IN THE MID 1990s, as popular sentiment began to turn against the poor, unloved carbohydrate, a backlash originated within the medical community. “Carbohydrate-reduced diets are nutritionally unbalanced,” they sputtered.
That sure sounded good. There are only three macronutrients, after all: protein, fat and carbohydrate. Severe restriction of any one of these runs the risk of an “unbalanced” diet. Of course, nutritional authorities had no similar compunction about severely restricting dietary fat. But that’s beside the point. Certainly any such diet is unbalanced. The more important concern is whether such diets are unhealthy.
So for the sake of argument, let’s say that carb-reduced diets are unbalanced. Does that imply that the nutrients contained within carbohydrates are essential for human health?
Certain nutrients are considered essential in our diet because our bodies cannot synthesize them. Without dietary sources of these nutrients, we suffer ill health. There are essential fatty acids, such as the omega 3 and omega 6 fats, and essential amino acids, such as phenylalanine, valine and threonine. But there are no essential carbohydrates and no essential sugars. Those are not required for survival.
Carbohydrates are just long chains of sugars. There is nothing intrinsically nutritious about them. Low-carbohydrate diets that focus on removing refined grains and sugars should be inherently healthier. Perhaps unbalanced, but not unhealthy.
Another criticism leveled at the low-carb diets is that much of the initial weight loss that dieters experience is water—which is true. High carbohydrate intake increases insulin, and insulin stimulates the kidney to reabsorb water. Lowering insulin therefore causes excretion of the excess water. But why is this bad? Who wants swollen ankles?
By the late 1990s, as the “new” low-carbohydrate approach fused with the prevailing low-fat religion, the Atkins diet v2.0 was born—a low-carb, low- fat and high-protein approach. Where the original Atkins diet was high in fat, this new bastard diet was high in protein. Most high-protein foods also tend to high fat too. But this new approach called for lots of boneless, skinless chicken breasts and egg-white omelets. Once you tired of that, there were protein bars and shakes. A high-protein diet made many worry about potential kidney damage.
High-protein diets are not recommended for those with chronic kidney disease, since the ability to deal with the breakdown products of proteins is impaired. However, in people with normal kidney function, there are no concerns. Several recent studies have concluded that a high-protein diet was not associated with any noticeable harmful effects on kidney function.1 The concerns about kidney damage were overblown.
The biggest problem with high-protein diets was that they didn’t really work for weight loss. But why not? The reasoning seems solid. Insulin causes weight gain. Reducing refined carbohydrates lowers blood sugar and insulin. But all foods cause insulin secretion. The Atkins v2.0 approach assumed that dietary proteins do not raise insulin since they do not raise blood sugars. This notion was incorrect.
The insulin response to specific foods can be measured and ranked. The glycemic index measures the rise in blood sugar in response to a standard portion of food. The insulin index, created by Susanne Holt in 1997, measures the rise in insulin in response to a standard portion of food, and it turns out to be quite different from the glycemic index.2 Not surprisingly, refined carbohydrates cause a surge in insulin levels. What was astounding was that dietary proteins could cause a similar surge. The glycemic index does not consider protein or fats at all because they do not raise glucose, and that approach essentially ignores the fattening effects of two out of the three major macronutrients. Insulin can increase independently of blood sugar.
With carbohydrates, there is a very tight correlation between blood glucose and insulin levels. But overall, blood glucose was responsible for only 23 percent of the variability in the insulin response. The vast majority of the insulin response (77 percent) has nothing to do with blood sugars. Insulin, not glucose, drives weight gain, and that changes everything.
This point is precisely where glycemic index diets failed. They targeted the glucose response with the assumption that insulin mirrored glucose. But this is not the case. You could lower the glucose response, but you didn’t necessarily lower the insulin response. In the end, the insulin response is what matters.
What factors (other than glucose) determine the insulin response?
Consider the incretin effect and cephalic phase.
THE INCRETIN EFFECT AND THE CEPHALIC PHASE
BLOOD SUGAR IS often assumed to be the only stimulus for insulin secretion. But we’ve long known this was false. In 1966, studies showed that oral or intravenous administration of the amino acid leucine causes insulin secretion.3 This inconvenient fact was promptly forgotten until it was rediscovered decades later.4
In 1986, Dr. Michael Nauck noticed something very unusual.5 A subject’s blood sugar response is identical whether a dose of glucose is given by mouth or intravenously. But, despite the same level of blood sugar, the subject’s insulin levels differ greatly. Remarkably, the insulin response to oral glucose was much more powerful.
