-
The effect of treating bacterial vaginosis on preterm labor
Christine C. Tebes1, Catherine Lynch1 and John Sinnott2
1Department of Obstetrics and Gynecology, and2Department of
Infectious Diseases, University of South Florida, Tampa, FL
Objective: Multiple studies suggest that bacterial vaginosis
(BV) causes preterm labor; yet its routine treatmentremains
controversial. In order to help to elucidate this controversy, we
performed a thorough review of studieswith levels of evidence
ranging from I to II–II.Methods: We searched for all of the studies
from the years 1994 to 2001 via Medline’s database, includingMD
Consult and Ovid Mednet.Results: Several trials discovered a
decrease in the incidence of preterm labor when BV was treated, but
mostof those trials were performed on women with a history of
preterm labor. However, the majority of trialsreviewed advise
against treatment of a general low-risk obstetric population, as
there was no significant decreasein preterm labor.Conclusions:
Therefore, based on the above studies and the current guidelines of
the Centers for DiseaseControl and Prevention (CDC), treating
pregnant women in high-risk populations who are diagnosed with
BVprovides the clinician with an opportunity to possibly prevent
preterm labor in this population. In nulliparouswomen without a
history of preterm birth, treatment is recommended if other risk
factors are present (e.g.gonorrhea or chlamydia). However, in the
general low-risk populations, routine screening is not
indicated.
Key words: INFECTION; MOBILUNCUS; ANTIBIOTICS; VAGINITIS;
PREMATURITY
Premature births remain a serious problem inthe USA, occurring
in 11% of all pregnancies.Preterm birth, defined as delivery of an
infantbefore 37 weeks’ gestation, is also the leadingcause of
neonatal mortality and morbidity in thedeveloped world. In the USA
alone, an excess of$4 billion1, or 57% of direct nursery costs2 can
beattributed to the care of premature infants. Inaddition, infants
who survive exhibit an increasedrisk of long-term morbidity,
chronic lung disease,cerebral palsy, developmental delay, and
visual andhearing impairment1,2.
Causes of preterm labor vary, but infection ishighly suspect. Up
to 80% of early premature birthsare associated with an intrauterine
infection priorto the rupture of membranes. Asymptomatic
bacteriuria, Neisseria gonorrhea, Chlamydia tracho-matis,
Trichomonas vaginalis and bacterial vaginosis(BV) have all been
associated with an increased riskof preterm birth3. The exception
is vaginal yeastinfections4. Risk factors for preterm delivery
arelisted in Table 1. Multiple studies suggest thatBV is a cause of
preterm labor, yet its routinetreatment remains
controversial5–8.
In conjunction with preterm delivery, BV inpregnant patients is
associated with prematurerupture of membranes, infection of the
amnionand chorion, histologic chorioamnionitis andinfection of
amniotic fluid. Flynn et al.5 noted a60% increase in the risk of
preterm delivery in thepresence of BV. Another study by Hillier et
al.6
associated BV with risk of spontaneous preterm
Infect Dis Obstet Gynecol 2003;11:123–129
Correspondence to: John Sinnott, MD, Department of Infectious
Diseases, University of South Florida, PO Box 1289, Tampa,FL
33601-1289, USA. Email: [email protected]
2003 The Parthenon Publishing Group 123
55Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:53
Color profile: DisabledComposite Default screen
-
birth by a factor of 1.5 to 3 in high-risk women,while other
studies suggest that BV almost doublesthe risk of spontaneous
preterm delivery7.
Although BV is present in almost 20% of preg-nant women, most
cases remain asymptomatic,and not all women with the condition will
deliverprematurely8. It is postulated that there exists a sub-group
of high-risk women (e.g. those who havevaginal colonization with
Mycoplasma hominis).African–American women may also exhibit200–300%
more BV than white populations7,8.The exact conditions under which
BV directlycorrelates with preterm labor are unknown5.
The purpose of this paper is to review therecent literature
addressing the associationbetween BV and preterm birth4.
METHODS
In order to perform this review, we searched theMedline
database, including MD Consult andOvid Mednet for the years
1994–2001. Some ofthe search terms included preterm labor, BV,
treat-ment, metronidazole, clindamycin, and vaginitis.
