OBJECTIVE: To evaluate the efficacy and safety of three antithrombotic agents on venous thromboembolism (VTE) after unilateral total knee arthroplasty. METHODS: From November 2011 to March 2014, 149 patients undergoing unilateral total knee arthroplasty for knee osteoarthritis were reviewed. Among them, there were 66 males and 83 females, ranging in age from 48 to 76 years old. All the cases were randomly divided into three groups including Aspirin group, low-molecular-weight heparin (LMWH) group, and rivaroxaban group, according to antithrombotic agents. Deep vein thrombosis (DVT), pulmonary embolism (PE) and bleeding complication (including wound ecchymosis, hematoma and other local complications, gastrointestinal, cardiovascular, urinary hemorrhage and other major bleeding events) of antithrombotic agents were observed and analyzed statistically at the 6 week, 8 week, and 12 week after operation. RESULTS: Among patients who received Aspirin (48 cases), 4 patients had DVT, in 1 patient had PE, and 2 patients had bleeding complication. Among 54 patients in low-molecular-weight heparin group, 3 patients had DVT, 1 patient had PE, and 3 patients had bleeding complication. While among those patients received the rivaroxaban (47 cases), 3 patients had DVT, 0 patient had PE, and 11 patients had bleeding complication. There were no statistically differences among three groups on DVT, and PE (P>0.05). The incidence of bleeding complication in rivaroxaban group was higher than the other two antithrombotic agents, and the difference among the three groups was statistically significant (P<0.05). CONCLUSION: Aspirin, low-molecular-weight heparin, and rivaroxaban could effectively reduce the incidence of VTE after total knee arthroplasty, and their efficacy was similar. Rivaroxaban has a higher incidence of bleeding complication and further clinical trials are required to be conducted to assess the safety of rivaroxaban in clinical.