sbn173_35_S1 1..273

1. 1 Diagnosis

AN EMPIRICAL EVALUATION OF THE ARIZONA

SEXUAL EXPERIENCE SCALE (ASEX) QUESTION-

NAIRE TO IDENTIFY SEXUAL DYSFUNCTION IN

PATIENTS OF THE SCHIZOPHRENIA SPECTRUM

Luciana Vargas Nunes, L. H. Dieckmann, F. S. Lacaz,

A. M. Costa, C. Daltio, R. A. Bressan, S. V. Nunes, J. J. Mari

Psychiatry, UNIFESP-Universidade Federal de Sa˜o Paulo, Sa˜o

Paulo, Brazil

Objective: The main aims of this paper are threefold: a) to assess the fre-

quency of sexual dysfunction in a sample of out-patients with schizophre-

nia and schizoaffective disorder under antipsychotic therapy; b) to

investigate the effect of different antipsychotics on sexual function;

and c) to evaluate the accuracy of the Arizona Sexual Experience Scale

(ASEX) to identify sexual dysfunction. Method: A cross-sectional study

of sexual function was conducted with one-year consecutive outpatients

from the Schizophrenia Program of the Universidade Federal de Sao

Paulo (Proesq), from February 2007 to January 2008. Patients, who

met DSM IV criteria for schizophrenia or schizoaffective disorder,

were asked to fulfill both the Arizona Sexual Experience Scale (ASEX)

and the Dickson Glazer Scale for the Assessment of Sexual Functioning

Inventory (DGSFi) at a single interview. Results: 137 patients entered the

study (86 were males and 51 were females). The sensitivity and specificity

of the ASEX at identification of sexual dysfunction in relation with

DGSFi scores were respectively: 80.8%, (95% CI = 70.0%–88.5%) and

88.1% (95% CI = 76.5%–94.7%), and the misclassification rate was

9,5%. The ROC curve comparing the ASEX and the DGSFi scores

revealed a value of 0.93 (CI = 0.879–0.970), with the optimum cutoff

point of ASEX for sexual dysfunction being 14/15. Sexual dysfunction

measured by ASEX was higher among females (79.2%) than for

males (33.3%), and this difference was statistically significant (v

27.41, df = 1, P < .001). Conclusion: Patients under antipsychotic treat-

ment showed a high level of sexual complaints, and the ASEX question-

naire showed to be an accurate instrument to identify sexual dysfunction

in an out-patient sample of patients with schizophrenia and schizoaffec-

tive disorders. Females showed a higher frequency of sexual dysfunctions

than males and sexual drive and ability to reach orgasm were the most

affected areas. The use of antipsychotics, especially the combinations

of medications, was more likely to impair sexual functioning.

ID: 540733

METHODS FOR DETERMINING INTER-RATER

RELIABILITY OF THE PANSS: A REVIEW OF THE

LITERATURE

Kia Crittenden, C. Yavorsky, F. Ockun, K. Wolanski, K. Kobak

Clinical Operations, MedAvante, Inc., Hamilton, NJ, USA

Inter-rater reliability (IRR) is an important consideration in CNS trials,

as poor reliability increases error variance, which reduces study power

and increases the risk for type II errors. The increasing placebo response

in schizophrenia points to renewed examination of the role of clinical

ratings. The method for evaluating reliability impacts the reliability fig-

ures, eg, observation of tapes results in artificially higher IRR than in-

dependent interviews, as the former artificially reduces information

variance. Thus, the methodology for determining IRR is a critical factor

in interpreting a study’s results. To date there has been no systematic

review of the methods used in determining reliability of the PANSS,

and the relationship between method and level of IRR. We searched

Pub Med using keywords ‘‘PANSS’’ and ‘‘reliability.’’ In only 53% of

studies (18/34), the method for determining reliability was described:

these included rating tapes (9/34;26%), joint interviews (8/34;24%),

and a combination of both methods (1/34;3%). No studies used indepen-

dent interviews. Rater training was done in 56% (19/34) of studies. IRR

total PANSS score ranged from 0.63 to 0.71 for tapes and from 0.92 to

0.99 for joint interviews. The positive subscale ranged from 0.56 to 0.99

(tapes) and 0.72 to 0.99 for joint interviews, and the negative subscale

ranged from 0.27 to 0.90 for tapes and from 0.63 to 0.92 for joint inter-

views. Methodology for testing IRR is not often cited. Tape rating is the

most common method reported. Reliability on the PANSS using joint

methods is good, but unknown with independent assessments.

ID: 550710

COMPARING SCHIZOPHRENIA AND BIPOLAR

PROBANDS: NEUROPSYCHOLOGY,

NEUROPHYSIOLOGY AND FMRI

John A. Sweeney

Psychiatry, University of Illinois, Chicago, IL, USA

A better characterization of similarities and differences between schizo-

phrenia and affective psychoses is needed to define their diagnostic

boundaries and to clarify their overlapping and unique pathophysiolo-

gies. We contrasted untreated first episode patients with schizophrenia,

bipolar disorder and psychotic depression, and matched controls, on

a battery of neuropsychological, oculomotor and fMRI measures. Pre-

viously published data from the Pittsburgh first episode study will be pre-

sented together with data from our new sample from Chicago that now

includes 40 schizophrenia patients, 22 bipolar patients and 15 unipolar

depression patients. In both samples, neuropsychological data indicate

similar profiles of deficit across cognitive domains across disorders

that were more severe in schizophrenia patients. Profiles of oculomotor

deficit were more variable. Measures most heavily dependent on prefron-

tal systems were impaired in all groups but showed greater deficit in

schizophrenia. Preliminary analysesoffMRIdatawithoculomotortasks

also suggest greater deficits in prefrontal systems in schizophrenia. Some

attention-dependent and sensorimotor tasks were equally or more im-

paired in one or both groups of patients with affective psychoses than

in schizophrenia. Patients with psychotic depression generally had the

least severe cognitive deficits, but uniquely manifested some specific mo-

tor system dysfunctions. These findings document neurobehavioral def-

icits in both affective and nonaffective psychotic disorders. The

similarities and differences in their severity and profile suggest that

the illnesses have somewhat different impacts across functional neural

systems, and also suggest promising phenotypic approaches for charac-

terizing unique and common neurocognitive deficits associated with

schizophrenia and psychotic affective disorders.

ID: 550510

MAJOR DEPRESSIVE EPISODES IN FIRST EPISODE

PSYCHOSIS

Kristin Lie Romm

, J. I. Rossberg

1,2

, O. A. Andreassen

1,2

E. A. Barrett

, A. Faerden

, I. Melle

1,2

Institute of Psychiatry, University of Oslo, Oslo, Norway;

Department of Research and Education, Division of Psychiatry,

Ullevaal University Hospital, Oslo, Norway;

Division of

Psychiatry, Aker University Hospital, Oslo, Norway

There are clear indications that depressive symptoms contribute to increased

morbidity and mortality in patients with schizophrenia spectrum disorders,

but the actual rate of patients experiencing clinically significant depressive

1. 1 Diagnosis 1

International Congress on Schizophrenia Research

symptoms in the early phases of the disorder (from before onset of psychotic

symptoms until start of first treatment) is not known. The main aims of this

study were to examine the prevalence of lifetime DSM-IV Major Depressive

Episode in a sample of patients with first episode schizophrenia spectrum

disorders, and to examine the demographic and clinical characteristics dis-

tinguishing this patient group (MDE) from patients that never had experi-

enced a Major Depressive Episode (Non-MDE). A total of 123 patients (age

18–65) from the ongoing longitudinal Thematic Organized Psychosis re-

search study (TOP) were included at the time of their first treatment. Diag-

noses were set according to SCID-I for DSM-IV (schizophrenia,

schizophreniform disorder, schizoaffective disorder and psychosis NOS).

Present symptom level was assessed by the Positive and Negative Syndrome

Scale (PANSS) and the Global Assessment of Functioning scale (GAF). Pre-

morbid functioning was measured by the Premorbid Adjustment Scale

(PAS). The duration of untreated psychosis (DUP) was measured as weeks

from first psychotic symptoms until start of first treatment. A total of 59

(48%) patients had experienced one or more MDE. Poor premorbid adjust-

ment in childhood and adolescence was significantly associated with more

depressive episodes. There were no significant differences between the MDE

and Non-MDE group for age, gender and DUP. With the exception of the

PANSS general score (which was higher in the MDE group), there were no

differences between the groups concerning positive and negative symptom

scores or GAF scores. This study shed more light on the frequency of MDE

among patients with first episode psychosis. The fact that as many as 48% of

the patients met the MDE criteria shows the importance of achieving more

knowledge about depressive symptoms among patients with first episode

psychosis.

ID: 550371

HEIGHTENED RATES OF SCHIZOAFFECTIVE

DISORDER AMONG IMMIGRANTS? A CROSS-

SECTIONAL STUDY

Akiah A. O. Berg

, I. Melle

1,2

, S. R. Aminoff

O. A. Andreassen

1,2

, E. Hauff

Institute of Psychiatry, University of Oslo, Oslo, Norway;

Department of Research and Education, Division of Psychiatry,

Ullevaal University Hospital, Oslo, Norway

Migration has been found to be a relevant risk factor in the development

of schizophrenia and other psychotic disorders. The purpose of this

study is to explore if migration is associated with particular symptoms

among psychosis patients exposed to migration (IM) compared to non-

immigrants (NI) patients. Our main hypothesis is that among an IM

group we will find a higher rate of schizophrenia-like psychotic symp-

toms in connection with mood episodes, suggesting possible heightened

rates of schizoaffective (SA) disorder compared to NI. We used natural-

istic cross-sectional data from a cohort of patients being treated for

schizophrenia- or bipolar-spectrum disorders in the greater Oslo region

(Thematic Organized Psychosis research (TOP) study). Exclusion crite-

ria was drug-induced psychosis, IQ < 70, or limited Scandinavian lan-

guage skills. Participants (age 18–65) received a DSM-IV diagnosis on

the basis of a SCID-I interview. Present symptom level was assessed

by the Positive and Negative Syndrome Scale (PANSS). Immigration

status was based on observed ethnicity, country of birth, mother tongue

and immigrant status of parents. The sample consisted of 435 partici-

pants with lifetime psychosis symptoms, whereof 27 % were 1st. or

2nd. generation IM. We found more IM with a DSM-IV diagnosis of

SA disorder (v

= 8214, df = 1; P < .005), and more NI with a diagnosis

of Bipolar I (BPI) (v

= 12,915, df = 1; P < .001). Logistic regression was

performed to explore main effects of diagnostic grouping for SA and BPI

in the context of independent variables. Results indicated that IM had

a7,2(CI= 2,4–21,6) times higher risk of receiving a SA disorder diag-

nosis (Wald = 12,609, df = 1, P < .001) than NI. ANOVA analyses of the

three PANSS subscales (positive, negative and general) showed similar

symptom patterns for IM and NI in the SA group, but significantly higher

positive (F = 18,848, df = 3, P < .001), and general symptoms (F = 9,396, df =

3, P < .001) for the BPI-IM group compared to the BPI-NI group. In con-

clusion we found that immigrant status was a significant predictor of re-

ceiving a SA diagnosis in our sample. Comparisons of symptom

patterns within diagnostic categories suggests that this finding was not

due to misclassification of affective disorders as belonging to the schizo-

phrenia spectrum in the IM group, but rather supports the hypothesis

that aspects of the migration experience are associated with specific clinical

symptom expression.

ID: 550226

IS SCREENING FOR PSYCHOSIS POSSIBLE IN

GENERAL POPULATION SAMPLES?

Tyrone D. Cannon

1,2

, R. Loewy

, C. E. Bearden

Psychology, UCLA, Los Angeles, CA, USA;

Psychiatry and

Biobehavioral Sciences, UCLA, Los Angeles, CA, USA;

Psychi-

atry, UCSF, San Francisco, CA, USA

The criteria used to ascertain youth at high clinical risk for psychosis em-

phasize the recent onset of sub-psychotic intensity psychotic-like symp-

toms. We have developed a self-report screening instrument (the

Prodromal Questionnaire or PQ) to provide an efficient screen for these

symptoms. Adolescents and young adults referred to a prodromal psychosis

research clinic completed the PQ and the Structured Interview for Prodro-

mal Syndromes. PQ positive symptom scores predicted a concurrent pro-

dromal or psychotic diagnosis with 92% sensitivity and 42% specificity. The

PQ thus shows good preliminary validity in detecting individuals with an

interview-based diagnosis of a prodromal or psychotic syndrome, but is not

sensitive to the threshold between prodromal and full psychosis. We also

assessed the rates of self-reported ‘‘prodromal’’ psychotic symptoms and

related distress in a college population (N = 1020). Participants’ responses

to the PQ were highly similar to the responses of non-psychotic-spectrum

patients in the original PQ validation sample, suggesting that the PQ may

perform similarly with a variety of populations. Applying the cutoff pro-

posed for screening treatment-seeking patients identified 43% of students,

while comparatively fewer participants (25%) endorsed items at the fre-

quency required for prodromal syndrome diagnosis by interview (ie,

weekly), and only 2% endorsed the positive symptom items as distressing.

Although attenuated psychotic experiences are commonly reported by

‘‘normal’’ young adults, frequent and distressing items identify a proportion

of students more consistent with the prevalence of psychotic-spectrum dis-

orders in the general population, which suggests a potential for future

screening of unselected samples.

ID: 550092

IMPRESSION-MANAGEMENT EFFECTS IN

PARANOIA ASSESSMENT

David Leland Roberts, J. M. Fiszdon, P. C. DeGeorge, C. Tek

Psychiatry, Yale University, New Haven, CT, USA

It is challenging to objectively measure undesirable personal characteristics

(eg, racism) due to participants’ tendency to portray a positive impression.

However, impression-management effects can be overcome with the use

of non-obvious measures. In schizophrenia, there is evidence linking para-

noia with an Externalizing-Personalizing Bias (EPB), a tendency to blame

negative experiences on the hostile intentions of others. However, this

finding is inconsistent, possibly due to participants’ efforts to appear

non-paranoid. The Social Cognition Screening Questionnaire (SCSQ)

uses non-obvious measurement to assess both EPB and efforts to under-

endorse EPB. The SCSQ is presented as a memory task consisting of

ten verbally-presented, second-person vignettes describing unpleasant

2 1. 1 Diagnosis

International Congress on Schizophrenia Research

social interactions. Each is followed by three Yes-or-No ‘‘memory’’ ques-

tions, one of which tacitly elicits a judgment about whether story characters

had negative intentionality toward the participant (eg, ‘‘Was the manager

trying to be rude to you?’’). Five of the vignettes include explicit evidence of

negative intentionality toward the participant, while five include no such

evidence. The SCSQ yields two scores: 1) An EPB score reflects endorse-

ment of negative intentionality in vignettes in which it is not present;

2) A Disavowal of Negative Intentionality (DNI) score reflects non-

endorsement of negative intentionality in vignettes in which it is present.

Outpatients with schizophrenia completed the SCSQ and PANSS (n =

38), and two established measures of paranoid attributional bias (n =

19). Convergent validity of the EPB score was supported by its correlations

with existing measures of paranoid attributional bias (r = .598; P = .011; and

r = .574; P = .010). The DNI score correlated significantly with the PANSS

suspiciousness item (r = .357; P = .033), whereas the EPB score did not (r =

.257; NS). Additionally, the DNI score discriminated between schizophre-

nia subgroups that were high (> 3; n = 14) versus low (< 3; n = 12) on the

PANSS suspiciousness item (t = 2.15; P = .042), whereas the EPB score did

not (t = 1.61; NS). This study provides preliminary evidence that individuals

with paranoia may under-report inferences of negative intentionality, and

that objective measurement of this effect may add incremental validity to

paranoia assessment in schizophrenia.

ID: 549924

ANTIDEPRESSANT EFFECT OF ANTIPSYCHOTICS

IN SCHIZOPHRENIA: COMPARISON OF TYPICAL

AND ATYPICAL ANTIPSYCHOTIC DRUGS

Antonio Reis Sa

, B. Avrichir, H. Elkis

Department and Institute of Psychiatry., Clinical Hospital, Faculty

of Medicine, University of Sa˜o, Brazil., Sa˜o Paulo, Brazil

Objective: Despite progress in the treatment of schizophrenia the efficacy of

atypical antipsychotics on depressive symptoms in schizophrenia is not well

established and for clarify this question we examined the effects of atypical

versus typical antipsychotics on depressive symptoms in a cohort study in

patients with schizophrenia. Methods: The data were drawn from a cohort,

naturalistic, observational study with 96 subjects diagnosed as being af-

fected by schizophrenia during a re-exacerbation phase. The patients

were taking typical or atypical antipsychotics. All subjects completed the

Calgary Depression Scale for Schizophrenia (CDSS) to rate the severity

of the depressive symptoms. The severity of schizophrenic symptoms

was rated by the Positive and Negative Syndrome Scale (PANSS) and

the Clinical Global Impression (CGI) severity and improvement scales.

Assessments of scales above were undertaken at baseline, 8 weeks, 16 weeks

and 24 weeks. Results: The PANSS total score higher than 70 and female

gender were significantly associated with the presence of depressive

symptoms. Global improvement of depressive symptoms was associated

with use of antipsychotics in general, however atypical antipsychotics

showed a statistically significant effect when compared to typical antipsy-

chotics. Conclusion: This observational study provides evidence that

atypical antipsychotics are more effective than typical antipsychotics on

the depressive symptoms in patients with schizophrenia.

ID: 549525

AFFECT IN SCHIZOPHRENIA: IMPLICATIONS FOR

DSM V AND RESEARCH

Dolores Malaspina

1,2

, J. W. Messinger

1,3

, V. Prudent

E. Mendelsohn

, D. Antonius

Psychiatry, New York University School of Medicine, New York,

NY, USA;

Psychiatry, Columbia University, College of Physicians

and Surgeons, New York, NY, USA;

Psychology, Long Island

University, Brooklyn, NY, USA;

Psychology, Emory University,

Atlanta, GA, USA

A flat or inappropriate facial affect has been a consistent finding among

a portion of people with schizophrenia from the earliest descriptions of

the disease. Flat affect (restricted, blunted) describes a lack of observable

emotional reactivity and inappropriate affect is the term for an incongruous

facial emotional expression. Kraepelin viewed flat affect as indicative of

a lack of emotional experience. Bleuler also observed incongruence between

emotional expression and self-reported experience in his schizophrenia

patients. Both Kraepelin and Bleuler agreed that abnormalities in affect

denoted a defective or diminished emotional expression. As such, anchored

ratings of the magnitude of affect abnormalities may be important indica-

tors of the success of new treatments for the core emotional deficits in the

disease. Additionally, as the truly heterogeneous nature of the schizophrenia

syndrome has been better appreciated, the possibility of using stable flat af-

fect as a categorical tool for subtyping schizophrenia has been supported in

numerous studies.

ID: 549399

ANTI-GAD65, NMDAR AND VGKC ANTIBODIES

DETECTED IN FIRST EPISODE PSYCHOSIS

Belinda Lennox

1,2

, M. Zandi

, S. Irani

, A. J. Coles

, A. Vincent

Psychiatry, University of Cambridge, Cambridge, United Kingdom;

Cameo, Early Intervention for Psychosis service, Cambridgeshire

and Peterborough NHS Foundation Trust, Cambridge, United

Kingdom;

Neurology, University of Cambridge, Cambridge, United

Kingdom;

Neuro-Immunology, University of Oxford, Oxford,

United Kingdom

There are many lines of evidence to suggest that schizophrenia may be

an autoimmune disorder. To date there have been no reports of path-

ogenic antibodies in cases of pure psychosis. However in cases of lim-

bic encephalitis, where psychosis is often present, there have been new

serum autoantibodies identified, which have been responsive to immu-

notherapy. Our aim was to screen serum from cases of first episode

psychosis for these antibodies. Methods: 16 patients under the care of

Cameo (www.cameo.nhs.uk). Serum collected within six months of

symptoms (range 2–26 weeks, mean 8 weeks). Serum was also col-

lected from 22 patients with chronic schizophrenia (length illness

>2 years) and 23 healthy controls. Voltage gated potassium channel

(VGKC)-Ab titres were measured byradioimmunoassay(Vincentet

al. 2004), and anti GAD65 antibodies measured by radio-immunopre-

cipitation assay using 125I (RSR Cardiff). Anti-NMDAR antibodies

were measured:HEK cells were transfected with NR1 and NR2B sub-

units and incubated with patient serum and a fluorescent antihuman

secondary to detect cell surface binding. Results: One case with

raised NMDAR antibodies, one with raised VGKC antibodies and

three with raised GAD antibodies were identified in the first episode

group. No antibodies were found in the chronic patients, or controls.

Discussion: These cases were all under the care of a first episode

psychosis service. None had any signs of organic illness. None had

features suggestive of limbic encephalitis. No antibodies were identi-

fied in chronic group, in keeping with other studies of autoimmune

mediated neurological disorders. Each of these antibodies identified

has validity in having an aetiological role in schizophrenia. If the

incidence found in this group is replicated, 31% of schizophrenia

may be autoimmune, and amenable to immunotherapy. We should

now aim to establish the frequency of these antibodies in all cases

of first episode psychosis, and introduce early recognition and

treatment.

ID: 549060

International Congress on Schizophrenia Research

1. 1 Diagnosis 3

NEUROPSYCHOLOGICAL TESTING AND

STRUCTURAL MAGNETIC RESONANCE IMAGING

IN THE DIAGNOSIS OF SCHIZOPHRENIA AFTER

A FIRST PSYCHOTIC EPISODE

Elissaios Karageorgiou

, R. Gollub

, N. Andreasen

, B.-C. Ho

J. Lauriello

, V. Calhoun

, S. C. Schulz

, A. P. Georgopoulos

1,2

Neuroscience, University of Minnesota, Minneapolis, MN, USA;

Psychiatry, University of Minnesota, Minneapolis, MN, USA;

Harvard University, Boston, MA, USA;

University of Iowa, Iowa

city, IA, USA;

University of New Mexico, Albuquerque, NM, USA

Introduction: Making an accurate diagnosis after a first psychotic episode

is important for initiating treatment and counseling patients and families.

Previous work by our group demonstrated the potential of applying linear

discriminant analysis (LDA) to structural magnetic resonance imaging

(sMRI) and neuropsychological (NP) testing in adolescents (Pardo et al.

2006). In this study, we developed sMRI and NP models for disease diag-

nosis and, additionally, cross-validated the accuracy of our modeling algo-

rithms. The analysis was made on a subject subsample characterized by

a first psychotic episode from the larger MIND Clinical Imaging Consor-

tium (MCIC) sample. Methods: Prospective first-episode schizophrenic

patients (47 controls; 28 patients) underwent structured interview evalua-

tions, sMRI scanning and NP testing. In the ensuing analysis 75 NP and 95

sMRI variables were used. We developed diagnostic models based on the

following algorithms. First, a stepwise-LDA was performed, where subsets

of the original variables were used in the final models. Second, an LDA was

performed on latent variables, which were in turn created by Principal

Component Analysis (PCA) and selected according to Humphrey-Ilgen

parallel analysis. Error estimation of the modeling algorithms was evalu-

ated by leave-one-out external-cross-validation. The described analyses

were performed on NP and sMRI variables both in combination and sep-

arately. Results: The following cross-validation results were obtained.

sMRI only. a) Stepwise-LDA: 53.6% sensitivity and 74.5% specificity, b)

PCA-LDA: 67.9% sensitivity and 72.3% specificity. NP only. a) Step-

wise-LDA: 78.5% sensitivity and 85.1% specificity, b) PCA-LDA: 78.5%

sensitivity and 91.5% specificity. Combined sMRI-NP. a) Stepwise-

LDA: 60.7% sensitivity and 72.3% specificity, b) PCA-LDA: 89.3% sensi-

tivity and 93.6% specificity. Conclusions: These results reveal that a) the

combination of sMRI and NP variables can significantly improve diagnos-

tic accuracy, b) the PCA-LDA approach accounting for more of the var-

iance in the data is more robust than Stepwise-LDA, c) NP variables are

more informative than sMRI indicating that cognitive deficits precede

structural anomalies. This work verifies the accuracy of the applied mod-

eling algorithms and, by extension, any of the models produced in the pro-

cess. To obtain an error estimate for a specific model to be used in clinical

practice, that model will have to be tested on a separate test sample.

ID: 548502

A PARADIGM SHIFT: CRITICAL CHANGES

ANTICIPATED IN THE CLASSIFICATION OF

PSYCHOTIC DISORDERS

William T. Carpenter

1,2

Psychiatry, University of Maryland School of Medicine, Baltimore,

MD, USA;

Maryland Psychiatric Research Center, University of

Maryland School of Medicine, Baltimore, MD, USA

DSM-V, when published in 2012, will shape the clinical care, education, and

research agendas for psychotic illnesses for the following decade. It was pre-

viously hoped that classification would be reorganized based on etiological

and pathophysiological evidence that would move diagnoses from syn-

dromes to disease entities. In the absence of sufficient evidence, critical issues

related to a more modest modification of DSM-IV are addressed. These

issues, briefly stated, are: 1) Criteria emphasize reality distortion and neglect

avolition, psychomotor and cognitive pathology; 2) Schizoaffective schizo-

phrenia is weak on validity and reliability, but is clinically useful; 3) Psycho-

sis NOS is over-utilized, and there is inadequate data on psychotic disorders

other than schizophrenia; 4) Data from early detection studies has not been

integrated with classification; 5) Similarities between cases of schizophrenia

and cases on the diagnostic borders raise questions regarding bipolar disor-

ders and schizophrenia spectrum personality disorders; and 6) The role of

affect pathology in classification of psychotic disorders. Addressing these

issues is an important challenge, but a more substantial basis for advancing

classification relates to a proposed paradigm shift. A dimensional system

may augmentdiagnostic classes. This presentation will outline psychopath-

ological dimensions based on evidence that domains of pathology represent

critical issues for clinical care and education, and provide a more heuristic

framework for research. Individual cases within a class vary in which

domains of pathology are manifest, and domains of pathology may better

explain across boundary similarities. Dimensions include some diagnostic

criteria (eg, negative symptoms) and some additional features (eg, depres-

sion, cognition). A tentative model for diagnostic criteria and pathological

dimensions for schizophrenia will be presented for audience critique.

ID: 548047

THE AUSTRALIAN SCHIZOPHRENIA RESEARCH

BANK (ASRB): THE DEVELOPMENT OF AN ELEC-

TRONICALLY DELIVERED CLINICAL ASSESS-

MENT BATTERY

Carmel Maree Loughland

1,2

, J. Richards

1,2

, M. Aphale

1,2

F. Henskens

1,2

, V. J. Carr

1,2

, S. V. Catts

1,4

, A. Jablensky

1,3

P. Michie

1,2

, B. J. Mowrey

1,4

, C. Pantelis

1,5

, U. Schall

1,2

R. J. Scott

1,2

, T. J. Lewin

1,2

Centre for Brain and Mental Health Research, Schizophrenia

Research Institute, Newcastle, NSW, Australia;

University of

Table.

sex/

age antibody

assay

level

(normal

range)

clinical

features

other

diagnostic

tests outcome

F/20 GAD65 3904cpm

(<2500)

3/7 elated

mood,

paranoid

psychosis

repeat GAD

after 18/

12 < 3u/l

remitting course

for 6 months,

treated with

Risperidone.

well since

M/20 GAD65 5581cpm

(<2500)

2/12

paranoid

psychosis,

catatonia

5/12 admission,

continued

negative

symptoms

M/25 GAD65 21622cpm

(<2500)

2/52

paranoid

psychosis

improved after

2/12 on

Olanzapine

disengaged

from follow up

F/22 VGKC 2300 at

1ul cpm

(<100)

2/52 paranoid

psychosis,

insomnia,

overactivity

MRI normal,

Naþ normal,

no evidence

paraneoplastic

syndrome

gradual

improvement

over 3/12 on

Risperidone

M/21 NMDAR þve 2/52 insomnia,

elated mood,

paranoid

and grandiose

delusions

No evidence

teratoma

Improved after

3/12

Olanzapine

International Congress on Schizophrenia Research

4 1. 1 Diagnosis

Newcastle, Newcastle, NSW, Australia;

University of Western

Australia, Perth, WA, Australia;

University of Queensland, Bris-

bane, QLD, Australia;

University of Melbourne, Melbourne, VIC,

Australia

The Australian Schizophrenia Research Bank (ASRB) was established in

2007 to collect cross referenced diagnostic, neuropsychological, anatomical

(MRI) and genetic data from 2000 people with schizophrenia and 2000

healthy age and gender matched controls. A clinical assessment battery

(CAB), administered over 3 hours, was developed and consists of a struc-

tured diagnostic interview, clinical and family history, measures of neuro-

psychological and cognitive performance (WTAR, WASI, RBANS,

COWAT), symptom (SANS) and general functioning (GAF) ratings.

This large scale data collection across multiple Australian sites necessitated

a quality-controlled, time and cost effective data collection strategy. To this

end, the ASRB has developed an electronic version of the assessment bat-

tery called eCAB. The eCAB was developed using SQL programming. Data

is entered directly into the software by assessment officers using a lap top

computer. Each assessment item is presented in a question and answer for-

mat, and the program allows navigation through questions with comment

boxes for note taking. Quality control parameters for data input and

a checking mechanism at the end of the assessment ensure all questions

are answered. Scoring formulae embedded into the program calculate

raw scores, standard scores and factor scores for the neuropsychological

evaluations and rating scales, reducing scoring time and error. The

OPCRIT algorithm (Craddock et al., 2006) is used to derive diagnostic con-

firmation based on DIP answers. Completed assessments are uploaded to

the ASRB central server in encrypted format for access by researchers. The

eCAB is a time and cost efficient method for collecting research data, and

significantly reduces data handling errors through automatic score calcu-

lations. It is well tolerated by patients and controls alike, and eliminates the

need for paper assessment storage in large scale projects such as the ASRB.

ID: 550847

A NEW DIAGNOSTIC CLASS: THE PRODROMAL

RISK SYNDROME FOR FIRST PSYCHOSIS

Thomas McGlashan

, J. Addington

, K. S. Cadenhead

T. D. Cannon

, B. A. Cornblatt

, R. Heinssen

, D. O. Perkins

L. J. Seidman

, M. T. Tsuang

3,8

, E. F. Walker

, S. W. Woods

Psychiatry/Medicine, Yale University, Branford, CT, USA;

Psychiatry, University of Toronto, Toronto, ON, Canada;

Psy-

chiatry, UCSD, San Diego, CA, USA;

Psychology/Psychiatry/

Behavioral Sciences, UCLA, Los Angeles, CA, USA;

Psychiatry,

Zucker Hillside Hospital, Long Island, NY, USA;

Schizophrenia

Spectrum Disorders Research Program, NIMH, Bethesda, MD,

USA;

Psychiatry, University of North Carolina, Chapel Hill, NC,

USA;

Psychiatry, Harvard Medical School, Boston, MA, USA;

Psychology/Psychiatry, Emory University, Atlanta, GA, USA

Since Falloon’s seminal pilot study in 1992 (1), research on early detection

and intervention in psychotic disorders has demonstrated that reliably iden-

tifiable clinical syndromes can be identified in help-seeking persons which

are also ‘‘prodromal’’, meaning they signal a very high risk for developing

(converting to) psychosis in the near future, ie, within 2–3 years (2). A Scale

of Prodromal Symptoms (SOPS) embedded within a Structured Interview

for Prodromal Syndromes (SIPS) has been developed by the presenter and

colleagues and used in ‘‘prodromal’’ clinics nationally, and internationally

in studies of the natural course and treatment of these clinical high risk

syndromes (3). This presentation will include: 1) a review of prodromal as-

sessment from Falloon (1) to Yung and McGorry (4) to Miller and McGla-

shan (3), 2) reliability studies of the SIPS and SOPS prodromal syndromes

(5), 3) early longitudinal studies validating high risk for psychosis in these

syndromes (4), 4) a later study of 372 patients meeting SIPS/SOPS prodro-

mal criteria (6) showing predictive accuracy for psychosis comparable to

other areas of preventive medicine and (7) an unpublished study comparing

the NAPLS prodromal sample with a) 195 natural subjects, b) 200 help-

seeking persons who do not develop psychosis, c) 40 familial (genetic)

high risk subjects, and d) 53 persons with Schizotypal personality disorder.

The groups are compared across the symptoms, functioning, comorbidity,

family history, and course of illness domains originally identified by Robins

and Guze as validating psychiatric syndromes. The data from these studies

collectively support the inclusion of the Prodromal Risk Syndromes for

First Psychosis in DSM-IV alongside traditional diagnoses in the Psychotic

Disorders spectrum. Specific criteria have been formulated. The clinical,

therapeutic, scientific, ethical, and political issues involved in adding a re-

liably identifiable and potentially treatable risk syndrome to the DSM-V

psychotic disorders will also be discussed (8).

References

1. Falloon. Psychiatry 1992;55.

2. McGlashan TH, Johannessen JO. Schizophr Bull. 1996;22:201.

3. Miller TJ, McGlashan TH, et al. Psychiatric Q. 1999;70:273.

4. Yung AR, McGorry PD. Schizophr Bull. 1996;22:354.

5. Miller TJ, et al. Schizophr Bull. 2003;29:703.

6. Cannon TD, et al. Arch Gen Psychiatry. 2008;65:28.

7. Woods SW, et al. Validity of the Prodromal Risk Syndromes for Psy-

chosis, submitted.

8. McGlashan TH. Early Intervention in Psychiatry. 2007;1:289.

ID: 551923

A TAXOMETRIC ANALYSIS OF NEGATIVE

SYMPTOMS IN THE WHO TEN-COUNTRY STUDY

OF SCHIZOPHRENIA

Amy Lynn Wilson, J. J. Blanchard

Psychology, University of Maryland, College Park, College Park,

MD, USA

It has been proposed that enduring primary negative symptoms, or deficit

symptoms, reflect a distinct subtype of schizophrenia—the deficit syndrome

(Carpenter et al., 1988). Although differences between deficit and nondeficit

schizophrenics are suggestive of a distinct and separate disease process

(Buchanan and Carpenter, 1994), these findings are also consistent with a

dimensional-only model where the deficit syndrome merely reflects an ex-

treme on a continuum of severity. Taxometric statistical methods (Waller

and Meehl, 1998) allow for a quantitative examination of whether a taxonic

or latent class model best describes negative symptoms in schizophrenia. Us-

ing suchmethods Blanchard et al. (2005) have shown that negative symptoms

are taxonic. In this study we sought to replicate these findings in a larger and

more diverse sample. The current investigation is a taxometric analysis of the

World Health Organization Ten-Country Study of Schizophrenia (Jablensky,

et al., 1992). Negative symptom data from a subset of 694 schizophrenia par-

ticipants within the WHO study were analyzed using the taxometric methods

of maximum covariance analysis (MAXCOV) and mean above minus below

a cut (MAMBAC). Results from MAXCOV and MAMBAC indicated a

latent class with base rates of 16% and 18%, respectively. These results

are consistent with a categorical view of negative symptoms in schizophrenia.

ID: 551893

SCHIZOPHRENIA COLLABORATIVE RESOURCE

(SCORE) TOOL: TRACKING PATIENT PROGRESS

Stephen Rodriguez

, C. A. Bossie

, W. Macfadden

J. T. Haskins

, J. Docherty

Ortho-McNeil Janssen Scientific Affairs, LLC, Titusville, NJ, USA;

Joan and Sanford I. Weill Medical College and Graduate School of

Medical Sciences of Cornell University, White Plains, NY, USA

Unlike many other chronic disorders, stages of illness and disease progres-

sion for patients with schizophrenia are not well characterized. This gap in

International Congress on Schizophrenia Research

1. 1 Diagnosis 5

understanding the course and stages of the illness impacts expectations, goal

setting, recognition of clinically relevant change, and ultimately overall pa-

tient management. We describe software, entitled the Schizophrenia Collab-

orative Resource Tool (SCoRe), that identifies the stage of illness, creates

a personalized patient profile, and tracks patient progress over time. A

group of clinicians who are experts in the treatment of patients with schizo-

phrenia were involved in the development of this tool. Literature reviews

were conducted and data from clinical trials were analyzed to guide the clini-

cians in describing stages of illness and associated symptom severity and

level of functioning. A software tool designed to capture information about

the status of patients with schizophrenia was then developed for use in clin-

ical practice. This interactive program captures information (from both cli-

nician and patient) on current interventions, symptom severity, social

functioning, stress tolerance, cognition, and physical health. The patient’s

illness is staged according to an algorithm that considers the type of clinical

intervention needed and the severity of core symptoms of schizophrenia.

Four stages of illness were defined based on both clinician input and

data analysis: 1) acute, 2) stabilization, 3) stable, and 4) remission. These

stages are reassessed at each subsequent visit. This information can then

be used to develop treatment goals and to track patient progress over

time. The program can also provide graphical outputs of an individual

patient’s clinical profile over time. SCoRe may help in the management

of patients with schizophrenia. It can identify milestones for improvement

throughout the course of illness toward which both the patient and clinician

can strive to achieve. Additionally, this software provides physicians with

a mechanism to record and track patient progress and assist with the

development of treatment goals. Supported by funding from Ortho-McNeil

Janssen Scientific Affairs, LLC.

ID: 551850

DEVELOPMENT AND INITIAL PSYCHOMETRIC

EXPLORATION OF A CLINICAL GLOBAL

IMPRESSION SCALE FOR SCHIZOAFFECTIVE

DISORDER (CGI-SCA)

Michael Allen

, J. Bartko

, J. P. Lindenmayer

, D. G. Daniel

C. M. Canuso

, C. Kosik-Gonzalez

, J. Adams

, D. Revicki

G. Garibaldi

University of Colorado School of Medicine, Denver, CO, USA;

Independent Consultant, Newville, PA, USA;

New York Univer-

sity School of Medicine, New York, NY, USA;

United BioSource

Corporation, McLean, VA, USA;

Ortho-McNeil Janssen Scientific

Affairs, LLC, Titusville, NJ, USA;

i3 Research, Midland, MI,

USA;

United BioSource Corporation, Bethesda, MD, USA;

Roche Pharmaceuticals, Nutley, NJ, USA

Currently, there is no rating scale specific to schizoaffective disorder (SCA).

This study tests the psychometric properties of a new scale, the Clinical

Global Impression for Schizoaffective Disorder (CGI-SCA), using the fa-

miliar 7-point format to measure positive, negative, manic, depressive

and global severity and change. Experts adapted the CGI and developed

a manual for evaluation of patients with SCA. Data for psychometric ex-

ploration of the CGI-SCA were gathered during training for an interna-

tional multicenter drug trial using the CGI, PANSS, HAM-D-21 and

YMRS. Exploratory data were obtained using 132 investigators from the

United States (N = 41), India (N = 35) and Eastern Europe (N = 56) rating

2 videotapes. Inter-rater reliability was initially evaluated using intra-class

correlation (ICC) and convergent and divergent validity using Cramer’s V.

Effect sizes were calculated for sensitivity to change. A questionnaire exam-

ined content validity. Power analysis using the training data suggested that

with sufficient variability among subjects, inter-rater reliability could be

demonstrated with 2 raters and 12 subjects. This formal assessment was ac-

complished using 12 sets of videotaped interviews and 2 additional raters.

Test-retest reliability was assessed with 10 randomly selected raters from

each region rating the same tapes on 2 occasions separated by a minimum

of 2 weeks and maximum of 4 weeks. Inter-rater reliability varied across the

3 geographic regions. ICC ranges: CGI-Manic, 0.829–0.948; CGI-Depres-

sive, 0.916–0.972; CGI-Positive, 0.245–0.759; CGI-Negative, 0.464–0.624;

CGI-Global, 0.005–0.428. Pairs of ratings, eg, YMRS and CGI-Manic,

converged in 5 of 6 cases (Cramer’s V 0.44–0.82) and diverged in 3 of 5,

eg, CGI-Positive vs CGI-Manic (Cramer’s V 0.64–0.76). Sensitivity to

change ranged in effect size from 2.7 to 12.7. Overall, raters in each region

concurred that the CGI-SCA was useful, with 90% or more agreeing that the

CGI-SCA captured a large fraction of the variance. Formal testing of inter-

rater reliability found ICCs of 0.5–0.883. Test-retest reliability ranged from

0.444 to 0.888. Initial psychometric examination indicates at least moderate

inter-rater and test-retest reliability for these symptom domains and good

sensitivity to change. Content validity was high in the United States, India

and Europe. These results suggest that the CGI-SCA may be useful in

clinical trials. Supported by Ortho-McNeil Janssen Scientific Affairs, LLC.

ID: 551816

DIMENSIONAL CLASSIFICATION OF SCHIZO-

PHRENIA

Patricia Beatriz Estani, P. B. Estani

Department of Neurocognition and Neurobiology of Schizophrenia,

Neuropsychiatric Medical Hospital Dr. L.S. Morra, Cordoba,

Argentina

Introduction: The objective of this research is to propose a simple, dimen-

sional classification of schizophrenia, based on the concept of endopheno-

types (Gottesman, 2003). Methods: A total of 1200 participants were the

subjects. Neuropsychological tests were applied . We operationalized two

neurocognitive endophenotypes: sensorial gating and cognitive perfor-

mance. Results: A: Dimensional Classification of Schizophrenia: Structural

dimensions.A1:Genetically-defined Dimensions,asa Working Group1.Neg-

ative Schizophrenia and 2. Positive Schizophrenia. B. The Longitudinal Diag-

nosis of the Disease: Longitudinal Dimensions. B1. The Continuous

Dimensions: The dimensions as phasesofa continuum from a pre-clinical state

to a more chronic, pathological state. 1) the schizophrenia prodrome. 2) first

episode ofillness. 3)The Spontaneous Amileoration Syndrome: typically active

phase or positive symptoms. 4) A progressive neurodegeneration and neuro-

deterioration phase. 5) Deficit Syndrome. Conclusions: The classification

allows full inclusionof the neurocognitive endophenotypes in the classification

ofschizophrenia,defining thegenotypestobenefitthe understandingandtreat-

ment of schizophrenia. The full explanation of this categorization will be pre-

sented along with the results of this pilot sample.

Reference

1. Gottesman II, Gould TD. The endophenotype concept in psychiatry:

etymology and strategic intentions. Am J Psychiatry. 2003 Apr 1 ;

160(4):636–45.

ID: 551775

SCHIZOPHRENIA AT THE BORDERS WITH

BIPOLAR DISORDER AND SCHIZOTYPAL

PERSONALITY

Larry J. Siever

Psychiatry, Mt. Sinai School of Medicine, New York, NY, USA

In order to delineate similarities and differences between schizophrenia

(SCZ) and schizophrenia spectrum personality pathology, on the one

hand, and SCZ and bipolar disorder (BD), on the other for the purposes

of defining the relationship between these disorders in DSM-V, external

validators, evaluated by imaging methodology including functional

International Congress on Schizophrenia Research

6 1. 1 Diagnosis

imaging and measurement of D

binding and raclopride displacement, bio-

markers, physiologic measures, cognitive status, genetic data, environmen-

tal risk factors, course, and treatment response will be compared between

schizotypal personality disorder (SPD) and SCZ and between BP and SCZ

incorporating new data from our and other laboratories. These studies sug-

gest that SCZ and SPD share structural and functional imaging character-

istics, cognitive profile, and asociality but are distinguished by the

neurobiology of dopaminergic activity, course of illness, comorbidity,

and treatment response. SPD and schizophrenic patients both show volume

decreases, with sparing of frontal lobe in SPD. SPD patients, in contrast to

schizophrenia patients overactivate instead of underactivate prefrontal cor-

tex, recruit compensatory regions (eg, BA10) doing cognitive tasks, are

more vigilance on a startle-blink task, and evidence lesser activation of

the dopamine system on provocation than SCZ patients. In contrast,

SCZ and BD, both of which may be characterized by psychosis, appear

to share dopaminergic hyperactivity and reduced gabaminergic markers

as measured in imaging and neurochemical studies, treatment response

to dopamine antagonists, genetic antecedents such as neuroregulin 1,

and altered physiology. BP and SCZ patients both share decreases in

grey matter, greater in SCZ patients, reduced NAA on MRS, abnormal

PPI, P300, and pursuit maintenance, while BP patients have better pre-

served cognitive function and intact MMN. SPD thus appears to be related

to schizophrenia as reflected as part of the schizophrenia spectrum while

bipolar disorder and schizophrenia share complementary characteristics

reflecting a diathesis to psychosis, reflected in neurotransmitter (DA,

GABA) abnormalities.

ID: 551702

STRESS REACTIVITY IN ULTRA-HIGH RISK

PATIENTS: THE INFLUENCE OF CLINICAL

SYMPTOM LEVELS, LIFE STRESS, SOCIAL

SUPPORT AND OTHER PSYCHOSOCIAL FACTORS

Gretchen Louise Sholty

, J. C. Sun

, K. I. Mathis

, P. Bachman

T. J. Williams

1,2

, S. E. Taylor

, K. H. Nuechterlein

1,2

T. D. Cannon

1,2

, C. E. Bearden

, C. M. Yee

1,2

Psychology, UCLA, Los Angeles, CA, USA;

Psychiatry and

Biobehavioral Sciences, UCLA, Los Angeles, CA, USA

From the perspective of the diathesis-stress model of schizophrenia, life

stress is likely a critical factor in the development of psychosis. It has been

suggested that the hypothalamic-pituitary-adrenal (HPA) axis may serve

as a biological pathway underlying the relationship between stress and

illness expression (eg, Corcoran et al., 2003). The present study sought

to elucidate this relationship in individuals at ultra-high risk (UHR)

for psychosis by examining HPA activity at rest and in response to a psy-

chosocial stressor. Previous studies investigating the stress response in

schizophrenia have shown that patients display higher resting cortisol rel-

ative to healthy individuals and a blunted response to stress (eg, Jansen

et al., 2000). Heightened resting cortisol levels also have been observed in

schizotypal personality disorder, possibly reflecting vulnerability to

schizophrenia (Walker et al., 1996). Although prior research has exam-

ined cortisol reactivity in patients with schizophrenia-spectrum disorders,

less is known about the relationship between risk status for psychosis and

the stress response. In the present study, the Trier Social Stress Test

(TSST) was administered to 33 UHR individuals, as determined by the

Structured Interview for Prodromal States (Miller et al., 2002), and 20

healthy comparison subjects. Salivary cortisol was sampled before and

immediately after the TSST. UHR participants were later subdivided

into those who subsequently developed a psychotic disorder (n = 10)

and those who did not (n = 23) over the following 12–24 months. Initial

results suggest that among the UHR individuals, those who converted to

psychosis had higher resting cortisol relative to non-converters and

healthy participants. Both UHR groups showed blunted cortisol reactiv-

ity to the TSST as compared to healthy individuals, who exhibited a robust

response to the stressor. These findings support the view that the stress

response may be a marker of evolving psychosis. It is unclear, however,

if the distinct patterns of HPA reactivity among UHR patients who con-

verted might be due to their vulnerability for psychosis, the acuity of their

illness and related increases in life stress, or a combination of these factors.

Behavioral ratings of life stress, clinical symptom levels and their rele-

vance to this hypothesis are examined, as well as relationships between

cortisol reactivity and self-report measures of coping style, social support,

and early life adversity.

ID: 551667

PERSONALITY TRAITS PREDICT PSYCHOTIC

TRANSITION IN ULTRA HIGH RISK

Peter M. Dingemans, R. van de Flierte, H. Becker, D. Nieman,

D. Linszen

Psychiatry, Academic Medical Center, Amsterdam Z.-O.,

Netherlands

This is the first study to deal with personality traits in the transition of Ultra

High Risk Subjects. In a Ultra High Risk group for psychosis (Dutch Pre-

diction Study for Psychosis: subjects (n = 77) rated the Schizotypal Person-

ality Questionnaire (SPQ) and a DSM-IV screener (PDQ-R)for personality

disorders at the start of the study. After a median follow up time of 24 months

transitioners were compared with non-transitioners on the SPC en SCID-II

subscales with regard to transition (yes/no) and time to transition (multivar-

iate survival with correction; two sided P = .10). SPQ-factors 2 (negative

schizotypy) and factor 3 (desorganization)turned out to be the best predic-

tors for transition (yes versus no) and speed of transition. Conclusion is that:

1) personality traits influence transition into psychosis; 2) schizotypy

concept is applicable to UHR, 3) SPQ is a better predictor for transition

than PDQ-R.

ID: 551291

AN INNOVATING TOOL FOR THE EARLY

DIAGNOSIS OF SCHIZOPHRENIA AND OTHER

PSYCHIATRIC DISORDERS

Helene Wilquin

1,2

, Y. Delevoye-Turrell

Lab. URECA, University of Lille Nord de France, Villeneuve

d’Ascq, France;

Clinic Lautreamont, Loos, France

The phase before the emergence of a first episode psychosis is today at-

tract a widespread interest in the field of clinical research of schizophrenia.

Indeed, it has been shown that the earlier patients are treated, the better

their outcome. It is also important from a preventive perspective. How-

ever, the prodromal phase is difficult to detect as it is characterised by

many none specific behavioural and cognitive deficits. The most widely

used approach is an interview for the detection of ultra-high risk

(UHR) patients (McGorry, Yung 1996). This test combines symptoms

and signs with other risk factors, eg, age, family history, personality dis-

order. This approach has revealed to be promising but its time consuming.

The aim of our work was to see if non-verbal motor tasks could be as

efficient in dissociating UHR, first episode and chronic patients with

schizophrenia. A load cell was used to quantify those motor disorders

and agency impairments that are known to be affected in schizophrenia

(Bulot et al. 2007; Delevoye-Turrell et al. 2003). The CAARMS (Compre-

hensive Assessment of At Risk Mental State—Yung et al., 2003) was used

International Congress on Schizophrenia Research

1. 1 Diagnosis 7

to divided the young patients into two distinct groups: UHR patients (n =

10) and patients after the onset of fully fledged psychosis (n = 10). Control

subjects included healthy and schizophrenic adults. The subjects’ task was

to use an object to hit (task 1) or resist (task 2) impacts produced by a col-

lision between the hand held object and a pendulum. The results of the hit

task provided the means to quantify motor preparation, motor prediction

and motor coordination. For the resist task, we further contrasted the case

when the pendulum was self released and released by the experimenter in

order to evaluate agency. The two groups of patients were impaired in the

execution of motor control. However, the pattern of the deficit was dif-

ferent compared to both healthy and chronic schizophrenic adults. A def-

icit of agency was observed only in the group of patients after the onset of

fully fledged psychosis. This deficit was similar to that observed in the

chronic schizophrenic; the UHR had similar results than that observed

in the healthy controls. Motor deficits are present very early in the evo-

lution of psychosis development while the agency deficit seems to be the

result of a first episode psychosis. These first results suggest that non-ver-

bal motor tasks may reveal an efficient tool to help for an early diagnosis

of schizophrenia.

ID: 551186

KRAEPELIN’S INFLUENCE ON THE

SCHIZOPHRENIA CONCEPT IN DSM

Stephan Heckers

Psychiatry, Vanderbilt University, Nashville, TN, USA

Emil Kraepelin has shaped diagnostic systems and research strategies in

psychiatry with the introduction of the diagnoses dementia praecox and

manic-depressive illness. He conjectured that clinical observation, exper-

imental studies of patients, and basic studies of disease mechanisms will

identify psychiatric disorders, with clearly delineated etiology, mecha-

nisms, and outcome. However, his separation of diagnostic classes and

the reliance on disease course and outcome for diagnostic classification

have been challenged repeatedly. Despite these concerns, DSM-III and

DSM-IV have continued a categorical classification of psychotic disor-

ders in the tradition of Kraepelin, emphasizing outcome and affective

symptoms for the subtyping of psychotic disorders and minimizing the

importance of psychomotor changes and avolition. The presentation

will review in detail how Kraepelin’s dementia praecox concept: 1) evolved

over several decades, 2) emphasized avolition and cognitive deficits and

deemphasized psychotic symptoms, and 3) continues to fuel schizophrenia

research even today. The presentation will review the implications of

Kraepelin’s scientific biases for the current DSM and will outline options

for DSM-V.

ID: 551175

VALIDATION OF ‘‘PRODROMAL’’ CRITERIA TO

DETECT INDIVIDUALS AT ULTRA HIGH RISK OF

PSYCHOSIS: 2 YEAR FOLLOW UP

Alison Yung

1,2

, B. Nelson

1,2

, C. Stanford

, M. Simmons

1,2

E. Cosgrave

1,2

, E. Killackey

, L. Phillips

, A. Bechdolf

J. Buckby

, P. McGorry

1,2

Orygen Research Centre, Parkville, VIC, Australia;

Psychiatry,

University of Melbourne, Melbourne, VIC, Australia

Background: Identification of individuals ‘‘prodromal’’ for schizophrenia

and other psychotic disorders relies on criteria that predict onset within

a brief period. Previous trials and biological research have been predicated

on the view that certain ‘‘ultra high risk’’ (UHR) criteria detect ‘‘the pro-

drome’’, but there is a need to test the validity of these criteria. Also, it

may be the fact of seeking help from a UHR service, rather than the UHR

criteria per se, which predicts onset of psychosis. Aim: To assess the pre-

dictive validity of the UHR criteria in a clinical population. A secondary

aim was to assess clinicians’ ability to detect UHR criteria. Method: Pres-

ence of UHR criteria was determined in 292 individuals at baseline. At 2

year follow up the number of new cases of psychotic disorder was assessed

in those meeting and not meeting the UHR criteria. Results: Help-seeking

per se was not a risk factor for development of psychotic disorder, rather

it was the UHRþ status at baseline that predicted psychosis onset within

2 years. The transition rate of 16% was much lower than in initial cohorts

(over 40%). Clinicians frequently missed UHRþve individuals and failed

to diagnose the UHR criteria in others. Conclusions: Although young

help-seekers meeting UHR criteria are at greater risk of psychotic disor-

der than those who do not meet them, caution is needed in their manage-

ment, since a high transition rate can no longer be assumed. Training

with a structured assessment tool may aid clinicians in detecting the

UHR criteria.

ID: 550971

IS LATE-ONSET SCHIZOPHRENIA A DISTINCT

SUBTYPE OF SCHIZOPHRENIA?

Dilip V. Jeste

1,2

, I. Vahia

, S. Golshan

, I. Fellows

, H. Kraemer

B. Palmer

Psychiatry, University of California, San Diego and VA San Diego

Healthcare System, La Jolla, CA, USA;

Psychiatry, VA San Diego

Healthcare System, San Diego, CA, USA;

Psychiatry, Stanford

University, Stanford, CA, USA

Introduction: Onset of schizophrenia in later life has been a controversial

topic. Different versions of DSM have taken varying positions on this issue,

usually without adequate data. An International Consensus Statement by

researchers in late-life schizophrenia recommended that age 40 be used as

the lower age cutoff for defining late-onset schizophrenia (LOS). Yet,

whether LOS is a distinct subtype of schizophrenia remains unsettled.

Methods: We analyzed data from over 800 community-dwelling patients

age > 40 years with DSM-III-R or DSM-IV diagnosis of schizophrenia

or schizoaffective disorder, studied at our NIMH-funded research center

at UCSD over a 12-year period. These included more than 700 patients

with EOS, and about 100 with LOS, using age of 40 at first manifestation

of prodromal symptoms as the cutoff. Baseline assessments included stan-

dardized instruments for psychopathology (Positive And Negative Syn-

drome Scale, Hamilton’s Depression scale), motor abnormalities

(Abnormal Involuntary Movement Scale), cognition (including Wechsler

Adult Intelligence Scale—Revised, California Verbal Learning Test, Wis-

consin Card Sorting Test), and health-related quality of life (Quality of

Well-Being scale). The patients were followed annually. Results: Compared

to EOS, the LOS patients were older, had a greater proportion of women,

shorter duration of illness, less severe positive symptoms, better perfor-

mance on abstraction/cognitive flexibility and psychomotor speed, and bet-

ter health-related quality of life. Most of these differences remained

significant after controlling for age, gender, and severity of negative symp-

toms. The EOS and LOS groups did not differ in education, ethnicity, se-

verity of depressive, negative, or motor symptoms, proportion of patients

with deficit syndrome, or in crystallized verbal knowledge and perceptual-

organization/visual construction. The diagnosis of most of the patients fol-

lowed longitudinally did not change. Discussion: Our results support the

notion that LOS may be a distinct subtype of schizophrenia. Limitations

of our study will be discussed. The implications of the results are clinical as

well as theoretical in that they may relate to pathophysiologic factors pres-

ent at menopausal or post-menopausal age of onset of LOS. Biological re-

search on identifying factors responsible for a delayed onset of

schizophrenia could be useful in opening possible new leads in prevention

and intervention strategies for schizophrenia.

ID: 552223

International Congress on Schizophrenia Research

8 1. 1 Diagnosis

2. 2. Phenomenology

BILINGUAL LANGUAGE UNDERSTANDING IN

SCHIZOPHRENIA

Debra Titone

, M. Libben

, V. Whitford

, A. Malla

, R. Joober

Psychology, McGill University, Montreal, QC, Canada;

Douglas

Hospital, McGill University, Montre

al, QC, Canada

Investigations of language in people with schizophrenia date back to the first

scientific accounts of the disorder. Virtually all of this work has focussed on

language in monolingual people despite the fact that bilingualism is more

prevalent globally than monolingualism. Basic studies of bilingualism are

relevant to theories of language and neurocognitive dysfunction in schizo-

phrenia, and to both the clinical assessment and treatment of bilingual indi-

viduals with schizophrenia. We present preliminary data from a study of

language in four French-English bilingual people diagnosed with schizo-

phrenia and four matched bilingual non-psychiatric controls. Participants

read short English sentences that were literally plausible (‘‘She kept her

desk’’), idiomatic (‘‘He had a lark’’), or anomalous (‘‘He hired the debt’’),

and they made speededdecisions aboutthe sensibility of each sentence. These

sentences also varied in the contextual constraint of the final word of literal

and idiomatic sentences (eg, ‘‘She drove the car’’ vs. ‘‘He bought the knife’’).

Further, a common feature of bilingual communicative interaction, code-

mixing, was experimentally simulated by replacing the final word of each sen-

tence with its French translation equivalent (eg, ‘‘He used his comb’’ vs. ‘‘He

used his peigne’’). Mixed design analysis of variance revealed that bilingual

people with schizophrenia were slower to make sensibility judgments overall,

less able to comprehend idiomatic language, and were less likely to capitalize

on high contextual constraint than bilingual non-psychiatric controls (all rel-

evant group interactions, P < .05). Thus, several of the most replicated find-

ings observed for monolingual people with schizophrenia were evident in this

preliminary bilingual sample. Moreover, comprehension of code-mixed sen-

tences was less accurate and slower for bilingual people with schizophrenia

than for non-psychiatric controls, especially for contextually constrained

sentences (all relevant group interactions, P < .05). This novel finding sug-

gests that language impairment in schizophrenia may be especially pro-

nounced under conditions that simulate normal bilingual communicative

interaction. Whether bilingual people with schizophrenia engage in less

cross-language interaction than normal, or experience more difficulty resolv-

ing cross-language interaction is a topic of future investigation.

ID: 539076

SELF-ESTEEM, ANXIETY, AND DEFEATIST PER-

FORMANCE BELIEFS AS RISK FACTORS FOR

PARANOIA

Dror Ben-Zeev

, E. Granholm

1,2

Department of Psychiatry, University of California, San Diego, San

Diego, CA, USA;

Psychology Service, VA San Diego Healthcare

System, San Diego, CA, USA

Paranoid delusions are one of the most prevalent psychiatric symptoms in

individuals suffering from psychosis. Psychological theories have implicated

a number of variables in the formation and maintenance of paranoid beliefs,

including self-esteem, anxiety, and negative beliefs about the self. Empirical

studies evaluating the relationship between paranoia and self-esteem in

clinical samples have yielded inconsistent findings. This may be due to

differential mechanisms in various symptomatic subgroups. The current

cross-sectional study examined the interrelationships between predictors

of persecutory ideation in people with psychosis who range across the con-

tinuum of paranoid symptom intensity. One hundred and ninety-nine com-

munity-dwelling people with schizophrenia (n = 140) or schizoaffective

disorder (n = 59) completed measures of psychotic symptoms (PANSS),

paranoia (Paranoia Scale-PS), anxiety (Beck Anxiety Inventory- BAI),

self-esteem (Self-Esteem Rating Scale, Sort Version-SERS-SF), and defeatist

performance beliefs (Performance Evaluation factor from the Dysfunctional

Attitudes Scale-DPAS). Participants were divided in to 3 groups: Paranoid

(PANSS ‘‘suspiciousness’’ score of 4 or more), Non Paranoid (PANSS ‘‘Sus-

piciousness’’ score of 3 or less, and a score of 4 or more on one of the PANSS

items ‘‘delusions’’, ‘‘hallucinatory behavior’’, ‘‘grandiosity’’, and ‘‘unusual

thought content’’), and Remitted (score of 3 or less on all described PANNS

items). Multiple regressionanalysesshowed thatin all three groups self-esteem

predicted paranoia. In the paranoid group the relationship between self-

esteem and paranoia was mediated by anxiety and not defeatist performance

beliefs, whereas in the other two groups this relationship was mediated by de-

featist performance beliefs but not anxiety. Participants from the paranoid

group were significantly more anxious than non paranoid and remitted par-

ticipants. Defeatist performance beliefs did not differ significantly across

groups. The findings suggest different treatment strategies for different

groups, and highlight the importance of anxiety reduction as a potential treat-

ment target for patients with more severe paranoia.

ID: 550695

EMOTION RECOGNITION AND SCHIZOPHRENIA

Maite Ferrin

1,3

, J. L. Rubio

, M. Ruiz-Veguilla

, M. L. Barrigon

L. Hernandez

Unidad de Investigacio

n de Neuropsiquiatria del desarrollo, Com-

plejo Hospitalario de Jae

n, Jaen, Spain;

Unidad de Investigacio

nde

Neuropsiquiatria del desarrollo, Hospital de Santa Ana, Motril,

Spain;

Department of child and adolescent psychiatry, Institute of

Psychiatry, London, United Kingdom

Schizophrenia is associated with deficits in the processing of affective infor-

mation from facial expressions and other socialcues. The purpose of the pres-

ent study was to explore whether schizophrenia patients show deficits in

recognition of specific facial expressions. Healthy comparison subjects, un-

affected siblings of schizophrenia patients, and schizophrenia patients were

tested with a computerized test of degraded facial affect recognition. Patients

demonstrated a worse recognition of facial expressions. Specifically with

greater impairments of matched positive (happy) and negative (angry)

whereas healthy subjects and unaffected siblings recognized negative and

positive expressions at an equal level of accuracy. These results suggest

that deficits in the processing of facial emotion recognition from social

cues may contribute to poor social functioning in schizophrenia patients

and must be identified as an important target for clinical intervention.

ID: 550494

SENSE OF HUMOR IN PATIENTS WITH

SCHIZOPHRENIA

Irina Falkenberg

1,2

, K. Kluegel

, M. Bartels

, B. Wild

Psychiatry and Psychotherapy, RWTH Aachen University,

Aachen, Germany;

Psychiatry and Psychotherapy, Tu

bingen

University, Tu

bingen, Germany

Introduction: Sense of humor is an important social skill. It can facilitate

social interaction and can help to cope with adverse situations. So far, it has

not been studied systematically in patients with schizophrenia. Methods: 18

patients with schizophrenia and 18 healthy controls were examined with

standardized self-assessment questionnaires to study potential group differ-

ences in humor type preferences, state and trait cheerfulness, seriousness

and bad mood as well as the use of humour as a coping strategy. Results:

The patients had more difficulties in understanding the test joke items but

patients and controls did not differ in their humor type preferences and

their use of humor as a coping strategy. The patients’ readiness to react

to funny stimuli with exhilaration was negatively correlated with the

International Congress on Schizophrenia Research

2. 2. Phenomenology 9

severity of accompanying depressive symptoms. Conclusion: If understood

and if no additional depression is present, patients with schizophrenia can

enjoy humorous stimuli as controls do. This indicates a close connection

between exhilaration and hedonia. The ability to use humour as a coping

strategy is unaffected by the disorder.

ID: 550463

FEWER NEUROLOGICAL SOFT SIGNS AMONG

FIRST EPISODE PSYCHOSIS PATIENTS WITH

HEAVY CANNABIS USE

Miguel Ruiz-Veguilla

, M. Ferrin

1,2

, M. L. Barrigon-Esteve

J. L. Rubio

, J. A. Becerra

, M. Gurpegui

, J. Cervilla

Unidad de Investigacion de Neuropsiquiatrı

a del Desarrollo,

Complejo Hospitalario de Jae

n, Jaen, Spain;

Department of Child

and Adolescent Psychiatry, Institute of Psychiatry, London, United

Kingdom;

Unidad de Investigacion de Neuropsiquiatria del

Desarrollo, Hospital de Santa Ana, Motril, Spain;

CIBERSAM

Granada (CIB07/09/0036), Division of Psychiatry and Institute of

Neurosciences, University of Granada, Granada, Spain;

Psychiatry

and Neurosciences Research Group (CTS-549), Institute of

Neurosciences, University of Granada, Granada, Spain

Background: Although neurological soft signs (NSS) have been consis-

tently associated with schizophrenia and a variety of risk factors, few stud-

ies have focused on the association between NSS and environmental factors

such as cannabis use, particularly in first episode patients. Methods: We

administered the Neurological Evaluation Scale (NES) to 92 patients dur-

ing their first episode of functional psychosis. Psychopathology was

assessed with the Positive And Negative Syndrome Scale (PANSS) and

the family history of psychotic disorder was established on the basis of

the Family Interview for Genetic Studies (FIGS). We also assessed lifetime

cannabis and cocaine use utilizing that specific section of the Composite

International 28 Diagnostic Interview. The outcome variable was the pres-

ence of high NSS, defined by a score above the median split of the NES

score (N21). Results: Most patients (80/92, 87%) presented a non-affective

psychosis. The presence of high NSS showed a significant independent as-

sociation with not having been a heavy cannabis user (OR = 8.3; 95% CI =

2.4–33.3), family history of psychosis (OR = 4.3; 95% CI = 1.2–14.9), male

sex 33 (OR = 4.0; 95% CI = 1.2–14.0), lower score in verbal fluency and

higher score in negative symptoms (both P < .01). Conclusion: Our cross-

sectional results support the hypothesis that potentially different pathways

associated with the emergence of first episode psychosis may exist, includ-

ing neurological premorbid alteration and environmental cannabis abuse.

ID: 550355

INITIAL RELIABILITY AND VALIDITY OF THE

NEW NIMH-MATRICS NEGATIVE SYMPTOM

RATING SCALE

Jack J. Blanchard

, C. Forbes

, A. Wilson

, B. Baker

, R. Gur

W. P. Horan

, A. Kring

, M. E. Bennett

Psychology, University of Maryland, College Park, MD, USA;

Psychiatry, University of Pennsylvania, Philadelphia, PA, USA;

Psychiatry, UCLA, Los Angeles, CA, USA;

Psychology, UC-

Berkeley, Berkeley, CA, USA;

Psychiatry, University of Maryland

School of Medicine, Baltimore, MD, USA

Negative symptoms in schizophrenia are a major determinant of the func-

tional impairments that characterize the disorder as well as a significant

source of distress for caregivers. Unfortunately, negative symptoms repre-

sent an unmet therapeutic need as no medications have shown to be effec-

tive in their treatment. One outcome of an NIMH consensus development

conference on negative symptoms (Kirkpatrick et al., 2006) was the recom-

mendation for the development of a new negative symptom assessment in-

strument that addresses conceptual and psychometric limitations of

existing instruments. The NIMH-MATRICS negative symptom work-

group has developed a new instrument, the Negative Symptom Rating

Scale (NSRS), which represents a substantial step forward in the assessment

of this critical symptom domain. The current study is the first assessment of

the newly developed NSRS. Forty individuals diagnosed with schizophre-

nia or schizoaffective disorder completed assessments of current symptoms

and social functioning. To examine convergent validity of the NSRS, in-

dependent assessments of negative symptoms were obtained with the Scale

for the Assessment of Negative Symptoms (SANS). Discriminant validity

was examined by assessing the independence of the NSRS ratings from psy-

chotic and depressed symptoms. Finally, it was expected that the NSRS

ratings would be related to assessments of social functioning in the com-

munity. Analyses indicated that the NSRS has high internal consistency

reliability (total scale alpha = .86) and high convergent correlations with

SANS scales (median r = .52, P < .01). The NSRS ratings were uncorrelated

with either psychotic or affective symptoms (all ps > .05). The NSRS dem-

onstrated significant correlations with social functioning in the community.

These initial findings indicate that the NSRS holds promise as a next-gen-

eration negative symptom scale.

ID: 550319

INSTRUMENTAL ASSESSMENT OF MOVEMENT

DISORDERS IN PATIENTS WITH SCHIZOPHRENIA,

THEIR HEALTHY SIBLINGS AND CONTROLS

J. Koning

1,2

, P. N. van Harten

, R. S. Kahn

Psychiatric Centre, Symfora Groep, Amersfoort, Netherlands;

Psychiatry, Rudolf Magnus Institute of Neuroscience, University

Medical Centre Utrecht, Utrecht, Netherlands

Background: Movement disorders as dyskinesia and parkinsonism are

regarded as core features of schizophrenia and highly related to the dopa-

mine system. Several studies suggested that the presence of these movement

disorders in siblings of patients would constitute (genetic) markers of vul-

nerability. However, most individual studies on first degree-relatives failed

to find this association, possibly due to insufficient sensitivity of the applied

clinical rating scales. Quantitative instrumental measurements can be more

sensitive to detect subclinical movement disorders (Cortese 2005). This

study compares prevalences of dyskinesia and parkinsonism in patients

with schizophrenia, healthy siblings and matched controls using instrumen-

tal assessments. Methods: Instrumental measurements of tongue and finger

instability as a measure of dyskinesia and resting tremor of the finger as

a measure of parkinsonism were carried out in 44 medicated patients

with a non-affective psychosis (70% schizophrenia), 44 healthy siblings

and 32 controls. Dyskinesia and parkinsonism were quantified using instru-

mental force variability, a method to measure the percentage of motor in-

stability and tremor (Caligiuri 1994). All subjects were screened for Axis-1

pathology using CASH-interviews. Siblings and controls were matched

according to age, sex and years of education. Results: Siblings displayed

significantly more finger instability and resting tremor than controls

(2.35% vs.1.97%; P = .02 and 0.30% vs. 0.25%; P = .01 respectively). There

was no difference in tongue instability (16.7% vs 16.9%; P = .88). In addi-

tion, patients not only demonstrated more finger instability and resting

tremor than siblings (2.80% vs. 2.35%; P = .03 and 0.43% vs 0.30% P <

.01 respectively), but also more tongue instability (21.2% vs. 16.7%; P <

.01). Demographic information of patients, siblings and controls are given

for age, years of education and percentage of males (27.2, 27.7 and 25.7

years), (12.1, 14.2 and 14.7 years) and (82%, 55% and 50%). Conclusion:

This is the first study that used instrumental assessments to compare prev-

alences of motor disorders between patients with schizophrenia, healthy

siblings and controls. It clearly shows that siblings have more dyskinetic

and parkinsonian movements of the fingers than controls suggesting a

(genetic) risk of developing schizophrenia. Instrumental measurement

International Congress on Schizophrenia Research

10 2. 2. Phenomenology

can be used for detecting subclinical movement disorders in high risk

populations.

ID: 550265

ASSESSING AUDITORY VOCAL HALLUCINA-

TIONS: THE PSYCHOMETRIC EVALUATION OF

THE AUDITORY VOCAL HALLUCINATION

RATING SCALE (AVHRS)

Agna A. Bartels, G. Van de Willige, J. A. Jenner, D. Wiersma

University Center for Psychiatry, University Medical Center

Groningen, Groningen, Netherlands

Introduction: Assessing auditory vocal hallucinations (AVH) is important,

both for clinical practice and research purposes. Questioning patients thor-

oughly about different dimensions of their voices may offer a solid base for

therapeutic interventions, whereas reliable instruments are useful for out-

come assessment and research. Our aim is to examine the inter-rater reli-

ability and concurrent validity of the Auditory Vocal Hallucination Rating

Scale (AVHRS), a more extensive version of an existing scale to assess voi-

ces. Methods: To examine the inter-rater reliability (IRR) 23 patients of the

Voices Outpatient Department (VOPD) of the University Medical Center

Groningen (the Netherlands) were assessed with the AVHRS, both about

the past month and about lifetime. AVHRS total scores were compared

with available PANSS scores—of all patients from September 2006 till Sep-

tember 2008—to investigate concurrent validity. Results: The IRR of the

AVHRS was good (lifetime) to excellent (past month). Correlations with

the PANSS will be presented at the conference. Face validity was excellent.

Conclusions: So far, we consider the AVHRS to be a very comprehensive

scale with favourable psychometric properties. The instrument is useful for

both clinical therapy and research.

References

1. Haddock G, McCarron J, Tarrier N, and Faraghe, EB. Scales to mea-

sure dimensions of hallucinations and delusions: The psychotic symptom

rating scale (PSYRATS). Psychological Medicine. 1999;29:879–889.

2. Jenner JA and Van de Willige G. The Auditory Vocal Hallucination

Rating Scale (AVHRS). Groningen: University Medical Center Groningen,

University Center for Psychiatry, University of Groningen; 2002.

ID: 550238

SOCIAL ANXIETY AND IDEAS OF REFERENCE IN

EARLY PSYCHOSIS

Gloria H. Y. Wong

, C. Hui

, C. Chiu

, E. Chen

Department of Psychiatry, The University of Hong Kong, Hong

Kong, China;

Department of Psychiatry, Queen Mary Hospital,

Hong Kong, China

An increased prevalence of social anxiety disorder is observed in patients

with psychosis (16%–38%, compared with 13% in the general popula-

tion). Despite important treatment implications, previous attempts in

understanding the comorbidity have been inconclusive. This study

aims to explore the relationship between social anxiety and other psycho-

pathological characteristics in psychotic disorders, with a conceptually

driven focus on the possible linkage between social anxiety and the sub-

threshold symptom ideas of reference. A consecutive sample of 137 out-

patients with early psychosis was assessed for social anxiety using the

Liebowitz Social Anxiety Scale (LSAS), and for ideas of reference using

a new instrument Ideas of Reference Interview Schedule (IRIS). Positive

symptoms, negative symptoms, depression and insight were measured

using the Scale for the Assessment of Positive Symptoms in Schizophre-

nia (SAPS), the Scale for the Assessment of Negative Symptoms in

Schizophrenia (SANS), Calgary Depression Scale for Schizophrenia

(CDSS) and the Scale to Assess Unawareness of Illness (SUMD). To

identify potential explanatory variables for social anxiety, univariate

analyses were first carried out between LSAS and age, sex and all symp-

tom subscales. Variables that displayed simple correlation with LSAS

(based on a lax criterion of P < .1) were then entered as independent

variables into a stepwise multiple regression analysis to determine the

symptoms specifically correlated with LSAS. Results suggested

a three-factor model, with IRIS included as the strongest explanatory

factor, while avolition and blunting were also related (adjust R2 =

0.18; P = .00). The association observed between social anxiety and neg-

ative symptoms is consistent with previous findings. This is the first

study to demonstrate that a proportion of social anxiety covaries

with ideas of reference. The nature of their relationship and the extent

to which ideas of reference may explain for the excess of social anxiety

prevalence in early psychosis need to be further investigated.

ID: 550218

CHARACTERISTICS OF HIGH FUNCTIONING

INDIVIDUALS WITH SCHIZOPHRENIA

Stephen R. Marder

1,2

, A. N. Cohen

1,2

, A. Hamilton

1,4

, E. Saks

S. Glynn

1,2

, D. Hollan

, J. Brekke

Psychiatry, Semel Institute at UCLA, Los Angeles, CA, USA;

Mental Illness Research, Education, and Clinical Center, VA

Greater Los Angeles, Los Angeles, CA, USA;

School of Law,

University of Southern California, Los Angeles, CA, USA;

Department of Anthropology, UCLA, Los Angeles, CA, USA;

School of Social Work, University of Southern California, Los

Angeles, CA, USA

Recent attention has focused on individuals who live remarkably produc-

tive lives with rich interpersonal relationships despite active symptoms. We

know very little about how these ‘‘high functioning’’ individuals with

schizophrenia successfully cope with their illness to achieve occupational

and social success. Therefore, we initiated a qualitative study of high func-

tioning individuals with schizophrenia who are living in Los Angeles. For

this study, ‘‘high functioning’’ was defined as employment in a professional,

technical, or managerial occupation or high-level functionality as a stay-at-

home caretaker or full-time student. Method: Subjects who responded to

flyers were evaluated according to entry criteria. Following informed con-

sent, subjects were evaluated using the SCID, the PANSS, a background

survey, and the BASIS-24. This was followed by a Person-centered Inter-

view. This interview uses a method that is used in psychological anthropol-

ogy to obtain in-depth, phenomenological perspectives. Results: Fourteen

subjects qualified for the study and were interviewed. Subjects included 1

MD, 2 PhD’s, 2 MA’s as well as 4 BA’s. The majority had managerial or

supervisory responsibilities in their current job. Nearly all of the subjects

had active symptoms that affected their everyday life. The majority could

point to cognitive or behavioral techniques used to manage symptoms

in the face of these responsibilities. Participants tended to be very familiar

with mental health recovery concepts and articulated the ways in which

these concepts interfaced with their ideas about their own mental health

experiences.

ID: 549746

AN EXPERIENCE SAMPLING STUDY OF THE

EXPRESSION OF POSITIVE AND NEGATIVE

SCHIZOTYPY IN DAILY LIFE

Thomas R. Kwapil

, N. Barrantes-Vidal

, L. H. Brown

I. Myin-Germeys

Department of Psychology, UNC-Greensboro, Greensboro, NC,

USA;

Department de Psicologia Clı

nica i de la Salut, Universitat

International Congress on Schizophrenia Research

2. 2. Phenomenology 11

Auto

noma de Barcelona, Barcelona, Spain;

Psychiatry and

Neuropsychology, Maastricht University, Maastricht, Netherlands

Psychometrically identified positive and negative schizotypy are differen-

tially related to psychopathology, personality, and social functioning. How-

ever,little is knownabouttheexperience and expression of schizotypyin daily

life. The present study employed the experience sampling method (ESM) to

assess positive and negative schizotypy in daily life in a nonclinical sample of

412 young adults. ESM is a structured diary technique in which participants

are prompted at random times during the day to complete an assessment of

their current experiences. As hypothesized, positive schizotypy was associ-

ated with increased negativeaffect,thought impairment, suspiciousness, neg-

ative beliefs about current activities, and feelings of rejection, but not with

social disinterest or decreased positive affect. Negative schizotypy, on the

otherhand,wasassociatedwithdecreasedpositiveaffectand pleasurein daily

life, increased negative affect, and marked decreases in social contact and in-

terest. Both positive and negative schizotypy were associated with the desire

to be alone when with others. However, this desire appeared to be moderated

by anxiety in positive schizotypy and by diminished positive affect in negative

schizotypy.The findings supportthe construct validityof a multidimensional

model of schizotypy and the use of psychometric inventories for assessing

these dimensions. ESM appears to be a promising method for examining

the daily life experiences of schizotypic individuals.

ID: 549731

EFFECTS OF PRE-ONSET CANNABIS USE ON

EARLY-COURSE FEATURES OF FIRST-EPISODE

NONAFFECTIVE PSYCHOSIS: PREMORBID

FUNCTIONING, THE PRODROME, AND ONSET OF

PSYCHOSIS

Michael T. Compton

,M.Pringle

, S. M. Goulding

M. L. Esterberg

, M. Kelley

, T. Stewart

,E.F.Walker

Department of Psychiatry and Behavioral Sciences, Emory University

School of Medicine, Atlanta, GA, USA;

Department of Psychology,

Graduate School of Arts and Sciences of Emory University, Atlanta,

GA, USA;

Department of Biostatistics, Rollins School of Public

Health of Emory University, Atlanta, GA, USA;

Department of

Psychology, Georgia State University, Atlanta, GA, USA

Background: Several studies suggest that pre-onset cannabis use results in

an earlier age at onset of psychosis, independent of effects of gender and use

of other drugs. However, little is known about the influence of prior can-

nabis use on other early-course epochs, including the premorbid and pro-

dromal phases. We examined the effects of prior cannabis use on: (1) early

adolescent (12–15 years) premorbid functioning, (2) late adolescent (16–18

years) premorbid functioning, (3) four features of the prodrome, and (4)

mode of onset and age at onset of psychosis. Methods: Participants in-

cluded urban, low-income, predominantly African American, first-episode

patients (n = 109) in a public-sector hospital. The assessment included the

Premorbid Adjustment Scale (PAS) and the Symptom Onset in Schizophre-

nia inventory, along with other measures. Potential confounders, including

gender, family history of psychosis, and prior nicotine and alcohol use, were

considered. Results: Those having used cannabis before or at age 15 years

had a mean PAS early adolescence social score of 1.21

6 0.85, compared to

1.75

6 1.27 in those who had not (t = 2.46, P = .02), indicating better social

functioning among those having used cannabis. On the other hand, those

having used cannabis before or at age 18 had a mean PAS late adolescence

academic score of 3.59

6 1.46, compared to 2.00 6 1.33 (t = 4.52, P <

.001), indicating poorer academic functioning among those having used

cannabis. Cannabis use prior to onset of psychotic symptoms was not as-

sociated with having had a prodrome, duration of prodrome, or number of

prodromal features experienced. Whereas 16 (42.1%) of those having used

cannabis daily had an acute mode of onset of psychosis, only 12 (20.0%) of

those without past daily use had an acute mode (v

= 6.53, P = .04).

Conversely, 9 (23.7%) of those who had used daily had a subacute

mode, compared to 26 (43.3%) of those who had not used daily. Pre-onset

cannabis use was associated with an earlier age at onset of psychotic as well

as prodromal symptoms. Conclusions: We replicated the finding that pre-

onset cannabis use is associated with an earlier age at onset. Further, these

findings suggest that cannabis use is associated with poorer premorbid so-

cial and academic functioning and an acute mode of onset of psychosis.

Further research is warranted to further elucidate the complex associations

between early-course features and cannabis use.

ID: 549715

THE RELATIONSHIP BETWEEN AWARENESS OF

TARDIVE DYSKINESIA AND INSIGHT INTO

MENTAL ILLNESS IN SCHIZOPHRENIA

Robin Emsley, D. J. Niehaus, P. P. Oosthuizen, L. Koen,

B. Chiliza

Psychiatry, Stellenbosch University, Cae Town, South Africa

Background: A striking feature of tardive dyskinesia TD is that a large per-

centage of patients display an unawareness or lack of concern for the move-

ment disorder, with reported rates ranging between 44% and 95%.5;6 Poor

insight into their mental illness is also a common and often striking symp-

tom in schizophrenia, with an estimated 50% to 80% of such individuals not

being convinced that they have a disorder.12 Both phenomena have been

likened to anosognosia, a neurological deficit characterised by unawareness

or denial of an impairment, and associated with damage to specific brain

areas.6 We investigated whether poor awareness of TD is related to poor

insight into mental illness and examined their clinical correlates. Methods:

The sample comprised 84 men and 46 women aged 45

6 11.4 yrs with

schizophrenia and TD. Motor disorders were assessed by means of the Ex-

trapyramidal Symptom Rating Scale (ESRS)18 and schizophrenia psycho-

pathology by means of the Positive and Negative Syndrome Scale

(PANSS)19. The level of awareness of TD was calculated by the sum of

two items on the ESRS scale that rate the patient’s subjective evaluation

of the intensity of dyskinesia of extremities. For assessing insight into men-

tal illness we used a single item on the PANSS scale (item G 12). Results: We

found high rates of poor awareness of TD (52%) and poor insight into their

mental illness (72%). Poor of awareness of TD and impairment of insight

into mental illness were significantly related, although regression analysis

indicated that relatively little variance is shared between them. In fact, cor-

relation profiles for level of awareness of TD and the level of impairment of

insight into the mental illness were very different, suggesting that the two

phenomena are not related. Discussion: Insight into mental illness was re-

lated to the presence and severity of core psychosis symptoms, while aware-

ness of TD was related to severity of TD and the presence of parkinsonism.

