Vaginal Lactobacilli, Microbial Flora, and Risk of Human Immunodeficiency Virus Type 1 and Sexually Transmitted Disease Acquisition (original) (raw)

Journal Article

,

1

Departments of Medicine, University of Washington

,

Seattle

Search for other works by this author on:

,

2

Departments of Biostatistics, University of Washington

,

Seattle

Search for other works by this author on:

,

4

Department of Medical Microbiology, University of Nairobi

Search for other works by this author on:

,

3

Departments of Epidemiology, University of Washington

,

Seattle

Search for other works by this author on:

,

6

Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh

,

Pittsburgh, Pennsylvania

Search for other works by this author on:

,

5

Coast Provincial General Hospital

,

Mombasa, Kenya

Search for other works by this author on:

,

5

Coast Provincial General Hospital

,

Mombasa, Kenya

Search for other works by this author on:

,

4

Department of Medical Microbiology, University of Nairobi

Search for other works by this author on:

,

4

Department of Medical Microbiology, University of Nairobi

Search for other works by this author on:

1

Departments of Medicine, University of Washington

,

Seattle

3

Departments of Epidemiology, University of Washington

,

Seattle

Reprints or correspondence: Dr. Joan K. Kreiss, Depts. of Epidemiology and Medicine, University of Washington, 325 Ninth Ave., Box 359909, Seattle WA 98104-2499 (iartp@u.washington.edu).

Search for other works by this author on:

a

Current affiliation: Park Nicollet Clinic, Minneapolis, MN.

Author Notes

Revision received:

13 August 1999

Published:

01 December 1999

• PDF
• Split View
• • Article contents
  • Figures & tables
  • Video
  • Audio
  • Supplementary Data
• Cite

Cite

Harold L. Martin, Barbra A. Richardson, Patrick M. Nyange, Ludo Lavreys, Sharon L. Hillier, Bhavna Chohan, Kishorchandra Mandaliya, J. O. Ndinya-Achola, Job Bwayo, Joan Kreiss, Vaginal Lactobacilli, Microbial Flora, and Risk of Human Immunodeficiency Virus Type 1 and Sexually Transmitted Disease Acquisition, The Journal of Infectious Diseases, Volume 180, Issue 6, December 1999, Pages 1863–1868, https://doi.org/10.1086/315127
Close

Navbar Search Filter Mobile Enter search term Search

Abstract

A prospective cohort study was conducted to examine the relationship between vaginal colonization with lactobacilli, bacterial vaginosis (BV), and acquisition of human immunodeficiency virus type 1 (HIV-1) and sexually transmitted diseases in a population of sex workers in Mombasa, Kenya. In total, 657 HIV-1—seronegative women were enrolled and followed at monthly intervals. At baseline, only 26% of women were colonized with Lactobacillus species. During follow-up, absence of vaginal lactobacilli on culture was associated with an increased risk of acquiring HIV-1 infection (hazard ratio [HR], 2.0; 95% confidence interval [CI], 1.2–3.5) and gonorrhea (HR, 1.7; 95% CI, 1.1–2.6), after controlling for other identified risk factors in separate multivariate models. Presence of abnormal vaginal flora on Gram's stain was associated with increased risk of both HIV-1 acquisition (HR, 1.9; 95% CI, 1.1–3.1) and Trichomonas infection (HR, 1.8; 95% CI, 1.3–2.4). Treatment of BV and promotion of vaginal colonization with lactobacilli should be evaluated as potential interventions to reduce a woman's risk of acquiring HIV-1, gonorrhea, and trichomoniasis.

The human vaginal ecosystem is dominated by Lactobacillus species [1, 2]. Lactobacilli are gram-positive rods that, in vitro, produce substances with antimicrobial properties, including lactacidin, acidolin, lactacin B, and hydrogen peroxide (H2O2) [3]. H2O2-producing lactobacillus strains play a pivotal role in controlling the microenvironment of the vagina and in inhibiting the overgrowth of potentially pathogenic organisms [3–7]. In vitro, H2O2-producing lactobacilli are cidal to human immunodeficiency virus type 1 (HIV-1), perhaps because of the reaction of H2O2 with myeloperoxidase and halides present in vaginal fluid [8].

