QUICKLINKS AND VIEW OPITONS
Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 2: High Cholesterol Causes Heart Disease
Wednesday, November 12, 2014 10:35 am Email this article
“In our need to understand, to explain, and to treat, the temptation to impute causality to association is pervasive and hard to resist. It is the most important reason for error in medicine.”
— Petr Skrabanek and James McCormick , Authors of Follies and Fallacies in Medicine
Large and small percentages
Framingham is a small town near Boston, Massachusetts. Since the early-1950s a large number of Framingham citizens have taken part in a study surveying all factors that may play a role in the development of atherosclerosis and heart disease. Among other things their cholesterol was measured frequently.
After five years the researchers made an observation, which should become one of the cornerstones in the cholesterol issue. When they classified the citizens into three groups with low, medium and high cholesterol values they saw that in the latter group more had died from heart attacks than in the two other groups. A high cholesterol level predicted a greater risk of a heart attack, they said; high cholesterol is a risk factor for coronary heart disease.
The predictive value of blood cholesterol levels was confirmed in the greatest medical experiment in history, the Multiple Risk Factor Intervention Trial, also called MR.FIT. In that trial researchers measured the blood cholesterol of more than 300.000 American middle-aged men.
Six years later the director of MR.FIT, professor Jeremiah Stamler and his coworkers from Chicago asked how many of these men had died and from what. The participants were then divided into ten groups of equal size, so-called deciles, according to their cholesterol values. The first decile thus consisted of the tenth of the men with the lowest cholesterol, the tenth decile of the tenth with the highest cholesterol.
The researchers analysis showed that in the tenth decile four times more men had died of a heart attack than in the first decile. Professor Stamler’s team put it in another way: “the risk of dying from a heart attack with cholesterol above 265 mg/dl (6.8 mmol/l) was 413 percent greater than with cholesterol below 170.”
With statistics you can change black to white, or vice versa; as any politician will tell you. Four hundred and thirteen percent! A frightening figure.
But let us look at the real figures and not only at the percentages. How many men had, in fact, died from a heart attack?
The total number was 2258, or 0.6 percent of the more than 300,000 men investigated. We could also describe these results by saying that 99.4 percent did not die from a heart attack.
Among those with the highest cholesterol value (the tenth decile) 494, or 1.3 percent, died from a heart attack. Said in another way, 98.7 percent of those with the very highest cholesterol values were alive after six years.
Among those with the lowest values, the first decile, ninety-five men, or 0.3 percent, died from a heart attack, while the rest, 99.7 percent survived. Thus, the difference in numbers of death between the first and the tenth decile was only one percentage point (99.7% — 98.7%).
One percentage point doesn’t have the same alarming effect as Dr. Stamler’s 413 percent, but both figures are correct because 1.3 is 413 percent of 0.3.
The excess of deaths was most pronounced in the tenth decile. It should be remembered that almost all individuals with the rare, inherited abnormality called familial hyper?cholesterolemia must have been included in the tenth decile. These people have considerably higher cholesterol values than normal individuals and some of them have severe atherosclerosis and cardiovascular disease in early life. A little less than one percent of humanity have familial hyper?cholesterolemia or some other kind of genetic problem that interfere with cholesterol metabolism. This means that about ten percent of the tenth decile (10 x 1 percent) were abnormal in this respect. Thus, in a complicated way, the statistics demonstrated what we already knew—that patients with an inborn error of cholesterol metabolism have a greater risk of dying from heart disease.
There are more ways that risk factor statistics can be used to magnify trivial differences. Let us go back to the Framingham study.
Great or small differences?
To illustrate the association between blood cholesterol and the risk of dying from a heart attack the researchers from Framingham constructed an interesting graph, shown in figure 2A. Two bell-shaped curves are seen. The horizontal axis, or x-axis, represents levels of blood cholesterol whole the vertical or y-axis concerns the number of individuals.
The curves in figure 2A are called Gaussian or “bell” curves. When plotted in a diagram like figure 2A, all measurements in biology usually produce a Gaussian curve with a distinctive parabolic or bell shape that rises slowly from the baseline on each side and then rapidly increases in slope at the center. The total area under the curve gives the number of individuals investigated. If, for example, you graph the heights of a random number of people, short individuals will be situated in the area beneath the curve’s left slope, and tall individuals will be situated in the area beneath its right slope. When a bell curve is symmetrical, the mean value of the group lies at the top of the curve.
Figure 2A. The distribution of the participants in the Framingham project according to their initial blood cholesterol. The solid line represents 1378 individuals without coronary disease at follow-up; the broken line represents 193 individuals who had coronary disease at follow-up. Data from Kannel and others. 
Let’s take a careful look at figure 2A. The broken line represents the cholesterol level of all middle-aged men in the Framingham study, who sixteen years after the start of the project had suffered a heart attack. The unbroken line represents the cholesterol levels of the men who had not. The first group consisted of 193 individuals, the other of 1378, but the curves are of an equal size because the vertical y-axis gives percent instead of the number of individuals.
The curve representing the patients is slightly asymmetrical with a little hump far to the right. Otherwise, the two curves appear identical except that the curve of the coronary patients is placed a little to the right; their cholesterol values are approximately 5-10 percent higher than the values of those who remained free of cardiovascular disease.
Most people probably think that those who have a heart attack almost always have large amounts of cholesterol in their blood. The curves demonstrate however that the difference is marginal. In fact, the graph shows that almost half of those who had a heart attack had low cholesterol.
According to the diet-heart idea this small difference in blood cholesterol is one of the most important causes of atherosclerosis and coronary heart disease. But as we shall see, the difference may be due to several other factors.
