Cholesterol, Animal Fats and Heart Disease: A Modern Myth?
(Chapter 10 of Dr. Ron's Book The Untold Story of Milk)by Ron Schmid, ND
The purpose of myth is to provide a logical model capable of overcoming a contradiction
(an impossible achievement if, as it happens, the contradiction is real).
Misconceptions about cholesterol and heart disease prevent many people from fully appreciating the value of raw milk and other vital foods. Many of you believe that cholesterol in foods can cause heart disease and wonder how unprocessed milk, butter and cream can be good for us since they are rich in animal fats and cholesterol. My answer is simple: “Because animal fats and cholesterol are good for you.” This is a difficult statement to make because so many of the health professionals and institutions we trust preach otherwise.
It is a surprising fact that the results of the major trials most often cited as evidence for the lipid hypothesis—the idea that cholesterol and animal fats cause heart disease—do not actually prove the theory. On the contrary, these studies contain serious flaws. Trial directors often manipulate statistics, sometimes in subtle ways but at other times in rather obvious ways, to give the appearance of success when results are at best equivocal. Conclusions about the trials often contain claims of success not actually supported by the data. Researchers then cite the summaries of unsupportive trials as evidence for further claims.1
The betrayal of trust described in the previous chapter pales in comparison to the campaign to persuade the public that cholesterol and animal fats cause heart disease. The ascendancy of the diet-heart idea has spurred the growth of the processed food industry and the concurrent demise of raw milk. An understanding of the cholesterol controversy will help us appreciate the value of vital foods and the reasons they have been lost.
The lipid hypothesis or diet-heart idea is a powerful and carefully cultivated myth that has played a major role in forcing thirty million farm families off the land since the end of World War II. I hope that having read this chapter, you’ll never again hesitate to eat good food, rich in cholesterol and saturated fat.
A Modern Myth - Page 2
The word myth has two meanings. The myths of ancient cultures are traditional narratives involving supernatural or imaginary persons, which embody deeply held ideas about natural or social phenomena. These are the myths that Carl Jung, the great Swiss psychiatrist and scholar, described so eloquently, springing like dreams from the depths of the collective unconscious. Jung, the founder of psychotherapy, called his work “the healing of souls.” Myths, said Jung, provided a window to the soul, a means of understanding humanity’s most basic truths.
Myth can also mean a widely held false notion. Thus the word myth can refer to something that is profoundly true—or patently false. The cholesterol theory of heart disease belongs to the latter category; it is a popular delusion that has engendered profound cultural changes in western society—from the way we eat and the way we farm to the way we think about our bodies and of nature herself.
We've explored many qualities of milk and other vital foods from domestic animals that live humane, natural and healthy lives. But those qualities are impossible for us to fully appreciate until we clarify issues concerning cholesterol and heart disease. The political and economic trends explained in the last chapter have fostered widespread acceptance of the diet-heart theory, the creature of persistent and unrelenting indoctrination against traditional foods.
The subject of nutrition and human diets—which should reveal its secrets through carefully executed scientific studies—has become highly politicized. The constant denigration of beef, animal fats, eggs, and locally produced dairy foods like raw milk, cream, cheese and butter—all of which were once the products of small farms—has no basis in good science but powerful support from monopolistic grain cartels, food processors, vegetable oil producers and pasteurized dairy manufacturers. Health professionals generally ignore the growing body of evidence indicting the ingredients of processed foods—liquid and hydrogenated vegetable oils, refined sweeteners, refined flour, processed milk products, preservatives and artificial flavorings—which are the largest contributors to modern disease, especially heart disease.
Cholesterol and Saturated Fat: Friend or Foe? - Page 3
Although the U.S. government’s National Cholesterol Education Program portrays cholesterol and saturated fat as villains, these substances are essential components of human biochemistry. In animals and humans, the cell membrane is the living boundary of the cell, the gate that lets the substances of life in and waste products out. Saturated fat and cholesterol are the major constituents of this membrane—without saturated fat and cholesterol, the cell cannot work properly. Saturated fat provides “stiffness” or integrity to the cell wall and cholesterol makes it waterproof. Thus cholesterol provides one of the main engines for life—a different chemistry on the inside and the outside of the cell.
The greatest concentration of cholesterol is found in the brain and nervous system, where electrical impulses course along the cholesterol rich membranes of the nerve cells. Cholesterol also provides the basic material for the creation of sex hormones and the adrenal hormones, which we use for everything from blood sugar regulation to dealing with stress. As for saturated fats, good science has discovered that they play many important roles—they enhance the immune system, protect us against pathogens, provide energy to the heart and are vital to the function of the kidneys, the liver, the brain and the lungs.2
Given the ubiquitous nature of cholesterol, is it not strange that so many of us have come to fear it? True, cholesterol can accumulate in undesirable places, for example in damaged areas of arteries where, along with other substances, it becomes a component of the plaques or atheromas associated with heart disease. Investigators have interpreted the presence of cholesterol in plaque as an indication that dietary cholesterol, found only in animal foods, is the cause of heart disease. But cholesterol is the body’s repair substance. Without cholesterol in the bloodstream, tears and irritations in the arteries would soon lead to aneurisms and ruptures. (One of the main constituents of plaque is calcium, but no one is blaming calcium for heart disease.) In recent years the political and medical establishments have presented the lipid hypothesis as proven fact, an impregnable structure built of the following tenets:
Cholesterol and animal fats in foods raise blood cholesterol, and high blood cholesterol causes cholesterol to build up in the arteries causing heart disease.
The higher the cholesterol levels in the blood, the faster heart disease occurs.
