Background: Cough persisting after a respiratory infection is common in children and is often managed as asthma. However, little is known about the pathophysiologic mechanisms of such cough and how it compares with asthma. Objective: We used the technique of induced sputum to examine the inflammatory index values associated with persistent cough or allergic asthma in children. We hypothesized that the sputum from children with persistent postinfectious cough would differ from that of children with allergic asthma in that the former would lack eosinophils compared with the latter. Study design: Sputum production was induced with hypertonic saline solution in 34 children: 12 with cough persisting for 1 month or more after an apparent respiratory tract infection, not treated with corticosteroid; 11 with untreated atopic asthma, not using inhaled corticosteroid; and 11 with treated atopic asthma using inhaled corticosteroid. Results: The percentage of eosinophils in the sputum of children with cough was significantly lower than in the sputum of children with untreated allergic asthma (median 0.5% vs 14.5%, P < .0001). Similarly, the percentage of eosinophils in the sputum of children with asthma treated with inhaled steroids was significantly lower compared with untreated asthmatic children (1.5% vs 14.5%, P < .0001). The peripheral blood eosinophils, serum eosinophil cationic protein, and nasal percent eosinophils of the patients with cough were also significantly lower than those from patients with untreated asthma. Methacholine challenge in 6 of the 11 cough patients tested showed mild-to-moderate hyperresponsiveness, whereas the other 5 had a negative methacholine challenge. Conclusions: Children with persistent postinfectious cough do not have airway eosinophilia typical of untreated asthma. Despite the absence of eosinophilic inflammation, some of the patients with chronic cough had reactive airways. These results suggest that postinfectious cough in children has different pathophysiologic features than allergic asthma and probably represents a different disease. (J Allergy Clin Immunol 2000;105:495-9.)