Download Diet's Impact on Blood Cholesterol: Public Health Recommendations and more Study notes Biochemistry in PDF only on Docsity!
The Rationale for Controlling Dietary Lipids in
the Prevention of Coronary Heart Disease
DORALIE L. SEGAL~
Despite the acknowledged association between blood cholesterol levels and coro- nary heart disease, many people throughout the Americas remain uncertain about the importance of that association and available countermeasures. This article reviews the subject’s historical background, results of recent studies, and current recommendations of leading expert groups in the United States. It also examines ways in which the body’s cholesterol level is influenced by changes in the dietary intake of calories, saturated and unsaturated fats, and cholesterol, and recommends the adoption of specific dietary measures as a way of markedly improving public health.
A
n association between blood choles- terol levels and the severity of ather- osclerosis was first proposed over 100 years ago. Since then, the results from various types of epidemiologic (longitu- dinal, cross-sectional, migration, and au- topsy) studies and clinical trials have supported a cause-and-effect relationship between blood cholesterol, atheroscler- osis, and coronary heart disease. In order to understand how dietary fac- tors may increase blood cholesterol lev- els, it is important to know some basic facts about fats, fatty acids, and cholesterol-which are collectively re- ferred to as “lipids.”
lThis article will also be published in Spanish in the BoIetin de ZuOjkina Sanitatia Panamericana,Vol. 109,
- Two lectures, one presented at a course on preventive cardiology at the XLIII Congress of the Brazilian Society of Cardiology (Bra&a, 27 Sep- tember 1987) and the other at a course in epidemi- ologv of chronic diseases with emuhasis on cardio- vascular disease that was sponsored by the Venezuelan Health h4inistrv and PAHO at Bar- quisimeto, Venezuela, and 21 November 1987, served as the basis for the material presented here. *Physiologist on special assignment to the Pan American Health Organization from the United States Food and Drug Administration.
Fat is present in most foods, regardless of whether the foods are of plant or ani- mal origin. The amount of fat present varies, ranging from trace amounts in leafy vegetables to predominant amounts in food ingredients such as cooking oils. Dietary fats provide the most concen- trated source of energy in the diet, each gram of fat contributing slightly more than nine kilocalories (kcal). Animal tis- sues, dairy products, and oils extracted from certain seeds and plants provide most of the fats in the diet-oils being defined as fats that are usually liquid at room temperature. The principal breakdown products of fats are fatty acids and glycerol. A fatty acid may be saturated or unsaturated. A saturated fatty acid contains no double bonds, and all available carbon positions are “saturated” with hydrogen atoms. The degree of unsaturation is determined by the number of carbon atoms within the fatty acid that are double-bonded to each other. A fatty acid with a single dou- ble bond is called “monounsaturated,” while a fatty acid with two or more dou- ble bonds is usually referred to as “poly- unsaturated.” Fats from plant as well as
Bulletin of PAHO 24(2J, 1990 197
animal sources usually contain both satu- rated and unsaturated fatty acids. How- ever, fats from animal sources usually contain more saturated fatty acids than those from plant sources. Cholesterol differs from fats and fatty acids in that it occurs in significant amounts only in animal tissues and is present in relatively small quantities. In- deed, when measured in animal foods it is typically measured in milligrams per serving, while fat is measured in grams per serving. Cholesterol, which is synthesized by the body, also differs chemically and functionally from fat. While both are lip- ids, cholesterol is a complex substance belonging to a class of chemical com- pounds called “sterols,” while fats be- long to a class called “glycerides.” Fats are associated mainly with the body’s production and storage of energy; in ad- dition, they supply the body with insula- tion and help cushion some of our inter- nal organs. In contrast, cholesterol plays an essential role in the synthesis of cell membranes, steroid hormones, and bile acids. On the negative side, elevated blood levels of cholesterol can produce adverse health effects of great conse- quence.
