Edoxaban for stroke prevention in routine practice patients with atrial fibrillation with and without atherosclerotic disease: a post-hoc sub-study of ETNA-AF-Europe

Abstract

Abstract Background Patients with atrial fibrillation (AF) and concomitant coronary and peripheral artery disease (CAD and/or PAD) have an increased risk of adverse health events compared to those without such atherosclerotic disease. The annual risks of major adverse cardiovascular outcomes for these patients treated with edoxaban in routine clinical practice remain uncertain. Purpose To determine the clinical characteristics and rates of key adverse events over multiple years of follow-up in patients with and without atherosclerotic disease (defined as CAD and/or PAD). Methods This is a post-hoc sub-study of ETNA-AF-Europe, a multinational, prospective cohort study on unselected patients with AF treated with edoxaban. In total, 13,164 (97%) of the 13,632 patients who provided informed consent were included in this substudy. The included patients were followed for a total of 48 months. Data were collected at baseline and yearly thereafter (±2 months). The present analysis compared baseline clinical characteristics (n [%] or mean [SD]) and clinical outcomes (events per 100 patient-years [%/year] with 95% confidence intervals [95% CI]) by CAD/PAD status at baseline. Results Of the 13,164 patients included in the analysis, 2973 had CAD and/or PAD (n=2762 CAD only). Patients with atherosclerotic disease were older (75.3 years vs 73.2 years), had higher CHA2DS2-VASc scores (3.8 vs 3.0) and HAS-BLED scores (2.8 vs 2.4), and more frequently received the reduced dose of 30 mg edoxaban than those without CAD/PAD (30.2% vs 21.0%) (Table). Cardiovascular comorbidities were in general more prevalent in those with atherosclerotic disease vs those without. Rates of stroke or systemic embolism (0.89%/year, 95%-CI 0.72–1.09 vs 0.59%/year, 95%-CI 0.52–0.68), myocardial infarction (0.73%/year, 95%-CI 0.58–0.92 vs 0.25%/year, 95%-CI 0.20–0.30), and cardiovascular (1.56%/year, 95%-CI 1.33–1.83 vs 0.85%/year, 95%-CI 0.76–0.95) as well as all-cause death (5.96%/year, 95%-CI 5.50–6.46 vs 3.56%/year, 95%-CI 3.37–3.77) were higher for patients with CAD/PAD vs those without such atherosclerotic disease (Figure). Conclusion Annual rates of thromboembolic and bleeding events were low for AF patients treated with edoxaban in routine practice. The higher rates of thrombotic and bleeding events in patients with CAD/PAD are most likely caused by differences in baseline intrinsic risks. Nonetheless, these differences, particularly the threefold higher rate of myocardial infarction, warrants further investigation.Baseline Characteristics