A group of 52 children with mild asthma were recruited from the allergy clinic at Seoul National University Children's Hospital. All subjects had a history of wheezing and dyspnea, and a methacholine PC₂₀ (provocative concentration causing a 20% fall in forced expiratory volume in one second [FEV₁]) of less than 16 mg/mL. The clinical severity of asthma was assessed according to the National Asthma Education and Prevention Program (8). The subjects whose clinical severity could not be accurately assessed were excluded. All patients lived in the Seoul Metropolitan area during the study period. Atopy was defined as the presence of at least one positive skin prick test result to a panel of 13 common aeroallergens, in the presence of positive and negative controls. The allergens tested were: 1) house dust mites (Dermatophagoides pteronyssinus and Dermatophagoides farinae), 2) animal danders (cat epithelium and dog epithelium), 3) pollens (mugwort, ryegrass, ragweed, hazel, alder, and oak), 4) molds (Aspergillus fumigatus and Alternaria alternata), and 5) cockroach (Blatella germanica). Parents provided written informed consent for their children to participate in the study.

At the beginning of the study, parents and children were instructed to complete diaries twice daily with regard to respiratory symptoms. The current day symptom scores reflected respiratory symptoms experienced that day (presence or absence) and the nighttime records associated with the preceding night's symptoms (filled out each morning). The number of as-needed short-acting bronchodilator use was also noted.

Participants were asked to record PEF each morning and each evening using a Mini-Wright adult type peak-flow meter (Clement Clarke Ltd., London, U.K.). After achieving documented reproducibility within 20 L/min, PEF recordings were taken twice daily for eight consecutive weeks (between March 8 and May 2). Morning values were obtained at 8:00 a.m. and evening values at 8:00 p.m. On each occasion, the best of three attempts was recorded. Subjects inhaled two doses of 200 µg salbutamol sulfate (Ventolin, Glaxo, U.K.) immediately afterwards, and all these measurements were repeated 15 min later. During the study, asthmatic patients were asked to refrain from taking any medication other than salbutamol sulfate, 200-400 µg, which was allowed as a rescue medication, but preferably not within 4 hr before the PEF measurements. At the time of the study, all patients were free of acute respiratory tract infections or asthma exacerbations for 4 weeks. All patients were asked to discontinue oral theophylline for 48 hr, long-acting β₂-agonists for 48 hr, and inhaled corticosteroids for 7 days before the start of the study. Diary cards with values missing for more than 10% of the eight weeks were excluded from the final analysis.

Methacholine bronchial challenges were carried out using a modification of the method described by Chai et al. (9). Concentrations (0.075, 0.15, 0.3, 0.625, 1.25, 2.5, 5, 10, and 25 mg/mL) of methacholine (Sigma-Aldrich, St Louis, MO, U.S.A.) were prepared by diluting with buffered saline (pH 7.4). A Rosenthal-French dosimeter (Laboratory for Applied Immunology, Baltimore, MD, U.S.A.) triggered by a solenoid valve set to remain open for 0.6 sec was used to deliver aerosol generated from a DeVilbiss 646 nebulizer with pressurized air at 20 psi. Each subject inhaled five inspiratory capacity breaths of buffered saline and increasing concentrations of methacholine until the FEV₁ fell by >20% of the post-saline value. FEV₁ was measured 1.5 min after each concentration was inhaled. The methacholine PC₂₀ was obtained from the log concentration-percent fall in FEV₁ curve by linear interpolation of the last two points. These tests were repeated within 1 month after the study period.

Daily information on Asian dust events during the study period was provided by the Korea Meteorological Administration from a station located in central Seoul. During the study period, five Asian dust days (March 10-11, March 30-31, and April 23) were identified in the Seoul Metropolitan area.

There were five Asian dust days during the length of the study, and the effects of dust events, if any, might have been delayed or manifested over a period of several days. To investigate potential effects between exposure and outcomes (time lapse), we examined a lag period of two days following each of the events (10, 11). The control days included all the days during the study period except the Asian dust days and the lag periods. Average levels of respiratory symptoms and pulmonary function parameters on the Asian dust days and during the lag period were compared with those of the control days. Daily symptom prevalence was calculated as the fraction (%) of children reporting symptoms or medication use, using data only from children with complete dairy information for each separate day. The proportion of decrements larger than 10% of morning PEF was calculated as the number of children experiencing a decrement divided by the total number of children reporting valid PEF measurements on each day of the study. The percentage of evening decrements was calculated in the same way. PEF diurnal variability, which was equal to the difference between morning and afternoon PEF values divided by the mean value, was calculated. Statistical comparisons of average daily reported respiratory symptoms and the PEF indices between each period were made with the Kruskall-Wallis non-parametric test. Correlation between daily average PM₁₀ concentration and PEF variability was examined using Pearson correlation tests. Methacholine PC₂₀ values obtained before and after the study period were log-transformed for statistical analysis, and the difference between the two was calculated by the paired t-test. All data analyses were performed using SPSS version 12.0 (SPSS Inc, Chicago, IL, U.S.A.). In each case, statistical significance was accepted when the two-sided p value was <0.05.

Of the 57 patients who were initially enrolled in this study, five were excluded because of unacceptable PEF records or having symptoms suggestive of respiratory infections during the study period. The characteristics of the 52 study subjects are shown in Table 1. The mean age of 32 boys and 20 girls was 10.5±3.2 yr. Atopy was identified in 45 (86.5%) subjects.

The daily prevalence of acute respiratory symptoms and signs was significantly higher during the Asian dust days than during the control days (p<0.05). More as-needed bronchodilator use was noted during the Asian dust days than during the lag period (p=0.016). Regarding the lag period, the daily prevalence of cough, sore throat, or eye irritation was significantly higher during the lag period than during the control days (p<0.05) (Table 2).

Table 3 summarizes the PEF parameters during each period. The mean morning PEF value (87.9±5.2%) and evening PEF value (90.4±5.4%) for % best during the Asian dust days were significantly (p<0.05) lower than those during the control days (morning, 94.2±3.2%; evening, 92.8±4.3%). The mean proportion of decrements larger than 10% of morning PEF was higher during the Asian dust days (57.3±16.0%) than during the control days (21.4±5.5%, p=0.015). The evening data showed similar results (40.7±18.3% vs. 21.0±4.3%, p=0.023). PEF diurnal variability during the Asian dust days was significantly higher than during the control days (10.1±3.5% vs. 5.5±2.2%, p=0.017), and it was augmented when post-bronchodilator values were included (15.8±6.8% vs. 10.9±7.2%, p=0.011). All PEF indices except the proportion of decrements of larger than 10% of evening PEF during the Asian dust days were significantly different from those during the lag period. There was a significant association between the degree of average PM₁₀ concentration and PEF diurnal variability during the study period (r=0.778, p=0.007) (Fig. 1).

Among the 52 subjects with a measurable PC₂₀ both before and after the study period, five subjects exhibited an increase in PC₂₀ and five subjects showed a decrease in PC₂₀ by a twofold magnitude (Fig. 2). The geometric mean of methacholine PC₂₀ after the study period (3.16 mg/mL) was not significantly different from that before the study period (2.82 mg/mL, p=0.290).