Author: Chloe Duncan, Andrew Kanawati, Peter Malek, Tess Reynolds 👨🔬
Affiliation: University of Sydney, Westmead Hospital, Image X Institute, Faculty of Medicine and Health, The University of Sydney 🌍
Purpose: Pedicle screw fixation, a standard spinal surgery procedure, has high misplacement rates (~40%) which decrease with surgeon experience. However, opportunities for surgical rehearsal, training, and research are limited. To solve this problem in a globally accessible manner, 3D-printed models have shown potential to realistically simulate the haptic/biomechanical properties of vertebrae. Here, we investigate the impact of varying cortical shell thickness on 3D-printed vertebral models (upper-thoracic, lower-thoracic, and lumbar) to accurately simulate real bone across multiple vertebral levels, replicating hands-on surgical experience.
Methods: Human T1, T2, T9, T10 and L4 cadaveric vertebrae were CT-imaged and STL files were generated using 3D Slicer. Files were modified using CHITUBOX Basic and Blender to generate models with varying cortical shell thicknesses of 0.25mm from 1.00mm to 2.00mm, with custom gyroid infills. 6 copies of each model were printed in resin (Durable, Form 4 Printer). Two surgeons completed pedicle screw fixation on the models and qualitatively rated the haptic feel of drilling and screw insertion. Peak insertion torque was measured using a digital torque screwdriver for quantitative assessment.
Results: For thoracic vertebrae, 1.50mm-1.75mm shell thickness provided the most similar haptic feel to bone. For L4 vertebrae, 1.75mm-2.00mm shell thickness provided the most similar haptic feel to bone. Across all vertebral levels, peak insertion torque increased with shell thickness. However, all shell thicknesses resulted in peak insertion torques much higher than that of the original cadaveric vertebrae. L4 models had lower peak insertion torque than the thoracic models.
Conclusion: This study is the first to investigate the effect of cortical shell thickness on the performance of 3D-printed synthetic models across multiple vertebral levels, revealing different ideal thickness between thoracic and lumbar vertebra. Future investigations should focus on varying the density and thickness of the custom gyroid infills to further increase model realism.