Early Insulin Resistance Predicts Subsequent Risk of Metabolic Syndrome

Insulin resistance and metabolic syndrome are most often thought of in connection with middle age. However, an article in the March 2009 issue of the journal Metabolism shows that these conditions get their start early in life.

A biracial group of 475 girls was studied first at ages 9-10 years and then at ages 18-19 years. In the first stage of the study, the invesigators determined the body mass index (BMI) of each girl. They also determined a score called the HOMA-IR, or homeostatis model assessment of insulin resistance. Insulin resistance can be measured using an oral glucose tolerance test (OGTT), but this requires several hours and numerous blood draws. In the HOMA-IR, the fasting levels of insulin and glucose are determined, multiplied, and converted into a score that correlates well with the insulin resistance determined by more technically-demanding methods. The girls were divided in two ways: (1) according to whether their BMI put them in the bottom, middle or top third of the group with regard to to relative obesity (2) according to whether their HOMA-IR score put them in the bottom, middle or top third with regard to relative insulin resistance.

Nine years later the girls were assessed for five metabolic syndrome risk factors. These included:

• Triglycerides
• High-density lipoprotein cholesterol
• Systolic/diastolic blood pressure
• Waist circumference
• Blood glucose

Each risk factor measurement was converted to a z score, which is a way of expressing the relationship of a specific measurement to the population average and the standard deviation of that particular variable. The five z scores were added for each individual. If the total z score was negative, that meant that the individual had a lower risk of metabolic syndrome. (The more negative the total, the better in this instance.) If the total z score was positive, that meant that the individual had a higher risk of metabolic syndrome.

The results were not surprising. The girls who had the lowest BMI and the lowest insulin resistance at ages 9-10 were found to have the lowest risk of metabolic syndrome at ages 18-19. As initial BMI and insulin resistance scores rose, the subsequent scores for risk of metabolic syndrome also rose. The girls who had the highest BMI and the highest insulin resistance at ages 9-10 had the highest risk of metabolic syndrome at ages 18-19. It appears that childhood obesity and insulin resistance interact to produce a higher risk of metabolic syndrome in early adulthood.

Taking it one step farther, the significance of metabolic syndrome in childhood was underlined by a study recently published in the journal Pediatrics, Metabolic Syndrome in Childhood Predicts Adult Cardiovascular Disease 25 Years Later. In this study, the investigators found that the incidence of cardiovascular disease for their 31 patients with pediatric metabolic syndrome was 19.4%, while the incidence was 1.5% for the subjects without metabolic syndrome as children. If these findings are correct, children with metabolic syndrome are significantly more likely than their peers to experience cardiovascular disease twenty five years later as adults.

The obvious question becomes, what happens to these children if intervention takes place between their pre-teen years and their forties? And what would that intervention be? Would it be pharmacological? Would it involve a change in dietary habits? There is a bias at this blog for at least trying a low-carb eating strategy to lower BMI, reduce insulin resistance, avert childhood metabolic syndrome and prevent the progression to adult cardiovascular disease. Whether or not that approach will be tested in the scientific community remains to be seen.

Reversal of Type 2 Diabetes?

Unfortunately there is still no cure for type 2 diabetes. But a recent study published by Eric Westman's group at Duke University suggests that a diet with less than 20 grams of carbohydrate per day or a low-calorie/low-glycemic index diet could enable some type 2 diabetics to decrease their medication and lower their hemoglobin A1c levels.

Investigators randomized 84 obese type 2 diabetic subjects into a Low-Carb group or a Low-Calorie/Low-Glycemic Index (Low Calorie/Low-GI) group. (Full disclosure: the study was funded by the Robert C. Atkins Foundation.) For a period of 24 weeks the Low-Carb group was instructed to eat an Atkins induction-level diet (less than 20 grams of carbohydrate per day, without caloric restrictions). During the same time the Low Calorie/Low-GI group was instructed to eat a reduced-calorie diet that contained 55% low-glycemic index carbohydrates. In the Low-Calorie/Low-GI group, the target caloric intake was set for each individual to provide 500 calories less than that required for weight maintenance.

Fifty eight percent of the enrollees completed the study. Prior to the study the average food intake was 2128 calories a day, with 245 grams in the form of carbohydrates. During the study, the Low-Carb completers ate a daily average of 1550 calories with 49 grams (13%) of the total coming from carbohydrate. The Low-Calorie/Low-GI completers averaged 1335 calories per day with 149 grams (45%) coming from carbohydrate. While compliance with protocol design was not perfect, the Low-Carb group ate about 16% more calories and about two-thirds fewer carbs than the Low-Calorie/Low-GI group did.

Over the 24 weeks of the study, the Low-Carb group lost an average of 24.4 pounds and the Low-Calorie/Low-GI group lost an average of 15.2 pounds. At the same time, both groups saw significant improvements in fasting glucose, fasting insulin and hemoglobin A1c. It is commonly believed that for type 2 diabetics, weight loss causes a drop in hemoglobin A1c. Interestingly, in this group of subjects there was no correlation at all between the amount of weight loss and the level of hemoglobin A1c. (The corresponding figure can be found here.)

Of the 50 patients who were taking insulin at the start of the study, eight in the Low-Carb group and three in the Low-Calorie/Low-GI group were able to decrease or discontinue their insulin altogether. (Click here for a tabular summary.)

The authors go to some lengths to show that the Low-Carb diet was superior to the Low-Calorie/Low-GI diet in this study. In any case, it appears that type 2 diabetics who are able to adhere to either of these diets in the long term can see a significant improvement in their symptoms and a decrease in the need for injected insulin.

Health Care Sacrifices

Heart disease is increasing in the United States. According to the following graph, hospital discharges for coronary heart disease went from about 1.4 million per year in 1970 to about 2.1 million per year in 2000.

In spite of the increasing incidence of heart disease, the death rate from heart disease is decreasing. According to the figure below, from the New York Times, the one-year heart attack survival rate in Medicare patients went from about 58% in 1986 to just under 70% in 2000. The second graph suggests the reason for this: Medicaid hospital spending doubled during that time period.

