Editorial

Declaration of conflicts of interest in scientific publicationsHans-Jurgen Moller ............................................................................................................................... 2

Review

Standardised rating scales in Psychiatry: Methodological basis, their possibilities andlimitations and descriptions of important rating scales

Hans-Jurgen Moller ............................................................................................................................... 6

Original Investigations

Isoprostenes as indicators of oxidative stress in schizophreniaAnna Dietrich-Muszalska, Beata Olas.................................................................................................... 27

A single session of rTMS over the left dorsolateral prefrontal cortexinfluences attentional control in depressed patients

Marie-Anne Vanderhasselt, Rudi De Raedt, Chris Baeken, Lemke Leyman, Hugo D’Haenen ............. 34

Intramuscular olanzapine versus short-acting typical intramuscular antipsychotics:Comparison of real-life effectiveness in the treatment of agitation

David J. Castle, Tudor Udristoiu, Chang Yoon Kim, Andrea Sarosi, Vladimir Pidrman,A. Nasser Omar, Juan Ignacio Rosales, Yuval Melamed, Turgut Isik,Jamie Karagianis , Tamas Treuer ........................................................................................................ 43

Tardive dyskinesia in a patient treated with quetiapineEmmanouil Rizos, Athanassios Douzenis, Rossetos Gournellis,

Christos Christodoulou, Lefteris P. Lykouras ..................................................................................... 54

Brief Reports

Clinical characteristics in long-term care psychiatric patients: A descriptive studyAnna Placentino, Luciana Rillosi, Emanuela Papa, Giovanni Foresti, Andrea Materzanini,

Giuseppe Rossi, Giovanni Battista Tura, Jorge Perez ......................................................................... 58

Abnormal microstructures of the basal ganglia in schizophrenia revealed by diffusion tensor imagingRyota Hashimoto, Takeyuki Mori, Kiyotaka Nemoto, Yoshiya Moriguchi, Hiroko Noguchi,

Tetsuo Nakabayashi, Hiroaki Hori, Seiichi Harada, Hiroshi Kunugi,Osamu Saitoh, Takashi Ohnishi.......................................................................................................... 65

Case Report

Catatonia as a risk factor for the development of neuroleptic malignant syndrome:Report of a case following treatment with clozapine

Thomas Paparrigopoulos, Elias Tzavellas, Panagiotis Ferentinos, Iraklis Mourikis, John Liappas ....... 70

Letters to the Editor

Adjunctive topiramate treatment for a refractory familial adolescent maniaFor-Wey Lung, Chun-Lin Liu, Chien-Shu Wang & Dong-Sheng Tzeng ................................................ 74

Electroconvulsive therapy for major depression in an elderly person with epilepsyKrzysztof Artur Kucia, Radosl aw Stepanczak, Beata Tredzbor ............................................................. 78

The World Journal of Biological PsychiatryVolume 10, No 1, 2009

Contents

www.informaworld.com/wjbpISSN 1562�2975

EDITORIAL

Declaration of conflicts of interest in scientific publications

Physicians’ and scientists’ conflicts of interest are

increasingly considered to be a serious problem,

particularly those related to sponsorship by the

pharmaceutical industry (Miller et al. 1999; Helm-

chen 2003; Moller 2006). In general terms a conflict

of interest exists when an author has interests,

financial or otherwise, that might inappropriately

influence his or her judgement, even if that judge-

ment is not influenced. Because of this, authors must

disclose potentially conflicting interests so that

others can make judgements about such effects.

A relationship between an author and sponsor that

includes some kind of financial support on a

personal or scientific level may result in a presenta-

tion of results that is biased in favour of the sponsor.

Given that such an author usually has a good

personal relationship with the sponsor, the author

may not even be aware of any bias and any bias is

usually unintentional.

To deal with this potential problem, most profes-

sional journals require that authors present their

conflicts of interest in a ‘‘disclosure’’ (Henderson

et al. 2003): the author names the source(s) of

financial support for the study or publication and

also states any other kind of financial or other support

he has received from the pharmaceutical company

concerned. Furthermore, the author is required to

mention any other relationships relevant to the topic

of the publication, such as support from other

companies. The reader should not assume from the

information that the author really does have a conflict

of interest; the ‘‘disclosure’’ simply serves to lay open

to the reader all the author’s relationships of relevance

to the matter at hand and in this way to sensitise the

reader to a possible author bias in the current

publication/presentation. However, it is often difficult

to decide whether and, if so, which types of conflicts of

interest should be presented. For example, should the

author focus only on financial or other support related

to the specific paper, or should he disclose all kinds of

financial or other supports by the industry? Some

journals do give clear instructions in this respect,

others do not. This leads to much uncertainty among

authors, putting them at risk to not being compre-

hensive enough regarding their disclosure.

The considerations about conflicts of interest

focus much too closely on financial support by the

pharmaceutical industry. Of course, material con-

flicts of interest also include those pertaining to the

author himself, e.g., any personal patents relevant to

the publication’s content. Furthermore, additional

influential factors are probably just as important, for

example focusing on a certain school of thought

(e.g., biological psychiatry versus psychotherapy),

rivalry between experts (Horton 1997), etc. How-

ever, it is much more difficult to quantify such

ideational and psychological factors than the pri-

marily ‘‘material’’ relationships with the sponsors of

a study, designed and organised by a pharmaceutical

company, or sponsoring by the pharmaceutical

industry of a study initiated by the author himself

(investigator-initiated trial, IIT). Since a ‘‘disclo-

sure’’ that is only concerned with financial support

by the pharmaceutical industry cannot fulfil such

ideational conflicts of interest, some colleagues

completely oppose a ‘‘disclosure’’ and have branded

it the ‘‘new McCarthyism’’ (Rothman 1993).

A disclosure of non-financial conflicts of interest is

in the field of psychiatry only demanded by very few

journals, such as the International Journal of Neuro-

psychopharmacology, for example. If one agrees that

not only financial but also ideational influences

should be disclosed, authors should be required to

name any allegiances to a specialisation or school of

thought. This may be of particular relevance if the

author is the official representative of a certain

professional scientific society: someone may be

biased in favour of the concepts or results of their

area of specialisation and not give adequate con-

sideration to other views. This is even more relevant

when an author writes about the areas of interest of a

different specialisation/school of thought; in such

cases, it is highly likely that he will view the other

area through his ‘‘tinted spectacles’’. In such cases,

one would have to differentiate between an uninten-

tional bias, which goes unnoticed by the author and

which simply happens because he is so closely

involved with his area of specialisation, and an

intentional bias, which he uses to promote his own

specialty at the cost of the competing one.

As mentioned above, bias may arise from an

individual having a narrow view of a subject or from

his working as an official representative of a certain

specialisation/school of thought. Mario Maj, the

The World Journal of Biological Psychiatry, 2009; 10(1): 2�5

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970802685032

current WPA president, recently underlined the need

for a disclosure of non-financial conflicts of interest

(Maj 2008). The competing areas of biological

psychiatry/psychopharmacology, psychotherapy and

social psychiatry are a relevant example in this

context. A respective ‘‘disclosure’’ would be indicated

even if such an author writes as an individual and not

in his official function (for example president of the

respective professional society). One could carry this

idea further and suggest that connections with a

political party, for example, should also be declared;

this may be particularly relevant in psychiatry. We

know that some parties have certain opinions about

psychiatry that lean more towards social psychiatry/

psychotherapy and less towards biological psychiatry/

psychopharmacotherapy, and vice versa. One would

be entitled to a certain extent to argue that presenta-

tion of these ideational influences in a conflict of

interest would go too far, and the argument of the

‘‘new McCarthyism’’ mentioned above would be

even more justified.

There are currently only a few tendencies to take the

issue so far. However, connections with health insur-

ance companies or political health institutions should

definitely be declared, not only material relationships.

For example, statements made by the drug commis-

sion of the physicians’ association about modern

psychoactive drugs also often show a tendency

towards a certain direction: in the view of many

experts, these statements do not correspond with

the available evidence but have obviously been

influenced mainly by health economic considerations

(Moller and Fritze 2008). It is thereby important to

ensure that such national or other public institutions

do not virtually completely deny the respective bias by

arguing that they are attempting to allocate resources

sensibly throughout the whole population: it must be

assumed that the interest of the government or

healthcare organisations, health insurance compa-

nies, etc., in keeping the cost of the health system as

low as possible may induce them to use a study design

or presentation that is biased from an objective,

scientific point of view (Maj 2008). This is becoming

increasingly obvious (Moller and Maier 2007). For

example, in the recent ‘‘effectiveness studies’’ on

second generation antipsychotics, which were sup-

ported by governments or certain healthcare organi-

sations, the advantages of modern antipsychotics over

first-generation antipsychotics were tendentiously

minimized not only in publications but sometimes

through the choice of scientific design (Moller 2007).

The first efforts to demand a ‘‘disclosure’’ were

definitely not far-reaching enough; they assumed

that the only distortions of the content of scientific

and educational publications, whether intended or

unintended, were related to the interplay between an

author and the pharmaceutical industry. Without

clearly defining what the ‘‘disclosure’’ should cover,

the regulations of the International Journal of

Neuropsychopharmacology, for example, demand a

wider approach to disclosure: ‘‘A conflict of interest

exists if authors, reviewers or editors have financial

or other interests (emphasis in original text) that could

inappropriately influence their power of judgement,

even if it is not actually influenced. For this reason,

authors, reviewers and editors must disclose their

potential conflicts of interest to allow others to form

an opinion about possible related implications’’. It

seems to be of special relevance that beside the broad

scope of the possible conflicts of interest, the journal

forces not only authors but also editors and

reviewers to disclose conflicts of interest.

Both the traditional focus of collaborations with

the pharmaceutical industry and the often one-sided

guidelines dealing with these collaborations have

been criticised. In a short commentary, Stossel

(2008) writes that there is usually a much lower

risk of bias in studies performed and published by

the pharmaceutical industry than, for example,

studies performed in a purely academic setting

without such external influences: ‘‘Conflict of inter-

est ideology purports to promote scientific rigour yet

is far from rigorous itself. Adverse outcomes objec-

tively ascribable to financial conflicts of interest are

almost non-existent, especially in the context of

overwhelmingly positive commercially driven medi-

cal advances. But purely academic research and

education are arguably less reliable than their

corporate or corporate sponsored counterparts.

They are not, for example, subject to stringent

Federal Drug Administration reporting require-

ments. Misconduct fells a single academic miscreant

but can bring down an entire company’’.

Even though this statement is probably too one-

sided and leans towards the other extreme, one

cannot avoid appreciating Stossel’s line of argumen-

tation. It is true that drug studies performed by the

pharmaceutical industry are very carefully monitored

and evaluated by the responsible authorities. One

only has to think of the sensational reports about

falsification of study results in academic medicine just

in the last ten years (Moller 2005). Such excrescences

are hardly imaginable within the pharmaceutical

industry. However, federal supervision of the phar-

maceutical industry is mainly concerned with study

design and conduct and not so much with publica-

tions. Publications of study results by the pharma-

ceutical industry may have a certain bias (Heres et al.

2006), so that demands for a respective ‘‘disclosure’’

are completely justified. However, one has to admit

that not only publications of studies performed by the

pharmaceutical industry can be biased.

Editorial 3

It is difficult enough to know where to draw the line

as far as disclosure of material influences is con-

cerned, but the situation becomes even more com-

plicated if ideational influences are also included.

The obligation to disclose these non-financial inter-

ests also comes increasingly into conflict with the

rights of the individual. Where should the limit be set

so that an author’s rights are not encroached upon,

for example to ensure that he does not become

‘‘transparent’’? On the other hand, if the focus is

solely placed on material relationships, there is a

danger that other important and formative influen-

cing factors will not be considered (Maj 2008).

In this context, attempts were made to design a

grading system for conflicts of interest. According to

Fava (2007), there is a ‘‘substantial conflict of inter-

est’’ if someone is an employee of a private firm, a

stockholder of a firm related to the field of research,

owns a patent directly related to the published work or

is a regular consultant or in the board of directors of a

firm. Fava does not consider occasional consultan-

cies, grants for performing an investigation, or receiv-

ing honoraria or refunds in specific occasions to be a

source of substantial conflict of interest. He states that

someone who has a leading position in a scientific

organization or is an editor of a medical journal must

be free of a substantial conflict of interest, and that

although such a ‘‘substantial’’ or ‘‘non-substantial’’

conflict of interest should not prevent something

being published, it should be disclosed (Fava 2007).

If one follows this grading approach, a one-time

invitation from a pharmaceutical company to a

congress may fall in the category of a ‘‘non-substantial

conflict of interest’’ or even below that.

The broad disclosure itself of all financial relation-

ships to pharmaceutical companies results in a

situation that pulls the serious objective of the

‘‘disclosure’’ movement in a different direction: the

misuse of a presentation of conflict of interest as a

seal of quality. At the moment this undesirable

development is especially obvious in the USA. At

the start of congress presentations, authors or speak-

ers proudly show one or even several ‘‘disclosure’’

charts crammed full of information showing their

manifold connections with the industry; the speaker

uses the disclosure to show in what high esteem he is

obviously held by the pharmaceutical industry. The

‘‘disclosure’’ thus becomes a seal of quality. I have

participated in presentations in which a speaker has

started his talk by presenting three ‘‘disclosure’’

charts filled with closely written text. One colleague

later got into great difficulties because he had

omitted an important ‘‘disclosure’’ (namely that he

had founded a biotechnology company that could

profit from certain parts of his findings) from one of

his publications. Thus, in individual cases over-

presentation of conflicts of interest may well be

accompanied by selective non-presentation of an

important conflict of interest.

It is important that the ‘‘disclosure’’ of cooperative

research activities with the pharmaceutical industry,

and associated financial relationships, does not result

in the author/speaker who reveals these relationships

being discredited. The disclosure does not mean that

the speaker really is biased, but gives the readers/

audience the chance to evaluate whether the state-

ments may be influenced by the special focus of

interest. A ‘‘duality of interests’’ is often involved, i.e.

a scientist who cooperates with a pharmaceutical

company is interested both in his primary responsi-

bility (the health of his patients, the success of his

research, or both) and in the cooperation with a

company that may have supported these objectives.

This duality of interests does not mean that one of

these primary responsibilities has to be sacrificed if

there is a conflict of interest with these competing

responsibilities (Komesaroff and Kerridge 2002; The

Royal Australasian College of Physicians 2006).

A further question is when a conflict of interest

should be disclosed and when not. One could say,

for example, that in all publications concerned with

psychopharmacotherapy, all possible related con-

flicts of interest should be stated, independent of

the drug directly referred to in the publication.

However, one could also limit this to only those

conflicts of interest that are related to the group of

psychoactive drugs being dealt with in the publica-

tion, e.g., neuroleptics, or even to only one particular

preparation. As regards other publications, e.g., on

the aetiopathogenesis of schizophrenia, it may be

justified to state that there is no conflict of interest

because a fair presentation of this topic is not

associated with the product-related interests of any

companies. But one could just as well claim that a

presentation on aetiopathogenesis, particularly if the

main emphasis is placed on certain neurobiological

factors, is actually also in the interests of the

pharmaceutical industry or of a certain pharmaceu-

tical company, and one would have to state any

relationship with the pharmaceutical industry as a

conflict of interest. The same applies vice versa to a

publication on psychotherapy, which appears

unsuspicious at first glance but may have been

written with a critical bias from the viewpoint of a

neurobiologically oriented psychiatrist and thus be

of interest to the pharmaceutical industry.

The thoughts about an additional disclosure of

conflicts of interest other than the material kind give

rise to the idea that a ‘‘disclosure’’ may be indicated

for many other topics, e.g., presentations on the

successes of psychotherapy.

4 Editorial

In contrast to the USA, other regions of the world

do not yet have a long tradition with presenting

conflicts of interest. Such ideas started there much

later than in the USA. Many European journals only

decided in the last few years to publish conflicts of

interest. Most of these have had a difficult job trying

to strike a sufficiently moderate path appropriate for

the European situation.

The problem is not as trivial as one might first

think. As discussed above, it is difficult to determine

what should be declared and what not. We need to

learn gradually so that every single author has time

to develop a good sense of what needs to be

disclosed. Even if one can provide a rough set of

criteria, there is still large room for personal discre-

tion, which depends on the individual author’s

sensibility. We have to raise this sensibility, for

example through discussions of the issue, as is

repeatedly happening at the moment. In my opinion,

and this is supported by discussions I have had with

my colleagues, this process will need years to

complete. Of course, it would be easy to demand

that ‘‘everything’’ has to be disclosed so that in the

end every publication includes a comprehensive and

perhaps highly differentiated presentation of con-

flicts of interest. But in my opinion this is exactly the

situation that would constitute the undesirable

development that can be seen in the USA today.

In order to clarify the problems faced, I would like

to report a personal experience. I authored a paper

on evidence-based medicine in psychopharma-

cotherapy (Moller and Maier 2007). In my opinion,

it was a totally ‘‘neutral’’ article (it did not name any

drugs, for example), so I did not think it necessary to

include a disclosure. The coordinating editor of the

journal then sent me a collegial note in which he

encouraged me to include a statement about con-

flicts of interest. I subsequently did so and included

the same details as I would for a presentation on the

treatment of depression in the USA, for example,

however without going into too much detail. Col-

leagues who consider the article to be ‘‘neutral’’ may

be surprised by my inclusion of conflicts of interest

and may even express criticism (‘‘Does he want to

make himself important? Does he want to give the

impression that he is particularly ethical?’’). These

were my exact thoughts when I initially submitted

the article without disclosing conflicts of interest.

As mentioned above, all of us have to learn to

disclose conflicts of interest sensibly without pre-

senting too little or too much. The latter may result

in the ‘‘transparent person’’ or even in the misuse of

the disclosure as a seal of quality. In my opinion, it is

not yet possible to reach a generally accepted

consensus about exactly what needs to be disclosed.

Further critical discussion is necessary, particularly

about the question whether only material conflicts

should be disclosed or also ideational conflicts and,

if so, to what extent.

Hans-Jurgen Moller

Chief Editor

Correspondence:

Prof. H.-J. Moller

Department of Psychiatry

Ludwig-Maximilians-University

Nussbaumstrasse 7

80336 Munich

Germany

Tel: �49 89 5160 5501 Fax: �49 89 5160 5522

Email: hans-juergen.moeller@med.uni-muenchen.de

References

Fava GA. 2007. Financial conflicts of interest in psychiatry. World

Psychiatry 6:19�24.

Helmchen H. 2003. Psychiater und pharmazeutische Industrie.

Nervenarzt 74:953�964.

Henderson C, Howard L, Wilkinson G. 2003. Acknowledgement

of psychiatric research funding. Br J Psychiatry 183:273�275.

Heres S, Davis J, Maino K, Jetzinger E, Kissling W, Leucht S.

2006. Why olanzapine beats risperidone, risperidone beats

quetiapine, and quetiapine beats olanzapine: an exploratory

analysis of head-to-head comparison studies of second-genera-

tion antipsychotics. Am J Psychiatry 163:185�194.

Horton R. 1997. Conflicts of interest in clinical research:

opprobrium or obsession? Lancet 349:1112�1113.

Komesaroff PA, Kerridge IH. 2002. Ethical issues concerning the

relationships between medical practitioners and the pharma-

ceutical industry. Med J Aust 176:118�121.

Maj M. 2008. Non-financial conflicts of interests in psychiatric

research and practice. Br J Psychiatry 193:91�92.

Miller FG, Pickar D, Rosenstein DL. 1999. Addressing ethical

issues in the psychiatric research literature. Arch Gen Psychia-

try 56:763�764.

Moller HJ. 2005. Are placebo-controlled studies required in order

to prove efficacy of antidepressants? World J Biol Psychiatry

6:130�131.

Moller HJ. 2006. Ethical aspects of publishing. World J Biol

Psychiatry 7:66�69.

Moller HJ. 2007. Do effectiveness studies tell us the real truth?

World J Biol Psychiatry 8:138�140.

Moller HJ, Fritze J. 2008. (Irrationale) Rationisierung von

Psychopharmaka im deutschen Gesundheitssystem. Zur Pro-

blematik von "Me-too"-Listen. Psychopharmakotherapie 15:

30�35.

Moller HJ, Maier W. 2007. Probleme der ‘‘evidence-based

medicine’’ in der Psychopharmakotherapie. Problematik der

Evidenzgraduierung und der Evidenzbasierung komplexer

klinischer Entscheidungsprozesse. Nervenarzt 78:1014�1027.

Rothman KJ. 1993. Conflict of interest. The new McCarthyism in

science. J Am Med Assoc 269:2782�2784.

Stossel TP. 2008. Has the hunt for conflicts of interest gone too

far? Yes. Br Med J 336:476.

The Royal Australasian College of Physicians. 2006. Guidelines

for ethical relationships between physicians and indusry. http://

www.racp.edu.au/page/publications-and-communications. 3rd.

Editorial 5

REVIEW

Standardised rating scales in Psychiatry: Methodological basis, theirpossibilities and limitations and descriptions of important rating scales

HANS-JURGEN MOLLER

Department of Psychiatry, University of Munich, Germany

AbstractStandardized rating scales are an important tool to improve the assessment of psychopathological symptoms in terms ofvalidity and reliability. Especially observer-rated scales are of primary importance in this respect. Self-rating scales can give ameaningful complementary view to the findings of observer-rated scales. Besides scales covering more or less the wholespectrum of psychopathological symptoms, specialised scales focussing only on symptoms of disorders like schizophrenia ordepression were developed. They are widely used, specifically for pragmatic reasons.

Key words: Rating scales, observer-rated scales, self-rating scales

Aims and Methods

Standardised methods of examination are used in

psychiatry to assess objectively and, in some cases, to

quantify psychopathological phenomena and other

clinically relevant domains, making it easier to

communicate these, to verify their status and to

analyse them statistically (Moller et al. 1996; Stie-

glitz and Baumann 1994). They are also essential to

develop models of psychopathology.

Major areas in which standardised procedures are

applied in psychiatry include the following: cross-

sectional quantitative description of psychopatholo-

gical abnormalities; assignment by a standardised

method of individual cases to diagnostic categories;

quantitative assessment of change over time in

psychopathological abnormalities (with or without

therapeutic interventions).

Those who favour intuitive phenomenological

methods have expressed the concern that applying

standardised methods cannot take sufficient account

of each patient’s individuality (Huber 1976). How-

ever, this criticism seems largely unfounded. There

are individual characteristics, which such standar-

dised methods will not fully capture; this is necessarily

the case, as standardised measurement instruments

tend to be constructed on the principle that a

symptom only qualifies for inclusion if it is present

in at least a specified minimum proportion of the

populations for which the instrument is intended.

However, when required, this deficit can be remedied

by using additional methods of investigation aimed at

capturing distinctive characteristics of individuals.

Indeed, reports of positive experiences with measure-

ment methods aimed specifically at the investigation

of individual cases may be as advanced as a counter-

argument. Such methods have a long history (Shapiro

1966; Brett-Jones et al. 1987).

Standardised measurement procedures can be

categorised on the basis of their methodologies into

standardised assessment instruments, systematic

behavioural analysis and objective tests in the

narrower sense of the word. The terms ‘‘standar-

dised assessment instrument’’ or ‘‘rating scale’’ are

applied to structured methods of assessing current

and/or past behaviour and/or experience, based on

lists of characteristics and, in some cases, descrip-

tions of these characteristics. The extent of standar-

disation ranges from a simple list of symptoms filled

in on the basis of a freely structured exploratory

interview to semi- or fully structured interview

schedules. These standardised assessment proce-

dures are especially suitable to examine the full

spectrum of psychiatric symptomatology; and, since

they are less restrictive than other procedures, are

additionally particularly practicable. A variety of

interview schedules are available and in general use.

Correspondence: Prof. Dr. Hans-Jurgen Moller, Department of Psychiatry, University of Munich, Nussbaumstrasse 7, 80336 Munich,

Germany. Tel: �49-89-5160-5501. Fax: �49-89-5160-5522. E-mail: hans-juergen.moeller@med.uni-muenchen.de

The World Journal of Biological Psychiatry, 2009; 10(1): 6�26

(Received 12 February 2008; accepted 9 June 2008)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970802264606

Systematic behavioural analysis involves using a

fixed set of categories to classify the quantity and

type of various forms of behaviour (including speech

and actions) occurring during a fixed observation

frame (methods involving sampling fixed periods of

time or particular events). This usually focuses on

manifest behaviour, and systems of categorisation

are often developed specifically to fit the particular

question being asked. This method has found

particular favour in the areas of behaviour therapy

and research about individual communication and

interactions.

Objective tests measure reactions to standardised

and fixed ‘‘stimulus material’’. They allow analysis of

specific particular psychological functions such as

perception, concentration, attention and intelli-

gence, usually from the point of view of perfor-

mance. This category includes tests of attention and

concentration, intelligence tests and a variety of

psychopathological incidences. These tests are said

to be objective as they cannot really be falsified by

the examiner or the subject and there are fixed

assessment criteria with corresponding methods of

data analysis and fixed norms.

Because they are very practicable, rating scales are

often preferred to other methods if the results of

patient examinations are to be performed in the

context of routine professional care. They are also

frequently applied in clinical psychiatric research,

such as clinical psychiatric research, clinical trials of

drugs, studies of longitudinal course, in routine

documentation or in epidemiological studies. Even

so, in terms of their level of precision, standardised

assessment measures are methodologically inferior

to objective tests and systematic behavioural analy-

sis. Despite the methodological superiority of these

latter methods, they tend to be included in clinical

psychiatric research only as supplementary measures

for the sake of completeness. An exception to this

can be found in the investigation of specific aspects

of cognitive functioning, e.g., the efficacy of drugs

on certain questions of differential diagnosis. This

limited use results not only from the amount of time

and effort involved in applying these tests. Particu-

larly in the case of objective tests, this limited use

results from the fact that the constructs which they

measure are rather more remote from the psychiatric

approach than the more complex phenomena issues

which can be described when using rating scales.

However, some would argue that they are closer to

the ‘‘core’’ of the disturbance.

Investigations of the relationship between the

three domains of assessment form the basis of

much research in psychopathology. Because stan-

dardised rating scales are so widely applied in

psychiatry practice and research, they will be the

focus of this article.

Scale construction, scoring methods and

quality criteria

Standardised methods of assessment or rating scales

allow description in terms of numerical values of

psychological abnormalities of various characteristic

forms. Different measurement scales allow the

degree of abnormality to be quantified to varying

extents. In the simplest instances, such as symptom

checklists, scales simply allow for rating of 0 or 1 to

be made for each symptom or complex of symptoms,

indicating whether or not it is present. More precise

assessment becomes possible if the construction of

the scale allows the severity of phenomena to be

described using a scale consisting of a series of levels.

Since there is a danger that different assessors will

base their evaluations on different standards, it is

important to establish a framework for the assess-

ment by providing anchor points, e.g., by giving

examples of situations which would be characteristic

for each point on the scale. Overly detailed assess-

ment, using an excessively broad scale is not mean-

ingful, as differences at the extreme end of the scale

cease to reflect real and significant variations in the

phenomena being examined, so that the differentia-

tions being made are not real ones.

In some assessment instruments the values at

which points on the scale are fixed may be varied

as required, so that a scale may be constructed which

is as finely differentiated as required; examples

include visual analogue methods of assessing sub-

jective well-being (Luria 1975). Since the measure-

ment of psychological phenomena is essentially

imprecise, a relatively coarse scale is usually ade-

quate, especially for comparisons between indivi-

duals, as a fine scale may have advantages for intra-

individual comparisons. However, any improvement

in measurement precision will generally be archived

not by refining the construction of the scale, but

rather by improving methods of measurement (von

Zerssen 1977). The values for characteristics be-

longing together (e.g., individual symptoms within a

syndrome) can be added to produce a summary

score. The extent to which characteristics belong

together to make up a syndrome is determined,

during the process of test construction (see below),

by applying multivariate statistical procedures such

as factor and cluster analysis. In some cases, before

adding up the figures for each characteristic to

produce a summary score, these figures will be

weighted to indicate the relative importance of

each characteristic within the syndrome. However,

if the characteristics have been shown to be relatively

Important rating scales in psychiatry 7

independent, theoretical or practical justification is

required for any such summation (Garety and

Hemsley 1987).

Measures of psychopathology obtained from stan-

dard rating scales generally have the level of mea-

surement of ordinal scales, i.e. they give only a rank

order and do not possess the measurement level of

an interval scale, in which there are equal intervals

between points on the scale. A fundamental problem

in measurement is that measurement instruments

with more detailed scales and higher levels of

precision tend to bring with them greater restrictions

regarding the phenomena to be measured. This

normally means increasing abstraction from the

theoretical or conventional understanding of the

phenomena/construct which is the starting point.

This is called the reliability/validity dilemma.

Standardised assessment instruments should meet

the following quality criteria derived from test theory

as far as possible (Lienert 1969; Fischer 1974; Sarris

and Rey 1981):

1. Objectivity: The results should not depend on

who carries out the assessment and analyses the

results. Procedure, analysis and interpretation

should be standardised so that, as far as

possible, the same results are obtained regard-

less of who administers the instrument, ana-

lyses it or interprets it.

2. Reliability: This refers to the reliability with

which a standardised assessment instrument

records a characteristic. When repeating the

measurement, the same result should be ob-

tained.

3. Validity: This is the extent to which the instru-

ment records what it is intended to record. The

connection between the results of measurement

and any external criteria available for assessing

what it is that should be measured should be as

close as possible.

4. Establishment of norms: Reference values for

different clinical groups and varying groups of

normal subjects and, where applicable, a re-

presentative sample of the general population

should be available.

5. Practicability: The amount of resources required

for administering standardised assessment in-

struments in terms of time, staff and material

should be as low as possible.

For a particular test, these criteria will not necessa-

rily be highly correlated with one another. For

example, a test of concentration may give a reliable,

but not a valid measurement of individual differ-

ences in performance when a test designed for

intellectually normal individuals is applied to people

with learning difficulties. In this group it no longer

functions as a measure of concentration but rather of

intelligence (Sarris and Lienert 1974). There is a

partial incompatibility inherent in the relationship

between reliability and validity (the reliability/valid-

ity dilemma): improvement in reliability is often

accompanied by a reduction in validity and vice

versa. This can be explained with other examples as

well.

Currently of special interest in the context psy-

chopharmacological trials is the approach to apply

observer-rated scales like the HAMD for depressive

patients, not in a face to face view between patient

and doctor, but using an interactive voice recording

system (IVRS; Kobak et al. 1996; Mundt 1997;

Kobak et al. 1999) by phone. This can increase the

reliability besides saving research budgets, but in this

context the HAMD is not used anymore like a real

observer-rated scale in the way it was constructed.

Due to this new approach the process becomes

similar to a self rating process and all its limitations

(see below!).

While for psychometric tests in the narrower sense

the availability of norms is largely taken for granted,

this has been approached with a great deal less rigour

for clinical rating scales. Thus, for example, the

Inpatient Multidimensional Psychiatric Scale

(IMPS; Lorr 1974) is almost the only observer-rated

scale measuring psychopathology for which norms

for a representative sample of the general population

are available (Hiller et al. 1986). For a number of

rating scales, reference values are available for

particular diagnostic groups. Referring to such

norms, or more precisely to reference values, has a

substantial impact on the interpretation of results.

For example, moderately high scores for the domain

of paranoid syndromes have quite a different sig-

nificance from moderately high scores for depressive

symptoms, in a way that depressive symptoms are

common in the general population, whereas para-

noid symptoms are not.

In producing norms for a standardised assessment

instrument, the usual starting point is the normal

distribution in order for it to be possible to derive

confidence intervals (see below) and to apply parti-

cular statistical tests, such as Pearson‘s product�moment correlation. Two values need to be known

to characterise a particular normal distribution: (1)

the mean of all scores obtained for the test and (2) a

measure of the extent of dispersion of these values.

Generally expressed in the form of standard proper-

ties of the Gaussian (normal) distribution allow the

proportion of subjects who will have a particular test

score to be calculated. Thus, for example, 68% of

patients will have a test value which falls within one

standard deviation either side of the mean and

8 H.-J. Moller

around 95% of patients will have a test value which is

no more than two standard deviations from the

mean. On the basis of the norm values it will

therefore be possible to calculate where a subject’s

score lies in relation to a reference population

(Figure 1).

Norm values for a particular test can be straight-

forwardly expressed by giving the mean and stan-

dard deviation. Once this information is available, a

statement may be made about the position of the

subject in relation to the reference population.

However, a disadvantage of referring to the numer-

ical value of the standard deviation for a particular

test is that it is difficult to compare the results

obtained by a particular subject for several different

tests. To allow comparisons of this kind to be made,

a z-value can be calculated; this is the result obtained

in a particular test expressed in terms of units of the

standard deviation for that test. Results obtained by

a particular subject in different tests may also be

compared using percentage rankings, by specifying

for each test what proportion of a reference popula-

tion has higher or lower scores for the test.

Various empirical methods may be used to test

whether the test quality criteria specified above have

been met. Appropriate ways of examining the

reliability of a test include test�retest reliability,

inter-rater reliability, the split-half correlation coeffi-

cient and internal consistency. To determine test�retest reliability, the same test is given to the same

group of people at two different time points. The

time between the two applications of the test

depends on the interval to which the test is intended

to apply. For tests with the aim to record enduring

personality traits, an interval between applications of

the test of 14 days to 1 year is recommended. For

tests with the aim to record rapidly fluctuating

characteristics (such as mood or subjective well-

being), a time span between several minutes and a

few hours is appropriate. Ideally, identical results

should be obtained for each measurement, but of

course this is not the case in practice, since

measurement errors occur necessarily (related to

strong influences caused by the test situation,

practice effects, etc.). The correlation between the

two values gives the test�retest reliability coefficient.

Deciding whether the reliability of a test is high

enough depends very much on the purpose of

administering the test (Lienert 1969). As a rule, a

reliability coefficient in excess of 0.8 is required.

Methods for which the test�retest reliability is below

0.5 are not generally useful. The measurement

accuracy of a test may be different for different

diagnostic groups (differential reliability).

Several different procedures also exist for deter-

mining the validity of a test, e.g., examination of

consensual validity, predictive validity, construct

validity and content validity. Consensual validity is

determined by correlating the results of applying the

tests to a sample of subjects with comparable data

obtained by methods other than the application of

the test (external criteria). For example, results for

the test may be correlated with corresponding scores

obtained for the same subjects for another test

examining the same psychological characteristics.

Whereas with consensual validity, test values and

external criteria are measured at the same time,

predictive validity is determined by investigating

whether events predicted on the basis of the test

results have actually happened. A classic example is

the correlation of test results from an intelligence test

with assessment at a later date of actual success at

school.

A requirement which needs to be met is that

following a translation of a scale from one language

into another, new validity tests must be carried out

with the translated version. This is also the case if the

scale is modified in any way.

Percentage of cases above the axis section, which are determined by standard deviations

Standard deviation

Cumultative percentage shares (%)rounded off Standard deviation

Percentage rankings

Typical standardised norms: z scale

Figure 1. Relationship between some frequently used standardised scales and the normal distribution curve.

Important rating scales in psychiatry 9

Taking all these test development procedures

together shows that test constructions really are a

time-consuming, long-lasting, complex process.

This is amongst others the reason why only a few

more or less well-validated rating scales are available.

The increasing fashion to generate and publish new

rating scales, e.g., in the context of certain research

prospects after a very short development period and

with low effort, is therefore not the right way. It is for

example not enough to take DSM-IV symptom

criteria for major depression, to provide a scoring

for these symptoms and to calculate the correlation

with values of corresponding cross-sectional scales,

e.g., with the HAMD, like it was done for SUM-D

(Sachs et al. 2002, 2007). Some correlations be-

tween SUM-D and traditional depression scales are

mentioned in Table I to emphasise this. The

correlation between the traditional scales and

SUM-D are quite low and unstable, depending

apparently on severity of depressive symptoms, while

the correlation between HAMD and MADRS is

high and stable.

Observer rating scales

Standardised assessment instruments relate to past

or current behaviour and experience. The extent of

psychological abnormalities is rated by using fixed

scales. These rating scales may focus on a single

aspect, e.g., anxiety (unidimensional scales) or on

several aspects (multidimensional scales) of psycho-

pathology. For each aspect of psychopathology

assessment may be based on a global rating or on

different elements within the aspect being assessed,

e.g., on individual symptoms of the depressive

syndrome. In this latter case the overall score of

the instrument is obtained by adding values for these

different elements.

The level of standardisation of standardised as-

sessment scales, also know as rating scales, falls

between that of unstructured clinical assessment and

that of objective tests. In some of these instruments,

standardisation is limited to providing guidelines

describing items and the categories used to assess

them, and to specifying a method of analysis (gen-

erally one or more summary scores are calculated).

In other scales, a time frame is also stipulated for the

assessment, and in some the framework in which

observation takes place is also fixed. In the latter

case, the instrument is referred to as a fully

structured or standardised interview. The more

extensive the standardisation procedures are under-

taken, the greater the reliability of an assessment

instrument generally becomes. However, a highly

standardised instrument tends to become less prac-

ticable. For this reason, both in everyday clinical use

and research, where resources are concerned, sim-

pler scales such as the Present State Examination

(PSE; Wing et al. 1974) are preferred to fully

structured instruments. The latter required a fully

structured interview, whereas the simpler rating

scales can be completed following a routine psychia-

tric interview. Particularly for the simpler rating

scales, inter-rater reliability for observer-rated in-

struments can be improved by systematic joint

training of the raters. In principle, fully structured

interview methods with extensive individual inter-

viewing ought to produce high inter-rater reliability

and should therefore have particular advantages in

multicentre, multinational studies in which great

discrepancies need to be taken into account, not only

in terms of how mental states are assessed, but also

in the psychiatric interview techniques applied.

Standardised assessment instruments may be

classified on the basis of who carries out the

assessment into self-rated and observer-rated instru-

ments. In observer-rated instruments, psychopatho-

logical abnormalities are identified by trained

assessors (e.g., doctors, psychologists, care staff,

lay people trained to administer the instrument) or

by relevant others (e.g., partner, relatives, friends).

The assessment refers to the behaviour and /or

experience of the patient and is based on the

assessor’s own observations and/or information gi-

ven by the patient. Observer-rated scales need to be

constructed in a way that they are appropriate to the

level of training of the particular types of inter-

viewers to be involved in their administration. Thus

Table I. Correlation coefficients of clinical-status assessments with standard rating scales (Sachs et al. 2002).

Depression scale Depressed Mixed/RC Roughening Recovering Recovered

Correlation coefficients of clinical assessments with standard rating scales

HRSD a SUM-D 0.617 (n�98) 0.695 (n�99) 0.869 (n�14) 0.756 (n�77) 0.788 (n�29)

MADRS a SUM-D 0.65 (n�42) 0.958 (n�8) 0.714 (n�4) 0.641 (n�30) 0.565 (n�10)

HRSD a MADRS 0.905 (n�42) 0.931 (n�8) 0.99 (n�4) 0.730 (n�30) 0.877 (n�10)

Non-parametric correlation coefficients of clinical assessments with standard rating scales

HRSD a SUM-D 0.630 (n�98) 0.638 (n�99) 0.693 (n�14) 0.8 (n�77) 0.784 (n�29)

MADRS a SUM-D 0.669 (n�42) 0.935 (n�8) 0.400 (n�4) 0.738 (n�30) 0.594 (n�10)

HRSD a MADRS 0.896 (n�42) 0.905 (n�8) 1 (n�4) 0.750 (n�30) 0.829 (n�10)

10 H.-J. Moller

there are scales for doctors trained in psychiatry,

e.g., the Brief Psychiatry Rating Scale (BPRS;

Overall and Gorham 1976), the Hamilton Depres-

sion Rating Scale (HAMD; Hamilton 1976), for

clinical psychologists, for care staff trained in psy-

chiatry, e.g., the Nurses’ Observation Scale for

Inpatient Evaluation (NOSIE; Honigfeld and Klett

1965) and also for patients’ relatives, e.g., the

Symptoms and Social Behaviour Rating Scales for

Relatives (Katz and Lyerly 1963).

Observer-rated scales mainly focus on the psycho-

pathological state, either with the global epidemio-

logical question of classifying each individual wholly

as a ‘‘case’’ or ‘‘non-case’’. Examples here are the

interview developed by Goldberg (1972), with the

recording of specific aspects of mental state such as

depression or anxiety. Or the Hamilton Depression

Scale or the Hamilton Anxiety Scale (Hamilton

1959, 1967). Or for assessment of the whole

spectrum of psychopathology, e.g., the AMDP

system (Arbeitsgemeinschaft fur Methodik und Do-

kumentation in der Psychiatrie 1995). In scales

which record the whole range of psychopathology,

there is usually an emphasis put on the detection of

symptoms of functional psychoses, while generally

only limited emphasis is put on other symptoms. For

specific measurements of the latter types of sympto-

matology, e.g., depression or anxiety (Moller 2003),

the use of scales which focus on them specifically is

recommended (Table II).

In addition to the mental state, domains such as

social adjustment may also be measured by observer-

rated scales (Morosini et al. 2000). Examples include

the Social Adjustment Scale (Weissman et al. 1981),

the Social Interview Schedule (SIS; Faltermaier et al.

1987; Moller et al. 1988) and the Disability Assess-

ment Schedule (Jablensky et al. 1980). These scales

offer the opportunity to access social findings in a very

differential way focussing on different domains and,

in the case of the SIS, involving also ratings of the

following three different aspects: objective condi-

tions, management and subjective satisfaction for

each domain (Figure 2). Although these scales

represent a sophisticated assessment procedure and

have an excellent psychometric standard, there is

unfortunately much reluctance to use them because

of the time consuming assessment. More widely used

is a global assessment easily to perform, the Global

Table II. Some classical observer-rated scales.

Domain Procedure Reference Abbrevation Characteristics

Several domains

in one scale

Brief psychiatric Rating Scale Overall and Gorham

(1976)

BPRS 18 symptom complexes,

overall score and 5 sub-scales

Comprehensive Psychiatric

Rating Scale

Asberg etal.(1978)

Kuny etal.(1982)

CPRS 65 items, 4 sub-scales,

2 overall scores

AMDP-System AMDP (1995) AMDP 140 items, 9 sub-scales,

3 overall scores

Depression Hamilton Depression Scale Hamilton 1976a HAMD 17-21 items, 2-6 subscales

Montgomery-Asberg Depression

Scale

Montgomery and

Asberg (1979)

MADRS 10 items, overall score

Mania Bech-Rafaelson Mania Scale Bech etal.1978, 1991 BRMAS 11 items, summary score

Young Mania Rating Scale Young 1978 YMRS 11 items, overall score

Schizophrenia Positive and Negative

Symptom Scale

Kay etal. (1988) PANSS 30 items, summary scores,

3 sub-scales

Anxiety disorders Anxiety Status Inventory Zung (1976a) ASI 20 items, summary score

Hamilton Anxiety Scale Hamilton (1976b) HAMA 14 items, summary score

Obsessive-compulsive

disorders

Yale-Brown Obsessive Compulsive

Scale

Goodman et al.

(1989a,b)

YBOCS 10-items

Dementia Alzheimer’s Dementia Assessment Mohs and Cohen 1988 ADAS 21 items, 1 summary score,

2 sub-scores

100

150

200

250

300

X

x

0 M

X

x

S

x

o

Figure 2. Box plot representations of SIS total scores for the

schizophrenic patients and matched controls, showing problems.

O, objective conditions; M, management; S, satisfaction; 50% of

all data are included in the box, the top of the box represents the

third quartile, the bottom the first quartile, the line in between;

the median; x, far outlier; 0, near outlier (Moller et al. 1988).

Important rating scales in psychiatry 11

Assessment Scale (GAS; Endicott et al. 1976), a later

version of which the Global Assessment of function-

ing (GAF; Endicott et al. 1976) is integrated in the

DSM-IV system for the rating of axis V. This scale

provides anchor points to describe the functioning

together with the symptom burden on a 100-point

global scale. The approach of the Social and Occupa-

tional functioning Assessment Scale (SOFAS) is

nearly identical, apart from the fact that it focuses

only on functioning without considering the symptom

burden (Goldman et al. 1992). A recent further

development is the Personal and Social performance

scale (PSP; Morosini et al. 2000), which follows the

same approach but differentiates the assessment on a

global rating of a few different domains.

