If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
In mid-20th century, several streams of knowledge converged to create the new academic discipline of cardiovascular disease epidemiology and the new practice of preventive cardiology. One stream was modern cardiology, with the ability to diagnose myocardial infarction, to characterize and count its victims, and to report vital statistics on cardiovascular causes of death. Another stream came from burgeoning clinical and laboratory research and greater understanding of the underlying processes of atherosclerosis and hypertension. A third stream came from the observations of intellectually curious “medical Marco Polos,” who brought back from travels their tales of unusual population frequencies of heart attacks, along with ideas about sociocultural causes. This led to more formal research about cardiovascular disease risk and causes among populations and about mechanisms in the clinic and laboratory. The broad river of investigation thus formed produced a risk paradigm of the multiple biologic, behavioral, and societal factors in causal pathways to the common cardiovascular diseases. An evidence base was built for sound clinical and public health approaches to prevention. Here, the author tells brief stories about 5 early and particularly observant world travelers and their influence on knowledge and thinking about prevention.
Modern epidemiologic observations and trials of cardiovascular disease (CVD) prevention have some of their roots in the voyages of medical scientists who, like Venetian explorers of yore, traveled to “exotic” lands to marvel over and report on “foreign” wonders. In the last century, a few found themselves, for one reason or another, in regions with remarkably low or high rates of CVD and with their curiosity aroused about possible causes. These “medical Marco Polos” became central actors in the early dramas of preventive cardiology and were originators of today's views on the environmental determinants of epidemic heart attack and thus the potential for prevention. Stories about their travels among contrasting cultures during the early and mid-20th century begin with that of Cornelis de Langen, of Groningen, a young internist assigned by the government of The Netherlands to practice and teach in the Dutch East Indies.
Cornelis D. de Langen
De Langen's adventure began in May 1914, when his ship left Rotterdam and sailed by a route we cannot document but that likely took him through the Suez Canal, then over the vast Indian Ocean into the Strait of Sudah, past Rakita, the islet remnant of Krakatua's explosion, to dock finally, after weeks at sea, in the bustling tropical port of Batavia, the Dutch colonial name for present-day Jakarta.
As chef de clinique in internal medicine at Groningen University, de Langen's promising academic career was interrupted by his assignment to the colonies. The young physician, guided by his Groningen chief and mentor, Albert Abraham Hijmans van den Bergh, was nevertheless well prepared to help the Javanese combat epidemic plague and to teach medicine to the local trainees.
Van den Bergh, at the beginnings of modern medicine, was the complete internist: intellectually curious, deeply informed in the biochemistry of disease, devoted to his patients, and honoring a tradition of professional duty. House officer de Langen practiced and taught within this solid tradition. Thus, the request of his government to serve in the colonies was met with dutiful acceptance. He departed his homeland for an indefinite stay in a world that would prove to be foreign not only in climate, culture, and language but also in the medical profile of its people.
Supported in Batavia by the Foundation for the Education of Indonesian Doctors (Figure 1) , de Langen soon established order in the clinic, with its clean-swept yards and pleasant semitropical surroundings. His family biographers recount how, with a color-coded chart of the ward beds, he could identify at a glance not only the patient and the diagnosis but the stage of hospitalization and therapy, allowing him to make rounds efficiently and effectively.
International medical academician Isidore Snapper, one of de Langen's closest lifetime colleagues, later corroborated others' impressions of the young doctor's suitability for the job, calling de Langen “a young man with a very original mind” and writing that he “always considered de Langen as an excellent example of the inspiration which Hijmans van den Bergh instilled into his pupils.”
It was not long before de Langen noted that his native Javanese patients rarely exhibited the common diseases with which he had been accustomed to dealing in Europe. Pursuing this clinical impression, he reviewed 10 years of admissions charts and found only 5 cases of acute gallbladder disease among many thousands of patients passing through the medical wards and only 1 case on the surgery service among 70,000 admissions surveyed.
After documenting the rarity of gallstones, de Langen wrote about vascular phenomena: “thrombosis and emboli, so serious in Europe, are most exceptional here. This is not only true of internal medicine, but also on surgery, where the surgeon needs take no thought of these dreaded possibilities among his native patients. Out of 160 major laparotomies and 5,578 deliveries in the wards, not a single case of thrombosis or embolism was seen.”
The following confirmatory account is taken from the memoir of another traveling Dutch internist, Isidore Snapper: “My friend and former co-resident in Groningen, C. D. de Langen, had discovered in 1916 in Indonesia, that the obligatory vegetarian Oriental, whose intake of cholesterol was practically zero, does not develop gallstones. Soon it appeared that not only cholesterol stones of the gall bladder but also arteriosclerosis and phlebitis (also lung emboli) hardly ever occur among the Orientals.”
De Langen documented the rarity of coronary syndromes in Java in a 1935 report to the second conference of the International Society of Geographic Pathology. There he described his survey finding of only 1 case of angina pectoris among Javanese and 6 among Chinese admitted during a 5-year period to the 500-bed Batavia Municipal Hospital.
Having thus established impressions of comparative CVD prevalence and incidence in the patient population, de Langen proceeded to observe characteristics among defined populations outside hospital, first comparing the average blood cholesterol level of Javanese, Dutch, French, and Germans, demonstrating similarities among the Western groups but much lower levels in the Javanese. He also found that Javanese stewards on the Dutch transport liners were closer “metabolically” to the Europeans than to other native Javanese.
In his seminal early report, published in Dutch in 1916, de Langen summed up clinical relations among diet, serum cholesterol, and atherosclerosis: “a cholesterol-rich diet and severe metabolic diseases, such as diabetes, obesity, nephritis, and arteriosclerosis, are associated with hypercholesterolemia.”
