Magnetic Resonance Imaging-Derived Microvascular Perfusion Modeling to Assess Peripheral Artery Disease

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Biology

Background:

• Computational fluid dynamics (CFD) models align well with contrast-enhanced magnetic resonance imaging (CE-MRI) in cardiovascular applications.
• Microvascular dysfunction is a key factor in peripheral artery disease (PAD) affecting skeletal muscle perfusion.
• Developing advanced computational models is crucial for understanding microvascular behavior in disease states.

Purpose of the Study:

• To develop and validate a biomechanical model of microvascular perfusion in skeletal calf muscles using CE-MRI data.
• To investigate differences in microvascular parameters between patients with PAD and healthy controls.
• To assess the association of computational microvascular model parameters with clinical markers of PAD severity and exercise capacity.

Main Methods:

• A computational microvascular model was developed using CE-MRI signal intensities from calf muscles.
• 56 participants (36 PAD patients, 20 controls) underwent CE-MRI and ankle-brachial index (ABI) testing at rest and after exercise.
• Key microvascular parameters (transfer rate constant, interstitial permeability, porosity, outflow filtration coefficient, microvascular pressure) were quantified and analyzed.

Main Results:

• Patients with PAD exhibited higher transfer rate constant, interstitial permeability, and microvascular pressure, with lower porosity and outflow filtration coefficient compared to controls (P≤0.014).
• All model parameters significantly correlated with ABI, claudication onset time, and peak walking time across all participants (P≤0.013).
• Within the PAD group, non-completers of treadmill exercise showed higher interstitial permeability and microvascular pressure, and lower porosity and outflow filtration coefficient than completers (P≤0.001).

Conclusions:

• Computational microvascular modeling parameters significantly differentiate PAD patients from controls.
• The model parameters are associated with clinical indicators of PAD severity and functional limitations.
• This computational approach holds potential for advancing the study of lower extremity ischemia and PAD.