The authors studied the association between incidence of primary cardiac arrest and daily measures of fine particulate matter (</=2.5 micro m) using a case-crossover study of 1,206 Washington State out-of-hospital cardiac arrests (1985-1994) among persons with (n = 774) and without (n = 432) clinically recognized heart disease. The authors compared particulate matter levels on the day of the cardiac event and the 2 days preceding the event with levels from matched reference days. The estimated relative risk for a 13.8- micro g/m(3) increase in fine particulate matter (nephelometry: 0.54 x 10(-1) km(-1 )bsp) on the day prior to cardiac arrest was 0.94 (95% confidence interval: 0.88, 1.02). Pollutant levels measured on the same day as the event and on the 2 days preceding the event demonstrated similar results. No increased risk was found among all cases with preexisting cardiac disease (odds ratio = 0.97, 95% confidence interval: 0.89, 1.07); however, an unexpected association appeared between current smokers with preexisting heart disease and increased particulate matter levels 2 days prior to the event (odds ratio = 1.29, 95% confidence interval: 1.06, 1.55). This association was not present in the 0- or 1-day lag analyses or in persons with other diseases. There was no consistent association between increased levels of fine particulate matter and risk of primary cardiac arrest.