Epidemiologic Reviews
2
Copyright © 2002 by the Johns Hopkins Bloomberg School of Public Health
All rights reserved
Bacterial Vaginosis in Pregnancy: Current Findings and Future Directions
Deborah B. Nelson1 and George Macones1,2
1 Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics and Epidemiology, University of Pennsylvania
School of Medicine, Philadelphia, PA.
2 Department of Maternal and Fetal Medicine, University of Pennsylvania Health System, Philadelphia, PA.
Received for publication June 6, 2002; accepted for publication November 19, 2002.
Abbreviation: BV, bacterial vaginosis.
INTRODUCTION
Bacterial vaginosis (BV) is an extremely prevalent vaginal
condition and the number one cause of vaginitis among both
pregnant and nonpregnant women (1). Although it is not a
reportable disease, current studies have found the prevalence
of BV among nonpregnant women to range from 15 percent
to 30 percent; up to 50 percent of pregnant women have been
found to have BV (2–5). However, the majority of cases of
BV are asymptomatic and remain unreported and untreated
(3, 6). Previously considered a benign condition, BV has
been related to many gynecologic conditions and complica-
tions of pregnancy including pelvic inflammatory disease,
posthysterectomy vaginal cuff cellulitis, endometritis, amni-
otic fluid infection, preterm delivery, preterm labor, prema-
ture rupture of the membranes, and, possibly, spontaneous
abortion (7–12). The role of asymptomatic, compared with
symptomatic, BV in both gynecologic and pregnancy-
related conditions has been less studied, although research
emphasis is shifting toward determining these independent
relations. In laboratory and clinical studies, BV has been
shown to ascend to the endometrium and invade the
placenta, but the complete impact of this migration in terms
of initial and sustained placental development and early fetal
development is unclear (13). The purposes of this article is to
review the background, diagnosis, and treatment of BV in
pregnancy; discuss the epidemiology and consequences of
BV in pregnancy; and outline current research findings and
future research directions focusing exclusively on the conse-
quences of BV in pregnancy.
BACKGROUND
BV is a polymicrobial, superficial vaginal infection
involving a reduction in the amount of hydrogen-peroxide-
Correspondence to Dr. Deborah B. Nelson, Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine,
921 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104-6021 (e-mail: dnelson@cceb.med.upenn.edu).
102
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
Vol. 24, No.
Printed in U.S.A.
DOI: 10.1093/epirev/mxf008
producing Lactobacillus and an overgrowth of anaerobic and
Gram-negative or Gram-variable bacteria (14, 15). The
reduced number of Lactobacillus promote overgrowth of
anaerobic bacteria, including Mycoplasma hominis,
Bacteroides species, Mobiluncus species, and Gardnerella
vaginalis (14, 15). Although most of these organisms are
present in small numbers in the normal vagina, Mobiluncus
is rarely found and is a sensitive marker for the diagnosis of
BV (16). On the other hand, Gardnerella has been reported
in up to 50 percent of women with no signs or symptoms of
BV; therefore, the finding of Gardnerella is not a definitive
diagnostic of BV (17, 18). In fact, it seems that the decrease
in Lactobacillus, as opposed to the increase in other organ-
isms, influences the vaginal flora and may be the most
important predictor in subsequent BV development (19).
Studies among nonpregnant women collecting serial
samples of vaginal flora have concluded that some events
(either behavioral, hormonal, or environmental) occur that
promote a change in the normal flora of the vagina. A recent
study incorporating repeat measures of vaginal flora concen-
trations among women throughout the menstrual cycle
reported a high rate of BV presentation during the follicular
phase of the menstrual cycle and a spontaneous resolution of
BV during the luteal phase. These results suggest that endog-
enous sex hormones may support and assist in sustaining
high levels of Lactobacillus and illustrate the potential for
sex hormones to influence the organisms present in the
vagina (6, 20, 21).
Currently, we know of no studies that have been
conducted among pregnant women to describe the changes
in vaginal flora or BV prevalence during gestation. Of
interest would be the assessment of BV prevalence by gesta-
tional age and the correlation between increasing sex
hormone levels and BV presentation. In addition, it is
Epidemiol Rev 2002;24:102–108

FIGURE 1. Diagnostic tests for bacterial vaginosis. KOH, potassium hydroxide.
currently unknown whether the proportion of symptomatic
and asymptomatic cases varies by gestational age. The pres-
ence of BV at a particular gestational age may be a factor in
the subsequent development of pregnancy complications,
and the risk for disease may change based on BV positivity
during different stages of gestation. For example, the risk of
preterm delivery due to BV in the first trimester, during early
fetal and placental development, may be different compared
with the risk of preterm delivery in the second and third
trimesters, during profuse placental functioning. These rela-
tions currently are unknown.
