Abstract
Pedicle screw malpositioning remains a frequent complication, with reported rates from 2% to 15%, often leading to revision surgeries. Analyzing mechanical resistance and torque encountered during screw insertion has been implicated as a promising approach for real-time detection. Five fresh-frozen human thoracolumbar spine specimens were utilized in this study. Using 3D-printed templates, correct trajectories were systematically compared against four defined malpositions (medial, lateral, superior, superolateral), with offsets ranging from 2.0 mm to 3.5 mm. Drilling, tapping, and insertion phases were conducted at a constant speed and defined feed force. Contrary to the anticipated behavior, malpositioned trajectories showed no statistically significant difference in peak torque compared to correct trajectories across all phases (e.g., tapping p=0.944, r=0.01; insertion p=0.693, r=0.05). Regional stratification between thoracic and lumbar spine also failed to yield significant differences. The only statistically significant difference was observed between the correct trajectory and the superolateral malposition during drilling (p=0.038). Under the tested standardized conditions, torque-based mechanical resistance during pedicle screw placement is generally not a reliable and consistent real-time indicator of malposition.