Structural Brain Abnormalities in Borderline Personality Disorder Correlate with Clinical Severity and Predict Psychotherapy Response

Although previous imaging studies in borderline personality disorder (BPD) have found brain abnormalities, the results have been inconsistent. This study aimed to investigate structural brain abnormalities using voxel-based morphometry (VBM) and cortical thickness (Cth) analyses in a large sample of patients with BPD. Additionally, we aimed to determine the correlation between structural abnormalities and clinical severity and to assess its potential value at predicting psychotherapeutic response. Sixty-one individuals with BPD and 19 healthy controls underwent magnetic resonance imaging. Participants with BPD completed several self-report clinical scales, received dialectical-behavioral therapy skills training and post-therapy changes in clinical scores were also recorded. Gray matter volume (GMV) and Cth differences between groups were compared. Within the BPD group, we further characterized the structural brain correlates of clinical severity and investigated the relationship between pre-therapy structural abnormalities and therapeutic response. As potential confounders we included age, sex, educational level, and total intracranial volume (the latter only in VBM analyses). Compared to controls, the BPD group showed a reduced GMV/Cth in prefrontal areas but increased GMV in the limbic structures (amygdala and parahippocampal regions). Prefrontal abnormalities correlated with higher baseline scores on impulsivity and general BPD severity. Increased GMV in the parahippocampal area correlated with a greater emotion dysregulation. Importantly, several baseline structural abnormalities correlated with worse response to psychotherapy. Patients with BPD showed a reduced GMV in the prefrontal areas but a greater GMV in the limbic structures. Several structural abnormalities (i.e. middle and inferior prefrontal areas, anterior insula, or parahippocampal area) correlated with clinical severity and could potentially be used as imaging biological correlates biomarkers to predict psychotherapy response.