Stroke risk reduction outweighed bleeding risk increase from vitamin K antagonist treatment among nonvalvular atrial fibrillation patients with high stroke risk and low bleeding risk

Abstract

Objective: Warfarin is widely used for stroke prevention in patients with nonvalvular atrial fibrillation (NVAF). We compared the rates of stroke and major bleeding in NVAF patients with a high stroke risk and low bleeding risk profile during warfarin treated (W+) and warfarin untreated (W−) periods.

Method: Insurance claims from six commercial, Medicaid or Medicare databases were analyzed from 2000 to 2014. NVAF patients treated with warfarin, with a CHADS₂/CHA₂DS₂-VASc score ≥2, and an ATRIA score ≤3 at baseline were identified. Incidence rate ratios (IRRs) of stroke and major bleeding were calculated for W + versus W− episodes of person-time, as well as for first 30 days versus beyond 30 days of W + episodes.

Results: Among 316,145 patients, anticoagulant prophylaxis with warfarin significantly reduced stroke risk, with IRRs ranging from 0.48 (95% CI: 0.46–0.51) to 0.80 (95% CI: 0.70–0.91), and increased major bleeding risk, with IRRs ranging from 1.13 (95% CI: 1.10–1.15) to 1.95 (95% CI: 1.10–3.45). Stroke and major bleeding rates were higher during the first 30 days of W + than beyond.

Conclusion: In NVAF patients at high risk for stroke and low risk for bleeding, our data confirm the effectiveness of anticoagulation for stroke prevention. The decrease in stroke risk of anticoagulation may outweigh the risk of major bleeding events, particularly among elderly patients. Potential risks of warfarin during initiation warrant attention, especially among patients who stop and start therapy repeatedly.

Keywords:

• Atrial fibrillation
• bleeding
• stroke
• warfarin

Transparency

Declaration of funding

This research was funded by Janssen Scientific Affairs LLC, Raritan, NJ, United States.

Declaration of financial/other relationships

F.L., G.G., D.P., N.M., and P.L. have disclosed that they are employees of Analysis Group Inc., a consulting company that has received research grants from Janssen Scientific Affairs. W.W.N. and A.A.P. have disclosed that they are employees of Janssen Scientific Affairs.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors would like to acknowledge Michel Cloutier, who was employed by Groupe d’analyse Ltee, Montreal, QC, Canada, at the time the study was conducted and who contributed to the analysis phase with funding from Janssen Scientific Affairs.