For some doctors in affluent countries the first question about prevention of coronary heart disease (CHD) nowadays is whether to write a prescription for one of the statins (simvastatin, pravastatin, fluvastatin, atorvastatin, etc) which
inhibit an early step of cholesterol biosynthesis in the body. Tables are available to show whether the 5- or 10-year risk justifies the cost of long term statin medication, but the relation of diet and CHD is still of primary importance for the majority of people. What we eat is bound up with the aetiology of CHD. Many people do not know their current plasma cholesterol, many coronary deaths occur before medical help and most countries cannot afford these expensive drugs.
Coronary heart disease is the largest single cause of death in Britain and the disease that causes most premature deaths, but it is only one-seventh as common in industrial Japan and rare in the masses in most developing countries. Its incidence must be environmentally determined because immigrant groups soon take on the incidence rate of their new country and there have been large changes in mortality over time. Coronary heart disease was uncommon everywhere before 1925 and then increased steadily in Western countries until the 1970s, except for a dip during the Second World War. Age-standardised mortality rates from coronary heart disease in the United States of America and Australia started to decline from 1966 and have reduced by more than 70%. In Britain rates are higher in Scotland and Ireland than in England, and higher in the north of England than the south. They have been
declining since 1979 and have fallen by about 25%. Most EU countries have shown similar recent modest reductions of coronary mortality, but in the countries of eastern Europe coronary mortalities have risen. They have, however, recently fallen in Poland and the Czech Republic.
Coronary heart disease death rates in six countries, for men aged 25-74
Coronary heart disease is a multifactorial disease, but diet is probably the fundamental environmental factor. The pathological basis is atherosclerosis, which takes years to develop. Thrombosis superimposed on an atherosclerotic plaque, which takes hours, usually precipitates a clinical event. Then whether the patient dies suddenly, has a classic myocardial infarct, develops angina, or has asymptomatic electrocardiographic changes depends on the state of the myocardium. Each of these three processes is affected by somewhat different components in the diet.
The characteristic material that accumulates in atherosclerosis is cholesterol ester. This and other lipids in the
plaque, such as yellow carotenoid pigments, come from the blood where they are carried on low density lipoprotein (LDL). In animals, including primates, atheroma can be produced by raising plasma cholesterol concentrations with high animal fat diets. Much of this cholesterol is present in modified macrophages that have the histological appearance of foam cells. Experimental pathology studies indicate that these cells only take up large amounts of LDL if it has been oxidised. This oxidation probably occurs within the artery wall.
People with genetically raised LDL-cholesterol (familial hypercholesterolaemia) tend to have premature coronary
heart disease. This is accelerated even more in homozygotes who have plasma cholesterols four times normal and all develop clinical coronary heart disease before they are 20.
Thousands of papers have been written on diet and CHD. Since early in the century scientists have suggested links between a series of dietary components and CHD. Some of these were subsequently found to be unconnected or of little importance, for example sucrose, soft water, milk. The latest component to be associated is in the news, but this does not mean that the older components have been disproved—just that well-established facts are not newsworthy.
Over 50 prospective (cohort) studies in more than 600 000 subjects in 21 countries have reported on risk factors associated with or protective against CHD. The three best established risk factors are: raised plasma total and LDL-cholesterol, cigarette smoking, and high blood pressure.
Two step reasoning
High plasma LDL- (and total) cholesterol is firmly established as a major risk factor for CHD, both from cohort study
epidemiology and from randomised controlled trials with statins. In turn, how diet affects LDL-cholesterol concentration can be—and has been—demonstrated in controlled human dietary experiments, in which one dietary component is changed in the experimental period, with control periods on either side or in parallel.
Within-population relation between plasma cholesterol and CHD and total mortality based on 6-year follow up of 350 000 US men.
Plasma total and low density lipoprotein cholesterol (LDL-cholesterol)
About three quarters of plasma total cholesterol is normally in LDL-cholesterol and the higher the total cholesterol the higher the percentage of LDL-cholesterol because HDL-cholesterol rarely exceeds 2 mmol/l (and never exceeds 3). The mean plasma total cholesterol of healthy adults ranges widely in different communities, from 2.6 mmol/l (Papua New Guinea highlanders) to 7.2 mmol/l (in east Finland some years ago). Only in countries whose average total cholesterol exceeds 5.2 mmol/l (200 mg/dl)—as in Britain—is coronary heart disease common.
