Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Fredricks, D. N. Articles by Marrazzo, J. M. Search for Related Content PubMed PubMed Citation Articles by Fredricks, D. N. Articles by Marrazzo, J. M.

 Previous Article  |  Next Article 

Journal of Clinical Microbiology, March 2009, p. 721-726, Vol. 47, No. 3
0095-1137/09/$08.00+0     doi:10.1128/JCM.01384-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Changes in Vaginal Bacterial Concentrations with Intravaginal Metronidazole Therapy for Bacterial Vaginosis as Assessed by Quantitative PCR

David N. Fredricks,^1,2* Tina L. Fiedler,¹ Katherine K. Thomas,³ Caroline M. Mitchell,^1,4 and Jeanne M. Marrazzo²

Fred Hutchinson Cancer Research Center,¹ Department of Medicine, University of Washington,² Center for AIDS and STDs, University of Washington,³ Department of Obstetrics and Gynecology, University of Washington, Seattle, Washington⁴

Received 19 July 2008/ Returned for modification 3 September 2008/ Accepted 5 January 2009

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Several fastidious bacteria have been associated with bacterial vaginosis (BV) using broad-range bacterial PCR methods such as consensus sequence 16S rRNA gene PCR, but their role in BV remains poorly defined. We describe changes in vaginal bacterial concentrations following metronidazole therapy for BV. Vaginal swabs were collected from women with BV diagnosed using Amsel clinical criteria, and vaginal fluid was assessed by Gram stain to generate Nugent scores. Follow-up swabs were collected 1 month after a 5-day course of vaginal 0.75% metronidazole gel and analyzed for 24 subjects with cured BV and 24 subjects with persistent BV. Changes in bacterial concentrations were measured using eight bacterium-specific 16S rRNA gene quantitative PCR assays. DNA from several fastidious BV-associated bacteria (BVAB) were present at high concentrations in the vagina prior to treatment. Successful antibiotic therapy resulted in 3- to 4-log reductions in median bacterial loads of BVAB1 (P = 0.02), BVAB2 (P = 0.0004), BVAB3 (P = 0.03), a Megasphaera-like bacterium (P < 0.0001), Atopobium species (P < 0.0001), Leptotrichia/Sneathia species (P = 0.0002), and Gardnerella vaginalis (P < 0.0001). Median posttreatment bacterial levels did not change significantly in subjects with persistent BV except for a decline in levels of BVAB3. The presence or absence of BV is reflected by vaginal concentrations of BV-associated bacteria such as BVAB1, BVAB2, Leptotrichia/Sneathia species, Atopobium species, Gardnerella vaginalis, and a Megasphaera-like bacterium, suggesting that these bacteria play an important role in BV pathogenesis and may be suitable markers of disease and treatment response.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial vaginosis (BV) is a very common condition, affecting 10 to 37% of women (24), with particularly high prevalences reported in areas of sub-Saharan Africa (3, 25) and among African-American women (51%) (15). BV is the most common cause of vaginitis and vaginal discharge and has been linked to several adverse health outcomes including preterm labor, pelvic inflammatory disease, and human immunodeficiency virus acquisition (12, 18, 19).

Women with BV experience a loss of vaginal Lactobacillus species that produce hydrogen peroxide (primarily Lactobacillus crispatus) and the acquisition of several facultative or anaerobic bacteria, but no single cultivated bacterium has been conclusively demonstrated to cause BV. The profound shifts in vaginal microbiota that characterize BV and the clinical response to antibiotics such as metronidazole or clindamycin suggest that anaerobic bacteria play a key role. The recent use of cultivation-independent methods (e.g., broad-range 16S rRNA gene PCR) to characterize the vaginal microbiota (5, 26) has identified novel fastidious bacteria that appear to be highly associated with BV, including three bacteria of the order Clostridiales that are not closely related to any previously identified bacteria and that we have designated BV-associated bacterium 1 (BVAB1), BVAB2, and BVAB3 (6). Other fastidious bacteria detected in subjects with BV using molecular methods include Atopobium vaginae, Leptotrichia/Sneathia species, and bacteria most closely related to Megasphaera species. Women with BV have a higher degree of vaginal bacterial species diversity (species richness) than women without BV (6, 7).