Lowering cholesterol helps prevent heart attacks and extend life.
Animal studies prove that cholesterol causes heart disease.
The anti-cholesterol campaign is based on good science and has the support of nearly all researchers.
We’ll now examine some of the history and research behind these claims.
Framingham and Ancel Keys - Page 4
Rising numbers of deaths from coronary heart disease occurred during the years following World War II, and a logical strategy for combatting the problem emerged. First, epidemiological studies would identify the risk factors associated with the disease, and then clinical trials would measure the effects of intervention.
The initial government-sponsored epidemiological investigation began in 1948. Researchers recruited two-thirds of the adult population of Framingham, Massachusetts as their living laboratory in order to pinpoint risk factors—observable characteristics or behaviors that demonstrated statistical correlations with heart disease.
It was in the Framingham study that high blood cholesterol first emerged as a risk factor for heart attack. Researchers divided the participants into three groups with relatively low, medium and high levels of blood cholesterol. Individuals in the latter group had slightly more heart attack deaths than those in the other two groups, thus indicating that high levels of cholesterol in the blood was indeed a risk factor, a predictor for coronary heart disease (CHD).3
But a risk factor is not necessarily a cause, even though a risk factor may change in parallel with the disease. For example, in the years following World War II, telephone usage increased dramatically in many of the population groups that experienced an increased incidence of heart disease. Telephone usage is a risk factor for heart disease, but obviously not a cause. But if the risk factor is the cause, it must move in parallel direction with the incidence of the disease, without exception.4 Thus, if butter or animal fat consumption declines while heart disease increases (which is what happened in the U.S.), then scientists must rule out butter or animal fats as a cause.
Although the Framingham study merely indicated that high cholesterol is a risk factor for heart disease (albeit a very weak one), it led to the belief that consumption of cholesterol and saturated fats is a cause of heart disease. Since all animal foods contain cholesterol and saturated fat, avoidance of these foods became the cornerstone of the National Cholesterol Education Program. The diet-heart idea thus allowed food companies to promote vegetable oils as healthy alternatives to traditional fats and a high-carbohydrate diet as a healthy alternative to a diet based on meat.
It was the work of Ancel Keys, the founding father of the lipid hypothesis, that allowed diet-heart proponents to make the leap from the Framingham study to the dietary guidelines of the National Cholesterol Education Program. As director of the Laboratory of Physiological Hygiene at the University of Minnesota, Keys published a series of articles linking heart disease with animal fat consumption, beginning in 1953. He constructed graphs showing linear relationships between the percentage of calories from fat and the incidence of heart attacks in several countries. In his Six Countries Study, for example, he was able to show an almost perfect relationship between fat consumption and deaths from heart attack in Japan, Italy, England and Wales, Australia, Canada and the U.S.5 Keys' tidy graphs were enormously influential. But Keys was able to show a perfect relationship only because he selected carefully, choosing the handful of countries that fit his hypothesis and ignoring many others that did not.6,7
Although the Framingham study has often been cited in support of lowfat, mostly plant-based diets, the researchers actually found no relationship between the foods eaten and cholesterol levels in the blood. “These findings suggest a cautionary note with respect to the hypotheses relating diet to serum cholesterol levels,” they wrote in their summary. “There is a considerable range of serum cholesterol levels within the Framingham Study Group. Something explains this inter-individual variation, but it is not the diet.” This statement appears in an unpublished manuscript written by Drs. William Kannel and Tavia Gordon, authors of the official published report.8
Actually, Keys himself once stated that there was “. . . no connection whatsoever between cholesterol in food and cholesterol in the blood. None. And we’ve known that all along.”9 He placed the blame for heart disease on saturated fats alone. But the Framingham investigators found no correlation of blood cholesterol levels with saturated fat either. In a statement published in 1992, Dr. William Castelli, then director of the Framingham project, made the following startling admission: “In Framingham, Mass., the more saturated fat one ate, the more cholesterol one ate, the more calories one ate, the lower the person’s serum cholesterol . . . we found that the people who ate the most cholesterol, ate the most saturated fat, ate the most calories, weighed the least and were the most physically active.”10 Most doctors today tell their patients that it is important to exercise. . . but they never mention the fact that eating saturated fats will make it easier to do so.
Today, most researchers admit that dietary cholesterol has only a slight influence on cholesterol levels in the blood—despite broad public acceptance of the dictum that we should restrict intake of cholesterol—because the body’s cholesterol level is regulated in the liver according to individual needs. The more cholesterol we eat, the less the liver makes, and vice-versa. Even an extreme diet generally lowers cholesterol by no more than about ten percent in the short run.11
I’ve observed that extended periods of protein deficiency in people who follow vegetarian or near vegetarian diets can lower cholesterol perhaps a bit more, even down to levels around 150 mg/dl for some people. This may be due to the lack of protein, needed for the production of cholesterol. (These patients are then distressed to learn that low cholesterol levels are a strong risk factor for cancer.)
