Identification of Gardnerella vaginalis by Molecular Methods in Women Diagnosed With Bacterial Vaginosis in Isfahan, Iran

Background : Bacterial vaginosis (BV) is one of the most common causes of abnormal vaginal discharge in women. The disease is characterized by an imbalance in the vaginal bacterial flora. We aimed to determine the frequency of Gardnerella vaginosis by cultivation and molecular method. Methods: In summer 2019, the vaginal secretion of 110 women with BV were collected and isolated for G. vaginalis , in Isfahan. The isolates were identified by the biochemical test. Polymerase chain reaction (PCR) was performed to detect G. vaginalis in vaginal secretions. Antibiotic susceptibility of the isolates was evaluated by disc diffusion method. Results: Gardnerella vaginalis was isolated from five samples among the 110 patients with symptoms of BV by cultivation. Based on molecular identification, G. vaginalis was shown in 32 (29.1%) samples. G. vaginalis isolates were resistant to clindamycin (20%) and amoxicillin/clavulanic acid (80%). All the isolates were sensitive to Metronidazole. All women with this infection were married and most (43.8%) belonged to the 25-30 year-old age group. A significant difference was found between participants with positive clue cell ( P < 0.01) and pH > 4.5 vaginal discharge ( P < 0.01) in the PCR-positive and the PCR-negative women. Conclusion: High prevalence of Gardnerella vaginalis in women with vaginosis confirms the important role of bacteria in the incidence of BV. Identification of pathogenic agents of G. vaginosis using molecular methods and determining their antibiotic susceptibility pattern is essential for proper treatment in different societies.

Since G. vaginalis is a bacterium in vaginal flora of healthy women, the isolation of this bacterium may not be used for BV diagnosis. But in many BV cases increased number of G. vaginalis is associated with the presence of BV (7,8). The increase in the prevalence of BV and the lack of clinical symptoms in most patients, which have resulted in failure to treat these cases or inadequate treatments, re-admission of patients and failure to control vaginal infections, as well as increased drug resistance bacteria (9), necessitate further studies in this regard.

Objectives
The aim of this study was to evaluate the frequency of G. vaginalis in patients with BV referring to Isfahan clinics using cultivation and the molecular method.

Sampling
In this study, from September to April 2018, 110 women with evident clinical symptoms of vaginal discharge were studied. Women who had symptoms of BV, according to

Identification of Gardnerella vaginalis by Molecular Methods in Women Diagnosed With Bacterial Vaginosis in Isfahan, Iran
Hormozgan Med J . Vol 25, No 4, 2021 176 hmj.hums.ac.ir http a gynecologist were included in the study. To diagnose BV, four Amsel criteria were used. These criteria included vaginal pH > 4.5, the presence of clue cells in Gram stain, homogenous white discharge, and fishy odor (10,11).
An informed consent was obtained from the participants. Women who had received any antiinflammatory medication or antibiotics for two weeks prior to the study or who were pregnant were excluded.
A questionnaire containing patient characteristics such as age, place of residence, education, occupation, marital status, and having a child was completed by each woman.

Vaginal Sample Collection
Vaginal specimens were collected during clinical examination. A sterile swab was saturated with vaginal secretion and immediately cultured on selective medium. Additional vaginal samples were collected for molecular identification.
Vaginal discharges were cultured on chocolate agar and brain heart infusion agar supplemented by starch. Plates were incubated at 37℃ for 48-72 hours in microaerophilic atmosphere with 5-10% CO 2 . After incubation, the gramvariable or gram-negative bacilli and small and transparent colonies were presumptively identified as G. vaginalis and differential biochemical tests were performed (12,13).

Primer Design and Polymerase Chain Reaction Assay
DNA was extracted from the vaginal swaps with the Aron-Gene Pars kit, according to the manufacturer's instructions. For specific molecular identification, the primers GVNM forward (5'-TTGACATGTGCCTGACGACTG-3') and GVNM reverse (5'-GCACCATGTCACCATGAAGCAA-3') were designed based on the conserved region of the 16S rRNA gene fragments. The BLAST analysis of these primers provided an identity value of 100% with the G. vaginalis gene.

Statistical Analyses
The collected data were analyzed using SPSS software, version 17.0, descriptive statistics, and Chi-square test. P≤0.05 was considered statistically significant.

