S162. MULTI-MODAL ANALYSIS OF THE EFFECTS OF URBAN UPBRINGING ON BRAIN STRUCTURE: THE FOR2107 COHORT. (18th May 2020)

• Record Type: Journal Article
• Authors: Nenadic, Igor; Weigardt, Lukas; Schmitt, Simon; Meller, Tina; Stein, Frederike; Brosch, Katharina; Dannlowski, Udo; Krug, Axel; Kircher, Tilo
• Is Part Of: Schizophrenia bulletin. Volume 46(2020)Supplement 1
• Page Start: S98
• Page End: S98
• Publication Date: 2020-05-18
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbaa031.228
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: Urbanicity has been identified as a major environmental risk factor for schizophrenia as well as other major mental disorders. While initial structural brain imaging studies have pointed to medial and lateral prefrontal cortices being associated with urbanicity during childhood and adolescence, most of these studies have been limited to smaller samples and single analysis methods. The present study used the large ongoing FOR2107 multi-centre study cohort (Kircher et al., 2018) to analyse associations of urban upbringing. Methods: We analysed a data set of n=625 healthy subjects without a current or previous psychiatric disorder (ascertained through SCID-I interviews), who underwent 3 Tesla MRI scanning, obtaining a high-resolution T1-weighted MPRAGE and a DTI scan. We subsequently also analysed a pilot samples of 42 patients with DSM-IV schizophrenia. We obtained data on urban upbringing (Lederbogen score; Lederbogen et al., 2014) for the first 15 years of life, as well as number of moves. T1 data were pre-processed using CAT12 software, using surface-based morphometric analysis of cortical thickness (CT) with 15 mm smoothing. DTI data were analysed using the TBSS approach in FSL software. We used general linear models to calculate multiple regression analyses using both linear and quadratic (non-linear) associations with urbanicity scores, followed up by analyses of correlations with number of relocations (as unspecific stress factors). Analyses of CT and DTI were each corrected for multiple comparisons using FWE. Results: In the healthy subject cohort we identified a significant negative linear correlation (p=0.042 FWE cluster-level; p=0.014 peak-level) between urban upbringing scores and cortical thickness (CT) in a right precuneus / posterior cingulate cluster (x/y/z=35;-88;-15), while non-linear analysis identified an additional trend in the left occipital cortex (p=0.073 FWE cluster-level; -17;-100;15). We did not find significant effects for number of relocations/moves. DTI analysis of fractional anisotropy (FA) showed a significant association (all p<0.05 FWE-corrected) for the uncinate fasciculus and inferior fronto-occipital fasciculus. CT analysis in the schizophrenia pilot cohort showed similar effects, but in a more dorsal precuneus cluster (6;-56;45) only at uncorrected levels. Discussion: Our study identified structural variation in cortical thickness in the precuneus / posterior cingulate cortex of healthy subjects, regions linked to abnormal DMN activity and stress. While the trend-level finding in schizophrenia patients was located in an adjacent cluster, our findings can be interpreted as these medial parietal lobe structure mediating this particular risk factor. Our findings argue against a more wide-spread non-specific effect, as seen in some earlier smaller studies, but points to distinct neuronal network as mediators of this particular risk. The identified brain regions are linked to stress. Unlike previous prefrontal findings, they suggest a new link to the precuneus, a central hub of the default mode network. Given that these effects were observed in clinically healthy subjects, our findings also carry implications for a better understanding of the macro-environment in adolescence.