by Joseph Brasco, MD


[ Page 1 | Page 2 | References ]

Unfortunately, the debate over the validity of this concept has primarily been waged in the media and lay publications and not in the scientific journals. Many of the popular books which support this position are gimmicky, and often, lack adequate scientific referencing. Yet, at their core is very important concept -- limiting the intake of carbohydrates, (especially as cereal grains and starches), will improve human health.

Some critics claim that reduced carbohydrate diets are a fashion trend. Well, this so called trend actually dates back some time. Anthropological study of early hominids has concluded that they lived as hunters-gathers. While nuts, seeds, vegetation and fruit made up an important part of the hunter- gather's diet, his mainstay was hunted or scavenged animal prey.

More recent evaluations of early man's nutritional patterns by Dr. Loren Cordain, estimate that as much as 65 percent of his calories were derived from animal products. Granted, early man was not eating corn fed Angus beef from Jewel, but he was eating the meat, the organs and the bones of his prey. Essentially, a high protein/fat diet. It was a mere 10,000 years ago that man began exploiting an agricultural niche.

This transition was made due to decreasing population of large game prey and an increasing population of humans. While undeniable good has transcended this dietary shift, i.e., growth of the human population, establishment of permanent settlements, the inception of civilization itself - man's health may have suffered in the transition. Generally, in most parts of the world, whenever cereal-based diets were first adopted as a staple food replacing the primarily animal-based diets of hunter-gatherers, there was a characteristic reduction in stature, a reduction in life span, an increase in infant mortality, an increased incidence of infectious disease, an increase in diseases of nutritional deficiencies (i.e., iron deficiency, pellagra), and an increase in the number of dental caries and enamel defects.

In a review of 51 references examining human populations from around the earth and from differing chronologies, as they transitioned from hunter-gathers to farmers, one investigator concluded that there was an overall decline in both the quality and quantity of life. There is now substantial empirical and clinical evidence to indicate that many of these deleterious changes are directly related to the predominately cereal-based diets of these early farmers. Since 99.99% of our genes were formed before the development of agriculture, from a biological perspective, we are still hunter-gathers.

Thus, our diet should reflect the sensibilities of this nutritional niche: lean meats; fish; seafood; low glycemic vegetables and fruit, (modern agriculture has significantly increased the sugar and starch content of vegetables and fruits over their Paleolithic counterparts), nuts and seeds - the evolutionary diet.

Glycemic Index

The term glycemic index, (GI) (a qualitative indicator of carbohydrate's ability to raise blood glucose levels), has seen a lot of mileage among the many non-ketogenic low carbohydrate diets. Most of these diets attribute the rise in obesity to the over consumption of high glycemic carbohydrates, and the subsequent over production of insulin. While this may be an oversimplification, there is growing evidence to support a relationship between GI and non-insulin dependent diabetes (NIDDM), and obesity. In a prospective study of 65,000 US women, researchers were able to demonstrate that the dietary GI was positively associated with the risk of NIDDM. The authors concluded that diets with a high GI increase insulin demand and thus cause hyperinsulinemia among patients with NIDDM, as well as in normal subjects. If chronic, this hyperinsulinemia can increase the risk for, as well as exacerbate NIDDM.

The issue of carbohydrates and insulin has more recently been addressed in a review article by Grundy. Grundy states that because secretion by pancreatic beta-cells is glucose sensitive, a high intake of carbohydrates has been reported to produce higher post prandial insulin levels. Moreover, it is possible that repeated stimulation of a high insulin output by high-carbohydrate diets could hasten an age-related decline in insulin secretion and lead to an earlier onset of NIDDM.

However, chronic hyperinsulinemia is not only associated with NIDDM, but is also related to a host of other medical conditions jointly known as Syndrome X. The constellation of disorders comprising Syndrome X include hypertriglyceridemia, increased LDL cholesterol, decreased HDL cholesterol, hypertension, hyperuricemia and obesity. If high GI carbohydrates in fact contribute to chronic hyperinsulinemia as multiple studies suggest, they are likely to be causative of these other conditions as well. In addition to their role in hyperinsulinemia, studies have also linked high GI foods with overeating.

One study found an inverse relationship between satiety and both glycemic and insulin index. In another study,it was found that voluntary energy intake after a high GI meal was 53% greater than after a medium GI meal and was 81% greater than after the low GI meal. The authors concluded that a high GI meal promotes excessive food intake in obese subjects. The literature clearly points to a role of high GI carbohydrates in the development of insulin resistance and its subsequent disorders.

However, GI is obviously not the whole story. One researcher examined the insulin demand generated by isoenergetic portions of common foods. While some of the results were predictable, i.e., the fact that glucose and insulin sources were highly correlated, some were unexpected, i.e., some protein-based foods induced as much insulin secretion as did some carbohydrate rich foods. At first glance, these results seem confounding. However, if one looks at the broader function of insulin, they are consistent.

Insulin is not just responsible for glucose disposal, but for storage and uptake of multiple nutrients. Whether these other nutrients can result in a chronic hyperinsulinemic state, as seen with high GI diets, is not known; it is unlikely due to their compensatory effect on glucagon. The other major difference between the insulin response of other nutrients versus carbohydrate is their effect on blood glucose.

While protein and fat stimulate insulin response, their effect on glucose is minimal. This lack of effect on blood sugar is more than trivial difference. It actually may be the glycosylation of end organs (especially the pancreatic beta-cells) that ultimately leads to NIDDM and its associated conditions. Thus, while a hyperinsulinemic state is not desirable for human health under any circumstance, the combination of hyperinsulinemia with impaired glucose homeostasis is likely to prove even more deliterious.

While the current literature would support limiting the consumption of high GI foods, GI certainly does not provide the final answer. If one was to follow this concept literally (as some popular books suggest) one could argue that potato chips at a GI of 50-59% were more beneficial than carrots at a GIU of 90-99%.

