Kyphosis recurrence after posterior short-segment fixation in thoracolumbar burst fractures

Recurrent kyphosis has been commonly seen after posterior short-segment pedicle instrumentation for a thoracolumbar fracture, but studies on this issue are relatively scarce, and the clinical significance of recurrent deformity is uncertain. No study has addressed the associations between the reduction of a burst fracture vertebra and the final recurrent kyphosis after implant removal. The aim of this study was to investigate the recurrent kyphosis after short-segment pedicle screw fixation in thoracolumbar burst fractures and to evaluate the effect of the degree of a vertebral reduction on the recurrent kyphotic deformity after implant removal.

Twenty-seven patients who had undergone posterior short-segment pedicle screw fixation for thoracolumbar junction burst fractures (T12-L2) were investigated retrospectively. The minimum follow-up period was 2 years (mean 2.7 years). Pain status was evaluated using the Denis pain scale. Changes in the anterior vertebral height ratio, vertebral wedge angle, upper intervertebral angle, lower intervertebral angle, Cobb angle, regional angle, and sagittal index were measured preoperatively, postoperatively, before implant removal, and at final follow-up. The correlation between the reduction of a fractured vertebra and the recurrent kyphotic deformity was also analyzed.

After the initial surgical correction, the reduced vertebral body (VB) height (anterior vertebral height ratio and vertebral wedge angle) remained stable until final follow-up, whereas the intervertebral disc space (the upper and lower intervertebral angles) collapsed, resulting in a progressive kyphotic deformity (Cobb angle, regional angle, and sagittal index). No significant correlation was found between the final kyphosis and pain scale, but the 8 patients with a sagittal index > 15 degrees showed a higher incidence of moderate to severe pain (P3-5 on the Denis pain scale) compared with the remaining 19 patients with a sagittal index < 15 degrees . Significant positive correlation was found between recurrent kyphosis and vertebral wedge angle (r = 0.850, p < 0.001) and the reduced vertebral height (r = -0.727, p < 0.001).

Given that the correction loss occurs primarily through disc space collapse, the amount of the final kyphotic deformity was predictable by the degree of the fractured vertebral reduction as seen on the lateral x-ray study. Surgeons who perform posterior reduction and fixation procedures should pay more attention to reducing the fractured vertebral wedge angle to its intact condition, rather than the segmental angular parameters. If the wedge angle of the fractured VB is unacceptable after reduction, additional reconstruction of the anterior column may be necessary.