The critical shoulder angle is associated with rotator cuff tears and shoulder osteoarthritis and is better assessed with radiographs over MRI

PURPOSE: The purpose of this study was to determine whether there is an association between the critical shoulder angle (CSA) and rotator cuff tears (RCT) as well as osteoarthritis (OA), and to compare the reproducibility and accuracy of CSA values as measured with radiograph and MRI. METHODS: Based on a pre hoc power analysis, three independent reviewers measured CSAs for three groups of 10 age-matched patients each (RCT group, OA group, and non-RCT/non-OA group) using radiographs and MRIs. Inter- and intra-observer agreements of CSAs were determined with intraclass correlation coefficients. ANOVA analysis was performed. RESULTS: Interobserver and intra-observer agreements on radiographs were 0.87 (95 % CI 0.78-0.93) and 0.91 (95 % CI 0.82-0.96), respectively. Significant differences in mean CSAs by radiographs were found between the groups (RCT 37.3° ± 2.6°; OA 28.7° ± 2.2°; non-RCT/non-OA 32.7° ± 2.5°; p < 0.001). CSAs obtained from MRI scans had interobserver agreement of 0.62 (95 % CI 0.43-0.78) and intra-observer agreement of 0.68 (95 % CI 0.50-0.80). There was a significant correlation between the patients' pathologies and CSAs measured by radiographs (r = 0.44; p = 0.02) and MRI (r = 0.50; p = 0.01). A significant difference in mean CSAs measured on radiograph versus MRI was seen in the OA group (28.7° ± 2.2° vs. 31.3° ± 4.4°; p = 0.01). CONCLUSION: CSA measurements obtained from radiographs demonstrated excellent interobserver agreement with less variability than CSAs from MRI, especially in OA patients. This study confirms an association between the CSA and RCTs as well as OA pathology. Assessing the CSA may be helpful during diagnostic evaluation of patients with shoulder pain and can help predict pathology. LEVEL OF EVIDENCE: Retrospective cohort study, Level IV.