BACTERIAL VAGINOSIS (BV) IS A clinical condition characterized by a shift in normal vaginal flora from aerobic to a predominantly anaerobic flora. It has been associated with serious sequelae such as spontaneous abortion,1 preterm delivery,1 and increased susceptibility to HIV and other sexually transmitted infections (STIs)2; however, more than 50% of women with BV are asymptomatic. Despite an increasing understanding of the pathogenesis, the etiology of BV remains unknown; and although BV has been associated with certain hormonal factors3 and sexual practices commonly observed with STIs,4–6 it remains unclear whether BV is the result of a sexually transmitted agent.
A number of bacterial organisms have been implicated in the etiology of BV, including Gardnerella vaginalis and Mobiluncus spp.; however, recently, a metronidazole-resistant anaerobe, Atopobium vaginae, has been reported in association with BV, which appears to be uncommon in women with normal flora.7,8 To examine the relationship of BV-associated organisms with sexual activity, we examined the prevalence of G. vaginalis and A. vaginae in vaginal samples from 44 asymptomatic women who had not yet initiated penetrative vaginal sex.
Female patients who were considered unlikely to have initiated vaginal sex by attending medical practitioners were offered participation in the study.9 The patients were recruited in 1991 for a study of assessment of the sexual transmission of human papillomavirus.9 The mean age of these women was 18 years (range, 13–41 years). The ethics committee for human research at the Royal Women's Hospital approved the study and each participant gave informed written consent. Each participant was given an envelope containing a tampon, a specimen jar with sterile transport medium, and a self-answer questionnaire; participants were identified by a number only. Each woman was asked to collect a tampon (in and out) specimen, place it into the specimen jar provided,9 and to complete the questionnaire.
Two questionnaires were used. A brief questionnaire asked if the participant had had sexual intercourse involving vaginal penetration. A more detailed questionnaire also asked about other forms of sexual activity, including contact with a partner's hands and genitals, oral sex, and genital-genital contact not involving penetration. The individual recruiting practitioner decided which questionnaire to give the participant on the basis of whether they felt the participant would be comfortable with the more detailed questionnaire. Questionnaires were completed in private, placed in an envelope and sealed, and were not seen by the recruiting doctor. Consent forms and questionnaires were not linked.
Sample Preparation and DNA Amplification
The tampon cell pellet was processed previously from the original study and stored at −70°C was used. A 200-μL aliquot of tampon pellet was extracted using the automated MagNA Pure LC (Roche Diagnostic, Mannheim, Germany) with the DNA Isolation Kit I protocol. Extracted DNA was amplified by polymerase chain reaction (PCR) for 3 targets by real-time PCR (5-μL aliquots of DNA for each reaction). Amplification and detection was carried out for the β-globin gene as an internal control to assess inhibition, sample adequacy, and integrity10 and as described previously for G. vaginalis.11 Amplification of A. vaginae was directed at a 430-bp region of the 16S rRNA gene sequence using primer AV3R12 and modified Ato-0291 primer13 by addition of 2 G to the 3′ end. In addition, 5′ nuclease probe nested for this region 5′FAM CAGATTTAACTCCTGACCTAACAGACC TAMARA 3′ was also designed. Amplification reaction also consisted of 1× Lightcycler Fast Start Reaction Mix (Roche Diagnostic) containing 3 mmol/L MgCl2 (final concentration). The samples were heated at 95°C for 10 minutes and cycled 50 times using parameters of 95°C for 0 seconds and 60°C for 60 seconds. All PCR assays included a positive and negative control and have been shown to be highly specific for the target amplified with analytical sensitivity of 10 copies per reaction.
Forty-four of the initial 55 samples from virginal women reported in 1991 had adequate stored samples available for analysis.9 Of these, 27 had answered the detailed questionnaire and all reported nonpenetrative sexual activity. The mean age of those answering the short questionnaire was 16 years compared with 20 years for those answering the detailed questionnaire (P = 0.003).
