Vaginal microbiome as a tool for prediction of chorioamnionitis in preterm labor: a pilot study

Abstract

Abstract Intra-amniotic infection (IAI) is a major cause of preterm birth with a poor perinatal prognosis. We aimed to determine whether analyzing vaginal microbiota can evaluate the risk of chorioamnionitis (CAM) in preterm labor cases. Vaginal discharge samples were collected from 83 pregnant women admitted for preterm labor. Based on Blanc’s classification, the participants were divided into CAM (stage ≥ II; n = 46) and non-CAM (stage ≤ I; n = 37) groups. The 16S rDNA amplicons (V1–V2) from vaginal samples were sequenced and analyzed. Using a random forest algorithm, the bacterial species associated with CAM were identified, and a predictive CAM (PCAM) scoring method was developed. The α diversity was significantly higher in the CAM than in the non-CAM group (P < 0.001). The area under the curve was 0.849 (95% confidence interval 0.765–0.934) using the PCAM score. Among patients at < 35 weeks of gestation, the PCAM group (n = 22) had a significantly shorter extended gestational period than the non-PCAM group (n = 25; P = 0.022). Multivariate analysis revealed a significant difference in the frequency of developmental disorders in 3-year-old infants (PCAM, 28%, non-PCAM, 4%; P = 0.022). Analyzing vaginal microbiota can evaluate the risk of IAI. Future studies should establish appropriate interventions for IAI high-risk patients to improve perinatal prognosis.