Correlation of MRI with arthroscopy for the evaluation of the subscapularis tendon: a musculoskeletal division’s experience

Abstract

To determine the accuracy of MR imaging for the evaluation of the subscapularis tendon as well as define imaging findings that will increase accuracy. Retrospective review of the MR and operative (OR) reports of 286 patients was conducted and reviewed for the presence/degree (partial (PT)/full-thickness (FT)) of tearing; only PT articular tears were included. The presence of a supraspinatus tear and time interval between surgery and MRI were also documented. All of the PT tears called on MRI were also reviewed to see if there was a statistically significant association between certain imaging characteristics and the presence of a tear in surgery. Statistical analysis included 95 % confidence intervals, Fisher’s exact, and exact Mann–Whitney tests. A total of 244 patients were included in the study with a total of 25 subscapularis tears, 16 PT and nine FT, and 219 intact tendons in arthroscopy; 20/25 tears and 200 intact sensitivity of 80%, specificity of 91%, accuracy of 90%, positive predictive value of 51%, and negative predictive value of 98 %. There was a significant association between the presence of a PT tear during arthroscopy and fluid-like signal within the tendon on more than one imaging plane (p<0.001) with an accuracy of 90%. This study reflects a musculoskeletal radiology section’s experience with the diagnosis of subscapularis tendon pathology, demonstrating that MRI could be used to accurately evaluate the subscapularis tendon. An understanding of certain imaging pitfalls and the presence of fluid-like signal on multiple imaging planes should increase the diagnostic accuracy of the radiologist evaluating the subscapularis tendon for the presence of a tear.