Other Studies - Page 5
More contradictory evidence emerged in the 1976 Tecumseh Study.12 Experienced dieticians questioned over two thousand people in the small Michigan town of Tecumseh and subsequently analyzed the responses. The participants were then divided into three groups according to their blood cholesterol levels. The results: food intake had no relationship with serum cholesterol. The low-cholesterol group ate just as much saturated fat as did the high cholesterol group. No association between diet and cholesterol levels was found in children either, in two studies conducted in 1978, one at the Mayo Clinic in Minnesota13 and the other in New Orleans.14
Two interesting studies involved bank tellers, trained to keep meticulous records. Ninety-nine men weighed all of their food for a week, recording the results. No connection between the food they ate and their blood cholesterol levels appeared. Later in the year, seventy-six of them agreed to do it again, and still no connection emerged. Researchers then selected the tellers whose records were most meticulous—but failed once again to find a connection between what they ate and the levels of cholesterol in their blood.15
During the late 1970s and early 1980s, scientists at the National Heart, Lung and Blood Institute tested the idea that restriction of foods high in animal fat and cholesterol will significantly lower blood cholesterol. Researchers screened three hundred sixty thousand men for participation in the massive Multiple Risk Factor Intervention Trial (MRFIT, or “Mister Fit”) and chose twelve thousand considered especially prone to heart attack.
The treatment group made drastic dietary changes in hopes of avoiding heart attack and premature death. They cut their intake of saturated fats by one quarter and of cholesterol by half, and they increased their intake of polyunsaturated vegetable oils by one third. Over the course of four years, their cholesterol levels went down by an average two percent, an amount that was statistically insignificant.16 Moreover, while those who had reduced their cholesterol experienced a slight reduction in deaths from heart disease, they had greatly increased deaths from other causes—cancer, stroke, accidents and suicides. Mr. Fit became Mr. Frustrated.
This trial had unlimited funding and personnel—including dieticians, behavior modification specialists and psychologists—and the results should have put an end to the idea that the boring and unappetizing diet they embraced so enthusiastically could lower cholesterol levels. Instead, the cholesterol proponents promoted their agenda with increasing vigor.
Proof that high levels of cholesterol in the blood lead to formation of plaques in the arteries has also eluded investigators. In 1936, a research team at New York University studied individuals who had died violent deaths. They found absolutely no correlation between the amount of cholesterol in the blood and the amount of blockage in the arteries. Those with low levels of cholesterol had just as much atherosclerosis as those with high cholesterol levels, and many individuals with high cholesterol readings had little or no blockage.17 The 1968 International Atherosclerosis Project, in which over twenty-two thousand corpses in fourteen nations were examined for plaques in the arteries, showed the same degree of atheroma in all parts of the world—in populations that consumed large amounts of fatty animal products and those that were largely vegetarian, and in populations that suffered from a great deal of heart disease and in populations that had very little or none at all.18
When scientists propose a theory, they must also accept the burden of proving that theory. Even one bit of contradictory evidence requires a reformulation of the theory—or its abandonment. But mountains of contradictory evidence have not led to the abandonment of the cholesterol myth, just more obfuscation and shriller voices.
Dietary Trends - Page 6
Dr. Paul Dudley White, a Harvard graduate and a brilliant cardiologist, served as physician to President Eisenhower and wrote a textbook, Heart Disease, published in 1943. Thirty-five years later, when I was a medical student myself, I discovered these words in White’s text:
“When I graduated from medical school in 1911, I had never heard of coronary thrombosis [heart attack].”19 A surprising statement, I thought, coming from a famous cardiologist. I soon discovered the reason for it: the first article about coronary thrombosis appeared in the Journal of the American Medical Association in 1912. Heart attacks were rare in those days, despite the fact that many people lived to advanced ages. When White introduced the electrocardiograph to his colleagues at Harvard, they advised him to focus on a more profitable specialty as heart disease was so rare. The machine recorded unusual patterns in the rhythm of the heart and evidence of possible blocked arteries: the challenge was to find enough patients to make a living with the new invention.
When White began his career, over half of all Americans lived on small farms with their dairy cows, fowl, pigs and beef cattle. Both country people and city folk typically feasted on fatty meats, raw whole milk, eggs and plenty of butter. They cooked in bacon fat and lard, and used lard in pie crusts and pastries. Over eighty percent of the fats in the American diet were animal fats.20 By 1970, that percentage had fallen to about sixty percent, while cholesterol consumption had not changed. But by that time heart disease caused nearly half of all deaths in America—an increase of about one thousand percent between 1930 and 1970.21 In other words, while consumption of animal fats declined, the incidence of heart disease increased—these facts alone are sufficient to disprove the theory that animal fats and cholesterol cause heart disease!
During the same years, consumption of refined vegetable oils increased four hundred percent while the use of sugar and processed foods skyrocketed. Many studies have linked these foods to heart disease and cancer, but you never hear about them on television or read about them in the women’s magazines.22
Comparison of dietary patterns and rates of heart disease in other countries reveals similar inconsistencies. In England, consumption of animal fat remained stable while the number of heart attacks increased one thousand percent between 1930 and 1970.23 In Yugoslavia, four times as many people died of heart disease in 1965 as in 1955, while saturated fat consumption fell twenty-five percent. During the same period, processed foods had become much more widely available.24 In Switzerland, heart attacks have declined in the years since World War II, while the consumption of animal fat has increased about twenty percent.25
Further evidence that something other than animal fat causes heart disease comes from studies of two tribes in Kenya, the Maasai and the Samburu, discussed in Chapter 8. These tribesmen are shepherds, and their diets consist almost entirely of raw milk, blood and meat. Adult men typically consume almost one gallon of rich milk per day, providing at least one-half pound of utterfat—that’s two sticks of butter daily! Heart disease is nonexistent among those on traditional diets and their cholesterol levels are about half the value of those of most Americans.26 If a diet rich in animal fat and cholesterol is the most important factor in causing heart disease, these people would die of heart attacks at least as often as Americans do.
