Composition of the vaginal microbiota in women of reproductive age--sensitive and specific molecular diagnosis of bacterial vaginosis is possible?

Elena Shipitsyna,Anders Hallén,Annika Roos,Magnus Unemo,Hans Fredlund,Lars Engstrand,Jørgen S Jensen,Raluca Datcu,Tom Coenye

doi:10.1371/journal.pone.0060670

Abstract

Background and ObjectiveBacterial vaginosis (BV) is the most common vaginal disorder, characterized by depletion of the normal lactobacillus-dominant microbiota and overgrowth of commensal anaerobic bacteria. This study aimed to investigate the composition of the vaginal microbiota in women of reproductive age (healthy women and women with BV), with the view of developing molecular criteria for BV diagnosis.Materials and MethodsVaginal samples from 163 women (79 control, 73 BV and 11 intermediate (Lactobacillary grade II flora) cases) were analyzed using 454 pyrosequencing of the hypervariable regions V3–V4 of the 16S rRNA gene and 16 quantitative bacterial species/genus-specific real-time PCR assays. Sensitivities and specificities of potential BV markers were computed using the Amsel criteria as reference standard for BV. The use of quantitative thresholds for prediction of BV, determined for both relative abundance measured with 454 pyrosequencing and bacterial load measured with qPCR, was evaluated.ResultsRelative to the healthy women, the BV patients had in their vaginal microbiota significantly higher prevalence, loads and relative abundances of the majority of BV associated bacteria. However, only Gardnerella vaginalis, Atopobium vaginae, Eggerthella, Prevotella, BVAB2 and Megasphaera type 1 detected at or above optimal thresholds were highly predictable for BV, with the best diagnostic accuracy shown for A. vaginae. The depletion of Lactobacillus species combined with the presence of either G. vaginalis or A. vaginae at diagnostic levels was a highly accurate BV predictor.ConclusionsQuantitative determination of the presence of G. vaginalis, A. vaginae, Eggerthella, Prevotella, BVAB2 and Megasphaera type 1 as well as the depletion of Lactobacillus was highly accurate for BV diagnosis. Measurements of abundance of normal and BV microbiota relative to total bacteria in vaginal fluid may provide more accurate BV diagnosis, and be used for test-of-cure, rather than qualitative detection or absolute counts of BV related microorganisms.

Highlights

Bacterial vaginosis (BV) is the most common vaginal disorder in women of reproductive age [1], commonly manifested as abnormal vaginal discharge
Only Gardnerella vaginalis, Atopobium vaginae, Eggerthella, Prevotella, BVAB2 and Megasphaera type 1 detected at or above optimal thresholds were highly predictable for BV, with the best diagnostic accuracy shown for A. vaginae
The depletion of Lactobacillus species combined with the presence of either G. vaginalis or A. vaginae at diagnostic levels was a highly accurate BV predictor

Summary

Introduction

Bacterial vaginosis (BV) is the most common vaginal disorder in women of reproductive age [1], commonly manifested as abnormal vaginal discharge. No single etiologic agent is known to be the cause of BV, and the syndrome is considered an ecological disorder of the vaginal microbiota. It is characterized by depletion of the normal lactobacillus-dominant microbiota, accompanied by a shift towards higher bacterial diversity and intense overgrowth of commensal anaerobic bacteria (100–1000-fold above normal) [15]. Bacterial vaginosis (BV) is the most common vaginal disorder, characterized by depletion of the normal lactobacillus-dominant microbiota and overgrowth of commensal anaerobic bacteria. This study aimed to investigate the composition of the vaginal microbiota in women of reproductive age (healthy women and women with BV), with the view of developing molecular criteria for BV diagnosis

Objectives

Methods

Results

Discussion

Conclusion