Prediction of sensitization to inhalant allergens in childhood: evaluating family history, atopic dermatitis and sensitization to food allergens

Background A family history of atopy is a poor predictor of sensitization to inhalant allergens and allergic disease during childhood. We recently identified early sensitization to food allergens, especially hen's egg, as a valuable predictor of subsequent sensitization to inhalant allergens. Objective (1) Whether prediction will be improved by in vitro allergy tests at 1 year of age in combination with family history and medical history data. (2) Comparison with the capacities of in vitro tests to predict sensitization to aeroallergens. Methods Of an observational birth cohort study (MAS) 49 children who were sensitized to inhalant allergens at 5 years of age and 116 non-sensitized controls were included in the present study. For the prediction of sensitization to inhalant allergens the following prognostic factors were evaluated: atopic family history (FH), atopic dermatitis (AD) during the first year of life, two ir? vitro allergy tests for specific IEE to common food allergens at I year of age (fx5 [Pharmacia] and single allergen specific tests (sIgE) for four allergens) and 'high' total serum IgE, defined by three different cut off points. Results The combination of medical history data and laboratory tests resulted in the best predictive discrimination. The positive predictive values (PPV) were higher if sensitization to food was detected by single allergen specific tests (PPV: 66%/75%/100% corresponding to the three evaluated risk groups) than by the qualitative fx5 (PPV: 46%/65%/100%). The negative predictive values were equal for both tests (69 and 92% for the two low risk groups). High total serum IgE had low predictive capacity. Conclusion During infancy the prediction of sensitization to inhalant allergens should be based on medical history data and allergy tests determining sensitization to food allergens. The in vitro tests improve the predictive discrimination, but the individual risk profile of the child must be considered for a reliable and valid prediction.