Computational analyses of different intervertebral cages for lumbar spinal fusion.

Lumbar spinal fusion is the most common approach for treating spinal disorders such as degeneration or instability. Although this procedure has been performed for many years, there are still important challenges that must be overcome and questions that need to be addressed regarding the high rates of non-union. The present finite element model study aimed to investigate the influence of different cage designs on the fusion process. An axisymmetric finite element model of a spinal segment with an interbody fusion cage was used. The fusion process was based on an existing mechano-regulation algorithm for tissue formation. With this model, the following principal concepts of cage design were investigated: (1) different cage geometries with constant compressive stiffness and (2) cage designs optimized to provide the ideal mechanical stimulus for bone formation, first at the beginning of fusion and then throughout the entire fusion process. The cage geometry substantially influenced the fusion outcome. A cage that created an optimized initial mechanical stimulus did not necessarily lead to accelerated fusion, but rather resulted in delayed fusion or non-union. In contrast, a cage made of a degradable material produced a significantly higher amount of bone and resulted in higher segmental stiffness. However, different compressive loads (250, 500 and 1000 N) substantially affected the amount of newly formed bone tissue. The results of the present study suggest that aiming for an optimal initial mechanical stimulus may be misleading because the initial mechanical environment is not preserved throughout the bone modeling process.