Between January 1989 and December 1994, 94 patients (96 knees) had a two-stage reimplantation for treatment of an infected total knee arthroplasty. All patients were treated with an interval antibiotic-loaded static cement spacer and had antibiotic-loaded bone cement for prosthesis fixation at the time of reimplantation. The purpose of this study was to assess the long-term risk of reinfection and the mechanical durability of these reimplantation arthroplasties. Patients were followed up for a median of 7.2 years (range, 2.5-13.2 years). At final followup, 15 knees (16%) had required reoperation. Nine knees (9%) had component removal for reinfection and six knees (6%) were revised for aseptic loosening. The median time to reoperation for reinfection was 1 year (range, 0.1-9.8 years). The risk of recurrent infection was not correlated with the type of organism, patient demographics, or method of prosthesis fixation at reimplantation. The survivorship free of implant removal for any reason was 90% (confidence intervals, 83.9-96.4%) at 5 years and 77.3% (confidence intervals, 65.5-89.6%) at 10 years. The survivorship free of implant removal for reinfection was 93.5% (confidence intervals, 88.5-98.7%) at 5 years and 85% (confidence intervals, 73.8-96.3%) at 10 years. Survival free of revision for mechanical failure (aseptic loosening or radiographic loosening) was 96.2% (confidence intervals, 92-100%) at 5 years and 91% (confidence intervals, 80.8-98.3%) at 10 years. These results suggest that the high likelihood of early success after two-stage reimplantation of an infected TKA is well maintained throughout long-term followup, with a modest rate of late recurrent infection or mechanical implant failure.