Diagnostic performance of MR arthrography with anterior trans-subscapularis versus posterior injection approach for subscapularis tendon tears at 3.0 T

To compare the diagnostic performance of shoulder magnetic resonance arthrography (MRA) with the anterior trans-subscapularis versus posterior injection approach to diagnose subscapularis tendon (SCT) tears. One hundred and sixty-seven arthroscopically confirmed patients (84 anterior and 83 posterior approaches) were included. Two readers retrospectively scored SCT tears. Proportions of correctly graded tears between MR arthrography and arthroscopy were calculated. Retrospective error analysis was performed. The sensitivity and specificity were 80 % (24/30) and 72 % (39/54) by reader 1, 73 % (22/30) and 76 % (41/54) by reader 2 in the anterior approach, and 86 % (30/35) and 79 % (38/48) by reader 1, 80 % (28/35) and 88 % (42/48) by reader 2 in the posterior approach, respectively. There were no significant differences in sensitivity and specificity between the two groups. Proportions of correctly graded tears of both readers were 48 % and 36 % in the anterior approach, and 70 % and 68 % in the posterior approach, respectively. The intratendinous collection of contrast material was not statistically significantly different between anterior (n = 8) and posterior (n = 3) approach group. For the MRA diagnosis of SCT tears, there was no significant difference between the anterior trans-subscapularis and the posterior approach. (orig.)