Since artificial joints are expected to operate for more than decades in human body, animal and clinical studies are not suitable for evaluation of their durability. Instead, in-vitro mechanical tests have been employed, but they cannot fully reproduce complex in-vivo mechanical and biochemical environment. For instance, lipids in synovial fluid have been known to be absorbed in ultra-high molecular weight polyethylene (UHMWPE) components of artificial joints in vivo, and recently it was found that absorbed lipids have potential to degrade UHMWPE. In order to assure clinical relevance of the in-vitro mechanical tests, understanding of the effect of the in-vivo environment on mechanical properties is indispensable. However, well-developed mechanical tests cannot be applied to retrieved components, because they require large specimens. In this study, we attempted to develop methods to evaluate mechanical properties of retrieved UHMWPE components.

We prepared five kinds of UHMWPE. Those are molded UHMWPE made from GUR 1020 resin without any further treatment, remelted highly crosslinked UHMWPE, annealed highly crosslinked UHMWPE, squalene absorbed UHMWPE which was prepared by immersing in squalene at 80°C for 7 days (SQ) and squalene absorbed and artificially aged UHMWPE which was prepared by artificially aging SQ at 80°C for 21 days in air (SQA). SQ and SQA were employed in this study to mimic lipid absorption and lipid induced degradation.

These materials were tested by two well-established mechanical tests, namely, tensile tests and compression tests, and two proposed mechanical tests that can be applied to retrieved components, namely, tensile punch tests and micro indentation tests.

It was possible to clearly identify the difference between materials by any of test methods used in this study. Stiffness obtained from tensile punch tests and elastic modulus obtained from micro indentation tests were shown to be highly correlated with elastic modulus obtained from compression tests except for SQA, which was inhomogeneous due to degradation at the surfaces. The results showed that the elastic modulus of the local surface could be evaluated by micro indentation tests, while the average of that of the entire specimen could be evaluated by compression tests. ield load, fracture load and maximum load obtained from tensile punch tests showed little correlation with yield stress, fracture stress and maximum stress obtained from tensile tests, respectively. These differences were considered to be attributed to the differences in a stress condition between these two test methods. It is multi-axial tension in tensile punch tests, while it is uniaxial in tensile tests.

Although some of the parameters obtained by tensile punch tests showed no or limited correlation with those obtained by tensile tests, it was possible to clearly identify the difference between materials by these proposed test methods. In particular, micro indentation tests could evaluate the mechanical properties very locally. These proposed test methods have the potential to provide useful information on mechanical properties of retrieved UHMWPE components.