Diet-heart issues in a pharmacological era

Abstract

Reduction of serum cholesterol low-density lipoprotein (LDL) is an important step in secondary and primary prevention of coronary heart disease, and clinical trials have now shown that cholesterol lowering drugs substantially lessen coronary risk within a few years.1Yusuf S Anand S Cost of prevention.in: The case of lipid lowering. Circulation. 93. 1996: 1774-1776Google Scholar These findings will undoubtedly increase the popularity of cholesterol lowering by drugs at the expense of dietary modification. However, properly applied dietary changes should be the first step in cholesterol lowering. In this short overview I focus on two elements of a healthy-heart diet—fatty acids and antioxidants. Fatty acids are a major determinant of serum cholesterol and LDL concentrations. Antioxidants may prevent oxidative modification of LDL, which is an important step in the development of atherosclerosis. I also discuss the effectiveness of dietary intervention. In trials of the first generation of cholesterol-lowering drugs only a modest reduction of cholesterol was achieved—about 10%. In recent trials, however, drugs have depressed cholesterol by about 25%, with substantial reductions in coronary heart disease endpoints. Furthermore, these benefits were achieved without the increase in non-cardiovascular mortality noted in the first-generation trials. Therefore there seems now to be a consensus that cholesterol lowering is effective and safe after myocardial infarction.2Oliver M Poole-Wilson P Shepherd J Tikkanen MJ Lower patients' cholesterol now.BMJ. 1995; 310: 1280-1281Crossref PubMed Scopus (46) Google Scholar The results of secondary prevention trials even suggest that cholesterol lowering is effective in a range that was previously viewed as normal— eg, serum cholesterol between 4 and 6 mmol/L.1Yusuf S Anand S Cost of prevention.in: The case of lipid lowering. Circulation. 93. 1996: 1774-1776Google Scholar Until lately, few clinicians favoured drug treatment for symptomless individuals with high cholesterol—ie, primary prevention. The WOSCOPS trial3The WOSCOPS Study GroupWest of Scotland Coronary Prevention Study: implications for clinical practice.Eur Heart J. 1996; 17: 163-164Crossref PubMed Scopus (22) Google Scholar will change this. In that trial, men with raised blood cholesterol but no evidence of previous myocardial infarction were randomised to drug or placebo, and the drug-treated group showed reductions in coronary-heart-disease and all-cause mortality comparable with those observed in the secondary prevention trials. The WOSCOPS trial investigators nevertheless concluded that dietary modification, for individuals and for populations, must remain the basis of a coronary prevention strategy. In all the major trials recently published, the cholesterol lowering agents were statins (HMG-CoA reductase inhibitors). These drugs did not increase non-cardiovascular mortality and there was no evidence of adverse effects on liver or skeletal muscle. This will promote cholesterol lowering by drugs with the side-effect of increasing health care costs and medicalisation of society. Therefore, dietary treatment should be the cornerstone of cholesterol lowering, especially in primary prevention. We must, however, acknowledge that in symptom-free high-risk persons even rigorously applied dietary changes may not lower serum cholesterol levels sufficiently. In that case drug treatment is needed to reduce the risk of myocardial infarction. In research on diet and coronary heart disease the cholesterol hypothesis has been prominent for decades. Originally the main determinant of serum cholesterol was thought to be dietary cholesterol. In the 1950s, however, metabolic ward studies by Groen in the Netherlands and by Kinsell, Ahrens, and Keys in the USA made clear that dietary fatty acids were much more important. The results of these intervention studies were summarised in the Keys equation, showing the substantial cholesterol-raising effect of saturated fat, especially fatty acids with 12–16 carbon atoms.4Keys A Anderson JT Grande F Serum cholesterol response to changes in diet: IV, particular saturated fatty acids in the diet.Metabolism. 1965; 14: 776-786Summary Full Text PDF PubMed Scopus (758) Google Scholar Since the mid-1960s a lot more information has emerged on the relations between diet, serum cholesterol, and coronary heart disease. Lipid research has shown that not only total cholesterol (or the highly correlated LDL fraction) is of importance in relation to coronary risk but also the high-density lipoprotein (HDL) fraction. The HDL fraction is an independent protective factor for coronary heart disease. Dietary controlled trials conducted between 1970 and 1991 confirmed that total cholesterol is raised more by specific fatty acids than by unsaturated fatty acids or carbohydrates.5Mensink RP Katan MB Effect of dietary fatty acids on serum lipids and lipoproteins.in: A meta-analysis of 27 trials. Arteriosclerosis Thrombosis. 12. 