| Background/PurposesThe interface coverage of the tibial component in total knee arthroplasty is affected by factors such as bone preparation,implantation technique and prosthesis design,however,there is a lack of systematic research regarding the intrinsic relationship among these factors and the stress distribution of the prosthesis system influenced by multi-factors.Also,the design features of the current mainstream prosthesis system are mostly based on research data from Western population,lacking the anatomical parameters of Chinese and the basis for the prosthesis design improvement.Therefore,a 3D model database of the lower extremities based on the local population was established and virtual surgery was carried out in order to evaluate the effect of tibial component posterior slope and rotational alignment on tibial coverage.Then the anatomical features of the tibia specimens from the local population on the osteotomy level were obtained by graphic fitting and the image registration algorithm was used to solve the optimal registration problem for general polygon extracted from the interface contour.At last,in orderto provide theoretical basis for clinical surgery and prosthesis design optimization,the finite element analysis was used to further explore the stress distribution in the prosthesis system under osteotomy with different posterior slope angles and different interface coverage conditions.Methods1.Full-length CT scan of lower extremities,MRI scan of both knees in one normal adult and CT scan of knees of three fresh corpses were conducted to obtain two-dimensional images.Then these images were saved as DICOM files and loaded into Mimics 17.0 for further reconstruction of the lower extremities and simulation model of the knee including bone,meniscus and ligament.During the reconstruction process,by referencing the default bone tissue threshold,reconstruction thresholds of(+50)HU,(-50)HU and(-100)HU are additionally selected to reconstruct corresponding 3D model in order to evaluate the effect of segmentation threshold selection on the reconstruction results.2.The tibial osteotomy was performed on 6 knee joint specimens of3 fresh corpses,and the geometric parameters of the resection plane on the osteotomy level of the tibial plateau and the humeral head were measured respectively.The data was compared with corresponding parameters obtained by 3D model reconstruction based on CT scan to evaluate the accuracy of 3D model measurements.3.CT scans of the lower extremities from 55 normal adults(101knees included)were performed and 3D model was reconstructed.The virtual osteotomy with high accuracy was conducted using the space vector method and the osteotomy contour under different posterior slope(0,3,5,7 degree)was obtained.The baseplates from Gemini prosthesis system were digitalized using X-ray scan.The interface coverage of different zones,under different posterior slope and specific alignment with different tibial baseplates,was achieved using manual match and statistically analyzed.4.The applicability of the Gemini and [E]8 prosthesis system to the local population was analyzed using measurements of the general anatomical features of the tibial 3D model specimens at the osteotomy level which was characterized by graphic fitting.The coherent point drift algorithm is used to solve the optimal registration problem for general polygon extracted from the interface contour,and the free rotational alignment and positioning of the four tibial prostheses under maximizing coverage are analyzed.5.Standard virtual osteotomy was carried out to the 3D reconstruction model of the lower extremities and simulation model of the knee and the Open Knee prosthesis system was implanted afterwards simulating normal total knee arthroplasty.All the components were loaded into Geomagic 2012 for model optimization and solid.The materials,load and constrains were all defined after the geometries weremeshed in Hypermesh 14.0,then the finite element static analysis model was established.6.By retaining and removing part of the attachment structure of the natural knee,implanting prosthesis to knee joint model with different posterior slope(0,3,5,7 degree)and adapting tibial coverage,the effect of the attachment structure on the stress distribution of the natural knee was analyzed and the comparison of stress distribution in knee prosthesis system under different posterior slope and bone-prosthetic interface coverage was made in Optistruct 14.0.Results1.The surface area of the reconstructed model gradually increased with the increase of the threshold value.There was no statistically significant difference between the default threshold group and the ±50HU threshold group,but the difference comparing with the-100 HU group was significant.The Intraclass correlation coefficient showed that the coefficient was 0.991 between the actual measurement of cadaver specimens and the CT reconstruction measurement,and the average error was 1.34±0.97%.2.The tibial mean cross sectional area(CSA)increased along with the increasing of slope angle.The tibial mean CSA,regardless of slope angle,was significantly greater in males as compared with females.Whatever the prosthesis type,there was a significant difference incoverage between Akagi and Insall alignments with a posterior slope from 0 to 7 degrees as the differences ranged from 1.26 to 2.22%.The coverage increased significantly from 85.19 to 87.22% and form 83.21 to85.96% for Insall and Akagi alignment,respectively.Besides,slopes of 5and 7 degrees provided a significantly higher coverage rate compared to slopes of 0 and 3 degrees.When using the anatomical and symmetrical component with a slope of 5 and 7 degrees,the difference between the underhang and overhang was significant in all the anteromedial,anterolateral,posteromedial and posteromedial zone except for the posteromedial zone using anatomical component with a slope of 7degrees.3.The tibial baseplate coverage was greater in males as compared with females.This difference was statistically significant except for the Akagi alignment with a slope of 0 degree for both symmetrical and anatomical design and the Insall alignment with a slope of 0 degree using symmetrical design.Regardless of the prosthesis type,the subgroup analysis showed a significantly smaller CSA of posterior cruciate ligament zone and larger underhang in the posteromedial zone for males as compared with females when aligning to the Insall line with slopes of5 and 7 degrees.Also,anteromedial underhang was more and overhang less in