Following fracture reduction and initial reconstitu-tion of spinal alignment, 1055 of correction over time is frequently observed after posterior instrumentati-on. The degree of stabiiity to provide a favorable en-vironment for protection of initial correction is not known.

A total of 40 patients with thoracolumbar burst fractures were divided into three groups. Posterior instrumentation was used in all groups. Group 1 inc-luded 8 patients treated by pedicle screw fixation, group 2 included 16 patients treated by hook fixation and group 3 included 16 patients treated by combi-ned hook and pedicle screw fixation. The preoperati-ve, early postoperative and 1-year follow-up lateral spinal radiographs were evaluated by measuring the local kyphosis angle (LKA) and the percentage of an-terior body height (%ABC).

Correction loss of LKA was 8.8° in group 1, 6.9° in group 2 and 3.2° in group 3. For protecting the initial correction of LKA, group 3 was superior to the ot-her groups (P<0.05) and group 1 and group 2 were the same (P>0.05). Correction loss of %ABC was 7.4 in group 1, 5.6 in group 2 and 3.5 in group 3. For protecting the initial correction of %oABC, group 3 was superior to the group 1 (P<0.05), and other groups were the same (P>0.05). Failure rate was 37.5% in group 1, 12.5% in group 2 and 6.2% in group 3.

For maintenance of initial carrection, posterior instrumentation with combined hook and pedicle screw fixation was superior to posterior instrumentation with hook fixation and posterior instrumentation with pedicle screw fixation.

Keywords: Thoracolumbar burst fracture, posterior fixation, instrumentation, stabiiity.