ID: 549413

SCHIZOTYPY: PSYCHOMETRIC PROPERTIES AND

FACTORIAL STRUCTURE OF THREE SELF-

REPORT RATING SCALES, DEMOGRAPHIC

CORRELATES, AND ASSOCIATIONS WITH

NICOTINE, ALCOHOL, AND CANNABIS USE

Sandra M. Goulding

, M. L. Esterberg

, E. B. McClure-Tone

M. T. Compton

School of Medicine, Department of Psychiatry and Behavioral

Sciences, Emory University, Atlanta, GA, USA;

Graduate School

of Arts and Sciences, Department of Clinical Psychology, Emory

University, Atlanta, GA, USA;

Graduate School of Arts and

Sciences, Department of Clinical Psychology, Georgia State

University, Atlanta, GA, USA

International Congress on Schizophrenia Research

12 2. 2. Phenomenology

Introduction: Schizotypy and schizotypal personality disorder are thought

to be genetically related to schizophrenia given that relatives of people

with schizophrenia are more likely to exhibit schizotypal traits than indi-

viduals without a family history of schizophrenia. This presentation gives

an overview of a recent survey of undergraduate students, focusing on: (1)

measurement issues regarding self-report rating scales of schizotypy, (2)

demographic correlates of schizotypy scores, and (3) associations between

self-reported schizotypy and reports of nicotine, alcohol, and cannabis

use. Methods: A survey was conducted with 825 undergraduate college

students. Measures included the Schizotypal Personality Questionnaire

(SPQ), the Perceptual Aberration Scale (PAS), and the Social Anhedonia

Scale (SAS). Demographic characteristics and substance use variables also

were assessed. Results: Confirmatory factor analysis (CFA) supported

a 4-factor structure of the SPQ in this sample. Correlations between scores

on the three schizotypy measures were relatively modest, though inter-

correlations among the four SPQ subscales were fairly high. Internal con-

sistency reliability coefficients were acceptable for all instruments. A CFA

of an abbreviated version of the SPQ (SPQ-B) revealed that the standard

3-factor solution may not fit the data better than a 1-factor solution. Male

gender was significantly associated with higher SAS scores, and race was

associated with several schizotypy measures. Students reporting past use

of cigarettes, alcohol, or cannabis generally had higher schizotypy scores

than those without past use of these substances. Specifically, higher levels

of disorganized schizotypy were associated with greater indices of use of

all three substances. Furthermore, lower negative schizotypy and higher

cognitive-perceptual schizotypy were selectively associated with nicotine

and cannabis use, respectively. Discussion: These findings indicate that

commonly used measures of schizotypy have adequate psychometric

properties in this sample; that scores on some dimensions of schizotypy

may differ by gender and race; and that the use of nicotine, alcohol, and

cannabis are associated with schizotypal features. Greater elucidation of

these associations, ideally using longitudinal research designs, could pro-

vide crucial information not only on the connection between substance

use and schizotypy, but also between substance use and schizophrenia.

ID: 549144

PARANOIA AND HALLUCINATIONS IN THE

CONTEXT OF DAILY LIFE: THE ROLE OF

EMOTIONAL EXPERIENCES AND SELF-ESTEEM

Viviane Thewissen

1,2

, R. Bentall

,J.a

Campo

, T. van Lierop

J. van Os

2,5

, I. Myin-Germeys

2,4

Faculty of Psychology, Open University of the Netherlands,

Heerlen, Netherlands;

Department of Psychiatry and Neuropsy-

chology, Maastricht University, Maastricht, Netherlands;

School

of Psychology, Bangor University, Bangor, Gwynedd, Wales, United

Kingdom;

Section Social Cognition, Mondriaan Zorggroep,

Heerlen, Netherlands;

Division of Psychological Medicine, Institute

of Psychiatry, London, United Kingdom

Emotional experiences—anxiety in particular—and self-esteem are consid-

ered to play an important role in the generation and maintenance of para-

noia. The goal of current study was to explore the dynamic association

between emotional experiences (anxiety, anger, depression) and self-esteem

on the one hand and paranoia on the other. In addition, it was investigated

whether the association is symptom-specific by looking at hallucinations as

well. The Experience Sampling Method, a structured self-assessment diary

technique, was used to examine the alleged association in the daily life of

158 individuals who ranged across the paranoia continuum. Multilevel re-

gression analyses were performed to investigate the association between

self-esteem and emotional experiences and paranoia / hallucinatory epi-

sodes ( = uninterrupted series of occurrence of paranoia / hallucination).

Results showed that self-esteem plays a significant role in the onset of

both paranoia and hallucinatory episodes. In addition, level of self-esteem

is significantly lower during both paranoia and hallucinatory episodes.

However, self-esteem only plays a significant maintaining role in the

duration of a paranoia episode. Paranoia and hallucinatory episodes are

both characterised by higher levels of anxiety and depression. However,

anxiety is most strongly associated with paranoia. Only paranoia episodes

are characterised by higher levels of anger. It can be concluded that emo-

tional experiences and self-esteem are associated with both paranoia and

hallucinations. The findings give support to the postulation that especially

anxiety is important in the relationship with paranoia, probably because

both anxiety and paranoia can be seen as treat beliefs.

ID: 549120

PHENOMENOLOGY OF SCHIZOPHRENIA AND

AFFECTIVE FIRST EPISODE PSYCHOSIS

Cherise Rosen, R. Marvin, J. L. Reilly, O. A. DeLeon,

P. J. Weiden, S. K. Keedy, M. S. Harris, H. Solari, J. A. Sweeney

Psychiatry/Center for Cognitive Medicine, University of Illinois at

Chicago, Chicago, IL, USA

Objective: This study compares the phenomenological dimensions of psy-

chotic disorders at initial presentation of first episode psychosis in both

schizophrenia and affective psychotic disorders. Method: Patients were

recruited from the University of Illinois Medical Center. Lifetime exposure

to psychotropics in all patients was less than 16 weeks, and patients were

evaluated free of pharmacological treatment. Diagnosis was determined us-

ing the DSM-IVtr and consensus procedure including treating clinicians.

All patients were assessed on standardized research instruments evaluating

symptoms and global functioning prior to starting drug treatment. 5-factor

scoring of PANNS dimensions included positive, negative, cognitive, ex-

citement, and depression. A composite Cluster score for PANNS dimen-

sions assessed anergia, thought disturbance, activation, paranoia and

depression. Results: Of the 110 first episode psychosis subjects who partic-

ipated, 56 were diagnosed with schizophrenia, 30 with bipolar disorder, and

24 with unipolar depression. PANSS 5-factor scores at baseline demon-

strated that schizophrenia and bipolar patients showed significantly

more positive symptoms than depressed patients (P < .001) while schizo-

phrenia and depressed patients demonstrated significantly more negative

symptoms than bipolar patients (P < .001). Additionally, patients with

schizophrenia reported greater cognitive symptoms than patients with ei-

ther bipolar or depression (P < .001). Lastly, PANSS cluster scores showed

that schizophrenia and bipolar patients exhibited significantly more

thought disturbance (P = .007), activation (P < .05), and paranoia (P =

.02) than patients with depression. Schizophrenia and depressed patients

exhibited greater anergia (P < .001) than bipolar patients. Conclusions:

This study compared patients with schizophrenia, bipolar disorder and uni-

polar depression at the time of their initial clinical presentation of psycho-

sis. Data emerging from this study suggests that phenomenology of these

disorders is overlapping with a broad co-occurrence of psychopathological

symptom dimensions. However, differences across groups were observed

with acute psychosis severity being greater in schizophrenia and bipolar

disorder than depression. The similarities of psychotic symptoms in schizo-

phrenia and bipolar disorder may suggest some common mechanisms un-

derlying psychosis in these disorders.

ID: 548869

A PHENOMENOLOGICAL STUDY OF NEGATIVE

SYMPTOMS IN SCHIZOPHRENIA: AMOTIVATION

AND ITS RELATIONSHIP TO FUNCTIONING

George Foussias

1,2

, S. Mann

1,2

, G. Remington

1,2

Centre for Addiction and Mental Health, Toronto, ON, Canada;

Department of Psychiatry and Institute of Medical Science, Faculty

of Medicine, University of Toronto, Toronto, ON, Canada

The negative symptoms of schizophrenia are comprised of two key symp-

tom subdomains: 1) diminished expression (affective flattening and

International Congress on Schizophrenia Research

2. 2. Phenomenology 13

poverty of speech); and 2) amotivation (apathy and anhedonia), and

contribute to functional impairment in this illness. Of note, there is ev-

idence that individuals with schizophrenia do not, in fact, experience an-

hedonia (ie. a consummatory pleasure deficit), but rather deficits in

anticipating pleasure. To date we are unaware of any investigation

that has concurrently examined amotivation and these different facets

of the experience of pleasure, and their relationship with functioning

in schizophrenia. This study aims to specifically examine these relation-

ships, with the hypothesis that amotivation bears the strongest relation-

ship to functional outcomes. Outpatients between the ages of 18 and 55

with a diagnosis of schizophrenia, on stable doses of antipsychotic med-

ication, were evaluated with the Scales for the Assessment of Positive

Symptoms (SAPS) and Negative Symptoms (SANS). Amotivation

was assessed with the Apathy Evaluation Scale—Clinician version

(AES-C), anticipatory/consummatory pleasure with the Temporal Expe-

rience of Pleasure Scale (TEPS), and cognition with the Brief Assessment

of Cognition in Schizophrenia (BACS). The Quality of Life Scale

(QLS) was used to evaluate current functioning. Analysis of data

from 11 subjects (mean age of 38, mean duration of illness of 13 years)

revealed significant relationships between QLS scores and total SANS

(r = .645, P = .032), Amotivation subdomain (r = .704, P = .016),

and AES-C (r = .771, P = .005). Other measures including SAPS total

score, Diminished Expression subdomain score, TEPS anticipatory/con-

summatory pleasure scores, and BACS composite score did not exhibit

any significant association with functional status. Negative symptoms

have been implicated in poor functional outcome, reflected here as

well with total SANS and QLS scores. The present data, albeit prelim-

inary, take this issue a step further though, examining different aspects of

thenegativesymptomdomain.Amotivation was strongly correlated

with current functioning, whereas this was not the case for the expressive

component of negative symptoms, or for anticipatory or consummatory

pleasure. These preliminary findings suggest that it is loss of drive, rather

than pleasure, that links negative symptoms to poor functional outcome.

ID: 546949

A META-ANALYSIS AND QUALITATIVE REVIEW

OF THE IMPACT OF FAMILY HISTORY OF

PSYCHOSIS ON PRESENTATION AND COURSE

OF ILLNESS IN SCHIZOPHRENIA

Michelle Esterberg, H. D. Trotman, E. F. Walker

Psychology, Emory University, Atlanta, GA, USA

Research on the influence of a positive family history of psychosis on pre-

sentation and course of illness in schizophrenia has been mixed, and to

date, there have been no comprehensive reviews that have attempted to

bring together this large area of research and determine the exact impact

of this potentially important etiologic factor. Therefore, the current re-

view qualitatively and quantitatively reviews four areas of research rela-

tive to positive family history: age at onset, presentation of positive and

negative symptoms, course of positive and negative symptoms, and treat-

ment response. Results showed that positive family history of psychosis

has a small but significant impact on several of these areas, including age

at onset and its relationship with gender, negative symptoms, course, and

treatment response. While only one moderator analysis was significant,

there were several interesting trends. The findings of the current review

are discussed in light of an extension of the diathesis-stress model for

the development of schizophrenia. Theoretical assumptions and empirical

research are reviewed to support the notion that family history operates

to influence presentation and course through the same mechanisms it

operates through to influence susceptibility. Finally, implications of

the current findings, limitations of the review, and directions for future

research are highlighted.

ID: 546059

AN ALTERNATE EXPLANATION TO ‘‘COGNITIVE

DYSMETRIA’’ IN SCHIZOPHRENIA

Robert G. Bota

Psychiatry, UMKC, Kansas City, MO, USA

Background: Schizophrenia is considered to be an illness in which there is

a generalized ‘‘cognitive dysmetria’’ in which cortical-cerebelar-thalmic-

cortical circuitry shows impairments when compared with individuals with-

out mental illness. Method: We reviewed the literature pertinent to circuitry

abnormalities in schizophrenia. Also, we looked for correlates with severity

of illness. Further we focused on described impairments in various domains

of insight in schizophrenia. Results: Form the data gathered we observed

that insight into the symptoms is less often impaired than insight into the

illness and the consequences of illness. Experiential aspect of perception is

obtained from processing primitive awareness through working memory

and referenced through the association areas. The reported difficulties

for schizophrenia in processing and encoding would lead in time to incre-

mental deteriorations in reality testing. Thinking can be understood as re-

lated to perception (impure cognition) and not related to experience (pure

cognition, eg, close systems such as matematics). We propose that even

though are severe dysfunction in each of the presented domains, they occur

at different degree of severity. The cognitive dysmetria can be explained as

a ‘‘schism’’ in the interplay of impaired perception and faulty cognition.

Discussion: Despite the fact that is a vast literature discussing neurological

abnormalities in schizophrenia, very limited integrative work was done.

Our hypothesis suggests a different way of assessing and understanding

each individual affected with this illness and ultimately proposes to correct

the specific dysfunctions to reduce the decalage between levels of cognition

and perception.

ID: 543795

IMPULSIVITY IN CO-OCCURING SCHIZOPHRENIA

AND SUBSTANCE USE

Stephanie C. Marcello

1,2

, R. Spiga

, S. Silverstein

Division of Schizophrenia Research, University of Medicine and

Dentistry of New Jersey- University Behavioral HealthCare,

Piscataway, NJ, USA;

Department of Behavioral Research,

Temple University Hospital, Philadelphia, OR, USA

Impulsivity is a risk factor associated with substance use disorders. On

paper-and-pencil measures, people with schizophrenia and substance use

disorders have been shown to be more impulsive than their non-using

counterparts. However, there has been little research elucidating the vari-

ous behavioral components which often define the concept of impulsivity.

These components include: temporal discounting, risk taking, underesti-

mating time and failure to inhibit extraneous responding. Specifically,

the study compared people with schizophrenia who did and did not use

cocaine on these behavioral impulsivity measures. One group had a positive

urine drug screen (UDS) for cocaine (N = 20). A second group had a neg-

ative UDS, but a positive cocaine history (N = 20). Finally, the third group

had no history of cocaine use (N = 20). Participants were assessed using

computer-based tasks. Those with a current or past history of cocaine

use had greater means on the Balloon Analogue Risk Task and engaged

in more risk-taking behaviors. People with schizophrenia with a current

or drug use history seemed to be less affected by loss and more attuned

to monetary gains. Contrary to our hypothesis, the study found that

patients with a positive UDS for cocaine selected larger, delayed rewards

over the smaller, immediate rewards. Performances on the immediate/delay

task also suggest greater attentiveness to magnitude of the monetary re-

ward. On the GoStop computer task, a trend was found toward significant

differences between the means for inhibited responses and total number of

International Congress on Schizophrenia Research

14 2. 2. Phenomenology

responses. However, no significant differences were found between groups

for time estimation. These data indicate that depending on the conditions,

people with comorbid schizophrenia and cocaine abuse may demonstrate

more or less impulsivity than people with either disorder alone. Implica-

tions for understanding the interaction between the CNS effects of cocaine

and psychosis will be discussed.

ID: 541250

INSIGHT INTO DEFICITS? THE DEFICIT SYN-

DROME IN FIRST-EPISODE SCHIZOPHRENIA

Hanan D. Trotman

1,2

, M. T. Compton

Department of Psychiatry and Health Behavior, Medical College of

Georgia, Augusta, GA, USA;

Department of Veterans Affairs,

Charlie Norwood VA Medical Center, Augusta, GA, USA;

Department of Psychiatry and Behavioral Sciences, Emory

University School of Medicine, Atlanta, GA, USA

Schizophrenia as it is currently conceptualized is a markedly heterogeneous

disorder. Researchers have long attempted to decrease this variability to

allow greater predictive power regarding illness course and treatment re-

sponse. The aim of the current research is to examine the deficit syndrome,

a category gaining increasing support in the literature. Studies of deficit

syndrome patients have indicated a history of poor functioning (Tiryaki

et al., 2003), but decreased clinical distress (Kirkpatrick et al., 1994, Fenton

and McGlashan et al., 1994) compared to nondeficit patients. Poor insight

in patients with deficit schizophrenia may be one explanation for these find-

ings. One study in a sample of patients with chronic illness found that deficit

patients had poor insight compared to nondeficit patients (Kirkpatrick

et al., 2000). We hypothesized that poor insight would be present early

in the course of illness in deficit patients. Selected clinical correlates of

the deficit syndrome were also examined. We used the Proxy for the Deficit

Syndrome (PDS) to categorize 109 newly diagnosed patients with nonaf-

fective psychosis into deficit (N = 31) and nondeficit (N = 78) groups

(Kirkpatrick et al., 1993) based on Positive and Negative Syndrome Scale

scores. Results indicate that first-episode deficit patients have poorer in-

sight than first-episode nondeficit patients after accounting for age, gender,

hallucinatory behavior, and delusional thinking. Deficit patients exhibited

greater negative symptoms, less severe positive symptoms, and less distress.

Deficit patients also demonstrated greater impairments in global indicators

of functioning (education, marital/relationship status, employment status).

These results were not confounded by greater substance abuse in the deficit

group. Results from the current study further validate the deficit syndrome

as a useful subcategory of schizophrenia in a sample of first-episode

patients. Results also suggest that the treatment of deficit patients may

be complicated by poor insight. Understanding heterogeneity in the early

course of schizophrenia may hasten treatment interventions and enhance

the development of new treatment targets.

ID: 550901

TRAUMA EXPOSURE AND POSITIVE SYMPTOM

SEVERITY IN PRODROMAL PATIENTS

Cheryl Corcoran, J. Thompson, D. Kimhy, S. Schobel,

D. Malaspina

Columbia, New York, NY, USA

Introduction: Although traumatic childhood experiences are commonly en-

dorsed by individuals with schizophrenia, and associated with psychotic

symptoms, little is known about childhood trauma in prodromal or clinical

high risk samples. Methods: We examined the prevalence of childhood

trauma and its symptom correlates in 30 clinical high risk patients, includ-

ing prodromal and depressive symptoms. Results: All forms of childhood

abuse were prevalent, particularly among ethnic minority participants,

among whom trauma was associated with subthreshold psychotic symp-

toms, in particular grandiosity. By contrast, childhood trauma was associ-

ated with depression only in Caucasian participants at heightened risk for

psychosis. Discussion: This exploratory study suggests childhood trauma is

common among prodromal patients and related to symptoms. Further

studies will entail the predictive value of traumatic experiences and their

neuroendocrine correlates.

ID: 551921

THE WINTER BLUES: SEASON OF BIRTH

PATTERNS IN SCHIZOTYPY

Bianca Iglesias, A. S. Cohen

Psychology, Louisiana State University, Baton Rouge, LA, USA

The season of birth effect, defined in terms of an elevated incidence of birth

during certain months, is one of the most replicated findings in schizophrenia

research. The majority of research has reported elevated rates of winter births

for individuals with schizophrenia dating back to 1929. Many theories have

attempted to explain this effect by including risk factors such as maternal

stress and influenza exposure during pregnancy. Although this effect has

been extensively studied in individuals with schizophrenia, it has not been

examined on individuals with schizotypy—those with a putative genetic vul-

nerability who don’t manifest symptomatology. In the current study, we ex-

amined individuals with schizotypy to see whether a pattern of birth rates is

exhibited.Theschizotypalpersonalityquestionnaire (SPQ) was administered

to 1135 undergraduates and comptuted dimensional positive, negative, and

disorganization scores for each individual, We examined 2 issues: 1) whether

schizotypal features were associated with season of birth and 2) how gender

differencesmightmoderatethisrelationship.Wefound that those with higher

positive schizotypy scoresshowed a higher incidenceof winter birth, however

only within males. This study has replicated in schizotypy what has been

foundinschizophrenia—thatwinterbirthsareassociatedwithpositivesymp-

tomotology. However, this relationship was only present for males.

ID: 551862

THEORY OF MIND ABILITIES AND SELF-

CONFIDENCE IN SCHIZOTYPY

Laura Brown, A. S. Cohen

Department of Psychology, Louisiana State University, Baton

Rouge, LA, USA

Social cognitive deficits have been well established in schizophrenia. How-

ever, there have been mixed results in research on social cognition in indi-

viduals with schizotypy, those with the underlying genetic vulnerability that

do not exhibit full-blown symptomology. The current study investigated

theory of mind, one component of social cognition involving the ability

to infer mental states. Participants included 81 psychometrically-identified

schizotypes and 35 controls who completed a computerized theory of mind

task (TOM). This task was designed to be more sensitive than those used in

previous research, and was composed of items from an existing published

hinting task with a true/false response format. We also has participants rate

how confident they were that their answers were correct. Schizotypes per-

formed similarly to controls on TOM, although they were significantly less

confident. Confidence ratings were not related to overall performance but

were positively correlated with performance on true/false items in which

a correct social inference was made. Results suggest that although schizo-

types may have intact theory of mind, they are less confident in their abil-

ities. Among possible interpretations are that: 1) social cognitive deficits

seen in patients with schizophrenia stem from later decline in general cog-

nition in conjunction with lower confidence in social abilities, 2) social cog-

nitive abilities serve as a resilience factor buffering against the eventual

International Congress on Schizophrenia Research

2. 2. Phenomenology 15

decline into psychosis, and that 3) schizotypes may be employing a compen-

satory strategy leading them to the correct responses.

ID: 551824

EVALUATION OF SPEECH MISATTRIBUTION BIAS

IN SCHIZOPHRENIA

Massoud Stephane

1,2

, M. Kuskowski

1,2

, K. McClannahan

C. Surerus

Psychiatry, VA Medical Center, Minneapolis, MN, USA;

Psychiatry, University of Minnesota, Minneapolis, MN, USA;

Psychiatry, University of Washington, seattle, WA, USA

Background: Attributing self-generated speech to others could explain the

experience of verbal hallucinations. This theory was based on experiments

that used non-linguistic, or speech perception paradigms. Speech genera-

tion paradigms compared self-other to other-self misattribution errors. Im-

pairment in the self-other distinction could cause both types of errors. In

this study we compare both self-other and other-self misattribution errors

to errors that are not related to the self-other distinction. Methods: 39

schizophrenia patients and 26 matched healthy controls were required to

distinguish between self-generated, other-generated, and non-generated

(self or other) sentences. The sentences were in the first, second or third

person and read in a male or female voice in equal proportion. Mixed mul-

tilevel logistic regression models were used to investigate the effect of group,

sentence source, pronoun, and gender of the heard sentences on response

accuracy. Results: Patients differed from controls in the recognition of self-

generated and other- generated sentences, but not in general recognition

ability. Pronoun was a significant predictor of response accuracy but with-

out any significant interaction with group. The gender of heard sentences

variable was not significant. Misattribution bias differentiated groups only

in the self-other direction. Conclusions: These data support the theory that

misattributing self-generated speech to other could result in verbal halluci-

nations. Syntactic (pronoun) but not acoustic (gender) manipulations affect

the recognition of self and other generated speech. This indirectly implicates

language, rather than sensory, disorder in the pathogenesis of AVH.

ID: 551803

THC ELICITS NEGATIVE SYMPTOMS IN HEALTHY

CONTROLS: AN EXPERIMENTAL STUDY

Paul D. Morrison, S. Kapur, R. M. Murray

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom

Background: It is well recognised that cannabis can induce transient

positive psychotic symptoms in healthy subjects and in people with schizo-

phrenia. The psychotogenic effects are attributable to the molecule delta-9-

tetrahydrocannibinol (THC). Less well understood is the effect of cannabis/

THC on the negative dimension. It has been suggested that patients who use

cannabis experience less negative symptomatology compared to those who

don’t. In contrast, there is some experimental evidence that THC may ac-

tually induce schizophrenia-like negative symptoms. Here we report the

effects of pure THC on the negative dimension in a sample of healthy

male subjects. Method: A validated structured instrument, The Community

Assessment of Psychic Experiences (CAPE-state version) was used to assess

drug effects on the negative dimension in 22 participants. The CAPE dis-

tinguishes between positive, negative and depressive dimensions. Pharma-

ceuticals were administered by intravenous injection under randomised,

double-blind placebo controlled conditions. The dose of THC (2.5 mg)

was chosen on the basis of previous studies. Ratings were made at baseline

and at 30, 80 and 120 minutes post-injection. Plasma levels of THC and its

major metabolites were measured over the course of the experiment.

Results: Nineteen participants completed the paradigm. Overall, subject-

rated scores on the CAPE negative dimension were increased from baseline

following THC but not placebo administration (Friedman’s v

= 25.3, df =

7, P = .001). At 30 minutes post THC, CAPE negative scores had increased

by a mean of 4 points, returning to baseline levels by 120 minutes post-in-

jection. There was wide inter-individual variation in responses to THC.

Overall, 70% of subjects reported an increase in the CAPE negative dimen-

sion following THC whilst 15% showed no change and 15% reported a de-

crease. Plasma levels of THC were similar to those observed following

inhalational use of cannabis. Conclusion: In a majority of healthy subjects,

pure THC elicits negative symptoms. Further work to explore the under-

lying mechanism is underway in our lab.

ID: 551761

PERSISTENCE OF RATER TRAINING EFFECTS IN

SCHIZOPHRENIA CLINICAL TRIALS

David Gordon Daniel

, J. J. Bartko

, M. H. Allen

Training and Education, UBC, McLean, VA, USA;

Statistics,

Private Consultant, Newville, PA, USA;

Psychiatry, University

of Colorado Health Sciences Center, Denver, CO, USA

Background: We report a retrospective assessment of rater drift in 9 CNS

trials involving the Positive and Negative Syndrome Scale (PANSS) and

Negative Symptom Assessment Scale (NSA-16) in schizophrenia. We hy-

pothesized that measures of rater adherence to scoring would be signifi-

cantly lower at study mid-point than at initiation. Method: All trials

included had at least one post-investigator meeting recertification or re-

fresher procedure at least 6 months (range approximately 6–19 months) af-

ter the initial approval. For each scale, raters were trained to rate these

scales by viewing at least one training lecture and viewing and rating at least

one videotaped practice subject interview using the scale, followed by de-

tailed feedback on proper scoring from an expert trainer utilizing an audi-

ence response system and expert panel consensus scoring explanations.

Subsequently, in order to be approved to rate, investigators were evaluated

on their ratings of an additional videotaped subject interview. Approxi-

mately 9–16 months later, raters were required to complete a self-

instructional review of rater scoring technique using internet based training

or instructional CD and retested on their ability to properly score a video-

taped interview. For purposes of this analysis, rater performance was eval-

uated as passing if 80% of the scale items were scored within one point of the

modal score or within 1 point of the expert consensus. The proportion of all

raters who scored within one point of the modal score or panel expert con-

sensus on at least 80% of the scale items was compared at study initiation

and mid-study using the chi-square statistic. Results: For schizophrenia

(PANSS and NSA-16), rater performance at mid-study compared to study

initiation was not statistically significant (v

=2.66; P NS; df = 1). At initial

testing 85.9% (n = 1198) of raters met criteria considered acceptable for

passing and 14.1 % (n = 197) did not. At retesting 88.6% (n = 528) of raters

met criteria considered acceptable for passing and 11.4% (n = 68) did not.

Discussion: A retrospective analysis of 9 CNS clinical trials in schizophre-

nia suggests relatively high levels of rater agreement may be sustainable 6–

19 months after initial training with an instructional review at mid-study.

ID: 551757

EXPLORING THE INTERPLAY BETWEEN

MARIJUANA USE, MEDICATION ADHERENCE,

AND PSYCHOTIC SYMPTOMS IN FIRST-EPISODE

PSYCHOSIS

Kia Faridi

1,2

, A. Malla

1,2

International Congress on Schizophrenia Research

16 2. 2. Phenomenology

Department of Psychiatry, McGill University, Montreal, QC,

Canada;

Clinical Research Division, Douglas Mental Health

University Institute, Montreal, QC, Canada

Marijuana use is a frequently present co-morbidity among people with a

first episode of psychosis (FEP), and has an impact on symptoms and on

outcome. The objective of this study was to examine, in the context of

a FEP program, the proportion of marijuana users who persist in using

during the first year in treatment. The study also examined the interaction

between drug use, medication compliance, and symptomatology. The

study was conducted at PEPP-Montreal, a specialized integrated treat-

ment service for all incident FEP cases within a defined area. Subjects

were 192 consecutive admissions, aged between 14 and 30, suffering

from at least two weeks of psychotic symptoms meeting DSM-IV syndro-

mal criteria, and with no history of antipsychotic treatment. Diagnoses,

including co-morbid substance use diagnoses, were determined using the

Structured Clinical Interview for DSM-IV (SCID-I), administered shortly

after program admission, and again at one year. Symptoms were evalu-

ated at regular intervals through one year of follow-up using the Positive

and Negative Symptom Scale (PANSS). Antipsychotic medication com-

pliance was also evaluated regularly based on patient report and corrob-

orated by collateral report and clinical notes. 56 subjects (29.2%) had

a current cannabis use disorder at baseline. Of the cannabis-using subjects

for whom information was available at one year (50), 29 (58.0%) had con-

tinued to use marijuana, while 21 (42.0%) had stopped. PANSS total score

did not differ between stable users compared to those who abstained,

either at baseline (75.1 vs. 77.6, t = 0.47, P = .64) or at one year

(56.2 vs. 50.3, t = 1.40, P = .17). The groups did not differ in their rate

of compliance at baseline (85.2% vs. 90.5%, Fisher’s P = .68). However,

at month 12 stable users were significantly more compliant than cannabis

users who had stopped (92.3% vs. 60.0%, Fisher’s P = .01). An ANCOVA

revealed that PANSS total at 12 months was associated with compliance

at 12 months, F

1,39

= 5.28, P = .03, but that the effect of stability of mar-

ijuana consumption remained significant after controlling for this covari-

ance, F

2,39

= 17.16, P < .001. These results suggest that FEP patients who

continue to use marijuana are more adherent to medications than abstaining

users, and that this effect is not explained by patients who, having stopped

marijuana, felt they no longer need medications to control symptoms.

ID: 551746

DEFINING THE RELATIONSHIP BETWEEN

CLINICAL RATINGS OF BLUNTED AFFECT AND

BEHAVIORAL CODING OF FACIAL AFFECT:

FREQUENCY, INTENSITY, OR DURATION?

Bryann Renee Baker, K. Llerena, S. Lin, C. Forbes,

J. J. Blanchard

University of Maryland, College Park, College Park, MD, USA

Negative symptoms within schizophrenia include profound changes in the

expression of emotion as reflected in the symptom domain of flat or

blunted affect. Despite the importance of this domain it remains unclear

how expression is altered. Specifically, it is not yet known whether dimin-

ished expression reflects changes in the frequency, duration, or intensity of

emotional expression. In order to better understand the nature of blunted

affect, the present study conducted a detailed assessment of facial affect

using the Facial Affect Coding System (FACES; Kring and Sloan, 2007).

A further refinement in this study was the use of the newly developed

NIMH-MATRICS Negative Symptom Rating Scale (NSRS) for the as-

sessment of negative symptoms. 40 outpatient schizophrenics were

assessed with diagnostic and symptom interviews including the NSRS

and BPRS. Videotaped interviews were subsequently rated for facial af-

fective displays using the FACES. Data analyses will examine the relation-

ship between clinical ratings of blunted affect and behavioral coding of

the frequency, duration, and intensity of facial affective displays. Further,

the association between other negative, psychotic, and affective symptoms

will be examined.

ID: 551706

CANNABIS USE AND CLINICAL AND

FUNCTIONAL OUTCOME IN PRODROMAL

ADOLESCENTS AND YOUNG ADULTS

Andrea Auther

1,2

, C. Smith

1,2

, P. Nagachandran

1,2

M. Akerman

1,2

, K. Candenhead

, B. Cornblatt

1,2

Psychiatry Research, The Zucker Hillside Hospital, Glen Oaks,

NY, USA;

Center for Translational Psychiatry, Feinstein Institute

for Medical Research, Manhasset, NY, USA;

Department of

Psychiatry, University of California, San Diego, San Diego, CA,

USA

The Recognition and Prevention Program has recently found that can-

nabis use is not related to worsening of positive symptoms or to conver-

sion to psychosis. However, theNorthAmericanProdromal

Longitudinal Study (NAPLS), a consortium of eight longitudinal pro-

dromal programs, has found that drug use in general is related to con-

version (Cannon et al. 2008). This poster will involve a more intensive

evaluation of the relationship between cannabis use and prodromal

symptoms, social and role functioning, and conversion to psychosis uti-

lizing the NAPLS retrospective database. Subjects include 330 prodro-

mal individuals who met criteria for Attenuated Positive Symptom

Syndrome according to the Scale of Prodromal Symptoms. Level of can-

nabis use was divided into those with no or low use and those with abuse

or dependence. Social and role functioning was assessed with rating

scales developed for NAPLS. Baseline and 12 month follow up data

were utilized. Of the 330 participants, 64 (19.4%) met the criteria for can-

nabis abuse or dependence and the remainder reported no or low levels of

cannabis use. Cannabis misusers were older at baseline but did not differ

from no/low users on other demographic variables. In terms of prodro-

mal symptoms, there were no significant differences between cannabis

misusers and no/low users on attenuated positive or negative symptoms

at baseline. However, cannabis misusers were significantly more likely to

convert to psychosis over follow up (38.2% vs. 24.9%; P = .038). 33% of

participants with follow up cannabis data (68/205) changed their level of

cannabis use over the follow up period. In terms of functioning, subjects

with cannabis misuse at baseline had significantly better social function-

ing than those who had no/low use (P = .021), but there were no differ-

ences on role functioning. These results indicate that cannabis misuse is

relatedtoconversiontopsychosisintheNAPLSretrospectivesample,

although cannabis misusers could not be distinguished from no/low users

on attenuated symptoms at baseline. Cannabis misusers were older

and had better social functioning. Further analyses will examine

whether prodromal symptoms and cannabis use fluctuate in relation

to each other.

ID: 551705

Reference

1. Cannon TD, Cadenhead K, Cornblatt B, et al. Prediction of psychosis

in youth at high clinical risk: a multisite longitudinal study in North

America. Arch Gen Psychiatry Jan. 2008;65(1):28–37.

ASSESSMENT OF NEUROLOGICAL SOFT SIGNS IN

POSITIVE AND NEGATIVE SCHIZOTYPY

Jessica Anna Kaczorowski

, N. Barrantes-Vidal

, T. R. Kwapil

International Congress on Schizophrenia Research

2. 2. Phenomenology 17

Psychology, University of North Carolina at Greensboro, Greens-

boro, NC, USA;

Departament de Psicologia Clı

nica i de la Salut,

Universitat Autonoma de Barcelona, Barcelona, Spain

Patients with schizophrenia often exhibit structural brain abnormalities, as

well as neurological soft signs, consistent with the conceptualization of

schizophrenia as a neurodevelopmental disorder. Neurological soft signs

are mild, presumably nonlocalizing, neurological impairments that are

inferred from performance deficits in domains such as sensory integration,

motor coordination, and motor sequencing. The vulnerability for schizo-

phrenia is presumed to be expressed across a broad continuum of impair-

ment referred to as schizotypy. It is hypothesized that nonpsychotic people

along the schizotypy continuum should exhibit elevated rates of neurolog-

ical soft signs. The present study examined the relation between psychomet-

rically identified positive and negative schizotypy and neurological soft

signs in a nonclinically ascertained sample of young adults (n = 177). Pos-

itive and negative schizotypy dimensional scores were derived from the Per-

ceptual Aberration, Magical Ideation, Physical Anhedonia and Revised

Social Anhedonia Scales. The neurological assessment was conducted using

the Neurological Evaluation Scale (NES). The battery consists of 26 tasks

that tap three neurological soft sign domains including sensory integration,

motor coordination, and motor sequencing. As hypothesized, negative, but

not positive, schizotypy was related to increased neurological soft signs as-

sociated with deficits in tasks related to fine and gross motor coordination,

motor sequencing, eye movement abnormalities, and memory recall. How-

ever, positive schizotypy was associated with deficits in a few tasks related

to sensory integration dysfunction. In general, the positive x negative schiz-

otypy interaction term was unrelated to soft signs. Recommendations are

offered to strengthen the utility of the NES for use with nonclinical samples,

including using a modified version of the battery and a continuous scoring

system. The findings support: a) the theory that the vulnerability for schizo-

phrenia is expressed across a broad continuum of subclinical and clinical

impairment referred to as schizotypy; b) the multidimensional structure of

schizotypy; and c) the notion that schizotypy may be a more appropriate

construct for understanding the etiology and development of spectrum psy-

chopathology than full-blown schizophrenia.

ID: 551536

ASSESSING SUICIDIDAL BEHAVIOR IN SCHIZO-

PHRENIA: PSYCHOMETRIC PROPERTIES OF

A BRIEF SELF-REPORT MEASURE

Jill Harkavy-Friedman

Psychiatry, Columbia University, NYSPI, New York, NY, USA

Individuals with schizophrenia are at risk for suicidal behavior. Careful and

ongoing assessment is important for detecting risk.The HASS is a 21-item

self-report measure of suicidal behavior assessing suicidal behavior over

the past 2 weeks (HASS I) and the patient’s whole life except the past 2 weeks

(HASS II). The strong psychometric properties and the application of this

measure will be presented. Data will be presented demonstrating the efficacy

of the HASS at identifying suicidal behavior and measuring change with

treatment. Over 140 individuals with Sz/SA disorder completed the

HASS. Three factors accounting for 59% of the variance were identified

for the HASS I: 1) thoughts of life, death and suicide; 2) suicidal behaviors

including attempts; 3) substance use. Four factors were identified for the

HASS II accounting for almost 67% of the variance: 1) Thoughts of suicide;

2) suicidal plans and actions; 3) substance use; and 4) thoughts of impulsive

suicidal behavior. Data will be presented regarding relationships with de-

pression, psychotic symptoms, impulsivity, aggression and the Beck Scale

for Suicidal Ideation. The findings will be compared with a sample of

adolescent high school students and with a sample of adolescents hospital-

ized for suicidal behavior in order to demonstrate the external validity of the

instrument. Sixty-five individuals were administered the HASS I on admis-

sion and discharge to the Schizophrenia Research Unit at New York State

Psychiatric Institute and the HASS II was administered on admission. The

HASS I detected attempts during the 2 weeks prior to admission and coupled

with the HASS II identified 100% of documented attempters in the sample.

There was a significant decrease in suicidal behavior from admission to

discharge (P < .001). The relationship between change in suicidal behavior

and change in other symptoms will be presented. The ability to assess suicidal

behavior in Sz regularly and thoroughly is important. The HASS is the only

self-report measure in the literature available at this time. The reliability is

excellent and a significant amount of validity data has been accrued. The

hope is that the availability of this measure will increase the likelihood

of the assessment of suicidal behavior in research as well as clinically. Sup-

ported by NIMH RO1 MH 56422-O1A, 2 P30 MH4674.

Reference

1. Harkavy JM, Asnis GM. Assessment of Suicidal Behavior: A new

instrument. Psychiatric Annals. 1989;19(7):382–387.

ID: 551535

THE PSYCHOTIC BRAIN-MIND BEYOND THE

KRAEPELINIAN DICHOTOMY: SIMILAR LEVELS

OF COGNITIVE BIZARRENESS FOUND IN THE

DREAM AND WAKING MENTATION OF PATIENTS

DIAGNOSED WITH BIPOLAR DISORDER AND

SCHIZOPHRENIA

Ivan Limosani, A. D’Agostino, M. L. Manzone, S. Scarone

Department of Psychiatry, Universita

degli Studi di Milano Azienda

Ospedaliera San Paolo, Milano, Italy

Similarities and differences were evaluated in terms of structural organiza-

tion of thought across two diagnostically distinct groups of actively psy-

chotic subjects, by measuring cognitive bizarreness in their dream and

waking mentation. Dreams reported upon awakening and narrative

responses to a standardized projective stimulus, the Thematic Apperception

Test (TAT), were collected from 20 acutely manic inpatients with psychotic

symptoms, 20 actively psychotic schizophrenic inpatients and 20 subjects

with no present or past history of mental illness, all adequately matched

in terms of age and education. The material was scored by two highly-trained

judges who evaluated the degree of bizarre elements referred to the formal

architecture of verbalized thoughts according to the dream bizarreness scale

(Scarone et al., 2008). A mean Bizarreness Density Index (BDI) was then

elaborated for each of the three experimental groups’ sets of dream and

TAT reports; analysis of the data indicated a substantial similarity in terms

of cognitive bizarreness in all reports, with the only statistically significant

outlier being the control group’s waking mentation (see Table). Cognitive

bizarreness can be interpreted as the phenomenological correlate of the spe-

cific activation pattern occurring in the brain during REM sleep, where

dream imagery is more vivid and storylines more articulated and incongru-

ous. Similar experimentally measurable and clinically objectivable depar-

tures from the physiological functioning of the brain-mind have been

shown to underlie dreaming and schizophrenia (Scarone et al., 2008).

The results found in this study support the view of psychotic mentation

as a correlate of the same type of neurobiological activation underlying

dream sleep. Moreover, waking levels of cognitive bizarreness appear to

be shared amongst acutely manic bipolar patients and actively psychotic

schizophrenic subjects, linking psychosis to a specific shift in brain-mind

functioning independent of diagnostic categories.

Reference

1. Scarone S, et al. The dream as a model for psychosis: an experimental

approach using bizarreness as a cognitive marker Schizophrenia Bulle-

tin. 2008;34(3):515–22.

International Congress on Schizophrenia Research

18 2. 2. Phenomenology

Table. Descriptive statistics (Mean BDI Standard Deviation)

DREAM BDI TAT BDI

Control subjects 0.29

6 0.20 0.11 6 0.11

Schizophrenic subjects 0.32

6 0.13 0.32 6 0.19

Manic bipolar subjects 0.33 6 0.11 0.29 6 0.15

ID: 551484

HIGH FREQUENCY BAND ACTIVITY IN RESTING

EEG AND THE RELATIONSHIP WITH SMOKING IN

SCHIZOPHRENIA

Alice Saperstein

, A. Summerfelt

, L. E. Hong

, R. Buchanan

G. K. Thaker

Psychology, University of Maryland, College Park, MD, USA;

Psychiatry, University of Maryland School of Medicine, Baltimore,

MD, USA

Evidence suggests that high nicotine dependence observed in schizophrenia

is related to its core neuronal deficits such as abnormalities in neural syn-

chronization and sensory gating deficits. Some of these neuronal deficits are

shown to mark schizophrenia liability, raising the possibility that the in-

creased nicotine dependence in schizophrenia is related to its etiological fac-

tors. In an ongoing study, we are examining the rates of nicotine

dependence in families of schizophrenia probands using the Fagerstrom

Test for Nicotine Dependence, and the relationships between nicotine de-

pendence and schizophrenia endophenotypes in these subjects. Rates of nic-

otine use were significantly greater in patients (n = 103) than controls (n =

77) with respect to a history of ever smoking (69.9% vs 46.8%) and current

smoking (58.3% vs 23.4%) with patients reporting greater levels of nicotine

dependence (FTND; F

1,103

= 20.33, P < .001). The two phenotypes, ie, ever

smoking and current smoking, were associated with moderate to high her-

itability among schizophrenia families (heritability estimates > 0.9) and

control families (> 0.5). We are in the process of examining their relation-

ships with schizophrenia endophenotypes; at present data are available only

in schizophrenia probands and control subjects on resting EEG measures

and P50 gating. Resting EEG (5 min. sample) were analyzed to yield ab-

solute power averages in alpha (8–10Hz), beta (12–28Hz) and gamma (30–

50Hz) frequency bands examined at electrode sites Cz and Fz with repeated

measures ANOVA to test for diagnostic differences and interactions with

smoking status. No main effect of diagnosis was found for power in the

alpha, beta, or gamma bands (all p values > 0.20), but classification by

never, former, or current smoker yielded a significant main effect in the

gamma band (F

2,173

= 3.34, P = .038) and a trend towards significance

for the beta band (F

2,174

= 2.62, P = .076); post-hoc tests demonstrated sig-

nificantly lower gamma band activity in current smokers than former

smokers (P = .037). There was a significant group by smoking status inter-

action for P50 ratio (F

1,81

= 9.12, P = .003) whereby sensory gating deficits

were normalized in patient but not control smokers. We are in the process

of analyzing data in the relatives. These results suggest that further analysis

of high frequency bandpower in resting EEG may provide important links

between neurophysiological deficits associated with schizophrenia and

smoking.

ID: 551482

SELF-DISORDERS IN THE PRODROME: A PILOT

STUDY AMONG HELP-SEEKING ADOLESCENTS

Danny Koren

, N. Reznik

, M. Herman

, R. Rozilio

, J. Parnas

Psychology, University of Haifa, Haifa, Israel;

Psychiatry,

University of Copenhagen, Copenhagen, Denmark

The overarching objective of this pilot study was to lay the ground for

a large-scale, community-based, prospective study that will examine the po-

tential of anomalous self-experiences as an early, pre-psychotic phenotype

of emerging schizophrenia. More specifically, the present study had two

interrelated goals: 1) to demonstrate the feasibility of identification of

anomalous self-experiences among non-psychotic help-seeking adolescents,

and 2) to examine the potential of these experiences to serve as an early risk

marker by examining their relationship with level of worry for a future psy-

chotic illness as rated by an experienced child psychiatrist, presence and

severity of prodromal symptoms, and decline in level of functioning.

The study was motivated by the view that present research is not guided

by a comprehensive psychopathological theory regarding the nature of

the pre-psychotic phase. To accomplish these goals we administered the Ex-

amination of Anomalous Self-Experience (EASE: Parnas et al. 2005) to

a group of 30 adolescents seeking help for emotional and behavioral prob-

lems. In addition, all participants were assessed with an extensive battery

measuring conventional genetic and clinical risk factors such as familial

predisposition, prodromal symptoms, emotional distress, psychosocial

functioning, and (meta-)cognitive functioning. All assessments were carried

out in complete blindness to the level of worry for future psychotic illness as

assessed by the intake psychiatrist. A wide range of anomalous self expe-

riences has been observed in about two thirds of the sample. The prevalence

and level of severity of these experiences were significantly correlated with

level of worry as assessed by the child psychiatrist, as well as with level of

prodromal symptoms, emotional distress, and depth of cognitive and meta-

cognitive deficits. If further validated, these preliminary findings suggest

that anomalous self-experiences can enrich current early detection models

by expanding the focus to the subjective experiences that may antecede the

more descriptive-observational at-risk phenomena.

ID: 551427

RATES AND CORRELATES OF PAST INCAR-

CERATION AMONG PATIENTS HOSPITALIZED

FOR A FIRST-EPISODE OF NONAFFECTIVE

PSYCHOSIS

Claire Ramsay, S. M. Goulding, B. Broussard, S. Cristofaro,

G. Abedi, M. T. Compton

Department of Psychiatry and Behavioral Sciences, Emory

University School of Medicine, Atlanta, GA, USA

Background: Population-based and clinical studies attest to the fact that

individuals with schizophrenia have high rates of arrest and incarceration.

However, little is known about the prevalence of arrest and incarceration in

first-episode psychosis samples, or social and clinical correlates of such le-

gal system involvement occurring prior to initial treatment seeking. This

study examined prevalence of arrest/incarceration and associations be-

tween incarceration and demographic, premorbid functioning, and clinical

variables in a first-episode sample. Methods: The sample consisted of 109

urban, predominantly African American patients with first-episode

nonaffective psychosis, hospitalized in a public-sector setting. Assessments

included the Premorbid Adjustment Scale and the Positive and Negative

Syndrome Scale (PANSS), along with a number of other measures. Results:

Rates of past arrest and incarceration were very high in this sample (77%

and 63%, respectively). Patients with a history of incarceration had

completed fewer years of education (10.9

6 2.1) compared to those who

had not been incarcerated (12.6

6 2.4; t = 3.9, P < .001). Past incarceration

was associated with poorer academic functioning in early (P = .020) and late

adolescence (P = .007), though no association was found with academic

functioning before the age of 11 years or social functioning during any pre-

morbid period. Past incarceration was significantly associated with the

presence of alcohol (P = .017) and cannabis (P = .003) abuse or dependence.

Of those having ever used cannabis, patients who had been previously in-

carcerated had a mean age at first reported cannabis use of 15.2

6 4.1, com-

pared to those who had not been arrested (17.0 6 3.6; t = 2.0, P = .049).

International Congress on Schizophrenia Research

2. 2. Phenomenology 19

There was a significant association between past incarceration and the total

number of Axis IV psychosocial and environmental stressors (P = .002),

PANSS positive symptoms, and PANSS general symptoms, each indicating

poorer functioning among those having been previously incarcerated. Dis-

cussion: Although criminalization of individuals with serious mental ill-

nesses is a widely recognized problem, it is typically conceptualized as

a function of chronicity of illness. These data indicate that even prior to

the first treatment contact, this sample had already accumulated substantial

psychosocial consequences in terms of arrest/incarceration. Each signifi-

cant correlate of past incarceration suggested grave implications for symp-

tomatic and social functioning.

ID: 551399

DIFFERENTIAL HEDONIC CAPACITY IN

SCHIZOPHRENIA AND BIPOLAR DISORDER.

Tyler Barrett Grove, I. F. Tso, S. F. Taylor

Psychiatry, University of Michigan, Ann Arbor, MI, USA

Emotional abnormalities characterize both affective and non-affective psy-

chosis, and both schizophrenia and bipolar disorder have poor functional

outcomes. However, socio-emotional and volitional impairment are central

to the Kraepelinian distinction between the groups, suggesting that emo-

tional measures should differentiate them. To investigate this question,

we administered several emotion function scales to patients with affective

and non-affective psychosis in order to differentiate hedonic capacity. Sta-

ble outpatients (43 with schizophrenia/schizoaffective disorder [SCZ]; 27

with bipolar disorder I/II [BP]) and 36 matched healthy subjects (HC)

were administered tests of hedonic capacity (Physical and revised Social

Anhedonia scales[PAS, RSAS]), basic drive and inhibition (Behavioral In-

hibition and Activation Scales [BIS-BAS]), and consummatory/anticipa-

tory hedonic capacity (Temporal Experience of Pleasure Scale –TEPS).

Subjects also completed a measure of anxiety/negative affect (State-trait

anxiety, STAI) and the Social Adjustment Scale (SAS). Twelve-month fol-

low-up data is reported from 33 SCZ and 18 BP subjects. Internal reliability

of the self-rating scales was favorable for both groups (alpha > 0.75). Both

patient groups exhibited significant negative affect (STAI: F = 26.7, P <

.000, SCZ = BP > HC), and significant functional social impairment

(SAS: F = 20.9, P < .000; SCZ = B, P < HC). On traditional anhedonia

measures, both groups were impaired (PAS: F = 14.7, P < .000, SCZ =

BP > HC; RSAS: F = 9.7, P < .000, SCZ = BP > HC). However, the

TEPS differentiated the groups (F = 5.8, P = .004, SCZ<B, P < HC), as

did BAS reward reinforcement (F = 6.8, P = .002, SCZ < BP = HC)

and BAS drive (F = 4.7, P = .01, SCZ < BP = HC). The BP group exhibited

greater inhibition than the HC group (F = 6.2, P = .003, BP > HC). The

TEPS did not show differential impairment according to anticipatory or

consummatory pleasure, i. e. no group by subscale interaction (P > .5).

However, in SCZ only the TEPS anticipatory subscale correlated with cli-

nician-rated anhedonia (r =  .33, P < .05), predicted current social adjust-

ment (r = .35, P < .05), and predicted social adjustment at 12 months (r =

.45, P < .01). Negative affect (STAI) predicted social adjustment in SCZ

(r = .58, P < .01) and BP (r = .56, P < .05). The findings illustrate the phe-

notypic overlap of the two groups, but also support the Kraepelinian poles

of affective and non-affective psychosis by way of hedonic capacity and

emotional drive.

ID: 551292

AN INVESTIGATION OF NEGATIVE AFFECT AND

POSITIVE SYMPTOMS OF PSYCHOSIS

Audra Crutchfield

, C. Neumann

, P. Wupperman

Psychology, University of North Texas, Denton, TX, USA;

Yale University, New Haven, CT, USA

Disturbances of negative affect, such as depression and anxiety, are con-

sistently identified as core features of both psychotic and psychosis-prone

(at-risk) individuals and represent important treatment considerations for

individuals with psychotic disorders. Vulnerability-stress models suggest

that individuals with a predisposition to psychosis may have heightened

emotional reactivity to environmental, biological, and psychosocial stres-

sors – and therefore may be more susceptible to negative affective states

(Asarnaw, Cromwell, and Rennick, 1978; Meehl, 1962; Walker and Difor-

io, 1997). Additionally, emerging evidence suggests a cross-sectional as-

sociation between negative affect and positive symptoms of psychosis,

such as hallucinations, delusions, and thought disorder; however, few lon-

gitudinal studies have examined these relationships (eg, Docherty, Grosh,

and Wexler, 1996; Docherty, Rhinewine, Nienow, and Cohen, 2001). The

current study examined cross-sectional and longitudinal associations be-

tween disturbances in negative affect (depression and anxiety) and posi-

tive symptoms of psychosis (hallucinations, delusions, and thought

disorder). Specifically, we hypothesized synchronous (cross-sectional)

associations between negative affect and positive psychosis and believed

these associations would remain stable over time after accounting for

baseline levels. It was also hypothesized that negative affective traits at

baseline would predict an increase in severity of positive symptoms of psy-

chosis among those individuals with a diagnosis of psychotic disorder at

subsequent time frames (10 and 20 weeks). Data were drawn from the

MacArthur Violence Risk Assessment study and included individuals

(N = 245) with a diagnosis of psychotic disorder and scores of individual

items on the Brief Psychiatric Rating Scale. Results of Confirmatory Fac-

tor Analysis and Structural Equation Modeling suggested good fit to the

data (CFI = .94, RMSEA = .05, SRMR = .05, BIC = 12023). Analyses

revealed moderately strong, statistically significant cross-sectional asso-

ciations between negative affect (Depression-Anxiety) and positive symp-

toms of psychosis. These correlations remained stable over time after

accounting for baseline and previous effects of the latent variables. Con-

trary to hypotheses, no significant cross-lagged effect was identified,

suggesting no directional relationship between negative affect and positive

psychotic symptoms.

ID: 551278

22Q11DELETION SYNDROME AND SCHIZOTYPAL

PERSONALITY DISORDER: BEHAVIORAL AND

PRODROMAL SYMPTOMS IN INDIVIDUALS AT

HIGH RISK FOR PSYCHOSIS

Daniel Shapiro

, J. F. Cubells

, K. Rockers

, O. Ousley

E. F. Walker

Department of Psychology, Emory University, Atlanta, GA,

USA;

Department of Psychiatry and Behavrioal Sciences, Emory

Autism Center, Emory University School of Medicine, Atlanta,

GA, USA

Adolescents with 22q11 Deletion Syndrome (22q11DS) and Schizotypal

Personality Disorder (SPD) are at increased risk for the development of

psychosis. This is evidenced by rates of illness much higher than those in

the general population, as well as by the presence, in both, of subthresh-

old psychotic symptoms. The current study compares the behavioral and

prodromal symptom profiles of adolescents in genetic (22q11DS) and

behavioral (SPD) high risk groups. While many studies have investigated

the behavioral and symptom profiles in each of these groups, no pub-

lished report has directly compared them with each other. Doing so

can potentially shed light on the relevance of etiological subtypes for un-

derstanding the pathogenesis of schizophrenia. The current study used

two measures: the Achenbach Child Behavior Checklist (CBCL; SPD:

n = 13, 22q11DS: n = 13), and the Structured Interview for Prodromal

Syndromes (SIPS; SPD: n = 23, 22q11DS: n = 23).Thetworiskgroups

International Congress on Schizophrenia Research

20 2. 2. Phenomenology

were matched on age, gender, and where possible, ethnicity. PRELIM-

INARY RESULTS: CBCL—Individuals in the SPD group had scores

that were in the clinically significant range, and higher than those in

the 22q11DS group on the anxious/depressed, social problems, thought

problems, attention problems, internalizing problems, and total prob-

lems scales. The 22q11DS group did not have any scores in the clinically

significant range. SIPS—individuals in the SPD group had more positive

symptoms and attenuated ideational richness, as well as fewer motor

problems. Both groups also showed heightened attention problems

and elevated anhedonia scores. Thus, results show both similarities

and differences between these two high risk groups, suggesting that

individuals with 22q11DS and those with SPD may share some of the

same risk factors for the development of schizophrenia. Results are

consistent with the notion that genes mapping the 22q11 chromosomal

region are involved in the pathogenesis of schizophrenia and

spectrum disorders.

ID: 551246

THE PATTERN OF CANNABIS USE AND WITH-

DRAWAL SYMPTOMS IN PEOPLE WITH SCHIZO-

PHRENIA

Douglas Lee Boggs

, D. L. Kelly

, J. Linthicum

, F. Liu

R. P. McMahon

, D. A. Gorelick

School of Medicine, Psychiatry, Maryland Psychiatric Research

Center, University of Maryland Baltimore, Baltimore, MD, USA;

Office of the Scientific Director, Intramural Research Program,

National Institute on Drug Abuse, National Institutes of Health,

Department of Health and Human Services, Baltimore, MD, USA

Chronic users of cannabis often report withdrawal symptoms after ab-

stinence from use (1), but it is unknown whether this occurs in people

with schizophrenia. This study used a structured questionnaire (Mari-

juana Quit Questionnaire (1)) to examine behavioral patterns of canna-

bis use and withdrawal symptoms. Fifty-four people in treatment for

schizophrenia or schizoaffective disorder who were > 18 years old

and reported at least one serious cannabis quit attempt participated

in the study. Volunteers were 40.1

6 10.9 years old (mean 6 SD),

66.76% male, 54% Caucasian, and 37% African-American. First

reported age of cannabis use was 16.5 6 6.1 yrs, and reported age of

onset of psychotic symptoms (n = 50) was 21.0

6 5.6 yrs. Ninety-two

percent (46/50) reported using cannabis before the onset of psychotic

symptoms. The number of lifetime uses of cannabis ranged from 1 to

> 10 000 (63% reported at least 500 uses). Seventeen percent met criteria

for cannabis abuse; 74% met surrogate criteria for cannabis dependence.

Cannabis use had a negative effect on quality of life: 87% (46/53)

reported experiencing psychological or emotional problems and 60%

(32/54) reported physical health problems related to cannabis use. With-

drawal symptoms were reported by 93%; 52% reported between 1–5

withdrawal symptoms, 28% reported 6–10 symptoms, and 20% reported

> 10 symptoms. The most common reported withdrawal symptoms

were: anxiety (50%), depression (46%), craving for cannabis (44%),

and boredom (39%), similar to the symptoms reported by users without

psychiatric comorbidity (1). Ten percent reported using cannabis to re-

lieve withdrawal symptoms. This suggests that withdrawal is a negative

reinforcer for relapse. We conclude that most people with schizophrenia

who use cannabis begin before the onset of psychosis and that with-

drawal symptoms are common in those who have attempted to quit.

Withdrawal symptoms are important in the clinical treatment of schizo-

phrenia and comorbid cannabis use disorders. Supported by the Intra-

muralResearchProgram,NIH,National Institute on Drug Abuse and

the NIDA Residential Research Support Services Contract

HHSN271200599091CADB.

Reference

1. Copersino ML, et al. Cannabis withdrawal among non-treatment-

seeking adult cannabis users. Am J Addict. 2006;15(1):8–14.

ID: 551198

SOCIAL WORLD INTERACTIONS: HOW COMPANY

CONNECTS TO PARANOIA

Dina Collip

, M. Oorschot

, V. Thewissen

, J. van Os

1,2

R. Bentall

, I. Myin-Germeys

1,4

Psychiatry and Neuropsychology, South Limburg Mental

Health Research and Teaching Network, EURON, Maastricht

University Medical Center, Maastricht, The Netherlands, Maas-

tricht, Netherlands;

Division of Psychological Medicine, Institute of

Psychiatry, London, United Kingdom;

School of Psychology,

University of Bangor, Bangor, United Kingdom;

School of

Psychological Sciences, University of Manchester, Manchester,

United Kingdom

Background: Many surveys and experimental studies have indicated that

social contact, even when it is neutral, triggers paranoid thinking in people

who score high on clinical or sub-clinical paranoia. However, little is known

about the relationship between paranoia and social environmental effects in

the context of daily life reality. The current study aims to investigate

whether contextual variables are predictive of momentary increases in

the intensity of paranoid thinking in a sample of subjects with clinical

and non-clinical paranoia. Methods: A sample of 37 control subjects, 37

high-schizotypy participants, and 33 paranoid patients were assessed

with the Experience Sampling Method (a structured diary technique)

assessing momentary social context, appraisal of social company, perceived

social threat and paranoia in daily life. Results: Multilevel analyses revealed

a significant interaction effect between group and being alone in the model

of momentary paranoia (v

(2) = 20.22, P < .00). Patients reported lower

levels of paranoia when in company than when alone (v

(1) = 13.84, P <

.000), whereas no such difference was found in the high-schizotypy group or

controls. With regard to the type of social company, differential associa-

tions were found for momentary paranoia (v

(2) = 28.83, P < .000), and

perceived social threat (v

(2) = 41.88, P < .000) in the three groups. Controls

and high-schizotypy individuals reported higher levels of perceived social

threat when with non-familiar compared to familiar individuals. Moreover,

the high-schizotypy group reported more paranoia in non-familiar com-

pany. The patients, on the other hand, reported no difference in the per-

ception of social threat between familiar and non-familiar contacts.

They even reported less paranoia when in non-familiar compared to famil-

iar company. Discussion: The data suggest that the development of

sub-clinical momentary paranoia is context-dependent. However, once

in a full-blown paranoid episode, momentary paranoia seems to become

more autonomous and independent of the social reality. Being with other

people, especially with non-familiar individuals, may then even serve as a

protective factor.

ID: 551181

DEVELOPMENT OF AN INTERVIEW-BASED

‘‘CO-PRIMARY’’ MEASURE OF COGNITION

Joseph Ventura

, R. M. Bilder

1,2

, S. R. Reise

, R. S. Keefe

Psychiatry, UCLA, Los Angeles, CA, USA;

Psychology, UCLA,

Los Angeles, CA, USA;

Psychiatry, Duke University, Los Angeles,

NC, USA

Background: The MATRICS initiative has reached a consensus that prac-

tical, reliable ‘‘co-primary’’ measures of cognition are needed to evaluate

change in cognitive functioning after new drug interventions (Green

International Congress on Schizophrenia Research

2. 2. Phenomenology 21

et al., 2008). However, for data to be meaningful there should be evidence

that the co-primary instrument effectively measures the appropriate

concept. Method: Data were collected as part of the MATRICS initiative

(n = 176) with the (SCoRS and CGI-CogS; total of 41 items), two interview-

based measures of cognition. Sources of information included a patient,

a caregiver, and the rater. The aim of these analyses was to develop a valid

but brief co-primary measure of cognition. The study investigators used

classical test theory (CTT) including factor analysis and modern test theory

approaches, such as Item Response Theory (IRT) and developed a 10-item

Cognitive Assessment Interview (CAI). The validity analysis focused on the

relationship between the CAI and objective measures of cognitive function-

ing (MCCB), functional outcome (Birchwood Social Functioning Scale),

and psychiatric symptoms (BPRS). Results: For the CAI, both the patient

(r = .82) and the caregiver (r = .95) ratings were highly correlated with the

rater’s composite. The 10-item CAI correlates highly (r = .87) with the com-

bined set of parent instruments (CGI-CogS and SCoRS; 41 items). In ad-

dition, the CAI is moderately correlated with objectively measured

cognition (r = .32) and with functional outcomes (r = .31) at levels

that are comparable the parent instruments. The CAI was correlated

with positive symptoms (r = .28), but not with depression (r = .06) or neg-

ative symptoms (r = .09). Conclusions: The CAI allows a rater to use expert

judgment to evaluate the patient’s or the caregiver’s report of cognitive

functioning. The CAI is experience-near, and given the moderate correla-

tion with neurocognition and functional outcome, shows indications of be-

ing valid. Although the CAI is correlated with positive symptoms, the

ratings are not unduly influenced by all types of symptoms. The CAI is brief

and repeatable, and therefore might be a valuable tool for assessing cog-

nitive functioning for schizophrenia patients who are participating in clin-

ical trials.

ID: 551133

THE ECOLOGICAL VALIDITY OF NEGATIVE

SYMPTOMS

Margreet Oorschot

, V. Thewissen

1,2

, J. van Os

1,3

I. Myin-Germeys

1,4

Department of Psychiatry and Neuropsychology, South Limburg

Mental Health Research and Teaching Network, EURON, Maas-

tricht University, Maastricht, Netherlands;

Faculty of Psychology,

Open Universiteit Nederland, Heerlen, Netherlands;

Division of

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom;

School of Psychological Sciences, University of Man-

chester, Manchester, United Kingdom

Flat affect and social anhedonia are assumed to be key characteristics of

negative symptomatology in psychotic disorders. However, the ecological

validity of these concepts remains unclear. This study compared flat affect

and social anhedonia as measured by a structured interview to emotional

experience and social behavior in the realm of daily life. In the current

study, emotional experience and social behavior of 178 patients diagnosed

with psychotic spectrum disorders and 164 controls were explored using the

Experience Sampling Method (a structured diary method). Social

anhedonic and blunted patients were identified using the Positive and Neg-

ative Symptom Scale. Flat affect in daily life was defined as diminished

emotional intensity. Social anhedonia in daily life was conceptualized using

1) frequency of being in company, 2) preference of being alone while in

company and 3) emotional reward of being in company. Multilevel regres-

sion analyses revealed that the patient group as a whole experienced more

intense negative and less intense positive emotions than controls. Blunted

patients, however, did not differ from non-blunted patients in emotional

intensity. Compared to controls, patients were more often alone and

showed a higher preference of being alone when in company. However, so-

cial company was associated with more positive and less negative emotions

in both patients and controls. No differences between social anhedonic and

non-anhedonic patients were found on all three indicators of social anhe-

donia. These results suggest that patients may experience problems with

positive emotions and social isolation in normal reality of daily life. How-

ever, these problems were not validly captured using a structured interview,

since no differences on social anhedonia and flat affect in daily life were

found between patients scoring high or low on the negative symptom items

of the structured interview.

ID: 551083

PRODROMAL SYMPTOMS IN CANNABIS,

KETAMINE AND COCAINE USERS

Celia J. A. Morgan

, S. Duffen

, S. Hunt

, L. Monaghan

O. J. Mason

, H. V. Curran

Clinical Psychopharmacology Unit, University College London,

London, United Kingdom;

Psychology, Univerisity College London,

London, United Kingdom

Three classes of drugs have been repeatedly associated with psychotic phe-

nomena: cannabis, the stimulants (cocaine and amphetamine), and the dis-

sociative anaesthetics (PCP and ketamine) but their users have never been

directly compared on psychotic symptoms. We used a pre-clinical or pro-

dromal measure of symptoms to compare non-psychotic users of these three

groups of drugs. 130 volunteers were rated on the Schizophrenia Proneness

Instrument-Adult (Schulz-Lutter et al. 2008; SPI-A): 29 dependent ‘skunk’

(high potency cannabis) users, 22 dependent cocaine users, 21 dependent

ketamine users, along with 28 ‘recreational’ poly-drug (ketamine, cannabis

and cocaine) users, and 30 drug-naı

ve controls. Group membership was

verified by hair and urine analysis. The SPI-A yields 6 basic symptom ‘fac-

tors’: affective symptoms, attentional symptoms, cognitive disturbances,

perception of self and surroundings, body perception and general percep-

tual changes. SPI-A data were also compared with existing data on prodro-

mal individuals (Kloster-Kotter et al., 2007). Ketamine users showed the

greatest number of affective and perceptual symptoms. Both ketamine

and cannabis users manifested the greatest levels of attentional problems

and cognitive disturbances. Cocaine users showed greater perceptual dis-

turbances than the cannabis group but fewer than ketamine users. Overall

dependent drug users exhibited more symptoms than recreational users on

every factor except disturbances in body perception. Recreational drug

users did not differ from non-drug users in affective symptoms and other

perceptual disturbances. Of the dependent groups, ketamine users pre-

sented with basic symptoms which were the most similar to prodromal indi-

viduals; ‘cognitive’ basic symptoms were also prominent in dependent

cannabis users; the symptom profile of cocaine users was the most subtle.

The drug-related elevation in symptoms was restricted largely to dependent

drug users, with much lower levels of symptoms in recreational drug users.

Few of these drug-dependent individuals will likely transition to psychosis

but these drug induced increases in basic symptoms may represent mech-

anism of risk in people with a predisposition to the disorder. In addition,

clinicians should be aware of the similarity of symptoms observed in drug

users who will likely not transition to psychosis and genuinely prodromal

individuals.

ID: 550978

CHILDHOOD TRAUMA, CORTISOL LEVELS AND

HPA AXIS FUNCTION IN DRUG-NAI

VE FIRST

EPISODE PSYCHOSIS PATIENTS

Belinda Garner

, C. Phassouliotis

, L. J. Phillips

, Y. Yun

C. Markulev

, S. Bendall

, R. Parslow

, C. Leong

P. D. McGorry

Psychiatry, ORYGEN Research Centre, Melbourne, VIC,

Australia;

School of Behavioural Sciences, University of

International Congress on Schizophrenia Research

22 2. 2. Phenomenology

Melbourne, Melbourne, VIC, Australia;

ORYGEN Youth Health,

NorthWestern Mental Health, Melbourne, VIC, Australia

Early life trauma is associated with increased vulnerability for psychiatric

disorders, possibly mediated by alterations in the hypothalamic-pituitary-

adrenal (HPA) axis. This study examined the relationship between child-

hood trauma and HPA axis function in 17 first-episode psychosis (FEP)

patients and 19 healthy controls using the low dose (0.25mg) Dexametha-

sone Suppression Test (DST). Blood samples were obtained at 9am before

and after administration of 0.25mg oral dexamethasone at 11pm. Suppres-

sion was defined as a cortisol level <5ug/dL. Childhood trauma was

assessed using the Childhood Trauma Questionnaire (CTQ). FEP patients

had significantly higher CTQ scores for emotional abuse (P < .001), sexual

abuse (P = .043) and physical neglect (P = .01) compared to controls. Base-

line cortisol levels tended to be lower in FEP patients compared to controls

(mean (SEM): FEP = 16.4 (1.4) ug/dL; controls = 20.6 (1.4) ug/dL; P = .054).

Six out of 17 FEP patients (35.3%) suppressed cortisol following ingestion

of dexamethasone compared to 2 out of 19 (10.5%) control participants

=3.18; P = .07). In the entire sample, there was a non-significant negative

correlation between baseline cortisol levels and total CTQ scores (0.2; P =

.2). In the low dose DST, percent cortisol suppression was significantly cor-

related with emotional abuse in FEP patients (r = .58, P = .038), but not in

control participants (r = .05; P = .8). No significant correlations were ob-

served for the other CTQ dimension scores. These preliminary findings

show for the first time that a subset of FEP patients have enhanced feed-

back inhibition of the HPA axis, which might be linked to prior exposure to

childhood trauma. Further studies with larger sample sizes are needed. Ex-

amination of the relationship between childhood trauma, HPA function

and the development of PTSD symptoms following the first psychotic

episode is currently underway.

ID: 550972

SEPARABLE DEVELOPMENTAL TRAJECTORIES IN

SCHIZOPHRENIA FROM WOMB TO GRAVE.

RESULTS FROM THE NORTHERN FINLAND 1966

BIRTH COHORT

Matti K. Isohanni

, P. Tanskanen

, A. Alara

isa

nen

E. Ja

skela

inen

, I. Isohanni

, J. Miettunen

Psychiatry, University of Oulu, Oulu, Finland;

Radiology,

University of Oulu, Oulu, Finland;

Oulu Polytechnic, Oulu, Finland

Subtle developmental (motor, emotional, structural, cognitive and behav-

ioural) abnormalities are often present in individuals who later develop psy-

chosis suggesting that some aspects of causation are established before

overt psychosis. However, their effect size and predictive power are small,

and the longitudinal trajectory of developmental factors can be difficult to

tease apart. The Northern Finland 1966 Birth Cohort Study examines the

pre- and postmorbid life-span developmental trajectory for schizophrenia

in a population-based cohort. We studied developmental pathways across

diagnostic groups using developmental markers at birth, at ages 1, 16, and

at ages 31, 34 and (in progress) 43 (genetics, brain morphology, cognitive

capacity, clinical status). The main results were: the schizophrenia group

had excess of achieved developmental milestones later and showed altered

patterns of development over time when compared with non-psychotic

controls. The associations between early development and post-onset

cognition/brain morphology differed in various diagnostic groups. Further-

more, we have identified evidence of progressive dysfunction in a distributed

network involving a fronto-striatal-cerebellar circuit (‘‘developmental dys-

metria’’) in schizophrenia. This is akin to the system underpinning the cog-

nitive dysmetria of schizophrenia and provides a mechanistic link between

the developmental dysmetria prior to schizophrenia. Schizophrenic persons

experience different trails also after the illness onset: separable outcomes in

terms of psychiatric symptomatology, somatic health and mortality. We

conclude that the developmental trajectories in schizophrenia from

womb to grave are distinctly different compared to controls and also within

the schizophrenia group. These findings emphasize both neurodevelopmen-

tal and neurodegenerative aspects of schizophrenia and the scientific value

of pathway variables collected in longitudinal birth cohort studies.

ID: 550963

THE RELATIONSHIP AMONG CHILDHOOD

TRAUMA, STRESS AND COPING IN THE FIRST

EPISODE PSYCHOTIC POPULATION:

PRELIMINARY RESULTS FROM SHARP STUDY

Yang Yun Master

1,2

, R. Parslow

, B. Garner

1,2

, L. Phillips

C. Phassouliotis

, C. Markulev

, C. Leong

, S. Bendall

1,2

G. Berger

2,3

, P. McGorry

1,2

ORYGEN Research Centre, Melbourne, VIC, Australia;

Department of Psychiatry, University of Melbourne, Melbourne,

VIC, Australia;

Department of Psychiatry, University Hospital,

Basel, Switzerland

SHARP (Stress, HPA function And Related Psychosis) is a project to inves-

tigate the relationship among childhood trauma, stress, coping, psychopa-

thology, HPA function, neuro-cognition, brain structure and medical

compliance in the first episode psychotic (FEP) population. This abstract

is the results about childhood trauma, stress, coping, medical compliance

and psychopathology. 50 first episode psychosis drug naı

ve or early treated

(less than 7 days of antipsychotic or antidepressant) patients and 40 matched

healthy controls were recruited from the ORYGEN Youth Health in Mel-

bourne, Australia. Their psychopathology (BPRS, SANS, HAMA and

HAMD), social function (GAF and SOFAS), medical compliance (Clinical

rating scale, Drug attitude Inventory), NEO personality inventory, trauma

(Childhood Trauma Questionnaire and List of Threatening Experiences)

and Stress and coping (Coping Inventory for Stressful Situations, Daily Has-

sles Scale, Perceived Stress Scale and Connor-Davidson Resilience Scale)

were measured at baseline and after 12 weeks follow up. The prevalence

of childhood trauma in FEP group is higher than in the control group. In

the FEP group, the level of daily hassle is positively correlated to perceived

distress at baseline. The bigger change of hassle total score and severity of

hassle is linked to the more severe negative symptoms at 12 week follow up.

Perceived stress at baseline is correlated to less severe negative symptom,

higher hassle and hassle severity and poor clinical rating score. Also higher

CTQ is linked to higher baseline hassle total score, severity of hassle and

higher perceived stress. It also related to lower mood symptom in the

FEP group. FEP with childhood traumatic history might experiences

more hassle, more severe hassles and perceived more stress when they

were suffering from the psychosis, also might report less mood symptom.

ID: 550957

DISORGANIZATION IN SCHIZOPHRENIA:

POSITIVE SYMPTOM OR NEUROCOGNITIVE

DEFICIT?

Joseph Ventura

, A. D. Thames

, G. S. Hellemann

Psychiatry, UCLA, Los Angeles, CA, USA;

Psychology, Alliant

International University, Los Angeles, CA, USA

Background: Several factor analytic studies have shown that in schizophre-

nia patients, disorganization is a separate dimension from positive (non-

disorganizing) symptoms termed, reality distortion. A prior meta-analysis

found that the strength of the relationship between positive symptoms and

neurocognition can vary depending on whether researchers were examining

‘‘reality distortion’’ or ‘‘disorganization’’ (Nieuwenstein et al., 2001). Sev-

eral studies on this topic have now been added to the literature. Methods:

Meta-analysis of 58 studies published in the English language (combined

n = 5,029) was conducted to determine the magnitude of the relationship

International Congress on Schizophrenia Research

2. 2. Phenomenology 23

between neurocognition and reality distortion, eg, delusions and hallucina-

tions as compared to disorganization, eg, formal thought disorder. Addi-

tional analyses were conducted to determine whether the strength of these

relationships differed depending on the neurocognitive domain under in-

vestigation. Results: The effect size of relationship between neurocognition

and reality distortion was small (r = .16) as compared to disorganization

which was moderately associated with neurocognition (r = .26). In four of

five MATRICS domains that were examined, neurocognition was more

closely related to disorganization (r’s range from .20 to .28) than

was reality distortion (r’s = .08 to .15). For disorganization, moderate

effects were found for reasoning and problem solving (r = .26) and verbal

memory (r = .28). Conclusions: The effect size of the relationship between

neurocognition and reality distortion was small compared to disorganiza-

tion which was moderately related in several MATRICS domains. These

findings support a dimensional view of positive symptoms indicating

that the strongest relationship to neurocognitive deficits is shown by the

category of disorganization. The cognitive domains of reasoning and prob-

lem solving, and verbal memory appear to be more conceptually linked to

disorganization than to reality distortion.

ID: 550927

DECREASED BDNF IN PATIENTS WITH FIRST

EPISODE TREATMENT NAI

VE PSYCHOSIS

Ripu D. Jindal

, A. K. Pillai

, S. R. Mahadik

, M. S. Keshavan

1,3

Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA;

Psychiatry, medical College of Georgia, Augusta, GA, USA;

Psychiatry, Harvard Medical School (BIDMC), Boston,

MA, USA

Objective: Brain-derived neurotrophic factor (BDNF) is a key factor that

may mediate neuronal proliferation, differentiation, survival and response

to stress. Several small studies show a decrease in BDNF levels in patients

with psychosis. However, studies of effect of antipsychotic agents on

BDNF have yielded conflicting results. Herein we report on BDNF levels

in first-episode treatment naı

ve psychosis. Method: Fasting serum BDNF

levels were measured in .43.patients with treatment naı

ve first episode psy-

chosis and 42 age matched healthy controls. Baseline levels in the 9 patients

were compared with their levels after 4–8 weeks of antipsychotic treatment

with second generation antipsychotic agents. Data were square root trans-

formed to ensure normality of distribution. Results: Serum BDNF levels

(square root) were reduced in the patients (9.95 þ 1.5) versus controls

(10.70 þ 1.6) (Mann Whitney U = 670.5; P = .04). The levels did not sig-

nificantly differ after treatment (before treatment 9 þ 1.43) versus after

treatment (8.9 þ 1.17); Mann Whitney U = 39.5; P = .93) Conclusions:

Low BDNF levels at the onset of psychosis suggest a possible role in

the pathogenesis of schizophrenia. Our failure to detect treatment may ei-

ther reflect inadequate statistical power, or a true lack of effect of the atyp-

ical antipsychotic agents on BDNF. The inadequacy of contemporary

treatments in addressing the core pathology of schizophrenia highlights

the need to examine alternative treatment targets. This publication was sup-

ported by funds received from AAGP SRI Alumni Award (P.I Ripu Jindal)

and the NIH/NCRR/GCRC grant #M01 RR00056.The clinical core staff

of the Center for the Neuroscience of Mental Disorders (MH45156) assisted

in diagnostic assessments.

ID: 551937

CHILDHOOD BEHAVIOR AND SCHIZOPHRENIA

SPECTRUM PSYCHOPATHOLOGY IN YOUNG GE-

NETIC HIGH-RISK RELATIVES

Srihari S. Bangalore

, D. M. Montrose

, D. Mermon

J. Miewald

, S. Eack

, K. Prasad

, V. Diwadkar

1,2

, M. Abela

R. Rajarethinam

, M. Keshavan

1,2

Psychiatry, University of Pittsburgh Medical Center, Pittsburgh,

PA, USA;

Psychiatry, Wayne State University, Detroit, MI, USA

Introduction: High frequency of psychopathological disorders are seen

among offsprings of schizophrenia and bipolar disorder. The Child Behav-

ior Checklist (CBCL) is a reliable measure of the assessment of changes in

behavior over a period of time as reported by parents and subjects. Schiz-

otypy scores in high risk patients have been shown to be related to premor-

bid psychopathological traits and is a specific predictor of later

development of schizophrenia spectrum disorder. We wanted to examine

the relation between schizotypy and CBCL scores as reported by parents

and subjects. Methods: Subjects for this study were recruited from the Pitts-

burgh Risk Evaluation Program (PREP) at the Western Psychiatric Insti-

tute and Clinic of the University of Pittsburgh Medical Center , as well as

Wayne State University, Detroit. The Chapman subscales evaluating mag-

ical ideation (MA), perceptual aberration (PA) and social anhedonia (SA)

were used to evaluate schizotypy scores using the Chapman psychosis

proneness scale. Internalizing and externalizing symptoms were evaluated

using the parent and child version of the CBCL. Seventy-four young rel-

atives, with a first-degree family history of schizophrenia or bipolar disor-

der, aged 10 to 24 years were evaluated. Results: Cross-sectional analyses

revealed that child internalizing subscale correlated with Chapman sub-

scales of magical ideation (Spearman r = 0.31, P = .007), perceptual aber-

ration (Spearman r = 0.29, P = .038), and social anhedonia (Spearman r =

0.438, P = .004). Strong correlations were also seen the child externalizing

subscales and Chapman subscales of magical ideation (Spearman r = 0.41,

P = .0004), perceptual aberration (Spearman r = 0.23, P = .052), and social

anhedonia (Spearman r = 0.37, P = .005). There was no correlation between

the Parent version of CBCL and Chapman subscales. Conclusions: Child

version of CBCL correlates better than parent version with schizotypy.

Given that schizotypy scores may reliably predict progression towards fu-

ture schizophrenia spectrum disorders, it is prudent to recognize and treat

internalizing and externalizing symptoms early into the prodrome. High-

risk offsprings may be better able to state and recognize their problems ear-

lier on before their parents could recognize it. The CBCL may be used as an

effective tool to screen these high risk children to aid early detection and

predict progression of the illness. This work was supported by MH064023

(Keshavan).

ID: 555210

SCHIZOTYPAL INDIVIDUALS DISPLAY DEFICITS

IN SOCIAL COGNITION

Diane C. Gooding, J. S. Peterman

Psychology, University of Wisconsin-Madison, Madison, WI, USA

Deficits in social information processing have been consistently observed in

schizophrenia patients. Social cognition is a multifaceted construct that

includes the ability to perceive others’ mental states (Theory of Mind), rec-

ognize affect, and accurately identify and interpret emotional meanings

from the environmental milieu. The extent to which these social cognition

deficits are present in individuals at risk for schizophrenia is not clear. The

purpose of the present study was to compare psychometrically-identified

individuals at heightened risk for the later development of schizophrenia

and spectrum disorders with controls in terms of their performance on

two tasks tapping social cognition. Individuals with excessively high (n =

33) and normal (n = 62) scores on the revised Social Anhedonia Scale

were administered the Reading the Mind in the Eyes Task (RMET) and

the Facial Affect and Social Cue Test (FASC). The Social Anhedonia group

showed significantly poorer performance on the RMET relative to the con-

trols, t(93) = 2.34, P < .05. The two groups did not differ in terms of the

FASC. We observed a small but significant association between the

two measures of social cognition. These findings indicate that individuals

at heightened risk for the later development of schizophrenia and schizo-

phrenia-spectrum disorders display deficits in social cognition. It is possi-

ble, however, that the deficits in social information processing are more

International Congress on Schizophrenia Research

24 2. 2. Phenomenology

circumscribed in at-risk samples. Our results will be considered in the larger

context of social cognition and social functioning deficits in the premorbid

and prodromal stages of schizophrenia.

ID: 553487

REEXAMINING THEORY OF MIND IMPAIRMENT

AND NEGATIVE SYMPTOMS IN SCHIZOPHRENIA:

A ROLE FOR ATTENTIONAL DYSFUNCTION?

Aaron Leonard Mishara, T. C. Greig, S. S. Nicholls, M. D. Bell

Psychiatry, Yale University School of Medicine, New Haven,

CT, USA

Abstract: Patients with prominent negative symptoms consistently perform

worse than other clinical groups on Theory of Mind (ToM) tasks. How-

ever, the nature of this relationship has not been closely examined. In this

study, we investigated the relationship between ToM and the Scale for the

Assessment of Negative Symptoms (SANS) components. We replicated

our findings in two independent samples and at two timepoints. Methods:

227 patients with DSM-III-R diagnosis of schizophrenia or schizoaffective

disorder in two independent samples were administered the Hinting Task

and the SANS. One sample (n = 137) was reassessed at 6 months. Pear-

son’s correlations were used to examine the relationships between the

Hinting Task and the SANS subscales and total scores. Subscale scores

that were significantly correlated with the Hinting task were entered

into a stepwise multiple regression to determine how much of the variance

they explained. Results: Results indicate a modest but significant relation-

ship between negative symptoms and the Hinting Task. Examination of

the components revealed that Attention is largely responsible. Alogia

made a minor contribution. Conclusions: ToM deficits are associated

with negative symptoms. A common mechanism may be the allocation

of attention to social events and socially relevant stimuli. Disrupted pro-

cessing at the level of implicit, ‘‘on line’’ representations may contribute to

impairments in attentional function as assessed by the SANS as well as

contributing to impaired ToM performance on the Hinting Task. The cur-

rent findings indicate that we should reexamine ToM impairment in

patients with negative symptoms and how it might be targeted in thera-

peutic interventions.

