BACKGROUND Approximately seventy per cent of asthmatic children from temperate climates, with normal lung function, have exercise-induced asthma (EIA). There is certain evidence to suggest that EIA may be less frequently encountered among children who live in tropical climates. Prevalence of EIA in Asian asthmatic children has never been thoroughly studied. OBJECTIVE To study the prevalence of EIA among Thai asthmatic children. METHODS A prospective study was performed to determine the prevalence of EIA in 44 Thai asthmatic children who were able to perform the spirometric maneuver. Subjects were randomly selected asthmatic children from the Pediatric Allergy Clinic, Department of Pediatrics, Siriraj Hospital. They were subjected to exercise testing on a steady state, motor-driven treadmill for 6 minutes (mean speed +/- SD = 3.7 +/- 0.4 km/h, mean level of inclination + SD = 15.0 +/- 5.2 degrees). The testing was conducted in a temperature-controlled (mean temperature +/- SD = 24.4 +/- 0.8 degrees C) and humidity-controlled environment (mean relative humidity +/- SD = 41.7 +/- 2.1%). Lung function tests were performed before exercise, immediately after and at 3, 5, 10, 15, 20 and 30 minutes after exercise. Results of the lung function test were calculated as per cent falls of forced expiratory volume in 1 sec (FEV1), peak expiratory flow rate (PEFR), and forced expiratory flow at 25 per cent-75 per cent (FEF50) from baselines. EIAs were diagnosed when drops of FEV1, PEFR and FEF50 were greater than 20 per cent, 25 per cent and 25 per cent from baseline values, respectively. RESULTS Of the 44 patients studied (31 boys and 13 girls; mean age 11.9 years), 34 per cent had mild asthma. Fifty-nine per cent had moderate asthma and 7 per cent had severe asthma. Eleven patients (25%) had EIA diagnosed by significant falls of FEV1's (26 +/- 12.6%), whereas, 13 patients (30%) and 20 patients (45%) had EIA defined by significant drops of PEFR's and FEF50's, respectively. A total of 23 patients (52%) had EIA by one or more diagnostic criteria. Peak times for EIA as diagnosed by FEV1, PEFR and FEF50 occurred at 3, 10, and 10 minutes respectively, after exercise. Most EIA episodes observed were of mild degree. CONCLUSIONS The prevalence of EIA in Thai children is much lower than figures reported in studies from Western countries. By using a significant fall of FEV1's as the diagnostic criteria, only 25 per cent were diagnosed as having EIA. By PEFR and FEF50 criteria, percentages of EIA increased to 30 per cent and 45 per cent respectively. Screening for EIA, therefore, may not be an appropriate diagnostic tool for the diagnosis of childhood asthma in tropical climates.