International variations in bronchial responsiveness in children: findings from ISAAC phase two.

RATIONALE: Bronchial responsiveness is an objectively measurable trait related to asthma. Its prevalence and association with asthma symptoms among children in many countries are unknown. OBJECTIVES: To investigate international variations in bronchial responsiveness (BR) and their associations with asthma symptoms and atopic sensitization. METHODS: Bronchial challenge tests were conducted in 6,826 schoolchildren (aged 8-12 years) in 16 countries using hypertonic (4.5%) saline. FEV(1) was measured at baseline and after inhalation for 0.5, 1, 2, 4, and 8 min. BR was analyzed both as a dichotomous (bronchial hyperreactivity, BHR, at least 15% decline in FEV(1)) and as a continuous variable (time-response slope, BR slope, individual decline in FEV(1) per log(min)). RESULTS: Prevalence of wheeze last year ranged from 4.4% in Tirana (Albania) to 21.9% in Hawkes Bay (New Zealand) and of BHR from 2.1% in Tirana to 48% in Mumbai (India). The geometric mean BR slope varied between 3.4%/log(min) in Tirana and 12.8%/log(min) in Mumbai and Rome (Italy). At the individual level, BHR was positively associated with wheeze during the past 12 months both in affluent countries (OR = 3.6; 95% CI: 2.7-5.0) and non-affluent countries (OR = 3.0; 1.6-5.5). This association was more pronounced in atopic children. There was a correlation (rho = 0.64, P = 0.002) between center-specific mean BR slope and wheeze prevalence in atopic, but not in non-atopic children. CONCLUSIONS: BR to saline in children varied considerably between countries. High rates of BR were not confined to affluent countries nor to centers with high prevalences of asthma symptoms. The association between wheeze and BHR at the individual level differed across centers and this heterogeneity can be largely explained by effect modification by atopy. Pediatr. Pulmonol. 2010; 45:796-806. (c) 2010 Wiley-Liss, Inc