0321 Pilot Study Detecting Patterns in REM Sleep in Healthy Adults for Later Comparison with Alzheimer’s Patients

Abstract

Introduction: Patients with Alzheimer's disease (AD) spend less time in REM sleep compared to healthy controls, while the number of REM sleep episodes does not appear to differ between AD and normal. With the advent of fitness trackers capable of detecting sleep stages, sleep architecture can now be tracked daily, passively, and continuously for long durations. This pilot is the first step in a larger scale human subjects study aimed at identifying relationships between daily behaviors and sleep patterns indicative of cognitive changes, proteomic signatures, and the regenerative potential of neural networks derived from induced pluripotent stem cells donated from patients with AD and healthy controls. Method(s): We analyzed 13 healthy volunteers' sleep and daily activity (5 male, 9 female) from October 2017 to the present, using a wearable fitness device. Volunteers were between the ages of 18 and 23, with 19 to 421 nights of sleep per subject recorded and an overall 2641 nights of data collected. Result(s): Average time in REM across all participants was 95.6 (+/- 15.4) min/night with an average REM period length of 22.2 (+/- 4.1) min. Subjects' REM comprised on average 22.3% of total sleep time with a mean of 65.1 (+/- 9.0) min between REM periods and 4.6 (+/- 0.6) average periods of REM per night. Mean time to REM onset was 106.9 (+/- 29.1) min. Conclusion(s): While wearable activity trackers are proven to be accurate detectors of sleep length and movement, further study on the accuracy of sleep stage detection will be necessary before long-term sleep stage data can be used as predictors of disease. The next stage of the project will compare activity tracker sleep stage data to polysomnography, the gold standard of objective sleep measurement. Once the parameters for accuracy of the current technology is established, this sleep stage data will be used in predictive algorithms to detect patterns of sleep architecture, specifically REM length and latency, that indicate early stages of AD.