The safety of this diabetes drug remains open to question.
A recent meta-analysis (Journal Watch May 24 2007) found that rosiglitazone increased the risk for cardiovascular death (by 64%) and for myocardial infarction (by 43%). The authors of that meta-analysis acknowledged numerous limitations. Investigators present a reanalysis, focusing on two additional methodologic limitations — concerns about combining results from heterogeneous study designs and subject populations, and concerns about the use of a method of meta-analysis that excluded studies with no events in either randomized group.
In a reanalysis that addresses the concern about the choice of the meta-analytic method, the increase in risk for cardiovascular death associated with rosiglitazone (17%–51% across several different analyses) was lower than that in the original study and no longer statistically significant. The increase in risk for MI (26%–36% across several different analyses) was also lower than that in the original study and no longer statistically significant. In addition, the investigators cite several other studies — an unplanned interim analysis of a randomized trial, a meta-analysis done by the maker of the drug, a Cochrane review, and a large observational study — with results ranging from no increase in risk, to nonsignificant increases, to a significant 30% increase in risk for cardiovascular ischemic events.
Comment: Rosiglitazone may increase the risk for cardiovascular death or MI. But it seems clear that, based on current data, this conclusion remains uncertain. The authors of this reanalysis, who note that their meta-analysis suffers from many of the same limitations as the prior one, call for prospective studies to settle the question.
— Richard Saitz, MD, MPH, FACP, FASAM
Published in Journal Watch General Medicine August 16, 2007
Diamond GA et al. Uncertain effects of rosiglitazone on the risk for myocardial infarction and cardiovascular death. Ann Intern Med 2007 Aug 6; [e-pub ahead of print].