INFLUENCE OF THE ROTATOR CUFF CONDITION O N SHOULDER ARTHROPLASTY RESULTS

Purpose of the study The purpose of this report was to study the reper cussions of rotator cuff condition on unconstrained shoulder arthropla sty results. Methods Between 1986 and 1993, 40 unconstrained shoulder prostheses were performed (24 total and 16 hemiarthroplasties). At the time of the operation, rotator cuff condition was reported to be norm al in 15 shoulders, atrophic in 10, scarred in 3, torn in 12. Clinical and radiographical results were analyzed and supplemented with a rota tor cuff echography. Postoperative follow-up averaged 42.25 months (12 to 97 months). Results When the rotator cuff was intact at the time o f operation, clinical results were the best at the time of review. Con stant score, overall mobility, forward elevation, external rotation wi th the elbow along the side were better for intact than for atrophic, scarred or torn rotator cuffs (p < 0.01). Ruptures repaired by local t issue transfer or trapezo deltoidal flap substitution (3 cases) gave s atisfactory results, whereas two dacron cuff prostheses failed. In the non repaired ruptures (7 cases), arthroplasty resulted in 3/4 fair or poor results. At the time of review, the total incidence of postopera tive rotator cuff tears was high (12 cases), functional repercussions were variable. Among the 15 rotator cuff tears observed at the time of the review, the long head of the biceps was present and in place in 9 cases and ruptured in 6 cases. In these latter cases, Constant score (p < 0.01), the forward elevation (p < 0.01), external rotation with e lbow along side (p < 0.05) and overall mobility (p < 0.05) were worse; moreover, humeral head superior migration was greater (p < 0.01).Disc ussion Our study confirms the superiority of clinical results when the cuff was intact at the time of the operation. The repair of rotator c uff tears seems to be justified. When the rupture has not been repaire d, clinical results were fair or poor in 5 cases out of 7, the initial rupture spread to the other tendons in 3 cases out of 7. Secondary ru ptures were frequent (12 cases). No predictive factor was identified, but an overlapping of the greater tuberosity appears to be detrimental . Among rotator cuff tears present at the time of review, long head of the biceps rupture jeopardized clinical results and was associated wi th a proximal migration of the humeral head which was significantly mo re severe than with an intact non displaced long head of the biceps. C onclusion During unconstrained shoulder prosthesis implantation, it is therefore recommend to systematically repair any associated rotator c uff rupture, in order to avoid any greater tuberosity overlap relative to the prosthetic head and to preserve the long head of the biceps te ndon which limits upward migration of the humeral head and improves pr osthetic kinetics.