Associations have been reported linking rotator cuff tears (RCTs) to both greater lateral extension of the acromion and greater inclination of the glenoid cavity. These two factors combined can be assessed using a recently introduced parameter, the critical shoulder angle (CSA). The primary objective of this study was to confirm the association linking a high CSA value to RCTs, and the secondary objective was to assess the reproducibility of CSA measurement using a goniometer.

The null hypothesis was that the CSA value in a group of patients with RCTs was not significantly different from that in patients with anterior shoulder instability and a Bankart lesion, taken as the general population for this study.

After a power estimation, we retrospectively included 28 patients with a mean age of 55.5 years who had surgery for RCTs and 27 patients with a mean age of 27.2 years who underwent anterior labral repair. Two surgeons used a goniometer to measure the CSA in each patient. Reproducibility was assessed based on Bland-Altman plots and Pearson's correlation coefficient.

The mean CSA was significantly higher (P=0.02) in the RCT group (36.4°±4.4°; range: 30°–46°) than in the labral-repair group (33.3°±3.8°; range: 25°–41°). Intra-observer reproducibility was 96.7% and inter-observer reproducibility was 95.5%.

Our results support previously published evidence that the CSA is significantly greater in patients with RCTs. Thus, an anatomical difference seems to exist between patients with RCTs and the general population. The CSA measured on a standard radiograph using a goniometer provides a reproducible assessment of this anatomical difference.

IV, case-control epidemiological study with a power estimation.