Study Objectives:Sleep deficiency is associated with Alzheimer’s disease (AD) pathogenesis. We examined the association of sleep architecture with anatomical features observed in AD: (1) atrophy of hippocampus, entorhinal, inferior parietal, parahippocampal, precuneus, and cuneus regions (“AD-vulnerable regions”) and (2) cerebral microbleeds.
Methods:In 270 participants of the Atherosclerosis Risk in the Communities Study, we examined the association of baseline sleep architecture with anatomical features identified on brain MRI 13∼17 years later. Sleep architecture was quantified as the proportion of slow wave sleep (SWS), proportion of rapid eye movement (REM) sleep, and arousals index using polysomnography. Outcomes included (1) volumetric measurements of each AD-vulnerable region and (2) the presence of any cerebral microbleeds (CMBs) and that of lobar CMBs, which are more specifically associated with AD. We analyzed the association of each sleep predictor with each MRI outcome, adjusting for covariates.
Results:Median age was 61, 53% female, 100% were White, and 47% had 16+ years of education. Mmedian times in SWS and REM were 17.4% and 21.5%, respectively. Having less SWS was associated with smaller volumes of the inferior parietal region (β=-44.18 mm3 per -1 percentage point [PP] of SWS, [95%CI=-76.62, -11.74]) and cuneus (β=-11.98 [=-20.92, -3.04] mm3 per -1 PP of SWS). Having less REM was associated with smaller volumes of the inferior parietal region (β=-75.54 [-129.36, -21.72] mm3 per 1 PP of REM) and precuneus (β=-31.92 [-63.78,-0.06] mm3 per 1 PP of REM). After FDR adjustments, lower SWS and REM were associated with significantly smaller inferior parietal region volumes. Arousal index was not associated with the volumes of AD-vulnerable regions. None of the sleep architecture variables were associated with CMBs or lobar CMBs.
Conclusions:Sleep deficiency is associated with the atrophy of the inferior parietal region, which is observed in early AD. Sleep architecture may be a modifiable risk factor for AD.