Bias and confounding in longitudinal measures of exhaled monoxides
BACKGROUND Fractional exhaled nitric oxide (FE(NO)) and exhaled carbon monoxide (ECO) have been proposed as markers of airway inflammation and oxidative stress. OBJECTIVE The aim of this study was to assess the effect of oral prednisone treatment on FE(NO) and ECO levels in a group of 30 asthmatic children with asthma exacerbation. METHODS Thirty asthmatic children with asthma exacerbation were treated with oral prednisone for 5 days (1 mg/kg/day). Before and after prednisone therapy, ECO was measured by means of a chemical analyzer and FE(NO) was measured by means of a chemiluminescence analyzer. ECO and FE(NO) were also measured in a group of healthy nonatopic children. RESULTS Before therapy, both ECO values and FE(NO) values were higher in asthmatic children (ECO, 3.2 +/- 0.2 ppm; FE(NO) online, 74.9 +/- 6.2 ppb; FE(NO) offline, 20.2 +/- 1.4 ppb) than in healthy controls (ECO, 2.0 +/- 0.2 ppm [P <.01]; FE(NO) online, 10.1 +/- 0.8 [P <.0001]; FE(NO) offline, 5.9 +/- 0.4 ppb [P <.0001]). An overlap in ECO values was found between healthy controls and asthmatic children. After prednisone therapy, there was a significant reduction in FE(NO) values (FE(NO) online, 40.6 +/- 4.6 ppb [P <.0001]; FE(NO) offline, 11.1 +/- 0.8 ppb [P < 0.0001]) and a slight but nonsignificant decrease in ECO values (2.7 +/- 0.2 ppm [P = not significant]) in the asthmatic group. No significant correlation between ECO values and FE(NO) values was found in either the asthmatic children or the controls. CONCLUSIONS After a course of prednisone therapy, in children with asthma exacerbation there is a significant decrease in FE(NO) but no significant change in ECO levels. This possibly suggests that ECO is less sensitive than FE(NO) to inhibition by corticosteroids.