Personality characteristics (Moller and von Zers-

sen 1987) and personality disorders can also be

described using observer-rated instruments, as de-

veloped in the new methods of assessment of

personality for the rating of axis II (personality

disorders) of DSM-III, DSM-IV and ICD-10

(Stangl et al. 1985; Sass 1986; Pfohl et al. 1989;

Loranger et al. 1994; Maffei et al. 1997; Jablensky

2002; Fossati et al. 2006; Trull et al. 2007).

Standardised assessment instruments may also be

used to document the side effects of treatment with

psychotropic drugs, as in the scales developed by

Simpson and Angus (1970) for the documentation

of extrapyramidal motor side effects, or in the UKU

Side Effect RATING Scale, which includes a full

spectrum of psychotropic drug side effects (Ling-

jaerde et al. 1987).

On the basis of multivariate statistical analysis

(factor and cluster analysis), the data obtained from

administering rating scales may be used to derive

factors. These factors identify groups of individual

symptoms which tend to occur together. If we

consider that the term ‘‘clinical syndrome’’ generally

refers to a group of symptoms which frequently

occur in combination, it then becomes apparent that

the factors extracted from rating scales relating to

mental state are conceptually identical to clinical

syndromes. Multivariate analysis of the data ob-

tained from different multidimensional psychiatric

rating scales applied to different samples of patients

has tended to repeatedly generate the same factors or

symptom clusters (Lorr et al. 1962; Mombour

1974a,b; Gebhardt et al. 1981; Cairns et al.

1982a,b): paranoid hallucinatory syndrome, manic

syndrome, depressive syndrome, apathetic syn-

drome, hypochondriac syndrome, phobic-obsessive

syndrome, amnesiac syndrome. For some well-

developed observer-rated scales it has been shown

that the factor structure also remains relatively stable

across different studies and for many of the factors

this is true even with repeated measurements in the

course of treatment (Baumann and Stieglitz 1983;

Moller and Hacker 1988; Steinmeyer and Moller

1992b). This invariability of the structure of factors

across different samples and time points is an

important aspect of the validity of a scale (factorial

validity). Different psychiatric diagnostic groups are

reflected in different characteristic syndrome profiles

when rating scales are applied.

It is important to bear in mind that identically

named syndromes from different scales may vary

greatly in terms of the items included, and the

correlation between analogous syndrome scores is

not always very big. The more syndromes are

represented in a scale, the wider the range of its

potential applications will be. However, in order to

address specific questions these broadly applicable

rating scales should be combined with other specific

observer-rating scales. In the interests of economy,

when research questions are narrow in scope (e.g., if

they concern depression), it makes sense to admin-

ister specific scales which focus on the syndrome of

interest, e.g., depression. As scales which measure

the same domain (e.g., depressive symptoms) some-

times focus on different aspects of this domain

(Mombour 1976), or have different scale-related

characteristics, certain questions may be best ad-

dressed by using a combination of them. In terms of

developing models of psychopathology, there is a

continuing debate on the relative merits of single-

symptoms versus syndrome research strategies

(Bentall 1992; Lenzenweger 1999). At present, the

approaches are best viewed as complementary.

When observer-rated instruments are adminis-

tered by professionally trained observers, it is usually

assumed that, in making the rating, the observer

decides how much weight is put on the information

the patient gives: e.g., an observable improvement in

general behaviour and demeanour is taken into

account in the rating, even if the patient gives no

clear report of this improvement. An advantage of

this expert assessment is that it reduces the scope for

inaccurate assessments resulting from the distortions

in patients’ perception of themselves, but on the

other hand it introduces the danger of distortions

which are related to the assessment (rater bias).

Systematic distortion in the assessor’s observa-

tions (Hasemann 1971) can result from the follow-

ing factors in particular:

1. Rosenthal effect: The assessor’s expectations

influences the result of the assessment.

2. Tendency on the part of the rater to system-

atically over- or under-rate the degree of dis-

turbance.

3. Halo effect: The results of assessment of one

characteristic are influenced by the rater’s

12 H.-J. Moller

knowledge of the subject’s other characteristics

or by the overall impression made by the

subject.

4. Logical errors: The result of the assessment is

influenced by assessors reporting only those

detailed observations which make sense to

them in the context of their theoretical and

logical preconceptions.

These errors may be partially compensated for by

combining observer-rated scales with self-rated

scales (von Zerssen 1979; von Zerssen and Moller

1980; von Zerssen 1982).

Most rating scales allow the description of current

state and, with repeated application, can also be

used to examine change over time.

Generally, any multidimensional instrument for

examining psychopathology allows trying to make a

diagnostic classification by applying specific algo-

rithms, e.g., to detect characteristic symptom pro-

files. As would be expected, the results of nosological

classification based solely on psychopathological

rating scales are not very satisfactory since diagnosis

is generally also based on information about clinical

history and on hypothesis about the cause of the

illness. The Category system of Wing et al. (1974) is

based on the PSE, together with a supplementary

scale describing history. This combination allows

satisfactory results to be obtained in the sphere of

functional psychoses and it has been applied in a

variety of large national and international research

projects (Wing et al. 1974).

In connection with the development of an oper-

ationalised diagnosis system such as the Diagnostic

and Statistical Manual III-R and IV (DSM-III-R,

DSM-IV), standardised instruments have been pro-

duced to allow examination of the aspects of clinical

history and psychopathology on which the operatio-

nalised diagnostic criteria are based. The Schedule of

Affective Disorders and Schizophrenia (SADS) was

developed specifically for the RDC (Spitzer et al.

1975).

A series of fully structured interview schedules

and diagnostic instruments developed in the last

decade allow ICD-10 and/or DSM-III and DSM-IV

diagnoses to be generated: the Composite Diagnos-

tic Interview (CIDI; Wittchen and Semmler 1991;

Wittchen and Semmler 1997), the Structured Clin-

ical Interview for DSM (SCID; Wittchen et al.

1991; Wittchen et al. 1997) and the PSE-based

Schedules for Clinical Assessment in Neuropsychia-

try (SCAN; WHO 1991, 1999). Overall, all these

instruments seem likely to lead to a considerable

increase in inter-rater reliability in mental state

assessment and diagnostic classification.

Self-rating scales

In self-rated instruments, patients themselves can

describe past or current behaviour and experience

on the basis of fixed rating scales. Self-rated scales

have the further advantage that their use is very

economical for the assessor and eliminates observer

bias. However, their use also introduces the dis-

advantage that conscious or unconscious tendencies

to falsify responses (e.g., tendencies to exaggerate or

conceal symptoms, the positive response bias, social

desirability effects) will have greater impact on

patients and are only partially detectable through

the use of control scales (so-called lie detector

scales).

Self assessment procedures are currently available

for different psychiatric disturbances, e.g., the Beck

Depression Inventory (Beck et al. 1986), the Clinical

Self-Rating Scales (CSRS; von Zerssen 1976a,b,c,d,

1986) or the Self-Report Symptom Inventory (SCL-

90; Derogatis 1977a). Self-rated scales in from of

visual analogue scales (so-called barometer scales on

which particular dimensions or current experience

are graphically represented) are especially useful for

intra individual studies of course over time (Luria

1975). Self-rating procedures are also used in several

personality scales, such as the MMPI (Hathaway

and McKinley 1963; Hathaway and McKinley

1989), Neo-PI and Neo-FFI (Costa and McCrae

1992). A problem with most personality scales in

clinical use is that, contrary to their goals, they do

not distinguish precisely between habitual person-

ality dispositions and current disturbances in beha-

viour (von Zerssen 1993, 1994).

As with observer-rated scales, self-rated scales can

also record domains other than psychopathological

abnormalities, e.g., social adjustment: the self rating

version of the Social Adjustment Scale (SAS; Weiss-

man and Bothwell 1976) or the Sheehan Disability

Scale (SDS; Rush 2000), a self rating version of the

observer rating scale SOFAS. Quality of life is

traditionally seen as a domain for a self rating

approach and several self rating instruments are

available (Bullinger et al. 1995; Bullinger and Kirch-

berger 1998; Pukrop et al. 2000; Pukrop et al. 2003),

even for the assessment of effects of psychotropic

drugs (National Institute of Mental Health 1976).

Apart from a few scales measuring current mental

state which, as with the Self-Report Symptom In-

ventory (SCL-90), record a very broad spectrum of

psychopathological symptoms, most self-rated scales

focus on specific aspects of disturbance of subjective

experience (Table III). Examples are depressive

symptom scales (Beck et al. 1961; Zung 1965; von

Zerssen 1976d) or measures of subjective well-being

(von Zerssen 1976b). One of the advantages of this

Important rating scales in psychiatry 13

approach is that the quantity of items is limited, a

particular strength where severely disturbed psychia-

tric patients are concerned. In order to obtain a

sufficiently clear view of the current psychological

state from a subjective point of view, it is always better

not only to present a checklist of adjectives describing

complaints but also to add other symptom-oriented

scales like the depression self-rating scale.

However, very precise differentiation between

different aspects of ‘‘subjective state’’ is probably

not generally meaningful (von Zerssen 1979) in

contrast to the detailed measurement of psychologi-

cal disturbances which may be made by observer

assessment. In fact, where results from clinical self-

rating scales are compared with observer-rated scales

administered by specialists, it seems that the various

dimensions of the subjective state, which self-rated

instruments describe, are more similar to one

another than the different aspects of psychopathol-

ogy delineated by clinical observer-rated assess-

ments. This is indicated by a joint factor analysis

of data from observer ratings and self-ratings of

mental state for example (von Zerssen and Cording

1978). This finding was reported from a study in

which the Inpatient Multidimensional Psychiatric

Scale (IMPS; Lorr 1974) was applied as an ob-

server-rated measure, while the CSRA (von Zerssen

1976a,b,c,d) was applied as a self-assessing measure.

The self-assessed data were mainly represented in a

single factor, the first to emerge, while the observer-

rated data were distributed across five further

factors. However, it cannot be concluded from this

secondary factor analysis (in which the primary

factors derived from the initial analysis of each scale

were also entered as variables) that self-assessment

simply produces a factor reflecting a global tendency

to complain rather than a differentiated picture of

subjective impairment. Primary factor analysis,

based on single items from the CSRS and also on

other self-rating instruments, indicates that certainly

a differentiation can be made at a subjective level

between different dimensions of disturbances, such

as depressiveness, paranoid tendencies and somatic

complaints. However, the depressiveness factor is

closely associated with the various other types of

subjective disturbance.

The level of agreement between self-assessment

and observer assessment is variable (Figures 3 and 4)

and depends amongst other things on the type of

disturbance and on symptom severity (Prusoff et al.

1972a,b; Bailey and Coppen 1976; White et al.

1984; Moller 1991, 2000; Sayer et al. 1993). For

example, when depressive symptomatology is severe,

as at the time of inpatient admission, an agreement is

substantially more limited than after partial remis-

sion of symptoms at the time of discharge. This is

probably connected to the greater limitation of the

capacity for self-observation among the severely

depressed. And probably also to the fact that

observers tend to recognise very severe depressive

symptoms on the basis of non-verbal evidence to a

greater extent than less severe depressive symptoms,

where the patients verbal reports are more impor-

tant. Compared to patients with endogenous depres-

sions, those with ‘‘neurotic’’ depressions (dysthymia)

show a greater tendency to overstate their symp-

toms. Degree of agreement between self-rating and

observer rating is substantially greater for the

amount of change, as measured in longitudinal

studies, e.g., in the context of treatment studies,

than when psychopathological phenomena are re-

corded at a single cross-sectional time point (von

Zerssen 1986; Moller and von Zerssen 1995).

Table III. Some classical self-rating scales.

Domain Procedure Abbrevation Author(s)

Several domains Self-Report Symptom Inventory SCL-90 SCL-90-R Derogatis et al. (1976) CIPS (1996)

Depression Beck depression Inventory BDI Beck et al. (1986)

Depressivitats Skala

(Depressive Symptom Scale)

DS von Zerssen (1976b, 1986)

Befindlichkeitsskala (Actual Mood Scale) Bf-S von Zerssen (1976c, 1986); CIPS

2005

Anxiety disorders Self-Rating Anxiety Scale SAS Zung (1976b), see also CIPS (1996)

State-Trait Anxiety Inventory STAI Spielberg (1983)

Figure 3. Responder under antidepressant therapy as assessed by

psychiatrists ratings (HAM-D) and self-ratings (AMS) (Moller

2003).

14 H.-J. Moller

The item ‘‘selection’’ is also of importance for the

achievable degree of correlation between self ratings

and observer ratings. It seems plausible that scales

which are coming close in their item ‘‘selection’’ on

both scales might have a higher degree of concor-

dance than, e.g., an observer-rated symptom scale

like the HAMD and an adjective mood scale for self

rating. However, due to the factors mentioned above

and a possible different interpretation of the wording

of symptom descriptions by doctors and patients,

there are still limitations of concordance. This can

lead to relevant differences in score values (Bailey

and Coppen 1976; Sayer et al. 1993; Uher et al.

2008) and in categories depending on the score

values, like, e.g., response or remission criteria. This

becomes evident (Figure 5) for example through a

study by Rush et al. (2003). For all these reasons,

caution is advised in applying self-rating scales

without using observer rating scales at the same

time, at least for measuring relevant outcome criteria

(Figure 5).

Multi-methodological diagnostic procedures in

which a combination of self-rated and observer-

rated scales are applied offer the best guarantee of

satisfactory description of both, subjective and

objective psychopathological states.

Measures of subjective well-being are of particular

interest in the area of treatment assessment, i.e.

particularly visual analogue scales, sometimes also

called barometer scales. These measure current

disturbances of psychological well-being and lend

themselves especially well to repeated measurement.

These methods allow a very good description at the

self-assessment level of response to a therapeutic

intervention. Modern methods of statistical analysis,

such as some of the procedures developed for time

series analysis, allow satisfactory analysis of such

data (Moller et al. 1987, 1989; Morley 1994).

Examples for important observer rating scales

AMDP system

The AMDP scales are most frequently used in the

German speaking psychiatry (Arbeitsgemeinschaft

fur Methodik und Dokumentation in der Psychiatrie

2000; Arbeitsgemeinschaft fur Methodik und Do-

kumentation in der Psychiatrie and Collegium

Internaationale Psychiatriae Scalarum 1990). It

developed from the attempt to turn the complete

system of the descriptive German psychopathology

based on Karl Jaspers, Kurt Schneider, etc., into an

observer scale that meets the requirements of

modern standards. This scale contains psychopatho-

logical forms as well as forms for anamnesis and

somatic complaints. The psychopathological form

comprises, in its approximately 100 items, the most

important symptoms of psychoses, affective disor-

ders, justifiable and other psychiatric disorders and

40 items on somatic and vegetative signs.

The AMDP System is a very comprehensive scale,

based on nine dimensions found through factor

analysis (Sulz-Blume et al. 1979; Baumann and

Stieglitz 1983; Bobon 1985; Troisfontaines and

Bobon 1987; Cuesta and Peralta 2001). It exten-

sively shows the complex psychopathological symp-

tomatic in cross-sectional and longitudinal sectional

figures (Egli et al. 2008), such as, e.g., the following

syndromes: paranoid-hallucinatory syndrome, man-

ic syndrome, hostility syndrome, depressive syn-

drome, apathy syndrome, vegetative syndrome,

obsessive syndrome. Contrary to the Present State

Examination (Wing et al. 1974), which mainly

targets the symptomatic of schizophrenic psychoses,

the AMDP System is not fully structured but leaves

Figure 4. Non-responder under antidepressant therapy as as-

sessed by psychiatrists ratings (HAM-D) and self-ratings (AMS)

(Moller 2003).

Figure 5. Differences between depressive self-rating and observer

rating scales in determining remission. Time to remission as

determined by total score at exit for 30-item Inventory of

Depressive Symptomatology, Self-Report [IDS-SR30] (514),

16-item Quick Inventory of Depressive Symptomatology, Self-

Report [QIDS-SR16] (56) and 24-item Hamilton Rating Scale

for Depression [HAM-D24] (58) (data set from nefazodone

study: Keller et al. (2000) New Engl J Med.

Important rating scales in psychiatry 15

the examining physician the possibility of a free

exploration. This facilitates the use in the clinical

practice. Syndrome shifts, e.g., from a schizophrenic

to an affective symptomatic, can be well evaluated

since the AMDP is not focused on a specific clinical

picture. The instrument is therefore preferably used

for diagnostics in course studies (Moller et al. 2002;

Bottlender et al. 2003). Although AMDP has been

translated into many languages (Guy and Ban 1979,

1982), it is mainly used in the German-speaking

psychiatry. For pragmatic reasons in clinical psy-

chopharmacology, shorter and internationally more

common observer scales are used, usually restricted

to one or only a few dimensions such as e.g.,

depressive syndrome, manic syndrome, etc.

The items are worded in psychiatric terminology,

explained in a glossary. ‘‘Subjective’’ and ‘‘objective’’

specifications are assessed together. As also known

from other observer scales, problems occur in this

process, specifically regarding the assessment of the

efficiency of the patient in different areas, e.g.,

memory, concentration, etc. This is because state-

ments regarding these areas can be characterized by

feelings of insufficiency which can develop in con-

nection with depressive syndromes (Busch et al.

1975). Attempts to solve these problems were

carried out through respective encoding laws.

Reliability studies proved satisfactory, especially

after respective assessment training (Gebhardt and

Helmchen 1973; Busch et al. 1975, 1980; Woggon

et al. 1978; Kuny et al. 1983; Renfordt et al. 1983).

Problems concerning the inter-rater reliability oc-

curred less regarding the existence or non-existence

of symptoms, but specifically regarding the grading.

This is also known from experiences with other

observer scales and caused by the fact that the

assessment of the intensity of the parameter value

and its duration should be jointly assessed.

Positive aspects of the validity of the scale are

amongst others the diagnostic differentiability of

various psychiatric groups, the correlation of the

results with simultaneously used other scales, and

the sensibility for therapy-related and progress-

related psychopathological changes, respectively

(Mombour et al. 1973; Bente et al. 1974; Pietzcker

et al. 1981; Baumann and Stieglitz 1983; Bottlender

et al. 2000; Moller et al. 2002).

BPRS

The ‘‘Brief Psychiatric Rating Scale’’ (BPRS; Overall

and Gorham 1962, 1976) was developed from two

much longer scales (such as IMPS), developed by

Lorr and colleagues. Mainly symptoms which

showed a significant change under psychopharma-

cological therapy were included. The original scale

consists of 16, the modified version consists of 18

items (Overall and Gorham 1976), which cover

symptom complexes mainly from the area of psy-

choses and to a minor degree also from depressions.

A factor structure with a total of five factors was

developed on the basis of factor analytical studies

conducted in patients with a diagnosis of schizo-

phrenia. The inter-rater reliability reached a satis-

factory coefficient (Overall and Gorham 1962;

Cicchetti and Aivono 1976).

During validity studies, patients of different diag-

nosis groups could be distinguished due to the

BPRS. The scores mostly correlated closely with

the analogue scores of other scales (Mombour et al.

1975; Freudenthal et al. 1977; Woggon et al. 1979).

The usability of this internationally very frequently

used scale for the diagnostic findings of psychophar-

macological therapy results was repeatedly proven

(Overall and Gorham 1972).

Despite its shortness and some shortcomings

regarding the item formulation, the scale offers

relatively good diagnostic possibilities. However, it

is not useful for a detailed diagnostic. Different to

broader scales, the danger of bias is given since

relatively broad categories have to be evaluated. In

the American psychiatry it was the mostly used scale

for standardized examinations of patients with

schizophrenic psychoses just because of this short-

ness and practicability.

Although the BPRS continues to be considered

‘‘gold standard’’ for the evaluation of schizophrenic

analysis, specifically in psychopharmacological re-

search, it is increasingly replaced by PANNS, the

‘‘Positive and Negative Syndrome Scale’’ (Kay et al.

1988), which better comprises the negative sympto-

matic beside productive symptomatic. However, the

BPRS is integrated in the PANSS, so that the BPRS

score can be calculated with PANNS data at hand.

PANSS

The PANSS consists of 30 items (symptoms), each

of which has seven levels of severity. The scope of

symptoms and evaluation basis (source of informa-

tion) of each item are described in accompanying

text, and the scale ranks are operationalised through

detailed symptom-specific information. There are

seven items each in the subscales for positive and

negative symptoms (P, N; score range from 7 to 49

points); 16 items make up the subscale to assess

general psychopathological symptoms (G; score

range from 16 to 112 points, Table IV). For each

of the three subscales (positive, negative and general

psychopathology scales), the scores of the respective

items are summated to give the respective subscale

score. The score value reflects the severity of

16 H.-J. Moller

symptoms in the respective symptom area. Although

not planned in the original publication of the scale,

often a total score is calculated from all 30 items.

Evaluations on the basis of multidimensional symp-

tom models are further on possible. The raw scores

can be transferred into percentile and t values, based

on an American random sample of 240 schizophre-

nic inpatients. Published study results can serve as

reference data for different patient groups. In a

number of studies, a high correspondence in the

evaluation of the different raters is reported (Muller

and Wetzel 1998). The factor analysis of PANNS in

different language versions, including German, sug-

gest by the majority a five-dimensional symptom

model. Mostly identified are separate components

for positive, negative and cognitive symptoms, as

well as for agitation/hostility and anxiety/depression,

at which random sample-related variations and

methodological variations of the item factor classifi-

cation are found (Lindenmayer et al. 1994; von

Knorring and Lindstrom 1995; Marder et al. 1997;

White et al. 1997; Mass et al. 2000; Wolthaus et al.

2000; Emsley et al. 2003). The PANSS has proven

treatment sensitive in a number of studies with

antipsychotic psychopharmaceuticals. Differences

in the treatment effects could be demonstrated in

comparison, both between classical and atypical

psychopharmaceuticals as well as between classical

and atypical psychopharmaceuticals and placebos.

Depending on the patient population and the test

matter, the assessment rests mainly on the positive

scale (McEvoy 1994), the negative scale (Lapierre

et al. 1999) or the total score (Moller et al. 1997;

Sanger et al. 1999).

HAMD

The Hamilton Depression Scale (HAMD) (Hamil-

ton 1960, 1967) became one of the first observer

rating scales for depression to gain worldwide accep-

tance, although its weaknesses are increasingly criti-

cized (see below). The original version of this scale

contains 17 items; later versions contain 21 or even 24

items. The formulation of the items is not always

precise enough and is considerably worse than in the

Montgomery�Asberg Depression Rating Scale

(Schmidtke et al. 1988), for example. Additional

information from relatives and friends, etc., can be

considered in the rating. In addition to the possibility

of calculating a total score, it is also possible to

calculate factor scores during the final analysis

(Hamilton 1960). However, there is no uniform

solution since the results of factor analytical evalua-

tions resulted in solutions of two to six factors

(Baumann 1976; Hamilton 1960; Hamilton 1967).

The inter-rater reliability can be seen as very high, at

least on the level of the total score (Hamilton 1960;

Waldron and Bates 1965). The correlation with the

Clinical Global Impression of the depressivity indi-

cates the validity of the scale (Welner 1972), as does

the sensitivity for the recording of antidepressant-

induced changes, which has been demonstrated in

numerous antidepressant studies (Figure 6). Refer-

ence values for various clinical samples are available,

but no norm values from a representative healthy

population. A literature review of control groups in

clinical trials of depression reported a mean HAMD-

17 score of 3.293.2 (SD) among healthy control

individuals (Zimmerman et al. 2004b).

However, some further problems of the scale still

remain unsolved for example regarding the content.

The scale does not record certain diagnostically

specific areas that are partially depicted in other

depression scales and therefore proves to be unsatis-

factory for a different diagnosis of depressive dis-

orders, particularly under the aspect of a differential

diagnosis. As regards the content, it is questionable

whether the characteristic ‘‘daytime fluctuations’’

makes a difference in the sense of higher depression

HAM-D total score

***P<0.001

Agomelatine

placebo

0

5

10

15

20

25

30

W6/8W4W2W0

N=358

N=363

∆=2.86***

∆=2.39***

∆=1.54***

LOCF

Figure 6. Efficacy of agomelatine in depression: combined

analysis of three placebo-controlled studies (Montgomery 2006).

Table IV. PANSS items.

P Positive Symptoms: 7 Items (P1 to P7)

Delusional ideas, formal thought disorder, hallucinations,

agitation, delusions of grandeur, distrust/persecution mania,

hostility.

N Negative Symptoms: 7 Items (N1 to N7)

Blunted affect, emotional withdrawal, lacking affective rapport,

social passiveness and apathy, difficulties with abstract thinking,

lack of spontaneity and fluency of speaking, stereotyped thinking.

G General psychopathology: 16 Items (G1 to G16)

Health concern, anxiety, guilt feelings, tension, mannerisms and

posturing, depression, motor retardation, uncooperativeness,

unusual thought content, disorientation, lack of attention, lack of

ability to judge and insight, weak will, lack of impulse control,

egocentrism, active social avoidance behaviour.

Important rating scales in psychiatry 17

score. This can lead to contradictions in the diagnosis

of the course of the disease in view of the clinical

experience that the most severe endogenous depres-

sions often show no daytime fluctuations at first and

that these only occur upon improvement of the severe

depressive mood. The fact that sleep disorders are

depicted with three items leads to an efficacy bias in

favour of sedating/sleep-inducing antidepressants in

antidepressant studies.

The scale was subjected to a critical test-theore-

tical analysis in order to investigate its homogeneity

and the stability of the factor structure in repeated

measurements during treatment (Bech 1981; Maier

et al. 1985; Steinmeyer and Moller 1992a).

Based on these psychometric analyses a number of

abbreviated versions of the full-length scales have

been suggested to be equivalent to their parent scales

in terms of their psychometric properties and sensi-

tivity to change (Ruhe et al. 2005; Ballesteros et al.

2007). Beside the Bech six-item version and a Maier

six-item version (Bech 1984; Maier et al. 1985), a

seven-item abbreviated version (HAMD-7) was de-

veloped for assessing depressive symptom severity

and remission (McIntyre et al. 2002, 2005). On the

basis of the analysis the Bech�Rafaelsen Melancholia

Scale was developed (BRMES) (Bech 1981, 1984),

which consists of 11 items, six of which are from the

original HAMD scale. Furthermore, unidimensional

sub-scores have been shown to outperform the parent

HAMD-17 in detecting treatment differences, lead-

ing to the suggestion that the use of these short scales

in studies would have required approximately one-

third less patients compared to the full HAMD-17

scale (Faries et al. 2000). In addition to these

abbreviated versions, a self-rating version of the full-

length scale was developed for pragmatic reasons

(Carroll et al. 1973, 1981) as well.

There is recently a strong focus on depression in

drug treatment studies to use remission as a relevant

efficacy criterion. The most widely used criterion for

remission is a HAMD-17 score of 57, which

corresponds to a HAMD-7 score of 53 (McIntyre

et al. 2002). It has variously been suggested that the

corresponding cut-off score on the MADRS scale

should be 58 of 59 (Carmody et al. 2006), B10

(Hawley et al. 2002), 510 (Zimmerman et al.

2004d), 511 (Bandelow et al. 2006). The ACNP

Task Force (Rush et al. 2006) recommended that if

the HAMD-17 scale was used, a score of 57 or 55

should be used as criteria for remission. However,

there is recent evidence to support the use of more

stringent criterion scores on both the HAMD and

MADRS scales (Moller 2008). A criticism of the use

of response as an outcome measure is that it can

identify a strongly heterogeneous population of

patients. However, defining remission using a

HAMD-17 cut-off score of 57 identifies a popula-

tion of remitters that is as heterogeneous as the

population of responders in terms of psychosocial

impairment (Zimmerman et al. 2007a,b). Patients

with HAMD-17 scores 52 (or MADRS scores 54)

show better psychological functioning than those

with scores of 3-7 (or MADRS scores 5�9) (Zim-

merman et al. 2004c, 2005). On the other hand, a

problem with the use of very stringent definitions of

remission is that healthy, non-depressed individuals

may show some degree of depressive symptoms. A

literature review of control groups in clinical trials of

depression reported a mean HAMD-17 score of

3.293.2 (SD) among healthy control individuals

(Zimmerman et al. 2004a).

MADRS

Although the HAMD is still widely accepted, the

Montgomery�Asberg Depression Rating Scale

(MADRS) (Montgomery and Asberg 1979) is

becoming increasingly important thanks to its con-

ciseness and particularly to its better definition of

characteristics. The aspect that the scale was con-

structed according to the principle ‘‘sensitivity to

change’’ is of advantage for treatment-related stu-

dies. The scale includes the following 10 items:

apparent sadness; reported sadness; inner tension;

reduced sleep; reduced appetite; concentration diffi-

culties; lassitude; inability to feel; pessimistic

thought; suicidal thoughts. The scale is supposed

to include the main symptoms of depressive illness,

even if certain important areas (e.g., psychomotor

retardation, tendency to somatise) have been

omitted as a result of the method of item selection

(Kearns et al. 1982). Overall, the factor analyses and

correlations with Hamilton scale (in particular with

the various subscale) show that the MADRS covers

more purely psychological symptoms than the

HAM-D (Montgomery and Asberg 1979; Kearns

et al. 1982).

Schmidtke et al. (1988) used a heuristic procedure

in which they subjected the correlation between

individual item scores (calculated from the raw

scores after dichotomization of items according to

various criteria such as less than the mean�0/��

mean�1; 0 and 0, 2 and 2�1; (Maier and Philipp

1985) for 57 different patient ratings by MADRS

and the HAMD conducted by the same physician to

a main axis factor analysis with varimax rotation.

The analyses all show that, despite the methodolo-

gical limitation that still remains because of the

similar calculations of the various factor analyses, the

MADRS items do not represent a unidimensional

scale. Four-factor solutions accounting for 51�54%

of the total variance in all analyses proved relatively

18 H.-J. Moller

stable. In these analyses, the aspects covered by the

MADRS items were classified under the headings

sadness/pessimistic thoughts, inner tension, inability

to feel and reduced appetite. In the construction

studies, the sensitivity to change was claimed to be

better than that of other procedures used simulta-

neously (Montgomery et al. 1978; Montgomery and

Montgomery 1980). In later studies, the sensitivity

of the MADRS for differences in the severity of

depression (Kearns et al. 1982) and change in the

symptoms of depression was again shown to be good

(Deloch 1986; Gutzmann 1986; Schmidtke et al.

1988). Reference values of several clinical samples

are available (Maier and Philipp 1985; Schmidtke et

al. 1988). The inter-rater reliability has been given

for different samples as 0.89 to 0.97 (Montgomery

et al. 1978; Montgomery and Asberg 1979).

HAMA

A number of symptoms that can be observed in

association with anxiety states have been collected for

the Hamilton Anxiety Scale (HAMA; Hamilton

1959, 1969; Lader and Marks 1974; Beneke 1987).

Thirteen symptom groups have been put together and

complemented by a fourteenth variable, namely

the patient’s observable behaviour during the inter-

view. The 14 groups of symptoms measure either

psychic or somatic effects of anxiety and are as

follows: anxious mood; tension; fears; insomnia;

intellectual; depressed mood; somatic (muscular);

somatic (sensory); cardiovascular symptoms; respira-

tory symptoms; gastrointestinal symptoms; genitour-

inary symptoms; automatic symptoms; behaviour at

interview. A factor score is calculated for both the

symptom groups assessing somatic anxiety and the

symptom groups assessing psychic anxiety (scores 1

and 2, respectively). The total raw score (score 3) can

be taken as a measure of the patient’s anxiety and

makes it possible to compare groups receiving differ-

ent treatments. Hamilton gave a very high coefficient

of correlation for the inter-rater reliability, namely

0.89 (following transformation of the mean of the

correlations between three raters).

Young Mania Rating Scale (YMRS)

The Young Mania Rating Scale (YMRS) is an

observer rating scale covering 11 items to assess

manic symptoms. The items were defined on the

basis of published descriptions of the core symptoms

of mania (manic bipolar disorder) and include the

symptoms covering the whole therapeutic index,

from mild to severe. The YMRS follows the style

of the Hamilton Depression Scale for Depression

and is made for trained clinician to be carried out in

a 15�30-minute interview. The severity rating for

each of the 11 items is based on the patients’

subjective report of his or her clinical condition

over the previous 48 hours and on clinical observa-

tions made during the course of the clinical inter-

view. The YMRS is today the most frequently used

rating scale to assess manic symptoms in clinical

studies (Young et al. 1978).

Yale Brown Obsessive-Compulsive Scale (Y-BOCS)

This rating scale (Goodman et al. 1989) is designed

to rate the severity and type of symptoms in patients

with obsessive-compulsive disorder (OCD). It is

intended for use as a semi-structured interview.

The scale consists of two parts, a symptom checklist

and the actual assessment scale. The symptom

checklist is used for the exploration of the obsessive

compulsive disorder which is the basis for the

subsequent total assessment.

The rating scale covers 10 main � or core � items

making up the total score, as well as nine additional

items providing further information but not ac-

counted for in the total score. With five or 10 main

items, obsessive thoughts and compulsive behaviour

are separately evaluated, according to: (a) time spent

on obsessions; (b) interference of obsessions; (c)

distress of obsessions; (d) resistance; and (e) control

over obsessions.

The cumulative value of all 10 main items shows

the severity of the obsessive symptomatic (0�40

points). Separate cumulative values for obsessive

thoughts (items 1�5) and compulsive behaviour

(items 6�10) can also be generated (each 0�20

points). Factor analyses suggest a 2�3 factorial

structure of the scale. McKay et al. (1995) found

two dimensions which represent obsessive thoughts

and compulsive behaviour, and therefore corre-

sponded with the specified structure of the scale.

Kim et al. (1994) extracted an additional factor,

specifically characterized by the two items to mea-

sure resistance. Y-BOCS has proven therapy sensi-

tive in different pharmacological and psychotherapy

studies (Schaible et al. 2001). As in the studies

published up to now, a cut-off value for a clinically

relevant symptomatic or as basis criteria for phar-

macological studies, a total score of 16 (out of 40) is

considered in simultaneously existing thought �and

behaviour obsessions.

Examples for important self-rating scales

SCL-90

The Self-Report Symptom Inventory 90 items

(SCL-90; Derogatis et al. 1973, 1974; Derogatis

1977b) is the revised version of the Hopkins

Important rating scales in psychiatry 19

Symptom Check List. The scale is used for the self-

rating of patients with respect to various burdening

symptoms. It allows nine ranges of syndromes to be

recorded and was especially constructed to register

of effects of drug treatment (Figure 7). It was used

respectively in various clinical studies with neuro-

leptics, tranquilizers and antidepressant (Volz et al.

2000; Moller et al. 2001). Reference values for

different clinical groups are available.

Paranoid Depression Scale (PDS)

The Paranoid Depression Scale (Arbeitsge-

meinschaft fur Methodik und Dokumentation in

der Psychiatrie 2000; CIPS (Collegium Internatio-

nale Psychiatriae Scalarum) 1990; von Zerssen

1976a,b,c,d), which is available in two parallel

forms, is composed of 43 items. It records the

degree of subjective impairment by emotional re-

duction of the type anxious-depressive mood � these

items are also on a separate depression scale � as well

as a distinct cognitive dimension to determine a

distrusting attitude and whether the subject is out of

touch with reality. Additionally there are eight

control items to measure disease denial and three

items to assess motivation. The values of the

individual items are summarized as factor values.

Amongst others, correlations of the paranoid scale

with the criterion of the affiliation to a group of

schizophrenic patients, correlations of the depres-

sion scale with the criterion of affiliation to a group

of patients with depressive mood, correlations with

relevant factors of other scales and sensitivity in the

recording of therapy-induced changes indicate the

validity. The depression scale is also available as a

separate scale, with 16 items, without the items of

the paranoid scale. Norm values are available for a

representative sample of the general population in

Germany, as well as reference values for various

clinical groups (physically ill, mixed psychiatric

groups, individual psychiatric diagnosis groups).

Beck Depression Inventory (BDI)

Another frequently used self-rating scale is the Beck

Depression Inventory (Beck et al. 1961; Beck and

Beamesderfer 1974), which was originally developed

as an observer rating scale. It has 21 items and is still

widely used. It has a special focus on cognitive

aspects of depression, which might explain the fact

that it is preferentially used in the context of

psychological treatments, while the application in

psychopharmacological studies is rare. The psycho-

metric criteria with respect to reliability and validity

are satisfactory (Steer et al. 1986). Reference values

for different clinical groups are available.

Self-Rating Anxiety Scale (SAS)

The Self-Rating Anxiety Scale (SAS) is the self-

rating version of the observer rating scale ‘‘Anxiety

Status Inventory’’, developed by the same author

(Zung 1971). The SAS contains 20 anxiety symp-

toms, five of which are affective and 15 somatic.

Some of the items are formulated positively with

respect to symptoms, other negatively. With respect

to content validity, items were selected using psy-

chiatric diagnosis criteria of different psychiatric

disorders that are associated with anxiety. And with

respect to criteria-related validity, patients with the

diagnosis of ‘‘anxiety symptoms’’ have a significantly

higher mean index (PB0.05) than patients with the

diagnosis of ‘‘schizophrenia’’, ‘‘depressive symp-

toms’’, ‘‘personality disorder’’ or ‘‘situational transi-

tional symptoms’’. The Taylor Manifest Anxiety

Scale (Taylor 1953) could not show this difference.

A group of 100 controls (57 men, 43 women, aged

18�62 years, mean 34 years) had a significantly lower

mean index than the diagnosis group listed above.

Correlations with other tests are as follows: r�0.66

with the observer rating form ASI (r�0.74 only in

patients with a diagnosis of anxiety), r�0.30 with

the Taylor Manifest Anxiety Scale. Reference values

for different clinical control samples are available.

State-Trait Anxiety Inventory (STAI)

The State-Trait Anxiety Inventory (STAI) is an

inventory for evaluating two different components

of anxiety (Johnson 1968; Spielberg 1983; Laux and

Glanzmann 1981). One scale (scale 1) is designed to

measure state anxiety and the other (scale 2) to

measure generalized (trait) anxiety. Depending on

the purpose both scales can be used together or on

their own. Scale 1 assesses a relatively stable person-

ality trait and is thus neither time nor situation

dependent. It provides an evaluation of the person’s

state at the time of testing. However, if the instruc-

tions are altered, scale 1 can also be used to assess

Figure 7. Course of median values of the SCL-90-R anxiety

subscale. Three different (n�218) treatment conditions (Moller

2003).

20 H.-J. Moller

state in any specific situation provided that the subject

can bring to mind the situation in question and is

willing to cooperate with the tester. The total scores

for scales 1 and 2 are calculated for each subject as

measures of state and trait anxiety, respectively. The

correlations with the IPAT anxiety scale, the Taylor

Manifest Anxiety Scale (TMAS) and the Zuckerman

Affect Adjective Checklist (AACL) are high (between

0.75 and 0.84). Studies on construct validity in which

the scale was administered repeatedly under different

conditions have shown point-bi-serial correlations of

r�0.60 and r�0.73. Test-retest reliability is reported

for trait as r�0.84 and state r�0.34 (men, interval

1h). Alpha coefficients of r�0.83 and 0.92 have been

found for internal consistency. Norm values are

available (Arbeitsgemeinschaft fur Methodik und

Dokumentation in der Psychiatrie 2000; CIPS (Col-

legium Internationale Psychiatriae Scalarum) 1990).

Adjective Mood Scale (AMS)

The Adjective Mood Scale contains 28 items (Ar-

beitsgemeinschaft fur Methodik und Dokumenta-

tion in der Psychiatrie 2000; CIPS (Collegium

Internationale Psychiatriae Scalarum) 1990; von

Zerssen 1976a,b,c,d) and is available in two parallel

versions. It records the degree of current impairment

of subjective well being. The scale is especially

indicated for course descriptions when tests are

frequently repeated. It is suitable for healthy subjects

and physically or psychically ill patients, particularly

for psychically ill patients with affective disorders.

The values of the individual items are summed

to give a total score, which gives the impairment of

the subjective well being. Higher inter- and intra-

individual correlations with global assessment of

the depressive mood as well as the sensitivity for

the recording of therapy-induced changes prove the

validity. There are norm values for a representative

sample of the general population of the former West

Germany and reference values for various clinical

groups.

Acknowledgements

None.

Declaration of Interest

Dr Moeller has received grants or is a consultant for

and on the speakership bureaus of AstraZeneca,

Bristol-Myers Squibb, Eisai, Eli Lilly, Glaxo-

SmithKline, Janssen Cilag, Lundbeck, Merck, No-

vartis, Organon, Pfizer, Sanofi-Aventis, Sepracor,

and Wyeth.

References

Arbeitsgemeinschaft fur Methodik und Dokumentation in der

Psychiatrie. 1995. Das AMDP-System. Manual zur Dokumen-

tation psychiatrischer Befunde. Gottingen: Hogrefe.

Arbeitsgemeinschaft fur Methodik und Dokumentation in der

Psychiatrie. 2000. Das AMDP-System. Manual zur Dokumen-

tation psychiatrischer Befunde. 7. ed. Gottingen, Bern, Tor-

onto, Seattle: Hogrefe.

Arbeitsgemeinschaft fur Methodik und Dokumentation in der

Psychiatrie, Collegium Internaationale Psychiatriae Scalarum.

1990. Rating Scales for Psychiatry. European Edition. Wein-

heim: Beltz Test GmbH.

Bailey J, Coppen A. 1976. A comparison between the Hamilton

Rating Scale and the Beck Inventory in the measurement of

depression. Br J Psychiatry 128:486�489.

Ballesteros J, Bobes J, Bulbena A, Luque A, Dal Re R, Ibarra N,

et al. 2007. Sensitivity to change, discriminative performance,

and cutoff criteria to define remission for embedded short

scales of the Hamilton depression rating scale (HAMD). J

Affect Disord 102:93�99.

Bandelow B, Baldwin DS, Dolberg OT, Andersen HF, Stein DJ.

2006. What is the threshold for symptomatic response and

remission for major depressive disorder, panic disorder, social

anxiety disorder, and generalized anxiety disorder? J Clin

Psychiatry 67:1428�1434.

Baumann U. 1976. Methodologic studies of the Hamilton rating

scale for depression. Arch Psychiatr Nervenkr 222:359�375.

Baumann U, Stieglitz RD. 1983. Testmanual zum AMDP-

System. Empirische Studien zur Psychopathologie. Berlin,

Heidelberg, New York: Springer.

Bech P. 1981. Rating scales for affective disorders: their validity

and consistency. Acta Psychiatr Scand Suppl 295:1�101.

Bech P. 1984. The instrumental use of rating scales for depres-

sion. Pharmacopsychiatry 17:22�28.

Beck AT, Beamesderfer A. 1974. Assessment of depression: the

depression inventory. In: Pichot P, editor. Psychological

measurements in psychopharmacology. Modern problems in

pharmacopsychiatry. vol 7. Basel: Karger. p 151�169.

Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. 1961. An

inventory for measuring depression. Arch Gen Psychiatry

4:561�571.

Beck AT, Rush AJ, Shaw BF, Emery G. 1986. Kognitive Therapie

der Depression, 2nd ed. Munich: Psychologie Verlags Union.

Beneke M. 1987. Methodological investigations of the Hamilton

Anxiety Scale. Pharmacopsychiatry 20: 249�255.

Bentall RP. 1992. Reconstructuring psychopathology. Psycholo-

gist 5:61�65.

Bente D, Feder J, Helmchen H. 1974. Multidimensionale

pharmakopsychiatrische Untersuchungen mit dem Neurolepti-

kum Perazin. 2. Mitteilung: Verlaufsprofile psychopatholo-

gischer und somatischer Merkmale (Untersuchungen mit

dem AMDP-System). Pharmakopsychiatr Neuropsychophar-

makol 7:8�17.

