Effect of an artificial disc on lumbar spine biomechanics: a probabilistic finite element study

The effects of different parameters on the mechanical behaviour of the lumbar spine were in most cases determined deterministically with only one uncertain parameter varied at a time while the others were kept fixed. Thus most parameter combinations were disregarded. The aim of the study was to determine in a probabilistic finite element study how intervertebral rotation, intradiscal pressure, and contact force in the facet joints are affected by the input parameters implant position, implant ball radius, presence of scar tissue, and gap size in the facet joints. An osseoligamentous finite element model of the lumbar spine ranging from L3 vertebra to L5/S1 intervertebral disc was used. An artificial disc with a fixed center of rotation was inserted at level L4/L5. The model was loaded with pure moments of 7.5 Nm to simulate flexion, extension, lateral bending, and axial torsion. In a probabilistic study the implant position in anterior–posterior (ap) and in lateral direction, the radius of the implant ball, and the gap size of the facet joint were varied. After implanting an artificial disc, scar tissue may develop, replacing the anterior longitudinal ligament. Thus presence and absence of scar tissue were also simulated. For each loading case studied, intervertebral rotations, intradiscal pressures and contact forces in the facet joints were calculated for 1,000 randomized input parameter combinations in order to determine the probable range of these output parameters. Intervertebral rotation at implant level varies strongly for different combinations of the input parameters. It is mainly affected by gap size, ap-position and implant ball radius for flexion, by scar tissue and implant ball radius for extension and lateral bending, and by gap size and implant ball radius for axial torsion. For extension, intervertebral rotation at implant level varied between 1.4° and 7.5°. Intradiscal pressure in the adjacent discs is only slightly affected by all input parameters. Contact forces in the facet joints at implant level vary strongly for the different combinations of the input parameters. For flexion, forces are 0 in 63% of the cases, but for small gap sizes and large implant ball radii they reach values of up to 533 N. Similar results are found for extension with a maximum predicted force of 560 N. Here the forces are mainly influenced by gap size, implant ball radius and scar tissue. The forces vary between 0 and 300 N for lateral bending and between 0 and 200 N for axial torsion. The parameters that have the greatest effect in both loading cases are the same as those for extension. Intervertebral rotation and contact force in the facet joints depend strongly on the input parameters studied. The probabilistic study shows a large variation of the results and likelihood of certain values. Clinical studies will be required to show whether or not there is a strong correlation of parameter combinations that cause high facet joint forces and low back pain after total disc replacement.