By John Gever, Senior Editor, MedPage Today
Published: February 18, 2013
Reviewed by Robert Jasmer, MD; Associate Clinical Professor of Medicine, University of California, San Francisco
Failure rates for hip implants were 29% higher for women than men in a large U.S. registry study after controlling for a variety of factors including device type, researchers said.
With a total of 35,140 patients undergoing primary total hip arthroplasty followed for a median of 3 years, the crude all-cause rate of failure (defined as subsequent revision surgery) was 2.3% for women (95% CI 2.1% to 2.5%) compared with 1.9% for men (95% CI 1.6% to 2.1%), according to Maria C.S. Inacio, MS, of the Southern California Permanente Research Group in San Diego, and colleagues.
After adjustments for age, body mass index, diabetes status, degree of presurgical symptom severity, implant fixation method, device category, and femoral head size, the authors calculated a hazard ratio (HR) for revision of 1.29 for women versus men (95% CI 1.11 to 1.51), they reported online in JAMA Internal Medicine.
The risk appeared most prominent for aseptic revision (HR 1.32, 95% CI 1.10 to 1.58) compared with septic failure (HR 1.17, 95% CI 0.81 to 1.68), the researchers found.
Larger femoral head sizes appeared especially problematic for women. For head sizes of 36 mm or more, the adjusted HR for failure in women versus men was 1.49 (95% CI 1.14 to 1.95), whereas differences in revision rates for smaller head sizes were not significant after adjustment.
Much of the increased risk for women also seemed concentrated in metal-on-metal implants, with a doubling in risk for women versus men (adjusted HR 1.97, 95% CI 1.29 to 3.00).
But that was primarily because of reduced risk of failure with metal-on-metal devices in men (adjusted HR 0.68 versus highly crosslinked polyethylene, 95% CI 0.45 to 1.02), whereas in women, the adjusted failure rates for metal-on-metal versus crosslinked polyethylene were similar, adjusted HR 1.07, 95% CI 0.72 to 1.60).
In an accompanying commentary, Diana Zuckerman, PhD, of the National Research Center for Women and Families in Washington, D.C., suggested that the study's clinical implications were relatively trivial.
She noted that most patients considering hip replacement are already suffering pain and limited mobility and have few other options.
"Knowing that their chances of success are lower than men's is not helpful to women who are unable to perform many activities of daily living," Zuckerman argued.
Instead, she said, what is needed is "long-term comparative effectiveness research based on large sample sizes, indicating which total hip arthroplasty devices are less likely to fail in women and in men, with subgroup analyses based on age and other key patient traits, as well as key surgeon and hospital factors."
But Glenn Don Wera, MD, of UH Case Medical Center in Cleveland, told MedPage Today that the study provided valuable insights into the reasons for higher failure rates in women.
He noted that the higher rate of revision in women was already known from Medicare data. In the current study, however, "they were able to control for a number of clinical factors, including the kind of prosthesis the patient had, the experience level of the surgeon, and the different institutions and the different prostheses they were using."
That the increased risk in women was still evident despite adjusting for those factors indicates that something else, such as women's generally smaller stature, is responsible, Wera suggested.
Revision Rate: Beyond Infection
Data for the current study came from the Kaiser Permanente system's registry of total joint replacements from 2001 to 2010. Procedures were performed at 46 hospitals in California, Hawaii, Oregon, Washington, and Colorado by 319 different surgeons.
The registry is the largest of its type in the U.S., the authors said and includes data on surgeons' and hospitals' arthroplasty procedure volumes; the patients and the implants they received (cemented, uncemented, or hybrid); and implant bearing surface, such as metal on metal, metal or ceramic on highly crosslinked polyethylene, or ceramic on ceramic. The researchers put the DePuy metal-on-metal hip resurfacing monoblock device into its own category.
Only patients undergoing unilateral procedures were included in the analysis.
About 58% of the 35,140 procedures were performed in women (mean age 65.7 versus 63.8 for men). Just over 60% of both sexes had scores of 1 or 2 on the American Society of Anesthesiologists index, with nearly all of the remainder having scores of 3 or more.
The age difference between men and women was statistically significant. In addition, women in the cohort tended to be slightly more likely to be white or Asian and to have osteoarthritis, rheumatoid arthritis, or dysplasia. They were less likely to be diabetic or obese and to have osteonecrosis or post-traumatic arthritis.
Not surprisingly, women were much less likely to have implant femoral head sizes of 36 mm or more (32.8% versus 55.4% for men, P<0.001). About twice as many men as women had metal-on-metal bearings (19.4% versus 9.6%), whereas ceramic or metal on highly crosslinked polyethylene were more popular for female patients (P<0.001).
The DePuy resurfacing implant was used in 1.3% of women versus 2.6% of men (P<0.001).
Preferences for fixation types also differed between men and women, with hybrid methods more common in the women and cementless fixation more common in men.
Mean surgeon and hospital volumes did not differ between sexes.
The authors noted that with no significant increase in risk of septic failure for women, their results mean that "factors other than infection" are responsible for the higher overall revision rate.
Limitations to the analysis include its observational design, the relatively short follow-up period, lack of data on some potential confounding factors, and the use of revision surgery as the definition of implant failure. Also, the researchers used relatively broad categories of implant type, conceding that design variations within these categories could have influenced the results.
The study was funded by the FDA.
Study authors and Zuckerman declared no relationships with commercial entities. Several study authors were Kaiser Permanente employees.
Primary source: JAMA Internal Medicine Source reference: Inacio M, et al "Sex and risk of hip implant failure: Assessing total hip arthroplasty outcomes in the United States" JAMA Intern Med 2013; DOI: 10.1001/jamainternmed.2013.3271.
Additional source: JAMA Internal Medicine Source reference: Zuckerman D "Hip implant failure for men and women: What and when we need to know" JAMA Intern Med 2013; DOI: 10.1001/jamainternmed.2013.19.