Bobon D. 1985. Contribution of the AMDP scale to quantitative

psychopathology. Acta Psychiatr Belg 85:5�249.

Bottlender R, Strauss A, Moller HJ. 2000. Prevalence and

background factors of depression in first admitted schizophre-

nic patients. Acta Psychiatr Scand 101:153�160.

Bottlender R, Sato T, Jager M, Wegener U, Wittmann J, Strauss

A, et al. 2003. The impact of the duration of untreated

psychosis prior to first psychiatric admission on the 15-year

outcome in schizophrenia. Schizophr Res 62:37�44.

Brett-Jones J, Garety P, Hemsley D. 1987. Measuring delusional

experiences: a method and its application. Br J Clin Psychol

26:257�265.

Important rating scales in psychiatry 21

Bullinger M, Kirchberger I. 1998. SF-36. Fragebogen zum

Gesundheitszustand. Gottingen: Hogrefe.

Bullinger M, Kirchberger I, Ware J. 1995. Der deutsche SF-36

Health Survey. Uberstzung und psychometrische Testung eines

krankheitsubergreifenden Instruments zur Erfassung der

gesundheitsbezogenen Lebensqualitat. Z Gesundheitswiss

3:21�36.

Busch H, Fandrich E, Freudenthal K. 1975. Zur Anwendung und

Weiterentwicklung des AMDP-Systems � Bericht uber ein

Symposium und Trainingsseminar. Pharmakopsychiatr Neu-

ropsychopharmakol 8:170�175.

Busch H, von Cranach M, Gulbinat W, Renfordt E, Tegeler J.

1980. Reliability of the AMDP-system. A preliminary report on

a multicentre exercise on the reliability of psychopathological

assessment. Acta Psychiatr Scand 62:382�392.

Cairns V, Faltermaier T, Wittchen HU, Dilling H, Mombour W,

von Zerssen D. 1982a. Some problems concerning the relia-

bility and structure of the scales in the inpatient multidimen-

sional psychiatric scale (IMPS). Arch Psychiatr Nervenkr

232:395�406.

Cairns V, von Zerssen D, Stutte KH, Mombour W. 1982b. The

stability of the symptom groupings in the Inpatient Multi-

dimensional Psychiatric Scale (IMPS). J Psychiatr Res 17:19�28.

Carmody TJ, Rush AJ, Bernstein I, Warden D, Brannan S,

Burnham D, et al. 2006. The Montgomery Asberg and the

Hamilton ratings of depression: a comparison of measures. Eur

Neuropsychopharmacol 16:601�611.

Carroll BJ, Fielding JM, Blashki TG. 1973. Depression rating

scales. A critical review. Arch Gen Psychiatry 28:361�366.

Carroll BJ, Feinberg M, Smouse PE, Rawson SG, Greden JF.

1981. The Carroll rating scale for depression. I. Development,

reliability and validation. Br J Psychiatry 138:194�200.

Cicchetti DV, Aivono SL. 1976. The Brief Psychiatric Rating

Scale (BPRS) rev: further evidence of its utility. Paper

presented at joint meeting of the American Statistical Associa-

tion. American Statistical Association, Boston.

CIPS (Collegium Internationale Psychiatriae Scalarum). 1990.

Rating scales for psychiatry. Weinheim: Beltz.

Costa PT, McCrae RR. 1992. Manual for the revised NEO

Personality Inventory (NEO-PI-R) and NEO Five-Factor

Inventory (NEO-FFI). Odessa: Psychological Assessment Re-

sources.

Cuesta MJ, Peralta V. 2001. Integrating psychopathological

dimensions in functional psychoses: a hierarchical approach.

Schizophr Res 52:215�229.

Deloch E. 1986. Vortrag auf dem Idom-Expertengesprach in

Estoril. G Selecta Bericht 42:3068�3070.

Derogatis LR, Lipman RS, Covi L. 1973. SCL-90: an outpatient

psychiatric rating scale�preliminary report. Psychopharmacol

Bull 9:13�28.

Derogatis LR, Lipman RS, Rickels K, Uhlenhuth EH, Covi L.

1974. The Hopkins Symptom Checklist (HSCL): a self-report

symptom inventory. Behav Sci 19:1�15.

Derogatis LR. 1977a. SCL. Administration, scoring and proce-

dures. Manual for the revised version and other instruments of

the psychopathology rating scale series. Baltimore, MA: John

Hopkins University School of Medicine.

Derogatis LR. 1977b. SCL-90. Administration, scoring and

procedures. Manual-I for the R(evised) version and other

instruments of the psychopathology rating scale series. Balti-

more, MA: Johns Hopkins University School of Medicine.

Egli S, Riedel M, Moller HJ, Strauss A, Lage D. 2008. Creating a

map of Psychiatric Patients based on Psychopathological

Symptom Profiles. in preperation

Emsley R, Rabinowitz J, Torreman M. 2003. The factor structure

for the Positive and Negative Syndrome Scale (PANSS) in

recent-onset psychosis. Schizophr Res 61:47�57.

Endicott J, Spitzer RL, Fleiss JL, Cohen J. 1976. The global

assessment scale. A procedure for measuring overall severity of

psychiatric disturbance. Arch Gen Psychiatry 33:766�771.

Faltermaier T, Hecht H, Wittchen HU. 1987. Die Social Inter-

view Schedule (deutschsprachige modifizierte Version). Re-

gensburg: Roderer.

Faries D, Herrera J, Rayamajhi J, DeBrota D, Demitrack M,

Potter WZ. 2000. The responsiveness of the Hamilton Depres-

sion Rating Scale. J Psychiatr Res 34:3�10.

Fischer G. 1974. Einfuhrung in die Theorie psychologischer

Tests. Bern: Huber.

Fossati A, Beauchaine TP, Grazioli F, Borroni S, Carretta I, De

Vecchi C, et al. 2006. Confirmatory factor analyses of DSM-IV

Cluster C personality disorder criteria. J Personal Disord

20:186�203.

Freudenthal K, Gebhardt R, Pietzcker A. 1977. A comparison of

two asssessment methods. Pharmakopsychiatr Neuropsycho-

pharmakol 10:64.

Garety PA, Hemsley DR. 1987. Characteristics of delusional

experience. Eur Arch Psychiatry Neurol Sci 236:294�298.

Gebhardt R, Helmchen H. 1973. Zur Zuverlassigkeit psycho-

pathologischer Symptomerfassung. Schweiz Arch Neurol Neu-

rochir Psychiatr 112:459�469.

Gebhardt R, Pietzcker A, Freudenthal K, Langer C. 1981. Die

Bildung von Syndromen im AMDP-System. Arch Psychiatr

Nervenkr 231:93�109.

Goldberg DP. 1972. The detection of psychiatric illness by

questionaire. London: Oxford University Press.

Goldman HH, Skodol AE, Lave TR. 1992. Revising axis V for

DSM-IV: a review of measures of social functioning. Am J

Psychiatry 149:1148�1156.

Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleisch-

mann RL, Hill CL, et al. 1989. The Yale-Brown Obsessive

Compulsive Scale. I. Development, use and reliability. Arch

Gen Psychiatry 46:1006�1011.

Gutzmann H. 1986. Vortrag auf dem Idom-Expertengesprach in

Estoril. G Selecta Bericht 42:3068�3070.

Guy W, Ban TA. 1979. The AMDP System. Berlin, Heidelberg,

New York: Springer.

Guy W, Ban TA. 1982. The AMDP System. Manual for the

assessment and documentation of psychopathology. Berlin:

Springer.

Hamilton M. 1959. The assessment of anxiety states by rating. Br

J Med Psychol 32:50�55.

Hamilton M. 1960. A rating scale for depression. J Neurol

Neurosurg Psychiatry 23:56�62.

Hamilton M. 1967. Development of a rating scale for primary

depressive illness. Br J Soc Clin Psychol 6:278�296.

Hamilton M. 1969. Diagnosis and rating of anxiety. In: Lader

MH, editor. Studies of anxiety. Ashford, Kent: Headly Brothers

Ltd. p 76�79.

Hamilton M. 1976. HAMD. Hamilton Depression Scale. In: Guy

W, editor. ECDEU assessment manual for psychopharmacol-

ogy. Rockville, MD: National Institute of Mental Health. p

193�198.

Hasemann K. 1971. Verhaltensbeobachtung. In: Heiss R, editor.

Handbuch der Psychologie. 3rd ed. vol 6. Gottingen: Hogrefe.

p 807�836.

Hathaway SR, McKinley JC. 1963. MMPI Saarbrucken. Hand-

buch zur deutschen Ausgabe des MMPI. Bern: Huber.

Hathaway SR, McKinley JC. 1989. MMPI-2. Minnesota Multi-

phasic Personality Inventory 2. Minneapolis, MN: University

of Minnesota.

22 H.-J. Moller

Hawley CJ, Gale TM, Sivakumaran T. 2002. Defining remission

by cut off score on the MADRS: selecting the optimal value. J

Affect Disord 72:177�184.

Hiller W, von Zerssen D, Mombour W, Wittchen HU. 1986. Die

IMPS. Weinheim: Beltz.

Honigfeld G, Klett CJ. 1965. The Nurses’ Observation Scale for

inpatient evaluation: A new scale for measuring improvement

in chronic schizophrenia. J Clin Psychol 21:65�71.

Huber G. 1976. Zur Problematik quantitativer Verlaufsbeobach-

tungen bei Schizophrenen. Psychopathometrie 2:61�66.

Jablensky A. 2002. The classification of personality disorders:

critical review and need for rethinking. Psychopathology

35:112�116.

Jablensky A, Schwarz R, Tomow T. 1980. WHO collaborative

study on impairments and disabilities associated with schizo-

phrenic disorders. A preliminary communication: objectives

and methods. Acta Psychiatr Scand 62:152�159.

Johnson DT. 1968. Effects of interview stress on measures of state

and trait anxiety. J Abnorm Psychol 73:245�251.

Katz MM, Lyerly SB. 1963. Methods of measuring adjustment

and social behaviour in the community. Psychol Rep 13:503�535.

Kay SR, Fiszbein A, Opler LA. 1988. Positive and negative

syndrome scale (PANSS) for schizophrenia. Schizophr Bull

13:21�76.

Kearns NP, Cruickshank CA, McGuigan KJ, Riley SA, Shaw SP,

Snaith RP. 1982. A comparison of depression rating scales. Br J

Psychiatry 141:45�49.

Kim SW, Dysken MW, Pheley AM, Hoover KM. 1994. The Yale-

Brown Obsessive-Compulsive Scale: measures of internal

consistency. Psychiatry Res 51:203�211.

Kobak KA, Greist JH, Jefferson JW, Katzelnick DJ. 1996.

Computer-administered clinical rating scales. A review. Psy-

chopharmacology (Berlin) 127:291�301.

Kobak KA, Greist JH, Jefferson JW, Mundt JC, Katzelnick DJ.

1999. Computerized assessment of depression and anxiety over

the telephone using interactive voice response. MD Comput

16:64�68.

Kuny S, Lucker MV, Baur P, Eichberger G, Woggon B. 1983.

Interrater reliability of AMDP and AMDP � symptoms. In:

Bobon D, Baumann U, Angst JW, Helmchen H, Hippius H,

editors. AMDP-system in pharmacopsychiatry. Basel: Karger.

p 143�160.

Lader MH, Marks IM. 1974. The rating of clinical anxiety. Acta

Psychiatr Scand 50:112�137.

Lapierre YD, Angus C, Awad AG, Saxena BM, Jones B,

Williamson P, et al. 1999. The treatment of negative symptoms:

a clinical and methodological study. Int Clin Psychopharmacol

14:101�112.

Laux L, Glanzmann P. 1981. Das State-Trait-Angstinventar

(STAI). Weinheim: Beltz Test GmbH.

Lenzenweger MF. 1999. Schizophrenia: refining the phenotype,

resolving endophenotypes. Behav Res Ther 37:281�295.

Lienert GA. 1969. Testaufbau und Testanalyse. Weinheim: Beltz.

Lindenmayer JP, Bernstein-Hyman R, Grochowski S. 1994. A

new five factor model of schizophrenia. Psychiatr Q 65:299�322.

Lingjaerde O, Ahlfors UG, Bech P, Dencker SJ, Elgen K. 1987.

The UKU side effect rating scale. A new comprehensive rating

scale for psychotropic drugs and a cross-sectional study of side

effects in neuroleptic-treated patients. Acta Psychiatr Scand

Suppl 334:1�100.

Loranger AW, Sartorius N, Andreoli A, Berger P, Buchheim P,

Channabasavanna SM, et al. 1994. The International Person-

ality Disorder Examination. The World Health Organization/

Alcohol, Drug Abuse, and Mental Health Administration

international pilot study of personality disorders. Arch Gen

Psychiatry 51:215�224.

Lorr M. 1974. Assessing psychotic behaviour by the IMPS. In:

Pichot P, Olivier-Martin R, editors. Psychological measure-

ments in psychopharmacology. Basel: Karger. p 50�63.

Lorr M, McNair DM, Klett CJ, Lasky JJ. 1962. Evidence of ten

psychotic syndromes. J Consult Psychol 26:185�189.

Luria RE. 1975. The validity and reliability of the visual analogue

mood scale. J Psychiatr Res 12:51�57.

Maffei C, Fossati A, Agostoni I, Barraco A, Bagnato M, Deborah

D, 1997. Interrater reliability and internal consistency of the

structured clinical interview for DSM-IV axis II personality

disorders (SCID-II), version 2.0. J Personal Disord 11:279�284.

Maier W, Philipp M. 1985. Comparative analysis of observer

depression scales. Acta Psychiatr Scand 72:239�245.

Maier W, Philipp M, Gerken A. 1985. Dimensions of the

Hamilton Depression Scale. Factor analysis studies. Eur Arch

Psychiatry Neurol Sci 234:417�422.

Marder SR, Davis JM, Chouinard G. 1997. The effects of

risperidone on the five dimensions of schizophrenia derived

by factor analysis: combined results of the North American

trials. J Clin Psychiatry 58:538�546.

Mass R, Schoemig T, Hitschfeld K, Wall E, Haasen C. 2000.

Psychopathological syndromes of schizophrenia: evaluation of

the dimensional structure of the positive and negative syn-

drome scale. Schizophr Bull 26:167�177.

McEvoy JP. 1994. Efficacy of risperidone on positive features of

schizophrenia. J Clin Psychiatry 55(Suppl):18�21.

McIntyre R, Kennedy S, Bagby RM, Bakish D. 2002. Assessing

full remission. J Psychiatry Neurosci 27:235�239.

McIntyre RS, Konarski JZ, Mancini DA, Fulton KA, Parikh SV,

Grigoriadis S, et al. 2005. Measuring the severity of depression

and remission in primary care: validation of the HAMD-7

scale. Can Med Assoc J 173:1327�1334.

McKay D, Danyko S, Neziroglu F, Yaryura-Tobias JA. 1995.

Factor structure of the Yale-Brown Obsessive-Compulsive

Scale: a two dimensional measure. Behav Res Ther 33:865�869.

Moller HJ. 1991. Outcome criteria in antidepressant drug trials:

self-rating versus observer-rating scales. Pharmacopsychiatry

24:71�75.

Moller HJ. 2000. Rating depressed patients: observer- vs self-

assessment. Eur Psychiatry 15:160�172.

Moller HJ. 2003. Scales used in depression and anxiety research.

In: Kasper S, den Boer JA, Sitsen JM, editors. Handbook of

depression and anxiety. New York, Basel: Marcel Decker.

p 789�808.

Moller HJ. 2008. Outcomes in major depressive disorder: the

evolving concept of remission and its implications for treat-

ment. World J Biol Psychiatry 9(2):102�14.

Moller HJ, Hacker H. 1988. Study concerning the sample

dependency and temporal variance of the factor structure in

the Inpatient Multidimensional Psychiatric Scale. Psycho-

pathology 21:281�290.

Moller HJ, von Zerssen D. 1987. Pramorbide Personlichkeit von

Patienten mit affektiven Psychosen. In: Kisker KP, Lauter H,

Meyer JE, Muller C, Stromgren E, editors. Affektive Psycho-

sen. Psychiatrie der Gegenwart. 5. 3 ed. Berlin, Heidelberg,

New York: Springer. p 165�179.

Moller HJ, von Zerssen D. 1995. Self-rating procedures in the

evaluation of antidepressants. Psychopathology 28:291�306.

Moller HJ, Leitner M, Dietzfelbinger T. 1987. A linear mathe-

matical model for computerized analyses of mood curves. An

empirical investigation on mood courses in depressive and

schizophrenic inpatients. Eur Arch Psychiatry Neurol Sci

236:260�268.

Important rating scales in psychiatry 23

Moller HJ, Schmid Bode W, Cording-Tommel C, Wittchen HU,

Zaudig M, von Zerssen D. 1988. Psychopathological and social

outcome in schizophrenia versus affective/schizoaffective psy-

choses and prediction of poor outcome in schizophrenia.

Results from a 5�8 year follow-up. Acta Psychiatr Scand

77:379�389.

Moller HJ, Blank R, Steinmeyer EM. 1989. Single-case evaluation

of sleep-deprivation effects by means of nonparametric time-

series analysis (according to the HTAKA model). Eur Arch

Psychiatry Neurol Sci 239:133�139.

Moller HJ, Engel RR, Hoff P. 1996. Befunderhebung in der

Psychiatrie: Lebensqualitat, Negativsymptomatik und andere

aktuelle Entwicklungen. Wien, New York: Springer.

Moller HJ, Bauml J, Ferrero F, Fuger J, Geretsegger C, Kasper S,

Kissling W, Schubert H. 1997. Risperidone in the treatment of

schizophrenia: results of a study of patients from Germany,

Austria, and Switzerland. Eur Arch Psychiatry Clin Neurosci

247:291�296.

Moller HJ, Volz HP, Reimann IW, Stoll KD. 2001. Opipramol for

the treatment of generalized anxiety disorder: a placebo-

controlled trial including an alprazolam-treated group. J Clin

Psychopharmacol 21:59�65.

Moller HJ, Bottlender R, Groß A, Hoff P, Wittmann J, Wegner U,

et al. 2002. The Kraepelinian dichotomy: preliminary results of

a 15-year follow-up study on functional psychoses: focus on

negative symptoms. Schizophr Res 56:87�94.

Mombour W. 1974a. Frequency of symptoms in psychiatric

illnesses. A comparative investigation with two rating scales

(IMPS and AMP-scale) � psychological-pathological findings

(author’s transl). Arch Psychiatr Nervenkr 219:133�152.

Mombour W. 1974b. Syndromes in psychiatric illnesses. A

comparative investigation with two rating scales (IMPS and

AMP-scale) (author’s transl). Arch Psychiatr Nervenkr

219:331�350.

Mombour W. 1976. Systematik psychischer Storungen. In:

Pongratz LJ, editor. Handbuch der Psychologie. vol 8/1 ed.

Gottingen: Hogrefe. p 116�153.

Mombour W, Gammel G, von Zerssen D, Heyse H. 1973.

Objectivation of psychiatric syndromes through the multi-

factorial analysis of the psychopathological status. Nervenarzt

44:352�358.

Mombour W, Kockett G, Fliege K. 1975. Die Brief Psychiatric

Rating Scale (BPRS) von Overall und Gorham bei akuten

paranoid-halluzinatorischen Psychosen. Untersuchungen zu

einer deutschen Uberstzung der BPRS. Pharmakopsychiatr

Neuropsychopharmakol 8:279�288.

Montgomery S, Asberg M, Jornestedt L, Thoren P, Traskman L,

McAuley R, et al. 1978. Reliability of the CPRS between the

disciplines of psychiatry, general practice, nursing and psychol-

ogy in depressed patients. Acta Psychiatr Scand 271 (Suppl):

29�32.

Montgomery SA. 2006. Why do we need new and better

antidepressants? Int Clin Psychopharmacol 21(Suppl 1):S1�10.

Montgomery SA, Asberg M. 1979. A new depression scale

designed to be sensitive to change. Br J Psychiatry 134:382�389.

Montgomery SA, Montgomery DB. 1980. Measurement of

change in psychiatric illness: new obsessional, schizophrenia

and depression scales. Postgrad Med J 56(Suppl 1):50�52.

Morley SJ. 1994. Single case methodology in psychopathology

therapy. In: Lindsay SJ, Powell GE, editors. Handbook of

clinical adult psychology. London: Routledge. p 723�745.

Morosini PL, Magliano L, Brambilla L, Ugolini S, Pioli R. 2000.

Development, reliability and acceptability of a new version of

the DSM-IV Social and Occupational Functioning Assessment

Scale (SOFAS) to assess routine social functioning. Acta

Psychiatr Scand 101:323�329.

Muller MJ, Wetzel H. 1998. Improvement of inter-rater reliability

of PANSS items and subscales by a standardized rater training.

Acta Psychiatr Scand 98:135�9.

Mundt JC. 1997. Interactive voice response systems in clinical

research and treatment. Psychiatr Serv 48:611�612.

National Institute of Mental Health. 1976. Clinical global

impressions. In: Guy W, editor. ECDEU assessment manual

for psychopharmacology. Rockville, MD: National Institute of

MEntal Health. p 217�222.

Overall JE, Gorham DR. 1962. The Brief Psychiatric Rating

Scale. Psychol Rep 10: 799�812.

Overall JE, Gorham DR. 1972. Applied multivariate analyses.

New York: McGraw-Hill.

Overall JE, Gorham DR. 1976. BPRS. Brief Psychiatric Rating

Scale. In: Guy W, editor. ECDEU assessment manual for

psychopharmacology. Rockville, National Institute of Mental

Health. p 157�169.

Pfohl B, Blum N, Zimmerman M, Stangl D. 1989. Structured

interview for DSM-III-R personality (SIDP-R). Iowa City, IA:

Department of Psychiatry, University of Iowa.

Pietzcker A, Gebhardt R, Freudenthal K, Langer C. 1981.

Diagnostische Differenzierungsmoglichkeiten von psycho-

pathologischen Symptomen bei schwierigen Differentialdiag-

nosen. Arch Psychiatr Nervenkr 230:141�157.

Prusoff BA, Klerman GL, Paykel ES. 1972a. Concordance

between clinical assessments and patients’ self-report in depres-

sion. Arch Gen Psychiatry 26:546�552.

Prusoff BA, Klerman GL, Paykel ES. 1972b. Pitfalls in the self-

report assessment of depression. Can Psychiatr Assoc J 17(2):

Suppl 2:SS101.

Pukrop R, Moller HJ, Steinmeyer EM. 2000. Quality of life in

psychiatry: a systematic contribution to construct validation

and the development of the integrative assessment tool

‘‘modular system for quality of life’’. Eur Arch Psychiatry

Clin Neurosci 250(3):120�132.

Pukrop R, Schlaak V, Moller-Leimkuhler AM, Albus M, Czernik

A, Klosterkotter J, et al. 2003. Reliability and validity of Quality

of Life assessed by the Short-Form 36 and the Modular System

for Quality of Life in patients with schizophrenia and patients

with depression. Psychiatry Res 119:63�79.

Renfordt E, Busch H, Cranach M, Gulbinat W, Tegeler J. 1983.

Special aspects of interrater reliability in the AMDP psycho-

pathological version. In: Bobon D, Baumann U, Angus JW,

Helmchen H, Hippius H, editors. AMDP-System in pharma-

copsychiatry. Basel: Karger. p 125�142.

Ruhe HG, Dekker JJ, Peen J, Holman R, de Jonghe F. 2005.

Clinical use of the Hamilton Depression Rating Scale: is

increased efficiency possible? A post hoc comparison of

Hamilton Depression Rating Scale, Maier and Bech subscales,

Clinical Global Impression, and Symptom Checklist-90 scores.

Compr Psychiatry 46:417�427.

Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B,

Klein DN, et al. 2003. The 16-Item Quick Inventory of

Depressive Symptomatology (QIDS), clinician rating (QIDS-

C), and self-report (QIDS-SR): a psychometric evaluation in

patients with chronic major depression. Biol Psychiatry

54:573�583.

Rush AJ, Kraemer HC, Sackeim HA, Fava M, Trivedi MH, Frank

E, et al. 2006. Report by the ACNP Task Force on response

and remission in major depressive disorder. Neuropsychophar-

macology 31:1841�1853.

Rush JA. 2000. Handbook of Psychiatric Measures. Arlington:

American Psychiatric Association.

Sachs GS, Guille C, McMurrich SL. 2002. A clinical monitoring

form for mood disorders. Bipolar Disord 4:323�327.

24 H.-J. Moller

Sachs GS, Nierenberg AA, Calabrese JR, Marangell LB, Wis-

niewski SR, Gyulai L, et al. 2007. Effectiveness of adjunctive

antidepressant treatment for bipolar depression. New Engl J

Med 356:1711�1722.

Sanger TM, Lieberman JA, Tohen M, Grundy S, Beasley C Jr,

Tollefson GD. 1999. Olanzapine versus haloperidol treatment

in first-episode psychosis. Am J Psychiatry 156:79�87.

Sarris V, Lienert GA. 1974. Konstruktion und Bewahrung von

klinisch-psychologischen Testverfahren. In: Rey ER, editor.

Klinische Psychologie. II. Methoden, Ergebnisse und probleme

der Forschung. Bern: Huber. p 286�351.

Sarris V, Rey ER. 1981. Allgemeine Grundlagen von klinisch-

psychologischen Testfaktoren. In: Rey ER, editor. Klinische

Psychologie. Stuttgart: Fischer. p 11�28.

Sass H. 1986. [Classification of personality disorders]. Nervenarzt

57:193�203.

Sayer NA, Sackheim HA, Moeller JR, Prudic J, Devanand DP,

Coleman EA, et al. 1993. The relations between observer-

rating and self-report of depressive symptomatology. Psychol

Assess 5:350�360.

Schaible R, Armbrust M, Nutzinger DO. 2001. Yale-Brown

Obsessive Compulsive Scale: Sind Selbst- und Fremdrating

aquivalent? Verhaltenstherapie 11:298�303.

Schmidtke A, Fleckenstein P, Moises W, Beckmann H. 1988.

Untersuchungen zur Reliabilitat und Validitat einer deutschen

Version der Montgomery-Asberg Depression-Rating Scale

(MADRS). Schweiz Arch Neurol Psychiatr 139:51�65.

Shapiro MB. 1966. The single case in clinical-psychological

research. J Gen Psychol 74:3�23.

Simpson GM, Angus JW. 1970. A rating scale for extrapyramidal

side effects. Acta Psychiatr Scand Suppl 212:11�9.

Spielberg CD. 1983. Manual for the State-Trait Anxiety Inven-

tory (Form X-I). Palo Alto, CA: Consulting Psychologists

Press.

Spitzer RL, Endicott J, Robins E. 1975. Clinical criteria for

psychiatric diagnosis and DSM-III. Am J Psychiatry 132:1187�1192.

Stangl D, Pfohl B, Zimmerman M, Bowers W, Corenthal C. 1985.

A structured interview for the DSM-III personality disorders. A

preliminary report. Arch Gen Psychiatry 42:591�596.

Steer RA, BECK AT, Garrison B. 1986. Applications of the Beck

Depression Inventory. In: Sartorius N, Ban TA, editors.

Assessment of depression. New Xork: Springer. p 123�142.

Steinmeyer EM, Moller HJ. 1992a. Facet theoretic analysis of the

Hamilton-D scale. J Affect Disord 25:53�61.

Steinmeyer EM, Moller HJ. 1992b. Facet theoretic analysis of the

Hamilton-D scale. J Affect Disord 25: 53�61.

Stieglitz RD, Baumann U. 1994. Psychodiagnostik psychischer

Storungen. Stuttgart: Enke.

Sulz-Blume B, Sulz KD, von Cranach M. 1979. [Stability of the

factor structure in the AMDP-Scale (author’s transl)]. Arch

Psychiatr Nervenkr 227:353�366.

Taylor JA. 1953. A personality scale of manifest anxiety. J Abnorm

Psychol 48:285�290.

Troisfontaines B, Bobon D. 1987. [Scales, factor analysis and

subscales of the French-language AMDP system]. Acta Psy-

chiatr Belg 87:23�60.

Trull TJ, Tragesser SL, Solhan M, Schwartz-Mette R. 2007.

Dimensional models of personality disorder: Diagnostic and

Statistical Manual of Mental Disorders Fifth Edition and

beyond. Curr Opin Psychiatry 20:52�56.

Uher R, Farmer A, Maier W, Rietschel M, Hauser J, Marusic A, et

al. 2008. Measuring depression: comparison and integration of

three scales in the GENDEP study. Psychol Med 38:289�300.

Volz HP, Moller HJ, Reimann I, Stoll KD. 2000. Opipramol for

the treatment of somatoform disorders results from a placebo-

controlled trial. Eur Neuropsychopharmacol 10:211�217.

von Knorring L, Lindstrom E. 1995. Principal components and

further possibilities with the PANSS. Acta Psychiatr Scand

Suppl 388:5�10.

von Zerssen D. 1976a. Kllinische Selbstbeurteilungs-Skalen

(KSbS) aus dem Munchener Psychiatrischen Informationssys-

tem (PSYCHIS Munchen). Algemeiner Teil. Weinheim: Beltz.

von Zerssen D. 1976b. Kllinische Selbstbeurteilungs-Skalen

(KSbS) aus dem Munchener Psychiatrischen Informationssys-

tem (PSYCHIS Munchen). Befindlichkeits-Skala. Weinheim:

Beltz.

von Zerssen D. 1976c. Kllinische Selbstbeurteilungs-Skalen

(KSbS) aus dem Munchener Psychiatrischen Informationssys-

tem (PSYCHIS Munchen). Beschwerden-Liste. Weinheim:

Beltz.

von Zerssen D. 1976d. Kllinische Selbstbeurteilungs-Skalen

(KSbS) aus dem Munchener Psychiatrischen Informationssys-

tem (PSYCHIS Munchen). Paranoid-Depressivitats-Skalen.

Weinheim: Beltz.

von Zerssen D. 1977. Konstitutionstypologische Forschung. In:

Strube G, editor. Die Psychologie des 20. Jahrhunderts, vol 5.

Zurich: Kindler. p 545�616.

von Zerssen D. 1979. Klinisch-psychiatrische Selbstbeurteilungs-

Fragebogen. In: Baumann U, Berbalk H, Seidenstucker G,

editors. Klinische Psychologie. Trends in Forschung und

Praxis, vol 2. Bern: Huber. p 130�159.

von Zerssen D. 1982. Personality and affective disorders. In:

Paykel ES, editor. Handbook of affective disorders. Edinburgh:

Churchill Livingstone. p 212�228.

von Zerssen D. 1986. Clinical Self-Rating Scales (CSRS) of the

Munich Psychiatric Information System (PSYCHIS

Munchen). In: Sartorius N, Ban TA, editors. Assessment of

depression. Berlin, Heidelberg, New York: Springer. p 279�303.

von Zerssen D. 1993. Normal and abnormal variants of pre-

morbid personality in functional mental disorders. Conceptual

and methodological issues. J Pers Disord 7:116�136.

von Zerssen D. 1994. Diagnostik der pramorbiden Personlichkeit.

In: Stieglitz RD, Baumann U, editors. Psychodiagnostik

psychischer Storungen. Stuttgart: Enke. p 216�229.

von Zerssen D, Cording C. 1978. The measurement of change in

endogenous affective disorders. Arch Psychiatr Nervenkr

226:95�112.

von Zerssen D, Moller HJ. 1980. Psychopathometrische Verfah-

ren in der psychiatrischen Therapieforschung. In: Biefang S,

editor. Evaluationsforschung in der Psychiatrie: Fragestellun-

gen und Methoden. Stuttgart: Enke. p 129�166.

Waldron J, Bates TJ. 1965. The management of depression in

hospital. A comparative trial of desipramine and imipramine.

Br J Psychiatry 111:511�6.

Weissman MM, Bothwell S. 1976. Assessment of social adjust-

ment by patient self-report. Arch Gen Psychiatry 33:1111�1115.

Weissman MM, Sholomskas D, John K. 1981. The assessment of

social adjustment. An update. Arch Gen Psychiatry 38:1250�1258.

Welner J. 1972. Eine internationale multizentrische Doppelblind-

Prufung eines neuen Antidepressivums. In: Kielholz P, editor.

Depressive Zustande. Vienna: Huber. p 209�219.

White J, White K, Razani J. 1984. Effects of endogenicity and

severity on consistency of standard depression rating scales. J

Clin Psychiatry 45:260�261.

White L, Harvey PD, Opler L, Lindenmayer JP. 1997. Empirical

assessment of the factorial structure of clinical symptoms in

schizophrenia. A multisite, multimodel evaluation of the

factorial structure of the Positive and Negative Syndrome

Scale. The PANSS Study Group. Psychopathology 30:263�274.

Important rating scales in psychiatry 25

WHO. 1991. Schedule for the clinical assessment in neuropsy-

chiatry. Geneva: WHO.

WHO. 1999. Schedule for the clinical assessment in neuropsy-

chiatry, Version 2.1. Genf: WHO.

Wing JK, Cooper JE, Sartorius N. 1974. Measurement and

classification of psychiatric symptoms. Cambridge: Cambridge

University Press.

Wittchen HU, Semmler G. 1991. Composite International

Diagnostic Interview (CIDI, version 1.0). Weinheim: Beltz.

Wittchen HU, Semmler G. 1997. Composite International

Diagnostic Interview (CIDI). Weinheim: Beltz.

Wittchen HU, Zaudig M, Spengler P. 1991. Wie zuverlassig ist

operationalisierte Diagnostik? � Die Test-Retest Reliabilitat des

Strukturierten Interviews fur DSM-III-R. Z Klin Psychol

20:136�153.

Wittchen HU, Wunderlich U, Gruschwitz S, Zaudig M. 1997.

Strukturiertes Klinisches Interview fur DSM-IV (SKID).

Gottingen: Hogrefe.

Woggon B, Baumann U, Angst J. 1978. Interrater-Reliabilitat von

AMDP-Symptomen. Arch Psychiatr Nervenkr 225:73�85.

Woggon B, Fleischhauer J, Widmer A. 1979. Der Einfluss von

Diagnose, Klinik und Geschlecht auf die Wirkung von Brom-

peridol. Int Pharmacopsychiatry 14:213�227.

Wolthaus JE, Dingemans PM, Schene AH, Linszen DH, Kneg-

tering H, Holthausen EA, et al. 2000. Component structure of

the positive and negative syndrome scale (PANSS) in patients

with recent-onset schizophrenia and spectrum disorders. Psy-

chopharmacology (Berlin) 150:399�403.

Young RC, Biggs JT, Ziegler VE, Meyer DA. 1978. A rating scale

for mania: reliability, validity and sensitivity. Br J Psychiatry

133:429�435.

Zimmerman M, Chelminski I, Posternak M. 2004a. A review of

studies of the Hamilton depression rating scale in healthy

controls: implications for the definition of remission in treat-

ment studies of depression. J Nerv Ment Dis 192:595�601.

Zimmerman M, Chelminski I, Posternak M. 2004b. A review of

studies of the Hamilton depression rating scale in healthy

controls: implications for the definition of remission in treat-

ment studies of depression. J Nerv Ment Dis 192:595�601.

Zimmerman M, Posternak MA, Chelminski I. 2004c. Defining

remission on the Montgomery-Asberg depression rating scale. J

Clin Psychiatry 65:163�168.

Zimmerman M, Posternak MA, Chelminski I. 2004d. Derivation

of a definition of remission on the Montgomery-Asberg

depression rating scale corresponding to the definition of

remission on the Hamilton rating scale for depression. J

Psychiatr Res 38:577�582.

Zimmerman M, Posternak MA, Chelminski I. 2005. Is the cutoff

to define remission on the Hamilton Rating Scale for Depres-

sion too high? J Nerv Ment Dis 193:170�175.

Zimmerman M, Posternak MA, Chelminski I. 2007a. Hetero-

geneity among depressed outpatients considered to be in

remission. Compr Psychiatry 48:113�117.

Zimmerman M, Posternak MA, Ruggero CJ. 2007b. Impact of

study design on the results of continuation studies of anti-

depressants. J Clin Psychopharmacol 27:177�181.

Zung WW. 1965. A self-rating depression scale. Arch Gen

Psychiatry 12:63�70.

Zung WW. 1971. A rating instrument for anxiety disorders.

Psychosomatics 12:371�379.

26 H.-J. Moller

ORIGINAL INVESTIGATION

Isoprostenes as indicators of oxidative stress in schizophrenia

ANNA DIETRICH-MUSZALSKA1 & BEATA OLAS2

1II Department of Psychiatry, Medical University of Lodz, Poland, and 2Department of General Biochemistry, University

of L odz, Lodz, Poland

AbstractObjective. Free radicals induce oxidative stress and damage to all types of biological molecules and may be involved inpathology of schizophrenia. A cell membrane dysfunction caused by lipid peroxidation can be secondary to a free radical-mediated pathology and may contribute to specific aspects of schizophrenic symptomatology and complications of itstreatment. Method. The aim of our study was to estimate oxidative stress in a group of schizophrenic patients by usingdifferent biomarkers of free radicals-induced lipid peroxidation (isoprostanes, thiobarbituric acid reactive substances(TBARS)). We also determined the products of enzymatic peroxidation of arachidonic acid, such as thromboxane B2

(TXB2) and its metabolite 11-dehydrothromboxane B2. Isoprostanes (IPs) are a family of novel prostaglandin isomers andare produced in free radical-catalysed reactions from arachidonic acid. They are useful as a specific, sensitive, chemicallystable, noninvasive index of free radical generation in vivo. We therefore assessed in schizophrenic patients and controlsubjects the level of urinary excretion of isoprostane � 8-epi-prostaglandin F2a (8-isoPGF2a) � a marker of lipid peroxidationinduced by free radicals using an immunoassay kit. We also studied the level of the other marker of enzymatic arachidonicacid peroxidation � 11-dehydrothromboxane B2 � in urine from schizophrenic patients and healthy volunteers. Moreover,we estimated the production of TBARS and TXB2 in plasma from schizophrenic patients and the control group. Patientshospitalised in the II Psychiatric Department of Medical University in Lodz, Poland, were interviewed with a specialquestionnaire (treatment, course of diseases, dyskinesis and other EPS). According to DSM-IV criteria, all patients haddiagnosis of paranoid type. All patients were treated with second-generation antipsychotic drugs (risperidone, clozapine,and olanzapine). Mean time of schizophrenia duration was about 2 years. Results. We observed a statistically increased levelof TBARS in plasma (P�0.000162) and isoprostanes (P�3.5�10�12) in urine of schizophrenic patients in comparisonwith the control group. The level of markers of enzymatic oxidation of arachidonic acid (TXB2 and its metabolite,11-dehydrothromboxane B2) did not change. This indicates that free radicals induce peroxidation of unsaturated fatty acidin schizophrenic patients. Conclusion. Considering the data presented in this study, we suggest that non-invasivemeasurement of 8-isoPGF2a is a valuable and sensitive (contrary to TBARS) indicator of oxidative stress status in vivoin schizophrenia.

Key words: Schizophrenic disorders, oxidative stress, isoprostanes, thromboxane B2, TBARS

Introduction

Schizophrenia is a severe psychiatric disorder,

with as yet unknown aetiopathogenesis, that affects

about 1% of the world population. It is characterized

by fundamental distortions in thinking and percep-

tion, and by inappropriate emotions. Schizophrenia

follows a variable course, with complete sympto-

matic and social recovery in about one-third of

cases. Schizophrenia can, however, follow a chronic

or recurrent course, with residual symptoms and

incomplete social recovery (WHO 1979, 2001). The

aetiology of schizophrenia remains elusive, although

several hypotheses have been suggested. Recent

findings suggest the role of abnormalities of mem-

brane phospholipids, and fatty acids caused by

oxidative stress, in the aetiopathogenic mechanisms

of schizophrenia (Fenton et al. 2000; Ross 2003;

Fendri et al. 2006). The role of membrane altera-

tions, caused by oxidative stress, was taken into

account by Reddy and Yao (1999) in the pathophy-

siology of schizophrenia. The neuronal membrane

contains a high amount of polyunsaturated fatty

acids (PUFA) that can be the site of oxidative stress

(Fendri et al. 2006). The phospholipid membrane

hypothesis of schizophrenia originates with sugges-

tions by Horrobin (1977) that schizophrenia might

be caused by prostaglandin excess or deficiency.

Correspondence: Dr A. Dietrich-Muszalska MD, PhD, II Department of Psychiatry, Medical University of Lodz, Czechoslowacka 8/10,

92-216 Lodz, Poland. Tel: �48 42 691 881 787. Fax: �48 42 675 7403. E-mail: tzn_lodz@post.pl

The World Journal of Biological Psychiatry, 2009; 10(1): 27�33

(Received 22 February 2007; accepted 22 March 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701361263

There is evidence that schizophrenia may be asso-

ciated with abnormal metabolism of membrane

phospholipids and polyunsaturated fatty acids,

particularly arachidonic acid, as precursors of

different eicosanoids (Fenton et al. 2000; Ross

2003). Oxidative neuronal membrane damage

measured by levels of lipid peroxidation products

leads to altered membrane dynamics and dysfunc-

tion, and may contribute to specific aspects of

schizophrenic symptomatology. Moreover, the pre-

sence of oxidative lipid products within the cell

membrane results in an unstable membrane struc-

ture, altered fluidity, permeability and impaired

signal transduction. In schizophrenia, phospholipids

and their metabolism, mainly in brain and in

other tissues, may be perturbed, and phospholipids

of neuron membranes altered; this may induce

changes in the levels of neurotransmitters and their

receptors, that in turn leads to their dysfunction.

In post-mortem brain tissues from schizophrenic

patients, reduced amounts of phosphatidylcholine

and phosphatidylethanolamine were observed

(Ross 2003). Lipid peroxidation products, together

with clinical symptoms, may be credible biological

markers of the cause of schizophrenia disorders

and useful for quicker diagnosis and specific therapy

(Fendri et al. 2006).

Free radicals (superoxide anion (/O+�2 ); the hy-

droxyl radical (HO+)), and other oxygen species, like

H2O2 and singlet oxygen (1DgO2) or organic per-

oxides generated by different mechanisms are

highly reactive. Under physiological conditions

they are constantly produced from normal oxidative

metabolism mainly in the mitochondrial electron

transport chain. A certain physiological level of

reactive oxygen species (ROS) is crucial for the

proper regulation of cell function (intracellular

signalling). Living organisms have developed several

ways to protect themselves from oxidant attacks.

The defence mechanisms include a variety of anti-

oxidant enzymes like glutathione peroxidase (GSH-

PX), catalase and superoxide dismutase (SOD).

An imbalance between free radical generation

and antioxidant defences in favour of the former

leads to oxidative stress. It is well known that free

radicals can oxidize proteins, lipids and nucleic

acids, leading to dysfunction and even cell death

(Lachance et al. 2001; Yao et al. 2001). In

schizophrenia, deregulation of free radical metabo-

lism and antioxidant capacity are altered (Yao et al.

1998), and low levels of antioxidant enzymes, such

as glutathione peroxidase and superoxide dismutase

have been noticed (Khan and Das 1997; Yao et al.

2001; Fendri et al. 2006). A common approach to

estimate oxidative stress in vivo is to determine the

end products of lipid peroxidation. Polyunsaturated

fatty acids, major components of membrane phos-

pholipids, are highly susceptible to free radical

attack and formation of peroxidation products.

Existence of peroxyradicals and lipid peroxide

intermediates within cell membranes results in

altered membrane structure, and their impaired

function and signal transduction. In schizophrenic

patients with symptoms of tardive diskinesia, lipid

peroxidation in plasma was found to be increased

(Peet et al. 1993). Some studies demonstrated an

increase of lipid peroxidation in the plasma of first

episode patients of schizophrenia, who had not

taken any antipsychotic drugs (Mahadik et al.

