brain changes.
These brain
changes may be epiphenomena secondary to stress-related changes in glucocorticoid hormones and not primary pathogenetic pathways. Nevertheless, they could have crucial clinical effects through diminishing neuronal and cortical function and so complicate recovery from the primary illness. These changes may possibly be reversed by therapies that protect from glucocorticoid-related neurotoxicity or which act to promote neuroprotective cell-signalling pathways. It will now be important to understand these illnesses in terms of both early and late events that involve both developmental and atrophic processes.
Bibliography Altshuler LL, Bartzokis G, Grieder T et al, Amygdala enlargement in bipolar disorder and hippocampal reduction in schizophrenia: an MRI study demonstrating neuroanatomic specificity. Arch Gen Psychiatry 1998; 55:663–664. Cotter D, Mackay D, Landau S et al, Glial cell loss and reduced neuronal size in the anterior cingulate cortex in major depressive disorder. Arch Gen Psychiatry 2001; 58:545–553. Cotter DR, Pariante CM, Everall IP, Glial cell abnormalities in major psychiatric disorders: the evidence and implications. Brain Res Bull 2001; 55:585–595. Drevets WC, Neuroimaging and neuropathological studies of depression: implications for the cognitive–emotional features of mood disorders. Curr Opin Neurobiol 2001; 11:240–249. Eastwood SL, Harrison PJ, Synaptic pathology in the anterior cingulate cortex in schizophrenia and mood disorders. A review and a Western blot study of synaptophysin, GAP-43, and the complexins. Brain Res Bull 2001; 55:569–578. Harrison PJ, The neuropathology of schizophrenia. A critical review of the data and their interpretation. Brain 1999; 122:593–624. Harrison PJ, The neuropathology of primary mood disorder. Brain 2002; 125:1428–1449. Ismail K, Murray RM, Wheeler MJ, O’Keane V, The dexamethasone suppression test in schizophrenia. Psychol Med 1998; 28:311–317. Lawrie SM, Whalley H, Byrne M et al, Brain structure change and psychopathology in subjects at high risk of schizophrenia. Schizophr Res 2000; 41:11.
Ch 13
7/4/05
3:58 pm
128
Page 128
Bipolar disorder: the upswing in research & treatment
McCarley RW, Wible CG, Frumin M et al, MRI anatomy of schizophrenia. Biol Psychiatry 1999; 45:1099–1119. Pantellis C, Velakoulis D, Suckling J et al, Left medial temporal lobe volume reduction occurs during the transition from high risk to first episode psychosis. Schizophr Res 2000; 41:35. Pariante CM, Miller AH, Glucocorticoid receptors in major depression: relevance to pathophysiology and treatment. Biol Psychiatry 2001; 49:391–404. Pearlson GD, Structural and functional brain changes in bipolar disorder: a selective review. Schizophr Res 1999; 39:133–140; discussion 162. Rajkowska G, Miguel-Hidalgo JJ, Wei J, Morphometric evidence for neuronal and glial prefrontal cell pathology in major depression. Biol Psychiatry 1999; 45:1085–1098. Rappaport JL, Giedd JN, Blumenthal J et al, Progressive cortical change during adolescence in childhood-onset schizophrenia. A longitudinal magnetic resonance imaging study. Arch Gen Psychiatry 1999; 56:649–654. Rosoklija G, Toomayan G, Ellis SP et al, Structural abnormalities of subicular dendrites in subjects with schizophrenia and mood disorders. Arch Gen Psychiatry 2000; 57:349–356. Sapolsky R, The possibility of neurotoxicity in the hippocampus in major depression: a primer on neuron death. Biol Psychiatry 2000; 48:755–765. Webster MJ, O’Grady J, Orthmann C, Weickert C, Decreased glucocorticoid receptor mRNA levels in individuals with depression, bipolar disorder and schizophrenia. Schizophr Res 2000; 41:111. Weinberger DR, Implications of normal brain development for the pathogenesis of schizophrenia. Arch Gen Psychiatry 1987; 44:660–669.
Ch 14
7/4/05
3:58 pm
Page 129
chapter 14
Cortisol in Chicago (from crime of passion to celebrity headline) Carmine M Pariante
Cortisol, the stress hormone that is life-saving in a variety of circumstances, is considered a ‘criminal’ by part of the scientific community. It has been accused of killing the brain, inducing hippocampal atrophy, blocking neurogenesis, causing cognitive abnormality, even of making one depressed. This has been a real press campaign against the poor hormone! In the 2003 award movie, Chicago, it takes a very good lawyer to get two beautiful murderers out of prison and ‘rehabilitated’ in the newspapers. Clearly, cortisol also needs a good lawyer. So, who is the defendant? The hypothalamic–pituitary–adrenal (HPA) axis, producing cortisol, is the main stress hormonal response system. As shown in Figure 14.1, HPA axis activity is governed by the secretion of corticotrophin releasing hormone (CRH) from the hypothalamus. CRH activates the secretion of adrenocorticotrophic hormone (ACTH) from the pituitary. ACTH, in turn, stimulates the secretion of glucocorticoids (cortisol in humans) from the adrenal glands, two small organs situated on top of the kidneys. Glucocorticoids interact with their corticosteroid receptors − the glucocorticoid receptor (GR) and the mineralocorticoid receptor (MR) − in almost every tissue in the body, and the best-known physiological effect is the regulation of energy metabolism. By binding to corticosteroid receptors in the brain, glucocorticoids also inhibit the secretion of CRH from the hypothalamus and ACTH from the pituitary (negative feedback). There is a variety of data produced over the past 30 years showing that the most severe phases of mental disorders such depression, mania and schizophrenia are associated with hyperactivity of the HPA axis and
Ch 14
7/4/05
3:58 pm
130
Page 130
Bipolar disorder: the upswing in research & treatment
increased concentrations of cortisol in the bloodstream. This hyperactivity has been demonstrated by findings such as the lack of suppression of cortisol secretion by dexamethasone, the increased level of CRH in the brain, the increased volume of the anterior pituitary gland, the increased volume of the adrenal gland and, of course, the increased level of the circulating cortisol.1–3 Indeed, we have recently demonstrated increased volume of the pituitary gland in two independent samples of patients at their first episode of psychosis, including psychotic depression, psychotic mania and schizophrenia.4,5 In fact, during a situation of stress, the pituitary cells producing ACTH (and other hormones participating in the stress response) proliferate, and the pituitary becomes macroscopically larger. In our first study,4 in an Australian sample, the first-episode subjects had pituitary volumes that were 10% larger than those of controls. However, because they were receiving antipsychotics, we could not exclude the hypothesis that the pituitary hyperplasia could be related, at least in part, to increased function of prolactin-producing cells activated by neuroleptic treatment. However, in the second study,5 conducted in London as part of the AESOP First-Onset Psychosis Study, this effect was present even in neuroleptic-free first-
Amygdala
Midbrain
Hippocampus GRs/MRs
–/+
Hypothalamus GRs
– –
CRF AVP
Circulating cortisol Pituitary GRs
–
ACTH Adrenal cortex
Figure 14.1 Schematic diagram of the hypothalamic–pituitary–adrenal axis, describing regulation and negative feedback (–) of cortisol via glucocorticoid receptors (GRs) and mineralocorticoid receptors (MRs). CRF, corticotrophin releasing factor; AVP, arginine vasopressin; ACTH, adrenocorticotrophic hormone.
Ch 14
7/4/05
3:58 pm
Page 131
Cortisol in Chicago
131
episode patients (> 15%), although it was more evident in patients receiving typical antipsychotics (> 30%). Having demonstrated that the severe phases of affective and psychotic disorders are characterized by HPA axis hyperactivity, we argue that there is no conclusive evidence that these elevated levels of cortisol can cause the psychiatric symptoms or induce the brain functional and anatomical abnormalities that are described in these patients. Indeed, we claim that the opposite is true: that patients have a hyperactive HPA axis as a compensatory mechanism, because their brain is resistant to the effects of circulating cortisol. The accepted explanation for HPA axis hyperactivity in depression and during stress (and possibly in psychosis) is reduced function of the GR and the MR that mediate the inhibition of CRH and ACTH secretion.3,6 According to this model, because of the reduced function of these receptors, the circulating cortisol is unable to inhibit HPA axis activity successfully (‘glucocorticoid resistance’). Consistent with this model, antidepressants (including lithium) directly increase the function of corticosteroid receptors in the brain, thus increasing the effects of cortisol on the brain, enhancing the negative feedback and reducing HPA axis activity.3,7,8 The reduction of GR levels in the brains of patients with schizophrenia and bipolar disorder,9,10 and the reduction of GR function in patients with depression and in subjects experiencing chronic stress,3,6 all suggest that elevated plasma cortisol levels represent a compensatory strategy. Moreover, recent studies have indicated that levels of cortisol in the brain of animals and humans are regulated by efflux systems at the blood–brain barrier, that are influenced by psychotropic drugs.7,11 For example, we have shown that only 20% of circulating cortisol can access the brain in the guinea pig, an animal that, like humans, has cortisol as the main stress hormone.7 This indicates that peripheral cortisol levels, as often assumed in studies, may not necessarily dictate cerebral levels. Furthermore, our work suggests that a mechanism by which antidepressants increase cortisol effects on the brain is by regulating these transporters and increasing the access of cortisol to the brain.7 Finally, further evidence ‘defending’ cortisol comes from studies using HPA axis manipulation to induce antidepressant effects. In fact, treatment with GR and MR agonists, including cortisol, has antidepressant effects in humans,12–14 again suggesting that these patients do not experience ‘too much cortisol’ in their brain. Surprisingly, even the therapeutic effects of RU-486, a GR antagonist, in the treatment of psychotic depression15 and bipolar disorder16 could be explained by the fact that this drug completely blocks the negative feedback, and thus causes a persistent elevation of cortisol levels and therefore an increase in cortisol signal in the brain.
Ch 14
7/4/05
3:58 pm
132
Page 132
Bipolar disorder: the upswing in research & treatment
Cortisol
Figure 14.2 Who is innocent?
Clearly, whether patients with major depression, bipolar disorder or schizophrenia have elevated or lowered activation of the corticosteroid receptors in the brain is yet to be conclusively elucidated. Nevertheless, we believe that the putative ‘neurotoxic’ effects of cortisol have been undeservedly emphasized. Defending cortisol is not only about getting an innocent out of Chicago jails (see Figure 14.2) but also about following all the available leads – in all directions – to understand the pathophysiology of mental disorders.17,18
Acknowledgements Our research is funded by the UK Medical Research Council (MRC), the National Alliance for Research on Schizophrenia and Depression (NARSAD), the Rosetrees Trust and the Psychiatry Research Trust.
References 1. 2.
Pariante CM, Depression, stress and the adrenal axis. J Neuroendocrinol 2003; 15:811–812. Cotter D, Pariante CM, Stress and the progression of the developmental hypothesis of schizophrenia. Br J Psychiatry 2002; 181:363–365.
Ch 14
7/4/05
3:58 pm
Page 133
Cortisol in Chicago 3. 4. 5.
6.
7.
8.
9.
10.
11. 12. 13.
14.
15. 16.
17.
18.
133
Pariante CM, Miller AH, Glucocorticoid receptors in major depression: relevance to pathophysiology and treatment. Biol Psychiatry 2001; 49:391–404. Pariante CM, Vassilopoulou K, Velakoulis D et al, Pituitary volume in psychosis. Br J Psychiatry 2004; 185:5–10. Pariante C, Brudaglio F, Danese A et al, Increased pituitary volume in patients of the AESOP First-Onset Psychosis Study. Schizophr Res 2004; 67(Suppl 1):99–100. Raison CL, Miller AH, When not enough is too much: the role of insufficient glucocorticoid signaling in the pathophysiology of stress-related disorders. Am J Psychiatry 2003; 160:1554–1565. Pariante CM, Thomas SA, Lovestone S et al, Do antidepressants regulate how cortisol affects the brain? 2003 Curt Richter Award Paper. Psychoneuroendocrinology 2004; 29:423–447. Pariante CM, Papadopoulos AS, Poon L et al, Four days of citalopram increase suppression of cortisol secretion by prednisolone in healthy volunteers. Psychopharmacology (Berl) 2004; 177:200–206. Xing GQ, Russell S, Webster MJ, Post RM, Decreased expression of mineralocorticoid receptor mRNA in the prefrontal cortex in schizophrenia and bipolar disorder. Int J Neuropsychopharmacol 2004; 7:143–153. Webster MJ, Knable MB, O’Grady J et al, Regional specificity of brain glucocorticoid receptor mRNA alterations in subjects with schizophrenia and mood disorders. Mol Psychiatry 2002; 7:985–94, 924. de Kloet ER, Vreugdenhil E, Oitzl MS, Joels M, Brain corticosteroid receptor balance in health and disease. Endocr Rev 1998; 19:269–301. Dinan TG, Lavelle E, Cooney J et al, Dexamethasone augmentation in treatment-resistant depression. Acta Psychiatr Scand 1997; 95:58–61. Bouwer C, Claassen J, Dinan TG, Nemeroff CB, Prednisone augmentation in treatment-resistant depression with fatigue and hypocortisolaemia: a case series. Depress Anxiety 2000; 12:44–50. DeBattista C, Posener JA, Kalehzan BM, Schatzberg AF, Acute antidepressant effects of intravenous hydrocortisone and CRH in depressed patients: a doubleblind, placebo-controlled study. Am J Psychiatry 2000; 157:1334–1337. Belanoff JK, Flores BH, Kalezhan M et al, Rapid reversal of psychotic depression using mifepristone. J Clin Psychopharmacol 2001; 21:516–521. Young AH, Gallagher P, Watson S et al, Improvements in neurocognitive function and mood following adjunctive treatment with mifepristone (RU-486) in bipolar disorder. Neuropsychopharmacology 2004; 29:1538–1545. Juruena MF, Cleare AJ, Pariante CM, Hypothalamic pituitary adrenal axis, glucocorticoid receptor function and relevance to depression. Rev Bras Psiquiatr 2004; 26:189–201. Juruena MF, Cleare AJ, Bauer ME, Pariante CM, Molecular mechanisms of GR sensitivity and relevance for affective disorders. Acta Neuropsychiatrica 2003; 15:354–367.
Ch 14
7/4/05
3:58 pm
Page 134
Ch 15
7/4/05
3:59 pm
Page 135
chapter 15
Biological factors sustaining hypothalamic–pituitary–adrenal axis overactivation in affective disorder: focus on vasopressin Timothy G Dinan, Sinead O’Brien and Lucinda Scott Overactivity of the hypothalamic–pituitary–adrenal (HPA) axis characterized by hypercortisolism, adrenal hyperplasia and abnormalities in negative feedback is the most consistently described biological abnormality in melancholic depression.1 Corticotrophin releasing hormone (CRH) and arginine vasopressin (AVP) are the main secretagogues of the HPA/stress system. CRH is a peptide of 41 amino acid residues, originally discovered and sequenced by Vale et al,2 which is produced in the medial parvicellular neurones of the paraventricular nucleus of the hypothalamus. These neurones project to the external zone of the median eminence, where CRH is released into the portal vasculature to act on CRH1 receptors at the anterior pituitary. CRH and AVP act synergistically in bringing about adrenocorticotrophin (ACTH) release from the corticotrophs of the anterior pituitary which in turn stimulates cortisol output from the adrenal cortex. Following its identification in 1954, vasopressin, a nonapeptide, was considered the principal factor in the regulation of ACTH release but, with the subsequent elucidation of the structure of CRH and the domination of the one neurone/one transmitter principle, the role of CRH came to overshadow that of AVP. This dominance has been reflected in neuroendocrine studies conducted in depression.
Arginine vasopressin neuroanatomy AVP is released following a variety of stimuli, including increasing plasma osmolality, hypovolaemia, hypotension and hypoglycaemia. It has powerful
Ch 15
7/4/05
3:59 pm
136
Page 136
Bipolar disorder: the upswing in research & treatment
antidiuretic and vasoconstrictor effects. AVP has also been implicated in learning and memory processes.3 Our knowledge of the functional activity and pharmacology of AVP and its receptors in the regulation of HPA activity rests largely on studies conducted in rodents. AVP is released from the magnocellular system and from the parvicellular neurones of the paraventricular nucleus (PVN). AVP produced by the parvicellular neurones of the PVN is secreted into the pituitary portal circulation from axon terminals projecting to the external zone of the median eminence.4 AVP-containing cell bodies in the PVN are co-localized with CRH-containing neurones. In control non-stressed rats, within the pool of CRH neurosecretory cells, 50% coexpress AVP.5 CRH+/AVP– neurones are mostly found medially and ventrally, whereas CRH+/AVP+ cells are located more dorsally and laterally in the PVN. The neurosecretory granules in CRH-neurones undergo remarkable changes in size and peptide composition under experimental manipulations of the HPA axis. The PVN serves as an important relay site. It receives projections from ascending catecholaminergic pathways including noradrenergic projections from the nucleus of the solitary tract and from the locus coeruleus. The PVN also receives input from areas of the limbic system, notably the bed nucleus of the stria terminalis, the hippocampus and amygdala. In primates, including humans, high levels of immunoreactive AVP neurones have been demonstrated in these areas and also in the pituitary intermediate lobe, granular layers of the cerebellum and dentate gyrus. Lower levels are found in the medial habenula, adenohypophysis, area postrema, pineal, subfornical and subcommisural organs. Whilst it is evident that almost all of the CRH in the hypophyseal portal blood originates from the hypothalamic PVN, the precise origin of AVP is more controversial. It is thought that the bulk of AVP derives from the PVN. However, morphological and neurochemical studies suggest that AVP from supraoptic magnocellular AVP-secreting cells also access the hypophyseal portal blood,4 although this has not been definitively shown in humans.
Vasopressin receptors As with other peptide hormones, AVP exerts its effects through interaction with specific plasma membrane receptors, of which three major subtypes have been identified. V1a receptors are widely distributed on blood vessels, and have also been found in the central nervous system (CNS), including the PVN. V2 receptors are predominantly located in the principal cells of the
Ch 15
7/4/05
3:59 pm
Page 137
HPA axis overactivation in affective disorder
137
renal collecting system, although there is some evidence for central V2 receptors also. The ACTH-releasing properties occur via the V3 (V1b) receptor subtype. In situ hybridization studies reveal that V3 receptor mRNA is expressed in the majority of pituitary corticotrophs, in multiple brain regions and a number of peripheral tissues including kidney, heart, lung, breast and adrenal medulla.6
Synergism of corticotrophin releasing hormone and arginine vasopressin Vasopressin has ACTH-releasing properties when administered alone in humans, a response which may be dependant on the ambient endogenous CRH level. Following the combination of AVP and CRH, a much greater ACTH response is seen and both peptides are required for maximal pituitary–adrenal stimulation. The precise nature of this synergism is incompletely understood, with most information deriving from animal studies. It has been demonstrated that CRH, through cAMP, increases POMC gene transcription and peptide synthesis and storage. There may also be distinct corticotroph populations in the anterior pituitary, some of which require both AVP and CRH for ACTH release. Antoni4 suggests that AVP plays a role in stimulating the primary nuclear transcripts induced by CRH at the pituitary corticotroph.
Corticotrophin releasing hormone and vasopressin regulation by glucocorticoids Glucocorticoids play a pivotal feedback role in the regulation of the HPA. Two types of cortisol-binding sites have been described in the brain. The type 1 receptor, which is indistinguishable from the peripheral mineralocorticoid receptor, is distributed principally in the septohippocampal region. The type 2 or glucocorticoid receptor has a wider distribution. These receptor systems provide negative feedback loops at a limbic, hypothalamic and pituitary level. Overall the type 1 receptor is thought to mediate tonic influences of cortisol or corticosterone, whilst the type 2 receptor mediates stress-related changes in cortisol levels. Under normal conditions the responsiveness of parvicellular neurones to stress is under marked inhibition by the low resting levels of glucocorticoids.
Ch 15
7/4/05
3:59 pm
138
Page 138
Bipolar disorder: the upswing in research & treatment
The sensitivity of CRH and AVP transcription to glucocorticoid feedback is markedly different.7 CRH mRNA and CRH1 receptor mRNA levels are reduced by elevated glucocorticoids, whereas V3 receptor mRNA levels and coupling of the receptor to phospholipase C are stimulated by glucocorticoids, effects which may contribute to the refractoriness of AVP-stimulated ACTH secretion to glucocorticoid feedback.
Effects of chronic stress on corticotrophin releasing hormone/arginine vasopressin Immobilization stress and the use of hierarchically structured colonies both function as ‘psychological stress’ paradigms. Studies employing these paradigms in rats have revealed a shift of the hypothalamic CRH/AVP signal in favour of AVP; there is enhanced AVP synthesis in CRH-producing cells of the parvicellular neurones of the PVN and AVP storage in the CRHcontaining nerve terminals in the external zone of the median eminence. An increased proportion of hypothalamic neurones co-expressing AVP is also observed. In contrast, CRH production remains unaltered or decreased. A similar pattern of response is seen following prolonged immobilization stress where a shift towards an increased AVP/CRH ratio is observed. Direct manipulations of the HPA underscore this thesis. Tizabi and Aguilera,8 using a minipump infusion of CRH, revealed a reduction in CRH receptor number which was enhanced by the simultaneous infusion of AVP. Repeated immobilization stress was found to bring about sustained elevations in V3 receptor mRNA in the pituitary, suggesting an upregulation of AVP receptors in situations of chronic stress. This may explain the preserved enhanced response to novel superimposed stress in these animal models.
Vasopressin in major depression Studies of HPA function in depression have revealed numerous abnormalities. These abnormalities are most pronounced in depressives with melancholic features.9 A potential role for AVP in affective illness was put forward in 1978 by Gold and Goodwin.10 The hypothesis was based on the observations that (1) AVP deficiency produces deficits of behaviour which are reversed when the peptide is replaced, and (2) well-developed systems exist for distribution of AVP throughout the CNS, rendering AVP a suitable
Ch 15
7/4/05
3:59 pm
Page 139
HPA axis overactivation in affective disorder
139
candidate for involvement in complex behavioural systems. They also described the symptom complexes in affective illness that AVP is known to influence, notably memory processes, pain sensitivity, synchronization of biological rhythms and the timing and quality of rapid eye movement (REM) sleep. A role for AVP was supported not only by the above spectrum of symptoms but also by dynamic tests of HPA activity, and, in particular, the ‘DEX/CRH’ test. Dexamethasone (DEX), a potent synthetic glucocorticoid, binds primarily to the glucocorticoid receptor on anterior pituitary corticotrophs and, by feedback inhibition, suppresses ACTH and cortisol secretion.11 In the DEX/CRH test, when healthy subjects are treated with dexamethasone prior to CRH infusion, the release of ACTH is blunted and the extent of blunting is proportional to the dose of DEX. Paradoxically, when depressives are pretreated with DEX they show an enhanced ACTH response to CRH. It was postulated that vasopressin (VP)-mediated ACTH release was responsible for this finding. The combined DEX/CRH test is estimated to have a sensitivity of 80% in differentiating healthy subjects from depressives.12
Basal measures and symptom profile There are relatively few data on plasma AVP levels in depression. An early report found no change in plasma AVP levels in depression.13 In contrast, van Londen et al14 reported basal plasma levels of AVP to be elevated. In that study, 52 major depressives and 37 healthy controls were compared; AVP concentrations were found to be higher in the depressed cohort, with greater elevation in in-patient compared with out-patient depressives and in those with melancholic features. A number of studies have shown a significant positive correlation between peripheral plasma levels of AVP and hypercortisolaemia in patients with unipolar depression.14
Postmortem studies A postmortem study of depressed subjects reported an increased number of vasopressin-expressing neurones in paraventricular hypothalamic neurones.15 Raadsheer et al16 have also shown an increase in number of CRHneurones in depressives.
