|| Checking for direct PDF access through Ovid
The purposes of this study were (a) to describe the detailed movements of the shoulder complex during the cocking phase in tennis serve and (b) to determine the contribution of the scapular rotation to the peak shoulder external rotation attained at the end of the cocking phase.Twenty tennis players performed flat tennis serves with the maximum effort, and three-dimensional kinematic data of the thorax, scapula, and humerus of the dominant side were recorded by an electromagnetic tracking device (Liberty; Polhemus) at 240 Hz. The humeral rotation with respect to the thorax (named as the resultant shoulder rotation) was decomposed into the scapular rotation with respect to the thorax and the humeral rotation with respect to the scapula (named as the glenohumeral rotation). The scapular rotation that contributed to attain the peak resultant shoulder external rotation was quantitatively determined, and the ratio of the glenohumeral external rotation to the scapular rotation that contributed to attain the peak resultant shoulder external rotation was computed to represent the scapulohumeral rhythm for shoulder external rotation.Of the three components of scapular rotations, the scapular posterior tilt was the primary component that contributed to the attainment of the peak resultant shoulder external rotation. The scapulohumeral rhythm for shoulder external rotation (scapular posterior tilt-glenohumeral external rotation) exhibited during the backswing phase was 1:2.3.The peak resultant shoulder external rotation was generated by a synchronized combination of the scapular posterior tilt and glenohumeral external rotation. A consistent pattern of three phases of the scapular posterior tilt and the glenohumeral external rotation that could be accurately modeled by three linear regressions was observed in preparation for the peak shoulder external rotation in tennis serve.