The consistent presence of so many different bacterial species in subjects with BV suggests, but does not prove, that BV is caused by communities of bacteria that include many uncultivated species. Since in vitro antibiotic susceptibility cannot be determined for uncultivated bacterial species, the response of fastidious vaginal bacteria to antibiotic therapy must be assessed using other approaches such as measuring bacterial concentrations in vivo. For instance, one can assess the impact of metronidazole therapy on vaginal concentrations of BVAB1 to determine if decreasing bacterial levels are associated with cure. Theoretically, particular bacterial species may decrease in concentration because they are directly susceptible to the antibiotic administered or because they are metabolically dependent on other species that are eradicated with antibiotic therapy. However, some vaginal bacteria present in women with BV may not be directly implicated in pathogenesis but may colonize an open vaginal niche vacated by Lactobacillus crispatus and other vaginal lactobacilli. Concentrations of such bacteria may not directly correlate with the presence of BV or response to antibiotic therapy.

We sought to determine how concentrations of vaginal bacteria change in women with BV by comparing women who were cured to women with persistent BV 1 month following vaginal metronidazole treatment. We hypothesized that concentrations of fastidious vaginal bacteria linked to BV would drop with cure of BV but would remain elevated in women with persistent BV. We used eight taxon-directed real-time quantitative PCR (qPCR) assays targeting both easily cultivated vaginal bacteria (Gardnerella vaginalis and Lactobacillus crispatus) and fastidious bacteria (BVAB1, BVAB2, BVAB3, Leptotrichia/Sneathia, Atopobium, and Megasphaera-like species) to measure levels of vaginal bacteria as reflected by bacterial rRNA gene concentrations.

(This work was presented in part at the Infectious Diseases Society of America Meeting in San Diego, CA, October 2007 and at the International Society for Sexually Transmitted Diseases Research Meeting, Seattle, WA, July 2007.)

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study population and definitions. This study was a nested case-control study of women with cured versus persistent BV. The study population was derived from women enrolled in an institutional review board-approved prospective longitudinal study of vaginal flora among women who have sex with women, a population with a high prevalence of BV. Participants were seen at the Women's Research Clinic in Seattle, WA, and were eligible for enrollment if they reported sex with at least one woman in the prior year and were 16 to 35 years of age. Report of vaginal symptoms was not required for eligibility. For this substudy, women were included if they had a diagnosis of BV, returned for a 1-month follow-up visit within 25 to 47 days by 7 March 2007, and had vaginal swabs available for analysis at both initial diagnosis and 1-month follow-up visits. All participants provided written informed consent and underwent standard pelvic examination with the collection of vaginal fluid for performance of tests of Amsel criteria. BV was diagnosed if at least three of four criteria (vaginal discharge, elevated vaginal pH, clue cells, and positive whiff test) were present (1). The presence of abnormal vaginal flora at initial diagnosis was later confirmed using Gram stains of vaginal fluid (20), with all subjects having Nugent scores of 4. Women with BV received a single 5-day course of intravaginal metronidazole gel (Metrogel [37.5 mg daily]) and were asked to return at 1 month for a test of cure, where study procedures were repeated, including the collection of vaginal swabs; they were asked to defer the scheduled return visit until the cessation of menses, if relevant. Clinical cure of BV was defined as having less than three Amsel criteria posttreatment, and clinically persistent BV was defined as having at least three Amsel criteria. Microbiological cure of BV was defined as having a Nugent score of 6 at 1 month posttreatment, and persistent BV was defined as having a Nugent score of 7.

Sample collection and DNA extraction. A foam swab (Catch-All; Epicentre) was used to collect vaginal fluid by rolling it along the lateral vaginal wall. Swabs were immediately frozen at –80°C. At the time of assay, swabs were thawed and washed with 2 ml of saline in a vortex mixer. A 1.0-ml aliquot of saline was centrifuged, and the pellet was subjected to DNA extraction (Ultra-Clean Soil kit; MoBio). DNA was eluted in 150 µl of buffer, and 2 µl of template DNA was used for each qPCR assay. Extracted DNA was frozen until assayed by qPCR. Swabs without human contact (sham digest controls) were subjected to DNA extraction in parallel with vaginal swabs, with one extraction control for every 12 swabs.