A better way of looking at carbohydrates is to return to the principles of the "evolutionary diet." Robert Crayhon, M.S., author and champion of the "Paleolithic diet", divides carbohydrates into two basic groups, paleocarbs and neocarbs. Paleocarbs include vegetables, fruits and perhaps tubers. Neocarbs (carbohydrates introduced within the last 10,000 years or less), include grains, legumes, and especially flour products, which did not exist for most of human history.

The worst of the neocarbs include sugar and white flour products. If we follow the simple guidelines of restricting ourselves to paleocarbs, we will in general be eating fiber rich, nutrient dense, low glycemic carbohydrates, the best nature has to offer.

Epidemiological Data

Another argument against carbohydrate restriction is based on epidemiological evidence, and the Pima Indians are frequently cited. The Arizona Pima Indians have received the attention of the medical community because of their prodigious rates of obesity, which is nearly 70% among the adult population. Along with the reputation of being one of the most obese people known, the Arizona Pima has a rate of diabetes 8 times the national average with nearly 50% of the adult population over 35 afflicted with this condition.

In spite of innumerable studies, examining the Pima from every imaginable vantage point, there has been no defining discovery explaining the Pima's plight. One hypothesis favored by Eric Ravussn, Ph. D, is that after generations of living in the desert, the only Pima who survived famine and drought were those highly adept at storing fat in times of plenty. These "thrifty" genes which once ensured the Pima's' survival are now at the root of his demise.

Although it is not known for certain what metabolic processes these "thrifty" genes control, insulin resistance and glucose homeostasis are thought to be at the heart of the matter. Since preagricultural, man's diet was primarily derived from animal sources (protein/fat), an insulin resistant genotype would have minimized glucose utilization and thus, proven to be of an evolutionary advantage. As primitive peoples have become acculturated and have assumed a modern diet, the constant supply of highly refined, high glycemic index carbohydrates has resulted in postprandial hyperinsulinemia and the subsequent diseases associated with this condition i.e. obesity, diabetes, cardiovascular disease, etc.

The Arizona Pima's diet prior to acculturation was essentially that of a hunter-gather with some subsistence farming: (chollacatus buds, honey mesquite, poverty weed, prickly pears, mule deer, white-winged dove, black-tailed jackrabbit, squawfish, and they raised wheat, squash and beans). However, by the end of the second World War, the Pima had almost entirely left their traditional lifestyle and adopted the typical American diet.

There are many problems with the typical American diet, and to blame the Pima's situation on just one element of that diet would be disingenuous. However, given the current scientific and anthropological studies, one could suggest that the high availability of sugar and highly refined, high glycemic carbohydrates (i.e. neocarbs), are at the core of the Pima's health crisis. It could also be extrapolated that, while the Pima's "thrifty" genes may work at a more accelerated pace, it is the same set of genes interacting with the same diet and producing the same results in the average American.

In 1991, the Pima's story became even more interesting. Peter Bennett FRCP, the lead epidemiologist studying the Arizona Pima, discovered in Sierra Madre, Mexico, the remnants of a tribe that once comprised the Southern half of the Pima Nation. However, unlike their Northern brothers, the Mexican Pima remained, in general, unacculterated and living a traditional lifestyle. Also, unlike their northern counterparts, the Mexican Pimas were not obese, nor did they share in the Arizona Pima's high rate of diabetes and degenerative diseases. This dichotomy has been termed the "Pima Paradox." Since the Mexican Pima consume a diet comprised mostly of beans, potatoes, corn tortillas and the occasional animal product, (i.e. chicken) , this has often been used as the epidemiological case study for the benefit of high carbohydrate diets in obesity management.

However, two issues confound this example. First, on average, the Mexican Pima's have 23 to 26 hours/week of occupational physical activity versus the Arizona Pima's 5 hours or less. Certainly, such high levels of activity could mitigate the hyperinsulinemic effects of the Mexican Pima's diet. The second issue is the "Enigma" within the "Paradox". Although the Mexican Pima does not have the health issues of the Arizona Pima, they still have a prevalence rate of diabetes at 6.4% (approximately 1.5x greater that the non Pima Mexicans), and a 13% incidence of obesity among the adult population.

While these numbers are impressive compared to the US population, and stellar compared to the Pima population, the question remains why should an essentially unacculturated population performing on average 23-26 hours of physical labor per week have any incidence of diabetes or obesity. When modern day hunter-gatherers were studied by anthropologists, incidence of these conditions were non existent, even among the eldest members of tribe. The "evolutionary diet" model would thus suggest, in spite of their improved health over the Arizona Pimas, the Mexican Pimas are still consuming a less than optimal diet.

Although conclusions drawn from epidemiological data can sometimes be misleading, the real message that can be taken from the Pimas is that as a species we have proclivity towards obesity, a proclivity that will vary based on our genetic stock. This genetic predisposition, while multifactorial in nature, probably centers around insulin resistance and glucose homeostasis. Since our preagricultural ancestors did not have ready access to simple carbohydrates, fats were the preferred source of caloric energy, and glucose conservation was evolutionarily advantageous.

In modern times, the detrimental combination of low physical activity, hypercaloric intake, and over consumption of neocarbs is at the root of our obesity crisis. A return to an evolutionary based diet - lean meats, seafood, fish, vegetables, fruits, (raw) nuts and seeds and moderate physical activity, will ultimately be the cure.

Health Risk Associated with reduced Carbohydrate Intake

Another argument against carbohydrate restriction focuses on the purported health risk of this dietary approach. Of the three macronutrients, protein, fat and carbohydrate, it is only carbohydrate that is nonessential to the human diet. Humans can exist for extraordinarily long periods of time without carbohydrate consumption as long as essential protein and fat needs are met. It is thus perplexing why nutritional dogma ascribes so many risks to the restriction of this non-essential nutrient.

