Non-Invasive Blood Flow Modelling Studies

Principal Investigator: Prof. C. Ross Ethier

Department of Mechanical and Industrial Engineering, University of Toronto, 5 King's College Rd., Toronto, Ontario, Canada

ethier(at)mie.utoronto.ca

Keywords: non-invasive blood flow, medical imaging, atherogenesis

Beginning and End of the Project: 1996-1998

Background and Aims

Background: Diseases of the cardiovascular system are the leading cause of death in Western society. It is well known that arteries have the ability to remodel in response to blood flow patterns, and that blood flow patterns play a role in the pathogenesis and progression of a number of arterial diseases, such as atherosclerosis and distal anastomotic intimal hyperplasia. It is also known that vascular endothelium responds to shear stress, i.e. to the stress exerted by flowing blood on the arterial wall. These facts have led many investigators to study blood flow (hemodynamic) patterns in the large arteries, with the goal of linking hemodynamically-induced wall shear stress distributions to arterial wall pathology.

Typically, animal models are used for such studies. Dogs are most commonly used in such experiments, but rabbits, domestic swine, and baboons are also frequently employed. Although animal models are useful, there are also limitations to their applicability.  Primary among these is the fact that some disease processes appear to differ in humans as compared to animals.  For this reason, as well as to minimize the use of animals in such studies, it would be optimal to use human subjects in experiments designed to correlate blood flow patterns and arterial pathology. To do so, we proposed to use three-dimensional medical imaging techniques (primarily magnetic resonance imaging) to visualize the arteries and computer modelling to determine the detailed blood flow patterns within the arterial segment of interest.

Aim: The project had three specific aims:

1. Validate and test our approach by using a standard modelled artery having known but complex geometry, so as to determine the spatial fidelity of the MR tissue imaging approach and the accuracy of the subsequent numerical blood flow simulations.
2. Refine our technique by carrying out tests using excised abdominal aortic specimens obtained from cadavers at autopsy, in order to optimize the efficiency of the tissue imaging/flow modelling process.
3. Carry out proof‑of‑concept experiments using human volunteers, focussing on blood flow and wall shear stress patterns in the abdominal aorta.  Numerically simulated aortic blood flow patterns will be compared to measurements made using MR flow imaging within our labs, as well as published data.

Methods and Results

The progress that was made towards each specific aim is listed below.

For Specific Aim 1: We successfully completed this aim. The results showed that excellent spatial fidelity and accuracy could be achieved. This has been documented in two publications:

1. J.A. Moore, D.A. Steinman and C.R. Ethier, "Computational Blood Flow Modelling: Errors Associated with Reconstructing Finite Elements Models from Magnetic Resonance Images", J.  Biomechanics, Vol. 31, pp.179-184, 1998.
2. J.A. Moore, D.A. Steinman, D.W. Holdsworth and C.R. Ethier, “Accuracy of computational hemodynamics in complex arterial geometries reconstructed from MRI”, Annals of Biomedical Engineering, Vol. 27, pp. 32-41, 1999.

For Specific Aim 2: We successfully completed this aim, although we slightly modified the procedure to study an end-to-side anastomosis specimen rather than an aortic specimen. This was documented in a publication:

1. J.A. Moore, D.A. Steinman, S. Prakash, K.W. Johnston and C.R. Ethier, “A numerical study of blood flow patterns in anatomically realistic and simplified end-to-side anastomoses”, ASME J. Biomechanical Engineering, Vol. 121, pp. 265-272, 1999.

For Specific Aim 3: We did not carry out the proof-of-concept experiments, since a former student of mine independently carried out these experiments in human volunteers using technology that we developed through the Doerenkamp Foundation grant.  This work has been very successful, see for example: Milner JS, Moore JA, Rutt BK, Steinman DA, Hemodynamics of human carotid artery bifurcations: computational studies with models reconstructed from magnetic resonance imaging of normal subjects. J Vasc Surg 1998 Jul;28(1):143-56. Further studies in human volunteers have been carried out in numerous labs using this approach.

Conclusions and Relevance for 3R: Support from the Foundation allowed us to develop, optimize and validate techniques for non-invasively determining detailed blood flow patterns. These techniques have been published and are now available to all investigators as a tool to minimize the use of animals in hemodynamic studies.