Bucking conventional wisdom, a University of Pennsylvania study concludes that natural resistance to the heart-protective effects of aspirin is rare, and questions the value of increasingly widely used screening tests for aspirin resistance.

The study also found that coated aspirin pills, touted as being easier on the stomach, can throw off the tests, which measure how much aspirin reduces the formation of blood clots.

"The incidence of true aspirin resistance is vanishingly small, and the idea of testing for it is seriously undermined by this study," said senior author Garret A. FitzGerald, an expert on blood clotting.

Long-term, low-dose aspirin use has been proven to reduce the risk of heart attack and stroke by about a third in people who have already suffered one. The ancient drug works by reducing the stickiness of platelets, the blood cells that clump together to form clots.

If resistant, patients may need higher doses of aspirin, the anti-platelet drug Plavix, or newer platelet-inhibiting drugs.

Although aspirin's cardiovascular benefits are less clear in lower-risk people, they account for most of the estimated 50 million Americans now on the therapy.

Some people have heart attacks or strokes despite taking aspirin, so researchers have studied individual biological response to anti-platelet therapy and found that it varies. However, there is no standard definition of aspirin resistance, and estimates of its prevalence vary wildly.

"I Googled 'aspirin resistance' this morning, and the estimates vary from 5 percent to 60 percent" of aspirin users, said FitzGerald, whose lab was the first to warn of the dangers of painkillers such as Vioxx.

In recent years, commercial blood and urine tests have been developed to screen for aspirin resistance by measuring markers of platelet activity.

VerifyNow, the market leader made by San Diego-based Accumetrics, uses a blood sample that is tested right in an office. A newer test, AspirinWorks, made by Denver-based Corgenix Medical Corp., uses a urine sample that is evaluated by a commercial lab.

The Penn study, published Tuesday in the journal Circulation, did not use these two commercial tests, but used the technology behind them as well as one other. Indeed, FitzGerald said, his lab originally identified the chemical biomarker that is the basis for the urine test.

The Penn study, which tested 400 healthy volunteers, set out to find people with an innate, or genetic, resistance to aspirin. To be considered resistant, volunteers had to consistently show a weak anti-clotting response, regardless of the type of screening test, the hours between the aspirin dose and the test, the use of coated or uncoated aspirin, or the dosage.

"We failed to find a single person who satisfied these criteria," the researchers wrote.

But they did find that up to 30 percent of volunteers were "resistant" depending on these variables. For example, of 27 people who tested resistant after single doses of aspirin, only one failed to respond to week-long treatment. And even that non-responder became a responder when aspirin was added directly to a sample of his blood in a test tube, rather than being swallowed.

Brian Bartolomeo, Accumetrics' market development manager, said testing is valuable for patients at high risk of a heart attack or stroke, because for them, clotting could be deadly. He said recent studies suggest 10 percent of them are resistant.

"Obviously, this study was in healthy volunteers so it's a bit difficult to make a direct translation to patients with a stroke or cardiac event," he said.

VerifyNow is used in more than 800 U.S. hospitals, and 100 doctors' offices, the company said.

Ann Steinbarger, Corgenix's senior vice president for marketing, said the study has no bearing on the reliability of AspirinWorks, which has just gotten regulatory approval for sale in China.

The Penn team "didn't use our test so it couldn't relate to reliability of our assay," she said. "They're looking at slightly different things."

But Chicago cardiologist Robert Bonow, past president of the American Heart Association, said: "The study goes a long way to reassuring us that aspirin resistance as a fundamental factor in some people does not appear to be present at all. Or if it is, it's in a very small number."