Objective: Therapeutic intraarticular injections are used in the management of hip osteoarthritis (OA). Some studies suggest that their use increases the risk of infection and subsequent revision surgery after primary total hip arthroplasty (THA), while others do not. We undertook this study to clarify the relationship between prior intraarticular injection and the risk of complication in a subsequent primary THA.
Methods: In a cohort of patients with hip OA who underwent a primary elective THA between 2002 and 2009, we identified those who received ≥1 intraarticular injection performed by a radiologist in the 5 years preceding their THA. Multivariable Cox proportional hazards models were used to determine the relationship between receipt of a presurgical injection (no injection, 1-5 years prior to THA, or <1 year prior to THA) and the occurrence of postsurgical joint infection and revision THA in the following 2 years, while controlling for confounders.
Results: Of 37,881 eligible THA recipients, 2,468 (6.5%) received an intraarticular injection performed by a radiologist within 5 years of their THA (1,691 at <1 year, 777 at 1-5 years). Controlling for age, sex, comorbidity, frailty, income, and provider volume, those who had an injection in the year preceding surgery were at increased risk of infection (adjusted hazard ratio [HR] 1.37, P = 0.03) and revision THA (adjusted HR 1.53, P = 0.03) within 2 years of the primary THA, relative to patients who did not. The association between prior injection and revision arthroplasty was attenuated and became nonsignificant (adjusted HR 1.41, P = 0.13) after occurrence of postoperative infection was controlled for in the regression model. No effect was found for injection 1-5 years prior to surgery.
Conclusion: Intraarticular injection in the year preceding THA independently predicted increased risk of infection leading to early revision surgery. Further studies are warranted to elucidate explanations for these findings.
Copyright © 2015 by the American College of Rheumatology.