Anteroposterior translation of the glenohumeral joint in various pathologies: differences between shoulder MRI in the adducted neutral rotation and abducted externally rotated positions

The present study was performed to determine the translation of the glenohumeral joint in patients with and without shoulder lesions by comparing the magnetic resonance images obtained in the conventional adducted neutral rotation position with those obtained in the abducted externally rotated position.

Two hundred and eighty-five consecutive shoulders without rotator cuff tears that had been subjected to magnetic resonance imaging (MRI) without arthrography in the abducted externally rotated position were reviewed retrospectively. Among them, 50 shoulders without pathology were selected at random to be compared with three shoulder pathology groups, comprising shoulders with superior labrum, anterior-to-posterior (SLAP) lesions without range of motion (ROM) limitation (group I, 47 shoulders), with massive rotator cuff tears without ROM limitation (group II, 20 shoulders), and with full-thickness subscapularis tendon tears without ROM limitation (group III, 20 shoulders). Glenohumeral translation in the anterior-to-posterior direction relative to the glenoid face was evaluated using a method based on the glenohumeral contact point (CP) and humeral head centre (HHC) in the adducted neutral rotation and abducted externally rotated views, which were measured by three orthopaedic surgeons. For each shoulder, the differences in translation for the glenohumeral CP and HHC between the adducted neutral rotation and abducted externally rotated views were calculated as relative posterior translation in millimetres.

The differences in ΔCP and ΔHHC between group I and the normal control group were not statistically significant. The differences in ΔCP (P = 0.001) and ΔHHC (P = 0.001) between group II and the normal control group were statistically significant. Additionally, the differences in ΔCP and ΔHHC between group III and the normal control group were not statistically significant.

The MRI in abducted externally rotated view in patients with SLAP lesions or full-thickness subscapularis tendon tears diagnosed by conventional MRI alone showed no significant glenohumeral posterior translation relative to the adducted neutral rotation view in the present study. However, the abducted externally rotated view in patients with massive rotator cuff tears showed significant glenohumeral anterior translation relative to the adducted neutral rotation view.

Diagnostic study, Level II.