Bacterial Vaginosis (BV) and the Risk of Incident Gonococcal or Chlamydial Genital Infection in a Predominantly Black Population

Abstract

Objective: 

The objective of this study was to assess in prospective data whether bacterial vaginosis (BV) is associated with gonococcal/chlamydial cervicitis.

Study: 

A total of 1179 women at high risk for sexually transmitted infections was followed for a median of 3 years. Every 6 to 12 months, vaginal swabs were obtained for Gram stain, culture of microflora, and Neisseria gonorrhoeae and Chlamydia trachomatis. A Gram stain score of 7 to 10 based on the Nugent criteria categorized BV.

Results: 

Baseline BV was associated with concurrent gonococcal/chlamydial infection (adjusted odds ratio, 2.83; 95% confidence interval [CI], 1.81–4.42). However, the association between BV and subsequent, incident gonococcal/chlamydial genital infection was not significant (adjusted relative risk [RR], 1.52; 95% CI, 0.74–3.13). Dense growth of pigmented, anaerobic Gram-negative rods (adjusted RR, 1.93; 95% CI, 0.97–3.83) appeared to elevate the risk for newly acquired gonococcal/chlamydial genital infection.

Conclusions: 

BV was common among a predominantly black group of women with concurrent gonococcal/chlamydial infection but did not elevate the risk for incident infection.

In Brief

Bacterial vaginosis, albeit common in a group of young, sexually active women, did not elevate the risk for developing incident gonococcal or chlamydial cervicitis.

BACTERIAL VAGINOSIS (BV) IS CHARACTERIZED by a 10⁴-fold increase in the numbers of vaginal bacteria and by a disequilibrium in vaginal microflora such that the normally predominant hydrogen peroxide-producing strains of lactobacilli are reduced and facultative and anaerobic organisms are increased.¹ The production of lactic acid by lactobacilli has long been considered the basis of a protective role of these organisms against vaginal infection and anaerobic overgrowth. Strains of hydrogen peroxide-producing lactobacilli may inhibit the growth of genital microorganisms.^2–4 Hydrogen peroxide-producing lactobacilli have been shown in vitro to kill bacteria including Gardnerella vaginalis and Prevotella bivia as well as Neisseria gonorrhoeae.

Chlamydia trachomatis and N. gonorrhoeae ascending from the lower to the upper genital tract are thought to be the most common precipitants of pelvic inflammatory disease (PID).⁵ These 2 pathogens can be isolated from the cervix or upper genital tract in approximately two thirds of PID cases on average.⁶ Although several studies have demonstrated that women with BV are more likely to have PID and that the presence of C. trachomatis and N. gonorrhoeae in the upper genital tract is frequently accompanied by anaerobic and facultative bacteria,^7–11 few have directly examined the link between BV and gonococcal/chlamydial genital infection.^2,12,13

We conducted a multicenter, prospective observational cohort study designed to examine whether, among women at high risk for sexually transmitted infections, BV is more common among women with concurrent gonococcal/chlamydial genital infection and whether BV is associated with the acquisition of incident infections by gonorrhea and chlamydia.

Materials and Methods

Patient Selection

The methods of subject enrollment, data collection, and follow-up have been reported in detail elsewhere.¹⁴ Briefly, women 13 to 36 years of age were recruited into the GYN Infections Follow-Through (GIFT) Study from 5 U.S. sites between May 1999 and June 2001. Human subjects approval was obtained at each participating institution and all women signed informed consent. Women were eligible for the GIFT study if they were not specifically seeking care for a sexually transmitted disease yet were considered at high risk based on a previous risk stratification paradigm for chlamydial cervicitis.¹⁵ Specifically, to be enrolled, a woman had to have a score of 3 points or more on an algorithm in which points were derived as follows: age 24 or less = 1; black race = 2; never pregnant = 1; 2 or more sexual partners = 1; douches at least once per month = 2; and any prior sexually transmitted infection, including N. gonorrhoeae, C. trachomatis, and Trichomonas vaginalis = 2. Of 2740 women screened for study entry, 853 (31.1%) did not meet these inclusion criteria. An additional 259 (9.5%) women were excluded on the basis of a priori criteria such as being pregnant, married, or virginal, or being on antibiotics at baseline. Among the 1628 women who were eligible for the study, 1179 (72.4%) completed a questionnaire and evaluation of their vaginal flora at baseline and are the focus of these analyses.

Microbiologic Methods for Evaluation of the Vaginal Flora

At baseline, study staff taught each subject how to self-obtain vaginal specimens with a Q-tip-like swab using a validated method.¹⁶ Smears from these swabs were Gram-stained and a score of 0 to 10 was assigned by laboratory staff, masked to any subject characteristics, in light of the relative proportions of large Gram-positive rods (lactobacilli), small Gram-negative or gram-variable rods (Bacteroides or Gardnerella), and curved Gram-variable rods (Mobiluncus).¹⁷ The results were scored by a standardized method as described by Nugent.¹⁷ A score of 0 to 3 was interpreted as consistent with normal vaginal flora; a score of 4 to 6, corresponding to disturbed flora, was designated as intermediate; and a score of 7 to 10 was considered to be BV.

