Eczema in childhood and adolescence

Abstract

Background: Eczema (atopic dermatitis) is an itchy inflammatory skin disease that affects many children and adolescents. Eczema associated with IgE antibodies is called atopic eczema (AE), while other eczema is called nonatopic eczema (non-AE). Having a filaggrin (FLG) mutation is associated with increased risk for eczema. Children with eczema more often develop food allergy, asthma, and rhinitis. It has been suggested that eczema is associated with non-allergic comorbidities such as ADHD.

Aims: To describe eczema in adolescence and the course of preschool eczema up to age 16 years in a population-based birth cohort, to investigate consequences of preschool eczema and FLG mutation later in childhood/adolescence and to study factors of importance for remission of preschool eczema.

Methods: We used data from a Swedish birth cohort recruited from the general population (n=4,089). The families provided information on background factors at inclusion when children were 2 months of age. Questionnaires regarding eczema during the preceding year were answered by parents at 1, 2, 4, and 8 years, and by parents and children at 12 and 16 years. Information on IgE sensitization at 4, 8, and 16 years and on FLG mutation was available for most children. Record-linkage to the Swedish prescribed drug register provided data on dispensed medication at age 10-18 years (range 8.6-19.9).

Results: The 12-month prevalence of eczema in adolescence was 10%, and of those affected 73% had mild, 17% moderate, and 10% severe eczema. The most common locations of eczema in adolescence were flexural surfaces (73%), neck (40%) and extensor surfaces of extremities (39%). Among adolescents with eczema, onset was most common before age 2 years (49%), but onset after age 12 years was also common (26%). A history of eczema at every previous follow-up was seen among 11% of the adolescents with eczema. In adolescence, AE was more common, had earlier onset and was more severe compared with non-AE, but there were no differences in seasonal variation or location of eczema. There was no association between PSE and ADHD medication at school age (adjOR 1.12, 95% CI: 0.80-1.56). Further, PSE was not associated with dispensed antidepressants, migraine medication or antiepileptics at school age. PSE was associated with IgE sensitization to both food and aeroallergens up to age 16 years (overall adjOR 2.30, 95% CI: 2.00-2.66). This association was significantly stronger among children with persistent PSE. Among sensitized children at age 4, 8, and 16 years, polysensitization was more common in children with PSE than those without PSE. FLG mutation was not associated with IgE sensitization up to age 16 years, with the exception of IgE sensitization to peanut at age 4 years (adjOR 1.88, 95% CI: 1.03-3.44). Half (51%) of children with PSE were in complete remission of eczema at school age. In prognostic multivariate models male sex and exclusive breastfeeding ≥4 months were positively associated with complete remission of PSE at school age. Persistent PSE, severe PSE, parental allergy, parental smoking, and FLG mutation were associated with reduced likelihood of complete remission. The prognostic models developed had correct classification rates of 63-65%.

Conclusions: Only half of children with eczema before age 4 years are in complete remission at school age. Of adolescents with eczema, more than 10% have severe eczema. In adolescence, AE and non-AE do not differ in seasonal variation or location. No significant association was found between eczema and dispensed ADHD medication, antidepressants, migraine drugs or antiepileptics. PSE, especially if persistent, is strongly associated with IgE sensitization to both foods and aeroallergens. FLG mutation is associated with IgE sensitization to peanut, but not to other allergens, indicating that FLG mutation in the absence of PSE does not play a major role in IgE sensitization. Finally, male sex and exclusive breastfeeding ≥4 months were positively associated with complete remission of PSE at school age, whereas persistent and/or severe PSE, parental allergy, parental smoking, and FLG mutation were associated with reduced likelihood of complete remission.