DIET AND HEART DISEASE
Coronary heart disease is the single most common cause of death and disabil- ity in 31 of PAHO’s 35 member countries. Indeed, it accounts for more deaths an- nually than any other disease, including all forms of cancer combined, and the trend seems to be increasing. Many young and highly productive people have symptomatic coronary heart dis- ease, while others have asymptomatic disease that has not yet been diagnosed. Besides being limiting, costly, and of- ten painful to the individuals involved, coronary heart disease constitutes an im-
portant direct and indirect drain on na- tional economies. Moreover, as popula- tions tend to age and become more urbanized, individual exposure to the risks of coronary heart disease can be ex- pected to increase. Consequently, the public health prognosis regarding coro- nary heart disease in the Americas is not good, and preventive measures are seri- ously needed. For many individuals, the severity of atherosclerosis increases in a linear fash- ion with rising plasma cholesterol levels. According to Grundy (I), when 60% of a coronary artery’s surface is covered with raised atherosclerotic lesions-or fibrous plaques that contain significant amounts of cholesterol and its fatty acid esters- one crosses a critical threshold for in- creased risk of clinical coronary heart dis- ease, as shown in Figure 1. When the arterial wall is injured me- chanically (as a result of smoking, hyper-
Figure 1. A diagram of the relationship between coronary atherosclerosis (based on the percent- age of the surface of coronary arteries covered with raised lesions) and age at different levels of plasma cholesterol. When 60% of the coronary artery surface is covered with lesions, patients enter the zone of markedly enhanced risk for clinical coronary heart disease. From S.M. Grundy, jAMA, 28 November 1986, vol. 256, no. 20, pp. 2849-2856 (1). Plasma cholesterol,q mg/dL(mmolll) (^) 350BOOS)
30 40 50 60 Age In years
70 80
198 Bullefin ofPAHO 24(2), 1990
adults derive less than 30% of their total caloric intake from fats, and that choles- terol intake be less than 100 mg per 1, calories, not to exceed 300 mg per day (2, 8). In 1984 the Inter-Society Commission for Heart Disease Resources in the United States published a statement that recommended reducing dietary choles- terol to no more than 250 mg per day, reducing total fat intake to less than 30% of total caloric intake, and adjusting fat intake so that saturated fats provided no more than 8% of the total calories (9). This commission was created in 1969 to develop guidelines for optimal medical resources to prevent and treat cardiovas- cular disease. Its members include the AHA, the American Medical Association, the American College of Cardiology, the American College of Physicians, the American Public Health Association, and 12 other medical associations. In 1982 the World Health Organization Expert Committee on Prevention of Cor- onary Heart Disease recommended that in countries with a high incidence of cor- onary heart disease, blood cholesterol levels should be lowered through pro- gressive changes in eating patterns. These patterns should include limitation of cholesterol intake to less than 300 mg per day, limitation of total fat energy to 30% of total energy, and limitation of sat- urated fat energy to less than 10% of total energy (10).
CLINICAL STUDIES
In January 1984 the National Heart, Lung, and Blood Institute (NHLBI) re- leased the results of a 7 to 10 year clinical study it had sponsored that was known as the Lipid Research Clinic’s Coronary Primary Prevention Trial. These results showed that reducing serum cholesterol by means of the cholesterol-lowering
drug cholestyramine (a bile acid seques- tram) resulted in a reduction in fatal and nonfatal heart attacks (II). In addition, the study results revealed a linear corre- lation between cholesterol reduction and a reduction in the risk of coronary heart disease. More specifically, the double-blind placebo-controlled study dealt with 3, hypercholesterolemic men at high risk of developing coronary heart disease. The men who were actively treated achieved a 19% reduction in coronary heart disease risk as a direct result of an 8.5% reduction in plasma cholesterol. Thus, on the aver- age, every 1% decline in plasma choles- terol yielded a 2% reduction in the risk of coronary heart disease (2 I). Before initiation of the study, all poten- tial participants were placed on a three- month diet that contained about 400 mg of cholesterol per day and that was de- signed to provide a polyunsaturated-to- saturated fatty acid ratio of approxi- mately 0.8. This diet resulted in a 3.5% reduction in total plasma cholesterol and a 4.0% reduction in LDL-cholesterol, the lipoprotein-cholesterol complex associ- ated with increased risk of coronary heart disease (12). NHLBI concluded that the study’s findings, “taken in conjunction with the large volume of evidence relat- ing diet, plasma cholesterol levels and coronary heart disease support the view that cholesterol lowering by diet also would be beneficial” (11). The study, however, followed men with no previous evidence of coronary heart disease. Consequently, to learn if the actual atherosclerosis disease process could be reversed or slowed by a con- comitant reduction in blood cholesterol levels, scientists from the University of Southern California’s Atherosclerosis Re- search Institute performed a two-year trial known as the Cholesterol Lowering Atherosclerosis Study.
