Pregnancy outcome after early detection of bacterial vaginosis
Brunella Guerraa, Tullio Ghia,*, Simona Quartaa, Antonio Maria Morselli-Labateb,
Tiziana Lazzarottoc, Gianluigi Pilua, Nicola Rizzoa
aDepartment of Obstetrics and Gynecology, Policlinico S.Orsola-Malpighi, Alma Mater Studiorum-University of Bologna, Italy
bDepartment of Clinical and Experimental Medicine, Section of Microbiology, Policlinico S.Orsola-Malpighi,
Alma Mater Studiorum-University of Bologna, Italy
cDepartment of Internal Medicine and Gastroenterology, Policlinico S.Orsola-Malpighi, Alma Mater Studiorum-University of Bologna, Italy
Received 6 April 2005; received in revised form 11 October 2005; accepted 30 December 2005
Objective: To assess if detecting bacterial vaginosis either in early pregnancy or at midtrimester may predict adverse pregnancy outcome in
women at risk for preterm delivery.
10 + 0 weeks) or in the second one (24–26 weeks). Adverse outcome was intended as miscarriage (?25 weeks), or premature delivery
(?36 + 6).
Results: The risk of adverse pregnancy outcome was significantly increased in women diagnosed at first trimester with bacterial vaginosis
(OR: 4.56; 95% CI: 2.54–8.93); the same finding at midtrimester did not increase significantly the risk of preterm delivery.
Conclusions: Early screening for bacterial vaginosis in pregnant women who experienced a preterm delivery may help in predicting the risk
of adverse outcome.
# 2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Bacterial vaginosis; Vaginal flora; Preterm delivery; Miscarriage
Ascendinginfections ofthe lowergenital tract inpregnant
women are associated with preterm delivery (PTD) [1–4].
Bacterial vaginosis in pregnancy may favour ascending
infectionsand isconsidereda risk factor for adverseoutcome
such as preterm delivery, preterm rupture of membranes or
miscarriage [5–7]. Bacterial vaginosis is more commonly
reported in women who deliver or rupture the membranes
prematurely [2,6–13]. In order to decrease the preterm
delivery rate, screening for bacterial vaginosis during the
second trimester of pregnancy followed by antibiotic
administration in the positive cases has been proposed. Such
with a previous PTD [14–16], is of dubious efficacy in the
general population [15,17–26]. Some observations have
suggested that premature delivery may be consequence of a
latent genital infection which has established very early in
vaginosis, the stronger the association with premature
delivery [10,11,13]. Moreover, when serial assessments have
been performed throughout the pregnancy, an increased risk
of preterm delivery was maintained by women who screened
positive at first evaluation and appeared negative at a second
the first smear and become positive later on, the risk of
preterm delivery is not significantly different from persis-
tently negative cases [27,28]. The explanation for these
findings may be that an abnormal vaginal flora since early
gestation may interfere with the immunologic surveillance
throughout the pregnancy . Furthermore, only an early
European Journal of Obstetrics & Gynecology and
Reproductive Biology 128 (2006) 40–45
* Corresponding author at: Clinica Ostetrica e Ginecologica dell’Uni-
versita ` di Bologna, Policlinico S.Orsola- Malpighi, Via Massarenti 13,
40100 Bologna, Italy. Tel.: +39 051 6364411; fax: +39 051 301994.
E-mail address: firstname.lastname@example.org (T. Ghi).
0301-2115/$ – see front matter # 2006 Elsevier Ireland Ltd. All rights reserved.
acquired vaginosis may trigger the biochemical events which
The aim of our study was to assess if detection of
bacterial vaginosis either in early pregnancy or at
midtrimester may predict a miscarriage or a premature
delivery in women at risk for preterm delivery.
We also searched for additional factors other than
vaginosis that may increase the risk of such unfavourable
2. Materials and methods
From September 1997 to December 2000, 264 pregnant
women with a previous preterm delivery were consecutively
referred to the antenatal Clinic of our Department of
Obstetrics and Gynecology.
pregnancy prior to 10 weeks of gestation in women with at
least a previous premature delivery (delivery > 25 + 0 and
?36 + 6 weeks of gestation).
Exclusion criteria were multiple pregnancies, drug abuse,
suspected or known mullerian anomalies.
Overall, 242 women fulfilled the inclusion criteria and
after sonographic confirmation of gestational age were
enrolled in the study.
