The association of sleep and cortical thickness in mild cognitive impairment.

Abstract

We investigated whether device-measured sleep parameters are associated with cortical thickness in older adults with probable mild cognitive impairment (MCI). We performed a cross-sectional, exploratory analysis of sleep and structural MRI data. Sleep data were collected with MotionWatch8© actigraphy over 7days. We computed average and variability for sleep duration, sleep efficiency, and fragmentation index. T1-weighted MRI scans were used to measure cortical thickness in FreeSurfer. We employed surface-based analysis to determine the association between sleep measures and cortical thickness, adjusting for age, sex, Montreal Cognitive Assessment (MoCA) score, and sleep medication use. Our sample included 113 participants (age=73.1 [5.7], female=72 [63.7%]). Higher fragmentation index variability predicted lower cortical thickness in the left superior frontal gyrus (cluster size=970.9mm2, cluster-wise p=0.017, cortical thickness range=2.1mm2 to 3.0mm2), adjusting for age, sex, MoCA, and sleep medication. Our results suggest that higher variability in sleep fragmentation, an indicator of irregular sleep pattern, is linked to lower cortical thickness. Future longitudinal studies are needed to determine the directionality of these associations.