ID: 551995

TOWARD A COMPREHENSIVE MODEL OF

NEUROLOGICAL SOFT-SIGNS AND PSYCHO-

PATHOLOGY IN SCHIZOPHRENIA

Marie-Odile Krebs

1,2

, D. Gourion

, M. C. Bourdel

1,2

, I. Amado

1,2

J. O. Olie

1,2

Lab Physiopathologie of Psychiatric Diseases, Inserm, Paris,

France;

SHU, Ho

pital Sainte Anne, Paris, France

The goal of this study was to explore interrelations between core dimen-

sions of schizophrenia, neurological soft-signs and age at onset, by using

structural modeling. Two hundred and thirty four patients with a DSM-IV

diagnosis of schizophrenia were included in the study. All subjects were

administered a semi-structured interview (DIGS) for diagnosis and age

at psychotic onset, the Positive and Negative Syndrome Scale (PANSS),

and neurological signs were assessed with a standardized scaleSeveral

structural equations analyses were performed. The best structural model

in terms of fit indices incorporated two latent components of the disease

and nine observed variables (TLI: 0.86; RMSEA:0.064). The first latent

component, Neurological Loading (NL), was mostly related to abnormal

motor coordination and motor integration scores (structural coefficients:

respectively .81 and .72). The second latent component, Schizophrenic Psy-

chopathology (SP), was strongly linked to the dimension of disorganisation

(.99). NL and SP were significantly inter-correlated (.49; P < .001).Age at

onset was significantly correlated to SP but not to NL. The close association

of neurological abnormalities with core schizophrenic dimensions, partic-

ularly disorganisation, suggest common physiopathological pathways.

ID: 551957

International Congress on Schizophrenia Research

2. 2. Phenomenology 25

3. 3. Drug Side Effects and Physical Illness

ANTIPSYCHOTIC STAY VS. SWITCH FOR

METABOLIC BENEFIT: PSYCHIATRIC COST

Jonathan M. Meyer,

, R. R. Henry

, D. L. Braff

Psychiatry, UCSD, San Diego, CA, USA;

Medicine, UCSD and

VA San Diego, San Diego, CA, USA

Background: Switching antipsychotics is one option for patients with an-

tipsychotic-induced metabolic effects,

yet there is concern that switching

stable patients carries a psychiatric cost, and may result in exacerbation.

Methods: Data were analyzed from a randomized, 6-month, open-label

switch study for overweight or obese nondiabetic schizophrenia patients

taking risperidone or olanzapine. Eligible subjects were randomized to

usual care non-switch (UC) or open-label switch (S) to ziprasidone.

Arms were balanced for gender and race/ethnicity. Imaging measures

of adiposity and endocrine measures were obtained at baseline and

6 months. The primary psychiatric outcome measure was total hospital

days in each arm, while secondary analyses examined change in total

PANSS and PANSS subscale scores. Results: To date, 51 subjects

have entered the randomization phase of the study: UC n = 22; S n =

29. Mean age of randomized subjects was 47.3, 78.4% were male,

66.7% white and 9.8% Hispanic. Mean BMI was 33.4 kg/m

, 63% smoked,

and 15.7% reported a 1st-degree family history of diabetes. Baseline total

PANSS score was 64, positive subscale 18, and negative subscale 15. At

baseline there were no significant between-group differences on any de-

mographic, adiposity or laboratory variable, nor on total PANSS, posi-

tive or negative subscale symptom scores. There were no significant

differences in total hospital days between the UC arm (26 days), and

the S arm (25 days), or in proportion of subjects hospitalized: UC

(14%) vs. S (21%) (v

= .338, P = .561). In a linear regression model

with change in total PANSS score from baseline to week 26 as the de-

pendent variable, neither treatment arm, nor baseline antipsychotic signif-

icantly predicted the outcome variable. Conclusions: Contrary to many

assumptions, in this 6-month, open-label, randomized trial, the hospital-

ization rate and total hospital days in non-switchers did not significantly

exceed that seen with careful antipsychotic switching. This inherent re-

lapse rate among stable patients with schizophrenia must be factored

into the stay vs. switch decision to change antipsychotic treatment for

possible metabolic benefit.

References

1. Meyer JM, et al. Change in metabolic syndrome parameters with anti-

psychotic treatment in the CATIE Schizophrenia Trial: prospective

data from phase 1. Schizophr Res 2008;101:273–86.

2. Essock S, et al. Effectiveness of switching antipsychotic medications.

Am J Psychiatry 2006;163:2090–5.

ID: 538155

REDUCED EPS RISK WITH LURASIDONE, A NOVEL

PSYCHOTROPIC AGENT UNDER DEVELOPMENT

FOR SCHIZOPHRENIA AND BIPOLAR DISORDER

Hiroyuki Nishikawa

, T. Ishibashi

, K. Tokuda

, T. Matsumoto

M. Ogasa

Dainippon Sumitomo Pharma Co., Ltd., Osaka, Japan;

Dainippon

Sumitomo Pharma America, Fort Lee, NJ, USA

Background: Lurasidone is a novel psychotropic agent with affinity for

dopamine D

and serotonin 5-HT

receptors. Compared with other atyp-

ical antipsychotics, lurasidone has high affinity at serotonin 5-HT

,5-

, and noradrenaline a

receptors; minimal affinity for a

adrenocep-

tors, dopamine D

and D

receptors, serotonin 5-HT

receptors, and a

adrenoceptors; and no affinity for histamine H

and cholinergic M

recep-

tors. The goal of the current series of studies was to evaluate the EPS-pro-

ducing potential of lurasidone compared to other antipsychotics in

standard animal models. Methods: The EPS potential of lurasidone

was evaluated in animal models. The mouse pole-test was used to measure

bradykinesia-inducing activity. The rat and mouse catalepsy was measured

by duration of immobility induced by lurasidone in a catalepsy apparatus.

The rat paw test was used to measure forepaw retraction time. Results: As

summarized in Table 1, below, lurasidone was much weaker than other

antipsychotic drugs in inducing catalepsy in both rats and mice. In the

mouse pole test, lurasidone was also far weaker than other antipsychotics

in inducing bradykinesia, and failed to cause a significant motor impair-

ment (delay in turning and pole-descending behaviors) even at 1000 mg/kg

(Table 1). Finally, in the rat paw test, even high doses of lurasidone did not

increase the forepaw retraction time. In contrast, other antipsychotic

drugs were associated with increased retraction times even at relatively

low doses (Table 1). Conclusion: The results of these pre-clinical studies

suggest that lurasidone has a low potential for causing clinically significant

EPS.

Table 1. Comparative ED50 (mg/kg) for induction of extrapyramidal side

effects across 4 animal models

Drug

Rat

Catalepsy

Mouse

Catalepsy

Mouse Pole

Test

Bradykinesia

Rat Paw

Test

Lurasidone >1000 >1000 >1000 >1000

Risperidone 20 0.85 3 30

Olanzapine 28 >10 10 30

Sertindole >300 >30 10 >1000

Quetiapine NT >300 NT NT

Ziprasidone 97 63 30 100

Clozapine >300 >30 >30 300

Haloperidol 12 2.0 1 1

Chlorpromazine 25 7.3 10 30

Thioridazine 890 42 30 >1000

ID: 550540

PREDICTORS OF CARDIOVASCULAR MORTALITY

IN HOSPITALIZED PATIENTS WITH SEVERE

MENTAL ILLNESS

Kimberly R. Warren

, F. Liu

, R. P. McMahon

H. J. Wehring

, R. C. Love

, K. M. Mackowick

, J. C. Shim

D. L. Kelly

Maryland Psychiatric Research Center, University of Maryland

School of Medicine, Catonsville, MD, USA;

Department of

Pharmacy Practice and Science, University of Maryland School of

Pharmacy, Baltimore, MD, USA;

Paik Hospital, Inje University,

Gimhae, South Korea

People with mental illness have an increased risk of comorbid medical

conditions (Sokal, et al., 2004) and premature death (Harris and Barra-

clough, 1998). Cardiovascular (CV) disease is the second leading cause of

death in patients with schizophrenia (Osbey et al., 2000). The present

study investigated predictors of CV mortality in hospitalized people

with severe mental illness: psychosis (predominantly schizophrenia

(SZ)), bipolar disorder (BD), or other mental illness (predominantly

depression and anxiety disorders ([OTH]). Clinical chart data was

International Congress on Schizophrenia Research

26 3. 3. Drug Side Effects and Physical Illness

abstracted and available on 1828 patients (n

= 1443, n

= 165, n

OTH

220) from inpatient State psychiatric hospitals between 1994 and 2000 in

order to assess risk factors for cardiovascular (CV) mortality. Deaths were

identified from the Social Security Death Index and death records were

collected for all decedents. Males (n = 1065) and females (n = 762) between

the ages of 35–65 were included. Groups were similar in age, smoking

status, and sex, however, the SZ group had more non-whites (41%)

than BD (29%) or OTH (22%). There were 60 (n

male

= 30, n

female

30) deaths in this sample. A Cox proportional hazards model, including

age>55, race (white vs. non-white), sex, smoking status, and diagnosis,

was used for multivariate survival analysis. Increased CV mortality

was associated with age>55 (hazard ratio (HR) = 4.1, v

= 15.76,

P < 0.001), white race (HR = 2.4, v

= 4.07, P < 0.05), and smoking

(HR = 2.0, v

= 3.63, P = .057). Compared with patients with SZ, increase

in CV mortality was not significant in patients with bipolar disorder

(HR = 1.7, v

= 0.71, P = .399), and was marginally elevated in the

OTH group (HR = 2.7 v

= 3.30, P = .069). Sex was not significantly

associated with CV mortality, P = .25. These findings are consistent

with numerous studies that suggest that older age and smoking are

risk factors for heart disease (Hennekens, 2007). Previous studies have

also indicated greater CV mortality risks for patients with anxiety and

depressive disorders than patients with schizophrenia (Harris and

Barraclough, 1998). Racial differences in CV mortality in patients with

severe mental illness require further investigation. This project was funded

by the National Institutes of Mental Health (NIMH R03 MH069871-01;

Kelly, PI) and the Advanced Centers for Intervention and Services

Research (NIMH P50 MH40279; Carpenter, PI).

ID: 550530

THE RELATIONSHIP OF BRAIN WEIGHT TO BODY

MASS INDEX (BMI) UPON AUTOPSY IN PEOPLE

WITH SEVERE MENTAL ILLNESS

Heidi Wehring

, F. Liu

, R. P. McMahon

, D. L. Boggs

R. C. Love

, D. Dickinson

3,4

, J. C. Shim

, D. Fowler

, D. L. Kelly

Maryland Psychiatric Research Center, University of Maryland

School of Medicine, Baltimore, MD, USA;

Pharmacy Practice and

Science, University of Maryland School of Pharmacy, Baltimore,

MD, USA;

University of Maryland School of Medicine, Baltimore,

MD, USA;

Research Core, VISN 5 Mental Illness Research,

Education, and Clinical Center, Baltimore, MD, USA;

Paik

Hospital, Inje University, Gimhae, South Korea;

Office of the Chief

Medical Examiner (OCME), Baltimore, MD, USA

Weight gain and other metabolic adverse effects of antipsychotic medica-

tions have been of concern in maintaining the physical health of patients

with schizophrenia. However, the potential effects of possible brain

changes during weight gain have received less attention. In the non-

schizophrenia population, it has been observed that an association exists

between high body mass index (BMI) and reduced gray matter, and BMI

has also been found to be inversely related to brain volume measured by

MRI. This association has not been reported thus far in patients with

schizophrenia. Also, people with first episode schizophrenia have been

shown to undergo changes in gray matter and ventricular volume over

time and some evidence suggests that brain volume changes may occur

with some antipsychotic medications. This study examined the relationship

between BMI and brain weight upon autopsy in 60 patients with severe

mental illness treated with antipsychotics. The population was 67% Cau-

casian, 70% male, with a mean age of 38.2

6 9.6 years. The mean BMI and

brain weights were 29.64 6 8.71 kg/m

and 1322.58 6 162.23 gm, respec-

tively. Mean brain weight was significantly higher in men as compared to

women, despite no gender differences in BMI (Kruskal-Wallis test,

=12.28, df = 1, P = .0005). No racial differences were evident. Normal

weight subjects (BMI = 18.5–24.9 kg/m

; n = 19) had a significantly greater

mean brain weight (1357.4

6 141.7 gm) as compared to those who were

considered obese (BMI  30 kg/m

) (1299.0 6 194.0 gm; n = 29)

(Kruskal-Wallis test, v

= 33.8, df = 1, P < .0001). Further characterization

found that in those with morbid obesity (Obesity class III BMI  40 kg/m

;

n = 5), the mean brain weight was the smallest among all obese categories

(1186.0

6 286.2 gm). There were no differences in brain weight by antipsy-

chotic treatment and no correlations between brain weight and body

weight were found by race, gender or drug treatment. The results of

this study suggest that further research is needed regarding the effects

of obesity on brain changes in people with severe mental illness. This pro-

ject was funded by the National Institutes of Mental Health (NIMH R03

MH069871-01; Kelly, PI) and the Advanced Centers for Intervention and

Services Research (NIMH P50 MH40279; Carpenter, PI).

ID: 550511

METHODS USED TO ASSESS ADVERSE EFFECTS IN

A SAMPLE OF REPORTS OF CLINICAL STUDIES OF

ANTIPSYCHOTIC MEDICATION

Alison Pope

, C. Adams

, C. Paton

, T. Weaver

, T. R. Barnes

Psychological Medicine, Imperial College London, London, United

Kingdom;

Cochrane Schizophrenia Group, Nottingham, United

Kingdom

Evidence-based decisions regarding the selection of antipsychotics (APs)

for the long-term treatment of schizophrenia are influenced by existing

knowledge regarding the relative safety and tolerability profiles of differ-

ent agents. However, collecting information about AP safety is method-

ologically challenging because there is no single assessment approach

capable of capturing the full range of adverse effects. Different methods

for collecting information about adverse effects have been shown to result

in different prevalence figures, which may represent a source of bias in

reports of AP safety and tolerability. To identify how investigators typ-

ically collect and report information about AP safety, we surveyed eligible

journal articles. To generate a representative sample, we searched the

Cochrane Schizophrenia Group’s register. Reports on 103 eligible clinical

studies of antipsychotic drug treatment (90 of which were double-blind)

were identified. The findings revealed that published rating scales were

frequently used to evaluate drug-induced extrapyramidal symptoms

(EPS), although relatively few studies employed several measures to assess

the full range of EPS. In contrast, published scales were rarely used to

assess general treatment emergent adverse effects and a large number

of reports failed to specify how such information had been collected. De-

spite broad uniformity in the choice of measures used to evaluate EPS,

findings from different studies remain difficult to interpret due to varia-

tions in reporting. Scores derived from rating scales tended to be presented

as statistics such as mean change, which do not provide clinicians with

meaningful information about the clinical or functional impact of the

movement abnormalities observed. After EPS, weight gain was the second

most frequently reported adverse effect. Despite increasing concerns about

other metabolic risks associated with APs, relatively few studies explicitly

referred to collecting data regarding markers of the metabolic syndrome.

Only a small minority of studies reported assessing adverse effects that

may be of particular concern to patients, eg, sexual dysfunction and aver-

sive subjective experiences. These findings highlight the need for more con-

sistent approaches to collecting and reporting information about the

adverse effects of APs. Such information is vital if meaningful compari-

sons are to be made between the safety and tolerability profiles of different

APs.

ID: 550476

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 27

GENE PROFILING REVEALS MAJOR

DIFFERENCES IN THE EFFECT OF ANTI-

PSYCHOTICS ON LIPID METABOLISM

Aouatef Bellamine

, C. Elosua

, C. Cao

, R. Baker

, A. Pikalov

O. Flint

Research and Development, Bristol-Myers Squibb, Princeton, NJ,

USA;

Mediacl Affairs, Otsuka America Pharmaceuticals,

Rockville, MD, USA

Background: Some anti-psychotics (APs) are associated with metabolic side

effects. We investigated the effect of aripiprazole and other APs on

gene expression in human hepatoma cells (HepG2) and primary human

adipocytes. Methods: Cultured HepG2 and differentiated human adipo-

cytes were exposed to Cmax and 10-fold Cmax of aripiprazole (ARI),

olanzapine (OLZ), quetiapine (QUE), risperidone (RIS), 9-OH risperi-

done (9-OH-RIS) or ziprasidone (ZIP) for 6 or 24 hrs. Total RNA

was isolated and used to evaluate expression levels for the treated vs ve-

hicle control. GeneChip hybridization Array technology (Affymetrix) (P

 0.01) was used. Statistical comparisons were made by ANOVA with

Dunnet’s correction. Results: In adipocytes, OLZ and QUE induced

genes involved in cholesterol transport and lipid synthesis. These genes

are unchanged by other APs and down-regulated by ARI. In HepG2,

PCA analysis grouped OLZ/QUE together but only QUE treatment

resulted in the up-regulation of squalene epoxidase, an enzyme convert-

ing cholesterol to oxysterols. ARI treatment led to the down-regulation

of this gene. Furthermore, in lipid synthesis studies, QUE treatment

shifted the cholesterol band to a higher molecular weight product cor-

responding to an oxysterol. Conclusions: AP treatment of both

HepG2 and adipocytes resulted in changes in expression levels of genes

involved in lipid metabolism and cholesterol reverse transport. In addi-

tion, in HepG2 QUE led to the shunt of cholesterol pathway to oxyster-

ols, known LXR ligands. The possible activation of the transcription

factor LXR by oxysterols could be the reason for the observed gene

up-regulation, contributing potentially to an increase in plasma lipids lev-

els and cardiovascular risks.

ID: 550415

PHARMACOGENETICS OF FOLIC ACID METABO-

LISM AND METABOLIC DISTURBANCES IN

SCHIZOPHRENIA

Vicki Ellingrod

, T. Grove

, S. F. Taylor

, J. Moline

D. D. Miller

, T. Holman

, J. Kerr

University of Michigan, Ann Arbor, MI, USA;

University of Iowa,

Iowa City, IA, USA;

Cleveland Clinics, Cleveland, OH, USA

Purpose: To examine the relationship between the methylenetetrahydrofo-

late reductase (MTHFR) 677C/T variant and metabolic syndrome in

schizophrenia subjects receiving antipsychotics for  12 months. This var-

iant is associated with aberrant folic acid metabolism and cardiovascular

disease risk due to alterations in the Aldomet cycle associated with homo-

cysteine and folic acid metabolism. Methods: 121 subjects were recruited

from the Universities of Iowa and Michigan for this cross-sectional anal-

ysis. They were screened for the metabolic syndrome (NCEP ATP-III cri-

teria) and genotyped for MTHFR 677C/T. Additionally serum folate,

vitamin B12, leptin, and homocysteine were measured in some subjects.

Results: 49 subjects (40%) met metabolic syndrome criteria. The groups

mean age (

6 s.d.) was 41.05 6 11.17 years, 78% (n = 94) were Caucasian,

61% (n = 74) were male, and 71% (n = 84) were receiving clozapine, olan-

zapine, risperidone or quetiapine. There were no differences in age, gender,

race, or antipsychotic exposure, between the genotype groups. The geno-

types were CC (53%), CT (395), TT (8%) and are in Hardy Weinberg (v

0.11, P = .74) Overall, for those receiving an atypical antipsychotic (AAP)

a significant relationship was found between metabolic syndrome and the

MTHFR TT genotype (v

= 6.81, P = .009). Specifically, subjects with this

genotype receiving an AAP had a relative risk of 2.42 (95% CI: 1.87–3.14)

for meeting metabolic syndrome criteria compared to those without this

genotype receiving AAPs. For subjects not receiving AAPs no relationship

between MTHFR genotype and metabolic syndrome was found (v

= 0.47,

P = .49). No statistically significant relationship between MTHFR geno-

type and homocysteine was found, however subjects with the TT genotype

did have higher levels (9.6 vs 10.3, P = .7). Conclusion: Overall, these

results confirm our group’s previous data. Aberrant function in MTHFR

related to poor folic acid metabolism and hyperhomocysteinemia may be

associated with antipsychotic associated metabolic complications. How-

ever, these results should be taken cautiously due to the small sample

size included in this study. Acknowledgements: This project was sup-

ported by the NIMH (K08MH64158), the NIH-NCRR, General Clinical

Research Centers Program (M01-RR-59 and UL1RR024986), a University

of Michigan College of Pharmacy Vahlteich Award, and Washtenaw

Community Health Organization (WCHO).

ID: 550377

SUBJECTIVE WELL-BEING AND D

RECEPTOR

OCCUPANCY IN PSYCHOTIC PATIENTS TREATED

WITH ANTIPSYCHOTICS: AN EXPERIENCE

SAMPLING STUDY

Johan Lataster

, J. van Os

1,2

, L. de Haan

, I. Myin-Germeys

1,4

Department of Psychiatry and Neuropsychology, South Limburg

Mental Health Research and Teaching Network, EURON, Maas-

tricht University Medical Center, Maastricht, Netherlands;

Divi-

sion of Psychological Medicine, Institute of Psychiatry, London,

United Kingdom;

Department of Psychiatry, Academic Medical

Center/de Meren, Amsterdam, Netherlands;

School of Psycho-

logical Sciences, University of Manchester, Manchester, United

Kingdom

Background: Blockade of dopamine D

receptors is a key element in the

therapeutic activity of antipsychotic medication, but may also reduce mo-

tivation and subjective well-being. Previous laboratory studies have related

receptor occupancy to subjective well-being measured with question-

naires. The current study aims to extend these findings to a broader domain

of well-being assessed in the realm of daily life. In addition, antipsychotic

agents differ in how tightly they bind to the D

receptor, possibly resulting

in different effects on subjective well-being. The differential effect of ‘tight’

and ‘loose’ binding drugs will be investigated. Methods: The Experience

Sampling Method (a structured diary technique) was used to assess

subjective experience in daily life in 111 patients with a diagnosis of

psychotic disorder. Patients were on current antipsychotic medication

and were divided into a loose (olanzapine; n = 41) and tight (haloperidol,-

risperidone; n = 70) binding agent user group based on the agents’ disso-

ciation constants at the D

receptor. D

occupancy values were estimated

using occupancy formulas extracted from previous literature. Based on the

tertiles, a three level D

occupancy variable was defined. Results: Multilevel

analyses showed a significant interaction between binding group and D

receptor occupancy (positive affect (PA):b = .05 (SE = .02), P < .009; neg-

ative affect (NA):b = .04(SE = .02), P < .03). For tight binding agent users,

a significant effect was found of D

receptor binding on PA and NA (PA:

(2) = 6.41, P < .042; NA:v

(2) = 29.73, P < .0001). There was a decrease in

PA in the middle group and an even larger decrease in the group with the

most D

receptor occupancy. A significant increase in NA was found in the

group with the most D

receptor occupancy. For loose binding agent users,

no main effect of D

occupancy on PA or NA was found. Although related

to symptom severity, effects could not be entirely explained in terms of clin-

ical symptom scores. Conclusions: These findings add ecological validity to

previous laboratory findings showing an association between D

receptor

International Congress on Schizophrenia Research

28 3. 3. Drug Side Effects and Physical Illness

occupancy and subjective well-being, particularly pertaining to the tight

binding agents. Furthermore, given that tight and loose binding agents

showed overlap in terms of overall receptor occupancy estimates, our

results suggest that the mechanism of dissociation from the D

receptor

rather than mere occupancy levels of the drug may determine a drug’s im-

pact on subjective well-being.

ID: 550288

THE NATIONAL REGISTER OF ANTIPSYCHOTIC

MEDICATION IN PREGNANCY (NRAMP)

Heather Gilbert, N. Marston, C. Gurvich, A. de Castella,

K. McCauley, J. Kulkarni

Monash Alfred Psychiatry Research Centre, The Alfred Hospital

and School of Psychology, Psychiatry and Psychological Medicine,

Monash University, Melbourne, VIC, Australia

Current data on the use of antipsychotic medication in pregnancy is limited.

Establishment of The National Register of Antipsychotic Medication in

Pregnancy (NRAMP) will provide evidence-based clinical guidelines for

the safest use of antipsychotic medication during pregnancy and for one

year postnatally. NRAMP is an Australia-wide observational, non-

interventional study involving female participants with a history of mental

illness, who take antipsychotic medication and who become pregnant or

have had a baby in the last 12 months. Information is collected via tele-

phone or face to face interviews during the pregnancy and for the first

year postpartum and includes demographic, social, family, medical, psychi-

atric, medication and obstetric history, as well as information on general

health and wellbeing for both mother and baby. This observational study is

current and ongoing, with 66 consented participants at different stages on

the study timeline; that is, either antenatal or postnatal. This is reflected in

the results to date, with regard to consenting data, pregnancy outcomes and

maternal and neonatal complications. Preliminary interpretation of these

results will therefore be spread across various time points: the antenatal

period, post-delivery and for one year postnatally. The resultant evi-

dence-based guidelines arising from The National Register of Antipsy-

chotic Medication in Pregnancy (NRAMP) have the potential to

provide regular, contemporary updates to clinical treating teams for the

management of women in this vulnerable population group. We plan to

fill a void in mental health services where currently there is a distinct

lack of information available to treating clinicians, with regard to providing

safe and timely care of women who take antipsychotic medication and be-

come pregnant. This study is supported by AstraZeneca, Janssen-Cilag,

Mayne Pharmaceuticals and the Australian Rotary Health Research Fund.

ID: 550183

THE INCIDENCE OF OBESITY AND DIABETES IN

PATIENTS TREATED WITH CLOZAPINE

Rammohan Rao Malesu

1,2

, N. Limet

Psychiatry, Blacktown Hospital, Blacktown, Sydney, NSW,

Australia;

Psychiatry, University of sydney, Sydney, NSW,

Australia

This is a 15 year retrospective study to determine the correlation between

Clozapine treatment and the development of diabetes and obesity . The

study cohort consisted of all 136 patients (77M;59F) registered at the Clo-

zapine outpatients clinic of the Blacktown Hospital in Sydney. All patients

were suffering from drug resistant Schizophrenia and were on Clozapine.

The data was collected from patient records, which included demographic

details, duration of Clozapine treatment, vital signs including weight, Blood

levels for Glucose, Cholesterol and Triglycerides. Patients were on Cloza-

pine treatment for between 1–15 years(mean 5.6 years) of which 24 patients

were on Clozapine for 10 years or more. As per the protocol, all patients

had routine investigations including blood levels for glucose, cholesterol

and triglycerides. Their weight was also recorded at least once a month.

Of this cohort, 10(5M:5F) had consistently elevated blood glucose levels,

34 (19M:15F) had high cholesterol and 55(39M:16F) had high triglycerides.

All the patients who had high glucose levels also had high cholesterol and

triglycerides and were overweight. These figures were compared to the prev-

alence of obesity and Type II diabetes in the general population of Austral-

ia. These figures support the fact that there is a higher incidence of obesity

and diabetes in patients treated with Clozapine even after adjusting for age

and gender.

ID: 550161

EFFECTS OF CLOZAPINE AND HALOPERIDOL ON

EXPRESSION OF DOWNSTREAM EFFECTORS OF

INSULINE SIGNALING: FOCUS ON Ped/Pea-15

Fabio Panariello

, G. Perruolo

, P. Formisano

, G. Muscettola

A. de Bartolomeis

Neuroscience, University of Neaples ‘‘Federico II’’, Neaples, Italy;

Dipartimento di Biologia e Patologia Cellulare e Molecolare and

Istituto di Endocrinologia ed Oncologia Sperimentale del CNR,

University of Neaples ‘‘Federico II’’, Neaples, Italy

Schizophrenia is a major highly debilitating psychiatric disorder with

a prevalence of about 1% (Mueser and McGurk, 2004). Several lines of

evidence have indicated that schizophrenic patients have an higher prev-

alence of impaired glucose tolerance, insulin-resistance, and type-2 diabe-

tes mellitus than general population (Citrome et al., 2005). Clinical and

experimental evidence indicates that first and second generation antipsy-

chotic drugs may contribute to impair glucose metabolism (Haupt and

Kane 2007). In our study we have shown, by in vitro (PC12 transplantable

rat pheochromocytoma cells and L6 skeletal muscle cells) paradigm, that

both haloperidol and clozapine reduced insulin-stimulated glucose uptake

in PC12 neuronal cells and in L6 skeletal muscle cells. Pre-treatment with

clozapine, but not with haloperidol, also prevented insulin effect on insulin

receptor and IRS-1/2 tyrosine phosphorylation. Furthermore in neurons

and skeletal muscle cells, both haloperidol and clozapine led to the acti-

vation of Akt and to increased expression of the Ped/Pea-15 protein, an

Akt substrate. This is paralleled by decreased PKC-f activity and deranged

insulin-stimulated glucose uptake. Similarly increased Akt activity and

Ped/Pea-15 levels, as well as reduced PKC-1 activity, were detected in cau-

date-putamen and in skeletal muscle of mice (3-months-old male C57/BL/

KsJ mice) treated with either haloperidol or clozapine. Our data suggest

that first and second generation antipsychotics may contribute to the im-

pairment of glucose metabolism by reducing insulin-mediated glucose up-

take in insulin-targeted tissues. The molecular mechanism implicated in

this effect involves intracellular effectors of the insulin signaling cascade.

For instance, the increase of phosphorilated Akt enhances Ped/Pea-15 lev-

els by increasing its phosphorylation on Ser-116; as a consequence, Ped/

Pea-15 inhibits the activation of atypical PKC-f and prevents insulin-

stimulated glucose uptake.

References

1. Citrome L, Blonde L, Damatarca C. Metabolic issues in patients with

severe mental illness. South Med J. 2005;98(7):714–20.

2. Haupt DW, Kane JM. Metabolic risks and effects of atypical antipsy-

chotic treatment. J Clin Psychiatry. 2007;68(10):e24.

3. Mueser KT, McGurk SR. Schizophrenia. Lancet. 2004;19:363(9426):

2063–2072.

ID: 549990

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 29

PHARMACOGENETICS OF THE METABOLIC

CONSEQUENCES OF ANTIPSYCHOTIC

TREATMENT - PREDICTION AND MECHANISM

Gavin P. Reynolds

Department of Psychiatry, Queen’s University Belfast, Belfast,

United Kingdom

There are substantial differences between individuals in the effects of an-

tipsychotic drugs. This is particularly true in the weight gain apparent fol-

lowing treatment with many of the second generation, as well as some of the

first generation, antipsychotics; equivalent treatment regimes can cause

profound fat deposition in some patients but minimal effects in others.

Such weight gain can lead to further morbidity as indicated by the devel-

opment of metabolic syndrome, identifying a set of risk factors for cardio-

vascular disease and diabetes. The pharmacology of antipsychotic-induced

weight gain is not fully defined, but antagonist action at 5-HT2C receptors,

as well as at histamine H1 and some other receptors, is strongly implicated.

The underlying mechanisms too are unclear but may involve hypothalamic

receptor antagonism interfering with the signalling of circulating hormones

such as leptin. These receptors and hormonal hypotheses provide good can-

didate genes for pharmacogenetic study of antipsychotic drug-induced

weight gain. In two pharmacogenetic studies of antipsychotic-induced

weight gain in initially drug-naive patients, we demonstrated a strong as-

sociation with weight gain of a promoter region polymorphism of the 5-

HT2C receptor, which we have also shown to influence gene expression

in neuronal cells in culture. The leptin gene also shows a functional pro-

moter polymorphism and this too is associated with antipsychotic-induced

weight gain, particularly in the longer term. Findings in a population of

chronically-treated patients with schizophrenia have revealed interacting

effects of the two polymorphisms on measures of obesity, and particularly

association of the leptin gene polymorphism with the incidence of metabolic

syndrome. This supports the interpretation that the leptin polymorphism

may have longer-term effects on these metabolic consequences of antipsy-

chotic treatment. Interestingly, not only the leptin polymorphism but also

that of the 5-HT2C receptor gene was associated with blood leptin concen-

trations prior to treatment. These findings point towards underlying mech-

anisms, as well as indicating the potential for genetic testing for the

propensity to develop weight gain and its severe metabolic consequences.

ID: 549828

COMPARISON OF LEPTIN LEVELS AND OBESITY

BETWEEN DRUG-NAI

VE FIRST-EPISODE

SCHIZOPHRENIA PATIENTS AND CHRONICALLY

MEDICATED PATIENTS

Peter F. Buckley

, S. P. Mahadik

1,2

, M. M. Khan

D. R. Evans

1,2

Department of Psychiatry, Medical College of Georgia, Augusta,

GA, USA;

Mental Health Section, Charlie Norwood VA Medical

Center, Augusta, GA, USA

Weight gain and metabolic disturbances now predominate the psychophar-

macology of schizophrenia. In this study, plasma levels of leptin were eval-

uated in relation to weight and psychopathology measures in never-

medicated first episode of psychosis patients (FEP; N = 23), patients

with chronic schizophrenia who were receiving clozapine, (N = 21), and nor-

mal subjects (N = 11). Patients were clinically assessed using the PANSS and

BPRS. Plasma leptin levels were determined by commercially available En-

zyme Linked ImmunoSorbent (ELISA) kits. Plasma leptin levels were not

significantly different between never-medicated FEP and matched normal

controls. However leptin levels were significantly higher in patients treated

with clozapine than in either control subjects or FEP patients (P = .065 and

.016, respectively). While not surprisingly, clozapine-treated patients had

significantly higher BMI than FEP patients or controls (29.86

6 3.6 vs

25.1

6 2.61 and 23.2 6 2.14, respectively), higher leptin levels in these

chronic schizophrenia patients did not correlate with higher BMI but

did correlate with psychopathology measures. In some contrast, leptin lev-

els in FEP patients were inversely related to psychopathology measures.

Consistent with the emergent literature, leptin levels may be a useful mea-

sure for examining energy metabolism and relationships to antipsychotic-

induced weight and metabolic disturbances in schizophrenia.

ID: 549709

PROLACTIN LEVELS AND SEXUAL SIDE EFFECTS

IN SCHIZOPHRENIA PATIENTS DURING

ANTIPSYCHOTIC TREATMENT

Falko Biedermann, M. Rettenbacher, A. Hofer, C. Ebenbichler,

S. Baumgartner, M. Edlinger, M. Hummer, G. Kemmler,

W. W. Fleischhacker

Biological Psychiatry of Innsbruck, Medical University of

Innsbruck, INNSBRUCK, Austria

Introduction: Sexual dysfunction occurs frequently in schizophrenia

patients treated with antipsychotic drugs. Prolactin levels can be elevated

under antipsychotic treatment and have been suggested to impair sexual

functioning through an action on the hypothalamic-pituitary-gonadal

axis altering sex hormone release. As there is limited evidence about the

association between prolactin levels and sexual dysfunction in schizophre-

nia patients receiving antipsychotic treatment we have prospectively inves-

tigated patients during treatment with second generation antipsychotics.

Methods: Thirty-nine schizophrenia patients (diagnosed according to

ICD 10) treated with amisulpride, clozapine, quetiapine, olazapine, risper-

idone or ziprasidone were included in a 4 week study. To quantify antipsy-

chotic side effects we used the Udyalg for Klinske Undersogelser (UKU)

Side Effect Scale. Prolactin levels were determined weekly. Results: Prolac-

tin levels in males rose significantly while no significant changes in prolactin

levels were observed in female patientsfrom baselineto week 4. Men reported

diminished sexual desire most frequently followed by orgastic dysfunction

and ejaculatory dysfunction. Women most often complained about dimin-

ished sexual desire, amenorrhea or orgastic dysfunction. The incidence of

sexual adverse events did not change significantly from baseline to week

4. At baseline a significant association was found between diminished sexual

desire and prolactin levels in men. At week 4 orgastic dysfunction and higher

prolactin levels showed a correlation in male and female patients. In addition

a relationship between changes of prolactin concentration and diminished

sexual desire was detected in female patients. Conclusion: Further studies,

including more Patients, which need to be studied over a larger observation

period, will be necessary to shed more light on sexual functioning in schizo-

phrenia patients and to differentiate between antipsychotics, which was not

the aim of this study.

ID: 549658

PROSPECTIVE LONG-TERM EVALUATION OF

SERUM CREATINE KINASE LEVELS AND

NEUROMUSCULAR DYSFUNCTION IN

SCHIZOPHRENIA PATIENTS: NEW RESULTS OF

FOLLOW-UP STUDY

Ilya Reznik

, L. Volchek

, M. Reznik

, H. Y. Meltzer

A. Weizman

Laboratory of Biological Psychiatry, Felsenstein Medical Research

Center, Bat-Yam, Israel;

Neurology Department, Barzilay Medical

Center, affiliated to Ben Gurion University, Ashkelon, Israel;

Psychiatry Division, Barzilay Medical Center, affiliated to Ben

Gurion University, Ashkelon, Israel;

Division of Psychopharma-

cology, Vanderbilt University Medical Center, Nashville, TN, USA

International Congress on Schizophrenia Research

30 3. 3. Drug Side Effects and Physical Illness

Background: There are multiple causes of hyper-CKemia (HCK)—the el-

evation of serum creatine kinase (SCK) activity above upper limits. These

increases may be related to underlying trait-like biological abnormalities in

psychosis, phasic episodes related biological abnormalities, or the effect of

drug treatment. Previously we found, that among HCKemic patients, the

majority was treated with atypical antipsychotic drug (AAPD) and some

of them had neuromuscular complaints. The purposes of this study were to

follow-up prospectively and to estimate the incidence and severity of prob-

able neuromuscular dysfunction in HCKemic schizophrenia and schizoaf-

fective patients. Methods: For this study, we selected 39 HCK-emic

patients suffering from schizophrenia or schizoaffective disorder. Eleven

patients, treated either with clozapine, olanzapine or perphenazine had

persistent/recurrent hyper-CKemia (PHCK). 25 patients had occasional

HCKemia (OHCK). Blood samples for CK determinations were collected

up to three years of follow-up. Each year, all patients were assessed neu-

rologically for possible muscular and peripheral nervous systems involve-

ment. Results: During the study, SCK activity in PHCK patients was

found to be persistently elevated: 400

6 200 (mean 6 SD) in range

250–950 IU/L. Five of these patients had complaints of muscular weak-

ness, early fatigue on exercise, some muscular pains and cramps. In two of

them clinical assessment revealed mild general muscular weakness, espe-

cially in the proximal parts of the limbs. This pattern was diagnosed as

probable mild myopathy. In majority of OHCK patients SCK activity

was reduced to the normal levels range. Some patients remained with oc-

casional HCK without any kind of neuromuscular pathology. Conclu-

sions: The results of this first long-term large-scale prospective

comparative study indicate that among HCK-emic schizophrenia and

schizoaffective patients, only PHCK pattern may consider to be a bio-

chemical marker of the neuromuscular dysfunction. Meanwhile, OHCK

is likely to be a benign phenomenon, has no clinical manifestations,

and does not need to be followed-up routinely. Further investigation of

neuromuscular dysfunction, its mechanisms, pathophysiological signifi-

cance and possible pharmacogenetics associations in schizophrenia

patients is certainly indicated. The relationship of HCK to mutations

in neuregulin, dysbindin or reelin—the genes related both to schizophrenia

and to muscle, is also of interest.

ID: 549626

CHANGES IN BODY MASS INDEX (BMI) DURING

TREATMENT WITH OLANZAPINE VERSUS

AMISULPRIDE: A PROSPECTIVE STUDY

Christian Georg Widschwendter, A. Hofer, S. Baumgartner,

M. Edlinger, M. A. Rettenbacher, W. W. Fleischhacker

Biological Psychiatry, Medical University Innsbruck, Innsbruck,

Austria

Adiposity (body mass index [BMI]  30), hyperlipidemia, hypertension

and increased blood glucose levels, which define the metabolic syndrome,

are common side effects of atypical antipsychotic treatment. The aim of

this prospective study was to examine changes in BMI during treatment

with olanzapine versus amisulpride over a twelve-week period (day 0,

week 4, week 8, week 12). 46 patients with schizophrenia (ICD-10) were

included into the study (olanzapine: n = 25, amisulpride: n = 21), 32 patients

(olanzapine: n = 14, amisulpride: n = 18) completed the trial. BMI increased

by 1.81

6 1.0 in the olanzapine group (baseline-BMI = 23.7 6 4.8), whereas

BMI decreased by 0.21

6 1.6 in the amisulpride group (baseline-BMI =

28.5 6 8.0). Analysis of covariance (adjustment for baseline differences)

showed a significant difference (P = .01) in BMI change to the disadvan-

tage of Olanzapine. These results support differences in metabolic

side effects of olanzapine and amisulpride and further emphasize the

need for a risk-benefit analysis regarding the choice of antipsychotics in

schizophrenia.

ID: 549363

RELATIONSHIP BETWEEN THE ADIPONECTIN

(APM1) 276G/T VARIANT AND METABOLIC

DISTURBANCES IN A SCHIZOPHRENIA

POPULATION RECEIVING ANTIPSYCHOTICS

Vicki Ellingrod

, M. Meden

, T. B. Grove

, S. F. Taylor

J. Moline

, D. D. Miller

, T. L. Holman

, J. Kerr

University of Michigan, Ann Arbor, MI, USA;

University of Iowa,

Iowa City, IA, USA;

Cleveland Clinics, Cleveland, OH, USA

Purpose: To explore the relationship between the adiponectin (APM1)

276G>T (rs1501299) variant and metabolic abnormalities in schizophrenia

subjects receiving antipsychotics. Methods: Subjects with schizophrenia, re-

ceiving antipsychotic treatment for  12 months were screened for the met-

abolic syndrome (NCEP ATP-III criteria), including blood pressure, height,

weight, and fasting laboratory measures (cholesterol panel, glucose, insulin,

hemoglobin A1C), and a DNA sample. From these measures, body mass

index (BMI) and a Homeostasis Model Assessment Insulin Resistance

(HOMA-IR) value was calculated. Results: A total of 110 schizophrenia

spectrum disorder subjects were recruited. The sample contained 68 men

(62%), with a mean age of 41

6 11.17 years, and 40% meeting metabolic syn-

drome criteria. Eight-one percent of subjects were Caucasian and 75% were

receiving olanzapine, clozapine, risperidone or quetiapine at the time of eval-

uation. Eight subjects with a known diabetes diagnosis were excluded from

the analysis. The T allele was present in 51% of subjects and 49% had the GG

genotype, which was in Hardy Weinberg (v

= 0.68, P = .41). There were no

differences in demographic or laboratory values between genotype groups

and no relationship between APM1 and a metabolic syndrome diagnosis

= 1.086, P = .29). However, subjects with a T allele had lower BMIs com-

pared to the GG genotype group (P = .007). Overall, a significant interaction

between the APM1 276 G>T genotype andBMI wasfound, wherethose with

the T allele had higher levels of insulin resistance (HOMA-IR) at similar

BMIs compared to the GG genotype group (F = 9.94, df = 3,104, P <

.0001). After controlling for BMI differences, this interaction remained sig-

nificant (P = .0035). Conclusions:Schizophrenia subjects are at increasedrisk

for type II diabetes and insulin resistance. New data suggests APM1

variation may play a role. We found the APM1 276T allele may increase in-

sulin resistance risk at similar BMIs compared to the GG genotype. These

results should be interpreted cautiously due to the small sample size, lack of

adiponectin levels, and the need for replication. Acknowledgements: Project

support obtained from the NIMH (K08 MH64158), the National Center for

Research Resources, General Clinical Research Centers Program (M01-RR-

59 and UL1RR024986), a University of Michigan College of Pharmacy

Vahlteich Award, and Washtenaw Community Health Organization

(WCHO).

ID: 549235

EPIDEMIOLOGICAL STUDY FOR THE

EVALUATION OF METABOLIC DISORDERS IN

PATIENTS WITH SCHIZOPHRENIA: THE

METEOR STUDY

Marc De Hert

, B. Falissard

, M. Mauri

, K. Shaw

T. Wetterling

Universitair Centrum St-Jozef,, Kortenberg, Belgium;

INSERM

U669, PSIGIAM,, ‘‘Paris Sud Innovation Group in Adolescent

Mental Health’’ Maison de Solenn, Paris, France;

Clinica

Psichiatrica,, University di Pisa, Pisa, Pisa, Italy;

Klinik fu

Psychiatrie und Psychotherapie—Gerontopsychiatrie, Vivantes

Klinikum Hellersdorf, Berlin, Germany;

Queen Alexandra

Hospital, Portsmouth Hospitals NHS Trust, Hants, United

Kingdom

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 31

A large, observational, multinational cross-sectional pharmacoepidemio-

logical study was initiated in 2006 in order to determine the prevalence

of diabetes and other metabolic disorders in patients with schizophrenia

receiving atypical antipsychotic drugs in Europe. Secondary objectives in-

cluded identification of risk factors for metabolic disorders. The study in-

cluded adult outpatients with schizophrenia (DSM IV TR) treated for at

least 3 months by an antipsychotic drug. In each country, patients treated

with classical or atypical antipsychotic drugs were recruited into two par-

allel strata (ratio of 1:3). The study consisted of a single visit at which a fast-

ing blood sample was taken and height, weight, waist and hip circumference

and blood pressure measured. Information was collected on socio-demo-

graphic variables, lifestyle, comorbidities and co-medications, metabolic

and cardiovascular risk factors, psychiatric history and antipsychotic

drug history. Biochemical parameters assessed included glucose, insulin,

HbA1c, triglycerides, total cholesterol, HDL cholesterol and apolipopro-

tein B. 2463 patients were included in 12 countries (Bulgaria, Czech Repub-

lic, Estonia, Germany, Greece, Italy, Latvia, Lithuania, Romania, Russian

Federation, Slovenia and Turkey). Median age of included patients was

41.0 years and 54.6% were male. Medical history revealed that 10.9% of

patients were treated for arterial hypertension, 7.1% of patients for a lipid

disorder, 0.3% for type I diabetes and 3.5% for type II diabetes. In addition,

26% of untreated patients presented biochemical evidence of dysglycemia,

67.7% of dyslipidemia and 38% had hypertension. A metabolic syndrome

was identified in 34% of the patients. No overall difference was observed in

the proportion of patients with glycemic and lipid disorders between the

two strata of antipsychotic treatment (classical or atypical). Mean weight

and weight gain, abdominal obesity and the proportion of patients with

BMI  25 kg/m2 were slightly higher in the atypical stratum compared

to the classical stratum. Hypertension was more frequent in the classical

stratum (47.3%) than in the atypical stratum (42.2%). The results observed

in this large epidemiological study emphasize the need for careful follow-up

of patients with schizophrenia treated with antipsychotic drugs to detect

metabolic disorders. The rate of patients with glycemic or lipid disorders

was very high and as such largely underdiagnosed. Funding: Sanofi-Aventis.

ID: 549033

BONE MINERAL DENSITY AND OSTEOPOROSIS

RISK IN OLDER PATIENTS WITH SCHIZOPHRENIA

Joo-Cheol Shim

, D. Jung

, S. Jung

, T. Min

,M.Oh

, B. Kong

D. L. Kelly

Psychiatry, Inje University, Busan, South Korea;

Psychiatry,

Dongseo Hospital, Masan, South Korea;

Psychiatry, Dongrae

Hospital, Busan, South Korea;

Psychiatry, Saemyung Hospital,

Busan, South Korea;

Psychiatry, Maryland Psychiatric Research

Center, Baltimore, MD, USA

Osteoporosis is a significant public health concern because of its morbidity

and mortality associated with fractures and its associated health care costs.

Patients with schizophrenia are at higher risk for developing osteoporosis as

compared to the general population. Aging and gender are risk factors for

developing bone loss. In our previous study, middle aged women had a sig-

nificantly greater rate of bone loss compared to men of similar ages (85.7 %

vs 50%; P < .05). In this study, we evaluated the prevalence of bone density

and gender differences in older patients (> 50 years) with a DSM-IV diag-

nosis of schizophrenia. Patients with nutritional impairments or a medical

disorder known to be risk factor for osteoporosis were excluded. Bone den-

sitometry testing was performed by dual energy x-ray absorptiometry

(DEXA) (GE Lunar 4500 scanner), in the lumbar spine (L1-L4) and in

the femoral neck, trochanter, and intertrochanteric regions of the proximal

right femur. WHO definitions were used for osteoporosis and osteopenia.

The mean age in this study (N = 312 patients (184 men, 130 women) was

56.7

6 8.1 years; duration of illness, 142.6 6 79.2 months; and chlorprom-

azine equivalent dose of antipsychotics, 589

6 518mg/day (no gender differ-

ences). Cigarette smoking was significantly higher in men compared with

women (66.9% vs, 11.3%; P < .001). The lifetime prevalence of fracture was

not significantly different in men and women (24.7% vs. 21.1%). Prevalence

of bone loss was 75.2% in men and 79.9% in women. The prevalence of

osteoporosis was significantly different between men and women patients

(P < .010); osteoporosis was higher in women as compared to men (39.5%

vs 27.2%), while the prevalence of osteopenia was higher in men (52.7% vs.

35.7%). Women had significantly lower bone density volumes in L1-L4,

neck and the trochanter region of the femur compared to men (P <

.001). T-scores, however, only in the trochanter region of the femur,

were significantly lower in women than men (P = .035). Smokers showed

significantly lower absolute bone density (0.72

6 1.22 vs. 0.78 6 0.15, P <

.0003) and t-scores (0.42

6 1.22 vs. 0.04 6 1.32, P < .003) than non-

smokers. In conclusion, bone density loss in women with schizophrenia

was more severe as compared to men, however, the prevalence of bone den-

sity loss and lifetime fractures in older men was also high, thus careful at-

tention is needed to monitor men as well as women with schizophrenia.

ID: 549015

PRELIMINARY RESULTS OF THE MEAC STUDY:

METABOLIC EFFECTS OF ANTIPSYCHOTICS IN

CHILDREN

Ginger Nicol, D. Haupt, K. Flavin, J. Schweiger, M. Hessler,

E. Hessler, M. Yingling, J. W. Newcomer

Psychiatry, Washington University School of Medicine, St. Louis,

MO, USA

The purpose of the ongoing Metabolic Effects of Antipsychotics in Chil-

dren study (MEAC) is to quantify changes in markers of adiposity and

insulin sensitivity associated with antipsychotic treatment. Antipsy-

chotic-naı

ve children ages 7–18 with clinically significant aggression

and/or irritability (a score of > 18 on the Aberrant Behavior Checklist

(ABC) irritability subscale) who are already planning to start antipsychotic

treatment for one or more DSM-IV psychiatric disorder(s) are randomized

to 12 weeks of treatment with risperidone, olanzapine or aripiprazole fol-

lowing baseline evaluations of adiposity, insulin sensitivity and metabolic

indices. Dual energy x-ray absorptiometry (DEXA), abdominal magnetic

resonance imaging, modified frequently-sampled oral glucose tolerance

tests (mOGTT), and stable isotopomer tracer methodology during hyper-

insulinemic-euglycemic clamp conditions, as well as fasting labs and an-

thropomorphic measures, are used to assess baseline and treatment-

emergent changes in adiposity and insulin sensitivity, as well as secondary

measures, over the 12 weeks of treatment. Preliminary analyses, pooling

treatment groups into an overall study sample, indicate that 12 weeks

of initial antipsychotic treatment is associated with a statistically signifi-

cant increase in adiposity as measured by DEXA percent body fat, with

a mean increase of 2.91%

6 3.53 (mean 6 SD) corresponding to

a mean DEXA total fat increase of 2.85 kg (

69.35). The primary measure

of insulin sensitivity, derived from hyperinsulinemic-euglycemic clamps,

indicated a statistically significant pooled-groups decrease in whole

body insulin sensitivity. Preliminary results also indicate that BMI percen-

tile and fasting triglyceride undergo changes similar to the patterns ob-

served in the primary measures of adiposity and insulin sensitivity.

Finally, treatment produced marked improvement in the ABC irritabil-

ity/aggression score. These preliminary results demonstrate important

effects of antipsychotic treatment on adiposity and measures of insulin sen-

sitivity within the first 12 weeks of treatment, relevant to long-term car-

diometabolic risk prediction. Clinically available surrogates for

adiposity and insulin sensitivity, BMI percentile and fasting plasma triglyc-

eride, respectively, also indicate a similar pattern of change during treat-

ment. Importantly, all three treatments produce clinically and statistically

significant improvement in mood and behavior.

ID: 548877

International Congress on Schizophrenia Research

32 3. 3. Drug Side Effects and Physical Illness

ANTIPSYCHOTIC POLYPHARMACY AND RISK OF

DEATH FROM NATURAL CAUSES IN PATIENTS

WITH SCHIZOPHRENIA

Lone Baandrup

, C. Gasse

, V. D. Jensen

, B. Glenthøj

M. Nordentoft

, H. Lublin

, A. Fink-Jensen

, A. Lindhardt

P. B. Mortensen

Center for Neuropsychiatric Schizophrenia Research, Psychiatric

Center Glostrup, Copenhagen University Hospital, Glostrup,

Denmark;

National Center for Register-based Research, University

of Aarhus, Aarhus, Denmark;

Psychiatric Center Bispebjerg,

Copenhagen University Hospital, Copenhagen N, Denmark;

The

Danish Medicines Agency, Copenhagen S, Denmark;

Psychiatric

Center Rigshospitalet, Copenhagen University Hospital,

Copenhagen, Denmark

Concomitant prescription of more than one antipsychotic agent (antipsy-

chotic polypharmacy) in the treatment of schizophrenia is prevalent al-

though monotherapy is generally recommended. Mortality from natural

causes is markedly increased in schizophrenia and the role of polypharmacy

in this remains controversial. The objective of this study was to investigate if

antipsychotic polypharmacy is associated with the excess mortality from

natural causes among patients with schizophrenia. The study was designed

as a population-based nested case-control study using Danish central regis-

ters. From the study population of 27633 patients with schizophrenia or

other mainly non-affective psychoses, aged 18–53 years, we identified

193 cases who died of natural causes within a 2-year period, and 1937

age- and sex-matched controls. Current drug use was defined as at least

one prescription filled within 90 days before the date of death or the index

date. The data were analysed by conditional logistic regression. The risk of

natural death did not increase substantially with the number of concur-

rently used antipsychotic agents compared with antipsychotic monother-

apy (no antipsychotics: adjusted odds ratio 1.48 [95% confidence

interval 0.89–2.46]; two antipsychotics: 0.91 [0.61–1.36]; three or more anti-

psychotics: 1.16 [0.68–2.00]). Current use of benzodiazepine derivatives

with long elimination half-lives (more than 24 hours) was associated

with increased risk of natural death in patients with schizophrenia treated

with antipsychotics (1.78 [1.25–2.52]). We conclude that antipsychotic pol-

ypharmacy was not associated with substantially increased risk of natural

death. However, benzodiazepines, particularly those with long elimination

half-lives, should be cautiously prescribed in combination with antipsy-

chotics. The authors were supported by the National Board of Health in

Denmark, the Wørzner Foundation, and an unrestricted grant from H.

Lundbeck, Denmark.

ID: 548790

PREDICTORS OF CARDIOVASCULAR MORTALITY

IN HOSPITALIZED PATIENTS WITH SEVERE

MENTAL ILLNESS

Kimberly R. Warren

, F. Liu

, R. P. McMahon

, H. J. Wehring

R. C. Love

, K. M. Mackowick

, J. C. Shim

, D. L. Kelly

Maryland Psychiatric Research Center, University of Maryland

School of Medicine, Catonsville, MD, USA;

Department of

Pharmacy Practice and Science, University of Maryland School of

Pharmacy, Baltimore, MD, USA;

Inje University, Gimhae, South

Korea

People with mental illness have an increased risk of comorbid medical con-

ditions (Sokal, et al., 2004) and premature death (Harris and Barraclough,

1998). Cardiovascular (CV) disease is the second leading cause of death in

patients with schizophrenia (Osbey et al., 2000). The present study inves-

tigated predictors of CV mortality in hospitalized people with severe mental

illness: psychosis (predominantly schizophrenia (SZ)), bipolar disorder

(BD), or other mental illness (predominantly depression and anxiety disor-

ders (OTH)). Clinical chart data was abstracted and available on 1828

patients (n

=1443, n

=165, n

OTH

=220) from inpatient State psychiatric

hospitals between 1994 and 2000 in order to assess risk factors for cardio-

vascular (CV) mortality. Deaths were identified from the Social Security

Death Index and death records were collected for all decedents. Males

(n = 1065) and females (n = 762) between the ages of 35–65 were included.

Groups were similar in age, smoking status, and sex, however, the SZ group

had more non-whites (41%) than BD (29%) or OTH (22%). There were 60

male

= 30, n

female

= 30) deaths in this sample. A Cox proportional hazards

model, including age>55, race (white vs. non-white), sex, smoking status,

and diagnosis, was used for multivariate survival analysis. Increased CV

mortality was associated with age>55 (hazard ratio (HR) = 4.1, v

15.76, P < 0.001), white race (HR = 2.4, v

= 4.07, P < 0.05), and smoking

(HR = 2.0, v

= 3.63, P = .057). Compared with patients with SZ, increase in

CV mortality was not significant in patients with bipolar disorder (HR =

1.7, v

= 0.71, P = .399), and was marginally elevated in the OTH group

(HR = 2.7 v

=3.30, P = .069). Sex was not significantly associated with

CV mortality, P = .25. These findings are consistent with numerous studies

that suggest that older age and smoking are risk factors for heart disease

(Hennekens, 2007). Previous studies have also indicated greater CV mor-

tality risks for patients with anxiety and depressive disorders than patients

with schizophrenia (Harris and Barraclough, 1998). Racial differences in

CV mortality in patients with severe mental illness require further investi-

gation. Funded by the National Institutes of Mental Health (NIMH R03

MH069871-01; Kelly, PI) and the Advanced Centers for Intervention and

Services Research (NIMH P50 MH40279; Carpenter, PI).

ID: 548416

HANDWRITING MOVEMENT KINEMATICS FOR

QUANTIFYING ANTIPSYCHOTIC-INDUCED EPS

Michael Caligiuri

, H. L. Teulings

, A. Niculescu

3,4

, C. E. Dean

J. B. Lohr

UCSD, La Jolla, CA, USA;

Neuroscript, LLC, Tempe, AZ, USA;

Indiana University, Indianapolis, IN, USA;

Indianapolis VA

Medical Center, Indianapolis, IN, USA;

Minneapolis VA, Min-

neapolis, MN, USA

Ongoing monitoring of antipsychotic-induced extrapyramidal side effects

(EPS) is important to maximize treatment outcome and improve medication

compliance. Conventional clinical assessments of EPS appear insensitive to

differences across various atypical antipsychotics and examiner bias can re-

duce their reliability. Instrumental methods emerged as a remedy to these

problems, however, these systems have had limited clinical utility because

of their complexity and cost. We developed a novel method based on quan-

tifying handwriting movements that overcomes previous limitations of com-

plexity and cost. In our study we test the hypothesis that handwriting

movement abnormalities are associated with severity of EPS but are inde-

pendent of age, severity of psychosis, depression, or other factors that

can bias the clinical assessment of EPS. We also tested whether handwriting

movements differed across four commonly used atypical antipsychotics.

Handwriting movements from 87 psychosis patients and 45 healthy compar-

ison subjects were quantified during a loop-drawing task. Participants were

instructed to draw continuous loops 2 cm high from left to right across a dig-

itizing tablet. Samples were recorded and analyzed using commercial soft-

ware. Kinematic variables included vertical loop size and peak velocity,

relative time to peak velocity, and pen contact duration per stroke

for both primary and secondary submovements. Patients with clinically

determined EPS exhibited smaller vertical heights (P < .001), lower peak ve-

locities (P < .05), longer times to peak velocity (P < .05), and increased du-

ration of pen contact (P < .05) compared to patients without EPS. Patients

treated with aripiprazole or risperidone exhibited significantly more im-

paired handwriting movements than patients treated with quetiapine or

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 33

olanzapine, especially for peak velocity (P < .01), whereas, clinical EPS

assessments were insensitive to these medication differences. Unlike clinical

EPS assessments, kinematic handwriting variables were independent of se-

verity of positive and negative symptoms of psychosis, depression, or demo-

graphic variables. These findings support the high specificity of our

kinematic handwriting measures to EPS. Differences between various anti-

psychotic medications suggest that handwriting kinematics may be useful in

measuring the effects of switching medication in patients with pre-existing

EPS. Research supported by NIH grant R44 MH073192.

ID: 547581

NON-PHARMACOLOGICAL INTERVENTION

FOR WEIGHT GAIN MANAGEMENT IN SEVERE

MENTAL DISORDERS: RESULTS FROM A LARGE

MULTICENTRIC STUDY

Cecı

lia Attux

, L. C. Martini

, A. M. Roma

, C. M. Arau

B. Petreche

, D. F. Cangucxu

, E. Mc Mullan

, M. G. Camargo

P. L. Castro

, P. A. Barbosa

, A. F. Reis

, R. A. Bressan

Psychiatry, Federal University of Sa˜o Paulo, Sa˜o Paulo, Brazil;

Endocrinology, UNIFESP—Federal University of Sa˜o Paulo, Sa˜o

Paulo, Brazil

Background: A high prevalence of metabolic abnormalities has been ob-

served in patients with severe mental disorders and schizophrenia. Obesity

interventions studies in this population are scarce. Objective: Evaluate the

effectiveness of a non-pharmacological intervention in controlling body

weight and metabolic parameters. Method: An open, multicentric, longitu-

dinal study was conducted on 57 mental health services in 34 cities in Brazil.

Patients included in this study received a 12-week 1-hour group interven-

tion. Topics such healthy nutrition, lifestyle, physical activity and self-es-

teem were discussed with patients and their relatives. Weight, high, waist

circumference and blood pressured were measured before and after the in-

tervention. A total of 656 patients participated in the intervention. All of

them were in use of some antipsychotic and/or mood stabilizers and had

some concern about weight gain. Results: 512 patients completed the inter-

vention. A significant weight loss (0.5

6 3.4kg, P < .000) and a significant

BMI reduction (0.2 6 1.4 kg/m2, P < .007) were observed. There was

a marked increase in physical activity after the intervention, at the baseline

55.2% of the patients were enrolled in some kind of exercise program, after

the intervention, 72.8% were regularly practicing physical activity. Conclu-

sion: To our knowledge, the Brazilian Wellness Program is the largest na-

tional, multicentric, naturalistic study of a non-pharmacological

intervention for severe mental disorder patients showing positive outcomes

on weight gain management, reducing diastolic blood pressure and increas-

ing physical activity. Efficacy and effectiveness studies are being conducted

to further inform the implementation of the wellness intervention as public

mental health policy.

Table. Clinical parameters at baseline and 3 months (n = 512)

Variable

Baseline

(mean 6 SD)

3 months

(mean 6 SD) Analysis P

Weight (kg) 83.2 (18.7) 82.7 (18) t = 3.549; df = 511 .000

BMI (kg/m2) 31.3 (5.6) 31.1 (5.6) t = 2.696; df = 510 .007

Waist (cm) Male 109.2 (15.2) 108.6 (14.1) t = 1.332; df = 151 .185

Waist (cm) Female 102.1 (15.5) 101.8 (13.7) t = 0.438; df = 305 .662

Systolic BP (mmHg) 119.6 (15.4) 118.6 (14.7) t = 1.653; df = 483 .099

Diastolic BP (mmHg) 79.8 (11.7) 78 (10.5) t = 3.655; df = 484 .000

BMI: body mass index; BP: blood pressure; SD: standard deviation; df:

degrees of freedom

ID: 546913

PIOGLITAZONE AS A TREATMENT FOR GLUCOSE

AND LIPID METABOLIC ABNORMALITIES IN

SCHIZOPHRENIA

Robert C. Smith

1,2

, N. Bark

7,8

, J. P. Lindenmayer

1,2

, J. M. Davis

H. Jin

5,9

, A. Shekhar

, J. Cornwell

, T. Viviano

, M. A. Caldwell

C. Li

Psychiatry, NYU Medical School, New York, NY, USA;

Psychatry, Manhattan Psychaitric Center, New York, NY, USA;

Psychiatry, Univ of Indiana Medical School, Indianpolis, IN,

USA;

Psychiatric Institute, Univ of Illinois Medical School,

Chicago, IL, USA;

Psychiatry, Univ of California , San Diego,

San Diego, CA, USA;

Psychiatry, Shanghai Mental Health

Center, Shanghai, China;

Psychiatry, Bronx Psychiatric Center,

Bronx, NY, USA;

Psychiatry, Albert Einstein College of

Medicine, Bronx, NY, USA;

Psychiatry, VA Medical Center, San

Diego, CA, USA

Diabetes, hyperglycemia, and hyperlipidemia more prevalent in psychiat-

ric patients and increased by treatment with antipsychotic drugs. Research

by our own and other groups have shown that some second generation

antipsychotics increase insulin resistance even in patients who do not meet

metabolic criteria for diabetes. Pioglitazone is a PPAR agonist which spe-

cifically acts by increasing insulin sensitivity, a mechanism of action dif-

ferent form other antidiabetic drugs, and has beneficial effects on

decreasing triglycerides and increasing HDL. We present preliminary

data from a double-blind placebo controlled study of pioglitazone as

a treatment for glucose and lipid abnormalities in schizophrenia and as-

sessment of its potential cognitive effects. Method: Patients with schizo-

phrenia or schizoaffective psychosis who were treated with antipsychotic

medications and had fasting glucose  100 mg/dl and fasting triglycerides

 120 mg/dl participated in a 12 week double blind study in which they

received pioglitazone or placebo. Fasting glucose, insulin, triglycerides

and other metabolic parameters were measured monthly and a glucose

tolerance test was performed at baseline and end of study. Cognitive func-

tions was assessed by RBANS, CPT, and RANDT and psychopathology

asessed by PANSS. Preliminary results from the first 29 patients showed

that by the end of treatment patients treated with pioglitazone had lower

fasting glucose and triglycerides, and higher fasting HDL than patients

treated with placebo. 46% of placebo treated patients had diabetic range

fasting glucose levels compared to 13% of pioglitazone patients. There

was no drug differences in serum insulin but pioglitazone produced an

increase in insulin sensitivity measured by the homo procedure. By the

end of treatment Pioglitazone treated patients had lower 1 and 2 hr glu-

cose glucose levels on the 75 mg GTT test than placebo patients. Piogli-

tazone patients showed a decrease in PANSS. There were no significant

drug effects on the cognitive performance.Pioglitazone may be a useful

treatment for reducing or reversing metabolic abnormalities in schizo-

phrenic patients treated with antipsychotic medication. Because pioglita-

zone increases insulin sensitivity and may prevent or delayed conversion

of impaired fasting glucose to diabetes it might be considered as a preven-

tive strategy in patients treated with some antipsychotic medications.

ID: 541974

GLUCOSE AND LIPID DISTURBANCES AFTER

1 YEAR OF ANTIPSYCHOTIC TREATMENT IN A

DRUG-NAI

VE POPULATION

Rocio Perez-Iglesias

, I. Mata

, J. M. Pelayo-Teran

J. A. Amado

, M. T. Garcia-Unzueta

, A. Berja

, O. Martinez-

Garcia

, J. L. Vazquez-Barquero

, B. Crespo-Facorro

Psychiatry, Marques de Valdecilla University Hospital, Santander,

Spain;

Endocrinology, Marques de Valdecilla University Hospital,

International Congress on Schizophrenia Research

34 3. 3. Drug Side Effects and Physical Illness

Santander, Spain;

Biochemistry, Marques de Valdecilla University

Hospital, Santander, Spain

The objective of this study is to examine the main metabolic side effects

induced by antipsychotic treatment in a cohort of first-episode drug-naı

subjects after the first year of treatment. A randomized, open-label, pro-

spective clinical trial was conducted. Participants were 164 consecutive

subjects included in a first episode program and never treated with anti-

psychotic medication. Patients were assigned to haloperidol, olanzapine

or risperidone. The main outcome measures were changes at one year in

fasting glucose parameters (glucose, insulin levels and insulin resistance

index) and changes in fasting lipid parameters (total cholesterol, triglycer-

ides, LDL cholesterol and HDL cholesterol). At one year follow-up 144

subjects were evaluated. There was a statistically significant increase in the

mean values of insulin levels (3.5 microU/ml; P = .004); HOMA insulin

resistance index (0.7; P = .013); total cholesterol (22.2 mg/dl; P < .001);

LDL-cholesterol (15.1mg/dl; P < .001) and triglycerides (36.6 mg/dl;

P < .001). No pathological values in fasting glucose plasma levels were

found at baseline and there were no changes after one year. Weight

gain was positively correlated with changes in insulin increase (r = .32,

P = .001); HOMA index (r = .34, P < .001) and triglyceride levels (r =

.23, P = .016) and negatively correlated with HDL changes (r = 0.25,

P = 0.007). We did not detect statistically significant differences between

treatments in any of the parameters evaluated. Fasting glycaemia and in-

sulin concentrations at baseline do not support the hypothesis that schizo-

phrenia is associated with an underlying abnormality in glucose

metabolism. To summarize, after the first year of antipsychotic treatment

we have observed an increase in insulin-resistance index and a worsening

lipid profile—but no clinically relevant illness was detected. The changes

in insulin and lipid parameters seem to be related to the magnitude of

weight gain. There were no significant differences between haloperidol,

olanzapine and risperidone concerning metabolic adverse effects after

the first year of treatment. Grants: Instituto de Salud Carlos III, FIS

00/3095 and SENY Fundatio Research Grant CI 2005-0308007, Funda-

cion Marques de Valdecilla A/02/07. No pharmaceutical company sup-

plied financial support.

ID: 541006

SELF-RATING OF PHARMACOLOGICAL

TREATMENT(SIDE-)EFFECTS IN SCHIZOPHRENIA:

A COMPARISON OF INSTRUMENTS

Hugo Wolters, R. J. van den Bosch, D. Wiersma, H. Knegtering

University Medical Centre Groningen, University of Groningen,

Groningen, Netherlands

Introduction: To support clinical practice as well as clinical research, self-

rating scales have been developed to evaluate desired as well as undesired

effects of antipsychotic treatment. Psychometric properties and other char-

acteristics of four identified self-rating scales including their relationship

with subjective quality of life will be compared. Method: Four scales of

self-report of treatment effects of antipsychotics were identified through

a MEDLINE and cross-references search: the Drug Attitude Inventory

(DAI-10), the Liverpool University Neuroleptic Side Effect Rating

Scale (LUNSERS), the Subjective Wellbeing to Neuroleptics (SWN)

and the Subjects’ Reaction to Antipsychotics questionnaire (SRA). All

scales were filled out by 320 patients with schizophrenia treated with

antipsychotics together with a subjective quality of life instrument, the

WHO-QoL BREF. Results: The DAI-10 contains ten items covering

many different aspects of treatment with antipsychotics, such as compli-

ance and attitudes towards neuroleptic treatment. The scope of the LUNS-

ERS is measuring undesired effects of antipsychotics, while the SRA

incorporates desired as well as undesired treatment effects. The SWN con-

sists mainly of items on aspects of wellbeing without a direct link to anti-

psychotics. The instruments show an acceptable internal reliability

(Cronbach’s alpha’s varying between .64–93) with the exception of the

DAI-10 (Cronbach’s alpha .52), probably due to the mix of experiences.

The scales did not strongly correlate (r between .11 and .51) with each other

except for the SRA-subscale undesired experiences and the LUNSERS (r =

.68, P < .01). Correlations with quality of life were statistically significant,

with the SWN showing the strongest one (r = .78, P < .01). Conclusion:

The lack of internal reliability and the conceptual mixture of items restrict

the use of the DAI-10. Its strength lies in measuring compliance and atti-

tudes towards neuroleptic treatment. The LUNSERS maybe very useful

to screen extensively side-effects of neuroleptic treatment, with the caveat

of overestimation of the frequency of these effects. The SWN is to be rec-

ommended as an alternative for a QoL-instrument in patients (not only

with schizophrenia) using antipsychotic medication. The SRA is recom-

mended if a clinician or researcher is specifically interested in undesired

as well as desired responses attributed to the antipsychotic medication.

ID: 538965

PREDICTORS OF MORTALITY IN PATIENTS WITH

SERIOUS MENTAL ILLNESS AND CO-MORBID

DIABETES

Clayton H. Brown

1,4

, J. Leith

, F. B. Dickerson

, D. R. Medoff

L. J. Fang

, J. Kreyenbuhl

2,4

, R. W. Goldberg

2,4

, W. Potts

L. B. Dixon

2,4

Epidemiology and Preventive Medicine, University of Maryland

School of Medicine, Baltimore, MD, USA;

Department of Psy-

chiatry, University of Maryland School of Medicine, Baltimore,

MD, USA;

Sheppard Pratt Health System, Baltimore, MD, USA;

Mental Illness Research Edcuation and Clinical Center

(MIRECC), VA Capitol Health Care Network (VISN 5),

Baltimore, MD, USA

Objective: Persons with serious mental illness (SMI) have a higher rate of

type 2 diabetes and a higher mortality rate than the general population. The

purpose of this study was to assess baseline demographic and health-related

factors in the prediction of all-cause mortality over a seven-year follow-up

period in a group of diabetes patients with serious mental illness and a group

of diabetes patients without serious mental illness. Methods: From 1999 to

2002, 300 patients with type 2 diabetes were recruited from community

mental health centers in the greater Baltimore region and nearby primary

care clinics. Of these, 100 had a diagnosis of schizophrenia, 101 a diagnosis

of major mood disorder, and 99 had no identified mental illness. Patient

deaths over an average seven-year period after baseline assessment were

identified using the Social Security Administration’s Death Master File

via the website www.rootsweb.ancestry.com. Bivariate associations be-

tween predictors and mortality were assessed in the serious mental illness

group and compared to those in the group without serious mental illness.

This was followed-up with multivariable logistic regression. Results: A total

of 21 % of persons in each group died in the follow-up period. Median age

of death was 60. Age, duration of diabetes, and hospitalization for a diabe-

tes-related condition in the 6 months prior to baseline were all predictive of

mortality. In the SMI group (but not in the Non-SMI group) decedents

were more likely to have been smokers than non-decedents (69% versus

49%, P = .021) and in the multivariable model, odds of mortality was

over two times greater among smokers than non-smokers (adjusted odds

ratio (AOR) = 2.3, 95% CI: [0.96, 5.3]). In the Non-SMI group those

who were prescribed insulin had over a four-fold greater odds of mortality

(AOR = 4.1, 95% CI: [1.3, 13.1]), however in the SMI group there was no

evidence that being prescribed insulin served as a diabetes severity marker

(AOR = 0.59, 95% CI: [0.21, 1.6]). Conclusions: Diabetes may contribute to

the excess mortality of persons with serious mental illness. To better address

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 35

this problem, clinicians caring for persons with serious mental illness need

to be especially vigilant regarding mortality risk when the diabetes illness of

their patients requires hospitalization and as their patients age. Further-

more, it is important that smoking cessation be aggressively promoted

among persons with SMI.

ID: 550828

SUBJECTIVE EXPERIENCE MEASUREMENT IN

PATIENTS WITH SCHIZOPHRENIA/SCHIZO-

AFFECTIVE DISORDER USING THE NOVEL

ANTIPSYCHOTIC MEDICATION EXPERIENCE

SCALE (NAMES)

Donna A. Wirshing, J. Rosen, J. M. Pierre, Z. D. Erickson,

L. H. Guzik, S. J. Mena

Psychiatry, VA Greater Los Angeles Healthcare System, Los

Angeles, CA, USA

Rating scales are effective tools to assess therapeutic efficacy and patient

subjective experience. Patients suffering from Schizophrenia benefit from

treatment with psychotropic medications to reduce symptomatology, to

reduce the frequency of hospitalizations and to provide for better com-

munity reintegration. However, medication nonadherence is problematic

for the effective treatment of these patients, and patients’ negative sub-

jective responses to medications correlate with treatment noncompliance.

Therefore, assessment of the level of patient subjective experience of

a medication is an important correlate of treatment efficacy. Existing rat-

ing scales for psychotropics in Schizophrenia (eg, Drug Attitude Inven-

tory (DAI) 30, AMDP-5) were developed to assess patient report of

side effects prior to the advent of atypicals. However, these scales may

now be less useful given the application of atypicals which have different

side effect profiles. We report here a preliminary study of the validity and

reliability of the Novel Antipsychotic Medication Experience Scale

(NAMES), a new patient self-rating questionnaire. The NAMES, DAI

30, and AMDP-5 questionnaires were administered to patients meeting

inclusion criteria: 1) patients with schizophrenia/schizoaffective disorder,

between 18 and 65 years old; 2) at least 1 month of monotherapy with

clozapine, fluphenazine, haloperidol, olanzapine, quetiapine, risperidone

or ziprasidone; 3) the absence of psychiatric co-morbidity. 101 outpatients

completed the study. The results from the NAMES questionnaire are cor-

related with results from the DAI 30 and AMDP-5, with Pearson Corre-

lation Coefficients of 0.89, and .83. The NAMES Standardized Cronbach

Coefficient Alpha (SCCA) was 0.81, while the DAI 30 SCCA was 0.46,

and the AMDP-5 was .32. Subjective measures assessed by the NAMES,

show that the scale is both reliable and valid. Based upon a measure of

internal consistency, the NAMES has a higher average inter-item corre-

lation than the DAI 30 or the AMDP-5. Further analysis may indicate

that NAMES is a better measure of patient subjective experience of treat-

ment with atypicals than other rating scales because of administration

protocol.

ID: 551878

PRE-DIABETES AND DIABETES ASSESSMENT IN

VETERANS WITH SCHIZOPHRENIA

Laurel A. Copeland

1,2

, J. E. Zeber

1,2

, M. L. Parchman

1,3

V. A. Lawrence

1,4

, B. J. Hosek

, A. L. Miller

VERDICT (HSRD), South Texas Veterans Health Care System,

San Antonio, TX, USA;

Department Psychiatry, University of

Texas Health Science Center at San Antonio, San Antonio, TX,

USA;

Department Family and Community Medicine, University of

Texas Health Science Center at San Antonio, San Antonio, TX,

USA;

Department Medicine, University of Texas Health Science

Center at San Antonio, San Antonio, TX, USA

In February 2004, the American Diabetes Association and American Psy-

chiatric Association issued a Consensus statement on metabolic sequelae of

antipsychotic agents, summarizing research over the previous ten years that

documented growing awareness of this connection. The Pre-diabetes in Vet-

erans with Schizophrenia study assessed monitoring of pre-diabetes and un-

diagnosed diabetes just prior to the Consensus statement, in a national

sample of 39 825 Veterans Health Administration outpatients with schizo-

phrenia, aged 50 or older, with no recorded diagnosis of diabetes and no use

of hypoglycemic medications (non-diabetic at baseline). Diagnosis of

schizophrenia was defined by one inpatient or two or more outpatient di-

agnoses over the one-year study period. The retrospective study used fiscal

year 2002 data. Blood glucose tests with same-day low-density lipoprotein

(LDL) tests were used as a proxy for fasting glucose, because patients are

required to be fasting for the LDL test. Test time of day was not available,

nor was verification of fasting status. Hemoglobin A1c tests were also

extracted, as some authors advocate their use in monitoring pre-diabetes.

Tests were categorized: 100–125 mg/dL (pre-diabetes fasting glucose), > =

126 (diabetes-indicative fasting glucose), 5.8%–6.4% (pre-diabetes A1c), >

= 6.5% (diabetes-indicative A1c). Multivariable regressions modeled receipt

of a fasting glucose test, receipt of an A1c test, test results consistent with

pre-diabetes, and test results consistent with diabetes. Patient descriptives

included average age of 60 years (SD 9.5), 60% white, 42% unmarried, with

comorbid hypertension (40%), dyslipidemia (25%), and depressive disor-

ders (22%). Approximately 27% of patients had fasting glucose tests, of

whom 42% were pre-diabetic; 9% had A1c tests, of whom 28% were

pre-diabetic per study definition. Test rates among patients on atypical

antipsychotics were slightly higher: 29% fasting glucose, 10% A1c, but

within those groups, the proportions testing pre-diabetic were identical

(42% and 28%, respectively). Both glucose and A1c testing were associated

with hypertension, dyslipidemia, and atypical antipsychotics. Among tested

patients, pre-diabetes was associated with cardiovascular disease, older age,

and being married. Dysglycemia was prevalent among schizophrenia

patients, yet glucose monitoring was uncommon. Overall, few schizophre-

nia patients were being adequately monitored for dysglycemia.

ID: 551835

AUTONOMIC SYSTEM DYSREGULATION IN

SCHIZOPHRENIA AND BIPOLAR MANIA

Brook Henry, A. Minassian, M. Paulus, M. Geyer, W. Perry

Psychiatry, UC San Diego, San Diego, CA, USA

Heart rate variability (HRV) reflects the interchange of the parasympa-

thetic and sympathetic autonomic nervous system (ANS). Severe psychiat-

ric disorders such as schizophrenia have been associated with dysfunction

of the ANS, including alterations in cardiac function and HRV. Recent

reports suggest that ANS dysregulation among psychiatric patients may

be related to responses to physical and psychological stressors. Studies con-

sistently demonstrate a correlation between psychosis and heightened anx-

iety states even when the stressful condition is relatively minor. Still, it is

unclear whether this association is due to disease-related autonomic ner-

vous system changes or a side effect of antipsychotic medication. In the

present study we examined the autonomic response of hospitalized individ-

uals with schizophrenia and bipolar-mania when exposed to a novel

environment. The patients were compared to healthy comparison subjects.

Subjects were fitted with a vest that contained an ambulatory monitoring

device that allowed us to obtain continuous recordings of cardiac, respira-

tory, and motor activity. Both schizophrenia and bipolar-manic subjects

exhibited increased heart rate, increased respiration, and decreased HRV

International Congress on Schizophrenia Research

36 3. 3. Drug Side Effects and Physical Illness

during a 15 minute exposure to a novel but benign environment. Spectral

analysis of cardiac data revealed that both groups showed an increase in the

Low Frequency/High Frequency ratio of the HRV signal compared to com-

parison subjects, indicating increased sympathetic nervous system activity.

Group differences in autonomic function remained significant even after

excluding variance due to locomotion, indicating that changes in cardiac

activity and respiration were independent of differences in physical activity.

Importantly, ANS function was not significantly affected by medications.

These findings indicate that individuals with both schizophrenia and bipo-

lar-mania exhibit an increased autonomic response to a novel environment.

The data suggest that both of these disorders may be characterized by

a common deficit in the ability to adapt to challenges in the environment,

which may have a direct impact on cognitive, affective, and behavioral

function.

ID: 551807

SUBSTANCE ABUSE IS ASSOCIATED WITH

INCREASED EXTRAPYRAMIDAL SYMPTOMS

IN SCHIZOPHRENIA: A META-ANALYSIS

Ste

phane Potvin

1,2

, P. J. Blanchet

, E. Stip

1,2

Centre de recherche Fernand-Seguin, Montreal, QC, Canada;

Psychiatry, University of Montreal, Montreal, QC, Canada;

Stomatology, University of Montreal, Montreal, QC, Canada

Background: Psychoactive substances (PAS) may interact with antipsy-

chotics in the development of extrapyramidal symptoms (EPS) in schizo-

phrenia, since PAS exert impacts on the basal ganglia. Clinical data have

been gathered about the effects of PAS on EPS in schizophrenia, producing

inconsistent results. This meta-analysis sought to determine whether PAS

enhance EPS in schizophrenia patients. Methods: A search of the literature

using computerized engines was undertaken. Studies were retained in the

analysis if: (i) they included schizophrenia patients with and without sub-

stance abuse; and (ii) they comprised a measure of EPS using valid instru-

ments. Results: Fifteen studies were available identified, involving 3373

patients. The composite analysis revealed a small and positive effect size

(g = 0.246), suggesting increased EPS in substance-abusing patients.

Dual-diagnosis patients were more frequently males than single diagnosis

patients. Thus, we performed a sub-analysis of studies with no confounders

(age, sex, negative symptoms, etc.). The pooling of these 10 studies

produced a moderate and positive effect size (g = 0.405). Discussion:

Our results show that PAS negatively impact on EPS in schizophrenia, es-

pecially when potential confounding factors are controlled. As such, these

results have implications for the prevention of EPS in schizophrenia and for

the design of future studies on the topic.