Patient survival is good. Increasing Medicare costs is bad. Some people have proposed limits on medical spending as a way to cope with the increased costs. One of these is Tom Daschle, who was recently nominated to be head of the Health and Human Services Department. Mr. Daschle's sudden withdrawal from that nomination was unexpected, and he had already provided much advice about acting quickly to ensure that health care legislation could be enacted without the debate and delay that had caused the 1994 health care overhaul to be abandoned.

In 2008 Mr. Daschle wrote a book called Critical: What We Can Do About the Health-Care Crisis. In it he advocated many of the provisions that have found their way into the 2009 economic stimulus package. These include moving medical record keeping to an electronic format that will be available to a new federal bureaucracy, the National Coordinator of Health Information Technology. This organization has a mandate to reduce costs and "help guide medical decisions at the time and place of care."

Mr. Daschle also recommended the creation of another bureaucracy to be modeled the after United Kingdom's National Institute for Clinical Excellence (NICE). This entity is found in the stimulus package as the Federal Coordinating Council for Comparative Effectiveness Research. It is described in the draft report for the bill as being expected to apply a cost-effectiveness standard to approve or reject treatments using a formula that divides the approximate cost for a treatment by the years the patient is expected to benefit from that treatment. The Council will enforce the standard by evaluating doctors and hospitals based on whether or not they are "meaningful users" of the new system. Translation: as Americans get older, they should expect to receive less medical care.

It is hard to contemplate health care reform built on the principle of denial of care to the elderly. However, Mr. Daschle has said that health care reform will not be pain free. He believes that Americans should be more like Europeans in their ability to accept hopeless diagnoses, to do without experimental treatments and to refrain from expecting too much from the health care system. It may be necessary in such a system to provide health care benefits for younger citizens and then ask citizens to be willing to sacrifice those benefits in their later years.

One hopes that all of this has very little to do with low-carbing. It is still permissible to buy eggs, cheese, meat and fresh vegetables. It is still permissible to avoid refined carbohydrates. As far as the health care system, following a low-carb lifestyle will prevent or lessen the impact of many medical conditions including diabetes, heart disease, gallstones, and dental cavities. And being able to achieve excellent health without resorting to medication and hospitalization is a good thing.

If I'm so fat, why am I always hungry?

Body fat serves a number of purposes. It cushions our organs, it insulates us, and it provides a way to store energy. A normal amount of body fat is about 10-20% for men and 15-25% for women. So far, so good.

But sometimes we get too fat. What then? It seems logical to expect that our appetites would decrease, the extra fat would be burned as energy, and we would return to a normal fat percentage once again. But as millions of overweight people can testify, that is not how it always works. They can eat a meal, leave the table feeling full, and two hours later they will be ravenously hungry. They look down at their rolls of excess body fat and wonder, "Why in the world am I hungry? Why can't I just use some of this fat instead of having to eat again?"

In most cases, the answer is one word. Insulin.

Insulin is a hormone which is secreted by the pancreas whenever we eat carbohydrates or protein. Insulin sends a message to the tissues of the body--store nutrients! It is not practical for us to eat continuously. So we eat discrete meals, use some of that energy immediately, and (in response to the signal from insulin) store the rest for use later. Several hours after a meal is consumed, our insulin levels will normally fall, and this will permit nutrients to come back out of storage until it is time for our next meal.

As people get older, the insulin response system may begin to break down. The pancreas has to secrete more and more insulin in order to store nutrients following a meal. The elevated insulin takes longer and longer to return to normal levels, until it stays somewhat elevated all the time. As insulin levels stay high, it becomes progressively harder for stored nutrients to be released between meals. In other words, energy (mostly in the form of fat) is being stored at mealtime, but the energy in the fat can no longer be released efficiently between meals. The result? If a person has excess fat stores but also has persistently high insulin levels, he will be less and less able to access the energy he has stored, but will be forced to eat frequently to provide his body with the energy it needs. Even though he is fat, because of elevated insulin, he will find that he is always (or almost always) hungry.

Obviously the next question is, if a fat person wants to overcome constant hunger, how does he address his elevated insulin? This post, Reversing Insulin Resistance, shows that simply eating a low-carbohydrate diet significantly reduced insulin levels in a small group of obese type 2 diabetic patients. For a more detailed answer, please see this post. It summarizes the three primary strategies for lowering insulin (eat low-carb, eat moderate protein and wait 5-6 hours between meals) and gives several more suggestions for bringing insulin levels (and hunger signals) back into a normal range.

Fatty Liver--Not Just for Geese Anymore

Foie gras, French for "fat liver," is a delicacy produced by force-feeding ducks or geese with corn meal. However, ducks and geese aren't the only species that can get a fatty liver.

Nonalcoholic fatty liver disease (NAFLD) is a condition that can occur in humans. NAFLD, defined as a liver fat content greater than 5.5%, is found in about one third of the U.S. adult urban population. As shown in the figure below, NAFLD (also called hepatic steatosis) progresses from fat deposits that cause liver enlargement, to fibrosis and the formation of scar tissue, to cirrhosis and the actual destruction of liver cells. It occurs in 45% of adult Hispanics, 33% of adult whites, and 24% of adult blacks.



According to the Mayo Clinic website, as diabetes and obesity increase, the incidence of NAFLD is increasing in both adults and children. Unfortunately there is no standard medical treatment for nonalcoholic fatty liver disease. Although several possible treatments are under investigation, none of them has yet proven effective.

With all of that in mind, there is encouraging news from a pilot study performed at Duke University by Eric Westman's group. Westman and his colleagues studied five obese patients who had been diagnosed with fatty liver disease by liver biopsy. They were instructed to follow a low-carbohydrate diet (less than 20 grams of carbohydrate per day) for six months. At the end of that time the patients were biopsied again and they showed significant reductions in liver fat and liver inflammation.

A recent study in Hepatology by Browning et al. explains why a ketogenic diet might reverse fatty liver disease. The investigators divided a group of 14 weight-loss patients into two groups. For two weeks 7 of the patients followed a calorie-restricted diet and the other 7 patients followed a carbohydrate-restricted diet. The scientists used radioactive tracers and NMR spectroscopy to determine how each group was performing gluconeogenesis in their livers.