qPCR. qPCR assays employed primers and a dual-labeled fluorogenic probe hydrolyzed during PCR (TaqMan format) for detecting each bacterium's specific 16S rRNA gene in a highly sensitive and specific fashion. All probes had a carboxyfluorescein (FAM) reporter at the 5' end and a tetramethyl-6-carboxyrhodamine (TAMRA) quencher or Black Hole quencher at the 3' end. Probes were used at a concentration of 150 nM in each PCR. Core reagent kits from Applied Biosystems were used to assemble master mixes containing buffer A, deoxynucleotide triphosphates (0.8 mM), magnesium (final concentration, 3 mM), uracil-N-glycosylase, and TaqGold polymerase. Forward and reverse primers were added to master mix at a concentration of 0.8 µM. All qPCRs underwent 45 cycles of amplification. Primers and probes were designed using multiple-sequence alignments of vaginal bacterial 16S rRNA gene sequences (6) (Table 1). qPCR assays targeted several vaginal bacteria that are significantly associated with BV or vaginal health (7), including a Megasphaera-like bacterium, Atopobium vaginae, the closely related Leptotrichia and Sneathia species (single assay), Gardnerella vaginalis, Lactobacillus crispatus, and three Clostridium-like bacteria which we have designated BVAB1, BVAB2, and BVAB3. Plasmids containing 16S rRNA genes from 45 vaginal bacterial species were employed to assess detection thresholds for each bacterium-specific qPCR assay, to generate standard curves for quantitation, and to assess cross-reactivity with nontarget species. Bacterial levels were expressed as 16S rRNA gene copies per swab. To meet criteria for assay implementation, each bacterium-specific qPCR was required to detect 10 16S rRNA gene copies per reaction of the target species in duplicate reactions and to show no detection when 1,000,000 16S rRNA genes copies from each of 44 nontarget vaginal bacterial species were employed as template DNA. Four no-template PCR controls were run with each qPCR assay. No-template controls consisted of all necessary PCR reagents for amplification except template DNA and were used to monitor for bacterial contamination of the PCR reagents.

View this table: [in this window] [in a new window]

TABLE 1. Primer and probe sequences used in qPCR assaysa

Extraction and amplification controls. Beta-globin gene PCR was performed on all extracted DNA samples to ensure that the swabs contacted a tissue surface during sampling and to document the presence of amplifiable DNA (8). In addition, amplification control qPCRs using exogenous DNA were performed on all samples to detect PCR inhibitors in a more sensitive and quantitative fashion. The amplification control qPCR assay targeted a segment of the jellyfish aequorin gene using specific primers and a TaqMan probe as previously described (14). A synthesized single strand of DNA coding for a portion of the jellyfish aequorin gene was added to PCR master mix at a concentration of 10⁵ copies per reaction. Inhibition in the amplification control PCR was defined as a >2-cycle increase in the threshold cycle for the extracted vaginal DNA sample compared to the no-template controls.

Statistics. Subjects' demographic and other nonmicrobial characteristics were presented by case or control status and compared using a t test for age and Fisher's exact test for other characteristics, all binary. The statistical significance of differences in pre- and posttreatment bacterial levels for each participant was evaluated using Wilcoxon signed-rank tests. Changes in quantities of vaginal bacterial DNA within cured patients were compared to changes within persistent patients using Wilcoxon rank-sum tests. Tests were performed on those subjects with specific bacteria detected at one or more of the two time points. Since a few women did not complete metronidazole treatment, analyses were also performed for only those women who reported completing the course of metronidazole. Analyses were performed using Stata 10.0.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Subject characteristics. Among 107 women returning 25 to 47 days after diagnosis with BV by Amsel criteria in the parent longitudinal study, persistent BV was found in 24 women (22.4%), with initial and follow-up swabs available for all 24 women. Initial and follow-up swabs from 24 women with clinically cured B were analyzed. Clinical cure (Amsel criteria) and microbiological cure (Nugent criteria) corresponded exactly in these 48 subjects (24 cured and 24 persistent). Five of 23 (22%) subjects with persistent BV and 1 of 20 (5%) cured subjects reported not completing metronidazole treatment (Table 2). Five subjects did not respond to the question regarding completion of treatment, including one subject in the persistent-BV group and four subjects in the cured group. One subject in the cured group was menstruating at the time of follow-up sample collection.

