- Email: [email protected]
Poster #M51 TEMPORAL LOBE WHITE MATTER ALTERATIONS IN SCHIZOPHRENIA: A DIFFUSION TENSOR IMAGING TRACTOMETRY FAMILY STUDY
Abstracts of the 4th Biennial Schizophrenia International Research Conference / Schizophrenia Research 153, Supplement 1 (2014) S1–S384
the healthy control group (irrespective of childhood trauma). Interestingly these abnormalities are equivalent to the ones we found comparing cases and controls irrespective of childhood trauma exposure (all p=0.01). Discussion: These preliminary results suggest that exposure to childhood trauma in people with psychosis may worsen their HPA axis response to a mild stressor. Furthermore, the analysis shows differences in brain structure between patients never exposed and patients exposed to trauma, highlighting a potential precious relationship between childhood abuse and brain structure. Remarkably the effect of childhood trauma on the brain structure seems to be present only in subject with vulnerability for psychosis and not in healthy controls. In order to further investigate this relationship, in future analyses we will explore whether the alterations in HPA we observed in individuals with childhood trauma mediate the brain abnormalities identified in these patients, thus representing a potential biological pathway by which a history of trauma increases the risk of psychosis.
Poster #M49 FUNCTIONAL AND GRAY MATTER ASYMMETRIES IN PATIENTS WITH SCHIZOPHRENIA AND BIPOLAR DISORDERS Sonia Dollfus 1 , Nicolas Delcroix 2 , Elise Leroux 3,4 , Mathieu Alary 3 , Annick Razafimandimby 3 , Perrine Brazo 1 , Pascal Delamillieure 1 1 CHU Caen; 2 CNRS, UMS 3408; 3 CNRS, UMR 6301, ISTS team; 4 CNRS, UMR 6301 ISTCT, ISTS team, GIP CYCERON, Bd Henri Becquerel, BP5229, F-14074 Caen cedex, France Background: It has been reported that patients with schizophrenia exhibit decreased functional hemispheric lateralization while patients with bipolar disorders do not. Few studies have evaluated the relationships between the hemispheric anatomical and functional asymmetry in language networks. The present study aimed to determine whether decreased leftward functional hemispheric lateralization could be related to the gray matter volume asymmetry. This investigation was the first to use an individual structural asymmetry index to analyze the gray matter specifically involved in a language network. Methods: Thirty-one right-handed patients with schizophrenia and 20 right-handed patients with bipolar disorders underwent a session of functional magnetic resonance imaging (fMRI) with a speech listening paradigm. Each group was matched with healthy subjects on gender, age and level of education. Maps of the Blood Oxygen Level Dependent (BOLD) signal contrast and structural maps (gray matter, white matter and CSF were generated in each participant in the MNI (Montreal Neurological Institute) space. Functional laterality indices (FLI) were calculated (Wilke, M. and Lidzba, K., 2007. LI-tool: a new toolbox to assess lateralization in functional MR-data. J Neurosci Methods. 163, 128–136) in each subject from the individual contrast maps. The gray matter volume asymmetry indices (GVAI) were computed in each subject from the gray matter maps masked with the individual functional maps. Anatomo-functional relationships were studied with ANCOVAs. Results: Patients with schizophrenia exhibited a significant decreased leftward functional hemispheric lateralization and patients with bipolar disorders did not. There was a positive correlation between GVAIs and FLIs (p<0.0001*) in healthy subjects, in patients with schizophrenia and in patients with bipolar disorders. Discussion: This study reports for the first time a significant relationship between the anatomical and functional asymmetry in healthy subjects, in patients with schizophrenia and in patients with bipolar disorders. While decreased leftward functional lateralization for language was observed in patients with schizophrenia compared to the control group, anatomofunctional relationships were observed in all groups of patients or healthy subjects and did not differ between groups. Thus, the decreased functional lateralization in patients with schizophrenia might not be due to a decreased gray matter volume asymmetry.
