The editorial is the latest salvo in a growing debate among doctors about the risks of fatal blood clotting and serious heart attacks associated with the latest generation of stents, which are drug-coated. The devices are sold by Boston Scientific and Johnson & Johnson. The drug coating is meant to reduce inflammation at the site of the stent, in hopes of preventing a recurrence of the arterial blockage that led to the insertion of the device.

The article, published as a guest editorial on a Web site of the American College of Cardiology, said patients faced a lower risk if treated with older, bare-metal stents that might work just as well in many cases.

Stents have become the preferred therapy for millions of Americans a year. And all stents carry some risk of clotting, particularly in the early days and weeks after insertion. The new focus of concern has involved the drug-coated stents and reports of clotting many months or even years after the devices are inserted.

The editorial was written by Drs. Sanjay Kaul and George A. Diamond of Cedars-Sinai Medical Center in Los Angeles.

Dr. Kaul said in an interview that research suggested the use of the drug-coated stents increased the risk of getting a clot by 0.6 percent annually compared with patients who received the bare metal stents.

With more than one million Americans annually receiving the stents and at least 80 percent of them getting the drug-coated versions, that worked out to at least an additional 2,160 deaths each year, the doctors calculated. The figure assumed that 45 percent of the people who suffered such clots had died.

“It’s eerily reminiscent of Vioxx,” said Dr. Kaul, referring to the popular painkiller that Merck withdrew from the market in 2004 after it was linked to heart attacks. “Initially we didn’t see the signal because the incidence is so rare.”

Dr. Kaul said that drug-coated stents were proper treatment for many patients but that they were being used far too frequently in cases where clinical data suggests that bare metal stents or long-term drug therapy would be safer. Only about 20 percent of the drug-coated devices are being inserted in patients who have the kinds of conditions studied in the clinical trials that led federal regulators to approve them, he said.

Advocates for drug-coated stents have suggested that the higher comparable risks reflect the increase of their use in diabetics and with other difficult cases. They also note that the clotting risks appear to be largely avoidable if patients continue to take aspirin and Plavix, an anticlotting drug, as is routinely prescribed for three to six months after stenting.

But concerns have grown since the beginning of September, when new reports at a scientific meeting in Barcelona, Spain, spotlighted the clotting problem. On Sept. 7, Boston Scientific disclosed that it had told the Food and Drug Administration this summer that the company’s most recent analysis of its accumulating data on its Taxus drug-coated stent had concluded that clotting risks were slightly higher than for bare metal stents.

Johnson & Johnson has said that data for its Cypher stent showed no such problem, although independent research has suggested otherwise. The food and drug agency plans to hold a two-day meeting of experts in early December to review the data and advise the agency on whether changes in labeling or use are needed.