Computational models for characterisation and design of patient-specific spinal implant

BACKGROUND CONTEXT: Spinal fusion is designed to reduce movements between vertebrae and therefore pain. The most used devices for this procedure are mainly made of titanium or polyether ether ketone (PEEK). However, the mismatch between devices, with standard shapes and materials, and the surrounding bones can lead to suboptimal outcomes. Computational models, namely, Finite Element Analyses (FEA), can be employed to optimise existing device and design more effective solutions. PURPOSE: The goal of this study was to compare the performance of different materials and material densities for spinal cages, and to design a novel geometry which can ideally match the anatomical characteristics of a patient. STUDY DESIGN/SETTING: Computational. PATIENT SAMPLE: Nil. OUTCOME MEASURES: Nil. METHODS: FEA were set up to simulate compression (400 N) and bending (7.5 Nm) on a generic cage design. Three materials were modelled: titanium, PEEK and polycarbonate. Polycarbonate was included as widely available within additive manufacturing techniques. For each of the cages, four designs were modelled with varying material filling density. Furthermore, a new cage was modelled to match the pre-operative computed tomography (CT) of a patient exactly. The patient-specific cage was also tested by means of FEA. RESULTS: Stress distribution was compared between all the three materials tested. Consistently, stresses increased with reducing material density. Stress peak values were lower than the respective risk of failure in all the simulated cases, confirming the feasibility of polycarbonate implants. The patient-specific design showed even stress distribution consistently within anatomical constraints. CONCLUSIONS: Computational analyses suggested the feasibility of a lighter, cheaper and patient-specific cage for spinal fusion