Childhood Trauma and Cortical Thickness in Bipolar Disorder with and without Psychosis: Effects by Diagnostic Group and Trauma Subtype

Abstract

Abstract Background Bipolar disorder affects about 2-3% of the population and is associated with cortical abnormalities, with bipolar disorder with psychosis (BDP) reported to show more pronounced changes. It remains unclear whether childhood trauma is associated with cortical thickness differences across bipolar disorder subgroups, particularly in BDP. Methods Using the publicly available PARDIP (Psychosis and Affective Research Domains and Intermediate Phenotypes) data set, we examined whether childhood trauma was associated with cortical thickness differences among 120 individuals: 33 with bipolar disorder without psychosis (BD; 36% male), 38 with bipolar disorder with psychosis (BDP; 47% male), and 49 healthy controls (HC; 61% male). Participants completed the Childhood Trauma Questionnaire-Short Form (CTQ-SF), and T1-weighted MRI scans were processed with FreeSurfer to extract cortical thickness across the 68 Desikan-Killiany regions. Linear mixed effects models tested associations between CTQ subtype scores and cortical thickness and trauma-by-diagnosis interactions, with sensitivity analyses for medication class and illness duration following the ENIGMA guidelines. Results No cortical thickness difference survived correction for multiple testing. At the nominal level, BD showed reduced thickness relative to HC in cingulate, frontal and occipital regions, whereas BDP showed greater superior temporal thickness relative to BD. Sexual abuse was nominally associated with reduced temporal thickness. Exploratory trauma-by-diagnosis interactions suggested region- and subtype-specific patterns that did not survive correction. Medication class and illness duration showed no significant effects after FDR correction. Conclusions These exploratory findings suggest nominal, region-specific associations between childhood trauma subtypes and cortical thickness in bipolar disorder. However, no findings survived correction for multiple comparisons, so they should be read as preliminary and need replication in larger samples.