Thoracic Endovascular Aortic Repair (TEVAR) is a minimally invasive surgical procedure, consisting of the intraluminal placement of one or more stent-grafts to treat thoracic aortic pathologies such as aneurysms and acute aortic syndromes. Although TEVAR is generally recommended over traditional open repair by current clinical guidelines for its lower invasiveness and reduced complication rates, several challenges remain in surgical planning to ensure the long-term performance of the implanted device. Computational simulations can play a key role in supporting this process by reproducing the deployment of the stent-graft within the aorta and predicting potential complications; however, their reliability must be proved through validation. To contribute to the expansion of the current literature in this field, this thesis aims to validate pre- and post-TEVAR fluid-structure interaction (FSI) simulations using in vitro 4D-flow MRI data, incorporating accurate reconstructions of both the aortic geometry and the stent-graft. The same patient-specific pathological aortic model was used for both pre- and post-TEVAR FSI simulations, 3D-printed using a compliant resin and integrated into a flow loop replicating physiological circulation. For the post-TEVAR model, a Medtronic Valiant Captivia stent-graft was implanted replicating the real intervention. Flow and pressure sensors were added to the experimental setup to acquire curves used as boundary conditions for numerical simulations. For validation purposes, 2D-flow and 4D-flow MRI acquisitions provided geometric and velocity information at three key cross-sections and within the whole aortic domain, before and after TEVAR. The comparison between experimental and simulated results showed good qualitative and quantitative agreement in terms of aortic distensibility and flow dynamics. Mean absolute percentage errors between normalized cross-sectional area curves were below 1% pre-TEVAR and around 2–3% post-TEVAR, while the Mann–Whitney U test confirmed no statistically significant differences between FSI and MRI through-plane velocity distributions at key time points of the cardiac cycle. Despite moderate discrepancies in through-plane velocity distributions observed during the phases of rapid flow variation, the overall results provide the first evidence in the literature supporting the reliability of FSI models in accurately reproducing realistic hemodynamic conditions of the thoracic aorta, in both pre- and post-operative configurations, including the integration of a device model within the fluid domain.
La Riparazione Endovascolare dell’Aorta Toracica (TEVAR) è una procedura chirurgica mini-invasiva che consiste nel rilascio intraluminale di uno o più stent-graft per il trattamento di patologie dell’aorta toracica, quali aneurismi e sindromi aortiche acute. Sebbene il TEVAR, per la sua minore invasività e i ridotti tassi di complicanze, sia generalmente raccomandato dalle linee guida cliniche rispetto alla chirurgia tradizionale a cielo aperto, la pianificazione dell’intervento presenta ancora diverse sfide da superare per garantire la corretta funzionalità del dispositivo nel lungo periodo. Le simulazioni computazionali possono supportare questo processo, consentendo di riprodurre il rilascio dello stent-graft all’interno dell’aorta e di prevedere possibili complicanze; tuttavia, l’affidabilità dei risultati deve essere dimostrata attraverso un processo di validazione. Per contribuire all’ampliamento della letteratura in questo ambito, la presente tesi ha come obiettivo la validazione di simulazioni di interazione fluido-struttura (FSI) pre- e post-TEVAR, con un’accurata modellazione sia dell’aorta sia dello stent-graft, utilizzando dati sperimentali ottenuti da acquisizioni in vitro di risonanza magnetica (RM) 4D-flow. Lo stesso modello aortico, paziente-specifico, è stato utilizzato per entrambe le simulazioni, stampato 3D con una resina elastica ed inserito in un circuito di flusso che permettesse di riprodurre la circolazione fisiologica. Per il modello post-TEVAR, uno stent-graft Medtronic Valiant Captivia è stato impiantato simulando la reale procedura chirurgica. Sensori di flusso e di pressione sono stati impiegati nell’apparato sperimentale per l’acquisizione delle curve utilizzate come condizioni al