Prediction of Thoracic Endovascular Aortic Repair Intervention: Proof of Concept using Digital Twin Technology

Thoracic endovascular aortic repair (TEVAR) has become the preferred treatment for thoracic aortic pathologies, with current planning based on geometric assessment, although it neglects device-aorta interaction. This manuscript reports the use of a validated high fidelity finite element methodology to replicate the TEVAR procedure. The study aimed to establish the credibility of the TEVAR digital twin for clinical use, providing a reliable tool for pre-procedural planning in patient specific cases.

Pre- and post-operative computed tomography angiography (CTA) images of patients who underwent TEVAR using the Valiant Captivia stent graft were analysed retrospectively. To ensure the accuracy and reliability of the TEVAR digital twin, the simulation results were rigorously validated against post-operative image reconstructions. The patient specific anatomy was segmented using the level set technique from pre-operative CTA images, and a digital twin of the corresponding stent graft model was generated. Explicit finite element numerical simulations were conducted to replicate the stent graft implantation. Simulation validation analysis compared the simulated stent configuration with the stent reconstructed from post-operative images, using the stent ring opening area percentage error (eOA%) as a quantitative metric.

Twenty patients were included. The simulations replicated stent graft implantation in typical patient specific anatomies, with eOA% consistently below 10% per stent ring, averaging 4%. Notably low errors were observed in the proximal and distal stent rings, critical for ensuring sealing (< 3.8% average). These analyses, completed in about 10 hours each, showed improved accuracy compared with values reported in the literature.

The credibility of a TEVAR digital twin is essential for its trustworthiness and integration into clinical practice. The findings, achieved in a short time, validate the use of numerical simulations for predicting short term stent graft post-operative geometry in patient specific anatomies. By enabling personalised planning, digital twins could help optimise procedure strategies, reduce complications, and improve TEVAR outcomes.