Cross-Sectional Area of the Rotator Cuff Muscles in MRI – Is there Evidence for a Biomechanical Balanced Shoulder?

OBJECTIVE To provide in-vivo evidence for the common biomechanical concept of transverse and craniocaudal force couples in the shoulder that are yielded by both the rotator cuff muscles (RCM) and the deltoid and to quantitatively evaluate and correlate the cross-sectional areas (CSA) of the corresponding RCM as a surrogate marker for muscle strength using MRI. MATERIALS AND METHODS Fifty patients (mean age, 36 years; age range, 18-57 years; 41 male, 9 female) without rotator cuff tears were included in this retrospective study. Data were assessed by two readers. The CSA (mm2) of all rotator cuff muscles was measured on parasagittal T1-weighted FSE sequence at two different positions (at the established "y-position" and at a more medial slice in the presumably maximal CSA for each muscle, i.e., the "set position"). The CSA of the deltoid was measured on axial intermediate-weighted FSE sequences at three positions. CSA measurements were obtained using 1.5 Tesla MR-arthrographic shoulder. Pearson's correlation for the corresponding CSA of the force couple as well as was the intraclass correlation coefficient for the inter- and intra-reader agreement was calculated. RESULTS The mean CSA was 770 mm2 (±167) and 841 mm2 (±191) for the supraspinatus (in the y- and set-positions, respectively) and 984 mm2 (±241) and 1568 mm2 (±338) for the infraspinatus. The mean CSA was 446 mm2 (±129) and 438 mm2 (±128) for the teres minor (in the y- and set-positions, respectively) and 1953 mm2 (±553) and 2343 mm2 (±587) for the subscapularis. The three measurements of the deltoid revealed a CSA of 3063 mm2 (±839) for the upper edge, 3829 mm2 (±836) for the lower edge and 4069 mm2 (±937) for the middle of the glenoid. At the set position Pearson's correlation of the transverse force couple (subscapularis/infraspinatus) showed a moderate positive correlation of r = 0.583 (p<0.0001) and a strong correlation when the CSA of the teres minor was added to the infraspinatus CSA (r = 0.665, p = 0.0008) and a strong positive correlation of the craniocaudal force couple (supraspinatus/deltoid) that ranged from r = 0.565-0.698 (p<0.0001). Inter-reader agreement (ranged from 0.841 to 0.997, p = 0.0007) and intra-reader agreement were excellent (ranged from 0.863 to 0.999, p = 0.0006). CONCLUSION The significant correlation of the CSA of the RCM that form the transverse (subscapularis/infraspinatus-teres minor) and craniocaudal (supraspinatus/deltoid) force couple measured by MR-arthrography supports the biomechanical concept of a dynamically balanced shoulder in patients with an intact rotator cuff.