ALCOHOL INTAKE AND THE RISK OF CARDIOVASCULAR-DISEASE IN MIDDLE-AGED JAPANESE MEN

Background and Purpose Understanding the effects of alcohol intake on stroke and other cardiovascular diseases is an important issue for pub lic health. Methods A 10.5-year prospective study of the relationship between alcohol intake and cardiovascular disease incidence was conduc ted in 2890 men, aged 40 to 69 years and free of a history of stroke a nd coronary heart disease, in three rural communities of Japan. Result s One hundred seventy-eight strokes (40 intracerebral hemorrhages, 18 subarachnoid hemorrhages, 104 nonhemorrhagic strokes, and 16 unclassif ied strokes), 34 coronary heart disease events, and 19 sudden unclassi fied deaths occurred. Drinkers of greater than or equal to 70 g/d etha nol had an approximately 2.5 times higher age-adjusted risk of all str oke than never-drinkers; the excess risk was more evident for hemorrha gic stroke than nonhemorrhagic stroke. When hypertension category, ser um total cholesterol level, cigarette smoking, and diabetes mellitus w ere taken into account, these excess risks were reduced but remained s ignificant for all stroke (2.0; 95% confidence interval, 1.3 to 3.1) a nd hemorrhagic stroke (3.4; 95% confidence interval, 1.2 to 9.2). A J- shaped relationship was suggested between alcohol intake and risk of n onhemorrhagic stroke; drinkers of <42 g/d ethanol had a slightly lower risk and heavy drinkers had a higher risk than never-drinkers. Curren t drinkers had a slightly lower risk of coronary heart disease than ne ver-drinkers, although the risk difference was not statistically signi ficant. The age-adjusted risk of sudden death was 10 times higher in h eavy drinkers than never-drinkers, and the excess risk did not change when the covariates were controlled for. Total cardiovascular disease showed a similar pattern as did all stroke. Conclusions Heavy drinking appeared to increase the risk of hemorrhagic stroke, in part due to h ypertension, and to increase the risk of sudden death, which was proba bly due to drinking per se. Light or moderate alcohol consumption seem ed to protect against nonhemorrhagic stroke and coronary heart disease .