• OBJECTIVE
    • To determine whether radiographic findings associated with thoracolumbar burst fractures could be predictors of failure of short-segment posterior instrumentation with insertion screw at the fracture level (SSPI-f).
  • METHODS
    • Seventy-five patients with thoracolumbar burst fracture surgically treated by SSPI-f were enrolled in the study and divided into 2 groups: a reduction group (n = 46) and a failed-reduction group (n = 29). Radiographic data including local kyphosis, Cobb angle, anterior vertebral height, posterior vertebral height (PVH), anterior/posterior vertebral height ratio, interpedicle distance (IPD), bony compress area, bony fracture area, and compress-fracture area of the fractured vertebra and clinical data including age and neurologic function were also analyzed. t test, Pearson χ2 test, and binary logistic regression were performed to compare the values.
  • RESULTS
    • The PVH in the failed-reduction group was smaller than that of the reduction group (83.5% ± 7.2% and 89.1% ± 5.4%, respectively) (P = 0.001). The IPD differed between the reduction and failed-reduction group (18.0% ± 4.1% and 25.8% ± 7.1%, respectively) (P < 0.001). There was a statistical difference between the 2 groups in delayed time before surgery (P = 0.008). There was a significant difference of bony fracture area and compress-fracture area of the fractured vertebra between the failed-reduction and reduction group (both P < 0.001). Binary logistic regression showed that IPD was a risk factor of reduction failure of SSPI-f (P = 0.001).
  • CONCLUSIONS
    • These results showed that increased IPD was a risk factor of failed-reduction of SSPI-f in managing thoracolumbar burst fractures, particularly for patients with neurologic deficit, whereas local kyphosis, Cobb angle, anterior vertebral height, PVH, anterior/posterior vertebral height ratio, bony compress area, bony fracture area, and compress-fracture area of the fractured vertebra were not.