Establishment of vaginal microbiota composition in early pregnancy and its association with subsequent preterm prelabor rupture of the fetal membranes

Abstract

Vaginal bacterial community composition influences pregnancy outcome. Preterm prelabor rupture of the fetal membranes (PPROM), which precedes 30% of all spontaneous preterm births, is associated with high vaginal bacterial diversity prior to rupture. The point at which vaginal bacterial diversity is established before PPROM is unknown. In this study, we use metataxonomics to longitudinally characterize the vaginal bacterial composition from as early as 6 weeks of gestation in women at high (n = 38) and low (n = 22) risk of preterm birth who subsequently experience PPROM and in women delivering at term without complications (n = 36). Reduced Lactobacillus spp. abundance and high diversity was observed prior to PPROM in 20% and 26% of women at low and high risk of preterm births respectively, but in only 3% of women who delivered at term. PPROM was associated with instability of bacterial community structure during pregnancy and a shift toward higher diversity predominately occurring during the second trimester. This was characterized by increased relative abundance of potentially pathogenic species including Prevotella, Peptoniphilus, Streptococcus, and Dialister. This study identifies reduced Lactobacillus spp. abundance and increasing vaginal bacterial diversity as an early risk factor for PPROM and highlights the need for interventional studies designed to assess the impact of modifying vaginal bacterial composition for the prevention of preterm birth.

Fig 1

Prospective recruitment of low-risk and high-risk pregnant women. One thousand five hundred and fifty patients were recruited prospectively, yielding 60 cases of PPROM. PPROM, preterm prelabour rupture of the fetal membranes.

Fig 2

Women destined to experience PPROM exhibit reduced Lactobacillus spp. abundance and increased richness. (A). Stacked bar chart displaying percentage abundance of the top 25 bacterial genera, Lactobacillus spp., diversity and richness of vaginal bacterial communities comparing the last sample taken before PPROM (n = 60) and samples taken at a similar gestation from women subsequently delivering at term matched for age, BMI and ethnicity (n = 36). Samples prior to PPROM exhibit significantly higher richness (P = 0.0086, Mann-Whitney) and a significantly higher proportion of women with Lactobacillus spp. intermediate or deplete communities (<75% abundance) P = 0.011, Fisher's Exact. (B). Total abundance of Lactobacillus spp. is significantly higher on average in pregnancies delivering at term P = 0.016, Mann-Whitney U test. BMI, body mass index; PPROM, preterm prelabour rupture of the fetal membranes.

Fig 3

Bacterial taxonomic groups discriminate between normal term delivery and women destined to experience PPROM. (A) Cladogram describing differentially abundant vaginal bacterial clades and nodes observed between women subsequently experiencing normal term delivery or PPROM as identified using LEfSe analysis. (B) The effect size for each differentially abundant species was estimated using LDA. Vaginal microbiota of patients prior to PPROM was enriched with Bacteroidales, Fusobacteriales, and Clostridiales whereas those with a term delivery were comparatively enriched in Lactobacillales. Comparison of relative abundance across the 4 differentially expressed bacterial genera showing (C) reduced Lactobacillales (P = 0.0044) and (D) increased Prevotella (P = 0.0014), (E)Peptoniphilus (P = 0.022) and (F)Dialister (P = 0.0227, Mann-Whitney U, 2-tailed) in women prior to membrane rupture compared to term delivery control_. LDA, latent discriminatory analysis; PPROM, preterm prelabour rupture of the fetal membranes.

Fig 4

Bacterial richness and the proportion of women with a Lactobacillus spp. deplete microbiome increases with gestation in pregnancies destined for PPROM. Richness (species observed), diversity (Inverse Simpson Index) and Lactobacillus spp. abundance of vaginal bacterial communities sampled at 6–11⁺⁶, 12–17⁺⁶, 18–23⁺⁶, 24–29⁺⁶, 30–36⁺⁶ weeks gestation in women with subsequent term delivery and PPROM. (A) Richness was significantly higher at 12–17⁺⁶ weeks compared to 24–29⁺⁶ weeks in women with uncomplicated term deliveries (P = 0.03, Mann-Whitney). (D) Conversely in women with subsequent PPROM richness was significantly higher at 24–29⁺⁶ in comparison to 6–17⁺⁶ weeks (P = 0.02, Mann-Whitney) (B, E) Diversity remained stable across gestation. There were no statistically significant differences in Lactobacillus spp. abundance across gestation. (C) In women with subsequent term delivery all samples were Lactobacillus spp. dominant above 24 weeks. (D) In women with subsequent PPROM the proportion of women with Lactobacillus spp. deplete communities increases above 24 weeks. PPROM, preterm prelabour rupture of the fetal membranes.

Fig 5

Vaginal bacterial communities demonstrate instability prior to PPROM. Longitudinal profiling of vaginal bacterial communities displaying the dominant bacterial species (top) and Lactobacillus abundance state (bottom) for each sample from women with subsequent term delivery and PPROM who were sampled on 2 or more occasions antenatally. Gestation at PPROM is denoted by the blue circle while delivery is denoted by the gray circle. The vaginal bacterial community remained stable throughout pregnancy for the majority (13/16, 81%) of women with subsequent term delivery. In contrast 37% of women with PPROM experienced transition from one state to another at some point prior to membrane rupture. PPROM, preterm prelabour rupture of the fetal membranes.

Fig 6

Vaginal bacterial communities with reduced Lactobacillus spp. abundance are associated with subsequent PPROM from 24 weeks of gestation. Stacked bar chart displaying percentage abundance of the top 25 bacterial genera, Lactobacillus spp., diversity and richness of vaginal bacterial communities at 6–11⁺⁶, 12–17⁺⁶, 18–23⁺⁶, 24–29⁺⁶, 30–36⁺⁶ weeks gestation for women with subsequent term delivery and PPROM. The proportion of women with a vaginal microbiome with reduced Lactobacillus spp. abundance is significantly higher in the subsequent preterm membrane rupture group compared to term delivery at 24–29⁺⁶ (P = 0.004) and 30–36⁺⁶ (P = 0.03, Fisher's Exact). PPROM, preterm prelabour rupture of the fetal membranes.

Fig 7

Vaginal bacterial communities devoid of Lactobacillus spp. from 6 weeks and reduced Lactobacillus spp. abundance beyond 24 weeks are risk factors for subsequent PPROM. Forest plot, relative risk of subsequent PPROM based upon the composition of the vaginal microbiome at each gestational age window. (A) Bacterial communities dominated by Lactobacillus spp. beyond 24 weeks reduce the relative risk of subsequent PPROM. (B) Bacterial communities with reduced Lactobacillus spp. abundance beyond 24 weeks increase the relative risk of PPROM. (C) A vaginal microbiome dominated by non-Lactobacillus spp. is associated with increased risk of PPROM at all gestational ages. PPROM, preterm prelabour rupture of the fetal membranes.

Fig 8

Vaginal bacterial communities with reduced Lactobacillus spp. abundance are equally distributed between women considered at low and high risk of PTB. Stacked bar chart displaying percentage abundance of the top 25 bacterial genera, Lactobacillus spp., diversity and richness of vaginal bacterial communities for the last sample taken before PPROM for women considered low and high risk of PTB in the index pregnancy. The proportion of women with bacterial communities deficient in lactobacilli is similar for women considered low and high risk for PTB (5/25, 20% vs 9/35, 26%, P = 0.76, Fisher's exact).