Longitudinal sleep instability predicts cognitive decline and is associated with increased MRI‐visible perivascular space burden

Abstract

AbstractBackgroundAlthough emerging data suggest a direct link between sleep duration, aging, and cognitive performance, the impact of longitudinal sleep patterns have on Alzheimer’s disease‐associated pathology and cognitive decline remains unclear.MethodUtilizing longitudinal data on self‐reported sleep duration collected within the Seattle Longitudinal Study, we used a Cox proportional hazard (CPH) regression to evaluate the relationship between longitudinal sleep factors on the development of cognitive decline. Additional analysis in a sub‐cohort utilized multivariate linear regression prediction analyses to evaluate the predictive value of these longitudinal sleep factors on the incidence of developing cognitive impairment over an interval of 10 years. A cohort within the study received MRI and FLAIR neuroimaging, enabling the assessment of the impact these longitudinal sleep factors have on subcortical white matter MRI‐visible perivascular space burden (PVS). A semi‐automated segmentation algorithm accounted for PVS total burden and volume, as well as average PVS width, length, and median volume across participants, blinded by group.ResultCPH analysis found years of education (HR 1.12, 95% CI 1.02, 1.22), APOE4 status (HR 2.18, 95% CI 1.37, 3.49), higher depression scores (HR 1.08, 95% CI 1.04, 1.13), short sleep phenotype (HR 1.78, 95% CI 1.03, 3.08) and increased sleep variability (HR 3.10, 95% CI 1.66, 5.80) were significantly associated with cognitive impairment. Further, when including these factors into our linear regression prediction analyses, age (p<0.0001), depression (p = 0.01) and sleep variability (p = 0.042) were significant predictors of cognitive impairment over a 10‐year period. Preliminary analysis showed significant association between high sleep variability and increased whole‐brain PVS volume (p = 0.04) and length (p = 0.03), with a significant correlation between sleep variability and median PVS volume (p = 0.04). Follow‐up analysis evaluated PVS features within the wider cohort.ConclusionThese findings suggest that variability in longitudinal self‐reported sleep duration exerts a previously unappreciated influence on downstream cognitive performance and pathological processes, potentially including glymphatic dysfunction. These findings elaborate the link between sleep instability, aging, cognitive decline and neuropathologic disease state. Future studies should evaluate these findings on an independent cohort to assess whether sleep variability of different time scales (weeks, years, decades) exerts similar effects on these outcomes.