ID: 551548

INFLAMMATORY MARKERS IN SCHIZOPHRENIA:

COMPARING ANTIPSYCHOTIC EFFECTS IN THE

CATIE SCHIZOPHRENIA TRIAL

Henry A. Nasrallah

, J. M. Meyer

, J. P. McEvoy

, V. G. Davis

D. C. Goff

, S. M. Davis

Psychiatry, University of California at San Diego, San Diego, CA,

USA;

Psychiatry, University of Cincinnati College of Medicine,

Cincinnati, OH, OH, USA;

Psychiatry, Duke University, Durham,

NC, USA;

Psychiatry, University of North Carolina, Chapel Hill,

NC, USA;

Psychiatry, Harvard University, Boston, MA, USA

Background: C-reactive protein (CRP), intercellular adhesion molecule-1

(ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and E-selectin are

systemic inflammatory markers (IM) that positively correlate with cardio-

vascular (CV) risk. Despite the known CV effects of atypical antipsy-

chotics, there is limited prospective data on IM changes during

treatment. Methods: IM outcomes were compared between antipsychotic

treatment groups in the CATIE Schizophrenia Trial phase 1 using subjects

with laboratory assessments at baseline and 3 months (n = 789). Results:

There were significant treatment differences in CRP, E-selectin and ICAM-

1 at 3 months, with an impact of baseline values on outcome. In overall

comparisons, olanzapine or quetiapine arms had the highest median levels

for each IM. In those with low baseline CV risk by CRP (< 1 mg/L), olan-

zapine was significantly different than perphenazine (P < .001) and risper-

idone (P = .001) after baseline adjustment. Among those with high baseline

E-selectin values, olanzapine was significantly different than perphenazine

(P = .005) and ziprasidone (P = .002), as was quetiapine vs. ziprasidone

(P = .004). Olanzapine and quetiapine also had higher 3-month ICAM-

1 levels than perphenazine in subjects with baseline ICAM-1 above the me-

dian, but the differences were not statistically significant after controlling

for multiple comparisons (P = .010 for both). The 18-month repeated

measures CRP analysis confirmed the significantly higher values for olan-

zapine in those with low baseline CRP. Conclusions: This analysis provides

further evidence for differential antipsychotic metabolic liabilities as mea-

sured by changes in systemic inflammation. CRP may emerge as a useful

target for CV risk outcomes in schizophrenia patients.

ID: 551428

MORPHOLOGICAL, CELLULAR AND

MOLECULAR CORRELATES OF ANTIPSYCHOTIC

MEDICATION EXPOSURE IN PRIMATE

NEOCORTEX

David A. Lewis

1,2

, K. A. Dorph-Petersen

1,3

, G. T. Konopaske

1,4

Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA;

Neuroscience, University of Pittsburgh, Pittsburgh, PA, USA;

Centre for Psychiatric Research, Aarhus University Hospital,

Risskov, Denmark;

McLean Hospital, Belmont, MA, USA

Schizophrenia is associated with a number of morphological, cellular and

molecular abnormalities in the cerebral cortex. Knowledge of the extent to

which prolonged treatment with antipsychotic medications contributes to

these disturbances is essential both for understanding the disease process

of schizophrenia and for gaining insight into the mechanisms underlying

the therapeutic and adverse effects of antipsychotic medications. To ad-

dress these issues, we exposed experimentally-naı

ve, young adult, male

macaque monkeys to twice-daily oral doses of haloperidol, olanzepine

or sham treatment for approximately two years. At steady state, trough

plasma levels of haloperidol and olanzapine were in the range reported to

be therapeutic in schizophrenia. Both the haloperidol and olanzapine-

exposed groups had significantly smaller total brain weight and volume.

At the cellular level, we found lower glial cell number and higher neuron

density in the cortex, without a difference in total neuron number, in the

antipsychotic-exposed monkeys; follow-up immunocytochemical studies

revealed a significant 21% lower astrocyte number with a non-significant

13% lower oligodendrocyte number in the antipsychotic-exposed mon-

keys. Similar effects were seen in both the haloperidol and olanzapine

groups. In contrast, none of the examined GABA-related transcripts

reported to be altered in schizophrenia differed across the three groups

of monkeys. Although species differences and other factors need to be

considered, these findings suggest that certain morphological and cellular

abnormalities identified in schizophrenia might be attributable to chronic

exposure to antipsychotic medications.

ID: 551304

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 37

SOMNOLENCE AND SEDATION IN ADOLESCENT

PATIENTS WITH SCHIZOPHRENIA TREATED WITH

ARIPIPRAZOLE IN AN ACUTE STUDY WITH LONG

TERM FOLLOW-UP

Suja J. Mathew

, R. D. McQuade

, W. H. Carson

, M. Nyilas

R. A. Forbes

, T. Iwamoto

, R. Mankoski

, A. Pikalov

Medical Affairs, Otsuka America Pharmaceutical, Inc, Rockville,

MD, USA;

Global Clinical Development, Otsuka Pharmaceutical

Development and Commercialization Inc., Princeton, NJ, USA;

Medical Strategy, Bristol Myers Squibb Co., Plainsboro, NJ,

USA

Background: Somnolence and sedation are relatively common side effects

of antipsychotics to which adolescents may be more sensitive. The purpose

of this study was to evaluate characteristics of somnolence and sedation in

adolescents with schizophrenia treated with aripiprazole. Methods: Data

are derived from a 6-week, double-blind, placebo controlled, randomized

clinical trial comparing two fixed doses of aripiprazole (10 and 30mg, N =

302) and a 26-week, open-label, follow-up study with flexible dosing (5-

30mg, N = 239). Somnolence and sedation were evaluated with respect

to time of onset, severity, dose-response, dose-reduction, discontinuation

rates, age, gender, and ethnicity. Results: In both the short- (ST) and

long-term (LT) studies, almost all reported cases of somnolence and se-

dation were classified as mild (ST: n = 25, 12%; LT: n = 24, 10%) or

moderate (ST: n = 8, 4%; LT: n = 9, 4%), and only one patient was

discontinued from the study due to somnolence. Most events (76%)

started early (within 2 and 4 weeks) and had resolved within the study

period. A possible dose-dependence relationship was seen, with more

events reported at the 30 mg dose. In the short-term trial, Blacks

reported substantially higher rates (35% in the 10 mg arm and 55%

in the 30 mg arm) than the overall population (12% in the 10 mg arm

and 22% in the 30 mg arm). No differences were noted in incidence strat-

ified by age or gender. In the long-term open-label study, the overall

incidence of somnolence and sedation was low (14%) and was the similar

between groups switched from aripiprazole 10 mg. 30 mg or placebo.

When sedation or somnolence occurred, most investigators took no ac-

tion or reduced the dose rather than discontinuing study drug. Conclu-

sions: Somnolence and sedation in adolescents with schizophrenia treated

with aripiprazole, if experienced, is typically mild, transient, and man-

ageable. In general, higher doses lead to an increased incidence in youn-

ger patients, and ethnicity may be a modifying factor and should be

evaluated further.

ID: 551235

PROSPECTIVE LONGITUDINAL EVALUATION OF

SERUM CREATINE KINASE LEVELS AND NEURO-

MUSCULAR DYSFUNCTION IN SCHIZOPHRENIA

PATIENTS: NEW RESULTS OF FOLLOW-UP STUDY

Ilya Reznik

, L. Volchek

, M. Reznik

, H. Y. Meltzer

A. Weizman

Laboratory of Biological Psychiatry, Felsenstein Medical

Research Center, Bat-Yam, Israel;

Neurology Department,

Barzilay Medical Center, affiliated to Ben Gurion University,

Ashkelon, Israel;

Psychiatry Division, Barzilay Medical Center,

affiliated to Ben Gurion University, Ashkelon, Israel;

Division of

Psychopharmacology, Vanderbilt University Medical Center,

Nashville, TN, USA

Background: There are multiple causes of hyper-CKemia (HCK)—the

elevation of serum creatine kinase (SCK) activity above upper limits. These

increases may be related to underlying trait-like biological abnormalities in

psychosis, phasic episodes related biological abnormalities, or the effect of

drug treatment. Previously we found, that among HCKemic patients, the

majority was treated with atypical antipsychotic drug (AAPD) and some of

them had neuromuscular complaints. The purposes of this study were to

follow-up prospectively and to estimate the incidence and severity of prob-

able neuromuscular dysfunction in HCKemic schizophrenia and schizoaf-

fective patients. Methods: For this study, we selected 39 HCK-emic

patients suffering from schizophrenia or schizoaffective disorder. Eleven

patients, treated either with clozapine, olanzapine or perphenazine had

persistent/recurrent hyper-CKemia (PHCK). 25 patients had occasional

HCKemia (OHCK). Blood samples for CK determinations were col-

lected regularly up to three years of follow-up. Each year, all patients

were assessed neurologically for possible muscular and peripheral ner-

vous systems involvement. Results: During the study, SCK activity in

PHCK patients was found to be persistently elevated: 400

6 200

(mean

6 SD) in range 250-950 IU/L. Five of these patients had com-

plaints of muscular weakness, early fatigue on exercise, some muscular

pains and cramps. In two of them clinical assessment revealed mild gen-

eral muscular weakness, especially in the proximal parts of the limbs.

This pattern was diagnosed as probable mild myopathy. In majority

of OHCK patients SCK activity was reduced to the normal levels range.

Some patients remained with occasional HCK without any kind of neu-

romuscular pathology. Conclusions: The results of this first long-term

large-scale prospective comparative study indicate that among HCK-

emic schizophrenia and schizoaffective patients, only PHCK pattern

may consider to be a biochemical marker of the neuromuscular dysfunc-

tion. Meanwhile, OHCK is likely to be a benign phenomenon, has no

clinical manifestations, and does not need to be followed-up routinely.

Further investigation of neuromuscular dysfunction, its mechanisms,

pathophysiological significance and possible pharmacogenetics associa-

tions in schizophrenia patients is certainly indicated. The relationship

of HCK to mutations in neuregulin, dysbindin or reelin—genes related

both to schizophrenia and to muscle, is also of interest.

ID: 551193

METABOLIC SIDE EFFECTS OF COMBINED

ANTIPSYCHOTIC TREATMENT: RESULTS FROM

A DOUBLE BLIND TRIAL OF ADJUNCTIVE

RISPERIDONE IN CLOZAPINE TREATED PEOPLE

WITH TREATMENT-RESISTANT SCHIZOPHRENIA

Charles Michael Richardson

, S. Feldman

, D. L. Kelly

M. P. Ball

, D. L. Boggs

, E. Weiner

, R. C. Conley

, J. M. Gold

R. P. McMahon

, R. W. Buchanan

Maryland Psychiatric Research Center, University of Maryland

School of Medicine, Baltimore, MD, USA;

Lilly Pharmaceuticals,

Indianapolis, IN, USA

Incomplete response to clozapine in people with treatment-resistant schizo-

phrenia poses a significant clinical challenge. The use of a second antipsy-

chotic has become widespread despite little evidence to support its efficacy

or safety. This analysis examines the safety of combined risperidone and

clozapine with respect to metabolic and prolactin-related side effects,

during a 16-week, placebo-controlled, randomized trial of adjunctive risperi-

done (4 mg/day) in clozapine partial responders. Eligible subjects were

those with DSM-IV schizophrenia or schizoaffective disorder who contin-

ued to manifest moderate illness severity (18-item BPRS total ^ 45 and

CGI Severity score ^ 4) and persistent psychosis (four BPRS psychosis

item total ^ 8, with one of these items ^ 4)despite adequate prior clozapine

treatment. Weight and blood pressure were measured every two weeks

and laboratory measures (fasting glucose, cholesterol, triglycerides, and

prolactin) were obtained at baseline, week 4, and endpoint. Eighty-six sub-

jects signed consent; 71 entered the 4-week evaluation phase; 65 were

International Congress on Schizophrenia Research

38 3. 3. Drug Side Effects and Physical Illness

randomized and entered the double-blind treatment phase (risperidone: 30;

placebo: 35). The majority of subjects were male (risperidone group: 63.3%;

placebo group: 71.4%) and Caucasian (risperidone group: 76.7%; placebo

group: 62.9%). The risperidone group was significantly older (48.3 (7.2) ver-

sus 43.6 (9.6)). Fifty-two subjects (80%) completed all 16 weeks of treatment

(risperidone: 25/placebo: 27). Adjunctive risperidone was not associated

with weight gain or blood pressure changes compared to placebo. The ris-

peridone group had a significantly greater increase in prolactin levels com-

pared to placebo (P < .001). The presentation will include complete data on

other metabolic parameters. In summary, the addition of risperidone to clo-

zapine was associated with no significant change in weight or metabolic side

effects relative to clozapine treatment alone, although prolactin levels were

significantly elevated in the group receiving adjunctive risperidone. Sup-

ported by R01 MH45074 the Stanley Medical Research Institute and

P30 MH068580. Janssen Pharmaceuticals provided study medication.

ID: 551129

EFFECTS OF ARIPIPRAZOLE ON METABOLIC

MEASURES IN PEDIATRIC AND ADOLESCENT

PATIENTS: A POOLED ANALYSIS OF PLACEBO-

CONTROLLED TRIALS

Andrei Pikalov

, R. A. Baker

, S. Kaplita

, J. Han

, S. Mathew

R. Mankoski

, W. H. Carson

, R. A. Forbes

, M. Nyilas

T. Iwamoto

, R. Owen

Otsuka America Pharmaceutical Inc., Rockville, MD, USA;

Bristol-Myers Squibb, Plainsboro, NJ, USA;

Bristol-Myers

Squibb, Wallingford, CT, USA;

Otsuka America Pharmaceutical

Inc., Rockville, MD, USA;

Otsuka Pharmaceutical Development

and Commercialization Inc., Princeton, NJ, USA;

Otsuka Phar-

maceutical Co., Ltd, Tokyo, Japan

The potential metabolic side effects of antipsychotics in children and ado-

lescents are of particular concern for not only clinicians, but also patients

and their caregivers in situations where an atypical agent may be beneficial.

In the present analysis, the metabolic effects of aripiprazole in pediatric and

adolescent patients from the aripiprazole worldwide development program

were evaluated. Metabolic measures in aripiprazole- and placebo-exposed

patients were compared with ANCOVA using LOCF. Aripiprazole pro-

duced no significant mean change (mg/dL) versus placebo in fasting plasma

glucose (aripiprazole: þ0.3, placebo: 1.2; P = .38), total cholesterol (aripi-

prazole: 5.8, placebo: 8.5; P = .22), or fasting triglycerides (aripiprazole:

2.5, placebo: -0.9, P = .79), whilst the change in body weight was significant

(aripiprazole: þ1.6 kg, n = 381, placebo: þ0.3 kg, n = 187, P < .001).

The results inform evidence-based clinical decisions, and reinforce the

need for realistic management of weight gain and metabolic monitoring

in all children/adolescents being treated with antipsychotic medications.

ID: 551116

BONE MINERAL DENSITY IN PATIENTS WITH

PSYCHOSIS AND NON-PSYCHOTIC SIBLINGS

Machteld Marcelis

, P. Habets

, M. van Kroonenburg

P. Domen

, J. van Os

1,3

Department of Psychiatry and Neuropsychology, Maastricht

University Medical Center, Maastricht, Netherlands;

Department

of Nuclear Medicine, Maastricht University Medical Center,

Maastricht, Netherlands;

Department of Psychological Medicine,

Institute of Psychiatry, London, United Kingdom

Background: Increased rates of altered bone mineral density (BMD), like

osteopenia and osteoporosis, have been reported in schizoprenia, but the

underlying mechanism are not yet fully elucidated. Reduced BMD, as a bi-

ological marker of diminished cumulative estrogen exposure, has been sug-

gested to be a marker of increased psychosis risk. In the present family study

we tested whether decreased BMD is also an indicator of genetic liability for

the disorder. Methods: As part of an observational longitudinal study,

BMD of the spine and femur was measured with dual-energy-X-ray-

absorptiometry (DXA) in a subsample of 42 patients with schizophrenia,

54 non-psychotic siblings of these patients and 39 controls. Preliminary,

multilevel random regression, analyses were performed to assess group dif-

ferences in bone mineral density of the lumbar spine and femur (baseline

measures). Results: In female patients, bone mineral density was signifi-

cantly lower than in controls, in those not using contraceptives. This effect

was found in both the lumbar spine (b = 0.74, P < .024) and femur (b =

0.66, P < .05). Female siblings also showed (trends of) decreased bone

mineral density in the lumbar spine and femur region compared to controls.

In male patients and siblings, bone mineral density was not significantly

altered compared to controls. Conclusions: Both female patients and their

non-psychotic female siblings may have decreased bone mineral density.

This data support not only that reduced BMD, as a biological marker

of diminished cumulative estrogen exposure, may be a marker of increased

psychosis risk, but also that reduced BMD may be a marker of the genetic

liability for schizophrenia.

ID: 551025

A GUIDELINE FOR WEIGHT GAIN PREVENTION —

A PILOT STUDY

Jair Borges Barbosa

, C. Attux

, L. C. Martini

, C. M. Arau

G. H. Kaio

, A. M. Roma

, R. F. Reis

, R. A. Bressan

Psychiatry, UNIFESP, Sao Paulo, Brazil;

Endocrinology,

UNIFESP, Sao Paulo, Brazil

Introduction: Antipsychotic induced weight gain is a fair ordinary and well

known side effect. Cardiovascular disease, glucose intolerance, and type 2

Diabetes Mellitus might be the outcome of this side effect. The weight gain

is more pronounced in the first 4 months of treatment and the psychiatrist

should actively help the patient to avoid it. During the psychiatric visit

clinicians have some difficulties on evaluating and orienting these aspects,

focusing one’s attention into psychopathologic aspects. Our aim was to de-

velop a quick supportive material and a guideline to: 1) optimize the psy-

chiatric visit focusing on detecting, preventing and intervening to avoid

medication induced weight gain; 2) to structure the clinician practice sug-

gesting four levels of intervention according to the severity of problem: diet-

ing, physical activities, binge eating control and pharmacologic treatment.

Methods: Two different materials were developed; first a practical informa-

tive booklet for the patients, with healthy dieting and physical activities

recommendations. Second, a practical guideline suggesting the physician

a few questions about weight gain and appetite increase is used to establish

a dialog with the patient and promote early detection of eating problems.

The guideline provides 4 modules of intervention empowering the physician

to promote dieting and physical activity habits changes, actively treating

binge eating and increased appetite whenever necessary. Results: In our pi-

lot study 5 physicians applied our methods to 10 schizophrenic outpatients,

9 patients finished the four months period. Mean age: 35,8, 80% were male,

mean weight: 97,7 Kg, and mean BMI: 30,9 Kg/m2 at the beginning of the

study; mean weight 95,2 Kg, and mean BMI 30,2 Kg/m2 at the end. One

patient was prescribed with topiramate, due to obesity, one with metformin,

due to glucose intolerance. Discussion: This pilot study showed that the

designed material was useful in clinical practice taking less than 5 minutes.

It helped the clinician to diagnose weight changes and empowered clinicians

to actively implement measures to avoid this trend. Our future goal is test-

ing this material for the weight gain prevention in larger samples at the be-

ginning of antipsychotics use. If the major weight gain occurs during the

first 4 months, we believe this is the best period for weight gain prevention

with life style modifications, and the introduction of medications to treat

glucose intolerance or obesity.

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 39

ID: 550873

THE ACUTE EFFECTS OF ATYPICAL ANTIPSY-

CHOTIC DRUGS ON GLUCOSE SENSITIVITY IN AN

ANIMAL MODEL

Heidi Noel Boyda

, L. Tse

, R. M. Procyshyn

2,1

, A. M. Barr

1,2

Faculty of Medicine, Anesthesiology, Pharmacology and

Therapeutics, University of British Columbia, Vancouver, BC,

Canada;

Faculty of Medicine, Psychiatry, BC Mental Health and

Addictions Research Institute, Vancouver, BC, Canada

The atypical antipsychotics are currently the drugs of choice for psychiatric

conditions including schizophrenia spectrum disorders and bipolar disor-

der. However, recent studies have shown that these compounds contribute

as risk factors for metabolic syndrome. Untoward effects induced by these

drugs, such as weight gain, hyperlipidemia, glucose intolerance and insulin

sensitivity occur within patients and rodents both acutely and on a chronic

time scale. The objective of this study is to examine these metabolic abnor-

malities through the use of a rodent model. To assess the effects of atypical

drugs on glucose sensitivity, adult female Sprague-Dawley rats were treated

with either a low or high dose of clozapine (2, 20 mg/kg) or risperidone

(0.5, 2.5 mg/kg) and compared to vehicle treated animals. Separate cohorts

of rats were subjected to an intraperitoneal glucose tolerance test (IGTT) at

1, 3 and 6 hours after antipsychotic drug treatment. For the IGTT, blood

samples were drawn every 15 min for a two hour period. Blood samples

were also stored for future measurement of drug and metabolite levels

by HPLC-UVD. The statistical analysis indicated that there was an effect

of both drug dose and time since treatment on glucose sensitivity, as mea-

sured by the IGTT. For both clozapine and risperidone, the high dose of

drug caused greater glucose resistance than the low dose and vehicle treated

animals, while the low dose drug caused greater insulin resistance than

the vehicle treated animals. This effect was greatest one hour after drug

administration and diminished over time. These data show that, at least

in an animal model, both an atypical drug with a well known metabolic

side-effect profile (clozapine) and a drug with more modest metabolic

side-effects (risperidone) can cause acute glucose resistance, and these

effects are dose-dependent. Ongoing studies will compare these findings

with additional typical and atypical antipsychotics after both acute and

chronic treatment.

ID: 550865

PERSISTENT ANTIPSYCHOTIC POLYPHARMACY

AND EXCESSIVE DOSING IN THE COMMUNITY

PSYCHIATRIC TREATMENT SETTING

Alasdair M. Barr

, S. McIsaac

,T.Wu

,R.Ko

, J. L. Johnson

A. H. Young

, W. G. Honer

, R. M. Procyshyn

Pharmacology, University of British Columbia, Vancouver, BC,

Canada;

Psychiatry, University of British Columbia, Vancouver,

BC, Canada;

Nursing and Health Behaviour Research Unit,

University of British Columbia, Vancouver, BC, Canada

Introduction: Clinical practice guidelines have been developed to assist

clinicians on the practical aspects of treating patients with schizophrenia,

including treatment resistant patients. Consistent among these references is

the recommendation that treatment resistant patients be trialed with clo-

zapine monotherapy before using strategies with limited or no evidence,

such as concurrent use of two antipsychotics (ie, antipsychotic polyphar-

macy). Antipsychotic polypharmacy practice may increase the risk of ad-

verse events and its benefits on mental health outcome remain largely

unsubstantiated. Methods: We investigated persistent antipsychotic poly-

pharmacy and excessive dosing in patients originally recruited for a study

that examined the relationship of tobacco use in the context of severe and

persistent mental illness. This study targeted adults who were receiving serv-

ices from a community mental health team within the Vancouver Coastal

Health Authority. BC PharmaNet (a province-wide network that that links

all British Columbian pharmacies to a central set of data systems) was used

to screen for eligible patients for this study and obtain data about prescrip-

tions filled over at least a 90 day period. Results: Of the 514 patients

recruited in the initial study, 397 patients met the inclusion criteria; 162 di-

agnosed with schizophrenia, 81 with schizoaffective disorder, 48 with major

depression, 77 with bipolar disorder, and 29 with psychosis not otherwise

specified. The highest incidence of persistent antipsychotic polypharmacy

was found in individuals diagnosed with schizoaffective disorder (32.1%),

followed by schizophrenia (30.9%), psychosis NOS (17.2%), bipolar dis-

order (15.6%) and major depression (14.6%). In the case of the atypical

antipsychotics, quetiapine was the agent associated with the greatest pro-

portion of patients treated with persistent antipsychotic polypharmacy

(46.1%) followed by risperidone (37.1%), olanzapine (32.7%), and then

clozapine (30.3%). Differences were noted between subjects treated by

monotherapy versus polypharmacy on concurrent treatment with alternate

medications. Discussion: The primary finding of this study was that 25% of

our outpatient sample was treated with persistent antipsychotic polyphar-

macy. The relationship of polypharmacy to excessive dosing remains an

important issue with respect to risk for adverse events, clinical efficacy

and economic concerns.

ID: 550830

DISCORDANT CARDIOMETABOLIC RISK OF

ATYPICAL ANTIPSYCHOTICS DESPITE SHARED

WORSENING OF BODY COMPOSITION DURING

FIRST-TIME USE IN CHILDREN AND

ADOLESCENTS

Christoph Ulrich Correll

1,2

, J. M. Kane

1,2

, V. Olshanskiy

B. Napolitano

, A. K. Malhotra

1,2

Psychiatry Research, The Zucker Hillside Hospital, Glen Oaks,

NY, USA;

Psychiatry, Albert Einstein College of Medicine, Bronx,

NY, USA

Background: Cardiometabolic effects of antipsychotics are problematic in

patients with prolonged antipsychotic exposure, but have been insuffi-

ciently studied in antipsychotic-naı

ve and pediatric patients. Methods:

Youths (N = 272), age 4–19 (mean: 13.9

6 3.6) years, prescribed atypical

antipsychotics for the first time, underwent assessments of

weight, fasting glucose, lipids and homeostatic model of insulin resistance

(HOMA-IR) at baseline, week 4, 8 and 12. Refusing/non-adherent patients,

determined by interview and absent antipsychotic blood levels, served as

controls. Results: After 10.3

6 3.1 weeks, weight increased by 9.2 6

0.6lbs = 7.9 6 9.4% with aripiprazole (N = 41), 18.0 6 1.3lbs = 15.5 6

17.9% with olanzapine (N = 45), 12.3 6 1.2lbs = 9.9 6 12.0% with quetiapine

(N = 36), and 10.9 6 0.6lbs = 10.0 6 11.1% with risperidone (N = 135). Con-

trols (N = 15) lost weight minimally (1.5 6 1.6lbs = 0.2 6 1.9%). Weight

gain >7%/>14% occurredin 58.4%/17.1% of patients onaripiprazole, 84.4%/

51.1% on olanzapine, 55.6%/30.6% on quetiapine, 64.4%/25.1% on risperi-

done, and 0% of controls. All metabolic parameters, except HDL-choles-

terol, increased significantly with olanzapine. With quetiapine and

risperidone, triglycerides and triglyceride/HDL-cholesterol ratio increased

significantly. With quetiapine, non-HDL cholesterol also increased signifi-

cantly. Metabolic changes were non-significant with aripiprazole and in con-

trols. New-onset dyslipidemia developed in 7.3%, 28.9%, 8.8%, and 19.4% of

youths on aripiprazole, olanzapine, quetiapine and risperidone, compared to

6.7% of controls (P = .033). In contrast to weight gain and dyslipidemia,

International Congress on Schizophrenia Research

40 3. 3. Drug Side Effects and Physical Illness

acquired insulin resistance (HOMA-IR>4.39: 2.9%–17.8%) and metabolic

syndrome (0%–6.5%) were relatively rare. Conclusions: Short-term, first-

time atypical antipsychotic use in pediatric patients produced significant

weight and triglyceride increases. Olanzapine had the most pervasive, aripi-

prazole had the most circumscript cardiometabolic liability. These results re-

quire careful clinical attention. Findings were more pronounced than

previously reported, highlighting the need to assess antipsychotic-naı

ve sam-

ples.

ID: 554130

WEIGHT MAINTENANCE BY BOOSTER

BEHAVIORAL SESSIONS IN SCHIZOPHRENIA

SUBJECTS WHO LOST WEIGHT IN A BEHAVIORAL

PROGRAM

Rohan Ganguli

1,3

, J. S. Brar

Psychiatry, University of Toronto, Toronto, ON, Canada;

Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA;

Psychiatry, Center for Addiction and Mental Health, Toronto,

ON, Canada

Introduction: Several controlled clinical trials, involving patients with

schizophrenia, have established that almost half of the participants can ben-

efit by losing weight. However, there have been no long term follow up

studies, so far, on whether the loss is maintained. We report the preliminary

two-year follow up results on subjects who participated in an NIMH-

funded clinical trial of weight reduction in schizophrenia. The study is on-

going and hence these data are from the 23 subjects who have reached the

two-year follow up time point. Methods: Subjects in this analysis are those

who participated in a behavioral weight loss intervention and who suc-

ceeded in losing at least 3% of their initial weight. In order to evaluate

the effectiveness of every other week individual booster sessions, subjects

we re-randomized to receive individual booster sessions every other week

for two years or to usual care with follow up weights gathered yearly.

Results: There was some weight regain in both groups, but considerably

less in the booster group, as compared to the controls. The subjects in

the booster group regained 3.1 lb in the first year and an additional

1.7 lb in the second year of follow up. However, even after two years their

mean weight was still 7 pounds below their weight at entry into the behav-

ioral treatment. Controls on the other hand regained an average of 9 lb in

the first year of follow up, already exceeding their average entry weight and

then regained an additional 4.9 lb pounds in the second year of follow up.

Thus, controls were at 5.8 pounds above their weight before entering treat-

ment while subjects in boosters were 7 pounds below their weight at entry

into the study. Conclusions: From these preliminary data, we conclude that

twice monthly booster sessions appear to be effective in preventing weight

regain, following intentional weight loss, for as long as two years following

treatment. It remains to be investigated whether less active intervention,

such as monthly boosters would be as effective, and whether more active

intervention, such as weekly contact would be even more effective. These

results reinforce the proposition that obesity is a chronic disorder and

requires on-going maintenance after successful weight loss.

Table. Weight Change (lb): Initial and Over Two Year-Followup

Randomized

Mean

Study Entry

Weight

Mean Weight

at Start of

Followup

Mean

Weight

at 1 year

Mean

Weight

at 2 years

Booster

Treatments

228.4 217.6 220.7 222.4

Usual Care 237.7 229.1 238.1 243.0

ID: 554356

International Congress on Schizophrenia Research

3. 3. Drug Side Effects and Physical Illness 41

4. 4. Electrophysiology

THE RELATIONSHIP BETWEEN GAMMA BAND

SYNCHRONY AND P300 BREAKS DOWN IN

SCHIZOPHRENIA

Judith M. Ford

, B. J. Roach

, R. E. Hoffman

D. H. Mathalon

Psychiatry, UCSF, San Francisco, CA, USA;

Psychiatry,

Yale University, New Haven, CT, USA

Auditory P300 amplitude reduction in schizophrenia is canonical and may

be explained by poor synchronization or reduced power of the underlying

neural activity. We asked if patients have reduced synchrony and power,

and whether together with P300 amplitude, they make unique or overlap-

ping contributions to the discrimination between patients and controls. We

also asked whether people who have large P300s have higher power and

greater synchrony of neural activity, and if the relationships between

P300 and power and synchrony are different in patients and healthy con-

trols. Methods. We recorded EEG data from 22 controls and 21 patients

with schizophrenia (DSM-IV) while they performed an auditory target (P =

.10) detection task. We used wavelet analyses to estimate total power and

synchrony of delta, theta, alpha, beta, and gamma activity in a 50ms win-

dow around the peak of the P300 to the target from the single trial data. We

measured P300 amplitude from the average of the single trials, in a 50ms

window around its peak. Results and Conclusions. P300 amplitude and

delta and theta synchrony were reduced in patients; delta power and syn-

chrony better distinguished between groups than P300 amplitude. In spite

of the strong correlation between P300 amplitude and gamma synchrony in

controls (r = .79), and in spite of P300 amplitude reduction in the patients,

gamma synchrony was not reduced in the patients. Whatever is contribut-

ing to P300 amplitude reduction in the patients (eg, gray matter volume

deficits, poor clinical state, or fluctuations in attention [Ford, 1999]) is

not significantly affecting gamma synchrony. The breakdown of a lawful

relationship among neurobiological or neuropsychological variables in

schizophrenia patients is not uncommon in the field of biological psychi-

atry. It suggests that the variables are affected by independent factors and

the relationships between them are attenuated by additional pathophysio-

logical processes.

ID: 540424

VISUAL ATTENTIONAL CAPTURE IN

SCHIZOPHRENIA: AN EVENT-RELATED

POTENTIAL STUDY

Maxime Bubrovszky

1,2

, M. Combel

, P. Thomas

1,2

Psychiatry, University Hospital of Lille, Lille, France;

Functional

Neurosciences and Pathology Laboratory, CNRS UMR 8160, Lille,

France

Selection of relevant information is disrupted by schizophrenia. In the vi-

sual modality, selection capacities can be studied by resistance to interfer-

ences: patients’ impairment is characterized by an abnormal persistence of

attentional capture effect when control subjects manage to ignore the same

irrelevant stimuli. We propose here to record event-related potentials

(ERP) during a visual attentional capture paradigm. We hypothesized

that electrophysiological indicator of capture will persist for patients.

We recruited 10 patients with schizophrenia (DSM IV-TR criteria) and

10 healthy controls. Subjects were instructed to detect as quickly as possible

the position of a black square which appears up or down on a screen. Dur-

ing the whole process two red circles were presented on both lateral side of

the screen. Capture was induced by lateral translation of one of the circles.

This translation occurred following randomized order for 48 of the 96 trials.

Behavioural effect was measured by difference of reaction time between the

two conditions: non-capture (square appearance) VS capture (square ap-

pearance and circle translation). Two sets of presentations were proposed.

During the first one, the translations occurred bilaterally (50/50, random-

ized order). During the second one all translations occurred unilaterally for

the right circle. ERP were recorded with 70 electrodes montage. After pre-

processing and averaging, we subtracted capture and non-capture condi-

tions. Relevant time periods and regions were defined when this subtraction

was significantly different from zero. Behavioural results are similar to

previous studies: both patients and controls are ‘‘captured’’ by bilateral

translation. For unilateral translation, patients stay captured. On contrary,

controls recover non-capture reaction time. Two time windows are selected

by electrical capture effect (maximal at frontal area). For bilateral capture,

no difference is detected. For unilateral, the later window (250 to 350 msec.)

differs between the two groups. Subtracted ERP for patients keep showing

capture effect for frontal region and are significantly higher than for

controls. Our results suggest that selectivity impairment for patients

could be linked to incapacity to ignore irrelevant stimulus. Patients persist

to engage important frontal activity to process irrelevant data. This impair-

ment could constitute a key to explore both fundamental perceptual deficits

and alteration of daily functioning.

ID: 550596

MIRROR NEURON FUNCTION IN A FIRST EPISODE

OF PSYCHOSIS POPULATION

Fiza Singh

, J. Pineda

, K. Cadenhead

Psychiatry, University of California at San Diego, La Jolla, CA,

USA;

Cognitive Science, University of California at San Diego,

La Jolla, CA, USA

Schizophrenia is a chronic debilitating illness characterized by the patient’s

inability to integrate into the social milieu, although the neural basis of this

impairment is not well understood. The recent discovery of mirror-neurons

has led to conjecture that the inability of patients with schizophrenia to

express themselves, participate in and be attuned to social cues results

from an inability to simulate the mind of others, thus misinterpreting mean-

ings and goals of social interactions. Mirror neurons fire specifically in re-

sponse to viewing actions of others and facilitate anticipation of social

interaction. Their function can be measured using EEG recordings and cal-

culating mu wave suppression index (MSI) in bilateral sensorimotor corti-

ces as previously shown in autism spectrum disorders. Using this paradigm

we conducted a pilot study with 5 normal controls and 5 subjects with a first

episode (FE) of psychosis. Standard EEG disk electrodes were applied to

the scalp and data recorded while the subject sat in a quiet room and

watched videos on a computer screen. Videos consisted of a moving

hand, two bouncing balls, a ball being passed between people, a cigarette

being taken out of a case, a person pretending to smoke a cigarette, a series

of moving lights that correspond to a moving person and visual white noise.

Preliminary data suggests that at baseline controls have higher MSI in the

right hemisphere compared to FE subjects (0.1 vs. .04 log [condition/base-

line]). In addition, when viewing a moving hand, controls did not show

hemispheric effects, whereas FE subjects showed a deficit in right hemi-

spheric MSI (controls vs. FE, 0.1 vs. .01 log [condition/baseline] respec-

tively). Taken together, the data suggest that MSI, a measure of motor

mirror neuron activity, is altered at baseline and in response to viewing

of actions of others in subjects who have experienced a first episode of psy-

chosis. Interestingly, a specific deficit in the right hemisphere of the senso-

rimotor cortex is consistent with this region’s previously established role in

social cognition. The loss of function in this region may be a consequence of

the psychotic episode or may predate the episode representing a develop-

mental abnormality that confers future risk of developing psychosis. To

investigate this hypothesis further, we plan to test more FE subjects to in-

crease the power of our study and test subjects with a prodromal syndrome

who are at risk of developing psychosis.

ID: 550572

International Congress on Schizophrenia Research

42 4. 4. Electrophysiology

PERIADOLESCENT MATURATION OF DOPAMINE

ACTIONS IN THE PREFRONTAL CORTEX

Patricio O’Donnell

, K.-Y. Tseng

Anatomy and Neurobiology, University of Maryland School of

Medicine, Baltimore, MD, USA;

Pharmacology and Experimental

Therapeutics, Rosalind Franklin University of Medicine and Science,

North Chicago, IL, USA

Several lines of evidence suggest that the prefrontal cortex (PFC) contin-

ues its maturation well into adolescence and beyond. Anatomical studies

from human and non-human primates reveal protracted changes in archi-

tecture and innervation. Dopamine fibers, in particular, as well as dopa-

mine (DA) receptors change their density during that late developmental

stage. In addition, markers of local circuit interneurons have also been

reported as changing during adolescence in monkeys. As the DA modu-

lation of local PFC circuitry may be critical for a number of high-order

functions and could be a primary element in schizophrenia pathophysi-

ology, we explored the changes in responses to DA of pyramidal neurons

and interneurons in the rat PFC with in vivo intracellular recordings and

whole-cell recordings from brain slices. D1 agonists enhance excitability

of pyramidal neurons, and effect more pronounced in slices from adult

rats. Co-activating D1 and NMDA receptors yielded persistent depolari-

zations resembling up states, but only in slices from adult rats. The DA

modulation of PFC interneurons also changed during adolescence. The

D2 agonist quinpirole switched from a negligible effect in slices from in-

fant (ie, PD < 35) rats to a strong increase in excitability in slices from

adult (PD > 55) rats. The data suggest that the ability of interneurons to

modulate pyramidal cell activity via feed-forward inhibition becomes re-

fined in the PFC during adolescence. A disruption in such maturation

could be responsible for the emergence of symptoms in conditions

such as schizophrenia.

ID: 550489

ABNORMAL PERIADOLESCENT MATURATION OF

INTERNEURON FUNCTION IN RATS WITH

A NEONATAL VENTRAL HIPPOCAMPAL LESION

Patricio O’Donnell

, K.-Y. Tseng

Anatomy and Neurobiology, University of Maryland School of

Medicine, Baltimore, MD, USA;

Pharmacology and Experimental

Therapeutics, Rosalind Franklin University of Medicine and Science,

North Chicago, IL, USA

It is clear that a strong developmental component is important for

schizophrenia pathophysiology, and there is converging evidence

from human and animal studies that such developmental component

may affect primarily inhibitory interneurons in several cortical regions.

Such deficit is likely to cause abnormal behaviors in animals and symp-

toms in humans. The time course of interneuron development may

provide some cues to symptom onset, as there are several critical de-

velopmental stages, including a late maturation during adolescence.

Here we explored the physiological properties of neurons in the pre-

frontal cortex (PFC) of adult and preadolescent rats with a neonatal

ventral hippocampal lesion (NVHL) or sham controls. Whole-cell

recordings in slices revealed an increased excitability of PFC pyramidal

neurons and an abnormal maturation of fast-spiking interneurons. D2

agonists fail to excite interneurons, yielding a hyper-excitable and noisy

PFC. Thus, this model reveals an onset of abnormal excitation/

inhibition balance that matches the onset of symptoms in humans;

ie, during late adolescence.

ID: 550475

NEUROPHYSIOLOGICAL MARKERS OF INTER-

NEURON DYSFUNCTION IN (NON-GENETIC)

RODENT DEVELOPMENTAL MODELS OF

SCHIZOPHRENIA

Patricio O’Donnell, A. J. Gruber, M. Roesch,

G. Schoenbaum

Anatomy and Neurobiology, University of Maryland School of

Medicine, Baltimore, MD, USA

Prefrontal cortical interneurons have been reported as abnormal in brains

from schizophrenia patients in several postmortem studies. Many develop-

mental and even pharmacological rodent models show remarkable conver-

gence in altering cortical interneuron function. As high-frequency

oscillations in EEG or field recordings are known to be related to synchro-

nized activity in fast-spiking interneurons, and are affected in schizophre-

nia, it is possible that neurophysiological testing can reveal interneuron

dysfunction in rodent models. Here, we tested for field potential activity

in rats with a neonatal ventral hippocampal lesion (NVHL) during a behav-

ioral session in which they had to choose a reward site based on an odor cue.

During the interval between the odor and the response, the prefrontal cor-

tex (PFC) exhibited an increase in activity in the beta band (1330 Hz) in

control rats, but not in NVHL rats. Combined with our recordings in slices

showing that PFC interneurons fail to mature during adolescence in NVHL

rats, the results indicate that EEG measures can identify abnormal balance

of excitation/inhibition in the PFC of rodent models of schizophrenia with

an approach similar to what has been used in humans.

ID: 550449

MISMATCH NEGATIVITY IS ASSOCIATED WITH

GENETIC, CLINICAL, COGNITIVE, AND

FUNCTIONAL ABNORMALITIES OF SCHIZO-

PHRENIA PATIENTS

Gregory A. Light, T. A. Greenwood, J. Sprock, M. Pela,

A. Rissling, K. S. Cadenhead, N. R. Swerdlow, D. L. Braff

Psychiatry, University of California, San Diego, La Jolla, CA, USA

Schizophrenia patients have widespread deficits ranging from abnormali-

ties in sensory processing to impairments in cognition and daily living. Mis-

match Negativity (MMN) is an EEG waveform that is a probe of the

earliest automatic stages of sensory information processing and can be eli-

cited in the absence of directed attention. To assess the genetic architecture

of MMN, a candidate gene analysis was performed on 203 schizophrenia

patients and 119 controls from the UCSD Schizophrenia Program using the

UCSD/Consortium on the Genetics of Schizophrenia (COGS) custom 1536

SNP chip. The initial single marker analyses revealed significant associa-

tions with 25 of 94 carefully selected schizophrenia-related genes (empirical

P < .01). These genes included BDNF, GRIN3A, DISC1, CTNNA2,

NRG1, and ERBB4. Several of these genes also interact molecularly

(eg, NRG1 and ERBB4). Patients had MMN deficits (P < .01) which

were associated with impaired performance on tests of working memory

(P < .01), verbal recall (P < .01), and negative symptoms (P < .01).

MMN deficits were also associated with reduced performance on a compre-

hensive functional skills assessment battery (P < .05), and significantly (P <

0.01) lower ratings on several measures of functional status (eg, indepen-

dence in living situation, managing finances, Scale of Functioning, Global

Assessment of Functioning Scale). In contrast, MMN was not associated

with performance on other cognitive measures or positive symptoms.

MMN deficits reflect neural dysfunction associated with the core cognitive,

clinical, functional, and even genetic abnormalities of schizophrenia

patients.

ID: 550407

International Congress on Schizophrenia Research

4. 4. Electrophysiology 43

PREPULSE INHIBITION AND P50 SUPPRESSION IN

YOUNG ADOLESCENTS AT RISK FOR PSYCHOSIS

Tim Ziermans, C. Kemner, M. Sprong, M. Magne

e, P. Schothorst,

H. van Engeland

Child and Adolescent Psychiatry, University Medical Center,

Utrecht, Netherlands

In schizophrenia response inhibition of repetitive auditory stimuli is

assumed to be disrupted by insufficient filtering of information. These

kind of sensory gating deficits have also shown to be present in patients

at risk for psychosis. However, the developmental course of these psycho-

physiological vulnerability markers is unclear and the timing of disease re-

lated changes remains unknown. In this study, we investigated auditory P50

suppression and prepulse inhibition (PPI) in a sample of young adolescents

(aged 12–18 years) at ultra high-risk for psychosis (UHR). P50 suppression

and PPI of the startle reflex were measured in UHR patients (n = 64) and

compared to a group of typically developing adolescents (n = 67), matched

for age, sex and IQ. PPI was significantly reduced in the total UHR group,

but did not distinguish between those making a transition to psychosis and

other patients. There were no group differences for P50 suppression. Our

results show that reduced PPI of the startle reflex is already present in

young adolescents at risk for psychosis. This suggests that PPI is a relatively

stable vulnerability marker, while changes in P50 suppression may only be

affected later during the illness course.

References

1. Brockhaus-Dumke A, Schultze-Lutter F, Mueller R, Tendolkar I,

Bechdolf A, Pukrop R, Klosterkotter J, Ruhrmann S. Sensory gating

in schizophrenia: P50 and N100 gating in antipsychotic-free subjects

at risk, first-episode, and chronic patients. Biol Psychiatry.

2008;64(5):376–384.

2. Quednow BB, Frommann I, Berning J, Maier KU, Wagner M. Im-

paired Sensorimotor Gating of the Acoustic Startle Response in the

Prodrome of Schizophrenia. Biol Psychiatry. May 29, 2008.

ID: 550294

CLINICAL AND COGNITIVE FEATURES OF EARLY

ONSET PSYCHOSIS: PILOT DATA FROM AN EARLY

PSYCHOSIS STUDY

Jean Starling

1,2

, A. Harris

2,3

, L. Williams

2,3

, C. Smith

M. Nagy

Psychological Medicine, Children’s Hospital Westmead, Sydney,

NSW, Australia;

Discipline of Psychological Medicine, University

of Sydney, NSW, Australia;

Brain Dynamics Centre, Sydney,

NSW, Australia

Background: Schizophrenia and other psychotic disorders are a major

cause of disability world wide. Much of this burden is due to cognitive

impairments, which predict functional outcome better than acute symp-

toms. The outcome is worse for younger patients, perhaps because they

are more likely to have neurocognitive abnormalities at presentation. Pos-

sible deficits include impaired social cognition and working memory, and

reduced gamma phase synchrony. Aims: To describe the clinical and neuro-

cognitive features of a small pilot cohort with early onset psychosis and to

identify differences in cognitive functioning in comparison with normal

children of the same age. Method: The subjects for this pilot study are four-

teen children and adolescents diagnosed with psychosis and fourteen nor-

mal controls matched for age, gender and years of education. The clinical

group have data from questionnaires on symptoms and premorbid func-

tioning (SCID, PSAS, RFS, PANSS, DASS, CDI, and YMRS). Subjects

and controls have data from the BRID battery of tests, including cognitive

and EEG measures. Results: The mean age at diagnosis was 13 (median 14),

and 64% were female. Fifty seven percent had affective psychosis, 39%

schizophrenia spectrum disorders and 14% Schizoaffective disorder or psy-

chosis NOS. The results of their performance on the BRID battery of tests

compared to controls is still being analysed and will be presented in this

poster.

ID: 550250

P3B AMPLITUDE REDUCTION IS NOT

PROGRESSIVE OVER THE COURSE OF

SCHIZOPHRENIA AFTER THE ONSET OF ILLNESS

Alp Ucok

, Y. K. Ergen

, M. Devrim Ucok

Psychiatry, Istanbul Medical Faculty, Istanbul, Turkey;

Physi-

ology, Istanbul Medical Faculty, Istanbul, Turkey

P3b is an event-related potential that is elicited by attended, task-relevant

stimuli. Auditory P3b amplitude reduction is a widely replicated finding in

schizophrenia. The findings of cross-sectional studies concerning the effects

of duration of illness on P3b processes are contraversial. In this study lon-

gitudinal data of first-episode schizophrenia patients were evaluated to de-

termine whether the P3b abnormality is increased over the course of

schizophrenia, consistent with a progressive pathophysiological process.

Subject groups comprised 11 patients with schizophrenia (10 women)

and 14 age- and education-matched healthy controls (7 women). Schizo-

phrenia patients and healthy controls were tested on two occasions with

mean intertest intervals of 6.2 and 5.7 years, respectively. Schizophrenia

patients were at the post-acute phase of their first-episode during the initial

testing. All patients were receiving atypical antipsychotics during the first

and second evaluations. All patients, but one, met remission criteria in the

period between the two evaluations suggesting that our sample consisted of

patients with better outcome. P3b was elicited during an oddball paradigm

consisted of standard (1000 Hz, probability 0.8) and target stimuli (1500

Hz, probability 0.2). The differences in P3b amplitude and latency between

the two evaluations in patients and controls were assessed with the analysis

of variance repeated measures design. Groups showed the expected parie-

tocentral distribution of P3b (P = .001). P3b amplitudes were reduced in

patients compared to controls (P = .03). P3b amplitude was not different

between the two evaluations. Most importantly, groups did not differ in

P3b amplitude difference between the first and second evaluations. No dif-

ference was found for P3b latency. Our finding of P3b amplitude reduction

in schizophrenia patients is consistent with previous studies. Our results

suggested that P3b reduction is not progressive over the course of schizo-

phrenia after the onset of illness at least in a sample with better outcome

who are receiving atypical antipsychotics. To our knowledge, the present

study is the first to report on the longitudinal course of P3b in schizophrenia

to delineate illness progression.

ID: 550214

MISMATCH NEGATIVITY IN PRODROME, FIRST

EPISODE AND ESTABLISHED SCHIZOPHRENIA:

RELATIONSHIP WITH STIMULUS TYPE,

GENERATOR SOURCES, GREY MATTER LOSS,

AND FUNCTIONAL OUTCOME

Ulrich Schall

1,2

, P. E. Rasser

1,2

, R. Atkinson

1,2

, R. Fulham

1,2

K. Helmbold

, J. Todd

1,2

, S. Halpin

, P. Johnston

, P. T. Michie

1,2

P. B. Ward

, P. M. Thompson

5,1

, V. Carr

1,2

Centre for Brain and Mental Health Research, University of

Newcastle, Newcastle, NSW, Australia;

Cognition and Connec-

tivity Research Panel, Schizophrenia Research Institute, Sydney,

NSW, Australia;

Swinburn University of Technology, Melbourne,

VIC, Australia;

Schizophrenia Research Unit, University of New

South Wales, Sydney, NSW, Australia;

Laboratory of Neuro

Imaging, University of California, Los Angeles, CA, USA

International Congress on Schizophrenia Research

44 4. 4. Electrophysiology

Mismatch negativity (MMN) is an early negative component of an event

related potential and considered a robust biological marker of schizophre-

nia. We recorded MMN in response to duration, frequency and intensity

deviance in 18 schizophrenia subjects and 18 pair-wise age- and gender-

matched healthy subjects. Patients were rank-ordered according to their

socio-occupational function levels. High-resolution structural magnetic

resonance images were acquired to generate average cerebral cortex models

using cortical pattern matching. MMN amplitude was found to be reduced

in schizophrenia patients and correlated with their reduced socio-occupa-

tional function levels. Only in patients, grey matter reduction in cortical

areas subserving auditory processing, motor organization and executive

function was correlated with reduced MMN amplitude in response to fre-

quency deviance. In a separate study, MMN generator source data were

derived from LORETA constrained source localisation performed on

current dipoles that were oriented perpendicular to the cortical surface

in realistic head models (CURRY). Generator sources were largely re-

stricted to the temporal lobes and reduced for duration deviance in young

people with a recent diagnosis of schizophrenia. Our preliminary data from

an ultra high-risk cohort of 72 referrals to a specialized mental health ser-

vice in Newcastle further suggests significantly reduced MMN to duration

deviance for those individuals who subsequently make a transition from

prodrome to schizophrenia. Our findings suggest a close association of

MMN to frequency deviance with cerebral pathology and clinical outcome

in established schizophrenia while reduced MMN to duration deviance

appears to be associated with the prodrome and first-episode phase of

the disorder. Supported by the National Health and Medical Research

Council of Australia and the Hunter Medical Research Institute.

ID: 550210

ELECTROPHYSIOLOGICAL CORRELATES OF

EMOTIONAL EXPERIENCE IN SCHIZOPHRENIA

William Powers Horan

1,4

, J. K. Wynn

1,4

, A. M. Kring

R. F. Simons

, S. M. Crosby

1,4

, P. Nori

1,4

, L. Mancini

1,4

C. Tripp

1,4

, M. F. Green

1,4

Psychiatry and Biobehavioral Sciences, UCLA, Los Angeles, CA,

USA;

Psychology, UC Berkeley, Berkeley, CA, USA;

Psychol-

ogy, University of Delaware, Newark, DE, USA;

Psychiatry, VA

Greater Los Angeles Healthcare System, Los Angeles, CA, USA

Individuals with schizophrenia have repeatedly been found to report nor-

mal levels of pleasure when directly exposed to evocative stimuli. However,

little is known about the neural correlates and time course of emotional

experience in schizophrenia. Following a translational research approach,

this study examined performance on a well-validated affective picture view-

ing task that elicits a characteristic pattern of Event Related Potentials

(ERPs) in healthy individuals that differs between emotional versus non-

emotional images. 38 stabilized outpatients and 37 healthy were shown

a set of 60 standardized pictures of pleasant, unpleasant, and neural images

for 6000 ms each while ERP’s were recorded. Participants also provided

reports of their emotional responses to the pictures. Replicating prior re-

search, the healthy controls self-reported emotional responses that matched

the intended valences of the pictures and they demonstrated significantly

larger P300s and Late Positive Potentials (LPPs; 500–100 ms post-picture

onset) for pleasant and unpleasant pictures compared to neutral pictures.

The patients’ pattern of self-reported emotions and P300s were similar to

those found in controls. However, there was a significant group x valence

interaction for LPPs (F

2,72

= 3.30, P < .05), indicating that patients showed

a smaller difference in LPPs between pleasant versus neutral pictures than

did controls. These findings support the validity of self-reports of normal

emotional responses to evocative stimuli in schizophrenia. They also sug-

gest that individuals with schizophrenia may show disturbances in affective

chronometry: while the similar mid-latency ERP’s (P300s) across groups

suggest intact attentional allocation to emotional stimuli in schizophrenia,

the difference in LPPs suggest a disruption in sustained attentional process-

ing of emotional stimuli.

ID: 550101

ELECTROPHYSIOLOGICAL CORRELATES OF

EMOTION ANTICIPATION IN SCHIZOPHRENIA

Jonathan K. Wynn

1,2

, W. P. Horan

2,1

, R. Simons

, A. Kring

P. Nori

, L. Mancini

, S. Crosby

, C. Tripp

, M. F. Green

MIRECC, VA Greater Los Angeles Healthcare System, Los

Angeles, CA, USA;

Psychiatry and Biobehavioral Sciences, UCLA,

Los Angeles, CA, USA;

Psychology, University of California,

Berkeley, Berkeley, CA, USA;

Psychology, University of

Delaware, Newark, DE, USA

Schizophrenia patients report intact in-the-moment experiences of emo-

tional events. However, initial research based on self-report data suggests

that individuals with schizophrenia show disturbances during the anticipa-

tion of forthcoming emotional events, particularly those involving pleasur-

able experiences. The current study examined the electrophysiological

correlates of emotional anticipation using a cued, reaction-time contingent

emotional picture viewing task. EEG data from 38 schizophrenia outpa-

tients and 34 healthy controls were collected during an emotional anticipa-

tion task. On each trial, participants were first shown one of three visual

cues (þ, , O) for 3000 ms, which indicated whether they would subse-

quently be shown a positive, negative or neutral picture. Participants

then made a button press as quickly as possible after cue-offset, which de-

termined how long the picture was shown (ie, a fast response resulted in

a 5000 ms picture presentation ; a slow response resulted in a 500 ms picture

presentation). A 4000 ms delay interval preceded picture presentation. Two

anticipatory EEG events were analyzed: the 500 ms prior to the button

press (termed Readiness Potential, RP) and the 500 ms prior to the onset

of the picture (termed Stimulus Preceding Negativity, SPN). For the RP,

analyses revealed a significant group main effect, F

1,70

= 7.40, P < .01, with

the RP for the controls being significantly larger than that for the patients (-

9.2 lV versus 5.4 lV). There were no other significant main effects or

interactions. For the SPN, analyses revealed a significant group main effect,

1,70

= 10.57, P < .01, with the SPN for the controls significantly larger than

that for the patients (5.4 lV versus 2.9 lV). There were no other sig-

nificant main effects or interactions. The current electrophysiological find-

ings provide further support for general disturbances in emotional

anticipation in schizophrenia. Emotion anticipation is an important aspect

of adaptive approach and avoidance behavior, and anticipatory dysfunc-

tions may contribute to the hedonic and motivational deficits commonly

seen in schizophrenia.

ID: 549985

ERP’S PATTERN AND PROGNOSTIC IN

SCHIZOPHRENIA

Albert Marie Boxus

, P. Nuss

Psychiatry, ASM, Limoux, France;

Psychiatry, CHU St. Antoine,

Paris, France

Objective: In a previous study*, we showed that clinical improvement with

SGA in schizophrenia is associated both cognitive change and by P300 and

P50 suppression. The aim of this study was to investigate an ERP’S pattern

associated with a good prognostic in schizophrenia all validated by PANNS

and GAF at 8 months. Methods: Fifty-five individuals (32 males and 23

females; mean age = 35.9) meeting DSM-IV criteria for schizophrenia, ad-

mitted for an acute relapse, were included in the study and observed within

eight months period from 2005/10 to 2006/07. They were treated with the

following antipsychotics (mean daily dosage) : aripiprazole (13.75 mg), ris-

ridone (5.06 mg), olanzapine (14.12 mg), amisulpride (1000 mg) and

International Congress on Schizophrenia Research

4. 4. Electrophysiology 45

clozapine (150 mg). Psychotropic drugs were prohibited except for benzo-

diazepines, zolpidem and anticholinergic medications. Clinical and electro-

physiological evaluations were performed before the start of treatment (T1)

and after remission (T2). Psychopathology was measured by the Positive

and Negative Syndrome Scale (PANSS). GAF was evaluated in T2.

Results: All patients are improved in T2. We separated them in 2 groups:

under and over 50% PANSS score reduction. In the ERP’S:the reaction

time P300 reduction, the increase of the amplitude of de P300 and the omis-

sions errors reduction are improved in the same way than the PANSS score

reduction and the increase of the GAF after treatment. Conclusions: Re-

action time P300 reduction, increase of the amplitude of P300 and omis-

sions errors reduction could constitute an ERP’S pattern for a good

prognostic in schizophrenia. These results need to be confirmed by a larger

scale and longer duration study. *The effect of second generation antipsy-

chotic drugs on event related potentials in schizophrenia: a preliminary

study: Albert BOXUS: Poster Nr 448 APA 2007 SAN DIEGO.

ID: 549876

References

1. Salisbury D, Volgmaier M, Seidman L, McCarley R. Topographic

abnormalities of P3 in schizotypal disorder. Biological Psychiatry.

1996;40:165–172..

2. Alteration of event related potentials in siblings discordant for schizo-

phrenia. Schizophrenia Research. 2000;41(2):325–33:.

USING TMS-EVOKED POTENTIALS TO PROBE

NEURAL RESPONSE MECHANISMS IN

SCHIZOPHRENIA

Bruce I. Turetsky, P. Crutchley, M. Bhati

Psychiatry, University of Pennsylvania, Philadelphia, PA, USA

Studies of the neuropathology of schizophrenia have implicated a failure of

appropriate neuronal inhibition and a disturbance of cortical connectivity

as two potential models of the illness. However, direct tests of these mech-

anisms in patients have been limited by the practical constraints of in vivo

clinical research, and empirical support has largely been indirect and infer-

ential. Recent developments in the use of transcranial magnetic stimulation

(TMS) as a neurophysiological probe of cortical activity have suggested

a means to test the functional integrity and responsivity of neural circuits

in a more direct manner. By combining single-pulse and paired-pulse TMS

with simultaneous multi-channel evoked potential recordings, we are able

to assess cortical reactivity and inhibitory and facilitative mechanisms

within localized brain areas. Using lagged cross-correlations of evoked ac-

tivity at different electrode sites, we can also assess neural connectivity

across brain regions. This can be all be done without requiring the subject

to attend to any sensory stimuli. This novel methodology has only been

applied to a limited extent in the study of healthy individuals, and even

less in the study of clinical populations. We will describe this methodology,

including our strategy for eliminating TMS pulse artifacts using Indepen-

dent Components Analysis. Our initial data, applying TMS over the left

prefrontal cortex, suggest the following: 1) Schizophrenia patients exhibit

abnormally sustained responses to single TMS pulses; 2) Control subjects

exhibit both cortical inhibition and facilitation following appropriate

paired-pulse stimulation; 3) Patients exhibit relatively normal cortical in-

hibition, but fail to exhibit cortical facilitation; 4) Control subjects exhibit

strongly correlated activity at homologous sites over the contralateral hemi-

sphere; 5) Patients do not show strongly correlated cross-hemisphere activ-

ity. The dual findings of sustained baseline excitation and failed prepulse

facilitation, in patients, are consistent with a model of tonically increased

glutamatergic tone in localized neuronal assemblies; this focal hyper-excit-

ability is associated with disrupted functional connectivity across brain

regions. These preliminary findings demonstrate that TMS-evoked poten-

tials hold tremendous promise as an in vivo method to investigate funda-

mental aspects of the pathophysiology of schizophrenia.

ID: 54971

SENSORIMOTOR GATING IN FIRST-EPISODE,

ANTIPSYCHOTIC-NAI

VE SCHIZOPHRENIA

PATIENTS: A LONGITUDINAL STUDY

Bodil Aggernaes, B. Oranje, B. Glenthoj

Center for Neuropsychiatric Schizophrenia Research, University

Psychiatric Center Glostrup, Copenhagen, Denmark

Deficits in information processing appear to be core features in the patho-

physiology of schizophrenia. Prepulse inhibition of the startle response (PPI)

is an operational measure of sensorimotor gating. In clinical studies, gener-

ally no effect has been found on PPI in schizophrenia patients treated with

typical antipsychotics, whereas some studies found improved PPI in patients

treated with atypical compounds, while others did not. These inconsistencies

might be due to differences in binding characteristics of the various com-

pounds used, in particular to their affinity for the D2 dopaminergic receptor.

Alternatively, it has been suggested that PPI deficits are stable vulnerability

indicators, insensitive to medicaltreatment. Since most previous studies were

cross-sectional, longitudinal studies on drug-naı

ve first-episode schizophre-

nia patients are warranted, to avoid the possible confounding of antipsy-

chotic medication and course of illness. In the present longitudinal study,

PPI of 29 antipsychotic-naı

ve, first-episode schizophrenia patients was com-

pared to that of 30 age and sex matched healthy controls at baseline, and after

six months of treatment with the atypical antipsychotic quetiapine. Since

a gender effect was found and relatively few females were included, it was

decided to use the male data only. Consistent with literature, the patients

at baseline showed significantly less PPI than the healthy controls. At

follow-up however, PPI from the 11 patients who completed the 6 months

treatment with quetiapine, did not differ significantly from that of the

healthy controls. Since the antipsychotic-naı

ve, first-episode schizophrenia

patients showed significantly less PPI compared to the healthy controls at

baseline, this indicates that PPI deficits are present at an early stage in

the development of schizophrenia. Furthermore, the results suggest that

Results

PANSS

score

reduction

Patients

number

GAF

score

Reaction

time P300

reduction

Increase

Amplitude

P300 lvolt

Latency

P300

reduction

lvolt

False

alarms

reduction

Omission

reduction

errors

P50

suppression

30 to 40% 10 64.8 10.3 0.5 9 1.4 0.3 3

40 to 50% 22 69.3 2 2.5 4.3 0.3 12

50 to 60% 18 71.9 19.6 2.1 6.3 0.5 1.2 0

More than

60%

5 74.2 8.1 5.1 3.3 2.4 2.8 0

International Congress on Schizophrenia Research

46 4. 4. Electrophysiology

PPI deficits in schizophrenia can be normalized by six months of treatment

with quetiapine.

ID: 549589

DEFICITS IN WORKING MEMORY PERFORMAN-

CES IN PEOPLE WITH CLINICAL HIGH-RISK FOR

PSYCHOSIS: ERP EVIDENCE

Su Young Lee

1,2

, K. R. Kim

1,2

, J. Y. Park

2,3

,H.H.Cho

J. S. Park

1,2

, H. K. Koo

, J. I. Kang

2,4

, E. Lee

1,2

,S.K.An

1,2

Psychiatry, Yonsei University College of Medicine, Seoul, South

Korea;

Section of Affect and Neuroscience, Institute of Behavioral

Science in Medicine, Yonsei University College of Medicine, Seoul,

South Korea;

Psychiatry, Gongju National Hospital, Gongju-si,

South Korea;

Psychiatry, Ilsan Hospital, National Health Insur-

ance Corporation, Goyang-si, South Korea

The aim of this study was to investigate whether people with clinical high-

risk for psychosis show more rapid decline of P3 amplitudes with increasing

working memory load of verbal and spatial n-back tasks. People with clin-

ical high-risk for psychosis (CHR), schizophrenia patients (SPR) whose du-

ration of illness is less than 5 years and normal controls (NC) were recruited.

Subjects were asked to perform verbal and spatial 0/1/2-back tasks. Re-

peated measures ANOVA with PROC mixed model were performed. P3

amplitudes of 0-back were larger than those of 1- and 2-back condition

(F = 12.8, P < .001). CHR as well as SPR showed smaller amplitudes of

P3 than that of NC (F = 13.8, P < .001). The amplitudes of P3 in CHR de-

clined more rapidly than those of NC when working memory load is

increased (F = 4.6, P = .001). These findings provide the neurophysiological

evidence that the working memory deficits may be already present in pro-

dromal phase of psychosis. The associations of declined P3 amplitudes of

working memory and psychopathology, social and role functioning will

be discussed.

ID: 549427

DURATION MISMATCH NEGATIVITY IN

PRODROMAL AND FIRST-EPISODE

SCHIZOPHRENIA

Carol Jahchan

, K. S. Cadenhead

, D. L. Braff

, G. A. Light

Joint Doctoral Program in Clinical Psychology, San Diego State

University/University of California–San Diego, San Diego, CA,

USA;

Psychiatry, University of California, San Diego, San Diego,

CA, USA

Background: Deficits in the automatic sensory discrimination as indexed by

mismatch negativity (MMN) are well documented in chronic schizophrenia

patients. The majority of previous studies, however, failed to detect MMN

impairments in the early stages of the illness. In fact, six out of seven studies

did not find a reduced pitch or duration MMN in patients with first-episode

schizophrenia. The only study (Brockhaus-Dumke et al., 2005) that exam-

ined MMN in the schizophrenia prodrome reported a trend for at-risk

patients to exhibit deficits on this paradigm. Methods: Nine at-risk individ-

uals (AR) and 15 patients with first-episode schizophrenia (FE), recruited

as part of the UCSD CARE (Cognitive Assessment and Risk Evaluation)

program, and 39 age-matched normal comparison subjects (NC) under-

went duration-deviant MMN testing (1000 Hz tones, standards: 50

msec, P = .90; deviants: 100 msec, P = .10) in the UCSD Schizophrenia

Program Laboratory. Results: Group differences were present in the am-

plitude of the MMN to duration deviants at fronto-central electrodes

2,60

= 4.60, P = .014). In contrast to previous reports showing no

MMN deficits in FE subjects, significant and large effect size (d = 0.85)

MMN reductions were observed in the FE (M = 2.03, SD = 1.64) relative

to the NC (M = 3.85; SD = 2.12) group. Consistent with Brockhaus-

Dumke et al.’s findings, the AR group’s MMN amplitudes (M = 3.32;

SD = 1.81) were intermediate between those of the NC group and the

FE group, but those differences did not reach statistical significance (d =

0.26). Conclusions: In contrast to previous studies, robust and large effect

size duration-deviant MMN deficits were observed in first-episode patients

with schizophrenia. Individuals at risk for developing schizophrenia

showed small and non-significant reductions relative to age-matched nor-

mal comparison subjects. We will assess MMN in a larger sample of at-risk

individuals and further explore the association between MMN and the de-

mographic, clinical, cognitive, and functional characteristics of the sample.

Future studies are needed to delineate the nature of MMN abnormalities

early in the course of schizophrenia and to clarify whether those deficits

reflect premorbid neuropathology or ongoing disease processes associated

with illness progression. Supported by MH60720, MH79777, NARSAD,

and the VISN-22 Mental Illness, Research, Education, and Clinical Center.

ID: 549205

A NEURAL ENSEMBLE ANALYSIS OF COGNITIVE

FAILURE

Jeremy Seamans

Psychiatry, University of British Columbia, Vancouver, BC, Canada

It is now well established that dysfunction of the prefrontal cortex (PFC) is

a key aspect of the cognitive impairments in schizophrenia. Patients exhibit

significant deficits in many aspects of cognition that rely on the PFC in-

cluding working memory, attention and executive function. Conceivably,

these deficits could occur for a number of reasons or there may be a single

underlying problem in PFC networks that affects multiple processes. As

a first step in addressing this issue we have begun to investigate the network

dynamics associated with the commission of errors on tasks of working

memory and executive function in rats. Based on multiple single-unit

recordings we found that ensembles of PFC neurons tracked each separable

cognitive component of the tasks by entering distinct and coherent activity

states. The organization and coherence of network activity states was

largely absent on trials where rats made numerous errors and the break-

down was especially evident at the choice points of the task but not

when the information was acquired or maintained over a delay. There is

also accumulating evidence that the dopamine (DA) modulation of the

PFC is crucial for optimal cognitive performance and this modulation

may be altered in schizophrenia. The PFC DA system is believed to exhibit

an inverted-U dose response curve such that too little or too much DA

becomes detrimental to performance. In order to examine the effects of arti-

ficially pushing DA levels to different parts of the curve, we injected rats

with either the catecholamine releaser amphetamine or the COMT inhibitor

Tolcapone (Tasmar). While Tolcapone improved working memory perfor-

mance, higher doses of amphetamine impaired it. Recordings of multiple

single-unit activity showed that the higher doses of amphetamine decreased

the ensemble organization making the overall network of recorded neurons

less coherent and synchronized across all task epochs. This was a different

dynamic than what was observed when untreated rats committed errors and

suggested that pushing the DA system to the ‘far end’ of the inverted -U

curve produces large-scale cortical disorganization. The next step is to de-

termine how network dynamics within the PFC change in the context of

popular animal models of schizophrenia.

ID: 548748

DURATION AND PITCH MMNS IN

SCHIZOPHRENIA AND THEIR RESPONSE TO

ACUTE NICOTINE TREATMENT

Louise Dulude

, A. Labelle

, V. *Knott

2,3

Psychology, University of Ottawa, Ottawa, ON, Canada;

Psychiatry, University of Ottawa, Ottawa, ON, Canada;

University of Ottawa Institute of Mental Health Research, Ottawa,

ON, Canada

International Congress on Schizophrenia Research

4. 4. Electrophysiology 47

While prevalence rates of smoking are reported as high as 90% in schizo-

phrenia (SZ), the reasons for increased cigarette use in this disorder are not

well understood. Nicotine is known to enhance cognition, including early

sensory and later attention-dependent processes, which are impaired in SZ.

Mismatch negativity (MMN) is a neurophysiologic index of pre-attentive

acoustic change detection that has been found deficient in SZ and has

shown sensitivity to smoking doses of nicotine. The objective of this study

was to compare MMNs elicited by both tone duration and pitch deviants in

smokers with and without SZ and to examine the acute effects of nicotine

on MMNs in SZ. Participants included 12 stable patients with SZ and 12

non-psychiatric controls, all cigarette smokers who were smoking a mini-

mum of 15 cigarettes per day. Mid-line (Fz, Cz, Pz) event-related potential

(ERP) recordings within both a duration-MMN paradigm with 100 ms

standard tones (95%) and 50 ms deviant tones (5%) and a pitch-MMN par-

adigm with 1000 Hz standard tones (95%) and 1100 Hz deviant tones (5%)

were acquired in controls, who were assessed in one session, and patients,

who were assessed in two sessions involving, double-blind randomized ad-

ministration of placebo (0 mg) and nicotine gum (4 mg). Relative to con-

trols, MMN amplitudes of patients were significantly reduced in both

duration and pitch paradigms. Nicotine in patients did not alter pitch

MMNs but it increased the amplitude of duration MMNs to a level com-

parable to that seen in controls. These findings suggest that nicotine

modulation of aberrant central mechanisms underlying early detection

of auditory changes in SZ may be feature specific and further studies

are required to elucidate its role in relation to smoking behavior in SZ.

ID: 548055

EVENT RELATED POTENTIAL PATTERNS

REFLECTING DYSFUNCTIONAL FACIAL AFFECT

PROCESSING IN SCHIZOPHRENIA PATIENTS

Seug-Hwan Lee

1,2

, E. Kim

Psychiatry, Inje Univ. Ilsan Paik Hospital, Goyang, South Korea;

Psychiatry, Clinical Emotion and Cognition Research Laboratory,

Goyang, South Korea

This study was designed to clarify the difference of facial affect processing

between patients with schizophrenia and normal controls. Three ERP com-

ponents were chosen as an index of three separate stages in the emotion

identification processing: P100 as a basic visual processing, N170 as a facial

structural encoding, and N250 as an facial affect decoding. Forty-two

patients who met the DSM-IV criteria for schizophrenia and 36 normal

controls were recruited. ERPs were recorded using 64 channels while par-

ticipants were presented with pictures of faces showing three different emo-

tional states such as happy, neutral, and fear. Participants were instructed

to press the button when they see the emotional face. Mean amplitude of

P100 response was not different between schizophrenia patients and normal

controls whereas the amplitudes of both N170 and N250 in patients were

significantly smaller compared to normal controls (F

1,76

= 9.615, P = .003

for N170; F

1,76

= 16.854, P < .001 for N250). In within group analysis, the

amplitudes of N170 and N250 of patients were significantly smaller on the

fear condition. Mean peak latency was not different across groups in nei-

ther P100 nor N250. However, the significantly shorter N170 latency was

found in schizophrenia patients compared with normal controls(F

1,76

8.195, P = .005). This study suggests that the basic visual processing is

not different between schizophrenia patients and normal controls. But fa-

cial structural encoding and facial affect decoding are dysfunctional in

schizophrenia patients compared with normal control.

ID: 546712

Reference

1. Wynn JK, Lee J, Horan WP, Green MF. Using event related potentials

to explore stages of facial affect recognition deficits in schizophrenia.

Schizophr Bull. 2008;34:679–687.

EVENT-RELATED SPECTRAL POWER DEFICITS

DURING SEMANTIC PROCESSING IN

SCHIZOPHRENIA

Michael Kiang

, M. Kutas

2,3

, D. L. Braff

, J. L. Hsu

, G. A. Light

Psychiatry and Behavioural Neurosciences, McMaster University,

Hamilton, ON, Canada;

Cognitive Science, University of Cali-

fornia, San Diego, La Jolla, CA, USA;

Neurosciences, University

of California, San Diego, La Jolla, CA, USA;

Psychiatry, Uni-

versity of California, San Diego, La Jolla, CA, USA;

Neurology,

Shin Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan

Background: Schizophrenia patients exhibit abnormal electroencephalo-

graphic (EEG) N400 event-related potential (ERP) responses to semanti-

cally related materials. We investigated whether event-related EEG

spectral power changes during semantic processing are also abnormal in

schizophrenia; and if so, the extent to which they are associated with

N400 abnormalities. Methods: The EEG was recorded from 16 schizophre-

nia patients and 16 normal control participants (NCPs), as they viewed

prime words, followed after 750 ms by related (eg, METAL > STEEL)

or unrelated (DONKEY > PURSE) target words, or nonwords (DRESS

> ZORES). Participants’ task was to indicate whether or not the target was

a word, via delayed button-press. Event-related spectral perturbation

(ERSP) was calculated for primes and targets, from 0–750 ms post-stim-

ulus onset and between 1–50 Hz. Results: Across all participants, words

elicited increased power from 1–15 Hz from 50–200 ms, albeit less so in

patients relative to NCPs for related targets. Across all participants, words

elicited decreased power from 1–20 Hz from 200–750 ms. Between 600–750

ms Hz, power decreases from 1–20 Hz were larger for unrelated versus re-

lated targets, but only in NCPs; in patients, larger power decreases for re-

lated targets correlated with disorganized symptoms. Overall, ERSP

deficits did not correlate with N400 abnormalities in patients. Conclusions:

Schizophrenia patients’ abnormally attenuated early power increases for

related targets may result from deficient use of primes to predict perceptual

features of upcoming related words. In the post-N400 interval, patients

exhibited deficits in modulating power decreases to targets as a function

of relatedness to primes, reflecting the propensity of more disorganized

patients to process related targets more like unrelated ones. ERSP and

N400 ERP abnormalities in patients, however, were not associated, impli-

cating distinct pathophysiological processes.

ID: 541887

DISTINCT NEURAL GENERATORS OF SENSORY

GATING IN SCHIZOPHRENIA: BASIC

MECHANISMS AND ATTENTIONAL EFFECTS

Terrance Williams

1,2

, C. M. Yee

1,2

, K. H. Nuechterlein

1,2

K. L. Subotnik

Psychiatry and Biobehavioral Sciences, University of California at

Los Angeles, Los Angeles, CA, USA;

Psychology, University of

California at Los Angeles, Los Angeles, CA, USA

Attentional and perceptual anomalies observed in schizophrenia have been

associated with underlying sensory gating deficits. Despite progress in iden-

tifying the neural generators contributing to P50 and its suppression,

a prominent measure of sensory gating, the neuroimaging methods utilized

and the results obtained have been quite varied. Through the implementa-

tion of electroencephalographic (EEG) source analysis, the current study

was designed to clarify the neural sources involved in the basic P50 suppres-

sion effect and its attentional modulation during a paired-stimulus design.

Results obtained from healthy individuals, recent-onset schizophrenia

International Congress on Schizophrenia Research

48 4. 4. Electrophysiology

patients, and chronic schizophrenia patients showed involvement of a neu-

ral network, consisting of the hippocampus, thalamus, dorsolateral pre-

frontal cortex (DLPFC) and superior temporal gyrus (STG), in the

generation of P50, its suppression and modulation by attention. P50 sup-

pression in healthy participants was positively correlated with hippocampal

and, to some extent, DLPFC activity. This pattern of relationships did not

extend to schizophrenia patients, such that recent-onset patients showed

a positive correlation between DLPFC activity and the P50 suppression

ratio, and chronic patients exhibited no associations. When healthy partic-

ipants directed voluntary attention to the first stimulus of a pair, thalamic

activity was positively correlated with the P50 suppression ratio whereas no

significant correlations were detected with any of the brain structures in the

patient groups, despite improvements in P50 suppression. When healthy

individuals directed attention to the second stimulus, suppression was

disrupted and associations involving the hippocampus, DLPFC and thal-

amus were no longer evident. Similar to the baseline condition, recent-onset

patients exhibited a positive correlation between DLPFC activity and the

P50 ratio score while chronic patients showed no relationships. The current

results support hippocampal and potentially DLPFC involvement in the

basic P50 suppression effect in healthy individuals. In schizophrenia

patients, the pattern of results is consistent with the presence of hippocam-

pal abnormalities as reflected by P50 deficits beginning with the early stage

of illness, whereas DLPFC involvement appears to deteriorate over the

course of schizophrenia. Results obtained during attentional modulation

of P50 further implicate disturbances in functional connectivity in

schizophrenia.

ID: 550842

COGNITIVE-PERCEPTUAL AND INTERPERSONAL

FEATURES OF SCHIZOTYPY ALIGN WITH

DISTINCT AUDITORY ERP ABNORMALITIES

Jason Karl Johannesen

1,2

, B. F. O’Donnell

3,4

, A. Shekhar

W. P. Hetrick

3,4

Psychiatry, Yale University, New Haven, CT, USA;

Psychology

Service, VA Connecticut Healthcare System, West Haven, CT,

USA;

Psychological and Brain Sciences, Indiana University,

Bloomington, IN, USA;

Psychiatry, Indiana University, Indian-

apolis, IN, USA

Schizotypy is conceptualized as a dimension of liability to schizophrenia,

characterized by similar but milder variants in perceptual, interpersonal,

and cognitive function. These features of schizotypy are detectable in

the general population and, at high levels, may define a latent class who

exhibit neurophysiological traits of the schizophrenia phenotype. This

study tested the hypothesis that individuals high in schizotypy express def-

icits similar to schizophrenia on event-related potential (ERP) measures of

sensory (N100) and attentional (P300) processing. Furthermore, we sought

to clarify how the Cognitive-Perceptual, Interpersonal, and Disorganiza-

tion factors of schizotypy, described by Raine and colleagues (1994), dif-

ferentially relate to these neurophysiological features. Sixty chronic

schizophrenia (SZ) and 69 healthy community volunteers, screened for

Axis I and Axis II disorders, completed the Schizotypal Personality Ques-

tionnaire (SPQ) and a two-tone auditory oddball task. ROC curves were

used to identify the cut point in each SPQ factor score distribution at which

50% of healthy subjects were classified as SZ (high schizotypy) and 50% as

normal (low schizotypy). High and low schizotypy subgroups were then

compared to the SZ sample on N100 (standard tones) and P300 (target

tones) measures by MANCOVA with age and SPQ total entered as cova-

riates. Main group comparisons replicated usual findings of N100 and P300

amplitude reductions and prolonged P300 latency in SZ. Schizotypy sub-

groups classified by Cognitive-Perceptual factor scores differed on N100

amplitude only, with high schizotypy intermediate to, and significantly dif-

ferent from, low schizotypy and SZ. Conversely, subgroups classified by

Interpersonal factor scores differed significantly in P300 amplitude, with

high schizotypy intermediate to low schizotypy and SZ. No appreciable dif-

ferences were observed between high and low schizotypy subgroups clas-

sified by either SPQ total or by Disorganization factor scores. These

results are consistent with previous evidence for auditory ERP abnormal-

ities in schizotypy, which appear to represent an intermediate point on the

continuum from normalcy to schizophrenia. Importantly, results allude

to distinct neurophysiological bases for two domains of schizotypy, with

Cognitive-Perceptual features associated with basic sensory/attentional

processing and Interpersonal features with more complex, context-relevant,

information processing deficits.

ID: 551910

COMPARING ELECTROPHYSIOLOGICAL ENDO-

PHENOTYPES IN SCHIZOPHRENIA AND BIPOLAR

DISORDER FAMILIES

Gunvant K. Thaker

1,2

, C. A. Tamminga

, E. Ivleva

, L. E. Hong

I. Wonodi

, A. Moates

Maryland Psychiatric Research Center, Baltimore, MD, USA;

Psychiatry, UT Southwestern Medical School, Dallas, TX, USA

The use of alternative phenotypes that index specific, heritable neurophys-

iological deficits associated with psychosis risk provide an approach to ex-

plore the extent of etio-pathophysiologic overlap between schizophrenia

(SZ) and bipolar disorder (BD). Several such distinct neurophysiological

deficits have been identified including abnormalities in smooth pursuit

eye movements (SPEM), pre-pulse inhibition (PPI), P50 gating, error

rate in antisaccade task, oculomotor delayed response (ODR), as well

impairments in several cognitive domains. In a set of studies we (1) exam-

ined to what extent these different physiological measures are independent

of each other. This was carried out in a Baltimore sample that included 98

families of schizophrenia and 23 families of healthy control probands.

Results suggest that SPEM, P50 gating, PPI and saccadic eye movement

measures are independent domains of neurophysiological deficits. (2) In

the same sample, measurements of SPEM and P50 gating were refined

and heritability estimates examined. SPEM measure was refined by covertly

stabilizing target image on the fovea thus isolating the predictive response.

We are in the process of analyzing data to obtain heritability estimates of

the refined SPEM measure. In the paired-click paradigm, we deconstructed

the single-trial gating response in to different frequency bands and noted

that the gating of alpha-theta band responses significantly separated SZ

probands and their 1st degree relatives from the healthy control subjects.

The heritability estimates associated with alpha-beta gating in SZ and con-

trol families ranged from 0.49–0.83, which were much higher than the tra-

ditional P50 gating measure. (3) These refined SPEM and sensory gating

measures as well as PPI and saccadic eye movement measures were exam-

ined in another family sample collected in Dallas that included families of

both SZ and BD probands. Preliminary analyses suggest that many of the

neurophysiological impairments observed in SZ families are also present in

BD families.

ID: 551896

ERROR-RELATED NEGATIVITY DURING OBSER-

VATION IN SCHIZOPHRENIA

Monica Constance Mann-Wrobel

1,2

, S. E. Morris

2,3

Psychology, University of Maryland, College park, MD, USA;

Psychology, VISN 5 Mental Illness Research Education and

Clinical Center, Baltimore, MD, USA;

Psychiatry, University of

Maryland, School of Medicine, Baltimore, MD, USA

International Congress on Schizophrenia Research

4. 4. Electrophysiology 49

Deficits in neurocognition and social behavior are well-documented fea-

tures of schizophrenia. There is evidence to suggest that cognitive impair-

ment contributes to problems in social functioning, but direct correlations

between these domains are modest and inconsistent. Social cognition has

received some empirical support as a mediator between these constructs, yet

there have been few investigations to date of the underlying neural pro-

cesses associated with socially-relevant stimuli and behavior. The recent

discovery of the mirror neuron system has provided a useful starting point

for describing brain activity involved in the interpretations of other’s

actions. This study examines whether error-related negativity (ERN), an

event-related potential (ERP) that follows the execution of an error on

a task and that has been consistently found to be reduced in amplitude

among schizophrenia patients, might reflect activity involved in the pro-

cessing of error-related activity of others and whether this activity has rel-

evance for social processing and functioning. In this study, schizophrenia

patients and psychiatrically healthy controls performed the flanker task

and then observed a confederate perform the same task. Participants

also completed measures of neurocognition, social cognition, community

social functioning, and symptomatology. It was hypothesized that 1)

schizophrenia patients would show reduced ERN amplitude compared

to controls while executing and observing the task, 2) ERN activity would

be distinguishable from activity of the primary motor cortex as indexed by

the lateralized readiness potential (LRP), 3) deficits in observational ERNs

would be related to impaired social cognition and social functioning. Pre-

liminary analyses suggest the presence of a robust ERN in control subjects

and reduced differentiation of activity following errors and correct

responses in patients while performing the task. During observation, al-

though there were apparent morphological group differences and a small

differentiation between correct and error trials, no significant group differ-

ences were observed. The relationship of the observation ERN with the

LRP, social cognition, social functioning and symptoms of schizophrenia

and the implications of this research will be discussed.