He further connected blood lipid levels with diet patterns in these early observations: “The question has frequently been asked whether cholesterol concentration of blood and bile is controlled by diet—the diet of the population in the Dutch East Indies differs very much from the Western European diet. The local diet is mainly vegetarian with rice as the staple, that is very poor in cholesterol and other lipids.”
Elsewhere, de Langen made perhaps the first comment on exceedingly low blood cholesterol values seen in wasting diseases and the effect of adding fat calories, as quoted from a letter to Snapper: “The food of the masses which visit our hospitals and polyclinics in Java contains but little cholesterin; the result is a smaller content of cholesterin in the blood. Especially in the very numerous patients with malaria and ancylostomiasis, the cholesterin content is quite often on the very low side. Whenever, in hospitals or elsewhere, we give people a diet rich in lipoids, the quantity of cholesterin in their blood rises at once.”
In any case, the rice-based, virtually meatless “food of the masses” to which de Langen referred, combined with a pattern of daily activity, added up to a picture of habits and health very different from the pattern of Europeans.
From observation to experiment
From these observations, de Langen moved in 1922, logically and apparently comfortably, to experimentation. In what was probably the first systematic human trial of the effects of diet change on serum cholesterol levels, he found an average 40 mg/dl increase in cholesterol in 5 Javanese natives who were shifted from a rice-based vegetarian cuisine to a 6-week regimen high in meat, butter, and egg fats.
Thus was a logical cycle completed, from observations on disease frequency, to cross-sectional relations of diet, blood-lipid levels, and disease rates, and finally to modification of blood lipid levels by dietary experiment. It might seem that already in the 1920s, de Langen had, with simple tools and broad curiosity, left to future study only the direct experiments in humans: the influence on disease risk of diet change and lipid lowering. That future, as we know, would be long in coming.
Perspective on de Langen's influence
Did de Langen's findings, published in Dutch, in fact influence the subsequent pursuit of causal and prevention investigation? His early reports were indeed referenced in other languages, including English and French, but most of what we know about the wake left by his pursuits comes from correspondence with his colleague and fellow Dutchman Isidore Snapper. Snapper had for a short period taken de Langen's place as chef de clinique at Groningen University Hospital before going on to a notable career at Amsterdam, then in China and in the United States. Snapper wrote about how de Langen's work and correspondence influenced his own observations in China: “In 1940, I confirmed De Langen.s results … by the observation that in North China, coronary disease, cholesterol [gall]stones and thrombosis were practically nonexistent among the poorer classes. They lived on a cereal-vegetable diet consisting of bread baked from yellow corn, millet, soybean flour and vegetables sauteed in peanut and sesame oil. Since cholesterol is present only in animal food, their serum cholesterol content was often in the range of 100 mg. per cent. These findings paralleled the observation of De Langen that coronary artery disease was frequent among Chinese who had emigrated to the Dutch East Indies and followed the high fat diet of the European colonists.”
It cannot be claimed that de Langen's remarkable multidisciplinary findings, or their powerful implications, inundated the medical and nutritional plain of the period leading up to midcentury. His early works on diet and disease were not published in English, which was particularly unfortunate because, according to Gerson (who made this comment in French), de Langen wrote and spoke “an elegant English, clear and simple, which made him a classic teacher of first rank.”
The translation lag time also may have contributed to what Snapper called “historical inaccuracies that necessarily result when a new generation tries to review a concept born half a century previously.”
However, de Langen provided, at the very least, insightful observations, testable hypotheses, background evidence, and, as I will document, intellectual impetus to other investigators who effectively pursued the same questions. De Langen's ideas about cultural differences in lifestyle and cardiovascular risk were unique among the early formulations of the diet–lipid–heart disease hypothesis. His studies in Java were among its first clinical-epidemiologic tests.
In any case, de Langen did not receive the credit by contemporaries that he deserved for his hypotheses or his rigorous testing of them in this early period of modern medicine. However, because of the influence of his findings on the thinking of his colleagues, we can trace a direct line—a Dutch dynasty, if you will—in the drama of diet and chronic diseases, from de Langen in 1916, to Snapper in 1941, to Groen in the early 1950s, and to other Dutch workers up until today. We can also trace the line from them to the American pioneers Ancel Keys and Paul White.
De Langen's innovative career led to his eventual appointment as rector of the School of Medicine in Batavia. Later, he succeeded van den Bergh as chief of internal medicine at Utrecht and cowrote with him and Snapper the first Dutch textbook of internal medicine (in 1946). One can only speculate why he abandoned further development of his seminal findings about the diet-lipid-atherosclerosis hypothesis. Perhaps he considered the matter largely settled.
(A symposium to honor the career of Cornelis de Langen was held in Leuven, Belgium, in October 1999, at which his contributions were gathered by his descendants and successors.
Early in the evolution of preventive cardiology came a scholarly admonition from the renowned dean of North American internists, George Minot of Boston City Hospital. He introduced the 1941 edition of Isidore Snapper's little tome, Chinese Lessons to Western Medicine (Figure 2), with this rare and lucid grasp of the sociocultural determinants of common diseases: “It became recognized in the final quarter of the 19th Century that there is a geography of disease. It has been proved that certain diseases reflect the character of the social and economic, as well as the geographic environment.”
Yet it would be several decades before the medical community as a whole, aware from clinical findings of causal influences on arterial diseases in individuals, began to take a serious interest in the causal-preventive implications of geographic and cultural differences in population rates of heart attack.
Isidore Snapper, Colleague of de Langen
Isidore Snapper, like de Langen, was educated at Groningen, where he too was profoundly influenced by Hijmans van den Bergh, who taught him clinical medicine and biochemistry, and also by the famous Groningen physiologist H. Jacob Hamburger. In these favored circumstances, Snapper became one of an early international cadre of internists and clinical investigators who applied biochemical knowledge and curiosity to clinical problems. Among his contributions while working with van den Bergh, for example, was the simple test he developed to discriminate direct from indirect bilirubin in serum, later used worldwide and called the van den Bergh reaction.
