3.6.1 Diet and blood cholesterol
Cholesterol is a determinant of CHD mortality, and its blood level is at least partly regulated by diet. However, few epidemiological studies have prospec-tively included analyses of the dietary habits of the studied populations in the evaluation of their risk.119 In the Seven Countries Study, marked differences in CHD mortality, dietary habits and cholesterol distribution were observed in the different cohorts.119 Cholesterol levels were high in Northern Europe and in the USA (an average level of 7mmol/L), and low in rural Japan (an average of 4mmol/L), and population cholesterol levels were positively associated with CHD mortality. Secondary prevention trials with statins in Northern Europe120 and Australia121 confirmed the importance of cholesterol by demonstrating a reduction by 25-30 per cent of the relative risk of CHD death in patients taking these drugs. Whether the effect of statins was entirely related to their effect on cholesterol remains unknown.
A major (and often underestimated) finding of the Seven Countries Study was the large difference in absolute risk of CHD death at the same level of serum cholesterol in the different cohorts. At a cholesterol level of about 6 mmol/L, CHD mortality was three times as high in Northern Europe as in Mediterranean Europe (18 per cent vs. 6 per cent). This suggested that factors other than cholesterol were playing an important role. Because of the similarity of the other traditional risk factors and the large differences in the dietary habits of the cohorts,122 it was proposed that the difference in CHD mortality between populations was mainly related to their dietary habits, through biological effects independent of cholesterol.123
This was the basis of a new 'diet-heart hypothesis' in which cholesterol was not the central issue.36,123 In fact, the first dietary trials designed for the secondary prevention of CHD were based on the hypothesis that a cardioprotective diet should primarily reduce cholesterol.36 While the investigators succeeded in reducing cholesterol, they failed to reduce CHD mortality.41 This was mainly attributed to an insufficient effect of the tested diets on cholesterol, and the conclusion was that cholesterol-lowering drugs should be preferred. However, none of the diets tested in these old trials was patterned from the traditional diets of populations protected from CHD (e.g. vegetarian, Asian or Mediterranean), although these diets are associated with low cholesterol.119,122 Also, no trial was aimed at testing the cholesterol-lowering effect of a typical Mediterranean diet, probably because this diet was (and often still is) mistakenly regarded as a high-fat diet, allegedly not appropriate to reduce cholesterol. Studies investigated the effect of the main lipid-related features of Mediterranean diets, for instance diets low in saturated and polyunsaturated fat but relatively high in monounsaturated fat.123-125 Certain aspects of the Mediterranean diet not related to lipids (for instance, the amount of fibre) were not investigated, although they influence lipid metabolism. The consensus now is that a diet low in saturated and poly-unsaturated fat but rich in oleic acid results in a significant reduction of total and LDL cholesterol, and also has an effect on triglycerides and a small positive or no effect on HDL cholesterol.123-125 It is not certain whether these results can be completely reproduced in patients with established CHD, as none of these studies were conducted in such patients. Finally, as discussed in the different sections of this text, the Mediterranean diet was shown to strongly reduce the risk of CHD complications in secondary prevention19 60 and should be one of the preferred dietary patterns adopted by post-infarct patients. In the Lyon trial, the lipid-lowering effect of the Mediterranean diet was not different from that of the prudent Western diet followed by the control group, because lipid-lowering drugs were widely used in the two randomised groups. This nonetheless suggested that the Mediterranean diet was cardioprotective through biological effects independent of its effect on cholesterol. In particular, data from the Lyon trial suggested that the Mediterranean diet might prevent SCD (see above, section 3.2).
Blood pressure is also related to CHD mortality, and HBP is a common problem in many Western countries. In fact, the relationship between blood pressure and CHD is continuous and there is no abrupt increase in risk at levels of blood pressure regarded as high.126 This suggests that efforts to prevent blood pressure-related diseases should be focused both on hypertensive and on nonhypertensive persons. In secondary prevention, most patients are taking some blood pressure-lowering drugs (often beta-blockers) as systematic post-infarct treatment. In addition, traditional approaches to control the epidemic of blood pressure-related CHD have largely concentrated on drug therapy in persons with hypertension. However, because of the many side effects, the rate of discontinuation is high with these classes of drugs.127 Clearly, a non-drug therapy, including lifestyle modifications, may have an important and expanding role as a complement of drug therapy, especially in the long term. Another point is the importance of even small differences in blood pressure in terms of outcome. For instance, a 5mmHg reduction in diastolic blood pressure has been shown to result in a 35—40 per cent lower risk of stroke.128 Thus, a significant clinical benefit can be expected even from a small decrease in blood pressure resulting from a dietary change.
