Finite Element Analysis And Clinical Research On Optimization Of Rod Curvature And Cross-link In Posterior Lumbar Interbody Fusion

ObjecitveCurrently,pedicle screw fixation is a common surgical technique in spine surgery.Although the majority of patients could obtain postoperative improvement significantly when this technique is applied in the treatment of spinal diseases,there have been numerous reports from literatuures on negative surgical outcomes of this technique since it was invented.Therefore,how to optimize surgical techniques to improve clinical outcomes,further to save medical resources and reduce family and social burden,has always been a hot topic and difficult problem in the field of spine surgery.Now there are two easily ignored and controversial issues,subjective or experiencebased rod contouring procedure and the necessity and optimal position of cross-link usage,in Posterior Lumbar Interbody Fusion(PLIF).Based on the two issues,this study conducted numerical simulation and a retrospective study to evaluate the impact of rod curvature or cross-link on PLIF in aspects of biomechanical properties and postoperative outcomes,and optimized rod contouring techniques during posterior spinal surgery to guide clinical practice.The objectives of this study were set forth as below:(1)To explore biomechanical effects of rod curvature on single-segment PLIF,aiming to provide references for appropriate rod curvature during intraoperative rod contouring by finite element analysis;(2)To evaluate the impact of rod curvature on the prognosis of PLIF by a retrospective study;(3)To optimize intraoperative rod contouring procedure to meet clinal requirements;(4)To evluate biomechanical effects of usage and position of cross-links on single-and multisegment PLIF,aiming to clarify the necessity and optimal position of cross-link usage.Method1.Based on the simulation of lumbosacral vertebrae model,finite element analysis(FEA)was adopted to evaluate biomechanical effects of four rod curvatures(RC)on the single-segment PLIF surgery including three kinds of contoured rods corresponding to three commonly-used LL measuring methods(i.e.,centroid,Cobb,posterior tangent)and straight rods.The measurements included Range of Motion(ROM)of the lumbosacral model,intersegmental rotation angle(IRA)of adjacent levels,intradiscal pressure(IDP)in adjacent segmental discs and maximum von Mises stress in fixation(MSF).2.A retrospective clinical study was conducted to compare patient characteristics between lumbar stenosis patients who underwent PLIF with satisfactory postoperative results and those who had poor recovery resulted from adjacent segment degeneration after PLIF and needed revision surgery(i.e.Normal and Revision Groups).Patient demographics(age,gender,height,weight and BMI),surgical characteristics(number of fused levels,surgical time,blood loss and hospital stay)and radiographic parameters(lumbar lordosis[LL],pelvic tilt [PT],pelvic incidence [PI],sacral slope [SS],PI-LL,Cobb angle of fused segments [Cobb],rod curvature [RC],Posterior tangent angle of fused segments [PTA] and RC-PTA)were analyzed.3.By referring to the normal spinal curvature,a convenient and accurate rod contouring procedure was invented by adopting surgical planning and digital projection technology.4.Single-(L4/5)and multi-segment(L3-S1)PLIF surgery models were simulated on the basis of lumbosacral vertebrae model.FEA was adopted to evaluate biomechanical effects of the necessity and position of the cross-link on the single-segment or multi-PLIF surgery.Single-segment PLIF surgery included two cases: With and Without cross-link,and multi-segment PLIF surgery consisted of four cases: Without,Upper,Middle,and Lower cross-links.The measurements included ROM of the lumbosacral model,IRA of adjacent levels,IDP in adjacent segmental discs and MSF.Results1.No difference was observed among four RC models in aspects of ROM of the lumbosacral model,IRA of adjacent levels and IDP in adjacent segmental discs in L4/5single-segmental PLIF surgery under the four actions of flexion,extension,lateral bending and axial rotation.Posterior tangent model had less MSF in flexion,extension and axial rotation compared to the other RC models.Stress concentration in all four RC models was mainly occurred at rods and thread run-out under lateral bending and axial rotation.2.In this retrospective study,53 and 44 patients with lumbar stenosis were identified in normal and revision groups,respectively.Patients in revision group had older median age(63.5 [Q1,Q3 53.75,67] versus 53 [Q1,Q3 43,64],P = 0.002)and suffered more blood loss(325 ml [Q1,Q3 237.5,500 ml] versus 300 ml [Q1,Q3 200,300 ml],P = 0.033)than patients in normal group.No significant difference was observed