Ch 15
7/4/05
3:59 pm
140
Page 140
Bipolar disorder: the upswing in research & treatment
Dynamic tests of hypothalamic–pituitary–adrenal function and the vasopressin-ergic system Dinan et al17 examined a cohort of depressed subjects on two separate occasions, with CRH alone, and with the combination of CRH and desmopressin (DDAVP). A significant blunting of ACTH output to CRH alone was noted. Following the combination of CRH and DDAVP, the release of ACTH in depressives and healthy volunteers was indistinguishable. It was concluded that, whilst the CRH1 receptor is downregulated in depression, a concomitant upregulation of the V3 receptor takes place. This is consistent with the animal models of chronic stress, described above, in which a switching from CRH to AVP regulation is observed. It is interesting that, in CRH1 receptor-deficient mice, basal plasma AVP levels are significantly elevated, AVP mRNA is increased in the PVN and there is increased AVP-like immunoreactivity in the median eminence.18 In a recent study we have provided further evidence for the upregulation of the anterior pituitary V3 receptor.19 Fourteen patients with major depression and 14 age- and sex-matched healthy comparison subjects were recruited. Desmopressin 10 µg was given intravenously and ACTH and cortisol release was monitored for 120 minutes. The mean ± SEM ACTH response in the depressives was 28.4 ± 4.3 ng/l and in the healthy subjects was 18.8 ± 4.9 ng/l (p = 0.04). The mean ± SEM cortisol response in the depressives was 261.8 ± 46.5 nmol/l and in the healthy subjects it was 107.3 ± 26.1 nmol/l (p < 0.01). This suggests that patients with major depression have augmented ACTH and cortisol responses to desmopressin indicating enhanced V3 responsivity. Treatment with fluoxetine decreases cerebrospinal fluid (CSF) levels of CRH and AVP.20 Chronic fluoxetine administration has also been shown to reduce hypothalamic AVP secretion in vitro. In contrast, Heuser and coworkers21 examined CSF concentrations of CRH, AVP and somatostatin in depressed patients and in healthy controls prior to, and following, amitriptyline treatment. In treatment responders, CSF CRH was reduced with no effect of amitriptyline on AVP or somatostatin levels. Larger studies are required before any conclusions can be drawn about an antidepressant effect on central AVP activity. The question arises as to what type of involvement AVP may have in HPA dysregulation in depressive illness. Is a change in AVP regulation a biological correlate, a component of aetiology or a response to another hormonal disequilibrium? An understanding of major depression as a disorder of the stress system is fundamental in attempting to elucidate these possibilities.
Ch 15
7/4/05
3:59 pm
Page 141
HPA axis overactivation in affective disorder
141
Activation of the HPA in situations of stress is a normal homeostatic mechanism. In healthy subjects this response ceases when it is no longer biologically relevant. It is on this basis that we suggest an important role for AVP in the pathophysiology of major depression. Extrapolating from animal models of chronic stress, in which there is a switch from CRH to AVP regulation of ACTH release, a similar occurrence in humans is readily conceptualized. Supportive evidence includes the demonstration in depressed subjects of increased numbers of CRH neurones in the PVN co-expressing AVP and the ability of DDAVP to normalize the blunting of CRH-induced ACTH release, suggesting pituitary V3 receptor upregulation.17 Of importance also is the relative refractoriness of AVP-stimulated ACTH release to circulating glucocorticoids. This would suggest that, in major depression, the constant forward drive of the HPA axis may depend on AVP activity unrestrained by elevated ambient cortisol levels. The switch in emphasis from CRH to AVPergic pituitary–adrenal regulation in situations of chronic stress results in a continued activation of the HPA axis after the initial CRH-mediated response to stress is biologically appropriate. This recruitment of AVP as the primary regulator of the HPA axis in chronic stress conditions may explain the hypercortisolaemia that is demonstrated in depressed subjects, even when CRH/ACTH release is reduced, probably secondary to pituitary CRH receptor downregulation. The V3 receptor would seem an appropriate target site for the development of future antidepressants. Preliminary studies of a V3 antagonist in animal models of depression provide support for this view.22
Acknowledgement Timothy Dinan is in receipt of a project grant from the Health Research Board to study vasopressin in depression.
References 1.
2.
Rubin R, Dinan TG, Scott LV, The neuroendocrinology of affective disorders. In: Pfaff D, Arnold AP, Etgen AM, Fahrbach SE, Moss RL, Rubin RT (eds), Hormones, Brain and Behaviour. Academic Press: New York, 2001. Vale W, Spiess J, Rivier C, Rivier J, Characterisation of a 41 residue ovine hypothalamic peptide that stimulates secretion of the corticotropin and betaendorphin. Science 1981; 213:1394–1399.
Ch 15
7/4/05
3:59 pm
142 3.
4. 5.
6.
7.
8.
9. 10. 11.
12. 13. 14.
15.
16.
17.
18.
Page 142
Bipolar disorder: the upswing in research & treatment De Wied D, Diamant M, Fodor M, Central nervous system effects of the neurohypophyseal hormones and related peptides. Front Neuroendocrinol 1993; 14:251–302. Antoni FA, Vasopressinergic control of pituitary–adrenal secretion comes of age. Front Neuroendocrinol 1993; 14:76–122. Whitnall M, Vasopressin co-exists in half of the corticotropin-releasing factor neurones in the external zone of the median eminence in normal rats. Neuroendocrinology 1987; 45:420–424. Grazzini E, Lodboerer AM, Perez-Martin A et al, Molecular and functional characterization of the V1b receptor in rat adrenal medulla. Endocrinology 1996; 137:3906–3914. Ma XM, Lightman S, Aguilera G, Vasopressin and corticotropin-releasing hormone gene responses to novel stress in rats adapted to repeated restraint. Endocrinology 1999; 140:3623–3632. Tizabi Y, Aguilera G, Desensitization of the hypothalamic–pituitary–adrenal axis following prolonged administration of corticotropin-releasing hormone and vasopressin. Neuroendocrinology 1992; 56:611–618. Dinan TG, Glucocorticoids and the genesis of depressive illness: a psychobiological model. Br J Psychiatry 1994; 164:365–371. Gold PW, Goodwin FK, Vasopressin in affective illness. Lancet 1978; 10:1233–1236. Cole MA, Kim PJ, Kalman BA, Spencer RL, Dexamethasone suppression of corticosteroid secretion: evaluation of the site of action by receptor measures and functional studies. Psychoneuroendocrinology 2000; 25:151–167. Heuser I, Yassouridis A, Holsboer F, The combined dexamethasone/CRH test: a refined laboratory test for psychiatric disorders. J Affect Dis 1994; 4:93–101. Gjerris A, Hammer M, Vendsborg P et al. Cerebrospinal fluid vasopressinchanges in depression. Br J Psychiatry 1995; 147:696–701. van Londen L, Goekoop J, van Kempen G et al, Plasma levels of arginine vasopressin elevated in patients with major depression. Neuropsychopharmacology 1997; 17:284–292. Purba J, Hoogendijk W, Hoffman M, Swaab D, Increased numbers of vasopressin- and oxytocin-containing neurons in the paraventricular nucleus of the hypothalamus in depression. Arch Gen Psychiatry 1996; 53:137–143. Raadsheer F, Hoogendijk W, Hofman M, Swaab B, Increased number of corticotropin-releasing hormone expressing neurons in the paraventricular nucleus of the hypothalamic paraventricular nucleus of depressed patients. Neuroendocrinology 1996; 18:436–444. Dinan TG, Lavelle E, Scott L et al, Desmopressin normalises the blunted ACTH response to corticotropin-releasing hormone in melancholic depression: evidence of enhanced vasopressinergic responsivity. J Clin Endocrinol Metab 1999; 84:2238–2246. Muller MB, Landgraf R, Preil J et al, Selective activation of the hypothalamic vasopressinergic system in mice deficient for the corticotropin-releasing hormone receptor 1 is dependent on glucocorticoids. Endocrinology 2000; 141:4262–4269.
Ch 15
7/4/05
3:59 pm
Page 143
HPA axis overactivation in affective disorder 19.
20.
21.
22.
143
Dinan TG, O’Brien S, Lavelle E, Scott LV, Further evidence of enhanced vasopressin V3 receptor responses in melancholic depression. Psychol Med 2004; 34:169–172. De Bellis M, Gold PW, Geriacoti T et al, An association of fluoxetine treatment with reductions in CSF corticotropin-releasing hormone and arginine vasopressin in patients with depression. Am J Psychiatry 1996; 150:656–657. Heuser I, Bissette G, Dettling M et al, Cerebrospinal fluid concentrations of corticotropin-releasing hormone, vasopressin, and somatostatin on depressed patients and healthy controls: response to amitryptiline treatment. Depress Anxiety 1998; 8:71–79. Griebel G, Simiand J, Serrandeil-Le Gal C et al, Anxiolytic- and antidepressantlike effects of the non-peptide vasopressin V1b receptor antagonist, SSR149415, suggest an innovative approach for the treatment of stress-related disorders. Proc Natl Acad Sci USA 2002; 30:6370–6375.
Ch 15
7/4/05
3:59 pm
Page 144
Ch 16
7/4/05
3:59 pm
Page 145
chapter 16
Cognitive dysfunction: cause or consequence of bipolar disorder? Samuel R Chamberlain and Barbara J Sahakian
Bipolar disorder is characterized by cycling between depression, euthymia and mania. The argument that alterations or dysfunctions in cognition are a core phenomenon comes from evidence on several levels. Cognitive dysfunction is central to the diagnosis of depressive and manic episodes using the Diagnostic and Statistical Manual IV (DSM-IV). Distractibility and poor decision-making are included in the diagnostic criteria for manic episodes, and diminished ability to concentrate and indecisiveness are included in the criteria for depression. Another tier of evidence comes from psychological models, in which abnormalities in cognition are often held to be important. In Beck’s Cognitive Model,1 aberrant cognitive schema develop during childhood and are activated in later years by stressful and unpleasant life events (Figure 16.1). The activation of these aberrant schema leads to systematic errors in logic, and the well-known triad of negative belief directed at self, world and future. Cognitive therapies developed from these psychological models2 aim to correct dysfunctional attitudes and negative automatic thoughts. From a top-level perspective, the emotional states of people with mania and depression form two extremes of an affective spectrum (Figure 16.2). The emotional status of a given individual can vary on this spectrum in response to life events: towards dysphoria in response to relationship break-up, or towards euphoria when celebrating achievements. However, the extreme emotional states of people with mania or depression differ in that the moods are disproportionate and cause gross impairments in social functioning. Indeed, the deficits in social functioning, and behaviour of bipolar patients more broadly, are in themselves suggestive of cognitive dysfunction. Collectively, whether one examines affective disorder from the perspective of overt syndromic behaviour, DSM-IV diagnosis, or psychological models, cognitive dysfunction is central to our understanding. The
Ch 16
7/4/05
3:59 pm
146
Page 146
Bipolar disorder: the upswing in research & treatment
All-or-nothing thinking (e.g. 'whole week's work is useless because of one error')
Overgeneralization (e.g. 'this always happens to me')
Cognitive distortions Selective abstraction
Arbitrary inference
(disregarding positive comments from people and placing great emphasis on criticisms)
(e.g. 'my relative didn't answer my call. She must be ignoring me')
Minimization (e.g. 'he complimented me because he was having a good day')
Activation of schema by life events in adolescence or adulthood
Personalization (attributing events/comments to self)
Negative cognitive schema: SELF, WORLD, FUTURE
Aversive events during childhood Environmental factors (e.g. family background)
Genetic susceptibility
Figure 16.1 Beck’s cognitive model of depression: an overview.
development of advanced neurocognitive testing coupled with functional neuroimaging has in recent years facilitated the reliable investigation of specific cognitive profiles between patient groups. It has been possible to identify different cognitive dysfunctions that are (1) common to depression and mania; (2) capable of differentiating depression from mania; and (3) shown to persist into full clinical remission. Early measures of cognitive functioning, including the Wisconsin Card Sorting Test (WCST),3 revealed cognitive inflexibility in manic and depressed patients. However, these findings are difficult to interpret as they confound multiple areas of cognition, and there is an ongoing need for more sensitive and specific neurocognitive tests. With the advent of automated neurocognitive testing, such as the Cambridge Neuropsychological Test Automated Battery (CANTAB),4 broad and substantial cognitive deficits have been identified that are common to manic and depressed patients. Frequent findings include deficits on measures of attention, executive functioning, memory, and psychomotor speed (Figure 16.3) (see reference 10 for review). Importantly, the deficits appear to worsen with increasing age, and it may be necessary to augment the usual psychopharmacological treatment of elderly depressed patients with agents such as methylphenidate11 or modafinil,
Ch 16
7/4/05
3:59 pm
Page 147
Cognitive dysfunction (1) (2)
(3)
Persistent low mood Feelings of: - helplessness - hopelessness - worthlessness Anhedonia
147
Euthymia
Dysphoria Depression
Mania Euphoria
Euthymia
(1) (2) (3) (4)
(5) (6)
Grandiosity Overactivity Distractibility Socially inappropriate behaviour Increased appetite Impaired insight
Figure 16.2 The affective spectrum, and bipolar cycling. An individual’s position on the affective continuum is determined by genetic, social and other factors; it can fluctuate in day-to-day life. Mania and major depressive disorder (MDD) represent two extremes of this spectrum. In bipolar disorder, there is cycling between MDD and manic states, with ‘euthymia’ as an intermediate mood state.
the latter of which has been shown to reverse some of the cognitive dysfunctions found in schizophrenia.12 The DSM-IV diagnosis of mania and depression allows for considerable flexibility within the criteria, and therefore it should not be surprising that the precise pattern of cognitive dysfunction is somewhat variable. As neurocognitive tasks continue to be developed and modified in the light of clinical and research findings, it becomes increasingly feasible to identify trait markers for specific disorders. Much of our research work is geared towards the development of neurocognitive tasks capable of identifying trait and state markers, in order to detect disease onset, monitor treatment response, and assess psychopharmacological treatment efficacy. Advanced neurocognitive batteries are now commonly used in psychiatric research worldwide. In future, we foresee a central role for these approaches in the everyday assessment of psychiatric patients, particularly as many psychiatric disorders, including schizophrenia, are being reconstrued as neurocognitive disorders in the light of recent research findings.
Ch 16
7/4/05
3:59 pm
148
Page 148
Bipolar disorder: the upswing in research & treatment
Study
a
b
c
d
e
Age
32
37.5
50
72
43
HAM-D
22
23
22
30
26
Pattern recognition memory Spatial recognition memory Motor latency Delayed matching to sample (DMTS) Tower of London (% correct) New Tower of London (% correct) New Tower of London (latency) Set-shifting (ID/ED) ED deficits Increasing age (unipolar patients) Bipolar patients a b c d e
20005
Sweeney et al, Purcell et al, 19976 Elliott et al, 19967 Beats et al, 19968 Rubinsztein et al, 20009
No deficits Deficits Not performed
Figure 16.3 Neurocognitive dysfunctions in unipolar and bipolar patients. The cognitive deficits in unipolar disorder are broad and variable, and appear to be worse with increasing age. Bipolar patients demonstrate a degree of cognitive dysfunction equivalent to older unipolar patients.
In order to assess whether cognitive dysfunctions in bipolar disorder are cause or consequence of the overt pathology, we need to ask whether the broad and substantial cognitive deficits common to both these conditions are ‘primary’ deficits, or can be explained by more fundamental impairments in cognition. For example, can cognitive deficits be explained by a narrowing of the attentional focus to specific depression-relevant environmental stimuli or thoughts, or can they be related to an abnormal response to negative feedback? Additionally, we need to examine whether a subset of the cognitive dysfunctions found in mania differ in character from those found in depression. Multiple studies have failed to identify a different pattern of deficits on measures of attention, memory, visuospatial ability and executive functioning. Where might we expect differences to manifest? On a behavioural level, both disorders share great overlap in terms of impairment in everyday functioning, but mania and depression are considered to be at two extremes of the affective spectrum. Whereas depression is characterized by persistent low mood, anhedonia and negative interpersonal
Ch 16
7/4/05
3:59 pm
Page 149
Cognitive dysfunction
149
bias, mania is characterized by grandiose ideas, distractibility, socially inappropriate behaviour and an excessive positive interpersonal bias. On this basis, it would seem logical that differential cognitive characteristics might be found on neurocognitive tasks where affective processes are involved – i.e. on ‘hot’ cognitive tasks13 involving the processing of emotionally coloured stimuli, or where subjects are given negative feedback about their ongoing performance. Indeed, key differences have been identified between manic and depressed patients on tasks examining attentional processing bias, response to negative feedback, and decision-making. In go/no-go tasks, subjects give a rapid motor response to stimuli that fit into a target category, and withhold motor responses to stimuli in distractor categories. The affective version of this task uses positively valenced (happy) words, negatively valenced (sad) words, and emotionally neutral (‘cold’) words. This allows attentional processes to be investigated, and is particularly important, given that the mood states of patients with bipolar disorder are indicative of dysfunctions within attentional processes. Healthy controls undertaking the affective go/no-go task show a differential neural response in the subgenual cingulate when responding to emotionally ‘hot’ words compared with neutral ‘cold’ words.14 This neural region is known to function abnormally in bipolar disorder.15 Depressed patients respond more rapidly to sad versus happy words, and manic patients respond more rapidly to happy versus sad words,16 consistent with attentional bias towards negative environmental stimuli in depression, and towards positive environmental stimuli in mania. Acute tryptophan depletion in healthy controls has been found to cause a slowing of response to happy words in the same go/no-go task, demonstrating a central role for serotonin in the modulation of these aspects of attentional processing.17 These attentional biases are linked with the ruminations and triad of negativity central to Beck’s cognitive approach. Excitingly, it has been possible to identify putative neural substrates underlying these attentional abnormalities in depressed patients, in whom elevated neural responses to sad target words are found in the rostral anterior cingulate extending to the anterior medial prefrontal cortex. Additionally, depressed patients show a differential neural response in the right lateral orbitofrontal cortex when responding to sad distractor words (Figure 16.4).18 The findings agree with previous work demonstrating the importance of the medial prefrontal and orbitofrontal cortex in emotional processing and behavioural inhibition. These structures are components of relatively segregated basal ganglia–thalamocortical loops, with strong connections to limbic regions known to be important in emotional regulation (for example, see reference 19).
Ch 16
7/4/05
3:59 pm
150
Page 150
Bipolar disorder: the upswing in research & treatment
Response to sad versus happy targets: medial prefrontal cortex extending from rostral anterior cingulate to the medial prefrontal cortex
Responses to sad versus happy distractors: right lateral orbitofrontal cortex
Figure 16.4 Affective go/no-go task – functional magnetic resonance imaging activations comparing depressed patients with healthy controls. From reference 18, with permission from American Medical Association.
When depressed patients make errors in their daily life, such errors are viewed in a pessimistic light and serve as focal points for rumination and negative automatic thoughts. The attributional psychological model provides a formalization for these findings, in which depressed patients inappropriately attribute the cause of failure to global, stable and internal factors rather than specific, unstable, external factors.20 For example, after failing an exam, rather than thinking ‘I failed one exam, I will be able to pass re-takes if I work harder, and it was a difficult exam’, typically a depressed patient might think ‘I failed one exam because I’m a failure’. These negative automatic thoughts contribute to the hopelessness and pervasive, persistent negative mood. Novel cognitive tasks have been developed that can render objective measures of these abnormal responses in patients, by examining the effect of negative feedback on task performance. In the New Tower of London task (NTOL) (Figure 16.5), two sets of ‘snooker’ balls are presented on the screen, and the subject has to decide on the minimum number of moves necessary to make the bottom set of balls match the top set within a specified rule-set. After thinking through the problem, the subject selects the corresponding number on the screen. When the subject responds incorrectly, the computer gives negative feedback, and allows the subject to think through the problem again and select a new response. Although patients
Ch 16
7/4/05
3:59 pm
Page 151
Cognitive dysfunction Delayed Matching-to-Sample Text (DMTS) (CANTAB)
New Tower of London (NTOL) (CANTAB)
•
• •
Per cent correct impaired in bipolar disorder, schizophrenia and Parkinson's disease Abnormal response to negative feedback in depression Task performance dependent on relatively frontal cortical structures
•
Per cent correct impaired in bipolar disorder, schizophrenia and Parkinson's disease Abnormal response to negative feedback in depression Task performance dependent on relatively posterior cortical structures
• •
Probabilistic reversal learning (CANTAB)
•
Abnormal response to false negative feedback in depression
151
Decision-making task
•
•
Depressed and manic patients: impairment on latency and strategy Only manic patients demonstrate impaired quality of decision-making
Figure 16.5 Neurocognitive tasks of use in differentiating mania from depression in bipolar disorder. CANTAB, Cambridge Neuropsychological Test Automated Battery.
with depression, schizophrenia and Parkinson’s disease all show impairments on this and similar tasks, in terms of percentage problems solved correctly, only depressed patients show an abnormal response to negative feedback;21,22 they are far more likely to fail a given problem if they have just
Ch 16
7/4/05
3:59 pm
152
Page 152
Bipolar disorder: the upswing in research & treatment
been given negative feedback. These findings are replicated on the Delayed Matching-To-Sample Test of Recognition Memory (DMTS),4 in which subjects memorize an abstract complex pattern and select the correct pattern from other patterns at a later time (Figure 16.5). Similarly, Steffens et al have found an abnormal response to negative feedback in a different task in elderly depressed patients.23 False negative feedback, in which subjects are told that they made an incorrect response when their response was in fact correct, selectively impairs depressed patients’ performance on a probabilistic reversal learning task (Figure 16.5). On this task, subjects have to select an arbitrarily assigned correct pattern from a choice of two. After each choice, feedback is given as to whether the choice was correct or incorrect. However, in a minority of cases (20% of trials) subjects are given false feedback. It is made clear in the task instructions before administering this task that the computer will give false feedback some of the time, yet depressed patients’ performance is still found to be persistently impaired by false negative feedback. Even on the last 10 trials, false negative feedback is still disrupting depressed patients, whereas controls are just ignoring it. The abnormal response to negative feedback in depression found on these various tasks is indicative of dysfunctional reward systems, and this is likely to contribute to the anhedonia and loss of motivation reported at a syndromic level. Bipolar patients in the manic phase show impaired decision-making, with excessive involvement in pleasurable activities with high potential for negative consequences. For example, they might opt to go on spending sprees with the family savings or make unwise business decisions. In many ways, these behaviours are akin to those found in patients with orbitofrontal/ventromedial prefrontal cortical damage consequential to brain injury.24,25 These patients often demonstrate normal performance on ‘cold’ cognitive tasks of memory, learning and executive function – such as the Wisconsin Card Sorting Task3 – but abnormal performance on ‘hot’ gambling tasks. In one version of a computerized gambling task, subjects make decisions based on variably weighted probabilities (Figure 16.5). A ‘token’ is hidden behind a red or blue box, and the subject bets a variable number of points on which coloured box the token is hidden behind. Both manic and depressed patients’ performance on this task is worse than controls’ in terms of response latency, and total number of points accumulated. However, only manic patients made suboptimal decisions – they make poor decisions that run contrary to logic whereas depressed patients do not.26 The extent of this deficit is correlated with the severity of the mania recorded by Young’s Mania Score, and functional imaging points to abnormalities
Ch 16
7/4/05
3:59 pm
Page 153
Cognitive dysfunction
153
in the dorsal anterior cingulate, frontal polar and right inferior frontal cortex as neural substrates of these suboptimal decisions in manic patients.27 Manic patients also behave in a risky way, in that they often lose all accumulated points on this task.26 In contrast to the broad and variable cognitive dysfunctions found during manic and depressive episodes, recent studies have identified specific cognitive dysfunctions that persist even when patients are considered to be fully recovered clinically. These residual cognitive deficits include: psychomotor slowing, impaired visual recognition memory and impaired sustained attention.8,9,28–31 One study found psychomotor slowing on a spatial working memory task that persisted into recovery in patients with seasonal affective disorder (SAD), despite other measures of cognitive dysfunction returning to baseline.28 Residual deficits have also been identified in bipolar patients who had been in remission for at least 4 months and met careful screening criteria, on a test of visuospatial recognition memory.9 The Rapid Visual Information Processing (RVIP) task4 is a measure of sustained attention, and requires subjects to detect odd or even sequences of digits by observing a white box in the centre of the screen. The task has been shown to activate a right frontoparietal neural network in healthy controls.32 Digits appear within the white box one at a time (Figure 16.6). Euthymic patients show impaired sustained attention, as measured by RVIP, even when the extent of remission is at peak.31 It has been argued that this may represent a trait marker for the development of bipolar disorder. These residual deficits may prove to be the greatest barrier to patient rehabilitation, as has been found to be the case in schizophrenia,10,33,34 and serve to limit the use of psychological treatments in certain contexts. Cognitive dysfunction is central to our understanding of bipolar disorder, as evidenced by the syndromic behaviour, DSM-IV diagnostic criteria, psychological models and neurocognitive findings. In particular, the coupling of advanced neurocognitive testing with functional imaging is leading an upsurge in research towards the synthesis of endophenotypes – intermediate constructs capable of being both downstream of overt syndromic behaviour and upstream of biochemical abnormalities and genetics. In many respects then, cognitive dysfunctions and their neural substrates can be viewed as being cause rather than consequence of the overt manifestation of bipolar disorder. The finding of residual cognitive deficits persisting into full clinical remission has huge ramifications for treatment, as do the differential performances of manic and depressed patients on ‘hot’ processing tasks. Advanced neurocognitive testing batteries are likely to play a key role in the generation of more objective means of detecting
Ch 16
7/4/05
3:59 pm
154
Page 154
Bipolar disorder: the upswing in research & treatment
• •
Impaired performance in bipolar patients even at peak of remission Impaired sustained attention, as measured by RVIP, may represent a trait marker for bipolar disorder
Figure 16.6 Rapid visual information processing tasks (RIVP) (Cambridge Neuropsychological Test Automated Battery (CANTAB)).
disease onset, monitoring recovery and relapse, and assessing the efficacy of novel and established pharmacological and psychological interventions, towards the synthesis of more efficient treatment algorithms for these highly prevalent and debilitating disorders.