1995, 1998). The onset of schizophrenia may be

associated with disturbances in phospholipid meta-

bolism and changes in the concentration of PUFA

(Rotrosen and Wolkin 1987; Glen et al. 1994; Yao

et al. 1994). Decreased level of membrane PUFA

observed in schizophrenia may be associated partly

with an increase of phospholipase A2 (PLA2)

activity (Ross 2003). PLA2 catalyses the removal

of PUFA from phospholipid molecules, which then

take part in signalling reactions or are recycled back

into the phospholipid molecules. The increased

PLA2 activity in schizophrenia (Ross 2003) may

result in reduced levels of enzyme phosholipid/fatty

acid substrate, with concomitant increases in the

abundance of PLA2-derived metabolites, lysopho-

spholipids and glycerophospholipids. This is con-

sistent with the phospholipid alteration that occurs

in schizophrenia. The role of PLA2 in schizophre-

nia has also been investigated using molecular

genetic approaches.

It should be mentioned that therapy with anti-

psychotic drugs may affect lipid metabolism (Sagara

1998). Classic antipsychotic drugs may exhibit

prooxidative effects (Jeding et al. 1995; Sagara

1998), whereas the second-generation antipsychotic

drugs do not exhibit such effects; moreover they may

have even antioxidative effects (Jeding et al. 1995;

Kropp et al. 2005).

The measurement of lipid peroxidation products

in vitro relies on indirect methods. The most widely

used marker of lipid peroxidation is the level of

malonyldialdehyde or conjugated dienes (which

readily decompose) in plasma or blood cells.

However, among the most widely used methods,

the commonly used assay for the detection of

malondialdehyde, based on thiobarbituric acid

reactive substances (TBARS), is not sensitive and

is nonspecific since other metabolites may react

with the thiobarbituric acid (TBA) reagent, thus

leading to an overestimation of the end product.

The specific and reliable markers of lipid peroxida-

tion in vivo seem to be isoprostanes (IPs) (Basu

2004; Davi et al. 2004) or neuroprostanes (Roberts

28 A. Dietrich-Muszalska & B. Olas

et al. 1998), therefore the aim of our study was to

determine the level of urinary excretion of F2-

isoprostanes (8-isoprostaglandin F2; 8-isoPGF2a) in

patients with schizophrenia paranoid type disorders

(DMS-IV criteria). F2-isoprostanes (F2-IPs), which

are formed during the free radical-induced oxida-

tion of arachidonic acid, can be divided into four

major regioisomeric groups comprising a total of 64

isomers (Roberts and Morrow 2002). It has been

shown that levels of isoprostanes, measured as 8-

isoprostaglandin F2, increased in different disorders

(Greco et al. 2000; Souvignet et al. 2000; Cra-

cowski et al. 2002; Pratico et al. 2004; Milne and

Morrow 2006; Morrow 2006). The aim of our

study was also to estimate oxidative stress in a

schizophrenic patient group, using commonly used

biomarkers of this process � thiobarbituric acid

reactive substances (TBARS) � and determine the

level of thromboxane B2 in plasma and 11-dehy-

drothromboxane B2 (TXB2) in urine as end-pro-

ducts of enzymatic (cyclooxygenase) peroxidation

of arachidonic acid.

Material and methods

The urine and plasma samples were taken from

healthy volunteers (39) and schizophrenic patients

(47) aged 26�36 (average: 31; SD�4.8 years).

The healthy subjects included those with no

psychiatric, neurological or somatic disorders, no

history of head injuries, no disorders of lipid or

carbohydrate metabolism, no allergy, with normal

body mass index (BMI), not taking any medications

or addictive substances (including tobacco, alcohol

and other drugs of abuse), on a balanced diet

(meat and vegetables), and from similar socio-

economic background, using no antioxidant supple-

mentation.

Patients hospitalized in the II Psychiatric Depart-

ment of Medical University in Lodz, Poland, were

interviewed with a special questionnaire (treatment,

course of diseases, dyskinesis and other extrapyr-

amidal syndromes). According to DSM-IV criteria,

all patients had diagnosis of paranoid type (in acute

period of psychosis). All patients were treated with

second-generation antipsychotic drugs (risperidone,

clozapine, olanzapine). The mean time of schizo-

phrenia duration was 2 years. Patients and control

subjects with significant medical illness were ex-

cluded.

All subjects signed a consent form for participa-

tion in the study, according to the protocol accepted

by the Committee for Research on Human Subjects

of the Medical University of Lodz (number RNN/

899/2000).

8-isoPGF2a estimation in urine

The level of 8-isoPGF2a was estimated in urine

samples from control subjects and from schizophre-

nic patients using an immunoassay kit (Oxis Inter-

national, Inc.) according to the manufacturer’s

instructions. 8-isoPGF2a in samples (100 ml) or

standards, competes for binding (to the antibody

coated on the plate) with 8-isoPGF2a conjugated to

horseradish peroxidase (HRP). The peroxidase ac-

tivity results in colour development in the substrate

when added. The intensity of the colour is propor-

tional to the amount of HRP-bound 8-isoPGF2a and

inversely proportional to the amount of 8-isoPGF2ain the samples or standards.

Thromboxane B2 and 11-dehydrothromboxane B2

estimation

The level of thromboxane B2 (in plasma) and 11-

dehydrothromboxane B2 (in urine) was determined

in samples from control subjects and from schizo-

phrenic patients using an EIA kit (Cayman Chemi-

cal) according to the manufacturer’s instructions.

The amount of thromboxane B2 and 11-dehydro-

thromboxane B2 in the samples is presented in

pg/ml.

TBARS estimation in plasma

Samples of plasma were transferred to an equal

volume of 20% (v/v) cold trichloroacetic acid in

0.6 M HCl and centrifuged at 1200�g for 15 min.

One volume of clear supernatant was mixed

with 0.2 vol of 0.12 M thiobarbituric acid in

0.26 M Tris, pH 7.0, and immersed in a boiling

water bath for 15 min. Absorbance at 532 nm

was measured and results are expressed as nmol

of TBARS (Wachowicz 1984).

The statistical analysis was done by several

tests. In order to eliminate uncertain data, Grubbs

test was performed. All the values in this study

are expressed as mean9SEM. The statistically

significant difference between the control group

and schizophrenic patients was done using the

Mann�Whitney test using StatSoft Inc. ‘‘Statistica’’

v. 6.0.

Results

Our studies have shown that the level of 8-isopros-

taglandin F2a in urine from schizophrenic patients

was markedly higher than the level of 8-isoPGF2ain the healthy control subjects (P�3.5�10�12)

(Figure 1). The increase of 8-isoprostaglandin

F2a was extremely large and reached about 380%

Isoprostenes as indicators of oxidative stress in schizophrenia 29

(Figure 1). In the control group, the level of

8-isoPGF2a in urine was 296.8928.5 pg/mg of

creatinine, whereas samples from schizophrenic pati-

ents contained 1432.09113.2 pg of 8-isoPGF2a/mg

of creatinine.

We also observed a statistically increased level of

TBARS (P�0.000162) in plasma of schizophrenic

patients in comparison with the control group

(Figure 2).

The level of thromboxane B2 in plasma from

schizophrenic patients (87.7912.5 pg/ml) was

very similar to the level in control plasma (87.29

14.2 pg/ml) (Figure 3). The amounts of 11-dehy-

drothromboxane B2 in urine of schizophrenic pa-

tients did not differ from the amounts measured in

control urine samples (Figure 3).

Discussion

The recent research on oxidative changes of different

biomolecules, including lipids, could explain the role

of free radicals in various pathophysiological pro-

cesses and disorders (Khan and Das 1997; Fenton

et al. 2000; Yao et al. 2001; Pratico et al. 2004). In

different neuropsychiatric disorders, including schi-

zophrenia, oxidative stress may be determined by

special biomarkers, as well as in plasma and blood

cells or urine (van Kammen et al. 1989; Reilly et al.

1998; Yao et al. 1998; Dietrich-Muszalska et al.

2005; Fendri et al. 2006). Our present study

demonstrates that the measurement of the level of

8-isoPGF2a in urine is the most specific, sensitive

biomarker of oxidative stress in schizophrenia. The

first demonstration that prostaglandin F2-like com-

pounds, termed isoprostanes, derived from arachi-

donic acid (5,8,11,14-eicosatetraenoic acid), are

produced in humans was by Morrow et al. (1990).

Since that time, F2-isoprostanes have been used as

clinical markers of nonenzymatic lipid peroxidation

in vivo in different human diseases (Greco et al.

2000; Souvignet et al. 2000; Cracowski et al. 2002;

Morrow 2006). F3-isoprostanes are formed from

eicosapentaenoic acid (Basu 2004; Rokach et al.

2004; Gao et al. 2006). Isoprostane-like compounds

derived from docosahexaenoic acid, and termed

F4-neuroprostanes, seem also to be promising

tools for assessing lipid peroxidation in neuro-

logical and psychiatric disorders (Roberts and

Morrow 2002). F2-isoprostanes are esterified on

phospholipids in cell membranes and then released

in free form by phospholipases. Among all these

0

200

400

600

800

1000

1200

1400

1600

1800

schizophrenic patients grouphealthy subjects group

urin

ary

8-is

o-P

GF 2

α [p

g/m

g cr

eatin

ine]

Figure 1. The level of 8-isoprostaglandin in urine from schizo-

phrenic patients and in control urine obtained from healthy

volunteers. Results are means9SEM. The difference is statisti-

cally significant (P�3.5�10�12).

healthy subjects group schizophrenic patients group

nmol

TB

AR

S/m

l

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

1.8

2

Figure 2. The level of TBARS in plasma from schizophrenic

patients and in control plasma obtained from healthy volunteers.

Results are means9SEM. The difference is statistically significant

(P�0.000162).

0

50

100

150

200

250

300

350

400

450

schizophrenic patients grouphealthy subjects group

pg T

XB

2or

its

met

abol

ite/m

l

thromboxane B211-dehydro thromboxane B2

Figure 3. The level of thromboxane B2 (in plasma) and its

metabolite 11-dehydrothromboxane B2 (in urine) from schizo-

phrenic patients and in control plasma or urine obtained from

healthy volunteers. Results are means9SEM. The difference is

not statistically significant (P�0.05).

30 A. Dietrich-Muszalska & B. Olas

F2-isoprostanes, 8-isoPGF2a also named isoprosta-

glandin F2a type III (Rokach et al. 1997) or 15-

F2t-IsoP (Taber et al. 1997), is present in plasma or

urine and is extensively used as a sensitive clinical

marker of lipid peroxidation. Elevated levels of

8-isoPGF2a have been described in different dis-

orders (Mahrabi et al. 2001; Cracowski et al. 2002);

and, in animal models of oxidative injury, increased

levels of up to 200 fold has been observed (Roberts

and Reckelhoff 2001). Increased levels of 8-iso-

PGF2a were also described in patients with hyperch-

olesterolemia, diabetes mellitus, hepatorenal syn-

drome, scleroderma and in neurological disorders

such as Alzheimer’s disease, Parkinson’s disease and

Creutzfeldt�Jakob disease (Greco et al. 2000; Sou-

vignet et al. 2000; Cracowski et al. 2002; Pratico et

al. 2004; Milne and Morrow 2006). The present

study, for the first time, provides evidence that

schizophrenic patients in acute period of psychosis

have extremely high levels of urinary isoprostanes.

Isoprostanes have been measured in biological

fluids such as urine or plasma (Oguogho et al.

2000; Rokach et al. 2004), and in different cells or

tissues (Cracowski et al. 2001, 2002; Davi et al.

2004). Among the biological fluids available, most

studies were performed on urine because of the

non-invasiveness of the procedure and the lack of

artificial formation of isoprostanes. Therefore, we

measured the level of 8-isoPGF2a in urine of patients

with schizophrenic disorders and in healthy control

subjects. One major advantage of urinary measure-

ments of 8-isoPGF2a is that this compound is not

formed ex vivo by autooxidation in urine, even when

urine samples are left at room temperature for

10 days. The concentration of 8-isoPGF2a is un-

changed during storage of urine samples at �208C(Cracowski et al. 2002). Measurement of F2-

isoprostanes in a single sample of urine represents

the daily isoprostane secretion in humans (Oguogho

et al. 2000). We observed, for the first time, that the

production of isoprostanes in schizophrenic patients

in the acute period of psychosis is extremely high

compared with the control group (Figure 1), and our

results indicate that in schizophrenic patients in-

creased production of 8-isoPGF2a reflects oxidative

stress and oxidative damage. This suggests that free

radicals and oxidative stress may contribute to

schizophrenic disorders. The source of oxidized

arachidonate seems not only to be neuron mem-

branes (Horrobin et al. 1991; Horrobin et al. 1996;

Reddy and Yao 1999; Yao et al. 2000), but also

peripheral cells. Blood platelets may be a peripheral

marker in psychiatric illnesses, and our earlier

studies also indicate that a correlation exists between

increased platelet lipid peroxidation and inhibition

of the activity of superoxide dismutase in schizo-

phrenic blood platelets (Dietrich-Muszalska et al.

2005). Moreover, our earlier experiments demon-

strate that the level of reactive oxygen species

(/O+�2 ; H2O2, singlet oxygen, organic radicals) in

blood platelets from schizophrenic patients in acute

period of psychosis was higher than in platelets from

healthy subjects, and an increase of free radical

production in platelets from schizophrenic patient

leads to lipid peroxidation measured by increased

level of TBARS (Dietrich-Muszalska et al. 2005).

Our present study confirmed that the level of

TBARS in plasma from schizophrenic patients was

higher than in the control group (Figure 2). Akyol

et al. (2002) also showed that plasma TBARS levels

were increased in schizophrenic patients.

Thromboxane A2 (TXA2), like other eicosanoids,

is produced enzymatically from arachidonic acid by

many cells, mainly by blood platelets and megakar-

yocytes, and causes irreversible platelet aggrega-

tion and vascular contraction. Thromboxane A2 is

formed when arachidonic acid is released from the

membrane phospholipids by the sequential actions

of phospholipases (phospholipase A2), and by the

action of enzymes such as prostaglandin synthase

and thromboxane synthase. TXA2, like most lipid

mediators, is not a circulating hormone. It is formed

in response to local stimuli and exerts its effects

within a short time from its biosynthesis. TXA2 is an

unstable molecule with a half-life of approximately

30 s. Because of its instability, TXA2 is rapidly

converted non-enzymatically to TXB2, which is then

very quickly metabolized to urinary metabolites for

clearance by the kidneys (Funk 2001). Therefore,

measurement of the TXB2 metabolite � 11-dehydro

TXB2 � in urine or plasma may give better estima-

tion of in vivo TXA2 production. The levels of TXB2

and its metabolite 11-dehydro TXB2 in patients with

schizophrenic disorder did not differ from the

control group (Figure 3). This indicates that in

schizophrenic patients arachidonic acid is not meta-

bolized via enzymes to form TXA2; only non-

enzymatic peroxidation of arachidonic acid caused

by free radicals is observed. Yao et al. (1992) and

Deminisch et al. (1987) showed that the turnover of

phosphatidylinositol and arachidonic acid is in-

creased in platelets of patients with schizophrenia.

Moreover, it should be underlined that all schizo-

phrenic patients were treated with second-genera-

tion antipsychotic drugs which do not induce lipid

peroxidation (Dietrich-Muszalska et al. 2004).

Considering the data presented in this study, we

suggest that measurement in urine of isoprostanes

(particularly 8-isoPGF2a), which are the end pro-

ducts of free radical-induced arachidonic acid per-

oxidation, is a specific, valuable and non-invasive

bioindicator of oxidative stress in schizophrenia and,

Isoprostenes as indicators of oxidative stress in schizophrenia 31

together with clinical symptoms, may be useful for

quicker diagnosis and specific therapy.

Acknowledgements

Supported by the grant 502-11-692 from Medical

University of Lodz, Poland, and by grant 506/810

from University of Lodz, Poland. The protocol was

passed by the Committee for Research on Human

Subjects of the Medical University of Lodz, number

RNN/899/2000.

Declaration of Interest

None.

References

Akyol O, Herken H, Zu E, et al. 2002. The indices of endogenous

oxidative and antioxidative processes in plasma from schizo-

phrenic patients. The possible role of oxidant/antioxidant

imbalance. Prog Neuro-Psychopharmacol Biol Psychiatr 26:

995�1005.

Basu S. 2004. Isoprostanes: novel bioactive products of lipid

peroxidation. Free Radic Res 38:105�122.

Cracowski JL, Devillier P, Durand T, Stanke-Labesque F, Bessard

G. 2001. Vascular biology of the isoprostanes. J Vasc Res 38:

93�103.

Cracowski JL, Durand T, Bessard G. 2002. Isoprostanes as a

biomarker of lipid peroxidation in humans: physiology, phar-

macology and clinical implications. Trends Pharmacol Sci

23:1�4.

Davi G, Falco A, Patrono C. 2004. Determinations of F2-

isoprostane biosynthesis and inhibition in man. Chem Phys

Lipids 128:149�163.

Deminisch L, Gerbaldo H, Gebhart P, Georgi K, Bochnik HJ.

1987. Incorporation of 14C-arachidonic acid into platelet

phospholipids of untreated patients with schizophrenic form

or schizophrenic disorders. Psychiatr Res 22:275�282.

Dietrich-Muszalska A, Rabe-Jablonska J, Zgirski A. 2004. Wpl yw

dzial ania olanzapiny (Zolafren) na peroksydacje lipidow w

ludzkich plytkach krwi i osoczu w ukl adzie in vitro. Psychiatria

i Psychologia Kliniczna 4:13�19.

Dietrich-Muszalska A, Olas B, Rabe-Jablonska J. 2005. Oxidative

stress in blood platelets from schizophrenic patients. Platelets

16:386�391.

Fendri C, Mechri A, Khiari G, Othman A, Kerkeni A, Gaha L.

2006. Oxidative stress involvement in schizophrenia pathophy-

siology: a review. Encephale 32:244�252.

Fenton WS, Hibbeln J, Knable M. 2000. Essential fatty acids,

lipid membrane abnormalities, and the diagnosis and treatment

of schizophrenia. Biol Psychiatr 47:8�21.

Funk CD. 2001. Platelet eicosanoids. In: Colman RW, Hirsh J,

Marder VJ, Clowes AW, George JN, editors. Hemostasis and

thrombosis. Basic principles and clinical practice. Chapter 30.

Philadelphia, PA: Lippincott Williams & Wilkins. pp 533�539.

Gao L, Yin H, Milne GL, Porter NA, Morrow JD. 2006.

Formation of f-ring isoprostane-like compounds (F3-isopros-

tanes) in vivo from eicosapentaenoic acid. JBC on line

M601035200, 1�19.

Glen AIM, Glen EMT, Horrobin DF, et al. 1994. A red cell

membrane abnormality in a subgroup of schizophrenic pa-

tients: evidence for two diseases. Schizophr Res 12:53�61.

Greco A, Minghetti L, Levi G. 2000. Isoprostanes, novel markers

of oxidative injury, help understanding the pathogenesis of

neurodegenerative diseases. Neurochem Res 25:1357�1364.

Horrobin DF. 1996. Schizophrenia as a membrane lipid dis-

order which is expressed throughout the body. Prostaglandins

Leukotr Essent Fatty Acids 55:3�7.

Horrobin DF. 1977. Schizophrenia as a prostaglandin deficiency

disease. Lancet 1:936�937.

Horrobin DF, Manku MS, Hillman H, Iain A, Glen M. 1991.

Fatty acid levels in the brains of schizophrenics and normal

controls. Biol Psychiatr 30:795�805.

Jeding I, Evans PJ, Akanmu D, et al. 1995. Characterization of the

potential antioxidant and pro-oxidant actions of some neuro-

leptic drugs. Biochem Pharmacol 49:359�365.

Khan NS, Das I. 1997. Oxidative stress and superoxide dismutase

in schizophrenia. Biochem Soc Trans 25:418S.

Kropp S, Kern V, Lange K, et al. 2005. Oxidative stress during

treatment with first- and second-generation antipsychotics.

J Neuropsychiatry 17:227�231.

Lachance PA, Nakat Z, Jeong WS. 2001. Antioxidants: an integ-

rative approach. Nutrition 17:835�838.

Mahadik SP, Mukherjee S, Correnti E, Scheffer R. 1995.

Elevated levels of lipid peroxidation products in plasma of

drug-naıve patients at the onset of psychosis. Schizophr Res

15:66.

Mahadik SP, Mukherjee S, Scheffer RS, Correnti EE, Mahadik

JS. 1998. Elevated plasma lipid peroxidation at the onset of

nonaffective psychosis. Biol Psychiatr 43:674�570.

Mehrabi MR, Serbecic N, Ekmekcioglu C, et al. 2001. The

isoprostane 8-epi-PGF2a is a valuable indicator of oxidative

injury in human heart valves. Cardiovasc Pathol 10:241�245.

Milne GL, Morrow JD. 2006. Isoprostanes and related com-

pounds: update 2006. Antioxidants Redox Signaling 8:1379�1384.

Morrow JD. 2006. The isoprostanes � unique products of

arachidonate peroxidation: their role as mediators of oxidant

stress. Curr Pharm Design 12:895�902.

Morrow JD, Hill KE, Burk RF, Nammour TM, Badr KF, Roberts

LJ. 1990. A series of prostaglandin F2-like compounds are

produced in vivo in humans by a non-cyclooxygenase, free

radical-catalyzed mechanism. Proc Natl Acad Sci USA

87:9383�9387.

Oguogho A, Ferlitsch A, Sinzinger H. 2000. LDL-apheresis

decreases plasma levels and urinary extretion of 8-epi-

PGF2a. Prostaglandins Leukotr Essent Fatty Acids 62:209�216.

Peet M, Laugharne J, Rangarajan N, Reynolds GP. 1993. Tardive

dyskinesia, lipid peorxidation, and sustained amelioration with

vitamin E treatment. Int Clin Psychophamacol 8:151�153.

Pratico D, Rokach J, Lawson J, FitzGerald GA. 2004. F2-

isoprostanes as indices of lipid peroxidation in inflammatory

diseases. Chem Phys Lipids 128:165�171.

Reddy RD, Yao JK. 1999. Membrane protective strategies in

schizophrenia: conceptual and treatment issues. In: Peet M,

Glen I, Horrobin DF, editors. Phospholipid spectrum disorder

in psychiatry. Lancashire, UK: Marius Press. pp 75�88.

Reilly MO, Lawson JA, FitzGerald GA. 1998. Eicosanoids and

isoeicosanoids: indices of cellular function and oxidant stress.

J Nutr 128:434�438S.

Robers LJ II, Morrow JD. 2002. Products of the isoprostane

pathway: unique bioactive compounds and markers of lipid

peroxidation. Cell Mol Life Sci 59:808�820.

Roberts LJ II, Reckelhoff JF. 2001. Measurement of F2-isopros-

tanes unveils profound oxidative tsress in adged rats. Biochem

Biohys Res Comm 287:254�256.

32 A. Dietrich-Muszalska & B. Olas

Roberts LJ II, Montine TJ, Markesbery WR, et al. 1998.

Formation of isoprostanes-like compounds (neuroprostanes)

in vivo from oxidation of docosahexaenoic acid. J Biol Chem

273:13605�13612.

Rokach J, Khanapure SP, Hwang SW, Adiyaman M, Lawson JA,

FitzGerald GA. 1997. The isoprostanes: a perspective. Pros-

taglandins 54:823�851.

Rokach J, Kim S, Bellone S, et al. 2004. Total synthesis of

isoprostanes: discovery and quantitation in biological systems.

Chem Phys Lipids 128:35�56.

Ross BM. 2003. Phospholipid and eicosanoid signaling distur-

bances in schizophrenia. Prostaglandins Leukotr Essent Fatty

Acids 69:407�412.

Rotrosen J, Wolkin A. 1987. Phospholipid and prostaglandin

hypotheses of schizophrenia. In: Meltzer HY, editor. Psycho-

pharmacology. The third generation of progress. New York:

Raven Press. pp 759�764.

Sagara Y. 1998. Induction of reactive oxygen species in neurons by

haloperidol. J Neurochem 71:1002�1012.

Souviget C, Cracowski JL, Stanke-Labesque F, Bessard G. 2000.

Are isoprostanes a clinical marker for antioxidant drug in-

vestigation. Fundam Clin Pharmacol 14:1�10.

Taber DF, Morrow JD, Roberts JL II. 1997. A nomenclature

system for the isoprostanes. Prostaglandins 53:63�67.

Wachowicz B. 1984. Adenine nucleotides in thrombocytes of

birds. Cell Biochem Funct 2:167�170.

van Kammen DP, Yao JK, Goetz K. 1989. Polyunsaturated fatty

acids, prostaglandins, and schizophrenia. Ann NY Acad Sci

559:411�423.

Yao JK, van Yasaei P, Kammen DP. 1992. Increased turnover of

platelet phosphatidylinositol in schizophrenia. Prostaglandins

Leukot Essent Fatty Acids 46:39�46.

Yao JK, van Kammen DP, Gurkis WJ. 1994. Correlation of fatty

acid abnormalities with clinical measures in schizophrenia.

Schizophr Res 11:124.

Yao JK, Reddy R, McElhinny LG, van Elhinny K. 1998. Reduced

status of plasma total antioxidant capacity in schizophrenia.

Schizophr Res 32:1�8.

Yao JK, Leonard S, Reddy R. 2000. Membrane phospholipids

abnormalities in portmorten brains from schizophrenic pa-

tients. Schizophr Res 42:7�17.

Yao JK, Reddy RD, Kammen DP. 2001. Oxidative damage and

schizophrenia. CNS Drugs 15:287�310.

Isoprostenes as indicators of oxidative stress in schizophrenia 33

ORIGINAL INVESTIGATION

A single session of rTMS over the left dorsolateral prefrontal cortexinfluences attentional control in depressed patients

MARIE-ANNE VANDERHASSELT1, RUDI DE RAEDT1, CHRIS BAEKEN2,

LEMKE LEYMAN1 & HUGO D’HAENEN2

1Department of Psychology, Ghent University, Ghent, Belgium, and 2Department of Psychiatry, Academic Hospital, Free

University of Brussels (VUB), Brussels, Belgium

AbstractDepressed patients are impaired in the ability to shift their focus of attention. This attentional control process is related todysfunctions in the dorsolateral prefrontal cortex (DLPFC). It has been proposed that a dorsal circuit plays an importantrole in the interaction between emotional and attentional information processing. However, because the different emphasisof fundamental cognitive neuroscience research and clinical research of repetitive transcranial magnetic stimulation (rTMS)over the DLPFC, little research has been done on the effects of rTMS on cognitive functioning after a single stimulationsession to explore the neural systems underlying depression. This study was conducted as a double-blind, placebo-controlled, crossover, within subjects design. Sixteen depressed patients performed a modified task switching paradigm,before and after receiving high frequency (HF) versus placebo rTMS over the left DLPFC. One session of HF- rTMS overthe left DLPFC had a specific beneficial effect on task-switching performance, whereas mood remained stable. Anti-depressant effects of rTMS could be related to the same neurochemical changes that underlie cognitive functioning.Therefore, task switching performance may provide a unique window into the extent of antidepressant effects which can beconsidered as second-order long-term effects possibly related to primary alternations in cognitive functioning.

Key words: rTMS, dorsolateral prefrontal cortex, executive functioning, mood, depression

Introduction

As a non-invasive technique to influence brain

circuits, repetitive transcranial magnetic stimulation

(rTMS) induces alterations of neuronal activity that

may affect mood and cognition (Moser et al. 2002).

Although rTMS revealed to be an interesting tool to

investigate cognitive functions in healthy subjects

(Vanderhasselt et al. 2006a), much rTMS research

has merely focused on the effects of rTMS as a

treatment procedure for major depression. Regard-

ing the antidepressant effects, research using rTMS

over left dorsolateral prefrontal cortex (DLPFC)

yielded promising results (for a review see Burt et al.

2002). A meta-analysis of Kozel and George (2002)

showed substantial clinical improvements in self-

reported mood measurements in left prefrontal

rTMS studies. On the contrary, an rTMS meta-

analysis of Martin et al. (2003), reported inadequate

and inconsistent evidence for the antidepressant

evidence in depression. They concluded that more

specific research is required concerning its under-

lying working mechanisms.

Cognitive functioning could represent the under-

lying fundamental working mechanism of rTMS.

However, research on the effects of rTMS on

cognitive functioning in depressed patients shows

inconsistent findings. We are aware of a small

number of rTMS studies where improved cognitive

performance was found in depressed patients (Mo-

ser et al. 2002). Triggs and co-workers (1999)

found an improvement in neuropsychological per-

formance following left-frontal rTMS after 10

weeks as well as after 3 months. Other researchers

(Loo et al. 1999) reported trends for improvement

in neuropsychological performance after 4 weeks of

active rTMS. However, as the authors mentioned,

these effects could be due to practice effects since

they used no placebo control condition. On the

other hand, an absence of cognitive effects after

several weeks of daily rTMS in depressed patients

Correspondence: Marie-Anne Vanderhasselt, Department of Psychology, Ghent University, Henri Dunantlaan 2, 9000 Ghent, Belgium.

Tel: �32 9264 8612. Fax: �32 9264 6489. E-mail: MarieAnne.Vanderhasselt@UGent.be

The World Journal of Biological Psychiatry, 2009; 10(1): 34�42

(Received 20 March 2007; accepted 14 November 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701816514

has frequently been reported (e.g. Speer et al.

2000). In some cases, cognitive improvement was

found in combination with no therapeutic effect of

rTMS (e.g. Padberg et al. 1999).

In contrast, O’Connor and co-workers (2005)

found that rTMS over a 2-week period improved

performance on cognitive tasks and that these

cognitive effects were greater in those patients

who showed a significant antidepressant effect of

rTMS. Given the interaction between cognition

and emotion, the causal status of the improved

cognitive effects could not be demonstrated because

it is widely recognised that improved mood also

influences cognitive functions (e.g. Boggio et al.

2005). There has recently been an increased inter-

est in the study of cognitive performance as a

marker of brain pathology in affective disorders

(Stuss et al. 2003). However, support for cognitive

functioning as underlying working mechanisms of

depression, have been inconclusive to date (for a

review see Martin et al. 2003). Since studying the

interface between cognition and emotion becomes

more prominent, research regarding immediate

changes in cognition after rTMS in depressed

patients, is most important. Studies investigating

the influence of a single rTMS session on mood

and cognition are very scarce.

Depression has been related to dysfunctions in

specific aspects of executive processes, such as

strategic attentional processes and selective set shift-

ing, whereas relative automatic processes remain

intact (Hartlage et al. 1993). Austin (2001) demon-

strated that depressed patients are impaired in their

ability to shift the focus of attention. This attentional

process can be studied using a task switching para-

digm that requires participants to rapidly switch

between two or more tasks across consecutive trials

(Arbuthnott and Frank 2000).When subjects switch

between tasks, they must both inhibit the previous

relevant task and re-engage in a different task. We

refer to the latter process as task set inhibition

(Arbuthnott and Frank 2000).

Functional magnetic resonance imaging (fMRI)

studies have reported decreased activation of the

left DLPFC in depression (Mayberg et al. 1999).

Executive functions such as task-set inhibition are

thought to depend on the left DLPFC (MacDonald

et al. 2000). Using task-switching paradigms,

several neuroimaging studies have shown that the

lateral prefrontal cortex is more active on task-

switch then on task-repetition trials (Sohn et al.

2000; Crone et al. 2006). In addition, rTMS

research has pointed out the role of the DLPFC

in overcoming inhibition of a previously performed

task during task switching in healthy volunteers

(Vanderhasselt et al. 2006b). However, analogous

and controlled neuropsychological rTMS research

within a depressed population is limited (Bermpohl

et al. 2006).

The aim of the present study was to evaluate the

specific influence of a single session of rTMS on task

switching and mood in depressed patients. We used

a task switching paradigm with three following

conditions. During two separate blocks of repetitive

trials, the participants were pretrained on two simple

tasks afforded by a set of auditory or visual stimuli.

The responses of these stimuli required mutually

incompatible response demands. In the third block

(task switching block), participants alternated be-

tween the two pretrained tasks (switch trials) or

repeated the same task (repetitive trials). During this

block, they had no previous knowledge which task

they had to perform, which requires continuous task

set inhibition.

In line with several lesion studies (Aron et al.

2004), neurophysiological studies (Garavan et al.

2002) and analogous studies in healthy subjects

(Vanderhasselt et al. 2006a), we predicted that high

frequency (HF)-rTMS over the DLPFC in de-

pressed patients, compared to sham placebo, would

have a primary influence on attentional control

processes. More specific, we expected that after

stimulation over the left DLPFC, the reaction time

on the switch trials in the task switching block, in

contrast to the trials in the first repetitive blocks,

would decrease. We are aware that mood mediates

executive functioning (Damasio, 1996) and that

this understanding is of great importance when

studying subjects with major depression. Therefore,

we will also evaluate if possible mood changes

mediate the effects. Based on evidence from recent

studies (e.g. Bermpohl et al. 2006), we predicted

that there would be no mood changes after a single

session of HF-rTMS over the left DLPFC.

To control for shortcomings mentioned in other

rTMS research, we made use of a sham-controlled

condition, a large time interval between stimulation

sessions, stimulation of one single region per session

in order to eliminate interaction effects with the

previous stimulation, brain imaging to determine the

exact position of stimulation, a large number of

pulses at high stimulation intensity and a large

uniform sample (Baeken et al. 2006, 2007).

Experimental procedure

This study was part of a larger project investigating

the influence of rTMS on different neuro-cognitive

markers.

rTMS and depression: influence on cognition 35

Participants

A total of sixteen right-handed depressed patients

(10 women and six men, mean age 42 years; SD�11.20) were enrolled in the study, which has been

approved by the ethics committee of the hospital

(UZ Brussel). After the nature of the procedure had

been fully explained, all participants gave written

informed consent before inclusion. They all under-

went a physical examination and an EEG. All

patients fulfilled the DSM-IV criteria for major

depression as confirmed by the Mini-International

Neuropsychiatric Interview (MINI) (Sheehan et al.

1998). Depression severity was measured using the

17-tem Hamilton Depression Rating Scale (HDRS)

(score at least�16). Right-handedness was evalu-

ated by the van Strien hand preference screening

questionnaire (Van Strien 2001). Patients were all

free of antidepressant medication. Some patients

were washed out for at least14 days (and for

minimally 3 weeks if they were on fluoxetine) before

the start of the study, other patients did not use

psychotropic medication. Only patients who did not

need rescue medication or concomitant therapies

during this period were included in the study.

Importantly, they had regularly contact with a

psychiatric to evaluate possible deterioration of their

mood.

An overview of demographic and clinical charac-

teristics of every patient is presented in Table I.

Design

This study was conducted as a double-blind, pla-

cebo-controlled crossover, within subjects design, in

which participants received 20 min of real (10-Hz)

or placebo (sham) rTMS over the left DLPFC.

Procedure

On the morning of stimulation, the investigation

started at around 09:00 h with a baseline mood

measurement. Patients were asked to indicate their

current mood state on visual analogue mood scales

(VAS). These VAS scales consisted of subscales for

‘‘depression’’, ‘‘anger’’, ‘‘fatigue’’, ‘‘vigour’’ and

‘‘tension’’. The participants were asked to describe

how they felt ‘‘at that moment’’ by indicating on

horizontal 10-cm lines whether they experienced the

five abovementioned mood states, from ‘‘totally not’’

to ‘‘very much’’.

Afterwards, participants performed a compu-

terised self-paced switching task, programmed in

Delphi. The device consists of a board (connected to

a computer) with a central pushbutton around which

eight pushbuttons are positioned in a semicircle. In

addition, a loudspeaker and a pedal are attached to

the device. For a sketch of the experiment, we refer

to Figure 1.

This paradigm contains three blocks, in which the

first two blocks consist of repetitive tasks (one block

with 28 visual stimuli and one block with 28

auditory stimuli) and the last block consists of switch

trials. Both motor responses involved different

modalities and thereby non-overlapping neural sys-

tems.

During the first block, participants were told that,

when they saw an illuminated pushbutton, they had

to remove their finger from the central pushbutton

and push out the light. At each visual trial, one out of

four of the eight pushbuttons could illuminate

randomly. For visual trials, two independent compo-

nents of the reaction time were recorded. Decision

time (DT), a central cognitive component, reflects

the time necessary to initiate a response and corre-

sponds to the time that elapses between stimulus

Table I. Demographic and clinical characteristics of the depressed patients.

Subject Age Gender Duration current episode At least failed trial Psychotropic medication during rTMS

1 35 F 1 year 1.00 None

2 22 F 2 months 0.00 None

3 38 M 5 months 3.00 None

4 55 F 1 year 1.00 None

5 51 F 1 year 1.00 None

6 51 F 2 year 3.00 None

7 35 M 3 year 2.00 None

8 61 M 2 year 1.00 Lendormin

9 34 M 7 year 5.00 None

10 52 F 5 year 3.00 None

11 45 F 2 year 3.00 None

12 48 F 5 year 3.00 Regulton, pantozol

13 25 M 2 year 3.00 None

14 34 M 11 year 3.00 None

15 42 F 4 year 3.00 None

16 53 F 2 months 1.00 None

36 M.-A. Vanderhasselt et al.

onset and the release of the central pushbutton.

Movement time (MT), a peripheral executive com-

ponent, represents the motor activity or the time that

is required to complete the response (Gorus et al.

2006). In the second block, participants were in-

structed to press their foot on a pedal when they

heard a buzzer. Participants were instructed to let

their foot hover over the pedal during the entire

experiment. In this task, only total reaction times can

be recorded.

The third block, the double task condition, was an

alternating switch block with 29 auditory and 28

visual stimuli that were randomly mixed. During the

switch block, 24 out of the 29 auditory signals and

25 out of the 28 visual signals were switch trials (only

switch trials were analysed). Because of a technical

problem, the first trials were lost and only 22

auditory and 22 visual switch trials could be

recorded. Patients were instructed to focus their

attention to the visual stimuli and to switch attention

when the auditory stimuli would appear. After each

response on a visual trial, they had to return their

finger to the central pushbutton as quick as possible,

which triggered stimulus onset asynchrony (SOA)

for the next trial. After each auditory signal, they had

to remove their foot of the pedal to trigger SOA for

the next trial. In each of the three tasks, SOA

differed randomly between 3000 and 6000 ms.

The same sequence was used for all participants.

The participants were instructed to respond as

quickly and as accurately as possible. If errors

occurred, stimuli were replaced by a new stimulus

in order to obtain the same amount of correctly

performed reactions for every participant. Delayed

reactions time latencies (�3000 ms) were removed

from the analyses. For a timeline of the experiment,

we refer to Figure 2.

Subsequently, HF-rTMS over the left DLPFC

was performed using a Magstim high-speed mag-

netic stimulator (Magstim Company Limited,

Wales, UK), connected to a specially designed figure

8-shaped coil. Before rTMS application, the motor

threshold (MT) of each subject was determined

individually using EMG. Stimulation intensity was

Auditory trials

Visual trials

Figure 1. A sketch of the task switching experiment.

*RETURN FINGER TO CENTRAL LIGHT**REMOVE FOOT FROM PEDAL

LIGHT* OR BUZZER**

RESPONSE:*PUSH OUT THE LIGHT: 4 out of 8 lights**PRESS FOOT ON PEDAL

SOA: 3000-6000 msec

Figure 2. Timeline of the task switching experiment.

rTMS and depression: influence on cognition 37

110% of MT of the right abductor pollicis brevis

muscle, stimulation frequency was 10 Hz (HF-

rTMS). Forty trains of 3.9 s duration, separated by

an intertrain interval of 26.1 s (1560 pulses per

session) were applied. The total stimulation time

was approximately 20 min. The precise left DLPFC

(Brodmann area 9/46) stimulation site and position

of the coil was defined under magnetic resonance

(MRI) non-stereotactic guidance. Perpendicular to

this point, the precise stimulation site on the skull

was marked and stimulated. Safety guidelines, based

on recent available safety studies on rTMS, were

followed (Wassermann 1998; Anand and Hotson

2002). Afterwards, approximately 10 min following

stimulation, the task switching paradigm was again

administered.

Real and sham stimulation were performed at the

same place on the skull, but for sham stimulation the

figure 8-shaped coil was held at an angle of 90% only

resting on the scalp with one edge, following recent

SHAM guidelines (Anand and Hotson 2002). Be-

cause the nature of the procedure had been ex-

plained in the informed consent before the start of

the study, subjects were fully aware that one of the

sessions was placebo. During stimulation, all parti-

cipants wore earplugs and were blindfolded to

guarantee that they could not see the difference

between the placebo sham and the real rTMS

procedure. The order of the stimulation conditions

sham (placebo)-real was counterbalanced with a

delay of 1 week between the two stimulation

sessions. The same individuals were stimulated at

the same moment of the day.

The mood scales (VAS) were used to record mood

at various stages of the experiment, respectively at

baseline (pre), immediately after rTMS (post1) and

after task-switching performance (930 min post

stimulation, post2).

Results

All patients tolerated the experimental procedure

well, only two reported mild headaches after the

real stimulation procedure. Debriefing after the

experiment revealed that participants tended to

believe after each session that this particular session

was the real stimulation. Significance level was set at

PB0.05.

Mood effects

Analysis of variance (ANOVA) was used to analyse

mood changes. Because of some missing values on

the mood scales, data of only 14 patients were

analysed. We used a 2�3 within-subjects ANOVA

with stimulation (rTMS-SHAM) and time (pre,

post1, post2) as within-factors and mood scores,

evaluated with the different VAS scales, as depen-

dent variables. As expected, no main effects reached

significance, neither the crucial interaction effects

between time and stimulation (Fanger(2,12) �1.994,

P�0.18, ns; Fvigor(2,12) �0.119, P�0.89, ns;

Ffatigue(2,12)�1.087, P�0.37, ns; Ftension(2,12)�2.080, P�0.17, ns; Fdepression(2,12)�0.509, P�0.62, ns).

Therefore, we conclude that there were no mood

changes from baseline caused by left prefrontal HF-

rTMS compared to ratings immediately after stimu-

lation and after the second task-switching task.

Reaction time on switch trials during the double task

(third block)

Switch effects were analysed using mixed ANOVAs.

The basic design was a 2�2�2 design with

stimulation condition (rTMS-SHAM), and time

(pre�post) as within-subjects factors and the order

of the stimulation condition as between-subjects factor.

The dependent variables were the mean reaction

time (in milliseconds) on auditory and visual (both

DT and MT) switch trials, corrected for individual

processing speed (RT of auditory and visual switch

trials of block 3 minus RTon repetitive trials of block

2 and block 1, respectively). The order of stimula-

tion yielded no main effect and was not implied in

any interaction effect (F(2,14) �2.614, P�0.12,

ns). Consequently, this factor was left out in all

further analyses.

Regarding our a priori assumptions, the crucial

interactions between stimulation condition ‘‘rTMS/

SHAM’’ and time ‘‘pre�post’’ for DT (F(1,15) �5.157, P�0.04, s) as well as the interaction between

stimulation condition ‘‘rTMS/SHAM’’ and time

‘‘pre�post’’ for ART (F(1,15)�7.261, P�0.02, s)

were significant. The interaction between stimula-

tion condition ‘‘rTMS/SHAM’’ and time ‘‘pre�post’’

for MT (F(1,15) �3.318, P�0.09, ns) was not

significant.

The significant interaction effects were further

analysed by paired t-tests to test our specific a priori

hypotheses. Paired t-tests indicated a significant

decreased reaction time for both auditory switch

trials (t(15)�3.301, P�0.01, s) and for the DT of

the visual switch trials (t(15)�3.457, P�0.01, s)

after the rTMS stimulation as compared the pre

rTMS task. For the SHAM placebo condition, we

found no significant pre�post differences for the

visual stimulus modality (t(15)�0.474, P�0.64,

ns) nor for the auditory stimulus modality (t(15)�0.35, P�0.73, ns). For RTs, we refer to Tables II

and III.

38 M.-A. Vanderhasselt et al.

Reaction time on repetitive trials during the single task

(first and second block)

To further verify our specific hypothesis, we addi-

tionally explored the influence of rTMS on the

reaction times during the two repetitive task blocks,

using ANOVA’s. The basic design was a 2�2

factorial ANOVA with stimulation condition (rTMS-

SHAM), and time (pre�post) as within-subject

factors. The dependent variables were the mean

reaction time (in milliseconds) on auditory and

visual repetitive trials.

