Saturday, December 27, 2014

The Nutrition Debate #274: “Should Everyone Take Metformin?”

I had to laugh. Of course, by “everyone” the title of this Medscape piece meant, I thought, as the “Initial Choice of Oral Glucose-Lowering Medication” for [treatment of type 2] Diabetes Mellitus.” But for a moment I thought that the authors meant “everyone everyone.” That’s the way the new guidelines for prescribing statins are being interpreted, at least for everyone over 39 years of age (and under 76), and that for the very dubious, almost exclusive purpose of lowering LDL-C. But if you want to read about that, you can go to The Nutrition Debate # 180, “The AHA/ACC Cholesterol Guidelines.”

No, this story, published in JAMA Internal Medicine, was an “observational cohort study [that] sought to determine the effect of initial oral glucose-lowering class on subsequent need to additional anti-hyperglycemia therapy.” The 15,516 participants, none of whom had previously been treated for diabetes, were started on 1) metformin, 2) a sulfonylurea, like glyburide or glipizide, 3) a TZD like Avandia, or 4) a DPP4, like Januvia and Onglyza. “The primary outcome was time to treatment intensification, defined as initiation of a different class of oral glucose-lowering medication,” Medscape said.

“Secondary outcomes included time to composite cardiovascular event (coronary heart disease, congestive heart failure, unstable angina, ischemic stroke, acute myocardial infarction [heart attack], or a revascularization procedure), congestive heart failure alone, an emergency department visit or hospital visit for hypoglycemia, and any other diabetes related emergency department visit.” That’s one heck of a scary list of secondary outcomes. Something to think about…

“The Winner, and Still Champion: Metformin,” the Medscape sub-head declared. Well, there should be no surprise there. But if you are now pre-diabetic, or when first diagnosed a type 2 diabetic, you aren’t (weren’t) started on Metformin, you might want to print this post out – better yet, go to the Medscape and JAMA Internal Medicine links above, print them out, and give them to your doctor and ask why you were not. I’d be interested to hear his or her answer.

The FINDINGS: “58% of the patients began therapy with metformin, 23% with a SU [sulfonylurea], 6% with a TZD and 13% with a DPP4.” During the “median follow-up of slightly more than 1 year, subsequent treatment intensification differed significantly by drug class. Of patients prescribed metformin, 25% required a second oral agent, compared to 37% of SU recipients, 40% of TZD recipients, and 36% of patients taking a DPP4.” This implies significantly different effectiveness.

The Medscape piece amplifies this finding: “Relative to metformin uses, the risk of treatment intensification was 68% greater among SU users, 61% greater among TZD users, and 62% greater among DPP4 users.” That’s really significant.

As to the “secondary outcomes,” Medscape states this simple but dramatic finding: “Also relative to metformin, SU use was associated with an increased risk for composite cardiovascular events, congestive heart failure, and hypoglycemia.”

Then this Medscape analysis of comparative effectiveness took an interesting turn. It began with this question, posed by another study: “Can People With Type 2 Diabetes Live Longer Than Those Without? A Comparison of Mortality in People Initiated With Metformin or Sulphonylurea Monotherapy and Matched Non-diabetic Controls.” This British study, which appeared in Diabetes Obesity and Metabolism, essentially asked, “Does Metformin Reduce Mortality?” Hmmm… Maybe the title of the Medscape piece was intended to mean “everyone everyone.” I read on to find out.

In this study, “patients initiating metformin therapy were compared with those initiating treatment with a SU, and both diabetic groups were compared with their matched nondiabetic controls.” Subsequent mortality was tracked for up to 5.5 years. The FINDINGS: “Crude death rates were substantially lower for metformin users that SU users” and, “All subgroup comparisons favored metformin over SU and were statistically significant.” But here’s the zinger.

“Perhaps the most striking finding was that survival time for controls [the non-diabetics who were not taking metformin] was 15% shorter than for matched metformin users. This finding was consistent across all subgroups, nearly all of which demonstrated statistical significance, and was particularly strong among patients with high comorbidity.”

“The protective effect of metformin relative to SUs was not a surprise,” the Medscape piece says. The remarkable finding was an apparent protective effect of metformin compared with nondiabetic individuals. Because of metformin’s favorable results among people with diabetes, it has been postulated that the drug may also provide benefit to people without diabetes.” Sort of like statin therapy… What are they saying? What am I saying? Everyone should take metformin??


  1. The main problem I have seen with Metformin for T2 diabetics is a total lack of education about what it can actually DO. I see people who think it works like insulin - "If I eat 2 cookies, I'll just taken an extra pill today." So they take the medication and still keep on eating high carb crap and...oh my...they just keep getting worse. Guess that calls for more medication, right? I think a better way to go than simply say everyone could benefit from metformin would be to find out why that seems to be, then figure out a way to duplicate the effect without the drug. I found I could no longer take metformin on a strict LCHF diet because the drug exacerbated very bad cramps. Why? Maybe because I wasn't storing glycogen in my liver on that diet, and the metformin compounded the problem? I don't know, and of course the doctor wasn't interested in finding out.

    1. I completely agree with your first point. Doctor’s are partly to blame, of course. They think, like Mr. Gruber, that Americans are too stupid to understand the principal mechanism of action of metformin, or at least the one most understood at the moment.

      My understanding is that the full mechanism and range of action of metformin is not understood. I haven’t studied this recently, but I always say that the main reason I take metformin is to deal with gluconeogenesis from eating too much PROTEIN at dinner. I want to suppress this unwanted glucose production from unused amino acids returned to the liver and stored there for this explicit purpose, as a backup source of glucose production when glycogen stores are low and nearly exhausted. After all, the body does require some glucose for various purposes where cells do not contain ATP.

      But patients think, as you say, that if they are being medicated for prediabetes or frank T2DM that that medication will suppress serum glucose from ingested CARBOHYDRATES. It does not.
      Of course, metformin apparently has other lesser and lesser known mechanisms of action, but suppression of gluconeogenesis is the one that needs to be told to patients so that they can understand that difference (between PROTEIN and CARBOHYDRATES), but most patients are low-information consumers of health information. They just want their doctors to be in charge of their healthcare. They are essentially in denial, and continue to be so at their own peril. But doctors also have a responsibility, and I’m not sure they are doing their part.

      Of course, my blog post was reproting on a epidemiological study that suggested that metformin has a benefit in terms of all cause mortality over a population of non- diabetics not taking it compared to a population that was taking it, which is an interesting finding and suggests yet another mechanism of action that I think does support further research. Of course, I like your idea too that once the mechanism is better understood, we could duplicate it in a way that did not require taking a drug. I am with you on that 100% as well.

  2. This is my first time I visit here. This is really awesome post ! please try to publish some more like this one.

    cardiology fellowship