Do patients treated surgically following acute anterior shoulder dislocations have less instability than those treated nonsurgically?
In this meta-analysis of only 5 studies, patients 15 to 39 years of age with acute anterior dislocations who were treated surgically were less likely to have recurrent instability than those who were treated nonsurgically.
Meta-analysis (randomized controlled trials)
These authors searched 3 databases to identify randomized trials that compared surgical and nonsurgical approaches to managing patients with their first acute anterior shoulder dislocation. To be included, the studies had to report the frequency of recurrent instability and be published in English in a peer-reviewed journal. The authors ignored the potential for publication bias by not searching the gray literature or by not formally assessing it statistically or graphically. They included 5 trials with 259 patients between the ages of 15 and 39 years; nearly 90% were men. The follow-up intervals ranged from 2 years to 10 years. Although the authors report the overall study quality was good, all were unmasked, and only one included an outcome assessed by study personnel who were masked to treatment allocation. Four of the studies used similar surgical approaches; in all the studies, the control group was treated with immobilization. The physical therapy protocols for each treatment group were identical within each study, though there were slight differences among the 5 studies. Only 6.4% of the surgically treated patients had recurrent instability compared with 46.6% of those treated nonsurgically (number needed to treat = 3; 95% CI 2 - 4) with no statistical heterogeneity. Although the authors report that each study provided data on complications, they did not report the pooled event rate. However, the authors specifically comment that one surgically treated patient developed adhesive capsulitis that resolved within 18 months, and another patient developed septic arthritis requiring arthroscopic irrigation and antibiotics. It is likely that with only 259 total patients, this study lacks the power to detect rare events.
Henry C. Barry, MD, MS
Michigan State University
East Lansing, MI