In order to study the determinants of respiratory system impedance and bronchodilator response in preschool children, a sample (n = 109) of healthy children (age 2.1-7.0 years) attending kindergarten was measured by using the impulse oscillometry. Their selection was based on a standardized questionnaire, negative skin prick test results and clinical examination, and sufficient cooperation. Triple measurements of respiratory resistance (Rrs) and reactance (Xrs) at 5, 10, 15 and 20 Hz, total respiratory impedance (Zrs), the resonance frequency (Fr) and the frequency dependence of resistance (dRrs/df) were performed, to determine individual mean values. Measurements were repeated after inhalation of 300 microg salbutamol (n = 89) or placebo (n = 19). At the baseline, Zrs and Rrs5-20 showed negative, and Xrs5-20, Fr and dRrs/df positive correlations with age, height and weight. However, logarithmic transformed height was the best independent variable for the regression equations of all the oscillometric variables. After inhalation of placebo, none of the oscillometric variables changed significantly. In the salbutamol group, the mean (SD) change in Rrs5 was -0.187 (0.124) kPa l(-1) s(-1) and -19.2 (10.2)%, corresponding to a lower reference limit of -36.9%. Both the within-test and between-test repeatabilities for the measurement of respiratory resistance were acceptable, for Rrs5 the coefficients of variation being 6.2 and 6.1%, respectively. As the overall success rate in our sample was high (89%), the forced oscillation technique seems to be a useful method in assessing respiratory function and bronchial lability in preschool children.