200 Bulletin ofPAHO 24(2), 1990
This study enrolled 162 nonsmoking men with normal blood pressures who had previously undergone coronary ar- tery bypass graft surgery and who were clinically stable. These subjects were ran- domly assigned either to a cholesterol- lowering drug intervention group (whose members received colestipol hy- drochloride plus niacin) or to a placebo group. Each group was given a cholesterol-lowering diet. The placebo group’s diet included less than 250 mg of cholesterol per day and provided 26% of the subject’s total energy intake in the form of fat (10% as polyunsaturated fat, 5% as saturated fat, and 11% as monoun- saturated fat). The drug group had a much stricter diet, one allowing only half the placebo group’s daily cholesterol in- take. Each subject was also given an angi- ogram (arterial X-ray) of his coronary, ca- rotid, and femoral arteries and his coronary bypass grafts, both before and after two years of study participation (13). The before-and-after results in the diet- only (placebo) group were statistically significant: Decreases were seen in the levels of total cholesterol (4%), triglycer- ide (5%), LDL-cholesterol (5%), and the ratio of LDL-cholesterol to HDL- cholesterol (6%). More dramatic de- creases were seen in the drug interven- tion group’s observed levels-except in the level of the HDL-cholesterol complex, which increased by 37% (23). Sixty-one percent of the men in the drug intervention group showed either no change or improvement of their ather- osclerotic status (16.2% actually im- proved). Along these same lines, 39% of those in the placebo group showed either no change or improvement (2.4% actu- ally improved). The results of this study demonstrated for the first time that non- smoking men with normal blood pres- sures and prior coronary bypass surgery
could actually reverse the atherosclerotic disease process through significant re- ductions in blood cholesterol levels (23). More generally, two relationships have emerged from these and other clinical studies involving reduced blood choles- terol levels. First, the NHLBI-sponsored study discussed above showed that every 1% reduction in plasma cholesterol yielded an average 2% reduction in the risk of coronary heart disease (II). And second, the Multiple Risk Factor Inter- vention Trial (MRFIT), another very large study sponsored by the National Insti- tutes of Health, data from which were applied to the Framingham Heart Study equation, showed that every 1 mgldl re- duction in serum cholesterol yielded an average 1% decrease in coronary heart disease mortality over a six-year period (24). Figure 2 shows the serum
Figure 2. The relationship between serum cho- lesterol and death from coronary heart disease in 361,662 men 35-57 years of age.
1
140 ' 160 180 ' 200 220 ' 260 ' 240 280 300 Serumcholesterol.mgidl
Source: M. j. Martin, S. B. Hulley, W S. Browner, L. H. Kuller, and D. Wentworth. Serum cholesterol, blood pressure, and mortality: Implications from a cohort of 361,662 men. Lancet 2:933, 1986.
Segal Dietary Lipids and Coronary Hearf Disease 201
Table 1. Dietary therapy for high blood cholesterol.
Recommended intake Nutrient Step 1 diet Step 2 diet Total fat Lessthan 30% of^ total^ calories^ Less^ than 30% of^ total^ calories Saturated fatty acids Less than 10% of total^ calories^ Less than 7% of total^ calories Polyunsaturated fatty acids Up to 10% of total calories Up to 10% of total calories Monounsaturated fatty acids 1 O-l 5% of total calories 1O-l 5% of total calories Carbohydrates 5040% of total calories 50-60% of total calories Protein 7O-20% of total calories 7O-20% of total calories Cholesterol Less than 300 mglday Less than 200 mglday Total calories To achieve andmaintain desirable To achieve and maintain desirable weight weight Source: United States,National Institutes of Health, National Heart, Lung, and Blood Institute (15).
