A retrospective matched-pair comparative analysis was done between 30 total knee arthroplasties following failed high tibial osteotomies and 30 total knee arthroplasties following failed unicompartmental knee arthroplasties. The groups were matched according to age, gender, type of prosthesis, primary disease, and length of followup. A minimum followup of 2 years was required for inclusion in the study, and the average followup was 3.8 years (range, 2-9 years). The Knee Society Knee Score for the high tibial osteotomy group was significantly higher than that for the unicompartmental arthroplasty group. More osseous reconstructions were required in the unicompartmental revisions. Difficulty with exposure was not significantly greater in the osteotomy group. Rates of component loosening were not significantly different between the groups. A failed unicompartmental knee arthroplasty and a failed high tibial osteotomy can be revised successfully to a total knee arthroplasty. The results confirm that revisions after unicondylar arthroplasty and high tibial osteotomy are technically demanding. In this series, the results of total knee arthroplasty following unicompartmental knee arthroplasty approached but did not equal those obtained after high tibial osteotomy.