The Man Who Touched His Own Heart

by Rob Dunn

  Keys seemed to be able to do anything he set his mind to. Not only could he make discoveries; he could also convince others to follow his lead. Just how far others would follow him was about to be tested. Keys decided he wanted to figure out how to “predict both short time and long range physiological results from a given mode of life, personal habits, activity, and diet.”2 This would require experiments with humans, experiments more extreme than those he was willing to conduct on himself.

  The most obvious immediate need in terms of understanding the influence of diet was in the context of war. World War II had reached its most awful moment. Soldiers of many nations were starving, as were civilians, but little was known about what happened to the body during such conditions or how to heal the body of someone who had been starved. Keys decided to enlist conscientious objectors, many of them Quakers, to be part of a study on starvation. The men gathered in the basement of the football stadium at the University of Minnesota where Keys had set up his lab. (The space was ample, though it vibrated with roars each time a touchdown was scored.) There, they were subjected to a variety of low-calorie or starvation diets. Men grew so hungry they became desperate, but the shame of leaving the project was akin to the shame of abandoning a loved country, and so no one left. Personally, Keys was torn about the trials he was inflicting upon these men. Yet, overall, the men themselves believed in their mission, even when it became most difficult. In later years and with very few exceptions, the men in the starvation trial said that if they had it to do over, they would participate in the study again. The privations they experienced were extreme (one man chopped his fingers off with an ax in order to go to the hospital and get out of the study without seeming weak), but so were the benefits their volunteerism offered society.3 The study resulted in the single most comprehensive treatise on starvation ever published. Medics used the results when starving concentration-camp prisoners were freed and needed to be fed. It is used today in treating individuals with extreme eating disorders. Keys wanted to do work that lasted and that mattered, and he succeeded.

  From his work on fish, K rations, and starvation, Keys came to contemplate something even more central to postwar humanity: he wanted to understand what everyone should eat, not just soldiers or starving people. To Keys, diet and lifestyle were a unifying feature of his research, even when he was studying eels or killifish. And, at least for humans, cholesterol seemed to be a key player in the story of diet, lifestyle, and health. Cholesterol, he knew, was associated in some way with atherosclerosis and heart disease. Keys wrote about the cholesterol in cow’s milk and its potentially negative effects as early as 1948. Initially, Keys did not know that there were different types of cholesterol.4 Nor did he know that the cholesterol humans consume has a relatively modest effect on the cholesterol levels in the blood, but he was clever enough to make his way.

  Everywhere Keys looked, he saw the importance of better understanding cholesterol, atherosclerosis, and heart disease. Around Keys in Minnesota, in the years after the war, heart disease seemed to be epidemic among business executives, and Keys well knew that those executives consumed, as if by law, rich diets dense and slippery with cholesterol—eggs, butter, milk, more butter. Keys, among others, thought the extent of this phenomenon new and troubling. Keys stalked the obituaries and found them full of wealthy, powerful men having heart attacks.5 In 1900, pneumonia was the leading cause of death, and by the time Keys was checking the obituaries, in the early 1950s, heart disease seemed to have taken its place.

  Keys came to believe that consuming more animal fat caused the body to produce too much cholesterol, which, in combination with the consumption of cholesterol itself, led to higher rates of atherosclerosis. Such a link had not yet been documented in the lab or among humans yet, but with his hypothesis in hand, Keys envisioned an informal test. He noted the Minneapolis businessmen he saw dying all around him and decided to observationally study 286 of them. He would give each man a physical examination, including blood tests for cholesterol levels, a few times a year for what would ultimately be twenty-five years. He also tracked their deaths. Keys could be persuasive. The men agreed. He had convinced conscientious objectors to starve themselves nearly to death in the name of the war against Germany. Now he just had to convince men to submit to blood tests and the like in the name of the war on heart disease. The men who agreed, including Bernie Bierman, the famous University of Minnesota football coach, and Edward John Thye, then the governor, were all excited to be part of Keys’s vision.