Percentage distribution of serum total cholesterol in British adults by sex
Dietary components that affect plasma LDL-cholesterol: type of fat
The major influence is the type of fat. Fats in the diet are mostly in the form of triglycerides (triacylglycerols): three
fatty acids joined to glycerol. The most abundant fatty acid(s) determine(s) the effect. Saturated fatty acids raise
LDL-cholesterol; these are mostly 12:0 (lauric), 14:0 (myristic), and 16:0 (palmitic). Palmitic may be less potent but is the most abundant of these saturated fatty acids in foods. 18:0 (stearic) has little or no cholesterol-raising effect.
Monounsaturated fatty acids—the main one is 18:1 (oleic)— in the natural cis configuration have an intermediate effect on LDL-cholesterol: lower than on saturated fatty acids, not as low as on linoleic.
Polyunsaturated fatty acids (PUFA), (with two or more double bonds) lower LDL-cholesterol. The most abundant of these in foods is 18:2 (linoleic) which belongs to the ω-6 (omega-6 or n minus 6, n-6) family of polyunsaturated fatty acids (first double bond, numbering from the non-carboxylic acid end is at 6th carbon). The omega-3 (ω-3) series of PUFAs are less abundant in most foods 18:3, ω-3, α-linolenic occurs in plants and some vegetable oils. 20:5, ω-3, eicosapentaenoic acid (EPA) and 22:6, ω-3, docosahexaenoic acid (DHA) are mostly obtained from fatty fish and fish oils. The cholesterol-lowering effect of ω-3 PUFAs is less important than their other properties.
Omega 3 and omega 6
In unsaturated fatty acids the double bond is normally in the cis configuration and the carbon chain bends at the double bond. If the configuration is trans, straight at the double bond, the fatty acid behaves biologically like a saturated fatty acid. The usual trans fatty acid is 18:1 trans (elaidic) acid, found in foods produced by hydrogenation in making older-type hard margarines.
Dietary cholesterol and phytosterols
Cholesterol is only found in animal foods. Dietary cholesterol has less plasma cholesterol-raising effect than saturated fats. This is because about half the plasma cholesterol comes from the diet and half is biosynthesised in the liver from acetate. When more cholesterol is absorbed it tends to switch off this endogenous synthesis.
Plant oils also contain sterols, but these are phytosterols, for example, β-sitosterol, campesterol, brassicasterol. These typically have one or two more extra carbons on the side chain of the cholesterol molecule. They interfere competitively with cholesterol absorption and are poorly absorbed themselves. Phytosterols in vegetable oils (200-500 mg/100 g) add a little to their cholesterol-lowering effect. They are also present in nuts and seeds. Some premium PUFA margarines (introduced 1999) are enriched with concentrated natural phytosterols (or-stanols) to enhance cholesterol lowering.
Overweight and obesity
Overweight people tend to have raised plasma triglycerides and to a lesser extent total and LDL-cholesterol. Weight
reduction by diet and/or exercise will usually reduce their cholesterol. Overweight, especially abdominal visceral
adiposity, is itself a direct risk factor for CHD.
The effect of dietary fibre depends on the type. Wheat fibre (bran or wholemeal breads) does not lower plasma cholesterol but viscous (“soluble”) types, pectin and guar and oat fibre, in large intakes, produce moderate cholesterol reductions. Although wheat fibre does not lower plasma cholesterol cohort studies consistently show less subsequent CHD in people who eat more wheat fibre and whole grain foods.
Most vegetable foods are low in protein. Soya is an exception. When soya protein replaces animal protein in the diet
there has usually been a reduction of plasma total and LDL-cholesterol. Although many human trials have been
carried out, the mechanism has been elusive.
Coffee contains small amounts of diterpenes (lipids), cafestol and kahweol—not caffeine—that raise plasma total and LDLcholesterol. Several cups a day of boiled, plunger or espresso coffee can raise the cholesterol but filtered or instant coffee does not—the diterpenes have been removed from the beverage.
Mechanisms for LDL-cholesterol lowering
Many complex experiments have been done to elucidate how different fatty acids affect LDL-cholesterol. The main
mechanism appears to be by effect on the number and activity of the LDL-receptors in cell membranes. Saturated fatty acids downregulate these receptors, so less cholesterol is taken up from the plasma; unsaturated fatty acids have the opposite effect. In overweight people there is increased secretion of very low density lipoprotein (VLDL) from the liver.
Cis unsaturated fatty acids are bent at the double bond(s), trans fatty acids are not (1). The relation between body mass index (weight/height2) and total cholesterol, HDL-cholesterol and triglycerides (all in mmol/l).
