Apixaban Reduces Bleeding Risk Versus Rivaroxaban in Venous Thromboembolism

In a randomized trial of 2,760 patients, apixaban resulted in a 3.3% bleeding rate compared to 7.1% for rivaroxaban over three months.

For Doctors in a Hurry

Researchers investigated the uncertain difference in bleeding risk between apixaban and rivaroxaban for treating acute venous thromboembolism.
This randomized, open-label trial assigned 2760 patients with acute venous thromboembolism to receive either apixaban or rivaroxaban for three months.
Clinically relevant bleeding occurred in 3.3% of apixaban patients versus 7.1% of rivaroxaban patients (relative risk 0.46; 95% confidence interval 0.33 to 0.65; P<0.001).
The authors concluded that apixaban significantly lowers the risk of clinically relevant bleeding compared to rivaroxaban during three months of treatment.
Clinicians treating acute venous thromboembolism can utilize apixaban rather than rivaroxaban to significantly reduce the risk of bleeding events over a three-month period.

Weighing Bleeding Risks in Venous Thromboembolism

Direct oral anticoagulants have largely replaced vitamin K antagonists as the standard of care for managing acute venous thromboembolism, offering comparable efficacy without the need for routine coagulation monitoring [1]. While these medications generally demonstrate favorable safety profiles, the risk of hemorrhage remains a primary clinical concern, particularly in vulnerable populations such as patients with end-stage renal disease. In this group, apixaban reduces major bleeding rates compared to warfarin (relative risk 0.53, 95% confidence interval 0.45 to 0.64) [2]. Similarly, in adults aged 75 years and older, network meta-analyses show that apixaban significantly decreases bleeding risk compared to rivaroxaban (relative risk 0.28, 95% confidence interval 0.09 to 0.86) [3]. Broad analyses encompassing 76,641 patients further indicate that the risk of major bleeding is significantly lower with apixaban than with dabigatran (odds ratio 2.10, 95% confidence interval 1.07 to 4.12) and edoxaban (odds ratio 2.64, 95% confidence interval 1.36 to 5.15) [4]. However, clinicians have long lacked definitive, head-to-head randomized trial data directly comparing the safety of the two most commonly prescribed options. A new international trial now provides direct comparative evidence to help physicians navigate this daily prescribing dilemma.

Trial Design and Dosing Protocols

Apixaban and rivaroxaban are the oral anticoagulants most frequently used to treat acute venous thromboembolism, yet uncertainty remains regarding their comparative bleeding risks. To address this knowledge gap, researchers conducted the COBRRA trial (ClinicalTrials.gov number NCT03266783), funded by the Canadian Institutes of Health Research and other organizations. The study utilized a prospective, randomized, open-label, blinded end-point design. This methodology means that while patients and their treating physicians knew which medication was being administered, the investigators assessing the clinical outcomes were completely unaware of the group assignments, thereby minimizing detection bias while reflecting real-world clinical practice. Eligible patients enrolled in the trial had acute symptomatic pulmonary embolism or proximal deep-vein thrombosis. During the enrollment phase, a total of 2760 patients underwent randomization. These individuals were assigned in a 1:1 ratio to receive either apixaban or rivaroxaban for a duration of 3 months. Specifically, 1370 patients were randomized to the apixaban group, while 1390 patients were randomized to the rivaroxaban group. The investigators utilized standard therapeutic dosing regimens for both medications. In the apixaban cohort, the drug was given at a dose of 10 mg twice daily for 7 days followed by 5 mg twice daily. Conversely, in the rivaroxaban cohort, the medication was given at a dose of 15 mg twice daily for 21 days followed by 20 mg daily. This 3-month treatment period allowed the researchers to directly compare the safety profiles of the two regimens during the highest-risk phase of acute anticoagulation therapy, providing actionable guidance for initial treatment decisions.

Primary Bleeding Outcomes

To evaluate the safety profiles of the two anticoagulants, the investigators established that the primary outcome was clinically relevant bleeding during the 3-month trial period. The researchers ensured standardized reporting by defining clinically relevant bleeding as a composite of major bleeding or clinically relevant nonmajor bleeding according to the International Society on Thrombosis and Haemostasis criteria. This composite metric captures both life-threatening hemorrhages and bleeding events that, while not classified as major, still require medical intervention, prompt a clinical evaluation, or necessitate a change in antithrombotic therapy. The trial results demonstrated a clear safety distinction between the two medications. A primary-outcome event occurred in 44 of 1345 patients (3.3%) in the apixaban group, whereas the same outcome occurred in 96 of 1355 patients (7.1%) in the rivaroxaban group. Statistical analysis revealed that the relative risk for the primary outcome of clinically relevant bleeding was 0.46 (95% confidence interval [CI], 0.33 to 0.65; P<0.001). Consequently, the findings indicate that among patients with acute venous thromboembolism, the risk of clinically relevant bleeding is significantly lower with apixaban than with rivaroxaban during the 3-month treatment period. For practicing physicians, these data provide direct evidence to inform anticoagulant selection, suggesting that apixaban may be the preferable agent to mitigate hemorrhagic complications during the critical early phase of treatment.

Secondary Outcomes and Adverse Events

Beyond the primary bleeding endpoints, the researchers evaluated overall mortality to ensure a comprehensive safety profile. Secondary outcomes included death from any cause, which remained exceptionally low across both treatment cohorts. Specifically, death from any cause occurred in 1 patient (0.1%) in the apixaban group and in 4 patients (0.3%) in the rivaroxaban group. Statistical analysis showed that the relative risk for death from any cause was 0.25 (95% CI, 0.03 to 2.26). The wide confidence interval crossing 1.0 indicates no statistically significant difference in mortality between the two oral anticoagulants during the three-month study period. For clinicians, this confirms that the observed reduction in bleeding with apixaban does not come at the cost of an increased mortality risk or compromised efficacy. The investigators also tracked other severe medical complications to determine if either medication carried hidden risks outside of hemorrhage. Serious adverse events unrelated to bleeding or venous thrombosis occurred in 36 patients (2.7%) in the apixaban group and in 30 patients (2.2%) in the rivaroxaban group. These similar rates suggest that both medications share comparable baseline safety profiles regarding non-hemorrhagic and non-thrombotic complications. Consequently, physicians can interpret the primary bleeding differences as a distinct pharmacological effect rather than a reflection of overall systemic toxicity, further supporting the safety of apixaban in the acute management of venous thromboembolism.

Study Info

Bleeding Risk with Apixaban vs. Rivaroxaban in Acute Venous Thromboembolism

Lana A. Castellucci, Vivien Chen, Michael J. Kovacs, Alejandro Lazo-langner, et al.

Journal New England Journal of Medicine

Published March 11, 2026