  Keys expected to see higher blood cholesterol levels in those men with diets richer in animal fat and cholesterol; that is just what he found. The men with diets richer in animal fat had more cholesterol in their blood. Because more fat was associated with more cholesterol and more plaques, Keys began to wonder if a change in diet could reduce cholesterol and plaque formation and therefore potentially prevent heart disease.

  Keys’s data suggested that a simple dietary change such as reducing saturated fat intake might decrease heart-disease risk, but he needed more direct evidence. For one thing, while there did seem to be a link between fat in the diet and cholesterol in the blood, maybe there were other factors at play. Also, the cholesterol levels in the businessmen, who had high rates of heart attacks, were not any higher than those in the general public. Keys decided to do an experiment in order to really isolate the effects of fats. He gained permission to manipulate the diet of thirty patients at the Hastings Hospital for the Insane. Fifteen patients were given diets rich in animal fats, akin to what patients who weren’t in the experiment received; fifteen were given diets in which the animal fats had been replaced with starches. The cholesterol in the individuals eating the animal-fat diets went up, while that in the individuals consuming the starches went down. Keys was as convinced of his idea as ever, even though in retrospect, the change appears very modest—statistically significant but perhaps not practically so. Cholesterol in the blood increases when animal fats are consumed. The cholesterol shepherds the animal fat, which isn’t soluble in water, through the blood. That the levels of cholesterol in the body increase when one eats animal fat is not, in and of itself, a malfunction. It is the body’s marvelous response to a hard-to-move compound, but somehow, Keys felt sure blood cholesterol levels must be associated with the extent to which high cholesterol leads to atherosclerosis.

  The executive study and the small experiment were useful in informing Keys’s intuition. They convinced him he was on the right track, but he needed something grander in order to convince the public and his colleagues. The ideal would have been to do a global experiment in which he assigned people around the world to one of several randomized diets and lifestyles. Such an experiment was not and will never be possible, not even for Keys.6 But Keys was going to try the next best thing, a pair of “natural experiments.”

  Keys’s first approach was to study what happened to people from countries around the world when they move to a new region with a new culture; Hawaii, in this case. This is the sort of natural experiment that evolutionary biologists and ecologists do. The British ecologist Charles Elton pioneered one version of this approach. He realized that when species from one place were put in another, they represented, however unintentionally, a test of ecological theory. Scientists could observe which species succeeded where and how fast they evolved. Keys, thinking again like an ecologist or a physiologist, saw immigrant humans in the same vein. These individuals from around the world would all be subjected to the diets of their new places, but they came from very different genetic, cultural, and dietary backgrounds. In theory, Keys could observe not only the influence of Hawaiian-American life but also whether that influence affected different peoples differently.

  Japanese immigrants to Hawaii were an interesting test of this hypothesis because Hawaiians, like other Americans, consume much more animal fat than do people in many other regions of the world, including Japan. Sure enough, when Japanese people, with an average cholesterol level of 12
0, moved to Hawaii, their cholesterol levels rose to about 183. Their risk of heart disease rose too. Similarly, when Japanese people moved to Los Angeles, where their diets were even more Americanized, their cholesterol levels increased to 220, and their rate of heart disease increased as well, even relative to that of Japanese people in Hawaii. This evidence was compelling, though it had several problems, the biggest of which was that becoming Hawaiian or Los Angeleno meant many other changes in addition to the shift to a diet with more animal fat. Diet is enmeshed in culture, and moving to a place means becoming part of it. A fattier diet was part of that culture, but so were many other things, such as more hours spent indoors, more smoking, a lack of vitamin D, increases in salt and sugar intake, altered exposure to parasites, and increased exposure to pollution, to name a few.

  Beginning in the late 1950s, Keys decided to do another natural experiment. He would compare the diets of people living in many different countries and, in turn, their rates of atherosclerosis and heart disease. His first two countries were Japan, where the diet is based on fish and rice with little animal fat, and Finland, where, as Keys would put it, people sometimes dined on meals consisting entirely of butter and cheese. When Keys compared these two regions (which he did by training doctors in both countries to perform a standardized set of repeated measurements on patients; once again, Keys was convincing people to work for a common goal, and once again, it was his goal), he found that in the country where people ate more fat, Finland, there was more heart disease. Keys then met more people and enlisted more countries in his study. By 1956, Greece, Yugoslavia (sites now in Croatia), Italy, and the Netherlands had signed on, leading the project to be called the Seven Countries Study; there were seven countries but even more individual sites.