contorno delle simulazioni numeriche. Ai fini della validazione, acquisizioni RM 2D- e 4D-flow hanno fornito informazioni sulla geometria e la velocità rispettivamente in tre sezioni trasversali e nell’intera aorta, prima e dopo il TEVAR. Il confronto tra risultati sperimentali e numerici ha mostrato un buon accordo qualitativo e quantitativo in termini di distensibilità aortica e fluidodinamica. L’errore percentuale assoluto medio tra le curve di area normalizzata è risultato inferiore all’1% nel caso pre-TEVAR e compreso tra il 2% e il 3% nel caso post-TEVAR, mentre il test di Mann–Whitney U ha confermato l’assenza di differenze statisticamente significative tra le distribuzioni di velocità ottenute tramite FSI e RM in istanti chiave del ciclo cardiaco. Nonostante alcune moderate discrepanze nelle distribuzioni di velocità durante le fasi in cui il flusso varia più rapidamente, i risultati complessivi forniscono la prima evidenza in letteratura a supporto dell’affidabilità dei modelli FSI nel riprodurre accuratamente condizioni emodinamiche realistiche dell’aorta toracica, sia nella configurazione pre-operatoria sia in quella post-operatoria, includendo anche l’integrazione di un modello di dispositivo all’interno del dominio fluido.
Validation of pre- and post-TEVAR FSI simulations using in vitro 4D-flow MRI data
Macioce, Emanuele
2024/2025
Abstract
Thoracic Endovascular Aortic Repair (TEVAR) is a minimally invasive surgical procedure, consisting of the intraluminal placement of one or more stent-grafts to treat thoracic aortic pathologies such as aneurysms and acute aortic syndromes. Although TEVAR is generally recommended over traditional open repair by current clinical guidelines for its lower invasiveness and reduced complication rates, several challenges remain in surgical planning to ensure the long-term performance of the implanted device. Computational simulations can play a key role in supporting this process by reproducing the deployment of the stent-graft within the aorta and predicting potential complications; however, their reliability must be proved through validation. To contribute to the expansion of the current literature in this field, this thesis aims to validate pre- and post-TEVAR fluid-structure interaction (FSI) simulations using in vitro 4D-flow MRI data, incorporating accurate reconstructions of both the aortic geometry and the stent-graft. The same patient-specific pathological aortic model was used for both pre- and post-TEVAR FSI simulations, 3D-printed using a compliant resin and integrated into a flow loop replicating physiological circulation. For the post-TEVAR model, a Medtronic Valiant Captivia stent-graft was implanted replicating the real intervention. Flow and pressure sensors were added to the experimental setup to acquire curves used as boundary conditions for numerical simulations. For validation purposes, 2D-flow and 4D-flow MRI acquisitions provided geometric and velocity information at three key cross-sections and within the whole aortic domain, before and after TEVAR. The comparison between experimental and simulated results showed good qualitative and quantitative agreement in terms of aortic distensibility and flow dynamics. Mean absolute percentage errors between normalized cross-sectional area curves were below 1% pre-TEVAR and around 2–3% post-TEVAR, while the Mann–Whitney U test confirmed no statistically significant differences between FSI and MRI through-plane velocity distributions at key time points of the cardiac cycle. Despite moderate discrepancies in through-plane velocity distributions observed during the phases of rapid flow variation, the overall results provide the first evidence in the literature supporting the reliability of FSI models in accurately reproducing realistic hemodynamic conditions of the thoracic aorta, in both pre- and post-operative configurations, including the integration of a device model within the fluid domain.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_Macioce_Tesi.pdf
non accessibile
Descrizione: Tesi
Dimensione
5.87 MB
Formato
Adobe PDF
|
5.87 MB | Adobe PDF | Visualizza/Apri |
|
2025_12_Macioce_Executive Summary.pdf
non accessibile
Descrizione: Executive summary
Dimensione
1.44 MB
Formato
Adobe PDF
|
1.44 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246422