November 17, 2008

Myth Busting: Your Brain and That "Required" 130 Grams of Carbohydrate

My email has been filed this past week with emails from people with diabetes whose doctors or nutritionists have told them that it is dangerous to eat less than 130 grams of carbohydrates a day.

It isn't true. In fact, for most people with diabetes the opposite is true: eating more than 130 grams of carbs a day guarantees blood sugars that are so high they raise your risk of blindness, amputation, kidney failure and heart attack.

The old wives tale that you must eat 130 grams of carbohydrate a day has no basis in science. Is is one of those factoids that has been passed from teacher to student in the health profession for generations--long after anyone remembers where it originally came from. As it turns out, it came from two sources, one was ignorance of how the body works and the other a problem common 25 years ago that has been solved by medical progress.

Let's look at the origins of the damaging myth that you have to eat 130 grams of carbohydrate every day:

1. Though the brain requires carbs, you don't need to eat carbs to provide your brain with carbs.

The brain is unique among organs in its need for glucose. All your other organs can run on ketones or free fatty acids, both the byproducts of the metabolizing fat--but your brain does does require a certain amount of glucose to keep functioning--somewhere around 120 grams.

But before you rush off to eat a bagel for breakfast, you need to know while your brain needs glucose, you do not need to EAT glucose to provide your brain with the glucose that it needs. That is because your liver has the remarkable ability to transform the protein that you eat into glucose. This process is called "gluconeogenesis."

You liver can transform 58% of every gram of protein you eat into glucose, so any deficit created by eating less than 130 grams of carbohydrate can be made up by eating enough extra protein to supply the liver with the raw material to make some carbohydrate on its own.

But (as they say on the infomercials) that's not all! After a period of prolonged glucose "starvation" the brain's requirement for glucose drops. It becomes far more efficient and is able to run largely on the ketones produced by metabolizing fat. At that point, the brain's requirement drops to a mere 40 grams of glucose. If you ate no carbohydrate at all, you could can easily get that 40 grams of glucose from eating 69 grams of protein, or slightly over ten ounces of a protein-containing food like meat, cheese, or eggs. If, like most of us who limit our carbohydrates to control our blood sugars, you eat 45 to 100 grams a day, you will already be eating enough glucose to keep your brain happy.

This ability of the liver to make glucose from protein and its ability to adapt to burning ketones has been known for a long time. You can find it discussed in complex scientific language on pages 279 and 282 of the textbook, "Understanding the Brain and Its Development." Though the chapter assumes that a lowering of carbohydrate would be caused by "malnutrition" the same mechanisms occur when you lower carbohydrate and eat enough protein and fat to avoid malnutrition.

With this information, you can see that a person with diabetes who is eating 40-80 grams of carbohydrate a day can easily meet their brain's need for carbohydrates by eating a slightly higher protein intake for three weeks and then dropping it to a moderate amount--no more than 12 ounces a day for many people.

I have put together a calculator which will let you figure out exactly how much protein you need to eat to give your body both enough glucose to run your brain and additional protein to repair your muscles: Protein Need Calculator. Check out your current diet and see whether you are getting enough protein.

In case you might worry that your brain might still suffer if you make this adaptation to glucose "starvation" you might find it interesting to know that, far from it being dangerous, neurologists have found that eliminating all carbohydrates from the diet and providing high quality protein and fat often cures forms of childhood epilepsy that are otherwise untreatable with any drug or surgery, often permanently. Many parents also report improvements in their children's behavior when they are placed on these extreme low carb diets. (NOTE: The epilespy extreme low carb diet also severely limits fluids, which is not the case with the diabetes mildly lower carb diet.)

It is worth noting that the authors two of the best-regarded and bestselling books about diabetes, Dr. Bernstein and Gretchen Becker, both eat considerably less than 130 grams of carbohydrate a day. So, for that matter, do I. Perhaps another side effect of limiting carbohydrates is uncontrollable bibliographia*.

2. Low carb diets posed some danger to people using insulin and insulin-stimulating drugs before blood sugar meters came into use.

There is another reason why many doctors and nutritionists were trained to believe that a diet of less than 130-140 g of carbohydrate a day was dangerous. It is because, until the middle 1990s, few people with diabetes had access to blood sugar meters. At the same time, their diabetes was usually treated with sulfonylurea drugs that stimulate uncontrolled insulin production or with antique insulin regimens that involved taking one or two large shots of mixed insulins each day.

Because these people could not check their blood sugar in real time and had large, often unknown doses of insulin coursing through their bloodstream, there was always the fear that they might get too much insulin and have a life-threatening attack of hypoglycemia.

Back in these bad old days the only way people with diabetes could test their blood sugar at home was by dipping test strips in urine. These strips changed color when the concentration of glucose in the urine rose. Since most people don't spill glucose into urine until their blood sugar has been quite high for several hours, the lowest blood sugar that could be identified with these strip was one somewhere between 160 and 180 mg/dl (8.9 mmol/L and 10 mmol/L). If a person's blood sugar dropped below this level, it was impossible to tell what it was. It might be a normal 130 or a near-fatal 32 mg/dl (6.7 or 1.8 mmol/L). Not only that, but the "blood sugar" reading you saw on a urine test lagged several hours behind your blood sugar at the moment of testing so they told you what your blood sugar had been an hour or two before, not at the time of the test.

