Comparative analysis of custom-made endograft simulation techniques: physical prototypes versus numerical simulations

The technical success of fenestrated endovascular aortic repair (FEVAR) relies on the exact fit of the custom-made endograft in the patient. Numerical simulation software has been made available (PLANOP™, PrediSurge) to digitally test the intended endograft design.

The distance between opposite peaks of the proximal endograft sealing ring were measured on computed tomography (CT) scans of the test implantation within a 3D printed model, on numerical simulation software, and on postoperative CT (reference). Two types of modeling were used for the aorta in the numerical simulation software: rigid and deformable. This resulted in four measurements: (1) CT scan of the physical endograft prototype implanted in a rigid printed silicone model of the aorta, (2) rigid numerical or finite element (FE) simulation of the endograft implanted in a rigid aortic model, (3) numerical or FE simulation with a deformable virtual aortic model, and (4) patient postoperative CT.

Ten patients were included in the study. The mean distance between peaks was 26.8 mm in the postoperative CT scan (reference). The distance in the rigid printed model was 23.8 ± 2.0 mm p < 0.003). The FE analysis with rigid aorta measured 23.8 ± 1.5 mm (p < 0.006). Measurements performed on CT of the physical prototype test implanted in the printed silicone model of the aorta, and measurements from FE analysis with rigid modeling, were not significantly different. Measurements in a simulated deformable aorta were 27.4 ± 2.1 mm (p  < 0.521). A high correlation (r = 0.81, p < 0.01) was found between measurements on postoperative CT and deformable aorta FE analysis.

Numerical simulation with a rigid aorta may be a suitable substitute for traditional test implantation of a non-sterile prototype within a 3D aortic model. This may help reduce time for graft planning and fabrication. The FE analysis with a deformable aorta was able to predict peak-to-peak distances at the proximal sealing ring more reliably. The effect on clinical outcomes and endoleak occurrence remains to be elucidated in future trials.