Two swabs, placed in an anaerobic transport vial, were also shipped to the microbiology laboratory for characterization of the following: Lactobacillus species, anaerobic Gram-negative rods, Gardnerella vaginalis, group B streptococcus, Enterococcus species, Escherichia coli, Candida species, Mycoplasma hominis, and Ureaplasma urealyticum. Lactobacilli were identified to the genus level on the basis of Gram stain morphology and production of lactic acid.¹⁸ The amount of growth for each of these microorganisms was recorded on a semiquantitative scale from 0 to 4.

DNA Amplification for Neisseria gonorrhoeae and Chlamydia trachomatis

DNA amplification for N. gonorrhoeae and C. trachomatis was performed using a strand displacement DNA Amplification (SDA) Assay (Becton Dickinson, Sparks, MD) from self-obtained vaginal swabs. All positive test results for gonococcal or chlamydial infection were reported to the clinical sites within 1 week of enrollment where infected subjects were treated.

Follow-Up

Among the 1179 subjects who had baseline assessment of BV, 26 (2.2%) had a baseline visit only and 6 (0.5%) had missing data on covariates used for statistical adjustment (see “Statistical Analysis” section). Among the remaining 1147 subjects, the median length of follow-up was 3.0 years (interquartile range: 2.4–3.4 years) and 88% of women were interviewed at their final, regularly scheduled contact.

Other Data Collection

Women were asked about demographic factors, including age, race, education, and income. They also reported relevant lifestyle behaviors such as tobacco smoking, number of sexual partners in the past 2 months, acquisition of a new partner in the past 2 months, contraception use, and sex during menses. Furthermore, they were requested to recall past episodes of sexually transmitted infections, including PID and gonococcal and/or chlamydial genital infections.

Statistical Analysis

To assess the association between BV and gonococcal/chlamydial infection, and the influence of other microbes (H₂O₂-producing lactobacillus, M. hominis, G. vaginalis, anaerobic Gram-negative rods–pigmented), both cross-sectional and prospective analyses were conducted. In cross-sectional analyses, logistic regression was used to estimate adjusted odds ratios of gonococcal/chlamydial infection at baseline in relation to BV and other microbes. Generalized estimating equations (GEEs) were similarly used among: 1) follow-up clinic visits that excluded the baseline visit, and 2) all clinic visits. This approach takes into account the correlation among multiple observations (clinic visits) per subject.^19,20

In prospective analyses conducted at the visit level of observation, discrete-time proportional hazards models²¹ fit by pooled logistic regression²² were used to assess the time-varying impact of BV and other microbes on acute risk of gonococcal/chlamydial infection. The “critical exposure” visit to estimate an acute effect from time-varying microbial status was the visit that immediately preceded the diagnosis of gonococcal/chlamydial infection and that occurred within 6 months of the diagnosis. All other visits among subjects who either did or did not experience gonococcal/chlamydial infection were considered to represent etiologically irrelevant exposure periods.

In all analyses, estimates of risk of gonococcal/chlamydial infection in relation to microbial status were adjusted for age, race, education, and history of PID, each of which was independently associated with gonococcal/chlamydial infection as determined by stepwise Cox regression. Clinical site findings were assessed and found not to be an effect modifier or an appreciable source of confounding.

Results

Baseline Subject Characteristics

Participants were predominantly 19 to 24 years of age (66%), black (75%), and with a household income of less than $20,000 (74%). At baseline, 428 (36%) women had a Gram-stained vaginal smear consistent with normal vaginal flora, 280 (24%) had intermediate flora, and 471 (40%) had BV. Forty-five percent of women had a history of gonococcal/chlamydial infection at study entry. As we have reported previously, detection of BV at baseline was significantly more common among women of black race; with less than a high school education and a low family income; and who were current smokers, had past PID and gonococcal/chlamydial infections, and had sex during menses.²³

Vaginal Microflora and Concomitant Gonococcal/Chlamydial Genital Infection

In cross-sectional analyses, after adjustment for age, race, years of education, history of PID, and time of follow-up, identification of gonococcal cervicitis, chlamydial cervicitis, or both was more common among women with intermediate vaginal flora (adjusted odds ratio [OR], 1.57–1.75) than women with normal flora and was most common among women with BV (adjusted OR, 1.93–2.83) (Table 1). This was true when the cross-sectional analysis occurred at baseline, at follow-up visits limited to the first episode of detected gonorrhea/chlamydial, or at any visit. Similarly, significant cross-sectional associations were observed between gonorrhea/chlamydia and lack of H₂O₂-producing lactobacilli, M. hominis, G. vaginalis, and pigmented anaerobic Gram-negative rods.

