Objective To investigate the amount of bone grafting, bone defect gap residual rates, and biomechanical stability of the injured vertebral body after reduction of thoracolumbar burst fractures, pedicle screw-rods fixation, and bone graft by bilateral pedicle or unilateral spinal canal. Methods Eighteen fresh lumbar spine (L1-5) specimens of calves (aged 4-6 months) were collected to establish the burst fracture model at L3 and divided into 3 groups randomly. After reduction and fixation with pedicle screws, no bone graft was given in group A (n=6), and bone graft was performed by bilateral pedicles in group B (n=6) and by unilateral spinal canal in group C (n=6). The amount of bone grafting in groups B and C was recorded. The general situation of bone defect gaps was observed by the DR films and CT scanning, and the defect gap residual rates of the injured vertebrae were calculated with counting of grids. The compression stiffness was measured by ElectreForce-3510 high precision biological material testing machines. Results The amount of bone grafting was (4.58 ± 0.66) g and (5.72 ± 0.78) g in groups B and C respectively, showing signficant difference (t=2.707, P=0.022). DR films and CT scanning observation showed large bone defect gap was seen in injured vertebrae specimens of group A; however, the grafting bone grains was seen in the “eggshell” gap of the injured vertebral body, which were mainly located in the posterior part of the vertebral body, but insufficient filling of bone graft in the anterior part of the vertebral body in group B; better filling of the grafting bone grains was seen in injured vertebral body of group C, with uniform distribution. The bone defect gap residual rates were 52.0% ± 5.5%, 39.7% ± 2.5%, and 19.5% ± 2.5% respectively in groups A, B, and C; group C was significantly lower than groups A and B (P lt; 0.05), and group B was significantly lower than group A (P lt; 0.05). Flexion compressive stiffness of group C was significantly higher than that of groups A and B (P lt; 0.05), but no significant difference was found between groups A and B (P gt; 0.05). Extension compressive stiffness in group C was significantly higher than that in group A (P lt; 0.05), but no significant difference was found between groups A and B, and between groups B and C (P gt; 0.05). The compression stiffness of left bending and right bending had no significant difference among 3 groups (P gt; 0.05). Conclusion Thoracolumbar burst fracture pedicle screws fixation with bone grafting by unilateral spinal canal can implant more bone grains, has smaller bone defect gap residual rate, and better recovery of flexion compression stiffness than by bilateral pedicles.