Chocolate: The Darker the Better.

EVERY WINTER, AROUND FEBRUARY 14, the food and health pages run amusing pieces on why women crave chocolate, or how it came to be a.s.sociated with love and romance- and might even get you some. I used to find these reruns tedious, especially when they used the term chocoholic, but I have new respect for chocolate, which I consider one of the great food-drugs, along with wine and cayenne peppers. Now I skip the s.e.xy quotes and scan the articles for hard facts on the dark, complex bean of the cacao tree, a native of lush jungles around the equator, from Hawaii to Venezuela to Nigeria. The tree and bean are called cacao - rhymes with cow. The liquor, powder, and drink are cocoa.

Everyone has his poison (or ought to), and I'd probably eat chocolate no matter what, but for chocolate lovers seeking nutri1 tional validation, here is a sneak preview at the good news: the saturated and monounsaturated fats in cocoa b.u.t.ter are good for cholesterol; cocoa powder is rich in antioxidants and contains mild antidepressants.

Twelve hundred years before Christ, the Olmec of modern Mexico grew cacao trees and made the beans into a drink laced with chilies, herbs, and honey. The Mayans and Aztecs believed the G.o.d of Agriculture brought the cacao tree from paradise. At weddings, they served a cocoa drink called xocalatl (warm or bitter liquid), admired for its ability to enhance energy, pa.s.sion, and s.e.xual performance. In 1527, Hernan Cortes found the cocoa bean in Montezuma's court and took it home to Spain, where it was doctored with sugar and vanilla.

Soon chocolate spread pleasure across Europe, and in 1753 the Swedish naturalist Carolus Linnaeus renamed the cacao tree Theobroma cacao, Greek for "elixir of the G.o.ds." But chocolate lovers were not content to leave the divine bean alone; they wanted to perfect it. In 1879, chocolate was rendered smooth and creamy by conching, a trick devised by Rodolphe Lindt of Berne, Switzerland, one of many chocolate inventor-industrialists. A small producer in Sicily called Bonajuto sells the grainy, unconched kind made with the nineteenth-century method. It didn't knock me out when I tried it. I'm all for traditional methods of production, but to me, Lindt was a genius; with chocolate, I think, the smoother, the better.

HOW CACAO BEANS BECOME CHOCOLATE.

First cacao beans are removed from the pods. The best chocolate makers ferment the beans to develop flavor. Next the beans are roasted (again for flavor), sh.e.l.led, and broken up into little shards called nibs. Nibs are then ground and heated to make cocoa liquor. If you remove all the cocoa b.u.t.ter from the liquor, what's left is pure cocoa powder. If you add more cocoa b.u.t.ter, plus sugar and vanilla, to the liquor, you get a chocolate bar.

The cacao tree is an unusual plant. Cocoa powder contains large amounts of calcium, copper, magnesium, phosphorus, and pota.s.sium, and more iron than any vegetable. It is very rich in polyphenols, particularly a group called flavonoids, which account for the rich pigment in red wine, cherries, and tea. These antioxidants promote vascular health, prevent LDL oxidation, lower blood pressure, reduce blood clots, and fight cancer. A one-and-a-half-ounce (40-gram) bar of milk chocolate contains as many antioxidants as a five-ounce (150-ml) gla.s.s of red wine. Polyphenols are found in cocoa solids, not cocoa b.u.t.ter. Thus pure cocoa powder has the most antioxidants by weight, then dark chocolate, and finally milk chocolate. White chocolate- made of cocoa b.u.t.ter without any cocoa powder- has none at all.

The polyphenols in chocolate may prevent obesity, diabetes, and high blood pressure- all risk factors for heart disease. In 2005, the American Journal of Clinical Nutrition reported that eating three and a half ounces (100 grams) of dark chocolate daily decreased blood pressure and significantly improved sugar metabolism by increasing sensitivity to insulin. Insulin sensitivity is desirable; recall that in diabetes, the cells are deaf to insulin. White chocolate did not have the same effect.29 ANTIOXIDANTS IN CHOCOLATE AND OTHER FOODS.

Each of the following contains about 200 milligrams of polyphenols. Note that chocolate from fermented beans contains more polyphenols, and dark chocolate contains more polyphenols than milk chocolate.

* 1.5 oz milk chocolate.

* 1 gla.s.s red wine (5 oz).

* 12 gla.s.ses white wine (5 oz each).

* 2 cups tea.

* 4 apples.

* 5 servings of onions.

* 3 gla.s.ses black currant juice.

* 7 gla.s.ses of orange juice.

Source: John Ashton and Suzy Ashton, A Chocolate a Day Keeps the Doctor Away.

