Can Shoulder Muscle Activity Be Evaluated With Ultrasound Shear Wave Elastography?

Quantitative assessment of rotator cuff muscle activity is important in the treatment of shoulder disorders. However, the known methods for assessing rotator cuff muscle activity thus far have been inaccurate, invasive, and inconvenient. (1) Does the activity of the deltoid, supraspinatus, and infraspinatus muscles measured using ultrasound shear wave elastography have a linear correlation with muscle activity assessed using generally used methods, including isokinetic dynamometry and electromyography? (2) Does the activity of the deltoid, supraspinatus, and infraspinatus muscles measured using shear wave elastography show good intraobserver and interobserver reliability? Twelve volunteers participated in intrasession reliability experiments. They were asked to perform isometric abduction, external rotation, and scaption contractions (defined as elevation of the arm within the plane of the scapula with neutral arm rotation) gradually increased from 0% to 75% of maximal voluntary contraction. The joint torque, electromyographic activity, and shear elastic modulus were synchronously measured in the middeltoid, supraspinatus, and infraspinatus muscles. The validity of the elastic modulus value was assessed using regression analysis between normalized torque and electromyographic root mean square values. For intraobserver and interobserver reliability measurements, repeated experiments were performed with the same protocol. The shear elastic modulus and normalized joint torque with isokinetic dynamometry showed a linear relationship in all muscles (deltoid, supraspinatus, and infraspinatus) and each of the ultrasonography planes (longitudinal and transverse) (mean R > 0.8 and p < 0.001 for all measurements). For the supraspinatus muscle, the mean slope of the relationship between shear elastic modulus in the longitudinal plane and the normalized joint torque during scaption contraction was 1.28 ± 0.39 kPa/%MVC (mean R = 0.93 ± 0.21, p < 0.001). Furthermore, similar results were obtained in relation to electromyography root mean square values (mean R > 0.8 and p < 0.001 in all measurements). For the supraspinatus muscle, the mean slope of the relationship between shear elastic modulus in the longitudinal plane and electromyographic (EMG) root mean square was 0.96 ± 0.27 kPa/%EMG (mean R = 0.91 ± 0.08, p < 0.001). The intraobserver and interobserver reliabilities were excellent in all positions (abduction, external rotation, and scaption) and in both the longitu