91. Structural Brain Volumes in Youth With Psychotic and Bipolar Spectrum Symptoms

Abstract

Background: Structural brain abnormalities have been observed in people with overt bipolar disorder and schizophrenia yet less is known about when during the developmental course of the illnesses such abnormalities manifest. Methods: The Philadelphia Neurodevelopmental Cohort is a community-based sample of 8740 youth, ages 8-21. A subsample of 989 underwent structural magnetic resonance neuroimaging, from which bilateral volumetric measures were calculated. In addition, study participants were assessed for psychiatric symptoms. Psychosis spectrum diagnosis was determined based on a combination of responses to the Prime Screen Revised, endorsement of K-SADS hallucinations or score of two standard deviations or more above age peers on the SIPS. Bipolar spectrum diagnosis was determined based on two or more depression and two or more mania symptoms on the K-SADS, taking into account duration and severity of symptoms. Comparisons of brain volumetric measures were made using linear regression models, adjusted for age and gender, with statistical signifcance calculated for multiple tests using false discovery rate correction. Results: For the 989 youths who underwent imaging, 117 endorsed bipolar spectrum symptoms and 113 endorsed psychosis spectrum symptoms. Each symptom group was compared to 650 participants who did not endorse either. After FDR correction, only one brain region (transverse temporal volume; P =.0001) was signifcantly reduced in the bipolar spectrum group relative to youth without bipolar/psychotic symptoms. In contrast, widespread volumetric reductions were observed in the psychosis spectrum group relative to youths without such symptoms, with signifcant (FDR-corrected) reductions in total brain volume and 27 of 34 regions of interest, most prominently in heteromodal association cortex and cingulate regions. Conclusion: Prior research on individuals with established illness has typically indicated more severe volumetric decrements in those with psychotic disorder compared with bipolar disorder. Our study suggests that these neuroanatomic patterns are observed in non-clinically ascertained individuals as well, as youth with psychotic spectrum symptoms, but not those with bipolar symptoms, show signifcant widespread volumetric defcits, in regions that largely recapitulate those observed in adult patients with established illness. Whether these early life brain abnormalities are associated with subsequent onset of overt psychosis or bipolar disorder remains to be determined. These fndings point to potential biological distinctions between psychotic and bipolar spectrum conditions, which may suggest additional biomarkers relevant to early identifcation.