Avandia and Heart Attacks
What Does the Research Say About Avandia and Heart Attacks?
The study that sparked the concern about heart attacks with Avandia was published online in The New England Journal of Medicine on May 21, 2007. The study is actually a "meta analysis," which simply means that the results of several previous studies were compiled into one big study, using statistical techniques. It combined several studies that compared Avandia to various other diabetes drugs or to a placebo (a "sugar pill").
This study reported that people taking Avandia were 43 percent more likely to have a heart attack, compared to people taking other diabetes medicines or a placebo. Although many news reports and Web sites are reporting that this study also suggested that there was an increased risk of death due to heart problems, the study does not support this conclusion. The study showed that the difference between Avandia and the other medications in term of death due to heart problems could be due to chance alone (that it was not a statistically significant difference).
Avandia and Heart Attacks: Is the Risk Real?While a 43 percent increase may seem huge, it is important to understand that the risk is still small. Only 86 out of 14,371 people taking Avandia had a heart attack. This is equivalent to about 0.6 percent, or about 6 in every 1,000 people. In comparison, for those taking other diabetes medications or a placebo, 72 out of 11,634 people had a heart attack. This is also about 0.6 percent. (The true risk may actually be lower, since this meta analysis did not include information from studies in which no heart attacks occurred.)
However, when studies are combined to make one enormous study, even tiny differences can seem huge, depending on the way the results are reported. When studies are combined, it is much more complicated than simply taking an average. Complex statistical tests are performed, such as the statistical test that concluded that the risk of heart disease was 43 percent greater for Avandia.
This study has been criticized for several reasons. First, some have suggested that this meta analysis study was not necessary. Meta analysis studies should be performed when many small studies (but no large studies) are available. With Avandia, there have been several very large studies performed. When large studies are available, meta analysis studies are not particularly useful or informative. Secondly, these large studies individually showed that any increase in the risk of heart attacks for those people taking Avandia was not statistically significant (that it could be due to chance).
Since that 2007 meta-analysis, the United States Food and Drug Administration (FDA) has been carefully examining this issue. Further meta-analyses have shown a small but statistically significant increase of heart attacks when Avandia was compared to a placebo. However, when Avandia was compared with other diabetes medications instead of a placebo, the risk was not statistically significant. It is unclear how exactly these results should be interpreted.
There are a few reasons why Avandia might increase the risk of heart attacks. First, Avandia can increase cholesterol (see Lipid Effects of Avandia). However, while Avandia increases LDL cholesterol ("bad cholesterol"), it usually also increases HDL ("good cholesterol"). Avandia can also cause weight gain (see Avandia and Weight Gain). Both of these effects can increase the risk of heart attacks. Or, Avandia could increase the risk of heart attacks by an unknown mechanism.