Bacterial vaginosis (BV) is a clinical condition that is characterized by decreased H2O2-producing lactobacilli and increased concentrations of anaerobic gram-negative rods, Gardnerella species, and genital mycoplasmas [1, 7, 9, 10]. In some studies, BV was associated with pelvic inflammatory disease [11] and preterm delivery of low-birth-weight infants [12]. Treatment of BV decreases the incidence of preterm delivery in women at risk for preterm birth [13].

Several observational studies among different populations have found associations between BV and HIV-1. A cross-sectional survey of female sex workers in Thailand found a positive association between a clinical diagnosis of BV and HIV-1 seropositivity [14]. More recently, cross-sectional studies of rural women in Uganda and pregnant women in Malawi both demonstrated a positive association between abnormal vaginal flora and HIV-1 seropositivity and a dose-response relationship between the severity of disturbance of the vaginal flora and HIV-1 serostatus [15, 16]. A recent longitudinal study of pregnant women in Malawi found that clinical BV was associated with increased risk of HIV-1 seroconversion during pregnancy and after delivery [17].

With regard to sexually transmitted diseases (STDs), a cross-sectional survey of pregnant women in the United States and a study of nongravid women in Sweden found associations between BV and Neisseria gonorrhoeae [9, 18]. A cross-sectional study of pregnant women in the United States found an inverse association between presence of lactobacilli and Chlamydia trachomatis, and H2O2-producing isolates were inversely associated with isolation of Mycoplasma hominis and Ureaplasma urealyticum [19]. There was also a trend for an inverse association between Lactobacillus isolation and presence of Trichomonas vaginalis [19]. To our knowledge, only 1 prospective study has been published on the association of clinical BV and HIV-1 risk, and no groups have reported data regarding Lactobacillus culture results and HIV-1 and STD risk [17]. Our prospective cohort study was designed to examine the association between vaginal colonization with lactobacilli, abnormal vaginal flora on Gram's stain, and the acquisition of HIV-1 and STDs.

Methods

Female sex workers in Mombasa, Kenya, attending a municipal STD clinic for routine check-ups were offered HIV-1 counseling and serologic testing. Women testing negative for HIV-1 antibodies were offered enrollment into a prospective cohort study. Details of the cohort have been described [20]. In brief, participants underwent structured interview, physical examination, pelvic examination, and STD screening. They were asked to return in 1 week for laboratory results and monthly thereafter for follow-up examination. At each follow-up visit, subjects were interviewed regarding interim sexual behavior, condom use, and physical symptoms, and physical examination, pelvic examination, STD screening, and HIV-1 serologic testing were performed. Any STD detected was treated according to the treatment guidelines of the Kenya Ministry of Health. Clinical BV, as defined by symptomatic characteristic vaginal discharge, pH >4.5, and presence of clue cells, was treated with oral metronidazole. Risk reduction counseling was conducted at each visit, and free condoms were provided.

Laboratory methods

STD screening consisted of vaginal wet preparation and potassium hydroxide microscopy for T. vaginalis, Candida species, and clue cells. Endocervical secretions were cultured for N. gonorrhoeae on modified Thayer-Martin media and tested for C. trachomatis antigen by EIA (Micro Trak II; Syva, San Jose, CA). Syphilis serologic testing was conducted by rapid plasma reagin (Becton Dickinson, Cockeysville, MD) and Treponema pallidum hemagglutination assay (Biotech Laboratories, Lightwater, UK). HIV-1 serology was done by ELISA (Detect HIV; BioChem ImmunoSystems, Montreal). Any sample testing positive on screening ELISA was confirmed by a second ELISA (Recombigen; Cambridge Biotech, Worcester, MA), and all incident HIV-1 infections were confirmed by HIV-1 Western blot (Cambridge Biotech).

Vaginal Gram's stains were prepared for evaluation of the vaginal flora by rolling a Dacron swab across the upper lateral wall of the vagina and rolling it onto a microscope slide, where it was allowed to air dry. Gram's stain scores of 0—10 were assigned on the basis of standardized criteria [21], with 0 representing a _Lactobacillus_-predominant flora and 10 a flora dominated by Gardnerella, Bacteroides, and Mobiluncus-like organisms (small gram-negative or variable or curved rods) [9]. Scores of 0–3 were considered normal, 4–6 borderline for BV, and 7–10 diagnostic of BV [21]. A score ⩾4 was used to define abnormal vaginal flora. Because of the vaginal hygiene practices in this population, vaginal discharge was rarely seen, and a clinical diagnosis of BV could not be made reliably. Thus, for all analyses in this study, BV was defined by laboratory criteria by use of the Gram's stain criteria described above. The definition of normal flora was based on studies done with Western women, and it is this definition of normal that was used as the comparison group for this population of African women.