Ketosis

Ketosis is a natural physiologic state induced during prolonged states of decreased glucose availability. It is triggered by severe coloric restriction or when carbohydrate intake falls below 20-30 grams, (most of the current low carbohydrate diets are nowhere near this level of restriction).

In ketosis, a set of elaborate metabolic processes occur which have the net result of decreasing insulin secretion, increasing glucagon secretion, switching off glycolysis, turning on lipolysis, switching muscles from glucose to almost entirely fatty acids for fuel, and ultimately providing ketone bodies (produced in the liver), markedly diminishing the need for glucose by the brain in particular and the body in general.

Ketosis was an absolutely vital survival mechanism for early man. It allowed him to survive periods of starvation as well as long periods of carbohydrate deprivation. Despite the role ketosis plays in normal human physiology, its' modern application has often been portrayed with multiple negative health connotations. However, both scientific and epidemiological data has failed to justify these concerns. The ketogenic diet has been used for nearly 70 years to treat refractory seizures in the pediatric population. Multiple recent studies have described nutritionally balanced, food varied versions of this diet.

One investigator looked at the health profiles of adults who had been treated during childhood with ketogenic diet. He found no evidence of adverse effects on cardiovascular function, including arteriosclerosis, hypertension or cardiac abnormalities. Blood cholesterol determinations were performed on these adults and all were normal. These studies thus fail to reveal any short term complication or long term sequelae associated with ketogenic diets.

In the mid twenties to late thirties, the famed anthropologist V. Stefansson chronicled the life and culture of the Eskimo in a series of books and journal articles. Of the many observations made by Stefansson, he was most intrigued with their diet and health. In spite of a nearly 100% animal based diet, the Eskimo people enjoyed an excellent state of well being and a freedom from many western diseases.

This observation was greeted with a high degree of skepticism in a scientific community that was becoming increasingly hostile toward the role of protein and fat in the American diet. To silence his critics, Steffansson devised a study whereby he would consume an all meat diet for one year. Under observation at Bellvue Hospital in New York City, Stefansson and a colleague did in fact consume for one year an all meat diet. At years end, to the surprise of the scientific community, both investigators were in excellent health. They demonstrated weight loss with reduction in body fat, normal kidney and liver function, and improvement in blood lipids (within the limits of diagnostic testing of the time).

The "Bellvue ward study" created quite a stir in the scientific community and was detailed in numerous articles appearing both in popular and professional literature. Although long term commentary cannot be made, this remarkable study certainly speaks to the short term safety of a ketogenic diet. Ample scientific, epidemologic and anthropological data exists to support the general safety of a ketogenic diet. However, this data does not exonerate all the modern inceptions of this diet.

Traditional cultures who consumed a largely animal based diet, derived a great deal of their vitamins and nutrients by consuming the organs, eyes, glands and gonads of their prey. Modern ketotic diets are primarily based on common American foods, i.e. meats, eggs and cheeses. They do not qualify the source of animal products (i.e. salmon versus bacon), and are usually overloaded with salt. In general, these diets are only concerned about limiting carbohydrate intake without overall regard to food quality.

In the most popular version of the ketogenic diet, Dr. Atkins New Diet Revolution, Dr. Atkin's writes "at the other end of the spectrum is a convenience food that sounds terrible fatty, but in fact, contains nearly none. Those are the maximizers of crispness - fried pork rinds - the zero carbohydrate consolation prize for corn or potato chip addicts. Virtually all the fat has been rendered off, leaving you with the protein matrix that held the pork fat together. Your pate, sour-cream based dips and guacamole find an exceedingly crisp and comfortable home atop a fried pork rind.

In spite of their potential physiologic benefits, the modern ketogenic diets with their unbalanced, nutrient poor and often absurd dietary suggestion are difficult to support. However, ketogenic diet based on evolutionary appropriate foods would be interesting to pursue in clinical practice. Lack of fruits, vegetables and grains Aside from the ketogenic diets, most other reduced carbohydrate programs allow for the ample consumption of vegetables and the modest consumption of low glycemic fruit, (the best sources of nutrients and phytonutrients available to man).

Of the major carbohydrate sources mentioned, only grain is heavily restricted. Although present diet dogma portrays grain as the quintessential food source, (it is at the base of the food pyramid after all), many nutritional scientist have called this assertion into question. In a work of prodigious proportions (342 literature citations), Dr. Loren Cordain examines mans double edged relationship with grain.

On one hand man is utterly dependent upon grain as a primary caloric source and yet grain may be at the core of many of our common maladies. As would be predicted by the evolutionary diet model, Dr. Cordain concludes that grain is biologically novel to the diet of mankind as it was introduced as a staple food only 10,000 years (or less) ago. Due to its relatively recent introduction, our species has not fully adapted physiologically to its digestion and metabolism.

In spite of the impressive nutrient profiles of grain, the vitamins and minerals often occur in forms that have low bioavaildality to the human digestive tract. In addition to these poorly utilizable nutrients, grain contains many secondary metabolic components commonly categorized as anti-nutrients.

Anti-nutrients are chemical compounds naturally occurring in grains, which evolved to protect the plants from predators. Processing and cooking does not not fully rid the grain of these elements, thus making them prominent in our diet. Recent scientific study has linked these anti-nutrients to a number of negative biological consequences which include: allergen based disorders; pancreatic hypertrophy and disruption of the gut cell wall tight junctions (thus exposing the systemic circulation to food allergens and gut flora).

