Thomas Magee, Presenter: Nothing to Disclose

Marc David Shapiro MD, Abstract Co-Author: Nothing to Disclose

David S. Williams MD, Abstract Co-Author: Nothing to Disclose

R. Richard Ramnath MD, Abstract Co-Author: Nothing to Disclose

MR imaging of the shoulder has been found to be very sensitive and specific for detection of full thickness supraspinatus tendon tears at 1.5 Tesla or lower field strength as compared with arthroscopy. Three Tesla MR imaging of the shoulder has not been specifically assessed. This study assesses the sensitivity and specificity of MR imaging at 3.0 Tesla for supraspinatus tendon tears as compared with arthroscopy.

Two experienced musculoskeletal radiologists retrospectively reviewed MR images of the shoulder in 100 consecutive patients. All patients had a proton density fat saturated axial, coronal T1 and fat saturated T2 weighted coronal and sagittal images performed. The radiologists read the MR images by consensus without knowledge of arthroscopy results. Scans were read as to whether there were full thickness or partial thickness supraspinatus tendon tears. If partial thickness supraspinatus tendon tears were seen on MR imaging, it was stated as to whether the partial thickness tear was articular or bursal in location. Sixty seven of the 100 patients went on to arthroscopy. After consensus review of MR images, arthroscopy results were compared with consensus MR interpretations.

Of the sixty seven patients who went on to arthroscopy, forty six had full thickness supraspinatus tendon tears. Eight of the sixty seven patients had partial thickness supraspinatus tendon tears. Six of these eight partial thickness tears were along the articular surface and two were along the bursal surface. Forty-five out of forty six full thickness tears seen at arthroscopy were seen on consensus MR reading. All eight partial thickness tears seen at arthroscopy were seen at consensus MR reading however one of the partial thickness articular surface tears seen at arthroscopy was read as a full thickness tear on consensus MR reading.

3.0 Tesla MR imaging of the shoulder is very sensitive and specific as compared with arthroscopy in detection of full thickness and partial thickness supraspinatus tendon tears.

Magee, T, Shapiro, M, Williams, D, Ramnath, R, Sensitivity and Specificity in Detection of Supraspinatus Tendon Tears with 3.0 Tesla MR Imaging of the Shoulder.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4404943.html