The Overlap of Affective and Schizophrenic Spectra

This page intentionally left blank

The Overlap of Affective and Schizophrenic Spectra

An increasing number of clinicians and researchers are now favoring an overlap between affective and schizophrenic spectra. In this book, an international team of experts discuss aspects of comorbidity, genetic models, clinical course, phenomenology and therapies. This is the first comprehensive overview of this area of overlap. Challenging cases presenting clinical and paraclinical features of both spectra are surprisingly numerous. Not only the phenomenology but also the course, outcome and treatment of such cases have their own characteristics. Recent research shows that the overlap also involves genetics and biological processes related to psychotic disorders. Within the overlap of affective and schizophrenic spectra it is possible to identify some groups of disorders having similar clinical and non-clinical features: the “Schizoaffective” group, “Acute and Transient Psychotic Disorder” or “Brief Psychosis,” and other groups found in so-called “Atypical Forms.” Andreas Marneros is Professor of Psychiatry in the Department of Psychiatry and Psychotherapy at the Martin Luther University in Halle-Wittenberg, Germany. He won the Kraeplin Research Prize in 2002 for his work on psychoses, especially schizoaffective and acute brief psychoses. Hagop Akiskal is Professor of Psychiatry and Director of the International Mood Center at the

University of California at San Diego. Among his awards are the Anna Monika Prize for research on the depressive spectrum, and the World Psychiatric Association’s Jean Delay Prize for bridging clinical and research perspectives in the bipolar spectrum.

The Overlap of Affective and Schizophrenic Spectra Andreas Marneros Department of Psychiatry and Psychotherapy, Martin Luther University, Halle-Wittenberg, Germany

Hagop S. Akiskal International Mood Center, University of California at San Diego, USA

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521858588 © Cambridge University Press 2007 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2006 ISBN-13 ISBN-10

978-0-511-26872-4 eBook (EBL) 0-511-26872-6 eBook (EBL)

ISBN-13 ISBN-10

978-0-521-85858-8 hardback 0-521-85858-5 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

1

List of contributors

page vii

Preface

page xi

The paradigma of overlapping affective and schizophrenic spectra: schizoaffective conditions

1

Andreas Marneros

2

The overlapping of the spectra: overlapping genes and genetic models

25

John R. Kelsoe

3

The continuum of psychosis and its genetic basis

43

T. J. Crow

4

Functional psychoses: molecular-genetic evidence for a continuum

55

Hans H. Stassen, Christian Scharfetter and Jules Angst

5

State- and trait-related deficits in sustained attention in bipolar disorder: are there any overlaps with schizophrenia?

79

Luke Clark

6

The concept of schizoaffective disorder: utility versus validity and reliability – a transcultural perspective

104

Ahmed Okasha

7

Phenomenological approaches to the schizoaffective spectrum

133

William Coryell

8

Clinical course of schizoaffective disorders

145

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros

9

Depressive syndromes in schizophrenia Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

v

156

vi

Contents

10

The overlapping of the spectra: brief and acute psychoses

182

Frank Pillmann and Andreas Marneros

11

Overlapping of the spectra: physical comorbidity between schizophrenia and affective disorders

207

William S. Stone, Andrea H. Roe and Ming T. Tsuang

12

The overlapping of the spectra suicide

224

Simavi Vahip

13

Biological treatment of schizoaffective disorders

248

Christopher Baethge

14

Psychological therapies and schizoaffective disorders

264

Jan Scott

Epilogue: The interface of affective and schizophrenic disorders: a cross between two spectra?

277

Hagop S. Akiskal

Index

293

Contributors

Hagop S. Akiskal International Mood Center University of California at San Diego VA Psychiatry Service 116A 3350 La Jolla Village Drive San Diego, CA 92161 USA

Ellen Bittner Department of Psychiatry Heinrich-Heine-Universität Düsseldorf Bergische Landstrasse 2 Postbox 12 05 10 D-40605 Düsseldorf Germany

Jules Angst Psychiatrische Universitätsklinik Zürich Postfach 68 CH_8029 Zürich Switzerland

Luke Clark Department of Experimental Psychology University of Cambridge Downing Street Cambridge CB2 3EB UK

Christopher Baethge Department of Psychiatry and Psychotherapy University of Cologne Medical School Kerpener Str. 69 50 924 Cologne Germany Antoni Benabarre Bipolar Disorders Program Institute of Neuroscience Hospital Clinic University of Barcelona Spain

vii

William Coryell University of Iowa Department of Psychiatry 500 Newton Rd Iowa City IA 52242 USA T. J. Crow SANE Prince of Wales International Centre Warneford Hospital Oxford OX3 7JX UK

viii

List of contributors

Wolfgang Gaebel Department of Psychiatry Heinrich-Heine-Universität Düsseldort Bergische Landstrasse 2 Postbox 12 05 10 D-40605 Düsseldorf Germany

Maria Reinares Bipolar Disorders Program Institute of Neuroscience Hospital Clinic University of Barcelona Spain Andrea H. Roe

John R. Kelsoe Department of Psychiatry, 0603 University of California, San Diego 9500 Gilman Drive La Jolla, CA 92093 USA

Christian Scharfetter Psychiatrische Universitätsklinik Zürich Postfach 68 CH-8029 Zürich Switzerland

Andreas Marneros Klinik und Poliklinik für Psychiatrie und Psychotherapie Martin Luther Universität Julius-Kühn-Strasse 7 06097 Halle/Saale Germany

Jan Scott P. O. Box 96 Department of Psychological Medicine Institute of Psychiatry Denmark Hill London, SE5 9AF UK

Ahmed Okasha Director, WHO Collaborating Center for Training and Research in Mental Health Institute of Psychiatry Ain Shams University Cairo Egypt

Hans H. Stassen Psychiatrische Universitätsklinik Zürich Postfach 68 CH-8029 Zürich Switzerland

Frank Pillmann Department of Psychiatry and Psychotherapy Martin Luther Universität Julius-Kühn-Strasse 7 06097 Halle/Saale Germany

Ming T. Tsuang Massachusetts Mental Health Center Division of Public Sector Psychiatry, Beth Israel Deaconess Medical Center and Harvard Medical School Departments of Psychiatry Boston, MA USA

William S. Stone

ix

List of contributors

Simavi Vahip Affective Disorders Unit Ege University Medicine Faculty Department of Psychiatry Izmir Turkey Eduard Vieta Bipolar Disorders Program Institute of Neuroscience Hospital Clinic University of Barcelona Spain

Wolfgang Wölwer Heinrich-Heine-Universität Düsseldorf Bergische Landstrasse 2 Postbox 12 05 10 D-40605 Düsseldorf Germany

Preface

Affective and schizophrenic disorders are not monolithic concepts. The idea of groups of disorders has proved to be useful. Today the concept of a spectrum of affective disorders and a spectrum of psychotic disorders is based on clinical and research findings. It has also been proposed that a kind of continuity exists between the two spectra. There are clinical bridges joining them, or perhaps some nosologic islands filling the gaps in-between. Possibly what is “in-between” represents a cross of the underlying dimensions of the two “voluminous” spectra, or a superposition of some of the contributory factors of one or that of the other. Psyche is like Physis – Nature. She does not take any leaps, even when she is ill. Emil Kraepelin himself realized the indistinct boundaries of the dichotomy of the manic-depressive and schizophrenic psychoses he had proposed at the turn of the nineteenth century. He thought that it was only partially true. One can read it in his paper published in 1920: “Die Erscheinungsformen des Irreseins”, which means “The manifestation types of insanity”. Among other relevant observations, he noted: No experienced psychiatrist will deny that there is an alarmingly large number of cases in which, despite the most careful observation, it seems impossible to arrive at a reliable diagnosis . . . We therefore will have to get used to the fact that the symptoms we have used so far are not sufficient to always reliably distinguish between manic-depressive insanity and schizophrenia, but that there are overlaps based on the origin of these symptoms from given preconditions.

Developments in psychiatry in the last century have actually confirmed the doubts of Emil Kraepelin, and have shown that the so-called Kraepelinian dichotomy is not the philosopher’s stone. Many scientists in various countries of the world have attempted to fill the gap with different concepts: in Germany with the “cycloid psychoses”, in France with the “bouffée délirante”, in Scandinavia with the “psychogenic” and “reactive psychoses”, in Japan with the “atypical psychoses”, in the United States with the “schizoaffective disorders” or with “remitting schizophrenia”. The major diagnostic systems of DSM xi

xii

Preface

and ICD accept the existence of an intermediate area called “schizoaffective”. Nevertheless the uncertainties in diagnosis and nosology remain. One of the reasons for these uncertainties is the inconsistent definition of schizoaffective psychopathology. To better understand the overlap between the affective and schizophrenic spectra it is necessary, among others, to consider the space in-between in a diachronic continuity. Psychiatric symptoms have to be considered in their evolution from fundamental structures such as temperament and personality. Modern dimensional models deliver both clinical and theoretical aspects for syndromes to depict the spectrum. In the contemporaneous evolution of psychiatry, the concept of a domain based on the overlap of the spectra is, to a varying degree, supported by genetic, biological, psychological, clinical, therapeutic and longitudinal findings. The purpose of this book is to present such findings, theories, and methods and to highlight their relevance and limitations regarding the interface of affective and schizophrenic disorders.

1

The paradigma of overlapping affective and schizophrenic spectra: schizoaffective conditions Andreas Marneros Department of Psychiatry and Psychotherapy, Martin Luther University Halle-Wittenberg

Broad–narrow–broad: the circuitry of certainties and uncertainties In the centuries between the great Greek founders of medicine and psychiatry, Hippocrates, Aretaeus of Cappadocia, Galenos of Pergamon or Soranos of Ephesos (see Longrigg, 1993; Marneros and Angst, 2000; Angst and Marneros, 2001) and the father of the modern psychiatric systematics, Emil Kraepelin, at the end of the nineteenth and the beginning of the twentieth century, physicians and psychiatrists described and allocated mental disorders according to broad criteria. Symptoms, which today in modern nomenclature are called “schizophrenic,” “affective,” “mood congruent” or “mood-incongruent,” were described as characteristics of the same disorder. Therefore, case reports published during this long historic period of more than 2400 years could, with the same strong arguments, be interpreted by modern psychiatrics as “pure schizophrenia,” or “pure affective disorder” or “typical schizoaffective.” At the end of the nineteenth century, Emil Kraepelin tried to clean the field, dichotomizing the so-called functional psychotic disorders into dementia praecox and manic-depressive illness (Kraepelin, 1896; 1899). The Kraepelinian dichotomy, which really was not very dichotomous, as Emil Kraepelin himself pointed out in 1920, received an epigonal strength by Kurt Schneider (1959) through the definition of “first-rank schizophrenic symptoms”: their existence confirms the diagnosis “schizophrenia” (provided that organic causal conditions can be ruled out). Kurt Schneider’s ascetic strength completed in some way Karl Jaspers’ hierarchical principle (1913). According to Jaspers’ principle, “schizophrenic” symptoms eliminate the diagnostic validity of “mood” symptoms (putting at the base of the hierarchical pyramid the “organic The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

1

2

Andreas Marneros

symptoms” – strongest – and on the top the “neurotic” symptoms – weakest). Jaspers’ and Schneider’s influence contributes essentially to the fact that the great majority of psychiatrists of that time ignored or forgot that Emil Kraepelin himself accepted a non-classifiable domain consisting of a mixture of both conditions: schizophrenia and affective disorders. More paradoxically, it has been ignored for a long time – even by his fellows – that Kurt Schneider himself described the “cases-in-between.” At the end of the twentieth century, it became certain that a clear dichotomy of the so-called functional mental disorders into schizophrenic and mood disorder was impossible. Clinical, prognostic, pharmacological, biological and genetic findings supported the existence of a “bridge” or a “continuum” between these disorders (Marneros et al., 1995a; 1995b). Schizophrenic and mood disorders show overlaps. One of the efforts to identify and describe this overlapping of the spectra results in the concept of schizoaffective disorders. The major diagnostic systems, International Statistical Classification of Diseases and Related Health Problems (ICD) and Diagnostic and Statistic Manual of Mental Disorders (DSM), accept the existence of an intermediate area called “schizoaffective.” Nevertheless, the uncertainties in diagnosis and nosology remain. One of the reasons for these uncertainties is the inconsistent definition of the schizoaffective area (see below). Another reason is that the definitions of affective disorders became broader, especially in DSM-IV: mood-incongruent symptoms, even Kurt Schneider’s “first-rank symptoms,” were declared compatible with the diagnosis “mood disorder,” making the diagnosis “schizoaffective” narrower. The relativization of the diagnostic power of symptoms caused a relativization of the diagnosis as well. The boundaries between schizophrenia, schizoaffective and mood disorders became more diffuse and confused. Nevertheless, the diffusion and confusion underline more intensively the overlapping of schizophrenic and affective spectra. Its identification, however, still remains a challenge (Marneros and Tsuang, 1986; Marneros and Angst, 2000; Marneros and Goodwin, 2005a). Empirical efforts to identify the schizoaffective overlap The large group of patients who manifest the symptoms or characteristics of both major disorders – schizophrenic and mood disorder – is a challenge for theorists and clinicians. These patients present a conceptual problem to theorists, a therapeutic problem to clinicians, and a classification problem to researchers. The problem of the “intermediate psychotic area” or the “cases-in-between” is even older than the term “schizoaffective” itself, which was originated by Kasanin (1933) (Maj, 1984; Angst, 1986; Perris, 1986; Pichot, 1986; Strömgren, 1986; Marneros and Angst, 2000; Angst and Marneros, 2001). Perhaps it began with Karl

3

The paradigma of overlapping spectra

Kahlbaum (1863) and lead through Kraepelin’s work to the clinical empiricists of the twentieth century. Karl Kahlbaum can be considered the first psychiatrist in modern times to describe some kind of “schizoaffective disorders” as a separate group in “vesania typica circularis,” applying both cross-sectional and longitudinal criteria (Kahlbaum, 1863). Nevertheless, states described as melancholia or mania by authors of the classic period, for instance by Hippocrates or Aretaeus of Cappadocia, and authors at the beginning of modern scientific psychiatry like Heinroth (1818), Griesinger (1845), or even the father of the word “psychiatry,” Johann Christian Reil (1803–12, in Marneros and Pillmann, 2005), were very often “schizoaffective” according to the modern nomenclature. Emil Kraepelin was also acquainted with cases between “dementia praecox” and “manic-depressive insanity” (Kraepelin, 1893; 1896; 1920). These “cases-in-between” (Schneider, 1959) were a problem for him, a nuisance, but on the other hand an interesting challenge. As is well known, Kraepelin dichotomized the so-called endogenous psychoses into two groups, namely “dementia praecox” (with a poor outcome) and “manic-depressive insanity” (with a favorable outcome). However, he already knew that not all cases of “endogenous” mental disorders could readily be classified into the two categories. Some cases of “mixed states,” “delirious mania,” and other mental disorders described by Kraepelin (1893; 1920) could be allocated to both categories or to neither of them. In a critical appraisal of his own taxonomy, Kraepelin wrote in his important paper of 1920 (“Die Erscheinungsformen des Irreseins” – “The Phenomenological Forms of Insanity”), that mental disorders can have elements of both groups, namely “dementia praecox” and “manic-depressive insanity” and that they can have a different course and a different prognosis than “dementia praecox” as well. He knew that the boundaries between the two groups of mental disorders are elastic and that there are bridges connecting them. His doubts became stronger in the wake of an investigation by his pupil and colleague Zendig. Zendig reported in his paper “Contributions to Differential Diagnosis of Manic-Depressive Insanity and Dementia Praecox” (1909) that approximately 30% of Kraepelin’s samples diagnosed with “dementia praecox” (using Kraepelin’s guidelines) had a course and outcome not corresponding to that of “dementia praecox”; Zendig attributed the good outcome to an incorrect diagnosis. Later Kraepelin saw in such cases a weakness of his dichotomy concept. He wrote: “The cases which are not classifiable (namely to manic-depressive insanity or dementia praecox) are unfortunately very frequent.” (Kraepelin, 1920, p. 26). Two pages later he made a decisive and for him certainly not easy statement: “We have to live with the fact that the criteria applied by us are not sufficient to differentiate reliably in all cases between schizophrenia and manic-depressive insanity. And there are also many overlaps

4

Andreas Marneros

in this area.” (i.e., between schizophrenia and affective disorders) (Kraepelin, 1920, p. 28). Eugen Bleuler (1911; 1924) recognized the occurrence of mood syndromes in patients he diagnosed as schizophrenic. For the most part, he ultimately concluded that schizophrenia was the illness in question and did not see these patients as a group, but rather as single, aberrant cases. Schneider (1959; 1973), in differentiating between schizophrenia and mood disorder (“cyclothymia” in his nomenclature), described “cases-in-between” in which the diagnoses of both schizophrenia or mood disorder could be made with equally strong arguments. He distinguished between concurrent and sequential forms of “cases-in-between,” and his definition bore great similarity to modern ones (Marneros et al., 1986a). Kasanin (1933) described a group of cases which are quite “atypical” for both schizophrenia and mood disorder; so he introduced the term “schizoaffective.” These cases are fairly young individuals, socially quite well integrated, who suddenly blow up in a dramatic psychosis and present a clinical picture which may be called either schizophrenic or affective and in whom the differential diagnosis is extremely difficult. These patients had often had a history of a previous attack in late adolescence, but otherwise had good premorbid adjustment. Onset was sudden, accompanied by emotional turmoil, a distortion of the outside world and, for some, the presence of false sensory impressions, but not passivity or withdrawal. Onset was often precipitated by a definite environmental stress. The duration of symptoms would be a few weeks or months, followed by full recovery. Kasanin’s definition was embedded in the attempt to pin down such “atypical” disorders (Marneros and Pillmann, 2005). The Norwegian psychiatrist Langfeldt (1939) added his definition of “schizophreniform psychoses,” whereby a patient’s inheritance was uncorrelated with his prognosis. He showed cyclothymic temperament, pyknic habitus, depressive symptoms, self-reference tendencies, cloudiness, incoherence, catatonic or pathoplastic features, and the illness had an acute onset characterized by distinct precipitants. Youth and good premorbid adjustment are absent from this definition. Kant (1940) described “recovered schizophrenics” as having more bipolar affectives than schizophrenics among their relatives, pyknic physique, psychogenic precipitants of the psychotic episode, acute or sub-acute onset, duration of several months, complete recovery, clouding of consciousness, psychotic experience, ideas of reference, and auditory hallucinations. Here, family history and duration of symptoms are considered for the first time. However, Vaillant (1962) found the family history of recovered schizophrenics heavy with unipolar affectives, while the patients themselves were also typically depressed, though the operational criteria he used included acute onset, confusion or disorientation during the acute episode,

5

The paradigma of overlapping spectra

good premorbid adjustment, a clear precipitating event, and remission to the best premorbid level. In common, these somewhat haphazardly derived criteria for schizoaffective disorder emphasize the sudden onset, presence of confusion or disorientation, and a good recovery. The confusion or disorientation suggests a greater affinity of the illness with schizophrenia, from which it primarily differs by virtue of outcome. Kleist (1928), continuing the tradition of Wernicke (1900) and in his tradition Leonhard (1957), described the “cycloid psychoses,” the more important features of which are the polarity of the symptoms and the favorable outcome (Perris, 1986; Marneros and Pillmann, 2005). As early as 1966, Jules Angst investigated the schizoaffective disorders (under the term “Mischpsychosen” – “mixed psychoses”) as a part of the affective disorders. This was an outlier’s position, not only against the “zeitgeist,” but also contrary to the opinion of his teacher Manfred Bleuler, who assumed them to be a part of schizophrenia. Later investigations by Angst and his group (1989; 1990), by Clayton et al. (1968), by other members of the Winokur group (Fowler et al., 1972), by Cadoret et al. (1974) and the comparative studies of Marneros and coworkers (1986 a–c; 1988 a–c; 1989a–e; 1991b) supported more and more the opinion that the relation between schizoaffective and affective disorders is stronger than the relation between schizoaffective and schizophrenic disorders. Using larger sample sizes and more sophisticated statistical methods, Astrup and Noreik (1966) published a series of reports from 1957 to 1966 in which they analyzed the outcome of more than 1200 cases of schizophrenia. Those schizophrenics who recovered (n  131) instead of deteriorating (n  416) were the ones who showed affective symptoms: elation, psychomotor agitation, flight of ideas, mood swings. Here, the affinity of the illness with affective disorder was seen as central. Applying Schneider’s first-rank symptoms along with Bleuler’s criteria to specify the schizophrenic dimension within schizoaffective disorder, Spitzer et al. (1978) wrote the following rules for the diagnosis: i.e the research diagnostic criteria (RDC) for schizoaffective disorder (manic/depressive subtypes). These included the full manic or depressive syndrome and at least one of the following symptoms suggestive of schizophrenia: delusions of being controlled or of thought broadcasting, insertion or withdrawal; non-affective hallucinations for several days straight or intermittently throughout a week; auditory hallucinations (a voice doing a running commentary on the subject or two or more voices conversing); more than a week of delusions or hallucinations without accompanying prominent depressive or manic symptoms; more than a week of marked formal thought disorder with blunted or inappropriate affect or delusions or hallucinations; or grossly disorganized behavior, but no prominent manic symptoms. In addition, the symptoms were to have a duration of at least one week, with a temporal overlap of affective with schizophrenia-like symptoms.

6

Andreas Marneros

In the nineteen seventies period of the twentieth century, it became common for the clinical diagnosis of schizoaffective disorder to require both affective and schizophrenic symptoms, combined with a minimum duration. Welner et al. (1977), Kendell and Gourlay (1970), Angst et al. (1979), Mendlewicz et al. (1980), Perris (1966), and Tsuang et al. (1976) provide examples. According to Welner et al. (1977), there should be enough affective symptoms to make the diagnosis of schizophrenia unlikely, yet not necessarily enough to meet the criteria for schizophrenia: sufficiently severe thought and behavior disorders, and at least one of the following: acute onset, episodic course, or confusion; and psychosis not associated with alcohol, drug abuse, or known organic brain disease. The findings of Welner et al. based on these criteria (1977; 1979) did not support the traditional association of schizoaffective disorder and good prognosis. Over 70% of 114 patients diagnosed as suffering from schizoaffective or related psychoses had a chronic course of illness, and over 80% of these chronic cases deteriorated. A related family study of 27 relatives with psychotic symptoms showed that 20 probands with both affective and schizophrenic symptoms also had a chronic course of illness. It must be noted, however, that neither a minimum duration of symptoms nor a complete remission of symptoms between episodes was required for the selection of these probands. A study by Himmelhoch et al. (1981) which supports these findings similarly neglected to require a minimum duration of symptoms. In the view of Kendell and Gourlay (1970), either schizophrenia or paranoid psychosis must first be present. For schizophrenia, one of the core symptoms (thought insertion, withdrawal, broadcasting, echoes, voices, delusions of control) or two objective signs, either behavioral (mannerisms, posturing, stereotypes, catatonic phenomena, or behavior suggesting hallucinations) or affective (suspicion, perplexity, blunting, or incongruity) or relating to speech (neologisms, incoherence, non-social speech) must be present. For paranoia, a delusion involving the external world, such as delusions of influence, persecution, reference, misinterpretation, etc., must be present and must be persistent and preoccupying; the patient must show conviction. Second, either depression or mania must be present. For depression, four items from a list of 16 must be fully rated (sadness, hopelessness, suicidal intent, loss of interest, inferiority, pathologic guilt, hypochondriacal delusions, nihilistic delusions, insomnia, muddled thoughts or poor concentration, morning depression, and loss of appetite, libido or emotions), and three signs must be evident (observed sadness, agitation, retardation). For mania, three fully rated items from the following list of five symptoms (euphoria, racing thoughts, tirelessness, delusions of special powers, delusions of grandiose identity) and seven signs (over-activity, distractibility, irreverent behavior, embarrassing behavior, hypomanic affect, pressure of speech, flight of ideas) must be present.

7

The paradigma of overlapping spectra

In Angst’s definition, as well as in Welner’s, both affective and schizophrenic symptoms should be present, each strong enough to make the opposite diagnosis unlikely. Patients should show a tendency towards remission with no marked defect, yet also towards recurrence. Angst studied the morbidity risk for schizophrenia among 1000 first-degree relatives of 150 schizoaffectives selected by these criteria. He found the risk to be 5.26%, compared with a risk of affective disorder of 6.7%. Full remission among schizoaffectives was less common (43%) than among those with bipolar disorders (73%) (Angst et al., 1979). If Welner’s findings suggest that schizoaffective disorder bears a relationship to schizophrenia by virtue of being similarly chronic, Angst’s findings rather suggest that schizoaffective disorder is equally related to pure schizophrenic and mood disorder. The criteria of Mendlewicz (1980) include episodic affective syndromes of the manic or depressive type and at least one schizophrenic episode not concurrent with an affective syndrome. The concept of atypical schizophrenia according to Tsuang et al. (1976) relies on a diagnosis of schizophrenia essentially based on the Feighner criteria, but with either short duration or the possibility of another diagnosis and either a previous remitting illness or affective symptoms at the time of admission. Tsuang et al. (1986) presented a detailed example of the application of similar criteria for the purpose of subtyping schizoaffective disorder. Despite the impression of catalogued precision which some of the above definitions gave and certain themes which recur among them, a number of them have been shown not to select the same group of patients. In 1979, Brockington and Leff reported the results of an objective comparison for some of these criteria. They tested the validity of eight different sets of criteria for schizoaffective disorder, including the CATEGO System (Wing et al., 1974), Kendell’s criteria (1970), Kasanin’s criteria (1933), Stephens’s criteria for “good prognosis schizophrenia” (Stephens et al., 1966), the study criteria of Welner, and Spitzer’s RDC for schizoaffective disorders (Spitzer et al., 1978). Based on information from blind interviews of 119 psychotic patients, who met at least one of the definitions, the researchers found a very low level of concordance between the eight different definitions of schizoaffective disorder compared with criteria for schizophrenia and affective disorder. One weakness of this study was the fact that it eliminated any criteria based upon longitudinal information. The authors concluded that it is highly unlikely that the diagnostic concept of schizoaffective disorder currently in use corresponds to anything coherent in nature. Later research by Brockington et al. (1980a–b) further undermined the concept of schizoaffective disorder. Their study of 32 patients meeting criteria for “schizomanic” psychosis made them conclude that, in terms of response to lithium treatment and overall outcome, most of their patients could be re-classified

8

Andreas Marneros

as manic. They subsequently analyzed family history, treatment response, and outcome for 76 patients who showed signs both of depression and of schizophrenia or paranoia. These patients presented a more intransigent problem. Many of them were ultimately re-diagnosed as schizophrenic or bipolar, but 20 eluded all attempts at reclassification. This might suggest a continuum model of psychiatric illness or the heterogeneity of schizoaffective disorder, but in any case does nothing to lessen confusion in the field. Some researchers have done away with the traditional categories altogether. Vaillant (1962) found six variables which yielded 82% accurate prognoses. These were: psychotic depressive heredity, symptoms suggesting a depressive psychosis, onset within six months before the fully developed illness, presence of precipitating factors, absence of schizoid personality before onset, and confusion or disorientation during the acute episode. In a 1964 study, he added a seventh variable, the fear of death. Individual case histories were rated on a scale of 0–7; a rating above 4 predicted good outcome and below 4 predicted bad outcome. This placing of cases on a continuum helps to offset the problem that, once a diagnostic scheme is in place, the labels associated with that diagnostic scheme can influence the observation of patients and the conceptualization of their illnesses in a biased manner. The continuum concept eliminated the need to create a new category of illness or a residual category in order to classify patients with schizoaffective features. However, diagnoses may then multiply without limit, and end up in conceptual chaos. The conceptual chaos has to be minimized also in regard to the relationship of schizoaffective psychoses to other “atypical” psychoses. Sometimes, the terms “cycloid psychosis,” “psychogenic” or “reactive psychosis”, and the French term “bouffée délirante” are used synonymously with schizoaffective psychosis (Marneros and Pillmann, 2005). However, there are very important differences between these psychoses – later called “acute and transient psychotic disorder” – and the psychopathologic picture of schizoaffective psychoses, so that they cannot be assumed to be identical (Perris, 1974; 1986; Pichot, 1986 and Strömgren, 1986; Marneros and Pillmann, 2005). One of the most voluminous and comprehensive studies of the schizoaffective area is the Cologne Study, carried out by Marneros, Deister and Rohde and published in a German monograph with an extensive English summary in 1991 (Marneros et al., 1991b). The Cologne Study is a naturalistic study comparing 402 patients with schizoaffective, affective, and schizophrenic disorders with an average illness duration of more than 25 years. The Cologne Study distinguishes between “episodes” (which is a cross-sectional diagnosis [Marneros et al., 1991b] – and “illnesses or disorders” [which is a longitudinal diagnosis]). The Cologne Study showed an intermediate position of schizoaffective disorders between schizophrenia and mood disorders on almost all investigated levels (premorbid, family, social,

9

The paradigma of overlapping spectra

personality, course and longitudinal outcome). It showed also that schizoaffective disorders have a very strong relation to mood disorders and they have to be distinguished into unipolar and bipolar forms (Marneros et al., 1990a–c). The ICD and DSM evolution: their strengths and weaknesses As Pichot (1986) pointed out, the official American history of the disorder “schizoaffective” can be followed in the successive editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM) of the American Psychiatric Association (APA). DSM-I (APA, 1952) describes the “schizoaffective type” among the “schizophrenic reactions.” The criteria used in DSM-I are different from Kasanin’s original description: no mention is made of sudden onset, shortness of episode, or complete recovery. DSM-II (APA, 1968) included the category “schizophrenia, schizoaffective type.” The definition, however, had become brief and non-committal: “Patients showing a mixture of schizophrenic symptoms and pronounced elation and depression.” The ICD-8, published in the same year, contained the same category. In 1978, the Task Force on Nomenclature and Classification of the APA published the draft of DSM-III. It included a special category, schizoaffective disorders, completely distinct from schizophrenic disorders. The criteria proposed as essential were “a depressive or manic syndrome . . . that preceded or develops concurrently with certain psychotic symptoms thought to be incompatible with a purely affective disorder”. The DMS-III draft stated, “The term schizoaffective has been used in many different ways. . . . At the present time there is a controversy as to whether this disorder represents a variant of Affective Disorder of Schizophrenia, a third independent nosological entity, or part of a continuum between pure Affective Disorder and pure Schizophrenia”. The separate listing is justified by “the accumulated evidence that individuals with a mixture of ‘affective’ and ‘schizophrenic’ symptoms, as compared with individuals diagnosed as having schizophrenia, have a better prognosis, a tendency towards acute onset and resolution, more likely recovery to premorbid level of functioning, and an absence of a prevalence increase of schizophrenia among family members . . .”. Two years later, in the final printed edition of DSM-III (APA, 1980), the category had practically disappeared. The manic episode and the major depressive episode now included cases with “mood-incongruent psychotic features” which, in the draft, would have belonged to the schizoaffective disorders. It is true that DSM-III has formally retained a category called schizoaffective disorders but, being without diagnostic criteria, it was considered as a residual class “for those instances in which the clinician is unable to make a differential diagnosis between Affective Disorders and either Schizophreniform Disorder or Schizophrenia.”

10

Andreas Marneros

A new category, schizophreniform disorder, appears. As Pichot (1986) pointed out, this category is very similar to Kasanin’s original schizoaffective psychosis as far as the evolution is concerned: “The duration . . . is less than six months . . . [there is] a tendency towards acute onset and resolution . . . recovery to premorbid levels of functioning,” but the symptomatic criteria are those of schizophrenia, with the exception of “a greater likelihood of emotional turmoil and confusion.” No mention of affective symptoms is made (Pichot, 1986). In DSM-III-R, published in 1987, schizoaffective disorders were born again – this time classified independently from both schizophrenia and affective disorders in the category “psychotic disorders not elsewhere classified” and with their own diagnostic criteria as well as with subtypes, namely bipolar type and depressive type. In DSM-IV, published in 1994, schizoaffective disorders belong to the category “other psychotic disorders” with almost the same diagnostic criteria and the same subtypes as in DSM-III-R. This time the mixed bipolar symptomatology is also recognized. ICD-9 (1976) continued the tradition of ICD-8 (1968). In ICD-10 (1991), after bouncing like a ping-pong ball during successive draft publications, schizoaffective disorders landed in a category of their own within schizophrenia and delusional disorders, with extensive description and with five subcategories: • Schizoaffective disorder, manic type • Schizoaffective disorder, depressive type • Schizoaffective disorder, mixed type • Other schizoaffective disorders • Schizoaffective disorders, unspecified The evaluation of the concept and the definition of schizoaffective disorders continue. Many aspects remain to be clarified, many questions still require answers. Most diagnostic systems only recognized the concurrent form of schizoaffective disorders, not the sequential ones. But operational research showed no differences in any investigated dimension between concurrent and sequential schizoaffective disorders (Marneros et al., 1988a–c; 1991b). We lost valuable time ignoring the sequential type of schizoaffective disorders. Nevertheless, the ongoing evolution of concepts and definitions of schizoaffective disorder also means continuing uncertainty. The question “What are the schizoaffective disorders?” remains unanswered. Schizoaffective disorders present in both DSM-IV and ICD-10 (APA, 1994; WHO, 1992) as unipolar or bipolar forms in a way similar to mood disorders (Marneros et al., 1989a–b, e; Marneros et al., 1990 a–c; Marneros et al., 2000). However, there are essential differences between DSM-IV and ICD-10 (Marneros and Goodwin, 2005b). While DSM-IV defines two subtypes based on longitudinal course, namely

11

The paradigma of overlapping spectra Table 1.1. Schizoaffective disorders according to ICD-10.

Schizoaffective disorders (F 25) G1. G2.

G3.

G4.

The disorder meets the criteria for one of the affective disorders (F30.–, F31.–, F32.–) of moderate or severe degree, as specified for each category. Symptoms from at least one of the groups listed below must be clearly present for most of the time during a period of at least two weeks (these groups are almost the same as for schizophrenia (F20.0–F20.3). Criteria G1 and G2 above must be met within the same episode of the disorder, and concurrently for at least part of the episode. Symptoms from both G1 and G2 must be prominent in the clinical picture. Most commonly used exclusion clause. The disorder is not attributable to organic mental disorder (in the sense of F00–F09) or to psychoactive substance-related intoxication, dependence, or withdrawal (F10–F19).

bipolar and depressive, the ICD-10 defines three types (i.e., manic, depressive and mixed) based on the most recent episode, rather than longitudinal course (WHO, 1992). These differences present a difficulty for cross-national research. Tables 1.1 and 1.2 illustrate how differently ICD-10 and DSM-IV handle the definition of schizoaffective disorder. While the main problem regarding the definition of ICD-10 concerns the longitudinal issue, the problem regarding DSM-IV concerns both – cross-sectional and longitudinal issues. The problem with the cross-sectional definition of DSM-IV concerns the time criteria for criterion B (during the period of illness there have been delusions or hallucinations for at least two weeks in the absence of prominent mood symptoms). Obviously, this is an attempt by the DSM-IV to separate schizoaffective disorders from psychotic mood disorders. The DSM-IV definition of mood disorders is broad, including even those with mood-incongruent symptoms (even first-rank schizophrenic symptoms) as the mood disorders. But the chronological criterion is an arbitrary one (two weeks of psychotic symptoms without mood disorders is schizoaffective; less than two weeks is psychotic mood disorder). One problem with this is that the beginning of a psychotic episode can be rarely assessed exactly. Every clinician knows that there is usually a gap of many days, weeks, or months between the beginning of a psychotic episode and admission to a hospital. Reconstruction of the psychopathological picture, retrospectively, is fraught with difficulty. Given the likelihood that the psychotic period would be underestimated, many patients who are really schizoaffective could be diagnosed as schizophrenic or as having psychotic mood disorder. Furthermore, the intensity of both concurrent syndromes can vary enormously

12

Andreas Marneros Table 1.2. Schizoaffective disorders according to DSM-IV.

Diagnostic criteria for 295.70 Schizoaffective Disorder (DSM-IV) A.

An uninterrupted period of illness during which, at some time, there is either a major Depressive Episode, a Manic Episode, or a Mixed Episode concurrent with symptoms that meet criterion A for Schizophrenia. Note: The Major Depressive Episode must include criterion A1: depressed mood. B. During the same period of illness, there have been delusions or hallucinations for at least two weeks in the absence of prominent mood symptoms. C. Symptoms that meet criteria for a mood episode are present for a substantial portion of the total duration of the active and residual periods of the illness. D. The disturbance is not owing to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition. Specify type: Bipolar Type: If the disturbance includes a Manic or a Mixed Episode (or a Manic or a Mixed Episode and Major Depressive Episodes). Depressive Type: If the disturbance only includes Major Depressive Episodes.

during an episode – it seems arbitrary to give chronological priority to the psychotic symptoms over the mood component. It is curious that DSM-IV rejected Jasper’s hierarchical diagnostic principle, which suggested a diagnostic superiority of schizophrenic symptoms over affective symptoms, but, regarding the chronological criterion of the schizoaffective definition, obviously made an exception! Considering what is now known about schizoaffective disorders, we repeatedly suggested that the definition of schizoaffective disorders should contain two components: one cross-sectional and one longitudinal (Marneros and Goodwin, 2005b). The cross-sectional definition should be the definition of an episode; while the longitudinal definition should be that of a disease or disorder. The cross-sectional definition of a schizoaffective episode should be based on the simultaneous occurrence of symptoms of a schizophrenic and a mood episode, independent of the chronological manifestation. Thus, we agree with the definition of ICD-10, which yields three types of schizoaffective episodes: schizodepressive, schizomanic and mixed. The longitudinal definition of the “schizoaffective disorder” should consider the sequential occurrence of mood and schizophrenic episodes during course. The longitudinal research demonstrates that the course of schizoaffective disorders can be very unstable because schizoaffective episodes, pure mood episodes, and pure schizophrenic episodes can each occur at different points in the patient’s longitudinal course (see Figure 1.1).

13

The paradigma of overlapping spectra

0

Figure 1.1.

2

4

6

8

10

12

14

16

18

20

22

24

26

manic

schizomanic

mixed schizoaffective

depressive

schizodepressive

schizophrenic

28

30

32

34

36

others

Halle Bipolarity Longitudinal Study (HABILOS). Longitudinal course of patients with mixed schizo-manic-depressive episodes.

In a recently published paper (Marneros and Goodwin, 2005b) we posed the question: what are such disorders when viewed longitudinally? Are they to be considered mood disorders because of the pure mood episodes, or schizophrenic disorders because of some pure schizophrenic episodes, or schizoaffective disorders because of some schizoaffective episodes? Relevant to this question is the finding that there are no differences between patients who have only had schizoaffective episodes, and those in whom schizoaffective episodes occur along with pure mood and schizophrenic episodes. There are therefore no differences between the “concurrent” and the “sequential” type of schizoaffective disorder (Marneros et al., 1986a; 1991a–b). Patients, who change from pure mood episodes to a pure schizophrenic episode and vice versa, do not differ from patients having schizoaffective episodes. In this sense, we suggest a longitudinal definition of schizoaffective disorders, including a concurrent and a sequential type (Marneros et al., 1986–2004): the “concurrent type” is characterized by the coincidence of schizophrenic and affective episodes, while the “sequential type” is characterized by the longitudinal change from schizophrenic to affective episodes and vice versa (Marneros et al., 1986a, 1988c, 1989a, 1991b, 2000; Marneros and Angst, 2000). How essential it is to make a longitudinal diagnosis is illustrated by the Halle Bipolarity Longitudinal Study (HABILOS) (Figure 1.1), in which the investigators tried to allocate disorders with manic symptomatology to “pure mood disorders” or “schizoaffective disorders,” according to DSM-IV, ICD-10, and to an “empirical

14

Andreas Marneros

Bipolar patients (n=277) Bipolar schizoaffective patientsa (n =23)

36.1% 8.3%

55.6%

a b

Figure 1.2.

Bipolar affective patients (n =100)

Bipolar unclassifiable patientsb (n =154)

Only schizoaffective episodes. Because of occurence of schizophrenic and affective episodes and/or schizoaffective episodes.

Halle Bipolarity Longitudinal Study (HABILOS). ICD-10 longitudinal classification of bipolar patients.

Bipolar patients (n =277)

36.1%

Bipolar schizoaffective patients (n =177)

Bipolar affective patients (n =100)

63.9%

Schizoaffective = occurence of schizoaffective episodes or sequentially schizophrenic and affective episodes.

Figure 1.3.

Halle Bipolarity Longitudinal Study (HABILOS). Emipirical longitudinal classification of bipolar patients.

definition” by the Marneros group (as defined above). Applying the ICD-10 definition, the findings illustrated in Figure 1.2 were produced. As shown, only 8.3% of the 277 patients could be allocated longitudinally as schizoaffective bipolar and 36.1% as affective bipolar; whereas the majority of patients (55.6%) could not be allocated longitudinally because of the occurrence of different types of episodes (schizophrenic, schizoaffective, affective) at different times. However, if we use the empirical definition with its cross-sectional and sequential aspect, all patients can be allocated: 36.1%, as in the ICD-10 categorization, could be allocated as bipolar mood disorder, and 63.9% could be allocated as schizoaffective disorder (see Figure 1.3).

15

The paradigma of overlapping spectra

One-day-fly or a diachronic continuity? Perhaps a polymorphism? It is evident – the old question still remains up-to-date (in spite of the efforts described above): what kind of illness should be diagnosed in a patient having a schizophrenic episode at first, but several months later a depressive episode, or a manic episode followed by a schizodepressive or a schizomanic episode? The polymorphism (change between different kinds of episodes during the longterm course) is not an exception, but very common (Marneros et al., 1991b; 1995a). There are some diagnoses which are more stable than others (i.e., schizophrenia) and others that are very unstable (mixed states, schizomanic episodes). But also the “atypical area,” which includes the schizophreniform psychoses and brief psychoses (or acute and transient psychotic disorders), is characterized by a polymorphism (Marneros and Pillmann, 2005). It seems that the polymorphism is the rule in the intermediate area, i.e., in the area of the overlapping of the spectra. The longitudinal investigation of monomorphous (having only the same type of episode during a long period of time) and polymorphous (having various types of episodes) schizoaffective disorders over a period of 25 years in the Cologne Study showed that the majority of schizoaffective disorders (70%) were polymorphous (Marneros et al., 1988a–c; 1989a–c; 1991b), i.e., more than one type of episode occurred during the course: schizophrenic, schizodepressive, schizomanic, manic, depressive and mixed. Less than one-third of the patients (30%) had a monomorphous course, with only one type of episode during the whole course. However, it has to be considered that a special group within the polymorphous schizoaffective disorders is formed by patients having a change from depressive (or schizodepressive) into manic (or schizomanic) or mixed symptomatology and vice versa: the bipolar schizoaffective disorders. Independent of the type of the initial episode, all other types of schizophrenic, schizodepressive, schizomanic, schizoaffective mixed and all types of affective episodes may occur. In the majority of polymorphous schizoaffective disorders the first change occurred already in the second episode (63%). In patients with polymorphous schizoaffective disorders, the first change occurred on average 7.5 years after first manifestation; in 72% of the cases within the first five years. Comparison between monomorphous and polymorphous schizoaffective disorders showed no significant differences in relevant sociodemographic premorbid features or features regarding course, patterns and outcome (Marneros et al., 1988a–c; 1989a–c; 1991b). Polymorphism, continuum and overlap: the existent reality Recently, discussing the overlap of the spectra after investigations concerning the brief polymorphic psychoses (Marneros and Pillmann, 2005), we drew the following conclusions, which we repeat here:

16

Andreas Marneros

Diagnosis is a central task in psychiatry – clinically, determining the approach to treatment and the prognosis; and in research, identifying the population of interest. However, psychiatric diagnoses are based solely on clinical grounds with no external validating parameters. Furthermore, there are several determinants driving current diagnostic practice that may not be helpful for the needs of genetic and biological research (Duffy and Grof, 2001). Disease entities have to be determined by identical symptoms, course, etiology and pathomorphology. The research process is meant to follow a constant approximation of this idea. Hence, outlining perspectives from a mere psychopathological point of view is a difficult task, since psychopathological validation of nosological concepts usually implies external validation with course, outcome, biological parameters, genetics and personality, and will probably result in outlining a type of mental disorganization, perhaps inherent with its structure, rather than producing stringent mental concomitants of particular biological disturbances. Conventions rather than objective findings mark the borders, whereas only the extremes are very discriminating. It is therefore not surprising that the debate concerning the controversy of psychotic continuum versus separate entity is as old as scientific psychiatry itself (Berrios and Beer, 1994). A clear-cut dichotomy between schizophrenia and mood disorders – the two biggest and classical pillars of the so-called idiopathic or endogenous psychoses – does not exist. This is not new, but quite old knowledge, as Craddock and Owen in “The beginning of the end for the Kraepelinian dichotomy” (2005) suggested. Even Kraepelin, the creator of the dichotomy of the psychoses, noted it in 1920. Yet, there are no specific findings correlated only with schizophrenia or mood disorders in all possible domains (Carpenter et al., 1995), such as psychopathology (Mundt, 1995), prognosis (Marneros et al., 1995b), genetics (Tsuang, 1995), psychopharmacology (Meltzer, 1995), neuroimaging (Andreasen and Flaum, 1995), biochemistry (van Kammen, 1995), histology (Benes, 1995) or other domains (Angst, 1995; Crow, 1995). The contributions in this book confirm the conclusions mentioned above. Perhaps this is not the result of our possibly insufficient methods of research, but the very nature of mental disorders. Obviously the recognition of mental disorders not belonging to the core of schizophrenia or mood disorders is not only a function of differing phenomenology, but also of many other factors like prognosis, gender distribution, family, and genetics. The schizoaffective disorders have the most prominent position within the group of non-typical non-core schizophrenia. The reason is that schizoaffective disorders challenge the dichotomy concept of schizophrenia versus mood disorders very radically through the presence of the characteristics of both groups, which theoretically have to be mutually excluded. Together with the fact that they are not rare, this is one of the possible explanations for why the nosological position of schizoaffective disorders has been intensely and

17

The paradigma of overlapping spectra

controversially discussed (Maj, 1984, see also contributions in Marneros and Tsuang, 1986; Marneros and Tsuang, 1990; Marneros et al., 1995a; Marneros and Goodwin, 2005a). But, as the World Health Organization complains, the Brief Polymorphic Psychoses have only rarely been a focus of research interest. The reasons for this are perhaps that they are less frequent than schizoaffective disorders and that the contrast in the symptomatology is not very obvious. Their psychotic symptomatology and good prognosis led to labels like “good prognosis schizophrenia” or “remitting schizophrenia,” which is a very superficial solution. But also putting them, as ICD10 and DSM-IV did, into a “schizophrenic spectrum,” is not rational. For schizoaffective psychoses, it is evident that their relationship to mood disorders is much closer than to schizophrenia. Nevertheless, they share features and findings of both disorders: schizophrenia and mood disorders. In this sense, it has been assumed that schizoaffective disorders belong to a psychotic continuum extending between schizophrenic and mood disorders (Marneros et al., 1995a–b). But for the Brief Polymorphic Psychosis, no similar discussion or proposal exists. In contrast, the creators and main presenters of a concept of non-schizophrenic Brief Polymorphic Psychosis, like Wernicke, Kleist and Leonhard, as well as their epigones, absolutely confirmed in an almost religious dogmatism that this group of psychoses is an independent nosological entity. However, the early research on the topic, unfortunately yet rare and somewhat unsystematic, and especially the Halle Study on Brief and Acute Psychoses (HASBAP), which is the most voluminous and most systematic comparative study on the topic, cannot confirm the assumption of an independent nosological entity (Marneros and Pillmann, 2005). Rather, the overlaps between Brief Polymorphic Psychosis and schizophrenia on the one hand, and Brief Polymorphic Disorders, schizoaffective and mood disorders (especially their mixed forms) on the other hand, are considerable. It seems that Brief Polymorphic Psychosis cannot be assumed to be a sub-group of schizophrenia. But neither are they a group of a so-called schizophrenic spectrum, whatever this means. There is also no reliable definition of the elastic and changeable schizophrenic spectrum (Andreasen and Flaum, 1995), nor is there a reliable definition of the elastic and changeable “affective spectrum” (Akiskal and Pinto, 2000). If we accept continuity between the various psychotic groups – but not only between them – then the position of the Brief Polymorphic Psychosis is obviously a position within the spectrum of the psychotic continuum. The research findings to date reject the idea that schizoaffective disorders and Brief Polymorphic Disorders are identical, which means they do not comprise a single bridge between schizophrenia and affective disorders. Instead, they connect not only the two classical mental disorders, but also the connections and relations with one another. There is, namely, a continuum within the continuum.

18

Andreas Marneros

The following recommendations and remarks of Duffy and Grof (2001) regarding bipolar disorders could be valid with some modifications to all kinds of mental disorders: Our psychiatric forefathers, such as Emil Kraepelin and Eugen Bleuler [but in our opinion also Carl Wernicke, Karl Kleist, Karl Leonhard or Valentin Magnan] were clinicians developing psychiatric diagnoses for the purpose of clinical description and clinical communication. When psychiatric research strengthened in the 1960s, operationalized diagnostic criteria for research were developed. But when the research criteria were “converted” by a group of committees into the DSM and ICD, we were given a system that tried to serve everyone: clinicians, psychiatric statisticians, administrators, insurance companies and others. As pointed out by Angst and Ernst (1993), we need to collect research data and develop diagnostic strategies independent of current diagnostic trends so that the value of this data will remain intact while the diagnostic fashion changes. The task of defining a phenotype closely tied to the underlying genes is further complicated by the existence of phenocopies (non-genetically based forms of the illness) and variable presentations of the disease genotype within and between affected individuals, possibly reflecting epistasis. Therefore, genetic studies require a conservative phenotype definition to be rigorously applied in order to identify a homogeneous sub-group of patients who likely share the same susceptibility genes. It is important to realize the limitations of our current diagnostic approach. We can hardly expect to find genes for a disorder unless the phenotypic description is reasonably valid, reliable and stable. Unfortunately, our profession has developed a habit of using the same diagnostic strategy towards very different ends. Research, treatment, statistics for public health tasks and disability pensions are all very different activities. What fits one will not well suit another. For example, while public health statistics seek to categorize every case, research usually requires exclusivity via selection/exclusion criteria in order to achieve homogeneity. (Duffy and Grof, 2001).

The new developments in science in general and psychiatry in particular, for instance molecular genetic studies, will not “overturn the traditional dichotomous view” – as Craddock and Owen (2005) “predict” – but confirm the long clinical experience and clinical studies indicating that the dichotomy is artificial and the continuum natural.

R E F E R E N C ES Akiskal, H. S. and Pinto, O. (2000). The soft bipolar spectrum: footnotes to Kraepelin on the interface of hypomania, temperament and depression. In Bipolar Disorders. 100 Years After Manic-depressive Insanity, ed. A. Marneros and J. Angst. Dordrecht, Boston, London: Kluwer, pp. 37–62. American Psychiatric Association (1952). DSM-I: Diagnostic and Statistical Manual of Mental Disorders. Washington: APA Mental Hospital Service. American Psychiatric Association (1968). DSM-II: Diagnostic and Statistical Manual of Mental Disorders (2nd edn). Washington: APA.

19

The paradigma of overlapping spectra American Psychiatric Association (1978). DSM-III (draft): Diagnostic and Statistical Manual of Mental Disorders (draft of 3rd edn). Washington: APA. American Psychiatric Association (1980). DSM-III: Diagnostic and Statistical Manual of Mental Disorders (3rd edn). Washington: APA. American Psychiatric Association (1987). DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders (3rd edn, revision).Washington: APA. American Psychiatric Association (1994). DSM-IV: Diagnostic and Statistical Manual of Mental Disorders (4th edn). Washington: APA. Andreasen, N. C. and Flaum, M. (1995). The schizophrenia spectrum: perspectives from neuroimaging. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 91–106. Angst, J. (1966). Zur Ätiologie und Nosologie endogener depressiver Psychosen. Eine genetische, soziologische und klinische Studie. Berlin, Heidelberg, New York: Springer. Angst, J. (1986). The course of schizoaffective disorders. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, p. 63. Angst, J. (1989). Der Verlauf schizoaffektiver Psychosen. In Schizoaffektive Psychosen, Diagnose, Therapie und Prophylaxe, ed. A. Marneros. Berlin, Heidelberg, New York: Springer. Angst, J. (1990). Recurrent brief depression: a new concept of depression. Pharmacopsychiatry, 23, 63–6. Angst, J. (1995). Psychotic continuum or distinct entities: discussion. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 87–8. Angst, J. and Ernst, C. (1993). Current concepts of the classification of affective disorders. International Clinical Psychopharmacology, 8, 211–15. Angst, J., Felder, W. and Lohmeyer, B. (1979). Schizoaffective disorders: I. Results of a genetic investigation. Journal of Affective Disorders, 1, 139–53. Angst, J. and Marneros, A. (2001). Bipolarity from ancient to modern times: conception, birth and rebirth. Journal of Affective Disorders, 67, 3–19. Astrup, C. and Noreik, K. (1966). Functional Psychoses: Diagnostic and Prognostic Models. Springfield, IL: Charles C. Thomas. Benes, F. M. (1995). Microscopic findings in the cortex and hippocampus of schizophrenic and schizoaffective patients. In Psychotic Continuum, eds A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 127–36. Berrios, G. E. and Beer, D. (1994): The notion of a unitary psychosis: a conceptual history. History of Psychiatry, 5, 13–36. Bleuler, E. (1911). Dementia praecox oder Gruppe der Schizophrenien. In: Handbuch der Psychiatrie, Spezieller Teil 4, ed. G. Aschaffenburg. Leipzig: Deuticke. Bleuler, E. (1924). Textbook of Psychiatry. Translation of 4th German edn by A. A. Vrill. New York: Macmillan. Brockington, I. F., Kendell, R. E. and Wainwright, S. (1980a). Depressed patients with schizophrenic or paranoid symptoms. Psychological Medicine, 10, 665–75. Brockington, I. F. and Leff, J. R. (1979). Schizoaffective psychosis: definitions and incidence. Psychological Medicine, 9, 91–9.

20

Andreas Marneros Brockington, I. F., Wainwright, S. and Kendell, R. E. (1980b). Manic patients with schizophrenic or paranoid symptoms. Psychological Medicine, 10, 73–83. Cadoret R. J., Fowler, R. C., McCabe, M. S. and Winokur, G. (1974). Evidence for heterogeneity in a group of good-prognosis schizophrenics. Comprehensive Psychiatry, 15, 443–50. Carpenter, W. T., Buchanan, R. W. and Kirkpatrick, B. (1995). Schizophrenia: disease entity, disease entities, or domains of psychopathology. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 137–49. Clayton, P. J., Rodin, L. and Winokur, G. (1968). Family history studies: III – Schizoaffective disorder, clinical and genetic factors – including a one to two year follow-up. Comprehensive Psychiatry, 9, 31–49. Craddock, N. and Owen, M. J. (2005). The beginning of the end for the Kraepelinian dichotomy. The British Journal of Psychiatry, 186, 364–6. Crow, T. J. (1995). Psychotic continuum or disease entities? The critical impact of nosology on the problem of aetiology. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 151–63. Duffy, A. and Grof, P. (2001). Psychiatric diagnoses in the context of genetic studies of bipolar disorder. Bipolar Disorders, 3, 270–5. Fowler, R. C., McCabe, M. S., Cadoret, R. J. and Winokur, G. (1972). The validity of good prognosis schizophrenia. Archives of General Psychiatry, 26, 182–5. Griesinger, W. (1845). Die Pathologie und Therapie der psychischen Krankheiten für Ärzte und Studierende. Stuttgart: Krabbe. Heinroth, J. C. A. (1818). Lehrbuch der Störungen des Seelenlebens oder der Seelenstörung und ihrer Behandlung – aus rationaler Sicht. Leipzig: Vogel. Himmelhoch, J. A., Fuchs, C. Z., May, S. J., Symons, B. J., Neil, J. F. (1981). When a schizoaffective diagnosis has meaning. The Journal of Nervous and Mental Disease, 169, 277–82. Jaspers, K. (1913–1973). Allgemeine Psychopathologie, 1st to 9th edn. Berlin, Heidelberg, NewYork: Springer. Kahlbaum, K. (1863). Die Gruppirung der psychischen Krankheiten und die Eintheilung der Seelenstörungen. Danzig: Kafemann. Kant, O. (1940). Types and analyses of the clinical pictures of recovered schizophrenics. The Psychiatric Quarterly, 14, 676–700. Kasanin, J. (1933). The acute schizoaffective psychoses. American Journal of Psychiatry, 13, 97–126. Kendell, R. E. and Gourlay, J. (1970). The clinical distinction between affective psychoses and schizophrenia. British Journal of Psychiatry, 117, 261–6. Kleist, K. (1928). Über zykloide, paranoide und epileptoide Psychosen und über die Frage der Degenerationspsychosen. Schweizer Archiv fur Neurologie und Psychiatrie, 23, 3–37. Kraepelin, E. (1893). Psychiatrie: ein kurzes Lehrbuch für Studirende und Ärzte. 4th edn. Leipzig: J. A. Barth. Kraepelin, E. (1896). Psychiatrie: ein kurzes Lehrbuch für Studirende und Ärzte. 5th edn. Leipzig: J. A. Barth. Kraepelin, E. (1899). Psychiatrie: ein kurzes Lehrbuch für Studirende und Ärzte. 6th edn. Leipzig: J. A. Barth.

21

The paradigma of overlapping spectra Kraepelin, E. (1920). Die Erscheinungsformen des Irreseins. Zeitschrift für die gesamte Neurologie und Psychiatrie, 62, 1–29. Langfeldt, G. (1939). The Schizophreniform State. Copenhagen: Munksgaard. Leonhard, K. (1957). Aufteilung der endogenen Psychosen. Jena: Akademie. Longrigg, J. (1993). Greek Rational Medicine. Philosophy and Medicine from Alcmaeon to the Alexandrians. London, New York: Routledge. Maj, M. (1984). The evolution of some European diagnostic concepts relevant to the category of schizoaffective psychoses. Psychopathology, 17, 158–67. Marneros, A. and Angst, J. (2000). Bipolar Disorders. 100 Years After Manic Depressive Insanity. Dordrecht, Boston, London: Kluwer Academic Publishers. Marneros, A., Andreasen, N. C. and Tsuang, M. T. (1991a). Negative Versus Positive Schizophrenia. Berlin, Heidelberg, New York: Springer. Marneros, A., Andreasen, N. C. and Tsuang, M. T. (1995a). Psychotic Continuum. Berlin, Heidelberg, New York: Springer. Marneros, A., Deister, A. and Rohde, A. (1986a). The Cologne study on schizoaffective disorders and schizophrenia suspecta. In Marneros, A. and Tsuang, M. T. (eds), Schizoaffective Psychoses. Berlin, Heidelberg, New York: Springer, pp. 123–42. Marneros, A., Deister, A. and Rohde, A. (1989a). Unipolar and bipolar schizoaffective disorders: A comparative study. I. Premorbid and sociodemographic features. European Archives of Psychiatry and Clinical Neuroscience, 239, 158–63. Marneros, A., Deister, A. and Rohde, A. (1990a). The concept of distinct but voluminous bipolar and unipolar diseases. Part I: The bipolar diseases. European Archives of Psychiatry and Clinical Neuroscience, 240, 77–84. Marneros, A., Deister, A. and Rohde, A. (1990b). The concept of distinct but voluminous bipolar and unipolar diseases. Part III: Unipolar and bipolar comparison. European Archives of Psychiatry and Clinical Neuroscience, 240, 90–5. Marneros, A., Deister, A. and Rohde, A. (1991b). Affektive, schizoaffektive und schizophrene Psychosen. Eine vergleichende Langzeitstudie. Berlin, Heidelberg, New York: Springer. Marneros, A., Deister, A. and Rohde, A. (2000). Bipolar Schizoaffective Disorders. In Bipolar Disorders. 100 Years After Manic Depressive Insanity, ed. A. Marneros, J. Angst, pp. 111–26. Dordrecht, Boston, London: Kluwer Academic Publishers. Marneros, A., Deister, A., Rohde, A., and Jünemann, H. (1989b). Unipolar and bipolar schizoaffective disorders: A comparative study. III. Long-term outcome. European Archives of Psychiatry and Clinical Neuroscience, 239, 171–6. Marneros, A., Deister, A., Rohde, A., Jünemann, H. and Fimmers, R. (1988a). Long-term course of schizoaffective disorders. Part I: Defintions, methods, frequency of episodes and cycles. European Archives of Psychiatry and Neurological Sciences, 237, 264–75. Marneros, A., Deister, A., Rohde, A., Jünemann, H. and Fimmers, R. (1988b). Long-term course of schizoaffective disorders. Part II: Length of cycles, episodes and intervals. European Archives of Psychiatry and Neurological Sciences, 237, 276–82. Marneros, A., Deister, A., Rohde, A., Jünemann, H. and Fimmers, R. (1988c). Long-term course of schizoaffective disorders. Part III: Onset, type of episodes and syndrome shift,

22

Andreas Marneros precipitating factors, suicidality, seasonality, inactivity of illness, and outcome. European Archives of Psychiatry and Neurological Sciences, 237, 283–90. Marneros, A., Deister, A., Rohde, A., Steinmeyer E. M. and Jünemann, H. (1989c). Long-term outcome of schizoaffective and schizophrenic disorders: a comparative study. Part I: Definitions, methods, psychopathological and social outcome. European Archives of Psychiatry and Neurological Sciences, 238, 118–25. Marneros, A. and Goodwin, F. K. (eds.) (2005a): Bipolar Disorders. Mixed States, Rapid Cycling and Atypical Forms. Cambridge: Cambridge University Press. Marneros, A. and Goodwin, F. K. (2005b). Bipolar disorders beyond major depression and euphoric mania. In Bipolar Disorders. Mixed States, Rapid Cycling and Atypical Forms, ed. A. Marneros and F. K. Goodwin. Cambridge: Cambridge University Press. Marneros, A. and Pillmann, F. (2005). Das Wort Psychiatrie wurde in Halle geboren. Stuttgart: Schattauer. Marneros, A., Rohde, A. and Deister, A. (1989d). Unipolar and bipolar schizoaffective disorders: a comparative study. II. Long-term course. European Archives of Psychiatry and Clinical Neuroscience, 239, 164–70. Marneros, A., Rohde, A. and Deister, A. (1990c). The concept of distinct but voluminous bipolar and unipolar diseases. Part II: The unipolar diseases. European Archives of Psychiatry and Clinical Neuroscience, 240, 85–9. Marneros, A., Rohde, A. and Deister, A. (1995b): Psychotic continuum under longitudinal considerations. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang, M. T. Berlin, Heidelberg, New York: Springer, pp. 17–30. Marneros, A., Rohde, A., Deister, A. and Risse, A. (1986b). Schizoaffective psychoses: the prognostic value of the affective component. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 155–63. Marneros, A., Rohde, A., Deister, A. and Risse, A. (1986c). Features of schizoaffective disorders: the “cases-in-between”. In Marneros, A. and Tsuang, M. T. (eds.). Schizoaffective Psychoses. Berlin, Heidelberg, New York: Springer, pp. 143–54. Marneros, A., Steinmeyer E. M., Deister, A., Rohde, A. and Jünemann, H. (1989e). Longterm outcome of schizoaffective and schizophrenic disorders: a comparative study. Part III: Social consequences. European Archives of Psychiatry and Neurological Sciences, 238, 135–9. Marneros, A. and Tsuang, M. T. (eds.) (1986). Schizoaffective Psychoses. Berlin: Springer. Marneros, A. and Tsuang, M. T. (1990). Affective and Schizoaffective Disorders. Similarities and Differences. Berlin, Heidelberg, New York: Springer. Meltzer, H. Y. (1995). Psychotic continuum or disease entities: perspective from psychopharmacology. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 31–55. Mendlewicz, J., Linkowski, P. and Wilmotte, J. (1980). Relationship between schizoaffective illness and affective disorders or schizophrenia: Morbidity risk and genetic transmission. Journal of Affective Disorders, 2, 289–302. Mundt, C. (1995): Psychotic continuum or distinct entities: perspectives from psychopathology.

23

The paradigma of overlapping spectra In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 7–15. Perris, C. (1966). A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatrica Scandinavica, (Suppl.), 194, 1–89. Perris, C. (1974). A study of cycloid psychoses. Acta Psychiatrica Sandinavica, (Suppl.), 253, 1–77. Perris, C. (1986). The case for the independence of cycloid psychotic disorder from the schizoaffective disorders. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, p. 272. Pichot, P. (1986). A comparison of different national concepts of schizoaffective psychosis. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, p. 8. Schneider, K. (1959). Clinical Psychopathology. New York: Grune and Stratton. Schneider, K. (1973). Klinische Psychopathologie, 10th edn. Stuttgart: Thieme. Spitzer, R. L., Endicott, J. and Robins, E. (1978). Research diagnostic criteria: rationale and reliability. Archives of General Psychiatry, 35, 773–82. Stephens, J. H., Astrup, C. and Mangrum, J. C. (1966). Prognostic factors in recovered and deteriorated schizophrenics. American Journal of Psychiatry, 122, 1116–21. Strömgren, E. (1986). Reactive (psychogenic) psychoses and their relations to schizoaffective psychoses. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, p. 260. Tsuang, M. T. (1995). Psychotic continuum: perspectives from family studies. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 57–66. Tsuang, M. T., Dempsey, G. M. and Rausher, F. (1976). A study of ‘atypical schizophrenia’: comparison with schizophrenia and affective disorder by sex, age of admission, precipitant, outcome, and family history. Archives of General Psychiatry, 33, 1157–60. Tsuang, M. T., Simpson, J. C. and Fleming, J. A. (1986). Diagnostic criteria for subtyping schizoaffective disorder. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 50–62. Vaillant, G. E. (1962). The prediction of recovery in schizophrenia. The Journal of Nervous and Mental Disease, 135, 534–43. van Kammen, D. P. (1995). Biochemical heterogeneity in schizophrenia: implications and research strategies of the state dependency model. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 107–26. Welner, A., Croughan, J., Fishman, R., and Robins, E. (1977). The group of schizoaffective and related psychoses: A follow-up study. Comprehensive Psychiatry, 18, 413–22. Welner, A., Welner, Z. and Fishman, R. (1979). The group of schizoaffective and related psychoses: IV. A family study. Comprehensive Psychiatry, 20, 21–5. Wernicke, C. (1900). Grundriss der Psychiatrie. Leipzig: Thieme. Wing, J. K., Cooper, J. E. and Sartorius, N. (1974). The Measurement and Classification of Psychiatric Symptoms. London: Cambridge University Press. World Health Organization (1968). ICD-8: International Statistical Classification of Diseases and Related Health Problems (8th edn). Geneva: WHO.

24

Andreas Marneros World Health Organization (1976). ICD-9: International Statistical Classification of Diseases and Related Health Problems (9th edn). Geneva: WHO. World Health Organization (1992). ICD-10: International Statistical Classification of Diseases and Related Health Problems (10th edn). Geneva: WHO. Zendig, E. (1909). Beiträge zur Differentialdiagnostik des manisch-depressiven Irreseins und der Dementia praecox. Allgemeine Zeitschrift für Psychiatrie, 6, 47–9.

2

The overlapping of the spectra: overlapping genes and genetic models John R. Kelsoe Department of Psychiatry, University of California, San Diego

Introduction The overlapping spectra of clinical presentation in bipolar disorder and schizophrenia described elsewhere in this volume suggest overlapping etiology and a spectrum of pathophysiological mechanisms. Genetic studies provide a powerful route to dissecting such pathophysiological mechanisms and elucidating the etiological relationship between these syndromes. Many genetic traits display phenotypes with a similar spectrum presentation, better described as a quantitative genetic trait than a Mendelian one. Yet both bipolar disorder and schizophrenia have characteristics of both quantitative and dichotomous genetic traits, as does the relationship between the disorders. Various studies of families and twins have been conducted in order to understand the genetic relationships between these psychotic illnesses and the most appropriate genetic model. More recently, these questions have been informed by molecular genetic methods to identify the specific susceptibility genes. Together these data suggest that the complexity of the overlapping phenotype is a reflection of complex genetic underpinnings. In this chapter, these possible models and family studies will be reviewed, as well as the more recent molecular genetic studies. Possible models and genetic mechanisms responsible for these spectra will be discussed. Genetic models Single major locus models

The genetic models to be considered fall into two major categories. In the Mendelian or major locus model, one gene of large effect is primarily responsible for the genetic susceptibility to illness in each affected individual. Such a The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

25

26

John R. Kelsoe

susceptibility gene may be transmitted in a dominant fashion in which only one susceptibility allele is necessary for disease, or a recessive fashion, where an individual must have two copies of the susceptibility allele in order to be vulnerable. If the susceptibility gene is on the X chromosome, then an X-linked form of transmission may be observed in which the disease is seen primarily in men or women depending on whether it is recessive or dominant respectively. These classic patterns of transmission can be modified by various factors that affect the penetrance of the gene. The penetrance is the probability that the phenotype will be manifest given that an individual inherits the susceptibility allele. The penetrance for both bipolar disorder and schizophrenia is between 40% and 60%. Hence, the inherited risk alleles are susceptibility factors and other non-heritable factors are required for expression of the trait. As the age of onset for each of these traits is in the early twenties, penetrance is usually modeled as increasing with age until a maximum is reached by about 40 years of age. In addition to reduced penetrance, other factors may modify the expression of a major locus so as to present as a spectrum phenotype. This is termed variable expressivity in that the same gene and susceptibility allele may have a variety of phenotypes. Numerous genetic traits may display such variable expressivity. Neurofibromatosis may be manifest as a café-au-lait spot on fundoscopic exam or as multiple skin tumors. This wide range of presentation is all the result of one gene (Viskochil, 2002). Many genetic mechanisms can produce such variability in expression. Different mutations in the disease gene can produce dramatically different phenotypic effects. Alternatively, though the primary susceptibility to disease is transmitted through one gene, other genes may operate as modifiers of its effect. Certain environmental factors may be required for a disease to be manifest, or may modify the expression of the susceptibility gene. It is therefore possible, though unlikely, that the wide range of phenotypic expression observed in psychosis could be owing to one gene. The major locus model may be further complicated by heterogeneity. Though one gene may be responsible for the genetic vulnerability in one family, other genes may operate in other families. Under such a model, there could be distinct bipolar genes and schizophrenia genes. The observed overlap in presentation could result from modifying factors just described. Alternatively, a group of genes may cause a unitary risk for psychosis, which in turn is variable in its presentation by such modifying factors. Quantitative trait polygenic models

The alternate model in classical genetics is the quantitative trait polygenic model. In this model, many genes are involved in the susceptibility to illness, each contributing a small amount to an individual’s risk. An individual’s vulnerability is the

27

Overlapping genes and genetic models

result of the combined effect of multiple inherited susceptibility alleles. The effect of susceptibility alleles may simply add together to produce a cumulative risk. Alternatively, the genes may interact with each other such that the combined effect is greater than the sum, termed an epistatic or multiplicative interaction. The resulting phenotype is typically quantitative in nature. This might reflect a varying severity of psychosis or mood lability. Such a phenotype is consistent with a spectrum of severity, but does not capture the qualitative differences in the bipolar or schizophrenic phenotypes. The multifactorial threshold model attempts to better model the qualitative variation in phenotype. In this model, the quantitative trait is a latent unitary vulnerability to psychosis. Different forms of illness result with increasing levels of vulnerability at different thresholds. This model requires assumptions of a quantitative relationship between different affective and psychotic disorders. Typically, the following relationship is assumed: major depression < bipolar II < bipolar I < schizoaffective disorder < schizophrenia. Though to some extent this ordering may reflect relative degrees of severity, it does not adequately capture the qualitative differences in these disorders. It also predicts a corresponding ordering of relative population prevalences which is not consistent with observed rates. Specifically, schizophrenia and schizoaffective disorder are more common in relationship with bipolar disorder than this model would predict. Genetic epidemiology Numerous epidemiological studies have been conducted to examine the heritability and mode of genetic transmission of various psychiatric disorders. However, only a relatively few studies have examined the question of overlap between schizophrenia and bipolar disorder. A selection of these studies with different designs is reviewed here. Family studies have examined the question of whether these illnesses “breed true.” Is bipolar disorder observed at an elevated rate in the families of probands with schizophrenia, and conversely is schizophrenia observed at an elevated rate in the families of probands with bipolar disorder? Figure 2.1 illustrates the results from one such study (Tsuang et al., 1980). Tsuang and colleagues employed direct interview and family history methods to estimate the frequency of mood and psychotic illness in controls and the families of probands with schizophrenia, mania, and depression. The risk of both bipolar disorder and schizophrenia was elevated in families of probands with schizophrenia, though the risk for bipolar disorder was less than that for schizophrenia. Similarly, the risk for bipolar disorder and schizophrenia was elevated in the relatives of probands with mania, though the risk for bipolar disorder was greater than that for schizophrenia. This indicates a

28

John R. Kelsoe

16

**

Morbid risk (%)

14 12 10 Bipolar

8 *

6 4

Schizophrenia

**

Unipolar

* *

2 0 SZ

Figure 2.1.

Mania Dep. Proband

Con.

Morbid risk in family members of probands with psychotic and affective disorders vs. controls. The morbid risk for bipolar disorder, schizophrenia and unipolar disorder is illustrated for probands with schizophrenia, mania, and controls. Morbid risk was assessed by direct interview and family history. The symbols * and ** indicate significant differences from the risk for control probands at the p<0.05 and p<0.01 levels respectively. (Tsuang, Winokur and Crowe, 1980).

partial overlap in familial transmission, or that the illnesses only partially breed true. Partially consistent with these results, Kendler and colleagues interviewed 1753 relatives of 349 probands with affective illness and schizophrenia as part of a systematic study of psychiatric illness in Roscommon County, Ireland (Kendler, Karkowski, and Walsh, 1998). They divided probands into six categories based on a latent class analysis of symptoms. Morbid risk for schizophrenia was increased in the relatives of probands with “classic schizophrenia” (5.1%) as compared to control probands, and in the relatives of those with schizomania-bipolar illness, though to a lesser extent (3.8%). Bipolar disorder was increased in the relatives of probands with schizomania-bipolar (4.4%), but not in those with “classic schizophrenia” (1.4%). However, other family studies have not supported such an overlap. Several studies of affective probands have failed to demonstrate an increase in schizophrenia in the relatives of bipolar probands (Weissman et al., 1984; Andreasen et al., 1987). Family study data have also been used to argue for a spectrum within affective disorders from non-psychotic to psychotic (Gershon et al., 1982). It has been reported that the overall degree of family risk, and hence, genetic loading, follows a gradient among different probands with schizoaffective and affective disorder: schizoaffective > bipolar I > bipolar II > unipolar. This gradient is consistent with a unitary genetic risk and a multiple threshold multifactorial model as outlined above. However, this study did not address possible generalizability to schizophrenia.

29

Overlapping genes and genetic models

Recently, Cardno et al. employed a novel twin strategy to address this question. Seventy seven monozygotic and 89 same sex dizygotic twin pairs were identified with psychotic illness (Cardno et al., 2002). A non-hierarchical approach to diagnosis was employed such that subjects could be placed in more than one of three categories: schizophrenia, schizoaffective, or manic. Each of the three across category groups of twins showed a higher correlation in monozygotic than in dizygotic twins consistent with a shared vulnerability. An independent pathway model analysis was most consistent with shared additive genetic effects and syndromeindependent environmental effects. Further support for a shared genetic susceptibility between schizophrenia and affective illness comes from prospective longitudinal follow-up studies of the children of patients with schizophrenia or affective disorders (Erlenmeyer-Kimling et al., 1997). In the New York High Risk Project, 84, 67 and 136 children at high risk for schizophrenia, affective illness and controls have been followed prospectively for over 20 years. Rates of schizophrenia and related psychoses occurred in the schizophrenia high-risk group but not the affective high-risk group. However, affective psychoses were observed at an elevated rate in both high-risk groups. Nonaffective psychosis was not elevated in either of the groups. In summary, these studies are consistent with a partial overlap in the genetic risk between schizophrenia and affective disorders, though the nature of this overlap varies between studies. This may be more pronounced for the occurrence of affective psychoses in the relatives of schizophrenics than for schizophrenia in the families of those with affective disorders. In particular, the overlap may primarily involve patients with schizoaffective disorders or psychotic affective disorders which have been variously defined in different studies. Genetic linkage studies Despite the partial overlap observed in many family studies, these data had widely been interpreted as showing that schizophrenia and bipolar disorder largely “breed true.” Therefore, it came as a surprise when linkage studies of both disorders began to identify the same chromosomal regions. Genetic linkage is said to be present when the alleles of a DNA marker and a trait locus are observed to cosegregate in families. Linkage implies that the gene for a disease or trait is physically nearby the marker on a chromosome. As the marker’s chromosomal location is known, the approximate location of the disease gene can be inferred. Though this concept had been developed nearly a century ago, its full exploitation awaited the development of molecular genetic methods to identify and genotype large numbers of polymorphic markers in the genome. The current approach is to genotype about 400 markers distributed around the genome at approximately 10

30

John R. Kelsoe

million base pair (Mb) intervals so as to cover the entire genome. Linkage may localize a disease gene to approximately 10–30 Mb in complex genetic traits. Such a region may include several hundred genes, and other genetic methods such as association must be employed to identify the specific gene. The strength of this approach, variously termed systematic gene mapping, genome scanning, positional cloning, or meiotic mapping, is that a gene is identified based on its location rather than its known function. A-priori knowledge of disease mechanism is not required and, hence, completely novel genes and molecular disease mechanisms can be discovered. Numerous genetic linkage studies of schizophrenia and bipolar disorder have been conducted. These are reviewed in detail elsewhere (Prathikanti and Mc Mahon, 2001; Mathews and Reus, 2003); and several recent large meta-analyses have been conducted (Badner and Gershon, 2002; Segurado et al., 2003). Numerous chromosomal regions have been implicated for both bipolar disorder and schizophrenia as illustrated in Figure 2.2. Early on in the use of this approach, several chromosomal regions, including 22q, 13q and 18p, were reported to be linked to both disorders (Pulver et al., 1994; Blouin et al., 1998; Detera-Wadleigh et al., 1999; Berrettini, 2000; Kelsoe et al., 2001; Schwab et al., 1998). This began to suggest the possibility that common genes were involved in the susceptibility of these two disorders (Kelsoe, 1999; Berrettini, 2000). Two of these regions, 13q and 22q, were also determined in one of the two existing meta-analyses to be the most robustly replicated regions in the genome for both disorders. Subsequently, several other regions were also identified for both disorders, such that approximately half of the genomic regions reported to be linked for each disorder were the same. These data must be interpreted in the light of many caveats. First, which, if any, of these reported linkages are real and replicated, is highly debated. The statistical requirements for genome-wide significance and replication have been debated, though some standards are generally accepted (Lander and Kruglyak, 1995). Linkage findings are highly inconsistent between studies, though a few regions have achieved genome-wide statistical significance in one study and significant replication in another. The two reported meta-analyses led to very different conclusions. Badner and Gershon (2002) employed a method designed to systematically determine if a significant linkage had been replicated. They found the most robust results for bipolar disorder on 22q and 13q; and for schizophrenia on 22q, 13q, and 8p. More recently, a larger set of studies was analyzed using a method called Genome Scan Meta-Analysis (GSMA) that includes all evidence across all studies in binned regions of the genome (Levinson et al., 2003). This strategy is able to identify regions that are significant in the combined analysis because of modest evidence across multiple studies, even though significance is not achieved in any single study. This analysis identified no region with significant linkage for bipolar

31

Overlapping genes and genetic models 1

3

2

5

4

6

8

7

9

1 1

1 0

Bipolar Schizophrenia

1 2 Figure 2.2.

1 3

1 4

1 5

1 6

1 7

1 8

1 9

2 0

2 1

2 2

X

Y

Overlapping linkage peaks in bipolar disorder and schizophrenia. The chromosomal locations for loci reported to be linked to bipolar disorder and schizophrenia. As linkage provides only approximate locations (10–30 Mb) overlap can be difficult to judge and may not represent the same genes. Yet, numerous regions are implicated for both disorders. Adapted and updated from Kelsoe (1999).

disorder, though the strongest evidence was found on 9p, 10q and 14q (Segurado et al., 2003). Analysis of studies of schizophrenia found genome-wide significance in 12 regions (Lewis et al., 2003). However, in these meta-analyses no regions were found to overlap between bipolar disorder and schizophrenia. The reason for such inconsistent results in linkage studies is unclear. However, it likely results from the large number of genes involved and a sampling effect. The difference in results between these two meta-analyses is likely more the result of the different samples of studies included than the analytic method. The question of overlap between bipolar disorder and schizophrenia might also be more powerfully addressed by a combined analysis of all schizophrenia and bipolar studies that tested for differences between the disorders. However, this has yet to be accomplished. Several other studies address whether psychosis in bipolar disorder may define a genetically distinct subtype of illness. Potash et al. examined a set of families with bipolar disorder for the familiality of psychosis (Potash et al., 2001). They found that families with a proband with psychosis were more likely to have family members whose bipolar illness also included psychotic symptoms. They then

32

John R. Kelsoe

extended this to linkage analysis and conducted a genome scan of their bipolar families, considering as affected only those with psychotic mood disorders (Potash et al., 2003). They detected linkage to two regions also implicated in schizophrenia on 13q and 22q. If the degree of genetic heterogeneity is quite high and there are many genes for both bipolar disorder and schizophrenia, then it is possible that some linkage peaks include separate and distinct genes for each disorder. The overlap of linkage peaks might then result from the proximity of distinct bipolar and schizophrenia genes rather than a true shared susceptibility. In fact, several regions in the genome such as 6p, 13q, and 22q may contain multiple susceptibility genes within a single linkage peak (Kelsoe et al., 2001; Straub et al., 2002; Shaw et al., 2003). Only the identification of specific genes for each disorder and the demonstration of their association in both disorders will prove a genetic overlap. Overlapping genes Recent success in identifying several specific susceptibility genes primarily for schizophrenia does, in fact, provide such evidence for overlap. Table 2.1 summarizes several genes, their chromosomal map positions, known functions and association results for both schizophrenia and bipolar disorder. Several of these are genes identified initially through positional cloning studies of schizophrenia. Subsequently, they have been reported to be also associated with bipolar disorder. Others are genes identified as candidates based on their known function. Dysbindin is a gene involved in cytoskeletal function that maps to 6p. Linkage to 6p was first identified in a sample of Irish families with schizophrenia (Straub et al., 1995). More recently, association and fine-mapping studies identified the dysbindin gene as the schizophrenia susceptibility gene in this region (Straub et al., 2002). Dysbindin is localized to both pre- and postsynaptic densities and has a role in receptor localization and trafficking of proteins involved in signaling (Husi et al., 2000; Inoue and Okabe, 2003). Through these presynaptic functions, dysbindin may play a role in modulating glutamate release (Numakawa et al., 2004). Dysbindin has also recently been shown to also be associated with psychotic bipolar disorder in an Israeli sample (Kohn et al., 2004). However, studies of major depression have been negative (Zill et al., 2004). Neuregulin 1 (NRG1) was similarly identified by positional cloning studies. Linkage of schizophrenia was first reported to 8p (Blouin et al., 1998). Neuregulin was later identified by association in studies in this region of the Icelandic population (Stefansson et al., 2002), and replicated in several other populations (Stefansson et al., 2002; Stefansson et al., 2003; Yang et al., 2003). Neuregulin 1 is one member of a family of growth factors that activate erbB receptors and may modulate a variety of

Pre- and postsynaptic cytoskeletal and receptor localization function; may modulate glutamate release Erb1 signaling, neuronal growth and synapse formation Indirectly modulates glutamate signaling by activating D-amino acid oxidase Cytoskeletal and centrosome function, possibly involved in receptor localization; important role in brain development Receptor desensitization

Dysbindin

G-protein signaling

RGS4

1q

22q

22q

1q

8p 13q

6p

Chromosome

Chowdari et al., 2002

Linkage to evoked potential and eye tracking dysfunction in schizophrenia families, MylesWorsley et al., 1999b Li et al., 1996

Blackwood et al., 2001

Stefansson et al., 2002 Chumakov et al., 2002

Straub et al., 2002

Schizophrenia

Rapid cycling bipolar disorder, Papolos et al., 1998 Chowdari et al., 2002 (Statistical trend)

Barrett et al., 2003

Hodgkinson et al., 2004

Association in psychotic bipolar disorder, Kohn et al., 2004 Green et al., 2004 Hattori et al., 2003

Bipolar

Association Studiesa

Notes: a First, primary or representative reference for the association for each disorder. b As this is a linkage rather than an association study, it is less certain that this is the responsible gene in this region.

Catabolism of catecholamines

COMT

GRK3

DISC1

Neuregulin 1 G72

Function

Gene

Table 2.1. Overlapping susceptibility genes for bipolar disorder and schizophrenia.

34

John R. Kelsoe

neuronal and synaptic functions including neuronal growth, synapse formation, and long-term potentiation (Moises et al., 2002; Corfas et al., 2004). Neuregulin 1 has recently also been reported to be associated with bipolar disorder in a British sample (Green et al., 2004). Chromosome 13q has been reported to be linked to both schizophrenia and bipolar disorder in several studies (Blouin et al., 1998; Detera-Wadleigh et al., 1999; Kelsoe et al., 2001). Additionally, G72 is a gene in this region that was shown to be associated with schizophrenia and presumably is responsible for the linkage signal in this region (Chumakov et al., 2002). The gene G72 is a novel gene that has been shown to activate D-amino-acid oxidase, and to thereby modulate D-serine regulation of glutamate neurotransmission. The gene has also been shown to be associated with bipolar disorder (Hattori et al., 2003; Chen et al., 2004c). The gene DISC1 was identified in a large Scottish family that transmitted a balanced translocation between chromosomes 1 and 11 along with psychiatric illness (Blackwood and Muir, 2004; Blackwood et al., 2001; Millar et al., 2001). Affected family members had a range of different psychiatric syndromes including schizophrenia, bipolar disorder, and recurrent major depression. Consistent with the phenotypes observed in this family, DISC1 has been reported to be associated with both schizophrenia and bipolar disorder (Hennah et al., 2003; Hodgkinson et al., 2004; Kockelkorn et al., 2004). Though its function remains unclear, DISC1 has been shown to associate with proteins relevant to centrosome and cytoskeletal function, suggesting that it may play a role in receptor localization (Millar et al., 2003; Morris et al., 2003; Millar et al., 2004; Schurov et al., 2004). It has also been shown to play an important role in brain development. The above genes were each first identified by positional cloning in studies of schizophrenia and later shown to be associated with bipolar disorder. The Gprotein receptor kinase 3 (GRK3) is a gene which we have identified in studies of bipolar disorder using a positional cloning approach. We have previously reported evidence of linkage to chromosome 22 in a study of 20 extended bipolar families (Kelsoe et al., 2001). These data suggested the presence of two distinct linkage peaks on chromosome 22, one of which mapped very close to the gene for GRK3. The gene GRK3 is a member of a family of receptor kinases that play an important role in homologous desensitization of G-protein coupled receptors (Gainetdinov et al., 2004). The relevance of this gene in neurotransmitter systems relevant to psychiatric illness was demonstrated in a microarray expression profiling study of an animal model of mania and psychosis. Twenty-four hours following methamphetamine administration, GRK3 expression was substantially increased in the prefrontal cortex in rats (Niculescu, et al., 2000). Sequencing of the GRK3 gene in bipolar probands identified several variants in the promoter of the gene (Barrett et al., 2003). One of these, P-5, was significantly associated with bipolar disorder

35

Overlapping genes and genetic models

in two independent samples. A study of eight large families with schizophrenia from Utah suggests that GRK3 may be linked to certain pathophysiological processes in schizophrenia (Myles-Worsley et al., 1999). A measure of inhibitory information processing comprising P50 evoked potentials and eye tracking was employed as a quantitative endophenotype in these families. The strongest evidence of linkage was obtained to a marker in the GRK3 gene. Linkage does not have the genomic resolution that an association study would have to definitively identify GRK3, as another nearby gene could be responsible. Furthermore, one study in a Japanese population found no evidence of association (Yu et al., 2004). However, this does suggest that GRK3 may play a role both in bipolar disorder and an aspect of the pathophysiology of schizophrenia. Several genes have been selected as possible candidate genes for bipolar disorder or schizophrenia based on their known function in neurotransmitter metabolism. These genes have then shown association to both bipolar disorder and schizophrenia. Two of these are illustrated as examples in Table 2.1. The above genes identified through a positional cloning approach likely are more important and have a larger effect on disease than such functional candidates because they were identified as having the strongest evidence of linkage in a scan of the entire genome in one or more studies. Yet, these functional candidates also seem to play a role in both disorders. Catechol-O-methyl transferase (COMT) plays an important role in the degradation of dopamine and norepinephrine by methylation. This is likely the primary mode of elimination of dopamine from the synapse in prefrontal cortex. A defect in its function would therefore result in elevated synaptic dopamine. A val/met substitution has been identified and shown to be associated with a difference in enzyme function (Lotta et al., 1995; Lachman et al., 1996a; Lachman et al., 1996b). Though it has been debated whether this variant or another is the primary effect on gene function (Chen et al., 2004a), this variant has subsequently been shown to be associated with both risk for schizophrenia (Li et al., 1996; Bray et al., 2003) and performance on neuropsychological tests of executive function in schizophrenia (Egan et al., 2001). Though studies of bipolar disorder and COMT have been mixed, several studies have reported evidence for association suggesting that this gene may play a role in both disorders (Papolos et al., 1998; Rotondo et al., 2002; Shifman et al., 2002). The gene RGS4 was identified as a possible candidate gene for schizophrenia by microarray expression profiling of postmortem brain (Mirnics et al., 2001). The expression levels of thousands of genes were measured using an oligonucleotidebased microarray method. The gene, regulator of G-protein signaling 4 (RGS4), was found to be significantly decreased in the cortex of ten patients with schizophrenia. The RGS4 gene is a member of a 20-gene family of RGS genes and plays a role in the regulation of signaling through dopamine and other G-protein coupled receptors. The RGS4 gene maps to chromosomal region 1q21–q22 which has been

36

John R. Kelsoe

linked to schizophrenia. Subsequently, RGS4 was shown to be associated with schizophrenia in two independent samples (Chowdari et al., 2002). This study also showed a trend for association in a set of bipolar families using family-based association methods. Several subsequent studies have replicated the association in schizophrenia (Chen et al., 2004b; Morris et al., 2004; Williams et al., 2004). The trend for association in bipolar disorder has yet to be replicated but suggests a possible shared gene. This work on RGS4, as with that for GRK3 described above, employed a novel strategy of using microarray expression profiling to identify candidate genes, which were subsequently confirmed with association studies. At this point in time, more genes have been identified for schizophrenia than for bipolar disorder. The genes described above, with the exception of GRK3, were first identified for schizophrenia, and then also later shown to be associated with bipolar disorder. This is an incomplete selection of genes intended to illustrate the emerging story of the overlap between bipolar disorder and schizophrenia at the gene level. Overlap at the family and linkage level is intriguing and suggestive, but only the identification of specific genes associated with both disorders provides convincing evidence of genetic overlap. Conclusion Together, the existing data support at least a partial genetic overlap between schizophrenia and bipolar disorder. The family and twin data suggest that a portion of the genetic vulnerability is shared. The linkage data implicate some regions as containing genes for both disorders. The most compelling data come from the recent identification of genes for each disorder. Though this is at an early stage and only a few genes have been identified with promising data, several of them show evidence of association to both disorders. If a substantial portion of the genes for these disorders is shared, then one must ask, what is it that distinguishes between these two illnesses? Some possible explanations and models are listed in Table 2.2. One possibility is that of allelic heterogeneity. Different mutations in the same gene may disrupt gene and protein function in different ways and lead to different phenotypes. Many examples exist in medical genetics of wide differences in disease presentation resulting from different mutations in the same gene. Different mutations in the CFTR gene, for example, can lead to cystic fibrosis or male sterility from agenesis of the vas deferens (Bienvenu et al., 1993). Though the data presented above argues strongly for the role of several genes in schizophrenia and bipolar disorder, clear functional variants are known for only one of them (COMT). It is likely that most susceptibility genes carry many mutations that can singly or together impact susceptibility. Therefore, the manner in which mutations in genes lead to

37

Overlapping genes and genetic models Table 2.2. How can the same genes cause bipolar disorder and schizophrenia? Some possible models of genetic overlap.

