Vertebral body spread in thoracolumbar burst fractures can predict posterior construct failure.

Background Context: The load sharing classification (LSC) laid foundations for a scoring system able to indicate which thoracolumbar fractures, after short-segment posterior-only fixations, would need longer instrumentations or additional anterior supports.

Purpose: We analyzed surgically treated thoracolumbar fractures, quantifying the vertebral body's fragment displacement with the aim of identifying a new parameter that could predict the posterior-only construct failure.

Study Design: This is a retrospective cohort study from a single institution.

Patient Sample: One hundred twenty-one consecutive patients were surgically treated for thoracolumbar burst fractures.

Outcome Measures: Grade of kyphosis correction (GKC) expressed radiological outcome; Oswestry Disability Index and visual analog scale were considered.

Methods: One hundred twenty-one consecutive patients who underwent posterior fixation for unstable thoracolumbar burst fractures were retrospectively evaluated clinically and radiologically. Supplementary anterior fixations were performed in 34 cases with posterior instrumentation failure, determined on clinic-radiological evidence or symptomatic loss of kyphosis correction. Segmental kyphosis angle and GKC were calculated according to the Cobb method. The displacement of fracture fragments was obtained from the mean of the adjacent end plate areas subtracted from the area enclosed by the maximum contour of vertebral fragmentation. The "spread" was derived from the ratio between this subtraction and the mean of the adjacent end plate areas. Analysis of variance, Mann-Whitney, and receiver operating characteristic were performed for statistical analysis. The authors report no conflict of interest concerning the materials or methods used in the present study or the findings specified in this paper. No funds or grants have been received for the present study.

Results: The spread revealed to be a helpful quantitative measurement of vertebral body fragment displacement, easily reproducible with the current computed tomography (CT) imaging technologies. There were no failures of posterior fixations with preoperative spreads <42% and losses of correction (LOC)<10°, whereas spreads >62.7% required supplementary anterior supports whenever LOC>10° were recorded. Most of the patients in a "gray zone," with spreads between 42% and 62.7%, needed additional anterior supports because of clinical-radiological evidence of impending mechanical failures, which developed independently from the GKC. Preoperative kyphosis (p<.001), load sharing score (p=.002), and spread (p<.001) significantly affected the final surgical treatment (posterior or circumferential).

Conclusions: Twenty-two years after the LSC, both improvements in spinal stabilization systems and software imaging innovations have modified surgical concepts and approach on spinal trauma care. Spread was found to be an additional tool that could help in predicting the posterior construct failure, providing an objective preoperative indicator, easily reproducible with the modern viewers for CT images.