CONCERN over Boris Yeltsin's health has focused attention on the links between heart disease and alcohol. Much more publicity has been given recently to the finding that two or three small drinks a day may reduce the risk of heart disease than to the warning that more than three drinks will raise blood pressure and increase the risk not only of heart disease but also of stroke and stomach and liver disorders.

The effect of alcohol on blood pressure is at first sight surprising. Alcohol causes flushing of the face and the assumption had been that it makes blood vessels open up, increasing blood flow and lowering the pressure. Over the past 20 years, however, evidence has accumulated that many men who drink heavily have high blood pressures and that the effect is proportional to the amount drunk. Swiss research reported in the New England Journal of Medicine has now provided an explanation for this. Three women and six men, aged around 27 and of normal weight, were given alcohol equivalent to three average drinks daily directly into the bloodstream. The investigators found that this stimulated the activity of the sympathetic nervous system - which deals with stress - and there was a substantial, sustained increase in the amount of nonadrenaline in the blood. The blood pressure rose slightly, and so did the heart rate.

Commenting on these results, the journal points out that the subjects were young and healthy and the experiment short term. Even so, the results provide a plausible explanation for the known association between regular drinking and raised blood pressure. The link seems to be direct, the effect seems to depend on the amount drunk, and the rise in blood pressure is reversible. When heavy drinkers - people who have six or more drinks every day - stop drinking completely their blood pressure falls within a few hours, but it rises again as soon as they resume drinking. The effects of very heavy and binge drinking are less clear, but there is evidence that someone who drinks a vast amount in one session risks an immediate stroke.

How does all this affect the belief that drinking in moderation is good for the heart? The reduction in the risk of coronary heart disease is thought to be due to an effect on the amounts of different types of cholesterol in the blood as well as possibly improving blood flow through the coronary arteries. Calculations have been done to show that any harm from the slight rise in blood pressure that may be expected from two drinks a day is outweighed by the protection given by the alcohol against heart attacks. The New England Journal concludes that in the absence of alcohol-related illnesses a drink or two a day still seems advisable from the standpoint of heart disease.

In Britain many experts remain cautious, mainly because they think two drinks a day is unrealistic: people who drink every day tend to take more than two small drinks. Doctors see so many patients with alcohol-related disease they are reluctant to encourage a belief that "safe" drinking is health-promoting.