Accuracy of sagittal oblique view in preoperative indirect magnetic resonance arthrography for diagnosis of tears involving the upper third of the subscapularis tendon

Abstract

Background: The purpose of this study was to investigate the accuracy of sagittal oblique view (SOV) magnetic resonance imaging grading for tears involving the upper third of the subscapularis tendons in correlation with arthroscopic findings. We also propose a schematic classification in the SOV for upper-third subscapularis tears.

Methods: The study analyzed 364 patients with an average age of 56.0 years. All patients underwent indirect magnetic resonance arthrography (MRA) before surgery. A slightly different magnetic resonance grading was introduced with SOV by musculoskeletal radiologists (Y.C.Y. and J.Y.Y.) and an orthopedic shoulder surgeon (H.Y.R.) who had no information about the clinical data. Subscapularis tendon tear classifications were all prospectively analyzed. Magnetic resonance grading was compared with the arthroscopic findings.

Results: The sensitivity, specificity, and accuracy of subscapularis tendon tear detection were 0.72, 0.77, and 0.75, respectively, for the radiologists. For the orthopedic surgeon, the corresponding values were 0.73, 0.83, and 0.79, respectively. The κ coefficients for interobserver agreement comparing magnetic resonance grade of the musculoskeletal radiologists with the arthroscopic grading showed fair values (κ value, 0.301). Interobserver agreement between the magnetic resonance grade of the orthopedic surgeon and arthroscopic grades was fair (κ value, 0.377). Interobserver agreement between the musculoskeletal radiologists and the orthopedic surgeon was moderate (κ value, 0.591).

Conclusions: The accuracy of SOV indirect MRA for detecting tears was 0.75 to 0.79. Although the correlation between magnetic resonance and arthroscopic grading was only fair, for most of the magnetic resonancegrading on SOV, the first facet view showed similar results compared with arthroscopy. The accuracy of determining whether subscapularis repair is indicated was 0.82 to 0.83 with SOV indirect MRA.

Keywords: MRA; MRI; Subscapularis; first facet; leading edge tear; sagittal oblique view.