STUDY DESIGN: This is a retrospective study conducted to evaluate the efficacy of single versus separate registration in assessing the pedicle screw accuracy in the computer-assisted lumbar spinal instrumentation. OBJECTIVES: To see if separate registration reduced lumbar pedicle screw misplacement. SUMMARY OF BACKGROUND DATA: Computer-assisted spinal instrumentation has been shown to improve pedicle screw installation accuracy, but 2.7% to 8% of screws still perforate the pedicular cortex. Suspected causes include differences in lumbar lordosis between preoperative CT scans and surgery. METHODS: Postoperative radiographs and CT scans were used to assess the accuracy of pedicle screw placement in 47 adult patients following computer-assisted lumbar spinal instrumentation. Twenty-two patients underwent single registration at one level, while the other 25 underwent registration at each level. RESULTS: The time required for a registration procedure on one level was 6 to 8 minutes, while the time required for application of a pedicle screw using computer-assisted techniques was an additional 6 to 10 minutes. The total number of screw placements was 118 in the single registration group and 130 in the separate registration group. In the former group, 85 (72%) pedicle screw placements were categorized as good, 28 (24%) were fair, and 5 (4%) were poor. All five poorly placed screws were placed in the lower lumbar or upper sacral spine with high mobility, and at levels without registration, with one causing root injury. In the latter group, 117 (90%) pedicle screw placements were good and 13 (10%) were fair. The difference in placement was found to be statistically significant (chi2, P = 0.0003). CONCLUSION.: Before the intraoperative real-time CT imaging is widely used, separate registration at each instrumented level during traditional computer-assisted lumbar spinal instrumentation is necessary to enhance the accuracy of screw placement.
STUDY DESIGN: This is a retrospective study conducted to evaluate the efficacy of single versus separate registration in assessing the pedicle screw accuracy in the computer-assisted lumbar spinal instrumentation. OBJECTIVES: To see if separate registration reduced lumbar pedicle screw misplacement. SUMMARY OF BACKGROUND DATA: Computer-assisted spinal instrumentation has been shown to improve pedicle screw installation accuracy, but 2.7% to 8% of screws still perforate the pedicular cortex. Suspected causes include differences in lumbar lordosis between preoperative CT scans and surgery. METHODS: Postoperative radiographs and CT scans were used to assess the accuracy of pedicle screw placement in 47 adult patients following computer-assisted lumbar spinal instrumentation. Twenty-two patients underwent single registration at one level, while the other 25 underwent registration at each level. RESULTS: The time required for a registration procedure on one level was 6 to 8 minutes, while the time required for application of a pedicle screw using computer-assisted techniques was an additional 6 to 10 minutes. The total number of screw placements was 118 in the single registration group and 130 in the separate registration group. In the former group, 85 (72%) pedicle screw placements were categorized as good, 28 (24%) were fair, and 5 (4%) were poor. All five poorly placed screws were placed in the lower lumbar or upper sacral spine with high mobility, and at levels without registration, with one causing root injury. In the latter group, 117 (90%) pedicle screw placements were good and 13 (10%) were fair. The difference in placement was found to be statistically significant (chi2, P = 0.0003). CONCLUSION.: Before the intraoperative real-time CT imaging is widely used, separate registration at each instrumented level during traditional computer-assisted lumbar spinal instrumentation is necessary to enhance the accuracy of screw placement.