Validation of model-predicted tibial tray-synthetic bone relative motion in cementless total knee replacement during activities of daily living

摘要

As fixation of cementless total knee replacement components during the first 4-6 weeks after surgery is crucial to establish bony ingrowth into the porous surface, several studies have quantified implant-bone micromotion. Relative motion between the tray and bone can be measured in vitro, but the full micromotion contour map cannot typically be accessed experimentally. Finite element models have been employed to estimate the full micromotion map, but have not been directly validated over a range of loading conditions. The goal of this study was to develop and validate computational models for the prediction of tray-bone micromotion under simulated activities of daily living. Gait, stair descent and deep knee bend were experimentally evaluated on four samples of a cementless tibial tray implanted into proximal tibial Sawbones (TM) constructs. Measurements of the relative motion between the tray and the anterior cortical shell were collected with digital image correlation and used to validate a finite element model that replicated the experiment. Additionally, a probabilistic analysis was performed to account for experimental uncertainty and determine model sensitivity to alignment and frictional parameters. The finite element models were able to distinguish between activities and capture the experimental trends. Best-matching simulations from the probabilistic analysis matched measured displacement with an average root mean square (RMS) difference of 14.3 mu m and Pearson-product correlation of 0.93, while the mean model presented an average RMS difference of 27.1 mu m and a correlation of 0.8. Maximum deviations from average experimental measurements were 40.5 and 87.1 mu m for the best-matching and average simulations, respectively. The computational pipeline developed in this study can facilitate and enhance pre-clinical assessment of novel implant components. (C) 2018 Elsevier Ltd. All rights reserved.