Gradient-echo versus fat-suppressed fast spin-echo MR imaging of rotator cuff tears.

Overview

OBJECTIVE: The purpose of this study is to compare fat-suppressed T2-weighted fast spin-echo sequences with T2-weighted multiplanar gradient-echo sequences in revealing rotator cuff disorders. MATERIALS AND METHODS: Oblique coronal fat-suppressed fast spin-echo images and gradient-echo images of 39 patients who underwent surgery of the shoulder were retrospectively reviewed. Images from the two techniques were reviewed separately. Each set of images was interpreted twice by four musculoskeletal radiologists independently and without knowledge of the surgical findings. Sensitivity and specificity for both sequences were determined. The kappa statistic was used to calculate intraobserver and interobserver agreement of interpretations. RESULTS: For detecting any tear, the sensitivity of the four readers was 71-96% using fast spin-echo and 58-100% using gradient-echo imaging. Confidence intervals showed no difference between the two sequences. Sensitivity of detecting full-thickness tears was 83-100% for both sequences. Interobserver agreement was good. CONCLUSION: T2-weighted gradient-echo and fat-suppressed T2-weighted fast spin-echo sequences depict rotator cuff tears equally well and take less time to acquire than conventional dual-echo spin-echo sequences.