References 1. 2. 3. 4. 5.
6. 7.
8.
Beck AT, The past and future of cognitive therapy. J Psychother Pract Res 1997; 6:276–284. Parker G, Roy K, Eyers K, Cognitive behavior therapy for depression? Choose horses for courses. Am J Psychiatry 2003; 160:825–834. Berg E, A simple objective technique for measuring flexibility in thinking. J Gen Psychol 1948; 39:15–22. Cambridge Cognition, Cambridge Neuropsychological Test Automated Battery (CANTAB). www.camcog.com. Sweeney JA, Kmiec JA, Kupfer DJ, Neuropsychologic impairments in bipolar and unipolar mood disorders on the CANTAB neurocognitive battery. Biol Psychiatry 2000; 48:674–684. Purcell R, Maruff P, Kyrios M, Pantelis C, Neuropsychological function in young patients with unipolar major depression. Psychol Med 1997; 27:1277–1285. Elliott R, Sahakian BJ, McKay AP et al, Neuropsychological impairments in unipolar depression: the influence of perceived failure on subsequent performance. Psychol Med 1996; 26:975–989. Beats BC, Sahakian BJ, Levy R, Cognitive performance in tests sensitive to frontal lobe dysfunction in the elderly depressed. Psychol Med 1996; 26: 591–603.
Ch 16
7/4/05
3:59 pm
Page 155
Cognitive dysfunction 9. 10. 11. 12.
13. 14.
15.
16. 17.
18.
19.
20. 21.
22.
23. 24. 25.
26. 27.
155
Rubinsztein JS, Michael A, Paykel ES, Sahakian BJ, Cognitive impairment in remission in bipolar affective disorder. Psychol Med 2000; 30:1025–1036. Tavares JV, Drevets WC, Sahakian BJ, Cognition in mania and depression. Psychol Med 2003; 33:959–967. Lavretsky H, Kumar A, Methylphenidate augmentation of citalopram in elderly depressed patients. Am J Geriatr Psychiatry 2001; 9:298–303. Turner DC, Clark L, Pomarol-Clotet E et al, Modafinil improves cognition and attentional set shifting in patients with chronic schizophrenia. Neuropsychopharmacology 2004; 29:1363–1373. Roiser J, Rubinsztein JS, Sahakian B, Cognition in depression. Psychiatry 2003; 2:43–47. Elliott R, Rubinsztein JS, Sahakian BJ, Dolan RJ, Selective attention to emotional stimuli in a verbal go/no-go task: an fMRI study. Neuroreport 2000; 11:1739–1744. Drevets WC, Ongur D, Price JL, Neuroimaging abnormalities in the subgenual prefrontal cortex: implications for the pathophysiology of familial mood disorders. Mol Psychiatry 1998; 3:220–226, 190–191. Murphy FC, Sahakian BJ, Rubinsztein JS et al, Emotional bias and inhibitory control processes in mania and depression. Psychol Med 1999; 29:1307–1321. Murphy FC, Smith KA, Cowen PJ et al, The effects of tryptophan depletion on cognitive and affective processing in healthy volunteers. Psychopharmacology (Berl) 2002; 163:42–53. Elliott R, Rubinsztein JS, Sahakian BJ, Dolan RJ, The neural basis of moodcongruent processing biases in depression. Arch Gen Psychiatry 2002; 59:597–604. Alexander GE, DeLong MR, Strick PL, Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986; 9:357–381. Abramson LY, Metalsky GI, Alloy LB, Hopelessness depression: A theory-based subtype of depression. Psychol Rev 1989; 96:358–372. Elliott R, Baker SC, Rogers RD et al, Prefrontal dysfunction in depressed patients performing a complex planning task: a study using positron emission tomography. Psychol Med 1997; 27:931–942. Elliott R, Sahakian BJ, Herrod JJ et al, Abnormal response to negative feedback in unipolar depression: evidence for a diagnosis specific impairment. J Neurol Neurosurg Psychiatry 1997; 63:74–82. Steffens DC, Wagner HR, Levy RM et al, Performance feedback deficit in geriatric depression. Biol Psychiatry 2001; 50:358–363. Starkstein SE, Mayberg HS, Berthier ML et al, Mania after brain injury: neuroradiological and metabolic findings. Ann Neurol 1990; 27:652–659. Bechara A, Tranel D, Damasio H, Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 2000; 123:2189–2202. Murphy FC, Rubinsztein JS, Michael A et al, Decision-making cognition in mania and depression. Psychol Med 2001; 31:679–693. Rubinsztein JS, Fletcher PC, Rogers RD et al, Decision-making in mania: a PET study. Brain 2001; 124:2550–2563.
Ch 16
7/4/05
3:59 pm
Page 156
156
Bipolar disorder: the upswing in research & treatment
28.
O’Brien JT, Sahakian BJ, Checkley SA, Cognitive impairments in patients with seasonal affective disorder. Br J Psychiatry 1993; 163:338–343. Silverstein ML, Harrow M, Bryson GJ, Neuropsychological prognosis and clinical recovery. Psychiatry Res 1994; 52:265–272. Ferrier IN, Stanton BR, Kelly TP, Scott J, Neuropsychological function in euthymic patients with bipolar disorder. Br J Psychiatry 1999; 175: 246–251. Clark L, Iversen SD, Goodwin GM, Sustained attention deficit in bipolar disorder. Br J Psychiatry 2002; 180:313–319. Coull JT, Frith CD, Frackowiak RS, Grasby PM, A fronto-parietal network for rapid visual information processing: a PET study of sustained attention and working memory. Neuropsychologia 1996; 34:1085–1095. Goldberg TE, Gold JM, Greenberg R et al, Contrasts between patients with affective disorders and patients with schizophrenia on a neuropsychological test battery. Am J Psychiatry 1993; 150:1355–1362. Zarate CA Jr, Tohen M, Land M, Cavanagh S, Functional impairment and cognition in bipolar disorder. Psychiatr Q 2000; 71:309–329.
29. 30. 31. 32.
33.
34.
Ch 17
7/4/05
4:00 pm
Page 157
chapter 17
The neural basis of cognitive function in bipolar disorder Vivienne Curtis
Despite the notion of apparent ‘recovery’ between episodes of bipolar disorder,1 impairments in cognitive task performance have been reported during remission.2,3 These deficits have been observed across a range of cognitive domains including declarative memory,4,5 attention6 and executive function.7–9 However, when residual symptomatology is controlled for a specific deficit in attention, executive function predominates.2,3,6,8 This implies that a dysfunction of the neuronal substrates underlying cognitive operations may exist alongside alterations in limbic function. The use of cognitive paradigms to elicit information about neural networks has been established within the literature relating to schizophrenia and is now beginning to be applied to studies of bipolar disorder. To date, only a handful of studies have examined working memory function in remitted bipolar disorder patients. Kusumo and Vaughan10 reported that ‘well’ bipolar disorder patients showed delay-dependent deficits on the Brown–Peterson paradigm, a test of retention in short-term (working) memory. Asarnow and MacCrimmon11 reported that bipolar disorder patients ‘free from major symptoms’, were impaired on an attentional target detection task only when the targets were embedded in longer stimulus arrays. In tasks specifically testing the integrity of the phonological loop, euthymic bipolar subjects have been shown to have no impairments of the Sternberg paradigm and the digits forward sub-test of the WAIS-R, but subjects have shown impairment on the digits backwards sub-test which is an index of the integrity of the central executive.8,12 Harmer et al13 observed impaired performance in remitted bipolar subjects undertaking a sustained attention task that was independent of working memory load, and have proposed that deficits in sustained attention rather than working memory may represent a core feature of bipolar disorder. Taken together, these studies converge on
Ch 17
7/4/05
4:00 pm
158
Page 158
Bipolar disorder: the upswing in research & treatment
a general concept of dysfunction within the attentional and executive control systems in euthymic bipolar disorder. The combination of a neuropsychological and neuroimaging approach can enable us to understand these dysfunctions further and investigate the effect of disease trait on the relationships between neuroanatomy and neuropsychology. We have recently undertaken a functional magnetic resonance imaging (fMRI) study using two tasks: the two-back task (dependent on the integrity of the central executive) and the Sternberg paradigm (dependent on the integrity of the phonological loop) in matched groups of euthymic bipolar patients.14 Twelve right-handed euthymic bipolar I males receiving lithium carbonate monotherapy were recruited and matched with 12 controls. The two-back task comprised a single working memory load contrasted with a baseline vigilance condition (Figure 17.1). The Sternberg paradigm used a parametric design incorporating variable working memory load with fixed delay between presentation of an array of items to be remembered and a target item (Figure 17.2). Functional activation data were acquired during performance of the tasks and were analysed to produce brain activation maps representing significant group differences in activation (ANOVA). Loadresponse curves were derived from the Sternberg task data set. There were no significant differences between the groups in demographics, clinical
Task parameters: • Random series of letters displayed on the screen at the rate of 1 every 2000 ms • Inter-stimulus interval of 1000 ms • Subject to press response button when the letter on the screen is the same as that occurring two previously
1000 ms 1000 ms 1000 ms 1000 ms
a
1000 ms
b a
'YES' Figure 17.1 Two-back task.
Ch 17
7/4/05
4:00 pm
Page 159
The neural basis of cognitive function in bipolar disorder
159
symptomatology or on-line performance of the tasks. In the two-back task, compared with controls, bipolar disorder patients showed reductions in bilateral frontal, temporal and parietal activation, and increased activations with the left precentral, right medial frontal and left supramarginal gyri. No between-group differences were observed in the Sternberg task at any working memory load. Common criticisms of clinical fMRI studies are that the results are confounded by differences in clinical state, medication and task performance between patient and control groups. Within our study we tried to limit these confounds at source wherever possible. Thus, we used standardized diagnosis to reduce variability within the subject groups and we actively screened subjects to detect the presence of significant misuse of alcohol or illicit substances. No statistically significant differences in training requirements for either group were observed outside the scanner or during on-line performance. In addition, our bipolar subjects had been euthymic for at least 4 weeks prior to scanning, and did not differ from controls on standardized clinical ratings prior to testing, facilitating an investigation of the bipolar ‘trait’ without ‘state’ confounds. One potential confounding factor is medication. Although the bipolar subjects had previously been exposed to different medications, they were all
Task parameters: • On each trial 1–5 digits are presented to the participant • Interval of 500 ms in which a plus sign is presented • Single probe letter presented for 1500 ms • Subject to state whether probe digit was a member of the original memory set (via dual response box) • 1500 ms inter-trial interval
1500 ms 500 ms 1500 ms
-384+ 4 1500 ms
'YES' Figure 17.2 The Sternberg task.
Ch 17
7/4/05
4:00 pm
160
Page 160
Bipolar disorder: the upswing in research & treatment
established on lithium monotherapy for at least 6 months prior to the scan. While no studies of the effect of lithium on the blood oxygen level dependent (BOLD) response in humans have been undertaken, chronic lithium administration has been demonstrated in the rat aorta to increase acetylcholine endothelium-dependent relaxation, and also to increase vascular contraction.15,16 Therefore, it is possible that there is an effect of lithium on vascular responsiveness and, consequently, on neurovascular coupling. Future specific investigations of the influence of medication and lithium in particular on cognitive activation effects will be required to quantify such a theoretical confound. There have been relatively few functional imaging studies using cognitive activation tasks in euthymic patients with bipolar disorder. The findings with respect to frontal lobe function have varied depending on experimental design and clinical state, but the anterior cingulate, supplementary motor area and dorsal and ventrolateral prefrontal cortices appear to be regularly implicated in both increases and decreases of network activation in bipolar subjects.2,17–19 During performance of the two-back task, neural networks were activated which were in keeping with those reported by other neuroanatomic20,21 and neuroimaging studies of working memory in healthy controls.22 The prefrontal regions which showed an attenuated response in the bipolar subjects are in keeping with those previously implicated in both the maintenance and the manipulation of information held in the working memory store in studies of healthy subjects23–26 and in other psychiatric groups27 and mood states.19 Furthermore, both structural neuroimaging and postmortem studies have suggested morphological alterations in these regions in bipolar subjects.2 The bipolar subjects showed regions of attenuated response beyond the left frontal lobes in the right middle temporal gyrus, cuneus/precuneus and the cerebellum. These regions have been inconsistently reported in other functional imaging working memory studies and may represent a variability of recruitment of other nodes within the working memory network. There were few regions of increased activation in our bipolar group during the two-back task. These included the left supramarginal gyrus, which has been implicated as part of the phonological loop.28–30 This altered pattern of response suggests that changes are not simply due to increased effort, but this might reflect bipolar subjects’ attempt to draw upon intact slave system resources to support executive task performance, and intact working memory capacity may scaffold euthymic bipolar disorder patients’ performance.25 Our subjects also had some small regions of increased activation
Ch 17
7/4/05
4:00 pm
Page 161
The neural basis of cognitive function in bipolar disorder
161
within the frontal lobes, a finding reported previously during a verbal fluency task.18 This hyperactivation of the frontal cortices has also been reported in schizophrenic subjects during working memory tasks and may be a reflection of inefficient use of the prefrontal networks.31 Callicott et al32 have suggested that in schizophrenic subjects the pattern of load response relationships may vary across different nodes of the working memory network and that regions may move through hyperfrontality to a hypofrontal response. In this way our findings of both hyper- and hypofrontality during task performance may be a reflection of compromised neural strategies in a subject group performing at their ceiling. While the Sternberg task has been studied less extensively than the twoback task (and with a greater heterogeneity of experimental designs) it has been shown to generate load-dependent changes in activation in the dorsolateral prefrontal cortex which are felt to reflect encoding of items.33,34 These findings have not been widely replicated, either within the literature or in our study, although the neural networks subserving task performance in our subjects were in keeping with those seen in neuroimaging studies of working memory tasks.35 Furthermore, we have found no regions of significant difference between bipolar and control groups while performing the Sternberg task. Despite the lack of a load-dependent increase in the BOLD signal evident in this study there was a main effect of memory load on the response time variables, suggesting that this paradigm served its parametric function. However, as neural networks were engaged to a similar degree in both groups, the current results indicate that the range of task difficulty fell within the range of achievable performance. It is therefore possible that BOLD signal changes may be no more sensitive as a measure of neural network function than performance in the behavioural task and so significant decrements in activation for the Sternberg task may accompany only significant decrements in performance. This study demonstrates how combining functional imaging can teach us about the neural basis of cognitive function in bipolar disorder and identifies a task-dependent alteration of prefrontal lobe function which we suggest is related to central executive rather than phonological loop function. This finding reinforces the importance of using more than one cognitive activation paradigm in a patient group to increase our understanding of the relationship between task demand and neural activity. While we have controlled for the confounding effects of mood symptoms, future studies of bipolar subjects are required to exclude additional confounds such as medication. However, our findings support the notion that, in bipolar disorder, failure to engage frontoexecutive function represents a core deficit.
Ch 17
7/4/05
4:00 pm
162
Page 162
Bipolar disorder: the upswing in research & treatment
References 1. 2. 3. 4.
5. 6. 7.
8. 9. 10. 11.
12.
13.
14.
15.
16.
17.
Goodwin FK, Jamison KR, Manic Depressive Illness. Oxford University Press: New York, 1990. Bearden CE, Hoffman KM, Cannon D, The neuropsychology and neuroanatomy of bipolar affective disorder: a critical review. Bipolar Disord 2001; 3:106–150. Ferrier IN, Thompson JM, Cognitive impairment in bipolar affective disorder: implications for the bipolar diathesis. Br J Psychiatry 2002; 180:293–295. Van Gorp WG, Altshuler L, Theberge DC et al, Cognitive impairment in euthymic bipolar patients with and without prior alcohol dependence Arch Gen Psychiatry 1998; 55:41–46. Rubinsztein JS, Michael A, Paykel ES, Sahakian BJ, Cognitive impairment in remission in bipolar affective disorder. Psychol Med 2000; 30:1025–1036. Clark L, Iverson SD, Goodwin GM, Sustained attention deficit in bipolar disorder. Br J Psychiatry 2002; 180:313–319. Hawkins KA, Hoffmann RE, Quinlan DM et al, Cognition, negative symptoms, and diagnosis: a comparison of schizophrenic, bipolar, and control samples. J Neuropsychiatry Clin Neurosci 1997; 9:81–89. Ferrier IN, Stanton BR, Kelly TP, Scott J, Neuropsychological function in euthymic patients with bipolar disorder. Br J Psychiatry 1999; 175:246–251. Zubieta JK, Huguelet P, O’Neil RL, Giordani BJ, Cognitive function in euthymic bipolar I disorder. Psychiatry Res 2001; 102:9–20. Kusumo KS, Vaughan M, Effects of lithium salts on memory. Br J Psychiatry 1977; 131:453–457. Asarnow RF, MacCrimmon DJ, Span of apprehension deficits during the postpsychotic stages of schizophrenia. A replication and extension. Arch Gen Psychiatry 1981; 38:1006–1011. Thompson JM, Gray JM et al, A component process analysis of working memory dysfunction in bipolar affective disorder. J Psychopharmacol 2000; 14(Suppl):A24. Harmer CJ, Clark L, Grayson L, Goodwin GM, Sustained attention deficit in bipolar disorder is not a working memory impairment in disguise. Neuropsychologia 2002; 40:1586–1590. Monks PJ, Thompson JM, Bullmore ET et al, A functional MRI study of working memory task in euthymic bipolar disorder: evidence for task-specific dysfunction. Bipolar Disord 2004; 6:550–564. Dehpour AR, Ghafourifar P, Samenian J et al, The effect of lithium on endothelial-dependent relaxation in rat isolated aorta. Gen Pharmacol 1995; 26: 1003–1007. Ullian ME, Walsh LG, Wong KC, Allan CJ, Potentiation of angiotensin IIstimulated vascular contraction by lithium. Am J Physiol 1995; 268:H2009–2016. Frangou S, Raymont V, Kingston J, Shergill S, Investigating the functional interface between the dorsal and central prefrontal circuitry in bipolar disorder. Bipolar Disord 2003; 5(Suppl):47–48.
Ch 17
7/4/05
4:00 pm
Page 163
The neural basis of cognitive function in bipolar disorder 18.
19.
20.
21.
22. 23.
24. 25.
26.
27.
28.
29. 30. 31. 32.
33. 34.
35.
163
Curtis VA, Dixon TA, Morris RG et al, Differential frontal activation in schizophrenia and bipolar illness during verbal fluency. J Affect Disord 2001; 66:111–122. Blumberg HP, Leung HC, Skullarski P et al, A functional magnetic resonance imaging study of bipolar disorder: state- and trait-related dysfunction in ventral prefrontal cortices. Arch Gen Psychiatry 2003; 60:601–609. Goldman-Rakic PS, Circuitry of primate prefrontal cortex and regulation of behaviour by representational memory. In: Plum F, Mountcastle V (eds), Handbook of Physiology – The Nervous System. American Physiological Society: Bethesda, 1987:373–417. Alexander GE, Delong MR, Strick PL, Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986; 9:357–381. Duncan J, Owen AM, Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci 2000; 23:475–483. Cohen JD, Forman SD, Braver TS et al, Activation of prefrontal cortex in a nonspatial working memory task with functional MRI. Hum Brain Mapp 1994; 1:293–304. Awh E, Jonides J, Smith EE et al, Dissociation of storage and rehearsal in verbal working memory: Evidence from PET. Psychol Sci 1996; 7:25–31. D’Esposito M, Ballard D, Aguirre GK, Zarahn E, Human prefrontal cortex is not specific for working memory: a functional MRI study. Neuroimage 1998; 8:274–282. Collette F, Salmon E, Van der Linden M et al, Regional brain activity during tasks devoted to the central executive of working memory. Cogn Brain Res 1999; 7:411–417. Goldberg TE, Weinberger DR, Effects of neuroleptic medication on the cognition of patients with schizophrenia: a review of recent studies. J Clin Psychiatry 1996; 57(Suppl 9):62–65. D’Esposito M, Postle BR, Rypma B, Prefrontal cortical contributions to working memory: evidence from event-related fMRI studies. Exp Brain Res 2000; 133:3–11. Becker JT, Mintun MA, Diehl DJ et al, Functional neuroanatomy of verbal free recall: a replication study. Hum Brain Mapp 1994; 1:284–292. Palescu E, Frith CD, Frackowiak RS, The neural correlates of the verbal component of working memory. Nature 1993; 362:342–345. Callicott JH, Bertolino A, Mattay VS et al, Physiological dysfunction of the prefrontal cortex in schizophrenia revisited. Cereb Cortex 2000; 10:1078–1092. Callicott JH, Mattay VS, Verchinski BA et al, Complexity of cortical dysfunction in schizophrenia: more than up or down. Am J Psychiatry 2003; 160:2209–2215. Braver TS, Cohen JD, Nystrom LE et al, A parametric study of prefrontal cortex involvement in human working memory. Neuroimage 1997; 5:49–62. Rypma B, D’Esposito M, The roles of prefrontal brain regions in components of working memory: effects of memory load and individual differences. Proc Natl Acad Sci USA 1999; 96:6558–6563. Veltman DJ, Rombouts SARB, Dolan RJ, Maintenance versus manipulation in working memory revisited: an fMRI study. Neuroimage 2003; 18:247–256.
Ch 17
7/4/05
4:00 pm
Page 164
Ch 18
7/4/05
4:00 pm
Page 165
chapter 18
Psychological treatments: does the evidence stack up? Jan Scott
Introduction To put some of the issues of this chapter into context, I will briefly review some of the morbidity and cost issues in bipolar disorders. I will then summarize outcomes of randomized controlled trials (RCTs) for bipolar disorder where psychological therapies have been added to medication and usually, but not exclusively, are compared with treatments as usual. I will then present some preliminary data from the recently completed Medical Research Council (MRC) study looking at a similar paradigm using cognitive therapy added to treatment as usual, compared with routine psychiatric treatment. Finally, I will draw some conclusions particularly highlighting how these data fit in with the recommendations given by treatment guidelines on bipolar disorders, such as those published by the American Psychiatric Association, the British Association of Psychopharmacology, and the Royal College of Psychiatrists of Australia and New Zealand Psychiatry (for example, see references 1 and 2).