Snapper took the opportunity of de Langen's departure for the Dutch East Indies to replace him as chef de clinique at Groningen, but only for a short time. When it became clear in 1917 that van den Bergh would leave Groningen for a more prestigious appointment at the University of Utrecht, Snapper surmised that he would not be in line for the Groningen chair of medicine. This assessment he attributed in part to his youth but also to the enmity he was aware was engendered by his personality, most tactfully described as “self-assured.” Thus, to the disappointment of van den Bergh, who had hoped Snapper would succeed him, the ambitious and insightful Isidore sought to advance his academic career at Amsterdam. There, Snapper indeed became a leader in internal medicine and clinical investigation, propounding his philosophy of generalist as distinguished from the new trend toward specialization.
As war clouds moved in, and after Hitler's invasion of Austria in 1938, Snapper described his home situation, saying that “Mrs. Snapper did not rest until I made a trip to the U.S. to look for another position.”
In New York, making use of contacts from earlier Rockefeller explorations of medical education in The Netherlands, Snapper went directly to the director's office of the Rockefeller Foundation. There he met the visiting head of the Peking Union Medical College, an institution supported from its inception by the foundation in its mission to “combine Chinese wisdom with American know-how.” On the spot, Snapper was offered the post of chief of medicine at Peking and so requested a leave of absence from Amsterdam University to take the post.
At this time, before war broke out, according to Snapper, Jewish academicians who “fled” their positions in Holland were considered (by those perhaps less threatened) to be “unpatriotic, unprincipled, opportunistic deserters.” Despite opprobrium, Snapper left home with his wife and the 4 youngest of their 5 children and in January 1939 took up his new position in Peking in what he described as a “beautifully built, well-organized medical school with excellent research facilities.” There, he promptly “fell in love with the Chinese way of life.”
As had de Langen in Java, Snapper soon observed the rarity of Western diseases among the Chinese, with the exception of a few successful merchants affected by diabetes and atherosclerosis. And he was equally systematic if more modern in his approach to documenting the cultural contrasts, describing, for example, what he called an “epidemiologic study” of electrocardiograms in the Peking Hospital: “I studied all these electrocardiograms without looking at the names and I found only a small number of electrocardiograms typical for myocardial infarction. When I had finished I … checked the names: all the coronary disease electrocardiograms belonged to Occidentals who had been admitted to the private pavilion! And among the poor Chinese, I did not find any electrocardiographic curves indicating coronary sclerosis, just as hardly any of our Chinese patients had anginal complaints. The pathologist [moreover] did not find arteriosclerosis at the autopsies of our patients. Although the absence of arteriosclerosis in Orientals has been found over and over again, there are still doubting Thomases.”
Snapper documented the vegetable nature of the northern Chinese diet and its association with the rarity of CVD cases, deriving details from Chinese studies that characterized the average daily diet of the poor in China as consisting of 2,100 cal, 88 g of protein, 47 g of fat, and 331 g of carbohydrate.
Snapper described the caloric sources: “As far as the poor classes are concerned, the caloric energy of their diet is almost entirely derived from cereals, that is, yellow corn and millet. Before grinding, these cereals are mixed with soybean flour which contains valuable amino acids. The fat of this diet is derived principally from the sesame and the peanut oil in which the vegetable is cooked. The vegetables vary with the season. Spinach is taken in early spring, small green cabbage in late spring, kohlrabi and string beans in summer, and Chinese cabbage in autumn and winter. Sometimes turnips and eggplant are eaten. Instead of the butter of the Western diet, the Chinese use considerable quantities of vegetable oil, in north China, especially peanut oil and sesame oil, sometimes even soybean oil, [which] is more popular in Manchuria.”
Snapper surmised, in the 1965 second edition of Chinese Lessons to Western Medicine, that the vegetable oil intake in China resulted in the higher unsaturated fatty acid levels found in plasma of the Chinese and concluded that the fatty acid content of diet may influence certain diseases. However, he cautioned, “Any suggestion about a relation between the presence of special unsaturated fatty acids in the diet and the remarkable scarcity in China of several diseases in the pathogenesis of which lipoides play an important role … would be completely speculative.”
Caution, however, did not stop him from informed speculation.
Snapper spoke early about cardiac disease trends, observing in this later edition that atherosclerosis, angina, and myocardial infarction seldom occurred in northern China, while those conditions were increasing in the United States and Europe, and he opined about possible causes: “It is difficult to give an explanation of this characteristic feature of geographic pathology. One can, of course, fall back on the equanimity of the Chinese, but the difference in nutrition in the Chinese and Westerners may be a better explanation. Atherosclerosis begins as a fatty infiltration in the intimae of the vessel walls. Quantitative and qualitative differences exist between the lipoide content of the Chinese and foreign diets…. The Chinese diet contains only small amounts of cholesterol but considerable quantities of unsaturated fatty acids, especially linoleic and linolenic acid. It is certain that the average cholesterol content of the blood of the Chinese is lower than that of the Westerners and this gives perhaps an indication of why the tendency of lipoide infiltration in the vessel wall is so much smaller among the Chinese.”
In this later edition of his book, Snapper was among the first to emphasize the possible importance of omega-3 as well as omega-6 fatty acids with respect to coronary heart disease and prevention of the atherosclerotic process. He also offered this vivid observation on the Chinese and dairy fat: “The likes and dislikes of different peoples are widely divergent and can never be predicted. The northern Chinese is unwilling to drink milk, designates butter as ‘cow oil,' and is unable to understand how educated persons can eat the ‘malodorous’ substance we designate as cheese.”
