A series of 123 revision total knee replacements performed at three centers and followed up for 2 to 4 years was evaluated. In cases in which exposure could be obtained with undue tension on the patellar tendon, the surgical approach was modified using either quadriceps turndown (14 cases) or tibial tubercle osteotomy (15 cases). The remaining 94 patients underwent a standard operative approach that consisted of a medial parapatellar capsular incision which in 31 cases was combined with a quadriceps snip. The results were compared using the Knee Society clinical score, a patient satisfaction survey, and a patellofemoral questionnaire. Postoperatively, the group of patients who had a quadriceps snip was equivalent to the group of patients who underwent a standard approach in every parameter measured, and the groups therefore were combined for comparison purposes. The patients who had quadriceps turndown and the tibial tubercle osteotomy had equivalent scores postoperatively both of which were significantly lower than the standard group. The group of patients who had quadriceps turndown had a significantly greater increase in arc of motion than the tibial tubercle osteotomy group. The tibial tubercle osteotomy group had a lower degree of extension lag but a higher percentage of patients who had difficulty with kneeling and stooping and a higher percentage of patients who had difficulty with kneeling and stooping and a higher percentage of patients who thought the surgery was unsuccessful in relieving pain and unsuccessful in returning them to normal daily activities.