2.2. Bacterial vaginosis
All the patients enrolled in the study accepted to undergo
screening of bacterial vaginosis. The presence of bacterial
the first visit (?9 + 6 weeks) and at midtrimester (24–26
weeks) all participants were offered a smear test of vaginal
fluid. Fluid was sampled from the vaginal walls by a cotton
swab prior standard examination, smeared on a slide, air
dried and then Gram stained.
The evaluation of each vaginal smear was blinded by the
laboratoryto the clinician and tothe patient. The presence of
bacteria was evaluated on the smear in a semiquantitative
manner using the 10 point score described by Nugent et al.
 and validated in pregnancy by Hillier et al. .
According to that, a score between 0 and 3 is obtained with
normal vaginal flora, a score between 4 and 6 is assigned to
intermediate flora, and a score between 7 and 10 to bacterial
vaginosis. We divided positive findings in two different
classes: abnormal flora (score ? 4, including intermediate
flora and bacterial vaginosis) and bacterial vaginosis
(score ? 7). Patients with positive findings were not treated.
2.3. Other risk factors
From the study group, a series of variables which have
shown to increase the risk of preterm delivery (putative risk
Afro-Caribbean, or others), maternal age (?20 years, 21–34
years, ?35 years), smoking, scholarity (?13 years or >13
years), historyof previous
loss ? 25 weeks of gestation), number of previous preterm
delivery (two or more versus one), number of partners who
fathered the pregnancies (two or more versus one).
Gestational age at delivery was investigated for each
pregnancy using the hospital records or telephonic inter-
views. We split up our population in cases with favourable
pregnancy outcome (delivery > 36 + 6 weeks of gestation),
and those with adverse pregnancy outcome (delivery ?
36 + 6 weeks of gestation).
Among adverse pregnancy events we distinguished
miscarriages (pregnancy loss ? 25 weeks of gestation),
and premature deliveries (delivery > 25 + 0 and ?36 + 6
weeks of gestation).
The study was carried out in accordance to the ethical
rules of St. Orsola-Malpighi General Hospital, Bologna,
Italy. A signed consent form was obtained by all patients
who participated to the study.
Means, standarddeviations, and frequencies were used as
descriptive statistics. The effects of factors influencing the
outcome and the vaginal flora were analyzed by means of
univariate logistic regression. Stepwise multivariate logistic
regressions, taking into account each of the different pattern
of vaginal flora together with all the other putative factors,
were also performed in order to identify independent factors
related to the outcome. The odds ratios (OR) and their 95%
confidence intervals (95% CI) computed by the logistic
regression were also reported.
Two-tailed P values less than 0.05 were considered
statistically significant. All statistical analyses were per-
formed by means of a personal computer and the SPSS for
Windows version 8.0 statistical software.
Of the 242 women who underwent the first testing
(?9 + 6 weeks of gestation) 147 presented abnormal flora
(60.7%) including 95 with bacterial vaginosis (39.3% of
group 38 (15.7%) women were missing because of earlier
miscarriage (n = 32) or because they did not attend the
vaginal sampling (n = 6).Amongthe remaining204 women,
B. Guerra et al./European Journal of Obstetrics & Gynecology and Reproductive Biology 128 (2006) 40–4541
59 (28.9%) had abnormal flora including 36 with bacterial
vaginosis (17.6% of overall cases). No new case of bacterial
vaginosis was picked up among the 147 women with a
normal or intermediate flora in the first trimester smear. All
the 88 women with vaginal score less than 4 at first trimester
confirmed the normal flora at the second testing, whereas
normal flora spontaneously resumed in 57 out of 116
(49.1%) with previous abnormal flora including 33 out of 68
(48.5%) with previous bacterial vaginosis.
A statistically significant association was noted for
abnormal flora with Afro-Caribbean ethnicity either at first
(OR: 4.99; 95% CI: 2.02–12.4) or at second (OR: 2.26; 95%
CI: 1.05–4.82) testing. Furthermore, abnormal flora app-
eared more frequent at midtrimester in pregnant of younger
age (OR: 6.6; 95% CI: 1.24–35.2).
