Like an endless courtroom drama, the role of diet in coronary heart disease has been hotly contested for nearly two generations. The indictment of the American diet as a possible malefactor was first drawn up in the years following World War II. Since then, a procession of scientists has taken the stand, presenting masses of data on the menus and bloodstreams of populations around the world.

Until 1984, the outcome of this scientific debate had been a succession of hung juries. Then the results of a long-awaited study were published. Called the Lipid Research Clinics Coronary Primary Prevention Trial (CPPT), the study showed that reducing abnormally high levels of cholesterol in the blood could reduce the incidence of heart attacks.

To the American Heart Association (AHA), the findings of the CPPT provided "smoking gun" evidence for dietary changes. The AHA has long advocated what is commonly known as the "prudent diet." It involves specific dietary measures to reduce the risk of heart disease, such as lowering the percentage of total dietary fat and choosing polyunsaturated fats over saturated fats. Saturated fats, which come primarily from foods of animal origin, tend to raise blood cholesterol levels. Polyunsaturated fats, which come mainly from vegetable sources, tend to lower blood cholesterol.

The link between coronary disease and blood cholesterol levels has been established by other studies of various population groups around the world. Groups with low blood cholesterol generally had lower rates of heart attack than those with high levels of blood cholesterol. At the same time, dietary experiments conducted with subjects in controlled settings suggest that there is a link between diet and blood cholesterol levels. In such studies, blood cholesterol levels can be influenced by the type of fat eaten and, to a lesser extent, by the amount of cholesterol consumed.

Logically, then, it appears that reducing the level of blood cholesterol—through diet, drugs, or other means—might reduce the risk of coronary heart disease. To physicians, this concept is known as the "lipid hypothesis." To people familiar with the AHA's dietary advice, it's the diet-heart theory.

Whatever the name, the ultimate goal is to prevent, retard, or even reverse the development of atherosclerosis—the buildup of fibrous, fatty deposits, called plaques, on the inner walls of major arteries. Atherosclerosis can reduce or cut off the blood flow in arteries serving major organs such as the heart or brain. When it affects the coronary arteries nourishing the heart, it can lead to heart attack and impaired heart function.

Elusive Proof

A theory requires proof before it can be accepted as fact. And, for many years, that proof remained elusive. More than a score of clinical trials failed to show convincingly that reducing blood cholesterol by diet or drugs reduced the incidence of a first heart attack. Yet a number of studies had strongly suggested that a change in diet could lower the incidence of second heart attacks.

Cholesterol, technically a sterol similar in structure to vitamin D and certain bile constituents, is found in all body cells and serves a number of important functions. It is a building block for cell membranes and steroid hormones, and it shares in other essential bodily processes. It's also an important component of arterial plaque, although its exact role in plaque formation is unclear.

Some of the clinical trials suggested that reducing blood cholesterol was beneficial. But, partly because of design weaknesses in many of the studies, the overall results were mixed and inconclusive.

Accordingly, even though several federal agencies and expert panels have advised cutting back on fat and cholesterol, many physicians tended to be skeptical. In a 1983 survey of 1600 physicians by the National Heart, Lung, and Blood Institute (NHLBI), which sponsored the CPPT study, 90 percent said that quitting smoking would have a "large impact" on the prevention or control of coronary disease and 80 percent said that reducing high blood pressure would—but only 39 percent said the same about reducing blood cholesterol levels. By the time a follow-up study was conducted in 1986, however, 64 percent of the physicians agreed with that statement about cholesterol.

The CPPT Study

Many public-health officials credit the CPPT results for changing minds about the importance of dietary measures to control cholesterol. Ironically, the study involved the test of a drug rather than diet. Nevertheless, it offers the strongest evidence to date for the value of blood-cholesterol reduction by any means. Even the CPPT's critics agree that the study was well designed and well executed. But there is disagreement over how the results should be interpreted.

The CPPT involved 3800 middle-aged men. All had blood cholesterol levels of at least 265 milligrams (per deciliter of blood)—enough to put them in the highest 5 percent of cholesterol levels among adult Americans. Although all were free of any sign of coronary disease on entry into the trial, they were considered to be at risk of heart attack because of their elevated cholesterol levels. (In the Framingham Heart Study, a long-running study of residents of Framingham, Massachusetts, men with an average blood cholesterol of 260 milligrams have experienced a heart attack rate three times as high as men with levels below 195. The average blood cholesterol among all Americans is 210 milligrams.)

