OBJECTIVE: The aim of this study was to compare the biomechanical performance of 6 pedicle screw internal fixation strategies for the treatment of burst fractures of the thoracolumbar spine using finite element (FE) analysis. METHODS: A finite element model of the T11-L3 thoracolumbar segment was established to simulate L1 vertebral burst fractures, and 6 models were conducted under multidirectional loading conditions: P2-D2, P1-D1, P2-D1,P1-D, P1-BF-D1, and P1-UF-D1. The range of motion (ROM) in the T12-L2 region and the von Mises stresses of pedicle screws and rods under the 6 internal fixation models were mainly analyzed. RESULTS: The maximum ROM and von Mises stress were obtained under flexion motion in all models. The P1-BF-D1 model had the least ROM and screw stress. However, when the injured vertebra was not nailed bilaterally, the P1-UF-D1 model had the smallest ROM; the maximum von Mises stress on the screw and rod was remarkably higher than that recorded in the other models. Moreover, the P2-D1 model had a ROM similar to that of the P1-D2 model, but with lower screw stress. The 2 models outperformed the P1-D1 model in all 6 conditions. The P2-D2 model had a similar ROM with the P2-D1 model; nevertheless, the maximum von Mises stress was not substantially reduced. CONCLUSIONS: The P1-BF-D1 model exhibited better stability and less von Mises stress on the pedicle screws and rods, thereby reducing the risk of screw loosening and fracture. The P2-D1 internal fixation approach is recommended when the fractured vertebrae are not nailed bilaterally.