View this table: [in this window] [in a new window]

TABLE 2. Subject characteristics by BV status at test-of-cure visit

PCR controls. Template-free PCR controls and DNA from sham digest control swabs (without human contact) were negative in the bacterial qPCR assays, ruling out bacterial contamination of PCR reagents or DNA extraction reagents and swabs. Beta-globin DNA was successfully PCR amplified from every vaginal sample, providing evidence that all swabs appropriately collected vaginal fluid. Sham digest control swabs were beta-globin PCR negative, as expected. Likewise, the amplification control qPCRs targeting exogenously added jellyfish DNA were successful for all samples, providing evidence that PCR inhibitors were not present at significant concentrations.

Assessment of bacterial qPCR assays. It is possible that the presence of nontarget bacterial rRNA gene sequences in extracted vaginal fluid DNA may falsely elevate levels of measured bacterial DNA due to poor specificity or may falsely decrease measured levels of bacterial DNA by interfering with amplification, such as through competition. To assess the impact of different bacterial rRNA gene sequences on the amplification efficiency of each bacterium-specific qPCR assay, pools of plasmids containing 16S rRNA gene sequences from 50 vaginal bacteria were added to qPCR reactions. The threshold cycle values of qPCRs containing the single target plasmid were compared to the threshold cycle values of the target plasmid in a background of high concentrations of nontarget bacterial plasmids (10⁵ copies/reaction). We found no evidence that the mixtures of bacterial rRNA gene sequences had an impact on threshold cycle values, confirming that the qPCR assays provide an accurate measurement of bacterial DNA levels. Studies performed with human DNA also confirm that there is no impact on the detection of vaginal bacterial DNA.

Bacterium-specific qPCR results. Table 3 shows median values for bacterial rRNA gene concentrations before and after antibiotic therapy in the two outcome groups of cured and persistent BV, with cure defined by the presence of less than three Amsel clinical criteria at follow-up and a Nugent score of <7 as well as by the prevalence of individual bacteria in each group. For participants cured of BV, median values for BV-associated bacterial DNA decreased to a level below our detection threshold (either 1,000 or 750 16S rRNA gene copies/swab) for most species, with 3- to 5-log reductions in median values. Concentrations of G. vaginalis DNA dropped 4 logs in the cured group, but the median level of bacterial DNA was still 40,000 16S rRNA gene copies/swab at follow-up. Differences in bacterial rRNA gene concentrations between BV diagnosis and follow-up for cured participants were statistically significant for BVAB1 (P = 0.02), BVAB2 (P = 0.0004), BVAB3 (P = 0.03), a Megasphaera-like bacterium (P < 0.0001), Atopobium species (P < 0.0001), Leptotrichia/Sneathia species (P = 0.0002), and Gardnerella vaginalis (P < 0.0001).

View this table: [in this window] [in a new window]

TABLE 3. Concentrations of bacteria in vaginal fluid expressed as 16S rRNA gene copies per swab at initial diagnosis (pretreatment) and 1 month after antibiotic treatment (posttreatment) in women with cured and persistent BV

In contrast, women with persistent BV after antibiotic treatment had higher concentrations of BV-associated bacterial DNA at follow-up than subjects who were cured, with species that included BVAB1, BVAB2, a Megasphaera-like bacterium, and Gardnerella vaginalis, with no significant differences in concentrations of bacterial DNA between diagnosis and follow-up. One exception of note was BVAB3, for which median bacterial DNA concentrations decreased significantly (P = 0.03), from 1.4 x 10⁶ to 1.9 x 10⁴ 16S rRNA gene copies per swab, in subjects with persistent BV. Although median BVAB3 DNA concentrations decreased in women with persistent BV, there was a higher prevalence of BVAB3 at diagnosis in the group with persistent BV (58.3%) than in the group with cured BV (25%) (P = 0.04 by Fisher's exact test).

Median concentrations of Lactobacillus crispatus DNA increased more than 2 logs in those participants cured of BV, from 4.6 x 10⁵ 16S rRNA gene copies/swab at the diagnosis of BV to 6.9 x 10⁷ copies/swab at the posttreatment visit (P = 0.05). On the other hand, L. crispatus was detected in only 10 of 24 women (42%) with cured BV at the 1-month follow-up visit. For comparison, L. crispatus was detected by PCR in 93.4% of subjects without BV from this clinic population. There was no significant change in median L. crispatus DNA levels in those with persistent BV, but only 2 of these 24 women (8.3%) had detectable L. crispatus levels at the follow-up examination.

