Comparison of short-segment versus long-segment fixation for the treatment of thoracolumbar burst fracture: a meta-analysis

Background: Thoracolumbar burst fracture is one of the most common spinal traumas. It was still indeterminate whether short-segment (SS) fixation was more effective and preferable than long-segment (LS) fixation. We therefore conducted this meta-analysis to assess the comparative safety and efficacy of SS and LS fixation for the treatment of thoracolumbar burst fracture. Methods: A literature search was conducted between January 1990 and September 2014. The sources of electronic searching included EMBASE, PubMed, Google Scholar, Ovid, Cochrane Library and Springer. Study eligibility was based on predefined criteria. The data were extracted by two authors independently and were managed using the Review Manager (RevMan) 5.2. A meta-analysis was conducted under the suggestion of PRISMA guideline standards. Results: Finally only 8 eligible articles were included in the current study, including a total of 455 thoracolumbar burst fracture patients who underwent the treatment with SS fixation (n=239) and LS fixation (n=216). The outcomes of this study indicated that final outcome in regard to implant failure, anterior body height loss (ABHL), cobb angle (CA) and restoration ratio of compromised canal is better in the LS fixation group than those of SS fixation group. However, SS group significantly reduced surgical time when compared with LS group. There was no significant difference regarding blood loss, complication rate, Sagittal Index (SI) and kyphotic angle (KA). Conclusions: The main results of radiographic indexes and implant failure are better in the LS fixation group than those of SS fixation group. Nevertheless, LS fixation prolonged the duration of operation. Clinical outcomes suggested that there was no difference between the SS and LS fixation. The choice of the appropriate method has to be made cautiously and on an individualized basis.