S207
Poster #M50 STRUCTURAL GREY MATTER AND WHITE MATTER DIFFERENCES IN INDIVIDUALS WITH PSYCHOTIC LIKE SIGNS FROM AN EPIDEMIOLOGICAL COHORT Mark Drakesmith 1 , Anirban Dutt 2 , Glyn Lewis 3 , Anthony S. David 2 , Derek K. Jones 1 1 Cardiff University; 2 Institute of Psychiatry, Kings College London; 3 University College, London Background: Identifying neuroanatomical correlates of psychotic symptoms in non-clinical population samples promises insights into the underlying neurobiology of psychosis unconfounded by illness and treatment. To explore this, we examined structural differences in the grey (GM) and white matter (WM) of individuals experiencing pre-clinical psychotic-like signs (PLIKS) and healthy controls without such symptoms. Methods: 252 subjects were selected from the Avon Longitudinal Study of Parents and Children (ALSPAC) birth cohort, on the basis of the presence or absence of psychotic experiences based on interviews conducted at 17 years of age. Those whose psychotic experiences were verified by trained researchers as suspected or definite (cases), and subjects with no such experiences (controls) were invited to undergo MRI scanning. At the time of scanning all subjects were 18 years old. All data were acquired on a 3T GE HDx MRI system. Structural data were acquired with a 3D-FSPGR sequence (TR = 7.8 ms, TE = 3.0 ms, FOV = 256×256 mm, 186 slices, voxel size = 1×1×1 mm3 ). HARDI data were acquired with a cardiac-gated EPI sequence (TE = 87 ms, 60 gradient orientations, b-value = 1200 smm-2, FOV = 96×96 mm, 60 slices, voxel-size = 1.6×1.6×2.4 mm). 3 PLIKS subjects and 1 control withdrew before the HARDI acquisition. GM volume was analysed using voxel-based morphometry (VBM) using the program FSL. GM volumes were tissue segmented, non-linearly registered to the MNI152 template and smoothed with a 2 mm3 Gaussian kernel. General linear model (GLM) analysis compared the PLIKS and control groups while co-varying for total brain volume. Statistics were corrected with cluster-enhanced permutation tests. HARDI data were analysed in ExploreDTI and corrected for motion, eddy current distortions and field inhomogeneities. Fractional anisotropy (FA), axial diffusivity (AD) and radial diffusivity (RD) was derived from a diffusion tensor model fitted using the RESTORE algorithm and corrected for partial volume effects. FA, AD and RD were analysed using tensor based spatial statistics (TBSS). DTI images were non-linearly registered to the FMRIB58_FA template and projected onto the template skeleton. Additional tractography was carried out using the damped Lucy-Richardson algorithm to identify tracts that traverse through statistically significant WM regions. Results: VBM analyses revealed a significant cluster of voxels with lower GM volumes in the left temporoparietal junction in the PLIKS group compared to controls without PLIKS. TBSS analysis data showed significant decreases in FA, AD and increases in RD in a white matter region in the left medial frontal lobe. Seeding tracts from this region implicate the left cingulum, anterior thalamic radiation and genu. Discussion: The results suggest some candidate biomarkers relevant to the psychosis spectrum. The lack of overlap in the GM and WM changes points to heterogeneity of processes potentially implicated in the development and pathogenesis of psychotic illness. The GM results may imply dysfunction in language processes while the WM results may implicate executive or motivation abnormalities.