saturated fat and cholesterol. When di- etary cholesterol is reduced, lowered blood levels usually will reflect this de- crease. Dietary intake of saturated fatty acids (SFA) has the most potent effect upon blood cholesterol levels, specifically LDL- cholesterol levels. Elevated SFA intake down-regulates the LDL receptor activity and also increases the rate of LDL syn- thesis (2, 2, 5, 26). Likewise, a positive energy balance that leads to obesity causes an overproduction of LDL and an increase in the rate of cholesterol synthe-
sis, resulting in an exaggerated increase in blood cholesterol and LDL-cholesterol levels (I, 2, 5, 16). When blood cholesterol levels are ele- vated in the mild to moderate range, di- etary excesses are suspected. More se- vere elevations are attributed to a genetic component (3, 5). When, for example, a group of people in one geographic area consumes a fairly uniform diet, and yet wide variations in plasma cholesterol lev- els exist, inherited differences in LDL re- ceptor activity are considered a major cause. Nevertheless, even when a ge-
Table 2. Risk status based on the presence of coronary heart disease (CHD) risk factors other than LDL-cholesterol. The patient is considered to have a high-risk status if he or she has one of the following: l Definite CHD, as indicated by the characteristic clinical picture and objective laboratory findings of either: Definite prior myocardial infarction, or Definite myocardial ischemia, such as angina pectoris l Two other CHD risk factors: Male sexa Family history of premature CHD (definite myocardial infarction or sudden death before 55 years of age in a parent or sibling) Cigarette smoking (currently smokes more than 10 cigarettes per day) Hypertension Low HDL-cholesterol concentration (below 35 mg/dl, confirmed by repeated measurement) Diabetes mellitus History of definite cerebrovascular or occlusive peripheral vascular disease Severe obesity (B 30% overweight) a Maleness is considered a risk factor in this scheme because the rates of CHD are three to four times higher in men than in women in the middledecades of life and roughly two times higher in the elderly. Hence, a man with one other CHD risk factor is considered to have a high-risk status,whereas a woman is not so considered unless she has two other CHD risk factors. Source:United States,National Institutes of Health, National Heart, Lung, and Blood Institute (15).
Segal Dieta y Lipids and Coronay Heart Disease 203
netic basis exists for elevated cholesterol levels, most individuals respond to di- etary changes (Z-3, 15, 16). Several prediction formulas have been developed in the United States that esti- mate the serum cholesterol level based on dietary intakes of polyunsaturated fatty acids (PUFAs), SFAs, and choles- terol. These formulas, developed for an average population of middle-aged North American males, illustrate numeri- cally the profound influence of SFAs on blood cholesterol levels (18). Dietary PUFAs reduce serum cholesterol levels by half the amount that SFAs increase them. Dietary cholesterol makes a much less significant contribution to serum cholesterol levels, as indicated by For- mula 2, where the effect of dietary cho- lesterol intake varies only with the square root of that intake. The most commonly used formula, known as a modified Keys formula or the Keys/Anderson formula, is used when one knows the percentage of total daily caloric intake accounted for by lipids. This formula is
- PSC (predicted serum cholesterol) = 41+ 164, and
- qb = 1.26(25-P) + 1.502; so PSC = 1.26(2S-P) + 1.5oz + 164
where: (b = a measure of the serum cholesterol elevating ability of the average in- dividual’s dietary pattern; S = saturated fatty acids expressed as a percentage of total daily caloric in- take; only those acids with carbon chain lengths of 12, 14, and 16 are included; P = total polyunsaturated fatty acids expressed as a percentage of total daily caloric intake; and Z = the square root of cholesterol in- take expressed as milligrams per 1,000 kcal of daily caloric intake.
204 Bulletin ofPAHO 24(2), 1990
Regarding the coefficient 1.26 in equa- tion 2, another coefficient (1.35) had pre- viously been used. However, this in- cluded stearic acid and fatty acids with fewer than 12 carbon atoms. It now ap- pears that only saturated fatty acids with 12, 14, and 16 carbon atoms affect serum cholesterol levels. To date, the reason why stearic acid does not behave like lauric, myristic, and palmitic acids is un- known. However, stearic acid does in- crease plasma triglyceride levels (28). Another form of the Keys/Anderson formula may be used when the daily in- takes in grams of saturated and polyun- saturated fatty acids are known, rather than the proportional share of these lip- ids in total caloric intake. This formula, which is typically more convenient to use, is 1,134(2Sg - Pg) 6= =^ 1. CE where: Sg = total daily intake of saturated fatty acids in grams; Pg = total daily intake of polyunsatu- rated fatty acids in grams; C E = total daily kcal intake; and Z = the square root of cholesterol in- take in mg per 1,000 kcal of daily caloric intake. It should be noted that Z can be ex- pressed as
z= 1,000 C Chol CE where C Chol = daily dietary cholesterol intake in mg and C E = total daily kcal intake. By knowing the relationships shown in these equations, one can predict how di- etary alterations such as decreases in SFAs and increases in PUFAs can be ex- pected to affect the blood cholesterol level.