One of the most curious of these negative processors associated with grain anti-nutrients is a phenomenon known as molecular mimicry. Molecular mimicry is when a similarity of structure is shared by products of dissimilar genes. When this phenomenon occurs within the human body, the potential for developing an autoimmune reaction is created. The main body of evidence implicates viral and bacterial pathogens as initiators of cross-reactivity and autoimmunity. However, there is an emerging body of literature supporting the view that dietary antigens including cereal grains may also induce cross-reactivity and hence autoimmunity by virtue of peptide structures homologous to those in the host.

The diseases that may share this common origin are numerous and varied. They may include everything from aphthous ulcers (canker sores), to rheumatoid arthritis to non-insulin dependent diabetes to multiple sclerosis. While many of these assertions may seem preposterous to a society reared on grain, evolutionary pressures would suggest otherwise. The primate gut was initially adapted to both the nutritive and defensive components of dicotyledonous plants rather that the nutritive and defense components of mono- cotyledons cereal grains.

Consequently, humans, like other primates, have had little evolutionary experience in developing a physiology that can both fully utilize and defend against the compounds which naturally occur in cereal grains. So, while the motives for limiting grains may be completely unrelated, many of the popular incarnations of reduced carbohydrate diets may be paying their readers a great - albeit - indirect service.

Increased Saturated Fats

Of all our nutritional mantras, the one most widely and emphatically proclaimed is the relationship between saturated fats and coronary artery disease. One would think a "fact" so ingrained in our social psyche would be supported by mountains of evidence. However, the reality is the data to support the "diet-heart hypothesis" is flimsy at best - non existent at worst. In an extensive review of existing studies, Ravnskov came to the conclusion that, "Few observations agree with the diet-heart idea, but a large number have falsified most effectively.

Man's diet possibly includes factors of importance to the vessels or the heart, but there is little evidence that saturated fatty acids as a group are harmful or that polyunsaturated fatty acids as a group are beneficial." In a similar review, Dr. Mary Enig was also unable to find a solid relationship between saturated fat consumption and coronary artery disease. She instead came to the conclusion that the inordinate increase in trans fatty acid consumption was more likely the causative factor.

When discussing the "dietary heart hypothesis", the work of Dean Ornish, M.D., is often cited as clinical evidence for the efficacy of dietary fat reduction. However, while Ornish is a major proponent of the "low fat diet", in his studies a number of coronary artery risk factors are addressed, in addition to the dietary changes. In Ornish's work, study participants underwent vigorous lifestyle changes, which included smoking cessation, stress management, exercise and a low-fat (near vegan) diet (the only animal products allowed were egg whites and one cup of non-fat milk or yogurt per day).

After following these changes for one year, the experimental group did show an overall regression of atherosclerotic plaque, Ornish's study is extraordinarily important because he was able to demonstrate, in quantifiable terms to the medical community, that lifestyle changes could be as powerful as drugs in managing a serious disease. However, to extrapolate that this study proves the value of the low fat diet is fallacious.

Ornish manipulates four separate variables in his study, all of which have purported association with cardiovascular disease. To suggest that any one variable or combination of variables is more important than the other cannot be concluded from Ornish's data. Even if diet alone is examined, there are multiple variables within the diet, that in and of themselves could have significance. Was it the omission of trans fatty acids (which have been linked to cardiovascular disease)? Was it the increase of antioxidants provided by the intake of fresh fruits and vegetables? Was it the fact that the experimental group experienced an average loss of 22 lbs?

Again, to conclude that it was the "low fat diet" which was primarily responsible for the experimental group's success (as the study is often interpreted), is quite disingenuous. A factor often overlooked in Ornish's work is the effect of low fat/high carbohydrate diets on lipid profiles. While it is true, the experimental group had an overall reduction in cholesterol, there was a concomitant reduction in HDL cholesterol with an increase in triglycerides.

Numerous recent studies have verified this dietary effect. Of these current studies, Berglund specifically looked at the response of the reduction in dietary total and saturated fats and HDL cholesterol subtypes. The study demonstrated a decrease in dietary total and saturated fat resulted in a significant decrease in HDL2 and HDL2b cholesterol concentrations. The authors concluded that the dietary changes suggested to be prudent for a large segment of the population will primarily affect the concentrations of the most prominent antiatherogenic HDL subpopulations.

Although definitive conclusions for the general population may be premature, in individuals demonstrating evidence of hyperinsulinemia and dyslipidemia (i.e. - Syndrome X) carbohydrate restriction is imperative for improved lipid profiles. In nutrition, as well as in life, balance is always the key. Nowhere is balance more crucial than in the discussion of dietary fats.

No diet, whether it be high fat - low fat (or anything in-between), if it promotes imbalances in fatty acid profiles, will in the long run have negative health consequences. In the mid '50s, the biochemist, anthropologist, and explorer Hugh Sinclair suggested an alternative explanation for the relationship between dietary fat and cardiovascular disease.

Sinclair noted that several people groups existed that consumed relatively high amounts of fat and yet were free of heart disease. Sinclair detailed the dietary habits of the Eskimos (previously discussed); the Masai people of Kenya who ate large quantities of ruminant milk and meat; and Jamaicans who ate large amounts of saturated fat in the form of coconut oil. All three groups, all consuming high fat diets, were relatively free from heart disease.

Sinclair suggested that the polyunsaturated profiles of these diets were protective, and concluded that the rise in cardiovascular disease was more related to their exclusion from the diet rather than the inclusion of saturated fats or cholesterol. Since Sinclair's day, our biochemical understanding of fat has increased exponentially. We now realize it is not just the polyunsaturated content of the diet, but the ratio of N-6 to N-3 polyunsaturates that may ultimately determine health.

