Supraspinatus tendon tears: comparison of 3D US and MR arthrography with surgical correlation

The objective of the study was to compare the diagnostic reliability of 3D US with MR arthrography in diagnosing supraspinatus tendon tears, with arthroscopic findings used as the standard. In a prospective study 50 patients who later underwent arthroscopic surgery of the rotator cuff were examined pre-operatively by 3D US with MR arthrography. The presence or absence of a full- or partial-thickness supraspinatus tendon tear and the tear size as demonstrated by each imaging and arthroscopy was recorded. The tear size was divided into three grades: small (< 1 cm), medium (1-3 cm), and large (> 3 cm). The arthroscopic diagnosis was a full-thickness tear in 40 patients, partial-thickness tears in 5, and intact supraspinatus tendon in 5. 3D US correctly diagnosed 35 out of 40 full-thickness tears and MR arthrography 39 out of 40 full-thickness tears. Regarding partial-thickness tears, 3D US underestimated 2 cases as no tear and overestimated 1 case as a full-thickness tear. MR arthrography underestimated 1 case as a partial-thickness tear and overestimated 2 cases as full-thickness and partial-thickness tears respectively. 3D US and MR arthrography yield a sensitivity for full-thickness tears of 87.5% and 97.5% with specificity of 90.0% and 90.0%. Based on the grading system, 3D US measurements correctly predicted the tear size of 23 (65.7%) of the 35 full-thickness tears and MR arthrography 30 (75.0%) of the 39 full-thickness tears. Three-dimensional ultrasound seems to be a promising imaging modality comparable to MR arthrography for the assessment of the supraspinatus tendon tears.