Burden of disease Of the top ten leading causes of disability worldwide amongst people aged 19–45, six are psychiatric disorders, and unipolar and bipolar disorders occupy the first and second positions, respectively, both coming above schizophrenia in terms of burden.3 Standardized mortality ratios suggest bipolar disorders again exceed schizophrenia if all causes of death are considered.4 Morbidity statistics suggest that about 2% of the population will
Ch 18
7/4/05
4:00 pm
166
Page 166
Bipolar disorder: the upswing in research & treatment
experience bipolar I, bipolar II or a bipolar spectrum disorder.4,5 Very importantly, patients with bipolar I and II disorders show a great deal of axis I and axis II co-morbidity, with about 30–50% demonstrating an anxiety, substance misuse or personality disorder.6–8 Suicide attempts are also very common, in about a third to a half of these individuals.4 Studies of social adjustment suggest that at least 30% of individuals with bipolar disorder do not return to their previous employment within a year of an episode, and in another 30%, if they do return to employment, they do not return to their previous level of functioning.9,10 In Britain, people with bipolar disorders rather than other severe mental disorders occupy most acute psychiatric bed days, and this is despite the fact that, in bipolar depression, people are often not admitted to hospital for treatment.8,11 Between episodes, individuals are highly symptomatic; Judd et al12 reported that persons with bipolar disorders had syndromal or subsyndromal depressive symptoms for about 50% of the time during 12 years of weekly follow-up ratings. This illustrates the enormous level of morbidity and mortality of bipolar disorders, yet the National Service Framework for UK Mental Health Services did not set one target specifically aimed at the care and treatment of bipolar disorders. Our MRC study13 highlighted the annual costs of care for individuals with bipolar disorders: 92% received one or more psychiatric out-patient contacts, the median number being five per year and 98% were receiving medication, with the median number of prescribed medications being about five (similar to reports from the USA). However, individuals rarely reported day care, additional support or therapy and few were living in supported accommodation. These cost figures contrast sharply with the distribution of resources in schizophrenia, where about 10–15% of the costs of care and
Citation Cochran 198416 Lam et al 200018 Perry et al 199917 Scott et al 200119 Fixed Combined (4) Random Combined (4)
N Total
Effect
p Value
28 23 68 42
0.12 0.02 0.59 0.38
0.02 0.00 0.34 0.17
161 161
0.28 0.22
0.00 0.01
0.01
0.1
1
10
Favours psychotherapy
Figure 18.1 Odds ratios of relapse rates in early trials.
100
Ch 18
7/4/05
4:00 pm
Page 167
Psychological treatments
167
resources are allocated to psychological treatments or other communitybased supports. In bipolar disorders, the majority of the health-care spend (about 53%) was actually accounted for by hospital admissions. This is quite important in the context of psychological treatments because the argument against adjunctive therapy is usually that it is too expensive, but the opportunity cost of a course of about 20 hours of therapy is probably only equivalent to that of 2 days of in-patient care and, as evidenced from recent research, most psychological treatments can significantly reduce hospitalizations compared with usual psychiatric treatment.
Psychological treatment studies A review of psychological treatment studies in bipolar disorders over the past 30 or 40 years shows that, up until very recently, the research was not of a high standard. Between 1960 and 1995, there were about 20 studies and the average sample size was about 20–25 subjects.14,15 Few of them were randomized trials, but the studies provided some indication that, with family therapy and group therapy, adherence to medication was improved and individuals with bipolar disorders did better subjectively if they received these additional inputs. However, there were no data that addressed the issue of relapse or measured social or functional outcomes in specific terms. The situation has changed quite dramatically in the past 4–5 years and there are now 17 or 18 RCTs ongoing across the world and four or five of those trials have over 100 participants, which may be a small number in terms of drug trials, but is a large number in terms of many psychological treatment studies. About eight of the completed studies have produced data that are accessible to be put into a pooled analysis of relapse rates for psychological treatments compared with usual psychiatric treatment (either routine or standardized). The first set of studies is put into rank order of the size of the studies and includes four or five relatively small-scale studies using a variety of approaches, but predominantly either cognitive behaviour therapy (CBT) or cognitive and behavioural techniques, or interpersonal social rhythms therapy (IPSRT).16 These small-scale studies demonstrate that psychological treatments appear to have some benefit in preventing relapse, although there is a hint that the briefer interventions17,18 are more effective in preventing mania rather than depression. As shown in Figure 18.1, a metaanalysis of relapse rates using a fixed effects model shows that, when data from all the studies are combined,17–20 the odds ratio for relapse in the
Ch 18
7/4/05
4:00 pm
168
Page 168
Bipolar disorder: the upswing in research & treatment
intervention as compared with the control group is 0.31 (95% confidence interval (CI) 0.15–0.64; p < 0.002). A similar statistically significant result is obtained for the random effects model. The three largest studies published in the literature used either CBT,21 family therapy22 or group psycho-education.23 A separate meta-analysis of outcome data from these RCTs using fixed and random effects models demonstrates that the odds ratio (OR) for relapse in the active as compared with the control treatment groups (OR 0.37; 95% CI 0.23–0.60; p < 0.001) is similar to that reported for the earlier studies (Figure 18.2). There appear to be some differences in odds ratios between studies, but these may relate to sample characteristics (e.g. proportion of participants who met criteria for bipolar I or bipolar II disorders) as well as similarities or differences in the style and content of the treatments. Importantly, these interventions all have a significant effect on rates of depressive relapses. Perhaps for the treatment of syndromal and subsyndromal symptoms of bipolar depression it is beneficial to use more complex and more extended interventions. The problem with these studies, and the same is true of the majority of medication trials, is that they report treatment efficacy in less representative groups of patients with bipolar disorders, rather than effectiveness with a clinical sample more typical of those being treated in general adult psychiatry services in the UK. Many of the RCTs were single centre or specialist centre studies, which often recruited selected sub-populations of individuals with bipolar disorders. In some studies, participants had to be euthymic for more than 1 year to be included and individuals with comorbid disorders were largely excluded. Some RCTs utilized self-report, postal returns of self-ratings or unblinded researcher assessments of outcomes, and the trial results often reported cross-sectional analyses of relapse rates rather than longitudinal data such as weekly symptom fluctuations or survival curve analyses. The outcome data reported in some of the
Citation 200323
Colom et al Lam et al 200321 Miklowitz et al 200022 Fixed Combined (3) Random Combined (3)
N Total
Effect
p Value
120 96 101
0.41 0.26 0.46
0.02 0.00 0.08
317 317
0.37 0.37
0.00 0.00
0.01
0.1
1
10
100
Favours psychotherapy
Figure 18.2 Odds ratios of relapse rates in recent larger-scale trials.
Ch 18
7/4/05
4:00 pm
Page 169
Psychological treatments
169
earlier studies made it difficult to disentangle relapses from dropouts and it was not always clear in the RCTs whether the results were from intent to treat or per protocol analyses.11 Overall, it seems that there is promising but not conclusive data about the benefits of psychological treatments in bipolar disorders. However, there is a need for a pragmatic large-scale trial looking at effectiveness, undertaken in a number of centres, with intensive follow-up assessments, including blind raters and with therapists trained in a way that is comparable to what usually happens in the NHS, i.e. with therapists competent in delivering psychological therapies then trained to deliver a therapy specifically designed for individuals with bipolar disorders. This is what we attempted in our MRC study.13
Preliminary report on the MRC study Just over 250 persons with bipolar disorder were recruited to the RCT from five centres across Britain. Potential participants were identified via case registers, hospital data or Care Programme Approach records and then the relevant community teams were approached to ask whether they would allow us to assess the individual for the trial. The main inclusion criterion was that the individual had had at least one manic episode in the previous 12 months, giving them at least a 50% risk of further relapse in the next 12–18 months. The sample was fairly typical of bipolar disorder populations (about 60% female), their mean age of onset was about 25 years, most had been diagnosed with bipolar disorder for about 17 or 18 years and they had had many previous episodes (median 12). Importantly, because we kept our inclusion criteria very broad, the only two factors that led to exclusion were inability to give informed consent or high risk of suicide. Thus, the population is much more heterogeneous than in any previous RCTs – about 30% of our population were currently unwell, about a third had a co-morbid axis 1 disorder, about a fifth had co-morbid substance misuse, 7–10% had antisocial personality disorder or borderline personality disorder, half had a lifetime history of substance misuse, 21% had a history of violence towards others, 22% had been in trouble with the police, including a small number who had been in prison, and a number of people (at least one in five) had made significant and severe suicide attempts with intent to kill themselves. The acute treatment phase lasted 6 months and individuals were randomly allocated to usual psychiatric care and treatment or usual treatment
Ch 18
7/4/05
4:00 pm
170
Page 170
Bipolar disorder: the upswing in research & treatment
plus 22 sessions of CBT. All subjects were then followed up every 8 weeks for a further 12 months. The primary outcome measure was time to first relapse. The main analysis showed no between-group differences in time to manic, depressive or any type of relapse. However, if the patient population was separated into those with and those without the adverse clinical features highlighted in this chapter (co-morbidity, severe suicide attempts, persistent syndromal or subsyndromal symptoms, multiple episodes, etc.), we discovered that those subjects with bipolar disorder and additional adverse clinical characteristics had a highly significant reduction in survival time compared with those with bipolar disorder but no evidence of additional adversity. The latter sub-group are typical of those who are usually included in all types of treatment trials, and had comparable reductions in relapse rates to those receiving CBT added to treatment as usual, having a relapse rate of less than half that of subjects receiving usual treatment alone. Those patients with complex bipolar disorders who are difficult to treat have a poorer outcome with and without adjunctive CBT.
Current treatment guidelines The main treatment guidelines on bipolar disorders, like those on other mood disorders (e.g. American Psychiatric Association24), acknowledge the role and importance of psychological therapies as well as medication. However, the guidelines all suggest that specialist psychological therapies are a precious resource that should therefore be targeted at the cases that are most difficult to treat. Whilst the notion that those with the most complex problems should be provided with the greatest input seems to make clinical sense, the data from the RCTs reviewed appear to undermine this idea. Persons who appear consistently to benefit from adjunctive psychological therapy are those at high risk of recurrence but do not have other complications or adverse clinical features that commonly accompany bipolar disorders; i.e. the best candidates for psychological therapies are those with relatively fewer previous episodes, who are at above average risk of a further relapse, but who do not appear to have done well with medication and out-patient support, either because of some ineffectiveness of the prescribed medication or because they are not taking medication. The problem is that this group represents only about one in five of the individuals with bipolar disorders in contact with general adult psychiatry services and, very importantly, the use of psychological therapies in this group would appear to go against current guideline recommendations on the use of adjunctive
Ch 18
7/4/05
4:00 pm
Page 171
Psychological treatments
171
psychological treatments. Further research is now needed to establish whether those individuals with more complex presentations of bipolar disorders require a longer course of therapy, e.g. CBT plus maintenance sessions, or whether a different model of psychological therapy needs to be introduced to help deal with the multiple psychological and social problems they confront over and above managing the consequences of bipolar disorder.
Conclusions The review of published trials on bipolar disorders offers evidence for the efficacy of adjunctive psychological treatments. Indeed, the evidence is that adding simple interventions, such as 6–10 sessions targeted at medication adherence or relapse prevention, is relatively cost-effective, particularly as manic relapses (the most frequent cause of hospitalizations) are significantly reduced. Self-help and psycho-education are also very useful and will probably avert some of the relapses in some individuals. The extended individual therapies such as CBT or family therapy appear to be useful for extinguishing subsyndromal or syndromal symptoms of bipolar depression. However, the main problem at the moment is that the treatment guidelines for bipolar disorders are based on extrapolations from unipolar disorders and there is simply not sufficient evidence on effectiveness to support this. The MRC study is the largest study that has ever been undertaken of psychological treatments in bipolar disorders and addresses this question of the benefits of such therapies in general clinical settings. The findings suggest it may be necessary to adapt the current psychological treatment models to take into account the heterogeneity of cases seen in day-to-day practice. At this time we have evidence that suggests that psychological therapies work very well for the patients with the middle group of outcomes, and that 30% do not respond to mood stabilizers alone, but we need more studies to establish what the best approach will be for the cases of bipolar disorders that are the most difficult to treat.
References 1.
American Psychiatric Association, Practice guideline for the treatment of patients with bipolar disorder (revision). Am J Psychiatry 2002; 159(Suppl 4):1–50.
Ch 18
7/4/05
4:00 pm
172 2.
3. 4. 5.
6.
7.
8. 9. 10. 11. 12.
13. 14. 15.
16. 17. 18.
19. 20.
Page 172
Bipolar disorder: the upswing in research & treatment Goodwin GM; Consensus Group of the British Association for Psychopharmacology, Evidence-based guidelines for treating bipolar disorder: recommendations from the British Association for Psychopharmacology. J Psychopharmacol 2003; 17:149–173. López AD, Murray CJ, The global burden of disease. Nat Med 1998; 4:1241–1243. Angst F, Stassen HH, Clayton PJ, Angst J, Mortality of patients with mood disorders: follow-up over 34–38 years. J Affect Disord 2002; 68:167–181. American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th edn. American Psychiatric Association: Washington, DC, 1994. Bieling PJ, MacQueen GM, Marriot MJ et al, Longitudinal outcome in patients with bipolar disorder assessed by life-charting is influenced by DSM-IV personality disorder symptoms. Bipolar Disord 2003; 5:14–21. Black DW, Winokur G, Hulbert J, Nasrallah A, Predictors of immediate response in the treatment of mania: the importance of comorbidity. Biol Psychiatry 1988; 24:191–198. Scott J, Psychotherapy for bipolar disorder – An unmet need? Br J Psychiatry 1995; 167:581–588. Gitlin MJ, Swendsen J, Heller TL et al, Relapse and impairment in bipolar disorder. Am J Psychiatry 1995; 152:1635–1640. Harrow M, Goldberg JF, Grossman LS et al, Outcome in manic disorders. A naturalistic follow-up study. Arch Gen Psychiatry 1990; 47:665–671. Scott J, Cognitive therapy for bipolar disorders. Expert Rev Neurother 2002; 2:573–581. Judd LL, Akiskal HS, Schlettler PJ et al, The long-term natural history of the weekly symptomatic status of bipolar 1 disorder. Arch Gen Psychiatry 2002; 59: 530–537. Scott J, Cognitive Therapy in Bipolar Disorders. ENCP: Stockholm, Sweden, 2004. Gutierrez MJ, Scott J, Psychological treatments for bipolar disorders: A review of randomised controlled trials. Eur J Psychiatry 2004; 254:92–98. Scott J, Gutierrez MJ, The current status of psychological treatments in bipolar disorders: a systematic review of relapse prevention. Bipolar Disord 2004; 6:498–503. Frank E, Interpersonal and social rhythm therapy prevents depressive symptomatology in bipolar I patients. Bipolar Disord 1999; 1(Suppl):13. Cochran SD, Preventing medical noncompliance in outpatient treatment of bipolar disorders. J Consult Clin Psychol 1984; 52:873-878. Perry A, Tarrier N, Morriss R et al, Randomised controlled trial of efficacy of teaching patients with bipolar disorder to identify early symptoms of relapse and obtain treatment. BMJ 1999; 318:149–153. Lam DH, Bright J, Jones S et al, Cognitive therapy for illness: a pilot study of relapse prevention. Cogn Ther Res 2000; 24:503–520. Scott J, Garland A, Moorhead S, A pilot study of cognitive therapy in bipolar disorders. Psychol Med 2001; 31:459–467.
Ch 18
7/4/05
4:00 pm
Page 173
Psychological treatments 21.
22.
23.
24.
173
Lam DH, Watkins ER, Hayward P et al, A randomized controlled study of cognitive therapy for relapse prevention for bipolar affective disorder: outcome of the first year. Arch Gen Psychiatry 2003; 60:145–152. Miklowitz DJ, Simoneau TL, George EL et al, Family-focused treatment of bipolar disorder: 1-year effects of a psychoeducational program in conjunction with pharmacotherapy. Biol Psychiatry 2000; 48:582–592. Colom F, Vieta E, Martinez-Aran A et al, A randomized trial on the efficacy of group psychoeducation in the prophylaxis of recurrences in bipolar patients whose disease is in remission. Arch Gen Psychiatry 2003; 60:402–407. American Psychiatric Association, Practice guideline for major depressive disorder in adults. Am J Psychiatry 1993; 150:1–26.
Ch 18
7/4/05
4:00 pm
Page 174
Ch 19
7/4/05
4:01 pm
Page 175
chapter 19
Lithium, the forgotten drug Mario Maj In this chapter I will address the following questions: (1) Is lithium to some extent a forgotten drug? (i.e. has the use of lithium in bipolar disorder declined drastically in recent years?); (2) If so, why has lithium been forgotten? (i.e. why has its use declined?). The answer to the first question is yes. Right now, in the USA, lithium is less frequently prescribed than valproate for the treatment of bipolar disorder. For instance, Scott–Levin’s Physician Drug and Diagnosis Audit showed that, in May 2002, 1858 prescriptions for bipolar disorder were for lithium and 2116 were for valproate.1 The decline in the use of lithium is also reflected in a similar way in the sale figures in European countries. For instance in Italy, in the year 2002, the maintenance treatment for bipolar disorder included in 59% of cases at least one antipsychotic (with olanzapine and risperidone surpassing haloperidol as the most frequently prescribed drugs of this group); in 37% of cases at least one anticonvulsant (with valproate surpassing carbamazepine as the most frequently prescribed drug of this group); in 32% of cases lithium and in 25% of cases any antidepressant. Several European clinicians probably now share the experience described a few years ago in the USA by Ronald Fieve: ‘For years, in an average week I have seen four to six bipolar patients who have been given smatterings of all the above drugs (anticonvulsants and new antipsychotics) for a few weeks to 2 years, by one to five psychiatrists. The patient and family report that lithium has not been used, or was used for a short time and has not worked’.2 Why has the use of lithium declined so drastically? Five factors have in my opinion been more or less important, and I will briefly discuss four of them. I will not discuss in detail the remaining one, i.e. the doubts recently expressed about the impact of lithium on the course of bipolar disorder in ordinary clinical conditions, because this issue has been the subject of several reviews (for example, reference 3). I do not believe that this factor contributed significantly to the recent decline in the use of the drug.
Ch 19
7/4/05
4:01 pm
176
Page 176
Bipolar disorder: the upswing in research & treatment
A much more powerful factor has been, in my opinion, the fact that lithium treatment is a very demanding one for both the psychiatrist and the patient. On the psychiatrist’s side, lithium treatment requires training, time, attention and general medical skills; on the patient’s side, it requires collaboration, tolerance of side-effects, diligence and persistence. The patient’s side, and in particular the burden of side-effects, has been covered extensively in the literature, whereas the psychiatrist’s side has not frequently been a focus of attention. I will refer now briefly to the situation of psychiatric practice in my country (Italy), arguing that some recent developments in this practice have worked against the use of a drug with the characteristics of lithium. Although the case of Italy may be extreme, some of the developments I will describe are occurring now or are going to occur in several other European countries. The first and the most significant development has been the move from hospital-based to community-based psychiatric practice; this means that the vast majority of Italian psychiatrists do not work now in a hospital environment, but in centres that are sometimes several hundreds of kilometres away from the nearest hospital. No laboratory is available in these centres and, if the consultation of another specialist (for instance, an endocrinologist or a nephrologist) is required, the patient has to be referred to the nearest hospital. Psychiatric training increasingly reflects this new situation, with an increasing focus on social psychiatry and community mental health care, and a decreasing emphasis on proper medical aspects (for instance, most young psychiatrists in Italy have never seen an electroconvulsive therapy session and several of them are not trained adequately in the use of lithium). On the other hand, the workload for psychiatrists operating in the public sector has increased significantly, which means that much less time can be devoted to the individual patient. Finally, the press and the public have been recently sensitized to incidents in medical practice, including those involving the use of drugs, and psychiatrists’ alertness to the risk of these incidents has increased significantly. In this situation, it is not surprising that psychiatrists show a preference for drugs that are easy to use, not problematic, not potentially toxic, unlikely to cause incidents, possibly ‘transnosographic’ (thus not requiring a sophisticated diagnostic assessment), which do not require frequent physical examinations or laboratory tests and which are unlikely to require consultations of other specialists. Lithium certainly does not correspond to this identikit, nor does clozapine, arguably another forgotten drug, which is received now by just 8% of Italian patients with a diagnosis of schizophrenia. The example of Italy may be extreme, but similar problems with training and a preference for drugs that are non-problematic and easy to use have been noted
Ch 19
7/4/05
4:01 pm
Page 177
Lithium, the forgotten drug
177
also in the USA. In that country, according to Fieve, ‘most residents are not adequately trained to the subtleties of lithium treatment… Therefore, on graduation they are poorly equipped and tend to underuse lithium as the first line of treatment. Instead, they begin the new bipolar patient on an antiepileptic, since it is easier to use and requires less knowledge’.3 Another powerful factor obviously contributing to the recent decline in the use of lithium has been the introduction of several new drugs which have been found effective in bipolar disorder and which, contrary to lithium, are backed by significant commercial interests. The availability of these new drugs is, of course, very welcome. For many years, there has been no significant alternative for bipolar patients who did not respond or responded partially to lithium or did not tolerate lithium or in whom lithium was contraindicated. Now several alternatives are available. Of course, drug companies producing these drugs pursue profit as their main objective but, if the drugs are effective, it can be said that to a large extent there is a convergence of interests between psychiatrists, companies, patients and patients’ families. It is also understandable that drug companies have their own marketing strategies and that psychiatrists are a target for these strategies, although a code of conduct for both psychiatrists and drug companies would be very useful in this respect. I believe, however, that we should make a much more significant effort to preserve the integrity and independence of our scientific publications and of our research in the area of treatment of bipolar disorder. I have witnessed in this area in recent years several unpleasant developments. I have seen several biased clinical guidelines, and many biased and tendentious reviews and editorials. I am aware of some cases of ghostwriting and of some attempts by sponsors to obtain changes in chapters of volumes in favour of their drugs. I have seen some cases of double or triple publication of the same study with different first authors. I have seen several journal supplements with papers that had not been peer-reviewed, and I am aware of some cases of selective publication. I have seen several clinical trials with evident biases in favour of new drugs and against the comparison drug, which was lithium: for instance, ‘enriched’ designs favouring the drug under testing; trials in which serum lithium levels up to 2.7 mEq/l were reached, so that not surprisingly there was a very high dropout rate in the lithium group due to intolerance; reports on clinical trials focusing on secondary outcome measures favouring the new drug, so that the trial was presented as positive, even if on the primary outcome measure the drug was not superior to placebo. These problems and biases affect only a minority of the available publications and trials, but I believe that even these relatively few cases are
Ch 19
7/4/05
4:01 pm
178
Page 178
Bipolar disorder: the upswing in research & treatment
of concern, because they may impair the credibility of the entire system. A recent paper published in the British Medical Journal,4 entitled ‘Evidence b(i)ased medicine – selective reporting from studies sponsored by pharmaceutical industry’, focusing on drugs for depression, represents a clear warning in this respect. A fourth, less powerful factor contributing to the decline in the use of lithium has been the recent change in the phenomenology and/or conceptualization of bipolar disorder and the fact that lithium is not equally effective throughout the ‘bipolar spectrum’. That lithium is less efficacious in atypical and complicated varieties of bipolar disorder than in the classical manic–depressive forms is now sufficiently documented. What is not documented, however, is that lithium is not efficacious in these atypical or complicated varieties, or that other drugs are superior to lithium in these forms, as is frequently stated in reviews and clinical guidelines. For instance, lithium does exert an impact on bipolar disorder with mood-incongruent psychotic features5 and in rapidly cycling bipolar disorder,6 and a recent metaanalysis showed no superiority for anticonvulsants over lithium in rapid cyclers.7 The fifth factor, which is more significant than commonly believed, is related to the treatment of acute mania. Lithium is regarded as the gold standard for the treatment of mania, and new drugs proposed for use in bipolar disorder are first of all tested for their antimanic activity, and usually compared to placebo or to lithium. The fact is, however, that lithium has never been the preferred drug for the treatment of acute mania in clinical practice. Most psychiatrists prefer to use antipsychotics and several of them show an inclination to continue to prescribe for maintenance treatment the same drug they used for the manic episode. This is a significant advantage that new generation antipsychotics may have over lithium, if clinicians will be convinced that these new drugs are as quick and powerful as standard antipsychotics in the acute management of mania. It is clear from the above discussion that lithium has been to some extent forgotten in the treatment of bipolar disorder. However, did lithium perform better or worse than other old and powerful drugs challenged by new developments, such as haloperidol as an antipsychotic or tricyclics as antidepressants? It seems that lithium has been less forgotten than these other drugs. For instance, in Italy, only 26% of patients with a diagnosis of schizophrenia received haloperidol for maintenance treatment in the year 2002, and only 7% of those with a diagnosis of depression received tricyclic antidepressants. On the research side, a Medline search for the year 2002 iden-
Ch 19
7/4/05
4:01 pm
Page 179
Lithium, the forgotten drug
179
tified 943 papers dealing with lithium, versus 446 for haloperidol and 171 for amitriptyline. Finally, I have seen no recent study showing a superiority of haloperidol over second-generation antipsychotics on a significant dimension of schizophrenia or a superiority of tricyclics over the new antidepressants on a significant dimension of depression, whereas I have seen, for instance, a recent study showing the significant superiority of lithium over valproate in reducing the risk for suicide in bipolar patients.8 All these may be regarded as signals that lithium has not been, and will probably not be, at least in the near future, completely forgotten.