The only significant main effect was for DT,

showing faster latencies in the SHAM condition

(F(1,15)�9.353, P�0.008, s). However, the crucial

interaction effects between stimulation condition

‘‘rTMS/SHAM’’ and time ‘‘pre�post’’ for DT

(F(1,15)�3.130, P�0.10, ns), for ART (F(1,15)�1.988, P�0.18, ns) and for MT (F(1,15)�1.596,

P�0.23, ns) were not significant. Reaction times of

the auditory trials and visual trials in the task

repetition blocks are presented in Table III.

Discussion

The influence of HF-rTMS in medication free

depressed patients might offer new avenues to study

the relationship between basic cognitive processes

and depression. To our knowledge, this is the first

study to explore the influence of a single session HF-

rTMS over the left DLPFC on mood and task

switching performance in depressed patients.

We found that one session of HF- rTMS over the

left DLPFC had a beneficial effect on task switching.

These results are in line with previous rTMS

research that related task switching performance of

healthy volunteers to activity in the DLPFC (Van-

derhasselt et al. 2006b). More specifically, we found

that reaction time latencies of switch trials during the

task switching block for both visual and auditory

trials significantly decreased after rTMS, whereas

sham yielded no effects. No differences on the

repetitive trials of the single task blocks were found,

which indicates that our results are not caused by a

general increased arousal. Moreover, peripheral

movement time was not influenced by the rTMS

procedure, which means that the effects are related

to central cognitive functioning. Since we used a

sham controlled crossover design, the improved

cognitive performance associated with HF rTMS

could not be related to a non-specific effect.

As predicted, after a single session of HF-rTMS in

depressed patients, no mood effects were found,

indicating that the beneficial cognitive effects are not

related to an immediate antidepressant effect of

rTMS. In line with the imbalance theory of depres-

sion, which is based on findings of a hypo-activity of

the left relative to the right prefrontal cortex in the

pathophysiology of depression (Drevets, 2000),

other studies reported that successive sessions of

Table II. Mean reaction time latencies approximate standard deviation of the task switching block in a SHAM control and an active rTMS

stimulation condition.

rTMS SHAM

Pre Post Pre Post

Auditory switch trials 420,13 (110,26) 363,94* (76,36) 439,35 (92,95) 419,67 (47,77)

Visual switch trials DT 442,48 (67,76) 425,47* (75,81) 414,34 (58,42) 400,99 (37,54)

Visual switch trials MT 282,33 (64,98) 279,42 (51,92) 290,03 (50,70) 310,39 (45,79)

For both auditory and visual switch trials, RTs before and after rTMS/SHAM were compared.

*RT for visual trails (DT) and auditory trials was decreased after rTMS at PB0.05. No other behavioural changes in response to rTMS or

SHAM reached statistical significance.

Table III. Mean reaction time latencies and approximate standard deviation of the task repetition blocks in a SHAM control and an active

rTMS stimulation condition.

rTMS SHAM

Pre Post Pre Post

Repetitive auditory trials 313,82 (67,69) 327,23 (79,01) 359,38 (103,46) 332,62 (66,20)

Repetitive visual trials; DT 362,21 (48,35) 401,46 (80,93) 342,89 (41,97) 335,19 (31,61)

Repetitive visual trials; MT 311,02 (63,68) 312,73 (62,05) 298,21 (57,09) 315,12 (74,24)

Auditory and visual trials of the respective repetitive blocks were compared before and after rTMS. No behavioural changes reached

statistical significance.

rTMS and depression: influence on cognition 39

HF-rTMS over the left DLPFC transiently increase

mood (Kozel and George 2002).

The current study provides a contribution to the

literature on cognitive control and attentional

processes that might be related to the underlying

antidepressant effect of rTMS, administered with

parameters typically used in clinical studies of

major depression.

Our results are in line with research from Haus-

mann and co-workers (2004), who used multiple

sessions of unilateral rTMS (HF over the left

DLPFC) as well as bilateral combined rTMS (HF

over the left DLPFC and low frequency over the

right DLPFC). They reported mild beneficial effects

on attention partly independent of its antidepressant

efficacy.

Wagner and co-workers (2006) stated that, as

there is good clinical evidence for a relationship

between stimulation intensity of rTMS over the left

DLPFC and its antidepressant efficacy (Padberg

et al. 2002), analyses of cognitive effects would be

particularly interesting. They suggest that increased

attentional control processes after rTMS might not

be limited to a period immediately after stimulation

but may possibly reflect primary neurochemical

alterations and, as a result, may be a sensitive

cognitive measure to trace short-term effects of

rTMS in humans (Wagner et al. 2006). Correspond-

ing to this consideration, recent findings by Pogarell

and colleagues (2006) demonstrated an increased

dopaminergic neurotransmission in the striatum as

an acute neurobiological antidepressant action of left

dorsolateral rTMS.

In addition, Bermpohl and colleagues (2006)

recently suggested that, given the link between

emotional and cognitive functions, a switch task

could be used as a rough indicator for the general

clinical state of depressed patients. This implies that

cognitive tasks may serve as a valuable tool for

studying acute rTMS effects in depressed patients

(Bermpohl et al. 2006).

A study of Moller and co-workers (2006) found

that the P300, a major endogenous brain event-

related potential (ERP) component which has been

found to be reduced in patients with depression, was

significant increased in amplitude after rTMS over

the left prefrontal cortex compared to sham stimula-

tion. An increase in P300 amplitude is indicative of

improved attentiveness (Sommer and Matt 1990;

Picton 1992, in Moller et al. 2006). However, no

significant antidepressant effects after 5 days

of stimulation were found in this study. Nevertheless,

similar rTMS procedures administering this treat-

ment over a longer period reported a clear antide-

pressant outcome (Gershon et al. 2003).

One could thus suggest that primary to the

antidepressant effect of rTMS, cognitive changes

emerge.

In addition, neuroimaging studies demonstrated

functional changes in blood flow within the DLPFC

and connected regions after HF-rTMS (Kimbrell et

al. 1999; Speer et al. 2000; Paus and Barrett 2004).

The antidepressant effects of rTMS might be related

to the same neurochemical changes in the neurocir-

cuitry that underlie cognitive attentional control

processes. It has already been demonstrated that

this dorsal circuit plays an important role in the

interaction between emotional and attentional in-

formation processing (Taylor and Fragopanagos

2005) and that dysfunctional activation in this area

of the brain may possibly contribute to the develop-

ment of affective disorders (George et al. 1994).

In sum, the use of task switching paradigms may

provide a unique window into the extent of anti-

depressant effects (Wagner et al. 2006) which can be

considered as second-order long-term effects possi-

bly caused by primary alternations in cognitive

functioning.

Future research combining rTMS with functional

brain imaging is necessary for providing evidence of

these cognitive changes as a marker of antidepres-

sant effects.

Acknowledgements

This study was supported by a grant from the

Scientific Fund W. Gepts UZ Brussel. The authors

thank Professor Dr R. Luypaert, P. Clerinx and

P. Van Schuerbeeck from the department of radi-

ology for the AZ VUB technical assistance in data

collection and MRI data analysis. The authors also

wish to thank Mark Moens and Marc Vercauteren

for the development of the Task-switching task

(www.subliemservice.be).

Declaration of Interest

None.

References

Anand S, Hotson J. 2002. Transcranial magnetic stimulation:

Neurophysiological applications and safety. Brain Cognition

50:366�386.

Arbuthnott K, Frank J. 2000. Executive control in set switching:

Residual switch cost and task-set inhibition. Can J Exp Psychol

Rev Can Psychol Exp 54:33�41.

Aron AR, Monsell S, Sahakian BJ, Robbins TW. 2004. A

componential analysis of task-switching deficits associated

with lesions of left and right frontal cortex. Brain 127:

1561�1573.

Austin MP, Mitchell P, Goodwin GM. 2001. Cognitive deficits in

depression � Possible implications for functional neuropathol-

ogy. Br J Psychiatry 178:200�206.

40 M.-A. Vanderhasselt et al.

Baeken C, Leyman L, De Raedt R, Vanderhasselt MA, D’haenen

H. 2006. Lack of impact of repetitive high frequency transcra-

nial magnetic stimulation on mood in healthy female subjects. J

Affect Disord 90:63�66.

Baeken C, De Raedt R, Leyman L, et al. 2007. The impact of one

session of HF-rTMS on Salivary Cortisol in Healthy Female

Subjects. World J Biol Psychiatry 13:1�5.

Bermpohl F, Fregni F, Boggio PS, et al. 2006. Effect of low-

frequency transcranial magnetic stimulation on an affective go/

no-go task in patients with major depression: Role of stimula-

tion site and depression severity. Psychiatry Res 141:1�13.

Boggio PS, Fregni F, Bermpohl F, et al. 2005. Effect of repetitive

TMS and fluoxetine on cognitive function in patients with

Parkinson’s disease and concurrent depression. Move Disord

20:1178�1184.

Burt T, Lisanby SH, Sackeim HA. 2002. Neuropsychiatric

applications of transcranial magnetic stimulation: a meta

analysis. Int J Neuropsychopharmacol 5:73�103.

Crone EA, Wendelken C, Donohue SE, Bunge SA. 2006. Neural

evidence for dissociable components of task-switching. Cere-

bral Cortex 16:475�486.

Damasio AR. 1996. The somatic marker hypothesis and the

possible functions of the prefrontal cortex. Phil Trans R Soc

London Ser B Biol Sci 351:1413�1420.

Drevets WC. 2000. Neuroimaging studies of mood disorders. Biol

Psychiatry 48:813�829.

Garavan H, Ross TJ, Murphy K, Roche RAP, Stein EA. 2002.

Dissociable executive functions in the dynamic control of

behavior: Inhibition, error detection, and correction. Neuro-

image 17:1820�1829.

George M, Ketter TA, Post RM. 1994. Activation studies in mood

disorders. Psychiatry Ann 24:648�652.

Gershon AA, Dannon PN, Grunhaus L. 2003. Transcranial

magnetic stimulation in the treatment of depression. Am J

Psychiatry 160:835�845.

Gorus E, De Raedt R, Lambert M, Lemper JC, Mets T. 2006.

Attentional processes discriminate between patients with mild

Alzheimer’s disease and cognitively healthy elderly. Interna-

tional Psychogeriatrics 18:539�549.

Hartlage S, Alloy LB, Vazquez C, Dykman B. 1993. Automatic

and effortful processing in depression. Psychol Bull 113:247�278.

Hausmann A, Pascual-Leone A, Kemmler G, et al. 2004. No

deterioration of cognitive performance in an aggressive uni-

lateral and bilateral antidepressant rTMS add-on trial. J Clin

Psychiatry 65:772�782.

Kimbrell TA, Little JT, Dunn RT, et al. 1999. Frequency

dependence of antidepressant response to left prefrontal

repetitive transcranial magnetic stimulation (rTMS) as a

function of baseline cerebral glucose metabolism. Biol Psychia-

try 46:1603�1613.

Kozel F, George M. 2002. Meta-analysis of left prefrontal

repetitive transcranial magnetic stimulation (rTMS) to treat

depression. J Psychiatr Pract 8:270�275.

Loo C, Mitchell P, Sachdev P, McDarmont B, Parker G,

Gandevia S. 1999. Double-blind controlled investigation of

transcranial magnetic stimulation for the treatment of resistant

major depression. Am J Psychiatry 156:946�948.

MacDonald AW, Cohen JD, Stenger VA, Carter CS. 2000.

Dissociating the role of the dorsolateral prefrontal and anterior

cingulate cortex in cognitive control. Science 288:1835�1841.

Martin JLR, Barbanoj MJ, Schlaepfer TE, Thompson E, Perez V,

Kulisevsky J. 2003. Repetitive transcranial magnetic stimula-

tion for the treatment of depression � Systematic review and

meta-analysis. Br J Psychiatry 182:480�491.

Mayberg HS, Liotti M, Brannan SK, et al. 1999. Reciprocal

limbic-cortical function and negative mood: Converging PET

findings in depression and normal sadness. Am J Psychiatry

156:675�682.

Moller AL, Hjaltason O, Ivarsson O, Stefansson SB. 2006. The

effects of repetitive transcranial magnetic stimulation on

depressive symptoms and the P-300 event-related potential.

Nordic J Psychiatry 60:282�285.

Moser DJ, Jorge RE, Manes F, Paradiso S, Benjamin MJ,

Robinson RG. 2002. Improved executive functioning following

repetitive transcranial magnetic stimulation. Neurology

58:1288.

O’Connor MG, Jerskey BA, Robertson EM, Brenninkmeyer C,

Ozdemir E, Leone AP. 2005. The effects of repetitive tran-

scranial magnetic stimulation (rTMS) on procedural memory

and dysphoric mood in patients with major depressive disorder.

Cogn Behav Neurol 18:223�227.

Padberg F, Zwanzger P, Thoma H, et al. 1999. Repetitive

transcranial magnetic stimulation (rTMS) in pharmacother-

apy-refractory major depression: comparative study of fast,

slow and sham rTMSI. Psychiatry Res 88:163�171.

Padberg F, Zwanzger P, Keck ME, et al. 2002. Repetitive

transcranial magnetic stimulation (rTMS) in major depression:

Relation between efficacy and stimulation intensity. Neuropsy-

chopharmacology 27:638�645.

Paus T, Barrett J. 2004. Transcranial magnetic stimulation (TMS)

of the human frontal cortex: implications for repetitive TMS

treatment of depression. J Psychiatry Neurosci 29:268�279.

Picton TW. 1992. The P300 wave of the human event-related

potentional. J Clin Neurophysiol 9:456�479.

Pogarell O, Koch W, Popperl G, et al. 2006. Striatal dopamine

release after prefrontal repetitive transcranial magnetic stimula-

tion in major depression: Preliminary results of a dynamic [I-

123] IBZM SPECT study. J Psychiatr Res 40:307�314.

Sheehan DV, Lecrubier Y, Sheehan KH, et al. 1998. The Mini-

International Neuropsychiatric Interview (M.I.N.I.): the devel-

opment and validation of a structured diagnostic psychiatric

interview for DSM-IV and ICD-10. J Clin Psychiatr 59(Suppl

20):22�33;quiz 34�57. M.I.N.I. Plus (Nederlandstalige vertal-

ing, Van Vliet, Leroy & Van Megen, 2000).

Stuss DT, Murphy KJ, Binns MA, Alexander MP. 2003. Staying

on the job: The frontal lobes control individual performance

variability. Brain 126:2363�2380.

Sohn MH, Ursu S, Anderson JR, Stenger VA, Carter CS. 2000.

The role of prefrontal cortex and posterior parietal cortex in

task switching. Proc Natl Acad Sci USA 97:13448�13453.

Sommer W, Matt J. 1990. Awareness of P300-related cognitive

processes � A signal detection approach psychophysiology.

Psychophysiology 27:575�585.

Speer AM, Kimbrell TA, Wassermann EM, et al. 2000. Opposite

effects of high and low frequency rTMS on regional brain

activity in depressed patients. Biol Psychiatry 48:1133�1141.

Taylor JG, Fragopanagos NF. 2005. The interaction of attention

and emotion. Neural Networks 18:353�369.

Triggs WJ, McCoy KJM, Greer R, et al. 1999. Effects of left

frontal transcranial magnetic stimulation on depressed mood,

cognition, and corticomotor threshold. Biol Psychiatry

45:1440�1446.

Vanderhasselt MA, De Raedt R, Baeken C, Leyman L, D’haenen

H. 2006a. The influence of rTMS over the left dorsolateral

prefrontal cortex on Stroop task performance. Exp Brain Res

18:1�4.

Vanderhasselt MA, De Raedt R, Baeken C, Leyman L, D’haenen

H. 2006b. The influence of rTMS over the right dorsolateral

rTMS and depression: influence on cognition 41

prefrontal cortex on intentional set switching. Exp Brain Res

172:561�565.

Van Strien. 2001. Handvoorkeur en taaldominantie. Neuropraxis

2:10�15.

Wagner M, Rihs TA, Mosimann UP, Fisch HU, Schlaepfer TE.

2006. Repetitive transcranial magnetic stimulation of the

dorsolateral prefrontal cortex affects divided attention imme-

diately after cessation of stimulation. J Psychiatr Res 40:315�321.

Wassermann EM. 1998. Risk and safety of repetitive transcranial

magnetic stimulation. In: Report and suggested guidelines

from the International Workshop on the Safety of Repetitive

Transcranial Magnetic Stimulation. Clin Neurophysiol 108

(Suppl 1):1�16.

42 M.-A. Vanderhasselt et al.

ORIGINAL INVESTIGATION

Intramuscular olanzapine versus short-acting typical intramuscularantipsychotics: Comparison of real-life effectiveness in the treatmentof agitation

DAVID J. CASTLE1, TUDOR UDRISTOIU2, CHANG YOON KIM3, ANDREA SAROSI4,

VLADIMIR PIDRMAN5, A. NASSER OMAR6, JUAN IGNACIO ROSALES7,

YUVAL MELAMED8, TURGUT ISIK9, JAMIE KARAGIANIS10 & TAMAS TREUER11

1St. Vincent’s Hospital, The University of Melbourne, Fitzroy, Australia, 2Department of Psychiatry, University of Medicine

and Pharmacy, Craiova, Romania, 3Department of Psychiatry, College of Medicine, University of Ulsan, Asan Medical

Center, Seoul, Korea, 4Kutvolgyi Clinical Center, Department of Clinical and Theoretical Mental Health, Semmelweis

University, Budapest, Hungary, 5Department of Psychiatry, College of Medicine, Palacky University, Olomouc, Czech

Republic, 6Faculty of Medicine, Ain Shams University, Ard El-Golf-Masr el gedeeda, Cairo, Egypt, 7Clinica San Rafael,

Santa Ursula, Mexico, 8Lev-Hasharon Mental Health Center, Netanya, affiliated to Sackler Faculty of Medicine, Tel-Aviv

University, Israel, 9Manisa Ruh Sagligi ve Hastaliklari Hastanesi, Manisa, Turkey, 10Eli Lilly Canada Inc., Toronto,

Canada, Memorial University of Newfoundland, Canada, and 11Lilly Hungaria Kft, Budapest, Hungary

AbstractObjective. To compare the effectiveness of intramuscular (IM) olanzapine and typical IM antipsychotics in naturalisticallytreated acutely agitated patients with schizophrenia or acute mania. Methods. During the acute phase, 2011 inpatients(including emergency settings) were assessed at 2, 24 and 72 h, and 7 days following initial injection and on oralantipsychotic transition. Mean change in agitation was assessed via Positive and Negative Symptom Scale-ExcitedComponent (PANSS-EC) and Clinical Global Impressions-Severity (CGI-S) scores. Response (]40% reduction inbaseline PANSS-EC score) was analysed using logistic regression. Results. Significantly greater decreases in PANSS-EC andCGI-S scores were observed in patients receiving IM olanzapine (n�1294) as their first injection compared with patientsreceiving other IM antipsychotics (n�717) (PB0.05; 2 h: effect size 0.1); IM haloperidol treatment (all assessments, PB0.05); and IM zuclopenthixol treatment (2 h, PB0.001). Higher response rates were observed with IM olanzapinecompared with other IM antipsychotics at 24 and 72 h, and 7 days (PB0.05). IM olanzapine was associated with fewerextrapyramidal side effects compared with other assessed IM antipsychotics. Conclusions. IM olanzapine provided somewhatmore effective control of acute agitation than other assessed IM antipsychotics.

Key words: Agitation, bipolar disorder, intramuscular antipsychotics, olanzapine, schizophrenia

Introduction

Agitation is a common acute symptom of schizo-

phrenia and bipolar mania (Meehan et al. 2001;

Wright et al. 2001; Breier et al. 2002). Severe

agitation may be accompanied by verbal and physi-

cal aggression, posing a danger to patients them-

selves and to others (Binder and McNiel 1999).

Thus, agitated patients require urgent treatment

which facilitates rapid alleviation of symptoms.

This imperative, along with the fact that the patient

might refuse oral medication, often requires the use

of parenteral medications. Typical antipsychotics

and benzodiazepines are commonly used alone or

in combination to relieve agitation, although treat-

ment with typical antipsychotics has been associated

with the development of extrapyramidal symptoms

(EPS), including acute dystonia and akathisia

(Glazer 2000); while benzodiazepine treatment can

result in excessive sedation (Bodkin 1990). Also the

product monograph of lorazepam, a medication

commonly used for managing agitation, contains a

warning recommending that it not be used in

psychotic reactions. Atypical antipsychotic agents

Correspondence: Dr Tamas Treuer, Lilly Hungaria Kft, H-1075 Budapest, Madach Imre utca 13�14, Hungary. Tel: �36 1 328 5127. Fax:

�36 1 328 5103. E-mail: treuert@lilly.com

The World Journal of Biological Psychiatry, 2009; 10(1): 43�53

(Received 9 July 2008; accepted 7 December 2008)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970802688051

have a more favorable EPS safety profile compared

with typical antipsychotics (Glazer 2000), with oral

formulations used widely for long-term maintenance

therapy. Hence, acute treatment with IM atypical

antipsychotics may facilitate the transition to oral

atypical maintenance therapy (Bartko 2006).

IM olanzapine has been shown to be safe and

effective in the acute management of agitation in

numerous randomized clinical trials (RCTs)

(Meehan et al. 2001; Breier et al. 2002; Mohr et al.

2005; Citrome 2007), typically reaching a mean

maximum plasma concentration 15�45 min after

injection, compared with 3�6 h for an equivalent

oral dose (Jones 2001; Wright et al. 2001). In patients

diagnosed with schizophrenia, IM olanzapine pro-

vided more rapid onset of action and fewer extrapyr-

amidal symptoms compared with IM haloperidol

(Wright et al. 2001; Breier et al. 2002). Furthermore,

alleviation of agitation was sustained following transi-

tion from IM olanzapine to oral antipsychotic treat-

ment (Wright et al. 2003). In patients with bipolar

mania, IM olanzapine provided significantly greater

improvements in agitation severity compared with IM

lorazepam for acute treatment (Meehan et al. 2001).

These studies suggest IM olanzapine may provide

more effective symptom relief compared with some

other available IM treatments for acute agitation.

Highly agitated patients are often excluded from

randomized clinical trials, as their capacity to

provide informed consent prior to receiving study

medication is likely to be limited. Therefore, the

efficacy of agents observed to reduce agitation

severity in RCTs may not necessarily extend to

severely ill patients in real-life inpatient or emer-

gency settings, who are likely to benefit most from a

rapid-acting agent (Preval et al. 2005; Centorrino

et al. 2007). Observational studies typically have

fewer inclusion and exclusion criteria than RCTs,

and thus may more closely represent treatment

effectiveness in usual clinical practice. Small scale

observational studies have found IM olanzapine to

be effective in the treatment of highly agitated

patients (Pascual et al. 2006; San et al. 2006;

Centorrino et al. 2007). However, IM olanzapine

must be evaluated in larger and more diverse

populations in order to properly assess its effective-

ness in the actual treatment of patients with acute

agitation.

The primary objective of this study was to

compare the effectiveness of IM olanzapine with

typical IM antipsychotic drugs as measured by mean

change in PANSS-EC scores at 2 h post-injection in

patients with acute agitation. Further effectiveness

and tolerability measures were also compared

between treatment groups at 2, 24 and 72 h, and

7 days post initial injection, and at time of transition

to oral treatment.

Methods

Study population

A total of 2011 patients aged ]18 years were

enrolled from sites across 15 countries: Australia,

Bulgaria, Canada, Czech Republic, Egypt, Hungary,

Israel, Korea, Mexico, Romania, Russia, Saudi

Arabia, Slovakia, Slovenia and Turkey. Patients

were considered eligible for the study if they had a

diagnosis that met an indication for olanzapine in

their local region (acute mania and/or schizophre-

nia); were inpatients (including emergency room

settings) requiring treatment with at least one

injection of a short acting IM antipsychotic; and

were not concurrently participating in a controlled

clinical trial. This study was approved and reviewed

by the Institutional Review Board at each site, and

was conducted in accordance with the ethical

principles that have their origin in the Declaration

of Helsinki and that are consistent with good clinical

practices. Written consent was obtained from all

patients when they were well enough, in their mental

state, to provide consent in an informed manner. In

the event that a patient refused consent to release of

data, any information that had been collected was

destroyed. Recruitment was conducted between

March 2004 and December 2005, with the final

patient visit conducted in June 2006.

Study design

This prospective, observational, non-interventional

study included two phases. In the acute phase,

following the initial study drug injection, patient

observations were reported at 2, 24 and 72 h, and 7

days, and at the time of transition to oral anti-

psychotic. In a subset of countries (Czech Republic,

Egypt, Romania, Russia, Saudi Arabia, Slovakia,

Turkey), patients entered the extension phase of the

study, where they were followed up to 6 months after

transition to oral antipsychotic medication. Here we

report only the acute phase results.

As this was a non-interventional observational

study, all treatment decisions, including initiation,

change or cessation of treatment, were made solely

at the discretion of the patient and the treating

psychiatrist, independently of the study sponsor.

Treatments were prescribed as part of the usual

standard of care and were not provided by the study

sponsor. Participation in this study in no way

influenced payment or reimbursement for any treat-

ment received by patients during the study.

44 D.J. Castle et al.

Antipsychotic polypharmacy and concomitant

medications were permitted throughout the study.

Treatment groups

For the purpose of statistical comparison, patients

were assigned to either IM olanzapine or other IM

antipsychotics groups based on the first study

injection administered regardless of subsequent

medications. The patients may have received sub-

sequent IM injection(s) with the same or different

drug(s) than that received initially.

Outcome measures

Patient assessments included the documentation of

demographic information and baseline psychiatric

history, medication and clinical status. Treatment

effectiveness was measured by the change from

baseline in Positive and Negative Symptom Scale-

Excited Component (PANSS-EC) (Kay and Sevy

1990) and Clinical Global Impressions-Severity

(CGI-S) (Guy 1976) scores at all time points of

assessment and at time to transition to oral medica-

tion. The primary efficacy measure was change in

PANSS-EC score at 2 h after the first study

injection.

Response was defined as ]40% reduction in

baseline PANSS-EC score. Relapse was determined

by requirement of additional IM injections following

initial response; and by investigator opinion for all

patients and for the responder population only. The

proportion of patients who required physical re-

straints at any time during the acute phase was also

recorded.

Tolerability measures

Tolerability was assessed by treatment-related fac-

tors affecting outcomes and the use of concomitant

medications to control these factors. Adverse events

were recorded at first assessment (including the

week prior to first assessment), and at all subsequent

assessments including the visit of transition to oral

antipsychotics. Concomitant psychotropic medica-

tion (including oral antipsychotic) use was reported

during the acute phase prior to transition to oral

antipsychotics including the week prior to the first

study injection.

Statistical analysis

An a priori sample size calculation was performed

assuming a 1-point difference between treatment

groups on change from baseline in PANSS-EC score

at 2 h with a standard deviation of 7.7. Taking a two-

sided 5% level of significance and 90% power

required a sample size of 1247 patients per treat-

ment group (2494 patients in total).

Treatment cohorts based on initial study injection

of IM olanzapine or other short-acting IM antipsy-

chotics were the primary comparison of interest. We

also report the three main treatment cohorts (IM

olanzapine, IM haloperidol, and IM zuclopenthixol

acetate (hereafter referred to as IM zuclopenthixol))

for patients given only one IM drug as initial study

treatment.

Pre-specified potential baseline confounders of

age, gender, region, diagnosis (acute mania or schizo-

phrenia), level of agitation (PANSS-EC ]20), pre-

sence of EPS, and recent use of antipsychotic

medication (oral or IM in past week, depot in past

30 days) were identified. All post-baseline compar-

isons using linear analysis of covariance (ANCOVA)

and logistic regression models included covariates for

treatment group, baseline severity and the pre-speci-

fied potential confounders. The Level of Agitation

subgroups were examined using logistic regression

with a treatment-by-level of agitation interaction

term.

Longitudinal analyses of change in continuous

effectiveness measures (PANSS-EC, CGI-S) were

conducted using restricted maximum likelihood

(REML)-based mixed-effects model. Independent

fixed effects of treatment group, visit, treatment-

by-visit interaction, baseline severity and the pre-

specified potential confounders were included in the

model. Longitudinal analyses of response were

conducted with logistic regression using generalized

estimating equations with covariates for treatment

group, baseline PANSS-EC and the pre-specified

potential confounders. All longitudinal analyses were

conducted using a pre-specified unstructured covar-

iance matrix.

Baseline comparisons between groups were con-

ducted using t-tests (assuming equal variance) for

continuous measures and the chi-square test or

Fisher’s exact test for discrete variables. Results are

reported as least squares means (LS means) or

adjusted odds ratios (OR), with accompanying

95% confidence interval, as appropriate.

Results

A total of 2011 patients entered the acute phase of

the study, of which 1294 patients (64.3%) received

IM olanzapine and 717 patients (35.7%) received

other IM antipsychotics as their first acute phase

injection. Figure 1 provides a summary of the patient

disposition during the study. Completion rates for

the acute phase were 98.0% (1268 patients) in the

IM olanzapine group and 96.8% (694 patients) in

the other IM antipsychotics group, with loss to

IM olanzapine vs typical antipsychotics 45

follow-up being the most common reason for

discontinuation (1.1%, n�23).

Table I summarizes the demographic and clinical

characteristics of patients at entry to the acute phase

of the study. Similar proportions of women and men

entered the acute phase, with an overall mean (SD)

age of 36.3 (11.65) (range 18�80) years. The

majority of patients had a diagnosis of schizophrenia

(70.1%), and approximately half of the patients were

considered free of concomitant antipsychotics,

meaning that they had not received oral or IM

antipsychotic medication in the past 7 days, and had

not received depot antipsychotic medication in the

past 30 days. Most patients were highly agitated

(79.9%) and of marked-to-severe illness at baseline

(overall CGI-S mean9SD�5.490.91). Patients

receiving IM olanzapine had significantly lower

PANSS-EC scores at baseline compared with pa-

tients receiving other IM antipsychotics (P�0.003),

although CGI-S scores were similar between treat-

ment groups. Fewer patients with high agitation

levels received IM olanzapine compared with other

IM antipsychotics (PB0.001), and women were

more likely to receive IM olanzapine rather than

Figure 1. Patient disposition.

46 D.J. Castle et al.

other IM antipsychotics (P�0.048). Regional dif-

ferences were also observed in the distribution of

patients according to treatment group (P�0.003).

All post-baseline comparisons between treatment

groups were adjusted for pre-specified potential

baseline confounders, including the observed differ-

ences in gender, region, and agitation level at base-

line.

Overall, mean symptom severity improved at all

time points of assessment (2, 24 and 72 h and 7 days)

during the acute phase for IM olanzapine-treated

cohort and those receiving other IM antipsychotics,

as measured by reductions in PANSS-EC and CGI-S

scores from baseline (Figure 2a and 3a). Patients

treated with IM olanzapine experienced significantly

greater symptomatic improvement compared with

patients receiving other IM antipsychotics at all time

points during the acute phase (all PB0.05). Two

hours post-injection, PANSS-EC mean (95% CI)

scores decreased by �6.53 (�7.03, �6.02) for IM

olanzapine-treated patients and �5.71 (�6.27,

�5.15) for patients receiving other IM antipsychotics

(an effect size of 0.1 (Cohen’s d)) (Cohen 1969);

CGI-S scores decreased by �0.67 (�0.75, �0.59)

and �0.56 (�0.65, �0.47), respectively. IM olan-

zapine-treated patients also experienced significantly

greater improvement in symptom severity at time of

transition to oral antipsychotic medication compared

with patients receiving other IM antipsychotics

(PANSS-EC: LS mean [95% CI] difference��0.73 [�1.29 to �0.18], P�0.010; CGI-S:

�0.21 [�0.32 to �0.11], PB0.001). IM olanzapine

was also associated with a significantly shorter time to

transition to oral antipsychotic medications com-

pared with other IM antipsychotic medications (P�0.009; median time in h interquartile range: 46.5

[21.1 to 97.0] and 48.0 [16.3 to 144.0], respectively).

Overall, there were significant differences in symp-

tom severity between the three most commonly used

monotherapy treatment groups (IM olanzapine, IM

Figure 2. Longitudinal analysis of mean change in PANSS-EC

score during acute phase comparing (a) patients receiving IM

olanzapine versus other IM antipsychotics and (b) patients

receiving initial study treatment of IM olanzapine (O), IM

zuclopenthixol (Z) or IM haloperidol (H) only.

Table I. Demographic and clinical characteristics at study entry.

Characteristic

IM olanzapine

(N�1294)

Other IM

antipsychotics (N�717) All (N�2011) P value

Mean age, years (SD) 36.1 (11.7) 36.7 (11.5) 36.3 (11.7) 0.268

Gender, number of women (%) 655 (50.6) 330 (46.0) 985 (49.0) 0.048

Region, n (%) 0.003

Africa and Middle East 332 (25.7) 176 (24.5) 508 (25.3)

Asia 80 (6.2) 73 (10.2) 153 (7.6)

Australia and Canada 251 (19.4) 141 (19.7) 392 (19.5)

Central and Eastern Europe 506 (39.1) 241 (33.6) 747 (37.1)

Latin America 125 (9.7) 86 (12.0) 211 (10.5)

Diagnosis, n (%) 0.705

Schizophrenia 911 (70.4) 499 (69.6) 1410 (70.1)

Mania, acute 383 (29.6) 218 (30.4) 601 (29.9)

High agitation level*, n (%) 996 (77.0) 609 (84.9) 1605 (79.9) B0.001

Extrapyramidal side effects, n (%) 89 (6.9) 58 (8.1) 147 (7.3) 0.318

Free of concomitant antipsychotics,$ n (%) 614 (47.4) 344 (48.0) 958 (47.6) 0.820

PANSS-EC score, mean (SD) 24.3 (6.0) 25.1 (5.6) 24.6 (5.9) 0.003

CGI-S score, mean (SD) 5.4 (0.91) 5.4 (0.90) 5.4 (0.91) 1.000

*High level of agitation was defined as PANSS-EC score ]20.

$No IM or oral antipsychotics in the past week and no depot antipsychotics in past 30 days.

PANSS-EC, Positive and Negative Symptom Scale-Excited Component; CGI-S, Clinical Global Impressions-Severity.

IM olanzapine vs typical antipsychotics 47

zuclopenthixol, or IM haloperidol monotherapy as

initial study treatment) in CGI-S scores at each time

point (all PB0.05), and in PANSS-EC scores for all

time points (PB0.05) except 7 days (P�0.112).

Patients receiving IM olanzapine demonstrated a

significantly greater improvement in symptom sever-

ity compared with IM haloperidol-treated patients at

all time points of assessment during the acute phase

(all PB0.05; Figure 2b and 3b), and at time

to transition (PANSS-EC: LS mean [95% CI]

difference�1.0 [0.4 to 1.7], P�0.002). IM olanza-

pine-treated patients also demonstrated significantly

greater improvement in symptom severity compared

with IM zuclopenthixol-treated patients at the 2-h

assessment (PB0.001 for both PANSS-EC and

CGI-S), although no statistically significant differ-

ences were observed at any other time point. Patients

treated with IM zuclopenthixol showed greater im-

provements in symptom severity compared with IM

haloperidol-treated patients at 2 h, as measured

by PANSS-EC (P�0.018) and CGI-S scores (P�0.010), and at 7 days via CGI-S scores only

(P�0.016).

There were no statistically significant differences

in the effectiveness of IM olanzapine and other IM

antipsychotics in the treatment of highly agitated

(PANSS-EC]20) (n�996 for IM olanzapine; n�609 for other IM antipsychotics) and less agitated

(PANSS-ECB20) patients (n�297 for IM olanza-

pine; n�108 for other IM antipsychotics) at 2 h

after initial injection (P�0.179). Similarly, no

statistically significant differences were observed in

the treatment effectiveness of IM olanzapine com-

pared with other IM antipsychotics between patients

with acute mania or schizophrenia at the 2 h

assessment (P�0.120).

As shown in Figure 4, a greater proportion of

patients receiving IM olanzapine met the criteria for

treatment response (]40% decrease in PANSS-EC)

compared with patients receiving other IM antipsy-

chotics during the acute phase, with statistically

significant differences at 24 and 72 h and 7 days

(all PB0.05). Of the patients who responded to

treatment, no statistically significant difference was

observed in the proportion of patients who relapsed

(as defined by the requirement for additional IM

injections) between IM olanzapine and other IM

antipsychotics at any time point during the acute

phase [22.1 vs. 24.4%; OR (95% CI)�0.891

(0.683, 1.162), P�0.393; NNT (95% CI)�44

(�42 to � to 15)]. However, significantly fewer

patients receiving IM olanzapine relapsed according

to investigator assessment compared with patients

receiving other IM antipsychotics [20.7 vs. 30.0%;

OR (95% CI)�0.602 (0.480, 0.754), P�B0.001;

NNT (95% CI)�11 (8, 20)], and also when relapse

was defined by both investigator assessment and the

requirement for additional IM medication [7.0 vs.

10.9%; OR (95% CI)�0.659 (0.444, 0.978), P�0.039, NNT (95% CI)�26 (15, 169)]. Similar

proportions of patients treated with IM olanzapine,

IM zuclopenthixol and IM haloperidol responded or

relapsed during the acute phase, although IM

olanzapine was associated with a somewhat greater

response compared to IM haloperidol (71.3 vs.

66.1%, OR [95% CI]�0.89 [0.70, 1.13], P�.324,

NNT [95% CI]�19 [10, 573]) and a somewhat

higher relapse rate compared with IM zuclopenth-

xiol (22.1 vs. 12.0%, OR [95% CI]�1.65 [0.81,

3.35], P�0.170, NNH [95% CI]�9 [5, 39]).

As per study inclusion criteria, all participating

patients received at least one injection of a short

acting IM antipsychotic during the acute study

Figure 3. Longitudinal analysis of mean change in CGI-S score

during acute phase comparing (a) patients receiving IM olanza-

pine versus other IM antipsychotics and (b) patients receiving

initial study treatment of IM olanzapine (O), IM zuclopenthixol

(Z) or IM haloperidol (H) only.

Figure 4. Percentage of patients who met the criteria for treat-

ment response (]40% reduction in baseline PANSS-EC score)

during the acute phase.

48 D.J. Castle et al.

phase; a low proportion of patients received only one

injection (IM olanzapine 33.2%, other IM antipsy-

chotics 36.7%). Of the 717 patients receiving other

IM antipsychotics, 436 patients (21.7%) received

IM haloperidol and 107 patients (5.3%) received IM

zuclopenthixol. The mean (SD) dose was 10.1

(1.97) mg for IM olanzapine, 9.1 (10.52) mg for

IM haloperidol and 93.3 (69.46) mg for IM zuclo-

penthixol. The remaining 174 patients (8.7%) were

treated with antipsychotic polypharmacy (94 pa-

tients, 4.7%) or other single IM antipsychotics (80

patients, 4.0%). The most commonly used other

single IM antipsychotics were levomepromazine (28

patients, 1.4%) and droperidol (20 patients, 1.0%).

The most common combinations of IM antipsycho-

tics were levomepromazine plus haloperidol (56

patients, 2.8%) and chlorpromazine plus haloperidol

(33 patients, 1.6%).

Adverse events (AEs) were observed in all treat-

ment groups. AEs were more commonly observed

for patients in the IM zuclopenthixol monotherapy

group (65.4% of patients), than for those receiving

IM olanzapine or IM haloperidol monotherapy

(34.4 and 41.7% respectively, PB0.001). Signifi-

cant differences were observed in the prevalence of

specific AEs across the treatment groups, particu-

larly regarding extrapyramidal side effects with

haloperidol and zuclopenthixol (Table II). There

were no deaths during the study. Further safety data

for this study will be reported elsewhere.

The majority of patients (62.7%) received at least

one concomitant oral antipsychotic medication dur-

ing the study phase (Table III). Patients initially

treated with IM olanzapine more commonly re-

ceived concomitant oral olanzapine (37.5%),

whereas patients receiving other IM antipsychotics

predominantly received concomitant oral haloper-

idol (16.9%) or oral olanzapine (16.3%). Compared

with IM olanzapine-treated patients, a higher pro-

portion of patients receiving other IM antipsychotics

were treated with concomitant depot antipsycho-

tics (5.0 vs. 1.5%). The majority of patients were

also receiving concomitant psychotropic medications

other than antipsychotics during the acute phase

(70.5%), as shown in Table IV. A higher proportion

of patients treated with other IM antipsychotics were

receiving concomitant psychotropic medications

[PB0.001; NNH (95% CI)�6 (5, 8)]. These

patients were significantly more likely to be receiving

anticholinergic [PB0.001; NNH (95% CI)�3 (3,

4)] or anxiolytic/hypnotic medication [P�0.023;

NNH (95% CI)�25 (�141 to � to 11)] compared

with IM olanzapine-treated patients. Many patients

were sufficiently agitated to require physical restraint

during the acute phase of the study. Although there

were fewer highly agitated patients in the IM

olanzapine-treated group, the differences between

groups in the proportion of patients requiring

restraint were not significant, with 31.6% of IM

olanzapine-treated patients and 37.6% of patients

receiving other IM antipsychotics requiring restraint.

Discussion

This prospective, transregional, observational study

demonstrates the effectiveness and tolerability of IM

olanzapine compared with other assessed IM anti-

psychotics in the naturalistic treatment of acute

agitation in patients with schizophrenia and bipolar

mania. To our knowledge, this represents the first

large scale multinational observational study of the

‘‘real-life’’ effectiveness and tolerability of IM olan-

zapine in the treatment of agitation. Few naturalistic

studies assessing the effectiveness of IM olanzapine

have been conducted, most with enrolment up to

approximately 300 patients (Preval et al. 2005;

Pascual et al. 2006; Bushe et al. 2007; Centorrino

et al. 2007; Daniel et al. 2007). Randomized

controlled clinical trials have reported rapid and

effective symptom reduction with IM olanzapine in

the treatment of agitated patients with schizophrenia

or bipolar mania (Meehan et al. 2001; Wright et al.

2001; Breier et al. 2002). However, the majority of

RCTs have excluded highly agitated patients, with

mean baseline PANSS-EC scores between 17 and 19

(Meehan et al. 2001; Wright et al. 2001; Breier et al.

Table II. Most frequently (overall frequency ]4%) reported adverse events during the acute phase (including the week prior to first

study injection).

Adverse event

IM

olanzapine (O)

(N�1294)

IM

haloperidol (H)

(N�436)

IM

zuclopenthixol (Z)

(N�107)

Overall

(N�1837)

P value

O vs H O vs Z Z vs H

Patients with ]1 event, n (%) 445 (34.4) 182 (41.7) 70 (65.4) 697 (37.9) 0.007 B0.001 B0.001

Akathisia, n (%) 57 (4.4) 31 ( 7.1) 22 (20.6) 110 (6.0) 0.032 B0.001 B0.001

Disturbance in attention, n (%) 90 (7.0) 32 (7.3) 8 (7.5) 130 (7.1) 0.829 0.843 1.000

Dystonia, n (%) 30 (2.3) 38 (8.7) 8 (7.5) 76 (4.1) B0.001 0.006 0.847

Parkinsonism, n (%) 75 (5.8) 50 (11.5) 10 (9.3) 135 (7.3) B0.001 0.140 0.609

Sedation, n (%) 301 (23.3) 103 (23.6) 45 (42.1) 449 (24.4) 0.896 B0.001 B0.001

IM olanzapine vs typical antipsychotics 49

2002), and mean baseline CGI scores of 4.6,

indicative of moderate illness severity (Meehan

et al. 2001). By comparison, recent observational

studies assessing acutely agitated patients in inpati-

ent and emergency settings included patients with

overall higher agitation levels (mean baseline

PANSS-EC scores�26.5 and 29) and greater illness

severity (mean baseline CGI scores�5.5 and 6.1)

compared with RCTs (San et al. 2006; Centorrino

et al. 2007). Similarly, the majority of patients in the

present study were rated as highly agitated (mean

PANSS-EC score�24.6) and marked-to-severely ill

(mean CGI score�5.4) at baseline. Thus, patients

enrolled in observational or naturalistic studies may

closely represent patients presenting in usual prac-

tice settings (Centorrino et al. 2007).

