TORONTO - A mere two weeks before a critical hearing on the future of a controversial diabetes medication, new studies have been published that add to mounting evidence that the drug, Avandia, poses what critics describe as an unnecessary risk to people who take it.

The studies, published electronically on Monday, echo what others before them have suggested: that people taking Avandia (its generic name is rosiglitazone) appear to be at higher risk of heart failure, stroke and probably heart attack than people taking a sister drug, Actos (pioglitazone).

Dr. David Juurlink, a Toronto-based drug safety researcher, said the totality of the evidence points to a need to get Avandia off the market.

"No one will ever miss this drug," said Juurlink, who was not an author of either of these studies, but who wrote an editorial accompanying them.

"The only people who will miss this drug are the people who have a financial interest in this drug. And those people have had far too loud a voice in this matter for a long time. And there are a lot of patients who are still taking this drug because the entire discussion has been slanted by those people."

Reacting to findings of the new studies, Avandia's maker said there are strengths and limitations in each type of analysis, and half a dozen randomized clinical control trials have suggested the drug does not increase the overall risk of heart attack, stroke or death.

A spokeswoman for GlaxoSmithKline Inc. said the company is looking forward to a "rigorous scientific discussion" when the U.S. Food and Drug Administration holds its hearing into Avandia on July 13 and 14.

The FDA panel is expected to consider whether to order an end to a large international study comparing Avandia and Actos -- known as the TIDE trial -- as well as whether Avandia should remain on the market.

In May, Juurlink and the U.S. consumer advocacy group Public Citizen called on the FDA to end the TIDE trial, saying it would be unethical to continue it.

Juurlink is a specialist in internal medicine and clinical pharmacology at Sunnybrook Health Sciences Centre. He published a study last summer which showed that people taking Actos for Type 2 diabetes were 23 per cent less likely to be hospitalized for heart failure and 14 per cent less likely to die than people taking Avandia.

The new studies were rushed to print by sister publications the Journal of the American Medical Association and Archives of Internal Medicine.

Juurlink's editorial appeared in JAMA, along with a study by Dr. David Graham and colleagues which compared health outcomes of more than 225,000 American seniors taking Avandia or Actos.

Graham's study didn't randomly assign people to take one or the other drug, but rather looked at the health records of people in the U.S. Medicare system using the drugs for diabetes control.

They found people on Avandia were 27 per cent more likely to have a stroke, 25 per cent more likely to suffer heart failure and 14 per cent more likely to die than seniors taking Actos.

"For every 60 people you treat with rosiglitazone, one extra person in that group is going to have one of those bad things happen to them because they're taking Avandia instead of Actos," explained Graham, an outspoken FDA drug safety researcher whose earlier work contributed to the market withdrawal of the painkiller Vioxx.

Given how many people take Avandia, that's a significant additional risk, he said. Graham argued that there are no benefits to Avandia that offset the increased cardiovascular risk. No studies suggest Avandia is in any way better than Actos, he insisted.

"By treating your patients with rosiglitazone, physicians are not helping their patients. They're placing them at increased risk and for no reason. There's no gain in it," Graham said.

One of the principal investigators of the TIDE trial countered that Graham's study and the other with which it was published do not answer the question TIDE was set up to investigate.

Dr. Hertzel Gerstein, a diabetes expert at McMaster University in Hamilton, said there is no way of knowing whether the patients getting Avandia in Graham's study were similar, healthwise, to those who got Actos. He believes the jury is still out and he still prescribes both drugs in his practice.

"My view of the evidence is that all of these analyses tell us that the question regarding the comparability of the drugs is unanswered," said Gerstein, who disclosed that he has received funds from GSK for scientific advice and for giving scientific presentations on diabetes prevention and care.

The other study was an update of earlier research that put Avandia on the FDA's radar.

In 2007, leading cardiologist Dr. Steven Nissen of the Cleveland Clinic published a meta-analysis of studies on the drug that prompted the FDA to convene an earlier panel on Avandia. At that time the FDA decided the evidence was not conclusive and asked GlaxoSmithKline to conduct the TIDE trial to come up with definitive answers.

In the new article, titled "Rosiglitazone Revisited," Nissen and colleague Kathy Wolshi pooled data from 56 randomized controlled trials which enrolled more than 35,500 patients. Just over half took Avandia and the rest received other diabetes control treatments.

In Nissen's analysis those taking Avandia were between 28 per cent and 39 per cent more likely to have a heart attack than those using other forms of diabetes control.

That's unacceptable in a population already at high risk of heart disease, he suggested, noting that 70 per cent of diabetics die of heart disease.

"A drug that increases the risk of heart attack by a third or more in a diabetic population is truly an enormous public health burden," said Nissen, who is chair of cardiovascular medicine at the Cleveland Clinic.