  The Seven Countries Study required Keys to fly around the world again and again coordinating activities. It was not easy. Each data point was hard fought, and there were so many data points; Keys and the vast team he brought together worked with more than ten thousand people (as he would later say, from Bantu tribesmen to Italian contadini). The results, and even the endeavor itself, were controversial, but Keys was, as ever, compelling. He began to gather endorsements—for lack of a better word—from influential people, such as Paul Dudley White, President Eisenhower’s personal physician. With White’s important blessing, the study gained more traction among doctors and scientists than it might have otherwise; it also gained funding. The study still continues to this day (minus a few countries that weren’t able to find funding to continue), but for Keys, the big result came after the first five years, at a point when he plotted a clean line on a graph showing that the countries where people had the most animal fat in their diets were also the countries where people had the most deaths due to heart disease, heart attacks in particular. Finland had the highest incidence of heart-attack deaths, 992 out of 10,000 deaths, and its people consumed the most saturated fat, animal fat. People in Crete, with their lifestyle of sun, relaxed days, and diets rich in vegetables and olive oil, had lower blood cholesterol, and just 9 deaths out of 10,000 were due to heart attacks. These patterns have held up over the years. As of twenty-five years after the beginning of the study, the number of deaths due to heart disease in a region was still strongly positively associated with saturated-fat consumption (particularly with butter consumption). Also, time added another insight—those deaths tended to be negatively associated with the consumption of olive oil, fish, and wine. The take-home, in its simplest form, was this: eat less saturated fat, eat more olive oil and fish, drink some wine.

  Based on their results, Ancel Keys and his wife, Margaret, who had by this time become intimately involved in both Ancel’s research and the marketing thereof, could simply have written a scientific paper. That would have been the standard next step, but Ancel Keys was not a standard man, nor were he and his wife a typical team. They wanted something bigger. They had, they thought, figured out a way to prevent atherosclerosis by lowering cholesterol through a change in diet. The original data from the Seven Countries Study were not data on atherosclerosis or even cholesterol—they were data on heart disease and fat intake—but for Keys, the link between the two was sufficiently compelling for him to advocate change. Most scientists never advocate anything; they never make suggestions for fear of not knowing quite enough, not yet. Keys would not be one of those scientists—he would hold forth, he would recommend.

  What the Keyses came to advocate was, in essence, a K ration for America—a diet that would get Americans through not war but their daily lives. The Keyses focused on the Mediterranean diet—something along the lines of what Southern Italians consumed, though with any vegetable oil taking the place of olive oil and other simplifications to make things easy. It was not the only diet that the Keyses thought might work (the Japanese in the study actually seemed to have the lowest cholesterol and longest lives), but it was one they thought might both lower cholesterol and spread person to person, engendering financial success for the Keyses and public-health success for humanity. While the Keyses’ version of the diet had features of the Mediterranean diet, it was not a Mediterranean diet, and it certainly did not entail an entire Mediterranean lifestyle. The Keyses had to assume that the elements they kept in their recommended diet maintained the region’s central character: its necessary influence on cholesterol and, they thought, heart disease. Their first book was Eat Well and Stay Well, which came out in 1959. The book, unlike any diet book ever written before or since, began with a two-hundred-page introduction to the work the Keyses had done and the science of diet and heart disease, followed by meals that would reduce cholesterol levels and the risk of death due to heart disease. The work was so compelling and so fully embraced that the lessons from the book became a part of how Americans ate, or at least how Americans tried to eat. It also became a part of how doctors thought about heart disease. In 1959, before Eat Well, the American Heart Association took the position that the link between saturated fat, cholesterol, and atherosclerosis was not well established. In 1961, after the publication of Eat Well, and with Keys by then on its board, the American Heart Association advised Americans to avoid saturated fat and cholesterol. Eschew, they and Keys said, butter, meat, egg yolks, and full-fat milk. The American Heart Association began to recommend a diet that bore a striking resemblance to that which the Keyses advocated in Eat Well. That same year, Keys was on the cover of Time magazine, portrayed as the genius behind the modern, scientifically informed diet.7

