Background Uncertainties about efficacy and safety of oral anticoagulant therapy (OAT) among older and frail medical patients with atrial fibrillation (AF) largely contribute to under-prescription of these drugs.Aims In this prospective observational cohort study, we investigated mortality, and ischemic and hemorrhagic events, in hospital-discharged older patients with AF.Methods Stroke and bleeding risk were evaluated using CHA2DS2-VASC and HAS-BLED scores. Comorbidity, frailty, cognitive and nutritional status and functional autonomy were evaluated using standardized scales. Independent associations between clinical variables, including OAT use, and all-cause mortality, fatal and non-fatal ischemic and hemorrhagic events, were evaluated. Further clinical outcomes comparison between patients treated with OAT and those untreated was performed after adjustment for significant differences in patient baseline characteristics with propensity score matching.Results Of 452 patients included (mean age 81.6 years, 54.9 % women, roughly 30 % cognitively impaired and/or functionally dependent, mean CHA2DS2-VASC and HAS-BLED scores 4.6 and 2.8, respectively), 151 (33.4 %) died during a mean follow-up period of 300.5 days; ischemic and hemorrhagic stroke occurred in 4.0 and 0.4 % of patients, respectively, and major bleedings in 6.2 %.Discussion After multivariate analysis, OAT at discharge was associated with lower overall mortality and reduced occurrence of ischemic stroke, the first finding being confirmed in propensity score matched analysis.Conclusions Among older vulnerable AF patients with high post discharge death rate, OAT was associated, among other multiple factors, with reduced mortality and lower occurrence of ischemic stroke.

In this retrospective cohort observational study, we investigated mortality, ischemic, and hemorrhagic events in patients ≥65 years with atrial fibrillation consecutively discharged from an Acute Geriatric Ward in the period 2010 to 2013. Stroke and bleeding risk were evaluated using CHA2DS2-VASC (congestive heart failure/left ventricular dysfunction, hypertension, aged ≥75 years, diabetes mellitus, stroke/transient ischemic attack/systemic embolism, vascular disease, aged 65 to 74 years, gender category) and HAS-BLED (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly, drugs/alcohol concomitantly) scores. Co-morbidity, cognitive status, and functional autonomy were evaluated using standardized scales. Independent associations among clinical variables, including use of vitamin K antagonist-based oral anticoagulant therapy (OAT), all-cause mortality, and fatal and nonfatal ischemic and hemorrhagic events, were evaluated. Further clinical outcomes comparison between patients treated with OAT and those untreated was performed after adjustment for significant differences in patient baseline characteristics with propensity score matching. Of 980 patients discharged (mean age 83 years, 60% women, roughly 30% cognitively impaired or functionally dependent, mean CHA2DS2-VASC and HAS-BLED scores 4.8 and 2.1, respectively), 505 (51.5%) died during a mean follow-up period of 571 days; ischemic and hemorrhagic stroke occurred in 82 (12.3%) and 13 patients (1.3%), respectively, and major bleedings in 43 patients (4.4%). Vitamin K antagonists' use was independently associated with reduced mortality (odds ratio 0.524) and with a nonsignificant reduction in incidence of ischemic stroke, without excess in bleeding risk. Similar findings were observed in the 2 propensity score-matched cohorts of patients. In conclusion, among vulnerable patients with atrial fibrillation ≥65 years with high post-discharge death rate, OAT was associated, among other multiple factors, with reduced mortality.