Effect of wear of bearing surfaces on elastohydrodynamic lubrication of metal-on-metal hip implants

The effect of geometry change of the bearing surfaces owing to wear on the elastohydrodynamic lubrication (EHL) of metal-on-metal (MOM) hip bearings has been investigated theoretically in the present study. A particular MOM Metasul bearing (Zimmer GmbH) was considered, and was tested in a hip simulator using diluted bovine serum. The geometry of the worn bearing surface was measured using a coordinate measuring machine (CMM) and was modelled theoretically on the assumption of spherical geometries determined from the maximum linear wear depth and the angle of the worn region. Both the CMM measurement and the theoretical calculation were directly incorporated into the elastohydrodynamic lubrication analysis. It was found that the geometry of the original machined bearing surfaces, particularly of the femoral head with its out-of-roundness, could lead to a large reduction in the predicted lubricant film thickness and an increase in pressure. However, these non-spherical deviations can be expected to be smoothed out quickly during the initial running-in period. For a given worn bearing surface, the predicted lubricant film thickness and pressure distribution, based on CMM measurement, were found to be in good overall agreement with those obtained with the theoretical model based on the maximum linear wear depth and the angle of the worn region. The gradual increase in linear wear during the running-in period resulted in an improvement in the conformity and consequently an increase in the predicted lubricant film thickness and a decrease in the pressure. For the Metasul bearing tested in an AMTI hip simulator, a maximum total linear wear depth of approximately 13 microm was measured after 1 million cycles and remained unchanged up to 5 million cycles. This resulted in a threefold increase in the predicted average lubricant film thickness. Consequently, it was possible for the Metasul bearing to achieve a fluid film lubrication regime during this period, and this was consistent with the minimal wear observed between 1 and 5 million cycles. However, under adverse in vivo conditions associated with start-up and stopping and depleted lubrication, wear of the bearing surfaces can still occur. An increase in the wear depth beyond a certain limit was shown to lead to the constriction of the lubricant film around the edge of the contact conjunction and consequently to a decrease in the lubricant film thickness. Continuous cycles of a running-in wear period followed by a steady state wear period may be inevitable in MOM hip implants. This highlights the importance of minimizing the wear in these devices during the initial running-in period, particularly from design and manufacturing points of view.