Comparison between masked stereolithography and fused deposition modeling 3D-printed synthetic vertebra for surgical training and simulation of pedicle screw fixation

Pedicle screw fixation remains the gold standard for posterior fixations of the spine for fusion and treatment of fractures. To master the technique, hands-on clinical experience is predominately supplemented via training on human cadavers, which is prohibitive due to high cost and limited availability. Here, we investigate the viability of producing 3D-printed synthetic vertebrae models with realistic pedicle screw insertion torque as an alternative to human cadavers to aid in surgical training and simulation. We considered synthetic vertebrae printed from two resin masked stereolithography (low and medium cost) and one polylactic acid fused deposition modeling 3D-printers (low cost). 3D mesh models were created from CT images of human cadaveric vertebra (T1, T2, and T9). For printing, the cancellous component of the vertebra was modelled by a gyroid lattice with ~25% infill density, while the cortical component was modelled by a solid shell of 1-mm and 2-mm thickness. Six copies of each vertebra/shell thickness/printer combination were produced, totaling 108 models. Two orthopedic surgeons with varying experience inserted pedicle screws into the synthetic vertebrae, measuring the peak insertion torque with a digital torque screwdriver and recording qualitative feedback. 3D-printed synthetic vertebra from resin (1-mm shell thickness) enabled realistic insertion torque measurements of the order of human cadaver vertebra, with the Durable resin providing the most consistent and closest to the original human cadaver bones peak insertion torque measurements and qualitative assessment between the two surgeons. PLA filament is not a suitable material for developing 3D-printed synthetic vertebra with realistic screw insertion torque.