DIAGNOSIS
Two diagnostic tests are commonly used for BV. Amsel
criteria, the test most commonly used in the clinical setting,
involves assessing four clinical conditions, with the exist-
ence of three or more conditions corresponding to a diag-
nosis of BV (figure 1). These conditions include an elevated
vaginal pH (>4.5), an amine or fishy odor when vaginal fluid
is prepared with 10 percent potassium hydroxide solution
(called the “whiff test”), the presence of clue cells on wet
mount, and a homogeneous vaginal discharge. In brief, clue
cells are vaginal epithelial cells coated with bacteria that
look speckled rather than translucent and have serrated or
unclear borders because of the adherent bacteria (22). The
updated Amsel criteria specify that at least 20 percent of
epithelial cells should be clue cells, although these amended
criteria may reduce the sensitivity of the test (7). In the past,
the Amsel criteria was the most common method of identi-
fying BV; however, there are inherent difficulties with each
of the individual parameters. Assessment of vaginal pH
lacks specificity because an increase in vaginal pH may be a
consequence of many other lower genital tract conditions,
conduct of the whiff test is subjective for each individual
clinician and lacks sensitivity, and identification of clue cells
may vary according to the skill and interpretation of the
microscopist and the quality of sample collection (11).
Epidemiol Rev 2002;24:102–108
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
Bacterial Vaginosis in Pregnancy 103
The second, more commonly used diagnostic test involves
a Gram stain of vaginal fluid and use of Nugent criteria to
identify a case of BV (figure 1). This method has been
shown to have a high sensitivity and specificity compared
with Amsel criteria (89 percent and 83 percent, respectively)
(23, 24). A vaginal swab is obtained, spread on a glass slide,
air dried, and later Gram stained. The amount of three
morphotypes is quantified and scored: Lactobacillus, Mobi-
luncus, and Gardnerella. For Lactobacillus, scores range
from 0 to 4; 0 indicates that 30 or more organisms were
found, and 4 indicates that no organisms were found in the
sample. In contrast, for Gardnerella, a score of 0 indicates
that no organisms were found and the highest score, 4, indi-
cates that 30 or more organisms were found. For Mobi-
luncus, scores range from 0 to 2, with a score of 2 indicating
that five or more organisms were identified in the sample. A
summary BV score is computed ranging from 0 to 10; this
score is dichotomized, with values of 7 and over indicating a
case of BV (25).
The validity of BV diagnosis given the method of swab
collection (either self-collected or provider collected) has
been examined (26). Using self-collected vaginal swabs may
be particularly attractive for epidemiologic studies because
this method does not require recruiting study participants
from clinical settings, and it facilitates repeat sampling for
studies designed to measure BV status at multiple points in
time. Two studies of nonpregnant women assessing the
validity of self-compared to provider-collected swabs to
detect BV found excellent validity when the two methods
were compared (26).
The Nugent criteria is the test most often used in epidemi-
ologic studies such as the large-scale ones conducted by the
Maternal-Fetal Medicine Network Units of the National
Institute of Child Health and Human Development (27, 28).
This method has several advantages that include 1) creating
a permanent record that can be subsequently reviewed to
confirm the diagnosis of BV and assess the reliability of the
reading; 2) reporting intermediate stages of BV, which is
 104 Nelson and Macones
particularly useful in longitudinal studies examining serial
vaginal fluid samples for changes in BV status; and 3) quan-
tifying the amount of the three individual organisms,
enabling assessment of the organism-specific risk of disease.
More recently, BV is beginning to be considered a condi-
tion with a spectrum of positivity. Currently, a case of BV is
classified as either positive or negative without an organism-
specific definition or an assessment of organism-specific
risk for disease. In addition, to our knowledge the relation
between BV and risk of disease has not been assessed recog-
nizing that the summary BV score itself, ranging from 0 to 7,
indicates a continuous degree of BV positivity. In the future,
studies should relate a dose-response disease risk to indi-
vidual BV scores as well as examine the affect of the sepa-
rate organisms on disease occurrence. Thus, we are in the
early stages of both identifying and diagnosing BV as well as
determining the relation between BV and adverse pregnancy
outcomes.
TREATMENT
What is meant by “cure”? Typically, a cure for BV refers
to resolution of symptoms and perhaps a repeat BV-negative
screen. We know from clinical studies that BV has both a
spontaneous resolution and recurrence, but the factors
contributing to this resolution or recurrence are unknown
(29). In addition, the behavioral, hormonal, or environmental
factors predisposing a woman to multiple, recurrent cases of
BV are also unclear. This section reviews current therapies
to treat BV during pregnancy. It is important to recognize
that the diagnostic tests we describe, which are used to detect
BV, collect fluid samples from the vagina; however, we
know that BV ascends to the upper genital tract. Therefore,
identifying factors that predict BV ascension and the effi-
cacy of treatment therapies to resolve lower and upper
genital tract infection are important. The efficacy studies
described refer to the cure rate for the initial case of BV and
do not attempt to follow the patient for recurrence. As many
as 30 percent of women relapse within 1 month of treatment,
with spontaneous relapse occurring more commonly among
women treated with topical compared with systemic antibi-
otics (29).
