Abstract
OBJECTIVES: This study aimed to investigate the efficacy and safety outcomes of patients with atrial fibrillation (AF) compared between those taking warfarin and non-vitamin K antagonist oral anticoagulants (NOACs) based on SAMe-TT(2)R(2) score. METHODS: AF patients using warfarin or NOACs were enrolled from Thailand's COOL-AF registry. A low SAMe-TT(2)R(2) score was defined as a score of 0-2. The efficacy outcomes were all-cause death, ischemic stroke (IS), transient ischemic attack (TIA), and/or systemic embolization (SE). The safety outcome was major bleeding (MB). The secondary outcome was a combination of cardiovascular (CV) death, IS/TIA/SE, or MB. Cox proportional hazards model was used to compare the event rate between the AF patients taking warfarin and NOACs according to SAMe-TT(2)R(2) score. RESULTS: A total of 2568 AF patients taking oral anticoagulants were enrolled. Warfarin and NOACs were used in 2340 (91.1%) and 228 (8.9%) patients, respectively. Among overall patients, 305 patients taking warfarin (13.0%) and 21 patients taking NOACs (9.2%) had the efficacy outcome, while 155 patients taking warfarin (6.6%) and 11 patients taking NOACs (4.8%) had the safety outcome. After adjustment for confounders, overall patients taking warfarin had significantly more secondary outcome than those taking NOACs (11.4% vs. 7.5%, respectively; adjusted hazard ratio: 1.74, 95% confidence interval: 1.01-2.99; p = 0.045) regardless of SAMe-TT(2)R(2) score. CONCLUSIONS: AF patients taking warfarin had a significantly higher CV death or IS/TIA/SE or MB compared to those taking NOACs regardless of SAMe-TT(2)R(2) score. The results of this study do not support the use of SAMe-TT(2)R(2) score to guide OAC selection.