Pullout strength of pedicle screws using cadaveric vertebrae with or without artificial demineralization

OBJECTIVES: To evaluate the differences in the pullout strength and displacement of pedicle screws in cadaveric thoracolumbar vertebrae with or without artificial demineralization. METHODS: Five human lumbar and five thoracic vertebrae from one cadaver were divided into two hemivertebrae. The left-side specimens were included in the simulated osteopenic model group and the right-side bones in a control group. In the model group, we immersed each specimen in HCl (1 N) solution for 40 minutes. We measured bone mineral density (BMD) using dual-energy X-ray absorptiometry and quantitative computerized tomography. We inserted polyaxial pedicle screws into the 20 pedicles of the cadaveric lumbar and thoracic spine after measuring the BMD of the 2 hemivertebrae of each specimen. We measured the pullout strength and displacement of the screws before failure in each specimen using an Instron system. RESULTS: The average pullout strength of the simulated osteopenic model group was 76% that of the control group. In the control and model groups, the pullout strength was 1678.87±358.96 N and 1283.83±341.97 N, respectively, and the displacement was 2.07±0.34 mm and 2.65±0.50 mm, respectively (p<.05). We detected positive correlations between pullout strength and BMD in the control group and observed a negative correlation between displacement and BMD in the model group. CONCLUSIONS: By providing an anatomically symmetric counterpart, the human cadaveric model with or without demineralization can be used as a test bed for pullout tests of the spine. In the simulated osteopenic model group, pullout strength was significantly decreased compared with the untreated control group. CLINICAL SIGNIFICANCE: Decreased bone mineral density may significantly reduce the pullout strength of a pedicle screw, even though the range is osteopenic rather than osoteoporotic.