LAUSR

AIMS Sleep disorders are associated with increased risk of cardiometabolic diseases in observational studies, but the causality remains unclear. In this study, we leveraged two-sample Mendelian randomization (MR) analyses to assess the causal associations of self-reported daytime napping, daytime sleepiness, and other sleep phenotypes with cardiometabolic diseases including ischemic stroke (IS), coronary artery disease (CAD), heart failure (HF), and type 2 diabetes mellitus (T2DM). METHODS We selected genetic variants as instrumental variables for self-reported daytime napping, daytime sleepiness, morning person, insomnia, short sleep duration, and long sleep duration from European-descent genome-wide association studies (GWASs). Summary statistics for cardiometabolic diseases originated from four different GWASs with a total of 2,500,086 participants. We used the inverse-variance weighted method to explore the role of self-reported sleep phenotypes on the etiology of cardiometabolic diseases in the main analyses, followed by several sensitivity analyses for robustness validation. RESULTS Genetically predicted self-reported daytime napping (T2DM: OR, 1.56 [95% CI, 1.21-2.02]), insomnia (IS: OR, 1.07 [1.04-1.11]; CAD: OR, 1.13 [1.08-1.17]; HF: OR, 1.10 [1.07-1.14]; T2DM: OR, 1.16 [1.11-1.22]) and short sleep duration (CAD: OR, 1.37 [1.21-1.55]) were causally associated with elevated risk of cardiometabolic diseases. Moreover, genetically determined self-reported daytime sleepiness (CAD: OR, 2.05 [1.18-3.57]; HF: OR, 1.82 [1.15-2.87]) and morning person (HF: 1.06 OR, [1.01-1.11]) had potential detrimental effect on cardiometabolic risks. CONCLUSIONS Self-reported daytime napping, insomnia, and short sleep duration had causal roles in the development of cardiometabolic diseases, while self-reported daytime sleepiness and morning person was the potential risk factor for cardiometabolic diseases.