The 1941 edition of Snapper's classic Chinese Lessons to Western Medicine came to the attention of those in the late 1940s who sought confirmation or refutation of their own ideas about cultural influences on population rates of heart attack. Ancel Keys and Francisco Grande in the Laboratory of Physiological Hygiene at the University of Minnesota, for example, often mentioned Snapper's book in 1950s conversations and also referred me to the remarkable early works of de Langen. Citations of these Dutchmen appear early among reports of these and other pioneers in CVD epidemiology. Thus, de Langen's and Snapper's early “epidemiologic” and clinical observations in China and in Java bore fruit long after, in the initiation of formal cross-cultural comparisons by Ancel Keys and Paul White and international colleagues of the Seven Countries Study and the studies of others who followed this line of research among contrasting occupations and cultures.
(Snapper's expansive scholarship in China came to an abrupt halt in 1941, when he and his family were interned by the Japanese. After a dramatic repatriation in exchange for Japanese prisoners of war, the family arrived in England in 1942 and immediately sought passage back to the United States. There Snapper was drafted for 2 years of service in the United States Army surgeon general's office to prepare public health and medical recommendations for Asian countries in anticipation of their liberation from the Japanese. Subsequently, he spent 8 years at Mt. Sinai Hospital in New York, where he became an international authority on infectious and tropical diseases and on medical education.
One of his biographers wrote of Snapper's demeanor in his new homeland: “Snapper enjoyed a considerable reputation in America, but he was also known for his authoritative and arrogant way of doing things. The story made the rounds that when asked why he did not have a diploma from the American Board of Internal Medicine, he answered, ‘Who on the Board would dare to examine me?’”
The Dutch Dynasty Continues: Johannes Juda Groen of Holland and Israel
Johannes Groen, like his colleague, countryman, and contemporary Isidore Snapper, was the son of a Dutch diamond cutter, an elite profession of Dutch and Belgian artisans often passed down among families. As Snapper's trainee in Amsterdam, Groen was inculcated with an academic tradition of biochemical and metabolic approaches to internal medicine and came thereby to a lifelong interest in diet, metabolism, blood lipids, and atherosclerosis. Following in Snapper's footsteps, he studied abroad, at St. Bart's and the London Middlesex Hospital and then, in the late 1930s under a Rockefeller Fellowship, at Harvard's Thorndike Memorial Laboratory.
Although his personal survival during the Second World War was protected by having a non-Jewish wife, Groen was nevertheless dismissed from his position in Amsterdam, as were most of his Jewish colleagues under the Nazi occupation. During this enforced period of reading and reflection, he developed an interest in psychosomatic aspects of disease that continued when he returned to be chief of internal medicine at the Wilhelmina Hospital from 1945 to 1948. Thereafter, he became professor at the Hadassah Hebrew University Medical School, returning finally to The Netherlands in 1968 to become professor of psychobiology at Leiden.
Groen's early research demonstrated hypercholesterolemia in a patient with prominent arcus senilis, a deposition of cholesterol in the circumference of the cornea. He recounted, “Once we had established this hypercholesterolemia, we asked ourselves if the course of the disease could be positively influenced by the use of a cholesterol-free diet … only animal products supply cholesterol to the body … such a diet consists of bread (preferably whole meal bread), vegetables, fruits, potatoes, cereals, peas, beans, and pure vegetable fat … in some cases we saw indeed a remarkable improvement.”
This is one of the first cholesterol-lowering diets prescribed, and Groen's regimen remains appropriate to the purpose many years later.
During this period, Groen also analyzed the per-capita fat consumption in The Netherlands dating from the prewar period. Comparing samples from blood donors over the years, he also established that there likely were significant differences in average cholesterol levels before and after the war, at ages <30 and >30 years, and between Jews and non-Jews. Thus, he already had a clear idea of the dietary and genetic components of elevated serum cholesterol.
Despite his early and keen interest in the relation of diet, blood lipids, and CVDs, Groen was unable to pursue it because of the wartime suspension of his professional status. However, on returning after the war to the academic scene, and with support of The Netherlands Organization for Research, he published in the Dutch nutrition journal Voeding, already in 1952, the results of a controlled dietary experiment on cholesterol modification. In that study among 60 healthy volunteer students, he compared the pattern of a “well-to-do diet, mostly animal fat,” a “sober normal diet,” and a vegetable-fat diet, demonstrating clear dietary effects on serum cholesterol level along with different intrinsic levels of blood cholesterol among subjects.
Soon after, he became internationally recognized as an authority on diet–lipid–heart disease relations.
Groen went on to study other possible environmental influences on serum cholesterol levels, including gender differences, menstruation, infectious diseases, and physical and emotional stress, concluding that dietary effects were independent of all these factors and that serum cholesterol level decreased during a stressful period but overcompensated during recovery.
Cultural comparisons: Sephardim and Ashkenazim, Benedictines and Trappists
When Groen became active at the new Hadassah Medical School in Israel in the late 1940s, he made cross-cultural comparisons that had a substantial and wide impact on awareness of possible cultural effects on disease risk. He found that the migrant Sephardic Israelis had far less atherosclerosis and fewer diabetic complications than the Ashkenazim, which he attributed to the former group's only recent departure from a nomadic lifestyle.
Then, beginning in 1953, Groen and coworkers in The Netherlands systematically compared Benedictine and Trappist monks, hypothesizing there would be a difference between the monastic orders in cholesterol levels and coronary heart disease frequency on the basis of known differences in their dietary fat intake (Figure 3). If, in contrast, the primary difference in serum cholesterol was due to psychosocial factors, they reasoned, no differences should be found between the 2 meditative monastic groups. The higher cholesterol level they found among Benedictines supported the idea that diet had the more important environmental influence. However, they also found in these relatively small populations that coronary disease was apparently rare in the 2 groups of monks.
