Behavioral problems after early life stress: contributions of the hippocampus and amygdala.

BACKGROUND: Early life stress (ELS) can compromise development, with higher amounts of adversity linked to behavioral problems. To understand this linkage, a growing body of research has examined two brain regions involved with socioemotional functioning-amygdala and hippocampus. Yet empirical studies have reported increases, decreases, and no differences within human and nonhuman animal samples exposed to different forms of ELS. This divergence in findings may stem from methodological factors, nonlinear effects of ELS, or both. METHODS: We completed rigorous hand-tracing of the amygdala and hippocampus in three samples of children who experienced different forms of ELS (i.e., physical abuse, early neglect, or low socioeconomic status). Interviews were also conducted with children and their parents or guardians to collect data about cumulative life stress. The same data were also collected in a fourth sample of comparison children who had not experienced any of these forms of ELS. RESULTS: Smaller amygdala volumes were found for children exposed to these different forms of ELS. Smaller hippocampal volumes were also noted for children who were physically abused or from low socioeconomic status households. Smaller amygdala and hippocampal volumes were also associated with greater cumulative stress exposure and behavioral problems. Hippocampal volumes partially mediated the relationship between ELS and greater behavioral problems. CONCLUSIONS: This study suggests ELS may shape the development of brain areas involved with emotion processing and regulation in similar ways. Differences in the amygdala and hippocampus may be a shared diathesis for later negative outcomes related to ELS.

Robust Automated Amygdala Segmentation via Multi-Atlas Diffeomorphic Registration.

Here, we describe a novel method for volumetric segmentation of the amygdala from MRI images collected from 35 human subjects. This approach is adapted from open-source techniques employed previously with the hippocampus (Suh et al., 2011; Wang et al., 2011a,b). Using multi-atlas segmentation and machine learning-based correction, we were able to produce automated amygdala segments with high Dice (Mean = 0.918 for the left amygdala; 0.916 for the right amygdala) and Jaccard coefficients (Mean = 0.850 for the left; 0.846 for the right) compared to rigorously hand-traced volumes. This automated routine also produced amygdala segments with high intra-class correlations (consistency = 0.830, absolute agreement = 0.819 for the left; consistency = 0.786, absolute agreement = 0.783 for the right) and bivariate (r = 0.831 for the left; r = 0.797 for the right) compared to hand-drawn amygdala. Our results are discussed in relation to other cutting-edge segmentation techniques, as well as commonly available approaches to amygdala segmentation (e.g., Freesurfer). We believe this new technique has broad application to research with large sample sizes for which amygdala quantification might be needed.