Tests of statistical significance noted above have focused on whether median levels of vaginal bacterial DNA were significantly different by comparing pretreatment and posttreatment values within an outcome group (cured or persistent BV). We also sought to determine if changes in median vaginal bacterial DNA concentrations from pre- to posttreatment were significantly different between outcome groups (cured versus persistent BV). The difference in changes in median vaginal bacterial DNA concentrations between women with cured BV and those with persistent BV was significant for all bacteria assayed except BVAB3 (P = 0.6), reflecting the fact that BVAB3 DNA concentrations decreased in both the cured- and persistent-BV groups. On the other hand, 9 of 14 subjects (64%) with persistent BV and BVAB3 at diagnosis still had BVAB3 detected at the follow-up examination.

Results comparing pretreatment and posttreatment levels of vaginal bacteria, both within an outcome group and between outcome groups, were very similar when the results were restricted to women who reported completing the full course of metronidazole treatment.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The pathophysiological process underlying BV remains poorly understood. In 1954, Gardner and Dukes first claimed that Gardnerella vaginalis (called Haemophilus vaginalis at the time) was the main etiological agent in nonspecific vaginitis, now known as BV (11). A follow-up report by those investigators in 1955 suggested that a pure culture of G. vaginalis inoculated into the human vagina could produce BV (4, 7, 10). However, a closer inspection of their data demonstrated that whole vaginal fluid from subjects with BV was a much more efficient inoculum, and most subjects inoculated with pure cultures of G. vaginalis did not develop BV. Despite the initial claims of Gardner and Dukes, Koch's postulates for disease causation have not been consistently fulfilled for any microbe in the syndrome BV. In particular, the ability to detect G. vaginalis in 50 to 60% of women without BV (7) violates Koch's second postulate regarding specificity for disease (4). Many additional cultivated bacteria have been isolated from women with BV, including Mobiluncus species, Mycoplasma hominis, and a variety of anaerobic bacteria such as Prevotella and Porphyromonas species (16, 21). The recent identification of several fastidious vaginal bacteria using sequence-based detection methods adds to the list of potential bacterial pathogens in BV (5, 6, 13, 26), but uncertainty regarding their contribution to pathogenesis persists.

In the present study, we sought to determine whether changes in quantities of vaginal bacteria on vaginal swabs, as reflected by bacterial DNA concentrations, are associated with the cure of BV after metronidazole therapy. We reasoned that vaginal concentrations of bacteria which are etiologically linked to BV would decrease in concentration with clinical cure but would remain elevated in women with persistent BV. Conversely, concentrations of bacteria that are not important in the pathogenesis of BV might tend to show a weaker direct quantitative relationship with clinical response. This reasoning follows sequence-based guidelines for collecting evidence of microbial disease causation elaborated by Fredricks and Relman in 1996 (most specifically, guideline 3) (9). More definite proof of causation would come from the fulfillment of Koch's postulates, but Koch's postulates cannot be fulfilled for uncultivated microbes such as many bacteria found in subjects with BV and, as noted above for Gardnerella vaginalis, are not completely specific. We emphasize that the mere correlation of vaginal bacterial concentrations with disease status does not prove that a causal relationship exists.

The qPCR data presented here demonstrate that there is a strong association between levels of fastidious BV-associated bacterial DNA and response to intravaginal metronidazole. This correlation is significant for BVAB1, BVAB2, a Megasphaera-like bacterium, Atopobium species, and Leptotrichia/Sneathia species but is also significant for the easily cultivated bacterium G. vaginalis. Interestingly, BVAB3 DNA concentrations decreased in both the women cured of BV and those with persistent BV, a finding that suggests that this bacterium may be less critical for pathogenesis or that its role in the pathogenic bacterial community may be replaced by other species. However, the presence of BVAB3 at baseline is a strong predictor of BV persistence, a finding that we have confirmed in a separate analysis (17). The failure of metronidazole to decrease levels of vaginal bacterial DNA in the group of subjects with persistent BV at the 1-month follow-up evaluation is an important finding that may have several explanations. First, it is possible that concentrations of BV-associated bacteria decreased during therapy but that bacteria were not eradicated, resulting in rebound after the end of antibiotic suppression. Second, BV-associated bacteria may have been eradicated from the vagina but reintroduced after the 5-day antibiotic course either from an endogenous reservoir or from a sexual partner. Third, specific BV-associated bacteria may have been more resistant to metronidazole in the subjects with persistent BV. Further studies are under way to explore some of these mechanisms for failure.