They learned that the carbohydrate-restricted group produced more of their glucose from lactic acid and amino acids than did the calorie-restricted group. Not only that, the carbohydrate-restricted group burned their liver fat to provide the energy required to perform gluconeogenesis, while the calorie-restricted group tended to use liver glycogen to fuel gluconeogenesis. The researchers also found that the low-carbohydrate group increased fat burning throughout their entire body.

While these studies do not offer conclusive proof that a low-carbohydrate diet should be used to counteract and possibly reverse nonalcoholic fatty liver disease, the findings are encouraging and will be followed up by Dr. Browning and his associates in their next study.

Sweetness and Light Dopamine

Just the taste of a sweet beverage can result in liking, wanting and sipping of more of the sweet beverage. (Please see the previous blogpost for details.) The liking-wanting-sipping phenomenon happens even when the sweet beverage is not actually consumed, and happens in less than a minute. What causes this reaction to the taste of sweet?

In 1989, L.H. Schneider observed that dopamine receptors in the brain are stimulated when rats are allowed to feed themselves sweet solutions. Since that time, many investigators have noticed a relationship between either the taste of sweet or the actual consumption of sweet and the response of dopamine receptors in the brain, both in rats and in humans.

Dopamine is a neurotransmitter found in the midbrain. It has many functions, but one of them is the ability to produce prolonged feelings of pleasure. Increased dopamine signaling is involved in the mechanism of addiction to cocaine, amphetamines, and nicotine. At a more moderate level, novel foods, sweet foods, and overeating also cause increases in brain dopamine.

In order to investigate possible rebound effects of overstimulation with dopamine followed by abrupt dopamine withdrawal, a group of rats was treated for five days with l-dopa (a substance which is converted to dopamine in the brain). The research is described in an article in the December 2008 issue of Nutrition & Metabolism. After treatment was completed, the rats in the previously-treated l-dopa group were compared with an untreated control group. Over the next 12 weeks both groups of rats were allowed to eat as much food as they desired. At the end of that time, the previously-treated rats had gained 15% more weight than the control group.

Why did this happen? The authors hypothesize that treatment with the dopamine precursor l-dopa caused overstimulation of the dopamine signaling system in the rats. This, in turn, caused downregulation of dopamine receptors and decreased endogenous dopamine production. When the l-dopa treatment ceased, the rats were left with few dopamine receptors and low endogenous dopamine production. To compensate for this, the rats used the mechanism of overeating to compensate for their relative dopaminergic deficiency.

Rats are not humans. Nevertheless, it is possible to suggest that eating or tasting sweet food causes an overstimulation of the dopamine signaling system. This produces a downregulation of dopamine signaling such that, in the absence of sweet, there is a noticeable decrease in energy, motivation and mood. Sweets are legal, cheap and easy to obtain. If the sweet-eater wishes to experience the pleasant feelings associated with dopamine overstimulation, it will be very easy to continue eating sweets. And if he is unable to wait several days to a week to allow the body to return to its normal level of dopamine production and receptor activity, it will be harder than he might have expected to eliminate sweet tastes from his diet.

Detecting, Liking and Wanting Sweetness

Those who read food labels realize that high fructose corn syrup is added to all sorts of products, from pickles to yogurt to spaghetti sauce. The Corn Refiners Association says this is because high fructose corn syrup enhances flavor and increases shelf life. A recent article in the journal Physiology and Behavior suggests that there may be another reason.

The article, called Modified sham feeding of sweet solutions in women with and without bulimia nervosa, was designed to show whether people who experience binge-eating episodes might overrespond to the stimulations of taste and smell. As it turns out, they do not, or at least they did not in this study. However, the study did produce an interesting outcome in terms of the way people respond to sweet tastes.

The study compared two groups of women--ten healthy women (termed NC, or Normal Control) and eleven women with Bulimia Nervosa (termed BN). The women were given solutions of cherry-flavored Kool-Aid sweetened with aspartame in concentrations of 0, 0.01, 0.03, 0.08 and 0.28%. (The 0.08% solution approximates the sweetness of commercial soda.) There were three trials in which the five solutions were prepared in five opaque containers, each fitted with a straw. The solutions were presented in a random order, using a one-minute access period during which the women could sip as much as they wanted of that particular solution, but they could not swallow it. They were instructed to spit out the solution into another opaque container. (The amount sipped and the amount spit out was later measured by the investigators.) The women were then asked to

1. Rate the sweetness of the solution

2. Rate how well they liked the solution

3. Rate how much they wanted more of the solution

Even though the solutions were presented in random order, both the Normal Control group and the Bulimia Nervosa group were able to accurately distinguish among the five levels of sweetness provided in the solutions. Again, although the solutions were presented randomly, both the Normal Control group and the Bulimia Nervosa group reported liking the solutions in direct proportion to how sweet the solutions were. Consistent with the self-reported preference rating, both groups sipped an increasing amount of the solution as the sweetness of the solution increased. (Remember, they were not allowed to swallow the solution, but they could sip as much of it as they wanted.) Finally, as shown in the graph below, both groups reported that they wanted more of the solution as the sweetness of the solution increased.
For both groups of women, more sweetness led to more liking, more sipping and more wanting. This was not a function of actually consuming the sweetened solutions, but simply of having the solutions in their mouths for a few seconds. Using this information, it not unreasonable to suggest that the increased use of another sweet substance, high fructose corn syrup, in all sorts of foods, may have the unintended result of producing more liking, more eating and more wanting of the products that contain it.

But you have to eat carbohydrates!

Perhaps you overdid it a bit at Christmas and are looking for a weight-loss plan. Although low-carb diets are currently out of fashion, they have one big advantage over low-calorie, low-fat diets. They work.

Ordinary people can lose weight and keep it off by following a low-carb eating plan. Not only that, they will notice an improvement in blood pressure, HDL cholesterol, triglycerides, and insulin responsiveness. So why don't more people attempt weight loss through the low-carb lifestyle?