Groen's publications in the early 1950s, and also his meeting Ancel Keys and colleagues in England and in Amsterdam in 1952, probably had to do with his increasing recognition in the new field and with his appointment in 1958 as chairman of the World Health Organization Expert Committee on Cardiovascular Disease and Hypertension. Later in life, Groen was quoted about his “missed boat,” the important, unexplained findings on the effects on blood cholesterol levels of different dietary fats and fatty acids: “Gradually I got the idea that not only the amount of fat, but also the kind of fat influences the serum cholesterol level in the development of coronary disease. At the end of the war not only less fat was eaten but [it was] also primarily vegetable fat. The idea that the kind of fat could have an influence had not yet been described.”
He later lamented that it was “a great pity” that he had not been able to follow up on his animal-vegetable fat hypotheses, noting, “we were on a route to an important breakthrough. Looking back, it became clear how important the discovery had been. Nobody in the world had established that the type of fat influences serum cholesterol level. We talked continuously about vegetable and animal fats but we didn't know what the chemical differences were between those two types of fat.”
Nonetheless, Groen felt that he had laid the foundation for the definitive studies on effects of fatty acids and their saturation that took place from the late 1950s onward, primarily in the United States.
In 1981, Groen received the van den Bergh Award for Achievements in Internal Medicine, the wording of which would warm the heart of any “true internist” as well as the hearts of those who knew and admired him: “The committee dealt in the spirit of the originators to propose a true internist with a broad and special clinical interest who practices the profession in all its variety, additionally bestowed with a biologically oriented imagination culminating with the development of new concepts of internal medicine. The work of this internist should in addition contain differently oriented scientific contributions. According to the committee, colleague Groen completely fulfilled this requirement.”
The award was made at the 50th anniversary of the Dutch Society of Internists and was attended by the family of Professor Hijmans van den Bergh, which touched Groen deeply. In his biography, Groen recalled a late-career encounter with his kindly mentor, van den Bergh, who, at a public event remarked to him, “You were educated by Snapper. [Therefore,] I consider you my intellectual grandson.”
“When I blushed,” Groen wrote, “he also suddenly became shy and, saying not another word, moved on.”
At the end of his career, Groen was reported by friends to have been embarrassed but nonetheless grateful to receive a knighthood from Queen Wilhelmina. He died in 1990.
(As for his long relation with Isidore Snapper, Groen recounted in his biography that his apprenticeship with him was difficult because of the latter's abrupt, challenging manner, which Snapper, apparently uniquely, thought encouraged independence and originality among his trainees. However, although he may have been intimidated, Groen especially admired Snapper's performance on ward rounds and patient demonstrations and sought to emulate him in those clinical “sports.” He wrote in 1954, “As tutor in my own specialty I am very much indebted to [Snapper]; also I am grateful to him for my spiritual development because he was responsible for my journey to London and later to the U.S.” However, he added, “as human beings we never became close to each other.”
Snapper's assessment of his student included an ultimate compliment. At Groen's PhD defense, he recalled Groen's diminutive physical stature. The greatly admired yet feared mentor remarked, “Assistant Groen came to us a tiny man. He became the internist taller than any of us.”
More lessons from medical travels were brought to an even wider audience by the dashing duo of Ancel Keys, Minnesota physiologist-nutritionist, and Paul Dudley White, Boston cardiologist and medical diplomat of renown. For several years in the early 1950s, they together and informally compared coronary disease prevalence along with diet, lifestyles, and blood cholesterol levels among Mediterranean countries, South Africa, Finland, the United States, Hawaii, and Japan. In doing so, they were among the small group of founders of CVD epidemiology and preventive cardiology.
Ancel Keys, physiologist of Minnesota, grew up in the Bay Area of California and was schooled at the University of California, Berkeley. After graduate-study peregrinations, including periods at the Scripps Institute in La Jolla, California, August Krogh's laboratory in Copenhagen, Joseph Barcroft's laboratory at Cambridge, and the Harvard Fatigue Laboratory, Keys found his true métier in human biology. The new home base for his life's work would be an unusual institution he named the Laboratory of Physiological Hygiene and which he established at the University of Minnesota in 1939, housed under its memorial football stadium.
The noted British nutritionist Hugh Sinclair visited the Laboratory of Physiological Hygiene during a wartime tour of the United States, and he was impressed with the scope of study, the multidisciplinary organization, and the methodologic rigor of Keys's physiologic hygiene. After the war, and after the publication of Keys's monograph The Biology of Human Starvation,
Sinclair invited his American colleague to spend a sabbatical year at Oxford, which Keys accepted for the academic year 1951–1952.
By that time, Keys, the physiologist, already had for several years begun to look beyond the laboratory to the population and larger culture, to what he called the “mode of life,” for causes of the apparent epidemic of “executive disease.” And he was already suggesting the strategies of physiologic hygiene: diet and lifestyle change as possible preventive measures.
The priority of his thinking about epidemic heart attack in postwar America is documented in a 1947 photograph (Figure 4), which is among the first depictions of the epidemiologic transition, long before the phenomenon received that label.
Keys recounted in his memoir that he was alerted to the public threat of heart attack in the late 1940s by frequent Minneapolis newspaper reports of sudden deaths among prominent Minnesota executives.
At the occasion of the photograph, he argued early for sociocultural causes of the epidemic before an audience of local leaders he sought as subjects for the first prospective study of characteristics influencing CVD risk, the Minnesota Business and Professional Men Study, which opened in the fall of 1947. This graphic presentation was clearly meant for lay understanding and local recruitment, rather than for “science,” because his argument was based on such crude data; inadequate numbers from only 1 community, unadjusted for age and gender, with data trends speculatively extrapolated a decade ahead of the meager evidence. The graphic represents, however, the idea and the priority of its declamation in 1947.