Pregnancy outcome was favourable in 132 cases (54.5%)
and unfavourable in the remaining 110 (45.5%). Among
these 110 pregnancies with unfavourable outcome, mis-
carriage occurred in 32 cases (13.2%), and premature
deliveryin78 (32.2%), with 17 (7.0%)of theselatterwomen
delivering prior to 32 completed weeks of gestation (early
The correlation between vaginal score at either visit and
pregnancy outcome is displayed in Table 1. An abnormal
result of vaginal testing prior to 10 completed weeks of
gestation was significantly associated with an unfavourable
The risk of both events was almost five-fold increased either
with abnormal flora (OR: 4.95; 95% CI: 2.77–8.84) or with
preterm delivery only, was also significantly increased with
abnormal flora (OR: 3.25; 95% CI: 1.75–6.04) as with
bacterial vaginosis (OR: 2.44; 95% CI: 1.24–4.79).
At midtrimester, the detection of abnormal flora was
again significantly more common among those patients who
delivered prematurely (OR: 2.29; 95% CI: 1.23–4.27)
whereas bacterial vaginosis only did not increase the risk of
In Table 2 pregnancy outcome is displayed according to
anamnestic data. A significant association with adverse
outcome is noted only for women with previous mis-
carriages (OR 1.86; 95% CI: 1.11–3.13).
The results of the multivariate stepwise logistic regres-
sions performed with the outcome as dependent variable and
the anamnestic data together with the vaginal flora as
independent variables, are reported in Table 3. The analyses
were performed twice taking into account either abnormal
flora or bacterial vaginosis within the set of the independent
variables. The OR, their 95% CI and the P values of the
independent variables that entered the stepwise procedures
are shown in the table. Multivariate analysis confirmed all
significant relationships between pregnancy outcome and
B. Guerra et al./European Journal of Obstetrics & Gynecology and Reproductive Biology 128 (2006) 40–45 42
Relationship between outcome and vaginal score obtained either at first or second trimester of pregnancy (univariate logistic regression)
Miscarriage or delivery ?36 + 6 weeks
No. of cases ORa(95% CI)
Delivery ?36 + 6 weeks
No. of casesP ORa(95% CI)P
Early pregnancy (?9 + 6 weeks; n = 242)
Normal flora (score < 4; n = 95)
Abnormal flora (score ? 4; n = 147)
Bacterial vaginosis (score ? 7; n = 95)
Midtrimester (24–26 weeks; n = 204)b
Normal flora (score < 4; n = 145)
Abnormal flora (score ? 4; n = 59)
Bacterial vaginosis (score ? 7; n = 36)
ND: not done.
aOR vs. normal flora.
bThirty-two patients with previous miscarriage and 6 patients who did not attend the second vaginal sampling were excluded.
Anamnestic risk factors and pregnancy outcome
Miscarriage or delivery ? 36 + 6 weeks
(n = 110)
No. of casesOR (95% CI)P
Afro-Caribbean (n = 43)
Others (n = 199)
?20 years (n = 10)
?21years (n = 232)
?13 years (n = 130%)
>13 years (n = 112)
Yes (n = 14)
No (n = 228)
More (n = 86)
1 (n = 156)
Number of previous PTDa
>1 (n = 75)
1 (n = 167)
0.62 (0.35–1.08) 0.090
Yes (n = 99)
No (n = 143)
vaginal flora which had been previously shown at univariate
analysis thus indicating that the influence of vaginal flora on
the outcome is independent on other possible confounding
factors. In early pregnancy, abnormal flora and previous
miscarriage, as well as bacterial vaginosis and previous
miscarriages were the two patterns of variables which
increased independently the risk of unfavourable outcome
(miscarriage or premature delivery)
On the other hand, as long as only premature delivery is
concerned, abnormal flora or bacterial vaginosis at early
pregnancy were the unique variables increasing the risk of
this event. No variable entered the stepwise procedure when
preterm delivery was related to the bacterial vaginosis at
midtrimester together with the other confounding factors.
Our data show that in women at risk for preterm delivery
an adverse outcome is more likely if bacterial vaginosis or
abnormal flora only are detected in the first trimester. In fact,
both preterm delivery and second trimester miscarriages
were more common among the patients who screened
positiveatfirst vaginalsmear (score ? 4) than inthe patients
who did not (score < 4). The risk of adverse pregnancy
outcome and of premature delivery only were in fact
significantly increased either in patients with abnormal flora
or in those with bacterial vaginosis. Moreover, a significant
association with miscarriage or preterm delivery was
confirmed for cases with a history of previous miscarriages.
As shown by multivariate analysis, the risk of an adverse
outcome was particularly high whether a patient with
previous miscarriages had a positive vaginal smear in early
pregnancy. In presence of both these predictors, the risk of
adverse outcome was 10-fold increased.
