An independent advisory panel to the U.S. Food and Drug Administration (FDA) decided this week there is not yet enough evidence to conclusively show that the painkiller naproxen has a lower risk of cardiovascular side effects than other nonsteroidal anti-inflammatory drugs (NSAIDs), such ibuprofen (Advil and Motrin) and celecoxib (Celebrex). All of these are used to treat arthritis as well as many other conditions.

The Arthritis Advisory Committee was asked to consider whether the warning about cardiovascular side effects should be removed from the label on naproxen – sold over-the-counter (OTC) as Aleve and by prescription as Naprosyn.

“There is certainly a lot of evidence that naproxen is safer than other [NSAIDs] in terms of causing heart attack and stroke, but being safer doesn’t mean it is perfectly safe. I think that is the big issue,” says committee member Donald Miller, chair of the pharmacy practice department at North Dakota State University in Fargo. .

The issue of NSAIDs’ safety profile resurfaced after a meta-analysis of more than 600 randomized trials was published in The Lancet last year. It found naproxen had a lower risk of cardiovascular side effects, including heart attack and stroke, than other OTC and prescription NSAIDs.

But after two days of meetings Feb. 10 and 11, the committee voted 16 to 9 to leave the labeling as is. It also voted to let a new study on the subject, currently underway at Cleveland Clinic, continue. The PRECISION (Prospective Randomized Evaluation of Celecoxib Integrated Safety vs. Ibuprofen or Naproxen) trial is directly comparing the risks of the medications on more than 20,000 patients at high risk of cardiovascular problems. The Lancet analysis, by contrast, was based on indirect comparisons.

The lead investigator of the PRECISION study, Steven Nissen, MD, chair of Cleveland Clinic’s department of cardiovascular medicine, testified before the FDA advisory panel. He says when his study is completed by the end of next year, the results will be based on stronger evidence than is currently available, including the studies reviewed for the meta-analysis.

“Trying to make decisions on the safety of NSAIDs based on these small observational studies and other evidence that is very weak has been the problem,” says Dr. Nissen. “It’s hypothesis-generating, but doesn’t prove anything.”

Dr. Nissen says his study, underway since 2006, needs to be completed for clarification on this issue. “If you make a regulatory decision based on incomplete evidence, you may make the wrong decision,” he says. “You want to get the definitive study done.”