Because of the international impact of his newly published tome on starvation, Keys was named chair of the 1951 Rome meeting of the United Nations Food and Agriculture Organization. There he propounded his newfound enthusiasm for diseases of overnutrition. And there his dietary theories were labeled an “American idea” and were either ignored or considered an annoyance by most of the serious-minded nutrition professionals present, still preoccupied with postwar deficiency states. One physician-physiologist from Naples, however, Gino Bergami, remarked privately to Keys that coronary events were rare among workers and peasants in his part of the world. When Ancel expressed skeptical interest, Bergami suggested he come to Naples to see for himself.
Soon after his return to wintry Oxford, Keys wrote to Bergami accepting the “invitation.” Not only intrigued by the idea of a country of “few coronaries,” he also was ready to escape the humid cold climate and academic ambiance that, he complained, featured too many routine lectures to medical students and practitioners, hardly adventuresome in a sabbatical year. Serendipitously, therefore, it may have been the combination of Bergami's casual comment and Keys's characteristic curiosity and restlessness that opened the next period of Keys's career and, with it, a new chapter in the epidemiology and prevention of CVDs.
Cultural comparisons worldwide
In the spring of 1952, Keys and his chemist wife Margaret left Oxford and drove their British Hillman Minx across France, the Alps, and Italy, to Naples. There, Bergami set them up with a laboratory and assigned them a young nutritionist assistant, Flaminio Fidanza, who was to become their longest term collaborator. While Margaret prepared a facility and a protocol for serum cholesterol determinations, Fidanza pursued the dietary survey. They proceeded to make a comparison of casual recruits among Italian laborers, university staff members, and Neapolitan banker friends of Bergami, in sufficient numbers to demonstrate statistical significance for serum cholesterol differences on the order of 10% between these groups that differed in habitual eating patterns.
Meanwhile, medical colleagues confirmed the rarity of coronary syndromes on the wards of public hospitals in Naples.
Keys's uncommon sense had led him early to a sine qua non of epidemiologic design: correlations among apparently contrasting rather than among homogenous samples. The Naples results were promising. Laborers and peasants who ate a typical workingman's Mediterranean diet of pasta, olive oil, legumes, fruits and vegetables, bread, and wine, had significantly lower values of serum cholesterol than did the bankers who ate “high on the hog.” A quick side trip to meet colleagues in Spain and make similar crude comparisons was confirmatory. Keys had become a modern-day “Marco Polo,” and with the same point of departure: Italy (Figure 5).
Paul Dudley White and Ancel Keys make a traveling team
Keys then invited Paul Dudley White of Boston, the leading international cardiologist of his day, to join these geographic comparisons, thereby opening medical portals worldwide. In the mid-1950s, the intrepid pair set off for adventures afar: to Sardinia in 1954, South Africa in 1955, Finland, Honolulu, Japan, Los Angeles, and back home to Minneapolis and Boston in 1956. They returned from each new junket with ever more engaging booty: congruent data and fresh ideas. And they got busy enlisting the leaders of world cardiology in the new activities of population studies and the new concepts of preventive cardiology.
Keys had met and hosted White in an early seminar on atherosclerosis held in Minneapolis in 1955, and among others, also the Cape Town physician and nutritionist John Brock, who intriguingly portrayed for those assembled the unique natural experiment of contrasting cultures in southern Africa: European-descended Caucasians, Bantu natives, and the so-called Cape Colored Indians and mixed-race groups. Brock and his biochemically sophisticated colleagues already had adopted multidisciplinary investigations among these populations and even had fed the varied local diets to Cape Town prisoners to study effects on serum cholesterol.
During his visit to Cape Town in 1955, Keys and White acquired another investigator colleague, Brian Bronte-Stewart, who joined them in world travels, in an ever widening foray into populations purportedly contrasting in traditional diet. On their way to Japan in 1956, the quartet of Ancel and Margaret, White, and Bronte stopped off in Honolulu to visit Nils Larsen, a leading internist, who quickly recruited a group of Hawaiian Japanese for diet-lipid-disease comparisons (Figure 6).
On reaching the Japanese mainland, with introductions arranged by Noboru Kimura of Kurume University, the team compared findings in miners, clerks, bankers, and professors. Kimura had approached Keys in 1952 with his own pioneering ideas about stroke and coronary disease in Japan and had visited the Minnesota laboratory just before the 1954 Washington World Congress of Cardiology, laden with a suitcase of Japanese electrocardiograms for classification. He was interested mainly in the relation of low dietary protein and high salt intake to stroke but went along with Keys's interest in coronary disease to adopt and join in Keys's systematic approach to field studies. (Local Japanese physicians never quite grasped why Keys, who was interested in heart attacks, came to study Japanese, who rarely had them.)
On their return to the United States from this extended Asian junket, Professor and Mrs. Keys, Bronte-Stewart, and White descended on the Japanese-American Hospital of Los Angeles, where they were able to survey Japanese immigrants, thus rounding out a picture of findings in Japanese under different conditions of cultural exposure, from the mainland, through Hawaii, and all the way to California.
The data from the informal surveys of the grand tour of the Marco Polos indicated to them important differences in diet, serum lipid levels, and coronary disease and stroke frequency, both within and among societies. They tended to confirm White's long-standing impressions of varied geographic pathology and Keys's new ideas about sociocultural determinants of the metabolic characteristics associated with vascular diseases.
Thus, after a quadrennium of such wanderings and taking of casual samples, more formal study was clearly called for. Keys set to work to plan the Seven Countries Study of some 12,000 middle-aged men in 14 cohorts of 7 countries contrasting widely in the amount and type of dietary fat intake, for which grand scheme he drafted me (as project officer) and many other interested colleagues internationally.
led Keys to an “ecologic fallacy” (i.e., the inappropriate application of group data to inference for individuals). Average α-lipoprotein levels that he had measured by electrophoresis (now characterized as high-density lipoprotein [HDL] or “good” cholesterol) were similar among the mainland, Hawaiian, and California Japanese. Beta, or low-density lipoprotein (LDL) cholesterol, in contrast, was greatly different and correlated directly with the hypothesized dietary fat–disease relation.