RESULTS
Hauth et al. (Level I study in 1995) treatedpregnant women with
a history of preterm birth
or weight < 50 kg and a positive diagnosis of BV7.Treatment
consisted of metronidazole (250 mgthree times a day for 7 days) and
erythromycin(333 mg three times a day for 14 days). The resultsof
this 2:1 double-blind randomization trialrevealed a decreased
incidence of preterm delivery(< 37 weeks) for the entire study
population (oddsratio, OR, 0.48; 95% confidence interval,
CI0.28–0.81), as well as for a subset of patients with aprevious
preterm delivery (OR, 0.48; 95% CI,0.25–0.90). Women who were
diagnosed with BVat the initial visit (24 weeks) and who
receivedantibiotics rather than placebo presented withfewer preterm
deliveries. Since this treatmentbenefit was only observed at
initial examination,the overall results do not support
mid-trimestertreatment with metronidazole and erythromycinin women
at risk for preterm delivery withoutBV. There are no data to
suggest that treatmentof low-risk pregnant women with BV
decreasesthe rates of prematurity7.
Over a period of 3 years, Morales et al. (level Istudy in 1994)
screened for BV in women between13 and 20 weeks with a singleton
gestation anda history of preterm birth via a
double-blindrandomized trial. The 80 women with a positivescreen
received either metronidazole (250 mgthree times a day) or a
placebo for 7 days. The studyshowed a significant decrease in
delivery priorto 37 weeks among women taking metronidazole(18%)
compared with those on placebo (39%)(p < 0.05). In the
metronidazole group there weresignificantly fewer hospital
admissions for pretermlabor, cases of premature rupture of
membranes,and low birth weights9.
In 1995, McGregor et al. (level I study)performed a prospective,
controlled treatment trialof 1260 subjects to study the effect of
clinda-mycin on pregnant women with BV. Womenwho were treated with
300 mg of clindamycinorally twice daily for 7 days showed a
reduction inpreterm birth (relative risk 0.5; 95% CI, 0.3–0.9),and
the authors recommended that women at riskfor preterm birth with BV
should be screened andtreated1.
In 1997, McDonald et al.10 conducted amulticenter, randomized,
placebo-controlled trialof 879 women at 19 weeks’ gestation, but
failedto demonstrate a reduced preterm birth rate
Bacterial vaginosis and preterm labor Tebes et al.
124 • INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY
History of preterm birthMultiple gestationBacterial
vaginosisUrinary tract infectionCrack cocaine usageAbruptionSecond-
and third-trimester vaginal bleedingMaternal age under 16 years or
over 40 yearsAfrican–American raceHistory of uterine
surgerySmokingStrenuous workPolyhydramnios and
oligohydramniosCervical infectionsCervical
incompetencePre-eclampsiaPremature preterm rupture of
membranesDomestic violenceRetained intrauterine device
Table 1 Risk factors for preterm labor1,2
56Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:54
Color profile: DisabledComposite Default screen
-
among pregnant women with BV or those with aheavy growth of
Gardnerella vaginalis (level I). Theintention-to-treat analysis
showed no differencebetween the treatment and placebo groups
inoverall preterm birth (31/429 [7.2%] vs. 32/428[7.5%]) or
spontaneous preterm birth rate (20/429[4.7%] vs. 24/428 [5.6%]).
However, amongwomen with a previous preterm birth, thosetreated
with oral metronidazole (400 mg twicedaily for 2 days at 24 weeks’
gestation, and again at29 weeks) demonstrated a marked reduction
inspontaneous preterm birth rate (OR, 0.14; 95%CI,
0.01–0.84)10.
A meta-analysis from the Cochrane databasedetermined that
preterm birth rates did not differsignificantly between treated and
non-treatedpregnant patients with BV (OR, 0.78; 95%CI, 0.60–1.02),
yet a subgroup of women with aprevious preterm birth demonstrated a
significantdecrease in the incidence of preterm birth, with anodds
ratio of 0.37 (95% CI, 0.23–0.60)3. Thismeta-analysis of 1504 women
from a total of fivetrials using amoxicillin, clindamycin and
metro-nidazole did not recommend screening and treat-ing all
pregnant women for BV in order toprevent preterm birth.
Carey et al.11 recently conducted a randomized,double-blind
clinical trial of the use of metro-nidazole to treat asymptomatic
BV. Treatmentdid not reduce the frequency of delivery before37
weeks’ gestation (relative risk in the metro-nidazole group, 1.0;
95% CI, 0.8–1.2). Thisclinical trial differed from previous
attemptsbecause it was larger (1953 subjects) and it studiedthe
general obstetrical population, not just womenwith a history of
preterm delivery. Treatmentinvolved a shorter course of
metronidazole therapywith two 2 g doses during a 48-hour period,
anda second treatment of the women between 24and 30 weeks’
gestation. These researchersrecommended the longer course of
therapy toeradicate upper genital tract organisms11.
