Shoulder Proprioception in Rehabilitation

Abstracts & Commentary

Synopsis: The study prospectively evaluated shoulder proprioception in patients with traumatic anterior instability who underwent anterior shoulder repair.

Sources: Edmonds G, et al. The effect of early arthroscopic stabilization compared to non-surgical treatment on proprioception after primary traumatic anterior dislocation of the shoulder. Knee Surg Sports Traumatol Arthrosc. 2003;11:116-121; Zuckerman JD, et al. The effects of instability and subsequent anterior shoulder repair on proprioceptive ability. J Shoulder Elbow Surg. 2003;12(2):105-109.

In 1994, Lephart and colleagues first compared the shoulder proprioception of patients who underwent surgical repair to that of healthy subjects, demonstrating that anterior shoulder stabilization procedures appeared to restore proprioception.1 Since Lephart et al did not perform a prospective, randomized trial, nor did they measure the same patients both before and after surgery, the effect of these procedures on proprioceptive function remains uncertain. The 2 studies reviewed here prospectively evaluated shoulder proprioception to address this question. Edmonds and colleagues studied patients with initial traumatic anterior shoulder dislocation treated with arthroscopic stabilization, and Zuckerman and colleagues used recurrent traumatic anterior instability patients treated with open stabilization.

Edmonds et al performed a prospective, randomized study comparing a nonoperative control group (n = 13) that underwent sling immobilization for 3 weeks followed by rehabilitation and another group who had first time, traumatic anterior dislocations (n = 11) treated by an arthroscopic, transglenoid multiple suture technique followed by the identical rehabilitation program. Measurements were performed (mean, 19 months; range, 3-36 months) with the use of a modified CybexTM dynamometer. Both threshold to detection of passive motion (TTDPM) and detection of passive motion (RPP) were measured at 90º abduction at both 30º and 60º of external rotation. Both tests were performed at 2º per second with the patient in the supine position. The re-dislocated subjects (1 of 11 surgical, 3 of 13 conservative) were not excluded from the study. Results showed that there were no significant differences between these 2 groups in either TTDPM or RPP.

In the second study, Zuckerman et al prospectively evaluated shoulder proprioception both before and after anterior shoulder stabilization surgery in patients with recurrent traumatic anterior instability. Thirty consecutive patients with multiple episodes of anterior dislocations (range, 5-12) were evaluated. Both TTDPM and RPP for the direction of shoulder flexion, abduction, and external rotation position were measured a week before surgery. Both tests were performed at 2º per second. Flexion and abduction were tested with patients in a seated upright position, whereas external rotation was tested in a standing position. All patients had open anterior capsulorrhaphy combined with capsulolabral repair. The subjects were retested both at 6 and 12 months after surgery.

Before surgery, significant deficits in both TTDPM and RPP were found when comparing the affected side to the unaffected side. Six months after surgery, RPP showed improvement of approximately 50% but still differed significantly from the unaffected side, while TTDPM was no longer significantly different from the uninvolved shoulder. One year after surgery, both TTDPM and RPP were equivalent to those of the uninvolved shoulder.

Comment by Ed McFarland, MD, and Atsushi Yokota, MD, PhD

The importance of these papers lies in the fact that they are among the first to have a prospective design, and the Edmunds paper was the first to have a control group. These studies demonstrated that proprioception may recover with or without surgery for anterior instability. However, this conclusion warrants further study due to limitations of these studies.

The study by Edmonds et al provides a valuable contribution to this field by suggesting that the surgical re-tensioning of the anterior capsule for patients with initial dislocation may not be a necessary condition for the restoration of shoulder proprioception. They found that their results did not change whether the patients who failed treatment (9% of surgical, 23% of conservative) were included in the statistical analysis. This suggests that either the measures were not discriminating or that the numbers were not adequate. Another weakness of this study was that Edmonds et al did not do a side-to-side comparison between the normal and unstable side. The published rate of redislocation after conservative therapy (23%) in this study was low compared to the other reports and may indicate that the patient groups were not equivalent. The use of a large number of subjects in the next phase of this study may resolve some of this uncertainty.

The strength of the study by Zuckerman et al was that it is a prospective study of shoulder proprioception both before and after surgery. Zuckerman et al confirmed the findings of previous studies, which have shown that shoulder proprioception is compromised in patients with recurrent anterior instability. In addition, they showed that proprioceptive function was restored equivalent to the unaffected side after the surgical procedure followed by a standard postoperative rehabilitation. The limitation of this study is that it remains uncertain whether the major cause of restoration of shoulder proprioception is surgical stabilization or postoperative intensive rehabilitation. The study by Edmonds et al would suggest that a nonoperative control group should be used for comparison. Furthermore, the effect of prolonged morbidity on the proprioceptive function is still uncertain. Although this study did not demonstrate that patients with more episodes had more loss of proprioception, it is possible that repetitive dislocation causes further damage to the mechanoreceptors surrounding the shoulder, which could deteriorate proprioceptive function.

The results of both studies suggest an important role of postoperative rehabilitation for restoration of shoulder proprioception. The recent study by Swanik and associates clearly demonstrated the efficacy of plyometric training for improvement of shoulder proprioception.2 Although it is still unknown whether the deficit of proprioception is a part of a primary pathology or a secondary phenomenon, restoration of proprioception is an important part of therapy for traumatic shoulder instability. Therefore, intensive rehabilitation should be performed for all patients.

When one tries to compare the results of previous studies concerning shoulder proprioception to those studies, one difficulty is interpreting the numerous parameters used for the measurement of proprioception, such as patients’ position (eg, upright position or supine position), starting position (eg, arm-at-side position or abduction and external rotation position), test speed, and method of immobilization of extremity. A major concern is that these factors could significantly affect the reported values for many of the measurements. Future studies will hopefully determine the most accurate and reproducible methods for measurement of shoulder proprioception.

Dr. McFarland, Associate Professor, Department of Orthopaedic Surgery, Johns Hopkins School of Medicine, Baltimore, MD, is Associate Editor of Sports Medicine Reports. Dr. Yokota is a Research Fellow in Orthopaedic Surgery at Johns Hopkins University.


1. Lephart SM, et al. Proprioception of the shoulder joint in healthy, unstable, and surgically repaired shoulders. J Shoulder Elbow Surg. 1994;3(6):371-379.

2. Swanik KA, et al. The effects of shoulder plyometric training on proprioception and selected muscle performance characteristics. J Shoulder Elbow Surg. 2002; 11(6):579-586.