Corticosteroid injection prior to surgery had no effect on 2-year outcomes following arthroscopic rotator cuff repair

Corticosteroid injections (CSIs) can be an effective nonsurgical treatment for patients with rotator cuff tears. Recent large database studies have raised concern that CSI may result in a higher reoperation rate, increased infection risk, and worse outcome after arthroscopic rotator cuff repair (ARCR). The purpose of this study was to evaluate the reoperation rate, incidence of postoperative infection, and two-year outcomes of patients undergoing ARCR with and without the use of preoperative CSI.

An institutional database generated from fellowship-trained orthopedic sports surgeons was retrospectively queried for patients who underwent ARCR with a minimum of two-year follow-up. Inclusion criteria consisted of 1) primary full-thickness rotator cuff tear and 2) preoperative and minimum two-year patient-reported outcome measures (PROMs). Of the 219 patients identified, 134 patients had preoperative subacromial CSI administered within one year of ARCR. Reoperation rate, number of injections, Visual Analog Scale, American Shoulder and Elbow Surgeons Score, Single Assessment Numeric Evaluation, and Veterans Rand 12-Item Health Survey Physical Component Score/Mental Component Score were compared between groups at six months, one year, and two years. Chi-square and t-tests were used to compare baseline differences, postoperative infections, and reoperations. A repeated measures Analyses of Covariance was used to measure differences between PROMs at each time point. Simple Analyses of Covariance were used for the two-year sub-analyses for patients receiving CSI within 90 days of surgery and if multiple preoperative CSI had been given (α ≤ 0.05).

There were no significant demographic differences between groups (P > .05). Preoperative use of subacromial CSI within one year prior to ARCR did not increase reoperation rate (P = .85) or impact PROMs at any timepoint. There were two reoperations during the study period in the CSI group (2 lysis of adhesions). No infections occurred in either cohort. No differences were found if injections were performed within 90 days of surgery or if more than one CSI was administered within the year prior to surgery (P > .05).

Our results show that preoperative CSI prior to primary ARCR did not increase risk of reoperation, infection, or influence PROMs with a minimum follow-up of 2 years.