The fats in chocolate- mostly monounsaturated and saturated- are also healthy. Chocolate is unctuous because cocoa b.u.t.ter melts at mouth temperature. The better chocolate bars have added cocoa b.u.t.ter, which is roughly equal parts stearic, palmitic, and oleic acid. Extra stearic acid is converted to oleic acid, so the net effect of these fats is good for cholesterol. A 2004 study in Free Radical Biology and Medicine found that chocolate increased HDL and reduced oxidation of cholesterol. Oxidized cholesterol causes atherosclerosis. Chocolate keeps well because saturated fats are stable, and any oxidation from heat or light is inhibited by cocoa's abundant polyphenols.

COCOA b.u.t.tER IS GOOD FOR YOUR HEART.

Is chocolate addictive? The Chocolate Information Center- a research lab funded by the Mars company- prefers to call the craving for chocolate a "strong desire." "Theoretically," its scientists write cautiously, chocolate could "contribute to feelings of well-being." Chocolate contains the stimulants theobromine and caffeine; tyramine and phenylethylamine, which are uppers; and anandamide, which mimics cannabinoids, natural pain killers. (Anandamide derives from the Sanskrit ananda, "internal bliss.") But the chocolate scientists note that these natural drugs are found only in trace amounts in chocolate. Moreover, they are also found in many foods we don't crave. Caffeine is an exception; it is certainly habit-forming, but there is much more caffeine in coffee and tea than in chocolate.

Craving a certain food is sometimes regarded as signaling an acute nutritional deficiency, but in the case of chocolate, I'm not convinced. Chocolate is rich in magnesium. If a woman needs- and thus craves- magnesium before her period, she might just as easily have a strong desire to eat magnesium-rich foods, like broccoli, tofu, and kidney beans. Rushing to the corner shop for broccoli, however, is not part of PMS folklore.

What often goes unmentioned in discussion about the desire for chocolate is sugar. I suspect that sugar plays a large part in the urge to finish a bar of chocolate once you've torn into it. Sugar quickly brings on a gentle high, which is why so many people crave it in low moments. Unfortunately, the high is often followed by a crash.

Studies show that capsules containing the compounds in cocoa don't satisfy the desire as well as real chocolate, which strongly suggests that the aroma, flavor, and creaminess of chocolate, more than any mood-enhancing chemicals, are key factors in chocolate craving. In Chocolate: A Bittersweet Saga of Dark and Light, Mort Rosenblum considers all these factors, including the mood-enhancing agents in chocolate, and concludes that chocolate fails the two tests of addiction: "It is not dangerous to the human organism. And no symptoms of withdrawal appear when consistently high consumption is abruptly stopped."

NAKED CHOCOLATE.

Nibs are little pieces of fermented, roasted, and sh.e.l.led cacao beans- the raw material of all chocolate. More like nuts than candy, crunchy nibs are about half cocoa b.u.t.ter. They have a tannic flavor, like espresso or red wine. Try them in place of chocolate chips in cookies and brownies or sprinkle on ice cream. Toss nibs on roasted pumpkin or add to Mexican mole. They pack a little jolt of caffeine.

How much chocolate is good for you? John Ashton and Suzy Ashton, the authors of A Chocolate a Day Keeps the Doctor Away, who are unabashed enthusiasts, recommend no more than two ounces (55 grams) of chocolate each day, preferably dark chocolate, which has less sugar and more antioxidants. It also has more flavor than milk chocolate, which is why connoisseurs the world over prefer it. Once you're used to the bold, complex flavors of chocolate unmasked by sugar, the average milk chocolate will seem cloying. Not, I hasten to add, that there is anything wrong with milk, which goes very nicely with chocolate. After all, ganache is nothing more than blended cream and chocolate. If you like milk chocolate, look for a bar with more cocoa than sugar, such as one made by Scharffen Berger, which is 41 percent cocoa, about twice the amount in a typical bar.

Sugar makes me fat and grumpy, so gradually I weaned myself from dark (70 percent cocoa) to very dark (85 percent) chocolate bars. Occasionally, I will make a cup of cocoa with unsweetened chocolate and very fresh gra.s.s-fed raw cream, which is just sweet enough to balance the bitterness. Do buy the best unsweetened and dark chocolate you can afford; without sugar there is no concealing shoddy chocolate. Cheap chocolate can be bitter or metallic.

With chocolate desserts I recommend the same thing: less sugar. When you make chocolate mousse (or any dessert), use half the sugar called for. You will taste the chocolate first, instead of the sugar. When sweetness is not the dominant sensation, your appreciation for a particular flavor will grow; chocolate may seem fruity, smoky, or herbal, for example. Other flavors will change subtly, too. Next to dark chocolate, savory foods such as almonds, coconut, and cream are almost sweet. If you're used to sugary desserts, you may find "half-the-sugar" desserts not quite to your taste at first. But I swear by this rule of thumb. I follow it every time I try a new recipe for ice cream or pumpkin pie or anything else. As my mother likes to say, "If it's sweet on the first bite, it will be too sweet by the last."