Model

Description

Allelic heterogeneity

Different mutations within the same gene might predispose to different disorders

Common genes – different environment

The same genes operate in both disorders, but different environmental factors result in different disorders

Gene – gene interaction

The same genes are involved in both disorders; different combinations of genes result in different disorders

Quantitative multiple threshold model

The same genes are involved in both disorders. Their effects add together along a single risk continuum. Above a certain threshold, bipolar disorder is manifest; a higher threshold results in schizophrenia

phenotype is likely to be quite complex. If this model is correct, then linkage, which is not specific to allele and disease mutation, would find overlap between disorders. However, association studies which are specific to allele and mutation may find different alleles, haplotypes and functional variants associated with each disorder. Another possibility is that much of the genetic vulnerability is non-specific and it is differences in environment and its interaction with genes that determine the specific phenotypic presentation. The twin study of Cardno et al. (2002) reviewed above provides some support for this idea by detecting common genetic factors and individual syndrome specific environmental factors playing a role in susceptibility. Gene-gene interactions might also explain the specificity to phenotype. Individual genes might not be specific to illness, but rather certain combinations of genes would result in specific phenotypes. A variant of this idea would be that some genes would increase risk in a non-specific way and their expression be modified by other genes whose effect is more specific to phenotype. Lastly, the multiple threshold model described above is compatible with such an overlap. In this model, the same genes operate to produce a cumulative susceptibility to psychotic illnesses. The more susceptibility alleles an individual carries, the greater their vulnerability. With increasing numbers of alleles, different thresholds are passed resulting in major depression < bipolar disorder < schizophrenia. Hence, it is the number of susceptibility genes that distinguish the different disorders. However, this model predicts that bipolar disorder is more common than schizophrenia, which is contrary to epidemiological observations.

38

John R. Kelsoe

It is increasingly likely from the emerging molecular data that there is some degree of genetic overlap between bipolar disorder and schizophrenia. This is consistent with both older and more recent family and twin data. Only the identification of more susceptibility genes and elucidation of their role in each disorder will enable a more detailed understanding of the genetic relationship among psychotic disorders. More detailed studies of environmental factors are also required. Specific environments may influence the phenotype in terms of both severity and presentation of illness. It is also possible that specific environmental factors may have interactions with specific genes. It is not clear which of the models reviewed here explain the emerging data for genetic overlap. Most likely each of the possibilities may operate to some extent. The prospect of the identification of an increasing number of susceptibility genes in the coming years promises exciting developments and a new understanding of the complex relationship between bipolar disorder and schizophrenia.

R E F E R E N C ES Andreasen, N. C., Rice, J., Endicott, J. et al. (1987). Familial rates of affective disorder. A report from the National Institute of Mental Health Collaborative Study. Archives of General Psychiatry, 44, 461–9. Badner, J. A. and Gershon, E. S. (2002). Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Molecular Psychiatry, 7 (4), 405–11. Barrett, T. B., Hauger, R. L. and Kennedy, J. L. (2003). Evidence that a single nucleotide polymorphism in the promoter of the G protein receptor kinase 3 gene is associated with bipolar disorder. Molecular Psychiatry, 8, 546–57. Berrettini, W. H. (2000). Susceptibility loci for bipolar disorder: overlap with inherited vulnerability to schizophrenia. Biological Psychiatry, 47 (3), 245–51. Bienvenu, T., Beldjord, C., Adjiman, M. and Kaplan, J. C. (1993). Male infertility as the only presenting sign of cystic fibrosis when homozygous for the mild mutation R117H. Journal of Medical Genetics, 30 (9), 797. Blackwood, D. H., Fordyce, A., Walker, M. T. et al. (2001). Schizophrenia and affective disorders–cosegregation with a translocation at chromosome 1q42 that directly disrupts brainexpressed genes: clinical and P300 findings in a family. American Journal of Human Genetics, 69 (2), 428–33. Blackwood, D. H. and Muir, W. J. (2004). Clinical phenotypes associated with DISC1, a candidate gene for schizophrenia. Neurotoxicity Research, 6 (1), 35–41. Blouin, J. L., Dombroski, B. A., Nath, S. K. et al. (1998). Schizophrenia susceptibility loci on chromosomes 13q32 and 8p21. Nature Genetics, 20, 70–3. Bray, N. J., Buckland, P. R., Williams, N. M. et al. (2003). A haplotype implicated in schizophrenia susceptibility is associated with reduced COMT expression in human brain. American Journal of Human Genetics, 73 (1), 152–61.

39

Overlapping genes and genetic models Cardno, A. G., Rijsdijk, F. V., Sham, P. C., Murray, R. M. and McGuffin, P. (2002). A twin study of genetic relationships between psychotic symptoms. American Journal of Psychiatry, 159 (4), 539–45. Chen, J., Lipska, B. K., Halim, N. et al. (2004a). Functional analysis of genetic variation in catechol-O-methyltransferase (COMT): effects on mRNA, protein, and enzyme activity in postmortem human brain. American Journal of Human Genetics, 75 (5), 807–21. Chen, X., Dunham, C., Kendler, S. et al. (2004b). Regulator of G-protein signaling 4 (RGS4) gene is associated with schizophrenia in Irish high density families. American Journal of Medical Genetics, 129B (1), 23–6. Chen, Y. S., Akula, N., Detera-Wadleigh, S. D. et al. (2004c). Findings in an independent sample support an association between bipolar affective disorder and the G72/G30 locus on chromosome 13q33. Molecular Psychiatry, 9 (1), 87–92. Chowdari, K. V., Mirnics, K., Semwal, P. et al. (2002). Association and linkage analyses of RGS4 polymorphisms in schizophrenia. Human Molecular Genetics, 11 (12), 1373–80. Chumakov, I., Blumenfeld, M., Guerassimenko, O. et al. (2002). Genetic and physiological data implicating the new human gene G72 and the gene for D-amino-acid oxidase in schizophrenia. Proceedings of the National Academy of Sciences USA, 99 (21), 13675–80. Corfas, G., Roy, K. and Buxbaum, J. D. (2004). Neuregulin 1-erbB signaling and the molecular/cellular basis of schizophrenia. Nature Neuroscience, 7 (6), 575–80. Detera-Wadleigh, S. D., Badner, J. A., Berrettini, W. H. (1999). A high-density genome scan detects evidence for a bipolar-disorder susceptibility locus on 13q32 and other potential loci on 1q32 and 18p11.2. Proceedings of the National Academy of Sciences USA, 96 (10), 5604–9. Egan, M. F., Goldberg, T. E., Kolachana, B. S. et al. (2001). Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia. Proceedings of the National Academy of Sciences USA, 98 (12), 6917–22. Erlenmeyer-Kimling, L., Adamo, U. H., Rock, D. et al. (1997). The New York High-Risk Project. Prevalence and comorbidity of axis I disorders in offspring of schizophrenic parents at 25-year follow-up. Archives of General Psychiatry, 54, 1096–102. Gainetdinov, R. R., Premont, R. T., Bohn, L. M., Lefkowitz, R. J. and Caron, M. G. (2004). Desensitization of G protein-coupled receptors and neuronal functions. Annual Review of Neuroscience, 27, 107–44. Gershon, E. S., Hamovit, J., Guroff, J. J. et al. (1982). A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands. Archives of General Psychiatry, 39, 1157–67. Green, E. K., Raybould, R., Macgregor, S. et al. (2004). Evidence that variation at the neuregulin gene (NRG1) influences susceptibility to bipolar disorder with maximal effect in individuals who experience mania and mood-incongruent psychotic features. American Journal of Medical Genetics, 130B, 27. Hattori, E., Liu, C., Badner, J. A. et al. (2003). Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. American Journal of Human Genetics, 72 (5), 1131–40. Hennah, W., Varilo, T., Kestila, M. et al. (2003). Haplotype transmission analysis provides evidence of association for DISC1 to schizophrenia and suggests sex-dependent effects. Human Molecular Genetics, 12 (23), 3151–9.

40

John R. Kelsoe Hodgkinson, C. A., Goldman, D., Jaeger, J. et al. (2004). Disrupted in schizophrenia 1 (DISC1): Association with schizophrenia, schizoaffective disorder, and bipolar disorder. American Journal of Human Genetics, 75 (5), 862–72. Husi, H., Ward, M. A., Choudhary, J. S., Blackstock, W. P. and Grant, S. G. (2000). Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nature Neuroscience, 3 (7), 661–9. Inoue, A. and Okabe, S. (2003). The dynamic organization of postsynaptic proteins: translocating molecules regulate synaptic function. Current Opinion in Neurobiology, 13 (3), 332–40. Kelsoe, J. R. (1999). Recent progress in the search for genes for bipolar disorder. Current Psychiatry Reports, 1 (2), 135–40. Kelsoe, J. R., Spence, M. A., Loetscher, E. et al. (2001). A genome survey indicates a possible susceptibility locus for bipolar disorder on chromosome 22. Proceedings of the National Academy of Sciences USA, 98 (2), 585–90. Kendler, K. S., Karkowski, L. M. and Walsh, D. (1998). The structure of psychosis: latent class analysis of probands from the Roscommon Family Study. Archives of General Psychiatry, 55, 492–9. Kockelkorn, T. T., Arai, M., Matsumoto, H. et al. (2004). Association study of polymorphisms in the 5’ upstream region of human DISC1 gene with schizophrenia. Neuroscience Letters, 368 (1), 41–5. Kohn, Y., Danilovich, E., Filon, D. et al. (2004). Linkage disequlibrium in the DTNBP1 (dysbindin) gene region and on chromosome 1p36 among psychotic patients from a genetic isolate in Israel: findings from identity by descent haplotype sharing analysis. American Journal of Medical Genetics, 128B (1), 65–70. Lachman, H. M., Morrow, B., Shprintzen, R. et al. (1996a). Association of codon 108/158 catechol-O-methyltransferase gene polymorphism with the psychiatric manifestations of velo-cardio-facial syndrome. American Journal of Medical Genetics, 67, 468–72. Lachman, H. M., Papolos, D. F., Saito, T. et al. (1996b). Human catechol-O-methyltransferase pharmacogenetics: description of a functional polymorphism and its potential application to neuropsychiatric disorders. Pharmacogenetics, 6, 243–50. Lander, E. and Kruglyak, L. (1995). Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nature Genetics, 11, 241–7. Levinson, D. F., Levinson, M. D., Segurado, R. and Lewis, C. M. (2003). Genome scan metaanalysis of schizophrenia and bipolar disorder, part I: Methods and power analysis. American Journal of Human Genetics, 73 (1), 17–33. Lewis, C. M., Levinson, D. F., Wise, L. H. et al. (2003). Genome scan meta-analysis of schizophrenia and bipolar disorder, part II: Schizophrenia. American Journal of Human Genetics, 73 (1), 34–48. Li, T. Sham, P. C., Vallada, H. et al. (1996). Preferential transmission of the high activity allele of COMT in schizophrenia. Psychiatric Genetics, 6 (3), 131–3. Lotta, T., Vidgren, J., Tilgmann, C. et al. (1995). Kinetics of human soluble and membranebound catechol O-methyltransferase: a revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry, 34 (13), 4202–10. Mathews, C. A. and Reus, V. I. (2003). Genetic linkage in bipolar disorder. CNS Spectrums, 8 (12), 891–904.

41

Overlapping genes and genetic models Millar, J. K., Christie, S., Anderson, S. et al. (2001). Genomic structure and localisation within a linkage hotspot of Disrupted In Schizophrenia 1, a gene disrupted by a translocation segregating with schizophrenia. Molecular Psychiatry, 6 (2), 173–8. Millar, J. K., Christie, S. and Porteous, D. J. (2003). Yeast two-hybrid screens implicate DISC1 in brain development and function. Biochemical and Biophysical Research Communications, 311 (4), 1019–25. Millar, J. K., James, R., Brandon, N. J. and Thomson, P. A. (2004). DISC1 and DISC2: discovering and dissecting molecular mechanisms underlying psychiatric illness. Annals of Medicine, 36 (5), 367–78. Mirnics, K., Middleton, F. A., Stanwood, G. D., Lewis, D. A. and Levitt, P. (2001). Disease-specific changes in regulator of G-protein signaling 4 (RGS4) expression in schizophrenia. Molecular Psychiatry, 6 (3), 293–301. Moises, H. W., Zoega, T. and Gottesman, I. I. (2002). The glial growth factors deficiency and synaptic destabilization hypothesis of schizophrenia. BMC Psychiatry, 2 (1), 8. Morris, D. W., Rodgers, A., McGhee, K. A. et al. (2004). Confirming RGS4 as a susceptibility gene for schizophrenia. American Journal of Medical Genetics, 125B (1), 50–3. Morris, J. A., Kandpal, G., Ma, L. and Austin, C. P. (2003). DISC1 (Disrupted-InSchizophrenia 1) is a centrosome-associated protein that interacts with MAP1A, MIPT3, ATF4/5 and NUDEL: regulation and loss of interaction with mutation. Human Molecular Genetics, 12 (13), 1591–1608. Myles-Worsley, M., Coon, H., McDowell, J. et al. (1999). Linkage of a composite inhibitory phenotype to a chromosome 22q locus in eight Utah families. American Journal of Medical Genetics, 88 (5), 544–50. Niculescu, A. B. 3rd, Segal, D. S., Kuczenski, R. et al. (2000). Identifying a series of candidate genes for mania and psychosis: a convergent functional genomics approach. Physiological Genomics, 4 (1), 83–91. Numakawa, T., Yagasaki, Y., Ishimoto, T. et al. (2004). Evidence of novel neuronal functions of dysbindin, a susceptibility gene for schizophrenia. Human Molecular Genetics, 13 (21), 2699–708. Papolos, D. F., Veit, S., Faedda, G. L., Saito, T. and Lachman, H. M. (1998). Ultra-ultra rapid cycling bipolar disorder is associated with the low activity catecholamine-O-methyltransferase allele. Molecular Psychiatry, 3, 346–9. Potash, J. B., Willour, V. L., Chiu, Y. F. et al. (2001). The familial aggregation of psychotic symptoms in bipolar disorder pedigrees. American Journal of Psychiatry, 158 (8), 1258–64. Potash, J. B., Zandi, P. P., Willour, V. L. et al. (2003). Suggestive linkage to chromosomal regions 13q31 and 22q12 in families with psychotic bipolar disorder. American Journal of Psychiatry, 160 (4), 680–6. Prathikanti, S. and McMahon, F. J. (2001). Genome scans for susceptibility genes in bipolar affective disorder. Annals of Medicine, 33 (4), 257–62. Pulver, A. E., Karayiorgou, M., Wolyniec, P. S. et al. (1994). Sequential strategy to identify a susceptibility gene for schizophrenia: report of potential linkage on chromosome 22q12–q13.1: Part 1. American Journal of Medical Genetics, 54, 36–43. Rotondo, A., Mazzanti, C., Dell’Osso, L. et al. (2002). Catechol o-methyltransferase, serotonin

42

John R. Kelsoe transporter, and tryptophan hydroxylase gene polymorphisms in bipolar disorder patients with and without comorbid panic disorder. American Journal of Psychiatry, 159 (1), 23–9. Schurov, I. L., Handford, E. J., Brandon, N. J. and Whiting, P. J. (2004). Expression of disrupted in schizophrenia 1 (DISC1) protein in the adult and developing mouse brain indicates its role in neurodevelopment. Molecular Psychiatry, 9 (12), 1100–10. Schwab, S. G., Hallmayer, J., Lerer, B. et al. (1998). Support for a chromosome 18p locus conferring susceptibility to functional psychoses in families with schizophrenia, by association and linkage analysis. American Journal of Human Genetics, 63, 1139–52. Segurado, R., Detera-Wadleigh, S. D., Levinson, D. F. et al. (2003). Genome scan meta-analysis of schizophrenia and bipolar disorder, part III: Bipolar disorder. American Journal of Human Genetics, 73 (1), 49–62. Shaw, S. H., Mroczkowski-Parker, Z., Shekhtman, T. et al. (2003). Linkage of a bipolar disorder susceptibility locus to human chromosome 13q32 in a new pedigree series. Molecular Psychiatry, 8 (5), 558–64. Shifman, S., Bronstein, M., Sternfeld, M. et al. (2002). A highly significant association between a COMT haplotype and schizophrenia. American Journal of Human Genetics, 71 (6), 1296–302. Stefansson, H., Sarginson, J., Kong, A. et al. (2003). Association of neuregulin 1 with schizophrenia confirmed in a Scottish population. American Journal of Human Genetics, 72 (1), 83–7. Stefansson, H., Sigurdsson, E., Steinthorsdottir, V. et al. (2002). Neuregulin 1 and susceptibility to schizophrenia. American Journal of Human Genetics, 71 (4), 877–92. Straub, R. E., Jiang, Y., MacLean, C. J. et al. (2002). Genetic variation in the 6p22.3 gene DTNBP1, the human ortholog of the mouse dysbindin gene, is associated with schizophrenia. American Journal of Human Genetics, 71 (2), 337–48. Straub, R. E., MacLean, C. J., O’Neill, F. A. et al. (1995). A potential vulnerability locus for schizophrenia on chromosome 6p24–22: evidence for genetic heterogeneity. Nature Genetics, 11, 287–93. Tsuang, M. T., Winokur, G. and Crowe, R. R. (1980). Morbidity risks of schizophrenia and affective disorders among first degree relatives of patients with schizophrenia, mania, depression and surgical conditions. British Journal of Psychiatry, 137, 497–504. Viskochil, D. (2002). Genetics of neurofibromatosis 1 and the NF1 gene. Journal of Child Neurology, 17 (8), 562–70. Weissman, M. M., Gershon, E. S., Kidd, K. K. et al. (1984). Psychiatric disorders in the relatives of probands with affective disorders. The Yale University–National Institute of Mental Health Collaborative Study. Archives of General Psychiatry, 41, 13–21. Williams, N. M., Preece, A., Spurlock, G. et al. (2004). Support for RGS4 as a susceptibility gene for schizophrenia. Biological Psychiatry, 55 (2), 192–5. Yang, J. Z., Si, T. M., Ruan, Y. et al. (2003). Association study of neuregulin 1 gene with schizophrenia. Molecular Psychiatry, 8 (7), 706–9. Yu, S. Y., Takahashi, S., Arinami, T. et al. (2004). Mutation screening and association study of the betaadrenergic receptor kinase 2 gene in schizophrenia families. Psychiatry Research, 125 (2), 95–104. Zill, P., Baghai, T. C., Engel, R., Zwanzger, P. et al. (2004). The dysbindin gene in major depression: an association study. American Journal of Medical Genetics, 129B (1), 55–8.

3

The continuum of psychosis and its genetic basis T. J. Crow SANE Prince of Wales International Centre, Warneford Hospital, Oxford

Kraepelin’s 1920 doubts about the binary principle It is remarkable that over a hundred years after its original formulation Kraepelin’s binary concept still dominates psychiatric nosology, even though he had apparently abandoned it. In 1920 he wrote: Perhaps it is also possible to tackle the difficulties which still prevent us from distinguishing reliably between manic depressive insanity and dementia praecox. No experienced psychiatrist will deny that there is an alarmingly large number of cases in which it seems impossible, in spite of the most careful observation, to make a firm diagnosis. Nevertheless it is becoming increasingly clear that we cannot distinguish satisfactorily between these two illnesses, and this brings home the suspicion that our formulation of the problem may be incorrect. (Kraepelin, 1920.)

The inadequacies of the original concept are easily demonstrated by contemporary operational criteria. Endicott et al. (1982) presented the findings of a study including 46 consecutive admissions to the Psychiatric Institute in New York with a diagnosis of schizophrenia by seven different operational sets of criteria. The variation in numbers was wide. By Taylor and Abrams criteria (1978), the most restrictive, only six cases were allocated a diagnosis of schizophrenia, whereas by the New Haven criteria of Astrachan et al. (1972), the most liberal, 44 cases were so diagnosed, a ratio of more than 7:1. The New Haven criteria may be considered somewhat extreme in their scope. The next broadest set was that of Carpenter et al. (1973) that included 24 cases, but this still gives a ratio of more than 4:1. Such variation clearly indicates that there is no agreement on what constitutes a schizophrenic illness. Some idea of the source of the variation can be discerned by applying the RDC (Research Diagnostic Criteria) – the modal system – for schizophrenia, schizoaffective and affective illnesses to the whole set of cases as indicated by the The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

43

44

T. J. Crow

Taylor & Abrams

6

RDC chronic

11

Feighner

12

RDC schizophrenia

18

DSM 3

19

Carpenter

24

New Haven

Schizophrenia

Figure 3.1.

46 cases

ed

ar

Schizoaffective

es s ep r D

Bi p

ol

c an i M

Ch r

on

ic Su b + -ac ac ut ut e e D ep re ss ed

44

Endicott 1982

Affective

Numbers of patients that are allocated to the category of schizophrenia by different systems of operational diagnostic criteria. The series relates to 46 consecutively admitted cases to the Psychiatric Institute in New York who met any of the diagnostic criteria. The categories on the bottom line are those defined by the Research Diagnostic Criteria. From Endicott et al. (1982).

bottom line in Figure 3.1. Whereas by RDC criteria for schizophrenia 18 cases are included, a further 14 are allocated to the schizoaffective category and the remaining 20 to the category of affective psychoses. As the definition of schizophrenia expands, therefore, it includes cases that by some criteria are regarded as “schizoaffective” and with further expansion it includes some of those allocated to the category of affective psychosis. A continuum concept is implicit in the findings of this elegant study. An evolutionary and ontogenetic concept In the same paper in which Kraepelin expressed his doubts about the binary concept, he also wrote: “. . . if we try to relate the syndromes of insanity to particular stages of personality, we have few hypotheses on which to base our argument. If these attempts are to be more than tentative groupings we must trace the phenomena of our inner life back to their roots in the psyche of the child, of primitive man, and of animals.” It seems he was considering a developmental, even an evolutionary, concept of the origin of psychosis. One can ask how far have we advanced along this path in the intervening hundred years? That the continuum is a developmental phenomenon can be seen in the relationship between age of onset and form of psychosis (Figure 3.2). Here we see that with varying systems of categorical diagnosis there are variations in age of onset.

45

The continuum of psychosis and its genetic basis Female schizophrenia Male schizophrenia

Male affective Female affective

Form of psychosis

Penrose

Eaton

Schizophrenia (% male)

Unipolar SA Bipolar affective Bipolar SA Schizophrenia

Marneros

Kendler

Unipolar affective

BP/ schizomania Schizophreniform Schizodepression Severe Hebephrenia depression Classical schiz. 24

Figure 3.2.

Maryland (46%) Victoria (51%) Denmark (62%)

26

28

30

32

Age of onset (years) 34

36

38

40

42

44

Age at onset and form of psychotic illness in four series (Penrose, 1991; Eaton et al. 1992; Marneros et al. 1995; Kendler et al. 1998). Penrose’s data relate to 5456 pairs of relatives with mental illness admitted to the Ontario (Canada) hospitals between 1926 and 1943. Eaton et al. included case-register data on schizophrenia from three centers (a fourth was omitted as it included onsets in old age). Marneros et al. included 355 cases of psychosis in a long-term follow-up study in Cologne, Germany. Kendler et al.’s data are from the Roscommon family study (1998). (Figure from Crow, 1998.).

For example Kendler et al. (1998) in their Irish family studies arrived a system of six categories of psychosis including classical schizophrenia and hebephrenia as representatives of the schizophrenic part of the spectrum, schizodepression and bipolar schizomania and perhaps schizophreniform psychosis as representatives of the intermediate states, and severe depression as representative of affective psychosis. What is notable about these categories is the relatively early age of onset. By their selection criteria these authors presumably defined more severely affected groups of patients. Marneros et al. (1995) with a later mean age of onset still found cases that met their criteria for schizophrenia with earlier age of onset than the bipolar schizoaffective patients, and these in turn had an earlier onset than the bipolar affectives. Unipolar schizoaffective illness had an earlier onset than unipolar affective illness. The general rule is that illnesses that are schizophrenic in form have an earlier onset than those that are schizoaffective, and these have an earlier onset than the affective illnesses. Thus age of onset and form of psychosis are related. They reflect some fundamental underlying variable. The second phenomenon that is apparent in Figure 3.2 is that onsets of schizophrenia in general are earlier in males than in females. This is apparent in the case

46

T. J. Crow

register studies of Eaton et al. (1992). It appears that the diagnostic criteria used in Denmark are more restrictive than those used in the case registers in Australia or the United States, and the percentage of males recorded in the former is higher. Earlier age of onset of psychosis in males is perhaps the best established and least explained epidemiological fact about psychosis. According to Rosenthal (1962; 1970), the relationship between age of onset and sex is linear over the age range 4–60 years. What is the genetic basis of these phenomena? Much endeavour has now gone into linkage studies in psychosis. A prevalent view is that there has been progress on a number of loci and even that genes have already been established, thus substantiating the view that “polygenes of small effect” contribute to the predisposition. I share the contrary view (Crow, 1995) that there is “a continuum of psychosis, a single human gene and not much else.” Are there genes for psychosis? Two meta-analyses of the genome scans in relation to schizophrenia and two in relation to bipolar disorder have now been completed. Taken together the findings are instructive: • Of the two meta-analyses in schizophrenia one draws attention to three and the other to nine possible loci of significance. Badner and Gershon (2002) pointed to 8p, 13q and 22q as the regions of greatest significance while by the method Lewis et al. (2003) adopted, the strongest finding was on 2p with significant evidence for linkage on 3p, 11q, 5q, 20p, all yielding stronger evidence than the strongest candidate (8p) in the first study. Thus the two meta-analyses applied to a material with an overlap of approximately 500 pedigrees out of a total of 681 (Badner and Gershon, 2002) or 1208 (Lewis et al., 2003) agree only on two out of ten chromosome arms identified to harbour a locus for schizophrenia. The strongest finding in one study ([2003] linkage to 2p) was not detected in the other analysis (Badner and Gershon, 2002). The reason is that this locus emerged from a large study ([DeLisi et al., 2002]; n  380 sibling-pairs) that was included only in the second meta-analysis. • While the meta-analyses of schizophrenia (Lewis et al., 2003) and of bipolar disorder (Segurado et al., 2003) were conducted with an identical methodology, the conclusions that the authors draw are quite different. Whereas the schizophrenia meta-analysis concluded that there were eight bins that yielded genome-wide evidence of linkage, the bipolar meta-analysis revealed no such evidence (the locations given in the table for this study are the nominal 5% by-bin significances) and those locations that yielded the largest rank sums showed no overlap with the bins identified for schizophrenia. This is surprising to those who expected that there would be some genetic commonality between these major phenotypes

47

The continuum of psychosis and its genetic basis

(Stassen et al., 1988; Gershon et al., 1988; Maier et al., 1993). It is the more surprising when it is appreciated that the core phenotype in both meta-analyses included schizoaffective diagnoses. • The two findings in the first analysis of bipolar disorder (Badner and Gershon, 2002) of 353 families were not replicated in the second meta-analysis (Segurado et al., 2003) that included 347 families from essentially the same published literature but identified four different locations of interest. Where does this leave the search for genes for psychosis? At the very least it is clear that no strong signal is emerging, and more importantly there is no obvious criterion for distinguishing signal from noise. Another way of approaching the problem is to compare the outcome in the individual studies. What I have done is to juxtapose the findings in the two largest sibling-pair studies (DeLisi et al., 2002; Williams et al., 2003) with respect to the main finding. Each study drew attention to two findings of interest – DeLisi et al. (2002) to a locus around the chromosome 2 centromere and on chromosome 10, and Williams et al. (2003) drew attention to a linkage on chromosome 17 and one also on chromosome 10. But the loci on chromosome 10 in the two studies are at opposite telomeres! Therefore there is no agreement between the two studies, and it has to be noted that the linkages in any case are relatively weak, barely reaching the criteria for genome-wide significance. Within these studies there was no support for loci on chromosome 1 (where the DISC1 gene has been identified as a candidate of interest), on chromosome 6 (Dysbindin is located here), chromosome 8 (Neuregulin) and chromosome 22 (Catechol-o-methyl transferase). Thus there is little support for candidate genes of current interest. A sex chromosomal locus? There is, however, a genetic phenomenon that deserves more attention than it has so far received. This is the same sex concordance effect which was first reported for psychosis by Mott (1910), was discussed by Penrose (1942) and Rosenthal (1962) and is present in affective as well as schizophrenic psychoses. It is, I believe, consistent with a sex chromosomal locus, i.e., with X and Y linkage. Same sex concordance arises because when a sex-linked gene is transmitted from a father it travels either on his X chromosome or on his Y. If on the Y, it is transmitted to his sons, whereas if on the X it is transmitted to his daughters. Thus a tendency arises for the gene to be associated within families with sex of proband. If the gene is transmitted from the mother on her X chromosome it may travel either to sons or daughters, indicating that (at least with dominant transmission) same-sex concordance arises from paternal transmission. I have previously summarized the evidence for same-sex concordance in psychosis (Crow, 1994). Among recent surveys in

48

T. J. Crow

affective disorder Angst et al. (1980) reported that “in 670 first-degree relatives of 95 probands with bipolar manic depressive disorders . . . the possibility has to be considered that relatives of the same sex as the proband show a somewhat higher morbidity risk than those of the opposite sex”; and Rice et al. (1987) reported that “in 187 families of bipolar probands . . . multifactorial analysis found significant heterogeneity for sex-specific siblings correlations (with brother–sister smaller than same sex correlations) . . .” While same sex concordance in schizophrenia has been discussed (Crow et al., 1989; 1990) it seems to have been overlooked how consistent the phenomenon is in affective disorders. There is a thin thread through the literature that schizophrenia is a disorder specific to the neocortex of Homo sapiens. This was hinted at by Crichton-Browne (1879) and by Southard (1915) and seems to have been made explicit by Miskolczy (1933). It has been supported more recently by Parfitt (1956) and Pearlson et al. (1995) in their concept of schizophrenia as a disorder of heteromodal association cortex. This, together with the same-sex concordance effect, may be the key to an advance in our understanding of the genetics of the continuum of psychosis. For if psychosis is linked to an X–Y homologous gene then the arrangement of genes on the sex chromosomes gives an important clue to recent changes in hominid evolution. A number of regions of X–Y homology are now established (Figure 3.3). In general the X chromosome is stable in structure across mammalian species; the X chromosome of the mouse looks remarkably similar in banding pattern to that of man. By contrast the Y varies considerably across species, having notable variations across the great apes and Homo sapiens. In general, variations on the Y chromosome have arisen as a result of translocations and deletions that have occurred at various points in mammalian evolution, and these translocations have often resulted from duplications of blocks on the X. Such rearrangements separate orders, families and sometimes species (Lambson et al., 1992). Of greatest interest is the 5Mb block in Xq21.3 that is represented as two regions on the Y chromosome short-arm, because this homology arose after the separation of the lineages leading to the chimpanzee and to Homo sapiens. Genes within this block are represented on the Y chromosome as well as the X in man, but on the X chromosome only in other primates and, as far as we know, other mammals. There were two major events affecting the constitution of this block of homology. The first was a duplicative translocation from the X that occurred at approximately five million years ago. This event duplicated on the Y short-arm of an intermediate hominid species (perhaps Homo ergaster or Homo erectus) although the timing is somewhat early for these species). Subsequently, although this cannot be dated, a paracentric inversion on the Y short-arm took most of the transposed

49

The continuum of psychosis and its genetic basis a 11.3 11.2 11.1 11.1 11.21 11.22 11.23

22.3 22.2 21.3

11.3 12 11.1 11.1

k Y

13 21.1 21.3

a b c d e f g h i j k Figure 3.3.

Yp PAR1 RPS4Y/RPS4X ZFY/ZFX Distal Yp11.2/Xq21.3 AMGY/AMGX Proximal Yp11.2/Xq21.3 Yq11.21/Xp22.3 SMCY/SMCX-HY antigen Yq11.22/Xq28 Yq11.23/Xq22.2 Yq PAR2

22.2

28

X

Regions of homology between X and Y chromosomes. (Figure adapted from Affara et al., 1996.) The Xq21.3/Yp Homo sapiens-specific region of homology is shown in bold. Homologous regions are labeled a to k according to their order of representation on the Y chromosome. Yp PAR1: pseudo-autosomal region 1 on the Y short-arm; Yq PAR2: pseudo-autosomal region 2 on the Y long-arm; gene acronyms are given in capitals – thus ZFYZinc finger protein on the Y; AMCX  amelogenin on the X. Note that outside PAR1 and PAR2, regular recombination between X and Y does not take place. Note also that most regions of homology have arisen as a result of translocations (reduplications) from the X to the Y that can be dated in mammalian evolution (see Lambson et al., 1992).

block and reversed it in direction to give a large distal block and a smaller proximal block of homology on the Y chromosome short-arm of modern human males. These events are of potential evolutionary significance. They could relate to differences between man and chimpanzee but they are singular events, i. e., they occurred only once and must presumably have been selected in the hominid lineage. Because these events occurred on the Y chromosome it may be thought that

50

T. J. Crow

they are relevant only to characteristics expressed in males. This is not the case for two reasons: 1 The presence of a gene sequence on the Y chromosome affects its X activation status on the inactive X in females. According to the rule that genes that are present on the Y chromosome are protected from X inactivation, gene dosage is apparently preserved between males and females. The mechanism is unknown. 2 The presence of the gene on the Y chromosome may influence the sequence on the X, on the assumption that there is an interaction between the protein products of the two genes. In other words, the presence of the gene on the Y, which can now vary, constitutes a new environmental component which the gene on the X must adapt to. Particular interest therefore attaches to the gene content of the Xq21.3/Yp11 homologous region and to the changes that genes within this region have undergone in the course of hominid evolution. The sequence is now established in the Human Genome project. Three genes have been identified: a TG interacting factor (TGIFLX and TGIFLY), a polyadenine binding protein (PABPC-5) and Protocadherin X and its homologue Protocadherin Y (PCDHX and PCDHY). Of these three genes TGIFLX and Y is expressed mainly in the testis, PABPC-5 has been eliminated from the Y chromosome by a subsequent deletion, and the Protocadherin X and Y gene pair is of particular interest because it is expressed in the brain, including the germinal cell layer of the embryonic cortex, and appears to have been under changed selective pressure. This gene pair is a member of the class of protocadherins, cell surface adhesion molecules that are thought to play a role in the interaction between cells. They may act as axonal guidance factors and thus determine the pattern of connections between cells in the developing brain. Protocadherin genes have three major domains: an extracellular series of tandem repeats (in this case seven), a transmembrane domain and a cytoplasmic domain that presumably acts as an effector in influencing the response of the target neurone to the interaction at the cell surface. A number of nucleotide differences between the gene sequence on the Y and that on the X have been established (Crow and Williams, 2004; Williams and Crow, 2002; Williams et al., 2006). Some of these relate to changes in the Y sequence as might be expected for a duplicated gene, but some relate to changes in the X sequence relative to the Y, consistent with the possibility that these reflect responses to the new selective environment created by the translocation. In a comparison of the gene sequences of PCDHX and PCDHY in man, with PCDHX in chimpanzee, bonobo, gorilla, orang-utan and mouse, there is evidence that selective pressures on both the X ectodomain and the Y cytodomain have changed in the course of hominid evolution. Specifically there have been six amino-acid changes in the PCDHX sequence and 15 amino-acid changes in the Y

51

The continuum of psychosis and its genetic basis

sequence since the original translocation. These changes have thus occurred in the last five million years of hominid evolution by contrast with two, and possibly three, amino-acid changes that occurred in the X sequence in the 15 million years following the separation of the orang-utan from the lineage leading to chimpanzees, gorillas and hominids. What has to be explained is the increase in brain size in the hominid lineage that appears to have started at the point of transition from Australopithecus to Homo ergaster or some other early hominid species. Encephalization quotients (brain to body weight ratios) following this transition are generally between 3.5 for Homo ergaster through to 5.4 for Homo sapiens and 5.7 for Homo neanderthalensis, by contrast with figures of between 2.3 and 3.1 for the Australopithecines – figures which are comparable to those of the other great apes. In that Protocadherin X/Y is expressed in the brain, this is clearly a candidate for these relative increases in brain size although five million years is too early a date for the Australopithecus–hominid transition: the original duplicative transposition has now been dated to 6 million years. It is therefore a candidate for the chimpanzee-australopithecus separation (Williams et al., 2006). The cerebral torque and the evolution of language What has to be explained, of course, is the presence of language in Homo sapiens and its relationship to cerebral asymmetry (Crow, 2004a). Cerebral asymmetry, manifest as the torque from right frontal to left occipital across the anteroposterior axis of the brain, reflected in the asymmetry of the planum temporale (Geschwind and Levitsky, 1968), is present in Homo sapiens but apparently absent in chimpanzees and other primates (Crow, 2004b). Whether asymmetry was present earlier in the hominid series is unclear. If it was, Protocadherin X/Y may have played a role, but clearly something important happened in the transition from a precursor hominid species (perhaps Homo heidelbergensis) to Homo sapiens. At this point the capacity for symbolic representation appeared. What was the nature of this transition? Some change that accounts for innovations in both males and females is necessary. All of the above amino-acid changes took place in the earlier hominid lineages. They are not polymorphic in modern human populations (Giouzeli et al., 2004). The paracentric inversion is a candidate for this critical role. As the direction of the sequence on the Y is relevant to critical epigenetic modifications intrinsic to protection from inactivation of the sequences on the X (Burgoyne and McLaren, 1985; Crow, 1991) this inversion must be supposed to have had a significant influence on gene expression in both males and females. It is therefore a candidate to the sapiens speciation event (Crow, 2002; 2004c).

52

T. J. Crow

Conclusions Psychosis must be conceived as a continuum that includes typically schizophrenic, schizoaffective and affective psychoses. These syndromes appear to be present in all human populations at approximately the same incidence. Age of onset determines the form and outcome of psychosis and is earlier in males than in females, consistent with a developmental concept. Although no genes for psychosis have been established through linkage studies, the tendency to concordance by sex for psychosis within families that applies to both affective and schizophrenic illnesses suggests a gene that is present on both X and Y chromosomes. This possibility drew attention to the Xq21.3/Yp duplicative transposition that occurred at 3.5 million years ago to create a hominid-specific region of X/Y homology. Within the region the gene pair Protocadherin X and Protocadherin Y is expressed in the brain and may be relevant to human-specific characteristics such as cerebral asymmetry and language. In comparison with sequences in the great apes, this gene pair has been shown to be subject to new selective pressures in hominid evolution, including particularly variations in the Y cytodomain and the X ectodomain. As the presence of the gene on the Y chromosome also influences the epigenetic status of the gene sequences on the X, the establishment of the gene pair has set up a new source of epigenetic modulation. It is proposed that the paracentric inversion (which at present cannot be dated) was the speciation event that generated the variation in Homo sapiens that relates to language and also psychosis. R E F E R E N C ES Affara, N., Bishop, C., Brown, W. et al. (1996). Report of the second international workshop on Y chromosome mapping 1995. Cytogenetics and Cell Genetics, 73, 33–76. Angst, J., Frey, R., Lohmeyer, B. and Zerbin-Rudin, E. (1980). Bipolar manic-depressive psychoses: results of a genetic investigation. Human Genetics, 55, 237–54. Astrachan, B. M., Harrow, M., Adler, D. et al. (1972). A checklist for the diagnosis of schizophrenia. British Journal of Psychiatry, 121, 529–39. Badner, J. and Gershon, E. S. (2002). Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Molecular Psychiatry, 7, 405–11. Burgoyne, P. S. and McLaren, A. (1985). Does X-Y pairing during male meiosis protect the paired region of the X chromosome from subsequent X-inactivation? Human Genetics, 70, 82–3. Carpenter, W. T., Strauss, J. S. and Bartko, J. J. (1973). Flexible system for the diagnosis of schizophrenia: Report from the WHO International Pilot Study of Schizophrenia. Science, 182, 1275–8. Crichton-Browne, J. (1879). On the weight of the brain and its component parts in the insane. Brain, 2, 42–67. Crow, T. J. (1991). Protection from X inactivation. Nature, 353, 710.

53

The continuum of psychosis and its genetic basis Crow, T. J. (1994). Con: The demise of the Kraepelinian binary system as a prelude to genetic advance. In Gershon, E. S. and Cloninger, C. R. (eds.), Genetic Approaches to Mental Disorders, pp. 163–92. Washington: American Psychiatric Press. Crow, T. J. (1995). A continuum of psychosis, one human gene and not much else – the case for homogeneity. Schizophrenia Research, 17, 135–45. Crow, T. J. (1998). From Kraepelin to Kretschmer leavened by K. Schneider: the transition from categories of psychosis to dimensions of variation intrinsic to Homo sapiens. Archives of General Psychiatry, 55, 502–4. Crow, T. J. (2002). The Speciation of Modern Homo Sapiens. Oxford: Oxford University Press. Crow, T. J. (2004a). Cerebral asymmetry and the lateralization of language: core deficits in schizophrenia as pointers to the gene. Current Opinion in Psychiatry, 17, 97–106. Crow, T. J. (2004b). Directional asymmetry is the key to the origin of modern Homo sapiens (the Broca-Annett axiom). Response to Lesley Rogers. Laterality, 9, 233–42. Crow, T. J. (2004c). Discussion: The genetics of psychosis is the genetics of the speciation of Homo sapiens. In Gattaz, W. F. and. Hafner, H. (eds.), Search for the Causes of Schizophrenia, Vol. 5, pp. 297–315. Darmstadt: Steinkopff-Verlag. Crow, T. J., DeLisi, L. E. and Johnstone, E. C. (1989). Concordance by sex in sibling pairs with schizophrenia is paternally inherited. Evidence for a pseudoautosomal locus. British Journal of Psychiatry, 155, 92–7. Crow, T. J., DeLisi, L. E. and Johnstone, E. C. (1990). In reply . . . a locus closer to the telomere? British Journal of Psychiatry, 156, 416–20. Crow, T. J. and Williams, N. A. (2004). What happened to Protocadherin/X and Protocadherin/Y in hominid evolution. Schizophrenia Research, 67, 29–30. (Abstract.) DeLisi, L. E., Shaw, S., Crow, T. J. et al. (2002). A genome-wide scan for linkage to chromosomal regions in 382 sibling pairs with schizophrenia or schizoaffective disorder. American Journal of Psychiatry, 159, 803–12. Eaton, W. W., Mortenson, P. B., Herrman, H. et al. (1992). Long-term course of hospitalisation for schizophrenia: I. Risk for hospitalisation. Schizophrenia Bulletin, 18, 217–28. Endicott, J., Nee, J., Fleiss, J. et al. (1982). Diagnostic criteria for schizophrenia: reliabilities and agreement between systems. Archives of General Psychiatry, 39, 884–9. Gershon, E. S., DeLisi, L. E., Hamovit, J. et al. (1988). A controlled family study of chronic psychoses: Schizophrenia and schizo-affective disorder. Archives of General Psychiatry, 45, 328–36. Geschwind, N. and Levitsky, W. (1968). Human brain: left-right asymmetry in temporal speech region. Science, 161, 186–7. Giouzeli, M., Williams, N. A., Lonie, L., DeLisi, L. E. and Crow, T. J. (2004). ProtocadherinX/Y, a candidate gene-pair for schizophrenia and schizoaffective disorder: a DHPLC investigation of genomic sequence. American Journal of Medical Genetics, 129B, 1–9. Kendler, K. S., Karkowski, L. M. and Walsh, D. (1998). The structure of psychosis: latent class analysis of probands from the Roscommon Family Study. Archives of General Psychiatry, 55, 492–9. Kraepelin, E. (1920). Die Erscheinungsformen des Irreseins (translated by H. Marshall as: Patterns of mental disorder. In Hirsch, S. R. and Shepherd, M. [eds.], Themes and Variations in European Psychiatry. Wright, Bristol, pp. 7–30, 1974). Zeitschrift Gesamte Neurologie Psychiatrie, 62, 1–29.

54

T. J. Crow Lambson, B., Affara, N. A., Mitchell, M. and Ferguson-Smith, M. A. (1992). Evolution of DNA sequence homologies between the sex chromosomes in primate species. Genomics, 14, 1032–40. Lewis, C. M., Levinson, D. F., Wise, L. H. et al. (2003). Genome scan meta-analysis of schizophrenia and bipolar disorder II: schizophrenia. American Journal of Human Genetics, 73, 34–48. Maier, W., Lichtermann, D., Minges, J. et al. (1993). Continuity and discontinuity of affective disorders and schizophrenia. Archives of General Psychiatry, 50, 871–83. Marneros, A., Andreasen, N. C. and Tsuang, M. T. (1995). Psychotic Continuum. Berlin: SpringerVerlag. Miskolczy, D. (1933). Uber das anatomische Korrelat der Schizophrenie. Zeitschrift für Neurologie, 147, 509–44. Mott, F. W. (1910). Hereditary aspects of nervous and mental disease. British Medical Journal, 2, 1013–20. Parfitt, D. N. (1956). The neurology of schizophrenia. Journal of Mental Science, 102, 671–718. Pearlson, G. D., Barta, P. E., Schlaepfer, T. E. et al. (1995). Heteromodal association cortex in schizophrenia – Grey matter changes and clinical correlates. Schizophrenia Research, 15, 1–2. Penrose, L. S. (1942). Auxiliary genes for determining sex as contributory causes of mental illness. Journal of Mental Science, 88, 308–16. Penrose, L. S. (1991). Survey of cases of familial mental illness. European Archives of Psychiatry and Neurological Science, 240, 315–24. Rice, J., Reich, T., Andreasen, N. et al. (1987). The familial transmission of bipolar illness. Archives of General Psychiatry, 44, 441–7. Rosenthal, D. (1962). Familial concordance by sex with respect to schizophrenia. Psychological Bulletin, 59, 401–21. Rosenthal, D. (1970). Genetic Theory and Abnormal Behaviour. New York: McGraw-Hill. Segurado, R., Detera-Wadleigh, S. D., Levinson, D. F. et al. (2003). Genome scan meta-analysis of schizophrenia and bipolar disorder III: bipolar disorder. American Journal of Human Genetics, 73, 49–62. Southard, E. E. (1915). On the topographical distribution of cortex lesions and anomalies in dementia praecox, with some account of their functional significance. American Journal of Insanity, 71, 603–71. Stassen, H. H., Scharfetter, C., Winokur, G. and Angst, J. (1988). Familial syndrome patterns in schizophrenia, schizoaffective disorder, mania, and depression. European Archives of Psychiatry and Neurological Science, 237, 115–23. Taylor, M. A. and Abrams, R. (1978). The prevalence of schizophrenia: A reassessment using modern diagnostic criteria. American Journal of Psychiatry, 135, 945–8. Williams, N. A. and Crow, T. J. (2002). Protocadherin X/Y structure and sequence in Homo sapiens relative to other great apes. Schizophrenia Research, 53, 69. Williams, N. M., Norton, N., Williams, H. et al. (2003). A systematic genomewide-linkage study in 353 sib pairs with schizophrenia. American Journal of Human Genetics, 73, 1355–67. Williams, N. A., Clou, J. P., Giouzeli, M. and Crow, T. J. (in press). Accelerated evolution of Protocadherin II X/Y: a candidate gene pair for cerebral asymmetry and language. American Journal of Medical Genetics B.

4

Functional psychoses: molecular-genetic evidence for a continuum Hans H. Stassen, Christian Scharfetter and Jules Angst Psychiatrische Universitätsklinik Zürich

Background The functional psychoses “schizophrenia,” “schizoaffective disorders,” “psychotic depression,” “bipolar illness,” and “atypical psychosis” are distinct diagnostic entities sharing a considerable overlap in their underlying clinical syndromes (Stassen et al., 1988; Maier et al., 1993; 1999; 2005; Faraone et al., 1995; Maziade et al., 1995, 2001; Scharfetter and Stassen, 1995; Loftus et al., 1998; Wildenauer et al., 1999; Berrettini, 2000; Pulver et al., 2000; Vogt et al., 2000; Vuoristo et al., 2000; Bailer et al., 2002; Glatt et al., 2003; Kendler, 2003; Schurhoff et al., 2003; Craddock et al., 2005). Across ethnicities, schizophrenia and bipolar illness each affect about one percent of the general population, causing the loss of the ability to work, to have close relationships, and to have a fulfilling life. Population frequencies appear to be largely independent of ethnicity and social factors, as suggested by numerous epidemiologic studies including the World Health Organization (WHO) program on the outcome of severe mental disorders (Jablensky et al., 1992), and the Epidemiologic Catchment Area (ECA) program on psychiatric epidemiology (Weissmann et al., 1988). However, physical and sociocultural environments may modify the course of functional psychoses since patients in developing countries have a more benign course than patients in developed countries (Leff et al., 1992). Functional psychoses are characterized by a distinct loss of functional activities that particularly include the ability to experience a normal range of affective reactions, to reason and remember, to concentrate, to formulate ideas, to cope in a normal work environment, and to have a normal family life. Also, mortality is higher among patients suffering from functional psychoses. This increased mortality is primarily owing to a dramatically elevated suicide rate which is found to lie The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

55

56

Hans H. Stassen, Christian Scharfetter and Jules Angst

between 10% and 15% across subdiagnoses. Yet the increased mortality involves other causes of death as well, such as a higher rate of accidental death, or a reduced engagement in illness-preventing and health-promoting behaviors. Approximately one-third of cases show a deteriorating course of illness with chronic signs and symptoms. In consequence, functional psychoses are among the most debilitating and persistent diseases known to humankind. Little is known about the etiology of functional psychoses, and a diagnostic differentiation based on biological markers or objective laboratory methods is currently not available. Accordingly, it is hardly surprising that only incomplete treatments for functional psychoses are in use and that there is no cure in the majority of cases. Pharmacological treatment, though effective, is unspecific in the sense that (1) agents that differ greatly in their biochemical and pharmacological actions have virtually the same efficacy in terms of the proportion of patients in whom they induce therapeutic response; (2) a large proportion of patients (35%–45%) exhibit a refractory clinical picture which is resistant to all treatment modalities; and (3) the question of predicting treatment response in the individual patient is not answerable for any of the currently available antipsychotics and antidepressants. Evidence from twin, family and adoption studies suggests that, ultimately, genetic markers may represent the most important trait-like characteristics in the etiology of functional psychoses (Heston, 1966; Kety et al., 1971; 1978; Mendlewicz and Rainer, 1977; Cadoret, 1978; Gottesman and Shields, 1977; Cadoret et al., 1985; Wender et al., 1986; Kendler et al., 1985; Gottesman and Bertelsen, 1989; Stassen et al., 1999a, Weisbrod et al., 2004). However, genetic predisposition appears to vary across patients and to “explain” no more than 25% to 60% of the observed phenotypic variance, depending on the onset of illness, severity of illness, and long-term course of clinical syndromes. Therefore, genetic predisposition has been conceptualized as acting non-specifically, by elevating vulnerability which, however, is neither necessary nor sufficient for the development of functional psychoses (e.g., Zubin and Spring, 1977; Zubin et al., 1983). High comorbidity rates support a non-specific vulnerability model of functional psychoses, as over half of patients in psychiatric treatment typically receive more than one diagnosis. Particularly high among patients suffering from functional psychoses are the rates of alcoholism, nicotine dependence and substance abuse, with a dramatic increase of illegal drug use during the past decades. There is also considerable evidence from both treatment studies and longitudinal community studies that episodic comorbidity can complicate treatment and that lifetime comorbidity can lead to a more severe illness course (Boyd et al., 1984; Bukstein et al., 1989; DeMilio, 1989; Keitner et al., 1991; Kessler et al., 1994; Cassano et al., 1998). Substance abuse can create serious problems in the treatment of functional psychoses because continued use often interferes with therapy (Rabinowitz et al., 1998).

57

Functional psychoses: evidence for a continuum

Research in the field of functional psychoses, while primarily driven by the needs of patients, has also a socioeconomic dimension. With data on population prevalence and on the use of mental health services, it has become possible to estimate the costs of providing care to people with functional psychoses. The direct costs (payment for treatment and related services) are just one issue. More important than direct cost is the cost to society resulting from impairment in the individuals through absence from work, unemployment, and the need for social services. Additional cost to society arises from dealing with accidents, crime and other problems related to functional psychoses. These indirect costs have been estimated to be three times as high as the direct costs (McGuire, 1991; McCrone et al., 1998). Psychopathology: diagnostic systems Research in the field of functional psychoses requires the unambiguous identification of cases: that is, the definition of criteria that enable one to reproducibly identify individuals as suffering from these disorders. Lacking “objective” laboratory tests, such as chemical, anatomic, physiologic, or psychological tests, the discipline of psychopathology has its focus on establishing a complete clinical picture of the spectrum of psychiatric disorders in hospitalized patients, as well as in the general population, in order to search for characteristics that distinguish “true” cases from individuals with mild or subclinical states, to identify syndromes and subtypes that have differing etiologies or differing responses to treatment, and to establish diagnostic classification systems that are standardized and meet “objective” validation criteria for prognosis and therapy. The psychopathological description of psychiatric disorders involves specifying the phenomenology or symptomatology, premorbid history, age at onset, course of the disorder, family history, sociodemographic variables, and precipitating factors. The clinical description thus involves issues of content validity (e.g., what is the domain of symptoms and signs chosen to represent a disorder), as well as of criterion validity (e.g., what is the relation between premorbid history, age at onset, family history, or exogenous factors to the measure of a disorder). Conceptualizing and “measuring”a disorder after onset relies on a framework for assessing (1) duration of symptomatology; (2) remission with focus on speed and completeness; (3) time course with focus on fluctuations in symptomatology; (4) recurrence; and (5) outcome with focus on the consequences of the disorder, such as impairment or disability. Despite the progress in assessing the psychopathological picture of psychiatric disorders through structured interviews and standardized instruments, no generally accepted criteria of caseness are currently available, so that cross-comparisons between previous community studies and family studies are difficult to accomplish. In particular, current criteria for functional psychoses have an appreciable effect on

58

Hans H. Stassen, Christian Scharfetter and Jules Angst

the prevalence of these disorders in the general population, and on the pattern of these disorders within families. The latter point becomes a major problem for genetic studies if small changes in the criteria result in a significantly altered distribution of cases within families, thus biasing not only the distinction between sporadic and familial cases, but also the search for potential modes of transmission in segregation analyses. In fact, although abundant evidence from twin, family and adoption studies suggests the existence of a significant genetic component in the etiology of functional psychoses, segregation analyses of thousands of families worldwide have revealed nonMendelian inheritance which is difficult to model as the parameters underlying such models are extremely sensitive to misclassifications on the phenotype level. The situation is similar with diagnostic systems for psychiatric disorders, although the standards for diagnostic schemes have been established more than 30 years ago. These standards encompass (1) establishing the clinical description of the disorder with symptoms and signs, premorbid history, age at onset, course of the disorder, sociodemographic variables, and precipitating factors; (2) establishing the relationship to laboratory tests; (3) the use of family history, since many psychiatric disorders run in families; thus, an increased prevalence of the same clinical phenomenology may serve as an external validator of a diagnostic entity; (4) the relation of diagnostic entities to outcomes, including treatment response; in this context it is implicitly assumed that individuals of the same diagnostic entity have similar outcomes, or that certain diagnostic sub-groups have particularly poor or good outcomes; (5) the distinction from other diagnostic entities that may share certain signs and symptoms, laboratory test results, or outcomes. During the past decades much emphasis has been laid on the development of operational criteria for psychiatric diagnoses. Operational criteria involve a sort of clinical algorithm with a set of operations that are to be performed on signs and symptoms in a pre-specified way, thus leading to an explicit definition of disorders, even in cases of unknown pathology. Undoubtedly, this method of approach has greatly improved the reproducibility of diagnoses as well as the inter-rater reliability when classifying patients with major psychiatric disorders. On the other hand, reproducibility and reliability do not necessarily ensure validity. Indeed, for the individual patient, psychiatric diagnoses based on operational criteria do not offer an advantage over traditional diagnostic systems in the prediction of response to treatment, onset of improvement, or long-term prognosis. This fact has been acknowledged, for example, by the authors of DSM-IV when they stated “the specified diagnostic criteria for each mental disorder are offered as guidelines for making diagnoses, since it has been demonstrated that the use of such criteria enhances agreement among clinicians and researchers” (American Psychiatric Association, 1994).

59

Functional psychoses: evidence for a continuum

Another crucial point is the illness concept in psychiatry, which includes two complementary ways of thinking about the development towards illness. One way is that symptoms and signs are ubiquitously present in the general population, at a mild level of intensity, making it difficult to distinguish normal and subclinical forms from manifestations of illness. Then, an increase in severity or intensity of symptoms in an individual would eventually lead to this individual reaching the threshold for “illness.” The central methodological problem inherent to this “continuum approach” is the question of how to distinguish between normal fluctuations and significant changes in the severity or intensity of symptoms. The complementary way of thinking about the development towards illness is that new groups of symptoms occur where none existed. Then, the stepwise acquisition of symptoms by an individual towards a certain configuration of symptoms would eventually lead to this individual meeting the definitions for diagnosis. The central methodological problem inherent to this approach is the question of how to decide upon presence or absence of symptoms. A unification of the two complementary concepts may lead to a more natural model of psychiatric illness, yet at the cost of additional methodological complexity. In fact, measuring a concept of interest as accurately as possible involves not only a theoretical understanding of the concept with all its implications but also an empirical assessment of the criteria chosen to operationalize the concept. Given the apparent lack of externally valid concepts of psychiatric illnesses and of clear-cut, externally valid standards for establishing caseness through either psychopathological instruments or operationalized diagnostic criteria, the question arises as to the appropriateness of the traditional yes–no model of caseness for studying familial aggregations of psychiatric disorders and their distribution within the general population. Syndrome-oriented approaches to psychopathology avoid the yes–no dichotomy of diagnostic schemes by modeling the natural history and time course of psychic disorders through quantitative measures (“syndromes”), and may therefore be better suited to assess the fine gradations of the within- and between-family similarity of psychopathologic patterns that do not reach diagnostic threshold. Thus, syndrome-oriented approaches extend the qualitative nature of caseness definitions in such a way that the loss of information inherent to the decision-making algorithm, when deciding about presence or absence of a particular diagnosis, can be avoided. Psychopathology: syndrome-oriented approaches Functional psychoses, like most other complex illnesses where multiple genetic and non-genetic factors contribute to pathogenesis, appear to aggregate but not to segregate in families when cases are identified through psychiatric diagnoses. That is,

60

Hans H. Stassen, Christian Scharfetter and Jules Angst

the genetic predisposition to functional psychoses does not follow simple Mendelian modes of transmission, and signs and symptoms that underlie these diagnoses appear to be, by themselves, neither necessary nor sufficient for developing a particular disorder. Therefore, the utility of current diagnostic entities for genetic studies seems limited, especially since psychiatric diagnoses incompletely cover the patients’ prodromal phase, age of onset, severity of illness, long-term course, and impairment. And, what is most disappointing from the clinical point of view, psychiatric diagnoses do not offer much information about a patient’s response to treatment and prognosis (Parker, 1996; Stassen and Angst, 1998). Given these facts it is not really surprising that psychotropic drugs that differ widely in their biochemical and pharmacological actions display virtually the same efficacy in the proportion of patients in whom they induce a therapeutic response. Quantitative, syndrome-oriented approaches to psychopathology extend the DSM-IV or ICD-10 phenotype definitions by replacing the dichotomy of the diagnostic schema with dimensional quantities. These quantities allow one to model the “psychopathological similarity” between subjects and to assess intra-familial patterns of psychopathology that do not reach diagnostic thresholds (Faraone et al., 1995, 2004a; Maier et al., 1999; Kendler, 2003; O’Donovan et al., 2003; Schurhoff et al., 2003). Based on methodological developments derived from previous investigations into the familiality of psychopathology syndrome patterns, we introduced some 15 years ago a syndrome-oriented approach to psychopathology which includes 16 syndromes (Angst et al., 1988) and leads to a representation of psychic disorders via 16-dimensional, quantitative “syndrome vectors” that can be extracted, for example, from the Diagnostic Interview for Genetic Studies (DIGS) rating instruments (Stassen et al., 1988; Nurnberger et al., 1994). In fact, multivariate-feature vector methods, combined with pattern recognition techniques, give a high resolution picture of complex quantitative phenotypes and allow one to correlate multilocus deviations in the genome sequence with quantitative scores on the phenotype level. The Zurich family study In a prospective family study ascertained through 269 index cases suffering from functional psychoses, 1501 first- and second-degree family members, and a 20-year follow-up along with the assessment of 105 offspring of the index cases, we applied a multivariate syndrome-oriented approach to psychopathology in order to derive a quantitative measure of the severity of illness and to partition the population into “natural” sub-groups. Furthermore, our interest focused on: (1) the question of increasing severity of psychoses across generations along with a shift towards earlier onset (anticipation); (2) the question of differences in the severity of

61

Functional psychoses: evidence for a continuum

psychoses depending on whether the illness has been transmitted by the maternal or paternal side (genomic imprinting); and (3) the question of age-of-onset shifts towards earlier onset of psychoses in successive birth cohorts (secular trends). As to the question of “natural” sub-groups, structural analyses revealed clearly distinguishable sub-groups of patients, characterized by differences in the familial aggregation of syndromes and long-term outcome. These sub-groups were similarly found among the “affected” first-degree relatives (Stassen et al., 1988; Scharfetter and Stassen, 1995; Stassen et al., 1997; Scharfetter et al., 1999). Specifically, we carried out a multivariate cluster analysis based on the 16-dimensional, quantitative syndrome vectors derived from the 269 index cases with a diagnosis of functional psychosis, on the one hand, and from the 350 first-degree relatives who exhibited psychopathologic features in our quantitative SSCL-16 syndrome scores derived through the Schedule for Affective Disorders and Schizophrenia (SADS): Syndrome Checklists (Angst et al., 1988), on the other. We searched for “natural” groupings that showed up equally among the index cases and the affected first-degree relatives. The analysis revealed three “core” clusters which were similarly present in the two populations under investigation. These three clusters encompassed patients across diagnostic entities: that is, patients with a diagnosis of schizophrenia, schizoaffective disorder or bipolar illness. The striking similarity of quantitative syndrome patterns between index cases (n136) and affected first-degree relatives (n116), as revealed by the largest “core” cluster, is demonstrated in Figures 4.1a, b. The fact that the same syndrome patterns, derived through independent cluster analyses, showed up equally among index cases and affected first-degree relatives underline that quantitative, syndrome-oriented measures are well suited to evaluate the fine gradations of intrafamilial psychopathology beneath diagnostic thresholds. In consequence, use of syndrome-oriented measures can be regarded as a high-resolution method of assessing the complex quantitative phenotypes underlying functional psychoses, thus enabling direct correlation between multilocus deviations in the genome sequence and quantitative scores on the phenotype level. Evidence for a common neurobiological basis? Several previous investigations have suggested that the gene for the 7-nicotinic receptor may play a role in the pathogenesis of schizophrenia and may be responsible for heavy smoking among schizophrenic patients, which is threefold that observed with normal controls (Leonard et al., 1998, 2001; DeLeon et al., 2002; Lyons et al., 2002; Poirier et al., 2002; Simosky et al., 2002; Fergusson et al., 2003; Freedman et al., 2003; Gault et al., 2003; Tsoh et al., 2003; Vanable et al., 2003; AudrainMcGovern et al., 2004; DeLuca et al., 2004; Etter et al., 2004; Faraone et al., 2004b;

62

Hans H. Stassen, Christian Scharfetter and Jules Angst

Observations = 136

"Schizophrenic" cluster: index cases 0 125

0

91.9%

2 Hypochondriasis

28

0

20.6%

3 Compulsion/phobia

10

0

7.4%

4 Delusions/unsystematic (mc)

18

0

13.2%

5 Delusions/unsystematic (mic)

1 Schizophrenic thought disorder

132

0

97.1%

6 Delusions/systematic (mc)

0

0

0.0%

7 Delusions/systematic (mic)

2

0

1.5%

8 Hallucinations/acoustic (mc)

5

0

3.7%

9 Hallucinations/acoustic (mic)

104

0

76.5%

10 Hallucinations/optical (mc)

12

0

8.8%

11 Hallucinations/optical (mic)

44

0

32.4%

12 Hallucinations/other (mc)

6

0

4.4%

13 Hallucinations/other (mic)

64

0

47.1%

14 Disorder of ego-consciousness

110

0

80.9%

15 Depressive syndrome

102

0

75.0%

16 Manic-like syndrome

49

0

36.0%

134

0

98.5%

84

0

61.8%

128

0

94.1%

38

0

27.9%

0

0

0.0%

29

0

21.3%

17 Schizophrenic incongruent affect 18 Inhibited-stupurous 19 Agitated-excited 20 Attempted suicide 21 Amnestic psychosyndrome 22 Other symptoms

10

20

30

40

50

60

70

80

90

100

Figure 4.1a. Striking similarity of quantitative syndrome patterns between index cases (n = 136) and affected first-degree relatives (n = 116), as revealed through multivariate cluster analysis of SSCL-16 syndrome scores (here: index cases). (The quantifiers “mc” and “mic” distinguish mood congruent and mood-in congruent syndromes) Observations = 116

"Schizophrenic" cluster: first-degree relatives 0 107

0

92.2%

2 Hypochondriasis

5

0

4.3%

3 Compulsion/phobia

4

0

3.4%

4 Delusions/unsystematic (mc)

3

0

2.6%

5 Delusions/unsystematic (mic)

104

0

89.7%

6 Delusions/systematic (mc)

0

0

0.0%

7 Delusions/systematic (mic)

1

0

0.9%

8 Hallucinations/acoustic (mc)

4

0

3.4%

9 Hallucinations/acoustic (mic)

75

0

64.7%

10 Hallucinations/optical (mc)

5

0

4.3%

11 Hallucinations/optical (mic)

17

0

14.7%

12 Hallucinations/other (mc)

3

0

2.6%

13 Hallucinations/other (mic)

29

0

25.0%

14 Disorder of ego-consciousness

39

0

33.6%

15 Depressive syndrome

34

0

29.3%

16 Manic-like syndrome

13

0

11.2%

17 Schizophrenic incongruent affect

98

0

84.5%

18 Inhibited-stupurous

49

0

42.2%

19 Agitated-excited

99

0

85.3%

20 Attempted suicide

13

0

11.2%

2

0

1.7%

11

0

9.5%

1 Schizophrenic thought disorder

21 Amnestic psychosyndrome 22 Other symptoms

10

20

30

40

50

60

70

80

90

100

Figure 4.1b. Striking similarity of quantitative syndrome patterns between index cases (n = 136) and affected first-degree relatives (n = 116), as revealed through multivariate cluster analysis of SSCL-16 syndrome scores (here: affected first-degree relatives).(mc = mood congruent and mic = mood-in congruent syndromes)

63

Functional psychoses: evidence for a continuum Table 4.1. Diagnosis-dependent allele distribution of markers D15S1360 and L76630 among schizophrenics, bipolars and schizoaffectives as compared to healthy controls (“bp” designates base pairs).

D15S1360

Controls Schizophrenics Bipolars Schizoaffectives

L76630

n

112 bp

116 bp

174 bp

172 bp

129 50 48 29

22.0% 34.3% 41.7% 51.8%

27.6% 16.7% 17.7% 7.1%

26.6% 30.2% 35.1% 42.6%

18.0% 16.7% 10.6% 0.0%

Ripoll et al., 2004; Aguilar et al., 2005; DeLeon and Dias, 2005). In a study of 129 healthy controls and 127 schizophrenic, schizoaffective and bipolar patients, we aimed (1) to confirm the potential association between schizophrenia and the 7nicotinic receptor; (2) to test the diagnostic specificity of 7-receptor subunits with respect to psychiatric diagnoses; and (3) to investigate potential receptor differences between smokers and non-smokers in the general population. Our analysis included the two dinucleotide polymorphisms D15S1360 and L76630 that are localized on chromosome 15 in a genomic fragment containing the 7-nicotinic receptor gene CHRNA7. Highly significant differences (p 0.0001) between the allele distributions of patients and controls were detected for these two markers, with all three diagnostic sub-groups contributing to the discrimination. An independently ascertained replication sample of 24 patients confirmed this finding. Our results suggested a non-specific, 7-related vulnerability to functional psychoses which depended on the severity of overall psychopathology, with no clearcut boundary between clinical diagnoses. This vulnerability was lowest among schizophrenics, intermediate among bipolars, and highest among schizoaffectives (Tables 4.1, 4.2). For the healthy control subjects, we found no association between the 7-nicotinic receptor gene and tobacco consumption and, consequently, no evidence that the differences between patients and controls could be explained through a genetic predisposition to tobacco consumption, thus indicating that the differences between patients and controls are more than just a smoker–non-smoker distinction (Stassen et al., 2000; 2004a). Specifically, we found a highly significant correlation (p 0.001) between genotypes and severity of overall psychopathology as reflected by the co-occurrence of SSCL-16 syndromes. Subsequent discriminant analysis using both polymorphisms yielded 62% correctly classified controls and 69% correctly classified patients (Stassen et al., 2000). These results suggest that the above quantitative, syndromeoriented measures apparently possess a biological component which enables not

64

Hans H. Stassen, Christian Scharfetter and Jules Angst Table 4.2. Allele distributions of dinucleotide polymorphism D15S1360 across the populations of patients and controls, as well as the diagnostic subgroups within these populations (“bp” designates base pairs and “Patients (repl.)” denotes the patients of the replication sample).

Fragment sizes of dinucleotide D15S1360

Controls Patients Patients (repl.) Schizophrenics Bipolars Schizoaffectives Smokers Non-smokers

108bp

110bp

112bp

114bp

116bp

118bp

1/0.4% 0 0 0 0 0 0 1/0.6%

0 4/1.6% 0 1/1.0% 3/3.1% 0 0 0

56/22.0% 104/40.9% 7/15.9% 35/34.3% 40/41.7% 29/51.8% 15/25.0% 38/21.1%

122/48.0% 104/40.9% 26/59.1% 46/45.1% 35/36.5% 23/41.1% 24/40.0% 90/50.0%

70/27.6% 38/15.0% 11/25.0% 17/16.7% 17/17.7% 4/7.1% 20/33.3% 47/26.1%

5/2.0% 4/1.6% 0 3/2.9% 1/1.0% 0 1/1.7% 4/2.2%

only an external validation of the method of approach but also the estimation of the amount of phenotypic variance explained by a given configuration of genes. Ethnicity-independent vulnerability A considerable number of epidemiologic studies have suggested similar population frequencies of functional psychoses across ethnicities and social systems. Accordingly, functional psychoses have been hypothesized as representing evolutionary old disorders that are ubiquitously distributed in the general population worldwide. To test this hypothesis through a quantitative approach we carried out a moleculargenetic study of an ethnically diverse population including nuclear families of European, African-American and Non-African-American origin, each ascertained through index cases suffering from schizophrenic psychoses. Our quantitative approach to ethnicity was based on the genotypic differences in genetic similarity between individuals; that is, we evaluated high-dimensional feature vectors made up of the allelic information of 20 uncorrelated, polymorphic markers (“ethnicity markers”). The genetic similarity between first-degree relatives was then one half, and the genetic similarity between subjects of the same ethnic group was higher than that between subjects of different ethnic groups. Thus, it also became possible to structurally decompose the ethnically diverse population into genetically homogeneous sub-groups (furnishing the concept of “biological ethnicity”). Principal component analysis led to a representation of the multivariate feature vectors in such a way that the (orthogonal) axes of the transformed vector space optimally accounted for the variance of the underlying features (“genetic diversity”).

65

Functional psychoses: evidence for a continuum

In the next step, we addressed the question of ethnicity-related differences of genomic polymorphisms within the scope of our recent schizophrenia study of 77 nuclear African- and non-African-American families* (396 genotyped subjects; 10 cM (Centimorgan) genome scan with 430 marker loci) (Stassen et al., 2004a). Detailed analysis of the underlying allelic variability yielded moderate differences between the African-American (mean value 4.825.29; n141) and nonAfrican-American (mean value 5.89  5.67; n111) families in terms of the multitude of observed genotypes across genomic regions. The somewhat higher genetic variability among the non-African-Americans (cf. Collins-Schramm et al., 2002) underlined the role of the USA as a genetic “melting pot.” No such differences were observed for the genetically more homogeneous European (Swiss-German) population (Stassen et al., 1999b). Given the genetic differences between the African- and non-African-American samples, it is not surprising that these two populations could easily be distinguished by genetic feature vectors at an overall performance of70% (20 polymorphisms), whereas this discrimination failed when comparing the samples of Non-African-Americans and Europeans. The ability of genetic feature vectors to assess and quantify ethnic differences leads in a very natural way to the concept of “biological ethnicity” which allows one to distinguish between ethnicityindependent and ethnicity-specific factors. We addressed the question of ethnicity-independent susceptibility to schizophrenic psychoses by treating the African-American families as “training” samples, while the genetically different Non-African-American families served as independent “test” samples. We searched for a configuration of susceptibility loci for which the betweensibling genetic similarity in a sufficiently large number of affected sibling-pairs deviated from the expected values in the two populations under comparison. In fact, the between-sibling similarities in affected sibling-pairs are expected to deviate from the expected value of “0.5” if the genomic regions under investigation contain vulnerability genes – that is, affected sibs are expected to share the same alleles at a vulnerability locus. The reference value “0.5” derives from the parent–offspring similarities, which are always “0.5,” irrespective of the affection status of parents and offspring. The question of protective factors can be addressed analogously by

* Part of the data was provided by the National Institute of Mental Health (NIMH) Human Genetics Initiative, with genotypes having been cleaned according to Zurich standards. The NIMH data and biomaterials were collected in three projects that participated in the NIMH Schizophrenia Genetics Initiative. From 1991–1997, the principal investigators and coinvestigators were: Harvard University, Boston, MA; Ming T. Tsuang, M.D., Ph.D., D.Sc.; Stephen Faraone, Ph.D.; and John Pepple, Ph.D.; Washington University, St. Louis, MO; C. Robert Cloninger, M.D., Theodore Reich, M.D., and Dragan Svrakic, M.D.; Columbia University, New York, NY; Charles Kaufmann, M.D., Dolores Malaspina, M.D., and Jill Harkavy Friedman, Ph.D.

66

Hans H. Stassen, Christian Scharfetter and Jules Angst

evaluating the between-sibling genetic similarity in a sufficiently large number of unaffected sibling-pairs. Our novel multivariate genotype-to-phenotype search strategy was based on a genetic similarity function that allowed us to quantify the inter-individual genetic distances d(xi,xj) between allelic genotype patterns xi, xj of subjects i, j simultaneously at a given number, n, of loci l1, l2 . . . ln. We found 12 vulnerability loci, on chromosomes 1, 4, 5, 6, 13, 14, 18 and 20, that were reproducible across the two samples under comparison and explained 18% of the variance. Therefore, these loci are likely to constitute a first approach to an ethnicity-independent, oligogenic vulnerability model of functional psychoses, thus giving rise to universal targets for treatment. The elevated vulnerability appeared to be non-specific and to act in such a way that exogenous factors became more likely to trigger the onset of functional psychoses or psychiatric disorders in general. This view was particularly supported by the fact that some of the genomic regions identified through this study also showed up with other psychiatric and somatic conditions, such as alcohol dependence (Stassen et al., 2004b) and psoriasis (Miyaoka et al., 2000). The psoriasis finding is of special interest given previous reports of associations between schizophrenia and psoriasis vulgaris, as well as of olanzapine-related (Ascari-Raccagni et al., 2000), SSRI-related (Gupta and Guptat, 2001), and alcoholrelated (Wolf et al., 1999) psoriasis. In other words, our search for ethnicity-independent susceptibility factors to functional psychoses yielded strong support for the existence of such factors, yet indicated the presence of some additional, nonspecific vulnerability as well. Response to treatment Current treatments of psychiatric disorders are insufficient, mainly because the pathophysiologies involved are unknown and the mechanisms by which the various classes of psychotropic drugs alleviate symptoms are poorly understood. In consequence, it is currently impossible to make any predictions of whether or not a particular patient will respond to a particular treatment. Furthermore, comparisons of the “average” time course of improvement suggest modest differences in efficacy of active compounds over placebo, rarely exceeding a 20% increase in the number of patients responding to treatment. For example, only a relatively small proportion of patients suffering from Major Depressive Disorder (MDD) appear to benefit from active treatment, as revealed by recent meta-analyses of FDA data on 10 030 MDD patients treated with various types of antidepressants and placebo, where the symptom reduction was found to be 47.9% with investigational drugs, 47.5% with established active comparators, and 35.5% with placebo. The placebo–drug differences reached statistical significance in fewer than half of the

67

Functional psychoses: evidence for a continuum

underlying 52 studies (Khan et al., 2001a, 2002). Similarly, analyses of FDA data on 10 118 psychotic patients treated with new antipsychotics (n7630), established antipsychotics (n1851) and placebo (n637) showed that the mean change in the total Brief Psychiatric Rating Scale (BPRS) score was 17.3% or less after six weeks of treatment (Khan et al., 2001b). Antipsychotic response started in the first week of treatment and accumulated over time, with the greatest improvement occurring in the first treatment week (Agid et al., 2003). Our investigations into the time characteristics of psychotropic drug response (Stassen and Angst, 1998; Stassen et al., 1999c; Angst and Stassen, 2001; Stassen and Angst, 2002; Szegedi et al., 2003, 2005), encompassing 3045 patients treated with antidepressants, neuroleptics, and placebo have demonstrated that effective psychotropic drugs merely trigger and maintain conditions necessary for recovery, with virtually identical time courses of improvement under all active compounds and placebo across genders (Table 4.3). Differences in efficacy between active drugs, as well as between active drugs and placebo, were reflected only by the proportion of patients showing a therapeutic response. Specifically, our results suggested that patients are likely to possess a biological predisposition that controls the time characteristics of recovery but not the response to a particular treatment. Consequently, response to currently available psychotropic drug treatment appears, to a substantial part, determined by an unspecific “remission-genetic” component, which is largely independent of diagnostic entities. In fact, psychotropic drugs seem to be “polyvalent” in the sense that they induce a therapeutic response under various clinical indications. For example, antidepressants are also effective in anxiety disorders and neuroleptics are successfully used in the treatment of bipolar illness. If effective psychotropic drugs merely trigger and maintain conditions necessary for recovery, then psychiatric patients obviously have a biological, presumably genetically determined predisposition that controls the time characteristics of recovery to a major extent. The very similar, approximately normally distributed time points of onset of improvement – under all treatment modalities and with no gender differences – particularly support the existence of a genetic basis underlying these time characteristics. On the other hand, it appears quite unlikely that the respective distributions “result” from one single gene or a few major genes. For this reason, we have proposed an oligogenic configuration of genes with non-linear interactions, where no single genomic locus is, by itself, either necessary or sufficient for the development of the phenotype. Specifically, we hypothesize that response to psychotropic drugs is to a major part determined by an oligogenic configuration of genes that controls the time characteristics of recovery, which are independent of treatment modality and triggered by various exogenous and endogenous factors (“recovery-genetic” component).

68

Hans H. Stassen, Christian Scharfetter and Jules Angst Table 4.3. Mean time to sustained response to therapy

Patients meeting criterion of 50% baseline score reduction Sample size Imipramine Moclobemide Placebo Amitriptyline Fluoxetine Oxaprotiline Mirtazapine Paroxetine

509 1020 316 167 444 120 138 134

Number of patients/percentage 311 547 100 96 211 47 78 71

60.9% 53.1% 31.3% 57.5% 47.5% 39.2% 56.5% 53.0%

Day of response 18.0 7.9 18.9 8.3 20.010.0 21.311.4 20.2 8.3 23.311.6 19.412.2 23.911.6

Notes: Time to sustained response in the “average” patient under antidepressant and placebo treatment, as derived from an “individual-case” analysis using “true” assessment days rather than the “design days” of the study protocols. Sustained response was defined as a 50% baseline score reduction without subsequent deterioration beyond 15% of achieved improvement score. Differences between treatment modalities appear to relate to the proportion of patients in whom a therapeutic effect is induced, but not to the onset of effect.

This model contrasts with pharmacological hypotheses aimed at explaining response to a particular active treatment through single-gene approaches (Kirchheiner et al., 2004). A typical example of an unsuccessful attempt to directly link the effect of an antidepressant to its primary site of action was the monoamine oxidase-A (MAO-A) hypothesis of moclobemide, where a molecular-genetic study yielded no significant relation to the respective genes (Müller et al., 2002). Since response rates are very similar across ethnicities, the postulated “recovery-genetic” component can be assumed to be ethnicity-independent, with “ethnicity” being implicitly defined by “natural” groupings on the genotype level which are generated by the specifics of individual allele combinations at polymorphic genomic loci and give rise to the concept “biological ethnicity” (Stassen et al., 2003). In contrast to the “recovery-genetic” component, a much smaller, ethnicityspecific “pharmaco-genetic” component apparently causes the inter-individual variation in dose, blood level, weight gain, and other side effects (e.g., Brockmöller et al., 2002; Baumann et al., 2004; Correll and Malhotra, 2004; Ciusani et al., 2004; Grasmader et al., 2004; Kirchheiner et al., 2004). To test the biological relevance of the “recovery-genetic” approach to psychotropic drug response, we carried out a molecular-genetic pilot study of (1) 105 patients suffering from MDD and treated with antidepressants; and (2) 152 patients suffering from functional psychoses and treated with neuroleptics. We

69

Functional psychoses: evidence for a continuum

found an oligogenic configuration of 23 genomic loci (out of 72 candidate genes selected on the basis of some a-priori evidence from other studies) having the maximal correlation with the quantitative phenotype “onset of improvement” among the patients treated with antidepressants. Using day 14 as the boundary between “early” and “late” improvement, the genetic vector space enabled discrimination between “early” improvers, “late” improvers, and non-improvers at a rate of 75% correctly classified patients. No single genomic locus was either necessary or sufficient for having the phenotype, since exclusion from the configuration did not change a subject’s classification, but merely reduced the overall correlation with the phenotype. The same oligogenic configuration of 23 genomic loci displayed a significant correlation with onset of improvement among the 152 patients treated with neuroleptics as well. The overall rates of correctly classified patients were only slightly worse compared to the patients under antidepressants: 70% overall, 58.7% among non-improvers, 85.4% among early improvers, and 65.9% among late improvers. Yet, most interestingly, there was some overlap with genomic regions previously identified as being linked to vulnerability to functional psychoses. Conclusions Clearly, the strength of modern diagnostic systems in psychiatry lies in nosology; that is, in the systematic collection of knowledge and in the description and differentiation of the complex phenomena which become manifest as psychiatric disorders. In consequence, the well-established diagnostic systems ICD-10 and DSM-IV have achieved a high degree of perfectionism, thus representing state-ofthe-art instruments for the daily work of clinicians worldwide. Yet these diagnostic systems do not offer clinicians reliable guidelines for therapy and prognosis of a particular patient who suffers, for example, from functional psychosis. Only statistical information is available on psychotropic drug treatment, so that it is currently impossible to make any prediction of how a particular patient will respond to a particular antipsychotic or antidepressant therapy. The reason for this disappointing situation originates from the fact that psychiatric disorders are very heterogeneous and do not form distinct entities like most somatic illnesses, while the mechanisms of action of psychotropic drugs are likewise not well enough understood. Moreover, little is known about the etiology of psychiatric disorders in general, and of functional psychoses in particular. Even though strong evidence from adoption and twin studies underlines the significance of genetic contributions to the development of functional psychoses, the usefulness of psychiatric diagnoses in the context of phenotype definition is rather limited, as they delineate the familial aggregation of these disorders insufficiently and do not elucidate the modes of genetic transmission across generations.

70

Hans H. Stassen, Christian Scharfetter and Jules Angst

Multidimensional quantitative approaches to disentangling the complex processes that underlie psychiatric disorders, their development, acute phase, and further course, apparently offer some major advantages over the qualitative taxonomy of diagnostic systems. Empirical data from our studies carried out over the past two decades clearly support this view, as the chosen multidimensional approach to psychopathology based on quantitative syndromes has turned out (1) to resolve the fine gradations of within-family psychopathologies; (2) to possess a strong biological component as revealed by significant correlations between the severity of illness and molecular-genetic quantities; (3) to model genetic predisposition across ethnicities well, compared with existing epidemiologic data; (4) to be related to the time course of recovery under psychotropic drug treatment; and (5) to lead to a generalized model of genetic vulnerability to functional psychoses, where none of the vulnerability factors by itself is necessary or sufficient for the development of the disorder. In summary, our results suggest a continuum of psychopathologic features in a multidimensional space with no clear-cut boundaries between diagnostic entities, where the diverse facets of psychotic and affective syndromes are related to different regions. Several consequences for the current “best practice” treatment of patients, and for further research, lie at hand. If psychotropic drugs merely trigger and maintain, in a non-specific way, conditions necessary for recovery, as suggested by our studies on almost 3000 patients suffering from major depression, clinicians should especially pay attention – in the sense of current “best practice” – to early improvement as the indicator of later response to treatment in acute depression. Here, improvement means a sustained decrease of the patient’s psychopathology on a quantitative scale, even if the patient does not feel better. If this kind of decrease is observed, the antidepressant therapy should be continued. With no signs of improvement by the end of week 1, a higher dose of the current treatment is recommended. Clinicians have to be on the alert if a patient did not show any improvement during week 2: this patient should then either be treated using augmentation strategies or combination treatments, or be switched to another treatment, by no later than the end of week 3. Our recent investigation into the time characteristics of recovery under antipsychotic treatment suggested similarly non-specific, trigger-like mechanisms of action for antipsychotic drugs as well. The lack of specificity is especially apparent in our recent molecular-genetic results which revealed some overlap between the genetic factors related to recovery under antidepressants and those related to recovery under antipsychotics. Furthermore, our molecular-genetic results on vulnerability to functional psychoses suggest that significant non-linear interactions between the genomic loci play a key role, perhaps in a way described by the theory of dynamic systems encompassing many degrees of freedom, where self-regulation provides a high level

71

Functional psychoses: evidence for a continuum

of stability and error-tolerance against temporary, short-term, or even long-term “failure” of single factors, and “failures” are compensated through the joint action of other factors of the configuration. In terms of dynamic systems theory, the observed trigger effect of psychotropic drugs may act to restore and maintain selfregulation, a view which is well supported by clinical experience. Evidently, the theory of dynamic systems and self-regulation is compatible with current hypotheses of the pathogenesis of functional psychoses. Hypotheses, which assume complex disorders influenced by multiple genes as well as multiple non-genetic factors, where genetic predisposition “explains” 25% to 60% of the observed phenotypic variance, and where the amount of explained variance depends on the onset of illness, severity of illness, and long-term course of clinical syndromes. Yet the genetic analysis of complex disorders influenced by multiple genes as well as multiple non-genetic factors is still an unresolved methodological issue (Burghes et al., 2001). In fact, standard phenotype-to-genotype research strategies enable the localization of genes coding for traits influenced by a few major genes. Yet these strategies have not been successful in elucidating the genetic background of complex disorders. Thus, if (1) the contributions of single loci are small; (2) the single loci are, by themselves, neither necessary nor sufficient for developing the phenotype; (3) significant interactions between the loci are involved; and (4) there exist different pathways to the phenotype in ethnically diverse populations, detecting genes by these strategies is very difficult or impossible. Unknown population admixture, which is typical for European and US-American studies, can also substantially reduce the power of studies that aim to link phenotype to genotype (Hoffmann et al., 2000). For complex disorders a genotype-to-phenotype research strategy, as put forward by quantitative, syndrome-oriented approaches to psychopathology, with multilocus, interacting genes may be more successful. Such a strategy is based on oligogenic models and evaluates high-dimensional “feature vectors” of a genetic vector space with respect to between-population, within-population, and withinfamily similarities. The analysis of within-population similarities allows one to structurally decompose an ethnically diverse population and to address the question of ethnicity-independent susceptibility, while the analysis of within-family similarities allows one to detect deviations from the expected values in affected and unaffected sibling-pairs. Once an oligogenic model has been identified, the observed genotypic variation (e.g., in terms of genotype structure along with “clusters” in the genetic vector space) can be correlated with the phenotype of the disorder, using the quantitative syndrome-oriented measures. Accordingly, the genotype-to-phenotype approach does not critically depend on the reliability of a psychiatrist’s decision about the presence or absence of a diagnosis, as is the case for standard linkage and association analyses of the phenotype-to-genotype

72

Hans H. Stassen, Christian Scharfetter and Jules Angst

research strategy. Specifically, the genotype-to-phenotype strategy allows one to correlate multilocus deviations in the genome sequence with quantitative scores on the phenotype level, so that the subjects’ individual characteristics of the disorder, their lifestyles, and other environmental details can be included in the genetic model as well (Stassen et al., 1999b, 2004a).

Acknowledgements This article summarizes more than 20 years of research in the field of psychiatric genetics and genetic epidemiology. It includes results of previously published studies which have been funded in part by grants from the Swiss National Science Foundation.

References Agid, O., Kapur, S., Arenovich, T. and Zipursky, R. B. (2003). Delayed-onset hypothesis of antipsychotic action: a hypothesis tested and rejected. Archives of General Psychiatry, 60 (12), 1228–35. Aguilar, M. C., Gurpegui, M., Diaz, F. J. and de Leon, J. (2005). Nicotine dependence and symptoms in schizophrenia: naturalistic study of complex interactions British Journal of Psychiatry, 186, 215–21. Angst, J., Scharfetter, C., Stassen, H. H. and Winokur, G. (1988). Schedule for Affective Disorders and Schizophrenia (SADS): Syndrome Checklists SSCL-16 and Supplement. Psychiatric University Hospital Zurich (available on request from http://www.bli.unizh.ch/BLI/Subhome/ frboegen.html). Angst, J. and Stassen, H. H. (2001). Do antidepressants really taken several weeks to show effect? In Antidepressants, ed. B. E. Leonard. Basel: Birkhäuser, pp. 21–30. American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders DSM-IV. Washington, DC: American Psychiatric Association. Ascari-Raccagni, A., Baldari, U., Rossi, E. and Alessandrini, F. (2000). Exacerbation of chronic large-plaque psoriasis associated with olanzepine therapy. Journal of the European Academy of Dermatology and Venereology, 14, 315–16. Audrain-McGovern, J., Lerman, C., Wileyto, E. P., Rodriguez, D. and Shields, P. G. (2004). Interacting effects of genetic predisposition and depression on adolescent smoking progression. American Journal of Psychiatry, 161 (7), 12224–30. Bailer, U., Leisch, F., Meszaros, K. et al. (2002). Genome scan for susceptibility loci for schizoprenia and bipolar disorder. Biological Psychiatry, 52 (1), 40–52. Baumann, P., Hiemke, C., Ulrich, S. et al. (2004). Therapeutic monitoring of psychotropic drugs: an outline of the AGNP-TDM expert group consensus guideline. Therapeutic Drug Monitoring, 26 (2), 167–70. Berrettini, W. H. (2000). Are schizophrenic and bipolar disorders related? A review of family and molecular studies. Biological Psychiatry, 48 (6), 531–8.

