Clinical and MRI outcomes 10 years after repair of massive posterosuperior rotator cuff tears

International audienceBACKGROUND: Massive rotator cuff tears are challenging to treat, with few or no studies on long-term outcomes of repair. The purpose of this study was to report 10-year outcomes following repair of massive posterosuperior rotator cuff tears, with and without extension into the subscapularis, and to determine prognostic factors that could influence clinical scores and retear rates. METHODS: The records of 234 patients who underwent repair of a massive posterosuperior rotator cuff tear at a total of 15 centers were retrieved. Patients were asked to return for evaluation at 10 years; 78 patients could not be contacted, 7 had died, and 19 had undergone a reoperation. A total of 130 patients (68% men) with a mean age (and standard deviation) of 56.1 ± 7.7 years (range, 26 to 79 years) were evaluated clinically, and 102 of them were also evaluated using magnetic resonance imaging (MRI). The tear was confined to the supraspinatus and infraspinatus tendons in 94 shoulders and also involved the superior portion of the subscapularis in 36 shoulders. Univariable and multivariable regressions were performed to determine whether 10-year total Constant-Murley scores and repair integrity were associated with patient characteristics, tear patterns, or repair techniques. RESULTS: In the study cohort, complications were noted in 14 shoulders (11%) (stiffness in 10 and infection in 4). For the 130 shoulders evaluated clinically, the mean total Constant-Murley score improved from 53.1 ± 15.9 (range, 14 to 83) preoperatively to 78.5 ± 11.3 (range, 36 to 98) at 10 years. Of the 102 shoulders evaluated using MRI, 32 had a retear (Sugaya type IV or V). Of the 19 shoulders that underwent a reoperation (excluded from the study cohort), 9 had a retear. The overall prevalence of retears was 34%. Multivariable regression analysis revealed a significant association between the 10-year Constant-Murley score and preoperative retraction of the infraspinatus tendon, but no association between retears and any of the variables. Involvement of the subscapularis had no significant effect on preoperative or postoperative Constant-Murley scores or retear rates. CONCLUSIONS: Patients who had repair of a massive posterosuperior rotator cuff tear maintained considerable improvements in clinical and radiographic outcomes at 10 years. Partial concomitant tears of the subscapularis did not affect the total postoperative Constant-Murley scores or retear rates