INTRODUCTION. Total shoulder arthroplasty (TSA) implants are used to restore function to individuals whose shoulder motions are impaired by osteoarthritis. To improve TSA implant designs, it is crucial to understand the kinematics of healthy, osteoarthritic (OA), and post-TSA shoulders. Hence, this study will determine in vivo kinematic trends of the glenohumeral joints of healthy, OA, and post-TSA shoulders. Methods. In vivo shoulder kinematics were determined pre and post-operatively for five unilateral TSA subjects with one healthy and a contralateral OA glenohumeral joint. Fluoroscopic examinations were performed for all three shoulder categories (healthy, OA, and post-TSA) for each subject shoulder abduction and external rotation. Then, three-dimensional (3D) models of the left and right scapula and humerus were constructed using CT scans. For post-operative shoulders, 3D computer-aided design models of the implants were obtained. Next, the 3D glenohumeral joint kinematics were determined using a previously published 3D to 2D registration technique. After determining kinematics, relative Euler rotation angles between the humerus and scapula were calculated in MATLAB® to determine range of motion (ROM) and kinematic profiles for all three shoulder categories. The ROMs for each category were compared using paired t-tests for each exercise. Also, the location of the contact point of the humerus on the glenoid was found. This allowed the vertical translation from the most superior to most inferior contact point (SI contact range) to be calculated as well as the horizontal translation from the most anterior to most posterior contact point (AP contact range). The SI and AP contact ranges for all shoulder categories were compared using paired t-tests for each exercise. Results. Abduction. According to preliminary results, the averages range of abduction for healthy, OA, and post-TSA shoulders was 51.5 °, 19.4°, and 56.7°, respectively. The average SI contact range of abduction for healthy, OA, and post-TSA shoulders was 14.1 mm, 16.4 mm, and 14.1 mm, respectively while the AP contact range was 10.0, 14., and 14.3, respectively. The ranges of abduction between healthy and OA and between OA and post-TSA shoulders, and the AP contact range for healthy and OA shoulders displayed statistically significant differences at the α=0.05 level. External Rotation. The averages range of External Rotation for healthy, OA, and post-TSA shoulders was 63.6°, 31.1°, and 44.5°, respectively. The averages SI contact range of External Rotation for healthy, OA, and post-TSA shoulders was 20.7, 12.7, and 15.9 mm, respectively while the averages AP contact range was 8.5,12.9 mm, and 13.8 mm, respectively. The ranges of abduction for healthy and OA as well as AP contact range for healthy and OA shoulders were statistically different at the α=0.05 level. Conclusions. This study's preliminary results indicate that healthy, OA, and post-TSA shoulders show statistically significant difference in kinematic trends including ROM and contact point translation. These differences may result from the varying geometries of each condition or from subjects altering kinematic trends to reduce pain in OA shoulders. In addition, this study may provide a reference for future studies analyzing the kinematics of post TSA shoulders