Specific amygdala and hippocampal subfield volumes in social anxiety disorder and their relation to clinical characteristics – an international mega-analysis

Abstract

Abstract Social anxiety disorder (SAD) has been associated with alterations in amygdala and hippocampal volume but there is mixed evidence for the direction of volumetric alterations. Additionally, little is known about the involvement of the distinct subfields in the pathophysiology of SAD. Volumetric data from a large multi-centre sample of 107 adult individuals with SAD and 140 healthy controls (HCs) was segmented using FreeSurfer to produce 9 amygdala and 12 hippocampal subfield volumes. Volumes were compared between groups using linear mixed-effects models adjusted for age, age-squared, sex, site and whole amygdala and hippocampal volumes. Subgroup analyses examined subfield volumes in relation to comorbid anxiety disorder, and comorbid major depressive disorder (MDD), psychotropic medication status, and symptom severity. In the full sample, SAD was associated with smaller amygdala volumes in the basal ( d=- 0.32, p FDR =0.022), accessory basal ( d=- 0.42, p FDR =0.005) and corticoamygdaloid transition area ( d=- 0.37, p FDR =0.014), and larger hippocampal volume in the CA3 ( d= 0.34, p FDR =0.024), CA4 ( d= 0.44, p FDR =0.007), dentate gyrus ( d= 0.35, p FDR =0.022) and molecular layer ( d= 0.28, p FDR = 0.033), compared to HCs. SAD without comorbid anxiety, in addition, demonstrated smaller lateral amygdala ( d=- 0.30, p FDR =0.037) and hippocampal amygdala transition area ( d=- 0.33, p FDR =0.027) relative to HCs. In SAD without comorbid MDD, only the smaller accessory basal amygdala remained significant ( d=- 0.41, p FDR =0.017). No association was found between subfield volume and medication status or symptom severity. In conclusion, we observed distinct patterns of volumetric differences across specific amygdala and hippocampal subfields, regions that are associated with sensory information processing, threat evaluation and fear generalization. These findings suggest a possible disruption in information flow between the amygdala and hippocampal formation for fear processing in SAD.