Association between vaginal microbiota in early pregnancy and gestational diabetes mellitus: a nested case-control study

BACKGROUND: To investigate the association between vaginal microbiota structure in early pregnancy and gestational diabetes mellitus (GDM) and to characterize microbial signatures for early screening for GDM. METHODS: The present study was a nested case-control study recruiting pregnant women from the Nanjing Gulou Maternal-Child Health Center, China. Vaginal swabs were collected before 20 weeks of gestation for 16S rRNA sequencing. Following 1:3 propensity score matching, 45 GDM cases and 135 controls were enrolled. The final analysis included 42 GDM cases and 121 controls. A random forest model was used to explore the genera of vaginal differential microbiota associated with GDM. Based on these findings, latent profile analysis (LPA) was conducted to explore potential types of vaginal microbiota, and logistic regression was used to analyze the association between vaginal microbiota types and GDM. RESULTS: The GDM group exhibited elevated alpha diversity (Chao1 index, P = 0.036), and the abundance of Aeromonas in early pregnancy was associated with the risk of GDM (aOR = 1.17, 95% CI: 1.01–1.29). LPA classified vaginal microbiota into five profiles: Type Ⅰ (high Aeromonas with low Lactobacillus), Type Ⅱ (Lactobacillus crispatus-dominant), Type Ⅲ (co-dominance of Lactobacillus iners and Lactobacillus jensenii), Type Ⅳ (co-dominance of Aeromonas and Lactobacillus crispatus), and Type Ⅴ (Lactobacillus iners-dominant). Compared with type II, type Ⅰ (aOR = 2.83, 95% CI: 1.25–6.40) or type IV (aOR = 3.02, 95% CI: 1.27–7.15) exhibited an elevated risk of GDM. Compared with Aeromonas abundance prediction, vaginal microbiota type-based prediction demonstrated superior diagnostic accuracy (Aeromonas-AUC = 0.629, 95% CI: 0.541–0.718; vaginal microbiota profiles-AUC = 0.775, 95% CI: 0.716–0.835). CONCLUSION: The composition and structure of vaginal microbiota in early pregnancy are different in the two groups. The vaginal microbiota in early pregnancy, which is characterized by co-dominated by Aeromonas and Lactobacillus crispatus, was associated with a higher risk of GDM. Vaginal microbiota type-based prediction demonstrated superior diagnostic accuracy than single bacteria.