BACKGROUND Cohort studies have reported increased risks of cardiopulmonary mortality from long-term air pollution exposure, but the evidence is limited and inconclusive. We studied the association between long-term exposure to source-specific air pollution and myocardial infarction (MI) in a case-control study of first-time MI cases and population controls age 45 to 70 years in Stockholm county in 1992 to 1994. METHODS Home addresses during several decades were combined with historical emission databases and dispersion models to obtain annual mean levels of pollutants from traffic and heating during 30 years for 1397 cases and 1870 controls. Nitrogen dioxide (NO2), carbon monoxide (CO), and particulate matter with an aerodynamic diameter less than 10 microm (PM10) were used as indicators of traffic emissions and sulfur dioxide (SO2) as an indicator of emissions from residential heating. RESULTS There was no association between long-term average air pollution exposure and overall MI, but an increased risk of fatal MI was suggested, especially for out-of-hospital death. After adjustment for cardiovascular risk factors, the odds ratio for fatal MI associated with a 5th to 95th percentile difference in 30-year average exposure was 1.51 (95% confidence interval = 0.96-2.16) for NO2, 1.22 (0.98-1.52) for CO, 1.39 (0.94-2.07) for PM10, and 1.24 (0.77-2.02) for SO2. For out-of-hospital death, the odds ratio related to NO2 exposure was 2.17 (1.05-4.51). CONCLUSIONS This study provides some support for an association between long-term air pollution exposure and fatal cardiovascular disease.