References 1.
2.
3. 4.
5.
6.
7. 8.
Goodwin FK, Rationale for long-term treatment of bipolar disorder and evidence for long-term lithium treatment. J Clin Psychiatry 2002; 63(Suppl 10):5–12. Fieve RR, Lithium therapy at the millennium: a revolutionary drug used for 50 years faces competing options and possible demise. Bipolar Disord 1999; 1:67–70. Maj M, The impact of lithium prophylaxis on the course of bipolar disorder: a review of the research evidence. Bipolar Disord 2000; 2:93–101. Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B, Evidence b(i)ased medicine – selective reporting from studies sponsored by pharmaceutical industry: review of studies in new drug applications. BMJ 2003; 326:1171–1173. Maj M, Pirozzi R, Bartoli L, Magliano L, Long-term outcome of lithium prophylaxis in bipolar disorder with mood-incongruent psychotic features: a prospective study. J Affect Disord 2002; 71:195–198. Baldessarini RJ, Tondo L, Hennen J, Effects of rapid cycling on response to lithium maintenance treatment in 360 bipolar I and II disorder patients. J Affect Disord 2000; 61:13–22. Tondo L, Hennen J, Baldessarini RJ, Rapid-cycling bipolar disorder: effects of long-term treatments. Acta Psychiatr Scand 2003; 108:4–14. Goodwin FK, Fireman B, Simon GE et al, Suicide risk in bipolar disorder during treatment with lithium and divalproex. JAMA 2003; 290:1467–1473.
Ch 19
7/4/05
4:01 pm
Page 180
Ch 20
7/4/05
4:02 pm
Page 181
chapter 20
Advantages and disadvantages of atypical antipsychotics or valproate in bipolar disorder John Cookson This chapter concerns the comparison between atypical antipsychotics and valproate used for acute mania and for the prophylaxis of bipolar disorder. The comparison with classical antipsychotics is also considered. Discussion of antipsychotics and valproate in acute mania is topical in relation to current guidelines for the management of bipolar disorder. The revised American Psychiatric Association Guidelines1 propose a choice between lithium, valproate or an atypical antipsychotic for less severe mania. For severe mania, the Guidelines advise a combination of an antipsychotic and valproate or lithium. However, it is arguable whether as a first-line treatment a combination is better than using one or other of these drugs alone, as is recommended by the Guidelines of the British Association for Psychopharmacology.2 The comparison of antipsychotics and valproate for mania should include consideration of speed of action, mechanism of action, the spectrum of symptoms that are controlled, what pattern of mania may be affected, whether the presence of psychotic features makes a difference, and whether mixed mania responds differently. The study of psychotic mania by Keck et al3 indicated that, if a sufficiently large dose of valproate (as semisodium valproate 20 mg/kg per day) is used from the start, there is a similar improvement with valproate or haloperidol (0.2 mg/kg per day). However, the generalizability of this finding may be limited, since haloperidol did not show its usual rapid onset of effect. There was improvement by only 12% in mania ratings after 24 hours, as opposed to a 30% improvement in 20 minutes after intravenous haloperidol administration in other studies, which was not due to sedation,
Ch 20
7/4/05
4:02 pm
182
Page 182
Bipolar disorder: the upswing in research & treatment
as was confirmed in our own study.4 A second study of valproate loading by Hirschfeld et al5 showed a delay in onset of antimanic effect with valproate of about 48 hours. For rapid tranquillization, which is often needed in acute mania, it is necessary to use drugs that provide initial control of the manic state within minutes or hours, rather than days. A commonly used approach is to use combined treatment with 1 or 2 mg of lorazepam and 5 or 10 mg of haloperidol, given intramuscularly. Haloperidol can work quite rapidly but often has unpleasant side-effects, including akathisia, and less often dystonia. To avoid the side-effects we require an intramuscular atypical antipsychotic, such as olanzapine or ziprasidone. In a placebo-controlled trial, Meehan et al studied intramuscular olanzapine 10 mg, with lorazepam 2 mg as the active comparator.6 Using a scale measuring excitement with items from the Positive and Negative Syndrome Scale (PANSS), there was rapid improvement after intramuscular olanzapine administration, evident within 30 minutes, and greater than with 2 mg of lorazepam. Thus, antipsychotics are capable of having an antimanic effect starting within minutes and developing over 2 hours. Belmaker and colleagues have reported that valproate controls status epilepticus quickly when given intravenously, but does not have an immediate antimanic effect.7 This suggests that valproate works indirectly. The common mechanism of action of antipsychotics is blockade of dopamine receptors. Some, such as haloperidol, also block α1 receptors and olanzapine also blocks histamine receptors. Valproate, on the other hand, works on second-messenger systems and may reduce dopamine release.8 Lithium may also reduce dopamine release.9 Thus, a combination of an antipsychotic drug and valproate or lithium might be expected to have synergistic effects, improving mania.
Efficacy in mania The number needed to treat (NNT) represents the number of patients who must be treated in order for one patient to achieve the defined response – usually a 50% reduction in score on a scale such as the 11-item Young Mania Rating Scale (YMRS) – as a result of the pharmacological effect of the drug. The NNT is calculated by dividing the difference in response rate between active drug and placebo into 100 and correcting to the next highest integer. The NNT thus provides a measure of the size of effect that can be expected of the drug in a clinical situation. For a drug to be a useful monotherapy as a first-line treatment in a common and severe disorder
Ch 20
7/4/05
4:02 pm
Page 183
Advantages and disadvantages of atypical antipsychotics or valproate
183
such as mania, we have argued that the NNT for 50% improvement in severity should be in the order of 2–4.10 Two studies have compared valproate with placebo in mania. Bowden et al11 randomized patients to placebo, valproate or lithium, using a 50% reduction in scores on a mania rating scale as the criterion for improvement (Table 20.1). The NNT for 50% improvement was 5. Even though valproate is recognized as a useful drug, it was not usually sufficient as monotherapy. The placebo-controlled study by Pope et al12 showed a more dramatic effect. The placebo response rate was low, this being the only modern randomized trial in mania, showing such a low placebo response rate, probably because the patients were lithium-resistant manic in-patients. The difference in response rate was large, giving a NNT of 3, with close confidence intervals. The first modern controlled trials of antipsychotics in mania were those on olanzapine designed under the guidance of Mauricio Tohen and financed by Eli Lilly & Co. These and other large-scale studies of atypical antipsychotics answer important questions. In most of the placebo-controlled studies, the criterion for response was a 50% improvement on the YMRS. In the first study by Tohen et al,13 olanzapine (up to 15 mg a day) had a NNT of 4, with a confidence interval of 3–10. The antipsychotics that have been shown to improve mania in parallelgroup placebo-controlled trials are: haloperidol (in two published studies as active comparator), olanzapine (in two), risperidone (in three), quetiapine (in two), ziprasidone (in two) and aripiprazole (in two). The most impressive result was in a study of risperidone conducted in India by Khanna et al.14 The dropout rate for patients on risperidone (average 5.8 mg/day) was low, allowing time for a high response rate, and NNT of 3 (2–4) (Table 20.2). This is a true reflection of the potential of antipsychotics in mania: a large and reliable antimanic effect.
Table 20.1 Number needed to treat in placebo-controlled trials of valproate in mania Treatment 3 weeks Lithium (n = 35) Placebo (n = 73) Divalproex (n = 68)
Response rate (%)
Criterion 50% less SADS-M Bowden et al, 199411
49 25 48
Divalproex (n = 23) 50% less MRS
45
Placebo (n = 20)
9
Pope et al,
199112
Difference Number needed from placebo (%) to treat (95% CI) 24
5 (3–22)
23
5 (3–14)
36
3 (2–9)
SADS-M, Schedule for Affective Disorders and Schizophrenia; MRS, mania rating scale.
Criterion for response
Sites Extra drugs
Numbers for ITT
Inclusion
LZP for 10 days
(5.6 mg/day) YMRS
50% reduction
YMRS ≥ 20
Dropouts
14.6
4.8
(%)
inefficacy
for
ITT, intention to treat; LZP, lorazepam; YMRS, Young Mania Rating Scale.
(n = 144)
Placebo
India
Mean modal
Khanna, Vieta et al, 200314 or mixed
(n = 146)
DSM-IV Manic
3 weeks
Risperidone
criteria
Duration Authors
Dropouts
2.1
3.4
(%)
events
for adverse
Other
12.5
2.8
(%)
dropouts
Response
36
73
(%)
Difference
37
placebo
from
Number
(2–4)
3
to treat
needed
184
(mean dose/day)
4:02 pm
Treatment
7/4/05
Table 20.2 Placebo-controlled parallel-group randomized trial of monotherapy with risperidone in mania
Ch 20 Page 184
Bipolar disorder: the upswing in research & treatment
Ch 20
7/4/05
4:02 pm
Page 185
Advantages and disadvantages of atypical antipsychotics or valproate
185
There have been two direct comparisons of olanzapine and valproate in mania, one conducted by Tohen et al for the manufacturers of olanzapine,15 the other for the manufacturers of valproate by Zajecka et al.16 Both studies appear to show that olanzapine produces a slightly greater and faster improvement on the YMRS. In the larger of the two studies, which also used a slightly higher dose of olanzapine, there was a significant difference. The whole range of manic symptoms improved with both drugs within 3 weeks, the smallest improvement being in the item for insight. Subdividing the sample into patients with pure or mixed manic episodes, or into those with or without a rapid-cycling course, or those with or without psychotic features, the degree of improvement was not found to depend on the type of mania, with one exception. Olanzapine was superior to valproate in the non-psychotic manic patients, whereas for the psychotic manic patients valproate and olanzapine were equally effective.15 However, the study was not statistically powered to detect differences between subgroups of manic patients.
Treatment side-effects In the study by Tohen et al, the side-effects more common with olanzapine than valproate were drowsiness, dry mouth and increased appetite; the side-effects more common with valproate were gastrointestinal disturbance and nausea.15 In the study by Zajecka et al, further side-effects, commoner with olanzapine, were drowsiness, weight gain, rhinitis (from noradrenaline α1 receptor blockade), oedema and slurred speech, probably related to drowsiness.16 Broader evidence would indicate that the side-effects of valproate include polycystic ovary syndrome (particularly in adolescent women), hair loss (usually transient and perhaps counteracted by zinc supplementation), fetal valproate syndrome (if used in pregnancy), spontaneous bruising or bleeding, pancreatitis and a risk (especially in children) of liver damage.
Depression in mania Most manic patients have additional depressive symptoms and some patients switch from mania to depression during treatment. This switch has been reported by Zarate et al17 to occur more often when patients are treated with a typical antipsychotic, and anticholinergic medication is withheld, than with placebo.
Ch 20
7/4/05
4:02 pm
186
Page 186
Bipolar disorder: the upswing in research & treatment
In the direct comparison between valproate and olanzapine, both drugs reduced depressive symptoms.15 Thus, as mania gets better, the accompanying depressive symptoms also usually improve, proportionately with the improvement in mania. Some drugs may be better at preventing a switch from mania into depression. In a comparison by Tohen et al18 of olanzapine with haloperidol in mania, patients on haloperidol had a 12% switch rate into depression within 6 weeks, while those on olanzapine had a 5% switch rate. The only other comparative study suggesting a lower switch rate in comparison with haloperidol is that by Brecher et al with quetiapine.19 There was a trend for olanzapine to be better than valproate in preventing a switch into depression, but this was not significant.15
Combination treatment Several studies have demonstrated an advantage of combining an antipsychotic with lithium or valproate, over lithium or valproate alone. Conversely, the addition of valproate to classical antipsychotics (mainly haloperidol) has been shown by Muller-Oerlinghausen et al20 to produce greater improvement than addition of placebo. The majority of these studies involved a design in which patients who had not responded to one drug administered for at least 3 weeks had additional treatment with the combination of drugs using a placebo control for the added drug. In some studies a proportion of patients commenced on the combination without previous treatment, and the control group received only the lithium, valproate or carbamazepine with placebo instead of antipsychotic. There have been no studies of combination treatment in which a control group receive only placebo. It is therefore not possible to determine directly the size of effect of giving combination treatment to drugfree patients with mania over giving placebo alone, or to determine the size of advantage of combination therapy over monotherapy initiated in drugfree patients with mania. The studies of combination treatment have proved that several antipsychotics provided additional efficacy when added to lithium or valproate compared with lithium or valproate alone. These are: haloperidol, risperidone, olanzapine and quetiapine. A similar study with ziprasidone was negative. In combination studies with olanzapine, this drug was shown by Tohen et al21 to provide additional efficacy when added in patients who had already received lithium or valproate for at least 3 weeks. The effect was
Ch 20
7/4/05
4:02 pm
Page 187
Advantages and disadvantages of atypical antipsychotics or valproate
187
statistically significant in the subgroup on valproate but only at trend level in the smaller subgroup on lithium. Risperidone in combination with lithium or valproate was shown by Sachs et al22 and Yatham et al23 to be more effective than either lithium or valproate alone, and as efficacious as haloperidol in combination with either lithium or valproate,22 in patients with manic or mixed episodes. The advantage of adding risperidone was far more evident in those who had already been on a mood stabilizer for at least 2 weeks before randomization, than in those who started a mood stabilizer shortly before starting risperidone or placebo.23 In total, these combination studies demonstrate that addition of the second component of combined treatment, achieves an added response with a NNT of 5 or 6. This would suggest that not every patient needs the combination, but that some patients do better with a combination, and others fail to respond even with combined antipsychotic and lithium or valproate.
Maintenance treatment The only large-scale placebo-controlled maintenance trial of valproate, by Bowden et al,24 was very disappointing. The benefit of valproate over placebo was very small and applied mainly to depressive symptoms in a secondary analysis. In the direct comparison by Tohen et al15 of olanzapine and valproate for acute mania, there was a small difference in favour of olanzapine, in terms of days spent with symptoms over the subsequent year of the study. By contrast, there is evidence from Tohen et al25 that continued treatment with olanzapine plus lithium or valproate, after remission of mania on a combination of the two, led to a lower risk of symptomatic relapse of bipolar disorder over the subsequent year.
Weight gain The study of valproate against olanzapine16 demonstrated that olanzapine produced more weight gain than valproate over the course of 12 weeks. In the 1-year follow-up of the study comparing olanzapine and valproate,15 by the end of the year patients showed similar weight gain with valproate or olanzapine. However, the extent of weight gain on olanzapine in this study (mean 6 kg) was considerably less than in studies of olanzapine in schizophrenia. Weight gain on olanzapine reached a plateau at about 19 weeks,
Ch 20
7/4/05
4:02 pm
188
Page 188
Bipolar disorder: the upswing in research & treatment
which is earlier than in schizophrenia, where the weight gain with olanzapine continued for up to 1 year.
Comparison of classical and atypical antipsychotics Placebo-controlled monotherapy trials of haloperidol in mania Two monotherapy studies have included haloperidol as an active comparator: the risperidone study of Eerdekens et al26 and the quetiapine study of Brecher et al.19 In the study of Eerdekens et al,26 the haloperidol dose started at 4 mg/day and was adjusted to 2–12 mg/day by day 5. The timecourse of improvement was similar to that with risperidone, and by day 21 the NNT for 50% improvement was 8 (95% confidence interval (CI) 4–36). This is far larger than one would expect with the most commonly used antimanic drug of the previous decade; this might be either because the mean modal dose of haloperidol was only 8 mg/day, or because the patients in the trial were in some ways not typical of routine clinic patients and were more resistant to treatment. A comparator group on haloperidol (up to 8 mg/day) was also included in the study by Brecher et al19 of quetiapine (up to 800 mg/day) and placebo, analysed at 3 and 12 weeks. At 3 weeks the response rate (50% reduction in YMRS score) on haloperidol, on a mean dose of only 5.2 mg/day, was 55% compared with 35% on placebo, giving a NNT of 5 (95% CI 3–16). There were more dropouts on placebo than on haloperidol or quetiapine, so that the analysis using last observations carried forward was biased in favour of the active drugs, and especially so after 3 weeks when more patients on placebo or haloperidol than on quetiapine dropped out. By 12 weeks the response rate on haloperidol was 70% and on placebo 39%, giving a NNT of 4 (95% CI 3–6). Side-effects in the form of extrapyramidal symptoms were much more common on haloperidol (59.6%) than on placebo (15.8%), as was akathisia (33.3% on haloperidol and 5.9% on placebo). Somnolence occurred more often with haloperidol (9.1%) than placebo (5%).
Comparative randomized controlled trials of antipsychotics in mania without placebo In a comparative trial in mania, in which additional lorazepam was permitted, risperidone showed similar efficacy to haloperidol or lithium.27
Ch 20
7/4/05
4:02 pm
Page 189
Advantages and disadvantages of atypical antipsychotics or valproate
189
In the largest randomized comparative study of haloperidol,18 it was compared with olanzapine over 6 and 12 weeks. Among patients on haloperidol (up to 15 mg/day, mean modal dose at the 6th week 7 mg/day), the proportion responding (50% reduction in YMRS score) at 6 weeks was 74%; the proportion showing syndromal remission (according to DSM-IV) was 44%. Improvement in mania scores (YMRS) was greater for haloperidol than for olanzapine at 6 weeks, but not different at 12 weeks. Extrapyramidal symptoms occur to a much less extent with olanzapine or quetiapine than with haloperidol. In the largest comparative trial (n = 234) treatment-emergent akathisia was observed in 40% on haloperidol and 10% on olanzapine, dystonia in 6.8% and 1.3%, and Parkinsonism in 54% and 13%, respectively.
Conclusions Atypical antipsychotics and valproate are effective in reducing the symptoms of mania, although neither is often sufficient as monotherapy to bring about marked improvement in severe mania. Valproate probably acts a little less quickly. Classical antipsychotics remain useful for the rapid control of a severely agitated manic person, and several are available for use intramuscularly, as are olanzapine and ziprasidone. Classical antipsychotics have the profound disadvantage of inducing extrapyramidal side-effects, including dystonia and akathisia, which is particularly unwelcome. Olanzapine and quetiapine may produce a lower switch rate into depression than haloperidol, and interestingly these two drugs may also have some antidepressant properties. However, both drugs (in the doses used in clinical trials) seem also to produce slower improvement in mania than haloperidol. The complexity of diagnosing and treating bipolar mixed states is discussed by Cookson and Ghalib.28 Antipsychotics are useful in bipolar disorder, both acutely in mania and in long-term prophylactic treatment. Valproate is useful in severe mania, but its value in prophylaxis is unproven and requires further investigation. Intriguingly, there is preliminary evidence that, whereas antipsychotics and (high doses of) valproate may be equally effective in severe or psychotic mania, antipsychotics including olanzapine may be more effective than valproate for controlling non-psychotic or milder mania. This might indicate that when patients are educated about what steps to take to avert early manifestations of recurrence of mania, as described by Perry et al,29 access
Ch 20
7/4/05
4:02 pm
190
Page 190
Bipolar disorder: the upswing in research & treatment
to an antipsychotic may be more important than valproate, and certainly more effective than increasing the dose of lithium they are taking.
References 1.
2.
3. 4.
5.
6.
7. 8.
9. 10.
11. 12. 13.
14.
Hirschfeld RMA, Bowden CL, Gitlin MJ et al, Practice guideline for the treatment of patients with bipolar disorder (Revision). Am J Psychiatry 2002; 159(4 Suppl):1–50. Goodwin GM, Evidence-based guidelines for treating bipolar disorder: recommendations from the British Association for Psychopharmacology. J Psychopharmacol 2003; 17:149–173; discussion 147. Keck PE, McElroy SL, Tugrul KC, Bennett JA, Valproate oral loading in the treatment of acute mania. J Clin Psychiatry 1993; 54:305–308. Cookson JC, Moult PJA, Wiles D, Besser GM, The relationship between prolactin levels and clinical ratings in manic patients treated with oral and intravenous test doses of haloperidol. Psychol Med 1983; 13:279–285. Hirschfeld RM, Allen MH, McEvoy JP et al, Safety and efficacy of oral loading divalproex sodium in acutely manic bipolar patients. J Clin Psychiatry 1999; 60: 815–818. Meehan K, Zhang F, David S et al, A double-blind, randomized comparison of the efficacy and safety of intramuscular injections of olanzapine, lorazepam, or placebo in treating acutely agitated patients diagnosed with bipolar mania. J Clin Psychopharmacol 2001; 21:389–397. Applebaum J, Levine J, Belmaker R, Intravenous fosphenytoin in acute mania. J Clin Psychiatry 2004; 64:408–409. Yatham LN, Liddle PF, Shiah IS et al, PET study of [(18)F]6-fluoro-L-dopa uptake in neuroleptic- and mood-stabilizer-naive first-episode nonpsychotic mania: effects of treatment with divalproex sodium. Am J Psychiatry 2002; 159:768–774. Ferrie L, Ferrier N, Young AH, 2003. Poster presentation at the Summer Meeting, British Association for Psychopharmacology. Cookson JC, Taylor D, Katona C, Placebo effects, evaluating evidence, and combining psychotherapy. In Use of Drugs in Psychiatry: The Evidence From Psychopharmacology. Gaskell Press: London, 2002:117–131. Bowden C, Brugger AM, Swann AC et al, Efficacy of divalproex vs. lithium and placebo in the treatment of mania. JAMA 1994; 271:918–924. Pope HG, McElroy SL, Keck PE, Hudson JI, Valproate in the treatment of acute mania: A placebo-controlled study. Arch Gen Psychiatry 1991; 48:62–68. Tohen M, Sanger TM, McElroy SL et al, Olanzapine versus placebo in the treatment of acute mania. Olanzapine HGEH Study Group. Am J Psychiatry 1999; 156:702–709. Khanna S, Vieta E, Lyons B et al, Risperidone in the treatment of acute bipolar mania: a double-blind, placebo-controlled study of 290 patients. Poster presented at the 16th Congress of the European College of Neuropsychopharmacology (ECNP), Prague, September 2003.
