U.S. drugs giant Merck & Co Inc. (MRK) should have pulled its Vioxx painkiller from the market four years ago because data showing it raised the risk of heart attacks has existed since 2000, Swiss scientists said on Friday.

In a report for British medical journal The Lancet (search), researchers at the University of Berne said there was substantial evidence of the dangerous side effects of the drug by the end of 2000, but the mounting data was not analyzed properly.

"Our findings indicate that rofecoxib (Vioxx) should have been withdrawn several years earlier," the scientists said.

Merck did not recall Vioxx, a COX-2 inhibitor taken by about 20 million Americans, from the worldwide market until five weeks ago.

Merck said in a statement it "was vigilant in monitoring and disclosing the cardiovascular safety of Vioxx (search) and we absolutely disagree with any implication to the contrary."

The company said it also disagrees with the conclusion that the drug should have been withdrawn several years ago.

"The company could and should have made the statement several years back, when the data we analyzed were readily available," Matthias Egger, a professor at the university's department for Social and Preventative Medicine, told a news conference.

Merck said the data contained in the Swiss analysis are not new and are essentially consistent with combined analyzes of clinical trials that the company had previously published.

The Swiss research follows newspaper reports earlier this week that Merck had tried to fight mounting concerns about the drug's safety in order to protect sales.

The U.S. Food and Drug Administration (FDA) recently published a study estimating that Vioxx could have caused about 28,000 heart attacks or deaths since it was approved in 1999.

Merck's shares have slumped around 40 percent since the recall, and analysts estimate that it could face a bill of between $10 billion and $15 billion in litigation.

INDEPENDENT EVALUATION

The Swiss scientists performed a meta-analysis on the data, taking in results from past studies, some of which were available on the FDA Web site. They found that patients who took Vioxx were at greater risk even after a few months, regardless of how much of the drug was taken.

Using 18 randomized controlled trials and 11 observational studies, researchers saw that heart-attack risk more than doubled when Vioxx was taken.

By the end of 2000, 52 heart attacks had occurred in 20,742 patients, the researchers said. Of these, 41 were patients using Vioxx. The increased heart-attack risks became apparent in studies that were evaluated by external watchdogs.

"It could be that without independent evaluation of the data, the assessment of adverse effects is biased so that the risks of a drug appear smaller," Egger said.

"We therefore recommend that all studies be carried out with independent external data evaluation."

Drug licensing authorities should review their procedures, the study concluded, to ensure that data released after the drug launch is analyzed for signs of side effects.

The authors also said an independent panel of experts should investigate why manufacturers and licensing authorities did not evaluate the data available on Vioxx sooner.

In an editorial in the journal, editor Richard Horton criticized both Merck and the FDA, saying they acted out of "ruthless, short-sighted, and irresponsible self interest."

"The licensing of Vioxx and its continued use in the face of unambiguous evidence of harm have been public-health catastrophes," said Horton.

"This controversy will not end with the drug's withdrawal."

Shares of Merck closed down 3 percent at $27.02 on the New York Stock Exchange on Thursday.

The study was part of a wider investigation into anti-inflammatory drugs and their side effects conducted by the University of Berne on behalf of the Swiss National Science Foundation. Full results are due mid-2006.