Large amounts of viscous (soluble) dietary fibre increase viscosity in the lower small intestine and reduce reabsorption of bile acids, so producing negative sterol balance, hence increased cholesterol→bile acids (cholestyramine effect). The mechanism for the potent plasma cholesterol-raising effect of coffee lipids has not yet been worked out (plasma aminotransferase goes up too); no animal model has been found.
Plasma high density lipoprotein cholesterol (HDL-cholesterol)
HDL-cholesterol is a potent protective factor in communities with high LDL- and total cholesterols.2 It appears to act by mobilising cholesterol from deposits in peripheral tissues, including arteries, and transporting it to the liver for disposal (“reverse cholesterol transport”). Levels of plasma HDLcholesterol do not explain the big differences of coronary disease incidence between countries; its concentration is often lower in countries with little coronary heart disease. But in countries with a high incidence of CHD and high plasma-LDLcholesterol, individuals with above average HDL-cholesterol have a lower risk of the disease. HDL-cholesterols are higher in women (related to oestrogen activity), a major reason why coronary disease usually affects women at older ages than men.
Low HDL-cholesterols are often associated with raised plasma triglycerides and the latter metabolic dysfunction may
compound the risk of coronary disease. HDL-cholesterols tend to be lower in overweight people, in those with diabetes, and in those who smoke. They may be reduced by a high carbohydrate (that is, low fat) diet. They are raised by alcohol consumption, by moderate or heavy exercise, by reduction of body weight, and by high fat diets.
Increased HDL concentration is the clearest reason why moderate alcohol consumption is associated epidemiologically with reduced risk of CHD. Note that above two drinks per day, total mortality goes up because of other diseases and accidents associated with alcohol.
When someone changes from a typical Western diet to a low fat (therefore high carbohydrate) diet LDL-cholesterol goes down, (good!) because percentage saturated fat was reduced, but HDL-cholesterol goes down as well (may not be so good). If instead the fat intake is maintained but saturated fat is replaced by polyunsaturated and monounsaturated fats, LDL also goes down but with little or no reduction of HDL-cholesterol. Changing fat type like this should give a lower risk of coronary disease but reducing total fat intake is better for the management of overweight.
If a patient has raised plasma triglycerides the first question is whether they had been fasting when the blood was taken. The next question is whether the hypertriglyceridaemia is a pointer to other risk factors that tend to be associated with it: high plasma cholesterol, overweight, lack of exercise, glucose intolerance, low-HDL-cholesterol or other metabolic disease (renal disease, hypothyroidism). A common cause of increased plasma triglycerides is excessive alcohol indulgence the evening before blood was taken.
The management of hypertriglyceridaemia consists of looking for and dealing with any of the common associations.
The non-pharmacological treatment is more exercise, fewer calories (weight reduction), and less alcohol. Reduced
carbohydrate is not advised; it implies an increased fat intake which can only increase lipaemia during the day. People with exaggerated postprandial lipaemia appear to have an increased risk of coronary heart disease. Fish oil (for example, Maxepa) is a nutritional supplement with a powerful plasma triglyceridelowering
effect and regular consumption of fatty fish also lowers plasma triglycerides.
- Triglycerides in the blood after overnight fast are mainly in VLDL (very low density lipoprotein), synthesised in the liver, hence endogenous. Triglycerides in casual blood samples taken during the day may be mainly in chylomicrons, after a fatty meal, and hence exogenous.
- In prospective studies, raised fasting triglycerides have often shown up as a risk factor for coronary heart disease in single-factor analysis. But hypertriglyceridaemia is likely to be associated with raised plasma cholesterol, or overweight/ obesity, or glucose intolerance, or lack of exercise or low HDL-cholesterol. When these are controlled, increased triglycerides is certainly not as strong a risk factor as hypercholesterolaemia but it has emerged in some studies as an independent coronary risk factor, more often in women.
Other risk factors
High blood pressure is discussed in chapter 2; overweight and inactivity in chapter. Increased levels of two of the coagulation factors, Factor VII and fibrinogen, have been clear in some prospective studies (they were not assayed in most studies). Factor VII activity is increased during alimentary lipaemia after a fatty meal and is persistent in people with hypertriglyceridaemia. Plasma fibrinogen is raised in people who smoke and in obesity; it is reduced by alcohol consumption.
