Computational Modelling of Mass Transport in Large Arteries

In the present study, a fluid-wall model was developed to simulate arterial mass transport of macromolecules in atherosclerosis-prone arteries. To incorporate time-dependent transport processes in the arterial wall, two time-averaged numerical procedures were proposed and compared. Model parameters, especially the transport properties of the arterial wall layers, were estimated using a simulation-based optimisation approach. Furthermore, shear-dependent transport properties, such as hydraulic conductivity and permeability were derived from relevant experimental data in the literature to incorporate the shear-dependence of trans-endothelial transport. The mathematical model and corresponding parameters were tested in idealised arterial geometries as preliminary investigations. The model was then applied to a patient-specific case, albumin transport in a human right coronary artery. The main outcomes of this study include the following. 1) The determination of four sets of model parameters, including albumin transport parameters under transmural pressure of 70 mmHg, LDL transport parameters under transmural pressures of 70 mmHg, 120 mmHg, and 160 mmHg. It was shown that the simulation-based optimisation approach led to much better parameter estimations than existing methods. 2) Preliminary model tests showed that the conventional steady flow assumption in the arterial mass transport models may not be adequate, especially when the bulk flow is highly disturbed. The LFTA numerical procedures provided an efficient way to incorporate the effects of flow pulsatility on shear-dependent trans-endothelial transport. Results suggested that the influence of transient haemodynamic conditions on macromolecular transport can be modelled as a time-averaged effect. 3) Albumin accumulation in the subendothelial layer was found to be co-localised with low WSS, implying that the arterial wall exposed to low WSS is atherosclerosis-prone due to a greater lipid accumulation. The accumulation was mainly due to a greater influx in the low WSS regions where the endothelium is more permeable. A number of limitations existed in the present study, including the use of a vascular scale model, parameter-related limitations, and limitations in the time-averaged numerical procedures.