• Guowei Li, Lip Gregory Y H GYH University of Birmingham Institute of Cardiovascular Sciences, City Hospital, Birmingham, UK. , Anne Holbrook, Yaping Chang, Torben B Larsen, Xin Sun, Jie Tang, Lawrence Mbuagbaw, Daniel M Witt, Mark Crowther, Lehana Thabane, and Levine Mitchell A H MAH Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton, Canada. .
• Guangdong Second Provincial General Hospital, 466 Newport Middle Road, Haizhu District, Guangzhou, 510317, Guangdong Province, China. lig.mileo@yahoo.com.
• Eur. J. Epidemiol. 2019 Feb 1; 34 (2): 173-190.

AbstractThe non-vitamin K antagonist oral anticoagulants (NOACs) have been increasingly prescribed in clinical practice for stroke prevention in patients with nonvalvular atrial fibrillation (AF). Direct comparisons between NOACs in trials are lacking, leaving an important clinical decision-making gap. We aimed to perform a systematic review and meta-analysis to summarize the evidence of observational studies for direct comparative effectiveness and safety amongst NOACs in patients with AF. Conference proceedings and electronic databases including MEDLINE, CINAHL, EMBASE and PUBMED were systematically searched. We included observational studies directly comparing individual NOACs in patients with nonvalvular AF who were aged ≥ 18 years for stroke prevention. Primary outcome included effectiveness outcome (stroke or systemic embolism) and safety outcome (major bleeding). Data were extracted in duplicated by two reviewers independently. A random-effects meta-analysis was conducted to synthesize the data from included observational studies. We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) to rate the overall quality of evidence for each outcome. Fifteen studies were included for qualitative synthesis, twelve studies for meta-analyses. It was found that rivaroxaban and dabigatran were similar with regard to risk of stroke or systemic embolism (Hazard ratio [HR] = 1.00, 95% CI 0.91-1.10; evidence quality: low), but rivaroxaban was associated with higher risk of major bleeding (HR = 1.39, 95% CI 1.28-1.50; evidence quality: moderate). Compared with apixaban, a significantly higher risk of major bleeding was observed with rivaroxaban (HR = 1.71, 95% CI 1.51-1.94; evidence quality: low). Apixaban was associated with lower risk of major bleeding, in comparison with dabigatran (HR = 0.80, 95% CI 0.68-0.95; evidence quality: low). No differences in risk of stroke or systemic embolism was observed between rivaroxaban versus apixaban, and apixaban versus dabigatran. In this study, apixaban was found to have the most favorable safety profile amongst the three NOACs. No significant difference was observed in risk of stroke or systemic embolism between the NOACs. Such findings may provide some decision-making support for physicians regarding their choices amongst NOACs in patients with AF.Registration PROSPERO (identifier: CRD42016052908).

45 keywords...

hide…