Prevalence of bacterial vaginosis in Portuguese pregnant women and vaginal colonization by Gardnerella vaginalis

Population, data and biological samples collection

This cross-sectional study involved pregnant women attending prenatal care consultations in two public Portuguese hospitals of the Northwest region (Hospital of Braga and Unidade Local de Saúde de Matosinhos), between May 2014 and May 2016. The inclusion criteria for the present study were: to be pregnant at any gestational age and to be older than 18 years. The exclusion criteria were: women who were unable to read or understand Portuguese and who did not accept to participate in study.

During prenatal care consultation, obstetricians invited all eligible women under their care to participate in the study. All participants signed the informed consent form, as approved by the respective local ethical committees. Afterwards, the participants were questioned by the obstetricians about their age, educational level (basic, secondary or university), if they had previous BV episodes, previous deliveries, history of preterm delivery, suffered of chronic disease or smoking and about its use of intimate hygienic products and vitamin supplements, and determined the current gestational period (first, second or third trimester). A short standardized questionnaire was used to collect these data. Finally, the obstetricians collected vaginal samples (only one sample point was taken for each participant), using sterile swab containing Amies transport medium with coal (VWR, Radnor, PA, USA). After collection, the swabs were immediately conserved at 4 °C and transported to the Laboratory of Research in Biofilms Rosário Oliveira of the University of Minho, where they were processed.

Gram staining and BV diagnosis

For BV diagnosis, a direct smear was performed by transferring the vaginal fluid present on the swab to a glass slide. Then, vaginal smears were heat-fixed and Gram stained (Spiegel, Amsel & Holmes, 1983). Afterwards, the smears were visualized using an Olympus BX51 microscope (Olympus Portugal SA, Lisboa, Portugal) under oil immersion objective (1,000× magnification) and, subsequently, graded in accordance with the Nugent scoring system (Nugent, Krohn & Hillier, 1991). In summary, on 10 microscopic fields, the following bacterial morphotypes were identified and quantified: large gram-positive rods (Lactobacillus spp.), small gram-variable rods (G. vaginalis), small gram-negative rods (Bacteroides spp.) and curved gram-variable rods (Mobiluncus spp.). The sum of each morphotype score allowed to classify vaginal flora into normal (score 0–3), intermediate (score 4–6) or BV (score 7–10).

Molecular detection of G. vaginalis

Each vaginal swab was then immersed in 2 mL of 0.9% (wt/v) of sodium chloride (Liofilchem; Roseto degli Abruzzi, Italy) and the content was suspended using vigorous vortexing. Afterwards, 1.5 mL of each diluted swab content was transferred into a tube that was centrifuged at 8,000 rpm, during 10 s, in order to allow for coal deposition at the bottom of the tube. Then, 0.6 mL of supernatant was collected to a new tube and this later was centrifuged at 13,500 rpm, during 5 min. At the end, the supernatant was discarded and the pellet was suspended in 0.1 mL of sterile ultra-pure water. This bacterial suspension was incubated in a heating block at 95 °C for 20 min in order to disrupt bacterial cell wall and release the cell content, making available the genomic DNA that is required for polymerase chain reaction (PCR). After incubation, it was immediately cooled on ice for 5 min and the cell suspension was centrifuged at 13,500 rpm, during 5 min. The supernatant was used as template for a multiplex PCR since we used the PCR to amplify different DNA sequences simultaneously (namely gene encoding 16S rRNA of G. vaginalis and aap gene of Staphylococcus epidermidis). Briefly, each PCR included 6 µL of Dream Taq PCR Master Mix 2X (Thermo Fisher Scientific, Waltham, MA, USA), 1 µL of DNA from the vaginal sample, 2 µL of distilled water DNase/RNAse free (Thermo Fisher Scientific), 0.5 µL of 5 µM forward (FW1) G. vaginalis primer , 0.5 µL of 5 µM reverse (RV1) G. vaginalis primer, 0.5 µL of 5 µM FW S. epidermidis primer, 0.5 µL of 5 µM RV S. epidermidis primer and 1 µL of complementary DNA of S. epidermidis. Also, a negative control (containing 1 µL of DNase/RNase free water instead of genomic DNA) and positive control (containing 1 µL of genomic DNA from pure culture of G. vaginalis) were included in each set of reactions. The tubes were placed in a thermocycler (Bio-Rad; Hercules, CA, USA) that was programmed with the following protocol: 94 °C for 2 min, 40 steps of 94 °C for 30 s, 60 °C for 30 s, 72 °C for 1 min and finally 72 °C for 5 min. The inclusion of the specific primers for the S. epidermidis aap gene acted as internal positive control of PCR procedure, in order to ensure that amplifiable DNA was successfully extracted and there are no PCR inhibitors on sample.

After the PCR reaction, amplified products were analyzed in 1% (wt/v) of agarose (SeaKem LE; Rockland, ME, USA) gel with 0.05 µL/mL midori green nucleic acid dye (Nippon Genetics Europe GmbH, Düren, Germany). The electrophoresis run for 50 min at 100 volts. Finally, the results were visualized using the ChemiDoc (Bio-Rad) system, according to the manufacturer’s instructions. Negative results for the amplification of the gene encoding 16S rRNA of G. vaginalis were confirmed using an independent set of primers (FW2/RV2). All primers used are described in Table 1 and were previously assessed for specificity (França et al., 2012; Henriques et al., 2012; Castro et al., 2017).

Statistical analysis

Data were analyzed with GraphPad Prism version 6 (GraphPad Software Inc., La Jolla, CA, USA). Firstly, we determined BV and G. vaginalis colonization prevalences and then we used Chi-square or Fisher’s exact test to verify whether BV positive status or G. vaginalis colonization were associated with some sociodemographic, medical, reproductive, behavioral or microbiological variables. A p-value <0.05 was used as threshold for statistically significance. The strength of association between BV diagnosis or G. vaginalis colonization with the study variables was assessed through calculation of odds ratios (OR) and their 95% confidence intervals (CI).

Ethical considerations

The study was approved by the ethics committees of Unidade Local de Saúde de Matosinhos (process 013/CE/JAS) and of Hospital of Braga (process SECVS 063/2014). All the study participants agreed through informed consent to collaborate voluntarily, anonymously and freely. To ensure confidentiality no personal data were recorded that could lead to identification of the participants.