Sulfonylureas Lose Again

Many people with diabetes have complained about the sulfonylurea drugs for years because these medicines encourage weight gain and can cause hypoglycemic episodes (lows). Now a large study from the United Kingdom has found that, compared to metformin (brand name Glucophage and others), sulfonylurea drugs are associated with higher risks of death and heart failure.


Sulfonylureas act by pushing struggling beta cells to produce more insulin. The increased insulin lowers blood glucose, creating the risk of lows. Their effect on glucose cannot be calculated as accurately as the effect of injected insulin. And they are thought by some people to wear the beta cells out, so you’ll eventually need insulin anyway.

There are two widely used “generations” of sulfonylureas. (A third was more recently developed.) First-generation sulfonylureas were developed in the 1940’s and became available in 1955. They include chlorpropamide (Diabinese), tolbutamide (Orinase), and tolazamide (Tolinase). Second-generation drugs came on-line in the 80’s. They include glipizide (Glucotrol) and glyburide (DiaBeta, Micronase, Glycron, and Glynase).

Second-generation sulfonylureas have been better in many ways. They are more potent and seem to have fewer side effects. But in the UK study, single-drug treatment with first- and second-generation sulfonylureas was associated with up to a 61% increased risk for death compared with metformin. Users of second-generation sulfonylureas had up to a 30% higher risk for congestive heart failure.

The study looked at more than 91,500 British men and women with Type 2 diabetes. They (or their medical records) were followed for an average of 7.1 years. The researchers controlled for sex, duration of diabetes, previous complications from diabetes, peripheral artery disease, cardiovascular disease, other medicines, body-mass index, cholesterol levels, systolic blood pressure, HbA1c, creatinine and albumin concentrations, and smoking status (whew!), so these results look pretty convincing.

In my opinion, the only reason to take a sulfonylurea is if you cannot take metformin, and even then, it’s worth asking about other alternatives (such as insulin or no meds at all). But everyone is different. If you are doing well on a sulfonylurea, you might want to stay on it. Talk with your doctor about your concerns.

And Now for Something Completely Different
This is fascinating. According to ScienceDaily, “researchers at the Salk Institute for Biological Studies in La Jolla, CA, found that when you eat may be just as vital to your health as what you eat.” According to lead researcher Satchidananda Panda, PhD, many of our bodies’ genes set their clocks by when we eat and fast, not by the rhythms of daylight and dark. These rhythms of eating and fasting control when we release different enzymes and other body chemicals.

This is a huge change from the old belief that “Circadian rhythms” control most body functions. Circadian rhythms help us wake up in the morning and go to sleep at night, for example.

According to the new theory, there is a “master clock” in the brain that responds to light. But there are subsidiary clocks [or “oscillators”] in the other tissues that respond to other factors such as feeding time.

According to Dr. Panda, the liver oscillator is especially important. It “tun[es] the activity of thousands of genes regulating metabolism and physiology,” he says. That means your body will be ready to digest and absorb food, produce insulin if you have any, and produce other chemicals needed to handle food well.

But if the oscillator is thrown off by wildly changing times of food intake, you won’t have the right chemical mix in your blood to benefit from food. The Salk researchers say their findings could explain why “shift workers are unusually prone to metabolic syndrome, diabetes, high cholesterol levels and obesity.”

Of course, if you take insulin, you probably try to eat at consistent times already. But here’s another reason, even if you’re not on insulin. You’ll get more benefit and less harm from your food if you eat according to your body’s internal clocks.

Learn more about the health and medical experts who who provide you with the cutting-edge resources, tools, news, and more on Diabetes Self-Management.
About Our Experts >>

  • CalgaryDiabetic

    Dear David.

    Right I the money I fully agree. The sulphonyl ureas should not be used in any except the most mild diabetics or very old people that can’t cope with injections and BG measurements. All other diabetics should be on metformin and if this does not bring the BG into the normal range. They should add a long lasting insulin and a short lasting one before meals.

  • Steve Parker, M.D.