73

Functional psychoses: evidence for a continuum Boyd, J. H., Burke, J. D., Gruenberg, E. et al. (1984). Exclusion criteria of DSM-III. A study of co-occurrence of hierarchy-free syndromes. Archives of General Psychiatry, 41, (10), 983–9. Brockmöller, J., Kirchheiner, J., Schmider, J. et al. (2002). The impact of the CYP2D6 polymorphism on haloperidol pharmacokinetics and on the outcome of haloperidol treatment. International Journal of Clinical Pharmacology and Therapeutics, 72 (4), 438–52. Bukstein, O. G., Brent, D. A. and Kaminer, Y. (1989). Comorbidity of substance abuse and other psychiatric disorders in adolescents. American Journal of Psychiatry, 146 (9), 1131–41. Burghes, A. H., Vaessin, H. E. and Chapelle de la, A. (2001). The land between Mendelian and multifactorial inheritance. Science, 293, 2213–14. Cadoret, R. J. (1978). Evidence for genetic inheritance of primary affective disorder in adoptees. American Journal of Psychiatry, 135 (4), 463–6. Cadoret, R. J., O’Gorman, T.W., Heywood, E. and Troughton, E. (1985). Genetic and environmental factors in major depression. Journal of Affective Disorders, 9 (2), 155–64. Cassano, G. B., Pini, S., Saettoni, M., Rucci, P. and Dell’Osso, L. (1998). Occurrence and clinical correlates of psychiatric comorbidity in patients with psychotic disorders. Journal of Clinical Psychiatry, 59 (2), 60–8. Ciusani, E., Zullino, D. F., Eap, C. B. et al. (2004). Combination therapy with venlafaxine and carbamazepine in depressive patients not responding to venlafaxine: pharmacokinetic and clinical aspects. Journal of Psychopharmacology, 18 (4), 559–66. Collins-Schramm, H. E., Kittles, R. A., Operario, D. J. et al. (2002). Markers that discriminate between European and African ancestry show limited variation within Africa. Human Genetics, 111, 566–9. Correll, C. U. and Malhotra, A. K. (2004). Pharmacogenetics and antipsychotic-induced weight gain. Psychopharmacology, 174 (4), 477–89. Craddock, N., O’Donovan M. C. and Owen M. J. (2005). The genetics of schizophrenia and bipolar disorder: dissecting psychosis. American Journal of Medical Genetics, 42 (3), 193–204. DeLeon, J., Becona, E., Gurpegui, M., Gonzalez-Pinto, A. and Diaz, F. J. (2002). The association between high nicotine dependence and severe mental illness may be consistent across countries, Journal of Clinical Psychiatry, 63 (9), 812–16. DeLeon, J. and Diaz, F. J. (2005). A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizoprenia Research, 76 (2–3), 135–57. DeLuca, V., Wong, A. H., Müller, D. J. et al. (2004). Evidence of association between smoking and alpha7 nicotinic receptor subunit gene in schizophrenia patients. Neuropsychopharmacology, 29 (8), 1522–6. DeMilio, L. (1989). Psychiatric syndromes in adolescent substance abusers. American Journal of Psychiatry, 146 (9), 1212–14. Etter, M., Mohr, S., Garin, C. and Etter, J. F. (2004). Stages of change in smokers with schizophrenia or schizoaffective disorder and in the general population. Schizophrenia Bulletin, 30 (2), 459–68. Faraone, S. V., Kremen, W. S., Lyons, M. J. et al. (1995). Diagnostic accuracy and linkage analysis: how useful are schizophrenia spectrum phenotypes? American Journal of Psychiatry, 152, 1286–90.

74

Hans H. Stassen, Christian Scharfetter and Jules Angst Faraone, S.V., Su, J., Taylor, L. et al. (2004b). A novel permutation testing method implicates sixteen nicotinic acetylcholine receptor genes as risk factors for smoking in schizophrenia families. Human Heredity, 57 (2), 59–68. Faraone, S. V., Su, J. and Tsuang, M. T. (2004a). A genome-wide scan of symptom dimensions in bipolar disorder pedigrees of adult probands. Journal of Affective Disorders, 82 Suppl. 1, S71–8. Fergusson, D. M., Goodwin, R. D. and Horwood, L. J. (2003). Major depression and cigarette smoking: results of a 21-year longitudinal study. Psychological Medicine, 33 (8), 1357–67. Freedman, R., Olincy, A., Ross, R. G. et al. (2003). The genetics of sensory gating deficits in schizophrenia. Current Psychiatry Reports, 5 (2), 155–61. Gault, J., Hopkins, J., Berger, R. et al. (2003). Comparison of polymorphisms in the alpha7 nicotinic receptor gene and its partial duplication in schizophrenic and control subjects. American Journal of Medical Genetics, 123 B(1), 39–49. Glatt, S. J., Faraone, S. V. and Tsuang, M. T. (2003). CAG-repeat length in exon 1 of KCNN3 does not influence risk for schizophrenia or bipolar disorder: a meta-analysis of association studies. American Journal of Medical Genetics, 121B (1), 14–20. Gottesman, I. I. and Bertelsen, A. (1989). Confirming unexpressed genotypes for schizophrenia. Risks in the offspring of Fischer’s Danish identical and fraternal discordant twins. Archives of General Psychiatry, 46 (10), 867–72. Gottesman, I. I. and Shields, J. (1977). Contributions of twin studies to perspectives on schizophrenia. In: Maher B.A. (ed) Contributions to the Psychopathology of Schizophrenia. New York: Academic Press, pp. 169–266. Grasmader, K., Verwohlt, P. L., Rietschel, M. et al. (2004). Impact of polymorphisms of cytochrome-P560 isoenzymes 2C9, 2C19 and 2D6 on plasma concentrations and clinical effects of antidepressants in a naturalistic clinical setting. European Journal of Clinical Pharmacology, 60 (5), 329–36. Guptaa, M. A. and Guptat, A. K. (2001). The use of antidepressant drugs in dermatology. Journal of the European Academy of Dermatology and Venereology, 15, 512–18. Heston L. L. (1966). Psychiatric disorders in foster home reared children of schizophrenic mothers. British Journal of Psychiatry, 112, 819–25. Hoffmann, K., Stassen, H. H. and Reis, A. (2000). Genkartierung in Isolatpopulationen. Medizinische Genetik¸12 (4), 428–37. Jablensky, A., Sartorius, N., Ernberg, G. et al. (1992). Schizophrenia: manifestations, incidence and course in different cultures. A World Health Organization ten-country study. Psychological Medicine Monograph Suppl. 20, 1–97. Keitner, G. I., Ryan, C. E., Miller, I. W., Kohn, R. and Epstein, N. B. (1991). 12-month outcome of patients with major depression and comorbid psychiatric or medical illness (compound depression). American Journal of Psychiatry, 148 (3), 345–50. Kendler, K. S. (2003). The genetics of schizophrenia: chromosomal deletions, attentional disturbances, and spectrum boundaries. American Journal of Psychiatry, 160 (9), 1549–53. Kendler, K. S., Masterson, C. C. and Davis, K. L. (1985). Psychiatric illness in first-degree relatives of patients with paranoid psychosis, schizophrenia, and medical illness. British Journal of Psychiatry, 147, 524–31.

75

Functional psychoses: evidence for a continuum Kessler, R. C., McGonagle, K. A., Zhao, S. et al. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry, 51 (1), 8–19. Kety, S. S. (1978). Schizophrenic illness in the families of schizophrenic adoptees: Findings from the Danish national sample. Schizophrenia Bulletin, 14, 217–22. Kety, S. S., Rosenthal, D., Wender, P. H. and Schulsinger, F. (1971). Mental illness in the biological and adoptive families of adopted schizophrenics. American Journal of Psychiatry, 128 (3), 302–6. Khan, A., Khan, S. and Brown, W. A. (2002). Are placebo controls necessary to test new antidepressants and anxiolytics? International Journal of Neuropsychopharmacology, 5 (3), 193–7. Khan, A., Khan, S. R., Leventhal, R. M. and Brown, W. A. (2001a). Symptom reduction and suicide risk in patients treated with placebo in antidepressant clinical trials: a replication analysis of the Food and Drug Administration Database. Interantional Journal of Neuropsychopharmacology, 4 (2), 113–18. Khan, A., Khan, S. R., Leventhal, R. M. and Brown, W. A. (2001b). Symptom reduction and sucide risk among patients treated with placebo in antipsychotic clinical trials: an analysis of the food and drug administration database. American Journal of Psychiatry, 158 (9), 1449–54. Kirchheiner, J., Nickchen, K., Bauer, M., Wong, M. L., Licinio, J., Roots, I. and Brockmoller, J. (2004). Pharmacogenetics of antidepressants and antipsychotics: the contribution of allelic variations to the phenotype of drug response. Molecular Psychiatry, 9 (5), 442–73. Leonard, S., Adler, L. E., Benhammou, K. et al. (2001). Smoking and mental illness. Pharmacology Biochemistry & Behavior, 70 (4), 561–70. Leonard, S., Gault, J., Adams, C., Breese, C. R., Rollins, Y., Adler, L. E., Olincy, A. and Freedman, R. (1998). Nicotinic receptors, smoking and schizophrenia. Restorative Neurology and Neuroscience, 12 (2–3), 195–201. Leonard, S., Gault, J., Hopkins, J. et al. (2002). Association of promoter variants in the alpha7 nicotinic acetylcholine receptor subunit gene with an inhibitory deficit found in schizophrenia Archives of Generla Psychiatry, 59 (12), 1085–96. Leff, J., Sartorious, N., Jablensky, A., Korten, A. and Ernberg, G. (1992). The International Pilot Study of Schizophrenia: five-year follow-up findings. Psychological Medicine, 22 (1), 131–45. Loftus, J., DeLisi, L. E. and Crow, T. J. (1998). Familial associations of subsyndromes of psychosis in affected sibling pairs with schizophrenia and schizoaffective disorder. Psychiatry Research, 80 (2), 101–11. Lyons, M. J., Bar, J. L., Kremen, W. S. et al. (2002). Nicotine and familial vulnerabiligy to schizophrenia; a discordant twin study. Journal of Abnormal Psychology, 111 (4), 687–93. Maier, W., Hofgen B., Zobel, A. and Rietschel M. (2005). Genetic models of schizophrenia and bipolar disorder: overlapping inheritance or discrete genotypes? European Archives of Psychiatry and Clinical Neuroscience, 255 (3), 159–66. Maier, W., Lichtermann, D., Miges, J. et al. (1993). Continuity and discontinuity of affective disorders and schizophrenia. Archives of General Psychiatry, 50, 871–83. Maier, W., Rietschel, M., Lichtermann, D. and Wildenauer, D. B. (1999). Family and genetic studies on the relationship of schizophrenia to affective disorders. European Archives of Psychiatry and Clinical Neuroscience, 249 Suppl. 4, 57–61. Maziade, M., Roy, M. A., Martinez, M. et al. (1995). Negative psychosis and disorganized

76

Hans H. Stassen, Christian Scharfetter and Jules Angst dimensions in patients with familial schizophrenia or bipolar disorder: continuity and discontinuity between the major psychoses. American Journal of Psychiatry, 152, 1458–63. Maziade, M., Roy, M. A., Rouillard, E. et al. (2001). A search for specific and common susceptibility loci for schizophrenia and bipolar disorder : a linkage study in 13 target chromosomes. Molecular Psychiatry, 6 (6), 684–93. McCrone, P., Thornicroft, G., Parkman, S., Nathaniel-James, D. and Ojurongbe, W. (1998). Predictors of mental health service costs for representative cases of psychosis in south London. Psychological Medicine, 28 (1), 159–64. McGuire, T. G. (1991). Measuring the economic costs of schizophrenia. Schizophrenia Bulletin, 17 (3), 375–88. Mendlewicz, J. and Rainer, J. D. (1977). Adoption study supporting genetic transmission in manic-depressive illness. Nature, 268 (5618), 327–9. Miyaoka, T., Seno, H., Inagaki, T. et al. (2000). Schizophrenia associated with psoriasis vulgaris: three case reports. Schizophrenia Research, 41, 383–6. Müller, D. J., Schulze, T. G., Macciardi, F. et al. (2002). Moclobemide response in depressed patients: association study with a functional polymorphism in the monoamine oxidase A promoter. Pharmacopsychiatry, 35 (4), 157–8. Nurnberger, J. I., Blehar, M. C., Kaufmann, C. A. et al. (1994). Diagnostic interview for genetic studies. Rationale, unique features, and training. NIMH Genetics Initiative. Archives of General Psychiatry, 51 (11), 849–59. O’Donovan, M. C., Williams, N. M. and Owen, M. J. (2003). Recent advances in the genetics of schizophrenia. Human Molecular Genetics, 12 Suppl. 2, R125–33. Parker, G. (1996). On brightening up: triggers and trajectories to recovery from depression. British Journal of Psychiatry, 168, 263–4. Poirier, M. F., Canceil, O., Bayle, F. et al. (2002). Prevalence of smoking in psychiatric patients. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 26 (3), 529–37. Pulver, A. E., Mulle, J., Nestadt, G. et al. (2000). Genetic heterogeneity in schizophrenia: stratification of genome scan data using co-segregating related phenotypes. Molecular Psychiatry, 5(6), 650–3. Rabinowitz, J., Bromet, E. J., Lavelle, J. et al. (1998). Prevalence and severity of substance use disorders and onset of psychosis in first-admission psychotic patients. Psychological Medicine, 28 (6), 1411–19. Ripoll, N., Bronnec, M. and Bourin, M. (2004). Nicotinic receptors and schizophrenia. Current Medical Research and Opinion, 20 (7), 1057–74. Scharfetter, C., Ernst, C., Guggenbühl, s. and Stassen, H. H. (1999). Syndrome-oriented approaches to resolving phenotypic heterogeneity in functional psychoses. Molecular Psychiatry, 4, Suppl. 1, 27. Scharfetter, C. and Stassen, H. H. (1995). Psychopathological concepts. Psychopathology, 28, Suppl. 1, 8–12. Schurhoff, F., Szoke, A., Meary, A. et al. (2003). Familial aggregation of delusional proneness in schizophrenia and bipolar pedigrees. American Journal of Psychiatry, 160 (7), 1313–19. Simosky, J. K., Stevens, K. E. and Freedman, R. (2002). Nicotinic agonists and psychosis. Current Drug Targets – CNS and Neurological Disorders, 1 (2), 149–62.

77

Functional psychoses: evidence for a continuum Stassen, H. H. and Angst, J. (1998). Delayed onset of action of antidepressants: fact or fiction? CNS Drugs, 9 (3), 177–84. Stassen, H. H. and Angst, J. (2002). Wirkung und Wirkungseintritt in der AntidepressivaBehandlung. In Therapie der affektiven Störungen, ed. H. Böker and D. Hell. Stuttgart, New York: Schattauer, pp. 141–65. Stassen, H. H., Begleiter, H., Beirut, L. et al. (2004b). Oligogenic approaches to the predisposition of alcohol dependence. A genome-wide search on 255 families. Neurological Psychiatry & Brain Research, 11, 13–22. Stassen, H. H., Begleiter, H., Porjesz, B. et al. (1999b). Structural decomposition of genetic diversity in families with alcohol dependence. In Goldin, L., Amos, C. I., Chase, G. A. et al. (eds.) Genetic Analysis Workshop 11: Analysis of genetic and environmental factors in common diseases. Genetic Epidemiology, 17, Suppl. 325–30. Stassen, H. H., Bridler, R., Hägele, S. et al. (2000). Schizophrenia and smoking: evidence for a common neurobiological basis? American Journal of Medical Genetics – Neuropsychiatric Genetics, 96, 173–7. Stassen, H. H. Bridler, R., Hell, D., Weisbrod, M. and Scharfetter, C. (2004a). Ethnicityindependent genetic basis of functional psychoses. A Genotype-to-phenotype approach. American Journal of Medical Genetics – Neuropsychiatric Genetics, 124, 101–12. Stassen, H. H., Coppola, R., Torrey, E. F. et al. (1999a). EEG differences in monozygotic twins discordant and concordant for schizophrenia. Psychophysiology, 36 (1), 109–17. Stassen, H. H., Angst, J. and Delini-Stula, A. (1999c). Fluoxetine versus moclobemide: crosscomparison between the time courses of improvement. Pharmacopsychiatry, 32, 56–60. Stassen, H. H., Hoffmann, K. and Scharfetter, C. (2003). Similarity by state/descent and genetic vector spaces: Analysis of a longitudinal family study. In Almasy, L., Amos, C. I., Bailey-Wilson, J. E. et al. (eds.). Genetic Analysis Workshop 13: Analysis of longitudinal family data for complex diseases and related risk factors. BMC Genetics, 4, S59, 1–6. Stassen, H. H., Ragaz, M. and Reich, T. (1997). Age-of-onset or age-cohort changes in the lifetime occurrence of depression? Psychiatiatric Genetics, 7, 27–34. Stassen, H. H., Scharfetter, C., Winokur, G. and Angst, J. (1988). Familial syndrome patterns in schizophrenia, schizoaffective disorder, mania, and depression. European Archives of Psychiatry and Clinical Neuroscience, 237, 115–23. Szegedi, A., Muller, M. J., Anghelescu, I. et al. (2003). Early improvement under mirtazapine and paroxetine predicts later stable response and remission with high sensitivity in patients with major depression. The Journal of Clinical Psychiatry, 64 (4), 413–20. Szegedi, A., Rujescu, D., Tadic, A. et al. (2005). The catechol-O-methyltransferase Va1108/158Met-polymorphism affects short-term treatment response to mirtazapine, but not to paroxetien in Major Depression. The Pharmacogenomics Journal, 5 (1), 49–53. Tsoh, J. Y., Lam, J. N., Delucchi, K. L. and Hall, S. M. (2003). Smoking and depression in Chinese Americans. American Journal of The Medical Sciences, 326 (4), 187–91. Vanable, P. A., Carey, M. P., Carey, K. B. and Maisto, S. A. (2003). Smoking among psychiatric outpatients: relationship to substance use, diagnosis, and illness severity. Psychology of Addictive Behaviors, 17 (4), 259–65. Vogt, I. R., Shimron-Abarbanell, D., Neidt, H. et al. (2000). Investigation of the human serotonin

78

Hans H. Stassen, Christian Scharfetter and Jules Angst 6 [5-HT6] receptor gene in bipolar affective disorder and schizophrenia. American Journal of Medical Genetics, 96 (2), 217–21. Vuoristo, J. T., Berrettini, W. H., Overhauer, J. et al. (2000). Sequence and genomic organization of the human G-protein Golfalpha gene (GNAL) on chromosome 18p11, a susceptibility region for bipolar disorder and schizophrenia. Molecular Psychiatry, 5 (5), 495–501. Weisbrod, M., Hill, H., Sauer, H. et al. (2004). Nongenetic pathologic developments of brainwave patterns in monozygotic twins discordant and concordant for schizophrenia. American Journal of Medical Genetics, 125, 1–9. Weissman, M. M., Leaf, P. J., Tischler, G. L. et al. (1988). Affective disorders in five United States communities. Psychological Medicine, 18 (1), 141–53. Wender, P. H., Kety, S. S., Rosenthal, D. et al. (1986). Psychiatric disorders in the biological and adoptive families of adopted individuals with affective disorders. Archives of General Psychiatry, 43, 923–9. Wildenauer, D. B., Schwab, S. G., Maier, W. and Detera-Wadleigh, S. D. (1999). Do schizophrenia and affective disorder share susceptibility genes? Schizophrenia Research, 39 (2), 107–11. Wolf, R., Wolf, D. and Ruocco, V. (1999). Alcohol intake and psoriasis. Clinics in Dermatology, 17, 423–30. Zubin, J. and Spring, B. (1977). Vulnerability – a new view of schizophrenia. Journal of Abnormal Psychology, 86 (2), 103–26. Zubin, J., Magaziner, J. and Steinhauer, S. R. (1983). The metamorphosis of schizophrenia: from chronicity to vulnerability. Psychological Medicine, 13 (3), 551–71.

5

State- and trait-related deficits in sustained attention in bipolar disorder: are there any overlaps with schizophrenia? Luke Clark Department of Experimental Psychology, University of Cambridge

Introduction The manic and depressed states of bipolar disorder have a clear impact upon cognitive function (Clark and Sahakian, 2006). Cognitive deficits are also widely regarded to represent a core feature of schizophrenic illness (Green et al., 2004). Neuropsychological testing aims to characterize the domains of cognitive impairment and cognitive sparing in a disorder, with the use of standardized quantitative assessment (Lezak, 1995). In understanding the neuropsychology of psychiatric disorders, it is useful to distinguish state and trait markers of impairment (Nuechterlein et al., 1992; Clark and Goodwin, 2004). A state marker is manifested during acute episodes (such as depression) and recovers on symptom remission. A state-related deficit is likely to correlate with symptom ratings (e.g., depression ratings on the Hamilton scale). A trait marker, on the other hand, persists during periods of symptom remission and is therefore more likely to be associated with the underlying neuropathology of a disorder. A trait marker may be completely stable across phases of illness, or may be state-modulated, whereby the deficit persists in the absence of symptoms but is exacerbated during acute episodes (Nuechterlein et al., 1992). According to the traditional Kraepelinian view of psychosis, bipolar disorder and schizophrenia represent categorically distinct disease entities, distinguished by the chronic, deteriorating course of schizophrenia contrasting with the episodic, fluctuating course of bipolar disorder (Kraepelin, 1899). Patients with bipolar disorder were thought to recover fully in the periods of neutral mood (known as euthymia) between episodes. This categorical view is challenged by a number of practical observations. First, many psychotic patients present with a combination The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

79

80

Luke Clark

of schizoaffective symptoms that refuse to be segregated into one of the two core diagnoses (Kendell and Gourlay, 1970). Second, patterns of familial aggregation do not indicate pure transmission of either diagnosis. Instead, rates of bipolar disorder are increased in the families of schizophrenic probands, and vice versa (Taylor, 1992). A dimensional model places the pure diagnoses of schizophrenia and bipolar disorder at either end of a continuum of psychosis, where the underlying etiological mechanisms are associated with psychosis as a whole (Crow, 1986). The symptomatic expression of this underlying pathology will be shaped by environmental and developmental processes, so that the majority of patients will lie at some point between the two extreme diagnoses, showing varying proportions of bipolar and schizophrenic symptoms. The dimensional view is represented (to some extent) in the current DSM-IV classification by the number of intermediary diagnoses lying between bipolar I disorder and chronic schizophrenia, although the DSM-IV system remains fundamentally categorical by nature. The nosological debate between categorical and dimensional models of psychosis has continued for over a century, but recent developments in neuroscientific research methods (including neuropsychological testing, alongside other techniques like structural and functional neuroimaging) have offered to shed new light on the dispute. Recent neuropsychological findings in bipolar disorder challenge one of the central tenets of the categorical model – that bipolar patients recover completely between episodes. Using sensitive measures of cognitive function, persisting trait deficits have been demonstrated in bipolar disorder patients tested during the euthymic phase (van Gorp et al., 1998; Ferrier et al., 1999; Clark et al., 2002). One domain of cognitive function that has been thoroughly researched in bipolar disorder is sustained attention (Clark and Goodwin, 2004). The purpose of this chapter is to review studies of sustained attention in bipolar disorder and schizophrenia, to inform the categorical–dimensional dispute. If the profile of deficit in bipolar disorder and schizophrenia was qualitatively distinct, this would support a categorical model. However, if this marker for bipolar disorder was shown to overlap with the cognitive dysfunction in schizophrenia, it would implicate overlapping etiological mechanisms in the two diagnoses. Sustained attention The recent National Institute of Mental Health – Measurement and Treatment Research to Improve Cognition in Schizophrenia (NIMH-MATRICS) review committee (Green et al., 2004), a team of experts in the field of neuropsychological function in schizophrenia, concluded that there were seven key domains of cognitive dysfunction in schizophrenia: working memory, attention/vigilance, verbal learning and memory, visual learning and memory, reasoning and problem

81

Bipolar disorder: any overlaps with schizophrenia?

solving, speed of processing, and social cognition. Interest in these areas of cognition is developing rapidly because they represent important potential targets for pharmacological intervention, and several domains are thought to be associated with functional outcome. Neuropsychological testing offers a quick and noninvasive research technique that can be administered without extensive training, in a clinic or outpatient setting. In comparison to schizophrenia, much less is known about cognitive function in bipolar disorder. A number of recent studies have investigated the domain of attention and vigilance (hereafter subsumed by the term sustained attention) in patients with bipolar disorder (see Table 5.1), and it is this domain that the current review will focus on. Sustained attention enables the focused processing of information from a single source, over an extended period of time that is usually several minutes. It is akin to the lay concept of concentration, and psychologists may refer to the process as vigilance. Research in human attention has proposed a specialized neural mechanism responsible for the sustenance of attention. This circuitry is thought to interact with distinct mechanisms for selective attention, shifting attention, and divided attention (Mesulam, 1981; Posner and Petersen, 1990; Desimone and Duncan, 1995). Sustained attention is assessed in the laboratory using variants of the Continuous Performance Test (CPT). In a precursor to the CPT, subjects were asked to monitor a clockface for double movements of the second hand (“respond whenever the second hand skips a beat”) (Mackworth, 1948). Performance on tests of this type reliably deteriorates over an extended period, a phenomenon known as the vigilance decrement. In CPTs, the subject is presented with a continuous stream of stimuli (e.g., letters or digits, presented at the rapid rate of one per second), which must be monitored for targets specified by the experimenter. Test duration is usually around ten minutes. In the original X-CPT (Rosvold et al., 1956), the subject was asked to respond whenever the letter X was presented. In the more difficult AX version, the subject responded only if the X was preceded by the letter A (Rosvold et al., 1956). In another common variant, the Identical Pairs task (CPT-IP) (Cornblatt et al., 1989), subjects must respond when any two consecutive stimuli are the same. In the Rapid Visual Information Processing (RVIP) task (Sahakian et al., 1989; Coull et al., 1996), subjects must respond to pre-specified digit sequences (e.g., 3–5–7) (which are displayed at the side of the screen to limit memory load). A shared feature across all variants is that targets are infrequent and unpredictable. Several variables are typically extracted in a CPT analysis. The key index of sustained attention is the number (or proportion) of targets correctly detected (“hits”). The mean response latency on these correct responses is recorded as an index of psychomotor speed. The number of false responses to non-targets (“false alarms”) is also reported, and the hit and false alarm rates are frequently converted using signal detection analysis, to derive the independent measures of target

82

Luke Clark

sensitivity (d’ or A’) and response bias (ln  or B’’) (Green and Swets, 1966; Grier, 1971). Target sensitivity provides an index of the subjects’ ability to discriminate a target from a non-target. Response bias indicates the likelihood of the subject making a response regardless of whether a target or non-target is presented. Response bias is traditionally associated with the level of motivation and fatigue, and in cases where the rate of false responding is very high, it is likely to indicate impulsivity. Optimal performance on a CPT requires the subject to maintain attention on a monotonous task for an extended period of time, resisting distraction from external stimuli in the environment and internal thoughts. An adequate level of arousal is necessary (Parasuraman, 1984; 1998), associated with ascending neurotransmitter projections from subcortical structures to the cortex (Robbins, 1998). Rapid analysis of sensory information is required, which in the case of a visual CPT places demands on posterior visual processing streams. Furthermore, components of executive control are required in order to maintain pre-specified targets in working memory, and to inhibit responding to non-targets and extraneous stimuli competing for neural resources (Manly and Robertson, 1997; Braver et al., 2002). These executive processes are typically associated with prefrontal cortex. Given the complex interaction of arousal, perceptual and executive processes in governing CPT performance, sustained attention deficits may arise from multiple sources. The demands on these components vary across the existing CPT variants, and can be manipulated within a task. The perceptual load can be increased by degrading (blurring) the visual stimuli with the introduction of white noise (the degraded-stimulus [DS] CPT) (Nuechterlein, 1983), or by increasing the number of stimuli presented on each trial (as in the Span of Apprehension task [Asarnow and MacCrimmon, 1981]). Working memory load can be increased or manipulated in a task like the RVIP by varying the number of target sequences to be attended, or in an identical-pairs task, by increasing the inter-trial interval (Cohen and Servan-Schreiber, 1992). The demands for inhibitory control can be manipulated by increasing target frequency and therefore response prepotency (Elvevag et al., 2000), or by inserting “catch” trials where a nontarget that closely resembles a target is occasionally presented (Swann et al., 2003). The effects of distraction can also be examined by providing irrelevant stimulation during the task (Cornblatt et al., 1989). Sustained attention deficits in schizophrenia

Reduced CPT target detection is one of the most robust cognitive deficits associated with the schizophrenic syndrome (Cornblatt et al., 1989; Cornblatt and Malhotra, 2001). Latency of responding to targets is reliably increased (Fleck et al., 2001), with less consistent effects on false responding or response bias (Rund et al., 1992; Cornblatt and Keilp, 1994). Studies in schizophrenia do not usually report an

83

Bipolar disorder: any overlaps with schizophrenia?

exaggerated vigilance decrement (Nuechterlein, 1983; Cornblatt et al., 1989; Addington and Addington, 1997); rather, target detection seems to be impaired from the start of the task. Relationship to symptoms

Sustained attention deficits in schizophrenia are impervious to symptom status. Several early studies showed that CPT deficits were clearly detectable in remitted patients (Wohlberg and Kornetsky, 1973; Asarnow and MacCrimmon, 1978). Nuechterlein et al. (1992) tested 17 schizophrenic patients on the DS-CPT, in the florid psychotic state and retested patients on symptom remission. In comparison to matched controls also tested on two occasions, the schizophrenic group was significantly impaired in both test phases and showed only a slight improvement on remission, which was not statistically significant. This finding was replicated by Addington and Addington (1997) using the same task in 59 patients retested at three months, and by Liu et al. (2002) in 41 patients tested on hospital admission and discharge. The improvement on retest was not statistically significant in any of the three studies, indicating that sustained attention deficit is a stable trait marker in schizophrenia. Other studies have concentrated more specifically on the relationship between sustained attention performance and medication status. Initial cross-sectional studies comparing medicated and unmedicated patient groups revealed conflicting results (Orzack and Kornetsky, 1971; Harvey et al., 1990; Earle-Boyer et al., 1991; Nestor et al., 1991), although it is clearly difficult to match the groups precisely. Longitudinal studies have tended to indicate minimal improvement with medication. For example, Cornblatt et al. (1999) found no improvement on the CPT-IP between the acute phase of hospitalization and after four weeks of treatment when symptoms had stabilized. These studies have mainly examined medication with typical antipsychotics, although Liu et al. (2000) reported a similar pattern in a 12-week trial with groups randomly assigned to receive haloperidol (a dopamine D2 antagonist) or risperidone (an atypical antipsychotic with combined dopamine and serotonin antagonism). Neither group of patients (n19 per group) showed significant CPT improvements over the course of the study, despite a significant improvement on clinical rating scales. Normative data indicated the schizophrenic patients fell 3–4 standard deviations below controls at all test points. Several studies have also reported the lack of any significant correlation between CPT target detection and symptom ratings, either in acute or remitted cases (Cornblatt et al., 1998; Liu et al., 2000; Liu et al., 2002). In each of these studies, positive and negative symptom ratings improved significantly over the duration of the experiment. In one study, Addington and Addington (1997) reported a modest association with negative but not positive symptom scores at follow-up. As a caveat

84

Luke Clark

of these longitudinal experiments, the order of testing could not be counterbalanced; all patients were less symptomatic on the second session. The effect of practice on the second session may confound within group comparisons. It seems unlikely, however, that practice effects could contribute to this pattern of results, given that patients showed no significant improvement. In conclusion, sustained attention deficit in schizophrenia appears to represent a stable trait marker that is unrelated to symptom status. Relationship to the disorder

A trait impairment need not be associated with the diagnosis per se: persisting cognitive deficits could plausibly arise from chronic hospitalization, medication, or as a neurotoxic effect of an enduring mental illness. The association between schizophrenia and sustained attention deficit is strengthened by the observation that it is apparent early in the course of schizophrenia, including patients at first episode (Albus et al., 1996; Cornblatt et al., 1998). Like chronic patients, first episode cases also failed to show CPT improvements with neuroleptic treatment (Cornblatt et al., 1998). Sustained attention deficit has also been detected in groups of subjects at high risk of developing schizophrenia, including the children of schizophrenic probands (Rutschmann et al., 1977; Cornblatt and Erlenmeyer-Kimling, 1985) and the adult first-degree relatives (i.e., siblings and parents) of schizophrenic probands (Franke et al., 1994; Chen et al., 1998; Laurent et al., 1999). The target detection scores in the groups of relatives typically lie at an intermediate point between the probands with schizophrenia and healthy controls with no family history. Given that these relatives share on average 50% of their genes with the proband, this is to be expected. As a result, the effect size of any group difference is smaller, and larger group sizes are necessary to detect statistically significant effects in high-risk groups. It remains unclear whether sustained attention scores in first-degree relatives are actually bimodally distributed, i.e., some relatives can be categorized “high risk” and others “low risk”. A further method of identifying high-risk subjects has been through questionnaire assessment of schizotypy, a personality trait varying across the normal population, which is associated with characteristics like paranoia, perceptual aberrations and thought disorganization. First-degree relatives of schizophrenic probands report higher self-report ratings of schizotypy than subjects without a family history of psychosis (Katsanis et al., 1990). Some studies have found schizotypy ratings to correlate inversely with CPT performance in relatives of schizophrenics (Grove et al., 1991; Keefe et al., 1997; Chen et al., 1998). In a nonclinical undergraduate population, high schizotypy subjects were more impaired at the CPT-IP than low schizotypy subjects (Lenzenweger et al., 1991). These studies illustrate a close relationship between the diagnosis of schizophrenia and target

85

Bipolar disorder: any overlaps with schizophrenia?

detection impairment on CPTs. The association appears to pre-date the onset of the disorder and indicate vulnerability. A further demonstration of this point comes from a prospective high-risk study by Cornblatt et al. (1999) which identified and monitored a group of children who were diagnosed subsequently with a schizophrenic disorder. Sustained attention deficit was apparent as early as at the age of 12, well before the manifestation of symptoms. CPT scores did not change at the point of onset of psychosis, and were roughly stable over a 15-year period spanning either side of the time of diagnosis. In summary, a wealth of research demonstrates stable deficits in sustained attention associated with the diagnosis of schizophrenia. Sustained attention in bipolar disorder Relationship to symptoms

The manic state of bipolar disorder is characterized by distractibility, restlessness, and impulsivity, and these symptoms can be quantified using a CPT. In a study that administered a comprehensive neuropsychological assessment to 15 manic inpatients and 30 matched healthy controls, target detection on the RVIP test and the verbal learning score (on the California Verbal Learning Test) were the two outcome measures that best distinguished the manic and control groups, classifying 87% of the manic patients correctly and 91% of subjects overall in a discriminant function analysis (Clark et al., 2001). In addition to impaired target detection, the manic patients also demonstrated slower response latencies and an increased rate of false alarms (reduced response bias) on the RVIP (Clark et al., 2001). The false alarm effect is likely to reflect increased impulsivity in the manic state. This finding has since been replicated by Swann et al. (2003) using the Immediate Memory Test-Delayed Memory Test (IMT-DMT) in 14 manic patients. This test resembles the CPT-IP but uses five digit strings on each trial. The subject must respond to consecutive identical strings, but the insertion of occasional “catch” trials where the consecutive strings differ by only one digit (e.g., 25453, 25452) makes this task particularly sensitive to impulsive responding. The rate of false alarms was significantly increased in the manic patients relative to controls, and target detection was reduced (Swann et al., 2003). An earlier study by Sax and colleagues (1995) compared CPT performance in symptomatic bipolar inpatients with pure manic (n17) and mixed (n13) symptoms with healthy controls (n  14). Impaired target detection was present in both bipolar groups, but only the pure manic group showed an increased rate of false responding, suggesting that the increase in impulsivity is specifically associated with the manic (rather than depressed) state. The Swann et al. (2003) study included a second group of bipolar patients tested in the remitted state (although the criteria for remission were not defined). The

86

Luke Clark

remitted patients showed a significant reduction in target detection relative to healthy controls. The rate of false responding in the remitted patients was, however, similar to controls and significantly less than the rate in the manic group. Further experiments support these findings. Clark et al. (2002) tested 30 outpatients with bipolar disorder on a neuropsychological assessment, including RVIP. Patients were defined as euthymic at the time of testing on the grounds of Hamilton Depression ratings 9 and Young Mania ratings 9. However, the bipolar group still reported significantly more affective symptoms on these scales than the healthy control group, as with many other studies. After controlling for these residual symptoms, the only significant difference between the two groups was impaired target detection on the RVIP. Nearly 50% of the euthymic group performed at least one standard deviation below the control mean on the percentage of targets detected, and 13% of the euthymic group were over two standard deviations below the control mean. When target detection was analyzed over the duration of task (in seven one-minute blocks), it was apparent that the vigilance decrement was also more pronounced in the euthymic group (group by time interaction effect, p 0.076). However, the euthymic group did not display the increase in false responding seen in manic inpatients in the earlier study by Clark et al. (2001) (control mean 1.9, manic group mean8.0, euthymic group mean1.6). This state-trait profile was also examined by Liu et al. (2002) who tested 15 bipolar patients on inpatient admission and again at discharge. Mood state at admission was not specified but these patients showed impaired target detection and an increased rate of false alarms. On discharge, target detection improved significantly but remained impaired relative to a large normative sample. False responding improved significantly back to the level of healthy performance. A further 53 remitted bipolar outpatients were tested on a single session only. In the combined group of patients at discharge and fully remitted patients (n68), target detection was impaired but the rate of false responding was normal. Overall, this series of studies demonstrates a highly consistent profile of sustained attention disturbance in bipolar disorder: a trait-related deficit in target detection coupled with a state-related increase in impulsivity specific to the manic state. Impaired target detection in remitted groups with bipolar disorder was also shown by Addington and Addington (1997), and Wilder-Willis et al. (2001). None of these studies in remitted patients reported a significant association between the target detection deficit and mood ratings at the time of testing (Wilder-Willis et al., 2001; Clark et al., 2002; Liu et al., 2002; Swann et al., 2003). It is likely that some degree of residual (subclinical) symptomatology was present in the remitted patients in each study. However, performance clearly worsens during acute illness episodes compared to remission (Liu et al., 2002; Swann et al., 2003; Clark and Goodwin, 2004). This is most clearly demonstrated in a study by Sax et al. (1998), who tested patients with

87

Bipolar disorder: any overlaps with schizophrenia?

first-episode affective psychosis (mostly bipolar) at hospital admission and again two months after discharge. The reduction in mania ratings (on the Young Mania Rating Scale) between admission and discharge was the only clinical rating seen to correlate with the improvement in CPT target detection. In conclusion, the deficit in target detection in bipolar disorder is exacerbated by the presence of manic symptoms, fitting the profile of a state-modulated trait marker. There is an additional state-related impairment in impulsive responding in mania (see Table 5.1). Given this state-dependency of sustained attention in bipolar disorder, it is difficult to separate the natural improvement in performance with symptom remission from any specific (i.e., pharmacological) influence of medication. Post-hoc cross-sectional analyses of patients on and off particular treatment regimes reveal no clear effects. In the Clark et al. (2002) study of euthymic patients, a comparison of patients receiving (n19) and not receiving lithium medication (n11) did not indicate any effect on CPT performance. Patients not receiving lithium remained significantly worse than controls. Wilder-Willis et al. (2001) also reported no effects of lithium, anticonvulsants and antipsychotics on cognitive performance in their remitted sample, although given that their total group constituted only 14 patients, these analyses may not be sufficiently powered. The study of manic inpatients (Clark et al., 2001) did not indicate any difference between patients receiving (n  7) and not receiving (n6) benzodiazepines, which may affect sustained attention via a sedative action. Target detection was also not correlated with the dosage of neuroleptic medication, in terms of chlorpromazine equivalence. It is worth noting that by the nature of the syndrome, it can be difficult to engage manic patients in any form of cognitive testing at hospital admission, and hence longitudinal designs involving a pre-treatment assessment may be unrealistic. Similarly, the withdrawal of prophylactic medication from remitted patients is unethical given the potential for triggering relapse, and the effect of any acute pharmacological challenge is likely to be buffered by the presence of prophylactic mood stabilizers. Larger studies are required to explicitly address the long-term cognitive effects of mood stabilizing medications in bipolar disorder. Relationship to the disorder

It is unclear to what extent sustained attention deficits develop with the progression of bipolar disorder. Clark et al. (2002) reported a correlation between target detection and illness chronicity in terms of both the number of bipolar episodes and the duration of illness. The studies of Liu et al. (2002) and Swann et al. (2003) failed to show any relationship with duration of illness or age of onset. Models of memory impairment in affective disorders have emphasized the progressive deterioration of memory function relating to the neurotoxic effects of increased cortisol levels in the hippocampus (Altshuler, 1993; MacQueen et al., 2003). Hypercortisolemia occurs

88

Luke Clark Table 5.1. Sustained attention performance in bipolar disorder: summary of experimental findings.

Reference

Clinical sample

Test

Finding

Notes

Rund et al. (1992)

28 schz, 19 BP depressed, 17 controls 17 BP manic, 13 BP mixed, 14 controls

SAT

Impaired target detection in BP and schz. Impaired target detection in both BP groups. Increased false responding in manic only. Impaired target detection in schz; BP intermediate between controls and schz. Tested on admission and after discharge; improvement in target detection correlated with YMRS change. Impaired target detection in BP.

Mood state not clearly defined in BP group. No effects on vigilance decrement.

Sax et al. (1995)

DS-CPT

Addington & Addington (1997)

59 schz, 40 BP remitted, 40 controls

X-CPT, SAT

Sax et al. (1998)

27 affective psychosis (16 BP), 31 controls

DS-CPT

Sax et al. (1999)

17 BP manic, 12 controls

DS-CPT

Clark et al. (2001)

15 BP manic, 30 controls

RVIP

Fleck et al. (2001)

20 BP manic, 20 schz, 20 controls Wilder-Willis et al. 14 BP remitted, (2001) 12 controls

DS-CPT

Clark et al. (2002)

RVIP

30 BP euthymic, 30 controls

DS-CPT

Impaired target detection in BP, increased false responding. Impaired target detection in BP and schz. Impaired target detection and RT in BP. Impaired target detection in BP. No effect on false responding.

Not significantly impaired at discharge.

Correlated with volume of prefrontal cortex. Significant after controlling for residual mood symptoms. Hit RT faster in BP than schz. No association with symptom ratings or illness chronicity. Correlated with number of episodes and duration of illness.

89

Bipolar disorder: any overlaps with schizophrenia? Table 5.1. (cont.)

Reference

Clinical sample

Test

Finding

Notes

Liu et al. (2002)

41 schz., 68 BP remitted

AX- CPT

Mood state not clearly defined in BP group.

Seidman et al. (2002)

87 schz., 15 BP manic, 94 controls

X-CPT

Robertson et al. (2003)

44 BP remitted adolescent onset, 30 UP remitted adolescent onset, 45 controls 22 BP euthymic, 12 BP manic, 35 controls

Connors CPT

Impaired target detection in BP. Schz. more impaired than BP. Impaired target detection in BP. Schz. more impaired than BP. No group effects observed.

Impaired target detection in both BP groups. Increased false responding in manic group only.

Not associated with illness chronicity.

Swann et al. (2003)

IMTDMT

Mood state not clearly defined in BP group.

Notes: BPBipolar disorder, UPUnipolar depression, schzschizophrenia, SATSpan of Apprehension test, DS-CPTDegraded Stimulus Continuous Performance Test, RVIPRapid Visual Information Processing, IMT–DMTImmediate Memory Test – Delayed Memory Test, RTReaction time, YMRSYoung Mania Rating Scale.

during depressive episodes. As such, verbal learning impairment in major depressive disorder correlates with the number and duration of episodes more closely than with the time elapsed since the first diagnosis (Kessing, 1998). Presumably, in patients with multiple episodes, memory impairment will become detectable during remission once substantial neurotoxicity has occurred. This is an example of a trait marker that is not directly related to the core etiology, but is a downstream effect of another pathological process (hypercortisolemia). Preliminary evidence indicates that sustained attention deficit is detectable early in the course of bipolar disorder, supporting a link with underlying neuropathology. Clark et al. (2002) showed a significant impairment in target detection in a subgroup of eight euthymic patients with short illness durations (40 months) and no more than two hospitalizations. Albus et al. (1996) showed a significant CPT deficit in patients tested during the first episode of affective psychosis. Sax et al. (1998) retested a similar

90

Luke Clark

group of patients (with substantial manic symptomatology at admission) two months after discharge. The comparison with healthy controls was not statistically significant in this study, but a moderate effect size (Cohen’s d0.67) indicates that the comparison was under-powered. In contrast, a study of adolescent-onset bipolar disorder failed to report any impairment on the Connors CPT relative to agematched healthy youths (Robertson et al., 2003). These cases all had illness onset between the age of 13 and 19 and were tested at a mean age of 19.4, during a period of remission. However, raw scores were not presented in this report and it is possible that a ceiling effect may have obscured a deficit in the bipolar group. The Connors CPT uses an unorthodox format similar to a Go-No Go task, where the subject must respond on every trial except when an X is presented. Consequently, the tendency to respond on every trial becomes automatic, and therefore the task may show improved sensitivity to false responding but reduced sensitivity to target detection impairment. Further studies are required, but there is a strong indication that sustained attention deficit is present early in the course of bipolar disorder. High-risk studies are needed to confirm whether this deficit actually conveys vulnerability to bipolar disorder, and pre-dates illness onset. A preliminary study (Clark et al., in press) in 28 first-degree relatives of probands with bipolar disorder failed to find any statistically significant impairment on the RVIP test relative to matched controls, but a modest effect size (Cohen’s d0.38) was apparent and the deficit is expected to attenuate with inheritance alone. A recent review by Glahn et al. (2004) highlights attentional dysfunction as one of the key areas for a candidate endophenotype to be targeted by further research. Direct comparisons of sustained attention in schizophrenia and bipolar disorder A small number of studies have directly compared sustained attention in groups of patients with schizophrenic and bipolar disorder diagnoses. The presence of psychotic features in cases with bipolar disorder is reliably associated with poorer performance (Liu et al., 2002). In first-episode cases with affective disorders (17 bipolar, 10 unipolar depression), patients with psychotic features were impaired on CPT target detection to a similar extent as schizophrenics (Albus et al., 1996). Patients without psychotic features did not differ significantly from controls. These findings are broadly consistent with a dimensional model of psychosis. However, in addition to psychotic features, CPT performance in bipolar disorder is also dependent critically on the mood state at the time of testing. A comparative study of a group of schizophrenic cases and a group of bipolar patients tested during a manic episode revealed a similar degree of impairment in CPT target detection (Fleck et al., 2001). Furthermore, neither group showed an exaggerated vigilance

91

Bipolar disorder: any overlaps with schizophrenia?

decrement relative to controls. The manic group was significantly faster than the schizophrenic group in this study, but response latency is more closely related to psychomotor speed than to sustained attention mechanisms per se. A comparison of bipolar patients tested during a depressive phase and patients with schizophrenia also revealed similar performance in the two groups on target detection (Rund et al., 1992). The schizophrenic group made more false responses than the bipolar group (Rund et al., 1992). In a third comparative study of 15 chronic, symptomatic patients with bipolar disorder (14 manic, 1 mixed), the schizophrenic group (n 87) performed significantly worse than the bipolar group on a multivariate score for sustained attention/vigilance that included both a CPT and a dichotic listening test (Seidman et al., 2002). In this study, however, the schizophrenic group scored significantly higher than the bipolar group on a symptom rating scale, and “virtually all subjects had low to moderate levels of overt psychosis” (Seidman et al., 2002, p. 35). In other studies where bipolar patients were tested during a period of symptom remission, the schizophrenic group performed significantly worse than the bipolar cases. The bipolar groups remain significantly impaired relative to controls (as described in the section ‘Relationship to symptoms’) and lie at an intermediate position between schizophrenics and controls (Addington and Addington, 1997). The state-dependency of sustained attention performance in bipolar disorder, compared to schizophrenia, was demonstrated in the study of Liu et al. (2002) where groups of bipolar (n15) and schizophrenic (n41) patients were tested on hospital admission and again at discharge. On admission, both groups were severely impaired at CPT target detection, with slightly greater deficits in the bipolar group (2.8 standard deviations below the mean for controls, compared to 2.5 standard deviations in the schizophrenic group). When re-tested at discharge, the bipolar group showed a significant improvement in target detection, whereas performance was unchanged in the schizophrenic group (the average value became slightly more impaired, in fact). In collapsing the bipolar group on discharge with a larger group of bipolar outpatients, the combined bipolar group (n 68) performed significantly better than the schizophrenic patients. Overall, these comparative studies substantiate the claim that schizophrenia and bipolar disorder are distinguished primarily by the symptomatic modulation of sustained attention performance. This is a qualitative distinction that supports a more categorical model of psychosis. In schizophrenia, sustained attention deficit is unrelated to the current symptom status. In bipolar disorder, sustained attention deficit is detectable during periods of well-defined euthymic mood, but this deficit becomes more pronounced during the manic and depressive states, when the magnitude of impairment is indistinguishable from schizophrenia. Direct comparative studies have not yet confirmed if the increased rate of false responding during the manic state is also specific to bipolar disorder (Sax et al., 1995; Clark et al., 2001;

92

Luke Clark

Swann et al., 2003). In the Liu et al. (2002) study, the rate of false responding in the bipolar group improved significantly from admission to discharge. At admission, the bipolar group was more impaired than the schizophrenic group (response biases were 1.6 standard deviations from the control mean, and 0.6 standard deviations, respectively) but a statistical value for this comparison was not reported. Given an earlier demonstration that the increased rate of false responding was specific to pure mania compared to mixed-state bipolar disorder (Sax et al., 1995), the increase in impulsivity is postulated to represent a disorder-specific neuropsychological marker that would also strengthen a categorical, rather than dimensional, viewpoint. Neuropsychological mechanisms of sustained attention impairment in psychosis Sustained attention deficits in schizophrenia and bipolar disorder are presumed to indicate aspects of neuropathology associated with these disorders. Impairments in CPT have been shown consistently in studies of organic brain damage (Rosvold et al., 1956) and are most associated with frontal lobe damage (Wilkins et al., 1987; Glosser and Goodglass, 1990). Right frontal lesions are particularly disruptive and are associated with an exaggerated vigilance decrement and diminished eventrelated potentials during task performance (Woods and Knight, 1986; Rueckert and Grafman, 1996). Functional neuroimaging studies in healthy subjects have also identified a right prefrontal cortex response during CPT performance, as part of a wider right-lateralized network including parietal cortex and thalamus (Coull et al., 1996; Paus et al., 1997; Coull et al., 1998; Sturm et al., 1999). Target detection rates correlate with the Functional Magnetic Resonance Imaging (fMRI) response in right frontal and right parietal cortex in healthy volunteers (Lawrence et al., 2003). On the basis of these studies, prefrontal cortex abnormalities have been postulated to underlie sustained attention deficits in schizophrenia and bipolar disorder. A structural MRI investigation in manic subjects showed that CPT target detection correlated with prefrontal cortex volume (Sax et al., 1999). The volume of prefrontal cortex was significantly reduced in the manic group relative to healthy controls. In this study, a single region of interest was defined and laterality was not assessed. The prefrontal cortex volume reduction is presumably stable over time and therefore indicates a trait (rather than state-related) abnormality. Using a whole-brain analysis of structural MRI data in 13 first-episode, neuroleptic naive patients with schizophrenics, Salgado-Pineda et al. (2003) showed that poor CPT performance was associated with reduced grey matter density in several foci including the left frontal cortex and thalamus (bilaterally), but also in several posterior cortical foci. Using functional imaging techniques (PET and fMRI) to assess brain responses during

93

Bipolar disorder: any overlaps with schizophrenia?

sustained attention performance, a number of studies have reported reduced glucose metabolism and reduced blood flow/blood oxygenation levels response in the frontal cortex of cases with schizophrenia relative to controls (Buchsbaum et al., 1992; Schroeder et al., 1994; Volz et al., 1999). However, patterns of hypofrontality in schizophrenia are notoriously difficult to interpret given the differences that exist in baseline performance: the reduced frontal response may be a corollary (rather than a cause) of sustained attention deficit. This artefactual account would predict hypofrontality in other psychiatric conditions associated with sustained attention deficit, but no functional imaging studies have investigated the neural correlates of CPT performance in bipolar disorder to enable a comparison. While prefrontal cortical dysfunction has been proposed to underpin CPT deficits in both schizophrenia and bipolar disorder, it is important to acknowledge that multiple mechanisms may be implicated in deficient performance including arousal, sensory processing and executive function (see section on “Sustained attention”). Cognitive studies have attempted to dissect CPT performance to isolate dysfunctional mechanisms in schizophrenia. Van den Bosch et al. (1996) reported a strong correlation between CPT target sensitivity and variability (standard deviation) in target reaction time in a group of schizophrenic patients. Increased variability of responding may arise as a result of motor dysfunction, or impaired alertness leading to attentional lapses. Cohen and Servan-Schreiber (1992) used a neural network simulation to model the performance of schizophrenic patients on a “double” CPT, where the target is defined by the stimulus on the previous trial (e.g., respond to “X” if it follows “A”). They hypothesized that deficient performance in schizophrenia on tasks of this type was associated with a specific component of working memory, the ability to represent and maintain the “context” of the previous trial in order to detect targets. By reducing gain in a working memory module holding the previous trial, Cohen and Servan-Schreiber mimicked the performance of schizophrenic patients on a manipulation where contextual demands were increased by expanding the inter-trial interval. Impaired context processing was postulated to be a function of dorsolateral prefrontal cortex. This was tested in a subsequent fMRI study using the AX-CPT in medication-naive patients with first-episode schizophrenia (Barch et al., 2001). Relative to healthy controls, the patients with schizophrenia showed blunted activation of dorsolateral prefrontal cortex during long-delay trials requiring the maintenance of context. This effect remained significant in two subgroups (schizophrenics and controls) matched for behavioral performance, suggesting that the effect was not confounded by baseline differences in performance (Barch et al., 2001). These data argue convincingly for a working memory impairment in schizophrenia associated with prefrontal cortical dysfunction. However, patients with schizophrenia are also impaired on other CPT variants that place only negligible demands on working memory and context

94

Luke Clark

processing. For example, the degraded stimulus CPT is particularly sensitive to schizophrenia (Nuechterlein et al., 1992), where the visual input is blurred using white noise. This implicates an additional impairment in schizophrenia at an earlier stage of information processing. Elvevag et al. (2000) used a series of variants of the double CPT to manipulate different components of sustained attention performance in a group of schizophrenic patients and healthy controls. Stimulus–response compatibility was manipulated by using an abstract task with low stimulus–response compatibility (respond when “1” is followed by “9”) and a task with inherent high compatibility (respond when “ready” is followed by “press”). The inhibitory component was manipulated by varying the proportion of target stimuli (17% or 50%) and the working memory component was manipulated by varying the delay between trials (1s or 3s). In a fourth condition, a dualtask component was inserted by requiring subjects to finger tap during CPT performance. All four manipulations successfully affected performance in the patients with schizophrenia and healthy controls. However, in this study there were no significant group by condition interactions, indicating that none of these manipulations tested an area of disproportionate weakness in schizophrenia. This study fails to support the model of Cohen, Barch and colleagues, and provides no evidence for executive function/prefrontal cortex mediation of the sustained attention impairment in schizophrenia. Elvevag et al. (2000) suggested that the deficit occurs during early processing at the stage of stimulus encoding. In bipolar disorder, the contribution of executive dysfunction to sustained attention depends critically on the phase of illness. The increased rate of false responding during the manic state is thought to indicate a breakdown of inhibitory control: manic patients are less able to withhold responses to non-target stimuli. Human lesion and functional neuroimaging data associate this type of motor inhibition with prefrontal cortex integrity, particularly the region of the inferior frontal gyrus in the right hemisphere (Aron et al., 2003; Rubia et al., 2003). Increased false responding can also be induced in healthy volunteers by administration of intravenous amphetamine (McTavish et al., 2001). This increase in false responding is attenuated by pre-dosing subjects with a tyrosine-free drink that depletes dopamine availability in the brain (McTavish et al., 2001). Intravenous amphetamine administration has been proposed as a model of manic symptomatology in healthy subjects (Jacobs and Silverstone, 1986), eliciting symptoms of euphoria and hyperactivity via increased dopamine neurotransmission (Moore, 1977; Drevets et al., 2001). Levels of the dopamine metabolite homovanillic acid are increased in the cerebrospinal fluid of manic patients (Post et al., 1978) and dopamine stimulants can trigger manic episodes (Murphy et al., 1971; Silverstone, 1984). Acute tyrosine depletion transiently reduced manic symptomatology in a group of manic inpatients (McTavish et al., 2001). These findings support a hypothesis that the

95

Bipolar disorder: any overlaps with schizophrenia?

increased rate of false responding in mania is associated with a state-related increase in dopamine neurotransmission in prefrontal cortex. During the euthymic phase of bipolar disorder, the residual deficit in sustained attention seems unrelated to executive function. Euthymic patients do not show the increased rate of false responding that characterizes the manic state (Clark et al., 2002; Liu et al., 2002; Swann et al., 2003). In several studies, euthymic patients perform in the range of healthy controls on other neuropsychological measures of frontal lobe function like planning, decision-making and self-ordered working memory (Rubinsztein et al., 2000; Clark et al., 2002). Harmer et al. (2002) used two variants of the Mackworth Clock Task to investigate specifically the contribution of working memory to the sustained attention impairment in the euthymic phase. In the original task (“WM”), the working memory load is negligible (“respond when the clock hand skips a beat”). The working memory load was increased substantially in a second task (“WM ”) where the clock hand changed color on each movement, and subjects were asked to respond to several color sequences (e.g., Green-Blue-Red). Although the WM task was considerably more difficult than the WM, the bipolar group were more impaired relative to controls on the WM task. This indicates that working memory does not mediate their impairment. Other mechanisms that could mediate the CPT impairment in the euthymic phase of bipolar disorder are decreased arousal/alertness, or an increased susceptibility to distraction. Distraction itself may arise from external sources in the environment or internal cognition. Manipulations of the CPT format will be able to separate these alternatives in future experiments. Arousal and distraction accounts are both broadly consistent with the observation of Clark et al. (2002) that the vigilance decrement is exaggerated in the euthymic phase: the detrimental impact of impaired arousal or increased distraction should accumulate with time spent on the task. Sustained attention in humans and other species has been associated with the integrity of the noradrenaline system. Noradrenergic dysfunction is an interesting candidate mechanism for sustained attention deficit in bipolar disorder, because reduced noradrenaline neurotransmission is a common feature of the depressive state. Both impaired arousal and increased distraction are compatible with models of noradrenaline disruption (indeed, distraction may arise from increased arousal). Noradrenergic neurons project from the locus coeruleus in the brainstem to cortical targets including prefrontal cortex. Electrophysiological recording from locus coeruleus neurons in the monkey has demonstrated that phasic and tonic components of cell firing in this region are closely associated with sustained attention performance (Aston Jones et al., 1999). Target stimuli were accompanied by a phasic burst of cell firing from noradrenaline neurons, while the rate of tonic firing was related to sustained attention according to an inverted-U function: at

96

Luke Clark

low levels, the monkeys were drowsy and inattentive, and at high levels, they were over distractible. Further studies using a rodent CPT analogue (the 5-choice serial reaction time; 5-CSRT) have shown that lesions of the ascending noradrenaline bundle impair sustained target detection when the inter-trial interval is unpredictable (as in the human CPT) and when distraction is increased by introducing bursts of auditory white noise (Carli et al., 1983; Cole and Robbins, 1992). Noradrenergic manipulations also modulate sustained attention in humans. For example, the mixed 1/2 agonist clonidine impaired target detection on the RVIP test (Coull et al., 1995) and increased attentional lapses (very long reaction times) on a test of selective attention (Smith and Nutt, 1996). Although an agonist, clonidine inhibits noradrenaline neurotransmission at low doses via presynaptic effects. There is also some evidence for a right-lateralized bias in the noradrenaline system (Robinson, 1979; Posner and Petersen, 1990). In summary, a wealth of evidence across several species implicates noradrenaline in the target detection component of sustained attention. As such, residual (trait) deficits in noradrenaline function may underlie the persisting target detection impairments in the euthymic phase of bipolar disorder, via a mechanism of decreased arousal or increased distraction. Conclusion Impaired target detection on CPTs is a robust neuropsychological marker of both schizophrenia and bipolar disorder. This may be taken, prima facie, as support for a dimensional model of psychosis where qualitatively similar dysfunction arises from a common etiology. However, some important features of the sustained attention profile differentiate schizophrenia and bipolar disorder. First, the acute episodes of bipolar mania are accompanied by a state-dependent increase in false responding, which is not present in remitted bipolar patients, and is not a reliable feature of schizophrenia. This cognitive variable implicates increased impulsivity in mania, which may be an effect of a state-related increase in dopamine neurotransmission. A second difference is in the symptomatic modulation of target detection impairments. In schizophrenia, sustained attention deficit is a stable trait marker that is not modulated by the presence of psychotic symptoms. In contrast, acute episodes of illness (mania or depression) in bipolar disorder exacerbate the residual impairment during the euthymic phase. During episodes of mania and depression, sustained attention is impaired to a similar degree to schizophrenia. However, in the remitted phase of bipolar disorder, sustained attention performance improves significantly and patients score at an intermediate point between schizophrenics and controls. This profile in bipolar disorder fits a state-modulated trait marker, and this may be linked to underlying

97

Bipolar disorder: any overlaps with schizophrenia?

neuropathology but also to the mechanisms that govern symptom formation. These two qualitative differences in the neuropsychological profile provide support for categorical models of psychosis. Moreover, target detection deficits on a CPT may arise from dysfunction in multiple brain mechanisms, and contrasting mechanisms are implicated in the deficits in schizophrenia and bipolar disorder. The deficit in schizophrenia may be associated with dysfunction in a component of the working memory system that maintains cognitive context across consecutive trials (Cohen and Servan-Schreiber, 1992). Dorsolateral prefrontal cortex is recruited during the maintenance of context in healthy volunteers, and this neural response is diminished in patients with schizophrenia (Barch et al., 2001). However, patients with schizophrenia are additionally impaired on sustained attention tasks that place negligible demands on working memory such as the degraded stimulus CPT, and these impairments may be further linked to dysfunction in posterior circuits that support the rapid encoding of incoming stimuli. The residual impairment in bipolar disorder may be closely linked to dysfunction in the brain mechanisms of arousal and/or distraction. Euthymic bipolar patients appear to show an exaggerated vigilance decrement on sustained attention tasks, a further feature of the cognitive profile that is not usually seen in schizophrenia. Reduced noradrenergic neurotransmission is one candidate mechanism for the residual deficit in bipolar disorder. Further research is needed to confirm the mechanisms that underpin sustained attention deficits in these two disorders, using experimental manipulations of CPT subcomponents alongside neuroscientific techniques including fMRI and psychopharmacological challenges. Models of dysfunction in schizophrenia benefit from a much larger body of research on the neuropsychological profile, compared to bipolar disorder. Neuropsychological studies in patients with bipolar disorder are further complicated by the critical role of phase of illness, which has not been adequately characterized in many previous experiments. Manic, depressed and even mixed-state patients may show differential effects and, in remitted patients, subclinical levels of affective symptomatology may still influence neuropsychological function and need to be controlled. The impact of mood stabilizing medications in bipolar disorder remains unclear, and very few studies to date have employed high-risk methodology to look at vulnerability to bipolar disorder. From a therapeutic perspective, the functional consequences of sustained attention deficit remain unclear although a considerable impact on everyday functioning may be predicted – from reading a book or completing household chores, to maintaining concentration in the workplace. Novel pharmacological treatments directed at sustained attention mechanisms may improve quality of life and protect against relapse, particularly in bipolar disorder where the deficit has been shown to vary naturally with illness state.

98

Luke Clark

Acknowledgements The author is indebted to the input of G. M. Goodwin and S. D. Iversen for Ph. D. supervision in the Department of Psychiatry, University of Oxford, with the support of a graduate studentship from the Medical Research Council of the United Kingdom. R E F E R E N C ES Addington, J. and Addington, D. (1997). Attentional vulnerability indicators in schizophrenia and bipolar disorder. Schizophrenia Research, 23, 197–204. Albus, M., Hubmann, W., Wahlheim, C. et al. (1996). Contrasts in neuropsychological test profile between patients with first-episode schizophrenia and first-episode affective disorders. Acta Psychiatrica Scandinavica, 94, 87–93. Altshuler, L. L. (1993). Bipolar disorder: are repeated episodes associated with neuroanatomic and cognitive changes? Biological Psychiatry, 33, 563–5. Aron, A. R., Fletcher, P. C., Bullmore, E. T., Sahakian, B. J. and Robbins, T. W. (2003). Stop-signal inhibition disrupted by damage to right inferior frontal gyrus in humans. Nature Neuroscience, 6, 115–16. Asarnow, R. F. and MacCrimmon, D. J. (1978). Residual performance deficit in clinically remitted schizophrenics: a marker of schizophrenia? Journal of Abnormal Psychology, 87, 597–608. Asarnow, R. F. and MacCrimmon, D. J. (1981). Span of apprehension deficits during the postpsychotic stages of schizophrenia. A replication and extension. Archives of General Psychiatry, 38, 1006–11. Aston Jones, G., Rajkowski, J. and Cohen, J. (1999). Role of locus coeruleus in attention and behavioral flexibility. Biological Psychiatry, 46, 1309–20. Barch, D. M., Carter, C. S., Braver, T. S. et al. (2001). Selective deficits in prefrontal cortex function in medication-naive patients with schizophrenia. Archives of General Psychiatry, 58, 280–8. Braver, T. S., Cohen, J. D. and Barch, D. M. (2002). The role of prefrontal cortex in normal and disordered cognitive control: a cognitive neuroscience perspective. In Principles of Frontal Lobe Function, ed. D. T. K. Stuss, Oxford: R. T. Oxford University Press. Buchsbaum, M. S., Haier, R. J., Potkin, S. G. et al. (1992). Frontostriatal disorder of cerebral metabolism in never-medicated schizophrenics. Archives of General Psychiatry, 49, 935–42. Carli, M., Robbins, T. W., Evenden, J. L. and Everitt, B. J. (1983). Effects of lesions to ascending noradrenergic neurones on performance of a 5-choice serial reaction task in rats; implications for theories of dorsal noradrenergic bundle function based on selective attention and arousal. Behavioural Brain Research, 9, 361–80. Chen, W. J., Liu, S. K., Chang, C. J. et al. (1998). Sustained attention deficit and schizotypal personality features in nonpsychotic relatives of schizophrenic patients. American Journal of Psychiatry, 155, 1214–20.

99

Bipolar disorder: any overlaps with schizophrenia? Clark, L., Iversen, S. D. and Goodwin, G. M. (2001). A neuropsychological investigation of prefrontal cortex involvement in acute mania. American Journal of Psychiatry, 158, 1605–11. Clark, L., Iversen, S. D. and Goodwin, G. M. (2002). Sustained attention deficit in bipolar disorder. British Journal of Psychiatry, 180, 313–19. Clark, L. and Goodwin, G. M. (2004). State- and trait-related deficits in sustained attention in bipolar disorder. European Archives of Psychiatry and Clinical Neuroscience, 254, 61–8. Clark, L., Kempton, M. J., Scarna, A., Grasby, P. M. and Goodwin, G. M. (in press). Sustained attention deficit confirmed in euthymic bipolar disorder, but not in first-degree relatives of bipolar patients or euthymic unipolar depression. Clark, L. and Sahakian, B. J. (2006) Neuropsychological and biological approaches to understanding bipolar disorder. In The Psychology of Bipolar Disorder, ed. S. Jones and R. P. Bentall. Oxford: Oxford University Press. Cohen, J. D. and Servan-Schreiber, D. (1992). Context, cortex, and dopamine: a connectionist approach to behavior and biology in schizophrenia. Psychological Review, 99, 45–77. Cole, B. J. and Robbins, T. W. (1992). Forebrain norepinephrine: role in controlled information processing in the rat. Neuropsychopharmacology, 7, 129–42. Cornblatt, B. A. and Erlenmeyer-Kimling, L. (1985). Global attentional deviance as a marker of risk for schizophrenia: specificity and predictive validity. Journal of Abnormal Psychology, 94, 470–86. Cornblatt, B. A., Lenzenweger, M. F. and Erlenmeyer-Kimling, L. (1989). The continuous performance test, identical pairs version: II. Contrasting attentional profiles in schizophrenic and depressed patients. Psychiatry Research, 29, 65–85. Cornblatt, B. A. and Keilp J. G. (1994). Impaired attention, genetics, and the pathophysiology of schizophrenia. Schizophrenia Bulletin, 20, 31–46. Cornblatt, B. A., Obuchowski, M., Schnur, D. and O’Brien, J. D. (1998). Hillside study of risk and early detection in schizophrenia. British Journal of Psychiatry (Supplement), 172, 26–32. Cornblatt, B. A., Obuchowski, M., Roberts, S., Pollack, S. and Erlenmeyer-Kimling, L. (1999). Cognitive and behavioral precursors of schizophrenia. Developmental Psychopathology, 11, 487–508. Cornblatt, B. A. and Malhotra, A. K. (2001). Impaired attention as an endophenotype for molecular genetic studies of schizophrenia. American Journal of Medical Genetics, 105, 11–5. Coull, J. T., Middleton, H. C., Robbins, T. W. and Sahakian, B. J. (1995). Clonidine and diazepam have differential effects on tests of attention and learning. Psychopharmacology (Berlin), 120, 322–32. Coull, J. T., Frith, C. D., Frackowiak, R. S. and Grasby, P. M. (1996). A fronto-parietal network for rapid visual information processing: a PET study of sustained attention and working memory. Neuropsychologia, 34, 1085–95. Coull, J. T., Frackowiak, R. S. and Frith, C. D. (1998). Monitoring for target objects: activation of right frontal and parietal cortices with increasing time on task. Neuropsychologia, 36, 1325–34. Crow, T. J. (1986). The continuum of psychosis and its implication for the structure of the gene. British Journal of Psychiatry, 149, 419–29. Desimone, R. and Duncan, J. (1995). Neural mechanisms of selective visual attention. Annual Review of Neuroscience, 18, 193–222.

100

Luke Clark Drevets, W. C., Gautier, C., Price, J. C. et al. (2001). Amphetamine-induced dopamine release in human ventral striatum correlates with euphoria. Biological Psychiatry, 49, 81–96. Earle-Boyer, E. A., Serper, M. R., Davidson, M. and Harvey, P. D. (1991). Continuous performance tests in schizophrenic patients: stimulus and medication effects on performance. Psychiatry Research, 37, 47–56. Elvevag, B., Weinberger, D. R., Suter, J. C. and Goldberg, T. E. (2000). Continuous performance test and schizophrenia: a test of stimulus-response compatibility, working memory, response readiness, or none of the above? American Journal of Psychiatry, 157, 772–80. Ferrier, I. N., Stanton, B. R., Kelly, T. P. and Scott, J. (1999). Neuropsychological function in euthymic patients with bipolar disorder. British Journal of Psychiatry, 175, 246–51. Fleck, D. E., Sax, K. W. and Strakowski, S. M. (2001). Reaction time measures of sustained attention differentiate bipolar disorder from schizophrenia. Schizophrenia Research, 52, 251–9. Franke, P., Maier, W., Hardt, J., Hain and C., Cornblatt, B. A. (1994). Attentional abilities and measures of schizotypy: their variation and covariation in schizophrenic patients, their siblings, and normal control subjects. Psychiatry Research, 54, 259–72. Glahn, D. C., Bearden, C. E., Niendam, T. A. and Escamilla, M. A. (2004). The feasibility of neuropsychological endophenotypes in the search for genes associated with bipolar affective disorder. Bipolar Disorders, 6, 171–82. Glosser, G. and Goodglass, H. (1990). Disorders in executive control functions among aphasic and other brain-damaged patients. Journal of Clinical and Experimental Neuropsychology, 12, 485–501. Green, D. M. and Swets, J. A. (1966). Signal Detection Theory and Psychophysics. New York: John Wiley & Sons. Green, M. F., Nuechterlein, K. H., Gold, J. M. et al. (2004). Approaching a consensus cognitive battery for clinical trials in schizophrenia: the NIMH-MATRICS conference to select cognitive domains and test criteria. Biological Psychiatry, 56, 301–7. Grier, J. B. (1971). Nonparametric indexes for sensitivity and bias: computing formulas. Psychological Bulletin, 75, 424–9. Grove, W. M., Lebow, B. S., Clementz, B. A. et al. (1991). Familial prevalence and coaggregation of schizotypy indicators: a multitrait family study. Journal of Abnormal Psychology, 100, 115–21. Harmer, C. J., Clark, L., Grayson, L. and Goodwin, G. M. (2002). Sustained attention deficit in bipolar disorder is not a working memory impairment in disguise. Neuropsychologia, 40, 1586–90. Harvey, P. D., Keefe, R. S., Moskowitz, J. et al. (1990). Attentional markers of vulnerability to schizophrenia: performance of medicated and unmedicated patients and normals. Psychiatry Research, 33, 179–88. Jacobs, D. and Silverstone, T. (1986). Dextroamphetamine-induced arousal in human subjects as a model for mania. Psychological Medicine, 16, 323–9. Katsanis, J., Iacono, W. G. and Beiser, M. (1990). Anhedonia and perceptual aberration in firstepisode psychotic patients and their relatives. Journal of Abnormal Psychology, 99, 202–6. Keefe, R. S., Silverman, J. M., Mohs, R. C. et al. (1997). Eye tracking, attention, and schizotypal symptoms in nonpsychotic relatives of patients with schizophrenia. Archives of General Psychiatry, 54, 169–76.

101

Bipolar disorder: any overlaps with schizophrenia? Kendell, R. E. and Gourlay, J. (1970). The clinical distinction between the affective psychoses and schizophrenia. British Journal of Psychiatry, 117, 261–6. Kessing, L. V. (1998). Cognitive impairment in the euthymic phase of affective disorder. Psychological Medicine, 28, 1027–38. Kraepelin, E. (1899). Psychiatrie. Ein Lehrbuch fur Studirende und Aerzte. Leipzig: J. A. Barth. Laurent, A., Saoud, M., Bougerol, T. et al. (1999). Attentional deficits in patients with schizophrenia and in their non-psychotic first-degree relatives. Psychiatry Research, 89, 147–59. Lawrence, N. S., Ross, T. J., Hoffmann, R. et al. (2003). Multiple neuronal networks mediate sustained attention. Journal of Cognitive Neuroscience, 15, 1028–38. Lenzenweger, M. F., Cornblatt, B. A. and Putnick, M. (1991). Schizotypy and sustained attention. Journal of Abnormal Psychology, 100, 84–9. Lezak, M. D. (1995). Neuropsychological Assessment, 3rd edn. New York: Oxford University Press. Liu, S. K., Chen, W. J., Chang, C. J. and Lin, H. N. (2000). Effects of atypical neuroleptics on sustained attention deficits in schizophrenia: a trial of risperidone versus haloperidol. Neuropsychopharmacology, 22, 311–19. Liu, S. K., Chiu, C. H., Chang, C. J. et al. (2002). Deficits in sustained attention in schizophrenia and affective disorders: stable versus state-dependent markers. American Journal of Psychiatry, 159, 975–82. Mackworth, N. H. (1948). The breakdown of vigilance during prolonged visual search. Quarterly Journal of Experimental Psychology, 1, 6–21. MacQueen, G. M., Campbell, S., McEwen, B. S. et al. (2003). Course of illness, hippocampal function, and hippocampal volume in major depression. Proceeding of the National Academy of Sciences USA, 100, 1387–92. Manly, T. and Robertson, I. H. (1997) Sustained attention and the frontal lobes. In Methodology of Frontal and Executive Function, ed. P. Rabbitt. Hove, UK: Psychology Press. McTavish, S. F., McPherson, M. H., Harmer, C. J. et al. (2001). Antidopaminergic effects of dietary tyrosine depletion in healthy subjects and patients with manic illness. British Journal of Psychiatry, 179, 356–60. Mesulam, M. M. (1981). A cortical network for directed attention and unilateral neglect. Annals of Neurology, 10, 309–25. Moore, K. E. (1977). The actions of amphetamine on neurotransmitters: a brief review. Biological Psychiatry, 12, 451–62. Murphy, D. L., Brodie, H. K., Goodwin, F. K. and Bunney, W. E., Jr. (1971). Regular induction of hypomania by L-dopa in “bipolar” manic-depressive patients. Nature, 229, 135–6. Nestor, P. G., Faux, S. F., McCarley, R. W. et al. (1991). Neuroleptics improve sustained attention in schizophrenia. A study using signal detection theory. Neuropsychopharmacology, 4, 145–9. Nuechterlein, K. H. (1983). Signal detection in vigilance tasks and behavioral attributes among offspring of schizophrenic mothers and among hyperactive children. Journal of Abnormal Psychology, 92, 4–28. Nuechterlein, K. H., Dawson, M. E., Gitlin, M. et al. (1992). Developmental processes in schizophrenic disorders: longitudinal studies of vulnerability and stress. Schizophrenia Bulletin, 18, 387–425.

102

Luke Clark Orzack, M. H. and Kornetsky, C. (1971). Environmental and familial predictors of attention behavior in chronic schizophrenics. Journal of Psychiatry Research, 9, 21–9. Parasuraman, R. (1984). The psychobiology of sustained attention. In Sustained Attention in Human Performance, ed. J. S. Warm. New York: Wiley & Sons. Parasuraman, R., Warm, J. S. and See, J. E. (1998). Brain systems of vigilance. In The Attentional Brain, ed. R. Parasuraman. Cambridge, MA: MIT Press. Paus, T., Zatorre, R. J., Hofle, N. et al. (1997). Time-changes in neural systems underlying attention and arousal during the performance of an auditory vigilance task. Journal of Cognitive Neuroscience, 9, 392–408. Posner, M. I. and Petersen, S. E. (1990). The attention system of the human brain. Annual Review of Neuroscience, 13, 25–42. Post, R. M., Lake, C. R., Jimerson, D. C. et al. (1978). Cerebrospinal fluid norepinephrine in affective illness. American Journal of Psychiatry, 135, 907–12. Robbins, T. W. (1998) Arousal and attention: psychopharmacological and neuropsychological studies in experimental animals. In The Attentional Brain, ed. R. Parasuraman. Cambridge, MA: MIT Press. Robertson, H. A., Kutcher, S. P. and Lagace, D. C. (2003). No evidence of attentional deficits in stabilized bipolar youth relative to unipolar and control comparators. Bipolar Disorders, 5, 330–9. Robinson, R. G. (1979). Differential behavioral and biochemical effects of right and left hemispheric cerebral infarction in the rat. Science, 205, 707–10. Rosvold, H. E., Mirsky, A. F., Sarason, I., Bransome, E. D. and Beck, L. H. (1956). A continuous performance test of brain damage. Journal of Consulting and Clinical Psychology, 20, 343–50. Rubia, K., Smith, A. B., Brammer, M. J. and Taylor, E. (2003). Right inferior prefrontal cortex mediates response inhibition while mesial prefrontal cortex is responsible for error detection. Neuroimage, 20, 351–8. Rubinsztein, J. S., Michael, A., Paykel, E. S. and Sahakian, B. J. (2000). Cognitive impairment in remission in bipolar affective disorder. Psychological Medicine, 30, 1025–36. Rueckert, L. and Grafman, J. (1996). Sustained attention deficits in patients with right frontal lesions. Neuropsychologia, 34, 953–63. Rund, B. R., Orbeck, A. L. and Landro, N. I. (1992). Vigilance deficits in schizophrenics and affectively disturbed patients. Acta Psychiatrica Scandinavica, 86, 207–12. Rutschmann, J., Cornblatt, B. and Erlenmeyer-Kimling, L. (1977). Sustained attention in children at risk for schizophrenia. Report on a continuous performance test. Archives of General Psychiatry, 34, 571–5. Sahakian, B., Jones, G., Levy, R., Gray, J. and Warburton, D. (1989). The effects of nicotine on attention, information processing, and short-term memory in patients with dementia of the Alzheimer type. British Journal of Psychiatry, 154, 797–800. Salgado-Pineda, P., Baeza, I., Perez-Gomez, M. et al. (2003). Sustained attention impairment correlates to gray matter decreases in first episode neuroleptic-naive schizophrenic patients. Neuroimage, 19, 365–75. Sax, K. W., Strakowski, S. M., McElroy, S. L., Keck, P. E. Jr. and West, S. A. (1995). Attention and formal thought disorder in mixed and pure mania. Biological Psychiatry, 37, 420–3.

103

Bipolar disorder: any overlaps with schizophrenia? Sax, K. W., Strakowski, S. M., Keck, P. E. Jr. et al. (1998). Symptom correlates of attentional improvement following hospitalization for a first episode of affective psychosis. Biological Psychiatry, 44, 784–6. Sax, K. W., Strakowski, S. M., Zimmerman, M. E. et al. (1999). Frontosubcortical neuroanatomy and the continuous performance test in mania. American Journal of Psychiatry, 156, 139–41. Schroeder, J., Buchsbaum, M. S., Siegel, B. V. et al. (1994). Patterns of cortical activity in schizophrenia. Psychological Medicine, 24, 947–55. Seidman, L. J., Kremen, W. S., Koren, D. et al. (2002). A comparative profile analysis of neuropsychological functioning in patients with schizophrenia and bipolar psychoses. Schizophrenia Research, 53, 31–44. Silverstone, T. (1984). Response to bromocriptine distinguishes bipolar from unipolar depression. Lancet, 1, 903–4. Smith, A., and Nutt, D. (1996). Noradrenaline and attention lapses. Nature, 380, 291. Sturm, W., de Simone, A., Krause, B. J. et al. (1999). Functional anatomy of intrinsic alertness: evidence for a fronto-parietal-thalamic-brainstem network in the right hemisphere. Neuropsychologia, 37, 797–805. Swann, A. C., Pazzaglia, P., Nicholls, A., Dougherty, D. M. and Moeller, F. G. (2003). Impulsivity and phase of illness in bipolar disorder. Journal of Affective Disorders, 73, 105–11. Taylor, M. A. (1992). Are schizophrenia and affective disorder related? A selective literature review. American Journal of Psychiatry, 149, 22–32. Van den Bosch, R. J., Rombouts, R. P. and van Asma, M. J. O. (1996) What determines continuous performance task performance? Schizophrenia Bulletin, 22, 643–51. van Gorp, W. G., Altshuler, L., Theberge, D. C., Wilkins, J. and Dixon, W. (1998). Cognitive impairment in euthymic bipolar patients with and without prior alcohol dependence. A preliminary study. Archives of General Psychiatry, 55, 41–6. Volz, H., Gaser, C., Hager, F. et al. (1999). Decreased frontal activation in schizophrenics during stimulation with the continuous performance test – a functional magnetic resonance imaging study. European Psychiatry, 14, 17–24. Wilder-Willis, K. E., Sax, K. W., Rosenberg, H. L. et al. (2001). Persistent attentional dysfunction in remitted bipolar disorder. Bipolar Disorders, 3, 58–62. Wilkins, A. J., Shallice, T. and McCarthy, R. (1987). Frontal lesions and sustained attention. Neuropsychologia, 25, 359–65. Wohlberg, G. W. and Kornetsky, C. (1973). Sustained attention in remitted schizophrenics. Archives of General Psychiatry, 28, 533–7. Woods, D. L. and Knight, R. T. (1986). Electrophysiologic evidence of increased distractibility after dorsolateral prefrontal lesions. Neurology, 36, 212–16.

6

The concept of schizoaffective disorder: utility versus validity and reliability – a transcultural perspective Ahmed Okasha Director, WHO Collaborating Center for Training and Research in Mental Health, Institute of Psychiatry, Ain Shams University, Cairo

Historical background Anthropologists, historians, and students of cross-cultural psychiatry have observed that every society has its own view of health and illness as well as its own classification of diseases (Okasha, 1988). Ever since the recognition by psychiatrists that clusters of symptoms repeatedly occur together, share a similar prognosis and often respond to the same lines of management, attempts have been made to find labels for these clusters. The labels tend to vary and increase in number every time a point of differentiation is recognized or believed to be recognized within these clusters. As the addition of labels was evidence of creative research in psychiatry, mental disorders and syndromes increased in number, and psychiatry seemed to enjoy the pride of having a name for almost every mental and psychological phenomenon. Yet psychiatrists were soon to recognize the drawbacks of such diversity in names and labels: they frequently seemed not to be speaking the same language and an insight was gained into the need for unifying psychiatric classification terms to enable the exchange of information and advice between psychiatrists worldwide. It was recognized that this is a difficult task, because the absence of physical signs and laboratory abnormalities in many mental disorders makes them much more dependent on the consensus of what is normal in a given society, what is abnormal, what is asocial and what is part of a disease (Sartorius, 1990). The high rate of comorbidities among our current psychiatric disorders undermines the hypothesis that our syndromes represent distinct etiologies. The high degree of short- and long-term diagnostic instability for many disorders and the The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

104

105

Utility versus validity and reliability

longitudinal variation in the diagnostic pattern throw doubt on our current classification. More important is the fact that lack of treatment specificity is the rule rather than the exception. To give two examples from genetic studies, we should refer to the work of Kendler’s study (1996) suggesting a common genetic basis for major depressive disorder and generalized anxiety disorder. Furthermore, the work by Berrettini (2000) indicates that three of the putative susceptibility loci associated with bipolar disorder also contribute to the risk of schizophrenia. Gaps in classification The present dilemma originates from our imprisonment in Kraepelin’s dichotomy between dementia praecox (schizophrenia) and manic depressive insanity (bipolar). Current evidence shows great overlap of their symptomatology, and the origin of the concept schizoaffective is to abridge the gap between the previously suggested two separate entities. Mood symptoms are prevalent in the prodroma, acute onset and postschizophrenic disorders. A high percentage of Schneiderian symptoms occur in manic disorder. In a recent article by Berrettini (2003) under the title “Bipolar disorder and schizophrenia: Not so distant relatives”, he explores the similarities in epidemiology as follows: (1) Life risks for SZ (Schizophrenia) and BPD (bipolar disorder) are estimated at about 1%; (2) both syndromes affect men and women equally; (3) onset of illness typically occurs between the ages of 15–25; (4) both are uncommon in the prepubertal period; (5) it is unusual for either disorder to arise de novo after the age of 50; (6) BPD and SZ are lifelong conditions; (7) spontaneous, lifelong remissions in either type of disorder are uncommon; (8) increased risk for suicide is shared by both; (9) treatments for BPD and SZ show overlap in the case of second-generation antipsychotics such as clozapine, as the prototype may have mood stabilizing properties; and (10) family and linkage studies are consistent with the concept that SZ and BPD share more genetic susceptibility. Multiple regions of the genome, including 18p11, 13q32 shared genetic susceptibility in both bipolar disorders and schizophrenia (Berrettini, 1996; Coryell and Zimmerman, 1988a). To have an illness or a disorder dependent on temporal overlap is a figment of imagination. Schizoaffective disorder is an outcome of our current nosological weakness, of our insistence of having the two entities BPD and SZ. If we have a psychotic spectrum we can locate many of our psychotic disorders with a rational explanation of the overlap. There are three disorders in addition to schizophrenia listed in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) in the

106

Ahmed Okasha

section “Schizophrenia and Other Psychotic Disorders.” The first, schizoaffective disorder, is a complex illness that has changed significantly over time. In its simplest definition, it is presently conceived as an illness with coexisting, but independent, schizophrenic (psychotic) and mood components. Schizoaffective disorder is seen primarily as part of a schizophrenia spectrum rather than an equal hybrid of mood and schizophrenia disorders. Schizophreniform disorder is a diagnosis that assumes another disorder that will replace it after six months. Most cases of schizophreniform disorder progress to either schizophrenia or schizoaffective disorder, with some cases being re-diagnosed as a non-schizophrenia spectrum illness (i.e., schizotypal or schizoid personality disorders), while a few resolve completely. Finally, the diagnosis of brief psychotic disorder describes an impairment in reality testing that lasts at least one day, but less than one month. All three disorders have a psychotic component, are often misunderstood, are incorrectly applied, and are not as well studied as schizophrenia, bipolar I disorder or major depressive disorder. The dilemma of schizoaffective disorder At the end of the century great strides have been made in clarifying the diagnostic criteria for many psychiatric illnesses. However, patients often do not fall neatly into set illness criteria. There are several approaches to dealing with such patients. One is to diagnose the patient with two distinct illnesses and treat those illnesses as separate problems. Another possibility is to consider that the patient has a primary illness and symptoms of a second illness that are not as important and might even resolve when the primary illness is treated. A third approach considers that the patient suffers from a distinct blended illness with its own history, diagnosis, and treatment. This last approach best represents the current orthodoxy in the diagnosis and treatment of patients with the DSM-IV diagnosis of schizoaffective disorder. Unfortunately, this approach is not easily applied, often making the diagnosis confusing and convoluted (Kaplan and Saddock, 2000). At the beginning of the twentieth century, patients with mental illness were grouped together as suffering from the common illness insanity. With the work of Emil Kraepelin, and Eugene Bleuler, distinct diagnostic groups began to emerge. Kraepelin was able to distinguish an unremitting, dementing illness in young patients that became known as schizophrenia, which he contrasted with an episodic illness of affect, now known as bipolar I disorder. However, there were patients who did not fit neatly into either category. Bleuler believed that the presence of any symptoms of schizophrenia, even when there was an affective component, was still schizophrenia. Patients with mixed features of schizophrenia and affective (mood) disorder were first described by

107

Utility versus validity and reliability

George Kirby (1913) and August Hoch (1921) in the early part of the century. In 1933, Jacob Kasanin introduced the term “schizoaffective psychosis” to describe a group of patients who had symptoms of both affective and schizophrenic illnesses. While he is credited for introducing the term, on subsequent review of these patients, all would now meet the diagnosis of a pure mood disorder. Nevertheless, the term schizoaffective disorder has survived, albeit in several different contexts. Kraepelin’s doubt became stronger through the investigation made by his scholar and colleague Zendig. Zendig (1909), in his paper “Contribution to differential diagnosis of manic-depressive insanity and dementia praecox” reported that approximately 30% of Kraepelin’s investigated sample with “dementia praecox” according to Kraepelin’s guidelines had a course and outcome not corresponding to that of “dementia praecox.” In such cases, Kraepelin saw a weakness in his dichotomy concept. Eugene and Manfred Bleuler (Bleuler, 1937; 1983) knew and described such overlaps and named them Mischpsychosen (mixed psychoses). In 1966, Jules Angst investigated the “mixed psychosis” notion as part of the affective disorders: a very emancipating and contrary step regarding the theories of his teacher. Kasanin’s cases (1933) have much more similarity to the bouffé délirante of French psychiatry (Pichot, 1986) or partially to what we call today “acute and transient psychotic disorder,” especially with the polymorphous type (ICD-F23) (Marneros et al., 2000; Pillmann et al., 2001). Schizoaffective disorder has been variously proposed to represent a subtype of schizophrenia (as in DSM-II), a subtype of manic-depressive illness (Clayton et al., 1968), or a separate entity (Walinder, 1972; Procci, 1976). The diagnosis of schizoaffective disorder has been in use for nearly 50 years, despite a lack of evidence to establish its validity. Does schizoaffective disorder merit distinction as a valid diagnostic category? What we actually call “schizoaffective disorder” was described in an almost identical way by Kurt Schneider in Germany in the 1930s as “cases-in-between.” However, in North America, long before Kasanin’s paper (1933) was published, similar cases were described and similar opinions existed: for example by Kirby (1913) or Hoch (1921). Three years after Kasanin’s publication, Hunt and Appel (1936) published their study in which they investigated all patients admitted to the Pennsylvania Hospital between 1919 and 1929 with mixed, schizophrenic, and affective symptomatology. They found that remission in these cases is twice as frequent as in schizophrenia, but 50% less frequent than in manic-depressive diseases. In 1963, Vaillant described all patients with so-called “remitting schizophrenia” as schizoaffective. Vaillant wrote: “Almost every schizophrenic patient with remission could be also diagnosed as having a ‘schizoaffective disorder’.”

108

Ahmed Okasha

The modern definitions of schizoaffective disorders according to DSM-IV and ICD-10 are not sufficient to define all the groups and sub-groups of schizoaffective disorders, nor are they sufficient to describe and define the schizoaffective phenomenon. One of the difficulties in using a diagnosis that depends on not being another diagnosis is that both depend on changes in the other. Schizoaffective disorder is affected by any changes in the diagnostic criteria of schizophrenia, affective disorder, or both. As psychotic affective disorders and schizophrenia have been better distinguished, those that fall through the “diagnostic cracks” have become clearer. In the second edition of DSM (DSM-II), schizoaffective disorder was a subtype of schizophrenia and denoted patients who had any mood symptoms while meeting the criteria for schizophrenia. In contrast, the Research Diagnostic Criteria (RDC) for schizoaffective disorder allowed as few as one symptom of schizophrenia in a patient who met the criteria for a full affective disorder. The third edition of DSM (DSM-III), influenced by studies in the United States and Great Britain, narrowed the diagnosis of schizophrenia and expanded the diagnosis of bipolar disorder. It allowed symptoms of schizophrenia to coexist with a mood disorder as long as these schizophrenic symptoms did not persist when the mood disorder resolved. Moreover, mood-incongruent psychotic symptoms could now exist in bipolar disorder. Finally, schizoaffective disorder moved from its place as a schizophrenia subtype to stand alone as a “psychotic disorder not elsewhere classified.” The revised third edition of DSM (DSM-III-R) expanded this notion by inserting the criterion that a patient with schizoaffective disorder must meet the criteria for schizophrenia for at least two weeks, independent of any mood syndrome. The DSM-IV classification has retained most of the DSM-III-R criteria but has stricter diagnostic criteria for schizophrenia. Patients must meet the symptoms of schizophrenia for at least one month as opposed to the previous one-week criterion. Schizoaffective disorder is now listed in the section “Schizophrenia and Other Psychotic Disorders.” The ICD-10 essentially describes the same disorder. The ICD10 schizoaffective disorders describe single as well as recurrent episodes. Subtypes include manic, depressed, and mixed types. The mixed type includes a cyclic schizophrenia and a mixed schizophrenic-mood psychosis. Schizoaffective disorder is one facet of the various caveats of international classifications. Slavish adoption of ICD-10 and DSM-IV definitions may have hindered research into the etiology of mental disorders. The concrete use of DSM-IV and ICD-10 classifications considered to be equivalent to “diseases” is more likely to obscure rather than elucidate research findings. We should encourage a research agenda that goes beyond our categorical current ways of thinking to a dimensional approach, integrating information from a wide variety of sources and technologies (Berrettini, 2003).

