RESUMO
BACKGROUND: It remains unclear how much coronary risk factors contribute to the prediction of myocardial infarction among Japanese populations. METHODS AND RESULTS: A nested case-control study of Japanese male workers aged 35 to 65 years was conducted between 1997 and 2000 in the Morbidity of Myocardial Infarction Multicenter Study in Japan (3M Study). Two hundred four myocardial infarctions were identified and two controls per case were selected by matching for age. We calculated odds ratios using a conditional logistic regression model, and constructed risk predictive models for the risk of myocardial infarction using coronary risk factors. The multivariable odds ratios (95 percent confidence intervals) of myocardial infarction were 2.02 (1.29-3.16) for high blood pressure, 2.33 (1.51-3.59) for high LDL-cholesterol, 4.16 (2.36-7.33) for low HDL-cholesterol, 1.49 (0.94-2.35) for high triglycerides, 1.46 (0.89-2.39) for high glucose, and 2.95 (1.90-4.59) for current smoking. A large reduction of the predictive value for myocardial infarction was shown after exclusion of high LDL-cholesterol (reduction of predictive value was -3.4%), further exclusion of low HDL-cholesterol (-7.1%), and further exclusion of current smoking (-16.4%). CONCLUSIONS: High LDL-cholesterol and low HDL-cholesterol as well as current smoking had high predictive values for myocardial infarction among Japanese middle-aged male workers.