ID: 551860

EXCITATORY-INHIBITORY BALANCE AND

MYELIN SIGNALING DETERMINE CRITICAL

PERIOD TIMECOURSE

Takao K. Hensch

1,2

, M. Fagiolini

, H. Morishita

Neurology, Children’s Hospital (Harvard Medical School),

Boston, MA, USA;

Molecular and Cellular Biology, Harvard

University, Cambridge, MA, USA

Much of our adult behavior reflects the neural circuits sculpted by experi-

ence in infancy and early childhood. At no other time in life does the sur-

rounding environment so potently shape brain function. An understanding

of how this plasticity waxes and wanes with age carries an impact far be-

yond neuroscience, including education policy, therapeutic approaches to

developmental disorders or strategies for recovery from brain injury in

adulthood. Our research aimed at the interface between cell biology and

neuroscience has achieved the first direct control over critical period timing

in any system (Hensch 2005). Strikingly, manipulation of inhibitory trans-

mission in the neocortex can delay the amblyopic effects of monocular vi-

sual deprivation (by gene-targeted reduction of GABA synthesis) or

accelerate plasticity onset (by cortical infusion of a positive GABA receptor

modulator, diazepam). Remarkably, several lines of evidence now indicate

that a specific, local inhibitory circuit (parvalbumin-positive basket cells)

may drive critical period onset in mouse visual cortex. Downstream of

this GABA trigger lies an extracellular proteolytic cascade (tPA-plasmin)

which mediates the enduring structural reorganization of cortical circuits.

Ultimately, these changes become fixed to consolidate plasticity. Myelin-

related factors and Nogo receptor (NgR) signaling, a major inhibitor of

CNS regeneration following axonal injury, contribute to the inability to

recover from amblyopia in the adult primary visual cortex. Indeed, we

find the NgR 1 knockout mouse to be the first animal model which spon-

taneously recovers from amblyopia simply by reopening the deprived eye.

Microarray profiling and anatomical studies of human post-mortem tissue

now implicate the same mechanisms that regulate critical period onset (par-

valbumin cells) and closure (myelin-NgR signaling) in the visual cortex as

more broad contributors to the etiology of cognitive and psychiatric disor-

ders in adulthood. We consider the unexpected possibility that schizo-

phrenic symptoms may reflect excessive and enduring brain plasticity

due to a failure to close early critical periods.

ID: 551833

THE EFFECT OF COLD PRESSOR STRESS ON P50

GATING: IMPLICATIONS FOR SCHIZOPHRENIA

RESEARCH

Sherrie DeAnna All

1,2

, A. M. Hartzell

, D. Charles

, R. J. Erwin

Department of Psychology, Rosalind Franklin University of

Medicine and Science, North Chicago, IL, USA;

Division of

Schizophrenia Research—UBHC, University of Medicine and

Dentistry of New Jersey, Piscataway, NJ, USA

The P50 auditory sensory gating deficit has received considerable attention

as a candidate endophenotype of schizophrenia and as a target of new drug

development. Furthermore, recent meta-analyses suggest that this deficit

has one of the highest effect sizes for differentiating people with schizophre-

nia from controls; however, these effects have been called into question as

reported effect sizes appear to vary. Furthermore, temporary impairments

in P50 expression and suppression can occur in practically anyone, and such

transient disruptions are thought to result from acute autonomic activa-

tion, which can create further variability in effect sizes. Most studies report

that acute stress leads to S2 suppression impairment in healthy adults.

However, some evidence suggests that S1 amplitude may also be reduced

by stress. The current study aimed to replicate and extend the one previous

study of cold pressor test (CPT) induced disruption of P50 gating in healthy

controls, using a large sample size in an attempt to identify potentially la-

tent S1 effects and to better characterize the relationship between P50

disruption and psychophysiological measures of autonomic arousal and

self-reported pain and negative affect. 40 healthy, non-smoking adults

with no history of psychiatric or neurological disease completed the exper-

iment at a small Midwestern medical school. Auditory evoked-potential,

heart rate, blood pressure, and self-report data were collected across three

experimental conditions: a resting baseline, immediately after a 3-minute

cold-pressor, and after a 25 minute recovery period. Data collection, reduc-

tion and P50 peak-picking methods will be discussed. Results of the current

study did not replicate the previously reported CPT induced changes for the

sample as a whole, a result that may have been related to methodological

differences between the two studies, which will be discussed. However,

when baseline gating was taken into account, S2 amplitudes changed sig-

nificantly across conditions. Change in P50 gating correlated significantly

with increased pain perception but no other self-report or cardiovascular

variables. These results support the need for further characterization of the

impact of stress on P50, as this relationship continues to be complicated and

unpredictable and can serve to further limit the reliability of the P50 mea-

sure in its use as an endophenotypic marker and outcome variable for ther-

apeutic interventions.

ID: 551809

ELECTROPHYSIOLOGICAL EVIDENCE OF AB-

NORMAL AUTOMATIC ACTIVATION OF SEMAN-

TIC NETWORKS IN SCHIZOPHRENIA

Daniel H. Mathalon

1,2

, B. J. Roach

, J. M. Ford

1,2

Psychiatry, University of California, San Francisco, San Francisco,

CA, USA;

Psychiatry Service 116d, San Francisco VA Medical

Center, San Francisco, CA, USA

International Congress on Schizophrenia Research

50 4. 4. Electrophysiology

Abnormal activation of semantic networks characterizes schizophrenia and

can be studied using the N400 event-related potential (ERP). N400 is

elicited by words that are not primed by the preceding context and pro-

vides a direct measure of the neural mechanisms underlying semantic prim-

ing. Semantic priming refers to facilitated semantic processing gained

through pre-exposure to semantic context, which can happen automati-

cally if the interval between the prime and target is very short. We pre-

dicted that, relative to healthy controls, schizophrenia patients would

have (1) overly inclusive semantic networks, reflected in a reduced (ie,

less negative) N400 to relatively unprimed words, and (2) deficits in their

use of semantic context, responding to primed words as if they were

unprimed, reflected in a an abnormally increased (ie, more negative)

N400 to primed words. EEG data were acquired from patients with

DSM-IV schizophrenia (n = 26) and age-matched healthy controls (n =

29) during performance of a picture-word verification (match vs. non-

match) choice-response task. Word targets were presented 325 ms after

a picture prime, which either matched (CAMELà’’camel’’), or did not

match the prime (‘‘InCat’’ = In Category Non-Match: CAMELà’’cow’’;

‘‘OutCat’’ = Out Category Non-Match: CAMELà’’candle’’). For each

subject, N400 was identified as the most negative peak between 300

and 500 ms following onset of unprimed words at electrode Pz. Although

primed words do not normally elicit an N400, voltage associated with

primed words was measured at the OutCat N400 peak latency. N400

was defined as the mean amplitude surrounding the N400 peak (

630

ms), relative to a 100 ms pre-picture baseline, for primed and unprimed

(InCat and OutCat) words. N400 data showed that both patients and con-

trols were sensitive to the difference between primed and unprimed words,

but patients were less sensitive than controls. Similarly, N400 data showed

that both groups were sensitive to the subtler difference between classes of

unprimed words (InCat vs. OutCat), but patients were less sensitive, espe-

cially those with prominent negative symptoms. Surprisingly, there were

no correlations between N400 and the severity of thought disorder or delu-

sions in patients. Thus, abnormal activation of semantic networks in

schizophrenia may not be specifically related to the abnormalities in

thought content and process that would seem, on theoretical grounds,

to be the most dependent on semantic processes.

ID: 551753

BETA AND GAMMA CONTRIBUTIONS TO P50

SUBCOMPONENT ABNORMALITIES IN

SCHIZOPHRENIA: CLINICAL CORRELATES

Aileen Hartzell

, S. Keedy

, S. D. All

, R. E. Gur

, R. J. Erwin

Psychology, Rosalind Franklin University of Medicine and Science,

North Chicago, IL, USA;

Psychiatry, University of Pennsylvania,

Philadelphia, PA, USA

Main: Studies examining impaired sensory gating of the P50 component of

the auditory evoked potential (AEP) in schizophrenia (SZ) typically use

a bandpass of 10–100 Hz. Recent evidence suggests that using this bandpass

groups two distinct frequency bands, beta (12–20 Hz) and gamma (30–50 Hz),

together and excludes important information about beta and gamma con-

tributions, which are theoretically associated with different stages of atten-

tion. Dysfunctional attention in schizophrenia is a prominent component

and serves as a potential vulnerability marker for the disorder. Impaired

attention is associated with negative and/or disorganized clinical symp-

toms. We previously found relationships between these symptoms in

patients and abnormalities in P50 subcomponents using a 10–100 Hz fre-

quency band. This study examined the relationship between P50 subcom-

ponents and clinical measures of SZ symptomatology using separate beta

and gamma filter settings. Participants and Methods: AEP data were col-

lected on 42 healthy adults and 39 SZ patients using a 32-channel recording

array with a paired click protocol (0.5sec ISI, 10sec ITI). Amplitudes de-

rived from a spatial PCA analysis of the data were correlated with clinical

measures of schizophrenia symptomatology. Results: Temporal and spatial

PCA analyses of control data revealed a fronto temporal subcomponent in

the beta analysis and both a midline and later frontotemporal subcompo-

nent in the gamma analysis. For beta, higher suppression P50 amplitudes

were not associated with greater disorganized symptoms and exploratory

analyses indicated an inverse relationship. For gamma, lower amplitudes

were not associated with greater positive symptoms. Conclusions: This

study suggests that using either a beta or a gamma filter does not capture

the relationship between clinical measures and P50 abnormalities in schizo-

phrenia as does the traditional bandpass measure. This suggests that con-

tributions from other frequencies may be most important in indexing P50

abnormalities in schizophrenia.

ID: 551740

GAMMA BAND SYNCHRONY AS AN

ENDOPHENOTYPE FOR SCHIZOPHRENIA

Scott R. Sponheim

1,2

, S. Aviyente

, S. S. Kang

, J. J. Stanwyck

E. M. Bernat

Mental Health, Minneapolis VA Medical Center, Minneapolis,

MN, USA;

Psychiatry, University of Minnesota,

Minneapolis, MN, USA;

Psychology, University of Minnesota,

Minneapolis, MN, USA;

Electrical and Computer

Engineering, Michigan State University, East Lansing, MI, USA

The synchronous activity of cortical regions has been posited as a means by

which cognitive functions are carried out by the brain. Both the power and

phase of high frequency gamma band activity is augmented during the per-

ception of objects and appears to reflect the entrainment of neural

responses during cognitive demands. Individuals with schizophrenia

have exhibited both diminished gamma band power and synchrony during

the viewing of simple objects and contours. We have found that unaffected

individuals with genetic liability for schizophrenia exhibit diminished

gamma band power when identifying simple objects in visual noise. To

determine whether aberrant synchrony of gamma activity during the iden-

tification of objects is also associated with genetic liability for schizophre-

nia we acquired electroencephalograms (EEGs) from schizophrenia

patients, first-degree biological relatives of schizophrenia patients, and

nonpsychiatric control subjects while they viewed objects in visual noise.

Schizophrenia patients and their relatives exhibited reductions in gamma

band power during the identification of target objects. To obtain uniform

resolution phase synchrony across brain regions we applied a novel re-

duced interference complex time-frequency distribution to avoid the

trade-off between time and frequency evident in wavelet analysis. Analyses

revealed diminished phase synchrony to be associated with decremented

gamma power in schizophrenia patients. Additional analyses will test

whether first-degree biological relatives exhibit similarly diminished

gamma band phase synchrony to that of schizophrenia patients, and

whether phase anomalies are associated with susceptibility genes related

to glutamatergic and dopamanergic systems. To date, findings suggest

that deficient neural communication may underlie object identification def-

icits in schizophrenia and that such a deficit may mark genetic liability for

the disorder.

ID: 551730

ELECTROPHYSIOLOGICAL ANALYSES OF

ADOLESCENTS AT RISK FOR SCHIZOPHRENIA

Franc Donkers, N. Lucena, A. Belger

Psychiatry, University of North Carolina at Chapel Hill, Chapel

Hill, NC, USA

Background: Puberty is a critical period for the maturation of the fronto-

limbic and fronto-striate circuits critical for executive function and affective

International Congress on Schizophrenia Research

4. 4. Electrophysiology 51

processing. Puberty also coincides with the emergence of the prodromal

signs of schizophrenia, possibly indicating an association between these

two processes. Methods: Event-related potentials (ERPs) and brain oscil-

latory activity in support of executive attention and affective processing

were compared in 24 healthy control (HC) subjects and 21 individuals

at genetic risk for schizophrenia (GHR) between the ages of 9 and 18 years.

Participants were matched for age, gender, ethnicity, and parental educa-

tion. Electrophysiological recordings were obtained from 32 EEG channels

while participants performed an emotional oddball task, where they iden-

tified rare visual targets while ignoring emotional and neutral distracter pic-

tures. Results: Our preliminary ERP results showed significant differences

between the high-risk and control groups in early sensory components, as

well as in late cognitive components. In addition, brain oscillatory activity

in low-frequency ranges was significantly reduced in the high-risk group.

Conclusion: Inefficient cortical information processing during puberty

may be an early indicator of altered brain function in children at risk

for schizophrenia, and may represent a vulnerability marker for illness on-

set. Longitudinal assessments will further inform about their predictive

value for illness onset in populations at high risk for psychotic illness.

ID: 551662

NEURAL RESPONSE TO IMAGES FEATURING

SYMBOLIC THREAT IN SCHIZOPHRENIA

PATIENTS: ERPS AND EVOKED AND INDUCED

EEG ACTIVITY

Peter Bachman

, K. L. Subotnik

, K. H. Nuechterlein

1,2

C. M. Yee

1,2

Psychology, UCLA, Los Angeles, CA, USA;

Psychiatry and

Biobehavioral Sciences, UCLA, Los Angeles, CA, USA

Investigation of emotional experience and motivated behavior in schizo-

phrenia includes reports of both intact functioning and illness-related ab-

normalities. Many of these abnormalities involve behaviors that are

clinically relevant (eg, display of flat affect, anxiety related to fears of per-

secution), but may prove quite complex with respect to their underlying

cognitive architectures. As a basis for further study of the defensive mo-

tivational system, and more specifically, how the brain processes emotion-

ally unpleasant information within the first second of stimulus perception,

we recorded electroencephalographic (EEG) data while first-episode

schizophrenia patients and healthy comparison subjects viewed neutral

and negative images featuring symbolic representations of threat. Repre-

sentations of threat varied systematically in order to permit examination of

whether particular content categories, ranging from high (eg, attack, mu-

tilation) to low imminent threat (eg, pollution, loss), elicit group-wise dif-

ferences in electrophysiological responses that are not apparent when only

super-ordinate affect categories (ie, negative, neutral) are considered.

Event-related potentials (ERPs) previously shown to be sensitive to affec-

tive intensity of stimuli were derived. Additionally, concomitant segments

of EEG data were subjected to time-frequency transformations to measure

frequency band-specific changes in signal power and phase. Collectively,

these measures reflect large-scale neural activity in primary and secondary

visual cortex, as well as an extended network of regions that are sensitive to

the motivational salience of stimuli and interact strongly with visual cortex.

Initial analyses suggest that both patients and controls show strong mod-

ulation of ERP response amplitude, and EEG power and phase (particu-

larly in the delta/theta and alpha frequency bands) by negative content

category. Group differences appear more subtle and do not appear to

be associated exclusively with any one type of negative image content.

Relationships between ERP and EEG measures and clinical variables,

as well as possibly interacting effects of gender, will also be considered.

Finally, implications of the potential influence of affect and arousal on

more cognitively demanding tasks will be discussed.

ID: 551645

GENETIC ASPECTS OF MOTOR INHIBITION IN

PATIENTS WITH SCHIZOPHRENIA AND THEIR

NON-PSYCHOTIC RELATIVES

Joel W. Nelson

, M. K. Hall

, V. M. Goghari

A. W. MacDonald

, J. J. Stanwyck

, S. R. Sponheim

1,2

Veterans Affairs Medical Center, Minneapolis, MN, USA;

Psychiatry, University of Minnesota, Minneapolis, MN, USA;

Psychology, University of Minnesota, Minneapolis, MN, USA

The basis of deficient cognitive control in schizophrenia has been charac-

terized in terms of impaired motor inhibition. Several investigations have

revealed that individuals with schizophrenia as well as their biological rel-

atives exhibit abnormal striatal activation associated with the inhibition of

motor and eye movements (Vink et al., 2006). In addition, recent work has

identified that neural activity during response inhibition is associated with

the catechol-O-methyltransferase (COMT) gene and presumably the avail-

ability of prefrontal dopamine (Kramer et al., 2007). We carried out

a study to determine whether aberrant neural responses of schizophrenia

patients and their biological relatives during motor inhibition were asso-

ciated with the Val158Met polymorphism of the COMT gene. The sample

consisted of patients with schizophrenia, their first degree relatives, and

healthy controls recruited for a family study based at the Minneapolis

VA Medical Center. We measured behavioral responses and event related

potentials (ERPs) while administering a stop-signal task. Whole blood

specimens were used to gather genetic information and to determine

the COMT genotype on subjects. Consistent with manipulation of re-

sponse demands behavioral data show an increase in response time for

GO trials as expectancy of inhibition (ie, stop signal) increases. To

date, electrophysiology data has been collected on 60 patients, 29 relatives,

and 49 controls. We expect to find aberrant N2 and P3a ERP components

in examining inhibition-related activity in schizophrenia patients and their

relatives. We also predict that the Val158Met polymorphism of the

COMT gene will associate with N2 and P3 abnormalities evident in

the patient and relative groups. If hypotheses are supported, evidence

would be consistent with the presence of anterior cingulate and prefrontal

deficits marking genetic liability for schizophrenia and a role of the

COMT gene and dopamanergic mechanisms in deficient inhibitory con-

trol in the disorder.

ID: 551618

NEURAL CORRELATES OF INHIBITORY MOTOR

CONTROL IN SCHIZOPHRENIA PATIENTS AND

THEIR BIOLOGICAL RELATIVES

Mallory Kate Hall

, J. W. Nelson

, V. M. Goghari

A. W. MacDonald

, J. J. Stanwyck

, S. R. Sponheim

2,4

Veterans Affairs Medical Center, Minneapolis, MN, USA;

Psychology, Veterans Affairs Medical Center, Minneapolis, MN,

USA;

Psychology, University of Minnesota, Twin Cities, Minne-

apolis, MN, USA;

Psychiatry and Psychology, University of

Minnesota, Twin Cities, Minneapolis, MN, USA

Several lines of research indicate that individuals with schizophrenia have

poor inhibitory control. Evidence suggests that deficient inhibition may

contribute to errant eye movements, poor episodic memory, and deficient

cognitive control in the disorder. Functional magnetic resonance imaging

studies of inhibitory control have revealed abnormal striatal and prefrontal

activation in schizophrenia patients and their biological relatives during in-

hibition of motor movements. Nonetheless, little is known about the

time-course of neural anomalies associated with poor inhibitory control

in schizophrenia. To investigate the relationship of neural events to defi-

cient inhibition in schizophrenia we carried out an event-related potential

(ERP) study using a motor response task previously used to examine

International Congress on Schizophrenia Research

52 4. 4. Electrophysiology

schizophrenia patients and their relatives (Vink et al. 2006). In the present

study participants completed a stop signal task that consisted of go only

blocks in which participants responded to every trial and go/stop blocks

where participants were required to inhibit their response to stop trials pre-

sented pseudo-randomly between go trials. Electrophysiological data has

been collected on 60 patients with schizophrenia, 29 first degree biological

relatives, and 49 community controls. We hypothesize that schizophrenia

patients and their relatives will exhibit impaired inhibitory control, and

that these abnormalities will be associated with the amplitude of P3a and

N2 ERP components. These components have been associated with decision

making and early response inhibition processes in other investigations and

appear to represent neural processes underlying response inhibition. A com-

parison will be made across schizophrenia, relative, and control groups con-

trasting ERP amplitudes elicited from go trials during go only blocks with

those during go/stop blocks, to investigate the effects of increased inhibitory

demand on motor control. To date, behavioral data are consistent with

slowed response inhibition associated with schizophrenia.

ID: 551616

THE EFFECTS OF INTRAVENOUS SYNTHETIC

D9-TETRAHYDROCANNABINOL ON P300 AND

N300 EVENT-RELATED POTENTIALS DURING

TARGET AND NOVELTY PROCESSING

Daniel Arvid Wilson

1,2

, A. L. Sumich

2,3

, S. Bhattacharyya

D. Ffytche

1,2

, J. Stone

, J. Nottage

, N. Tunstall

, S. Kapur

P. McGuire

, R. Murray

, P. D. Morrison

Psychological Medicine and Psychiatry, Institute of Psychiatry,

King’s College London, London, United Kingdom;

Clinical

Neuroscience, Institute of Psychiatry, King’s College London,

London, United Kingdom;

Psychology, Institute of Psychiatry,

King’s College London, London, United Kingdom

This study investigates the effect of intravenous synthetic D9-tetrahydro-

cannabinol (THC) on P300 and N300 event-related potentials (ERPs) to

target and novel stimuli in a 3-stimulus auditory oddball paradigm. Can-

nabis use is implicated as a risk factor for psychotic episodes and persistent

psychosis. An acute challenge of THC leads to a state modelling of psy-

chotic symptoms; however, the precise neural mechanisms underlying

the effect of THC on behaviour and cognition are unclear. One of the

most robust neurobiological abnormalities observed in schizophrenia is

a reduction in auditory P300 amplitude. Specifically, reductions in frontal

and left temporal P300 amplitude have been associated with psychotic

symptoms. The current study used ERP methods to test a THC model

of psychosis. A randomised double-blind placebo controlled within subject

design was used to investigate the effect of 1.25 mg intravenous THC rel-

ative to placebo on ERPs in 10 healthy adults. Auditory P300 and N300

amplitudes were measured in response to target and novel stimuli in

a 3-stimulus auditory oddball task. The positive and negative syndrome

scale (PANSS) was used to assess psychiatric symptoms. Data were ana-

lysed using repeated measures analysis of variance (ANOVA). Bilateral

frontal and temporal P300 amplitude in response to targets was reduced

under THC compared to placebo conditions. A more specific right hemi-

sphere effect was seen for novelty P300 amplitude (P3a) and N300. There

were no significant effects for P3a and P3b peak latency, perhaps indicating

individual memory encoding and response strategies. Left temporal P300

effects were not found in the THC condition, indicating their possible spec-

ificity to schizophrenia. There was a trend for an increase in positive symp-

toms during the THC condition. A trend was found for reduced response

time to targets in the THC condition compared to placebo. The current

results support THC as a neuropharmacological model of schizophrenia.

It is hypothesised that THC-mediated P300 amplitude reductions underlie

working memory deficits during THC intoxication. These deficits are

thought to impair the perception of novelty, which affects context updating

and memory storage. These results may translate to psychosis, in that in-

creased endocannabinoid signalling may be responsible for reduced P300

amplitude in psychosis.

ID: 551614

AUDITORY HALLUCINATIONS AND THE

CORTICAL RESPONSE TO AUDITORY

STIMULATION AND SPEECH IN SCHIZOPHRENIA:

AN ELECTROPHYSIOLOGICAL IMAGING

(LORETA) REGION OF INTEREST STUDY

Crystal Villeneuve

1,4

, D. J. Fisher

1,3

, D. K. Shah

1,3

, A. Labelle

V. J. Knott

Neuroelectrophysiology Laboratory, Institute of Mental Health

Research, Ottawa, ON, Canada;

Schizophrenia Program, Institute

of Mental Health Research, Ottawa, ON, Canada;

Department of

Psychology, Carleton University, Ottawa, ON, Canada;

Institute of

Cognitive Science, Carleton University, Ottawa, ON, Canada

Gamma (c) frequencies have been associated with the sensory encoding and

processing of auditory information, with auditory hallucinations (AHs) be-

ing associated with excessive c. This finding may reflect a hyperconnectivity,

which in turn could be associated with excessive processing and impaired

sensory integration. Imaging studies have implicated the superior temporal

cortex (STC), the anterior cingulate cortex (ACC) and the prefrontal cortex

(PFC) in AHs, and it appears that the abnormality in which the left PFC

influences the left STC is a result of the failure of the ACC to modulate their

interactions. Given this background, the purpose of this study was to use

gamma band oscillations (GBOs) to examine inter-regional synchrony in

hallucinating (HP) and non-hallucinating (NP) schizophrenia patients dur-

ing passive recording conditions. We were also interested in examining how

these inter-regional synchronies varied with the increasing of hallucinations.

As a number of studies have suggested that hallucinations worsen in the pres-

ence of speech and especially in the presence of unintelligible speech, the

study also examined GBOs in response to external speech presented during

background noise conditions, and during a silence condition. There were 3

groups of 12 participants: HPs, NPs and healthy controls. EEG recordings

were taken under 3 sound conditions presented randomly, and a separate

recording under silent conditions. The sound conditions consisted of fre-

quent and rare consonant-vowel syllables being presented once every ½ sec-

ond during no-noise or simultaneously with white noise or traffic noise. For

each of these 4 conditions, scalp electrical activity was recorded with a 40

channel system and artifact free activity was submitted to spectral analysis

for computation of c amplitudes, which were plotted across the scalp. Low-

resolution electromagnetic topography (LORETA) analysis was used to es-

timate current source density (CSD) and its distribution in the cortex. CSD

were also calculated for regions of interest (ROI) and these were used to com-

pute coherence values between PFC, STE and ACC regions. These ROI val-

ues were computed for the background noise conditions, as well as for the

discrete (external speech) sounds. Relative to NPs, we expect greater GBO

and greater GBO synchrony in HPs. Furthermore, we expect this pattern to

be more evident in the speech/noise conditions.

ID: 551577

AROUSAL EXPERIENCE IN SCHIZOPHRENIA:

SUBJECTIVE AND PHYSIOLOGICAL

ASSOCIATIONS

Daniel Antonius

, D. Malaspina

, F. Tremeau

1,2

, K. A. Nolan

1,2

Department of Psychiatry, New York University School of Medi-

cine, New York, NY, USA;

Nathan Kline Institute for Psychiatric

Research, New York, NY, USA

Background: Since Bleuler it has been widely acknowledged that disturbed

affective arousal is a primary characteristic of schizophrenia. However,

International Congress on Schizophrenia Research

4. 4. Electrophysiology 53

despite a large number of studies on affective (emotional) experience, rel-

atively little research has focused on the experience of arousal in schizophre-

nia and findings are inconsistent. Moreover, it remains unclear whether

underlying physiological responsivity in schizophrenia individuals (SZ) is

in concordance with subjective experiences of arousal, and how this com-

pares to non-schizophrenic individuals. Method: Skin conductance (SC)

was assessed in 20 medicated SZ and 30 control subjects (CS) during ex-

posure to picture (n = 40), sound (40), and printed word (40) stimuli

that varied in level of arousal and valence connotation. Results: Subjective

ratings indicated a similar response pattern in SZ and CS, but SZ demon-

strated an attenuated subjective arousal experience; very calm stimuli were

experienced as less calm, and very arousing stimuli as less arousing. For

physiological findings, SZ exhibited overall reduced SC responses to all

stimuli. Despite this, SZ and CS showed a similar SC response pattern

to stimuli rated very calm and moderately arousing; lower SC responsivity

to very calm stimuli. A marked difference between groups was evident for

stimuli rated as very arousing. As expected, CS exhibited increased SC

responsivity to all stimuli rated as very arousing. SZ showed a decrease

in SC responsivity to very arousing stimuli based on normative ratings;

lowest SC responsivity to stimuli rated as very arousing that also had a nor-

mative rating of very arousing. Background: Our findings suggest a phys-

iological deficit in arousal experience in schizophrenia, which may be

particularly related to highly arousing stimuli. Other schizophrenia studies

have demonstrated deficits or attenuated responses in neurophysiological

processing of stimuli of high emotional valence (fear), and the findings pre-

sented here suggest that similar anomalies may exist for highly arousing

stimuli (stressors, threats). This dissociation between physiological and

subjective arousal experiences may contribute to the restricted affective

arousal often associated with schizophrenia.

ID: 551576

EFFECTS OF ATTENTION ON ERP COMPONENTS

FOLLOWING P50 SUPPRESSION IN SCHIZOPHRE-

NIA PATIENTS

Maria Jalbrzikowski

, T. J. Williams

1,2

, P. Bachman

, W. Lau

K. L. Subotnik

, K. H. Nuechterlein

1,2

, C. M. Yee

1,2

1. Psychology, University of California, Los Angeles, Los Angeles,

CA, USA;

Psychiatry and Biobehavioral Sciences, University of

California, Los Angeles, Los Angeles, CA, USA

Aberrations in attention and sensory perception have long been associated

with schizophrenia. It has been proposed that these attentional deficits may

be due to an inability to filter out extraneous sensory information, resulting

in an inability to process and integrate sensory stimuli from the environ-

ment. A deficit in this filtering process, as indexed by the P50 component

of the event-related potential (ERP), has been demonstrated in patients

with schizophrenia across different phases of the illness. Recent research

suggests that disruption in P50 suppression in patients with schizophrenia

may be associated with abnormalities in subsequent neurocognitive pro-

cesses (Boutros, Korzyukov, Jansen, Feingold, and Bell, 2004). The first

phase of the present study was designed to elucidate whether disruption

of P50 suppression influences later ERP components in patients with

schizophrenia. Ratio measures of successive components (eg, N100,

P200) were examined to determine if suppression is also diminished at these

later components. To establish whether suppression of these components

may be related to one another, the second phase of this study involved

single-trial analysis in healthy participants and patients with schizophrenia.

The third phase of this study evaluated the effects of directing voluntary

attention during a P50 paired-stimulus paradigm. The P50 ratio in schizo-

phrenia has been shown to be transiently enhanced through the manipula-

tion of attention to the first stimulus in a pair and diminished by directing

voluntary attention to the second stimulus (Yee et al., in submission). Pre-

liminary results indicate that in comparison to healthy participants, schizo-

phrenia patients showed increased ratio values at later ERP components,

paralleling results finding disruption of P50 suppression. Additionally, it

appears that directing attention to the first stimulus may not necessarily

enhance suppression at these later components in either sample, but that

voluntary attention directed to the second click resulted in reduced suppres-

sion of later ERP components, with the strongest effects observed in

healthy comparison subjects. These findings suggest that although gating

abnormalities extend beyond the P50 component, it may not be possible to

sustain attentional or top-down modulation in schizophrenia beyond a very

early stage of auditory processing, consistent with abnormalities in fronto-

temporal connectivity (Mathalon, Heinks, and Ford, 2004).

ID: 551515

INCREASING VISUAL STIMULUS COMPLEXITY

AND EVENT-RELATED POTENTIALS IN

SCHIZOPHRENIC PATIENTS

nter Franz Heinz, A. Schneider, E. Hu

ttermann, M. Rubly

Psychiatry and Psychotherapy, Saarland University Hospital,

Homburg, Germany

Disturbances of global information processing as well as disturbances of

specialized networks are fundamental in schizophrenia. The aim of this

study was the comparative examination in healthy subjects and schizo-

phrenic patients of different ERP-generating networks by gradually in-

creasing the complexity of a visual stimulus. We examined 19 patients

(6f, 13m) and 15 (7f, 8m) healthy subjects, matched for age and education.

ERP were elicited by black and white patterns (21x16 squares) of increasing

complexity: i) regular checkerboard ( = simple stimulus), ii) irregular ran-

dom pattern ( = complex pattern stimulus), and iii) random pattern with

embedded letter ( = complex semantic stimulus). The number of black/white

squares was 1:1 for all stimuli. All stimuli had the same task relevance (but-

ton press). ERP were analysed for 2000ms (500ms pre-trigger) from

Fz,Cz,Pz, and from C3/C4 and P3/P4. Bandpass was 0.02 to 30 Hz, sam-

pling frequency 256 Hz. ERP were evaluated by comparisons of the curves

(Blair and Karniski) and of the area under the curves (AUC). In both

groups, all visual stimuli generated positive ERP between 300–500 ms,

larger in healthy subjects (8lV—10lV) than in patients (5l V—6lV), poten-

tials remaining positive with about 2lV till the end of analysis time(1500

ms). In healthy subjects only, the ERP of both complex stimuli differed

significantly from the simple stimulus by a distinct negative component

of about 4lV between 400–600 ms (N500). The ERP of the complex se-

mantic stimulus differed significantly from the complex pattern stimulus by

a late positive component between 900–1500 ms. Both effects were not

found in schizophrenic patients. Stimuli of same modality and task rele-

vance but different complexity generate different ERP, pronounced in

healthy subjects only. The introduction of irregularity (simple to complex

pattern stimulus) seems to generate an informational content, which seems

to activate an information-related ERP generator, leading to the N500

component. The further introduction of a semantic content seems to

activate a further neuronal processor, resulting in a prolonged positive

potential after 600 ms. Both of these effects are missed in schiophrenic

patients. This reveals a distributed disturbance of information processing

in schizophrenia, affecting multiple neuronal networks.

ID: 551510

COMT AND HIPPOCAMPAL FUNCTION: THE COMT

INHIBITOR TOLCAPONE MODULATES FIELD EX-

CITATORY POSTSYNAPTIC POTENTIALS IN CA1

SCHAFFER COLLATERAL AND PERFORANT

PATHWAYS

Elizabeth Tunbridge

, P. J. Harrison

, O. Paulsen

Department Psychiatry, University of Oxford, Oxford, United

Kingdom;

Department of Physiology, Human Anatomy and

Genetics, University of Oxford, Oxford, United Kingdom

International Congress on Schizophrenia Research

54 4. 4. Electrophysiology

Catechol-O-methyltransferase (COMT) metabolises catecholamines and

has been shown to modulate prefrontal dopamine levels. Inhibition of

COMT improves PFC-dependent task performance and a functional poly-

morphism in the COMT gene (Val158Met) has been linked with perfor-

mance on a range of cognitive tasks. COMT has therefore attracted

attention as a possible therapeutic target for treating cognitive dysfunction

in schizophrenia. More recently, evidence has emerged for a role for COMT

outside of the PFC: the Val158Met polymorphism has been associated with

hippocampal activation during emotional processing, consistent with the

high expression of COMT mRNA in this region. Therefore, we investigated

the effect of the COMT inhibitor tolcapone on CA1 Schaffer collateral and

perforant pathway field excitatory postsynaptic potentials (fEPSPs) in hip-

pocampal slices. Hippocampal slices (350 lm thickness) were prepared

from male Hooded Lister rats aged between P35–P42, transferred to an in-

terface chamber and superfused continuously with artificial cerebrospinal

fluid (aCSF). After resting for two hours, stimulating electrodes were

placed in the stratum radiatum and the stratum-lacunosum moleculare

of CA1, to evoke fEPSPs via the Schaffer collateral and perforant path-

ways, respectively. Field potentials were monitored via a recording elec-

trode placed in the stratum radiatum of CA1. Tolcapone was perfused

in aCSF once fEPSP slopes had stabilised. Between 30 and 60 minutes

post-perfusion, 20 lm tolcapone dramatically suppressed fEPSP slope in

both the Schaffer collateral and the perforant pathways (by approximately

60% after 60 minutes). Investigation of the effects of other doses of tolca-

pone, and their reversibility by dopamine receptor antagonists, are ongo-

ing. We demonstrate a striking effect of the COMT inhibitor tolcapone on

both Schaffer collateral and perforant pathway field potentials. These pre-

liminary data support emerging evidence of a role for COMT in modulating

hippocampal function, in addition to its importance in PFC. Our research is

currently investigating the extent to which tolcapone’s effects on Schaffer

collateral and perforant pathway fEPSPs is dopamine-dependent and the

dose-dependency of this effect. Such studies are critical, given the marked

non-linearity of dopamine’s effects on cortical function, and the potential

roles of COMT in both the pathophysiology and pharmacotherapy of

schizophrenia.

ID: 551470

THE EFFECT OF INTRAVENOUS SYNTHETIC THC

ON EEG OSCILLATIONS DURING A SELF-PACED

MOTOR TASK

Judith Nottage, D. Ffytche, S. Bhattacharyya, J. Stone, A. Sumich,

N. Tunstall, D. Wilson, P. McGuire, S. Kapur, P. Morrison

Institute of Psychiatry, King’s College London, London, United

Kingdom

Electroencephalogram (EEG) oscillations have been shown to be altered in

schizophrenia and it has been hypothesised that deficits in synchrony may be

involved in the disease process. It has previously been reported that in a self-

paced motor task inter-trial coherence at gamma frequencies is reduced in

subjects with schizophrenia. An acute challenge of D9-tetrahydrocannabinol

(THC) can trigger psychotic symptoms, and cannabis use is a risk factor for

development of schizophrenia. The purpose of this study was to investigate

the effect of acute THC on the synchrony and amplitude modulation of EEG

oscillations during a self-paced motor task. A randomised, double-blind,

within subject, design was used with 10 healthy adults. THC intravenous

(1.25 mg), or placebo, was administered prior to participants completing

a self-paced button pressing task. EEG was recorded using a 64 channel elec-

trode cap based on the 10–20 system. Wavelet analysis was used to determine

amplitude, spatial synchrony and inter-trial coherence of the Mu and gamma

rhythms at frontal, central and parietal electrode sites. EEG oscillations were

found to be significantly altered in the drug state compared to placebo. Ab-

errant neural oscillations may be related to the positive psychotic symptoms

and changes in consciousness produced by THC.

ID: 551326

ACTIVATION OF CB1-R DISRUPTS SENSORY

GATING AND NEURONAL SYNCHRONIZATION

IN RATS—RELEVANCE TO SCHIZOPHRENIA

Mihaly Hajos

Neuroscience, Pfizer Global Research and Development, Groton,

CT, USA

Impaired auditory gating and abnormal neuronal synchrony are considered

as indicators of dysfunctional information processing in schizophrenic

patients, and possible underlying mechanisms of their impaired sensory

and cognitive functions. Since cannabinoid receptors and endocannabi-

noids have been linked to psychiatric disorders, including schizophrenia,

the aims of the present experiments were to evaluate the effects of canna-

binoid-1 (CB1) receptor activation on sensory gating and neuronal oscil-

lations in rats. In our experiment, auditory sensory gating has been

recorded from the hippocampus (CA3) and entorhinal cortex in anaesthe-

tized rats. Neuronal network activity was recorded from the hippocampus

(CA1, CA3), medial septum, entorhinal cortex and medial prefrontal cor-

tex. Firing rate and oscillatory activity of septal single units were monitored

simultaneously with hippocampal and cortical field potential oscillations in

anaesthetized rats. In the auditory gating experiment, event-related spectral

perturbations and inter-trial coherence analysis were applied to hippo-

pcampal and cortical recordings. In freely-moving rats, hippocampal

and cortical oscillations were monitored in home cage and in rats exposed

to novel environment. Effects of systemic administration of the selective

CB1 receptor agonist CP-55940 were evaluated on these parameters.

Thus, CP-55940 significantly disrupted auditory gating both in the hippo-

campus and entorhinal cortex via interfering with event-related network

oscillations in anaesthetized rats. Spontaneous theta field potential oscilla-

tions were disrupted in the hippocampus and entorhinal cortex, with simul-

taneous interruption of theta-band oscillations of septal neurons.

Administration of the CB1 receptor antagonist AM-251 reversed both

the agonist-induced gating deficit and the diminished oscillations. In

freely-moving rats, CP-55940 significantly reduced theta and gamma power

in the hippocampus, whereas in the entorhinal cortex only gamma power

was attenuated. However, novelty-induced theta, beta and gamma activities

were significantly diminished by CP-55940 both in the hippocampus and

entorhinal cortex. Our data indicate that activation of CB1 receptors inter-

feres with neuronal network oscillations and impairs sensory gating func-

tion in the limbic circuitry, further supporting the connection between

cannabis abuse and increased susceptibility of developing schizophrenia

spectrum disorders.

ID: 551293

THE STABILITY OF PREPULSE INHIBITION OF THE

STARTLE REFLEX IN SCHIZOPHRENIA: A 5 YEAR

FOLLOW-UP STUDY OF ANTIPSYCHOTIC NAI

VE,

FIRST-EPISODE SCHIZOPHRENIA PATIENT

PRELIMINARY RESULTS

Trine Bjorg Hammer, B. Oranje, B. Glenthoj

University Psyciatric Center Glostrup, Center for Neuropsychiatric

Schizophrenia Research, Glostrup, Denmark

Deficits in information processing appear to be core features in the path-

ogenesis of schizophrenia. Prepulse inhibition of the startle response (PPI)

is an operational measure of sensorimotor gating. PPI deficits are generally

regarded as endophenotypes for schizophrenia. However, the stability of

PPI both over time, and during the course of antipsychotic treatment, is

still largely unclear. In an earlier study, we reported that neither a three

months treatment with zuclopenthixol (typical antipsychotic) nor with ris-

peridone (atypical antipsychotic) improved the PPI deficits of antipsychotic

naı

ve, first-episode patients with schizophrenia. The current study is

International Congress on Schizophrenia Research

4. 4. Electrophysiology 55

a 5 year follow-up investigation of that original study. At baseline, 25 drug-

naive first-episode schizophrenic patients and 23 healthy controls matched

for gender and age, participated in the project. Three PPI measures were

examined at baseline, after 3 month of randomized antipsychotic treatment

(risperidone or zuclopenthixol) and after 5 years follow-up. Currently, 18

patients and 16 healthy controls were reexamined at the 5 year follow-up.

Patients showed PPI deficits compared to healthy controls at baseline. At

the 5 years follow-up no significant group differences were found although

only one of the PPI measures improved significantly in the patients. The

current preliminary results support the notion that PPI deficits are funda-

mental trait markers of schizophrenia that are already present at an early

stage in the development of the disease. However, the deficits seem to di-

minish over time, most probably due to the disease process it self, or to the

effects of antipsychotic treatment.

ID: 551250

TIME-FREQUENCY ANALYSIS OF PREATTEN-

TIONAL COGNITION IN SCHIZOPHRENIA

PATIENTS IN TAIWAN

Jung Lung Hsu

1,2

, M. H. Hsieh

1,3

, C. M. Liu

3,4

, M. J. Chiu

3,5

H. G. Hwu

3,4

, F. S. Jaw

Institute of Biomedical Engineering, National Taiwan University,,

Taipei, Taiwan;

Neurology, Shin Kong WHS Memorial Hospital,

Taipei, Taiwan;

Department of Psychiatry, National Taiwan

University Hospital, Taipei, Taiwan, Taipei, Taiwan;

Department

of Psychiatry, National Taiwan University College of Medicine,

Taipei, Taiwan, Taipei, Taiwan;

Department of Neurology, Na-

tional Taiwan University Hospital, Taipei, Taiwan, Taipei, Taiwan

Disturbances of auditory information processing have repeatedly been

shown in schizophrenia. Event-related potentials are currently utilized to

study information processing under the influence of neuropsychiatric dis-

orders and are candidate endophenotypes of schizophrenia. This study tried

to focus on event-related potential components such as P50 and mismatch

negativity (MMN) between control and schizophrenia groups in Taiwan. In

addition, this study also compared the time-frequency analysis and aver-

aged event-related potentials of these event-related potentials. In 50 schizo-

phrenia patients and 50 controls EEG was recorded using an auditory

paired click paradigm for P50 and duration MMN (500msec SOA) for

MMN. Participants are seated in a comfortable recliner in a sound-atten-

uating, electrically shielded booth and instructed to relax with his/her eyes

open and to focus on a fixation point (P50 session) or the video monitor

(MMN session). The stimuli are generated by and data is recorded by Neu-

roscan STIM and ACQUIRE system. Electrodes are used at up to 36 re-

cording sites. Auditory stimuli are presented to subjects binaurally via foam

insert earphones. Written informed consent were all obtained from the

above subjects. In addition to the effect of clozapine versus other antipsy-

chotics was confirmed, the relationship of psychopathology dimensions

were also explored. In time-frequency analysis of single trial MMN, schizo-

phrenia have lower inter-trial phase coherence than controls. The induced

gamma activity did not show large difference between controls and

patients. The results support the view that deficits at sensory preattentive

information processing are essential in schizophrenia patients. These def-

icits may also play a role in the basic integrated neural network activity.

ID: 551095

SOURCE LOCALIZATION OF SENSORY GATING:

CONCURRENT ASSESSMENT OF EEG AND FMRI,

PRELIMINARY RESULTS

Nikolaj Bak

, B. Y. Glenthøj

, E. Rostrup

, H. Larsson

B. Oranje

Center for Neuropsychiatric Schizophrenia Research (CNSR),

University of Copenhagen, Psychiatric Center Glostrup, Copenha-

gen, Denmark;

Functional Imaging Unit, Department of Clinical

Physiology and Nuclear Medicine, Glostrup Hospital, Copenhagen,

Denmark

Reduced sensory gating appears to be among the core features in schizo-

phrenia. The sources of sensory gating however are largely unknown. The

aim of the current study is to identify these sources. To our knowledge this

is the first study in which concurrent EEG and fMRI assessment of P50

suppression was performed. Twenty healthy male volunteers were tested

with identical paradigms in two separate sessions: an EEG setting, and

an EEG concurrent with fMRI setting. Instead of the classical P50 suppres-

sion paradigm an especially constructed paradigm was used for the fMRI

environment, in which the auditory stimuli were replaced by weak electrical

stimuli. Two interstimulus intervals (ISIs) were used: 500 and 1000 ms. The

preliminary results from the 16 subjects who were analyzed in the EEG set-

ting and 9 from the fMRI setting so far showed significant P50 suppression

only in the 500 ms interval, not in the 1000 ms ISI. The contrast between the

1000 ms and 500 ms ISI revealed activation in the temporal region based on

the results from the 9 subjects who were analyzed so far in the fMRI setting.

The results indicate that the EEG data between the two settings are com-

patible. Furthermore the electrical P50 suppression paradigm gives results

that are consistent with classical auditory paradigms. The fMRI results sug-

gest that the temporal cortex is involved in P50 suppression.

ID: 551072

SOPRES (STUDY ON THE PSYCHOPATHOLOGICAL

PROGRESS OF THE PRE-PSYCHOTIC STATE) IN

TAIWAN: EVENT-RELATED POTENTIAL

FINDINGS

Ming H. Hsieh

1,2

,C.C.Liu

, C. M. Liu

2,3

, M. J. Chiu

H. G. Hwu

2,3

, F. S. Jaw

Institute of Biomedical Engineering, National Taiwan University,

Taipei, Taiwan;

Department of Psychiatry, NTU Hospital, Taipei,

Taiwan;

Department of Psychiatry, National Taiwan University

College of Medicine, Taipei, Taiwan;

Department of Neurology,

NTU Hospital, Taipei, Taiwan

A prospective study Study On the psychopathological progress of the PRE-

psychotic State, SOPRES was initiated in Taiwan and sponsored by Taiwan

National Health Research Institute since 2006. Four psychopathological

stages (First episode psychosis; Very high risk stage; Intermediate risk

stage; Very early stage) and normal control were recruited for comparison

of P50/N100/MMN results. Ninety-six prodrome patients and fifty controls

were recruited. Participants are seated in a comfortable recliner in a sound-

attenuating, electrically shielded booth and instructed to relax with his/her

eyes open and to focus on a fixation point (P50 and N100 session) or the

video monitor (MMN session). The stimuli are generated by and data is

recorded by Neuroscan STIM and ACQUIRE system. Electrodes are

used at up to 36 recording sites. Auditory stimuli are presented to subjects

binaurally via foam insert earphones.P50 and N100 suppression were

assessed with an auditory double-click paradigm, while MMN were given

with standard (90%, 50 msec duration) and deviant (10%, 100 msec dura-

tion) tones presented in pseudorandom order and 500msec SOA. Certained

P50/N100 suppression indices and MMN differed between groups. Among

the analyzed P50 parameters, the P50 suppression ratio reached significant

differences between healthy controls and very high risk stage patients, and

between healthy controls and the first episode psychosis subgroup. The

results support the view that deficits of sensory preattentive information

processing may be essential in prodromal patients, especially those with

higher clinical risk.

ID: 551038

International Congress on Schizophrenia Research

56 4. 4. Electrophysiology

EVIDENCE FOR NEURODEGENERATION?

FURTHER REDUCTION IN P300 AMPLITUDE

OBSERVED OVER 3 YEAR FOLLOW-UP IN FIRST

EPISODE SCHIZOPHRENIA

Anthony W. F. Harris

1,2

, K. Brown

, J. Brennan

L. M. Williams

1,2

Psychological Medicine, University of Sydney, Sydney, NSW,

Australia;

Brain Dynamics Centre, Westmead Millenium Institute,

Sydney, NSW, Australia;

Department of Child and Adolescent

Psychiatry, Sydney Children’s Hospital, Sydney, NSW, Australia

A reduction in the amplitude of the P300 waveform in the auditory oddball

paradigm in schizophrenia is a well replicated finding. However very few

studies have examined the stability of this finding over time and none in

a group of patients with their first episode of schizophrenia. This study fol-

lowed up 25 subject with their first episode of schizophrenia (male = 17,

female = 8) over a 2–3 year period as part of a larger program of research.

The subjects with schizophrenia were noted to have decreased P300 ampli-

tude at their initial recording when compared to normal controls. This re-

duction in P300 amplitude increased over the 2–3 year follow-up despite

a reduction in symptomatology. This finding was not explained by changes

in overall medication burden. This study suggests that the reduction in P300

amplitude continues to develop over the early stages of schizophrenia. This

is at odds with other work that has observed stability in the P300 compo-

nent over similar periods of time in subjects with chronic schizophrenia.

One possible explanation for this finding is that the P300 amplitude changes

reflect an ongoing pathological process that has stabilised with chronic

illness.

ID: 551016

TREATMENT RESISTANT P300 ABNORMALITIES

IN A LARGE GROUP OF ANTIPSYCHOTIC NAI

VE,

FIRST-EPISODE PATIENTS WITH SCHIZOPHRENIA

Bob Oranje, B. Aggernaes, B. Y. Glenthoj

CNSR, University Psychiatric Center Glostrup, Glostrup, Denmark

Evidence is accumulating that cognitive deficits form core features in

schizophrenia. It has been suggested that treatment with atypical antipsy-

chotics can ameliorate these deficits. However, studies have often been

confounded by patients either being medicated or chronically ill, making

it hard to differentiate between medication effects and progress of the

disease. In addition they frequently suffer from low subject populations,

giving rise to power issues. In the present study the influence of a six

months treatment period with quetiapine (atypical antipsychotic) was in-

vestigated on psychophysiological parameters of selective attention in

a large group of first episode, antipsychotic naı

ve schizophrenia patients

and age and gender matched healthy controls. Thirty-four antipsychotic

naı

ve patients with first-episode schizophrenia and 40 age and sex

matched healthy controls were tested in a selective attention paradigm

at baseline and at 6 months follow-up. The patients were treated with

quetiapine during the period between baseline and follow-up, the controls

received no treatment. Both at baseline and at follow-up, the patients

showed highly significant reduced P300 amplitudes compared to the

healthy controls. No treatment effects were found. The results indicate

that deficits in P300 amplitude, and thus attention deficits, are present

at an early stage in the development of schizophrenia. Furthermore,

the results indicate that a 6 months treatment period with quetiapine

does not ameliorate these attention deficits. The results are consistent

with the concept that P300 amplitude deficits represent stable endophe-

notypes for schizophrenia.

ID: 550938

EFFECTS OF WORD FREQUENCY ON SEMANTIC

MEMORY IN SCHIZOPHRENIA

Ruth Condray

1,2

, S. R. Steinhauer

1,2

, G. J. Siegle

1,2

M. S. Keshavan

1,3

, D. M. Montrose

, B. Chakraborty

Department of Psychiatry, University of Pittsburgh School of

Medicine, Pittsburgh, PA, USA;

Medical Research Service, VA

Pittsburgh Healthcare System, Pittsburgh, PA, USA;

Department

of Psychiatry, Wayne State University, Detroit, MI, USA

Semantic memory deficit may represent a biological marker for schizo-

phrenia. Findings based on the N400 component of the event-related

brain potential, which reflects on-line access and integration processes,

indicate patients are compromised in discriminating the semantic features

of words, and that executive function may play a key role. Typical an-

tipsychotic medication (haloperidol) appears to improve, but not normal-

ize, patients’ N400 priming effect, which is consistent with moderating

vulnerability-trait status. This hypothesis requires examination based

on newer generation medications. In the present study, N400 was elicited

with a semantic-priming paradigm designed to engage executive processes

(slow presentation rate, resource-demanding task). N400 was recorded in

14 schizophrenia outpatients (M = 34.3 yrs) and 14 healthy controls (M =

31.4 yrs) while they performed a delayed letter probe task that was in-

cidental to the linguistic characteristics of word pairs, which varied in

semantic association (related, unrelated) and word frequency (very low

to very high). Prime-target stimulus-onset asynchrony = 1 s. At testing,

patients were receiving clinician’s-choice atypical antipsychotics; patients

and controls were group-matched for age, gender, and demographics.

The N400 priming effect differed between groups (P < .01), with only

controls showing the typical pattern of greater N400 negativity for un-

associated than associated words. N400 to word frequency also differed

between groups (P < .05); controls’ N400 became more negative with

decreasing word frequency (P < .05) while patients failed to discriminate

word frequency (P = n.s.). Cumulative findings showing N400 priming

disturbance in patients receiving both traditional (typical) and newer gen-

eration (atypical) antipsychotics are consistent with the hypothesis that

semantic memory deficit represents a moderating trait for schizophrenia.

Results also support the assumed importance of executive processes for

semantic memory function in schizophrenia, and implicate a failure to

engage brain processes associated with the access and integration of

stored information.

ID: 550924

EFFECTS OF CHRONIC EXPOSURE TO CANNABI-

NOIDS ON SENSORY GATING AND NEURAL

SYNCHRONIZATION IN HUMANS.

Patrick David Skosnik, C. R. Edwards, G. P. Krishnan,

W. P. Hetrick, B. F. O’Donnell

Psychological and Brain Sciences, Indiana University, Bloomington,

IN, USA

Recent research has shown that schizophrenia (SZ) may be associated

with abnormalities of the endogenous cannabinoid system. This line of

evidence is in accord with numerous longitudinal studies, which have dem-

onstrated increased rates of psychosis in long-term cannabis users. How-

ever, it remains unclear how chronic exposure to cannabinoids is related

to the emergence of SZ-like perceptual/cognitive features. Given that can-

nabinoid receptors (CB1) are thought to modulate neural oscillations nec-

essary for sensory and cognitive processes, it has been postulated that

a possible neurophysiological link between SZ symptomatology and can-

nabis’ neurobehavioral effects involves alterations in neural synchrony.

The current set of studies therefore assessed the effect of chronic cannabis

use on neural synchronization (using EEG) during paired-click auditory

International Congress on Schizophrenia Research

4. 4. Electrophysiology 57

P50 gating and broadband (5–50 Hz) auditory steady-state responses

(ASSRs), both tasks which have been consistently shown to be aberrant

in SZ. As expected, it was found that cannabis users exhibited increased

SZ spectrum traits (schizotypy) compared to drug-naive controls, and that

overall levels of cannabis chronicity (total years of use) was positively cor-

related with schizotypy scores. In terms of auditory P50 gating, cannabis

users demonstrated abnormal gating in a pattern similar to that previously

reported in SZ. Moreover, time-frequency analyses revealed that the can-

nabis-related gating deficit observed at click S2 was accompanied by al-

tered theta and gamma oscillations. Similarly, the effect of chronic

cannabinoids on the ASSR revealed a selective decrease in spectral power

at 5 Hz (theta-range) and 40 Hz (gamma-range). Taken together, these

frequency-specific deficits in the generation and maintenance of theta

and gamma-range neural synchrony are in concurrence with recent

work in animals, demonstrating a general role of the cannabinoid system

in mediating GABAergic network oscillations. Further, these data indi-

cate that chronic cannabis use, mostly likely via CB1 downregulation/de-

sensitization, induces psychotomimetic features through dysregulation of

the GABA system.

ID: 550913

HERITABILITY OF ACOUSTIC STARTLE,

PREPULSE INHIBITION AND ONSET LATENCY

IN SCHIZOPHRENIA AND CONTROL FAMILIES

Erica Joan Duncan

1,2

, W. Hasenkamp

, M. P. Epstein

A. Green

2,1

, L. Wilcox

, W. Boshoven

2,1

, B. Lewison

1,2

Mental Health Service, Atlanta VA, Decatur, GA, USA;

Psychiatry and Behavioral Sciences, Emory University School of

Medicine, Atlanta, GA, USA;

Human Genetics, Emory University

School of Medicine, Atlanta, GA, USA

Background: Patients with schizophrenia have difficulty screening out ir-

relevant stimuli, and often have the experience of sensory flooding. These

‘‘gating deficits’’ may contribute to the thought disorder, cognitive frag-

mentation, and hallucinations that are so debilitating to these individuals.

The acoustic startle response is a reflex contraction of the skeletal muscles

in response to a sudden acoustic stimulus. The modulation of this reflex by

a preliminary nonstartling stimulus is termed prepulse inhibition of acous-

tic startle (PPI), a paradigm used as an operational measure of sensorimo-

tor gating. Many patients with schizophrenia have impaired PPI, and

several lines of evidence suggest that PPI may represent a heritable endo-

phenotype in this disease. The purpose of this study was to examine her-

itability of acoustic startle measures in a sample of schizophrenia and

control families. Methods: We examined baseline startle, habituation,

PPI, and onset latency of the startle response in 40 schizophrenia patients

(SCZ), 58 of their first-degree relatives (SCZ-FAM), and 100 healthy con-

trols (CONT) from 45 families. The startle session consisted of pulse alone

trials to evaluate baseline startle, and prepulseþpulse trials with interstim-

ulus intervals of 30, 60 or 120 ms between the prepulse and pulse. Herita-

bility analyses were conducted using a variance-component approach that

adjusted for age, gender, race and smoking status. We also examined differ-

ences in startle measures between subject groups. Results: We found a sig-

nificant heritability of 45% for PPI at the 60-ms interval, and 67% for

baseline startle. Onset latency heritability estimates ranged between 39%

and 90% across trial types. There was no significant heritability of star-

tle habituation. We did not find differences between CONT and either

SCZ or their families for PPI, baseline startle, or habituation. Onset

latency was longer in SCZ and SCZ-FAM than CONT. Conclusions:

PPI at the 60-ms interval, baseline startle, and onset latency were sig-

nificantly heritable in this sample of schizophrenia and control families.

No PPI impairments were detected in SCZ or SCZ-FAM compared to

CONT. Latency was longer (ie, slower) in SCZ and SCZ-FAM than in

CONT. Our data indicate that assessing family members of schizophre-

nia patients in a between group design may not be sufficient to detect

the heritability of putative endophenotypes. Supported by VA Merit Re-

view Grant (E Duncan).

ID: 550870

ALTERED D2 MODULATION OF PREFRONTAL

CORTICAL INTERNEURONS IN DEVELOPMENTAL

ANIMAL MODELS OF SCHIZOPHRENIA

Patricio O’Donnell

, C. Feleder

, K.-Y. Tseng

Anatomy and Neurobiology, University of Maryland School of

Medicine, Baltimore, MD, USA;

Pharmacceutical Sciences,

Albany College of Pharmacy, Albany, NY, USA;

Pharmacology

and Experimental Therapeutics, Rosalind Franklin University of

Medicine and Science, North Chicago, IL, USA

We have recently shown that rats with a neonatal ventral hippocampal

lesion (NVHL) exhibit an abnormal periadolescent maturation of pre-

frontal cortical interneurons. Specifically, activation of D2 dopamine

receptors becomes strongly excitatory on parvalbumin GABA interneur-

ons during adolescence in naive rats, and this maturation does not occur

in rats that wit a NVHL. Adult NVHL rats also display sensorimotor

gating deficits and other behavioral anomalies similar to those observed

with prenatal maternal infection. We tested whether an immune chal-

lenge localized to the hippocampus would cause similar deficits with in-

tra-hippocampal injections of the bacterial endotoxin LPS during

postnatal day 7–8. Adult rats with LPS, but not sham treatment

exhibited deficits in prepulse inhibition of the acoustic startle response

and increased levels on cytokines in several brain regions. Whole-cell

recordings from fast-spiking interneurons revealed that slices obtained

from sham treated animals exhibited the normal D2-mediated increase

in excitability, whereas similar recordings in slices from LPS-treated

rats failed to yield that response. Thus, the data show that activation

of immune responses in the ventral hippocampus can yield anomalies

similar to what a lesion or prenatal immune challenge may cause, sug-

gesting that the hippocampal region may be a critical target of maternal

infection.

ID: 551940

THE EFFECT OF ATTENTIONAL TASK ON

PREATTENTIVE ERP MEASURES OF SENSORY

PROCESSING IN SCHIZOPHRENIA PATIENTS

Sung-Hyouk Park

1,2

, D. L. Braff

, G. A. Light

Department of Psychiatry, University of California, San Diego,

CA, USA;

Department of Psychiatry, Seoul National Hospital,

Seoul, South Korea

Mismatch Negativity (MMN) and P3a are considered preattentive and au-

tomatic event-related potential components because they can be elicited in

the absence of directed attention. Deficits on these measures reflect abnor-

malities in the earliest stages of sensory information processing. The degree

of automaticity of MMN and P3a measures in nonpsychiatric subjects and

schizophrenia patients, however, has been a topic of debate. The aim of the

present study was to assess the role of attentional task manipulations on

MMN and P3a in nonpsychiatric comparison subjects and schizophrenia

patients. Standard (P = .84, 1000 hz, 50 ms), duration deviant (DD, P = .08,

1000 Hz, 100 ms), and white noise deviant (WND; P = .08, white noise,

50 ms) stimuli were presented with a fixed SOA of 500 ms across 4 identical

International Congress on Schizophrenia Research

58 4. 4. Electrophysiology

paradigms that varied in attentional task. During ERP recording, subjects

1) passively watched a silent video (Video), 2) performed the identical-pairs

CPT (CPT), or were instructed to press a button when they heard either the

3) DD (Auditory Target Detection-DD; ATD-DD) or 4) WND (Auditory

Target Detection-WND; ATD-WND). 20 schizophrenia patients and 20

control subjects participated. The patients had significant reductions in

both MMN (F

1,33

= 5.47, P < .05) and P3a (F

1,33

=12.62, P < .001) across

conditions. No significant group X condition interactions were present for

MMN or P3a amplitudes or latencies. Attentional (condition) effects were

observed for both MMN (F

1,33

= 90.39, P < .001) and P3a (F

2,32

=2.58, P =

.09) where amplitudes in Video and ATD conditions were larger than

in CPT conditions. A significant effect of condition on MMN (F

2,32

20.36, P < .001) and P3a (F

2,32

= 13.29, P < .001) latency was present

with longer latencies observed in the ATD condition. The results of the

present study indicate that there are subtle effects of attention on ERP com-

ponents that can be elicited in the absence of directed attention (ie, ‘‘pre-

attentive’’). Across all tasks and conditions, schizophrenia patients had

robust MMN and P3a deficits in response to task irrelevant stimuli.

ID: 555045

IMPACT OF NEUROCHEMICAL MANIPULATION

ON SENSORY GATING IN HEALTHY SUBJECTS

WITH LOW GATING LEVELS

Philipp A. Csomor, F. X. Vollenweider

Clinical Research, Psychiatric University Hospital Zurich, Zurich,

Switzerland

Deficient early information processing has been considered a central fea-

ture of schizophrenia spectrum disorders. A fundamental feature of infor-

mation processing is the ability gate extraneous stimuli and to attend to

salient features of the environment. Two operational measures of gating

are prepulse inhibition (PPI) and suppression of the P50 event-related po-

tential (P50 suppression). PPI refers to the attenuation of the startle re-

action elicited by an intense pulse stimulus when its presentation is

preceded by a weak prepulse. Similarly, in P50 suppression the first stim-

ulus not only produces an auditory evoked potential (AEP) but also acti-

vates gating, resulting in a suppression of the P50 AEP to the second

stimulus. Patients with schizophrenia exhibit deficits in PPI and P50 sup-

pression. Since PPI and P50 can be induced in healthy volunteers, patients

and rodents, and many findings related to the neural regulation of gating

in animal studies have been supported by studies in humans, these para-

digms represent excellent tools for translational research. These gating

measures provide a unique opportunity to characterize the neurochemical

basis of information processing and might be useful for the discovery of

novel compounds with antipsychotic properties. We have developed

a model to investigate the possible effects of antipsychotics on PPI

and P50 suppression in healthy volunteers rather than in patients. Study-

ing healthy volunteers has the potential to overcome confounding effects

of previous medication exposure in patients, the wide range in severity of

psychopathology and the generally non-random allocation of patients to

treatment regimens. We previously found that the antipsychotic clozapine

increased PPI in subjects exhibiting low baseline sensorimotor gating. In

contrast, the antipsychotic haloperidol did not increase PPI in subjects

exhibiting low baseline gating levels. Furthermore, haloperidol increased

P50 suppression in those subjects with low P50 suppression levels, and

reduced P50 gating in individuals with high P50 gating levels. The poster

summarizes the results of further psychoactive compounds, including ari-

piprazole, risperidone and amisulpride which have been tested on our

model of gating in healthy humans exhibiting low baseline gating. The

data elucidate the applicability of our translational model to serve as a use-

ful tool for the assessment of the efficacy of novel pharmacotreatment

strategies for patients with schizophrenia.

ID: 554152

MATURATION OF TASK-RELATED NEURAL

SYNCHRONY AND ITS RELEVANCE FOR THE

DEVELOPMENT AND PATHOPHYSIOLOGY OF

SCHIZOPHRENIA

Peter J. Uhlhaas

1,2

, F. Roux

, C. Tillmann

, E. Rodriguez

M. Wibral

, F. M. Leweke

, W. Singer

Department Neurophysiology, Max-Planck Institute for Brain

Research, Frankfurt, Germany;

Department of Psychiatry, Johann

Wolfgang Goethe University, Frankfurt, Germany;

Laboratorio de

Neurociencias, Pontificia Universidad Cato

lica de Chile, Santiago,

Chile;

Department of Psychiatry, University of Cologne, Cologne,

Germany;

MEG-Lab, Johann Wolfgang Goethe University,

Frankfurt, Germany

Schizophrenia is associated with an onset of symptoms during late adoles-

cence and early adulthood, suggesting that aberrant brain maturation dur-

ing this period may be involved in the expression of psychosis. In the current

study, we examined task-related neural synchrony during human develop-

ment as well as its impairment in patients with schizophrenia with electro-

and magnetoencephalography (EEG, MEG) to examine the developmental

role of synchronized, oscillatory activity in the pathophysiology of schizo-

phrenia. Development of neural synchrony was investigated during percep-

tual integration in healthy children, adolescents and adult participants (N =

68) between 6–21 years. Perceptual integration was assessed with Mooney

faces which consist of degraded pictures of human faces where all shades of

grey are removed. EEG-recordings were analysed for spectral power as well

as for phase-synchronisation of induced oscillations. Behavioral data

showed that improved detection rates and reaction times with age were ac-

companied by pronounced increases in spectral power and phase-syn-

chrony in the theta-, beta- and gamma-band. This development

occurred in two distinct phases the transition being characterized by

a marked reorganization of network topology and reduction of neural syn-

chrony during adolescence. Following late adolescence, we observed signif-

icant increases of spectral power and phase-synchronization in the theta-,

beta- and gamma-band during early adulthood. EEG/MEG experiments

using the same experimental paradigm show that patients with first-episode

and chronic schizophrenia are specifically impaired in indexes of neural

synchrony that undergo important changes during adolescence. Thus,

patients with schizophrenia are characterised by reduced high-frequency

gamma-band activity and decreased long-range synchronisation in the

theta- and beta-band during perceptual organisation. These data suggest

close relations between the expression and the increase of temporal

precision of synchronous, oscillatory activity, the reorganization and mat-

uration of functional networks. Specifically, we propose that the pro-

nounced changes in neural synchrony during adolescence reflect

a critical developmental period that is associated with a major rearrange-

ment of functional networks that may be related to the development of

schizophrenia during the transition from adolescence to adulthood.

ID: 552124

MULTIBLOCK AUDITORY APPROACH IN

SCHIZOPHRENIA PATIENTS

Murat Ozgoren

, A. Oniz

, S. Taslica

, A. Aktener

, B. Akdede

K. Alptekin

Department of Biophysics, Dokuz Eylul University, Faculty of

Medicine, Izmir, Turkey;

Department of Psychiatry, Dokuz Eylul

University, Faculty of Medicine, Izmir, Turkey

The schizophrenia research area is one of the most challenging ones among

the other pathologies, as the disorder requires advanced and combined

efforts to verify onset and progress. This validates any method to tackle

this problem with parameters including cost effectiveness, clinical relative

International Congress on Schizophrenia Research

4. 4. Electrophysiology 59

ease of application, test-retest etc. Accordingly, a multi-level auditory bat-

tery is being developed in our laboratory with different features. The presen-

tation will provide the resultsfrom the same individuals mainly in three major

test divisions: (A) the modified optimal MMN, (B) the modified DL and (C)

simple auditory event related potentials. The modified MMN was applied

while watching a black and white silent movie. The stimuli consisted of du-

ration, intensity, frequency and location deviants applied in a pseudo-ran-

dom design.The modified DL paradigm consisted of consonant vowels (/ba/,

/da/, /ga/, /ka/, /pa/ and /ta/) presented through a headphone diotically or

dichotically. The simple AERP consisted of stimuli of sinusoidal waveforms

presented binaurally, where the target occurred in 20 percent of cases (1500

Hz standard, 1600 Hz target). The EEG recording was done using a 64-chan-

nel Neuroscan amplifier and the Embedded Microcontroller based Interac-

tive Stimulus Unit (EMISU). The subject group was formed of 13 patients

(age 23–54 years, mean 37.0) with schizophrenia (DSM-IV criteria) and 13

healthycontrol subjects (age 21–50years, mean 34.2).The preliminary results

were as follows: (A) duration and frequency deviations to be diminished in

schizophrenia patients in respect to controls (P < .05). The MMN procedure,

effectively distinguished the four above mentioned auditory specifications in

both patients and controls. The grand average wave forms before subtraction

point to a distinctly different wave pattern with clear responses in the con-

trols. (B) The DL pointed to Right Ear Advantage both for patients and for

controls. The error rates were higher for patients than controls (P < .05). C)

The simple P300 componentwas diminishedin patients (P < .05). The overall

results could carry out various auditory components in a joined battery de-

sign. The cost-effectiveness, comparative ease of application, objective as-

sessment features can become useful parameters for monitoring the brain

information processing in schizophrenia as well as other cases. (Supporting

projects DEU.2006.KB.SAG.038–017).

ID: 552106

International Congress on Schizophrenia Research

60 4. 4. Electrophysiology

5. 5. Eye Movement Physiology

SCHIZOPHRENIA AND BIPOLAR PATIENTS

MANIFEST A COMMON INTERMEDIATE

PHENOTYPE OF DEFICIENT SENSORIMOTOR

TRANSFORMATION

Rebekka Lencer

2,1

, M. H. Harris

, J. L. Reilly

, M. S. Keshavan

J. A. Sweeney

Psychiatry and Psychotherapy, University of Luebeck, Luebeck,

Germany;

Center for Cognitive Medicine, University of Illinois at

Chicago, Chicago, IL, USA;

Psychiatry, Harvard Medical School,

Boston, MA, USA

Background: Pursuit eye tracking deficits that are regarded as intermediate

phenotype for schizophrenia may result from distinct disturbances of neu-

rocognitive processes including visual motion processing, sensorimotor

transformation, or the utilization of predictive mechanisms. Which of these

disturbances are the primary causes of the eye tracking phenotype, and

whether they are present in other psychotic disorders is yet unclear. Meth-

ods: First-episode patients with either bipolar disorder (N = 33), major de-

pression with psychosis (N = 23) or schizophrenia (N = 73) were matched to

111 healthy subjects. Participants performed three smooth pursuit para-

digms at different target speeds (4°,8°,16°,24°,32°) to evaluate sensori-

motor transformation and predictive pursuit abilities. All patients were

unmedicated at the time of testing with 108 patients being antipsychotic-

naı

ve and 21 patients having a median life-time exposure to antipsychotics

of two weeks. Results: Pursuit initiation gain was impaired in patients with

bipolar disorder and schizophrenia (P = .014) but bipolar patients initiated

pursuit later than all other groups (P = .007). Visual motion processing dur-

ing pursuit initiation as indicated by the saccade error in a step-ramp task

was not impaired in any group. While maintenance pursuit gain with un-

predictable ramp targets was reduced in all three patient groups compared

to healthy participants (P < .001), bipolar patients exhibited more severe

impairments than did patients with schizophrenia (P < .001) or psychotic

major depression (P < .001). No difference between groups was found for

smooth pursuit of predictive pursuit. Conclusions: During initiation and

maintenanceofpursuitofunpredictabletargets,which aredependentondirect

transformation of sensory input signals into motor commands, disturbances

weremoreseverein patients withbipolar disorder thaninpatients withschizo-

phrenia althoughbothgroups wereimpaired.In patients withpsychoticmajor

depression,sensorimotor transformation impairments weredetected,butonly

during maintenance pursuit. These findings imply that psychotic disorders,

especially bipolar disorder and schizophrenia, both manifest a common inter-

mediate phenotype of impaired sensorimotor transformation abilities point-

ing to disturbances in fronto-striatal circuits. In contrast, core visual motion

processing related to extrastriate area V5 appears to be unaffected in all three

patient groups with psychosis.

ID: 549036

TAKING ANIMAL BEHAVIORAL

PHARMACOLOGY PARADIGMS INTO CLINICAL

TRIALS TO EVALUATE COGNITIVE OUTCOMES

John A. Sweeney, J. Reilly, S. Hill, M. Harris, C. Rosen,

P. J. Weiden

Psychiatry, University of Illinois, Chicago, IL, USA

Selecting cognitive tasks for schizophrenia trials has many challenges.

Investigators typically choose between well-standardized neuropsycholog-

ical tests that tap into a generalized cognitive deficit vs. tasks that evaluate

particular cognitive functions known to be regulated by specific brain cir-

cuits and transmitter systems chosen on the basis of drug mechanism of ac-

tion. This presentation will make the case for the benefits of the latter

approach using tests of sensorimotor skills, working memory, procedural

learning and voluntary inhibitory control. It will contrast findings from neu-

ropsychological and translational oculomotor testing and fMRI studies in

untreated first episode patients examined before and 6-weeks after antipsy-

chotic treatment. Data will be presented from our Pittsburgh and Chicago

studies (now 60 total patients) where similar recruitment strategies, treat-

ment and cognitive measures were used. The data shows robust pretreatment

neuropsychological deficits in schizophrenia and marginal change after

treatment, consistent with many studies. In the same patients, while the mag-

nitudes of pretreatment oculomotor and neuropsychological deficits were

similar, oculomotor measures had significantly greater sensitivity to the neu-

rocognitive impact of antipsychotic treatment. Further, unlike the profile of

modest consistent change of neuropsychological performance after treat-

ment, oculomotor findings indicated greater post-treatment changes,

some beneficial and some adverse. The neurophysiological findings are con-

sistent with behavioral pharmacology research with nonhuman primates

performing the same tasks after D2 antagonists are given. Preliminary

data with fMRI studies using oculomotor tasks show treatment effects

that are larger even than laboratory neurophysiological testing. These

data will be used to illustrate the greater sensitivity and selectivity of trans-

lational neurophysiological measures compared to neuropsychological tests

for assessing the neurocognitive effect of antipsychotic treatment. Such neu-

roscience measures may be useful cognitive outcomes in clinical trials—per-

haps especially in Phase 2 proof of concept studies. Translational outcome

measures that build on animal model platforms in which the neurophysiol-

ogy and neurochemistry of brain systems supporting task performance are

well characterized may be especially useful in this regard.

ID: 550508

IS MOTION PERCEPTION DEFICIT IN SCHIZO-

PHRENIA A CONSEQUENCE OF EYETRACKING

ABNORMALITY?

Elliot Hong

, K. A. Turano

, H. B. O’Neill

, L. Hao

, I. Wonodi

R. P. McMahon

, G. K. Thaker

Department of Psychiatry, University of Maryland, Maryland

Psychiatric Research Center, Baltimore, MD, USA;

Wilmer Eye

Institute, Lions Vision Center, Johns Hopkins School of Medicine,

Baltimore, MD, USA

Background: Studies have shown that schizophrenia patients have motion

perception deficit, which is correlated with poor eyetracking performance.

These findings suggest that motion perception may cause eyetracking ab-

normality in schizophrenia. However, eye movement closely interacts with

motion perception. In this study, we examined how the known eyetracking

difficulties in schizophrenia patients may interact with their motion percep-

tion. Method: Two speed discrimination experiments were conducted in

a within-subject design. In Experiment 1, the stimulus duration was

150ms to minimize the chance of eyetracking occurrence. In Experiment

2, the duration was increased to 300ms, increasing the possibility of eye

movement intrusion. Regular eyetracking performance was evaluated in

a third experiment. Results: At 150ms, speed discrimination thresholds

did not differ between schizophrenia patients (n = 38) and controls (n =

33). At 300ms, patients had significantly higher thresholds than controls

(P = .03). Furthermore, frequencies of eyetracking during the 300ms stim-

ulus were significantly correlated with speed discrimination in controls (P =

.01) but not in patients, suggesting that eyetracking initiation may benefit

controls but not patients. The frequency of eyetracking during speed dis-

crimination was not significantly related to regular eyetracking perfor-

mance. Conclusions: Speed discrimination, per se, is not impaired in

schizophrenia patients. The observed abnormality appears to be a conse-

quence of impairment in generating or integrating the feedback information

from eye movements. This study introduces a novel approach to motion

perception studies, and highlights the importance of concurrently measur-

ing eye movements to understand interactions between these two systems;

International Congress on Schizophrenia Research

5. 5. Eye Movement Physiology 61

the results argue for a conceptual revision regarding motion perception ab-

normality in schizophrenia.

ID: 550203

BEHAVIORAL CHANGES FOLLOWING

DAILY PRACTICE OF SACCADE TASKS IN

SCHIZOPHRENIA

Madison Moore

, B. P. Austin

, K. A. Dyckman

,Q.Li

M. T. Amlung

, F. Meyer

, B. A. Clementz

, J. E. McDowell

Psychology, University of Georgia, Athens, GA, USA;

Department of Psychology, Bio-Imaging Research Center,

University of Georgia, Athens, GA, USA;

Departments of Psy-

chology and Neuroscience, Bio-Imaging Research Center, University

of Georgia, Athens, GA, USA;

Department of Psychiatry, Mas-

sachusetts General Hospital, Charlestown, MA, USA;

Department

of Psychology, University of Konstanz, Konstanz, Germany

It has been well documented that people with schizophrenia show impair-

ment in tasks requiring executive control such as inhibition. A simple test of

inhibition is the antisaccade task, which requires a glance towards the mir-

ror image of a peripheral cue. The goal of the current study is to determine

how practice on the antisaccade task changes performance on that task and

on related tasks known to assess executive control. Participants with schizo-

phrenia and healthy comparison subjects were assigned a single saccade

task to practice daily—either antisaccades or prosaccades (glances towards

a cue)—over a two-week period. Generalized executive control was evalu-

ated at pre- and post-test using two tasks. First, an ocular motor delayed

response (ODR) task was selected as a direct measure of changes due to

practice on a related, but distinctly different, saccade task. Second, the Wis-

consin Card Sorting Test (WCST) was selected as a means of evaluating

whether changes in executive control could be generalized beyond saccade

tasks. Preliminary results suggest that the healthy comparison group that

practiced antisaccades showed improved antisaccade performance across

the duration of the study, as well as improved performance on other exec-

utive function measures. The schizophrenia group that practiced antisac-

cades also showed an improvement in performance across executive

control measures. The schizophrenia group that practiced prosaccades,

however, showed poorer performance specifically on the ODR task by

making more anticipatory saccades. In sum, practice with saccade tasks

resulted in modest improvement in saccade task performance in both

schizophrenia and comparison groups. An exception is that the schizophre-

nia prosaccade-practice group showed more disrupted performance. This

study provides evidence that saccadic performance can be malleable within

certain parameters. The results are also are interesting for their practical

implications. If improvement on other tasks of executive control continues

to be seen in the schizophrenia antisaccade practice group, it suggests that

purposefully practicing executive control may be explored as a means of

improving activities of daily living. This research was supported in part

by a grant from the National Institute of Mental Health (MH076998)

and by the University of Georgia Honors Program’s Center for Undergrad-

uate Research Opportunities and the CURO 2008 Biomedical and Health

Sciences Research Fellowship.

ID: 551748

SCHIZOPHRENIA AND PSYCHOTIC BIPOLAR

DISORDER, BUT NOT PSYCHOTIC DEPRESSION,

SHARE INTERMEDIATE PHENOTYPE FOR

RESPONSE SUPPRESSION DEFICITS DURING

EARLY COURSE OF ILLNESS

Margret S. H. Harris; J. L. Reilly, C. Rosen, S. K. Keedy,

O. DeLeon, R. Marvin, P. J. Weiden, J. A. Sweeney

Department of Psychiatry, Center for Cognitive Medicine,

University of Illinois at Chicago, Chicago, IL, USA

Background: Response suppression deficits due to reduced voluntary

control of behavior have been investigated using antisaccade tasks and

are considered a promising intermediate phenotype for schizophrenia.

Disturbances in frontostriatal systems are believed to be the cause of

this deficit. Recently, multiple similarities across schizophrenia and psy-

chotic mood disorders have been demonstrated, including responsiveness

to antipsychotic medications, MRI morphometry, genetic linkage findings,

and cognitive performance deficits. Yet, few oculomotor studies have in-

vestigated the diagnostic specificity of this effect across psychotic disorders.

Methods: Unmedicated first-episode psychosis patients with schizophrenia

(N = 31), psychotic bipolar disorder (N = 21), or psychotic major depression

(N = 15) and 45 age- and IQ-matched healthy participants performed gap/

overlap antisaccade and visually-guided saccade control tasks. Results:

Both schizophrenia and psychotic bipolar patients showed similar and

higher antisaccade error rates than healthy individuals and psychotic de-

pression patients whose performance did not differ. No impairment of

the latencies of correct antisaccades was evident for any patient group.

Also, no group differences were observed for the accuracy or speed of vi-

sually-guided saccades, indicating intact sensory-motor systems across pa-

tient groups. For all groups, gap trials resulted in greater antisaccade error

rates and shorter antisaccade latencies as well as less accurate and faster

reflexive saccades compared to overlap trials. Conclusions: Results suggest

that a reduced ability to suppress context inappropriate behavior, reflected

in increased antisaccade error rates, may represent a shared intermediate

phenotype for schizophrenia and psychotic bipolar disorder. Contrary

to one of our previous studies, psychotic depression patients did not dem-

onstrate antisaccade performance impairments. These findings suggest that

both schizophrenia and psychotic bipolar disorder patient groups may ex-

perience a disruption of the prefrontal-striatal circuitry, via similar or dif-

ferent pathophysiologies, causing comparable cognitive abnormalities

across the disorders. Supported by R01 MH62134, R01 MH080066,

NARSAD.

ID: 551541

THE RELATIONSHIP BETWEEN OCULOMOTOR

MEASURES OF PREDICTIVE PURSUIT AND NEU-

ROPSYCHOLOGICAL MEASURES OF WORKING

MEMORY IN PATIENTS WITH SCHIZOPHRENIA

AND BIPOLAR DISORDER

Amanda F. Moates

, E. Ivleva

, D. Cole

, R. Gonzales

M. Cullum

, L. E. Hong

, G. Thaker

2,1

, C. Tamminga

Psychiatry, UT Southwestern Medical Center at Dallas, Dallas,

TX, USA;

Maryland Psychiatric Research Center, University of

Maryland Medical School, Baltimore, MD, USA

Oculomotor deficits have been well-documented in patients with schizophre-

nia (SZ) and a subset of their unaffected relatives, likely reflecting alterations

within the prefrontal cortex (FEF) and genetic liability for the disease. Inves-

tigation of specific components of the smooth pursuit eye movement system

has shown deficits in the predictive pursuit mechanism as seen by a reduced

gain (ratio of the eye velocity to target velocity) in response to a briefly invis-

ible target. Newer techniques, such as a novel covert stabilization technique

developedby Hong et al. (2007),allow moreprecise measurement ofthe brain

mechanismunderlyingthis predictivemechanismby stabilizing the targetim-

age on the fovea and eliminating overt awareness on target removal. Al-

though fewer studies of eyetracking dysfunction have been conducted in

patients with bipolar disease (BD), and particularly their families, some

data have suggested deficits in affective disorders that parallel those seen

in SZ. Independently, deficits in working memory have been established in

both groups of patients and family members. The working memory system

International Congress on Schizophrenia Research

62 5. 5. Eye Movement Physiology

has been hypothesized to be the cornerstone of executive function processes,

mediatedinthe prefrontal cortex, and found to beimpairedinthesedisorders.

