Eosinophilia, interleukin-5, and tumour necrosis factor-alpha in asthmatic children

Background: There are few paediatric studies of the interrelationships between inflammatory markers and asthma severity. We therefore assessed the relationships between eosinophil-associated markers, cytokines, and asthma severity in asthmatic children aged 8-12 years. Methods: Forty-five children were tested twice, 2 weeks apart. Asthma severity was measured in terms of symptoms, lung function, medication needs, and histamine responsiveness. Peripheral inflammatory markers measured included eosinophil numbers, serum ECP, IL-5, and TNF-alpha and mononuclear cell IL-5, and TNF-alpha production. Results: Histamine responsiveness was correlated with circulating eosinophils (r = 0.56, P = 0.0001) and serum ECP (r = 0.54, P = 0.003). Eosinophilia was increased in children with severe as opposed to mild airway hyperresponsiveness (P = 0.02) and those who lost days at school as opposed to those who did not (P = 0.01). There were no other associations between markers of asthma severity and inflammation. Children taking inhaled corticosteroids had lower serum IL-5 levels than those on beta -agonists +/- cromolyn (mean and 95% CI: 20.5 [11.7-35.7] pg/ml vs 64.3 [26.6-155.4] pg/ml; P = 0.04). Cellular IL-5 production correlated with serum TNF-alpha (r = 0.63, P = 0.0062) and IL-5 (r = - 0.59, P = 0.005). Conclusions: Serum levels of TNF-alpha and IL-5 were not related to peripheral eosinophilia and asthma severity in these children but were related to their own cellular production ex vivo. This study confirms that eosinophilia is the index of inflammation that is most closely related to the clinical severity of childhood asthma.