From left to right are: Dr. Jacky Hayles, longtime co-worker, daughter Sarah Nurse, Dr. James Maller, key collaborator, daughter Emily Nurse, Sir Paul Nurse, and former postdoctoral fellows Dr. Chris Norbury, Dr. Kathy Gould, Dr. Pierre Thuriaux, and Dr. Sergio Moreno.
Naproxen not helpful in preventing heart disease: study
A large retrospective study led by Wayne A. Ray, Ph.D., professor of Preventive Medicine and director of the Division of Pharmacoepidemiology, indicates that non-aspirin, non-steroidal anti-inflammatory drugs do not have a protective effect on the risk of coronary heart disease.
Results of the study, published in a recent issue of the journal The Lancet, suggest that the drugs naproxen, ibuprofen, and other non-selective NANSAIDs should not be used in lieu of aspirin for cardioprevention, at least not until randomized clinical trials provide support for such a practice.
Though there is reason to believe that such drugs might be candidates for cardioprotection, there are complicating factors involved. According to Ray, it has been known for some time that NANSAIDs have effects that might have “cardiovascular consequences.” These drugs affect heart function in four ways.
First, some of the non-aspirin drugs can act like aspirin in that they inhibit activation of platelets, which are small, sticky blood cells important to the clotting process that sometimes go astray and gum up arteries. Inhibition of such errant clotting could prevent heart attacks in the same way aspirin does.
Second, NANSAIDs reduce inflammation, and there is a growing school of thought that coronary artery disease might have an inflammatory component. Put simply, if these drugs reduce inflammation and inflammation causes heart attack, then maybe NANSAIDs can prevent heart attack.
Those are the two good things that they do. Now for the other side of the coin.
At high doses, NANSAIDs inhibit the body’s synthesis of prostacyclin, which is a naturally occurring platelet inhibitor. This effect could actually promote heart attacks. And, NANSAIDs have side effects such as hypertension, which also could promote heart attacks.
“So, from an overview of 50,000 feet,” Ray said, “there are four NANSAID effects, two possibly good and two possibly bad. And people didn’t really know how all these effects would play out clinically.”
The potential for physiological conflict initially drew Ray and his colleagues to this area of study, then results from a large clinical trial comparing the new cyclooxygenase-2 (COX-2)-selective drug, rofecoxib (VIOXX), with naproxen further piqued their interest. In this trial, which was designed to test the gastrointestinal safety of rofecoxib, the scientists unexpectedly found a four- to five-fold greater rate of heart attack in the group taking rofecoxib.
The trialists interpreted this finding as a protective effect of naproxen, though because of the study design, there was no way to rule out a harmful effect by rofecoxib as the cause. The explanation became widely circulated, Ray said, to the point that there were articles in the lay press suggesting that people could substitute naproxen for their daily aspirin taken to reduce the chance of heart attack.
Motivated to determine the overall effect of NANSAIDs, Ray and his collaborators designed a cohort study using information from the TennCare/Medicaid database that the department of Preventive Medicine has worked with for over 30 years. The database allowed the investigators to quickly identify groups of drug takers from records of prescriptions filled, and to get information on clinical outcomes from records of hospitalizations, emergency room visits, and death certificates.
The researchers identified two groups of individuals, one taking NANSAIDs and one not taking the drugs. They evaluated the health records of the approximately 180,000 patients in each group, which had been collected over a period of 11 years.
In a cohort study, those taking a drug are doing so because a doctor told them they needed it; those not taking the drug are doing so for a variety of reasons. So there may be important differences between the users and the non-users. This type of study is considered a good source of information on clinical effects of various exposures, Ray said.
The two cohorts were very comparable. They had a mean age of 64 years, 70 percent were female, and they had substantial levels of existing cardiovascular disease. About 22 percent had serious heart disease in the past, such as heart attack, stroke, or heart failure. About two-thirds were taking some kind of medication to treat cardiovascular problems, such as hypertension or angina.
When the researchers looked at clinical outcome, which was either acute heart attack or death due to coronary heart disease, Ray said, they found a total of 6362 events in the cohorts.
“And when we looked at these events by NANSAID use,” he said, “we found absolutely—at the level of our statistical power—no association between NANSAID use and outcome.”
So in the overall analysis, the study revealed no effect, either harmful or beneficial of all NANSAIDs. And there was no evidence of any specific protective effects of naproxen.
“We concluded that naproxen was unlikely to be a highly effective drug for preventing coronary heart disease,” Ray said. “Certainly not as protective as aspirin. And it certainly should not be used for this purpose unless there is further research demonstrating such efficacy.”
Funding for the study came from the FDA and from the Centers for Education and Research in Therapeutics (CERTs), which is a new set of organizations created by the Agency for Healthcare Research and Quality (AHRQ), an agency of the Department of Health and Human Services. CERTs was established specifically to perform research in the public interest, funding studies focused on patient outcomes associated with pharmaceutical therapeutics. Vanderbilt was one of the initial four centers funded by CERTs in 1999.
Co-authors on the publication include: Dr. Marie R. Griffin, professor of Preventive Medicine and Medicine; C. Michael Stein, M.B., associate professor of Pharmacology and Medicine; James R. Daugherty, M.S., analyst/programmer in Preventive Medicine; and Kathleen Hall, B.S., computer systems analyst in Preventive Medicine.