Clinical Outcome of Edoxaban vs. Vitamin K Antagonists in Patients With Atrial Fibrillation and Diabetes Mellitus: Results From a Multicenter, Propensity-Matched, Real-World Cohort Study

J Clin Med. 2020 May 27;9(6):E1621. doi: 10.3390/jcm9061621.


Diabetes mellitus (DM) is a chronic metabolic disease which is independently associated with unfavorable clinical outcomes in patients with atrial fibrillation (AF). Few real-world data are available about the clinical performance of non-vitamin K oral anticoagulants (NOACs) among patients with atrial fibrillation and diabetes. The aim of our propensity score-matched cohort study was to compare the safety and effectiveness of Edoxaban versus well-controlled vitamin K antagonists (VKAs) therapy among this population. In this study, we considered patients with AF and diabetes on Edoxaban or VKAs therapy included in the multicenter Atrial Fibrillation Research Database (NCT03760874). The occurrence of major bleedings (MB) and thromboembolic events (a composite of ischemic stroke, transient ischemic attack, systemic embolism) was respectively considered primary safety and effectiveness outcome. We identified 557 AF patients with diabetes who received Edoxaban (n: 230) or VKAs (n: 327) treatment. After propensity score matching analysis, 135 Edoxaban and 135 VKA recipients with similar clinical characteristics were evaluated. The mean follow-up was 27 ± 3 months. The incidence rate of thromboembolic events (TE) was 3.0 per 100 person-years (1.11 in Edoxaban vs. 1.9 in the VKA group, hazard ratio (HR): 0.59; 95% confidence interval (CI), 0.14 to 2.52; p = 0.48). The incidence rate of major bleedings (MB) was 3.7 per 100 person-years (1.2 in Edoxaban vs. 2.7 in the VKA group, HR: 0.43; 95% CI, 0.10 to 1.40; p = 0.14). The incidence rate of intracranial hemorrhage was 0.35 per 100 person-years in Edoxaban vs. 0.74 in the VKA group (HR: 0.49; 95% CI: 0.05 to 5.54; p = 0.56). A positive net clinical benefit (NCB) of Edoxaban over VKAs was found (+1.39). Insulin therapy (HR: 1.76, p = 0.004) and glycated hemoglobin (HR: 1.17, p = 0.002) were found to be independent predictors of TE; moreover, the concomitant use of antiplatelet drugs (HR: 2.41, p = 0.001) was an independent predictor of MB. Conclusions: Our data support the hypothesis of the safety and efficacy of Edoxaban for use in patients with AF and diabetes, justified by a favorable NCB over VKAs.

Keywords: atrial fibrillation; bleeding risk; diabetes mellitus; direct oral anticoagulants; edoxaban; stroke prevention.