Long versus short segment with intermediate screw fixation for burst fractures of thoracolumbar junction: radiological and clinical results.

Abstract

Long segment fixation has been frequently used to treat thoracolumbar burst fractures. In our study we want to compare the long and short segment with intermediate screw fixation of thoracolumbar junction burst fractures in relation to radiological and clinical outcomes. We collected the data of 115 patients, with thoracolumbar junction (T11-L2) burst fracture A3 or A4, according to AO Classification. Patients were divided into two groups. Group A was treated by long segment fixation. Group B was treated by short segment fixation. At admission, after surgery, and at 12-month follow-up the patients were radiographically assessed for local kyphotic angle using the Cobb method. Patients were clinically evaluated with the Visual Analogue Scale (VAS) for back pain. The mean difference of the preoperative, immediate postoperative, and 12-month follow-up Cobb angle was significant in both groups (P=0.018). The mean difference of the preoperative and immediate postoperative Cobb angle, Δ1, was significantly greater in group A than in group B (P=0.038), indicating that the Cobb angle correction immediately postoperative was better in patients with double level fixation. The mean difference of the immediate postoperative and 12-month follow-up Cobb angle, Δ2, was significantly greater in group A than in group B (P=0.007), indicating that the maintenance of local Cobb angle correction was better in patients with double level, long fixation. There was no difference in VAS values between group B (single) and group A (double) segment fixation immediately postoperatively (P=0.356) or after 12 months (P=0.147). In A3 and A4 thoracolumbar junction fractures, long segment fixation can correct the local kyphosis Cobb angle and maintain the correction at 12-month follow-up better than short segment fixation with intermediate screws in the fractured vertebra. However, the radiological difference was not predictive of clinical results.