520 patients with osteoarthritis in 1040 knees were randomized to undergo computer-navigated total knee arthroplasty for one knee and conventional total knee arthroplasty (TKA) for the other knee. 200 patients received press-fit condylar posterior cruciate-retaining mobile-bearing knee prosthesis, and 320 patients received a posterior-stabilized high-flexion fixed-bearing knee prosthesis. The results of the study suggest that computer-navigated total knee arthroplasty did not achieve significantly better clinical function, alignment, or survivorship, when compared to conventional method.