Plate Fixation Compared with Nonoperative Treatment for Displaced Midshaft Clavicular Fractures: A Multicenter Randomized Controlled Trial.

BACKGROUND

The use of operative treatment for clavicular fractures is increasing, despite varying results in previous studies. The aim of this study was to compare plate fixation and nonoperative treatment for displaced midshaft clavicular fractures with respect to nonunion, adverse events, and shoulder function.

METHODS

In this multicenter, prospective, randomized controlled trial, patients between 18 and 60 years old with a displaced midshaft clavicular fracture were randomized between nonoperative treatment and open reduction with internal plate fixation. The primary outcome was evidence of nonunion at 1 year. Other outcomes were secondary operations, arm function as measured with the Constant shoulder score and Disabilities of the Arm, Shoulder and Hand (DASH) score, pain, cosmetic results, and general health status. Outcomes were recorded at 6 weeks, 3 months, and 1 year following trauma.

RESULTS

One hundred and sixty patients were randomized. The rate of nonunion was significantly higher in the nonoperatively treated group than in the operatively treated group (23.1% compared with 2.4%; p < 0.0001), as was the rate of nonunion for which secondary plate fixation was performed (12.9% compared with 1.2%; p = 0.006). The rate of secondary operations was 27.4% in the operatively treated group (16.7% for elective plate removal) and 17.1% in the nonoperatively treated group (p = 0.18). Nineteen percent of the patients in the operatively treated group had persistent loss of sensation around the scar. No difference was found between the groups with respect to the Constant and DASH scores at all time points.

CONCLUSIONS

For patients with a diaphyseal fracture of the clavicle displaced at least 1 shaft width, plate fixation improves the chances that the bone will heal; however, the rate of patients who need a second operation is considerable. In addition, the procedure does not improve shoulder function or general symptoms, and it does not decrease limitations compared with nonoperative treatment in a sling.

LEVEL OF EVIDENCE: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.