Poster #M51 TEMPORAL LOBE WHITE MATTER ALTERATIONS IN SCHIZOPHRENIA: A DIFFUSION TENSOR IMAGING TRACTOMETRY FAMILY STUDY Vina Goghari 1 , Thibo Billiet 2 , Stefan Sunaert 2 , Louise Emsell 2 1 University of Calgary; 2 Leuven University Background: Individuals with schizophrenia consistently show abnormalities in white matter (WM) tracts compared to controls. Evidence also suggests that the non-psychotic relatives of individuals with schizophrenia show WM dysfunction. Given the relevance of the temporal lobe to the pathophysiology of schizophrenia, we investigated three key tracts associated with the temporal lobe: (1) the inferior longitudinal fasciculus, which connects the temporal lobe with the occipital lobe; (2) uncinate fasciculus,
S208
Abstracts of the 4th Biennial Schizophrenia International Research Conference / Schizophrenia Research 153, Supplement 1 (2014) S1–S384
which connects the temporal lobe with the prefrontal cortex; and (3) fornix, which connects the hippocampus with the hypothalamus. Furthermore, most previous studies investigating WM changes in schizophrenia have used voxel-based analysis, typically of a highly restricted WM skeleton (eg TBSS) or whole-tract mean values from tractography, thereby losing anatomical specificity. The goal of this investigation therefore, was to use sophisticated diffusion tensor imaging (DTI) tractometry analysis to examine along-tract microstructural differences in three major temporal lobe WM tracts in a family study of schizophrenia. Inclusion of both patients and family members allowed a better examination of genetic (familial) liability, as well as disease-related processes. Methods: Twenty-five patients with schizophrenia, 24 non-psychotic healthy relatives, and 27 community controls participated. DTI data were acquired (3T GE) along 60 gradient directions with b-value 1300 s/mm2 and corrected for motion and distortion using ExploreDTI. DTI parameters fractional anisotropy (FA) and mean and radial diffusivity (MD, RD) were estimated. A population-based FA-template was constructed, to which all parameter maps were registered. The aforementioned temporal lobe tracts were isolated using targeted tractography. The fornix was further subdivided into left/right anterior columns, body and left/right fimbriae. Parameter values for each subject were calculated at evenly distributed points along the length of the tract masks in template space. ANCOVA was used to investigate the main effect of group membership on mean tract values (between-subject factors: group, gender, covariate: age), whilst the group:position interaction term in a 2-way ANOVA was used to identify group differences at each point along each tract. Results: Both patients with schizophrenia and non-psychotic relatives demonstrated lower mean and radial diffusivity in their fornix compared to controls using age and gender as covariates. Tract profiles illustrated group differences along the entire length of the fornix rather than localized differences. No other significant differences were found between groups. Discussion: We found both schizophrenia patients and their biological relatives had differences in two indices associated with WM microstructure, compared to controls along the length of the fornix; however, in an unexpected direction. Possible reasons for this surprising finding could be a true biological difference, volumetric differences present between groups, or methodological issues associated with DTI analyses in typically sized schizophrenia study populations. Further investigation of the relationship between the DTI parameters, white matter volume and anatomical structures bordering the fornix will be presented to understand these findings. The results of this analysis may allow a better understanding of not only these results, but also other findings of fractional anisotropy changes in schizophrenia.
Poster #M52 NEUROLOGICAL SOFT SIGNS AND BRAIN MORPHOLOGY IN PATIENTS WITH CHRONIC SCHIZOPHRENIA Christina J. Herold 1 , Marc M. Lässer 2 , Lena A. Schmid 2 , Ulrich Seidl 3 , Marco Essig 4 , Philipp A. Thomann 5 , Johannes Schröder 1 1 Institute of Gerontology/Section of Geriatric Psychiatry, University of Heidelberg; 2 Section of Geriatric Psychiatry, University of Heidelberg, Germany; 3 Center for Mental Health, Department of Psychiatry, Stuttgart, Germany; 4 University Hospital Erlangen, Department of Neuroradiology; 5 Department of General Psychiatry, Heidelberg Background: While neurological soft signs are frequently described in patients with schizophrenia the underlying cerebral changes are rather scarcely investigated especially in patients with a chronic course of the disease. Methods: Therefore we examined a group of 37 patients with a duration of the disease of 28.7 years (±11.3 years) and 24 healthy subjects matched for age, gender and education. Neurological soft signs were assessed with the Heidelberg Scale after remission of the acute symptoms and correlated to grey matter volume using optimized voxel-based morphometry (VBM), the preprocessing steps of VBM have been improved with the DARTEL-algorithm. Results: 18 patients had a NSS sum-score of 10.1 (±10.3), 19 patients had a NSS sum-score of 28.6 (±10.6), while the NSS sum-score in the control group was significantly reduced with 4.1 (±3.5). The NSS low-score patient group had significantly increased grey matter volume than the NSS high-
score patient group in the inferior occipital gyrus, the anterior lobe of the right cerebellum (culmen) and the left thalamus. Discussion: These preliminary results are similar to those patterns found in patients with a first-episode and support at least partly the model of cognitive dysmetria with a dysfunctional prefrontal-thalamic-cerebellar circuitry in schizophrenia.