Furthermore, as previously reported either on selected as
on unselected groups [2,6–13,18–23,28–29], our data
confirm an association between an abnormal vaginal testing
at second trimester and the risk of premature delivery. The
occurrence of preterm delivery appeared significantly more
likely in the group with abnormal flora (score ? 4), but in
patients with bacterial vaginosis (score ? 7) the risk of
labouring prior to 37 weeks was not further increased. This
finding may be possibly due to the small number of cases of
bacterial vaginosis we picked up at this stage. In fact, 24–26
weeks of pregnancy is a relatively late timing for the
midtrimester smear and previous studies had demonstrated
that the more advanced the gestational age at testing the
lower the detection rate of vaginosis and the lower its
predictive value for preterm delivery [13,18,28]. However,
of note is that having a positive smear at midtrimester
(score ? 4) in pregnant with previous preterm delivery
doubles the risk of recurrence.
Some differences with previous studies need to be
specifically addressed. A rate of 45% for adverse outcome
(miscarriage and preterm delivery) is higher than what
generally quoted even in high risk patients. Moreover, the
frequency of vaginosis in our group of pregnant women is
trimester a 39% of bacterial vaginosis is notably high,
compared to the rate of 20% which is generally quoted by
were in fact unexpected, some reasonable explanations may
The early timing of patient recruitment (?9 + 6 weeks)
compared to previous studies may justify the higher rate of
adverse pregnancy outcome. In fact, due to the earlier
enrollment a higher number of spontaneous miscarriages is
expected. Most of them are possibly unrelated to vaginosis
and reflect the pregnancy selection which takes place in the
Moreover, the very early timing of vaginal sampling may
patients tested at 24–26 weeks, the rate of bacterial vaginosis
falls to 18% which is comparable to previous studies. A
the pregnancy, in absence of any medical therapy, had been
previously reported [27,29] and may be related to the
B. Guerra et al./European Journal of Obstetrics & Gynecology and Reproductive Biology 128 (2006) 40–45 43
Results of multivariate logistic regression for the evaluation of the factors influencing the outcome
Pattern of independent variables entered the procedureMiscarriage or delivery
< 36 + 6 weeks
Delivery < 36 + 6 weeks
OR (95% CI)P OR (95% CI)P
Early pregnancy (?9 + 6 weeks)
>1 previous preterm deliveries
Midtrimester (24–26 weeks)
ND: not done; None: no variables entered the procedure.
to vaginosis was noted at midtrimester.
As mentioned above, a higher preterm delivery rate had
been previously reported, either in the general population or
in high-risk women, when bacterial vaginosis was detected
in pregnancy [2,6–13,18,28,34,36]. In a selected population
our study seems to confirm such data. However, a main
difference with previous studies is that vaginal samples have
been collected early in pregnancy, between 7 and 10
completed weeks of pregnancy. In previous studies
gestational age at the time of vaginal smear varied between
12 and 28 weeks of pregnancy. Only in a few studies vaginal
testing had been started in the first trimester, from 10 weeks
on [10,11,18]. In those studies, the correlation of positive
cases with preterm delivery appeared stronger than for
patients diagnosed in the second trimester. Our paper
confirms that if vaginosis is detected early in pregnancy an
adverse outcome is more likely to occur than in cases in
which vaginosis is detected later on.
It would be interesting on a large scale to test the
and the preterm delivery rateinthegeneral population.Ifthis
association is also confirmed in the general population, this
in the second trimester has proved mostly inefficient in
decreasing the preterm delivery rate [15,17,18,19]. On the
other hand, some large randomised trials recently reported a
vaginosis was detected and cured between the first trimester
(12–15 weeks) and 20 weeks of gestation [20–23]. However,
some consecutive metaanalysis of the published trials have
repeatedly shown that screening and treating all pregnant
women for bacterial vaginosis is not effective in preventing
preterm birth [24–26]. On the other hand, for women with a
treatment of bacterial vaginosis may reduce the risk of
[25–26]. In particular, the anticipation of the therapy to the
first trimester in positive patients may prove beneficial .
