Effectiveness and Safety of the Coadministration of Rifampin and Warfarin versus Direct Oral Anticoagulants: A Cohort Study

Abstract

Introduction. Pharmacokinetic studies have shown that rifampin reduces the levels of oral anticoagulants during the initiation of coadministration, raising concerns about an increased thrombotic risk, but there are limited comparative clinical outcomes between rifampin and warfarin compared with direct oral anticoagulants (DOACs). This study aimed to evaluate the effectiveness and safety of concurrent use of rifampin and warfarin versus DOACs, with assessments of outcome-associated factors and oral anticoagulant (OAC) management quality. Methods. A total of 142 patients given rifampin plus warfarin (n = 56) or DOACs (n = 86) for over 7 days were included, and their clinical data and outcomes were compared. Results. The median Charlson Comorbidity Index and HAS-BLED (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile INR, elderly, drugs/alcohol concomitantly) score of the two groups were 2 and 3, respectively. The incidence rate of composite ischemic or thromboembolic events was 2.16 and 1.44 per 10,000 patient-days in the warfarin and DOAC groups, respectively, with an adjusted hazard ratio (HR) of 0.41 (95% confidence interval [CI] 0.02–7.34). The incidence rate of composite major bleeding or clinically relevant nonmajor bleeding events was 1.58 and 1.52 per 10,000 patient-days in the warfarin and DOAC groups, respectively, with an adjusted HR of 1.12 (95% CI 0.32–4.45). The risk of composite bleeding events increased with a higher HAS-BLED score (HR: 1.62, 95% CI: 1.02–2.63). Moreover, 34.3% of warfarin users maintained a percent time in therapeutic range of above 50%. Furthermore, 77.9% of DOAC users received appropriate dosing. Conclusion. No significant differences were observed in terms of the incidence of thrombotic or bleeding events between the two groups during coadministration. In addition, a higher HAS-BLED score was associated with a greater risk of bleeding events regardless of the class of OACs used. Finally, close monitoring of bleeding events should be considered.