BACKGROUND: The current understanding of the neural correlates of borderline personality disorder (BPD) is limited, but some evidence suggests that alterations in limbic structures play a role in adult BPD. The developmental course of structural neural differences in BPD is unknown. Whether there is specificity for structural alterations in BPD compared to other psychiatric presentations, such as major depressive disorder (MDD), remains unexplored. In the current study, we examined childhood trajectories of 2 limbic regions that have been implicated in BPD, hippocampal and amygdala volume, as they relate to adolescent BPD symptoms compared to MDD symptoms. METHODS: Participants (n = 175
85 [48.6%] female) were from a 17-year longitudinal study of preschool depression. Participants completed up to 5 magnetic resonance imaging scans from late childhood through adolescence. General linear models were used to examine the relationship between gray matter volume intercepts/slopes and BPD symptoms to understand the influence of the developmental trajectory of brain regions on BPD. Separate models were used to examine the relationship between MDD symptoms and volume intercepts to assess diagnostic specificity. RESULTS: Lower childhood amygdala volume (intercept
age 13 centered) across scans was associated with higher adolescent BPD symptoms (β = -0.25, adjusted p = .015). There was no relationship between the slope of amygdala volume and BPD symptoms. There was no relationship between hippocampal volume and BPD or any relationship between amygdala or hippocampal volume and MDD symptoms during adolescence. CONCLUSIONS: Our findings add evidence that supports the role of alterations in amygdala structure in BPD development. Decreased amygdala volume as early as age 13 may be an early indicator of the development of BPD during adolescence.