The results of this study supported those ofMcDonald et al.10 in
that metronidazole did notreduce preterm delivery among the
generalpopulation of pregnant women, nor did it reducehospital
admissions for preterm labor or pretermpremature rupture of the
membranes, postpartumendometritis, clinical intra-amniotic
infections,
vaginal infections serious enough to require treat-ment, and
treatment with tocolytic drugs. Theplacebo and metronidazole groups
did not differwith regard to incidence of admissions to the
neo-natal intensive-care unit, passage of meconium, orfetal
death/neonatal death. This study foundthat routine screening of
pregnant women forasymptomatic BV and treating the latter with
ashort course of orally administered metronidazoledid not reduce
the risk of preterm birth, despite itseffectiveness in eradicating
BV. Studies do notsupport the use of metronidazole treatment
inpregnant women with asymptomatic BV who areconsidered to be at
either high or low risk forpreterm delivery11. Studies by McDonald
et al.10,Morales et al.9 and Hauth et al.7 did find evidenceof a
decreased incidence of recurrent pre-term delivery among women with
a priorpreterm delivery treated with metronidazole
ormetronidazole/erythromycin.
The findings of Joesoef et al.12 concur with thoseof Carey et
al.11. This multicenter, double-blind,randomized,
placebo-controlled trial failed todemonstrate a reduction in
preterm delivery(level I, 1995) in 745 women between 14 and26
weeks’ gestation who were diagnosed with BV(via Gram stain score
> 6) with 2% clindamycinvaginal cream or with a placebo cream
for 7 days.Although 15% of the women treated withclindamycin had a
preterm delivery, only 13.5% ofplacebo patients did (OR, 1.1; 95%
CI, 0.7–1.7).The authors proposed that the increased frequencyof
preterm delivery might be caused by a transientincrease in vaginal
colonization by Escherichi coliand Enterococcus 1 month after
therapy. Since E. coliis linked to an increased risk of preterm
delivery,this may explain the value of 15%, comparedwith the 13.5%
difference found in the study byJoesoef et al.12.
In 1994, McGregor et al.13 evaluated 271women between 16 and 27
weeks’ gestation in adouble-blind trial (level I). Women who
werediagnosed with BV were treated with 2%clindamycin vaginal cream
or placebo for 7 days.Although 2% clindamycin vaginal cream
waseffective in treating BV during pregnancy(p = 0.001) it was not
permanent, as the conditiongradually returned, and treatment did
not reducethe risk of prematurity during the second trimester.
Bacterial vaginosis and preterm labor Tebes et al.
INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY • 125
57Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:54
Color profile: DisabledComposite Default screen
-
Possible reasons cited include the inadequatepower of the study,
the timing of treatment, andthe use of local vaginal
treatment13.
In 2001, Kekki et al.14 performed a multicenter,randomized,
double-blind, placebo-controlledtrial (level I) which showed that
the treatment ofBV in early pregnancy with vaginal clindamycinfor 7
days did not decrease the rate of pretermdeliveries or peripartum
infections. In 375 ran-domized subjects, preterm delivery occurred
in5% of the clindamycin group and 4% of theplacebo group (OR, 1.3;
95% CI, 0.5–3.5). Theefficacy of intravaginal clindamycin can be as
highas 90% in studies on non-pregnant women, and issimilar to that
of oral metronidazole. The subjectsincluded a low-risk population
of healthy womenwith singleton pregnancies and without a historyof
preterm delivery14. The results are similar to theconclusion of
McGregor et al.13 that topical treat-ment in early pregnancy
reduced vaginal fluidmucinase and sialidase, but failed to reduce
the rateof preterm births14.
A recent randomized controlled trial byKurkinen-Raty et al.15
assessed the efficacy ofvaginal clindamycin in reducing preterm
birth(level I). Of the 1956 women without a history ofpreterm
delivery who were screened at the firstantenatal visit (gestational
week 12), 143 womentested positive for BV. After randomization
andtreatment with clindamycin or placebo, thepreterm birth rate in
the clindamycin group was13.7%, compared with 6.0% in the placebo
group(OR, 2.5; 95% CI, 0.6–10). This study thereforesupports the
evidence that vaginal clindamycintreatment of BV in the first
trimester does notreduce the risk of preterm birth15.