Lactobacillus cultures were obtained by rolling a Dacron swab across the upper lateral wall of the vagina and inoculating directly onto Rogosa agar plates. The plates were stored in candle jars for transport to the laboratory, where they were incubated in anaerobic jars at 37°C for ⩾72 h. Colonies growing on the Rogosa agar were identified as Lactobacillus species, on the basis of colonial morphology and Gram's stain appearance. Three Lactobacillus colonies from each plate were stored in skim milk at −20°C for batch testing for H2O2 production. H2O2 production was tested by subculturing the Lactobacillus isolate on tetramethylbenzidine (TMB) agar containing horseradish peroxidase [1]. After anaerobic incubation of the subculture for 72 h, the culture plate was exposed to room air, at which time the appearance of a blue pigment indicated oxidation of TMB by horseradish peroxidase in the presence of H2O2. Any Lactobacillus isolate having blue pigment detectable by the naked eye was classified as an H2O2-producing isolate.

Data analysis

All data were collected on standardized forms and were double entered on a 486 desktop computer with a data entry system (SPSS, Chicago) and then verified by individual line listing. Spearman correlation coefficients were used to determine the correlation between lactobacilli culture and Gram's stain scores. Cox proportional hazards regression was used to examine the relationship between Lactobacillus status and HIV-1 infection. Andersen-Gill style proportional hazards regression with robust variance estimates were used for STDs and other recurring events [22]. Data were analyzed by use of SPSS and S-Plus (StatSci, Division of MathSoft, Seattle) software.

Because the presence or absence of lactobacilli in the vaginal environment is dynamic [7, 23], a time-dependent summary measure of Lactobacillus status was created by combining a subject's Lactobacillus results at the current visit and the prior visit. For the regression analyses, the risk category was defined by the absence of detectable lactobacilli at both the current and prior visits, and the reference category was defined by the presence of lactobacilli at either the current or prior visit (or both). Multivariate models of Lactobacillus status, Gram's stain score, and STD outcomes were created by including all covariates that were associated with the outcome of interest in univariate regression analysis with P ⩽ .05.

Results

Study population characteristics

This analysis included 657 women who were part of an ongoing prospective cohort study in Mombasa. A total of 621 person-years of follow-up were accumulated during 3805 follow-up visits. Subjects were followed a median of 6.4 months (range, 0.6–42.5); the median intervisit interval was 35 days. The median number of follow-up visits was 3 (range, 1–33). The baseline demographic and sexual exposure characteristics of this population are shown in table 1. The median percentage of condom use was 100% (range, 0%–100%), and 37% of women used condoms for <75% of sexual encounters. Only 1% of women used vaginal drying agents, but internal vaginal cleansing was practiced by 96% of women, and 77% used soap, detergents, or disinfectants for vaginal hygiene.

Table 1

Enrollment characteristics of subjects.

Despite the relatively low degree of sexual exposure reported, these women were at high risk for STDs, as evidenced by a baseline prevalence of 5.7% for gonorrhea, 3.9% for chlamydia, and 5.9% for trichomoniasis. The incidence of STDs remained high during follow-up. Despite risk reduction counseling, condom promotion and distribution, and STD treatment, the annual incidence of gonorrhea (16.7%), chlamydia (16.3%), and trichomonas (32.4%) remained high. In addition, 68 women seroconverted for antibodies to HIV-1 during follow-up, yielding an annual incidence rate of 11.0%.

Lactobacillus status and BV. Lactobacilli were isolated from 26% of women at enrollment and at 26% of follow-up visits (table 2). Lactobacilli that produced H2O2 were present at only 11% of enrollment visits and 10% of follow-up visits. BV was diagnosed by Gram's stain criteria (score, 7–10) at 36% of enrollment visits and at one-third of follow-up visits. Borderline BV (Gram's stain score, 4–6) was found at another one-third of visits, and normal flora patterns (Gram's stain score, 0–3) were found at only 34% of follow-up visits.

Table 2

Vaginal flora characteristics at enrollment and over follow-up.