Thus, according to Keys in 1956, only LDL or β-lipoprotein cholesterol was of pathogenetic interest in atherosclerosis, which idea he maintained with customary forcefulness and for too long a time. Group data, without several conditions of congruence with those from subjects, and without plausible mechanisms, are not necessarily applicable to individuals. Keys was correct, however, in that LDL, not HDL, accounted well and best for population differences in coronary heart disease rates and, he supposed, might be therefore a main and even necessary factor causing mass differences in atherosclerosis and heart attacks. He erred in assuming that HDL was necessarily unimportant with regard to individual risk, at least within cultures at high risk overall from high average LDL cholesterol levels. At the time, there was little inkling of the different functional roles of lipoprotein fractions or of the differing densities of HDL. Also, the LDL receptors had not yet been discovered.
The overall contributions of “medical Marco Polos”
Thus, by the mid-1950s, Keys and White, their predecessors, and their expanding team of traveler-investigators worldwide had demonstrated social-class differences in serum cholesterol level within several cultures as well as remarkable cross-cultural differences in eating patterns and average serum cholesterol levels. These findings were accompanied by apparent differences in the prevalence of coronary heart disease ascertained by informal cross-sectional surveys of hospital medical wards. They rounded out that sketch by showing graded differences of risk characteristics among Japanese on the mainland and Japanese migrants to Honolulu and Los Angeles, thereby, they believed, indicating the predominant power of cultural exposures, while “holding genes constant.”
Keys and White, using their international academic prominence, inserted these newer concepts and findings of population differences and cultural influences in atherosclerosis into the mainstream of cardiology by presenting the first symposium on CVD epidemiology before a plenary session of the World Congress of Cardiology in the fall of 1954 in Washington, D.C. Panel members included working colleagues Keys, White, and Kimura and also Jeremy Morris of London, Gunnar Biörck of Sweden, and John Higginson of the United Kingdom, experts who embraced clinical, laboratory, and the “new” CVD research strategies of epidemiology. There in Washington, the population evidence was aired in extenso for the first time before the community of cardiologic leaders. It was quickly documented in a small green volume, now a rarity, with the unique title Cardiovascular Epidemiology.
Keys A. White P.D. Cardiovascular Epidemiology Selected Papers from the Second World Congress of Cardiology and Twenty-Seventh Annual Scientific Sessions of the American Heart Association. Hoeber-Harper,
New York, New York1956
The duo of Keys and White went on to organize the Research Committee of the International Society of Cardiology, which became the major disseminator of research guidelines and training through the annual 10-day International Seminars in CVD Epidemiology and Prevention, which, in turn, became a major means for recruitment of worldwide interest in preventive cardiology.
Reaction and Resistance
Medical attention to the early findings of the Dutch travelers was modest and selective but generally positive. Keys's views more often elicited negative reactions. Criticism of his early data and arguments became pointed, often severe and personal, and began as early as the 1952 conference of nutritionists and diabetologists in Amsterdam. Although he reviewed, well and at length, the combined clinical, bench, and epidemiologic evidence and presented well-formulated ideas of the interrelations among habitual diet, lipid metabolism, and atherosclerotic diseases,
he also tended to propound his arguments in a blunt and cocksure manner. In Amsterdam, he was immediately challenged from the floor, and in discussion, his arguments were overwhelmingly rejected. A single exception in that vast international audience came from the diminutive, mild-mannered Dutchman Johannes Juda Groen.
In that public session and later at a private dinner, Groen supported Keys's ideas about cultural and dietary influences on vascular diseases. Furthermore, Groen had, among his other research endeavors, called attention in the early 1950s to wartime changes in nutrition and CVD trends in The Netherlands. And in the same year as the Amsterdam meeting, Groen had published a controlled feeding experiment modifying serum cholesterol levels.
Thus, Keys and Groen early found their independent activities and intellectual wavelengths virtually synchronous.
Soon after the boisterous Amsterdam reaction, Keys gave essentially the same lecture in New York City on the invitation of Ernst Boas and Frederick Epstein, before internists of the Mt. Sinai Hospital, after which it was promptly published.
At Mt. Sinai, it may have been the anthropologic views of the noted cultural anthropologist Franz Boas and his internist son Ernst that created a more receptive audience to Keys's arguments for sociocultural influences. In fact, Ernst Boas and Epstein already were engaged in cross-cultural comparisons of lifestyle and coronary disease rates among Italian and Jewish garment workers in the city.
For whatever reason, Keys's message received immediate understanding and an enthusiastic reception at Mt. Sinai. Subsequently, he often declared immodestly that Fred Epstein at that moment had become a Keysian “disciple.”
More meaningful for this story, however, was Keys's confrontation at the first meeting of the World Health Organization Expert Committee on the Pathogenesis of Atherosclerosis, held in Geneva in the fall of 1955. On arrival, Keys had little reason to expect a frontal attack on his dietary hypothesis of coronary heart disease. After all, the assembly of medical leaders from around the globe included Keys's friends and like-minded collaborators, such as Johannes Groen and Noboru Kimura, and Haqvin Malmros of Sweden, as well as Jerry Morris of London and James Watt from the United States National Heart Institute, all experts involved in early population findings on lifestyle and disease and at least sympathetic to public health concepts and strategies.
Others at the meeting, however, notably Sir George Pickering, Regis Professor of Medicine at Oxford, found themselves unconvinced by Keys's evidence and arguments and were more than a little put off by his apparent certitude on the matter of diet.
