Take a prominent Minneapolis cardiologist, pair him with a UC Berkeley professor with access to supercomputers and throw in some technology used to design cars and fighter jets, and what do you get? A way to possibly predict and prevent stroke after heart procedures.

That's the hope of Dr. Robert Schwartz of the Minneapolis Heart Institute Foundation and Shawn Shadden, an assistant professor at Berkeley. They are developing computer models that not only show how particles leave the heart and enter the blood vessels of the brain — potentially causing stroke — but how some particles are more likely to cause damage than others.

"It's neat because it's showing a direct applicability of engineering to medical problems," Schwartz said.

It's also big stuff. Interventional cardiologists are increasingly turning to less invasive procedures to make repairs to the heart and coronary arteries. Using catheters to clear blocked arteries has proved less invasive, allowing patients to get back on their feet and back to their lives more quickly. Medical technology companies are racing to develop all kinds of procedures that can be performed this way, including replacing diseased aortic valves in the heart itself. Called transcatheter aortic valve replacement (TAVR), some analysts are predicting a $2.4 billion worldwide market for the procedure.

But TAVR also leads to an increased risk of debilitating stroke — although Schwartz said no one has been able to prove the exact connection.

"It has to be particulates, although nobody has proved it," he said. "When you open the balloon and deploy the valve, you get a shower of junk — just running a catheter through there, you knock junk loose."

But how much junk? And where does it go? And which of the particles are more likely to cause a stroke? To answer those questions, Schwartz needed some help.

Enter Shawn Shadden, assistant professor of mechanical engineering at the University of California, Berkeley.

Schwartz and Shadden were brought together when the doctor approached people he knew at Stanford University for some advice on how to attack the stroke problem. He had images of the heart and brain circulatory systems from real patients. Shadden, a graduate student at the time, had built a complex model of the aorta and the blood vessels in the brain.

Using computational fluid dynamics, which uses numerical methods and algorithms to solve and analyze problems that involve fluid flows, scientists create models to see how airplanes and race cars will perform in wind tunnels.

Schwartz and Shadden used the same thing to look at what might happen when particles are scraped loose from the aorta by a catheter.

Continuing their work as Shadden left Stanford for the Illinois Institute of Technology in Chicago, then to his current position at Berkeley, they looked at what happened with different size particles. They found that the biggest particles leaving the aorta do not necessarily pose the greatest risk for stroke. Shadden's modeling showed those particles tended to linger near the aorta or go down toward the legs.

The tiniest particles, too, appear no more likely to cause stroke. They go everywhere blood goes, washing through the circulatory system like tiny specks on a fast-moving river.

The danger zone: 1 mm

The particles most likely to cause a stroke, they found, were medium-sized particles of about 1 millimeter in size. "That 1 millimeter is big enough to cause a pretty massive stroke," Shadden said.

They also are looking at data that show how differences in patient anatomy may lead to differences in where a stroke occurs — left brain or right brain.

More research needs to be done to validate their findings outside the computer. But they have published their early findings and have elicited interest from Keystone Heart, an Israeli medical technology company developing screens to be used during TAVR procedures to prevent stroke.

Schwartz has written grant proposals to the American Heart Association and the National Institutes of Health.

"This is a science project right now," Schwartz said of where the work is leading. "The commercial application may be, eventually, in diagnosis."

Dr. Augusto Pichard is an interventional cardiologist at Washington Cardiology Center in Washington D.C. He has performed thousands of procedures to open clogged coronary arteries and repair the heart. What Schwartz and Shadden are learning could be critical information for improving patient safety, he said.

"It provides information that, to this point, I have never had," he said. "Patterns of blood flow and the flow into important branches can be different within each patient under different conditions. To this point, we have had no knowledge of that."

Greater understanding of blood flow and the particles that could contribute to stroke could prevent major problems, Pichard said.

"TAVR is the greatest new development in medicine. But one of the risks of TAVR is stroke. Of course, we would like there to be no stroke," he said. "Now, we are doing it blindly. We don't know how [stroke] happens or when it happens. With this model being developed, we could."

James Walsh • 612-673-7428