Poster #M53 DIAGNOSING SCHIZOPHRENIA USING NEUROIMAGING: A META-ANALYSIS OF MULTIVARIATE PATTERN RECOGNITION STUDIES Joseph Kambeitz 1 , Lana Kambeitz-Ilankovic 1 , Leucht Stefan 2 , Steven Wood 3 , Christos Davatzikos 4 , Berend Malchow 1 , Peter Falkai 1 , Nikolaos Koutsouleris 5 1 Department of Psychiatry, LMU University of Munich, Germany; 2 Department of Psychiatry, Technical University Munich; 3 School of Psychology, University of Birmingham; 4 University of Pennsylvania; 5 Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich Background: Numerous studies have applied novel multivariate statistical approaches to the analysis of brain alterations in patients with schizophrenia. However, these studies differ with respect to multiple aspects such as the demographical characteristics of the investigated populations or the methodological details of the conducted analysis. As a result, the diagnostic accuracy of the reported predictive models differ largely, making it difficult to evaluate the overall potential of these studies to inform clinical diagnosis. Methods: We conducted a comprehensive literature search to identify all studies reporting performance of neuroimaging-based multivariate predictive models for the differentiation of patients with schizophrenia from healthy control subjects. A bivariate random-effects meta-analytic model was implemented. The robustness of the results as well as the effect of potentially confounding continous variables (e.g. age, gender ratio, year of publication) was investigated by adding moderator variables to the bivariate regression model. All computations were performed using the R statistical programming language with the package mada. Results: The final sample consisted of n=37 studies studies including n=1491 patients with schizophrenia and n=1488 healthy controls. Meta-analysis of the complete sample showed a sensitivity of 80.7% (95%-CI: 77.0 to 83.9%) and a specificity of 80.2% (95%-CI: 83.3 to 76.7%). There was no evidence for a publication bias and no effect of year of publication. Separate analysis for the different imaging modalities showed similar diagnostic accuracy for the structural MRI studies (sensitivity 77.3%, specificity 78.7%), the fMRI studies (sensitivity 81.4%, specificity 82.4%) and resting-state fMRI studies (sensitivity 86.9%, specificity 80.3%). Moderator analysis showed significant effects of age of patients on sensitivity (p=0.021) and of positive-to-negative symptom ratio on specificity (p=0.028) indicating better diagnostic accuracy in older patients and patients with positive symptoms. There was no significant effect of gender-ratio and no significant difference between different multivariate statistical approaches (all p>0.1). Discussion: Our analysis indicate an overall sensitivity and overall specificity of around 80% of neuroimaging-based predictive models for differentiating schizophrenic patients from healthy controls. This finding was robust against the inclusion of potential confounding factors. Thus our results underline the potential applicability of neuroimaging-based predictive models for the diagnosis of schizophrenia.
