The only person definitely known to have died as a consequence of an association with a low-carbohydrate diet is Dr. Herman Tarnower, author of the Scarsdale diet, although, as they used to say on the old TV detective shows, the immediate cause of death was lead poisoning. His girlfriend shot him. Not that folks haven’t been looking for other victims. The Atkins diet is still the bête noire of physicians, at least those who aren’t on it — a study published a few years ago said that physicians were more likely to follow a low carbohydrate diet when trying to lose weight themselves, while recommending a low fat diets for their patients.
But doctors, after all, have first to avoid harm, so when I gave a talk at at the European Society for the Study of diabetes (EASD) held in Vienna in 2008 on treating diabetes with carbohydrate restriction, someone in the audience was concerned about constipation. Constipation. This, in a disease that is the major cause of acquired blindness and second only to accidents as a cause of amputations. Is that what all the “concerns” are about? Is that really risk? Isn’t constipation just one of those minor intermittent annoyances that flesh is heir to? I don’t know whether low carb people have more constipation than anybody else but even if they do, let me try to explain what I think constitutes real risk. I’m going to ask: What would it be like if there were real risk in a low-carbohydrate diet?
Moderation in All Things. Except Drugs.
You always think of the best answer a few hours later. (I am grateful to Paula Nedved for the information that the French call it l’esprit d’escalier (literally, staircase wit). Another audience question that I got at the EASD conference was whether a more moderate version of the diet that I was talking about wouldn’t be better. My answer: “No. If you use a moderate diet. You get moderate results.” Not that bad, but, of course, it was on the way back to the hotel that I realized what I should have said:
“A low carbohydrate diet is moderate. If you want know what is extreme, it is putting into your body, a thiazoladinedione (TZD), a chemical unknown on the face of the earth until twenty-five years ago. A TZD is extreme. But TZD’s work. They lower HbA1c and increase insulin sensitivity and there is reduced risk of hypoglycemia. In other words, extreme is good if you are trying to deal with an extreme situation. Moderation gives you moderate results.”
That’s what I should have said.
But what if somebody had pointed out, as in the Perspective in the Permanente Journal: “However, the TZDs — specifically, rosiglitazone — have faced a great deal of criticism because of the discovery of worrisome adverse affects…. The most debated side effect is whether rosiglitazone causes heart attacks.”
Risk? Oui, d’ACCORD.
To get a sense of what it might be like if low carbohydrate diets actually caused risk, we should look at therapeutic interventions that have produced demonstrable problems. I was surprised to see the headline in one of the media outlets that I subscribe to.
The story is that the National Institutes of Health stopped the “intensive blood glucose-lowering arm” of the ACCORD trial. The name stands for Action to Control Cardiovascular Risk in Diabetes and, the cause for stopping that part of the trial was unexpected deaths, about half of which, according to the NIH, “were from cardiovascular diseases, such as heart attack, sudden cardiac death, stroke, heart failure, or another cardiovascular disease condition.” The intensive glucose-lowering, however, meant intensive drug treatment. For a branch of medicine built on pharmacology, such a failure must surely be a problem. Imagine if that were found in the (imaginary) low-carb arm of the trial. The NIH would be on the phone saying that Low-Carb kills. Low-carbohydrate groups, however, are never included in large trials. If you ask, they will say it is too risky but, really, one suspects it is because the low-carbohydrate diet might do as well in long-term trials as they do in short-term trials and ‘twould have anger’d any heart alive to hear the men deny’t.
In the current case, this would be very likely because what was targeted was control of blood glucose in people with diabetes which even the American Diabetes Association admits is best done by lowering carbohydrate intake. What to do with the ACCORD study? Simple, if you can’t hit the target, change the target. If intensive drug treatment to lower blood glucose has bad side effects, it must have failed because we shouldn’t have been targeting glucose. The announcement from ACCORD was received with amazement as if this were a great new scientific discovery. Elizabeth Nabel, then head of the National Heart Lung and Blood Institute issued a telephone report:
“… ACCORD is the first major clinical trial to study whether lowering a raised blood sugar level, to a level similar to that seen in people without diabetes, reduces the risk of cardiovascular disease. We now have one part of the answer to this question. The study will continue to examine other ways to lower the risk of cardiovascular disease, in high-risk adults, with Type 2 diabetes using blood pressure and lipid lowering approaches.”
And, as described by Gina Kolata in the Times:
“Dr. John Buse, the vice-chairman of the study’s steering committee and the president of medicine and science at the American Diabetes Association, described what was required to get blood sugar levels low:…’Many were taking four or five shots of insulin a day,” he said. “Some were using insulin pumps. Some were monitoring their blood sugar seven or eight times a day.’ They also took pills to lower their blood sugar, in addition to the pills they took for other medical conditions and to lower their blood pressure and cholesterol.”
Paraphrasing all the people to whom the original has been attributed: “If you can’t drive it in with your hammer, maybe it’s not a nail.” You can probably see where this post going: the major health risk associated with a low-carbohydrate is that you might be talked out of it and get into a program where you will do something else.
