Estimation of the load sharing pattern in the shoulder - a comparison between electromyographical measurements and biomechanical model calculations

The purpose of this study was to constitute to the validity of a three-dimensional biomechanical shoulder model using EMG recordings from ten relevant shoulder muscles. EMG data from 17 subjects, participating in two studies concerning voluntary reduction of the descending part of the trapezius muscle activity, were used. Measured changes in EMG activity were compared to calculated changes in muscle activity. The study evaluated ten shoulder muscles: the supraspinatus muscle, the infraspinatus muscle, the anterior and the middle part of the deltoid muscle, the levator scapulae muscle, the major and the minor parts of the rhomboid muscle, the middle part of the serratus anterior muscle and the descending and the transverse parts of the trapezius muscle. The comparisons were made in six standardized arm positions and individual anthropometrical data were considered. The relative change in force was calculated for each combination of muscle, arm position and level of voluntary reduced force of the descending part of the trapezius muscle and compared to the corresponding relative change in EMG activity. There was a passable agreement between model calculations and EMG measurements for all muscles except for the levator scapulae muscle and the supraspinatus muscle. Model calculations indicated a larger change for the levator scapulae muscle and a smaller change for the supraspinatus muscle compared to the EMG measurements. It is concluded that shoulder muscle forces estimated by the model acceptably agree with the measured EMG activity, except for the levator scapulae muscle and the supraspinatus muscle. The closest correlation was found for the middle and the anterior part of the deltoid muscle, the upper part of the infraspinatus muscle and the middle part of the serratus anterior muscle.