Brain structural covariance network centrality in maltreated youth with PTSD and in maltreated youth resilient to PTSD

Child maltreatment is a major cause of pediatric posttraumatic stress disorder (PTSD). Previous studies have not investigated potential differences in network architecture in maltreated youth with PTSD and those resilient to PTSD. High-resolution magnetic resonance imaging brain scans at 3 T were completed in maltreated youth with PTSD (n = 31), without PTSD (n = 32), and nonmaltreated controls (n = 57). Structural covariance network architecture was derived from between-subject intraregional correlations in measures of cortical thickness in 148 cortical regions (nodes). Interregional positive partial correlations controlling for demographic variables were assessed, and those correlations that exceeded specified thresholds constituted connections in cortical brain networks. Four measures of network centrality characterized topology, and the importance of cortical regions (nodes) within the network architecture were calculated for each group. Permutation testing and principle component analysis method were employed to calculate between-group differences. Principle component analysis is a methodological improvement to methods used in previous brain structural covariance network studies. Differences in centrality were observed between groups. Larger centrality was found in maltreated youth with PTSD in the right posterior cingulate cortex; smaller centrality was detected in the right inferior frontal cortex compared to youth resilient to PTSD and controls, demonstrating network characteristics unique to pediatric maltreatment-related PTSD. Larger centrality was detected in right frontal pole in maltreated youth resilient to PTSD compared to youth with PTSD and controls, demonstrating structural covariance network differences in youth resilience to PTSD following maltreatment. Smaller centrality was found in the left posterior cingulate cortex and in the right inferior frontal cortex in maltreated youth compared to controls, demonstrating attributes of structural covariance network topology that is unique to experiencing maltreatment. This work is the first to identify cortical thickness-based structural covariance network differences between maltreated youth with and without PTSD. We demonstrated network differences in both networks unique to maltreated youth with PTSD and those resilient to PTSD. The networks identified are important for the successful attainment of age-appropriate social cognition, attention, emotional processing, and inhibitory control. Our findings in maltreated youth with PTSD versus those without PTSD suggest vulnerability mechanisms for developing PTSD.