Eczema susceptibility and composition of faecal microbiota at 4 weeks of age: a pilot study in Chinese infants

DEAR EDITOR, Early-life microbial exposure plays an important part in the development of the immune system, reducing the likelihood of allergy. Several studies have investigated alterations in the relative abundance of bacterial species in the gut microbiota. Escherichia coli has been reported to be more abundant in infants who develop atopic eczema than in healthy infants. Compared with controls, a reduced abundance of Bifidobacterium has also been reported in allergic patients. Reduced gut microbial diversity in early life is also important in modulating allergy risk. Over time, the methods used to detect bacteria have changed, from low-throughput bacterial cultures to terminal restriction fragment length polymorphism to denaturing gradient gel electrophoresis to more efficient 16S rRNA pyrosequencing. More recently, next-generation sequencing (NGS) and bioinformatics tools have facilitated the nontargeted sequencing and unbiased analysis of the entire gut microbiome. The faecal microbiome is likely to be ethnicityand diet-specific; however, there are limited metagenomics data available in Asians. This prospective pilot study made use of NGS to characterize the stool microbiome in Hong Kong Chinese neonates and to investigate faecal microbial factors associated with the subsequent development of eczema. In total, 149 Chinese term neonates with an uneventful perinatal course and born in our hospital in September 2012 were followed for 9 months. The occurrence of allergic diseases was ascertained by use of the Chinese International Study of Asthma and Allergy in Childhood questionnaire. A diagnosis of eczema was confirmed according to the criteria of Hanifin and Rajka. This study was approved by the clinical research ethics committee of the Chinese University of Hong Kong, and the parents of all the neonates gave informed written consent for their infants to participate. Fifteen infants with eczema and 15 nonallergic controls, all 9 months old, were identified, and their frozen random stool samples (~1 g each) collected at 4 weeks of age were retrieved for NGS with the Ion 318 Chip v2 by Ion PGM System [Life Technologies, CA, U.S.A. (Appendix S1)]. All participants were formula-fed term infants with a normal birth weight (≥ 2 5 kg). They must also not have received antibiotics during the neonatal period. The Shannon–Weaver diversity index was used to measure the microbial diversity of the stool samples. Briefly, this index was calculated by the formula Σlog(p)p, where p denotes the relative abundance of species. It incorporates the richness and evenness of species in the community, typically with a value between 1 5 and 3 5. Differences in this index and different bacterial taxa at the genus level between eczema and controls at 9 months were analysed by the Mann–Whitney U-test. Statistical significance was defined as P < 0.05. Table S1 describes infants whose stool samples were subjected to NGS. Owing to insufficient DNA, five controls were excluded; therefore, 15 cases and 10 controls were sequenced. All clinical features were matched between these two groups.