Are Corticosteroid Injections Too Risky for Knee Arthroscopy Patients?

    A recently published study of Medicare patients undergoing knee arthroscopy found that those who received an intraoperative corticosteroid injection were at higher risk for infection. Can this finding be generalized to the typical younger, healthier knee arthroscopy patients?


    Mehul R. Shah, MD


    Cancienne JM, Gwathmey FW, Werner BC. Intraoperative corticosteroid injection at the time of knee arthroscopy is associated with increased postoperative infection rates in a large Medicare population. Arthroscopy. 2016 Jan;32(1):90-5. PMID: 26553960


    The authors of this article investigated the safety of utilizing intraoperative corticosteroid injections during knee arthroscopy.

    The retrospective study examines a cohort of patients from a national Medicare insurance database who underwent knee arthroscopy, with or without a corticosteroid injection, from 2005 to 2012.

    The researchers identified 2,866 patients who underwent knee arthroscopic procedures and a concomitant corticosteroid injection by searching for a large joint injection CPT code (20610) and a “J” code for corticosteroid medication.

    A matched control group of 170,350 patients was created from those who underwent the same arthroscopic knee procedures but without a concomitant corticosteroid injection.

    The incidence of postoperative infection was measured at 3 and 6 months after surgery and was found to be higher in the corticosteroid injection group at both time periods: 0.66% for the injection group vs. 0.25% for the control group at 3 months (odds ratio 2.6) and 1.92% vs. 0.54%, respectively, at 6 months (odds ratio 3.6). These findings were statistically significant at P < 0.0001 for each time period.

    Clinical Relevance

    Knee arthroscopy is one of the most common orthopaedic procedures performed in the US. Patients who have concurrent degenerative changes are likely to have poorer outcomes, and given this, many surgeons utilize intraoperative corticosteroids in arthritic knees to decrease postoperative pain, opioid analgesic use, and recovery time. The risk of infection after knee arthroscopy, fortunately, remains a rare occurrence.

    Prior studies on corticosteroids and infection risk have been underpowered, with conflicting results. Many of these studies have proposed that the risk of infection increases with corticosteroid administration due to the local and systemic immunosuppression.

    More recent studies have reported good pain control after corticosteroid injection, without infection. One must use caution in interpreting these results, however, as the studies were likely underpowered to assess infection risk given its low incidence.

    The present study is not without its limitations. It uses a retrospective cohort and, therefore, is susceptible to selection bias. Compared with patients in the control group, patients who received corticosteroid injections may have had worse disease and underwent more extensive surgery, which led to the corticosteroid administration. These confounding factors may have contributed to the increased infection risk.

    Additionally, the database utilized does not represent the general population of knee arthroscopy patients, who are typically younger and healthier than the Medicare population. Therefore, the risk of infection may not be as severe as indicated in this study.

    Given the results of this study, corticosteroid injections during knee arthroscopy should be done with extreme caution. Though these injections may provide significant pain relief, there is now statistically significant data showing they increase infection risk.

    Author Information

    Mehul R. Shah, MD is an Assistant Professor of Orthopaedic Surgery, Division of Sports Medicine, at NYU Langone Medical Center – Hospital for Joint Diseases, New York, NY