Lactobacillus status was a dynamic variable in this population. After visits at which H2O2-producing lactobacilli were cultured from an individual, H2O2-producing strains were found at only 24% of next visits and H2O2-negative strains at an additional 21%. After a visit at which no lactobacilli were detected, 78% of next visits were culture negative, 14% were positive for H2O2-negative lactobacilli, and 8% were positive for H2O2-producing strains. Of the women studied, 44% were consistently culture negative for lactobacilli; only 3% were persistently culture positive.

The correlation between lactobacilli culture and BV (Gram's stain ⩾7) was −0.11 (P <.001), and the correlation between lactobacilli culture and abnormal flora (Gram's stain ⩾4) was −0.11 (P < .001). The correlation between H2O2-producing lactobacilli and BV was −0.13 (P < .001), and the correlation between H2O2-producing lactobacilli and abnormal flora was −0.17 (P <.001).

Lactobacillus status, vaginal flora, and risk of HIV-1 and STDs. The univariate associations between the absence of lactobacilli (vs. presence of either H2O2-producing or H2O2-negative strains of lactobacilli) and acquisition of HIV-1 and STDs are displayed in table 3. Absence of lactobacilli was associated with a significantly increased risk of HIV-1 infection during follow-up (hazard ratio [HR], 1.9; 95% confidence interval [CI], 1.1–3.2). A significant association was also found between absence of lactobacilli and acquisition of gonorrhea (HR, 1.7; 95% CI, 1.1–2.6). As expected, absence of lactobacilli was significantly associated with BV(HR, 1.3; 95% CI, 1.1–1.5). In addition, there was an inverse association between absence of lactobacilli and vaginal candidiasis (HR, 0.7; 95% CI, 0.6–0.9).

Table 3

Univariate association between absence of lactobacilli, abnormal vaginal flora, and incidence of human immunodeficiency virus type 1 (HIV-1) and genital tract infections.

To evaluate the effects of H2O2 production by lactobacilli, the data were next analyzed according to Lactobacillus H2O2 production status. Results of univariate regression analyses are shown in table 4. For these models, the reference category was presence of any H2O2-producing strains at either the current or previous visit. Compared with these women, women without lactobacilli had a 2.5-fold increased risk of acquiring HIV-1 infection (95% CI, 1.1–5.8). Women with only H2O2-negative strains were at intermediate risk (HR, 1.5; 95% CI, 0.6–3.9), although this was not significantly different than the reference group. Compared with the reference category, women without lactobacilli were also at higher risk for both gonorrhea (HR, 1.7; 95% CI, 0.9–3.3) and BV (HR, 1.4; 95% CI, 1.1–1.7). For gonorrhea and BV, women with H2O2-negative lactobacilli had HRs similar to those of women with H2O2-producing strains.

Table 4

Univariate association between H2O2 production status of lactobacilli and incidence of human immunodeficiency virus type 1 (HIV-1) and genital tract infections.

Results from the final multivariate proportional hazards models are shown in table 5. The final model for HIV-1 acquisition included gonorrhea, genital ulcer disease, candida, working in a bar, contraceptive use, number of sex partners per week, and <75% condom use. In the model in which presence or absence of lactobacilli were compared, the adjusted HR for HIV-1 seroconversion for women lacking lactobacilli was 2.0 (95% CI, 1.2–3.5). When lactobacilli were analyzed by H2O2-production status, a stepwise relationship was seen. The HR for women with H2O2-negative strains was 1.7 (95% CI, 0.6–4.7), whereas women without lactobacilli had an HR for HIV-1 infection of 2.8 (95% CI, 1.1–4.7).

Table 5

Multivariate association between absence of lactobacilli, H2O2 production status of lactobacilli, abnormal flora, and human immunodeficiency virus type 1 (HIV-1) and gonorrhea acquisition.

The multivariate model for gonorrhea included cervicitis, cervical mucopus, abnormal vaginal discharge, use of drying agents, number of sex partners per week, and sexual frequency. In this model, the adjusted HR for acquisition of gonorrhea for women lacking lactobacilli was 1.7 (95% CI, 1.1–2.6). When analyzed by H2O2-production status, no trend for increased risk was seen among women with H2O2-negative lactobacilli (HR, 1.2; 95% CI, 0.6–2.7), whereas women lacking lactobacilli had a risk of gonorrhea of 1.9 (95% CI, 0.9–3.8).

