Abstract
BACKGROUND: This study aimed to evaluate the causal impact of common modifiable lifestyles on obstructive sleep apnea (OSA), which is beneficial for recommendations to prevent and manage OSA. METHOD: Published genome-wide association study (GWAS) summary statistics were used to perform two-sample Mendelian randomization (MR). Variants associated with each exposure of smoking, drinking, and leisure sedentary behaviors at the genetic level were used as instrumental variables (IVs). Then, inverse-variance weighting (IVW) was considered the primary result for causality. Moreover, several complimented approaches were also included to verify the observed associations. MR-PRESSO and MR-Egger intercept were applied to test the horizontal pleiotropy. To assess heterogeneity, Cochran's Q test by IVW and MR-Egger were applied. RESULTS: Regular smoking history increased OSA risk in all applied approaches [OR (95% CI)(IVW) = 1.28 (1.12, 1.45), p = 1.853 × 10(-4)], while the causality of lifetime smoking index [OR (95% CI)(IVW) = 1.39 (1.00, 1.91), p = 0.048], alcohol intake frequency [outliers removed OR (95% CI)(IVW) = 1.26 (1.08, 1.45), p = 0.002], and coffee intake behavior [OR (95% CI)(IVW) = 1.66 (1.03, 2.68), p = 0.039] on OSA risk were not always consistent in other approaches. In addition, no robust causal associations were observed for the effect of sedentary leisure behaviors on OSA risk. In sensitivity analysis, we observed no sign of horizontal pleiotropy or heterogeneity. CONCLUSION: Ever regularly smoking has a robust causal role in increasing OSA risk, which should be discouraged as precautions from developing OSA.