1992: 911-919Google Scholar These trials showed in addition that all fatty acids raise HDL cholesterol more than do carbohydrates, but the effect diminished with increasing unsaturation. As well as saturated fatty acids, the monounsaturated trans fatty acids can increase serum cholesterol. Trans fatty acids are a byproduct of saturation of polyunsaturated fatty acids either in ruminant animals (such as cows) or in margarine production. Mensink and Katan6Mensink RP Katan MB Effect of dietary trans fatty acids on high density and low-density lipoprotein cholesterol levels in healthy subjects.N Engl J Med. 1990; 323: 439-445Crossref PubMed Scopus (1129) Google Scholar showed that these fatty acids increase LDL cholesterol and decrease HDL cholesterol and may therefore increase coronary risk. In a cholesterol-lowering diet, saturated and trans fatty acids can be replaced by different classes of unsaturated fatty acids (ie, monounsaturated and N-3 and N-6 polyunsaturated fatty acids) or by carbohydrates. Sources of saturated and trans fatty acids are hard margarines and dairy and meat products; of monounsaturated fatty acids, olive oil; of N-3 polyunsaturated fatty acids, fish; and of N-6 polyunsaturated fatty acid, sunflower oil. The metabolic ward studies in the 1950s and 1960s suggested that polyunsaturated fatty acids were superior to monounsaturated fatty acids because of their additional cholesterol lowering effect.4Keys A Anderson JT Grande F Serum cholesterol response to changes in diet: IV, particular saturated fatty acids in the diet.Metabolism. 1965; 14: 776-786Summary Full Text PDF PubMed Scopus (758) Google Scholar However, later work indicates only a slight and unimportant difference in their effects on total and HDL cholesterol.5Mensink RP Katan MB Effect of dietary fatty acids on serum lipids and lipoproteins.in: A meta-analysis of 27 trials. Arteriosclerosis Thrombosis. 12. 1992: 911-919Google Scholar N-3 and N-6 polyunsaturated fatty acids are called essential fatty acids, small amounts being needed for brain development and for prevention of skin lesions, respectively. There is also evidence that consumption of fish, a major source of N-3 polyunsaturated fat intake, once or twice a week lowers coronary risk.7Kromhout D Bosschieter EB De Lezenne Coulander C The inverse relation between fish consumption and 20-year mortality from coronary heart disease.N Engl J Med. 1985; 312: 1205-1209Crossref PubMed Scopus (1824) Google Scholar A low intake of linoleic acid, a major determinant of N-6 polyunsaturated fatty acids in Western diets, is associated with increased coronary risk in cultures characterised by a low N-6 polyunsaturated fat intake—Scotland, for example.8Wood DA Riemersma RA Butler S et al.Linoleic and eicosapentaenoic acids in adipose tissue and platelets and risk of coronary heart disease.Lancet. 1987; i: 177-183Summary Scopus (181) Google Scholar This means that a population intake of at least 4% is needed to prevent coronary heart disease. Replacement of saturated and trans fatty acids by carbohydrates, rather than unsaturated fatty acids, will lower not only total and LDL cholesterol but also HDL cholesterol. Some researchers have therefore argued for replacement by unsaturated fatty acids. In my opinion, there is nothing to choose between these options. Populations with diets low in saturated fatty acids have similar LDL and HDL levels and similarly low coronary risk whether energy needs are made up chiefly with unsaturated fatty acids (eg, Mediterranean diet) or carbohydrates (Japanese diet). In a landmark paper entitled “Beyond cholesterol”, Steinberg and co-workers9Steinberg D Parthasarathy S Carew TE Khoo JC Witzum JL Beyond cholesterol.in: Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 320. 1989: 915-924Google Scholar hypothesised that LDL lipoprotein that has undergone oxidative damage is considerably more atherogenic than native LDL. If this hypothesis is correct the development of atherosclerosis and its complications will be a simultaneous function of plasma levels of LDL and the rate at which LDL undergoes oxidative modification. Hence the combination of LDL-lowering and antioxidant therapy could be additive or synergistic. There is tremendous interest in the possible protective role of antioxidants against coronary heart disease. Which dietary elements should we be looking at? Most of the evidence points in the direction of vitamin E; nonetheless, it is equivocal.10Greenberg ER Sporn MB Antioxidant vitamins, cancer and cardiovascular disease.N Engl J Med. 1996; 334: 1189-1190Crossref PubMed Scopus (87) Google Scholar In some prospective cohort studies dietary vitamin E intake, but not the use of supplements, is protective; in others, a protective effect has been observed only for vitamin E supplement users. In the CHAOS trial, treatment with a high dose of vitamin E did reduce subsequent risk of non-fatal myocardial infarction in patients with established ischaemic heart disease.11Stephens NG Parsons A Schofield PM et al.Randomised controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study (CHAOS).Lancet. 