Given that situation, doctors and nutritionists urged people with diabetes to eat a lot of carbohydrate, enough to make sure they wouldn't suffer dangerous hypos no matter how much insulin was in their blood stream. To make it possible to check their blood sugar with urine test strips, many were urged to eat enough carbohydrate to keep their blood sugar above the 160 mg/dl level that showed up on urine test strips--which most people with diabetes can do if they eat at least 130 grams of carbohydrate a day.

But we have blood sugar meters now that can tell us exactly what our blood sugar is, any time we fear it might be dropping. We also have drugs for Type 2 diabetes that do not cause dangerous hypos, and much more controllable insulin regimens available to those of us who need to use insulin.

So for many of us hypos are not even an issue, and even for those of us using insulin, they are no where near the threat they used to be. That is why there is no longer any reason for anyone to ever eat a diet intended to push their blood sugar up over the high end of the normal range, which is 140 mg/dl (7.7 mmol/L). Doing so raises your likelihood of developing complications.

Bottom line: There is no reason to eat any set amount of carbohydrates. The best carbohydrate intake level is the one that keeps your blood sugar at a safe and normal level. This flyer will explain to you what that level is and how to reach those safe and normal blood sugar levels. If you live in a part of the world that uses the mmol/L measurements, download this version:

Even the American Diabetes Association--notorious for its hyper-conservative stance on diet--has stated in its 2008 practice recommendations that the low carb diet appears to be safe for people with diabetes.

If you are still battling a doctor or nutritionist whose last education about diet and diabetes took place decades ago, bring them some of the research studies proving the safety of low carb diets you'll find at HERE.

Bibliographia: medical term for writing books.


Anonymous said...


I was always taught (came from the Joslin) the conversion rate for protein was about 40% and for fat about 10%. As you know, entirely contrary to what any educators other than Dr. Bernstein tells patients. I'm just curious ... where did you get the higher protein glucose conversion rate?


Doris Dickson

Jenny said...


I'm almost certain I got that 58% from an earlier edition of Dr. Bernstein's book, though I note he no longer gives the conversion factors in the latest version. It was also repeated in dozens of discussions on the old newsgroup back in the late 1990s when it was full of highly argumentative people who spent their liesure hours looking up journal citations and arguing about complex physiological pathways. Lyle Macdonald, author of the book, The Ketogenic Diet also posted on the topic online back then.

The confusion comes because although the body can convert up to 58% of the dietary protein to carb, it doesn't have to unless it really needs glucose, so scientists come up with a lot of different numbers when they measure glucose produced after meals in various rodents and people.

Jenny said...

Note: I deleted an otherwise interesting comment because it linked to a discussion that ended up with the promotion of a much overhyped bogus substance that I do not want to promote on this blog.

The gist of the comment was that there is a theory that the brain in people with Alzheimers can still process ketones, but not glucose. This may or may not be true, but ketogenic diets do NOT improve brain function in people with Alzheimer's.

Nothing we have right now improves function in people with Alzheimer's in any significant way.

I really hate the way that hucksters selling expensive fad health products take advantage of the love and desperation of those watching their loved ones sink into dementia by telling them their product will help.

Anonymous said...

Dear Jenny, are you aware of the book "The Ketogenic Diet: A Complete Guide for the Dieter and Practitioner" ( It really answers many of the ketogenic related issues, in an accessible and (very) scientific way.

Jenny said...

I'm not only aware of Lyle McDonald's excellent book, I own a copy signed by the author.

However, I would not urge my blog readership to buy this book for several reasons. It is very expensive. It has not been updated since 1998, and much of the book is about using the diet as part of a body building training regimen that is not appropriate for people with diabetes.

That said, when it came out (and when I bought it in 1998) it was a revolutionary book because it cut through the myths spread by the low carb diet doctors (particularly Atkins) and explained a lot about the physiology of low carb dieting.

Unknown said...

"It becomes far more efficient and is able to run largely on the ketones produced by metabolizing fat. At that point, the brain's requirement for protein drops to a mere 40 grams of glucose."

Is that the point where the brain fog lifts?

Jenny said...


I don't experience brain fog when lowering carbs into the ketogenic range, so I can't speak to this issue.

Anonymous said...

Couldn't agree more. My typical pattern on the Healthy High Carb Low Fat diet was a BG spike followed by an insulin spike followed by a BG crash followed by a neuroendocrine dump.

Now I don't know what part of this was responsible for which symptoms, all I know is that the amount of protein I now eat keeps my BG level and my brain actually works best in the 70 - 90 range, maybe it prefers ketones. I am much more focussed, my ADD is much reduced, my depression is much reduced, and physically I have *more* energy.

The carbs I eat are principally to fine tune my BG as the control system is somewhat laggy, around 60 - 100 g/day carefully applied over time keeps me an order of magnitude more functional than the dietician's 250 - 300g and so far some of my diabetes symptoms have resolved and the rest have gotten no worse.

