The Three-dimensional Finite Element Analysis And Related Clinical Research Of Congenital Scoliosis Coronal Lumbosacral Region Flexibility

Objectives To establish normal lumbosacral and different flexibility of congenital scoliosis coronal lumbosacral region three-dimensional finite element model, and comparative analysis three kinds of models in mechanical changes under different load. Making a retrospective study between preoperative coronal angle changes in the lumbosacral region and postoperative trunk shift after congenital scoliosis surgery. The finite element model of effective were verified by the clinical study results and to evaluate feasibility and application value of preoperative coronal lumbosacral region intervertebral angle changes in making surgery planning by biomechanical and clinical practice.Methods Select one normal subject and two congenital scoliosis patients with different coronal lumbosacral region flexibility. Get DICOM image data by CT scanning for lumbosacral region. Import it into 3d reconstruction software MIMICS, according to the selection of CT image gray value of each group, extract the corresponding organization through the threshold and generate the 3d model. After simplified handling by GEOMAGIC, we import the 3d lumbosacral region model into the finite element software ABAQUS to make a mechanics analysis under different load condition. We retrospectively reviewed the records of 921 patients who underwent posterior congenital scoliosis correction and internal fixation at our institution between January 2010 and June 2014. Patients were included in the analysis if the postoperative coronal trunk shift(TS was defined as the vertical distance from the midpoint of the sacrum to the C7 plumb line) was ≥ 20 mm immediately after posterior scoliosis correction, and complete follow-up data was available. Patients who had undergone an osteotomy were not included to exclude the impact of differences in the extent and method of osteotomy on postoperative spinal deviation. Standing anteroposterior (AP), lateral, and bending radiographs of the full spine were obtained preoperatively, and standing AP and lateral radiographs of the full spine were obtained postoperatively and at final follow-up. The coronal trunk shift, lumbosacral region(L4-5 and L5-S1) intervertebral interspace angles and Cobb angle on preoperative, postoperative, and final follow-up radiographs were also recorded. The intervertebral interspace angle change was defined as the sum of the absolute values of the L4-S1 intervertebral angle changes in the coronal plane on lateral-bending radiographs of the lumbosacral region. Groups were defined by two methods:lumbosacral region preoperative intervertebral interspace angle change≥ 10°(A-1) and<10°(A-2), and TS<20 mm(B-1) and≥20 mm(B-2) in the coronal view at final follow-up. All statistical analyses were performed with statistics software. Finally, to evaluate feasibility and application value of preoperative coronal lumbosacral region intervertebral angle changes in making surgery planning by using different mechanical characteristics of finite element model and retrospective clinical study results.Results There were no change in the stress distribution when compensatory model and uncompensatory model under different lateral force, but displacement changes were significantly, especially in compensatory model. The stiffness is also much smaller in compensatory model which mean it is easier to deformation for compensatory model. In a retrospective clinical study, A median negative correlation were found between TS at the final follow-up and total variance of bending (P= 0.001,r = -0.51). The sum of absolute value of lumbosacral region preoperative intervertebral interspace angle change was 10.3±6.1°.Thus, we divided patients into two groups: lumbosacral region preoperative intervertebral interspace angle change≥10°(A-1) and <10°(A-2). TS at the final follow-up was significantly greater in the A-2 group than the A-1 group (P= 0.001). When patients were classified by TS at the final follow-up (< 20 mm vs.≥ 20 mm), the bending L4-L5 change and bending L5-S1 change were all significantly higher in the< 20 mm group than in the ≥ 20 mm (P< 0.01). There were significant differences in the Cobb angle change of congenital scoliosis thoracic and lumber curves. Significant differences were happended in the A-2 and B-1 thoracic curve at final follow-up, but no significant differences in A-1 and B-2. The lumber curve change of two groups were quite similar with the change of thoracic curve.Conclusion Postoperative TS which will not compensate spontaneously can be predicted by measurement of the preoperative intervertebral interspace angle in the congenital scoliosis lumbosacral region.