Posterior decentering of the humeral head on Shoulder MR arthrography: Significant association with posterior synovial proliferation

OBJECTIVE. The purpose of this study is to analyze MRI findings associated with posterior decentering of the humeral head on shoulder MR arthrography. MATERIALS AND METHODS. A total of 255 MR arthrograms were obtained during a 10-month period. MR arthrograms in patients with posterior decentering of the humeral head (n = 33) were reviewed and compared with those of randomly selected control group without posterior decentering (n = 66). MR arthrograms were retrospectively evaluated by two observers for posterior factors (posterior synovial proliferation, posterior capsular thickening, and posterior labral abnormality), rotator cuff factors related to fatty degeneration, glenoid version, and anterior factors (subcoracoid bursa effusion and rotator interval tear). The chi-square test, Fisher exact test, and linear-by-linear association were used for comparison of categoric data; the t test was used for comparison of the glenoid version; and multivariate stepwise logistic regression analysis was performed. RESULTS. At univariate analysis, posterior synovial proliferation (27.3% [9/33] in the posterior decentering group vs 6.1% [4/66] in the control group; p = 0.003); posterior capsule thickening (21.2% [7/33] vs 0.0% [0/66]; p < 0.001); fatty infiltration of the supraspinatus, infraspinatus, and teres minor (linear-by-linear association values, 7.944, 10.496, and 5.985, respectively; p = 0.005, 0.001, and 0.014, respectively); and rotator interval tear (51.5% [17/33] vs 30.3% [20/66]; p < 0.04) were more frequently found in the posterior decentering group, with a statistically significant difference. At multivariate analysis, only the posterior synovial proliferation was significantly associated with posterior decentering of the humeral head (odds ratio, 7.675; 95% CI, 2.159-27.288). CONCLUSION. Posterior decentering of the humeral head is most significantly associated with posterior synovial proliferation. In addition, rotator cuff interval abnormalities and rotator cuff atrophy are associated with posterior decentering of the humeral head to a lesser extent. Awareness of the association of the posterior decentering of the humeral head with the factors described here will facilitate the effective interpretation of routine MR arthrograms in daily practice.