Stroke risk and suboptimal thromboprophylaxis in Chinese patients with atrial fibrillation: Would the novel oral anticoagulants have an impact?

Abstract

BackgroundThe risk of stroke associated with atrial fibrillation (AF) is higher in Far Eastern population than in Western population, and warfarin use suboptimal. There is uncertainty whether the novel oral anticoagulants (NOACs) would have a major impact on stroke prevention in Far Eastern populations with AF. ObjectivesWe investigated current antithrombotic therapy use on stroke and bleeding risk, determinants of warfarin use and performed a modeling analysis of the net clinical benefit of the NOACs (apixaban, dabigatran) in a large cohort of Chinese patients with AF. MethodsWe studied 1034 Chinese patients (27.1% female, median age 75years, interquartile range [IQR]: 63–83) with AF who were followed-up for an average of 1.9years (IQR: 1.43–2.64). Stroke/thromboembolism (TE), and major bleeding associated to antithrombotic treatment were investigated. A modeling analysis was performed for the net clinical benefit balancing major bleeding against stroke/TE for dabigatran 110mg bid, dabigatran 150mg bid and apixaban, using their respective recent clinical trial outcome data. ResultsUsing a Cox proportional hazard model, the Hazard Ratio [HR (95% confidence interval, CI)] for stroke/TE compared to no-antithrombotic therapy (no ATT) was 1.27 (0.65–2.50) on aspirin, 1.40 (0.35–3.57) on clopidogrel, 1.52 (0.72–3.23) on dual antiplatelets and 1.65 (0.76–3.57) on warfarin. The risk for major bleeding was 0.35 (0.14–0.85) on aspirin, 0.74 (0.24–2.29) on clopidogrel, 0.35 (0.11–1.10) on dual antiplatelets, and 0.88 (0.36–2.17) on warfarin. Binary logistic regression analysis showed persistent/permanent AF (Odds Ratio, OR, 2.03 [95%CI 1.05–3.92], p=0.035) was associated with warfarin use, but age ≥75years (0.26 [0.16–0.42], p<0.001), aspirin (0.18 [0.12–0.27], p<0.001) and clopidogrel (0.17 [0.08–0.33], p<0.001) were independent determinants of non-use of warfarin.On modeling net clinical benefit (per 100 person-years [95% CI]), apixaban use compared to antiplatelet agents or no ATT was 3.29 (2.15–4.30) using Singer's method and 2.08 (1.18–3.21) with Connolly's method amongst high-risk patients. The use of dabigatran 110mg bid and 150mg bid compared to antiplatelet agents could reduce an additional 18.1 stroke/TE and 24.3 stroke/TE events, respectively. Compared to warfarin, dabigatran 150mg bid had the best net clinical benefit. ConclusionsThere was suboptimal stroke prevention with no difference between antiplatelet and OAC treated patients, perhaps reflecting an inappropriate Target INR range. On modeling analyses, the use of the NOACs (apixaban and dabigatran) could provide better stroke prevention compared to antiplatelet (or warfarin) use in this Chinese AF population, with a positive net clinical benefit.