The CPPT participants were randomly assigned to a treatment or a control group of 1900 men each. The treatment group took daily doses of a cholesterol-reducing drug, cholestyramine, while the controls received an indistinguishable placebo. Both groups followed a cholesterol-reducing diet for an average of 7.4 years. Over the course of the trial, the treatment group had fewer heart attacks than the control group (155 versus 187) and fewer deaths from heart attack (30 versus 38). Because of the large number of participants, the difference between the two groups was found to be statistically significant.

Taken in conjunction with other evidence, said the National Heart, Lung, and Blood Institute, the findings "support the view that cholesterol lowering by diet also would be beneficial." According to the NHLBI, the results could be "narrowly interpreted" to apply only to the use of a specific drug in middle-aged men with cholesterol levels above 265 milligrams. "The trial's implications, however, could and should be extended to other age groups and to women."

The NIH Panel's Verdict: Reduce Blood Cholesterol

Those recommendations drew little criticism—or even much attention—until late 1984, when the NHLBI joined in convening a panel of medical and public-health experts to address key questions about lowering blood cholesterol. Such meetings, called "consensus development conferences," are held periodically at the National Institutes of Health (NIH) to thrash out reasonable policies on controversial health issues. In this instance, the panel reached a unanimous verdict, coming out wholeheartedly for reducing blood cholesterol levels, especially by diet.

It has been established "beyond a reasonable doubt," said the NIH panel, that lowering elevated blood cholesterol levels will reduce the risk of heart attacks from coronary disease. "This has been demonstrated most conclusively in men with elevated blood cholesterol levels, but much evidence justifies the conclusion that similar protection will be afforded in women with elevated levels." The panel recommended that at individuals with blood cholesterol levels judged to be in high-risk or moderate-risk categories receive dietary treatment, and if necessary, drug treatment to reduce those levels. It defined the two categories in terms of blood cholesterol levels at specific ages.

The panel also concluded that blood cholesterol levels of most Americans are "undesirably high," largely because of high dietary intake of "calories, saturated fat, and cholesterol." It advised that all Americans except children under two follow a diet identical to the latest version of the American Heart Association's prudent diet. Children under two were excluded because of possible deleterious effects on normal growth and development.

The Recommended Diet

Specifically, the diet would pare fat intake to 30 percent of total calories consumed (from an average of about 40 percent now) and cut saturated fats to less than 10 percent of calories. Polyunsaturated fats would be limited to 10 percent of calories, and daily cholesterol intake would be held to 250 to 300 milligrams (roughly the amount in one egg). Total calories would also be reduced, if necessary, to correct obesity and maintain ideal body weight.

In practical terms, the diet means eating more fruit, vegetables, and grain products, and much less food from animal sources—especially fatty meats, dairy products, eggs, and rich baked goods. It also means favoring fish and poultry over beef, lamb, and pork, and limiting portions to roughly 4 to 6 ounces.

The panel based its conclusions on various types of evidence, including experimental animal data and population studies as well as clinical trials. One study sponsored by the NHLBI, for example, had shown that treatment with cholestyramine helped to slow the progression of arterial plaques in middle-aged male cardiac patients with elevated cholesterol levels. However, the "keystone in the arch," as the panel chairman put it, was the CPPT.

Mixed Support for Universal Change in Diet

Despite the unanimity on the NIH panel, other medical experts view this "keystone" as a shaky support—especially for the panel's sweeping dietary advice to the entire public. For one thing, there's disagreement over whether beneficial effects achieved with a drug can be assumed for diet as well. Some contend that a cholesterol-lowering diet would be similar in action to the drug cholestyramine; others voice the opposite view.

The biggest bone of contention between the panel and its critics has been whether findings in high-risk, middle-aged men should be applied to the public at large. Virtually all clinical trials of cholesterol-lowering therapy have focused on middle-aged men. There's only limited clinical evidence available about the effects of reducing blood cholesterol in women or older people. And there's none at all for children or for people without elevated cholesterol levels. No one knows for sure whether reducing cholesterol levels in these groups is beneficial, harmful, or irrelevant.

Another study sponsored in part by the NHLBI was reported in 1987. Called the Cholesterol-Lowering Atherosclerosis Study (CLAS), it tested the effect of a similar cholesterol-lowering drug therapy on 162 middle-aged men who had previously undergone coronary bypass surgery. The control group received a placebo and followed the AHA diet; the treatment group received colestipol and niacin and followed an even more rigorous diet. Rather than counting heart attacks and deaths, the CLAS used angiograms (X rays of blood vessels) to evaluate the effect of treatment at the level of the arterial wall.

