Researchers looking at the safety of various arthritis and headache remedies said on Thursday they had found some more evidence that the popular analgesic naproxen may cause heart problems.

The researchers themselves cautioned about reading too much into their findings, and outside experts discounted them.

Nonetheless, the researchers said they thought it was important to present their evidence to help experts sort out what the real dangers are.

The issue is important because companies have been forced to withdraw several drugs in a new class called COX-2 inhibitors because they may cause strokes and heart attacks, and drug makers face multi-million dollar liability suits.

And the U.S. Food and Drug Administration has ordered strict warnings to be put on packages of painkillers known as non-steroidal anti-inflammatory drugs or NSAIDS, including naproxen, a popular NSAID sold over-the-counter by Bayer under the brand name Aleve.

The warning labels were ordered because of some studies that suggested NSAIDS might also raise heart risks.

“Particularly for safety data, ‘truth’ may come in small doses. We firmly believe that results from trials should be published regardless of the direction, magnitude, or statistical significance of the observed results,” said Barbara Martin of the John Hopkins University School of Public Health, who worked on the study.

Writing in the Public Library of Science journal PLoS Clinical Trials, Martin and colleagues said they looked at data from a trial designed to see if Celebrex might help prevent Alzheimer’s disease in high-risk patients.

The 2,500 elderly volunteers were given either COX-2 inhibitor Celebrex, naproxen or a placebo for up to 3-1/2 years.

It was stopped when concerns rose over the safety of COX-2 inhibitors.

The trial showed no increased or decreased heart risk for Celebrex, known generically as celecoxib, compared to placebo. But it did show an increased risk for naproxen.

Over three years, 5.5 percent of the patients getting Celebrex had a stroke, heart attack or were diagnosed with congestive heart failure, while 5.6 percent of those getting a placebo did. In comparison, 8.25 percent of the naproxen patients had such a serious cardiovascular incident or death.

Dr. Steve Nissen of the Cleveland Clinic in Ohio said the trial’s findings were “completely unreliable.”

“The stopping of the study early was improper and it led to erroneous conclusions,” Nissen said in a telephone interview. “We know from very large epidemiological (population) studies and all sorts of studies that naproxen is a safe drug, perhaps the best of the NSAIDS,” added Nissen, an outspoken expert on the safety of NSAIDS and COX-2 inhibitors.

Just last week, a much smaller study led by Dr. Michael Schiff of the Denver Arthritis Clinic suggested that naproxen did not raise the risk of heart attack and stroke.

“We don’t have the full answers,” Nissen said.

He has a safety trial with 21,000 patients underway to compare naproxen with Celebrex and placebo.