Direct Oral Anticoagulants vs. Vitamin-K Antagonists in the Elderly With Atrial Fibrillation: A Systematic Review Comparing Benefits and Harms Between Observational Studies and Randomized Controlled Trials

Background: The publication of high-quality observational studies (OSs) has fueled reassessment of the treatment effects of direct oral anticoagulants (DOACs) in the elderly with atrial fibrillation (AF). Methods: The MEDLINE, EMBASE, and Cochrane Library databases were systematically searched (through July 1, 2019) for eligible OSs and randomized controlled trials (RCTs) that reported effectiveness outcomes [stroke or systemic embolism (SE)] or safety outcomes [intracranial hemorrhage (ICH), major bleeding, gastrointestinal bleeding (GIB), myocardial infarction (MI), and all-cause mortality] for DOACs and vitamin-K antagonists (VKAs) in elderly AF patients. A random-effects model was applied to calculate adjusted hazard ratios (HRs) for OSs and relative risks (RRs) for RCTs. Interaction analyses and the ratio of HR (RHR) were used to assess and compare OSs and RCTs. Results: A total of 32 studies involving 547,419 patients were included. No significant difference in treatment effect estimates was found between 27 OSs and 5 RCTs [P(interaction) > 0.05 for each and all 95% confidence interval (CI) of RHR crossed 1.0]. Compared with VKAs, DOACs significantly reduced risk for stroke/SE (OSs, HR: 0.87, 95% CI: 0.81–0.94; RCT, RR: 0.82, 95% CI: 0.67–0.96), and ICH (OSs: 0.47 [0.37–0.57]; RCTs: 0.47 [0.31–0.63]), without increasing risk for GIB (OSs: 1.21 [0.98–1.43]; RCTs: 1.34 [0.91–1.77]), and all-cause mortality (OSs: 1.01 [0.92–1.11]; RCTs: 0.94 [0.87–1.00]). Among OSs, DOACs significantly decreased risk for major bleeding (0.87 [0.77–0.98]) and MI (0.89 [0.79–0.99]). It was found that dabigatran, but not other DOACs, significantly increased risk for GIB (1.48 [1.23–1.72]). Conclusions: DOACs were demonstrated to be more effective and safer than VKAs in elderly AF patients, whereas dabigatran users had a 48% increase in risk for GIB.