  The relationship between the number of deaths due to coronary heart disease in the Seven Countries Study and the average quantity of butter and margarine (two key contributors to saturated fat) consumed in each region. After accounting for the effects of saturated fats, countries in which individuals consume more olive oil (white circles) or fish (gray circles; Japan) have fewer deaths due to coronary heart disease than would be expected based on their saturated fat alone.

  The Keyses’ book ushered in a change in the American relationship to food; if you are alive today in America, your perspective on food has been shaped by the Keyses. The Keyses convinced Americans—including scientists and doctors—that saturated fat was bad. Animal fat was regarded as particularly bad, but all fat came to be suspect; this was a novel idea. Before the Keyses, people ate the food of their respective cultures. There was, of course, good food and bad food, but goods and bads were handed down from one generation to the next. There were no experts at the dinner table and few pronouncements heard on the news about what would kill you or allow you to live forever. The Keyses helped to create the seed of a culture in which the experts on what we should eat were scientists.

  In retrospect, we can identify complexities associated with the story of diet and heart disease that the Keyses offered America, complexities people missed in their attempts to simply consume the advice as though gobbling up K rations. The first had to do with the science. Early on, Keys had become relatively confident about the relationship between dietary levels of satura
ted fat and cholesterol, blood cholesterol levels, atherosclerosis, and heart disease. The Seven Countries Study data supported what Ancel Keys had already come to believe through his observations and small experiments and those of others. The data were in line with what Keys expected and what many analyses confirmed, but, as later analyses have shown, they were just part of the puzzle, a puzzle with many pieces. The pieces the Keyses figured out seem to hold together, and they hold up to the scrutiny of time, but they are just the edge pieces of a very big puzzle.

  In one unfinished section of the puzzle were more pieces of the cholesterol story. When the Keyses started working, there was just one kind of cholesterol. There is still just one, but that one type of cholesterol travels around the body in multiple forms, forms that were first identified by John Gofman, a physicist who worked with Glenn Seaborg on the plutonium nuclear bomb and then went on to get his medical degree. What Gofman figured out was that the two most common ways of travel for cholesterol are LDL and HDL, low-density lipoproteins and high-density lipoproteins. LDL and HDL are familiar acronyms, but most of us think about them in the wrong way. These are not types of cholesterol; they are more like types of boats into which cholesterol is incorporated for the ride. Cholesterol and triglycerides are not soluble in water (or blood) and they can’t go it alone through your blood vessels very well. High-density lipoproteins have a relatively big hunk of protein and relatively little in the way of cholesterol and triglycerides, and so they are dense, heavy for their size. Low-density lipoproteins are the opposite, with lots of cholesterol and triglycerides relative to the amount of heavy protein; they are less dense, light for their size. The body uses these two boats in different ways. Cholesterol in HDL tends to be removed from the blood and ferried to the liver, where it is broken down and excreted in bile. Cholesterol in LDL, by contrast, tends to be carried to the organs. Having more HDL boats can actually help reduce your LDL because they gather up the cholesterol left over in organs or in artery walls and drag it to the liver to be degraded. But it is more complex. Cholesterol appears to travel around the body in at least three other forms, and each of these forms plays a specific and only partially understood role. What is more, the real problem is not the LDL cholesterol itself. The real problem is that the LDL cholesterol, in bouncing around the body, gets damaged—oxidized—in much the way that a firm, fresh pear turns brown. The oxidization of the LDL leads white blood cells to swoop in, and it is the swooping white blood cells that initiate plaque formation. The white blood cells produce a kind of foam that is itself oxidized, attracting even more white blood cells. All of this is balanced, to a lesser or greater degree, by HDL, which can gather up the cholesterol in the LDL and drag it to the liver. When it does, the plaques can break up, the immune system having finally realized that everything is cool.