Magnetic resonance imaging based modeling of microvascular perfusion in patients with peripheral artery disease.

Abstract

Peripheral artery disease (PAD) is associated with an increased risk of adverse cardiovascular events, impaired lower extremity blood flow and microvascular perfusion abnormalities in the calf muscles which can be determined with contrast-enhanced magnetic resonance imaging (CE-MRI). We developed a computational model of the microvascular perfusion in the calf muscles. We included 20 patients (10 PAD, 10 controls) and utilized the geometry, mean signal intensity and arterial input functions from CE-MRI calf muscle perfusion scans. The model included the microvascular pressure (p_v), outflow filtration coefficient (OFC), transfer rate constant (k^t), porosity (φ), and the interstitial permeability (K_tissue). Parameters were fitted and the simulations were compared across PAD patients and controls. Intra-observer reproducibility of the simulated mean signal intensities was excellent (intraclass correlation coefficients >0.995). k^t and K_tissue were higher in PAD patients compared with controls (4.72 interquartile range (IQR) 3.33, 5.56 vs. 2.47 IQR 2.10, 2.85; p = 0.003; and 3.68 IQR 3.18, 4.41 vs. 1.81 IQR 1.81, 1.81; p < 0.001). Conversely, porosity (φ) was lower in PAD patients compared with controls (0.52 IQR 0.49, 0.54 vs. 0.61 IQR 0.58, 0.64; p = 0.016). Porosity (φ) was correlated with the ankle brachial index (r = 0.64, p = 0.011). The proposed computational microvascular model is robust and reproducible, and essential model parameters differ significantly between PAD patients and controls.