Background: Previous epidemiological studies have shown acute effects of increased amounts of ambient air pollution on the prevalence of respiratory symptoms in children with respiratory disorders. We investigated whether children with bronchial hyperresponsiveness (BHR) and relatively high serum concentrations of total IgE (>60 kU/L, the median value) are susceptible to air pollution.
Methods: We collected data from children during three winters (1992-95) in rural and urban areas of the Netherlands. Lower respiratory symptoms (wheeze, attacks of wheezing, shortness of breath), upper respiratory symptoms (sore throat, runny or blocked nose), and peak expiratory flow were recorded daily for 3 months. The acute effects of airborne particulate matter with a diameter of less than 10 microm, black smoke, sulphur dioxide, and nitrogen dioxide were estimated by logistic regression.
Findings: 459 (73%) of 632 children had complete data. Of these, 26% had BHR and relatively high (above median) serum total IgE, 36% had no BHR and total IgE of 60 kU/L or less, 15% had BHR and total IgE of 60 kU/L or less, and 23% had a total IgE of more than 60 kU/L but no BHR. In children with BHR and relatively high serum total IgE the prevalence of lower respiratory symptoms increased significantly by between 32% and 139% for each 100 microm/m3 increase in particulate matter, and between 16% and 131% for each 40 microm/m3 increase in black smoke, SO2, or NO2. Decrease in peak expiratory flow of more than 10% in that group was more common with increased airborne particulate matter and black smoke. There were no consistent positive or negative associations between increased air pollution and prevalence of respiratory symptoms or decrease in peak expiratory flow in the other three groups of children.
Interpretation: Children with BHR and relatively high concentrations of serum total IgE are susceptible to air pollution. Although our odds ratios were rather low (range 1.16-2.39) the overall effect of air pollution on public health is likely to be substantial since these odds ratios refer to large numbers of people.