Both dietary extremes discussed fail to introduce balance in this ratio. High carbohydrate diet due to their high grain and plant content will ultimately be low in N-3 fats (especially long chain N-3 fats - i.e. EPA/DHA), thus unbalancing the N-6/N-3 ratio. Low carbohydrate diets, in their popular form, rely heavily on commercially raised grain-fed meats and poultry (the fatty acid profile of the meat from wild game, free range beef and poultry have a significantly higher N-3 to N-6 ratio), eggs (free range hens also make better eggs) and cheeses.

A diet based on these foods will also greatly unbalance the N6/N3 ratio. Although the precise ratio remains controversial, the N6/N3 ratio should probably be in the range of 4-3/1 to optimize human health, western diets rich in vegetable oils, cereal grains and grain fed live stock, drive this ratio to an unprecedented 50-10:1. This imbalance may have implications in a host of diseases, including hyperinsulinemia, artherosclerosis and tumorgenesis.

When the diets of hunter-gatherer populations are studied, authors have concluded that their N6/N3 ratio varied between 4:1 to 1:1. This ratio appears to be biologically optimal. Based on these considerations, investigators, have advocated a return to dietary ratios of ancestral humans. A diet based on lean meats (wild game or free range livestock), fish, raw nuts and seed, vegetables, low glycemic fruit (paleocarbs) - "an evolutionary diet" - not only will be helpful in the management of obesity, but in a host of other common western diseases, including cardiovascular disease.

Dietary Protein and Cardiovascular Disease

Multiple recent studies have demonstrated the benefit of dietary fats (especially N-3 polyunsaturates and monounsaturates) in cardiovascular disease and in the reduction of cardiovascular risk factors. A more recent study trend has examined the possible beneficial role of dietary protein.

Wolfe has published numerous articles demonstrating the positive effects of the isocaloric substitution of protein for carbohydrate on lipid profiles. His studies have demonstrated a decreased LDL-C, an increased HDL-C, and reduction of triglycerides, thus reversing the dietary effects of increased carbohydrates. Wolfe states that substitution of carbohydrate for fat in the diet results in a reduction in HDL apoprotein transport rates along with increased catabolism of apolipoprotein A-1.

The decreases in plasma VLDL and LDL resulting from substitution of protein for carbohydrate in the diet may relate to either increased catabolism or decreased production. Thus, according to Wolfe's work, the simple dietary substitution of protein for carbohydrate could have profound health benefits.

Wolfe's data has recently been validated by Hu. In this study the dietary habits of over 80,000 women were examined. After controlling for variables, high protein intakes were associated with lowered risk of ischemic heart disease. Both animal and vegetable protein sources were protective. This inverse association was noted in women on both low fat or high fat diets. Wolfe's and Hu's work both indicate that dietary protein has cardioprotective properties independent of those of dietary fat.

Given the multiple health benefits ascribed to N-3 polyunsaturates and the evolving data regarding dietary protein - fish may be one of the best foods for human consumption. In a fascinating piece of epidemiological work, Marcovina compared 2 racially homogenous Bantu populations from Tanzania. The only appreciable difference between the groups was their dietary habits. The Bantu living closer to the shore had a predominantly fish based diet, while the inland Bantu consumed an essentially vegan diet (a diet devoid of animal products ). When plasma lipoprotein (a) (an independent cardiovascular risk factor) levels were compared, those among the fish eating population were 40% lower. This suggests another cardioprotective aspect of fish consumption.

In a recent study by Mori, he demonstrated the inclusion of fish in a weight loss program yielded greater results than either fish consumption or weight loss alone in their obese subjects. The experimental group in their study demonstrated improved glucose, insulin and lipid metabolism, as well as greater reductions in blood pressure, heart rate and weight loss versus controls. This study suggests a novel approach to the dietary management of obesity and NIDDM.

Perhaps the most influential of the studies looking at the benefits of fish, was the Diet and Reinfarction Trial (also known as the DART trial). In this study, the authors demonstrated that the addition of a modest amount of fish (2-3g of EPA per week or the equivalent of 300g of fatty fish per week) reduced post myocardial infarction mortality by about 29% when compared to controls.

One of the more interesting aspects of the study was that the control group was instructed on the standard fat reduction diet and on average had lower cholesterol levels than did the experimental group. The authors theorized that the fish oils had a favorable effect on clotting mechanisms and blood platelets, as well as a potential anti-arrhythmic effect on the ischemic heart. The results of this study are profound, especially given the modest and otherwise innocuous interventions undertaken.

Given the evidence of the benefit of N-3 polyunsaturates, coupled with the potential benefits of dietary protein, fish clearly is a biologically superior food source. The isocaloric substitution of fish for dietary carbohydrates is not only evolutionary appropriate, by may have untoward health benefits from weight control to improved glucose homeostasis to cardiovascular disease prevention.

Risk of Osteoporosis

Of all the potential negative side effects of dietary protein, the issue of osteoporosis is perhaps the most difficult to resolve. The literature is greatly divided on the topic, and clear recommendations are hard to find. In a recent study, Munger found that the intake of dietary protein, specifically from animal sources was associated with a reduced incidence of hip fractures in post menopausal women.

In the articles' discussion, a brief review of protein's controversial role in osteoporosis was undertaken. In the studies showing a potential benefit (as in the author's paper), it has been theorized that dietary protein may strengthen bone by its effect on the structure and function of bone-related proteins.

In studies demonstrating a negative effect, it has been argued that dietary protein (especially in the form of animal based protein) is a primary source of acid ash, which results in the acidification of urine. In order to buffer the urine and maintain acid-base homeostasis, calcium salts are mobilized from the skeleton, resulting in a net calciuria. Over time, this buffering of endogenous acids may contribute to a progressive decline in skeletal mass and, ultimately, lead to osteoporosis.

