Effect of Maternal Intestinal Floras on Neonatal Birth Weight

Abstract

Background Birth weight is closely related to individual health. Low birth weight is a high-risk factor for early neonatal death. Macrosomia is associated with higher risk of maternal and infant complications and various chronic diseases in adulthood. Therefore, it is very important to identify the influencing factors of neonatal birth weight. Objective To investigate the relationship between maternal intestinal flora and neonatal birth weight. Methods Participants were 516 singleton mothers and their babies〔24 with low birth weight (LW group) , 479 with normal birth weight (NW group) and 13 with macrosomia (OW group) 〕 delivered at term in Guangzhou Women and Children's Medical Center from January to September 2017. Maternal intestinal flora and clinical laboratory test parameters were collected, and the composition and diversity of intestinal flora were analyzed using QIIME. LEfSe analysis was used to compare the relative abundance of intestinal flora at the genus level of the mothers of three groups of babies to identify the flora with significant intergroup differences. MaAslin was used to assess associations of maternal laboratory test parameters and microbial genera. The Boruta was used to build models for predicting three types of neonatal birth weight using maternal laboratory test parameters and intestinal flora OTUs, to assess the association of maternal intestinal floras and neonatal birth weight. Results The analysis of maternal intestinal floras showed that the abundance of Firmicutes was the highest at the phylum level, and Faecalibacterium was significantly enriched at the genus level. There were no significant differences in Simpson's Diversity Index and Shannon Diversity Index at the phylum level across the maternal intestinal floras of three groups of babies (P>0.05) . LEfSe analysis found that compared with intestinal flora of mothers of LW group, the intestinal flora of mothers of NW group showed significantly enriched Streptococcus and Roseburia (P<0.05) , and significantly reduced abundance of Bacillaceae, Raphanus, Methanosphaera, Barnesiella and Paraprevotella (P<0.05) , while the intestinal flora of mothers of OW group demonstrated significantly enriched Closrtidiaceae and Alistipes as well as significantly reduced abundance of Barnesiella (P<0.05) . Compared with intestinal flora of mothers of NW group, the intestinal flora of mothers of OW group indicated significantly enriched Megamonas, Coprococcus, Veillonellaceae, cc-115, Closrtidiaceae and Alistipes, and significantly reduced abundance of Blautia and Eggerthella (P<0.05) . The area under ROC curve (AUC value) based on laboratory test parameter OTUs model and intestinal flora OTUs model was 0.62 and 0.77, respectively, in discriminating LW from NW, and was 0.65, and 0.78 respectively, in discriminating OW from NW. Conclusion Neonatal birth weight varied by the features of maternal intestinal floras. The OTUs model based on maternal intestinal flora could distinguish the neonatal birth weight. Maternal intestinal flora may be a good predictor of neonatal birth weight.

Keywords

• infant health|birth weight|gastrointestinal microbiome|lefse analysis|random forests