Median vaginal concentrations of Lactobacillus crispatus DNA increased in subjects with cured BV but remained below our detection threshold for all but two subjects with persistent BV. On the other hand, only 10 of 24 subjects (42%) with cured BV had vaginal colonization with L. crispatus at the 1-month follow-up visit. It is also interesting that five subjects with BV who were cured had high levels of L. crispatus DNA (>10⁵ 16S rRNA gene copies/swab) both before and after antibiotic treatment. Although these data support the view that L. crispatus is an important component of the normal vaginal microbiota, they also indicate that detectable vaginal colonization with this species is not the sole determinant of cure as defined by clinical criteria. From a clinical standpoint, this also indicates that vaginal colonization by adequate quantities of L. crispatus may first require the effective eradication of BV-associated bacteria.

There are some limitations of this study. First, we performed qPCR assays targeting six fastidious bacteria recently associated with BV using molecular methods and two cultivated bacteria that have been associated with BV (G. vaginalis) or vaginal health (L. crispatus). There are many additional vaginal bacteria that could be assayed using the qPCR platform and which may play a role in the pathogenesis of BV. Future research should seek to measure how these bacterial concentrations correlate with disease status. Second, our detection threshold for each assay was 750 to 1,000 copies per swab. The use of a larger fraction of vaginal fluid DNA for each assay would reduce the detection thresholds but would also compromise one's ability to run multiple assays. Third, the results of this study may not be generalizable to women treated for BV with oral antibiotics or to women having sex with men only. One of the strengths of this study is the extensive use of PCR controls to monitor for false-positive and false-negative results, thereby increasing the reliability of the bacterial qPCR data reported.

Several studies have used broad-range 16S rRNA gene PCR with clone library analysis to assess changes in the vaginal microflora in longitudinally collected samples from subjects with BV (5, 6), but this approach detects only the dominant species and is not reliably quantitative. Other studies have used taxon-directed conventional PCR assays to describe the prevalences of various bacterial species in vaginal samples, and while this method is much more sensitive at detecting minority bacterial species than broad-range PCR, it is also not reliably quantitative (7, 21). There are few published studies that have used targeted qPCR methods to describe levels of vaginal bacteria, but most such studies were cross-sectional (22, 23, 27, 28); even fewer studies have examined how vaginal bacterial concentrations change with antibiotic therapy (2). Ferris at al. previously applied an Atopobium vaginae qPCR assay to vaginal samples from six subjects with BV and showed that a low threshold cycle detected during qPCR with posttreatment samples (indicative of high concentrations of bacterial DNA) was associated with treatment failure, although absolute levels of bacteria or bacterial DNA were not reported (5).

In conclusion, median vaginal concentrations of bacterial DNA from several fastidious species decreased 3 to 5 logs between the diagnosis of BV and cure at 1 month following a 5-day course of intravaginal metronidazole, including BVAB1, BVAB2, BVAB3, Leptotrichia/Sneathia species, Atopobium species, and a Megasphaera-like bacterium, and these differences were highly significant for BVAB2, the Megasphaera-like bacterium, Leptotrichia/Sneathia species, and Atopobium species. Most cured subjects experienced eradication of these bacteria. Median concentrations of G. vaginalis DNA also significantly decreased, while median concentrations of L. crispatus DNA significantly increased in cured subjects. In contrast, subjects with persistent BV had persistently elevated levels of fastidious BV-associated bacterial DNA. These data demonstrate that BV status correlates with concentrations of fastidious vaginal bacterial DNA, and thus bacteria, and provide supporting evidence that these bacteria are important in the pathogenesis of BV. BV has a high rate of relapsing or recurrent disease. Efforts to understand this enigmatic condition will benefit from a greater understanding of the complex bacterial populations found in the human vagina and how they change over time. Furthermore, the detection of particular vaginal bacteria or consortia by PCR may be useful for the diagnosis or monitoring of the response to antibiotic therapy.

 
This work was supported by the National Institutes of Health NIAID grants to D.N.F. (R01 AI061628) and J.M.M. (R01 AI052228).

The Fred Hutchinson Cancer Research Center has a pending patent application in this area.

