Content area
Abstract
Abstract
Sleep plays a key role in the restoration and preservation of optimal brain function, whereas sleep loss causes fatigue and impairs a range of neurobehavioral functions. Accumulating evidence suggests that the human brain can be characterized as a small-world network. However, the effects of sleep deprivation (SD) and recovery sleep on human brain network organization remain unclear. Here we used resting-state functional magnetic resonance imaging (fMRI) to evaluate changes in brain network topology after sleep deprivation.
We used graph theory to analyze resting-state fMRI data from 51 health adults (34.1 ± 9.0y, 29 males) who participated in a 5-day and 4-night in-laboratory controlled study. Thirty-eight participants completed an experimental protocol including 36-hour acute total SD (TSD) after one night of 9-hour baseline sleep, and followed by two nights of recovery sleep. They were scanned three times on the mornings of day 2 after baseline sleep, day 3 during SD, and day 5 after recovery sleep. Thirteen participants completed a non-SD control protocol and were scanned three times on the equivalence days. Three network metrics, including small-worldness (σ), global efficiency, and local efficiency were calculated for each subject using GRETNA toolbox and compared between baseline, SD, and recovery conditions.
All three brain network metrics were significantly reduced after one night of TSD compared to baseline (all p <0.001). After two nights of recovery sleep, network small-worldness and global efficiency returned to baseline level, whereas local efficiency was not fully restored to baseline level. Brain network small-worldness and global efficiency changes correlated with self-reported fatigue level increases during sleep loss (both p <0.005). No changes in brain network metrics were found among the fMRI scans in the control group.
Our results suggest that one night of TSD significantly impairs topological properties of brain small-world network. Two nights of recovery sleep fully restored global but not local properties of brain network organization.
Supported in part by NIH grants R01-HL102119, R01-MH107571, R01-NR004281, CTRC UL1RR024134, P30-NS045839, and the PENN IOA Pilot Project.