The LDL oxidation hypothesis of atherogenesis predicts that if LDL carries more lipid-soluble antioxidants they should provide some protection against CHD. The principal antioxidant in LDL is -tocopherol, vitamin E (average
7 tocopherol molecules per LDL particle). Its concentration can be raised by intake of vitamin E supplements. In vitro (outside the body) extra vitamin E delays the oxidation of LDL (by copper). In two large prospective studies, one in US nurses, the other in health professionals, those with high intakes of vitamin E experienced less subsequent CHD. But these high intakes of vitamin E were achieved by taking supplements, and people who regularly take vitamin supplements are likely to have more health conscious lifestyles than the average citizen.
Five large randomised controlled prevention trials, in Western populations, with acronyms ATBC,14 GISSI,15 HOPE,
PPP, and CHAOS involving 56 000 subjects have now been reported. There was no reduction of cardiovascular disease or mortality. LDL contains smaller amounts of carotenoids, which are also lipid-soluble antioxidants. But supplements of β-carotene have also not prevented CHD in large randomised controlled trials.
Polyunsaturated fatty acids, 18:2, 20:5 and 22:6 are more susceptible to peroxidation in vitro than saturated or
monounsaturated acids but in the whole body there is a lot of evidence that PUFA intake is negatively associated with CHD.
No significant benefit from vitamins C and E and b-carotene in MRC/BHF secondary prevention trial in over 20 000 subjects
Homocysteine metabolism in humans. Enzymes
In the inborn error of metabolism homocystinuria, plasma homocysteine is so high that it spills into the urine and vascular diseases are among the complications. Then during the 1990s evidence accumulated (many case-control studies and several prospective studies) that lesser degrees of elevated plasma homocysteine (above 16 mol/l total homocysteine, tHcy) are a largely independent risk factor for CHD. They also increase the risk of cerebral and peripheral arterial diseases and even venous thrombosis.18 Raised plasma homocysteine appears to
both damage the endothelium and increase liability to thrombosis.
Homocysteine is an intermediary metabolite of the essential amino acid, methionine (it is methionine minus its terminal methyl group). Folic acid is co-factor for the enzyme in a pathway that re-methylates homocysteine back to methionine.
In apparently well-nourished people folic acid lowers elevated homocysteine by about a quarter.19 A dose of 0.5 mg or even 200 μg folic acid is effective. Plasma homocysteine is also increased in mild vitamin B-12 deficiency. Folic acid may be a safe, inexpensive way of reducing vascular disease. Randomised controlled trials are under way.
Dangerous arrhythmia is one of the two major causes of death in CHD. Over half the deaths occur before the arrival of paramedical or medical help. Then in the ambulance or coronary care unit the treatment of ventricular fibrillation saves lives. Developments in nutrition research are showing, with animal experiments, that electrical instability of ischaemic myocardium is influenced by the fatty acid pattern of the diet and hence of myocardial membranes. In rats or marmoset monkeys fed polyunsaturated oils, fewer animals had sustained ventricular arrhythmia when a coronary artery was tied, than in animals that had been fed on saturated fat or (monounsaturated) olive oil.20 The fish oil group were more resistant to arrhythmia than the sunflower oil group (ω-6 linoleic acid). Canola oil containing linolenic acid (18:3, ω-3), the plant ω-3 fatty acid, also appears to reduce arrhythmias.
Kang and Leaf have studied the mechanism of the fatty acid effect with cultured, neonatal, rat ventricular myocytes whose spontaneous contractions are recorded by a microscope and video camera. Eicosapentaenoic acid (20:5, -3) and the plant oil ω-3 acid, 18:3 (linolenic) as well as linoleic acid (18:2, ω-6) prevent tachyrhythmia induced by a variety of chemicals known to produce fatal ventricular fibrillation in humans. It appears that polyunsaturated fatty acids act by binding to sodium channel proteins in the membrane and altering their electrical charge.
The reduction of deaths outside hospital has been a striking feature in countries where coronary death rates have reduced. This may be explained, at least partly, by an anti-arrhythmic effect of increased -6 polyunsaturated fat intake (national fish intakes have not increased).
Platelet function and thrombosis
In patients with symptomatic CHD tests of platelet function have usually indicated activation. Available tests of platelet function are not on lists of risk factors predicting coronary disease; they are in vitro tests and are inevitably indirect. However platelet activation is of course a central phenomenon in myocardial infarction or recurrent angina, so that any diet that reduces platelet aggregation should reduce the risk of coronary disease.