One popular misconception centers on the notion that the so-called Mediterranean diet is low in animal fat. First promoted by Dr. Keys—who seems never to have heard of salami—this theory has inspired a number of books and articles claiming that the traditional diets of Italy, Greece, Spain, Portugal and France are rich in vegetables, fruit, bread, pasta and olive oil, and contain some cheese and wine but little other animal food and saturated fat.
Anyone who has traveled to the Mediterranean region soon discovers that the inhabitants enjoy eggs, meat, fish, sausage, butter, cream, full-fat cheese, rich patés and lard on a daily basis. In fact, consumption statistics for these countries show that from about 1960 to 1990, saturated fat increased an average of about forty-five percent—increased prosperity likely meant more money for the purchase of animal products by more of the population, more money to enjoy the good life. Meanwhile deaths from coronary heart disease decreased over the same period by an average of about twenty percent.27
Proponents of the lipid hypothesis have singled out eggs as a particularly dangerous component of the diet because egg yolk is richer in cholesterol than any other food. Fear of cholesterol has reached the point that university researchers who want to give healthy study participants two eggs a day must often first consult the university’s ethics committee. While this may raise a smile, I do understand it, for my new patients often find it absolutely astounding and all but accuse me of attempted homicide when I suggest that they can eat all the eggs they wish with impunity.
Now that the public is widely fearful of eggs, it should be easy to produce evidence that egg consumption is dangerous. Does research show that individuals who have heart disease have eaten more cholesterol from eggs and other cholesterol-rich foods than individuals with no symptoms of the disease? Not at all! Ten major studies comparing the consumption of cholesterol by heart disease patients and healthy people found no statistically significant differences. The actual figures for the ten studies showed that the heart disease groups had eaten about two percent less cholesterol: the mean cholesterol consumption in the heart disease groups was 506 milligrams per day, and in the control groups it was 518 milligrams per day.28
Heart disease is more common in rich countries than in poor ones, and in richer countries people consume more animal food. But within the richer countries, poor people die much more frequently of heart attacks, despite eating far fewer foods rich in cholesterol and animal fats. People in prosperous countries also eat more sugar and refined foods, smoke more, perform less manual labor, eat the products of industrial agriculture and are exposed to more environmental pollution—all factors that could account for the greater incidence of heart disease in rich countries compared to poor ones.29,30
Many research trials about heart disease followed in the footsteps of the Framingham study, studies involving hundreds of thousands of people. During the 1950s and early 1960s, most of the trials looked at dietary patterns. The evidence for the lipid hypothesis in those trials was not good. In fact, over thirty published studies involving over one hundred fifty thousand people have failed to show any difference in animal fat consumption between those who develop heart disease and those who do not.31
Intervention trials, in which tens of thousands of people were put on cholesterol-lowering diets, have with one exception failed to reduce the number of heart attacks. The one exception was the Oslo trial, published in 1981. Both the researchers and the media drew optimistic conclusions based on these fragile findings, findings that have never been confirmed in other trials.
This lack of evidence is in obvious conflict with popular conceptions and the position of government, medical and pharmaceutical industry spokesmen, but has been well documented by a vocal minority of qualified researchers and authors.32,33 Uffe Ravnskov, MD, PhD, a Swedish researcher who has extensively studied the literature on the diet-heart idea, lists almost three dozen brave scientists who have published studies describing flaws in the modern cholesterol myth.
Ravnskov points out that in many studies researchers have manipulated statistics in ways that magnify trivial differences to make them appear significant, using a concept called “relative risk.”34 For example, if one person out of a thousand in a group with low cholesterol dies of a heart attack compared to two people in a group with high cholesterol, the difference is only one-tenth of one percent but the “relative risk” of having high cholesterol is said to be one hundred percent, because two is one hundred percent greater than one. These exaggerated figures are then cited in the article summaries—often the only part physicians take time to read.
Despite the lack of evidence and obvious manipulation of trial results, the National Heart, Lung, and Blood Institute insists that “Diet is the cornerstone of treatment of high-risk cholesterol levels,” and urges Americans to drastically reduce their consumption of cholesterol and animal fats.
Cholesterol Lowering—Not Such a Good Idea - Page 7
What about the idea that whatever the diet, high cholesterol levels in the blood cause heart attacks, and lowering cholesterol levels can reduce the incidence of coronary heart disease? This is the basic argument used to justify millions of prescriptions for cholesterol-lowering drugs, and for current recommendations that millions more people should be so treated, including children as young as eight years old.36 It seems reasonable to expect good evidence for this argument.
Earlier, we discussed how high blood cholesterol first emerged as a risk factor for heart disease in the Framingham study. Because individuals with higher levels of cholesterol suffered slightly more heart attacks, researchers concluded that high cholesterol was a risk factor, a predictor for heart disease. But many trials found only a weak association of high cholesterol with heart disease. As with the diet studies, researchers manipulated statistics in ways that magnified trivial differences, and made exaggerated claims in the summaries of the articles to justify unwarranted conclusions unsupported by the actual research.37
Many articles promoting cholesterol-lowering measures make reference to the summary of one particular major article about the Framingham study, published in 1987 in the Journal of the American Medical Association (JAMA).38 But the actual text of the article reveals that nearly half of those who had heart attacks during the Framingham study were in the group that had low cholesterol. Comparing all of the groups, women with low cholesterol died as often as women with high cholesterol. Furthermore, men between the ages of forty-eight and fifty-seven with cholesterol levels between one hundred eighty-three and two hundred twenty-two died more often than those with levels between two hundred twenty-two and two hundred sixty-one. The authors also noted
that individuals whose cholesterol had decreased without treatment over the thirty years of the study had a greater risk of dying than individuals whose cholesterol had increased. The crucial statement, “For each 1 mg/dl drop of cholesterol there was an 11 percent increase in coronary and total mortality,” was buried in the body of the paper and did not appear in the summary.
