• eosinophils;
  • asthma;
  • childhood;
  • atopy;
  • risk factors;
  • epidemiology

Background Although peripheral blood eosinophilia is associated with risk of asthma, the relation with atopy has not been established.

Objective To assess the relationship between eosinophils and chronic asthma in childhood, and to determine the factors associated with eosinophil levels over time.

Methods Percent eosinophils/300 white blood cell (WBC) count (‘eos’) was measured at 9 months, 6 years and 11 years in subjects participating in the prospective Tucson Children's Respiratory Study. Children were classified based on the number of measurements in which they had low (≤ 2%) or high (> 5%) eosinophils, as follows: (1) Persistently low eos (n = 130); (2) Low eos (intermittently low or consistently moderate, but never high, n = 317); (3) Intermittently high eos (n = 192); and (4) Persistently high eos (n = 17). Only children with ≥ 2 eos measurements were included in the analysis. Chronic asthma was defined as medical doctor (MD)-diagnosed asthma with reports of wheezing during the previous year, on ≥ 3 questionnaires completed between 2 and 13 years of age. Children with at least one positive skin prick test (SPT; ≥ 3 mm) at age 6 or 11 were considered ‘atopic’.

Results Chronic asthma was linearly related to longitudinally ascertained eosinophils (trend χ2P < 0.001) with prevalence ranging from 5.8% among children with persistently low eos to 37.5% among children with persistently high eos. This relation was independent of atopy. Parental history of asthma was associated with both chronic asthma (P < 0.001) and with longitudinal eosinophil status (P < 0.001). After adjusting for atopy and gender, there was a 70% increase in asthma risk with each increase in longitudinal eosinophil level. This stepwise increase was reduced to 48% when parental asthma was added to the model.

Conclusion Longitudinal eosinophil levels are linearly associated with chronic asthma in childhood, independent of atopy. The strong association between parental asthma and eosinophil status suggests that genetic background may be an important determinant of eosinophilic response.