In the present study, patients treated with IM

olanzapine experienced significantly greater improve-

ments in symptom severity compared with those

treated with other IM antipsychotics at the 2-h

post-injection assessment (mean PANSS-EC score

change��6.53; mean CGI-S score change��0.67). Similar results were observed when IM

olanzapine-treated patients were compared with IM

haloperidol- or IM zuclopenthixol-treated patients at

2 h. Previous RCTs reported greater reductions in

PANSS-EC scores (mean change��7.7 and �9.4)

at 2 h after injection in patients treated with IM

olanzapine than those receiving IM haloperidol

(Wright et al. 2001; Breier et al. 2002), although

these studies enrolled patients with overall lower

agitation levels. In a previous observational study,

Centorrino and colleagues (2007) reported signifi-

cant symptomatic improvement in clinically agitated

patients (with diagnosis of schizophrenia or bipolar

mania) at two hours after the IM olanzapine injection

(mean PANSS-EC score change��9.9). San and

colleagues (2006) also reported a reduction in mean

PANSS-EC scores (�9.6) at 2 h in agitated patients

with schizophrenia or bipolar mania treated with IM

olanzapine. Both of these studies observed greater

improvements in symptom severity in comparison to

the present study; a pattern which is also reflected in

the response rates for each study.

In the present study, IM olanzapine was asso-

ciated with a greater response rate compared to

other IM antipsychotics at the majority of assess-

ments. Despite this, the responses rates for IM

olanzapine-treated patients were lower in the present

study (20%, 2 h; 34%, 24 h) when compared with

the response rates (73�80%) observed in previous

RCTs (Wright et al. 2001; Breier et al. 2002).

However, the response rates in the present study

are closer to those observed in previous naturalistic

studies (48�49%), which may also be explained

through the recruitment of patients with greater

agitation severity in these observational studies

(San et al. 2006; Centorrino et al. 2007), and/or

the inclusion of patients who may have been

excluded from RCTs for other reasons, such as

substance abuse.

In the study reported here, approximately 19% of

patients receiving IM olanzapine and 26% of those

receiving other IM antipsychotics required five or

more injections. This may be in part because the

patients receiving IM olanzapine were less agitated

at baseline. A low proportion of patients required

only one injection of IM antipsychotic in each group

during the acute phase (33%, IM olanzapine; 37%,

other IM antipsychotics), in a similar finding to the

Table III. Concomitant oral antipsychotic medication taken during the acute phase (by ]5% of all patients).

IM olanzapine

(N�1294)

Other IM antipsychotics

(N�717)

Overall

(N�2011)

Patients with ]1 medication, n (%) 760 (62.7) 422 (62.7) 1182 (62.7)

Olanzapine, n (%) 454 (37.5) 110 (16.3) 564 (29.9)

Haloperidol, n (%) 95 (7.8) 114 (16.9) 209 (11.1)

Risperidone, n (%) 114 (9.4) 87 (12.9) 201 (10.7)

Quetiapine, n (%) 80 (6.6) 73 (10.8) 153 (8.1)

Chlorpromazine, n (%) 58 (4.8) 71 (10.5) 129 (6.8)

Table IV. Concomitant psychotropic medication taken during the acute phase.

IM olanzapine

(N�1294)

Other IM

antipsychotics

(N�717)

Overall

(N�2011)

Odds ratio

(95% CI) P value

Patients with ]1 medication, n (%) 754 (64.6) 565 (80.5) 1319 (70.5) 0.416 (0.331, 0.524) B0.001

Anticholinergics, n (%) 162 (13.9) 298 (42.5) 460 (24.6) 0.190 (0.149, 0.242) B0.001

Antidepressants, n (%) 79 (6.8) 40 (5.7) 119 (6.4) 1.065 (0.710, 1.597) 0.761

Anxiolytics/hypnotics, n (%) 556 (47.6) 362 (51.6) 918 (49.1) 0.794 (0.650, 0.968) 0.023

Mood stabilizers, n (%) 316 (27.1) 186 (26.5) 502 (26.8) 1.105 (0.872, 1.400) 0.410

50 D.J. Castle et al.

26% of patients requiring a single injection in an

RCT assessing patients with bipolar mania (Meehan

et al. 2002). This finding is in contrast to those

reported in other naturalistic studies (93�96%),

which were conducted using patients presenting to

psychiatric emergency services only (San et al. 2006;

Centorrino et al. 2007), and RCTs (76%), which

were conducted using patients with more moderate

agitation levels (Breier et al. 2002). It was suggested

by San and colleagues (2006) that this discrepancy

may be explained through the greater availability of

more specifically trained staff in psychiatric emer-

gency settings in comparison to general inpatient

services, with both settings utilized within the

present study. Given the high level of agitation

severity within the present naturalistic study, as

compared to the lower agitation levels observed in

the controlled circumstances of registration trials

where very little concomitant medication is typically

used, it is important to note that no deaths were

reported during the acute study phase.

There were notable differences in the concomitant

medication profile between treatment groups within

the present study. Overall, fewer patients in the IM

olanzapine group were prescribed concomitant psy-

chotropic medication, irrespective of initial agitation

severity. Patients receiving other IM antipsychotics

were more likely to be receiving comorbid antic-

holinergic or anxiolytic/hypnotic medication com-

pared with IM olanzapine-treated patients. Similar

results were observed in RCTs comparing orally

disintegrating, standard oral or IM olanzapine to

other treatments in the acute management of

severely agitated or aggressive patients (Belgamwar

and Fenton 2005), suggesting that the more favour-

able extrapyramidal symptom and agitation relief

profiles of IM olanzapine compared to other IM

antipsychotics may lessen the need for certain

concomitant medications in the management of

acute agitation.

Overall, the present study had a lower frequency

of concomitant benzodiazepine use (49% anxioly-

tics/hypnotics � data on concomitant use of IM

benzodiazepines were not collected) compared with

a previous observational study (67%) (Centorrino

et al. 2007). Only 4�16% of patients had received

concomitant benzodiazepines in previous RCTs

(Wright et al. 2001; Breier et al. 2002), as con-

comitant medication use is typically limited in these

studies. Centorrino and colleagues (2007) suggested

that their superior response rate may be attributed to

benzodiazepine enhancement of the agitation-redu-

cing effects of IM olanzapine without producing

excessive sedation, although it is important to note

that the olanzapine product label warns against

combining parenteral benzodiazepines with IM

olanzapine. Similarly high response rates have been

observed over various RCTs and observational

studies with markedly different concomitant benzo-

diazepine use profiles, suggesting that other explana-

tions for variations in response rates should be

considered, such as efficacy differences in the main

treatments used. This study revealed some differ-

ences in locally preferred practices of medication

choices, which will be reported elsewhere, which

could potentially explain outcome differences in a

smaller study. Also, medication decisions and ad-

ministration methods for agitated patients may in

some jurisdictions be influenced by local treatment

and reimbursement guidelines, as well as the avail-

ability of medications in the emergency room setting.

This study has some limitations common to most

observation studies, including lack of placebo con-

trol, open-label unrestricted dosing, and unrest-

ricted concomitant medication use. The authors

acknowledge that patients receiving IM olanzapine

were less agitated at baseline than those receiving

other IM antipsychotics. However, as baseline illness

severity was pre-specified in the statistical modelling

and as there is greater potential for improvement

amongst patients with greater illness severity, the

lower level of baseline agitation amongst IM olanza-

pine patients would be not expected to influence the

clinical significance of the results observed for this

medication. All number needed to treat statistics

presented within this analysis are unadjusted. Since

the treatment cohorts were not balanced in terms of

demographic and clinical characteristics at study

entry these NNTs may be biased and as such should

be interpreted with caution. The authors also

acknowledge that the acetate form of IM zuclo-

penthixol has a duration of action of approximately

48 h and typically reaches maximum effect after 8 h;

parameters which are dissimilar to IM olanzapine

(duration of action �30 h, maximum effect after

�30 min) and some other medications used in this

study. IM zuclopenthixol was included in this study

as it continues to be used in the management of

agitation, and is still considered by many clinicians

as a short acting agent, despite the expected delayed

onset of action and longer duration of action

compared to other agents. These characteristics

could be seen as either advantageous or disadvanta-

geous for IM zuclopenthixol within the present

study. The delayed onset of action of IM zuclo-

penthixol did not prevent it from demonstrating

superiority to haloperidol in terms of symptomatic

improvement at 2 h, and IM olanzapine was

associated with a somewhat higher, yet not statisti-

cally significant, relapse rate compared with IM

zuclopenthxiol. However, the authors acknowledge

that the parameters described above may limit the

IM olanzapine vs typical antipsychotics 51

interpretation of the comparative performance of IM

zuclopenthixol within the present study.

Antipsychotic polypharmacy increases the diffi-

culty associated with establishing a definitive link

between medications taken and observed outcomes.

Although all post-baseline comparisons were ad-

justed for potential baseline confounders in this

study, care should be taken in the interpretation of

causality and associations between treatment and

outcome. Additionally, although the majority of

patients were highly agitated, only patients who

were well enough to provide written informed

consent were included in the study. Therefore, the

results of this study may not be generalizable to

patients who have higher levels of agitation than

those reported here. Also, the observed effect size

(Cohen’s d) for the primary comparison of PANSS-

EC score at 2 h is considered small, so physicians

should consider the clinical significance of the

presented findings. However, given the life-threaten-

ing nature of agitation, it could be argued that even a

small advantage could be important in clinical

practice.

In conclusion, IM olanzapine provided more

effective relief of acute agitation than other IM

antipsychotics assessed in this multinational obser-

vational study, and demonstrated distinct benefits

over other assessed IM antipsychotics in terms of

several adverse events. The results of this study

reinforce the relative advantage of IM olanzapine

over the assessed IM typical antipsychotics in the

treatment of acute agitation.

Acknowledgements

We wish to thank all participating investigators and

their patients. We thank Richard Walton for statis-

tical assistance and Renee Granger for medical

writing support. This study was supported by Eli

Lilly and Company, Indianapolis, IN.

Declaration of Interest

This study (F1D-VI-S046) was sponsored by Eli

Lilly and Company, Indianapolis, IN. Jamie Kar-

agianis is a full-time employee and holds shares of

Eli Lilly and Company. Tamas Treuer is an em-

ployee of Eli Lilly and Company. David Castle has

received grant monies from Eli Lilly, Janssen Cilag,

Roche, Allergen, and Bristol-Myers Squibb; has

received Travel Support and Honoraria for talks

and consultancy from Eli Lilly, Bristol-Myers

Squibb, Astra Zeneca, Lundbeck, Janssen Cilag,

Pfizer, Organon, Sanofi-Aventis, and Wyeth; and is

an advisory board member for Lundbeck, Eli Lilly,

Bristol-Myers Squibb, and Astra Zeneca. Tudor

Udristoiu has received an honorarium from Eli Lilly.

References

Bartko G. 2006. New formulations of olanzapine in the treatment

of acute agitation. Neuropsychopharmacol Hung 8:171�178.

Belgamwar RB, Fenton M. 2005. Olanzapine IM or velotab for

acutely disturbed/agitated people with suspected serious mental

illnesses. Cochrane Database Syst Rev 2:CD003729.

Binder RL, McNiel DE. 1999. Emergency psychiatry: contem-

porary practices in managing acutely violent patients in 20

psychiatric emergency rooms. Psychiatr Serv 50:1553�1554.

Bodkin JA. 1990. Emerging uses for high-potency benzodiaze-

pines in psychotic disorders. J Clin Psychiatry 51(Suppl):41�46.

Breier A, Meehan K, Birkett M, David S, Ferchland I, Sutton V,

et al. 2002. A double-blind, placebo-controlled dose-response

comparison of intramuscular olanzapine and haloperidol in the

treatment of acute agitation in schizophrenia. Arch Gen

Psychiatry 59:441�448.

Bushe CJ, Taylor M, Mathew M. 2007. Intramuscular olanzapine

� A UK case series of early cases. Ann Gen Psychiatry 6:11.

Centorrino F, Meyers AL, Ahl J, Cincotta SL, Zun L, Gulliver

AH, et al. 2007. An observational study of the effectiveness and

safety of intramuscular olanzapine in the treatment of acute

agitation in patients with bipolar mania or schizophrenia/

schizoaffective disorder. Hum Psychopharmacol 22:455�462.

Citrome L. 2007. Comparison of intramuscular ziprasidone,

olanzapine, or aripiprazole for agitation: a quantitative review

of efficacy and safety. J Clin Psychiatry 68:1876�1885.

Cohen J. 1969. Statistical power analysis for the behavioral

sciences. New York: Academic Press.

Daniel J, Chamberlain J, Castle D. 2007. The pharmacological

management of behavioural disturbance in psychosis: a natur-

alistic study. Australas Psychiatry 15:380�384.

Glazer WM. 2000. Extrapyramidal side effects, tardive dyskinesia,

and the concept of atypicality. J Clin Psychiatry 61(Suppl

3):16�21.

Guy W. 1976. ECDEU assessment manual for psychopharmacol-

ogy, revised. Bethesda, MD: US Department of Health,

Education, and Welfare.

Jones B, Taylor CC, Meehan K. 2001. The efficacy of a rapid-

acting intramuscular formulation of olanzapine for positive

symptoms. J Clin Psychiatry 62 Suppl 2:22�24.

Kay S, Sevy S. 1990. Pyramidical model of schizophrenia.

Schizophr Bull 16:537�545.

Meehan K, Zhang F, David S, Tohen M, Janicak P, Small J, et al.

2001. A double-blind randomized comparison of the efficacy

and safety of intramuscular injections of olanzapine lorazepam

or placebo in treating acutely agitated patients diagnosed with

bipolar mania. J Clin Psychopharmacol 21:389�397.

Meehan KM, Wang H, David SR, Nisivoccia JR, Jones B, Beasley

CM Jr, et al. 2002. Comparison of rapidly acting intramuscular

olanzapine, lorazepam, and placebo: a double-blind, rando-

mized study in acutely agitated patients with dementia.

Neuropsychopharmacology 26:494�504.

Mohr P, Pecenak J, Svestka J, Swingler D, Treuer T. 2005.

Treatment of acute agitation in psychotic disorders. Neuroen-

docrinol Lett 26:327�335.

Pascual JC, Madre M, Puigdemont D, Oller S, Corripio I, Diaz A,

et al. 2006. A naturalistic study: 100 consecutive episodes of

acute agitation in a psychiatric emergency department. Actas

Esp Psiquiatr 34:239�244.

Preval H, Klotz SG, Southard R, Francis A. 2005. Rapid-acting

IM ziprasidone in a psychiatric emergency service: a natur-

alistic study. Gen Hosp Psychiatry 27:140�144.

52 D.J. Castle et al.

San L, Arranz B, Querejeta I, Barrio S, De la Gandara J, Perez V.

2006. A naturalistic multicenter study of intramuscular olan-

zapine in the treatment of acutely agitated manic or schizo-

phrenic patients. Eur Psychiatry 21:539�543.

Wright P, Birkett M, David S, Meehan K, Ferchland I, Alaka K,

et al. 2001. Double-blind, placebo-controlled comparison of

intramuscular olanzapine and intramuscular haloperidol in the

treatment of acute agitation in schizophrenia. Am J Psychiatry

158:1149�1151.

Wright P, Meehan K, Birkett M, Lindborg SR, Taylor CC, Morris

P, et al. 2003. A comparison of the efficacy and safety of

olanzapine versus haloperidol during transition from intramus-

cular to oral therapy. Clin Ther 25:1420�1428.

IM olanzapine vs typical antipsychotics 53

ORIGINAL INVESTIGATION

Tardive dyskinesia in a patient treated with quetiapine

EMMANOUIL RIZOS, ATHANASSIOS DOUZENIS, ROSSETOS GOURNELLIS,

CHRISTOS CHRISTODOULOU & LEFTERIS P. LYKOURAS

Second Department of Psychiatry, University of Athens, Medical School, ‘‘Attikon’’ General Hospital, Greece

AbstractQuetiapine is an atypical antipsychotic that is believed to have a low D2 binding affinity in striatal and extrastriatal regions.We report the case of a female patient with the diagnosis of schizoaffective disorder (using DSM-IV-TR criteria) whoinitially received amisulpride for 3 months, discontinued gradually because of persistent and distressing extra-pyramidalsymptoms, and who developed tardive dyskinesia 3 months later after the initiation of quetiapine. A trial with ziprasidoneresulted in a further worsening of tardive dyskinesia symptoms. A further trial with aripiprazole, improved her tardivedyskinesia symptoms. Although, it is under consideration the possibility that the improvement could have been due to thediscontinuation of quetiapine, we conclude that aripiprazole improved the TD symptoms.

Key words: Tardive dyskinesia, quetiapine, schizoaffective disorder

Introduction

Tardive dyskinesia (TD) is a severe side effect

induced by conventional antipsychotics and affects

approximately 20�30% of schizophrenic patients

(Egan et al. 1997). Although it is usually mild,

1�8% of patients may develop severe TD (Correl et

al. 2004). Atypical antipsychotics cause TD less

often when compared to conventional antipsychotics

(Casey 1995; Kane 2001). Although atypical med-

ication is less likely to induce TD, several cases of

TD have been reported in patients treated with

atypical antipsychotic drugs (Lykouras et al. 1999;

Llorca et al. 2002).

Quetiapine is a novel atypical antipsychotic agent

with a mainly high affinity for 5-HT (serotonin) 2A

receptors (5-HT2A), histamine (H1) and noradre-

nergic (a1) receptor. Contrary to amisulpride, que-

tiapine has a much lower affinity for D2/D3

dopaminergic receptors in the striatal regions (Ste-

phenson et al. 2000; Nemeroff et al. 2002). Reports

have suggested that chronic blocking of D2 and

especially of D3 dopamine receptors (dopamine

hypersensivity hypothesis) is associated with the

development of dyskinesias and TD in patients

with chronic schizophrenia (Casey 1995; Steen

et al. 1997; Krebs et al. 1998; Bakker et al. 2006).

It has been reported that quetiapine, when gradually

introduced, improves extra-pyramidal and dystonic

symptoms caused by other antipsychotic medication,

whilst it still provides adequate psychotic symptoms

control (Emsley et al. 2004). It has also been

suggested that switching to quetiapine may be

effective in treating patients with EPS and patients

with established TD (Saccheti et al. 2003; Emsley et

al. 2004; Gourtzis et al. 2005). However, TD in

patients treated with quetiapine have also been

reported (Sharma 2003). We present the case of a

patient who developed tardive dyskinesia three

months after the initiation of quetiapine treatment

and was eventually managed with aripiprazole.

Case report

Mrs D. a 56-year-old married female, was referred to

our University outpatient Psychiatric Department,

in October 2005 complaining of anxiety, insomnia,

agitation and poor apetite. Her psychiatric history

revealed moderate depressive episodes with psycho-

tic symptoms as well as manic episodes with

psychotic symptoms. These episodes started at the

age of 40. Although this was her first psychiatric

referral, it was established that she had a 10-year

history of untreated DSM-IV-TR schizoaffective

Correspondence: Professor Lefteris P. Lykouras, University of Athens, Medical School � 2nd Department of Psychiatry, Rimini 1, Athens

12462, Greece. Tel: �30 2 105832426. Fax: �30 2 105326453. E-mail: panpsyclin@attikonhospital.gr

The World Journal of Biological Psychiatry, 2009; 10(1): 54�57

(Received 9 November 2006; accepted 21 March 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701362550

disorder, bipolar type. She was agitated, suspicious

and depressed. She admitted to auditory hallucina-

tions and expressed ideas of persecution, guilt and

death without suicidal ideation. She had no overt

cognitive deficits and there was no evidence of

dementia. There was no history of drug or alcohol

abuse or dependence. There was no Axis II diag-

nosis. Physical examination revealed no abnormal-

ities. Laboratory tests were unremarkable except

elevated cholesterol. A brain CT revealed no ab-

normalities. A trial of amisulpride (1000 mg/day),

venlafaxine (75 mg/day) and mirtazapine (60 mg/

day) was initiated to control her affective and

psychotic symptomatology. Six weeks later her

affective and psychotic symptoms had remitted.

However, she had some extrapyramidal symptoms

of moderate severity: tremor, cogwheel rigidity,

reduced facial expression and akathisia. Tremor

could be attributed to amisulpride, and presumably

the other EPS symptoms; a causative role of novel

antidepressants in the induction of tremor cannot be

excluded. An anticholinergic agent (4 mg biperiden

hydrochloride) was added and 2 weeks later her

extrapyramidal symptoms were minimally improved.

The dose of amisulpride was reduced to 800 mg/day

and 2 weeks later to 400 mg/day, but the tremor did

not improve. Because of this persistent and distres-

sing tremor, it was decided to gradually discontinue

amisulpride treatment and she was switched to

quetiapine titrated up to 400 mg/day. Four weeks

later her EPS symptoms were improved and biper-

iden was discontinued. Three months after the

initiation of quetiapine, abnormal movements of

the jaw, lips, mouth and the upper extremities were

noted. Her mental state was stable. The Abnormal

Involuntary Movements Scale (AIMS) score at that

time was 20. It was decided to gradually discontinue

quetiapine treatment. Following this, there was a

minor improvement of TD symptoms (AIMS score:

15) within 3 weeks. In order to provide antipsychotic

cover and help her with the TD symptoms it

was decided to initiate 60 mg/day of ziprasidone.

This, resulted in further worsening of the TD

symptoms (AIMS score: 24), and 1 week later

ziprasidone treatment was discontinued (AIMS

score: 15). Following this, our patient was disap-

pointed and left for her summer holidays. During

this time she manifested mild but troublesome

depressive and anxiety symptoms and decided to

resume quetiapine at her own accord at the dose of

200 mg/day. Her TD worsened, when she returned

her AIMS was 18 and quetiapine was discontinued.

She was given aripiprazole and the dose was titrated

up to 15 mg/day. Within 3 weeks, there was an

improvement in the severity of TD symptoms

(AIMS score: 15). After 4 weeks following the

same medication there was a further improvement

of TD symptoms (AIMS score: 6), whilst her mental

state remained unchanged.

Discussion

A MEDLINE search revealed two case reports and a

series of six cases of quetiapine-induced TD. The

first reported case was that of 44-year-old female

with schizophrenia resistant to treatment with con-

ventional antipsychotics for many years (Ghelber et

al. 1999). Treatment with quetiapine for 6 months

resolved the psychosis but she developed TD.

Ghaemi et al. (2001) reported a young woman

with bipolar, type I, disorder who had no previous

exposure to typical neuroleptics and developed TD 6

weeks after the initiation of quetiapine (Ghaemi

et al. 2001). In another study, six patients were

reported, who suffered from mood disorders and

who developed TD after the initiation of quetiapine

(Sharma 2003).

Quetiapine has a low affinity for the D2 and D3

dopamine receptors and a high Koff time indicating

dissociation from D2 receptors (Kapur-Seeman

2001). In a recent positron emission tomography

(PET) study, quetiapine at 450 and 750 mg/day led

to 30 and 41% occupancy of D2 receptors in the

putamen (Gefvert et al. 2001). In a recent single

positron emission tomography (SPECT) study,

D2/D3 receptor occupancy was 32% in the striatum

and 60% in the temporal cortex with a minimum

dose of 450 mg (Stephenson et al. 2000). Despite

the above-mentioned reports, in our case TD

emerged 12 weeks after quetiapine initiation.

Advanced age, female sex, mood disorder, high

neuroleptic dose, EPS, diabetes and continuous

prolonged exposure to neuroleptics are risk factors

for TD (Kane 1999, 2004). According to the

literature, Kane proposed that medication that

causes acute EPS is more likely to cause TD. With

the exception of exposure to conventional antipsy-

chotics, our patient had all the other risk factors. In

addition, before treatment with quetiapine our

patient was given amisulpride at quite a high dosage

(Kane 2001).

The pharmacological effects of amisulpride are

related to the blockade of D2 and D3 dopamine

receptors, mainly in limbic structures (Cudennec

et al. 1997; Scatton et al. 1997). These two types of

dopamine receptors are the only ones for which

amisulpride has been shown to have a relevant

affinity. At a low dosage, amisulpride blocks the

pre-synaptic dopamine auto-receptors, and their

blockade leads to an increase in dopamine release

while, at high doses (like the ones used in our

case), it blocks post-synaptic dopamine receptors

Dyskinesia in a patient treated with quetiapine 55

in the same way as conventional antipsychotics

(Schoemaker et al. 1997). Higher plasma concen-

trations induce marked binding in striatal and

extrastriatal regions (Xiberas et al. 2001; Bressan

et al. 2003). According to the above, amisulpride

could have operated as a risk factor for the emer-

gence of TD in our patient.

Ziprasidone, when administered, caused dete-

rioration of TD. It appears that our patient was

very sensitive to elevated D2 occupancy. This is

underlined by the worsening of TD symptoms after

inititiation of ziprasidone. One could speculate that

dyskinesia was caused by ziprasidone therapy. There

are five reports (six patients) of ziprasidone-related

TD (Ananth et al. 2004; Keck et al. 2004; Mendhe-

kar et al. 2005; Papapetropoulos et al. 2005; Sharma

et al. 2005). Rosenquist, et al. (2002) described a

case of TD in a patient who was treated with

ziprasidone. This patient had previous TD. Ananth

et al. (2004) reported the occurrence of TD in two

patients on long-term ziprasidone treatment. Symp-

toms of TD were first manifested in our patient

following quetiapine administration. Ziprasidone

only caused deterioration of the TD symptoms

induced by quetiapine. Beyond that, quetiapine

seems to have played a causative role in the

emergence of TD seen in our patient, since her

own reinstitution of the drug exacerbated TD

symptomatology.

It is worth noting that, an association between the

risk of TD and a specific D3 receptor polymorphism

Ser9Gly homozygosity or heterozygosity for the

DRD3gly allele of D3 genetic locus has been reported

with antipsychotic medication. This Ser9Gly poly-

morphism of dopamine D3 receptor gene, presents

in 22�24% of patients with TD compared with 4�6%

in patients without TD (Segman et al. 1999). Recent

studies implicate the development of TD with other

genetic polymorphisms such as dopamine D4 recep-

tor gene, catalysing enzyme catechol-O-methyltrans-

ferase (COMT) gene and dopamine transporter

gene (DAT) (Srivastava et al. 2006).

Therapy with antipsychotics has always been

accompanied by treatment with novel antidepres-

sants and this may have influenced neurotransmitter

receptor sensitivity. Up to our knowledge there is no

published report on the induction of TD by venla-

faxine and mirtazapine, either on their own or in

combination. In our case, the patient continued

receiving these antidepressants throughout the re-

ported period.

The low affinity for D2 receptors casts doubt as to

whether dopamine receptor hypersensitivity could

be the underlying mechanism for the induction of

TD in our patient. According to Sharma (2003),

since quetiapine has a relatively low affinity for D2

receptors but a high affinity for 5-HT2A, H1 and a1

receptors, alternative aetiological hypotheses, parti-

cularly involving histaminergic and noradrenergic

receptors, should be considered.

Our patient was also genotyped on the basis of the

DRD3gly allele and she was found to be heterozygous

(DRD3ser-gly genotype) which has been associated

with an increased likelihood towards manifestation

of tardive dyskinesia in schizophrenia patients (Lerer

et al. 2002).

We consider as limitations of our case report that

drug levels were not determined in the plasma due to

lack of appropriate technical support. TD can

appear several months following discontinuation of

neuroleptic treatment. In this regard, our patient

could have developed TD as a result of amisulpride

treatment, and quetiapine may not have influenced

its emergence. This possibility cannot be entirely

ruled out. In any case, the fact that her TD was

worsened during the time she took quetiapine on her

own initiative for a second time, suggests a primary

role of quetiapine in the emergence of TD.

This case report underlines the need for psychia-

tric vigilance when prescribing quetiapine in patients

with mood or schizoaffective disorders, especially

when these individuals also have other risk factors

for the development of TD.

Acknowledgements

None.

Declaration of Interest

None.

References

Ananth J, Burgoyne KS, Niz D, Smith M. 2004. Tardive

dyskinesia in 2 patients treated with ziprasidone. J Psychiatry

Neurosci 29:467�469.

Bakker RP, van Harten P, van Os J. 2006. Antipsychotic-induced

tardive dyskinesia and the ser9Gly polymorphism in the DRD3

gene: A meta analysis. Schizophr Res 83:185�192.

Bressan RA, Erlandsson K, Jones HM, Mulligan R, Flanagan RJ,

Ell PJ, Pilowski LS. 2003. Is regionally selective D2/D3

dopamine occupancy sufficient for atypical antipsychotic effect?

An in vivo quantitative B123�epidepride SPET study of

amisulpride-treated patients. Am J Psychiatry 160:1413�1420.

Casey DE. 1995. Tardive dyskinesia. In: Borroni E, Kupfer DJ,

editors. Psychopharmacology. The fourth generation of pro-

gress. Nashville, TN: The American College of Neuropsycho-

pharmacology. pp 1497�1501.

Correl CU, Leucht S, Kane JM. 2004. Lower risk for tardive

dyskinesia associated with second generation antipsychotics:

a systematic review of 1-year studies. Am J Psychiatry 161:

414�425.

56 E. Rizos et al.

Cudennec A, Fage D, Benavides J, Scatton B. 1997. Effects of

amisoulpride, an atypical antipsychotic which blocks preferen-

tially presynaptic dopamine autoreceptors, on integrated func-

tional cerebral activity in the rat. Brain Res 12:257�265.

Egan MF, Apud J, Wyatt RJ. 1997. Treatment of tardive

dyskinesia. Schizophr Bull 23:583�609.

Emsley R, Turner HJ, Schronen J, Botha K, Smit R, Oosthuizen

PP. 2004. single-blind, randomized trial comparing quetiapine

and haloperidole in the treatment of tardive dyskinesia. J Clin

Psychiatry 65:696�701.

Gefvert O, Lundberg T, Wisebgren IM, et al. 2001. D2 and

5-HT2A receptor occupancy of different doses of quetiapine in

schizophrenia: a PET study. Eur Neuropsychopharmacol

11:105�110.

Ghaemi SN. 2001. Quetiapine-related tardive dyskinesia. Am J

Psychiatry 158:1737.

Ghelber D, Belmaker RH. 1999. Tardive dyskinesia with quetia-

pine. Am J Psychiatry 156:796�797.

Kane JM. 1999. Tardive Dyskinesia in affective disorders. J Clin

Psychiatry 60(Suppl 5):43�47.

Kane JM. 2001. Extrapyramidal side effects are unacceptable. Eur

Neuropsychopharmacol 11(Suppl 4):397�S403.

Kane JM. 2004. Tardive dyskinesia rates with atypical antipsy-

chotics in adults: prevalence and incidence. J Clin Psychiatry

65(Suppl):S16�20.

Kapur S, Seeman P. 2001. Does fast dissociation from the

dopamine d(2) receptor explain the action of atypical anti-

psychotics?: a new hypothesis. Am J Psychiatry 158:360�369.

Keck ME, Muller MB, Binder EB, Sonntag A, Holsboer F. 2004.

Ziprasidone-related tardive dyskinesia. Am J Psychiatry

161:175�176.

Krebs MO, Sautel F, Bourdel MC, et al. 1998. Dopamine D3

receptor gene variants and substance abuse in schizophrenia.

Mol Psychiatry 4:337�341.

Lerer B, Segman RH, Fangerau H, et al. 2002. Pharmacogenetics

of tardive dyskinesia: combined analysis of 780 patients

supports association with dopamine D3 receptor gene Ser9-

Glypolymorphism. Neuropsychopharmacology 27(1):105�119.

Llorca PM, Cherau I, Bayle FJ, Lancon C. 2002. Tardive

dyskinesia and antipsychotics: a review. Eur Psychiatry 17:

129�138.

Lykouras L, Yannakis R, Hatzimanolis J, Christodoulou GN.

1999. Two cases of risperidone-induced tardive dyskinesia and

a review of the literature. Eur Psychiatry 14:245�247.

Mendhekar DN. 2005. Ziprasidone-induced tardive dyskinesia.

Can J Psychiatry 50:567�568.

Nemeroff CB, Kinkead B, Goldstein I. 2002. Quetiapine:

preclinical studies, pharmacokinetics, drug interactions and

dosing. J Clin Psychiatry 63(Suppl 13):5�11.

Papapetropoulos S, Wheeler S, Singer C. 2005. Tardive dystonia

associated with ziprasidone. Am J Psychiatry 162:2191.

Rosenquist KJ, Walker SS, Ghaemi SN. 2002. Tardive Dyskinesia

and ziprasidone. Am J Psychiatry 159:1436.

Saccheti E, Valsecchi P. 2003. Quetiapine, clozapine, and

olanzapine in the treatment of tardive dyskinesia induced by

first-generation antipsychotics: a 124-week case report. Int Clin

Psychopharmacol 18:357�359.

Scatton B, Claustre Y, Cudennec A, et al. 1997. Amisoulpride:

from animal pharmacology to therapeutic action. Int Clin

Psychopgarmacology 12(Suppl 2):S29�36.

Schoemaker H, Claustre Y, Fage D. 1997. Neurochemical

characteristics of amisulpride, an atypical dopamine D2/D3

receptor antagonist with both presynaptic and limbic selectivity.

J Pharmacol Exp Ther 280:83�97.

Segman R, Neeman T, Heresco-Levy U, et al. 1999. Genotyping

association between the dopamine D3 receptor and tardive

dyskinesia in chronic schizophrenia. Mol Psychiatry 4:418�428.

Sharma A, Ramaswamy S, Dewan VK. 2005. Resolution of

ziprasidone-related tardive dyskinesia with a switch to aripipra-

zole. Prim Care Companion J Clin Psychiatry 7:36.

Sharma V. 2003. Treatment-emergent tardive dyskinesia with

quetiapine in Mood Disorders. J Clin Psychopharmacol

23:415�416.

Srivastava V, Varma PG, Prasad S, et al. 2006. Genetic suscept-

ibility to tardive dyskinesia among schizophrenia subjects: IV.

Role of dopaminergic pathway gene polymorphisms. Pharma-

cogenet Genom 16:111�117.

Steen VM, Lovelie R, MacEvan T, McCreadie RG. 1997.

Dopamine D3 receptor gene variant and susceptibility to

tardive dyskinesia in schizophrenic patients. Mol Psychiatry

2:139�145.

Stephenson CM, Bigliani V, Jones HM, et al. 2000. Striatal

and extra-striatal D(2)/D(3) dopamine receptor occupancy

by quetiapine in vivo. [(123)I]-epidepride single photon emis-

sion tomography (SPET) study. Br J Psychiatry 177:408�415.

Xiberas X, Martinot JL, Mallet L, et al. 2001. In vivo extrastriatal

and striatal D2 dopamine receptor blockage by amisulpride in

schizophrenia. J Clin Psychopharmacol 21:207�214.

Dyskinesia in a patient treated with quetiapine 57

BRIEF REPORT

Clinical characteristics in long-term care psychiatric patients:A descriptive study

ANNA PLACENTINO1, LUCIANA RILLOSI2, EMANUELA PAPA1, GIOVANNI FORESTI3,

ANDREA MATERZANINI4, GIUSEPPE ROSSI2, GIOVANNI BATTISTA TURA2 &

JORGE PEREZ1

1Biological Psychiatry Unit and Dual Diagnosis Ward IRCCS, Centro San Giovanni di Dio, Fatebenefratelli, Brescia, Italy,2Psychiatric Unit IRCCS, Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy, 3AFaR Institute, Sacro Cuore di

Gesu, Fatebenefratelli, San Colombano al Lambro (MI), Italy, and 4Psychiatric Unit, Mellino Mellini Hospital, Chiari

(BS), Italy

AbstractObjective: The purpose of this study was to describe the clinical characteristics and related comorbid conditions ofpsychiatric patients admitted to residential facilities (RFs) and their impact on the levels of functioning of such patients.Methods: We assessed 426 patients admitted to residential facilities, by using SCID-I, SCID-II, BPRS, GAF and DAS.Results: The most common diagnostic category was schizophrenia/psychotic disorders (41.8%), followed by affectivedisorders (35.4%), personality disorders (14.1%), and other disorders (8.7%). In addition 33.3% had a psychiatriccomorbidity, and 62.6% had a medical comorbidity. Low levels of functioning were significantly correlated with bothmedical and psychiatric comorbid conditions. Conclusions: Comorbidity is common in patients requiring long-termresidential care. Thus improved detection and treatment of these conditions in people with severe mental illnesses will havesignificant benefits for their psychosocial functioning and overall quality of life.

Key words: Severe mental disorders, residential care, comorbid conditions, level of functioning

Introduction

Psychiatric disorders usually have their onset early in

life, and often turn into a chronic or recurrent course,

causing serious impairments and great human suffer-

ing. The most common psychiatric disorders are

among the 10 leading causes of disability in the

developed countries, probably causing increasingly

greater health, social and economic problems in the

coming years (Hyman 2000).

This picture has become even more complex,

considering the recent studies showing that besides

the principal diagnosis most patients have concur-

rent psychiatric and/or medical disorders, as well

as psychosocial and environmental complications

(Olfson and Pincus 1996; Pincus et al. 1999; Dixon

et al. 2000; Iosifescu et al. 2003; Lambert et al.

2003; Baigent 2005; Khan et al. 2005; Chwastiak

et al. 2006). It has been reported (Kessler et al.

1995; McDermuth et al. 2001; Lyketsos et al. 2002;

Melartin et al. 2002; Bruce et al. 2005; Sim et al.

2006) that patients with comorbid conditions have

treatment complications, chronic course, worse out-

come and poor psychosocial functioning as com-

pared with individuals with a single disorder.

Although, previous research carried out in different

clinical settings have documented rates of concurrent

psychiatric or medical disorders in psychiatric pa-

tients (Dixon et al. 1999; Zimmerman and Mattia

2000; Koran et al. 2002; Evans and Charney 2003;

Lambert et al. 2003; Pierce and Wilson 2004; Sokal

et al. 2004; Kessler et al. 2005; Khan et al. 2005;

Chwastiak et al. 2006), less clinical information are

available about those patients requiring long-term

residential care.

Therefore, the purpose of the current study was to

describe the clinical characteristics and related

comorbid conditions of psychiatric patients admitted

to residential facilities (RFs). We also examined

Correspondence: Jorge Perez, MD, PhD, Biological Psychiatry Unit and Dual Diagnosis Ward, IRCCS Centro S. Giovanni di Dio,

Fatebenefratelli, Via Pilastroni 4, 25125 Brescia, Italy. Tel: �39 30 3501719. Fax: �39 30 3501592. E-mail: jperez@fatebenefratelli.it

The World Journal of Biological Psychiatry, 2009; 10(1): 58�64

(Revised 25 July 2006; accepted 27 March 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701367922

whether comorbid conditions could affect the levels

of functioning in such patients.

Methods

The sample consisted of 426 consecutive patients

admitted to a public and three private residential

facilities (RFs) during a period of 22 months. The

residential settings in Italy constitute a heteroge-

neous system (de Girolamo et al. 2002; Santone

et al. 2005). In the present study the residential

facilities were characterized by providing long-term

accommodation and high care (facilities with 24-h

staff cover.). Each RF was constituted by three or

four independent houses with 10 residents per

house. Patients admitted to these RFs have generally

several difficulties living in a community due to their

psychopathological problems (i.e. negative symp-

toms, aggression behaviour), have disabilities in

daily living skills and lack of adequate family

support. Dementia and severe mental retardation

were excluded, while acute conditions were referred

to a Mental Health Hospital. The average time spent

in the current RFs was 15.8 months ( SD 3.7).

Public and private RFs had the same admission

criteria and both were entirely paid by the public

funds.

All patients were eligible if they or their key-

caregivers had provided a signed informed consent,

after having understood the aim of the study.

Patients were assessed within 7 days of admission

or when their clinical conditions permitted evalua-

tion by the structured Clinical Interview for DSM-

IV (SCID-I and SCID-II) (First et al. 1994a,b), to

generate both principal and comorbid psychiatric

diagnosis. Clinical information was collected directly

from each patient and from at least one close relative

as co-informant. Any definitive documentation (e.g.,

discharge summaries of prior hospital admissions for

episodes of illness) that could be obtained was

considered in assessing the diagnosis.

Psychiatric comorbidity was defined by DSM-IV

criteria, considering both Axis I and II disorders.

The severity of psychiatric symptoms was assess-

ed by the Brief Psychiatric Rating Scale (BPRS)

(Roncone et al. 1999). The overall assessment of the

individual social and occupational functioning was

obtained from the Global Assessment of Function-

ing Scale (GAF, DSM-IV Axis V) (APA 1994); in

addition, patient disability was assessed by using the

first two sections (overall behaviour and social role

functioning) of the Disability Assessment Schedule

(DAS-II) (WHO 1988). All the instruments were

administrated by trained psychiatrists.

The medical health of each subject was documen-

ted by the medical history, a physical examination,

electrocardiography, blood and serum chemical

analyses (including hepatic and renal profiles). The

psychiatrist referred patients with medical problems

to the department of medicine for an evaluation

and treatment recommendations. Through these

assessments the current physical disorders (medical

comorbidities) were collected; the current and im-

portant physical disorders were those that required a

treatment or a continued medical surveillance and

could threaten life or exacerbate a psychiatric

disorder or had potentially significant consequences

for health or functioning.

Statistical evaluations were performed using the

SPSS version 13.0 software package. Descriptive

analyses (i.e. percentages, mean and standard devia-

tion) examined the frequency of sociodemographic

and clinical data. One-way ANOVA was used to

analyse the mean differences between groups on the

following variables: levels of functioning and comor-

bid conditions as well as mean number of comorbid-

ities and diagnostic category; post hoc analysis

were performed with Bonferroni test. Pearson and

Spearman correlations were respectively used for

parametric and non-parametric variables to investi-

gate the correlation between comorbidity, levels of

functioning and the severity scale. P values 50.05

were considered statistically significant in the above

analyses.

Results

In this study, 426 patients were enrolled; all of them

were Caucasian and living in Northern Italy. Socio-

demographic data are shown in Table I.

On admission the principal diagnoses were schi-

zophrenia/psychotic disorders (41.8%), affective

Table I. Socio-demographic data (n�426).

Gender

Male 222 (52.1%)

Female 204 (47.9%)

Mean age 47.7 (SD 14.7)

Marital status

Never married (i.e. single) 197 (46.2%)

Married 134 (31.5%)

Divorced or separated 67 (15.8%)

Widowed 28 (6.5%)

Mean years of education 8.2 (SD 2.4)

Living arrangement

Immediate family 275 (64.5%)

Alone 100 (23.5%)

Other (relatives, friends) 51 (12%)

Occupational arrangement

Retired (social security for mental disorders) 218 (51.2%)

Unemployed 109 (25.6%)

Psychosocial rehabilitation program 84 (19%)

Paid employment 15 (14.1%)

Psychiatric patients in long-term care 59

disorders (35.4%), personality disorders (14.1%),

substance use disorders (3.5%), mental disability

(3.1%) and obsessive-compulsive disorders (2.1%).

The mean total BPRS score was 48.7 (SD 9.9)

showing moderate psychiatric symptoms, while the

mean total GAF score was 37.4 (SD 11.1), marking

an evident impairment in several areas. DAS1

(overall behaviour section) showed that 42.7% of

the patients had mild/moderate disability and 57.3%

had severe disability, while in the DAS2 (social role

functioning section) 20.9% of the patients had mild/

moderate disability and 79.1% of the patients had

severe disability.

The mean age of onset of the psychiatric illness

was 30.9 years (SD 14.9), whereas that of primary

admission to the mental health care services was

33.8 years (SD 15.3).

Several patients had a history of attempted suicide

(28.1%) and aggressive behaviour (34.2%).

