Analysis of In Vivo Glenohumeral, Scapular and Clavicle Kinematics of Reverse Total Shoulder Arthroplasty Patients During Shoulder Internal Rotation

This study aimed to analyze the kinematics of shoulder internal rotation in RTSA, compare high and low shoulder internal rotation groups, and explore the factors necessary for shoulder internal rotation.

This study included 10 patients with RTSA (five men and five women, mean age; 79.3 [71–90] years). Equinoxe (Exactech, Gainesville, FL, USA) was used in all cases. Single-plane fluoroscopic X-ray imaging was performed. The participants were asked to assume a sitting position (starting position), and the shoulder joint was internally rotated to the maximal hand-behind-back position. The three-dimensional (3D) models of the humeral component, scapula, and clavicle were created using a computer-aided design model and software based on computed tomography information. Subsequently, a coordinate system was defined for each 3D model. The 3D model of the artificial joint was fitted to the contours of the fluoroscopic X-ray images. The glenohumeral joint angle (humeral component relative to the scapular component) was determined using the Euler angle. The scapular component (upward rotation, external rotation, and posterior tilting), glenohumeral (abduction, flexion, and internal rotation), and clavicular (protraction, lateral rotation, and posterior tilt) angles were calculated at the starting and maximal hand-behind-back positions. For data analysis, the patients were categorized by constant shoulder internal rotation score. Those above the waist were categorized as the high group, and those at or below the sacroiliac joint were categorized as the low group. Each joint angle in the starting and hand-behind-back positions was compared between the groups using analysis of variance, with a statistical significance level of 0.05.

The high group comprised four men and one woman, whereas the low group included one man and four women. The scapular component and glenohumeral and clavicle angles were not significantly different between the groups. 

The patients were divided into the high and low groups based on the constant shoulder internal rotation score; however, there was no difference in shoulder girdle kinematics between both groups. We should focus not only on the shoulder girdle, but also on other adjacent joints.

The influence of other joints and shoulder joints should be considered to improve shoulder internal rotation motion in patients with RTSA.