No risk after forty-seven
Thirty years after the first cholesterol measurement in Framingham the researchers again asked themselves what had happened. This time, a few more of those with high cholesterol had died. I use the words few for a reason. On average one percent of all men with high cholesterol died each year during these 30 years. During the first ten years about a quarter of one percent died each year. As time passed the percentage that died each year naturally grew larger ad larger. Among those with the lowest cholesterol values only half as many died; and as in almost all earlier investigations women with low cholesterol died equally often as did women with high cholesterol.
But these figures concerned all causes of death. Nothing was said about heart mortality! (Why did the researchers from Framingham forget to tell about heart mortality, the main issue of the whole project?)
Now to the most interesting point. For men above forty-seven no difference was seen. Those who had low cholesterol at the age of forty-eight died just as often as those with high cholesterol.
Thus, the Framingham study showed that if you reach age forty-seven, it doesn’t matter whether your cholesterol is high or low! I have never met any believer in the diet-heart idea who has even raised an eyebrow when confronted with this astonishing fact.
Blood cholesterol is usually at its highest level at about the age of fifty. It is after this age that heart attacks usually appear, increasing in frequency year by year. After age fifty atherosclerosis also accelerates, but the first signs of atherosclerosis in the artery appear much earlier, between the ages of 20 and 30.
Atherosclerotic lesions are a kind of inflammation involving the smooth muscle cells, the elastic fibers and the white blood cells. In the early stages cholesterol may not be present at all. Much later, usually after age 50, cholesterol and various lipids may be deposited in the lesions, eventually resulting in the dangerous raised lesions.
With these facts in mind how do you explain that high cholesterol is dangerous at the age of 30, but not after 47? If high cholesterol produces atherosclerosis because its level in the blood is a little higher than usual, why is high cholesterol a risk factor at the age of thirty, where cholesterol is rarely found in he arteries, but not after 47, the period of life where most of arterial cholesterol is produced?
Furthermore, few die from a heart attack before the age of 48, and most of them who do die are diabetics or have a rare, genetic problem. More than 95 percent of all heart attacks occur in people older than 48. If cholesterol has importance only for the very few who have a heart attack before 48, why should the rest of us worry about high-fat food and blood cholesterol?
The Framingham findings are not a rare exception. High cholesterol has no importance in old Australians either, according to a study by Dr. L. A. Simons and his coworkers at St. Vincent’s hospital in Sydney. Similar findings were uncovered in a study by Dr. Peter Zimetbaum and his coworkers at the Albert Einstein College of Medicine in the Bronx, NY. They found that neither total nor LDL cholesterol predicted the risk of having a heart attack or any other cardiovascular disease in very old men. Curiously, the authors concluded that, “The findings of this study suggest that an unfavorable lipoprotein profile increases the risk for cardiovascular morbidity and mortality.”
Unfortunately this happens all too frequently. Researchers get a result that is contrary to the cholesterol hypotheses, and yet they write conclusions indicating hat their findings are in support. These misleading conclusions are most often written up in the summary of the papers, the only part of the paper that most doctors and researchers are likely to read. To find the contradictory results, you have to read the whole paper and meticulously study the tables.
In the elderly high cholesterol even seems to be protective. This was the surprising finding of Dr. Harlan Krumholz at the Section of Cardiovascular Medicine at Yale University and his coworkers. They followed 997 elderly men and women living in the Bronx, NY for four years. During a four year period about twice as many of the individuals with low cholesterol had a heart attack or died from one, compared to those with the highest cholesterol levels.
Let me return to the study of the Framingham group. Perhaps you think that the cholesterol campaign was cancelled after the results of he Framingham study came in. Not at all. The reason low cholesterol levels were associated with greater mortality, said the investigators, was that people with low cholesterol levels were dying of other diseases. But their results contradicted that explanation. Wrote the authors: “Those whose cholesterol had decreased by itself during these 30 years ran a greater risk of dying than those whose cholesterol had increased. To cite the report: For each 1 mg/dl drop of cholesterol there was an 11% increase in coronary and total mortality.”
Thus, not only total mortality but also coronary mortality had increased.
Now, stop for a moment! For many years we have been told how important it is to lower our cholesterol to prevent coronary heart disease. But the Framingham study demonstrated that if the cholesterol decreases by itself, the risk of dying increases.
Few people know about this alarming finding and the study is rarely discussed in the medical reviews of cholesterol and heart disease. Even worse, when the study is noted, it is cited as supporting the diet-heart idea! Consider the joint statement by the American Heart Association and the National Heart, Lung, and Blood Institute in their review entitled The Cholesterol Facts: “The results of the Framingham study indicate that a 1% reduction…of cholesterol [corresponds to a] 2% reduction in CHD risk.”
Please go back to the citation from the Framingham report. Yes, you are right. According to the original report mortality increased, and by 11% for each 1mg/dl reduction in blood cholesterol. But the review stated that mortality decreased.
Your next thought might be that the distinguished authors of the review referred to another of the numerous reports from Framingham, but they did not. And as we shall see, this was not their only “mistake.” For example, in 1987, the same authors published a new report concerning the 30 years of follow-up in Framingham. Without presenting anything other than complicated ratios and statistical calculations, and without referring to their previous report, they stated: “The most important overall finding is the emergence of the total cholesterol concentration as a risk factor for CHD in the elderly.”
Isn’t it strange that the cholesterol liner continues its voyage without any reactions from the passengers or crew? The few who have observed that the ship is leaning are calmed by the captain’s assurances that it has only struck an iceberg.
Rule with many exceptions
Most supporters of the diet-heart idea think that the increased risk of coronary heart disease is present at all cholesterol levels. This concept is of course pleasant to the drug producers, for it implies that almost everyone should be treated, including those with a normal cholesterol.