As regards the influence of diet on blood pressure-related CHD complications, data from the Seven Countries Study again provide major information.129 CHD mortality varies greatly among populations at each level of systolic or diastolic blood pressure in that study.129 At a diastolic blood pressure level of 90mmHg, CHD mortality was three times as low among Mediterraneans as among the populations from the USA and Northern Europe. On the same reasoning as for cholesterol, it is presumed that the protective factor is the diet of Mediterraneans. The same reasoning probably applies for the Asian (Japanese) diet. Another question is whether dietary factors influence blood pressure. High sodium intake and binge alcohol drinking certainly increase blood pressure.130 Whether the Mediterranean diet pattern may specifically influence (decrease?) blood pressure is unknown, although this dietary pattern has been reported to protect the arterial endothelium that is responsible for the production and release of several major vasodilators, including nitric oxide.131 In the (Mediterranean diet) Lyon trial, the extensive use of blood pressure-lowering drugs in both groups prevented any effect on blood pressure from becoming apparent. However, recent research has emphasised the powerful role of total diet in hypertension.130 An adequate intake of minerals (sodium, potassium, magnesium and calcium), rather than the sole restriction of sodium, was proposed as the focus of dietary recommendations.132 In this approach, however, the direct role of high sodium intake on the myocardium is not fully taken into account (see above the section about CHF and LVH). Other studies suggested that dietary n-3 fatty acids may lower blood pressure in subjects with hypertension.133 The responses were proportional to the changes in phospholipid n-3 fatty acids whereas n-6 fatty acids had no effect, which suggests that the effect did result specifically from the n-3 family. These data implied that in addition to their benefits through mechanisms such as the prevention of ventricular arrhythmias (see above, section 3.2), n-3 fatty acids may be helpful in modulating (endothelial) factors regulating blood pressure.
Another trial, the Dietary Approaches to Stop Hypertension (DASH), tested the effect on blood pressure of either a diet rich in fruits and vegetables or a 'combination' diet rich in fruits, vegetables and low-fat dairy products, and with reduced saturated and total fat.134 Although this 'combination' diet was not a typical Mediterranean diet, its main characteristics can be included among those recommended in a Mediterranean diet trial. In the first DASH trial, sodium intake was kept constant and the 'combination' diet decreased systolic blood pressure by 5-6mmHg in subjects with normal blood pressure; in those with mild hypertension, the blood pressure reduction was twice as great - about 12 mm Hg. Reductions of this magnitude are similar to those observed with antihypertensive medications, but they are obtained at a very much lower cost, particularly in terms of side effects. In fact, the first DASH trial confirmed the meta-analyses as well as earlier indications from observational studies, suggesting that dietary factors other than sodium markedly affect blood pressure.132 In a second trial, the DASH investigators studied the effects of different levels of dietary sodium in conjunction with the DASH diet.135 As before, the DASH diet substantially lowered systolic and diastolic blood pressure. In addition, at any level of sodium intake, blood pressure was lower among patients following the DASH diet than among those following the control diet.135 Thus, combining a reduction of sodium intake to levels below 100mmol per day and the DASH diet lowers blood pressure to a greater extent than either of the two separately. Whether these dietary changes may reduce the risk of CHD remains to be demonstrated.
3.6.3 Diet and 'endothelial dysfunction'
One mechanism that may contribute to the association between high blood pressure and CHD is called 'endothelial dysfunction'. The endothelium, the innermost layer of all blood vessels, is critical in determining the contractile state of the underlying smooth muscle.136 Through the release of a number of substances, the endothelium modulates several other functions, including platelet aggregation, leucocyte adhesion and migration, smooth muscle cell proliferation and lipid oxidation, all of which participate in the atherosclerotic process. The term 'endothelial dysfunction' has been used to describe a constellation of abnormalities in these regulatory actions of the endothelium, and 'endothelial dysfunction' has been reported in conditions such as hypercholesterolaemia, HBP, diabetes and hyperhomocysteinaemia. For instance, in patients with HBP, there is an imbalance in the bioactivity of endothelial factors with proatherosclerotic (endothelin-1) and antiatherosclerotic (nitric oxide) actions137 that may explain why HBP is a risk factor for CHD, regardless of whether endothelial dysfunction is a cause or a consequence of HBP. Coronary endothelial dysfunction by itself was indeed shown to be of prognostic significance in patients with CHD.138 Endothelium-derived nitric oxide (NO) plays an important role in the regulation of tissue perfusion, and evidence is accumulating that NO-dependent vasodilatation and NO availability are impaired in the coronary arteries of patients with CHD or with CHD risk factors such as high blood cholesterol or homocysteine levels.
