Total hip arthroplasty after operative treatment of an acetabular fracture.

Sixty-six primary total hip arthroplasties were performed to treat post-traumatic osteoarthrosis that had developed following an acetabular fracture and subsequent open reduction and internal fixation. The mean age of the patients at the time of the total hip arthroplasty was fifty-two years (range, nineteen to eighty years). The arthroplasty was performed with cement in forty-four hips and without cement in twenty hips; in the remaining two hips, the acetabular component was inserted without cement and the femoral component was inserted with cement (a so-called hybrid procedure). Scarring from a previous procedure, retained hardware, heterotopic bone, and residual osseous deformity and deficiency made the procedure more complex than routine total hip arthroplasty in most patients. However, only one of the sixty-six procedures was associated with an operative complication. Three patients were lost to follow-up. The remaining sixty-three patients were followed for a mean of 9.6 years (range, two to twenty years). The mean duration of follow-up was 14.9 years for the acetabular components inserted with cement, 11.6 years for the femoral components inserted with cement, 4.6 years for the femoral components inserted without cement, and 3.9 years for the acetabular components inserted without cement. The mean Harris hip score improved from 49 points preoperatively to 93 points at the latest follow-up evaluation for the forty-six patients who did not have a revision procedure after the index arthroplasty. Seventeen patients had a revision; sixteen revisions were performed because of aseptic loosening of one or both components (nine acetabular and eleven femoral components). Mechanical failure (radiographic loosening or revision due to aseptic loosening) occurred in twenty-five hips. As determined with use of the Kaplan-Meier method, the ten-year survival rate, with revision due to aseptic loosening as the end point, was 78 per cent (95 per cent confidence interval, 66 to 92 per cent) for the prosthesis as a whole (that is, no revision of either component), 87 per cent (95 per cent confidence interval, 76 to 99 per cent) for the acetabular component, and 84 per cent (95 per cent confidence interval, 72 to 97 per cent) for the femoral component. An age of less than fifty years (p = 0.02), a weight of eighty kilograms or more (p = 0.047), and large residual combined segmental and cavitary deficiencies in the acetabular bone (p < 0.0001) were significant risk factors for revision because of aseptic loosening. At the ten-year follow-up, none of the twenty-two acetabular components that had been inserted without cement had been revised or demonstrated radiographic loosening. The ten-year rate of failure due to aseptic loosening was higher than that in many reported series of total hip arthroplasties performed for other indications; this was probably partly because of the young mean age of the patients, the high number of patients who had Charnley class-A involvement, and the predominantly male cohort.