This is not true, however. In most studies, the increased risk is present only above a certain level. As a matter of fact, the relationship between the cholesterol level of the blood and the risk of coronary heart disease seems to be rather unsystematic. Women, for instance, should stop worrying immediately, because high cholesterol is not a risk factor for the female sex. Few words have been aimed at this peculiar fact in the vast literature on cholesterol. When it is mentioned at all, it is said that the female sex hormones protect against heart attacks.
In fact, it seems more dangerous for women to have low cholesterol than high. Together with his team Dr. Bernard Forette, a French researcher from Paris, France, found that old women with a very high cholesterol lived the longest. The death rate was more than five times higher for women with very low cholesterol. The French doctors warned, of course, against lowering the cholesterol in elderly women, but they could as well have warned against cholesterol lowering in any women, or, to be more precise, at all.
It is also notable that whereas high cholesterol has a slight association with increased risk for men in the US, it has no association for men in Canada. This conclusion was reached by Dr. Gilles Dagenais and his team in Quebec after having followed almost 5000 healthy middle-aged men for 12 years. They explained away their surprising finding away by assuming that more than twelve years were needed to see the harmful effect of high cholesterol; obviously they were ignorant of the result from the 30-year follow-up study results.
Neither is blood cholesterol important for those who already had a heart attack. For instance, Dr. Henry Shanoff and his team at the University Hospital of Toronto studied 120 men ten years after their recovery from a heart attack and found that those with low cholesterol had suffered a second one just as often as those with high cholesterol. Many others have confirmed their findings.
In Sweden, Professors Lars-Erik Böttiger and Lars A. Carlson at the Karolinska Hospital found that the risk of coronary heart disease was higher for men with the highest cholesterol, but the risk was considerably lower than in Framingham. They also found that if all kinds of vascular disease caused by atherosclerosis were considered the risk was not increased at all. Those with low cholesterol died as often from vascular disease as those with high.
And there are more exceptions, for instance the Maori people, who originally are Polynesians, but have migrated to New Zealand for several hundred years ago. Unlike the native Polynesians, Maoris often die from a heart attack, but they do it whether their cholesterol is low or high.
In Russia, a low cholesterol is associated with an increased risk of coronary heart disease. This was the surprising finding of Dr. Dmitri Shestov from the Russian Academy of Medical Sciences in St. Petersburg. Dr. Shestov and his colleagues, one of them was Professor Herman Tyroler from the Department of Epidemiology at the University of North Carolina, had also analyzed HDL and LDL cholesterol, the “good” and the “bad” cholesterol. They found that a low LDL cholesterol was also associated with an increased risk and this was not due to low levels of HDL-cholesterol. In fact, those with low LDL values had the highest HDL values.
Thus, a high cholesterol is dangerous for Americans but not for Canadians, Stockholmers, or Maoris, and a low cholesterol is dangerous for Russians. A high cholesterol is dangerous for men, but not for women; it is dangerous for healthy men, but not for coronary patients; and it is dangerous for men of thirty, but not for those of forty-eight and may even be beneficial for older people. Such discrepancies indicate that the association between high cholesterol and coronary heart disease is not due to simple cause and effect. The most likely interpretation is that a high cholesterol is not dangerous by itself but a marker for something else.
Many scientists who are critical of the diet-heart idea still have the impression that an increased level of cholesterol in the blood may be dangerous just because of its association with coronary mortality. Few know that the association is unsystematic and even rather weak. And even if the association had been both systematic and strong, this would not prove that it is the high cholesterol level itself that causes atherosclerosis or heart disease. There are at least five other plausible explanations for the higher cholesterol of patients with coronary heart disease.
Guilt by association
Familial hypercholesterolemia is one of them. Individuals suffering from this disease run a greater risk of dying early from a heart attack and they also have a raised cholesterol level. It is a widespread dogma that the increased cholesterol level, by promoting atherosclerosis, is the direct cause of their troubles. But as I shall discuss later it is questionable if the vascular changes seen in familial hyper-cholesterolemia is the same as atherosclerosis.
Smoking generates a slight increase of the blood cholesterol but may induce heart disease by several other mechanisms, for instance by producing many free radicals. Smoking may induce a heart attack and an elevated cholesterol level.
Being overweight increases blood cholesterol a little, and weight reduction lowers it a little. Excess weight means an excess burden to the heart. Excess weight may induce a coronary and an elevated cholesterol level.
High blood pressure is also associated with changes of blood cholesterol. Hypertension, untreated or treated, is seen in about a third of all individuals with a cholesterol level above 260 mg/dl but only in 15-20 percent of those with a cholesterol less than 220 mg/dl. High blood pressure, or rather the underlying cause, such as stress, may provoke a heart attack and raise blood cholesterol.
But the sharpest rise in cholesterol is seen as a result of emotional stress. An academic exam, blood sampling or surgery, conflicts at work or at home, loss of a spouse or a close friend, and various types of performance demands, have been found to increase the cholesterol level by ten to fifty percent. Psychological stress may provoke a heart attack (for instance by spasm of the coronary vessels) and an elevated cholesterol level. A likely explanation is that during stress more cholesterol is produced by the liver because cholesterol is used in the manufacture of various stress hormones.
That a high cholesterol is a risk factor for heart disease may have other explanations, but none of them are ever discussed in the papers written by the proponents of the diet-heart idea.
“Look at Finland and Japan”
Perhaps you will ask why it is so scary that death from a heart attack is more common in Western societies, since dying from a heart attack may not be too bad. After all, most of us prefer to die quickly, without spending many years in a nursing home, crippled or senile. And remember that coronary heart disease is a disease of old age. In fact, on average, those who die from a heart attack have lived just as long as other people. Nevertheless, since the association between average blood cholesterol and death rates from coronary disease in various countries has been used as an argument for the diet-heart idea, let us look at some of the facts.
