0533 Sleep Spindle Features in Obstructive Sleep Apnea: Results from the Sleep Heart Health Study

Abstract Introduction Obstructive sleep apnea (OSA) is associated with altered cognitive function and dementia, but the mechanisms underlying these associations remain unclear. One proposed mechanism is that OSA, by chronically fragmenting sleep, decreases the number and activity of sleep spindles, which are important for memory consolidation. Few studies have examined sleep spindle characteristics in people with OSA and with conflicting results. This study aims to examine sleep spindle characteristics in people with and without OSA from a multicenter community-based prospective cohort study, the Sleep Heart Health Study (SHHS). Methods The analysis was conducted on 5804 subjects of which 2830 had no OSA (apnea-hypopnea index (AHI) < 5, 61±12 years, 65% women), and 2974 had OSA (AHI ≥5, 65±11 years, 39% women). Spindle density (n/min) and frequency (Hz) were measured during non-rapid eye movement stage 2 from the left central electroencephalographic derivation. Arousal index (AI, n/hr) was used as a measure of sleep fragmentation and Epworth Sleepiness Scale (ESS) as a measure of subjective daytime sleepiness. Linear regression models adjusted for age, sex, BMI and education level were used to compare spindle features in OSA vs non OSA, and to examine the association between spindle features with AHI, oxygen desaturation degree and duration, AI and ESS in OSA group. Results In fully adjusted models, spindle density was lower in OSA compared to no OSA (-0.067 spindles per minute, p=< 0.001). In OSA group, lower sleep spindle density was associated with higher AHI (β= -0.003, p=< 0.001), higher AI (β= -0.002, p=0.020) and higher ESS score (β= -0.007, p=0.007) indicating higher daytime sleepiness. Spindle density was not associated with oxygen desaturation degree or duration. Spindle frequency was similar between OSA and non OSA. Conclusion These findings indicate that spindle density is reduced in people with OSA compared to those without OSA, and this reduction correlates with OSA severity (AHI) and levels of daytime sleepiness. Reduced spindle density in OSA might contribute to OSA-related cognitive impairment and future work is needed to investigate the relationship between reduced spindle density and cognitive function in OSA. Support (if any) Center for Circadian and Sleep Medicine