Reverse shoulder arthroplasty (RSA) is a common procedure used to restore function in patients with rotator cuff arthropathy. Few studies have examined shoulder motion after RSA while performing functional activities, and all of them report maximum or end range-of-motion (ROM) without continuous kinematic data. The study aimed to present a comprehensive description of functional motions in RSA patients, determine the repeatability of motion between and within subjects, and examine correlations between motion and patient-reported outcomes (PROs). It was hypothesized that increased range-of-motion would correlate with better patient-reported outcome scores.
Twenty-six patients received RSA. Each subject completed 5 shoulder motions for at least 3 repetitions each: abduction, hand-to-head, hand-to-back, internal/external shoulder rotation with the arm in 90° abduction, and circumduction. Reflective markers placed on the torso, shoulder and humerus were tracked using conventional motion capture and used to calculate shoulder abduction, plane of elevation, and internal/external rotation for each motion. Total ROM, maximum/minimum ROM, and continuous kinematic waveforms were calculated, and correlations between motion and implant characteristics as well as PROs of ASES, Constant-Murley Score (CMS), VAS, and Brophy scores were determined.
The largest component of motion during each motion was: 102.8 ± 24.9° of abduction during the abduction motion, 112.7 ± 62.4° of shoulder rotation during the hand-to-head motion, 73.2 ± 25.6° of internal/external shoulder rotation during the rotation motion, and 107.7 ± 29.2° in plane of elevation during the hand-to-back motion. During circumduction, the shoulder moved through 101.0 ± 13.6° of abduction, 101.5 ± 21.9° of plane of elevation, and 112.4 ± 17.0° of rotation. Greater abduction was associated with increased CMS during abduction (r = .467, p = .038) and hand-to-back motions (r = .456, p = .043). Greater abduction during hand-to-back motion also correlated with better ASES scores (r = .573, p = .008). Increased humeral retroversion was associated with lower abduction (r = -.47, p = 0.042), higher plane of elevation (r = -.496, p = .031), and more shoulder rotation (r = -.469, p = .043) during circumduction. Humeral retroversion also was correlated with more abduction during hand-to-head motion (r = -.512, p = .025). No correlations between ROM and tilt, lateralization, or eccentricity of the implant were found.
Increased abduction ROM during the tasks of abduction and hand-to-back motion was associated with better PROs. The functional movements showed greater shoulder rotation ROM in the hand-to-head compared to hand-to-back motion, while the opposite trend was seen for plane of elevation indicating different strategies are used in these functional movements. Additionally, circumduction is a complex cross adduction motion that has not previously been described and was shown to have a correlation with humeral retroversion for all measured motions. This indicates that humeral retroversion may be an important factor that determines ROM for various movements.
Improving abduction during functional tasks following RSA may be necessary for optimizing outcomes. Humeral retroversion may be important for determining total ROM for complex functional motions.