Hypothesis: Results of the reverse shoulder prosthesis on pain are generally satisfying; however, active range of motion (ROM) is often not optimal, especially after revision. A kinematic and clinical analysis of the reverse prosthesis was performed to provide more precise information on its glenohumeral motion pattern. We hypothesized that the difference in the primary and revision cases is due to differences in the motion in the glenohumeral joint. Materials and methods: The motion pattern of 31 patients with a reverse prosthesis (35 shoulders, 19 primary and 16 revisions) was measured during 3 active ROM tasks-forward flexion, abduction, and axial rotation. Average age was 71 ± 8 years (range, 58-85 years). Average follow-up was 23 ± 14 months (range, 4-63 months). Kinematic measurements were performed with a 3-dimensional electromagnetic tracking device. Clinical evaluation was performed by obtaining Constant score, Disabilities of Arm, Shoulder and Hand (DASH) score, and the Simple Shoulder Test (SST). Acromial-prosthetic distance was measured on anteroposterior radiographs. Results: Primary placed prostheses showed significantly better active glenohumeral motion than revisions for forward flexion (71° ± 18° vs 53° ± 26°, P < .05), abduction (64° ± 15° vs 46° ± 24°, P < .05), and active external rotation (31° ± 25° vs 13° ± 16°, P < .05). Constant score improved for the whole group from 24 (range, 5-47) to 50 (range, 8-87; P < .001), for the primary group from 28 (range, 13-47) to 60 (range, 8-87; P < .001) and for revisions from 20 (range, 5-47) to 38 (range, 11-73; P < .001). Acromial-prosthetic distance showed no significant correlation for active glenohumeral motion. Five shoulders with a deficient teres minor muscle showed no significant decrease of external rotation. Conclusion: Active ROM is better in primary placed prosthesis, and this difference takes place mainly in the glenohumeral joint. In all our patients, Constant scores improved significantly postoperatively. However, we could not find any clinical correlating parameters to explain this difference. © 2011 Journal of Shoulder and Elbow Surgery Board of Trustees.