9.

Beyond Cholesterol.

What Is Cholesterol?

WHEN I TOLD MY FRIEND Wendell Steavenson, an Anglo-American writer, I was writing a book about why b.u.t.ter is good for you, her comment was typically arch. "Cholesterol," she said, mock solemn, "only exists in America." I knew exactly what she meant. In much of the world- perhaps Wendy was thinking of London, Baghdad, Beirut, or Tbilisi, to name but a few cities where she has lived- people aren't morbidly afraid of traditional foods like cream and lamb. Not yet, anyway.

The American anticholesterol campaign, which British experts gently mock as "know-your-number" medicine, baffles foreigners. Here at home, it mostly inspires anxiety. When you sit down to eat with health-conscious Americans, the subject of cholesterol is hard to avoid, but the conversation seldom moves beyond weak jokes about clogged arteries.

We have been taught to fear this scoundrel called cholesterol, and fear it we do, yet most people have no idea what cholesterol is. It is often called a fat, but cholesterol is actually a sterol, a kind of alcohol. Cholesterol is part of all animal cell membranes. It makes up much of brain and nervous tissue, and it's an important part of organs including the heart, liver, and kidney. It is so vital to the developing brain that defects in cholesterol metabolism cause mental r.e.t.a.r.dation.1 Cholesterol is necessary to make vitamin D, bile acids (which digest fats), adrenal hormones, and the s.e.x hormones estrogen and testosterone.

These roles are well known, at least to cholesterol experts. Then, in a college textbook, I came across a striking statement: cholesterol is a repair molecule. At first I didn't understand it- or quite believe it. After all, cholesterol is known for doing damage, not healing. But apparently that's not quite how cholesterol works. Let me explain by introducing the famous lipoproteins.

Low-density lipoprotein (LDL), often called the "bad" cholesterol, and its counterpart, high-density lipoprotein (HDL), or "good" cholesterol, are not forms of cholesterol at all, but vehicles. Like little boats with a waxy cargo, LDL and HDL ferry cholesterol around the body. LDL carries cholesterol from the liver to the tissues (including blood), and HDL carries cholesterol from the tissues back to the liver.2 In every healthy person, the lipoproteins help cholesterol go about its ch.o.r.es, digesting fats here and making estrogen there. The body needs both LDL and HDL. According to the Journal of American Physicians and Surgeons, the "good" and "bad" cholesterol story is "overly simplistic and not supported by the evidence." 3 Repair is one of cholesterol's many tasks. When arterial walls are damaged, cholesterol rushes to the scene on a dinghy piloted by LDL to fix them.4 As the authors of Human Nutrition and Dietetics describe low-density lipoproteins, "their role is to deliver cholesterol to tissues for the vital functions of membrane synthesis and repair." In Know Your Fats, Mary Enig writes: "Cholesterol is used by the body as a raw material for the healing process. This is the reason injured areas in the arteries (as in atherosclerosis) or the lungs (as in tuberculosis) have cholesterol along with several other components (such as calcium and collagen) in the 'scar' tissue [they've] formed to heal the 'wound.'"

This would explain why high LDL is sometimes linked to heart disease. Many people with heart disease have damaged arteries, and cholesterol travels on LDL to heal them. Just because you see fire fighters at burning buildings does not mean they start fires.

Cholesterol comes from two sources. The body makes cholesterol in the brain and the liver, which makes about 1,500 milligrams daily. The other source is the diet. Only animal foods contain cholesterol. It is stored in the fat of dairy foods and the muscle of animal protein. Thus beef, pork, and poultry, although they vary in fat content, have similar amounts of cholesterol (about 20 milligrams per ounce). That means tr.i.m.m.i.n.g the fat from meat will reduce the fat but not cholesterol content, whereas skim milk contains less cholesterol than whole milk. Egg yolks and milk are particularly rich in cholesterol because baby animals need large amounts to build brain cells.

Experts once thought that eating cholesterol raised blood cholesterol. In 1968, they advised us to limit dietary cholesterol to 300 milligrams daily. This figure was not only unrealistic- with 275 to 300 milligrams of cholesterol, eating one egg would put you at or near the limit- but also arbitrary, as Gina Mallet discovered. Now we know that blood cholesterol is largely determined by metabolism- how the body makes, uses, and disposes of cholesterol. "The amount of cholesterol in food is not very strongly linked to cholesterol levels in the blood," says a report from the Harvard School of Public Health.5 How much cholesterol do you need to eat? In theory, none; your body will make enough. But there are good reasons to consume cholesterol, which is harmless in its natural form. First, infants and children under two don't produce enough; cholesterol must be part of their diets, which is why breast milk has plenty.6 The second reason to consume cholesterol applies to all ages: avoiding it entirely would mean shunning highly nutritious foods. Liver, meat, shrimp, b.u.t.ter, and eggs offer complete protein, omega-3 fats, and vitamins A, B12, and D- all vital nutrients not found in plants. It is not possible to separate the cholesterol from the nutrients in these foods. Older people may even benefit from eating cholesterol. In 1995, researchers found that cholesterol in eggs aids older people with declining memory.7 The body aims to keep cholesterol levels steady. Thus the more cholesterol you eat, the less the liver makes, and the less you eat, the more the liver makes. That explains why vegans and vegetarians, who eat few animal foods or none at all, can have high cholesterol. Moreover, up to 50 percent of cholesterol is determined by genes, not diet.8 As people with a family history of high blood cholesterol know, diet has little effect on blood levels.

