Metal-on-metal hip resurfacing could be clinically beneficial, cost effective in younger men and women

Bozic KJ. Clin Orthop Relat Res. 2010;468(9):2301-2312.

Metal-on-metal hip resurfacing arthroplasty resulted in modest improvements in quality-adjusted life-years at a small incremental cost, according to researchers at the University of California, San Francisco.

Kevin J. Bozic, MD, MBA, and colleagues constructed a Markov decision model to compare the quality-adjusted life-years and costs of hip resurfacing arthroplasty (HRA) and total hip arthroplasty (THA) over 30 years. They included men and women aged 50 years or older undergoing HRA or THA for advanced hip osteoarthritis.

Using sensitivity analyses, the researchers assessed the effect of patient characteristics, clinical outcome probabilities, quality of life and costs on the discounted incremental costs, incremental clinical effectiveness and the incremental cost-effectiveness ratio (ICER) of HRA compared to THA.

The findings showed that metal-on-metal HRA produced small improvements in quality-adjusted life-years at a small incremental cost. For men younger than 65 years and women younger than 55 years, the procedure had an ICER less than $50,000 per quality-adjusted life year gained. The factors that had the biggest effect on costs and quality of life were metal-on-metal HRA and THA failure rates, device costs and the difference in quality of life after conversion from HRA to THA compared to THA.

“Metal-on-metal HRA could be clinically advantageous and cost effective in younger men and women,” the authors wrote. “Further research on the comparative effectiveness of metal-on-metal HRA vs. THA should include assessments of the quality of life and resource use in addition to the clinical outcomes associated with both procedures.”

The role of metal-on-metal hip resurfacing is under increasing scrutiny given the high success rate of traditional THA as well as reports of high failures rates with some resurfacing devices and at some centers. In this setting, the authors examine the issue from a different point of view, that of decision-analysis methodology. Given the prominent role of comparative effectiveness research in current health care policy, this study is timely in introducing this terminology and methodology to orthopedic surgeons.

The results of this type of complex analysis are highly dependent on the underlying assumptions and data selected for input into the decision analysis. In this study, the sum of data from the Australian joint registry was used with the focus on the key variables of age and gender. Recent data indicates that component size is probably more important than gender in predicting revision rate. In addition, the Australian registry identifies certain components as having a “higher than expected failure rate.” Most of these components were never approved for use in the United States, and two have been recently withdrawn from the marketplace. In addition, results of resurfacing vary widely depending on the surgeons, the center and the device.

There are inherent dangers in using pooled data from registries that include known outliers. It would be interesting to repeat a study such as this excluding implants, centers and possibly even surgeons that are obvious outliers in terms of clinical results including revision rates.

— Robert L. Barrack, MD
Charles and Joanne Knight Distinguished Professor
Chief of Service, Department of Orthopaedic Surgery
Washington University School of Medicine