109

Utility versus validity and reliability

The study of schizoaffective disorder (SAD) can perhaps contribute seriously in answering the question of a “psychotic continuum” or “separate entities” of mental disorders. The most important criticisms of the definitions of DSM-IV and ICD-10 for SAD are the following: First, there is no argument for the chronological distinction regarding the coexistence of schizophrenic and affective symptomatology. Second, both diagnostic systems do not involve the longitudinal aspect in their definitions. Operational research has shown that there is no difference on all levels of investigation; for example, family history, premorbid personality, premorbid social adaptation, pattern of illness course, frequency and type of persistent alterations and response to prophylaxis between patients having only a schizodepressive episode in their course and patients changing between schizophrenic and affective episodes. It has also been well established that schizoaffective disorders must be divided into unipolar and bipolar types, in exactly the same way as the pure affective disorders. Longitudinal clinical research has shown that bipolar schizoaffective disorders must be distinguished from the unipolar type. Bipolar schizoaffective disorders have very similar premorbid and sociodemographic features and patterns to the pure affective bipolar disorders in terms of course, treatment and prophylaxis, but they differ significantly from the unipolar schizoaffective disorders, in exactly the same way as bipolar affective disorders differ from unipolar affective disorders. Nosological definition Schizoaffective disorders are yet a controversially discussed but existing nosological category. They are also an insufficiently investigated but nevertheless fascinating psychotic area. Many years ago, Marneros called the schizoaffective disorder a “nosological nuisance but a clinical reality.” Marneros (2003a, 2003b) stated that the present definitions of the International Classification of Disease and the Diagnostic and Statistical Manual of Mental Disorders are not sufficient enough to define schizoaffective disorders, especially because of deficits on the longitudinal axis. Schizoaffective disorders occupy a position between schizophrenia and pure mood disorders, especially regarding prognosis and premorbid and sociodemographic variables. Schizoaffective disorders are recurrent disorders and, therefore, need prophylactic treatment. Suicidal symptomatology is extremely frequent in patients with schizodepressive episodes. The most severe type of schizoaffective disorder is the schizoaffective mixed type. Schizoaffective patients with mixed episodes retire more frequently and at younger ages than other patients with bipolar disorder.

110

Ahmed Okasha

He suggested two types of schizoaffective disorder that should be distinguished: the “concurrent” and the “sequential” type. The first includes people having only a coincidence of schizophrenic and affective symptoms. The “sequential” type is defined as the schizoaffective disorder under a longitudinal aspect subsuming disorders with a symptoms change between different episodes. The DSM-III-R was a new attempt to split the group of patients identified by the RDC as having schizoaffective disorder, by redefining the distinction proposed in the RDC (the RDC affective subtype corresponding to DSM-III-R mood disorders with mood-incongruent psychotic features and the RDC schizophrenic subtype corresponding to DSM-III-R “true” schizoaffective disorder) (Kendler et al., 1989). The crucial differential criterion is whether the mood-incongruent psychotic features mainly occur when the criteria for a major affective syndrome are fulfilled (affective disorder with psychotic features) or whether psychotic features are also present for two weeks or longer after the affective syndrome has remitted or before the affective syndrome will become apparent (schizoaffective disorder). Validity of classification Harrison et al. (1980) compared patients who met widely accepted criteria for the diagnosis of schizoaffective disorder (manic type) with patients meeting rigorous criteria for manic disorder and schizophrenia, using three methods of validation: family history, short-term treatment response, and long-term outcome. No significant differences were found between patients with manic disorder and schizoaffective disorder. However, consistent and often highly significant differences separated patients with schizophrenia from those with manic disorder and schizoaffective disorder. The findings suggest that schizoaffective disorder, as currently defined, is not a valid and independent entity. The authors suggest that psychotic disorders not diagnosable as manic-depressive illness or schizophrenia and without an apparent organic basis would best be called “undiagnosed” or “atypical” psychosis. Further, while proposals for new diagnoses or for subtyping of schizophrenia or manic-depressive illness should be encouraged, these should undergo rigorous screening for validity before being accepted into clinical use. It is thus crucial to develop a nosology for these disorders, in which each diagnostic entity is shown to be valid. How does one establish the validity of a psychiatric diagnosis? Lacking the histological, pathological, or biochemical evidence used to validate diagnostic entities in most branches of medicine, psychiatry must rely on other validating criteria. These have been reviewed by Robins and Guze (1970), who enumerated five criteria for establishing diagnostic validity in psychiatric illness: clinical description, laboratory studies, delimitation from other disorders, follow-up study, and family study.

111

Utility versus validity and reliability

Currently, fairly well validated criteria exist for the diagnosis of schizophrenia and manic-depressive illness, including the criteria of Feighner et al. (1972), the Research Diagnostic Criteria (RDC) (Spitzer et al., 1978) and the criteria of the 1978 and 1979 drafts of DSM-III. However, it is widely recognized that many patients presenting with psychotic syndromes are difficult to classify under schizophrenia or bipolar disorders, and the clinician may escape with the diagnosis of schizoaffective disorder. No significant differences could be found between the manic and schizoaffective patients on any variables related to prognosis, premorbid history, family history or treatment response. These results would seem to question the validity of the schizoaffective diagnosis and suggest that schizoaffective disorder might be in distinguishable from manic disorder. On all three validating criteria, no significant differences were found between manic disorder and schizoaffective disorder, manic type. On the other hand, on the same three validating criteria, consistent, often highly significant differences emerged, separating patients with schizophrenia from both schizoaffective and manic patients. Etiology It is difficult to determine a cause of a disease that has changed so much over time. One might conjecture that the etiology of schizoaffective disorder as currently defined might be similar to the etiology of schizophrenia. Thus etiological theories of schizoaffective disorder would include some genetic and environmental causation. Molecular genetic studies of schizoaffective disorder have lagged behind recent studies of the genetics of schizophrenia and bipolar I disorder. Available family studies have reported that families of schizoaffective probands have significantly higher rates of relatives with mood disorder than families of schizophrenia probands. Similarly, these schizoaffective probands have more psychotic symptoms than families of mood disorder probands. The results of these family studies have argued that schizoaffective disorder is a unique disorder, separate from schizophrenia and mood disorders. Family studies Maier et al. (1992) investigated whether the distinction between schizoaffective disorder and affective disorders with mood-incongruent psychotic features as described in DSM-III-R is reflected by an aggregation of schizophrenia in the families of probands with the former disorder and an aggregation of affective disorders mainly among the relatives of probands with the latter type of disorders.

112

Ahmed Okasha

With regard to the familial aggregation of schizophrenia, the DSM-III-R distinction emerged as valid. However, the risk of unipolar affective disorders was enhanced in the families of all of the sub-groups of patients studied. The unipolar/bipolar distinction in both DSM-III-R and DSM-IV diagnostic groups was reflected by distinct patterns of bipolar disorders in the relatives. The results partly support the DSM-III-R dichotomy of schizoaffective disorder and affective disorders with mood-incongruent psychotic features. Although the differences between these two diagnostic groups were significant, the magnitude of the differences remained relatively modest. Some investigations (Maj et al., 1991; Kendler et al., 1995a) have assessed the diagnostic validity of schizoaffective disorder as defined by DSM-III-R. Gershon et al. (1988) and Maj et al. (1991) examined the risks for schizophrenia and mood disorder among first-degree relatives of schizoaffective disorder patients (depressed type) compared with patients with schizophrenia, psychotic depression, or non-psychotic depression. These authors noted that the relatives of the schizoaffective disorder group had the same risk of schizophrenia as those of the schizophrenia group. An earlier study of the relatives of patients with chronic schizoaffective disorder and schizophrenia also revealed no difference in the relatives’ risk of schizophrenia. In addition, a similar incidence of mood disorder (both bipolar and unipolar) was found in relatives of probands of both groups (schizoaffective disorder and schizophrenia) (Gershon et al., 1988). Taken together, these findings support the notion that schizoaffective disorder corresponds closely to schizophrenia. Ninety-one consecutively admitted patients with schizophrenia (n  21), schizoaffective depression (n  43), and psychotic depression (n  27) entered a blind family study along with 36 never-ill controls (Coryell and Zimmerman, 1988a). Though schizophrenia spectrum disorders clustered within families, they were not significantly more prevalent in the families of schizophrenic probands. In contrast, morbid risks for affective disorder clearly separated the families of psychotically depressed probands from the families of both schizophrenics and controls. Family study data for schizoaffective probands indicated links to both affective disorder and schizophrenia and suggested, as well, that a small number of patients with schizoaffective disorder may carry a genetic liability to both conditions. Coryell (1986) reviewed studies that have attempted to diagnose schizoaffective disorder in relatives. Only three studies (Baron et al., 1982; Angst, 1973; Endicott et al., 1986) reported a higher rate in the relatives of schizoaffective probands than in the relatives of comparison groups, and none of these group differences reached significance. However, schizoaffective probands, when compared with never-ill controls, had heavier familial loading for both schizophrenia and affective disorder, a finding now frequently reproduced in the literature. These data firmly support

113

Utility versus validity and reliability

the traditional separation of schizophrenia and psychotic affective disorder. Schizophrenics did not have a greater familial loading for affective disorder, nor did depressed probands have a greater loading for schizophrenia than did controls. These data clearly do not settle the debate over whether narrowly defined schizophrenia is a familial illness or not. The most parsimonious reconciliation of the various studies would probably hold that some schizophrenias are familial and other schizophrenias are not. Besides the re-evaluation of the Iowa 500 study by Kendler et al. (1989), this family study is the second in the literature to test the subdivision of the DSM-III-R categories of schizoaffective disorder and psychotic affective disorders by the unipolar/bipolar dichotomy. Both studies agree that this subdivision is a valid one; elevated rates of bipolar conditions are found only if the probands belong to a bipolar group. The study was further in agreement with the re-evaluated Iowa 500 study: in both reports DSM-III-R bipolar schizoaffective disorder is as closely linked to schizophrenia by familial risk figures as is unipolar schizoaffective disorder. In contrast to this conclusion, studies exclusively relying on the RDC suggest that bipolar schizoaffective disorder, but not unipolar schizoaffective disorder, is more a variant of affective than of schizophrenic disorders (Levinson and Levitt, 1987). The requirement in DSM-III-R, but not in the RDC, that schizoaffective disorder must meet the cross-sectional criteria of a full-blown schizophrenic syndrome at one time might explain these discrepant results. It remains for larger studies to determine the feasibility and value of subdividing schizophrenia in this way. Finally, these data accord with an under-recognized consistency in the literature: many schizoaffective patients have what is, in fact, affective disorder, and others have schizophrenia, while a small proportion may be expressing both illnesses and even some yet unrecognized and unrelated conditions. Genetic studies Tsuang (1979) studied 35 sibling-pairs (“sib-pairs”), where every sixth pair was independently diagnosed as having schizophrenia (SC), affective disorder (AD), or schizoaffective disorder (SA). The observed numbers of same diagnosis pairs (ADAD, SASA, SCSC) were compared with the numbers expected if the three disorders were genetically independent. The results showed a significant deficiency only in the observed number of SASA pairs, which suggests that schizophrenia and affective disorder are genetically different, whereas schizoaffective disorder is not. To test whether schizoaffective disorder is a variant of affective disorder or schizophrenia, the observed number of ADSA and SCSA pairs were compared against the expected numbers. No significant differences were found, which suggests that

114

Ahmed Okasha

schizoaffective disorder is genetically heterogeneous consisting at least of two subtypes, one a variant of affective disorder, the other a variant of schizophrenia. Bertelsen and Gottesman (1995) reviewed the literature relevant to genetic predisposition to schizoaffective disorder. Results from family, twin and adoption studies were somewhat divergent, but generally supported the concept that schizoaffective disorder was either a phenotypic variation or an expression of a genetic interform between schizophrenia and mood disorder. Kendler et al. (1995a) also examined the notion of an underlying continuum of liability to the “schizophrenia spectrum” of illness and concluded that schizoaffective disorder lies along a continuum of a range of disorders that varied in severity. Genetic epidemiological studies reveal that relatives of bipolar probands are at increased risk for recurrent unipolar, bipolar, and schizoaffective disorders, whereas relatives of probands with schizophrenia are at increased risk for schizophrenia, schizoaffective, and recurrent unipolar disorders. The overlap in familial risk may reflect shared genetic susceptibility. Recent genetic linkage studies have defined confirmed bipolar susceptibility loci for multiple regions of the human genome, including 4p16, 12q24, 18p11.2, 18q22, 21q21, 22q11–13, and Xq26. Studies of schizophrenia kindreds have yielded robust evidence for susceptibility at 18p11.2 and 22q11–13, both of which are implicated in susceptibility to bipolar disorder. Similarly, confirmed schizophrenia vulnerability loci have been mapped, too, for 6p24, 8p, and 13q32. Strong statistical evidence for a 13q32 bipolar susceptibility locus has been reported. Thus, both family and molecular studies of these disorders suggest shared genetic susceptibility. These two groups of disorders may not be as distinct as current nosology suggests. So if there is a suspicion that schizophrenia and bipolar are two distinct entities, would it be feasible to consider SA disorder as a separate entity (Berrettini, 2000)? Incidence of schizoaffective disorder There is no psychiatric epidemiological study of the incidence or prevalence of schizoaffective disorder in a general population. Even if there were such studies, older reports might not be useful, because the diagnosis (and therefore the incidence and prevalence) would have changed over time. Prevalence rates for consecutive patients diagnosed in a psychiatric treatment setting are available. These numbers range from 2 to 29%, a potentially significant cohort requiring treatment (Keck et al., 2001). The estimated lifetime prevalence of schizoaffective disorder is possibly in the range of 0.5 to 0.8%. In the inpatient settings of New York State psychiatric hospitals, approximately 19 of 6000 patients had a diagnosis of schizoaffective disorder (Levinson et al., 1999).

115

Utility versus validity and reliability

Symptomatology Williams and associates (1987) suggest that the disorder represents a variant of schizophrenia or mood disorder, a view that is also shared by Taylor (1992) and Kendler et al. (1995a). Lapierre (1994) holds the opinion that schizoaffective disorder represents a phenotypic variation of either schizophrenia or mood disorder and over the long term becomes a subtype of either one. Alternatively, some view schizoaffective disorder as the simultaneous expression of schizophrenia and a mood disorder. However, Kendler and colleagues (1995b) observe that schizoaffective patients differ significantly from both schizophrenia and mood disorder patients. In an effort to discriminate affective from schizophrenic illness, Taylor and Amir (1994) examined similarities and differences between DSM-III-diagnosed schizoaffective disorder or schizophrenia patients and mood disorder patients on the basis of ratings of psychopathology. Results differentiated chronic schizophrenia from mood disorder patients, but schizoaffective disorder patients overlapped with both groups. The schizoaffective disorder patients with unipolar depression more closely resembled the schizophrenia patients, while the schizoaffective disorder patients with bipolar depression were more similar to the mood disorder group. Different conclusions were suggested by another study by Kendler et al. (1995a) that examined the clinical features, outcome, and familial psychopathology among schizoaffective disorder patients compared with patients with schizophrenia or mood disorder. Results showed that schizoaffective disorder patients differed significantly from both schizophrenia and mood disorder patients. Specifically, although schizoaffective disorder patients had positive psychotic symptoms similar to schizophrenia patients, the schizoaffective disorder patients had a greater number of affective symptoms, fewer negative symptoms, and a better global course and outcome than did the schizophrenia patients. Furthermore, relatives of probands with schizoaffective disorder had higher rates of schizophrenia compared with mood disorder probands. These authors concluded that schizoaffective disorder was a syndrome that differed meaningfully from both schizophrenia and mood disorder, according to DSM-III-R criteria. This finding of better global outcome among schizoaffective disorder patients compared with schizophrenia patients was also seen in a Scandinavian study (Retterstol, 1991). However, a diagnosis of mood disorder or the presence of mood symptoms in addition to a psychotic illness was associated with a more favorable outcome relative to a primary diagnosis of a psychotic disorder among all patients. In a small sample comparing 27 schizoaffective disorder patients with 27 bipolar disorder patients, first-rank symptoms and mood-incongruent psychosis did not differ between the two groups (Strakowski et al., 1999). Whaley’s study (2002)

116

Ahmed Okasha

examined intercategory agreement among the symptom clusters associated with diagnoses of affective disorder, schizoaffective disorder, and schizophrenia in 156 African-American patients. The symptom clusters for diagnoses of schizoaffective disorders overlapped in 41% with those of affective disorders and in 71% with those of schizophrenia. The results support the assumption that schizoaffective disorder falls within the spectrum of schizophrenia disorders, drawing implications for studies involving the differential diagnosis of affective disorder and schizophrenia in African-Americans and those from other cultures. At present, Marneros et al. are carrying out a longitudinal prospective investigation on mixed states. One of the subjects of this study is the comparison of affective and schizoaffective mixed states. Mixed schizoaffective disorders are almost as frequent as the affective mixed ones. A schizo-manic-depressive mixed episode is an indicator of higher severity of the illness, as in pure affective mixed bipolar episodes. Some authors, especially in earlier descriptions, allocated the schizoaffective disorders to schizophrenia, but research during recent decades has shown increasingly that schizoaffective disorders are more similar to affective than to schizophrenic disorders, building a bridge between different disorders and stimulating the basis for research and discussion on a psychotic continuum and an independent nosological entity. Cognitive functions Some work has been done to compare the patterns of neuropsychological performance among schizoaffective disorder and schizophrenia patients. Bornstein and colleagues (1990) examined the pattern of cognitive deficit among paranoid and non-paranoid schizophrenia patients, as well as schizoaffective disorder patients compared with that of an age-matched healthy control group. The non-paranoid schizophrenia group had the greatest neuropsychological impairment, whereas the paranoid schizophrenia and schizoaffective disorder patients had a similar level of cognitive impairment. Other research suggests that frontal and/or subcortical dysfunction is common in both schizoaffective disorder and schizophrenia patients and that both groups exhibit comparable levels of deficits on tests of global mental status, attention, problem solving, and verbal and non-verbal fluency (Beatty et al., 1993). Recent work in the Clinical Research Center at the University of California, San Diego, has documented that the overall levels of psychopathology and cognitive impairment in patients with psychotic depression are similar to those in schizophrenia patients but greater than those in non-psychotic depressed patients (Jeste et al., 1996). Evans et al. (1999) suggest that schizoaffective disorder represents a variant of schizophrenia in terms of clinical symptoms, family history and treatment.

117

Utility versus validity and reliability

Diagnostic criteria The first step in the identification of more homogeneous types of schizoaffective disorders was the development of the Research Diagnostic Criteria (RDC) (Spitzer et al., 1975), in which “mainly affective” and “mainly schizophrenic” types were identified. The type is based on the temporal relationship between the occurrence of the affective and the schizophrenic-like symptoms and on the quality of the premorbid adjustment (Maj et al., 1991). The validity of this dichotomy has been supported by family studies (Baron et al., 1982) and drug response studies, but the data provided by outcome studies have been less clear. Some investigations have confirmed the validity of the dichotomy (Coryell et al., 1987; 1988b) and others have shown no significant difference between patients with mainly affective and mainly schizophrenic schizoaffective disorder (Grossman et al., 1991; Coryell and Zimmerman, 1988a). Over the decades, SA patients were often classified as having atypical schizophrenia, good prognosis schizophrenia, remitting schizophrenia, or cycloid psychosis. Inherent within these diagnoses was the implication that they shared similarities to schizophrenia and also appeared to have a relatively better course of illness. With the advent of effective treatment of bipolar disorder with lithium salts, some of these patients started responding to lithium, and the term schizoaffective disorder gained further momentum and evolved in the direction of bipolar disorder. Unfortunately, this lack of diagnostic clarity has plagued the diagnosis of schizoaffective disorder, such that there is not much that is known about the illness. Evans et al. (1999) compared outpatients aged 45 to 77 years with DSM-III-R diagnosis of schizoaffective disorder (n  29), schizophrenia (n  154), or nonpsychotic mood disorder (n  27) on standardized rating scales of psychopathology and a comprehensive neuropsychological test battery. The schizophrenia patients had more severe positive symptoms than the schizoaffective and mood disorder patients. The neuropsychological performances of the two psychotic groups were more impaired than those of the non-psychotic mood disorder patients. Finally, on the basis of a discriminate function analysis, the schizoaffective patients were more likely to be classified as having schizophrenia than a mood disorder. These findings suggest that schizoaffective disorder may represent a variant of schizophrenia in clinical symptom profiles and cognitive impairment. Validity of diagnosis Benabarre et al. (2001) analyzed the demographic, clinical and prognostic variables that determine the validity of the diagnosis of schizoaffective disorder bipolar type. They analyzed and compared 138 outpatients: 67 with type I bipolar disorder, 34

118

Ahmed Okasha

with schizoaffective disorder bipolar type and 37 with schizophrenia. Their results reaffirmed that, from the standpoints of demographics, clinical features, and prognosis, schizoaffective disorders bipolar type can be classified as a phenotypic form at an intermediate point between bipolar I disorder and schizophrenia. These results emphasize the importance of longitudinal follow-up in the diagnosis and assessment of psychotic syndromes. Although cross-sectional symptoms were closer to the schizophrenia spectrum, the course of the illness resembled more that of bipolar patients, resulting in an intermediate outcome. Concerns have been expressed about the reliability and validity of the DSM-IV criteria for schizoaffective disorder, but no systematic study has been published up to now. The Cohen’s kappa for the individual items of the DSM-IV definition of schizoaffective disorder, manic episode and major depressive episode was evaluated in 150 patients independently interviewed by two psychiatrists using the Composite International Diagnostic Interview. The Cohen’s kappa was 0.22 for the diagnosis of schizoaffective disorder, 0.71 for that of a manic episode, and 0.82 for that of a major depressive episode. The inter-rater reliability of the DSM-IV criteria for schizoaffective disorder is not satisfactory. The better outcome of DSM-IV schizoaffective disorder compared with schizophrenia seems to depend more on the inclusion (in the definition of schizophrenia but not in that of schizoaffective disorder) of the six-month duration and functional impairment criteria than on the different symptomatological patterns of the two conditions (Maj et al., 2000). Taylor and Amir (1994) used discriminate function analyses to assess the ability of systematically recorded psychopathology to distinguish 167 DSM-III schizophrenics from 74 affective patients. They divided the schizophrenics into chronic and schizoaffective sub-groups. They discriminated chronic schizophrenics from affectives, but schizoaffectives overlapped both groups. Schizoaffective unipolars were like chronic schizophrenics, and schizoaffective bipolars were like affectives. However, these discriminations also substantially overlapped, and among nonaffective positive features formal thought disorder was best at discrimination. Their findings do not fully support the present classification system, and suggest that its emphasis on hallucinations and delusions is overvalued. Outcome studies Outcome studies of schizoaffective disorder have taught us much about the longterm consequences of the syndrome, and it is argued whether they provided some indication of the potential usefulness of maintaining “schizoaffective disorder” as a diagnostic category, separate from schizophrenia and major affective disorder. In a review of outcome studies that compared schizoaffective patients to schizophrenic or affective patients, we found consistent results despite wide variations in diagnostic

119

Utility versus validity and reliability

criteria, length of follow-up and demographic characteristics. Global measures of outcome show that schizophrenic patients are more impaired than schizoaffective patients, who in turn are more impaired than affective patients. However, we realize that among schizophrenia there is a sub-group with a good outcome. Thus, outcome cannot be a criterion for delineating SA disorder. However, studies of specific outcome domains such as symptomatology, social functioning, and occupational functioning indicate that schizoaffective disorder is heterogeneous and that subtyping by polarity (e.g., schizoaffective-manic vs. schizoaffective-depressed) accounts for some of this variance. The consistency of these findings in the face of methodological variability suggests that it would be premature to classify schizoaffective patients with schizophrenia or affective disorder, also that strict diagnostic criteria for schizoaffective disorder are at best preliminary and need to be thoroughly validated (Samson et al., 1988). Grossman et al. (1991) assessed whether the outcome of schizoaffective disorder is more similar to that of schizophrenia or that of affective disorders. The authors conducted a prospective follow-up study of 101 schizoaffective, schizophrenic, bipolar manic and depressed patients assessed at three times: during hospitalization and two and four to five years later. Overall, schizoaffective patients showed some similarities to both schizophrenic and bipolar manic patients. Schizoaffective patients had somewhat better overall post-hospital functioning than patients with schizophrenia, somewhat poorer functioning than bipolar manic patients, and significantly poorer functioning than patients with unipolar depression. The data suggest that when mood-incongruent, schizophrenic-like psychotic symptoms are present in the acute phase, they predict considerable difficulty in outcome, even when affective syndromes are also present, as in schizoaffective disorder. It is likely that schizoaffective disorder is not just a simple variety of affective disorder. Impact of culture Ethnicity, culture, and socioeconomic factors all contribute to the prevalence, causation, clinical presentation, and outcome of psychiatric disorders. The Arab social historian Ibn Khaldoun in the fourteenth century was the first author to give a clear description of the relationship between mental health and culture (Murphy, 1986). He described the effect of urbanization on Islamic tribe warriors when they moved from nomadic life to life in towns. The movement was associated with an increase in the prevalence of psychological ailments: namely jealousy, suspiciousness, selfindulgence and fear of others. He viewed this behavior as a reaction to the change of social structure. In his view, the tribal system failed to adjust to the process of urbanization. Such failure was, in Ibn Khaldoun’s view, at the origin of decline of the Islamic civilization.

120

Ahmed Okasha

Eastern cultures emphasize social integration more than autonomy, i.e., the family and not the individual is the unit of society. Dependence is more natural and infirmity less alien in these cultures. When affiliation is more important than achievement, how one appears to others becomes vital and shame becomes a driving force more than guilt. In the same manner, physical illness and somatic manifestations of psychological distress become more understood, acceptable and evoke a caring response rather than a vague complaint of psychological symptoms, which can be either disregarded or considered a stigma of being “soft” or worse, even “insane” (Table 6.1) (Okasha, 2000). The World Health Organization (WHO) collaborating centre in the Eastern Mediterranean Regional Office (EMRO) at the Institute of Psychiatry, Ain Shams University, Cairo, within the WHO collaborating research and field study on ICD10 collected data from eight Arab centers, which investigated a total of 233 patients using the local psychiatric interview schedules and diagnosed according to ICD-10 criteria. Inter-rater reliability was found to range between an almost perfect (0.81–1) to substantial agreement (0.61–0.80) (using the kappa coefficient) in diagnosing organic mental disorders, substance-use disorders, schizophrenic, schizotypal and delusional disorders, affective disorders, and neurotic and stress-related disorders. The categories of psychological development and child and adolescent disorders were diagnosed less frequently and the agreement between raters was lower. Though no culture-bound syndromes were encountered in any of the centers, difficulties in diagnosis using the research criteria were identified in the domain of simple schizophrenia and dissociative versus conversion disorders. None of the schizophrenic spectrum was diagnosed as schizoaffective, which may clarify the way mood and thought disorders are diagnosed in a certain cultural context (Okasha et al., 1993). To this point, Wig comments that Third World psychiatry is not fully satisfied by some areas in the classification such as child psychiatry and personality disorders or the concepts underlying the classification of “hysterical disorders,” a domain that still seems to be influenced by the American and European schools of thought (Wig, 1990). Organic, psychotic and mood disorders enlisted in the ICD-10, especially with the inclusion of the category of acute polymorphic psychotic disorders, have offered a place for the ever-so-famous culturally bound syndromes of the Third World, which are characterized by a polymorphic clinical picture of acute onset, usually following stress, and resolving almost completely after some time. Although the clinical picture might be colored by local beliefs or superstitions, it is basically a stress reaction that acquires its characteristics from what is prevalent in a particular culture. A prominent example thereof is the possession syndrome, which can take several clinical pictures with a common denominator of being possessed by a spirit that either gives

121

Utility versus validity and reliability Table 6.1. Psychosocial markers in traditional versus Western societies (Okasha, 2000).

Traditional society

Western society

Family and group oriented Extended family (not as geographical as before, but conceptual)

Individual oriented Nuclear family

Status determined by age and position in the family, care of elderly

Status achieved by own efforts

Relationship between kin obligatory

Relationships determined by individual choice

Arranged marriages with an element of choice dependent on interfamilial relationship

Choice of marital partner determined by interpersonal relationship

Extensive knowledge for distant relatives

Extensive knowledge restricted only to close relatives

Decision making dependent the family

Autonomy of individual

Locus of control external

Locus of control internal

Respect, and holiness for the decision of the physician

Doubt in doctor–patient relationship

Malpractice suing is rare

Suing for malpractice is common

Deference is God’s will

Deference self-determined

Doctor–patient relationship is still healthy

Mistrust in doctor–patient relationship

Individual can be replaced; the family should continue and the pride is in the family tie

The individual is irreplaceable, self-pride

Pride in family care for the mentally ill patient

Community help responsible for care of mentally ill

Dependence on God in health and disease; attribution of illness and recovery to God’s will

Health and disease are self-determined

orders to the patient or induces a marked change of behavior. Although the clinical picture is that of florid psychotic symptoms, the condition usually resolves completely with removal or management of the stressor (Okasha, 1992). Islamic cultures value the collectivity of the community rather than the individuality of its member citizens. Decisions are not taken on an individual level but on a familial, tribal or communal level to the best of the perceived collective interest. Regarding family structure in traditional societies, namely the extended family, decision making is group and family oriented and the Western attitude towards individual autonomy does not exist. The external laws of control, the dependence on God in health and disease and attribution of illness and recovery to God’s will

122

Ahmed Okasha

maintain a healthy doctor–patient relationship; which makes trust, confidence and compliance a characteristic attitude in traditional societies. The following is a very crude contrast to highlight the main differences. However, it should be noted that those differences are the mainstream norm and not an absolute description of a stereotyped behavior. Traditional cultures and mental health Many countries share a common social and cultural heritage. Language, religion, and geography are some of the shared features shaping many cultures. Islamicspeaking countries, albeit with some variation, also share the socio economic and cultural characteristics of the rest of the developing world. However, homogeneity is by no means the norm. Agrarian regions around the Nile, Palestine, and the Euphrates have a long-documented history of settled communities and civilizations. These civilizations almost perished around three centuries after the birth of Christianity in favor of a long dark period lasting until the Islamic enlightenment of the Middle Ages. From the fifteenth century onwards, Islamic countries went again into four centuries of stagnation under Ottoman rule, caught the dawn of industrial revolution very late, and are still tentatively grappling with democratic values. The rest of the Islamic world is an enormous desert with a long history of nomadic life that only very recently started to settle in urban conglomerations fuelled by petrodollars in the East and a drive for employment and economic prosperity in the West. In the nineteenth and the first half of the twentieth centuries, Western imperialism set the tone for divergence of language and foreign cultural influence, especially in North West Africa (Amin et al., 2001). Both the convergent and divergent backgrounds of Islamic countries cannot be ignored in any account of the origins and characteristics of psychiatric disorders in this particular region of the world. Convergence can only be exemplified by the eruption of Islamic fundamentalism almost simultaneously all over the Islamic world. Divergence can be exemplified again by the contrast between highly westernized communities, for example in Lebanon, and the highly traditional conservative communities that occupy large parts of the Arabian Peninsula. Since then, a steady stream of publications indicated that depression in Africans is not rare but that it takes a different form from that observed in Europeans. Again, less verbalization of depressive feelings, rare self-castigation, frequent use of projection, prominence of hypochondriasis, and the scarcity of suicide were all described as characteristic of African depression. Still later, the balance tipped in the opposite direction so that an excess of depressive disorders has been diagnosed in African subjects (Orley and Wing, 1979) and Arab women (Ghubash et al., 1992)

123

Utility versus validity and reliability

compared to other similar epidemiological surveys from the West, e.g., Bebbington et al. (1981), which makes it difficult to diagnose schizodepressive disorder. The Arab family is the main social institution that has inputs relevant to clinical psychiatry. It contributes much more than the Western family to mental development, illness behavior, pattern, and management. The roles of schools and outof-home care institutions are more significant in the West. Recent budget realignments in Western countries drastically compromised their social welfare system, and services for the elderly, the poor and the sick had to be taken over by “someone else.” Western mental health professionals rediscovered the role of the family in this respect, which had long been abandoned for the sake of individuation of human beings. The family care, which was phased out through industrialization and communism alike, is now actively re-sought (El-Islam, 1998a; 2001). Individuals who have socially unconventional behavior, e.g., drinking alcohol, premarital sexual practice, or the non-observance of public rules for the fasting month may be taken to psychiatrists by their family elders in some Gulf communities for assessment of their mental state. The psychologization of their social deviance (El-Islam, 1998b) not only preserved a facade of adherence to the Islamic code but also avoided the punishment of code breakers by the legal system based on this code and saved the family the shame and inability to arrange the marriages of code breakers. Historically, a number of studies have been reported comparing the manifestations of schizophrenia across cultures (Varma, 2000). Differences in the symptomatology of schizophrenia in Arab culture have been a subject of many investigations, with interesting results. While Taleb et al. (1996) had found few clinical differences in his comparative study between schizophrenic patients from Morocco and France, Gawad et al. (1981) showed a number of important differences in the diagnosis of schizophrenia in Egypt compared to the USA and the UK. They studied the cross-national differences in symptom importance in the diagnosis of schizophrenia among the three countries. They found that restriction and incongruity of affect ranked first in the Egyptian study compared to the British and American ones. Their results are in agreement with other studies that stated that what is normal emotional expression in an Anglo-Saxon culture may suggest a schizoid reduction of emotional response in a Mediterranean culture (Lehman, 1967). The top ten symptoms in the Egyptian hierarchy for diagnosis of schizophrenia were incongruity and restricted affect, formal thought disorder, thought block, thought withdrawal, incoherence, passivity feeling, neologism, hallucination, delusions, and ideas of reference. While in the British study, formal thought disorder ranked first, followed by incongruity of affect, neologism, thought block, passivity of feeling, paranoid delusions, other delusions, thought withdrawal, and ideas of reference (Willis and Bannister, 1965). The Americans ranked symptoms

124

Ahmed Okasha

of importance for diagnosis of schizophrenia as follows: formal thought disorder, delusions, paranoid delusions, incongruity of affect, hallucinations, ideas of reference, neologism, depersonalization and thought block (Khalil, 2001; Edwards, 1972). The cultural and religious heritage absorbs many features, which would otherwise be considered symptomatic of a psychiatric disorder (El-Islam, 1980; Barakat, 1993; Frindlay, 1994). Impact of culture on diagnosis The past 50 years have seen a response to problems of lack of agreement between psychiatrists, and great efforts were and are constantly being poured into the development of a unified nosological language that enables a psychiatrist in the Far East of the world to understand what his/her colleague in the Far West is talking about. This need is ever more urgent in a region that shares a common language, common culture, history, and social context (Okasha and Seif El Dawla, 1992). The religious nature of upbringing and education in Egypt, the emphasis on religious rituals, and the warding-off of blasphemous thoughts through repeated religious phrases such as “I seek refuge with the Lord from the accursed Satan” can explain the high prevalence of religious obsessions and repeating compulsions among our Egyptian sample, even if the subjects are not practicing their religious duties. To elaborate further, Moslems, who constitute almost 90% of the Egyptian population, are required to pray five times a day. Each prayer is preceded with a ritualistic cleansing process (El Woodoo or ablution), which implies the washing of several parts of the body in specific order, each three times. This ablution is annulled by any form of excretion or ejaculation and, for some radical Moslems, by any contact with the opposite sex even as slight as a handshake. Women are not allowed to pray or touch the Koran during their menstruation, after which they should clean their bodies through a ritualistic bath. The prayers themselves are different in length and consist of certain phrases and “souras” from the Holy Koran that have to be read in a certain sequence. The emphasis on cleanliness or ritual purity is the cornerstone of most of the compulsive rituals. The number of prayers and the verbal content can be the subject of scrupulousness, checking, and repetition. The ritualistic cleansing procedures also can be a source of obsessions and compulsions about religious purity, e.g., in some compulsives the color red (reminder of menstruation) may trigger a compulsive washing. Another evidence of the religious connotation inherent in obsessive compulsive disorder (OCD) in Moslem culture lies in the term “El Weswas”. This term is used in reference to the devil, and at the same time is used as a name for obsessions.

125

Utility versus validity and reliability

It is also characteristic of a conservative society like that of Egypt to expect sexual obsessions to be among the most frequent in female patients. Although it is accepted socially (but prohibited religiously) for Egyptian males to have a wide range of sexual freedom in all stages of their lives, sexual matters remain an issue of prohibition, sin, impurity, and shame for Egyptian women. This is to the extent that radical Moslem individuals demand an ablution before prayer, for both men and women, if the man touches or salutes a woman with as little as a handshake. The female gender is surrounded by so many religious and sexual taboos that the issue becomes a rich pool for worries, ruminations, and cleansing compulsions in women susceptible to developing OCD. Christians represent approximately one-tenth of the population in Egypt, which is equivalent to their number in our sample (about 10%). The presenting symptoms were almost similar in terms of obsessions, where religious and sexual thoughts were predominant. However, there was a marked difference in rituals, which were more frequent in Moslems, emphasizing the role of ritualistic Islamic upbringing as compared with Christian upbringing in our community (Okasha et al., 1994). The current classification of hysterical symptoms in this context is unsatisfactory. The classical division into conversion and dissociation is arbitrary and unsatisfactory. Saxena (1987), for example, referred to hysterical fits, one of the most common presentations of hysteria in EMRO, as really a combination of conversion and dissociation. Also, Okasha (1988) reports hysterical symptoms to be one of the most common “neurotic” manifestations in Egypt, amounting to approximately 11% of all patients attending the Ain Shams psychiatric outpatient clinic. It occurs in all social classes and its presentation is multiple rather than monosymptomatic. It might thus be more useful to classify into acute and chronic hysterical conditions, rather than conversion and dissociation, as the latter division does not help either in management or in predicting outcome (Okasha et al., 1993). Another domain in which regional psychiatrists found a difficulty in using the ICD-10 was that of negative symptoms, and the fact that they constituted a main bulk in the diagnosis of several disorders such as simple schizophrenia, schizotypal disorder, and residual schizophrenia, leaving the differentiation to the duration and past psychiatric history of the patient, both of which can be unreliable or even unavailable in cases, for example, where the patient is the sole informant and has no adequate information about those two points. Within the three diagnostic categories, simple schizophrenia seemed to occupy a unique position. The ICD-10 states that “the subject must never have met the criteria of schizophrenia or another psychotic disorder.” It is the experience of psychiatrists working in EMRO that cases that later develop into frank simple schizophrenia could present at an earlier stage of the disorder with a prodroma of vague hypochondriasis and some forms of

126

Ahmed Okasha

thought disorder, which later gives way to a marked blunting of affect, avolition and anhedonia. We feel that the current description of simple schizophrenia needs a link to the broader group of schizophrenias in the ICD-10, especially that as it is not mentioned at all in DSM-IV. Expressed emotions Family climate plays an important role in the outcome of schizophrenic patients (Frindlay, 1994). In this respect, the concept of expressed emotions has gained respectable ground in the field of psychiatry. A series of studies have demonstrated that family EE (expressed emotions) predicts patient’s symptomatic relapse, both in Anglo and American settings (Brown et al., 1962; Wig et al. 1987; Vaughn and Leff, 1976). Abol-Magd (1993) and Kamal (1995) unexpectedly reported that there is no association between emotional overinvolvement and relapse in Egyptian schizophrenic patients. Moreover they proved that family expressed emotions, mainly criticism and hostility, seem to be an independent risk factor for relapse of their patients. Their results seem to be similar to the findings of Maosheng et al. (1998) in the Chinese culture. The presence of high warmth appeared to be associated with lower admission rates and better psychosocial adjustment (Abol-Magd, 1993; Bertrando et al., 1992). Leff (1989) suggested that high values of warmth may have an opposite effect to that of high EE, enhancing the patient’s psychosocial adjustment. Okasha et al. (1994) conducted a study to determine the value of families’ expressed emotion and patients’ perception of family criticism in predicting relapse in Egyptian depressed patients and to evaluate transcultural differences in the assessment of these measures. The relation of family criticism to relapse was statistically significant. Although this result replicates previous findings, the criticism score that differentiated relapsers and nonrelapsers was a score of 7, which is much higher than reported in previous studies. Expressed emotion is a prognostic factor that should be assessed with consideration of the specific culture and infra-familial patterns. Conclusion Different studies regarding the concept of schizoaffective (SA) disorder show diversity among authors in considering it a subtype of schizophrenia or bipolar or in between. Genetic and outcome studies did not reveal a definite conclusion. The reliability and validity of the concept throws some doubt as to its independence. The symptomatology of SA disorder is confounded by the overlap of mood and psychotic symptoms. The reliability and validity of the term SA disorder is still controversial. The fact that schizophrenia, bipolar disorder and SA disorder can be

127

Utility versus validity and reliability

treated with mood stabilizers and second-generation antipsychotics induced clinicians to over-diagnose SA disorder for the sake of simplicity and utility. Junior psychiatrists are inclined to diagnose SA disorder whenever they find an admixture of schizophrenia and mood symptoms, regardless of the temporal relations of symptoms, instead of looking deeply and examining thoroughly their patients to differentiate between thought or mood disorders. Some academics in some cultures deny junior psychiatrists the opportunity to diagnose SA disorder unless their clinical experience is mature and sufficient to make this diagnosis. As seen from this text, in some cultures the depressive, manic, or emotional blunting are perceived in a different way, that makes the dependence on symptoms rather futile and may lead to a variable diagnosis. The best example of the sociocultural influence on the diagnosis of SA is revealed in studies of EE and the presence of relapse (Okasha et al., 1993). An EMRO study during the field studies of ICD-10 did not show any SA diagnosis among Arab psychiatrists with a high reliability in their diagnoses. The role of the family in the detection, genesis, diagnosis and perception of symptoms may affect the diagnosis of SA disorder. Outcome studies in schizophrenia and depression show better outcome in developing than developed countries, in spite of less medication, less hospitalization and less maintenance. This can be attributed to the family role and their beliefs regarding the origin of mental illness. Being a member of the advisory group of DSM and ICD, I can predict the tendency to change the current categorical diagnosis into a dimensional one. This may create an important position for SA in the psychotic spectrum. This initiative may also give more utility, reliability and validity for the diagnosis of SA disorder and many other psychiatric disorders, especially neurotic related disorders.

R E F E R E N C ES Abol-Magd, S. (1993). Psychiatric Assessment of Families of Schizophrenic Patients. MD thesis, Cairo University, Faculty of Medicine. Amin, Y., Hamdi, E. and AbouSaleh, M. (2001). Depression in the Arab world, In Images in Psychiatry: An Arab Perspective, ed. A. Okasha and M. Maj. Cairo: World Psychiatric Association, pp. 89–122. Angst, J. (1966). Zur Ätiologie und Nosologie endogener depressiver Psychosen. Eine genetische, soziologische und klinische Studie. Berlin: Springer. Angst, J. (1973). The aetiology and nosology of endogenous depressive psychoses. Foreign Psychiatry, 2, 1–108. Barakat (1993). The Arab World : Society, Culture, and State. Berkeley and Los Angeles, California: University of California Press, pp. 97–118.

128

Ahmed Okasha Baron, M., Gruen, R., Asnis, L. and Kane, J. (1982). Schizo-affective illness, schizophrenia and affective disorders: Morbidity risk and genetic transmission Acta Psychiatrica Scandinavica, 65, 253– 62. Beatty, W. W., Jocic, Z., Monson, N. and Staton, R. D. (1993). Memory and frontal lobe dysfunction in schizophrenia und schizoaffective disorder. The Journal of Nervous and Mental Disease, 181, 448–53. Bebbington, P., Hurry, J., Tennant, C., Sturt, E., and Wing, J. (1981). Epidemiology of mental disorders in Camberwell. Psychological Medicine, 11, 561–79. Benabarre, A., Vieta, E., Colom, F. et al. (2001). Bipolar disorder, schizoaffective disorder and schizophrenia: epidemiologic, clinical and prognostic differences. European Psychiatry, 16 (3), 167–72. Berrettini, W. (1996). Genetic studies of bipolar disorders: new and recurrent findings. Molecular Psychiatry, 1 (3), 172–3. Berrettini, W. (2000). Susceptibility loci for bipolar disorder: overlap with inherited vulnerability to schizophrenia. Biological Psychiatry, 47 (3), 245–51. Berrettini, W. (2003). Bipolar disorder and schizophrenia: Not so distant relatives? World Psychiatry, 2, 68–72. Bertelsen, A. and Gottesman, I. I. (1995). Schizoaffective psychosis: Genetical clues to classification. American Journal of Medical Genetics, 60, 7–11. Bertrando, P., Beltz, J., Bressi, C. et al. (1992). Expressed emotions and schizophrenia in Italy: A study of an urban population. British Journal of Psychiatry, 161, 223–9. Bleuler, E. (1937). Lehrbuch der Psychiatrie, 6th edn. Berlin: Springer. Bleuler, E. (1983). Lehrbuch der Psychiatrie, 15th edn. Berlin: Springer. Bornstein, R. A., Nasrallah, H. A., Olson, S. C. et al. (1990). Neuropsychological deficit in schizophrenic subtypes: Paranoid, nonparanoid and schizoaffective subgroup. Psychiatry Research, 31, 15–24. Brown, G. W., Birely, J. and Wing, J. K. (1962). Influence of family life on the course of schizophrenia. British Journal of Psychiatry, 151, 156–73. Clayton, P.J., Rodin, L. and Winokur, G., (1968). Family history studies-III. Schizoaffective disorder, clinical and genetic factors including a one- to two-year follow-up, Comprehensive Psychiatry, 9, 31–49. Coryell, W. (1986). Schizoaffective and schizophreniform disorders. In Medical Basis of Psychiatry, ed. G. Winokur and P. Clayton. Philadelphia: WB Saunders Co., pp. 102–14. Coryell, W., Grove, W., vanEerdewegh, M., Keller, M. and Endicott, J. (1987). Outcome in RDC schizoaffective depression: the importance of diagnostic subtyping. Journal of Affective Disorders, 12 (1), 47–56. Coryell, W. and Zimmerman, M. (1988a). The heritability of schizophrenia and schizoaffective disorder. A family study. Archives of General Psychiatry, 45, 323–7. Coryell, W. and Zimmerman, M. (1988b). Diagnosis and outcome in schizoaffective depression: a replication. Journal of Affective Disorders, 15, 21–7. Edwards, G. (1972). Diagnosis of schizophrenia: An Anglo-American comparison. British Journal of Psychiatry, 120, 385–90.

129

Utility versus validity and reliability El-Islam, M. F. (1980). Arabic cultural psychiatry. Transcultural Psychiatric Research Review, 19: 5–24. El-Islam, M. F. (1998b). Clinical applications of cultural psychiatry in Arab gulf communities. In Clinical Methods in Transcultural Psychiatry, ed. S. O. Okpaku. Washington DC: American Psychiatric Press, Inc., pp. 155–70. El-Islam, M. F. (1998a). Rediscovery of the traditional role of the family. Transcultural Psychiatry Newsletter, 16, 3. El-Islam, M. F. (2001). Social psychiatry and impact of religion. In Images in Psychiatry. An Arab Perspective, ed. A. Okasha and M. Maj. Cairo: World Psychiatric Association, pp. 21– 35. Endicott, J., Nee, J., Coryell, W. et al. (1986). Schizoaffective, psychotic and non-psychotic depression: Differential familial association. Comprehensive Psychiatry, 27, 1–13. Evans, J. D., Heaton, R. K., Paulsen, J. S. et al. (1999). Schizoaffective disorder: a form of schizophrenia or affective disorder? Journal of Clinical Psychiatry, 60, 874–82. Feighner, J. P., Robins, E., Guze, S. B. et al. (1972). Diagnostic criteria for use in psychiatric research. Archives of General Psychiatry, 26, 57–63. Frindlay, A. M. (1994). The Arab World. Guildford and King’s Lynn, UK: Biddles Ltd, pp. 126–89. Gawad, M., Rakhawy, Y., Mahfouz, R. and Howaidy, M. (1981). Relative symptom importance in the diagnosis of schizophrenia. The Egyptian Journal of Psychiatry, 4 (1), 39–56. Gershon, E. S., DeLisi, L. E., Hamovit, J. et al. (1988). A controlled family study of chronic psychoses. Archives of General Psychiatry, 45, 328–36. Ghubash, R., Hamdi, E. and Bebbington, P. (1992). The Dubai Community Psychiatric Survey: I. Prevalence and socio-demographic correlates. Social Psychiatry and Psychiatric Epidemiology, 27, 53–61. Grossman, L. S., Harrow, M., Joseph, F. et al. (1991). Outcome of schizoaffective disorder at two long-term follow-ups: Comparisons with outcome of schizophrenia and affective disorders. American Journal of Psychiatry, 148, 1359–65. Harrison, G., Pope, J. R., Joseph, F., Bruce, M. and Axelrod, A. C. S. W. (1980). Schizoaffective disorder: an invalid diagnosis? A comparison of schizoaffective disorder, schizophrenia, and affective disorder. American Journal of Psychiatry, 137, 8. Hoch, A. (1921). Benign Stupor: A Study of New Manic-Depressive Reaction Type. New York: Macmillan. Hunt, R. C. and Appel, K. E. (1936). Prognosis in the psychoses lying midway between schizophrenia and manic-depressive psychoses. American Journal of Psychiatry, 93, 313–39. Jeste, D. V., Heaton, S. C., Paulsen, J. S. et al. (1996). Clinical and neuropsychological comparison of psychotic depression with nonpsychotic depression and schizophrenia. American Journal of Psychiatry, 153, 490–6. Kamal, A. (1995). Variables in expressed emotions associated with relapse: A comparison between depressed and schizophrenic samples in the Egyptian community. Current Psychiatry, 2 (2), 211–15. Kaplan, H. I. and Saddock, B. (2000). Comprehensive Textbook of Psychiatry, 7th edn. Lippincott: Williams & Wilkins. Kasanin, J. (1933). The acute schizoaffective psychoses. American Journal of Psychiatry, 13, 97–126.

130

Ahmed Okasha Keck, P. E. Jr., Reeves, K., Harrigan, E. and Ziprasidone study group (2001). Ziprasidone in the short-term treatment of patients with schizoaffective disorder: Results from two double-blind, placebo-controlled, multicenter studies. Journal of Clinical Psychopharmacology, 21, 27–35. Kendler, K. S. (1996). Major depression and generalized anxiety disorder. Same genes, (partly) different environments – revised. British Journal of Psychiatry, 168 (Suppl. 30), 68–75. Kendler, K. S., McGuire, M., Gruenberg, A. M. and Walsh, D. (1995a). Examining the validity of DSM-III-R schizoaffective disorder and its putative subtypes in the Roscommon family study. American Journal of Psychiatry, 152, 755–64. Kendler, K. S., Neale, M.C. and Walsh, D. (1995b). Evaluating the spectrum concept of schizophrenia in the Roscommon family study. American Journal of Psychiatry, 152, 749–54. Kendler, K. S., Spitzer, R. L. and Williams, J. B. W. (1989). Psychotic disorders in DSM-III-R. American Journal of Psychiatry, 146, 953–62. Khalil, A. (2001). Schizophrenia across Arab culture. In Images in Psychiatry: An Arab Perspective, ed. A. Okasha and M. Maj. Cairo: World Psychiatric Association, pp. 65–88. Kirby, G. H. (1913). The catatonic syndrome and its relation to manic-depressive insanity. The Journal of Nervous and Mental Disease, 40, 694–704. Lapierre, Y. D. (1994). Schizophrenia and manic-depression: separate illnesses or a continuum? Canadian Journal of Psychiatry, 39, 559–64. Leff, J. (1989). Controversial issues and growing points in research on relatives’ expressed emotions. Internal Journal of Social Psychiatry, 35, 133–45. Lehman H. E. (1967). Clinical features of schizophrenia. In Comprehensive Text Book of Psychiatry, ed. A. M. Freedman and H. I. Kaplan. Baltimore: The Williams and Wilkins Company. Levinson, D. F. and Levitt, M. E. M. (1987). Schizoaffective mania reconsidered. American Journal of Psychiatry, 144, 415–25. Levinson, D. F., Umapathy, C. and Musthaq, M. (1999). Treatment of schizoaffective disorder and schizophrenia with mood symptoms. American Journal of Psychiatry, 156, 1138–48. Maier, W. Lichtermann, D., Minges, J. et al. (1992). Schizoaffective disorder and affective disorders with mood incongruent psychotic features: Keep separate or combine? Evidence from a family study. American Journal of Psychiatry, 149, 12. Maj, M., Starace, F. and Pirozzi, R. (1991). A family study of DSM-III-R schizoaffective disorder, depressive type, compared with schizophrenia and psychotic and nonpsychotic major depression. American Journal of Psychiatry, 148, 612–16. Maj, M., Pirozzi, R., Formicola, A. M., Bartoli, L. and Bucci, P. (2000). Reliability and validity of the DSM-IV diagnostic category of schizoaffective disorder: preliminary data. Journal of Affective Disorders, 57, 95–8. Marneros, A. (2003a). Schizoaffective disorder: clinical aspects, differential diagnosis, and treatment. Current Psychiatry Reports, 5, 202–5. Marneros, A. (2003b). The schizoaffective phenomenon: the state of the art. Acta Psychiatrica Scandinavica, 108 (Suppl. 418), 29–33. Marneros, A., Pillmann, F., Haring, A. and Balzuweit, S. (2000). Die akuten vorübergehenden psychotischen Störungen. Fortschritte der Neurologie-Psychiatrie, 68, 22–25. Moasheng, R. and Zaijn Hand Mengze, X. (1998). Emotional expression among relatives of schizophrenic patients. Chinese Journal of Psychiatry, 31, 237–9.

131

Utility versus validity and reliability Murphy, H. B. M. (1986). The historical development of transcultural psychiatry. In Transcultural Psychiatry, ed. J. L. Cox, London: Croom Helm. Okasha, A. (1988). Clinical Psychiatry. Cairo: Anglo Egyptian Bookshop, pp. 1151–3. Okasha, A. (1992). Mental health services in Egypt. World Psychiatry, 2, 15–17. Okasha, A. (2000). The Impact of Arab culture on psychaitric ethics. In Ethics Culture and Psychiatry: International Perspectives, ed. A. Okasha, J. Arboleda-Florez and N. Sartorius. USA: American Psychiatric Press, Inc., pp. 15–28. Okasha, A., Sadek, A., Al Haddad, M. and Abdel, Maugoud, M. (1993). Reliability of psychiatric diagnosis among Arab psychiatrists: a comparative study between ICD-9, ICD-10 and DSMIII-R. British Journal of Psychiatry, 162, 621–6. Okasha, A. and Seif El Dawla, A. (1992). Reliability of ICD-10 research criteria: an Arab perspective. Acta Psychiatrica Scandinavica, 86, 484–8. Okasha, A., El Akabawi, A., Wilson, A., Youssef, I. and Seif El Dawla, A. (1994). Expressed emotion, perceived criticism, and relapse in depression: a replication in an Egyptian community. American Journal of Psychiatry, 151, 1001–5. Orley, J. and Wing, J. K. (1979). Psychiatric disorders in two African villages. Archives of General Psychiatry, 36, 513–20. Pichot, P. (1986). A comparison of different national concepts of schizoaffective psychosis. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin: Springer. Pillmann, F., Haring, A., Balzuweit, S., Bloink, R. and Marneros, A. (2001). Concordance of acute and transient psychotic disorders and cycloid psychoses. Psychopathology, 34, 305–11. Procci, W. R. (1976). Schizoaffective psychosis: fact or fiction? A survey of the literature. Archives of General Psychiatry, 33, 1167–78. Retterstol, N. (1991). Course and outcome in paranoid disorders. Psychopathology, 24, 277–86. Robins, R. and Guze, S. (1970). Establishment of diagnostic validity in psychiatric illness: its application to schizophrenia. American Journal of Psychiatry, 126, 983–7. Samson, J. A., Simpson, J. C. and Tsuang, M. T. (1988). Outcome studies of schizoaffective disorders. Schizophrenia Bulletin, 14, 543–54. Sartorius, N. (1990). Sources and traditions of psychiatric classification: introduction. In Sources and Traditions of Classification in Psychiatry, ed. N. Sartorius, A. Jablensky, D. A. Regier, J. D. Bruke and R. M. A. Hirschfeld. USA: Hogrefe & Huber Publishers, pp. 1–6. Saxena, S. (1987). Simple dissociative disorder. A subcategory in DSM-III-R. American Journal of Psychiatry, 144, 524–5. Spitzer, R. L., Endicott, J. and Robins, E. (1975). Research Diagnostic Criteria (RDC) for a Selected Group of Functional Disorders, 2nd edn. New York: Biometrics Research, New York State Psychiatric Institute. Spitzer, R. L., Endicott, J. and Robins, E. (1978). Research Diagnostic: Criteria (RDC) for a Selected Group of Function Disorders, 3rd edn. New York: Biometrics Research, New York State Psychiatric Institute. Strakowski, S. M., Keck, P. E. Jr., Sax, K. W., McElroy, S. L. and Hawkins, J. M. (1999). Twelvemonth outcome of patients with DSM-III-R schizoaffective disorder: comparison to matched patients with bipolar disorder. Schizophrenia Research, 35, 167–74.

132

Ahmed Okasha Taleb, M., Rouillon, F., Petitjean, F. and Gorwood, P. (1996). Cross-cultural study of schizophrenia. Psychopathology, 29, 85–94. Taylor, M. A. (1992). Are schizophrenia and affective disorder related? A selective literature review. American Journal of Psychiatry, 149, 22–32. Taylor, M. A. and Amir, N. (1994). Are schizophrenia and affective disorder related? The problem of schizoaffective disorder and the discrimination of the psychoses by signs and symptoms. Comprehensive Psychiatry, 35, 420–9. Tsuang, M. T. (1979). Schizoaffective disorder: dead or alive. Archives of General Psychiatry, 36, 633–4. Vaillant, G. (1963). Manic-depressive heredity and remission in schizophrenia. British Journal of Psychiatry, 109, 746–9. Vaughn, G. E. and Leff, J. P. (1976). The influence of family and social factors on the course of psychiatric illness: a comparison of schizophrenia and depressed neurotic patients. British Journal of Psychiatry, 129, 124–37. Varma, V. K. (2000). Transcultural perspective on schizophrenia, epidemiology, manifestation and outcome. Report of section of transcultural psychiatry. Cairo: World Psychiatric Association. Walinder, J. (1972). Recurrent familial psychosis of the schizoaffective type. Acta Psychiatrica Scandinavica, 48, 274–83. Whaley, A. L. (2002). Symptom clusters in the diagnosis of affective disorder, schizoaffective disorder, and schizophrenia in African-Americans. Journal of the National Medical Association, 94, 313–19. Wig, N. (1990). The third world perspective on psychiatric diagnosis and classification. In Sources and Traditions of Classification in Psychiatry, ed. N. Sartorius, A. Jablensky, D. A. Regier, J. D. Burke and R. M. A. Hirschfeld. USA: Hogrefe & Huber Publishers, pp. 181–210. Wig, N. N., Mendon, D. K., Bedi, H. et al. (1987). Expressed emotions and schizophrenia in North India: Cross cultural transfer of ratings of relatives. British Journal of Preventive and Social Medicine, 16, 55–68. Williams, P. V. and McGlashan, T. H. (1987). Schizoaffective psychosis, I: Comparative long-term outcome. Archives of General Psychiatry, 44, 130–7. Willis, J. H. and Bannister, D. (1965). The diagnosis and treatment of schizophrenia: a questionnaire study of psychiatric opinion. British Journal of Psychiatry, 111, 1165–71. Zendig, E. (1909). Beiträge zur Differentialdiagnostik des manisch-depressiven Irreseins und der Dementia praecox. Allgemeine Zeitschrift für Psychiatrie, 6, 47–9.

7

Phenomenological approaches to the schizoaffective spectrum William Coryell University of Iowa, Department of Psychiatry, Iowa City

Classification The coexistence of symptoms that characterize schizophrenia with those that characterize affective disorder poses a number of knotty questions. Schizoaffective disorder is the most enduring of the various labels used for this overlap but the definitions of this term have varied and have identified groups that have not overlapped well (Brockington and Leff, 1979). Most have specified either the presence of schizophrenia-like psychotic features within an affective syndrome, or the occurrence of psychotic features when affective symptoms are quiescent. Much of the research over the past several decades has employed the definition contained in the Research Diagnostic Criteria (RDC) (Spitzer et al., 1978). In this system the presence within a major depressive or manic syndrome of a Schneiderian first-rank symptom such as a delusion of passivity or thought broadcasting, persistent non-affective hallucinations, or definite instances of formal thought disorder, warrants a diagnosis of schizoaffective disorder. The presence of any clear psychotic feature when affective symptoms are absent also suffices and this defines the mainly schizophrenic subtype of schizoaffective disorder. In contrast, the DSM-IV definition is much narrower and encompasses only those who would meet the RDC for the mainly schizophrenic subtype of schizoaffective disorder. Competing views Schizoaffective disorder may be considered in one of three or more ways. According to a dichotomous view, the diagnosis applies to individuals who, in fact, have either schizophrenia or an affective disorder. Given the lack of diagnostically useful laboratory tests and the scarcity of diagnostically selective treatments, clarification of this uncertainty in a given case must await observation over time. The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

133

134

William Coryell

A fully remitting course with episodes consistently comprised of affective symptoms indicates an affective disorder. One that settles into a chronic psychosis without full remissions would, instead, be viewed by most as schizophrenia. Alternatively, spectrum views have taken two forms. One, articulated by Crow (1987, 1990a, 1990b), posits that a single genetic diathesis underlies all functional psychosis and that variations in this gene account for differences in the severity of eventual defect states. According to an alternative position, genetic and/or perinatal diatheses predisposing to a schizophrenia-like illness coexist with diatheses predisposing to affective disorder symptoms. The relative predominance of one over the other then determines whether the phenomenology and eventual outcome for a given individual is more suggestive of schizophrenia or of affective disorder. The possibility that schizoaffective disorder comprises a third psychosis has been generally dismissed because the preponderance of relevant family studies have failed to show significantly higher rates of schizoaffective disorder among the relatives of schizoaffective probands than among the relatives of either schizophrenic or affective disorder probands (Tsuang et al., 1977; Baron et al., 1982; Andreasen et al., 1987; Coryell and Zimmerman, 1988b; Kendler et al., 1995). Moreover, the single gene view is difficult to reconcile with the numerous family studies that have found far more schizophrenia, and far less affective disorder, among the relatives of schizophrenia probands than among the relatives of probands with affective disorder. It may well be that a hypothesis that combines these positions would best fit the overall findings of twin, family, and follow-up studies. Specifically, some individuals with a diagnosis of schizoaffective disorder may have an underlying pathophysiology that resembles that of patients with psychotic affective disorders; others may have, at this fundamental level, schizophrenia, while still others may have some pathophysiological admixture of the two. Few would argue that current diagnostic criteria are optimal, or that they are routinely applied with high precision. Indeed, the few studies that have systematically considered the diagnostic stability of schizoaffective disorder have shown substantial shifts over time to diagnoses of either affective disorder or of schizophrenia (Coryell and Zimmerman, 1988b; Fennig et al., 1994; Winokur et al., 1996). On the other hand, the majority of patients with a baseline diagnosis of schizoaffective disorder appear to retain this diagnosis on follow-up. Prediction of archetypal outcomes Historical and cross-sectional features

Uncertainty over its nature notwithstanding, clinicians who treat patients with schizoaffective disorder must recommend treatment and address the questions of patients and family members concerning eventual outcomes. Both prognostic

135

Phenomenological approaches

formulations and treatment selection hinge on judgments as to whether the patient is likely to follow a course indicative of an affective disorder or one of schizophrenia. If the former seems more likely, the clinician will tend to emphasize modalities, such as antidepressants, mood stabilizers, and ECT that are thought to be of more benefit for affective disorder than for schizophrenia. Follow-up studies have the potential to identify prognostically useful phenomena necessary for such judgments. Many, though, have simply contrasted the outcomes of patients with schizoaffective disorder to those of comparison groups. Others that have assessed the predictive value of individual phenomena have done so using only global outcome measures. This is problematic because the amount of morbidity experienced by patients with affective disorder can vary widely, even if no patient assumes a course of chronic psychosis. Likewise, some patients with schizophrenia experience much less morbidity than others with the same diagnosis even if no one achieves full remission. Yet, it is the predictions of these more archetypal outcomes that are most likely to influence treatment decisions. The following will therefore focus on those studies that have tested the value of individual, clinical, and historical features in the prediction of either chronic psychosis or complete remission during a follow-up of one year or more. Each of the studies in Tables 7.1 and 7.2 applied operational criteria to symptoms and history assessed at baseline, and described a sample comprising either patients with schizoaffective disorder, or a diagnostically mixed sample of patients with psychotic features. For those cohorts with baseline depressive syndromes, the quality of symptoms in cross-section rarely emerged as a predictor of full recovery (Table 7.1). Rather, most associated a failure to recover with baseline episode duration and/or a history of psychotic features that occurred when depressive symptoms were minimal. Seven studies attempted to predict the specific outcomes of chronic psychosis or a diagnostic change to schizophrenia (Table 7.3). Again, historical features such as pre-existing chronicity or the occurrence of schizophrenic symptoms outside of affective syndromes were most often important. Conclusions regarding the prediction of outcome in psychotic patients with manic syndromes (Tables 7.2 and 7.4) are more tenuous because only a few have used full recovery, diagnostic shift, or chronic psychosis as outcomes. In comparison to findings from depressed samples, individual symptoms seemed more likely to emerge as predictors though no two studies identified the same ones. A number of caveats apply to the data in these tables. First, the studies described in them varied considerably in the array of potential predictors tested and some only considered mood-congruence. The latter sort of study gives unwarranted importance to mood-congruence relative to other possible predictors. Second, the follow-ups ranged in length from one year to eight years and it is widely recognized that short- and long-term predictors may differ substantially. Third, diagnostic

136

William Coryell Table 7.1. Predictors of recovery in schizoaffective or psychotic depression.

Study

Sample

Length of follow-up

Predictors

Brockington et al., 1980a

Depressive schizophrenic symptoms, n76

6 yrs

Poor work record, gradual onset, schizophrenic symptoms outside of depressive episodes

Brockington et al., 1982

DSM-III major depression with psychotic features, n25

6 yrs

Mood-incongruence

Coryell et al., 1987

RDC SA depression, n40

2 yrs

RDC chronic or mainly schizophrenic subtype

Coryell and Zimmerman, 1988

RDC SA depression, n39

1 yr

RDC chronic or mainly schizophrenic subtype low GASa score

Tsuang and Coryell, 1993

DSM III-R non-manic psychosis, n92

8 yrs

Longer duration schizophrenic symptoms outside of depressive episodes, mood-incongruence

Notes: a GASSADS (Schedule for Affective Disorders and Schizophrenia) Global Assessment Score.

Table 7.2. Predictors of recovery in psychotic patients with mania or mixed polarity.

Study

Sample

Length of follow-up

Predictors

Brockington et al., 1980b

Manic schizophrenia symptoms, n32

6 yrs

Delusions with hallucinations, passivity phenomena, exacerbation of previous symptoms

Rabiner et al., 1986

RDC first episode of psychosis, n64

1 yr

Longer duration, premorbid asociality

Robinson et al., 2004

RDC first episode of schizophrenia or schizoaffective disorder, n119

5 yrs

Longer duration, poorer global cognition after stabilization

137

Phenomenological approaches Table 7.3. Predictors of chronic psychosis in schizoaffective or psychotic depression.

Study

Sample

Length of follow-up

Predictors

Brockington et al., 1980a

Depression with schizophrenic schizophrenia or paranoid symptoms, n76

6 yrs

Family history suggesting schizophrenia, personality suggesting schizophrenia, gradual mode of onset, nuclear syndrome

Brockington et al., 1982

DSM-III major depression with psychotic features, n25

6 yrs

Mood-incongruence

Coryell et al., 1990a RDC schizoaffective depression psychotic MDD, n103

5 yrs

Lowest level of optimal functioning in previous five years, longer duration, poor adolescent friendship patterns, history of moodincongruent symptoms, to the relative exclusion of other symptoms

Tsuang and Coryell, 1993

DSM-III-R major depression with psychotic features or schizoaffective depression, n43

8 yrs

Schizophrenic symptoms without depression

Winokur et al., 1996

RDC non-manic psychosis, n70

6 yrs

Preponderance of schizophrenia-like symptoms

compositions also varied widely and this is likely to have affected findings in a number of ways. Mood-congruence, for instance, may have more importance in a group without schizoaffective cases than in groups consisting largely of individuals with schizoaffective disorder. Fourth, only a few studies tested predictors specifically in patients who had been in manic episodes at baseline. Manic syndromes are much more likely to include psychotic features than are depressive syndromes and the presence of psychotic features in depression appears to be much more predictive of the subsequent persistence of psychosis, particularly in nonacute cases (Coryell et al., 2001). The prediction of full recovery and of chronic psychosis would therefore be expected to differ according to whether the presenting syndrome is one of mania or depression. Indeed, the two sets of authors who used

138

William Coryell Table 7.4. Predictors of chronic psychosis in psychotic patients with mania or mixed polarity.

Study

Sample

Length of follow-up

Coryell et al., 1990

RDC schizoaffective mania mania with psychotic features, n70

5 yrs

Fennig et al., 1996

DSM-III-R depression or mania with psychotic features, n84

2 yrs

Predictors

Lowest level of optimal functioning in previous 5 yrs, lower GASa score, longer duration, loosening of associations, formal thought disorder in the absence of prominent manic symptoms Mood-incongruence

Notes: GASSADS Global Assessment Score.

a

identical methods in outcome prediction for both psychotic depression and for psychotic mania found very little overlap in the predictors of full recovery or of chronic psychosis (Brockington, 1980a; 1980b; Coryell et al., 1990a; 1990b). Mood-congruence

Among phenomenological variables assessed in cross-section, mood-congruence (MC) has received the most attention as a potential outcome predictor in psychotic affective disorders. Because the DSM-IV criteria for Major Depressive Disorder (MDD) or mania with mood-incongruent (MI) psychotic features overlaps substantially with the RDC definition of schizoaffective disorder, a review should give particular attention to this distinction. Kendler (1991) did this over a decade ago. In thirteen published comparisons of patients with mood-congruent features to patients with mood-incongruent psychotic features, outcomes were uniformly at least somewhat better for the moodcongruent groups though differences reached statistical significance in only three of the studies. Subsequent reports have sustained this consistency but with more robust group differences (Tohen et al., 1992; Tsuang and Coryell, 1993; Fennig et al., 1996). Five of the comparisons described by Kendler (Rosenthal et al., 1980; Brockington et al., 1983; van Praag and Nijo, 1984; Winokur, 1990; Coryell et al., 1990b; Kendler, 1991), and two of the subsequently published studies (Fennig et al., 1996; Tohen et al., 1992) were of manic patients.

139

Phenomenological approaches Table 7.5. Severity of melancholic symptoms among patients with psychotic or schizoaffective depression.

Symptoms n Distinct quality of mood Lack of reactivity a.m. dysphoria Anhedonia Guilt Negative evaluation of self Terminal insomnia Psychomotor retardation Psychomotor agitation Anorexia Weight loss

Sample 1

Sample 2

76

76 MDSA * MDSA * MDSA ** MDSA **

MCMI * MCMI ** MDSA **

MCMI *

MDSA ** MDSA **

Notes: * p0.05 ** p0.01

Affective symptom quality

In contrast to the focus on the nature of psychotic features as prognostically meaningful, little attention has been given to the quality of affective symptoms. More is probably warranted, though. Two reports used the intensity of individual melancholic symptoms to compare groups of psychotically depressed patients. One separated groups by mood-congruence (Coryell and Tsuang, 1985) and the other compared patients with MDD and psychotic features to patients with RDC schizoaffective disorder (Coryell and Zimmerman, 1986). In neither study did any melancholic symptom have a greater mean severity in a mood-incongruent or schizoaffective group than it did in the comparison group with mood-congruent or psychotic major depressive disorder. Rather, patients with mood-congruent psychotic features and, in particular, those with RDC schizoaffective disorder, had melancholic symptoms that were significantly less prominent than they were in the comparison groups (Table 7.5). Historical and cross-sectional features: summary

A distillation of the preceding review to determine which clinical features have most frequently predicted either a failure to recover, a chronic psychosis, or a diagnostic revision to schizophrenia shows that the simple historical variable of chronicity is the one most often associated with these outcomes (Rabiner et al,

140

William Coryell

1986; Coryell et al., 1987; Coryell and Zimmerman, 1988b; Coryell et al., 1990a, 1990b; Tsuang and Coryell, 1993; Robinson et al., 2004). Nearly as many studies have found significant associations between these outcomes and a history of schizophrenia-like psychotic features, or psychotic features in general, that preceded or followed affective syndromes (Brockington et al., 1980a; Coryell et al., 1987; Coryell and Zimmerman, 1988a; Coryell et al., 1990a; Tsuang and Coryell, 1993). A measure of best psychosocial functioning in recent years may, in many cases, have been one of insidious onset. Taken together, these two features also have often predicted non-recovery in chronic psychosis (Brockington et al., 1980a, 1980b, Coryell et al., 1990a, 1990b; Winokur et al., 1996). HPA-axis hyperactivity as a predictor

Though there is some consistency in the follow-up literature to guide clinicians in their estimations of the course a patient will follow, an uncomfortable level of uncertainty often exists in the evaluation of schizoaffective patients, particularly near the onset of their illness. Might certain biological measures be prognostically useful? A number of physiological abnormalities have been identified in patients with affective disorders in comparison to control groups. Other abnormalities have seemed to characterize patients with schizophrenia relative to well controls. Some have been identified with more consistency than others have been but several factors have limited their potential value for these clinical problems such as the one posed here. Many of the required measurements are simply impractical because they are too invasive, expensive, or complex. Others have failed to show a reasonable level of diagnostic specificity when psychiatrically ill comparison groups were used instead of well controls. Hypothalamic–pituitary–adrenal (HPA) axis hyperactivity is probably the most widely researched of the abnormalities associated with depressive disorders. Among the numerous methods used to detect this, the dexamethasone suppression test (DST) probably offers the best balance between validity and clinical practicality. The test was at one point widely thought to be a promising tool for distinguishing melancholia from other psychiatric disorders (Carroll et al., 1981), but the subsequent failure of many studies to demonstrate diagnostic specificity (Hirschfeld et al., 1983) led to a general disillusionment with its clinical potential. Several lines of investigation, though, indicate that an assessment of HPA-axis function has value in the prediction of recurrence, completed suicide, and chronic psychosis. The last of these is relevant to the current topic. An extensive review of DST non-suppression rates across diagnostic categories was described in six studies that included cohorts with psychotic depression and in nine that described patients with schizoaffective depression (Sharma et al., 1988). A pooling of data yields 109 subjects with psychotic MDD and 87 with

141

Phenomenological approaches Table 7.6. Baseline DST results and long-term outcome in diagnostically mixed samples of inpatients with psychosis.

DST result

Coryell et al., 1989 MDD with psychotic features/SA-dep., n No. (%) psychotic at 1 yeara Coryell et al., 1992 MDD with psychotic features/SA-dep, n No. (%) with chronic psychosis at 8 yearsb Targum et al., 1983 DSM-III schizophreniform, n No. (%) not recovered at 1 yearc

Suppressor

Non-suppressor

23 16 (69.6)

21 4 (19.0)

21 18 (85.7)

26 13 (50.0)

8 6 (75.0)

9 1 (11.1)

Notes: p0.0001 b p0.0001 c p0.04 a

schizoaffective depression. Non-suppression rates in these groups were 67.0% and 41.4%, respectively ( 2 12.9, df1, p0.0001). Moreover, psychotic depression is the subtype of depressive disorder most consistently associated with HPA-axis hyperactivity (Nelson and Davis, 1997). Together, these findings suggest that the presence or absence of HPA-axis hyperactivity should have predictive value among patients with schizoaffective depression or with major depressive disorder and mood-incongruent psychotic features, because such patients may follow a course consistent with schizophrenia or one indicative of an affective disorder. Two studies (Table 7.6) have been designed to test this possibility and both found that, among patients with psychotic depression or schizoaffective disorder, the absence of HPA-axis hyperactivity predicted chronic psychosis one or eight years later (Tsuang and Coryell, 1993; Coryell and Zimmerman, 1989). Another, earlier study followed a small group of patients who had been evaluated during an initial episode of DSM-III schizophreniform disorder, and it described a similar relationship between baseline HPA-axis activity and recovery from psychosis at one year (Targum, 1983). Investigators in one of the above studies undertook a careful re-evaluation of subjects after one year of follow-up (Coryell et al., 1988a). The raters were blind to baseline DST results but used all baseline and follow-up information to re-assess diagnosis. Shifts from schizophrenia to schizoaffective disorder were noted in five patients and a shift from psychotic major depressive disorder to schizoaffective

142

William Coryell

disorder in another seven. Four of the former (80.0%), but only two (28.6%) of the latter had post-dexamethasone cortisol values greater than 5 mg/dl at their baseline assessment. The relationship between diagnosis and DST result was therefore clearer at follow-up than at baseline. At baseline assessment, DST results were abnormal in 6 of 18 (33.3%) patients with a schizophrenia diagnosis and in 14 of 26 (53.8%) patients with a diagnosis of psychotic MDD ( 2 2.2, df1, p0.1). After diagnoses were revised one year later, two of the thirteen (15.4%) now considered schizophrenic and 12 of the 19 (63.2%) now thought to have MDD had had abnormal DST results at baseline ( 2 7.2, df1, p0.01). Conclusions Though the fundamental relationship between schizoaffective disorder and other categories of psychotic illness remains a matter for debate, clinicians must select from an array of treatments that have been tested almost exclusively in samples of typical MDD, bipolar affective disorder or schizophrenia. The expectation that the eventual course will suggest an affective disorder rather than schizophrenia will therefore strongly influence treatment selection. A review of the relevant follow-up study shows that pre-existing chronicity, a history of psychotic features that have occurred outside of affective episodes, poor premorbid functioning, an insidious onset, and a predominance of mood-incongruent psychotic features all increase the risk that a patient’s eventual outcome will be more typical of schizophrenia. This review cannot determine the sensitivity, specificity, or interactions of these features to such an outcome because the studies have varied widely in the diagnostic mix of samples followed and in the array and definitions of the potential predictors that were tested. Likewise, this review cannot determine whether a particular combination of those predictors would provide better sensitivity and specificity than any one predictor alone. The possibility that a clinically accessible measure of HPA-axis activity would comprise an additional tool to determine prognosis clearly deserves additional study.

R E F E R E N C ES Andreasen, N. C., Rice, J., Endicott, J. et al. (1987). Familial rates of affective disorder. A report from the National Institute of Mental Health Collaborative Study. Archives of General Psychiatry, 44 (5), 461–9. Baron, M., Gruen, R., Asnis, L. and Kane, J. (1982). Schizoaffective illness, schizophrenia and affective disorders: morbidity risk and genetic transmission. Acta Psychiatrica Scandinavica, 65 (4), 253–62. Brockington, I. F., Helzer, J. E., Hillier, V. F. and Francis, A. F. (1982). Definitions of depression: concordance and prediction of outcome. American Journal of Psychiatry, 139 (8), 1022–7.

143

Phenomenological approaches Brockington, I. F., Hillier, V. F., Francis, A. F., Helzer, J. E. and Wainwright, S. (1983). Definitions of mania: concordance and prediction of outcome. American Journal of Psychiatry, 140 (4), 435–9. Brockington, I. F., Kendell, R. E. and Wainwright, S. (1980a). Depressed patients with schizophrenic or paranoid symptoms. Psychological Medicine, 10 (4), 665–75. Brockington, I. F., Leff, J. P. (1979). Schizo-affective psychosis: definitions and incidence. Psychological Medicine, 9 (1), 91–9. Brockington, I. F., Wainwright, S., Kendell, R. E. (1980b). Manic patients with schizophrenic or paranoid symptoms. Psychological Medicine, 10 (1), 73–83. Carroll, B. J., Feinberg, M., Greden, J. F. et al. (1981). A specific laboratory test for the diagnosis of melancholia. Standardization, validation, and clinical utility. Archives of General Psychiatry, 38 (1), 15–22. Coryell, W., Endicott, J., Keller, M. and Andreasen, N. C. (1985). Phenomenology and family history in DSM-III psychotic depression. Journal of Affective Disorders, 9 (1), 13–18. Coryell, W., Grove, W., vanEerdewegh, M., Keller, M. and Endicott, J. (1987). Outcome in RDC schizo-affective depression: the importance of diagnostic subtyping. Journal of Affective Disorders, 12 (1), 47–56. Coryell, W., Keller, M., Lavori, P. and Endicott, J. (1990a). Affective syndromes, psychotic features, and prognosis. I. Depression. Archives of General Psychiatry, 47 (7), 651–7. Coryell, W., Keller, M., Lavori, P. and Endicott, J. (1990b). Affective syndromes, psychotic features, and prognosis. II. Mania. Archives of General Psychiatry, 47 (7), 658–62. Coryell, W., Leon, A. C., Turvey, C. et al. (2001). The significance of psychotic features in manic episodes: a report from the NIMH collaborative study. Journal of Affective Disorders, 67 (1–3), 79–88. Coryell, W. and Tsuang, D. (1992). Hypothalamic-pituitary-adrenal axis hyperactivity and psychosis: recovery during an 8-year follow-up. American Journal of Psychiatry, 149 (8), 1033–9. Coryell, W. and Tsuang, M. T. (1985). Major depression with mood-congruent or moodincongruent psychotic features: outcome after 40 years. American Journal of Psychiatry, 142 (4), 479–82. Coryell, W. and Zimmerman, M. (1986). Demographic, historical, and symptomatic features of the nonmanic psychoses. Journal of Nervous and Mental Disease, 174 (10), 585–92. Coryell, W. and Zimmerman, M. (1988b). The heritability of schizophrenia and schizoaffective disorder. A family study. Archives of General Psychiatry, 45 (4), 323–7. Coryell, W. and Zimmerman, M. (1989). HPA axis hyperactivity and recovery from functional psychoses [see comments]. American Journal of Psychiatry, 146 (4), 473–7. Coryell, W., Zimmerman, M., Winokur, G. and Cadoret, R. (1988a). Baseline neuroendocrine function and diagnostic stability among patients with a nonmanic psychosis. European Archives of Psychiatry and Clinical Neuroscience, 237 (4), 197–9. Crow, T. J. (1987). Continuum of psychosis and the gene. British Journal of Psychiatry, 151, 125–6. Crow, T. J. (1990a). The continuum of psychosis and its genetic origins. The sixty-fifth Maudsley lecture. British Journal of Psychiatry, 156, 788–97. Crow, T. J. (1990b). Nature of the genetic contribution to psychotic illness – a continuum viewpoint. Acta Psychiatrica Scandinavica, 81 (5), 401–8. Fennig, S., Bromet, E. J., Karant, M. T., Ram, R. and Jandorf, L. (1996). Mood-congruent versus

144

William Coryell mood-incongruent psychotic symptoms in first-admission patients with affective disorder. Journal of Affective Disorders, 37 (1), 23–9. Fennig, S., Kovasznay, B., Rich, C. et al. (1994). Six-month stability of psychiatric diagnoses in first-admission patients with psychosis. American Journal of Psychiatry, 151 (8), 1200–8. Hirschfeld, R. M., Koslow, S. H., Kupfer, D. J. (1983). The clinical utility of the dexamethasone suppression test in psychiatry. Summary of a National Institute of Mental Health workshop. JAMA, 250 (16), 2172–4. Kendler, K. S. (1991). Mood-incongruent psychotic affective illness. A historical and empirical review. Archives of General Psychiatry, 48 (4), 362–9. Kendler, K. S., McGuire, M., Gruenberg, A. M. and Walsh, D. (1995). Examining the validity of DSM-III-R schizoaffective disorder and its putative subtypes in the Roscommon Family Study. American Journal of Psychiatry, 152 (5), 755–64. Nelson, J. C. and Davis, J. M. (1997). DST studies in psychotic depression: A meta-analysis. American Journal of Psychiatry, 154 (11), 1497–503. Rabiner, C. J., Wegner, J. T. and Kane, J. M. (1986). Outcome study of first-episode psychosis. I: Relapse rates after 1 year. American Journal of Psychiatry, 143 (9), 1155–8. Robinson, D. G., Woerner, M. G., McMeniman, M., Mendelowitz, A. and Bilder, R. M. (2004). Symptomatic and functional recovery from a first episode of schizophrenia or schizoaffective disorder. American Journal of Psychiatry, 161 (3), 473–9. Rosenthal, N. E., Rosenthal, L. N., Stallone, F., Dunner, D. L., Fieve, R. R. (1980). Toward the validation of RDC schizoaffective disorder. Archives of General Psychiatry, 37 (7), 804–10. Sharma, R. P., Pandey, G. N., Janicak, P. G. et al. (1988). The effect of diagnosis and age on the DST: a meta-analytic approach. Biological Psychiatry, 24 (5), 555–68. Spitzer, R. L., Endicott, J. and Robins, E. (1978). Research Diagnostic Criteria (3rd Edn.). Biometrics Research. New York: State Department of Mental Hygiene. Targum, S. D. (1983). Neuroendocrine dysfunction in schizophreniform disorder: correlation with six-month clinical outcome. American Journal of Psychiatry, 140 (3), 309–13. Tohen, M., Tsuang, M. T and Goodwin, D. C. (1992). Prediction of outcome in mania by moodcongruent or mood-incongruent psychotic features. American Journal of Psychiatry, 149 (11), 1580–4. Tsuang, D. and Coryell, W. (1993). An 8-year follow-up of patients with DSM-III-R psychotic depression, schizoaffective disorder, and schizophrenia. American Journal of Psychiatry, 150 (8), 1182–8. Tsuang, M. T., Dempsey, G. M., Dvoredsky, A. and Struss, A. (1977). A family history study of schizo-affective disorder. Biological Psychiatry, 12 (3), 331–8. van Praag H. M. and Nijo, L. (1984). About the course of schizoaffective psychoses. Comprehensive Psychiatry, 25 (1), 9–22. Winokur, G., Black, D. W. and Nasrallah, A. (1990). The schizoaffective continuum: Nonpsychotic, mood congruent and mood incongruent. In Affective and Schizoaffective Disorders, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 23–32. Winokur, G., Monahan, P., Coryell, W. and Zimmerman, M. (1996). Schizophrenia and affective disorder – distinct entities or continuum? An analysis based on a prospective 6-year follow-up. Comprehensive Psychiatry, 37 (2), 77–87.

8

Clinical course of schizoaffective disorders Maria Reinares1, Eduard Vieta1, Antoni Benabarre1and Andreas Marneros2 1 2

Bipolar Disorders Program, Institute of Neuroscience, Hospital Clinic, University of Barcelona Department of Psychiatry and Psychotherapy, Martin Luther University Halle-Wittenberg

Introduction The term “schizoaffective psychosis” was first introduced by Kasanin (1933) when he described a group of patients with good premorbid functioning who developed acute psychoses with a mixture of psychotic and affective symptoms, but fully recovered after a few months. While Kasanin is credited with introducing the term, it is defined differently now. Schizoaffective disorder is a complex illness whose definition has changed significantly over time. Despite the continued attempts to better define and classify schizoaffective disorder, much controversy and conflicting results remain. Unfortunately, schizoaffective disorders have been poorly investigated. Kahlbaum (1863) is usually considered the first psychiatrist in modern times to describe schizoaffective disorders as a separate group (Angst and Marneros, 2001). As Tsuang and Simpson (1984) reported, empirical findings are often contradictory and have at times supported the idea that schizoaffective disorder is (a) a variant of schizophrenia; (b) a variant of affective disorder; (c) a different and heterogeneous diagnostic group. Regarding classification systems, in DSM-II schizoaffective disorder was included in the group of schizophrenic disorders, although some studies with controversial results were published (Procci, 1976; Harrow, 1984). While in DSM-III schizoaffective disorders had only the state of a remnant category, in DSM-III-R schizoaffective disorders were extended to a “true” entity with specific diagnostic criteria (Jäger et al., 2004). The DSM-III-R suggests that its diagnosis should be considered in those pathologies that do not fulfill either schizophrenia or mood disorder criteria, but at other times manifest themselves as schizophrenia and mood disorder, and often present with psychotic symptoms in the absence of mood alteration. More specifically, the presence of criteria indicating delusions and The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

145

146

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros

hallucinations for at least two weeks without significant symptoms of mood alteration is necessary. The DSM-IV criteria are similar: schizoaffective disorder is listed in the category of “schizophrenia and other psychotic disorders.” The ICD-10 also has a separate diagnostic category for schizoaffective disorders; this category is part of the group “schizophrenia, schizotypical and delusional disorders.” In contrast to ICD-10, the criteria for schizoaffective disorders in DSM-IV require the presence of delusions or hallucinations for at least two weeks in the absence of prominent mood symptoms. According to Marneros (2003a), the most important criticisms of the definitions of DSM-IV and ICD-10 are that there is no argument for the chronological distinction regarding the coexistence of schizophrenic and affective symptomatology, and also the fact that both diagnostic systems do not involve the longitudinal aspect in their definitions. Based on longitudinal findings, the author defines two types of schizoaffective disorders: concurrent, characterized by the coincidence of schizophrenic and affective symptomatology; and sequential, characterized by the change of different phases. Regarding criteria proposed by DSM-IV for schizoaffective disorder, Maj et al. (2000) found the inter-rater reliability of such criteria unsatisfactory. Longitudinal follow-up studies are essential for resolving diagnostic instability problems in the current diagnostic system (González-Pinto et al., 1998). Clinical course of schizoaffective disorder Schizoaffective disorders compared with schizophrenia and affective disorders

There have been different attempts to classify schizoaffective disorder into two groups, one group with predominant affective symptoms and another group with schizophrenic symptoms dominating. Tsuang and Fleming, in 1996, suggested another category: the undifferentiated. The validity of these attempts is still controversial. An important clinical question is the difference in prognosis between affective or non-affective symptoms in psychosis. Winokur et al. (1996) provided indirect evidence for differential outcomes based on subtypes. Subjects with schizophrenia, affective disorder, or schizoaffective disorder were evaluated at intake, one year later, and after six years. Several outcome measures were used. Patients were divided into groups that had consistent affective diagnoses or consistent schizophrenic diagnoses at each of the assessment times, and there was a group with inconsistent diagnosis. The authors found that recovery from psychosis was more common in the group with consistent diagnosis of affective disorder than it was in the group with consistent diagnosis of schizophrenia. The inconsistent group was intermediate, but was closer to the consistent affective group. Samson et al. (1988) reviewed ten studies that had been carried out between 1963 and 1987 on the clinical course in schizophrenic and schizoaffective patients.