Of the 44 virginal women, 20 (45%) had G. vaginalis and 3 (7%) had A. vaginae detected. Among the 27 who completed the detailed questionnaire, 12 (46%) had G. vaginalis and none had A. vaginae detected. G. vaginalis was significantly more likely to be detected in those who had participated in oral sex (odds ratio [OR] = 22; 95% confidence interval [CI] = 2.2–222) and hand-genital contact without penetration (P = 0.02) but not genital-genital contact (Table 1).
An analysis stratified by whether participants had had hand-genital contact is shown in Table 2. All participants who had in engaged in oral sex had also engaged in hand-genital contact. The proportion of participants who had G. vaginalis was not significantly higher among those who had had both oral sex and hand-genital contact compared with those who had only hand-genital contact (OR = 8.8; 95% CI = 0.8–100).
Our study suggests that both G. vaginalis and A. vaginae are present in women who have not engaged in penetrative vaginal intercourse and is the first study to demonstrate that the presence of G. vaginalis is associated with nonpenetrative intimate sexual contact in virginal women. It was not possible to determine whether it was oral sex or hand-genital contact that was associated with the detection of G. vaginalis because both were highly correlated. Nevertheless, our results suggest that G. vaginalis is common and that routine transmission may occur before the onset of penetrative vaginal sex.
Our study has a number of limitations. First, it relied on self-report to both establish the virginal status of the participants and the nature of their nonpenetrative contact. The fact that questionnaires were not numbered or linked to consent forms, and never seen by the treating doctor, would increase the likelihood of accurate answers. In addition, the absence of any HPV infection,9 which is a common infection in sexually active women, supports the virginal status of these participants.
Second, the sample size of this study was small, involving only 44 women of whom 27 answered the detailed questionnaire. This meant that the study had limited power to detect a difference and in particular may explain why the odds ratio of 8.8 for G. vaginalis among those who reported both oral sex and hand-genital contact was not significant. However, our observation in the unstratified analysis, that oral sex and hand-genital contact were associated with G. vaginalis, even with small numbers, indicates the strength of the association. All women were asymptomatic, but grading of vaginal flora was not performed, so we are unable to comment on the presence or absence of BV in the study population. In addition, the absence of an age-matched group who had engaged in penetrative vaginal sex means we are unable to compare the prevalence of these organisms between adolescents who have and have not engaged in penetrative vaginal sex. Lastly, the specimens that were analyzed were collected 15 years ago, and it is possible that changes in behavior over this time may have changed the pattern of infection among women today.
A number of studies have assessed the prevalence of G. vaginalis among nonsexually active or virginal women. However, the definition of sexual activity has often been restricted to penetrative vaginal sex, and there has been failure to report on nonpenetrative sexual practices. Studies have yielded conflicting results. Shafer et al reported that up to one third of nonsexually active adolescents harbored G. vaginalis but significantly less than in sexually active adolescents (60%) (P = 0.03),14 whereas a study of 120 asymptomatic 14- to 17-year-old females found no significant difference in the prevalence of G. vaginalis (17%) detected by culture in young females reporting no penetrative sexual activity (n = 52) compared with sexually active females (34%).15 No information on nonpenetrative sexual practices such as oral sex or hand-genital contact was reported in either study.
To our knowledge, there have been no previous studies examining the prevalence of A. vaginae in virginal women. Studies conducted in adult women with normal vaginal flora have reported a prevalence of A. vaginae detected by PCR that ranges from 8% to 20%, significantly lower than that reported for women with BV, 54% to 78%.7,8 Among the 44 virginal women in our study, only 3 had A. vaginae (7%); this was not significantly different from the prevalence of 12% seen in 173 sexually active women aged 17 to 54 years of age with normal flora (Nugent score 0–3) attending our clinical service in 2002–2004 (CS Bradshaw, personal communication); however, we do not have data on a comparative adolescent group. We were unable to determine if the 3 virginal women had engaged in nonpenetrative sexual practices, because they did not answer the detailed questionnaire.