In 1999, French et al.16 evaluated the associationbetween BV,
first-trimester vaginal bleeding andpreterm labor in 1100 pregnant
women who wereenrolled in a prospective observational study
(levelII–II). It was determined that treatment of BV
withclindamycin (300 mg orally twice daily for 7 days)significantly
reduced the risks of preterm birthamong women without
first-trimester bleeding(relative risk, 0.37; 95% CI, 0.16–0.88).
Althoughthe overall population of women with both BVand
first-trimester vaginal bleeding experiencedreductions in preterm
birth, the finding was not
statistically significant (relative risk, 0.52; 95%CI,
0.18–1.55)16.
Several authors have proposed that the timingof the diagnosis
and/or treatment is important ifpreterm labor is to be decreased. A
level II–IIstudy by Meis et al.17 highlighted the effect oftiming
of diagnosis on BV and the incidence ofpreterm labor. Women who
tested negative at 24weeks but were positive when tested again at
28weeks had the highest likelihood of preterm birth(OR, 2.53; 95%
CI, 1.32–4.85; p = 0.005)17.
Riduan et al.18 found the opposite to be true(level II–I trial).
Women who tested positive forBV at 24 weeks and received
antibiotics showed asignificantly decreased incidence of preterm
birth(OR, 0.44; 95% CI, 0.11–1.91). The authorsfound that the
presence of BV at 16 weeks wasmore predictive of preterm delivery
than itspresence at 28 to 32 weeks18.
Table 2 provides a concise summary of thearticles reviewed
above.
DISCUSSION
Preterm labor may be classified as either physio-logic or
pathologic. Physiologic preterm labordescribes a normal initiating
factor that occurs tooearly in pregnancy, while pathologic preterm
laborresults from an abnormal initiating factor withtiming being a
distinguishing factor. The earlier inpregnancy preterm labor
occurs, the more likelyit is that a pathologic etiology exists.
Priorto 16 weeks’ gestation, BV is related to pretermdelivery by a
risk factor of 5 to 7.5. After 26 weeks’gestation, the risk factor
drops to 1.4 to 1.919,20.
BV, an overgrowth of anaerobic species thatproduce protease,
phospholipase A2 andcollagenases in the vagina21, is found in 800
000pregnant women each year11. There is disruptionof the vaginal
ecosystem that results in increasedlevels of anaerobes. BV alters
the vaginal flora bydecreasing the number of
hydrogen-peroxide-producing Lactobacillus acidophilus
organisms.Consequently, the levels of G. vaginalis, M.hominis, and
Mobiluncus species increase ratherthan remaining in their normal
state of suppres-sion. The metabolic by-products of these
Bacterial vaginosis and preterm labor Tebes et al.
126 • INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY
58Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:54
Color profile: DisabledComposite Default screen
-
organisms include amines, which increase thevaginal pH, and
exfoliation of vaginal epithelialcells results22.
Although the exact mechanism is not known,studies have shown
that BV can cause infection ofthe upper genital tract, which acts
as a prematurebirth trigger6. BV and T. vaginalis are also
asso-ciated with many microorganisms that producephospholipase A2
and C or phospholipase-likeactivity, and affected patients show
increased levelsof sialidase, phospholipase A2, prostaglandin E2and
interleukin–1(beta). The rise in levels ofthese enzymes may result
in the decidual or fetalmembrane cell fatty acid tissue stores
releasingarachidonic acid2, a precursor of the
uterotonicprostaglandins.
Other explanations of an association betweenBV and preterm labor
include activation offetal and/or maternal inflammatory responses
orproteolytic enzymes. Elevated vaginal or cervicallevels of
endotoxin, mucinase, sialidase and
interleukin-1(alpha) are found in women with BV,which suggests
that the microorganisms producecytokines6. These cytokines and the
release ofinterleukin-1(beta) and tumor necrosis factorinduce
cyclooxygenase II, an enzyme that pro-duces prostaglandins involved
in parturition.Proteolytic enzymes that may overcome maternalmucous
membrane defenses and impair fetalmembrane strength and elasticity
includecollagenases, immunoglobulin A proteases,elastases,
mucinases and/or sialidases2.
Several studies have concluded that screeningand treating for BV
is futile, and one of the largestclinical trials to date found no
difference in thetreated low-risk pregnant population comparedwith
the placebo11. However, since the clinicaldiagnosis of BV need not
be symptomatic, screen-ing all pregnant women who have risk factors
maybe unnecessary and not cost-effective21, whilethe most
advantageous time to screen and theoptimum dosage of antibiotic is
uncertain.
