The aim of this study is to decide the optimal level of fusion with comparing the results between the short segment fusion and long segment fusion treated with pedicle screw instrumentation, including fractured vertebra in thoracolumbar junctional fractures.

From February 2000 to November 2009, fifty three patients with junctional fracture of thoracolumbar spine were treated with pedicle screws and posterior fusion at our hospital. They were divided into two groups, the short segment group and long segment group. Preoperatively, immediate postoperative and last follow-up lateral radiological evaluation was done by measuring the correction and loss of segmental kyphosis, wedge angle, body compression rate and instrumented vertebra angle. In addition, operation time and amount of intraoperative bleeding were measured.

There were no significant differences of statistical analysis regarding the radiological variables between the two groups, especially the loss of corrected segmental kyphosis, wedge angle, body compression rate and instrumented vertebra angle (p>0.05). However, operative time in the short segment group (234 minutes) was shorter than the long segment group (284 minutes), and there was statistical significance (p=0.002).

We recommend the short segment transpediculr instrumentation one level above and one level below, including the fractured vertebra for thoracolumbar junctional fracture with 6 points or less of the load-sharing score.