Poster #M54 THE ROLE OF A FOXP2 VARIANT ON BRAIN STRUCTURE AND SPEECH PRODUCTION – A DTI STUDY Axel Krug 1 , Davide Laneri 2 , Bruno Dietsche 1 , Heidelore Backes 1 , Stephanie Witt 3 , Marcella Rietschel 3 , Jens Sommer 1 , Tilo Kircher 1,2 , Arne Nagels 2 1 University of Marburg; 2 Dept. of Psychiatry and Psychotherapy, Philipps-University Marburg; 3 ZI Mannheim Background: Mutations in severe speech disturbances. have been associated with related psychopathology in
the FOXP2 gene have been associated with In addition, several variants within this gene schizophrenia in general as well as speechpatients. The aim of the current study was
the healthy control group (irrespective of childhood trauma). Interestingly these abnormalities are equivalent to the ones we found comparing cases and controls irrespective of childhood trauma exposure (all p=0.01). Discussion: These preliminary results suggest that exposure to childhood trauma in people with psychosis may worsen their HPA axis response to a mild stressor. Furthermore, the analysis shows differences in brain structure between patients never exposed and patients exposed to trauma, highlighting a potential precious relationship between childhood abuse and brain structure. Remarkably the effect of childhood trauma on the brain structure seems to be present only in subject with vulnerability for psychosis and not in healthy controls. In order to further investigate this relationship, in future analyses we will explore whether the alterations in HPA we observed in individuals with childhood trauma mediate the brain abnormalities identified in these patients, thus representing a potential biological pathway by which a history of trauma increases the risk of psychosis.
Poster #M49 FUNCTIONAL AND GRAY MATTER ASYMMETRIES IN PATIENTS WITH SCHIZOPHRENIA AND BIPOLAR DISORDERS Sonia Dollfus 1 , Nicolas Delcroix 2 , Elise Leroux 3,4 , Mathieu Alary 3 , Annick Razafimandimby 3 , Perrine Brazo 1 , Pascal Delamillieure 1 1 CHU Caen; 2 CNRS, UMS 3408; 3 CNRS, UMR 6301, ISTS team; 4 CNRS, UMR 6301 ISTCT, ISTS team, GIP CYCERON, Bd Henri Becquerel, BP5229, F-14074 Caen cedex, France Background: It has been reported that patients with schizophrenia exhibit decreased functional hemispheric lateralization while patients with bipolar disorders do not. Few studies have evaluated the relationships between the hemispheric anatomical and functional asymmetry in language networks. The present study aimed to determine whether decreased leftward functional hemispheric lateralization could be related to the gray matter volume asymmetry. This investigation was the first to use an individual structural asymmetry index to analyze the gray matter specifically involved in a language network. Methods: Thirty-one right-handed patients with schizophrenia and 20 right-handed patients with bipolar disorders underwent a session of functional magnetic resonance imaging (fMRI) with a speech listening paradigm. Each group was matched with healthy subjects on gender, age and level of education. Maps of the Blood Oxygen Level Dependent (BOLD) signal contrast and structural maps (gray matter, white matter and CSF were generated in each participant in the MNI (Montreal Neurological Institute) space. Functional laterality indices (FLI) were calculated (Wilke, M. and Lidzba, K., 2007. LI-tool: a new toolbox to assess lateralization in functional MR-data. J Neurosci Methods. 163, 128–136) in each subject from the individual contrast maps. The gray matter volume asymmetry indices (GVAI) were computed in each subject from the gray matter maps masked with the individual functional maps. Anatomo-functional relationships were studied with ANCOVAs. Results: Patients with schizophrenia exhibited a significant decreased leftward functional hemispheric lateralization and patients with bipolar disorders did not. There was a positive correlation between GVAIs and FLIs (p<0.0001*) in healthy subjects, in patients with schizophrenia and in patients with bipolar disorders. Discussion: This study reports for the first time a significant relationship between the anatomical and functional asymmetry in healthy subjects, in patients with schizophrenia and in patients with bipolar disorders. While decreased leftward functional lateralization for language was observed in patients with schizophrenia compared to the control group, anatomofunctional relationships were observed in all groups of patients or healthy subjects and did not differ between groups. Thus, the decreased functional lateralization in patients with schizophrenia might not be due to a decreased gray matter volume asymmetry.
