BIRMINGHAM, ENGLAND – Contrary to expectations, metal-on-metal hip resurfacing for osteoarthritis was associated with higher patient survival at 10 years than was total hip arthroplasty in a large, population-based study.
Cumulative mortality rates were 2.8% for hip resurfacing versus 7.3% for cemented total hip replacement (THR; hazard ratio, 0.51). Ten-year mortality rates comparing hip resurfacing to uncemented THR were 2.6% and 3.2%, respectively (HR, 0.64).
Furthermore, the number needed to treat with hip resurfacing to prevent 1 excess death was 29 when compared to cemented THR, and it was 88 when compared to uncemented THR.
"Patients who received a metal-on-metal resurfacing [MoMR] procedure seem to have a long-term survival advantage compared to patients receiving cemented or an uncemented THR," said Dr. Adrian Kendal of the National Institute for Health Research (NIHR) Musculoskeletal Biomedical Research Unit at the University of Oxford, England.
"Our findings were robust after adjustment for known confounders," Dr. Kendal said at the British Society for Rheumatology annual conference. Propensity matching was used in the trial, which took age, gender, comorbidity, rurality, and social deprivation into account.
For the study, data from the English Hospital Episode Statistics database were obtained and linked to Office for National Statistics mortality records for all adults (over age 18) undergoing elective primary hip replacement for osteoarthritis in National Health Service hospitals in England and Wales between April 1999 and March 2012.
After propensity score matching, there were 91,633 procedures performed, of which 12,580 were MoMR, 37,740 were cemented THR, and 41,312 were uncemented THR.
In response to a comment that perhaps people opting for MoMR were more likely to be younger, more active, and hence more likely to exercise, Dr. Kendal conceded that other factors might exist that could have affected survival.
Speculating about why there might be such a difference in survival, he said: "I personally don’t think it’s just the use of cement, because that doesn’t explain the group that received an uncemented total hip replacement."
He added that the way the femur is prepared during THR might be important, regardless of whether or not cement is used. The known risk of thrombotic consequences also could affect survival. In addition, health care inequality might be important, as resurfacing procedures are less common than THR, perhaps because of the lack of specialized centers or dedicated teams.
Commenting on the findings after their presentation, consultant rheumatologist Dr. Alex MacGregor, of the University of East Anglia, Norwich, England, noted that similar data were published on this topic last year (BMJ 2012;344:e3319), but the results had proved somewhat controversial as the authors had a conflict of interest in favor of hip resurfacing.
Dr. MacGregor, who is a member of the National Joint Registry Steering Committee, has been involved in a subsequent reanalysis of the paper’s findings and said that the results will be made public later in the year.
"One of my concerns [with this study] is the use of the 10-year mortality endpoint. If these resurfacing procedures are saving lives, then you would expect to see a survival benefit sooner, say at 90 days," Dr. MacGregor said.
Dr. Kendal responded that they tried to account for this, but the answer will need to come from a properly organized, randomized controlled trial.
"We don’t have a conflict of interest here. If anything, we were perhaps looking for the opposite effect; we were expecting to see an increased mortality rate in the resurfacing group," Dr. Kendal said. "That was not the case as it turned out, so I am reasonably confident that our data support the findings of that BMJ article."
Dr. Kendal and Dr. MacGregor reported no conflicts of interest.