Bacterial vaginosis and preterm labor Tebes et al.
INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY • 127
Author Year Study type Effect on preterm labor DrugLevel
ofevidence
Hauth et al.7
Morales et al.9
McGregor et al.1
McDonald et al.10
Cochrane database3
Joesoef et al.12
Carey et al.11
McGregor et al.13
Kekki et al.14
Kurkinen-Raty et al.15
French et al.16
1995
19941995
1997
1991 to 2001
1995
2000
1994
2001
2000
1999
Double-blind
Placebo controlledProspective controlledtreatment
trialRandomized, placebo,controlled
Meta-analysis ofcontrolled trials
Double-blind,randomized, placeboRandomized,double-blind
Randomized,double-blindDouble-blind, placebo
Randomized,controlledProspective,observational
Decreased in high-risk
Decreased in high-riskDecreased in women athigh riskDecreased
only in subsetwith a history of pretermlaborNo decrease in
generalpopulation, but a decreasein those at high riskNo decrease
in low-riskpopulationAsymptomatic generalpopulation
withoutdecreaseNo decrease in women at16 to 27 weeks' gestationNo
decrease in low-riskpopulationNo significant decrease inlow-risk
populationDecreased in low-riskpopulation, but notstatistically
significant
Metronidazole/erythromycinMetronidazoleClindamycin
Metronidazole
Amoxicillin,metronidazoleand/or clindamycinMetronidazole
Metronidazole(short course)
Clindamycin
Clindamycin
Clindamycin
Clindamycin
Level I
Level ILevel I
Level I
Level I
Level I
Level I
Level I
Level I
Level I
Level II–II
Table 2 Studies that have evaluated the effect of treatment of
bacterial vaginosis on the incidence of preterm labor
59Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:55
Color profile: DisabledComposite Default screen
-
Although the The Cochrane Library’s review is anexcellent
resource, our paper differs in that wehave investigated different
aspects of this approach.We have included not only level I but also
level IIstudies. Furthermore, we included two studiesthat
investigated the effects of timing on the treat-ment of BV17,18.
However, ultimately our conclu-sion is similar to that of the
Cochrane review3.
To the clinicians who decide to screen forand treat BV, the most
effective antibiotic appearsto be oral metronidazole, which can
besafely administered during pregnancy. Vaginal
clindamycin therapy has shown slight butstatistically
non-significant increases in pretermbirth rate for the reasons
discussed previously23.
Analysis of the reviewed studies supports thecurrent guidelines
of the Center for DiseaseControl. Treating pregnant women in
high-riskpopulations who have been diagnosed with BVcan prevent
preterm labor in these individuals.In nulliparous women without a
history of pre-term birth, the recommendation is to treat ifother
risk factors, such as gonorrhea or chlamydia,are present.
REFERENCES1. McGregor JA, French JI, Parker R, et al.
Preven-
tion of premature birth by screening and treatmentfor common
genital tract infections: results of aprospective controlled
evaluation. Am J ObstetGynecol 1995;173:157–67
2. Andolsek KM, Kelton GM. Risk assessment. PrimCare
2000;27:71–103
3. Brocklehurst P, Hannah M, McDonald H. Inter-ventions for
treating bacterial vaginosis inpregnancy. The Cochrane Library.