By use of Gram's stain evaluations of vaginal microflora, presence of abnormal vaginal flora was significantly associated with HIV-1 infection (HR, 1.7; 95% CI, 1.0–2.7; table 3). A significant association was also found with T. vaginalis infection (HR, 2.1; 95% CI, 1.5–2.9), and an inverse association was found with vaginal candidiasis (HR, 0.8; 95% CI, 0.6–1.0). In a multivariate model of HIV-1 acquisition, as described above, the adjusted HR for HIV-1 acquisition among women with abnormal flora was 1.9 (95% CI, 1.1–3.1; table 5). For trichomonas infection, the adjusted HR for abnormal vaginal flora was 1.8 (95% CI, 1.3–2.4) in a model that included chlamydia, abnormal vaginal discharge, method of contraception, work place, age at first sexual intercourse, circumcision, and use of drying agents.

Discussion

This study demonstrated significant relationships between vaginal colonization with Lactobacillus species and risk of acquisition of HIV-1 infection and N. gonorrhoeae. For HIV-1 infection, a stepwise increase in risk was observed for women with H2O2-producing lactobacilli, women with non-H2O2-producing organisms, and women lacking vaginal lactobacilli. Thus, for HIV-1 prevention, any lactobacilli appeared to be beneficial, although H2O2-producing strains provided the most benefit. It remains to be shown whether H2O2 is the protective substance or whether it is simply a marker for some other protective property of the organism. For gonorrhea, no such stepwise association was seen; that is, H2O2-positive and H2O2-negative strains were equally protective, suggesting that some characteristic of lactobacilli other than H2O2 production mediates the association. This study also found that the overall state of vaginal microbial flora was significantly associated with risk of HIV-1 infection and trichomoniasis.

If the observed association between vaginal colonization with lactobacilli and HIV-1 infection is causal, then by what mechanisms could this relationship occur? A review of possible interactions between the vaginal flora and HIV-1 infections was recently published by Hillier [24]. First, microbicidal properties of lactobacilli may be directly protective for HIV-1 infection. In vitro studies have shown that lactobacilli can be cidal to HIV-1 via the peroxidase-halide system [8], and other substances may be produced by lactobacilli that have antimicrobial properties. For example, lactic acid is produced by all Lactobacillus species, regardless of H2O2 status, and HIV-1 is inactivated at acid pH ranges [25]. In addition, an acidic environment results in decreased activation of T lymphocytes, which may result in decreased lymphocyte susceptibility to HIV-1 infection [26].

Second, organisms associated with BV may increase a woman's susceptibility to HIV-1 infection. Succinate, produced by the obligate anaerobic gram-negative rods associated with BV, can inhibit polymorphonuclear leukocyte function [27]. Some BV-associated organisms also produce sialidases, which directly stimulate lymphocytes and modify leukocyte oxidative bursts, and mucinases, which may allow pathogens to bypass the protective mucous barrier of the genital epithelium [28]. U. urealyticum, which is almost universally present in women with BV, produces IgA proteases, which can alter the effectiveness of the mucosal immune system [29]. Significant associations with HIV-1 infection were seen both with Lactobacillus culture results and overall vaginal Gram's stain score, indicating that either of the above proposed mechanisms—that is, microbicidal effects of lactobacilli or altered susceptibility to infection due to abnormal flora—or a combination of both may play a role in the observed findings.

The inverse association seen between vaginal colonization with lactobacilli and risk of gonorrhea infection could be a result of antimicrobial properties of the lactobacilli, competition for microbial nutrients in the vagina, or decreased susceptibility in the host. In vitro studies have documented that all lactobacilli inhibit the growth of N. gonorrhoeae at an acidic pH [30]. This is consistent with the H2O2-independent association of lactobacilli with risk of gonorrhea infection seen in this study. Although lactobacilli were significantly associated with risk of gonorrhea, vaginal Gram's stain score was not, suggesting that something specific to the lactobacilli, rather than the presence of organisms associated with BV, mediated the association.

The observed positive association between Lactobacillus culture result, vaginal Gram's stain score, and presence of vaginal candida seems paradoxical. That a healthier vaginal environment would be more likely to harbor fungal elements is counterintuitive. However, this association has been seen elsewhere [9], and it is hypothesized that gram-negative bacteria, such as those present in BV, may be inhibitory to candida [31].

