0290 Cholesterol Levels Predict Sleep Quality, Total Sleep Duration, and REM Sleep Duration Among Black Adults

Abstract Introduction Both cardiovascular disease and sleep disturbances disproportionately affect Blacks. Yet, early detection and prevention remain challenging. Recent work has linked cardiovascular risk to sleep metrics (i.e., sleep duration and quality), but investigations of blood lipid levels in Black populations remain scarce. Furthermore, most published studies rely on laboratory-based polysomnography, although portable PSG and wearable sleep-tracking devices allow for accurate monitoring in real-world environments. Novel wearable sleep devices enable investigators to capture sleep patterns and ascertain their associations with cardiovascular disease. In this study, we investigated whether cardiovascular disease markers are associated with sleep metrics among Black adults. Methods Sample included 204 Black adult participants from two NIH-funded studies (MOSAIC and ESSENTIAL). Sleep metrics (sleep quality index (SQI) and rapid eye movement (REM) and sleep duration) were recorded via a SleepImage Ring for 7 consecutive nights. Fasting lipid profiles (total cholesterol, high-density lipoproteins (HDL), low-density lipoproteins (LDL), triglycerides) were obtained based on plasma obtained through venipuncture and a total cholesterol: HDL ratio was calculated as: total cholesterol/HDL. A linear regression model assessed whether biomarkers of cardiovascular risk were associated with sleep outcomes. Results TC: HDL-c ratios (F [9,195] =5.55, R²=0.20, p< 0.001) and LDL (F [9,194] =5.33, R²=0.20, p< 0.001) values were inversely correlated with SQI. Specifically, lower TC: HDL-c ratios (β=-0.16, p=0.02) and LDL (β=-0.14, p=0.04) values were associated with increased SQI. Additionally, TC:HDL-c ratios (F[9,195]=3.40, R²=0.14, p< 0.001) and LDL values (F[9,194]=2.44, R²=0.10, p=0.01) predicted REM sleep duration; specifically, lower TC:HDL-c ratios (β=0.24, p< 0.001) and LDL values (β=0.15, p=0.04) were linked to reductions in REM sleep. Furthermore, lower TC: HDL-c ratios (F [9,195] =2.37, R²=0.10, p=0.01) predicted longer sleep duration (β=0.26, p< 0.001), whereas LDL values did not (F [9,194] =1.20, R²=0.05, p=0.30). Finally, triglyceride values did not show significant associations with any sleep metrics. Conclusion Reduced levels of plasma-based lipids (LDL, total cholesterol: HDL) significantly predicted increased sleep quality as well as total sleep duration and REM sleep duration. Integrating wearable-derived sleep metrics with cholesterol biomarkers may be useful in enabling clinicians to identify at-risk individuals before disease develops, thus facilitating application of available preventive strategies. Support (if any)