Association of Sleep Traits With Venous Thromboembolism: Prospective Cohort and Mendelian Randomization Studies

Abstract

Previous research indicates an association between sleep traits and venous thromboembolism (VTE) risk, though causal relationships remain uncertain. This study evaluated combined and independent associations between sleep traits and VTE risk using UK Biobank data and explored the causal associations between sleep traits and VTE through two-sample Mendelian randomization (MR) analyses. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the associations between the healthy sleep score, as well as individual sleep traits (including sleep duration, insomnia, daytime sleepiness, snoring, and chronotype), and VTE risk were calculated using Cox proportional hazards regression models. Additionally, the two-sample MR analyses used the inverse-variance weighted method to determine odds ratios (ORs) and 95% CIs for causal associations. In the cohort analysis, 314 077 VTE–free participants were followed for a median of 12.3 years, during which 7176 VTE cases occurred. In comparison to those with a sleep score of 0–1, participants with a score of 5 were associated with a 30% lower risk of VTE (HR: 0.70; 95% CI: 0.61–0.80). A U-shaped association was noted between sleep duration and VTE risk. Both short (≤ 6 h) and long (≥ 9 h) sleep durations increased VTE risk. Excessive daytime sleepiness, snoring, and evening chronotype also elevated VTE risk. MR analyses supported a causal relationship for short sleep duration (OR: 1.24; 95% CI: 1.04–1.47) with VTE risk, while other sleep traits showed no causal association. These findings underscore the importance of optimal sleep in reducing VTE risk.