0560 Pre-sleep Rumination and Resting-State Connectivity with the Bed Nucleus of Stria Terminalis in Individuals with Insomnia Symptoms

Abstract Introduction Projections from the central amygdala to the bed nucleus of the stria terminalis (BNST) are implicated in slower, longer-lasting perseverative anxiety symptoms. The BNST’s role in sustained hyperarousal may delay sleep by promoting rumination prior to sleep onset, as is common in insomnia disorder. We hypothesized that BNST connectivity to regions of the default mode network (DMN) at baseline would be associated with elevated pre-sleep ruminations. Furthermore, we hypothesize that reductions in pre-sleep rumination following active continuous theta-burst stimulation (cTBS) will be associated with decreased, or no, BNST functional connectivity (FC). Methods Participants were 20 individuals (12 women; 19 to 39 years old) with self-reported insomnia (i.e., ISI≥15, PSQI≥6, and/or ESS≥11). To assess pre-sleep rumination, participants filled out the Glasgow Content of Thought Inventory (GCTI) prior to coming to the laboratory for cTBS overnight visits counterbalanced between one sham and the other active 40-second cTBS. The target region was the left inferior parietal lobule, a posterior region the DMN. MRI scans were conducted before and after each stimulation type. Results Seed-to-voxel connectivity with a seed placed in the BNST shows that higher baseline GCTI severity is associated with positive BNST FC to a precuneus cluster with center at +10 -40 + 06 (r =.701, p<.001, T =7.05, p-FDR < 0.001) and negative BNST FC connectivity to a left frontal pole cluster with center at -14 + 58 + 34 (r =-.657, p<.001; T = -6.25, p-FDR = 0.045). Reductions in GCTI following active cTBS, compared to pre-active and sham conditions, is associated with negative FC of the BNST with a cluster including the posterior cingulate gyrus and the precuneus with center at -02 -40 42 (r =-.736, p<.05, T = -5.48, p-FDR < 0.001). Conclusion BNST connectivity with the precuneus was associated with baseline levels of pre-sleep rumination and this effect was modulated only following active cTBS. Specifically, individuals with steeper reductions in pre-sleep rumination severity following cTBS exhibit more negative connectivity between the BNST and this posterior region of the DMN. As hypothesized, BNST connectivity to regions of the DMN predicts pre-sleep rumination and this connectivity reverses following cTBS. Support (if any) Funded by Department of Defense: W81XWH-19-PRMRP-DA