For researchers working on the Penn State artificial heart, one problem is hemolysis, or damage to red blood cells. Observed in laboratory simulations and in animals implanted with the heart, hemolysis can result in clotting.

What causes it?

It's hard to spot a flaw in the heart itself: a fist-sized shell of plastic containing a seamless sack made of a special biocompatible polymer. There are no rough spots, no discontinuities.

"Initially, there were two hypotheses," says Juan-Carlos Maymir, a Ph.D. candidate in bioengineering working in the lab of Penn State chemical engineering professor James Tarbell. "First, that when blood came into the heart, some of it might tend to pool at the bottom of that plastic casing. Maybe this blood would tend to clot.

"The other idea was that if blood entered the heart too fast, the shear stress -- the pressure exerted against a surface by a fluid moving across it -- might be enough to cause thrombosis."

"Tim Baldwin, a previous graduate student here, looked for sites of shearing or pooling. He didn't find any. That was encouraging for the heart, but it didn't explain the problem."

Then, Maymir says, Baldwin began looking specifically at fluid flow around the artificial heart valves. He hands over a sample valve, about the size of a quarter: It's a thin ring of titanium fitted with a round plastic occluder, a poker chip on hinges so that it flips easily open and closed. (The same valves are used in valve replacement surgery involving natural hearts.)

"Tim noticed that when the valve is closed, there's a gap between the plastic and the metal. Here, hold it up."

Sure enough, a thin ring of light shines through. When the valve is in place and closed, Maymir explains, incoming blood creates pressure across its top, forcing fluid through this tiny gap. The result is a turbulent jet, "just like what happens when you put your thumb over the nozzle of a garden hose."

Maymir measured the velocity of these jets using a technique called laser Doppler velocimetry (LDV).

"It depends on the same principle as the Doppler shift," he says, and gives a familiar example, the one about sitting at a railroad crossing when a train roars by and noticing the drop in pitch. If you can measure that drop -- the change in frequency of the sound waves -- you can determine the train's velocity.

"LDV is a similar technique involving laser light." Two laser beams focused on the valve produce an interference pattern: a cross-hatch of light and dark areas. A blood particle passing over the pattern will reflect this changing light. By measuring the frequency of reflective bursts, you can measure the velocity of flow.

The technique showed that the jets do indeed produce enough shear stress to cause hemolysis. Since no two valves have exactly the same gap, Maymir also looked at how gap size affects the stress produced. Not surprisingly, given the garden-hose analogy, he found that a larger gap produces less stress.

"Now we need to figure out how to use this information," he says. "The trend is to say the larger the gap the better, but that's only up to a certain point. A big gap also means leaking of fluid, and lower efficiency. So we're looking for the happy medium."

Meanwhile, another of Tarbell's graduate students, Debbie Sneckenberger, is investigating another valve-related problem.

As Sneckenberger explains, some prosthetic valves exhibit pitting: a pattern of tiny holes or scars in the titanium housing. The likely culprit, she says, is cavitation.

"In any fluid flow, if you have a sudden drop in pressure that's significant enough, you'll have the spontaneous formation of bubbles. When these bubbles collapse, as the pressure rises again, there's energy released." Given the right conditions, this energy can be enough to "dig away" at the surrounding materials.

Cavitation is a common problem around submarine propellers, where the spinning blade creates rising and falling pressure in the water. An artificial heart valve, too, presents a perfect site for the effect. As Sneckenberger explains, "The valve is open, blood flows in, the valve closes -- causing the blood to suddenly stop. There's a drop in pressure." Then the valve is opened again, and pressure jumps back up. "So you have rising and falling pressure with every beat of the heart."

The pitting that results creates a rough surface that may actually tear open red blood cells as they rub against it. "We have shown in vitro that if we create a cavitation situation, it is strong enough to do damage to red blood cells.

"What I'm trying to do now is find the most important design parameters for reducing cavitation."

Sneckenberger tested valves with different gap sizes, different "geometries," and occluders made of different materials in a "mock circulatory loop" that models the heart in action. She quantified the amount of cavitation produced by each valve by measuring a characteristic fluctuation in the pressure drop that occurs when the valve is closed.

What she found was that valves with occluders made of Delrin, a Dupont plastic, exhibited less cavitation than those made of pyrolitic carbon, a more rigid material. "When the carbon valve closes it might create a greater pressure drop, just because it's more rigid," she suggests. "Or it may be that Delrin actually absorbs some of the energy created."

Combined with Maymir's results, Sneckenberger's studies point to a need for a closer scrutiny of artificial heart valves.

"We suspect that hemolysis is occurring because of a problem with the valves," Maymir says, "but we don't know whether it's because of cavitation or turbulent jets. It could be either -- or both."

Juan-Carlos Maymir is a Ph.D. student in bioengineering, 23 Fenske Laboroatory, University Park, PA 16802; 814-865-3648. Deborah S. Sneckenberger will receive an M.S. in bioengineering in December 1995. In January 1996, she will enter the Ph.D. program in mechanical engineering at Georgia Tech.

Faculty advisers for the projects reported were John M. Tarbell, Ph.D., professor of chemical engineering; Steven Deutsch, Ph.D., senior scientist, Applied Research Laboratory; John F. Gardner, Ph.D., associate professor of mechanical engineering; David B. Geselowitz, Ph.D., professor of bioengineering; Gerson Rosenberg, Ph.D., professor of bioengineering at the Hershey Medical Center; and David R. Stinebring, Ph.D., associate research engineer at the Applied Research Laboratory.

Funding for the projects is from the National Heart, Lung and Blood Institute of the National Institutes of Health.