Congenital heart defects (CHDs) are diagnosed in 8 in 1000 newborns worldwide. Due to the growing rate of patients with failed Fontan physiology, the hemodynamics of patient with severe CHD illness has recently become of interest . Studies focusing pulmonary arteriovenous malformations (PAVM) and the hemodynamics of templates for Fontan patients have been conducted through computational fluid dynamics (CFD) [2, 3]. In these studies, different surgical alternatives are compared for a surgical stage, while the improvements from the previous surgical stages have not been emphasized strongly.
In this study three complex single ventricle patients having Interrupted IVC and azygous vein continuation are recruited for cardiac MRI imaging. An experimentally validated  openFoam modified solver compiled with user defined passive scalar equations was employed to determine the hepatic factor distribution in the pre-surgical model for each of the virtual surgery scenarios. The patient-specific anatomic data for boundary conditions (inflow for inlets and pressure-resistance for outlets) was used to reach the desired flow distribution.
A variety of virtual surgery options were modeled by running 11-14 different CFD models for each patient and the sensitivity of hepatic flow distributions plus power loss were calculated in each simulation. For brevity, the results of two different surgical strategies for Patient "A", a 14 year old male with dual superior vena cava (SVC) with an interrupted inferior vena cava (IVC) and azygous continuation, is shown in Figure 1.
Figure1: Two surgical alternatives for Hepatic Vein and Right Glenn surgery for Patient "A" (a) Intra-atrial IVC baffle (Diameter=15mm) anastomoses (b) Single hepatic baffle (Diameter=17mm) to left azygous with R-SVC Glenn closer to left.
This patient undergoes the R-SVC Glenn and IVC baffle-anastomoses in the current stage of therapy. The preferred surgical scenario is the Figure 1 (a) configuration where the hepatic flow is split among the right and left pulmonary arteries by 37% and 63% respectively. Power loss is computed to be 3.03 mW for this surgical scenario which is relatively low compared to the other surgical scenarios (e.g Figure 1 (b)).
For verification, the results have been compared with the post operating data both qualitatively and quantitatively and show good agreements. Furthermore, the sensitivity of the optimal baffle design to the superior inflows underscores the need to characterize both preoperative anatomy and hepatic flows to identify the best option. Presented experience is performed in a real clinical setting and confirmed the feasibility of pre-surgical approach.
Oguz, G.N., et al., J. of Medical Devices,2017.11(2): p.021012-021012-8.
Marsden, A.L., et al, The J. of Thoracic and Cardiovascular Surgery,2009.137(2):p394-403.e2.
Pekkan, K., et al., Annals of biomedical engineering,2009.37(1):p.50-63.