Introduction: In a previous study (ARCO, 2002), we reported that the clinical results of revision total hip arthroplasty for osteonecrosis patients were less satisfactory than those found for a matched group of osteoarthritis patients. The aim of this study was to evaluate the potential factors that may have contributed to these findings.

Materials and Methods: This study included 34 hips in 30 osteonecrosis patients who had undergone revision of a femoral total hip arthroplasty component. There were 19 men (22 hips) and 11 women (12 hips) who had a mean age of 46.1 years (range, 28 to 69 years). The surgeries were performed between March 1984 and January 2001. Most femoral stems (91%) were implanted without cement. Prostheses were of different stem lengths, but most (97%) were proximally porous-coated. The mean follow-up was 8.2 years [range, 0.1 (a re-revision) to 19.8 years]. A physical examination as well as patient and physician outcome forms were collected at each visit. Preoperative x-rays were categorized according to the technique of Della Valle and Paprosky. A Kaplan-Meier survival analysis was performed (PEPI statistical software package).

Results: Risk factors for osteonecrosis included 15 corticosteroid, 8 alcohol, 7 trauma, and 4 unknown. This was the first revision in 27 cases, second revision in 5 cases, and third revision in 2 cases. Preoperatively, the defects included 4 Type I, 9 Type II, 15 Type IIIA, 2 Type IIIB, 1 Type IV, and 3 unknown types. Of the 34 hips, the femoral component was re-revised in 12 cases. One of the failures was the only fully porous coated stem that was implanted. One of the 3 cemented implants failed, as compared to 11 of the 31 implanted without cement. Survival rates were 90.9% (74.4%–97.1%) at 5 years, 54.8% (24.9%–81.6%) at 10 years, 54.8% (19.9%–85.6%) at 15 years, and 27.4% (1.7%–88.9%) at 20 years. There was no relationship between frequency of re-revision and defect category, risk factors, or age.

Discussion: Although there was a high failure rate (12/34; 34%) in this patient cohort, over 50% survived at least 10–15 years. The lack of a relationship between the patient age or the extent of defect and re-revision suggest that other factors concerning this disease need to be examined.