Institute of Medical and Biological Engineering, School of Mechanical Engineering, University of Leeds, Leeds LS2 9JT, UK.
J Biomech. 2011 Apr 7;44(6):1108-16. doi: 10.1016/j.jbiomech.2011.01.027. Epub 2011 Feb 16.
Laboratory joint wear simulator testing has become the standard means for preclinical evaluation of wear resistance of artificial knee joints. Recent simulator designs have been advanced and become successful at reproducing the wear patterns observed in clinical retrievals. However, a single simulator test can be very expensive and take a long time to run. On the other hand computational wear modelling is an alternative attractive solution to these limitations. Computational models have been used extensively for wear prediction and optimisation of artificial knee designs. However, all these models have adopted the classical Archard's wear law, which was developed for metallic materials, and have selected wear factors arbitrarily. It is known that such an approach is not generally true for polymeric bearing materials and is difficult to implement due to the high dependence of the wear factor on the contact pressure. Therefore, these studies are generally not independent and lack general predictability. The objective of the present study was to develop a new computational wear model for the knee implants, based on the contact area and an independent experimentally determined non-dimensional wear coefficient. The effects of cross-shear and creep on wear predictions were also considered. The predicted wear volume was compared with the laboratory simulation measurements. The model was run under two different kinematic inputs and two different insert designs with curved and custom designed flat bearing surfaces. The new wear model was shown to be capable of predicting the difference of the wear volume and wear pattern between the two kinematic inputs and the two tibial insert designs. Conversely, the wear factor based approach did not predict such differences. The good agreement found between the computational and experimental results, on both the wear scar areas and volumetric wear rates, suggests that the computational wear modelling based on the new wear law and the experimentally calculated non-dimensional wear coefficient should be more reliable and therefore provide a more robust virtual modelling platform.
实验室联合磨损模拟器测试已成为人工膝关节耐磨性临床前评估的标准手段。最近的模拟器设计已经得到了改进，可以成功地再现临床回收中观察到的磨损模式。然而，单一的模拟器测试可能非常昂贵，并且需要很长时间才能运行。另一方面，计算磨损建模是解决这些限制的另一种有吸引力的解决方案。计算模型已广泛用于人工膝关节设计的磨损预测和优化。然而，所有这些模型都采用了经典的 Archard 磨损定律，该定律是为金属材料开发的，并任意选择了磨损因子。众所周知，对于聚合物轴承材料，这种方法并不普遍适用，由于磨损因子对接触压力的高度依赖性，实施起来很困难。因此，这些研究通常不是独立的，缺乏普遍的可预测性。本研究的目的是为膝关节植入物开发一种新的计算磨损模型，该模型基于接触面积和独立实验确定的无量纲磨损系数。还考虑了交叉剪切和蠕变对磨损预测的影响。将预测的磨损量与实验室模拟测量进行了比较。该模型在两种不同的运动学输入和两种具有弯曲和定制设计的平面轴承表面的胫骨插入物设计下运行。结果表明，新的磨损模型能够预测两种运动学输入和两种胫骨插入物设计之间的磨损量和磨损模式的差异。相反，基于磨损因子的方法无法预测这些差异。计算结果与实验结果之间的良好一致性，无论是在磨损痕迹区域还是体积磨损率方面，都表明基于新磨损定律和实验计算的无量纲磨损系数的计算磨损建模应该更可靠，因此提供了一个更强大的虚拟建模平台。
