4. Atherosclerosis and coronary heart disease have nothing to do with the diet
An excerpt from my previous
book The Cholesterol Myths (out of print).
National consensus committees in many countries have declared that atherosclerosis and coronary heart disease can be prevented by an appropriate diet. Although the scientific evidence for this message is surprisingly meager, if present at all, it has gained status as established wisdom.
The definition of the prudent diet has changed considerably with time. Initially, it was considered important to reduce dietary fat of all kinds. This advice was based on a review paper by Ancel Keys (49), the main designer of the so-called diet-heart idea. In his review Keys presented a perfect curvilinear correlation between the mortality from coronary heart disease and the consumption of fat in six countries, but his curve was based on a selection of countries that fit his hypothesis and it has not been confirmed in studies including many more countries (50).
The prudent diet was redefined a few years later based on a new study by Ancel Keys, Seven Countries (51). According to that study the total fat intake was unimportant; heart mortality in these seven countries was best predicted by the intake of saturated fat . But within each country no association was seen. In Finland and Greece for instance, heart mortality in two districts varied with a factor five and seven, respectively, despite similar diets and other risk factors. Furthermore, no correlation was found between the diet and the major electrocardiographic findings. Considering that all electrocardiograms were analysed in the American study center this finding should carry more weight than the correlation with the clinical diagnosis, settled as it was by local doctors with varying competence and diagnostic habits.
The seven countries were admittedly selected by Keys. Such selection may be helpful to illustrate an idea at a preliminary stage, but a proof of causality demands random data. In more recent studies, including many more countries, the association was weak, absent, or inverse (52).
Conclusions from associations between national food consumption data and disease should be drawn with care. Most important, assumed intake of animal fat may be falsely high in prosperous countries, because available fat is not the same as fat eaten, but includes fat consumed by pet animals, fat discarded in the kitchen or on the plate, and fat which has never reached the consumer. With all certainty, these amounts are larger in prosperous countries.
The finding that an increased intake of polyunsaturated fatty acids, also called PUFA, can lower the serum cholesterol concentration in laboratory experiments has led to the belief that they would lower the risk of coronary heart disease also. Consequently, an increased intake of PUFA has been advised as an important part of the prudent diet. Initially, no limit was put to such intake, but by the years the limit has been lowered successively. Most recently, an upper limit of only 7 cal% was recommended because a high intake of PUFA promotes cancer, infections and testicular damage in rats (53). The average food intake in most western countries includes that amount of PUFA.
There is little evidence that an increased intake of PUFA protects against heart attacks. In Seven Countries intake of PUFA was not associated with heart mortality, and studies of patients with coronary heart disease have shown that if anything, they eat more PUFA than do healthy individuals, see below.
If heart attacks are caused by eating too much animal fat or saturated fat, a rising intake in a population should of course be followed by more heart attacks and a decreasing intake by fewer attacks. No consistent pattern has been found, however. In a few countries the changes have followed each other and the data from these countries have been used eagerly to support national diet counceling. But in many countries fat consumption has changed whereas heart mortality has not, or vice versa; in many countries they have even changed at opposite directions (54).
In Switzerland, for instance, coronary mortality decreased after World War II. During the same period intake of animal fat increased by 20 per cent (55).
In England, the intake of animal fat has been relatively stable since at least 1910 while the number of heart attacks increased ten times between 1930 and 1970 (56).
In the US coronary mortality increased about ten times between 1930 and 1960, leveled off during the sixties and has since decreased. During the decline of mortality from coronary heart disease the consumption of animal fat declined, but so it did during the previous thirty years of sharply rising mortality (57). In Framingham the decline of coronary mortality was balanced by an increased number of non-fatal heart attacks (58) suggesting an effect of better treatment rather than an effect of dietary changes.
In Japan coronary heart disease is uncommon, allegedly due to the lean Japanese diet. A large study of Japanese emigrants (59) is often used as evidence because after migration to the United States these emigrants died from heart attacks almost as often as did Americans. The increased heart mortality after migration was not associated with the diet or the serum cholesterol, however, but with the cultural upbringing; those who lived according to Japanese traditions were protected. Most surprising, emigrants who stuck to the Japanese tradition, but ate the fat American food ran a smaller risk than those who were accustomed to the American way of life but ate the lean Japanese food (60).
If dietary fats were important this should obviously be reflected in the diet of patients who have had a heart attack. The following table gives the results from 13 studies, where the diet of patients with coronary heart disease was compared with the diet of healthy control individuals of the same age and sex. The amounts of dietary fats are given in percent of total calories. Asterisks means that the difference found was statistically significant. NS means that no absolute figures were given in the report, but that the difference was not statistically significant.
As you see, the differences were very small and in most cases due to chance. In only one study patients ate more saturated fatty acids than did healthy controls, but in the same study and in a further three, patients ate more polyunsaturated fatty acids, contrary to what was predicted from the message we have heard for so many years.
The crucial test is the controlled, randomised trial. Eight such trials using diet as the only treatment has been performed (76), but neither the number of fatal or non-fatal heart attacks were reduced significantly in any of these trials, not even if the results were added in a meta-analysis. A recent, small trial, which included the addition of alfa-linolenic acid to the diet, was succesful (77), but in that trial the serum cholesterol concentration was unaltered by the diet leaving us with more questions than answers.
Links to more critical comments about the diet
Some of my scientific papers about this and similar issues:
questionable role of saturated and polyunsaturated fatty acids in
cardiovascular disease. J Clin Epidemiol 1998;51:443-460.
Read also a dissent to the paper: Golomb BA. Dietary fats and heart
disease-dogma challenged? and my answer; same journal and same issue.
This paper won the Skrabanek award 1999
popular-scientific books, where you can read much more:
© Uffe Ravnskov