Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry of England and Wales

Background Total hip replacement (THR) is extremely common. Some prostheses fail, particularly in younger patients, and need to be revised, most commonly for loosening secondary to wear or dislocation. Surgeons have tried to address these problems by implanting large diameter metal-on-metal bearing surfaces. Our aim was to assess if metal-on-metal bearing surfaces lead to increased implant survival compared with other bearing surfaces in stemmed THR and, additionally, if larger head sizes result in improved implant survival. Methods We analysed the National Joint Registry of England and Wales for primary hip replacements (402 051, of which 31 171 were stemmed metal-on-metal) undertaken between 2003 and 2011. Our analysis was with a multivariable flexible parametric survival model to estimate the covariate-adjusted cumulative incidence of revision adjusting for the competing risk of death. Findings Metal-on-metal THR failed at high rates. Failure was related to head size, with larger heads failing earlier (3.2% cumulative incidence of revision [95% CI 2.5-4.1] for 28 mm and 5.1% [4.2-6.2] for 52 mm head at 5 years in men aged 60 years). 5 year revision rates in younger women were 6.1% (5.2-7.2) for 46 mm metal-on-metal compared with 1.6% (1.3-2.1) for 28 mm metal-on-polyethylene. By contrast, for ceramic-on-ceramic articulations larger head sizes were associated with improved survival (5 year revision rate of 3.3% [2.6-4.1] with 28 mm and 2.0% [1.5-2.7] with 40 mm for men aged 60 years). Interpretation Metal-on-metal stemmed articulations give poor implant survival compared with other options and should not be implanted. All patients with these bearings should be carefully monitored, particularly young women implanted with large diameter heads. Since large diameter ceramic-on-ceramic bearings seem to do well we support their continued use.