lowers LDL-cholesterol by substituting for saturated fatty acid intake. Other studies dispute this finding and claim that it does have its own independent ef- fect. It has also been found that besides lowering LDL-cholesterol, linoleic acid lowers HDL-cholesterol (2, 3, 5, 8, 15). Linoleic acid belongs to the omega- class of PUFAs. (The other class of PUFAs is omega-3, the omega number indicating the number of carbon atoms between the first double bond and the methyl end of the fatty acid.) PUFAs currently account for about 6% of the calories in the U.S. diet, and many groups now recommend that it should be increased to 10% (3, 2, 4, 9, 20, 25). Previously, linoleic acid had been en- thusiastically recommended as a choles- terol lowering agent. This PUFA is found abundantly in several vegetable oils- notably safflower, sunflower seed, soy- bean, and corn oil. These oils, while high in PUFAs and low in saturated fatty ac- ids, are also very rich in total calories. Increasing their share in the diet requires a compensatory decrease elsewhere; oth- erwise, the resulting weight gain would facilitate a rise in blood cholesterol. In this regard, the recommendation is that the increased PLEAS replace some of the saturated fats. More recently, however, concern has been expressed about the long-term ef- fects of linoleic acid. No population group appears to have consumed more than 10% of its calories in the form of linoleic acid over a long time period, and the literature on linoleic acid has reported potential problems (1, 14, 16). For exam- ple, it has been reported that linoleic acid is carcinogenic in laboratory animals, that it may increase one’s risk of gallstones if used in large amounts, and that it may suppress the immune system. Also, in- gestion of linoleic acid in large quantities is known to alter the cell membranes. Therefore, because we do not yet know
all the possible consequences resulting from long-term, high-dose consumption of this most common dietary PLEA, most experts feel justified in recommending that the intake of PUFAs not exceed 10% of the daily caloric intake. The other class of PUFAs, the omega- fatty acids, are found abundantly in fish oils. The two most common of these highly polyunsaturated fatty acids in the diet are eicosapentaenoic acid and doco- sahexaenoic acid. Recent claims that omega-3 oils are use- ful in the prevention of coronary heart disease appear premature. Actually, their value in reducing total serum cholesterol levels is unpredictable, contrary to claims now being made in the promotion of commercial supplements. Several studies indicate that the omega-3 oils may not lower LDL-cholesterol any more than the omega-6 PUFAs, and also indicate that they appear to lower protective HDL- cholesterol levels. Other studies have found them to raise LDL-cholesterol lev- els (1, 15). By inhibiting the synthesis of very low-density lipoproteins (VLDL), omega-3 oils do reduce VLDL levels, and they are effective in reducing plasma tri- glyceride levels. Omega-3 oils also appear to have an inhibitory effect on platelet aggregation and stickiness, thereby delaying clotting time and the potential for thrombosis (1, 15). This, in fact, may be their most im- portant feature in protecting against cor- onary heart disease. Fish oils may also have anti-inflammatory properties that tend to protect blood vessel walls against lipid deposition damage (5). Large long-term intakes of the omega- fatty acids have yet to be evaluated for adverse effects on various bodily sys- tems. Therefore, organizations providing expert advice do not recommend that the general public use high-dose supple- ments of fish oils to lower cholesterol lev- els or protect against coronary heart dis-
206 Bulletin ofPAHO 24(Z), 1990
ease. One concern is that large doses of omega-3 oils might cause hemorrhagic complications-especially in persons us- ing large doses of aspirin or other antico- agulants. However, public consumption of fresh fish should be encouraged because fish is an excellent substitute for meat. Even though some fish contain almost as much dietary cholesterol as many cuts of meat, fish are very low in saturated fatty acids. Also, population studies indicate that in those geographic areas where fresh fish consumption is very high, the risk of cor- onary heart disease is greatly reduced. Regarding monounsaturated fatty ac- ids, the primary monounsaturated fatty acid in the U.S. diet is oleic acid. This acid is found primarily in olive oil, rape- seed oil (canola oil), and peanut oil. Mo- nounsaturates had long been thought “neutral’‘-neither raising nor lowering blood cholesterol levels. More recent evi- dence indicates that when oleic acid is substituted for saturated fatty acids in the diet, it produces the same degree of LDL- cholesterol reduction as linoleic acid. However, oleic acid leaves the HDL- cholesterol level unchanged, whereas linoleic acid reduces it-at least initially (5, WI. Population studies have shown that in Mediterranean areas such as Greece and southern Italy, where olive oil intakes are quite high, the incidence of coronary heart disease is relatively low. Of course, it may be that the high oleic acid intakes are simply substituting for saturated fatty acids. Experts currently recommend that monounsaturated fatty acids, mainly oleic acid, should account for about lo- 15% of total daily calories (15).
