Clinical and radiological outcomes of osteoarthritis twenty years after rotator cuff repair.

ABSTRACT

PURPOSE: Outcomes of open or arthroscopic rotator cuff repairs are well reported in the literature. The purpose of the study was to evaluate the prevalence and clinical impact of osteoarthritis 20 years following rotator cuff repair. The hypothesis was that, at long follow-up, most shoulders would have developed gleno-humeral osteoarthritis. METHODS: The authors retrospectively recalled all 322 patients, operated for rotator cuff tears in 1994 at 6 different centres, for clinical and radiographic assessment. At 20 years of follow-up, 24 were re-operated (5 arthroplasty) and 53.4% were lost to follow-up. This left 126 patients, had been clinically assessed, had Magnetic Resonance Images (MRI) that allowed anatomic assessment of tendon healing (Sugaya), fatty infiltration (Goutallier), and X rays in order to analyse arthritis without head migration (Samilson) and with head migration (Hamada and Fukuda). Only patients with complete data were selected. RESULTS: Mean aged was 52.3 years (25.3-68.6) at index operation. The Constant score was 45.3±19.6 preoperatively to 67.4±18.7 points at 20 years. The SSV was 73.5±21 postoperatively. The rate of osteoarthritis was 29%. Osteoarthritis was associated with a significant inferior Constant score compared to the non-arthritic group (61 versus 71 points, p=0.02), mainly due to a significant lower strength (5.4 versus 8.7 points, p=0.007). Massive rotator cuff tears were significantly associated with a higher rate of osteoarthritis. Only 4,8% patients after cuff repear needed a reverse shoulder arthroplasty. Significantly less osteoarthritis was observed when the rotator cuff repair remained intact. Suprasupinatus retear had a significant influence on fatty infiltration of the infrasupinatus muscle and on the progression towards osteoarthritis. CONCLUSION: Twenty years after open rotator cuff repair, the rate of osteoarthritis was 29%. Massive rotator cuff tears were significantly associated with a higher rate of osteoarthritis. Less osteoarthritis was observed when suprasupinatus healed. LEVEL OF EVIDENCE IV.