There are considerable data from epidemiologic, clinical, and experimental studies which indicate that the consumption of wine and other alcoholic beverages lowers the risk of atherosclerosis and coronary heart disease (CHD) quite markedly. Some of the proposed mechanisms for such protection include improving the blood lipid profile, especially increasing HDL-cholesterol (the "good cholesterol"), decreasing thrombosis or clot formation within arteries, and perhaps improving endothelial function of arteries, reducing the risk of diabetes, and acting through many other mechanisms.

Consumption patterns (alcohol consumed with meals or on an empty stomach, amount consumed at one time, daily consumption versus binge-drinking, etc.) may not be the same for different beverages and may explain some of the purported advantages of wine.

Interactions of Alcohol with other Dietary Factors

When considering the role of both alcohol and diet in the prevention of CHD, it should be noted that there are data indicating that many of the antioxidants and other phenolic compounds in foods are better absorbed or become more biologically active in the presence of alcohol. Of particular interest is the recent report by Rimm et al from the Nurses Health Study that while higher levels of folate were associated with less coronary heart disease, the protection was many times greater among drinkers than among abstainers. In this study, going from subjects in the first (lowest) quintile to those in the highest quintile of folate intake, the risk of CHD was reduced by only about 15% among non-drinkers but by almost 80% among women consuming 15 g/day or more of alcohol.

In the Lyon Diet Heart Study, wine consumption predicted blood levels of vitamin E better than vitamin E intake. These and other studies suggest that moderate alcohol consumption may enhance the effects of vitamins and other components of a healthy diet.

The Traditional Mediterranean-Type Diet and CHD

While wine is known to be a common component of the so-called typical, Mediterranean-type diet, it is unclear the extent to which the lower CHD rates in Mediterranean countries are the result of the wine consumed there and the extent that other aspects of the diet protect against CHD. Even after many decades of research, there is no consensus regarding the etiological roles of diet in CHD or the best diet for its prevention. Some tenets are widely accepted, however. One is that diets high in saturated fat are associated with higher serum cholesterol, and in turn, higher CHD mortality. This has been confirmed not only in comparisons between populations, such as the Seven Countries Study and the Ni-Hon-San study, but also within populations.

An early finding in cross-cultural comparisons was that populations with a relatively high ratio of polyunsaturated to saturated fatty acids in the diet seemed to be at reduced risk of CHD. These observations led to several trials of diet in either the primary or secondary prevention of CHD in which the experimental group increased the P/S ratio from the usual 0.2-0.4 to greater than 1.0, sometimes greater than 2.0. These trials, in general, showed results that were unimpressive in terms of reductions of CHD incidence or, especially, mortality.

Recent scientific reports have suggested that there may be advantages to the traditional "Mediterranean-type diet" similar to the diet of subjects in Crete who participated in the Seven Countries Study. This diet emphasizes the consumption of more grains, fruits, and vegetables, and uses olive oil, containing high levels of monounsaturated fats and antioxidants, as the primary source of fat. Meat and dairy products are markedly reduced, but total fat intake is not specifically restricted. It is hypothesized that this diet reduces CHD through its effects on thrombogenesis, effects on ventricular irritability, and possibly through actions on vascular endothelial function, inflammation, and/or lipid oxidation. Further, while long-term compliance is a recognized major problem with low-fat diets, there are suggestions that a Mediterranean-type diet is much easier for individuals to adhere to over time.

A key feature of the Mediterranean-type Diet is the use of olive oil, high in monounsaturated and low in polyunsaturated fats, as the primary source of fat in the diet. A fifteen-year follow up from the Seven Countries Study showed that all-cause death rates were negatively associated with the ratio of monounsaturated to saturated fatty acids, that CHD death rates were low in cohorts with olive oil as the main fat, and that by far the lowest overall death rates (one half of US rate) and CHD death rates (5% of US rate!) were in the men from Crete. The diet in Crete at the time consisted of moderate amounts of fish and alcohol, relatively large amounts of bread, legumes, and fruit, and minimal dairy fat. It was characterized by a moderately high total fat intake (36.1% of total calories) but low in saturated fat (7.7%).The men in Crete had serum cholesterol levels and smoking rates comparable to rural populations in Western countries with much higher CHD rates.

