Analysis of the influence of species, intervertebral disc height and Pfirrmann classification on failure load of an injured disc using a novel disc herniation model.

Background Context: Annular repair devices offer a solution to recurrent disc herniations by closing an annular defect and lowering the risk of reherniation. Given the significant risk of neurologic injury from device failure it is imperative that a reliable preclinical model exists to demonstrate a high load to failure for the disc repair devices.

Purpose: To establish a preclinical model for disc herniation and demonstrate how changes in species, intervertebral disc height and Pfirrmann classification impacts failure load on an injured disc. We hypothesized that: (1) The force required for disc herniation would be variable across disc morphologies and species, and (2) for human discs the force to herniation would inversely correlate with the degree of disc degeneration.

Study Design: Animal and human cadaveric biomechanical model of disc herniation.

Methods: We tested calf lumbar spines, bovine tail segments and human lumbar spines. We first divided individual lumbar or tail segments to include the vertebral bodies and disc. We then hydrated the specimens by placing them in a saline bath overnight. A magnetic resonance images were acquired from human specimens and a Pfirrmann classification was made. A stab incision measuring 25% of the diameter of the disc was then done to each specimen along the posterior intervertebral disc space. Each specimen was placed in custom test fixtures on a servo-hydraulic test frame (MTS, Eden Prarie, MN) such that the superior body was attached to a 10,000 lb load cell and the inferior body was supported on the piston. A compressive ramping load was placed on the specimen in load control at 4 MPa/sec stopping at 75% of the disc height. Load was recorded throughout the test and failure load calculated. Once the test was completed each specimen was sliced through the center of the disc and photos were taken of the cut surface.

Results: Fifteen each of calf, human, and bovine tail segments were tested. The failure load varied significantly between specimens (p<.001) with human specimens having the highest average failure load (8154±2049 N). Disc height was higher for lumbar/bovine tail segments as compared to calf specimens (p<.001) with bovine tails having the highest disc height (7.1±1.7 mm). Similarly, human lumbar discs had a cross sectional area that was greater than both bovine tail/calf lumbar spines (p<.001). There was no correlation between disc height and failure load within each individual species (p>.05). Cross sectional area and failure load did not correlate with failure load for human lumbar spine and bovine tails (p>.05) but did correlate with calf spine (r=0.53, p=.04). There was a statistically significant inverse correlation between disc height and Pfirrmann classification for human lumbar spines (r=-0.84, p<.001). There was also a statistically significant inverse relationship between Pfirrmann classification and failure load (r=-0.58, p=.02).

Conclusions: We have established a model for disc herniation and have shown how results of this model vary between species, disc morphology, and Pfirrmann classification. Both hypotheses were accepted: The force required for disc herniation was variable across species, and the force to herniation for human spines was inversely correlated with the degree of disc degeneration. We recommend that models using human intervertebral discs should include data on Pfirrmann classification, while biomechanical models using calf spines should report cross sectional area. Failure loads do not vary based on dimensions for bovine tails.

Clinical Significance: Our analysis of models for disc herniation will allow for quicker, reliable comparisons of failure forces required to induce a disc herniation. Future work with these models may facilitate rapid testing of devices to repair a torn/ruptured annulus.