    I love studies like the UK study above. Now that we have something like nine different classes of drugs for type2 diabetes, how to we choose which to use? The UK study helps us decide.

    The American Association of Clinical Endocrinologists and the American College of Endocrinology have just issued a joint statement with specific drug recommendations. Their algorithm is quite detailed. Here are a few highlights some of your readers might not know about:

    -REGULAR human insulin is not recommended
    -NPH insulin is not recommended
    -The following should be used earlier and more frequently: GLP-1 agonists (exenatide) and DPP-4 inhibitors (sitagliptin and saxagliptin)
    -sulfonylureas are a lower priority
    -metformin is still a key drug


    Reference: Statement by an American Association of Clinical Endocrinologists/American College of Endocrinology consensus panel on type 2 diabetes mellitus: An algorithm for glycemic control. Endocrine Practice, 15 (2009): 540-559.

  • David Spero RN

    Hi Steve,

    So what kind of insulin do they recommend? Or do they say it’s best always to use the newest, most expensive drugs, and not use insulin? I just wonder about that — a few drugs haven’t been around that long. Some nasty effects may still come out. And the cost is high. Thanks for keeping the community appraised of reports like this.

  • CalgaryDiabetic

    I agree with steve that NPH is hard to use I ended up injecting the daily dose divided into 6 separate injections over 24 hours to give the basal dose without annoying lows. Lantus is much easier to use I can split the daily dose into 2 injections at 7 am and 7 pm seems to work fine not sure if spliting into 3 would give any further benefit. My Endo said a lot of his patients had problems with NPH giving them lows, it can see that used in the prescribed manner I couldnot make any sense out of the BG readings very scarry.

    Dear David it is interesting to read about the liver oscillations this is also true about the pancreas that supplies its basal insulin as an oscillation which unfortunately you cannot reproduce with injections as done currently. See insulin oscillations on wiki.

  • Steve


    Just curious… do you support and approve of the ADA food pyramid?

    Also… have you looked at the recipes on this website and if so do you approve?

    Steve Cooksey

  • Harry…………………….

    I have no problem with glyberide!

  • Angela Norton

    I disagree with an earlier comment:
    “Sulfonylurea’s should be reserved for use in the elderly”. It can be extremely dangerous in this population due to diminished renal function, and higher likelyhood of developping CHF.
    I have another way of looking at the insulin vs oral debate: insulin is a hormone, and orals are drugs. Our pancreas when working produces axactly the correct amount of insulin for our needs under different hormonal circumstances. We are just not able to respond with exogenous insulin quite as efficiently, but we are getting there with the new analog insulin’s, insulin pumps, and continous glucose monitors.
    There are a number of studies showing earlier use of insulin is benificial.

  • Carol

    As a nurse educator, I have been a proponent of early use insulin for years. I work to convince patients of the benefits (better control, at times less money) of an injectable vs a pill. I have never been able to figure out why people are so resistant to taking an injection. Those patients who do agree to see me and start insulin are surprised at how little, if any pain, is involved. I can compare the reluctant to an old wives’ tale; myths are hard to break once they have been established. This seems to be true of insulin injections and waiting far too long to begin. It’s up to the doctors to convince patients of the benefits. Depending on initial lab results and other findings, I would usually start with metformin then move to a long-acting insulin. Monitor and add premeal when this combo no longer is sufficient. Of course, meeting with an RD for a Meal Plan and encouraging activity are also important components as well. I have found glipizide more tolerable in the older population than glyburide.
    OH RN

  • Terrance

    With all the side effects from the pills I went to insulin in pen and am happier with the results and Less side effects.

  • Marcie

    Actually, I’ve read in a couple of places that this higher incidence of heart problems has been known of for some time. The first time I rad it was probably about 5 years ago.

    I’ve been on Byetta since it hit the shelves, and I love it. Yes, I’m still taking metformin, but at a minimal dose now. I read just a few days ago, though, that they’re dropping the requirement that a person be on metformin if they’re going to be taking Byetta.

    Personally, I’m very impatiently waiting for once-a-week Byetta to come out.