Abstract: Objective To observe the long-term biomechanical conformation after thoracolumber spine fracture cured by posterior approach short-segmental fixation with fusion or non-fusion technique,and to compare long-term clinical outcomes of the two methods.Methods From February 2004 to December 2006,79 patients with one level thoracolumber vertebral fracture(TVF) were treated with two operation methods:posterior approach short-segmental fixation(control group,non-fusion group) and posterior approach short-segmental fixation combined with grafting bone for fusion(experiment group,fusion group).All cases were followed up 1 week,3 months,1 year post-operation and 3 months after the fixation being removed.Results Before operation,there were no statistical significances between the two groups in the height of the anterior vertebral body,posterior vertebral body,and the cobb's angle,as well as the time point of 1 week after operation and 3 months after operation(all P > 0.05).One year after the operation,there were statistical significances in the height of the anterior vertebral body and the cobb's angle(P < 0.05).There were statistical significances between the two groups in the height of the anterior vertebral body(non-fusion:83.3±5.3,fusion:94.2±3.1),the cobb's angle(non-fusion:9.3±5.8,fusion:2.4±3.7) 3 months after the fixations were removed out(P < 0.05).Conclusion The effect of posterior approach short-segmental fixation combined with grafting bone for fusion is better than that of posterior approach shortsegmental fixation.