Biomechanical comparisons of spinal fracture models and the stabilizing effects of posterior instrumentations

In this study, the authors evaluated the stiffness of motion segments in intact spines in two spine fracture models, and with each of five implant systems used for posterior fixation of thoracolumbar spine fractures. The devices represented a cross-section of types, including those employing sublaminar wires with and without laminar hooks, pedicle screws, plates, and rods. Two spine fracture models, one partially and one totally destabilized, were used in the tests of the instrumentation. Stiffness, or the magnitude of load needed to produce a unit displacement of the construct in the direction of the applied load, was measured in flexion, extension, lateral bending, and torsion in combination with a compressive force. Both horizontal plane shear and angular displacements were measured in the two fracture patterns. All evaluations were made by testing the difference in stiffness for statistical significance among groups. The results showed significant differences in stiffness without instrumentation among intact spines, partly destabilized spines (anterior two-thirds of disk and posterior ligaments removed), and totally destabilized spines (only anterior longitudinal ligament intact). The implant/spine constructs were least stiff relative to the intact spine in torsion, followed in increasing order of stiffness with flexion, lateral bending, and extension. In the Roy-Camille plate with six-screw fixation was found to produce the stiffest construct, followed by wired Harrington rods, C-rods and J-rods, and the Vermont internal fixator.