 
* Corresponding author. Mailing address: Program in Infectious Diseases, D3-100, Vaccine and Infectious Diseases Institute, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue, North, Box 19024, Seattle, WA 98109. Phone: (206) 667-6706. Fax: (206) 667-4411. E-mail: dfredric{at}fhcrc.org

Published ahead of print on 14 January 2009.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Amsel, R., P. A. Totten, C. A. Spiegel, K. C. Chen, D. Eschenbach, and K. K. Holmes. 1983. Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic associations. Am. J. Med. 74:14-22.[CrossRef][Medline]
2
2. Bradshaw, C. S., S. N. Tabrizi, C. K. Fairley, A. N. Morton, E. Rudland, and S. M. Garland. 2006. The association of Atopobium vaginae and Gardnerella vaginalis with bacterial vaginosis and recurrence after oral metronidazole therapy. J. Infect. Dis. 194:828-836.[CrossRef][Medline]
3
3. Bukusi, E. A., C. R. Cohen, A. S. Meier, P. G. Waiyaki, R. Nguti, J. N. Njeri, and K. K. Holmes. 2006. Bacterial vaginosis: risk factors among Kenyan women and their male partners. Sex. Transm. Dis. 33:361-367.[CrossRef][Medline]
4
4. Evans, A. S. 1993. Causation and disease. Plenum Medical Book Company, New York, NY.
5
5. Ferris, M. J., J. Norori, M. Zozaya-Hinchliffe, and D. H. Martin. 2007. Cultivation-independent analysis of changes in bacterial vaginosis flora following metronidazole treatment. J. Clin. Microbiol. 45:1016-1018.[Abstract/Free Full Text]
6
6. Fredricks, D. N., T. L. Fiedler, and J. M. Marrazzo. 2005. Molecular identification of bacteria associated with bacterial vaginosis. N. Engl. J. Med. 353:1899-1911.[Abstract/Free Full Text]
7
7. Fredricks, D. N., T. L. Fiedler, K. K. Thomas, B. B. Oakley, and J. M. Marrazzo. 2007. Targeted PCR for detection of vaginal bacteria associated with bacterial vaginosis. J. Clin. Microbiol. 45:3270-3276.[Abstract/Free Full Text]
8
8. Fredricks, D. N., and D. A. Relman. 1999. Paraffin removal from tissue sections for digestion and PCR analysis. BioTechniques 26:198-200.[Medline]
9
9. Fredricks, D. N., and D. A. Relman. 1996. Sequence-based identification of microbial pathogens: a reconsideration of Koch's postulates. Clin. Microbiol. Rev. 9:18-33.[Abstract]
10
10. Gardner, H. L., and C. D. Dukes. 1955. Haemophilus vaginalis vaginitis: a newly defined specific infection previously classified "nonspecific" vaginitis. Am. J. Obstet. Gynecol. 69:962-976.[Medline]
11
11. Gardner, H. L., and C. D. Dukes. 1954. New etiologic agent in nonspecific bacterial vaginitis. Science 120:853.[Free Full Text]
12
12. Hillier, S. L., R. P. Nugent, D. A. Eschenbach, M. A. Krohn, R. S. Gibbs, D. H. Martin, M. F. Cotch, R. Edelman, J. G. Pastorek II, A. V. Rao, et al. 1995. Association between bacterial vaginosis and preterm delivery of a low-birth-weight infant. N. Engl. J. Med. 333:1737-1742.[Abstract/Free Full Text]
13
13. Hyman, R. W., M. Fukushima, L. Diamond, J. Kumm, L. C. Giudice, and R. W. Davis. 2005. Microbes on the human vaginal epithelium. Proc. Natl. Acad. Sci. USA 102:7952-7957.[Abstract/Free Full Text]
14
14. Khot, P. D., D. L. Ko, R. C. Hackman, and D. N. Fredricks. 2008. Development and optimization of quantitative PCR for the diagnosis of invasive aspergillosis with bronchoalveolar lavage fluid. BMC Infect. Dis. 8:73.[CrossRef][Medline]
15
15. Koumans, E. H., M. Sternberg, C. Bruce, G. McQuillan, J. Kendrick, M. Sutton, and L. E. Markowitz. 2007. The prevalence of bacterial vaginosis in the United States, 2001-2004; associations with symptoms, sexual behaviors, and reproductive health. Sex. Transm. Dis. 34:864-869.[CrossRef][Medline]
16