If you have normal cholesterol metabolism, you may eat real foods without fear. As we've seen, people who eat traditional foods rich in cholesterol and saturated fat don't have high cholesterol or more heart disease. Other traditional foods, meanwhile, have salutary effects on health in general and heart disease in particular; they include fish, red wine, chocolate, and olive oil.

Industrial foods are the real villain in heart disease. The main offenders are trans fats, corn oil, and sugar. As we've seen, trans fats promote atherosclerosis and clotting; polyunsaturated vegetable oils lower HDL; and sugar depletes B vitamins and raises triglycerides. All these effects are bad for the heart. The actual culprits are easy to spot for anyone who cares to read more than casually about diet and heart disease. So why did a perfectly useful molecule called cholesterol, which we've been consuming in liver, eggs, and shrimp for three million years, take the fall?

How Cholesterol Became the Villain.

THE IDEA THAT DIET contributes to heart disease is not new. In 1908, a young Russian medical researcher, M. A. Ignatovsky, fed rabbits a diet of animal protein, and when the bunnies developed arteriosclerosis, he blamed the protein. In 1913, a group of rival doctors followed by feeding cholesterol to rabbits, with similar results, including fat and cholesterol deposits in arteries, and they guessed that cholesterol, not protein, was responsible for the arteriosclerosis in Ignatovsky's rabbits as well as theirs.

Animal experiments can, of course, be very useful, but in this case researchers may have reached the wrong conclusions. Unlike humans, who are born to eat cholesterol, rabbits are herbivores with no ability to metabolize it. When you force-feed rabbits with cholesterol, their blood cholesterol rises ten or twenty times higher than the highest values ever seen in humans; the effect is like poisoning. "Cholesterol is deposited in the arteries of the rabbit, but these deposits do not even remotely resemble those found in human atherosclerosis," says Dr. Uffe Ravnskov, the author of The Cholesterol Myths.

Later, in human trials, researchers deliberately used oxidized cholesterol to demonstrate that dietary cholesterol causes atherosclerosis.

Oxidized cholesterol, like oxidized or rancid polyunsaturated vegetable oil, is damaged and unhealthy. As I've mentioned elsewhere, it's no secret in cholesterol circles that oxidized cholesterol, found in powdered eggs, powdered milk, and fried foods, causes arterial plaques.9 Dr. Kilmer McCully, author of The Heart Revolution, says that "pure cholesterol, containing no oxy-cholesterols, does not damage arteries in animals."

Nevertheless, by the 1950s the cholesterol theory was well established: it was thought that eating cholesterol raises blood cholesterol and causes arteriosclerosis. Then, in a significant development, researchers concluded that cholesterol was not acting alone. The revised cholesterol theory had two parts: first, saturated fats (as opposed to unsaturated fats) raise cholesterol, and second, elevated cholesterol causes arteriosclerosis.

Ancel Keys, the professor we met in an earlier chapter on fats, led the campaign against saturated fats. A prolific writer and speaker, Keys spent his last years in Naples, presumably enjoying the Mediterranean diet he had become famous for promoting. When he died in 2004, at the age of one hundred, his influence on diet and disease was rightly considered vast. Theodore Van Itallie wrote this tribute in Nutrition and Metabolism: "For those of us who worked . . . to call attention to the relationship of serum total cholesterol to risk of coronary heart disease (CHD), and to the cholesterol-raising effects of certain saturated fats, Keys will always be one of the major prophets who provided the early evidence that atherosclerosis is not an inevitable concomitant of aging, and that a diet high in saturated fat . . . can be a major risk factor for CHD."10 In the 1950s, Keys made a series of contradictory statements about fats.11 He said that all fats raise cholesterol; yet elsewhere, he wrote that saturated fats raise cholesterol and polyunsaturated oils lower it. Keys said that animal fats caused heart disease; elsewhere, he wrote there was no difference between animal fats and vegetable oils in their effects. Clearly, his data were inconsistent. Nevertheless, Keys focused on one hypothesis: that a diet high in fat and saturated fat caused heart disease.