The most common oral treatment for BV in both pregnant
and nonpregnant women is metronidazole (30). The indi-
vidual cure rate given a 7-day, twice-daily course of 500 mg
of metronidazole ranges from 84 percent to 96 percent, and
the cure rate given a 2-g single dose of metronidazole is 54–
62 percent (31). Previously, there was concern regarding use
of metronidazole during the first trimester of pregnancy
before the completion of organogenesis (32, 33). However, a
small study examining children exposed in utero to metro-
nidazole suggested no evidence of long-term teratogenic
effects (34). Although it was long considered an effective
treatment for the resolution of symptoms related to BV, an
interesting article recently reported that high concentrations
of metronidazole, greater than 5,000 µg/ml, completely
suppressed the growth of Lactobacillus and that concentra-
tions of 1,000–4,000 µ/ml significantly inhibited the growth
of Lactobacillus (35). Therefore, the dose of metronidazole
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
may be important in determining both the cure and the recur-
rence rate.
The second systemic treatment for BV is oral clindamycin.
The one known clinical trial conducted describing the effi-
cacy of oral clindamycin reported that a 300-mg, twice-daily
course of clindamycin for 7 days resulted in a 94 percent
cure rate (31). Of note, all of the efficacy studies have been
conducted among nonpregnant women, with the assumption
that the cure rate for BV among pregnant woman is similar.
The two topical treatments for BV include metronidazole
0.75 percent vaginal gel and clindamycin 2 percent vaginal
cream. A twice-daily, 5-day therapy of vaginal metronida-
zole had a reported cure rate of 75–81 percent, while treat-
ment with clindamycin cream was reported to resolve 82–96
percent of cases of BV (36). The efficacy of the 3-day treat-
ment and the 7-day treatment of clindamycin cream has been
found to be equally effective and well tolerated in treating
BV (37). Again, these topical treatments result in the resolu-
tion of lower genital tract infection but do not treat BV
occurring in the upper genital tract. What is the correlation
between lower genital tract resolution and upper genital tract
resolution? Currently, the answer is unknown.
EPIDEMIOLOGY: RISK FACTORS
Much information is known regarding the microbiology
and identification of BV; however, limited information
exists concerning the factors or behaviors that increase a
woman’s risk for BV during pregnancy. The current predic-
tors of BV have been limited to race, sexual activity, socio-
economic status, and perhaps vaginal douching. Most of the
epidemiologic studies conducted to date to determine risk
factors for BV have concentrated on symptomatic cases and
included results from women seeking care in sexually trans-
mitted disease clinics or inner-city obstetric offices. General-
izability of the current literature is unclear since
asymptomatic cases have not been examined fully and the
current data represent only a subset of women of reproduc-
tive age. Nonetheless, the reported prevalence of BV among
pregnant women ranges from 10 percent to 35 percent, with
higher rates occurring among African-American women,
low-income women, or women with prior sexually trans-
mitted diseases (10, 38, 39). The Vaginal Infections and
Prematurity Study, which measured BV among pregnant
women between 23 and 26 weeks of gestation, found a 2.0-
to 2.5-fold increased risk of BV among African-American
compared with White pregnant women (10). Numerous
studies have confirmed at least a twofold increased risk of
BV among African-American women presumably due to
environmental/behavioral exposures or stressors (40, 41).
Women of lower socioeconomic status and women self-
reporting higher levels of psychosocial stress also have
increased rates of BV. In recent studies among obstetrics
populations, the reported prevalence of BV ranged from a
low of 10 percent among private patients to a high of 35
percent among women reporting low monthly incomes and
low educational levels, although these studies did not adjust
for race (42, 43). Culhane et al. assessed the role of chronic
maternal stress, as measured by the Cohen perceived stress
scale, and found that independent of sociodemographic and
Epidemiol Rev 2002;24:102–108

behavioral factors, chronic maternal stress remained a signif-
icant predictor of BV among pregnant women (44).
Epidemiologic studies have found that early sexual
activity, a high number of lifetime sexual partners, women
with a new sexual partner, and women with a prior sexually
transmitted disease are also at increased risk of BV (45–47).
BV is more prevalent among women with a prior or current
sexually transmitted disease. However, the occurrence of
BV may be the direct consequence of exposure to the infec-
tious pathogen, not the sexual behavior. In fact, many patho-
gens have been shown to change vaginal flora by reducing
the concentration of Lactobacillus and promoting anaerobic
bacteria proliferation and subsequent BV development (29).
