Smokers in Norway who took such drugs for at least six months had twice the risk of dying of a heart attack, stroke or other heart-related problem.
The findings came from a study of whether these pain relievers could prevent oral cancer. The drugs did, in fact, cut the risk of developing oral cancer in half, but the deaths that were prevented were offset by the increased risk of dying from cardiovascular disease, according to the study, reported Monday at an American Association for Cancer Research conference in Anaheim.
The study was relatively small — 908 people — and involved people prone to heart problems and cancer because they smoked. But specialists said it supports the Food and Drug Administration's recent decision to warn about long-term use of all such painkillers except aspirin.
The findings add to the suspicion that the heart risk extends beyond the so-called cox-2 drugs — Bextra, Vioxx and Celebrex — to the larger family of medications known as non-steroidal anti-inflammatory drugs, or NSAIDs, which include naproxen, ibuprofen and virtually all other over-the-counter pain relievers except acetaminophen or Tylenol.
In recent months, many doctors have switched patients to these over-the-counter medications, believing them to be safer alternatives.
"To the best of our knowledge, these are the first data to support putting a box warning on NSAIDs, not just cox-2s," said Dr. Andrew Dannenberg, a Cornell University scientist who helped do the Norway study.
Vioxx was pulled from the market last fall when a different study, testing whether it could prevent colon cancer, showed a higher heart disease risk. The FDA recently persuaded Pfizer to stop selling Bextra for safety reasons, but Pfizer's Celebrex was allowed to remain. Some researchers from the new study in Norway have been Pfizer consultants.
Some results from the Norway study were presented in February hearings at the FDA, but consisted of only one summary slide and "nobody really picked up on it," said Dr. Raymond DuBois, a Vanderbilt University cancer expert who had no role in the work.
The fresh information presented on Monday should persuade others to study these over-the-counter drugs in more depth, DuBois said.
Oral cancer is a deadly and disfiguring disease that is an especially big problem in Asia. Worldwide, about 275,000 cases and 127,000 deaths occur each year. In the United States, about 29,370 new cases and 7,320 deaths are expected this year.
Dr. Jon Sudbo of the Norwegian Radium Hospital in Oslo, who led the research, and others wanted to see if NSAIDs could prevent oral cancer, since studies showed they could help ward off other types of cancer. They chose more than 3,000 people at high risk for oral cancer because of their smoking habits from the Norwegian Cancer Registry, a database of more than 123,000 people.
Over 20 years, 454 of them developed oral cancer, and they were compared with 454 similar people who did not develop the disease. Of these 908 people, 277 had used NSAIDs daily for at least six months. They had a two-thirds lower risk of developing oral cancer, but scientists were puzzled because deaths were not lower.
A deeper look revealed that the NSAID users were dying at twice the rate of the others from heart-related problems. There were 42 cardiovascular deaths among the 263 painkiller users and 41 deaths among the much larger group of 562 people who had never taken such drugs.
Risk was highest among ibuprofen users, who were nearly three times more likely to die of cardiovascular disease than non-NSAID users. Aspirin was the only NSAID that did not seem to raise the risk, but the numbers of aspirin users in the study were small.
Asked what advice he would give smokers about using such painkillers, Sudbo replied: "If you think you need them use them, but you have to be careful."
Doctors said the NSAID dose made no difference in risk, but others said this should be studied in future experiments.
"There are important uncertainties. One of them is the doses and one is what the effect would be in the larger population," said Dr. Michael Thun, chief of epidemiology for the American Cancer Society, who had no role in the study.
Thun said he would propose a larger study on over-the-counter NSAIDs using the cancer society's huge database, which contains information on more than 1.2 million Americans.
In other news at the conference, a large study suggested that cholesterol-lowering statin drugs might help prevent the most serious types of prostate cancer — those that kill or spread throughout the body.
Men who took statins had half the risk of advanced prostate cancer as men who did not take such drugs, reported Elizabeth Platz of Johns Hopkins University, who did the study with Harvard University researchers and the National Cancer Institute.
It involved 34,438 men in the Health Professionals Follow-up Study, which has been going on at Harvard since 1986. None of the men had prostate cancer at the start of this particular study in 1990, but 2,074 developed it over the next decade. Of these, 283 were advanced, including 206 that were fatal or had spread widely in the body.
While statins cut the chances of this, they made no difference in the risk of developing cancer that remains confined to the prostate. Although several large other studies have concluded statins may cut the risk of cancer, specialists say there is still not enough evidence yet to recommend them for this purpose.