To examine the relationship between use of oral contraceptives and the risk of dying from myocardial infarction, we made a comparison of deceased patients and live patients (women aged less than 45) identified for the Myocardial Infarction Causality case-control study, using data obtained from general practice medical notes. There were 422 live patients and 110 deceased patients of women with a myocardial infarction with data available. The adjusted odds ratio for exposure to second generation oral contraceptives and risk of death within 28 days of a myocardial infarction compared with no oral contraceptive use was raised (2.88, 95% confidence interval 1.22-6.77), and this effect was not seen for other types of oral contraceptives including third generation oral contraceptive formulations. In absolute terms, between 47,000 and 71,000 women would have to be exposed to a second generation pill for one year to result in one extra death from myocardial infarction, and this risk applies mainly to smokers. The results suggest a slightly increased relative risk of death among those having a myocardial infarction associated with exposure to second generation oral contraceptives, but this represents a small absolute risk. Further work is required before any change in contraceptive practice should be advocated.