The relationship between coffee consumption and acute myocardial infarction (AMI) was analyzed using data from a case-control study conducted in 1988 to 1989 within the framework of the GISSI-2 trial on streptokinase versus alteplase and heparin versus no heparin in the treatment of AMI. A total of 801 male patients with AMI and 792 control subjects who were hospitalized in several Italian regions for diseases unrelated to known or potential risk factors for cardiovascular diseases were included. Compared with coffee nondrinkers, the multivariate relative risks (RRs), after allowance for age, education, body mass index, smoking habits, alcohol consumption, family history of AMI, cholesterol level, history of diabetes, and hypertension, were 0.8 (95% confidence interval (CI), 0.5 to 1.2) for consumption of one cup/d, 1.3 (95% CI, 0.9 to 2.0) for two cups/d, 1.8 (95% CI, 1.1 to 2.7) for three cups, 2.5 (95% CI, 1.5 to 4.1) for four cups, and 2.6 (95% CI, 1.6 to 4.2) for five cups or more. The trend in risk with dose was statistically significant (P < 0.001). Duration of coffee consumption was not associated with the risk of AMI. The RRs for daily coffee consumption were elevated across strata of various covariates, including age, smoking habits, cholesterol level, diabetes, and hypertension, with a particularly elevated (although not significantly heterogeneous) estimate in patients younger than 50 years (RR, 5.7; 95% CI, 3.0 to 10.9 for four or more cups/d). The RR in patients who drank four or more cups of coffee per day and were current smokers was 8.1 (95% CI, 5.1 to 13.0), suggesting an unfavorable effect on the combination of cigarette smoking and high coffee intake on the risk of AMI.