S207
Poster #M50 STRUCTURAL GREY MATTER AND WHITE MATTER DIFFERENCES IN INDIVIDUALS WITH PSYCHOTIC LIKE SIGNS FROM AN EPIDEMIOLOGICAL COHORT Mark Drakesmith 1 , Anirban Dutt 2 , Glyn Lewis 3 , Anthony S. David 2 , Derek K. Jones 1 1 Cardiff University; 2 Institute of Psychiatry, Kings College London; 3 University College, London Background: Identifying neuroanatomical correlates of psychotic symptoms in non-clinical population samples promises insights into the underlying neurobiology of psychosis unconfounded by illness and treatment. To explore this, we examined structural differences in the grey (GM) and white matter (WM) of individuals experiencing pre-clinical psychotic-like signs (PLIKS) and healthy controls without such symptoms. Methods: 252 subjects were selected from the Avon Longitudinal Study of Parents and Children (ALSPAC) birth cohort, on the basis of the presence or absence of psychotic experiences based on interviews conducted at 17 years of age. Those whose psychotic experiences were verified by trained researchers as suspected or definite (cases), and subjects with no such experiences (controls) were invited to undergo MRI scanning. At the time of scanning all subjects were 18 years old. All data were acquired on a 3T GE HDx MRI system. Structural data were acquired with a 3D-FSPGR sequence (TR = 7.8 ms, TE = 3.0 ms, FOV = 256×256 mm, 186 slices, voxel size = 1×1×1 mm3 ). HARDI data were acquired with a cardiac-gated EPI sequence (TE = 87 ms, 60 gradient orientations, b-value = 1200 smm-2, FOV = 96×96 mm, 60 slices, voxel-size = 1.6×1.6×2.4 mm). 3 PLIKS subjects and 1 control withdrew before the HARDI acquisition. GM volume was analysed using voxel-based morphometry (VBM) using the program FSL. GM volumes were tissue segmented, non-linearly registered to the MNI152 template and smoothed with a 2 mm3 Gaussian kernel. General linear model (GLM) analysis compared the PLIKS and control groups while co-varying for total brain volume. Statistics were corrected with cluster-enhanced permutation tests. HARDI data were analysed in ExploreDTI and corrected for motion, eddy current distortions and field inhomogeneities. Fractional anisotropy (FA), axial diffusivity (AD) and radial diffusivity (RD) was derived from a diffusion tensor model fitted using the RESTORE algorithm and corrected for partial volume effects. FA, AD and RD were analysed using tensor based spatial statistics (TBSS). DTI images were non-linearly registered to the FMRIB58_FA template and projected onto the template skeleton. Additional tractography was carried out using the damped Lucy-Richardson algorithm to identify tracts that traverse through statistically significant WM regions. Results: VBM analyses revealed a significant cluster of voxels with lower GM volumes in the left temporoparietal junction in the PLIKS group compared to controls without PLIKS. TBSS analysis data showed significant decreases in FA, AD and increases in RD in a white matter region in the left medial frontal lobe. Seeding tracts from this region implicate the left cingulum, anterior thalamic radiation and genu. Discussion: The results suggest some candidate biomarkers relevant to the psychosis spectrum. The lack of overlap in the GM and WM changes points to heterogeneity of processes potentially implicated in the development and pathogenesis of psychotic illness. The GM results may imply dysfunction in language processes while the WM results may implicate executive or motivation abnormalities.
