Objectives:Increased humeral retroversion in the throwing shoulder is considered to exist in many baseball players and is often considered as a cause of internal rotation deficit. However, retroversion angles using two-dimensional (2D) methods utilized in previous studies did not correlated with a true three-dimensional measurement technique (3D retroversion) in our study. The purposes of this study were to determine 1) if significant side-to-side differences in retroversion exist, and 2) the relationships between retroversion and glenohumeral range of motion (ROM) in baseball players with a shoulder disorder.Methods:Fourteen male baseball players (age: 21.4 ± 1.5 years; mean ± standard deviation) with a throwing shoulder disorder were enrolled, including 11 in the collegiate and 3 in the adult levels, comprising 5 pitchers and 9 position players. The mean age when the athletes had started playing baseball was 9.1 years (range, 6 to 10 years) and the mean playing experience was 12.1 years (range, 9 to 17 years). Patients with history of fracture or surgery in the shoulder or elbow were excluded. Outcome measurements included a 3D retroversion angle and glenohumeral ROM in bilateral shoulders. Patients underwent CT scan at 1.0 mm slice pitch for the bilateral humerus and geometric bone models were created. The 3D retroversion angle was defined as an angle between the projected humeral neck line (connecting the spherical center of the humeral head and the cylindrical axis of the humeral shaft) onto the horizontal plane and the flexion-extension axis of the distal humerus obtained by a cylinder-fitting method for the capitulum and trochlea (intratester reliability: ICC 0.98, standard error of measurement: 1.7º). The CT slices for the proximal and distal humerus were used to measure a 2D-CT retroversion angle using a reported convention. An indirect measurement of retroversion was performed using a bicipital grove-ulna angle by an ultrasound images dusing a reported convention. Glenohumeral ROM measurements included internal rotation (IR) and horizontal adduction (HAd) in the supine position with the scapula manually stabilized on the treatment table. Independent t tests were used to determine if significant side-to-side differences exist in retroversion and glenohumeral ROM. Pearson correlation coefficients were used to determine the association between retroversion and glenohumeral ROM in the throwing shoulders. Statistical analysis was performed with PASW Statistics 18 (SPSS, Inc, Chicago, IL). An alpha level of 0.05 was set a priori for statistical significance.Results:No significant difference was found in 3D retroversion between the throwing and non-throwing shoulders (P = .182). The throwing shoulders demonstrated decreased IR and HAd compared with the non-throwing shoulders (P < .001 and P = .002, respectively). There were no significant correlations between retroversion and glenohumeral ROM for the throwing shoulders (Table). The 2D-CT retroversion and bicipital grove-ulna angles were greater in the throwing shoulder (P = .007 and .009, respectively).Conclusion:Contradicted to the previous studies, no side-to-side difference in retroversion and no association between retroversion and glenohumeral ROM were observed in the baseball players with a shoulder disorder. We believe that deficit in shoulder IR and HAd may be attributed to soft tissue tightness. The results of the previous studies involving humeral retroversion should be reassessed.