One big reason is that people are told (incorrectly) that their body requires them to eat a certain amount of carbohydrate every day. The brain requires glucose to function, and several other tissues require carbohydrate-derived fuels as well. If a person doesn't eat any carbohydrates, where does this fuel come from?

It comes from proteins. After proteins are eaten, they are broken down into their building blocks, which are called amino acids. These amino acids can be reassembled into other proteins. Some of them can be converted into fat. And thanks to a metabolic process called gluconeogenesis, up to 400 grams of them per day can be converted into carbohydrates.

Since the brain requires only about 120 grams of carbohydrates a day, that's well within a safe range. It does take a few days for the enzymes of gluconeogenesis to be induced, permitting conversion of enough glycogenic amino acids into glucose to supply the tissues that require it. During the transitional period, typical symptoms may include body aches, nausea and headaches. These will subside after about three days, and energy levels and alertness will improve noticeably. To figure out how much protein to eat to permit adequate carbohydrate synthesis, Jenny Ruhl has provided this convenient Protein Need Calculator.

If you need to start a weight-loss plan, there are many good diets available, including many versions of low-carb diets. Check out your options and pick the diet that suits you the best. And if you choose the low-carb lifestyle, be assured that thanks to gluconeogenesis, you can safely drop most or even all of the carbohydrates from your menus!

Type 1.5 Diabetes

Type 1.5 diabetes, as the name implies, falls between type 1 and type 2 diabetes. It manifests some of the symptoms of both types, and it is important because it affects about 15% of those diagnosed with diabetes.

Type 1 diabetes is characterized by the presence of autoantibodies against insulin or against certain components of the insulin-producing system such as glutamic acid decarboxylase (GAD), tyrosine phosphatase or the islet cells themselves. These autoantibodies cause the patient's own immune system to kill the beta cells of the pancreas, making the patient unable to produce any endogenous insulin.

Type 2 diabetes is characterized by insulin resistance and diminished production of insulin by the pancreas. If a patient is well-managed, this kind of diabetes can be controlled for many years by diet, excercise and oral medication.

Type 1.5 diabetes has characteristics of type 1 and type 2. Like type 2, its onset is in adulthood, and the pancreas is able to produce insulin for several years. Like type 1, it often occurs in thin people, and it often involves autoantibodies to GAD or islet cells. Unlike type 1, in type 1.5 diabetes, the autoantibodies work much more more slowly. However, their destruction of beta cells is relentless, and within 5-10 years of diagnosis, patients with type 1.5 diabetes will require insulin.


Type 1.5 diabetes is sometimes called Mature Onset Diabetes of the Young (MODY), Latent Autoimmune Diabetes in Adults (LADA), slow onset type 1 diabetes, or double diabetes. As might be expected, each of these terms involves specific criteria and there is disagreement about who falls into which category. Leaving the questions about specific terminology aside, why should it matter that about 15% of the diabetic population is neither type 1 nor type 2?

It matters because type 1.5 diabetics are often misdiagnosed as type 2 diabetics. It matters because type 1.5 diabetics will initially respond to dietary modifications and oral medication, but because their condition stems from the death of beta cells rather than insulin resistance, eventually they will not. Physicians who are not familiar with type 1.5 diabetes may not understand why a misdiagnosed type 1.5 diabetic has stopped responding to standard treatments and may assume that the patient is no longer complying with their instructions. (This happens often enough that physicians have good reason to suspect noncompliance as an explanation for poor diabetic control.) If the patient knows that he or she is following the doctor's guidelines, it might be a good idea to ask for autoantibody tests to see if destruction of the pancreatic beta cells is taking place. As in every other aspect of health care, it becomes important for the patient to become an active participant in the monitoring and management of his condition.

Type 2 Diabetes

Type 2 diabetes is a condition affecting slightly more than 7% of Americans. It used to be called adult-onset diabetes, but we now know it can occur in children as well as in adults. Another term for it was non-insulin-dependent diabetes mellitus (NIDDM), but about 30% of type 2 diabetics are treated with insulin.

Type 1 diabetes begins with death of insulin-producing beta cells in the pancreas. Typically this happens over a period of weeks or months and is not reversible. Type 2 diabetes may begin with a diminished production of insulin by the pancreas. Several risk alleles have been identified and have been found to have an additive effect, resulting in decreased glucose sensitivity in the pancreas of people who have not yet developed overt diabetes. This condition may go undetected for years, only becoming evident when the patient starts to experience insulin resistance. Often this happens during pregnancy, when the mother's body becomes less responsive to insulin during the third trimester. Gestational diabetes is a temporary condition, but can be an early indicator of a predisposition to type 2 diabetes. More often, as a person ages and becomes overweight, his or her muscles, liver and pancreas gradually develop increasing resistance to the action of insulin, and the symptoms of hyperglycemia start to appear. Insulin resistance in the liver may lead to overproduction of glucose by the liver, causing even more hyperglycemia. Hyperglycemia, in turn, may initiate a process called apoptosis (programmed cell death) in the pancreas, resulting in a pancreas that is significantly smaller and less able to produce insulin than the pancreas of a person who does not suffer from type 2 diabetes.

Unlike type 1 diabetes, the progression to type 2 diabetes usually occurs over several years. It does not typically result in the death of all the beta cells of the pancreas, but the ability of the pancreas to regulate blood glucose is significantly compromised. The presence of a genetic component is even more significant than it is in type 1 diabetes. In fact, if a person has a relative with type 2 diabetes, chances are 80% that the person will develop type 2 diabetes in his or her lifetime. With that in mind, it is a good idea to monitor fasting and postprandial glucose levels to see if blood sugars are starting to trend outside the normal range. If symptoms are detected early and the condition is managed appropriately, the progression to fullblown type 2 diabetes can be slowed or perhaps prevented altogether.

If you would like to do some more reading on the progression to type 2 diabetes, here are some links:

Metabolic Syndrome

Reactive Hypoglycemia

Reversing Insulin Resistance

(For those who are interested, the picture at the top is a Texas snowman.)