In 1953, Keys published a famous paper known as the Six Countries Study, placing fat and cholesterol at the center of the debate about diet and heart disease. Keys presented a diagram of fat consumption and death from heart disease in six countries; it appeared to show that the more fat people ate, the more deaths from heart disease. j.a.pan was at the low end of the graph, which swept smoothly upward, and the United States was at the top. But the diagram didn't tell the whole story. There were, in fact, data on fat and heart disease from twenty-two countries, but Keys omitted the other sixteen.12 Instead of forming a convincing upward curve, the twenty-two data points were scattered all over. He declined to cite Finland and Mexico, for example, where fat consumption was similar. Finland had seven times the rate of heart disease as Mexico.

In 1970, Keys published another famous paper- the Seven Countries Study- which appeared to demonstrate a link between cholesterol and heart disease in fifteen populations in seven countries. "The correlation is obvious," said Keys. But when Ravnskov plotted Keys's raw data into a graph, the correlation fell apart. The link is even weaker when you compare groups within countries. On the Greek island of Corfu, for example, people died five times more often from a heart attack than their fellow Greeks living on nearby Crete, although cholesterol on Corfu was lower.

Keys was undeterred and went on to advocate what he called the Mediterranean diet. "The heart of what we now consider the Mediterranean diet is mainly vegetarian," he wrote. "Pasta in many forms, leaves sprinkled with olive oil, all kinds of vegetables in season, and often cheese, all finished off with fruit and frequently washed down with wine." But the traditional Mediterranean diet is not chiefly vegetarian; beef, lamb, goat, pork, game, poultry, liver, and fish are common fare.

Pork- to name only one meat- is eaten throughout the region, where pigs (never fussy eaters) thrive on scrubby land. Italy and Spain are famous for cured pork (prosciutto and serrano ham) and for sausages, which require extra lard. The Spanish make sweet lard cakes called mantecados, while Italian bakers use strutto (rendered lard) much the way American and British cooks once did. In Tuscany, lardo di Colonnata - lard aged in marble with herbs- is eaten straight. In France, warm pork fat dressing and some kind of bacon are de rigueur in the cla.s.sic salads p.i.s.senlit au lard and frisee aux lardons, made with the bitter greens dandelion and endive, respectively.

Olive oil is also traditional in the Mediterranean, of course. It's eaten liberally on Crete, for example. ("My G.o.d, how much oil you use!" Keys is said to have exclaimed when he saw a green salad drowning in olive oil on the island.) But traditional Mediterranean cuisine includes many other fats, too. In northern Italy, b.u.t.ter is typical, and lard is eaten in central regions. In sprawling Provence, which spans the Mediterranean coast and the Alps, olive oil and lard are common, and Gascons are famous for duck and goose fat.

Although Keys was a central figure in the cholesterol hypothesis, he was not always invited to mingle with the nutrition establishment. Some people think this relative ostracism was due to his loner character and indifference to politics. I wonder whether Keys was excluded because proponents of the cholesterol hypothesis were threatened by his lifelong habit of independent thinking.

The medical professor Stephen Phinney remembers a hallway encounter with Keys in the mid-1980s, shortly after the Lipid Research Clinic Coronary Prevention Trial demonstrated that the drug cholestyramine reduced cholesterol and coronary mortality. Keys showed Phinney a paper in which he examined HDL and mortality in Minnesota businessmen. In the paper, Keys wrote that HDL levels predicted heart-related deaths, but not death from all causes. Phinney recalled: "Dr. Keys was fuming, because this ma.n.u.script had been rejected by the major medical journals. Having set the cholesterol-lowering juggernaut in motion, the nutrition establishment was not about to let him sully the picture by demonstrating that it was not the only factor that determined important outcomes such as longevity. In his early 80s, Dr. Keys was still way out ahead of the consensus."13 Later, Phinney told me why Keys, a nutritional epidemiologist, was unique in his field. "He understood the complexity of nutritional metabolism, whereas the pharmacologists either sought to reduce its complexity or ignored it. Pharmacology is a reductionist discipline- you always want to purify your drug and precisely define its target and its mechanism," said Phinney. "This helps explain why diet and nutrition struggle for acceptance in the medical mainstream. The Mediterranean diet works better than atorvastatin"- the statin sold as Lipitor, which lowers LDL-" because it breaks the reductionist rule by harnessing the power of a combination of nutrients working against both cholesterol and inflammation."14 Ancel Keys left a substantial legacy- and a complicated one. He was right to praise antioxidant-rich vegetables, monounsaturated olive oil, and fish. In 1994, the famous Lyon study found the greatest protection against heart disease in the Mediterranean diet was provided by omega-3 fats found in fish.

Yet Keys also set the stage for a battle against the alleged dangers of saturated fats in traditional foods such as b.u.t.ter- dangers that were oversold, if not by Keys himself then certainly by the medical-pharmaceutical complex, which took up the anticholesterol campaign with enthusiasm befitting a crusade.