In the Seven Countries study Keys pointed to the association between blood cholesterol and heart mortality. The correlation is obvious, Professor Keys wrote and he illustrated his words with a graph.
It is not apparent from Keys’s paper how the graph is constructed, but it is possible to draw a graph oneself by using the numbers from his tables. I have drawn such a graph choosing the hard data, meaning the number who died from coronary heart disease in the various districts, and compared them with the blood cholesterol values (figure 2B). If the diet-heart idea is correct heart attacks should, of course, be rare in the districts where cholesterol was low and common where it was high. But as seen from my chart, they were very far from that. Oddly, this important chart is not included in Keys’s paper, although his paper is loaded with more or less relevant graphs.
Figure 2B. Five year mortality of coronary heart disease and mean blood cholesterol in 15 populations in the Seven-Countries study. The populations that have been connected with lines are from the same country; see text. The figure is constructed from table data in the report of the Seven-Countries study. Note the great differences in coronary mortality at similar blood cholesterol levels.
There is a notable scattering of the points in the figure. Note for instance that in the districts where fatal heart attacks were uncommon (less than 100 per 100.000) the cholesterol levels vary between the lowest and the second highest value. It is difficult from this figure to see that the number of heart deaths and the level of blood cholesterol are related. Possibly they are statistically related, but we should be skeptical about correlations that depend on just one or two observations. For instance, cover the point labeled Karelen with your hand and the slight impression of an association disappears completely.
Any suggestion of an association also disappears if you look at each country individually (symbols representing various populations in the same country are connected with a line). In Crevalcore, Italy, the number of deaths from heart disease was 2.5 times greater than in Montegiorgio, Italy, although the average blood cholesterol was identical. In Slavonia three times more died from heart disease than in Dalmatia, although the mean cholesterol in Slavonia was only insignificantly higher. In Finland people living in Karelia died five times more often from a heart attack than people living in the area of Turku although blood cholesterol differed only little. And finally, on the Greek island Corfu people died five times more often from a heart attack than on nearby Crete, although their cholesterol was lower.
If you go to the tables of Ancel Keys’s paper again and do a little calculating you will discover another surprising finding: no correlation was found between the diet and the major electrocardiographic findings at entry. Considering that all electrocardiograms were analyzed in the American study center this finding should carry more weight than the correlation with the clinical diagnosis or the diagnosis on the death certificate, settled as they were on location by various physicians with varying competence and diagnostic habits.
But in his conclusion Ancel Keys wrote that the only factor which could explain the great differences between the number of heart attacks in the sixteen areas was the cholesterol level of the blood. And again and again the Seven Countries study is mentioned as a proof of the diet-heart idea: Look at Finland and Japan.
Instead, look at the figure once again. It is almost heart-breaking. Eager to prove his hypothesis, Keys unintentionally covered up one of the most interesting tracks. I suspect that many of you already have asked the question that Keys had the opportunity to answer twenty years ago. It is doubtful that the question can be answered so many years later. But, to ask it anyway, what is the factor that protects the inhabitants of Dalmatia, Montegiorgiu, Turku and Crete from coronary heart disease and that is absent in Slavonia, Crevalcore, Karelia and on Corfu?
Or, to turn the question around: why does death from coronary heart disease occur 3-5 times more often in the latter areas? We can blame none of the well-known risk factors because Keys found that they were evenly present in each pair of areas. If Ancel Keys and his coworkers had concentrated all their efforts in these eight places, if they had observed, investigated, questioned and turned every stone upside down they might perhaps have found something helpful to mankind.
This is the most tragic aspect of the cholesterol folly. Interesting side tracks are left unexplored, observations which do not fit with the idea are put aside, and any opportunity for a new discovery is allowed to slip by.
The Japanese Paradox
In Japan cholesterol levels are very low, and few people die from heart disease. This has been known for a long time and was confirmed by the Seven Countries study. Professor Keys also found that Japanese emigrants to the mainland USA had high blood cholesterol and died almost as often from heart attacks as Americans did, while the figures for Japanese emigrants to Hawaii lay somewhere in between.
Professor Keys was convinced that the difference was caused by the food, which in Japan was lean, while on Hawaii, and especially in the continental US, it was rich in animal fat.
As usual Keys had no other explanation. And what he did not mention was that while coronary mortality increased after the Japanese had migrated to the US, stroke mortality decreased just as much and total mortality decreased much more.
There is an alternative explanation to the increased coronary mortality after migration from Japan to the US.
As I mentioned earlier, calories from animal fat are usually expensive, and such food is therefore mostly consumed in rich countries, such as the Western, industrialized nations.
Calories from carbohydrates and vegetable oils are cheaper and such food therefore predominates in poor countries with a low degree of industrialization and a low standard of living. When Ancel Keys gathered his data in the early 1960s Japan belonged to this category. It was still a poor nation, successfully recovering from war, not the rich nation triumphant in industry that we know today.
Immigrants from a poor country are exposed in their new, richer country to many other things besides high-fat food. A multitude of factors in the Western environment or lifestyle may adversely effect the heart and the blood vessels, such as less physical activity, more stress, and more environmental and industrial pollutants.
