Atrial fibrillation (AF) ranks the most prevailing type of cardiac rhythm disorder and AF patients are associated with a significantly increased risk of stroke compared to others. This study is designed to assess the relative efficacy of several clinical events prevention anticoagulants in patients with AF. Conventional pairwise meta-analysis was performed with fixed-effect model initially, then network meta-analysis was performed with random-effects model within results illustrated by cumulative odds ratios (ORs) and corresponding 95% credible interval (CrI). The rank probabilities of each treatment outcomes were summarized by the surface under the cumulative ranking curve (SUCRA). We conducted a systematic review and collected key clinical data from 37 studies with respect to 5 anticoagulant treatments for AF. Patients treated with rivaroxaban and apixaban are associated with a reduced risk of stroke compared to those treated with warfarin (OR 0.72, 95% CrI 0.53 to 0.88; OR 0.68, 95% CrI 0.48 to 0.91). Rivaroxaban (SUCRA = 0.712) appears to be the most preferable one with respect to vascular events, and both apixaban (SUCRA = 0.720) and rivaroxaban (SUCRA = 0.678) are preferable to others with respect to stroke. Dabigatran outperforms others with respect to the outcome of mortality (SUCRA = 0.695), hemorrhage events (SUCRA = 0.747), and myocardial infarction (SUCRA = 0.620). In conclusion, dabigatran has a noticeable and comprehensive advantage compared to others with respect to preventing several complications including hemorrhage events, myocardial infarction, and mortality. In addition, apixaban may be the best choice of preventing stroke, and rivaroxaban is more preferable to others with respect to preventing vascular events.
Copyright © 2016 Elsevier Inc. All rights reserved.