Results
Baseline characteristics
Sixty-two asthmatic children and adolescents completed the study. Seven did not attend on the second day because of transport difficulties and were excluded. All individuals were able to complete the six-minute EVH and there was no need to interrupt the 30-minute FEV1 evaluation because of respiratory symptoms.
No patient was excluded because of a low predicted FEV1level. Twelve were prescribed a low-dose (200mcg/day) continuous-use corticosteroid inhaler (beclomethasone, which is distributed free of charge by the government) but none were using it regularly and none had used it for at least one week before testing. No patient was using long-acting beta-2 agonists and all had albuterol for recovery. General data are presented in Table 1.
Mean laboratory temperature was 25.5oC (+ 2.8oC) and relative humidity 58.7% (+ 6.2%), with no differences between test days (p>0.05).
Reproducibility and agreement
Twenty-six of the 62 patients responded positively for EIB after EVH on both days (i.e., had a FEV1fallmax%> 10%), 17 on one day only (5 on the first day and 12 on the second), and 19 responded negatively on both days. The overall agreement was 72.5% (95% CI 61.6%, 83.6%) and the positive and negative agreement proportions were 41.9% and 30.6% respectively.
No difference was observed in mean FEV1fallmax% after EVH between visits for the group as a whole, with low bias (mean difference between visits for FEV1fallmax%) but with wide LOAs. The same was found for the AUC0-30min (Table 2 and Figure 1). Seventeen out of 26 patients with a positive response on both test days (65.4%) experienced a decrease in FEV1 > 15%, and, in 10 of these, the reduction was > 20%. In this sub-group of individuals, bias was also small and LOAs were wide for both FEV1falmax% and AUC0-30min (Table 2). The mean FEV1fallmax% in those individuals testing positive for EIB on one visit only (divergent group) was 17.7%+ 13.5% on the positive day (either visit one or two) and was statistically different from that observed in those individuals testing positive for EIB on both days (p = 0.016 for visit 1 / p = 0.021 for visit 2).
For the group as a whole, the intra-class correlation coefficient (ICC) for the FEV1fallmax% between the two visits was 0.854 (95% CI 0.758, 0.912; p<0.001) and was 0.858 (95% CI 0.764, 0.915; p<0.001) for AUC0-30min.
There was a significant correlation between FEV1fallmax% and AUC0-30min for both visits (visit 1: r = 0.91, p < 0.001 and visit 2: r = 0.89, p < 0.001).
Confounding factors
Potential differences in baseline FEV1 (as a percentage of predicted) and achieved ventilation rate (as a percentage of the calculated target) between visits were considered confounding factors that could interfere with the FEV1 response after EVH. As shown in Table 3, no differences between visits were observed in these two parameters either between groups or within groups. A more detailed overview of the individual target ventilation rates achieved on both visits for each group is provided in Figure 2.
Baseline asthma control status measured using the ACT score was also considered a potentially confounding variable but showed no difference between groups (Table 3). There were no differences either between groups in terms of age and BMI (p=0.624 and p=.0957, respectively - Kruskal-Wallis) or sex (p=.0738 - Chi-square).
Severity of EIB and recovery
During visit one, FEV1 did not return spontaneously to baseline levels by the thirtieth minute in 15 of the individuals testing positive on both days (n=26), and, during visit two, in 14. In those testing positive on one day only, this occurred in 6/17. These individuals were given 400mcg inhaled albuterol and FEV1returned to baseline values (within 10%) in all of these. Among those testing positive on both days, the severity of FEV1fallmax% was graded as mild on Visits 1 and 2 in 16 and 13, respectively, as moderate in 8 and 11, and as severe in the same two individuals on both days. Fourteen of the individuals testing positive on one visit only had a mild FEV1fallmax% response and three had a moderate response. There was no need to interrupt the FEV1 measurements after EVH due to respiratory complaints or oxygen desaturation below 94%.