Ch 20
7/4/05
4:02 pm
Page 191
Advantages and disadvantages of atypical antipsychotics or valproate 15. 16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
191
Tohen M, Baker RW, Altshuler LL et al, Olanzapine versus divalproex in the treatment of acute mania. Am J Psychiatry 2002; 159:1011–1017. Zajecka JM, Weisler R, Sachs G et al, A comparison of the efficacy, safety, and tolerability of divalproex sodium and olanzapine in the treatment of bipolar disorder. J Clin Psychiatry 2002; 63:1148–1155. Zarate CA, Tohen M, Double-blind comparison of the continued use of antipsychotic treatment versus its discontinuation in remitted manic patients. Am J Psychiatry 2004; 161:169–171. Tohen M, Goldberg JF, Gonzalez-Pinto AM et al, A 12-week double-blind comparison of olanzapine versus haloperidol in the treatment of acute mania. Arch Gen Psychiatry 2003; 60:1218–1226. Brecher M, Huizar K, ElBaghdady A, Quetiapine monotherapy for acute mania associated with bipolar disorder. Poster presented at the Fifth Annual International Conference on Bipolar Disorder, Pittsburgh, PA, 12–14 June 2003. Muller-Oerlinghausen B, Retzow A, Henn FA et al, Valproate as an adjunct to neuroleptic medication for the treatment of acute episodes of mania: a prospective, randomized, double-blind, placebo-controlled, multicenter study. J Clin Psychopharmacol 2000; 20:195–203. Tohen M, Chengappa KN, Suppes T et al, Efficacy of olanzapine in combination with valproate or lithium in the treatment of mania in patients partially nonresponsive to valproate or lithium monotherapy. Arch Gen Psychiatry 2002; 59:62–69. Sachs GS, Grossman F, Ghaemi SN et al, Combination of a mood stabilizer with risperidone or haloperidol for treatment of acute mania: a double-blind, placebo-controlled comparison of efficacy and safety. Am J Psychiatry 2002; 159:1146–1154. Yatham LN, Grossman F, Augustyns I et al, Mood stabilisers plus risperidone or placebo in the treatment of acute mania: International, double-blind, randomised controlled trial. Br J Psychiatry 2003; 182:141–147. Bowden CL, Calabrese JR, McElroy SL et al, A randomized, placebo-controlled 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Divalproex Maintenance Study Group. Arch Gen Psychiatry 2000; 57:481–489. Tohen M, Chengappa KN, Suppes T et al, Relapse prevention in bipolar disorder I disorder: 18-month comparison of olanzapine plus mood stabilizer versus mood stabilizer alone. Br J Psychiatry 2004; 184:337–345. Eerdekens M, Karcher K, Grossman F, Kramer M, Risperidone monotherapy in acute bipolar mania. Poster presented at the Meeting of the World Psychiatric Association, 2003. Segal J, Berk M, Brook S, Risperidone compared with both lithium and haloperidol in mania: a double blind randomized controlled trial. Clin Neuropharmacol 1998; 21:176–180. Cookson JC, Ghalib S, The treatment of bipolar mixed states. In: Mixed States. 2004 Marneros A, Goodwin FK, eds. Perry A, Tarrier N, Morriss R et al, Randomised controlled trial of efficacy of teaching patients with bipolar disorder to identify early symptoms of relapse and obtain treatment. BMJ 1999; 318:149–153.
Ch 20
7/4/05
4:02 pm
Page 192
Ch 21
7/4/05
4:02 pm
Page 193
chapter 21
Is electroconvulsive therapy still given in bipolar disorder and does repetitive transcranial magnetic stimulation offer more? Andrew Mogg, Savitha Eranti, Graham Pluck and Declan M McLoughlin The evidence for the efficacy of electroconvulsive therapy (ECT) in depression is substantial. A recent systematic review and meta-analysis concluded that ECT is an effective treatment for depression and probably more effective than drug therapy.1 However, there are fewer research data available for the use of ECT in mania and in bipolar depression. These issues are addressed below before considering the role of repetitive transcranial magnetic stimulation (rTMS) as a potential alternative to ECT.
Electroconvulsive therapy and mania ECT is rarely used to treat mania. In the late 1970s, about 2% of referrals for ECT in Great Britain were for mania compared with 83% for depression.2 As will be seen, these figures have not changed much in subsequent years. Not surprisingly, there have been relatively few studies of ECT for mania. For example, the most recent comprehensive review on the topic was published nearly a decade ago.3 This review examined the use of ECT in mania since its introduction about 50 years previously. The authors identified just 16 primary studies, most of which were naturalistic case series or comparisons. Of these studies, seven were before 1950 and lacked certainty about both the diagnosis and the criteria for response to treatment. Data
Ch 21
7/4/05
4:02 pm
194
Page 194
Bipolar disorder: the upswing in research & treatment
combined from six retrospective studies published since the 1970s indicated that ECT was associated with either remission or marked clinical improvement in 85% of manic patients. There have been only three prospective randomized controlled studies examining the efficacy of ECT in mania. Small and colleagues4 randomly allocated 34 hospitalized patients to receive either lithium carbonate or bilateral ECT, followed by maintenance lithium carbonate. Blinded clinical ratings showed greater improvement for patients treated with ECT compared with lithium carbonate during the first 8 weeks but no significant difference between the groups after 8 weeks. A subsequent trial randomly assigned patients who had not responded to either lithium or a neuroleptic to four groups: one pharmacotherapy group (lithium and haloperidol) and three different ECT groups (bilateral, right unilateral and left unilateral); the authors reported that 59% of the ECT patients achieved complete remission compared with none in the pharmacotherapy group.5 Sikdar and colleagues randomly assigned 30 manic patients to receive eight ECT sessions, either real or simulated, together with chlorpromazine pharmacotherapy.6 The group receiving real ECT improved significantly more than the group receiving chlorpromazine alone. These three prospective trials have all been small, involving only 91 patients in total. This probably reflects both the difficulty of recruiting to ECT trials in general and specifically the difficulty of obtaining informed consent from acutely manic patients. Combining the data from all the available literature, i.e. both retrospective and prospective studies, Mukherjee and colleagues found that 80% of all the patients from this very mixed group either achieved remission or had a marked clinical improvement.3 Current American Psychiatric Association (APA) guidelines suggest that, owing to the availability of lithium and anticonvulsant and antipsychotic agents, ECT in mania is mainly reserved for medication-resistant patients.7 However, it should be noted that to date there have been no comparison studies of ECT with the newer anti-manic agents. The APA guidelines also include some specific indications, such as recommending ECT for manic delirium, and also suggest that ECT may be effective for rapid cycling in bipolar disorder. However, the evidence base for both these indications is extremely limited. Recent guidance from the National Institute of Clinical Excellence (NICE) in the UK concluded that, although the evidence for ECT in mania is less robust than in depression, the available data are enough to recommend its use in mania where other treatments have proved ineffective.8
Ch 21
7/4/05
4:02 pm
Page 195
Electroconvulsive therapy and repetitive transcranial magnetic stimulation
195
Electroconvulsive therapy and bipolar depression As mentioned previously, there is firm evidence that ECT is an effective antidepressant treatment. However, is there any difference in outcome between bipolar depressed patients and unipolar depressed patients? The majority of studies carried out have been retrospective analyses of case series and the results have not been consistent across studies. For example, Black et al9 reviewed patient records over a 12-year period in a university hospital and found that unipolar and bipolar depressed patients received the same number of treatments (mean of nine) and that ECT had equivalent efficacy in both groups. In contrast, an earlier study suggested that ECT was more efficacious in the unipolar group of depressed patients compared with the bipolar group.10 A recent study from New York compiled data from three randomized controlled trials of ECT carried out during the 1990s.11 It included 162 patients who had unipolar depression and 66 with bipolar depression, all of whom had a high degree of treatment resistance. The patients were admitted to hospital, taken off medications and remained drug-free for at least a week. They all underwent a stimulus dosing protocol and were randomized to bilateral or right unilateral ECT, with a variety of seizure thresholds that were the primary focus of the original studies. Taking into account the different forms of ECT used, the authors found no difference in outcome between the unipolar depressed and bipolar depressed groups; approximately 60% from each group were initial responders to ECT. The one significant difference between the groups was that the bipolar depressed patients received fewer ECT treatments than the unipolar depressed patients, a mean difference of around 1.5 in the number of treatments received. Using data collected prospectively on 133 patients over a 7-year period, Grunhaus et al in Israel have recently reported a similar overall response rate (57%) to ECT with no difference between bipolar and unipolar depressed patients.12 Another important observation in this study, relating to a concern clinicians may have, was that emergent hypomania in bipolar depressed patients treated with ECT was a rare event, experienced by only about 3% of patients.
Findings from London Most of the published data about ECT and bipolar disorder originates from the USA. Some recent data from the South London and Maudsley NHS
Ch 21
7/4/05
4:02 pm
196
Page 196
Bipolar disorder: the upswing in research & treatment
Trust in South London provides a British perspective. ECT use at this mental health trust, as in the rest of the UK, reached its peak in the mid 1950s when nearly 35% of all patients admitted had ECT. Following the introduction of neuroleptics and tricyclic antidepressants, this proportion rapidly declined and over the past 10 years the use of ECT has reached a plateau with less than 1% of all patients admitted to hospital now receiving ECT. We have examined the case notes and ECT records of patients having ECT over a 3-year period from 1999 to 2001. Operationalized diagnoses were applied on the basis of symptoms recorded in the notes. Over this period, 150 courses of ECT were administered to 130 patients. Figure 21.1 summarizes the diagnoses of these patients. The vast majority of patients (77%) had a diagnosis of unipolar depression; 11% of patients were bipolar depressed while only two patients (1.5%) had mania during this 3-year period. We compared the unipolar and bipolar depressed patients on a number of factors (Table 21.1) There were no differences between the groups in terms of age and sex. They both had similar levels of treatment resistance and the percentage of patients per group having treatment without consent under the Mental Health Act 1983 was equal. However, there were some interesting differences. There was a trend towards the mean number of treatments (seven versus nine) being less in the bipolar depressed group. The mean length of hospital stay of the bipolar group was only about 75% that of the unipolar group and there was a significant difference in terms of the actual response to ECT at the end of treatment course; 85% of the bipolar group and 57% of the unipolar group had either a complete recovery or major improvement following ECT. These
Unipolar depression Bipolar – depressed Bipolar – manic Other
Figure 21.1 Electroconvulsive therapy according to diagnosis, 1999–2001 (n = 130).
Ch 21
7/4/05
4:02 pm
Page 197
Electroconvulsive therapy and repetitive transcranial magnetic stimulation
197
Table 21.1 Comparison of unipolar and bipolar depressed patients having electroconvulsive therapy Unipolar
Bipolar
t Test
Number of patients
88
13
Mean age (SD)
68 (13)
73 (9)
NS
Mean no. of antidepressants (SD)
2.2 (1.3)
2.7 (1.3)
NS
Mean no. of treatments (SD)
9 (6.1)
7 (3.6)
NS
Mean length of hospital stay (SD)
200 days
145 days
NS
(178)
(165)
% consenting to treatment
62%
69%
NS
% recovery/major improvement
57%
85%
p = 0.05
NS, not significant.
data are broadly in accordance with previous reports and provide further support for considering ECT for patients with a bipolar depressive presentation.
Transcranial magnetic stimulation – a potential alternative to electroconvulsive therapy? Transcranial magnetic stimulation (TMS) is a potential alternative to ECT for the treatment of depression.13 Like ECT it electrically stimulates the brain, but it does not induce a seizure or require anaesthesia. TMS uses Faraday’s principle of electromagnetic induction to induce secondary electrical activity within the brain. A brief powerful electrical current is passed through a copper wire coil placed in a hand-held device on the scalp. This electric current generates a magnetic field that passes unimpeded through the skull and induces secondary electrical currents in the immediately underlying cerebral cortex tissue. TMS can thus be used focally to stimulate selected cortical regions. Repetitive TMS (rTMS) involves the repeated administration of TMS to induce sustained effects such as change in mood. ‘Fast’ rTMS (> 1 Hz) has an activating effect, while ‘slow’ rTMS (< 1 Hz) has an inhibitory effect on the underlying cortex. Much of the therapeutic work using TMS for depression has been attempting to target and modify specific neuronal circuitry thought to be relevant to mood disorder. Thus, fast rTMS of the left dorsolateral
Ch 21
7/4/05
4:02 pm
198
Page 198
Bipolar disorder: the upswing in research & treatment
prefrontal cortex has been widely used in studies of depression. It has become increasingly clear that this can have an antidepressant effect but the extent, duration and how best to harness it remain to be resolved.
Repetitive transcranial magnetic stimulation in bipolar depression There is rapidly expanding literature about rTMS and unipolar depression, but relatively little is known about rTMS in bipolar depression.13 The largest study to date randomized 23 patients with bipolar depression to receive either real or placebo rTMS over the left prefrontal cortex on a daily basis for 2 weeks.14 The investigators did not find a significant benefit of real rTMS compared with placebo. However, the placebo group used a coil angled away from the scalp rather than a specially designed placebo coil. There is now good evidence from functional neuroimaging studies that the angling away of a real coil still induces stimulation in the brain area beneath the scalp and hence in this study the placebo coil may not have been a true placebo. The study found that rTMS was well tolerated and there were no significant adverse effects.
Repetitive transcranial magnetic stimulation in mania There are very few data available on the use of rTMS in mania. It is interesting to note that for rTMS, as for any other antidepressant treatment, there are some case reports of induced manic symptoms, both after individual treatment sessions and following longer courses of treatment.15,16 To date there are only two small randomized studies of rTMS in acute mania. The first, from Grisaru and colleagues in Israel, randomized 16 patients to left or right prefrontal fast rTMS for 2 weeks.17 They reported significantly more improvement in patients treated with right rather than with left prefrontal rTMS, suggesting laterality opposite to that seen in depression. The same group have performed a larger study in which 25 patients were randomized to either real or sham fast rTMS; this time all patients received stimulation over the right dorsolateral prefrontal cortex.18 There was no difference between real and placebo treatment groups. Therapeutic courses of rTMS involve the patient coming to the clinic every weekday for 2–4 weeks. For patients with mania who are likely to be unsettled, agitated and irritable, it seems likely that practical problems would arise in trying to deliver the treatment, as a high degree of cooperation is required during treatment sessions lasting about half an hour. In conclusion, rTMS may have an increasing role to play in the treatment of bipolar depression, but its use in mania is likely to remain limited.
Ch 21
7/4/05
4:02 pm
Page 199
Electroconvulsive therapy and repetitive transcranial magnetic stimulation
199
References 1.
2. 3.
4.
5. 6. 7. 8. 9. 10.
11. 12.
13. 14.
15.
16. 17.
18.
The UK ECT Review Group, Efficacy and safety of electroconvulsive therapy in depressive disorders: a systematic review and meta-analysis. Lancet 2003; 361:799–808. Pippard J, Ellam L, Electroconvulsive treatment in Great Britain. Br J Psychiatry 1981; 139:563–568. Mukherjee S, Sackeim HA, Schnur DB, Electroconvulsive therapy of acute manic episodes: a review of 50 years’ experience. Am J Psychiatry 1994; 151:169–176. Small JG, Klapper MH, Kellams JJ et al, Electroconvulsive treatment compared with lithium in the management of manic states. Arch Gen Psychiatry 1988; 45:727–732. Mukherjee S, Sackeim HA, Lee C, Unilateral ECT in the treatment of manic episodes. Convuls Ther 1988; 4:74–80. Sikdar S, Kulhara P, Avasthi A, Singh H, Combined chlorpromazine and electroconvulsive therapy in mania. Br J Psychiatry 1994; 164:806–810. Rasmussen K, The practice of electroconvulsive therapy: recommendations for treatment, training, and privileging (second edition). J ECT 2002; 18:58–59. Sanders RD, Deshpande AS, Mania complicating ECT. Br J Psychiatry 1990; 157:153–154. Black DW, Winokur G, Nasrallah A, ECT in unipolar and bipolar disorders: a naturalistic evaluation of 460 patients. Convuls Ther 1986; 2:231–237. Homan S, Lachenbruch PA, WinokurG, Clayton P, An efficacy study of electroconvulsive therapy and antidepressants in the treatment of primary depression. Psychol Med 1982; 12:615–624. Daly JJ, Prudic J, Devanand DP et al, ECT in bipolar and unipolar depression: differences in speed of response. Bipolar Disord 2001; 3:95–104. Grunhaus L, Schreiber S, Dolberg OT et al, Response to ECT in major depression: are there differences between unipolar and bipolar depression? Bipolar Disord 2002; 4(Suppl 1): 91–93. Lisanby SH, Kinnunen LH, Crupain MJ, Applications of TMS to therapy in psychiatry. J Clin Neurophysiol 2002; 19:344–360. Nahas Z, Kozel FA, Li X et al, Left prefrontal transcranial magnetic stimulation (TMS) treatment of depression in bipolar affective disorder: a pilot study of acute safety and efficacy. Bipolar Disord 2003; 5:40–47. Nedjat S, Folkerts HW, Induction of a reversible state of hypomania by rapidrate transcranial magnetic stimulation over the left prefrontal lobe. J ECT 1999; 15:166–168. Sakkas P, Mihalopoulou P, Mourtzouhou P et al, Induction of mania by rTMS: report of two cases. Eur Psychiatry 2003; 18:196–198. Grisaru N, Chudakov B, Yaroslavsky Y, Belmaker RH, Transcranial magnetic stimulation in mania: a controlled study. Am J Psychiatry 1998; 155:1608–1610. Kaptsan A, Yaroslavsky Y, Applebaum J et al, Right prefrontal TMS versus sham treatment of mania: a controlled study. Bipolar Disord 2003; 5:36–39.
Ch 21
7/4/05
4:02 pm
Page 200
Ch 22
7/4/05
4:02 pm
Page 201
chapter 22
Improving outcome by selecting effective long-term treatment Paul Grof
Introduction If we took the literature on bipolar disorders at face value, we would have to conclude that over time the outcome of long-term treatment has become much worse. While the earlier reports showed very promising results with 75% of bipolar patients benefiting from long-term lithium treatment,1 recent findings have been very discouraging, suggesting response rates between 20 and 35%. However, these numbers are not directly comparable, primarily because of major shifts in the patients under study. Bipolar disorders are now diagnosed much more broadly and more frequently, as a bipolar spectrum, with negative implications for outcomes. In this brief chapter I will review the observations indicating that the current unsatisfactory outcome can be markedly improved by recognizing the striking heterogeneity of bipolar disorders and by selecting the treatment for each individual according to a characteristic clinical profile.
Heterogeneity of bipolar disorders The recognition that the bipolar disorder diagnosis includes a heterogeneous collection of illnesses has been emerging for some time, both from the classical studies (e.g. reference 2) and more recent investigations (e.g. references 3–5). So far this concept has been neglected by mainstream psychiatry. Angst analysed the data from his long-term studies of bipolar illness and found three subtypes (Dm, MD and Md) that differed markedly not only in
Ch 22
7/4/05
4:02 pm
202
Page 202
Bipolar disorder: the upswing in research & treatment
the clinical course, but also in a number of other important clinical characteristics. Patients of the Dm type experienced major depressive episodes accompanied only by occasional hypomanias; these patients were predominantly women (over 80%) and their age of onset was in the late thirties on average, significantly later than the other two groups. The Md group experienced primarily or exclusively hospitalized manic episodes and in comparison with the other two groups had significantly less chronicity, mortality and suicidal behaviour. In large multicentre studies Bellivier et al4 identified and replicated three subtypes of bipolar I patients, based on the age of onset of bipolar illness. Benazzi6 then reported three similar subgroups from a large series of bipolar II patients. In a comprehensive review Alda5 also found support for the existence of three main types of bipolar disorder that differed with respect to clinical presentation, course of illness, family history and possibly longterm treatment response. Together these examples of heterogeneity raise a question: can we improve the outcome of stabilizing treatments by respecting the types of bipolar disorder? This important question could best be resolved in a specifically designed, long-term, cross-over clinical trial, with proven stabilizing treatments, and focusing on the characteristics of bipolar patients. However, given the ethical and feasibility dilemmas involved, such a trial may never happen. In the meantime we need to treat bipolar patients effectively. In addition, evidence-based medicine should not be limited to drug trials; it has to utilize all valid information.7 Much practically useful information about treatment responsiveness of bipolar patients can be gained from patient-oriented studies, from some drug studies and from extensive clinical observations.
Heterogeneity of responders Here I briefly review data extracted from a series of unequivocal responders to three main types of long-term treatment for bipolar disorders – lithium, lamotrigine and olanzapine (as a representative of atypical neuroleptics). All three have been shown in controlled, double-blind trials to be effective in groups of patients with bipolar disorders.8–10 The body of emerging data appears to show that unequivocal responders to long-term monotherapies, such as lithium, lamotrigine and atypical neuroleptics have distinct clinical profiles. The differences include clinical presentation and course of illness,
Ch 22
7/4/05
4:02 pm
Page 203
Improving outcome by selecting effective long-term treatment
203
co-morbidity and in particular family history, thus suggesting that these are clinically relevant subtypes of bipolar disorders. While the characteristics of unequivocal lithium responders have been known for some time,11 the likely features of patients who benefit from lamotrigine and olanzapine have emerged mainly from two studies, one performed in Ottawa, and the other in Halifax. In Ottawa a consecutive series of patients was studied who were diagnosed as suffering from bipolar disorder according to DSM-IV criteria, required long-term prophylaxis with medication, and were treated in our programme for 3 or more years. To assess the characteristics of unequivocal responders to prototypic longterm treatment, the diagnosis and co-morbidity of each patient was assessed with the Schedule for Affective Disorders and Schizophrenia – Lifetime Version (SADS-L) interview and family history with Schedule for Affective Disorders and Schizophrenia – Family History (SADS-FH) completed with the patient. In addition, available and consenting first-degree relatives were blindly interviewed using the SADS-L. To be classified as a good responder, each patient had to receive a score on the Alda scale12 of 7 or better; 112 patients met all criteria and had 756 evaluated relatives. With regard to family history, loading was strikingly different for each group of responders. When the responders to long-term treatment with three different medications were compared, only the relatives of lithium responders had a significant excess of bipolar disorders. While the firstdegree relatives of bipolar patients responding to lamotrigine had an excess of anxiety, panic, substance abuse and alcohol addictions, the relatives of those benefiting from olanzapine had no excessive bipolar or anxiety disorders but a high rate of chronic psychotic illnesses. In parallel with the differences in family history between responder groups, there were corresponding dissimilarities in co-morbid disorders. Like their relatives, the lamotrigine responders also had more problems with substance and alcohol addictions as well as anxiety disorders, while a history of mood-incongruent psychotic symptoms was more prevalent among the olanzapine responders. There were also differences in the pre-treatment clinical course in that lithium responders presented with an episodic, fully remitting course and often had a predominance of depressive over manic episodes. On the other hand, lamotrigine and olanzapine responders tended to have mostly a nonepisodic course with significant residual symptoms and exacerbations, with predominance of manic episodes. The observations on lithium and lamotrigine responders in Ottawa were in good agreement with the findings from Halifax.13
Ch 22
7/4/05
4:02 pm
204
Page 204
Bipolar disorder: the upswing in research & treatment
The approaches in the two centres had several aspects in common: for example, both teams evaluated the outcome of long-term treatment with the same Alda scale that takes into consideration the risk of recurrences and various aspects of treatment such as length and compliance; also, both teams blindly evaluated the first-degree relatives of the responders, for the purpose of genetic studies.
Clinical characteristics of responders From these and earlier studies it is possible to derive the main clinical characteristics for bipolar patients responding to each of the major stabilizing treatments: lithium, olanzapine and lamotrigine. These characteristics can be useful in clinical practice in order to identify the treatment of choice for an individual bipolar patient.
Lithium Responders to lithium stabilization present with depressive and manic episodes of the classical type, without mood-incongruent symptoms, clearly sad depressions and often euphoric manias. In their family history, they tend to have bipolar disorders with an episodic course. They, themselves, have an episodic full-remitting course and, if the course has been extensive, one can usually see a predominance of depressions. Finally, these patients have relatively rare co-morbid conditions (Table 22.1).
Lamotrigine The characteristics of responders to lamotrigine prophylaxis are different. In the presentation they often have atypical features: the depression is described as emotional emptiness and apathy, indifference, slow motivation
Table 22.1 Characteristics of responder to long-term lithium treatment Clinical course: episodic, fully remitting, predominance of depressions Family history: bipolar disorders, with episodic course Co-morbidity: relatively rare, as in the general population Presentation: classical, as described in earlier textbooks (e.g. depressions with sadness, manias with euphoria, absence of mood-incongruent psychotic symptoms)
Ch 22
7/4/05
4:02 pm
Page 205
Improving outcome by selecting effective long-term treatment
205
and hypomanias as activations without euphoria. These patients often have anxiety and panic disorders or substance use disorders in their family history; the course of illness is non-episodic and often entails residual symptoms. Similarly, these patients have substantial co-morbidity similar to their family history (Table 22.2).
Olanzapine Olanzapine responders again have atypical features characterizing both their depressions and their manias, and one can often identify moodincongruent psychotic symptoms in their history or in their acute presentation. Family history, if positive, tends to show psychotic disorders or chronic psychiatric disorders. The clinical course has residual symptoms between the episodes of depressions and manias, and the history, if fully developed, shows more manias than depressions. Co-morbidity is frequent, particularly with alcoholism and substance abuse. We have preliminary evidence to suggest that the clinical features of responders to olanzapine may generalize to indicate response to other atypical neuroleptics (Table 22.3).