However, Wachman and Bernstein, the two authors who originally postulated this mechanism for osteoporosis, theorized that by increasing the dietary alkaline ash this process could be halted. In a study by Sebastian., he was able to reduce calicuria and improve overall calcium/phosphorous balance by the administration of potassium bicarbonate as a buffering agent to postmenopausal women consuming an acid promoting diet. The authors suggest that potassium bicarbonate could be administered long-term as a novel means of preventing and treating postmenopausal osteoporosis.

In a 4-year longitudinal study by Tucker, he was able to demonstrate that a greater bone mineral density was associated with increased dietary potassium and magnesium levels, as well as increased consumption of fruits and vegetables. The authors concluded that this positive association was due to the beneficial effects of potassium and magnesium on calcium balance and bone metabolism, as well as the buffering properties of increased alkaline ash in the form of fruits and vegetables.

Given the divergent nature of the theories, it is highly probable that both have merit. With respect to protein's beneficial effects, protein is certainly necessary for proper bone matrix formation and metabolism. It is likely a chronic suboptimal intake will jeopardize this function. One could conjecture that the studies finding a negative association between protein and osteoporosis have somehow highlighted this aspect of the equation. Those studies finding a positive association between protein and osteoporosis are probably looking at the endogenous acid production issue.

In an article by Remer, he calculated the potential renal acid load (PRAL) of frequently consumed foods in order to help dietitians design diets of varying urinary pH. On their list, animal protein sources (as expected) were calculated to increase PRAL. However, grain products, legumes and dairy products (especially hard cheeses) also increased PRAL. In fact , according to Remer's data brown rice had a greater PRAL than any of the meat products examined (with the exception of canned corned beef - if you want to call that meat).

Perhaps the most ironic of all, was Remer's finding that cheeses had the highest of the calculated PRALs. Parmesan, cheddar, and processed American cheese had PRALs almost 2 times any meat product. In light of Remer's data, the relationship of protein and osteoporosis cannot fully be determined without addressing the total dietary PRAL. The type of protein being consumed (lean meats vs. Processed meats vs. Cheese) and the other foods in the diet are likely to significantly affect the study's outcome.

The protein osteoporosis controversy was addressed in a review article by Spencer. According to the author, numerous studies have been published on the calcium-losing effect of protein. However, several aspects of the study conditions have to be considered in the interpretation of the results. Some of these are the type of protein, such as purified proteins (which seem not to promote calciuria): the duration of the study (there may be a transient increase in calciuria followed by a normalization or reduction); whether the phosphorous (which has an independent calcium sparing effect) intake remained the same, was increased, or decreased; whether the diets were under strict control or with outpatient volunteers; whether the protein intake was changed from a low to a high protein intake or was changed from a normal to a high protein intake; and whether excessively high protein intakes were used.

All these factors affect urinary calcium excretion during high protein consumption. After reviewing the available data, based on the aforementioned criteria, the authors concluded, "to our knowledge, no convincing data have been published showing that a high protein diet, using complex proteins for prolonged periods of time under strictly controlled dietary conditions, causes calcium loss."

It is quite obvious that the role of dietary protein in calcium homeostasis is complex and multifactorial in nature. However, given the work of Remer, it may actually be the net PRAL of the diet that is most important in influencing the development of osteoporosis, rather than the diet's absolute protein content. Since most of the current low carbohydrate diets encourage the ample consumption of vegetables, this is likely to offset any potential acidifying effects of increased dietary protein.

In fact, given most individuals do not consume enough vegetables and fruits, these diets are likely to promote better acid-base balance then the average American diet. Unlike the more modified low carbohydrate diets, modern ketogenic diets may pose a risk for calciuria since they rely heavily on animal protein, cheeses, and cured meats, and are usually not salt restricted (the Cl ion- not the Nat ion - can also cause a renal acid load and subsequently calciuria).

However, since most people are in ketosis for only a short period of time (after which they are theoretically supposed to transition into a modified low carbohydrate diet), it is unlikely that these diets will significantly contribute to an individual's overall risk for osteoporosis.

Kidney and Liver Damage

While it is generally accepted that people with pre existing kidney and liver disease will benefit from some level of protein restriction there is no data to support proposition that increased dietary protein will actually cause kidney or liver damage. In a study by Blum, he examined the kidney function of a group of healthy individuals consuming an ad lib. high-protein diet, as compared to a group of healthy vegetarians (Isn't that an oxymoron?). At the study's end, the authors concluded that protein does not affect kidney function in normal kidneys, and it does not influence the deterioration of kidney function with age.

The relationship of protein and the liver is somewhat more complex. Although there is no evidence that increased dietary protein will cause permanent liver damage, there is an actual dietary "protein ceiling". According to Rudman there is a lever at which dietary protein intake can exceed the liver's ability to metabolize it to the urea, thus leading to a build up of intermediary metabolites. These metabolites can subsequently lead to a toxic state in the affected individual.

The level of protein at which this will occur varies, but it is thought to be possible when protein makes up 30-40% of the calories in an eucaloric diet (the percent calories from protein can be higher in a hypocaloric diet).

"Rabbit Starvation" (a term coined by V. Stefansson to describe the phenomenon of excessive dietary protein) often occurred among explorers who would live for long periods of time on extremely low fat small game animals (i.e. rabbits). The condition was marked by nausea, vomiting, weight loss and fatigue. "Rabbit Starvation" was reversible when the percentage of daily calories from protein began to drop. Although the "Rabbit Starvation" phenomenon could effect an individual consuming a ketogenic diet, it is highly improbable.

In general, if one is consuming commercially available meats (even chicken), the percentage of calories from fat would be too high to induce this condition. In the modified low carbohydrate diets, due to the varied food sources, the risk of protein toxicity, for all practical purposes, is non-existent.

Conclusion

A critical reading of the current literature certainly supports the dietary trends of decreased carbohydrate intake (especially of neocarbs), increased protein intake, and increased fat intake (especially of monounsaturates and N-3 polyunsaturates). The data that supports these contentions comes from a wide spectrum of disciplines, including the basic sciences, medical science, epidemiology, and anthropology.