G. vaginalis and A. vaginae are present in women who have not engaged in penetrative vaginal intercourse, and G. vaginalis was associated with nonpenetrative intimate sexual contact in virginal women. Our results suggest that transmission of these organisms may occur before the onset of penetrative vaginal sex, and although a number of studies have evaluated vaginal microflora in sexually experienced and inexperienced adolescent females, none have reported an association with specific nonpenetrative sexual practices.14,15 Finally, our results should be seen as hypothesis-generating only because of the small sample size, biases associated with analyzing specimens collected to test a different hypothesis, and specimens that were 15 years old. We would encourage others to undertake a prospective study designed specifically to test the hypothesis.
1. Hay PE, Lamont RF, Taylor-Robinson D, et al. Abnormal bacterial colonisation of the genital tract and subsequent preterm delivery and late miscarriage. BMJ 1994; 308:295–298.
2. Sewankambo N, Gray RH, Wawer MJ, et al. HIV-1 infection associated with abnormal vaginal flora morphology and bacterial vaginosis. Lancet 1997; 350:546–550.
3. Keane FE, Ison CA, Taylor-Robinson D. A longitudinal study of the vaginal flora over a menstrual cycle. Int J STD AIDS 1997; 8:489–494.
4. Smart S, Singal A, Mindel A. Social and sexual risk factors for bacterial vaginosis. Sex Transm Infect 2004; 80:58–62.
5. Nilsson U, Hellberg D, Shoubnikova M, et al. Sexual behavior risk factors associated with bacterial vaginosis and Chlamydia trachomatis
infection. Sex Transm Dis 1997; 24:241–246.
6. Bradshaw CS, Morton AN, Garland SM, et al. Behavioural practices associated with bacterial vaginosis in comparison to vaginal candidiasis: Implications for a sexually transmitted pathogen. Obstet Gynecol 2005; 106:105–114.
7. Ferris MJ, Masztal A, Aldridge KE. Association of Atopobium vaginae
, a recently described metronidazole resistant anaerobe, with bacterial vaginosis. BMC Infect Dis 2004; 4:5.
8. Verhelst R, Verstraelen H, Claeys G, et al. 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 2004; 4:16.
9. Fairley CK, Chen S, Tabrizi SN, et al. The absence of genital human papillomavirus DNA in virginal women. Int J STD AIDS 1992; 3:414–417.
10. Tabrizi SN, Cohenford MA, Lentrichia BB, et al. Evaluation of real-time polymerase chain reaction assays for confirmation of Neisseria gonorrhoeae
in clinical samples tested positive in the Roche COBAS AMPLICOR assay. Sex Transm Dis 2004; 80:68–71.
11. Zariffard MR, Saifuddin M, Sha BE, et al. Detection of bacterial vaginosis-related organisms by real-time PCR for Lactobacilli
, Gardnerella vaginalis
and Mycoplasma hominis.
FEMS Immunol Med Microbiol 2002; 34:277–281.
12. Ferris MJ, Masztal A, Martin DH. Use of species-directed 16S rRNA gene PCR primers for detection of Atopobium vaginae
in patients with bacterial vaginosis. J Clin Microbiol 2004; 42:5892–5894.
13. Harmsen HJ, Wildeboer-Veloo AC, Grijpstra J, et al. Development of 16S rRNA-based probes for the Coriobacterium
group and the Atopobium
cluster and their application for enumeration of Coriobacteriaceae
in human feces from volunteers of different age groups. Appl Environ Microbiol 2000; 66:4523–4527.
14. Shafer MA, Sweet RL, Ohm-Smith MJ, et al. Microbiology of the lower genital tract in postmenarchal adolescent girls: Differences by sexual activity, contraception, and presence of nonspecific vaginitis. J Pediatr 1985; 107:974–981.
15. Bump RC, Sachs LA, Buesching WJ 3rd. Sexually transmissible infectious agents in sexually active and virginal asymptomatic adolescent girls. Pediatrics