Background: Many guidelines indicated that to continue using anticoagulant drugs to reduce the incidence of symptomatic venous thrombosis (SVT) in total knee arthroplasty (TKA) patients after discharged. Objective: This retrospective analysis included 407 consecutive patients who underwent TKA at the Lanzhou University Second Hospital in China. Cases history from Jan. 2013 to Dec. 2016. Methods: The group A were taken hypodermic injection of low molecular heparin for 5-10 days after surgery at the hospitalization time, then continued oral rivaroxaban after hospital discharge. The group B were only taken hypodermic injection of low molecular heparin for 5-10 days at the hospitalization time. Baseline characteristics of patients, total complications of SVT，any venous thromboembolic disease (VTE), deep venous thrombosis (DVT), pulmonary embolism (PE), lower limbs swelling, bleeding and mortality following hospital discharge were compared between the two groups. Main outcome measures: The main outcome measures were incidence of bleeding and death duration of in-hospital and out-hospital stay in this study. Results: After analysis, the results showed that the incidence of SVT patients had no significant difference between the two groups. Moreover, the incidence of VTE, major/minor bleeding, total–DVT, PE, fatal PE and death all had no significant differences between the two groups. But the patients of no smoking or no drinking were superior than the smoking or drinking in the incidence of symptomatic thrombosis. Conclusion: There is no enough evidence showed that the TKA patients given anticoagulant after hospital discharged had beneficial to decrease the risk of SVT incidence. Furthermore, the smoking and drinking would increase the risk of symptomatic thrombosis for TKA patients.