The primary goal of this research is to characterize the relationship between

deficits in these two areas across both diagnostic groups and relatives, as

working memory is thought to contribute to the predictive pursuit mecha-

nism, and both may be promising endophenotypic markers of psychosis.

Thus far 147 patients have been enrolled with diagnoses spanning the diag-

nostic categories as follows: 46 SZ, 30 schizophrenia relatives, 34 BD, 25

bipolar relatives, and 12 comparison subjects. A broad battery of neuropsy-

chologicaltestsandoculomotortaskswereperformedaspart oflarger genetic

phenotyping study. Results from neuropsychological measures of working

memory (Wechsler Letter-NumberSequencing and Spatial Span)will be cor-

related with predictive pursuit measures of eyetracking, specifically predic-

tive acceleration and predictive gain. Datafrom both a traditional measureof

predictive pursuit abnormalities and from a newer novel covert stabilization

technique developed by Hong et al. will be presented.

ID: 551212

PREDICTION IN SMOOTH PURSUIT OF

SCHIZOPHRENIC PATIENTS

Andreas Sprenger

, W. Heide

1,3

, M. Nagel

, R. Lencer

Neurology, University Luebeck, Luebeck, Germany;

Psychiatry

and Psychtherapy, University Luebeck, Luebeck, Germany;

Neurology, City Hospital Celle, Celle, Germany

Background: The mechanisms of smooth pursuit eye movement (SPEM)

deficits in patients with schizophrenia are still under debate. Pursuit blank-

ing experiments have been used to test prediction components of pursuit

performance (eg, Becker and Fuchs, Barnes et al.). A recent pursuit blank-

ing fMRI study (Nagel et al. 2007) indicates that temporal (STG), frontal

(FEF) and cingulate brain areas (ACC) are more activated during predic-

tive pursuit in schizophrenia patients compared to healthy controls, that

seem to rely more on cerebellar mechanisms (Crus VIIIA). In a laboratory

experiment we wanted to test the oculomotor functions at different target

velocities and blanking durations in order to obtain more evidence for the

role of prediction during smooth pursuit eye movements of patients with

schizophrenia. Methods: Twenty patients and 20 age- and gender-matched

controls performed a pursuit blanking task at two target velocities (10, 15°/

s) and two blanking intervals (666, 1000 ms). Eye movements were recorded

by high resolution infrared videobased oculography. Results: Patient’s

smooth pursuit velocity was significantly lower compared to healthy con-

trols. After the target was blanked off, the deceleration of eye movement

velocity started after the same time interval in both groups. However, the

deceleration velocity was significantly slower in the patient group than in

controls. The acceleration after reappearance of the target started earlier in

controls than in patients but the acceleration itself was the same in both

groups. Conclusion: When a movement has to be continued after the stim-

ulus has disappeared, patients with schizophrenia seem to rely to a higher

extent on mechanisms of prediction and velocity storage than healthy sub-

jects. Our results could explain why patients with schizophrenia are consid-

erably impaired during pursuit of randomized step-ramp but much less

during highly predictable sinusoidal stimuli.

ID: 552599

International Congress on Schizophrenia Research

5. 5. Eye Movement Physiology 63

6. 6. Epidemiology

MENSTRUAL DISORDERS IN SCHIZOPHRENIA,

POSSIBLE CAUSES AND CONSEQUENCES

Rikus Knegtering, J. Ouwehand, S. Sytema, D. Wiersma

Psychiatry, University Medical Center Groningen, Rob Giel

Research Center, BCN, University of Groningen, Groningen,

Netherlands

Menstrual Disorders (MDs) like oligomenorrhea or amenorrhea are often

found in women with schizophrenia. Although MDs are often considered

to be antipsychotic medication related, a systematic literature review did

not support this view (Ouwehand and Knegtering in preparation). The

understanding of mechanisms involved in MDs may increase our under-

standing of the pathogenesis of MDs and schizophrenia. Method: After

being treated for 6 weeks with an antipsychotic information on MDs (us-

ing the ASFQ (Knegtering et al. 2003)), medication and blood levels of

antipsychotics and hormones were evaluated in all women with schizo-

phrenia aged 17 till 40 years attending the University Medical Center

of Groningen. Results: 71 women fulfilled inclusion criteria (including

no known organic causes for MD), mean age 27 (SD7) years). 22 women

used Oral Contraceptives (OC). 43 used prolactin elevating antipsychotics

(classical antipsychotics, risperidone), 28 prolactin sparing (clozapine,

quetiapine, olanzapine, sertindole, aripiprazole). Women using prolactin

elevating antipsychotics reported more often MDs (13 out of 27 =

(48%)) than women using prolactin sparing antipsychotics (4 out of 22

= (18%)), (v

, P = .015). OC and age did not predict MD. Prolactin levels

predicted MDs (ANOVA F = 5.278, P = .025). Independent of prolactin

levels these women with schizophrenia reported more MDs in comparison

to epidemiological comparable groups (prevalence in normal population

of same age is about 1% (van der Linden et al. 2004)). Conclusion: Pro-

lactin elevation during treatment with antipsychotics for six weeks was

associated with more frequent MD. Prolactin may not be to only factor

involved. Before the introduction of antipsychotics high numbers of MD

have been reported in women with schizophrenia. More studies are war-

ranted to understand the pathophysiology of schizophrenia and MD in

women.

References

1. Knegtering H, Castelein S, Teske A, et al.. The Development of the An-

tipsychotic and Sexual Functioning Questionnaire (ASFQ). In Kneg-

tering H. Antipsychotic treatment and sexual functioning, role of

prolactin. Ph.D thesis University of Groningen, The Netherlands;

2003. online http://irs.ub.rug.nl/ppn/254939104.

2. Van der Linden MW, Westert GP, De Bakker DH, Schellevis FG.

Second National Survey in General Practice. Utrecht/Bilthoven, The

Netherlands: Nivel/RIVM, 2004. http://www.nivel.nl/.

ID: 539311

TIME-TO-PREGNANCY AND RISK OF

SCHIZOPHRENIA IN OFFSPRING

Mark G. Opler

1,3

, K. Ornstein

, M. Perrin

1,3

, K. Kleinhaus

1,3

S. Harlap

1,3

, D. Malaspina

1,3

Psychiatry, New York University Medical Center, New York, NY,

USA;

Epidemiology, Columbia University, New York, NY, USA;

Environmental Medicine, New York University Medical Center,

New York, NY, USA

Abstract: Several studies have reported advanced paternal age as a risk

factor for schizophrenia. Given the likely role of paternal reproductive

health in this relationship, other factors associated with male fertility

should be examined. One variable that has been utilized in prior studies

of fertility, time-to-pregnancy may serve as a useful indicator. Using data

from the prospective Jerusalem Perinatal Study (JPS), a pregnancy co-

hort, data from prenatal interviews was linked to systematically collected

data on mental health outcomes in offspring. Examining those subjects

with intentionally conceived pregnancies, a modest increase in risk of

schizophrenia in offspring was measured in those with time-to-pregnancy

>6 months (OR = 1.63, 95% CI: 1.02, 2.60). When the relationship was

examined on the basis of parents’ ages at conception, it was discovered

that the risk was further increased for subjects with fathers >35 years old

and increased time-to-pregnancy (OR = 2.49, 95% CI: 1.05, 5.89). Ma-

ternal age did not appear to have the same influence on risk associated

with time-to-pregnancy in this sample. For subjects from more reli-

giously observant families, the risk associated with increased time-to-

pregnancy was statistically significant (OR = 2.09, 95% CI: 1.07, 4.1),

possibly due to reduced misclassification. These findings suggest that bi-

ological mediators related to paternal reproductive health, associated

with increased paternal age and increased time-to-pregnancy may be re-

sponsible for increasing the risk of schizophrenia. Additional research

using both epidemiologic approaches and laboratory methods are cur-

rently underway.

ID: 550801

CLINICAL AND FUNCTIONAL OUTCOME IN

SCHIZOPHRENIA AMONG A PREDOMINANTLY

TREATMENT-NAI

VE COHORT IN RURAL

ETHIOPIA

Abebaw Fekadu

1,2

, D. Kebede

, T. Shibre

, A. Alem

Psychological Medicine, King’s College London, Institute of

Psychiatry, London, United Kingdom;

Department of Psychiatry,

Faculty of Medicine, Addis Ababa University, Addis Ababa,

Ethiopia;

World Health Organization, Regional Office for Africa,

Brazzaville, Congo

Background: The established view that schizophrenia has a favorable out-

come in developing compared to developed countries has been recently

challenged. However, systematic studies from developing country settings

are scarce. We present the clinical and functional outcome of schizophrenia

among a predominantly treatment-naive cohort in a rural community set-

ting in Ethiopia. Methods: Cohort was identified in a 2-stage sampling de-

sign using key informants and measurement-based assessment. After

screening 68 378 adults, ages 15–49 years, 321 cases with schizophrenia

(82.7% men and 89.6% treatment-naive) were identified and systematically

followed-up for 3.4 years. Results: Almost a third (30.8%) of cases were

continuously ill and most of the remaining cohort experienced an episodic

course. Only 5.7% of the cases enjoyed a near-continuous complete remis-

sion although 27.4% were in complete remission in the month of the final

follow-up assessment. Living in a household with 3 or more adults, later

age of onset, and taking antipsychotic medication for at least 50% of the

follow-up period independently predicted complete remission. Functioning

and other measures of health related quality of life were significantly di-

minished in cases as compared to the general population of the area at

baseline and follow up. The level of functioning observed in cases from

Butajira was also lower than that reported for cases from developed coun-

tries. Conclusion: Both the functional and clinical outcome of schizophre-

nia in this setting appears to be worse than reported elsewhere in

developing countries. Additionally, functional outcome in this setting

appears to be even worse than that reported in developed countries.

Thus our findings support the observation that the outcome of schizophre-

nia in developing countries may be heterogeneous rather than uniformly

favorable.

ID: 550785

International Congress on Schizophrenia Research

64 6. 6. Epidemiology

SCHIZOPHRENIA AMONG THE BORANA

PASTORALIST COMMUNITY IN SOUTHERN

ETHIOPIA: A RE-VISIT OF CASE ASCERTAINMENT

METHODOLOGY

Teshome Shibre Kelkile

, S. Teferra

, M. Araya

, C. Morgan

Department of Psychiatry, Faculty of Medicine, Addis Ababa

University, Addis Ababa, Ethiopia;

Health Services and Population

Research Department, King’s College London, Institute of Psychi-

atry, London, United Kingdom

There are few reports on the prevalence of psychotic disorders among

isolated population groups. The major challenge is establishing a meth-

odologically sound system of case identification. In our population

based survey of the Borana pastoralist community in Ethiopia, we found

an overall prevalence of psychiatric disorders, assessed using the Com-

posite International Diagnostic Interview (CIDI), of 21.6%. However,

no case of schizophrenia was identified. We set out to investigate whether

this was because of the absence of schizophrenia or a function of the how

serious mental disorder is conceptualised in the Borana pastoralist com-

munity. We conducted 6 Focus Groups (FG) with key members of the

Borana pastoralist community, in order to investigate how serious men-

tal disorder is conceptualized. Subsequently, the FG participants were

used as Key Informants (KI) to identify cases with a possible psychotic

disorder, based on their conceptualization. Cases identified by KIs were

interviewed by a trained psychiatrist using a semi-structured interview

instrument, the Schedules for Clinical Assessment in Neuropsychiatry

(SCAN) to confirm presence of disorder. Out of 56 individuals invited

to the FGs, 49 (87.5%) participated. Participants were 25–60 years of age

and 26 (46.4%) were women. The concept of serious mental disorder

among the FG participants was informed by traditional beliefs and attri-

butions. However, there was a marked overlap between the behavioural,

emotional and biological signs identified by FG participants and those

described in modern classification systems. Following the FG, partici-

pants identified 6 individuals with schizophrenia and 13 with a psychotic

mood disorder that were subsequently confirmed through SCAN inter-

views. Concepts of serious mental disorder in the Borana pastoralist

community reflected traditional beliefs and practices. Nevertheless, indi-

viduals with a psychotic disorder were easily identified by key inform-

ants. The inability of the CIDI to identify any cases of schizophrenia

in our initial prevalence survey cannot be fully attributed to the distinct

local conceptualizations of these disorders. Studies of psychotic disor-

ders in such communities may benefit from combining structured inter-

views with the key informant method.

ID: 550758

REHOSPITALIZATION RATES OF PATIENTS

DISCHARGED WITH HALOPERIDOL OR

OLANZAPINE OR RISPERIDONE OR SULPIRIDE OR

ZIPRASIDONE OR CLOZAPINE AND ITS RELATION

TO PREVIOUS TREATMENT

Ana Paula Werneck

, H. Elkis

, J. A. Pinto

Department and Institute of Psychiatry, University of Sa˜o Paulo,

Sa˜o Paulo, Brazil;

Institute of Mathematics and Statistics,

University of Sa˜o Paulo, Sa˜o Paulo, Brazil

It is well established that antipsychotics represent the mainstay of treatment

for schizophrenia. An important outcome variable for drug effectiveness

is relapse/rehospitalization prevention. Based on the previous review we

hypothesize that second generation antipsychotics are superior to conven-

tional antipsychotics for preventing relapse and rehospitalizations. Hence,

the aim of this study was to observe rehospitalization rates of patients dis-

charged on a regimen of haloperidol or olanzapine or risperidone or sul-

piride or ziprasidone or clozapine. Data were collected from charts and

hospital database. Survival curves were estimated by the product-limit

(Kaplan-Meier) formula. A total of 118 patients entered the selection

and were followed-up from discharge date (between January 2000 and De-

cember 2004) until December 2007. The percentage of patients remaining

discharged were 70% for haloperidol, 86% for olanzapine, 59% for risper-

idone, 90% for sulpiride, 75% for ziprazidone and 75% for clozapine. There

were no significant differences in readmission rates among groups (Log

Rank = 10,84, df = 5, P = .055). The results demonstrate lower rehospital-

ization rates for sulpiride and olanzapine and lower rate for risperidone,

with clozapine, haloperidol and ziprasidone in an intermediate position.

Risperidone along with haloperidol and olanzapine were the most frequent

choices for naı

ve patients and clozapine, according to our expectations, was

started in most patients with history of use of three or more antipsychotics

previous to hospitalization. It is interesting to observe that there is a signif-

icant trend that could indicate differences among antipsychotics in rehospi-

talization rate that is not statistical different in this study due to a small

sample size.

ID: 550754

SPECIFICITY OF ASSOCIATION BETWEEN

CHILDHOOD ABUSE AND PSYCHOSIS IN

A CLINICAL FIRST-EPISODE SAMPLE

Helen Fisher

, C. Morgan

, K. Morgan

, P. Dazzan

, G. Doody

T. Craig

, J. Leff

, P. McGuffin

, P. Jones

, R. Murray

P. Fearon

Psychological Medicine, Institute of Psychiatry, KCL, London,

United Kingdom;

Division of Psychiatry, University of Nottingham,

Nottingham, United Kingdom;

Department of Psychiatry,

University of Cambridge, Cambridge, United Kingdom

Childhood adversity has been associated with increased risk of developing

psychotic symptoms in adulthood but these studies are methodologically

limited. Therefore, this study sought to explore the prevalence of childhood

abuse amongst those with and without a psychotic disorder using detailed

assessments of a large epidemiological case-control sample. Information

relating to different types of childhood maltreatment (parental neglect

and antipathy, physical and sexual abuse) was obtained using the Child-

hood Experiences of Care and Abuse Questionnaire (CECA.Q). Data

were collected on 182 first-presentation psychosis cases and 246 epidemi-

ologically-matched controls drawn from two UK centres as part of the

ÆSOP study. Initial exploration of the data demonstrated that there

was good test-retest reliability over 7 years (physical abuse: k = 0.634,

P < 0.001; sexual abuse: k = 0.590, P < .01) and convergent validity (spec-

ificity: 100% physical abuse and 98% sexual abuse) for reports of childhood

abuse from participants diagnosed with psychosis. Analysis revealed that

psychosis cases were three times more likely to report severe physical abuse

from mother that commenced prior to 12 years of age, even after adjust-

ment for other forms of adversity, gender, ethnicity, age at interview, study

centre and parental social class (OR = 3.15, 95% CI = 1.13–8.78, P = .028).

Further, there was a trend for reports of severe childhood sexual abuse to

be associated with around a two-fold increased risk for psychosis but this

failed to reach significance after adjustment for all confounders (OR = 2.09,

95% CI = 0.95–4.96, P = .065). Paternal physical, parental neglect and an-

tipathy did not significantly predict psychosis case status. These findings

provide more comprehensive evidence of the association between childhood

adversity and development of psychosis in adulthood and lend support to

the hypothesis that it is particularly intrusive early life events which are

important in the aetiology of psychotic disorders. Further analyses are

underway to explore the potentially mediating role of familial susceptibility

and candidate genes.

ID: 550749

International Congress on Schizophrenia Research

6. 6. Epidemiology 65

DESCRIPTIVE EPIDEMIOLOGY AND COMORBID-

ITY OF SCHIZOPHRENIA VS. BIPOLAR DISORDER

IN THE NATIONAL HOSPITAL DISCHARGE

SURVEY (NHDS)

Natalya Weber

, D. N. Cowan

, S. A. Bedno

, M. A. Cavicchia

R. Yolken

, D. W. Niebuhr

Epidemiology, Walter Reed Army Institute of Research, Silver

Spring, MD, USA;

Neurovirology, Johns Hopkins University

School of Medicine, Baltimore, MD, USA

Introduction: Medical morbidity and mortality rates in schizophrenia and

bipolar disorder patients are elevated compared to the general US popu-

lation. More than 50% of patients with schizophrenia and bipolar disorder

have been diagnosed with one or more comorbid conditions. In addition

to providing descriptive epidemiology, the purpose of this study is to iden-

tify serious medical comorbidity occurring more frequently among those

hospitalized with schizophrenia than those with bipolar disorder. Meth-

ods: Using the 1979–2003 National Hospital Discharge Survey

(NHDS), we evaluated temporal trends in the proportional morbidity

(PM) of schizophrenia and bipolar disorder, demographic characteristics

and the most frequent comorbid conditions among hospitalizations with

schizophrenia, compared to hospitalizations with bipolar disorder. The de-

mographic characteristics and comorbidity are presented for the NHDS

sample of 5 731 450 records. Results: Percent of hospitalized with schizo-

phrenia versus bipolar disorder was higher among males (PM ratio 1.5)

than females, blacks (2.7) and other (1.6) compared with whites, age group

45–64 (9.8) followed by 15–44 (9.3) and 65þ (8.0) compared with those

younger than 15, Northeast and West (1.2) followed by Midwest (1.1)

compared with South. There was a significant increase over time in the

proportion of discharges with schizophrenia and bipolar disorder and

a significant decrease in the mean length of hospital stay. Compared

with bipolar disorder discharges, those with schizophrenia have a signifi-

cantly higher proportion of comorbidity with the following system groups

of medical conditions: neoplasms (overall PM ratio 1.1), diseases of blood-

forming organs (1.2), respiratory (1.1), genitourinary (1.1) systems, as well

as diseases of the skin and subcutaneous tissue (1.2). In addition, we iden-

tified a number of serious specific medical conditions occurring at least 1.5

times more frequently among discharges with schizophrenia. Discussion:

We found that compared with bipolar disorder, those hospitalized with

schizophrenia were more likely to have some serious comorbidity both

at the system level and for certain specific conditions. These finding

may justify higher scrutiny for those conditions during clinical examina-

tion of schizophrenia patients. Closer attention to prevention, early diag-

nosis and treatment of comorbid conditions may decrease medical

morbidity and mortality and improve the prognosis of schizophrenia

patients.

ID: 550643

ADULT-ONSET SCHIZOPHRENIA IN THE US

MILITARY: PATTERNS BY SEX, RACE AND AGE

David N. Cowan

, N. S. Weber

, S. A. Bedno

, M. A. Cavicchia

R. Yolken

, D. W. Niebuhr

Epidemiology, Walter Reed Army Institute of Research, Silver

Spring, MD, USA;

School of Medicine, Johns Hopkins University,

Baltimore, MD, USA

Background: There is substantial uncertainty about some aspects of the

epidemiology of schizophrenia, and data are sometimes conflicting.

There is limited prevalence data on US civilian population, and less

incidence data, especially for subgroups. US prevalence is approximately

1%, with conflicting data for sex, age, and racial groups. Incidence

estimates include 0.1–0.4/1000/yr (Jablensky, 1992 WHO international

estimate), and 0.08.43/1000/yr (10th–90th%) (McGrath, 2004, interna-

tional estimate). The prevalence of severe mental illnesses among

military personnel is lower than among the general population due to

self-deferral, medical screening at entry, and rapid attrition from

military. Methods: We used hospitalization data from the Defense Med-

ical Epidemiology Database for 1998–2007. Only first event hospitaliza-

tions were included. Analyses were conducted at 3-digit ICD-9 code

level. This report is a descriptive and graphical presentation of the

data to provide insight into patterns of disease. Results: During the study

period (1998–2007) there were 1,995 first-time military hospitalizations

for schizophrenia, occurring across 13 787 901 person years (PY), for an

overall incidence density (ID) 0.15/1000/PY. Over the study period the

rate of first time hospitalization for schizophrenia changed little, with

a marginally significant decrease among men (0.16 to 0.14, P = .07).

There were no changes over time by race or age. The female:male

IDR was 1.1 (P > .10). The ID ratio (IDR) among blacks (compared

to whites) was 2.4 (P < .01). Relative to those 40þ,theIDRwas14.4

(P < 0.01) for <20, decreasing to 2.4 (P < .01) for those 35–39. Among

whites rates decreased consistently with age; among blacks rates in-

creased from <20 to 20–24, then declined. At all age groups rates

were higher among blacks than whites. Among men rates declined con-

sistently with age; among women rates were flat for ages 20–24, 25–29,

and 30–34, and then declined. For younger personnel rates were higher

among men; among older they were higher among women, with rates

crossing at the 25–29 age group. Discussion: There are substantial dif-

ferences in patterns of schizophrenia by sex, race, and age. While overall

patterns provide some insight, only by reviewing data on subgroups can

these more specific patterns be discerned. These analyses clearly indicate

that there are certain groups at relatively high risk, and may provide guid-

ance in early detection of schizophrenia.

ID: 550641

SYSTEMATIC REVIEW AND META-ANALYSIS OF

POPULATION-BASED STUDIES OF PREMORBID

INTELLIGENCE AND SCHIZOPHRENIA

Golam Khandaker

, J. H. Barnett

2,3

, I. R. White

, P. B. Jones

2,5

Psychiatry in Learning Disability, Avon and Wiltshire Mental

Health Partnership NHS Trust, Bristol, United Kingdom;

Department of Psychiatry, University of Cambridge, Cambridge,

United Kingdom;

Psychiatric and Neurodevelopmental Genetics

Unit, Massachusetts General Hospital, Boston, MA, USA;

MRC

Biostatistics Unit, University of Cambridge, Cambridge, United

Kingdom;

Cameo (Cambridge Early Intervention in Psychosis

Service), Cambridge and Peterborough NHS Foundation Trust,

Cambridge, United Kingdom

Introduction: Previous studies have reported lower premorbid intelligence

in children and adolescents who will later develop schizophrenia. (1)

Method: We systematic reviewed the literature until August 2008, and

to the increase external validity of the research question only selected pop-

ulation-based prospective, retrospective and nested case-control studies for

meta-analysis. Included studies reported a standardised measure of intelli-

gence completed in childhood or well before any symptoms of illness. Sep-

arate meta-analysis was completed for full-scale, verbal and performance

IQ scores. Regression analysis was carried out to examine a dose-response

relationship between low IQ and risk of schizophrenia. Meta-regression

was used to examine associations between study effect sizes, mean age

of onset, and age at IQ assessment. Results: We identified 12 studies com-

prising 2654 cases and 728143 controls. Full-scale, verbal and performance

IQ all showed significant premorbid decrements (effect size 0.43–0.45) in fu-

ture cases of schizophrenia. There was no significant difference between pre-

morbid verbal and performance IQ in schizophrenia cases. A clear dose

International Congress on Schizophrenia Research

66 6. 6. Epidemiology

response effect was observed between low IQ and risk of schizophre-

nia, while risk increased by 3.7% with decrease in each IQ point

(Coef. 0.037, SE 0.0013, P < .0001). There was a significant associ-

ation between magnitude of IQ decrement and age at onset of illness (P =

.0007) and a marginal trend for greater decrement when IQ was assessed at

later ages (P = .06). Discussion: Compared with their peers, children who

will later develop schizophrenia show a consistent decrement of around

6–7IQpoints in both verbaland non-verbalIQ.Thereis alsoa dose response

relationship between low IQ and risk of schizophrenia. Associations with

age of onset and age of testing, suggests a progressively widening gap

between normal and schizophrenic brain development from birth until

symptom onset.

Reference

1. Aylward E, Walker E, Bettes B. Intelligence in schizophrenia:

meta-analysis of the research. Schizophr Bull. 1984;10(3):430–459.

ID: 550635

SOCIAL CHARACTERISTICS OF NEIGHBORHOODS

AND THE INCIDENCE OF PSYCHOTIC DISORDERS

Wim Veling

, E. Susser

, D. March

, H. W. Hoek

1,2

Center for Early Psychosis, Parnassia Psychiatric Institute, The

Hague, Netherlands;

Department of Epidemiology, Mailman

School of Public Health, Columbia University, New York, NY, USA

One of the most consistent findings in the epidemiology of schizophrenia

is that those who live in cities are at greater risk. It is much less clear which

risk factors drive this increased risk, because few attempts have been made

to move beyond the crude measure of urbanicity. Social factors across

different levels of organization may be involved, in particular those in-

volving availability of and access to social capital. We conducted a first

contact incidence study of psychotic disorders between 1997 and 2005 in

the city of The Hague, the Netherlands (n = 618) and investigated asso-

ciations between individual and neighborhood characteristics and the in-

cidence of psychotic disorders, using multilevel Poisson regression

analysis. First, after adjustment for individual level age, sex, marital status

and ethnic minority status, measures of neighborhood socioeconomic

level, residential mobility, population density, voter turnout at local elec-

tions and crime level were all significantly associated with the incidence of

psychotic disorders in a dose-response fashion. We calculated a cumula-

tive score of neighborhood social disadvantage, based on these five indi-

cators and divided it into tertiles (low, medium and high social

disadvantage). The individual level adjusted incidence rate ratio (IRR)

of psychotic disorders was 1.89 (95% Confidence Interval = 1.38–2.60)

for individuals living in the socially most disadvantaged neighborhoods

and was 1.35 (95% CI = 0.95–1.92) for those living in moderately disad-

vantaged neighborhoods compared to those living in the most advantaged

areas (Wald v

[df = 2], 18.19, P < 0.0005). Second, we investigated the

effect of the ethnic composition of neighborhoods on the incidence of psy-

chotic disorders, not only among ethnic minorities, but also among native

Dutch. In neighborhoods with a very high proportion of non-Western eth-

nic minorities, the incidence rate for native Dutch was 35.4 per 100 000

(95% CI = 18.2–62.1), indirectly standardized to age, sex and neighbor-

hood socioeconomic level. In neighborhoods with a relatively low propor-

tion of ethnic minorities, the incidence rate for native Dutch was 21.1

(18.4–24.2). Among ethnic minorities, the pattern was opposite. These

results are suggestive for an ethnic density effect both in ethnic minorities

and in native Dutch. Thus, neighborhood social characteristics strongly

influence the incidence of psychotic disorders within an urban environ-

ment in interplay with individual factors.

ID: 550624

THE RATE OF TB IN INDIVIDUALS WITH

SCHIZOPHRENIA: CLUES TO INFLAMMATORY

MECHANISMS?

Marie Casey Olseth

, D. R. Hanson

1,2

Psychiatry, University of Minnesota, Minneapolis, MN, USA;

Psychiatry, VA Medical Center, Minneapolis, MN, USA

Previously our group hypothesized that some psychotic disorders are the

result of disruption of the microvascular circulation of the brain brought

on by inflammatory processes. If inflammation is a contributing factor to

the development of psychosis, individuals with chronic inflammatory ill-

nesses, should display a higher incidence of psychosis as compared to

the general population. One such cause of chronic inflammation and acti-

vation of cytokine systems is tuberculosis (TB). TB susceptibility is genet-

ically mediated in part, based on twin and family studies in the literature.

TB infections generate circulating inflammatory agents that can have dis-

tant effects. Objective: To determine if individuals with schizophrenia have

a higher rate of tuberculosis as compared to individuals without schizo-

phrenia but who share similar contagion-risk factors. Method: An analysis

of the rates of tuberculosis in a historic (born 1860–1935) population of

psychiatric patients who had been institutionalized long term in Minnesota

State Hospitals. Complete psychiatric and medical records including autop-

sies were available. Psychiatric diagnoses were updated independently by

two clinicians according to RDC criteria. Results: The rate of TB in people

with schizophrenia was approximately 20% compared to 0.5 % in affective

illnesses and 0.5% in organic psychoses. Conclusions: The implication from

this study is that the individuals with schizophrenia have a higher incidence

of tuberculosis. There are at least two possibilities to consider. Chronic cir-

culating inflammatory processes may impact the brain leading to schizo-

phrenia. Alternatively, schizophrenia and TB susceptibility may both be

consequences of inherited perturbations of the immune system and innate

inflammatory processes triggered by unspecified environmental or epige-

netic factors.

ID: 550604

PRENATAL EXPOSURE TO TOBACCO IN PATIENTS

WITH FIRST EPISODE PSYCHOSIS

Geoffrey Smith, K. P. Good, G. W. MacEwan, L. C. Kopala,

T. S. Ehmann, A. E. Thornton, A. M. Barr, D. J. Lang,

S. W. Flynn, W. G. Honer

University of British Columbia, Vancouver, BC, Canada

Smoking during pregnancy impairs placental development and is associ-

ated with fetal undernourishment, slow fetal growth, and low birthweight.

In addition, nicotinic aceylcholine receptors are expressed early in fetal de-

velopment and animals exposed to nicotine in-utero have abnormal neuro-

nal proliferation and differentiation, and show developmental disruptions

in cholinergic and catecholaminergic systems. The neuronal consequences

of prenatal nicotine exposure in humans are not well defined but those born

to mothers who smoked during pregnancy are at increased risk for a range

of physical, behavioral, cognitive, and psychiatric problems and are more

likely to initiate smoking in adolescence. Many of these same problems are

observed in patients with psychosis, but have not been investigated in re-

lationship to maternal smoking. The goals of the present study were 1. to

determine if the mothers of first episode psychotic (FEP) patients were

more likely to have smoked during pregnancy than other women, and 2.

to determine if the known correlates of this early risk factor are observed

in FEP patients. An initial FEP sample (n = 80), a geographically distinct

FEP replication sample (n = 96), and a healthy control group (n = 58) were

recruited, and the results from a 1993 Canadian smoking survey were ac-

quired. Birth history was obtained from maternal report. FEP, control and

survey groups were compared for the proportion who were prenatally ex-

posed to nicotine. Exposed and unexposed individuals were compared for

International Congress on Schizophrenia Research

6. 6. Epidemiology 67

health-related and cognitive correlates of exposure to nicotine. The percent-

age of mothers who smoked during pregnancy was similar in the two FEP

samples (30%, 33%) and significantly greater than that of the control and

survey samples (15%, 19%). Prenatal tobacco exposure was associated with

low birthweight, poor academic achievement during childhood and ciga-

rette smoking in all three samples, and with obesity in FEP patients.

The present findings suggest that patients with psychosis are more likely

to have been prenatally exposed to nicotine than is observed in the general

population. As with the general population, prenatal exposure to nicotine is

associated with an increased risk of health-related and cognitive problems

in psychotic patients. The relatively high proportion of psychotic patients

exposed to nicotine in utero suggests that the negative correlates of this

early risk are likely to be commonly observed in patients with psychosis.

ID: 550573

GLUCOSE TOLERANCE IN NEWLY DIAGNOSED,

ANTIPSYCHOTIC-NAIVE PATIENTS WITH NON-

AFFECTIVE PSYCHOSIS

Brian W. Kirkpatrick

, E. Fernandez-Egea

, C. Garcia-Rizo

E. Parellada

, A. Justicia

, M. Bernardo

Psychiatry and Health Behavior, Medical College of Georgia,

Augusta, GA, USA;

Psychiatry, Cambridge and Peterborough

Foundation Trust., Cambridge, United Kingdom;

Psychiatry,

FUNDACIO PRIVADA CLINIC PER A LA RECERCA

BIOMEDICA, Barcelona, Spain

Some studies suggest patients with schizophrenia have an increased risk of

diabetes prior to antipsychotic use. Small sample sizes and the potential for

confounding by hypercortisolemia have decreased confidence in those

results. We administered a two-hour oral glucose tolerance test to 50 newly

diagnosed, antipsychotic-naı

ve patients with nonaffective psychosis and 50

matched control subjects. The patients had significant increases in two-

hour glucose (respective means [SD] for the patient and control groups

were 111 mg/dL [35.2] and 82 mg/dL [19.3]; t = 5.15, P < .001). The prev-

alence of impaired glucose tolerance or diabetes was 16% of patients vs. 0%

of control subjects (P < .003). These differences could not be attributed to

differences in cortisol concentrations, smoking, gender, neighborhood of

residence, demographic variables, body mass index, aerobic conditioning,

ethnicity, socioeconomic status, or age. Fasting glucose and hemoglobin

A1c concentrations did not differ between the two groups. Patients with

nonaffective psychosis appear to have an increased prevalence of abnormal

glucose tolerance prior to antipsychotic treatment. These underlying

problems may contribute to the metabolic side effects of antipsychotic

medications.

ID: 550568

ALCOHOL INDUCED PSYHOTIC DISORDER AND

DELIRIUM IN THE GENERAL POPULATION

Jonna Pera

, K. Kuoppasalmi

, S. Pirkola

1,3

, S. I. Saarni

S. Koskinen

,T.Ha

rka

nen

,J.Lo

nnqvist

1,3

, J. Suvisaari

1,2

Department on Mental Health and Alcohol Research, National

Public Health Institute, Helsinki, Finland;

Department of Social

Psychiatry, Tampere School of Public Health, University of Tam-

pere, Tampere, Finland;

Department of Psychiatry, University of

Helsinki, Helsinki, Finland;

Department of Health and Functional

Capacity, National Public Health Institute, Helsinki, Finland

Objective: We estimated the lifetime prevalence (LTP) of alcohol-induced

psychotic disorder and delirium in the general population, and compared

subjects with alcohol dependence with and without psychosis. Method: In

a nationally representative survey of 8028 persons aged 30 or over, the life-

time diagnosis of alcohol dependence was assessed in the CIDI interview

(1). In the re-assessment of psychotic disorders (2), the whole sample

was screened using CIDI, health examination and national registers.

Best-estimate DSM-IV-TR diagnoses were formed by combining SCID-I

interview and case note data. Prior hospital treatments were collected

from the National Hospital Discharge Register. Number of deaths during

eight year follow-up was obtained from the National Register of Deaths.

Results: The LTP was 0.41% (95% CI = 0.29–0.57) for alcohol-induced psy-

chotic disorder (n = 31) and 0.18% (95% CI = 0.11–0.32%) for alcohol-in-

duced delirium (n = 14). 6 subjects had had both diagnoses, and when each

individual was only counted once, the LTP for the total alcohol-induced

psychotic syndromes (AIPS) was 0.51% (95% CI = 0.38–0.70). The highest

LTP (1.77% 95% CI = 1.06–2.94) was found among men aged 45–54.

Among subjects with the diagnosis of alcohol dependence (n = 482) the

LTP for AIPS was 4.79%, and for all psychotic disorders 9.17%. Being

45–54 years, belonging to low socioeconomic groups, reporting mental

health or alcohol problems in fathers, having had high number of hospital

treatments or hospital treatments for alcohol-related disorders were asso-

ciated with AIPS among alcohol dependents. 35.9% and 7.0% of subjects

with AIPS and with alcohol dependence died in the follow-up, respectively.

The hazard ratios for mortality among subjects with AIPS were 13.0 com-

pared with the total population and 9.4 compared with other alcohol

dependents. Conclusions: Alcohol-induced psychotic disorder and delirium

are common in the general population, especially among working-aged

men. They are associated with high number of hospital treatments and

with high mortality.

References

1. Pirkola SP, Poikolainen K, Lo

nnqvist JK. Currently active and remit-

ted alcohol dependence in a nationwide adult general population–

results from the Finnish Health 2000 study. Alc Alc. 2006;41:315–20.

2. Pera

J, Suvisaari J, Saarni SI et al. Lifetime prevalence of psychotic

and bipolar I disorders in a general population. Arch Gen Psychiatry.

2007;64:19–28.

ID: 550544

BDNF VAL66MET GENOTYPE, MALE GENDER

AND ADOLESCENT CANNABIS USE DETERMINE

AGE OF ONSET OF PSYCHOSIS: A STUDY IN

A LARGE SAMPLE OF PSYCHOTIC PATIENTS

Jeroen Decoster

, M. de Hert

, G. Kenis

, J. van Os

R. van Winkel

1,2

UPC KULeuven, campus Kortenberg, Kortenberg, Belgium;

Department of Psychiatry and Neuropsychology, Maastricht

University Medical Centre, Maastricht, Netherlands

Several factors have been suggested to decrease age at onset of psychosis,

including male gender, cannabis and other drug use and BDNF Val66Met

genotype. The BDNF Val66Met polymorphism, of which Met-substitution

is associated with reduced intracellular trafficking and activity-dependent

BDNF secretion, was genotyped in a sample of 587 patients with psychosis.

Survival analyses with time from birth to age at first admission as indicator

for survival time were fitted as a function of several predictor variables, as

were multivariate Cox regression models. BDNF Val66Met genotype was

significantly associated with survival time (log-rank test: v

= 4.3, P = .039),

with Met-carriers being admitted for the first time a mean of 1.2 years ear-

lier. Gender, adolescent cannabis use (<16 yrs), lifetime stimulant, cocaine

or psychedelics use were also significantly associated with survival time (log

rank test P < .05). In the multivariate Cox regression models BDNF Val66-

Met Met genotype (Hazard Ratio (HR) 1.23, P = .026), adolescent cannabis

use (HR 1.59, P = .005) and male gender (HR 1.58, P < .001) remained

significant predictors of earlier age at onset. Results did not considerably

change when analyses were restricted to patients with an age at onset below

35 years. No interaction effects were observed. These results confirm find-

ings of earlier studies suggesting an effect of BDNF Val66Met genotype,

International Congress on Schizophrenia Research

68 6. 6. Epidemiology

cannabis use before age 16 (but not later cannabis use) and male gender on

age of onset of psychosis in a mainly Caucasian population.

ID: 550516

PATERNAL AGE IS ASSOCIATED WITH

INCREASED MORTALITY RISK IN PSYCHOTIC

DISORDERS

Brian James Miller

, J. Pihlajamaa

, M. Henriksson

, H. Hannele

M. Huttunen

, A. Tanskanen

, M. Cannon

, J. Haukka

,J.

Suvisaari

, B. Kirkpatrick

Psychiatry and Health Behavior, Medical College of Georgia,

Augusta, GA, USA;

Mental Health and Alcohol Research,

National Public Health Institute, Helsinki, Finland;

Psychiatry,

Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin,

Ireland

Increasing paternal age is a replicated risk factor for schizophrenia in the

offspring, while schizophrenia is associated with increased mortality. A pre-

vious study of men with psychosis found that paternal age was correlated

with suicide. We examined paternal age and mortality in N = 834 patients

with psychosis born in Helsinki, Finland, between January 1, 1951 and

December 31, 1960. Subjects were followed until June 2006 (ages

46–55). Offspring of fathers who were under 20 years old were omitted

from the analyses because of small numbers. In multivariate Cox propor-

tional hazard models, after adjustment for maternal age, gender, and age of

onset of psychotic symptoms, paternal age (as a continuous variable) was

a significant predictor of suicide mortality (Hazard Rate Ratio (HRR) =

1.05, 95% CI = 1.01–1.10, P = .020). By contrast, increasing maternal

age was associated with decreased risk of suicide mortality (HRR =

0.923, 95% CI = 0.87–0.98, P = .012). In males, paternal age was associated

with increased suicide risk (HRR = 1.05, 95% CI = 1.00–1.11, P = .050), and

maternal age was associated with decreased suicide risk (HRR = 0.90, 95%

CI = 0.83–0.97, P < .01). In females, the effect of paternal age on suicide risk

was not statistically significant, although the point estimate was the same as

in males (HRR = 1.054, 95% CI = 0.97–1.15, P = .22). Paternal is associated

with a linear increase in suicides in males with psychosis. The pathway(s)

associated with this increased mortality remain to be determined. The rel-

ative contribution of biological and psychosocial factors are unclear.

ID: 550435

EVIDENCE THAT ONSET OF CLINICAL PSYCHOSIS

IS THE OUTCOME OF PROGRESSIVELY MORE

PERSISTENT SUBCLINICAL PSYCHOTIC

EXPERIENCES: IMPLICATIONS FOR EARLY

INTERVENTION

Jim Van Os

1,2

, M. G. Dominguez

, M. Wichers

, R. Lieb

3,5

H. U. Wittchen

3,4

Department Psychiatry and Neuropsychology, Maastricht Uni-

versity, The Netherlands, Maastricht, Netherlands;

Division of

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom;

Clinical Psychology and Epidemiology Unit, Max

Planck Institute of Psychiatry, Munich, Germany;

Institute of

Clinical Psychology and Psychotherapy, Technical University

Dresden, Dresden, Germany;

Epidemiology and Health Psychol-

ogy, Institute of Psychology, University of Basel, Basel, Switzerland

Objective: It may be possible to intervene early in the prodromal phase of

psychosis. However, very little is known about the course of psychometric

risk indicators of psychosis in epidemiological samples of young people. To

examine the hypothesis that developmental expression of psychometric risk

in the form of subclinical psychotic experiences is usually transitory, but in

some instances may become abnormally persistent and progress to a clinical

psychotic state. Design: Prospective cohort study. Expression of psychosis

was assessed, using self-report and interview, four times over a period of 8.4

years. General population of adolescents, 845 individuals aged 14–17 years

at baseline, in Munich, Germany (Early Developmental Stages of Psycho-

pathology Study). A distinction was made between psychotic experiences

with impairment (clinical psychosis; defined as psychotic experiences with

Help-seeking and/or Dysfunction) and without impairment (subclinical

psychosis). Transition from subclinical psychosis at T0-T2 to clinical psy-

chosis at T3 was examined as a function of the level of prior persistence of

subclinical psychosis (present never, once, twice or thrice at T0, T1 and T2).

Results: The more subclinical psychosis persisted over the period T0-T2, the

greater the risk of transition to clinical psychosis at T3 in a dose-response

fashion (subclinical psychosis expression once over T0-T2: OR = 1.5 (CI:

0.6–3.7), post-test probability (PP) = 5%; twice: OR = 5.0 (CI: 1.6–15.9),

PP = 16%; at all three measurements: OR = 9.9 (CI: 2.5–39.8), PP =

27%. Of all clinical psychosis at T3, a third (38.3%) was preceded by sub-

clinical psychotic experiences at least once, and a fifth (19.6%) at least twice.

Conclusions: A significant proportion of psychotic disorder may be concep-

tualised as the rare poor outcome of a common developmental phenotype

characterised by persistence of psychometrically detectable subclinical psy-

chotic experiences. Early intervention efforts may target the factors that

make the common and transitory developmental expression of subclinical

psychosis persist.

ID: 550401

OBSTETRIC FACTORS, PRENATAL INFECTIONS

AND SCHIZOPHRENIA IN ADOLESCENCE AND

EARLY ADULTHOOD

Philip Nielsen, M. G. Pedersen, C. B. Pedersen, P. B. Mortensen

National Centre for Register-based Research, University of Aarhus,

Aarhus, Denmark

Several studies have reported an association between obstetric complica-

tions and schizophrenia. Especially children born small for gestational

age (SGA) have been reported to have an increased risk of schizophrenia.

No studies have however quantified whether the effect of SGA was related

to individual or familial factors. In addition, we also studied whether indi-

viduals exposed to prenatal infections have an increased risk of schizophre-

nia. Data from three population based registers, the Danish Medical

Birth Register, the Danish Psychiatric Central Register and the Danish Na-

tional Hospital Register were used. We used an historical population based

cohort design and selected all individuals born between 1979 and 1996 (n =

921 397). We identified all individuals who were listed in the Psychiatric

2006 with schizophrenia (n = 1823). We accounted for the potential con-

founding effect of parents age at child’s birth, urbanicity, parents history

of at least one psychiatric diagnosis. Individuals being small for gestational

age are more likely to develop schizophrenia, than individuals not being

small for gestational age (RR = 1.3). Results further indicate that having

an older sibling who were also small for gestational age increased the risk

for schizophrenia (RR = 2.0). Maternal infection during pregnancy in-

creased the risk by 1.4. Schizophrenia risk was associated with maternal

infections requiring hospitalization, especially infections during second tri-

mester. The association between being small for gestational age and schizo-

phrenia may be due to risk factors aggregating in families and shared by

schizophrenia and SGA, rather than an effect of SGA per se. This work

was supported by the Stanley Medical Research Institution.

ID: 550356

International Congress on Schizophrenia Research

6. 6. Epidemiology 69

ARE SELF-REPORTED AUDITORY HALLUCINA-

TIONS AMONG ADOLESCENTS VERIFIED ON

CLINICAL INTERVIEW?

Mary Cannon

1,2

, I. Kelleher

, A. Murtagh

, M. Blanchard

M. Harley

Department of Psychiatry, Royal College of Surgeons in Ireland,

Dublin, Ireland;

Department of Psychiatry, Beaumont Hospital,

Dublin, Ireland

The prevalence of auditory hallucinations among adolescents ranges from 5–

30% in population-based surveys. The rate of such self-reported hallucina-

tions varies with developmental age and degree of certainty required. In order

to assess the significance of these self-reported symptoms we need to know

howmany of theseadolescentswill turn out to have psychotic-type symptoms

on clinical interview. Weadministered aself-reportquestionnaire comprising

sevenquestions assessing psychotic-type symptoms (rated as‘yes’, ‘maybe’ or

‘no’) to 334 11–13 yearold school children attending primary schools in Dub-

lin.Thirty-eightpercentofthe sample answered ‘yes’ or ‘maybe’to at leastone

ofthese symptoms. With regard to auditory hallucinationsspecifically:15.6%

ofthe sampleanswered ‘yes’ and 20.1% answered‘maybe’tothisquestion.We

interviewedadolescentswhoanswered‘yes’ to2ormore questions(n= 22)and

a comparison group of adolescents who endorsed one or none of these symp-

toms (n = 22), using the semi-structured K-SADS diagnostic interview sched-

ule administered by a trained psychologist or psychiatrist. Subjects were

interviewed within 1–2 months of completing the self-report questionnaire.

Symptoms were rated as ‘definite’ and ‘possible’ at a consensus meeting in-

volving 3 clinicians. On interview, 17 adolescents were found to experience

‘definite’ psychotic symptoms and 5 experienced ‘possible’ psychotic symp-

toms. We found that answering ‘yes’ to the auditory hallucination question

(Have you ever heard voices or sounds that no-one else can hear?) on the

screener questionnaire was highly predictive of ‘definite’ clinician-rated au-

ditory hallucinations on interview (PPV 92.3%; NPV 95.7%; Sensitivity

92.3%; Specificity95.7%)and was alsohighlypredictiveof any clinician-rated

‘definite’ psychotic symptom on interview: (PPV 100%; NPV 84.6%; Sensi-

tivity 76.4%; Specificity 100%). Answering ‘maybe’ to the auditory halluci-

nations question was not predictive of clinician-ratedpsychotic symptoms on

interview.Our resultsindicate that adolescentsarewilling to reveal psychotic-

like symptoms using questionnaire methods and that the answers can be con-

firmed by clinical interview. In this age-group self-report of hearing voices is

the most useful predictor of psychotic symptoms on subsequent interview.

This research was supported by the Health Research Board (Ireland).

ID: 550353

STRUCTURAL BRAIN ABNORMALITIES IN

PATIENTS WITH SCHIZOPHRENIA AFTER 2ND

TRIMESTER FETAL EXPOSURE TO IL-8

Lauren M. Ellman

, D. F. Raymond

, S. Vinogradov

W. S. Kremen

, J. H. Poole

4,2

, D. M. Kern

, C. A. Schaefer

A. S. Brown

Psychiatry, College of Physicians and Surgeons of Columbia

University, New York, NY, USA;

Psychiatry, University of

California, San Francisco, San Francisco, CA, USA;

Psychiatry,

University of California, San Diego, San Diego, CA, USA;

Neuropsychology, Defense and Veterans Brain Injury Center,

Veterans Affairs Health Care System, Palo Alto, Palo Alto, CA,

USA;

Division of Research, Kaiser Permanente, Oakland, CA, USA

Maternal infection during pregnancy has been repeatedly associated with

increased risk for schizophrenia. Nevertheless, most viruses do not cross the

placenta; therefore the damaging effects on the fetus appear to be related to

maternal antiviral responses to infection, such as proteins linked to inflam-

mation (ie, proinflammatory cytokines). In the present cohort, fetal expo-

sure to increases in maternal levels of the proinflammatory cytokine

interleukin-8 (IL-8) was related to increased risk of schizophrenia spectrum

disorders (SSD). The purpose of the present study was to determine

whether fetal exposure to IL-8 led to structural brain abnormalities among

SSD patients. Methods: Subjects were 17 cases diagnosed with SSD and 8

controls from a birth cohort study, the Developmental Insult and Brain

Anomaly in Schizophrenia (DIBS) study. Psychiatric morbidity was deter-

mined through semi-structured interviews (DIGS) and medical records re-

view. Levels of IL-8 were ascertained from sandwich enzyme-linked

immunosorbent assays from archived prenatal sera collected during the

2nd trimester of pregnancy. Coronal T1-weighted magnetic resonance

images were acquired using a 1.5-Tesla Siemens system and 3D MP-

RAGE sequences. Volumes were determined through manually tracing

regions of interest (ROIs). ROIs were divided by intracranial volumes to

form ratios and then log-transformed. Results: Results indicated that fetal

exposure to increases in IL-8 were associated with significant increases in

ventricular cerebrospinal fluid (CSF) and significant decreases in right cau-

date, left entorhinal cortex, and right posterior cingulate volumes. In con-

trast, there were no significant effects of IL-8 on any ROIs for controls.

Conclusion: Results indicate that cases are preferentially sensitive to fetal

exposure to IL-8 compared to controls. Among cases, fetal exposure to IL-8

appears to affect dopamine-rich structures implicated in motor functioning

and memory, cognitive domains commonly disrupted among schizophrenia

patients. Fetal exposure to increases in IL-8 also was related to the most

replicated brain anomaly in schizophrenia, increases in ventricular CSF.

Overall, these findings suggest that there is likely a genetic or environmental

factor associated with schizophrenia that renders the aforementioned fetal

brain structures especially sensitive to maternal inflammatory immune

responses.

ID: 550332

THE DEVELOPMENTAL TRAJECTORY TO

SCHIZOPHRENIA—EVIDENCE FOR GENETIC AND

ENVIRONMENTAL INFLUENCES

Mary Catherine Clarke

, A. Tanskanen

, M. O. Huttunen

R. Murray

, M. Cannon

Psychiatry, The Royal College of Surgeons in Ireland, Dublin,

Ireland;

Mental Health and Alcohol Research, National Public

Health Institute, Helsinki, Finland;

Division of Psychological

Medicine, Institute of Psychiatry, London, United Kingdom

Childhood developmental abnormalities have frequently been described in

schizophrenia. We wished to determine whether childhood developmental

abnormality in schizophrenia is genetically mediated. We also wished to

elucidate the relationship between obstetric complications and develop-

mental deficits. The study population comprised all those born in Helsinki

between 1962 and 1969 who had developmental records archived in the Hel-

sinki City Archives. Through linking the Finnish Population Register and

the Finnish Hospital Discharge Register we identified 189 individuals who

had received a diagnosis of schizophrenia and 115 unaffected siblings who

were born between 1962 and 1969 and had developmental information

available. We also identified a control group, matched to cases and siblings

on gender and year of birth (n = 304). Prospective, standardised develop-

mental data for each subject was retrieved from the archives of ‘Well Baby’

clinics. The majority of children in Finland attend these clinics for monthly

check-ups in their first year of life. The age in months at which the following

series of motor developmental milestones were achieved significantly dif-

fered between cases, sibling high-risk and control groups: turns-over, raises

into a sitting position, raises to stand, walks supported and pincer grip.

Cases were significantly slower to reach these milestones compared to sib-

lings, who in turn were significantly slower to achieve them compared to

matched controls. Apgar scores were significantly lower among the cases

than the siblings or comparison groups. There was also an additive

effect between low apgar score and motor developmental delay such

International Congress on Schizophrenia Research

70 6. 6. Epidemiology

that exposure to both risk factors increased risk for schizophrenia to a sig-

nificantly greater extent than exposure to only one of these risk factors.

These findings support delayed motor development in the first year of

life as an endophenotype for schizophrenia, indicate that motor deficits

may be a genetic marker of vulnerability to schizophrenia and suggests

that obstetric complications may additively interact with this endopheno-

type. Funding: The Wellcome Trust, NARSAD and The Health Research

Board of Ireland.

ID: 550298

PSYCHIATRIC OUTCOMES AMONGST OFFSPRING

WITH PARENTAL HISTORY OF MENTAL

DISORDER

Kimberlie Dean

, H. Stevens

, P. B. Mortensen

, R. M. Murray

E. Walsh

, C. B. Pedersen

Forensic Mental Health Science, Institute of Psychiatry, Kings

College London, London, United Kingdom;

Division of Psycho-

logical Medicine and Psychiatry, Institute of Psychiatry, Kings

College London, London, United Kingdom;

National Centre for

Whilst parent-offspring associations for specific mental disorders are well

established, knowledge of the broader range of psychiatric outcomes

amongst offspring with parental history of serious mental illness (SMI)

is not. The aim of the current study was to examine the full range of po-

tential mental health outcomes occurring in a large national cohort,

amongst offspring of parents with SMI compared to the general popula-

tion. The cohort was obtained from linked Danish national registers

and consisted of all offspring born in Denmark between1980 and 1992. Off-

spring were followed from their 14th birthday for the development of

a range of psychiatric disorders (including schizophrenia, bipolar disorder,

schizoaffective disorder, affective disorder, anxiety and somatoform disor-

ders, personality disorder, substance abuse, and mental retardation) based

on outpatient and inpatient hospital data. Offspring exposure groups were

defined on the basis of parental history of mental illness and compared to

those offspring without any parental history. Poisson regression modeling

was used to obtain rate ratios for each exposure/outcome pair with account

taken of offspring gender, age and calendar year. Parental SMI (in one or

both parents) was found to be positively associated with all of the offspring

psychiatric outcomes examined. Offspring with parents (one or both) with-

out SMI but with a history of other psychiatric illness were also found to be

at increased risk of developing any of the range of psychiatric outcomes.

The strongest associations were found between history of parental illness

in both parents and offspring illnesses. Offspring of two parents with SMI

were almost 19 times more likely to develop schizophrenia for example. The

elevated risks were not confined to concordant parent-offspring associa-

tions (eg, offspring of two SMI parents were also 9 times more likely to

develop substance misuse disorders). The gender of the affected parent

had only limited impact on offspring risk of illness. The impact of parental

history of serious mental illness is not confined to elevated offspring risk of

concordant disorders. Offspring are at increased risk of a wide range of

mental disorders. This implies an important role for environmental factors

and genetic factors giving rise to broad, as well as specific vulnerabilities, in

the aetiology of familial risk.

ID: 550285

SCREENING FOR SUBCLINICAL PSYCHOSIS IN

THE GENERAL ADOLESCENT POPULATION

Ian Kelleher, M. Harley, A. Murtagh, M. Blanchard, M. Cannon

Psychiatry, Royal College of Surgeons in Ireland, Dublin 9, Ireland

The aim of this study was to design a psychometric instrument with good

sensitivity and specificity for the identification of subclinical psychotic

symptoms in the general adolescent population. A screening questionnaire

with seven questions designed to assess psychotic symptoms was adminis-

tered to 334 children aged 11–13 years. (1) ‘‘Some people believe that their

thoughts can be read by another person. Have other people ever read your

mind?’’; (2) ‘‘Have you ever had messages sent just to you through TV or

radio?’’; (3) ‘‘Have you ever thought that people are following or spying on

you?’’; (4) ‘‘Have you ever heard voices or sounds that no one else can

hear?’’; (5) ‘‘Have you ever felt you were under the control of some special

power?’’; (6) ‘‘Have you ever seen things that other people could not see?’’;

(7) ‘‘Have you ever felt like you had extra-special powers?’’. Answers were

scored as follows: No = 0 points; Maybe = 0.5 points; Yes = 1 point. Using

stratified random sampling we called adolescents with scores greater than

or equal to 2 on the screener for clinical interview. We also called a com-

parison sample of adolescents who scored 0 or 1 on the screener for inter-

view. In total, 44 adolescents were interviewed using the K-SADS interview

schedule by a trained clinical interviewer (psychiatrist or psychologist). The

results of these interviews were used to conduct sensitivity and specificity

analyses for the questionnaire in the identification of psychotic symptoms.

Twenty two participants reported no psychotic symptoms at interview, 17

reported ‘definite’ psychotic symptoms, and 5 reported ‘possible’ psychotic

symptoms. Symptoms were rated as ‘definite’ or ‘possible’ by a consensus

meeting after the interview. Three questions from the screening question-

naire (questions 3, 4 and 6) proved to have good predictive power for the

detection of ‘definite’ psychotic symptoms on clinical interview. Table 1

shows that endorsement of these 3 core questions provided good predictive

validity for subclinical psychotic symptoms. Our results suggest that a sim-

ple 3-item pen and paper questionnaire could be used to accurately screen

for subclinical psychotic symptoms in early adolescence.

Table. Predictive validity of three core questions for psychotic symptoms

PPV NPV Sensitivity Speciﬁcity

Endorsement of at least one question 64% 95% 94% 67%

Endorsement of at least two questions 88% 89% 93% 82%

Endorsement of all three questions 100% 82% 65% 100%

ID: 550275

UNEMPLOYMENT AND RISK OF PSYCHOSIS IN

BLACK AND MINORITY ETHNIC (BME) GROUPS

Rina Dutta, R. M. Murray, J. Boydell

Psychological Medicine and Psychiatry, Institute of Psychiatry,

London, United Kingdom

Introduction: Unemployment is closely related to both relative deprivation

and social drift of those predisposed to psychosis. Within a stress-vulner-

ability model of psychosis, unemployment would be plausible as a risk fac-

tor as well as early consequence of disease onset. However it has hardly been

studied as an epidemiological risk factor for psychosis. Some black and mi-

nority ethnic groups with high rates of psychosis also have high rates of

unemployment. The purpose of this study was to test the hypotheses

that (i) the incidence of psychosis is higher in unemployed people and

(ii) unemployed people from BME groups have a higher incidence of psy-

chosis than unemployed White people Method: All patients who presented

for the first time with an ICD-10 diagnosis of any psychotic illness, in a de-

fined area of southeast London from 1998–2004 were identified. Electoral

ward level data from 2001 census was used to calculate crude and stand-

ardised incidence rates of psychosis amongst unemployed people, by eth-

nicity (indirectly standardised to (a) the total employed population and (b)

the employed white population, stratifying for age and gender). Standar-

dised incidence ratios were calculated for each ethnic group. Results: Of

International Congress on Schizophrenia Research

6. 6. Epidemiology 71

the 178 patients with psychosis, 100 cases occurred amongst employed

people and 78 amongst the unemployed. The incidence of psychosis

was very high amongst unemployed people and extremely high amongst

Black Caribbean and Black African people. 3.4% (95% CI = 2.2–5.0%) of

unemployed Black Caribbean people become psychotic per year and

2.5% (95% CI = 1.6–3.9%) of Black African people. Standardised to

the employed population of the area, White unemployed had an almost

5-fold increased risk of psychosis (SIR 4.96; 95% CI = 2.71–8.32)

compared to a 25-fold increase for Black Caribbean (SIR 24.9; 95%

CI = 16.2–36.4) and more than 18-fold increase for Black Africans

(SIR 18.7; 95% CI = 11.9–28.1). More markedly, when standardised

to the employed White population of the area, White unemployed had

an almost 12-fold increased risk of psychosis (SIR 11.8; 95% CI =

6.4–19.7), compared to a 60-fold increase in Black Caribbeans (SIR

60.1; 95% CI = 39.3–88.1) and more than 40-fold increase in Black

Africans (SIR 40.7; 95% CI = 25.8–61.1). Conclusions: The incidence

of psychosis is higher overall in people who are unemployed. Unem-

ployed people from BME groups have a much higher incidence of psy-

chosis than White people.

ID: 550260

HEARING VOICES IN CHILDHOOD: A

PREVALENCE AND CASE-CONTROL STUDY OF

AUDITORY HALLUCINATIONS IN MIDDLE

CHILDHOOD

Jack A. Jenner

, A. A. Bartels-Velthuis

, G. van de Willige

J. van Os

2,3

University Center for Psychiatry, University Medical Center

Groningen, Groningen, Netherlands;

Department of Psychiatry,

Maastricht University, Maastricht, Netherlands;

Division of

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom

Introduction: Voice hearing does occur in middle childhood, but little is

known about its prevalence, etiology and consequences. Our aim is to in-

vestigate prevalence and characteristics of auditory vocal hallucinations

(AVH) in 7-and-8-year-olds, as well as associations with pre- and perinatal

influences, early development and current behaviour problems. Methods:

AVH were assessed with the Auditory Vocal Hallucination Rating Scale in

3870 children; prospectively recorded data on pre- and perinatal complica-

tions, early development and current problem behaviour were analyzed in

children with AVH and matched controls. Results: The one-year preva-

lence of AVH was 9%. 15% of AVH children reported severe suffering

and AVH elicited more anxiety in girls. The prevalence for AVH was

higher in rural areas but AVH were more severe, more often externally

attributed and of greater functional impact in the urban environment.

There was some evidence for an association with prenatal maternal in-

fection and slower motor development in early life. Only weak associa-

tions were apparent between AVH and current problem behaviour.

Conclusions: AVH in 7/8 year olds are prevalent but of functional con-

sequence in only a minority. Nevertheless, there may be continuity with

more severe psychotic outcomes given severe suffering in a subgroup,

associations with early developmental deviance and evidence for a poorer

prognosis in an urban environment.

References

1. Jenner JA and Van de Willige G. The Auditory Vocal Hallucination

Rating Scale (AVHRS). Groningen: University Medical Center Gro-

ningen, University Center for Psychiatry, University of Groningen;

2002.

2. Poulton R, Caspi A, Moffitt TE, Cannon M, Murray R, and Harring-

ton H. Children’s self-reported psychotic symptoms and adult schizo-

phreniform disorder: a 15-year longitudinal study. Archives of

General Psychiatry, 2000;57:1053–1058.

3. Scott J, McNeill Y, Cavanagh J, Cannon M, and Murray R. Exposure

to obstetric complications and subsequent development of bipolar dis-

order: Systematic review. British Journal of Psychiatry. 2006;189:3–11.

4. Van Os J, Hanssen M, Bijl RV, and Ravelli A. Strauss (1969) revisited:

a psychosis continuum in the general population? Schizophrenia Re-

search. 2000;45:11–20.

ID: 550233

PSYCHOLOGICAL SCALES AS VULNERABILITY

INDICATORS FOR SCHIZOPHRENIA

Jouko Miettunen

, J. Veijola

1,2

, M. Isohanni

, T. Paunio

N. Freimer

,E.Ja

skela

inen

, A. Taanila

5,6

, J. Ekelund

3,7

M. R. Ja

rvelin

5,8

, L. Peltonen

3,9

, M. Joukamaa

10,11

D. Lichtermann

Department of Psychiatry, University of Oulu, Oulu, Finland;

Academy of Finland, Helsinki, Finland;

Department of Molecular

Medicine, National Public Health Institute, Helsinki, Finland;

Center for Neurobehavioral Genetics, Semen Institute for Neuro-

science and Human Behavior, UCLA, Los Angeles, CA, USA;

Institute of Health Sciences, University of Oulu, Oulu, Finland;

Unit of General Practice, Oulu University Hospital, Oulu, Finland;

Department of Mental Health and Alcohol Research, National

Public Health Institute, Helsinki, Finland;

Department of

Epidemiology and Public Health, Imperial College, London, United

Kingdom;

Human Genetics Section, Wellcome Trust Sanger

Institute, Cambridge, United Kingdom;

Social Psychiatry Unit,

Tampere School of Public Health, University of Tampere, Tampere,

Finland;

Department of Psychiatry, Tampere University Hospital,

Tampere, Finland;

Methadone maintenance clinic ‘‘Cafe

Ersatz’’,

Bonn, Germany

We study the predictive power and associations of several schizotypal and

temperament scales with respect to familial risk and developing of schizo-

phrenic psychoses in a population-based birth cohort. Temperament and

Character Inventory, Physical Anhedonia Scale, Social Anhedonia Scale,

Perceptual Aberration Scale, Hypomanic Personality Scale, Bipolar II

Scale, and Schizoidia Scale were included in the 31-year follow-up survey

of the prospective Northern Finland 1966 Birth Cohort (N = 4,926). We

compared subjects without any previous hospitalisations (N = 4,786) to

those with previous hospital diagnoses for schizophrenia (N = 39) (concur-

rent validity) and to those who in the eight year long follow-up were hos-

pitalised due to schizophrenic psychosis (N = 12) (predictive validity). We

also compared the subjects with schizophrenic psychoses and subjects with

other psychiatric disorders (discriminant validity, N = 101 for concurrent

and N = 72 for predictive analyses). Familial risk was estimated through

parental diagnoses of psychosis in hospital discharge register data (1972

to 2005). The Perceptual Aberration Scale was the best scale for concurrent

(Effect Size, d = 1.89) and discriminant validity (d = 0.64). Subjects having

a high score in Hypomanic Personality Scale (OR 10.72; 95% CI = 2.87–

40.06) were in the highest risk for developing schizophrenic psychoses.

High harm avoidance (d = 1.14 for concurrent cases) was a dominant tem-

perament characteristic in schizophrenia when compared to controls.

Controls with familial risk differed mostly from those without familial

risk by having low physical anhedonia (d = 0.20) and high hypomania

(d = 0.18). Subjects with schizophrenic psychoses differed in most of the

scales from healthy controls and in a few from subjects with other psychi-

atric disorders. Many of the scales were useful predictors of developing

schizophrenic psychoses; however scales were not very diagnosis specific.

These scales are probably not useful as screening instruments but some

(eg, Perceptual Aberration Scale) can be used as intermediate phenotypes

when studying schizophrenia and related psychoses.

ID: 550223

International Congress on Schizophrenia Research

72 6. 6. Epidemiology

ADVANCED PATERNAL AGE IS ASSOCIATED

WITH IMPAIRED NEUROCOGNITIVE OUTCOMES

DURING INFANCY AND CHILDHOOD IN A

GENERAL POPULATION BIRTH COHORT.

John Joseph McGrath

1,4

, A. Barnett

, C. Foldi

, T. H. Burne

1,4

D. W. Eyles

1,4

, S. Buka

Queensland Centre for Mental Health Research, University of

Queensland, Wacol, QLD, Australia;

Institute of Health and

Biomedical Innovation and School of Public Health, Queensland

University of Technology, Brisbane, QLD, Australia;

Department

of Community Healt, Brown University, Providence, RI, USA;

Queensland Brain Institute, University of Queensland, Brisbane,

QLD, Australia

Background: Advanced paternal age (APA) is associated with an increased

risk of neurodevelopmental disorders such as autism and schizophrenia, as

well as with dyslexia and impaired intelligence. The aim of the study was to

examine the relationship between paternal age and neurocognitive measures

of children using a large birth cohort. Method: A sample of white singleton

children (n = 25,131) was drawn from the US Collaborative Perinatal Project.

The outcome measures were assessed at 8 months, 4 and 7 years (Bayley

scales, Stanford Binet IQ, Graham-Ernhart test, Wechsler Intelligence Scale

for Children, Wide Range Achievement Test). The main analyses examined

the relationship between neurocognitive measures and paternal or maternal

age when adjusted for potential confounding factors. Results: Advanced

paternal age showed significant associations with poorer scores on all meas-

ures of the neurocognitive measures apart from Stanford Binet Intelligence

Scale. The findings were consistent in direction and effect size at all three

ages. In contrast, advanced maternal age was generally associated with better

scores on these same measures. Discussion: The offspring of older fathers

show subtle impairments on tests of neurocognitive ability during infancy

and childhood. In light of secular trends related to delayed fatherhood,

the mechanisms underlying these findings warrant closer scrutiny.

ID: 550204

NEURODEVELOPMENTAL RISK FACTORS FOR

SCHIZOPHRENIA: POPULATION-BASED COHORT

STUDIES

Tyrone D. Cannon

, L. Ellman

, A. Brown

, S. Buka

, R. Yolken

F. Torrey

Psychology and Psychiatry and Biobehavioral Sciences, UCLA,

Los Angeles, CA, USA;

Department of Psychiatry, Columbia

University, New York, NY, USA;

Department of Epidemiology,

Brown University, Providence, RI, USA;

Neurobiology, Johns

Hopkins University, Baltimore, MD, USA;

Stanley Foundation

Medical Research Institute, Chevy Chase, MD, USA

Inspired in part by the pioneering work of Barbara Fish and Sarnoff Med-

nick, a number of studies have observed that obstetric complications are

associated with increased risk for schizophrenia later in life. Such factors

are also related to increased severity of certain neuropathological features

of schizophrenia, including hippocampal and cortical gray matter reduc-

tion. However, it remains to be determined whether such complications re-

sult from the schizophrenia genotype (as in G-E covariation) or interact

with genetic susceptibility to the disorder in disease expression. We con-

ducted searches of the Finnish National Hospital Discharge and Perinatal

Registers to ascertain pregnancy and birth information on 6471 offspring of

parents with schizophrenia in the country of Finland, compared with 25 884

gender matched controls born on the same day and in the same obstetric

units as the case offspring. A number of pregnancy and birth complications

were found to be elevated in the offspring with schizophrenic mothers.

However, there were no such increases among offspring of schizophrenic

fathers. These results are consistent with the view that the occurrence of

such complications is independent of genetic predisposition to schizophre-

nia, but may be influenced by maternal health risk behaviors, such as smok-

ing. In another study, we examined whether neurotrophic factors, which are

stimulated as part of a neuroprotective response to fetal distress, are dif-

ferentially expressed in cord blood samples at the time of birth following

particular obstetric stressors among individuals who developed schizophre-

nia as adults, as compared with demographically matched controls. This

study utilized a nested case-control design including 111 cases with psy-

chotic disorders (70 with schizophrenia) and 333 controls matched for gen-

der, race, and date of birth, drawn from the Philadelphia cohort of the

National Collaborative Perinatal Project. Among controls, birth asphyxia

was associated with a significant (10%) increase in BDNF in cord samples,

while among cases, asphyxia was associated with a significant (20%) de-

crease in BDNF. These findings provide serologically based prospective ev-

idence of disrupted neurotrophic signaling in response to birth asphyxia in

the molecular pathogenesis of schizophrenia and encourage search for

genes that confer heightened susceptibility via disruptions in BNDF.

ID: 550055

LEPTIN AND ADIPONECTIN CONCENTRATIONS

IN ANTIPSYCHOTIC-NAI

VE PATIENTS WITH

NONAFFECTIVE PSYCHOSIS

Wesley Seabolt

, C. Garcia-Rizo

, E. Fernandez-Egea

M. Bernardo

2,3

, B. Kirkpatrick

Department of Psychiatry and Health Behavior, Medical College of

Georgia, Augusta, GA, USA;

Schizophrenia Program, Department

of Psychiatry, Hospital Clinic, Barcelona, Spain;

Institute of

Biomedical Research Agusti Pi i Sunyer (IDIBAPS), Barcelona,

Spain;

Department of Psychiatry, University of Cambridge, CB2

0QQ Cambridge, United Kingdom

Some evidence suggests newly diagnosed, antipsychotic-naı

ve patients with

nonaffective psychosis have an increased risk of diabetes. Diabetes is asso-

ciated with abnormal circulating concentrations of the adipokines leptin

and adiponectin. We measured leptin and adinponectin in 32 fasting, newly

diagnosed, antipsychotic-naı

ve patients with nonaffective psychosis and

control subjects (N = 53). The patients had a significant increase in leptin

concentrations (P > .001 for the variable of patients vs. control subjects)

after accounting for the variance due to age, gender, socioeconomic status,

body mass index, smoking, cortisol, and aerobic conditioning (as measured

by resting heart rate); the two groups were very similar with regard to race

and neighborhood of residence so these two variables were not included in

the multivariate analysis. Using the same variables, we found no significant

patient/control difference in adiponectin. The difference in leptin, which

could not be accounted for by a difference in body mass index, coupled

with the evidence for an increased risk of diabetes, raises the possibility

that patients with schizophrenia have increased visceral fat.

ID: 550050

THE FUTURE OF HIGH RISK RESEARCH

Assen V. Jablensky

Centre for Clinical Research in Neuropsychiatry / School of

Psychiatry and Clinical Neurosciences, The University of Western

Australia, Perth, WA, Australia

During the five decades since Dr Barbara Fish initiated her groundbreaking

research into the neurobiological antecedents of schizophrenia in children,

International Congress on Schizophrenia Research

6. 6. Epidemiology 73

an impressive volume of literature spanning epidemiology, developmental

neurology, cognitive psychology, neurophysiology and behavioural genet-

ics has brought forth novel insights into the nature of ‘‘high risk’’. Recent

research has not only vindicated the original evidence for an inherited neu-

rointegrative vulnerability and the syndrome of pandysmaturation, but has

by extension contributed to a more general understanding of schizophrenia

as a basically neurodevelopmental disorder. Yet important gaps in knowl-

edge remain. Neurobehavioural and mental disorders other than schizo-

phrenia have until recently been less incisively investigated, and the

study of gene-environment interactions and epigenetic regulation (both pre-

dicted by Dr Barbara Fish) is only now beginning to gather pace. The future

of this area of research is bright and full of hope: the working alliance be-

tween developmental and clinical neuroscience, epidemiology and genetics

is likely to unravel many of the present complexities and ultimately lead to

practical means of risk containment and reduction.

ID: 549620

GENETIC AND ENVIRONMENTAL INFLUENCES

ON THE OVERLAP BETWEEN PREMORBID IQ,

PREMORBID SOCIAL ADJUSTMENT AND

SCHIZOPHRENIA: A POPULATION-BASED TWIN

AND SIBLING STUDY

Abraham Reichenberg

, M. Weiser

, A. Taylor

, J. Rabinowitz

G. Lubin

, E. Kravitz

, J. M. Silverman

, D. A. Greenberg

A. S. David

, R. M. Murray

, M. Davidson

Psychological Medicine and Psychiatry, Institute Of Psychiatry,

London, United Kingdom;

Psychiatry, Sheba Medical Center, Tel

Hashomer, Israel;

Social, Genetic and Developmental Psychiatry,

Institute of Psychiatry, London, United Kingdom;

Mental Health,

Israeli Defence Force, Tel Aviv, Israel;

Social Work, Bar Ilan

University, Ramat Gan, Israel;

Psychiatry, Mount Sinai School of

Medicine, New York, NY, USA;

Biostatistics, Columbia Univer-

sity, New York, NY, USA

Background: Low premorbid IQ and abnormal premorbid social adjust-

ment are consistent risk factors for schizophrenia in both prospective

and cross-sectional studies. There are many potential explanations for

the correlation between premorbid functioning and schizophrenia. As

far as we know, no previous study has used the twin method to test

competing explanations for the relation. In this study we used data from

a population of twins and siblings in the Israeli Conscripts Cohort to ex-

amine the extent to which the covariation between premorbid functioning

and schizophrenia is due to genetic overlap or environmental effects. Meth-

ods: Using the Israeli Conscripts Cohort, we ascertained 7321 pairs of

same-sex monozygotic and dizygotic twins, and 251 500 pairs of same-

sex siblings with information on intellectual and behavioral functioning

at age 17. Linkage to a national psychiatric hospitalization registry iden-

tified 58 twin pairs discordant for schizophrenia, and 5 concordant twin

pairs. There were 2234 discordant siblings-pairs, and 71 concordant sib-

lings-pairs. Results: A significant phenotypic correlation (r = .15) was ob-

served between premorbid IQ and schizophrenia. Genetic factors were the

main source of this correlation. Premorbid processing speed ability had the

greatest genetic correlation with schizophrenia, followed by non-verbal rea-

soning ability. Both genetic and environmental factors contributed to the

premorbid social adjustment <–> schizophrenia relationship (r = .14 and

r = .06, respectively). Conclusions: Findings suggest that specific candi-

date genes and neurobiological processes should be tested in relation to

both premorbid functioning and schizophrenia phenotypes. Further study

of the potential mechanisms through which environmental factors confers

risk for both premorbid functioning and schizophrenia is needed.

ID: 549618

SELF-REPORTED PSYCHOTIC SYMPTOMS

IN THE COMMUNITY, AND RISK OF LATER

HOSPITALIZATION FOR NON-AFFECTIVE

PSYCHOTIC DISORDERS

Mark Weiser

, N. Werbeloff

, M. Drukker

, J. Van Os

B. P. Dohrenwend

, R. Yoffe

, I. Levav

, M. Davidson

Psychiatry, Sheba Medical Center, Tel Hashomer, Ramat Gan,

Israel;

Psychiatry and Neuropsychology, Maastricht University,

Maastricht, Netherlands;

Psychiatry, Columbia University, New

York, NY, USA;

Division of Mental Health Services, Ministry of

Health, Jerusalem, Israel

Research suggests that low-grade psychotic experiences in the general pop-

ulation are a common phenomenon, and previous studies indicate that per-

sons experiencing low grade psychotic experiences are at increased risk for

suffering from clinically diagnosed psychotic disorders. The aim of the cur-

rent study is to assess the predictive validity of self-reported psychotic

symptoms on risk for later psychosis, and to identify the effect of poor ed-

ucation achievements (less then 12 years of formal education) or cannabis

abuse on risk for later hospitalization, in those persons experiencing psy-

chotic symptoms. This study utilizes data from an epidemiological study of

young adults aged 25–34 conducted in the 1980#s in Israel. A random sam-

ple of 4914 subjects were screened for psychiatric symptoms using a stan-

dardized clinical interview. The interview included questions regarding

psychotic symptoms (delusions, hallucinations, thought broadcasting,

etc). Data on later psychiatric hospitalization was ascertained in 2008 using

a psychiatric hospitalization registry, enabling up to 25 years follow-up.

Similar to data from previous studies, 25% of the subjects interviewed

reported having experienced one or more psychotic symptoms during their

lifetime. Self-reported psychotic symptoms increased risk of later hospital-

ization for psychosis (OR = 3.04, 95% CI: 1.19–7.76). In terms of popula-

tion attributable risk, 34% of hospital admissions for psychotic disorders

could be attributed to the presence of psychotic symptoms at baseline. In

individuals with psychotic symptoms, poor educational achievements or

cannabis abuse did not further increase risk for later hospitalization. In

individuals aged 25–34, a quarter of the population suffers from psychotic

symptoms, and these symptoms are associated with increased risk of later

hospitalization for a psychotic disorder. These data suggest that screening

for psychotic symptoms might enable early identification and intervention,

hence decreasing hospitalizations for psychotic illness. Lower educational

attainment or a history of drug abuse did not increase risk for later hospi-

tailzation among those with psychotic symptoms.

ID: 549612

BIRTH WEIGHT AND RISK OF SCHIZOPHRENIA OR

OTHER MENTAL DISORDER: IS IT CONFINED TO

SMALL BABIES?

Kathryn Mary Francis Abel

, S. Wicks

, E. Susser

, C. Dalman

M. G. Pedersen

, P. B. Mortensen

, R. T. Webb

1,5

Centre for Women’s Mental Health, University of Manchester,

Manchester, United Kingdom;

Psychiatric Epidemiology, Depart-

ment of Public Health Sciences, Karolinska Institutet, Stockholm,

Sweden;

Department of Epidemiology, Mailman School of Public

Health, University of Columbia and New York State Psychiatric

Hospital, New York, NY, USA;

National Centre for Register-based

Research, NCRR, University of A

rhus, A

rhus, Denmark;

Health

Methodology Research Group, University of Manchester,

Manchester, United Kingdom

Studies have shown a link between low birthweight and adult mental illness,

but have focused on the WHO threshold of 2.5kg, and on schizophrenia as

International Congress on Schizophrenia Research

74 6. 6. Epidemiology

the sole outcome. Our objectives were to examine whether: (i) increased risk

for adult schizophrenia is specifically associated with birthweight <2.5kg

rather than being spread across the birthweight range; (ii) excess risk is con-

fined to schizophrenia, or also linked with other adult mental illness. We

used a population-based cohort design using data from two adjacent Scan-

dinavian countries: all singleton live births in Sweden during 1973–1985 and

in Denmark 1979–1986 (N = 1.64 million in total). These births were linked

to comprehensive national registers of inpatient psychiatric treatment, with

follow-up to December 31, 2002 in Sweden and June 30, 2005 in Denmark.

Our main outcome measures were the odds of birthweight less than 4000–

4499g in consecutive 500g strata down to 500–1499g among people admit-

ted with any psychiatric diagnosis, schizophrenia or affective disorder ver-

sus the general population. For all diagnostic groups combined, there was

a clear trend in increasing odds ratios with decreasing birthweight across

the birthweight range: 3500–3999g OR = 1.1; 3000–3499g OR = 1.2; 2500–

2999g OR = 1.5; 2000–2499g OR = 1.8; 1500–1999g OR = 2.0; 500–1499g

OR = 2.2. This pattern was mirrored in the two specific diagnostic groups.

We found no indication of specificity of effect with either schizophrenia or

affective disorder. Country of birth, year of birth and sex had a modest

combined confounding effect in the inter-country cohort. In Sweden, socio-

economic status also had little confounding effect, although maternal an-

tenatal smoking was a stronger confounder. Even when the cohort was

restricted to term births, the patterns of birthweight risk across the normal

range persisted. Our findings suggest that the association between birth-

weight and risk of mental disorder occurs across the normal birthweight

range and that there is no evidence of a threshold effect. We do not find

that risk is specific to a particular diagnosis. Findings suggest that the range

of (environmental and other) causes of poor fetal growth interact with more

specific causes of individual diagnoses (genes, postnatal experiences) to en-

able expression of a particular mental disorder. Future prevention strate-

gies might usefully target the majority at medium risk rather than a high

risk minority.

ID: 549598

PREVALENCE OF CELIAC DISEASE AND GLUTEN

SENSITIVITY IN THE CATIE STUDY POPULATION

Nicola Gerardo Cascella

, D. Kryszak

, B. Bhatti

, P. Gregory

D. Kelly

, J. McEvoy

, A. Fasano

, W. W. Eaton

Psychiatry, Johns Hopkins University, Baltimore, MD, USA;

Medicine, University of Maryland, Baltimore, MD, USA;

Psychia-

try, University of Maryland, Baltimore, MD, USA;

Psychiatry, Duke

University, Durham, NC, USA;

Mental Health, Bloomberg School of

Public Health, Johns Hopkins University, Baltimore, MD, USA

Celiac Disease (CD) and schizophrenia (SZ) share approximately overlap-

ping prevalence but epidemiologic data show higher prevalence of CD

among SZ patients. The reason for this higher co-occurrence is yet un-

known but the clinical knowledge about the presence of immunologic

markers for CD or gluten intolerance in SZ patients may have implica-

tions for treatment. We evaluated antibody prevalence to gliadin

(AGA), transglutaminase (tTG) and endomysium (EMA). AGA, tTG

and EMA was assayed in 1410 SZ patients who were part of the Clinical

Antipsychotic Trials of Intervention Effectivness (CATIE) study and 6796

controls. Psychopathology in SZ patients was assessed using the Positive

and Negative Symptoms Scale or PANSS. Chi-square test was applied to

assess for difference in the frequency of IgA AGA and tTG antibodies.

Regression analyses were conducted to predict PANSS scores from

AGA and tTG antibodies adjusting for age, gender, and race. Among

SZ patients 22% had moderate to high levels of IgA AGA compared

to 7.9% of the controls (v

= 1234.6, df = 2, P < .001.) Moderate to

high levels of tTG antibodies were present in 4.7% of SZ patients vs.

