Хирургия позвоночника (Dec 2020)
Comparison of pedicle screw placement using O-arm naviga- tion and navigational templates in an animal model experiment
Abstract
Objective. To perform a comparative analysis of experimental pedicle screw placement using custom-made 3D-printed navigational templates and using O-arm (cone-beam computerized tomograph, CBCT) and navigation station. Material and Methods. The experiment was performed on five fresh anatomical specimens of the lamb thoracic and lumbar spine. In Group 1, 44 screws were inserted using O-arm and Stealth Station S7 navigation system, and in Group 2, 72 screws were inserted using 3D-printed navigational templates. The main comparison criterion was the safety of implantation assessed based on a grade (0 to 3) of cortical bone perforation on postoperative CT. The extra comparing criteria were the time of implantation and summary radiation exposure required for screw placement. In Group 2, the accuracy of implantation was analyzed by assessing the deviation (mm) of the actual screw trajectory from the planned one at the point of entry into the vertebra and at the intersection of the screw axis with the anterior cortical layer of the vertebral body (end point), and by measuring the angles between the trajectories. The results were evaluated for normal distribution and subjected to statistical analysis for paired independent groups using the Kruskal-Wallis test and Chi-square in the Statistica 10 software. Results. Analysis of the safety revealed significant difference between the groups (p < 0.0001). In Group 2 there were not any cases of cortical bone perforation, in Group 1 (O-arm) grade 0 was registered for 28 (64 %) screws, grade 1 for 7(16 %) screws, grade 2 for 4 (9 %) screws, and grade 3 for 5 (11 %) screws. The average time of one screw placement was 81.00 (64.50; 94.00) sec in Group 1 and 40.75 (33.50; 52.25) sec in Group 2, p < 0.001. In Group 2, the mean deviation of the entry point was 0.50 (0.34; 0.87) mm, and of the end point – 1.10 (0.66; 1.93) mm. The mean angle between the planned and actual trajectories was 2.76 (0.80; 4.89) degrees in the axial plane and 2.62 (1.43; 4.35) degrees in the sagittal plane. The average design time for one template was 8.75 (8.00; 9.75) min, and 3D printing time – 60 (57; 69) min. The approximate material cost for one template printing was 45 rubles, for one anatomical specimen of lamb thoracic and lumbar spine – 390 rubles. The CT dose index (CTDI) for the O-arm was 8.99–9.01 mGy, and dose length product (DLP) for one model (3 scans) was 432 mGr × sm. In Group 2, there was no intraoperative X-ray control, the CTDI for preoperative CТ was 10.37– 10.67 mGy, and DLP was 459–477 mGr × sm. Conclusion. The results of the experiment on a lamb spine biomodel showed that pedicle screw placement with 3D custom-made navigational templates is associated with better results of the safety and the speed of implantation as compared to that with intraoperative O-arm navigation. This justifies the 3D-printed template using in case of increased mobility of the spine during implantation, where the accuracy of CT navigation is significantly reduced. In clinical practice, these conditions correspond to transpedicular fixation of the cervical spine and screw fixation of the C1–C2 vertebrae.
Keywords
