To compare the risk of readmissions for major bleeding within one year between apixaban and rivaroxaban as a component of triple antithrombotic therapy.

This study was a multicenter, retrospective cohort study conducted at two academic medical centers in the Western New York and New York City region between July 1, 2011 and September 25, 2019. Adult patients were included if they were diagnosed with atrial fibrillation or venous thromboembolism and discharged on new triple antithrombotic therapy. The primary outcome compared the rates of 1-year readmission for major bleeding between apixaban and rivaroxaban groups. Secondary outcomes included rate of ischemic outcomes. Time to event analysis was determined with a Kaplan–Meier plot and Cox proportional hazard ratios (HR).

A total of 378 patients were included in the study, 212 in the apixaban group and 166 in the rivaroxaban group. Within 1 year, readmission for major bleeding events occurred in six (2.8%) patients in the apixaban group and four (2.4%) patients in the rivaroxaban group (P = 1.000). After adjustment, the major bleeding event rate was not statistically significantly different between apixaban and rivaroxaban [adjusted hazard ratio (aHR) 0.68, 95% confidence interval (CI) 0.12–3.77; P = 0.6624]. Higher albumin levels were identified to be protective against major bleeding related readmission events (aHR 0.18, 95% CI 0.05–0.63; P = 0.0072). The ischemic outcome occurred in seven (3.3%) patients in the apixaban group and three (1.8%) in the rivaroxaban group (P = 0.7368).

Use of apixaban or rivaroxaban in a triple antithrombotic regimen was not associated with bleeding or ischemic outcomes.