Indoor microbiome, air pollutants and asthma, rhinitis and eczema in preschool children - A repeated cross-sectional study
Yu Sun,Yi Meng,Zheyuan Ou,Yanling Li,Mei Zhang,Yang Chen,Zefei Zhang,Xingyi Chen,Peiqiang Mu,Dan Norbäck,Zhuohui Zhao,Xin Zhang,Xi Fu
DOI: https://doi.org/10.1016/j.envint.2022.107137
Abstract:Background: Indoor microbiome exposure is associated with asthma, rhinitis and eczema. However, no studies report the interactions between environmental characteristics, indoor microbiome and health effects in a repeated cross-sectional framework. Methods: 1,279 and 1,121 preschool children in an industrial city (Taiyuan) of China were assessed for asthma, rhinitis and eczema symptoms in 2012 and 2019 by self-administered questionnaires, respectively. Bacteria and fungi in classroom vacuum dust were characterized by culture-independent amplicon sequencing. Multi-level logistic/linear regression was performed in two cross-sectional and two combined models to assess the associations. Results: The number of observed species in bacterial and fungal communities in classrooms increased significantly from 2012 to 2019, and the compositions of the microbial communities were drastically changed (p < 0.001). The temporal microbiome variation was significantly larger than the spatial variation within the city (p < 0.001). Annual average outdoor SO2 concentration decreased by 60.7%, whereas NO2 and PM10 concentrations increased by 63.3% and 40.0% from 2012 to 2019, which were both associated with indoor microbiome variation (PERMANOVA p < 0.001). The prevalence of asthma (2.0% to 3.3%, p = 0.06) and rhinitis (28.0% to 25.3%, p = 0.13) were not significantly changed, but the prevalence of eczema was increased (3.6% to 7.0%; p < 0.001). Aspergillus subversicolor, Collinsella and Cutibacterium were positively associated with asthma, rhinitis and eczema, respectively (p < 0.01). Prevotella, Lactobacillus iners and Dolosigranulum were protectively (negatively) associated with rhinitis (p < 0.01), consistent with previous studies in the human respiratory tract. NO2 and PM10 concentrations were negatively associated with rhinitis in a bivariate model, but a multivariate mediation analysis revealed that Prevotella fully mediated the health effects. Conclusions: This is the first study to report the interactions between environmental characteristics, indoor microbiome and health in a repeated cross-sectional framework. The mediating effects of indoor microorganisms suggest incorporating biological with chemical exposure for a comprehensive exposure assessment.