The Influence of Head Size and Sex on the Outcome of Birmingham Hip Resurfacing

Background Hip resurfacing has gained popularity for the treatment of young and active patients who have arthritis. Recent literature has demonstrated an increased rate of revision among female patients as compared with male patients who have undergone hip resurfacing. The aim of the present study was to identify any differences in survival or functional outcome between male and female patients with osteoarthritis who were managed with metal-on-metal hip resurfacing. Methods A prospective collection of data on all patients undergoing Birmingham Hip Resurfacing at a single institution was commenced in July 1997. On the basis of the inclusion and exclusion criteria, 1826 patients (2123 hips, including 799 hips in female patients and 1324 hips in male patients) with a diagnosis of osteoarthritis who had undergone the procedure between July 1997 and December 2008 were identified. The variables of age, sex, preoperative Oxford Hip Score, component size used, surgical approach, lead surgeon, and surgeon experience were analyzed. A multivariate Cox proportional hazard survival model was used to identify which variables were most influential for determining revision. Results The mean duration of follow-up was 3.46 years (range, 0.03 to 10.9 years). The five-year cumulative survival rate for the 655 hips that were followed for a minimum of five years was 97.5% (95% confidence interval, 96.3% to 98.3%). There were forty-eight revisions. Revision was significantly associated with female sex (hazard rate, 2.03 [95% confidence interval, 1.15 to 3.58]; p = 0.014) and decreasing femoral component size (hazard rate per 4-mm decrease in size, 4.68 [95% confidence interval, 4.36 to 5.05]; p < 0.001). Revision was not associated with age (p = 0.88), surgeon (p = 0.41), surgeon experience (p = 0.30), or surgical approach (p = 0.21). A multivariate analysis including the covariates of sex, age, surgeon, surgeon experience, surgical approach, and femoral component size demonstrated that sex was no longer significantly associated with revision when femoral component size was included in the model (p = 0.37). Femoral component size alone was the best predictor of revision when all covariates were analyzed (hazard rate per 4-mm decrease in size, 4.87 [95% confidence interval, 4.37 to 5.42]; p < 0.001). Conclusions The present study demonstrates that although female patients initially may appear to have a greater risk of revision, this increased risk is related to differences in th