As can be seen in Table II, 33.3% of patients had a

psychiatric comorbidity: the most common diagnos-

tic category was personality disorders (42.9%)

followed by substance use disorders (39.4%), affec-

tive disorders (8.5%), anxiety disorders (7.1%), and

eating disorders (2.1%). Almost all the patients

(79%) smoked tobacco, with a daily mean of 17.8

(SD 12.3) cigarettes. Alcohol drinking was reported

for 27.6% patients, with a mean weekly consump-

tion of 5.7 units (SD 14.2), while 1.5% patients

reported occasional use of cannabis, without meet-

ing abuse criteria.

In addition, at least a physical disorder was

noticed for 62.6% of patients. In particular, 53.7%

had one physical illness and 48.3% of patients had

two or more. The frequency of medical illnesses in

patients aged between 40 and 60 years was 75.1%,

followed by those over 60 years ( 67%) and those

under 41 years (57 patients, 40.9%). The most

frequent were cardio-circulatory diseases followed

by endocrine/metabolic diseases and gastrointestinal

system disorders (Table III).

As can be seen in Figure 1 at least one psychiatric

or medical concurrent disorders was found in 85.7%

of patients with a principal diagnosis of substance

use disorder; 70.4% of affective patients, 66.6% of

patients with personality disorders and 53.3% of

patients with psychosis. Among them, significant

difference were found in the mean number of

comorbidities (F(4,425)�4.71, P�0.001); patients

with substance use disorders had in mean 1.81 (SD

1.13) comorbidities followed by personality disor-

ders 1.63 (SD 1.61), affective disorders 1.57 (SD

1.49), and psychotic disorders 0.92 (SD 1.26)

comorbidities.

Table II. Number of patients with a psychiatric comorbidity

(n�142).

Personality disorders 61 42.9%

Borderline 27

Paranoid 11

Histrionic 9

Substance use disorders 56 39.4%

Alcohol use disorders 41

Substance use disorders 15

Affective disorders 12 8.5%

Major depression 7

Dysthimia 5

Anxiety disorders 10 7.1%

Generalized anxiety disorders 6

Obsessive compulsive disorders 4

Eating disorders (Bulimia) 3 2.1%

Table III. Frequency of physical conditions found in patients

admitted to residential facilities (n�404).

Cardio-circulatory disorders 121 30%

Essential hypertension 68

Ischemic heart disease 39

Arrhythmia 7

Deep venous thrombosis 3

Dilated cardiomyopathy 2

Secondary hypertension 2

Endocrine, nutritional or metabolic diseases 115 28.50%

Diabetes 37

Obesity 31

Hyperlipidemia 26

Osteoporosis 15

Hypothyroidism, Thyroidism 6

Gastrointestinal disorders 74 18.40%

Gastric ulcer 37

Alcoholic liver diseases 19

Hepatitis 14

Pancreatic diseases 1

Chronic gastritis 2

Crohn’s disease 1

Respiratory disorders 30 7.50%

Chronic bronchitis, Emphysema 17

Bronchial asthma 12

Allergic asthma 1

Neurological disorders 28 7%

Epilepsy 12

Cerebrovascular diseases 8

Neuropathy 3

Meningitis 3

Huntington’s chorea 2

Hematopoiesis or hemostatis disorders 9 2.20%

Physical injury 7 1.80%

Sensory disorders 6 1.50%

Immune disorders 5 1%

HIV 2

Rheumatoid arthritis 1

Lupus erythematosus 1

Sclerosis 1

Kidney and urinary disorders 4 0.90%

Neoplasm 3 0.70%

Psoriasis 2 0.50%

60 A. Placentino et al.

The levels of functioning measured with GAF and

DAS scales were statistically different stratifying the

sample for the number of comorbid conditions

(Table IV).

We found a negative correlation between the

overall GAF scores and the psychiatric/medical

comorbidities (r��0.313, PB0.001). In addi-

tion, DAS1 (overall behaviour section) was, posi-

tively correlated with both medical (r��0.291,

PB0.001) and psychiatric comorbid conditions

(r��0.207, PB0.001). Similarly, DAS2 (social

role functioning section) was positively correla-

ted with both medical (r��0.251, PB0.001)

and psychiatric comorbid conditions (r��0.183,

PB0.001).

Not significant correlation was found between

BPRS score and the total number of comorbid

conditions

All the patients were receiving at least a medica-

tion with a mean of 4.7 (SD 2.3) medications

prescribed per patient. A total of 22.2% of all the

patients were receiving one medication; 21.3% were

receiving two; 19.4% were receiving three; and

36.7% were receiving four or more. The most

frequent psychotropic medications were antipsycho-

tic followed by antidepressants and mood stabilizers;

polypharmacy was common and each patient was

taking 2.5 (SD 1.1) of psychotropic medications.

Discussion

Careful comorbidity information, along with patient

demographics, psychological and social risk charac-

teristics are essential for comprehensive risk adjust-

ment which in turn is necessary for observational

and mental health services research, including a

comparison of the outcomes of different treatments

and quality assessment. Consonant with that, during

the last years, several studies reported that psychia-

tric patients assessed in different clinical settings had

one or more comorbid psychiatric and medical

conditions (Wolf et al. 1988; Rounsaville et al.

1991; Dixon et al. 1999; Pincus et al. 1999;

McDermut et al. 2001; Melartin et al. 2002; Hofer

et al. 2004; Sokal et al. 2004; Sim et al. 2006).

However, little is known about the clinical features

of those patients requiring long-term residential

care. The results of the current study show that

comorbid conditions are common among patients

admitted to RFs. We found that in addition to the

principal diagnosis, 33.3% of the patients had

psychiatric comorbidities, while more than the half

of the sample (62.6%) also had medical comorbid-

ities. These clinical conditions were correlated with

poor levels of psychosocial functioning and overall

quality of life.

The principal diagnostic categories reported in

our study are in line with previous studies (Lehman

et al. 1986; Zissi and Barry 1997; Fenton et al. 1998;

Hyman 2000; de Girolamo et al. 2005; Kim et al.

2006) showing that besides psychosis, mood and

personality disorders are frequently reported among

patients admitted to residential facilities.

It has been reported that psychiatric comorbidities

are underdetected in clinical practice due to the

process of establishing diagnosis with unstructured

Figure 1. Comorbidity rates observed in patients admitted to RFs.

Table IV. Overall comorbid conditions and level of functioning (means and one-way ANOVA).

No comorbidity

(N�137)

One comorbidity

(N�119)

Two comorbidities or more

(N�161)

GAF Mean 45.2 (SD 11.7) Mean 35.2 (SD 11.0) Mean 32.1 (SD 11.4) F (2, 425)�7.97; PB0.001

DAS1 Mean 2.64 (SD 1.0) Mean 2.88 (SD 0.9) Mean 3.35 (SD 0.9) F (2, 425)�20.22; PB0.001

DAS2 Mean 2.75 (SD 1.0) Mean 3.13 (SD 0.8) Mean 3.60 (SD 0.7) F (2, 425)�35.14; PB0.001

Psychiatric patients in long-term care 61

interview (Zimmerman and Mattia 1999; Wilk et al

2006). In our study, using structured interview, we

found 33.3% of the patients had psychiatric comor-

bidities and the most recurrent coexisting psychiatric

disorders were personality disorders, substance

abuse, and affective disorders. Although, no specific

data were reported about psychiatric comorbidities

in RFs patients, previous researches have shown that

patients in others clinical settings had one or more

psychiatric comorbid condition (Wolf et al. 1988;

Rounsaville et al. 1991; McDermut et al. 2001) with

a frequency similar to that observed in our sample.

Unlike early studies showing that long stay psy-

chiatric patients had lower levels of physical dis-

orders (Lelliott et al. 1996) and higher level of motor

disabilities (de Girolamo 2005), we found that

62.6% of the patients had medical comorbid condi-

tions, with higher frequency of cardio-circulatory

and metabolic diseases. These apparently contra-

dictory findings might be related either to the

characteristics of the RFs or to different methods

of assessment.

It is worth mentioning that, although medical

complications in patients over 65 years could be

aged-related, in our sample the higher frequency of

medical disorders was recorded in patients aged

between 40 and 60 years. This is an interesting

finding considering previous studies reporting that

physical illness can impact negatively on the symp-

toms of mental illness, can affect the treatment as

well as the quality of life, and the subsequent

management of the medical illness (Dixon et al.

1999; Hofer et al. 2004; Sokal et al. 2004; Sim et al.

2006).

In the present study we were unable to determine

the onset of the comorbid conditions so that the true

influence of such conditions on the course and

outcome of the principal diagnosis is unclear. More-

over, it is unknown whether coexisting diseases are

in some way related to adverse effects of psycho-

tropic medications. Despite these important limita-

tions, our results suggest that comorbid conditions

worsened the global functioning of long-term pa-

tients. These finding are in line with those reports

documenting that patients with comorbid conditions

have, poor psychosocial functioning, longer duration

of hospitalization and worse clinical course and

outcome (Lehman et al. 1989; Kessler et al. 1995;

Melartin et al. 2002; Barbato et al. 2004; Bruce et al.

2005).

Although more studies are necessary for a better

understanding of clinic course and outcome of RFs

patients, our findings underscore the need for early

intervention to prevent comorbid conditions that

contribute to poor functioning in this population.

Such clinical approach can be added to mental

health systems or incorporated into decision making

programs to help the patient in consultation and to

improve his making decision. Improved descriptions

of the patient with severe mental illnesses results in

improved prognostic stratification, thus given a more

accurate estimates of treatment effectiveness

through the outcomes research and by the results

of the observational studies. With regard to this,

previous researches have demonstrated that patients

with comorbid psychiatric and medical conditions

have poorer response to treatment (Carrasco et al.

1992; Thomas et al. 1999; Cacciola et al. 2001;

Pohjasvaara et al. 2001; Barbato et al. 2004; Mulder

et al. 2006).

We are aware of the much more work required

before we can define evaluation methods for patients

requiring residential care and devise ways to con-

vince program administrators and staff to implement

them. In this context, our results, albeit preliminary,

highlighted that studies are needed in the following

areas: (i) we need to determine the most common

pathways to comorbidity, critical periods of vulner-

ability and its relationship with psychosocial factors,

and how and why these differ by demographic

characteristics and principal diagnostic category.

Longitudinal studies that identify the most common

developmental sequences will demonstrate when

preventive intervention may be most beneficial; (ii)

we must develop more effective treatments for

comorbid conditions and identify which treatments

work best for special populations.

In conclusion, comorbid conditions are common

in patients requiring long-term residential care call-

ing for a greater awareness in clinicians of these

conditions, which are often under-recognized, un-

der-diagnosed and untreated. Improved detection

and treatment of coexisting psychiatric or medical

conditions in people with severe mental illnesses will

have significant benefits for their psychosocial func-

tioning and overall quality of life.

Acknowledgements

This study was supported by a grant of the Italian

Ministry of Health to Biological Psychiatric Unit

(RC 2006/2008) IRCCS-FBF Brescia, Italy.

Declaration of Interest

None.

References

American Psychiatric Association. 1994. Diagnostic and statistical

manual of mental disorders. 4th ed.Washington, DC: APA.

62 A. Placentino et al.

Baigent MF. 2005. Understanding alcohol misuse and comorbid

psychiatric disorders. Curr Option Psychiatry 18:223�228.

Barbato A, D’Avanzo B, Rocca G, Amatulli A, Lampugnani D.

2004. A study of long-stay patients resettled in the community

after closure of a psychiatric hospital in Italy. Psychiatr Serv

55:67�70.

Bruce SE, Yonkers KA, Otto MW, et al. 2005. Influence of

psychiatric comorbidity on recovery and recurrence in general-

ized anxiety disorder, social phobia, and panic disorder: a

12-year prospective study. Am J Psychiatry 162:1179�1187.

Cacciola JS, Alterman AL, Rutherford MJ, MacKay JR, Mulvaney

FD. 2001. The relationship of psychiatric comorbidity to

treatment outcomes in methadone maintained patients. Drug

Alcohol Depend 61:271�280.

Carrasco JL, Hollander E, Schneier FR, Liebowitz MR. 1992.

Treatment outcome of obsessive compulsive disorder with

comorbid social phobia. J Clin Psychiatry 53:387�391.

Chwastiak L, Rosenheck R, Leslie D. 2006. Impact of medical

comorbidity on the quality of schizophrenia pharmacotherapy

in a national VA sample. Med Care 44:55�61.

de Girolamo A, Picardi R, Micciolo R, Falloon I, Fioritti A,

Morosini P, PROGRES Group. 2002. Residential care in Italy.

National survey of non�hospital facilities. Br J Psychiatry

181:220�225.

de Girolamo G, Picardi A, Santone G, Falloon I, Morosini P,

Fioritti A, Micciolo R, PROGRES Group. 2005. The severely

mentally ill in residential facilities: a national survey in Italy.

Psychol Med 35:421�31.

Dixon L, Postrado L, Delahanty J, Fischer PJ, Lehman A. 1999.

The association of medical comorbidity in schizophrenia with

poor physical and mental health. J Nerv Ment Dis 187:

496�502.

Dixon L, Weiden P, Delahanty J, et al. 2000. Prevalence and

correlates of diabetes in national schizophrenia samples.

Schizophr Bull 26:903�912.

Evans DL, Charney DS. 2003. Mood disorders and medical

illness: a major public health problem. Biol Psychiatry 54:

177�180.

Evans M, Hammond M, Wilson K, Lye M, Copeland J. 1997.

Treatment of depression in the elderly: effect of physical illness

on response. Int J Geriatr Psychiatry 12:1189�1194.

Fenton WS, Mosher LR, Herrell JM, Blyler CR. 1998. Rando-

mized trial of general hospital and residential alternative care

for patients with severe and persistent mental illness. Am J

Psychiatry 155:516�522.

First MB, Spitzer RL, Gibbon M, Williams JBW. 1994a.

Structured Clinical Interview for DSM-IV Axis I Disorders,

Patient Edition (SCID-P, vers 2.). New York, NY: New York

State Psychiatric Institute Biometrics Research.

First MB, Spitzer RL, Gibbon M, Williams JBW, Benjamin L.

1994b. Structured Clinical Interview for DSM-IV Axis II

Personality Disorders (SCID-II, vers 2). New York, NY: New

York State Psychiatric Institute Biometrics Research.

Hofer A, Kemmler G, Eder U, Edlinger M, Hummer M,

Fleischhacker WW. 2004. Quality of life in schizophrenia: the

impact of psychopathology, attitude toward medication, and

side effects. J Clin Psychiatry 65:932�939.

Hyman S. 2000. Mental illness: genetically complex disorders of

neural circuitry and neural communication. Neuron 28:

321�323.

Iosifescu DV, Nierenberg AA, Alpertb JE, et al. 2003. The impact

of medical comorbidity on acute treatment in major depressive

disorder. Am J Psychiatry 160:2122�2127.

Kessler RC, Foster C, Saunders W, Stang PE. 1995. The social

consequences of psychiatric disorders, I. Educational attain-

ment. Am J Psychiatry 152:1026�1032.

Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE.

2005. Prevalence, severity, and comorbidity of 12-month

DSM-IV disorders in the National Comorbidity Survey Re-

plication. Arch Gen Psychiatry 62:617�627.

Kim EY, Cho E, June KJ. 2006. Factors influencing use of home

care and nursing homes. J Adv Nurs 54:511�517.

Khan AA, Jacobson KC, Gardner CO, Prescott CA, Kendler KS.

2005. Personality and comorbidity of common psychiatric

disorders. Br J Psychiatry 186:190�196.

Koran LM, Sheline Y, Imai K, et al. 2002. Medical disorders

among patients admitted to a public-sector psychiatric inpa-

tient unit. Psychiatr Serv 53:1623�1625.

Lambert TJ, Velakoulis D, Pantelis C. 2003. Medical comorbidity

in schizophrenia. Med J 78:67�70.

Lehman AF, Possidente S, Hawker F. 1986. The quality of life of

chronic patients in a state hospital and in community resi-

dences. Hosp Comm Psychiatr 37:901�907.

Lehman AF, Myers CP, Corty E. 1989. Assessment and

classification of patients with psychiatric and substance abuse

syndromes. Hosp Community Psychiatry 40:1019�1025.

Lelliott P, Audini B, Duffet R. 2001. Survey of patients from an

inner-London health authority in medium secure psychiatric

care. Br J Psychiatry 178:62�66.

Lyketsos CG, Dunn G, Kaminsky MJ, Breakey WR. 2002.

Medical comorbidity in psychiatric inpatients: relation to

clinical outcomes and hospital length of stay. Psychosomatics

43:24�30.

McDermuth W, Mattia J, Zimmerman M. 2001. Comorbidity

burden and its impact on psycosocial morbidity in depressed

outpatients. J Affect Disord 65:289�295.

Melartin TK, Rytsala HJ, Leskela US, Lestela-Mielonen PS,

Sokero TP, Isometsa ET. 2002. Current comorbidity of

psychiatric disorders among DSM-IV major depressive dis-

order patients in psychiatric care in the Vantaa Depression

Study. J Clin Psychiatry 63:126�134.

Mulder RT, Joyce PR, Frampton CM, Luty SE, Sullivan PF.

2006. Six months of treatment for depression: outcome and

predictors of the course of illness. Am J Psychiatry 16:95�100.

Olfson M, Pincus HA. 1996. Outpatient mental health care in

nonhospital settings: distribution of patients across provider

groups. Am J Psychiatry 153:1353�1356.

Pierce D, Wilson I. 2004. Psychiatric comorbidity in general

practice. Aust Fam Physician 33:217�220.

Pincus HA, Zarin DA, Tanielian TL, et al. 1999. Psychiatric

patients and treatments in 1997: findings from the American

Psychiatric Practice Research Network. Arch Gen Psychiatry.

56:441�9.

Pohjasvaara T, Vataja R, Leppavuori A, Kaste M, Erkinjuntti T.

2001. Depression is an independent predictor of poor long-

term outcome post-stroke. Eur J Neurol 8:315�319.

Roncone R, Ventura J, Impallomeni M, et al. 1999. Reliability of

an Italian standardized and expanded Brief Psychiatric Rating

Scale (BPRS 4.0) in raters with high vs. low clinical experience.

Acta Psychiatr Scand 100:229�236.

Rounsaville BJ, Anton SF, Carrol K, Budde D, Prusoff BA, Gawin

F. 1991. Psychiatric diagnoses of treatment- seeking cocaine

abusers. Arch Gen Psychiatry 152:1026�1032.

Santone G, de Girolamo G, Falloon I, Fioritti A, Micciolo R,

Picardi A, Zanalda E, PROGRES Group. 2005. The process of

care in residential facilities � a national survey in Italy. Soc

Psychiatry Psychiatr Epidemiol 40:540�50.

Sim K, Chan YH, Chua TH, Mahendran R, Chong SA, McGorry

P. 2006. Physical comorbidity, insight, quality of life and global

functioning in first episode schizophrenia: a 24-month, long-

itudinal outcome study. Schizophr Res 88:82�9.

Psychiatric patients in long-term care 63

Sokal J, Messias E, Dickerson FB, et al. 2004. Comorbidity of

medical illnesses among adults with serious mental illness who

are receiving community psychiatric services. J Nerv Ment Dis

192:421�427.

Thomas VH, Melchert TP, Banken JA. 1999. Substance depen-

dence and personality disorders: comorbidity and treatment

outcome in an inpatient treatment population. J Stud Alcohol

60:271�277.

Wilk JE, West JC, Narrow WE, et al. 2006. Comorbidity patterns

in routine psychiatric practice: is there evidence of under-

detection and underdiagnosis? Compr Psychiatry 47:258�64.

Wolf AW, Schubert DSP, Patterson MB, Marion B, Grande TP.

1998. Associations among major psychiatric diagnoses.

J Consult Clin Psychol 56:289�295.

World Health Organization. 1988. WHO psychiatric disability

assessment schedule (WHO/DAS) with a guide to its use.

Geneva.

Zimmerman M, Mattia JI. 1999. Psychiatric diagnosis in clinical

practice: is comorbidity being missed? Compr Psychiatry

40:182�91.

Zimmerman M, Mattia JI. 2000. Principal and additional DSM-

IV disorders for which outpatients seek treatment. Psychiatr

Serv 51:1299�1304.

Zissi A, Barry MM. 1997. From Leros asylum to community-

based facilities: levels of functioning and quality of life among

hostel residents in Greece. Int J Soc Psychiatry 43:104�115.

64 A. Placentino et al.

BRIEF REPORT

Abnormal microstructures of the basal ganglia in schizophreniarevealed by diffusion tensor imaging

RYOTA HASHIMOTO1�3, TAKEYUKI MORI3,4, KIYOTAKA NEMOTO4,

YOSHIYA MORIGUCHI4, HIROKO NOGUCHI3, TETSUO NAKABAYASHI5,

HIROAKI HORI3, SEIICHI HARADA5, HIROSHI KUNUGI3, OSAMU SAITOH5 &

TAKASHI OHNISHI3,4,6

1The Osaka-Hamamatsu Joint Research Center For Child Mental Development, Osaka University Graduate School of

Medicine, 2Department of Psychiatry, Osaka University Graduate School of Medicine, 3Department of Mental Disorder

Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4Department of Radiology,

National Center Hospital for Mental, Nervous, and Muscular Disorders, National Center of Neurology and Psychiatry,5Department of Psychiatry, National Center Hospital for Mental, Nervous, and Muscular Disorders, National Center of

Neurology and Psychiatry, and 6Department of Investigative Radiology, Research Institute, National Cardiovascular Center,

Osaka, Japan

AbstractThere has been a hypothesis that deficits in the basal ganglia�thalamic system may play an important role in thedysfunctional goal-directed behaviour in schizophrenia. By using diffusion tensor imaging, we measured fractionalanisotropy (FA) values in the basal ganglia�thalamic system in 42 schizophrenics and 42 matched controls to investigatemicrostructural tissue alterations in the basal ganglia�thalamic system in schizophrenia. Schizophrenics had significantlylower FA values in the bilateral globus pallidus and left thalamus compared to controls, suggesting that schizophrenics mighthave microstructural abnormalities in globus pallidus and thalamus. These data support the notion that myelinationabnormalities in basal ganglia�thalamic system are related to the pathophysiology of schizophrenia.

Key words: Schizophrenia, diffusion tensor imaging, basal ganglia, globus pallidus, MRI

Introduction

Schizophrenia often demonstrated movement ab-

normalities, such as catatonia, pacing and other

stereotyped behaviours considered to be associated

with basal ganglia dysfunction. The basal ganglia

regulates not only motor behaviours but also aspects

of cognitive and limbic behaviours. There has been a

hypothesis that deficits in the basal ganglia�thalamic

system may play an important role in the dysfunc-

tional goal-directed behaviour in schizophrenia

(Andreasen 1999). In fact, several studies demon-

strated abnormalities in the basal ganglia in schizo-

phrenic brains, including the volume reductions of

the pallidum internum of postmortem brains of

patients with schizophrenia (Bogerts et al. 1985),

higher volumes in the globus pallidus of previously

treated patients with schizophrenia than the healthy

comparison subjects and the neuroleptic-naive pa-

tients (Gur et al. 1998), fMRI evidence for basal

ganglia dysfunction in subjects with schizophrenia

(Menon et al. 2001), abnormality of oligodendro-

glial cells in caudate nucleus in schizophrenia

(Uranova et al. 2001), and positive correlation

between globus pallidus and the severity of global

symptoms in neuroleptic-naive patients (Spinks

et al. 2005).

Diffusion tensor imaging (DTI) is a relatively new

technique, and it is useful for evaluating white

matter abnormalities in schizophrenia. We have

reported progressive changes of white matter integ-

rity in schizophrenia using DTI (Mori et al. 2007).

Recently, this technique was applied to investigate

Correspondence: Ryota Hashimoto, MD, The Osaka-Hamamatsu Joint Research Center for Child Mental Development, Osaka University

Graduate School of Medicine, D3, 2-2, Yamadaoka, Suita, Osaka, 565-0871, Japan. Tel: �81 6 6879 3074. Fax: �81 6 6879 3059.

E-mail: hashimor@psy.med.osaka-u.ac.jp

The World Journal of Biological Psychiatry, 2009; 10(1): 65�69

(Received 4 July 2007; accepted 17 October 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701762536

abnormalities of the subcortical regions in neurode-

generative diseases. Patients with Parkinson’s disease

had significantly decreased fractional anisotropy

(FA) in the region of interest along a line between

the substantia nigra and the lower part of the

putamen/caudate complex, in which the nigrostriatal

dopaminergic neurons are lost in Parkinson’s dis-

ease, demonstrating its possibility to detect micro-

structural tissue alterations (Yoshikawa et al. 2004).

To investigate possible microstructural abnormalities

in the basal ganglia�thalamic system in schizophre-

nia, we measured FA values in the basal ganglia and

the thalami in schizophrenics and in normal controls

for comparison, as a sub-analysis of our previous

study (Mori et al. 2007).

Material and methods

Subjects and clinical assessments

Forty-two patients with DSM-IV schizophrenia

(26 male and 16 female, one left hander, mean

age: 40.099.3 years old, education: 13.092.9 years,

mean duration of illness; 16.899.0 years, mean

daily dose of antipsychotics (chlorpromazine equiva-

lent): 1005.19735.3 mg/day) (Association 1994)

and 42 controls (26 male and 16 female, one left

hander, mean age: 39.299.0 years old, education:

17.193.5 years) were participated in our study.

Written informed consent was obtained from all

the subjects. This study has been approved by the

local ethics committee and has therefore been

performed in accordance with the ethical standards

laid down in the 1964 Declaration of Helsinki. All

the normal subjects were screened by a question-

naire on medical history and excluded if they had

neurological, psychiatric or medical conditions that

could potentially affect the central nervous system.

We employed the Japanese version of National Adult

Reading Test (JART) as a convenient tool to

measure IQ for participants (premorbid IQ for

schizophrenics). Patients had fewer years of educa-

tion (two-sample t-test, PB0.0001), lower scores of

JART (controls: 78.8911.5, schizophrenics: 58.79

25.3, two-sample t-test PB0.001).

Neuroimaging analysis

MR studies were performed on a 1.5-Tesla Siemens

Magnetom Vision Plus system. Axial DTI scans

aligned to the plane containing anterior and poster-

ior commissures were acquired with a pulsed-gradi-

ent, spin-echo, single-shot echo planar imaging

(EPI) sequence (TR/TE�4000/100 ms, 256�256

matrix, FOV 240 mm, b�1000 s/mm2, NEX�4, 20

slices, 5 mm slice thickness, 1.5 mm gap). Diffusion

was measured along six non-collinear directions,

because six directions were maximum number of

this Vision Plus system. For each of six gradient

directions, four acquisitions were averaged. Four

acquisitions without diffusion weighting (b�0) were

also averaged. Additionally, a three-dimensional

volumetric acquisition of a T1-weighted gradient

echo sequence with a gapless series of thin sagittal

sections using an MPRage sequence (TR/TE�11.4/

4.4 ms; flip angle, 158; acquisition matrix, 256�256; NEX�1, FOV 315 mm; slice thickness 1.23

mm) was acquired for evaluating the volume of grey

matter (GM), WM and cerebrospinal fluid (CSF)

space. Seven diffusion images acquired as above by

an in-house script described previously (Mori et al.

2007) on Matlab 6.5 software (Mathworks, Inc.,

MA, USA). Then, the FA images were spatially

normalized using high-dimensional-warping algo-

rithm (Ashburner et al. 1999) and were matched

to the FA template image (Figure 1, top). To make

the FA template image, we warped FA images of

four normal subjects (other than 42 control subjects)

to the single-subject T1 template (skull stripped

image) using spatial normalization function of

SPM2 and averaged the four warped FA images.

The transformed FA images were smoothed with a

Gaussian kernel (the filter size, full-width half-

maximum: 6�6�6 mm).

Since our interest was FA changes in the basal

ganglia and thalamus, we excluded other brain areas

by using an explicit mask (Figure 1, top). The

resultant FA maps were analyzed using Statistical

parametric mapping 2 (SPM2), which implements a

‘general linear model’. To test hypotheses about

regional population effects, data were analyzed by a

two-sample t-test without global normalization.

JART scores were treated as nuisance variables.

Furthermore, we performed correlational analyses

between duration of illness, age of onset, total daily

dose of antipsychotic drugs (chlorpromazine equiva-

lent) and FA value in the schizophrenics. Our a

priori hypothesis is limited to the basal ganglia;

however, investigation of the FA changes within this

ROI is null hypothesis. Thus, we used PB0.05,

corrected for multiple comparisons with Family�Wise Error rate (FWE) within basal ganglia as a

statistical threshold.

Results

In comparison with controls, schizophrenics had

significantly lower FA values in the bilateral globus

pallidus (GP) (Figure 1, bottom). Increased FA

values in schizophrenics were not found in any

regions (data not shown).

A correlational analysis in the schizophrenics

demonstrated a significantly negative correlation

66 R. Hashimoto et al.

between duration of illness and FA in the left head of

the caudate nucleus (t value�4.77, Talairach co-

ordinate x, y, z: �11, �17, �6). However, there is

no significant correlation between duration of illness

and FA values in the GP and the thalamus. There

was no significant correlation between FA values in

the basal ganglia�thalamic system with age of onset

or total daily dose of antipsychotic drugs.

Discussion

In this study, we found a significantly reduced FA

value in the bilateral GP and left thalamus in

schizophrenics compared to controls. We consider

that reduced FA may reflect microstructural ab-

normalities in the basal ganglia�thalamic system in

schizophrenia. A previous fMRI study suggested that

GP itself may be the primary locus of the functional

deficits in the basal ganglia and may be dysfunctional

in schizophrenia (Menon et al. 2001). A postmortem

study of basal ganglia morphology reported that only

the GP were smaller in schizophrenics than in

controls (Bogerts et al. 1985). These studies indi-

cated functional and structural abnormalities in GP

in schizophrenia. Our data, reduced FA in GP in

schizophrenia, were obtained using a size-adjusted

high-dimensional warping method (Ohnishi et al.

2006). Our results, microstructural abnormalities in

the GP in schizophrenia, are consistent with pre-

vious reports.

Although the underlying mechanisms remain to

be clarified, previous DTI studies in parkinsonism

have well demonstrated ongoing pathological

changes in neurodegenerative diseases, suggesting

that this technique has the potential to detect

microstructural alterations in the basal ganglia

(Yoshikawa et al. 2004). Since pathological findings

of schizophrenia are still ambiguous, the underlying

pathological changes of reduced FA values in

schizophrenia are unclear. However, multiple lines

of evidence now converge to implicate oligoden-

droglia and myelin in schizophrenia (Davis et al.

Figure 1. Top: A half of the explicit mask is displayed onto mean FA images of warped FA images obtained from 42 controls (dark blue:

caudate nucleus; yellow: putamen; light blue: globus pallidus; red: thalamus). Even after averaging, the mean images are not blurred. Since

globus pallidus is traversed by numerous myelinated nerve fibres, it shows higher FA value than other parts of basal ganglia. Bottom: The

SPM {t} is displayed onto mean axial FA images of 42 schizophrenics. A significant reduction of FA value in schizophrenia was noted in the

bilateral globus pallidus (right GP: t value�6.52, Talairach coordinate x, y, z: 18, �2, �2, left GP: t value�6.37, Talairach coordinate x,

y, z: �18, �3, �2) and left thalamus (t value�4.96, Talairach coordinate x, y, z: �18, �33, 10).

Basal ganglia abnormality in schizophrenia 67

2003). We assume that damage of myelinated nerve

fibres may contribute to FA reduction in the basal

ganglia�thalamic system. The GP is traversed by

numerous myelinated nerve fibres that give it the

pale appearance for which it is named, and has rich

connections to the putamen and the thalamus.

These histological characteristics of the GP may

contribute to its higher FA values. A qualitative

electron microscopic study reported the density of

concentric lamellar bodies (an indicator of damage

of myelinated fibres) was dramatically increased in

the caudate nucleus in schizophrenia, as compared

to controls (Uranova et al. 2001). Such pathologi-

cal changes seem to explain decreased FA values in

the schizophrenic brain. However, there have been

no data on whether GP also have alterations of

myelinated fibres. Further pathological studies need

to be conducted to draw a firm conclusion on this

matter.

Although some studies demonstrated abnormal-

ities of GP in neuroleptic-naıve schizophrenics

(Spinks et al. 2005), abnormalities in the basal

ganglia have been considered to relate to antipsy-

chotic medication (Gur et al. 1998). In this study,

FA changes in the GP and thalamus were not

associated with the duration of illness or the daily

dose of antipsychotic drugs, suggesting that FA

changes in these regions might be independent of

medication with neuroleptics. Guidelines for the

biological treatment of schizophrenia developed by

an international Task Force of the World Federation

of Societies of Biological Psychiatry recommended

atypical antipsychotics as first line drugs (Falkai et

al. 2005, 2006). The differential treatment effects on

brain morphology could be due to typical antipsy-

chotics-associated toxicity or greater therapeutic

effects of atypical antipsychotics (Lieberman et al.

2005). It would be interesting to compare patients

treated with atypical antipsychotics to those with a

history of typical antipsychotics treatment; however;

the subgroup of patients that were only treated with

atypical antipsychotics or the subgroup of patients

that were only treated with typical antipsychotics

were too small to investigate a possible difference

between two groups in FA in our sample. To

conclude whether observed change of FA value is a

result of medication or relates to the pathophysiol-

ogy of schizophrenia itself, longitudinal studies on

treated schizophrenics, and studies on neuroleptic-

naive schizophrenics should be conducted.

There is a limitation to our study: we used a 1.5-

Tesla Siemens Magnetom Vision Plus system, which

is a relatively old system. We chose six gradient

directions, which is quite low, as this number is the

maximum number of directions in this system. Slice

thickness of 5 mm and 1.5-mm slice gaps are

methodological drawbacks to this study. The reason

why we used a slice thickness of 5 mm and 1.5-mm

slice gaps is to cover the whole brain as in our

previous study (Mori et al. 2007). There may be a

partial volume effect in our mapping parameters,

although we minimized the problem by using the

high-dimensional warping algorithm.

Our data suggest that patients with schizophrenia

might have microstructural abnormalities in globus

pallidus and thalamus. The DTI study may be a

promising method to investigate microstructural

abnormalities in schizophrenia.

Acknowledgements

We are grateful to Osamu Takizawa (Siemens) for

supporting the development of a program for

calculation of FA values.

This work was supported in part by Grants-in-

Aid from the Japanese Ministry of Health, Labor

and Welfare (H17-kokoro-007 and H16-kokoro-

002), the Japanese Ministry of Education, Culture,

Sports, Science and Technology, and Core research

for Evolutional Science and Technology of Japan

Science and Technology Agency, Japan Foundation

for Neuroscience and Mental Health.

Declaration of Interest

None.

References

Andreasen NC. 1999. A unitary model of schizophrenia: Bleuler’s

‘fragmented phrene’ as schizencephaly. Arch Gen Psychiatry

56:781�787.

Ashburner J, Andersson JL, Friston KJ. 1999. High-dimensional

image registration using symmetric priors. Neuroimage 9:619�628.

American Psychiatric Association. 1994. Diagnostic and statistical

manual of mental disorders. 4th ed. (DSM-IV). Washington,

DC: American Psychiatric Association.

Bogerts B, Meertz E, Schonfeldt-Bausch R. 1985. Basal ganglia

and limbic system pathology in schizophrenia. A morphometric

study of brain volume and shrinkage. Arch Gen Psychiatry

42:784�791.

Davis KL, Stewart DG, Friedman JI, et al. 2003. White matter

changes in schizophrenia: evidence for myelin-related dysfunc-

tion. Arch Gen Psychiatry 60:443�456.

Falkai P, Wobrock T, Lieberman J, Glenthoj B, Gattaz WF,

Moller HJ. 2005. World Federation of Societies of Biological

Psychiatry (WFSBP) guidelines for biological treatment of

schizophrenia, Part 1: acute treatment of schizophrenia. World

J Biol Psychiatry 6:132�191.

Falkai P, Wobrock T, Lieberman J, Glenthoj B, Gattaz WF,

Moller HJ. 2006. World Federation of Societies of Biological

Psychiatry (WFSBP) guidelines for biological treatment of

schizophrenia, part 2: long-term treatment of schizophrenia.

World J Biol Psychiatry 7:5�40.

Gur RE, Maany V, Mozley PD, Swanson C, Bilker W, Gur RC.

1998. Subcortical MRI volumes in neuroleptic-naive and

68 R. Hashimoto et al.

treated patients with schizophrenia. Am J Psychiatry 155:1711�1717.

Lieberman JA, Tollefson GD, Charles C, et al. 2005. Antipsycho-

tic drug effects on brain morphology in first-episode psychosis.

Arch Gen Psychiatry 62:361�370.

Menon V, Anagnoson RT, Glover GH, Pfefferbaum A. 2001.

Functional magnetic resonance imaging evidence for disrupted

basal ganglia function in schizophrenia. Am J Psychiatry

158:646�649.

Mori T, Ohnishi T, Hashimoto R, et al. 2007. Progressive changes

of white matter integrity in schizophrenia revealed by diffusion

tensor imaging. Psychiatry Res 154:133�145.

Ohnishi T, Hashimoto R, Mori T, et al. 2006. The association

between the Val158Met polymorphism of the catechol-O-

methyl transferase gene and morphological abnormalities of

the brain in chronic schizophrenia. Brain 129:399�410.

Spinks R, Nopoulos P, Ward J, Fuller R, Magnotta VA, Andreasen

NC. 2005. Globus pallidus volume is related to symptom

severity in neuroleptic naive patients with schizophrenia.

Schizophr Res 73:229�233.

Uranova N, Orlovskaya D, Vikhreva O, et al. 2001. Electron

microscopy of oligodendroglia in severe mental illness. Brain

Res Bull 55:597�610.

Yoshikawa K, Nakata Y, Yamada K, Nakagawa M. 2004. Early

pathological changes in the parkinsonian brain demonstrated

by diffusion tensor MRI. J Neurol Neurosurg Psychiatry

75:481�484.

Basal ganglia abnormality in schizophrenia 69

CASE REPORT

Catatonia as a risk factor for the development of neuroleptic malignantsyndrome: Report of a case following treatment with clozapine

THOMAS PAPARRIGOPOULOS, ELIAS TZAVELLAS, PANAGIOTIS FERENTINOS,

IRAKLIS MOURIKIS & JOHN LIAPPAS

Department of Psychiatry, Athens University Medical School, Eginition Hospital, Athens, Greece

AbstractCatatonia is characterized by the predominance of psychomotor abnormalities and shares many clinical, biological andtreatment response features with the neuroleptic malignant syndrome (NMS), a rare adverse reaction to psychoactivemedications. It has been advocated that the two conditions should be placed along the same spectrum of disorders. A case ofa 49-year-old woman, who developed NMS while on low dose clozapine soon after recovering from catatonia, is presented.The potential relationship between catatonia and NMS is discussed in the light of the existing literature, and attention isdrawn to the risk for clozapine-induced NMS in catatonic patients.

Key words: Catatonia, clozapine, neuroleptic malignant syndrome, risk factor

Introduction

Catatonia was first described by Karl Ludwig

Kahlbaum (Kahlbaum 1874) as a psychomotor

syndrome with motor, affective and behavioural

symptoms, such as akinesia, catalepsy, stupor, mut-

ism, negativism, perseveration, stereotypies, man-

nerisms, echo-phenomena, waxy flexibility, and

automatic obedience (Fink and Taylor 2001). At

the beginning of the 20th century Kraepelin and

Bleuler subsumed catatonia under dementia praecox

considering it as a motoric subtype of schizophrenia.

Consequently, because of the predominance of

movement disturbances, catatonia was regarded by

several investigators as a type of extrapyramidal

disorder and its pathophysiology has been related

to the basal ganglia (Kleist 1943).

Neuroleptic malignant syndrome (NMS) is a rare

(0.2% of psychiatric patients) adverse reaction to

psychoactive medications, particularly neuroleptics

(Caroff and Mann 1991). NMS typically presents

with signs of altered mental state, muscle rigidity,

tremor, tachycardia, and hyperpyrexia, with conco-

mitant elevated serum creatine kinase and leucocy-

tosis. NMS has been related to all neuroleptic

medications including the atypicals (Hasan and

Buckley 1998; Ananth et al. 2004; Chakraborthy

and Johnston 2004; Leibold et al. 2004; Zalsman et

al. 2004; Dew et al. 2005; Franzen et al. 2006;

Kobayashi et al. 2006; Norgard and Stark 2006).

Purportedly, the condition is attributed to an abrupt

and profound reduction in dopaminergic neuro-

transmission, caused by dopamine-blocking drugs

(Lazarus et al. 1989; Mann et al. 2000; Wang and

Hsieh 2001).

We present the case of a 49-year-old woman who

developed NMS while on low-dose clozapine, soon

after recovering from catatonia. In this context, the

potential relationship between catatonia and NMS is

discussed, and attention is drawn to the risk for

clozapine-induced NMS in such patients.

Case report

A 49-year-old married woman was admitted to our

hospital for an acute psychotic exacerbation with a

clinical picture dominated by marked psychomotor

disturbance. The patient manifested catalepsy, ne-

gativism, as evidenced by motiveless resistance to

instructions, maintenance of a rigid posture,

and mutism, as well as stereotyped movements,

Correspondence: Thomas Paparrigopoulos, MD, Assistant Professor of Psychiatry, Athens University Medical School, Department of

Psychiatry, Eginition Hospital, 74, Vas. Sofias Ave. 115 28 Athens, Greece. Tel/Fax: �30 210 72 89 324. E-mail: Thomas.

Paparrigopoulos@iop.kcl.ac.uk & egslelabath@hol.gr

The World Journal of Biological Psychiatry, 2009; 10(1): 70�73

(Received 23 January 2007; accepted 15 February 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701287369

mannerisms, and echo-phenomena; delusional ideas

of reference, neologisms and thought blocking could

also be detected whenever communication with the

patient was somehow possible. From her history, a

first psychotic episode had occurred 15 years pre-

viously, with two more recent hypomanic episodes 5

and 2 years previously. Two suicide attempts were

also recorded, 10 and 2 years previously; the latter

took place after her son shot himself. DSM-IV-TR

diagnoses either of schizophrenia, catatonic type, or

of schizoaffective disorder, bipolar type, with current

catatonic symptoms, were considered to be the most

pertinent (American Psychiatric Association 2000).

Physical examination and routine laboratory tests at

admission were normal. Because the patient’s psy-

chotic symptoms had been reported refractory to

several previously tried antipsychotics, and clozapine

had not been previously associated with the devel-

opment of NMS when administered to patients with

catatonic symptoms, treatment with clozapine was

initiated and slowly titrated upward to 200 mg/day.

A remarkable improvement of catatonic and psycho-

tic symptoms was observed within 10 days. How-

ever, 15 days later, the patient suddenly developed

severe hypertonia, ‘‘cog-wheel’’ rigidity, a flexed

posture, and tremor. Furthermore, she had elevated

body temperature (38.1�38.78C) and she was con-

fused and disoriented in time and place. Laboratory

tests showed leucocytosis (WBC 14600/ml), elevated

creatine kinase (CK: 1960 U/l; normal values: 24�190 U/l) and increased aldolase (14.7 U/l; normal

values: 1.5�12.0 U/l). The remaining serum chem-

istry, including CK-MB, was within the normal

range. No source of infection or other clinical signs

of pathology were found. A diagnosis of NMS was

suggested and treatment with i.m. diazepam (15 mg/

day) was started. Following 24 h of treatment,

clinical symptoms significantly subsided and bio-

chemical indices improved (WBC: 12800/ml, CK:

1275 U/l, aldolase: 13.2 U/l). Diazepam was stopped

and switched to lorazepam (7.5 mg/day per os, in

divided doses). The patient fully recovered and all

laboratory measures returned to normal after

2 weeks. Treatment with 4 mg/day lorazepam was

continued and the patient was discharged from the

hospital.

