Association of the Cervical Microbiota With Pregnancy Outcome in a Subfertile Population Undergoing In Vitro Fertilization: A Case-Control Study

Abstract

The microorganisms of the reproductive tract have been implicated to affect in vitro fertilization (IVF) outcomes. However, studies on the reproductive tract microbiota of infertile women are limited and the correlation between cervical microbiota and IVF outcome remains elusive. This study aimed to characterize the cervical microbiota of IVF patients undergoing embryo transfer (ET) and assess associations between the cervical microbiota and pregnancy outcomes while exploring the underlying contributing factors. We launched a nested case-control study of 100 patients with two fresh or frozen-thawed cleavage embryos transferred per IVF cycle. Cervical swabs were collected on the day of ET and divided into four groups according to clinical pregnancy outcomes. Variable regions 3 and 4 (V3-V4) of the 16S rRNA gene were amplified and sequenced on the Illumina MiSeq platform. In fresh IVF-ET cycles, the clinical pregnancy group (FP, n = 25) demonstrated higher α diversity (P = 0.0078) than the non-pregnancy group (FN, n = 26). Analysis of similarity (ANOSIM) revealed a significant difference in β diversity between the two groups (R = 0.242, P = 0.001). In frozen-thawed ET cycles, though not significant, similar higher α diversity was found in the clinical pregnancy group (TP, n = 27) compared to the non-pregnancy group (TN, n = 22) and ANOSIM analysis showed a significant difference between the two groups (R = 0.062, P = 0.045). For patients in fresh IVF-ET groups, Lactobacillus, Akkermansia, Desulfovibrio, Atopobium, and Gardnerella showed differentially abundance between pregnant and non-pregnant women and they accounted for the largest share of all taxa investigated. Among them, Lactobacillus was negatively correlated with the other genera and positively correlated with serum estradiol levels. Logistic regression analysis suggested that the composition of the cervical microbiota on the day of ET was associated with the clinical pregnancy in fresh IVF-ET cycles (P = 0.030). Our results indicate that cervical microbiota composition has an impact on the outcome of assisted reproductive therapy.

Keywords: 16S r RNA; IVF (in vitro fertilization); cervical microbiota; infertility; pregnancy.

Figure 1

Shows the distribution of the microbiota in the FP, FN, TP, and TN groups. (A) Venn diagram: represents the OTU distribution of the four groups. (B–D)  The average relative abundance of the sample microbiota in four groups at the level of phylum, genus, and species. (E) The relative abundance of the 100 sample microbiota at the genus level. FN, fresh IVF-ET cycle non-pregnancy; FP, fresh IVF-ET cycle pregnancy; TN, frozen-thaw ET cycle non-pregnancy; TP, frozen-thaw ET cycle pregnancy.

Figure 2

Shows the α diversity of the FP, FN, TP, and TN groups. (A) Dilution curve: FP, FN group Chao diversity index dilution curve (B) Dilution curve: TP, TN group Chao diversity index dilution curve (C) Box diagram, FP, FN group Shannon diversity index (D) Box plot: FP, FN group Simpson diversity index (E) Box plot: TP, TN group Shannon diversity index (F) Box plot: TP, TN group Simpson diversity index. FN, fresh IVF-ET cycle non-pregnancy; FP, fresh IVF-ET cycle pregnancy; TN, frozen-thaw ET cycle non-pregnancy; TP, frozen-thaw ET cycle pregnancy.

Figure 3

Shows the β diversity of the microbiota in the pregnancy and non-pregnancy groups of the fresh and frozen-thawed cycles. (A) FP, FN group microbiota unweighted PCoA analysis chart (B) TP, TN group microbiota unweighted PCoA analysis chart (C) FP, FN group microbiota unweighted Anosim analysis chart (D) TP, TN group microbiota unweighted Anosim analysis chart. FN, fresh IVF-ET cycle non-pregnancy; FP, fresh IVF-ET cycle pregnancy; TN, frozen-thaw ET cycle non-pregnancy; TP, frozen-thaw ET cycle pregnancy, PCoA, Principal co-ordinates analysis; ANOSIM, Analysis of similarity.

Figure 4

Shows the differential abundance and association analysis among the cervical microbiota of the fresh pregnancy group and the non-pregnancy group. (A) LEfSe analysis chart: The LDA score of the genera which showed differentially abundance between pregnant and non-pregnant women in fresh IVF-ET cycle. (B) Spearman correlation coefficient analysis: the important patterns and relationships among the genera abundance with Top 30 in FP, FN groups (C) Spearman thermal map analysis of the correlation between the serum sex hormone levels and genera which were detected in LEfSe analysis. FN, fresh IVF-ET cycle non-pregnancy; FP, fresh IVF-ET cycle pregnancy. E₂, estradiol; LDA, Linear discriminant analysis; LEfSe, Linear discriminant analysis effect size analysis; P, progesterone. ⁺ P < 0.05; *P < 0.01.