The Cholesterol Skeptics.

THE OLD ADVICE- b.u.t.ter high cholesterol heart attack- was too crude to be accurate. New advice is more subtle; even the conventional wisdom holds that margarine is "worse than" b.u.t.ter, and we know that high HDL is good. Yet I have some sympathy for the researchers who reduced the message to an antib.u.t.ter slogan. They meant well. If you labor in a complex field- and heart disease is certainly that- the appet.i.te for simple answers can be maddening.

People will ask, If b.u.t.ter doesn't cause heart disease, what does? Well, I venture, genes, lack of exercise, inflammation, free radicals, smoking, and industrial foods like trans fats, sugar, and corn oil. I believe this is accurate and reasonably complete; I hope it's also brief enough to keep people from nodding off. I don't envy doctors, who are forced by the clock and by anxious patients to reduce complex disease etiology, diagnosis, and treatment to three-minute summaries.

On diet and disease, this is as simple as I can make it without doing injustice to accuracy or uncertainty: traditional foods are good for you. There are various ways to go about proving this. You can feed people corn or coconut oil, and see that corn oil lowers HDL and coconut oil doesn't. You can observe whether people who eat extreme diets (e.g., all meat) get heart disease. And so on. Such studies convinced me that you can eat whatever you want- except industrial foods. If that satisfies you, close this book and enjoy b.u.t.ter and eggs. Those who would like to know more about heart disease may wish to read on.

The quintessential disease of civilization, heart disease was rare before 1900, and rare it remains in preindustrial groups. In the United States, the first heart attack was reported in 1912, and by midcentury heart disease was the nation's biggest killer. Today cardiovascular diseases- conditions of the heart and blood vessels, including angina, stroke, congestive heart failure, and heart disease- are still the leading cause of death. All cardiovascular diseases combined kill about a million Americans a year, men and women in pretty much equal numbers. Heart disease alone is responsible for five hundred thousand deaths every year.15 There is also good news. Since peaking in the late 1960s, the death rate from cardiovascular diseases has fallen. In 1999, the mortality rate was less than 40 percent of the rate in 1950.16 Back then, heart disease was an acute, often fatal condition. After a heart attack, patients were simply sent home to rest and to die. Today doctors are adept at various treatments- clot-busting drugs, tiny balloons to open arteries, bypa.s.s surgery- so that heart disease, though still prevalent, is more often chronic than fatal. These advances keep many of the sixty-four million Americans with heart disease alive longer.

In the first stage of heart disease, angina, blood flow to the heart is restricted. When blood flow stops, that's a myocardial infarction, or heart attack. Together, angina and a heart attack are what doctors call coronary heart disease. Arteriosclerosis, or hardening of the arterial walls, is partly a function of age; with time, the smooth, elastic arterial cells become fibrous and stiff. Arteriosclerosis may be a protective measure to prevent the arteries from expanding under the pressure of blood; veins, which carry blood to the heart at much lower pressure, don't stiffen in this way. When arterial walls become thick and swollen, it's called an atheroma-, many atheromas are known as atherosclerosis. Atheromas, which contain calcium, cholesterol, and fats, may burst, causing blood clots or heart attacks.

The cholesterol hypothesis holds that saturated fats raise cholesterol and cholesterol clogs arteries, but a number of researchers, some of whom belong to a network called the International Committee of Cholesterol Skeptics, are doubtful. "The truth is that the cholesterol theory has never been proven," says Dr. Kilmer McCully, whom I quoted earlier on oxidized cholesterol in powdered eggs. "Elevation of blood cholesterol is a symptom- not a cause- of heart disease." Here is Ravnskov, a leading skeptic and the author of The Cholesterol Myths: "When the cholesterol campaign was introduced in Sweden in 1989,1 was very surprised.

Having followed the scientific literature about cholesterol and cardiovascular disease superficially for years, I could not recall any study showing that high cholesterol was dangerous to the heart, or that any type of dietary fat was more beneficial or harmful than another. I became curious and started to read more systematically. Anyone who reads the literature in this field with an open mind soon discovers that the emperor has no clothes."17 At first I thought the skeptics might be few in number, but I found them all over. In 1978, a National Inst.i.tutes of Health conference held to discuss the drop in death rates from heart attacks since the 1960s was unable to account for the decline by changes in fat and cholesterol consumption or blood cholesterol.18 Yet this statement got little attention. In 1998, a British National Health Service review found that blood cholesterol alone was a "relatively poor predictor of individual risk."19 The authors concluded that for the general population, "cholesterol screening is unlikely to reduce mortality and can be misleading or even harmful."