Although sexually transmitted diseases and BV commonly
coexist, particularly trichomoniasis and BV, BV is not
considered a sexually transmitted disease (48–51). For
example, a study among school-age girls found similar rates
of BV among virgin girls and nonvirgin girls (12 percent and
15 percent, respectively) (49). Furthermore, although the
anaerobic organisms in excess in cases of BV have been
cultured from the male sexual partners of women with BV,
treatment of male sexual partners is not a reliable way to
reduce the recurrence of BV in these women (51). However,
a small study of monogamous lesbian women concluded that
the likelihood of one partner having BV was 20 times greater
if the other partner was BV positive (odds ratio = 19.7, 95
percent confidence interval: 2.1, 588.0), supporting the early
finding by Gardner and Dukes that BV is transmissible
through direct inoculation of vaginal secretions (50, 52).
Some behaviors, such as vaginal douching, have been
examined as potential risk factors for BV. Among nonpreg-
nant woman, self-reported vaginal douching has been
reported to increase the risk of BV (53, 54). Holzman et al.
found more than a twofold increased risk of BV among
nonpregnant women who self-reported vaginal douching in
the prior 2 months (4). No known studies have been
published to date examining the role of douching and BV
development among pregnant women. Vaginal douching
may change the vaginal flora, reduce the amount of Lactoba-
cillus, and create an environment promoting excessive
anaerobic growth; on the other hand, the act of douching
may be a consequence of the symptoms of BV (i.e., vaginal
discharge and odor) or a current sexually transmitted disease
(53, 54). Currently, prospective studies are under way to
answer these questions.
EPIDEMIOLOGY: ADVERSE PREGNANCY OUTCOMES
BV is very prevalent among reproductive-age women, but,
for a common condition, the subsequent risk of adverse preg-
nancy outcomes is marginal (11, 56). The finding of a rela-
tively low risk for a variety of events may in fact be due to
the imprecise definition of exposure. As discussed previ-
ously, BV is a syndrome with degrees of positivity. The
current literature has examined the relation between BV
positivity and health outcome, but currently we know of no
studies that have examined the organism-specific risks for
disease. In addition, it is unclear whether BV is the risk
factor for disease or whether exposure to BV or the various
microorganisms causes inflammatory changes that are the
Epidemiol Rev 2002;24:102–108
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
Bacterial Vaginosis in Pregnancy 105
necessary event predicting adverse outcomes. We do know
that BV diagnosed from the lower genital tract has been
related to 1) an increased potential for other vaginal patho-
gens to gain access to the upper genital tract, 2) the presence
of enzymes that reduce the ability of leukocytes to reduce
infection, and 3) an increased level of endotoxins stimulating
cytokine and prostaglandin production (57–61). In fact,
Imseis et al. reported higher vaginal levels of interleukin-1
beta, an inflammatory cytokine, among pregnant women
with BV, and Spandorfer et al. found higher levels of both
cervical interleukin-1 beta and interleukin-8 cytokine levels
among nonpregnant women with BV (62, 63). Future BV
research should attempt to define BV exposure and to outline
the inflammatory consequences of BV exposure and risk of
adverse pregnancy outcomes. The next section reviews the
studies conducted to date examining the role of BV and preg-
nancy outcomes.
The vast majority of epidemiologic research designed to
examine the role of BV and adverse pregnancy outcomes has
focused on the risk of preterm delivery, although many of
these studies incorrectly combined preterm labor and
preterm, premature rupture of the membranes. In any case,
these studies have consistently shown a twofold increased
risk of preterm delivery among women diagnosed with BV,
particularly BV diagnosed in the early second trimester (11,
55, 56). A recent meta-analysis reviewing studies examining
the role of BV and the risk of preterm delivery reported a
summary odds ratio of 1.6, indicating a 60 percent increased
risk of preterm delivery among pregnant women with BV. A
smaller number of studies have assessed the relation between
BV and the outcomes of premature labor, low birth weight,
and premature rupture of the membranes. One study exam-
ining several pregnancy outcomes related to BV diagnosed
during the first trimester of pregnancy reported a 2.6-fold
increased risk of preterm labor (95 percent confidence
interval: 1.3, 4.9), a 6.9-fold increased risk of preterm
delivery (95 percent confidence interval: 2.5, 18.8), and a
7.3-fold increased risk of preterm, premature rupture of the
membranes (95 percent confidence interval: 1.8, 29.4) (11).
Another study found that BV diagnosed in the second
trimester was associated with an increased risk of preterm
delivery and premature rupture of the membranes and that
BV accounted for 83 percent of the attributable risk for
preterm birth (64).
