Doctors report an increase in potentially deadly blood clots in patients' arteries that have been implanted with drug-coated stents.

Stents have revolutionized cardiac care: The tiny wire-mesh tubes, which prop open arteries that have been cleared of blockages caused by fatty deposits, save many patients from the risk and trauma of open-heart bypass surgery. The newest kind, coated with drugs, prevents the growth of scar tissue inside the stent.

Signs of increased clotting could reduce cardiologists' and patients' enthusiasm for drug-coated stents, which are by far the most popular kind. Since hitting the market in 2003, drug-coated stents have been implanted in three million people world-wide.

The risk of blood clots with both bare-metal and drug-coated stents has been known for some time. Until recently, it was considered a short-term problem, controllable with prescription anticlotting drugs for as long as six months.

This week, the principal investigators for clinical trials of two drug-coated stents -- Johnson & Johnson's Cypher stent and Boston Scientific Corp.'s Taxus stent -- presented new evidence of longer-term blood clots before a gathering of cardiologists at the Transcatheter Cardiovascular Therapeutics meeting in Washington.

Separately, the investigators presented data from four trials of each stent, involving a total of about 5,100 patients in eight trials. The researchers said the drug-coated stents seemed to cause clots at the same rate as earlier-generation bare-metal stents -- about 0.7 percent -- for the first 18 months after implantation. For the period from 18 months to three years, drug-coated stents add a "small but real increase" in clots, amounting to an additional one in 200 patients, or 0.5 percent, said Gregg Stone, of Columbia University Medical Center in New York, who is Boston Scientific's lead investigator. "This is a problem that we have to deal with," Dr. Stone said.

The evidence isn't statistically significant, Dr. Stone said, because the total number of patients in the trials was relatively low. But in a follow-up email, he wrote that if the eight studies "were to be combined into one dataset, which has not been done, the difference may indeed become statistically significant."

J&J and Boston Scientific say their drug-coated stents are safe and the clot rates for both types are essentially the same. "We have looked at a lot of data and we have seen no significant difference between our (drug-coated) stent and our bare-metal stent," said Paul Donovan, a spokesman for Boston Scientific. "The therapeutic benefit and the safety profile of (drug-coated) stents are maintained over time." Several doctors at the meeting also said they weren't concerned yet.

But some doctors say they are troubled by clots they are seeing in patients as many as two years after receiving drug-coated stents. At the Cleveland Clinic, cardiology chief Eric Topol says many patients have come in 15 months, 18 months or 24 months after the stent procedure with a clot. "It never happened in the bare-metal era," he said.

Dr. Topol and others say they now tell patients they must take anticlotting drugs after stent operations for a much longer period of time than previously was deemed necessary -- sometimes indefinitely.

They are doing so with apprehension. Some patients are resistant to the drugs. Others told to take them don't comply, because the drugs -- usually aspirin or a newer drug, Plavix -- can cause serious side effects, such as stomach bleeding and rashes. Patients often have to stop taking anticlotting drugs for dental work, minor surgery or after an accident or stroke, to prevent hemorrhaging. When they stop taking the drugs, the risk of blood clots returns.

An added risk of one in 200 patients may not sound like much. According to cardiologist Robert Schwartz, of the Minneapolis Heart Institute and Foundation, the risk "must be countered with the large numbers of stents being implanted." With millions of patients getting the devices, he says, thousands could be running the risk of a lethal blood clot. "We can't ignore the problem any longer," Dr. Schwartz says. "We probably need Plavix forever" in a lot of patients.

Labeling for J&J's Cypher stent says anticlotting drugs should be used for three months after insertion; guidance on Boston Scientific's Taxus stent is six months. Dr. Stone said "there is not enough data" to change these recommendations.

"No one knows when it is safe to come off" anticlotting drugs, says cardiologist Aloke Finn, of Massachusetts General Hospital in Boston. Dr. Finn says recipients of drug-coated stents at his hospital now are kept on the drugs indefinitely. "Every cardiologist at this hospital is going to tell you 'I'm concerned,' " he added.

Herman Gold, another Massachusetts General cardiologist, said he has gone back to using bare-metal stents in about 40 percent of his patients, because he wants to spare as many as he can the risk "for a late event for which there is no good treatment. It's unpredictable and it's deadly."

Ramon Quesada, of the Baptist Cardiac & Vascular Institute in Miami, prefers not to implant a drug-coated stent if he knows the patient can't afford anticlotting drugs or might not take the drugs consistently. "You're doing a disservice" by implanting a drug-coated stent in those patients, he said.

In a recent article published in the Journal of the American Medical Association, Italian researchers studied 2,229 patients implanted with drug-coated stents and found that 29, or 1.3 percent, had suffered a clot in the stent by nine months, with seven of them getting the clot later than three months after implantation. Of the patients who had clots, 45 percent died. The researchers also found that 29 percent of the patients with a clot had stopped taking their anticlotting medications, leading the researchers to conclude that stopping the drugs was the No. 1 factor linked to the serious side effect.

If drug-coated stents do put some patients at a higher risk of late clots, the question is why. Renu Virmani, a pathologist at CV Path, a Gaithersburg, Md., company that provides cardiac tissue analysis for clinical trials, said she believes it may be because the tissue in the artery isn't healing properly after implantation.

Drug-coated stents are supposed to work by preventing aggressive scar-tissue growth, called restenosis, which can block the stent, while at the same time allowing a thin layer of tissue to heal over the stent in the artery. Healing protects the area from forming blood clots.

At this week's conference, Dr. Virmani, using samples from her large collection of stents from human cadavers, presented an analysis from 40 patients with drug-coated stents who had died from cardiac causes. Of those, 24 had died of a blood clot in their stent. Fourteen had suffered from a late clot -- which Dr. Virmani defines as a clot happening beyond a 30-day postimplantation period. Half of the patients had the clot six months after implantation.

All patients had more than one stent implanted -- including bare-metal and drug-coated stents -- so Dr. Virmani was able to see how the same patient's tissue reacted to the two different types. In those who died of late clots, all the blood clots were found in the drug-coated stents, with none in the bare-metal stents, she said.

In the drug-coated stents that had clots, Dr. Virmani found that only about 38 percent of their surface had been healed over by tissue. In contrast, the drug-coated stents that didn't have blood clots had 75 percent of their surface covered. Drs. Finn and Gold from Massachusetts General Hospital worked with Dr. Virmani on the analysis. Dr. Finn is Dr. Virmani's son.

Dr. Virmani's data must be taken with caution, said Brian Firth, vice president of medical affairs at the Johnson & Johnson unit that makes the Cypher. "You have a couple of cases of something" and "no denominator," meaning no way of measuring rates of incidence, he said.

Others say concern over late clots with drug-coated stents has been blown out of proportion. Pedro Moreno, a cardiologist at Mt. Sinai Medical Center in New York, said during a presentation that one explanation for the increase in clots seen in stents is that longer portions of arteries are being treated. A higher frequency of late clots could come from a more-aggressive practice with stents in general, he added, rather than from something drug-coated stents inherently do to arteries.

Growing Concern

Recent studies may indicate more risk of blood clots from drug-coated stents.

Increased risk of blood clots in drug-coated stents, compared to metal stents, after 18 mos.: 0.5 percent

Mortality rate from blood clot in stent: up to 45 percent

Patients with drug-coated stents world-wide: 3 million

Main drug-coated stent brands: Cypher (J&J) and Taxus (Boston Scientific)

Sources: Journal of the American Medical Association, May 2005; Martin Leon, Columbia University Medical Center