Risk factors for developing atopic dermatitis.

The aim of this thesis was to investigate possible risk factors affecting the development of AD. AD is a frequent disease among children and has a substantial impact on the lives of both the child and its family. A better understanding of the disease would enable better treatment, prevention and information to the families involved. Previous risk factor studies have been hampered by an unsuitable study design and/or difficulties in standardization when diagnosing AD, which limit their conclusions. In paper I, we conducted a traditional cross-sectional analysis testing 40 possible risk factors for developing AD at 3 years of age. Our data suggested a strong heredity of AD and confirmed the risk associated with the non-functional FLG allele mutations after adjustments for confounders. Besides this mother's dermatitis and father's allergic rhinitis were found to increase the risk of AD. Perinatal exposure to dog was the only environmental exposure that significantly reduced the disease manifestation, suggesting other, yet unknown environmental factors affecting the increasing prevalence of AD in children. Length at birth was shown to be inversely associated with the risk of later developing AD. This traditional risk factor analysis led to two borderline significant results: duration of exclusive breastfeeding and mother's alcohol intake during the 3rd trimester. Since these possible two risk factors could neither be rejected nor accepted, we decided to do two in-depth studies, further investigating these, using longitudinal data information and data analysis instead of the traditional cross-sectional approach (paper II & III). In paper II, we investigated the risk of developing AD and wheezy symptoms until age 2 years depending on duration of breastfeeding. We found an increased risk of AD, but a protective effect on wheezy disorders in infancy from exclusive breastfeeding. The effect of exclusive breastfeeding on the risk of development of AD was significant after adjustment for demographics, FLG variants R501X and 2282del4 status, parent's AD and pets at home (RR 2.09, 95% CI 1.15-3.80, p=0.016). In addition, there was a significant effect of duration of exclusive breastfeeding (p=0.043), as the relative risk of AD was increased in proportion to increased duration of breastfeeding. The risk associated with exclusive breastfeeding was not explained by the fatty acid composition of mother's milk, though a trend showed higher risk of AD if mother's milk had low concentrations of n-3 fatty acids. In paper III, we found that alcohol intake during pregnancy was associated with a significantly higher risk of developing AD in the offspring, with the effect persisting throughout the whole 7 years follow-up period (HR 1.44, 95% CI 1.05-1.99, p=0.024). The increased risk was still significant after confounder adjustment for mother's education, AD and smoking habits during the 3rd trimester. There was no association between alcohol intake during pregnancy and other atopic endpoints (wheeze episodes, asthma, allergic rhinitis, blood eosinophil count, total IgE, sensitization, cord blood IgE and nasal eosinophilia). However, the underlying explanation was not clear. The thesis is based on data collected as part of the ongoing COPSAC cohort. The cohort is a longitudinal, prospective birth cohort following 411 children born to mothers with asthma. This selection of high-risk children restricts the interpretation of the results and they cannot necessarily be expanded to apply to the general population.