We aimed to describe induced sputum cell counts in healthy nonasthmatic children, and to compare these to children with controlled and uncontrolled asthma. Following clinical assessment and spirometry, ultrasonically nebulized hypertonic saline was used to induce sputum from children with asthma (n=50) and without asthma (n=72). Sputum was dispersed and cell counts performed to yield total and differential cell counts. Specific stains were used for eosinophil and mast cell counts. All of the children with asthma were receiving inhaled and/or oral corticosteroids. Current asthma control was assessed in terms of symptoms and lung function. Children were classified as controlled on inhaled corticosteroids (no current symptoms, normal lung function n=15), current symptomatic asthma (n=16) and asthma exacerbation (n=11). It was found that eosinophils comprised a median 0.3% (interquartile range (IQR): 0, 1.05) of cells in sputum from healthy children. Sputum eosinophils (4.3% (IQR: 15, 14.1) p=0.0005) and epithelial cells (14% (IQR: 6, 19.4) p=0.0005) were significantly higher in children with asthma than in nonasthmatic children. Children whose asthma was controlled, as well as those with symptoms, had more sputum eosinophils and epithelial cells than the nonasthmatics. Mast cells were found in the sputum of only four of the 42 children with asthma. This study demonstrates that eosinophilic airway inflammation and epithelial damage can occur in children with asthma. Airway inflammation persists even in those children who are receiving inhaled corticosteroids, have normal lung function and good symptomatic control of their disease.