Phillips AM, Smart C, Groom AFG. Acromioclavicular dislocation: conservative or surgical therapy. Clin Orthop. 1998;353:10–17.
Among patients with acromioclavicular (AC) dislocation, does surgical intervention produce better outcomes than conservative therapy?
Studies were identified by a MEDLINE search (1966–1997) and a manual search of the reference lists of each relevant study identified. The medical subject heading of acromioclavicular dislocation was used as the primary search term.
The search was limited to English-language journals listed in Index Medicus. Studies were included if they described severely displaced dislocations of the AC joint, mostly characterized as grade III injuries (Allman or Rockwood classification) or if there was at least 1-cm displacement of the clavicle. If more than 1 study included the same group or subgroups of patients, the study with the best assessed methods was used. Studies were divided into 4 classifications: group 1, randomized trials of surgery versus conservative therapy; group 2, nonrandomized trials of surgery versus conservative therapy; group 3, surgical trials only; and group 4, conservative trials only.
Data-extraction and study quality-assessment procedures were not explained in detail. The primary outcome measures were overall outcome, return to work, return to premorbid activities, complications, and radiographic features. Secondary measures were pain, range of motion, and strength. RevMan software (version 1.05; Cochrane Centre, Oxford, UK) was used for statistical analysis.
Specific search criteria identified 600 articles for review, of which 24 met inclusion and exclusion criteria: 2 in group 2, 3 in group 3, 14 in group 4, and 5 in group 4. A total of 1172 patients were represented (surgical treatment = 833, mean = 43.7 months' follow-up; conservative treatment = 339, mean = 60.4 months' follow-up). Both surgically and conservatively treated patients reported similar overall satisfactory outcome (88% surgical versus 87% conservative). Patients with surgical treatment reported longer time to return to work and premorbid activities. Among patients treated surgically, 59% had additional surgery, 6% had wound breakdown, 20% had fixation failure, and 3% reported residual deformity. Only 1% of conservatively treated patients reported wound problems, 6% had additional surgery, and 37% reported residual deformity. In only 1 study did the authors report the incidence of posttraumatic arthritis: 25% among surgically treated and 43% among conservatively treated patients. Analysis of secondary outcomes suggests that both groups had little or no pain (93% surgical, 96% conservative) but more conservatively treated patients had normal to near-normal range of motion (95% versus 86%) and normal strength (92% versus 87%). Conservative treatment of AC dislocations is 21% more likely to result in a satisfactory outcome than surgical treatment (odds ratio = 0.79, 95% confidence interval = 0.36, 1.71). The need for additional surgery is 7.4 times more likely and infection is 3.2 times more likely with surgical management.
These data suggest that the current evidence does not support surgical treatment of grade III AC dislocations with respect to overall patient satisfaction as well as clinical outcomes such as pain, range of motion, and strength.