between normal and revision groups in aspects of gender,height,weight,BMI,surgical time and hospital stay.As for radiographic parameters,RC and RC-PTA were significantly lower in revision group compared with normal group(10.71° ± 7.98° versus 22.04° ± 7.9°,P < 0.001;-8.4° [Q1,Q3-19.43°,-4.8°] versus 0° [Q1,Q3-6°,5.9°],P < 0.001).Multivariate regression analysis revealed that lower age(OR=0.94;95% CI: 0.89-0.99;P=0.0187),higher RC(OR=1.35;95%CI: 1.20-1.51)and lower PTA(OR=0.91;95% CI: 0.85-0.96)were related to higher odds of better surgical outcome.The area under the ROC curve(AUC)for predicting outcomes of surgery by RC classifier was 0.851(0.769-0.932)and the optimal cut-off value was 13.70.Under this optimal cut-off value,the sensitivity was 0.89,specificity was 0.75,the positive predictive value was 0.81,negative predictive value was 0.85,positive likelihood ratio was3.55 and Negative likelihood ratio was 0.16.The AUC for predicting outcomes of surgery by RC-PTA classifier was 0.801(0.713-0.888)and the optimal cut-off value was-1.30.Under this optimal cut-off value,the sensitivity was 0.60,specificity was 0.89,the positive predictive value was 0.86,negative predictive value was 0.65,positive likelihood ratio was5.31 and Negative likelihood ratio was 0.45.3.This study designed an intraoperative rod contouring device for posterior spinal fixation.The intraoperative image of pedicle screw fixiation was imported into the surgical planning system.Then the spinal alignment was corrected and the rod shape was simulated by matching the rod curvature with the normal spinal curvature.The simulated rod equivalent to the size of a real rod was projected to the operating table.Surgical assistants,nurses or technicians could directly perform the rod contouring procedure according to the projection.4.(1)In single-segment PLIF model,there was no significant difference in aspects of ROM of the lumbosacral model,IRA of adjacent levels and IDP in adjacent segmental discs regardless of the application of cross-links.Under the flexion and extension,the MSF in the With case was slightly lower than that in the Without case.However,under lateral bending and axial rotation,a higher MSF was found in the With case compared with that in the Without case.In particular,the difference was more pronounced under lateral bending and axial rotation.High stress areas were mainly concentrated on bilateral rods in lateral bending and thread run-out under the axial rotation.The usage of connection also reduced stress concentration on two lower screws under the axial rotation.(2)In L3-S1 multi-segment PLIF model,a very small difference was found in ROM of the lumbosacral model and IRA of adjacent levels among Without,Upper,Middle,and Lower cases under the flexion,extension or lateral bending.But compared with the case without cross-link,a slightly reduction of ROM of the lumbosacral model and IRA of adjacent levels was observed in the Upper,Middle,and Lower cases under the axial rotation,with more pronounced in Upper case.There was no significant difference in IDP among cases under all actions.Smaller MSF was observed in three cases with cross-link under the axial rotation compared with the case without cross-link,and the Middle and Lower positions had markedly lower values of MSF on the posterior fixation under the lateral bending and axial rotation,respectively.Stress concentration in all four RC models was mainly occurred at bilateral rods and thread runout of upper and lower screws under lateral bending and axial rotation.The usage of a crosslink reduced stress concentration on bilateral rods under the axial rotation.Conclusion1.All four RC models showed similar biomechanical behaviors under four actions,but the posterior tangent model was relevantly superior to other three RC models in the aspect of fixation stress distribution.2.Compared with lumbar stenosis patients who underwent PLIF surgery with satisfactory outcome,those who had poor prognosis resulted from adjacent segment degeneration after PLIF and needed revision surgery had significantly higher age,more blood loss,smaller RC and larger RC-PTA.RC and RC-PTA were demonstrated to be pretty good indicators for predicting postoperative outcomes.3.Surgical planning and projection technology could be used to optimize intraoperative rod contouring procedure to solve clinical issues like subjectivity,uncertainty and poor standardization during the operation.4.The application of cross-links was not beneficial for the single-segment PLIF,while it was found highly advantageous for the multi-segment PLIF in anti-rotation and risk reduction of internal fixation failure.Moreover,the usage of cross-links at the middle or lower positions resulted in less risk of internal fixation failure.