Experimental versus computational analysis of micromotions at the implant-bone interface.

In total hip arthroplasty, micromotions at the implant-bone interface influence the long-term survival of the prosthesis. These micromotions are often measured using sensors that are fixed to the implant and bone at points that are remote from the interface. Given that the implant-bone system is not rigid, errors may be introduced. It is not possible to assess the magnitude of these errors with the currently available experimental methods. However, this problem can be investigated using the finite element method (FEM). The hypothesis that the actual interface micromotions differ from those measured in the experimental manner was tested using a case-specific FE model, validated against deflection experiments. The FE model was used to simulate an 'experimental' method to measure micromotions. This 'experimental' method was performed by mimicking the distance between the measurement points; the implant point was selected at the interface while the bony point was at the outer surface of bone. No correlation was found between the micromotions computed at the interface and when using remote reference points. Moreover, the magnitudes of micromotions computed with the latter method were considerably greater. By reducing the distance between the reference points the error decreased, but the correlation stayed unchanged. Care needs to be taken when interpreting the results of micromotion measurement systems that use bony reference points at a distance from the actual interface.