Sixty-three total knee replacements were performed after a failed tibial osteotomy. The goal of this study was to compare the perioperative problems and the outcome of this group of patients (study group) to a group of patients with primary arthroplasties matched for age, gender, length of follow-up, weight, and preoperative Charnley class. Operative problems were more frequently encountered in the study group, with 7 tibial tubercle elevations and 15 lateral retinaculum releases needed, whereas lateral retinaculum release was necessary for only 1 knee in the control group. Outcome was assessed using both the International Knee Society (IKS) scoring system and Hospital for Special Surgery (HSS) knee score. The follow-up period averaged 4.6 years. The IKS score of the control group was significantly higher, averaging 80.9 +/- 13.8, whereas it was 74.4 +/- 14.8 for the study group (P = .0001). Among the parameters included in the knee score, only pain was significantly different with the control group (P = .03). The IKS function score and the HSS score were not statistically different. Conversion of a failed tibial osteotomy is a technically demanding procedure. Careful preoperative planning is needed. Results, especially on pain, appeared to be inferior to those for primary arthroplasties.