Assessment of pedicle screw placement accuracy in thoracolumbosacral spine using freehand technique aided by lateral fluoroscopy: Results of postoperative computed tomography in 114 patients

Background context Pedicle screw fixation is currently widely used in spine surgery for various pathologies. Increasing screw placement accuracy would improve the outcomes. Purpose To determine the accuracy rate of screw placement in a group of patients who underwent pedicle screw fixation with conventional techniques. Study design A case series. Patient sample It includes patients undergoing posterior spinal fixation with pedicle screw insertion. Outcome measures include the accuracy of screw placement in pedicles defined by postoperative computed tomography (CT). Methods After surgery, an axial thin-cut CT scan was performed in all patients. Screw position was classified as correct when the screw was completely surrounded by the pedicle cortex and incorrect when any part of the screw was outside the pedicle boundaries. Results Seven hundred seventy screws were inserted at vertebral levels T7-S1 of 114 patients between March 2012 and December 2012. There were three wound infections and one death. Eighteen screws were diagnosed as having an incorrect position (2.3%). The highest accuracy was observed in levels L4 and L5 (0.8% inaccuracy rate for each), whereas the highest inaccuracy rate was observed in T9. The mean inaccuracy rate was 10.5% for levels T7-T9, 3.5% for levels T10-L2, and 0.9% for levels L3-S1. The differences were statistically significant. Only one screw (5%) needed revision. Conclusions The results of our study show that conventional methods for pedicle screw placement remain safe and accurate, with best results obtained in the lumbosacral spine, followed by the thoracolumbar junction. Nonetheless, results are less accurate in the midthoracic spine.