In his doctoral thesis about coronary heart disease in Japanese immigrants, British physician Dr. Michael Marmot presents some interesting insights into the relationship between blood cholesterol levels and social factors, eating habits and lifestyle. Dr. Marmot demonstrated that it was not the food that raised the cholesterol of the Japanese immigrants, nor high cholesterol values that increased their risk of coronary heart disease. He found that if they maintained their cultural traditions, they were protected against heart attacks, even though their cholesterol increased as much as in Japanese immigrants who adopted a Western lifestyle and who died from heart attacks almost as often as did native-born Americans. The most striking of Dr. Marmot’s findings was that emigrants who maintained the Japanese traditions but preferred high-fat American food ran a smaller risk of heart disease than those who became accustomed to the American way of life but ate the lean, Japanese food.
Thus, according to Dr. Marmot’s study, there is something in the Japanese way of living that protects against coronary heart disease, and it is not the food.
Dr. Marmot himself points to the traditional Japanese culture, which is still a major factor shaping life in present-day Japan. In particular, the Japanese place great emphasis on group cohesion, group achievement, and social stability. Members of the stable Japanese society enjoy support from other members of their society and thus are protected from the emotional and social stress that Marmot believes to be an important cause of heart attacks. The Japanese traditions of togetherness contrast dramatically with the typical American emphasis on social and geographic mobility, individualism, and striving ambition.
Ignoring embarrassing data
We do not know whether Dr. Marmot’s explanation is correct or not. However, if his findings are correct, the diet-heart idea must be wrong. But Dr. Marmot’s results, as well as the many other embarrassing contradictory findings, have been ignored in the official reviews.
In 1979 for instance, the American Health Foundation organized an international conference with the aim of finding “the optimal cholesterol level.” In a later step, the level was to be lowered in the countries where the conference participants thought it was too high. The written report from this conference presented a meticulous account of Dr. Marmot’s results including detailed information about the Japanese food, cholesterol levels and risk of heart disease. But Marmot’s message (and the epidemiological data on which it was founded), was ignored.
Let’s look at a few more examples of the researchers sweeping contradictory findings under the rug. A large review written in 1984 by Dr. William Kannel, head of the Framingham project, and his colleagues, stated that there is a strong association between population means of total cholesterol and CHD incidence, and here Dr. Kannel refers to Ancel Keys’s Seven Countries study. (Total cholesterol simply means cholesterol. The term total cholesterol is used in texts where subfractions such as LDL- and HDL-cholesterol are also mentioned.)
The largest review, Diet and Health, published in 1989, comes from the prestigious National Research Council and concluded as follows:
“Epidemiological findings among populations and for individuals within populations consistently indicate a strong, continuous, and positive relationship between TC [Total Cholesterol] levels and the prevalence and incidence of, as well as mortality from, atherosclerotic CHD.”
Many supporters of the diet-heart idea seem to have stopped thinking critically. Perhaps, as members of a worldwide alliance that includes many distinguished researchers, they have become overconfident, too willing to assume that an idea must surely be valid if great numbers of people endorse it.
But if they were familiar with the way science has progressed through the centuries, they would know that the truth cannot be ruled out by the majority or decreed by consensus.
Another myth: “good” and “bad” cholesterol
Cholesterol is a peculiar molecule. It is often called a lipid or a fat, but the chemical term for a molecule such as cholesterol is alcohol, although it doesn’t behave like alcohol. Its numerous carbon and hydrogen atoms are put together in an intricate three-dimensional network, impossible to dissolve in water. All living creatures use this indissolubility cleverly, incorporating cholesterol into their cell walls to make cells waterproof. This means that cells of living creatures are able to regulate their internal environment undisturbed by chemical changes in the surrounding milieu. The fact that cells are waterproof is especially critical for normal functioning of nerves and nerve cells. Thus, the highest concentration of cholesterol in the body is found in the brain and other parts of the nervous system.
Because cholesterol is insoluble, it circulates in the blood inside spherical particles composed of fats (lipids) and proteins, the so-called lipoproteins. Lipoproteins are easily dissolved in water because their outside is composed mainly of water-soluble proteins. The inside of the lipoproteins is composed of lipids, and here we have room for water-insoluble molecules like cholesterol. Like submarines, lipoproteins carry cholesterol from one place in the body to another.
These submarines, or lipoproteins, are categorized according to their density. The best known are HDL (High Density Lipoprotein), and LDL (Low Density Lipoprotein).
The main task of HDL is to carry cholesterol from the peripheral tissues, including the artery walls, to the liver. Here it is excreted with the bile, or used for other purposes, for instance as a starting point for the manufacture of important hormones.
The LDL submarines mainly transport cholesterol in the opposite direction. They carry it from the liver, where most of our body’s cholesterol is produced, to the peripheral tissues, including the artery walls. When cells need cholesterol, they call for the LDL submarines, which then deliver cholesterol into the interior of the cells.
Between 60 and 80 percent of the cholesterol in the blood is transported by LDL and is called “bad” cholesterol. Only 15-20 percent is transported by HDL and is called “good” cholesterol. A small part of the circulating cholesterol is transported by other lipoproteins.
You may ask why a natural substance in our blood with important biologic functions, is called “bad” when it is transported from the liver to the peripheral tissues by LDL, but “good” when it is transported in the other direction by HDL. The reason is that a number of follow-up studies have shown that a lower-than-normal level of HDL-cholesterol and a higher-than-normal level of LDL-cholesterol are associated with a greater risk of having a heart attack, and conversely, that a higher-than-normal level of HDL-cholesterol and a lower-than-normal LDL-cholesterol are associated with a smaller risk. Or, said in another way, a low HDL/LDL ratio is a risk factor for coronary heart disease.
By now you know that a risk factor is not necessarily the same as the cause, and that something may provoke a heart attack and at the same time lower the HDL/LDL ratio. Let us have a look at some factors known to influence this ratio.
