Abstract: Objective To investigate the effect of bony tunnels created by pedicle screw implantation on the load-bearing capacity of vertebral bodies in thoracolumbar fracture surgery, and to clarify the biomechanical effects of the “tunnel shield” phenomenon in the surgical treatment of thoracolumbar fractures. Methods Ten 3-month-old Bama miniature pigs were randomly divided into a unilateral screw placement group （n=5） and a bilateral screw placement group （n=5）. Posterior lumbar surgery was performed, and screws with a diameter of 2.5 mm were implanted into the pedicles of L1-L5 （unilateral placement in the unilateral group, bilateral placement in the bilateral group）. Spinal specimens were obtained at staged time points postoperatively. The degree of corticalization of the screw tunnel walls was quantified by CT scanning, and three-dimensional finite element models of the L2 vertebral body were established to analyze stress distribution within the screw tunnels. A hydraulic testing machine was used to compare the biomechanical strength between screwed vertebrae and normal vertebrae. A total of 83 patients with thoracolumbar fractures who underwent pedicle screw removal were enrolled and divided into three groups based on the screw placement method in the injured vertebrae： unilateral screw placement group, bilateral screw placement group, and non-screw placement group. Spinal X-ray and CT scans were performed before internal fixation removal, immediately after removal, and at the final follow-up to measure the morphology of the injured vertebrae and adjacent vertebrae, as well as screw tunnel parameters. The effects of different screw placement methods on spinal stability and screw tunnels were analyzed. Results After pedicle screw implantation into the vertebral bodies of miniature pigs, the “tunnel shield” effect led to corticalization of the screw tunnel walls. With prolonged screw retention time, the corticalization of the tunnel walls became more pronounced, and the maximum load-bearing stress of the vertebrae gradually increased, though it remained lower than that of normal vertebrae. Under the same screw tunnel diameter, as the tunnel wall thickness increased, the stress on cortical bone in both the unilateral and bilateral tunnel models gradually decreased, while the stress on cancellous bone gradually increased, and the stress on the tunnel walls gradually decreased. Clinical data analysis revealed that at the final follow-up, the loss of anterior vertebral height ratio in the bilateral screw placement group was significantly lower than those in the other two groups [bilateral group （4.14±3.79）%, unilateral group （7.95±5.85）%, non-screw group （8.89±2.54）%; P=0.022], and the increase in segmental kyphotic Cobb angle was also significantly smaller than those in the other two groups [bilateral group （3.64±2.68）°, unilateral group （6.38±3.48）°, non-screw group （4.74±1.51）° P=0.019]. Conclusion The bony tunnel created by the “tunnel shield” effect can enhance vertebral stability. Bilateral screw placement demonstrates a tendency to be superior to unilateral screw placement in maintaining anterior vertebral height and segmental Cobb angle.

Key words: thoracolumbar fractures, pedicle screw, tunnel shield, biomechanics