BACKGROUND: Vitamin K antagonists are the only oral anticoagulants approved to prevent valve thrombosis and valve-related thromboembolism in patients with mechanical heart valves. Whether patients with an On-X mechanical aortic valve can be safely anticoagulated with apixaban is unknown. METHODS: Patients with an On-X aortic valve implanted at least 3 months before enrollment were randomly assigned to receive apixaban 5 mg twice daily or warfarin (target international normalized ratio 2.0 to 3.0). The primary efficacy end point was the composite of valve thrombosis or valve-related thromboembolism with coprimary analyses comparing apixaban with warfarin for noninferiority and comparing the apixaban event rate with an objective performance criterion (OPC). RESULTS: The trial was stopped after 863 participants were enrolled owing to an excess of thromboembolic events in the apixaban group. Most (94%) participants took aspirin. A total of 26 primary end-point events occurred, 20 (in 16 participants) in the apixaban group (4.2%/patient-year; 95% confidence interval [CI], 2.3 to 6.0) and 6 (in 6 participants) in the warfarin group (1.3%/patient-year; 95% CI, 0.3 to 2.3). The difference in primary end-point rates between the apixaban and warfarin groups was 2.9 (95% CI, 0.8 to 5.0); noninferiority and OPC success criteria were not met. Major bleeding rates were 3.6%/patient-year with apixaban and 4.5%/patient-year with warfarin. CONCLUSIONS: Apixaban did not demonstrate noninferiority to warfarin and is less effective than warfarin for the prevention of valve thrombosis or thromboembolism in patients with an On-X mechanical aortic valve. (Funded by Artivion; ClinicalTrials.gov number, NCT04142658.)