OTHER CONSIDERATIONS
Mention should also be made of other dietary considerations. Obesity has been cited as an independent factor in hyperli-
pidemia, and weight loss is one of the most effective techniques for reducing blood cholesterol levels in overweight in- dividuals. Weight reduction slows down rapid rates of cholesterol and lipoprotein synthesis; and in many people it lowers LDL-cholesterol, raises HDL-cholesterol, and reduces plasma triglycerides. In some extremely sensitive individuals, high-risk LDL-cholesterol levels will re- turn to normal values merely with a re- duced caloric intake, weight reduction, and then weight maintenance at a de- sired body weight. Therefore, caloric re- striction and maintenance of desirable weight is especially recommended for overweight individuals at high risk (1,15, 22). Gnmdy (2) has called attention to other high-risk mechanisms not yet fully iden- tified that alter cholesterol and lipopro- tein metabolism but that are not reflected in the measurements of plasma choles- terol levels. Following ingestion of di- etary cholesterol, for example, the cho- lesterol is transported in the gut by chylomicrons-another type of lipopro- tein. After the triglycerides are split off from the chylomicrons, cholesterol-rich “chylomicron remnants” remain, These particles, along with other postprandial lipoproteins (beta-VLDL apolipoprotein A-similar to LDL), are all suspected of being atherogenic. Also, a high dietary fat intake may promote cholesterol ab- sorption, cholesterol synthesis, and the growth of atherosclerotic lesions. It may also increase blood pressure and shift platelet metabolism toward the clotting portion of the equation (2,2). An additional concern about fat con- sumption has surfaced in recent years. Epidemiologic studies have repeatedly shown an association between dietary fat and the occurrence of cancer at several sites. While the data are not entirely con- sistent, and hence the relationship be- tween dietary fat and cancer is not clear,
Segal Dietay Lipids and Coronay Heart Disease 207
Prevention Trial. National Institutes of Health, Bethesda, Maryland, 12 January
Blankenhom, D. H., S. A. Nessim, R. L. Johnson, M. E. Sanmarco, S. l? Azen, and L. Cashin-Hemphill. Beneficial effects of combined colestipol-niacin therapy on coronary atherosclerosis and coronary ve- nous bypass grafts. JAt~%4257:3233-3240,
United States, National Institutes of Health, Multiple Risk Factor Intervention Group. Statistical design considerations in the NHLI Multiple Risk Factor Inter- vention Trial (MRFIT). Journal of Chronic Diseases30:261-275,1977.
United States, National Institutes of Health, National Heart, Lung, and Blood Institute. Report of the National Choles- terol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Arch Intern Med 148:36-69, 1988.
McNamara, D. J. Predictions of plasma cholesterol response to dietary choles- terol. Am J Clin Nutr41:657-658,1985.
United States, National Institutes of
Health. Lipid Research Clinic’s Population Studies Data Book (Vol. 2). NIH Publication No. 82-2014. Bethesda, Maryland, 1982.
- Anderson, J. T., D. R. Jacobs, Jr., N. Fos- ter, Y. Hall, D. Moss, L. Mojonnier, and H. Blackburn. Scoring systems for evalu- ating dietary pattern effect on serum cho- lesterol. Prev Med 8:525-537,1979.
- Keys, A. Serum cholesterol response to dietary cholesterol. Am J Clin Nutr 40:351- 359,1984.
- Hegsted, D. M., R. B. McGandy, M. L. Myers, and F. J. Stare. Quantitative ef- fects of dietary fat on serum cholesterol in man. Am J Clin Nutr 17:281-295,1965.
- Fletcher, E. S., N. Foster, J. T. Anderson, F. Grande, and A. Keys. Quantitative esti- mation of diets to control serum choles- terol. Am J Clin Nutr 20:475-492,1967.
- McNamara, D. J. Effects of fat-modified diets on cholesterol and lipoprotein me- tabolism. Annu Rev Nutr 7:273-290,1987.
- United States, National Research Council, Committee on Diet, Nutrition, and Can- cer, Assembly of Life Sciences. Diet, Nu- trition, and Cancer. National Academy Press, Washington, D.C., 1982.
Segal Dietary Lipids and Coronay He& Disease 209