Following up an observation that the rarity of coronary disease in Greenland Eskimos might be due to their heavy
consumption of marine fat, it was discovered that eicosapentaenoic acid (20:5, ω-3) or EPA, a principal fatty acid
of fish oil, displaces arachidonic acid (20:4, ω-6) in platelets, so that when stimulated they produce an inactive thromboxane TXA3 instead of the active TXA2 derived from arachidonic acid. EPA is only present in traces in the body fat of land animals and is absent from vegetable oils. In human experiments fish oil also reduced the levels of PAI-I, plasminogen activator inhibitor-1. Fish oil is therefore a pharmaceutical alternative (for example Maxepa) to
aspirin to reduce the tendency to thrombosis. Results have been mixed in trials with fish oils to see if they delay
restenosis after coronary angioplasty.
Dietary components associated directly with coronary disease in cohort epidemiological studies
Most of the many prospective studies involving coronary heart disease have not measured diet. It is much more complex and expensive to estimate all the different foods, and thence to compute all the nutrients, than to measure blood pressure or plasma lipids. Of all the parts of a total diet there have been most reports of alcohol intake. It is simpler to include in a questionnaire than to tackle the intricacies of type of fat intake.
In the minority of prospective studies that did report on foods or food components, most have used food frequency
questionnaires (chapter 12), which are easier to handle than open-ended dietary records. Another method, occasionally used, is to measure objective biomarkers of food intake such as plasma fatty acid pattern. Interpretation of associations in the table must allow for uncertainties in assessing usual food intake, and confounding between different food components and with lifestyle. Vitamin E findings have not been confirmed in randomised controlled trials.
Dietary components directly related to CHD
Adding a statin to the diet
Treatment with statins lowers raised plasma cholesterol by average 20% and LDL-cholesterol 25%, without lowering
HDL-cholesterol, and reduces subsequent CHD events significantly. Statin treatment has also been shown to reduce
CHD events by about 24% in people who had survived a myocardial infarction and had average plasma cholesterols of around 5.4 mmol/l.
Note that a statin is prescribed (as the manufacturers state) as an adjunct to diet and normally after a proper trial of a cholesterol lowering diet. The dietary principles described in this chapter lower plasma cholesterol by different mechanisms from the HMG COA reductase inhibition by statins. Parts of diets used to protect against CHD do not act by lowering LDL-cholesterol, for example, only by diet and exercise can overweight be treated.
Statins are very expensive at present, either for the patient or the health service, and we do not yet know if there might be long-term complications. Put very simply the indications for adding a statin to diet are for patients with:
- existing clinical CHD
- two or more coronary risk factors and high plasma cholesterol
- no or one coronary risk factor and very high plasma cholesterol.
In assessing the plasma cholesterol, LDL-cholesterol should be used or total cholesterol/HDL-cholesterol (after repeat measurements in a good laboratory). Risk factors are diabetes, hypertension, smoking, strong family history.
The dietary prescription
Reduction is not essential for improving plasma lipids but should reduce coagulation factors and daytime plasma
triglycerides and contribute to weight reduction.
Saturated fatty acids
Principally 14:0, 16:0 and 12:0 should be substantially reduced from around 15% of dietary energy in many Western
diets to 8-10%.
Polyunsaturated fatty acids
Mainly linoleic acid (18:2, ω-6): they should be about 7% of dietary energy (present British level), up to 10%. Omega-3 polyunsaturated fatty acids should be increased, both 20:5 and 22:6 from seafoods and 18:3 from canola rapeseed)
Monounsaturated fatty acids
Ideal intake if total fat 30%, saturated 10% and polyunsaturated 8% would be 12% of total dietary energy.
Trans fatty acids
With the help of margarine manufacturers these have been reduced. The Department of Health recommends no more than 2% of dietary energy. Avoid older hard margarines.
This boils down to the question of egg yolks. Eggs are a nutritious, inexpensive and convenient food. The Department of Health recommends for the general population no rise in cholesterol intake.
Restriction to under 6 g/day is advised for the general population (100 mmol Na). It is more important for coronary
The Department of Health recommends at least twice a week, preferably fatty fish. It should not be fried in saturated fat.
Eat plenty of high fibre and whole grain cereal foods, including oatmeal.
Vegetables and fruit
These are low in fat, and contain pectin and other fibres, flavonoids and other antioxidants, and they contain folate.
Expert Committees in Britain and the USA recommend five servings of different vegetables and fruit per day (400 g/day average weight).
(Not salty soy sauce) recommended.
In moderation, one or two drinks per day is beneficial for middle-aged people at risk of CHD but cannot be
recommended for the general population because of the greater danger of accidents in younger people and of all the
medical complications of excessive intake.
