Outcomes of revision reverse total shoulder arthroplasty after failed anatomic total shoulder arthroplasty

Reverse total shoulder arthroplasty (rTSA) is the mainstay of treatment after failed anatomic TSA. The purpose of this study was to demonstrate patient-reported outcomes and re-revision rates after rTSA performed for failed anatomic total shoulder arthroplasty.

Between 2002 and 2022, 267 shoulders underwent revision rTSA after failed TSA in this single surgeon institutional database. From this cohort, 188 shoulders (70.41%) had minimum 1-year clinical follow-up and were included in outcomes analysis. Outcomes including American Shoulder and Elbow Surgeons score, Simple Shoulder Test, stability score, visual analog scale pain, visual analog scale function, and patient satisfaction were collected both pre-operatively and post-operatively. Range of motion (ROM) measurements were recorded including forward flexion, abduction, external rotation, and internal rotation. Demographics and surgical factors including implant characteristics, use of bone graft, and indication for revision rTSA were included. Re-revisions were identified, and risk factors were analyzed. Statistical analysis including student's t-test, chi-square analysis, logistic regression, and multivariate stepwise regression was performed.

The average age at revision rTSA was 67.97 years, and average follow-up period was 61.72 months. Glenoid component loosening (40.82%) and rotator cuff failure (37.45%) were the most common indications for revision rTSA. There were improvements in American Shoulder and Elbow Surgeons scores at 1-year follow-up (pre-operative 30.41 ± 16.64, post-operative 66.04 ± 24.93, P < .001). All ROM parameters improved from the pre-operative period to 1-year follow-up, with forward elevation improving from 70° to 129° (P < .001). Abduction improved from 64° to 119° (P < .001). External rotation improved from 27° to 38° (P = .009). Internal rotation, which was graded on a 0-8 scale, improved from 3.06 to 3.92 (P = .005). A total of 28 patients (10.49%) underwent re-revision at a mean time of 33.3 months. Baseplate failure was the most common re-revision indication (n = 14, 5.2%). Use of a larger, load sharing glenosphere was associated with a lower risk of baseplate failure requiring re-revision (2.4% vs. 6.2%, P < .001).

Revision rTSA after failed TSA demonstrates significant improvements in patient-reported outcomes and ROM at an average follow-up period of greater than 5 years. Baseplate failure is the most common reason for re-revision. In this implant family, larger glenospheres were associated with lower baseplate-related re-revision in exploratory analysis. Use of larger glenosphere sizes (40-44 mm) is associated with lower rates of baseplate failure in this population within an implant system where the glenosphere is designed to contact the underlying glenoid bone. This finding should not be generalized to implant systems that do not share this design characteristic.