Results
We included 110 women with a mean ± SD age of 30.6 ± 6.2 years (range: 21-45 years) diagnosed with vaginosis.
Gardnerella vaginalis was isolated from 5 (4.5%) samples based on cultivation. The colony morphology and gramstain (for clue cell) are shown in Figure 1. The biochemical test of culture-positive samples were catalase-negative, oxidase-negative, starch/hydrolysis-positive, and glucose and maltose fermentation-positive.
Gardnerella vaginalis was found in 32 (29.1%) samples through the PCR method ( Figure 2). All vaginal samples with a positive culture for G. vaginalis also showed positive molecular identification.
The results of antibiotic sensitivity pattern of isolates showed that G. vaginalis were 20% resistant to clindamycin and 80% to amoxicillin/clavulanic acid. All isolates were sensitive to metronidazole (Table 1).
All the participants were married and most (49.1%) belonged to the age group of 25-30 years. All women had a normal marriage with their spouses. There was no relationship between age and BV (P = 0.200). The  Table 2).
We found no statistically significant association between place of residence, college education, occupation, having a child, and abortion history and PCR-positive and PCRnegative participants. The frequency of clue cell, amine odor and pH > 4.5 in vaginal secretions is presented in Figure 3.
Chi-square test showed a significant difference in pH > 4.5 in PCR-positive and PCR-negative groups (P = 0.017). The Whiff test was positive in 87.5% of PCRpositive and 86.92% of PCR-negative samples for G. vaginalis. There was no significant difference between Whiff test between the two groups (P = 0.208). Clue cells were observed in 87.5% of PCR-positive and 25.64% of PCR-negative groups (P < 0.001).

Discussion
BV, caused by G. vaginalis, is an infectious disease characterized by increased vaginal discharge, itching, and burning during urination or itching around the outside of the vagina, or both. In addition to annoying symptoms, this bacterium can increase the risk of many upper genital tract infections. Therefore, rapid diagnosis and antimicrobial treatment of this infection are very important.
Based on the Amsel criteria, all the women who participated in this study had BV. The results of bacterial culture showed that the prevalence of G. vaginalis among the 110 women with BV symptoms was 4.5%. Since G. vaginalis is a choosy bacterium, it is always difficult to isolate it using conventional culture methods. For example, Farajzadeh Sheikh and Hemmati reported that the rate of isolation for this bacterium in a culture medium was 23.8% (16). However, this bacterium was isolated from 29.1% of the samples when PCR was used as a more sensitive and accurate method. There are different   (21). However, Kalantari et al reported that the highest prevalence of BV was among women aged 20-30 years (22). Although all of these studies concluded that the sexual activity of women of these ages was the main cause of affliction with BV, none of them found a significant relationship between age group and BV. Other studies showed that there was a significant relationship between level of education and affliction with BV, which was attributed to awareness of and compliance with health issues (21,23). Ramezani Tehrani reported that the prevalence of BV in employed women was lower than unemployed ones (24), which is not consistent with the findings of the present study. We found no significant relationship between the prevalence of BV and the participants' place of residence, college education, employment status, having a child, and history of abortion in the two groups of afflicted with (PCR-positive) and not afflicted with (PCR-negative) G. vaginalis.
Previous studies have reported different antibiogram results for isolates of G. vaginalis. All strains were sensitive to metronidazole and 80% of them were sensitive to clindamycin (25). Goldstein reported that 28% of the clinical isolates of G. vaginalis were resistant to metronidazole, whereas all of the isolates were sensitive to clindamycin and ampicillin/sulbactam (26). Moreover, Maghsoudi showed that 75% of G. vaginalis isolates were resistant to clindamycin (27). On the other hand, another study showed that 9% of the isolates of this bacterium were sensitive to ampicillin (25). This discrepancy is probably related to different patterns of antibiotic consumption in different regions and the prevalence of resistant strains of G. vaginalis. The results of this study indicated that increased vaginal pH ( > 4.5) was significantly related to positive whiff test results and infection with G. vaginalis in patients with BV. Accordingly, although some studies have shown that G. vaginalis is among the vaginal bacterial flora, increased vaginal pH and positive whiff test seem to be appropriate criteria for the diagnosis of BV, which is consistent with our findings.
Given the importance of early diagnosis and specific treatment of G. vaginalis to eradicate this organism, gynecologists and obstetricians are recommended to prescribe a test on the stained smear of vaginal discharge (to view the clue cells) and measure its pH in women aged over 25 years. Therefore, the use of molecular methods in the definitive diagnosis of G. vaginosis may be more appropriate.