Abstract

OBJECTIVE: There is a gap in the knowledge concerning oral anticoagulation (OAC) in atrial fibrillation (AF) patients with a non-high risk of stroke. CHA2DS2VASc and CHADS2 scores generated imprecise risk estimates for low risk patients. We aimed to assess OAC in patients with low risk by CHADS2 and reclassified as high-risk by CHA2DS2VASc. PATIENTS AND METHODS: In this study, retrospective nationwide population-based study, data were obtained from the Turkish claims and utilization management system. Patients with non-valvular AF (n=451,113) between 2007 and 2012 sub-divided into those with a CHA2DS2VASc≥1 and CHADS2=0 (n=41,273) who were off-warfarin (n=29,448) and on-warfarin (n=11,825). Stroke and systemic embolism, major bleeding, all-cause mortality, net clinical benefit (NCB) and ultimate NCB (UNCB) were assessed. RESULTS: Of the total cohort (mean age 66.1 ± 14.1 years, 56.1% female), CHA2DS2VASc improved the net reclassification index of observed 5-year composite thromboembolic endpoint by 6.9% (p<0.05). CHA2DS2VASc reclassified 9.7% low risk patients as high risk. Among reclassified-high-risk category (patients with a CHA2DS2VASc score of ≥1 and CHADS2 score of 0), major bleeding for that prescribed warfarin was 3% and higher than the rate of thromboembolism among those off-warfarin. NCB (-0.035) and UNCB (-0.021) were negative. Death and hospitalization at 1 year were significantly higher for on-warfarin group. CONCLUSIONS: Clinical outcomes, net clinical benefit indices are negative; rates of death and hospitalization were significantly higher for OAC in reclassified category. This emphasizes the importance of greater attention to balancing the risks and benefits of OAC in patients with low risk by CHADS2 and reclassified as high-risk by CHA2DS2VASc.