Preoperative Corticosteroid Injections Within 4 Weeks of Arthroscopic Shoulder Procedures Are Associated With Increased Postoperative Infection Rates

Abstract

Purpose: To refine the understanding of the effect of timing of corticosteroid injections (CSIs) and shoulder arthroscopy on postoperative infection.

Methods: An insurance database was used to determine all patients who underwent shoulder arthroscopy for a 5-year period with an associated preoperative ipsilateral corticosteroid injection. Patients were stratified into cohorts based on timing of preoperative CSI: (1) 0-<2 weeks, (2) 2-<4 weeks, (3) 4-<6 weeks, and (4) 6-<8 weeks. Patients were pooled to include all patients who had a CSI less than 4 weeks and those longer than 4 weeks. A cohort of patients who never had a corticosteroid injection before undergoing arthroscopy were used as a control. All patients had a follow-up of 2 years. Multivariable regression analyses were performed using R Studio with significance defined as P < .05.

Results: Multivariate logistic regression showed a greater odds ratio (OR) for postoperative infection in patients who received CSI 0-<2 weeks before shoulder arthroscopy at 90 days (3.10, 95% confidence interval [CI] 1.62-5.57, P < .001), 1 year (2.51, 95% CI 1.46-4.12, P < .001), and 2 years (2.08, 95% CI 1.27-3.28, P = .002) compared with the control group. Patients who received CSI 2-<4 weeks before shoulder arthroscopy had greater OR for infection at 90 days (2.26, 95% CI 1.28-3.83, P = .03), 1 year (1.82, 95% CI 1.13-2,82, P = .01), and 2 years (1.62, 95% CI 1.10-2.47, P = .012). Patients who received CSI after 4 weeks had similar ORs of infection at 90 days (OR 1.15, 95% CI 0.78-1.69, P = .48) 1 year (OR 1.18, 95% CI 0.85-1.63 P = .33), and 2 years (OR 1.09, 95% CI 0.83-1.42, P = .54), compared with the control cohort.

Conclusions: The present study shows the postoperative infection risk is greatest when CSIs are given within 2 weeks of shoulder arthroscopy, whereas CSIs given within 2-<4 weeks also portend increased risk, albeit to a lesser degree. The risk of postoperative infection is not significantly increased when CSIs are given more than 1 month before surgery.

Level of evidence: Level III, retrospective comparative, prognosis study.