Altered amygdala subregion-related circuits in treatment-naïve post-traumatic stress disorder comorbid with major depressive disorder

Abstract

Individuals with both post-traumatic stress disorder and major depressive disorder (PTSD+MDD) often show greater social and occupational impairment and poorer treatment response than individuals with PTSD alone. Increasing evidence reveals that the amygdala, a brain region implicated in the pathophysiology of both of these conditions, is a complex of structurally and functionally heterogeneous nuclei. Quantifying the functional connectivity of two key amygdala subregions, the basolateral (BLA) and centromedial (CMA), in PTSD+MDD and PTSD-alone could advance our understanding of the neurocircuitry of these conditions. 18 patients with PTSD+MDD, 28 with PTSD-alone, and 50 trauma exposed healthy controls (TEHC), all from a cohort who survived the same large earthquake in China, underwent resting-state functional magnetic resonance imaging. Bilateral BLA and CMA functional connectivity (FC) maps were created using a seed-based approach for each participant. The analysis of covariance of FC was used to determine between-group differences. A significant interaction between amygdala subregion and diagnostic group suggested that differences in connectivity patterns between the two seeds were mediated by diagnosis. Post-hoc analyses revealed that PTSD+MDD patients showed weaker connectivity between right BLA and (a) left anterior cingulate cortex/supplementary motor area, and (b) bilateral putamen/pallidum, compared with PTSD-alone patients. Higher CMA connectivities left ACC/SMA were also observed in PTSD+MDD compared with PTSD-alone. An inverse relationship between the connectivity of right BLA with right putamen/pallidum and MDD symptoms was found in PTSD+MDD. These findings indicate a relationship between the neural pathophysiology of PTSD+MDD compared with PTSD-alone and TEHC and may inform future clinical interventions.

Keywords: Amygdala; Fear processing; Major depressive disorder; Post-traumatic stress disorder; Resting-state functional connectivity.

Fig. 1

Brain regions of significance from the group × subregion interaction for the BLA and CMA seed regions in the right hemisphere (FWE cluster-corrected threshold, cluster-level p < 0.05 when using a primary cluster determining threshold of p < 0.001) on resting-state functional connectivity. Three gray matter clusters were identified: the right putamen/pallidum, the left ACC/SMA and the left putamen/pallidum. ACC, anterior cingulate cortex; SMA, supplementary motor area; BLA, basolateral amygdala; CMA, centromedial amygdala.

Fig. 2

Box and Whisker plot showing group differences of the mean zFC values and standard deviations extracted from the BLA and CMA pathways. * p < 0.05 (FDR corrected); zFC, Fisher z transformed functional connectivity; BLA, basolateral amygdala; CMA, centromedial amygdala; ACC, anterior cingulate cortex; SMA, supplementary motor area.

Fig. 3

Whole-brain voxel-wise resting-state function connectivity profiles with left basolateral amygdala (BLA) and centromedial amygdala (CMA) seeds and right and BLA and CMA seeds are displayed in the PTSD + MDD group, the PTSD-alone group and TEHC group, separately (FWE corrected, p < 0.05).

Fig. 4

Correlations between HAMD scores and zFC values of BLA- right putamen/ pallidum connectivity (r = −0.570, p = 0.014, uncorrected). HAMD, Hamilton Depression Rating Scale; zFC, Fisher z transformed functional connectivity; BLA, basolateral amygdala.