Issue 4. Oxford:Update Software, 2001
4. Goldenberg RL, Rouse DJ. Prevention ofpremature birth. N Engl
J Med 2001;339:313–20
5. Flynn CA, Helwig AL, Meurer LN. Bacterialvaginosis in
pregnancy and the risk of prematurity:a meta-analysis. J Fam Pract
1999;48:885–92
6. Hillier SI, Nugent RP, Eschenbach DA, et al. Asso-ciation
between bacterial vaginosis and pretermdelivery of a
low-birth-weight infant. N Engl J Med1995;333:1737–42
7. Hauth JC, Goldenberg RL, Andrews WW, et al.Reduced incidence
of preterm delivery withmetronidazole and erythromycin in womenwith
bacterial vaginosis. N Engl J Med 1995;333:1732–6
8. Lamont RF. Antibiotics for the prevention ofpreterm birth. N
Engl J Med 2000;342:581–3
9. Morales WJ, Schorr S, Albritton J. Effect ofmetronidazole in
patients with preterm birth inpreceding pregnancy and bacterial
vaginosis: aplacebo-controlled, double-blind study. Am JObstet
Gynecol 1994;171:345–9
10. McDonald HM, O’Loughlin JA, Vigneswaran R,et al. Impact of
metronidazole therapy on pre-term birth in women with bacterial
vaginosisflora (Gardnerella vaginalis): a randomized, placebo
controlled trial. Br J Obstet Gynaecol 1997;104:1391–7
11. Carey JC, Klebanoff MA, Hauth JC, et al.Metronidazole to
prevent preterm delivery inpregnant women with asymptomatic
bacterialvaginosis. N Engl J Med 2000;342:534–40
12. Joesoef MR, Hillier SL, Wiknjosastro G, et al.Intravaginal
clindamycin treatment for bacterialvaginosis: effects on preterm
delivery and low birthweight. Am J Obstet Gynecol
1995;173:1527–31
13. McGregor JA, French JI, Jones W, et al. Bacterialvaginosis
is associated with prematurity and vaginalfluid mucinase and
sialidase: results of a controlledtrial of topical clindamycin
cream. Am J ObstetGynecol 1994:170:1048–59
14. Kekki M, Kurki T, Pelkonen J, et al. Vaginalclindamycin in
preventing preterm birth andperipartal infections in asymptomatic
women withbacterial vaginosis: a randomized, controlled
trial.Obstet Gynecol 2001;97:643–8
15. Kurkinen-Raty M, Vuopala S, Koskela M, et al.A randomized,
controlled trial of vaginal clinda-mycin for early pregnancy
bacterial vaginosis.Br J Obstet Gynaecol 2000;107:1427–32
16. French JI, McGregor JA, Draper D, et al. Gesta-tional
bleeding, bacterial vaginosis and commonreproductive tract
infections: risk for pretermbirth and benefit of treatment. Obstet
Gynecol1999;93:715–24
17. Meis PJ, Goldenberg RL, Mercer B. The pretermprediction
study: significance of vaginal infections.Am J Obstet Gynecol
1995;173:1231–5
18. Riduan JM, Hillier SL, Utomo B, et al. Bacterialvaginosis
and prematurity in Indonesia: associationin early and late
pregnancy. Am J Obstet Gynecol1993;169:175–8
Bacterial vaginosis and preterm labor Tebes et al.
128 • INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY
60Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:56
Color profile: DisabledComposite Default screen
-
19. Kurki T, Sivonen A, Renkonen OV, et al. Bacterialvaginosis
in early pregnancy and pregnancy out-come. Obstet Gynecol
1992;80:173–7
20. Hay PE, Lamont RF, Taylor-Robinson D, et al.Abnormal
bacterial colonization of the genital tractand subsequent preterm
delivery and late mis-carriage. Br Med J 1994;308:295–8
21. McCoy M, Katz C, Vern L, et al. Bacterial vaginosisin
pregnancy: an approach for the 1990s. ObstetGynecol Surv
1995;50:482–8
22. Egan ME, Lipsky MS. Diagnosis of vaginitis.Am Fam Physician
2000;62:1095–104
23. Burtin P, Taddio A, Ariburnu O, et al. Safetyof
metronidazole in pregnancy: a meta-analysis.Am J Obstet Gynecol
1995;172:525–9
RECEIVED 11/06/02; ACCEPTED 05/02/03
Bacterial vaginosis and preterm labor Tebes et al.
INFECTIOUS DISEASES IN OBSTETRICS AND GYNECOLOGY • 129
61Z:\Customer\PARTHEN\IDOG\A4577 - IDOG Vol 11 No 2 - June
2003.vp09 July 2003 14:29:56
Color profile: DisabledComposite Default screen
-
Submit your manuscripts athttp://www.hindawi.com
Stem CellsInternational
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
MEDIATORSINFLAMMATION
of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Behavioural Neurology
EndocrinologyInternational Journal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Disease Markers
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
BioMed Research International
OncologyJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Oxidative Medicine and Cellular Longevity
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
PPAR Research
The Scientific World JournalHindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Immunology ResearchHindawi Publishing
Corporationhttp://www.hindawi.com Volume 2014
Journal of
ObesityJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Computational and Mathematical Methods in Medicine
OphthalmologyJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Diabetes ResearchJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Research and TreatmentAIDS
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Gastroenterology Research and Practice
Hindawi Publishing Corporationhttp://www.hindawi.com Volume
2014
Parkinson’s Disease
Evidence-Based Complementary and Alternative Medicine
Volume 2014Hindawi Publishing
Corporationhttp://www.hindawi.com