Abstract
BACKGROUND: Although ambient air pollution has been linked to reduced lung function in healthy children, longitudinal analyses of pollution effects in asthmatic patients are lacking. OBJECTIVE: We sought to investigate pollution effects in a longitudinal asthma study and effect modification by controller medications. METHODS: We examined associations of lung function and methacholine responsiveness (PC20) with ozone, carbon monoxide (CO), nitrogen dioxide, and sulfur dioxide concentrations in 1003 asthmatic children participating in a 4-year clinical trial. We further investigated whether budesonide and nedocromil modified pollution effects. Daily pollutant concentrations were linked to ZIP/postal code of residence. Linear mixed models tested associations of within-subject pollutant concentrations with FEV1 and forced vital capacity (FVC) percent predicted, FEV1/FVC ratio, and PC20, adjusting for seasonality and confounders. RESULTS: Same-day and 1-week average CO concentrations were negatively associated with postbronchodilator percent predicted FEV1 (change per interquartile range, -0.33 [95% CI, -0.49 to -0.16] and -0.41 [95% CI, -0.62 to -0.21], respectively) and FVC (-0.19 [95% CI, -0.25 to -0.07] and -0.25 [95% CI, -0.43 to -0.07], respectively). Longer-term 4-month CO averages were negatively associated with prebronchodilator percent predicted FEV1 and FVC (-0.36 [95% CI, -0.62 to -0.10] and -0.21 [95% CI, -0.42 to -0.01], respectively). Four-month averaged CO and ozone concentrations were negatively associated with FEV1/FVC ratio (P < .05). Increased 4-month average nitrogen dioxide concentrations were associated with reduced postbronchodilator FEV1 and FVC percent predicted. Long-term exposures to sulfur dioxide were associated with reduced PC20 (percent change per interquartile range, -6% [95% CI, -11% to -1.5%]). Treatment augmented the negative short-term CO effect on PC20. CONCLUSIONS: Air pollution adversely influences lung function and PC20 in asthmatic children. Treatment with controller medications might not protect but rather worsens the effects of CO on PC20. This clinical trial design evaluates modification of pollution effects by treatment without confounding by indication.