Type 1 Diabetes

The pancreas is a multifunctional organ that sits below and behind the stomach. As an exocrine organ, it empties digestive enzymes into the gut at the level of the duodenum. The pancreas is also an endocrine organ, synthesizing insulin in its beta cells and glucagon in its alpha cells, and secreting those hormones into the blood.

Normally the pancreas performs its functions silently and efficiently. However in some cases the beta cells of the pancreas are vulnerable to an attack by the body's own immune system. For about three million Americans, the pancreatic beta cells are no longer functional, resulting in a condition called type 1 diabetes.

The cause of type 1 diabetes is not entirely clear. The peak age of diagnosis is 14 years, but type 1 diabetes can develop at any age. There is a genetic contribution--twenty percent of patients have a relative who also has the disease. The presence of other autoimmune disorders is a predisposing factor, and childhood viral infections such as rubella, cytomegalovirus and coxsackie B may trigger the condition. In any case, in these patients the body's immune system targets the beta cells of the pancreas and renders them unable to produce insulin.

Before Drs. Banting and Best discovered insulin in 1921, the diagnosis of type 1 diabetes was a death sentence. Victims had continuous thirst and voracious appetites, but without endogenous insulin to control their blood sugar, they wasted away because they could not properly utilize the food they ate. A very low carbohydrate diet could forestall the inevitable for several years, but most died before the age of 30.

With the advent of exogenous insulin therapy, a new set of problems arose for type 1 diabetics. Blood glucose could be controlled, but if it was poorly controlled, it would eventually result in eye, kidney and nerve diseases. If too much insulin was injected, it could result in severe hypoglycemia and even death. Dr. Richard K. Bernstein is a physician who developed type 1 diabetes in 1946 at the age of 12. He tells a fascinating story, showing how difficult it was in those days to match insulin dose to blood glucose level. The advent of the glucometer enabled patients to monitor their blood glucose much more accurately, but it also led to the temptation to eat large amounts of carbohydrates and then "cover" the carbs with injected insulin.


Today the American Diabetes Association recommends that diabetics eat 25-35% of calories from fat, 15-20% from protein and 45-55% from carbohydrates. By contrast, Dr. Bernstein proposes that eating a large number of calories as carbohydrates produces a large variability in blood glucose and a high level of difficulty in controlling blood sugar. To counteract this, he suggests something called the Laws of Small Numbers, which entails eating only small amounts of slow-acting carbohydrate, and no fast-acting carbohydrate at all.

Which approach is correct? No definitive scientific comparison is available, but this review notes that when the ACCORD study attempted tight glycemic control in type 2 diabetics through drug therapy, the study had to be terminated because of high mortality. By contrast, the reviewers cite a long list of references that indicate that dietary control of hyperglycemia is able to improve many of the long-term consequences of diabetes. Type 1 diabetics are unable to completely avoid the use of exogenous insulin, but the strategy of eating very small amounts of carbohydrate that require only small amounts of insulin, appears to be worth serious consideration.

Metformin

Metformin (brand name Glucophage) is a member of the class of antidiabetic drugs called biguanides. Unlike the sulfonylureas such as glipizide, metformin does not decrease blood glucose by increasing the plasma concentration of insulin. Instead it works in several other ways to accomplish its purpose.

Metformin exerts its main effect, suppression of gluconeogenesis, by inhibiting the ATP production of the mitochondrial respiratory chain. Mitochondria are little organelles inside most of the cells of our bodies. Their job? To convert the precursor molecule ADP (adenosine diphosphate), into ATP (adenosine triphosphate), a molecule that is used to provide the energy required for many metabolic processes. Our bodies produce a little ATP by breaking down glucose in a process called glycolysis. But most of our ATP is provided when two-carbon units enter the tricarboxylic acid (TCA) cycle and are burned in the mitochondrial respiratory chain to produce carbon dioxide and water. From that link, here is a pictorial representation of the mitochondrial respiratory chain.


When metformin interferes with the conversion of ADP to ATP, the ratio of ATP to ADP decreases. When this ratio decreases, there is a resultant decrease in the activity of pyruvate carboxylase, which is the first enzyme used in the process of gluconeogenesis. The inhibition of pyruvate carboxylase significantly decreases the amount of gluconeogenesis the liver can perform. As we have seen previously, when the liver becomes insulin resistant, it will raise blood glucose by continuing to do gluconeogenesis even when blood sugar levels are normal. Although metformin does nothing directly to reverse insulin resistance in the liver, it is able to use the complex series of events beginning with the inhibition of ATP production in mitochondria to partially block the synthesis of excess glucose by the liver.

(As an aside, the inhibition of gluconeogenesis may cause an increase of lactic acid in the blood, lactic acid being one of the building blocks used for gluconeogenesis. In extreme cases this can lead to lactic acidosis, but the phenomenon is relatively rare with metformin.)

The second major effect of metformin is that it is able to decrease blood glucose by improving glucose uptake in muscle cells. Glucose cannot pass into muscle cells simply by diffusion; it requires specfic transport proteins to carry it into the cell. Studies have shown that metformin increases the number of the glucose transporters GLUT1 and GLUT4 in the plasma membrane of muscle cells. More glucose transport proteins means more glucose can be moved into insulin-resistant muscle cells, which in turn lowers blood glucose.

Although metformin has several other actions that reduce blood glucose, these two are the major ones. Unlike injected insulin, or oral drugs that increase insulin secretion, metformin does not cause an increase in insulin resistance, nor does it cause weight gain. However, it is important to note that metformin does not reverse insulin resistance. It simply acts to lower blood glucose in a non-insulin dependent manner.

Alcohol and the Low-Carb Lifestyle

About six weeks ago, Woodswalker asked me to write a post addressing the role of alcohol in a low-carb diet. This is a huge topic, but I will present a few thoughts for consideration. If you have questions or comments, please remember that I am a biochemist, not a psychiatrist.

Alcohol, also known as ethanol, contains seven calories per gram. That's somewhat less than fat at nine calories per gram, and quite a bit more than carbs and protein at four calories per gram. Pure grain alcohol contains zero carbs. It is not an essential food. The metabolism of ethanol is fairly straightforward.

