Atopic dermatitis (AD) has a well-established association with skin colonization or infection by Staphylococcus aureus, which can exacerbate the disease. However, a causal relationship between specific changes in skin colonization during the first years of life and AD development still remains unclear. In this prospective birth cohort study, we aimed to characterize the association between skin colonization and AD development in 149 white infants with or without a family history of atopy. We assessed infants clinically and collected axillary and antecubital fossa skin swabs for culture-based analysis at birth and at seven time points over the first 2 years of life. We found that at age 3 months, S. aureus was more prevalent on the skin of infants who developed AD later on. S. aureus prevalence was increased on infants' skin at the time of AD onset and also 2 months before it, when compared with age-matched, unaffected infants. Furthermore, at AD onset, infants testing positive for S. aureus were younger than uncolonized subjects. In conclusion, our results suggest that specific changes in early-life skin colonization may actively contribute to clinical AD onset in infancy.

The influence of environmental factors on atopic dermatitis (AD) has been investigated in many cross-sectional studies. It remains however unclear if they could influence AD development early in life. This prospective birth cohort study aimed to monitor aspects of family lifestyle and child's nutrition within a Caucasian population and to assess its association with AD development over the first two years of life. Genetic predisposition was evaluated based on family history and profilaggrin genotyping. Of 149 included children, 36 developed AD.Infants with a family history of atopy developed AD 2.6 times more frequently (30 of 97) than infants without atopic predisposition (6 of 52). Genotyping was carried out on 50% of the children included. Profilaggrin mutations (R501X, 2282del4, R2447X and S3247X) were infrequent in our population. Lower incidence of AD was observed in infants exposed to a damp housing environment, lower household income and smoking mothers with a higher but not with a lower education level. Conclusion :Family history of atopy was a significant risk factor for AD regardless of the most common, currently defined, FLG mutations. Humidity at home and passive smoking seem associated with AD development in infancy.