Revision knee arthroplasty due to aseptic loosening

PURPOSE OF THE STUDY Anatomic changes associated with aseptic loosening make conditions for revision of total knee arthroplasty more difficult. The aim of this study was to evaluate the results of revision total knee replacement at an average follow-up of 6.1 years. MATERIAL AND METHODS A total of 97 revision knee replacements due to aseptic loosening carried out in the years 1992 to 2003 were evaluated. The group included 46 men and 51 women at an average age of 66.8 years. The average preoperative Knee Society Score (KKS) was 31 points and the Functional Score (FS) was 22 points. There were 41 minor operations for AORI type I defects, 49 moderately serious procedures for AORI type II defects and seven major operations for AORI type III defects. In minor procedures standard components were implanted in 14 patients, standard components with cemented stems with extension were used in nine, and polyethylene plateau exchange was carried out in 18 patients. For moderately serious procedures, posterior stabilized components with extended cemented stems were used in 15 patients, revision implants with cementless stems in 26 patients and standard components with cemented stems in eight patients. In seven patients with major surgery, the hinged type of prosthesis was employed. Radiographic results were evaluated on the basis of Ewald's classification. RESULTS Clinical findings showed improvement of the average KKS from 31 to 74 points at follow-up of 6.1 years. Functional outcomes improved, as shown by the average FS, from 22 to 67 points. Fifteen patients were not satisfied with the outcome of surgery, the causes being aseptic loosening in four, deep infection in eight and pain due to progression of radiolucent lines of the tibia in three patients. DISCUSSION The results of revision surgery with component replacement because of aseptic loosening are worse in comparison with those of primary total knee replacement. The average KSS score after revision surgery was 74 points at 6.1-year follow-up, whereas after primary surgery it was 92 points at 6.5 years. The average FS score after revision was 67 points, as compared with 86 points at 6.5 years after primary surgery. Complications involving infection occurred in 8.2% of the revision cases, but only in 0.8% of the primary operations. The authors used modular systems because these provide more options. Good outcomes were achieved with morselized bone grafting for filling cavitary defects. In patients with large defects in tibial or femoral metaphyses, posterior stabilized components and cementless intramedullary stems were used with good results. CONCLUSIONS The authors recommend to avoid filling large bone defects with bone cement. They prefer bone grafting. In the case of good quality metaphyseal bone, they use standard components or posterior stabilized components with or without additional cemented or cementless short stem extensions. In the case of poor quality metaphyseal bone with defects, they use revision implants with cementless long stems. The authors have achieved good results with off-set stems.