In 1990, the following statement appeared in the journal Circulation, published by the American Heart Association and the National Heart, Lung and Blood Institute: “The results of the Framingham study indicate that a 1% reduction of cholesterol corresponds to a 2% reduction in CHD risk.”39 Compare this with the quote above from the JAMA report that very few have read. No wonder so many physicians are convinced that cholesterol causes heart disease, in spite of the lack of evidence.
The Framingham authors themselves used similar tactics in 1987 to present revised, politically correct conclusions. The following statement appeared in a report concerning thirty years of follow-up, published in the American Journal of Cardiology: “The most important overall finding is the emergence of the total cholesterol concentration as a risk factor for CHD in the elderly.” The authors made no mention of several studies showing that high cholesterol levels in the elderly are unrelated to heart disease and may even be protective.40
Numerous other studies have confirmed the lack of correlation that emerged in the later years of the Framingham project. For example, a study in Sydney, Australia found that cholesterol levels had no predictive value in Australian men over the age of seventy-four.41 A study at the Albert Einstein College of Medicine in New York produced similar findings. Yet the authors of the latter study concluded, “The findings of this study suggest that an unfavorable lipoprotein profile increases the risk for cardiovascular morbidity and mortality.”42 As indicated earlier, the summaries of articles often contain conclusions having no apparent logical connection with data presented in the body of the paper.
Similar findings on the elderly came to light in work done at Yale University, where researchers studied almost one thousand elderly men and women over a four-year period: twice as many of the participants with low cholesterol had heart attacks as those with the highest cholesterol levels.43
These results have led Dr. Ravnskov and other researchers to conclude that high cholesterol may actually be protective.44 Although such thinking may well seem heretical, it is quite logical; the human body naturally seeks to heal itself, and increased cholesterol may be a normal protective mechanism when the body is under various types of stress. Consider, for example, the results of a French study reported in The Lancet in 1989. Women with low cholesterol had a death rate more than five times higher than that of women with very high cholesterol. In fact, the researchers noted that old women with very high cholesterol live the longest.45 At a 1992 National Heart, Lung and Blood Institute conference on cholesterol in women, participants looked at every study published about the risk of having high or low cholesterol. They concluded that mortality was higher for women with low cholesterol than for women with high cholesterol.46
As for men, many studies have similarly failed to show relationships between high cholesterol and heart disease. A research team published their findings in the Canadian Journal of Cardiology in 1990 after following over five thousand healthy middle-aged Canadian men for twelve years. Their conclusion: high cholesterol had no relationship with increased risk for coronary heart disease for the period of the study.47
Proponents of the lipid hypothesis state that high cholesterol levels are particularly dangerous for individuals who have already had a heart attack, but many studies contradict that notion. A team at the University Hospital of Toronto followed one hundred twenty men for ten years after their heart attacks and found no difference in the incidence of a second attack in those with high cholesterol compared to those with low. Many other published articles confirm these findings.48
Such articles have had no apparent impact on the American cholesterol juggernaut. In 1989 the National Research Council published Diet and Health, the largest official review on heart disease. The review concluded that there was “a strong, continuous and positive relationship between total cholesterol levels and the prevalence and incidence of, as well as mortality from, atherosclerotic CHD.”49
These tactics explain how supporters of the lipid hypothesis have convinced people to lower their cholesterol as much as possible. I run a chemistry screen panel on most of my new patients, and most of them express initial pleasure if their cholesterol level is below the new standard of 200 mg/dl. I then must explain that these results are not necessarily positive. Official American medicine appears to practice truth by decree. The practicing physician who does not recommend cholesterol-lowering drugs for, let us say, his elderly female patient who tests a bit above the official norm of 200 mg/dl, is guilty of failing to follow standard-of-care guidelines. He may be subject to disciplinary action from a medical review board and even the loss of his license, or perhaps a civil suit from a zealous family and their lawyers should his patient have a heart attack.
HDL and LDL - Page 8
Coincident with the belief that high blood cholesterol causes heart disease is the concept of good and bad cholesterol, HDL and LDL (high density lipoprotein and low density lipoprotein). Lipoproteins are particles composed of fats (lipids) and proteins that carry cholesterol through the bloodstream. HDL is thought to carry cholesterol from the tissues to the liver, where it is used for a variety of purposes or is excreted in the bile. LDL carries cholesterol from the liver, where most of the body’s cholesterol is manufactured, to the peripheral tissues.
Because a few studies have indicated that relatively higher levels of LDL are associated with greater risk of having a heart attack, LDL has been dubbed “bad” cholesterol and HDL “good.” A low HDL/LDL ratio was thus established as a risk factor for coronary heart disease.
