European Primary Care Cardiovascular Society

Real-world safety data show differences between anticoagulants in AF patients

Lip GYH et al., Thromb Haemost 2016

Real-world comparison of major bleeding risk among non-valvular atrial fibrillation patients initiated on apixaban, dabigatran, rivaroxaban, or warfarin
A propensity score matched analysis

Lip GYH, Keshishian A, Kamble S, et al.
Thromb Haemost 2016; 116: published online ahead of print


Vitamin K antagonists (VKAs) reduce stroke and systemic embolism by 64% and all-cause mortality by 26%, compared to placebo in patients with atrial fibrillation (AF) [1]. However, VKAs have many drug and food interactions and require routine INR monitoring, and these limitations result to under-treatment in 30-50% of AF patients [2,3]. Non-VKA oral anticoagulants (NOACs) have similar or superior efficacy to VKAs, they do not require routine INR monitoring and have fewer food and drug interactions [4]. There are limited comparative real-world data available for the NOACs dabigatran, rivaroxaban, apixaban, and edoxaban.
In this study, the risk of major bleeding was evaluated in 45,361 newly-anticoagulated non-valvular AF (NVAF) patients initiating on warfarin (n=15461, 34.08 %), rivaroxaban (17801, 39.24 %), dabigatran (4661, 10.28 %), or apixaban (n=7438, 16.40 %) using propensity score matching (PSM).

Main results

  • Pre-matched data showed that warfarin patients were generally older, had more comorbidities and higher risk profiles compared to NOAC users (all P-values<0.001).
  • The lowest incidence rate of major bleeding was observed among apixaban initiators (2.38 per 100 person-years) in the apixaban-warfarin matched cohort and the highest was observed for warfarin initiators (5.09 per 100 person-years) in the warfarin-rivaroxaban matched cohort.
  • Cox proportional hazard models to compare the risk of major bleeding between warfarin and NOACs yielded a lower risk with apixaban vs. warfarin (HR: 0.53; 95%CI: 0.39–0.71), dabigatran vs. warfarin (HR: 0.69; 95%CI:0.50–0.96), but no difference between rivaroxaban vs. warfarin (HR: 0.98; 95%CI: 0.83–1.17).
  • Comparison of incidence rates of major bleeding (ranging from 2.42 to 4.24) in NOAC-to-NOAC matched cohorts, yielded the following hazards: rivaroxaban vs. apixaban HR: 1.82; 95%CI: 1.36–2.43, dabigatran vs. apixaban HR: 1.41; 95%CI: 0.93–2.14, rivaroxaban vs. dabigatran HR: 1.05; 95%CI: 0.74–1.49.


In a real-world observational study using propensity score matched cohorts, initiation with apixaban or dabigatran was associated with significantly lower risk of major bleeding compared to warfarin initiation among newly anticoagulated NVAF patients. In comparisons between NOACs, rivaroxaban initiation was associated with significantly higher risk of major bleeding compared to apixaban initiation. These data represent the first real-world PSM comparative safety analysis assessing the risk of major bleeding between NOACs.
Find this article online at Thromb Haemost


1. Hart RG, Pearce LA, Aguilar MM. Adjusted-dose warfarin versus aspirin for preventing stroke in patients with atrial fibrillation. Ann Intern Med 2007; 147:590–592.
2. De Caterina R, Husted S, Wallentin L, et al. Vitamin K antagonists in heart disease:Current status and perspectives (Section III). Thromb Haemost 2013; 110:1087–1107.
3. Ogilvie IM, Newton N, Welner SA, et al. Underuse of oral anticoagulants in atrial fibrillation: a systematic review. Am J Med 2010;123: 638–645.
4. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet 2014; 383: 955–962.