Should be instant or filtered.
1. Truswell AS. Cholesterol controversy. BMJ 1992; 304: 912-13.
2. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med 1989; 320:915-24.
3. Keys A, Aravanis C, Blackburn H et al. Seven countries: a multivariate analysis of death and coronary heart disease. Cambridge, Massachusetts: Harvard University Press, 1980.
4. Martin MJ, Hulley SB, Browner WS, Kuller LH, Wentworth D. Serum cholesterol, blood pressure and mortality implications from a cohort of 361 662 men. Lancet 1986; ii: 933-6.
5. Gregory J, Foster K, Tyler H, Wiseman M. The dietary and nutritional survey of British adults. London: HMSO
6. Thelle DS, Shaper AG, Whitehead TP, Bullock DG, Ashby D, Patel J. Blood lipids in middle-aged British men. Br Heart J 1983; 49: 205-13.
7. Truswell AS. Cereal grains and coronary heart disease. Eur J Clin Nutr 2002; 56: 1-14.
8. Hendricks HFJ, Westrate JA, Van Vliet T, Meijer GW. Spreads enriched with three different levels of vegetable oil sterols and the degree of cholesterol lowering in normocholesterolaemic and mildly hypercholesterolaemic subjects. Eur J Clin Nutr 1999; 53: 319-27.
9. Urgert R, Meybom S, Kuilman M et al. Comparison of effect of cafetiere and filtered coffee on serum concentrations of liver aminotransferases and lipids: six month randomised controlled trial. BMJ 1996; 314: 1362-6.
10. Knuiman JT, West CA. Differences in HDL cholesterol between populations: no paradox? Lancet 1983; i: 296.
11. Boffeta P, Garfinkel L. Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study. Epidemiology 1990; 1: 342-8.
12. Tunstall-Pedoe H, Woodward M, Tavendale R, Brook RA, McClusky MK. Comparison of the prediction by 27 different factors of coronary heart disease and death in men and women of the Scottish heart health study: cohort study. BMJ 1997; 315: 722-9.
13. Miller GJ. Postprandial lipid metabolism and thrombosis. Proc Nutr Soc 1997; 56: 739-44.
14. Rapola JM, Virtamo J, Ripatti S et al. Randomised trial of tocopherol and carotene supplements on incidence of
major coronary events in men with previous myocardial infarction. Lancet 1997; 349: 1715-20.
15. GISSI-Prevenzione Investigators. Dietary supplement with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Lancet 1999; 354: 447-55.
16. Hu FB, Stampfer MJ, Manson J et al. Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med 1997; 337: 1491-9.
17. Heart Protection Study Group. MRC/BHF Heart Protection Study of antioxidant vitamins supplementation in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002; 360: 23-33.
18. Bouskey CJ, Beresford SAA, Omenn GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intake. JAMA 1995; 274: 1049-57.
19. Homocysteine Lowering Trialists’ Collaboration. Lowering blood homocysteine with folic acid based supplements: metaanalysis of randomised trials. BMJ 1998; 316: 894-8.
20. McLennan PL. Relative effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on cardiac
arrhythmias in rats. Am J Clin Nutr 1993; 57: 207-12.
21. Kang JX, Leaf A. Antiarrhythmic effect of polyunsaturated fatty acids. Recent studies. Circulation 1996; 94: 1774-80.
22. Beaglehole R, Stewart AW, Jackson R. Declining rates of coronary disease in New Zealand and Australia. Am J Epidemiol 1997; 145: 707-13.
23. Burr ML, Fehily AM, Gilbert JF et al. Effects of changes in fat, fish and fibre intakes on death and myocardial reinfarction:Diet and Reinfarction Trial (DART). Lancet 1989; ii: 757-61.
24. Truswell AS. Review of dietary intervention studies: effect on coronary events and on total mortality. Aust NZ J Med 1994; 24: 98-106.
25. de Lorgeril M, Renaud, Mamalle N et al. Mediterranean alphalinolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 1994; 343: 1454-9.
26. Sacks FM, Pfeffer MA, Moye LA et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996; 335: 1001-9.
27. Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel of Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486-97.
28. Department of Health. Nutritional Aspects of Cardiovascular Disease. Report on the Cardiovascular Review Group, Committee on Medical Aspects of Food Policy. London: HMSO, 1994.
29. National Heart Forum. At least five a day. Strategies to increase vegetable and fruit consumption. London: The Stationery Office, 1997.