To make a further quantitative assessment of the relationship between cigarette smoking and the risk of myocardial infarction, a multicentric case-control study was conducted in Italy between September 1988 and June 1989 within the framework of the GISSI-2 trial. Ninety hospitals in various Italian Regions participated. 916 cases of acute myocardial infarction with no history of ischaemic heart disease and 1106 controls admitted to hospital for acute conditions not related to known or suspected risk factors for ischaemic heart disease were studied. Relative risks (RR) and 95% confidence intervals (CI) of myocardial infarction according to various measures of tobacco smoking, were adjusted for identified potential confounding factors using multiple logistic regression. Compared to lifelong non-smokers, the RR was 1.3 (95% CI 1.0 to 1.9) for ex-smokers, 2.0 (95% CI 1.4 to 2.9) for current smokers of less than 15 cigarettes per day, 3.1 (95% CI 2.2 to 4.2) for 15-24 cigarettes per day and 4.9 (95% CI 3.4 to 7.1) for 25 or more cigarettes per day. No trend in risk was evident for the duration, the RR being around 3 for subsequent categories. There was a significant interaction between smoking and age. Below the age 45, smokers of 25 or more cigarettes per day had a 33 times higher risk than non-smokers, compared to 7.5 at in the age group 45-54, 4.4 between the ages 55-64 and 2.5 at the age of 65 or over. The risk estimates were higher in women (RR for > or = 25 cigarettes per day = 10.1), in subjects in the lowest cholesterol tertile (RR = 11.9), with no history of diabetes (RR = 6.8), hypertension (RR = 9.5), no family history ischaemic heart disease (RR = 9.1) and low body mass index (RR = 9.3). The importance of smoking is confirmed as a cause of acute myocardial infarction: about 50% of all nonfatal infarctions in this Italian population could be attributable to cigarette smoking.(ABSTRACT TRUNCATED AT 250 WORDS)