147

Clinical course of schizoaffective disorders

Patients with schizophrenia showed poorer outcomes in global, marital, social and occupational dimensions. Differences were also noted regarding hospital course and symptoms. In contrast, their review of eleven outcome studies comparing schizoaffective disorder with affective disorder showed that affective patients had equal or better outcomes than schizoaffective patients. Apart from the methodological differences that existed between the separate studies, and specifically the inclusion criteria that were used, schizoaffective disorder was more frequently associated with an intermediate clinical course between the schizophrenic and schizoaffective patients. Marneros et al. (1990) compared the psychopathological and social status over the long-term course (mean: 25.1 years) of 101 schizoaffective disorders with 148 schizophrenic and 106 affective disorders according to modified DSM-III criteria. They found that schizoaffective disorders occupy an intermediate position. However, the outcome of schizoaffective disorders was far more similar to that of affective than to that of schizophrenic disorders with regard to both psychopathological as well as psychosocial outcome variables. In the Grossman et al. (1984, 1991) studies, schizoaffective patients had a somewhat better overall post-hospital functioning than patients with schizophrenia, somewhat poorer functioning than bipolar manic patients, and significantly poorer functioning than patients with unipolar depression. When Marneros et al. (1998) studied the clinical course of the three disorders, the outcomes were measured through psychotic symptoms, reduction of energetic potential, qualitative and quantitative affective disturbances and other behavioral disturbances. As in previously mentioned studies, the schizophrenia group showed a higher proportion of patients with a worse clinical course (93.2%) compared with schizoaffective disorders (49.5%) and the affective group (35.8%). Regarding persisting alterations such as residual aspects, they usually seem to be milder in patients with schizoaffective disorder but more severe than in patients with mood disorders (Marneros, 2003b). Harrow et al. (2000) studied 210 patients with schizoaffective disorders, schizophrenia, bipolar manic disorders and depression who were assessed at hospitalization and then followed up four times over ten years (at 2, 4.5, 7.5 and 10 years after the index hospitalization). Schizoaffective outcome was better than schizophrenic outcome and poorer than outcome for psychotic affective disorders. Patients with mood-incongruent psychotic symptoms during index hospitalization showed significantly poorer subsequent outcome. With the objective of determining the epidemiological, clinical and prognostic differences among bipolar, schizophrenic and schizoaffective disorder, Benabarre et al. (2001a) carried out a study with a sample of 138 patients who met Research Diagnostic Criteria (RDC) for type I bipolar disorder (n67), schizoaffective disorder bipolar type (n34) and schizophrenia (n37). Results showed that the age

148

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros

of onset was significantly higher for bipolar patients. Schizoaffective patients showed significantly the greatest number of episodes since onset, followed by bipolar patients and schizophrenics. The characteristics of the recurrences regarding mood were more similar between bipolar and schizoaffective patients. The presence of psychotic symptoms in the first episode was more frequent in schizoaffective and schizophrenic patients. The existence of psychotic symptoms at any point in the course of the illness, and the congruence of these symptoms with mood, differentiated between bipolar disorder and other psychoses, but did not show significant differences between schizoaffectives and schizophrenics. Delusions were more frequently seen in schizoaffective and bipolar patients, and hallucinations were more present in schizophrenics, followed by schizoaffective and bipolar patients. Finally, the analysis of prognostic variables revealed that schizoaffectives occupied an intermediate position between bipolar and schizophrenic patients with respect to occupational functioning and autonomy. Taken together, the results obtained supported the hypothesis that the bipolar-type schizoaffective disorders could be an intermediate phenotype between bipolar I disorder and schizophrenia from the epidemiological, clinical and prognostic viewpoint. Schizoaffective patients seem to present a cross-sectional clinical pattern very similar to those with schizophrenia but their longitudinal evolution resembles that of bipolar I patients (Benabarre et al., 2001b). Although most studies have shown the intermediate course of schizoaffective disorder, there are some categories in which it resembles schizophrenia or affective disorder more closely. Tsuang and Coryell (1993) reported that schizoaffective disorders have a prognosis resembling that of schizophrenia. Williams and McGlashan (1987) had found similar results. Both Kendler et al. (1995) and Atre-Vaidya and Taylor (1997) observed similarities between psychotic symptoms in schizoaffective disorders and schizophrenia. Other research suggests that schizoaffective disorder represents a variant of schizophrenia in terms of clinical symptoms, family history, treatment regimen and cognitive impairment (Evans, 1999). However, Samson et al. (1988) observed that the schizoaffective outcome was almost equivalent to the affective in most dimensions. In a Marneros study published in 1992, 70% of schizoaffective patients rated as good or excellent in social adjustment; this data is not much different to that of the 84% of affective patients who received the same rating. Both groups differed significantly from the schizophrenic group, only 44% of whom showed good or excellent outcome. Nor did schizoaffective patients differ from the affective ones in aspects such as body language, affectiveness, conversation skills and capacity for cooperation; nevertheless both groups were rated as significantly less impaired than the schizophrenic group. Regarding the course, the only difference between schizoaffective and affective subjects found by Angst and

149

Clinical course of schizoaffective disorders

Preisig (1995) was a greater frequency in episodes requiring hospitalization among schizoaffectives. Pope et al. (1980) and Möller et al. (1989) found a favorable outcome for both affective and schizoaffective disorders in relation to schizophrenia. Similarly, when a sample of schizophrenic, schizoaffective and affective patients was assessed at the time of first hospitalization and about 12–15 years later (Möller et al., 2000, 2002), results showed that the course of schizophrenia showed greater deterioration than that of affective or schizoaffective psychosis. Negative symptoms had the highest impact on global functioning and the severity of illness only in schizophrenia. Jager et al. (2003) also found more pronounced negative symptoms and a lower global functioning in schizophrenic patients compared to persistent delusional disorders, acute and transient psychotic disorders and schizoaffective disorders. The specificity of negative symptoms to distinguish prognostic differences between the three groups begins to come into question, although even the pure bipolar patients could show this symptomatology (Maziade et al., 1995), together with diverse cognitive dysfunctions (Martínez-Arán et al., 2000; Salazar-Fraile et al., 2004). In a recent study, Jäger et al. (2004) included 241 first-admitted inpatients who fulfilled ICD-10 criteria for schizophrenia or affective disorders. Patients were examined at the time of first hospitalization and then followed up after 15 years. With respect to the first hospitalization, schizoaffective disorders were distinguishable from both schizophrenia and affective disorders in regard to the clinical picture. Schizodepressive patients showed a higher mean value in the total score of the depressive syndrome than patients with schizophrenia. Compared with depressive patients they showed a higher mean value in the total score of the paranoidhallucinatory syndrome and a lower one in the total score of the depressive syndrome. Schizomanic patients showed a significantly lower mean value in total score of the negative syndrome and a higher one in total score of the manic syndrome than patients with schizophrenia. Compared with manic patients they showed a higher mean value in total score of the paranoid-hallucinatory syndrome. However, with respect to the long-term outcome, schizoaffective disorders had a prognosis similar to that of affective disorders, but quite different from that of schizophrenia. Patients with schizoaffective disorders differ from those with schizophrenia because they have a more favorable outcome. When schizoaffective disorders were compared with affective disorders significant differences were only found in the total scores of the manic and depressive syndromes. No significant differences with regard to the long-term outcome – global assessment, social outcome and clinical picture – were found when schizoaffective disorders were compared with the sub-group of affective disorders with psychotic symptoms. The same was true for the comparison between patients with schizomanic or mixed subtype and those with schizodepressive subtype.

150

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros

Unipolar and bipolar schizoaffective disorders

Schizoaffective disorders are divided into unipolar and bipolar types. Bipolar schizoaffective disorders have similar premorbid and sociodemographic features and patterns to the pure affective bipolar disorders in terms of course, treatment, and prophylaxis, but they differ from the unipolar schizoaffective disorders in the same way as bipolar affective disorders differ from unipolar affective disorders. Angst and Preisig (1995) observed that when compared with unipolar disorders (unipolar depression and schizodepressive disorder), bipolar (bipolar and schizobipolar) disorders showed more periodicity, characterized by a greater number of total episodes, more episodes per year with shorter episodes and cycles. There are also mixed bipolar schizoaffective episodes that are almost as frequent as the affective mixed ones; but mixed schizoaffective patients became ill earlier than mixed affective patients, they have longer episodes, and seem to be impaired more severely than patients with mixed affective episodes. They also have a more unfavorable outcome (Marneros, 2003a). In a recent study, Marneros et al. (2004) found that 25% of bipolar affective and 32% of bipolar schizoaffective patients had at least one schizomixed episode during the illness course; they also suggested that schizoaffective mixed states are the most severe type within the bipolar spectrum. Patients with bipolar schizoaffective disorders have significantly more episodes, a higher annual frequency of episodes, more cycles, and a higher annual frequency of cycles than unipolar schizoaffective disorders. Regarding long-term outcome, it is dependent on the number of episodes – the more frequent the episodes, the more unfavorable the long-term outcome (Marneros and Angst, 2000). Prognostic indicators Regarding prognostic indicators, a classic study performed by Vaillant (1964) found that a good premorbid level, the existence of triggering factors, an acute onset, confusion symptoms, affective symptoms, and a familial history of affective disorder could predict remission in approximately 80% of cases of “remitting schizophrenia.” Later, Levitt and Tsuang (1988) reported that the existence of affective symptoms and family history of affective disorder were connected with a better outcome. Similarly, variables associated with schizophrenic symptoms or a family history of schizophrenia were related with a worse outcome. Findings of Himmelhoch et al. (1981) showed that “interepisodic thought disorders” in-between acute affective episodes were associated with an unfavorable outcome. Similar results were reported by Brockington et al. (1980), who found that the presence of schizophrenic symptoms at some time without depression was an important predictor for a poor outcome. However, the presence of mood symptoms in addition to a psychotic

151

Clinical course of schizoaffective disorders

illness has been associated with a more favorable outcome relative to a primary diagnosis of a psychotic disorder (Retterstol, 1991). McGlashan and Williams (1990) pointed out that both better developed premorbid instrumental skills and fewer typically schizophrenic symptoms were the two dimensions that regularly predicted better outcome among schizoaffective psychosis. Marneros et al. (1993) reported that the most important factors influencing the development of persisting psychosocial alterations in schizoaffective disorder proved to be the absence of pure melancholic episodes during course and life events at onset, the presence of first-rank schizophrenic symptoms, schizomanicdepressive mixed episodes, a higher number of episodes, and a higher annual frequency of episodes. Other factors that may be prognostic of poor outcomes include poor interepisode recoveries, persistent psychotic symptoms in the absence of affective features, poor premorbid social adjustment, chronicity, and a higher number of schizophrenia-like symptoms (Tsuang et al., 2000). Some authors have suggested that psychotic symptoms incongruent with mood could be associated with worse clinical course (Harrow and Grossman, 1984; Grossman et al., 1991; Tohen et al., 1992; Harrow et al., 2000). The findings from Bottlender et al. (2002) showed that a duration of untreated psychosis longer than 12 months was independently and significantly associated with a poorer outcome from first inpatient treatment. Premorbid functioning might have an additional influence on outcome but this influence seemed to be dependent on the diagnostic category – it had a greater effect on outcome in schizoaffective patients than in schizophrenic patients. Finally, in the last decade a number of therapies have been developed for this condition, which may help to improve the short- and long-term outcome of the disorder (Tran et al., 1999; Janicak et al., 2001; Keck et al., 2001; Vieta et al., 2001; SwainstonHarrison and Perry, 2004). Conclusions Considering the controversial classification of schizoaffective disorder, determining the long-term course and prognosis is difficult. Most studies show that as a group, patients with schizoaffective disorder seem to have an intermediate prognosis between schizophrenic patients and patients with mood disorders. However, schizoaffective disorders are not homogeneous. It has been presumed that an increasing presence of schizophrenic symptoms predicted worse prognosis. Nevertheless, because schizoaffective disorder is a complex illness whose definition has changed significantly over time and over the different diagnostic systems, the underlying diagnostic criteria have to be taken into account when analyzing the differences between studies.

152

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros

R E F E R E N C ES Angst, J. and Preisig, M. (1995). Course of a clinical cohort of unipolar, bipolar and schizoaffective patients. Results of a prospective study from 1959 to 1985. Schweizer Archiv für Neurologie und Psychiatrie, 146, 5–16. Angst, J. and Marneros, A. (2001). Bipolarity from ancient to modern times: conception, birth and rebirth. Journal of Affective Disorders, 67, 3–19. Atre-Vaidya, N. and Taylor, A. M. (1997). Differences in the prevalence of psychosensory features among schizophrenic, schizoaffective, and manic patients. Comprehensive Psychiatry, 38 (2), 88–92. Benabarre, A., Vieta, E., Colom, F. et al. (2001a). Bipolar disorder, schizoaffective disorder and schizophrenia: epidemiologic, clinical and prognostic differences. European Psychiatry, 16, 1–6. Benabarre, A., Vieta, E., Reinares, M., Torrent, C. and Comes, M. (2001b). El curso clínico del trastorno esquizoafectivo. In Trastorno esquizoafectivo, ed. A. González-Pinto. Madrid: Aula Médica, p. 93–105. Bottlender, R., Sato, T., Jäger, M. et al. (2002). The impact of duration of untreated psychosis and premorbid functioning on outcome of first inpatient treatment in schizophrenic and schizoaffective patients. European Archives of Psychiatry and Clinical Neuroscience, 252, 226–31. Brockington, I. F., Kendell, R. E. and Wainwright, S. (1980). Depressed patients with schizophrenic or paranoid symptoms. Psychological Medicine, 10, 665–75. Evans, J. D., Heaton, R. K., Paulsen, J. S. et al. (1999). Schizoaffective disorder: a form of schizophrenia or affective disorder?. The Journal of Clinical Psychiatry, 60, 974–82. González-Pinto, A., Gutiérrez, M., Mosquera, F. et al. (1998). First episode in bipolar disorder: misdiagnosis and psychotic symptoms. Journal of Affective Disorders, 50, 41–4. Grossman, L. S., Harrow, M., Fudala, J. and Meltzer, H. Y. (1984). The longitudinal course of schizoaffective disorders: a prospective follow-up study. The Journal of Nervous and Mental Disease, 172, 140–9. Grossman, L. S., Harrow, M., Goldberg, J. F. and Fichtner, C. G. (1991). Outcome of schizoaffective disorder at two long-term follow-ups: comparisons with outcome of schizophrenia and affective disorders. American Journal of Psychiatry, 148, 1359–65. Harrow, M. and Grossman, L. (1984). Outcome in schizoaffective disorders: a critical review and reevaluation of the literature. Schizophrenia Bulletin, 10, 87–108. Harrow, M., Grossman, L. S., Herbener, E. S. and Davies, E. W. (2000). Ten-year outcome: patients with schizoaffective disorders, schizophrenia, affective disorders and moodincongruent psychotic symptoms. British Journal of Psychiatry, 177, 421–6. Himmelhoch, J. M., Fuchs, C. Z., May, S. J., Symons, B. J. and Neil, J. F. (1981). When a schizoaffective diagnosis has meaning. The Journal of Nervous and Mental Disease, 169, 277–82. Jäger, M., Bottlender, R., Strauss, A. and Möller, H. J. (2003). On the descriptive validity of ICD10 schizophrenia: empirical analysis in the spectrum of non-affective functional psychoses. Psychopathology, 36,152–9.

153

Clinical course of schizoaffective disorders Jäger, M., Bottlender, R., Strauss, A. and Möller, J. (2004). Fifteen-year follow-up of ICD-10 schizoaffective disorders compared with schizophrenia and affective disorders. Acta Psychiatrica Scandinavica, 109, 30–7. Janicak, P. G., Keck, P. E., Davis, J. M. et al. (2001). A double-blind, randomized, prospective evaluation of the efficacy and safety of risperidone versus haloperidol in the treatment of schizoaffective disorder. Journal of Clinical Psychopharmacology, 21, 360–8. Kahlbaum, K. (1863). Die Gruppirung der psychischen Krankheiten und die Eintheilung der Seelenstörungen. Danzing: Kafemann. Kasanin, J. (1933). The acute schizoaffective psychoses. American Journal of Psychiatry, 13, 97–126. Keck, P. E., Reeves, K. R. and Harrigan, E. P. (2001). Ziprasidone in the short-term treatment of patients with schizoaffective disorder: results from two double-blind, placebo-controlled, multicenter studies. Journal of Clinical Psychopharmacology, 21, 27–35. Kendler, K. S., McGuire, M., Gurneberg, A. M. and Walsh, D. (1995). Examining the validity of DSM-III-R schizoaffective disorder and its putative subtypes in the Roscommon family study. American Journal of Psychiatry, 152, 755–64. Levitt, J. J. and Tsuang, M. T. (1988). The heterogeneity of schizoaffective disorder: implications for treatment. American Journal of Psychiatry, 145, 926–36. Maj, M., Pirozzi, R., Formicola, A. M., Bartoli, L. and Bucci, P. (2000). Reliability and validity of the DSM-IV diagnostic category of schizoaffective disorder: preliminary data. Journal of Affective Disorders, 57, 95–8. Marneros, A. (2003a). The schizoaffective phenomenon: the state of the art. Acta Psychiatrica Scandinavica, 108 (Suppl. 418): 29–33. Marneros, A. (2003b). Schizoaffective disorder: clinical aspects, differential diagnosis, and treatment. Current Psychiatry Reports, 5, 202–5. Marneros, A., Deister, A. and Rohde, A. (1990). Psychopathological and social status of patients with affective, schizophrenic and schizoaffective disorders after long-term course. Acta Psychiatrica Scandinavica, 82, 325–8. Marneros, A., Deister, A. and Rohde, A. (1992). Comparisons of long-term outcome of schizophrenic, affective and schizoaffective disorders. British Journal of Psychiatry, 161 (Suppl. 18), 44–51. Marneros, A., Rohde, A. and Deister, A. (1993). Factors influencing the long-term outcome of schizoaffective disorders. Psychopathology, 26, 215–24. Marneros, A., Rohde, A., Deister, A. (1998). Frequency and phenomenology of persisting alterations in affective, schizoaffective and schizophrenia disorders: a comparison. Psychopathology, 31, 23–8. Marneros, A. and Angst, J. (2000). Bipolar Disorders. 100 Years after Manic Depressive Insanity. Dordrecht, Boston, London: Kluwer Academic Publishers. Marneros, A., Röttig, S., Wenzel, A., Blöink, R. and Brieger, P. (2004). Affective and schizoaffective mixed states. European Archives of Psychiatry and Clinical Neuroscience, 254, 76–81. Martínez-Arán, A., Vieta, E., Colom, F. et al. (2000). Cognitive dysfunctions in bipolar disorder: evidence of neuropsychological disturbances. Psychotherapy and Psychosomatics, 69, 2–18.

154

Maria Reinares, Eduard Vieta, Antoni Benabarre and Andreas Marneros Maziade, M., Roy, M. A., Martínez, M. et al. (1995). Negative, psychotic, and disorganized dimensions in patients with familial schizophrenia or bipolar disorder: continuity and discontinuity between the major psychoses. American Journal of Psychiatry, 152, 1458–63. McGlashan, T. H. and Williams, P. V. (1990). Predicting outcome in schizoaffective psychosis. The Journal of Nervous and Mental Disease, 178, 518–20. Möller, H. J., Hohe-Schramm, M., Cording-Tommel, C. et al. (1989). The classification of functional psychoses and its implications for prognosis. British Journal of Psychiatry, 154, 467–72. Möller, H., Bottlender, R., Wegner, U., Wittmann, J. and Strau, A. (2000). Long-term course of schizophrenic, affective and schizoaffective psychosis: focus on negative symptoms and their impact on global indicators of outcome. Acta Psychiatrica Scandinavica, 102 (Suppl. 407), 54–7. Möller, H., Bottlender, R., Gross, A. et al. (2002). The Kraepelinian dichotomy: preliminary results of a 15-year follow-up study on functional psychoses: focus on negative symptoms. Schizophrenia Research, 56, 87–94. Pope, H. S., Lipinski, J. F., Cohen, B. M. and Axelrod, D. T. (1980). “Schizoaffective disorder”: an invalid diagnosis? A comparison of schizoaffective disorder, schizophrenia and affective disorder. American Journal of Psychiatry, 137, 921–7. Procci, W. R. (1976). Schizoaffective psychosis: fact or fiction? Archives of General Psychiatry, 33, 1167–78. Retterstol, N. (1991). Course and outcome in paranoid disorders. Psychopathology, 24, 277–86. Salazar-Fraile, J., Tabarés-Seisdedos, R., Selva-Vera, G. et al. (2004). Recall and recognition confabulation in psychotic and bipolar disorders: evidence for two different types without unitary mechanisms. Comprehensive Psychiatry, 45, 281–8. Samson, J. A., Simpson, J. C. and Tsuang, M. T. (1988). Outcome of schizoaffective disorders. Schizophrenia Bulletin, 14, 543–54. Swainston-Harrison, T. and Perry, C. M. (2004). Aripiprazole: a review of its use in schizophrenia and schizoaffective disorder. Drugs, 64, 1715–36. Tohen, M., Tsuang, M. T. and Goodwin, D. C. (1992). Prediction of outcome in mania by moodcongruent or mood-incongruent psychotic features. American Journal of Psychiatry, 149 (11), 1580–4 Tran, P. V., Tollefson, G. D., Sanger, T. M. et al. (1999). Olanzapine versus haloperidol in the treatment of schizoaffective disorder. Acute and long-term therapy. British Journal of Psychiatry, 174, 15–22. Tsuang, D. and Coryell, W. (1993). An 8-year follow-up of patients with DSM-III-R psychotic depression, schizoaffective disorder, and schizophrenia. American Journal of Psychiatry, 150, 1182–8. Tsuang, M. T. and Fleming, J. A. (1986). Long-term outcome of schizophrenia and other psychoses. In Search for the Causes of Schizophrenia, p. 88–97. Tsuang, M. T. and Simpson, J. C. (1984). Schizoaffective disorder: concept and reality. Schizophrenia Bulletin, 10, 14–25. Tsuang, M. T., Stone, W. S. and Faraone, S. V. (2000). Schizoaffective and schizotypical disorders. In Gelder, M. G., López-Ibor, J. J. and Andreasen, N. (eds.) New Oxford Textbook of Psychiatry. Vol. 1. New York: Oxford University Press, pp. 636–42.

155

Clinical course of schizoaffective disorders Vaillant, G. (1964). Prospective prediction of schizophrenic remission. Archives of General Psychiatry, 11, 509–18. Vieta, E., Herraiz, M., Fernández, A. et al. (2001). Efficacy and safety of risperidone in the treatment of schizoaffective disorder: initial results from a large, multicenter surveillance study. Journal of Clinical Psychiatry, 62, 623–30. Williams, P. V. and McGlashan, T. H. (1987). Schizoaffective psychosis, I: comparative long-term outcome. Archives of General Psychiatry, 44, 130–7. Winokur, G., Monahan, P., Coryell, W. and Zimmerman, M. (1996). Schizophrenia and affective disorder – distinct entities or a continuum? An analysis based on a prospective 6-year followup. Comprehensive Psychiatry, 37, 77–87.

9

Depressive syndromes in schizophrenia Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer Department of Psychiatry, Heinrich-Heine-University Düsseldorf

Introduction Depressive symptoms are frequently observed in schizophrenia. On average, a depression-like syndrome is occurring during the longitudinal course of schizophrenia in about 25% of the patients (Addington et al., 2002). Nevertheless, depressive symptoms do not belong to the diagnostic criteria for schizophrenia (ICD-10, DSM-IV), and it is relatively recently that stronger attempts have been made towards cross-sectional and longitudinal assessment with the new technologies to better understand the course and nature of this syndrome. The question still remains, however, whether depressive symptomatology in schizophrenia is the same as in depression. This question is at the core of this chapter which aims at analyzing similarities and differences in affective-emotional dysfunction in depressive syndromes. The chapter reviews findings from different empirical perspectives and thereby contributes to the ongoing discussion on the nosological specificity of depression in schizophrenia. Differential diagnosis According to Siris et al. (2001), differential diagnosis is hampered by difficulties associated with distinguishing between depressive symptomatology and certain features of schizophrenia (particularly negative symptoms), side-effects of treatment (e.g., akinesia), and difficulty of recognizing depression in the presence of florid psychosis. The following chapter gives a short overview on those factors to be considered in the differential diagnosis of depressive symptoms in schizophrenia. Medical/organic factors

Several medical and organic factors can present with (secondary) depression in patients with schizophrenia. These include anemias, carcinomas, metabolic The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

156

157

Depressive syndromes in schizophrenia

abnormalities, endocrine disorders, infectious diseases, cardiovascular or neurological disorders, medication (such as beta-blockers or other antihypersensitive agents) or acute and chronic substance abuse (alcohol, cannabis, cocaine, or narcotics) (Siris and Bench, 2003). Neuroleptic-induced dysphoria

“Pharmacogenic” theories have suggested that depression observed in psychosis may be a drug-induced akinetic dysphoria. Dopamine synapses are involved in “reward” pathways. Their blockade by conventional neuroleptic drugs could theoretically lead to anhedonia and, perhaps, depression (Siris, 2000). But several studies point out that depressive symptoms emerge before initiating antipsychotic treatment and tend rather to subside during treatment. Comparison of patients with and without neuroleptic medication did not find a difference in the prevalence of depressive symptoms (Siris, 2000). Voruganti and Awad (2004) investigated the temporal, phenomenological, biochemical, and neuroanatomical relation between dysphoria and extrapyramidal side effects (EPSEs). They came to the conclusion that neuroleptic-induced dopaminergic blockade mediates both dysphoria and EPSEs; but whereas blockade of D2 receptors in the nigrostriatal system is responsible for EPSEs, impaired dopamine function in the nucleus accumbens seems to give rise to dysphoria. Hence, drug-induced dysphoria is not a variant of EPSE, but both are variants of drug-induced dopaminergic blockade. Accordingly, under second-generation antipsychotics which rely much less exclusively on dopaminergic blockade for their therapeutic activity, dysphoria and EPSE have become less frequent. Neuroleptic-induced akinesia or akathisia

Akinesia and akathisia are often discussed as confounding symptoms in the identification of depressive symptoms in schizophrenia. Akinesia as a treatment side effect is primarily caused by conventional antipsychotic medication, sometimes accompanied by blue mood and social withdrawal. The differential diagnosis can be clarified by treatment “ex juvantibus” with antiparkinsonian medication. Akathisia is characterized by psychomotor agitation and does not respond well to antiparkinsonian drugs. According to Hausmann and Fleischhacker (2002), diagnostic difficulties occur when one is confronted with its abortive form, where patients exert more signs of dysphoria than motor restlessness. This may be mistaken for agitated depression and both suicidal ideation and behavior have been associated with it. To distinguish the psychomotor agitation from depressive symptoms an attempt at treatment with benzodiazepines or betablockers should be made.

158

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

Negative symptoms

Severe depressive symptoms apparently have an overlap with symptoms of the negative syndrome in schizophrenia: diminished interest, pleasure, energy, or motivation along with psychomotor retardation and impaired ability to concentrate. For a better discrimination of depressive symptoms from negative symptoms, Siris (2000) pointed out that “blunted affect, for example, suggests negative symptoms, whereas distinct blue mood or cognitive features such as guilt or suicidal thoughts suggest depression.” Similarly, Sands and Harrow (1999) state that “depression can be differentiated from negative symptoms when mood symptoms, including sad or blue emotional states, negative self-esteem, guilt, and pessimism are considered.” Anhedonia is usually claimed as a major component of negative symptoms in schizophrenia. Within the tripartite model of depression and anxiety, which nonetheless holds that anhedonia represents a relatively specific marker of depression, Joiner et al. (2003) have tested the hypothesis that major depression exceeds schizophrenia in the amount of anhedonia. As predicted, patients with major depression scored significantly higher on a three-item anhedonic symptoms subscale of the (self-report) Beck Depression Inventory (BDI) than did patients with schizophrenia. Since there was no difference between the two groups on the BDI total Sumscore or the BDI non-anhedonic symptoms Sumscore, it might be concluded that anhedonia – as measured by self-report – is more frequently related to depressive symptomatology in major depression than in schizophrenia. Initial or relapse prodromes

Investigations on prodromal symptoms of psychotic relapse describe concordantly the appearance of depression-like symptoms in the pre-psychotic phase of a reexacerbation-like anxiety, withdrawal, guilt, and shame accompanied by dysphoric mood (e.g., Tollefson et al., 1999). As Siris (2000) pointed out, “the appearance of a depression-like state as a prodrome of a new psychotic episode is a short-lived phenomenon before being superseded by more prominent and definitive psychotic symptoms.” In accordance with their relapse preceding appearance, patients are presenting with depressive symptoms in the prodromal state even before the first onset of a psychotic episode. Häfner et al. (1992) showed in a large retrospective study that 81% of all patients develop depressive symptoms at some time during the early course of the disease; 42% of patients in the initial prodromal phase experience depressive symptoms. Schizoaffective depression

The term “schizoaffective disorder” was first used in the early 1930s to describe patients showing an overlap of features of schizophrenia and affective illness (Kasanin, 1933). Today, the DSM-IV describes a schizoaffective disorder in patients

159

Depressive syndromes in schizophrenia

in whom a full affective syndrome coincides with the florid psychotic syndrome but who also have substantial periods of psychosis in the absence of an affective syndrome (Siris, 2000). However, the debate continues as to whether schizoaffective disorder should be considered a type of schizophrenia, a type of affective disorder, a domain on a continuum between schizophrenia and affective disorder, a cooccurrence of two distinct diatheses, or an erroneous concept altogether (Siris, 2000). Clinical symptomatology and illness course

Depressive symptoms are frequent in schizophrenia in all phases of the illness (Zisook et al., 1999), and it has been estimated that approximately 25–80% of patients with schizophrenia exhibit depressive symptoms (El Yazaji et al., 2002). Prevalence studies report widely varying numbers ranging from a low of 7 to 30% and 33 up to 75% (Hausmann and Fleischhacker, 2002). Depressive symptoms are most frequently associated with the acute phase of the illness, with studies suggesting a point prevalence as low as 20% and as high as 80% (Baynes et al., 2000). Lower rates of depressive symptoms are observed in the chronic phase of the illness, where the prevalence rates range from 4% (Addington et al., 1992) to an estimated mean for chronic in-patients of 15% (Leff, 1990). Bressan et al. (2003) found a high rate of depressive episodes occurring during the stable phase of schizophrenia (16.3%) similar to previous findings (Baynes et al., 2000). An overview on the relation between different symptom dimensions and depressive rating scales (self- vs. observer-ratings) in different illness phases is given in Table 9.1. The emergence of depressive symptoms in individuals with schizophrenia has been associated with poor outcome, increased medication usage, greater morbidity at hospital discharge, increased hospital readmission rates, and early relapse. In addition, depression, and especially the associated symptom of self-reported or perceived hopelessness, is an important factor in the assessment of suicide risk (Siris et al., 2001). With respect to illness course, Sands and Harrow (1999) in their study on illness course after 7.5 years have found evidence for poorer overall outcome, greater work impairment, lower activity, greater dissatisfaction and more suicidal tendencies in schizophrenia patients with depression compared to those with schizoaffective depression, psychotic depression or non-psychotic depression. Despite no differences in rates and severity of depressive symptoms, depressed patients with schizophrenia exhibit poorer post-hospital adjustment, less employment, more re-hospitalizations and more psychosis compared to patients with major depression (Sands and Harrow, 1999). Hence, differences among depression in schizophrenia, schizodepression or major depression disorders only seem to appear when regarding outcome variables like psychosocial adjustment.

160

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer Table 9.1. Correlations between different scales and symptom dimensions.

Authors

Depression subjective Symptom scale vs. objective dimension

illness phase Correlations

El Yazaji et al. CDSS (2002) HAM-D

objective objective

PANSS acute phase (5-dimensional)

Lancon et al. (2001)

CDSS

objective

PANSS acute phase (3-dimensional) stable phase

Müller et al. (2001)

BRMES (3-dimensional)

objective

PANSS acute phase (3-dimensional) -

Baynes et al. (2000)

BDI HAM-D

subjective objective

SANS EPSE

stable, chronic phase

CDSS and PANSS-D, HAM-D and PANSS-D, PANSS-N CDSS and PANSS-P, CDSS and PANSS-P, PANSS-N, PANSS-G “retardation” significant correlation with PANSS-N (in SZ, SD, MD), “core depression” and “unspecific depression” no correlation to PANSS-P, PANSS-N BDI no correlations with SANS or EPSE, HAM-D significant correlation with SANS and EPSE

List of abbreviations: CDSS: Calgary Depression Rating Scale for Schizophrenia HAM-D: Hamilton Depression Rating Scale PANSS: Positive and Negative Syndrome Scale PANSS-N: Positive and Negative Syndrome Scale – Negative subscale PANSS-P: Positive and Negative Syndrome Scale – Positive subscale PANSS-G Positive and Negative Syndrome Scale – General symptoms PANNS-D: Positive and Negative Syndrome Scale – Depressive subscale BRMES: Bech–Rafaelsen–Melancholia Scale SZ: Schizophrenia SD: Schizodepression MD: Major Depression BDI: Beck Depression Inventory SANS: Scale for the Assessment of Negative Symptoms EPSE: Extrapyramidal Side Effects

161

Depressive syndromes in schizophrenia

The relation among different clinical symptom dimensions seems to depend on the scales used as well as on the illness phase under investigation (acute, post-acute, remitted, or chronic). Accordingly, the Calgary Depression Rating Scale for Schizophrenia (CDSS), specifically developed for schizophrenia, although initially reported to be unidimensional (Addington et al., 1992), has been found to correlate with the Positive and Negative Syndrome Scale – Positive subscale (PANSS-P) in the acute phase, but with PANSS-P and PANSS-N (Negative subscale) as well as with PANSS-G (General symptoms) in the stable phase (Lancon et al., 2001) and with EPSE (Peralta and Cuesta, 1999). On the other hand, the factor analytically derived PANSS-D (Depression subscale) Sumscore as an orthogonal symptom dimension in schizophrenia has been shown to be correlated with the Hamilton Depression Rating Scale (HAM-D) and the CDSS (El Yazaji, et al. 2002). The relationship between depressive and negative symptoms may be especially phase- and scale-dependent (Salokangas et al., 2002). Accordingly, other scales have also been found to correlate with negative symptoms, EPSEs etc. The Bech–Rafaelsen Melancholia Scale (BRMES) has been evaluated as three-dimensional in acute schizophrenia (Müller and Wetzel, 1998), exhibiting dimensionspecific symptom correlates (Müller et al., 2001). Whereas the retardation factor was found to correlate significantly with the PANSS-P score in non-residual schizophrenia with depressive symptoms, schizodepressive disorder, and major depression with psychotic features, neither their core depressive symptom factor nor the accessory depressive symptom factor showed any significant correlations with negative or positive symptoms or extrapyramidal symptom ratings in any of the three disorders. As the authors conclude, the results suggest that depression represents a heterogeneous symptom domain with unique relationships of components to positive and negative symptoms across nosological borders (Müller et al., 2001). Subjective and objective depression rating scales in particular seem to have different correlates (Baynes et al., 2000; Dollfus et al., 2000). Assessments with the subjectively based BDI in stable, chronic schizophrenic patients (Baynes et al., 2000) have shown that those with BDI scores or17 have significantly higher scores on the Brief Psychiatric Rating Scale (BPRS) factors for anxiety/depression and hostility/suspiciousness, as well as the positive symptom and thought disturbance factors. No difference was found in the scores of the Scale for the Assessment of Negative Symptoms (SANS) and no correlations between BDI and SANS. However, significant associations were observed between the objectively based HAM-D and SANS scores as well as between HAM-D and EPSE scores. The authors concluded that rating scales such as the HAM-D with emphasis on the objective features of depression may not be the most useful, whereas scales with a focus on patients’ view of themselves may be more useful in differentiating depressive from negative symptoms. Dollfus et al. (2000) have

162

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

similarly reported that subjectively assessed extrapyramidal symptoms are unrelated to negative symptoms (SANS), whereas objectively assessed EPSEs are correlated with depression ratings (Montgomery–Asberg Depression Rating Scale [MADRS]). Facial mask and bradykinesia account for 41% of the variance of the SANS, but for only 17% of the MADRS. Their conclusion is that the subjective component of EPSEs can be extremely interesting and should be given more consideration than the objective one in evaluating EPSEs. Peralta and Cuesta (1999) conclude that “rating scales not measuring constructs other than those they are intended to measure are clearly necessary. Another possibility is to introduce ratings of patients’ subjective experiences, which may be useful in discriminating disorders with similar behavioral manifestations.” Kurt Schneider, the famous German psychopathologist, in his Clinical Psychopathology (1971) writes: “All mental processes are only accessible through expression, through expressive language, writing, facial expression and the other motor activities. Expression may also be of diagnostic significance”. . . . But he continues: “Overall it may be said that for diagnostic purposes, properly gathered modes of experience take priority over abnormalities of expression, the latter turning into impressions in the observer with all sources of subjective error”. (Translation by the authors.) It is in this context that the definition of affect according to DSM-IV should be considered in the further discussion: Affect is a pattern of observable behaviors that is the expression of a subjectively experienced feeling state (emotion). Experimental psychopathology Sources of information in psychopathology are generally subjective-verbal, motor and physiological. Hence, methodological approaches to psychopathology are selfratings, observer ratings, behavioral observation, and measurement of (re)action. Especially in the field of depressive states and similar-looking syndromes such as affective flattening or Parkinsonian symptoms – conceived as having psychomotor retardation as a common denominator (Benson, 1990; see Figure 9.1) – for their mutual delineation it seems essential to use the full spectrum of available methodological approaches. Accordingly, the use of a laboratory approach to assess behavior under defined experimental conditions in addition to clinical interviewing and application of subjective and objective rating scales has been recommended (Gaebel and Wölwer, 1992; 1996). This approach avoids confounding of information from different information channels (Polzer and Gaebel, 1993) and thereby allows segregation of overlapping behavioral phenotypes as a clue to the identification of underlying neurobiological dysfunctions. The following example from our own research has

163

Depressive syndromes in schizophrenia

Affective flattening

Psychomotor retardation Parkinsonism

Figure 9.1.

Depression

Psychomotor retardation: a common denominator of different clinical syndromes? (From Benson, 1990.)

adopted some essential principles of this approach and contributes to answering the question of nosological specificity of depression in schizophrenia. Affective behavior was studied in acute inpatients with schizophrenia (n 34) and major depression (n23) – both diagnosed according to Research Diagnostic Criteria (RDC) – and in healthy controls (n22) under standardized laboratory conditions (Gaebel and Wölwer, 1992; Wölwer et al., 1996). All assessments took place twice (at admission [T0] and four weeks later [T1]). Spontaneous facial activity, body- and object-focused gestures, speech activity (experimental interview condition), voice characteristics (voice pitch during reading of emotional standard sentences), and voluntary facial activity (simulation and imitation of emotional faces) were recorded on video- and audiotape, and were either coded (facial activity, hand movements, speech activity) or directly measured (voice pitch modulation) offline. In addition to expressive behavior, facial affect recognition was assessed. Facial activity was evaluated by using the “Facial Action Coding System” (FACS; Ekman and Friesen, 1978). The clinical psychopathological status was assessed with BPRS, SANS (self- and observer-rating version), HAM-D, and EPSE Rating Scales. Affective behavior both in schizophrenia and in depression was significantly different from normals in almost every respect, i.e., spontaneous facial activity as well as gestural and speech activity were reduced, and voice pitch was less modulated. Additionally, voluntary facial imitation and simulation performance were impaired, and the total number of facial Action Units (AUs), as indicators of mimic

164

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer Observer rating (Global score)

Self-rating (Subjective experience)

6

6

5

5

4

4

3

3

* *S

2

D

D S

1

1

0

0 T0

Figure 9.2.

2

T1

T0

T1

SANS-based assessment of affective flattening; D  depression, S  schizophrenia. (From Gaebel and Wölwer, 1996.)

expressivity, was reduced (Wölwer et al., 1996). In depression, but not in schizophrenia, speech modulation improved and body-focused gestures increased between T0 and T1. Facial affect recognition was significantly impaired in schizophrenia only (Gaebel and Wölwer, 1992; Gaebel and Wölwer, 1996). Quantitative reduction in spontaneous affective behavior and poorer quality of voluntary facial activity were relatively stable over time – despite clinical improvement – but were generally more pronounced in schizophrenia compared to depression. However, none of the behavioral measures differentiated between schizophrenia and depression at T0. Neuroleptic side-effects did not contribute significantly to the general reduction of expressive behavior. In schizophrenia, HAM-D sumscores were lower than in major depression, which was mainly owing to significantly lower item scores for depressive mood or depressive inhibition, whereas HAM-D item scores less specific for depression (e.g., work impairment, paranoid symptoms, depersonalization, lack of insight, anxiety, activation) were as high as or even higher than in major depression. On the other hand, whereas objective SANS ratings of affective flattening did not differ between both groups at T0, subjective ratings in schizophrenia were incongruently low compared to depression (see Figure 9.2). As revealed by regression analysis, in schizophrenia objectively rated depressive symptoms are related to subjectively felt blunting of affect, whereas in depression they are related to objective (and subjective) reduction in expression. On the other hand, in schizophrenia objectively rated negative symptoms are related to objective (and subjective) reduction in expression (as well as to akinesia), whereas in depression they are related to subjectively felt blunting of affect. In summary (see Figure

165

Depressive syndromes in schizophrenia

SCHIZOPHRENIA SCHIZOPHRENIA

ObjectivelyObjectively rated rated psychopathology psychopathology

Negative Negative DepressiveDepressive Akinesia Akinesia symptomssymptoms symptomssymptoms

Sources ofSources information for of information for objectivelyobjectively rated rated psychopathology psychopathology

Subjective Subjective flattening of affect of affect flattening Objective reduction expressivity Objectivein reduction in expressivity

DEPRESSION DEPRESSION ObjectivelyObjectively rated rated psychopathology psychopathology

Figure 9.3.

DepressiveDepressive Negative Negative symptomssymptoms symptomssymptoms

Subjective and objective sources of information for observer ratings. (From Gaebel and Wölwer, 1996.)

9.3), depressive symptomatology ratings in schizophrenia are especially related to subjective experience; in depression they are related to objective and subjective correlates (for negative symptoms it is just the opposite). In conclusion, whereas affect recognition was impaired in schizophrenia only, facial expressive functions were persistently disturbed – despite clinical improvement – in a trait-like fashion both in schizophrenia and depression. However, this behavioral similarity seems to differently inform psychopathological observer ratings: in major depression the reduction of affective expression becomes associated with depressive symptoms, whereas in schizophrenia it becomes associated with negative symptoms. Concerning the nosological debate on depression in schizophrenia, depressive symptoms seem to reflect rather a subjective impression of affective flattening combined with unspecific depressive symptoms instead of more specific depressive symptoms like depressive mood or depressive inhibition in major depression. Although (depression in) schizophrenia and major depression cannot be differentiated by means of objectively assessed facial expressivity, the results on subjective experience of affective flattening demonstrate that there is a difference between the two disorders. Hence, in accordance with Peralta and Cuesta (1999), patients’ subjective experience may be important in discriminating disorders with similar behavioral manifestation (Gaebel and Wölwer, 2004).

166

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

Neurocognition, neuroimaging, genetics Neurocognition

A key cognitive function that has repeatedly been found to be impaired in schizophrenia is memory. Recent models have also emphasized the role and the potential influence of depressive symptoms in memory. Table 9.2 gives an overview of recently published studies about neurocognitive dysfunctions and their relationship with depressive symptoms in schizophrenia and major depression. Brebion et al. (1997); Brebion et al. (2000) from their research on memory dysfunction in schizophrenia reported that deficits in verbal memory (effortful deep encoding reflected by semantic clustering) are related to depression (HAM-D) and – independently – to psychomotor retardation and lack of motivation. As expected from findings in depression, depressive symptoms in schizophrenia are consistently linked with all measures reflecting deficits in deep encoding, even when positive and negative symptoms are statistically considered. The authors conclude that the links between severity of depressive symptoms and effortful memory processes in depression seem to be applicable to schizophrenia (Brebion et al., 1997), and they suggest that the observed verbal memory impairment in schizophrenia is a consequence of depressive symptoms, rather than a feature of the schizophrenic disorder itself (Brebion et al., 2001). On the contrary, Hill et al. (2004) by applying a neuropsychological test battery to patients with psychotic unipolar depression, nonpsychotic unipolar depression, schizophrenia (without depressive symptoms), and healthy controls, found that patients with psychotic depression had a neuropsychological dysfunction comparable – but less severe – to that of patients with schizophrenia. The cognitive profile for non-psychotic unipolar depression differed dramatically from that for psychotic depression and was more consistent with the cognitive profile for the healthy comparison group. The authors conclude that “the similar neuropsychological profiles for schizophrenia and psychotic depression suggest that these psychotic disorders may have common pathophysiological features. The dramatic differences in performance between the patients with psychotic depression and those with non-psychotic depression point to a marked distinction in neurocognitive function associated with the expression of psychosis in depressed patients.” Accordingly, the advent of psychotic symptoms seems to be responsible for neurocognitive disturbances and not – as pointed out by Brebion et al. (2000) – that of depressive symptoms. Holthausen et al. (1999) reported that PANSS-D is correlated with objective (STROOP colors, TMT-A/B, Fingertapping), but even more so with subjective cognitive measures (distractibility, overload, diminished attentional control). In conclusion, they speculate that depressive symptoms are accompanied by less efficient information processing, as a result of which patients have to invest more effort,

167

Depressive syndromes in schizophrenia Table 9.2. Neurocognitive findings in schizophrenia patients with depression.

Authors

Results

Conclusion

Brebion et al. (2001)

Deficits in verbal memory (deep encoding) are linked to depressive symptoms in schizophrenia.

Verbal memory impairment in schizophrenia is a consequence of depressive symptoms.

Hill et al. (2004)

Cognitive profiles for non-psychotic major depression differ from psychotic depression and schizophrenia.

The advent of psychotic symptoms is responsible for the neurocognitive impairments in schizophrenia.

Kohler et al. (1998a)

Poorer attentional performance in schizophrenia patients with HAM-D or  18.

Depressed mood, regardless of its etiology, involves frontal cortex, which is also required for attentional processing.

Bozikas et al. (2004)

PANSS-Cognitive dimension correlates to executive function, attention, verbal memory, PANSS-N correlates with verbal memory and auditory attention, PANSS-D shows no correlation with the neurocognitive tests

The relationship between psychopathology and cognitive deficits suggests that they reflect distinct constructs.

which, in turn, results in subjective cognitive complaints in spite of normal results on neuropsychological tasks. This points to the importance of subjective measures in cognitive research. Kohler et al. (1998a) investigated whether patients with schizophrenia and moderate depressive symptoms exhibit more severe neuropsychological deficits than their counterparts without depression. They found poorer attentional performance (CPT, vigilance component) in schizophrenia patients with HAM-Dor18 and concluded that this supports “the notion that depressed mood, regardless of its etiology, involves similar brain systems, namely frontal cortex, which is also required for attentional processing. In schizophrenia, depression is associated with greater disturbance in attentional functioning, and we conclude that schizophrenia patients with depression exhibit neuropsychological impairment consistent with depression superimposed on impairment associated with schizophrenia.” Bozikas et al. (2004) hypothesized – because of the observed correlations between cognitive impairments and negative symptoms, but not depressive symptoms in schizophrenia – that “negative symptoms and cognitive functioning may result from the same underlying factors, but stressors in the environment may effect them differentially.” They concluded that “the modest relationship between

168

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

psychopathology and cognitive deficits that we found suggest that these deficits, although related to the symptoms of schizophrenia, are distinct constructs” (Bozikas et al., 2004). Neuroimaging (a) Structural imaging

Despite consistent descriptions of depressive symptoms in schizophrenia, little is known about their neurobiological underpinnings. One study which examined the association of depression in schizophrenia with measures of brain anatomy is the magnetic resonance imaging (MRI) study by Kohler et al. (1998b). The authors found in the “high depression group” (patients with HAM-D-scoreor18) larger bilateral temporal lobe volumes in comparison with a “low depression group” (HAM-D-score 18) and a healthy control group. The results indicate that the brain morphology of depression in schizophrenia has features in common with major depression. Smith et al. (2003) found patients with first-episode schizophrenia having significantly larger hippocampal fissure (HF) volumes compared to healthy volunteers, the HF size being significantly associated with anxietydepression symptoms during illness onset. The major finding of Salokangas et al. (2002) was that patients with first-episode schizophrenia had reduced left frontal gray matter volume compared with patients with severe psychotic or non-psychotic depression and healthy controls. Additionally, they found that white matter volume differentiates between patients with nonpsychotic (depression) and psychotic disorders (schizophrenia and psychotic depression), a significant interaction of diagnosis, region and hemisphere for white matter volume. Accordingly, “white matter volume may be an important factor in the differential diagnosis of psychotic and non-psychotic disorders in general, and between psychotic and non-psychotic depression specifically” (Salokangas et al., 2002). (b) Functional imaging

In the study by Kohler et al. (1998b), patients additionally underwent fluorodeoxyglucose-positron emission tomography (FDG-PET) measurement of cerebral glucose metabolism. The “high depression” group showed significantly decreased laterality (left minus right of metabolic activity) in the anterior cingulate. Whereas healthy subjects showed relatively higher left hemispheric metabolism in this region, this gradient was diminished in the high depression group and increased in the low depression group. Therefore, depression in schizophrenia seemed to be associated with relative decrease in left compared to right cingulate activity. These results are consistent with the findings by Bench et al. (1992) who found a left lateralized decrease in dorsal cingulate function. Since the cingulum

169

Depressive syndromes in schizophrenia

and mesial temporal regions form the limbic lobe, which is the instrumental part of the neural circuitry responsible for emotional processing, these results of functional imaging studies are particularly meaningful. A number of other functional imaging studies have been mainly concerned with a dysfunction of the dopaminergic system. Bressan et al. (2002) reported on a significant correlation between striatal dopamine D2-receptor occupancy (SPECT) and depressive symptoms (BPRS-D) in schizophrenia under typical neuroleptics. As mentioned above, higher striatal D2-receptor blockade by first-generation antipsychotics may contribute to the emergence of depressive symptoms in schizophrenia. Hietala et al. (1999) studied depressive symptoms and presynaptic dopamine function in neuroleptic-naive schizophrenic patients. Striatal dopamine synthesis capacity (FDOPA [fluorodopa] uptake, Ki values) was studied with PET. Whereas patients showed an overall increased FDOPA uptake in the striatum, left striatal FDOPA uptake (Ki) values in particular were significantly negatively correlated with depressive symptoms (PANSS-D). Accordingly, depressive symptoms even in neuroleptic-naive first-admission schizophrenia seem to be associated with low presynaptic dopamine function. This finding may have drug treatment implications for those patients prone to developing depression under predominant D2-receptor antagonist type antipsychotics. From mood induction studies with functional magnetic resonance imaging (fMRI), Schneider et al. (1998) reported that schizophrenia patients – unlike healthy controls – did not show amygdala activation during sadness, despite matched ratings to controls indicating a similar negative affect. Mayberg (2002) by means of FDG-PET contributed to the localization of the pathophysiological processes of unipolar depression. Compared with healthy controls, depressed patients demonstrated hypometabolism in the dorsal and ventral prefrontal cortex (F9/46), the inferior parietal region (P40), and the anterior cingulate cortex (CG24), as well as the interior insula and posterior cingulate. A discrete area in the rostral cingulate cortex (CG24a) predicted eventual antidepressant treatment response: in those who responded, glucose metabolic activity in this area was high; in non-responders, it was low. Post-treatment, in the state of full remission, relative hyperactivity persisted in this area. (c) Theoretical considerations

According to more sophisticated modern neurobiological theories, emotion processing and affect regulation occur within complex “emotion circuits” in the brain (LeDoux, 2000; Davidson et al., 2000), comprising prefrontal cortex (PFC), anterior cingulate cortex (ACC), amygdala, insular cortex and hippocampus as key structures (Davidson and Irwin 1999, Davidson et al. 2002) (see Figure 9.4). Each of these regions appears to play a separate function in emotion. Davidson and Irwin

170

Figure 9.4.

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

a

b

c

d

Key brain regions involved in affect and mood disorders: (a) orbital prefrontal cortex (OPFC) and ventromedial prefrontal cortex (VMPFC); (b) dorsolateral prefrontal cortex (DLPFC); (c) hippocampus and amygdala; (d) anterior cingulate cortex (ACC).

(1999) more specifically identified circuits underlying positive and negative emotion in the human brain, placing emphasis on the prefrontal cortex (PFC) and the amygdala as two key components of this circuitry. Abnormalities in activation of prefrontal regions in depression have been reported more frequently than for any other brain region, mostly in the direction of decreased bilateral or predominantly left-sided activation (Davidson and Irwin, 1999). Miller and Cohen (Miller and Cohen, 2001) integrating various findings from anatomical, physiological and neuroimaging studies in their “theory of prefrontal function” concluded that the PFC maintains the representation of goals and the means to achieve them. Thus enabling affect-guided anticipation, the PFC is likely to function as an affective working memory (Davidson et al. 2000). Besides the PFC, the ACC seems to play an important role with regard to affect regulation. Several authors propose the ACC as a bridge between attention and emotion, because it contributes to selective attention, affect and specific social behaviors (e.g., Devinsky et al., 1995, cited by Davidson et al., 2002). Davidson and Irwin (1999) additionally emphasize its role with regard to attention to subjective and conscious emotional responses.

171

Depressive syndromes in schizophrenia

Interconnectivity PFC

disturbed function MAJOR DEPRESSION ?

Figure 9.5.

ACC

disturbed function DEPRESSION IN SCHIZOPHRENIA ?

Different subtypes of depression? (From Davidson et al., 2002.)

As PFC and ACC are extensively interconnected, PFC function in depression could be disturbed through dysfunction of the PFC itself, of ACC, or of both structures (Davidson et al., 2002). Accordingly, Davidson et al. (2002) speculated that there might exist different subtypes of depression characterized by defects localized in one or other of the two structures. A depression of the ACC-subtype might cause patients not to experience the conflict between their current state and the demands of everyday life (see Figure 9.5). Patients with a PFC-subtype depression on the other hand might experience this conflict and suffer because of it, but nevertheless are not able to activate the PFCbased mechanisms to organize and guide behavior towards the resolution of the conflict. The incongruence between objectively assessed and subjectively felt affective flattening in schizophrenia (see Figure 9.2) possibly originates from dysfunctions of the ACC, whereas depressive patients might show a PFC-based dysfunction, as they experience their affective disturbances more congruently. Another important neural structure is the amygdala (LeDoux, 2000; Davidson et al., 2002). Although LeDoux (2000) states that the amygdala is an important component of the system involved in acquisition, storage and expression of fear memory, other researchers do not restrict its function to negative affect (Davidson et al., 2002). The amygdala has been consistently identified as playing a crucial role in both the perception of emotional cues and the production of emotional responses (Davidson and Irwin, 1999). Genetic findings

Several family studies (e.g., Kendler et al., 1993) found a familial association between schizophrenia and psychotic affective illness, but not affective illness in general. A familial relationship between schizophrenia and unipolar depressive disorder (Maier et al., 1993) and the occurrence of psychotic affective illness in parents of offspring with schizophrenia (Kendler et al., 1996) has been described too.

172

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

Kendler and Hays (1983) have reported that schizophrenia patients with firstdegree relatives with unipolar depression have a significantly higher likelihood of developing depression. The findings, however, could not be replicated by other investigators (Guze et al., 1983; Berrios and Bulbena, 1987). Sibling-pair studies feature another assessment approach. Kendler et al. (1997) determined whether the clinical manifestation of schizophrenia and other psychotic disorders are correlated in affected sibling-pairs. Contrary to others (e. g. DeLisi et al., 1987) the authors found three symptom factors – negative, positive and affective symptoms – significantly correlated in concordant sibling-pairs. Cardno et al. (1998) searched for associations of schizophrenia subtypes, psychotic symptoms (defined by SAPS and SANS), affective episodes and demographic variables in 109 sibling-pairs with DSM-IV schizophrenia or schizoaffective disorder and found no significant associations. The results of family as well as sibling-pair studies consistently describe a familial association between schizophrenia and affective disorders, supporting the assumption of a joint genetic etiology of depression and schizophrenia. Molecular genetic studies also contribute to the question about the biological association between depression and schizophrenia. Asherson et al. (1998) reported results on linkage of markers on chromosome 4p in a sample of 24 multiply affected families with schizophrenia and affective symptoms. They concluded that a broad phenotype including unipolar depression, bipolar disorder, schizoaffective disorder and schizophrenia when accompanied by significant affective symptoms can result from mutations within a gene in this region. The dopamine D5-receptor gene lies in the region identified by the linkage studies and is therefore a major candidate for a putative disease gene. Kendler et al. (2000) identified in patients from families with linkage to chromosome 8p significantly more affective deterioration, poorer outcome, more thought disorder, and fewer depressive symptoms than affected individuals from other families. According to the authors, these results raise the possibility that the putative susceptibility gene for schizophrenia localized in the 8p22–8p21 region may predispose individuals to the core dementia praecox syndrome. In research on the role of the serotonin transporter gene (5-HTTLPR) as a primary candidate in major psychosis, Serretti et al. (1999a, 1999b) assessed the presumed associations to a number of psychopathological conditions in schizophrenia. The finding that the 5-HTTLPR variants were not associated with excitement, depressive, disorganized, and delusional symptomatology factors, led the authors to the conclusion that the serotonin transporter gene is not a liability for the symptomatology of schizophrenia (Serretti et al., 1999b). In further research into the role of D3- and D4-receptor gene variants (DRD3, DRD4) and their association to the symptomatology of schizophrenia disorders (using mania,

173

Depressive syndromes in schizophrenia

depression, delusion, and disorganization as symptom factors), no association with the D3 and D4 gene variants could be demonstrated (Serretti et al., 1999c). Regarding the differentiation in self- and observer-ratings in schizophrenia, Golimbet et al. (2004) focused on the relation between 5-HTTLPR polymorphism and clinical presentations of schizophrenia, specifically those related to the depression-anxiety spectrum. Patients with the “ss” genotype (deletion variant) scored significantly higher on PANSS-items depression and guilt feelings, as compared with those of the “ll” genotype (insertion variant), and the frequency of the “ss” genotype was reduced in patients with no depression or no guilt feelings. However, with respect to self-ratings on depression and anxiety, no differences between genotypes were obtained, although patients with higher PANSS-D scored significantly higher on self-ratings. Obviously, the “ss” genotype is more likely to be associated with clinically rated depression and related traits, than with corresponding characteristics rated subjectively by schizophrenia patients. Treatment approaches From the viewpoint of depression in schizophrenia, there are a number of reasons to recommend the use of second-generation antipsychotic drugs. First, atypical antipsychotic drugs have a greatly reduced extrapyramidal side-effect profile, which is likely to overlap with depressive symptoms. Second, atypical agents rely much less exclusively on dopaminergic blockade for their therapeutic activity. The assumed relationship between dopaminergic blockade and dysphoria in schizophrenia can be avoided under atypical antipsychotics. Third, atypicals have advantages in treating negative symptoms, which sometimes appear similar to depression (Siris, 2000). A number of studies have in fact reported on the more favorable antidepressive effect of atypicals, e.g., olanzapine vs. haloperidol (Tollefson et al., 1998), ziprasidone vs. placebo (Keck et al., 1998), risperidone vs. haloperidol (Peuskens et al., 2000), olanzapine vs. risperidone (Tollefson et al., 1999), amisulpride vs. flupentixol (Müller et al., 2002) and amisulpride vs. haloperidol risperidone (Peuskens et al., 2002). In particular, the latter two studies on amisulpride demonstrate that a sole effect on D2/D3-receptors may also be sufficient to improve depressive symptomatology in schizophrenia. As has been shown by means of path analysis, the more favorable antidepressive effect, e.g., for olanzapine, seems to be direct effect, separable from effects on positive, negative, and motor symptoms (Tollefson et al., 1998). Aripiprazole is a new high-affinity partial agonist at D2-receptors, with partial agonist activity at 5-HT1A-receptors, and antagonist properties at 5-HT2Areceptors. Aripiprazole-treated patients exhibit significantly greater improvement in negative and depressive symptoms than those receiving haloperidol (Green,

174

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

2004). The beneficial effects on depression and anxiety may be related to aripiprazole’s serotonergic effects (Carson et al., 2003). In case antipsychotics alone do not improve depressive symptoms, antidepressants are recommended. Levinson et al. (1999) report on results which suggest evidence supporting adjunctive antidepressant treatment for schizophrenic or schizoaffective patients who develop a depressive syndrome after remission of acute psychosis, but there were mixed results for treatment of subsyndromal depression. According to an international survey, US psychiatrists prefer (Siris et al., 2001) for the treatment of depression in schizophrenia – in descending order – fluoxetine (62%), sertraline (62%), paroxetine (49%), trazodone (18%), nortryptiline (18%), buproprion (16%), and venlafaxine (15%). Overall, 33% of respondents reported rarely or never prescribing a combination of antidepressant and antipsychotic drugs. The most common reason was the risk of exacerbating the psychotic disorder (Addington et al., 2002). Finally, a number of psychotherapeutic approaches are used for depression in schizophrenia. From the above survey, frequently used approaches by USpsychiatrists are – in descending order – psychoeducation, family interventions, non-focused support, task-oriented rehabilitation, interpersonal therapy, supportive psychodynamic therapy and cognitive-behavioral approaches (Siris et al., 2001). Summary and conclusion Epidemiological studies have documented a depression-like syndrome occurring during the longitudinal course of schizophrenia in approximately 25% of the patients (Addington et al., 2002). The occurrence of depressive symptoms in the course of illness is associated with worse outcome, i.e., impaired functioning, personal suffering, higher rates of relapse or re-hospitalization, and suicide. Differential diagnosis requires the distinguishing of depression from other clinical features of schizophrenia (florid psychosis, negative symptoms) or side-effects of treatment (akinesia). The correlation with various symptom dimensions (e.g., negative, positive symptoms) seems to depend on the rating scales used regarding their item content and information source (self- vs. observer-rating) as well as the illness phase (acute, post-acute, remitted, or chronic). Therefore, it has been recommended to use especially developed unidimensional scales or orthogonal scale factors under consideration of subjective and objective access to information. Neurobiological correlates and genetic findings point to certain similarities with depression in primary affective disorders. From a treatment perspective, second-generation antipsychotic drugs should be first choice but, in case of insufficient response, may need combination with antidepressants. Not the least because of the suicidal risk, psychotherapeutic approaches are necessary to avert demoralization and disap-

175

Depressive syndromes in schizophrenia

pointment. Both in schizophrenia (with and without depression) and in depression, disturbances of affect (i.e., objective emotional expression) and emotion (i.e., subjective emotional experience) are important components of the clinical syndrome. In schizophrenia, there seems to be a persistent dissociation between experience and expression of emotions, whereas in depression there is a temporary, but concordant reduction of both emotional aspects. Obviously, for diagnosing a depressed state in schizophrenia subjective experience of expressive dysfunction seems to suffice, whereas in depression both subjective experience and objective expressive dysfunction are needed. In other words, there seems to be a methodological and a functional difference: given diagnostic non-blindness of raters, implicit diagnostic concepts may confound with psychopathological ratings – ratings of depressive symptomatology in schizophrenia seem more likely to be inferred from subjective experience of expressive deficits, whereas in true depression they are inferred both from subjective and objective measures; for negative symptoms it is just the opposite. Besides these potential methodological confounds, however, a functional difference might be inferred from the data. Given a simplified model of a brain circuit subserving integrated emotional-affective behavior with an interrelated emotional, i.e., a feeling module, and an affective, i.e., an expression module, various kinds of disturbance/dissociation may occur: either the emotional module or the expression module may be – temporarily or persistently – deficient, or the connection between both may also be deficient. Whereas in depression both modules seem to be temporarily hypofunctional (pointing to their potential intactness including their interconnections, with the primary dysfunction possibly being at the emotional module), in schizophrenia (and particularly in schizophrenic depression) a complex and more persistent dysfunction of the whole circuit seems to exist. The expressive module seems to be dysfunctional in a quantitative and qualitative sense, clinically relating to negative symptoms including inadequate affect. However, schizophrenia patients may either be subjectively aware of this dysfunction (as in true depression), or not. A less pronounced awareness deficit seems to be present in those who are rated as being more depressed. Hence, subjective “insight” in expression deficits – irrespective of objective deficits – seems similarly to characterize depressed schizophrenia patients and depression patients. However, it is the degree and pattern of subjective/objective disturbances which characterize the similarities and differences between these two types of “depression”. Whether these two types coincide with the ACC- and PFC-subtypes of depression as conceptualized by Davidson et al. (2002) remains to be empirically shown (see Figure 9.6). Whether awareness relates to the expressive module, to the emotional module, or to more extrinsic cognitive modules (which are particularly disturbed in schizophrenia), remains to be demonstrated. Hypo- and/or dysfunctions of the same

176

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer

DEPRESSIVE SYMPTOMS

Objective

+

+

Subjective

-

+

ACC- SUBTYPE

Depressive symptoms in schizophrenia?

Figure 9.6.

PFC- SUBTYPE

Depressive symptoms in major depression?

Subjective and objective symptoms of depression and their hypothetical relationship to different depression subtypes.

and/or different emotional/affective modules together with or without those in other (cognitive) modules may explain the differences in subjective and objective psychopathological findings and their clinical, neurobiological and genetic correlates. Further research – starting from an appropriate bio-behavioral phenotype and overcoming previous methodological shortcomings – will be able to clarify the nosological status of depression in schizophrenia. It then will be possible to decide, whether (1) depression is a homogeneous entity – irrespective of the diagnostic context – referring to changes in activity patterns of underlying brain modules; whether (2) there are subtypes even within a diagnostic category depending on the respective modules affected; or whether (3) it is a heterogeneous construct altogether depending on (colored by) the diagnostic context and all the respective brain modules involved. In other words, it then will be shown from a neurobiological perspective whether depression in schizophrenia or any other condition is the same as in true depression.

R E F E R E N C ES Addington, D., Addington, J., Maticka-Tyndale, E. and Joyce, J. (1992). Reliability and validity of a depression rating scale for schizophrenics. Schizophrenia Research. 6, 201–8. Addington, D., Azorin, J. M., Falloon, I. R. et al. (2002). Clinical issues related to depression in schizophrenia: an international survey of psychiatrists. Acta Psychiatrica Scandinavica. 105, 189–95. American Psychiatric Association (1994): DSM-IV: Diagnostic and Statistical Manual of Mental

177

Depressive syndromes in schizophrenia Disorders. Washington, DC: APA. Asherson, P., Mant, R., Williams, N. et al. (1998). A study of chromosome 4p markers and dopamine D5 receptor gene in schizophrenia and bipolar disorder. Molecular Psychiatry. 3, 310–20. Baynes, D., Mulholland, C., Cooper, S. J. et al. (2000). Depressive symptoms in stable chronic schizophrenia: prevalence and relationship to psychopathology and treatment. Schizophrenia Research. 45, 47–56. Bench, C. J., Friston, K. J., Brown, R. G. et al. (1992). The anatomy of melancholia – focal abnormalities of cerebral blood flow in major depression. Psychological Medicine. 76, 89–93. Benson, D. F. (1990). Behavioral aspects of movement disorders: Neuropsychiatry, Neuropsychology, and Behavioral Neurology. 3, 1–2. Berrios, G. E. and Bulbena, A. (1987). Post psychotic depression: the Fulbourn cohort. Acta Psychiatrica Scandinavica. 76, 89–93. Bozikas, V. P., Kosmidis, M. H., Kioperlidou, K. and Karavatos, A. (2004). Relationship between psychopathology and cognitive functioning in schizophrenia. Comprehensive Psychiatry. 45, 392–400. Brebion, G., Amador, X., Smith, M., Malaspina, D., Sharif, Z. and Gorman, J. M. (2000). Depression, psychomotor retardation, negative symptoms, and memory in schizophrenia. Neuropsychiatry, Neuropsychology, and Behavioral Neurology. 13, 177–83. Brebion, G., Gorman, J. M., Malaspina, D., Sharif, Z. and Amador, X. (2001). Clinical and cognitive factors associated with memory task performance in patients with schizophrenia. American Journal of Psychiatry. 158, 58–764. Brebion, G., Smith, M. J., Amador, X., Malaspina, D., and Gorman, J. M. (1997). Clinical correlates of memory in schizophrenia: differential links between depression, positive and negative symptoms, and two types of memory impairment. American Journal of Psychiatry. 154, 1538–43. Bressan, R. A., Chaves, A. C., Pilowsky, L. S., Shirakawa, I. and Mari, J. J. (2003). Depressive episodes in stable schizophrenia: critical evaluation of the DSM-IV and ICD-10 diagnostic criteria. Psychiatry Research. 117, 47–56. Bressan, R. A., Costa, D. C., Jones, H. M., Ell, P. J. and Pilowsky, L. S. (2002). Typical antipsychotic drugs – D2 receptor occupancy and depressive symptoms in schizophrenia. Schizophrenia Research. 56, 31–6. Cardno, A. G., Jones, L. A., Murphy, K. C., et al. (1998). Sibling pairs with schizophrenia or schizoaffective disorder: associations of subtypes, symptoms and demographic variables. Psychological Medicine. 28 (4), 815–23. Carson, W. H., Archibald, D. G., Manos, G., Kostic, D., Marcus, R. and Stock, E. G. (2003). Shortterm efficacy of aripiprazole on depression/anxiety in schizophrenia. Poster presentation. APA, San Francisco. Davidson, R. J. and Irwin, W. (1999). The functional neuroanatomy of emotion and affective style. Trends in Cognitive Sciences. 3, 11–21. Davidson, R. J., Pizzagalli, D., Nitschke, J. B. and Putnam, K. M. (2002). Depression: perspectives from neuroscience. Annual Review of Psychology. 53, 545–74. Davidson, R. J., Putnam, K. M. and Larson, C. L. (2000). Dysfunction in the neural circuitry of

178

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer emotion regulation: a possible prelude to violence. Science. 289, 591–4. DeLisi, L. E., Goldin, L. R., Maxwell, M. E., Kazuba, D. M. and Gershon, E. S. (1987). Clinical features of illness in siblings with schizophrenia or schizoaffective disorder. Archives of General Psychiatry. 44 (10), 891–6. Dollfus, S., Ribeyre, J. M. and Petit, M. (2000). Objective and subjective extrapyramidal side effects in schizophrenia: their relationships with negative and depressive symptoms. Psychopathology. 33, 125–30. Ekman, P. and Friesen, W. V. (1978). Facial action coding system: a technique for the measurement of facial movement. Palo Alto: Consulting Psychologists Press. El Yazaji, M., Battas, O., Agoub, M. et al. (2002). Validity of the depressive dimension extracted from principal component analysis of the PANSS in drug-free patients with schizophrenia. Schizophrenia Research. 56, 121–7. Gaebel, W. and Wölwer, W. (1992). Facial expression and emotional face recognition in schizophrenia and depression. European Archives of Psychiatry and Clinical Neuroscience. 242, 46–52. Gaebel, W. and Wölwer, W. (1996). Affektstörungen Schizophren Kranker. Stuttgart: Kohlhammer. Gaebel, W. and Wölwer, W. (2004). Facial expressivity in the course of schizophrenia and depression. European Archives of Psychiatry and Clinical Neuroscience. 254, 335–42. Green, B. (2004). Focus on aripiprazole. Current Medical Research and Opinion. 20 (2), 207–13. Golimbet, V. E., Alfimova, M. V., Shchebatykh, T. V. et al. (2004). Serotonin transporter polymorphism and depressive-related symptoms in schizophrenia. American Journal of Medical Genetics. 126, 1–7. Guze, S. B., Cloninger, C. R., Martin, R. L. and Clayton, P. J. (1983). A follow-up and family study of schizophrenia. Archives of General Psychiatry. 40, 1273–6. Häfner, H., Riecher-Rössler, A., Hambrecht, Maurer. K., et al. (1992). IRAOS: an instrument for the assessment of onset and early course of schizophrenia. Schizophrenia Research. 6, 209–23. Hausmann, A. and Fleischhacker, W. W. (2002). Differential diagnosis of depressed mood in patients with schizophrenia: a diagnostic algorithm based on a review. Acta Psychiatrica Scandinavica. 106, 83–96. Hietala, J., Syvalahti, E., Vilkman, H. et al. (1999). Depressive symptoms and presynaptic dopamine function in neuroleptic-naive schizophrenia. Schizophrenia Research. 35, 41–50. Hill, S. K., Keshavan, M. S., Thase, M. E. and Sweeney, J. A. (2004). Neuropsychological dysfunction in antipsychotic-naive first-episode unipolar psychotic depression. American Journal of Psychiatry. 161, 996–1003. Holthausen, E. A., Wiersma, D., Knegtering, R. H. and van den Bosch, R. J. (1999). Psychopathology and cognition in schizophrenia spectrum disorders: the role of depressive symptoms. Schizophrenia Research. 23 (39), 65–71. Joiner, T. E., Brown, J. S. and Metalsky, G. I. (2003). A test of the tripartite model’s prediction of anhedonia’s specificity to depression: patients with major depression versus patients with schizophrenia. Psychiatry Research. 119, 243–50. Kasanin, J. (1933). The acute schizoaffective psychoses. American Journal of Psychiatry. 90, 97–126. Keck, P. Jr., Buffenstein, A., Ferguson, J. et al. (1998). Ziprasidone 40 and 120 mg/day in the acute exacerbation of schizophrenia and schizoaffective disorder: a 4-week placebo-controlled trial.

179

Depressive syndromes in schizophrenia Psychopharmacology. 140, 173–84. Kendler. K. S., Myers, J. M., O’Neill, F. A. et al. (2000). Clinical features of schizophrenia and linkage to chromosomes 5q, 6p, 8p, and 10p in the Irish study of high-density schizophrenia families. American Journal of Psychiatry. 157, 402–8. Kendler, K. S. and Hays, P. (1983). Schizophrenia subdivided by the family history of affective disorder. A comparison of symptomatology and course of illness. Archives of General Psychiatry. 40, 951–5. Kendler, K. S., Karkowski-Shuman, L., O’Neill, F. A. et al. (1997). Resemblance of psychotic symptoms and syndromes in affected siblings pairs from the Irish study of high-density schizophrenia families: evidence for possible etiologic heterogeneity. American Journal of Psychiatry. 154, 191–8. Kendler, K. S., Karkowski-Shuman, L. and Walsh, D. (1996). The risk for psychiatric illness in siblings of schizophrenics: the impact of psychotic and non-psychotic affective illness and alcoholism in parents. Acta Psychiatrica Scandinavica. 94, 49–55. Kendler, K. S., McGuire, M., Gruenberg, A. M. et al. (1993). The Roscommon family study. I. Methods, diagnosis of probands, and risk of schizophrenia in relatives. Archives of General Psychiatry. 50, 527–40. Kohler, C., Gur, R. C., Swanson, C. L., Petty, R. and Gur, R. E. (1998a). Depression in schizophrenia: I. Association with neuropsychological deficits. Biological Psychiatry. 43, 165–72. Kohler, C., Swanson, C. L., Gur, R. C., Mozley, L. H. and Gur, R. E. (1998b). Depression in schizophrenia: II. MRI and PET findings. Biological Psychiatry. 42, 173–80. Lancon, C., Auquier, P., Reine, G., Bernard, D. and Addington, D. (2001). Relationship between depression and psychotic symptoms of schizophrenia during an acute episode and stable period. Schizophrenia Research. 47, 135–40. LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience. 23, 155–84. Leff, J. (1990). Depressive symptoms in the course of schizophrenia. In Depression in Schizophrenia, ed. L. E. DeLisi. Washington, DC: American Psychiatric Press, pp. 1–23. Levinson, D. F., Umapathy, C. and Musthaq, M. (1999). Treatment of schizoaffective disorder and schizophrenia with mood symptoms. American Journal of Psychiatry. 156, 1138–48. Maier, W., Lichtermann, D., Minges, J. et al. (1993). Continuity and discontinuity of affective disorders and schizophrenia: results of a controlled family study. Archives of General Psychiatry. 50, 871–83. Mayberg, H. (2002). Depression, II: Localization of pathophysiology. American Journal of Psychiatry. 159, 1979. Miller, E. K. and Cohen, J. D. (2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience. 24, 167–202. Müller, M. J. and Wetzel, H. (1998). Dimensionality of depression in acute schizophrenia: a methodological study using the Bech-Rafaelsen Melancholia Scale (BRMES). Journal of Psychiatric Research. 32, 369–78. Müller, M. J., Wetzel, H. and Benkert, O. (2002). Differential effects of high-dose amisulpride versus flupentixol on latent dimensions of depressive and negative symptomatology in acute schizophrenia: an evaluation using confirmatory factor analysis. International Clinical

180

Wolfgang Gaebel, Ellen Bittner and Wolfgang Wölwer Psychopharmacology. 17, 249–61. Müller, M. J., Szegedi, A., Wetzel, H. and Benkert, O. (2001). Depressive factors and their relationships with other symptom domains in schizophrenia, schizoaffective disorder, and psychotic depression. Schizophrenia Bulletin. 27, 19–28. Peralta, V. and Cuesta, M. J. (1999). Negative, parkinsonian, depressive and catatonic symptoms in schizophrenia: a conflict of paradigms revisited. Schizophrenia Research. 40, 245–53. Peuskens, J., Möller H. J. and Puech, A. (2002). Amisulpride improves depressive symptoms in acute exacerbations of schizophrenia: comparison with haloperidol and risperidone. European Neuropsychopharmacology. 12, 305–10. Peuskens, J., Van Baelen, B., De Smedt, C. and Lemmens, P. (2000). Effects of risperidone on affective symptoms in patients with schizophrenia. International Clinical Psychopharmacology. 15, 343–9. Polzer, U. and Gaebel, W. (1993). Beeinflussen sich visuelle und auditive Wahrnehmung bei der psychopathologischen Beurteilung von Ausrucksmerkmalen psychiatrischer Patienten? Nervenarzt. 64, 193–8. Salokangas, R. K., Cannon, T., Van Erp, T. et al. (2002). Structural magnetic resonance imaging in patients with first-episode schizophrenia, psychotic and severe non-psychotic depression and healthy controls. British Journal of Psychiatry. 181, 58–65. Sands, J. R. and Harrow, M. (1999). Depression during the longitudinal course of schizophrenia. Schizophrenia Bulletin. 25, 157–71. Schneider, K. (1971). Clinical Psychopathology. Stuttgart: Georg Thieme. Schneider, F., Weiss, U., Kessler, C. et al. (1998). Differential amygdala activation in schizophrenia during sadness. Schizophrenia Research. 34 (3), 133–42. Serretti, A., Lattuada, E., Catalano, M. and Smeraldi, E. (1999b). Serotonin transporter gene is not associated with symptomatology of schizophrenia. Schizophrenia Research. 35, 33–9. Serretti, A., Lattuada, E., Cusin, C. et al. (1999c). Dopamine D3 receptor gene is not associated with symptomatology of major psychoses. American Journal of Medical Genetics. 88 (5), 486–91. Serretti, A., Lilli, R., Di Bella, D. et al. (1999a). No interaction between serotonin tranporter gene and dopamine receptor D4 gene in symptomatology of major psychoses. American Journal of Medical Genetics. 88 (5), 481–5. Siris, S. G. (2000). Depression in schizophrenia: perspective in the era of “atypical” antipsychotic agents. American Journal of Psychiatry. 157, 1379–89. Siris, S. G. and Bench, C. (2003). Depression and schizophrenia, In Schizophrenia, ed. S. R. Hinsch and D. R. Weinberger. Oxford: Blackwell, pp. 142–67. Siris, S. G., Addington, D., Azorin, J. M. et al. (2001). Depression in schizophrenia: recognition and management in the USA. Schizophrenia Research. 47, 185–97. Smith, G. N., Lang, D. J., Kopala, L. C. et al. (2003). Developmental abnormalities of the hippocampus in first-episode schizophrenia. Biological Psychiatry. 53, 555–61. Tollefson, G. D., Andersen, S. W. and Tran, P. V. (1999). The course of depressive symptoms in predicting relapse in schizophrenia: a double-blind, randomized comparison of olanzapine and risperidone. Biological Psychiatry. 46 (3), 365–73. Tollefson, G. D., Sanger, T. M., Beasley, C. M. and Tran, P. V. (1998). A double blind, controlled

181

Depressive syndromes in schizophrenia comparison of the novel antipsychotic olanzapine versus haloperidol or placebo on anxious and depressive symptoms accompanying schizophrenia. Biological Psychiatry. 43 (11), 803–10. Voruganti, L. P. and Awad, A. G. (2004). Is neuroleptic dysphoria a variant of drug-induced extrapyramidal side effects? Canadian Journal of Psychiatry, 49, 285–9. Wölwer, W., Streit, M., Polzer, U. and Gaebel, W. (1996). Facial affect recognition in the course of schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 246, 165–70. Zisook, S., McAdams, L. A., Kuck, J. et al. (1999). Depressive symptoms in schizophrenia. American Journal of Psychiatry. 156, 1736–42.

10

The overlapping of the spectra – brief and acute psychoses Frank Pillmann and Andreas Marneros Department of Psychiatry and Psychotherapy, Martin Luther University Halle-Wittenberg

Introduction Brief and acute psychoses – as differentiated from schizophrenia and affective disorders – have long intrigued psychiatrists. Their nosological allocation has been a matter of debate. The World Health Organization (WHO) tried to solve this problem by creating the ICD-10 category F23: “acute and transient psychotic disorders” (ATPD). The ATPD is defined as a psychotic disorder of acute onset (within two weeks) and of limited duration (not longer than three months). For a diagnosis of ATPD, major affective disorders, organic disorders, and schizophrenia have to be ruled out. The creation of this category represented an attempt not only to define and operationalize such psychotic conditions, but also to integrate some national concepts such as the cycloid psychoses of German psychiatry, the bouffée délirante of French psychiatry, the psychogenic psychosis of Scandinavian psychiatry, the atypical psychosis of Japanese psychiatry and the remitting or goodprognosis schizophrenia of USA psychiatry into one single category (Marneros and Tsuang, 1986; Marneros et al., 2000; Pillmann et al., 2000; 2001; Marneros and Pillmann, 2004). The DSM-IV provides the category of brief psychotic disorder (BPD). A diagnosis of BPD requires the presence of delusions, hallucinations, disorganized speech, or grossly disorganized or catatonic behavior, with a duration of at least one day but less than one month and eventual full return to premorbid functioning. Again, schizophrenia, mood disorder and schizoaffective disorder have to be ruled out. By definition, patients fulfilling the DSM-IV BPD criteria also fulfill the ICD-10 criteria for ATPD, but not those of other ICD-10 psychotic groups. In other words: all DSM-IV (BPD) patients are also ICD-10 ATPD, but the same is not true vice versa. Our knowledge of the nosological assignment of brief and acute psychoses is largely unclear, in particular their relation to schizophrenia, The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

182

183

Overlapping spectra: brief and acute psychoses

schizoaffective disorders, and other psychotic disturbances (Susser and Wanderling, 1994; Jørgensen et al., 1996; Marneros et al., 2000; Marneros and Pillmann, 2004). Therefore, different psychiatric schools have assigned brief and acute psychoses a different place in their nosological systems. These different approaches include attempts to classify brief and acute psychoses as psychogenic or reactive psychoses, to regard them as minor or less severe forms of schizophrenia, or to place them within the affective disorders (Pillmann and Marneros, 1998). The concept of reactive/psychogenic psychoses proved particularly popular in Scandinavian countries (Strömgren, 1986a). August Wimmer, professor of psychiatry at the University of Copenhagen, published a monograph on psychogenic psychoses in 1916 (Wimmer, 1916; see also Pillmann, 2004). Later, the term “reactive” was used interchangeably with “psychogenic,” and finally “reactive” was preferred because it conveyed fewer theoretical assumptions. Although brief and acute psychoses comprise only a sub-group of psychogenic psychoses, they can be regarded as prototypical for the concept (Jørgensen et al., 1997; Ungvari and Mullen, 2000). The concept of reactive (psychogenic) psychosis was never fully accepted in English-language psychiatry. However, a category of “brief reactive psychosis” was introduced in DSM-III (APA, 1980). In DSM-IV it was replaced by “brief psychotic disorder.” For this diagnosis a triggering event is no longer necessary. Both in DSMIV and in ICD-10 a marked stressor can be coded optionally, emphasizing the historical relationship of brief and acute psychoses to the psychogenic/reactive psychoses. Outside Scandinavia, however, most authors would allocate brief and acute psychoses within the broad category that was formerly named “endogenous” psychoses. One possibility is to stress the relationship of brief and acute psychoses to schizophrenia and to view them as minor forms of schizophrenia or as closely related disorders. Such a view can be expressed by concepts such as “good-prognosis schizophrenia” (Vaillant, 1964), “atypical psychosis” (Manschreck and Petri, 1978; Hatotani, 1996; Remington et al., 1990) or “schizophreniform disorder” (Langfeldt, 1939; Strakowski, 1994; Sanchez, 1987). Another alternative is to place brief and acute psychoses with the affective disorders. One influential approach that views brief and acute psychoses as bipolar disorders is represented by the concept of cycloid psychoses. This concept was introduced in 1924 by Karl Kleist (Kleist, 1924; Kleist, 1926; Kleist, 1953), drawing in part on nosological entities described by his teacher Wernicke, such as motility psychosis and anxiety psychosis (Pillmann et al., 2000). Karl Leonhard later modified the concept and included it in his complex nosological system which differentiates more than 30 forms of endogenous psychoses (Leonhard, 1961; Leonhard, 1995). In this system, the cycloid psychoses comprise three distinct forms of psychoses which have in common a phasic course, remission without

184

Frank Pillmann and Andreas Marneros

residual symptoms, and bipolarity: anxiety-elation psychosis, hyperkineticakinetic motility psychosis, and excited-retarded confusion psychosis. Thus, for Leonhard, cycloid psychoses were instances of bipolar disorders. In Sweden, Carlo Perris (Perris, 1974) published a clinical and genetic study of 60 patients with cycloid psychoses that confirmed the validity of the concept. The introduction of ATPD can be seen as an attempt to create a category within ICD-10 to accommodate those patients that would have been diagnosed as having cycloid psychoses in the tradition of Leonhard (Leonhard, 1961) and Perris (Perris, 1974). The limitations of a strictly categorical nosology for the classification of the functional psychoses have been described early by Karl Hoche (Hoche, 1913) and Kurt Schneider (1950). The latter stated that for the endogenous psychoses no differential diagnosis but only differential typology was possible. A number of clinical facts attest against sharp categorical distinctions: the lack of a sharp boundary between diagnostic entities, syndromatic change during longitudinal course, genetic overlap, and, in the pharmacological era, the lack of nosological specificity of psychopharmacological agents. All this prompted the introduction of the concept of the “psychotic continuum” (Marneros et al., 1995). For schizophrenic psychoses and related disorders, a schizophrenic spectrum was postulated (Reich, 1976). Spectrum disorders of schizophrenia were meant to comprise schizoaffective disorders, atypical psychoses and schizotypical disorder (Maier et al., 1994; Faraone et al., 1999). Whether there is a continuum between nonschizophrenic and schizophrenic disorders is still a matter of debate (Schatzberg, 2004). Already Kraepelin conceived manic-depressive illness as a rather broad concept. Later, George Winokur was an important proponent of the affective spectrum (Winokur, 1972). The special position of bipolar illness has a long conceptual history (Marneros and Angst, 2000). For modern psychiatry it was confirmed by Angst (1966) and Perris (1966). Even bipolar illness, however, proved not to be a homogenous entity. The classical category of bipolar disorders was extended by the concept of bipolar II disorders (Dunner et al., 1976). Meanwhile, the notion of the bipolar spectrum has been extended further. For some authors it does not only encompass affective episodes with different combinations of maniform, dysphoric and depressive components, but also variants of temperament and personality disorders including borderline personality disorder (Marneros, 2001; Akiskal, 2002). The concepts of the schizophrenic and the affective spectrum, however, have not solved the problem of the nosological allocation of brief and acute psychoses. Rather, we have to answer the question of the relationship of brief and acute psychoses to the schizophrenic and the affective spectrum. This includes the possibility that brief and acute psychoses form a link between the schizophrenic and the affective spectrum.

185

Overlapping spectra: brief and acute psychoses

In this chapter, data from the Halle Study on Brief and Acute Psychoses (HASBAP) are reported, so far the most systematic investigation of clinical features, course and outcome of brief and acute psychoses. The results of the HASBAP will be reviewed with special regard to the nosological relationship of brief and acute psychoses to the schizophrenic and the affective spectrum. Methods The HASBAP combines three methodological approaches: (1) The prospective approach, studying a consecutively recruited inpatient sample with a diagnosis of “Acute and Transient Psychotic Disorder” as the ICD-10 operational definition for brief and acute psychoses. (2) A case-control design in which every patient of the original index cohort was matched for age and gender with two clinical groups and a non-clinical control group. (3) The longitudinal approach for all three clinical groups, comprising now three waves of follow-up investigations. In the first phase of the HASBAP, we identified all consecutive cases (n42) fulfilling ICD-10 criteria of ATPD (F23), treated as inpatients at the Department of Psychiatry and Psychotherapy at Martin Luther University Halle-Wittenberg, Germany, during a five-year period. The hospital is situated in the city of Halle, Germany, and takes patients from the city as well as from the surrounding communities, which comprise both rural and industrial areas. It can be said that the Halle University Hospital serves a large municipal and suburban area with a non-selective admissions policy. Moreover, ATPD are acute and usually dramatic psychotic states that – considering the German health care system – nearly always lead to inpatient treatment. The sample of the HASBAP can be regarded as representative of a clinical inpatient population with ATPD. With some restrictions, the findings of this study might also be regarded – because of the above-mentioned reasons – as a reasonable approximation of ATPD in general, and not only for inpatients. ICD-10 research criteria were strictly applied. The recruitment procedure has been described previously (Pillmann et al., 2002; Marneros and Pillmann, 2004). Briefly, patients with a clinical discharge diagnosis of ATPD were considered for inclusion in the study. All diagnoses were reviewed on the basis of a checklist incorporating ICD-10 research criteria. Only subjects who fulfilled ICD-10 research criteria of ATPD (WHO, 1993) were included in the study. The control groups were gender and age matched and comprised: (a) patients with an acute episode of “positive” schizophrenia (b) patients with an acute episode of bipolar schizoaffective disorders (c) surgical patients without any mental disorder

186

Frank Pillmann and Andreas Marneros

To avoid confounding by pre-existing chronicity for the schizophrenic control group, only patients with an episode of “positive” schizophrenia were selected. “Positive” schizophrenia (PS) was defined as an episode of schizophrenia with positive symptoms such as hallucinations or delusions (F20.0, F20.2, F20.3); patients with chronic schizophrenia or residual schizophrenia (F20.5) were excluded. To avoid spurious differences resulting from demographic differences between the index patients and controls, we chose a case-control design, in which the control groups were matched for gender and age with the index patients. The procedures described above resulted in a control group that is not representative for patients with schizophrenia in general. However, the design of the study ensures that any differences are not simply caused by demographic bias or pre-existing chronicity. We also recruited a control group of patients with bipolar schizoaffective disorder. We only included bipolar forms of schizoaffective disorder (BSAD) for two reasons. First, the restriction was necessary to obtain a more homogeneous sample because of the heterogeneity of schizoaffective disorder. Second, a bipolar characteristic of symptoms has been suggested to be an important feature in concepts of brief remitting psychoses related to ATPD (Pillmann et al., 2001). Follow-up investigations took place at predetermined times 2.5 1.3 years, 4.9  1.4 years and 7.01.5 years after the index episode, or 8.6 7.8 years, 10.57.3 years and 12.47.3 years after the first episode. Data from at least one point of follow-up could be obtained from 91.7% of the original sample. At the third followup, 78.6% of the original sample or 88.0% of the surviving patients could be examined. For the evaluation of sociobiographical features we used a semi-structured interview already used in earlier studies (Marneros et al., 1991). The level of general functioning during the previous week was assessed using the Global Assessment Scale (GAS) (Endicott et al., 1976). Relapse was defined as the occurrence of a major affective syndrome or of psychotic symptoms leading either to hospitalization or to outpatient treatment, including psychiatric medication and a disruption of daily activities. The ICD-10 diagnoses were assessed with the Schedules for Clinical Assessment in Neuropsychiatry (SCAN) (van Gülick-Bailer et al., 1995). Social disability was measured using the German version of the WHO Disability Assessment Schedule (WHO-DAS, Jung et al., 1989). The DAS evaluates functioning in a variety of social roles by means of a structured interview. Global functioning, functioning in general behavioral domains, and functioning in special roles (e.g., work, household, and marriage) are measured on three separate scales ranging from 0 to 5, with a higher score designating a greater degree of handicap. Good inter-rater reliability has been demonstrated (Jung et al., 1989).

