Rosiglitazone Controversy - ADOPT Rosi and Shatter Your DREAMs?

Ryan Bradley, ND, MPH     June, 2007


Our knowledge of complementary and alternative medicine treatments in diabetes is limited and we need to learn more. This much I am the first to admit.

Yet, when the New England Journal of Medicine article publishes research finding Avandia® (rosiglitazone, GlaxoSmithKline) may actually increase the risk of heart attack and cardiovascular-related death (1), I get frustrated and a little angry. The study combined the results of several recent studies and found that people with diabetes taking rosiglitazone have an average 43% increased risk of heart attack and a strong trend suggesting higher chance of death from cardiovascular causes. Diabetes itself increases risk of heart disease, heart attack and heart-related death, and now I read FDA-approved “treatments” further increases the risk. The FDA continues to use tax dollars to prematurely approve medications without adequate safety testing and monitoring while drug companies continue to aggressively market medications as “advances” in treatment.

The recent findings about rosiglitazone have raised significant questions about the competency of the FDA, the marketing ease of drug companies and the impact both of these factors have on the advice your doctors give you for your diabetes care.

I hope this article helps clarify some of the context surrounding the current controversy- and helps you ask your doctors for sound recommendations. 


What is PPAR Anyway?

Rosiglitazone is an example of a relatively new class of medications used in the “treatment” of diabetes called thiazolidinediones (TZDs) or PPAR-gamma agonists. PPAR is an abbreviation for: Peroxisome Proliferator Activated Receptors. When substances (both endogenous and exogenous) bind to these receptors it triggers numerous biochemical changes - including changes in gene expression. When genes are read in the body it sends a message to the machinery to make more proteins; every protein has a specific function - some functions include enzymes, hormones and structural building blocks. In the case of PPAR-gamma, agents that bind these receptors (like the TZDs) trigger changes in gene expression leading to the creation of enzymes that help regulate insulin sensitivity, inflammation and improve glucose breakdown (metabolism). TZD-class medications, like rosiglitazone, act as true “insulin sensitizers” and do help reduce blood sugar substantially (2).


ADOPT Rosiglitazone?

A very large, recent clinical trial called the ADOPT trial was the first clinical study to compare three oral medications for diabetes against each other, i.e. a “head-to-head” study (2). There have long been concerns in the medical community that some medications may “exhaust the insulin-producing beta cells” of the pancreas and which oral medication to be used in the treatment of diabetes has been unknown.

The ADOPT trial randomized participants into three groups: one group received metformin, one group received glyburide, and the third group received rosiglitazone. The participants were followed for a median time of four years. The study endpoints included time to treatment failure (treatment failure was defined as how long after the drug was started did fasting blood sugar rise over a certain level), side effects and cardiovascular events (including heart attack, stroke and congestive heart failure).

In my opinion, the main findings of the ADOPT trial was that all drugs fail. The amount of time it takes the drug to fail however was different - with five year failure rates being 15%, 21% and 34% for rosiglitazone, metformin and glyburide respectively.

The other findings were that patients on glyburide, although the medication fails at a higher rate, seem to have a lower rate of cardiovascular events than people on the other two medications (rosiglitazone and metformin had similar rates), however people on glyburide had more episodes of hypoglycemia than either of the other two medications. Also, those people taking rosiglitazone had higher rates of lower leg swelling and weight gain, while women taking rosiglitazone also had a higher rate of fractures.

Because it appears to last longer than other medications, some companies began using this study as rationale for using rosiglitazone as the first medication to prescribe for someone with new diabetes, despite the negative consequences of rosiglitazone.


Whose DREAM?

Another recent trial called the DREAM (Diabetes REduction Assessment with ramipril and rosiglitazone Medication) trial also studied rosiglitazone, but studied it for the prevention of diabetes in people with pre-diabetes (3). In this study, people with mild elevations in their fasting or post-meal blood sugar (elevations that put them at high risk for developing diabetes) were randomized to receive either rosiglitazone or placebo for three years. The study endpoint was the number of people in each group who went on to develop diabetes. The results of the DREAM trial showed the people who received rosiglitazone had a 60% reduced risk of developing diabetes compared to those who received a placebo. However, once again, the rosiglitazone had increased weight gain and in this study had significantly higher rates of developing congestive heart failure (even though they did not develop diabetes). 