The cholesterol ratio caper
As mentioned above, people who reduce their body weight also reduce their cholesterol. A review of 70 studies showed that, on average, weight reduction lowers cholesterol by about 10 percent, depending on the amount of weight loss. Interestingly, it is only cholesterol transported by LDL that goes down; the small part transported by HDL goes up. In other words, weight reduction increases the ratio between HDL and LDL-cholesterol.
An increase of the HDL/LDL ratio is called favorable by the diet-heart supporters; cholesterol is changed from “bad” to “good.” But is it the ratio or the weight reduction that is favorable? When we become fat, other harmful things occur to us. One is that our cells may become less sensitive to insulin, so that some of us even develop diabetes. And people with diabetes are much more likely to have a heart attack than people without diabetes, because atherosclerosis and other vascular damage may occur early in diabetics, even in those without lipid abnormalities. In other words, overweight may increase the risk of a heart attack by mechanisms other than an unfavorable lipid pattern, while at the same time overweight lowers the HDL/LDL ratio.
You may also recall that smoking increases cholesterol a little. Again, it is LDL-cholesterol that increases, while HDL-cholesterol goes down, resulting in an unfavorable HDL/LDL ratio. What is certainly unfavorable is chronic exposure to the fumes from H paper and tobacco leaves. It should be obvious that instead of considering a low HDL/LDL ratio as bad, we should consider the possibility that smoking in and of itself is bad. Smoking may provoke a heart attack and, at the same time, lower the HDL/LDL ratio.
Exercise decreases the so-called bad LDL-cholesterol and increases the so-called good HDL-cholesterol. In well-trained individuals the good HDL is increased considerably. In a comparison between distance runners and sedentary individuals, Dr. Paul D. Thompson and his team from Providence, Rhode Island, found that the athletes on average had a 41 percent higher HDL-cholesterol level. Most population studies have shown that physical exercise is associated with a lower risk of heart attacks and a sedentary life with a higher risk. A well-trained heart is better guarded against obstruction of the coronary vessels than a heart always working at low speed simply because the vascular channels in the well-trained heart are broader; remember the wide coronary arteries of the Masai people who ran all day long after their cattle in Kenya. A sedentary life may predispose people to a heart attack and, at the same time, lower the HDL/LDL ratio.
Univariate and multivariate
Thus, the risk of having a heart attack is greater than normal for people with high LDL-cholesterol, but so is the risk for fat, sedentary and smoking individuals. And since such individuals usually have elevated levels of LDL cholesterol, it is, of course impossible to know whether the increased risk is due to the previously mentioned risk factors (or to risk factors we do not yet know) or to the high LDL-cholesterol. A calculation of the risk of high LDL-cholesterol that does not consider other risk factors is called a univariate analysis and is, of course, meaningless.
To prove that high LDL-cholesterol is an independent risk factor, we should ask if fat, sedentary, smoking individuals with a high LDL-cholesterol level are at greater risk for coronary disease than fat, sedentary, smoking individuals with low or normal LDL-cholesterol.
Using complicated statistical formulas, it is possible to do such comparisons in a population of individuals with varying degrees of the risk factors and varying levels of LDL-cholesterol, a so-called multivariate analysis. If a multivariate analysis of the prognostic influence of LDL-cholesterol also takes body weight into consideration, it is said to be adjusted for body weight.
A major problem with such calculations is that we know a great number of risk factors, but not all of them. Another problem is that the data generated by these and other complicated statistical methods are almost impossible for most readers, including most doctors, to comprehend. For many years researchers in this area have not presented primary data, simple means, or simple correlations. Instead, their papers have been salted with meaningless ratios, relative risks, p-values, not to mention obscure concepts such as the standardized logistic regression coefficient, or the pooled hazard rate ratio. Instead of being an aid to science, statistics are used to impress the reader and cover the fact that the scientific findings are trivial and without practical importance. Nevertheless, let us have a look at some of the studies.
The “good” one
Publications almost beyond counting have studied the prognostic value of the “good” HDL cholesterol. The reason is, of course, that it is hard to find any prognostic value. If HDL-cholesterol had a heart-protecting effect of real importance, it would not be necessary to use the taxpayers’ money to demonstrate the effect again and again in expensive studies. To be brief I shall tell you only about a few of the largest studies.
In 1986 Dr. Stuart Pocock and his team from London and Birmingham, England published a report concerning more than 7000 middle-aged men in 24 British towns. The men had been followed for about four years after a detailed analysis of their blood. During this period 193 of the men had suffered a heart attack. As in most previous studies, these men had on average a lower HDL-cholesterol at the beginning than the men who did not have a heart attack. The mean difference between the cases and the other men was about 6 percent. This difference was small of course, but thanks to the large number of individuals studied it was statistically significant.
But this was a univariate analysis and as mentioned, the difference could therefore be explained in many ways. A multivariate analysis adjusted for age, body weight, cigarette smoking and non-HDL-cholesterol reduced the difference to an insignificant 2 percent. This means that those who had suffered a heart attack had a lower HDL-cholesterol mainly because they were older, fatter and smoked more than those who had not had a heart attack.
The British scientists compared their findings with those of six other studies. In five of them the differences were just as small. Only one study found a considerable difference, but it included 39 individuals only and thus was highly susceptible to bias. Dr. Pocock and his colleagues concluded that a low HDL-cholesterol level is not a major risk factor for coronary heart disease.
Their results were challenged in 1989 by nine American scientists headed by Dr. David Gordon at the National Heart, Lung and Blood Institute. They had analyzed the predicative value of HDL-cholesterol in four large American studies, including more than 15,000 men and women. They thought that the British scientists had used an incorrect way to adjust their figures. Using another formula, the American researchers wrote, showed HDL-cholesterol to be a much better predictor.
