A recent study found that older adults with human immunodeficiency virus infection and atrial fibrillation who received warfarin or rivaroxaban had a significantly higher risk of major bleeding compared with those receiving apixaban. The findings suggest that apixaban may present a safer anticoagulation option in this high-risk patient population.
Published in JAMA Internal Medicine, researchers analyzed Medicare claims data from 2013 to 2020, evaluating three separate cohorts: 2,683 patients for warfarin vs apixaban, 2,176 for rivaroxaban vs apixaban, and 1,787 for rivaroxaban vs warfarin. All were individuals aged 50 or older with HIV and atrial fibrillation who initiated oral anticoagulants. Using a new-user, active-comparator design with propensity score overlap weighting, the study simulated a randomized trial.
Results showed warfarin users had a 2.6-fold increased major bleeding risk versus apixaban users (HR, 2.60; 95% CI, 1.51-4.49), rising to nearly sevenfold among those on antiretroviral therapy (ART) (HR, 6.68; 95% CI, 2.78-16.02). Similarly, rivaroxaban was linked to a higher bleeding risk than apixaban (HR, 2.15; 95% CI, 1.18-3.94), with risk nearly fivefold in ART users (HR, 4.83; 95% CI, 2.11-11.08). No significant difference was observed between rivaroxaban and warfarin users.
"In a national U.S. cohort of patients aged 50 years or older with HIV and AF, we found that apixaban may be associated with a lower risk of major bleeding compared with both warfarin and rivaroxaban," the authors concluded. "Such associations appear to be stronger in the subgroup of patients with concomitant use of ART, raising concerns for potential drug interaction between ART and OACs."
While no significant differences in ischemic stroke or all-cause mortality rates were observed, gastrointestinal bleeding followed the same trend, with warfarin (HR, 2.99; 95% CI, 1.52-5.90) and rivaroxaban (HR, 3.38; 95% CI, 1.57-7.25) posing higher risks than apixaban.
The study also noted that a considerable proportion of patients (29%-31% across cohorts) were not receiving ART at the time of anticoagulant initiation, which the authors identified as an important finding worthy of further investigation.
"Studies have also shown that ART may induce or inhibit the metabolism of warfarin, increasing the risk for drug interactions and causing adverse reactions," the researchers noted in their discussion of potential mechanisms.
The authors acknowledged the study's observational nature and potential residual confounding but emphasized that it fills a critical evidence gap, as prior trials excluded HIV patients.
Full disclosures are available in the published study.