The results showed significant benefits from treatment. "Deterioration in overall coronary status was significantly less in drug-treated subjects than placebo-treated subjects." Actual improvement in coronary status was noted in 16 percent of the treatment group, as opposed to only 2 percent of the controls. The results indicated that actual regression of atherosclerosis is possible and demonstrated, for the first time, "a logical, mechanical explanation for benefits from blood cholesterol-lowering therapy."

"At this time," the researchers concluded, "we advocate measures to lower blood cholesterol levels in all postcoronary bypass patients." Despite the select population studied, they stated that their results "also support and extend" the cholesterol-lowering treatment goals developed in 1984 by the NIH consensus development conference.

Still, the questions remain: How much can diet do alone? And how would individuals other than the classic middle-aged male respond?

The Evidence from Epidemiology

The rationale for advising dietary changes in untested groups is based partly on epidemiology, the medical specialty that investigates the incidence and suspected causes of disease in various populations. Since World War II, studies of populations around the world have shown strong relationships between dietary fat, blood cholesterol, and cardiovascular disease.

Statistical associations can identify potentially related factors, but they can't prove a cause-and-effect relationship. Heart-disease rates have been associated positively or negatively with many factors, ranging from smoking and wine consumption to national income and the number of cars per 100 people. Scientists take some of these associations seriously and dismiss others. Even a highly plausible association requires confirming evidence from other research. With most groups other than middle-aged men, confirming data from clinical trials are either sparse or nonexistent.

Next to population studies, the most significant evidence implicating diet in heart disease comes from animal experiments. Diets high in cholesterol and fat have produced a form of atherosclerosis in various species, including such primates as rhesus monkeys. There's no consistency of effect from species to species, however, or even within species. Animals vary widely in their response to such diets. Rhesus monkeys are mainly vegetarians. Meat-eaters such as dogs and cats show the least sensitivity to fat and cholesterol, while rabbits seemingly develop plaques if they trip over-an Easter egg.

The Role of Other Risk Factors

Meanwhile, a number of other considerations complicate the issue of dietary influence in coronary disease.

Whereas the risk of lung cancer is dramatically increased by cigarette smoking, and high blood pressure strongly affects the risk of stroke, the risk of heart attack is influenced by many different factors. Cigarette smoking, high blood pressure, and elevated blood cholesterol are three major ones. Other influences include age, male gender, diabetes, a family history of heart disease, a sedentary life-style, decreased high-density lipoprotein cholesterol, and possibly behavioral and social factors.

Consequently, while diet tends to get most of the press coverage, it's not nearly as crucial as advertised. True, it can affect the blood cholesterol level—but that's only one of several important risk factors in coronary disease.

A substantial part of the data showing the effects of dietary cholesterol or fat on blood cholesterol levels comes from "metabolic ward" studies—controlled experiments conducted in hospitals or similar settings, using formula diets with precise quantities of cholesterol or fat. In contrast, the effects of dietary cholesterol and fat are not as clear-cut among people eating ordinary foods on free-choice diets. A mixed diet contains a variety of substances that have opposing effects on blood cholesterol. So the effect of a food may not depend solely on the amount of fat or cholesterol in it.

Furthermore, while fat-controlled diets have reduced blood cholesterol in clinical trials, such reduction requires serious dedication and sustained effort—not merely switching from butter to margarine or cutting out a single food such as eggs.

Good Cholesterol and Bad Cholesterol

Another confounding factor is the distribution of cholesterol in the blood. This involves HDL and LDL, sometimes popularly referred to as "good cholesterol" and "bad cholesterol."

Cholesterol circulates in the blood linked to large molecules called apoproteins, or carrier proteins. Low-density lipoprotein (LDL), a form of cholesterol-carrying protein in the apoprotein B family, seems to promote atherosclerosis by depositing cholesterol in the arterial wall. About two-thirds or more of total blood cholesterol is transported in LDL. High-density lipoprotein (HDL), part of the apoprotein A family, appears to protect against the disease process by removing cholesterol from the arterial wall. Increasing evidence now indicates that the relative distribution of cholesterol among those two types of lipoprotein is a better gauge of coronary risk than the total blood cholesterol level.