The first pathway happens mainly in the liver and is constitutive. ADH is the enzyme alcohol dehydrogenase. ALDH is the enzyme acetaldehyde dehydrogenase, and TCA stands for the tricarboxylic acid (Krebs) cycle. The second pathway is also found in the liver and is inducible--that is, it can be upregulated if the body is required to detoxify large amounts of alcohol on a consistent basis. MEOS stands for mitochondrial ethanol oxidizing system.

If a meal is consumed that contains alcohol, carbs, protein and fat, the calories from the alcohol will be processed first. This means that fat will not be used for energy until all the calories from the ingested alcohol have been burned. If a signficant number of calories of alcohol are ingested, this will postpone or even prevent fat burning. Drinking hard (i.e., distilled) liquor by itself does not affect insulin secretion, but when hard liquor is consumed with food, it increases insulin resistance and insulin secretion. Hard liquor also contains quite a few calories per ounce. By contrast, an ounce of mixed drinks, wine or beer will have fewer calories from ethanol. However, mixed drinks, wine and beer all contain carbohydrates, and, if they are consumed in quantity, will result in insulin secretion and eventual weight gain.

According to Dr. Michael Eades (see the comment at 31 October 2008, 21:34), a single glass of dry wine per day can improve insulin sensitivity and can assist with weight loss. For those who can stop at one glass of wine, that's great. But remember that alcohol is a psychoactive drug, and as such, it lowers inhibitions. In the low-carb context, it is important to note that alcohol can lower inhibitions against consuming carbs, and inhibitions against consuming a second glass of wine as well.

Alcohol stops gluconeogenesis. Gluconeogenesis is the process used by the liver to keep blood glucose levels within normal limits. If a person consumes lots of carbohydrates, an alcohol-induced cessation of gluconeogenesis will probably not even be noticed. However, if a person consumes alcohol while doing very low-carb, he is likely to experience a fall in blood sugar followed by a compensatory release of adrenaline. This can lead to heart palpitations which will be relieved by drinking orange juice or eating a high-glycemic food. Unfortunately, this regimen is not conducive to longterm success on a low-carb diet. If a low-carber notices that alcohol consumption is followed by the symptoms of low blood sugar, it may be necessary for him to drink less than a full serving to minimize the undesirable side effects.

More on Insulin Control

The previous post discussed a three-legged stool approach to dealing with reactive hypoglycemia. The three legs of the stool are critical for lowering blood insulin and restoring insulin responsiveness. They are:

-Eating low-carb
-Eating moderate protein
-Waiting 5-6 hours between meals

Overall, these three appear to be the most important strategies for lowering blood insulin and restoring insulin responsiveness. However, a scan of low-carb websites suggests some additional ideas for improving insulin control.





-Avoid "sweet"
For 18 days Jimmy Moore did a "Sweet"-Free Challenge. He avoided all artificial sweeteners, including those in diet soda. The taste of sweet, even if it comes in a zero-calorie product, can be enough to trigger an insulin release from the pancreas.

-Be careful with alcohol
Dr. Mike Eades discusses alcohol consumption in the comments section of a recent post at his blog. In response to a commenter, Lowcarb convert, Dr. Mike says Studies have shown that a glass of wine per day helps with weight loss, but if you can’t stop with just one - and I’m one of those who has difficulty in doing so - cold turkey may be the better strategy. In response to that, another poster, Tom, says For me, wine is a gateway drug…to carbs! Alcohol lowers inhibitions, including inhibitions against eating carbs, and eating carbs leads to the release of insulin. A word to the wise is sufficient.

-When you eat carbs, make them low-glycemic carbs
Dr. William Davis discusses Quieting the insulin storm in a recent post at his blog. He points out that some foods, like wheat and cornstarch, have a higher glycemic index than table sugar. The higher the glycemic index, the more rapidly blood sugar will rise, and the more insulin will be released by the pancreas in response.

-Eat healthy fats at every meal
Healthy fats make up the caloric difference between an individual's daily caloric need and the calories provided by low carbs plus moderate protein. Fats provide energy, promote satiety and can be consumed with no insulin required whatsoever.

-Avoid eating a large volume of food at one sitting
In his book The Diabetes Solution, Dr. Richard Bernstein discusses the fact that simply overstretching the stomach causes the release of insulin. This effect happens without reference to what type of food is consumed. It occurs whenever the stomach has been distended--by overeating or by eating large servings of high fiber foods. (According to Dr. Bernstein, it even happens when the stomach is distended with air.) To avoid oversecreting insulin, it is preferable to avoid eating one large meal and two small ones, but instead keep all three meals at a similar volume of food.

Reactive Hypoglycemia--An Experiment?

Reactive hypoglycemia is a condition which is characterized by unusually low blood sugar that occurs one to four hours following a meal. The symptoms are the typical ones for low blood sugar--shakiness, light-headedness, weakness, confusion, anxiety, depression, hunger, pounding heartbeat and sweating.

Progressive development of insulin resistance is often the cause of reactive hypoglycemia. When the pancreas becomes insulin resistant, it is unable to release the proper amount of insulin in response to the stimulus of carbohydrates and proteins. Sometimes the pancreas will overshoot its estimate of the amount of insulin needed to store ingested carbs and proteins. The excess insulin produces hypoglycemia and its associated symptoms. A more detailed explanation of the process can be found in my original post called Reactive Hypoglycemia. (Be sure to read the comments section.)

Reactive hypoglycemia can be diagnosed with a glucose tolerance test. If the test is positive, the patient will typically be advised to eat every 2-3 hours to relieve the symptoms. Although freqent ingestion of food does keep blood sugar from falling too low, it will not provide a long-term resolution of the underlying problem.

One of my readers, Alex, who blogs at Low Carb New England, entered the discussion on the original post with the story of how he has been dealing with reactive hypoglycemia since he was a teenager. Over the years he has systematically tried many different approaches and has taken careful note of what result each personal experiment has produced. To summarize briefly, Alex initially tried eating less sugar and eating frequent meals, but eventually he gained nearly 100 pounds. Next he investigated low-carb eating. By using the Atkins diet, he lost weight and many of his symptoms improved considerably. In an effort to reduce the remaining symptoms, Alex tried eliminating artificial sweeteners and caffeine, and this helped somewhat. He also tried extremely low-carb and even no-carb eating, which didn't help.