The one dietary program that addresses these principles in full, is the so called "evolutionary diet." The modern inception of this prehistoric lifestyle would favor the consumption of lean meats (preferably wild game or non-grain fed, free-range domesticated animals), fish, seafood, vegetables, fruits, raw nuts, and seed. Notably absent from this dietary genre are dairy products, cereal grains, beans, legumes and concentrated sweets (except for perhaps the occasional foray into raw honey!).

Adherence to these dietary guidelines will not only address obesity, but may also prove helpful in the management of everything from NIDDM to diseases of autoimmunity to cardiovascular illnesses. The guidelines are broad, but can be made quite specific depending on the goals, lean body mass, activity level, and overall health of the patient.

In the last few years, there has been a literal explosion of data in the nutritional sciences. Sometimes when addressing this data, we are put in the uncomfortable situation of realizing that today's facts are rapidly becoming tomorrow's fiction. However, by keeping an open mind and always questioning what we think we know, we will be able to provide our patients with the best and most innovative care possible.

DR MERCOLA'S COMMENT:

My congratulations to Dr. Brasco for compiling such an outstanding review of the concerns that some have when confronted with the "low carb" diet. Dr. Brasco is an internist and gastroenterologist and I believe one of the best in the country. It is a strange paradox of medicine that most GI specialist know virtually nothing about nutrition. That is certainly not true of Dr. Brasco who is clearly one of the leading nutritional GI specialists in the country. I could not recommend him more highly if you need a specialist.

I typically warn my patients that the diet recommended is NOT low carbohydrate but full of vegetables which are the good carbohydrates. Dr. Brasco provides an incredible review of the literature and some very sound scientific support for what appears to be the diet most of us were designed to eat.

I frequently explain to patients that part of the reason for the confusion on the carbohydrate issue is the fact that not all carbohydrates are created equal. The glycemic index mentioned above is one science tool that is used to explain this, but most patients have a hard time with this concept. I give them an analogy to think of grains and most below ground vegetables as a simple train. Each car in the train represents a simple sugar molecule which is easily broken down once it reaches the digestive system.

I then ask them to visualize that same train but this time stacked 20 to 50 high with other trains and each train care interconnected to the cars above them. This is an accurate representation of the much more highly complexed and branched sugar molecules that are present in most above ground vegetables. They have multiple bonds connecting each of the sugar molecules and take the body a long time to break them down. The extra time allows the body to slowly use the sugar and thus not have to secrete large amount of insulin to store the excess.

References

1. Cohen, MN (1989) Health and the Rise of Civilization, Yale University Press.

2 Solbrig (1994)So Shall You Reap, Island Press.

3 Stringer,C. (1996) African Exodus, Henry Hold and Company Inc.

4 Eaton, SB et al. (1997) "Paleolithic nutrition revisted: a twelve-year retrospective on its nature and implications." European Journal of Clinical Nutrition; 51:207-216.

5 Broadhurst, CL et al (1998) "Rift Valley lake fish and shellfish provided brain-specific nutrition for early homo." British Journal of Nutrition; 79:3-21.

6 Unpublished data from Cordain, L.

7 Cordain, L. (1999) "Cereal grains:Humanity's double-edged sword." World Rev Nutr Diet; 84:20-73.

8 Cohen, MN (1987) "The significance of long-term changes in human diet and food economy." Food and Evolution. Toward a Theory of Human Food Habits. pp 261-283. University Press.

9 Cassidy, CM (1980) "Nutrition and health in agriculturalist and hunter-gathers: a case study of two prehistoric populations." Food and Evolution. Toward a Theory of Human Food Habits. pp 117-145. Redgrave Publishing Company.

10 Diamond, L (1992) The Third Chimpanzee: The Evolution and Future of The Human Animal. pp 180-191. Harper Collins.

11 Eaton, SB et al. (1985) "Paleolithic Nutrition, a consideration of its nature and current implications." N Engl J Med; 312:283-289.

12 Salmeron, J et al. (1997) "Dietary fiber, glycemic load, and risk of non-insulin-dependent diabetes mellitus in women." JAMA; 277:472-477.

13 Grundy, S (1999) "The Optimal ratio of fat-to carbohydrate in the diet." Annual Review of Nutrition; 19:325-341.

14 Reaven, G (1995) "Pathophysiology of Insulin resistance in human disease." Physiological Reviews; 3:473-484.

15 Rasmussen, OW et al. (1993) "Effects on blood pressure, glucose and lipid levels of a high-monounsaturated fat diet compared with a high-carbohydrate diet in NIDDM subjects." Diabetes Care; 16:1565-1571.

16 Jenkins, DJA et al. (1987) "Metabolic effects of a low glycemic index diet." Am J Clin Nutr; 46:968-975.

17 Holt, S et al. (1994) "Glycemic index, satiety and the cholecystokinin response." Am J Clin Nutr; 3(S):787S.

18 Ludwig, D et al. (1999) "High glycemic index foods, overeating and obesity." Pediatrics; 103(3):E26-E31.

19. Holt S et al. (1997) "An insulin index of foods: the insulin demand generated by 100 KJ portions of common foods." AM J Clin Nutr; 66:1264-1275.

20. Rossetti L (1990) "Glucose Toxicity." Diabetes Care; 13: 610-630.

21. Crayhon R (1998) The Carnitine Miracle pp. 34-35 M. Evans and Company, Inc.

22. Gladwell M (1998) "The Pima Paradox." The New Yorker Magazine; Feb 2:44-57.

23. Trevision R et al (1998) "The epidemiology of diabetes Mellitus." Nephrology, Dialysis Transplantation; 13 suppl 8:2-5..