Abnormal vaginal flora and STDs share epidemiologic risk factors, so it is possible that STDs and risk behavior may confound the observed association between vaginal flora abnormalities and HIV-1 infection. However, a notable strength of this study was its prospective design with frequent follow-up visits and careful collection of data regarding potentially confounding factors. Use of those potentially confounding variables that were associated with HIV-1 seroconversion (sexual exposure variables, gonorrhea, and genital ulcer disease) in the multivariate analyses did not alter the association between lactobacilli, abnormal flora, and HIV-1 infection.

The prevalence of vaginal Lactobacillus species was extremely low, and the prevalence of abnormal vaginal flora and BV was extraordinarily high in this population of African prostitutes. High rates of BV were observed previously in cohorts of African women [15–17]. Reasons for this are unclear, although vaginal hygiene practices and sexual behavior may play a role. It is also possible that hygiene practices may have decreased the recovery rate of lactobacilli in this study. Whatever the reasons, it appears that the absence of vaginal lactobacilli may increase the efficiency of male-to-female HIV-1 infection. The treatment of BV and the promotion of vaginal colonization with lactobacilli represent new potential intervention strategies for the prevention of HIV-1 infection that warrant evaluation in clinical trials. If these strategies are found to be effective, active diagnosis and treatment of BV could be incorporated into existing STD control programs. Strategies to improve the effectiveness and durability of BV treatments should also be developed. To date, there is no evidence that BV treatment can reduce the rate of HIV-1 infection in women, although this is an important question for further study. Vaginal microbicides consisting of naturally occurring Lactobacillus organisms are currently being developed and evaluated. With the increasing expansion of the HIV-1 epidemic to women worldwide, novel interventions, such as those suggested by this study, offer exciting opportunities for new methods of HIV-1 control.

References

, , , et al.

Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis

,

J Clin Microbiol

,

1989

, vol.

27

(pg.

251

-

6

)

, , .

Bacteriology of the vagina

,

Scand J Urol Nephrol

,

1984

, vol.

86

Suppl

(pg.

23

-

39

)

, , , .

Control of the microbial flora of the vagina by H2O2 generating lactobacilli

,

J Infect Dis

,

1991

, vol.

164

(pg.

94

-

100

)

, , , , .

Is there a role for lactobacilli in prevention of urogenital and intestinal infections?

,

Clin Microbiol Rev

,

1990

, vol.

3

(pg.

335

-

44

)

, , .

Emerging role of lactobacilli in the control and maintenance of the vaginal bacterial microflora

,

Rev Infect Dis

,

1990

, vol.

12

(pg.

856

-

72

)

.

The vaginal ecosystem

,

Am J Obstet Gynecol

,

1991

, vol.

165

(pg.

1163

-

8

)

, , , et al.

Hydrogen peroxide-producing lactobacilli and acquisition of vaginal infections

,

J Infect Dis

,

1996

, vol.

174

(pg.

1058

-

63

)

, .

Viricidal effect of Lactobacillus acidophilus on human immunodeficiency virus type 1: possible role in heterosexual transmission

,

J Exp Med

,

1991

, vol.

174

(pg.

289

-

92

)

, , , .

Vaginal Infections and Prematurity Study Group

Characteristics of three vaginal flora patterns assessed by Gram stain among pregnant women

,

Am J Obstet Gynecol

,

1992

, vol.

166

(pg.

938

-

4

)

, , , , .

The normal vaginal flora, H2O2-producing lactobacilli, and bacterial vaginosis in pregnant women

,

Clin Infect Dis

,

1993

, vol.

16

Suppl 4

(pg.

S273

-

81

)

.

Role of bacterial vaginosis in pelvic inflammatory disease

,

Clin Infect Dis

,

1995

, vol.

20

Suppl 2

(pg.

S271

-

5

)

, , , et al.

Association between bacterial vaginosis and preterm delivery of a low-birth-weight infant

,

N Engl J Med

,

1995

, vol.

333

(pg.

1737

-

42

)

, , , , .

Reduced incidence of preterm delivery with metronidazole and erythromycin in women with bacterial vaginosis

,

N Engl J Med

,

1995

, vol.

333

(pg.

1732

-

6

)

, , , et al.

Bacterial vaginosis and HIV seroprevalence among female commercial sex workers in Chiang Mai Thailand

,

AIDS

,

1995

, vol.

9

(pg.

1093

-

7

)

, , , et al.