In telling me the story some years after the meeting, Pickering described the Geneva discussion as “lively, tending to tangents and tirades.” At a critical moment in Keys's presentation, Sir George was moved to interrupt him with a query along these lines: “Tell us, Professor Keys, if you would be so kind, what is the single best piece of evidence you can cite in support of your thesis about diet and coronary heart disease?”
Keys fell headlong into the debater's trap. He proceeded to cite a single piece of evidence (we do not know which piece). Sir George and the assembled peers were easily able to diminish that bit of evidence and promptly did so. By then it was too late for Keys to summon the totality and congruence of the clinical, bench, and epidemiologic evidence and to construct thereby a convincing argument for his hypothesis.
As the now legendary story has it, Keys was so stung by this exchange that he left the Geneva meeting intent on gathering definitive evidence to establish, or to refute, the “diet-heart hypothesis.” Out of this singular, humiliating experience may well have come the challenge, the motivation, and eventually, the implementation of the Seven Countries Study.
It stimulated, as well, the Minnesota series of critical metabolic experiments that Keys and colleagues set in operation in the 1950s, which demonstrated predictable effects of different dietary fatty acids on blood cholesterol level, an essential component of the diet–lipid–heart disease hypothesis.
An arm-chair epidemiologic strategy carried out in the comfort of home or office also took place during this period to complement the early cross-cultural travel adventures. Incomplete as it was, it too was not an effort to be sneezed at. Ancel Keys did his share of such ruminations, illustrated in the 1947 photograph (Figure 4) showing apparent mortality trends in Minneapolis from 1926 to 1946 and their extrapolation into the 1950s, an important idea supported by weak evidence.
Another example, seen in Figure 7, was graphed for his lecture in Amsterdam in 1952 and at Mt. Sinai Hospital in New York in January 1953, when Keys plotted the ecologic associations of international Food and Agriculture Organization data on diet fat consumption and World Health Organization vital statistics on degenerative CVD deaths, from countries where he had some confidence in the diet data quality.
for the obvious deficiencies of such ecologic attempts among few select populations. In contrast, it was respected by others as a very early tentative step and as only part of a broader review of all relevant evidence about diet, blood lipids, and CVDs. More important, the review was immediately followed by increasingly strong empiric research on the diet-heart relations by Keys and colleagues and by many others working in laboratory, clinic, and populations, the efforts of whom culminated in the risk factor paradigm so useful to practice and public policy today.
A process leading to guidelines for causal inference from correlations
Moreover, Keys's Mt. Sinai review and his simplistic ecologic correlation of diet fat and disease thrust the diet-heart argument into vigorous and fruitful international debate among the top epidemiologic-statistical leadership. Fortuitously, and in a traceable direct line of successive published commentary following Yerushalmy and Hilleboe's
critique of Keys's ecologic correlation, the debate on diet-heart associations led to progressive refinements of the logic for arriving at causal inference from statistical correlations. This discussion, with the evolution of guidelines, greatly strengthened the interpretation of evidence and the strategies of design and analysis in chronic disease epidemiology.
United States Department of Health and Human Services Smoking and Health: Report of the Advisory Committee of the Surgeon General of the Public Health Service.
U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health,
However, in the 1950s, much of the scientific community found Keys's early proposals too radical a departure from the passive clinical paradigm that atherosclerosis was a natural phenomenon of aging. At any rate, the academy insisted on an elusive experimental proof and resisted making public policy and taking health action in the face of the perceived uncertainties.
The “Medical Marco Polos” in Perspective
The world travelers depicted here combined curiosity, imagination, ingenuity, and great energy to achieve useful observations across cultures, either in the course of assigned tasks or in purposeful explorations. Understanding accelerated with each step in this peripatetic and romantic period of CVD epidemiologic history, with its bright ideas and their mixed reception and with its increasing evidence about the relation of culture to disease risk. Formal population studies of “proper” design were proposed. The thinking of Keys and colleagues, their theses and informal findings of population contrasts, cried out for translation into studies with formal design, appropriate samples, objective and quantitative measurement, and sophisticated analysis.
Contemporaneously, these international efforts were joined by pioneering studies of healthy subjects within cultures, which had the most visible impact on chronic disease epidemiology and preventive cardiology. The long-term, prospective, United States Public Health Service–run Framingham Heart Study of a community cohort, begun in 1948, was the study of greatest renown; it flourishes today. It was joined early by Jeremy Morris's occupational comparisons among London transport workers in 1949, myriad other prospective studies, and trials of cholesterol lowering in the 1950s in Scandinavia, Continental Europe, and the United States. The river of research on CVD prevention swelled and rapidly brought evidence for the environmental and cultural, as well as biologic causes of and preventive measures for the common CVDs.
Modern cardiologic diagnosis and awareness of epidemic heart attack converged with advances in understanding of the pathophysiology of atherosclerosis and with the keen observations and causal hypotheses of a handful of clinical and laboratory investigators, public health officials, and “medical Marco Polos.” Formal beginnings of preventive researches in CVD can thus be placed firmly within a few years and a few studies at mid-20th century, from which have evolved the sound public health discipline of CVD epidemiology and the active practice of preventive cardiology.
These and other stories are slated for a book on the origins of research in prevention of heart attacks, to be published by Oxford University Press.
Keys A. White P.D. Cardiovascular Epidemiology Selected Papers from the Second World Congress of Cardiology and Twenty-Seventh Annual Scientific Sessions of the American Heart Association. Hoeber-Harper,
New York, New York1956
United States Department of Health and Human Services
Smoking and Health: Report of the Advisory Committee of the Surgeon General of the Public Health Service.
U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health,
This work was supported in part by Publications Grant 5G13-LM008214-02 from the National Institutes of Health , Bethesda, Maryland; The Frederick Epstein Fund of Zurich, Zurich, Switzerland; and the Councils on Epidemiology and Prevention of the American Heart Association, Dallas, Texas, and the International Society of Cardiology.