A growing body of literature has begun to suggest an
increased risk of spontaneous abortion among pregnant
women with BV (12, 65, 66). Studies have reported a three-
to fivefold increased risk of spontaneous abortion among
pregnant women with BV in the first trimester, although
these studies were hampered by small sample size (12, 65).
Two additional studies among high-risk pregnant women
also reported an increase in spontaneous abortion among
women diagnosed with BV (66, 67). A study enrolling
women undergoing infertility treatment found more than a
twofold increased risk of spontaneous abortion among
women with BV after adjustment for maternal age, prior
livebirth, and self-reported cigarette smoking (relative risk =
2.67, 95 percent confidence interval: 1.26, 5.63) (66).
Numerous clinical trials have examined the efficacy of
oral and topical BV treatment to reduce the risk of preterm
 106 Nelson and Macones
delivery and have found no reduced risk among pregnant
women receiving topical treatments for BV (68, 69).
Although controversial, the studies of oral therapies have
suggested different therapeutic approaches for symptomatic,
asymptomatic, and high-risk pregnant women. Symptomatic
pregnant women with BV are treated to alleviate symptoms,
with prevention of adverse events (i.e., preterm birth,
preterm labor, premature rupture of the membranes) desir-
able but not well documented (31). The treatment of asymp-
tomatic BV-positive pregnant women and the possible
reduction in adverse pregnancy outcomes are also unclear.
Three separate placebo-controlled randomized clinical trials
indicated a reduction in the risk of preterm delivery
following treatment with metronidazole; however, in two
studies, the reduction was found in only the small subset of
high-risk asymptomatic pregnant women (70–72). In fact, a
meta-analysis of all randomized controlled trials of BV in
pregnancy found no benefit to BV treatment in average-risk
women for any pregnancy outcome (73). In addition, a
recent clinical trial did not find a reduction in the occurrence
of preterm delivery among either high-risk or low-risk
asymptomatic pregnant women following treatment with
oral metronidazole (28). In clinical practice, high-risk
asymptomatic pregnant women are commonly screened in
the early second trimester and are treated with oral metro-
nidazole, but the benefit of this therapy in reducing the
woman’s risk of preterm delivery remains unclear (74).
CONCLUSION
BV is an enormous public health problem, accounting for
the majority of cases of vaginitis and vaginal discharge in the
United States. Although symptomatic BV is an extremely
prevalent vaginal condition among pregnant women, the true
magnitude is not known because more than one half of BV
cases are asymptomatic. Studies developed to quantify the
prevalence of symptomatic and asymptomatic BV among
pregnant woman and to determine whether BV differs by
gestational ages would be useful. Given the limited informa-
tion regarding the factors and behaviors placing a woman at
increased risk for BV in pregnancy, additional research in
this area is also needed. It is essential to conduct large-scale
epidemiologic studies to determine behavioral, hormonal, or
environmental factors and/or comorbidities that increase a
woman’s risk for BV in pregnancy and to explore the associ-
ation of race, socioeconomic status, sexually transmitted
diseases, and vaginal douching with BV presentation.
Current studies relating BV to adverse pregnancy
outcomes have measured and categorized BV positivity
from samples of the lower genital tract. These lower genital
tract measurements of BV have been moderately related to
the risk of adverse events, with the corresponding assump-
tion being the eventual ascension of these organisms to the
upper genital tract. The biologic or environmental events
promoting the ascension of lower genital tract bacteria to the
upper genital tract are of extreme interest but have not been
adequately examined to date. In addition, the predictors of
recurrent BV and spontaneous resolution of BV are impor-
tant. Only when these microbiologic, epidemiologic, and
sociologic determinants of BV are examined fully will we as
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
a public health community begin to understand and prevent
the occurrence of BV in pregnancy.
REFERENCES
1. Rein MF, Holmes KK. Non-specific vaginitis, vulvovaginal
candidiasis, and trichomoniasis clinical features, diagnosis and
management. Curr Clin Top Infect Dis 1983;4:281–315.
2. Fleury FS. Adult vaginitis. Clin Obstet Gynecol 1981;24:407–
38.
3. McGregor JA, French JI. Bacterial vaginosis in pregnancy.
Obstet Gynecol Surv 2000;55(5 suppl 1):1–19.
4. Holzman C, Leventhal JM, Qiu H, et al. Factors linked to bac-
terial vaginosis in nonpregnant women. Am J Public Health
2001;91:1664–70.
5. Cristiano L, Coffetti N, Dalvai G, et al. Bacterial vaginosis:
prevalence in outpatients, association with some micro-organ-
isms and laboratory indices. Genitourin Med 1989;65:382–7.
6. Amsel R, Totten PA, Spiegel CA, et al. Non-specific vaginitis:
diagnostic and microbial and epidemiological associations. Am
J Med 1983;74:14–22.
