Eosinophilic inflammation has a central role in the pathogenesis of asthma. We aimed to elucidate whether elevated blood eosinophil count (B-EOS), eosinophil cationic protein in serum (S-ECP) or in nasopharyngeal aspirate (NPA-ECP) predict later asthma after hospitalization for wheezing in infancy. In 1992-1993, 100 infants aged <24 months were hospitalized for wheezing associated with respiratory infection. B-EOS, S-ECP and NPA-ECP were measured on admission. Asthma status was evaluated at the follow-up visits at the median ages of 4.0, 7.2 and 12.3 yr. Twenty (25%) of 81 children had asthma at all three visits and were considered to have persistent childhood asthma (PCA). Children with B-EOS >or= 0.450 x 10(9) cells/l had a 2.9-fold PCA risk compared with other children. The risk was 6.1-fold when S-ECP was >or=20.0 microg/l and 6.7-fold when NPA-ECP was >or=815.0 ng/g. By these cut-off limits, all these markers were specific (75-93%), but not very sensitive (30-58%) in predicting PCA. At least one marker was elevated in 75% of the children with PCA. The respective figure for NPA-ECP alone was 58%. In adjusted analyses, only elevated NPA-ECP was an independent risk factor for PCA (OR 4.09). In conclusion, eosinophil activity in early life predicts the development of childhood asthma after hospitalization for wheezing in infancy. The results highlight NPA-ECP as an independent predictor of the persistence of asthma at school age.
(c) 2009 John Wiley & Sons A/S