187

Overlapping spectra: brief and acute psychoses

Deficits in psychological function perceived during the interview were evaluated by means of the Psychological Impairments Rating Schedule (WHO-PIRS), a rating instrument also developed by the WHO (Biehl et al., 1989a). The observer rates the patient’s behavior during the interview. Seventy-five items from ten domains are integrated into three scales: a general scale, a subscale to measure the level of activity, and a further subscale measuring involvement in communication behavior. Scores on these scales range from 0 to 5, with a higher score indicating a greater degree of handicap. The reliability and validity of the instrument have been demonstrated (Biehl et al., 1989a; Biehl et al., 1989b). The level of general functioning was assessed using the Global Assessment Scale GAS (Spitzer et al., 1976; Endicott et al., 1976). The GAS is a widely used rating scale for the evaluation of the overall functioning of a subject during a specified period on a continuum from severe psychiatric illness to health. The inter-rater reliability of the scale is well documented (Endicott et al., 1976). The HASBAP also obtained a detailed family history in order to investigate the hereditary component of ATPD. However, the relatives themselves were not investigated because after the reunification of Germany, a significant number of the East German population moved to Western Germany or abroad. Halle is located in Eastern Germany and has lost nearly 70 000 residents. The danger of bias (through investigating mainly “non-mobile” relatives) was too high. Subjects were questioned about psychiatric illness in all first-degree relatives. In most cases, it was possible to make a provisional diagnosis and to differentiate between non-organic psychotic or major affective disorders and other psychiatric disorders, such as alcohol dependency, using ICD-10 criteria as a guideline. Inter-rater reliability estimates for the central outcome variables were calculated in our group. For this purpose, 15 interviews were double coded by two raters (interviewer-observer method). Kappa values for categorical items exceeded 0.80 for all items. For quantitative outcome scales, intra-class correlation coefficients were excellent, ranging from 0.86 (GAS) to 0.94 (DAS global score). Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 9.0. For continuous measures we tested for differences between the diagnostic groups using an analysis of variance (ANOVA) followed by post-hoc comparisons with Scheffé’s procedure. The 2 tests were used for contingency tables of categorical data. Bivariate comparisons of continuous data were performed with two-tailed t-tests. For the comparison of outcome measures at different points of follow-up non-parametric tests for paired samples were calculated. Significance was assumed at p  0.05. All subjects provided written informed consent. The study protocol was approved by the local ethics committee.

188

Frank Pillmann and Andreas Marneros

Results Gender and age at onset

Sociodemographic data and data on the illness course before the index episode are given in Table 10.1. Almost 79% of the patients with ATPD were females (Table 10.1). With regard to gender distribution, there is an interesting difference between ATPD, schizophrenia, and schizoaffective disorder in the general population. From the gender distribution of all inpatients with non-organic psychotic and affective disorders treated in the Halle University Hospital from 1993 to 1997 (Figure 10.1), it is evident that females are over-represented in ATPD. There were no differences regarding age at onset between ATPD (35.8 years) and PS (35.3 years), but both of them differed significantly in this respect from BSAD (Table 10.1). Patients with BSAD became ill younger (28.6 years). The finding that there were no differences between ATPD and PS regarding age at onset of the disorder is at variance with data from epidemiological studies. These show a considerably lower age at first episode for patients with schizophrenia than in the present control group. As discussed below, this finding is perhaps influenced by the selection of the PS group, especially by gender and positive symptomatology. Features of the acute episode

All patients with ATPD had either an abrupt onset (i.e., within 48 hours) of the index episode (42.9%) or an acute one (less than two weeks; 57.1%). An abrupt onset was found only rarely both in PS (11.9%) and in BSAD (9.5%). Patients with ATPD somewhat more frequently showed acute stress within two weeks before onset of the index episode than patients with PS or BSAD, but the difference was slight and not statistically significant (Table 10.2). The symptoms of the full-blown episode are shown in Table 10.3. By definition all patients with ATPD have to be psychotic, presenting hallucinations and/or delusions. The other two groups also present – according to the design of our study – psychotic symptoms. Delusions were generally equally frequent in all groups, but delusions of being influenced (including experiences of passivity) were found to be more frequent in schizophrenia. The frequency of hallucinations was significantly higher in the group of schizophrenic patients than in the other two groups. Importantly, the delusions of the ATPD group in comparison to the other two groups showed significantly more often a turbulent, changeable and dramatic character with rapidly changing topics. Surprisingly, the schizophrenic first-rank symptoms were almost equally represented at a high level in all three psychotic groups.

189

Overlapping spectra: brief and acute psychoses Table 10.1. Gender, age at onset and previous episodes in ATPD and comparison groups.

ATPD n42 n (%)

PS n42 n (%)

BSAD n42 n (%)

Post-hoc comparisonsa p

Gender female male

33 (78.6) 33 (78.6) 33 (78.6) 9 (21.4) 9 (21.4) 9 (21.4) MeanS.D. MeanS.D. MeanS.D. Age at first episode (years) 35.811.1 35.313.9 28.610.8 Number of previous episodes Age at index admission (years)

1.63.0

1.82.0

5.23.4

41.212.5

41.112.4

42.412.0

n.s.b

0.011 ATPDBSAD* ATPDBSAD**  0.001 PSBSAD** n.s.

Notes: ATPDacute and transient psychotic disorders; PSpositive schizophrenia; BSADbipolar schizoaffective disorder. a ANOVA: pair-wise comparisons with significant differences (Scheffé procedure) are indicated; *p0.05, **p0.01, n.s.no significance. b 2

test. Schizophrenia

**

Schizoaffective-bipolar Affective-bipolar Schizoaffective-unipolar

**

Affective-unipolar

***

ATPD 1:2

** 1:1

2:1

4:1

Female : male ratio

Figure 10.1. Gender distribution according to diagnostic category in all non-organic psychotic and nonorganic major affective disorders (ICD-10 F2, F3) during the period of investigation at Martin Luther University 1993–97 (n 1036 inpatients). Significance of deviation from equal distribution: *p0.05, **p0.01, ***p0.001.

One of the most frequent symptoms in ATPD was anxiety, which was significantly more frequent than in schizophrenia or in bipolar schizoaffective disorders. This finding is compatible with the concordance between acute polymorphic psychotic disorder, which is a sub-group of ATPD in ICD-10, and cycloid

190

Frank Pillmann and Andreas Marneros Table 10.2. Characteristics of the index episode.

ATPD n42 n (%) Onset of index episode abrupt (within 48 hours) acute (within 14 days) subacute (more than 14 days) Acute stress within 14 days

PS n42 n (%)

Statistical analysisa

BSAD n42 n (%)

18 (42.9)

5 (11.9)

4 (9.5)

24 (57.1) 0

15 (35.7) 22 (52.4)

19 (45.2) 19 (45.2)

4 (9.5)

0

0

ATPDPS** ATPDBSAD** ATPDPS* ATPDPS** ATPDBSAD** n.s.

Notes: ATPDacute and transient psychotic disorders; PSpositive schizophrenia; BSADbipolar schizoaffective disorder. a 2

test: only pair-wise comparisons with significant differences are indicated; *p  0.05, **p  0.01, n.s.no significance.

psychoses, especially the “anxiety-elation psychosis” of the Wernicke-KleistLeonhard school (Pillmann and Marneros, 1998; Marneros and Pillmann, 2004). More than two-thirds of patients with ATPD experienced quickly changing moods. In almost 29% of the patients, the changes of mood had a bipolar form with an alternation between a depressive and a maniform pole. Although also in BSAD, 21% of patients alternated during the episode between depression and manic symptomatology and vice versa (“mixed states”). The changes in ATPD were more dramatic, more unstable and of shorter duration than in bipolar schizoaffective mixed states. Frequency and types of relapses

As a relapse of illness we defined any psychotic or major affective episode either leading to inpatient treatment or being treated on an outpatient basis. Outpatient episodes, to count as relapse, had to entail interruption of daily activities plus frequent visits to the psychiatrist plus requirement of pharmacological intervention. Episodes determined at follow-up by use of WHO-SCAN were also included. Figure 10.2 shows that relapses during the prospective follow-up period were frequent in both groups and approached 80% of patients. There was no difference regarding the time until the first relapse between ATPD and PS in the total sample or in the subsample of first-episode patients (log-rank test). Figure 10.3 shows how many of the patients who experienced a relapse had a monosyndromal or a polysyndromal course. A monosyndromal course is defined as a course during which only episodes of the same type as the index episode recur. A polysyndromal course must contain at least one other type of episode.

191

Overlapping spectra: brief and acute psychoses Table 10.3. Symptomatology of the index episode.

ATPD n42 n (%)

PS n42 n (%)

BSAD n42 n (%)

Statistical analysisa

Productive psychotic symptoms (hallucinations and delusions) Hallucinations

42 (100)

42 (100)

42 (100)

n.s.

32 (76.2)

39 (92.9)

32 (76.2)

Delusions Delusions of being influenced First-rank symptoms Rapidly changing delusions

41 (97.6) 21 (50.0) 30 (71.4) 20 (47.6)

41 (97.6) 30 (71.4) 36 (85.7) 0

38 (90.5) 23 (54.8) 34 (81.0) 1 (2.4)

Affective disturbance Disturbance of drive and psychomotor disturbances Depressed mood Maniform symptoms Anxiety

42 (100) 36 (85.7)

41 (97.6) 39 (92.9)

42 (100) 41 (97.6)

PSATPD* PSBSAD* n.s. PSATPD* n.s. ATPDPS*** ATPDBSAD*** n.s. n.s.

31 (73.8) 32 (76.2) 32 (76.2)

37 (88.1) 24 (57.1) 22 (52.4)

30 (71.4) 36 (85.7) 22 (52.4)

Rapidly changing mood

29 (69.0)

1 (2.4)

5 (11.9)

Thought disorder Bipolar character of symptoms

36 (85.7) 12 (28.6)

41 (97.6) 1 (2.4)

41 (97.6) 9 (21.4)

n.s. BSADPS** ATPDPS* ATPDBSAD* ATPDPS*** ATPDBSAD*** n.s. ATPDPS*** BSADPS**

Notes: ATPDacute and transient psychotic disorders; PSpositive schizophrenia; BSADbipolar schizoaffective disorder. a Only pair-wise comparisons with significant differences ( 2 test or Fisher’s exact test, twotailed) are shown; *p  0.05, **p 0.01, ***p  0.001, n.s.no significance.

As Figure 10.3 shows, 20.5% of the patients had no further episode and another 28.2% had only ATPD episodes. These 48.8% can be considered to show a monosyndromal course during the prospective observation time. An inspection of the episodes showed that the majority of the patients who relapsed had another ATPD episode. Twelve patients (30.8% of all patients) had one or more affective episodes, five patients (12.8%) had schizoaffective episodes and another five patients (12.8%) had schizophrenic episodes during follow-up. In conclusion, while in somewhat less than half of the patients the course after the index episode is monosyndromal, the majority display a polysyndromal course with affective and schizoaffective episodes. Conversion to schizophrenia, however, is infrequent and concerns 12.8% of the patients.

192

Frank Pillmann and Andreas Marneros Patients without further episodes 1.0

0.8

0.6 Population PS ATPD

0.4

0.2 0.0 0

2

4

6

8

10

Duration of follow-up in years

Figure 10.2. Time to first relapse with psychotic or major affective episode. Kaplan–Meier survival curve.

Relapses of patients with ATPD as index episode during 7-year follow-up 20.5%

No relapse Only ATPD episodes

28.2%

At least one... ...ATPD episode

53.8% 30.8%

...affective episode ...schizoaffective episode

12.8%

...schizophrenic episode

12.8%

0

5

10

n =39 15

20

25

Number of patients Percentages refer to all ATPD patients with follow-up (n =39)

Figure 10.3. Types of episodes from the index episode until the end of the prospective follow-up in all ATPD patients.

The outcome of ATPD

The outcome of ATPD and control groups at the first follow-up is shown in Table 10.4. At this point of follow-up, also the mentally healthy control group was examined, enabling us to make the comparison with a matched group of people without mental disorder. Scores on global functioning (GAS), disability (DAS) and psychological impairment (PIRS) are given.

193

Overlapping spectra: brief and acute psychoses Table 10.4. Outcome of ATPD, clinical control groups and mentally healthy controls two years after the index episode (when the mentally healthy controls were also examined). ATPD MeanS.D.

PS MeanS.D.

BSAD MeanS.D.

CT MeanS.D.

p

Post-hoc comparisonsa

Global Assessment Score (GAS)b

81.415.4

66.220.7

74.312.9

95.84.2

0.001

ATPDPS** CTATPD** CTPS** CTBSAD**

Disability Assessment Schedule (DAS)c

0.821.11

1.741.54

1.160.96

0.170.44

0.006

ATPDPS** CTPS** CTBSAD**

Psychological Impairments Rating Schedule (PIRS)c

0.340.51

0.820.94

0.650.51

0.0260.080 0.009

ATPDPS** CTPS** CTBSAD**

Notes: ATPDacute and transient psychotic disorders; PSpositive schizophrenia; BSADbipolar schizoaffective disorder; CTmentally healthy controls. a ANOVA: pair-wise comparisons with significant differences (Scheffé procedure) are indicated; *p0.05, **p0.01, n.s.no significance. b Possible scores range from 0 to 100, higher score indicating better functioning. c Possible scores range from 0 to 5, higher score indicating greater deficit.

After a mean duration of illness of more than eight years since first episode and more than two years since index episode, significant differences among the groups were found regarding outcome. According to GAS, the ATPD patients had the best prognosis, followed by BSAD patients (see Table 10.4); PS patients had the worst prognosis. The difference between ATPD patients and patients with PS was statistically significant. According to the DAS, the majority of patients with ATPD had significantly better social adjustment than PS. The difference between ATPD and mentally healthy controls was not statistically significant. Interestingly, according to the PIRS, patients with ATPD did not differ significantly regarding psychological deficits from mentally healthy people or from BSAD patients, but showed more impairment than schizophrenic patients. The PS and BSAD patients differed significantly from mentally healthy people but not from each other.

194

Frank Pillmann and Andreas Marneros

Good global functioning at the first follow-up in patients and mentally healthy controls ** ** **

** **

100% 80% 60% 40%

100%

84% 45%

20%

51%

0% ATPD n =38

PS n =38

BSAD n =37 CONTR. n =42

Figure 10.4. Proportion of patients and controls at the first follow-up with good global functioning (GAS 70). **p  0.01.

Another way of visualizing these differences is to calculate the proportion of patients or controls with a good or very good global functioning defined by a GAS score70. These results are shown in Figure 10.4. In conclusion, the outcome of ATPD is also significantly more favorable than that for schizophrenic patients. In some respects, the outcome in ATPD is not significantly different from the status of mentally healthy controls matched for gender and age. Differences in outcome between ATPD and BSAD are less marked. The long-term stability of these differences will be the topic of the next section. Long-term course – differences to schizophrenia

In order to evaluate the longitudinal development of ATPD in relation to schizophrenia at multiple follow-up points, we performed an analysis on all patients for whom data on all three follow-up points were available. The number of patients in the following analysis is therefore somewhat lower than the numbers in the preceding analyses. As a comprehensive outcome measure, global functioning was assessed with the global assessment scale at the different points of follow-up (Figure 10.5). The GAS score in ATPD was 82.1 14.9 at the first follow-up, 82.014.0 at the second and 82.113.8 at the third follow-up (change from first to third follow-up n.s.), that for PS was 65.419.7 at the first, 59.421.3 at the second follow-up and 58.7 28.7 at the third follow-up (significance of difference between first and third follow-up p0.016). The score difference between ATPD and PS at the third follow-up was highly significant (p0.001). In conclusion, there was a significant decrease in global functioning from second to third follow-up for PS patients, while

195

Overlapping spectra: brief and acute psychoses Global Assessment Score (GAS) *** n.s. 100 *

80

1. Follow-up

60

2. Follow-up 40

3. Follow-up

20 0 ATPD n =34

PS n =30

Figure 10.5. Mean values on the Global Assessment Scale. Scores range from 1 to 100, with higher values indicating better global functioning. Within-group comparisons between first and last follow-up evaluated using Wilcoxon statistics. Between-group comparisons of status at third follow-up using Mann-Whitney U-test; *p0.05, ***p0.001, n.s.  no significance.

ATPD patients remained unchanged. Group differences between ATPD and PS patients were highly significant (p0.001). To further differentiate the development of the long-term course we compared the symptomatic status at different points of follow-up. Data assessed with the PANSS were only available for the second and third follow-up. The positive symptom score (Figure 10.6) in ATPD was 7.82.7 at the second and 8.02.5 at the third follow-up (change n.s.); that for PS was 10.75.8 at the first and 11.5 6.5 at the second follow-up (n34, p0.044). The score difference between ATPD and PS at the third follow-up was highly significant (p0.001). The negative symptom score (Figure 10.7) in ATPD was 10.46.4 at the second and 10.76.4 at the third follow-up (change n. s.); that for PS was 17.69.2 at the second and 19.39.0 at the third follow-up (significance for change p0.013). The score difference between ATPD and PS at the third follow-up was highly significant ( p  0.001). In conclusion, there was a significant increase in PANSS positive and negative symptoms from second to third follow-up for PS patients, while ATPD patients remained unchanged. Group differences were highly significant for all measures (p 0.001). Social disability was assessed with the WHO Disability Assessment Schedule (Figure 10.8). The DAS score in ATPD was 0.761.02 at the first follow-up, 0.60 0.96 at the second and 0.741.08 at the third follow-up (change from first to third follow-up n.s.); that for PS was 1.771.46 at the first, 2.101.40 at the second follow-up and 2.231.55 at the third follow-up (significance of difference

196

Frank Pillmann and Andreas Marneros ** *

14 12

n.s.

10 8

2. Follow-up 3. Follow-up

6 4 2 0 ATPD n =34

PS n =30

Figure 10.6. Mean values on the PANSS positive scale. Scores range from 7 to 49 with higher values indicating more severe symptoms. Within-group comparisons between first and last follow-up evaluated using Wilcoxon statistics. Between-group comparisons of status at third follow-up using Mann-Whitney U-test; *p0.05, **p0.01.

PANSS – negative symptoms

*** 25

*

20 n.s. 15

2. Follow-up 3. Follow-up

10 5 0 ATPD n =34

PS n =30

Figure 10.7. Mean values on the PANSS negative scale. Scores range from 7 to 49 with higher values indicating more severe symptoms. Within-group comparisons between first and last follow-up evaluated using Wilcoxon statistics. Between-group comparisons of status at third follow-up using Mann-Whitney U-test; *p0.05, ***p0.001.

between first and third follow-up p0.016). The score difference between ATPD and PS at the third follow-up was highly significant (p0.001). In conclusion there was a significant decrease in global functioning from first to third follow-up for PS patients, while ATPD patients remained unchanged. Group differences between ATPD and PS patients were highly significant for all measures. Summarizing the data presented above, the findings in the control group of the present study confirm a modest degree of deterioration in the long-term course of

197

Overlapping spectra: brief and acute psychoses

Disability Assessment Score (DAS)

*** * 2.5 2.0 1.5

1. Follow-up

n.s.

2. Follow-up

1.0

3. Follow-up

0.5 0.0 ATPD n =34

PS n =30

Figure 10.8. Mean values on the Disability Assessment Scale. Scores range from 0 to 5 with higher values indicating more severe social disability. Within-group comparisons between first and last follow-up evaluated using Wilcoxon statistics. Between-group comparisons of status at third follow-up using Mann-Whitney U-test; *p0.05, ***p0.001.

schizophrenia as evidenced by an increase in symptoms and social disability and a decrease in global functioning from first to last follow-up, although the effect is numerically modest. The same effect was not present in the index group of patients with ATPD, reinforcing the notion of long-term stability in this group. The family history

In the HASBAP, a detailed family history for each subject was obtained using all available information, including the subjects’ own reports and information from the case records. With respect to the limited nature of the information available on the secondary cases, and to avoid “false diagnoses,” broad diagnostic categories were created. The category “psychotic disorders” includes schizophrenia, schizoaffective and other psychotic disorders including brief psychoses (characterized by delusions or hallucinations). The category “major affective disorders” included all major affective (manic or depressive) disorders. The category “psychotic or major affective disorders” included the former two categories and all relatives who probably suffered from a major affective or psychotic disorder, but in whom a differential diagnosis was not possible owing to insufficient data. The category “alcohol abuse” included all relatives with clinically relevant abuse or dependency. The category “any mental disorders” comprised all above-mentioned categories and all other relevant psychiatric disorders. For comparison with other studies, data on morbid risk corrected by the age and number of first-degree relatives were reported according to the method proposed by Weinberg (Faraone et al., 1999). As the risk period, the

198

Frank Pillmann and Andreas Marneros Table 10.5. Mental disorders in the family (number and proportion of affected first-degree relatives; morbid risk calculated with Weinberg correction).

Diagnosis of the index proband ATPD Number of first-degree relatives with sufficient information

PS

BSAD

CONTR.

Statistical analysisa

198

186

209

247

118.5

103.0

127.0

139.5

24 (20.3)

28 (27.2)

22 (17.3)

5 (3.6)

ATPDCONTR.*** PSCONTR.*** BSADCONTR.***

Relatives with psychotic disordersc

4 (3.4)

5 (4.9)

1 (0.8)

1 (0.7)

n.s.

Relatives with major affective disordersc

3 (2.5)

0

6 (4.7)

0

BSADCONTR.* BSADPS*

Relatives with psychotic or major affective disorderc, d

8 (6.8)

8 (7.8)

11 (8.7)

1 (0.7)

ATPDCONTR.* PSCONTR.** BSADCONTR.**

7 (5.9)

6 (5.8)

2 (1.6)

1 (0.7)

Corrected number of relativesb Relatives with any mental disorderc

Relatives with alcohol abusec

ATPDCONTR.* PSCONTR.*

Notes: ATPDacute and transient psychotic disorders; PSpositive schizophrenia; BSADbipolar schizoaffective disorder; CONTR.mentally healthy controls. a Only pair-wise comparisons significant in 2 test or Fisher’s exact test are shown; *p  0.05, **p  0.01, ***p  0.001, n.s.no significance. b Calculated according to the method of Weinberg with an assumed period of risk from age 18–60 years. c n (morbid risk in %). d Includes relatives with psychotic or major affective disorder (no differentiation possible). This number therefore may exceed the sum of relatives with psychotic disorder and relatives with major affective disorder.

age range from 18 to 60 years was defined. The results of this analysis are given in Table 10.5. Rates for “any mental disorder” were increased in all clinical groups in comparison to the mentally healthy controls (the relative morbid risk for ATPD was 5.6, the risk for PS was 7.8, and the risk for BSAD was 4.8). The differences between the clinical groups and the healthy controls were all highly significant. Rates for the broad category of psychotic or major affective disorders were also increased

199

Overlapping spectra: brief and acute psychoses

(relative morbid risk for ATPD was 9.7, the risk for PS was 11.1 and the risk for BSAD was 12.3). The differences between the clinical groups and the healthy controls were all significant. Numbers were too small to give statistically significant differences between patient groups (except for a higher number of affectively ill relatives in BSAD than in PS). Qualitatively, however, it can be seen that in the relatives of probands with ATPD, both psychotic and major affective illnesses occur. Discussion Discussion of the main findings

The HASBAP was designed to answer the question: in which regard is the category “acute and transient psychotic disorders”, created by the ICD-10, different from schizophrenia and from other non-organic psychotic disorders, for example schizoaffective disorders? Therefore, ATPD was compared with schizophrenic conditions presenting with similar symptoms, namely those of “positive schizophrenia”, including hallucinations and delusions. Another comparison was carried out with schizoaffective disorders, especially with bipolar ones, in light of the polymorphous symptomatology and bipolar mood changes in some cases of ATPD (Marneros et al., 2000; Pillmann et al., 2001). The results of this study showed some similarities, but also relevant differences, between ATPD and the other two psychotic groups. A relatively characteristic feature of ATPD seems to be a dramatic and unstable phenomenological picture with rapidly changing delusional topics and rapidly changing mood. Bipolarity of mood states is frequent, showing some similarities to “schizoaffective mixed states” (Marneros and Angst, 2000). In addition, anxiety is one of the symptoms encountered more frequently in the ATPD group than in the other psychotic groups. The polymorphous and conflicting symptomatology of ATPD shows similarities with the so-called “cycloid psychoses” (Perris, 1986; Pillmann et al., 2001). With respect to the temporal course of the episode, not only is the duration of symptoms shorter in ATPD than in the other groups, but also the onset is usually abrupt and, by definition, always acute. Accordingly, the development from first symptoms to full symptomatology is very rapid. Some cases of ATPD are preceded by stressful life-events. So far, some cases of ATPD (but not the majority) have some similarities to the “reactive psychoses” of Scandinavian psychiatry (Strömgren, 1986b; Ungvari and Mullen, 2000). Finally, ATPD occurs more often in females than in males – in contrast to schizophrenia and bipolar schizoaffective disorders (Leung and Chue, 2000). Perhaps this finding is an interesting indication of biological mechanisms underlying the pathogenesis of ATPD, e.g., modifying influences of gonadal hormones (Häfner et al., 1998a). Age at onset in ATPD was considerably higher than in matched controls with BSAD, but did not differentiate between ATPD and the control group with PS.

200

Frank Pillmann and Andreas Marneros

However, it has to be underlined that the group of schizophrenic patients involved in this study was characterized by “positive” symptoms and that, owing to the matching, the great majority were females. Both positive symptomatology and female gender are associated with higher age at onset (Faraone et al., 1994; Häfner et al., 1998a; Leung and Chue, 2000; Riecher-Rössler and Rössler, 1998). When studies relying on large unselected treated samples or on epidemiological samples are considered, the mean age of onset for schizophrenia in general is consistently reported to be lower than in the present sample of ATPD (Marneros et al., 1991; Jablensky et al., 1992; Häfner et al., 1998b). Thus, it can be assumed that the age at onset of ATPD is higher than in unselected patients with schizophrenia, but the difference disappears once gender and positive symptomatology are controlled for. As we have shown elsewhere, some premorbid characteristics such as educational level and the presence of a stable heterosexual relationship differentiate between schizophrenic and ATPD patients, but not between ATPD and schizoaffective patients (Marneros et al., 2003; Marneros et al., 2002). It can be assumed that the premorbid functional level of patients with ATPD is very similar to that of those with schizoaffective disorders. In this respect, ATPD shows similarities to the mentally healthy population but differs significantly from schizophrenia. The outcome of ATPD is also significantly more favorable than that of schizophrenic patients. Differences in outcome between ATPD and BSAD were less marked. As stated above, the majority of schizophrenic patients investigated were females and all of them had so-called positive schizophrenia – two factors both associated with a more favorable outcome (Riecher-Rössler and Rössler, 1998). Nevertheless, outcome in schizophrenia remained significantly less favorable than in the other two psychotic groups. In all three psychotic groups the frequency of mental disorders in first-degree relatives was significantly higher than that in relatives of the mentally healthy group, but there were no statistically significant differences among the relatives of the psychotic probands. In spite of this last finding, it seems that at least on clinical, prognostic and therapeutic grounds ATPD has to be delineated both from schizophrenia and from schizoaffective disorders – although ATPD and schizoaffective disorders have strong similarities regarding premorbid adaptation and outcome. Other factors, including symptomatology, age at onset, and gender distribution differ significantly between ATPD and schizoaffective disorders. We will now consider the consequences of these findings for position of acute and transient psychoses within the psychotic and/or affective spectrum. Are ATPD psychogenic disorders?

The notion that acute and transient psychotic disorders are essentially psychogenic disorders, independent from the schizophrenic and/or affective spectrum, has to be

201

Overlapping spectra: brief and acute psychoses

rejected. Indeed, some episodes of ATPD are preceded by stressful life-events, but only about 10% of ATPD episodes are precipitated by acute stress according to the strict definition of ATPD. Delusions and/or hallucinations are obligatory in ATPD by definition, but also first-rank symptoms are hardly less frequent than in schizophrenia. In the long run, most cases of ATPD run a relapsing course, and major affective or schizoaffective episodes are quite often found during the later course of the illness. Therefore, although psychogenic triggers may play a role in some cases, on the whole ATPD cannot be regarded as entirely psychogenic disorders. The absence of preceding stress in the majority of cases, the typical psychopathological features and the course of the disorder suggest its integration into the affective and/or schizophrenic spectrum. What is the relationship of ATPD to the schizophrenia spectrum?

The obligatory presence of delusions and/or hallucinations and the empirically found frequency of first-rank symptoms indicate the relationship of ATPD to the schizophrenic spectrum. There are, however, significant differences between ATPD and schizophrenia regarding: • Gender distribution: ATPD are disproportionately often found in females. • Premorbid level of functioning and premorbid social interactions: these are much more favorable in ATPD than in schizophrenia. • Onset, development, duration, and phenomenology, as well as structure of symptomatology: ATPD are characterized by rapid onset, short duration and an instable symptomatology including rapidly changing delusions and quickly changing affect, often including severe anxiety. • Lack of deterioration during long-term course in ATPD as opposed to schizophrenia. The differences between ATPD and schizophrenia just mentioned cannot simply be explained by ATPD being a minor or less severe form of schizophrenia. In addition to the sociodemographic and symptomatological differences, the severity of productive symptoms in ATPD seems as high as in schizophrenia. Also relapses are quite frequent in ATPD. Therefore, the features of ATPD cannot be simply explained as an attenuation of schizophrenic symptomatology. The “lower severity” applies only to some features (e.g., long-term deterioration), while in some respects ATPD are just different from schizophrenia. Are ATPD related to the affective (bipolar schizoaffective) spectrum?

ATPD and bipolar schizoaffective disorders have strong similarities regarding: • premorbid level of functioning • premorbid interactions, and • outcome

202

Frank Pillmann and Andreas Marneros

Acute and transient or brief polymorphic psychoses

Schizoaffective disorders

Schizophrenia

Mood disorders

Figure 10.9. Acute and transient/brief polymorphic psychoses and schizoaffective disorders and their relation to the schizophrenic and affective spectrum.

Also, in the longitudinal course of patients with ATPD, affective and schizoaffective episodes occur relatively frequently, showing similarities to the course of schizoaffective disorder. But there are also important differences between ATPD and bipolar schizoaffective disorders: • Gender: the strong preponderance of female patients in ATPD is not found in bipolar disorder. • Age at onset: ATPD on average have a much later onset than BSAD. • Mode of onset: the onset is much more acute in ATPD. • Symptomatology: although some ATPD patients show bipolar features, the rapidly changing affect and rapidly changing delusions seem to be a rather specific phenomenon. Therefore, there seem to be strong relations of ATPD to the affective spectrum, in particular to its bipolar forms. At the same time, there are important differences that preclude the allocation of ATPD exclusively to the affective spectrum. Conclusions Brief and acute psychoses (by the ICD-10 operationalized as ATPD) seem to have a relationship both to the schizophrenic spectrum and to the affective spectrum. They constitute a “bridge” between the schizophrenic and the affective spectrum, in a similar way as schizoaffective disorders (Figure 10.9). In other words, brief and

203

Overlapping spectra: brief and acute psychoses

acute psychoses have to be located on a continuum of affective and psychotic disorders (Marneros et al., 1995; Marneros and Pillmann, 2004; Crow, 1995; 2000; Angst and Marneros, 2001). Their existence strongly suggests that the schizophrenic spectrum and the affective spectrum cannot meaningfully be conceptualized as separate but rather as overlapping spectra. R E F E R E N C ES Akiskal, H. S. (2002). The bipolar spectrum – the shaping of a new paradigm in psychiatry. Current Psychiatry Reports, 4 (1), 1–3. Angst, J. (1966). Zur Ätiologie und Nosologie endogener depressiver Psychosen. Monograph Gesamtgeb Neurol Psychiat, Heft 112. Berlin, Heidelberg, New York: Springer 1966. Angst, J. and Marneros, A. (2001). Bipolarity from ancient to modern times: conception, birth and rebirth. Journal of Affective Disorders, 67 (1–3), 3–19. APA (1980). Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. Washington DC: American Psychiatric Association. Biehl, H., Maurer, K., Jablensky, A., Cooper, J. E. and Tomov, T. (1989a). The WHO Psychological Impairments Rating Schedule (WHO/PIRS). I. Introducing a new instrument for rating observed behaviour and the rationale of the psychological impairment concept. British Journal of Psychiatry. Suppl., 7, S68–70. Biehl, H., Maurer, K., Jung, E. and Krumm, B. (1989b). The WHO Psychological Impairments Rating schedule (WHO/PIRS). II. Impairments in schizophrenics in cross-sectional and longitudinal perspective – the Mannheim experience in two independent samples. British Journal of Psychiatry. Suppl., 7, S71–7. Crow, T. J. (1995). Psychotic continuum or disease entities? The critical impact of nosology on the problem of aetiology. In Psychotic Continuum, ed. A. Marneros, N. C. Andreasen and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 151–63. Crow, T. J. (2000). Bipolar shifts as disorders of the bi-hemispheric integration of language: implications for the genetic origins of the psychotic continuum. In Bipolar Disorders. 100 Years After Manic-depressive Insanity, ed. A. Marneros and J. Angst. Dordrecht, Boston, London: Kluwer, pp. 335–48. Dunner, D. L., Fleiss, J. L. and Fieve, R. R. (1976). The course of development of mania in patients with recurrent depression. American Journal of Psychiatry, 133 (8), 905–8. Endicott, J., Spitzer, R. L., Fleiss, J. L. and Cohen, J. (1976). The global assessment scale. A procedure for measuring overall severity of psychiatric disturbance. Archives of General Psychiatry, 33 (6), 766–71. Faraone, S. V., Chen, W. J., Goldstein, J. M. and Tsuang, M. T. (1994). Gender differences in age at onset of schizophrenia. British Journal of Psychiatry, 164 (5), 625–9. Faraone, S. V., Tsuang, M. T. and Tsuang, D. W. (1999). Genetics of Mental Disorders. New York; London: Guilford Press. Häfner, H., an der Heiden, W., Behrens, S. et al. (1998a). Causes and consequences of the gender difference in age at onset of schizophrenia. Schizophrenia Bulletin, 24 (1), 99–113.

204

Frank Pillmann and Andreas Marneros Häfner, H., Maurer, K., Löffler, W. et al. (1998b). The ABC Schizophrenia Study: a preliminary overview of the results. Social Psychiatry and Psychiatric Epidemiology, 33 (8), 380–6. Hatotani, N. (1996). The concept of “atypical psychoses”: special reference to its development in Japan. Psychiatry and Clinical Neurosciences, 50 (1), 1–10. Hoche, A. (1913). Die Bedeutung der Symptomenkomplexe in der Psychiatrie. Zeitschrift für die gesamte Neurologie und Psychiatrie, 14, 540–51. Jablensky, A., Sartorius, N., Ernberg, G. et al. (1992). Schizophrenia: manifestations, incidence and course in different cultures. A World Health Organization ten-country study. Psychological Medicine. Monograph supplement, 20, 1–97. Jørgensen, P., Bennedsen, B., Christensen, J. and Hyllested, A. (1996). Acute and transient psychotic disorder: comorbidity with personality disorder. Acta Psychiatrica Scandinavica, 94, 460–4. Jørgensen, P., Bennedsen, B., Christensen, J. and Hyllested, A. (1997). Acute and transient psychotic disorder: a 1-year follow-up study. Acta Psychiatrica Scandinavica, 96 (2), 150–4. Jung, E., Krumm, B., Biehl, H., Maurer, K. and Bauer-Schubart, C. (1989). Mannheimer Skala zur Einschätzung sozialer Behinderung, DAS-M. Weinheim: Beltz. Kleist, K. (1924). Über die gegenwärtigen Strömungen in der klinischen Psychiatrie. Allgemeine Zeitschrift für Psychiatrie, 81, 389–93. Kleist, K. (1926). Über cycloide Degenerationspsychosen, besonders Verwirrtheits- und Motilitätspsychosen. Zentralblatt für die gesamte Neurologie und Psychiatrie, 44, 655–7. Kleist, K. (1953). Die Gliederung der neuropsychischen Erkrankungen. Monatsschrift für Psychiatrie und Neurologie, 125, 526–54. Langfeldt, G. (1939). The Schizophreniform States. Kopenhagen: Munksgaard. Leonhard, K. (1961). Cycloid psychoses – endogenous psychoses which are neither schizophrenic nor manic-depressive. Journal of Mental Science, 107, 633–48. Leonhard, K. (1995). Die Aufteilung der endogenen Psychosen und ihre differenzierte Ätiologie, 7th edn. Stuttgart: Thieme. Leung, A. and Chue, P. (2000). Sex differences in schizophrenia; a review of the literature. Acta Psychiatrica Scandinavica. Suppl., 401, S3–38. Maier, W., Lichtermann, D., Minges, J. and Heun, R. (1994). Personality disorders among the relatives of schizophrenia patients. Schizophrenia Bulletin, 20 (3), 481–93. Manschreck, T. C. and Petri, M. (1978). The atypical psychoses. Culture, Medicine and Psychiatry, 2, 233–68. Marneros, A. (2001). Expanding the group of bipolar disorders. Journal of Affective Disorders, 62 (1–2), 39–44. Marneros, A., Andreasen, N. C. and Tsuang, M. T. (eds.) (1995). Psychotic Continuum. Berlin: Springer. Marneros, A. and Angst, J. (2000). Bipolar disorders: roots and evolution. In Bipolar Disorders. 100 years after manic-depressive insanity, ed. A. Marneros and J. Angst. Dordrecht, Boston, London: Kluwer Academic Publishers, pp. 1–35. Marneros, A., Deister, A. and Rohde, A. (1991). Affektive, schizoaffektive und schizophrene Psychosen. (English abstract) Berlin: Springer. Marneros, A. and Pillmann, F. (2004). Acute and Transient Psychoses. Cambridge: Cambridge University Press.

205

Overlapping spectra: brief and acute psychoses Marneros, A., Pillmann, F., Haring, A. and Balzuweit, S. (2000). Die akuten vorübergehenden psychotischen Störungen. Fortschritte der Neurologie-Psychiatrie, 68 Suppl. 1, S22–5. Marneros, A., Pillmann, F., Haring, A., Balzuweit, S. and Blöink, R. (2002). The relation of “acute and transient psychotic disorder” (ICD-10 F23) to bipolar schizoaffective disorder. Journal of Psychiatric Research, 36 (3), 165–71. Marneros, A., Pillmann, F., Balzuweit, S., Blöink, R. and Haring, A. (2003). What is schizophrenic in ATPD? Schizophrenia Bulletin, 29 (2), 311–23. Marneros, A. and Tsuang, M. T. (eds.) (1986). Schizoaffective Psychoses. Berlin: Springer. Perris, C. (1966). A study of bipolar (manic-depressive) and unipolar depressive recurrent depressive psychoses. Acta Psychiatrica Scandinavica, 42 (Suppl. 194), 194. Perris, C. (1974). A study of cycloid psychoses. Acta Psychiatrica Scandinavica. Suppl., 253, 1–77. Perris, C. (1986). The case for the independence of cycloid psychotic disorder from the schizoaffective disorders. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg: Springer, pp. 272–308. Pillmann, F. (2004). Book review: Wimmer, August – Psychogenic Psychoses. Edited and translated with introduction by Johan Schioldann. History of Psychiatry, 15, 501–5. Pillmann, F., Arndt, T., Ehrt, U. et al. (2000). An analysis of Wernicke’s original case records: his contribution to the concept of cycloid psychoses. History of Psychiatry, 11, 355–69. Pillmann, F., Haring, A., Balzuweit, S., Blöink, R. and Marneros A. (2001). Concordance of acute and transient psychotic disorders and cycloid psychoses. Psychopathology, 34 (6), 305–11. Pillmann, F., Haring, A., Balzuweit, S., Blöink, R. and Marneros, A. (2002). The concordance of ICD-10 acute and transient psychosis and DSM-IV brief psychotic disorder. Psychological Medicine, 32, 525–33. Pillmann, F. and Marneros, A. (1998). Acute and transient psychotic disorders: development of concepts. European Psychiatry, 14 (Suppl. 4), S321. Reich, W. (1976). The schizophrenia spectrum: a genetic concept. Journal of Nervous and Mental Disease, 162 (1), 3–12. Remington, G., Menuck, M., Schmidt, P. and Legault, S. (1990). The remitting atypical psychoses: clinical and nosologic considerations. Canadian Journal of Psychiatry, 35, 36–40. Riecher-Rössler, A. and Rössler, W. (1998). The course of schizophrenic psychoses: what do we really know? A selective review from an epidemiological perspective. European Archives of Psychiatry and Clinical Neuroscience, 248 (4), 189–202. Sanchez, E. G. (1987). Schizophreniform disorder: a diagnostic dilemma. Southern Medical Journal, 80 (2), 223–7. Schatzberg, A. F. (2004). Non-schizophrenic psychoses: common and distinguishing features. Journal of Psychiatric Research, 38 (1), 1–2. Schneider, K. (1950). Klinische Psychopathologie. (3. Auflage.) Stuttgart: Thieme. Spitzer, R. L., Gibbon, M. and Endicott, J. (1976). The global assessment scale. Archives of General Psychiatry, 33, 768. Strakowski, S. M. (1994). Diagnostic validity of schizophreniform disorder. American Journal of Psychiatry, 151, 815–24. Strömgren, E. (1986a). Psychogene Psychosen. Nervenarzt, 57, 88–95.

206

Frank Pillmann and Andreas Marneros Strömgren, E. (1986b). Reactive (psychogenic) psychoses and their relations to schizoaffective psychoses. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg, New York: Springer, pp. 260–71. Susser, E. and Wanderling, J. (1994). Epidemiology of nonaffective acute remitting psychosis vs. schizophrenia: sex and sociocultural setting. Archives of General Psychiatry, 51, 294–301. Ungvari, G. S. and Mullen, P. E. (2000). Reactive psychoses revisited. Australian and New Zealand Journal of Psychiatry, 34 (3), 458–67. Vaillant, G. E. (1964). An historical review of the remitting schizophrenias. Journal of Nervous and Mental Disease, 138, 48–56. van Gülick-Bailer, M., Maurer, K. and Häfner, H. (eds.) (1995). Schedules for Clinical Assessment in Neuropsychiatry. Bern: Huber. WHO (1993). The ICD-10 Classification of Mental and Behavioral Disorders: Diagnostic Criteria for Research. Geneva: WHO. Wimmer, A. (1916). Psykogene Sindssygdomsformer. In St. Hans Hospital 1816–1916: Jubilæumsskrift, ed. A. Wimmer. Copenhagen: Gad., pp. 85–216. Winokur, G. (1972). Depression spectrum disease: description and family study. Comprehensive Psychiatry, 13 (1), 3–8.

11

Overlapping of the spectra: physical comorbidity between schizophrenia and affective disorders William S. Stone, Andrea H. Roe and Ming T. Tsuang Massachusetts Mental Health Center Division of Public Sector Psychiatry, Beth Israel Deaconess Medical Center and Harvard Medical School Departments of Psychiatry

Introduction As the theme of this chapter demonstrates, the schizophrenia and affective disorder spectra overlap across multiple dimensions. The nature and extent of this common ground has long been subject to debate, and itself reflects broader questions about the conceptualization of the major psychotic disorders. In this light, overlaps in clinical, epidemiological, cognitive, genetic/neurobiological and other relevant areas serve both to better delineate the specific disorders, and also to further our understanding of the nature of severe mental illness. One area that may shed light on both spectra, and on the nature of the liability to develop them, involves physical comorbid conditions. This chapter focuses on the issue, first by reviewing representative evidence that rates of certain related physical disorders are elevated in both schizophrenia and in affective illness. Poor glucose regulation will be emphasized in this discussion as an example of a physical condition that occurs in both spectra and may be inherent to one or both disorders. While physical disorders may influence many features of psychiatric illness (e.g., their course, their nature or the severity of clinical symptoms), common elements in their potential to modulate liability for schizophrenia and affective illness will be emphasized, along with conceptual and clinical implications of the comorbid relationships. Physical comorbidity in schizophrenia and affective spectrum disorders Associations between physical problems and mental disorders are not new. Harris and Barraclough (1998) noted, for example, that premature deaths in patients The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

207

208

William S. Stone, Andrea H. Roe and Ming T. Tsuang

with mental disorders have been recorded in English vital statistics for 150 years. Similar observations were reported earlier in the last century (e.g., Philips, 1934), and more recently (Koran et al., 1989). Nevertheless, physical problems in major psychiatric conditions have received increased attention in the last several years, for several reasons (Harris and Barraclough, 1998; Brown et al., 2000; Ryan and Thakore, 2002; Marder et al., 2004; Kilbourne et al., 2004). Among these: (1) comorbid physical problems tend to go relatively undetected in these individuals (Koran et al., 1989; Jeste et al., 1996; Subramaniam et al., 2003), and thus untreated; (2) the development of physical problems associated with psychopharmacologic medication has heightened awareness and interest in the topic (Baptista et al., 2001; Newcomer et al., 2002; Marder et al., 2004); and (3) comorbid medical conditions and treatments used to ameliorate them may impact patient outcomes for psychiatric disorders through adverse effects on treatment compliance and tolerability. For each of these reasons and others, excess levels of illness and mortality have long been associated with mental illness. Individuals with schizophrenia, for example, demonstrate 20% greater mortality than the general population (Newman and Bland, 1991). Of particular significance for the present discussion, Brown (1997) showed in a meta-analysis that about 60% of the excess mortality was produced by natural causes, while only 28% were attributable to suicide or accidents. Harris and Barraclough (1998) showed excess mortality in bipolar patients as well, with 46% of the excess attributable to natural causes. Not surprisingly, mortality rates are affected by multiple factors, including gender and type of mood disorder. Osby et al. (2001) showed, for example, that while males and females with bipolar disorder in Sweden were more likely to die from natural causes, only females with unipolar disorder were more likely to die from natural causes. If the only significance of these mortality figures involved the common vulnerability of individuals with schizophrenia and bipolar disorder to receive less treatment for medical conditions, then recognition of that point, in and of itself, would merit attention. It could also lead to better patient outcomes, both for the general health of patients, and for their psychiatric conditions. The issue may be significant for additional reasons, however, that are more related to the nature and etiology of the underlying medical problems. If, for example, physical health problems result primarily from environmental determinants such as a relative lack of detection on the part of health care providers and/or a vulnerability to adopt unhealthy and risky lifestyles on the part of patients, then elevated mortality rates, though important in a practical sense, are less likely to reveal conceptual similarities between the schizophrenic and affective disorder spectra. On the other hand, to the extent that physical problems reflect an inherent relationship to psychiatric conditions, they are more likely to shed light on the nature of the psychiatric disorders themselves,

209

Overlapping spectra: physical comorbidity

including perhaps the vulnerability to develop them and the types of intervention that might attenuate or even prevent their expression. This chapter will emphasize the latter possibility. In this context, there are at least two related reasons to focus on physical problems that occur in both psychiatric spectra. The first involves the central theme of this chapter, which is to further our understanding of schizophrenia and affective illness through an examination of dimensions in which they overlap. The inclusion of common physical vulnerabilities in this discussion has the potential to extend our conceptualization of these psychiatric problems. The second reason relates to the first, but differs in emphasis. A focus on physical problems that occur in both spectra provides a way of validating the potential of particular physical problems themselves to contribute to severe mental illness. From this perspective, the issue is less one of whether the relationship between the physical illness and the psychiatric conditions reflects an inherent relationship (such as a common etiology), but more whether particular physical conditions contribute to the development of schizophrenia or affective illness in vulnerable individuals. We will also focus on several disorders that are related to each other, both to provide a fuller, more coherent picture of physical disorders in the schizophrenia and affective spectra, and because together they comprise a separate entity – the “metabolic syndrome” (Reaven, 1988) – that itself is a significant feature of both spectra. After a brief discussion of physical problems that are associated with both psychiatric spectra, a more detailed consideration of one of them, glucose dysregulation, will follow as a representative example. In general, elevated rates of multiple physical disorders are more the rule than the exception in both conditions. Causes of excess mortality in schizophrenia include, among others, infectious, endocrine (including diabetes), cardiovascular, respiratory and digestive disorders (Harris and Barraclough, 1998). Depression and bipolar disorder are also associated with multiple medical conditions such as cardiovascular disease, endocrine disorders, respiratory disorders, immunological abnormalities and cancer (Harris and Barraclough, 1998; Horrobin and Bennett, 1999; Kilbourne et al., 2004). Cardiovascular disorders

Cardiovascular disease represents a broad class of disorder that is elevated in the schizophrenia and affective disorder spectra. Black and Fisher (1992) demonstrated that up to 20% of excess mortality in patients with DSM-III-R schizophrenia was attributable to cardiovascular disease. Excess mortality from cardiovascular problems has been demonstrated convincingly in both genders (Ryan and Thakore, 2002). These problems likely reflect multiple etiologies, such as ventricular arrhythmias related to antipsychotic medications (Henderson et al., 1991), abnormal heart rate variation (Lovett Doust, 1980), platelet dysfunction (Holt et al., 2004; Kaiya,

210

William S. Stone, Andrea H. Roe and Ming T. Tsuang

1991), sedentary lifestyles and/or other physical problems such as diabetes that are related to the metabolic syndrome (Reaven, 1988, Reaven, 2004). Disorders in the affective disorders spectrum are also associated with cardiovascular disease. Klumpers et al. (2004), for example, reported an increase in cardiovascular risk factors such as hypertension in a cohort of bipolar outpatients. All patients were on lithium monotherapy, but duration of treatment or disorder had no significant effect on the incidence of hypertension. In a meta-analysis of cohort studies, Rugulies (2003) examined the association between depression and myocardial infarction or coronary death and found a significant correlation in seven of eleven studies. In depressed subjects, the relative risk for the onset of coronary heart disease ranged from 1.50 to 4.16, and the aggregate relative risk was 1.64. Wulsin and Singal (2003) also reported an overall relative risk of 1.64 in an independent meta-analysis. Dyslipidemia

Dyslipidemias refer to disorders of lipoprotein metabolism, and are risk factors for coronary heart disease, hypertension, diabetes, obesity and the metabolic syndrome (Davidson, 2002). Their role in the schizophrenia and affective disorder spectra is somewhat unclear. In schizophrenia, elevated cholesterol or triglycerides are associated with antipsychotic treatment (e.g., Heiskanen et al., 2003), but these effects may be related to weight gain or obesity (Ryan and Thakore, 2002; Holt et al., 2004). While the nature or magnitude of dyslipidemia may reflect an inherent vulnerability to the effects of these medications, more research in drug-naive subjects is necessary to determine the role of altered lipid profiles in schizophrenia. In this regard, a recent study with drug-naive, first-episode patients with schizophrenia did not demonstrate altered levels of lipids (Ryan et al., 2003). A recent Finnish study, however, did show higher total cholesterol levels in a schizophrenic group (n 31) that included a small number of subjects who did not receive medication (n5), compared to a healthy control group (Saari et al., 2004). Although the number of subjects involved is too low to draw conclusions, it was notable that the mean cholesterol level and range of levels in the medication-free sub-group was comparable to those of subjects in typical- and atypical-antipsychotic medication sub-groups. Like the schizophrenia spectrum, the affective disorders spectrum also shows a mixed, complicated picture. Elevated triglycerides were one of several abnormalities in subjects with schizoaffective disorder, bipolar type, who met criteria for the metabolic syndrome, compared to schizoaffective subjects who did not meet criteria (Basu et al., 2004). Some studies, however, demonstrated an inverse relationship between serum total cholesterol concentration and depression, with lower total cholesterol levels in patients with depressive symptoms (Morgan et al., 1993;

211

Overlapping spectra: physical comorbidity

Partonen et al., 1999), major depression (Olusi and Fido, 1996; Borgherini et al., 2002) and bipolar disorder (Atmaca et al., 2002). Ghaemi et al. (2000) reported a relationship between lower cholesterol levels in current manic and depressive episodes, but not in mixed episodes. Other studies found no association between cholesterol concentration and depressed mood (Ledochowski et al., 2003; McCallum et al., 1994) or depression (Apter et al., 1999). Observations also vary according to the concentration of high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol types. Some studies report lower levels of HDL cholesterol in patients with major depression (Maes et al., 1997; Chen et al., 2001), while others report the opposite (Olusi and Fido, 1996). Further research is necessary to establish the connection between dyslipidemia and different affective disorders, and also with the same affective disorder in different phases. Obesity

Despite an epidemic of obesity in many parts of the world and numerous environmental factors that predispose individuals with major mental illness to gain weight (e.g., antipsychotic medications, negative symptoms or other illnessrelated causes of inactivity, poor diet and/or low levels of physical exercise), obesity per se is not generally elevated in schizophrenia, although when it is present, it is more likely to occur in women than in men (Allison et al., 1999; Ryan and Thakore, 2002; Holt et al., 2004). Certain types of obesity, however, are expressed more clearly at higher rates. Thakore et al. (2002; Thakore, 2004) showed, for example, that a group of drug-naive patients had a higher body mass index (BMI) than a control group. Although total and subcutaneous body fat did not differ between groups, abdominal fat in the schizophrenic group was over 3.4 times higher than it was in the control group. Elevated levels of abdominal fat are also associated with several mood disorders (McElroy et al., 2004). Among these, Elmslie et al. (2000) showed higher levels in male and female euthymic, bipolar I patients, and Thakore et al. (1997) demonstrated elevated abdominal fat in non-obese women with DSM-III-R major depression. Weber-Hamann et al. (2002) also demonstrated higher visceral fat in women with major depression, but only in those who were hypercortisolemic. Metabolic syndrome. Individually, the medical problems described above, and others such as poor glucose regulation and insulin resistance, are significant. They are also important, however, because they tend to cluster together in individuals and to elevate the risk for other components of the cluster. This group of metabolic abnormalities has been referred to in several ways, such as “syndrome X” and the “insulin resistance syndrome,” in addition to the metabolic syndrome (Reaven, 1988; Reaven, 2004; Sacks, 2004; Holt et al., 2004). Although exact definitions of

212

William S. Stone, Andrea H. Roe and Ming T. Tsuang

the syndrome vary (Ford and Giles, 2003; Sacks, 2004), its components usually include abnormal glucose and/or insulin metabolism, excess weight and especially visceral fat, dyslipidemia and/or cardiovascular disease. Many components of the syndrome share risk factors that may contribute to their development, and to the development of the syndrome, such as dysfunction of the hypothalamic-pituitaryadrenal axis and inflammation (Ryan and Thakore, 2002; Sacks, 2004; Rosmond, 2005). The prevalence of the syndrome depends on the precise diagnostic criteria used to define it, but regardless of which definition is used, it is relatively common in the general population. Data from the Kuopio Ischemic Heart Disease Risk Factor Study in Finland showed, for instance, rates of 8.8% to 14.3% in middleaged men without a recent history of heart disease, cancer, or diabetes (Lakka et al., 2002). The overall prevalence in the United States was higher, at 23.9%, using the National Cholesterol Education Program/Adult Treatment Panel III definition (Ford and Giles, 2003). Both age (e.g., the rate for 20–29 years of age was 7%; the rate for over 60 years of age was over 40%) and ethnicity (e.g., the rate among African-American men was 16.5%; the rate among Mexican-American women was 36.3%) are among the factors that modulate prevalence rates (Ford and Giles, 2003). Like the components of the syndrome discussed above, the metabolic syndrome itself is elevated in the schizophrenia spectra and probably in the affective disorder spectra. Recent studies in patients with schizophrenia showed rates of 51% and 22% in American and Taiwanese samples, respectively (Littrell et al., 2003), and 37% in a Finnish sample (Heiskanen et al., 2003). Basu et al. (2004) reported that 42.4% of a sample of schizoaffective subjects, bipolar subtype, met criteria for the metabolic syndrome. To our knowledge, prevalence rates of the syndrome have not been reported yet for other affective disorders, although the rates are likely to be elevated since rates for the components of the syndrome are elevated. Higher rates of the metabolic syndrome and its components in schizophrenia and affective disorders have significant implications for elevated mortality and morbidity rates in these conditions. They do not, however, explain the nature of the relationship between the medical and psychiatric conditions, or possible implications of medical problems for the disorders themselves. To explore this issue further, another related problem, glucose dysregulation, will be considered next. Glucose dysfunction was selected for this purpose both because it is related to the other medical conditions, including the metabolic syndrome, and because multiple lines of evidence relate glucose function to aspects of function in the psychiatric spectra. The discussion will include a consideration of whether glucose dysregulation might be inherent to the psychiatric spectra, but will not focus on likely underlying etiological factors, such as insulin insensitivity or dysregulation of the hypothalamic-pituitary-adrenal axis. In some instances,

213

Overlapping spectra: physical comorbidity

relationships between glucose function and schizophrenia will be emphasized if corresponding data are not available to assess relationships with affective disorders. Glucose regulation Like the association between mental illness and other forms of physical illness, the association between glucose dysfunction and psychiatric illness is not a recent discovery. Well over a century ago, Henry Maudsley wrote, “Diabetes is a disease which often shows itself in families in which insanity prevails” (Maudsley, 1879). Schizophrenia in particular has a well-documented association with diabetes and with impaired glucose regulation in general. As early as 1919, Kooy (1919) reported a link between hyperglycemia and schizophrenia in ten patients. Since then, the observation of glucose dysregulation in schizophrenia has been repeated many times. Rates of type 2 diabetes typically range from about 15% (Mukherjee et al., 1996; Dixon et al., 2000) to 21% (Subramaniam et al., 2003) in patients with schizophrenia, and are two to four times higher than rates in the general population (Bushe and Holt, 2004). Mukherjee (1996) reported that increasing age was a more important factor for the development of diabetes than the type of antipsychotic medication patients received. Rates of impaired glucose tolerance in the absence of frank diabetes are also elevated in schizophrenia. Ryan et al. (2003), for example, reported impaired glucose tolerance in 15.4% of a first-episode group of subjects compared to 0% in a control group. The finding is particularly significant because the subjects were young and drug-naive. Similar to the findings for schizophrenia, rates of diabetes are elevated in affective disorders as well. Both Lilliker (1980) and Cassidy et al. (1999) showed that the rate of diabetes is about three times higher in hospitalized patients with bipolar disorder than in the general population. Moreover, Eaton et al. (1996) demonstrated that major depression was associated with a 2.23 relative risk for diabetes, and data from the NIMH Collaborative Depression Study showed that firstdegree relatives of patients with major depressive disorder exhibited a 1.87 relative risk for diabetes (Moldin et al., 1993). In a retrospective analysis of psychiatric inpatient records, Regenold et al. (2002) reported a significant elevation in type 2 diabetes rates in patients with bipolar I disorder and especially with schizoaffective disorder. Moreover, the converse finding also occurs: subjects with diabetes are two to three times more likely to express major depression than those without diabetes (Anderson et al., 2001). As with the other medical problems discussed above, elevated rates of glucose dysregulation in mental disorders do not explain the nature of the relationship between the conditions. While these observations are consistent with the possibility that impaired glucose regulation is in some way related to schizophrenia, factors

214

William S. Stone, Andrea H. Roe and Ming T. Tsuang

such as poor diet, substance abuse, stress and/or lack of exercise are common in the schizophrenia and affective disorders spectra, and are likely to contribute to this impairment (Ryan and Thakore, 2002). This point is especially evident in the light of rising rates of obesity and the metabolic syndrome in the general population (Lakka et al., 2002; Ford and Giles, 2003). Further, consistent with the notion that glucose dysregulation occurs secondary to environmental causes, the single largest source of interest in the topic derives from hyperglycemic/diabetic effects associated with several of the newer, atypical antipsychotic medications (Popli et al., 1997; Hagg et al., 1998; Lindenmayer and Patel, 1999; Baptista et al., 2001; Newcomer et al., 2002). In addition to environmental contributory causes, however, there are also reasons to consider endogenous, illness-related etiological factors in these complex, multifactorial disorders. In schizophrenia, for example, some of the earlier observations of impaired glucose regulation were made prior to the development of neuroleptic treatments (e.g., Braceland et al., 1945), and some of the later ones were made in un-medicated patients (Schimmelbusch et al., 1971; Ryan et al., 2003) and in patients treated with typical neuroleptics (e.g., Mukherjee et al., 1989). Interestingly, impaired glucose regulation in schizophrenia is not limited necessarily to an association with the newer pharmacological treatments, even in patients who take these medications. Although there are methodological problems with many of the early studies (e.g., diagnostic definitions were less reliable, and many studies did not report relationships between body weight/adiposity and glycemic control), recent studies with appropriate methodological controls confirm relationships between antipsychotic medications and impaired glucose regulation (Newcomer et al., 2002). These findings raise the possibility that vulnerability to impaired glucose regulation in schizophrenia is related to the underlying disorder, as well as to the medication used to treat it. Consistent with this view, Mukherjee et al. (1989) reported elevated rates of type 2 diabetes in the relatives of patients with schizophrenia, and Wright et al. (1996) demonstrated elevated levels of type 1 diabetes in relatives of patients. Moreover, gestational diabetes during pregnancy is related to the later development of schizophrenia in offspring (Gunnell et al., 2003), and problems in pregnancy more generally increase rates of schizophrenia and impaired glucose regulation in adult offspring (Newsome et al., 2003; Holt et al., 2004). Taken together, these findings provide further evidence that impaired glucose regulation may be related to schizophrenia spectrum disorders – or their precursor conditions – in the absence of psychosis or antipsychotic medications. There is also evidence that prenatal and perinatal complications increase the risk for bipolar disorder (Buka and Fan, 1999), although the association appears to be weaker than it is in schizophrenia (Walker et al., 2002).

215

Overlapping spectra: physical comorbidity

We explored this issue recently from a different perspective, by performing linkage analyses on a set of genes that code for enzymes that are involved in the regulation of glucose metabolism (Stone et al., 2004a). This approach minimizes the effects of environmental variables, such as medications and diet, and raises the issue of whether impaired glucose regulation might be an intrinsic feature of schizophrenia. Data were utilized from the NIMH Genetics Initiative for Schizophrenia data set (Cloninger et al., 1998). A genome scan with 459 markers spaced at an average interval of 10 cM (centiMorgans) was conducted using a linkage analysis program. Data from European- and African-American groups were analyzed separately, with the genome-wide significance of these genes to schizophrenia assessed using permutation testing. When results were adjusted for multiple testing within and across ethnic groups, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase 2 (PFKFB2; chromosome 1q32.2) achieved genome-wide significance (p0.04), and hexokinase 3 (HK3; chromosome 5q35.3) showed evidence suggestive of linkage (p0.09). In the European-American sample, PFKFB2, hexokinase 3 (HK3), and pyruvate kinase 3 (PK3; chromosome 15q23) achieved significance at the 0.05 level. None of the genes showed significance in the African-American sample. These results, while in need of replication, provide converging support for the view that genes that regulate glucose metabolism may also influence susceptibility to schizophrenia. The question of whether glucose dysregulation occurs at elevated rates in relatives of patients with schizophrenia remains to be explored. Notably, Holt (2004) suggested that chromosomal susceptibility regions between schizophrenia and type 2 diabetes might overlap, although the hypothesis remains untested. There are at least two additional reasons to consider an inherent relationship between glucose regulation and the schizophrenia and affective disorders spectra. The first of these involves effects of glucose on dopamine (DA) transmission (Dwyer et al., 2001). In studies with rats, glucose suppressed DA-firing in individual midbrain neurons (Saller and Chiodo, 1980), while DA receptor-binding increased in streptozotocin-induced diabetes (Lozovsky et al., 1981). The pattern of decreased firing and increased number or sensitivity of postsynaptic receptors describes a state of receptor supersensitivity that contributes to poorly regulated and intermittently excessive DA activity. Insulin, however, prevents increases in haloperidol-induced DA receptors (Lozovsky et al., 1985), and to some extent normalizes DA transmission. Levin (2000) confirmed recently that glucose modulates DA release from the substantia nigra through an ATP-sensitive K channel. His data also suggested that low glucose availability (as may occur in chronic hyperglycemia or diabetes) was associated with decreased DA release, while high availability was associated with increased DA release. These findings suggest that impaired glucose regulation may destabilize subcortical DA neurotransmission,

216

William S. Stone, Andrea H. Roe and Ming T. Tsuang

which is consistent with the possibility that glucose may modulate dopamine function in ways that influence the expression of dopamine-related symptoms (e.g., positive symptoms such as psychosis). Another reason to consider an inherent relationship between glucose and schizophrenia involves the response to the administration of glucose. Several studies show that poor glucose regulation is related to poor memory performance in rodents and people (Stone et al., 1990; Convit et al., 2003). In addition, glucose administration improves memory deficits in schizophrenia. Newcomer et al. (1999) and Stone et al. (2003) both found increases in verbal declarative memory performance following glucose administration (compared with saccharin administration) in double-blind crossover designs. Moreover, we showed recently that glucose administration increased brain activation in selective temporal and frontal lobe regions during a verbal encoding task (Stone et al., 2005). These findings indicate that declarative memory performance and several brain regions that mediate it are sensitive to the effects of circulating glucose. It is likely that this sensitivity reflects, at least in part, deficits in glucose regulation and availability at times when they are needed to support task performance. In this view, additional glucose administration facilitates memory by compensating for glucose deficits in brain regions that mediate declarative memory functions (Stone et al., 2004b). Declarative memory is also one of several cognitive functions that are impaired in bipolar disorder (Seidman et al., 2002), but glucose administration has not (yet) been shown to improve declarative memory in this disorder (Newcomer et al., 1999). Implications and conclusions The evidence for comorbid physical disorders in the schizophrenia and the affective disorders spectra is compelling. Moreover, the nature of the physical disorders is significant because it contributes to excess mortality and morbidity in both spectra. Two points are particularly relevant. First, the spectra share vulnerabilities to many of the same comorbid physical conditions. Although the field is only starting to understand the meaning of these associations, the heightened vulnerability to common, life-shortening physical conditions represents a significant dimensional overlap of the schizophrenia and affective disorders spectra. The second point is related to the first and involves the possibility that, at least to some extent, certain physical disorders are inherent concomitants of the psychiatric spectrum conditions. At present, the evidence for this hypothesis is somewhat stronger for schizophrenia than it is for the affective disorders, although it has also been studied more in schizophrenia. The nature of the comorbid relationship remains to be determined, but at least two possibilities seem unlikely. First, as demonstrated by the converging epidemiological, genetic and neurobiological

217

Overlapping spectra: physical comorbidity

evidence relating impaired glucose regulation and schizophrenia, the likelihood that the comorbid conditions are completely separate and occur by chance seems low. Second, the circumscribed nature of the relationship (e.g., not all individuals with schizophrenia or affective disorders show impaired glucose regulation or cardiovascular disease, and most individuals with diabetes do not have spectrum psychiatric conditions) suggests that the physical and psychiatric symptoms are not primarily expressions of the same underlying pathophysiology. In contrast, a third, more likely relationship involves common or overlapping pathological elements between otherwise distinct psychiatric spectrum conditions and physical disorders. As our linkage study indicated (Stone et al., 2004a), this overlap could include genetic components. This view is most consistent with the pattern of epidemiological, clinical and family history data described in this paper. More broadly, it is also consistent with the multifactorial, polygenetic etiology that is likely to underlie most cases in the schizophrenic and affective disorders spectra (Gottesman, 2001), in which the nature and extent of comorbid relationships varies dimensionally between individuals or families. This type of relationship is similar to one proposed by McElroy et al. (2004) to explain the nature of the relationship between obesity and mood disorders. To the extent that this view is correct, several theoretical and clinical implications may be derived. First, physical comorbidities related to the metabolic spectrum or its individual components might contribute to the vulnerability to develop disorders in the schizophrenic and affective disorders spectra. Conceptually, this means that the nature of the liability to develop severe mental disorders extends beyond both clinical psychiatric symptoms and neurobiologically based endophenotypic expressions. Moreover, it identifies another potentially common dimension in the etiology of spectrum disorders. From a treatment standpoint, comorbid physical problems may eventually provide useful targets for early intervention or even prevention strategies. For example, if glucose dysregulation/type 2 diabetes is increased in families with a schizophrenic member and increases the likelihood of impaired dopamine transmission in vulnerable individuals (as discussed above), then interventions to regulate glucose could potentially lower the possibility of expressing subsequent psychosis. A second, related implication of comorbid physical disorders is that they may exacerbate the expression and/or the course of the disorders themselves. As demonstrated most clearly by the discussion on glucose regulation, for example, impaired regulation or availability of glucose may contribute to the presence and severity of cognitive deficits. This point raises the possibility that adjunct interventions to attenuate physical deficits, administered in addition to antipsychotic medications, may be useful in attenuating core cognitive problems in spectrum disorders (Stone et al., 2003; Stone et al., 2005).

218

William S. Stone, Andrea H. Roe and Ming T. Tsuang

Finally, regardless of the nature of the relationship between spectrum disorders and comorbid physical problems, the shared vulnerability of individuals with schizophrenia and affective disorders to develop physical problems, which may be overlooked in the context of their psychiatric problems, highlights the need for increased attention to medical disorders in spectrum conditions. Hopefully, the increased clinical attention to physical problems will further our understanding of this important overlap between the spectra.

R E F E R E N C ES Allison, D., Fontaine, K., Heo, M. et al. (1999). The distribution of body mass index among individuals with and without schizophrenia. Journal of Clinical Psychiatry, 60, 215–20. Anderson, R., Freedland, K., Clouse, R. and Lustman, P. (2001). The prevalence of comorbid depression in adults with diabetes: a meta-analysis. Diabetes Care, 24, 1069–78. Apter, A., Laufer, N., Bar-Sever, M. et al. (1999). Serum cholesterol, suicidal tendencies, impulsivity, aggression, and depression in adolescent psychiatric inpatients. Biological Psychiatry, 46, 532–41. Atmaca, M., Kuloglu, M., Tezcan, E., Ustundag, B. and Bayik, Y. (2002). Serum leptin and cholesterol levels in patients with bipolar disorder. Neuropsychobiology, 46, 176–9. Baptista, T., Kin, N., Beaulieu, S. and de Baptista, E. A. (2001). Obesity and related metabolic abnormalities during antipsychotic drug administration: mechanisms, management and research perspectives, Pharmacopsychiatry, 35, 207–19. Basu, R., Brar, J. S., Chengappa, K. N. R. et al. (2004). The prevalence of the metabolic syndrome in patients with schizoaffective disorder – bipolar subtype. Bipolar Disorders, 6, 314–18. Black, D. W. and Fisher, R. (1992). Mortality in DSM-IIIR schizophrenia. Schizophrenia Research, 7, 109–16. Borgherini, G., Dorz, S., Conforti, D., Scarso, C. and Magni, G. (2002). Serum cholesterol and psychological distress in hospitalized depressed patients. Acta Psychiatrica Scandinavica, 105, 149–52. Braceland, F. J., Meduna, L. J. and Vaichulis, J. A. (1945). Delayed action of insulin in schizophrenia. American Journal of Psychiatry, 102, 108–10. Brown, S. (1997). Excess mortality of schizophrenia. A meta-analysis. British Journal of Psychiatry, 171, 502–8. Brown, S., Inskip, H. and Barraclough, B. (2000). Causes of the excess mortality of schizophrenia. British Journal of Psychiatry, 177, 212–17. Buka, S. L. and Fan, A. P. (1999). Association of prenatal and perinatal complications with subsequent bipolar disorder and schizophrenia. Schizophrenia Research, 39, 113–19. Bushe, C. and Holt, R. (2004). Prevalence of diabetes and impaired glucose tolerance in patients with schizophrenia. British Journal of Psychiatry, 47 (Suppl.), S67–71.

219

Overlapping spectra: physical comorbidity Cassidy, E., O’Halloran, D. and Barry, S. (1999). Insulin as a substance of misuse in a patient with insulin dependent diabetes mellitus. British Medical Journal, 319, 1417–18. Chen, C., Lu, F., Wu, J. and Chang, C. (2001). Correlation between serum lipid concentrations and psychological distress. Psychiatry Research, 102, 153–62. Cloninger, C. R., Kaufmann, C. A., Faraone, S. V. et al. (1998). Genome-wide search for schizophrenia susceptibility loci: The NIMH genetics initiative and millennium consortium. American Journal of Medical Genetics (Neuropsychiatric Genetics), 81, 275–81. Convit, A., Wolf, O. T., Tarshish, C. and de Leon, M. J. (2003). Reduced glucose tolerance is associated with poor memory performance and hippocampal atrophy among normal elderly. Proceedings of the National Academy of Sciences of the United States of America, 100, 2019–22. Davidson, M. (2002). Risk of cardiovascular disease and sudden death in schizophrenia. Journal of Clinical Psychiatry, 63 (Suppl. 9), 5–11. Dixon, L., Weiden, P., Delahanty, J. et al. (2000). Diabetes and schizophrenia. Schizophrenia Bulletin, 26, 903–12. Dwyer, D. S., Bradley, R. J., Kablinger, A. S. and Freeman, A. M., 3rd (2001). Glucose metabolism in relation to schizophrenia and antipsychotic drug treatment. Annals of Clinical Psychiatry, 13, 103–13. Eaton, W., Armenian, H., Gallo, J., Pratt, L. and Ford, D. (1996). Depression and risk for onset of type II diabetes. A prospective population-based study. Diabetes Care, 19, 1097–102. Elmslie, J. L., Silverstone, J. T., Mann, J. I., Williams, S. M. and Romans, S. E. (2000). Prevalence of overweight and obesity in bipolar patients. Journal of Clinical Psychiatry, 61, 179–84. Ford, E. S. and Giles, W. H. (2003). A comparison of the prevalence of the metabolic syndrome using two proposed definitions. Diabetes Care, 26, 575–81. Ghaemi, S., Shields, G., Hegarty, J. and Goodwin, F. (2000). Cholesterol levels in mood disorders: high or low? Bipolar Disorders, 2, 60–4. Gottesman, I. I. (2001). Award for Distinguished Scientific Contributions. American Psychologist, 56, 864–77. Gunnell, D., Rasmussen, F., Fouskakis, D., Tynelius, P. and Harrison, G. (2003). Patterns of fetal and childhood growth and the development of schizophrenia and psychosis in young males: a cohort study. American Journal of Epidemiology, 158, 291–300. Hagg, S., Joelsson, L., Mjorndal, T. et al. (1998). Prevalence of diabetes and impaired glucose tolerance in patients treated with clozapine compared with patients treated with conventional depot neuroleptic medications. Journal of Clinical Psychiatry, 59, 294–9. Harris, E. and Barraclough, B. (1998). Excess mortality of mental disorder. British Journal of Psychiatry, 173, 11–53. Heiskanen, T., Niskanen, L., Lyytikainen, R., Saarinen, P. I. and Hintikka, J. (2003). Metabolic syndrome in patients with schizophrenia. Journal of Clinical Psychiatry, 64, 575–9. Henderson, R. A., Lane, S. and Henry, J. A. (1991). Life-threatening ventricular arrhythmia (torsades de pointes) after haloperidol overdose. Human Experimental Toxicology, 10, 59–62. Holt, R. I. G., Pevelert, R. C. and Byrne, C. D. (2004). Schizophrenia, the metabolic syndrome and diabetes. Diabetic Medicine, 21, 515–23. Horrobin, D. F. and Bennett, C. N. (1999). Depression and bipolar disorder: Relationships to impaired fatty acid and phospholipid metabolism and to diabetes, cardiovascular disease,

220

William S. Stone, Andrea H. Roe and Ming T. Tsuang immunological abnormalities, cancer, ageing and osteoporosis: Possible candidate genes. Prostaglandins, Leukotrienes and Essential Fatty Acids, 60, 217–34. Jeste, D. V., Gladsjo, J. A., Lindamer, L. A. and Lacro, J. P. (1996). Medical comorbidity in schizophrenia. Schizophrenia Bulletin, 22, 413–27. Kaiya, H. (1991). Prostaglandin E1 suppression of platelet aggregation response in schizophrenia. Schizophrenia Research, 5, 67–80. Kilbourne, A., Cornelius, J., Han, X. et al. (2004). Burden of general medical conditions among individuals with bipolar disorder. Bipolar Disorders, 6, 368–73. Klumpers, U. M. H., Boom, K., Janssen, F. M. G., Tulen, J. H. M. and Loonen, A. J. M. (2004). Cardiovascular risk factors in outpatients with bipolar disorder. Pharmacopsychiatry, 37, 211–16. Kooy, F. (1919). Hyperglycaemia in mental disorders. Brain, 42, 214–88. Koran, L. M., Sox, H. C., Marton, K. I. et al. (1989). Medical evaluation of psychiatric patients. I. Result in a state mental health system. Archives of General Psychiatry, 46, 733–40. Lakka, H.-M., Laaksonen, D. E., Lakka, T. A. et al. (2002). The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. Journal of the American Medical Association, 288, 2709–16. Ledochowski, M., Murr, C., Sperner-Unterweger, B., Neurauter, G. and Fuchs, D. (2003). Association between increased serum cholesterol and signs of depressive mood. Clinical Chemistry and Laboratory Medicine, 41, 821–4. Levin, B. E. (2000). Glucose-regulated dopamine release from substantia nigra neurons. Brain Research, 874, 158–64. Lilliker, S. (1980). Prevalence of diabetes in a manic-depressive population. Comprehensive Psychiatry, 21, 270–5. Lindenmayer, J.-P. and Patel, R. (1999). Olanzapine-induced ketoacidosis with diabetes mellitus. American Journal of Psychiatry, 156, 1471. Littrell, K., Petty, R., Hilligoss, N. et al. (2003). Insulin resistance and syndrome X among schizophrenic patients. In Meeting of the American Psychiatry Association. San Francisco, California. Lovett Doust, J. W. (1980). Sinus tachycardia and abnormal cardiac rate variation in schizophrenia. Neuropsychobiology, 6, 305–12. Lozovsky, D. B., Kopin, I. J. and Saller, C. F. (1985). Modulation of dopamine receptor supersensitivity by chronic insulin: implication in schizophrenia. Brain Research, 343, 190–3. Lozovsky, D. B., Saller, C. F. and Kopin, I. J. (1981). Dopamine receptor binding is increased in diabetic rats. Science, 214, 1031–3. Maes, M., Smith, R., Christophe, A. et al. (1997). Lower serum high-density lipoprotein cholesterol (HDL-C) in major depression and in depressed men with serious suicidal attempts: relationship with immune-inflammatory markers. Acta Pscyhiatrica Scandinavica, 95, 212–21. Marder, S. R., Essock, S. M., Miller, A. L. et al. (2004). Physical health monitoring of patients with schizophrenia. American Journal of Psychiatry, 161, 1334–49. Maudsley, H. (1879) The Pathology of Mind. London: Macmillan. McCallum, J., Simons, L., Simons, J. and Friedlander, Y. (1994). Low serum cholesterol is not associated with depression in the elderly: data from an Australian community study. Australian and New Zealand Journal of Medicine, 24, 561–4.

221

Overlapping spectra: physical comorbidity McElroy, S. L., Kotwal, R., Malhotra, S. et al. (2004). Are mood disorders and obesity related? A review for the mental health professional. Journal of Clinical Psychiatry, 65, 634–51. Moldin, S., Scheftner, W., Rice, J. et al. (1993). Association between major depressive disorder and physical illness. Psychological Medicine, 23, 755–61. Morgan, R., Palinkas, L., Barrett-Connor, E. and Wingard, D. (1993). Plasma cholesterol and depressive symptoms in older men. Lancet, 341, 75–9. Mukherjee, S., Decina, P., Bocola, V., Saraceni, F. and Scapicchio, P. (1996). Diabetes mellitus in schizophrenic patients. Comprehensive Psychiatry, 37, 68–73. Mukherjee, S., Schnur, D. B. and Reddy, R. (1989). Family history of type 2 diabetes in schizophrenic patients. Lancet, 1, 495. Newcomer, J. W., Craft, S., Fucetola, R. et al. (1999). Glucose-induced increase in memory performance in patients with schizophrenia. Schizophrenia Bulletin, 25, 321–35. Newcomer, J. W., Haupt, D. W., Fucetola, R. et al. (2002). Abnormalities in glucose regulation during antipsychotic treatment of schizophrenia. Archives of General Psychiatry, 59, 337–45. Newman, S. C. and Bland, R. C. (1991). Mortality in a cohort of patients with schizophrenia: a record linkage study. Canadian Journal of Psychiatry, 36, 239–45. Newsome, C. A., Shiell, A. W., Fall, C. H. et al. (2003). Is birth weight related to later glucose and insulin metabolism? A systematic review. Diabetic Medicine, 20, 339–48. Olusi, S. and Fido, A. (1996). Serum lipid concentrations in patients with major depressive disorder. Biological Psychiatry, 40, 1128–31. Osby, U., Brandt, L., Correia, N., Ekbom, A. and Sparen, P. (2001). Excess mortality in bipolar and unipolar disorder in Sweden. Archives of General Psychiatry, 58, 844–50. Partonen, T., Haukka, J., Virtamo, J., Taylor, P. and Lonnqvist, J. (1999). Association of low serum total cholesterol with major depression and suicide. British Journal of Psychiatry, 175, 259–62. Philips, R. J. (1934). Physical disorder in 164 consecutive admissions to a mental hospital: the incidence and significance. British Medical Journal, 2, 363–6. Popli, A. P., Konicki, P. E., Jurjus, G. J., Fuller, M. A. and Jaskiw, G. E. (1997). Clozapine and associated diabetes mellitus. Journal of Clinical Psychiatry, 58, 108–11. Reaven, G. (1988). Role of insulin resistance in human disease. Diabetes, 37, 1595–607. Reaven, G. (2004). The metabolic syndrome or the insulin resistance syndrome? Different names, different concepts, and different goals. Endocrinology and Metabolism Clinics of North America, 33, 283–303. Regenold, W., Thapar, R., Marano, C., Gavirneni, S. and Kondapavuluru, P. (2002). Increased prevalence of type 2 diabetes mellitus among psychiatric inpatients with bipolar I affective and schizoaffective disorders independent of psychotropic drug use. Journal of Affective Disorders, 70, 19–26. Rosmond, R. (2005). Role of stress in the pathogenesis of the metabolic syndrome. Psychoneuroendocrinology, 30, 1–10. Rugulies, R. (2003). Depression as a predictor for coronary heart disease. A review and metaanalysis. American Journal of Preventive Medicine, 23, 51–61. Ryan, M. C. M., Collins, P. and Thakore, J. H. (2003). Impaired fasting glucose tolerance in firstepisode, drug-naive patients with schizophrenia. American Journal of Psychiatry, 160, 284–9.

222

William S. Stone, Andrea H. Roe and Ming T. Tsuang Ryan, M. C. M. and Thakore, J. H. (2002). Physical consequences of schizophrenia and its treatment: the metabolic syndrome. Life Sciences, 71, 239–57. Saari, K., Jokelainen, J., Veijola, J. et al. (2004). Serum lipids in schizophrenia and other functional psychoses: a general population northern Finland 1966 birth cohort survey. Acta Psychiatrica Scandinavica, 110, 279–85. Sacks, F. M. (2004). Metabolic syndrome: epidemiology and consequences. Journal of Clinical Psychiatry, 65 (Suppl. 18), 3–12. Saller, C. F. and Chiodo, L. A. (1980). Glucose suppresses basal firing and haloperidol-induced increases in the firing rate of central dopaminergic neurons. Science, 210, 1269–71. Schimmelbusch, W., Mueller, P. and Sheps, J. (1971). The positive correlation between insulin resistance and duration of hospitalization in untreated schizophrenia British Journal of Psychiatry, 118, 429–36. Seidman, L. J., Kremen, W. S., Koren, D. et al. (2002). A comparative profile analysis of neuropsychological functioning in patients with schizophrenia and bipolar psychoses. Schizophrenia Research, 53, 31–44. Stone, W. S., Faraone, S. V., Su, J. et al. (2004a). Evidence for linkage between regulatory enzymes in glycolysis and schizophrenia in a multiple sample. American Journal of Medical Genetics – Part B: Neuropsychiatric Genetics, 127B, 5–10. Stone, W. S., Glatt, S. J. and Faraone, S. V. (2004b). The biology of schizotaxia. In Early Clinical Intervention and Prevention of Schizophrenia, eds. W. S. Stone, S. V. Faraone and M. T. Tsuang. Totowa, New Jersey: Humana Press, pp. 339–53. Stone, W. S., Seidman, L. J., Wojcik, J. D. and Green, A. I. (2003). Glucose effects on cognition in schizophrenia. Schizophrenia Research, 62, 93–103. Stone, W. S., Thermenos, H. W., Tarbox, S. I., Poldrack, R. A. and Seidman, L. J. (2005). Medial temporal and prefrontal lobe activation during verbal encoding following glucose ingestion in schizophrenia: A pilot fMRI study. Neurobiology of Learning and Memory, 83, 54–64. Stone, W. S., Wenk, G. L., Olton, D. S. and Gold, P. E. (1990). Poor blood-glucose regulation predicts sleep and memory deficits in normal aged rats. Journal of Gerontology: Biological Sciences, 45, B169–173. Subramaniam, M., Chong, S. A. and Pek, E. (2003). Diabetes mellitus and impaired glucose tolerance in patients with schizophrenia. Canadian Journal of Psychiatry, 48, 345–7. Thakore, J. (2004). Metabolic disturbance in first-episode schizophrenia. British Journal of Psychiatry, 47 (Suppl.), S76–9. Thakore, J. H., Mann, J. N., Vlahoos, J., Martin, A. and Reznek, R. (2002). Increased visceral fat distribution in drug-naive and drug-free patients with schizophrenia. International Journal of Obesity Related Metabolic Disorders, 26, 137–41. Thakore, J. H., Richards, P. J., Reznek, R. H., Martin, A. and Dinan, T. G. (1997). Increased intraabdominal fat deposition in patients with major depressive illness as measured by computed tomography. Biological Psychiatry, 41, 1140–2. Walker, J., Curtis, V. and Murray, R. M. (2002). Schizophrenia and bipolar disorder: similarities in pathogenic mechanisms but differences in neurodevelopment. International Clinical Psychopharmacology, 17 (Suppl. 3), S11–19.

223

Overlapping spectra: physical comorbidity Weber-Hamann, B., Hentschel, F., Kniest, A. et al. (2002). Hypercortisolemic depression is associated with increased intra-abdominal fat. Psychosomatic Medicine, 64, 274–7. Wright, P., Sham, P. C., Gilvarry, C. M. et al. (1996). Autoimmune diseases in the pedigrees of schizophrenic and control subjects. Schizophrenia Research, 20, 261–7. Wulsin, L. R. and Singal, B. M. (2003). Do depressive symptoms increase the risk for the onset of coronary disease? A systematic quantitative review. Psychosomatic Medicine, 65, 201–10.

