Arthroscopic glenoid removal for aseptic loosening in total shoulder arthroplasty

Background Glenoid loosening is the most common long-term complication of total shoulder arthroplasty (TSA) and frequently necessitates revision. Though arthroscopic glenoid removal is an accepted treatment option for glenoid loosening, there is a paucity of outcomes literature available. The purpose of this study was to report the long-term clinical and radiographic outcomes of arthroscopic glenoid removal for failed or loosened glenoid component in TSA. We hypothesized that arthroscopic glenoid removal would produce acceptable clinical and patient-reported outcomes while limiting the need for further revisions. Methods This was a retrospective analysis of 11 consecutive patients undergoing 12 arthroscopic glenoid removals for symptomatic glenoid loosening by a single orthopedic surgeon between March 2005 and March 2018. Indication for arthroscopic glenoid removal included symptomatic glenoid loosening with radiographic evidence of a 1-2 mm radiolucent line around the glenoid. Shoulder range of motion, functionality (American Shoulder and Elbow Surgeons, Simple Shoulder Test), and pain (visual analog scale [VAS]) were evaluated. Radiographs were assessed for glenohumeral subluxation, humeral superior migration, and glenohumeral offset following glenoid removal. Results The mean follow-up period since arthroscopic glenoid removal was 55 months (range, 20-172 months). Glenoid component removal significantly reduced forward elevation, with a mean decrease from 147 ± 13° preoperatively to 127 ± 29° postoperatively (P= .031). However, there was no significant change in external rotation (44 ± 9° vs. 43 ± 19°; P= .941) or internal rotation (L4 vs. L4; P= .768). Importantly, glenoid removal significantly decreased VAS pain scores from 7 ± 3 preoperatively to 5 ± 3 postoperatively (P= .037). Additionally, improvement in ASES approached statistical significance, increasing from 33 ± 25 preoperatively to 53 ± 28 postoperatively (P= .055). With regard to radiographic outcomes, there was no evidence of glenohumeral subluxation and humeral superior migration developed in 1 patient. However, there was significant medialization of the greater tuberosity relative to the acromion, with a mean lateral offset of 6 ± 7 mm preoperatively and −2 ± 4 mm postoperatively (P= .002). Two patients required conversion to reverse TSA for persistent pain. There were no complications. Discussion These findings suggest that arthroscopic glenoid removal for symptomatic glenoid loosening is a viable option to improve pain while limiting the need for additional reoperations and decreasing the risks associated with revision arthroplasty. However, continual follow-up to monitor medialization is recommended. Level of Evidence Level IV; Case Series; Treatment Study