The detection of the capsular tear at the undersurface of the extensor carpi radialis brevis tendon in chronic tennis elbow: the value of magnetic resonance imaging and computed tomography arthrography

Abstract

This study compared the diagnostic efficacy of magnetic resonance imaging (MRI) and computed tomography arthrography (CTA) in the assessment of capsular tears at the undersurface of the extensor carpi radials brevis tendon in chronic tennis elbow using arthroscopy as a gold standard. Because of the higher spatial resolution of CT, we hypothesized that CTA is superior to MRI for assessing capsular tears. We retrospectively reviewed 19 consecutive patients with chronic tennis elbow with preoperative MRI and CTA studies who underwent arthroscopic surgery. Three observers with different levels of training and experience (musculoskeletal radiologist, experienced elbow surgeon, and hand fellow) evaluated the capsular tear by MRI and CTA in a blinded manner. The results of the MRI and CTA were compared and the agreement among the 3 observers was determined using an intraclass correlation coefficient (ICC). Then, the results of the MRI and CTA examinations were compared with the intraoperative findings of the arthroscopic examination. The sensitivity, specificity, and κ value were calculated. The ICC of CTA (0.855) was superior to MRI (0.645). The sensitivity, specificity, and κ value of CTA were superior to those of MRI in each of the 3 observers. The κ value was 0.79, 0.89, and 0.79 for CTA, and 0.48, 0.48, and 0.27 for MRI for the radiologist, surgeon, and fellow, respectively. CTA was a reliable and accurate diagnostic modality compared with MRI to detect the capsular tear in patients with chronic tennis elbow. CTA was less influenced by the observer's experience.