Correlating Patient-Specific Computational Modeling of Altered Hemodynamics to Cellular Events in Endocardial Fibroelastosis

All articles of this category (opens in new window) Background: Endocardial Fibroelastosis (EFE), a subendocardial thickening characterized by excessive collagen and elastic fiber deposition resulting from hemodynamic alterations, impedes the left ventricle’s (LV) adaptive growth and function, ultimately leading to failure of biventricular (BiV) circulation. While surgical EFE resection is integral to treatment, evidence on the optimal extent and location of removal is limited. We aim to correlate localized hemodynamic measurements with cellular alterations to aid in devising individualized treatment plans for high-risk patients prone to EFE. Methods: We conducted a retrospective analysis involving six patients (aged: 1.5–5 years), four with histologically confirmed EFE in the presence of neonatal critical aortic stenosis and two without cardiac anomalies at Boston Children’s Hospital. Ventricular geometries were reconstructed from real-time three-dimensional echocardiography (RT3DE) and Doppler ultrasound data using a dedicated preprocessing pipeline, enabling the anatomical reconstruction of the ventricle (i.e., mitral and aortic valves, atrium, and aorta). These geometries were utilized in a moving mesh framework to perform computational fluid dynamics (CFD) simulations. Hemodynamic markers were correlated with histologically measured tissue characteristics. Results: Preliminary simulations demonstrate the feasibility of generating patient-specific computational models of intracardiac blood flow using RT3DE and Doppler data. Time-averaged wall shear stress (TAWSS) had a more heterogeneous distribution in the EFE patients, with values up to 8 Pascal in the septal regions of the ventricle, indicative of higher mechanical exposure of endothelial cells. The mean energy loss was about twice as high in EFE patients compared with healthy controls, indicating less efficient cardiac function. The median maximal gradient across the stenotic aortic valve was 56.5 mm Hg (IQR: 47.8–66.8 mm Hg). Histologically, resected EFE had higher amounts of collagen deposition in comparison to patients without cardiac anomalies (mean ± SD: 29.7 ± 14.6% vs. 1 ± 1% of total tissue area). Similarly, elastin deposition was increased (19.1 ± 21.6% vs. 1.3 ± 1.5%). Conclusion: This comprehensive integration of advanced 3D modeling, hemodynamic, cellular, and clinical data effectively combines computational simulation imaging with cellular alterations in patients with EFE. This approach serves as a guideline for individualized decision-making regarding EFE resection and optimizing the outcomes of these patients. Publication History Article published online: 17 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany