[The U.S. Department of Health, Education, and Welfare, Public Health Service,
presents T-1880 MCMLXIX]
[A National Medical Audiovisual Center Production]
[Dr. Jeremiah Stamler:] I'm Jeremiah Stamler, the Executive Director
of the Chicago Health Research Foundation
and Associate Professor, Department of Medicine, Northwestern University Medical School.
I'm convinced that a mass of evidence exists demonstrating that diet is one of the key factors
responsible for the current epidemic of severe mature, severe premature atherosclerotic disease
in the United States population, and accounting for the very high incidents of heart attacks
in the middle decades of life in our country.
[Dr. Mark Altschule:] I'm Dr. Mark D. Altschule, Boston, Massachusetts.
I'm Assistant Professor of Medicine at Harvard Medical School, Lecturer in Medicine at Yale University.
I'm convinced that diet cannot have a primary role in the pathogenesis of atherosclerosis.
[Dr. David Miller:] I'm Dr. David Miller from the Atlanta Field Investigations Unit of the Heart Disease and Stroke Control Program,
U.S. Public Health Service at the National Communicable Disease Center in Atlanta.
I'm here to serve as moderator for these discussions of this critically important subject.
[concepts and controversies in modern medicine]
[Diet and Atherosclerotic Disease: Are They Related?]
[Narrator:] Welcome to the exploration of Concepts and Controversies in Modern Medicine,
one of a series of programs dedicated to examining the uncertain, candidly recognizing
that much of today's teaching is necessarily based upon opinion and that the opinions
of eminent physicians in a given field vary widely.
The National Medical Audio-Visual Center believes that openly airing such opposing views
is a basic responsibility of medical communications.
Dr. David Miller, National Communicable Disease Center, will act as moderator of this presentation.
[Dr. David Miller:] First, Dr. Jeremiah Stamler will present some of the evidence from his work,
and from that of others, suggesting strongly that diet and atherosclerotic diseases are causally related,
and that appropriate changes in diet offer the possibility of prevention of these diseases.
Dr. Stamler.
[Dr. Jeremiah Stamler:] Let me, Dr. Miller, first point out the scope of the problem confronting us:
600,000 coronary deaths a year in the United States.
165,000 of them in persons under 65 years of age, three men for every woman,
with about twice as many non-fatal, crippling events.
This is one of the highest rates in the world, not only compared to the less developed countries
of the world, but also compared to the developed countries of Europe, the British Commonwealth,
and so forth, we rank very near the top.
I think only the Finns beat us out in this regard and I guess we can say we're like Avis,
we're second but we're trying harder.
Now, study after study of these international differences, both analyses of United Nations data,
World Health Organization data, the very important International Autopsy Study done by our colleagues
at the Louisiana State University Medical School, together with the very important field studies
of Dr. Keyes and his group in 12 different populations of Europe plus the Japanese,
all show a very close correlation among the pattern of a diet, particularly the saturated fat intake,
the dietary cholesterol intake, the serum lipids of the populations under study,
and the incidence and mortality rates as well as the prevalence rates of coronary heart disease.
That is of fatal and non-fatal atherosclerosis of the major coronary vessels.
Now this is one of the most important pieces of evidence.
I would underline the fact that the distribution of serum cholesterol in the population,
the mean levels, the rates of hypercholesterolemia differ markedly among these countries,
and along with that, the coronary rates differ and only diet can explain these differences.
For example, countries like the Greek islands of Corfu, of Crete, which have been studied intensely,
Southern Italy, Dalmatia, Japan, have much lower mean serum cholesterol and triglyceride
and beta life of protein levels than we do.
Their coronary rates are much lower in middle age.
This correlates directly with their saturated fat and cholesterol intakes.
It's importantly related also to their intake of total calories, but there's a great deal of evidence
to implicate the fat component of the diet as the most important one.
This is not to say that this is the only dietary factor, nor is it to say that diet is the only factor.
But it certainly is a key, and an important one.
Cigarette smoking also enters into the picture, and physical inactivity as well as high blood pressure
and other risk factors may as well.
I want to stress that diet is involved in at least four ways, not just in relation to serum lipids.
Diet is of course related to overweight.
Diet and exercise, or lack of exercise, are both related to overweight.
Overweight, as well as high serum lipids, are coronary risk factors.
In addition, diet is related through overweight to risk of high blood pressure.
Fat people get more high blood pressure than people who are lean throughout life.
And diet is related to the risk of diabetes, again, at least because fat people
get more diabetes than lean people.
And hyperlipidemia, overweight, hypertension, and hyperglycemia or diabetes,
are all major coronary risk factors.
I, therefore, find it very difficult to believe that diet is unrelated.
We also have the massive evidence of experimental atherosclerosis,
namely that the sine qua non for the reproduction of severe atherosclerosis
and our problem is not just atherosclerosis in general.
All of us have some of that. None of us at this moment are sick from it.
The thing that makes for morbidity and mortality is severe atherosclerosis, and the only way one
can produce severe atherosclerosis in species in the laboratory, avian, mammalian, herbivorous,
omnivorous, carnivorous, including monkeys, primates, is by feeding diets high in saturated fats,
high in total fats, high in cholesterol, and raising the blood lipid levels,
particularly the blood cholesterol levels.
And unless man is radically different from all other species and is a departure
from the basic laws of evolution, I cannot but regard this evidence as extremely important.
Finally, we know from a vast series of epidemiologic and clinical investigative studies
that by altering the kinds of diets we're eating at the present time,
we can shift the distribution of serum lipids to the left.
Or in plain, simpler English, we can lower the lipids.
Not just of hyperlipidemic people, but of whole populations.
This was shown in the National Diet Heart Study in a double-blind, controlled, open population study
in an institution and, at the present time, at least four major long-term field trials in New York,
in Chicago, in Los Angeles, and Helsinki indicate that such nutritional changes are associated
with decreased rate of coronary disease in middle-aged individuals.
In brief summary then, all of these pieces of information strongly indicate to me that diet
is one of the key factors in the contemporary epidemic and that to end this epidemic,
we must do something about the diet as well as other major risk factors,
such as cigarette smoking and hypertension.
[Pause]
[Dr. David Miller:] Thank you, Dr. Stamler.
Next, Dr. Mark Altschule will present some of his views, emphasizing the inadequacies of the dietary data
and casting doubt upon the hypothesis that diet is causally related to atherosclerotic disease.
Dr. Altschule?
[Dr. Mark Altschule:] Well, first let me say that there are certain aspects of the dietary problem
that all agree upon.
If a man's too fat, we try to reduce him. If he has diabetes, we treat it with diet.
If he has hypertension, we may treat it with diet.
But what I'd like to emphasize is that these are only peripheral and, in many cases,
very subsidiary factors.
The main etiologic factor cannot be diet and there's very little evidence to support the view
that there is any connection.
For example, the vast amount of animal work that has been done with the feeding of cholesterol
is imperfect in several respects.
In order to get any significant amount of atherosclerosis in some of these animal species,
one must elevate the blood cholesterol level to about between 2,000 and 4,000 milligrams.
Now, it's extremely rare for any man, sick or well, to have a cholesterol level over 1,000 and most of us
run around 250 or 300.
So that applying the data obtained at levels of two, three, or four thousand to what obtains in man,
where 300 is considered high, I don't think is very good logic.
Moreover, there's some question about whether the cholesterol that's introduced into the diet
of these animals in order to produce atherosclerosis is quite the same as the cholesterol we eat,
because it has to be separated, and processed, and purified, and there's a lot of school of thought
that believes that certain impurities develop in the handling of that cholesterol; there is in fact a toxic
or toxic product, damaged blood vessels, and induced disease
and it's not the cholesterol per se.
But, at any rate, I would say that the animal experiments are most unconventional,
and particularly with respect to the fact that certain species will have one strain in which
a very high cholesterol diet does produce atherosclerosis, but the same species with another strain
can take exactly the same diet and have no atherosclerosis at all.
With respect to the statistical studies, the epidemiologic studies in man,
they, too, of course, are imperfect.
The authors of the report of the Helsinki study and the Los Angeles study, in their own reports,
said that their groups were not matched.
The one from Los Angeles, in particular, referred to the fact that one of their groups
smoked more than the other and they hadn't taken that into account before starting it.
They said they didn't think it made any difference, I don't know on what basis.
The statistical studies that would be required to establish in man that the kind of diet
that Dr. Stamler refers to is the cause, the primary cause, of atherosclerosisis, is so huge
that the likelihood of their ever being carried out is remote.
There have been papers on that subject by epidemiologists and I won't go into them
in any detail because they've already been covered.
The matter of getting the blood cholesterol down is interesting, but I'm not sure how important it is,
because we don't really know where the cholesterol goes after it leaves the body.
For example, if you just add a few parts per million of cadmium to the drinking water of a rat,
you lower the blood cholesterol by about 40 or 50 percent, but the cholesterol all goes
into the blood vessels, and these rats have much more atherosclerosis than the rats
without that cadmium in the drinking water.
So that, when reports say that the cholesterol is diminished by 10, 15, or 20 percent,
unless such a report can say where the cholesterol goes, I don't think I'd be willing to accept it
as evidence of anything except an interesting laboratory finding, and laboratory findings,
unless they are directly related to clinical phenomena, can be extremely misleading.
As far as I know, there's only one phenomenon in which a lowering of a blood cholesterol
has been related to an increase in cholesterol breakdown, and that is with exercise.
If you give radioactively-tagged cholesterol to an exercising animal or man, you can show
that carbon dioxide is coming off and therefore, the cholesterol is being broken down.
Other ways of lowering cholesterol levels in the blood don't carry that implication really at all.
I think the emphasis should not necessarily be, in fact shouldn't at all be, on lowering a cholesterol.
The emphasis should be on influencing the cardiac status, and when a patient is put on a diet
with a change in fat count and so on, if blood cholesterol goes down 15 percent but he loses 40 pounds
in weight, if his cardiac status improves, I'd be more likely to relate to, more inclined to relate it
to the fact that he's lost some weight rather than to the fact that his blood level has changed.
I don't believe that levels, changes of 15 or 20 percent in levels really amount to very much.
First place, the blood cholesterol level's a very variable thing and if you were to take your blood
from a patient who's not likely to enjoy being stuck, blood cholesterol will go up.
You get a false high reading.
The current practice of advancing upon a patient with a large syringe, a large needle,
a toothy smile and a comment, "This won't hurt a bit," is not likely to change the blood cholesterol very much,
except in the upward direction.
Let us say that whatever evidence I've read does not support the view that diet is of
primary importance in the genesis of atherosclerosis.
[Dr. David Miller:] Thank you, Dr. Altschule.
You've raised some important questions from animal experimental work,
implying that much of this work is with overloaded animals, from epidemiologic work,
just pointing out the imperfections of some it.
I shan't summarize anymore, but I'll turn now to Dr. Stamler
for his comment on some of Dr. Altschule's points.
[Dr. Jeremiah Stamler:] Well, each of them could be dealt with at some length.
I'd like, for a moment, for us to discuss the experimental aspects.
I think that it is true that many of the studies have dealt with massive additions of cholesterol and fat
to the diet, but this is not the exclusive situation.
As a matter of fact, old Anichkov himself, the man who began the experimental work,
was one of the first to feed relatively small quantities of cholesterol--in natural form, let me add,
not as crystalline chemical material, but as eggs or milk or animal tissue.
And he did so over long periods of time with only modest elevations, Dr. Altschule,
of blood cholesterol into the two, three, and four hundred range, not into the couple thousand range.
Without the production of massive organ cholesterosis,
but with the production of significant atherosclerosis.
Dr. Katz, Dr. Pick, and I did the same thing with chickens in the laboratory at Michael Reese.
What we did was to give less of the cholesterol and fat in the diet over a longer period of time,
thereby mimicking in the composition of the diet very closely the kind of diet that's being eaten
at the present time in the United States.
Dr. Bruce Taylor has done the same thing with monkeys, and I guess one of the neatest
experiments in this regard is by Dr. Robert Wissler at the University of Chicago, who's fed two different
diets to primates, first the usual American diet,
no synthetic cholesterol added.
The chow, the grub such as we eat it everyday, day in and day out, most of us in the United States today,
versus the so-called prudent diet of the New York Anti-Coronary Club, the kind of diet that we've
been recommending through the Heart Association through our Coronary Prevention Program.
In those primates, only a moderate elevation of cholesterol ensued in the first group,
eating the usual American diet.
Again, no organ cholesterosis, but, again, definite atherosclerosis developed.
So I think there's no doubt from primate studies, from rabbit studies, from chicken studies, that you
can reproduce the lesions of human atherosclerosis in a variety of species by dietary
means of a longer duration, but sort of corresponding to what goes on over our lifespan
because it does take three, four, five decades from the time we begin eating the American diet,
so that our blood cholesterol rises from the chord blood level of 80 up to about 180 or 190 by age 20,
and up to about 230 by age 30.
This is the American pattern.
We used to say that this was physiological, this was the normal human response,
until we began to go out and look at other at other populations.
In the Japanese, it goes from cord blood 80 to 140 or 150 by middle age.
In Southern Italy, to 160 or 170, where the total life expectancy is better than ours.
The total death rate in middle age is much less than ours.
The coronary death rate, much, much less than ours.
So I think this argument about cholesterosis and massive feeding
is only part of the truth on this question.
[Dr. Mark Altschule:] I agree also with Dr. Stamler about the invalidity of the studies in which
such high amounts of cholesterol were given, and cholesterol was deposited in the blood vessels
and [?] cholesterosis developed.
What I was referring to, rather, was the fact that cholesterol, even in smaller amounts,
produces anemia in experimental animals, and in view of the fact that a considerable amount of evidence
has been accumulated, it shows that the endothelium of the arteries is a barrier to cholesterol infiltration
until damaged and the fact that hypoxia is one factor that will damage the lining.
I think cholesterol may be toxic without producing this massive deposit that we both deplore
in these experiments, and I might add that the hypoxic factor of damage to the endothelium probably explains
the propensity of heavy smokers to develop atherosclerosis because they carry about
a 10 percent saturation of carbon monoxide in their blood all the time,
which is enough to damage the vessels.
I'd like to pick up one other statement that Dr. Stamler made.
He said that the experimental animal developed a kind of atherosclerosis
that was much like the human.
However, in the conference held in New York about a year ago at the New York Academy of Sciences,
one of the experimenters with primates made a specific statement that animal atherosclerosis
has never reproduced what we saw in man.
That can be taken both ways.
It might be species different or differences or it might be a truly significant difference
between the experimental and the actual.
I'd like to enlarge a bit on the matter of the endothelium as a barrier.
The hypothesis was developed chiefly in Scandinavia, but has appeared in American
literature and paper by Liversmith & Newman, and I think the point is well made that a host of factors
may damage endothelium and a lot of cholesterol, which would come through in very small amounts,
to accumulate and that the cholesterol is not the cause of the atherosclerosis.
It's the lesion of the endothelium itself.
And along those lines, we have the experiments by Pavic on the production of arterial lesions
by humeral factor, not involving cholesterol at all.
And, of course, when it comes to a consideration of what the possible primary factor
in atherogenesis is, I think we can have recourse only to the work of [?], which studies the
hemodynamic factors and draws out the importance of them.
I should like also to point out that although cholesterol levels may be predictive
in Framingham, Mass., they're not at all predictive in Hawaii, and so, when we come to factors
that are predictive in one place and not in another,
I think we have to look behind them for their true significance.
[Dr. David Miller:] I wonder if we might, at this point, agree perhaps to disagree on some of the animal
experimental and pathologic work and get into the epidemiologic work and especially the subject of
what might be or what might not be recommended to patients or to the American public in the way of
dietary change in the hope that it might forestall some of this morbidity and mortality.
[Dr. Mark Altschule:] Well, I agree.
[Dr. David Miller:] Dr. Stamler, would you?
[Dr. Jeremiah Stamler:] I'd like to start in that discussion with a set of facts.