24. Haffner AM (1998) "Epidemiology of type 2 Diabetes: Risk Factors." Diabetes Care; 21 suppl 3:c3-6.

25. Colagiuri S et al. (1997) "The Metabolic Syndrome: From inherited Survival Trait to a health care Problem." Experimental and Clinical Endocrinology and Diabetes; 105 suppl 2: 54-60.

26. Valencia M et al. (1999) "The Pima Indians in Sonora, Mexico." Nutrition Reviews; 57:S55-S58.

27. Eaton SB et al. (1988) "Stone Agers in the Fast Lane: Chronic Degenerative diseases in Evolutionary Perspective." The American Journal of Medicine; 84:739-749.

28. Price WA (1939) Nutrition and Physical Degeneration (6th edition). pp. 59-72. Keats Publishing, Inc.

29. Lieb CW (1929) "The effects on human beings of a twelve month exclusive meat diet." JAMA; 93:20-22.

30. Stryer L (1995) Biochemistry (4th edition) pp. 775-778. WH Freeman and Company.

31. Stanley S (1998) Children of the Ice Age pp. 188-248. WH Freeman and Company.

32. Carroll J et al. (1998) "The ketogenic diet: a practical guide for caregivers." J AM Diet Assoc; 98:316-321.

33. Nebeling L et al. (1995) "Implementing a ketogenic diet based on medium-chain triglyceride in pediatric patients with Cancer." J AM Diet Assoc; 95:693-697.

34. Prasad AN et al. (1998) "Diet Therapy of Epilepsy in the Nineties; Renewed experience with the Ketogenic Diet." Nutrition Research; 18(2):403-416.

35. Howell E (1985) Enzyme Nutrition pp. 44-46, Avery Publishing Group, Inc.

36. Atkins R (1992) Dr. Atkin's New Diet Revolution pp. 280-281. Avon Books.

37. Milton K (1987) "Primate diet and gut morphology: Implications for hominid evolution." In Harris M. Ross EB: Food and Evolution Philadelphia, Temple University Press.

38 Ravnskov, U (1998) "The Questionable Role of Saturated and Polyunsaturated Fatty Acids in Cardiovascular Disease." J Clin Epidemiol; 51(6) :443-460.

39 Enig, M (1993 ) "Diet, serum Cholesterol and Coronary Heart Disease": in Mann G Coronary Heart Disease.

40 Ornish, D (1990) "Can lifestyle changes reverse coronary heart disease?" Lancet; 336:129-133.

41 Lichtenstein, A et al. (1999) "Effects of Different Forms of Dietary Hydrogenated Fats on Serum Lipoprotein Cholesterol Levels" NEJM; 340:1933-1940.

42 Ascherio, A et al. (1999) "Trans Fatty Acids and Coronary Heart Disease." NEJM; 340: 1994-1998.

43 Knekt, P et al. (1994) "Antioxidant vitamin intake and coronary mortality in a longitudinal population study." Am J Epidemiology; 139:1180-1189.

44 Verlangieri, AJ et al. (1985) "Fruit and vegetable consumption and cardiovascular mortality." Med Hypoth; 16:7-15.

45 Kanders , B et al. (1992) "Reducing Primary Risk factors by Therapeutic Weight loss." In Wadden T et al Treatment of the Seriously Obese Patient. pp 213-230. The Guilford Press.

46 Dreon, DM et al. (1999) "A very-low-fat diet is not associated with improved lipoprotein profiles in men with a predominance of large, ion-density lipoproteins." Am J of Clin Nutr; 69:411-418.

47 Golay, A et al. (1996) " Weight-loss with low or high carbohydrate diet?" International Journal of Obesity; 20:1067-1072.

48 Sheard, N (1995) " The Diabetic Diet: Evidence for a New Approach." Nutrition Reviews; 53(l) :16-18.

49 Williams, PT (1999) "Low-fat diets, lipoprotein subclasses and heart disease risk." Am J of Clin Nutr; 70: 949-950.

50 Berglund, L (1999) " HDL-subpopulation pattern in response to reductions in dietary total and saturated fat intakes in healthy subject." Am J of Clin Nutr; 70:992-1000.

51 Broadhurst, CL (1997) "Balanced intakes of natural triglycerides for optimum nutrition: an evolutionary and phytochemical perspective." Medical Hypotheses : 49:247-261.

52 Yehuda et al. (1998) "Fatty Acids and Brain Peptides." Peptides; 19(2):407-419.

53 Pond, C (1998) The Fats of Life. pp 289-293. Cambridge University Press.

54 Eaton, SB et al. (1988) The Paleolithic Prescription. Harper and Row Press.

55 Pan, DA et al. (1994) "Dietary Fats, Membrane Phospholipids and Obesity." Journal of Nutrition: 124:1555-1565.

56 Brouwer, BAJ et al. (1998) "Alpha-linolenic Acid does not augment long-chanin polyunsaturated fatty acid omega-3 status." Prostaglandins Leukot Essent Fatty Acids; 59(5):329-334.

57 Yam, D et al. (1996) "Diet and Disease - The Israeli Paradox: Possible dangers of a High Omega-6 poly unsaturated fatty acid diet." Isr. J Med Sci; 32:1134-1143.

58 De Lorgeril, M et al. (1990) " Effect of a Mediterranean Type of diet on the Rate of cardiovascular Complications in Patients with coronary artery disease." JACC ; 28(5):1103-1108.

59 DeLoreril, M et al. (1994) "Mediterranean alphja-linolenic acid-rich diet in secondary prevention of coronary heart disease." LANCET; 343:1454-1459.

60 Spiller, G et al (1998) "Nuts and Plasma Lipids: An Almond-Based Diet Lowers LDL-C while preserving HDL-C." J AM Coll of Nutr; 17(3):285-290.