PARIS – After total hip or knee replacement, men are at higher risk than women for heart attack, infection, and revision surgery, according to a large database analysis.
The elevated risk was particularly apparent following total knee arthroplasty (TKA), Dr. Gillian Hawker said at the World Congress on Osteoarthritis.
Dr. Gillian Hawker
Prior research suggests that younger patients and males who have joint replacement are at higher risk of certain complications such as early revision, and that joint replacement is underutilized in women.
Based on the current study, underuse of total joint arthroplasty among women versus men does not appear to be explained by higher surgical risk, she said.
The investigators used administrative databases to analyze 97,445 patients who underwent primary, elective total joint arthroplasty in Ontario, Canada, between 2002 and 2009. In all, 59,564 patients (39% male) underwent TKA and 37,881 patients (46% male) had total hip arthroplasty.
Knee arthroplasty
Men had more acute myocardial infarctions within 90 days of undergoing TKA than did women (1.1% vs. 0.8%), more infection within 2 years of surgery (1.1% vs. 0.7%), and more revision TKA within 2 years (1.5% vs. 1%), said Dr. Hawker, professor of medicine at the University of Toronto and physician-in-chief of medicine, Women’s College Hospital, Toronto.
At baseline, the men were similar in age to women, but less likely to be frail (4% vs. 6.7%) and more likely to have a Charlson Comorbidity Index score of 2 or more (5.7% vs. 3.4%).
In Cox regression analysis, the risk for acute MI was significantly increased (hazard ratio, 1.79; P less than .0001) after adjustment for age, sex, income quartile, rurality, frailty, Charlson score, and hospital volume, she said.
Men were also at significantly increased risk for infection (HR, 1.67; P less than .0001) and revision TKA within 2 years (HR, 1.49; P less than .0001), after further adjustment for surgeon volume.
The investigators, led by colleague and orthopedic surgery resident Bheeshma Ravi, Ph.D., postulate that the increased risk of infection and revision among men is due to the high-impact activities that men may engage in after their prosthesis.
No sex differences were found for venous thromboembolism within 90 days or periprosthetic fracture within 2 years.
Hip arthroplasty
Men undergoing total hip replacement had higher rates of early acute MI than women (0.9% vs. 0.7%), but lower rates of periprosthetic fracture within 2 years (0.3% vs. 0.5%), Dr. Hawker said.
At baseline, male patients were younger than their female counterparts (65 years vs. 70 years) and less likely to be frail (3.5% vs. 6.6%), but more likely to have a Charlson score of 2 or more (5.3% vs. 3.7%).
After full adjustment, men were at significantly increased risk for acute MI (HR, 1.64; P less than .0001) and reduced risk for periprosthetic fracture (HR, 0.52; P = .0005), Dr. Hawker said at the meeting, which was sponsored by the Osteoarthritis Research Society International.
"We think the potential explanations for acute MI after both hip and knee replacement may be additional cardiovascular risk factors," she said. "We did not control for preexisting cardiovascular risk; that is something we are doing now."
The study excluded patients who had a pre–joint replacement fragility fracture, but because of the quality of the data postoperatively and the availability of drug benefit data, "we don’t feel we have adequate control for the presence of osteoporosis," Dr. Hawker observed.
No differences were found between sexes in the hip replacement cohort for infection, death, venous thromboembolism, dislocation, or revision.
Session comoderator Dr. Martin Englund, of Lund (Sweden) University, commented that the study was well conducted and the findings very interesting, but he cautioned against generalizing the results too broadly.
"We have seen before, that results from Sweden are not necessarily the same as in North America," he said in an interview. "I’m sure these results are probably very generalizable in Canada and in that type of health care setting, but may not be the same elsewhere. ... These are also things that might change over time. So we just need to keep monitoring outcomes and repeating these studies and adjust our treatment to what we find."
Dr. Ravi reported fellowship support from the University of Toronto, Canadian Institutes of Health Research, and Women’s College Research Institute.