Periprosthetic joint infection (PJI) is one of the most devastating complications following total hip arthroplasty. The purposes of this study were to determine risk factors for PJI after primary total hip arthroplasty for osteoarthritis using a Canadian population-based database collected over 15 years and to determine the incidence of PJI, the time to PJI following primary total hip arthroplasty, and whether the PJI rate had changed over 15 years.
We performed a population-based cohort study using linked administrative databases in Ontario. We included all primary total hip arthroplasties performed for osteoarthritis in patients who were ≥55 years of age. We used a Cox proportional hazards model to analyze the effect of surgical and patient factors on the risk of developing PJI. We calculated 1, 2, 5, and 10-year PJI rates. We used the Cochran-Armitage test to assess the evidence of trends in PJI rates over time.
A total of 100,674 patients who were ≥55 years of age underwent a primary total hip arthroplasty for osteoarthritis. The cumulative incidence for PJI at 15 years was 1.44% (95% confidence interval [CI], 1.38% to 1.50%). Risk factors associated with the development of PJI include male sex (hazard ratio [HR], 1.43 [95% CI, 1.30 to 1.51]), type-2 diabetes mellitus (HR, 1.51 [95% CI, 1.31 to 1.70]), and being discharged to convalescent care (HR, 1.36 [95% CI, 1.05 to 1.77]). Sixty-two percent of PJI cases occurred within 2 years after the surgical procedure and 98% occurred within 10 years. The rate of PJI following primary total hip arthroplasty did not change over the 15 years of our study period.
The risk of developing PJI following primary total hip arthroplasty did not change in 15 years, despite improvements in other arthroplasty outcomes. Male sex, type-2 diabetes mellitus, and discharge to convalescent care were associated with an increased risk of PJI. The surgical approach, income quintile, and use of bone-grafting or cement were not significantly associated with increased risk of infection in our cohort.
Level of Evidence:
Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.