Patient-Specific 3D-0D Coupled Cardiovascular Model to Inform Pulmonary Valve Replacement in Tetralogy of Fallot Patients

  • Arjoune, Tahar (Technical University of Munich)
  • Meierhofer, Christian (German Heart Center Munich)
  • Stern, Heiko (German Heart Center Munich)
  • Ewert, Peter (German Heart Center Munich)
  • Gee, Michael (Technical University of Munich)

Please login to view abstract download link

One of the most common late consequences after repair of Tetralogy of Fallot (ToF) is pulmonary valve regurgitation. Significant pulmonary valve regurgitation progressively induces right ven- tricular enlargement, dysfunction, and is an important factor in late morbidity and mortality. While it has been shown that pulmonary valve replacement improves symptoms and functional status in these patients, the optimal timing and indications for the replacement intervention after repaired ToF are still debated. Therefore, a patient-specific 3D-0D coupled computa- tional model of the individualized cardiovascular mechanics aims to support the physicians in their decision-making by assessing the preoperative cardiovascular function of the patient and predicting its postoperative development. The 3D-0D coupled closed-loop model of the heart and the vascular system delivers valuable quantities of interest for the physicians, such as blood pressure, blood flow and the contraction of the myocardium [1]. The heart muscle is modeled as a 3D nonlinear anisotropic elastic solid with both active and passive material components. The patient-specific geometry of the ventricular myocardium is extracted from MRI data. The entire vascular system is modeled with 0D lumped parameter windkessel models. The valve models are also part of the 0D sub-model and are based on the Bernoulli formulation proposed in [2], which can describe the regurgitation of leaky valves and is parametrized in a physically interpretable way. The whole nonlinear system of equations is solved with a monolithic Newton solver, where the conservation of mass and momentum enforces the coupling between 0D and 3D variables. The patient-specific parameters of the windkessel, valve and active stress models are calibrated using inverse problem frameworks based on several preoperative measurements of blood pressure and flow, as well as on cine MRI. Results of the postoperative state prediction using a ToF patient calibrated model are physio- logically meaningful and correlate with the postoperative measurements.