Actually, the patients who screened positive in the second
trimester are mostly infected since the first trimester. This
was documented in this study and previously reported in
other studies where serial assessment of vaginal flora had
been performed . In women who screened positive in the
late midtrimester, antibiotic failure in affecting pregnancy
vaginosis. A long-standing contamination of the lower
genital tract in fact is more likely to have triggered the
a recent one. The role of a long standing subclinical
inflammation in the genesis of preterm labour has been
highlighted by Ghezzi et al. , who showed that higher
amniotic levels of C-reactive protein in low-risk patients
undergoing genetic amniocentesis are significantly asso-
ciated with the risk of a premature delivery. Moreover, the
risk of preterm labour seems to be higher in women with
uterine contractions but intact membranes who show
increased levels of insulin-like growth factor-binding
protein-1 (IGFBP-1) in the cervical secretion . The
activation and release of IGFBP-1 by decidua is part of the
biochemical cascade which is elicited by a inflammatory
Therefore, clearing a bacterial vaginosis detected early in
pregnancy may have a positive impact on the outcome of the
pregnancy, if fetal inflammation has not yet been elicited. In
population, new randomised controlled trials assessing the
effect of early vaginosis eradication on the pregnancy
outcome would be necessary in the short term.
 Minkoff H. Prematurity: infection as an etiologic factor. Obstet
 Minkoff H, Grunebaum AN, Schwarz RH, et al. Risk factors for
prematurity and premature rupture of membranes: a prospective study
of the vaginal flora in pregnancy. Am J Obstet Gynecol 1984;150(8):
 Carey JC, Yaffe SJ, Catz C. The Vaginal Infections and Prematurity
Study: an overview. Clin Obstet Gynecol 1993;36(4):809–20.
 Terzidou V, Bennett PR. Preterm labour. Curr Opin Obstet Gynecol
 Eschenbach DA, Gravett MG, Chen KC, Hoyme UB, Holmes KK.
Bacterial vaginosisduringpregnancy.An association withprematurity
and postpartum complications. Scand J Urol Nephrol Suppl 1984;86:
 Gravett MG, Hummel D, Eschenbach DA, Holmes KK. Preterm labor
associated with subclinical amniotic fluid infection and with bacterial
vaginosis. Obstet Gynecol 1986;67(2):229–37.
 Gravett MG, Nelson HP, DeRouen T, Critchlow C, Eschenbach DA,
Holmes KK. Independent associations of bacterial vaginosis and
Chlamydia trachomatis infection with adverse pregnancy outcome.
 Goldenberg RL, Iams JD, Mercer BM, et al. The preterm prediction
study: the value of new vs standard risk factors in predicting early and
all spontaneous preterm births. NICHD MFMU Network. Am J Public
 Hillier SL, Martius J, Krohn M, Kiviat N, Holmes KK, Eschenbach
DA. A case–control study of chorioamnionic infection and histologic
chorioamnionitis in prematurity. N Engl J Med 1988;319(15):
 Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson
J. Abnormal bacterial colonisation of the genital tract and subsequent
preterm delivery and late miscarriage. BMJ 1994;308(6924):295–8.
 Kurki T, Sivonen A, Renkonen OV, Savia E, Ylikorkala O. Bacterial
vaginosis in early pregnancy and pregnancy outcome. Obstet Gynecol
 Kurkinen-Raty M, Vuopala S, Koskela M, et al. A randomised
controlled trial of vaginal clindamycin for early pregnancy bacterial
vaginosis. Br J Obstet Gynecol 2000;107(11):1427–32.
 Leitich H, Bodner-Adler B, Brunbauer M, Kaider A, Egarter C,
Husslein P. Bacterial vaginosis as a risk factor for preterm delivery:
a meta-analysis. Am J Obstet Gynecol 2003;189(1):139–47.
 Hauth JC, Goldenberg RL, Andrews WW, DuBard MB, Copper RL.
Reduced incidence of preterm delivery with metronidazole and ery-
thromycin in women with bacterial vaginosis. N Engl J Med
 McDonald HM, O’Loughlin JA, Vigneswaran R, et al. Impact of
metronidazole therapy on preterm birth in women with bacterial
B. Guerra et al./European Journal of Obstetrics & Gynecology and Reproductive Biology 128 (2006) 40–45 44
vaginosis flora (Gardnerella vaginalis): a randomised, placebo con- Download full-text
trolled trial. Br J Obstet Gynaecol 1997;104(12):1391–7.
 Morales WJ, Schorr S, Albritton J. Effect of metronidazole in patients
with preterm birth in preceding pregnancy and bacterial vaginosis: a
placebo-controlled, double-blind study. Am J Obstet Gynecol 1994;
171(2):345–7 [Discussion 8–9].
