Operative and nonoperative treatment methods of burst fractures were compared regarding canal remodeling. The entire series consisted of 18 patients, with seven in the operative treatment group and 11 in the nonoperative treatment group. All fractures were studied with computed tomography (CT) at the postoperative (operative treatment group) or postinjury (nonoperative treatment group) and the latest follow-up. All patients were followed up for greater than or equal to 18 months. There was no statistical difference between postoperative and postinjury canal areas (p = 0.0859). However, a significant difference was found between the rates of remodeling (p = 0.0059). Although spinal canal remodeling occurred in both groups, the resorption of retropulsed fragments was less favorable in nonoperative treatment group.