Investigating the link between near-wall hemodynamic descriptors and oxygen transport in human carotid arteries

Atherosclerosis is a vascular disease of the wall of large and medium-sized arteries, leading to inflammation and thickening of the arterial wall with a progressive occlusion of the lumen due to the formation of the atheromatic plaque, a deposition of fatty material on the luminal surface. Several local biological, biomechanical and systemic factors are involved in the onset and progression of the atherosclerotic disease. In particular, near-wall mass transport plays a key role in vascular disease. In detail, oxygen transport is markedly involved in the initiation of the atherosclerotic disease, as hypoxic conditions promote the inflammation of the arterial wall. As for biomechanical factors, local hemodynamics, and, in particular, wall shear stress (WSS) magnitude and topology are the main actors involved in the development of the disease. For this reason, in the last decades, computational fluid dynamics (CFD) has been widely adopted for the study of the complex patterns in the blood flow field and their link to atherosclerosis, providing a huge arsenal of hemodynamic descriptors as markers of atherosclerotic lesion localization and development. Recently a marked interest emerged regarding the WSS vector field topological skeleton with its capability to capture features reflecting cardiovascular complex patterns and having a direct link with adverse vascular biological events. In this context, the aim of this work is to test the capability of near-wall hemodynamic descriptors based on WSS magnitude and topology to provide a reliable template of the near-wall oxygen transport in personalized computational models of human carotid bifurcations, overcoming the high computational cost needed to solve the mass transfer with advection-diffusion equations. To do that, starting from a larger dataset, two carotid bifurcation geometries were selected based on the amount of flow disturbances and meshed to properly model near-wall oxygen transport/transfer. Personalized boundary conditions were prescribed at the inlets and the outlets of the models to perform unsteady-state computational fluid dynamics simulations. The three canonical WSS-based descriptors of disturbed flow (i.e., TAWSS, OSI and RRT), and the WSS topological skeleton features (normalized WSS divergence) were used to investigate near-wall hemodynamics; while the oxygen flux at the arterial wall was estimated based on the Sherwood number (Sh). Final results highlight a marked co-localization between luminal regions exposed to negative normalized WSS divergence and hypoxic condition at the vascular wall (identified by low Sh). As expected, canonical WSS-based descriptors confirmed to be moderate markers of local hypoxia. These observations testify the capability of Eulerian-based WSS topological skeleton to provide an effective template of near-wall oxygen transport/transfer, with a significative reduction of both the computational cost and the complexity associated to classical techniques.