Localizing the structural and functional alteration networks associated with borderline personality disorder.
BACKGROUND Although structural and functional alterations in borderline personality disorder (BPD) have been widely reported in neuroimaging studies, the findings remain heterogeneous. Emerging consensus suggests that these distributed alterations may converge on a common pathological network associated with BPD, providing a network-level framework for elucidating the neural basis of the disorder. METHODS We integrated findings from 96 publications covering 2325 patients with BPD and 2356 healthy controls across gray matter volume, task-induced activation, and resting-state activity modalities. Using the functional connectivity network mapping (FCNM) approach, we projected the reported alterations onto a large-scale functional connectome (n = 872) to localize brain alteration networks associated with BPD. We then evaluated the spatial relationship between these networks and 19 neurotransmitter distribution maps. RESULTS We identified three distinct networks. The gray matter volume alteration network involved the precentral/postcentral gyrus and temporal gyrus, predominantly within the somatomotor (12.24%) and subcortical networks (8.65%). The task-induced activation network was mainly located in the insula/putamen, primarily associated with the ventral attention network (18.83%). The resting-state network encompassed the superior frontal gyrus, precuneus, and angular gyrus, dominated by the default mode network (53.86%). The structural network exhibited spatial correlations with the serotonin transporter, while the task-induced network was associated with noradrenergic and cholinergic systems. CONCLUSIONS These findings suggest BPD might be characterized by structural deficits in somatic-emotional integration and functional dysregulation in salience detection and self-referential processing. This network-level framework helps address previously existing inconsistencies and suggests that these brain alterations may be related to specific neurotransmitters.