WASHINGTON - A study testing whether Celebrex or naproxen would reduce the risk of Alzheimer's disease (news - web sites) was halted Monday after researchers noted an increase in heart attack and stroke among participants who were taking naproxen, an over-the-counter pain reliever on the market for nearly 30 years. Officials at the National Institutes of Health (news - web sites) said the study was stopped after three years when it was noticed that patients taking naproxen, sold under the brand name Aleve, had a 50 percent greater incidence of cardiovascular events — heart attack or stroke — than patients taking placebo. Another factor, officials said, was the announcement last week that advertising for Celebrex was being halted after a study found that high doses of the drug was associated with an increase in heart attack risk. Data from the Alzheimer's study, however, did not indicate an increased risk for heart attack or stroke.

Celebrex and naproxen are both pain relievers commonly used to treat arthritis. Naproxen has been approved for sale, first as a prescription and then as an over-the-counter drug, since 1976.

Efforts to obtain reaction Monday night produced no answers at phone numbers for Bayer Healthcare, the maker of Aleve, a popular version of naproxen.  Dr. Sandra Kweder of the Food and Drug Administration (news - web sites) said the NIH study is the first to show that naproxen might increase the risk of heart attack or stroke.   "We agree that this is confusing," she said. The FDA (news - web sites) will closely evaluate the findings from the NIH study, said Kweder, but that no new regulatory action with naproxen is expected within the next few days.  She said patients who routinely take naproxen should follow the drug directions carefully, including the instruction not to take it for more than 10 days, and to consult a doctor if there continues to be pain.  The Alzheimer's disease study was being conducted by the National Institute on Aging, an arm of the NIH. It called for 2,500 patients, aged 70 or older and who had a family history of Alzheimer's, to take either Celebrex, naproxen or placebo.

The group was divided and each division, or arm, was assigned to receive one of the drugs or placebo. The drugs were blinded, which means the patients did not know which medication they were taking.   The goal was to determine if the pain-relieving drugs lowered the risk of developing Alzheimer's disease. The study started three years ago and was to continue for a few more years. Officials said the patients in the study will be monitored for developing Alzheimer's, but will not be given the test drugs.  Dr. Elias A. Zerhouni, the director of the National Institutes of Health, said the study linking heart attack to Celebrex last week was a major factor in deciding to suspend the Alzheimer's study.  He said there was a question whether patients in the study would continue to take their medicine since they knew they might be taking Celebrex.

Suspending the study, Zerhouni said, "is the prudent thing to do."   John Breitner of the Veterans Affairs medical facility in Seattle and the University of Washington, an investigator in the trial, said only preliminary data is available. But he said it suggests that about 70 patients of the 2,500 suffered stroke or heart attack, including 23 deaths, and that there were about 50 percent more such events in the naproxen arm of the study than in the placebo division.