Risk factors and characteristics of respiratory and allergic phenotypes in early childhood.

BACKGROUND

Unsupervised approaches can be used to analyze complex respiratory and allergic disorders.

OBJECTIVE

We investigated the respiratory and allergic phenotypes of children followed in the Pollution and Asthma Risk: An Infant Study (PARIS) birth cohort.

METHODS

Information on respiratory and allergic disorders, medical visits, and medications was collected during medical examinations of children at 18 months of age; biomarker data were also collected (total and allergen-specific IgE levels and eosinophilia). Phenotypes were determined by using latent class analysis. Associated risk factors were determined based on answers to questionnaires about environmental exposures.

RESULTS

Apart from a reference group, which had a low prevalence of respiratory symptoms or allergies (n=1271 [69.4%]), 3 phenotypes were identified. On the basis of clinical signs of severity and use of health care resources, we identified a mild phenotype (n=306 [16.7%]) characterized by occasional mild wheeze and 2 severe phenotypes separated by atopic status. The atopic severe phenotype (n=59 [3.2%]) included 49 (83%) children with wheezing and was characterized by a high prevalence of atopy (61% with allergenic sensitization) and atopic dermatitis (78%). In contrast, atopy was rare among children with the nonatopic severe phenotype (n=195 [11%]); this group included 88% of the children with recurrent wheezing. Risk factors for respiratory disease included parental history of asthma, male sex, siblings, day care attendance, exposure to tobacco smoke or molds, indoor renovations, and being overweight, although these factors did not have similar affects on risk for all phenotypes.

CONCLUSION

Atopy should be taken into account when assessing the risk of severe exacerbations (that require hospital-based care) in wheezing infants; precautions should be taken against respiratory irritants and molds and to prevent children from becoming overweight.