Table 22.2 Characteristics of responder to long-term lamotrigine treatment Clinical course: non-episodic, with residual symptoms, mostly depressions (often ‘bipolar II’ type) Family history: anxiety and panic disorders, substance and alcohol addictions Co-morbidity: high, anxiety and panic disorders, substance and alcohol addictions Presentation: atypical, non-textbook features (e.g. depressions characterized by anergia or emotional emptiness, hypomanias by general speeding without euphoria)
Table 22.3 Characteristics of responder to long-term olanzapine treatment Clinical course: non-episodic, with residual symptoms, overactive episodes often more frequent than depression Family history: non-remitting or psychotic disorders Co-morbidity: alcohol abuse or addictions Presentation: atypical, non-textbook features, often history or presence of moodincongruent psychotic symptoms
Ch 22
7/4/05
4:02 pm
206
Page 206
Bipolar disorder: the upswing in research & treatment
Divalproex The data on divalproex are missing from these investigations, mainly because the evidence from controlled clinical trials for long-term efficacy of divalproex is not available.14
Selectivity of responses There has been a line of investigation documenting that the response to some stabilizers has been associated with specific clinical correlates, for example for lithium,11,15 carbamazepine,16 valproate17 and lamotrigine.13 In addition, however, there is a growing body of literature indicating that good responses to each of these substances are selective and often mutually exclusive. For example, excellent lithium responders failed on long-term carbamazepine and vice versa.18 Post et al19 made a similar observation: most patients with a good acute response to carbamazepine had a clear history of non-response to lithium. Bowden et al20 found that previous lithium responders did well on lithium but not on divalproex. Similarly, Swann et al21 noted that responders to valproate had evidence of prior non-response to lithium. Tohen et al22 observed that olanzapine succeeded in patients who had failed previously on lithium and divalproex. Despite some methodological limitations of these observations, together they provide a credible picture of a degree of selectivity among these medications. These data do not support the often reported clinical impression that many bipolar patients require combination treatment in order to get well. The issue of selectivity is also indirectly supported by the clustering of prophylactic responses in families. For lithium there is evidence that response to long-term treatment markedly clusters in families.12 Evidence is also slowly mounting that affected children of bipolar parents tend to benefit markedly from the long-term treatment to which the parent responded (A. Duffy et al, submitted for publication).
Discussion The critical task in the long-term management of bipolar disorders is matching the patient and the effective treatment. Much of the current literature stresses that bipolar patients should be treated by a combination of
Ch 22
7/4/05
4:02 pm
Page 207
Improving outcome by selecting effective long-term treatment
207
medications and that the right combination can be established by following an algorithmic sequence, adding one drug at a time. This recommendation is based on practical experience that many bipolar patients fail on the initial medication, and respond only after several more drugs are added. However, such anecdotal observations have more than one possible explanation. At present, there is much misunderstanding about long-term treatment of bipolar disorders. In particular, the natural course of atypical forms has not been well described and there are major misinterpretations about indications for effective lithium treatment. As a result, the interpretation of the outcome of treatment is often incorrect. Clinicians use combinations because they experience treatment failures during the initial stage of treatment of a bipolar patient. Very frequent initial failures should be expected. In DSM-IV-diagnosed bipolar disorders, long-term monotherapy chosen by the current practice of trial-and-error is effective in one-third of patients at best,23 but can be markedly improved by treating according to clinical profile of the patient and family as described above. Observations supporting this are particularly convincing for lithium and clozapine. The majority of bipolar patients who have been correctly selected for lithium treatment and adequately monitored can be completely stabilized with lithium monotherapy. On the other hand, some patients benefiting substantially from an atypical neuroleptic or lamotrigine will intermittently or sometimes chronically require an addition of an antidepressant or another psychotropic drug. A combination of medications appears to be indicated, particularly in bipolar patients who are treatment-resistant to monotherapy, do not tolerate it well, or do not have clinical characteristics helpful for a clear treatment choice. Evidence is lacking that combinations of several medications are necessary in the majority of bipolar patients, despite the current practice. It is difficult to justify exposing patients to the side-effects of several drugs if mood stabilization can be achieved without a multiple combination. Heterogeneity of bipolar disorders poses a major problem for the interpretation of the results of clinical trials. The results of any long-term drug trial may depend as much on the composition of the patient sample and the proportion of the subtypes as on the efficacy of the tested drug. Currently we are developing a computer program that forecasts the treatment outcome based on the individual bipolar patient’s clinical characteristics. This program should be ready for predictive testing soon.
Ch 22
7/4/05
4:02 pm
208
Page 208
Bipolar disorder: the upswing in research & treatment
Conclusions There are at least three distinct types of bipolar disorder that markedly differ both in clinical characteristics and in treatment outcome. Comprehensive clinical assessment is needed in order to identify the type; the patient can then be matched with a more effective long-term treatment. Despite prevailing practice, evidence is lacking so far that polypharmacy with multiple stabilizers is necessary for the majority of bipolar patients. However, for patients who can achieve stabilization by one primary medication, whether it is lithium, lamotrigine or atypical neuroleptics, it is difficult to justify exposing them to the side-effects of complex polypharmacy.
References 1.
2. 3. 4.
5. 6. 7. 8. 9. 10. 11.
12. 13.
Schou M, Thomsen K, Lithium prophylaxis of recurrent endogenous affective disorders. In: Johnson FN (ed), Lithium Research and Therapy. Academic Press: New York, 1975:63–84. Angst J. The course of affective disorders. II. Typology of bipolar manic– depressive illness. Arch Psychiatrie Nervenkrankheiten 1978; 226:65–73. Angst J, Gamma A, Benazzi F et al, Diagnostic issues in bipolar disorder. Eur Neuropsychopharmacol 2004; 13(Suppl 2):S43–S50. Bellivier F, Golmard J, Rietschel M et al, Age at onset in bipolar I affective disorder: further evidence for three subgroups. Am J Psychiatry 2003; 160:999–1001. Alda M, The phenotypic spectra of bipolar disorder. Eur Neuropsychopharmacol 2004; 14:S94–S99. Benazzi F, Toward better probing for hypomania of bipolar II disorder. Int J Methods Psychiatr Res 2004; 13:1–9. Jenicek M, Clinical Case Reporting in Evidence-based Medicine, 2nd edn. Oxford UniversityPress: New York, 2001. Schou M, Perspectives on lithium treatment of bipolar disorder: Action, efficacy, effect on suicidal behaviour. Bipolar Disord 1999; 1:5–10. Calabrese JR, Shelton MD, Rapport DJ et al, Long-term treatment of bipolar disorder with lamotrigine. J Clin Psychiatry 2002; 63(Suppl 10):18–22. Tohen M, Chengappa KNR, Suppes T et al, Olanzapine cotherapy in prevention of recurrence in bipolar disorder. Eur Psychiatry 2002; 17(Suppl 1):109. Grof P, Alda M, Grof E et al, The challenge of predicting response to stabilizing lithium treatment: The importance of patient selection. Br J Psychiatry 1993; 163(Suppl 21):16–19. Grof P, Duffy A, Cavazzoni P et al, Is response to prophylactic lithium a familial trait? J Clin Psychiatry 2002; 63:942–947. Passmore M, Garnham J, Duffy A et al, Phenotypic spectra of biopolar disorder in responders to lithium versus lamotrigine. Bipolar Disord 2003; 5:110–114.
Ch 22
7/4/05
4:02 pm
Page 209
Improving outcome by selecting effective long-term treatment 14.
15. 16.
17. 18. 19. 20. 21.
22.
23.
209
Bowden CL, Calabrese JR, McElroy SL et al, A randomized, placebocontrolled, 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Arch Gen Psychiatry 2000; 57:481–489. Grof P, Hux M, Grof E, Arato M, Prediction of response to stabilizing lithium treatment. Pharmacopsychiatria 1983; 16:195–200. Greil W, Kleindienst N, Erazo N, Muller-Oerlinghausen B, Differential response to lithium and carbamazepine in the prophylaxis of bipolar disorder. J Clin Psychopharmacol 1998; 18:455–460. Calabrese JR, Shelton MD, Bipolar diorders and the effectiveness of novel anticonvulsants. J Clin Psychiatry 2002; 63(Suppl 3):5–9. Grof P, Lithium update: selected issues. In: Ayd F, Taylor JT, Taylor BT (eds), Affective Disorders Reassessed. Ayd Medical Publications: Baltimore, 1983. Post RM, Denicoff KD, Frye MA, Everich GS, Re-evaluating carbamazepine prophylaxis in bipolar disorder. Br J Psychiatry 1997; 170:202–204. Bowden CL, Brugger AM, Swann AC et al, Efficacy of divalproex versus lithium and placebo in the treatment of mania. JAMA 1994; 271:918–924. Swann AC, Bowden CL, Calabrese JR et al, Mania: differential effects of previous depressive and manic episodes on response to treatment. Acta Psychiatr Scand 2000; 101:444–451. Tohen M, Chengappa KNR, Suppes T et al, Efficacy of olanzapine in combination with valproate or lithium in the treatment of mania in patients partially nonresponsive to valproate or lithium monotherapy. Arch Gen Psychiatry 2002; 59:62–69. Garnham J, Munro A, Teehan A et al, Bipolar disorder: assessing treatment response in a naturalistic setting. J Bipolar Disord 2001; 3(Suppl 1):37.
Ch 22
7/4/05
4:02 pm
Page 210
Ch 23
7/4/05
4:03 pm
Page 211
chapter 23
Is what we offer to patients half acceptable? Rachel Perkins
The catalogue of deficits and dysfunction described in this volume might suggest that someone with a diagnosis of bipolar disorder – like myself – has little constructive to contribute. It would be easy to dismiss my analysis as nothing other than a manifestation of my psychopathology. However, before doing this, it is perhaps worth remembering that I am also a senior provider of mental health services – a consultant clinical psychologist and clinical director – who has written over 150 papers and four books. One might ask whether this is possible in the presence of the cognitive deficits and dysfunctions that have been described. The primary concern of people with bipolar disorder is to be able to live the lives they wish to live and do the things they want to do.1 While people may desire freedom from debilitating symptoms this is only part of the story – retaining or rebuilding a decent life is at least (if not more) important. People want: ‘…safe, pleasant and affordable housing, well paying and fulfilling jobs … to be treated with dignity and respect, to have control over their lives and to have genuine choices. They want to feel good about themselves and to have the opportunity to achieve things that all of us do.2 Having a job is central to the physical, psychological and social well-being of most people;3 therefore, the analysis presented here will focus on employment. However, it must be emphasized that many of the arguments are equally applicable to other important life domains. Typically, the guiding philosophy – organising principle – of mental health services remains one of ‘symptom’ and ‘cure’: the identification of difficulties and dysfunctions and the reduction of these by pharmacological,
Ch 23
7/4/05
4:03 pm
212
Page 212
Bipolar disorder: the upswing in research & treatment
psychological or other means. Within this framework, four implicit assumptions can be discerned. First, it is assumed that the primary task of the assessment process is to catalogue a person’s symptoms and problems. Second, it is assumed that a person’s symptoms must be controlled before they can be helped to resume their life: people must be ‘well’ before they can go back to work and other social roles. Third, it is assumed that until a person’s symptoms can be reduced they must be looked after and protected from harm. Fourth, it is assumed that if a person’s symptoms can be eliminated then they will automatically be able to resume their ‘normal’ lives. In this chapter I will argue that such assumptions are unfounded and that a framework based on symptoms and cures provides an inadequate basis for the development of services that can enable people with bipolar disorder to pursue their ambitions and rebuild decent, satisfying and valued lives. Lives cannot be built on a foundation of deficits and dysfunctions – they are constructed on a base of talents and possibilities. As Chadwick has argued: ‘Deficit-obsessed research can only produce theories and attitudes which are disrespectful of clients and are also likely to induce behaviour in clinicians such that service users are not properly listened to, not believed, not fairly assessed, are likely treated as inadequate and are also not expected to become independent and competent individuals in managing life’s tasks.’4 It is equally, if not more, important that an assessment process identifies the strengths that are the building blocks for recovery. A focus on symptoms and problems saps confidence, causes people to lose sight of their potential and possibilities and can readily result in the hopelessness and despair that are associated with increased suicide risk.5,6 Mental health services are replete with people who have given up on themselves and their futures: a tragic waste of human potential. A focus on deficits and dysfunctions too readily results in clinicians having low expectations about what people with bipolar disorder can achieve. Such low expectations generate a vicious cycle – a self-fulfilling prophecy – that both erodes hope and diminishes opportunity (Figure 23.1). If the expert professionals believe that people with bipolar disorder are, for example, unlikely to be able to hold down a job (or at best can only manage lowlevel, low-stress positions) then this has two effects. First, people with the disorder believe them and give up applying for jobs – ‘If the experts say I cannot work then what hope is there?’ Second, employers believe them and
Ch 23
7/4/05
4:03 pm
Page 213
Is what we offer to patients half acceptable?
213
Expert professionals say that people with bipolar disorder are unlikely to be able to work
Employers believe that people with bipolar disorder cannot work – so don't employ them
People with bipolar disorder believe that they cannot work and give up trying to get jobs
Very few people with bipolar disorder are in employment
Figure 23.1 The vicious cycle of low professional expectations: a self-fulfilling prophecy.
are reluctant to hire people with the disorder – ‘If the experts say they cannot work then what is the point in employing them?’ If people with bipolar disorder see little point in applying for jobs, and employers are reluctant to hire them, then this guarantees that there will be few people with the diagnosis in employment. This, in turn, confirms the low expectations of employers, clients and mental health professionals in a vicious cycle of despondency. All can say ‘I told you so’: interpret low employment rates as confirmation of the low probability of people with the disorder being able to work. At a general level, the quest for a ‘bipolar credit’ – akin to ‘schizophrenic credit’4,7 – is as important as the search for deficits. The relationship between creativity and bipolar disorder might furnish a promising basis for such endeavours.8,9 At a more specific level, assessments must reveal an individual’s strengths, possibilities and ambitions if they are to provide a useful foundation for helping them to rebuild their lives. The assumption that a person’s symptoms must be controlled – their problems minimized – is equally problematic. The reduction of symptoms and problems is neither a necessary nor a sufficient condition for enabling people to work or resume other social roles. On the one hand, a person can gain and sustain employment even in the face of continuing, or recurring, symptoms. On the other hand, even if a person’s symptoms can be eliminated completely, this does not mean that they will be able to gain work. Symptom reduction does little to control the prejudice and discrimination that pervade society and operate to exclude
Ch 23
7/4/05
4:03 pm
214
Page 214
Bipolar disorder: the upswing in research & treatment
people from employment (and other social roles) on the basis of a history of mental disorder as well as its continued presence. However, randomized controlled trials have demonstrated that as many as 70% of those with serious mental health problems can successfully gain and sustain employment if offered the right kind of support: ‘Individual Placement with Support’; evidence-based supported employment.10,11 It is the type of support rather than diagnosis or symptoms that are the primary determinant of employment outcomes. Research has failed to find a relationship between any specific client factors – including diagnosis, symptomatology, disability status or prior hospitalization – and the outcomes of supported employment.11 Furthermore, in the time that it takes for a person’s symptoms to be fully controlled, it is likely that people will have lost the things that they value in life: friends, partners, leisure activities and, most especially, work. The longer a person is off work for illness reasons, the less likelihood there is of them returning to work.12 Time is of the essence: after 6 months of sickness absence, the probability that a person will return to work falls to 50%; after 1 year to 25% and after 2 years to 10%.13 Although there may be brief periods of acute crisis when it may be appropriate to relieve people of their social roles, these should be kept to an absolute minimum. The assumption that people cannot work (or engage in other roles) unless or until their symptoms have been controlled is likely to reduce their chances of retaining valued social roles and/or rebuilding their lives. It is important that services actively assist people to do the things they want to do at the earliest opportunity, whether or not their symptoms persist or recur. A desire on the part of clinicians to protect those who have continuing/recurring symptoms and deficits may, in some instances, be appropriate. It is perhaps understandable that some people with bipolar disorder may be reluctant to take risks for fear of precipitating a relapse. However, building a life – pursuing ambitions – necessarily involves taking risks: noone can ever gain qualifications, get a job, or form a relationship without experiencing anxiety and risking failure. A life in which stress is minimized in order to reduce the risk of relapse is not necessarily a satisfying one: each individual needs to weigh the relative costs and benefits for themselves. Clinicians must be prepared actively to encourage people to explore their possibilities and pursue valued ambitions, and to support them in taking the associated risks, rather than always counselling caution. If the services are to enable people with bipolar disorder to rebuild meaningful, satisfying and valued lives then it is necessary to move away from an emphasis on the identification and amelioration of symptoms and
Ch 23
7/4/05
4:03 pm
Page 215
Is what we offer to patients half acceptable?
215
deficits. While the importance of treating distressing and disabling symptoms should not be minimized, such endeavours do not offer an adequate organizing principle for mental health services or the development of their underpinning research base. In place of symptom identification and cure, a more appropriate guiding principle for services might better be framed in terms of recovering a meaningful, satisfying and valued life: enabling people to do the things they want to do, live the lives they wish to lead and access those opportunities that non-disabled citizens take for granted. The treatment of symptoms, the reduction of deficits, may form part, but only part, of such an enterprise. Perhaps, after Shepherd,14 it would be preferable to view the treatment of symptoms as part of the assessment process: helping to define those ongoing impairments that need to be accommodated if a person is to do the things they wish to do in life. Such an approach requires a move away from a focus on deficit and dysfunction towards an emphasis on skills and possibilities; a move away from a focus on care towards an emphasis on opportunity; a move away from prescribing what is good for people (what they should do, what help and support they need) to enabling people to take control of their own lives and the assistance they receive to live them. It also requires a change in the balance of research endeavours. The identification of deficits and dysfunctions and optimal ways of alleviating these can no longer take pride of place on the research agenda. Instead, there needs to be an increased focus on the best ways of enhancing social role functioning and optimal ways of minimizing the impact of the disorder on the person’s life: this may be achieved by decreasing symptoms, it may not – that is an empirical question that has not, to date, been adequately addressed. Attention is required to issues such as the ways in which the hope and optimism necessary for rebuilding a decent life can be fostered and the ways in which the social (and financial) chaos wrought by episodes of mania or depression can be minimized or mitigated. Under both US and UK disability rights legislation, employers, educators and the providers of goods and services are required to make ‘reasonable adjustments/accommodations’ to facilitate access for people with mental health problems such as bipolar disorder. Yet very little research has been conducted into the sorts of adjustments that might be most beneficial to those with the disorder. While there is a mounting body of literature concerning the types of services that can best assist people with more serious mental health to retain, regain and sustain employment, none of this has been specific to bipolar disorder and for other areas of people’s lives the research evidence is virtually non-existent.
Ch 23
7/4/05
4:03 pm
216
Page 216
Bipolar disorder: the upswing in research & treatment
When one strays outside the domain of symptoms a vast expanse of uncharted territory comes into view. Dangerous chasms in knowledge and research become evident and these take on a particular significance in relation to the development and refinement of treatment standards and practice guidelines (such as those being developed by the UK National Institute for Clinical Excellence). These can only be based on the best research evidence available. Those things that have attracted the attention of clinical academics and researchers will therefore be included; those areas where research interest has been lacking will be excluded. If outcome research has primarily addressed the efficacy of different psychological and pharmacological interventions then practice guidelines can focus only on these areas. Those things that are equally, if not more, important for people with bipolar disorder – work, friendship, intimate relationships, parenting, social and leisure activities – will be excluded simply because they have not appeared on research agendas. While attention remains focused on deficits and dysfunctions, views of those with bipolar disorder will remain negative and pessimistic. While intervention is considered solely in terms of symptom reduction, then arenas critical to well-being and quality of life will remain ignored. If what we offer to people with bipolar disorder is to be half acceptable – if the lives of people with the disorder are really to be improved – then it is imperative that we broaden the focus of mental health services and the research base that underpins them.
References 1. 2.
3. 4. 5. 6.
7.
Repper J, Perkins R, Social Inclusion and Recovery. A Model for Mental Health Practice. Ballière Tindall: London, 2003. Bond GR, Psychiatric rehabilitation outcome. In: The Publication Committee of International Association of Psychosocial Rehabilitation Services (IAPSRS), (eds), An Introduction to Psychiatric Rehabilitation. International Association of Psychosocial Rehabilitation Services: Columbia, MD, 1994: 490–494. Royal College of Psychiatrists, Employment Opportunities and Psychiatric Disability. Council Report CR111. Royal College of Psychiatrists: London, 2002. Chadwick PK, Schizophrenia: The Positive Perspective. In Search of Dignity for Schizophrenic People. Routledge: London, 1997. Drake RE, Cotton PG, Depression, hopelessness and suicide in chronic schizophrenia. Br J Psychiatry 1986; 148:554–559. Beck AT, Brown G, Berchick RJ et al, Relationship between hopelessness and ultimate suicide: a replication with psychiatric outpatients. Am J Psychiatry 1990; 147:190–195. Claridge GS, Origins of Mental Illness. Blackwell: Oxford, 1985.
Ch 23
7/4/05
4:03 pm
Page 217
Is what we offer to patients half acceptable? 8. 9. 10. 11. 12.
13. 14.
217
Jamison KR, Touched with Fire: Manic–depressive Illness and the Artistic Temperament. The Free Press: New York, 1993. Jamison KR, Manic depressive illness and creativity. Sci Am 1995; 272:62–67. Crowther RE, Marshall M, Bond GR, Huxley P, Helping people with severe mental illness to obtain work: systematic review. BMJ 2001; 322:204–208. Bond GR, Supported employment: evidence for an evidence-based practice. Psychiatr Rehabil J 2004; 27:345–359. Niemeyer L, Jacobs K, Reynolds-Lynch K et al, Work hardening, past, present and future – The work programs special interest section national work hardening outcome study. Am J Occup Ther 1994; 48:327–339. Clinical Standards Advisory Group, Back Pain. HMSO: London, 1994. Shepherd G, Institutional Care and Rehabilitation. Longman: London, 1984.
