Patients and Methods

We conducted a retrospective review of the pre-operative MR imaging and surgical records of 87 patients (87 shoulders) who underwent arthroscopic repair of a large-to-massive rotator cuff tear. Patients were divided into two groups, based on the surgical outcome of the repair. Of the 87 patients, 53 underwent complete repair (Group I) and 34 an incomplete repair (Group II). Pre-operative MR images were reviewed to quantify several variables. Between-group differences were evaluated and multiple logistic regression analysis was used to calculate the predictive value of significant variables. The reparability index (RI) was constructed using the odds ratios of significant variables and a receiver operating characteristic curve analysis performed to identify the optimal RI cutoff to differentiate between the two groups.