Click to open pdf in new window
Dr. Matthias Grieder PD¹; Prof. Dr. Leila Soravia PhD¹; Prof. Dr. Maria Stein PhD²; Dr. Marcel Meyer PhD³
¹ University Hospital of Psychiatry and Psychotherapy, University of Bern; ² University of Bern; ³University Hospital of Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland
Objective:
In alcohol use disorder (AUD), compromised amygdalar function has been linked to excessive stress and anxiety, both of which are contributing factors to poor treatment outcomes. Despite its proximity to the amygdala and its crucial role in the hormonal regulation of anxiety and stress, the bed nucleus of the stria terminalis (BNST) has yet to be probed on subjective stress, which also applies to its connection to the amygdala.
This study, therefore, investigated the relationship between subjective stress and BNST-amygdala functional connectivity in AUD and healthy controls (HC) using resting-state fMRI.
Methods:
Fifty adults (N[AUD] = 23; N[HC] = 27) underwent eight minutes of resting-state fMRI in a 7T scanner. Regions of interest (ROI) were the bilateral BNST and the left and right amygdala. A ROI-to-ROI functional connectivity analysis yielded functional connectivity strength by Fisher-transformed bivariate correlation coefficients from a general linear model.
These coefficients were estimated separately for each pair of ROIs, characterising the association between their fMRI signal timeseries. Group-level analyses were performed using multivariate parametric statistics with random-effects across subjects and sample
covariance estimation (Emo-Check subscale stress).
Results were adjusted, using a combination of a corrected p-FDR < 0.05 connection-level threshold and a familywise corrected p-FDR < 0.05 cluster-level threshold.
Results:
Elevated subjective stress was associated with increased functional connectivity between the left and right amygdala in AUD compared to HC. Conversely, heightened subjective stress was linked to decreased functional connectivity between amygdalae and the BNST in AUD compared to HC (all connections p < 0.01).
Conclusion:
These findings suggest that under stress, the amygdalae may become hypersynchronised between the hemispheres in AUD, possibly reflecting heightened emotional reactivity or dysregulated anxiety processing. Notably, higher stress was linked to weaker connectivity between the amygdala and BNST in AUD, compared to HC. This may imply that the regulatory influence of the BNST on the amygdala is reduced under stress in AUD, possibly contributing to emotional dysregulation and impaired stress adaptation.