Direct oral anticoagulants in very elderly and high-bleeding-risk patients with atrial fibrillation often excluded from oral anticoagulation therapy: a nationwide population-based cohort study.

Aims: In the Edoxaban Low-Dose for Elder Care Atrial Fibrillation Patients (ELDERCARE-AF) trial, very low-dose edoxaban (15 mg once daily) showed better efficacy and positive net clinical benefit compared with placebo in very elderly, high-bleeding-risk patients with atrial fibrillation (AF). However, there are limited data to generalize these results into daily practice. We aimed to investigate the optimal oral anticoagulant (OAC) strategy for the best net clinical benefit in ELDERCARE-AF-like patients.

Methods and results: We used the Korean nationwide claims database to identify AF patients aged ≥80 years from 2014 to 2017 who had one or more ELDERCARE-AF trial inclusion criteria. The risks of ischaemic stroke, major bleeding, all-cause death, and net clinical outcome were evaluated. Primarily, we compared patients without OAC therapy (non-OAC group) to patients with direct OAC (DOAC) therapy (DOAC group), using a propensity score weighting method. A total of 23 858 patients were included (no OAC: 16 575; warfarin: 2390; DOAC: 4893). Among the DOAC group, 69% used low dose, and 9% used very low dose. Compared with the non-OAC group, the DOAC group showed lower risks of ischaemic stroke (hazard ratio, 95% confidence interval: 0.81, 0.68-0.96) and all-cause death (0.91, 0.86-0.96) and a higher risk of major bleeding (1.44, 1.21-1.70). The DOAC group was associated with a lower risk of net clinical outcome compared with the non-OAC group (0.93, 0.88-0.98).

Conclusion: In very elderly, high-bleeding-risk patients with AF, DOACs that were prescribed in usual clinical practice showed better effectiveness and positive net clinical benefit compared with no OAC treatment.