Design, fabrication, and evaluation of single- and multi-level 3D-printed non-covering cervical spinal fusion surgery templates

BackgroundCervical spinal fusion surgeries require accurate placement of the pedicle screws. Any misplacement/misalignment of these screws may lead to injuries to the spinal cord, arteries and other organs. Template guides have emerged as accurate and cost-effective tools for the safe and rapid insertions of pedicle screws.Questions/PurposesNovel patient-specific single- and multi-level non-covering templates for cervical pedicle screw insertions were designed, 3D-printed, and evaluated.MethodsCT scans of two patients were acquired to reconstruct their 3D spine model. Two sets of single-level (C3-C7) and multi-level (C4-C6) templates were designed and 3D-printed. Pedicle screws were inserted into the 3D-printed vertebrae by free-hand and guided techniques. For single-level templates, a total of 40 screws (2 patients × 5 vertebrae × 2 methods × 2 screws) and for multi-level templates 24 screws (2 patients × 3 vertebrae × 2 methods × 2 screws) were inserted by an experienced surgeon. Postoperative CT images were acquired to measure the errors of the entry point, 3D angle, as well as axial and sagittal plane angles of the inserted screws as compared to the initial pre-surgery designs. Accuracy of free-hand and guided screw insertions, as well as those of the single- and multi-level guides, were also compared using paired t-tests.ResultsDespite the minimal removal of soft tissues, the 3D-printed templates had acceptable stability on the vertebrae during drillings and their utilization led to statistically significant reductions in all error variables. The mean error of entry point decreased from 3.02 mm (free-hand) to 0.29 mm (guided) using the single-level templates and from 5.7 mm to 0.76 mm using the multi-level templates. The percentage reduction in mean of other error variables for, respectively, single- and multi-level templates were as follows: axial plane angle: 72% and 87%, sagittal plane angle: 56% and 78%, and 3D angle: 67% and 83%. The error variables for the multi-level templates generally exceeded those of the single-level templates. The use of single- and multi-level templates also considerably reduced the duration of pedicle screw placements.ConclusionThe novel single- and multi-level non-covering templates are valuable tools for the accurate placement of cervical pedicle screws.