Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 May 26; 334(7603): 1073.
PMCID: PMC1877896

Study indicates diabetes drug linked to cardiovascular death

A meta-analysis of 42 trials of the type 2 diabetes drug rosiglitazone (Avandia) has shown a significantly raised risk of myocardial infarction and an increase in cardiovascular deaths that did not quite reach statistical significance (New England Journal of Medicine 2007 May 21 doi: 10.1056/NEJMoa072761).

The analysis by Steven Nissen and Kathy Wolski, of the Cleveland Clinic, Cleveland, Ohio, included 15 560 patients randomly assigned to regimens that included rosiglitazone, and 12 283 patients assigned to regimens that did not. The mean age of patients was 56 years, and the mean baseline glycated haemoglobin concentration was about 8.2%.

Patients receiving rosiglitazone had an odds ratio for myocardial infarction of 1.43 (95% confidence interval 1.03 to 1.98, P=0.03). The odds ratio for death from cardiovascular causes was 1.64 (0.98 to 2.74, P=0.06).

An editorial (doi: 10.1056/NEJMe078099) criticised the Food and Drug Administration for approving the drug on the basis of only its efficacy in glycaemic control in short term studies and also for a poor system for reporting adverse events after marketing.

Bruce Psaty of the University of Washington in Seattle and Curt Furberg of Wake Forest University in Winston-Salem, North Carolina, also say that the Food and Drug Administration Revitalization Act, now moving through Congress, should be strengthened so that the FDA can follow a drug through its life cycle, with new emphasis on postmarketing.

The FDA posted a safety alert on its website, cautioning patients taking the drug who have heart disease or are at high risk of a heart attack to talk to their doctors (

The alert said that data from controlled clinical trials showed “a potentially significant increase in the risk of heart attack and heart-related problems in patients taking Avandia. However, other published and unpublished data from long-term clinical trials of Avandia . . . provide contradictory evidence about the risks in patients treated with Avandia . . . For these reasons, FDA is not asking GlaxoSmithKline . . . to take any specific action at this time.”

Rosiglitazone was approved by the FDA in 1999 to treat type 2 diabetes. The disease affects as many as 20 million US residents, and there are 1.5 million new cases each year.

Rosiglitazone is a thiazolidinedione drug that lowers blood glucose by increasing insulin sensitivity in peripheral tissues. Other actions of such drugs are not completely understood, and it is not clear whether the other drug in this class, pioglitazone, has less, similar, or greater risks, the FDA said.

GlaxoSmithKline said it “strongly disagrees” with the papers and defended the drug on its website ( It said the Nissen paper was “a meta-analysis, which is not the most rigorous way to reach definite conclusions about adverse events. Each study is designed differently and looks at unique questions.”

A better way “to examine the safety and benefits of a medicine is to conduct large scale, long-term clinical trials in patients with the disease.” Several trials of this type are going on, and two trials and a comprehensive analysis of more than 33 000 people with diabetes had shown no increase in cardiovascular risk, the statement said.

An editorial in the BMJ by Carl Heneghan, deputy director of the Centre for Evidence-Based Medicine at Oxford University, last October, warned that the finding that rosiglitazone could prevent diabetes in people at risk of getting diabetes needed to be tempered with possible adverse effects of the drug, such as heart failure and perhaps the risk of medicalising a lifestyle problem (2006;333:764-5 doi: 10.1136/bmj.38996.709340.BE).

Articles from The BMJ are provided here courtesy of BMJ Group