But in one of the four studies the number of fatal heart attacks was identical in the first and second HDL tertile (individuals were classified into three groups, or tertiles, according to their HDL-cholesterol). In one of the studies the number of fatal cases was identical in the second and the third tertile, and in one study more deaths were seen in the third tertile (those who had the largest amount of the “good” cholesterol) than in the second tertile. And these figures were the unadjusted ones.
After adjustment the differences were even smaller. In three of the four studies, the differences lost statistical significance. And remember that the figures were not adjusted for physical activity, not to mention the risk factors we do not know yet.
Dr. Pocock and his colleagues returned with a new analysis later the same year, now using the same way of analyzing as had Dr. Gordon and his colleagues. At that time the participants in the study had been followed for 7.5 years and a total of 443 heart attacks had occurred.
This time a difference was noted between the HDL-cholesterol of the heart patients and the others. The difference was small but statistically significant, even after adjustment. However, the largest difference was noted for total cholesterol. The authors therefore concluded that a determination of HDL-cholesterol may be of marginal additional value in screening and in intervention programs for risk of coronary heart disease. They could also have added that they did not adjust for all risk factors so that the difference could as well be due to the heart patients being, for instance, more stressed or less active physically than the others.
And even if the difference had remained after adjustment of all the known risk factors the crucial question is if HDL-cholesterol has any importance whatsoever. From what you have read till now and especially from the chapters to come, you will realize that there is little or no evidence, that blood cholesterol plays a role in coronary heart disease. If total cholesterol, a better predictor than HDL-cholesterol, is unimportant, how could HDL-cholesterol be important?
I am tempted to discuss the many other studies which did not find HDL-cholesterol a good predictor. But to avoid boring you, I shall mention only the 24-year follow-up of the Finnish group in the Seven Countries study, because this is one of the longest follow-up studies of HDL-cholesterol. A total of 518 healthy men from three areas in Finland were followed. Table 2A gives the number of fatal heart attacks and the starting average HDL-cholesterol in each area.
Table 2A. Mean HDL-cholesterol in three Finnish areas and coronary mortality 24 years later.
||Mortality from Coronary
heart disease Per 1000 men 24 years later
If HDL cholesterol were good for the heart, the smallest number of men should have died from heart attacks in North Karelia, where HDL-cholesterol was the highest. Instead, the number was twice as large as in Helsinki.
In their paper, Keys and his co-authors mentioned two equally large and long-lasting studies that also found lower HDL-cholesterol in the healthy survivors than in those who had died from coronary disease. But in the summary of the paper they wrote: “These 24-year findings are not necessarily in conflict with reports in the literature on an inverse relationship between coronary heart disease incidence and HDL cholesterol.”
It is fortunate that low HDL-cholesterol itself does not increase the risk of a heart attack, because the prudent diet has a surprising effect on the HDL-cholesterol level. A French study by Dr. Frédéric Fumeron and his colleagues in Paris and Lille investigated the effect of two different diets on 36 healthy individuals. One diet contained 70 grams of butter, the other 70 grams of sunflower margarine; otherwise the diets were similar. Each individual ate both diets for three weeks; half of them started with the butter diet, half with the “prudent” margarine diet.
As in many previous studies, analysis of the blood lipid levels before and after each period showed that the “prudent” diet lowered blood cholesterol. But it also lowered the “good” HDL-cholesterol, especially two of its subfractions called HDL-2 and LpA-I. Other studies have shown that these subfractions are especially “good.” The authors also reviewed seven other studies with similar results.
The “bad” one
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 atherosclerosis and CHD. These words you will find in the large review Diet and Health.
A scientific review is, like this book, an analysis of what has been done and what has been written about a certain subject. Reviews usually do not present observations or experiments performed by the authors themselves. Reviews help researchers by sparing them the tedious work of seeking the primary observations themselves in the library or in the electronic data bases. Furthermore, in their papers researchers can refer to a few reviews instead of to a large number of original works. But, of course, the researcher must be sure that the reviews are complete and correct and that they give a balanced view.
Reviews by distinguished scientific bodies are supposed to meet such standards. Therefore, you are probably wondering how the authors of Diet and Health had reached their conclusion about LDL-cholesterol. I wondered too, when I started to untangle the HDL-LDL issue, because extensive reading had not yet given me the answer.
The fact is that very few analyses of LDL-cholesterol have been published. For example, in the hundreds of reports from the Framingham study very little is mentioned about LDL-cholesterol. An odd fact because all participants had this cholesterol fraction measured at the start and again later in the study.
Diet and Health is the official, most authoritative and supposedly most reliable review from the National Research Council in Washington. I was confident that its highly qualified authors would have the answer. What was their evidence? Upon which observations or experiments did they base their statements about the dangers of LDL-cholesterol?
Diet and Health cites four publications. First, inn 1973 Dr. Jack Medalie and his team at the Tel Aviv University in Israel published a five-year follow-up study of 10,000 Israeli male government and municipal employees. Among a large number of factors relevant to the study of coronary heart disease they had measured total and LDL-cholesterol. According to Diet and Health LDL cholesterol has the strongest relationship to risk of heart disease. However, the Israeli study did not support these words, because total cholesterol, not LDL-cholesterol, had the strongest relationship to risk of coronary disease.
The second paper claimed by the Diet and Health authors to support the idea about the dangerous LDL cholesterol, was a 1977 report from the Framingham Study. This study concerned HDL cholesterol, however. Only logistic regression coefficients (a statistical concept unknown to most doctors) for coronary disease on LDL-cholesterol were given, and one of the conclusions of the paper was that LDL-cholesterol… is a marginal risk factor for people of these age groups (men and women above 50 years). Some of the coefficients were indeed low. For women above the age of 70 it was negative, which means that women at that age ran a greater risk of having a heart attack if their LDL-cholesterol was low than if it was high. Thus, there was no support either from that paper.
