Heterogeneity of outcomes within diabetic patients with atrial fibrillation on edoxaban: a sub-analysis from the ETNA-AF Europe registry.

Background: Recent data have suggested that insulin-requiring diabetes mostly contributes to the overall increase of thromboembolic risk in patients with atrial fibrillation (AF) on warfarin. We evaluated the prognostic role of a different diabetes status on clinical outcome in a large cohort of AF patients treated with edoxaban.

Methods: We accessed individual patients' data from the prospective, multicenter, ETNA-AF Europe Registry. We compared the rates of ischemic stroke/transient ischemic attack (TIA)/systemic embolism, myocardial infarction (MI), major bleeding and all-cause death at 2 years according to diabetes status.

Results: Out of an overall population of 13,133 patients, 2885 had diabetes (22.0%), 605 of whom (21.0%) were on insulin. The yearly incidence of ischemic stroke/TIA/systemic embolism was 0.86% in patients without diabetes, 0.87% in diabetic patients not receiving insulin (p = 0.92 vs no diabetes) and 1.81% in those on insulin (p = 0.002 vs no diabetes; p = 0.014 vs diabetes not on insulin). The annual rates of MI and major bleeding were 0.40%, 0.43%, 1.04% and 0.90%, 1.10% and 1.71%, respectively. All-cause yearly mortality was 3.36%, 5.02% and 8.91%. At multivariate analysis, diabetes on insulin was associated with a higher rate of ischemic stroke/TIA/systemic embolism [adjusted HR 2.20, 95% CI 1.37-3.54, p = 0.0011 vs no diabetes + diabetes not on insulin] and all-cause death [aHR 2.13 (95% CI 1.68-2.68, p < 0.0001 vs no diabetes]. Diabetic patients not on insulin had a higher mortality [aHR 1.32 (1.11-1.57), p = 0.0015], but similar incidence of stroke/TIA/systemic embolism, MI and major bleeding, vs those without diabetes.

Conclusions: In a real-world cohort of AF patients on edoxaban, diabetes requiring insulin therapy, rather than the presence of diabetes per se, appears to be an independent factor affecting the occurrence of thromboembolic events during follow-up. Regardless of the diabetes type, diabetic patients had a lower survival compared with those without diabetes.