4D Motion of the Aorta in the Chest and Simulated Wall Stress Distribution in Relation to Aortic Events

Overall design The study population will consist of patients to be examined for proximal aortic conditions (dissection or aneurysm) requiring surgery or endovascular intervention. A patient cohort will be recruited to explore the effects of surgical or endovascular prosthesis on the adjacent aorta, and its impact on late clinical outcomes.

Patient recruitment Patients who have proximal aortic conditions and are referred to the aortic team in Royal Brompton and Harefield Hospitals will be screened for eligibility. For this pilot study and in view of the volume of aortic surgery in the Trust (approx. 100 cases per year), 30 patients will be recruited in the first year of this project to be followed for at least one year for clinical outcomes.

Imaging protocol After obtaining written informed consent, a new dynamic (multiphase ECG-gated contrast- enhanced) CT imaging protocol will be used for participants to replace the standard (ECG- gated contrast-enhanced) CT protocol both prior to and after surgical or endovascular intervention. The therapeutic pathway and clinical decision making will not be affected by the new imaging protocol. Study participants will receive the same standard of treatment and care as well as follow up surveillance as all other patients not participating in this study. The same dynamic CT image protocol will be offered prior to hospital discharge.

Image processing and motion analysis Image processing and analysis will be performed offline, and thus not affect the clinical pathway or delay standard treatment. Dynamic images will be extracted from the Trust PACS. Suitable sets of images will be anonymized, reconstructed using a Trust approved software and downloaded to a protected area on a Trust server for offline analysis. The analysis of 4D motion of the aortic root will be conducted using image processing software authorised by the Trust (3D Slicer). Detailed analysis requires the identification of end- systolic and end-diastolic frames, followed by measurements in 6 degrees of freedom (3 directions of displacement and 3 axial rotations in a global coordinate system). The raw motion data will be transferred to a patient-specific anatomical coordinate system (identified by the individual sinotubular junction) for further statistics with Matlab (Mathworks, USA).

Finite element modelling The spatial patient-specific model will be reconstructed from CT image and then automatically meshed for further simulation. The finite element analysis will be performed by using a commercial structural mechanics solver (Abaqus; Dassault Systèmes, France). The 4D motion data will be applied as the boundary condition to describe the motion of the ascending aorta in the model. The pulsatile pressure load from the blood flow will be applied to the inner surface of the model. Virtual surgery or endovascular intervention simulation will then be performed on this model to determine key biomechanical parameters, such as principal stresses and shear strain, which later will be correlated to clinical adverse events during follow-up. The results will allow us to elucidate the role of biomechanical changes in clinical events after surgery/intervention and to identify which parameter might predict adverse outcome in these patients.