1996; 347: 781-786Summary PubMed Scopus (1669) Google Scholar However, no discernible effect was noted on cardiovascular and all-cause mortality. A promising area of research is that of non-nutritive antioxidants present in plants. Evidence is accumulating that foods of plant origin protect against chronic diseases including coronary heart disease,12National Research Council Diet and health.in: Implications for reducing chronic disease risk. National Academy Press, Washington DC1989Google Scholar and one area of particular interest is the flavonoids, of which more than 4000 different types have been identified. We showed in the Zutphen Elderly Study13Hertog MGL Feskens EJM Hollman PCH Katan MB Kromhout D Dietary antioxidant flavonoids and risk of coronary heart disease.in: The Zutphen Elderly Study. Lancet. 342. 1993: 1007-1011Google Scholar that the intake of flavonoids was inversely related to occurrence of coronary heart disease, and this inverse association was confirmed in a prospective study in Finland. However, prospective studies in the USA, as yet unpublished, have not revealed an association between flavonoid intake and coronary endpoints.14Muldoon MF Kritchevsky SB Flavonoids and heart disease.in: Evidence of benefit still fragmentary. BMJ. 312. 1996: 458-459Google Scholar Also, in the Caerphilly Prospective Study, in Wales, no association was found between flavonoid intake and mortality from coronary heart disease (unpublished results). This was an unexpected finding because of the high consumption of tea, a rich source of flavonoids. However, the British habit of adding milk to tea may have been responsible for this negative finding; the antioxidant activity of tea is totally inhibited by milk.15Serafini M Ghiselli A Ferro-Luzzi A In vivo antioxidant effect of green and black tea in man.Eur J Clin Nutr. 1996; 50: 28-32PubMed Google Scholar Therefore, although the work on flavonoids is promising, much more research is needed before definite statements can be made about their protective power against coronary heart disease. Trials are of great importance in showing whether or not associations are causal—for instance, the association between serum cholesterol and coronary risk. However, once causality has been established, observational studies provide information about the theoretical maximum effect of long-term differences in diet on serum cholesterol and coronary risk. In this context cross-cultural studies are especially informative. The Seven Countries Study was started in the late 1950s under the leadership of Ancel Keys. Between 1958 and 1964 sixteen cohorts of in total 12 763 middle-aged men were examined and they were then followed for 25 years. Dietary information was collected in small random samples of each cohort during the baseline survey. Duplicate portions of the foods consumed were chemically analysed and the average intake of major fatty acids was determined. In 1987 food composites representing the average food consumption of each cohort were chemically analysed in one laboratory. This provided the possibility of determining with state-of-the-art chemical analytical technology the average intake of different nutrients such as fatty acids and vitamins and non-nutritive substances such as flavonoids. At population level, dietary saturated fatty acids were strongly associated with 25-year coronary heart disease mortality rates.16Kromhout D Menotti A Bloemberg B et al.Dietary saturated and trans fatty acids and cholesterol and 25-year mortality from coronary heart disease: the Seven Countries Study.Prev Med. 1995; 24: 308-315Crossref PubMed Scopus (428) Google Scholar Also, significant associations were observed between the population intake and the trans fatty acid elaidic acid, dietary cholesterol, and 25-year mortality rates from coronary heart disease mortality. The independent effect of saturated fatty acids, elaidic acid, and dietary cholesterol could not be studied because of high correlations between these compounds. We also studied the association between the population intake of antioxidants and 25-year coronary heart disease mortality rates. Population average intakes of vitamin E, β-carotene, and vitamin C were unrelated to 25-year coronary heart disease mortality rates. However, the population intake of flavonoids was inversely related to 25-year coronary heart disease mortality.17Hertog MGL Kromhout D Aravanis C et al.Flavonoid intake and longterm risk of coronary heart disease and cancer in the Seven Countries Study.Arch Intern Med. 1995; 155: 381-386Crossref PubMed Scopus (1727) Google Scholar At the population level the combined effect of saturated fat intake, flavonoid intake, and cigarette smoking correlated strongly with 25-year mortality from coronary heart disease (figure). If the new generation of drugs reduce serum cholesterol by 25% can dietary intervention approach this efficacy? In most dietary intervention trials the reductions in serum cholesterol have been much more modest.18Truswell AS Review of dietary intervention studies: effect on coronary events and mortality.Aust NZ J Med. 1994; 24: 98-106Crossref PubMed Scopus (58) Google Scholar However, even in the absence of cholesterol lowering, substantial reductions may be achieved in coronary-heart-disease and all-cause mortality. An example is a secondary prevention trial in which a Mediterranean N-3 polyunsaturated fat enriched diet was used.19de Lorgeril M Renaud S Mamelle N et al.Mediterranean alpha-linolenic acid rich diet in secondary prevention of coronary heart disease.Lancet. 