Variations in rest‐activity cycle length and peak timing during Alzheimer’s progression

Abstract

AbstractBackgroundRest‐activity rhythm is commonly used to assess the circadian regulation and its changes with aging and Alzheimer’s disease (AD). Traditional rhythmicity analyses such as Cosinor analysis assume a rhythm of a constant waveform while ignoring the variations in the length and peak timing between different cycles/days. We introduced a novel rhythmicity analysis based on the uniform phase empirical mode decomposition (UPMEMD) that can provide detailed information of each individual cycle in rest‐activity rhythm. We tested whether variations in rest‐activity cycles change during aging and at different stages of AD.MethodParticipants were 1,065 older adults (baseline age = 80.9±7.3[SD], 807 females) from the Rush Memory and Aging Project. Actigraphy (∼10‐day) was performed annually for up to 13 years. For each recording, the UPMEMD was used to extract the rhythm of ∼24‐h; the length and peak timing were obtained from each cycle; and standard deviations (SD) of cycle lengths and peak timings were computed for each participant. Linear mixed effects models were performed to quantify the within‐subject changes over time in SD of cycle lengths, and separately, SD of peak timings, with adjustments of age, sex, and education years. Change points at diagnoses of mild cognitive impairment (MCI) and Alzheimer’s dementia were considered in the mixed models to examine whether the rates of changes were altered after the diagnoses.ResultSD of cycle length increased by 0.04 hours/year when participants were cognitively intact (95%CI: 0.02‐0.05, p<0.001); the rate increased by an additional 0.04 hours/year after the onset of MCI (95%CI: 0.02‐0.06, p<0.001) and increased further by 0.18 hours/year (95%CI: 0.12‐0.24, p<0.001) after the onset of AD. Moreover, SD of peak timing increased by 0.04 hours/year when the participants were cognitively intact (95%CI: 0.02‐0.05, p<0.001); the rate was accelerated by 0.06 hours/year after the onset of MCI (95%CI: 0.03‐0.08, p<0.001) and further accelerated by 0.26 hours/year (95%CI: 0.20‐0.32, p<0.001) after the onset of AD.ConclusionInterdaily variability in rest‐activity rhythm increased with aging, and the ageing‐related circadian disturbance was accelerated as AD progressed. The markers of rest‐activity rhythm based on proposed rhythmicity analysis may facilitate identification of AD and monitoring of its progression.