THURSDAY, March 17 (HealthDay News) -- A new analysis confirms that those who take the diabetes drug Avandia are more likely to develop heart problems and die than those who take a similar type of diabetes medication.
"The effect on public health may be considerable," the analysis authors wrote.
Avandia (rosiglitazone) helps control blood sugar levels in the body, and is prescribed for patients with type 2 diabetes. Studies in recent years have differed about whether it boosts the risk of heart disease and death.
In response to growing concerns over the drug's cardiovascular effects, U.S. health officials in September restricted the use of Avandia to patients with type 2 diabetes who cannot control their disease on other medications.
In the new analysis, researchers looked at the results of 16 studies that involved 810,000 users of Avandia or Actos (pioglitazone), a similar diabetes medication. The findings appear online March 17 in the BMJ.
The researchers found a "modest but statistically significant increase" in the odds of certain heart conditions in those who took Avandia. The risk of heart attack rose by 16 percent and increased 23 percent for congestive heart failure. Overall, mortality rates rose 14 percent.
Avandia is a common drug, with about 3.8 million prescriptions a year in the United States.
The analysis authors cautioned that Actos carries its own risks. Like Avandia, it's been linked to a doubling of risk of fracture in women. And researchers worry that it may slightly boost the risk of bladder cancer.
"Patients should review this study with their physician to gain a better understanding of how their personal circumstances may or may not be reflected in the study, and talk to their physicians about the risks versus the benefit of taking the drug," said Dr. Debra Wertz, outcomes research manager for HealthCore Inc., the research subsidiary of the insurance company WellPoint.
Wertz co-authored a 2010 study that found no significant differences in risk of heart attack, heart failure or death in patients who took the two drugs.
"One thing to consider is that the HealthCore study evaluated a population that is commercially insured and potentially younger [most were under 65] and healthier" than the studies analyzed in the meta-analysis, Wertz noted.