Poster #M51 TEMPORAL LOBE WHITE MATTER ALTERATIONS IN SCHIZOPHRENIA: A DIFFUSION TENSOR IMAGING TRACTOMETRY FAMILY STUDY Vina Goghari 1 , Thibo Billiet 2 , Stefan Sunaert 2 , Louise Emsell 2 1 University of Calgary; 2 Leuven University Background: Individuals with schizophrenia consistently show abnormalities in white matter (WM) tracts compared to controls. Evidence also suggests that the non-psychotic relatives of individuals with schizophrenia show WM dysfunction. Given the relevance of the temporal lobe to the pathophysiology of schizophrenia, we investigated three key tracts associated with the temporal lobe: (1) the inferior longitudinal fasciculus, which connects the temporal lobe with the occipital lobe; (2) uncinate fasciculus,
S208
Abstracts of the 4th Biennial Schizophrenia International Research Conference / Schizophrenia Research 153, Supplement 1 (2014) S1–S384
which connects the temporal lobe with the prefrontal cortex; and (3) fornix, which connects the hippocampus with the hypothalamus. Furthermore, most previous studies investigating WM changes in schizophrenia have used voxel-based analysis, typically of a highly restricted WM skeleton (eg TBSS) or whole-tract mean values from tractography, thereby losing anatomical specificity. The goal of this investigation therefore, was to use sophisticated diffusion tensor imaging (DTI) tractometry analysis to examine along-tract microstructural differences in three major temporal lobe WM tracts in a family study of schizophrenia. Inclusion of both patients and family members allowed a better examination of genetic (familial) liability, as well as disease-related processes. Methods: Twenty-five patients with schizophrenia, 24 non-psychotic healthy relatives, and 27 community controls participated. DTI data were acquired (3T GE) along 60 gradient directions with b-value 1300 s/mm2 and corrected for motion and distortion using ExploreDTI. DTI parameters fractional anisotropy (FA) and mean and radial diffusivity (MD, RD) were estimated. A population-based FA-template was constructed, to which all parameter maps were registered. The aforementioned temporal lobe tracts were isolated using targeted tractography. The fornix was further subdivided into left/right anterior columns, body and left/right fimbriae. Parameter values for each subject were calculated at evenly distributed points along the length of the tract masks in template space. ANCOVA was used to investigate the main effect of group membership on mean tract values (between-subject factors: group, gender, covariate: age), whilst the group:position interaction term in a 2-way ANOVA was used to identify group differences at each point along each tract. Results: Both patients with schizophrenia and non-psychotic relatives demonstrated lower mean and radial diffusivity in their fornix compared to controls using age and gender as covariates. Tract profiles illustrated group differences along the entire length of the fornix rather than localized differences. No other significant differences were found between groups. Discussion: We found both schizophrenia patients and their biological relatives had differences in two indices associated with WM microstructure, compared to controls along the length of the fornix; however, in an unexpected direction. Possible reasons for this surprising finding could be a true biological difference, volumetric differences present between groups, or methodological issues associated with DTI analyses in typically sized schizophrenia study populations. Further investigation of the relationship between the DTI parameters, white matter volume and anatomical structures bordering the fornix will be presented to understand these findings. The results of this analysis may allow a better understanding of not only these results, but also other findings of fractional anisotropy changes in schizophrenia.
Poster #M52 NEUROLOGICAL SOFT SIGNS AND BRAIN MORPHOLOGY IN PATIENTS WITH CHRONIC SCHIZOPHRENIA Christina J. Herold 1 , Marc M. Lässer 2 , Lena A. Schmid 2 , Ulrich Seidl 3 , Marco Essig 4 , Philipp A. Thomann 5 , Johannes Schröder 1 1 Institute of Gerontology/Section of Geriatric Psychiatry, University of Heidelberg; 2 Section of Geriatric Psychiatry, University of Heidelberg, Germany; 3 Center for Mental Health, Department of Psychiatry, Stuttgart, Germany; 4 University Hospital Erlangen, Department of Neuroradiology; 5 Department of General Psychiatry, Heidelberg Background: While neurological soft signs are frequently described in patients with schizophrenia the underlying cerebral changes are rather scarcely investigated especially in patients with a chronic course of the disease. Methods: Therefore we examined a group of 37 patients with a duration of the disease of 28.7 years (±11.3 years) and 24 healthy subjects matched for age, gender and education. Neurological soft signs were assessed with the Heidelberg Scale after remission of the acute symptoms and correlated to grey matter volume using optimized voxel-based morphometry (VBM), the preprocessing steps of VBM have been improved with the DARTEL-algorithm. Results: 18 patients had a NSS sum-score of 10.1 (±10.3), 19 patients had a NSS sum-score of 28.6 (±10.6), while the NSS sum-score in the control group was significantly reduced with 4.1 (±3.5). The NSS low-score patient group had significantly increased grey matter volume than the NSS high-
score patient group in the inferior occipital gyrus, the anterior lobe of the right cerebellum (culmen) and the left thalamus. Discussion: These preliminary results are similar to those patterns found in patients with a first-episode and support at least partly the model of cognitive dysmetria with a dysfunctional prefrontal-thalamic-cerebellar circuitry in schizophrenia.