The only testing of a Mediterranean-type diet in a clinical trial was the Lyon Diet Heart Study, a secondary dietary intervention trial initiated in 1988 by de Lorgeril and Renaud of the INSERM research unit in Lyon. The Lyon study recruited men and women who were less than age 70, were able to complete an exercise test, and were clinically stable after having had an initial myocardial infarction (MI) within the preceding 6 months. That trial was ended prematurely because of the striking differences in rates of cardiac death, non-fatal MI, and total mortality being seen between the two groups. The Cox proportional-hazards risk ratios for all of these rates were between 0.24 and 0.30 for subjects on the Mediterranean diet in comparison with subjects in the control group, who had not been intervened upon in the trial but had been advised by their cardiologists to consume what is essentially the NCEP Step-One diet.

One reason for the apparent success of the Mediterranean-type diet in the Lyon Study may relate to the fact that the diet was very palatable and the recommended changes were readily assimilated into the lifestyle of the participants, leading to a high level of continuing compliance.

Is the Traditional Mediterranean-Type Diet Appropriate for Americans?

Over the past three years, my associates and I have been involved in pilot work developing and testing an American version of the Mediterranean-type diet (the "Med-Diet"). The levels of fat and olive oil in the diet developed proved to be acceptable to pilot subjects. With few exceptions, subjects had no difficulty in finding the foods being recommended.

We initially tested the effects of the Med-Diet on macro- and micro-nutrient intake in nine healthy volunteers (whose usual diet was low in fat and cholesterol) who faithfully adopted the Med-Diet for one week. When on the Med-Diet, there was a 11% decrease in the intake of animal protein and a 11% increase in vegetable protein; carbohydrates increased by 7%, soluble fiber by 45%, insoluble fiber by 27%, and total fiber by 32%. When on the Med-Diet, total fat increased 10%, with a 23% reduction in saturated fat intake, a 7% reduction in polyunsaturated fat, and a 64% increase in monounsaturated fat intake. We also tested the adequacy of the Med-Diet for essential nutrients, in terms of meeting the recommended dietary allowances (RDA) specified in the United States. When the subjects switched from their usual diets to the Med-Diet, they increased their intakes of calcium, iron, magnesium, zinc, and vitamins A, E, C, D, riboflavin, folic acid, and B12.

More recently, we have tested the acceptability of the Med-Diet among subjects with CHD. Dietary analyses based on 12 subjects with CHD randomly assigned to the two diets are available; 6 subjects were tested in each group. Based on 3-day food records at baseline and two months after being on the diets, total fat intake decreased from 21.0 to 19.5% of calories for the Low Fat-Diet and increased markedly (from 21.0 to 31.5%) for the Med-Diet, with most of the latter coming from an increase in monounsaturated fat from 11.4 to 18.4% of calories. Carbohydrates changed from 58.0 to 62.0% for the Low Fat-Diet and from 62.0 to 53.0% for the Med-Diet. The changes in vitamin and fiber intake were extremely variable, providing no consistent pattern in this small initial group of subjects.

While much of this work has been predominantly among Caucasians, we have addressed specific issues of African-American and Hispanic- American populations, both of whom have different pre-intervention diets and cultural food practices

A total of 17 subjects have been on the diets for at least two months and have completed a questionnaire reflecting their experience with the diets. We present the results of the overall assessments among these subjects, as well as for 14 subjects in whom only the Med-Diet was tested. We conclude that the studies thus far completed indicate that

(a) an American version of a traditional Mediterranean-type diet has been developed that meets all nutrient guidelines for essential nutrients;

(b) the costs are comparable with costs associated with following a low-fat, low-cholesterol diet;

(c) the diet is found to be highly acceptable to subjects with CHD and they state that they would be willing to adopt such a diet for the long-term prevention of further cardiovascular events.

The role of such a diet in the primary or secondary prevention of CHD has yet to be tested among Americans. Further, it is not known the extent to which the addition of red wine to such a diet will improve its ability to lower the risk of CHD.

Professor R. Curtis Ellison MD is Chief of the Evans Section of Medicine and Epidemiology and professor of Medicine and Public Health at the Institute on Lifestyle and Health, Boston University School of Medicine