Georgia Tech Researchers Use Numerical Models to Examine Blood Flow Patterns in Artificial Heart Valves
Findings Could Lead to Better Valve Design and Testing
Posted June 5, 2003 | Atlanta
Numerical modeling techniques, developed at the Georgia Institute of Technology to simulate the flow of water around hydraulic structures such as bridge foundations, are being used to better understand the complex patterns of blood flow through artificial mechanical heart valves.
The research could yield the most accurate description to date of the turbulent flow environment blood cells and platelets are exposed to as they pass through an implanted mechanical heart valve - information that researchers hope will lead to improvements in current mechanical valve testing and design procedures, and help reduce the potential for thromboembolic complications.
Backed by a four-year, $1.4 million grant from the National Institutes of Health, the research is being spearheaded by Ajit Yoganathan, Regents' professor in the Wallace H. Coulter School of Biomedical Engineering at Georgia Tech and Emory University and an internationally renowned expert on artificial heart valves and cardiovascular fluid mechanics research; and Fotis Sotiropoulos, an associate professor in the School of Civil and Environmental Engineering and a leading expert in computational fluid mechanics.
To date, the fluid mechanics of heart valves have been largely studied using experimental approaches. The Georgia Tech research is the first comprehensive attempt to develop advanced Computational Fluid Dynamics (CFD) techniques and apply them in conjunction with experiments to study blood flow turbulence in heart valves.
"Our numerical simulations can provide descriptions of the blood flow at a level of detail that far exceeds the insight one can get from experiments alone," Sotiropoulos said. "We will be able to go on a virtual journey along with platelets and blood cells through the valve and identify design elements that induce turbulence patterns, which could be harmful to blood elements. This cannot be done experimentally. Yet, we must rely on experiments to make sure that our virtual blood flow environment closely represents reality."
The team is working to adapt advanced CFD modeling techniques developed for simulating turbulent flows past bridge foundations in natural rivers and through hydraulic turbines in hydropower installations to prosthetic heart valves. In spite of many common elements with hydraulic engineering application, the heart valve problem is so complex that its solution necessitates new advancements and innovation in computational algorithms.
The heart valve consists of two leaflets that are free to open and close as the blood rushes back and forth during the cardiac cycle. To model this fluid/structure interaction, the team is developing sophisticated techniques to account for the coupling of leaflet motion with the blood flow. Unlike hydraulic engineering applications, where the water flow is fast enough to remain chaotic and turbulent all the time, the beating heart causes the blood flow through the valve to change its speed and direction continuously.
For the most part, the flow is ordered and laminar except for a brief period of time near peak systole when the blood flow becomes chaotic and turbulent. It is during this brief interval when engineers suspect that the flow environment could become most hazardous to blood elements. The team will pioneer the development and application of advanced models of turbulence that can accurately model flows continuously transitioning back and forth from a laminar to a turbulent state.
Since 1960, cardiac surgeons have been implanting artificial heart valves in patients who require heart valve replacement. There are two major types available: mechanical valves, which are made from man-made materials; and bioprosthetic valves, which are made from animal tissues.
All mechanical valves are made up of an orifice ring and occluders, either one or two "leaflets" through which blood flows through the heart in a forward direction. In the ring is a single movable disc, or two "leaflets," which open and close much like a natural heart valve to control the flow of blood. Current mechanical heart valves are made from biocompatible materials such as titanium and pyrolytic carbon. A soft fabric sewing ring, which is attached to the valve orifice is utilized by the cardiac surgeon to suture the valve into the patient's heart at the correct anatomical location.
Even though artificial heart valves have saved millions of lives over the past four decades, modern designs are less than ideal. To date, more than 50 different designs have been developed, a majority of them have failed in their clinical utility. The flow of blood across modern mechanical heart valves, such as bileaflet and tilting disc designs, is more turbulent than normal blood flow, which can lead to blood clots and stroke. This requires valve recipients to be on lifelong anticoagulant therapy (blood thinners).
"In the 21st century we need to design and evaluate medical devices such as artificial heart valves using state-of-the-art engineering and biological tools" said Yoganathan, who directs Georgia Tech's Cardiovascular Fluid Mechanics Laboratory. "The marriage of engineering and biological experimental techniques with computational analysis tools is critical to the effort to produce better medical devices."
A unique aspect of the Georgia Tech approach is that it will rely heavily on a close synergy between modelers and experimentalists to produce experimental data sets for validating and refining the simulation tools. In the past, CFD modeling and experiments were conducted independently of each other. The Georgia Tech team will design experiments tailor-made to provide the precise information needed to validate and improve the CFD models. Such synergy would ensure the development of a reliable modeling tool, which, coupled with rapid advancements in computational power, will pave the way for incorporating CFD into a virtual design environment. Engineers and medical doctors will be able to use these computational tools to interactively modify a given valve design and assess the hemodynamic implications of their modifications in real-time.