Leon David Rybak MD, Abstract Co-Author: Nothing to Disclose

Renata La Rocca Vieira MD, Presenter: Nothing to Disclose

Timothy Folsom Shepard MD, Abstract Co-Author: Nothing to Disclose

Graham Charles Wiggins DPhil, MS, Abstract Co-Author: Nothing to Disclose

Michael Paul Recht MD, Abstract Co-Author: Nothing to Disclose

Christian Glaser MD, Abstract Co-Author: Nothing to Disclose

To assess performance of an intermediate-w fat saturated (FS) isotropic TSE sequence (SPACE) in comparison to a standard 2D protocol and arthroscopy in direct MR shoulder arthrography (MRA) at 1.5T

An intermediate-w FS 3D TSE sequence (resolution 0.65 x 0.582 mm³; acq time 7:17) was added to routine shoulder MRA (triplanar T1-w TSE FS, coronal T2 TSE FS; resolution 3 x 0.272 mm³, B0=1.5T, 4-ch shoulder coil, time 17:15) in 28 consecutive symptomatic patients (23 m, 5 f, mean age=28). 11 patients had subsequent arthroscopy. Triplanar SPACE reformats were created (1 x 0.542 mm³). 2 radiologists independently graded the SPACE and 2D images for visualization (Lickert scale:1-5) of normal anatomy as well as detection (counts) of pathology (labrum, cartilage, cuff and glenohumeral ligaments) assigning confidence levels to readings. The 2D TSE and SPACE images were evaluated at separate sessions. Statistical analysis was performed using exact Wilcoxon test.

SPACE provided significantly better visualization of the biceps anchor, subscapularis insertion and articular cartilage and was equivalent to 2D TSE for other structures. Confidence scores for detecting pathology were comparable for SPACE (4.34 ± 0.74) and 2D images (4.43 ± 0.80) (p=0.768), but significantly more findings (p=0.0054) were reported on SPACE (7.5 ± 5.03 per patient, 210 total) than 2D TSE (5.57 ± 4.86 per patient, 156 total). While sensitivity for lesion detection as referenced by arthroscopy was comparable, there was a slightly higher specificity for the 2D protocol (86%) than SPACE (81%) (p=0.04). % concordance between readers for all findings and reader confidence was 59.4/71.4 and 25/31.7 for SPACE/2D TSE

3D isotropic TSE for shoulder MRA at 1.5T was equal to, or better than, standard 2D sequences for depiction of normal anatomic detail and was equally sensitive and slightly less specific in lesion detection. The interreader concordance for SPACE was comparable to 2D TSE for confidence levels, but was worst for interreader variability with respect to findings. This may be attributable to a lack of familiarity with the image characteristics and contrast of SPACE.

SPACE provides equal or better anatomic detail at shoulder MRA. With refinement and greater reader familiarity, it has potential to augment or replace 2D TSE protocols with significant time savings.

Rybak, L, Vieira, R, Shepard, T, Wiggins, G, Recht, M, Glaser, C, 3D Isotropic Intermediate-weighted TSE: A New Approach for Shoulder Arthrography at 1.5 T? Comparison to Standard 2D TSE.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9012614.html