[Comparison between T2*- and T2-weighted images in diagnosing rotator cuff tears]

This study was performed to determine the merits of T2*-weighted images in diagnosing rotator cuff tear, compared with T2-weighted images. T2-and T2*-weighted images were obtained in 10 asymptomatic volunteers and 94 patients with symptoms referable to the rotator cuff. The increased signal with full thickness of the rotator cuff was not shown on either T2- or T2*-weighted images in the volunteers. These findings on T2-weighted images and on T2*-weighted images were observed in 33 and 58 of 94 patients with symptoms, respectively. Every patient who showed these abnormal findings on T2-weighted images had the abnormal findings on T2*-weighted images. These findings on T2*-weighted images were wider than those on T2-weighted images in 20 of 33 patients. Surgical findings were available in 21 of 94 patients. Rotator cuff tears were surgically confirmed in 20 patients whose MR images showed increased signal lesions on both T2- and T2*-weighted images. On the other hand, one patient who did not have rotator cuff tear showed increased signal lesion with full thickness on T2*-weighted images, but not on T2-weighted images. We think increased signal lesions on T2-weighted images may strongly suggest rotator cuff tear, whereas those on T2*-weighted images are not specific.