7. Eschenbach DA, Hillier S, Critchlow C, et al. Diagnosis and
clinical manifestations of bacterial vaginosis. Am J Obstet
Gynecol 1988;158:819–28.
8. Abner KP, Hessol NA, Padian NS, et al. Risk factors for plasma
cell endometritis among women with cervical Neisseria gonor-
rhoeae, cervical Chlamydia trachomatis, or bacterial vaginosis.
Am J Obstet Gynecol 1998;178:987–90.
9. Meis PJ, Goldenburg RL, Mercer B, et al. The preterm predic-
tion study: significance of vaginal infections. Am J Obstet
Gynecol 1995;173:1231–5.
10. Hillier SL, Martius J, Krohn M, et al. A case-control study of
chorioamniotic infection and histologic chorioamnionitis in
prematurity. N Engl J Med 1988;319:972–8.
11. Kurki T, Sivonen A, Renkonen OV, et al. Bacterial vaginosis in
early pregnancy and pregnancy outcome. Obstet Gynecol 1992;
80:173–7.
12. Hay PE, Lamont RF, Taylor-Robinson D, et al. Abnormal bac-
terial colonization of the genital tract and subsequent preterm
delivery and late miscarriage. BMJ 1994;308:295–8.
13. Peipert JF, Montagno AG, Cooper AS, et al. Bacterial vaginosis
as a risk factor for upper genital tract infection. Am J Obstet
Gynecol 1997;177:1184–7.
14. Eschenbach DA, Dazick PR, Williams BL, et al. Prevalence of
hydrogen peroxide producing Lactobacillus species in normal
women and women with bacterial vaginosis. J Clin Microbiol
1989;27:251–6.
15. Spiegel CA, Amsel R, Eschenbach D, et al. Anaerobic bacteria
in nonspecific vaginitis. N Engl J Med 1980;303:601–7.
16. Hillier SL, Critchlow CW, Stevens CE, et al. Microbiological,
epidemiological and clinical correlates of vaginal colonisation
by Mobiluncas species. Genitourin Med 1991;67:26–31.
17. Hill GB. The microbiology of bacterial vaginosis. Am J Obstet
Gynecol 1993;169(2 pt 2):450–4.
18. Aroutcheva AA, Simoes JA, Behbakht K, et al. Gardnerella
vaginalis isolated from patients with bacterial vaginosis and
from patients with health vaginal ecosystems. Clin Infect Dis
2001;33:1022–7.
19. Klebanoff SJ, Hillier SL, Eschenbach DA, et al. Control of the
microbial flora of the vagina by H202-generating lactobacilli. J
Infect Dis 1991;164:94–100.
20. Hay PE, Morgan DJ, Iscon CA, et al. A longitudinal study of
bacterial vaginosis and other changes in the vaginal flora during
pregnancy. Br J Obstet Gynecol 1994;101:1048–53.
Epidemiol Rev 2002;24:102–108

21. Hay PE, Ugwamadu A, Chowns J. Sex, thrush and bacterial
vaginosis. Int J STD AIDS 1997;8:603–8.
22. Lichtman R, Duran P. The vulva and vagina. In: Lichtman R,
Papera S, eds. Well woman care. Norwalk, CT: Appleton &
Lange, 1990.
23. Schwebke JR, Hillier SL, Sobel JD, et al. Validity of the vagi-
nal Gram stain for the diagnosis of bacterial vaginosis. Obstet
Gynecol 1996;88:573–6.
24. Mastrobattista JM, Bishop KD, Newton ER. Wet smear com-
pared with Gram stain diagnosis of bacterial vaginosis in
asymptomatic pregnant women. Obstet Gynecol 2000;96:504–
6.
25. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing
bacterial vaginosis is improved by a standardized method of
Gram stain interpretation. J Clin Microbiol 1991;29:297–301.
26. Schwebke JR, Morgan SC, Weiss HL. The use of sequential
self-obtained vaginal smears for detecting changes in the vagi-
nal flora. Sex Transm Dis 1997;24:236–9.
27. Carey JC, Klebanoff MA, Hauth JC. Metronidazole to prevent
preterm delivery in pregnant women with asymptomatic bacte-
rial vaginosis. N Engl J Med 2000;342:534–40.
28. Goepfert AR, Goldenberg RL, Andrews WW, et al. The Pre-
term Prediction Study: association between cervical interleukin
6 concentration and spontaneous preterm birth. National Insti-
tute of Child Health and Human Development. Maternal-Fetal
Medicine Units Network. Am J Obstet Gynecol 2001;184:483–
8.
29. Hay PE. Recurrent bacterial vaginosis. Sex Transm Dis 1998;
16:769–73.