Poster #M53 DIAGNOSING SCHIZOPHRENIA USING NEUROIMAGING: A META-ANALYSIS OF MULTIVARIATE PATTERN RECOGNITION STUDIES Joseph Kambeitz 1 , Lana Kambeitz-Ilankovic 1 , Leucht Stefan 2 , Steven Wood 3 , Christos Davatzikos 4 , Berend Malchow 1 , Peter Falkai 1 , Nikolaos Koutsouleris 5 1 Department of Psychiatry, LMU University of Munich, Germany; 2 Department of Psychiatry, Technical University Munich; 3 School of Psychology, University of Birmingham; 4 University of Pennsylvania; 5 Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich Background: Numerous studies have applied novel multivariate statistical approaches to the analysis of brain alterations in patients with schizophrenia. However, these studies differ with respect to multiple aspects such as the demographical characteristics of the investigated populations or the methodological details of the conducted analysis. As a result, the diagnostic accuracy of the reported predictive models differ largely, making it difficult to evaluate the overall potential of these studies to inform clinical diagnosis. Methods: We conducted a comprehensive literature search to identify all studies reporting performance of neuroimaging-based multivariate predictive models for the differentiation of patients with schizophrenia from healthy control subjects. A bivariate random-effects meta-analytic model was implemented. The robustness of the results as well as the effect of potentially confounding continous variables (e.g. age, gender ratio, year of publication) was investigated by adding moderator variables to the bivariate regression model. All computations were performed using the R statistical programming language with the package mada. Results: The final sample consisted of n=37 studies studies including n=1491 patients with schizophrenia and n=1488 healthy controls. Meta-analysis of the complete sample showed a sensitivity of 80.7% (95%-CI: 77.0 to 83.9%) and a specificity of 80.2% (95%-CI: 83.3 to 76.7%). There was no evidence for a publication bias and no effect of year of publication. Separate analysis for the different imaging modalities showed similar diagnostic accuracy for the structural MRI studies (sensitivity 77.3%, specificity 78.7%), the fMRI studies (sensitivity 81.4%, specificity 82.4%) and resting-state fMRI studies (sensitivity 86.9%, specificity 80.3%). Moderator analysis showed significant effects of age of patients on sensitivity (p=0.021) and of positive-to-negative symptom ratio on specificity (p=0.028) indicating better diagnostic accuracy in older patients and patients with positive symptoms. There was no significant effect of gender-ratio and no significant difference between different multivariate statistical approaches (all p>0.1). Discussion: Our analysis indicate an overall sensitivity and overall specificity of around 80% of neuroimaging-based predictive models for differentiating schizophrenic patients from healthy controls. This finding was robust against the inclusion of potential confounding factors. Thus our results underline the potential applicability of neuroimaging-based predictive models for the diagnosis of schizophrenia.
Poster #M54 THE ROLE OF A FOXP2 VARIANT ON BRAIN STRUCTURE AND SPEECH PRODUCTION – A DTI STUDY Axel Krug 1 , Davide Laneri 2 , Bruno Dietsche 1 , Heidelore Backes 1 , Stephanie Witt 3 , Marcella Rietschel 3 , Jens Sommer 1 , Tilo Kircher 1,2 , Arne Nagels 2 1 University of Marburg; 2 Dept. of Psychiatry and Psychotherapy, Philipps-University Marburg; 3 ZI Mannheim Background: Mutations in severe speech disturbances. have been associated with related psychopathology in
the FOXP2 gene have been associated with In addition, several variants within this gene schizophrenia in general as well as speechpatients. The aim of the current study was