Oral administration almost never has a stronger effect than intravenous. Intravenous infusions have 100 percent bioavailability, meaning that all of the infusion is delivered directly into the blood. When given by mouth, many medicines are incompletely absorbed or partially deactivated by the liver before reaching the bloodstream. For this reason, intravenous delivery tends to be much more effective.
However, in this situation, the opposite was true. Oral glucose was far, far better at stimulating insulin. Furthermore, this mechanism had nothing to do with the blood sugar level. This phenomenon had not previously been described. Intensive research efforts revealed that the stomach itself produces hormones called incretins that increase insulin secretion. Since the intravenous glucose bypasses the stomach, there is no incretin effect. The incretin effect may account for 50 percent to 70 percent of the insulin secretion after oral glucose intake.
Rather than simply being a mechanism for food absorption and excretion, the gastrointestinal tract, with its nerve cells, receptors and hormones, functions almost like a “second brain.” The two human incretin hormones described so far are glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). Both hormones are deactivated by the hormone dipeptidyl peptidase-4. The incretins are secreted by the stomach and small intestine in response to food. Both GLP-1 and GIP increase insulin release by the pancreas. Fats, amino acids and glucose all stimulate incretin release and thus, increase insulin levels. Even non-nutritive sweeteners, which have no calories at all, can stimulate the insulin response. Sucralose in humans, for example, raises the insulin level 22 percent higher.6
The incretin effect starts within minutes of ingestion of nutrients into the stomach and peaks at roughly sixty minutes. The incretins have other important effects as well. They delay emptying of stomach contents into the small intestine, which slows down glucose absorption.
The cephalic phase is another pathway of insulin secretion independent of glucose. The body anticipates food as soon as it goes in your mouth and long before nutrients hit the stomach. For example, swishing a sucrose or saccharin solution around your mouth and spitting it out will increase your insulin level.7 While the importance of the cephalic phase is unknown, it highlights the significant fact that there are multiple glucose-independent pathways of insulin release.
The discovery of these new pathways was electrifying. The incretin effect explains how fatty acids and amino acids also play a role in stimulating insulin. All foods, not just carbohydrates, stimulate insulin. Thus, all foods can cause weight gain. And hence we get major confusion with calories.
High-protein foods can cause weight gain—not due to their caloric content, but rather to their insulin-stimulating effects. If carbohydrates are not the only or even the major stimulus to insulin, then restricting carbohydrates may not always be as beneficial as we believed. Substituting insulin-stimulating proteins for insulin-stimulating carbohydrates produces no net benefit.
Dietary fat, though, tends to have the weakest insulin-stimulating effect.
DAIRY, MEAT AND THE INSULIN INDEX
PROTEINS DIFFER GREATLY in their capacity to stimulate insulin,8 with dairy products in particular being potent stimuli.9 Dairy also shows the largest discrepancy between the blood glucose and insulin effect. It scores extremely low on the glycemic index (15 to 30), but very high on the insulin index (90 to 98). Milk does contain sugars, predominantly in the form of lactose. However, when tested, pure lactose has minimal effect on either the glycemic or insulin indexes.
Milk contains two main types of dairy protein: casein (80 percent) and whey (20 percent). Cheese contains mostly casein. Whey is the byproduct left over from the curds in cheese making. Bodybuilders frequently use whey protein supplements because it is high in branched-chain amino acids, felt to be important in muscle formation. Dairy protein, particularly whey, is responsible for raising insulin levels even higher than whole-wheat bread, due largely to the incretin effect.10 Whey protein supplementation increased GLP-1 by 298 percent.11
The insulin index shows great variability, but nevertheless, there are some general patterns. Increasing carbohydrate consumption leads to increased insulin secretion. This relationship forms the basis of many low- carbohydrate and glycemic index diets, and also explains the well-known propensity of starchy and sugary foods to cause obesity.
Fatty foods can also stimulate insulin, but pure fats, such as olive oil do not stimulate insulin or glucose. However, few foods are eaten as pure fat. It may be that the protein component of fatty foods drives the insulin response. It