Interestingly, folic acid therapy, either as chronic oral supplementation or as acute intra-arterial administration of the active form of folic acid (5-methyltetrahydrofolate), restores the impaired endothelial function even in patients with CHD risk factors but normal serum folic acid and homocysteine levels.139"140 Also, a deficient NO-synthase cofactor, tetrahydrobiopterin (BH4), may be involved in blunted endothelium-dependent vasodilatation in humans.141 In case of BH4 deficiency, uncoupling of the L-arginine-NO pathway is observed, resulting in increased formation of oxygen radicals. Intra-arterial or intra-coronary infusion ofBH4 was shown to improve endothelial dysfunction in patients with various clinical manifestations.141,142 The key point here is that the active form of folic acid is involved in the endogenous regeneration of BH4, suggesting a major interaction between the arginine-NO-synthase pathway and folic acid. Thus an adequate amount of folates in the diet appears to be crucial to protect the endothelium (and for the prevention of 'endothelial dysfunction') and, in general, for the prevention and treatment of CHD far beyond a simple effect on homocysteine (see above, section 3.5).
Another cause of endothelial dysfunction in the context of traditional risk factors of CHD (including HBP, dyslipidaemias and insulin resistance) is an elevated level of ADMA (asymmetric dimethyl arginine), which also inhibits the production of NO.143 Whereas blood ADMA levels seem to be influenced by dietary fats, further studies are required to clarify the relations between the dietary factors (folates, antioxidants) involved in the regulation of NO-synthase and ADMA metabolism.
A diet rich in vegetables and fruits and the traditional Mediterranean diet provide large amounts of folates. Consumption of legumes (drier beans and peas, fresh peas, peanuts, peanut butter, lentils), which is another typical Mediterranean habit, is a major source of folates and has been shown to be associated with a reduced risk of CHD.144 Also, tree nuts, such as walnut and hazelnut (common ingredients of vegetarian and Mediterranean diets), can help supply folates. Most nuts are also rich in arginine, an amino acid serving as a substrate for the synthesis of NO.
Because of the importance of NO in cardiovascular diseases, there has been growing interest over the past 10 years in using arginine to prevent and treat cardiovascular diseases.145 Compelling evidence shows that enteral or parenteral administration of arginine reverses endothelial dysfunction associated with major CHD risk factors in a way very similar to that observed with folic acid and
BH4. Endothelial arginine is derived from the plasma, via intracellular synthesis with citrulline as a precursor, and from the net degradation of intracellular proteins. However, food is the ultimate source of arginine for the body. Dietary arginine intake (the main source being animal products for the Western population) by an adult has been estimated to be around 5 g/day.145 Because of arginase activity in the intestine and of the limited digestibility of protein-bound arginine, it is assumed that only 50 per cent of dietary arginine enters systemic circulation. The adequate daily arginine requirement is difficult to assess and probably varies in accordance with the presence or absence of CHD risk factors, which are often associated with the presence of endogenous NO-synthase inhibitors. It probably also varies with the amount of folates in the diet, since the active form of folic acid is essential for the regeneration of BH4.
The amount of arginine found in the typical Western diet appears at best barely sufficient to cover the daily requirements of a healthy individual. Furthermore, in the presence of CHD risk factors or established CHD, the intake of arginine would tend to decrease as patients turn away from animal foods which, though protein-rich, are considered 'unhealthy for the heart' in other respects. The exact requirements of these patients have yet to be determined but previous studies suggested that in order to reverse endothelial dysfunction, an amount in the order of 6-9 g above dietary supplies might be necessary. Other factors, such as impaired intestinal absorption, competition with other amino acids (in particular lysine) for cell transport, and the amount of folates in the diet, were not taken into account in these calculations. Whatever the clinical condition, nuts are a convenient natural source of arginine, not only because of the high concentration of arginine in most nuts but also because it is associated with high levels of folates and vitamin B6, the major cofactors involved in the catabolism or recycling of homocysteine. In a certain sense, we can say that nuts are 'natural' functional foods.
People with type 1 diabetes are rather rare among the patients concerned by the secondary prevention of CHD. Apart from the dietary prevention recommended for non-diabetic patients, there are no specific dietary issues beyond the usual diabetic diet to control blood glucose once the insulin treatment is correctly administered. In contrast, the number of people with type 2 diabetes or insulin resistance is increasing rapidly and cardiologists have to manage that specific problem in more and more post-infarct patients.
