0286 Cross-Sectional and Longitudinal Relationships between Rest-Activity Rhythms and Metabolic Biomarkers in Older Men: The Osteoporotic Fractures in Men Sleep Study

Abstract

Introduction: Growing evidence has linked disturbances of the sleep-wake cycle to cardiometabolic conditions. However, most previous studies focused on nighttime sleep and few have examined different aspects of the 24-hour rest-activity rhythms in relation to metabolic health. We examined both the cross-sectional and prospective associations between rest-activity rhythms and metabolic biomarkers in the Outcomes of Sleep Disorders in Older Men (MrOS Sleep) study. Method(s): We used wrist-worn actigraphy to measure rest-activity rhythms over ~5 days at baseline (2003-2005). We applied an extended cosine model to calculate amplitude (peak-to-nadir difference in activity), acrophase (time of peak activity), mesor (minimum+1/2 amplitude), and pseudo-F statistics of goodness-of-fit (a measure of the strength of rhythmicity). Leptin, adiponectin, glucose and insulin were measured from fasting blood at baseline and at a follow-up visit (2007-2009). We used multiple linear regression to examine the associations between baseline rest-activity rhythm parameters and metabolic markers at baseline (N=2,538), and changes in markers between baseline and follow up (N=985). Result(s): Results from the cross-sectional analysis suggest that a lower amplitude, mesor and pseudo-F are associated with higher leptin and insulin (betaQ1 vs q4 (95% CI): amplitude, 1.30 (0.52, 2.08) for leptin (ng/ml), 1.17 (0.19, 2.15) for insulin (uIU/ml); mesor, 1.02 (0.25, 1.78), 1.20 (0.23, 2.16); pseudo-F, 1.39 (0.62, 2.17), 1.21 (0.24, 2.18), all p-for-trend < 0.02). When compared to normal acrophase (12:30-16:30), an earlier acrophase is associated with higher levels of fasting glucose at baseline (4.66 (0.36, 8.97), mg/ dL). In addition, early acrophase is associated with a significant decrease in fasting glucose between follow-up and baseline while late acrophase was associated with a significant increase (multivariable adjusted mean change, early: (-7.23 (-13.03,-1.43); normal,-0.40 (-1.33, 0.53), late, 6.02 (0.26, 11.78), p-for-trend=0.001). Conclusion(s): Multiple parameters of the rest-activity rhythms exhibit cross-sectional relationships with metabolic markers. A later acrophase may be a risk factor for higher increase in fasting glucose over ~6 years of follow up while an early acrophase appears to be protective.