It’s over for rosiglitazone.
Sold in the U.S. as Avandia, rosiglitazone is a drug used to control type 2 diabetes either alone or in combination with insulin, metformin, or a sulfonylurea. It has only one competitor in its class: pioglitazone (sold as Actos).
Both drugs in the thiazolidinedione class (aka TZDs or glitazones) increase the risk of heart failure. Prior studies had suggested that rosiglitazone increases the risk of heart attack, heart failure, and death. Research suggested that pioglitazone actually reduces the risk of heart attack, stroke, and death.
A study just published in the Journal of the American Medical Association directly compared clinical use of rosiglitazone and pioglitazone. Investigators looked at Medicare data involving over 227,000 patients, average age 74, average follow-up of 105 days.
Rosiglitazone comes out the loser: users had significantly higher risk of stroke, heart failure, and death. Risk of heart attack trended a bit higher in the rosi users but did not reach statistical significance.
The researchers also calculated the composite risk of suffering either a heart attack, stroke, heart failure, or death: rosiglitazone risk was about 18% higher compared to pioglitazone.
What do these numbers mean from a practical viewpoint? The researchers calculated a “number needed to harm.” Treat 60 patients with rosi and 60 with pio for one year; the rosi group will have one extra event—heart attack, stroke, heart failure, or death—compared with the pio users.
Why put up with that risk? There’s no good reason. Especially when pioglitazone is available.
Implications
If you take rosiglitazone, ask your doctor to find an alternative or switch you to pioglitazone. Soon.
Clearly, we don’t know all of the adverse effects of many of the drugs doctors prescribe, whether for diabetes or other illnesses. We balance the good with the bad, and that equation changes over time.
Rosiglitazone’s manufacturer may pull the drug off the market voluntarily. If not, the FDA will do it. Cardiovascular disease—e.g., heart attacks, strokes, heart failure—kills 68% of diabetics. The last thing we need is a drug that increases that risk.
Within a month, you’ll see ads on U.S. television from trial lawyers asking if you or a loved one has been hurt by rosiglitazone. “If so, call this toll-free number now…”
Reference: Graham, D., Ouellet-Hellstrom, R., MaCurdy, T., Ali, F., Sholley, C., Worrall, C., & Kelman, J. (2010). Risk of Acute Myocardial Infarction, Stroke, Heart Failure, and Death in Elderly Medicare Patients Treated With Rosiglitazone or Pioglitazone JAMA: The Journal of the American Medical Association DOI: 10.1001/jama.2010.920
Diabetic Mediterranean Diet 
Given rosi’s liabilities, I’ve never understood why it was such a hot drug.
I’ve been hearing rumors of this for years. A big deal, though!
In the JAMA study at hand, use of pioglitazone was about twice that of rosiglitazone.
Avandia was the first and only med my doctor prescribed for me upon diagnosis (no mention of diet or exercise). Right after I started low carbing is when the first warnings were coming out on Avandia and I told the doctor I wanted to stop taking it. His response was “It’s Avandia or Insulin!!” which sounded both histrionic and ridiculous – bviously there were other options because my 2 sisters are both on other meds. However, at the end of the visit, he said my a1c was 5.8 so I could try going without meds at all. Wish I’d made him promise to eat his hat!
DarMA-
Your comment is sad and a little funny at the same time. By my count, we have 10 different classes of drugs we can use to treat type 2 diabetes. [They’re not all first line drugs, of course.]
Most clinic-based physicians are under tremendous time pressure to get patients in and out of the exam room quickly. That’s one of many excuses for the “It’s my way or the highway” attitude. It’s not a GOOD excuse, but . . .
-Steve
Do you get the AMA daily digests? Now internal GSK emails from 2001 indicate they knew rosiglitazone was no better than Actos, yest riskier to the heart, and “per sr. Mgmt request, these data should not see the light of day to anyone outside GSK. ”
I’m never surprised to hear these kinds of things given the billions of dollars involved, but as a physician it is always disheartening.
I hadn’t run across that, Emily.
Quite disturbing, if true. Perhaps even criminal.
-Steve
My info was from an email digest of news the AMA sends out daily. However, the story with those quotes is being reported by the New York Times and the Wall Street Journal, so I imagine they did their fact-checking!
NOT to defend GSK, but emails are easily taken out of context, especially when referring to studies. Perhaps they thought that study flawed, poor execution, confounders, etc, etc, etc.
Regardless, the proof is in the data at hand with rosi and as Steve stated above, there is no reason whatsoever to prescribe it. With the lawsuits that are going to line up, I’d be amazed at any doctor that was foolhardy enough to even think of prescribing it.