[Shoulder arthroplasty for non-operated anterior shoulder instability with secondary osteoarthritis]

The purpose of this study was to analyze the natural history of osteoarthritis of the shoulder joint secondary to non-operated anterior instability, to evaluate the clinical and radiological outcome after arthroplasty for this indication, and to compare results with data reported in the literature.

Twenty-eight Aequalis shoulder arthroplasties were performed for osteoarthritis secondary to anterior shoulder instability in patients who had not undergone any prior stabilizing surgical procedure. This multicentric series was reviewed retrospectively at a mean follow-up of 44 months (range 24-87 month). Preoperative and last follow-up clinical status was compared using the Constant score. Standard operative technique was used and anatomic position of the implants was achieved in 27 patients. Total shoulder arthroplasty was performed in eighteen patients and simple humeral implantation in ten. Patients were divided into two sub-groups for analysis of outcome (age at first dislocation > or < 40 years).

There were four complications including one early loosening that required revision arthroplasty. Outcome was excellent or good in 54% of the cases. Group 1 (age at first dislocation < 40 years) included 17 patients and exhibited male predominance (13/17), young age at first dislocation (mean 28.5 years), high number of dislocations (mean 13), and a long natural history of osteoarthritis (mean 28 years). In this sub-group, imaging disclosed constructive osteoarthritis reflecting a long history. There was one case with a full-thickness tear of the supraspinatus. Outcome was excellent or good in 64% of these 17 patients (14 underwent total shoulder arthroplasty). The sub-group of 11 patients whose first dislocation occurred after the age of 40 years exhibited female predominance (9/11), a small number of dislocations (mean 1.3), and a short natural history of osteoarthritis (mean 6 years). In this sub-group, imaging disclosed six cases of basically chondrolytic osteoarthritis reflecting a short evolution. There were seven cases of full-thickness tears of the supraspinatus. Outcome was excellent or good in 36% of these 11 patients (4 underwent total shoulder arthroplasty).

The forty-year age cutoff allowed us to distinguish two populations. The first population of patients whose first dislocation had occurred before the age of 40 years was predominantly male, with a long history of constructive osteoarthritis without rotator cuff tears. Most of these patients were treated with total shoulder arthroplasty which gave results similar to those obtained with arthroplasty for primary centered osteoarthritis. The second sub-group of patients aged over 40 years at the time of the first dislocation were predominantly female with a short history of minimally constructive osteoarthritis and frequent rotator cuff tears. Humeral implants were used for most of these patients and gave less favorable results than in the former sub-group, probably due to the high rate of rotator cuff tears.