Also the third paper concerned HDL-cholesterol only. Thus, no support either.
The fourth reference was to the National Cholesterol Education Program, which produced another large review without original data. One of its conclusions was that, “a large body of epidemiological evidence supports a direct relationship between the level of serum total and LDL-cholesterol and the rate of CHD.” I became excited, thinking, “At last, the evidence.”
The large body of evidence was to be found in three references. The first one was another large review without original data, Optimal resources for primary prevention of atherosclerotic disease, with Dr. Kannel as the first author. I shall return to their review.
The next reference was yet a large review, but nothing in that review was said about the matter.
The last reference was a report from the Honolulu Heart Study. The conclusion of that paper was that, “both measures of LDL-cholesterol were related to CHD prevalence, but neither appeared to be superior to total cholesterol.”
Before I discuss Dr. Kannel’s review I shall mention another conclusion in the National Cholesterol Education program: “The issue of whether lowering LDL-cholesterol levels by dietary and drug interventions can reduce the incidence of CHD has been addressed in more than a dozen randomized clinical trials.” This is a most misleading statement because at that time, in 1988, only four such trials had included an LDL-cholesterol analysis, and only in one of them the number of heart attacks was lowered significantly.
Let me now return to the review by Dr. Kannel and colleagues, the one used as evidence by the authors of the National Cholesterol Education Program, which in turn was used as evidence by the authors of Diet and Health. Almost nothing was written about LDL-cholesterol in Dr. Kannel’s review except for the following: “Longitudinal studies within populations show a consistent rise in the risk of CHD in relation to serum total cholesterol and LDL-cholesterol at least until late middle-age.”
A little more cautious conclusion than in Diet and Health, it may seem, but even for this prudent statement the evidence was weak. References to six studies were given. In two of them LDL-cholesterol was not analyzed or mentioned at all. In two reports LDL-cholesterol was only correlated to the prevalence of heart disease, which means that it was not a longitudinal study. In one report two tables was aimed at the subject (tables 8 and 9) and showed that the predictive power of LDL-cholesterol was statistically non-significant. In one study LDL-cholesterol was predictive for heart disease, but only for men between 35 and 49 and only for women between 40 and 44.
In conclusion, the “large body of evidence” can be reduced to one single study, which showed a predictive value for LDL-cholesterol but for a few age groups only. The only valid conclusion therefore is that LDL-cholesterol is neither centrally nor causally important, it has not the strongest and most consistent relationship to risk of CHD, it has not a direct relationship to the rate of CHD, and it has not been studied in more than a dozen randomized trials.
You can read more about the above in one of my papers.
Here are links to the other chapters in the book.
- Cholesterol Myths by Uffe Ravnskov, MD, PhD: Forward to Book by Michael Gurr, PhD
- Cholesterol Myths by Uffe Ravnskov, MD, PhD: Author’s Foreword
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Introduction: The Diet-Heart Idea: A Die-Hard Hypothesis
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 1: High-Fat Foods Cause Heart Disease
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Triglycerides
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 2: High Cholesterol Causes Heart Disease
- Cholesterol Myths by Uffe Ravnskov MD: Familial hypercholesterolemia—not as risky as you may think
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 3: High-Fat Foods Raise Blood Cholesterol
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 4: High Cholesterol Blocks Arteries
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Myth 5: Animal Studies Prove the Diet-Heart Idea
- Cholesterol Myths by Uffe Ravnskov MD, PhD: Cholesterol lowering in children
- Cholesterol Myths by Uffe Ravnskov MD: Myth 6: Lowering Cholesterol Will Lengthen Your Life (Part 1)
- Cholesterol Myths by Uffe Ravnskov MD: Myth 6: Lowering Cholesterol Will Lengthen Your Life (Part 2)
- Cholesterol Myths by Uffe Ravnskov MD PhD: Myth 7: The Statins — Gift to Mankind (Part 1)
- Cholesterol Myths by Uffe Ravnskov MD PhD: Myth 7: The Statins — Gift to Mankind (Part 2)
- Cholesterol Myths by Uffe Ravnskov MD PhD: “The most exact data base”—the screenee
- Cholesterol Myths by Uffe Ravnskov MD PhD: Myth 8: Polyunsaturated Oils are Good for You
- Cholesterol Myths by Uffe Ravnskov MD PhD: Dr. Ornish and The Lifestyle Heart trial
- Cholesterol Myths by Uffe Ravnskov MD PhD: Myth 9: The Cholesterol Campaign is Based on Good Science
- Cholesterol Myths by Uffe Ravnskov MD PhD: Insider Insight
- Cholesterol Myths by Uffe Ravnskov MD PhD: Myth 10: All Scientists Support the Diet-Heart Idea
- Cholesterol Myths by Uffe Ravnskov MD PhD: Epilogue
- Cholesterol Myths by Uffe Ravnskov MD PhD: References
This chapter is from the book
The Cholesterol Myths: Exposing the Fallacy that Saturated Fat and Cholesterol Cause Heart Disease
by Uffe Ravnskov, MD, PhD.
Dr. Ravnskov has given me the permission to share this version of his book to help educate the world about the cholesterol campaign.
Information about Uffe Ravnskov, MD, PhD is posted here.
More information about Cholesterol Myths is posted on his website here.
Dr. Ravnskov posted his book for free here.
Uffe Ravnskov, MD, PhD is the founder of The International Network of Cholesterol Skeptics (THINCS.org) which can be found here.
Articles on the same subject can be found here:
Please feel free to share your comments about this article.
© Copyright 2003-2021 - Larry Hobbs - All Rights Reserved.