12

The overlapping of the spectra: suicide Simavi Vahip Affective Disorders Unit, Ege University Medicine Faculty, Department of Psychiatry, Izmir

Introduction The overlap between mood disorders and schizophrenia is a focus of interest in many aspects. Phenomenological and biological issues, including genetic and treatment issues, are the main areas of research in this overlap. One of the most important issues that might have potential clues to help us understand the secrets of the overlap is suicide. It is well known that suicide is a common and dramatic outcome both in mood disorders and in schizophrenia. The risk for suicide in patients with mood disorders has been reported as significantly higher than in patients with other psychiatric disorders (Henriksson et al., 1993; Mortensen et al., 2000). Rates range from 30 to 70% for mood disorders among all suicides (Goodwin and Jamison, 1990; Tondo et al., 2003). Longitudinal follow-up studies of patients with mood disorders have given more dramatic results despite inconsistencies of the reported rates. There are analyses in literature which resulted in different directions: more suicidal risk for unipolar depressives compared to bipolar disorder patients (Harris and Barraclough, 1997) or vice versa (Rihmer and Kiss, 2002). Beyond the inconsistencies in rates, definitions of suicidal behavior, and differentiation of subtypes of mood disorders, it is obvious that suicidal behavior is the most important feature of these disorders. About 19% risk for committed suicide in bipolar disorder patients (Goodwin and Jamison, 1990; Tondo et al., 2003) and almost similar rates for unipolar depressive patients are enough to indicate the importance of suicide in mood disorders. On the other hand, there is no doubt that suicide is higher in patients with schizophrenia than in the general population (Siris, 2001). Most of the studies indicate suicide rates of about 10% for schizophrenic patients (Tsuang 1978; Drake and Ehrlich, 1985; Nyman and Jonsson, 1986; Westermeyer et al., 1991; Meltzer and The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

224

225

Overlapping spectra: suicide

Okayli, 1995; Stephens et al., 1999; Siris 2001). The risk for suicide is also high in schizoaffective disorder, which can be considered as part of a continuum or an overlapping category between mood and schizophrenic disorders (Marneros et al., 1988; Angst et al., 1990). Owing to limitations of the descriptive approach in psychiatry and frequent changes in classifications and diagnostic criteria, the diagnosis of schizoaffective disorder has not received as much interest as it deserves. I believe that evaluation of the suicide issue in schizoaffective disorder from a comprehensive perspective of epidemiology, demographic and clinical correlates, biological findings, and treatment issues, taking into consideration both psychopharmacological and psychosocial aspects, will give us a more complete picture of our current understanding and help us catch some clues for future research on the overlap of mood disorders and schizophrenic disorders. Epidemiology of suicide in the overlapping spectra Suicide in mood disorders

Many reviews of psychological autopsy studies showed that mood disorders, mainly depressive episodes, were the most frequent diagnoses at the time of deaths of suicide victims (Henriksson et al., 1993; Mortensen et al., 2000; Rihmer and Kiss, 2002; Angst et al., 2002; Balázs et al., 2003). In their first review of 17 studies (14 follow-up, 2 population survey and 1 family study), Guze and Robins (1970) reported 12–19% suicide rates for depression. They estimated that by the time all patients had died, about 15% would have committed suicide. This rate was at least 30 times greater than the general population. After two decades, Goodwin and Jamison (1990) extended the review (30 studies) and reported a result of a 19% (range 9–60%) completed suicide rate for depressive patients. There are findings showing a hierarchy in lifetime suicide rates of depressed patients: 9% in patients who were hospitalized for a suicidal tendency, 4% in hospitalized depressed patients without specification of a suicidal tendency, 2.2% in mixed inpatient/outpatient depressives, and 0.5% for non-affective patients (Bostwick and Pankratz, 2000). The unipolar–bipolar distinction was not taken into consideration in many studies and reviews. In a recent meta-analysis, the risk for suicide was analyzed separately in unipolars and in bipolars. From 23 reports, including more than 8000 patients, authors found a 20-fold increased risk for suicide in unipolar patients, while they found a 15-fold risk in bipolars (14 reports in more than 3700 patients) (Harris and Barraclough, 1997). However, Rihmer and Kiss (2002) did not consider these results precise enough for a conclusion, mainly because bipolar II diagnosis was not taken into account separately and had probably been included in the unipolar group. They reviewed the only six studies in the literature which analyzed

226

Simavi Vahip

bipolar I and bipolar II patients separately. Their article reported contrary results: lifetime prevalence of suicide attempts for unipolar, bipolar I and bipolar II patients were, respectively, 12%, 17%, and 24%. There are very few studies considering these three diagnoses separately among depressed suicide victims. Rihmer and colleagues’ previous study (1990) reported 53% first-episode depression or recurrent unipolar depression, 46% bipolar II depressions, and 1% bipolar I depressions in their sample of 100 deaths. Suicide in schizophrenia

Before looking at data on suicide in schizoaffective disorder, a look at suicide in schizophrenia is essential. Although there are reported rates of suicide as low as 2% (Black et al., 1985; Stephens et al., 1999), it is widely reported between 4–13% (around 10% in general) for patients with schizophrenia (Tsuang, 1978; Drake et al., 1984; Nyman and Jonsson, 1986; Westermeyer et al., 1991; Meltzer and Okayli, 1995; Stephens et al., 1999; Siris, 2001). Beyond inconsistencies about the rate, there is no doubt that suicide is higher in patients with schizophrenia than in the general population (Siris, 2001). There is at least one possible factor responsible for these inconsistencies. Despite a certain ratio of schizoaffective patients in most of the studies (Altamura et al., 2003), it is interesting that there seems to be no hesitation to generalize the results as consequences of schizophrenia. This has caused confounding of the data. Suicide in schizoaffective disorder

It is not easy to generalize on the suicide rate for schizoaffective patients because of changes in the disorder’s definition over time and because of insufficient studies. The classification and the diagnostic criteria that were used need to be considered in any evaluation of schizoaffective disorder. There is a very limited number of studies that directly addressed the suicidal behavior in schizoaffective disorder. Many studies included schizoaffective patients in their schizophrenic sample and they did not even give separate data for them. However, Marneros and colleagues reported suicidal symptomatology in 65% of their 72 schizoaffective patients, in 24% of all their episodes, which were mainly schizodepressive (Marneros et al., 1988). A study (Radomsky et al., 1999) conducted with 1048 consecutively admitted psychiatric inpatients reported a 30.2% lifetime history of suicide attempts and 7.2% suicide attempts in the previous month. Interestingly, 42.8% of schizoaffective patients showed lifetime suicide attempts, while an additional 27.0% had lifetime suicidal ideation only. Among all psychotic patients, two diagnoses were associated with the highest recent and lifetime suicidal behavior: major depression with psychotic features and schizoaffective disorder. Considering

227

Overlapping spectra: suicide

ideation and attempt together as suicidal behavior, patients with schizoaffective disorder had the second highest rate of recent (within one month, 53.4%) and lifetime (69.8%) suicidal behavior among five different psychotic diagnoses (major depression – psychotic, schizoaffective, schizophrenia, bipolar and other psychotic disorders). In schizoaffective disorder patients, suicidal behavior rates were lower than in psychotic major depressed patients but slightly higher than in psychotic bipolars, and markedly higher than in schizophrenic patients, and also higher than in other psychotic disorder patients. Another striking result was higher rates of medical dangerousness of the most recent suicide attempts in patients with schizophrenia spectrum psychoses. The number of studies that directly addressed suicide in schizoaffective disorders is very limited and most of the data are derived from schizophrenia and psychosis studies, which generally include schizoaffective patients in their samples. One of the most important studies which included a substantial number of schizoaffective patients in its sample is the InterSePT study (Potkin et al., 2003). Out of a total of 980 patients, 371 were schizoaffective disorder patients. This was a prospective, two-year industry-sponsored study, comparing efficacy of clozapine and olanzapine on suicide risk. Patients from 67 centers in eleven countries were recruited. The main selection criterion was a high risk for suicidality: attempt or hospitalization to prevent suicide within the previous three years, or moderate to severe suicidal ideation within one week before baseline. As a result of selecting high-risk patients for suicide, 83% of all (schizophrenia and schizoaffective) patients had a lifetime history of attempted suicide at least once and 84% had a hospitalization to prevent a suicide attempt. The results indicated that a diagnosis of schizoaffective disorder increases the risk of having a suicide-related event by 41.7% relative to a diagnosis of schizophrenia. Additionally, it was found that the greater the baseline levels of depression and anxiety symptoms, the greater the risk for subsequent suicidal behavior (Potkin et al., 2003). Fenton and colleagues (1997) retrospectively re-diagnosed the Chestnut Lodge Follow-Up Study sample according to DSM-III criteria, and evaluated outcome data that were collected from subjects and/or significant others for an average of 19 years (range: 6–32), blind to diagnoses. They found no significant difference in rates of any suicidal behavior (completed, attempted or ideation) among different diagnostic groups of schizophrenia spectrum disorders in which they considered schizophrenia, schizoaffective disorder, schizophreniform disorder, and schizotypal personality. Completed suicide, suicide attempts, and suicidal ideation for schizoaffective patients diagnosed according to DSM-III were found, respectively, to be 9%, 24%, and 42%. A long follow-up study carried out by Angst and colleagues (1990) questioned two important controversial issues: (1) Are committed suicides constant or do they

228

Simavi Vahip

decrease or increase with age? (2) Are they linked to or independent from diagnosis? Data was collected from a group of hospitalized patients from the Universities of Zurich and Bonn, where the patients were followed-up for many years. Only for the Zurich group of patients, rates of committed suicides among all deaths for unipolar (UP) depressives, bipolar (BP) depressives, and schizoaffective patients were, respectively, 20%, 12%, and 19%. Large numbers of patients (133 UP, 110 BPdepressives and 145 schizoaffectives) and a long follow-up period which allowed the death of almost half of the patients (64% of UP, 54% of BP and 47% of schizoaffective patients) are the strong sides of the study. The Zurich criteria were used as diagnostic criteria for schizoaffective disorders: 502 schizophrenic patients, who were followed-up for a long period of time in Bonn, were re-diagnosed according to the Zurich criteria to identify schizoaffective patients and were added to the first group to find answers to the questions mentioned above. The authors concluded that the risk of suicide over a lifetime is constant and independent from diagnosis (affective, schizophrenic, schizoaffective and borderline). Additionally, when deaths because of other reasons were compared with deaths because of suicide, the only difference was found in the final GAS scores (poorer functioning) in the schizoaffective group. When compared to affective patients, schizoaffective patients showed many similarities related to the course. Henriksson and colleagues (1993) investigated the prevalence rates of mental disorders among a random sample of suicide victims from a nationwide suicide population in Finland and reported depressive disorders as the most prevalent, with a ratio of 59%, while the ratio for schizoaffective disorders was 3%. The method of retrospective psychological autopsy has some limitations by its nature. Retrospective determination of diagnoses, mainly with the information from the source of significant others may be biased to underestimate some diagnoses which have uncertainties in their definition and borders. This may as well be true for schizoaffective disorder. The in-patient population shows relatively severe symptomatology and is at high risk for suicide. There are many studies that investigated rates of different diagnoses and possible risk factors in in-patient populations (Modestin et al., 1992; Krupinski et al., 1998; Deisenhammer et al., 2000; Spießl et al., 2002). Of 30 committed suicides, 6.7% had the diagnosis of schizoaffective disorder (Spießl et al., 2002). However, in many of the studies, data related to patients with schizoaffective disorder were not differentiated. For example, in a study with 44 inpatients who committed suicides, 27.3% were diagnosed as schizophrenic. This sample included an undefined portion of schizoaffective patients (Deisenhammer et al., 2000). A study comparing suicides and serious suicide attempts found that they formed two overlapping populations, far more alike than different. Interestingly, the author used DSM-III-R and reported a higher risk of death by

229

Overlapping spectra: suicide

Factors for general population

Factors related to psychotic and/or chronic mental disorders

Factors related to depression

Suicidal behavior Figure 12.1. Main sources of suicidal behavior in schizoaffective patients.

suicide in males with non-affective psychosis, despite inclusion of schizoaffectives in the group (Beautrais, 2001). Harkavy-Friedman and colleagues (2004) found that both history of major depressive episode and diagnosis of schizoaffective disorder were significantly more in attempters compared to non-attempters in their study investigating the role of depression on suicidal behavior in schizophrenia and schizoaffective disorder. More interestingly, when they removed schizoaffective patients and examined the pure schizophrenic patients, the significance of past major depression between attempters and non-attempters disappeared. Beyond the differences in the definitions and criteria used in different studies, it is clearly obvious that suicidal behaviors are very frequent in schizoaffective disorders. Possible risk factors for suicide in the overlapping spectra Data are very limited on possible risk factors for patients with schizoaffective disorder. Regarding the unestablished definition and borders of the diagnosis, we are far from the conclusive definitions of risk factors specific to schizoaffective disorders. However, its chronicity, phenomenological components (psychotic and affective) and related characteristics, by nature, give us the opportunity to approach the issue (Figure 12.1) and to borrow data from suicide studies conducted with mood and schizophrenic disorder patients. Factors reported to be associated with suicidal behavior in schizophrenia and schizoaffective disorder are shown in Table 12.1. There is no doubt that depressive syndromes and/or symptoms have a great impact on suicidal behavior both in patients with schizophrenia and in patients with schizoaffective disorder. In schizophrenia literature, several different sources of depression or depression-like symptoms were reported (Siris, 2001). Biological diathesis is the prominent one, which has great impact on the diagnosis of schizoaffective disorder (Siris, 2000; 2001). The biological diathesis will be discussed in a separate section. A second possible source for depression and/or depressive symptoms and suicidal behavior in schizophrenia is certain psychological reactions. Siris (2001)

230

Simavi Vahip Table 12.1. Some of the reported correlates of suicidal behavior in schizophrenia/schizoaffective disorder.

Demographic factors Younger age ( 45 years of age) Males  females (committed suicide) Whites  blacks (?) Never married/divorced/widowed Psychosocial factors Poorer premorbid functioning (via impulsiveness or poor problem-solving skills) Higher premorbid functioning (via experience of loss) Dissatisfaction with quality of life Stressful life events Social isolation, lack of social support, recent loss of social support Increased insight for the disorder and its possible consequences Unemployment/part-time employment Diagnosis- (illness-) related factors Age proximity to the onset of illness (especially first ten years) Depression (syndrome/symptoms) Comorbid substance-use disorder Previous suicide attempts Suicidal command hallucinations Paranoid subtype/delusions In-patients Frequent exacerbations and hospitalizations Higher rates of positive symptoms Lower rates and severity of negative symptoms High levels of psychopathology and functional impairment after discharge Less impairment of cognitive functions Long period of untreated psychosis Comorbid personality disorders Longer duration of psychotic symptoms Panic symptomatology Time-related factors Age proximity to the onset (especially first ten years) Throughout lifetime Periods after acute exacerbations After discharge, and six months after hospitalization

231

Overlapping spectra: suicide

described disappointment reactions of two types – acute and chronic, which occur when a loss or frustration intervenes in life to spoil an expectation. As it is well known from the general suicide literature that life events (Rich et al., 1991; Heilä et al., 1999) and particularly losses (Heilä et al., 1999; Saarinen et al., 1999) may have consequences such as depression and/or suicide in schizophrenia. Most probably the risk is even greater in schizoaffective disorder patients. Regarding chronic reactions, many studies reported important sources of depression and/or depressive symptoms and/or suicidal behavior. Drake and colleagues (1984) emphasized the importance of younger age and high self-expectations along with a high premorbid functioning as possible risk factors for certain psychological reactions. Severe disappointment over failed expectations (Westermeyer et al., 1991), pervasive distress, loneliness and dissatisfaction (Cohen et al., 1990), despair and dissatisfaction with the treatment results (Virkunnen, 1976), hopelessness and awareness of the gradual deterioration of abilities (Nyman and Jonsson, 1986) are some of the related factors. However, some reports suggested the contrary, such as correlating poor premorbid functioning and increased suicide risk (Modestin et al., 1992). Two concepts have been noted as possible links between schizophrenia (and also possibly schizoaffective disorder) and suicide in this context: hopelessness and loss of control. Hopelessness, either as an important part of a depressive syndrome or seldom as a consequence of multiple psychosocial events and processes, is highly correlated with suicidal behavior both in mood disorders and in schizophrenic disorders. Many authors mentioned the importance of hopelessness in predisposed patients with various psychiatric disorders to suicidal behavior (Beck et al., 1975; Westermeyer et al., 1991; Meltzer and Okayli, 1995; Siris, 2001; Kim et al., 2003). Loss of control is also conceptually related to hopelessness and has been noted to be a core feature of demoralization (Birchwood et al., 1993; Siris, 2001). One of very few studies investigating a substantial number of schizoaffective patients (n  371) and schizophrenic patients (n  609) gave limited information about separate risk factors for schizoaffective patients. Considering that there was no significant difference between the two diagnostic groups in many ways (such as mean baseline awareness levels), patients were put together and results were given for all. One of the main findings was related to awareness of disorder and its contribution to depression and suicide-related events (Bourgeois et al., 2004). Two controversial findings were reported in this study: first, awareness of a psychiatric disorder (condition) at baseline was associated with increased risk of suicide events over the two-year follow-up period; and second, changes in awareness, which was associated with treatment, decreased the risk of suicide. Depression and hopelessness levels were reported as mediators of the first effect. Authors interpreted the

232

Simavi Vahip

second result by mentioning the complexity of interactions among awareness, depression, hopelessness and suicide (Bourgeois et al., 2004). The role of insight on suicidal behaviors in psychotic patients including schizoaffectives has been investigated in many studies. Most of the studies suggest insight as a possible risk factor leading to suicidal behavior, possibly through hopelessness and depression (Kuo et al., 2004) both in pure schizophrenic patients (Kim et al., 2003) and in the schizophrenic/schizoaffective mixed group (Schwartz and Smith, 2004). However, there are also contradictory findings (Yen et al., 2002). Pini and colleagues (2001) compared insight with illness measured by a highly detailed instrument (The Scale to Assess Unawareness of Mental Disorders) among four psychotic sub-groups and reported interesting findings: poorer insight in schizophrenia patients than patients with schizoaffective disorder and patients with unipolar depression, but insight did not differ from patients with bipolar disorder. Both insight to retrospective awareness of past mental disorder and insight to retrospective awareness of response to medications were significantly poorer in schizophrenia patients than schizoaffective disorder patients. In this regard, insight and related events as potential mediators leading to suicidal behavior may have special importance for schizoaffective patients. No matter how it is named and defined, possible pathways from insight to suicide explain very similar processes. Demoralization theory is one of them, which hypothesizes that as patients become more acutely aware of psychotic symptoms and associated symptomatology, demoralization and fears about their future lives may develop (Drake et al., 2004; Schwartz and Smith, 2004). Schwartz (2001) defined this complex interaction as insight-demoralization-depression-suicidality syndrome. Another possible source for depression and/or suicidal behavior is medication. Extrapyramidal side effects of neuroleptics such as akathisia have been reported to lead to suicidal behavior (Shear et al., 1983; Drake and Ehrlich, 1985; Atbas¸og˘lu et al., 2001). With the description of akinesia occurring either with or without the large muscle stiffness component as a basal-ganglia related decrement in the ability to initiate or sustain motor behaviors by Rifkin and colleagues (1975) and by Van Putten and May (1978), akinesia had also been accepted as an important side-effect of neuroleptics, which is related to depressed mood. The link was described as a lack of experiences (indistinguishable from anhedonia), lead by reduced ability to initiate and sustain motor behaviors and self-recrimination about laziness (an equivalent of reduced self-esteem or guilt) which lead to the diagnosis of depression (Siris, 2001). However, the discussion of whether the result is only a phenocopy of depression or actual depression is ongoing and the general tendency is to support the first hypothesis (Van Putten and May, 1978; Bermanzohn and Siris, 1992; Siris, 2001).

233

Overlapping spectra: suicide

Is there a depression caused by neuroleptics beyond the involvement of akathisia and akinesia? It is not easy to differentiate depression from akathisia and akinesia and give a definite answer (Bermanzohn and Siris, 1992; Siris, 2001). However, considering the relationship between the dopaminergic system and pleasure, dopaminergic blockage by neuroleptics can hypothetically induce depression. There are findings both supporting and rejecting this hypothesis (Siris, 2001). As the schizoaffective patients (at least the depressive subtype) may probably have a particular biological diathesis to depression (Levitt and Tsuang, 1988; Coryell et al., 1990; Taylor, 1992), this hypothetical neuroleptic-induced depression can be more relevant, prevalent and important for this group. As an analogy, this can be similar to a higher risk for tardive dyskinesia of mood disorder patients compared to schizophrenic patients (Casey, 1988; Dog˘aner et al., 1996). There are some other factors which have been reported to be associated with suicidal behavior in schizophrenia and/or schizoaffective disorder patients (Table 12.1). Most frequently reported ones are presence of psychosis, particularly paranoid delusions; comorbid substance, particularly alcohol use disorders; comorbid panic attacks or disorder; and a series of psychosocial factors such as lack of social support and dissatisfaction of quality of life (Siris, 2001). With regard to possible risk factors, the InterSePT study indicated some important predictors for suicidal behavior for the whole group: diagnosis of schizoaffective disorder, past and current use of alcohol or substance abuse, cigarette smoking, number of lifetime suicide attempts, and number of hospitalizations in the previous 36 months to prevent suicide (Potkin et al., 2003). Depression in schizophrenia is also differentiated according to the time when it occurs: during the prodrome of psychotic episodes, contemporaneously with psychosis, or as a post-psychotic depression phenomenon (Siris, 2001). However, in most of the studies, this differentiation was not mentioned. Regarding the diathesis of schizoaffective patients for depression, this may have greater potential importance and deserves to be studied. As a conclusion, it can be said that there are three groups of risk factors for schizoaffective patients: factors valid also for general population, factors related to a psychotic and chronic mental disorder, and factors mainly related to depression. It is not clear yet whether there are some correlates of suicidal behaviors specific to schizoaffective patients. Preventive factors for suicidal behavior are also as important as risk factors and main reported preventive factors for mood and schizophrenic/schizoaffective patients are shown in Table 12.2. There is also a need for more studies to find out their importance for schizoaffective patients.

234

Simavi Vahip Table 12.2. Potential preventive factors of suicidal behavior.

Adequate treatment of psychosis Adequate treatment of depression Close monitoring Social support Employment Cultural attitudes and other cultural preventive factors

Genetic findings of suicide in the overlapping spectra Are there general genetic factors predisposing specifically to suicidal behavior? Do some patients with different diagnoses have an additional genetic diathesis for suicide? Many of these questions do not have clear enough answers, yet. However, many studies pointed out the importance of familial if not genetic factors despite the highly complex interrelated factors contributing to suicidal behavior such as demographic, psychological, social, clinical, treatment, etc. A recent overview by Baldessarini and Hennen (2004) underlined the challenge of separating suicide-specific heritability apart from psychiatric illnesses. Despite this challenge, they pointed out the possibility of suicide specific-inheritence in several different types of data such as family studies, twin studies, and adoption studies. Of more than 20 controlled family studies which they evaluated, they reported nearly five-fold greater relative risk of suicidal acts among relatives of index cases with suicidal behavior compared to relatives of non-suicidal controls. The relative risk was greater for completed suicide than for attempts. At least in seven studies, one of which was conducted with a high number of twins – with the Twin Registry of Australia, of 1538 monozygotic (MZ) and 1199 dizygotic (DZ) twin pairs (Statham et al., 1998) – higher average concordance was found for suicidal behavior among co-twins of suicidal identical twins compared to fraternal twins or to relatives of other suicidal subjects. And in very few adoption studies a greater risk of completed suicides (but not of attempts among biological, compared to adoptive relatives of probands) among biological relatives of probands versus controls, was outlined. Molecular genetic studies – most of them related to proteins required for central serotonergic neurotransmission – were interpreted as tentative and inconclusive owing to inconsistencies of related research findings. Beyond this interpretation, most investigated genes are tryptophan hydroxylase genes, serotonin transporter genes, and genes for serotonin receptors. Polymorphisms of the tryptophan hydroxylase gene enzyme (the rate-limiting enzyme for neuronal biosynthesis of serotonin (5-hydroxytryptamine; 5-HT) and genes for some 5-HT receptors

235

Overlapping spectra: suicide

(1A, 2A, 1B/1D) were reported in association with suicidal behavior. These polymorphisms may have implications not only specific to suicide but in the context of psychiatric conditions such as depression, bipolar disorder, alcoholism, impulsivity, and violence (Åsberg and Forslund, 2000; Baldessarini and Hennen, 2004). However, considering the existence of negative results, we are far from a conclusion. If we do not consider family studies indicating the importance of heredity of schizoaffective disorders in general (Baron et al., 1982; Maier et al., 1992; Kendler et al., 1995), genetic studies directly addressing genetics of suicidal behavior in schizoaffective disorder are unfortunately absent. Other neurobiological findings The same question as in genetics is valid for the neurobiology of suicide in schizoaffective disorder: are there any specific neurobiological findings? The answer is again inconclusive. An enormous number of studies investigated neurobiological parameters associated with suicide in general to find common elements. Serotonergic dysfunction independent from depression has been shown in schizophrenic patients with a history of suicide attempt compared to patients without a history of suicide attempt and controls (Corrêa et al., 2002). This well-designed study supported the hypothesis that serotonergic dysfunction is at least partly associated with suicidal behavior regardless of the diagnosis. Beyond the methodological discussions, the serotonin system, the norepinephrin system, neuropeptides, cortisol, and cholesterol are the most investigated ones among many neurobiological parameters. The stronger evidence exists for the serotonin system. Association of low CSF 5-hydroxyindolacetic acid (5-HIAA) levels, with especially highly lethal and violent attempts and committed suicides is the most confirmed finding (Asberg and Forslund, 2000). In addition, there are platelet and challenge studies which mostly gave results in a similar direction (Corrêa et al., 2002). Postmortem brain studies also indicated 5-HT related changes in great consensus, mainly in prefrontal cortex (reduced serotonergic innervation and a consequent upregulation of postsynaptic receptor sites) and brain stem (lower 5-HT and 5-HIAA levels) (Mann et al., 1996). Possible connections with changes in the serotonergic system and suicide are not clear, yet. However, depression, certain behavioral dimensions (e.g., aggression, violence, impulsivity), substance-use disorders, particularly alcohol use, may act as intermediate pathways to suicide (Asberg and Forslund, 2000). The high prevalence of suicidal behavior in schizoaffective patients may be because of sharing these serotonin related correlates. Another correlate of suicide is a higher activity of the hypothalamic– pituitary–adrenal (HPA) system. Beyond the discussions of its sensitivity and

236

Simavi Vahip

specificity, the dexamethasone suppression test (DST) was widely used as an indicator of the HPA system. Although there are some findings supporting non-suppression in suicidal patients (Banki et al., 1984; Plocka-Lewandowska et al., 2001), this is mostly valid for primary major depressive patients and may even be a longterm trait indicator of suicide (Coryell and Schlesser, 2001). It is not precise whether DST non-suppression is associated with depression and/or suicides superimposed on schizophrenia and other psychiatric disorders. However, findings supporting HPA hyperactivity in schizophrenic patients who had a history of attempted suicide, suggest a possible diathesis for suicidal behavior in schizophrenic patients (Plocka-Lewandowska et al., 2001). In this regard, schizoaffective suicidal patients need to be studied for the possible hyperactivity of the HPA axis, both as a state and trait indicator. Despite relatively new studies on the relationship between cholesterol and suicidal behavior, there is strong supporting evidence (Åsberg and Forslund 2000, Atmaca et al., 2003). From population studies (Lindberg et al., 1992) to diagnosis-specific studies (Atmaca et al., 2003) there is growing evidence for this association. Whether it is related to low food intake in depressed and suicidal patients, or because of a connection via lower brain serotonin and poorer suppression of aggression (Lindberg et al., 1992; Muldoon et al., 1992), this issue should not be neglected in future studies. Findings such as significantly lower serum leptin and cholesterol in schizophrenics than in control cases, in schizophrenic patients with recent suicide attempt compared to patients without suicide attempt, and in violent attempters than in non-attempters (Atmaca et al., 2003) suggest promising areas for suicide studies in schizoaffective patients, too. Siris (2000) proposed the stress-vulnerability model as a potential integrating concept for biological and psychological factors and also for the continuum model and categorical model. This well-known model of schizophrenia explains the psychosis of schizophrenia as a final common path of neuropsychiatric decompensation (Lindenmeyer et al., 1991; Siris, 2000). Vulnerability to schizophrenic psychosis occurs on a continuum: from a strong vulnerability to a slight vulnerability. And in between a small group of individuals become psychotic if stressed enough. Possible stressors can be biological or psychosocial. Activation of an affective diathesis, such as depression, could act as a sufficient stressor precipitating psychosis in people with otherwise modest vulnerabilities. Siris supported this with some observations: depression-like symptoms observed early in the course of certain psychotic decompensations; association of dysphoria with positive symptoms rather than with negative symptoms in schizophrenia, and reduced psychotic exacerbations with antidepressants given for treating and preventing postpsychotic depression (Siris et al., 1994).

237

Overlapping spectra: suicide

Treatment issues of suicidal behavior in the overlapping spectra Some of the main questions regarding the treatment issues in the overlapping spectra are as follows: (1) Does the depressive component of schizoaffective disorder respond to antidepressants as it does in major depressive disorder? (2) Is there any medication that is specific for reducing the suicide risk? (3) Do mood stabilizers have suicide preventing effects in schizoaffective patients? (4) What kinds of psychotherapeutic and psychosocial interventions can be helpful to reduce the suicide risk in patients with schizoaffective disorder? Despite common occurrence and reports of depressive features in schizophrenia and schizoaffective disorder, there is a limited number regarding controlled studies of their treatment. Moreover, literature is somewhat unprecise and chaotic with regard to differentiation of syndrome versus symptom, timing of depressive features, and especially diagnostic criteria. It is not easy to draw a conclusion. However, it is obvious that depression is one of the major risk factors for suicide, and clinicians act to treat it in schizophrenic and schizoaffective patients without sufficient evidence derived from controlled studies. Levinson and colleagues (1999) evaluated the controlled and large open studies (where few controlled studies addressed a particular issue) in their excellent review. The authors concluded the following: typically and probably with greater efficacy on the depressive component, atypical antipsychotic drugs are most effective for the acute and maintenance treatment of patients with schizoaffective disorders and schizophrenic patients with mood symptoms. Optimization of antipsychotic treatment is more likely to be effective than the routine use of adjunctive antidepressants or mood stabilizers. The only exception is the use of antidepressants in patients who present a full depressive syndrome after stabilization of psychosis. Data supporting the use of antidepressants or mood stabilizers for subsyndromal depressive symptomatology or for depression during acute schizophrenic or schizoaffective exacerbations are very limited. Literature regarding the treatment of schizoaffective disorders and schizophrenia with depression generally addresses the efficacy of drugs on depression or psychotic symptoms in acute phases or in long-term maintenance. In general, a separate effect on suicidal behavior is not provided. Considering depression as a high risk factor for suicide, treatment of depression in schizophrenic and schizoaffective patients is important. Data about antidepressant efficacy in patients with schizoaffective disorder, depressed type, is very limited. In one of the studies (Brockington et al., 1978), diagnoses were concurrent schizophrenia or paranoid psychosis and depression, and chlorpromazine plus placebo was as effective as amitriptyline alone or amitriptyline plus chlorpromazine combination. Another study, conducted with schizophrenia with concurrent

238

Simavi Vahip

depression, reported that adding antidepressants to haloperidol resulted in an incomplete or prolonged resolution of the psychosis (Kramer et al., 1989). Cases in these two studies probably were not schizoaffective according to current classifications. These two studies indicate, once again, the importance of the diagnostic criteria that are used. Nevertheless, to my knowledge, these are the only two controlled studies that focused on the antidepressant drugs in such patients. With the introduction of atypical antipsychotics, the scene has changed and many studies have been conducted. For example, significantly more improvement in long-term treatment with clozapine in schizoaffective patients than patients with schizophrenia; and in patients with schizoaffective-bipolar type than patients with depressed type of schizoaffective disorders was reported (Banov et al., 1994). Similar findings were reported in acute and in treatment-refractory patients with clozapine (Levinson et al., 1999). A prospective, double-blind controlled study compared risperidone versus haloperidol plus amytryptiline combination and favored the combination in major depression with psychotic features, while there was no advantage of risperidone in patients with schizoaffective disorder, depressed type. Studies carried out with schizophrenic patients with depressive symptoms showed significantly more improvement with many atypical antipsychotics (e.g., olanzapine, risperidone, clozapine) compared to typical antipsychotics (Levinson et al., 1999). There is no controlled study testing the addition of antidepressants to antipsychotics. Most of the available studies showed no significant difference over antipsychotics with combination of antidepressants and antipsychotics. However, there are some early reports where the benefit of combination was higher, especially in outpatients (Levinson et al., 1999; Siris, 2000). Regarding the antidepressant properties of some mood-stabilizers, this is also a promising area of research. However, this issue is understudied, too. A preliminary report about the efficacy of lamotrigine in schizoaffective patients is encouraging (Erfurth et al., 1998). Is there any medication that has a specific suicide risk reducing effect? There has been great interest in specific antisuicidal effects of psychotropic drugs in recent years. Lithium is the most extensively studied and favored drug in this respect (Müller-Oerlinghausen, 2001). Re-evaluation of long-term maintenance studies and a series of meta-analysis have demonstrated a 13-fold reduction of suicide rates with lithium (Ernst and Goldberg, 2004). It is not clear yet whether this is part of its general mood-stabilizing effect or a specific antisuicidal effect. However, some possible ways were proposed for the explanation of this effect, such as reduction of impulsivity and aggressivity in mood disorders, moreover, across different diagnostic groups; serotonergic and antidopaminergic activity; personal interaction associated with the physician’s close monitoring, etc. (Baldessarini et al., 2001;

239

Overlapping spectra: suicide

Müller-Oerlinghausen, 2001; Tondo et al., 2001; Ernst and Goldberg, 2004). However, antisuicidal efficacy of lithium has not been well-documented for schizophrenic and schizoaffective patients. A second compound which has strong evidence for antisuicidal efficacy is clozapine. Its antisuicidal effect is well documented for patients with schizophrenia and schizoaffective disorder. As a consequence of strong evidence for this effect, for the first time, the Food and Drug Administration (FDA) approved a psychotropic drug for reduction of suicide risk in schizophrenic and schizoaffective patients. There are many studies supporting the antisuicidal effect of clozapine. Initial studies reported a significant reduction in the number of suicide attempts (Meltzer and Okayli, 1995), completed suicides (Meltzer and Okayli, 1995; Reid et al., 1998; Munro et al., 1999), and decreased suicidality in association with improvement in depression and hopelessness and reduction in lethality of suicide attempts (Meltzer and Okayli, 1995). The International Suicide Prevention Trial (InterSePT) is the most important study on the issue. It was a multi-center, prospective, randomized, parallel-group, two-year, open-label study. However, the raters of suicidal behavior were blind to drugs. Diagnosis of schizoaffective disorder was the highest risk factor in predicting suicidal behavior in follow-up. The study compared the efficacy of clozapine (n  479) versus olanzapine (n  477) among patients with schizophrenia and schizoaffective disorder who were at high risk for suicide. Results indicated significantly less suicidal behavior and many risk-related parameters in patients on clozapine as compared to olanzapine. Completed suicide rates were not different in the two groups. In a Cox proportional hazards regression model a 26% reduced risk for suicide attempt or hospitalization to prevent suicide for patients randomized to clozapine treatment compared with olanzapine treatment was reported (Meltzer et al., 2003). In a model, it has been estimated that clozapine could result in the saving of more than 10 250 life-years over the next 20 years, or an average of 53 lives per year in the United Kingdom (Duggan et al., 2003). The mechanisms underlying this effect are not clear, yet. Some possible explanations are an intrinsic antidepressant activity of clozapine (Meltzer and Okayli, 1995); conceptualizing suicidality as a separate domain that is related to but independent of depression and psychosis (Lindenmeyer et al., 2003); and modulations of the relationship between serotonin, dopamine and noradrenaline and their metabolites (Wagstaff and Perry, 2003). Is this effect unique to clozapine or do other members of the atypical antipsychotics group also have this effect? Despite the pharmacological (e.g., effects on serotonergic system) and clinical (efficacy in depressive symptoms in many diagnoses) similarities of other members with clozapine, the answer is not sufficiently clear yet. In a retrospective study, it was found that schizophrenic patients who attempted

240

Simavi Vahip

suicide were more frequently prescribed typical antipsychotics, whereas patients who did not attempt suicide were frequently prescribed atypical antipsychotics (Altamura et al., 2003). This study supports the widespread belief that atypicals in general have at least antidepressant efficacy if not antisuicidal efficacy compared to typical antipsychotics. However, Khan and colleagues’ study of the Food and Drug Administration (FDA) database gave contrary results. The authors assessed symptoms and risks of suicide and suicide attempts in psychotic patients in placebo controlled studies (a total of 10 118 patients used olanzapine, risperidone, quetiapine fumarate, typical antipsychotics, or placebo). Despite a general exclusion of patients with risk of suicide and close monitoring in the drug studies, there were 26 committed suicides and 51 attempted suicides in the whole group. Interestingly, rates of suicide and attempted suicide did not differ significantly between the placebo-treated and the drug-treated groups. Moreover, there was no significant difference in rates of suicide and attempted suicide among atypical antipsychotics, typical antipsychotics and placebo (Khan et al., 2001). Another study from the Netherlands, which evaluated 7 152 patients in 31 studies, found no significant difference between placebo and active treatment in suicide and suicide attempts (Storosum et al., 2003). Beyond methodological discussions on these two analyses, at least two atypical antipsychotics deserve to be evaluated separately with existing studies related to suicidal behavior: olanzapine and risperidone. In a 28-week study, which compared olanzapine versus risperidone in 339 patients with schizophrenia, schizoaffective or schizophreniform disorder, secondary analyses showed significantly lower suicide attempt rates with olanzapine than with risperidone (Tran et al., 1997). Additionally, a 2.3-fold reduced risk in a one-year period (Glazer, 1998) and a significantly greater improvement in suicidal thoughts (Beasley et al., 1998) were reported with olanzapine compared to patients on haloperidol. Data on risperidone is not encouraging enough. Muller-Siecheneder and colleagues (1998) compared risperidone versus haloperidol plus amytriptyline combination in patients with schizoaffective disorders depressive type, major depression, psychotic type and non-residual schizophrenia with major depressive symptoms, and reported the overall superiority of the combination to risperidone. This effect was mainly attributed to the major depression-psychotic group. Suicidal ideation was not different between the groups. Although there are some reports on the reduction of depressive symptoms in manic or psychotic patients with quetiapine, ziprasidone, and aripiprazole, there is no systematic data available to test their effects on suicidal behavior (Ernst and Goldberg, 2004). Although antidepressant properties, to a certain degree at least, in some diagnoses, such as bipolar disorder and major depression with psychotic features, is widely accepted, the antisuicidal effect of atypical antipsychotics other than clozapine need further systematic and controlled studies.

241

Overlapping spectra: suicide

Another important class of drugs, which may have an antisuicidal effect, is antidepressants. This is a controversial issue, too. In addition to their antidepressant efficacy in depression, which is one of the main causes of suicide, a short-term reduction in suicide risk is well documented (Ernst and Goldberg, 2004). However, their long-term potencies for preventing or inducing suicidal risk need more sophisticated studies, especially in different diagnostic categories. Future requirements Since suicidal behavior is a frequent and important component of the overlapping spectra of schizophrenia and mood disorders, there is great need for more systematic studies on suicidal behavior of schizoaffective disorders. Some of the questions which need to be investigated are as follows: • Are there any epidemiological differences in suicidal behaviors between schizoaffective disorders and pure schizophrenia or bipolar disorder with psychotic features? • There are some factors and/or correlates, which lead to suicidal behavior for the general population (e.g., lack of social support, unemployment, experiencing loss) and for severe and chronic mental disorders (e.g., frequent hospitalizations, social and functional impairment owing to illness, etc.). However, there are particularly high-risk factors for some disorders (e.g., major depression with psychotic features, bipolar II disorders), mostly owing to their biological nature and clinical features. Are there any factors or correlates specific to schizoaffective patients? • It has been generally reported that the depressive subtype of schizoaffective disorder has the highest risk among different subtypes of schizoaffective disorders for suicidal behavior. However, little is known about the rates for different subtypes. What are the rates of suicidal behavior in different subtypes of schizoaffective disorders, especially in the bipolar type? • Relatively preserved cognitive and functional status has frequently been reported in schizoaffective patients compared to schizophrenia. How are these functions related to suicidal behavior in schizoaffective patients? • Possibly, there are many routes leading to suicidal behavior of schizoaffective patients. Can we differentiate several of them? • What about differential efficacy of psychotropic drugs on suicidal behavior in schizoaffective patients? Controlled data are needed for atypical antipsychotics, mood stabilizers, and antidepressants. • Could there be any specific psychosocial/psychotherapeutic interventions for schizoaffective patients? One can add many more questions on suicidal behavior in schizoaffective patients, which need to be investigated. However, without increasing the reliability

242

Simavi Vahip

and the validity of the diagnosis, there will be many inconsistent findings and most of the questions will remain unanswered. Maybe we are at the limits of the descriptive approach and findings from the fields of genetic and neurobiology will open the route to more valid and reliable definitions in this overlapping spectrum of schizophrenia and mood disorders. On the other hand, suicide is such an important issue that it cannot wait until then, particularly in regard to the detection of risk factors, pharmacological and psychosocial interventions.

R E F E R E N C ES Altamura, A. C., Bassetti, R., Bignotti, S., Pioli, R. and Mundo, E. (2003). Clinical variables related to suicide attempts in schizophrenic patients: a retrospective study. Schizophrenia Research, 60, 47–55. Angst, F., Stassen, H. H., Clayton, P. J. and Angst, J. (2002). Mortality of patients with mood disorders: follow-up over 34–38 years. Journal of Affective Disorders, 67, 167–81. Angst, J., Stassen, H. H., Gross, G., Huber, G. and Stone, M. H. (1990). Suicide in affective and schizoaffective disorders. In Affective and Schizoaffective Disorders: Similarities and Differences, ed. A. Marneros and M. T. Tsuang, pp. 168–85. Berlin, Heidelberg: Springer-Verlag. Åsberg, M. and Forslund K. (2000). Neurobiological aspects of suicidal behaviour. International Review of Psychiatry, 12, 62–74. Atbas¸og˘lu, C., Schultz, S. K. and Andreasen, N. C. (2001). The relationship of akathisia with suicidality and depersonalization among patients with schizophrenia. Journal of Neuropsychiatry and Clinical Neuroscience, 13, 336–41. Atmaca, M., Kulog˘lu, M., Tezcan, E. and Üstündag˘, B. (2003). Serum leptin and cholesterol levels in schizophrenic patients with and without suicide attempts. Acta Psychiatrica Scandinavica, 108, 208–14. Balázs, J., Lecrubier, Y., Csiszér, N., Koszták, J. and Bitter I. (2003). Prevalance and comorbidity of affective disorders in persons making suicide attempts in Hungary: importance of the first depressive episodes and of bipolar II diagnoses. Journal of Affective Disorders, 76, 113–19. Baldessarini, R. J. and Hennen J. (2004). Genetics of suicide: An overview. Harvard Review of Psychiatry, 12, 1–13. Baldessarini, R. J., Tondo, L. and Hennen, J. (2001). Treating the suicidal patient with bipolar disorder. Reducing suicide risk with lithium. Annals of the New York Academy of Sciences, 932, 24–38. Banki, C. M., Arato, M., Papp, Z. and Kurcz, M. (1984). Biochemical markers in suicidal patients. Investigations with cerebrospinal fluid amine metabolites and neuroendocrine tests. Journal of Affective Disorders, 144, 341–50. Banov, M. D., Zarate, C. A., Tohen, M. et al. (1994). Clozapine therapy in refractory affective disorders: polarity predicts response in long-term follow-up. Journal of Clinical Psychiatry, 55, 295–300.

243

Overlapping spectra: suicide Baron, M., Gruen, R., Asnis, L. and Kane, J. (1982). Schizoaffective illness, schizophrenia and affective disorders: morbidity risk and genetic transmission. Acta Psychiatrica Scandinavica, 65, 253–62. Beasley, C. M., Sayler, M. E., Kiesler, G. M. et al. (1998). The influence of pharmacotherapy on self-directed and externally-directed aggression in schizophrenia. Schizophrenia Research, 29, 28–9. Beautrais, A. L. (2001). Suicides and serious suicide attempts: two populations or one? Psychological Medicine, 31, 837–45. Beck, A. T., Kovacs, M. and Weissman, A. (1975). Hopelessness and suicidal behavior. An overview. Journal of the American Medical Association, 234, 1146–9. Bermanzohn, P. C. and Siris, S. G. (1992). Akinesia: a syndrome common to Parkinsonism, retarded depression, and negative symptoms. Comprehensive Psychiatry, 33, 221–32. Black, D. W., Winokur, G. and Warrack, G. (1985). Suicide in schizophrenia: the Iowa record linkage study. Journal of Clinical Psychiatry, 46, 14–17. Birchwood, M., Mason, R., MacMillan, F. and Healy, J. (1993). Depression, demoralization and control over psychotic illness: a comparison of depressed and non-depressed patients with a chronic psychosis. Psychological Medicine, 23, 387–95. Bostwick, J. M. and Pankratz, V. Z. (2000). Affective disorders and suicide risk: a re-examination. American Journal of Psychiatry, 157, 1925–32. Bourgeois, M., Swendsen, J., Young, F. et al. and the InterSePT Study Group. (2004). Awareness of disorder and suicide risk in the treatment of schizophrenia: results of the international suicide prevention trial. American Journal of Psychiatry, 161, 1494–6. Brockington, I. F., Kendell, R. E., Kellett, J. M., Curry, S. H. and Wainwright, S. (1978). Trials of lithium, chlorpromazine and amitriptyline in schizoaffective patients. British Journal of Psychiatry, 133, 162–8. Casey, D. E. (1988). Affective disorders and tardive dyskinesia. L’Encéphale, 14, 221–6. Cohen, L. J., Test, M. A. and Brown, R. L. (1990). Suicide and schizophrenia: data from a prospective community treatment study. American Journal of Psychiatry, 147, 602–7. Corrêa, H., Duval, F., Mokrani, M-C. et al. (2002). Serotonergic function and suicidal behavior in schizophrenia. Schizophrenia Research, 56, 75–85. Coryell, W. and Schlesser, M. (2001). The dexamethasone suppression test and suicide prediction. American Journal of Psychiatry, 158, 748–53. Coryell, W., Keller, M., Lavori, P. and Endicott, J. (1990). Affective syndromes, psychotic features, and prognosis. I. Depression. Archives of General Psychiatry, 47, 651–7. Deisenhammer, E. A., DeCol, C., Honeder, M., Hinterhuber, H. and Fleischhacker, W. W. (2000). In-patient suicide in psychiatric hospital. Acta Psychiatrica Scandinavica, 102, 290–4. . Dog˘aner, I.., Noyan, A., Akdeniz, F., Pirildar, S¸. and Tug˘lular, I. (1996). Iki uçlu duygudurum bozuklug˘unda klinik ve sag˘altim özellikleri ile geç diskinezi ilis¸kisi. Türk Psikiyatri Dergisi, 7, 93–100. Drake, R. E. and Ehrlich, J. (1985). Suicide attempts associated with akathisia. American Journal of Psychiatry, 142, 499–501. Drake, R. E., Gates, C., Cotton, P. G. and Whitaker, A. (1984). Suicide among schizophrenics: who is at risk? Journal of Nervous and Mental Disease, 172, 613–17.

244

Simavi Vahip Drake, R. J., Pickles, A., Bentall, R. P. et al. (2004). The evolution of insight, paranoia and depression during early schizophrenia. Psychological Medicine, 34, 285–92. Duggan, A., Warner, J., Knapp, M. and Kerwin, R. (2003). Modelling the impact of clozapine on suicide in patients with treatment-resistant schizophrenia in the UK. British Journal of Psychiatry, 182, 505–8. Erfurth, A., Walden, J. and Grunze, H. (1998). Lamotrigine in the treatment of schizoaffective disorder. Neuropsychobiology, 38, 204–5. Ernst, C. L. and Goldberg, J. F. (2004). Antisuicide properties of psychotropic drugs: a critical review. Harvard Review of Psychiatry, 12, 14–41. Fenton, W. S., McGlashan, T. H., Victor, B. J. and Blyler, R. (1997). Symptoms, subtype, and suicidality in patients with schizophrenia spectrum disorders. American Journal of Psychiatry, 154, 199–204. Glazer, W. M. (1998). Formulary decisions and health economics. Journal of Clinical Psychiatry, 59, 23–9. Goodwin, F. K. and Jamison, K. F. (1990). Manic-Depressive Illness. Washington DC: Oxford University Press. Guze, S. B., Robins, E. (1970). Suicide and primary affective disorders. British Journal of Psychiatry, 117, 437–8. Harkavy-Friedman, J. M., Nelson, E. A., Venarde, D. E. and Mann, J. J. (2004). Suicidal behavior in schizophrenia and schizoaffective disorder: Examining the role of depression. Suicide and Life-Threatening Behavior, 34, 66–76. Harris, E. C. and Barraclough, B. (1997). Suicide as an outcome for mental disorders. A metaanalysis. British Journal of Psychiatry, 170, 205–8. Heilä, H., Heikkinen, M. E., Isometsä, E. T. et al. (1999). Life events and completed suicide in schizophrenia: A comparison of suicide victims with and without schizophrenia. Schizophrenia Bulletin, 25, 519–31. Henriksson, M. M., Aro, H. M., Marttunen, M. J. et al. (1993). Mental disorders and comorbidity in suicide. American Journal of Psychiatry, 150, 935–40. Kendler, K. S., McGuire, M., Gruenberg, A. M. and Walsh, D. (1995). Examining the validity of DSM-III-R schizoaffective disorder and its putative subtypes in the Roscommon family study. American Journal of Psychiatry, 152, 755–64. Khan, A., Khan, S. R., Leventhal, R. M. and Brown, W. A. (2001). Symptom reduction and suicide risk among patients treated with placebo in antipsychotic clinical trials: an analysis of the food and drug administration database. American Journal of Psychiatry, 158, 1449–54. Kim, C. H., Jayathilake, K. and Meltzer H. Y. (2003). Hopelessness, neurocognitive function, and insight in schizophrenia: relationship to suicidal behavior. Schizophrenia Research, 60, 71–80. Kramer, M. S., Vogel, W. H., DiJohnson, C. et al. (1989). Antidepressants in “depressed” schizophrenic inpatients. Archives of General Psychiatry, 46, 922–7. Krupinski, M., Fischer, A., Grahmann, R. et al. (1998). Risk factors for suicides of inpatients with depressive psychoses. European Archives of Psychiatry and Clinical Neuroscience, 248, 141–7. Kuo, W-H., Gallo, J. J. and Eaton, W. W. (2004). Hopelessness, depression, substance disorder, and suicidality: A 13-year community-based study. Social Psychiatry and Psychiatric Epidemiology, 39, 497–501.

245

Overlapping spectra: suicide Levinson, D. F., Umapathy, C. and Musthaq, M. (1999). Treatment of schizoaffective disorder and schizophrenia with mood symptoms. American Journal of Psychiatry, 156, 1138–48. Levitt, J. J. and Tsuang, M. T. (1988). The heterogeneity of schizoaffective disorder: implications for treatment. American Journal of Psychiatry, 145, 926–36. Lindberg, G., Rastam, L., Gullberg, B. and Eklund, G. A. (1992). Low serum cholesterol concentration and short-term mortality from injuries in men and women. British Medical Journal, 305, 277–9. Lindenmeyer, J. P., Czobor, P., Alphs, L. et al. and InterSePT Study Group (2003). The InterSePT scale for suicidal thinking reliability and validity. Schizophrenia Research, 63, 161–70. Lindenmeyer, J. P., Grochowski, S. and Kay, S. R. (1991). Schizophrenic patients with depression: psychopathological profiles and relationship with negative symptoms. Comprehensive Psychiatry, 32, 528–33. Maier, W., Lichtermann, D., Minges, J. et al. (1992). Schizoaffective disorder and affective disorders with mood-incongruent psychotic features: keep separate or combine? Evidence from a family study. American Journal of Psychiatry, 149, 1666–73. Mann, J. J., Underwood, M. D. and Arango, V. (1996). Postmortem studies of suicide victims. In Biology of Schizophrenia and Affective Disease, ed. S. J. Watson. Washington DC: American Psychiatric Press, pp. 197–221. Marneros, A., Rohde, A., Deister, A., Fimmers, R., and Jünemann, H. (1988). Long-term course of schizoaffective disorders, Part III: Onset, type of episodes and syndrome shift, precipitating factors, suicidality, seasonality, inactivity of illness, and outcome. European Archives of Psychiatry and Neurological Sciences, 235, 283–90. Meltzer, H. Y., Alphs, L, Gren, A. I. et al. and the InterSePT Study Group. (2003). Clozapine treatment for suicidality in schizophrenia. Archives of General Psychiatry, 60, 82–91. Meltzer, H. Y. and Okayli, G. (1995). Reduction of suicidality during clozapine treatment of neuroleptic-resistant schizophrenia: Impact on risk-benefit assessment. American Journal of Psychiatry, 152, 183–90. Modestin, J., Zarro, I. and Waldvogel, D. (1992). A study of suicide in schizophrenic in-patients. British Journal of Psychiatry, 160, 398–401. Mortensen, P. B., Agerbo, E., Erikson, T., Qin, P. and Vestergaard-Nielsen, N. (2000). Psychiatric illness and risk factors for suicide in Denmark. Lancet, 355, 9–12. Muldoon, M. F., Kaplan, J. R., Manuck, S. B. and Mann, J. J. (1992). Effects of a low-fat diet on brain serotonergic responsivity in cynomolgus monkeys. Biological Psychiatry, 31, 739–42. Munro, J., O’Sullivan, D., Andrews, D. et al. (1999). Active monitoring of 12,760 clozapine recipients in the UK and Ireland: beyond pharmacovigilance. British Journal of Psychiatry, 175, 576–80. Müller-Oerlinghausen, B. (2001). Arguments for the specificity of the antisuicidal effect of lithium. European Archives of Psychiatry and Clinical Neuroscience, 251 (Suppl. 2), 72–5. Muller-Siecheneder, F., Muller, M. J., Matthias, J. et al. (1998). Risperidone versus haloperidol and amitriptyline in the treatment of patients with a combined psychotic and depressive syndrome. Journal of Clinical Psychopharmacology, 18, 111–20 Nyman, A. K. and Jonsson, H. (1986). Patterns of self-destructive behaviour in schizophrenia. Acta Psychiatrica Scandinavica, 73, 252–62.

246

Simavi Vahip Pini, S., Cassano, G. B., Dell’Osso, L. and Amador, X. F. (2001). Insight into illness in schizophrenia, schizoaffective disorder, and mood disorders with psychotic features. American Journal of Psychiatry, 158, 122–5. Plocka-Lewandowska, M., Araszkiewicz, A. and Rybakowski, JK. (2001). Dexamethasone suppression test and suicide attempts in schizophrenic patients. European Psychiatry, 16, 428–31. Potkin, S. G., Alphs, L., Hsu, C. et al. and the InterSePT Study Group. (2003). Predicting suicidal risk in schizophrenic and schizoaffective patients in a prospective two-year trial. Biological Psychiatry, 54, 444–52. Radomsky, E. D., Haas, G. L., Mann, J. J. and Sweeney J. A. (1999). Suicidal behavior in patients with schizophrenia and other psychotic disorders. American Journal of Psychiatry, 156, 1590–5. Reid W. H., Mason, M. and Hogan, T. (1998). Suicide prevention effects associated with clozapine therapy in schizophrenia and schizoaffective disorder. Psychiatry Services, 40, 1029–33. Rich, C. L., Warsradt, G. M., Nemiroff, R. A., Fowler, R. C., and Young, D. (1991). Suicide, stressors, and the life cycle. American Journal of Psychiatry, 148, 524–7. Rifkin, A., Quitkin, F. and Klein, D. F. (1975). Akinesia: a poorly recognized drug-induced extrapyramidal behavioural disorder. Archives of General Psychiatry, 32, 672–5. Rihmer, Z., Barsi, J., Arato, M. and Demeter, E. (1990). Suicide in subtypes of primary major depression. Journal of Affective Disorders, 18, 221–5. Rihmer, Z. and Kiss, K. (2002). Bipolar disorders and suicidal behaviour. Bipolar Disorders, 4 (Suppl. 1), S21–5. Saarinen, P. I., Lehtonen, J. and Lönnqvist, J. (1999). Suicide risk in schizophrenia: An analysis of 17 consective suicides. Schizophrenia Bulletin, 25, 533–42. Schwartz, R. C. (2001) Self-awareness in schizophrenia: its relationship to depressive symptomatology and broad psychiatric impairments. Journal of Nervous and Mental Disease, 189, 401–3. Schwartz, R. C. and Smith, S. D. (2004). Suicidality and psychosis: the predictive potential of symptomatology and insight into illness. Journal of Psychiatric Research, 38, 185–91. Shear, K., Frances, A. and Weiden, P. (1983). Suicide associated with akathisia and depot fluphenazine treatment. Journal of Clinical Psychopharmacology, 3, 235–6. Siris, S. G. (2000). Depression in schizophrenia: perspective in the era of “atypical” antipsychotic agents. American Journal of Psychiatry, 157, 1379–89. Siris, S. G. (2001). Suicide and schizophrenia. Journal of Psychopharmacology, 15, 127–35. Siris, S. G., Bermanzohn, P. C., Mason, S. E. and Shuwal, M. A. (1994). Maintenance imipramine therapy for secondary depression in schizophrenia: a controlled trial. Archives of General Psychiatry, 51, 109–15. SpießI, H., Hübner-Liebermann, B. and Cording, C. (2002). Suicidal behaviour of psychiatric inpatients. Acta Psychiatrica Scandinavica, 106, 134–8. Statham, D. J., Heath, A. C., Madden, P. A. F. et al. (1998). Suicidal behaviour: an epidemiological and genetic study. Psychological Medicine, 28, 839–55. Stephens, J. H., Richard, P., McHugh, P. R. (1997). Long-term follow-up of patients hospitalized for schizophrenia, 1913–40. Journal of Nervous and Mental Diseases, 185, 715–21. Stephens, J. H., Richard, P. and McHugh, P. R. (1999). Suicide in patients hospitalized for schizophrenia: 1913 to 1940. Journal of Nervous and Mental Disease, 187, 10–14.

247

Overlapping spectra: suicide Storosum, J. G., Van Zwieten, B. J., Wohlfarth, T. et al. (2003). Suicide risk in placebo vs. active treatment in placebo-controlled trials for schizophrenia. Archives of General Psychiatry, 60, 365–8. Taylor, M. A. (1992). Are schizophrenia and affective disorder related? A selective literature review. American Journal of Psychiatry, 149, 22–32. Tondo, L., Isacsson, G. and Baldessarini, J. (2003). Suicidal behaviour in bipolar disorder. CNS Drugs, 17, 491–511. Tondo, L., Hennen, J. and Baldessarini, J. (2001). Lower suicide risk with long-term lithium treatment in major affective illness: a meta-analysis. Acta Psychiatrica Scandinavica, 104, 163–72. Tran, P. V., Hamilton, S. H., Kuntz, A. J., Potvin, J. H., et al. (1997). Double-blind comparison of olanzapine versus risperidone in the treatment of schizophrenia and other psychotic disorders. Journal of Clinical Psychiatry, 17, 407–18. Tsuang, M. T. (1978). Suicide in schizophrenics, manics, depressives, and surgical controls. Archives of General Psychiatry, 35, 153–5. Van Putten, T. and May, P. R. A. (1978). “Akinetic depression” in schizophrenia. Archives of General Psychiatry, 35, 1101–7. Virkkunen, M. (1976). Attitude to psychiatric treatment before suicide in schizophrenia and paranoid psychoses. British Journal of Psychiatry, 128, 47–9. Wagstaff, A. J. and Perry, C. M. (2003). Clozapine: in prevention of suicide in patients with schizophrenia or schizoaffective disorder. CNS Drugs, 17, 273–80. Westermeyer, J. F., Harrow, M. and Marengo, J. T. (1991). Risk for suicide in schizophrenia and other psychotic and nonpsychotic disorders. Journal of Nervous and Mental Disease, 179, 259–66. Yen, C-F., Yeh, M-L., Chen, C-S. and Chung, H-H. (2002). Predictive value of insight for suicide, violence, hospitalization, and social adjustment for outpatients with schizophrenia: A prospective study. Comprehensive Psychiatry, 43, 443–7.

13

Biological treatment of schizoaffective disorders Christopher Baethge Department of Psychiatry and Psychotherapy, University of Cologne Medical School

Introduction One example or model of the overlap of affective and schizophrenic spectra could be assumed to be the schizoaffective disorders (SAD). This topic is extensively dealt with in the first chapter of this book, so to avoid redundancy we refer to this chapter as well as to books extensively discussing the topic, such as Marneros and Tsuang 1986, 1990, Marneros, et al. 1995a, Marneros and Angst et al., 2000 as well as Marneros and Goodwin, 2005 a, b (see Chapter 1 references). In contrast to schizophrenia and recurrent affective disorders, SAD is rarely the focus of treatment guidelines. The standard textbooks of psychopharmacology contain no chapters on the treatment of schizoaffective disorders (e.g., Schatzberg and Nemeroff, 2004). Therefore, this chapter aims at summarizing the results pertaining to acute and long-term treatment of SAD. After a brief methodological part, the results regarding acute treatment of SAD episodes will be reviewed followed by an overview of SAD long-term treatment. The chapter will end with a summary of preliminary treatment guidelines that can be inferred from the data reviewed. Methods of review

This review is restricted to the efficacy and effectiveness of pharmacological treatments for schizoaffective disorders. Methodologically it is a non-systematic update of the most recent systematic review of SAD long-term treatment (Baethge, 2003a). Owing to SAD’s separated nosological standing in DSM and ICD, this overview includes only studies reporting on SAD patients in particular; that is, differentiated from patients with schizophrenia and affective disorders. In order to exclude the most anecdotal forms of evidence, case reports and studies reporting on less than The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

248

249

Biological treatment of schizoaffective disorders

five individuals were not taken into consideration. Also, congress abstracts and reports in languages other than English or German were excluded. For the purpose of this study, long-term studies were defined as studies covering at least six months of treatment. The review is based on searches of literature databases, on a hand search of the bibliographies in the papers retrieved, and on a hand search of psychiatric handbooks. The search terms for Medline (1966–2001) and EMBASE (1980–2001) were: schizoaffective, schizo-affective, long-term, maintenance, prophylactic. The Medline search was repeated in 2005 (August) including queries for the following terms: schizoaffective, valproate, lamotrigine, topiramate, olanzapine, risperidone, amisulpride, clozapine, aripiprazole, quetiapine, and ziprasidone. The articles were divided into acute treatment trials and long-term studies for clinical reasons: a schizoaffective patient can suffer from a predominantly affective episode, whereas the long-term course of the illness might be characterized by mainly schizophrenic SAD. In addition, long-term treatment differs from acute treatment in several ways that often affect treatment decisions, e.g., questions of dose, side effects, compliance. Therefore, a substance that might have served well as an acute treatment is not always a good choice for long-term treatment (e.g., haloperidol in schizophrenia or valproate in bipolar disorder). Diagnostic issues

The concept of SAD underwent many changes since Kasanin introduced the term schizoaffective disorders into psychiatry (Kasanin, 1933). For the first time, the DSM-III-R of 1987 offered a set of criteria that were generally accepted and widely adopted in clinical practice (Baethge, 2003a). Earlier the Research Diagnostic Criteria (RDC) of the seventies (Spitzer et al., 1977) were frequently used in research. Before RDC, there were many different, sometimes idiosyncratic, diagnostic criteria (Baethge, 2003). This may have rendered the diagnosis less reliable. The study of Greil and coworkers (1997) on SAD long-term treatment highlights the diagnostic difficulties that have to be taken into consideration when reviewing the research on SAD treatment: of the 90 patients that were diagnosed with SAD according to ICD-9, 74 also had a SAD-diagnosis in the terms of RDC. Among those patients, only 17 also fulfilled DSM-III-R-criteria for SAD. Similarly, Maj et al. (2000) documented that 33 of 48 SAD patients according to ICD-9 were also considered schizoaffective after RDC had been used. In the same vein, interrater reliability of the SAD diagnosis has been shown to be low, even if modern diagnostic criteria are used (Faraone et al., 1996). In the study by Maj and colleagues (2000) Cohen’s kappa for the DSM-IV-diagnosis of SAD was reported to be 0.22.

250

Christopher Baethge

The criteria of DSM-IV (American Psychiatric Association 1994) and ICD-10 (World Health Organization 1992) differ in that the DSM emphasizes that there must be a period of at least two weeks with delusions or hallucinations without prominent affective psychopathology. As a result, study samples differ according to the diagnostic system used (Baethge, 2003a). In this overview, particularly in the review of long-term treatment, studies are separated into two kinds: those before modern diagnostic criteria came into use, starting with RDC, and in those that were used prior to the publication of the Research Diagnostic Criteria. Acute treatment Lithium

Lithium had been introduced into psychiatry before widespread diagnostic criteria for SAD were established. Therefore, many lithium trials are not “modern” in the sense defined above. In an overview, Goodnick and Meltzer (1984) report on more than two dozen lithium studies in SAD acute treatment, controlled and uncontrolled. In this chapter, however, only the most important controlled trials are covered. In a very small placebo-controlled double-blind trial (n8) Alexander and coworkers (1979) found lithium to be superior to placebo in schizoaffective patients. Three studies compared lithium with chlorpromazine: Johnson and coauthors (1971; n13) reported that among “excited schizoaffective” patients – likely to be considered schizomanic today – lithium was inferior to chlorpromazine with regard to BPRS and CGI ratings after three weeks. However, larger studies by Prien et al. (1972) and by Brockington et al. (1978) reached a different conclusion. The first group compared lithium and chlorpromazine in 83 excited schizoaffectives (author’s criteria). They divided the sample with regard to excitation into a “highly active” group and a “mildly active” group. Chlorpromazine (CPZ) was superior after three weeks of treatment owing to a higher drop-out rate of lithium patients in the “highly active” group (regarding several ratings, e.g., BPRS). Both compounds, however, had very similar effects among mildly excited patients. The treatment-related drop-outs in the lithium group occurred in the first ten days of the trial. This time pattern points to the lag time between the start of lithium and the onset of its effect. Brockington and coworkers (1978), in a study of 19 schizomanic patients found lithium to be as effective as CPZ. Those findings were supported by a four-week RDC-based study with 31 patients by Braden et al. (1982): similar effectiveness was reported for lithium and CPZ, except for patients who were highly active. In the highly active group CPZ was superior.

251

Biological treatment of schizoaffective disorders

In sum, lithium appears to be a treatment option in acute SAD mania as long as the patients are not too excited. Lithium seems to be more useful in mainly affective schizoaffectives, in particular, in mainly affective schizomanic patients and less so in schizodepressed patients although data in this regard is sparse (Small et al., 2003; Alexander et al., 1979; Goodnick and Meltzer, 1984). Judged from the results of early studies, lithium appears to be almost as effective in the treatment of SAD mania as in bipolar mania (Goodnick and Meltzer, 1984). However, in their own open and non-randomized trial of lithium versus placebo in SAD mania, Goodnick and Meltzer (1984) found that in patients with schizoaffective mania (n30), time to response to lithium was considerably longer (two-fold) than in bipolar mania (n 41). Not much is known about what predicts successful lithium treatment in SAD. It might prove clinically useful, nevertheless, to use lithium particularly in affective SAD patients with a family history of bipolar disorder and with an episodic – as opposed to a chronic – course of the illness. It is important to note that this is an analogy to bipolar affective disorder (Passmore et al., 2003). Among other clinical characteristics that suggest a consideration of lithium, suicidality should be particularly mentioned (Baldessarini et al., 2003; Müller-Oerlinghausen et al., 2005). Lerner and Mintzer (1988) reported that the combination of haloperidol and lithium was superior to the combination of haloperidol and placebo in a small group of depressed schizophrenic and schizoaffective patients (RDC) with regard to BPRS improvement. The combination of lithium and antipsychotics has also been shown to be effective in SAD mania (Biederman et al., 1979). Data recommending lithium as a monotherapy for the treatment of schizodepression is insufficient, in part because only a small number of patients has been included in studies. Results show that response rates are considerably lower than in SAD mania (Goodnick and Meltzer, 1984). Antidepressants

Brockington and coworkers (1978) compared chlorpromazine and the antidepressant amitriptyline as well as the combination of the two drugs in 41 schizodepressive patients (pre-RDC criteria). Of note, CPZ-mono was associated with the best results, albeit non-significantly. On the other hand, treatment response was very modest with most of the patients not fully recovering during four weeks. Kramer et al. (1989), in a group of 58 patients with a cross-sectional diagnosis of SAD depression (but a longitudinal diagnosis of schizophrenia), found the addition of desipramine or amitriptyline to haloperidol to be worse than haloperidol alone. This underlines that schizodepression is a psychopathology particularly difficult to treat (Goodnick and Meltzer, 1984; Levinson et al., 1999). It is also one that has not been subjected to many studies.

252

Christopher Baethge

There is little evidence from randomised controlled trials (RCTs) that antidepressants are of benefit for patients with schizophrenia and depressive syndromes (Whitehead et al., 2002). Also, in a controlled study with patients suffering from late-life psychotic depression, the combination of nortriptyline and perphenazine was not more effective than perphenazine alone (Mulsant et al., 2001) indicating that adding an antidepressant to an antipsychotic is not always helpful, although it can be (Spiker et al., 1985). Findings from schizophrenia and psychotic depression do not necessarily apply to schizoaffective disorder. However, they show that more is not necessarily merrier and they emphasize the need for research. Anticonvulsants

There is solid evidence from controlled studies for the effectiveness of the anticonvulsants valproate and carbamazepine in the acute treatment of acute mania in bipolar disorder. Controlled studies of gabapentin, topiramate, and lamotrigine have been negative in this regard (for review see, for example, Moseman et al., 2003; Frye and Post, 2005; Ketter, 2005). However, only very little evidence supports the use of these agents in the acute treatment of schizoaffective episodes. Bogan and coworkers (2000), in a retrospective study of 20 schizoaffective patients (bipolar type) who had received valproate in addition to ongoing antipsychotic treatment, wrote that the drug was well tolerated and that three-fourths displayed CGI improvement. It is noteworthy that in a recent review of RCTs of valproate in schizophrenia and schizophrenia-like psychoses Basan and Leucht (2004) did not find sufficient data to support the use of valproate. The authors write that its effects for the sub-group of schizoaffective disorder were unknown. As a result, the rising use of valproate in the acute treatment of SAD (Flynn et al., 2002) appears unjustified. In Okuma and coworker’s four-week, double-blind RCT of 35 SAD patients (using their own SAD criteria) the combination of carbamazepine plus antipsychotics was not superior to placebo plus antipsychotics (Okuma et al., 1989). Here, too, a recent review of carbamazepine’s merits in schizophrenia and SAD showed that there is not enough supporting evidence from RCTs regarding monotherapy or adjunctive therapy for acute schizoaffective disorder (Leucht et al., 2004). The role of oxcarbazepine, carbamazepine’s better tolerated keto-derivative, in the acute treatment of schizoaffective disorders (Dietrich et al., 2003) has not been established yet. There is anecdotal evidence from a retrospective chart review and one open case series (Chengappa et al., 1999; Gupta et al., 2000) that topiramate might be useful; however, similar to the other anticonvulsants, positive controlled trials are lacking. Lamotrigine might turn out to be valuable in the acute treatment of schizodepressed episodes in SAD – a condition particularly difficult to treat (Levinson et al., 1999) – but no controlled study has been published so far.

253

Biological treatment of schizoaffective disorders

Antipsychotics

Although there is an almost incalculable host of double-blind randomized studies of atypical antipsychotics including for patients with schizophrenia and schizoaffective disorder, only very few present data for SAD patients. It is unclear how the picture would change if that had happened. The CPZ findings by Braden et al. (1982), Brockington et al. (1978), and Prien et al. (1972) cited above are among the few data about antipsychotics in the acute treatment of SAD in particular. However, it emerges that classical antipsychotics – if one takes chlorpromazine as a proxy for this group – are at least as effective as lithium in this group of patients. In highly excited patients CPZ seems to be superior; in mildly excited SAD patients CPZ was as effective as lithium. In general, olanzapine and risperidone were superior to haloperidol, and ziprasidone was better than placebo in the treatment of acute schizoaffective episodes (Janicak et al., 2001; Keck et al., 2001; Tohen et al., 2001). Tohen and co-authors (2001), in a post hoc analysis of 177 patients with SAD (bipolar type) compared olanzapine and haloperidol. Of note, they found that olanzapine was not statistically superior in the manic sub-group but was superior in the depressed sub-group. Janicak et al. (2001), in their trial comparing risperidone versus haloperidol (n 62) also found the newer antipsychotic to be superior to haloperidol in severely depressed SAD patients, whereas no other superiority was observed. To the knowledge of the author, there are no studies presenting data regarding SAD acute treatment for quetiapine and aripiprazole (see, for example, Mensink and Slooff, 2004). Also, there are no head-to-head comparisons of different newer generation antipsychotics. As a result, there is no first-line antipsychotic to recommend. The notion from independent studies (that is, not sponsor-driven) in schizophrenia research that olanzapine and clozapine might be slightly superior to other antipsychotics (Volavka et al., 2002; Lieberman et al., 2005) is still under debate, and its applicability to SAD is even more unclear. However, it is encouraging that, in the comparison with haloperidol, results for olanzapine and risperidone point to an improved effect in depressed patients. Ziprasidone, too, seems to help patients with the depressive type of schizoaffective disorder when compared with placebo (Keck et al., 2001). Long-term treatment Lithium

Eight of 25 studies on SAD long-term treatment with lithium included in this overview have used diagnostic criteria developed since RDC came into use, whereas many of the remaining 17 studies were published in the sixties and seventies. Among those early studies some showed that lithium maintenance treatment

254

Christopher Baethge

reduces the number of SAD episodes when compared with the time prior to medication (for example, Baastrup and Schou, 1967; Angst et al., 1970). In sum, the evidence that lithium works at all in SAD is derived from six prospective mirror image studies involving 242 lithium patients with an average follow-up time of 1.4 years (range: 0.9–2.3). Only one study (n33) used modern diagnostic criteria (Maj, 1988). The author is not aware of any study indicating that lithium does not work in SAD long-term treatment. The modern studies included 170 patients, and the number of patients included in all lithium studies is 607. On average the follow-up was 3.0 years (range: 0.9–6.8). However, there were only five prospective and controlled studies, three of which were also randomized. In the earliest RCT (Mattes and Nayak, 1984) lithium and fluphenazine were compared, in a double-blind fashion and over one year, in a sample of patients with SAD of the mainly schizophrenic subtype (n14; diagnostic criteria: RDC). Of the seven patients receiving lithium six had a recurrence, whereas two in the fluphenazine group (completers n5) suffered from a recurrence – a statistically significant difference. In both of the other RCTs lithium versus carbamazepine was studied: Bellaire et al. (1990) carried out a one-year study with 17 patients that they labeled as schizoaffective (DSM-III, no formal criteria). They observed three recurrences in the carbamazepine group (n8) and five among the nine patients taking lithium (difference non-significant). In both groups an improvement in illness course was observed when the study period was compared with the time prior to treatment. Greil and coworkers (1997) carried out the largest study in the field so far: over a follow-up time of 30 months they treated 47 patients with carbamazepine and 43 with lithium in an open fashion. As for the group of SAD patients diagnosed according to ICD-9 (no formal criteria), there were no differences between both treatments. However, when RDC was applied, as mentioned above, a diagnosis of SAD was upheld in 74 patients (mainly affective: 62; mainly schizophrenic: 12), and among the RDC schizodepressive patients (n35) carbamazepine was significantly superior to lithium in preventing recurrences. In the schizomanic group (n39) no differences occurred. An intention-to-treat-analysis of the whole group resulted in no differences between carbamazepine (15 drop-outs) and lithium (5 drop-outs). The patients subjectively felt more satisfied with carbamazepine than with lithium. Other studies helped to shed light on important clinical questions. Maj (1988), in a prospective and naturalistic study with 33 patients, found that two-thirds of mainly affective patients had no relapse during follow-up (two years) in contrast to only 30 percent in the group of mainly schizophrenic SAD participants. In accordance with the impressions from the lithium studies in SAD acute treatment, in

255

Biological treatment of schizoaffective disorders

Maj’s study schizodepressive patients (n19) had a significantly worse outcome than the group of schizobipolar patients (n14). In the study with the longest observation period in this field (6.8 years), we reported on 49 SAD patients according to ICD-10, 41 of whom were lithium patients, and 8 of whom were treated with carbamazepine (Baethge et al., 2004). We found that in this sub-group of patients lithium treatment was remarkably effective over the long run, and also well tolerated. However, compared with bipolar or unipolar depressed patients from the same center (Baethge et al., 2003b, c), SAD patients had a worse long-term outcome with considerable need for comedication. No predictors of treatment failure or success were found. Sex, age, duration of illness, polarity, illness severity, and family history were not related to outcome. Most samples comprised both SAD and bipolar patients and comparisons between both groups were reported in eight studies: three found lithium to be more effective in SAD patients; the remaining five came to the opposite conclusion. Evidence for a diminished efficiency of lithium in predominantly schizophrenic SAD derives from three publications (Mattes and Nayak, 1984; Kuefferle and Lenz, 1983; Maj, 1988). To the knowledge of the author, the opposite has not been shown so far. Anticonvulsants

Regarding the comparison of carbamazepine and lithium two controlled studies indicate, in general, similar efficacy in SAD (Bellaire et al., 1990; Greil et al., 1997). Both studies have been described in the previous section on lithium. It seems plausible to assume that carbamazepine, being as effective as lithium in two studies, would outperform placebo or would come out as effective in a mirror-image study. However, only Emrich’s very small study (n6) contains mirror-image results indicating carbamazepine’s effectiveness versus no treatment (Emrich, 1990). As for valproate, however, the published data are not sufficient to draw definite conclusions. Naturalistic studies reveal encouraging results for valproate as an addon medication (Puzynski and Klosiewicz, 1984; Emrich et al., 1985; Schaff et al., 1993; DelBello, 2000) but only very few patients undergoing monotherapy with valproate have been described in the literature (Hayes, 1989). There are no RCTs for valproate in SAD long-term treatment. In Emrich and coworkers’ small study (1985) the combination of lithium and valproate was less effective in SAD patients than in purely affective patients. Schaff et al. (1993), in a retrospective chart review of five SAD patients, reported that valproate add-on was favorable in four patients and poor in one patient. Anticonvulsants are increasingly used in SAD (Flynn et al., 2002) but only little research on their efficacy has been published. For lamotrigine, gabapentin, and topiramate no studies were found that met the inclusion criteria of this overview.

256

Christopher Baethge

Antipsychotics

Similar to the situation in SAD acute treatment already outlined, many studies on antipsychotic long-term treatment in schizophrenia have included patients suffering from SAD. However, only very few report on the SAD patients in particular. Although many new antipsychotics have been introduced, this situation has not changed much during the last few years. In the systematic review of SAD long-term treatment cited above (Baethge, 2003a), I included two open studies, both on clozapine, carried out before modern diagnostic criteria came into use, and eleven studies that were using criteria from RDC on. The most important of those publications is Mattes and Nayak’s (1984) comparison of lithium and fluphenazine showing that the antipsychotic was superior in mainly schizophrenic SAD patients. This study has been described in the section on lithium. Of the other trials five studied clozapine (McElroy et al., 1991; Banov et al., 1994; Frankenburg and Zanarini, 1994; Zarate et al., 1995; Ciapparelli et al., 2000), one risperidone (Vieta et al., 2001), one olanzapine (Narendran et al., 2001), one compared, in a double-blind design, haloperidol versus olanzapine (Tran et al., 1999), and two studies presented, among other data, results on antipsychotics as a group (Grossman et al., 1991; DelBello et al., 2000). In the meantime – to the knowledge of the author – no study has been added to the literature that would considerably change the conclusions of the systematic review (Baethge, 2003a). Under the assumption that results should be presented for the SAD group in particular, it is remarkable that the overall quality of antipsychotic studies in SAD is surprisingly poor. This is a distinct difference to the research situation in schizophrenia and bipolar disorder. Since none of the studies on antipsychotics was placebo controlled it is important to note that only one observational study (Banov et al., 1994, on clozapine) referred to prophylactic efficacy proper: that is, compared the time of clozapine treatment with the time before. In the only RCT, the extension of an acute treatment trial with SAD patients according to DSM-III-R, Tran et al. (1999) compared olanzapine and haloperidol. This study had a high drop-out rate (54 patients of the 110 who entered the extension phase) and in a last-observation-carried-forward analysis the olanzapine group had better results in the Montgomery Asberg depression scale (no significant differences in BPRS and PANSS). However, in a retrospective analysis Narendran and coauthors (2001) found no positive effect of olanzapine add-on treatment in a group of 47 SAD patients (DSM-IV) who were followed up for 1.5 years (for discusssion, see Baethge, 2003b). As regards number of studies, clozapine is the best-studied substance for SAD long-term treatment (Baethge, 2003a). In the largest of those studies (Banov et al., 1994), raters blind to diagnosis and baseline information examined 81 DSM-III-R

257

Biological treatment of schizoaffective disorders

SAD participants (out of 193 patients of all psychoses included in the study) after, on average, 20 months. Most patients were prescribed comedication. Affective and schizoaffective patients had no better outcome than schizophrenic patients on the basis of patient satisfaction, CGI, and re-hospitalization rate but had higher social functioning. Interestingly, response was inversely correlated with depressive symptoms. Previously, McElroy et al. (1991) reported that, in a retrospective analysis, 22 out of their 25 SAD patients had shown a “moderate” or “marked” response to clozapine during roughly three years of treatment. Conclusions One of the most important results of this overview is that study quality is unsatisfactory and in long-term treatment even worse than in acute treatment. There are only very few RCTs in this field, and there is no RCT that compared the most promising substances in SAD treatment. For example, in long-term treatment there is demand for a study that compares, in a balanced sample of mainly affective and mainly schizophrenic SAD patients, a promising mood stabilizer (lithium or carbamazepine) with an antipsychotic. The non-existence of such a study might indicate that SAD has not been fully accepted as a nosologically separated disorder in psychiatry. This impression is also supported by the fact that SAD patients are only rarely analyzed separately in mixed samples of predominantly schizophrenic or bipolar study samples. In some treatment studies SAD patients have results very similar to bipolar or schizophrenic participants of the same study, whichever one might be the main diagnosis in a given sample (Baethge, 2003a). This observation might, if proven, call into question one pillar of the SAD concept – as far as one considers a distinct therapeutic pattern a prerequisite or at least a supporting feature of nosological concepts. However, only a psychopathologically balanced study as proposed above will support or reject this hypothesis. The unsatisfying research situation notwithstanding, I will try to summarize the results of this overview in order to present recommendations for clinical practice. Those recommendations have to be considered as highly preliminary. The basis of the recommendations is the result that psychopathology seems to predict treatment outcome: for example, in acute treatment mild excitation was associated with good response to lithium; in long-term treatment mainly affective schizodepression was associated with a better response to carbamazepine. Clinical recommendations

It is very important to correctly assess the psychopathology. As a result misdiagnoses might be avoided and patients who are better characterized as schizophrenic or bipolar can be treated as per the guidelines for schizophrenia or bipolar affective

258

Christopher Baethge Table 13.1. Guidelines for acute and long-term treatment of schizoaffective disorder on the basis of the studies reviewed in this overview. Note the importance of re-evaluating the diagnosis and subtyping schizoaffective disorder. Owing to the lack of sufficiently powered and designed RCTs in this field the recommendations have to be considered highly preliminary.

Condition Acute Episodes mainly affective episodes schizomanic (-bipolar) schizodepressive mainly schizophrenic schizomanic schizodepressive Long-term treatment mainly affective schizobipolar schizodepressive mainly schizophrenic schizobipolar schizodepressive

Treatment

lithium, antipsychotics (when highly excited) antidepressants, antipsychotics (olanzapine, risperidone, ziprasidone) antipsychotics antipsychotics, ECT

lithium, carbamazepine carbamazepine, lithium antipsychotics (olanzapine) antipsychotics

disorder. Patients with a secured diagnosis of SAD should then be subtyped. In principle, one should subtype according to two poles: mainly affective versus mainly schizophrenic and schizomanic (or, for illness course: bipolar) versus schizodepressed. For the long-term diagnosis it is important to keep in mind that SAD is a versatile disorder and the illness course – treated or untreated – can change considerably (Maj and Perris, 1990) (see Table 13.1). Acute episodes

For mainly affective schizomanic episodes lithium and antipsychotics are the best evaluated medications and should be the first line treatment. In highly excited schizomanic syndromes antipsychotics are superior to lithium. In mainly schizophrenic schizomania antipsychotics are probably the best choice, although there are only few data. A difficult to treat condition is the schizodepressed syndrome. Predominantly affective schizodepressed episodes might benefit from treatment with antidepressants or antipsychotics. Olanzapine’s, risperidone’s, and ziprasidone’s promising results with regard to schizodepression have to be viewed as preliminary with

259

Biological treatment of schizoaffective disorders

regard to schizoaffective disorder. Electroconvulsive therapy (ECT) has not been sufficiently studied in this sub-group. However, given its efficacy in major depression and in schizophrenia it might be a promising treatment option. Long-term treatment

First choice for predominantly affective schizobipolar patients is lithium. Carbamazepine was as effective as lithium in one RCT (and another underpowered RCT) but is less well studied. From one RCT, however, it appears that carbamazepine might be superior to lithium in the sub-group of mainly affective schizodepressive patients (in this study, according to RDC). A family history of bipolar disorder and pronounced suicidality also supports lithium. This, however, is an analogy to bipolar disorder (Passmore et al., 2003) and has not been shown for SAD. In patients with mainly schizophrenic SAD antipsychotics are the long-term treatment of choice and, in the absence of sufficient data, treatment decisions have to be made in analogy to schizophrenia. For SAD, olanzapine is the only drug that has been tested in a RCT in long-term treatment. From observational studies there are encouraging results for the use of clozapine in SAD long-term treatment. Combination treatment is inceasingly common (Kupfer et al., 2002). However, it has not been well studied (Baethge et al., 2005). For SAD, therefore, combinations should be deployed only after monotherapy fails. Combinations of lithium and haloperidol (Lowe and Batchelor, 1990) as well as lithium and carbamazepine (Bocchetta et al., 1997) have been reported to be safe and effective.

R E F E R E N C ES Alexander, P. E., van Kammen, D. P. and Bunney, W. E. (1979). Antipsychotic effects of lithium in schizophrenia. American Journal of Psychiatry, 136, 283–7. American Psychiatric Association (1987). Diagnostic and Statistical Manual of Mental Disorders, 3rd edn, revised (DSM-III-R). Washington, DC: American Psychiatric Association. American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th edn. (DSM-IV). Washington, DC: APA. Angst, J., Weis, P., Grof, P., Baastrup, C. and Schou, M. (1970). Lithium prophylaxis in recurrent affective disorders. British Journal of Psychiatry, 116, 604–13. Baastrup, P. C. and Schou, M. (1967). Lithium as a prophylactic agent. Archives of General Psychiatry, 16, 162–72. Baethge, C. (2003a). Long-term treatment of schizoaffective disorder: Review and recommendations. Pharmacopsychiatry, 36, 45–56. Baethge, C., Baldesarini, R. J., Mathiske-Schmidt, K. et al. (2005). Long-term combination therapy versus monotherapy with lithium and carbamazepine in 46 bipolar I patients. Journal of Clinical Psychiatry, 66, 174–82.

260

Christopher Baethge Baethge, C., Gruschka, P., Berghöfer, A. et al. (2004). Prophylaxis of schizoaffective disorder with lithium or carbamazepine: outcome after long-term follow-up. Journal of Affective Disorders, 79, 43–50. Baethge, C., Smolka, M., Gruschka, P. et al. (2003b). Does prophylaxis-delay in bipolar disorder influence outcome? Results from a long-term study of 147 patients. Acta Psychiatrica Scandinavica, 107, 260–7. Baethge, C., Gruschka, P., Smolka, M. N. et al. (2003c). Effectiveness and outcome predictors of long-term lithium prophylaxis in unipolar major depressive disorder. Journal of Psychiatry and Neuroscience, 28, 355–61. Baldessarini, R. J., Tondo, L. and Hennen, J. (2003). Lithium treatment and suicide risk in major affective disorders: update and new findings. Journal of Clinical Psychiatry, 64 (Suppl. 5), 44–52. Banov, M. D., Zarate, C. A., Tohen, M. et al. (1994). Clozapine therapy in refractory affective disorders: Polarity predicts response in long-term follow-up. Journal of Clinical Psychiatry, 55, 295–300. Basan, A. and Leucht, S. (2004). Valproate for schizophrenia. Cochrane database of systematic reviews (Online), 1, CD004028. Bellaire, W., Demisch, K. and Stoll, K. D. (1990). Carbamazepin vs. lithium. Munich Medicalische Wischenschrift, 132 (Suppl. 1), S82–6. Biederman, J., Lerner, Y. and Belmaker, R. (1979). Combination of lithium carbonate and haloperidol in schizoaffective disorder. A controlled study. Archives of General Psychiatry, 36, 327–33. Bocchetta, A., Chilotti, C., Severino, G., Ardau, R. and del Zompo, L. (1997). Carbamazepine augmentation in lithium-refractory bipolar patients: A prospective study on long-term prophylactic effectiveness. Journal of Clinical Psychopharmacology, 17, 92–6. Bogan, A. M., Brown, E. S. and Suppes, T. (2000). Efficacy of divalproex therapy for schizoaffective disorder. Journal of Clinical Psychopharmacology, 20, 520–2. Braden, W., Fink, E. B., Quals, C. B., Ho, C. K. and Samuels, W. O. (1982). Lithium and chlorpromazine in psychotic inpatients. Psychiatry Research, 7, 69–81. Brockington, I. F., Kendell, R. E., Kellet, J. M., Curry, H. and Wainwright, S. (1978). Trials of lithium, chlorpromazine and amitriptyline in schizoaffective patients. The British Journal of Psychiatry, 133, 162–8. Chengappa, K. N., Rathore, D., Levine, J. et al. (1999). Topiramate as add-on treatment for patients with bipolar mania. Bipolar Disorders, 1, 42–53. Ciapparelli, A., Dell’Osso, L., Pini, S. et al. (2000). Clozapine for treatment-refractory schizophrenia, schizoaffective disorder, and psychotic bipolar disorder: A 24-months naturalistic study. Journal of Clinical Psychiatry, 61, 329–34. DelBello, M. P., Lopez-Larson, M. P., Getz, G. E. and Strakowski, S. M. (2000). Treatment of schizoaffective disorder with divalproex sodium (letter). Schizophrenia Research, 46, 77–9. Dietrich, D. E., Kropp, S. and Emrich, H. M. (2003). Oxcarbazepin in der Behandlung affektiver und schizoaffektiver Erkrankungen. [Oxcarbazepine in the treatment of affective and schizoaffective disorders.] Fortschritte der Neurologie-Psychiatrie, 71, 255–64.

261

Biological treatment of schizoaffective disorders Emrich, H. M. (1990). Alternatives to lithium prophylaxis for affective and schizoaffective disorders. In Affective and Schizoaffective Disorders, ed. A. Marneros and M. T. Tsuang. Berlin: Springer-Verlag, pp. 262–73. Emrich, H. M., Dose, M. and von Zerssen, D. (1985). The use of sodium valproate, carbamazepine and oxcarbazepine in patients with affective disorders. Journal of Affective Disorders, 8, 243–50. Faraone, S. V., Blehar, M., Pepple, J. et al. (1996). Diagnostic accuracy and confusability analyses: an application to the diagnostic interview for genetic studies. Psychological Medicine, 26, 401–10. Flynn, J., Grieger, T. and Benedek, D. M. (2002). Pharmacologic treatment of hospitalized patients with schizoaffective disorder. Psychiatric Services, 53, 94–6. Frankenburg, F. R. and Zanarini, M. C. (1994). Uses of clozapine in nonschizophrenic patients. Harvard Review of Psychiatry, 2, 142–50. Frye, M. A and Post, R. M. (2005). Valproate. In Comprehensive Textbook of Psychiatry, 8th edn, ed. B. Sadock and V. Sadock. Philadelphia: Lippincott Williams & Wilkins, Volume 2, pp. 2756–66. Goodnick, P. J. and Meltzer, H. Y. (1984). Treatment of schizoaffective disorders. Schizophrenia Bulletin, 10, 30–48. Greil, W., Ludwig-Mayerhofer, W., Erazo, N. et al. (1997). Lithium vs. carbamazepine in the maintenance treatment of schizoaffective disorder: a randomised study. European Archives of Psychiatry and Clinical Neuroscience, 247, 42–50. Grossman, L. S., Harrow M., Goldberg, J. F. and Fichtner, C. G. (1991). Outcome of schizoaffective disorder at two long-term follow-ups: Comparison with outcome of schizophrenia and affective disorders. American Journal of Psychiatry, 148, 1359–65. Gupta, S., Masand, P. S., Frank, B. L., Lickwood, K. L., Keller, P. l. (2000). Topiramate in bipolar and schizoaffective disorders: weight loss and efficacy. Primary Care Companion to the Journal of Clinical Psychiatry, 2, 96–110. Hayes, S. G. (1989). Long-term use of valproate in primary psychiatric disorders. Journal of Clinical Psychiatry, 50, 35–9. Janicak, P. G., Keck, P. E. Jr., Davis, J. M. et al. (2001). A double-blind, randomized, prospective evaluation of the efficacy and safety of risperidone versus haloperidol in the treatment of schizoaffective disorder. Journal of Clinical Psychopharmacology, 21, 360–8. Johnson, G., Gershon, S., Burdock, E. I., Floyd, A. and Hekimian, L. (1971). Comparative effects of lithium and chlorpromazine in the treatment of acute manic states. British Journal of Psychiatry, 119, 267–76. Kasanin, J. (1933). The acute schizo-affective psychoses. American Journal of Psychiatry, 57, 41–8. Ketter, T. A. (2005). Gabapentin, lamotrigine, topiramate. In Comprehensive Textbook of Psychiatry, 8th edn, ed. B. Sadock and V. Sadock. Philadelphia: Lippincott Williams & Wilkins, Volume 2, pp. 2746–56. Keck, E. J. Jr., Reeves, K. R., Harrigan, E. P. and the Ziprasidone Study Group (2001). Ziprasidone in the short-term treatment of patients with schizoaffective disorder: Results from two doubleblind, placebo-controlled, multicenter studies. Journal of Clinincal Psychopharmacology, 21, 27–35.

262

Christopher Baethge Kramer, M. S., Vogel, W. H., DiJohnson, C. et al. (1989). Antidepressants in “depressed” schizophrenic inpatients: a controlled trial. Archives of General Psychiatry, 46, 922–8. Küfferle, B. and Lenz, G. (1983). Classification and course of schizo-affective psychoses. Psychiatrica Clinica, 16, 169–77. Kupfer, D. J., Frank, E., Grochochinsky, V. J. et al. (2002). Demographic and clinical characteristics of individuals in a bipolar disorder case registry. Journal of Clinical Psychiatry, 63, 120–5. Lerner, Y. and Mintzer, Y. (1988). Lithium combined with haloperidol in schizophrenia patients. British Journal of Psychiatry, 153, 359–62. Leucht, S., Kissling, W. and McGrath, J. (2004). Lithium for schizophrenia revisited: A systematic review and meta-analysis of randomized controlled trials. Journal of Clinical Psychiatry, 65(2), 177–86. Leucht, S., McGrath, J., White, P. and Kissling, W. (2003). Carbamazepine for schizophrenia and schizoaffective psychoses. Cochrane Database of Systematic Reviews (Online), 3, CD001258. Levinson, D. F., Umapathy, C. and Musthaq, M. (1999). Treatment of schizoaffective disorder and schizophrenia with mood symptoms. American Journal of Psychiatry, 156, 1138–48. Lieberman, J. A., Stroup, T. S., McEvoy, J. P. et al. (2005). Clinical antipsychotic trials of intervention effectiveness (CATIE) investigators. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. The New England Journal of Medicine, 353, 1209–23. Lowe, M. R. and Batchelor, D. H. (1990). Lithium and neuroleptics in the management of manic depressive psychosis. Human Psychopharmacology, 5, 267–74. Maj, M. (1988). Lithium prophylaxis of schizoaffective disorders: a prospective study. Journal of Affective Disorders, 14, 129–35. Maj, M. and Perris, C. (1990). Patterns of course in patients with a cross-sectional diagnosis of schizoaffective disorder. Journal of Affective Disorders, 20, 71–7. Maj, M., Pirozzi, R., Formicula, R. M., Bartoli, L. and Bucci, P. (2000). Reliability and validity of the DSM-IV diagnostic category of schizoaffective disorder: Preliminary data. Journal of Affective Disorders, 57, 95–8. Mattes, J. A. and Nayak, D. (1984). Lithium versus fluphenazine for prophylaxis in mainly schizophrenic schizo-affectives. Biological Psychiatry, 19, 445–9. McElroy, S. L., Dessain, E. C., Pope, H. G. et al. (1991). Clozapine in the treatment of psychotic mood disorders, schizoaffective disorder, and schizophrenia. Journal of Clinical Psychiatry, 52, 411–14. Mensink, G. J. R. and Slooff, C. J. (2004). Novel antipsychotics in bipolar and schizoaffective mania. Acta Psychiatrica Scandinavica, 109, 405–19. Moseman, S. E., Feeman, M. P., Misiaszek, J. and Gelenberg, A. J. (2003). Mood stabilizers. In Psychiatry, 2nd edn, ed. A. Tasman, J. Kay and J. A. Lieberman. Chichester: John Wiley & Sons, Volume 2, pp.1965–89. Müller-Oerlinghausen, B., Felber, W., Berghöfer, A., Lauterbach, E. and Ahrens, B. (2005). The impact of lithium long-term medication on suicidal behavior and mortality of bipolar patients. Archives of Suicide Research, 9, 307–19. Mulsant, B. H., Sweet, R. A., Rosen, J. et al. (2001). A double-blind randomized comparison of nortriptyline plus perphenazine versus nortiptyline veresus placebo in the treatment of psychotic depression in late life. Journal of Clinical Psychiatry, 62, 597–604.

