Evaluation of linear finite-element analysis models' assumptions for external fixation devices

Abstract

Linear finite-element models (FEMs) have enjoyed an increased use in orthopaedic research, including the use for modelling external fixation devices. These fixator FEMs depend on a number of basic assumptions concerning the overall fixation frame stability and the components' rigidity. Among the more important ones are: (i) rigid fixation at both ends of the pin and sidebar; (ii) that the sidebar can be treated essentially as a rigid entity, with all bending occurring in the bone pins; and (iii) that the system can be treated as linearly elastic. Prior work done by the authors questions some of these assumptions. Thus, this study sought an empirical evaluation of the validity of some of these a priori assumptions. A Hoffmann single half-frame was tested in its standard form and then according to a stepwise protocol wherein the frame was welded to eliminate any possible points of instability. These tests looked at the stability and rigidity in various modes (axial compression, torsion, and medial-lateral and anterior-posterior four-point bending). The basic assumptions concerning the frame stability, frame rigidity and the frame's response to loads were found to be erroneous. Component failure was common under minimal loads and statistically significant differences ( p < 0.05) of up to 75% were noted in frame rigidity among the various frame forms tested. Thus, considerable caution must be exercised when employing the FEM technique for evaluating the fixator properties.