Purpose: To retrospectively determine the diagnostic accuracy of magnetic resonance (MR) arthrography of the shoulder in the evaluation of lesions of the biceps pulley and to evaluate previously described and new diagnostic signs.
Materials and methods: Institutional review board approval was obtained; the requirement for informed consent was waived. MR arthrograms of 80 consecutive patients (mean age, 34.2 years; 53 male, 27 female) with arthroscopically proved intact or torn pulley systems were assessed for the presence of a pulley lesion by three radiologists who were blinded to arthroscopic results. Criteria evaluated were displacement of the long head of the biceps tendon (LHBT) relative to the subscapularis tendon on oblique sagittal images (displacement sign), medial subluxation of the LHBT on transverse images, nonvisibility or discontinuity of the superior glenohumeral ligament (SGHL), presence of biceps tendinopathy, and rotator cuff tears adjacent to the rotator interval.
Results: There were 28 pulley lesions noted at arthroscopy. For observers 1, 2, and 3, respectively: MR arthrography showed a sensitivity of 89%, 86%, and 82% and a specificity of 96%, 98%, and 87% in the detection of pulley lesions. Nonvisibility or discontinuity of the SGHL was sensitive (79%, 89%, and 79%) and specific (83%, 79%, and 75%). With the displacement sign, sensitivity was 86%, 82%, and 75% and specificity was 96%, 98%, 90%. Tendinopathy of the LHBT on oblique sagittal images showed a sensitivity of 93%, 82%, 64%; specificity was 81%, 96%, and 85%. Subluxation of the LHBT was insensitive (36%, 50%, and 64%) but specific (100%, 98%, and 96%).
Conclusion: MR arthrography is accurate in the detection of pulley lesions; the displacement sign, nonvisibility or discontinuity of the SGHL, and tendinopathy of the LHBT on oblique sagittal images are the most accurate criteria for the detection of pulley lesions.