Prevalence of obstructive sleep apnea in venous thromboembolism: a systematic review and meta-analysis

Purpose Previous studies revealed that the concomitant prevalence of obstructive sleep apnea (OSA) and venous thromboembolism (VTE) was high, but the results were inconclusive due to various limitations. We aimed to systematically review the literature on the prevalence of OSA in patients with VTE. Methods Relevant studies were identified on multiple electronic databases through July 2018. The DerSimonian-Laird random effects model was used to calculate the pooled prevalence of OSA, moderate-to-severe OSA, and severe OSA in VTE patients, respectively. Sensitivity analysis was performed based on diagnostic methods of OSA and races. Results A total of 11 studies comprising 895 patients were available for the meta-analysis, but one study was excluded because of the between-study heterogeneity in the following analysis. The pooled prevalence of OSA, moderate-to-severe OSA, and severe OSA in VTE patients were 70% (95% CI = 65%, 75%), 41% (95% CI = 29%, 54%), and 19% (95% CI = 15%, 23%), respectively. Sensitivity analysis indicated that the prevalence was similar in different diagnostic methods, but the contributions of races to OSA were complex. Although the lower prevalence of all OSA and moderate-to-severe OSA as compared with Western countries, Asian countries have similar or even a little bit higher prevalence of severe OSA. Conclusions Findings from this meta-analysis supported that the prevalence of OSA in VTE patients was strikingly high. Screening for OSA in patients with VTE is necessary for developing effective treatment strategies.