ObjectivesCritical shoulder angle (CSA) and acromial index (AI) are two radiographic signs that can influence the risk of rotator cuff tears and the outcomes of repair. The purpose of this study was to determine the influence of CSA and AI on massive cuff tears and on the functional outcomes after repair. The hypothesis was that CSA and AI will be higher in posterosuperior compared to anterosuperior tears. MethodsCSA and AI were retrospectively measured on radiographs of patients who underwent repair of two rotator cuff tendons. Functional outcomes were evaluated using American Shoulder and Elbow Surgeons (ASES) score and Simple Shoulder Test (SST) at least six months postoperatively. Patients were divided according to the tendons repaired into anterosuperior group and posterosuperior group. Radiographic measurements and functional outcomes were compared. Patients in the posterosuperior group were subdivided into low or high CSA (cut-off value= 39), and into low or high AI (cut-off value = 0.75). All available preoperative magnetic resonance images were reviewed and graded according to Goutallier classification. Multivariate analysis was used to determine the influence of CSA, AI and Goutallier grade on functional outcomes. ResultsEighty six patients were included. Both CSA and AI were statistically significantly higher in the posterosuperior group (p=0.0143 and 0.0052 respectively). After a mean follow up of 33 months, ASES and SST were significantly better in patients with Goutallier grades 0-1 than grades>1 (multivariate p = 0.03 and 0.009 respectively). No statistically significant differences were found between low and high CSA and AI groups in terms of functional outcomes of the posterosuperior group after repair (multivariate p = 0.9). ConclusionHigher critical shoulder angle and acromial index seem to increase the risk of posterosuperior more than anterosuperior rotator cuff tears. Neither of these radiographic parameters influenced the functional outcomes of massive posterosuperior tears after repair. Level of EvidenceLevel III