Measurement of leukotriene E 4 (LTE 4 ) in urine is a noninvasive method for assessing changes in the rate of total body cysteinyl leukotriene production. Eosinophil protein X (EPX) has been used to assess eosinophil activity and monitor inflammation in bronchial asthma. The aim of the study was to look for differences in urinary LTE 4 and EPX concentrations between children with stable atopic asthma and healthy controls and to compare asthmatic children with different disease severity. In addition the relationship was evaluated between urinary LTE 4 amd EPX levels and lung function.LTE 4 was also measured (enzyme immunoassay) together with EPX (radioimmunoassay) in urine and lung function tests were carried out in children with mild asthma (steroid-naive) (n=49), moderate to severe asthma (using inhaled steroids) (n=31) and healthy control subjects (n=28).Urinary leukotriene E 4 (LTE 4 ) was significantly higher in children with asthma than in controls (median [25±75 percentile] 238.5 (126.5±375.7) SD 191.8 versus 189 (51±253.2) SD 131.7 pg . mg -1 creatinine; p=0.021). Urinary EPX was also significantly increased in asthmatic children compared with controls (85.5 [64±131.5] SD 76.2 versus 48.5 [43.2±90] 112.1 mg . mmol -1 creatinine; p=0.006). There were no differences in urinary LTE 4 and EPX between the group of mild and the group of moderate to severe asthmatic children. There were significant associations between the urinary LTE 4 and intrathoracic gas volume (ITGV), residual volume (RV), forced expiratory volume in one second (FEV1), forced expiratory capacity (FVC) and maximum expiratory flow rate at 25% of vital capacity (MEF25).Urinary EPX was only correlated with maximum expiratory flow rate at 75% of vital capacity (MEF75). Thus measurement of urinary LTE 4 may predict the degree of airflow obstruction in asthmatic children. Urinary LTE 4 and EPX are useful markers of airway inflammation and can be helpful in guiding asthma management. There was no correlation between LTE 4 and EPX levels. Eur Respir J 2000; 16: 588±592.