Clinical Characteristics and Outcomes of Very Elderly Patients With Atrial Fibrillation at High Bleeding Risk　― The Fushimi AF Registry ―

<b><i>Background:</i></b> The ELDERCARE-AF trial demonstrated that low-dose edoxaban prevented stroke or systemic embolism (SE) in very elderly Japanese patients with non-valvular atrial fibrillation (NVAF) in whom standard oral anticoagulant therapy was inappropriate because of high bleeding risk. The aim of this study was to elucidate the characteristics and outcomes of such patients in routine clinical practice. <b><i>Methods and Results:</i></b> Data were extracted from the Fushimi AF Registry for ELDERCARE-eligible NVAF patients aged ≥80 years, with a CHADS₂ score ≥2 and ≥1 bleeding risk factors, as shown in the ELDERCARE-AF trial. ELDERCARE-eligible patients (n=549; 12.8% of the entire cohort, 52.9% of those aged ≥80 years and with CHADS₂ score ≥2) were less often male, were older, had more comorbidity and higher risk scores than non-eligible patients from the entire cohort (n=3,734). The crude incidence (% per patient-year) of adverse events was significantly higher in ELDERCARE-eligible than non-eligible patients (stroke/SE, 4.8% vs. 2.0%; major bleeding, 3.6% vs. 1.9%; all-cause mortality, 15.5% vs. 3.9%; cardiovascular death, 2.7% vs. 0.6%; all log-rank P<0.001). Compared with non-eligible patients aged ≥80 years and with a CHADS₂ score ≥2 (n=488), the incidence of stroke/SE, all-cause mortality, and cardiovascular death remained significantly higher in ELDERCARE-eligible patients. <b><i>Conclusions:</i></b> Patients with NVAF who met the inclusion criteria of the ELDERCARE-AF trial were common in routine clinical practice, and had poor clinical outcomes.