Stress analysis of a canine spinal motion segment using the finite element technique.

Canine models have been frequently employed to investigate the in vivo effects of a surgical procedure. Various studies indicate that canine models can provide a successful in vivo biological model for these studies. Use of canine models for the biomechanical studies of the spine, however, have been questioned because of different loading conditions on the canine and human spines originated from posture differences between canine and human. Similarities between the stress distributions within the canine and human motion segments under physiological loads will strengthen the use of canine models for the studies of spine biomechanics. In the present study, finite element models of the canine intact and stabilized motion segments were developed to investigate these aspects. Comparison of model predicted flexion angle, axial stiffness, and facet contact force for the canine intact L6-L7 motion segment revealed good agreement with the corresponding parameters experimentally measured under the similar loading conditions. Similar stress distributions within the intact canine and human models were found from the predicted results in response to the physiological load. Stabilizing and stress-shielding effects of a pedicle screw-plate-type fixation device [variable spinal plating (VSP)] on the stabilized motion segment were also similar for the canine and human stabilized models. Furthermore, maximum stresses in the pedicle screws were found at the junction between the bone screw and the integrated nut of the inferior screw in both the canine and human stabilized models. This corresponds to the location of pedicle screw breakage reported in the literature. These findings suggest that a canine is a suitable model for the biomechanical studies of the lumbar spine.