HIV-1 infection associated with abnormal vaginal flora morphology and bacterial vaginosis

,

Lancet

,

1997

, vol.

350

(pg.

546

-

50

)

, , , et al.

HIV infection and disturbances of vaginal flora during pregnancy

,

J Acquir Immune Defic Syndr Hum Retrovirol

,

1999

, vol.

20

(pg.

52

-

9

)

, , , et al.

Bacterial vaginosis and disturbances of vaginal flora: association with increased acquisition of HIV

,

AIDS

,

1998

, vol.

12

(pg.

1699

-

706

)

.

Prevalence of bacterial vaginosis and its association with genital infections, inflammation, and contraceptive methods in women attending sexually transmitted disease and primary health clinics

,

Int J STD AIDS

,

1990

, vol.

1

(pg.

86

-

94

)

, , , .

The relationship of hydrogen peroxide-producing lactobacilli to bacterial vaginosis and genital microflora in pregnant women

,

Obstet Gynecol

,

1992

, vol.

79

(pg.

369

-

73

)

, , , et al.

Hormonal contraception, sexually transmitted diseases, and risk of heterosexual transmission of human immunodeficiency virus type 1

,

J Infect Dis

,

1998

, vol.

178

(pg.

1053

-

9

)

, , .

Reliability of diagnosing bacterial vaginosis is improved by a standardized method of Gram stain interpretation

,

J Clin Microbiol

,

1991

, vol.

29

(pg.

297

-

301

)

, .

Cox's regression model for counting processes: a large sample study

,

Ann Stat

,

1982

, vol.

10

(pg.

1100

-

20

)

, , , et al.

Use of nonoxynol-9 and changes in vaginal lactobacilli

,

J Infect Dis

,

1998

, vol.

178

(pg.

441

-

5

)

.

The vaginal microbial ecosystem and resistance to HIV

,

AIDS Res Hum Retroviruses

,

1998

, vol.

14

Suppl 1

(pg.

S17

-

S21

)

, , .

Disinfection and inactivation of the human lymphotropic virus type III/lymphadenopathy-associated virus

,

J Infect Dis

,

1985

, vol.

152

(pg.

400

-

3

)

, .

Human vaginal leukocytes and the effects of vaginal fluid lymphocyte and macrophage defense functions

,

Am J Obstet Gynecol

,

1992

, vol.

166

(pg.

720

-

6

)

, , , , .

Succinic acid, a metabolic by-product of Bacteroides species, inhibits polymorphonuclear leukocyte function

,

Infect Immun

,

1985

, vol.

48

(pg.

402

-

5

)

, , , et al.

Bacterial vaginosis is associated with prematurity and vaginal fluid mucinase and sialidase: results of a controlled trial of topical clindamycin cream

,

Am J Obstet Gynecol

,

1994

, vol.

170

(pg.

1048

-

60

)

, , .

Immunoglobulin A protease activity of Ureaplasma urealyticum

,

J Clin Microbiol

,

1984

, vol.

19

(pg.

255

-

8

)

, , .

Effects of H2O2-producing lactobacilli on Neisseria gonorrhoeae growth and catalase activity

,

J Infect Dis

,

1994

, vol.

170

(pg.

1209

-

15

)

, .

Microbial flora associated with Candida albicans vulvovaginitis

,

Obstet Gynecol

,

1980

, vol.

55

(pg.

397

-

401

)

Presented in part: XII International Conference on AIDS, Geneva, July 1998 (abstract 33148).

Informed consent was obtained from all subjects, and human experimentation guidelines of the Universities of Nairobi and Washington were followed.

Financial support: NIH (grants AI-33873, D43-TW00007, and T22-TW00001; subcontract AI-35173 through Family Health International).

Author notes

a

Current affiliation: Park Nicollet Clinic, Minneapolis, MN.

b

Deceased.

© 1999 by the Infectious Diseases Society of America

Topic:

• trichomoniasis
• gonococcal infection
• gram's stain
• follow-up
• hiv-1
• lactobacillus
• sexually transmitted diseases
• vagina
• microbial colonization
• hiv-1 infection

Citations

Views

Altmetric

Metrics

Total Views 13,178

11,798 Pageviews

1,380 PDF Downloads

Since 11/1/2016

Citations

646 Web of Science

×

Email alerts

More on this topic

Citing articles via

• Latest

• Most Read

• Most Cited

More from Oxford Academic