Motion analysis study on sensitivity of finite element model of the cervical spine to geometry

Numerous finite element models of the cervical spine have been proposed, with exact geometry or with symmetric approximation in the geometry. However, few researches have investigated the sensitivity of predicted motion responses to the geometry of the cervical spine. The goal of this study was to evaluate the effect of symmetric assumption on the predicted motion by finite element model of the cervical spine. We developed two finite element models of the cervical spine C2–C7. One model was based on the exact geometry of the cervical spine (asymmetric model), whereas the other was symmetric (symmetric model) about the mid-sagittal plane. The predicted range of motion of both models—main and coupled motions—was compared with published experimental data for all motion planes under a full range of loads. The maximum differences between the asymmetric model and symmetric model predictions for the principal motion were 31%, 78%, and 126% for flexion–extension, right–left lateral bending, and right–left axial rotation, respectively. For flexion–extension and lateral bending, the minimum difference was 0%, whereas it was 2% for axial rotation. The maximum coupled motions predicted by the symmetric model were 1.5° axial rotation and 3.6° lateral bending, under applied lateral bending and axial rotation, respectively. Those coupled motions predicted by the asymmetric model were 1.6° axial rotation and 4° lateral bending, under applied lateral bending and axial rotation, respectively. In general, the predicted motion response of the cervical spine by the symmetric model was in the acceptable range and nonlinearity of the moment–rotation curve for the cervical spine was properly predicted.