The medical records of 130 children (mean age, 10.63 years) with mild intermittent atopic asthma who were given methacholine and AMP challenge tests were reviewed retrospectively. All 130 children had histories of recurrent wheezing and cough, favorable responses to inhaled corticosteroids and/or inhaled bronchodilators, and had been diagnosed with asthma. This study was performed at the Childhood Asthma and Atopy Center of Asan Medical Center Children's Hospital from May 2009 to May 2011. The study protocol was approved by the Institutional Review Board of Asan Medical Center, and all parents and guardians provided written informed consent following a detailed explanation of the study.

Bronchial challenge testing with AMP followed by methacholine was performed in all recruited children. Methacholine was administered immediately after recovery of pulmonary function. None of the subjects showed symptoms of severe bronchial narrowing; thus, no bronchodilators were administered after the challenge tests. Subjects did not use any medications for asthma at least 2 months before the bronchial challenge test. A pulmonary function test with a bronchodilator was performed once after bronchial challenge testing using methacholine. Basal lung function, including forced vital capacity (FVC), FEV1, peak expiratory flow, and forced expiratory flow at the mid-portion of FVC were also measured. FVC and FEV1 were measured each time when there was a concentration change in the methacholine and AMP tests. Fresh solutions of methacholine and AMP were prepared in buffered saline solution at concentrations of 0.625, 1.25, 2.5, 5, 10, and 25 mg/mL for methacholine and 3.125, 6.25, 12.5, 50, 100, 200, and 400 mg/mL for AMP. The methacholine PC20 and AMP PC20 were calculated by interpolation between adjacent data points when the FEV1 decreased >20%. Censored values of 50 mg/mL for the methacholine PC20 and 800 mg/mL for AMP were given to children who did not have 20% declines in FEV1 after inhaling maximal concentrations of methacholine (25 mg/mL) or AMP (400 mg/mL). Children were considered to have BHR when the methacholine PC20 was <8 mg/mL12,13 and the AMP PC20 was <200 mg/mL.

Skin-prick testing was performed on the backs of children using standard methods. Commercial extracts of the following common allergens were used: mites (Dermatophagoides pteronyssinus and D. farina), molds (Alternaria, Aspergillus, Cladosporium, and Penicillium), pollens (grasses, trees, weed, ragweed, mugwort, oak, beech, nettle, willow, elm, pine, hop, elder, hazel, oats, lambs quarter, ash, alder, birch, timothy, and rye grass), foods (apple, beef, chicken, egg white, codfish, crab, lobster, milk, mushroom, oyster, peach, peanut, pork, mussel, shrimp, strawberry fruit, tomato, tuna, walnut, and wheat flour), dog and cat epithelia, and cockroach. Histamine and isotonic saline were used as positive and negative controls, respectively. A wheal diameter that was greater than the positive control was considered a positive response when the positive control was greater than 3 mm. Any child with a response to one or more allergens was considered atopic.

Total eosinophil counts and the percentage of blood eosinophils were measured with an automated blood analyzer. Serum total IgE was measured by a fluorescence enzyme immunoassay using the ImmunoCAP system (Phadia AB, Uppsala, Sweden).

The levels of IgE and eosinophil counts were log-transformed before analyses to normalize the distributions. The values or frequencies between the two groups were compared with a Student's t-test. Pearson's correlation test was used to evaluate the relationship between the PC20 of methacholine and that of AMP with values from the pulmonary function test (PFT) and inflammatory markers. A P value ≤0.05 was considered statistically significant. SPSS ver.18 was used for the analysis (Chicago, IL, USA).

In total, 130 children who were given skin-prick testing, methacholine challenge testing, and AMP challenge testing were enrolled. All children had positive reactions to at least one of the allergens in the skin-prick test and, thus, were considered to have atopy. Table 1 shows the baseline characteristics of the enrolled children. More boys (71.5%) were included than girls (28.5%), and the mean age of the participants was 10.63±3.50 years (range, 6-18 years). A total of 30% of the children had family histories of allergic disease, 70% had allergic rhinitis, and 23.8% had atopic dermatitis (Table 1). Approximately 86% (111/130) of patients were positive for the methacholine challenge, and 93.8% (114/122) of patients were positive for the AMP challenge (Table 3).

The methacholine PC20 and AMP PC20 values were weakly correlated with each other (r²=0.189, P<0.001, Figure). Furthermore, the PC20 values for methacholine and AMP were both correlated with the PFT parameters (Table 2). In particular, FEV1, FEV1/FVC%, and maximum mid-expiratory flow (MMEF) were significantly correlated with the PC20 of methacholine and AMP (methacholine: P=0.001, 0.011, and 0.001, respectively; AMP: P=0.008, 0.046, and 0.001, respectively).

Table 3 shows no differences in the levels of three inflammatory markers in the group with positive BHR to methacholine (PC20<8 mg/mL) or the group with negative BHR to methacholine (PC20≥8 mg/mL). In contrast, significant differences were observed between the group with positive BHR to AMP (PC20<200 mg/mL) and the group with negative BHR to AMP (PC20≥200 mg/mL) for log total eosinophil count, percentage of eosinophils, and log IgE (P=0.023,<0.001, and 0.020, respectively) (Table 3). The co-occurrence of allergic rhinitis or atopic dermatitis was not significantly correlated with the PC20 of methacholine or AMP (allergic rhinitis with methacholine PC20: P=0.905; with AMP PC20: P=0.909, atopic dermatitis with methacholine PC20: P=0.404; with AMP PC20: P=0.238 respectively, data not shown). The sex distribution did not make a significant difference in inflammatory markers between the groups with positive or negative BHR to AMP or methacholine. In contrast, body mass index (BMI) was significantly higher in the group with positive BHR to methacholine than that in the group with negative BHR to methacholine (P=0.027); however, BMI was not significantly different in the group with positive BHR to AMP and the group with negative BHR to AMP (P=0.62, Table 3).