The movable target.
In any melt-down there’s always somebody who saw it coming. Stephen Havas’s comment in Archives of Internal Medicine described how
“led to numerous stories in the mass media noting how experts in the field were stunned by the findings.1 Yet the results of this trial were predictable. For almost 40 years, there has been evidence that intensive lowering of glucose levels in patients with type 2 diabetes mellitus (DM) can lead to significant harm and has limited benefits.”
This would certainly be news to people who have consistently lowered their blood glucose by restricting dietary carbohydrate and found great benefit and no harm.
“Unlike blood glucose level, there is strong evidence that controlling high BP and high blood cholesterol levels significantly reduces both macrovascular and microvascular complications in persons with type 2 DM. Clinicians should therefore focus more on controlling these other risk factors than on glucose levels.” [my italics]
I don’t know whether there are physicians who think that it is either/or, but how should we target cholesterol? You’re kidding. Right?
Good news – bad news.
The headline in Medpage Today tells us that “Large Study Affirms Safety of Statins” which should be good news since everybody has somebody close to them who is on statins. The story is about a meta-analysis — that is where you have a bunch of weak studies, most of which have shown nothing or very little and you average them on the assumption that many wrongs make a valid scientific conclusion.
The study “confirmed that statins as a class are well tolerated, although safety profiles vary from agent to agent, researchers found.” Although? It turns out “Statin therapy was associated with increased odds of diabetes and elevations of liver enzymes compared with placebo” but, the good news: “there were no differences in development of myalgia or cancer, elevations in creatine kinase, or” — and here, I am not sure whether this is good news or bad news — “there were no … discontinuations because of adverse events.” The problem in statins revolves around the concept of “gain outweighs risk,” so maybe there should have been a few discontinuations rather than suffering adverse events on the chance that it was worth it. And there are more disclaimers: “There is this risk of some diabetes and you have to check some liver function tests, which we’ve known about, but …” The figure shows what I call the interlocking pharmacologic risk paradigm.
Cover-up – the triumph of acronyms.
“I’m not making this up.”
— Anna Russell, Wagner’s Ring Cycle.
What the Permanente Journal, cited above, was referring to was Avandia® (Rosiglitazone). I don’t know what researchers in carbohydrate restriction would do if the “concerns” of the nay-sayers had actually panned out — if there had actually been risk of the magnitude of the side-effects of drugs. I doubt that we would have convened a committee of experts to tell us that it was really ok. The FDA, on the other hand, has an Endocrinologic and Metabolic Drugs Advisory Committee and Drug Safety and Risk Management Advisory Committee. The committee has at its disposal one of the most remarkable programs called Risk Evaluation and Mitigation Strategy, or REMS. Part of this is the Elements to Assure Safe Use (ETASU) program which, according to one of the health bulletins that I subscribe to “was initiated in 2010 because of concerns about cardiovascular (CV) safety.” Armed with what is now the REMS/ETASU, most of the panel was for keeping Avandia® out there, to be freely prescribed again. Five of the members voted to continue the current REMS/ETASU guidelines which restrict Avandia® use but only 1 member (the consumer representative), voted to remove rosiglitazone altogether.
“Mitigation Strategy” is the scary part. (The strategy is in the spin from the committee report, not in the treatment). I don’t know if it is pronounced as words, as in the strict definition of an acronym, but REMS/ETASU is some kind of record. In a previous post on the language of food, I pointed out how French-derived or Latinate words are widely used to give an aura of seriousness that might not be deserved. Risk-Evaluation-and-Mitigation-Strategy-Elements-to-Assure-Safety, translated into Anglo-Saxon-derived words might be We Make You Think it’s Good (WMY-TIG ).
So, in the end, if there really were risk, my mail-box would have constant headlines, about low-carb diets, of the following type:
|Research ties metformin to cognitive impairment in diabetes||DiabetesPro SmartBrief||Sep 10, 2013|
|HPS2-THRIVE: High myopathy risk with niacin/laropiprant||theheart.org/Lipid/metabolic||Feb 26, 2013|
|Niacin/laropiprant products to be suspended worldwide||theheart.org/Lipid/metabolic||Jan 11, 2013|
|Statins linked to fatigue in randomized study||theheart.org/Lipid/metabolic||Jun 11, 2012|
|Statins Linked to Cataracts in Large, Retrospective Study||theheart.org/Cardiology||Sep 20, 2013|
But there would be good news too, along these lines:
|Most patients with statin intolerance can eventually tolerate therapy||Healio/Endocrine Today||Sep 10, 2013|
|STOMP: Atorvastatin doesn’t reduce muscle strength||heheart.org||Dec 4, 2012|
It is not known how we might implement REMS/ETASU to determine if the normalization of blood glucose, weight loss and improvement in atherogenic dyslipidemia as well as removal from drugs that are seen in low carbohydrate diets are really balanced by the threat of constipation. Nonetheless, it seems clear that the real risk of low-carb diets is that you will tell your doctor that you are on such a diet and they will come up with their recommendations based on the FDA-REMS/ETASU.