Changes in Default Mode Network connectivity following sleep deprivation in patients with depression

668 

Abstract Submission 

Artemis Zavaliangos-Petropulu¹, Noor Al-Sharif¹, Paloma Pfeiffer², Brandon Taraku¹, Ashish Sahib¹, Amber Leaver³, Randall Espinoza⁴, Katherine Narr^1,5

¹Department of Neurology, Geffen School of Medicine at the University of California, Los Angeles, CA, ²Department of Neurology, Geffen School of Medicine at the University of California, Los Angeles,, CA, ³Department of Radiology, Northwestern University, Chicago, IL, ⁴Jane and Terry Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Bio, Los Angeles, CA, ⁵Jane and Terry Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Biobehavioral Sciences, Los Angeles, CA

Artemis Zavaliangos-Petropulu  
Department of Neurology, Geffen School of Medicine at the University of California
Los Angeles, CA

Noor Al-Sharif  
Department of Neurology, Geffen School of Medicine at the University of California
Los Angeles, CA

Paloma Pfeiffer  
Department of Neurology, Geffen School of Medicine at the University of California
Los Angeles,, CA

Brandon Taraku  
Department of Neurology, Geffen School of Medicine at the University of California
Los Angeles, CA

Ashish Sahib  
Department of Neurology, Geffen School of Medicine at the University of California
Los Angeles, CA

Amber Leaver  
Department of Radiology, Northwestern University
Chicago, IL

Randall Espinoza  
Jane and Terry Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Bio
Los Angeles, CA

Katherine Narr  
Department of Neurology, Geffen School of Medicine at the University of California|Jane and Terry Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Biobehavioral Sciences
Los Angeles, CA|Los Angeles, CA

The mechanism driving the strong but transient antidepressant effects of total sleep deprivation (TSD) in patients with major depressive disorder (MDD) remains unclear (Ioannou et al., 2021). There is some evidence that suggests TSD impacts the default mode network (DMN) in both healthy controls (HC)(Lunsford-Avery et al., 2020) and MDD populations(Bosch et al., 2013). The DMN appears central to depression pathophysiology(Brakowski et al., 2017), and is often found to be hyperactive(Kaiser et al., 2015). Here, we examined how 24 hours of TSD modulates DMN connectivity in patients with MDD and HC. We hypothesized that changes in the DMN may contribute to the antidepressant effects of TSD.

Participants included N=60 individuals with MDD (age=32.6土11.6, 45% female) and N=54 HC (age=32.2土12.0, 43% female). All patients and a subset of HC (N=15, age=30土12.5, 40% female) participated in a monitored 24-hour sleep deprivation session at the UCLA Clinical and Translational Research Center. Brain imaging and mood assessments (Hamilton Depression Rating Scale; HDRS(Hamilton 1960)) were collected at baseline and 24 hours later (Figure 1a). T1- and T2-weighted images (voxel size(VS)=0.8mm3) and resting state fMRI (VS=2mm3), acquired on a Siemens 3T Prisma using Human Connectome Project (HCP) sequences, were preprocessed with the HCP minimal preprocessing pipeline(Glasser et al., 2013). Post-processing included ICA + FIX(Salimi-Khorshidi et al., 2014) and MSMALL alignment(Robinson et al., 2018). fMRI data was parcellated using the Schaefer 100 Yeo 17 Network atlas(Schaefer et al., 2018) to generate z-scored correlation matrices. Changes in HDRS and global signal average (GSA) were assessed using paired t-tests. Analyses were restricted to connectivity within-DMN nodes. Mixed effect models tested the effect of 1) time and 2) time*diagnosis, covarying for age, sex, and GSA. Linear regression models tested for 1) associations between change in connectivity and change in HDRS in patients and controls separately and 2) baseline differences between patients and controls, adjusting for age, sex, and GSA. All analyses were FDR corrected.

HDRS significantly improved in MDD (t=10.3, p=2.2e-16), but not in HC (p>0.05)(Figure 1b). GSA significantly changed over time for both MDD (t=5.1, p=3.89e-6) and HC (t=3.3, p=0.005)(Figure 1c). Trending (p<0.05) changes after TSD showed 4 connections increasing and 32 decreasing in MDD, 2 increasing and 8 decreasing in HC, and an interaction between time*diagnosis for 15 connections, though these did not pass FDR correction. However, significant associations were observed between improvements in HDRS and increases in right hemisphere retrosplenial cortex (RSP)-medial PFC (t=4.1, p=0.0002) and right-medial PFC to left-RSP (t=3.9, p=0.0003) connectivity in MDD (Figure 2a-c). In HC, trending (p<0.05) negative correlations between change in connectivity and HDRS were identified between left ventral PFC with the right inferior parietal lobe and right RSP with one positive correlation between left hemisphere RSP to Precuneus (Figure 2d-f). At baseline, MDD patients showed 35 connections (including RSP-medial PFC) with trending (p<0.05) greater connectivity compared to HC (Figure 2g-h).

MDD patients showed significantly improved depressive symptoms following TSD, while mood remained stable in HCs. TSD induced significant changes in DMN resting state functional connectivity in both patients and HC. These findings are in line with previous reports suggesting TSD modulates the DMN(Lunsford-Avery et al. 2020; Bosch et al. 2013). Changes in RSP-medial PFC, which differed by diagnosis at baseline, were exclusively correlated with improvements in mood in patients. Thus, core nodes of the DMN may be relevant to the antidepressant effects of TSD. Further research in larger samples is necessary to confirm these findings.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

fMRI Connectivity and Network Modeling ²

FUNCTIONAL MRI

Sleep

Treatment

Other - Major Depressive Disorder