CFD Modeling and Visualization Aid Stent Research
- Thursday, 01 July 2010
Once considered a medical miracle, today’s coronary stent installations are now viewed as low-risk and commonplace procedures. In the U.S. alone, more than one million patients with heart disease receive stent implants every year.
While stents save lives, side effects like blood clots and heart attacks remain fairly common, and research in this area is complicated by the complexities of the human arterial system — there are simply too many variables and too much data to weigh and analyze. To help evaluate and analyze the mountain of complex data, researchers are using computational fluid dynamics (CFD) modeling in conjunction with visualization technology to better study how stents and drugs delivered from stents can treat arteries on the one hand, but cause blood clots on the other.
A pioneer in stent research, Dr. Elazer R. Edelman, professor of health sciences and technology at Massachusetts Institute of Technology (MIT) and professor of medicine at Harvard Medical School, is working with Dr. Vijaya B. Kolachalama, post-doctoral associate at MIT's Edelman Laboratory, to improve stents. Together, they are testing and employing the use of CFD modeling that could ultimately lead to the design of customized stents for each patient.
Current Stent Applications
Angioplasty allows surgeons to reopen a blocked artery by inflating a balloon at the end of a catheter inserted into the patient’s artery to compact atherosclerotic plaque against the artery walls, and then install a small metal-mesh tube called a "stent" to keep the plaque from snapping back into the artery. Stents prevent this recoil but tissue grows over the stent as part of a healing response. This can restrict blood flow through the artery and cause blockage. In 25 to 50 percent of cases, the reaction is so severe that another procedure is required.
To prevent re-blocking, scientists began coating the stents with drugs, often imbedded in a thin polymer material for time release. Called "drug-eluting stents," these implants have reduced the need for repeat procedures to under 10 percent — but they create new issues for a small percentage of patients who suffer from life-threatening side effects, including blood clots and heart attacks.
The problem lies in the way the metal mesh stents must sit in the artery to deliver the drug; the stents lie against the wall of the artery but still protrude into the artery lumen. The mesh-like structure of the stent creates alternating flow disruptions similar to how rocks create white water in a flowing stream. The drug comes off the mesh stents at high concentrations and, once released, is subject to areas of high and low flow.
Observing that the chance of blood clots rises along with the amount of drug delivered, researchers realized the need for an effective method for identifying and predicting drug delivery patterns from stents in complex arterial vessels.
Though human trials and animal experiments provide valuable data related to the biological response to stents and drugs, it is difficult to examine the issues of drug distribution variability from these studies alone. Most current imaging technology cannot provide a complete view of the drug patterns, and patient variability renders it impossible to track the large number of variables at play.
CFD modeling has proven to be a perfect tool for integrating physiological data and complex non-intuitive conditions. The goal was to simulate what would happen in an artery under multiple complex but physiologically relevant conditions.
By observing the arterial drug distribution patterns for various settings, the MIT researchers discovered that the drug released from the stent does not uniformly reach all regions of the vessel. This non-uniformity depends on where the stent is placed in the artery as well as the blood flow that is entering the vessel.
The duo gathered massive amounts of biometric data from several sources and in many formats to create mathematical models that would then be solved against multiple variables or parameters. The simulation results, however, were in numerical format, making it nearly impossible for the human brain to interpret.
Seeing is Believing
Enter visualization. Using sophisticated post-processor visualization software, Drs. Edelman and Kolachalama were able to generate visuals that clearly showed how the artery and drugs would behave under different conditions — similar to looking at the visual results from an X-ray. The resulting images provided insight into how drugs deposited from a stent are affected by numerous factors such as the positioning of the stent, changes in blood flow where arteries meet, and blood flow changes created by the stent itself.