Some experts focus on the ratio of total cholesterol to HDL cholesterol. For example, a total cholesterol level of 240 milligrams might appear to represent a greater risk than a total level of 200 milligrams. Yet a person with a total level of 240 milligrams that is one-fourth HDL cholesterol (60 milligrams) would actually have a lower risk than a person with a 200-milligram level that is only one-fifth HDL (40 milligrams). So the person at 240 milligrams would have a ratio of 4 (240/60); the other person, a ratio of 5 (200/40). And the higher the ratio, the greater the risk.

The level of HDL cholesterol alone is also a strong indicator of coronary-disease risk. On average, the higher the level, the lower the risk.

Studies show that women, lean people, nonsmokers, people who consume two or more alcoholic beverages daily, and people who exercise regularly have relatively higher HDL levels than, respectively, men, obese people, smokers, nondrinkers, and sedentary people. A diet high in fat tends to increase both HDL and LDL, although not necessarily in the same proportions. Conversely, a low-fat diet tends to reduce both, again not necessarily in tandem.

A specific fraction of HDL cholesterol, called HDL2, appears to be the part that is protective against heart attack. But the factors that raise or lower HDL2 levels are not yet clear. Research suggests that an extended program of regular aerobic exercise raises HDL2 and that moderate drinking does not. Quitting smoking also increased HDL2 in one study.

More recently, the National Cholesterol Education Program, a group of experts convened to study policy, has decided to focus on LDL cholesterol. Individuals with LDL cholesterol less than 130 milligrams are considered to be at low risk. Those with levels between 130 and 160 are "borderline high," 160 to 190 "moderately high," and above 190 milligrams at "very high" risk. Currently, however, it is technically difficult to measure LDL cholesterol; determinations are arrived at by calculation. There is probably no need to determine LDL cholesterol when total cholesterol is less than 200 milligrams.

Further investigation is needed about these factors and others—especially the effects of diet.

What Is the Best Menu?

Both critics and advocates of dietary measures seem to agree on one point: Various types of diet might lower blood cholesterol, but no one knows for sure which diet is best.

The AHA's prudent diet is low in fat and cholesterol and emphasizes foods with complex carbohydrates, such as vegetables and fruit. The diets of Greenland Eskimos and the Japanese are high in fish and fish oils, which lower cholesterol and have other effects on blood. Diets high in polyunsaturates, as in some vegetarian diets, are another alternative, as are those associated with Mediterranean countries, which tend to be high in monounsaturated fats such as olive oil.

Each of these diets has been linked with low rates of coronary disease in at least some populations. But there is little information about how they compare with one another in efficacy and safety—or how consistently Americans would adhere to them.

Meanwhile, scientific investigation is continuing to refine prevailing ideas on the presumed advantages or disadvantages of specific foods. For example, in fat-controlled diets, cod, flounder, and other low-fat fish have been recommended over oily, fatty fish such as salmon and mackerel. But there is increasing evidence that the fatty fish may have a beneficial effect on coronary disease.

Similarly, monounsaturated fats were long thought to be neutral in their effect on blood cholesterol. But recent research suggests that they can have a cholesterol-lowering effect similar to that of polyunsaturateswithout lowering HDL levels as polyunsaturates do. This finding may give a boost to such products as olive oil at the expense of safflower oil and corn oil.

In addition to advising reductions in total and saturated fat, current AHA guidelines recommend limiting dietary cholesterol to 250 to 300 milligrams a day. In effect, that can mean only two or three eggs per week, since an egg contains roughly 250 milligrams of cholesterol. A public-health approach that discourages cholesterol consumption might well benefit people who are sensitive to cholesterol in their diet. But for other people, dietary cholesterol in itself has minimal, if any, adverse effects on blood cholesterol, even with moderate egg consumption.

The body, in fact, produces cholesterol on its own—about 800 to 1500 milligrams daily. Generally, when a person eats more cholesterol, the body responds by producing less. When a person eats less cholesterol, the body produces more. That natural response helps to regulate blood cholesterol levels. It doesn't work efficiently, though, in a significant number of people—and these are the ones who might benefit from limiting dietary cholesterol.

Eggs are relatively low in calories and high in important nutrients, making them especially suitable for diets to reduce obesity. Weight loss among people who are overweight is one of the most effective ways to reduce blood cholesterol. Paring excess weight also helps to lower high blood pressure and control diabetes—two other risk factors for coronary disease.

