[Shoulder arthroplasty for osteoarthritis after prior surgery for anterior instability: a report of 27 cases]

The purpose of this study was to analyze the natural history of shoulder osteoarthritis secondary to prior surgery for anterior instability, to evaluate clinical and radiological results of shoulder prosthesis, and to compare results with reports in the literature.

Twenty-seven shoulder prostheses (Aequalis) were implanted in patients with osteoarthritis of the shoulder after prior instability surgery. These shoulders were reviewed retrospectively at a mean follow-up of 46 months (range 24-48). Prior surgery had been performed with coracoid block in two-thirds of the shoulders and with soft-tissue procedures in one-third. Twenty-four shoulders had had one instability procedure, several procedures had been performed in three. The Constant score was used to assess clinical status preoperatively and at last follow-up. Male gender predominated (16/27), and mean age at first dislocation was 31.6 years (4 patients experienced their first dislocation after the age of 60 years). The patients had a mean 18.2 dislocations. The natural history of osteoarthritis was long: 24.2 years on the average. Mean age at arthroplasty was 55.8 years. The preoperative images demonstrated an iatrogenic factor in 5 shoulders and constructive osteoarthitis in 24. Posterior glenoid wear was observed in 24% of the cases, fatty degeneration of the subscapularis in 45% and full-thickness tears of the supraspinatus in 4 cases. The anterior scar tissue was released and implants were positioned anatomically in 26 cases. Total arthroplasty was used for 21 shoulders and a humeral prosthesis for 6.

There were five complications including three anterior prosthetic instability, all three in patients who were over 60 years of age at their first instability surgery (one of these three patients required revision arthroplasty). Outcome was good or excellent in 56% of the shoulder with a mean weighted Constant score of 83%: mean gain in pain score=8.3 points, in anterior elevation=51 degrees, in external rotation=33.5 degrees. The type of prior instability surgery and preoperative active external rotation had no determining effect on outcome. Factors correlated significantly with outcome were fatty degeneration of the rotator cuff muscles, particularly the subscapularis.

Osteoarthritis of the shoulder has a long natural history, 26 years in our patients excepting those whose first dislocation occurred after the age of 60 years. Posterior glenoid wear, described by others, was only found in 24% of the shoulders in this series. Surgical dissection and release of anterior scar tissue was difficult. The overall results of arthroplasty were good and were correlated with fatty degeneration of the rotator cuff muscles, particularly the subscapularis, but not with the type of instability surgery. These results were less satisfactory than reported in series of arthroplasty for primary centred osteoarthritis: complications occurred in 18% of the shoulders with three cases of anterior dislocation. Unlike reports of arthroplasty after instability surgery where many of the patients had several operations, overall results in our series were not compromised by a large number of multiple interventions.