Predictability of early atopy by cord blood-IgE and parental history

Background Atopic family history and cord blood IgE have been used as predictors of atopic disease in newborns for about 20 years, but at least for cord blood IgE the sensitivity has been shown to be very low. The objective of this paper was to evaluate whether parental history and cord blood-IgE were more accurate predictors for the appropriate atopic phenotypes in the infants rather than for any atopy. Methods A total of 1314 newborn infants was recruited in six German obstetric departments in 1990 and followed-up for 2 years. Four hundred and ninty-nine (38%) were at high risk for atopy with at least two first degree atopic family members and/or elevated cord-blood IgE concentrations. Results The cumulative incidence of atopic dermatitis over the first 2 years of life (AD24) amounted to 20.1% and there was a significant association with AD history of the mother (OR 2.5, 95%CI 1.46-4.26) and of the father (OR 3.53, 95%CI 1.90-6.54). The cumulative incidence of recurrent wheezing in the first 2 years of life (RW24) amounted to 16.1%, and was positively associated with asthma history (OR 2.11, 95%CI 1.33-3.60) and sensitization history (OR 1.64, 95%CI 1.34-2.36) of the mother, but with neither for the rather. RW24 was less prevalent in girls than in boys (OR 0.64, 95%CI 0.47-0.89). Thirty-one per cent of infants were sensitized (CAP test value > 0.35 kU/L) against at least one of nine food or inhalative allergens (S24) and this was significantly associated with cord blood-IgE value (OR 2.43, 95%CI 1.69-3.49), and sensitization history of the mother (OR 1.64, 95%CI 1.18-2.41). Using multiple logistic regression analysis, the prediction of AD24 by AD of parents, of RW24 by asthma of parents, and of sensitization by cord blood IgE was of low accuracy. Conclusion The predictive capacity of parental history and cord blood IgE is not high enough to recommend them as screening instruments for primary prevention. The majority of atopic manifestations and of sensitization occur in infants with no demonstrable risk at birth.