The Effect of BMI and Gender on Bleeding Events when Rivaroxaban Is Administered for Thromboprophylaxis Following Total Hip and Total Knee Arthroplasty

Rivaroxaban is approved in Europe and the United States for thromboprophylaxis following total joint arthroplasty. As the rate of obesity increases, confirming safety and efficacy in this patient population is paramount. This retrospective chart review assessed the efficacy and safety of rivaroxaban between two body mass index (BMI) groups: normal or overweight (< 30 kg/m²) and obese or morbidly obese (≥30 kg/m²). Safety outcome was a major bleeding event, defined as a decrease in hemoglobin of at least 2 g/dL from postoperative day 1(POD 1) to discharge or a blood transfusion of at least two units. Efficacy outcome was venous thromboembolism within 35 days postoperatively. There were 68 (68/1,241; 5.48%) major bleeding events. There was no significant association between major bleeding events and BMI in the univariable analysis (<i>p</i> < 0.36). However, after adjusting for other factors in the multivariable model, there was a significant interaction between BMI and gender (<i>p</i> < 0.001). Among males, the incidence of major bleeding events was three times greater in obese/morbidly obese subjects as compared with normal/overweight male subjects (odds ratio [OR]: 3.09, 95% confidence interval [CI]: 1.25, 7.62). Among females, incidence of having a major bleeding event was almost two times greater in normal/overweight subjects as compared with obese/morbidly obese female subjects (OR: 2.17, 95% CI: 1.10, 4.35). Incidence of venous thromboembolism was 0.4% in each group. The authors' study results highlight a previously unexplored association between BMI and gender-dependent differences in bleeding outcomes when rivaroxaban is used for thromboprophylaxis following total joint arthroplasty.