2.4% of the controls (v

= 68.9, df =2, P < .001.) Regression analyses failed

to predict PANSS scores from AGA and tTG antibodies. Persons with

schizophrenia have higher titers of antibodies related to celiac disease

and gluten sensitivity. Our findings might have implications for the treat-

ment of SZ patients given that a gluten-free diet may contribute to the

improvement of symptoms.

ID: 549343

COMPLEX GENETIC VARIATION IN BDNF AND

COMT GENES INTERACTS WITH ADOLESCENT

CANNABIS USE IN THE DEVELOPMENT OF

PSYCHOSIS.

Ruud van Winkel

1,2

, G. Kenis

, J. Allardyce

, I. Myin-Germeys

C. Henquet

, M. De Hert

, J. van Os

Maastricht University Medical Centre, Maastricht, Netherlands;

UPC KULeuven, Leuven, Belgium

Interaction between adolescent cannabis use (<16 yrs) and BDNF func-

tional haplotypic variation can be plausibly hypothesized, given the impor-

tance of both the endocannabinoid system and neurothrophins in processes

of neurodevelopment, neuroplasticity and cognition. Furthermore, gene-

environment interaction between adolescent cannabis use and COMT

Val158Met was previously suggested. 2 SNPs within the BDNF gene

were genotyped: the functional Val66Met polymorphism, of which Met-

substitution is associated with reduced intracellular trafficking and activity-

dependent BDNF secretion, and rs988748, which was reported to form

a haplotype with Val66Met. Five polymorphisms in the COMT gene

were also genotyped, four of which were recently described to form three

common functional haplotypes that better index COMT enzymatic activity

than Val158Met alone (rs6269, rs4633, rs4818, Val158Met). rs737865, a

polymorphism associated with schizophrenia in a recent meta-analysis

(Allen et al. Nature Genetics, 2008), was also genotyped. Case-only hap-

lotype analysis in 587 psychotic patients revealed a significantly higher

odds for adolescent cannabis use for the BDNF Met/rs988748_G haplotype

compared to the other BDNF haplotypes (OR 2.04, P = .033) as well as

a trend for a higher odds for the 4-SNP COMT high activity ‘LPS’ haplo-

type compared to the other COMT haplotypes (OR 1.60, P = .055). Re-

markably, the association between COMT LPS haplotype and

adolescent cannabis use was dependent of rs737865 genotype (LPS/

rs737865_T: OR 2.11, P = .016; LPS/rs737865_C: OR 1.25, P = .469),

and there was a trend towards antagonistic BDNF x COMT interaction

(b =0.83, P = .065). The associations between adolescent cannabis use

and BDNF and COMT haplotypes remained significant and similar in

magnitude after controlling for other psychotogenic drugs, and the

gene-by-gene antagonistic interaction now also reached statistical signifi-

cance (b =1.04, P = .048). No significant association was found between

BDNF or COMT haplotypes and cannabis use started after the age of 16, or

with use of any of the other substances. In conclusion, complex haplotypic

variation within BDNF and COMT interacts with adolescent cannabis use

in the development of psychosis, and these haplotypes may act through in-

dependent pathways. Therefore, these findings may serve as a ‘magnifying

glass’ in order to elucidate the pathophysiological mechanisms driving the

psychotogenic effects of cannabis.

ID: 549322

INCREASED PREVALENCE OF ANTIBODIES TO

TOXOPLASMA IN RECENT ONSET PSYCHOSIS

Faith Dickerson

, C. Stallings

, A. Origoni

, A. Sullens

, C. Copp

S. Khushalani

, R. Yolken

Stanley Research Program, Sheppard Pratt, Baltimore, MD, USA;

Pediatrics, Johns Hopkins School of Medicine, Baltimore,

MD, USA

Background: Increased rates of infection with Toxoplasma gondii have

been found in individuals with recent onset psychosis but this issue has

International Congress on Schizophrenia Research

6. 6. Epidemiology 75

not been studied in samples in the United States. Methods: We measured

IgG class antibodies to Toxoplasma gondii in N = 127 individuals with

recent onset psychosis, N = 457 individuals with multi-episode schizophre-

nia, and N = 164 individuals with multi-episode bipolar disorder. We com-

pared serological evidence of Toxoplasma infection in these groups with

that of N = 283 control individuals without a history of psychiatric

disorder. All participants were from central Maryland in the US. Within

the recent onset group, we also determined the correlates of Toxoplasma

seropositivity. Results: There was a higher prevalence of serological evi-

dence of Toxoplasma infection in individuals with recent onset psychosis

(21.3%) than in controls (13.8%) (OR 2.06, 95% CI = 1.1–3.7, P < .020 ad-

justed for age, gender, race, and education). Within the recent onset group,

the prevalence of Toxoplasma seropositivity was highest among individuals

with a diagnosis of schizophrenia (29.6%) compared with bipolar disorder

(18%) or other diagnosis (15.2%). Within the recent onset group, Toxo-

plasma seropositivity was correlated with receipt of olanzapine (OR 3.2,

95% CI = 1.2–9.0, P < .025 adjusted for age, gender, race, education,

and diagnosis). Serological evidence of Toxoplasma infection within the

recent onset group did not differ by age, gender, or race (all P > .1).

The seroprevalence of Toxoplasma in the individuals with multi-episode

schizophrenia and multi-episode bipolar disorder did not differ signifi-

cantly from that of the controls. Conclusions: Our results add to the grow-

ing body of literature indicating that individuals with recent onset psychosis

have an increased rate of exposure to Toxoplasma gondii compared to con-

trols living in the same geographic region. Our findings support a possible

role for Toxoplasma gondii in the etiopathogenesis of schizophrenia and

provide a rationale for trials of anti-Toxoplasma medications in individuals

with schizophrenia.

ID: 549197

ENVIRONMENTAL AND FAMILIAL DETERMI-

NANTS OF ADVERSE NEUROPSYCHIATRIC OUT-

COMES IN HIGH RISK CHILDREN OF MOTHERS

WITH SCHIZOPHRENIA AND OTHER PSYCHOSES

Vera A. Morgan

1,2

, S. Zubrick

, C. Bower

, M. Croft

, G. Valuri

J. Griffith

, A. Jablensky

1,2

Neuropsychiatric Epidemiology Research Unit, University of

Western Australia, School of Psychiatry and Clinical Neurosciences,

Perth, WA, Australia;

Centre for Clinical Research in Neuropsy-

chiatry, University of Western Australia, School of Psychiatry and

Clinical Neurosciences, Perth, WA, Australia;

Telethon Institute

for Child Health Research, Perth, WA, Australia

This study is designed to untangle environmental and familial contribu-

tions to risk of adverse neuropsychiatric outcomes (intellectual disability;

psychiatric morbidity) in genetically high risk children of women with

psychoses. We linked the Statewide midwives register (308 022 births)

and psychiatric case register (79 599 women) to identify 3174 high risk

children born 1980–92 to mothers with schizophrenia, bipolar disorder

or unipolar depression, and a comparison group of 3129 children born

to unaffected mothers. Psychiatric status and cognitive level were deter-

mined for mothers, fathers and children, and data collected on obstetric

complications. Time-to-event data were analysed using Cox regression.

Of 6303 children, 750 had had a psychiatric illness: 18.1% of schizophre-

nia offspring, 20.1% of bipolar offspring, 16.0% of unipolar offspring

and 5.6% of comparison children. Maternal psychosis and obstetric com-

plications were independently associated with risk of psychiatric illness

in offspring. In this young cohort, 57 children had a diagnosis of psy-

chosis: 3.1% of schizophrenia offspring, 1.2% of bipolar offspring,

0.9% of unipolar offspring and 0.4% of comparison children. Maternal

schizophrenia (OR 8.0, CI 3.6–18.2) and bipolar disorder (OR 2.6, CI

1.1–6.0) were associated with risk of psychoses. The effect for obstetric

complications variables was not significant although the odds were el-

evated. There were 129 children with intellectual disability (AAMR cri-

teria). Odds of intellectual disability was increased 3-fold in case

compared to comparison children. Parental cognitive level, maternal

psychosis, paternal psychiatric illness, and labour/delivery complica-

tions were independent risk factors for intellectual disability. Case chil-

dren were more likely than comparison children to have psychiatric

illness co-occurring with intellectual disability (P < .000). In conclusion,

children of mothers with psychoses were at significantly greater risk of

neuropsychiatric outcomes compared to children of unaffected mothers.

Environmental insults and familial liability exerted independent effects

on risk of intellectual disability and of psychiatric illness across all case

groups. Risk of psychoses was differentially distributed but numbers

were small and further analysis is warranted using a dataset under con-

struction of half a million children.

ID: 548805

CHILDHOOD TRAUMA AND INCREASED STRESS-

SENSITIVITY IN PSYCHOSIS: AN EXPERIENCE

SAMPLING STUDY

Inez Myin-Germeys

1,2

, M. Lardinois

, T. Lataster

, J. van Os

Department of Psychiatry, Maastricht University, Maastricht,

Netherlands;

school of Psychological Sciences, Manchester Uni-

versity, Manchester, United Kingdom

Background: Increasing epidemiological evidence suggests that traumatic

experiences in childhood are associated with the development of clinical

and sub-clinical psychosis in adulthood. However, the mechanism through

which trauma leads to an increased risk for psychosis remains unclear. Pre-

vious studies have demonstrated that increased sensitivity to daily life stress

is part of the underlying vulnerability for psychosis. The current study

hypothesises that early trauma increases the risk for psychosis through sen-

sitizing people for the small stresses of daily life. Method: The sample con-

sisted of three groups: patients with a diagnosis of psychotic disorder with

a recent onset (<10 years) (n = 47), their first degree relatives (n = 43) and

healthy control subjects (n = 48). The Experience Sampling Method (ESM;

a structured diary technique) was used to assess stress-reactivity in daily life

defined as (1) emotional reactivity (increase in negative affect) and (2) psy-

chotic reactivity (increase in psychosis intensity) in reaction to stress. Child-

hood trauma was assessed using the Childhood Trauma Questionnaire

(CTQ). Results: Group was significantly associated with trauma level

(F = 8.14, P < .0001), with patients reporting significantly more trauma

compared to the relatives and controls. A history of severe trauma signif-

icantly increased the emotional (B = 0.11 (SE = 0.03), P < .001) and the

psychotic (B = 0.17 (SE = 0.03), P < .001) reaction to stress in the patients.

Conclusions: The results support previous findings that patients with psy-

chosis report increased levels of childhood trauma compared to controls. In

addition, the findings suggest a mechanism of sensitization underlying the

trauma—psychosis association, since a history of childhood trauma results

in increased psychotic en emotional reactions to stress.

ID: 548793

BARBARA FISH’S LANDMARK STUDIES

AND THEIR INFLUENCE ON PARENT-BASED

HIGH-RISK STUDIES DURING THE PAST 50 YEARS

Thomas McNeil

1,2

Psychiatric Epidemiology, Lund University, Lund, Sweden;

Psychiatry, University of Western Australia, Perth, WA, Australia

With the assistance of Barbara Fish herself, the speaker summarizes the

landmark studies of Barbara Fish beginning in the 1950’s, and discusses

their place in the zeitgeist of American psychiatry at that time, their

International Congress on Schizophrenia Research

76 6. 6. Epidemiology

influence on the ensuing multitude of parent-based high-risk studies

beginning in the 1970’s, following the NIMH and Danish studies in

the mid-1960, and their indelible foot prints observed in ongoing high-

risk studies of that type. Fish’s concept of ‘‘pandysmaturation’’ (PDM)

is for example applied to the speaker’s prospective high-risk study

data, within the framework of the now pervasive ‘‘neurodevelopmental’’

model for serious psychopathology.

ID: 548789

PATIENTS WITH FIRST EPISODE PSYCHOSIS

TAKE MORE AND STRONGER CANNABIS THAN

CONTROLS.

Marta M. Di Forti, C. Morgan, M. Valeria, S. Stilo, R. Handley,

S. Luzi, T. Marques, M. Aas, P. Corinne, R. Manuela,

A. Demjaha, B. Wiffen, A. Kolliakou, P. Dazzan, C. Pariante,

J. Powell, R. Murray

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom

Background: Epidemiological studies have reported that the increased risk

of developing psychosis in cannabis users is dose related. In addition,

experimental research has shown that the active constituent of cannabis

responsible for its psychotogenic effect is Delta-9-Tetrahydrocannabinol

(THC) (Murray et al. 2007). Recent evidence has suggested an increased

in potency (% TCH) in the cannabis seized in the UK (Potter et al.

2007). Hypothesis: We predicted that first episode psychosis patients

are more likely to use higher potency cannabis and more frequently

than controls. Methods: We collected information concerning socio-demo-

graphic, clinical characteristics and cannabis use (age at first use, frequency,

length of use, type of cannabis used) from a sample of 191 first-episode

psychosis patients and 120 matched healthy volunteers. All were recruited

as part of the Genetic and Psychosis (GAP) study which studied all patients

who presented to the South London and Maudsley Trust. Results: There

was no significant difference in the life-time prevalence of cannabis use

or age at first use between cases and controls. However, cases were more

likely to be regular users (P = .05), to be current users (P = .04) and to

have smoked cannabis for longer (P = .01). Among cannabis users, 86.8%

of 1st Episode Psychosis Patients preferentially used Skunk/Sinsemilla

compared to 27.7% of Controls. Only 13.2 % of 1st Episode psychosis

Patients chose to use Resin/Hash compared to 76.3% of controls

(OR* = 7.4, 95% CI = 3.4–17.2, Adjusted for age, gender, ethnicity, level

of Education and employment status).The concentration of TCH in

these in South East London, ranges between 8.5 and 14% (Potter et

al. 2007). Controls (47%) were more likely to use Hash (Resin) whose

average TCH concentration is 3.4% (Potter et al. 2007). Conclusions:

Patients with first episode psychosis have smoked higher potency canna-

bis, for longer and with greater frequency, than healthy controls.

ID: 546713

THE SCHIZOPHRENIA PARADOX REVISITED:

LIFETIME REPRODUCTIVE OUTPUT OVER TWO

GENERATIONS IN A SWEDISH COHORT BORN

1915–1929

James Hunter MacCabe

, D. A. Leon

, R. M. Murray

, I. Koupil

Psychiatry, Institute of Psychiatry, King’s College London, Lon-

don, United Kingdom;

Non-communicable Disease Epidemiology

Unit, London School of Hygiene and Tropical Medicine, London,

United Kingdom;

Centre for Health Equity Studies, Stockholm

University, Stockholm, Sweden

Background: Schizophrenia is a highly heritable disorder of early onset,

which severely curtails reproductive output, yet it persists at a relatively

constant population prevalence over time (the schizophrenia paradox).

It has been suggested that the fitness disadvantage associated with schizo-

phrenia might be compensated by increased fitness in unaffected relatives,

otherwise the prevalence of schizophrenia would fall to zero in a few gen-

erations. Little is known about reproductive success in affective psychoses.

To study these questions with a minimum of bias requires longitudinal pop-

ulation data over several generations. Methods: We conducted a historical

cohort study using a Swedish birth cohort of 12 168 individuals born

1915–29, and followed up until 2002. By linking to data from population

registers, we identified patients with schizophrenia (n = 58) or affective

psychosis (n = 153) and their unaffected siblings, and compared their bi-

ological fitness over two generations with that of the population, adjust-

ing for a range of socio-demographic variables from throughout the

lifespan. This is the first study to use prospectively collected data across

the entire reproductive lifespan of a defined population, and to include

data on grandchildren as well as children. Results: There was strong

evidence that patients with schizophrenia had fewer children (fertility

ratio = 0.42 [95% confidence interval (CI) = 0.31–0.56]) than the healthy

population. Much of this reduction was related to low marriage rates in

schizophrenic patients. The unaffected siblings of schizophrenic patients

showed no evidence of any compensatory increase in fitness. There was

a trend towards enhanced fertility among the offspring of schizophrenia

patients, although this was not enough to compensate for the reduced

fertility in patients, so schizophrenic patients still had fewer

grandchildren overall than the healthy population (Fertility Ratio =

0.52 [95% CI = 0.33–0.80]). Patients with affective psychosis did not differ

from the general population on any fertility measure, and neither did their

siblings and offspring. Conclusions: Schizophrenia, but not affective psy-

chosis, is associated with reduced biological fitness. This persists into the

second generation, but with tentative evidence of compensatory increase

in fitness in the offspring of schizophrenic patients.

ID: 543922

EARLY EXPOSURE TO CANNABIS AND RISK

FOR PSYCHOSIS IN YOUNG ADOLESCENTS IN

TRINIDAD

Monique Konings

1,2

Department of Psychiatry and Neuropsychology, Maastricht

University, Maastricht, Netherlands;

Department of psychiatry,

GGZ Eindhoven, Eindhoven, Netherlands

Introduction: Cannabis is worldwide the most frequently used illicit drug,

in particular among young people. Cannabis use increases the risk for psy-

chosis, but these psychotogenic effects of cannabis may be restricted to ex-

posure during early adolescence. Method In a general population based

study, 472 participants (aged 12–23 years) were randomly selected from dif-

ferent schools in Trinidad (Republic of Trinidad and Tobago). Subjects

completed questionnaires on past and current cannabis use and psychotic

symptoms (using the Community Assessment of Psychic Experiences,

a self-report questionnaire for psychotic experiences). Results Cannabis

use increased the risk to experience psychotic symptoms and this effect

was conditional upon early exposure. Exposure before but not after the

median age of 14 years predicted psychotic symptoms (respectively b:

0.71, 95% CI = 0.22; 1.19, P = .004 and b: 0.11, 95% CI = 0.57; 0.36,

P = .66). The developmental effect of cannabis use was independent of

use of other drugs or current use of cannabis. Conclusion Early onset of

cannabis use is associated with a greater risk of developing psychotic symp-

toms than later onset of use. It seems that cannabis early exposure effects

are developmental in nature and that they do not simply reflect effects of

long-term exposure associated with earlier onset of use. Thus, early adoles-

cence may be a critical period with regard to the psychotogenic effect of

cannabis. In addition, this is the first study, to our knowledge, to show psy-

chotogenic effects of early cannabis exposure in a non-western society,

International Congress on Schizophrenia Research

6. 6. Epidemiology 77

where cannabis use has a long, non-stigmatized history for medical and

recreational use.

ID: 540988

GENDER DIFFERENCES OF PERSONS WITH

SCHIZOPHRENIA IN RURAL CHINA

Mao Sheng Ran

, W. J. Mao

, C. P. Tang

, F. R. Lin

,L.Li

S. H. Hu

, C. L. Chan

, E. Y. Chen

, Y. Conwell

University of Guam, Mangilao, GU, USA;

Chengdu Mental

Health Center, Chengdu, China;

Xinjin Mental Hospital, Chengdu,

China;

University of Hong Kong, Hong Kong, China;

Department

of Psychiatry, University of Rochester, Rochester, NY, USA

The gender characteristics of persons with schizophrenia in rural commu-

nity are still unknown. The goal of this study is to explore the gender differ-

ences of persons with schizophrenia in a longitudinal study in rural China.

Methods: A 10-year follow-up investigation among a 1994 cohort (n = 510)

of patients with schizophrenia was conducted in Xinjin County, Chengdu,

China. All the patients and their informants (100%) were followed up in

2004 using Patients Follow-up Scale (PFS). Results: The results indicated

that male patients had significantly higher level of education and rates of

unmarried and divorced than females. Female patients had significantly

more family members. Male patients had significantly worse social func-

tion. The rate of homelessness was significantly higher in male than female

(Log-rank test, v

= 4.00, P < .05). The mortality rate was significantly

higher in male {2913 per 100 000 person-years (95% CI = 2174–3652)}

than female patients {1661 per 100 000 person-years (95% CI = 1150–

2172)} (Hazard ratio 2.0, 95% CI = 1.3–3.2, P < .005). Although there

were no significant differences of suicidal attempts between male and fe-

male patients, the suicide rate was significantly higher in male {753 per

100 000 person-years (95% CI = 373–1133)} than female patients {249

per 100 000 person-years (95% CI = 50–448)} (Hazard ratio 3.1, 95%

CI = 1.2–8.0, P < .05). Conclusions: The outcome of male patients is worse

than females in rural China. Higher mortality and suicide in male than

female patients may contribute to the higher prevalence of schizophrenia

in women than in men in China. Specific intervention strategies for

improving the outcome of the illness should be developed for male and

female patients with schizophrenia. Acknowledgments: This work was

supported in part by GRIP 1 R01 TW007260–01 (M.S. Ran, PI) from

the Fogarty International Center of NIH.

Reference

1. Ran MS, Chen EYH, Conwell Y, Chan CLW, Yip PSF, Xiang MZ,

Caine ED. Mortality in Persons with Schizophrenia in Rural China:

A 10-Year Cohort Study. British Journal of Psychiatry. 2007;190: 237–242.

ID: 550836

A FURTHER EXPLORATION OF THE GENDER

DIFFERENCES IN THE AGE OF ONSET OF

SCHIZOPHRENIA

Margriet van der Werf

, J. Boydell

, J. van Os

1,2

, R. M. Murray

J. Allardyce

Department of Psychiatry and Neuropsychology, Maastricht

University, Maastricht, Netherlands;

Depatment of Psycological

Medicine, Institute of psychiatry, London, United Kingdom

Background: Whether the well-replicated association between male gender

and early onset of schizophrenia constitutes a biological characteristic of

the disorder (deLisi, 1992) or reflect a differential risk factor profile across

men and women modulating the timing of onset remains unclear. Methods:

All new cases who met RDC diagnostic criteria for schizophrenia, in Cam-

berwell and in Dumfries and Galloway over 20 years (1979–1998; n = 615)

were retrospectively identified. A standardised checklist and the OPCRIT

computer algorithm were used to generate diagnoses and extract premorbid

risk factors. The association of gender with age of onset was examined using

univariate and multivariate statistics. As age of onset was non-normally

distributed Mann Whitney U test was used in univariate and an amixture

analyses was used to generate age of onset groups for ordinal logistic re-

gression. Results: All risk factors were strongly associated with age of onset

in univariate analyses except winter birth and family history of other psychi-

atric disorders. Admixture analyses identified 3 age of onset classes: an early

onset (age at first presentation <42 year), a late onset (age 42–64 years) and

a very late onset schizophrenia group (after age of 64). Ordinal logistic

regressinggenderon age ofonset found a highlysignificant association across

the age groups with the highest probability of effect estimate varying with

gender in the early onset group (0.69 reducing to 17.38 in the LATE and

a probability of only 13.42 in the VERY LATE onset group). The probabil-

ities of variation in effect of gender across the age groups was very much re-

duced after adjusting for being single, unemployed , premorbid social and

occupational adjustment, premorbid cannabis use, urban environment, eth-

nic minority status, family history of schizophrenia, with the probability of

change in effect of just 0.12 in the early onset group, 0.07 in the late onset

group and 0.05 in the very late onset group (the proportional odds assump-

tion was valid in both sets of analyses). Conclusion: This study supports the

assertion that the association between age of onset and gender can be largely

explained by differential exposure of other known risk factors, and is consis-

tent with previous studies which have demonstrated a strong confounding

effect of social demographic variables (Jablensky and Cole, 1997).

ID: 551904

MORE PREMORBID CANNABIS USE WITH THE VAL

ALLELE OF THE CATECHOL-O-METHYLTRANS-

FERASE GENE ASSOCIATED WITH INCREASED

RISK FOR SCHIZOPHRENIA

Marjolaine Masse

1,2

, A. Liu

, R. Joober

1,2

, S. King

1,2

McGill University, Montreal, QC, Canada;

Douglas Mental

Health University Institute, Montreal, QC, Canada

Current thinking is that environmental influences interact with vulnerabil-

ities in some individuals to trigger the onset of certain illnesses, such as

schizophrenia. Literature on the subject supports cannabis use, and to

some extent, the COMT Val allele, as risk factors for schizophrenia.

Recently, adolescent cannabis use was found to interact with COMT geno-

type, increasing the risk for schizophrenia (Caspi et al., 2005). However, few

cases were found, greatly reducing subjects in each cell, and cannabis use was

dichotomized into use and non-use. Because the literature suggests a dose-

response relationship between cannabis use and schizophrenia, and because

onset of use has to precede the prodrome to accommodate a causal model of

cannabis, we believed greater methodological precision could increase inter-

nal validity. Prodrome onset was carefully dated, and total premorbid can-

nabis use was estimated in 72 schizophrenia patients and 63 controls.

Logistic regressions, to explain the risk of being a patient, showed a signif-

icant interaction between the Val allele (Val/Val, Val/Met) and cannabis use:

greater cannabis use increasing significantly the risk of being a patient in Val

allele carriers compared to homozygous Met allele carriers. No effect of can-

nabis use on schizophrenia was found for the Met/Met genotype. These

results support the hypothesis of a genetic vulnerability working in combi-

nation with environmental influences to increase the risk for schizophrenia.

ID: 551843

CANNABIS USE IN YOUNG PEOPLE AT HIGH RISK

OF DEVELOPING PSYCHOSIS

Fern Louise Day

, C. Pugh

, L. R. Valmaggia

, M. M. Byrne

S. Bhattacharyya

1,2

, P. K. McGuire

1,2

International Congress on Schizophrenia Research

78 6. 6. Epidemiology

Outreach and Supportin South London, South Londonand Maudsley

NHS trust, London, United Kingdom;

Psychological Medicine, In-

stitute of Psychiatry, King’s College London, London, United Kingdom

Strong evidence exists to show that cannabis use is associated with psychosis

and, as cannabis is the most widely used of illicit substances, this constitutes

a serious public health concern. The nature of the association between can-

nabis and psychosis has been the topic of much discussion and it has been

tempting to assume a causal relationship. In order to understand better the

role of cannabis in the development of psychotic symptoms and disorder, it

is important to investigate cannabis use in people who are at high risk of

developing psychosis. Information about lifetime cannabis and other sub-

stance use was obtained from 122 participants at ‘ultra high’ risk of devel-

oping psychosis. This information was used to investigate of the

relationship between cannabis use and symptom severity, level of function-

ing, and the subsequent onset of psychosis. The rate of cannabis use was

high: 78.7% of participants reported having tried it at least once. Only

18.5% of the sample was currently using cannabis, but the majority of those

that had tried it reported a past period of regular use. Lifetime or current

cannabis use was not associated with positive symptoms scores. However,

those that had used cannabis heavily, even in the past, presented with higher

scores on the PANSS positive subscale item for suspiciousness and perse-

cution (median score 3 vs 1; P = .026). No significant differences were found

in the rate of transition to psychosis between cannabis users and non-users.

However, early-onset cannabis users who also used drugs such as amphet-

amine and ecstasy tended to be more likely to become psychotic (P = .064).

The results show that heavy cannabis use is associated with greater levels of

persecutory ideation in people at high risk of developing psychosis. How-

ever, cannabis use alone does not appear to increase further the risk psy-

chosis within this population. It might be that cannabis use in the

context of other substance use, particularly stimulants, is more important.

ID: 551830

DO FIRST- AND SECOND-GENERATION

MIGRANTS HAVE A SIMILARLY ELEVATED RISK

OF DEVELOPING PSYCHOTIC DISORDERS? A

SYSTEMATIC REVIEW

Francxois Bourque, A. Malla

Psychiatry, McGill University, Montreal, QC, Canada

Background: It has been replicated that migration is a risk factor for psy-

chotic disorders such as schizophrenia. However, it is not clear yet whether

the risk increase found among first-generation immigrants persists or even

increases among second-generation migrants. This determination is crucial

to provide viable explanations for the correlation between migration and

psychotic disorders, and to provide insights into potential social causation

mechanisms. Objectives: a) to examine the magnitude of risk of developing

psychotic disorders among second-generation migrants in comparison with

first-generation migrants and reference groups; b) to explore potential

mechanisms underlying these findings. Methods: Four databases were

searched for relevant articles between 1977 and 2008. Potential studies

were screened by two independent reviewers in a two-fold process. Studies

were included if they reported numerators and denominators from a defined

area, differentiated first- and second-generation migrants, and provided

age-corrected data. Data extraction and quality analysis was performed

on included studies. Mean weighted relative risk for psychotic disorders

was computed for both first- and second-generation migrants, first using

all available effect sizes, then using only those studies reporting rates for

both migrant generations in order to reduce heterogeneity. Results: The pre-

liminary analysis of the selected studies suggest that the increased risk for

psychotic disorders observed among first-generation migrants is also present

among second-generation migrants, with a trend toward a further risk in-

crease. However, there remains significant heterogeneity across studies.

Conclusion: The persistence of an elevated risk among second-generation

migrants suggests that post-migration factors have likely more impact

than the migratory process per se or pre-migration factors, thus supporting

social causation mechanisms. Urbanicity, childhood adversity, discrimina-

tion and social defeat are among the proposed social factors of relevance to

both first- and second-generation migrants.

ID: 551796

CHANGES IN INCIDENCE OF PSYCHOSIS IN A LO-

CALITY OVER TIME—CLUES TO SOCIOENVIRON-

MENTAL AETIOLOGY

J. Boydell

Health Service and Population Research Department, Institute of

Psychiatry, King’s College, London, United Kingdom

Since the earliest research into psychosis, positive associations have been

repeatedly found between social adversity and incidence rates of schizo-

phrenia. It has been difficult however to completely exclude the possibility

of reverse causality. Natural experiments provide such an opportunity and

have produced some evidence that hostility towards minority groups in

Israel and Canada is linked to increased rates of psychosis. The social con-

text in which people live changes in response to political and economic

events. This can then be investigated as a natural experiment. 9/11 and sub-

sequent events in Europe including Islamophobia were such major events.

If social hostility is a factor in the aetiology of psychosis in the short term,

it would be predicted that rates of psychosis in Europe would drop

amongst people of Caribbean descent (who are overwhelmingly Christian)

and rise amongst people who identify themselves as Muslims. The Cam-

berwell Case Register has collected every incidence case of psychosis from

a geographically defined area of South East London for over four decades

up to 2004. The Black Caribbean and Black African populations who have

very high rates of schizophrenia and other psychoses, are overwhelmingly

Christian. The purpose of this study was to test whether rates fell amongst

Black Caribbean and Black African people after 2001. A binary variable

was created to distinguish cases as presenting from 1993 up to and includ-

ing 2001 or from 2002 to 2004. Poisson regression modelling was carried

out to estimate the effect of this variable on incidence and adjust for age,

gender and ethnicity if these improved the model. First the cases were strat-

ified into the two ethnic group categories and the Poisson regression mod-

elling repeated for each stratum. The analysis then proceeded to test for an

interaction between ethnicity (as the binary variable white and black ethnic

minority) and date of presentation, adjusting for age and gender. The

results for the overwhelmingly Christian Black Minority Ethnic population

show protective effects for onset post-2001 (IRR = 0.54, 95% CI = 0.33–

0.87, P = .012), age, and to a lesser extent, gender. Thus, the incidence of

psychosis in the Black Minority Ethnic population in South-East London

was significantly lower in the years 2002 to 2004 inclusive compared to the

years 1993 to 2001. This lends support to the theory of social hostility al-

though other explanations are possible.

ID: 551793

NEIGHBORHOOD SOCIAL CHARACTERISTICS

AND THE INCIDENCE OF PSYCHOTIC DISORDERS

Wim Veling

, E. Susser

, D. March

, H. W. Hoek

1,2

Center for Early Psychosis, Parnassia Psychiatric Institute, The

Hague, Netherlands;

Department of Epidemiology, Mailman

School of Public Health, Columbia University, New York, NY, USA

One of the most consistent findings in the epidemiology of schizophrenia is

that those who live in cities are at greater risk. It is less clear which risk

factors drive this increased risk, because few attempts have been made

International Congress on Schizophrenia Research

6. 6. Epidemiology 79

to move beyond the crude measure of urbanicity. Social factors across

different levels of organization may be involved, in particular those involv-

ing availability of and access to social capital. We conducted a first contact

incidence study of psychotic disorders between 1997 and 2005 in the city of

The Hague, the Netherlands (n = 618) and investigated associations be-

tween individual and neighborhood characteristics and incidence of psy-

chotic disorders, using multilevel Poisson regression analysis. First, after

adjustment for individual level age, sex, marital status and ethnic minority

status, measures of neighborhood socioeconomic level, residential mobility,

population density, voter turnout at local elections and crime level were all

significantly associated with the incidence of psychotic disorders in a dose-

response fashion. We calculated a cumulative score of neighborhood social

disadvantage, based on these five indicators and divided it into tertiles (low,

medium and high social disadvantage). The individual level adjusted inci-

dence rate ratio (IRR) of psychotic disorders was 1.89 (95% Confidence

Interval, 1.38–2.60) for individuals living in the socially most disadvantaged

neighborhoods and was 1.35 (95% CI = 0.95–1.92) for those living in mod-

erately disadvantaged neighborhoods compared to those living in the most

advantaged areas (Wald x2 [df = 2], 18.19, P < .0005). Second, we inves-

tigated the effect of the ethnic composition of neighborhoods on the inci-

dence of psychotic disorders, not only among ethnic minorities, but also

among native Dutch. In neighborhoods with a very high proportion of

non-Western ethnic minorities, the incidence rate for native Dutch was

35.4 per 100 000 (95% CI = 18.2–62.1), indirectly standardized to age,

sex and neighborhood socioeconomic level. In neighborhoods with a rela-

tively low proportion of ethnic minorities, the incidence rate for native

Dutch was 21.1 (18.4–24.2). Among ethnic minorities, the pattern was op-

posite. These results are suggestive for an ethnic density effect both in ethnic

minorities and in native Dutch. Thus, neighborhood social characteristics

strongly influence the incidence of psychotic disorders within an urban en-

vironment in interplay with individual factors.

ID: 551769

EXPOSURE TO PRENATAL MATERNAL STRESS

PREDICTS DEVELOPMENT RESEMBLING THAT OF

HIGH RISK AND PRE-SCHIZOPHRENIC CHILDREN

Suzanne King

1,2

, D. P. Laplante

, A. Brunet

1,2

, D. J. Walder

A. Ciampi

Research Centre, Douglas Institute, Montreal, Quebec, QC,

Canada;

Psychiatry, McGill University, Montreal, QC, Canada;

Epidemiology and Biostatistics, McGill University, Montreal,

Quebec, QC, Canada;

Psychology, Brooklyn College of the City

University of New York, Brooklyn, NY, USA

Because retrospective studies suggest that stress to the pregnant mother

increases risk for schizophrenia, our goal was to determine whether pre-

natal maternal stress explains variance in childhood characteristics seen in

pre-schizophrenic and high risk populations using a prospective ap-

proach. We have followed more than 150 children whose mothers

were pregnant during the January 1998 Quebec ice storm crisis, contact-

ing families up to 10 times since shortly after the disaster to determine if

effects could be seen on the children’s cognitive, behavioural, or physical

development. Ice storm stress was separated into objective aspects of ex-

posure (eg, days without power, financial loss) and subjective reaction

(PTSD-like symptoms rated from the Impact of Event Scale—Revised).

At the 2005 ICOSR we reported that greater prenatal maternal stress

from the ice storm was associated with poorer IQ and language at 2

years, more behavioural problems at 3 years, and greater dermatoglyphic

asymmetry. Here, we report results through age 5 years of age. Results

indicate significant effects of the severity of maternal objective stress from

the ice storm on cognitive and language development at age 5 years:

WPPSI IQ scores were significantly lower for the high stress group

than for the low stress group, especially in verbal IQ; and the high pre-

natal stress group had significantly lower Peabody Picture Vocabulary

Test scores than the low stress group. Also at 5 years, children from

the high objective stress group exhibited significantly more severe

autistic-like symptoms, whether rated by parents or teachers, than those

in the low stress group. Effects of exposure are also seen on physical

markers. The ratio of the 2nd (index) to 4th (ring) fingers (the 2D:4D

ratio) is a sexually dimorphic trait, with women having lower ratios

than men. Our results show that prenatal stress from the ice storm pre-

dicted this ratio in 5 year olds. Results also show that exposure to the ice

storm was associated with a feminization of the ratio; other studies show

that schizophrenic men and women tend to have more feminine ratios

than controls. In conclusion, we find that prenatal stress from a natural

disaster is associated with many traits seen in pre-schizophrenic or high

risk children. Supported by grants from the Stairs Memorial Fund

(McGill University), Canadian Psychiatric Research Foundation, and

Canadian Institutes of Health Research.

ID: 551765

THE SOCIAL DETERMINANTS OF HIGH RATES OF

PSYCHOSIS IN THE UK BLACK CARIBBEAN AND

BLACK AFRICAN POPULATIONS: A SYNTHESIS OF

FINDINGS FROM THE AESOP STUDY

C. Morgan

, J. B. Kirkbride

, G. Hutchinson

, P. Dazzan

K. Morgan

, G. Doody

, P. Jones

, J. Leff

, R. M. Murray

P. Fearon

Health Service and Population Research Department, Institute of

Psychiatry, King’s College, London, United Kingdom;

University of

Cambridge, Cambridge, United Kingdom;

University of the West

Indies, St. Augustine, Trinidad and Tobago;

Institute of Psychiatry,

King’s College, London, United Kingdom;

University of West-

minster, London, United Kingdom;

University of Nottingham,

Nottingham, United Kingdom

The incidence of schizophrenia and other psychoses is elevated in migrant

and ethnic minority groups. The ÆSOP study was established to test hy-

potheses concerning social and biological factors that might explain the in-

creased incidence of these disorders in the UK Black Caribbean and Black

African populations. ÆSOP is a multi-centre incidence and case-control

study of first-episode psychosis. Cases were all individuals presenting to

specialist mental health services for a first time with a psychotic disorder

over a two-year period. Controls were a group of randomly selected, pop-

ulation-based volunteers recruited over the same time period. We collected

data on clinical presentation, ethnicity and a range of potential social and

biological risk factors. ICD-10 and DSM-IV diagnoses were determined by

consensus, blind to the subject’s ethnicity. Over the study period, 535 in-

cident cases and 391 controls were identified. Compared with the White

British population, the incidence of all psychoses was significantly higher

in both Black Caribbean (Rate Ratio (RR) 6.7) and Black African (RR 4.1)

populations. Broadly, Black Caribbean and Black African cases had more

manic and positive symptoms and fewer negative symptoms. We investi-

gated the potential contribution of social and biological factors by testing

whether the effect, or prevalence, of risk indicators was higher in the Black

populations. For indicators of childhood and adult social adversity, we

found a similarly increased odds of psychosis in all ethnic groups. However,

all indicators of social adversity (eg, separation from a parent before the age

of 16, index of adult social disadvantage) were more common in the Black

Caribbean and, to a lesser degree, Black African populations, compared

with White British. In addition, in area level analyses, we replicated the

finding that rates of schizophrenia in Black groups are highest in areas

of low ethnic density. For biological indicators (eg, minor physical anom-

alies, neurological soft-signs), there was no evidence either that effects were

higher, or the variables more common, in the Black groups. Our findings

provide strong evidence that the high rates of psychosis in the UK Black

International Congress on Schizophrenia Research

80 6. 6. Epidemiology

Caribbean and Black African populations are, in part at least, a conse-

quence of greater exposure to individual and area level social risk factors

over the life course. This further supports a role for social adversity in the

aetiology of psychosis in general.

ID: 551726

EFFECTS OF PRENATAL INFECTION ON

NEUROPSYCHOLOGICAL AND NEUROANA-

TOMIC OUTCOMES IN SCHIZOPHRENIA

Alan Stuart Brown

, L. M. Ellman

, S. Vinogradov

W. S. Kremen

, R. F. Deicken

, J. H. Poole

, W. Y. Tsai

D. M. Kern

, C. A. Schaefer

College of Physicians and Surgeons of Columbia University,

Columbia University/NYSPI, New York, NY, USA;

Department of

Psychiatry, University of California, San Diego, CA, USA;

Division of Research, Kaiser Permanente, Oakland, CA, USA;

Department of Psychiatry, University of California, San Francisco,

CA, USA

In previous work, our research group and other investigators have identified

associations between several prospectively identified in utero infectious and

inflammatory exposures and risk of schizophrenia. In the present study, we

sought to examine whether in utero exposure to herpes simplex viruses [her-

pes simplex virus type 1 (HSV1), herpes simplex virus type 2 (HSV2), and

cytomegalovirus (CMV)], and interleukin-8 (IL-8) are related to neuropsy-

chological and neuroanatomic abnormalities that have been associated with

schizophrenia in previous studies. In the Developmental Insult and Brain

Anomaly in Schizophrenia (DIBS) study, archived maternal serum speci-

mens were assayed for each of these herpesviruses and for IL-8 among cases

of schizophrenia from a large birth cohort. Cases were assessed in adulthood

with comprehensive neuropsychological testing and magnetic resonance im-

aging in order to relate these infections and IL-8 to measures of executive

function/working memory and to neuromorphologic outcomes. In utero ex-

posure to HSV1, HSV2, and CMV were each associated with diminished per-

formance on the auditory N-back (2-back condition) and exposure to HSV1

was related to an increase in errors on the Wisconsin Card Sort Test, suggest-

ing deficits in working memory and executive function in exposed cases. Pre-

natal exposure to HSV-2 and CMV were related to diminished thalamic

volume. Elevated IL-8 in utero was associated with an increase in ventricular

cerebrospinal fluid (CSF) and decreases in volumes of the right caudate, left

entorhinal cortex, and right posterior cingulate. These findings suggest that

prenatal exposure to infection and cytokine activation disrupt the develop-

ment of neural circuitry essential to the performance of executive function

and working memory tasks among patients with schizophrenia, and impair

the genesis of neuroanatomic structures associated with this disorder. This

suggests pathogenic mechanisms by which in utero infection contributes to

risk of schizophrenia, and provides further validation for a role of these expo-

sures in the etiology of this disorder.

ID: 551699

A SYSTEMATIC REVIEW OF GENE-URBANICITY

INTERACTIONS IN PSYCHOSIS

Jim van Os

1,2

Department of Psychiatry and Neuropsychology, School of Mental

Health and Neuroscience, Maastricht University Medical Centre,

EURON, SEARCH, Maastricht, Netherlands;

Division of

Psychological Medicine, Institute of Psychiatry, London,

United Kingdom

The finding that the rate of psychotic disorder is higher in children and

adolescents growing up in an urban environment is well replicated and un-

likely to be confounded entirely by gene-environment correlation due to

selective drift to urban areas in those at genetic risk for psychosis. ‘‘Urban-

icity’’ is a proxy for an as yet unidentified environmental factor (s) prevalent

in urban areas, and, if causal, may contribute to up to 20%–30% of the in-

cidence of psychotic disorder in some countries. For this reason, urbanicity

is an interesting factor to study in the context of GxE. We conducted a sys-

tematic review of the literature and identified four studies in the Nether-

lands, Germany, Israel and Denmark that attempted to examine gene-

urbanicity interactions using epidemiological designs and indirect measures

of genetic risk. All studies found evidence for gene-urbanicity interaction.

The Dutch and Danish studies reported that the increase in incidence as-

sociated with urbanicity in individuals without a family history of psychosis

was much smaller than in those with a family history. The German study

found an effect of urbanicity on psychosis rates only in individuals with

a psychometric psychosis liability. The study from Israel used cognitive im-

pairment as indicator for genetic risk and found a nearly tenfold difference

in increase in incidence associated with urbanicity between the cognitively

non-vulnerable group and the cognitively vulnerable group. Clearly, the

possibility of interaction between an environmental exposure in urban areas

and genetic risk is in need of further study, focusing on (i) the precise nature

of the urban exposure, for example growing up in an area lacking in trust

and cohesion, (ii) the psychological and neurobiological mechanism of the

environmental exposure in order to develop rational hypotheses about

gene-environment interaction, (iii) the nature of the genetic variation

involved and ultimately (iv) the mechanism of the gene-environment

interactions.

ID: 551690

META-ANALYSIS OF PATERNAL AGE AND

SCHIZOPHRENIA RISK IN THE OFFSPRING

Erick Messias

, B. Miller

, J. Miettunen

,J.Lo

nen

A. Alara

isa

nen

,M.Ja

rvelin

3,4

, H. Koponen

,P.Ra

nen

B. Kirkpatrick

, M. Isohanni

Psychiatry and Health Behavior, Medical College of Georgia,

Augusta, GA, USA;

Psychiatry, University of Oulu, Oulu, Finland;

Epidemiology, Public Health, and Primary Care, Imperial College,

London, United Kingdom;

Public Health and General Practice,

University of Oulu, Oulu, Finland;

Psychiatry, University of

Kuopio, Kuopio, Finland

Advancing paternal age (APA) is a well-replicated and relatively robust

risk factor for schizophrenia in the offspring. In regards to autism risk

in the offspring, paternal age has a sexually dimorphic effect, with

a much greater impact on risk of autism in females than in males. A recent

meta-analysis of paternal age and schizophrenia risk in the offspring did

not adjust for gender and study design effects, and did not use the same

age classes across studies. We calculated pooled estimates using meta-anal-

ysis, adjusting for these factors and including data from additional studies.

Articles were identified by PubMed searches and review of references. We

attempted to contact authors of all studies published after 1992 to request

summary data stratified by gender and the same five-year paternal age

groups. Studies where greater than 80% of cases had a diagnosis of

non-affective psychosis, and who provided data in the requested format

were included. Data from the Northern Finland 1966 Birth Cohort, which

have not been previously published, were also included. A total of 8 stud-

ies met our criteria including 5 cohort and 3 case-control studies. Within

the cohort studies, males and females were considered separately. In both

cohort and case-control studies, we found a J-shaped association between

paternal age and schizophrenia risk in the offspring (reference paternal age

of 25–29). Effect sizes were similar between cohort and case-control stud-

ies. There was no evidence of sexual dimorphism for the paternal age ef-

fect. Further studies are needed to investigate mechanisms for this

association, including the increased risk in offspring of fathers under

age 25.

International Congress on Schizophrenia Research

6. 6. Epidemiology 81

Table. Risk of schizophrenia by paternal age for cohort and case-control

studies

Paternal

Cohort Studies

Case-Control

Studies

Age Cases Total Crude 95% CI Cases Controls Crude 95% CI

(years) (n)(n)RR (n)(n)OR

<20 313 49826 1.41 1.26–1.59 179 3398 1.37 1.17–1.60

20–24 2862 561986 1.15 1.09–1.20 1491 38032 0.99 0.92–1.08

25–29 4554 1025448 1.00 Reference 2420 62752 1.00 Reference

30–34 3415 754591 1.02 0.97–1.07 1812 43505 1.20 1.10–1.31

35–39 1902 366779 1.17 1.11–1.23 952 22697 1.11 1.02–1.20

40–44 927 154443 1.35 1.26–1.45 477 10121 1.24 1.12–1.38

45–49 348 55829 1.40 1.26–1.56 182 3806 1.23 1.06–1.44

50–54 136 17260 1.77 1.50–2.10 67 1160 1.48 1.16–1.90

>54 71 6858 2.33 1.85–2.94 36 322 2.87 2.03–4.06

—— ——

Total 14528 2993020 7616 185793

ID: 551689

CHILDREN OF PARENTS WITH AFFECTIVE AND

NON-AFFECTIVE PSYCHOSES: A PROSPECTIVE

STUDY OF CHILDHOOD INTERNALIZING AND

EXTERNALIZING BEHAVIOR

Jo-Ann Donatelli

, L. J. Seidman

2,3

, J. M. Goldstein

2,4

M. T. Tsuang

, S. L. Buka

1,6

Community Health, Brown University, Providence, RI, USA;

Psychiatry, Harvard Medical School, Massachusetts Mental

Health Center, Division of Public Psychiatry, Beth Israel

Deaconness Medical Center, Boston, MA, USA;

Athinoula

Martinos Center for Biomedical Imaging, Massachusetts General

Hospital and Massachusetts Institute of Technology, Charlestown,

MA, USA;

Connors Center for Women’s Health and Gender

Biology, Harvard Medical School at Brigham and Women’s

Hospital, Boston, MA, USA;

Psychiatry, University of California

at San Diego, Psychiatry, Center for Behavioral Genetics, San

Diego, CA, USA;

Society, Human Development, and Health,

Harvard School of Public Health, Boston, MA, USA

It is generally accepted that children of parents with schizophrenia or other

forms of psychosis are at heightened risk for a range of behavioral prob-

lems. However, it remains unclear whether offspring of parents with differ-

ent forms of psychosis (eg, schizophrenia, other non-affective psychoses,

and affective psychoses) have distinct forms of behavioral problems (ie, in-

ternalizing and externalizing). This investigation examined behavioral

observations at ages 4 and 7 years of children of psychotic (n = 281)

and non-psychotic parents (n = 188). There were no significant differences

between groups in behavior observed at age 4. At age 7, compared to chil-

dren of unaffected parents, children of parents with psychosis had a signif-

icantly elevated risk for externalizing problems {adjusted odds ratio (aOR)

2.8 (95% CI = 1.5–5.6)}, in particular for offspring of parents with schizo-

phrenia (aOR = 4.4; 95% CI = 1.7, 12.5). This increase in risk for external-

izing problems was observed for females only (aOR = 8.1; 95% CI =

2.5–26.3). In contrast, male children were at increased risk for internalizing

problems (aOR = 3.6; 95% CI = 1.6–8.3). Additional analyses examined

rates of individual symptoms of internalizing and externalizing behavioral

problems in relation to parental psychosis. Male high-risk offspring had

higher rates of shyness and withdrawal compared to low-risk male off-

spring. Female high-risk offspring exhibited more variable mood or unsta-

ble emotional responses compared to low-risk female offspring.

Implications for treatment of parents with psychotic disorders and high-

risk children are discussed.

ID: 551664

MATERNAL IRON DEFICIENCY AND RISK OF

SCHIZOPHRENIA IN OFFSPRING

Alan Stuart Brown

, B. J. Insel

, C. A. Schaefer

, I. McKeague

E. S. Susser

College of Physicians and Surgeons of Columbia University,

Columbia University/NYSPI, New York, NY, USA;

Division of

Research, Kaiser Permanente, Oakland, CA, USA

Iron is essential for brain development and functioning. Iron deficiency in

early life leads to long-lasting neural and behavioral deficits in infants and

children, and disruptions in myelination and dopamine function. Hence, we

examined the effects of early iron deficiency on the risk of schizophrenia in

adulthood. We hypothesized that maternal iron deficiency, assessed by ma-

ternal hemoglobin concentration ascertained during pregnancy, increases

the susceptibility to schizophrenia in the offspring. The data were drawn

from a large, population-based birth cohort. Hemoglobin was prospec-

tively assayed in virtually all pregnancies. Mean hemoglobin concentration

over pregnancy was the primary exposure. Mean maternal hemoglobin 

10.0g/dl was associated with a nearly fourfold, statistically significant in-

creased rate of schizophrenia (adjusted rate ratio, 3.73; 95% CI = 1.41–

9.81, P = .008), adjusting for maternal education and ethnicity. For every

one g/dl increase in mean maternal hemoglobin, a twenty-seven percent de-

crease in the rate of schizophrenia was observed (95% CI = 0.55–0.96; P =

.024). Further analysis revealed that the effect was specific to the third tri-

mester. These novel findings suggest that maternal iron deficiency may be

a risk factor for schizophrenia in the offspring. Since iron deficiency anemia

occurs commonly in pregnancy, these findings may have important impli-

cations for prevention of schizophrenia in the population.

ID: 551659

CNS INFECTIONS DURING CHILDHOOD AND RISK

FOR SCHIZOPHRENIA: A CASE CONTROL STUDY

Mark Weiser

, N. Werbeloff

, D. Halperin

, A. Levine

, G. Livni

R. Yoffe

, M. Davidson

Psychiatry, Sheba Medical Center, Ramat Gan, Israel;

Pedicat-

rics, Wolfson Medical Center, Houlon, Israel;

Pediatrics A,

Schneider Children’s Medical Center, Petach Tikva, Israel;

Mental

Health, Ministry of Health, Jerusalem, Israel

Some studies found a positive association between viral or bacterial CNS

infections during childhood and risk for later schizophrenia, psychotic dis-

order or mood disorder with psychotic symptoms (1,2). A large, recently

published Swedish study found a slightly increased risk for nonaffective psy-

chotic illness and schizophrenia with previous viral CNS infections, but no

increased risk with bacterial infections (2). The aim of this study was to assess

the risk for schizophrenia in early adulthood after viral or bacterial CNS in-

fection during childhood. The association between CNS infections in

childhood and hospitalization for schizophrenia was examined by identify-

ing 3500 persons who had been hospitalized as children for meningitis. Con-

trols were 6400 persons who as children had been hospitalized for gastro-

enteritis. They were followed for later hospitalization for schizophrenia

using the National Psychiatric Hospitalization Case Registry in Israel.

Cox Regression analysis was used to assess risk for later schizophrenia.

Among the 9970 individuals who were included in the study, 6371

(63.9%) were diagnosed with gastroenteritis and 3599 (36.1%) were diag-

nosed with CNS infections. Overall, no association was found between

International Congress on Schizophrenia Research

82 6. 6. Epidemiology

CNS infections during childhood and risk of later hospitalization for schizo-

phrenia. Post hoc analyses found that children hospitalized for a bacterial

CNS infection between the ages of 1–4 years were at increased risk of later

hospitalization for schizophrenia (adjusted HR = 5.27, 95% CI: 1.4–19.93)

(table). However this finding is based on 3 cases. Bacterial CNS infections at

an early age could be associated with a slightly increased risk for schizophre-

nia later in life.

ID: 551584

THE RELATIONSHIP BETWEEN

NEURODEVELOPMENTAL AND SOCIAL FACTORS

IN FIRST EPISODE PSYCHOSIS

Paul Fearon

, C. Morgan

, P. Dazzan

, T. Lloyd

, J. Leff

P. B. Jones

, R. M. Murray

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom;

Psychiatry, University of Nottingham, Nottingham,

United Kingdom;

Psychiatry, University of Cambridge, Cambridge,

United Kingdom

Introduction: Previous work has demonstrated an association between a)

markers of neurodevelopmental impairment, and b) markers of early and

later social adversity and psychosis. We sought to explore, in an epidemi-

ological sample of first episode psychosis cases, whether there was a rela-

tionship between specific markers of neurodevelopmental impairment and

social adversity, or whether these factors comprised entirely distinct path-

ways to the disorders. Methods: All cases from the UK MRC AESOP study

with complete social and biological data (N = 213: M = 113; F = 100) were

included in this analysis. Two markers of neurodevelopmental impairment

(total Neurological Soft Sign (NSS) score and total Minor Physical Anom-

aly (MPA) score) and two markers of early (prolonged parental separation

in childhood) and later (cumulative index of long term social disadvantage

prior to onset) social disadvantage were chosen. Correlations between each

neurodevelopmental and each social item were calculated, firstly in the

whole sample, and then stratified by gender and by diagnosis. Results:

We found no evidence of any correlation between NSS scores for any social

variable, or for parental separation and any neurodevelopmental variable.

However, there was evidence of a significant and consistent positive corre-

lation between higher MPA scores and greater later social adversity (Spear-

man’s Rho 0.204; P = .003) in the whole sample. This finding held for both

men (0.214; P = .02) and women (0.208; P = .04). In terms of diagnosis, this

held true only for schizophrenia (0.22; P = .01) but not for either manic

psychosis or depressive psychosis. Conclusion: Our findings suggest a spe-

cific relationship between the presence of higher scores of MPAs and higher

levels of long term social disadvantage, in both men and women destined to

develop schizophrenia. The precise nature of this relationship, and its ap-

parent diagnostic specificity, remain to be elucidated.

ID: 551558

THE PREVALENCE OF PSYCHOTIC SYMPTOMS IN

CHILDREN AND ADOLESCENTS: A SYSTEMATIC

REVIEW

Dearbhla Connor, M. Harley, N. Devlin, M. Cannon

Psychiatry, Royal College of Surgeons in Ireland, Dublin, Ireland

There is evidence that psychosis exists on a continuum in the population in

both adult and child and adolescent samples. A large number of studies have

been carried out examining the prevalence rates of psychotic symptoms in

adult populations, however studies in child and adolescent samples are more

scarce. To date, no systematic review of the literature on psychotic symp-

toms specifically in childhood and adolescence has been carried out. This

review aimed to establish prevalence rates of psychotic symptoms in children

and adolescents. A systematic review of all published literature on the prev-

alence of psychotic symptoms in children and adolescents was conducted

using electronic databases Ovid Medline, Pubmed and Psychinfo with an

agreed battery of search terms: young people, adolescents, teenagers, child,

psychotic symptoms, psychosis, paranoia, delusions, grandiosity, unusual

beliefs/ideations, bipolar, positive and negative symptoms, dissociative dis-

orders, prevalence. Reference lists of the relevant papers were examined to

identify further articles. We included both clinical and community samples

of children and adolescents. We excluded studies reporting prevalence rates

of psychotic symptoms in children and adolescents with organic illnesses and

in schizophrenic and selected samples at high risk for psychosis.126 papers

were located. Twenty nine relevant studies were identified: 14 from clinical

and 15 from community settings. Methods used to identify psychotic symp-

toms included self-report questionnaires, clinical interview, standardised in-

terview instruments and review of patient charts or a combination of

methods. Reported prevalence of psychotic symptoms for clinical samples

ranged from 0.4% to 81% across ages 2 to 19 years (outpatient samples 0.4 to

28 %, inpatient samples 1.1 to 81%). Prevalence rates in community samples

varied from 2% to 58.9% spanning ages 7 to 19 years (rates determined

by questionnaire 6% to 58.9% and at interview 9% to 50%). Differing

methodology across studies is reflected in a wide variation of rates of psy-

chotic symptoms both within and across clinical and community samples.

Research in this area would benefit from a standardised developmentally

appropriate screening instrument with high sensitivity for identifying psy-

chotic symptoms in children and adolescents. Future studies could employ

a two stage process to screen large samples of children and adolescents and

determine rates of psychotic symptoms at interview.

ID: 551476

Table. Risk for hospitalization for schizophrenia among individuals hospitalized for CNS infections in childhood (by type) compared to individuals

hospitalized for Gastroenteritis in childhood, by age groups.

Age of

Childhood

Hospitalization

Not later

hospitalized

(N = 9,898)

Later

Hospitalized

(N = 72)

Unadjusted

HR (95% CI)

Adjusted

(95% CI)

<1 Gastroenteritis 3436 28 1 1

CNS infections- bacterial 224 3 1.58 (0.48–5.19) 1.57 (0.48–5.15)

CNS infections- viral 748 3 0.49 (0.15–1.6) 0.46 (0.14–1.51)

1–4 Gastroenteritis 1616 8 1 1

CNS infections- bacterial 103 3 5.26 (1.39–19.88) 5.27 (1.4–19.93)

CNS infections- viral 936 3 0.64 (0.17–2.4) 0.65 (0.17–2.46)

>4 Gastroenteritis 1271 12 1 1

CNS infections- bacterial 80 0 – –

CNS infections- viral 1484 12 0.88 (0.39–1.95) 0.72 (0.32–1.6)

*Adjusted for gender

International Congress on Schizophrenia Research

6. 6. Epidemiology 83

DO NEIGHBOURHOOD CHARACTERISTICS

AFFECT THE DURATION OF UNTREATED

PSYCHOSES?

James B. Kirkbride

, D. Lunn

, C. Morgan

, J. Lappin

R. Abdul-Al

, P. Dazzan

, K. Morgan

, P. Fearon

D. Speigelhalter

, R. M. Murray

, P. B. Jones

Department of Psychiatry, University of Cambridge, Cambridge,

United Kingdom;

MRC Biostatistics Unit, University of

Cambridge, Cambridge, United Kingdom;

Division of Psycholog-

ical Medicine, Institute of Psychiatry, London, United Kingdom

Background: Family involvement in help-seeking has been shown to be as-

sociated with a shorter duration of untreated psychoses, but it is unknown

whether factors at the neighbourhood-level also affect the duration of un-

treated psychoses [DUP]. We sought to test this hypothesis. Methods: All

cases of first-episode psychoses which occurred over a two-year period in 32

neighbourhoods of Southeast London were included in the study. DUP was

estimated for all cases. We fitted a number of multilevel Gaussian (normal)

Bayesian models, which adopted different assumptions with regard to spa-

tial variability of DUP to asses whether there was evidence of heterogeneity

in DUP by neighbourhood, having accounted for a priori individual-level

confounders. Model fit was assessed via the Deviance Information Crite-

rion. Results: 314 subjects with first episode psychoses were included in the

analysis. DUP was heavily skewed with a median of 72 days (inter-quartile

range: 22–323). The median DUP between wards varied from 19 to 631.5

days. We transformed DUP onto the log scale to normalise the outcome

variable in order to fit our Bayesian models. After adjustment for individual

differences in age, sex and ethnicity, there was no evidence of significant

variation in DUP between neighbourhoods. A Bayesian model with indi-

vidual-level predictors, but without any hierarchical effects provided the

best fit to the data. As previously reported, ethnicity minority status

and increased age were significantly associated with longer DUP. Discus-

sion: Neighbourhood factors do not appear to be associated with the du-

ration of untreated psychoses, suggesting that the predictors of DUP are

rooted in the individual and family, not at the societal level.

ID: 551448

NEUROPSYCHOLOGICAL FUNCTION AND PATH-

WAYS TO CARE: FINDINGS FROM THE AESOP

FIRST EPISODE PSYCHOSIS STUDY

Dionne Harleston

, C. Morgan

, P. Fearon

, R. Murray

P. Dazzan

, K. Morgan

, P. Jones

, G. Doody

, J. Zanelli

A. Reichenberg

Division of Psychological Medicine, Institute of Psychiatry, King’s

College London, London, United Kingdom;

Westminster Univer-

sity, London, United Kingdom;

Nottingham University, Notting-

ham, United Kingdom;

Cambridge University, Cambridge, United

Kingdom

Numerous studies have found that patients with psychosis perform poorly

on neuropsychological tests compared with healthy controls. We sought to

investigate whether there was a relationship between cognitive function and

pathways to treatment. We also sought to investigate any differences by

demographic factors. 593 Participants, recruited as part of the three centre

AESOP study (Aetiology and ethnicity of schizophrenia- a longitudinal ep-

idemiological study of FEP) who made contact with services performed

a battery of neuropsychological tests to assess intellectual function (Na-

tional Adult Reading Test-Nart and the Wechsler Adult Intelligence scale,

Revised-WAIS-R). Participants intellectual functioning was compared

with their type of admission. Data on pre-morbid and current intellectual

function was then compared with data also gathered on clinical presenta-

tion. Any correlations in terms of socio-demographic characteristics

(namely ethnicity, age and gender) and socio-economic status were ex-

plored. Analysis was undertaken using t-tests and ANOVAs. No Signifi-

cant differences were found in pre-morbid intellectual function between

patients whose admission was compulsory versus non-compulsory. There

was however a significant relationship between current intellectual func-

tioning and pathways to care (P = .014). Patients who were voluntarily ad-

mitted had a mean full scale IQ of 91.38 compared with patients who were

compulsorily admitted who had a mean full scale IQ of 86.04. Those with

lower cognitive function were more likely to be compulsorily admitted than

being voluntarily admitted. There was additionally a trend for lower IQ and

those who had police involvement (P = .043). There was no significant re-

lationship between Criminal justice or GP referral and GP involvement.

These findings suggest poor performance on neuropsychological testing;

in this case current low intellectual functioning is associated with being

compulsorily admitted and having police involvement. These findings raise

the question of what role cognitive function actually plays in enabling peo-

ple to negotiate a positive pathway to care. There is potentially a need for

increased support and information from agencies for those with lower cog-

nitive function who may be less able to communicate with and navigate the

mental health system. Funded by the UK Medical Research Council.

ID: 551446

SYMPTOM CHARACTERISTICS IN AFRICAN-

AMERICAN AND EUROPEAN-AMERICAN SUB-

JECTS WITH SCHIZOPHRENIA: PRELIMINARY

RESULTS

Sarah Bergen

, A. Fanous

5,7

, R. Amdur

5,7

, S. Aggen

, A. Sanders

D. Levinson

, P. Gejman

, K. Kendler

1,6

Genetics/Virginia Institute for Psychiatric and Behavioral Genetics,

Virginia Commonwealth University, Richmond, VA, USA;

Psy-

chiatry/Center for Psychiatric Genetics, Northwestern University,

Chicago, IL, USA;

Psychiatry and Behavioral Sciences/Center for

Psychiatric Genetics, Northwestern University, Chicago, IL, USA;

Psychiatry and Behavioral Science/Genetics of Brain Function,

Stanford University, Palo Alto, CA, USA;

Psychiatry, Georgetown

University, Washington, DC, USA;

Psychiatry/Virginia Institute

for Psychiatric and Behavioral Genetics, Virginia Commonwealth

University, Richmond, VA, USA;

Washington VA Medical Center,

Washington, DC, USA

The heterogeneous presentation of schizophrenia makes symptom-level

analyses particularly important in designing studies exploring the etiol-

ogy and pathophysiology of this disorder. We used phenotypic data from

1192 African-Americans (AAs) and 1289 European-Americans (EAs)

with schizophrenia drawn from the Genetic Association Information

Network (GAIN) schizophrenia study. Factor analysis in MPlus yielded

five symptom factors in both the AA and EA cohorts: Schneiderian

symptoms, auditory hallucinations, delusions, visual hallucinations,

and negative symptoms. Additional analyses suggested that the lack

of measurement invariance in the five factor structure between the

AA and EA samples is largely due to noninvariance of the negative

factor loadings and thresholds. Demographic variables such as age of on-

set, duration of illness, and sex displayed different patterns of correla-

tions with the factors in the AA and EA samples. Earlier age of onset

predicted higher scores for all factors except the negative symptom factor

in the EA cohort which showed no significant relationship. Duration of

illness positively correlated with auditory hallucinations and negative

symptoms in both samples but visual hallucinations only in the AA sub-

jects. The only significant sex differences were higher levels of auditory

and visual hallucinations in the AA females. The extent to which these

populations differ in symptom expression may be a function of differen-

tial environmental exposures or genetic predispositions between groups.

International Congress on Schizophrenia Research

84 6. 6. Epidemiology

Subsequent analyses incorporating genome wide genotypic data will aid

in distinguishing these possibilities.

ID: 551416

PREVALENCE RATES AND CORRELATES OF

PSYCHOTIC-LIKE SYMPTOMS AMONG ADOLES-

CENTS REFERRED TO A CHILD AND ADOLESCENT

PSYCHIATRIC OUTPATIENT SERVICE

Nina Devlin

1,2

, M. Harley

2,3

, D. Beattie

, M. Iqbal

C. Fitzpatrick

1,3

, M. Cannon

School of Medicine,Catherine McAuley Research Centre,

University College Dublin, Catherine Mc Auley Research Centre,

Dublin 1, Ireland;

Department of Psychiatry, Royal College of

Surgeons in Ireland, Smurfit Building, Beaumont Hospital, Dublin 9,

Ireland;

Child and Adolescent Psychiatry, Mater Hospital, James

Joyce Street, Dublin 1, Ireland

During the past decade, there has been growing interest in psychotic-like

symptoms among young people. For years, such symptoms were not asked

about by child and adolescent mental health professionals as it was consid-

ered meaningless in the absence of a psychotic diagnosis. However, there is

now compelling evidence from population-based cohorts that self reported

psychotic symptoms in early adolescence are associated with a higher risk of

psychotic illness in adulthood. Prevalence studies of psychotic symptoms

among clinical child and adolescent samples are few and to date, have

shown prevalence rates of psychotic symptoms ranging from 4.5% to

33%. We wished to investigate the prevalence and correlates of psy-

chotic-like symptoms in a sample of adolescents referred to an Irish child

and adolescent mental health outpatient service. Over a one year period

consecutive referrals of 12–15 year olds to a child and adolescent mental

health service in North Dublin were asked to participate in the ‘Welcome’

Study. Subjects underwent a comprehensive assessment conducted by a psy-

chiatrist or psychologist. The interview protocol included the K-SADS PL,

Clinical Global Assessment Scale, the Clinical Global Impressions scale,

the Adolescent Wellbeing scale, the Parental Stress Scale and a psychosis

screening questionnaire. The psychosis screening questionnaire is derived

from the DIS-C: and comprises the following questions: Q1/ ‘‘Have other

people ever read your mind?’’,Q2/ ‘‘Have you had messages sent to you via

the TV/radio?’’ Q3/ ‘‘Have you ever thought people were spying on you or

following you?’’ and Q4/ Have you ever heard voices that other people can’t

hear?’’ Forty-four adolescents have participated in the study so far of whom

48.8% have responded positively to one or more of the psychosis screening

questions. Analysis of responses to individual questions was as follows: Q1

‘‘mind-reading’’16.3%; Q2 ‘‘TV/radio 0%; Q3 ‘‘spying’’ 25.6%; Q4 ‘‘voices’’

18.6%. Overall 95.4% of participants had a psychiatric diagnosis of whom

56.8% had co-morbid diagnosis and 4.5% (n = 2) were diagnosed with a psy-

chotic disorder .Of note no-one was originally referred to the service with

suspected psychosis. We conclude that the prevalence of psychotic-like

symptoms are far higher than previously thought among young adolescents

referred to a child and adolescent psychiatry service. These symptoms

should be routinely enquired about as part of the initial assessment.

ID: 551298

PRENATAL EXPOSURE TO RETINOL (VITAMIN A)

AND EXECUTIVE FUNCTIONING IN ADULT

OFFSPRING WITH SCHIZOPHRENIA

Poonam Nina Banerjee

, D. M. Kern

, W. S. Kremen

S. Vinogradov

, R. F. Deicken

, J. H. Poole

, C. A. Schaefer

A. S. Brown

New York Psychiatric Institute, Columbia University, New York,

NY, USA;

Department of Psychiatry, University of California, San

Francisco, CA, USA;

Division of Research, Kaiser Permanente,

Oakland, CA, USA;

Department of Psychiatry, Center for Be-

havioral Genomics, University of California, San Diego, CA, USA

The aim of this study is to evaluate evidence for the theory that elevated

maternal retinol (vitamin A) exposure levels, particularly during the third

trimester, are associated with decreased executive functioning in adult off-

spring with schizophrenia. Prenatal exposure to excess retinol levels and

related compounds (retinoids) has been associated with toxicity to the cen-

tral nervous system as well as congenital fetal anomalies including hydro-

cephaly and spina bifida. Several studies have shown that retinoid

dysregulation may be an important factor in the etiology of schizophrenia,

since congenital anomalies similar to those caused by retinoid dysfunction

are found in patients with schizophrenia and their relatives, and numerous

schizophrenia candidate genes are regulated by retinoic acid (Goodman,

1998). In the present study, we assessed the relationship between serolog-

ically documented prenatal exposure to retinol from archived maternal

sera, and a series of measures of executive functioning, in 20 individuals

with schizophrenia from a large and well-characterized birth cohort.

The cases were derived from the Developmental Insult and Brain Anomaly

in Schizophrenia (DIBS) study. Sera were assayed for retinol by an estab-

lished protocol and cases were administered a standard neuropsychological

test battery including measures of executive function. Elevated retinol levels

among schizophrenia cases were associated more total errors on the Wis-

consin Card Sort Test (P = .01), including both perseverative errors (P = .03)

and non-perseverative errors (P = .03), and lower test scores on the Ruff

Figural Fluency test (P = .09). These findings provide compelling evidence

that retinoids, already implicated in the etiology of schizophrenia and con-

genital malformations, may also lead to executive function deficits in

patients with schizophrenia.

ID: 551282

THE RELATIONSHIP BETWEEN ETHNICITY AND

PSYCHOTIC SYMPTOMS AS MEASURED BY THE

SCAN: PILOT DATA FROM THE GENETICS AND

PSYCHOSIS (GAP) STUDY

Corinne Prescott

Psychiatry, Institute of Psychiatry, King’s College London, London,

United Kingdom

Background: The presence of underlying symptom dimensions in the psy-

chopathology of psychosis is well established, and the relationship between

ethnicity and psychotic symptoms has been demonstrated in previous stud-

ies (Janssen et al. 2003 1, Veling et al. 2007 2). Methods: The Item Group

Checklist (IGCL) of the Schedules for Clinical Assessment in Neuropsychi-

atry (SCAN) Version 2.1. was used to derive scores for 59 patients for the

SCAN. The symptoms investigated comprised the following IGCL Item

Groups: Depressive delusions and hallucinations (Item Group 13) Nonspe-

cific auditory hallucinations (IG 21) Non-affective auditory hallucinations

(IG 24) Miscellaneous delusions (IG 28) Delusions of reference (IG 29)

Delusions of persecution (IG 30) IGCL scores were dichotomised prior

to analysis into present and absent. Subjects: All 59 subjects in the present

sample were recruited to the GAP study from the inpatient services of the

South London and Maudsley NHS Foundation Trust. All met diagnostic

criteria for psychosis (ICD-10). and were recruited during the period May

2005 to June 2006. The mean age of the sample was 27.73 years (sd. 7.82)

and the group comprised 40 males and 19 females. The sample was cate-

gorised by ethnicity into Black 63%) and White/Other (37%). The Black

sample comprised Black African, Black Caribbean, and Black Other.

The White/Other sample comprised White British, Mixed, Indian, Bangla-

deshi and Caucasian. Results: Of the six groups of positive symptoms ana-

lysed, a positive rating of presence for one symptom reached significance:

There was a trend for several symptom groups to be more common in the

International Congress on Schizophrenia Research

6. 6. Epidemiology 85

black patients but these did not reach statistical significance. However, de-

pressive delusions and hallucinations (IG13) were found to be associated

with ethnicity (P = .025; Pearson’s x2(1) = 5.629), being more common

in the white/other group. Conclusion: There was a tendency for several

groups of symptoms to be rated more commonly in the black patients

but these differences were not significant. However, depressive delusions

and hallucinations were more common in the white/other patients.

ID: 551188

TIME COURSE OF ILLICIT DRUG USE AMONG FEP

PATIENTS: DATA FROM THE PICOS-VENETO

STUDY AFTER ONE YEAR OF FOLLOW-UP

Rodolfo Mazzoncini, A. Lasalvia, S. Tosato, C. Bonetto,

D. Cristofalo, K. De Santi, M. Bertani, S. Bissoli, G. Marrella,

L. Lazzarotto, M. Tansella, M. Ruggeri

Section of Psychiatry and Psychological Medicine, Department of

Medicine and Public Health, University of verona, Verona, Italy

Introduction: Substance misuse prevalence rates are higher among psy-

chotic patients than in the general population. Emerging evidences suggest

a casual role of drug exposure in developing psychosis. Of great concern

and clinical relevance is whether or not the pattern of substance use among

these subjects is changing over time and how it affects the course of illness.

This study aims to determine substance misuse prevalence rates variation

over 1 year in first episode psychotic (FEP) patients. Methods: The pre-

sented data are part of the PICOS-Veneto project, a multi-site study aiming

to characterise course and outcome of psychosis from the first contact.

Assessments were performed using standardized measures including: clin-

ical evaluation, socio-demographic characteristics and illicit drugs use.

Results: 143 FEP patients were evaluated at baseline and at 1 year follow

up. At baseline 19.6% was positive for any drug use, 18.4% cannabis, 5.0%

cocaine, 5.7% hallucinogens. At one year follow up, 9.1% was positive for

drug misuse, 9.1% cannabis, 1.4% cocaine, 0.7% hallucinogens. If com-

pared with the general population, prevalence rates, among FEP subjects,

were significantly higher at the baseline: cannabis 18.4% vs 8.0%; cocaine

5.0% vs 1.2% and hallucinogens 5.7% vs 0.3%. After one year drug use

prevalence rates are similar to those observed in the general population.

Conclusions: Substance misuse prevalence rates are significantly higher

among FEP patients than in the general population. A significant reduction

in this behaviour is observed one year after the psychosis onset. This finding

could help in understanding the meaning of the association between

psychosis and drug use.

References

1. Wade D, Harrigan S, McGorry PD, Burgess PM, Whelan G. Impact

of severity of substance use disorder on symptomatic and functional

outcome in young individuals with first-episode psychosis. J Clin

Psychiatry. 2007;68(5):767–74.

2. Verdoux H, Tournier M, Cougnard A. Impact of substance use on the

onset and course of early psychosis. Schizophr Res. 2005 Nov 1;

79(1):69–75.

ID: 551153

MORTALITY AND SUICIDE IN SCHIZOPHRENIA.

FINDINGS FROM THE NORTHERN FINLAND 1966

BIRTH COHORT STUDY

Antti Alara

isa

nen, J. Miettunen, E. Ja

skela

inen, H. Koivumaa-

Honkanen, P. Ra

nen, M. Isohanni

Department of Psychiatry, University of Oulu, Finland, Oulu,

Finland

Our aim was to analyze the mortality and especially suicide rate and pre-

dictors in schizophrenia and other psychotic disorders in the Northern

Finland 1966 Birth Cohort. The Northern Finland 1966 Birth Cohort

Study is based upon 12 068 pregnant women and their 12 058 live-

born children. Those who were alive at the age of 16 (n = 10 934) were

followed up to the age of 39 years. We used validated hospital diagnoses

until 1997 and information on mortality until 2005. Risk ratios (RR) were

calculated having healthy subjects as comparison group. Crude and ad-

justed hazard ratios (HR) and their 95% confidence interval (95% CI) for

suicide by school performance were computed using Cox regression which

takes the time to the event (suicide) account. RR of death for all cohort

members suffering from psychosis (n = 155) was 6.19 (95% CI: 3.96–9.49).

For schizophrenia patients (n = 100) RR was 7.07 (4.24–11.42). Due the

young age of cohort members leading cause of death was suicide. At the

end of year 2005 (age 39) 7 of 100 schizophrenia patients had commit

a suicide. Case fatality (CF, number of suicides divided by number of

subjects) was 7%. CF for females was 2.9% (1/35) and for males 9.2%

(6/65). Suicide risk was high especially for males and in early phase of

psychotic disease; two thirds of suicides in schizophrenia occurred within

three years after onset of illness. Suicide rate of schizophrenia patients

wasnearly2-foldhighersixyearsafter first discharge from psychiatric

care (P = .005) compared to hospital treated patients with non-psychotic

disorder. Among individuals with psychosis predictors of mortality were

violent criminality and male gender, while good school performance was

a predictor of suicide. For psychotic persons having good school perfor-

mance (highest 20%), the adjusted hazard ratio (HR) for suicide was 3.56

(0.97–13.05) compared with the remaining 80%. In the non-psychotic

population (97% without psychiatric hospitalization), accordingly, ad-

justed HR was 0.28 (0.07–1.16). Interaction (school performance x psy-

chiatric diagnosis) was significant (P = .01) even when adjusted with

gender, social class and age of onset of illness. This is the first study

of suicide risk in a prospectively followed population-based cohort of

individuals with schizophrenia. The suicide rate for patients with

new-onset schizophrenia followed until the age of 39 is high. Great ma-

jority of the suicides took place during the first years of the illness.

ID: 551135

A COMPARISON OF SYMPTOM PROFILES BE-

TWEEN DIFFERENT ETHNIC GROUPS IN A FIRST

ONSET PSYCHOSIS STUDY

Arsime Demjaha

, C. Morgan

, K. Morgan

, P. Dazzan

G. Hutchinson

, S. Landau

, P. Jones

, R. Murray

, P. Fearon

Psychological Medicine, Institute of Psychiatry, London, United

Kingdom;

Department of Psychology, Westminster University,

London, United Kingdom;

Psychiatry Unit, University of

the West Indies, Trinidad, Trinidad and Tobago;

Department

of Psychiatry, University of Cambridge, Cambridge, United

Kingdom

The aim of the present study was to investigate whether the symptom

profile in first onset psychosis patients from various ethnic groups

differs from that of their White British counterparts. We recruited

536 patients as part of a multi-centre, population based, incidence

study of psychosis, of which 43% were White British, 29% Black Carib-

bean, and 13% Black African. Psychopathology was assessed using the

Schedules for Clinical Assessment in Neuropsychiatry (SCAN). A Prin-

cipal Axis Factor analysis was performed on symptom scores, using Var-

imax rotation. The symptom dimension scores were then compared

between different ethnic groups employing regression analyses. Factor

analysis gave rise to a five factor solution of manic, reality distortion,

negative, depressive, and disorganisation symptom dimensions account-

ing for 47% of total variance. Regression analyses revealed significant

differences between ethnic groups. Specifically, Black Caribbean

patients experienced more reality distortion and fewer negative symp-

toms than White British (P = .045, and P = .010 respectively).

International Congress on Schizophrenia Research

86 6. 6. Epidemiology

With regards to severity, Black Africans tended to have more severe

manic and reality distortion and less severe disorganisation symptoms

when compared with White British (P = .002, P = .003, and P = .021 re-

spectively). Our results indicate that there are significant variations in

symptom profiles between Black Caribbean, Black African and White

patients with first onset psychosis. These may reflect differences in un-

derlying aetiological factors.

ID: 551122

DO CANNABIS AND URBANICITY INTERACT IN

CAUSING PSYCHOSIS?

Cecile Henquet

, R. Kuepper

, M. Konings

, R. Lieb

2,3

H. U. Wittchen

2,3

, J. van Os

1,4

Department of Psychiatry and Neuropsychology, Maastricht

University Medical Center, Maastricht, Netherlands;

Clinical

Psychology and Epidemiology Unit, Max Planck Institute of Psy-

chiatry, Munich, Germany;

Institute of Clinical Psychology and

Psychotherapy, Technical University, Dresden, Germany;

Division

of Psychological Medicine, Institute of Psychiatry, London, United

Kingdom

Background: Cannabis use and the urban environment are risk factors for

psychotic illness. It is unknown, however, whether these environmental

factors may interact with each other as well on psychosis risk. Method:

Prospective data on 2438 adolescents and young adults (aged 14 to 24

years) from the Early Developmental Stages of Psychopathology

(EDSP) study were analyzed. Urbanicity was defined at baseline as the

German city of Munich (4061 persons per square mile) versus the sur-

roundings of Munich (553 persons per square mile). Cannabis use and

psychotic symptoms were assessed by means of the Composite Interna-

tional Diagnostic Interview at baseline, at T2 (four years later), and again

at T3 (eight years later). Logistic regression analysis was used to investi-

gate the interaction between urbanicity and T2 cannabis use on psychosis

outcome at T3. Results: Cannabis use was associated with an increased

risk of expression of psychosis, even after exclusion of individuals with

previous experiences of psychotic symptoms at T2 and after adjustment

for age, gender, socio-economic status, use of other drugs other, and

childhood trauma (OR = 1.61, 95% confidence interval 1.03 to 2.52,

P = .036). The effect of cannabis on psychosis outcome, however, was

much stronger in those who grew up in an urban area compared to those

who grew up in a rural area (risk difference 9.6%; test for interaction v

5.69, df = 1, P = .02). Neither urbanicity nor pre-existing psychotic expe-

riences predicted later cannabis use. Discussion: These results show that

environmental factors may interact with each other in causing psychosis.

The effects of cannabis on psychosis outcome seem to be particularly

detrimental for those young adults who are growing up in an urban

environment.

ID: 551108

INFECTIONS OF TOXOPLASMA GONDII AND THE

LATER DEVELOPMENT OF SCHIZOPHRENIA IN

MOTHERS

Marianne Pedersen

, H. Stevens

, B. Nørgaard-Pedersen

P. B. Mortensen

National Centre for Register-based Research, University of Aarhus,

Aarhus C, Denmark;

Statens Serum Institut, Copenhagen,

Denmark

Several studies suggest that infectious diseases including Toxoplasma gon-

dii can influence the risk of schizophrenia. Our own previous study (1)

showed that children, who were born with high level of Toxoplasma gon-

dii-specific immunoglobulin (IgG), have a slightly increased risk of schizo-

phrenia. The elevated level of IgG in the neonatal samples in our previous

study indicates that the mother has a history of T. gondii infection. The aim

of this study is to investigate if mothers, to children born with a high level of

IgG, suffer an increased risk of schizophrenia. Children born in five coun-

ties in Denmark from 1992 to 1995 were screened for T. gondii at 5–10 days

post-partum (N = 53,881). Their mothers were followed from the day they

gave birth until first diagnosis of schizophrenia during 1992 to 2007. Pre-

liminary results indicate a general association between Toxoplasma gondii

and later psychiatric illness (RR = 1.21 (95% CI: 1.13; 1.29)). The relative

risk for schizophrenia was of similar magnitude but not statistically signif-

icant. Final analyses will be presented at the conference. Supported by the

Stanley Medical Research Institute.

Reference

1. Mortensen PB, Pedersen BN, Waltoft BL, et al. Toxoplasma gondii as

a risk factor for schizophrenia: analysis of filter paper blood samples

obtained at birth. Biol Psychiatry, 2007;61:688–693.

