BS14 Image-based computatinoal simulations of foetal heart function to understand congenital malformations and foetal heart intervention

BackgroundSome Congenital Heart Malformations develops because of cardiac abnormalities during mid-gestation, which prevents normal development for the rest of gestation to lead to the malformation at birth. An example is foetal critical aortic stenosis, where an outflow obstruction causes high left ventricle (LV) pressures, low myocardial strains, and severe mitral regurgitation (MR). These abnormal conditions cause hypoplastic left heart Syndrome (HLHS) by birth in most cases, and are thus evolving HLHS cases. In such cases, catheter-based foetal aortic balloon valvuloplasty in utero interventions was shown to be promising in relieving the abnormal biomechanics, significantly reducing chances of single ventricular birth. However, the biomechanical nature of stenosis and intervention remain poorly characterized, even though they both have significant biomechanical effects. Further, our ability to predict outcomes of disease or intervention is very limited. We hypothesize that advanced image-based biomechanical simulations can improve our understanding, and can be used as a tool to better predict intervention outcomes. Here, we present preliminary work towards testing this hypothesis.Methods4D echocardiography images of foetal hearts from both healthy and diseased (critical aortic stenosis) foetuses were analysed for numerical reconstruction of the LV and its motion, using validated motion tracking algorithms. Image-based patient specific Finite Element Modeling of the LV was conducted to determine the biomechanical effects of various individual features of foetal aortic stenosis. This model featured both active tension and passive stiffness of myocardium, spatially varying myofiber orientations, and a simplified Windkessel model to describe ventricular-vascular coupling. Image-based computational fluid dynamics modeling of the LV was also conducted to understand flow patterns and forces in the LV during disease, compared to healthy hearts.ResultsFetal aortic stenosis alone was found to elevate LV pressures by 10-20 mmHg, and could drastically decrease stroke volume and myocardial strain, depending on severity. These effects were moderated down by MR. Our modelling indicated that stenosis alone could not lead to regurgitation velocities as high as clinical observations, unless hypertrophic wall thickening occurred. Indeed, our clinical data showed approximately 105±37% wall thickening. Modelling further indicated that this typical extent of hypertrophy pro