All of these findings are common to most but probably not all people following what are still largely considered to be dangerously low carb diets. Fortunately *some* medical professionals have noticed, there seem to be far more reports of sensible dietary advice today than even a couple of years ago.

Anonymous said...

Hi Jenny,

I have the 2003 version of Dr. Bernstein's book. It still said 37%.

I've read a few articles on the Ketogenic diet ... hugely fascinating topic especially as far as disease "cures" go. As restrictive "diets" go --- wow. diabetics have no right to complain. Even Dr. Bernstein's low carb requirements are nothing compared to that.

Again, thanks for your efforts. Trust me; I know what it takes to write one article.


Jenny said...


I have seen that 58% number cited in a few research studies. I'll nail it when I have a chance which is not now.

That ketogenic diet for epilepsy is incredibly intense. They do it for kids who can die if their seizures aren't controlled who don't respond to drugs. It works for enough of them to be worth doing.

Bernstein would seem like a pigout after that diet, just like my 100 - 110 g diet with insulin feels like a non-stop carb-fest after my 7 years of eating much, much lower.

Anonymous said...

Dear Jenny, just to say that Lyle's ketogenic book costs 50 dollars, not expensive if you consider its valuable information. It now has 10 years but its not outdated at all. Thanks for your great blog. I'm not diabetic but I'm certainly learning a lot from reading your excelent articles!

Anonymous said...

Dear Jenny,
I have a feed and read all your blogs.Thanks for that :)

From this one I understand that in your opinion the brain needs glucose and that the liver will provide it (?)

The brain of the human fetus is much larger (in proportion) and just runs fine on ketone bodies.

Here is a review from an expert : during starvation the adult human brain just runs fine on ketone bodies.

My two cents :
with a bit training and / or low calorie diet and /or fasting schemes the adult human brain does not need glucose.Not to mention all the other benefits of fasting like improving insulin resistance, shifting the hypothalamus set points for satiety to a bit healthier BMI.

Annu Rev Nutr. 2006;26:1-22.
Fuel metabolism in starvation.

This article, which is partly biographical and partly scientific, summarizes a life in academic medicine. It relates my progress from benchside to bedside and then to academic and research administration, and concludes with the teaching of human biology to college undergraduates. My experience as an intern (anno 1953) treating a youngster in diabetic ketoacidosis underscored our ignorance of the controls in human fuel metabolism. Circulating free fatty acids were then unknown, insulin could not be measured in biologic fluids, and beta-hydroxybutyric acid, which was difficult to measure, was considered by many a metabolic poison. The central role of insulin and the metabolism of free fatty acids, glycerol, glucose, lactate, and pyruvate, combined with indirect calorimetry, needed characterization in a near-steady state, namely prolonged starvation. This is the main topic of this chapter. Due to its use by brain, D-beta-hydroxybutyric acid not only has permitted man to survive prolonged starvation, but also may have therapeutic potential owing to its greater efficiency in providing cellular energy in ischemic states such as stroke, myocardial insufficiency, neonatal stress, genetic mitochondrial problems, and physical fatigue.

PMID: 16848698

Thanks for reading :)

Anonymous said...

".. while your brain needs glucose, you do not need to EAT glucose to provide your brain with the glucose that it needs. That is because your liver has the remarkable ability to transform the protein that you eat into glucose."

But why should we make the liver convert protein into glucose for this purpose? Why not get the needed glucose by eating the required amount of carbohydrates, which places less burden on the liver and produces less waste products to be removed from the body by the kidney? Does the body in anyway distinguish the glucose in blood by its origin, that is, if it is from the carbohydrate we ate or if it came from protein that the liver converted to glucose (rather inefficiently even if 58%)? The amount of insulin (if required) should be the same for both cases if the body does not distinguish glucose by its origin.


Jenny said...


The answer to your question is that some people with Type 2 diabetes do not have access to excellent medical care or have doctors who refuse to prescribe insulin, so they can't tolerate more than a very small amount of carbohydrate without going high.

I do better with a slightly higher carb intake myself, but I can only eat that way when I inject insulin, because I will spike with more than 10 grams eaten at once. For years I could not find a doctor who would prescribe me insulin because my fasting blood sugar was near normal. So I had no choice but to eat a very low carb diet.

Anonymous said...

G'day Jenny.

Some further research suport for you.

I just came across this, I think you'll find it interesting and relevant to this subject::
“Low-carb diets alter glucose formation by the liver.
A new study shows that a low-carbohydrate diet changes hepatic energy metabolism.”

After you read the article, you will find the fulltext pdf of the research paper here:
Alterations in Hepatic Glucose and Energy Metabolism as a Result of Calorie and Carbohydrate Restriction.

Cheers, Alan

Anonymous said...

It's the liver's job to convert stuff into other stuff.

I shall have to take some Clever Pills and read Alan's paper again but it seems to agree with my version of reality, my BG is much more level when I'm running on ketones and getting most of my glucose from protein.