263

Biological treatment of schizoaffective disorders Narendran, R., Young, C. M., Valenti, A. M. et al. (2001). Olanzapine therapy in treatmentresistant psychotic mood disorders: A long-term follow-up study. Journal of Clinical Psychiatry, 62, 509–16. Okuma, T., Yamashita, I., Takahashi, R. et al. (1989). A double-blind study of adjunctive carbamazepine versus placebo on excited states of schizophrenic and schizoaffective disorders. Acta Psychiatrica Scandinavica, 80, 250–9. Passmore, M. J., Garnham, J., Duffy, A. et al. (2003). Phenotypic spectra of bipolar disorder in responders to lithium versus lamotrigine. Bipolar Disorders, 5, 110–14. Prien, R. F., Caffey, Jr. E. M. and Klett, C. J. (1972). A comparison of lithium carbonate and chlorpromazine in the treatment of excited schizo-affectives. Archives of General Psychiatry, 27, 182–9. Puzynski, S. and Klosiewicz, L. (1984). Valproic acid amide in the treatment of affective and schizoaffective disorders. Journal of Affective Disorders, 6, 115–21. Schaff, M., Fawcett, J. and Zajecka, J. M. (1993). Divalproex sodium in the treatment of refractory affective disorders. Journal of Clinical Psychiatry, 54, 380–4. Schatzberg, A. F. and Nemeroff, B. (2004). The American Psychiatric Publishing Textbook of Psychopharmacology, 3rd edn. Washington, DC: American Psychiatric Publishing. Small, J. G., Klapper, M. H., Malloy, F. W. and Steadman, T. M. (2003). Tolerability and efficacy of clozapine combined with lithium in schizophrenia and schizoaffective disorder. Journal of Clinical Psychopharmacology, 23, 223–8. Spiker, D. G., Weiss, J. C., Dealy, R. S. et al. (1985). The pharmacological treatment of delusional depression. American Journal of Psychiatry, 142, 430–6. Spitzer, R. L., Endicott, J. and Robins, E. (1977). Research Diagnostic Criteria for a Selected Group of Functional Disorders, 3rd edn. New York: New York State Psychiatric Institute. Tohen, M., Zhang, F., Keck, P. E. Jr. et al. (2001). Olanzapine versus haloperidol in schizoaffective disorder, bipolar type. Journal of Affective Disorders, 67, 133–40. Tran, P. V., Tollefson, G. D., Sanger, T. M. et al. (1999). Olanzapine versus haloperidol in the treatment of schizoaffective disorder. British Journal of Psychiatry, 4, 15–22. Vieta, E., Goikolea, J. M., Corbella, B. et al. (2001). Risperidone safety and efficacy in the treatment of bipolar and schizoaffective disorders: Results from a 6-month, multicenter, open study. Journal of Clinical Psychiatry, 62, 818–25. Volavka, J., Czobor, P., Sheitman, B. et al. (2002). Clozapine, olanzapine, risperidone, and haloperidol in the treatment of patients with chronic schizophrenia and schizoaffective disorder. American Journal of Psychiatry, 159, 255–62. Whitehead, C., Moss, S., Cardno, A. and Lewis, G. (2002). Antidepressants for people with both schizophrenia and depression. Cochrane Database of Systematic Reviews (Online), 2, CD002305. World Health Organization (1992). The ICD-10 Classification of Mental and Behavioural Disorders. Clinical Descriptions and Diagnostic Guidelines. Geneva: WHO. Zarate, C. A., Tohen, M., Banov, M. D., Weiss, M. K. and Cole, J. O. (1995). Is clozapine a mood stabilizer? Journal of Clinical Psychiatry, 56, 108–12.

14

Psychological therapies and schizoaffective disorders Jan Scott Department of Psychological Medicine, Institute of Psychiatry, London

Introduction This chapter explores psychological therapies in schizoaffective disorders. In order to provide a coherent overview the focus will be exclusively on cognitive behavior therapy (CBT). This approach is the most widely investigated adjunctive therapy for severe mental disorders and there have been large-scale randomized controlled treatment trials (RCTs) in acute and chronic schizophrenia, acute and chronic depression, and in moderately severe and highly recurrent and complex bipolar disorders (for a review, see Roth and Fonagy, 1996). As such, a review of the characteristics of this model – and how this can be applied across the spectrum of severe mood and psychotic disorders – can inform thinking on the benefits and drawbacks of dimensional versus categorical approaches to diagnosis. Interestingly, there are only two small-scale contemporary studies of psychological therapy for schizoaffective disorders (Garety et al., 1994; Theilemann, 1993), so much of the chapter will draw on the experiences reported from the clinical manuals of CBT and the RCTs that have included patients with co-existing psychotic and mood symptoms. These patients are more likely to be found in the RCTs of schizophrenia, as the presence of psychotic symptoms is frequently an exclusion criterion in RCTs of CBT in unipolar or bipolar disorders. Depressed mood is a common accompaniment of all phases of schizophrenia and is often one of the primary outcome measures in RCTs of CBT (Fowler et al., 1995). Hyperactive episodes can also occur in schizophrenia, and although not so frequently monitored systematically in RCTs, such features have not led to exclusion of participants from trials.

The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

264

265

Psychological therapies and schizoaffective disorders

The increasing acceptance of cognitive behavior therapy Until recently, severe mood and psychotic disorders were regarded as biological illnesses best treated with medications (Prien and Potter, 1990; Scott, 1995a; Sensky et al., 2000). This approach is gradually changing for two reasons. First, in the past three decades, there has been a greater emphasis on stress-diathesis models. This has led to the development of new etiological theories that recognize that genetic models offer necessary but not sufficient theories of onset and also emphasize psychological and social aspects of vulnerability and risk. It has also increased the acceptance of brief psychological therapies, such as CBT, as an adjunct to medication for individuals with medication-refractory schizophrenia, and severe and chronic affective disorders (Scott and Wright, 1997). Second, there is a significant efficacy-effectiveness gap for pharmacological treatments targeting these patient populations. For example, mood stabilizer prophylaxis protects about 60% of individuals against relapse in research settings, but protects only 25–40% of individuals against further episodes in clinical settings (Dickson and Kendell, 1986). Similarly, despite research evidence of significant benefits of pharmacotherapy, 30% of individuals with depressive disorder are still symptomatic after two years of treatment with antidepressant medication and 50% of individuals with schizophrenia experience a further relapse over a five-year period. The introduction of newer medications has not improved prognosis (Scott, 1995b). This less impressive symptomatic and functional outcome in day-to-day practice has increased interest in non-pharmacological treatment approaches. Cognitive models have the advantage of allowing the clinician to focus on the unique experiences of an individual patient. While diagnosis is part of the assessment, they tend to emphasize the specific characteristics of a syndrome that differentiate it from another disorder; as such they may have less utility when patients present with features that overlap a number of syndromes. The cognitive model allows the clinician to produce a coherent representation of the patients’ experiences and draw up a logical treatment plan. Differences in the approach to similar cases may be related to the relative prominence of certain symptom clusters or the degree of distress or impairment caused. Some patients may show more prominent changes in mood (depression or elation) and psychomotor activity (poverty or excitation), suggesting the initial therapy focus should be on the mood disorder elements, whilst others may show psychomotor poverty accompanied by more prominent disorganization and reality distortion suggesting behavioral targeting of activity and cognitive targeting of psychotic symptoms would be more justified. However, the underlying cognitive model suggests that a similar explanation can be given for the maintenance of psychotic or affective features in mood, schizoaffective and schizophrenic disorders.

266

Jan Scott

The basic elements of the cognitive model Beck’s original cognitive model (now referred to as the linear processing model) suggests that emotional responses to events are determined by the conscious meaning the individual gives to that experience (Beck, 1976; Beck et al., 1979). This notion differs from the psychoanalytic approach, which emphasizes subconscious drives, and from the behavioral model of environmental determinism. Thus, the model emphasized that emotional reactions were a response to conscious processing of material and that once an abnormal mood state develops the processing of information may be biased owing to selective attention etc. This model also acknowledges that such reactions and processes occur in all individuals. Therefore, mental disorders and even psychotic experiences would be seen as part of a continuum. Beck hypothesized that cognitive vulnerability to mental disorders centers on prepotent dysfunctional underlying beliefs (e.g., “I am inadequate”) that develop from early learning experiences, and drive thinking and behavior. Those who have adaptive beliefs do not experience the extreme or abnormally persistent responses to events that may characterize mental disorders. It is suggested that individuals with a belief that they are inadequate or weak will be more likely to attend to evidence that supports this notion and may also tend to blame themselves for negative experiences, leading to depressed mood. Others may still develop a cognitive style where they tend to externalize power or control, increasing the likelihood that they will blame others and/or experience paranoid reactions when stressed. It is hypothesized that life events that have a specific meaning for that individual may activate these beliefs. For example, an individual who experiences neglect or abuse in childhood may hold a belief that “I’m unlovable.” As such, they may later experience depressed mood in the face of rejection by a significant other. Beck’s model also suggests that mood states are accentuated by patterns of thinking that amplify the initial mood shift. If individuals become depressed they then become more negative in how they see themselves, their world and their future (called the negative cognitive triad). Their immediate thoughts show a number of key information-processing biases, and they jump to negative conclusions, overgeneralize from one event to another, interpret situations in all-or-nothing terms, and self-blame to an excessive degree. This further depresses their mood and leads to a further cascade of negative thoughts (the vicious cycle). Finally, the changes in behavior that are recognized as symptoms of mental disorders (such as avoidance of social interaction, inactivity, hypervigilance etc.) may be regarded as a cause or a consequence of mood shifts and dysfunctional thinking. Beck’s later publications on cognitive models highlight the importance of biological factors in the onset of illness episodes, often seeing the cognitive distortions as factors that maintain the psychotic or affective symptoms (Beck, 1983; Beck,

267

Psychological therapies and schizoaffective disorders

1996). In addition, work on attributional style has established that individuals who experience delusions “jump to conclusions” when drawing inferences, underutilize disconfirming data, and tend to attribute negative events to external causes (Morrison, 1998). The classical cognitive characteristics of a delusion are that it is a culturally unacceptable explanation of an experience that attributes the cause of the experience outside the individual (Morrison, 2002). Delusions should therefore be amenable to structured reasoning and behavioral approaches (Scott et al., 1993; Kingdon and Turkington, 1994). In contrast, many individuals who experience persistent affective disturbances make internal, self-blaming attributions for negative events, which may then promote depressive reactions. Individuals with bipolar disorders show a cognitive style similar to those with unipolar disorders, but those who develop hypomania or mania often misattribute the actual symptoms they experience to other aspects of themselves, e.g., a man who had undertaken a longhaul flight was experiencing jet lag and the disruption to his circadian rhythms precipitated the prodromal symptoms of hypomania. However, rather than attributing his sleep disturbance, increased restlessness and energy to a bipolar relapse, he interpreted them as evidence that he was “a chief executive in the making, full of ideas, drive and determination” and took the opportunity to work through the night leading to a further escalation in his symptoms. Using cognitive behavior therapy in schizophrenia and mood disorders The basic premise of CBT in schizophrenia and bipolar disorders does not differ from its use in anxiety, depression, and other mental disorders. The same is true for schizoaffective disorders. The key characteristics of the therapy are the same as shown in Table 14.1. Cognitive behavior therapy focuses on understanding the links between events, thoughts, feelings, and behavior. For example, Mr. A develops the belief that the police are harassing him; he becomes frightened and depressed and isolates himself. Because of his isolation, he becomes more depressed and is unable to check out with other people the evidence for his belief and his preoccupation. His low mood leads him to negatively distort his assessment of evidence, setting up a self-perpetuating cycle. Therapeutic interventions can address each of these areas, but an immediate attempt to tackle Mr. A’s beliefs may well be counterproductive. Persons with psychosis often anticipate skepticism – they expect the psychiatrist or mental health professional to assume they are “mad.” So the initial approach needs to be one that actively encourages the persons to detail “their side of the story”. How did they arrive at the conclusions (i.e., delusional beliefs) that they now hold so firmly? What were the circumstances that prevailed at the time? What were the events in their lives – sometimes apparently innocuous – that led to the emergence of their beliefs?

268

Jan Scott Table 14.1. Key characteristics of cognitive behavior therapy.

The therapy offers a specific formulation of the individual’s problems. The model of therapy is shared openly with the patient. There is rational use of techniques in a logical sequence. There is an emphasis on skill development and transfer of learning outside of therapy sessions. Change is attributed to the patient’s rather than the therapist’s efforts.

Differences in using CBT with patients with psychosis are generally a matter of emphasis, in addition to some specific techniques to supplement those used for anxiety and unipolar depression. Agenda-setting may need to be more flexible, simple, and sometimes less explicit; and the length and frequency of sessions may also need to vary. For example, if the patient is agitated or is having problems with concentration, brief 15–20 minute sessions may be most useful. Occasionally during the engagement and assessment period, sessions even longer than an hour can be fruitful in exploring beliefs constituting a complex delusional system. A typical session with a patient with schizophrenia or severe mood disorder lasts from 30 to 45 minutes. When patients become unsettled in sessions, the therapist usually will discuss terminating the session with them and/or may move to discuss less emotive topics unless the patient is particularly keen to pursue the topic (Scott et al., 1993). The number of sessions offered is usually 20 or more (Scott and Wright, 1997). The assessment process involves the active use of guided discovery, usually with the therapist prompting but, as far as possible, the patient actively participating. The assessment phase of CBT seeks to develop an understanding of vulnerability factors (e.g., genetic predisposition, personality factors) and precipitants (e.g., illicit drug use exacerbating psychotic symptoms, stress-related incident, or sometimes just a casual critical comment). A full psychiatric history and examination eliciting predisposing, precipitating, perpetuating and protective factors provides the key information on which to base a formulation that forms the template for the use of CBT. The dimensional nature of hallucinations and delusions are explored, e.g., degree of distress, conviction, preoccupation, interference with daily living and the explanations that the patient has for these symptoms. The first episode is particularly discussed as it frequently sheds light on current symptoms. The content of delusions and hallucinations often becomes understandable when the circumstances prevailing prior to the first episode are clarified. The contribution of any available medical information, care-givers and others who may have knowledge of the early period can be invaluable in providing information with which to prompt the patient’s memory. The assessment also promotes engagement as it establishes a working alliance. Many patients will be enthusiastic about discussing

269

Psychological therapies and schizoaffective disorders

their beliefs or talking about their voices, especially when this has not been encouraged in previous treatment efforts. However, some patients find engaging in CBT to be more difficult and a gentle non-directive supportive approach may need to extend over several weeks to establish trust. The development of an individualized formulation begins from the start of the assessment procedure. The information gathered is assembled into a preliminary formulation which in persons with psychosis needs to give sufficient emphasis to personal history and, especially, the initial episode. Discussing events that the patient finds perplexing is a useful way in which to begin to develop the therapeutic relationship (Morrison, 2002). It is then possible to re-examine the information with the patient and look at ways of testing whether, or to what degree, the statements or beliefs are accurate. Where mood symptoms are particularly prominent the therapist may initially focus on achieving some degree of emotional stabilization; with hypomanic symptoms this would be achieved through regularizing daily activities and the sleep–wake cycle and managing exposure to stressful or overstimulating environments. In severe depression, cognitive and behavioral techniques may be used to provide rapid symptom relief from the first session and key problems such as high levels of agitation, insomnia, and difficulties with concentration, or profound hopelessness may be targeted. When mood improvement is achieved, thoughts and beliefs may be explored. For example, it emerged that Mr. A had developed his belief about the police after being interviewed twice by them as part of their routine inquiries. He had started to become suspicious that they were harassing him and then reported these suspicions to a family member. This person was sufficiently concerned to inform a doctor. The patient then refused to talk to the doctor and became angry and hostile accusing “everyone . . . [of] . . . being out to get me. . .” The situation deteriorated and eventually led to Mr. A’s involuntary admission to a hospital under police escort. Nine years later, in a CBT session, he was able to review these events and gradually change his explanation for the meaning of these events. The patient began to entertain the possibility that he might have over-reacted to the initial circumstances. This allowed him to start to alter his behavior – he began going out from the hostel where he was living, having become quite reclusive, and eventually he was able to lead a more independent life. Initially, Mr. A only marginally modified the degree of conviction in his beliefs about what had occurred, but his behavior changed much more substantially; and with this change he engaged in more interactions and activities outside the hostel which helped him identify further evidence, essentially the absence of police “harassing me,” that undermined his earlier assumptions that there was a police conspiracy against him. It is essential to understand the meaning of the beliefs and hallucinations to the individual. This process involves gathering detailed information, as mentioned

270

Jan Scott

above, about the initial episode and its antecedents. These circumstances often involve one of, at least, four scenarios (Scott et al, 2004): (a) Gradual deterioration of functioning and onset of symptoms, but often presenting for initial treatment with an acute episode. The patient may have oversensitivity to stressful circumstances leading to pronounced negative symptoms. (b) Distressing, condemnatory hallucinations (e.g., “you are evil”) which are related to past traumatic experiences, as a prominent feature. (c) Drug-precipitated initial psychotic experience with persisting or recurrent relapses or symptoms occurring, even in the absence of hallucinogens. (d) Anxiety or depressive symptoms developing out of stressful circumstances (where there is a belief that something significant is about to happen) and then, often abruptly, crystallizing into a delusional belief which provides “meaning” to their experiences and may then become entrenched and systematized. In individuals who show manic affect, the beliefs and ideas that need to be identified relate to their misattribution of their symptoms to their “positive personality” or “strength of character” rather than recognizing and attributing their change in energy, sleep and speed of thought to the onset of an episode of illness. Some patients jump to conclusions – forming delusions involving a high level of conviction on the basis of minimal evidence. Kingdon and Turkington (1994) describe a CBT method for delusions that initially involves engagement, building trust, assessment and formulation. From this, the links between events and circumstances, and the feelings, behavior, and beliefs that emerged are examined. A range of explanations, including the delusional ones, chosen by the patient is discussed. Then reality testing, involving examination of evidence, logical inquiry and reasoning, is used – as far as possible led by or elicited from the patient. For example, if a patient expressed the belief that there is a silicon chip in his brain, the discussion might focus on why the patient believed it was there, how it had got there, why someone would put it there, what its function could be, how it worked, what power source it has, etc. (On one occasion, a patient had a calcified pineal gland, which had been seen on X-ray and not explained to her in a way she understood, so a delusional explanation about a brain implant developed.) Feelings of depression and hopelessness often accompany the belief that external and powerful forces are controlling the individual. Hence developing alternative explanations may also help resolve affective symptoms. The reasoning process in CBT may allow the patient to reconsider beliefs but, more commonly, assists in developing a therapeutic relationship, improving compliance with services and medication and allowing behavior to change (e.g., often patients begin to socialize more). At this stage, underlying concerns may emerge, such as past traumatic events, or unfulfilled goals such as to have a career or an

271

Psychological therapies and schizoaffective disorders

intimate relationship. Core beliefs (schemas) about themselves may also come to the surface or be specifically elicited by the use of a variety of cognitive therapy techniques. Exploration of traumas is an area where caution is needed – emotional arousal is well recognized as a precipitant of psychotic symptoms and so can worsen rather than assist and can lead to disengagement from therapy. Where distressing events or memories emerge, work on associated beliefs (e.g., “I’m to blame”) rather than the most distressing belief may be more possible. However the main focus should be on reducing immediate distress. Working with delusional beliefs therefore involves structured, consistent, collaborative discussion that is non-confrontational and aimed at allowing the person to review their beliefs, possible alternatives, and their subsequent behavior. Again it is clear that alleviation of depression or reduction in arousal or activation level is important to allow this work to be undertaken in those with schizoaffective disorders. Beliefs rarely change immediately in therapy sessions but may do so over time. Frequently, the process of jointly examining beliefs assists with engagement; then other underlying issues (e.g., isolation, poor self-esteem, or specific distressing events) may emerge and become targets for therapy. These themes, particularly low self-esteem, are commonly observed in bipolar disorders as well as unipolar disorders and are also a prominent feature of other psychoses. However, the patient may not be fully aware of this or may resist working with them until he or she is more engaged in therapy. Sometimes an inference-chaining approach (asking a series of linked questions to get to the core of an issue) can uncover such issues. (E.g., in a patient who believed, wrongly, that she was a member of the French aristocracy, the approach taken was: “We probably need to agree to differ on whether you are a member of the French aristocracy, but if people did believe you, why would that be important to you?”). In answering such questions, patients often describe being respected or no longer being alone; in this case, she did not feel respected. So she was asked: “Who is it that you think does not respect you?” The response to such a question may be general – “everyone” – but often is more specific – e.g., in this instance, “My father doesn’t respect me.” She can then be asked: “Why do you think he doesn’t?” “Is there anything you can do about it?” or “How do you feel when you think that of him?” The person will often allow the therapist to work directly with material that relates to the delusional belief, in this case being respected by her father, but which is only accessible by using a process that includes full discussion of it. Techniques that may be helpful for hallucinations include reattribution (improving insight into hallucinations as being internally generated and owing to a “chemical imbalance”), development of specific coping strategies, and modification of beliefs about the omnipotence of voices and their content. Early

272

Jan Scott

Assessment

Engagement

Explanation of antecedents

Formulation

Techniques for delusions Socratic questioning Reduction of affective investment Reality testing Inference chaining/schema work Techniques for hallucinations Analysis and reattribution of origin Alternative hypothesis Diary keeping and distraction Anxiety reduction and coping skills enhancement

Techniques for mood symptoms Depression Activity scheduling Mastery and pleasure ratings Detection and modification of negative automatic thoughts Hypomania Self-regulation and stabilization of social rhythms Behavioral interventions to reduce arousal and activation Refocus thinking on negative aspects and potential damaging consequences of plans

Medication adherence as a coping strategy

Enhancing self-esteem

Relapse prevention strategies

Figure 14.1. Cognitive behavior therapy in schizoaffective disorders (based on Turkington and Kingdon, 1999).

studies showed symptomatic improvements with both focusing on hallucinations and distraction techniques (Fowler et al., 1995). More recent studies emphasize that assessment and formulation are vital to help understand the development and impact of hallucinations on the person’s life. Key critical events in childhood and adolescence may have established the hallucinatory experiences, although sometimes it is years later that they become sufficiently distressing and disabling to lead to psychiatric presentation. Focus on beliefs about the events – usually related to shame, anger, and fear – rather than the events themselves may be the only effective

273

Psychological therapies and schizoaffective disorders

way to work with them. Such work may be more effectively performed after reattribution of the source of the voices has occurred. This strategy involves looking at beliefs the person has about voices. This process is relatively non-threatening to most patients and can form part of the assessment of the hallucinations. It is helpful for patient and therapist to agree about the characteristics of the hallucinations. Are they perceived as speech (as they generally are) “like me speaking to you now” (maybe louder or quieter, etc.) or visions – “as if seen on a film/TV or someone walking in front of you?” Then ‘“Does anyone else hear them? If not, why not?” “How do they do it? How do they communicate directly with you?” This type of questioning often allows alternative, normalizing explanations to be introduced. Drawing these different elements together, Turkington and Kingdon (1999) have drawn up a template for CBT interventions in schizoaffective disorders (see Figure 14.1). Outcome research There is considerable evidence that CBT benefits individuals with severe mental disorders in the short term and also leads to sustained health gains over 12–48 month follow-up. For example, Tarrier et al. (1993), and Sensky et al. (2000) demonstrate that CBT may reduce positive symptoms of schizophrenia but also significantly reduce negative symptoms and depressive features. There have been two large-scale randomized controlled trials of CBT in chronic depressive disorders which included patients who had or previously experienced melancholic or endogenous disorders (Paykel et al., 1999; Keller et al., 2000) and DeRubeis et al. (1999) and Thase et al. (1997) offer important meta-analyses of the use of combined pharmacotherapy and CBT in severe and chronic depressions. These studies suggest CBT is an effective treatment for severe depression, but very few studies included subjects with psychotic depression and only a few papers give details of such an approach (Scott, 1988; Scott, 2000). Outcome studies of CBT in bipolar disorders suggest the most robust effect is on the prevention of relapse (Lam et al., 2003; Scott et al., 2005), rather than being indicated in the acute stages, and there are problems in using standard CBT techniques when the content of delusions may change frequently. However, the two small-scale studies in schizoaffective disorders suggest that the therapy can be beneficial and there is therefore a case for more studies in this area (Theilemann, 1993; Garety et al., 1994). Conclusions Cognitive behavior therapy has specific characteristics that may benefit patients with complex and severe presentations. Its collaborative, educational style and its use of a

274

Jan Scott

step-by-step approach and of guided discovery make it acceptable to individuals who wish to take an equal and active role in their therapy (Beck et al., 1979; Scott, 1995a). Many individuals with severe disorders resist and challenge a didactic approach to treatment (Miklowitz and Goldstein, 1990). Because the individuals play an active role in developing the formulation of their own beliefs and problems, the interventions used appear rational and logical, giving CBT a sense of coherence. The structured approach to each session with agenda setting, prioritization of problems for discussion and joint development of homework tasks enables patients to retain their focus on specific topics or issues even when exhibiting psychotic symptoms (Kingdon and Turkington, 1994). The structure enables individuals with hypomanic symptoms to retain their focus on the session despite being distractible (Scott, 1995a). Clients with depression also cope better with a structured approach as this tends to contain their sense of hopelessness and helplessness, which might otherwise lead to such overwhelming levels of anxiety that the individual is unable to engage in therapy. Cognitive behavior therapy can also encourage skills development and an increased sense of self-efficacy and control. These features of CBT may be particularly helpful to individuals who experience low self-esteem and perceive a loss of identity because they are viewed as a person with a severe or chronic mental disorder. For researchers in schizoaffective disorders, cognitive models offer an informative approach to exploring the dimensions and phenomenology of this disorder through individual case formulations. This is particularly useful as it allows clinicians to explore the psychological mechanisms and determinants of thought content that are shared across the spectrum of mood disorders and schizophrenia and provides data on trans-diagnostic processes, as well as a new perspective on what may be the underlying factors that will more clearly distinguish between these syndromes.

R E F E R E N C ES Beck, A. T. (1976). Cognitive Therapy and the Emotional Disorders. Madison: International Universities Press. Beck, A. T. (1983). Cognitive therapy for depression: new perspectives. In Treatment of Depression: Old Controversies and New Approaches, ed. P. J. Clayton and J. E. Barrett. New York: Raven Press, pp. 58–74. Beck, A. T. (1996). Beyond belief: a theory of modes, personality and psychopathology. In Frontiers of Cognitive Therapy, ed. P. Salkovskis. London: Guilford Press, pp. 12–33. Beck, A. T., Rush, A. J., Shaw, B. F. and Emery, D. (1979). Cognitive Therapy of Depression. New York: Guilford Press. DeRubeis, R. J., Gelfand, L. A., Tang, T. Z. and Simons, A. D. (1999). Medications versus cognitive behaviour therapy for severely depressed outpatients: mega-analysis of four randomised comparisons. American Journal of Psychiatry, 156, 1007–13.

275

Psychological therapies and schizoaffective disorders Dickson, W. E. and Kendell, R. E. (1986). Does maintenance lithium therapy prevent recurrences of mania under ordinary clinical conditions? Psychological Medicine, 16, 521–30. Fowler, D., Garety, P. and Kuipers, L. (1995). Cognitive Therapy of Psychoses. Chichester, UK: Wiley. Garety, P. A., Kuipers, L., Fowler, D., Chamberlain, F. and Dunn, G. (1994). Cognitive behavioural therapy for drug-resistant psychosis. The British Journal of Medical Psychology, 67, 259–71. Keller, M., McCullough, J., Klein, D. et al. (2000). The acute treatment of chronic depression: A comparison of nefazadone, cognitive behavioural analysis system of psychotherapy, and their combination. New England Journal of Medicine, 342, 1462–70. Kingdon, D. and Turkington, D. (1994). Cognitive Behavioural Therapy of Schizophrenia. London: Guilford Press. Lam, D. H., Bright, J., Jones, S. et al. (2003). Cognitive therapy for bipolar illness. Archives of General Psychiatry, 60, 145–52. Miklowitz, D. and Goldstein, M. (1990). Behavioural family treatment for patients with bipolar affective disorder. Behaviour Modification, 14, 457–89. Morrison, P. (2002). A Casebook of Cognitive Therapy in Psychosis. New York: Taylor & Francis. Morrison, T. (1998). Cognitive behaviour therapy for psychotic symptoms in schizophrenia. In Treating Complex Cases, ed. N. Tarrier, A. Wells and G. Haddock. Chichester: John Wiley & Sons, pp. 27–41. Paykel, E., Scott, J., Teasdale, J. et al. (1999). Prevention of relapse in residual depression by cognitive therapy: a controlled trial. Archives of General Psychiatry, 56, 829–35. Prien, R., and Potter, W. (1990). NIMH workshop report on the treatment of bipolar disorders. Psychopharmacology Bulletin, 26, 409–27. Roth, A. and Fonagy, P. (1996). What Works for Whom? A Critical Review of Psychotherapy Research. London: The Guilford Press. Scott, J. (1988). Chronic depression. British Journal of Psychiatry, 153, 287–97. Scott, J. (1995a). Psychotherapy for dipolar disorder: an unmet need? British Journal of Psychiatry, 167, 581–8. Scott, J. (1995b). Psychological treatments of depression: an update. British Journal of Psychiatry, 167, 289–92. Scott, J. (2000). New evidence in the treatment of chronic depression. New England Journal of Medicine, 342, 1518–20. Scott, J., Byers, S. and Turkington, D. (1993). The chronic patient. In Cognitive Therapy with Inpatients, ed. J. Wright, M. Thase, A. T. Beck and J. Ludgate. New York: Guildford Press, pp. 107–34. Scott, J., Kingdon, D. and Turkington, D. (2004). Cognitive therapy with schizophrenia. In Review of Psychiatry, Volume 20, ed. A. Frances and R. Hales. Washington, DC: American Psychiatric Association Press, pp. 61–88. Scott, J., Paykel, E., Morris, R. et al. (2006). Cognitive therapy for severe and recurrent bipolar disorder. British Journal of Psychiatry, 188, 313–32. Scott, J., and Wright, J. (1997). Cognitive therapy with severe and chronic mental disorders. In Review of Psychiatry. Volume 16, ed. A. Frances and R. Hales. Washington, DC: American Psychiatric Association Press, pp. 321–47.

276

Jan Scott Sensky, T., Turkington, D., Kingdon, D. et al. (2000). A randomised controlled trial of cognitive behavioural therapy for persistent symptoms in schizophrenia resistant to medication. Archives of General Psychiatry, 57, 165–72. Tarrier, N., Beckett, R., Harwood, S. et al. (1993). A trial of two cognitive behavioural methods of treating drug resistant residual psychotic symptoms in schizophrenic patients: I–Outcome. British Journal of Psychiatry, 162, 524–32. Thase, M. E., Greenhouse, J. B., Frank, E. et al. (1997). Treatment of major depression with psychotherapy or psychotherapy-pharmacotherapy combinations. Archives of General Psychiatry, 54, 1009–15. Theilemann, S. (1993). Modification of cognitive disorders in schizophrenic and schizoaffective psychoses using cognitive therapy in comparison with sociotherapy. Nervenarzt, 64, 587–93. Turkington, D. and Kingdon, D. (1999). A systematic approach to cognitive therapy in schizoaffective disorders. In Cognitive Psychotherapy of Psychotic and Personality Disorders, ed. C. Perris and P. McGrory. Chichester: John Wiley & Sons Ltd., pp. 257–67.

Epilogue

The interface of affective and schizophrenic disorders: a cross between two spectra? Hagop S. Akiskal International Mood Center, University of California at San Diego

Introduction This monograph, based on the Proceedings of the November 2004 Frankfurt Conference, has greatly enriched our current knowledge of the interface between affective and schizophrenic disorders. As conference organizers and editors, Professor Marneros and I brought together some of the most incisive minds working on this interface along many lines of clinical and basic investigation. These lines of evidence have revealed both continuities and discontinuities between the affective and schizophrenic spectra. Perhaps what is “in-between” represents a cross of the underlying dimensions of the two “voluminous” spectra, or superposition of some of the contributory factors of one on the other. This epilogue will not give rise to firm conclusions. Rather, in examining several main lines of research, it will reformulate them with an eye towards the future of this field. Historical context In the first century ad, Aretaeus of Cappadocia (1856 translation) described the intimate relationship between melancholia and mania, thereby indicating a relationship between what we call today unipolar and bipolar disorders. Angst and Marneros (2001) contend that Aretaeus actually went further than that in his depiction of mania, spilling over into the realm of what some today may consider schizophrenia. Here is the passage on the transition of euphoric mania into severe psychosis: The patient may become excitable, suspicious, and irritable; hearing may become sharp, get noises The Overlap of Affective and Schizophrenic Spectra, ed. Andreas Marneros and Hagop Akiskal. Published by Cambridge University Press. © Cambridge University Press 2006.

277

278

Hagop S. Akiskal and buzzing in the ears; or may have visual hallucinations; bad dreams . . . and his sexual desires may get uncontrollable; aroused to anger, he may become wholly mad and run unrestrainedly, roar aloud; kill his keepers, and lay violent hands upon himself.

Affectiveologists, such as myself, are likely to diagnose the above picture as severe psychotic mixed mania, and the perceptual disturbances to be largely moodcongruent. In the DSM-IV schema (American Psychiatric Association, 1994), this could be regarded as mood-incongruent, but certainly not schizophrenic. The DSMIV is of course neo-Kraepelinian in the master’s broad formulation of the affective psychoses (Kraepelin, 1899). Curiously, the DSM-IV formulation also obliquely hints at Karl Jaspers’ (1913) concept of “understandability” (though without directly invoking his specific conceptualization that would be difficult to operationalize): I am referring to whether the psychotic experiences arise understandably from the pathologic mood change. Bleuler (1911), in turn, in his broad formulation of schizophrenia (which subsumed a mixture of manic depression and schizophrenia under the latter), may have invoked a schizophrenic psychosis as the most appropriate diagnosis for the Aretaeus case. For Kraepelin, the fate of such patients during follow-up was decisive: recovery without significant deterioration was evidence for manicdepressive psychosis. For Bleuler, outcome was unimportant; he based himself on the phenomenology of the illness, relying on presumably pathognomic signs and symptoms, which were based on a theory of dissociation of thought from affect, behavior and volition; it is also noteworthy that delusions and hallucinations were only of secondary importance to his diagnostic approach to schizophrenia. Kurt Schneider (1950) subsequently developed his own set of “first-rank” symptoms, relying on special delusional and perceptual disturbances, thereby highly restricting the terrain of manic-depressive psychosis, which he, paradoxically, broadly termed “cyclothymia”. This in part reverts back to the French contribution, a century earlier, whereby Falret (1854) and Baillarger (1854) emphasized, respectively, a circular and a biphasic course with free intervals. Although the French contribution described the core features of classical manic-depressive psychosis – i.e., a severe but cyclic or periodic illness with free intervals – it may have anticipated also what today is described as rapid cycling, a relatively uncommon expression of bipolar disorder. We find the more profound implications of circular insanity in a casual, but prophetic remark made by Falret, to the effect that one day it would be known that many melancholics in the community will be considered as milder cases of folie circulaire by virtue of brief periods of elation at the tail end of the melancholia; this he described as brief, joyful periods, “moments de gaieté” (Figure 15.1). The latter anticipated today’s concept of bipolar II (Dunner et al., 1976). It is equally instructive that Valentin Magnan (1893), his disciple, described mixed

279

The interface of affective and schizophrenic disorders

Falret – La folie circulaire

Baillarger – La folie à double forme

“Moments de gaieté” The melancholic forms in the community?

Figure 15.1. The evolution of a hereditary cyclic psychosis in double form (Baillarger, 1854; Falret, 1854).

Prevalence (%)

Temperaments 10

UP > 5 (+ others) BP IV+ BP III + BP II + BP I + SB BP III + BP II+ BP I + SB

5

BP II+ BP I +SB BP I +SB SB*

0 Genetic loading

(*Schizo-Bipolar)

Figure 15.2. Prevalence of bipolar phenotypes as a function of genetic loading (UP  unipolar; BP  bipolar).

states linking folie à double forme with folie circulaire, thereby somewhat obscuring the free interval. These historical remarks suggest that although phenomenology and course can be used to distinguish the affective from the schizophrenic psychoses, neither represent a definitive “test”. In an address to the French National Academy of Medicine at the 150th anniversary of Falret’s and Baillarger’s historical contributions, the present author formulated a broad bipolar spectrum that spans from temperament to schizobipolar psychosis (Akiskal, 2004). This is shown in Figure 15.2, where unipolar and bipolar subtypes occupy intermediary positions in the spectrum. Returning to Kraepelin, by invoking the concept of “dementia praecox,” he formulated an important psychopathologic feature, i.e., the fundamental importance of cognitive dysfunction in schizophrenia. Another German psychiatrist, Wilhelm Griesinger (1967 translation), who had preceded Kraepelin by several decades, in his concept of Einheitspsychose (unitary psychosis) had already formulated the

280

Hagop S. Akiskal

continuum between melancholia, mania, psychosis, and dementia. (Today, such a continuum involving all the functional psychoses is best represented in the largely theoretical position of Crow [1990].) Kraepelin’s concept of manisch-depressives Irresein (manic-depressive psychosis) is modern in one other fundamental respect. Although he did not use the term “spectrum”, he nonetheless defined a continuum between the severest psychosis and low-grade affective traits. Here we reproduce a few excerpts from his opening passage on manic depressive psychosis, which: Includes . . . the whole domain of so-called periodic and circular insanity . . . mania, the greater part of melancholia and . . . amentia [confusional psychosis]. Lastly we include here certain colorings of mood, some of them periodic, some of them continuously morbid . . . [which] pass over without sharp boundary into the domain of personal disposition.

Kraepelin avoided the term “temperament.” Kretschmer (1936) went further than his master in postulating that endogenous psychoses are nothing other than exaggerated expressions of temperaments. He further singled out the “cyclothymic” temperament as the underlying disposition to manic-depressive psychosis and the “schizothymic” as that of schizophrenia. In contemporary times, schizophrenologists were to rephrase the latter as “schizotypal” (Meehl, 1962; Parnas et al., 2005). Both temperaments in today’s terminology represent, respectively, the earliest “behavioral endophenotypes” of the bipolar and schizophrenic spectra. Thus far, we have in the main described continental European contributions. The American psychiatrist Kasanin (1933) envisaged a schizoaffective disorder which had an outgoing temperament preceding a confusional psychosis with relatively good prognosis (Table 15.1). The Scandinavian dual concepts of schizophreniform (Langfeldt, 1939) and reactive (Strömgren, 1986) psychoses cover an overlapping nosological terrain. The same can be said about the Japanese concept of “atypical psychoses” (Mitsuda, 1965). The culture-bound amok syndrome (Saint Martin, 1999) is another one, apparently still relevant even in the United States. This is not an exhaustive list of all the interface conditions between affective and schizophrenic disorders, but it is meant to convey to the reader that these “in-between” conditions have been a clinical dilemma for psychiatrists from diverse countries. What is important is that they can be reliably identified by clinicians (Okasha et al., 1993). Both psychological and somatic contributory factors have been described to define pathoplastic factors that could alter the expression of affective and schizophrenic psychoses. Despite such plurality of descriptive and somatic attempts to characterize this interface, these psychoses have defied further characterization on

281

The interface of affective and schizophrenic disorders Table 15.1. Kasanin’s concept of schizoaffective.

• • • • •

Acute features of both psychoses Emotional turmoil and confusion Precipitating stress Sudden onset

Full recovery • Good premorbid personality: “Lack of withdrawal or passivity” Source: Kasanin, 1933.

which unanimous agreement could be reached. Marneros, together with US and German colleagues (1986, 1991, 2004), has devoted much of his career to this interface: his work represents an exemplary model of dissection along multidimensional clinical and course parameters. Confusion, or otherwise altered consciousness, even stupor, represent common but not universal characteristics of these interface psychoses. Examples of this “organic” feature include Kahlbaum’s (1874) catatonia and the French concept of bouffée délirante (Magnan, 1893), as well as the oneiroid states (Mayer-Gross, 1924). Some today even invoke the possibility of “epileptoid” origins (Monroe, 1970; Blumer, 1984). Modern formulations of the interface The foregoing conditions have been officially designated as the polymorphous (i.e., rapidly alternating irritable, anxious, depressive, euphoric and psychotic features) termed “cycloid psychoses” (WHO, 1992). This concept derives from many sources, but reaches its more elaborate formulation in Leonhard (1961). His work represents a challenge to the Kraepelinian dichotomy of the endogenous psychoses. His contribution goes beyond the dichotomy and even the third psychosis option, to include a large gamut of psychotic conditions, such as unsystematic and systematic schizophrenias and bipolar disorders, catatonias, and cycloid psychoses. Although some research has been done to validate his model, particularly by Beckmann’s group (1990) and that of Brockington (1982), much of the postLeonhard literature has dealt with the cycloid psychoses and the monopolarbipolar distinction. Angst (1966), Perris (1966), Winokur and colleagues (1969) have attempted to validate the unipolar–bipolar distinction on the basis of primarily family history, being more depressive in what they have termed “unipolar depression” and more bipolar in those they have termed “bipolar disorder.” However, the work of many contemporary authors such as Gershon and colleagues (1982) has challenged their

282

Hagop S. Akiskal Table 15.2. Morbidity risk in first-degree relatives by proband diagnosis

Probands SA BP I BP II UP Normal

% SA

% BP I

% BP II

% UP

% Total

6.1 1.1 0.6 0.7 0.5

10.7 4.5 2.6 1.5 0

6.1 4.1 4.5 1.5 0.5

14.7 14.0 17.3 16.6 5.8

38 24 25 20 7

Source: Gershon et al., 1982.

dichotomous distinction in the affective realm. Table 15.2 from their work summarizes the plurality of affective and psychotic diagnoses within the family, largely irrespective of proband diagnosis. Indeed, the most common illness in the families of schizoaffective, bipolar I, bipolar II and unipolar, as well as normal probands, is depression. Furthermore, a gradient from high familial genetic loading to more modest loading can be seen, respectively, in the probands stratified from schizoaffective to normal. The present author (Akiskal, 1983) has argued that such data suggest a spectrum between psychotic, bipolar and unipolar disorders. Figure 15.2 is a graphic representation of this broad continuum. Other evidence for a spectrum concept of affective disorders is the gradient of affective disorders arising from an examination of the prospective course of cyclothymic individuals (Table 15.3): indeed, the designation “cyclothymic-bipolar spectrum” was first used in this paper (Akiskal, 1977). We have subsequently shown that hypomania could be as short as two days (Akiskal et al., 1979), spontaneous or “induced”, and that it could be favored by antidepressants (and other somatotherapies for depression), not only in depressive disorders (Akiskal et al., 1983), but among anxious-neurotic (Akiskal et al., 1978) and borderline personality (Akiskal et al, 1985) patients. Such data suggest a broader spectrum of affective dysregulation than classical unipolar-bipolar disorders. Finally, our work has shown that isolated hypomanic symptoms of short duration are commonly observed in both bipolar II and unipolar patients (Benazzi and Akiskal, 2001; Akiskal and Benazzi, 2003). This would suggest that many so-called unipolar patients are bipolar by virtue of intra-depressive hypomanic symptoms rather than outside the time frame of a depressive episode. Such data and considerations do suggest a spectrum of bipolar disorders which subsume much of the large terrain of unipolarity (see Figure 15.1). A model to depict this spectrum, expanded from Akiskal and Pinto (1999), is summarized in Table 15.4. For purposes of this discussion, the most interesting aspects of this dimensional model are at the two extreme ends. Type VI represents affectively disinhibited and labile states in the setting of the early course of dementing disorders of late onset (Akiskal, 2005). Type 1⁄2 represents the interface between

283

The interface of affective and schizophrenic disorders Table 15.3. Prospective course of cyclothymic temperament.

Manic episodes Hypomanic episodes Depressive episodes Cycling on TCAs

(3/46) (6/46) (11/46) (11/25)

6% 16% 24% 44%

Notes: Evidence for a cyclothymic-bipolar spectrum, Akiskal et al., 1977. Table 15.4. Proposal for a bipolar (BP) spectrum.

0 BP 1⁄2 BP I BP I 1⁄2 BP II BP II 1⁄2 BP III BP III 1⁄2 BP IV BP V BP VI

Schizophrenia Schizobipolar Mania Protracted hypomania Spontaneous hypomania Cyclothymic depressions Antidepressant-associated hypomania Mood swings with polysubstance abuse Hyperthymic depression Recurrent unipolar mixed states Mood instability in early dementia

psychotic bipolar depression of Type I and schizophrenia represented by Type 0 (Akiskal and Pinto, 2000). We believe this model to represent a quantitative logic which is in line with current oligogenic models of affective and schizophrenic disorders in molecular genetic and functional genomic investigations (Gershon, 2000; Niculescu et al., 2000). Schizoaffective disorders, used generically for the interface psychoses, represent a middle area between bipolar and schizophrenic disorders. This means that one can envisage psychoses beyond bipolar I, which do not quite make the schizoaffective label and could be characterized as mood-incongruent in their acute phase only (Type 3⁄4), versus psychoses in between schizoaffective and schizophrenic disorders, which could conceivably be characterized by a schizophrenic baseline and affective (particularly depressive) symptoms in the course of such a disorder (Type 1⁄4). Other permutations would be possible in the Type 0–I interface, and numerical designations might supplant the clinical terminology that has been used to describe them (Table 15.5). It would be relevant to point out that Kasanin’s concept of schizoaffective is closer to European and ICD-10 concepts of the interface psychoses. The DSM-IV definition of schizoaffective is very broad and Maj et al. (2000) have found it to be of very low reliability. Since at least 50% of manic-depressive patients exhibit

284

Hagop S. Akiskal Table 15.5. Affective disorder vs. schizophrenia

• • • • • • • • •

Chronic mania (manic dementia) Delusional depression Mood-incongruent bipolar (schizo-bipolar?) Delirious mania Confusional psychosis Puerperal psychosis Catatonia Cycloid psychosis (third psychosis?) “Secondary” depression in schizophrenia (schizo-depressive?)

DSM-IV psychotic features (Hantouche et al., 2003), in practice the DSM-IV schizoaffective category is largely characterized by interval schizophrenic signs and symptoms in the absence of affective symptoms. The broad range of schizoaffective disorders can be usefully divided into at least schizomanic (schizobipolar) and schizodepressive components (Clayton, 1982; Marneros and Tsuang, 1986). As described in this chapter, schizomania seems to lie at the extreme end of the psychotic bipolar disorders. Narrowly defined, schizoaffective disorder appears to have almost as good a prognosis as bipolar disorders (Angst et al., 1980). It is schizodepression then which is a very broad and an uncertain diagnosis. Some probably refer to schizophrenia with secondary depression (the post-psychotic depression of DSM-IV). Others might be a hodgepodge of miscellaneous conditions that clinicians are either unable or unwilling to disentangle. What is the origin of Type 1⁄2? To answer this question clinically, we first need to address the question of what it is that characterizes schizophrenia compared to affective disorder. One can take a categorical or dimensional approach (Ketter et al., 2005). If we were to consider the affective temperaments such as cyclothymia to represent the long-term trait of the affective, in particular, the bipolar spectrum disorders, trait cognitive dysfunction might be considered the hallmark of schizophrenia. This is not to say that the latter does not occur in bipolar disorder, but that, whatever the neurodevelopmental dysfunction is in schizophrenia, it is of earlier onset (congenital?), more pervasive and enduring in the latter; whereas in the bipolar spectrum it would be present largely at the severest end of the spectrum (i.e., the psychotic bipolar I disorders), and that it might behave in a “state-like” fashion. Others have invoked “defect state” taken as a proxy for the ravages that can be caused by neurodevelopmental cognitive dysfunction, to conceptualize unipolar

285

The interface of affective and schizophrenic disorders

psychotic depression at one pole and schizophrenia with defect state at the other. Interestingly, Malla (1995) states that “negative symptoms are associated only with schizophrenia without affect” and that “variation in affect can be a primary determinant in the type of psychosis.” More broadly, what characterizes schizophrenia is derailment of thought processes with poverty of content on repeated examination, as well as such negative symptoms as paucity of speech, apathy, and flattening on repeated examination that is unexplained by depression or classical neuroleptics (Andreasen and Akiskal, 1983). With respect to age at onset as a defining characteristic of two types of schizophrenia, Murray et al. (1992) stated that adult-onset psychosis is relapsing and remitting and related to affective disorder, with female preponderance and positive symptoms; by contrast, the early-onset type could be considered “congenital schizophrenia” with aberrant brain development during fetal and neonatal life, further characterized by male predominance and poor outcome. Not having definitive data about the underlying biologic structure of affective and schizophrenic spectra (though more is published about the latter, e.g., Parnas et al. [1982], Cadenhead et al. [2002], Siever and Davis [2004]), our aim in attempting to characterize the origins of the cross between these spectra has to be more modest and to be based on what our partial literature review has shown to be the fundamental characteristics of each of these spectra. These could involve, among others, familial–genetic factors, neurodevelopmental abnormalities, pathoplastic temperamental factors, as well as psychological and somatic formative influences. Using the family-history strategy, it has been shown that schizophrenic patients with depressive episodes have a significant excess of familial depression (Subotnik et al., 1997). Maier et al. (1993) have shown that schizophrenic probands have an increased familial risk for unipolar depression and schizo-unipolar disorder, but not for bipolar disorder. Both sets of studies suggest at least some “mixing” of putative “psychosis” and “depression” genes. As far as molecular genetics, the genes underlying affective and schizophrenic disorders could be both distinct and overlapping (Berettini, 1998; Gershon, 2000). It is also plausible that “clock genes” (Bunney and Bunney, 2000) could modify the expression of “mood genes” or “psychosis genes” to lead to seasonal or cyclical rather than chronic or deteriorating disorders. A provocative possibility is the “operation of the schizophrenia susceptibility gene, neuregulin 1, across traditional diagnostic boundaries to increase risk for bipolar disorder” (Green et al., 2005). Another possibility is that there is an oligogenic threshold of additive broad behavior genes whereby fewer genes would spell out depression, a greater number would fashion bipolarity, and the largest number would translate into schizophrenic psychosis (Gershon, 2000). A last possibility that should be mentioned is that “temperament genes” might

286

Hagop S. Akiskal

moderate the differential expression of affective and schizophrenic psychopathology (Kelsoe, 2003; Evans et al., 2005). As formulated herein, the foregoing examples do not necessarily represent specific hypotheses, but ideas for future investigation. As regards temperament, in a more clinical vein, we have already mentioned the good prognosis of Kasanin’s (1933) psychotic patients with outgoing premorbid temperament. Contrary-wise, the French EPIMAN-II group (Azorin et al., 2006, in press) has shown how an irritable temperament can adversely influence the course of mania. Thus, mood-incongruent mania is characterized by female gender, remitting but shorter free intervals, greater instability of mood, more mixed-anxious features, all of which are associated with an irritable temperament. We have formulated that, in line with the Berner (1982) hypothesis, the emotional reactions of others to this temperament render these individuals paranoid. Prospective epidemiologic data (Messias et al., 2001) too, support an analogous formulation for schizophrenia spectrum disorders with secondary depression. There might be psychological and somatic precipitants or formative factors which influence how an affective disorder might acquire a “pseudo-schizophrenic” characteristic, which might qualify for the phenotypic label of “schizoaffective” or “schizophrenic” (Akiskal and Puzantian, 1979). These include substance (especially stimulant) abuse, anticholinergic agents used to counteract the side effects of neuroleptic drugs, sleep deprivation, metabolic disturbances, concurrent epileptic or temporal lobe pathology, head trauma to the prefrontal lobes, and superimposition of psychosis on mental retardation. It is finally relevant to point out that a somatic precipitant, such as a puerperal onset, might impart a better prognosis to post-partum psychoses (Rohde and Marneros, 1992). Decline of function (even defect state) in bipolar I disorder could arise from neurodevelopmental abnormalities shared with schizophrenia (a discussion of which is beyond the scope of this epilogue), as well as from social factors, such as extreme discord between the patient with bipolar I disorder and the family (Akiskal, 2005). This could lead to loss of social resources and support, resulting in uneven or lost productivity, financial and legal complications, alienation of family and friends, loss of healthcare, chaotic, unstable social existence, “prostitution,” and homelessness. Kraepelin (1899) was aware of such a deteriorating course in bipolar disorder, and towards the end of his career expressed doubt whether the dichotomy between manic-depressive and schizophrenic disorders could be maintained in its original formulation. This is proof that, far from being “rigid” – as many who may not have read all his work contend – he was actually rather “open-minded”.

287

The interface of affective and schizophrenic disorders

Concluding remarks With respect to the main theme of this monograph, Karl Jaspers (1913) stated: Whereas one set of investigators . . . makes the dementia praecox group impossibly wide, the other set . . . narrows down the dementia praecox group considerably. Thus for many years the border between manic-depressive insanity and dementia praecox has vacillated considerably without anything new emerging.

After more than a century and a half of research highlighted in this book, it is relevant to ask whether we are any closer to better understanding of this interface. I submit that perhaps we have better insight about the research directions we need to take towards that goal. Whether we have achieved such insight is a judgment for the reader to make. R E F E R E N C ES Akiskal, H. S. (1983). The bipolar spectrum: New concepts in classification and diagnosis. In Psychiatry Update: The American Psychiatric Association Annual Review, 2, ed. L. Grinspoon. Washington, DC: American Psychiatric Press, pp. 271–92. Akiskal, H. S. (2004). De la Folie circulaire (à double forme) au spectre bipolaire: la tendance chronique à la récidive dépressive [From circular insanity (in double form) to the bipolar spectrum: chronic tendency for depressive recurrence.] Le Bulletin de l’Académie nationale de Médecine, 88, 285–96. Akiskal, H. S. (2005). Mood disorders: Clinical features. In Sadock and Kaplan’s Comprehensive Textbook of Psychiatry, edn VIII, ed. B. J. Sadock and V. A. Sadock. Baltimore, MD: Lippincott, Williams & Wilkins, pp. 1611–52. Akiskal, H. S. and Benazzi, F. (2003). Family history validation of the bipolar nature of depressive mixed states. Journal of Affective Disorders, 73, 113–22. Akiskal, H. S., Bitar, A. H., Puzantian, V. R., Rosenthal, T. L. and Walker, P. W. (1978). The nosological status of neurotic depression: A prospective three-to-four year examination in light of the primary-secondary and unipolar-bipolar dichotomies. Archives of General Psychiatry, 35, 756–66. Akiskal, H. S., Chen, S. E., Davis, G. C. et al. (1985). Borderline: An adjective in search of a noun. Journal of Clinical Psychiatry, 46, 41–8. Akiskal, H. S., Djenderedjian, A. H., Rosenthal, R. H. and Khani, M. K. (1977). Cyclothymic disorder: Validating criteria for inclusion in the bipolar affective group. American Journal of Psychiatry, 134, 1227–33. Akiskal, H. S. and Pinto, O. (1999). The evolving bipolar spectrum: Prototypes I, II, III, IV. Psychiatric Clinics of North America, 22, 517–34. Akiskal, H. S. and Pinto, O. (2000). The soft bipolar spectrum: Footnotes to Kraepelin on the interface of hypomania and depression. In Bipolar Disorders: 100 Years After Manic Depressive Insanity, ed. A. Marneros and J. Angst. Dordrecht: Kluwer Academic Publishers, pp. 37–62.

288

Hagop S. Akiskal Akiskal, H. S., Pinto, O. and Lara, D. R. (2005). “Bipolarity” in the setting of dementia: bipolar type VI? http://www.medscape.com/viewarticle/495795 01/06/2005 (5 pp). Akiskal, H. S. and Puzantian, V. R. (1979). Psychotic forms of depression and mania. Psychiatric Clinics of North America, 2, 419–39. Akiskal, H. S., Rosenthal, R. H., Rosenthal, T. L. et al. (1979). Differentiation of primary affective illness from situational, symptomatic, and secondary depressions. Archives of General Psychiatry, 36, 635–43. Akiskal, H. S., Walker, P. W., Puzantian, V. R. et al. (1983). Bipolar outcome in the course of depressive illness: Phenomenologic, familial, and pharmacologic predictors. Journal of Affective Disorders, 5, 115–28. American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, 4 (DSM-IV). Washington DC: American Psychiatric Press. Andreasen, N. C. and Akiskal, H. S. (1983). The specificity of Bleulerian and Schneiderian symptoms: A critical re-evaluation. Psychiatric Clinics of North America, 6, 41–54. Angst, J. (1966). Zur Ätiologie und Nosologie Endogener Depressiver Psychosen. Berlin, Heidelberg, New York: Springer. Angst, J., Felder, W. and Lohmeyer, B. (1980). Course of schizoaffective psychoses: results of a follow-up study. Schizophrenia Bulletin, 6, 579–85. Angst, J. and Marneros, A. (2001). Bipolarity from ancient to modern times: conception, birth and rebirth. Journal of Affective Disorders, 67, 3–19. Aretaeus of Cappadocia. (1856). The Extant Works of Aretaeus the Cappadocian. London: Sydenham Society. Azorin, J. M., Akiskal, H. S. and Hantouche, E. (2005). The mood-instability hypothesis in the origin of mood-congruent versus mood-incongruent psychotic distinction in mania: validation in a French National Study of 1090 patients. Journal of Affective Disorders, in press. Baillarger, J. (1854). Notes sur un genre de folie dont les accès sont caractérisés par deux périodes régulières, l’une de dépression, l’autre d’excitation. Le Bulletin de l’Académie Nationale de Médecine, 19, 340. Beckmann, H., Fritze, J. and Lanczik, M. (1990). Prognostic validity of the cycloid psychoses. A prospective follow-up study. Psychopathology, 23, 205–11. Benazzi, F. and Akiskal, H. S. (2001). Delineating bipolar II mixed states in the Ravenna-San Diego collaborative study: The relative prevalence and diagnostic significance of hypomanic features during major depressive episodes. Journal of Affective Disorders, 67, 115–22. Berettini, W. (1998). Progress and pitfalls: bipolar molecular linkage studies. Journal of Affective Disorders, 50, 287–97. Berner, P. (1982). Psychiatrische Systematik. Dritte Auflage. Bern: Huber. Bleuler, E. (1911). Dementia praecox oder Gruppe der Schizophrenien. Leipzig, Wien: Deuticke. Blumer, D. (1984). Psychiatric Aspects of Epilepsy. Washington DC: American Psychiatric Press. Brockington, I. F., Perris, C., Kendell, R. E., Hillier, V. E. and Wainwright, S. (1982). The course and outcome of cycloid psychosis. Psychological Medicine, 12, 97–105. Bunney, W. E. and Bunney., B. G. (2000). Molecular clock genes in man and lower animals: possible implications for circadian abnormalities in depression. Neuropsychopharmacology, 22, 335–45.

289

The interface of affective and schizophrenic disorders Cadenhead, K. S., Light, G. A., Geyer, M. A., McDowell, J. E. and Braff, D. L. (2002). Neurobiological measures of schizotypal personality disorder: defining an inhibitory endophenotype? American Journal of Psychiatry, 157, 869–71. Clayton, P. J. (1982). Schizoaffective disorders. The Journal of Nervous and Mental Disease, 170, 646–50. Crow, T. J. (1990). The continuum of psychosis and its genetic origins. The sixty-fifth Maudsley lecture. British Journal of Psychiatry, 56, 788–97. Dunner, D. L., Gershon, E. S. and Goodwin, F. K. (1976). Heritable factors in the severity of affective illness. Biological Psychiatry, 11, 31–42. Evans, L., Akiskal, H. S., Keck, Jr., P. E. et al. (2005). Familiality of temperament in bipolar disorder: support for a genetic spectrum. Journal of Affective Disorders, 85, 153–68. Falret, J. P. (1854). Mémoire sur la folie circulaire. Le Bulletin de l’Académie Nationale de Médecine, 19, 382–415. Gershon, E. S. (2000). Bipolar illness and schizophrenia as oligogenic diseases: implications for the future. Biological Psychiatry, 47, 240–4. Gershon, E. S., Hamovit, J., Guroff, J. J. et al. (1982). A family study of schizoaffective, bipolar I, biplar II, unipolar, and normal control probands. Archives of General Psychiatry, 39, 1157–67. Green, E. K., Raybould, R., Macgregor, S. et al. (2005). Operation of the schizophrenia susceptibility gene, neuregulin 1, across traditional diagnostic boundaries to increase risk for bipolar disorder. Archives of General Psychiatry, 62, 642–8. Griesinger, W. (1967). Mental Pathology and Therapeutics (translated by C. L. Robertson and J. Rutherford). London: New Sydenham Society. Hantouche, E. G., Azorin, J-M., Châtenet-Duchêne, L. et al. (2003). Caractérisation de la manie dans une cohorte nationale de 1090 patients dans «EPIMAN-II Mille»: Fréquence des soustypes cliniques, début et errances diagnostiques. Annales Médico-Psychologiques, 161, 359–66. Jaspers, K. (1913). Allgemeine Psychopathologie. Berlin: Spinger. Kahlbaum, K. (1874). Die Katatonie. Berlin: Kirschwald. Kasanin, J. (1933). The acute schizoaffective psychoses. American Journal of Psychiatry, 13, 97–126. Kelsoe, J. R. (2003). Arguments for the genetic basis for the bipolar spectrum. Journal of Affective Disorders, 73, 183–97. Ketter, T. A., Wang, P. W., Becker, O. V., Nowakowska, C. and Yang, Y. (2005). Psychotic bipolar disorders: dimensionally similar to or categorically different from schizophrenia? Journal of Psychiatric Research, 38, 47–61. Kraepelin, E. (1899). Psychiatrie: ein Lehrbuch fur Studirende und Ärzte, 6. vollst. umgearb. Aufl. Leipzig: J. A. Barth. Kretschmer, E. (1936). Physique and Character. London: Kegan, Paul, Trench, Trubner and Co. Ltd. Langfeldt, G. (1939). The Schizophreniform States. Kopenhagen: Munksgaard. Leonhard, J. (1961). Cycloid psychoses – endogenous psychoses which are neither schizophrenic nor manic-depressive. Journal of Mental Science, 107, 633–48. Magnan, V. (1893). Leçons Cliniques sur les Maladies Mentales. 2. Aufl. Paris: Battaille.

290

Hagop S. Akiskal Maier, W., Lichtermann, D., Minges, J. et al. (1993). Continuity and discontinuity of affective disorders and schizophrenia. Results of a controlled family study. Archives of General Psychiatry, 50, 871–83. Maj, M., Pirozzi, R., Formicola, A. M., Bartoli, L. and Bucci, P. (2000). Reliability and validity of the DSM-IV diagnostic category of schizoaffective disorder: preliminary data. Journal of Affective Disorders, 57, 95–8. Malla, A. K. (1995). Negative symptoms and affective disturbance in schizophrenia and related disorders. Canadian Journal of Psychiatry, 40, S55–9. Marneros, A. (1991). (English abstract) Affektive, Schizoaffektive und Schizophrene Psychosen. Berlin: Springer. Marneros, A. and Angst, J. (2000). Bipolar Disorders. 100 Years After Manic Depressive Insanity. Dordrecht, Boston, London: Kluwer Academic Publishers. Marneros, A. and Goodwin, F. K. (2005). Bipolar Disorders. Mixed States, Rapid Cycling and Atypical Forms. Cambridge: Cambridge University Press. Marneros, A. and Pillmann, F. (2004). Acute and Transient Psychoses. Cambridge: Cambridge University Press. Marneros, A. and Tsuang, M. T. (eds.) (1986). Schizoaffective Psychoses. Berlin: Springer. Marneros, A. and Tsuang, M. T. (1990). Affective and Schizoaffective Disorders. Similarities and Differences. New York: Springer. Marneros, A., Tsuang, M. T. and Andreasen, N. C. (1995). Psychotic Continuum. Berlin, Heidelberg, New York: Springer. Mayer-Gross, W. (1924). Selbstschilderung und Verwirrtheit. Die oneiroide Erlebnisform. Berlin: Springer. Meehl, P. E. (1962). Schizotaxia, schizotypy, schizophrenia. American Psychology, 17, 827–38. Messias, E., Kirkpatrick, B., Ram, R., Tien, A. Y. (2001). Suspiciousness as a specific risk factor for major depressive episodes in schizophrenia. Schizophrenia Research, 47, 159–65. Mitsuda, H. (1965). The concept of “atypical psychosis” from the aspect of clinical genetics. Acta Psychiatrica Scandinavica, 41, 372–7. Monroe, R. R. (1970). Episodic Behavioral Disorders: A Psychodynamic and Neurophysiologic Analysis. Cambridge, MA: Harvard University Press. Murray, R. M., O’Callaghan, E., Castle, D. J., Lewis, S. W. (1992). A neurodevelopmental approach to the classification of schizophrenia. Schizophrenia Bulletin, 18, 319–32. Niculescu, A. B. 3rd, Segal, D. S., Kuczenski, R. et al. (2000). Identifying a series of candidate genes for mania and psychosis: a convergent functional genomics approach. Physiological Genomics, 9, 83–91. Okasha, A., El Dawla, A. S., Khalil, A. H. and Saad, A. (1993). Presentation of acute psychosis in an Egyptian sample: a transcultural comparison. Comprehensive Psychiatry, 34, 4–9. Parnas, J., Licht, D. and Bovet, P. (2005). Cluster A personality disorders: a review. In Personality Disorders. WPA Series Evidence and Experience in Psychiatry, ed. M. Maj, H. S. Akiskal, J. E. Mezzich and A. Okasha. Chichester: John Wiley & Sons Ltd, pp. 1–74. Parnas, J., Schulsinger, F., Schulsinger, H., Mednick, S. A., Teasdale, T. W. (1982). Behavioral precursors of schizophrenia spectrum. A prospective study. Archives of General Psychiatry, 39, 658–64.

291

The interface of affective and schizophrenic disorders Perris, C. (ed). (1966). A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatrica Scandinavica, 194, 9–14. Rhode, A. and Marneros, A. (1992). Schizoaffective disorders with and without onset in the puerperium. European Archives of Psychiatry and Clinical Neuroscience, 242, 27–33. Saint Martin, M. D. (1999). Running amok: a modern perspective on a culture-bound syndrome. Primary Care Companion to the Journal of Clinical Psychiatry, 1, 66–70. Schneider, K. (1950). Klinische Psychopathologie. Stuttgart: Thieme. Siever, L. J. and Davis, K. L. (2004). The pathophysiology of schizophrenia disorders: perspectives from the spectrum. American Journal of Psychiatry, 161, 398–413. Strömgren, E. (1986). Reactive (psychogenic) psychoses and their relations to schizoaffective psychoses. In Schizoaffective Psychoses, ed. A. Marneros and M. T. Tsuang. Berlin, Heidelberg: Springer, pp. 260–71. Subotnik, K. L., Nuechterlein, K. H., Asarnow, R. F. et al. (1997). Depressive symptoms in the early course of schizophrenia: relationship to familial psychiatric illness. American Journal of Psychiatry, 154, 1551–6. Winokur, G., Clayton, P. J. and Reich, T. (1969). Manic Depressive Illness. St. Louis, MO: C.V. Mosby. World Health Organization. (1992). The International Classification of Disease (ICD-10): classification of mental and behavioral disorders. Geneva, Switzerland: WHO.

Index

ACC-subtype depression 171, 175–6 acute and transient psychotic disorders 8, 15, 182–4 see also brief and acute psychoses acute treatment of schizoaffective disorders 250–3, 257–9 affective disorders association with diabetes 213 family genetic risk 27–9 spectrum concept 281–4 see also bipolar disorder; depressive syndromes affective spectrum, relationship of brief and acute psychoses 184–203 age of onset and form of psychosis 44–6 and sex 45–6 akathisia, drug-induced 157 akinesia, drug-induced 157 amisulpride 173 amitriptyline treatment of schizoaffective disorders 251 treatment of suicidal behavior 237, 238, 240 amok syndrome 280 amygdala, role in depressive syndromes 169–70, 171 Angst, Jules 5, 7 anterior cingulate cortex (ACC) ACC-subtype depression 171, 175–6 role in depressive syndromes 169–71 anticonvulsant medications acute treatment of schizoaffective disorders 252 long-term treatment of schizoaffective disorders 255 antidepressant medications for depressive syndromes in schizophrenia 174 for schizoaffective disorders 251–2 antipsychotic medications acute treatment of schizoaffective disorders 253 and glucose dysfunction 214 long-term treatment of schizoaffective disorders 256–7 Aretaeus of Cappadocia 1, 3, 277–8 aripiprazole 173–4, 240 treatment of schizoaffective disorders 253

293

atypical antipsychotic drug treatments 173–4 atypical psychosis 8, 182, 183, 280 bipolar disorder 18 chromosome regions implicated 30–8 effects of hypercortisolemia 87, 89–90, 211 genome scan meta-analyses 46–7 mechanisms of sustained attention impairment 92–3, 94–6 progressive memory impairment 87, 89–90 state-modulated trait marker 86–7 state-related deficit 86–7 susceptibility genes overlap with schizophrenia 32–8 sustained attention deficit 81, 85–92 trait marker 87, 89 bipolar schizoaffective disorders 15 relationship to brief and acute psychoses 200, 201–2 bipolar spectrum of disorders 184, 281–4 Bleuler, Eugen 4, 18, 106–7, 278 bouffée délirante 8, 182 brain evolution cerebral torque 51–2 increase in size 50–2 language development 51–2 brief and acute psychoses acute and transient psychotic disorders (ICD10) 182–4 brief psychotic disorder (DSM-IV) 182–3 classification systems 182–4 overlapping of the spectra 199–203 psychotic continuum concept 184 relationship to affective spectrum 184–203 relationship to schizophrenic spectrum 184–203 brief and acute psychoses study (HASBAP) 185–203 family history and morbid risk 197–9 features of the acute episode 188–90, 191, 199 frequency and types of relapses 190–2 gender and age at onset 188, 189, 199–200 longitudinal development in relation to schizophrenia 194–7 methodology 185–7

294

Index brief and acute psychoses study (HASBAP) (cont.) outcome comparisons between groups 192–4 premorbid functional level 200 psychogenic aspect 200–1 relationship to bipolar schizoaffective disorders 200, 201–2 relationship to schizophrenia spectrum 201 results of study 188–99 brief polymorphic disorders 17 brief polymorphic psychoses 15, 17 brief psychotic disorder (DSM-IV) 106, 182–3 see also brief and acute psychoses buproprion 174 candidate susceptibility genes, findings for 46–7 carbamazepine, treatment of schizoaffective disorders 252, 254–5 cardiovascular disorders, as physical comorbidities 209–10 catechol-O-methyl transferase gene see COMT categorical models of psychosis 79–80 chlorpromazine 237 treatment of schizoaffective disorders 250, 251, 253 cholesterol, association with suicidal behavior 236 chromosomes, regions implicated in psychoses 30–8, 46–7 classification see DSM-IV; ICD-10; psychiatric classification clinical recommendations acute treatment of schizoaffective disorders 257–9 long-term treatment of schizoaffective disorders 257–9 clozapine treatment of schizoaffective disorders 253, 256–7 treatment of suicidal behavior 238, 239 cognitive behavior therapy (CBT) 264–74 cognitive model 265–7 influence of stress-diathesis models 265 interest in non-pharmacological treatments 265 key characteristics 267–73 outcome research 273 psychological aspects of vulnerability and risk 265 social aspects of vulnerability and risk 265 use in schizoaffective disorders 264, 267–74 use in schizophrenia and mood disorders 267–74 Cologne Study 8–9 comorbidity see physical comorbidity COMT (catechol-O-methyl transferase) susceptibility gene 33, 35, 36, 47 Continuous Performance Test (CPT) for sustained attention 81–2 continuum of psychoses 15–18, 43–4, 79–80, 184, 281–4 cultural influences on diagnosis culturally-bound syndromes of the Third World 120–1

culture attitudes towards illness 119–22 hysterical symptoms classification 125 impact on mental health 119–22 influence on diagnosis of psychoses 123–6 influence on manifestation of psychoses 124–6 psychosocial markers of traditional society 119–22 psychosocial markers of Western society 120 religious ritual and obsessive compulsive disorder 124–5 traditional cultures and mental health 122–4 cycloid psychoses 5, 8, 182, 183–4, 199, 281 cyclothymia 4, 278, 280 cyclothymic-bipolar spectrum 282–3 dementia praecox 1, 3, 279 depressive syndromes differential diagnosis 156–62 clinical symptomatology and illness course 159–62 correlations between rating scales and symptom dimensions 159–62 drug-induced akathisia 157 drug-induced akinesia 157 drug-induced dysphoria 157 illness phase-dependent symptom relationships 159–61 importance of subjective ratings 161–2, 165, 166–7 initial or relapse prodromes 158 medical/organic causes 156–7 medication-induced conditions 157 negative symptoms 158 neuroleptic-induced conditions 157 psychomotor agitation 157 relationship with negative symptoms 159–62 schizoaffective depression 158–9 substance abuse-induced conditions 157 depressive syndromes in schizophrenia ACC-subtype depression 171, 175–6 antidepressant treatments 174 atypical antipsychotic drug treatments 173–4 candidate genes 172–3 differential diagnosis 156–62 ‘emotion circuits’ in the brain 169–71 experimental psychopathology 162–5 family studies 171–2 frequency of occurrence 156 functional neuroimaging 168–9 genetic findings 171–3 molecular genetic studies 172–3 neurocognitive impairments 166–8 PFC-subtype depression 171, 175–6 psychotherapeutic approaches 174 role of the amygdala 169–70, 171 role of the anterior cingulate cortex 169–71 role of the hippocampus 169, 170 role of the insular cortex 169 role of the prefrontal cortex 169–71 sibling pair studies 172

295

Index structural neuroimaging 168 treatment approaches 173–4 desipramine, treatment of schizoaffective disorders 251 diabetes association with affective disorders 213 association with schizophrenia 213–16 diagnostic criteria for schizoaffective disorders DSM-IV 2, 10–14 ICD-10 2, 10–14 diagnostic systems for functional psychoses 57-61 quantitative, syndrome-oriented approaches 59–72 dichotomous genetic traits 25 dichotomy concept 15–18 see also Kraepelin, Emil dimensional model of psychosis 80 DISC1 susceptibility gene 33, 34, 47 DNA markers, genetic linkage studies 29–30 dopamine transmission, effects of glucose dysfunction 215–16 DSM, evolution of criteria for schizoaffective disorder 9–10, 18 DSM-IV definition of schizoaffective disorder 2, 10–14, 108–9 dysbindin susceptibility gene 32, 33, 47 dyslipidemia, as physical comorbidity 210–11 dysphoria, drug-induced 157 endogenous psychoses 183–4 epigenetic modulation, Protocadherin X and Y genes 52 epistatic (multiplicative) gene interactions 27 ethnicity, and population frequency of functional psychoses 55 ethnicity-independent genetic vulnerability for functional psychoses 64–6 evolution see hominid evolution expressed emotions in the family, and risk of relapse 126 familial genetic loading 282 family climate, and psychosocial adjustment 126 family expressed emotions, and risk of relapse 126 family genetic risk, schizophrenia and affective disorders 27–9 family studies, suicidal behavior 234–5 fluoxetine 174 flupentixol 173 fluphenazine, treatment of schizoaffective disorders 254 folie à double forme 279 folie circulaire 278–9 functional psychoses comorbidity rates 56 complex quantitative phenotypes 60–1, 62 definition and characteristics 55 diagnostic systems 57–61 direct and indirect costs to society 57 elevated suicide rate 55–6 environmental modification 55 ethnicity-independent genetic vulnerability 64–6

illness concept 59 in developing countries 55 mortality rates 55–6 neurobiological basis 61, 63–4 non-specific vulnerability model 56 oligogenic vulnerability model 66, 71–2 operational criteria for diagnosis 58 pharmacological treatment 56 population frequencies 55 quantitative, syndrome-oriented approaches 59–72 sociocultural modification 55 time characteristics of psychotropic drug response 66–9 Zurich family study 60–1, 62 G72 susceptibility gene 33, 34 gabapentin, treatment of schizoaffective disorders 252, 255 genetic components of physical comorbidities 215, 217 genetic epidemiology family studies 27–9 heritability of psychiatric disorders 27–9 mode of transmission of psychiatric disorders 27–9 schizophrenia and affective disorder risk 27–9 genetic heterogeneity, effects on genetic linkage studies 32 genetic linkage studies 29–32 chromosome regions implicated 30–2 common regions for schizophrenia and bipolar disorder 30–2 DNA markers 29–30 effects of genetic heterogeneity 32 findings for candidate genes 46–7 Genome Scan Meta-Analysis (GSMA) 30–1 genome scanning 30 inconsistency between studies 30–2 meiotic mapping 30 overlapping susceptibility genes 32–8 positional cloning 30 systematic gene mapping 30 genetic models epistatic (multiplicative) gene interactions 27 factors affecting susceptibility gene expression 26 multifactorial threshold model 27 penetrance of the susceptibility allele 26 quantitative trait polygenic models 26–7 single major locus (Mendelian) models 25–6 spectrum phenotype 26 susceptibility gene transmission 25–6 variable expressivity in genetic traits 26 X-linked transmission 26 genetic overlap allelic heterogeneity model 36–7 common genes–different environment model 37 gene–gene interaction model 37 multiple threshold model 37 possible models 36–8

296

Index genetic traits dichotomous 25 quantitative type 25 genetic vulnerability for functional psychoses, ethnicity-independent 64–6 Genome Scan Meta-Analysis (GSMA) 30–1 genome scanning 30 gestational diabetes 214 glucose dysfunction and antipsychotic medications 214 and problems before birth 214 as physical comorbidity 211–16 association with schizophrenia 213–16 effects on dopamine transmission 215–16 effects on memory performance 216 in metabolic syndrome 211–12 in psychiatric illness 213–16 good-prognosis schizophrenia 182, 183 GRK3 (G-protein receptor kinase 3) susceptibility gene 33, 34–5, 36 Halle Bipolarity Longitudinal Study (HABILOS) 13–14 Halle Study on Brief and Acute Psychoses (HASBAP) 17, 185–203 haloperidol 83, 173 treatment of schizoaffective disorders 251, 253, 256 treatment of suicidal behavior 238, 240 hebephrenia 45 heritability of psychiatric disorders 27–9 hippocampus, role in depressive syndromes 169, 170 Hippocrates 1, 3 hominid evolution cerebral torque and evolution of language 51–2 epigenetic modulation 52 mechanism for increase in brain size 50–2 sapiens speciation event 49–52 sex chromosome alterations 48–52 HPA (hypothalamic-pituitary-adrenal) axis dysfunction and suicidal behavior 235–6 in metabolic syndrome 212 predictive value 140–2 hypercortisolemia 87, 89–90, 211 hysterical symptoms, classification difficulties 125 ICD, evolution of criteria for schizoaffective disorders 9, 10, 18 ICD-10 classification of simple schizophrenia 125–6 criteria for schizoaffective disorders 2, 10–14, 108–9 difficulties with use of negative symptoms 125–6 illness concept in psychiatry 59 insular cortex, role in depressive syndromes 169 insulin resistance syndrome, as physical comorbidity 211 interface between spectra clinical dilemma of “in-between” conditions 277–81

continuum of psychoses 15–18, 43–4, 79–80, 184, 281–4 cycloid psychoses 281 cyclothymic-bipolar spectrum 282–3 distinctions between schizophrenia and affective disorders 284–7 DSM-IV schema 278 familial genetic loading evidence 282 historical context 277–81 modern formulations 281–7 schizoaffective disorders 282–7 spectrum of affective disorders 281–4 spectrum of bipolar disorders 281–4 unipolar-bipolar distinction 281–2 Islamic culture attitudes towards illness 120–2 community and family orientation 120–2 effects of urbanization 119 family role in mental health 123, 126 hysterical symptoms classification 125 influence on diagnosis of psychoses 123–6 influence on manifestation of psychoses 124–6 influence on psychiatric disorders 122–4 psychologization of social deviance 123 religious and sexual taboos relating to women 124–5 religious ritual and obsessive–compulsive disorder 124–5 Jaspers, Karl 1–2, 12, 278, 287 Kahlbaum, Karl 3, 281 Kasanin, Jacob 2, 4, 107, 280–1 Kraepelin, Emil 1–4, 18 challenges to dichotomy principle 16, 280 concept of dementia praecox 1, 3, 279 concept of manic-depressive illness 1, 3, 184, 280 dichotomy of endogenous psychoses 1–4 doubts about binary principle 3–4, 43, 44, 106–7, 286 influence of Kraepelin’s dichotomy 105, 278 lamotrigine 238 treatment of schizoaffective disorders 252, 255 lithium 87, 238–9 acute treatment of schizoaffective disorders 250–1, 253 long-term treatment of schizoaffective disorders 253–5 long-term treatment of schizoaffective disorders 253–7, 257–9 major depressive disorder, response to drug treatments 66–7 males, age of onset of schizophrenia 45–6 manic-depressive illness 1, 3 manic syndromes, predictive value of individual phenomena 135–8 medications see pharmacological treatment meiotic mapping 30 memory, progressive impairment in bipolar disorder 87, 89–90

297

Index memory performance, effects of glucose dysfunction 216 metabolic syndrome, as physical comorbidity 211–13 mode of transmission of psychiatric disorders 27–9 monomorphous schizoaffective disorders 15 mood disorders see affective disorders; bipolar disorder; depressive syndromes mortality rates for psychiatric patients 207–9, 216 multifactorial threshold genetic model 27 neuregulin 1 (NRG1) susceptibility gene 32–4, 47 neurobiology of suicidal behavior 235–6 neurofibromatosis, variability in gene expression 26 noradrenegic dysfunction, and sustained attention deficit 95–6 nortriptyline 174 treatment of schizoaffective disorders 252 obesity, as physical comorbidity 211 obsessive–compulsive disorder, and religious ritual 124–5 olanzapine 173 treatment of schizoaffective disorders 253, 256 treatment of suicidal behavior 238, 239, 240 oligogenic vulnerability model 66, 71–2 overlap between spectra see interface between spectra overlapping susceptibility genes 32–8 oxcarbazepine, treatment of schizoaffective disorders 252 paroxetine 174 penetrance of the susceptibility allele 26 perphenazine, treatment of schizoaffective disorders 252 PFC-subtype depression 171, 175–6 pharmacological treatment of functional psychoses 56 pharmacological treatment of schizoaffective disorders 248–59 acute treatment 250–3, 257–9 amitriptyline 251 anticonvulsants (acute) 252 anticonvulsants (long-term) 255 antidepressants 251–2 antipsychotics (acute) 253 antipsychotics (long-term) 256–7 aripiprazole 253 carbamazepine 252, 254–5 chlorpromazine 250, 251, 253 clinical recommendations (acute) 257–9 clinical recommendations (long-term) 257–9 clozapine 253, 256–7 desipramine 251 diagnostic issues 249–50 fluphenazine 254 gabapentin 252, 255 haloperidol 251, 253, 256 lamotrigine 252, 255 lithium 250–1, 253–5

long-term treatment 253–7, 257–9 nortriptyline 252 olanzapine 253, 256 oxcarbazepine 252 perphenazine 252 quetiapine 253 risperidone 253, 256 topiramate 252, 255 valproate 252, 255 ziprasidone 253 phenotype spectrum 26 physical comorbidity and vulnerability for psychiatric disorders 209, 217 cardiovascular disorders 209–10 dyslipidemia 210–11 genetic components 215, 217 glucose dysfunction 211–16 HPA axis dysfunction in metabolic syndrome 212 indications for interventions 209, 217 insulin resistance syndrome 211 metabolic syndrome 211–13 mortality rates for psychiatric patients 207–9, 216 obesity 211 overlap in schizophrenia and affective disorders 207–18 potential to contribute to mental illnesses 209, 217 premature deaths among psychiatric patients 207–9, 216 syndrome X 211 polymorphous schizoaffective disorders 15 positional cloning 30 possession syndrome 120–1 prefrontal cortex abnormalities and sustained attention deficit 92–5 PFC-subtype depression 171, 175–6 role in depressive syndromes 169–71 pregnancy problems and glucose dysfunction in offspring 214 and schizophrenia in offspring 214 gestational diabetes 214 premature deaths among psychiatric patients 207–9, 216 Protocadherin X and Y gene pair 50–2 source of epigenetic modulation 52 psoriasis vulgaris, as a vulnerability factor 66 psychiatric classification cultural influences on diagnosis 104 diagnostic instability of disorders 104–5 DSM-IV and ICD-10 108–9 effects of comorbidities 104 epidemiological similarities 105 gaps in 105–6 historical approaches 106–7 influence of Kraepelin’s dichotomy 105–6 need for common terminology 104 origin of schizoaffective disorder concept 105–6 psychiatric illness, association with glucose dysfunction 213–16 psychogenic psychosis 8, 182, 183, 200–1

298

Index psychological therapies see cognitive behavior therapy psychomotor agitation, in depressive syndromes 157 psychopathology diagnostic systems 57–61 illness concept 59 operational criteria for diagnosis 58 quantitative, syndrome-oriented approaches 59–72 psychosis age of onset and form of psychosis 44–6 age of onset and sex 45–6 continuum concept 15–18, 43–4, 79–80, 184, 281–4 developmental concept 44–6 dichotomy concept 15–18 see also Kraepelin, Emil findings for candidate genes 46–7 inadequacies of the binary principle 43–4 possible sex chromosomal gene effect 47–52 predictive value of HPA axis hyperactivity 140–2 predictive value of individual phenomena 135–40 same sex concordance effect 47–8 see also functional psychoses psychotherapeutic approaches 174 see also cognitive behavior therapy psychotropic drug response ‘pharmaco-genetic’ component 68 possible non-specific drug mechanisms 67–9, 70 prediction 66, 69 ‘recovery-genetic’ component 67–9 significance of early improvement 70 time characteristics 66–9 quantitative genetic traits 25 quantitative, syndrome-oriented approaches to diagnosis 59–72 quantitative trait polygenic models 26–7 quetiapine treatment of schizoaffective disorders 253 treatment of suicidal behavior 240 RDC (research diagnostic criteria) for schizophrenia 43–4 reactive psychosis 8, 183, 280 religious ritual and obsessive–compulsive disorder 124–5 remitting schizophrenia 182 RGS4 susceptibility gene 33, 35–6 risperidone 83, 173 treatment of schizoaffective disorders 253, 256 treatment of suicidal behavior 238, 240 same sex concordance effect, in psychosis 47–8 schizoaffective disorders bipolar schizoaffective disorders 15 challenge to dichotomy concept 15–18 classification 106–7, 133, 145–6

classification validity 110–11 clinical course 146–50 cognitive deficit pattern 116 Cologne Study 8–9 concurrent type 10–14 continuum concept 8, 15–18 criticism of DSM-IV and ICD-10 definitions 108–10 cross-sectional definitions 11–14 cultural effects on manifestation and diagnosis 123–6, 127 diagnostic criteria 117 diagnostic validity 117–18 dichotomous view 133–4 DSM-IV and ICD-10 definitions 108–9 DSM-IV criteria 2, 10–14 DSM-IV criteria reliability and validity 118 etiology 111 evolution of DSM criteria 9–10, 18 evolution of ICD criteria 9, 10, 18 expressed emotions and relapse risk 126 family studies 111–13 genetic heterogeneity 113–14 genetic studies 113–14 history of efforts to define 1–10, 18 ICD-10 criteria 2, 10–14 incidence 114 interface between spectra 282–7 longitudinal definitions 10–14 monomorphous (episodes of same type) 15 nosological definition 109–10 origin of the concept 105–6 outcome based on subtypes 146, 149–50 outcome comparison with schizophrenia and affective disorders 146–50 outcome studies 118–19 polymorphous (episodes of various types) 15 prediction of archetypal outcomes 134–42 predictive value of historical features 134–40 predictive value of HPA axis hyperactivity 140–2 prevalence 114 prognostic indicators 150–1 Research Diagnostic Criteria (RDC) 108, 117 sequential type 10–14 spectrum views 134 suicide epidemiology 226–9 symptomatology 115–16 traditional cultures and mental health 122–4 unipolar and bipolar types 109, 150 schizophrenia age of onset and sex 45–6 and problems in pregnancy 214 association with diabetes 213–16 association with glucose dysfunction 213–16 chromosome regions implicated 30–8 diversity of operational diagnostic systems 43–4 expressed emotions and relapse risk 126 family genetic risk 27–9 genome scan meta-analyses 46–7 manifestation and diagnosis across cultures 123–6

299

Index predictive value of HPA axis hyperactivity 140–2 predictive value of individual phenomena 135–40 suicide epidemiology 226 susceptibility genes overlap with bipolar disorder 32–8 sustained attention deficit 80–1, 82–5, 90–4 trait marker 83–5 see also functional psychoses; psychosis schizophrenia spectrum, relationship of brief and acute psychoses 184–203 schizophreniform disorder 106, 183 schizophreniform psychosis 15, 280 schizotypal temperament 280 Schneider, Kurt 1–2, 3, 4, 5, 107, 278 serotonergic dysfunction, and suicidal behavior 235 sertraline 174 sex chromosomes events with evolutionary implications 48–52 Protocadherin X and Y gene pair 50-2 regions of X–Y homology 48–52 simple schizophrenia, classification 125–6 single major locus (Mendelian) genetic models 25–6 social deviance, psychologization of 123 state markers 79 state-modulated trait marker for bipolar disorder 86–7 state-related deficit in bipolar disorder 86–7 stress-vulnerability model of schizophrenia 236 suicidal behavior and HPA axis hyperactivity 235–6 and serotonergic dysfunction 235 association with cholesterol 236 drug treatment issues 237–41 effects of mood stabilisers 237, 238 family studies 234–5 future investigations for schizoaffective disorder 241–2 neurobiological findings 235–6 possible antisuicidal medications 238–41 possible genetic factors 234–5 possible risk factors for 229–34 preventative factors 233–4 stress-vulnerability model 236 treatment of depression in schizoaffective disorder 237–8 treatment of depression in schizophrenia 237–8 suicide epidemiology in mood disorders 225–6 in schizoaffective disorder 226–9 in schizophrenia 226 suicide risk, overlap between mood disorders and schizophrenia 224–5 susceptibility genes allelic heterogeneity model 36–7 common genes–different environment model 37 COMT (catechol-o-methyl transferase) 33, 35, 36

DISC1 33, 34 dysbindin 32, 33 factors affecting expression 26 findings for candidate genes 46–7 G72 33, 34 gene–gene interaction model 37 GRK3 (G-protein receptor kinase 3) 33, 34–5, 36 multiple threshold model 37 neuregulin 1 (NRG1) 32–4 overlap between schizophrenia and bipolar disorder 32–8 possible models of genetic overlap 36–8 RGS4 33, 35–6 transmission 25–6 sustained attention 80–2 Continuous Performance Test (CPT) 81–2 false responses to non-targets 81 mean response latency 81 response bias 82 target detection (‘hits’) 81 target sensitivity 81–2 vigilance decrement 81 sustained attention deficit comparison of schizophrenia and bipolar disorder 90–2 in bipolar disorder 81, 85–92 in schizophrenia 80–1, 82–5, 90–2 neuropsychological mechanisms of impairment 92–6 noradrenegic dysfunction 95–6 prefrontal cortex abnormalities 92–5 syndrome-oriented approaches to diagnosis 59–72 syndrome X, as physical comorbidity 211 systematic gene mapping 30 Third World, culturally-bound syndromes 120–1 topiramate, treatment of schizoaffective disorders 252, 255 trait markers 79 for bipolar disorder 86–7, 89 for schizophrenia 83–5 trazodone 174 treatment see cognitive behavior therapy; pharmacological treatment; psychotherapeutic approaches unipolar-bipolar distinction 281–2 valproate, treatment of schizoaffective disorders 252, 255 variable expressivity in genetic traits 26 venlafaxine 174 X chromosome, stability of structure 48 X-linked transmission of susceptibility genes 26 Y chromosome, variability across species 48–52 ziprasidone 173 treatment of schizoaffective disorders 253 treatment of suicidal behavior 240