Figure 1.
Lactobacillus spp. and the vaginal pathogenic bacteria quantified by species-specific quantitative PCR (qPCR).
The relative abundance of the vaginal pathogenic bacteria and Lactobacillus were compared to the copy number of total bacteria. The relative abundance of L. crispatus, L. iners and L. jensenii was compared to the copy number of Lactobacillus. The Mann-Whitney U-test was used to evaluate statistical difference between the BV-positive and BV-negative groups. * indicates P<0.05; ** indicates P<0.01.
Figure 2.
Receiver operating characteristic (ROC) curves for bacterial molecular counts used to predict bacterial vaginosis (BV).
The closer the area under the curve (AUC) is to 1.0, the better the bacterial counts predict BV. A. vaginae (E) and G. vaginalis (G) counts have the best predictive power for BV, while Eggerthella (H), Leptotrichia/Sneathia (I) and Prevotella (J) have the better predictive power for BV. (A) Lactobacillus ROC curve; (B) L. crispatus ROC curve; (C) L. iners ROC curve; (D) L. jensenii ROC curve; (F) Megasphaera typeIROC curve.
Table 1.
Sensitivity, specificity and predictive values of the vaginal pathogenic bacteria and Lactobacillus for the prediction of BV.
Figure 3.
Correlations between Lactobacillus spp., the vaginal pathogenic bacteria and vaginal pH.
(A) Lactobacillus. (B) L. crispatus. (C) L. iners. (D) L. jensenii. (E) Atopobium vaginae. (F) Garnerella vaginalis. (G) Leptotrichia/Sneathia. (H) Megasphaera typeI. (I) Eggerthella. (J) Prevotella. The Pearson correlation (R) and probability (P) were used to evaluate statistical importance.
Figure 4.
Correlations between Lactobacillus spp., the vaginal pathogenic bacteria and Nugent scores.
(A) Lactobacillus. (B) L. crispatus. (C) L. iners. (D) L. jensenii. (E) Atopobium vaginae. (F) Garnerella vaginalis. (G) Leptotrichia/Sneathia. (H) Megasphaera typeI. (I) Eggerthella. (J) Prevotella. The Pearson correlation (R) and probability (P) were used to evaluate statistical importance.