females as compared with males.With a slope of 7 degrees,the symmetrical design showed significantly more anterolateral overhang infemales as compared with males,while the anatomical design showed more anteromedial overhang in males as compared with females.4.According to the 3D reconstruction model of lower extremities based on the local population,the average mediolateral diameter(ML)of all tibia resection plane was 77.7±6.6mm,the average anteroposterior diameter(AP)was 46.8±4.0mm,the average medial plateau anteroposterior diameter(MAP)was 46.2±4.2mm,and the average lateral plateau anteroposterior diameter(LAP)was 44.5±4.1mm.The measurements for males were significantly larger than for females,but the average ML/AP ratio were 1.66±0.04,and the average MAP/LAP ratio was 1.04±0.07,no significant difference between male and female was found.The MAP was 1.8±3.0 mm longer than the LAP on average with a statistically significant difference,and there was a significant positive correlation between them.In addition,there were 15 cases of the lateral plateau with the anteroposterior diameter larger than the medial,and 24 cases of the two sides had the anteroposterior diameter difference less than 2 mm,which accounted for 37.9% of the total.5.Based on local population,the anatomical design showed higher applicability than the symmetrical design on MAP,LAP MAP/LAP,ML,AP and ML/AP.Regression analysis showed that for tibial baseplate of which the LAP was less than 45-50 mm,the difference between its MAP and LAP should be enlarged.On the contrary,for the LAP which waslarger than 45-50 mm,the difference should be reduced accordingly.Also,the current tibial baseplate should reduce the absolute value of the ML and its ratio to AP(currently 1.8)to more suitable for the local population.6.When placed for maximal tibial coverage using algorithm registration,only 23% of the tibial components were malrotated and the actual numbers of malrotated subjects for each design were similar.The average tibial coverage,in the maximizing situation,achieved across all specimens and implants was 85.62%±3.65%.The Gemini system had a worse performance on coverage comparing to the [E]8 system.The mean degree of rotation related to Insall line was-0.73°±4.53° for all subjects.The average percentage position of the implant AP axis was26.95%±14.71% of the full patellar tendon from the medial edge,which was located between axis medial 1/3 and axis medial 1/6 of the patella tendon,but the individual differences were large.7.In the natural standing position,the stress of the femur was stable with or without collateral ligaments and the fibula,the local maximum stress was located in the anterior medial aspect of the shaft on the distal third.For the tibia without collateral ligaments and the fibula,the main stress was in the medial aspect of the shaft on the distal third.The maximum stress and concentration area of the tibial plateau and meniscus was mainly on the medial side.The fibula had an influence on the stressdistribution of the tibial shaft,but the presence of structures regarding the collateral ligaments and the fibula had little effect on the stress distribution of the joint space and interface.8.In the virtual total knee arthroplasty using a simplified knee model,when posterior slope increasing from 0 to 7 degrees,the peak Von Mises stress of the femoral prosthesis was maximum 123 MPa at slope of 3degrees,and the minimum 15 MPa at slope 5 degrees.With the increasing of the posterior slope,it showed a tendency that increased at first and then decreased,concentrating on the lateral side.The peak Von Mises stress of the bearing increased from 11 Mpa to 29 Mpa,the distribution was stable and concentrated on the posterolateral side but the number of stress concentration point decreased.The peak Von Mises stress of the tibial prosthesis increased from 15 Mpa to 17 Mpa,the distribution was stable and concentrated on the posterolateral side between the prosthesis and the bone rather than the bearing,the number of stress concentration point decreased as well.The peak Von Mises stress in the prosthesis system was much larger than that of the femur and tibia and was most affected by the femoral prosthesis.9.The enhanced bone-prosthesis interface coverage was simulated by sizing up the bearing and the tibial prosthesis by 3%,the results showed that the peak Von Mises stress of the femoral prosthesis decreased from 123 Mpa to 44 Mpa only under the slope of 3 degrees.Thecorresponding change of the bearing and the tibial prosthesis was a slight rise instead,and the distribution and number of the concentration point was not significantly changed.Conclusion1.The reconstruction results of the 3D model are stable only when the segmentation threshold selection is limited to a small range.Computer-aided and manually modified threshold segmentation is able to reconstruct a realistic and reliable 3D model.2.The virtual surgery can be realized with high accuracy by applying space vector calculation.The interface coverage increased along with the increasing of the slope angle from 0 to 7 degrees and the current tibial baseplate design has better tibial coverage when aligning to the Insall line with slopes of 5 and 7 degrees.There is a significant mismatch difference between males and females,gender difference should be taken into account and given priority for prosthesis design.3.The contours of the resection plane with posterior slope of 7degrees based on the tibial specimens from local population present large individual difference.In order to be compliant with the anatomical features of the local population,a tibial prosthesis design differentiation should be focused on the variety of the baseplate among different size numbers and the reduction of the absolute mediolateral diameter.4.Image registration algorithm is helpful to contour match betweenbaseplate and tibial resection plane.With current prosthesis design,approximating the maximum tibial coverage in total knee arthroplasty does not necessarily result in implant malrotation.The tibial baseplates has better performance on coverage when the anteroposterior axis positioned between the medial 1/3 and medial 1/6 of the tibial tuberosity.5.With the increasing of the posterior slope from 0 to 7 degrees,the stress distribution of the prosthesis shows a tendency that increases at first and then decreases,concentrating on the lateral side.Simulating the increased interface coverage does not have a significant impact on peak stress,stress distribution. |
PDF Full Text Request