The effect of pedicle screw redirection after lateral wall breach - A biomechanical study using human lumbar vertebrae

Background context: Currently, pedicle screw segmental fixation of the spine is considered a standard of care for a number of conditions. Most surgeons employ a free-hand technique using various intraoperative modalities to improve pedicle screw accuracy. Despite continued improvements in technique, pedicle breach remains a frequent occurrence. Once a breach is detected intraoperatively, the most common corrective maneuver is to medially redirect the pedicle screw into the pedicle. To our knowledge, the biomechanical impact of medially redirecting a pedicle screw after a lateral pedicle breach has not been examined. Purpose: To compare the fixation strength of perfectly placed pedicle screws to the fixation strength of pedicle screws that were correctly placed after having been redirected (RD) following a lateral pedicle breach. Study design/setting: A biomechanical study using human lumbar vertebrae. Methods: Ten fresh human lumbar vertebrae were isolated from five donors. Each vertebra was instrumented with a monoaxial pedicle screw into each pedicle using two different techniques. On one side, a perfect center-center (CC) screw path was created using direct visualization and fluoroscopy. A 6.0-mm-diameter cannulated tap and a pedicle probe were used to develop the pedicle for the 7.0-mm-diameter by 45-mm-long cannulated pedicle screw, which was placed using a digital torque driver. On the contralateral side, an intentional lateral pedicle wall breach was created at the pedicle-vertebral body junction using a guide wire, a 6.0-mm-diameter cannulated tap, and a pedicle probe. This path was then redirected into a CC position, developed, and instrumented with a 7.0-mm-diameter by 45-mm-long cannulated pedicle screw: the RD screw. For each pedicle screw, we assessed four outcome measures: maximal torque, seating torque, screw loosening, and post-loosening axial pullout. Screw loosening and axial pullout were assessed using an MTS machine. Results: The biomechanical cost of a lateral pedicle breach and the requirement to redirect the pedicle screw are as follows: an overall drop of 28% (p<.002) in maximal insertion torque and 25% (p<.049) in seating torque, a drop of 25% (p<.040) in resistance to screw loosening, and a drop in axial pullout force of 11% (p<.047). Conclusions: Compared with a CC lumbar pedicle screw, an RD lumbar pedicle screw placed after a lateral wall breach is significantly weaker in terms of maximal insertional torque, seating torque, screw loosening force, and axial pullout strength. These significant decreases in biomechanical properties are clearly important when RD pedicle screws are placed at the cephalad or caudal end of a long construct. In this situation, augmentation of the RD screw is an option.