Discussion

Catatonia and NMS have been reported to share

clinical features (Fink 1996; Carroll and Taylor

1997), biological characteristics, i.e. neurochemical,

electrophysiological findings (Carroll and Goforth

1995; Jahanshahi et al. 1995; Mann et al. 2000;

Nisijima and Ishiguro 1995; Northoff et al. 1996;

Northoff et al. 2000; Wang and Hsieh 2001), and

treatment response patterns (Lew and Tollefson

1983; Kontaxakis et al. 1988; Miyaoka et al. 1997;

Koch et al. 2000). Therefore, many investigators

advocate that the many clinical and pathophysiolo-

gical similarities between the two conditions place

both of them along the same spectrum of disorders

(White 1992; Fink 1996).

Neuroleptic malignant syndrome and catatonia

not only may present with a similar clinical picture,

but their severity may also widely vary. For instance,

the symptoms of catatonia may range from mild to

lethal. Especially, a severe form of catatonia known

as malignant or lethal catatonia is considered by

many authors to be clinically inseparable from NMS

(White 1992; Fink 1996; Carroll and Taylor 1997).

Furthermore, the origin of catatonic symptoms has

been regarded either as spontaneous/psychogenic or

induced by neuroleptics (Woodbury and Woodbury

1992; Tsai et al. 2005). On the other hand, mild

forms of NMS may be difficult to distinguish from

neuroleptic-induced catatonia (Kontaxakis et al.

1990), which, if left untreated, may progress to

NMS (Woodbury and Woodbury 1992). Conse-

quently, it has been proposed that these two condi-

tions are likely to be part of a so-called ‘neuroleptic

toxicity spectrum’ (Tsai et al. 2005). However, it is

noteworthy that DSM-IV-TR does not provide a

separate diagnostic category for neuroleptic or drug-

induced catatonic disorder.

In terms of biological findings, elevated CK is

present both in catatonia and NMS (Northoff et al.

1996). However, massive serum CK increases have

been reported with atypical antipsychotic drugs in

the absence of other signs of NMS (Meltzer 2000).

In addition, similar electroencephalographic ab-

normalities have been reported in both conditions

(Lazarus et al. 1989; Caroff et al. 1991; Carroll and

Boutros 1995).

The neurochemical basis of NMS and catatonia

has not been clarified; multiple neurotransmitter

systems may be involved. Dopaminergic and GA-

BAergic pathways have been implicated in the

pathophysiology of catatonia (Wetzel et al. 1987;

Fricchione 1989; Lazarus et al. 1989; Caroff et al.

1991; White 1992). The relation of catatonia to

dopamine is supported by the clinical observation

that behavioural symptoms of catatonia are consis-

tent with distinct dysfunction of the frontal lobes

mediated by dopaminergic neurotransmission (Tay-

lor 1990). On the other hand, motor disturbances in

catatonia may result from reduced GABAergic

activity in specific brain areas; this is consistent

with the robust responsiveness of catatonia to

electroconvulsive therapy (Mann et al. 1986; Phil-

brick and Rummans 1994) and benzodiazepine

administration (Francis et al. 1997), which has also

Catatonia and the development of neuroleptic malignant syndrome 71

been reported for NMS (Kontaxakis et al. 1988;

Miyaoka et al. 1997).

This report adds to the literature of cases in which

NMS is preceded by symptoms of catatonia (White

and Robins 1991; White 1992; Raja et al. 1994). In a

total of nine cases, aged 20�61 years, mostly

females, neuroleptics were administered to treat

catatonic and other psychotic symptoms; however,

they worsened catatonia or converted it to NMS

(Table I). Therefore, NMS could be considered as a

neuroleptic-induced exacerbation of pre-existing

catatonia (Fink and Taylor, 2003; Caroff et al.

2004). As shown in the table, it is remarkable that

both high- and low-potency neuroleptics, clozapine

excluded, have been implicated in the development

of NMS, that the time from neuroleptic administra-

tion to the emergence of NMS was quite variable

(from 6 hours to 22 days), and that no diagnosis of

schizophrenia was recorded. To the authors’ best

knowledge, this is the first case in the literature

reporting that catatonic symptoms can exacerbate

and progress to NMS following clozapine adminis-

tration, which has been rarely associated with NMS

(Sachdev et al. 1995).

A possible explanation of the temporal association

between catatonia and NMS is that both conditions

are different variants of the same disorder or parts of

a spectrum; therefore, catatonia should also be

considered as a potential risk factor for the develop-

ment of NMS. This point of view is corroborated by

clinical observations indicating that NMS may pre-

sent either with or without catatonic symptoms, with

different underlying pathophysiology and treatment

response (Carroll and Lee 2004); furthermore,

catatonic symptoms have been seldom reported to

follow NMS after it subsides (Dent 1995; Caroff

2000). An alternative explanation might be that

various nonspecific factors typically associated with

catatonia, such as dehydration, agitation, physical

exhaustion, presence of an affective disorder, or

concurrent medical illness, are also risk factors for

NMS (Rosebush and Stewart 1989).

In conclusion, catatonia should be suspected as a

potential risk factor for the development of NMS

and therefore should be treated with caution.

Finally, clinicians should be alerted that even

clozapine might exacerbate catatonic symptoms

and precipitate progression to NMS.

Acknowledgements

None.

Declaration of Interest

None.

References

American Psychiatric Association. 2000. Diagnostic and statistical

manual of mental disorders (DSM-IV-TR), 4th ed. Text

revision. Washington, DC: American Psychiatric Association.

Ananth J, Parameswaran S, Gunatilake S. 2004. Side effects of

atypical antipsychotic drugs. Curr Pharm Des 10:2219�2229.

Caroff SN, Mann SC, Lazarus A, Sullivan K, MacFadden W.

1991. Neuroleptic malignant syndrome: Diagnostic issues.

Psychiatr Ann 21:130�147.

Caroff SN, Mann SC, Keck PE Jr, Francis A. 2000. Residual

catatonic state following neuroleptic malignant syndrome. J

Clin Psychopharmacol 20:257�259.

Caroff SN, Francis A, Fricchione GL, Mann SC, editors. 2004.

Catatonia: From psychopathology to neurobiology. Washing-

ton, DC: American Psychiatric Press.

Caroll BT, Boutros NN. 1995. Clinical electroencephalograms in

patients with catatonic disorders. Clin Electroencephalogr

26:60�64.

Caroll BT, Goforth HW. 1995. Serum iron in catatonia. Biol

Psychiatry 38:776�777.

Carroll BT, Lee JW. 2004. Catatonia is a risk factor for

neuroleptic malignant syndrome. J Clin Psychiatry 65:1722�1723.

Caroll BT, Taylor RE. 1997. The non dichotomy between lethal

catatonia and neuroleptic malignant syndrome. J Clin Psycho-

pharmacol 17:235�238.

Chakraborty N, Johnston T. 2004. Aripiprazole and neuroleptic

malignant syndrome. Int Clin Psychopharmacol 19:351�353.

Dent J. 1995. Catatonic syndrome following recovery from

neuroleptic malignant syndrome. J Intellect Disabil Res 39(Pt

5):457�459.

Table I. Published cases of catatonia preceding NMS.

Reference Age Sex Psychiatric diagnosis* Neuroleptic used Time**

White and Robins (1991) 25 F � Clothiapine 80 mg i.v. 72 h

20 M � Clothiapine 80 mg i.v. 36 h

20 M � Clothiapine 80 mg i.v. Few h

47 F � Chlorpromazine 100 mg i.m. 48 h

33 F � Clothiapine 80 mg i.v. 72 h

White (1992) 33 F � Haloperidol 10 mg i.v. 6 h

Raja et al. (1994) 23 F Psychotic depression Haloperidol 6�8 mg 192 h

61 M Bipolar disorder Clothiapine 50 mg p.o., haloperidol 4 mg i.v. 96 h

59 F Psychotic depression Haloperidol 2 mg p.o. 22 days

*Apart from catatonia.

**Time elapsed between neuroleptic administration and emergence of NMS.

72 T. Paparrigopoulos et al.

Dew RE, Rosenquist PB, McCall WV. 2005. Aripiprazole for

agitation in a 13-year-old girl with neuroleptic malignant

syndrome. Int J Adolesc Med Health 17:187�188.

Fink M. 1996. Neuroleptic malignant syndrome and catatonia.

One entity or two? Biol Psychiatry 39:1�4.

Fink M, Taylor MA. 2001. The many varieties of catatonia. Eur

Arch Psychiatry Clin Neurosci 251 (Suppl):I8�13.

Fink M, Taylor MA. 2003. Catatonia: A clinician’s guide to

diagnosis and treatment. Cambridge: Cambridge University

Press.

Francis A, Divadeenam K, Petrides G. 1997. Advances in the

diagnosis and treatment of catatonia. Convuls Ther 12:259�261.

Franzen D, Burkhard J, Corti N, Schupbach D, Fontanel D,

Staubli M. 2006. Neuroleptic malignant syndrome after 30

years treatment with clozapine: a rarely seen differential

diagnosis on intensive care units. Anasthesiol Intensivmed

Nottfallmed Schmertzher 41:125�127.

Fricchione G. 1989. Catatonia: a new indication for benzodiaze-

pines? Biol Psychiatry 26:761�765.

Hasan S, Buckley P. 1998. Novel antipsychotics and the neuro-

leptic malignant syndrome: a review and critique. Am J

Psychiatry 155:1113�1116.

Jahanshahi M, Jenkins H, Brown RG, Marsden CD. 1995. Self-

initiated versus externally triggered movements. Brain

118:913�933.

Kahlbaum KL. 1874. Die Katatonie oder das Spannungsirresein.

Berlin: Verlag August Hirshwald.

Kleist K. 1943. Die Katatonie. Nervenarzt 16:1�10.

Kobayashi A, Kawanishi C, Matsumura T, et al. 2006. Quetia-

pine-induced neuroleptic malignant syndrome in dementia

with Lewy bodies: a case report. Prog Neuropsychopharmacol

Biol Psychiatry 30:1170�1172.

Koch M, Chandragiri S, Rizvi S, Petrides G, Francis A. 2000.

Catatonic signs in neuroleptic malignant syndrome. Compr

Psychiat 41:73�75.

Kontaxakis VP, Vaidakis NM, Christodoulou GN, Valergaki HC.

1990. Neuroleptic-induced catatonia or a mild form of

neuroleptic malignant syndrome? Neuropsychobiology 23:

38�40.

Kontaxakis VP, Christodoulou GN, Markidis MP, Havaki-Kon-

taxaki BJ. 1988. Treatment of a mild form of neuroleptic

malignant syndrome with oral diazepam. Acta Psychiatr Scand

78:396�398.

Lazarus A, Mann SC, Caroff SN, editors. 1989. The Neuroleptic

malignant syndrome and related conditions. Washington, DC:

American Psychiatric Press.

Leibold J, Patel V, Hasan RA. 2004. Neuroleptic malignant

syndrome associated with ziprasidone in an adolescent. Clin

Ther 26:1105�1108.

Lew TY, Tollefson G. 1983. Chloropromazine-induced neurolep-

tic syndrome and its response to lorazepam. Biol Psychiatry

18:1441�1446.

Mann SC, Caroff SN, Bleier JR, Welz WKR, Kling MA,

Hayashida M. 1986. Lethal catatonia. Am J Psychiatry

143:1374�1381.

Mann SC, Caroff SN, Fricchione G, Campbell C. 2000. Central

dopamine hypoactivity and the pathogenesis of neuroleptic

malignant syndrome. Psychiatr Ann 30:363�374.

Meltzer HY. 2000. Massive serum creatine kinase increases with

atypical antipsychotic drugs: what is the mechanism and the

message? Psychopharmacology (Berl) 150:349�350.

Miyaoka H, Shishkura K, Otsubo T, Moramatsu D, Kamijima K.

1997. Diazepam-responsive neuroleptic malignant syndrome: a

diagnostic subtype? Am J Psychiatry 153:882 (letter).

Nisijima K, Ishiguro T. 1995. Cerebrospinal fluid levels of

monoamine metabolites and gamma-aminobutyric acid in

neuroleptic malignant syndrome. J Psychiatr Res 29:233�244.

Norgard NB, Stark JE. 2006. Olanzapine-associated neuroleptic

malignant syndrome. Pharmacotherapy 26(8):1180�1182.

Northoff G, Wenke J, Pflug B. 1996. Increase of serum creatinine

phosphokinase in catatonia: an investigation in 32 acute

catatonic patients. Psychol Med 26:547�553.

Northoff G, Pfennig A, Krug M, Leschinger A, Borgets B. 2000.

Delayed onset of late movement-related cortical potentials and

abnormal response to lorazepam in catatonia. Schizophr Res

44:193�211.

Philbrick KL, Rummans TA. 1994. Malignant catatonia. J

Neuropsychiatry Clin Neurosci 6:1�13.

Raja M, Altavista MC, Cavallari S, Lubich L. 1994. Neuroleptic

malignant syndrome and catatonia. A report of three cases. Eur

Arch Psychiatry Clin Neurosci 243:299�303.

Rosebush P, Stewart T. 1989. A prospective analysis of 24

episodes of neuroleptic malignant syndrome. Am J Psychiatry

146:717�725.

Sachdev P, Kruk J, Kneebone M, Kissane D. 1995. Clozapine-

induced neuroleptic malignant syndrome: review and report of

new cases. J Clin Psychopharmacol 15:365�371.

Taylor MA. 1990. Catatonia. A review of the behavioural

neurologic syndrome. Neuropsychiatr Neuropsychol Behav

Neurol 3:48�72.

Tsai JH, Yang P, Yen JY, Chen CC, Yang MJ. 2005. Zotepine-

induced catatonia as a precursor in the progression to

neuroleptic malignant syndrome. Pharmacotherapy 25:1156�1159.

Wang HC, Hsieh Y. 2001. Treatment of neuroleptic malignant

syndrome with subcutaneous apomorphine monotherapy. Mov

Disord 16:765�767.

Wetzel H, Heuser I, Benkert O. 1987. Stupor and affective state:

alleviation of the psychomotor disturbances by lorazepam and

recurrence of symptoms after Ro 15-1788. J Nerv Ment Dis

175:240�242.

White DAC. 1992. Catatonia and the neuroleptic malignant

syndrome � a single entity? Br J Psychiatry 161:558�560.

White DAC, Robins AH. 1991. Catatonia: harbinger of the

neuroleptic malignant syndrome. Br J Psychiatry 158:419�421.

Woodbury MM, Woodbury MA. 1992. Neuroleptic-induced

catatonia as a stage in the progression toward neuroleptic

malignant syndrome. J Am Acad Child Adolesc Psychiatry

31:1161�1164.

Zalsman G, Lewis R, Konas S, Loebstein O, Goldberg P,

Burguillo F, et al. 2004. Atypical neuroleptic malignant

syndrome associated with risperidone treatment in two adoles-

cents. Int J Adolesc Med Health 16:179�182.

Catatonia and the development of neuroleptic malignant syndrome 73

LETTER TO THE EDITOR

Adjunctive topiramate treatment for a refractory familialadolescent mania

FOR-WEY LUNG1�3, CHUN-LIN LIU4, CHIEN-SHU WANG1 &

DONG-SHENG TZENG1,5

1Department of Psychiatry, Military Kaohsiung General Hospital, Kaohsiung, Taiwan, 2Graduate Institute of Behavior

Sciences, Kaohsiung Medical University, Taiwan, 3Department of Psychiatry, National Defense Medical University, Taipei,

Taiwan, 4Tsyr-Huey (Loving) Mental Hospital, Kaohsiung Jen-Ai (Love) Homes, Kaohsiung County, Taiwan, and5Graduate Institute of Occupational Safety and Health, Kaohsiung Medical University, Kaohsiung, Taiwan

AbstractAdolescent mania is often misdiagnosed. This case study describes the clinical course and diagnostic reclassification fromschizophrenia to bipolar disorder in a 15-year-old girl. This case study also describes the pedigree of the siblings, familialaggregation, and anticipation of mood disorders. In addition, we present the successful use of topiramate, a newantiepileptic drug, which is increasingly being used as a mood stabilizer in paediatric bipolar disorder. The efficacy oftopiramate in this case supports its role as a promising agent in treatment-resistant adolescent mania associated with familialaggregation.

Key words: Adolescent mania, diagnostic reclassification, familial aggregation, topiramate

Introduction

Juvenile mania is controversial because of ques-

tions regarding the reliability and validity of the

diagnosis. In 20�30% of patients with bipolar I

disorder, their first depressive episode is before the

age of 20. Traditionally, the diagnostic criteria for

adult mania have been applied to diagnose juvenile

bipolar disorder. However, while diagnosing juve-

nile depression, it is difficult to determine whether

the first episode of depression will follow a bipolar

or a unipolar course (Jack 2000). Children with

bipolar disorder often present with an ‘‘atypical’’

clinical picture, particularly when the assessment

relies on the adult criteria, with irritability, mixed

presentation, and chronicity (Wozniak et al. 2001).

The presentation of hypomania or mania may

differ between children and adults. There is

considerable disagreement about the most appro-

priate diagnosis for children with a constellation of

chronic and severe irritability, hyperactivity, and

abnormal mood, typically characterized by sadness

and anger.

Juvenile bipolar I disorder is commonly comorbid

with other psychiatric and behavioural disorders.

Comorbidity with attention-deficit/hyperactivity dis-

order, and the overlapping symptomatology, often

complicate diagnosis. The different treatment stra-

tegies for these diagnoses can result in conflicting

treatments (Giedd 2000). These factors make it

likely that large numbers of patients are either not

being diagnosed or are being misdiagnosed. The lack

of a well-established longitudinal history may in-

crease the miscategorization of psychotic mania as

schizophrenia (Calderoni et al. 2001).

A novel antiepileptic, topiramate, whose possible

mechanism of action is through the blockade of

sodium channels and glutamate-mediated neuroex-

citation (Bourgeois 1998), has shown efficacy in the

treatment of paediatric affective disorder (DelBello

et al. 2002; Schuler-Springorum et al. 2002). These

findings suggest that juvenile treatment resistance

and familial mania may be stabilized by topiramate.

Here, we present a case of successful treatment of

the use of topiramate in a refractory juvenile bipolar

disorder.

Correspondence: For-Wey Lung, MD, ScD, Department of Psychiatry, Military Kaohsiung General Hospital, No. 2 Chung Cheng 1st Rd.,

Kaohsiung City, Taiwan. Tel: �886 7 7490056. Fax: �886 7 7490786. E-mail: forwey@seed.net.tw

The World Journal of Biological Psychiatry, 2009; 10(1): 74�77

(Received 24 December 2006; accepted 8 February 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701268864

Case report

The proband, Ms Shu, is a 15-year-old single female

patient. She is 165 cm tall and weighs 65 kg. In her

premorbid state, she was characterized as an intro-

vert and weak-willed. Her academic performance

was excellent. At the age of 13, she was selected as

the leader of her class. Because of conflicts with her

classmates, she became frustrated. She then devel-

oped dysphoria, insomnia, poor self-confidence, and

poor attention. She attempted to commit suicide

once. In April 2002, she began to complain of

nausea, vomiting, and headache. She was first

introduced to Pediatrics LMD for treatment, yet

she showed no improvement. In July 2002, she was

referred to the psychiatric department in a medical

centre where major depression was diagnosed. She

received six courses of electroconvulsive treatment

(ECT). Labile mood, suicide ideation, and self-

injury persisted after the above treatment. In Octo-

ber 2002, she first reported auditory hallucinations

(AHs) consisting of a second-person voice making

comments. She was admitted again and another

psychiatrist diagnosed her with schizophrenia. After

treatment with sulpride, 200 mg 4�QN for 1 week,

the above symptoms partially improved during

the period of admission. After leaving hospital, she

was still adhering well to sulpiride treatment at

home. Unfortunately, 1 month after discharge, the

patient began talking to herself and began exhibit-

ing repetitive speech. She was then evaluated at

another medical center, where she received quetia-

pine 100 mg 2�QN, quetiapine 25 mg 1�TID,

and paroxetine 20 mg 1�BID for 3 months. How-

ever, no significant improvement in symptoms was

noted. In March 2003, the patient was admitted to

our psychiatric ward for further evaluation and

treatment.

Intervention procedure and results

Ms Shu’s initial presentation to our ward included

labile mood, irritability, poor impulse control, AH,

strong suicide ideation, and hyperactivity. She re-

ceived five courses of ECT treatments in the first

week. However, the above symptoms persisted.

Further systemic review and psychological exams

were done. She had no systemic disease, surgical

history, or history of substance abuse. Physical

examination and neurological examination were

essentially negative. Laboratory examination and

neuroradiological examination were also within the

normal range. WISC-III was performed and her

total IQ was 102. B�G test showed no evidence of

brain injury. WCST also showed no impairment in

abstract thinking. A comprehensive review of her

history was done. Most of the patient’s symptoms

were mood shift and behavioural problems. Irrit-

ability was her main symptom. Furthermore, she

was noted with no first-rank symptoms. The diag-

nosis was revised to bipolar I disorder with manic

episodes. Treatment was revised to lithium 300 mg

1�TID, carbamazepine 200 mg 1�TID, alprazo-

lam 1 mg 2�TID, and zolpidem 10 mg 2�QN. No

improvement was evident after 6 weeks of treatment.

Topiramate 600 mg/day was added to the above

regimen. Four weeks later, irritability and other

symptoms had almost completely subsided. Ms

Shu returned to school and graduated from junior

high school in 2004. She continued regular follow-

up at our ambulatory clinic for 2 years. Her parents

reported that her mental status, social skills, inter-

personal relationships, and school performance re-

mained normal.

Ms Shu’s elder sister, 18 years old, was also

diagnosed with bipolar I disorder with onset at the

age of 12. Her younger sister was aged 11. Both her

sisters also received treatment at our psychiatry

department. All the three siblings had similar

symptoms and course, including irritable mood,

severe agitation, violent behaviour, persecutory de-

lusion, and AH. At first, the siblings were treated for

paranoid schizophrenia. However, drug response

was poor until the diagnosis was changed to juvenile

mania and topiramate was added to the treatment

regimen.

Further investigation of the family medical history

indicated that her maternal grandfather, 71 years

old, had been diagnosed with major depression at

age 68 following the death of his wife. In addition,

Ms Shu’s maternal grandmother, 63 years old, was

diagnosed with major depression at age 58 following

symptoms of dyspnoea, general weakness, loss of

interest, and suicide ideation. Ms Shu’s mother,

aged 42, was diagnosed with major depression at the

age of 39. The family pedigree is shown in Figure 1.

Discussion

The strong family distribution of mood disorder in

our case is similar to those reported by Rice et al.

(1987), who found significant heterogeneity for sex-

specific correlations using a multifactorial analysis.

They found that same-sex siblings had a higher

correlation for psychiatric disorders than opposite-

sex siblings. The siblings in our case study belong to

the same age-at-onset subgroup, identified by ad-

mixture analysis in the study of Bellivier et al.

(2003). Age of onset has been shown to be corre-

lated in affected siblings, suggesting that some

familial vulnerability factors may be age specific

Anticipation in juvenile mania 75

(Leboyer et al. 1998). Siblings also often have the

same degree of psychosis (O’Mahony et al. 2002).

The sibling’s mother and grandparents were all

diagnosed with major depression, although their

clinical course and symptoms are not well known.

It is possible they also have bipolar I disorder. Judd

et al. (2003) reported that the longitudinal sympto-

matic course of bipolar I disorder is chronic and

dominated by depressive rather than hypomanic or

cycling/mixed symptoms. Gershon et al. (1982)

found that anger is a common symptom in bipolar

depression and can lead to the misdiagnosis of major

depression. Studies have shown that 40% of patients

with bipolar depression who express symptoms of

anger are misdiagnosed with major depression

(Ghaemi et al. 2000). Studies have indicated that

further examination of patients with major depres-

sion in order to detect a history of mania can

decrease the chances of misdiagnosis.

Familial aggregation of psychotic symptoms has

been established (Potash et al. 2001). It is signifi-

cantly more common for psychotic probands to have

at least one relative with an affective disorder with

psychotic symptoms, than probands without psy-

chosis. Cluster analysis of psychotic subjects across

all families revealed significant familial aggregation.

The fragile X syndrome and Huntington’s disease

are characteristic of these disorders with an increas-

ing severity and a decreasing age of onset over

succeeding generations. This phenomenon, known

as ‘‘anticipation’’ is present in bipolar disorder (Rich

1995). An intergenerational difference in the age

of onset, between 9 and 13.5 years, and an increase

in the severity of symptoms have been documented

in bipolar disorder (McInnis et al. 1993). Previous

studies showed that anticipation of bipolar I dis-

order predominantly occurred on the paternal side

(Grigoroiu-Serbanescu et al. 1997). However, our

study found no evidence for male�male, parent�offspring transmission, as reported by Pekkarinen

et al. (1994).

Multiple phenotypes of juvenile mania exist,

including a narrow type exhibited by patients who

meet the full DSM-IV diagnostic criteria for hypo-

mania or mania, plus the duration and the hallmark

symptoms of elevated mood or grandiosity. The

intermediate phenotypes are (1) hypomania or

mania not otherwise specified, in which the patient

has clear episodes and hallmark symptoms, but the

episodes are between 1 and 3 days in duration, and

(2) irritable hypomania or mania, in which patients

have demarcated episodes with irritable but not

elevated mood. The broad phenotype is exhibited

by patients with a chronic, nonepisodic illness,

which does not include the hallmark symptoms of

mania but shares the narrower phenotype and

symptoms of severe irritability and hyperarousal

(Leibenluft et al. 2003).

The common and predominant symptoms in our

three juvenile cases of the affective disorder were

behavioural problems and irritability. AH, consisting

only of a second-person voice making comments,

was noted exclusively in the active stage with mood

factors. The majority of parents of children with

affective disorders report that their children have a

mixture of abnormal mood states, such as irritability,

euphoria, and depression, for more than 75% of the

day (Wozniak et al. 2001). In fact, the predromal

symptoms of mania in adolescence tend to be

behavioural (Lam et al. 2001). Therefore, this case

study highlights that clinicians should evaluate

unusual affective and behavioural changes when

diagnosing juvenile mania.

Although, this single-case trial could be a limita-

tion in this study, her two sisters had similar

symptoms and treatment courses. This provides

evidence that topiramate is efficacious and well

tolerated as an adjunctive treatment for refractory,

familial, and adolescent mania. Furthermore, we

found that neither the proband nor her sibling

showed any weight gain with this treatment. This

may be due to the side effect of reduced appetite

associated with topiramate. Adolescents with mood

disorders who have gained weight on other neuro-

leptic drugs may benefit from topiramate for both

mood stabilization and body weight control (Guille

and Sachs 2002; Schuler-Springorum et al. 2002).

However, further randomly controlled trials are still

necessary for topiramate use in juvenile mania.

8

1 2 3 4

6

7 9

Figure 1. The diagnoses and family pedigree of mood disorder in

Shu’s families. (2) A 63-year-old female diagnosed with major

depression at age 58; (3) a 71-year-old male diagnosed with major

depression after the death of his wife; (6) a 42-year-old female

diagnosed with major depression at age 39; (7) the elder sister of

Ms Shu, the proband, 18 years old, diagnosed with bipolar I

disorder with onset at the age of 12 (the predominant syndrome

and the course were similar to those of Ms Shu); (8) Proband, Ms

Shu, 15 years old, diagnosed with bipolar I disorder based on the

clinical course; (9) the younger sister of Ms Shu, 11 years old, also

diagnosed as having bipolar I disorder, her predominant syndrome

and disease course are similar to the proband.

76 F.-W. Lung et al.

Acknowledgements

None.

Declaration of Interest

None.

References

Bellivier F, Golmard JL, Rietschel M, et al. 2003. Age at onset in

bipolar I affective disorder: further evidence for three sub-

groups. Am J Psychiatry 160:999�1001.

Bourgeois BF. 1998. New antiepileptic drugs. Arch Neurol

55:1181�1183.

Calderoni D, Wudarsky M, Bhangoo R, et al. 2001. Differentiat-

ing childhood-onset schizophrenia from psychotic mood dis-

orders. J Am Acad Child Adolesc Psychiatry 40:1190�1196.

DelBello MP, Kowatch RA, Warner J, et al. 2002. Adjunctive

topiramate treatment for pediatric bipolar disorder: A retro-

spective chart review. J Child Adolesc Psychopharmacol

12:323�330.

DelBello MP, Findling RL, Kushner S, et al. 2005. A pilot

controlled trial of topiramate for mania in children and

adolescents with bipolar disorder. J Am Acad Child Adolesc

Psychiatry 44:539�547.

Gershon ES, Hamovit J, Guroff JJ, et al. 1982. A family study of

schizoaffective, bipolar I, bipolar II, unipolar, and normal

control probands. Arch Gen Psychiatry 39:1157�1167.

Ghaemi SN, Boiman EE, Goodwin FK. 2000. Diagnosing bipolar

disorder and the effect of antidepressants: A naturalistic study. J

Clin Psychiatry 61:804�808.

Giedd JN. 2000. Bipolar disorder and attention-deficit/hyperac-

tivity disorder in children and adolescents. J Clin Psychiatry

61(Suppl l9):31�34.

Green T, Shoval G, Weizman A. 2005. The treatment of mood

stabilizers in children and adolescents suffering from bipolar

affective disorder. Harefuah 144:810�815.

Grigoroiu-Serbanescu M, Wickramaratne PJ, Hodge SE, Milea S,

Milhailescu R. 1997. Genetic anticipation and imprinting in

bipolar I illness. Br J Psychiatry 170:162�166.

Guille C, Sachs G. 2002. Clinical outcome of adjunctive

topiramate treatment in a sample of refractory bipolar patients

with comorbid conditions. Prog Neuropsychopharmacol Biol

Psychiatry 26:1035�1039.

Jack AG. 2000. Biological therapies. In: Sadock BJ, Sadock VA,

editors. Comprehensive textbook of psychiatry. New York:

LW & W Press.

Judd LL, Akiskal HS, Schettler PJ, et al. 2003. A prospective

investigation of the natural history of the long-term weekly

symptomatic status of bipolar II disorder. Arch Gen Psychiatry

60:261�269.

Lam D, Wong G, Sham P. 2001. Prodromes, coping strategies and

course of illness in bipolar affective disorder�a naturalistic

study. Psychol Med 31:1397�1402.

Leboyer M, Bellivier F, McKeon P, et al. 1998. Age at onset

and gender resemblance in bipolar siblings. Psychiatry Res

81:125�131.

Leibenluft E, Charney DS, Towbin KE, Bhangoo RK, Pine DS.

2003. Defining clinical phenotypes of juvenile mania. Am J

Psychiatry 160:430�437.

McInnis MG, McMahon F, Chase G, Simpson S, Ross C,

DePaulo JR Jr. 1993. Anticipation in bipolar affective disorder.

Am J Hum Genet 53:385�390.

O’Mahony E, Corvin A, O’Connell R, et al. 2002. Sibling pairs

with affective disorders: Resemblance of demographic and

clinical features. Psychol Med 32:55�61.

Pekkarinen P, Bredbacka PE, Terwilliger J, Hovatta I, Lonnqvist J,

Peltonen L. 1994. Evidence for a susceptibility locus for

manic-depressive disorder in Xq26. Am J Hum Genet

55(Suppl 3):A27.133.

Potash JB, Willour VL, Chiu YF, et al. 2001. The familial

aggregation of psychotic symptoms in bipolar disorder pedi-

grees. Am J Psychiatry 158:1258�1264.

Rich T. 1995. Genetic linkage studies of bipolar disorder. Curr

Opin Psychiatry 8:3�6.

Rice J, Reich T, Andreasen NC, et al. 1987. The familial

transmission of bipolar illness. Arch Gen Psychiatry 44:

441�447.

Schuler-Springorum M, Hebebrand J, Wehmeier PM, Re-

mschmidt H. 2002. Is topiramate effective for weight loss in

neuroleptic-induced obesity? Two case reports. Z Kinder

Jugendpschiatr Psychother 30:51�58.

Vieta E, Torrent C, Garcia-Ribas G, et al. 2002. Use of

topiramate in treatment-resistant bipolar spectrum disorders.

J Clin Psychopharmacol 22:431�435.

Wozniak J, Biederman J, Richards JA. 2001. Diagnostic and

therapeutic dilemmas in the management of pediatric-onset

bipolar disorder. J Clin Psychiatry 62(Suppl 14):10�15.

Anticipation in juvenile mania 77

LETTER TO THE EDITOR

Electroconvulsive therapy for major depression in an elderly personwith epilepsy

KRZYSZTOF ARTUR KUCIA, RADOSL AW STEPANCZAK & BEATA TREDZBOR

Department of Psychiatry and Psychotherapy, Medical University of Silesia, Katowice, Poland

AbstractThe case of a 72-year-old woman with a history of 40 years of epilepsy and medication-refractory severe depression isdescribed. Despite the chronicity of the present depressive episode, mild MRI pathology and somatic complications,especially pneumonia and drug-induced hyponatraemia, we observed rapid and complete remission of depressive symptomsin the course of ECT. Neither cognitive impairment nor a perceptible influence on the neurological illness was seen, and noincrease in seizure threshold has been observed during the course of 2 years maintenance ECT treatment. This article isoffered in an attempt to enrich the clinical literature in this field and therefore encourage psychiatrists to consider ECT andMECT as a safe and efficacious option in epileptic patients with major depressive disorder.

Key words: Major depressive disorder, epilepsy, electroconvulsive therapy, maintenance ECT

Introduction

The incidence of psychiatric diseases is much higher

in patients with epilepsy than in the general popula-

tion. According to most authors depression is the

most common psychiatric comorbidity in this pa-

tient group (Harden 2002; Kanner and Balabanov

2002; Preuter and Norra 2005). Additionally, de-

pression in epileptics is more frequent and severe

than in patients with other chronic medical and

neurological conditions, and the suicide rate is

particularly high (Lambert and Robertson 1999;

Harden 2002; Preuter and Norra 2005).

Electroconvulsive therapy (ECT) is not contra-

indicated in patients with epilepsy and may be safely

used in those with severe, refractory or psychotic

depression (Lambert and Robertson 1999; Harden

2002).

ECT is widely used all over the world but there is

still scant published data on the use of ECT in the

treatment of psychiatric syndromes in epileptic

patients.

Case report

Ms W, 72-year-old woman was admitted to the in-

patient psychiatric ward because of a major depres-

sive episode, following repeated failure to respond to

outpatient treatment efforts and threats of suicide.

She presented markedly depressed mood, slowed

psychomotor activity, anergy, anhedonia, impaired

concentration, internal tension, groundless anxiety,

trembling, constipation and stomach ache. She

reported decreased appetite and loss of weight

(5 kg during the last month). Transient suicidal

ideation was present. She had initial insomnia,

early morning awakening and diurnal variation in

mood.

There was no evidence of hallucinations or delu-

sions. She denied illicit drug or alcohol use. This

most recent, third episode, began 7 months ago.

Adequate trials of atypical (venlafaxine and mirta-

zapine) and tricyclic antidepressants in combination

with neuroleptics and benzodiazepines yielded

minimal response. Augmentation of antidepressant

therapy with antipsychotics (sulpiryde, and then

with olanzapine) during outpatient treatment was

chosen because of prominent anxiety features and

gastrointestinal complaints (gastrointestinal exami-

nations revealed no organic pathology). Despite

applied pharmacotherapy minimal symptoms re-

duction was observed. Additionally, Ms W was

suspected of having olanzapine-induced glucose dys-

regulation.

Correspondence: Krzysztof Artur Kucia, MD, PhD, Department of Psychiatry and Psychotherapy, Medical University of Silesia, Ziol owa

45/47 Str., 40-635 Katowice, Poland. Tel: �48 32 205 9260. E-mail: krzykuc@mp.pl

The World Journal of Biological Psychiatry, 2009; 10(1): 78�80

(Received 8 January 2007; accepted 6 March 2007)

ISSN 1562-2975 print/ISSN 1814-1412 online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

DOI: 10.1080/15622970701320582

There was a history of depression in the patient’s

family: her mother and sister were hospitalised due

to depression.

Her past medical history was significant for lasting

40 years epilepsy (complex partial seizures), coron-

ary artery disease and hypertension.

Her daily medication on admission included: tra-

zodone, clorazepate, zolpidem, primidone (500 mg),

carbamazepine (600 mg), bisoprolol, indapamide,

omeprazole.

The physical and neurological exams were essen-

tially normal, and routine screening laboratory

values were within normal limits.

EEG demonstrated discreet pathology of parox-

ysmal activity (basal activity a�8�9 Hz, with encore

of slow activity theta, mainly from temporal and

occipital regions).

MRI revealed mild dilatation of pericerebral

spaces in parieto-temporal regions and bilateral

mild dilatation of perivascular spaces in subcortical

nuclei.

The diagnosis of major depressive disorder, re-

current, severe without psychotic features was estab-

lished.

During the first weeks of hospitalisation cloraze-

pate was discontinued and trazodone was titrated up

to 300 mg daily without any therapeutic effect. GP

consultant judged that antidepressant potentiation

with lithium or tri-iodothyronine was contraindi-

cated because of cardiovascular risks. Augmentation

strategy with 60 mg buspirone proved ineffective. In

an effort to obtain reduction of anxiety symptoms,

400 mg perazine were added to antidepressant,

unfortunately with no clinical improvement. The

patient still presented sorely depressed mood, ten-

dency to cry, definite psychomotor retardation,

anxiety, suicidal thoughts and gastrointestinal com-

plaints. She refused to take part in therapeutic

activities, spending the most of the day in bed.

Due to loss of appetite she limited the ration.

At this point, because of the severity of the

presentation and the lack of response to medication,

following complex consideration of the advantages

and disadvantages of ECT compared with alterna-

tive treatments, after written consent the patient was

referred to electroconvulsive therapy. Trazodone,

buspirone and zolpidem were gradually withdrawn.

Following neurological consultation, carbamaze-

pine was changed to oxcarbazepine 450 mg because

of the side effect of hyponatraemia.

Anaesthesia was induced with thiopental followed

by succinylcholine and oxygenation. The bifrontal

ECT treatments were administered twice weekly

(SpECTrum 5000Q device) by monitoring four-

channel EEG, ECG and oxygen saturation. ECT

seizure threshold was estimated with use of struc-

tured stimulus dosage titration procedure described

by Coffey et al. (1995) at the first (192 mC) and

eleventh (192 mC) ECT. Subsequent treatments

were given at 1.5 times the seizure threshold. During

maintenance ECT (MECT), seizure threshold was

determined every 3 months and remained stable

(192 mC).

Ms W underwent 11 ECTs delivered at a charge of

288 mC. Motor seizure length ranged between 21

and 57 s, and assessed per EEG ranged between 35

and 120 s. The patient was maintained on her

antiepileptic medications throughout the ECT treat-

ment period.

After the third ECT session, visible improvement

of mood and activity was observed. Insomnia and

suicidal thoughts disappeared. After the fourth ECT

the patient started smiling spontaneously, began to

talk to others and read magazines. She was more

active, and enjoyed attending art and music therapy.

Anxiety and somatic complaints disappeared.

Because of pneumonia and intensified hypona-

traemia, on the day of the sixth planned ECT session

the patient was transferred to the Department of

Internal Diseases. Oxcarbazepine was replaced with

valproate semisodium 600 mg daily.

Unfortunately her mental state worsened so she

was transferred back to the psychiatric ward.

After a break of 3 weeks, caused by the somatic

disturbances described above, the sixth and seventh

ECT sessions were applied, leading to improvement

of the patient’s mental condition. Finally the patient

underwent 11 sessions of ETC with a complete

remission of symptoms.

After discharge we choose MECT as a sole anti-

depressive regimen. The continuation protocol was

guided by a schedule of decreasing treatment, with

the first MECT after 1 week, the second treatment

2 weeks later, the third treatment 3 weeks later and

then monthly. In the ensuing two years, Ms W had

no episodes of affective illness. According to her

neurologist there was neither a noticeable change in

the frequency of sporadic spontaneous partial sei-

zures she had had before undergoing ECT nor in the

follow-up EEG made twice a year (a�8�9 Hz,

encore of activity theta mainly from temporal and

occipital regions).

Discussion

Many literature reviews have described the effective-

ness and safety of ECT in treatment of major

depressive disorders. ECT and MECT are good

options for elderly patients, particularly those who

are drug refractory, medication-intolerant or medi-

cally ill (Rabheru and Persad 1997).

ECT for major depression in an elderly epileptic 79

However, there have been reports of spontaneous

seizures in patients after ECT, worsening of epilepsy

with ECT is unlikely (Lambert and Robertson 1999;

Preuter and Norra 2005; Lunde et al. 2006).

Furthermore, several studies have shown that the

seizure threshold tends to rise during the course of

treatment, therefore some consider ECT to be an

effective anticonvulsant (Lambert and Robertson

1999; Harden 2002).

Despite the chronicity of the illness in the de-

scribed case, poor previous medication response and

the MRI findings described above, ECT provided

rapid antidepressant response. It is interesting to

note that, in spite of antiepileptic medication

changes, her seizure threshold remained stable in

the course of ECT. Neither there was alteration of

seizure threshold detected during MECT.

Cognitive side effects of ECT are a concern in

considering the use of this method in elderly

patients. To minimalise possible cognitive impair-

ment, bifrontal electrode placement was chosen.

A bifrontal placement has been reported in several

case series to have efficacy equal to, or greater

than, standard bifrontotemporal placement, with

fewer cognitive side effects (Lawson et al. 1990;

Letemendia et al. 1993; Bailine et al. 2000; Green-

berg and Kellner 2005).

Neither significant confusion, memory impair-

ment, arrhythmia nor any other adverse effects

typical of electroconvulsive therapy were observed

in the course of 11 ECTs as well as during 2 years of

MECT.

Epilepsy does not represent a significant risk

factor for ECT as long as it is diagnosed and well

treated (Hsiao et al. 1987; Harden 2002; Kanner

and Balabanov 2002; Preuter and Norra 2005). For

our patient, who suffered from medication-resistant

depression coexisting with epilepsy, ECT proved to

be beneficial and led to satisfactory control of her

mental status without perceptible influence on the

neurological illness.

Acknowledgements

None.

Declaration of Interest

None.

References

Bailine SH, Rifkin A, Kayne E, et al. 2000. Comparison of

bifrontal and bitemporal ECT for major depression. Am J

Psychiatry 157:121�123.

Coffey CE, Lucke J, Weiner RD, Krystal AD, Aque M. 1995.

Seizure threshold in electroconvulsive therapy: I. Initial seizure

threshold. Biol Psychiatry 37:713�720.

Greenberg RM, Kellner CH. 2005. Electroconvulsive therapy:

A selected review. Am J Geriatr Psychiatry 13:268�281.

Harden CL. 2002. The co-morbidity of depression and epil-

epsy: epidemiology, etiology, and treatment. Neurology 59

(Suppl 4):48�55.

Hsiao JK, Messenheimer JA, Evans DL. 1987. ECT and

neurological disorders. Convuls Ther 3:121�136.

Kanner AM, Balabanov A. 2002. Depression and epilepsy: how

closely related are they? Neurology 58(Suppl 5):27�39.

Lambert MV, Robertson MM. 1999. Depression in epil-

epsy: etiology, phenomenology and treatment. Epilepsia 40

(Suppl 10):21�47.

Lawson JS, Inglis J, Delva NJ, Rodenburg M, Waldron JJ,

Letemendia FJJ. 1990. Electrode placement in ECT: Cognitive

effects. Psychol Med 20:335�344.

Letemendia FJ, Delva NJ, Rodenburg M, et al. 1993. Therapeutic

advantage of bifrontal electrode placement in ECT. Psychol

Med 23:349�360.

Lunde ME, Lee EK, Rasmussen KG. 2006. Electroconvulsive

therapy in patients with epilepsy. Epilepsy Behav 9:355�359.

Preuter C, Norra C. 2005. Mood disorders and their treatment

in patients with epilepsy. J Neuropsychiatry Clin Neurosci

17:20�28.

Rabheru K, Persad E. 1997. A review of continuation and

maintenance electroconvulsive therapy. Can J Psychiatry 42:

476�484.

80 K.A. Kucia et al.