ID: 551087

PSYCHOPATHOLOGY DURING CHILDHOOD AND

ADOLESCENCE PREDICTS DELUSIONAL-LIKE

EXPERIENCES IN ADULTS: A 21 YEAR BIRTH

COHORT STUDY

James Scott

1,2

, J. Welham

1,3

, G. Martin

1,2

, W. Bor

, J. Najman

M. O’Callaghan

, G. Williams

, R. Aird

, J. McGrath

1,7

Department of Psychiatry, University of Queensland, Brisbane,

QLD, Australia;

Child and Youth Mental Health Service, Royal

Children’s Hospital, Brisbane, QLD, Australia;

Queensland Centre

for Mental Health Research, The Park Centre for Mental Health,

Brisbane, QLD, Australia;

Kids in Mind, Mater Children’s Hos-

pital, Brisbane, QLD, Australia;

School of Population Health,

University of Queensland, Brisbane, QLD, Australia;

School of

Public Health, Queensland University of Technology, Brisbane,

QLD, Australia;

Queensland Brain Institute, University of

Queensland, Brisbane, QLD, Australia

Community surveys have shown that many otherwise well individuals re-

port delusional-like experiences. The aim of this study was to examine the

association between psychopathology during childhood and adolescence as

predictors of subsequent delusional-like experiences in young adults. The

study was based on a birth cohort of 3617 young adults born between 1981

and 1984. Psychopathology was assessed at age 5 and 14 by mothers using

the Child Behaviour Check List (CBCL), and by the cohort members with

the Youth Self Report (YSR) at age 14 delusional-like experiences were

measured using the Peters Delusional Inventory at the 21 year follow up.

The association between childhood symptoms and later delusional-like

experiences was examined using logistic regression adjusted for gender

and age at last assessment, as well as drug use and the presence of nonaf-

fective psychotic disorder at age 21. High CBCL scores at 5 and 14 years

predicted high levels of adult delusional-like experiences; (Highest versus

other quartiles OR and 95% CI: 1.25; 1.00–1.57: 1.85; 1.482.30 respec-

tively). Those in the highest quartile of YSR scores at 14 years were nearly

four times more likely to have high levels of delusional-like experiences at

age 21 (OR and 95% CI: 3.71; 2.92–4.71). Adolescent onset psychopathol-

ogy and continuous psychopathology through both childhood and adoles-

cence strongly predict delusional-like experiences at age 21. Delusional-like

experiences and hallucinations at 14 years were significantly associated with

adult delusional-like experiences. The general pattern of associations per-

sisted when adjusted for previous drug use or the presence of nonaffective

International Congress on Schizophrenia Research

6. 6. Epidemiology 87

psychoses at age 21. Our study found that psychopathology during child-

hood and adolescence predicts adult delusional-like experiences. Under-

standing the biological and psychosocial factors that influence this

developmental trajectory may provide clues to the pathogenesis of

psychotic-like experiences.

ID: 551057

ARE NEURAL TUBE DEFECTS IN SIBLINGS A RISK

FACTOR FOR SCHIZOPHRENIA: A POPULATION-

BASED TEST OF THE ROLE OF LOW FOLATE

DURING PREGNANCY IN RELATION TO

SCHIZOPHRENIA RISK

Michael Skaarup Pedersen

, G. Davey-Smith

A. M. N. Andersen

, P. B. Mortensen

National Centre for Register-based Research, University of Aarhus,

Aarhus C, Denmark;

Department of Social Medicine, University of

Bristol, Bristol, United Kingdom;

Institute of Public Health, Uni-

versity of Southern Denmark, Odense, Denmark

Mothers with low folate intake and TT homozygote mothers for the meth-

yltetrahydrofolate reductase (MTHFR) gene are at increased risk for hav-

ing children with neural tube defects. Neural tube defects as well as

schizophrenia have been found to occur with increased frequency in

cohorts exposed to famine during fetal life, and a meta-analysis examining

the polymorphisms in schizophrenics and controls found the odds ratio

for TT homozygotes to be 1.48 (95% CI = 1.18–1.86). This supports

the hypothesis that folate status may be an important determinant of

schizophrenia risk. In order to test this hypothesis in a population-based

setting, we used the Danish population-based registers to examine schizo-

phrenia risk in the Danish population using the occurrence of neural tube

defects in siblings as a proxy for low folate intake or MTHFR TT homo-

zygosity in the mother. We based our data on the Danish civil registration

system. Through this system it is possible to establish links between indi-

viduals, their parents, and through the parents with their sibs for persons

born in Denmark 1952 or later. Exposure was determined by diagnosed

neural tube defects identified through the National Patient Register and

the Register of Congenital Malformations. Follow-up was made through

the Danish Psychiatric Central Register, covering all psychiatric hospitals

and since 1995 all psychiatric outpatient wards. Outcome was schizophre-

nia, ICD-8 295, ICD-10 F20, 25. We also studied the broader group of

non-affective psychoses as a potential outcome. Results will be available

at the conference.

ID: 551026

EARLY VIRAL INFECTIONS AS A RISK FACTOR

FOR SCHIZOPHRENIA

Preben Bo Mortensen

, C. B. Pedersen

, D. M. Hougaard

B. N. Petersen

, O. Mors

, A. Børglum

, E. F. Torrey

R. H. Yolken

National Centre for Register-based Research, University of Aarhus,

Aarhus C, Denmark;

State Serum Institute, Copenhagen,

Denmark;

Centre for Psychiatric Research, Aarhus University

Hospital, Aarhus, Denmark;

Institute of Human Genetics,

University of Aarhus, Aarhus, Denmark;

Stanley Neurovirology

Lab, Johns Hopkins University, Baltimore, MD, USA

Several studies have suggested that exposure to infections in utero or during

early childhood is a risk factor for the onset of schizophrenia later in life.

Several studies have been based on ecological data or information based

upon maternal recall, and only few cohorts exist where it is possible to fol-

low-up individuals for whom there is serological evidence of early infec-

tions. We studied the role of Herpes simplex virus 1 and 2 and

cytomegalovirus in a nationwide birth cohort in Denmark. We identified

all cases in a Danish psychiatric hospital or outpatient clinic. For each

case, we selected one control from the same cohort born on the same

day as the case and of the same gender, and alive on the data of first ad-

mission with schizophrenia for the case. We initially identified 923 schizo-

phrenia patients, but there was only sufficient biological material available

from the dried blood spot samples in the neonatal biobank for PKU screen-

ing for 602 cases and 602 controls. Those included in the study and those

who had to be excluded due to lack of biological material did not differ

significantly by established risk factors for schizophrenia including family

history of mental illness and urbanicity of place of birth. At the Stanley

Neurovirology Laboratory type specific antibodies to the gG2 glycoprotein

of viral infection were measured by enzyme immuno assay and an assay

value < 0.2 optical density unit was predefined as indicating exposure to

viral infection. There was an increased risk associated with HSV2, RR

1.51 (95% CI = 1.08–2.10). There was no general association between

CMV and schizophrenia (RR 1.10), but a significant interaction by gender

where the risk was increased in males (RR 1.49, 95% CI = 1.05–2.11), con-

trary to what was found in women (RR 0.76, 95% CI = 0.52–1.11) for

HSV1. We only found a weak and non-significant association between

HSV1 and schizophrenia (RR 1.12, 95% CI = 0.89–1.40). We did, however

find an effect in the group of individuals with foreign born mothers, ie, sec-

ond generation immigrants (RR 3.72, 95% CI = 1.30–10.61), both com-

pared to HSV1 seronegative second generation immigrants and to

seropositive and seronegative native Danes. We identified both general

and subgroup specific associations between early viral infections and

schizophrenia risk. This may indicate an interaction between early infec-

tions and other subgroup specific genetic or environmental risk factors

in the etiology of schizophrenia.

ID: 551013

CANNABIS USE AND TRUE INCIDENCE OF

PSYCHOTIC SYMPTOMS IN A POPULATION-BASED

SAMPLE

Rebecca Kuepper

, J. van Os

1,3

, H. U. Wittchen

4,5

, R. Lieb

4,5

C. Henquet

1,2

Departement of Psychiatry and Neuropsychology, South Limburg

Mental Health Research and Teaching Network, EURON,

Maastricht University Medical Center, Maastricht, Netherlands;

Division Addiction Care, Mondriaan Zorggroep, South Limburg,

Netherlands;

Division of Psychological Medicine, Institute of

Psychiatry, London, United Kingdom;

Clinical Psychology and

Epidemiology Unit, Max Planck Institute of Psychiatry, Munich,

Germany;

Institute of Clinical Psychology and Psychotherapy,

Technical University, Dresden, Germany

Research has consistently shown that adolescent cannabis use increases the

risk for later development of psychotic symptoms or schizophrenia with

a factor of two. However, it still remains equivocal whether the relationship

between cannabis use and psychosis is causal or follows the so-called self-

medication hypothesis. In the current study, prospective data from the

Early Developmental Stages of Psychopathology study (N = 2438, age

at baseline 14 to 24 years) were analyzed. Substance use and psychotic

symptoms were assessed at baseline and at follow-up at T2 (four years later)

and again at T3 (eight years later), by means of the Composite International

Diagnostic Interview. Logistic regression analyses were conducted to inves-

tigate the effect of T2 cannabis use on true incidence of psychotic symptoms

at T3, by only including individuals who were cannabis naive at baseline

and had no sign of psychotic symptoms at T2. Analyses revealed that can-

nabis use at T2 significantly increased the risk for psychotic symptoms at

T3, after adjusting for age, gender, socio-economic status, urbanicity, use of

International Congress on Schizophrenia Research

88 6. 6. Epidemiology

other drugs, and childhood trauma (OR = 1.57, 95% confidence interval

1.13 to 2.17, P = .007). The effect of T2 cannabis use on T3 psychosis out-

come remained both strong and significant when only including individuals

who were cannabis naı

ve at baseline and had no previous psychotic expe-

riences at T2 (OR = 1.88, 95% confidence interval 1.11 to 3.16, P = .018).

Testing the reverse association (ie, the effect of psychotic symptoms at T2

on cannabis use at T3) did not reveal significant results (OR = 0.89, 95%

confidence interval 0.67 to 1.19, P = .45), speaking against self- medication

mechanisms. The data presented here confirm that adolescent cannabis use

increases the risk to develop psychotic symptoms later in life. Self-

medication mechanisms are very unlikely to explain this association.

ID: 551007

BIPOLAR DISORDER, SCHIZOAFFECTIVE

DISORDER AND SCHIZOPHRENIA OVERLAP;

A NEW COMORBIDITY INDEX

Thomas Munk Laursen, E. Agerbo, C. B. Pedersen

University of Aarhus, Aarhus, Denmark

Objective: Growing evidence of an etiologic overlap between schizophrenia,

schizoaffective disorder and bipolar disorder has become increasingly dif-

ficult to disregard. We investigated the magnitude of the overlap between

the clinical diagnoses of bipolar affective disorder, schizoaffective disorder,

and schizophrenia over a 30 year period, based on the entire Danish pop-

ulation. We introduced a new comorbidity index, measuring the magnitude

of the overlap between the three disorders, enabling us to make a new con-

tribution to the discussion of whether they are distinct or overlapping dis-

ease entities. Method: We established a register-based cohort study of more

than 2,5 million persons born in Denmark. Risks for the 3 psychiatric dis-

orders were estimated by survival analysis, using the Aalen-Johansen

method. Results: Overall, N = 12 734 were admitted with schizophrenia,

N = 4205 with bipolar disorder and N = 1881 with schizoaffective disor-

der. A female bipolar patient’s risk of also being admitted with a schiz-

oaffective disorder, before the age of 45 was approximately 103 times

higher than that of a woman at the sameageinthegeneralpopulation.

Thus, we defined the comorbidity index between schizoaffective disorder

and bipolar disorder at age 45 to be 103. The index between schizophrenia

and schizoaffective disorder was 80 at age 45, and between schizophrenia

and bipolar disorder 20. Similar results were found for men. A strong

effect modification by age was present resulting in a higher index for

younger patients. Conclusions: The proposed comorbidity index revealed

a large overlap between the diagnostic categories of schizophrenia, schiz-

oaffective disorder, and bipolar disorder. There was a large overlap be-

tween schizophrenia and schizoaffective disorder as well as a large

overlap between bipolar disorder and schizoaffective disorder. But,

more surprisingly it was clear that a substantial overlap between bipolar

disorder and schizophrenia was present. The study supports the existence

of an overlap between bipolar disorder and schizophrenia.

ID: 551000

IS THE INCREASED RISK OF SCHIZOPHRENIA

IN IMMIGRANTS AN ARTIFACT OF SELECTIVE

MIGRATION?

Carsten B. Pedersen

, P. B. Mortensen

, E. Cantor-Graae

National Centre Register-based Research, Aarhus University,

Aarhus, Denmark;

Department of Health Sciences,

Lund University, University Hospital Malmo

, Malmo

, Sweden

Background: Many studies have shown that foreign immigrants have an

increased risk of developing schizophrenia in their host country as com-

pared to their native counterparts. Although, the factor(s) responsible

for the increased risk remain unknown, it has been hypothesized to include

environmental factors associated with the process of migration. However, if

people migrating from one country to another have a higher liability

(genetic and/or environmental) of developing schizophrenia, then an alter-

native explanation for the increased risk associated with migration could be

that immigrants have a higher pre-existing liability of developing schizo-

phrenia, or in other words that migration per se may not increase the

risk of schizophrenia, but the increased risk associated with migration is

due to selective migration. Objective: The aim of this study was to assess

whether risk factors for schizophrenia were also risk factors for emigration.

If risk factors for schizophrenia also increase the risk of emigration, we may

have evidence that immigrants are a pre-selected group of people who have

a higher liability to develop schizophrenia as compared to the native coun-

terparts. Below we focus on the potential association of the degree of ur-

banization at place of birth on the risk of emigration from Denmark; urban

place of birth has consistently been shown to also increase the risk of de-

veloping schizophrenia. At the conference results concerning other risk fac-

tors for schizophrenia will be presented, including a history of mental illness

in a family member. Methods: We followed all people born in Denmark

1971–1991 for the first emigration from Denmark 1971–2006. Results:

Among children (0–15 yrs) the greater the degree of urbanization at place

of birth, the higher the risk of emigration from Denmark. Conclusion: The

higher the degree of urbanization of place of birth, the higher the risk of

schizophrenia and emigration from Denmark, suggesting that people

migrating from Denmark are expected to carry a higher liability to develop

schizophrenia than their native counterparts. Whatever the unknown

factor(s) responsible for the urban-rural differences in the risk of schizo-

phrenia, some of these may also be responsible for the increased risk of

schizophrenia among immigrants. Supported by the Stanley Medical Re-

search Institute.

ID: 550964

EARLY DEVELOPMENTAL RISK FACTORS FOR

VOICE-HEARING AMONG 7 AND 8 YEAR OLD

CHILDREN

Agna A. Bartels

, J. A. Jenner

, G. Van de Willige

, J. Van Os

2,3

D. Wiersma

University Center for Psychiatry, University Medical Center

Groningen, Groningen, Netherlands;

Maastricht University,

Maastricht, Netherlands;

Division of Psychological Medicine,

Institute of Psychiatry, London, United Kingdom

The accurate identification of children and adolescents who experience hal-

lucinations may allow us to target those at higher risk for psychotic illness in

adulthood. In 2002/2003 auditory vocal hallucinations (AVH) were

assessed in a population-based sample of 3870 children in the province

of Groningen, the Netherlands. Voice-hearing was assessed with the Audi-

tory Vocal Hallucination Rating Scale (AVHRS; Jenner and Van de

Willige, 2002). To investigate associations between voice-hearing and

pre- and perinatal influences, early development and current behaviour

problems, for a case-control sample data were gathered from the Infant

Health Service records, and from parental completed Child Behaviour

Checklists (CBCL). A 5-year follow-up study is currently being carried

out to investigate prevalence, persistence and characteristics of AVH in

the case-control sample of now 12 and 13 year old children, and their re-

lationship with social behaviour. At baseline, the one-year prevalence of

AVH was 9%. 15% of AVH children reported severe suffering and AVH

elicited more anxiety in girls. The prevalence for AVH was higher in rural

areas, but AVH were more severe, more often externally attributed and of

greater functional impact in the urban environment. There was some evi-

dence for an association with prenatal maternal infection (OR 2.07, 95% CI

= 1.04–4.05; P = .04) and slower motor development in early life (OR 1.22,

95% CI = 1.02–1.46; P = .03). Only weak associations were apparent be-

tween AVH and current problem behaviour. We conclude that there

may be continuity between AVH in childhood and more severe psychotic

International Congress on Schizophrenia Research

6. 6. Epidemiology 89

outcomes given the severe suffering in a subgroup, associations with early

developmental deviance and evidence for a poorer prognosis in an urban

environment.

References

1. Jenner JA, Bartels-Velthuis AA, Van de Willige G, Van Os J, and

Wiersma D. Hearing voices in childhood: a prevalence and case-control

study of auditory hallucinations in middle childhood. 2008;(submitted).

2. Jenner JA and Van de Willige G. The Auditory Vocal Hallucination

Rating Scale (AVHRS). Groningen: University Medical Center Gro-

ningen, University Center for Psychiatry, University of Groningen.

2002.

ID: 550962

THE ANTECEDENTS OF SCHIZOPHRENIA:

A REVIEW OF BIRTH COHORT STUDIES

Joy Lillian Welham

, M. Isohanni

, P. Jones

, J. McGrath

Queensland Centre for Mental Health Research, University of

Queensland, Brisbane, QLD, Australia;

Department of Psychiatry,

University of Oulu, Oulu, Finland;

Department of Psychiatry,

University of Cambridge, Cambridge, United Kingdom;

Queens-

land Brain Institute, University of Queensland, Brisbane, QLD,

Australia

Background: Birth cohort studies suggest that individuals who develop

schizophrenia differ from the general population on a range of developmen-

tal indices. We summarize key findings from birth cohort studies to identify

areas of convergence and areas requiring further research. Method: We ex-

amined studies based on general population birth cohorts where data are

collected prospectively from birth or childhood, and whose outcome is

schizophrenia or related disorders. We searched various electronic data-

bases to identify such studies using the search parameters (schizo* OR

psych*) AND (birth cohort). We also examined references from relevant

articles and previous reviews. Results: We found 10 birth cohorts from

6 countries with studies adopting adult schizophrenia as an outcome.

They provide relatively consistent evidence that, as a group, children

who develop schizophrenia in adulthood have behavioural disturbances

and psychopathology, intellectual and language deficits and early motor

delays. Some studies provide evidence of changes in educational perfor-

mance and physical growth. Birth cohort studies also find evidence of

a wide range of putative risk factors for schizophrenia. Conclusions: Birth

cohort studies provide important, convergent insights into how the de-

velopmental processes of individuals with schizophrenia differ from their

peers. Some more recent studies examine developmental trajectories;

that is examining individual rather than group measures across develop-

ment—a particular strength of birth cohort designs. Larger cohorts and

new paradigms, together with modern epidemiology and biomedical sci-

ence, are illuminating the developmental pathways to schizophrenia.

ID: 550905

IMPACT OF GENETIC VULNERABILITY FOR

PSYCHOSIS, FETAL INFLAMMATION, AND

GENDER ON AGE 7 COGNITIVE AND BEHAVIORAL

FUNCTIONING

Sara Cherkerzian

, S. L. Buka

2,7

, J. L. Donatelli

, A. Remington

M. T. Tsuang

4,6

, J. M. Goldstein

2,5

Connors Center for Women’s Health and Gender Biology, Division

of Women’s Health, Department of Medicine, Brigham and

Women’s Hospital, Harvard Medical School, Boston, MA, USA;

Department of Community Health, Brown University, Providence,

RI, USA;

Department of Psychiatry, Beth Israel Deaconess

Medical Center, Boston, MA, USA;

Harvard Institute of

Psychiatric Epidemiology and Genetics, Boston, MA, USA;

Athinoula Martinos Center for Biomedical Imaging, Massachusetts

General Hospital and Massachusetts Institute of Technology,

Charlestown, MA, USA;

Department of Psychiatry, Center for

Behavior Genomics, University of California at San Diego, San

Diego, CA, USA;

Department of Society, Human Development,

and Health and of Epidemiology, Harvard School of Public Health,

Boston, MA, USA

Recent work on the etiology of schizophrenia has focused on premorbid

and prodromal endophenotypes to characterize early identifiers of risk.

This high-risk study assessed the impact of genetic vulnerability, fetal in-

flammation, and gender on age 7 offspring cognitive and behavioral out-

comes. Using data from the longitudinal New England Family Study

(NEFS) cohort, we identified 185 parents with psychosis and 126 comparable

control parents. There were 169 (87 males, 82 females) offspring among these

psychotic parents (‘High-Risk’) and 247 (125 males, 122 females) among the

controls. Fetal inflammation, parental psychosis, their interaction and gen-

der were together modeled as predictors of age 7 cognitive and behavioral

functioning in the offspring. Least-squares means, fixed effects and covari-

ance parameters were estimated using mixed linear models accounting for

intrafamilial correlation. Cognitive functioning was significantly lower

among high-risk offspring and those exposed to fetal inflammatory condi-

tions, operationalized as IQ, academic achievement, and perceptual-motor

and verbal-conceptual functions (eg, there was a 5.5 point mean IQ difference

between offspring exposed and unexposed to fetal inflammation (P < .0002);

and there was a 3.5 point mean IQ difference between the high-risk and con-

trol offspring (P < .003). The fixed-effects parameters for inflammation were

significant across all cognitive measures; genetic vulnerability was significant

for IQ (P < .01), and interactions for verbal/conceptual functioning (P < .02)

and inhibition (P < .02). Both of these interactions were found to be signif-

icant for males but not for females. Genetic vulnerability for psychosis and

fetal inflammation are independently associated with lower cognitive func-

tioning even by age 7, and together have an interactive effect on both verbal/

conceptual cognitive functioning and dysregulation of affective expression

primarily in males.

ID: 551933

EVALUATION OF PERINATAL AND PEDIATRIC

RISK FACTORS FOR THE DEVELOPMENT OF

SCHIZOPHRENIA

Raul Escamilla

Servicios Clı

nicos. Clı

nica de Esquizofrenia, Instituto Naciona de

Psiquiatrı

a, Mexico city, Mexico

Schizophrenia is a disorder that affects in a significant degree the psicoso-

cial functioning of the pacient in many areas. One of the major risks to have

this disorder is related with biological proximity; identical twins (sharing

100% of genetics) do not show a 100% concordance for schizophrenia,

which suggests that environmental or epigenetic factors that are not well

enough characterized, could have a determinant rol during neurodevelop-

ment phase on the origin of this disorder. Method. This case-control study

explored the association between perinatal risk, pediatric risk neurological

conditions (seizures on early infancy, cranial traumatism) and the risk

for schizophrenia. Case totaled 100 patients with schizophrenia diagnosis,

every case was individually matched by gender and nearly age with one

sibling without a psychiatric diagnosis in axis 1 (DSM IV criteria). Results:

Adjusted analyses showed that the risk of schizophrenia was associated

with fetal suffering/distress (x

= 34.0, 1 gl, P < .001Fisher P < .001).

The logistic regression model was able to discriminate cases from controls

(general efficacy 86.5%, regarding Hosmer y Lemeshow model), being more

accurate to predict subjects without schizophrenia (93%). Results suggest

International Congress on Schizophrenia Research

90 6. 6. Epidemiology

that this perinatal and early childhood conditions are associated with the

risk of schizophrenia and seem to act not only independently. Conclusions:

The following complications were significantly associated with schizo-

phrenia: Cranial traumatism, seizure on early infancy, complications of

delivery (abnormal labor and delivery, fetal suffering or distress and in-

cubator use); even in a isolated way or integrated in a synergistic model.

A combination of diverse disciplines and approaches will be needed to

elucidate the mechanisms underlying these important associations in or-

der to establish preventive measures. The development and implementa-

tion of optimal strategies for recording important events during

pregnancy, delivery and early childhood is needed to identify subjects

on risk for schizophrenia.

Table. Predictors factors for schizophrenia: logistic regresio

n model

Independent risk b D. E. b Exp (b) IC 95% Exp (b) P

Cranial traumatism 3.7 0.8 39.1 8.3–183.3 <0.001

Seizure on early

infancy

3.3 1.1 27.0 3.0–240.9 0.003

Abnormal labor

and delivery

2.4 1.2 10.7 1.0–111.3 0.05

Fetal suffering

or distress

1.5 0.8 4.6 0.93–22.7 0.06

Incubator use 3.2 1.5 0.04 0.002–0.73 0.03

ID: 555523

ECT USE IN A PUBLIC HOSPITAL IN SYDNEY

Natalie Gabrielle Limet, R. R. Malesu

Psychiatry, Blacktown Hospital, Sydney, NSW, Australia

The aim of the study was to determine the use of ECT in a public hospital in

Sydney. ECT data over a 5 and a half year period at a public hospital ser-

vicing a suburb of 271 710 population (2006 census) in Western Sydney was

analysed. The data collected was: the type of ECT given (eg. bilateral / uni-

lateral etc), the indications (diagnoses), gender, age and legal status of

patients who received ECT. During this period (January 2003 to August

2008) 916 ECTs were administered of which 483 were right unilateral

and, 433 bilateral. None of the patients received bifrontal ECT. 816

ECTs were administered to females whilst only 100 ECTs were given to

males ie, F:M ratio was 8:1. This ratio of females: males was strikingly dif-

ferent from the female: male admission ratio of only 3:2. The majority (504)

of the patients who received ECT suffered from depression whilst 130 suf-

fered from schizophrenia, 274 suffered from schizoaffective disorder and 8

suffered from mania. Only a quarter (248) of ECTs were given voluntarily

while the rest (668) were given to involuntary patients who were under the

NSW Mental Health Act 2007. About 10% (90) of the patients were 65 years

or over of which 50 ECTs were given to patients over 80 years of age. The

use of ECT in this centre is much less when compared to its use in the other

centres in the same Area Health Service. Although current practices suggest

Bifrontal ECT is preferable to the other types, none of the patients were

given bifrontal ECT. In addition only half the patients received Unilateral

ECT despite unilateral ECT usually having less side effects and less memory

impairment. It has become apparent that the use of ECT in this centre needs

further evaluation and implementation of more up-to-date practices.

ID: 554143

International Congress on Schizophrenia Research

6. 6. Epidemiology 91

7. 7. Genetics, Basic

ASSOCIATION OF THE BRAIN-DERIVED NEURO-

TROPHIC FACTOR (BDNF) VAL66MET POLYMOR-

PHISM WITH HIPPOCAMPAL N-ACETYL

ASPARTATE (NAA/CHO) LEVEL AND VERBAL

MEMORY CAPACITY

Oliver Gruber

, H. Scherk

, M. Backens

, T. Schneider-Axmann

T. Wobrock

, A. Hasan

, W. Reith

, J. Meyer

, P. Falkai

Department of Psychiatry, Georg August University, Goettingen,

Germany;

Department of Neuroradiology, Saarland University,

Homburg, Germany;

Department of Neuro-Behavioral Genetics,

University of Trier, Trier, Germany

Background: The brain-derived neurotrophic factor (BDNF) is a key reg-

ulator of neural plasticity and has been suggested to be involved in the path-

ophysiology and pathogenesis of schizophrenia and depression, with

particular emphasis on dysfunctions of the hippocampal formation. In

the present study, we investigated the impact of the val66met polymor-

phism in the BDNF gene on hippocampal functioning, in particular on hip-

pocampal n-acetyl aspartate, a biochemical marker of neuronal integrity, as

well as on verbal memory capacity, which depends on the functional integ-

rity of the hippocampus. Methods: 192 subjects gave written informed con-

sent to participate in the study (67 schizophrenic, 45 bipolar and 33 OCD

patients, and 47 healthy subjects). Proton magnetic resonance spectroscopy

was performed in the left hippocampus according to a standardized algo-

rithm, verbal memory was assessed using the VLMT, and the subjects were

genotyped with regard to the val66met polymorphism (rs6265) of the

BDNF gene. Results: Met allele carriers showed lower relative n-acetyl as-

partate (NAA/Cho) levels (6.2%; F

1,117

= 3.9, P = .05) and worse verbal

memory performance (6.6%; F

1,141

= 3.8, P = .05) as compared to homo-

zygous val/val carriers. The effect on verbal memory capacity was most pro-

nounced in the schizophrenic patients (17.0%). Conclusions: These

findings provide further evidence for a crucial role of BDNF in hippocam-

pal functions, particularly in hippocampus-mediated verbal short-term

memory. Although association studies between the BDNF gene and schizo-

phrenia have been inconclusive, our results are consistent with the assump-

tion that changes in BDNF functioning may contribute to the development

of schizophrenia.

ID: 546112

A CHROMOSOME 15Q13–14 2-BASE PAIR POLY-

MORPHISM IN THE PARTIAL DUPLICATION OF

THE ALPHA 7 NICOTINIC ACETYLCHOLINE GENE

IS ASSOCIATED WITH SCHIZOPHRENIA

Melissa L. Sinkus

, S. Leonard

1,2

, M. J. Lee

, J. Gault

, J. Logel

M. Short

, A. Olincy

1,6

, R. G. Ross

1,6

, L. E. Adler

1,6

R. Freedman

1,6

, S. L. Christian

, J. Lyon

, D. H. Ledbetter

B. Sullivan

Psychiatry, University of Colorado at Denver, Aurora, CO, USA;

Pharmacology, University of Colorado at Denver, Aurora, CO,

USA;

Neurosurgery, University of Colorado at Denver, Aurora,

CO, USA;

Genetics, University of Chicago, Chicago, IL, USA;

Genetics, Emory University, Atlanta, GA, USA;

Psychiatry,

Veteran’s Affairs Medical Center, Denver, CO, USA;

University of

Appalachia College of Pharmacy, Oakwood, VA, USA

Multiple genetic linkage studies support the hypothesis that the 15q13–14

chromosomal region contributes to the etiology of schizophrenia. Among

the putative candidategenes in this area are the alpha 7 nicotinicacetylcholine

gene (CHRNA7) and a partial duplication of alpha 7 (CHRFAM7A). We

have previously reported that a large chromosomal segment including the

CHRFAM7A gene locus, but not the CHRNA7 locus, is deletedin someindi-

viduals. In addition, the CHRFAM7A gene locus also contains a polymor-

phism consistingof a 2 base pair (2-bp) deletion at position497–498of exon 6.

The purpose of this study was to determine whether the 2-bp polymorphism

or the segmental deletion of CHRFAM7A is associated with schizophrenia.

We have employed a PCR-based method to quantify the copy number of

CHRFAM7A and the presence of the 2-bp polymorphism in a large, mul-

ti-ethnic population of Caucasian, African-American and Hispanic descent

forassociation of this locus with schizophrenia.The presence of the 2-bppoly-

morphism was associated with schizophrenia (allele frequency P = .014, ge-

notype P = .055). Association of the 2-bp polymorphism with schizophrenia

was highest in the African-American population (genotype P = .018, allele

frequency P = .028). Association of the 2-bp deletion with schizophrenia

also reached statistical significance in the Caucasian population (genotype

P = .028, allele frequency P = .033). The association of the 2-bp deletion

with schizophrenia did not reach significance in the Hispanic population.

Comparison of P50 auditory gating test-to-conditioning (T/C) ratios in con-

trols revealed no significant difference in the T/C ratio in those individuals

with the 2-bp deletion (18.3

6 2.3 vs 27.4 6 6.7, P = .213). CHRFAM7A

copy number was not associated with schizophrenia or with P50 auditory gat-

ing in any of the ethnicities or in the group as a whole. The 2-bp deletion poly-

morphism is in strong linkage disequilibrium with an inversion of the entire

duplicate geneand surrounding loci. We conclude thatthe presence ofthe 2 bp

deletion at the CHRFAM7A locus may therefore represent larger genomic

rearrangements that may contribute an increased risk for developing schizo-

phrenia by epistatic influences, perhaps involving CHRNA7.

ID: 550808

DIFFERENTIAL REGULATION OF THE CHRNA7

GENE IN SCHIZOPHRENIC SMOKERS

Sherry Leonard

1,2

, S. Mexal

, S. Stephens

, R. Berger

R. Freedman

1,2

, R. G. Ross

Psychiatry, University of Colorado at Denver, Aurora, CO, USA;

Research Division, Veterans Affairs Medical Research Service,

Denver, CO, USA;

Cenomed Biosciences, Irvine, CA, USA;

Institute for Behavioral Genetics, Boulder, CO, USA

The a7 neuronal nicotinic receptor gene (CHRNA7) has been implicated in

the pathophysiology of schizophrenia by genetic and pharmacological stud-

ies. A mutation 1831bp upstream of exon one in the CHRNA7 gene is asso-

ciated with schizophrenia, after correction for multiple testing (Caucasians,

P = .003; African Americans, P = .001). Genotype for this SNP predicts a pos-

itive response to the a7* receptor agonist, DMXB-A. Expression of the a7*

receptor, as measured by [125I]a-bungarotoxin binding is decreased in post-

mortem hippocampus, cortexand reticular nucleus of the thalamusin schizo-

phrenic subjects compared to non-mentally ill controls. Most schizophrenic

patients are heavy smokers. Smoking changes the expression of multiple

genes and differentially regulates gene expression in schizophrenic hippo-

campus.This differentialregulation is in specific patterns; expression is either

lower or higher in schizophrenic non-smokers, compared to control non-

smokers, and is brought to control levels or normalized in schizophrenic

smokers. Several schizophrenia candidate genes are differentially regulated

by smoking in schizophrenia, including CHRNA7. We have examined the

effects of smoking on CHRNA7 in additional subjects by qRT-PCR and

western blot, finding that smoking differentially regulates expression of

both mRNA and protein. CHRNA7 mRNA and protein levels are signifi-

cantly lower in schizophrenic non-smokers, compared to control non-

smokers (P < .05). Core promoter polymorphisms in the CHRNA7 gene,

associated with both the P50 deficit and with schizophrenia, are significantly

more abundant in the schizophrenic non-smokers with low levels of

International Congress on Schizophrenia Research

92 7. 7. Genetics, Basic

CHRNA7mRNA (P < .01). In schizophrenicsmokers, both mRNAand pro-

tein and are brought to control levels. Although a7 subunit levels are ade-

quate in schizophrenic smokers, low surface expression of the a7*

receptor suggests aberrant assembly or trafficking of the receptor. We ana-

lyzed the expression of the nicotinic receptor chaperone gene, RIC3 in hip-

pocampus. RIC3 expression, though not different in schizophrenic

postmortem hippocampus was decreased in smokers.

ID: 550768

MITOCHONDRIAL VARIANTS IN SCHIZOPHRE-

NIA, BIPOLAR DISORDER, AND MAJOR

DEPRESSIVE DISORDER

Marquis P. Vawter

, B. Rollins

, M. Martin

, A. Sequeira

S. Watson

, A. Schatzberg

, H. Akil

, R. Myers

, E. Jones

D. Wallace

, W. Bunney

Psychiatry and Human Behavior, University of California, Irvine,

Irvine, CA, USA;

Molecular and Behavioral Neuroscience Institute,

University of Michigan, Ann Arbor, MI, USA;

Psychiatry, Stan-

ford University, Palo Alto, CA, USA;

Stanford Human Genome

Center, Stanford University, Palo Alto, CA, USA;

Neuroscience

Center, University of California, Davis, Davis, CA, USA;

Molecular and Mitochondrial Medicine and Genetics, University of

California, Irvine, Irvine, CA, USA

An increasing number of brain studies implicate mitochondrial dysfunction

in psychiatric disorders. The present study evaluated mtDNA variants in

brain tissue including substitutions, synonymous and non-synonymous,

and rare variants, which could predispose to bipolar disorder (BD), schizo-

phrenia (SZ) and major depressive disorder (MDD). Genomic DNA from

dorsolateral prefrontal cortex (DLPFC) from a total of 77 subjects (12

with BD, 14 withSZ, 15 with MDD, and36 controls) was studied for mtDNA

sequence variations using Affymetrix GeneChip Mitochondrial Resequenc-

ing Arrays 2.0. The microarray resequencing of mtDNA was 100% concor-

dant with conventional capillary electrophoresis sequencing results for 103

mtDNA variants. There was 99.997% concordance for 3 individuals that

were completely resequenced using both methods. In 45 000 sequences there

was 1 discordant base call. The mitochondrial microarray however showed

a large N callrate, which maybe improved by usingsimilar PCR conditionsas

the National Institute of Standards and Technology. DLPFC from subjects

with schizophrenia had an increased rate (22% higher compared to the con-

trol rate (P = .0017)) of synonymous base substitutions in mtDNA. A novel

risk factor for BD and MDD was found in the ND4L gene at T10652C. In

addition, there was a significant increase of the mtDNA common deletion in

brain DLPFC in the ND4L T10652C carriers. Postmortem brain pH, used as

a quantitative trait, showed significant association (P = .004) with three

mtDNA SNPs (tRNA Leu, ND4, and ND5). These 3 mtDNA SNPs are eth-

nic-specific haplotype-defining polymorphisms for the super haplogroup

cluster (U, K, UK). This super-haplogroup showed a significantly increased

postmortem pH (mean 7.0

6 0.18 SD) compared to the other haplogroups

combined. Focusing on haplogroup susceptibility factors in psychiatric dis-

orders and considering mtDNA variants in brain may lead to innovative

treatments that improve mitochondrial health and brain function.

ID: 550673

A SUSCEPTIBILITY GENE FOR PSYCHIATRIC

DISORDERS INCLUDING SCHIZOPHRENIA,

BIPOLAR DISORDER AND GILLES DE LA

TOURETTE IN 13Q14

Chantal Me

rette

2,1

, C. Emond

, Y. C. Chagnon

2,1

, M. A. Roy

2,1

A. Bureau

3,1

, A. Fournier

, M. Maziade

2,1

Centre de recherche Universite

Laval Robert-Giffard, Quebec, QC,

Canada;

Department of Psychiatry, Laval University, Quebec, QC,

Canada;

Department of Social and Preventive Medicine, Laval

University, Quebec, QC, Canada

The idea that schizophrenia (SZ) and bipolar disorder (BP) share a com-

mon genetic ethiology has emerged from the fact that several vulnerability

loci or genes may be common to the two disorders. Moreover, the nosol-

ogy of major psychoses is challenged by the findings that SZ and BP share

several neurobiological, neuropsychological and clinical phenotypic char-

acteristics. A genome-scan in a sample of 48 SZ and BP families from

Eastern Quebec was carried out with one of the main objective to test

this hypothesis. Model-based and model-free linkage analysis were per-

formed in this sample and provided a nonparametric linkage (NPL) value

of 5.21 at marker D13S1247, a score which clearly suggests that 13q14,

where lies D13S1247 locus, contains one or several genes underlying

both mood disorders and schizophrenia. Our data thus suggest a suscep-

tibility locus in 13q13-q14 that may be shared by schizophrenia and mood

disorder. That locus would be additional to another well documented and

more distal 13q locus where the G72/G30 gene is mapped. In a parallel

project on the genetic study of Gilles de la Tourette (TS), we recently com-

pleted a full genome scan in a large Eastern Quebec family (127 members)

in which 20 family members were definitely affected by Gilles de la Tour-

ette (TS) and 20 others showed tic disorders. One of the most striking fea-

tures of the results from this linkage study is that the best finding was also

obtained in 13q14 with the exact same marker (D13S1247) that yielded the

best significant evidence for linkage with SZ and BP. Indeed, in 13q14, we

observed 3 markers linked to TS with a maximum parametric lod score

max

) above 1, including a Z

max

of 3.01 at marker D13S1247. Moreover,

the nonparametric linkage analysis (NPL) was consistent with the para-

metric one and yielded a maximum NPL entropy score of 3.58 again

at D13S1247. Given our actual results obtained with TS, we can now ex-

pand on our previous conclusion and stipulate that 13q14 may harbor

a more general susceptibility gene for psychiatric disorders including

SZ, BP and now TS.

ID: 550663

ASSOCIATION WITH SCHIZOPHRENIA OF THE

BIPOLAR DISORDER GENE PURINERGIC

RECEPTOR P2X, LIGAND-GATED ION CHANNEL,

7 LOCATED AT 12Q24

Yvon C. Chagnon, A. Bureau, M. A. Roy, C. Merette,

M. Maziade

Laval University Robert-Giffard Research Center, Quebec, QC,

Canada

Purpose of the study: The purinergic receptor P2X, ligand-gated ion chan-

nel, 7 (P2RX7) gene is a positional candidate for bipolar disorder (BP) since

its location at 12q24 where strong linkages with BP were reported

(Morissette et al., 1999; Shink et al., 2005a,b). In fact, P2RX7

showed a strong association in French-Canadian bipolar families at the

P2RX7 non-synonymous single nucleotide polymorphism (SNP)

rs2230912 (P-value .0007), which results from an over-transmission of

the mutant G-allele to affected offspring (Barden et al. 2006). Numerous

convergences are observed between BP and schizophrenia (SZ) for

phenotypic and genetic characteristics. For instance, we observed a linkage

at 12q23.1 with a mixed phenotype (CL) including both BP and SZ

(Maziade et al. 2005). We aimed to evaluate if P2RX7 could also be

associated to SZ. Methods: We analyzed 23 SNPs located within

P2RX7, including rs2230912, in 247 unrelated SZ and 150 unrelated con-

trols. A Chi2 analysis was used to evaluate the Hardy-Weinberg equilib-

rium within SZ and controls, and to compare genotypic and allelic

International Congress on Schizophrenia Research

7. 7. Genetics, Basic 93

frequencies between them. We aimed to confirm some of these results in our

multigenerational families densely affected by SZ and BP using Family-

based association test (FBAT). Results: We observed differences between

genotypic frequencies at four SNPs (.0017  P  .035), and differences be-

tween allelic frequencies at 8 SNPs (.0003  P  .04), including rs2230912

previously associated to BP. Using FBAT in families, we observed no as-

sociation with BP, while a weak association between rs2230912 and our

composite phenotype CL was detected, and between a second SNP and

SZ (P = .03 for both). However, the power of our multigenerational families

to test for association is relatively low, more particularly for BP, and for

so we cannot confirm or infirm previous association reported with BP.

Conclusion: We concluded that P2RX7 is also associated to SZ in

French-Canadians.

ID: 550584

A TRANSLATIONAL APPROACH TO PRIORITIZE

CANDIDATE GENES FOR SCHIZOPHRENIA:

CONVERGENCE OF ASSOCIATION AND GENE

EXPRESSION ANALYSES

Paul D. Shilling, T. A. Greenwood, D. L. Braff, G. A. Light,

K. S. Cadenhead, M. A. Geyer, J. Sprock, N. R. Swerdlow

Department of Psychiatry, Univ of California San Diego, La Jolla,

CA, USA

Background: We have reported gene expression values in the nucleus accum-

bens (NAC) of two rat strains that exhibit differential sensitivity to the pre-

pulse inhibition-disruptive effects of the dopamine (DA) agonist,

apomorphine (APO) (Shilling et al., Biol Psych 2008). In the present study,

we used this database to prioritize a list of 29 genes that had been identified

using the custom COGS/UCSD SNP chip, based on their associations

with schizophrenia and/or at least one of the following potential endophe-

notypes for this disorder: prepulse inhibition (PPI), startle magnitude,

startle habituation, P50 S1 amplitude, or P50 suppression (Greenwood

et al. 2008). Because schizophrenia patients exhibit PPI deficits compared

to normal comparison subjects, genes associated with strain differences

in sensitivity to APO-induced PPI deficits in this animal model provide

a rationale to prioritize specific schizophrenia-associated genes for further

investigation. Methods: Affymetrix 230 2.0 chips were used to measure

gene expression in the NAC. Twenty-seven of the 29 genes identified by

association analysis in human subjects were represented on this gene

chip. Gene expression of these 27 genes in the NAC was analyzed based

on data collected from this animal model of differential PPI sensitivity

to apomorphine in Long Evans (LE) vs. Sprague Dawley (SD) rats. Sta-

tistical analysis of microarrays (SAM) was employed to identify signif-

icant between strain differences in expression using a false discovery rate

(FDR) < 5 %. Results: SAM analyses revealed that 10 of the genes as-

sociated with at least one of the potential schizophrenia endophenotypes

also exhibited significant gene expression differences in the NAC at

a FDR <5%. Interestingly, all 10 genes were upregulated in the LE

compared to SD rats. These genes include COMT, NRG1, GRIN2B

and GRID2. Gene expression/behavioral correlations will also be pre-

sented. Discussion: Specific genes that were associated with endopheno-

types for schizophrenia were also differentially expressed across strains

exhibiting differences in PPI sensitivity to a DA agonist. This conver-

gent translational approach provides a potentially powerful means to

prioritize candidate genes for investigation in future studies, and ulti-

mately might facilitate the identification of genes that contribute to

the development of schizophrenia. Supported by MH42228 and

MH68366.

ID: 550558

UNDERSTANDING NEUROCHEMISTRY OF

SCHIZOPHRENIA ENDOPHENOTYPES: GENETIC

ANIMAL MODELS OF HYPERDOPAMINERGIA

AND HYPOGLUTAMATERGIA

Raul R. Gainetdinov

1,2

Neuroscience, Italian Institute of Technology, Genoa, Italy;

Cell

Biology, Duke University, Durham, NC, USA

There is growing understanding that abnormalities in multiple neurotrans-

mitter systems can be involved in the pathophysiology of schizophrenia.

Based on pharmacological observations, two major neurochemical hypoth-

eses have been proposed to explain schizophrenia. The dopamine hypoth-

esis suggests that either an excess of dopamine or an increased sensitivity to

the neurotransmitter is the underlying pathological mechanism. A compet-

ing notion, the hypoglutamatergic hypothesis of schizophrenia, suggests

that decreased glutamatergic transmission underlies schizophrenia, based

on the ability of NMDA receptor antagonists to recapitulate certain endo-

phenotypes of schizophrenia. In an attempt to understand the potential

contribution of various neurotransmitter systems to particular endopheno-

types of schizophrenia several genetically modified animal models have

been investigated. A mouse in which the dopamine transporter gene

(DAT-KO) has been inactivated provides a model of hyperdopaminergia.

This mouse displays hyperactivity, perseverations in cognitive tasks and de-

ficient sensorimotor gating. These behavioral deficits can be corrected by

antipsychotic drugs and as such recapitulate particular endophenotypes of

schizophrenia related to positive symptoms. A mouse, which carries a hy-

pomorphic allele of the NR1 subunit of the NMDA receptor provides

a model for a hypofunctioning glutamate system. NR1 mutant mice display

more complex set of behavioral abnormalities that include mild hyperac-

tivity, social dysfunctions, deficient sensorimotor gating and cognitive im-

pairment. These aberrant behaviors can be ameliorated more effectively by

atypical rather than typical antipsychotics and thus NR1 deficient mice

may have translational value to understand endophenotypes of schizophre-

nia related to negative symptomatology. Here I will discuss how these and

other recent genetic animal models of aberrant neurotransmission may be

instrumental to decipher the contribution of specific neurochemical abnor-

malities to certain endophenotypes of schizophrenia.

ID: 550503

ANALYSIS OF GENE EXPRESSION IN TWO LARGE

SCHIZOPHRENIA COHORTS IDENTIFIES

MULTIPLE EXPRESSION CHANGES ASSOCIATED

WITH NERVE TERMINAL FUNCTION

Jackie de Belleroche

, P. Maycox

, F. Kelly

, A. Taylor

, J. Reid

R. Viknaraja

, N. Jones

, M. Lennon

, C. Davies

, J. Hagan

C. Scorer

, C. Angelinetta

, T. Akbar

, A. Mortimer

, T. Barnes

S. Hirsch

Division of Neuroscience and Mental Health, Imperial College

London, London, United Kingdom;

Psychiatry CEDD,

GlaxoSmithKline, Harlow, United Kingdom

Schizophrenia is a severe psychiatric disorder with a world wide prevalence

of 1%. The pathophysiology of the illness is not understood, but is thought

to have a strong genetic component with some environmental influences on

aetiology. To gain further insight into disease mechanism, we used micro-

array technology to determine the expression of over 30 000 mRNA tran-

scripts in post-mortem tissue from a brain region associated with the

pathophysiology of the disease- Brodmann Area 10 (anterior prefrontal

cortex) in 28 schizophrenic and 25 control patients. We then compared

our study (Charing Cross Hospital Prospective Collection: CCHPC)

with that of an independent prefrontal cortex dataset from the Harvard

Brain Bank (HBB). We report the first direct comparison between two

International Congress on Schizophrenia Research

94 7. 7. Genetics, Basic

independent studies in which 51 genes have been identified that are com-

mon between the schizophrenia cohorts and of these, 49 show the same di-

rection of disease-associated regulation. In particular, changes were

observed in gene sets associated with synaptic vesicle recycling/ transmitter

release and cytoskeletal dynamics. This strongly suggests multiple, small

but synergistic changes in gene expression that affect nerve terminal

function.

ID: 550256

ALTERATION OF MICRORNA BIOGENESIS IN

SCHIZOPHRENIA AND ITS IMPLICATIONS FOR

SYNAPTIC STRUCTURE AND FUNCTION

Murray Cairns

1,2

, N. J. Beveridge

1,2

, A. Carroll

1,2

, P. Tooney

2,1

E. Gardiner

1,2

, D. Santarelli

2,3

Schizophrenia Research Institute, Sydney, NSW, Australia;

School of Biomedical Sciences, The University of Newcastle,

Callaghan, NSW, Australia;

Hunter Medical Research Institute,

Newcastle, NSW, Australia

Our analysis of cortical microRNA (miRNA) expression in postmortem

tissue from the superior temporal gyrus (STG) and the dorsolateral pre-

frontal cortex (DLPFC) has shown that there is a broadly based schizo-

phrenia-associated increase. The scope of this anomaly was suggestive of

an alteration to miRNA processing and we can confirm that we have iden-

tified a corresponding increase in the expression of mRNA for the

DiGeorge Critical Region 8 (DGCR8) gene in both regions of the brain.

This gene encodes a major component of the microprocessor complex,

which is thought to be a rate-limiting step in the miRNA biogenesis path-

way. The link between altered miRNA expression and biogenesis in schizo-

phrenia was even stronger in the DLPFC, with a statistically significant,

increase in mRNA for Drosha, also part of the microprocessor, and Dicer

another RNase III synonymous with post-transcriptional gene silencing.

Quantitative real-time RT-PCR analysis supported the view that there

was a schizophrenia-associated increase in mature miRNA expression in

both cortical regions and that many of the altered miRNAs were also com-

mon to both regions. The biological significance of these findings are pro-

found as each miRNA has the capacity to regulate the expression hundreds

of target genes. This was exemplified by upregulation of the entire miR-15

family including miR-15a, miR-15b, miR16 and miR-195. This functionally

convergent group is predicted to have many target genes involved in syn-

aptic structure and function and many are also associated with schizophre-

nia. We have substantiated some of these target gene relationships through

luciferase reporter gene assays and gene expression profiling in cells trans-

fected with synthetic miRNA and miRNA inhibitors.

ID: 550221

DIFFERENTIAL NEUROBEHAVIOURAL EFFECTS

OF ACUTE AND CHRONIC CANNABINOID

TREATMENT ON HETEROZYGOUS NEUREGULIN

1 MUTANT MICE

Tim Karl

1,2

, B. Aurelie

1,3

, G. Hunt

, J. Micheau

, J. Arnold

1,3

Neuroscience, Schizophrenia Research Institute, Darlinghurst,

Sydney, NSW, Australia;

Neuroscience Research Program, Garvan

Institute of Medical Research, Darlinghurst, Sydney, NSW,

Australia;

Department of Pharmacology, University of Sydney,

Newtown, Sydney, NSW, Australia;

Department of Psychological

Medicine, University of Sydney, Newtown, Sydney, NSW,

Australia;

Centre de Neurosciences Inte

gratives et Cognitives,

Universite

Bordeaux, Talence, Bordeaux, France

The human neuregulin 1 (NRG1) gene is a schizophrenia candidate gene

and mice mutant for transmembrane domain (TM) Nrg1 exhibit a marked

schizophrenia-related behavioural phenotype. Schizophrenia with its multi-

factorial aetiology has a concordance rate of 30–50% for monozygotic twins

highlighting the fact that neither environment nor genetics alone are suf-

ficient to cause schizophrenia but a combined action is likely. Cannabis

use may increase the risk of developing schizophrenia by unmasking the

disorder in genetically vulnerable individuals. Using a multi-factorial ani-

mal model strategy, we investigated the neurobehavioural phenotype of

male heterozygous TM Nrg1 mutant mice (Nrg1 HET) and their wild

type-like (WT) littermates, which were treated acutely and chronically

with either vehicle or cannabinoids [acute: 5 and 10 mg/kg delta-9-tetrahy-

drocannabinol (THC) / chronic: 0.4 mg/kg CP 55 490 (CP)]. All mice were

tested in a variety of tasks for locomotion, exploration, anxiety and senso-

rimotor gating [ie, prepulse inhibition (PPI)]. Fos B/deltaFos B as well as c-

Fos expression analyses determined neuronal correlates for the behavioural

effects of acute and chronic cannabinoid treatment. Acutely, Nrg1 HETs

were more sensitive to the locomotor suppressant actions of THC and

expressed a greater THC-induced enhancement in %PPI compared to

WT mice. Mutants were also more susceptible to the anxiogenic effects

of THC. THC selectively increased c-Fos expression in the ventrolateral

septum (LSV), the central nucleus of the amygdala and the paraventricular

nucleus of Nrg1 HETs. Chronically, Nrg1 HET mice developed more rap-

idly tolerance to CP-induced hypothermia and locomotor suppression com-

pared to WTs. Conversely, tolerance to the anxiogenic effect of CP was only

observed in WT mice. In addition, a selective increase in Fos B/deltaFos B

expression was detected in the LSV of Nrg1 HET mice following chronic

CP exposure, with no corresponding effect seen in WT mice. Acute but not

chronic CP treatment facilitated PPI in Nrg1 HET mice and decreased it in

WT mice. These data suggest that a variation in the Nrg1 gene alters the

sensitivity to the neurobehavioural effects of cannabinoids and results in

differential adaptive processes to repeated cannabinoid exposure. Overall,

our findings support the idea of an interactive relationship between neure-

gulin and the cannabinoid system.

ID: 548541

PILOT STUDY OF THE PERCEPTIONS OF GENETIC

RISK ESTIMATION AND ASSOCIATED REPRO-

DUCTIVE DECISIONS IN CHINESE

Huijie Li

1,3

, R. C. Chan

1,2

, W. G. Honer

, J. C. Austin

4,5

Neuropsychology and Applied Cognitive Neuroscience Laboratory,

Institute of Psychology, Chinese Academy of Sciences, Beijing,

China;

Key Laboratory of Mental Health, Institute of Psychology,

Chinese Academy of Sciences, Beijing, China;

Graduate School,

Chinese Academy of Sciences, Beijing, China;

Department of

Psychiatry, Centre for Complex Disorders, University of British,

Vancouver, BC, Canada;

Early Psychosis Intervention Programme,

Peace Arch Hospital, White Rock, BC, Canada

Background: Schizophrenia is a polygenic illness affecting about 1% of the

population across different countries. First-degree relatives of schizophre-

nia are more prone to be affected. However, relatively few studies have been

launched to examine the relationships between risk perception, reproduc-

tive decisions and attitudes towards predictive testing in a Chinese sample.

Aim: to investigate relationships between risk perception, reproductive

decisions and attitudes towards predictive testing, for patients with schizo-

phrenia and their first-degree nonpsychotic relatives, respectively Method:

A Chinese version of a tailor-made questionnaire (Austin et al., 2006) cap-

turing specific questions on perceptions of psychotic disorders was admin-

istered to 57 patients with schizophrenia and 25 corresponding first-degree

nonpsychotic relatives in an outpatient clinic in Shantou. Results: Only

a quarter of nonpsychotic relatives of patients with schizophrenia perceived

risk accurately and more than half of the relatives overestimated/underes-

timated the genetic risk. For the patients, only 12.3% perceived genetic risk

International Congress on Schizophrenia Research

7. 7. Genetics, Basic 95

accurately. 32% of the relatives and 28.1% of the patients reported that fam-

ily history had affected reproductive decisions. Majority of the relative sup-

ported prenatal genetic testing for psychosis as well as genetic testing in

general, while no more than half patients agreed with the prenatal and gen-

eral genetic testing. Conclusions: Most of the first-degree nonpsychotic rel-

atives estimated the risk inaccurately among the current Chinese sample.

Facilitating accurate risk perception through genetic counseling could sig-

nificantly impact reproductive decisions, and the appropriate impact repro-

ductive decisions, and the appropriate use of genetic tests in the future in

China.

ID: 546390

MICRORNA CHANGES IN THE PREFRONTAL AND

TEMPORAL CORTICES IN SCHIZOPHRENIA

Natalie Jane Beveridge

1,2

, A. P. Carroll

1,2

, P. A. Tooney

1,2

M. J. Cairns

1,2

Schizophrenia Research Institute, Sydney, NSW, Australia;

School of Biomedical Sciences, University of Newcastle, Callaghan,

NSW, Australia

microRNA (miRNA) are small RNA molecules that are expressed in a de-

velopmental and tissue-specific manner, and are thought to regulate at least

a third of all human genes through the sequence-specific base pairing with

the 3#-UTR of specific target mRNAs. Expression profiling experiments

have shown that many mammalian miRNAs appear to be crucial to the

development of the brain and the nervous system. It has been suggested

that abnormalities in the miRNA system could potentially alter brain struc-

ture and function, and may contribute to the development of disorders such

as schizophrenia. Gene expression studies in our laboratory have identified

a general trend toward gene down-regulation in schizophrenia, and it is pos-

sible that this could in part be a consequence of post-transcriptional gene

silencing. To investigate this hypothesis, miRNA expression was analysed

in postmortem cortical grey matter of the superior temporal gyrus (STG)

and dorsolateral prefrontal cortex (BA9; DLPFC); from both schizophre-

nia and control subjects using a high throughput microarray platform. This

study identified a global up-regulation of miRNA expression in the STG

and DLPFC and many expression changes were validated using a custom-

designed real-time RT-PCR method. Investigation of primary and precur-

sor miRNA transcripts as well as genes involved in the miRNA biogenesis

pathway suggest that altered mature miRNA expression may be a conse-

quence of aberrant miRNA processing. Using in silico target gene analysis,

numerous schizophrenia candidate genes were found to have putative miR-

NA binding sites for these altered miRNA within their 3#-UTR (eg, BDNF,

RELN, HTR2A, DRD1, DLG4, VSNL1, GRIN3A, GRM7, CHRM1 and

PLEXNA2). A luciferase reporter gene assay was established and many

miRNA recognition elements were functionally validated. This data sug-

gests that alterations in post-transcriptional gene silencing may play a sig-

nificant role in schizophrenia-associated changes in gene expression.

ID: 550907

TIMING OF PRENATAL DEVELOPMENTAL

PERTURBATIONS INDUCED BY MUTANT DISC1

RESULTING IN SCHIZOPHRENIA-LIKE BEHAV-

IORAL AND PATHOLOGICAL ALTERATIONS

Yavuz Ayhan

, M. W. Vogel

, C. A. Ross

, M. V. Pletnikov

Psychiatry and Behavioral Sciences, Johns Hopkins University,

Baltimore, MD, USA;

MPRC, University of Maryland,

Catonsville, MD, USA

Data from neuropathologic, neuroimaging, neuropsychological and epide-

miologic studies suggest that early neurodevelopmental alterations may

contribute to the dysfunction in schizophrenia and related disorders. We

used a genetic mouse model as an experimental system to demonstrate

the contribution of neurodevelopmental insult on adult pathology. Previ-

ously, we have generated an inducible mouse model of the human mutation

of Disrupted-In Schizophrenia 1 (mhDISC1), using a Tet-off system. The

expression of mhDISC1 is restricted to forebrain neurons. Although the

existence of mhDISC1 results in schizophrenia-like alterations, such as en-

larged ventricles, the timing for the generation of these alterations remains

incompletely understood. To evaluate the critical time points for adult

pathological alterations, we generated three groups of mice with expression

of mhDISC1 prenatally and postnatally; prenatally only; or without expres-

sion. The expression of the mhDISC1 was regulated by feeding mice with

doxycycline-containing food. Prenatal but not postnatal expression of

mhDISC1 was associated with decreased social interaction and increased

aggressiveness. Mutant mice with prenatal expression displayed enhanced

sensitivity to a NMDA antagonist, MK-801, or amphetamine. Prenatal ex-

pression was also associated with depressive features as indicated by in-

creased immobilization time and decreased latency to immobility in

Porsolt Forced Swim Test. In addition, prenatal but not postnatal expres-

sion of the mutant protein produced decreased total cortical volume, re-

duced parvalbumin immunoreactivity, and increased dendritic spine

density in granule cells of the dentate gyrus. We believe these results provide

experimental support for neurodevelopmental hypothesis of schizophrenia

and related disorders.

ID: 551899

DISC-1 PROTEINS ENCODED BY LEU607PHE

ALLELES SHOW A DIFFERENTIAL SUB-CELLULAR

LOCALISATION AND DIFFERENTIALLY

MODULATE NEUROTRANSMITTER RELEASE

Sharon L. Eastwood

, C. A. Hodgkinson

, P. J. Harrison

Department of Psychiatry, University of Oxford, Oxford, United

Kingdom;

Laboratory of Neurogenetics, NIAAA, Rockville, MD,

USA

Like other putative schizophrenia susceptibility genes, disrupted-in-

schizophrenia 1 (DISC-1) may impact on synaptic functioning, in part

through effects on microtubules. The Leu607Phe substitution is one of

the few polymorphisms in DISC-1 which is coding, and which has been

associated to a schizophrenia spectrum phenotype. The latter property

increases the prior probability that is a functional variant, but direct evi-

dence for this is lacking. Given the links between DISC-1, microtubules,

and synapses, we have used an in vitro model system to examine whether

the polymorphism affects neurotransmitter release, microtubule dynamics,

or the sub-cellular localisation of the encoded DISC-1 protein. SH-SY5Y

cells were chosen for this study as they are human and have a neuronal

phenotype. Undifferentiated SH-SY5Y cells were transfected with con-

structs containing either V5-tagged Leu607 or Phe607 DISC-1. Twenty

four after transfection, [3H] noradrenaline neurotransmitter release from

the cells was examined using an established assay. Perfusate was collected

after basal (5mM Kþ) and stimulated (100 mM Kþ) release, the adherent

cells solubilized, and the quantities of released and non-released [3H] nor-

adrenaline determined by liquid scintillation counting. Western blots were

used to quantify tyrosinated and detyrosinated alpha-tubulin, markers of

dynamic and stable microtubules respectively. SH-SY5Y cells were also

grown on slides, transfected with V5-tagged Leu607 or Phe607 DISC-1

and anti-V5 immunofluorescence performed to characterise their sub-cel-

lular distribution. Consistent with prior observations, Leu607- and Phe607-

DISC-1 were differentially distributed, with only Leu607 DISC-1 showing

localisation to the centrosome, an organelle important in microtubule or-

ganization. Transfection of Phe607 DISC-1 decreased basal noradrenaline

release, whilst Leu607 DISC-1 had no effect. The results reveal that the

Leu607Phe polymorphism affects the subcellular localisation of DISC1

and basal neurotransmitter release by SH-SY5Y cells. The latter finding

suggests that modulation of neurotransmitter release may be added to

International Congress on Schizophrenia Research

96 7. 7. Genetics, Basic

the many roles that DISC-1 plays in the brain, and that may contribute to

its role in genetic pathophysiology. Ongoing studies are being conducted to

determine if the variants are associated with differences in the proportion of

stable to dynamic microtubules and with other evidence for altered micro-

tubular and synaptic functioning.

ID: 551792

HOW DISC1 MOUSE MODEL CAN HELP ADDRESS

THE COMPLEX PATHOGENESIS OF

SCHIZOPHRENIA

Mikhail Pletnikov

Psychiatry and Behavioral Sciences, Johns Hopkins University

School of Medicine, Baltimore, MD, USA

Animal models are instrumental in elucidating the neurobiology of schizo-

phrenia. Despite of significant advances, heterogeneity of etiological fac-

tors and the complex pathogenic mechanisms of this psychiatric disease

call for further development of animal models. Recent discoveries in human

genetics have provides a new framework for developing models with strong

etiological and neurobiological validity. In addition, a better elucidation of

environmental factors that contribute to the disease allows for modeling

gene-environment interactions using relevant genetic risk factors and iden-

tified environmental challenges. I will describe new findings with our mouse

models based on inducible expression of mutant human Disrupted-In-

Schizophrenia (DISC1). Two main directions in using this models will

be overviewed, ie, the timing of the effects of mutant DISC1 on neurode-

velopment and interactions of mutant DISC1 with an environmental factor

that induces strong immune response during pregnancy. For the first

project, I will evaluate the new data to indicate that majority of subtle

schizophrenia-like neurobehavioral alterations in mutant mice, such as en-

hanced sensitivity to a NMDA antagonist, MK-801, or amphetamine as

well as the decreased total cortical volume and reduced numbers of parval-

bumin-positive cells, are due to prenatal expression of mutant DISC1. For

the second project, I will discuss the new findings about possible synergistic

interactions between mutant DISC1 and prenatal poly IC treatment to

activate immune response in pregnant mouse dams. Compared to saline-

treated DISC1 mice, poly IC-challenged DISC1 mice exhibited increased

anxiety-like and depression-like responses associated with the decreased

density of dendritic spines in the dentate gyrus of the hippocampus. I

will argue that DISC1 mouse model provides new avenues for advancing

mechanistic studies of the complex pathogenesis of schizophrenia and re-

lated neurodevelopmental disorders.

ID: 551678

THE ROLE FOR THE CIRCADIAN TIMING SYSTEM

AND THE GENE CLOCK IN INFORMATION

PROCESSING

Robert Gonzalez

Psychiatry, UT Southwestern, Dallas, TX, USA

Background: The precision of the circadian timing system is in large part

dictated by endogenous molecular clocks and the expression of circadian

genes. There is recent scientific evidence that suggests an association be-

tween variations in circadian genes, bipolar disorder, and schizophrenia.

The circadian genes have been shown to be important in the processing

of somatosensory signals and in systems that affect information processing,

such as the dopaminergic system. Both neuranatomical and behavioral sim-

ilarities have been reported between circadian gene knockout mice and

patients who are diagnosed with schizophrenia. Similarities include reduced

hippocampal volume, enlarged ventricles as well as impairments in PPI and

memory. The Clock 3111 T/C SNP has been associated with variations of

activation patterns in the go no go task in bipolar depressed patients. An

increased frequency of the C allele has been reported in patients with schizo-

phrenia, particularly in disorganized and residual sub-groups. These find-

ings suggest that endogenous clocks can process sensory stimuli and play

a key role on sensory driven adaptive behavior as well as information pro-

cessing. Aim: To test the associations between Clock 3111 T/C genotype

and both clinical and neuropsychiatric measures of information processing

in subjects with psychotic illnesses. Methods: Subjects: Subjects with

a SCID-IV confirmed diagnosis of Bipolar type I with psychosis, Schizo-

phrenia, and schizoaffective disorder enrolled in a phenotyping study at the

University of Texas Southwestern Medical Center at Dallas Department of

Psychiatry’s Division of Translational Research. Genotyping: All Subjects

will be genotyped for the Clock 3111 T/C SNP (rs1801260). Measures of

Working Memory: (1) WAIS-III Digital Symbol Coding, (2) WAIS-III

Letter-Number-Sequencing, (3) WMS-III Spatial Span. Measures of Exec-

utive Functioning: (1) Wisconsin Card Sorting Test, (2) Trails B Trail Mak-

ing Test. Measures of Neuropsychiatric Functioning: (1) PPI, (2) Eye

Tracking Statistical Analyses: We will test for associations between Clock

3111 T/C genotype (T/T, T/C, C/C) and both clinical and biologic measures

of cognitive functioning. Significance: Circadian genes may serve a more

varied purpose as a component of a central integrator system and therefore

be important in information processing in psychotic illnesses.

ID: 551598

ASSOCIATIONS AND INTERACTIONS BETWEEN

A NETWORK OF DOPAMINERGIC GENE

POLYMORPHISMS IN SCHIZOPHRENIA: A -

FOLLOW-UP IN AN INDIAN SAMPLE

Chowdari Kodavali

, P. Semwal

, M. E. Talkowski

, J. Wood

S. Deshpande

, B. K. Thelma

, V. L. Nimgaonkar

1,2

Department of Psychiatry, Univ of Pittsburgh, Pittsburgh, PA,

USA;

Department of Human Genetics, Univ of Pittsburgh, Pitts-

burgh, PA, USA;

Department of Psychiatry, Dr. RML Hospital,

New Delhi, India;

Department of Genetics, University of Delhi,

New Delhi, India;

Center for Human Genetics Research, Harvard

University, Boston, MA, USA

We have earlier evaluated the hypothesis that dopaminergic polymor-

phisms are risk factors for schizophrenia (SZ) in Caucasian samples

from the USA and Bulgaria (Talkowski, et al. Hum. Mol. Genet. 2008).

The most promising associations are detected with SLC6A3 (alias

DAT), DRD3, COMT and SLC18A2 (alias VMAT2). Here we present

the follow-up analyses in the collaborative Indian trio sample (n = 601 fam-

ilies). Unscreened cord blood samples (n = 520) were collected from live

birth at Lok Nayak Hospital, New Delhi. The SZ diagnoses were based

on DSM IV criteria, similar to the US sample. Informed consent was

obtained from the probands and family members at Ram Manohar Lohia

Hospital, New Delhi. Snplex and SnapSHOT assay methods were em-

ployed for SNP genotyping. Our goals were, a) to seek replication at

SNP level based on our earlier work in the US and Bulgarian sample,

b) to understand the linkage disequilibrium (LD) patterns in the Indian

sub-population and c) to test the epistatic interactions reported. We

have evaluated a total of 41 SNPs. Suggestive associations were detected

in the same direction reported earlier at SLC18A2 (rs363338, trends test

P = .075). At DRD3 and SLC6A3, trends were noted with the opposite

alleles (respectively, rs324030, P = .084; rs403636, P = .058). We also ob-

served additional SNPs with suggestive associations at SLC18A2

(rs363399, rs10082463 and rs363285, P < .1). The LD patterns were similar

to the US and Bulgarian samples. Earlier reported epistatic SNP interac-

tions at all four genes between seven locus pairs (p  0.05), were not noted.

However, we observed other epistatic interactions, the majority being be-

tween SNPs at SLC6A3 and COMT. They need to be explored further at

a functional level. Overall, we suggest the importance of DA genes and their

interaction, as risk factors for SZ in the Indian sample.

ID: 551488

International Congress on Schizophrenia Research

7. 7. Genetics, Basic 97

DISC1 EXPRESSION AND FUNCTION IN THE

HIPPOCAMPUS

Kate D. Meyer

1,2

, J. A. Morris

1,2

Human Molecular Genetics, Department of Pediatrics, Children’s

Memorial Research Center, Chicago, IL, USA;

Integrated

Neuroscience Program, Northwestern University, Chicago, IL, USA

Disrupted In Schizophrenia-1 (DISC1) is a highly promising schizophrenia

susceptibility gene which is involved in several processes important to neu-

rodevelopment, such as neurite outgrowth and neuronal migration. DISC1

is expressed in several brain areas implicated in schizophrenia, with partic-

ularly strong expression in the hippocampus. Immunohistochemical anal-

ysis of Disc1 expression in the mouse hippocampus reveals strong

expression in areas CA1-CA3 and the dentate gyrus throughout develop-

ment, with expression becoming more well-defined as these structures ma-

ture. We have also discovered that Disc1 is expressed by precursors

migrating toward the dentate gyrus in early development. Given that func-

tional and anatomical abnormalities in the hippocampus are consistently

observed in the schizophrenic brain, there is an intriguing possibility

that DISC1 abnormalities confer an increased risk for schizophrenia by dis-

rupting the development of the hippocampus. To investigate this possibil-

ity, we have performed functional studies in the mouse hippocampus that

indicate a role for Disc1 in its development.

ID: 551367

ALTERNATIVE SPLICING OF A NOVEL CASSETTE

EXON IN THE DOPAMINE TRANSPORTER IS

ASSOCIATED WITH SCHIZOPHRENIA

Vishwajit L. Nimgaonkar

, M. E. Talkowski

, M. Chen

L. McClain

, M. Bamne

, J. Wood

, D. Lewis

, M. Owen

M. O’Donovan

, G. Kirov

, L. Georgieva

, D. Toncheva

P. Papasaikas

, J. Lopez

Department of Psychiatry, University of Pittsburgh School of

Medicine, Pittsburgh, PA, USA;

Center for Human Genetics Re-

search, Harvard University, Pittsburgh, PA, USA;

Department of

Biological Sciences, Carnegie Mellon University, Pittsburgh, PA,

USA;

Department of Psychological Medicine, Cardiff University,

Pittsburgh, PA, USA;

Medical University of Sofia, Sofia, Bulgaria

We aimed to refine previously detected, replicable associations and epistatic

interactions between intronic variations within the dopamine transporter

(SLC6A3, ‘‘DAT’’) and schizophrenia. In fine-mapping analyses, we

obtained near complete common variation information within these

regions. To date, 375 variations have been catalogued, including 164 com-

mon variants which we captured at r2 < 0.95 using 88 ‘tag’ SNPs. Signif-

icant associations with schizophrenia were replicated with several linkage

disequilibrium (LD) clusters spanning introns 3 and 4 of the gene between

two Caucasian cohorts (494 cases / 540 controls from the U.S., 659 trios

from Bulgaria). A primate-specific computational model predicted a novel

108 base pair cassette exon defined by a potential recursive splice site and

a standard 5’ splice site motif within intron 3. The predicted exon was

flanked within 600 nucleotides by schizophrenia-associated SNPs which

fell within predicted splicing regulatory signals. Alternative transcripts

have not been previously identified at DAT. We detected alternative splic-

ing of the cassette exon (E3b) in cell transfection experiments and verified

the inclusion of E3b in endogenous DAT transcripts in human substantia

nigra tissue using RT-PCR assays. Differential inclusion of E3b was ob-

served between constructs bearing schizophrenia risk alleles in this region

and a construct with non-risk alleles. Using tag SNPs to represent all plau-

sible risk alleles spanning the cassette exon and predicted regulatory

sequences, we detected a significant association between schizophrenia

and a common haplotype in both cohorts. E3b introduces multiple in-frame

stop codons into the mRNA which truncates the DAT open reading frame

and may serve as a mechanism to negatively regulate DAT production.

Exon E3b is conserved among primates and many other placental mammals

but appears to have been lost in the Glires clade. We conclude that alter-

native splicing of a novel cassette exon could alter DAT function in the

human brain and confer risk for schizophrenia.

ID: 551264

SCHIZOPHRENIA GENETICS: NUMBER OF GENES,

ENVIRONMENTAL INFLUENCES AND DISEASE

PENETRANCE UNDER EPISTATIC MODELS

Esben Agerbo

, B. E. Madsen

, C. Wiuf

University of Aarhus, National Centre for Register-based Research,

Aarhus, Denmark;

Bioinformatics Research Center, University of

Aarhus, Aarhus, Denmark

Schizophrenia is a complex disease with a prevalence of about 1% in the

general population. Little is known about the genetics underlying schizo-

phrenia, the number of genes and the interplay between genes. We aim to

get insight into the genetics of schizophrenia by investigating a range of

genetic models and their fit to empirical data. In particular, we consider

additive, multiplicative and network models as well as combinations of

additive and network models. These models are evaluated by

calculating a likelihood-based score of how well they fit observed prev-

alences from pooled epidemiological studies of families where at least one

member suffers from schizophrenia. (1) Our models are functions of

number of genes, allele frequencies and environmental influence where

the disease penetrance is determined as a function of the particular ge-

netic model and environmental influence. The data are best explained by

models containing few genes in epistasis, and furthermore, models with

recessive genes are generally superior. Particularly, best fits are obtained

under a one-gene additive model combined with network models with

recessive genes, which is in keeping with recent findings on a de novo

copy number variant expressing a rare chromosomal deletion. (2) An-

other striking result is that the percentage of individuals in the general

population predisposed to schizophrenia is between 5% and 7% in the

best fit model. Compared to a prevalence of 1%, this indicates that

the environment has an important influence on the development of

schizophrenia: only one in six (five to seven) predisposed individuals

develops schizophrenia.

References

1. Gottesman II. Schizophrenia genesis: The origins of madness. New

York: Freeman WN;1991.

2. Stefansson H, Rujescu D, Cichon S, Pietilainen OP, Ingason A, Stein-

berg S, et al. Large recurrent microdeletions associated with schizo-

phrenia. Nature 2008:11;455(7210):232–236.

ID: 551137

CONVERGING EVIDENCE FOR DPYSL2 AS A RISK

FACTOR FOR SCHIZOPHRENIA

Catherine Louise Winchester

1,2

, L. H. O’Donovan

1,2

M. E. Bailey

, R. Hunter

1,3

, J. A. Pratt

1,2

, B. J. Morris

1,2

PysRING, Universities of Glasgow and Strathclyde, Glasgow,

United Kingdom;

Faculty of Biomedical and Life Sciences,

University of Glasgow, Glasgow, United Kingdom;

Psychiatry,

NHS Greater Glasgow and Clyde, Glasgow, United Kingdom

The genetic basis of schizophrenia is not clearly understood and no caus-

ative mutations have been definitively identified. It is highly heritable and

involves the interaction of multiple genes of small effect and their interplay

with the environment. Understanding the number of genes and their role

and interaction in the pathophysiology, susceptibility to the disease or par-

ticular clinical symptoms and antipsychotic metabolism remains a crucial

International Congress on Schizophrenia Research

98 7. 7. Genetics, Basic

issue in understanding the disease and developing novel drug strategies. We

identified DPYSL2 as a candidate gene for schizophrenia from a transcrip-

tome screen of the prefrontal cortex of a rodent phencyclidine (PCP) model

of schizophrenia developed in our laboratory 1,2. This PCP model produces

a pattern of metabolic hypofunction, neurochemical changes and behaviou-

ral deficits in the prefrontal cortex that closely mirror the cognitive deficits

of schizophrenia 1,2. DPYSL2 is an intriguing candidate gene as it maps to

8p21, a region of the human genome, that has been repeatedly implicated by

linkage analyses and association studies. In addtion its role in axonal for-

mation and guidance in the developing nervous system is highly consistent

with the neurodevelopmental theory of schizophrenia. To further charac-

terise the function of DPYSL2 we have validated specific siRNAs to knock-

down the expression of DPYSL2 in vitro and in vivo. In addition, we have

performed a case-control association study of DPYSL2 in DNA samples

from 500 UK patients with DSM4 schizophrenia and 500 controls using tag

SNPs selected from HapMap3 and the ABI SNPBrowser4. Genotyping of

SNPs in several haplotype blocks has been completed and association

analysis at both single SNP and haplotype levels is underway. Our initial

findings include the observation of a modest association with a 3’-UTR

SNP in DPYSL2. In conclusion, combining neurobiological and genetic

association studies provides a powerful means to understanding the con-

tribution of individual candidate genes to the pathophysiology of

schizophrenia. This work was funded by formerly Mitsubishi Pharma

Co. (currently Mitsubishi Tanabe Pharma Co.) and NHS Scotland RandD

for Mental Health Research.

References

1. Cochran SM, et al. Neuropsychopharmacology 2003;28(2):265–275.

2. Pratt JA, et al. British Journal of Pharmacology 2008;153:S465–470.

3. Altshuler D, et al. Nature 2005;437(7063):1299–320.

4. http://www.allsnps.com/snpbrowser.

ID: 551097

DISCOVERY AND INITIAL CHARACTERIZATION

OF A NOVEL D-AMINO ACID OXIDASE MRNA

VARIANT

Louise Verrall

1,2

, K. Vernon

, M. Takeda

R. Hashimoto

, P. J. Harrison

Psychiatry, Oxford University, Oxford, United Kingdom;

Physiology, Anatomy and Genetics, Oxford University, Oxford,

United Kingdom;

Psychiatry, Osaka University, Osaka, Japan

D-amino acid oxidase (DAO) degrades D-serine, the endogenous NMDA

receptor(NMDAR) co-agonist. SinceD-serine levels andNMDAR function

may be reduced in schizophrenia DAO is a pathophysiological candidate.

This is supported by its genetic associations with and gene expression alter-

ations

1,2

in schizophrenia and its potential benefit as a therapeutic target. We

previously showed regional differences in DAO cellular localization and an-

tigenicity between human cerebellum and cortex

. One potential explanation

may be the regional expression of alternative splice variants. We have de-

scribed splice variants for other schizophrenia genes

3,4

, of interest since

they may represent novel therapeutic targets for the disorder. This study

therefore investigated novel DAO mRNA variants in human brain. We

used exon-exon PCR and primers designed to novel human DAO sequences

deposited on the Aceview database. PCR products were purified, sequenced

and any novel variants analysed in silico and expressed in mammalian vectors

in vitro. A novel variant, denoted DAO_vG, was detected in human cerebel-

lum and lymphoblasts but could not be detected in cortex. Sequencing

revealed DAO_vG comprises a portion of Intron8–9 of DAO at its 5# end

continuouswithexons9, 10, 11andthe 3#UTR of theDAO gene.PCRexperi-

mentswith primers flanking Intron8–9 verified that DAO_vGis notthe result

of retention of the entire Intron8–9 but is likely transcribed from a novel start

site. In silico analysis revealed the presence of two single nucleotide polymor-

phisms (SNPs) in the 5’ sequence of DAO_vG and showed that one of them

alters binding of 3 transcription factors. Current studies are investigating

whether DAO_vG mRNA levels are regulated by these SNPs or are altered

in schizophrenia using lymphoblasts from Japanese schizophrenia (n = 45)

and control (n = 45) subjects. A mammalian expression vector containing

DAO_vG has been generated and ongoing studies are characterizing the

function of DAO_vG transfected into HEK293 cells. In conclusion, we re-

port a novel mRNA variant of DAO, detected in human cerebellum but not

cortex. Ongoing studies are investigating its transcriptional regulation, ex-

pression in schizophrenia and functional characterization. Supportedby Ox-

ford University.

References

1. Verrall et al. Eur J Neurosci. 2007;26:1657.

2. Burnet et al. Mol Psych. 2008;13:658.

3. Sartorius et al. J Neurochem. 2006;96:1139.

4. Tunbridge et al. Am J Med Genet B Neuropsychiatr Genet. 2007;

144B:834.

ID: 551059

POLYMORPHIC VARIANTS OF THE TNXB

ASSOCIATED WITH COGNITION AND

SCHIZOPHRENIA SUSCEPTIBILITY

Karine R. Mayilyan

1,2

, R. Vakkalanka

, B. Kolachana

D. R. Weinberger

CBDB/ Genes, Cognition and Psychosis Program, IRP, NIMH,

NIH, Bethesda, MD, USA;

Institute of Molecular Biology,

Armenian National Academy of Sciences, Yerevan, Armenia

The TNXB is a gene of an extracellular matrix protein tenascin X (TNX)

spanning 68.2 kb on 6p21.3 in HLA-III region. TNX is important for the

proper deposition of collagen fibers in dermis and its deficiency causes an

autosomal recessive form of Ehlers-Danlos syndrome. TNX is also known

to be associated with growth of central and peripheral nerves. Recently, an

association between two SNPs of the TNXB (rs1009382 and rs204887) and

schizophrenia was observed in family trios from the UK. With moderate

significance, the results have been replicated in the Japanese but not in

the Chinese population. We have genotyped 5 SNPs of the TNXB in 51

schizophrenic patients (SP) þ 684 members of 176 European-American

families from the CBDB/NIMH ‘sibling study’. SNP genotyping was

performed on the ABI PRISM 7900HT System by using TaqManÒ

SNP Genotyping Assays chemistry (ABI). Single SNP and three-marker

haplotypes association analyses were calculated using the TDTPHASE

(version 2.37) based on 100 000 permutations. The FBAT was used to

test for a marker association with the clinical phenotype. Quantitative trait

analysis (ie, with cognitive phenotypes) was performed in parent-affected

child trios using QTDT (version-2.4.3). TNXB rs2071293 (C/T) and

rs185819 (Arg[T/C]His) SNPs were associated with schizophrenia.

FBAT analyses showed under-transmission of ‘‘C’’ and ‘‘T’’ alleles of

the SNPs to SP (z = 2.855; P < .005; z = 2.757; P < .004, respectively).

Two combinations of three-marker haplotypes were contributing to schizo-

phrenia as well involving those SNPs (1. rs2269429-rs204887-rs185819:

global P < .009;

rs185819-rs204900-rs2071293: global P < .008). As single

markers the other rs2269429, rs204887, rs204900 SNPs did not show any

difference in the transmission. Healthy subject’s data suggested a significant

association of the rs185819-rs204900-rs2071293 haplotype of the TNXB

with cognitive parameters, eg, verbal (global P < .01) and visual memories

(global P < .004), memory span (global P < .015), processing speed (global

P < .04) and results of Nback (global P < .04) and card sorting (global P <

.01) tests. We have not been able to replicate previous finding in this

European-American sample. However, our data have indicated two novel

SNPs of TNXB gene in association with schizophrenia either in or out of

haplotypes. Strikingly, the haplotype involving those SNPs was also

implicating to cognitive functioning. KRM acknowledges the USA Federal

Government contract #HHSN271200700340P.

ID: 550921

International Congress on Schizophrenia Research

7. 7. Genetics, Basic 99

TIMING OF PRENATAL DEVELOPMENTAL

PERTURBATIONS INDUCED BY MUTANT DISC1

RESULTING IN SCHIZOPHRENIA-LIKE

BEHAVIORAL AND PATHOLOGICAL

ALTERATIONS