Case Presentation:

A 59 year‐old man presented with three days of progressive dyspnea after right total knee arthroplasty. He had a history of Child‐Pugh class B alcoholic cirrhosis complicated by portal hypertension and encephalopathy and stage three chronic kidney disease. He was doing well undergoing rehabilitation at home until eleven days postoperatively when he presented to the emergency department with worsening dyspnea and hypoxia. Lungs had decreased breath sounds and diffuse crackles bilaterally and right leg revealed expected post surgical edema. Over the next twelve hours he had progressive hypoxic respiratory failure and was intubated. Chest radiography revealed bilateral patchy perihilar infiltrates. Coagulation studies revealed INR 3.9, PTT 45. WBC was 24; platelets were normal. Creatinine was 1.8. Admission culture data collected revealed no growth. Diagnostic bronchoscopy two days later revealed blood in his airways and hemorrhagic bronchoalveolar lavage results consistent with diffuse alveolar hemorrhage. Of note, he had been switched to rivaroxaban 10 mg daily 3 days prior to presentation from subcutaneous enoxaparin for venous thromboembolism prophylaxis. The patient was treated with rivaroxaban cessation and fresh frozen plasma. After a complicated hospital course, he was extubated and discharged to complete further rehabilitation.


Rivaroxaban is an oral factor Xa inhibitor being used more frequently by hospitalists for stroke prophylaxis in atrial fibrillation, VTE treatment and VTE prophylaxis following hip and knee arthroplasty. It is partially excreted unchanged by the kidneys and the remainder is hepatically metabolized and interacts with other inhibitors of cytochrome P450 CYP3A4 and P‐glycoprotein. The RECORD 3 and RECORD 4 trials demonstrated the safety and efficacy of rivaroxaban 10mg compared to subcutaneous enoxaparin for thromboprophylaxis after total knee arthroplasty. Bleeding events were not significantly different between the groups; however, these trials specifically excluded patients with hepatic and renal dysfunction. Looking at more medically complex patients, the MAGELLAN study showed rivaroxaban to be noninferior versus enoxaparin for thromboprophylaxis in medically ill patients. In that study, there were significantly more clinically relevant bleeding events in the rivaroxaban group at day 10 and day 35.


Once daily oral rivaroxaban is a new drug for thromboprophylaxis that has some advantages over previously used medications such as enoxaparin with its subcutaneous route and warfarin with its many dietary and medications interactions. However, rivaroxaban and other newer oral anticoagulants also have medications interactions via hepatic metabolism and renal clearance. Patients with renal and liver disease, particularly advanced disease, may not be suitable candidates for rivaroxaban due to the potential for increased bleeding risk and decreased drug clearance. Furthermore, drug interactions via cytochrome P450 CYP3A4 should be carefully considered in all patients. More studies are needed to determine the safety and efficacy of rivaroxaban in medically complex postoperative patients. Until then, caution should be used when selecting patients for thromboprophylaxis with rivaroxaban and hospitalists should maintain a high index of suspicion for bleeding complications in medically complex patients.