The Cochran Firm Legal Blog

This week, three studies were published on GlaxoSmithKline's blockbuster diabetes drug Avandia (rosiglitazone). Their results were announced at the 2010 meeting of the American Diabetes Association. Two of the studies have shown that Avandia (rosiglitazone) is a dangerous drug whose use is associated with a higher risk of cardiovascular events such as myocardial infarction (MI, or heart attack), stroke, heart failure, and death, while the third seemed to show that there was no such risk. One study, a follow-up by cardiologist Steven E. Nissen, who dropped the first big cardiovascular bomb on Avandia in 2007, is a meta-analysis of all randomized trials of Avandia that were 24 months or longer. This study, published in the Archives of Internal Medicine, showed that there was a statistically significant increase in risk of MI (28%), but not cardiovascular or all-cause deaths. Notably, when Avandia's RECORD study (see below) was excluded, the relative risk of MI and cardiovascular death jumped up (to 139% and 146%, respectively).

The second study is a review of data from Medicare patients by FDA scientists. The FDA data reviewed patient records for 227,571 Medicare patients who were on diabetes control drugs Avandia or Actos (proglitazone) showed that there was no statistically-significant difference in MI, but heart failure seemed to be increased by 27% and CV death by 14%.

A third study, also published this week, showed that there was no increased risk of cardiovascular events, and adds to the long string of "denial studies" designed by GSK. It was already included in the list of studies assembled by the company to refute claims put forward by the large observational studies and meta-analyses that cast doubt on Avandia. The authors of this study, designated BARI 2-D, noted that its data "does not suggest a significant cardiovascular hazard and may suggest a beneficial effect on ischemic cardiovascular events associated with treatment with rosiglitazone [Avandia]." However, one important factor in this latest study is that Avandia was not randomly assigned, but was deliberately prescribed by physicians, who would more likely prescribe it to patients with a lower risk of cardiovascular events, which makes this study essentially worthless for determining the heart attack risk associated with the drug.

The BARI 2-D study joins five other studies in GSK's list of controlled clinical trials that show "rosiglitazone does not increase the overall risk of heart attack, stroke[,] or death." However, taking these studies together is exactly what a meta-analysis does, and it shows there is an increased risk. And if you look at the studies individually, every one seems designed with a small enough sample size to be statistically inconclusive about the risk of cardiovascular events associated with Avandia so that GSK can continue saying that its drug "does not increase the overall risk" of this dangerous side effect.

Consider, for example, the RECORD study. The study was designed to evaluate the risk of cardiovascular events due to Avandia use (after all, it's in the name: RECORD stands for "Rosiglitazone Evaluated for Cardiovascular outcomes in ORal agent combination therapy for type 2 Diabetes"), but as early as 2004, the company knew the study was "underpowered" to resolve the question ("Staff Report on GlaxoSmithKline and the Diabetes Drug Avandia"). But instead of designing a new study instead of or in addition to RECORD, the company persisted with RECORD for five years, and even published interim, inconclusive results after three years in an attempt to refute Nissen's meta-analysis.

Another study cited by GSK as showing no relative risk of Avandia is the ACCORD study. The ACCORD study was called off early because of the high mortality rate of its participants. This was not due to Avandia, according to the study's author, but because the intensive therapy method of diabetes treatment (a controversial approach) was flawed. However, another study on the approach, the ADVANCE Study, showed no inherent risk of intensive therapy for diabetes. One distinguishing factor between these two studies is that over 90% of the intensive therapy group for ACCORD were given Avandia, whereas the ADVANCE study used other medications for its intensive therapy. Looking at the data, even the ACCORD study suggests a small, not statistically-significant, increase in fatal or nonfatal heart failure.

If GSK really wanted to resolve the issue of Avandia's heart-attack risk, it could design a study to do so, but it has not. Since nearly the introduction of Avandia, GSK seems to have been working to create a fog of studies, each of which seems to show that Avandia is not associated with heart risks while concealing a growing body of evidence that it is.

If you have suffered a heart attack associated with Avandia, or a side effect associated with a different dangerous pharmaceutical drug, the lawyers of The Cochran Firm stands ready to help you pursue compensation for your injuries. To learn more, please contact us today for a free initial consultation.