Given the hoopla over the side effects associated with Glaxo diabetes pill, this isn’t terribly surprising. Nonetheless, the Department of Veteran Affairs made its decision earlier this month after conducting its own review and concluded that, for some patients, Avandia “patients, may not afford the same margin of safety as alternative drug therapies,” according to The New York Times.

This is another blow to Glaxo, which is already struggling to cope with a drug that was once a $2 billion seller in the US. Since May, however, when the New England Journal of Medicine published a study suggesting Avandia could increase the risk of heart attacks by 43 percent, Avandia sales have plunged by an estimated 60 percent, the paper notes.

The VA accounts for 8 percent of Avandia sales, a Glaxo spokeswoman tells the paper. The agency, which treated more than five million veterans last year, issued more than 161,000 Avandia scrips from September 2006 to August 2007, Terry Jemison, a VA spokesman, tells the paper.

As part of that review, an agency advisory panel overwhelming recommended in July that the drug remain on the market. But such recommendations are not binding, and the agency is deeply split, with some officials pushing for Avandia’s withdrawal. One safety officer at the agency, David Graham, argued to the advisory panel that Avandia had caused up to 205,000 heart attacks and strokes since its approval in 1999.

Mary Anne Rhyne, the Glaxo spokeswoman, says it’s too soon to gauge the impact of the VA decision, because it’s not clear how many VA patients currently using Avandia would choose to keep taking it. “We are surprised and disappointed by the V.A. central office decision,” she says. “We do not believe it is in the best interest of patients.”

Glaxo argues there were shortcomings in the methodology of the New England study, conducted by Steve Nissen of the Cleveland Clinic, and insists that Avandia’s effect on the risk of heart attacks is similar to that of other oral medications for diabetes, including the older, widely used drugs metformin and sulfonylurea. Nissen’s findings, however, seemed to have been bolstered by another analysis, conducted at Wake Forest University and published in September in The Journal of the American Medical Association. That study’s conclusion was similar to his.

Jon LeCroy, a pharma analyst at Natixis Bleichroeder, says that before Nissen’s article last May, about one million scrips were being written each month for Avandia. As of September, Avandia prescriptions had declined to 426,000 a month, he says. But he adds there hasn’t been a corresponding increase in the use of other drugs for diabetes, indicating that some patients who had stopped taking Avandia had not replaced it in their drug regimens. Diabetes patients often take more than one medication at a time for their conditions.

Actos, a drug similar to Avandia that some studies indicate does not carry the same heart risk, has picked up 100,000 prescriptions a month, but that does not account for the drop of almost 600,000 in Avandia prescriptions. The Glaxo spokeswoman says a survey had shown that 50 percent of patients who discontinued Avandia didn’t begin another therapy as a substitute, and questioned whether those patients were placing themselves at risk for uncontrolled diabetes.