OBJECTIVE: The purposes of this study were to describe the imaging appearance of transverse thickening along the articular surface of the supraspinatus and infraspinatus tendons presumed to represent the rotator cable on MR arthrographic images obtained with the shoulder in abduction and external rotation (ABER) and in the non-ABER position and to assess the role of the rotator cable in the diagnosis of rotator cuff tears. MATERIALS AND METHODS: The study group comprised 54 patients who underwent arthroscopy of the shoulder and preoperative MR arthrography in which ABER images were obtained. Two blinded reviewers independently examined the ABER and non-ABER images for the presence of the rotator cable and rotator cuff tears. RESULTS: In intact rotator cuffs, the presumed rotator cable was not well visualized on non-ABER images. In one case of prominent rotator cable, MR arthrography showed no tear on non-ABER images, but at arthroscopy, a partial-thickness undersurface tear was found. On non-ABER images, both readers readily visualized the rotator cable in eight and 10 of 18 cases (44% and 56%) of partial-thickness undersurface tear and four of five cases (80%) of full-thickness tear. On ABER images, both readers readily visualized the rotator cable in 15 and 17 of 31 cases (48% and 55%) of intact cuff, 14 and 15 of 18 cases (78% and 83%) of undersurface tear, and four and five cases (80% and 100%) of full-thickness tear. Interobserver agreement on cable visualization was almost perfect (kappa = 0.86). CONCLUSION: On non-ABER MR arthrographic images, thickening along the articular side of the supraspinatus and infraspinatus tendons presumed to represent the rotator cable suggests the presence of a partial-thickness rotator cuff tear. On ABER images, the cable is well defined in intact and torn rotator cuffs.
OBJECTIVE: The purposes of this study were to describe the imaging appearance of transverse thickening along the articular surface of the supraspinatus and infraspinatus tendons presumed to represent the rotator cable on MR arthrographic images obtained with the shoulder in abduction and external rotation (ABER) and in the non-ABER position and to assess the role of the rotator cable in the diagnosis of rotator cuff tears. MATERIALS AND METHODS: The study group comprised 54 patients who underwent arthroscopy of the shoulder and preoperative MR arthrography in which ABER images were obtained. Two blinded reviewers independently examined the ABER and non-ABER images for the presence of the rotator cable and rotator cuff tears. RESULTS: In intact rotator cuffs, the presumed rotator cable was not well visualized on non-ABER images. In one case of prominent rotator cable, MR arthrography showed no tear on non-ABER images, but at arthroscopy, a partial-thickness undersurface tear was found. On non-ABER images, both readers readily visualized the rotator cable in eight and 10 of 18 cases (44% and 56%) of partial-thickness undersurface tear and four of five cases (80%) of full-thickness tear. On ABER images, both readers readily visualized the rotator cable in 15 and 17 of 31 cases (48% and 55%) of intact cuff, 14 and 15 of 18 cases (78% and 83%) of undersurface tear, and four and five cases (80% and 100%) of full-thickness tear. Interobserver agreement on cable visualization was almost perfect (kappa = 0.86). CONCLUSION: On non-ABER MR arthrographic images, thickening along the articular side of the supraspinatus and infraspinatus tendons presumed to represent the rotator cable suggests the presence of a partial-thickness rotator cuff tear. On ABER images, the cable is well defined in intact and torn rotator cuffs.