Heart risks from taking Merck & Co Inc's painkiller Vioxx could have been detected more than three years before the company withdrew the drug from the market in September 2004, had the data been openly available, U.S. researchers said on Monday.

They said their analysis illustrates the need for quick, public disclosure of drug safety data.

You could have known there were marked safety problems with Vioxx as early as 2000 and the signal only grew stronger, said Dr. Harlan Krumholz of Yale University School of Medicine, who worked on the study published in the Archives of Internal Medicine.

Merck voluntarily withdrew Vioxx from the market in September 2004 after a clinical trial found the blockbuster drug increased the risk of heart attacks and strokes in long-term users of the medicine.

In November 2007, Merck signed a $4.85 billion deal to settle thousands of claims for heart attacks, strokes and deaths allegedly caused by the drug.

The study draws on company data from more than 30 clinical trials conducted by Merck between 1996 and 2004 comparing Vioxx, or rofecoxib, to a placebo or dummy pill.

Krumholz gained access to the data through his role as a paid witness for plaintiffs in lawsuits involving heart attacks or strokes that occurred after taking the drug.

Most of the information we are using in this study was never published, or if it was published, they never included the key safety data, he said.

Of the 30 studies, 18 were published before September 2004, when the drug was withdrawn, six were published after that, and six were never published.

We're suggesting this has to change, Krumholz said.

The team pooled data from randomized clinical trials that compared Vioxx to a placebo.

They pooled the trials in the order the studies were done, and after each one, they analyzed the data to see if they could detect any signal that would suggest the drug was raising the risk of heart attacks and strokes.

They found a trend toward increased heart risks compared with placebo as early as December 2000, and a statistically significant signal by June 2001, nearly three and a half years before the company pulled the drug from the market.

Merck rejects the conclusions of the study, saying its own analysis of the data done while the drug was on the market did not show an increased risk of blood clots.

There is nothing new here. We studied Vioxx before and after it was on the market. We studied it extensively using more rigorous methods than these authors used and we didn't see any cardiovascular risk, Ron Rogers, a Merck spokesman, said in a telephone interview.

He said the first time they saw a risk was in September 2004, and the company voluntarily pulled the drug from the market within a week of those results.

He rejects the methodology used by Krumholz and colleagues, which counted heart problems reported by doctors, and not those that were confirmed by outside experts.

Krumholz said the methods used by the team represent a well-accepted approach that passed a tough peer review process.

He said the purpose of the study was not to point a finger at Merck, but to argue for better ways of tracking the safety of drugs once they are approved by the U.S. Food and Drug Administration.

SOURCE: Archives of Internal Medicine, November 23, 2009.