Study Design: Cadaveric biomechanical study., Objective: To determine fatigue behavior of cortical bone trajectory (CBT) pedicle screws., Summary of Background Data: Cortical bone trajectory screws have been becoming popular in spine surgery; however, the long-term fatigue behavior of the new CBT screws remains understudied and limitations not well defined., Methods: Twelve vertebrae from six cadaveric lumbar spines were obtained. After bone mineral density (BMD) measurements, each vertebral body was instrumented with screws from each group, that is, CBT (4.5 × 25 mm) or standard pedicle screw (6.5 × 55 mm). A load (± 4 Nm sagittal bending) was applied under displacement control at 1 Hz. Each construct was loaded for 100 cycles or until 6° of loosening was observed. After fatigue testing, the screws were pulled out axially at 5 mm/min., Results: The standard pedicle screw showed better resistance against 100 cycle loading compared with the CBT screws (P < 0.001, 6.9° ± 4.8° vs. 15.2° ± 5.5°, respectively). The standard pedicle screw testing usually required more than 100 cycles of loading to achieve the critical loosening (3592 ± 4564 cycles), whereas the CBT screw never exceeded 100 cycles (84 ± 24 cycles) (P = 0.002). Increased BMD was significantly associated with a higher number of cycles and less loosening. The standard pedicle screw group had a higher postfatigue pullout load than the CBT screw group (P = 0.001, 776 ± 370 N and 302 ± 232 N, respectively)., Conclusion: The standard pedicle screw had a better fatigue performance compared with the CBT screw in vertebra with compromised bone quality. The proper insertion of the CBT screw might be prevented by the laminar anatomy depending on the screw head design. The CBT screw damaged the bone along its shaft by rotating around a fulcrum, located at either the pars, pedicle isthmus, or the junction of the pedicle and superior endplate, contingent upon the strength of the bone.