BioEd Online

A study of the arthritis drug Vioxx has shown that it may have caused up to 140,000 cases of heart disease in the United States before it was withdrawn from sale. A separate study finds that many of the victims may have received the drug unnecessarily.

The estimate comes from David Graham, a researcher at the US Food and Drug Administration (FDA) in Rockville, Maryland, and his colleagues. The painkiller rofecoxib, sold under the name Vioxx by New Jersey-based manufacturer Merck, was pulled from the market in September 2004. The decision was made after a study of Vioxx's effect on colorectal cancer revealed that it increased patients' risk of heart attacks.

In their study, published in The Lancet¹, Graham and his team examined the medical records of 1.4 million Californians who had used anti-inflammatory drugs. Users of Vioxx, were between 1.6 and 3.6 times more likely to experience heart trouble than users of a rival drug, celecoxib, which Pfizer manufactures under the name Celebrex.

Graham's group estimated the number of incidents likely to have occurred between the drug's debut in 1999 and its withdrawal from the market. By combining data from Merck's clinical trials with data on how many patients took Vioxx, the group concluded that the drug had caused between 88,000 and 140,000 cases of coronary heart disease or heart attacks.

Many of these complications could have been avoided if FDA regulators had done more when problems with the drug surfaced in 2000, says Graham. At the time, they recommended simply that a warning be added about the drug's possible complications. "This new study shows the societal cost of not taking appropriate regulatory action," Graham says.

Newer not better

In a related study, a group led by Randall Stafford of Stanford University's Prevention Research Centre in Palo Alto, California, found that many patients taking Vioxx did not even need the drug.

Vioxx and other drugs in its class were initially developed as alternatives to pain relievers that caused internal bleeding in some patients. But by studying two federal surveys, Stafford and his colleagues found that almost two-thirds of new prescriptions written between 1999 and 2002 went to patients who were not at risk of internal bleeding. By 2002 only 39% of patients receiving the class of drugs that includes Vioxx actually needed that type, rather than earlier types of painkiller.

"From the perspective of both the patient and the doctor I think there is a perception that newer drugs are better," Stafford says. Aggressive marketing by drug companies, combined with faith in more recent discoveries, meant that far more people were taking the drug than actually needed it, he says. His research is published in the Archives of Internal Medicine².

Both Stafford and Graham agree that the FDA needs to be more vigilant about watching drugs after they have been approved and controlling the advertising blitz that often occurs after they go to market. "It is time that we learn from these lessons," Stafford says.