Abstract
Introduction
Our previous study suggested that vertebral fracture nonunion(VFNU) could be classified into stable and unstable types based on whether there are fracture fragment in the middle column of fractured vertebrae, and we hypothesize that vertebral augmentation could restore the stability of fractured vertebrae in unstable type of VFNU, this study is to determine the effect of percutaneous vertebral augmentation on stable and unstable vertebral fracture nonunion by three-dimensional finite element analysis.
Material and Methods
An elderly senior, who suffers from osteoporosis, but owns the normal alignment and profile of thoracolumbar spine was included into this finite element study and accepted CT examination. The finite element three-dimension model of physiological thoracolumbar spine (T11-L3) was established and investigated with Mimics, Rapidform, and Abaqus softwares, then wedge-shaped osteotomy was performed in L1 vertebral body to stimulate intravertebral bone defect, which is the main feature of VFNU, to make the pathological finite element model of type A VFNU, which is named pathological model A. Furthermore, the posterior–inferior region of vertebral body was cut to build a piece of separated fracture, with the purpose of simulation of the pathological feature of type B VFNU and establishment of corresponding Finite element model which is called as Pathological model B. After that, imitated bone cement augment was performed in pathological model A and B, distributing cement in the whole intervertebral cleft or 5 mm ahead posterior edge of vertebral body, to establish four therapy finite element model, which were called as therapy model A1, B1, A2, and B2. All of the models were calculated under the same axial loads of 400 N, to investigate the validity of physiological model and compare the mechanical status of different models.
Results
The physiological finite element model owns good mechanic conformity with the reported models. Compared with physiological model, there was notable stress, strain concentration in the L1 vertebral body of pathological model A, mainly distributing beside the posterior cortex and in the posterior inferior part of the vertebral body, which indicated that there was a risk of further fracture in this model. Although there was also stress and strain concentration in pathological model B, the extent and level of stress and strain were lower than the corresponding value in the pathological model A, furthermore, concentration mainly located in the posterior–inferior fracture piece, which implied that the separated posterior–inferior fracture piece of vertebral body intended to dislocate, and the stress, strain concentration were dispersed after pathological Model A became Model B. According to calculation of therapy model A1, A2, the stress and strain concentration could be dispersed with bone cement augmentation, and the mechanic conduction function of L1 vertebral body could be restored after imitated PVP, which indicated that vertebroplasty could be used effectively in VFNU patients with reliable mechanic stability. However, on the basis of the calculation of therapy model B1 and B2, the tendency of stress, strain concentration, and displacement of posterior–inferior fracture piece could be found, even though the whole concentration could be dispersed by cement augmentation, which implied that there was risk of dislocation of posterior–inferior fracture piece after bone cement augmentation in pathological model B.
Conclusion
A favorable mechanical performance was revealed in stable VFNU after PVA, but the risk of redislocation of posterior–inferior bony fragments might be increased result from PVA.