As you might imagine, the cholesterol skeptics have not received a hearty embrace from the medical and pharmaceutical establishment. In Finland, supporters of the anticholesterol campaign belittled Ravnskov's book on television, and then- literally- set the book on fire. When I read The Cholesterol Myths, I got excited too, but it didn't put me in the mood to burn books. Quite the opposite: it made me want to buy them. I began to read medical journals and textbooks, and soon I was a skeptic, too.

Does saturated fat raise cholesterol? Not in unhealthy ways. Early studies did show that certain saturated fats, when compared with polyunsaturated oils, raise total cholesterol, but now we know that total cholesterol is a poor predictor of heart disease. In fact, saturated fats raise HDL and polyunsaturated oils lower it. The National Cholesterol Education Program is clear about the virtues of HDL: "the higher, the better." The general effect of saturated fats is to restore a healthy balance of HDL and LDL. Coconut oil, for example, raises HDL if it's low, and lowers LDL if it's high. As we've seen, certain saturated fats (stearic acid in beef and chocolate, and palmitic acid in b.u.t.ter and coconut oil) are good for HDL and LDL ratios.20 There is abundant evidence from traditional diets to absolve saturated fats. In Nigeria, for example, the Fulani get half their calories from fats, half of which are saturated. Despite what the theory predicts, they have low LDL.21 Does high blood cholesterol predict heart disease? In a striking number of cases, the link is weak. Since 1948, researchers have studied the residents of Framingham, Ma.s.sachusetts, a city near Boston. After a few years, directors of the now famous Framingham Heart Study reported findings that became the bedrock of the cholesterol hypothesis: when they sorted people by low, normal, and high cholesterol, those with high levels had more fatal heart attacks. But almost half the heart attack patients had normal or low cholesterol. In Russia, a twelve-year study of more than sixty-four hundred men found those with low cholesterol had more heart disease.22 A study in rural China found that neither cholesterol nor LDL was linked to heart disease.23 I could cite many other examples- and the cholesterol skeptics do, at length- but the point, I hope, is clear: in Ma.s.sachusetts, Russia, and China, something other than high cholesterol must be to blame for a large number of heart disease cases.

Cholesterol may be a concern for a relatively small group of people: younger men at high risk of heart disease, such as those who've already had one heart attack. In 1987, the Framingham data showed an a.s.sociation between high cholesterol and mortality for men under forty-seven. But for men older than forty-seven and for all women, there was no a.s.sociation between cholesterol and death rates from all causes, including heart disease.24 According to James Wright of the University of British Columbia, compared with high blood pressure, obesity, diabetes, and smoking, cholesterol is the weakest risk factor for women and heart disease.

The Center for Medical Consumers believes that the heart disease-awareness campaign exaggerates risks for women. One hears that cardiovascular diseases kill almost five hundred thousand women a year in the United States, but nearly 80 percent of heart-related deaths occur in women older than seventy-five. a.s.sociate Director Maryann Napoli described what the Framingham researchers found: "Cholesterol was identified as one, but only one, of 240 risk factors that included male baldness, creased ear lobes, and being married to a highly educated woman. Research focused on cholesterol because it is a modifiable risk factor (translation: drug industry opportunity). Though the Framingham Study found a strong a.s.sociation between . . . cholesterol and heart disease only in young and middle-aged men, the entire population was . . . instructed to fear this particular risk factor."

The vast set of data from the prestigious, long-running Framingham study will provide rich research material for years to come. For now, consider this fact. According to Dr. William Castelli, director of the Framingham study, "the more saturated fat one ate, the more cholesterol one ate . . . the lower the person's serum cholesterol."25 When Castelli made this astonishing admission in 1992, it didn't make news.

Diet First, Then Medication.

SUPPOSE YOU WERE A DOCTOR, and the patient sitting before you is a fit woman in her midsixties. According to official guidelines, she has "high" cholesterol of 261 milligrams and "borderline high" LDL of 153 milligrams, but her HDL and triglycerides are great. Current advice from the National Cholesterol Education Program is to lower total cholesterol and LDL aggressively.26 What should you do?

Some doctors would write a prescription for a statin drug, which blocks the liver from making cholesterol. Statins are the best-selling drug in the United States, worth sixteen billion dollars a year. Pfizer, which makes Lipitor, spends sixty million dollars a year marketing statins to consumers and employs thousands of sales representatives to promote it to doctors.27 About fifteen million Americans take statins, but under 2004 guidelines, twice that many- thirty-five million people- are candidates. Cardiologists call statins "aspirin for the heart" and joke about putting them in the water supply.

In this not-quite-hypothetical case, the patient happens to be my mother, and I was glad her doctor didn't prescribe a statin.