But as we’ve seen, a risk factor is not necessarily a cause. Many factors that change levels of HDL and LDL, including weight loss, blood pressure, smoking and exercise, appear to also affect heart disease risk. Many studies have attempted to establish prognostic value for levels of HDL and LDL, but with no success. In one of the largest, British scientists looked at seven thousand men for over four years and also reviewed six other large trials before concluding that HDL and LDL levels were not major risk factors, much less causative factors, for coronary heart disease.50,51
Yet the U.S. government’s Diet and Health publication referred to above is unequivocal in its insistence that LDL should be lowered: “LDL has the strongest and most consistent relationship to individual and population risk of CHD, and LDL-cholesterol is centrally and causally important in the pathogenetic chain leading to CHD.”52 One would expect some very good evidence to back up such a strong comment. The authors cite four publications. Dr. Ravnskov analyzed these four and all of the studies to which the four refer, and found that evidence for increased LDL-cholesterol as a risk factor or causative factor for coronary heart disease does not appear in any of them. Ravnskov points out numerous obvious errors and shows how the conclusions of various studies are often at odds with data presented in the papers themselves. Proponents of the lipid hypothesis cite the conclusions of one flawed study after another to support their pronouncements, studies in which statistically insignificant data are repeatedly cited as significant. In short, no evidence indicates that increased LDL-cholesterol causes heart disease or is even a legitimate risk factor—no more a risk factor than high total cholesterol.
How to Give a Monkey a Heart Attack - Page 9
Proponents of the lipid hypothesis often cite animal studies as proof that cholesterol causes heart disease. Thousands of scientists have carried out cholesterol experimentation on laboratory animals. But no other mammal utilizes cholesterol in quite the same way humans do. Vegetarian animals do not normally eat foods containing cholesterol, and when they are force-fed cholesterol-rich foods, the level of cholesterol in their blood skyrockets. Changes may be induced in the arteries of some animals, rhesus monkeys for example, that vaguely resemble atherosclerosis in humans. Eminent scientists have criticized the notion that this observation can be used as proof that animal fats and cholesterol cause heart disease in humans as naïve and ill-founded.53
Rabbits are the animals most commonly used in cholesterol experiments. When forced to eat cholesterol-rich food, this vegetarian animal’s blood cholesterol rises to levels ten to twenty times higher than the highest ever seen in humans. Cholesterol is deposited in the arteries, among other places, but the deposits do not resemble the lesions of human atherosclerosis. Yet rabbit studies are frequently cited as proof that elevated serum cholesterol in humans causes heart disease.
During the 1960s, scientists used wild rhesus monkeys captured from the jungle in feeding experiments designed to induce what the researchers called atherosclerosis. They fed huge amounts of cholesterol-rich foods to twenty-seven animals who were kept in small individual cages, reinforced with solid metal sheets, in a Chicago basement laboratory. The unhappy monkeys ate little or went on long hunger strikes and threw their food around their cages.
Blood samples were taken from the groin arteries only with great difficulty, for the monkeys resisted violently—screaming, urinating and defecating. After four years of this treatment, one of the monkeys died of a heart attack, an animal reported to be especially hyperactive and extremely nervous. The researchers then published an article, “Fatal Myocardial Infarction in a Monkey Fed Fat and Cholesterol,” in the Archives of Pathology.54 The scientists and many others who have cited this study in later papers considered diet and high serum cholesterol levels, and nothing else, to be the factors that led to the monkey’s death.
Other scientists with a more integrated view of health and disease, and a more complete understanding of the literature on animal studies, believe otherwise. In a thorough review based on a comprehensive study of the experiments involving cholesterol in animals, William Stehbens, MD, a professor in the Department of Pathology at the Wellington School of Medicine in New Zealand, wrote the following: “Any pathologist of independent mind and free from preconceived ideas would conclude that human atherosclerosis and the lesions induced [in animals] by the dietary overload of cholesterol and fats are not one and the same disease.”55
The Pharmaceutical Trials - Page 10
We’ve looked at a number of aspects of the Framingham study, but perhaps most important is the light it sheds on today’s bottom-line belief that lowering your cholesterol will extend your life. This is the belief that has millions of Americans taking cholesterol-lowering drugs and millions more thinking about it.
The four most important risk factors that emerged from the Framingham study, in order of importance, were age, sex, high blood pressure and elevated cholesterol. Recall that the relationship for cholesterol was quite weak. Among the women in the study under the age of fifty-five, it was not a risk factor at all. For both men and women, the association weakened greatly with advancing age. The majority of heart disease deaths occurred among those with average cholesterol levels and there was no relationship between life expectancy and cholesterol levels for men or women after age forty-eight. The only notable cholesterol link involved the small minority of young and middle-aged men with levels of cholesterol of 280 mg/dl or higher; these men had heart disease at a rate about three times higher than men of the same age with average cholesterol levels.
Other epidemiological studies confirmed the finding that elevated cholesterol was a risk factor, but a weak one. This finding was clear by the early 1960s, but the question of whether or not lowering cholesterol would prevent or alleviate heart disease remained unanswered.56
The first great trial to address this question was the Coronary Drug Project, which included over eight thousand men who had suffered at least one heart attack. Sponsored by the National Heart, Lung and Blood Institute, the trial began in 1967 and tested several drugs, including clofibrate, one of the early cholesterol-lowering drugs. Treatment with the drug succeeded in lowering cholesterol levels, but after seven years just as many deaths occurred in those treated with clofibrate as in the control group. Many people in the treatment group suffered from severe side effects.
Follow-up studies years later revealed that the number of heart attack deaths in the group that had received the drug during the seven-year trial was greater than in the control group. Similar findings emerged in the follow-up studies of other large trials with clofibrate—findings that are not mentioned in the many subsequent articles that refer to the original papers.57
Another major trial was conducted by the Upjohn Company, this one on their new cholesterol-lowering drug, Colestipol.58 Doctors in hospitals selected over two thousand patients with high cholesterol and then allowed Upjohn scientists to select which patients would receive the drug and which the placebo—in defiance of proper trial protocol, which calls for random selection. They achieved apparently remarkable results, results never achieved before or after in such trials; the number of heart attacks for men in the treatment group was half the number of those in the control group. However, subsequent analysis of laboratory test records, conducted by independent scientists, showed that the Upjohn scientists had selected a larger number of individuals with familial hypercholesterolemia, an inborn error of cholesterol metabolism, for the control group.59 These people are much more prone to heart attacks than others, and many die young. By placing larger numbers of these people in the untreated control group, the Upjohn scientists manipulated the protocols so that the treatment group would fare well in comparison. Thus, they managed to achieve the results they wanted in the trial. Scientists can be very clever, as we have seen in the preceding chapter.