Eventually Alex realized that he needed to limit his protein intake to the amount recommended by Drs. Mike and Mary Dan Eades in The Protein Power Lifeplan. (Remember, eating protein also causes insulin release, and thus can contribute to insulin resistance.) Again, his symptoms improved, but were not entirely gone. The final piece of the puzzle seemed to arrive when he read my first comment under the Reactive Hypoglycemia post and decided to try waiting 5-6 hours between meals to allow his insulin levels to come back to baseline and give his body a chance to re-establish a normal level of insulin sensitivity.

After a month of using this three-legged stool approach (low-carb/moderate-protein/5-6 hours between meals) to dealing with reactive hypoglycemia, Alex has finally experienced relief from the symptoms of reactive hypoglycemia. He gives a much more complete version of the story on his blog in a post called "I'm back!" (For those who don't have access to The Protein Power Lifeplan, another method of calculating one's daily protein need can be found here.)

And now I've reached the main point of this post. If the three-legged stool approach (illustrated in the picture above) has worked for Alex, would it work for anybody else out there who has reactive hypoglycemia? Each leg of the stool is designed to reduce insulin resistance and, one hopes, to restore some degree of insulin sensitivity in muscle, liver, brain and pancreas. If any of my readers is interested in trying to follow this plan for a few weeks or a month, I would be very interested in getting your feedback. If this method actually works, it's possible that a series of anecdotal experiences could convince a low-carb researcher to design a study to see if using the three legs of the stool is an improvement over frequent feeding as a way to treat reactive hypoglycemia. If these informal personal experiments don't work, that's also important information. It's possible that there are other pieces of the puzzle that aren't obvious, or perhaps that the mechanisms of reactive hypoglycemia are different from one individual to the next. If you decide to try this, please be very careful, and please don't do anything that would put your health in danger. That said, if you try it and you have observations you would like to share, please put them into the comments and we shall see where this might lead.

Does Exercise Produce Weight Loss?

Common wisdom suggests that exercising will cause a person to lose weight. Superficially this makes sense. A 150 pound person at rest will use about 60 calories an hour. If this person jogs at 5 mph for an hour, he or she will use an additional 540 calories per hour. Because a pound of fat represents 3500 calories, a faithful jogger should lose a pound every 6.5 days. However, as exercisers can attest, this does not seem to work out in the real world. Why would that be?

1. Vigorous exercise can produce physical stress. Stress in turn causes the release of cortisol, which stimulates carbohydrate synthesis (gluconeogenesis) for quick energy. Gluconeogenesis produces an elevation in blood glucose which then stimulates insulin release. If this sequence happens repeatedly during days and months of an ongoing exercise program, it becomes more and more likely that the chronically physically-stressed person will start gaining weight.

2. Vigorous exercise can cause fatigue. The person who exercises may be expending more calories during his workout, but if he becomes exhausted by his efforts, he may compensate by conserving energy (being more sedentary or even napping) during his other daily activities.

3. Exercise in the form of resistance training may cause the exerciser to overestimate how much energy his body consumes post-exercise. A 2006 article by Ralph La Forge states that, for the non-athlete, the excess post-workout oxygen consumption is less than 100 calories per day.

4. Vigorous exercise may cause the body's homeostatis mechanisms for fat storage to overcompensate. Exercise activates the enzyme lipoprotein lipase (LPL) in muscle tissue, allowing muscles to take up fatty acids as fuel. Once the exercise stops, the activity of LPL in muscle decreases and the activity of LPL in fat tissue increases. Calories will be pulled into fat cells and stored there to prepare for the next round of exercise. Although an individual's appetite might be depressed immediately after a workout session, later in the day there may be a more-than-compensatory drive to eat to replace lost fat stores.

5. Exercise plus frequent meals can cause weight gain. Eating frequently prevents both leptin levels and insulin levels from returning to baseline. As earlier posts have discussed, persistently elevated leptin levels can hinder satiety signals and cause excess consumption of calories. Elevated insulin levels will produce storage of those excess calories as muscle and as fat. Underweight bodybuilders use exercise plus frequent meals as a method to gain weight. However, without careful monitoring, overweight body builders can also gain weight on this regimen.


Exercise is a good thing. It can strengthen the heart and lungs, elevate mood, create a better physique and improve stamina. But for a number of very good reasons, exercise by itself does not necessarily produce weight loss, and if the circumstances are right, it may even result in weight gain.

Transgenerational Obesity

Americans are getting fatter. According to a recent article in Obesity, by 2030, 86.3% of American adults will be overweight. The average adult BMI (Body Mass Index) is now 28, which is in the overweight range. By 2030 the average adult BMI is predicted to be 31.4, which is well into the obese range.

One factor that contributes to this phenomenon is the fact that fat mothers produce fatter offspring. In some senses this is not surprising. If there are genes that predispose to obesity, those will be passed down from parents to children. If there are lifestyle choices that contribute to an increased BMI, children will learn those by example from their parents.

What is not expected is the presence of a multiplier effect in generational obesity. In both rats and mice that are susceptible to obesity, a fat rodent mother gives birth to offspring which will become fatter than she was, and they, in turn, give birth to pups which grow up to be fatter than they were. The same phenomenon happens in humans. It is harder to observe in humans because it takes decades to progress from mother to daughter to granddaughter, while rodents can easily produce several generations in a few months or years. In rodents it is also much easier to control for genetic and environmental factors.

In both rodents and humans, the cause of the multiplier effect has not been established. It could result from a relatively high blood sugar in the mother causing the fetus to produce extra insulin-secreting cells in the developing pancreas. This could lead to increased insulin resistance and subsequent obesity as the child matures into an adult. It is possible that the high levels of leptin in an obese mother could cause her fetus to become leptin resistant. Methylation studies described in the rat reference above, suggest that maternal obesity may produce long-term modifications in the regulatory regions of obesity-related genes in her offspring. These modifications would be epigenetic, not mutational modifications. In other words, the actual DNA coding sequence is not changed, but while the fetus is in the womb, the 3-dimensional conformation of its DNA is modified, causing obesity-related genes to be more or less easily expressed even after the baby is born.

