Management of primary anterior shoulder dislocations using immobilization

Reference/Citation : Paterson WH, Throckmorton TW, Koester M, Azar FM, Kuhn JE. Position and duration of immobilization after primary anterior shoulder dislocation: a systemic review and meta-analysis of the literature. J Bone Joint Surg Am. 2010;92(18):2924-2933.

Does an optimum duration and position of immobilization after primary anterior shoulder dislocation exist for reducing recurrence rates?

MEDLINE/PubMed, EMBASE, and Cochrane databases were searched up to December 2009 without limitations. The search terms for all databases used were shoulder AND dislocation and shoulder AND immobilization.

Criteria used to include articles were (1) English language, (2) prospective level I or level II studies (according to Journal of Bone & Joint Surgery guidelines), (3) nonoperative management of initial anterior shoulder dislocation, (4) minimum follow-up of 1 year, and (5) rate of recurrent dislocation as a reported outcome.

A standardized evaluation method was used to extract data to allow assessment of methods issues and statistical analysis to determine sources of bias. The primary outcome was the recurrence rate after nonoperative management of anterior shoulder dislocation. Additional data extracted and used in subanalyses included duration and position of immobilization and age at the time of initial dislocation. Data were analyzed to determine associations among groups using 2-tailed Fisher exact tests. For pooled categorical data, relative risk of recurrent dislocation, 95% confidence intervals, and heterogeneity using the I(2) statistic and χ(2) tests were calculated for individual studies. The Mantel-Haenszel method was used to combine studies and estimate overall relative risk of recurrent dislocation and 95% confidence intervals. The statistical difference between duration of immobilization and position was determined using z tests for overall effect. Pooled results were presented as forest plots.

In the initial search of the databases, the authors identified 2083 articles. A total of 9 studies met all of the criteria and were included in this review. In most of the studies, age was a risk factor for recurrence. Patients less than 30 years of age were more likely to sustain a recurrent dislocation than patients more than 30 years of age. In 5 studies (n = 1215), researchers found no difference in recurrence of shoulder dislocation when immobilized in internal rotation (IR) for less than 1 week (41%, 40 of 97) compared with more than 3 weeks (37%, 34 of 93) in patients less than 30 years of age (P = .52). Authors of 3 studies (n = 289) compared the effect of immobilization in IR versus external rotation (ER), and whereas they found no statistical difference, a trend appeared toward reduced recurrence rates in ER but not IR (P = .07). The rate of recurrent dislocation was 40% (25 of 63) in patients treated with IR sling immobilization and 25% (22 of 88) in patients immobilized in ER.

Overall, the investigators found that younger age (<30 years) was a predictor of recurrent dislocations, immobilization for more than 1 week did not improve recurrence rates, and an apparent trend existed toward decreased recurrence rates with ER rather than IR. According to the review and meta-analysis by Paterson et al, the level of evidence for recommendations regarding optimal duration and position of immobilization to reduce the risk of recurrent dislocation was therapeutic level II. This level of evidence was appropriate because the review included only prospective studies of level I or II and a minimum follow-up of 1 year.