LAUSR

Background Little is known about the effects of anticoagulation in patients with atrial fibrillation (AF) and left atrial enlargement (LAE). Methods Data of patients with AF were retrieved from Chang Gung Research Database during 2007–2016. We excluded patients who were not using oral anticoagulants, used anticoagulants for <30 days, used ≥2 agents concomitantly or switched anticoagulants, had left atrial diameter missing from their data, were aged <65, had received valve surgeries, had mitral stenosis, or had a history of cancer. The primary outcomes were ischemic stroke (IS)/systemic embolism (SE), major bleeding, and death from any cause. Results We identified 40,777 patients who received a diagnosis of AF. After the exclusion criteria were applied, 6,445 patients remained, 4,922 with LAE, and they were followed up for 2.4 ±1.9 years. The mean age of the patients was 77.32 ± 0.18 in the NOAC group and 76.58 ± 6.91 in the warfarin group (p < 0.0001); 48.24% of patients in the NOAC group and 46.98% of patients in the warfarin group were men (p > 0.05). The mean CHA2DS2-VASc score was 3.26 ± 1.05 in the NOAC group and 3.07 ± 1.12 in the warfarin group (p < 0.0001). The mean HAS-BLED score was 3.87 ± 3.81 in the NOAC group and 3.86 ± 3.80 in the warfarin group (p > 0.05). Furthermore, the mean LA diameter was 4.75 ± 0.63 cm in the warfarin group and 4.79 ± 0.69 cm in the warfarin group (p > 0.05). Among patients with LAE, NOAC was associated with significantly reduced IS/SE events (CRR = 0.63, 95% CI = 0.52–0.77), no difference in major bleeding (CRR = 0.91, 95% CI = 0.78–1.05), and significantly reduced death from any cause (aHR = 0.65, 95% CI = 0.52–0.80) compared with warfarin. Conclusions In elderly patients with AF and LAE, NOAC was associated with reduced IS/SE and death from any cause compared with warfarin, whereas no difference in major bleeding was observed between these treatments.