The results of this study were widely praised because no medication has ever demonstrated similar preventive actions for diabetes, though diet and lifestyle change have demonstrated similar protective benefits - the Diabetes Prevention Program study showing a 58% reduction in diabetes from lifestyle changes (4).


So What Should I Do Now? Are There Natural PPAR Agents?

These two studies, ADOPT and DREAM, were being used to aggressively market rosiglitazone as a good first line option of the treatment of diabetes, and although never formally marketed for the treatment of diabetes, many drug company representatives were conveniently sliding the results of the DREAM trial into their conversations with doctors. The recent New England Journal of Medicine study finding increased heart attack risk (1), casts significant doubt in my mind regarding the safe use of rosiglitazone in the treatment of diabetes - and completely eliminates the possibility that I would ever recommend this medication for diabetes prevention (rosiglitazone has never been FDA approved for preventing diabetes).

There may be natural PPAR-gamma agents. Preliminary test-tube research has suggested several herbs and nutrients have PPAR-gamma activity including mulberry, Korean ginseng, phytanic acid (a breakdown product of chlorophyll from vegetables), conjugated linoleic acid, banaba, tumeric and licorice (5,6,7,8). However, only a few of these natural products have been studied in patients with diabetes, and therefore cannot be recommended as a substitute therapy. In fact, conjugated linoleic acid (CLA) has been tested in clinical studies in patients with diabetes and was actually found to worsen insulin sensitivity (9). More research is needed to determine the effectiveness of these products in diabetes.


Therefore I recommend the following:

If you are on rosiglitazone, it is appropriate for you to ask your doctor about alternative treatments, i.e. other medications. If your doctor dismisses your question, you should find another doctor because you deserve to get a straight, informed answer.

If you are not on rosiglitazone, until we know more about its safety I discourage people from starting it.


Other options for treatment include:

Focus on changing elements of your diet and exercise program that need to be reinvigorated. I think this recent study gives another lesson - you can only trust yourself to take care of your health. Diet and exercise remain as effective as medications and even insulin. Eat lots of vegetables; eliminate white, carbohydrate dense foods (bread, pasta, rice, potatoes, refined sugar); and exercise every day.

Other oral medications that have not been shown to cause the same safety issues. These medications include metformin and medications like glyburide, which can be used at least short term for improvements in blood sugar with confidence (though not necessarily without any side effects).

Consider talking with your doctor about starting insulin. Although insulin is often seem as a last resort, early use of insulin has advantages. Often one injection per day can help people lower their blood sugar to a healthy range and avoid using multiple oral medications. Also insulin needles are so thin now they really do not hurt.



People with diabetes need effective treatments to lower blood sugar and reduce risk of complications, and yet seemingly safe, less expensive, natural therapies do not seem to get adequate research funding from the National Institutes of Health or many private research foundations. Fortunately, Diabetes Action grants research dollars to researchers studying promising therapies because of its true commitment to people with diabetes and their families - and a true commitment to eliminate bias in medical research for the benefit of people affected by diabetes. This commitment deserves your support.

Although many promising treatments remain to be studied in more depth in diabetes, several treatments are always effective at lowering blood sugar and reducing cardiovascular risk: exercise; following a healthy, vegetable-rich, carbohydrate limited diet; managing stress, and loosing weight when needed. 



  1. Nissen and Wolski, New Eng J Med, 2007 
  2. Kahn et al., New Eng J Med, 2006
  3. DREAM Study Group, Lancet, 2006
  4. Diabetes Prevention Program Study Group, New Eng J Med, 2002
  5. Park et al. Life Sciences, 2005
  6. McCarty et al. Medical Hypotheses, 2001
  7. Chintharlapalli et al. Mol Cancer Ther, 2007
  8. Cheng et al. Chin Med J, 2007
  9. Moloney et al. Am J Clin Nutr, 2004