The relationship of developmental adaptations with asthma, allergic rhinitis/conjunctivitis, and atopic dermatitis: multivariable analysis
In the multivariable analysis, preterm birth was positively associated with asthma between 7 and 11 years (ORs, 1.29–1.36) (Figure 3A) (Table 2a). Meanwhile, overweight gain was positively associated with asthma at 9 and 10 years (ORs, 1.17–1.21) (Figure S3A). For allergic rhinitis/conjunctivitis, its positive association with preterm birth and overweight gain was only seen at 8 and 5.5 years, respectively (Figures 3B and S3B) (Table 2b). No significant associations between preterm birth and overweight gain were determined with atopic dermatitis (Figures 3C and S3C) (Table 2c). The results of the confounding factors are shown in Table 2.
Discussion
To the best of our knowledge, this is the first longitudinal study that has highlighted the similarities and differences in the age-specific associations of the three atopy-related diseases with early life factors using a large and a long-term cohort. The protective effect of daycare attendance during infancy in preventing the development of childhood asthma attenuated after 10 years old, while the protective effect of exposure to infectious agents in preventing the development of allergic rhinitis/conjunctivitis remained throughout childhood. This effect was not observed in atopic dermatitis. In contrast, a history of severe airway infection showed an increased risk for developing asthma throughout childhood. In addition, preterm birth was a significant risk factor for developing asthma between the ages of 7 and 11 years, with a small effect on allergic rhinitis/conjunctivitis and atopic dermatitis. Overweight gain in infancy did not significantly affect the development of atopy-related diseases.
Daycare attendance has prevented the development of asthma in selected age groups, and its protective effect remained throughout childhood in allergic rhinitis/conjunctivitis, whereas the opposite was seen in atopic dermatitis. Contact with other infants in daycare or siblings at home reflects frequent exposure to infectious agents12. In a study comparing the characteristics between Amish and Hutterites 13, the mechanism of protection against asthma has been determined to be related with innate immunity mediated by microbes in dust. The transient protective effect of daycare attendance in this study confirmed the result of a previous longitudinal study 5 that investigated the relationship between airway symptoms and early daycare attendance. On the contrary, early life daycare attendance increased the subsequent risk of developing atopic dermatitis 14. This inverse relationship could be related to the differences of barrier intensity and indigenous microbiota between the airway and the skin. A previous study reported that early life daycare attendance may cause non-atopic eczema with a different pathogenesis from that of atopic diseases15. This shows that a number of different mechanisms play a role in the development of allergic airway diseases and skin disease associated with daycare attendance. On the other hand, this current study did not indicate an association between older siblings and the prevention of developing of asthma in children, which is in line with a report by Nicolaou et al 16.
The effect of an episode of severe airway infection on the subsequent development of asthma and allergic rhinitis/conjunctivitis was harmful. Some patients may have early colonization of pathogenic bacteria in the airways, which may then predispose them to bronchiolitis with symptoms severe enough to warrant hospitalization resulting in an increased susceptibility to developing childhood asthma. This was opposite to the protective effect of exposure to microbes when attending daycare. The mechanisms of the inverse association of severe airway infection and daycare with allergic airway diseases remain unknown. However, a highly variable microbial composition in early life 17 that is induced by commensal and pathogenic bacteria and environmental microbes could be the difference on whether the immune system becomes susceptible or resistant to allergic airway diseases. The airway18 and gut microbiomes 19, 20 were also suggested to be one of the features which may potentially alter the phenotype of immune responses. In addition, Illi et al. suggested that the number and sites of infections early in life may determine the risk of subsequent asthma 21. A complicated network may be involved in the association between infection and allergic diseases.
This study confirmed the fact that babies born before 37 weeks of gestation are prone to suffer from childhood asthma 9. A meta-analysis of European children pointed out that preterm birth in infants was associated with asthma and that preterm birth mainly explains the association between a lower birth weight and asthma in childhood 9. It has been regarded that patients with immature lung growth in early life experience altered lung structure such as bronchial wall thickening, sub-pleural opacities and fibrosis, increased respiratory symptoms 22, and ongoing respiratory morbidity 23, eventually resulting in lung dysfunction and chronic lung diseases 24-27 in childhood and adulthood that exhibit hyperreactivity with chemokine and cytokine stimulation 28.
Moreover, our study indicated that overweight gain did not seem to be as relevant to the development of asthma in children, although the cause of obesity-related asthma in children was shown to occur in infancy in a European birth cohort study. This might be explained by the small number of obese children and adults in Japan 29.
Our study has some limitations. First, we did not measure the bronchial hyperresponsiveness to diagnose asthma 5. The prevalence of asthma in this study was similar to that of asthma in another survey in 2002 that assessed children with asthma across 11 prefectures in Japan 30. Second, we did not have a replication population. Instead, we assessed 3 atopic diseases with allergic manifestations using different age groups. Furthermore, we used a cohort study that covers a large population.
In conclusion, we have demonstrated markedly different patterns in the relationship of the potent factors in infancy and the development of atopy-related diseases in childhood. A dynamic interaction is supposed to be involved between the genetic attributes, organ maturation, environmental factors, barrier dysfunction, microbiome abnormalities, and immune system. Further cohort studies are required to determine whether preventing or delaying early life influences in growing years could modify atopy-related diseases progression.
Acknowledgments: The authors would like to thank participants and the Japanese Ministry of Health, Labour, and Welfare for their contributions.