Ch 23
7/4/05
4:03 pm
Page 218
Index
7/4/05
4:03 pm
Page 219
Index adolescents, changes in amygdala of 23 impaired recognition of facial expression in 38 MRS studies of bipolar disorder in 22 adrenocorticotrophic hormone (ACTH), role in hypothalamic–pituitary–adrenal axis 116, 129, 130, 135 affective instability, neural activity and 43 neural basis of 44 affective spectrum 147 age, as predictor of recovery 16 at onset 6 differences in incidence with 1 ECT use and 196 gender and age at onset 1, 2, 4 grey matter volume and 23 hippocampal volume reduction with increasing 126 hyperintense lesions and aging 22 alcoholism, as predictor of functional outcome 19 occupational status after discharge and 12 remission length and 13 risk of relapse and 11 amygdala, abnormalities in 22 changes in adolescents and children 23 heterogeneity within 33 increased activity in 51 neural responses to facial expression in 38 neuroanatomical changes in 51 structural changes in 41 anger recognition 38 animal studies, compromised BDNF expression in transgenic mice 107 compromised CREB expression in transgenic mice 107 compromised TrkB expression in transgenic mice 108 cortisol in guinea pigs 131
Index
7/4/05
4:03 pm
220
Page 220
Bipolar disorder: the upswing in research & treatment
transgenic mouse models 103–10 anterior cingulate, abnormalities in 21 activity and facial expression recognition 38 attentional processing in 149 increased activity in 51 neuroanatomical changes in 51 neuron size in 124 neuropathology in 60, 61 synaptic pathology in 61 anticonvulsants, use and functional outcome 14 antidepressants, cognitive function and 42 effect on CREB–BDNF–TrkB pathway 104, 105 effect on neurotransmitters 103 use and functional outcome 14 antipsychotics, advantages and disadvantages of 181–90 basal ganglia enlargement and 32 cf lithium in mania 178 cf placebo 183 cf valproate 181–90 classical cf atypical 188 combination treatment with lithium 186 with valproate 186 comparative trials 188 extrapyramidal symptoms 189 grey matter volume changes and 32 prescription at discharge 18 rapid response to 182 use and functional outcome 14 anxiety, co-morbidity with bipolar disorder 166 arginine vasopressin, effects of 135–6 in major depressive disorder 138, 139 levels in depression 139 neuroanatomy 135 response to chronic stress 138 role in hypothalamic–pituitary–adrenal axis 135 role in hypothalamic–pituitary–adrenal axis dysregulation 140 synergism with corticotrophin releasing hormone 137 aripiprazole cf placebo 183 attentional processes, dysfunction in 149 in bipolar disorder 157 autonomic response, lack in major depressive disorder 40 autopsy results 59–66
Index
7/4/05
4:03 pm
Page 221
Index basal ganglia, enlargement and antipsychotic use 32 BDNF gene 84, 85 see also brain-derived neurotrophic factor; neurotransmitters Beck’s cognitive model of depression 145, 146 biased medicine 178 bipolar cycling 147 bipolar disorder, advantages and disadvantages of antipsychotics or valproate in 181–90 age at onset 6 brain abnormalities in 21–4 brain volume changes in 27, 30, 31 care costs 166 cf schizophrenia 4 changes in synaptic protein levels in 124, 125 clinical epidemiology 1–7 clinical epidemiology study in Camberwell 1–7 co-morbidities 166 cognitive dysfunction and cause or consequence 145–54 current treatment guidelines 170 ECT use in 193–8 functional outcome 9–20 gender differences in age at onset 6 gender differences in presentation 3 genetic basis to brain abnormalities in 93–101 genetics of 64, 69–75, 77–88 glia-based origin of 65 glial cell changes in 124 heterogeneity of 201 hypothalamic–pituitary–adrenal axis and 115–21 dysfunction in 118 imaging studies of 27–34, 37–44 incidence rate 3 living with 211 longitudinal course 9 mifepristone use in 120 morbidity 165 mortality 165 neural basis for cognitive function in 157–61 neuropathology 59–66 and cortisol dysregulation in 123 pathophysiology of 51–6, 115 peak age at onset 3 postmortem findings 59–66 psychological treatments 165–71 repetitive transcranial magnetic stimulation use in 193–8 services for patients with 215 subcortical regions in 37–44 transgenic mouse models 103–10
221
Index
7/4/05
4:03 pm
222
Page 222
Bipolar disorder: the upswing in research & treatment
use of lithium in 175–9 vicious cycle of low expectations 213 brain abnormalities, activity 37–41 amygdala 22 anatomical specificity and precision in studies of 63 anterior cingulate grey matter reduction 21 as endophenotypes 94 attentional processing and 149 changes in volume 29 by brain region 30 cingulate gyrus 21 expansion of subcortical region 37–44 frontolimbic changes 23 functional imaging studies of 158–61 genetic basis to 93–101 genetic liability and grey matter volume 97, 98 genetic liability and white matter volume 97, 99 grey matter reduction in children and adolescents with bipolar disorder 22, 23 heterogeneity within brain structures 33 hyperintense lesions 22 in bipolar disorder 21–4 meta-analysis of 27–34 neural function as predictor 43 neuropathology 59–66 previous illness history and 42 response to medications 42 stress and 123–7 structural 41 brain morphometry 94 brain-derived neurotrophic factor (BDNF) 83, 84 activity 84 antidepressant effects of 106, 110 map of gene 85 response to antidepressants 105 response to immobilization stress 109 role in depression 103, 104 role in other disorders 86 transgenic mouse models with compromised expression of 107 Brown–Peterson paradigm 157 Camberwell, clinical epidemiology study in 1–7 Cambridge Neuropsychological Test Automated Battery 146, 151 cAMP response element-binding protein (CREB) response to antidepressants 105 see also CREB–BDNF–TrkB pathway transgenic mouse models with compromised expression of 107 carbamazepine, in lithium responders 206
Index
7/4/05
4:03 pm
Page 223
Index children, changes in amygdala of 23 MRS studies of bipolar disorder in 22 chlorpromazine, cf ECT 194 chromosomal regions involved in bipolar disorder 81 chromosome 13q 74, 81 chromosome 22q 74, 81 cingulate gyrus, abnormalities in 21 citalopram and cognitive function 42 clinical epidemiology 1–7 cognitive behaviour therapy 167, 168 MRC study 170 cognitive dysfunction, see under cognitive function cognitive function, age and increased deficit 146, 147 assessment of deficits in bipolar disorder 148 cause or consequence of bipolar disorder 145–54 development in childhood 145 effect of cortisol on 119 effect of glucocorticoids on 118 effect of medication on 42 effect of mifepristone on 120 focus on deficit and dysfunction is negative 212 hypercortisolaemia and 121 impaired decision-making in mania 152 impairment in affective disorders 118 in Cushing’s disease 119 in recovered bipolar disorder patients 153 linked with functional imaging 153 measures of 151 measures of deficit 146 neural basis for 157–61 neurocognitive tasks as measure of dysfunction 147 residual deficits 153 combination therapies 186 use for improved outcome 207 comorbidity, at first episode presentation 17 with bipolar disorder 166 complexin-I level in anterior cingulate cortex 124 COMT gene 64, 83 corticotrophin releasing hormone, description 135 regulation by glucocorticoids 137 response to chronic stress 138 role in hypothalamic–pituitary–adrenal axis 116, 129, 130 synergism with arginine vasopressin 137
223
Index
7/4/05
4:03 pm
224
Page 224
Bipolar disorder: the upswing in research & treatment
cortisol, bad press for 129–32 effect on brain 129 effect on frontal lobes 119 effects at cellular level 117 regulation of levels 131 costs, annual care costs for bipolar disorder 166 CREB, see under cAMP response element-binding protein CREB–BDNF–TrkB pathway, description 104 response to antidepressants 105 response to stress 104 role in antidepressive therapy 110 role in depression 109, 110 computed tomography (CT) of bipolar disorder 27 Cushing’s disease, cognitive function in 119 hippocampal volume reduction in 126 DAAO gene 74 decision-making tasks 53, 151 declarative memory in bipolar disorder 157 Delayed-Matching-to-Sample test 151 dementia, hyperintense lesions in 22 depression, arginine vasopressin levels in 139 automatic negative thoughts in 150 characteristics of 148 cognitive dysfunction in cf that in mania 148 glucocorticoid receptors for 120 hippocampal volume change in 32 in mania 185 level as predictor of recovery 16 neurotrophin hypothesis of 104 predictors of cf those for mania 17 remission length and presence in index episode 13 risk of relapse after 11 risk of relapse to 15 role of CREB–BDNF–TrkB pathway in 109, 110 role of neurotransmitter deficiency in 103 syndromal and subsyndromal 166 transgenic mouse models for 109 desmopressin, dynamic tests of hypothalamic–pituitary–adrenal axis with 140 despair in mice 108 diagnosis, hierarchy of 71 disgust, recognition of 38 dopamine, role in depression 103 DRD4 receptor gene 83
Index
7/4/05
4:03 pm
Page 225
Index
225
drug companies, power of to change prescribing patterns 177 electroconvulsive therapy (ECT) as an antidepressant 195 cf chlorpromazine 194 cf lithium 194 cf neuroleptics 194 clinical trials of 193, 194 effect on CREB–BDNF–TrkB pathway 104 guidelines for use 194 use by diagnosis 196 use in bipolar cf unipolar depression 197 use in bipolar disorder 193–8 use in mania 193 emotion, neural responses to 37 neural structures required in processing of 44 emotion-processing tasks, neural responses to 40–1 response to emotional target words 40 role of amygdala in 22 employment 211 focus on dysfunction cf function 212 vicious cycle of low expectations 213 endophenotypic markers, genetic liability and grey matter volume 97, 98 genetic liability and white matter volume 97, 99 implications for identification 100 epidemiology, functional outcome of bipolar disorder 9–20 incidence study in south-east London 1–7 of schizophrenia 1 epilepsy, hyperintense lesions in 22 ethnicity, differences in schizophrenia incidence 1 incidence of bipolar disorder and 4 executive function, impairment in affective disorders 118 in bipolar disorder 157 previous illness history and 42 facial expression, enhanced recognition of disgust in 38 impaired recognition of fear in 38 neural responses to in bipolar disorder 37 neural responses to recognition 37–40 normal brain responses cf bipolar disorder and major depressive disorder 39 familial factors, genetic liability scores 96 family studies, familial links of schizoaffective disorder to bipolar disorder 70
Index
7/4/05
4:03 pm
226
Page 226
Bipolar disorder: the upswing in research & treatment
family therapy 168 fear, brain responses to recognition of expression 39 recognition of 38 fluoxetine 140 functional imaging, 43 confounding factors in 159 during Iowa Gambling Task 55 during N-back task 54 linked with cognitive function 153 Maudsley Bipolar Disorder Project 52–6 of Sternberg paradigm 158, 159 of two-back task 158 functional neuroimaging, emotional processing 43 functional outcome, definitions of 15 G72 gene 64, 74 gambling tasks, Iowa Gambling Task 52, 53 neural function during 55 results in mania 152 Wisconsin Card Sorting Task 146, 152 gender, age at onset and 1, 2, 4 as predictor of recovery 16 differences in incidence with 1 differences in schizophrenia incidence 1 ECT use and 196 genetic association studies, description 78, 79 in bipolar disorder 81 genetics, association studies, description 78 in bipolar disorder 81 basis to brain abnormalities 93–101 definition of phenotype 77 effects on hypothalamic–pituitary–adrenal axis 117 explanation for overlap between disorders 71 familial incidence of bipolar and unipolar disorders 69 future developments 87 genetic liability scores 96 glucocorticoid receptor gene expression in mood disorders 126 linkage studies, description 78 in bipolar disorder 80 molecular findings in bipolar disorder 74 molecular genetics in bipolar disorder 82
Index
7/4/05
4:03 pm
Page 227
Index
227
of bipolar disorder 64, 77–88 of neurotransmitters in bipolar disorder 83 of schizophrenia 64 polygenic liability-threshold model 72 polymorphism Val66Met 85, 86 relation between bipolar disorder schizophrenia and unipolar depression 69–75 susceptibility genes 93 glial cells, as basis of neuropathology in bipolar disorder 64, 65 changes in bipolar disorder 124 changes in major depressive disorder 124 decrease in subgenual cingulate cortex 61 types and role 124 glucocorticoid receptors 117 activity of 129 agonists and cortisol 131 antagonists for therapy 119 gene expression 126 impaired function in depression 126 glucocorticoids, effect on cognitive function 118 regulation of corticotrophin releasing hormone by 137 regulation of vasopressin by 137 role in hypothalamic–pituitary–adrenal axis 129, 130 role in neuronal changes 126 go/no-go tasks 149 imaging of response to 150 grey matter, antipsychotics and volume deficit 32 genetic liability and volume 97, 98 imaging of 98 lithium and volume changes 32 protects against changes in 21 loss with age 23 reduction in bipolar disorder 21 reduction in children and adolescents with bipolar disorder 22, 23 volume changes 41, 51 group psycho-education 168 growth-associated protein-43, level in anterior cingulate cortex 124 guidelines, biased 177 British cf American 181 current treatment guidelines 170 for acute mania 181 for use of ECT 194 guinea pig, cortisol in 131
Index
7/4/05
4:03 pm
228
Page 228
Bipolar disorder: the upswing in research & treatment
haloperidol, cf olanzapine 189 cf placebo 183 cf valproate in psychotic mania 181 combination treatment with 186 effect on depression after mania 186 placebo-controlled monotherapy 188 use with lorazepam 182 happiness, brain responses to recognition of expression 39 heterogeneity, causes of 33 in brain 33 of bipolar disorder 201 of responders 202 hippocampus, BDNF and CREB mRNA levels in 105 neuronal changes in 61 structural changes in 41 synaptic pathology in 61 volume changes in 31, 32 volume reduction and glucocorticoid receptors 126 hospitalization, length of as a predictor of recovery 16 hypercortisolaemia 118 effect on cognitive function 121 role of vasopressin in 135–41 hyperintense lesions 22 hypothalamic–pituitary–adrenal axis, biological factors in hyperactivity 135–41 control of activity 129, 130 cortisol and its receptors 117 description of 116 development of therapies for dysfunction of 119 dynamic tests of 140 dysfunction in bipolar disorder 118 genetic effects on 117 hyperactivity of and mood disorder 126 reasons for hyperactivity of 131 response to stress 116, 130 role in bipolar disorder, 115–21 hypothalamic–pituitary–adrenal system, role in depression 103 imaging, see under computed tomography; functional neuroimaging; magnetic resonance imaging; magnetic resonance spectroscopy imipramine, effect in mice 108 incidence 37 clinical epidemiology study in London 1–7 increase over time 6 variation in 1
Index
7/4/05
4:03 pm
Page 229
Index interpersonal social rhythms therapy 167 Iowa Gambling Task 52, 53 neural function during 55 Kraepelinian dichotomy 73 lamotrigine, as long-term treatment 202 cf lithium 203 clinical characteristics of responders 204, 205 lateral ventricles, enlargement of 27, 29 right, volume changes in 29, 30, 31 total volume changes 31 linkage analysis, description 78, 79 in bipolar disorder 80 lithium 175–9 as long-term treatment 202 cf ECT 194 cf lamotrigine 203 cf olanzapine 203 cf valproate 179 clinical characteristics of responders 204 cognitive function and 42 combination treatment with antipsychotic 186 comparison with other drugs 177 decline in use 175 demands of treatment, on patient 176 on psychiatrist 176 effects on functional imaging studies 160 efficacy throughout bipolar spectrum 178 grey matter volume changes and 32 in carbamazepine responders 206 in combination with antipsychotics 187 monotherapy cf combination therapies 207 neurogenesis and 63 prescription at discharge 18 reasons for decline in use 176 selective publication as reason for decline in use 177 subgenual cingulate gyrus changes and 42 use and functional outcome 14 use in acute mania 178 use in Italy 176 lorazepam use with haloperidol 182 McLean–Harvard first-episode study 15–19
229
Index
7/4/05
4:03 pm
230
Page 230
Bipolar disorder: the upswing in research & treatment
magnetic resonance imaging (MRI), brain volume changes on 29 for brain abnormalities in bipolar disorder 21–4 go/no-go task activation 150 meta-analysis 27–34 neuropathology studies with 60 white matter hyperintensities in bipolar disorder 125 white matter hyperintensities in major depressive disorder 125 magnetic resonance spectroscopy description of technique 22 for brain abnormalities in bipolar disorder 21–4 maintenance treatment 187 major depressive disorder, autonomic responses in 40 changes in synaptic protein levels in 124, 125 genetic links to bipolar disorder and schizophrenia 65 glial cell changes in 124 hippocampal volume reduction in 126 neuropathology of 123 position on affective spectrum 147 response to facial expression in 39 vasopressin in 138 mania, age at onset 6 characteristics of 149 cognitive dysfunction in cf that in depression 148 decision-making impairment in 152 depression in 185 ECT use for 193 efficacy of valproate in 182, 183 first-episode, four-year follow-up after 9–14 incidence and gender 5 predictors of functional outcome after 14 psychosis during 3 recovery after 15 incidence rate 3 position on affective spectrum 147 predictors of cf those for depression 17 relapse and recovery from 10 repetitive transcranial magnetic stimulation in 198 risk of relapse after 10, 11 risk of relapse to 15 severity and brain activity 38 use of lithium as gold standard 178 valproate cf antipsychotics for 181 MAOA gene 83 Maudsley Bipolar Disorder Project 52–6
Index
7/4/05
4:03 pm
Page 231
Index Maudsley Family Study of Psychosis 95 medication, as confounding factors in functional imaging studies 159 combination treatments 186 effect on neuronal changes on 63 fall in lithium usage 175 monotherapy cf combination therapies 207 valproate cf antipsychotics 181–90 mifepristone, effect on cognitive function 120 effect on cortisol levels 131 use in bipolar affective disorder 120 use in depression 120 migrants, incidence in 2 incidence increase and 6 mineralocorticoid receptors 117 activity of 129 agonists and cortisol 131 mitogen-activated protein (MAP) kinase, role of BDNF in cascade 110 Modified Location Code Index 11 Modified Vocational Status Index 11 molecular genetics in bipolar disorder 82 mood congruence, as predictor of mania but nor depression 17 risk of relapse and 10, 13 mood induction, neural responses to 40 morbidity 165 mortality 37, 165 MRC study 169–71 MRI, see under magnetic resonance imaging MRS, see under magnetic resonance spectroscopy N-acetylaspartate (NAA) levels in bipolar disorder 22 N-back task 53 neural function during 54 neural function, during Iowa Gambling Task 55 during N-back task 54 neural responses to facial expression 37 neuroanatomy, functional 51 of arginine vasopressin 135 structural 51 neurocognitive dysfunction 146, 148 neurogenesis and lithium 63 neuroleptics, attention deficits and 42
231
Index
7/4/05
4:03 pm
232
Page 232
Bipolar disorder: the upswing in research & treatment
cf ECT 194 neuropathology 59–66 commonalities between disorders 125 of bipolar disorder 123 of major depressive disorder 123 neurotransmitters, BDNF 83, 84 CREB–BDNF–TrkB pathway 104 deficiencies 103 G72/G30 locus and DAAO 83 genetics of in bipolar disorder 83 neurotrophin hypothesis 103 neurotrophin, role in depression 104, 105 transgenic mouse models of 103–10 neurotrophins, antidepressant effects of 106 New Tower of London task 150, 151 norepinephrine, role in depression 103 obsessive–compulsive disorder, BDNF gene in 86 occupational status, alcoholism and 12 as predictor of new episodes 17 factors affecting 19 functional outcome and 14 risk of relapse and 11 olanzapine, 182 as long-term treatment 202 cf haloperidol 189 cf lithium 203 cf placebo 183 cf valproate 185 clinical characteristics of responders 205 combination treatment with 186 controlled trials with 183 effect on depression after mania 186 in combination with lithium 187 side-effects of 185 orbitofrontal cortex, neural responses in 149 neuron size in 124 neuropathology in 60, 61 outcome, clinical characteristics of responders 204 four-year follow-up after first-episode of mania 10 functional 9–20 definitions of 15 factors affecting 19
Index
7/4/05
4:03 pm
Page 233
Index importance of 20 predictors of 14 risk factors for poor 10, 11, 12 heterogeneity of responders 202, 203 improvement of by long-term treatment selection 201 neural function as a predictor of 43 selectivity of responses 206 paraventricular nucleus 136 pathophysiology 51–6 of bipolar disorder 115 personality disorder, co-morbidity with bipolar disorder 166 phenotype, definition of 77 endophenotypes 93, 94 placebo trials 183 polygenes 71 polymorphism Val66Met 85, 86 positron emission tomography (PET) neuropathology studies 60 post-traumatic stress disorder, hippocampal volume reduction in 126 postmortem findings 59–66 corticotrophin releasing hormone-expressing neurones 139 CREB levels 105 phase of illness at time of death and 62 vasopressin-expressing neurones 139 predictors, endophenotypic markers 94, 95 genetic liability scores 96 implications for identification 100 neural abnormalities as 43 of functional outcome after mania 14 of outcome, identification of 9 of recovery 16 of relapse 13, 17 prefrontal cortex, activity and facial expression recognition 38 activity during semantic cf orthographic tasks 40 changes in 23 decreased activity in 51 functional imaging studies 52–6 neuroanatomical changes in 51 neuron size in 124 response to emotion in 38 structural changes in 41 presentation, variations in 78 probabilistic reversal learning 151 psychological treatments, current treatment guidelines 170
233
Index
7/4/05
4:03 pm
234
Page 234
Bipolar disorder: the upswing in research & treatment
MRC study 169–71 studies of 167 use in bipolar disorder 165–71 psychosis, as predictor of functional outcome 19 during first-episode mania 3 hypothalamic–pituitary–adrenal axis hyperactivity in 130, 131 remission length and presence in index episode 13 risk of relapse and incidence 10, 12 verbal impairment in 42 psychotropics, neuroanatomy and 42 prescription at discharge 18 use and functional outcome 13 quetiapine, cf placebo 183 combination treatment with 186 Rapid Visual Information Processing task 153, 154 recovery, after first-episode mania 15 definitions of 15 predictors of 16 syndromal cf functional 15, 19 relapse, after mixed episode 17 occupational status as predictor of 17 odds ratios for 166, 168 predictors of 13 risk factors for 10, 11 time to onset 18 remission, probability of remaining in 10 risk factors for shortened 13 risk for relapse from 10 repetitive transcranial magnetic stimulation bipolar disorder use in 193–8 description 197 in bipolar depression 198 in mania 198 residential status, factors affecting 19 functional outcome and 14 risk of relapse and 11 risperidone, cf placebo 183, 184 combination treatment with 186 in combination with lithium 187
Index
7/4/05
4:03 pm
Page 235
Index in combination with valproate 187 RU-486, see under mifepristone schizoaffective disorder, familial links to bipolar disorder 70 schizophrenia BDNF gene in 86 brain volume changes in 29, 31 cf bipolar disorder 4 clinical epidemiology studies of 1 gender differences in age at onset 6 genetic relation to bipolar disorder 69–75 genetics of 64 hyperintense lesions in 22 knowledge of cf bipolar disorder 1 seasonal affective disorder 153 serotonin, role in attentional processing 149 role in depression 103 serotonin transporter (5-HTT) gene 83 side-effects, avoidance of with antipsychotics 182 of lithium 176 of olanzapine cf valproate 185 weight gain 187 social functioning in bipolar disorder 145 status epilepticus, valproate for 182 Sternberg paradigm 157 brain regions active during 161 functional imaging studies of 158, 159 stress, effect on brain 123–7 effect on CREB–BDNF–TrkB pathway activity 104 effect on CREB–BDNF–TrkB pathway in mice 109 effects on corticotrophin releasing hormone and arginine vasopressin 138 hypothalamic–pituitary–adrenal axis response to 116, 130 subcortical regions, activity and facial expression recognition 38 activity in response to emotive scenes 40 expansion of 37–44 subgenual cingulate cortex, glial cells in 61 lithium and 42 neuronal changes in 61 neuropathology studies with 60 structural changes in 41 subgenual prefrontal cortex, grey matter reduction in 21 heterogeneity within 33
235
Index
7/4/05
4:03 pm
236
Page 236
Bipolar disorder: the upswing in research & treatment
substance abuse, comorbid with bipolar disorder 17 comorbidity with bipolar disorder 166 development after mania 19 suicide incidence in bipolar disorder 166 susceptibility genes 93 synapses, changes in bipolar disorder 124 changes in major depressive disorder 124 function of, genetics and 64 pathology of, 61 synaptic protein levels in anterior cingulate cortex 124 synaptophysin level in anterior cingulate cortex 124 transcranial magnetic stimulation 193–8 cf ECT 197 transgenic mouse models 103–10 compromised BDNF expression 107 compromised CREB expression 107 compromised TrkB expression 108 CREB–BDNF–TrkB pathway studies in 106 treatment, importance of speed in 214 triplets, bipolar disorder and schizophrenia in 70 tryptophan, effect on attentional processing 149 twin studies 70 bipolar disorder and schizophrenia in triplets 70 two-back task, brain regions active during 160 functional imaging studies of 158 tyrosine receptor kinase B (TrkB), response to antidepressants 106 role in CREB–BDNF–TrkB pathway 104 unipolar depression, genetic relation to bipolar disorder 69–75 Val66Met polymorphism 85, 86 valproate, advantages and disadvantages of 181–90 cf antipsychotics 181–90 cf haloperidol in psychotic mania 181 cf lithium 175, 179 cf olanzapine 185 cf placebo 183 combination treatment with antipsychotic 186 delayed onset 182 efficacy in mania 182 in combination with antipsychotics 186, 187 maintenance treatment 187
Index
7/4/05
4:03 pm
Page 237
Index mechanism of action cf antipsychotics 182 side-effects of 185 weight gain on 187 vasopressin, dynamic tests of vasopressin-ergic system 140 in major depressive disorder 138 regulation by glucocorticoids 137 role in hypothalamic–pituitary–adrenal axis hyperactivity 135–41 vasopressin receptors 136 verbal impairment, in psychosis 42 weight gain 187 white matter, genetic liability and volume 97, 99 hyperintensities in 27, 65 and outcome 43 on MRI 125 volume changes and 32 imaging of 99 Wisconsin Card Sorting Task 146, 152 working memory, in bipolar disorder 157 ziprasidone 182 cf placebo 183 combination treatment with, 186
237
Index
7/4/05
4:03 pm
Page 238