Skip to main content

20-05-2019 | Rheumatology | News | Article

Editor's pick

Glucocorticoid use may increase infection risk in RA patients undergoing joint replacement

medwireNews: Analysis of two US administrative databases indicates that glucocorticoid use is associated with a dose-dependent increase in postoperative infection risk in patients with rheumatoid arthritis (RA) undergoing elective total knee or hip replacement.

However, the risk for infection was similar among people treated with different biologics, report the researchers in the Annals of Internal Medicine.

Michael George (University of Pennsylvania, Philadelphia, USA) and colleagues used the Medicare and Truven MarketScan databases to analyze infection rates among 9911 biologic-treated patients undergoing 10,923 procedures between 2006 and 2015. The most commonly used biologic agent was infliximab (31.0%), followed by etanercept (27.0%), adalimumab (18.3%), and abatacept (16.1%), and 43.0% of patients also received glucocorticoids within 90 days prior to surgery.

Serious infections resulting in hospitalization occurred within 30 days of knee or hip replacement in 9.0% of 7929 procedures in the Medicare dataset and 4.7% of 2994 procedures in MarketScan, most frequently urinary tract infection (4.1% overall), skin and soft tissue infection (1.1%), and pneumonia (0.9%), while prosthetic joint infections (PJI) occurred within 1 year of 2.0–2.6% of operations.

George and team found that glucocorticoid use was associated with a dose-dependent increase in the risk for serious infection, PJI, non-urinary tract serious infection, and hospital readmission within 30 days in both databases. For instance, the 298 patients receiving glucocorticoids at a dose of 5–10 mg/day had a 1.32-fold increased risk for hospitalized infection relative to the 1834 who did not receive glucocorticoids in propensity-weighted analyses, while the 118 patients taking glucocorticoids at a daily dose above 10 mg had a 2.10-fold increased risk.

These findings indicate that “[m]inimizing glucocorticoid exposure before surgery should be a primary focus of perioperative medication management,” write the researchers.

However, “[o]utcomes were similar in patients who received different biologics,” they report. Propensity-weighted risk for serious infection, PJI, non-urinary tract serious infection, and readmission were not significantly different among patients treated with different biologics, with the exception of a significantly lower PJI rate among patients treated with rituximab versus abatacept in the Medicare database. Nevertheless, the investigators note that they “could not precisely assess PJI risk” due to the small number of patients receiving rituximab (n=423 overall).

The researchers emphasize that patients receiving methotrexate alongside a biologic, representing 45.6% of the study population, did not have an elevated risk for postoperative infection, although they were not able to determine whether the drug was withdrawn preoperatively.

Writing in an accompanying editorial, Bheeshma Ravi and Gillian Hawker, both from the University of Toronto in Ontario, Canada, say that the study provides “convincing evidence to support lack of variability in postoperative infection risk across the available biologic therapies,” and “should also be reassuring regarding the lack of evidence for increased infection risk with methotrexate–biologic cotherapy.”

They note, however, that “the study does not resolve the question of whether withholding biologic therapies in the perioperative period actually reduces patients’ overall risk for infection complications,” and did not provide a stratified analysis by joint.

“[W]hether infection risk differs for recipients of hip versus knee replacement receiving biologic therapies remains unclear,” the editorialists conclude.

By Claire Barnard

medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group

Ann Intern Med 2019; doi:10.7326/M18-2217
Ann Intern Med 2019; doi:10.7326/M19-1088

See also: