Decreases in slow wave sleep associate with a higher risk of incident Alzheimer’s disease dementia in a community sample

Abstract

AbstractBackgroundSleep disturbance is common in dementia, although it is unclear whether changes in sleep architecture precede dementia onset. We examined the prospective association of changes in sleep architecture with the risk of incident Alzheimer’s disease (AD) dementia in the community‐based Framingham Heart Study (FHS).MethodsOur sample comprised a subset of 294 FHS Offspring participants who participated in the Sleep Heart Health Study Visit 1 (1995‐1998) and Visit 2 (2001‐2003), were dementia‐free and aged over 60 years at the time of Visit 2. Stages of sleep were quantified using home‐based overnight polysomnography. Annualized change in the sleep metrics were computed by subtracting Visit 1 from Visit 2 measures divided by the time interval (in years) between the two visits. Surveillance for incident AD dementia commenced at visit 2 and continued for up to 17 years (mean follow‐up 11 (SD, 4)). We used a series of proportional hazards models to relate changes in sleep to the risk of incident AD dementia, adjusting for age, sex, APOE ε4 allele status, smoking status, and sleeping medication, antidepressant and anxiolytic use.ResultsParticipant mean age was 69 years (SD, 6, 49% male). Over 17‐years follow‐up, we observed 27 cases of AD dementia. The two sleep assessments were separated by a mean of 5.2 years (SD, 0.4). Levels of wake after sleep onset increased by a mean of 4.8 minutes (SD, 11.7) per year whereas both sleep efficiency and the percentage of slow‐wave sleep declined by 0.4% (SD, 2.3) and 0.6% (SD, 1.5) per year, respectively. Each percentage reduction in slow‐wave sleep associated with a 30% increase in the risk of incident AD dementia (HR, 0.70; 95% CI: 0.54, 0.91, P=0.007). Changes in other stages of sleep (stage 1, 2 and REM), sleep fragmentation (sleep efficiency and wake after sleep onset), and respiratory disturbances (apnea‐hypopnea index) did not associate with dementia risk.ConclusionsBy linking decreases in slow‐wave sleep to the risk of clinical AD up to 17 years later, our results extend observations suggesting that slow‐wave sleep is important for synaptic plasticity underlying memory and for glymphatic clearance of neurotoxins that potentially lead to AD dementia.