Air pollution and myocardial infarction. Strasbourg France, 1984-89

Background: Many studies have shown positive associations between urban air pollution, mortality and hospitalizations for cardiovascular diseases. Thi s study tried to estimate short term effects of ambient air pollution an my ocardial infarction on the basis of data collected in a morbidity registry. Methods: The daily number of myocardial infarctions between 1984 and 1989 w as supplied by the Monica registry (Bas-Rhin). The pollution variables were daily mean and maximum of hourly measures of carbon monoxyde (CO), sulfur dioxide (SO2), particles (PM13), nitrogen monoxyde (NO) and nitrogen dioxid e (NO2), mean and maximum of hourly measures of ozone (O-3) between 10 a.m. and 6 p.m. Other data were influenza epidemics, daily temperature and humi dity. The analysis was a Poisson regression controlling for trend, season, meteorological factors and pollutant, using nan parametric smoothing. Influ ence of day of week and influenza were controlled through dummy variables. Results: For the hourly maximum of NO with a 5 day lag the association was statistically significant (square root relation): for an increase from perc entile 25 (63 mug/m(3)) to percentile 75 (189 mug/m(3)), the RR was 1.087 ( 95% CI: 1.014-1.166). The association was significant during winter with a RR of 1.129 (95% CI: 1.028-1.241)for a percentile 25 (101 mug/m(3)) to perc entile 75 (265 mug/m(3)) increase. A positive linear association was found with daily maximum of NO2 during winter with a 5 day lag: the RR, for an in crease from percentile 25 (59 mug/m(3)) to percentile 75 (107 mug/m(3)) was 1.095 (95% CI: 1.015-1.181). For the daily mean and maximum of O-3, a posi tive association was found but it was not robust. For other pollutants, no association was found. Conclusions: The association between NO2 and coronary events, hospitalizati ons and mortality, has been shown in several studies but not in all. This s econdary pollutant could be a proxy for small particles.