Research is raising doubts about a popular form of stroke therapy that aims to snatch a blood clot from a patient's brain and restore blood flow before serious damage is done.

In clinical trials presented this week at the American Heart Association's International Stroke Conference, researchers said that mechanically removing a blood clot from a stroke patient's brain was no more useful than the older treatment of giving a clot-dissolving medicine called tissue plasminogen activator, or tPA.

The devices, first approved by the Food and Drug Administration in 2005, have taken the stroke community by storm and are widely used in stroke centers to clear clots that have cut off blood to the brain. But the trial results suggest that, for the earliest generation of the devices at least, the benefits of retrieving clots through blood vessels have been overstated.

"For the stroke field this is a really big deal," said Walter Koroshetz, deputy director of the National Institute of Neurological Disorders and Stroke.

The studies, which also were published online by the New England Journal of Medicine, touched off a flurry of debate. The device-driven therapy has become a major source of revenue and prestige for hospitals that offer it. Stroke care that uses a retrieval device typically costs close to $25,000 -- about twice as much as the use of tPA alone.

Stroke is the fourth-leading cause of death in the United States. About 800,000 people suffer a stroke each year, and about 130,000 die. Nearly 90 percent of strokes occur when a clot blocks an artery in the brain, starving the region downstream of blood and oxygen.

The largest of the clinical trials compared outcomes in stroke patients who were randomly assigned to receive either the medicine alone or the medicine along with the endovascular therapy. That trial, conducted in 58 hospitals in the United States, Canada, Australia and Europe, was stopped ahead of schedule after safety watchdogs concluded the invasive devices offered no benefit but did present small risks.

The 434 patients who were treated with endovascular therapy were just as likely to die and to be disabled three months after their strokes -- and were no less likely to suffer bleeding in the brain -- compared with the patients who got the medicine alone. Patients treated with the devices were more likely to experience bleeding in their brains. Further, many practitioners believe that newer clot-retrieving devices work better than the ones used in the trials.

The findings represent "one of the most significant announcements in stroke in the past five years," said Dr. Patrick Lyden, chairman of neurology at Cedars-Sinai Medical Center in Los Angeles. "We need to take a very careful look at who's going to get these procedures."

The Washington Post contributed to this report.