Study design: Histologic experiments and biomechanical tests were performed in human cadaveric lumbar spine models.
Objective: To determine (1) the anatomic structures of lumbar endplates, (2) the relationship between bone mineral density (BMD) and biomechanical properties of lumbar endplates, and (3) the influence of spinal level on the failure loads of lumbar endplates.
Summary of background data: Previous works have shown that the posterolateral corners of the lumbar endplates are stronger than the anterior and central regions. Information on the microstructures of lumbar endplates and the effects of both BMD and spinal level on biomechanical properties of lumbar endplates would be valuable for spine surgeons and implant designers to avoid subsidence.
Methods: Twenty fresh human cadaver lumbar vertebrae specimens were collected for HE staining and the sections through the specimens were examined under light microscope. The areas of the pore structures in the lumbar endplates were measured and analyzed by statistical methods. Sixty-five cadaver lumbar vertebrae were evaluated with dual energy radiograph absorptiometry and according to their BMD, all vertebrae were divided into 3 groups. Indentation tests were performed at 27 standardized test sites in endplates of these vertebrae using a 1.5-mm-diameter, hemispherical indenter with a rate of 12 mm/min. The failure load at each test site was determined using the load-displacement curve. Spearman analysis was used to evaluate the correlation between BMD and failure loads of lumbar endplates. Factorial analysis of variance was performed to reveal the effects of both BMD and spinal level on failure load distribution on the lumbar endplates.
Results: (1) The peripheral regions were thicker than the central regions of lumbar endplates. The central regions were porous, as the fused trabeculae and peripheral regions had fewer and smaller pore structures. (2) The tests showed that there were significant differences in failure load between 3 groups (P < 0.01). BMD was positively correlated with the failure load of the lumbar endplate. (3) The strongest position of the lumbar endplate was the posterolateral region closest to the pedicles and the failure loads tended to increase from anterior to posterior regions of the lumbar endplates. (4) The failure load distribution did not change with the BMD decrease. (5) The differences of the failure loads between the lumbar segments were significant (P < 0.05). The failure loads of the lumbar endplates showed an increasing tendency from L1-L5 segments.
Conclusion: The differences in the anatomic structures of different regions are the histologic foundation of biomechanical properties of lumbar endplates. It is necessary that BMD should be measured before operation and implants should be placed in posterior-lateral regions of lumbar endplates so that subsidence syndrome can be reduced as much as possible. In addition, the fact that the endplates of the upper lumbar segments with the lower strength have higher risk of subsidence should be concerned about by spine surgeons and implant designers.