Judicious use of corticosteroid injections prior to shoulder arthroplasty does not compromise outcomes at a minimum of 2 years following surgery

The use of total shoulder arthroplasty is continuing to rise with its expanding indications. For patients with chronic conditions, such as glenohumeral arthritis and rotator cuff arthropathy, nonoperative treatment is typically done prior to arthroplasty and often includes corticosteroid injections (CSIs). Recent studies in the shoulder arthroplasty literature as well as applied from the hip and knee literature have focused on the risk of periprosthetic infection. Literature is lacking as to whether the judicious use of corticosteroids in the year prior to arthroplasty influences patient-reported outcomes (PROs). The purpose of this study was to determine if preoperative CSIs prior to shoulder arthroplasty affected 2-year PROs. Retrospective review of anatomic and reverse total shoulder arthroplasty (RSA) patients (n = 230) was performed at a single institution including multiple surgeons. Patients were included if they had preoperative and a minimum of 2-year postoperative PROs, including: American Shoulder and Elbow Surgeons (ASES), visual analog scale, Single Assessment Numeric Evaluation, Veteran's RAND 12 Physical Component Score, and Veteran's RAND 12 Mental Component Score. Patients were included in the injection group if they had received an injection, either glenohumeral or subacromial, within 12 months prior to arthroplasty (inject = 134). Subgroup analysis included anatomic (total shoulder arthroplasty [TSA] = 92) and RSA (RSA = 138) as well as those with no injection within 12 months prior to surgery. An analysis of variance was used to compare outcomes between patients who received an injection and those who did not prior to TSA and RSA. There were 230 patients included with 134 patients in the injection group and 96 in the no injection group. Patients who received an injection in the year prior to arthroplasty displayed a significantly higher ASES (82 [16.23 standard deviation] vs. 76 [19.43 standard deviation], P