Critical Shoulder Angle Measurement Accuracy in Magnetic Resonance Imaging versus Radiographs: A retrospective cross-sectional study

Background: The Critical Shoulder Angle (CSA) is a radiographic measurement that has proven valuable in assessing shoulder pathologies such as osteoarthritis and rotator cuff tears. Despite its clinical utility, the consistency of CSA measurements across different imaging modalities, particularly magnetic resonance imaging (MRI) versus radiographs, remains debated. Objective: This retrospective cross-sectional study aims to assess the accuracy and reliability of CSA measurements obtained from routine shoulder MRI compared to standard radiography. Methods: A total of 210 patients with both coronal oblique MRI images and true anteroposterior (AP) radiographs of the shoulder were included in the study. CSA measurements were independently performed by two observers using a standardized method, and inter- and intra-observer agreement was evaluated. Statistical analyses were conducted to determine differences in CSA measurements between the two modalities and assess agreement levels. Results: The mean CSA measured on radiographs (33.7 ± 4.4) was significantly higher than that on MRI (32.8 ± 4.06, p=0.04). Inter- and intra-observer agreements were higher for radiographic measurements (ICC > 0.95) compared to MRI, where variability was noted, particularly concerning gender differences. CSA values showed no significant differences based on shoulder side (right vs. left) in either modality. Conclusion: CSA measurements were more accurate and consistent on radiographs than MRI, suggesting that radiography remains the preferred modality for assessing CSA in clinical settings. The study highlights the need for standardized imaging techniques to improve measurement consistency across modalities.