Computationally efficient modelling of hip replacement separation due to small mismatches in component centres of rotation

Patient imaging and explant analysis has shown evidence of edge loading of hard-on-hard hip replacements in vivo. Experimental hip simulator testing under edge loading conditions has produced increased, clinically-relevant, wear rates for hard-on-hard bearings when compared to concentric conditions. Such testing, however, is time consuming and costly. A quick running computational edge loading model (Python Edge Loading (PyEL) - quasi-static, rigid, frictionless), capable of considering realistic bearing geometries, was developed. The aim of this study was to produce predictions of separation within the typical experimental measurement error of ∼0.5 mm. The model was verified and validated against comparable finite element (FE) models (including inertia and friction) and pre-existing experimental test data for 56 cases, covering a variety of simulated cup orientations, positions, tissue tensions, and loading environments. The PyEL model agreed well with both the more complex computational modelling and experimental results. From comparison with the FE models, the assumption of no inertia had little effect on the maximum separation prediction. With high contact force cases, the assumption of no friction had a larger effect (up to ∼5% error). The PyEL model was able to predict the experimental maximum separations within ∼0.3 mm. It could therefore be used to optimise an experimental test plan and efficiently investigate a much wider range of scenarios and variables. It could also help explain trends and damage modes seen in experimental testing through identifying the contact locations on the liner that are not easily measured experimentally.