OBJECTIVETo summarize the mid-term follow-up results of revision of total knee arthroplasty and compare the different strategies for infective revisions.

METHODSAll of 45 patients (47 operated knees) lived in Beijing were treated from April 1989 to October 2010 in Arthritis Clinic and Research Center, Peking University People's Hospital. There were 6 male and 39 female patients, who aged from 31 to 77 years (mean (62 ± 11) years). The function of knee, satisfaction and imaging then were compared retrospectively. American Knee Society Scores (KSS), Western Ontario & McMaster University Osteoarthritis Index (WOMAC), the medical outcomes study item short form health survey (SF-36) scales and satisfaction/pain visual analogue scales (VAS) of patients were evaluated. The patients were divided into infection group (33 patients, 34 knees) and non-infection group (12 patients, 12 knees) according to the indication of revision of total knee arthroplasty and compared by t-tests.

RESULTSThe time from operation to follow-up was 1 year and 2 months to 17 years. The mid-term follow-up time was 8 years 3 months. There were significant improvements of KSS clinical and function scores (from 66.9 ± 28.0 and 44.4 ± 37.6 to 25.4 ± 24.2 and 10.0 ± 24.8, t = 7.043 and 3.797, both P = 0.001). Patients of infection group had lower KSS clinical and function scores than non-infection group before operation, and lower Society Function (t = 2.225, 3.520 and 2.885, P = 0.035, 0.002 and 0.007). About the septic group, the II-stage group had significant better post-operation KSS function scores, Society Function, physical component summary, WOMAC functional score and WOMAC score than I-stage group (t = 2.160-3.268, P = 0.004-0.042). The 1-year, 2-year, 6-year, 17-year survival rate were 83.6%, 78.7%, 62.1%, 44.5%.

CONCLUSIONSRevision total knee arthroplasty is an effective method for solving the failure of primary total knee arthroplasty. It can improve the pain and activity difficulty following the failure of primary total knee arthroplasty, and partially improve function along with quality of life. The results of non-infection group are better than infection group. There may be better results for II-stage revision total knee arthroplasty than I-stage revision. Both I-stage and II-stage revision total knee arthroplasty are effective.

Adult ; Aged ; Arthroplasty, Replacement, Knee ; Female ; Follow-Up Studies ; Humans ; Knee Joint ; surgery ; Male ; Middle Aged ; Pain Measurement ; Quality of Life ; Reoperation ; Retrospective Studies ; Surgical Wound Infection ; epidemiology