A Prudent Approach to the Prudent Diet

Despite various unanswered questions about the prudent diet, it deserves consideration by men with moderate or high levels of blood cholesterol as a possible risk-reducing measure. It avoids extremes and aims for a balanced variety of nutrients. It is unlikely to harm this population, and it may do some good.

A similar diet 'was used for some 1200 high-risk men in a five-year clinical trial in Oslo, Norway. That study reported a significant reduction in heart-attack rates, but the benefit from diet is difficult to assess because the trial included smoking reduction as well. There were no apparent ill effects from the diet.

For men with "low-risk" blood cholesterol levels, there's less to gain from the prudent diet, and there's no proof as yet that it will reduce their already low coronary risk. There's also little likelihood of harm for those who want to try the diet as a way of hedging their bets. However, a cholesterol-lowering diet is inadvisable for men with total blood cholesterol below 180 milligrams. In some epidemiological studies, cholesterol levels below that level are associated with a higher risk of mortality from causes other than coronary disease, particularly colon cancer.

For women, the possible value of the prudent diet is still unclear. There is little information about the effects of cholesterol-lowering therapy in women, and no clinical studies at all about the efficacy or safety of the diet for them. Women of any age have a lower risk of heart attack than their male counterparts. The difference is especially pronounced before menopause. Among white American adults under 45, for example, men have about 10 times the heart-attack rate of women.

Many women not only have less to gain from a fat-controlled diet, but in fact may incur more risk from it than men do. Meat and dairy foods, the targets for major cholesterol cutbacks, are important sources of iron and calcium, two nutrients that are already inadequate or marginal in many women's diets.

Despite such considerations, however, women with elevated blood cholesterol levels are at increased risk of coronary disease. Although there's no proof that the AHA diet will be beneficial, it's a reasonable precaution for women with levels in the high-risk or moderate-risk range, particularly after menopause.

Cautions for Children and the Elderly

Perhaps the most controversial AHA recommendation is to extend the prudent diet to children two years of age and older. The NIH panel endorses that policy, but the American Academy of Pediatrics does not. Both groups agree on modifying diets for children with abnormally high cholesterol levels or inherited metabolic disorders. The disagreement centers on what healthy children should eat.

The main reason for starting a low-fat diet at age two is to foster that eating habit. "It is desirable to begin prevention in childhood because patterns of lifestyle are developed in childhood," the NIH panel said. Moreover, a specific diet may be easier to adhere to if all family members eat the same foods.

The American Academy of Pediatrics disagrees with that approach. Alvin M. Mauer, M.D., chairman of the Academy's committee on nutrition, spoke to CU in 1985. "There is no evidence that diet in childhood influences the development of atherosclerosis in adult men," he said. "Is it necessary to start the diet in childhood rather than in the early twenties?" The American Academy of Pediatrics has recently altered its view and recommends that children with elevated blood cholesterol levels be counseled on diet. What worried Mauer is that dietary restrictions may compromise growth and development, not only in young children, but also during the major growth spurt of adolescence. "Meat and eggs are good sources of iron and other nutrients needed for expanding blood volume and muscle growth," he said. "Dairy foods are a major source of calcium needed for bone development." Mauer also pointed out that the concept of family meals applies more to young children than to adolescents, most of whom eat only one meal a day at home.

Pediatricians who support adopting the AHA diet in children are confident that it's adequate for growth and development. But until enough experience is logged with high-risk children receiving dietary treatment, this assumption may be premature.

Caution is also warranted at the opposite end of the life span. The elderly often have special nutrition problems, including that of simply obtaining a balanced diet. Eggs, for example, are high in cholesterol, but they are also nutritious and cheap. Moreover, dental problems, chronic disorders, and economic or social constraints may partly dictate food choices. What older people may need least is another set of restrictions on what they should or shouldn't eat—particularly since the prudent diet hasn't been tested in this age group.

Meanwhile, people who want to try the AHA diet as a possible risk-reducing measure should bear in mind that it's not a self-sufficient program. Elimination of cigarette smoking, control of high blood pressure and diabetes, avoidance of a sedentary life-style, and reduction of obesity should all be part of a comprehensive program to reduce the risk of coronary disease.

Measurement of blood cholesterol, including its HDL and LDL fractions, should be one part of any overall evaluation of coronary risk. Since laboratory measurement may vary, at least two separate determinations should be made before accepting a value as accurate.

Luis Treacy

Copyrighted material; do not reprint without permission.

 

View all articles by Luis Treacy