News Article
Metformin May Be Cardioprotective, Meta-Analysis Suggests
Among oral diabetes medications, only metformin appears possibly to be cardioprotective, and only rosiglitazone appears possibly to harm the cardiovascular system, according to a meta-analysis of 40 studies.
Other oral diabetes agents appear to have no significant specific cardiovascular (CV) effects, positive or negative, according to the meta-analysis presented in the Oct. 27 issue of the Archives of Internal Medicine.
However, even these qualified findings on these drugs' specific cardiovascular effects must be considered "inconclusive" because clinical studies of the issue have been too small, too short term, or too flawed to provide definitive answers, according to Elizabeth Selvin, Ph.D., of Johns Hopkins University's Bloomberg School of Public Health, Baltimore, and her associates.
Dr. Selvin and her colleagues performed the meta-analysis because the cardiovascular effects of oral diabetes medications are unclear and, in some cases, controversial. They identified only 40 randomized, controlled trials of these drugs that reported on CV events and CV mortality; most of these were not statistically powered to assess CV events, and only two had follow-up for longer than 2 years.
In general, "there have been few trials of oral diabetes therapies that have lasted longer than 6 months," and reporting of CV events has been consistently "poor," the investigators said.
Nevertheless, after pooling the results of the 40 studies, they found that metformin was associated with a statistically significant decrease in CV mortality, with an odds ratio of 0.74, compared with placebo or any other oral diabetes therapy.
In addition, "when compared with any other diabetes agent or placebo, rosiglitazone was the only therapy associated with a possible increase in the risk of CV morbidity or mortality, but these results were not statistically significant," Dr. Selvin and her associates said (Arch. Intern. Med. 2008;168:2070-80).
"Larger, longer-term studies taken to hard end points and better reporting of CV events in short-term studies will be required to draw firm conclusions about major clinical benefits and risks related to oral diabetes agents," they noted.
In an editorial comment, Dr. David M. Nathan of Massachusetts General Hospital, Boston, said that this study adds to "the cottage industry of meta-analyses" and sheds some light on the issue.
However, "in the end ... the conclusions drawn will be disappointing for health care practitioners who want a clear answer to the question, ‘Is it safe?' " he said (Arch. Intern. Med. 2008;168:2064-6).
Overall, the meta-analysis results are in accord with the current consensus algorithm, favoring the use of metformin and the sulfonylureas over thiazolidinediones and limiting the use of rosiglitazone because of its CV risks, Dr. Nathan said.
No conflicts of interest were reported by Dr. Selvin or Dr. Nathan.
All articles have been reviewed by members of our Editorial Board or independent referees.
You need Adobe Acrobat Reader installed on your computer to read the PDF version of these articles. If you don't have Acrobat Reader, click here to download it free.
