Lopes Vitor M M, Neto Maria A, Amaro Ana M, Roseiro Luis M, Paulino M F
CEMMPRE, Center for Mechanical Engineering, Materials and Process, University of Coimbra, Rua Luís Reis Santos, 3030-788 Coimbra, Portugal.
Polytechnic Institute of Coimbra, Coimbra Institute of Engineering - Mechanical Engineering Department, Rua Pedro Nunes- Quinta da Nora, 3030-199, Coimbra, Portugal.
Med Eng Phys. 2017 Aug;46:96-109. doi: 10.1016/j.medengphy.2017.06.001. Epub 2017 Jun 20.
The primary aim of this work was to validate the "numerical" cortex material properties (transversely isotropic) of synthetic femurs and to evaluate how the strain level of the cancellous bone can be affected by the FE modeling of the material's behavior. Sensitivity analysis was performed to find out if the parameters of the cortex material affect global strain results more than the Polyurethane (PU) foam used to simulate cancellous bone. Standard 4th generation composite femurs were made with 0.32g/cm solid PU foam to model healthy cancellous bone, while 0.2g/cm cellular PU was used to model unhealthy cancellous bone. Longitudinal and transversal Young's moduli of cortical bone were defined according the manufacturer data, while shear modulus and Poisson's ratios were defined from the literature. All femurs were instrumented with rosette strain gauges and loaded according to ISO7206 standards, simulating a one-legged stance. The experimental results were then compared with those from finite element analysis. When cortical bone was modelled as transversely isotropic, an overall FE/experimental error of 11% was obtained. However, with isotropic material the error rose to 20%. Strain field distributions predicted inside the two bone models were similar, but the strain state of a healthy cancellous bone was much more a compression state than that of unhealthy bone, the compression state decreased about 90%. Strain magnitudes show that average strain-levels of cancellous bone can be significantly affected by the properties of the cortical bone material and, therefore, simulations of femur-implanted systems must account for the composite behavior of the cortex, since small shear strains would develop near isotropic cancellous bone-implant interfaces. Moreover, the authors suggest that changing the volume fraction of glass fibers used to manufacture the cortical bone would allow a more realistic osteoporotic synthetic femurs to be produced.
这项工作的主要目的是验证合成股骨的“数值”皮质材料特性（横向各向同性），并评估有限元建模的材料行为如何影响松质骨的应变水平。进行了敏感性分析，以确定皮质材料的参数是否比用于模拟松质骨的聚氨酯（PU）泡沫对整体应变结果的影响更大。标准的第四代复合股骨采用0.32g/cm的实心PU泡沫制成，以模拟健康的松质骨，而0.2g/cm的多孔PU用于模拟不健康的松质骨。皮质骨的纵向和横向杨氏模量根据制造商数据定义，而剪切模量和泊松比则从文献中定义。所有股骨均安装了应变片，并根据ISO7206标准加载，模拟单腿站立。然后将实验结果与有限元分析结果进行比较。当将皮质骨建模为横向各向同性时，获得的有限元/实验总体误差为11%。然而，采用各向同性材料时，误差上升至20%。两个骨模型内部预测的应变场分布相似，但健康松质骨的应变状态比不健康骨更处于压缩状态，压缩状态降低了约90%。应变大小表明，皮质骨材料的特性会显著影响松质骨的平均应变水平，因此，股骨植入系统的模拟必须考虑皮质骨的复合行为，因为在各向同性的松质骨-植入物界面附近会产生小的剪切应变。此外，作者建议，改变用于制造皮质骨的玻璃纤维体积分数，将能够制造出更逼真的骨质疏松症合成股骨。
