Functional results of isolated femoral revision of hip resurfacing arthroplasty.

Background: Conversion of a failed femoral resurfacing component is reportedly a straightforward procedure; however, little has been published regarding the functional results following revision. Our primary aim was to compare the functional results for a group of patients who had had isolated femoral component revision after a failed hip resurfacing with those for a group of patients with a surviving hip resurfacing. Our secondary aim was to identify whether the mode of failure affects functional outcome.

Methods: Between 1997 and 2002, data were prospectively collected on 5000 Birmingham Hip Resurfacing procedures. One hundred and seventy-seven hips were revised, and, of those, seventy-six had an isolated femoral component revision. We reviewed the modes of failure and the post-revision clinical outcomes for this subgroup (the revised implant group) and compared the results with those for the patients who had a surviving hip resurfacing implant (the surviving implant group).

Results: The median Harris and Merle d'Aubigné hip scores were significantly better in the surviving implant group than in the revised implant group (median Harris score, 96 compared with 82 [p < 0.001]; median Merle d'Aubigné score, 17 compared with 14 [p < 0.001]). When we analyzed outcomes following revision, we found that the mode of failure affected outcome. Patient satisfaction and clinical outcomes were worse following revision because of femoral component loosening in comparison with revision because of femoral neck fracture or revision because of femoral head collapse or osteonecrosis. In these three subgroups, the median Harris hip scores were 66, 87, and 92, respectively, and the median Merle d'Aubigné scores were 10, 14, and 15, respectively. Six of fifteen patients in the femoral loosening group believed that they were worse or much worse after the revision than they had been before the primary procedure, compared with four of twenty-five patients in the femoral neck fracture group and two of twenty-two patients in the femoral head collapse or osteonecrosis group. Four patients (four hips) experienced complications as a consequence of revision surgery (three deep infections and one case of loosening of an uncemented femoral stem).

Conclusions: The functional results for patients who had revision of the femoral component of a current-generation metal-on-metal resurfacing arthroplasty were worse than those for patients with a surviving hip resurfacing. Patients who had revision because of femoral component loosening had worse outcomes than those who had revision because of femoral neck fracture or femoral head collapse or osteonecrosis.