Comparative Clinical Outcomes of Edoxaban in Adults With Nonvalvular Atrial Fibrillation

Background: A critical appraisal of all pooled evidence regarding novel oral anticoagulants (NOACs) for stroke prevention regardless of publication status or study design has not been conducted yet. Being the latest addition to NOACs, the data on edoxaban are especially scarce. Study Question: What are the comparative clinical outcomes of edoxaban versus warfarin and other NOACs apixaban, dabigatran, or rivaroxaban in adults with nonvalvular atrial fibrillation? Data Sources: Randomized controlled trials (RCTs), observational studies, and network meta-analyses were identified in PubMed, EMBASE, the Cochrane Library, Pharmapendium, Elsevier Clinical Pharmacology, and the clinicaltrials.gov trial registry in June 2018. Study Design: Rapid review per a priori developed protocol, direct frequentist random-effects meta-analysis of aggregate data, grading the quality of evidence per the Grading of Recommendations Assessment, Development and Evaluation working group approach. Results: Direct 4 RCTs (23,021 patients) suggest that edoxaban is noninferior to warfarin in prevention of stroke and systemic embolism [pooled relative risk (RR): 0.65, 95% confidence interval (CI): 0.23–1.81, 2 RCTs] and reduces the risk of cardiovascular mortality (RR: 0.87, 95% CI: 0.78–0.97, 1 RCT), major cardiovascular morbidity (RR: 0.90, 95% CI: 0.82–0.98, 2 RCTs), and major bleeding events (RR: 0.80, 95% CI: 0.71–0.91, 1 RCT) but increases the risk of gastrointestinal bleeding (RR: 1.21, 95% CI: 1.01–1.46, 1 RCT) and anemia (RR: 1.45, 95% CI: 1.05–1.99, 3 RCTs). Edoxaban is superior to warfarin in patients with increased risk of bleeding with warfarin because of variants in CYP2C9 and VKORC1 genes. Indirect evidence does not allow valid conclusions regarding comparative superiority of NOACs. The quality of evidence was downgraded because of reporting bias, small number of events, and indirectness in comparisons. Conclusions: Edoxaban is a welcome addition to the NOAC's armamentarium. However, the comparative data with other novel NOACs are mostly nonexisting, and urgently needed for better individual patient assessment.