There is also evidence that the cycle of increasing transgenerational obesity can be broken. In 2006 an article in Pediatrics described a group of 113 obese mothers who had undergone biliopanceatic diversion (BPD) surgery for weight loss. This group of mothers had 45 childen before the surgery and 172 after the surgery. All were followed for 2-18 years. Comparing the 172 children born after BPD surgery with the 45 born before it, the prevalence of obesity decreased by 52% and severe obesity decreased by 45%. The effect was gender-specific, with the prevalence of overweight in the daughters decreasing from 56% to 42% and in the sons decreasing from 50% to 25%.

Apparently obesity does not have consequences just for the obese mother, but its effects extend into the lives of her children as well.

I'm Back!

Thanks to all the people who left excellent comments on my archived posts. Each one of your observations has been very much appreciated. As I've realized during this month of real-life challenges, the most important information about the science of low-carbing can already be found on the previous posts in this blog.

Woodswalker tactfully pointed out that the subject of low-carbing may not be inexhaustible. She's right. There are lots of interesting aspects of low-carb that we will address in the future, but the basics involve a few well-known principles of biochemistry and physiology. Medical students have learned about these for years, but once they graduate into the world of practicing medicine, they seem to absorb the dogma about low-fat/low-cholesterol/low-calories and forget about their original training.

So, a word to the wise. Go back often. Review frequently. Remind yourself of the many scientific reasons we have for low-carbing. Most of the mainstream medical and nutritional community hasn't caught up to us yet. Someday it will. In the meantime, let's do all we can to keep ourselves on the path to good health that low-carbing provides.

Eat Fat to Lose Fat

Many aspects of the low-carb lifestyle are surprising. For example, the successful low-carber soon learns that he or she must eat fat to lose fat. Why would that be?

A possible explanation comes from a couple of studies published last summer in Cell Metabolism, one by Inagaki et al and one by Badman et al. For those who are interested in the specifics, see the PDFs here and here. These studies were performed in mice but were quite exhaustive and appear to have application to humans as well.

The requirement for eating fat to lose fat begins with a cellular receptor called peroxisome proliferator-activated receptor-alpha or PPAR-alpha for short. PPAR-alpha is a protein found inside liver cells. When dietary fat diffuses into the liver cell as fatty acids, the fatty acids are able to bind to PPAR-alpha and activate it. Activated PPAR-alpha then binds to another protein called the retinoid X receptor or RXR, and these dimerized proteins in turn are able to bind to the cell's DNA. In so doing, they enhance the production of a third protein--fibroblast growth factor-21 or FGF-21.

FGF-21 is secreted by the liver and produces several effects. In white adipose tissue, it stimulates lipid breakdown. The breakdown of stored lipids allows them to be used as fuel. In the liver, FGF-21 upregulates ketone body production. Ketone bodies provide another source of fuel. The two studies showed that production of FGF-21 was greatly enhanced when the mice were fed a low-carb/high-fat (ketogenic) diet. When few carbohydrates are provided in the diet, but the diet does contain fat, mice are able to switch to an efficient mode that allows them to consume stored fat for energy. Mice are not people, and the usual admonition applies--more research is required. But the observation that a ketogenic (low-carb/high-fat) diet allows mice to produce lots of FGF-21, mobilize fat stores and upregulate ketone production suggests an explanation for what low-carbers know by experience--you have to eat fat to lose fat.

Calories Count

One of the great things about low-carbing is that (at the beginning anyway) low-carbers don't need to count calories.

Low-carbers do have to learn what a normal portion size is--a portion of macadamia nuts is 1/4 cup, not half a bag. A portion of cucumbers is 1/2 cup, not a whole cucumber. Low-carbers also need to learn that it's okay to subtract fiber carbs from their carbohydrate count. Once that's accomplished, it becomes a simple matter to look up various foods, figure out the carb content and add up the number of carbs consumed in a day. The target number is normally in the double digits, which is a fairly easy calculation for those of us who are arithmetically challenged.

Since low-carbers count grams of carbohydrate, does that mean that for low-carbers calories don't count? No. Calories do count.

Typically in a low-calorie versus a low-carb scientific study, the low-calorie group is given a target number of daily calories while the low-carb group is given a target number of daily carbs. When the results are tabulated, the net caloric intake will be compared between the two groups. Rather surprisingly, the two groups will have ingested almost the same number of calories. Examples are the recent study published in the New England Journal of Medicine and the A to Z Weight Loss Study published last year in JAMA. See Table 2 in each link for comparisons of daily energy intake from group to group.

Why do low-carbers unconsciously limit calories when they count carbs? One reason is the action of the signaling hormone leptin, discussed in the previous two posts. As low-carbers become more sensitive to the signals provided by leptin, they have an improved ability to perceive satiety. Their brains detect the leptin released by their fat stores and turn off the hunger signal at a caloric level that will allow them to use some of their fat stores for energy. The study group that eats a low-calorie diet without carbohydrate restriction will have a harder time getting the satiety signal. The participants in that group will have to turn off their eating at an intellectual level. When they have eaten the allowed number of calories, they have to consciously make themselves stop eating.

In comparison studies of weight loss, both the carb counters and the calorie counters end up eating approximately the same number of calories. Both groups lose weight in approximate proportion to their decrease in energy consumption. One of the considerations in choosing a weight-loss diet is the ease of complying with the diet. A controlled carbohydrate diet severely restricts the consumption of carbohydrate-containing foods but allows the dieter to eat to satiety. By contrast, a controlled calorie diet allows the dieter to eat balanced portions of whatever he or she wants, but requires the dieter to stop eating even if satiety has not been reached. As always, it's a tradeoff. The dieter decides which parameters are most important to him or her and chooses the diet that best fits those needs.