Mom doesn't need to worry about her total cholesterol or her LDL. First, total cholesterol is a poor predictor of heart disease in women and older people. Second, her ratio of total cholesterol to HDL puts her in the "below average" risk category. Third, in women and men over sixty-five, high LDL means longer life.28 Statins are highly effective at reducing LDL, and some studies show they can reduce the risk of dying of a heart attack. But some researchers have doubts. Benefits of statins for total mortality- or death from all causes, the gold standard in epidemiology- are small or nonexistent. In 2004, Britain was the first country to approve over-the-counter sales of a statin. The Lancet objected, noting that five major trials found that death from all causes was similar with and without statins.29 "Statins have not been shown to provide an overall mortality benefit," wrote the Lancet editors.

Statins may work for a fairly small group. "The people who benefit are middle-aged men who are at high risk or have heart disease," Dr. Beatrice Golomb said. "The benefits do not extend to the elderly or to women." Golomb, who describes herself as "pro-statin," is a medical professor at the University of California and the lead investigator in a large federally financed statin study. (Diabetics may also benefit from statins, but it seems sensible to treat the most common form, type 2 diabetes, with diet first, because diet causes it.) As with any treatment, benefits must be weighed against costs. Side effects of statins include muscle weakness, nerve damage, kidney failure, liver damage, and memory loss. A rare but serious side effect is the potentially fatal muscle-wasting disorder rhab-domyolysis. In 2001 a statin linked to thirty-one rhabdomyolysis deaths was withdrawn. Statins deplete coenzyme Q10 an antioxidant found in fish, pork, heart, and liver. Used to prevent and treat heart disease in the United States and j.a.pan, CoQ10 prevents LDL from being oxidized.30 "The first thing I do with new heart patients," says cardiologist Dr. Peter Langsjoen, who has reviewed many studies on CoQ10, "is take them off statins."31 What const.i.tutes healthy cholesterol levels is also a matter of debate. The National Cholesterol Education Program says total cholesterol under 200 milligrams is "desirable." But this target is not particularly useful. First, "total cholesterol" is not really cholesterol at all, but a composite number equal to HDL, LDL, and 20 percent of triglycerides. We now know that total cholesterol does not predict heart disease. Despite the bad press, "high LDL" is a poor predictor, too. Other readings such as triglycerides, blood sugar, and C-Reactive Protein (CRP) are more useful.

Some doctors think it's unwise to reduce cholesterol at all costs.

Attention has focused on "the supposed danger" of high blood cholesterol for fifty years, says Barry Groves, a British researcher on obesity, diabetes, and heart disease, while "the dangers of low blood cholesterol levels have largely been ignored."32 Older people, for example, benefit from high cholesterol. Death rates in the elderly from all causes, including heart disease, are greater with low cholesterol.33 Thus the Lancet advises doctors to be "cautious" about reducing cholesterol in people over sixty-five.34 Low cholesterol is linked to respiratory disease, HIV, depression, and death by violence or suicide. Low cholesterol is also a.s.sociated with another serious cardiovascular disease: stroke.35 Cholesterol protects against infection, a well-known risk factor for heart disease. Infection leads to inflammation, which appears in the arterial walls of heart disease patients with normal cholesterol. A good measure of inflammation is CRP, a risk factor for heart disease.36 Women with high CRP and healthy cholesterol have twice as many heart attacks.37 Inflammation is caused by excess omega-6 fats, smoking, and gum disease, another risk factor for heart disease. Exercise reduces both inflammation and CRP, which is produced in fat cells.

Clearly, there is much more to learn from the lab than just our LDL and HDL levels. Heart disease has many causes. That means there are no simple answers in diagnosis, prevention, and treatment. On the positive side, there are potentially many cures. For example, if my mother wanted to lower her LDL without taking drugs, she could eat more fish. Omega-3 fats reduce LDL, raise HDL, lower triglycerides, prevent clots, reduce blood pressure, and fight inflammation. Fish is powerful stuff- and it has no side effects.

Another nutritional approach to reducing LDL (if it worries you) is eating soy, almonds, oats, barley, okra, and eggplant. Dubbed the portfolio diet, this regimen compares favorably with statins in lowering LDL.38 University of Toronto researchers gave people with high cholesterol three treatments: one group ate a diet "very low" in saturated fat, the second ate the portfolio diet, and the third took statins. The statin treatment and the portfolio diet were equally good, each reducing LDL by about 30 percent. The diet low in saturated fat was the least effective, reducing LDL by only 8 percent. (The low-saturated fat diet, I noted, was also heavy on industrial foods: sunflower oil, fat-free cheese, egg subst.i.tutes, liquid egg whites, and "light" margarine.) How does the portfolio diet work? Almonds are rich in monounsaturated fat, which lowers LDL. Soy isoflavones lower LDL. The viscous fiber in whole grains, okra, and eggplant also lowers LDL, perhaps by mopping up bile acid, which forces the liver to use up cholesterol to make more bile acid. The t.i.tle of the editorial to accompany this small but promising study in the Journal of the American Medical a.s.sociation was clear enough: "Diet First, Then Medication."

A Disease of Deficiency.