TABLE 1: Cross-Sectional Analyses of Microbes and Their Relation to Documented Gonorrhea/Chlamydia

Vaginal Microflora Immediately Preceding Incident Gonococcal/Chlamydial Infection

BV at the visit immediately preceding (but not concurrent with) the first episode of genital infection did not significantly elevate the risk for acquiring incident N. gonorrhoeae/C. trachomatis infection after adjustment for relevant compounding factors (adjusted relative risk [RR], 1.52; 95% confidence interval [CI], 0.74–3.13) (Table 2). Moreover, after adjustment, there was no significant trend toward alteration of vaginal microecology (normal to intermediate to BV) in relation to gonococcal/chlamydial cervicitis (P = 0.25). Similarly, absence of H₂O₂-producing lactobacilli and denser growth of G. vaginalis and M. hominis did not significantly increase the risk for acquiring gonococcal/chlamydial infection. However, denser growth of pigmented, anaerobic Gram-negative rods (3–4+ growth) appeared to raise the risk for a subsequent, first episode of gonococcal/chlamydial genital infection (adjusted RR, 1.93; 95% CI, 0.97–3.83; P = 0.09).

TABLE 2: Cohort Analysis of Time-Varying Effect of Microbial Status and Subsequent Risk of Gonorrhea/Chlamydia (GC/CT) Within 6 Months

Discussion

In this cohort of high-risk women, BV was significantly associated with concurrent gonococcal/chlamydial genital infection but not significantly associated with the development of these infections. A stronger trend toward incident risk of infection was observed for pigmented, anaerobic Gram-negative rods. Aerobic Gram-negative rods, pigmented, are a category of microorganisms typically including the species Porphyromonas, Prevotella, and Bacteroides.

An association between BV and gonococcal/chlamydial cervicitis has been demonstrated in a limited number of cross-section studies of pregnant women^13,24 and nonpregnant women^2,12 and in 1 prospective cohort study of female sex workers in Kenya.²⁵ Weissenfeld et al. studied 255 nonpregnant women who reported recent sexual contact with a gonococcal- or nongonococcal urethritis-diagnosed male partner.¹² Women with BV, as compared with normal vaginal flora, were 4 times more likely to test positive for N. gonorrhoeae (by culture) and 3 times more likely to test positive for C. trachomatis (by DNA amplification). Women with hydrogen peroxide-producing lactobacilli were less likely to have concurrent gonorrhea or chlamydia. Similarly, Moi, in a study of 2128 women attending a sexually transmitted disease clinic, found BV to be associated with gonococcal/chlamydial infection.²⁶ These findings confirmed an earlier cross-sectional study.²

However, these research designs cannot exclude the possibility that aberrancies within the vaginal flora follow, rather than precede the bacterial sexually transmitted infections, i.e., gonorrhea and chlamydia, that initiate PID. Similarly, studies detecting BV-related bacteria in the upper genital tract of women with PID^8,11,27–29 might represent an overgrowth subsequent to, rather than before, gonococcal or chlamydial upper genital tract infection.

In the one other prospective study to examine vaginal flora and incident sexually transmitted infections, absence of lactobacilli was found to increase the risk for acquiring N. gonorrhoeae (adjusted hazard ratio, 1.7; 95% CI, 1.1–2.6) but not C. trachomatis.²⁵ Abnormal vaginal flora (BV score >4) did not significantly increase the risk for gonorrhoeae or chlamydia acquisition. Our study did not separate gonorrhea from chlamydia as a result of small numbers of women with gonorrhea. Because chlamydia was 3 to 4 times more common in our population, our results generally replicate, and extend to other vaginal microflora, those in the previous prospective study.

The strongest risk found in our data was for pigmented, anaerobic Gram-negative rods. This is consistent with a precious study in which Hillier et al. demonstrated an association between anaerobic Gram-negative rods and histologic endometritis, independent of BV.³⁰ This suggests that specific components of the microecology that comprises BV may be more predictive for developing morbidity than BV per se.

Strengths of our study include the large number of women studies, enrollment of a high-risk population enhancing study power, use of consistent and standardized enrollment data collection protocols, collection of biomarkers of effect, and relatively long-term and complete longitudinal data collection, which permitted assessment of BV status as a time-varying, as well as a fixed, baseline characteristic. Weaknesses include the observational nature of the study, making it impossible to exclude unmeasured confounding. Furthermore, the relatively long intervals between vaginal microbiologic assessments allow for a somewhat gross assessment of the impact of variation of vaginal flora over time. Finally, although we treated known gonococcal/chlamydial lower genital infections, we did not perform “test of cure” and therefore some infections detected during follow up may have been prevalent rather than incident. We did not provide notification of diagnosis or treatment for bacterial vaginosis, yet some women may have independently sought BV treatment, an intervention we did not record. This may have resulted in some underestimate of the association between BV and gonococcal/chlamydial genital infection.

In summary, among predominantly young, black women followed longitudinally, we found BV and gonococcal/chlamydial infection to coexist, whereas the evidence that BV preceded these bacterial sexually transmitted infections was not strong.