A similar trial to test the cholesterol-lowering agent clofibrate was conducted under the sponsorship of the World Health Organization in the 1970s. This trial involved some thirty thousand healthy, middle-aged European men. Researchers selected ten thousand with the highest cholesterol levels, treating half with clofibrate and half with a placebo. Over the course of five years, an equal number in both groups died of heart attacks. But the total number of deaths from all causes was one hundred twenty-eight in the clofibrate group, and only eighty-seven in the placebo group. Five years after the trial, the number of heart attack deaths was also larger in the treatment group.
But clofibrate became a top-selling drug for years, and it is still recommended in many countries. The reason? In the initial five-year trial, one hundred seventy-four of the people taking a placebo were reported to have suffered from a non-fatal heart attack, compared with one hundred thirty-one of those treated with clofibrate.60 Articles about the trial emphasized this fact, as did promotional literature for clofibrate. But the interpretation of what constitutes a non-fatal heart attack is somewhat subjective; even experienced physicians sometimes diagnose gastrointestinal symptoms or hiatus hernia as a mild heart attack. Given the bias we have seen in many of the studies discussed, it seems quite possible that investigators exaggerated the number of non-fatal heart attacks in the placebo group.
Consistently poor results with cholesterol-lowering drugs did not prevent the National Heart, Lung and Blood Institute from embarking on another massive trial with a new drug in the late 1970s, the Lipid Research Clinics Coronary Primary Prevention Trial (LRC trial). This time cholesterol levels were measured in nearly one half million middle-aged men, and the highest eight-tenths of one percent—about four thousand individuals—became participants in the actual trial. The drug tested was cholestyramine, which half of the men received; the other half took a placebo.
Blood cholesterol in the treatment group went down by over eight percent. Deaths from heart attacks occurred in 1.7 percent of this group, compared to 2.3 percent in the control group. Nonfatal heart attacks numbered ten percent, compared to just over eleven percent in the control group. These results were so close that they could have resulted by chance; they had no statistical significance. The trial was a flop.
But in the summary of the paper, the researchers used the concept of relative risk to make these trivial differences seem important. They compared the absolute number of incidents in the treatment group with the absolute number of incidents in the control group while ignoring the total number of men involved. Fatal heart attacks were said to have been lowered by thirty percent, and nonfatal heart attacks by nineteen percent.61 Furthermore, there were eleven deaths from other causes in the treatment group, compared with only four in the untreated group, but the researchers failed to mention this fact in the summary. If they had used the relative risk measurement for the side effects, they would have reported a one hundred seventy-five percent increase in deaths from other causes in the group receiving the drug.62
If the results of the trial were not due to chance, and if the millions of men in the United States with blood cholesterol as high as those in the trial received the same treatment, perhaps two hundred fewer heart attack fatalities would occur in the country each year. Yet in a 1990 letter that appeared in the Atlantic Monthly, Daniel Steinberg, chairman of the conference that initiated the National Cholesterol Education Campaign, cited the LRC trial and claimed that one hundred thousand lives could be saved each year according to “a large number of studies” with statistical significance.63 In direct contradiction, a few months earlier, the physician who had served as director of the LRC trial stated in a medical journal that the trial had not reduced the number of heart disease deaths, and that “further gains in life expectancy are unlikely.”64
The side effects experienced in the LRC trial are another story. Normally, the untreated control group receives a placebo that has no side effects. Symptoms unrelated to treatment are then assumed to occur in both the treatment and the control groups with equal frequency. The true percentage of side effects can then be calculated by subtracting the percentage of side effects in the control group from that in the treatment group.
But in the LRC trial the percentage of extremely unpleasant side effects experienced by both groups was alarmingly high. Over two-thirds of the men taking the drug had nausea, vomiting, bloating, abdominal pain and heartburn, and nearly half had constipation or diarrhea. Nearly half of those in the control group experienced similar symptoms, indicating that the placebo contained some sort of active compound (the researchers did not reveal what the placebo was).65 It is hard to believe that the use of an active placebo was inadvertent and not deliberate.
However, the treatment subjects did suffer more symptoms than the controls, and a greater number were admitted to hospitals for treatment of nervous system disorders. This finding is consistent with other studies in which treatment with cholesterol-lowering drugs resulted in a markedly higher incidence of suicide, depression and violent death. Although it is clear that cholesterol reduction adversely affects the central nervous system, this possibility received no mention in the LRC trial report.
The side effects of cholestyramine (the drug used in the LRC trial) are rarely mentioned. In his letter to the Atlantic Monthly, Steinberg wrote, “The drugs in current use for lowering cholesterol levels have remarkably few side effects and, to my knowledge, no fatal side effects.”
The relative risk statistical manipulations that researchers used in the LRC trial led to new levels of aggressiveness in the treatment of “high cholesterol.” Recall that only men with the upper 0.8 percent of cholesterol values were treated in the trial. Less than one percent of all people have familial hypercholesterolemia, the inborn error of cholesterol metabolism, but a substantial majority of the men treated in the LRC trial had this problem. A treatment that had proved to be of marginal use for these men even with statistical manipulation would now be extended to the general population.