16. Marrazzo, J. M., L. A. Koutsky, D. A. Eschenbach, K. Agnew, K. Stine, and S. L. Hillier. 2002. Characterization of vaginal flora and bacterial vaginosis in women who have sex with women. J. Infect. Dis. 185:1307-1313.[CrossRef][Medline]
17
17. Marrazzo, J. M., K. K. Thomas, T. L. Fiedler, K. Ringwood, and D. N. Fredricks. 2008. Relationship of specific vaginal bacteria and bacterial vaginosis treatment failure in women who have sex with women. Ann. Intern. Med. 149:20-28.[Abstract/Free Full Text]
18
18. Myer, L., L. Denny, R. Telerant, M. Souza, T. C. Wright, Jr., and L. Kuhn. 2005. Bacterial vaginosis and susceptibility to HIV infection in South African women: a nested case-control study. J. Infect. Dis. 192:1372-1380.[CrossRef][Medline]
19
19. Ness, R. B., K. E. Kip, S. L. Hillier, D. E. Soper, C. A. Stamm, R. L. Sweet, P. Rice, and H. E. Richter. 2005. A cluster analysis of bacterial vaginosis-associated microflora and pelvic inflammatory disease. Am. J. Epidemiol. 162:585-590.[Abstract/Free Full Text]
20
20. Nugent, R. P., M. A. Krohn, and S. L. Hillier. 1991. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of Gram stain interpretation. J. Clin. Microbiol. 29:297-301.[Abstract/Free Full Text]
21
21. Schwebke, J. R., and L. F. Lawing. 2001. Prevalence of Mobiluncus spp among women with and without bacterial vaginosis as detected by polymerase chain reaction. Sex. Transm. Dis. 28:195-199.[CrossRef][Medline]
22
22. Sha, B. E., H. Y. Chen, Q. J. Wang, M. R. Zariffard, M. H. Cohen, and G. T. Spear. 2005. Utility of Amsel criteria, Nugent score, and quantitative PCR for Gardnerella vaginalis, Mycoplasma hominis, and Lactobacillus spp. for diagnosis of bacterial vaginosis in human immunodeficiency virus-infected women. J. Clin. Microbiol. 43:4607-4612.[Abstract/Free Full Text]
23
23. Sha, B. E., M. R. Zariffard, Q. J. Wang, H. Y. Chen, J. Bremer, M. H. Cohen, and G. T. Spear. 2005. Female genital-tract HIV load correlates inversely with Lactobacillus species but positively with bacterial vaginosis and Mycoplasma hominis. J. Infect. Dis. 191:25-32.[CrossRef][Medline]
24
24. Sobel, J. D. 1997. Vaginitis. N. Engl. J. Med. 337:1896-1903.[Free Full Text]
25
25. Taha, T. E., R. H. Gray, N. I. Kumwenda, D. R. Hoover, L. A. Mtimavalye, G. N. Liomba, J. D. Chiphangwi, G. A. Dallabetta, and P. G. Miotti. 1999. HIV infection and disturbances of vaginal flora during pregnancy. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 20:52-59.[Medline]
26
26. Verhelst, R., H. Verstraelen, G. Claeys, G. Verschraegen, J. Delanghe, L. Van Simaey, C. De Ganck, M. Temmerman, and M. Vaneechoutte. 2004. Cloning of 16S rRNA genes amplified from normal and disturbed vaginal microflora suggests a strong association between Atopobium vaginae, Gardnerella vaginalis and bacterial vaginosis. BMC Microbiol. 4:16.[CrossRef][Medline]
27
27. Zariffard, M. R., M. Saifuddin, B. E. Sha, and G. T. Spear. 2002. Detection of bacterial vaginosis-related organisms by real-time PCR for lactobacilli, Gardnerella vaginalis and Mycoplasma hominis. FEMS Immunol. Med. Microbiol. 34:277-281.[CrossRef][Medline]
28
28. Zozaya-Hinchliffe, M., D. H. Martin, and M. J. Ferris. 2008. Prevalence and abundance of uncultivated Megasphaera-like bacteria in the human vaginal environment. Appl. Environ. Microbiol. 74:1656-1659.[Abstract/Free Full Text]

---

Journal of Clinical Microbiology, March 2009, p. 721-726, Vol. 47, No. 3
0095-1137/09/$08.00+0     doi:10.1128/JCM.01384-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Fredricks, D. N. Articles by Marrazzo, J. M. Search for Related Content PubMed PubMed Citation Articles by Fredricks, D. N. Articles by Marrazzo, J. M.