The U.S. Food and Drug Administration recommended limited use of the painkilling drugs Celebrex and Bextra yesterday, urging patients who are at high risk for heart attacks and strokes to avoid the drugs if possible.
In its public health advisory, which followed a spate of conflicting studies about the potential risks of pain relievers, the agency did not rule out use of COX-2 inhibitors, the class to which the two drugs belong.
Patients who are prone to gastric bleeding, a well-known side effect of such older anti-inflammatory drugs as aspirin, naproxen and ibuprofen, might be candidates for the COX-2 drugs, the FDA said.
The agency also advised caution on use of the older drugs. Patients should use them "in strict accordance with the label directions," not exceeding recommended doses or taking them for longer than 10 days without consulting a physician.
"The actions that we're recommending in the advisory are for more limited prescribing than would currently be included in the approved labeling for these products," John Jenkins, director of the FDA's office of new drugs, said in a news briefing.
The FDA said doctors should carefully consider a patient's risk of heart disease before prescribing Bextra or Celebrex. With some studies indicating the drugs could raise a person's risk, the agency suggested that patients who are already susceptible to heart attacks and strokes might not be good candidates for them.
In a rapid-fire sequence that has confused doctors and patients alike, new studies have called into question the safety of drugs that millions of people depend on to function.
First, Merck & Co. withdrew Vioxx, the best-selling COX-2 inhibitor, after a study linked it to heart attacks. Next, other studies suggested similar problems with two related drugs, Bextra and Celebrex, made by Pfizer.
The pharmaceutical giant agreed last week to stop advertising Celebrex to consumers but did not withdraw the product. Finally, on Tuesday, the National Institutes of Health announced that it had stopped giving Celebrex and naproxen to volunteers in a three-year Alzheimer's prevention trial.
Volunteers on naproxen were showing a 50 percent higher rate of heart attacks and stroke than those taking Celebrex or a placebo. The study did not find that Celebrex posed a cardiovascular risk, but researchers said they couldn't expect volunteers to take the drug in light of the other studies' results.
Many doctors greeted the news about naproxen with skepticism, noting that other studies found that it actually protected the heart. Some also said concern over Celebrex might be overblown because two studies found no link to heart disease. Also, they noted, the use of low-dose aspirin might reduce any small risk that does exist.
The FDA is reviewing data on all the drugs to determine whether additional regulatory action is needed, the agency said. An advisory committee is to meet in February to review the evidence and suggest action.
News about painkillers and their potential risks is likely to affect arthritis patients more than people with occasional headaches and joint pain: Many take daily doses for years, perhaps for the rest of their lives.
Dr. Joan Bathon, a rheumatologist who heads the arthritis program at the Johns Hopkins School of Medicine, said she was composing a consumer advisory for a Hopkins Web site when she heard the FDA message: "I was basically saying the same thing. For each individual patient, we really have to weigh pros and cons. We can't eliminate an entire class or a drug based on the evidence right now."
Increasingly, doctors are prescribing drugs in tandem, one to treat the condition and another to counteract the first drug's possible side effects. For instance, patients taking naproxen might take a separate drug - such as Prilesec or Prevacid - to prevent gastric bleeding. Those taking a COX-2 inhibitor might take a low-dose aspirin to mitigate the risk of heart attack.
In some cases, she said, doctors might have to prescribe three-drug regimens. Physicians have been doing this to some extent anyway, but recent events have made the calculations more complicated than ever.
Dr. Marc C. Hochberg, head of rheumatology at the University of Maryland School of Medicine, said the FDA is taking a sensible course at a time when safety issues remain unclear.
"I think they're following their marching orders to recommend a safe, prudent approach to physicians," Hochberg said. Hochberg intends to continue prescribing Celebrex - he uses Bextra much less - but only after weighing the risks and benefits for each individual, which he says should be the standard practice anyway.
"I evaluate their individual risk for cardiovascular disease, GI complications and other side effects, as well as their need to be on their medications. And then we make a decision."
Mark Aronson, a general internist at Beth Israel Deaconess Medical Center in Boston, said the FDA issued "a very wise set of recommendations."
In a recent editorial in Annals of Internal Medicine, Aronson had recommended that patients avoid COX-2 inhibitors until the safety questions could be answered. He is now telling patients to use all non-steroidal anti-inflammatories carefully.
"My recommendation would be to use these drugs for a short time, if at all," he said.
FDA safety officer David Graham, a frequent critic of the agency, said the recommendations did not go far enough. He said many patients use pain relievers chronically, and the agency's statement would probably do little to change that.
"To assume that people will follow this is wishful thinking," he said.
Graham noted, for example, that in 2002, the FDA changed Vioxx's label, warning patients that high doses might increase heart attack risk. But people continued to take just as much as before, he said.
Graham suggested that the agency now take stronger action to limit use of COX-2 inhibitors.
"The FDA is saying they can't be sure these drugs are safe," he said. "Well, if they're concerned, perhaps people shouldn't be using [the drugs] at all."
Erika Niedowski and the Associated Press also contributed to this article.