Early failure of precoated femoral components in primary total hip arthroplasty

Background: In an effort to decrease the rate of aseptic loosening, certain cemented femoral components were designed to have a roughened or textured surface with a methylmethacrylate precoating. Reports differ as to whether this step has increased or decreased the rate of failure. This study was designed to evaluate this issue. Methods: Five hundred and fourteen hips treated with a cemented Harris Precoat stem (Zimmer, Warsaw, Indiana) were evaluated clinically and radiographically and compared with 254 hips treated with an uncoated Harris Design-2 stem (Howmedica, East Rutherford, New Jersey). Prostheses that had been removed at revision were examined. The cementing and surgical techniques were identical and the population demographics were similar for these two groups. Results: The mean durations of follow-up were 8.4 and 13.5 years for the Precoat and uncoated Design-2 stems, respectively. At those times, at least forty-nine (9.5%) of the 514 Precoat components and at least ten (3.9%) of the 254 uncoated Design-2 stems had failed (p = 0.006). Five Precoat stems fractured, and no uncoated Design-2 stems fractured. Component failure was associated with use in young, active, heavy men with a diagnosis of avascular necrosis and generally with the use of smaller components. The cementing technique was satisfactory in the majority of the patients, and there were no qualitative differences in cementing technique between the hips that failed and those that did not. The mechanisms of failure of the Precoat prostheses included bone-cement loosening, focal osteolysis, stem fracture, and prosthesis-cement debonding. Fractures of smaller components occurred as a result of fatigue failure and were associated with good distal fixation but proximal stem loosening. Conclusions: The rate of failure of roughened, precoated, cemented femoral components was considerably higher and occurred earlier than that of femoral components that were neither textured nor precoated with methylmethacrylate. Younger patients with avascular necrosis had a higher risk of failure; however, this factor alone did not completely explain the differences in outcome between these two components. The causes of aseptic loosening are multifactorial and may be related to component design and size as well as to precoating and surface finish.