Characterization of Vaginal Microbiota in Women With Recurrent Spontaneous Abortion That Can Be Modified by Drug Treatment

Abstract

Objective: The role of vaginal microbiota in recurrent spontaneous abortion (RSA) remains unknown. The purpose of this study was to investigate characteristics of vaginal microbiota and the effects of drug treatment on vaginal microbiota of patients with RSA.

Methods: A case-control study was performed, in which non-pregnant patients who experienced RSA were selected and divided into untreated and drug-treated groups. Drug-treated patients were subdivided into the metformin group, metformin plus aspirin group, and other drugs group. Healthy women who had live births and never experienced spontaneous abortion were enrolled in the control group. Characteristics of vaginal microbiomes of patients with RSA and healthy women and the impact of drug treatment on the microbiome was evaluated via 16S rRNA gene sequencing of the V3-V4 region using the Illumina MiSeq platform.

Results: Women who underwent RSA had lower microbial richness than healthy women. Compared to controls, the relative abundance of seven taxa (Megasphaera, Sneathia sanguinegens, Pseudomonas, Sphingomonas, Rhodococcus, Burkholderia- Caballeronia-Paraburkholderia, and Corynebacterium_1) in the patient's vaginal microbiota changed significantly, which may be closely related to RSA. The composition of the vaginal microbial community in RSA patients was altered by drug treatment. Metformin combined with aspirin treatment significantly increased the relative abundance of vaginal Lactobacillus spp. in patients.

Conclusion: An altered vaginal microbiome composition might be associated with RSA, which could be modified by drug treatment. The effect of metformin combined with aspirin on vaginal Lactobacillus is worthy of attention.

Keywords: 16S rRNA gene; RSA; aspirin; metformin; recurrent spontaneous abortion; vaginal microbiome.

Figure 1

The diversities of vaginal microbiota in the three groups are shown by the Chao index (A), Shannon–Wiener index (B), and principal coordinate analysis (PCoA) plot created based on the unweighted (C) and weighted (D) UniFrac distances. The values show the percentages of total community variation explained. (A, B) p-values were calculated using the one-sided Wilcoxon rank-sum tests. *p < 0.05.

Figure 2

Taxonomic classification of the vaginal microbiota at the genus level among different groups.

Figure 3

Comparison between patients with recurrent spontaneous abortion (RSA) and controls. The effect size for each differentially abundant taxon was computed using linear discriminant analysis (LDA), which indicated its contribution to group differentiation. OTUs are presented in red and green when the taxa were significantly more abundant in the controls and RSA, respectively. (A) Differentially abundant vaginal taxa detected in patients not taking any medicine and controls. (B) Differentially abundant taxa detected in patients who did take medicine and controls.

Figure 4

The effect of different medications on vaginal microbiota. The diversity of the vaginal microbiota in each group is shown by the Chao index (A), Shannon–Wiener index (B), and principal coordinate analysis (PCoA) plot created based on the unweighted (C) and weighted (D) UniFrac distances. (E) Differentially abundant vaginal taxa detected using LDA in samples taken from the metformin and control groups. (F) Differentially abundant taxa detected in patients who received metformin combined with aspirin and controls. (A, B) P values were calculated using the one-sided Wilcoxon rank-sum tests. *p < 0.05 and **p < 0.01.

Figure 5

The proportion of the Community State Types (CSTs) in each group; NM group (n = 65), DT group (n = 43), control group (n =18).