The elderly would be treated, even though there was no evidence that treatment was of any use for them. Women would be treated, even though the LRC trial had not studied women, and all previous studies had indicated that high cholesterol is not a risk factor in women. Men with cholesterol only slightly above normal (by then defined as 200 mg/dl) would now be treated. All of these groups were targeted for treatment in the LRC trial report. Children, bypassed at the time, would be included later.66
Statins Everyone? The Cancer Risk - Page 11
One of my patients said to me recently, “Well, they used to have some problems with the cholesterol-lowering drugs, but the probles have been solved. The new drugs don’t have side effects and they work much better.”
New drugs that inhibit the body’s production of cholesterol and a number of other important substances were introduced in the late 1980s and have become the drugs of choice for cholesterol-lowering today. Known as the statins, they include Zocor®, Pravachol®, Lipitor®, Lescol® and Mevacor.® These are the new wonder drugs, supposedly free of serious side effects and marvelously effective, lowering cholesterol by as much as forty percent or more. Several large clinical trials have shown small but statistically significant reductions in the number of heart attack deaths and the number of nonfatal heart attacks. Overall mortality has been reduced as well with the drugs, because fewer people in the treatment groups died from stroke.
But ironically, the reports of the trials provide strong evidence that cholesterol levels do not matter and that the statins achieve their results through mechanisms other than lowering cholesterol. Here is the evidence:
The drugs were as effective for the elderly as they were for younger individuals, and as effective for women as for men; almost all earlier studies have shown that high cholesterol is not a risk factor for women and the elderly.
Statin treatment reduced the number of strokes; all studies have shown that high cholesterol is a negligible risk factor for stroke.
Patients who had suffered a previous heart attack gained some protection; previous studies have shown that high cholesterol is at most a weak risk factor for these individuals.
People with either high or low cholesterol gained some protection taking the statins, indicating that it was not the lowering of cholesterol that afforded protection.
Finally, it did not matter whether the cholesterol level was lowered a lot or a little; the risk of heart attack was lowered to the same degree in either case.67
The likely explanation for the effectiveness of the statins lies in the fact that they inhibit the production of cholesterol by inhibiting the body’s production of mevalonate. This substance is the precursor not only of cholesterol but also of a number of other important molecules, including those that regulate the activity of the smooth muscle cells lining the arteries, the platelets that control blood clotting, and substances that slightly reduce the inflammation involved in atherosclerosis. Several researchers have demonstrated that it is these and other statin effects that are responsible for their modest clinical success—and not their characteristic of lowering cholesterol.68
Whatever the science, the statins are popular, and more people than ever are now taking drugs to lower cholesterol. But the scientific literature exaggerates the benefits of the statins using the same statistical methods described above.
Keep in mind also that large and expensive trials for these drugs are paid for and conducted by the drug companies. Company money pays for the scores of meetings, conferences and workshops that accompany the trials, and company money pays the professionals to attend and speak at the meetings and do the field work for the trials.69
Tens of millions of healthy middle-aged and younger people are now targets for treatment with statins without being told that these drugs may cause cancer. In 1996, an article titled “Carcinogenicity of lipid lowering drugs” appeared in the Journal of the American Medical Association. A careful review showed that all of the statins caused cancer growth in rodents, and that the blood levels of the drug associated with cancer in rodents were close to those of patients treated with statins. Because the period between exposure to a carcinogen and the diagnosis of cancer is often ten or twenty years, the authors recommended that the drugs be used only for patients at very high risk for heart disease. They further cautioned individuals with a life expectancy of greater than ten to twenty years to avoid the drugs.70
This advice, of course, directly conflicts with advertising promotions now targeting healthy middle-aged and younger people with levels of cholesterol in the 200 mg/dl range.
The effect these medications may have on people over the next twenty years is a great unknown, but the early indications are not good. In one large clinical trial, twelve women in the treatment group got breast cancer during the trial, compared with only one in the control group. “These findings could be an anomaly,” the authors of the study wrote.71 But the difference was highly statistically significant. The package inserts for the statins detail various other side effects—liver problems, muscle wasting, neuropathy, all of which can make life miserable—but make no mention of cancer.
The published results of the first and the largest statin trial to date are even more disconcerting. Preliminary findings in the EXCEL trial, the Expanded Clinical Evaluation of Lovastatin, came out in 1991 after the first year of the trial. In fine print the authors revealed that there had been thirty-two deaths from all causes in the treatment group of six thousand six hundred individuals compared with only three in the placebo group of one thousand six hundred fifty individuals, or two-and-one-half times more deaths in the treatment group, a figure that would be highly significant if it extended throughout the trial.
But in over twenty reports about the trial published since then, none have mentioned the final outcome of the trial, and no further data about the increased mortality seen in the first year in the treatment group have been forthcoming.72
Failed Leadership -Page 12 of 12
The cholesterol hypothesis has had an enormous influence on the fate of millions of America's dairy farmers over the last fifty or sixty years. Most Americans no longer understand or want whole raw milk, butter and cheese, all rich in cholesterol, fat and crucial fat-soluble vitamins so lacking in most American diets. This unfortunate dietary change is due in large part to government policies promulgating the diet-heart idea while encouraging the production and marketing of large qualities of cheap, second-rate, reduced-fat milk and impeding the production and marketing of high-quality raw milk. If government policy toward small dairy farmers since around the end of World War II can be characterized as “Let them eat cake,” then the demonization of cholesterol and animal fats is the icing on the cake. We turn now to the story of the farms and the farmers.