OBJECTIVE： To observe the effects and safety of different treatment courses of rivaroxaban in prevention and treatment of deep venous thrombosis （DVT） after total hip replacement.  METHODS： The clinical data of 127 patients underwent selective total hip replacement in our hospital during Dec. 2014-Dec. 2015 were selected and analyzed retrospectively. They were divided into group A [83 cases， subdivided into group A1 （40 cases） and A2 （43 cases） by treatment course] and group B （44 cases） according to anticoagulant duration. Group A was given Rivaroxaban tablet 10 mg， po， qd， 6-8 h after surgery； the treatment course of group A1 was 5 weeks （35 d） while that of group A2 was 2 weeks （14 d）. Group B was given Enoxaparin sodium injection 5 000 AxaIU，sc，qd， 12d before surgery； and then given Enoxaparin sodium injection 5 000 AxaIU，sc，qd， after surgery； the treatment course lasted for 2 weeks. The incidence of DVT were compared among 3 groups as well as thrombin time （TT）， activated partial thromboplastin time （APTT）， hemoglobin （Hb） content， whole blood viscosity， plasma viscosity， platelet count and anticoagulant safety 3 days before surgery and after treatment. RESULTS： The incidence of DVT in group A1 was 0， which was significantly lower than that in group A2 （9.30％） and group B （11.36％）， with statistical significance （P＜0.05）； there was no significant differences between group A2 and group B （P＞0.05）. Whether 3 d before surgery or after treatment， there was no statistical significance in PT， APTT， Hb content， whole blood viscosity， plasma viscosity and platelet count among 3 groups （P＞0.05）. No major bleeding events or pulmonary embolism was found in 3 groups. The incidence of mild bleeding events in group A was 5.00％， which was slightly higher than that in group A2 （2.33％） and group B （2.27％）， without statistical significance （P＞0.05）； there was no statistical significance between group A2 and group B （P＞0.05）. CONCLUSIONS： Rivaroxiban long-term anticoagulation after total hip replacement has no significant effect on coagulation function and platelet， not only decrease the incidence of DVT， but also will not increase no bleeding with good safety.