1994; 343: 1454-1459Summary PubMed Scopus (1787) Google Scholar The positive effect of this trial was probably due to a protective effect of the N-3 polyunsaturated fatty acid, α-linolenic acid, on thrombogenesis. When patients on a Western diet with more than 15% of energy from saturated fat go on a step I diet of the National Cholesterol Education Program in the USA, with less than 10% of energy from saturated fat and less than 300 mg cholesterol, serum cholesterol lowering of about 10% is generally obtained.20National Cholesterol Education Program Second report of the expert panel on detection, evaluation and treatment of high blood cholesterol in adults (adult treatment panel II).Circulation. 1994; 89: 1329-1445Google Scholar Cholesterol lowering effects of this order were also obtained at the population level in Finland and the USA during a 20-year period; so these changes are practical and can be achieved if populations are motivated to change from a high saturated fat diet to a diet lower in saturated fat and richer in vegetables and fruits. Larger decreases in serum cholesterol, up to 20%, can be obtained by switching to a step II diet.20National Cholesterol Education Program Second report of the expert panel on detection, evaluation and treatment of high blood cholesterol in adults (adult treatment panel II).Circulation. 1994; 89: 1329-1445Google Scholar This diet contains less than 7% of energy from saturated fat and less than 200 mg cholesterol. It is also rich in plant foods and high in carbohydrates. Most patients will lose weight on a step II diet, an additional cholesterol lowering measure. In the Lifestyle Heart Trial21Ornish D Brown SE Scherwitz LW et al.Can lifestyle changes reverse coronary heart disease?.in: The Lifestyle Heart Trial. Lancet. 336. 1990: 129-133Google Scholar cholesterol lowering of more than 20% has been obtained with such a diet.21Ornish D Brown SE Scherwitz LW et al.Can lifestyle changes reverse coronary heart disease?.in: The Lifestyle Heart Trial. Lancet. 336. 1990: 129-133Google Scholar The extent to which cholesterol declines depends on the intensity of the dietary change; and the rewards in terms of coronary-heart-disease and all-cause mortality are similarly dose-dependent. In a review of seventeen dietary intervention studies Truswell18Truswell AS Review of dietary intervention studies: effect on coronary events and mortality.Aust NZ J Med. 1994; 24: 98-106Crossref PubMed Scopus (58) Google Scholar showed that the odds ratio of intervention versus control groups was 0·94 for total deaths and 0·87 for coronary events. But in the seven trials with most effective cholesterol lowering the odds ratios were 0·89 for all deaths and 0·70 for coronary events. These results suggest a dose-response relation between degree of cholesterol lowering in relation to coronary events and all-cause mortality. From these trials there is no indication of an excess of all-cause mortality; on the contrary. In most reports on coronary primary and secondary prevention, patients are advised to try a step I diet for 3-6 months. At this point they usually show only a small reduction of serum cholesterol and are advised to go on cholesterol lowering drugs. We do, however, know that such disappointing results with dietary therapy are much more likely to be due to poor compliance than to metabolic resistance. Dietary therapy is the area of competence of registered dieticians, who should be involved from the start of the therapy. Cholesterol-lowering diets also need to be continued for years to achieve maximum effect. Therefore the diet of the patient should be the starting point for dietary modification. Having ascertained the patient's likes and dislikes, the physician and the dietician should design a diet that is palatable and can be kept up for a long time. Effective cholesterol lowering by dietary modification is possible. In the future even larger reductions in cholesterol may be obtained if new products (such as margarines enriched with the plant sterol sitostanol) come onto the market. Although definite statements about the protective effect of antioxidants can not be made, an increased consumption of vegetables and fruits can be recommended. In changing people's eating habits one must ensure that the proposed diet is palatable and that it is compatible with good health in general as well as low coronary risk. Cholesterol lowering by drugs is needed for hypercholesterolaemic patients in whom dietary measures fail. But the first choice should be cholesterol lowering by diet.