30. Joesoef MR, Schmid GP, Hillier SL. Bacterial vaginosis:
review of treatment options and potential clinical indications
for therapy. Clin Infect Dis 1999;28(suppl 1):S57–S65.
31. Greaves WL, Chungafung J, Morris B, et al. Clindamycin ver-
sus metronidazole in the treatment of bacterial vaginosis.
Obstet Gynecol 1988;72:799–802.
32. Rusha M, Shwak P. Introduction of lung tumors and malignant
lymphoma in mice by metronidazole. J Natl Cancer Inst 1972;
48:721–9.
33. Cavaliere A, Bacci M, Amorisi A, et al. Introduction of lung
tumors and lymphomas in BALB/C mice by metronidazole.
Tumori 1983;69:379–82.
34. Thapa PB, Whitlock JA, Brockman Worell KG, et al. Prenatal
exposure to metronidazole and risk of childhood cancer: a ret-
rospective cohort study of children younger than 5 years.
Cancer 1998;83:1461–8.
35. Simoes JA, Aroutcheva AA, Shott S, et al. Effect of metronida-
zole on the growth of vaginal Lactobacillus in vitro. Infect Dis
Obstet Gynecol 2001;9:41–5.
36. Ferris DG, Litaker MS, Woodward L, et al. Treatment of bacte-
rial vaginosis: a comparison of oral metroxidazole, metrotrida-
zole vaginal gel and clindamycin vaginal cream. J Fam Pract
1995;41:443–9.
37. Sobel J, Peipert JF, McGregor JA, et al. Efficacy of clindamy-
cin vaginal ovule (3-day treatment) vs. clindamycin vaginal
cream (7-day treatment) in bacterial vaginosis. Infect Dis
Obstet Gynecol 2001;9:9–15.
38. Rauh VA, Culhane JF, Hogan VK. Bacterial vaginosis: a public
health problem for women. J Am Med Womens Assoc 2000;55:
220–2.
39. Schwebke JR. Vaginal infections. In: Goldman MB, Hatch MC,
eds. Women and health. San Diego, CA: Academic Press, 2000.
40. Schmid GP. The epidemiology of bacterial vaginosis. Int J
Gynaecol Obstet 1999;67:S17–S20.
41. Goldenberg RL, Klebanoff MA, Nugent R, et al. Bacterial col-
onization of the vagina during pregnancy in four ethnic groups:
vaginal infections and prematurity study group. Am J Obstet
Epidemiol Rev 2002;24:102–108
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
Bacterial Vaginosis in Pregnancy 107
Gynecol 1996;174:1618–21.
42. Minkoff H, Grunebaum A, Schwarz R. Risk factors for prema-
turity and premature rupture of membranes: a perspective study
of vaginal flora in pregnancy. Am J Obstet Gynecol 1984;150:
965–72.
43. Gardner HJ, Dampeer T, Dukes C. The prevalence of vaginitis.
Am J Obstet Gynecol 1957;73:1080–7.
44. Culhane JF, Rauth V, McCollum KF, et al. Maternal stress is
associated with bacterial vaginosis in human pregnancy.
Matern Child Health J 2001;5:127–34.
45. Barbone F, Austin H, Louv WC, et al. A follow-up study of
methods of contraception, sexual activity and rates of trichomo-
niasis, candidiasis and bacterial vaginosis. Am J Obstet Gyne-
col 1990;163:510–14.
46. Moi H. Prevalence of bacterial vaginosis and its association
with genital infections, inflammation and contraceptive meth-
ods in women attending sexually transmitted disease and pri-
mary health clinics. Int J STD AIDS 1990;1:86–94.
47. Paavonen J, Miettinen A, Stevens CE, et al. Mycoplasma hom-
inis in non-specific vaginitis. Sex Transm Dis 1983;45:271–5.
48. Franklin TL, Monif GR. Trichomonas vaginalis and bacterial
vaginosis: coexistence in vaginal wet mount preparations from
pregnant women. J Reprod Med 2000;45:131–4.
49. Bump RC, Buesching WJ 3rd. Bacterial vaginosis in virginal
and sexually active adolescent females: evidence against exclu-
sive sexual transmission. Am J Obstet Gynecol 1988;158:935–
9.
50. Berger BJ, Kolton S, Zenilman JM, et al. Bacterial vaginosis in
lesbians: a sexually transmitted disease. Clin Infect Dis 1995;
21:1402–5.
51. Morris MC, Rogers PA, Kinghorn GR. Is bacterial vaginosis a
sexually transmitted infection? Sex Transm Infect 2001;77:63–
8.
52. Gardner HL, Dukes CD. Hemophilus vaginalis vaginitis. Am J
Obstet Gynecol 1955;69:962–7.
53. Merchant JS, Oh MK, Klerman LV. Douching a problem for
adolescent girls and young women. Arch Pediatr Adolesc Med
1999;153:834–7.
