Abstract Background Pulmonary hypertension associated with interstitial lung disease (ILD-PH) is a progressive and often fatal condition characterized by elevated pulmonary vascular resistance despite heterogeneous parenchymal involvement. The underlying mechanisms remain incompletely understood, and whether proximal anatomic distortion drives the hemodynamic burden remains uncertain. Computational fluid dynamics (CFD) offers a means to quantify pulmonary vascular geometry and flow patterns, providing insight into structure-function relationships across disease phenotypes. In this study, we applied CFD modeling to compare pulmonary arterial anatomy and hemodynamics between ILD-PH, idiopathic pulmonary arterial hypertension (PAH), and a healthy control, testing the hypothesis that vascular distortion is the primary driver of ILD-associated pulmonary hypertension. Methods Three-dimensional pulmonary artery reconstructions were generated from contrast-enhanced CT angiography using SimVascular, an open-source platform for vascular modeling. Models were constructed from age and sex matched patients as follows: one ILD-PH patient, one idiopathic PAH patient, and healthy control. Meshing and steady-state CFD simulations were performed using subject-specific inlet and outlet boundary conditions derived from right heart catheterization and echocardiographic data. Quantitative metrics included mean and maximum vessel tortuosity, average luminal diameter, wall shear stress (WSS), and pressure gradients across successive vascular generations. Results Average vessel tortuosity was similar between the ILD-PH, PAH, and control models, indicating comparable overall vessel curvature. However, maximum tortuosity was reduced in both ILD-PH and PAH, suggesting attenuation of localized geometric distortion. Both disease models exhibited diffuse pulmonary arterial dilation, consistent with remodeling and compliance changes observed in chronic pulmonary hypertension. Global WSS was markedly reduced in ILD-PH and PAH compared to the healthy control, reflecting diminished shear-mediated endothelial signaling. Simulated pressure drops across the pulmonary arterial tree were minimal in both disease models and lower than in the healthy control, indicating preserved proximal flow transmission and implicating distal vascular remodeling as the dominant contributor to elevated pulmonary resistance. Conclusion CFD modeling reveals that proximal anatomical distortion does not account for the hemodynamic impairment seen in ILD-PH. The combination of global arterial dilation, reduced WSS, and minimal proximal pressure gradients supports a pathophysiologic model centered on distal microvascular rarefaction rather than proximal obstruction. These findings demonstrate the utility of CFD in distinguishing structural from microvascular contributions to pulmonary hypertension and highlight its potential role in refining disease phenotyping and therapeutic targeting. This abstract is funded by: None
Gilani et al. (Fri,) studied this question.