IADR Abstract Archives
Can Oral Microbiome Predict Low-birth Weight Infant Delivery?
Objectives: Pathogenic oral microbiomes were recognized to be associated with adverse pregnancy outcomes including low birth weight (LBW, infant weighing < 2500g). The study aimed to identify underlying oral microbiota factors contributing to the LBW and develop a prediction model using machine learning.
Methods: In a prospective cohort of 580 Chinese pregnant women, a nested case-control study was conducted with selected 23 LBW delivery cases and 23 healthy delivery control. Age and smoking habit of the women were matched. Saliva samples were collected at early- and late-pregnancy. Their demographic profiles (age, BMI, education level, and smoking habit), systemic health status, and periodontal status were recorded. Microbiome profiles of the saliva were analyzed through 16S rRNA gene sequencing.
Results: In the LBW case group, the relative abundance of Streptococcus was over-represented (median 0.259 vs. 0.116) and Saccharibacteria_TM7 was under-represented (median 0.033 vs. 0.068) significantly than in the controls (p<0.001, p=0.015 respectively). Ten species were identified as microbiome biomarkers of LBW by LEfSe analysis, which included 6 species within the genus of Streptococcus, three species of Leptotrichia buccalis, Gemella sanguinis and Granulicatella adiacens (all LDA score>3.5) as risk biomarkers, and one species of Saccharibacteria TM7 as a beneficial biomarker (LDA=-4.5). The machine-learning model based on these 10 distinguished oral microbiota species could predict LBW, with an accuracy of 82%, sensitivity of 91% and specificity of 73% (AUC-ROC score 0.89, 95% CI: 0.75-1.0). Results of α-diversity showed that mothers who delivered LBW infants had less stable salivary microbiota construction throughout pregnancy than the control group (measured by Shannon, p=0.048; and Pielou’s, p=0.021), however the microbiome diversity did not improve the prediction accuracy of LBW.
Conclusions: Several salivary microbiota biomarkers can assist in the prediction of adverse pregnancy outcome, including one protective predictor. The developed machine-learning model demonstrates good prediction accuracy.
Methods: In a prospective cohort of 580 Chinese pregnant women, a nested case-control study was conducted with selected 23 LBW delivery cases and 23 healthy delivery control. Age and smoking habit of the women were matched. Saliva samples were collected at early- and late-pregnancy. Their demographic profiles (age, BMI, education level, and smoking habit), systemic health status, and periodontal status were recorded. Microbiome profiles of the saliva were analyzed through 16S rRNA gene sequencing.
Results: In the LBW case group, the relative abundance of Streptococcus was over-represented (median 0.259 vs. 0.116) and Saccharibacteria_TM7 was under-represented (median 0.033 vs. 0.068) significantly than in the controls (p<0.001, p=0.015 respectively). Ten species were identified as microbiome biomarkers of LBW by LEfSe analysis, which included 6 species within the genus of Streptococcus, three species of Leptotrichia buccalis, Gemella sanguinis and Granulicatella adiacens (all LDA score>3.5) as risk biomarkers, and one species of Saccharibacteria TM7 as a beneficial biomarker (LDA=-4.5). The machine-learning model based on these 10 distinguished oral microbiota species could predict LBW, with an accuracy of 82%, sensitivity of 91% and specificity of 73% (AUC-ROC score 0.89, 95% CI: 0.75-1.0). Results of α-diversity showed that mothers who delivered LBW infants had less stable salivary microbiota construction throughout pregnancy than the control group (measured by Shannon, p=0.048; and Pielou’s, p=0.021), however the microbiome diversity did not improve the prediction accuracy of LBW.
Conclusions: Several salivary microbiota biomarkers can assist in the prediction of adverse pregnancy outcome, including one protective predictor. The developed machine-learning model demonstrates good prediction accuracy.