We have just looked at both our own study and the People's Gas Company, and in the pool of data of six studies,
Framingham, Albany, Los Angeles Civil Servants, the Minneapolis Businessmen,
the Western Electric Study in Chicago, and our gas company study, and we're impressed by the fact
that there are, unfortunately, a minority of Americans with serum cholesterol levels under 250.
The average for this cut being about 210, with normal blood pressure and without cigarette smoking.
And these people fare remarkably differently from the great mass of Americans with high lipid levels
with or without one another of these factors.
In fact, if one does five-, seven-, and ten-year life tables on these people, the coronary death rate,
the sudden death rate, is about one-sixth that of the population that has the traits I've mentioned
and total mortality is about a half.
Now I think we can make one simple recommendation.
As a matter of fact, I find this to be universally supported, and I hope Dr. Altschule supports it,
and that is certainly we should go after those high- risk people who are extremely coronary-prone
in our population and, aside from controlling their hypertension and their overweight and their
hyperglycemia and their cigarette smoking, do something also to shift their lipids over to the
low-risk group so that we can buy them some protection against the very high risk they have
at the present time.
From all our experience, a reduction in saturated fat intake and cholesterol intake,
if you will the Mediterraneanization--I'm stumbling over that hard word--of the American diet,
which incidentally permits very pleasant eating and which can lower their cholesterol
15 or 20 percent very easily for high-lipid people.
This kind of approach should certainly be fruitful for high-risk people.
[Dr. David Miller:] May I stop now and ask Dr. Altschule his reaction to that?
What, in view of the imperfections of the data thus far, what do you think should be recommended
for now and the future?
[Dr. Mark Altschule:] Well, I think that people should not be pigs and they should eat reasonably
and if a man's fat, he should be thinned down.
But beyond that, I don't see that there's any need to limit the kind or specific kind of food.
We all know that there are large variations, even within families, of the selection of food, and yet
a whole family may have coronary disease. In fact, they may all die of it before they're very old.
The matter of an excess of unsaturated fat bothers me a little because it can become toxic,
and whether or not unsaturated fat can give rise to intimal lesions is still a matter of discussion,
but in large amounts it quite definitely is toxic in animals.
Another thing that I think is somewhat important is the fact that eating has more than caloric significance,
it has a social and emotional significance, and I have dined with some friends
of mine who are on this low-cholesterol, high- unsaturated fatty acids kick
and I'm quite sure I'll never go back to dinner at those homes again.
[Dr. David Miller:] Let me attempt to summarize, first by emphasizing some of the points
on which we agree:
that morbidity and mortality from atherosclerotic disease presents a tremendous health problem in
this country and in other industrialized countries;
that central to the core of understanding these diseases...of their cause lies disturbed lipid
metabolism and that there are multiple other causes which are complexly interrelated
and very difficult to separate and enumerate any one.
We have not agreed as to the role played by diet.
I think Dr. Stamler believes that it is of primary importance and I believe you believe
that it is of secondary importance.
We have agreed, I believe, that something needs to be done more effectively than we now are doing
about these disease and I hope that at future time, at some future time, we may be able
to come to this sort of agreement.
[Narrator:] We thank Dr. Jeremiah Stamler, Dr. Mark Altschule, and Dr. David Miller
for their interesting analysis of a critical problem in patient care.
In subsequent programs we shall continue to record equally significant
concepts and controversies in modern medicine.
The opinions expressed on this program do not necessarily constitute endorsement by the
Department of Health, Education, and Welfare, the Public Health Service, or its constituents.
[Jeremiah Stamler, M.D., Executive Director Chicago Health Research Foundation;
Assoc. Professor, Department of Medicine, Northwestern University Medical School]
[Mark Altshule, M.D., Assistant Clinical Professor of Medicine Harvard University;
Lecturer in Medicine, Yale University School of Medicine]
[David Miller, M.D., Chief, Atlanta Field Investigations Unit, Heart Disease
and Stroke Control Program, National Communicable Disease Center, Atlanta, Georgia]
[concepts and controversies in modern medicine]
[THE END T-1880]