LOCKDOWN-ASSOCIATED CHANGES IN THE DEFAULT MODE NETWORK IN DEPRESSION AND HEALTHY INDIVIDUALS DURING THE COVID-19 PANDEMIC IN AUSTRIA

Major threatening life events can impact the functionality of brain networks as revealed by neuroimaging studies [1]. The COVID-19 pandemic was an unprecedented global event that was associated with extensive restrictions to daily life. These measures greatly challenged the whole population including specific subgroups such as patients with depression. Here, we investigated the impact of pandemic-associated lockdowns on functional connectivity (FC) of major functional brain networks.

We hypothesized that there would be an effect of restriction measures on the default-mode network (DMN) and associated brain regions in patients with recurrent depression (rMDD) and healthy individuals (HI).

For each participant up to three MRI measurements were performed between September 2020 and June 2021, during different degrees of lockdown measures (before, during, and after). All subjects received a structural brain scan and three 7-minute single-shot gradient-recalled EPI measurements on a 3T Siemens Magnetom Prisma Scanner (with TE/TR = 30/2050ms, series length: 201 frames, matrix size 100x100x35)). An adapted version of the HCP minimal processing pipeline [2], along with ICA-AROMA and global signal regression for the removal of spatially diffuse noise was applied; denoised fMRI time- series were mapped to midthickness surfaces (CIFTI format) generated with FreeSurfer, and spatially smoothed with geodesic Gaussian kernels (σ = 1.75 mm). Subgenual anterior cingulate cortex (SgACC) FC was derived using a previously validated weight-map method that uses priors of sgACC FC from 1200 HCP subjects, to impute reliable sgACC FC estimates via a weighted average of sgACC correlated time- courses [3, 4]. Network-wise sgACC FC values were derived using the Glasser parcellation (360 bilateral parcels) [5] with a prior established network assignment based on a large data set of healthy subjects and a clustering algorithm that was calibrated upon its performance in well-delineated sensory-motor networks. The same parcellation scheme was used to derive within-network FC for all 12 functional networks [6]. Group differences and changes over time were assessed with linear mixed models (LMM).

We analysed resting-state MRI data from 19 patients with rMDD (13 females, mean age 38 years (SD± 12.5) and 26 HIs (10 females, mean age 27.9 years (SD±5.3). LMMs revealed significant changes over time in the DMN (F=4.3, p>0.017), but no interaction effect of time*group (F1.99, p>0.1). Post-hoc tests revealed a significant decrease in FC in the DMN only between time point 2 and 3 (p=0.014, corrected for 3 post-hoc tests). SgACC weight-map FC analyses revealed widespread changes of the sgACC with other DMN nodes (i.e., 10 DMN parcels).

Analyses revealed changes in functional connectivity within the DMN, a brain network heavily implicated in MDD and associated with introspection and rumination [7]. The connectivity changes were most pronounced when comparing timepoints with hard and light lockdown measures. Further, we were able to confirm that the sgACC, a key DMN brain-region causally implicated in depression and stress-regulation [8] changed connectivity over time towards other DMN nodes. However, the data at hand represent preliminary findings. Future analyses should investigate how these changes might relate to psychological outcome.

1. Bolsinger, J., et al., Neuroimaging Correlates of Resilience to Traumatic Events-A Comprehensive Review. Front Psychiatry, 2018. 9: p. 693.

2.Glasser, M.F., et al., The minimal preprocessing pipelines for the Human Connectome Project. Neuroimage, 2013. 80: p. 105-24.

3. Elbau, I.G., et al., Functional Connectivity Mapping for rTMS Target Selection in Depression. Am J Psychiatry, 2023. 180(3): p. 230-240.

4. Cash, R.F.H., et al., Personalized connectivity-guided DLPFC-TMS for depression: Advancing computational feasibility, precision and reproducibility. Hum Brain Mapp, 2021. 42(13): p. 4155-4172.

5. Glasser, M.F., et al., A multi-modal parcellation of human cerebral cortex. Nature, 2016. 536(7615): p. 171-178.

6.Ji, J.L., et al., Mapping the human brain's cortical-subcortical functional network organization. Neuroimage, 2019. 185: p. 35-57.

7.Sheline, Y.I., et al., The default mode network and self-referential processes in depression. Proceedings of the National Academy of Sciences, 2009. 106(6): p. 1942-1947.

8.Mayberg, H.S., et al., Deep brain stimulation for treatment-resistant depression. Neuron, 2005. 45(5): p. 651-60.