 Carey JC, Klebanoff MA, Hauth JC, et al. Metronidazole to prevent
preterm delivery in pregnant women with asymptomatic bacterial
vaginosis.NationalInstitute of ChildHealthand HumanDevelopment
Network of Maternal-Fetal Medicine Units. N Engl J Med 2000;
 Joesoef MR, Hillier SL, Wiknjosastro G, et al. Intravaginal clinda-
mycin treatment for bacterial vaginosis: effects on preterm delivery
and low birth weight. Am J Obstet Gynecol 1995;173(5):1527–31.
 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(4):1048–59 [Discussion 59–60].
 Leitich H, Brunbauer M, Bodner-Adler B, Kaider A, Egarter C,
Husslein P. Antibiotic treatment of bacterial vaginosis in pregnancy:
a meta-analysis. Am J Obstet Gynecol 2003;188(3):752–8.
 Ugwumadu A, Manyonda I, Reid F, Hay P. Effect of early oral
women with abnormal vaginal flora and bacterial vaginosis: a rando-
mised controlled trial. Lancet 2003;361(9362):983–8.
 Lamont RF, Duncan SL, Mandal D, Bassett P. Intravaginal clinda-
mycin to reduce preterm birth in women with abnormal genital tract
flora. Obstet Gynecol 2003;101(3):516–22.
 Kiss H, Petricevic L, Husslein P. Prospective randomised controlled
trial of an infection screening programme to reduce the rate of preterm
delivery. BMJ 2004;329(7462):371.
 Brocklehurst P, Hannah M, McDonald H. Interventions for treating
bacterial vaginosis in pregnancy. Cochrane Database Syst Rev
 McDonald H, Brocklehurst P, Parsons J, Vigneswaran R. Antibiotics
for treating bacterial vaginosis in pregnancy. Cochrane Database Syst
Rev 2003;(2):CD000262 [Review].
 McDonald H, Brocklehurst P, Parsons J. Antibiotics for treating
bacterial vaginosis in pregnancy. Cochrane Database Syst Rev
 Hay PE, Morgan DJ, Ison CA, et al. A longitudinal study of bacterial
vaginosis during pregnancy. Br J Obstet Gynaecol 1994;101(12):
 Gratacos E, Figueras F, Barranco M, et al. Spontaneous recovery of
bacterial vaginosis during pregnancy is not associated with an
improved perinatal outcome. Acta Obstet Gynecol Scand 1998;
 Ugwumadu AH. Bacterial vaginosis in pregnancy. Curr Opin Obstet
 Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial
vaginosis is improved by a standardized method of gram stain inter-
pretation. J Clin Microbiol 1991;29(2):297–301.
 Hillier SL, Krohn MA, Nugent RP, Gibbs RS. Characteristics of three
vaginal flora patterns assessed by gram stain among pregnant women.
Vaginal Infections and Prematurity Study Group. Am J Obstet Gyne-
 Cristiano L, Rampello S, Noris C, Valota V. Bacterial vaginosis:
prevalence in an Italian population of asymptomatic pregnant women
and diagnostic aspects. Eur J Epidemiol 1996;12(4):383–90.
 Mead PB. Epidemiology of bacterial vaginosis. Am J Obstet Gynecol
1993;169(2 Pt 2):446–9.
 Kekki M, Kurki T, Pelkonen J, Kurkinen-Raty M, Cacciatore B,
Paavonen J. Vaginal clindamycin in preventing preterm birth and
peripartalinfectionsin asymptomatic womenwith bacterialvaginosis:
a randomized, controlled trial. Obstet Gynecol 2001;97(5 Pt 1):
 Ghezzi F, Franchi M, Raio L, et al. Elevated amniotic fluid C-reactive
protein at the time of genetic amniocentesis is a marker for preterm
delivery. Am J Obstet Gynecol 2002;186(2):268–73.
 Guaschino S, Ricci E, Franchi M, et al. Treatment of asymptomatic
bacterial vaginosis to prevent pre-term delivery: a randomised trial.
Eur J Obstet Gynecol Reprod Biol 2003;110(2):149–52.
 Kekki M, Kurki T, Karkkainen T, Hiilesmaa V, Paavonen J, Rutanen
EM.Insuline-likegrowth factor-bindingprotein 1in cervical secretion
as a predictor of preterm delivery. Acta Obstet Gynecol Scand 2001;
binding protein-1 in cervix as a marker of infectious complications in
pregnant women with bacterial vaginosis. Lancet 1999;353(9163):
B. Guerra et al./European Journal of Obstetrics & Gynecology and Reproductive Biology 128 (2006) 40–4545