• S K Goergen, W G Bradley, J Liu, J K Tam, M D Dubin, P J Pema, L M Teresi, J Jordan, and D Morrison.
• Memorial MR Center, Long Beach Memorial Medical Center, CA 90806, USA.
• J Comput Assist Tomogr. 1996 Sep 1;20(5):829-33.

PurposeOur goal was to determine the accuracy of MRI in the diagnosis of infraspinatus tendon injury and more specifically to determine if the antero-posterior extent of a rotator cuff tear is predictive of infraspinatus tearing.MethodThe MR images of 41 shoulders with surgically proven supraspinatus tears at surgery were retrospectively reviewed. The following were assessed for each of the 41 studies: the number of oblique coronal images on which a tendon defect could be seen, the angle subtended by the tear on axial images (the rotator cuff "axial angle"), and the extent of signal abnormality on sagittal images.ResultsThe rotator cuff axial angle was 75.6 degrees in patients with infraspinatus tendon tears (ITTs) versus 40 degrees in those without ITTs, and this difference was significant (p < 0.001, t = 3.06). The mean number of oblique coronal images (obtained with a 4 mm slice and 1 mm gap) showing signal abnormality was 5.4 in the ITT group versus 2.9 in those without ITTs, and this difference was also significant (p < 0.001, t = 4.45). The mean sagittal extent of the tendon abnormality was 24.6 mm in the ITT group and 11.6 mm in those without ITTs, but the difference was not significant (p > 0.05, t = 1.1364).ConclusionThe axial angle and the number of oblique coronal images in which signal abnormality was present were significantly related to a higher incidence of infraspinatus tears.

17 keywords...

hide…