Background Proper rotational alignment during total knee arthroplasty (TKA) is important for adequate postoperative patellofemoral and tibiofemoral kinematics,as well as for achieving balanced flexion space at 90°.The effects of computer navigation-assisted total knee replacement and conventional total knee arthroplasty on rotational alignment,mechanical axis,component position and clinical outcomes were compared.Methods Two methods were used in 82 patients and the rotation of the femoral and tibial components in the transverse plane,the combined rotation of the two components,the mismatch between them,and the mechanical axis of the lower limb were analyzed.All of these parameters were measured from postoperative radiographs and computed tomography images.Functional outcomes were compared at 6 weeks and 6 months postoperatively.Results Significant differences were found between the two techniques (P ＜0.05) in the following parameters:average rotation of the femoral component ((1.51±3.55)° vs.(-0.63±3.04)°); combined rotation of the femoral and tibial components (2.85±4.07)° vs.(0.28±3.43)°); and mismatch between the femoral and tibial components ((1.44±4.55)° vs.(-0.43±2.86)°).Differences in the rotation of the tibial component were not statistically significant.The prevalence of outliers (malalignment ＞±3° intemal/external rotation) of the femoral component (31.7％ vs.12.5％) and the tibial component (36.6％ vs.15％) were significantly reduced when the navigation system was used (P＜0.05).In addition,while patients in the navigation group had significantly better mechanical axis and functional outcomes at 6 weeks after surgery (P ＜0.05),there was no significant difference between the two groups (P ＞0.05) with respect to functional outcomes at 6 months.Conclusion The navigation system exhibited higher accuracy than the conventional technique in the transverse and coronal plane,and provided better early functional outcomes.