54. Bruce FC, Fiscella K, Kendrick JS. Vaginal douching and pre-
term birth: an intriguing hypothesis. Med Hypotheses 2000;54:
448–42.
55. Hillier SL, Nugent RP, Eschenbach DA, et al. Association
between bacterial vaginosis and preterm delivery of a low-
birth-weight infant. N Engl J Med 1995;333:1737–42.
56. Gratocos E, Figueras F, Barranco M, et al. Spontaneous recov-
ery of bacterial vaginosis during pregnancy is not associated
with an improved perinatal outcome. Acta Obstet Gynecol
Scand 1998;77:37–40.
57. Briselden M, Moncla B, Stevens C, et al. Sialidases (neuramini-
dases) in bacterial vaginosis and bacterial vaginosis-associated
microflora. J Clin Microbiol 1992;30:663–6.
58. McGregor JA, French JI, Jones W, 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;170:1048–59.
59. Platz-Christensen JJ, Mattsby-Baltzer I, Thomsen P, et al.
Endotoxin and interleukin-1c in the cervical mucus and vaginal
fluid of pregnant women with bacterial vaginosis. Am J Obstet
Gynecol 1993;169:1161–6.
60. Platz-Christensen JJ, Brandberg A, Wiqvist N. Increased pros-
taglandin concentrations in the cervical mucus of pregnant
women with bacterial vaginosis. Prostaglandins 1992;43:133–
4.
61. Cauci S, Monte R, Driussi S, et al. Impairment of the mucosal
immune system: IgA and IgM cleavage detected in vaginal
washings of a subgroup of patients with bacterial vaginosis. J
 108 Nelson and Macones
Infect Dis 1998;178:1698–706.
62. Imseis HM, Greig PC, Livengood CH 3rd, et al. Characteriza-
tion of the inflammatory cytokines in the vagina during preg-
nancy and labor and with bacterial vaginosis. J Soc Gynecol
Investig 1997;4:90–4.
63. Spandorfer DS, Neuer A, Giraldo PC, et al. Relationship of
abnormal vaginal flora, proinflammatory cytokines and idio-
pathic infertility in women undergoing IVF. J Reprod Med
2001;46:806–10.
64. Purwar M, Ughade S, Bhagat B, et al. Bacterial vaginosis in
early pregnancy and adverse pregnancy outcome. J Obstet
Gynecol Res 2001;27:175–81.
65. McGregor JA, French JL, Parker R, et al. Prevention of prema-
ture birth by screening and treatment for common genital tract
infections: results of a prospective controlled evaluation. Am J
Obstet Gynecol 1995;173:157–67.
66. Ralph SG, Rutherford AJ, Wilson JD. Influence of bacterial
vaginosis on conception and miscarriage in the first trimester:
cohort study. BMJ 1999;319:220–3.
67. Llahi-Camp JM, Rai R, Ison C, et al. Association of bacterial
vaginosis with a history of second trimester miscarriage. Hum
Reprod 1996;11:1575–8.
68. Kurkinen-Raty M, Vuopala S, Koskela M, et al. A randomised
Downloaded from https://academic.oup.com/epirev/article-abstract/24/2/102/535045
by guest
on 22 April 2018
controlled trial of vaginal clindamycin for early pregnancy bac-
terial vaginosis. BJOG 2000;107:1427–32.
69. Kekki M, Kurki T, Pelkonen J, et al. Vaginal clindamycin in
preventing preterm birth and peripartal infections in asympto-
matic women with bacterial vaginosis: a randomized, con-
trolled trial. Obstet Gynecol 2001;97(5 pt 1):643–8.
70. Hauth JC, Goldenberg RL, Andrews WW, et al. Reduced inci-
dence of preterm delivery with metronidazole and erythromy-
cin in women with bacterial vaginosis. N Engl J Med 1995;333:
1732–6.
71. Morales WJ, Schorr S, Albritton J. Effect of metronidazole in
patients with preterm birth in preceding pregnancy and bacte-
rial vaginosis: a placebo-controlled, double blind study. Am J
Obstet Gynecol 1994;171:345–7.
72. McDonald HM, O’Loughlin JA, Vigneswaran R, et al. Impact
of metronidazole therapy on preterm birth in women with bac-
terial vaginosis flora (Gardnerella vaginalis): a randomized
placebo controlled trial. Br J Obstet Gynecol 1997;104:1391–7.
73. Guise JM, Mahon SM, Aickin M, et al. Screening for bacterial
vaginosis in pregnancy. Am J Prev Med 2001;20(3 suppl):62–
72.
74. Guidelines for treatment of sexually transmitted diseases.
MMWR Morb Mortal Wkly Rep 1998;47(RR-1):1–118.
Epidemiol Rev 2002;24:102–108