Patient Specific Fluid and Solid Mechanics Modeling of Pulmonary Hypertension

Pulmonary hypertension (PH) is an idiopathic vascular condition characterized by high blood pressure in the pulmonary arteries and typically affects 15 - 50 persons per million. Aberrant hemodynamics trigger remodeling of the extracellular matrix in the pulmonary arteries, which substantially compromises arterial compliance and eventually leads to right heart failure. Currently, PH diagnosis requires a standard of care right heart catheterization (RHC), which is highly invasive and considerably limits the cardiologist’s ability to safely diagnose and monitor the disease. Developing safe and repeatable non-invasive surrogates for RHC would be of great importance and have high translational potential. Imaging methods, such as cardiac magnetic resonance imaging (MRI), have shown promise as potential non-invasive substitutes, while having a limited spatial resolution. Alternatively, in silico models can play a critical role in identifying disease markers. Therefore, computing the patient specific pulmonary hemodynamics and wall mechanics may lead to the identification of biomarkers associated with a developing PH condition.

In this work, we devised a three-fold methodology where we inter-dependently quantified the hemodynamics (via CFD), ex vivo material properties (via planar biaxial tensile testing), and wall mechanics (via finite element analysis - FEA), occurring in the human pulmonary vasculature during PH. RHC data obtained retrospectively from 32 adult PH patients was utilized to inform the spatiotemporally varying patient specific hemodynamics over three cardiac cycles. The patient specific vascular geometries were reconstructed from segmented chest computed tomography angiography. Material properties of the pulmonary arteries were assessed ex vivo by planar biaxial tensile testing of porcine pulmonary artery specimens. The pressure fields obtained from the CFD simulations were mapped onto the endoluminal surface of the vasculature and used with a FEA solver to quantify the stress and strain at the wall. From the results of the CFD and FEA simulations, we evaluated multiple simulation-based indices and evaluated their association with clinical metrics measured or calculated during RHC. We also characterized the influence of spatiotemporal localization of the simulation-based indices on these associations. The primary outcome of this research is the statistically significant correlations found between the clinical metrics and the hemodynamic and wall mechanics indices obtained from the in silico models of the pulmonary vasculature. We contemplate that the comprehensive analysis presented in this work will provide the critical inputs in directing the advancement of future imaging methods that can be used as surrogates for PH invasive diagnostic procedures.