The association between poor sleep health and Alzheimer's disease structural neuroimaging biomarkers

Poor sleep may be a risk factor for neurodegeneration and Alzheimer's disease (AD). Few studies have examined objectively measured sleep with structural neuroimaging measures. Vanderbilt Memory and Aging Project participants (N = 407; median age: 70 years) wore ActiGraph accelerometers for 10 days to estimate sleep regularity, timing, efficiency, duration, wake-after-sleep onset, and awakening length. Volume in brain regions of interest (ROIs) and AD signatures were quantified using 3T brain magnetic resonance imaging (MRI). Cross-sectional linear regression models were adjusted for sociodemographic and lifestyle factors, depression, cognitive status, and cardiovascular risk. ROI and McEvoy models were adjusted for total intracranial volume. Greater sleep irregularity (βHippocampus = -0.12, p = 0.005; βMcEvoy = -0.07, p = 0.024) and longer awakening length (βHippocampus = -0.11, p = 0.009; βParietal = -0.08, p = 0.012) were associated with smaller volumes in ROIs related to AD. More irregular and fragmented sleep was associated with smaller volume in ROIs vulnerable to AD, indicating a potential link between poor sleep and neurodegeneration. Associations between sleep and brain health are poorly understood. Gray matter atrophy by irregular sleep may imply Alzheimer's disease (AD) decline. Increased sleep irregularity is associated with smaller brain region volume. Sleep disruption, estimated by awakening length, is linked to AD-related brain volume.