Related Experiment Video
Updated: Apr 13, 2026
Author Spotlight: Simulating Pediatric Cardiac Surgery Using a Neonatal Piglet Model
Published on: May 26, 2023
Computational Modeling of Neonatal Cardiopulmonary Bypass Hemodynamics With Full Circle of Willis Anatomy
Senol Piskin1, Akif Ündar2, Kerem Pekkan1,3
1Department of Mechanical Engineering, Koc University, Istanbul, Turkey.
Insights
Accurate head-neck blood perfusion during neonatal cardiopulmonary bypass (CPB) requires including the 3D Circle of Willis (CoW). Cannula placement significantly impacts blood flow distribution, crucial for surgical strategy and minimizing complications.
Area of Science:
- Biomedical Engineering
- Computational Fluid Dynamics
- Pediatric Cardiology
Background:
- Cardiopulmonary bypass (CPB) is vital for congenital heart defect (CHD) repair in neonates.
- Cannula placement in CPB critically influences hemodynamic performance.
- Previous computational studies simplified the neonatal cerebral system.
Purpose of the Study:
- To integrate a 3D patient-specific cerebral model, including the Circle of Willis (CoW), with a hypoplastic neonatal aortic arch for enhanced CPB simulations.
- To compare hemodynamic differences between innominate artery (IA) and ductus arteriosus (DA) cannulation configurations.
- To assess the impact of CoW inclusion on predicting head-neck blood perfusion.
Main Methods:
- Developed a 3D patient-specific computational model integrating MRI-derived cerebral vasculature (including CoW) with a neonatal aortic arch.
- Utilized a commercial, experimentally validated CFD solver for flow simulations at Re=2150.
- Applied steady velocity inlet boundary conditions for the CPB pump and resistance boundary conditions for all outlets.
Main Results:
- Inclusion of the 3D CoW is essential for accurate head-neck blood perfusion prediction.
- Significant differences in flow distribution were observed between IA- and DA-cannula configurations.
- DA-cannulation resulted in higher flow to the ascending aorta (198%) and cerebral arteries (91% right, 68% left) compared to IA-cannulation. Retrograde vertebral artery flow occurred with IA-cannulation but not DA-cannulation.
Conclusions:
- The integrated 3D CoW model is critical for preoperative planning of neonatal CPB cannulation strategies.
- Cannula configuration significantly alters blood flow dynamics, impacting perfusion to vital organs.
- Findings support improved neonatal arterial cannula design to mitigate CPB-related cerebral complications.
Abstract:
Cardiopulmonary bypass (CPB) procedure is employed to repair most congenital heart defects (CHD). Cannulation is a critical component of this procedure where the location and diameter of cannula controls the hemodynamic performance. State-of-the-art computational studies of neonatal CPB employed an isolated aortic arch region by truncating the three-dimensional (3D) patient-specific cerebral system. The present work expanded these studies where the 3D patient-specific MRI reconstruction of the cerebral system, including the Circle of Willis (CoW), is integrated with a hypoplastic neonatal aortic arch. The inlet of the arterial cannula is assigned a steady velocity boundary condition of the CPB pump, while all outlets are modeled as resistance boundary conditions, thus allowing acute comparisons between different cannula configurations. Three-dimensional (3D) flow simulations in the aortic arch model are performed at a Reynolds number of 2150 using an experimentally validated commercial solver. Results demonstrate that the inclusion of 3D CoW is essential to predict the accurate head-neck blood perfusion and therefore critical in deciding the neonatal aortic cannulation strategy preoperatively. Using this integrated model two CPB configurations are studied, where the cannulas were placed at innominate artery (IA) (IA-cannula configuration) and ductus arteriosus (DA) (DA-cannula configuration). Configuration change produced significant differences in flow splits and local hemodynamics of blood flow throughout the whole aortic arch, neck and cerebral arteries. Percent flow rate differences between the IA- and DA-cannula configurations are computed to be: 19%, for descending aorta, 198% for ascending aorta (perfusing coronary arteries), 91% for right anterior cerebral artery, and 68% for left anterior cerebral artery. Another important finding is the retrograde flow at vertebral arteries for IA-cannula configuration, but not for DA-cannula. These results may help to translate better neonatal arterial cannulae design for minimizing cerebral complications during CPB procedures.
