Hippocampal and amygdala subfield volumes in obsessive-compulsive disorder by medication status

Abstract

Background: Although it has been suggested that the hippocampus and amygdala (HA) are involved in the neurobiology of obsessive-compulsive disorder (OCD), volumetric findings have been inconsistent, and little work has been undertaken on the volumetry of the heterogeneous anatomic units of HA, with their specific functions and cytoarchitecture, in OCD. We sought to explore potential sources of heterogeneity in brain volumes by performing a separate analysis for people with and without psychotropic medication use, as well as the association of subfield volumes with OCD symptom severity.

Methods: We segmented T ₁-weighted images from people with OCD and healthy controls in the OCD Brain Imaging Consortium to produce 12 hippocampal subfields and 9 amygdala subfields using Free-Surfer 6.0. We assessed between-group differences in subfield volume using a mixed-effects model adjusted for age and quadratic effects of age, sex, site, and whole HA volume. We also performed subgroup analyses to examine subfield volume in relation to comorbid anxiety and depression, medication status, and symptom severity. We corrected all analyses for multiple comparisons using the false discovery rate (FDR).

Results: We included images from 381 people with OCD and 338 healthy controls. These groups did not significantly differ in HA subfield volumes. However, medicated people with OCD had significantly smaller volumes in the hippocampal dentate gyrus (p _FDR = 0.04, d = -0.26) and molecular layer (p _FDR = 0.04, d = -0.29), and larger volumes in the lateral (p _FDR = 0.049, d = 0.23) and basal (p _FDR = 0.049, d = 0.25) amygdala subfields, than healthy controls. Unmedicated people with OCD had significantly smaller volumes in the hippocampal cornu ammonis sector 1 (p _FDR = 0.02, d = -0.28) than controls. We did not detect associations between any subfield volume and OCD severity.

Limitations: We used cross-sectional data, which limits the interpretation of our analysis.

Conclusion: Differences in HA subfields between people with OCD and healthy controls are dependent on medication status, in line with previous work on other brain volumetric alterations in OCD. This emphasizes the importance of considering psychotropic medication in neuroimaging studies of OCD.

Figure 1

(A) Lateral and (B) medial visualization of amygdala–hippocampal Freesurfer subfield segmentation from right hemisphere of a single representative healthy control, using 3DSlicer (https://www.slicer.org/). The hippocampal fimbria was not included in our analysis., CA = cornu ammonis; GC ML DG = granule cells in the molecular layer of the dentate gyrus; HATA = hippocampus–amygdala transition area.

Figure 2

(A) Mixed effect size estimates (d) for hippocampal and amygdala subfield volumes between participants with obsessive–compulsive disorder (OCD; n = 381) and healthy controls (n = 338) and (B) between medicated participants with OCD (n = 161) and healthy controls (n = 291). Data presented with standard errors (SEs). *p < 0.05. AAA = anterior amygdaloid area; CA = cornu ammonis; CAT = corticoamygdaloid transition area; DG = granule cell layer of dentate gyrus; HATA = hippocampus–amygdala transition area; ML = molecular layer. Supporting data are presented in Table 4 and Appendix 1, Tables S1.1a and S1.7a.