Type 2 diabetes mellitus is associated with a three- to fourfold increase in the incidence of CHD,146 and the risk of CHD death is as high in people with diabetes without CHD as in those without diabetes and with established CHD.147 The decline in CHD mortality in most Western populations has been mainly attributed to reduction of risk factors, owing to dietary changes in particular. The smaller decline in CHD mortality among people with diabetes, particularly women, may be due to less effective changes in risk factors for those people.
Apart from calorie restriction, the composition of the diet of people with type 2 diabetes remains controversial. The emphasis currently is on a diet low in saturated fatty acids. A reduction of the total fat intake is also suggested when weight loss is a primary issue. Thus, as most type 2 diabetics should lose weight, a low-fat diet is commonly prescribed. However, many physicians are still reluctant to recommend such a diet because they think that a diet high in monounsaturated fats improves metabolic control in these patients better than a low-fat, high carbohydrate diet, and should be preferred.148 On the basis of a meta-analysis, it is clear that diets high in monounsaturated fat improve the lipoprotein and blood glucose profiles and also lower blood pressure.148 This type of diet may also reduce the susceptibility of LDL particles to oxidation and thereby reduce their atherogenic potential; in addition, it does not induce weight gain, provided that energy intake is controlled.
Thus, in theory, diets low in saturated fatty acids but rich in monounsaturated fats (two of the main characteristics of the Mediterranean diet are advantageous for the prevention of CHD in people with diabetes. Curiously, no diabetic diet has ever been tested in this way for the prevention of CHD in diabetic patients.
An important message from the UK Prospective Diabetes Study and other recent trials is that in the prevention of CHD in people with type 2 diabetes, it is unwise to focus on single risk factors.149 All known risk factors should be tackled simultaneously, including hyperlipaemia and hypertension. Also, because of a high risk of SCD in people with diabetes, specific recommendations aimed at preventing SCD should be given (see above). Classical risk factors fail to explain the excess CHD rate in Indians as compared with Europeans, although the high prevalence of diabetes in India may play a part.150 When exploring the contribution of dietary fatty acids in Indian people with diabetes, large differences in phospholipids fatty acids were noted, with lower concentrations of n-3 fatty acids151 suggesting an explanation for their high CHD mortality.
Considering all of these observations, it seems that the optimal diabetic diet may be a low-calorie Mediterranean diet. Not only does this diet protect the heart, improve lipid profiles and reduce blood pressure, but certain components (n-3 fatty acids in association with vegetables and legumes) have also been shown to improve glucose tolerance and prevent the apparition of overt diabetes.152 These human data confirm animal research that showed the importance of n-3 fatty acids in the action of insulin in various experimental models.153 Thus, although further studies are required, in particular about the physical structure of foods to modulate glucose metabolism and insulin resistance,154 it is clear that people with diabetes should be instructed in the basic principles of the Mediterranean diet.
3.6.5 Diet, overweight and obesity
Another question is why has the incidence of type 2 diabetes increased so rapidly? Considerable epidemiological evidence points to excess caloric intake and physical inactivity as the major reasons. There is no room here to discuss the problem of obesity and overweight in people with established CHD. Obesity and overweight are obviously associated with a clustering of CHD risk factors and weight reduction results in favourable changes in the risk factor profile of most individuals. However, it is also obvious that weight reduction efforts have met with limited success in the general population and that the treatment of obesity is complex and difficult. There is no reason (and no published data) to believe that the situation is different among patients with CHD. In addition, there are controversies regarding the efficacy, benefits and consequences of high-carbohydrate or low-fat or very low-fat or high-protein diets, all of which have been proposed as the best way to reduce weight. Thus, despite the fact that having (or maintaining, or reaching) a healthy body weight has been claimed a major goal of the dietary prevention of CHD in primary prevention,155 we do not believe that this is necessarily true in secondary prevention as well. As discussed in the previous sections of this text (and summarised in the last section), there are other dietary priorities for which we have scientific evidence showing that people with CHD get immediate and major clinical benefits by adhering to them. Of course, if someone with CHD is obese and/or asks spontaneously for a slimming diet, that issue should be appropriately addressed. In line with the principle of 'clinical rather than surrogate efficacy', there is little scientific data on the topic 'slimming diet in CHD patients'. Indeed, no dietary (or pharmacological) trials have been conducted so far to test whether weight loss is associated with improved prognosis in patients with CHD. Finally, whether a high-protein or high-carbohydrate diet or any other diet, in addition to a low energy diet, is preferable is an open question.
Was this article helpful?