Catastrophic osteolysis in total knee replacement. A report of 17 cases.

One hundred eighty-five total knee revisions were done by the 2 senior authors from 1988 to 1994. Of these, 17 patients were identified as having severe osteolysis. The average age was 62.7 years, and the average weight was 90.8 g. All patients had osteoarthritis. Preoperative aspiration was negative in all patients. Eight patients had a metal-backed patella, 8 had a cementless implant, 4 had a cemented implant, and 5 had a hybrid implant. The average time interval from the index surgery to radiographic evidence of lysis was 56 months. Radiographic review showed gross polyethylene wear in 10 patients, loose tibial implants in 8, fractured baseplates in 2, and evidence of a metallic synovitis in 2. Retrieved polyethylene liners had a average thickness of 7 mm, and none were conforming. Osteolytic defects were defined and classified according to Engh's criteria. The histology showed a foreign body reaction with intracellular refractile polyethylene particles. Four cases had metallic debris. The exposure at the time of revision required a rectus snip in 4, V-Y-quadricepsplasty in 2, and tibial tubercle osteotomy in 1. Posterior stabilized implants were used in 65% and a constrained implant in 30%. Lytic defects were reconstructed with cement only in 47% of knees, allograft in 30%, and metallic wedges in 35%. Based on this report, the authors conclude that younger, overweight patients seem to be at higher risk of implant failure and that osteolysis is not restricted to cementless implants. Furthermore, because review of the radiographs leads to a constant underestimation of the lytic defect, the surgeon must be prepared to deal with complex revisions.