Probiotics for the treatment of bacterial vaginosis (Review) The Cochrane Library

Probiotics for the treatment of bacterial vaginosis (Review)
Senok AC, Verstraelen H, Temmerman M, Botta GA
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2009, Issue 4
http://www.thecochranelibrary.com
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Probiotic + Metronidazole versus placebo + Metronidazole, Outcome 1 Microbiological cure.
Analysis 2.1. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 1 Microbiological
cure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.2. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 2 Resolution of
symptoms as reported by patient. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.3. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 3 Adverse effects. .
Analysis 3.1. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules), Outcome 1
Microbiological cure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.2. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules), Outcome 2 Resolution
of symptoms as reported by patient. . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.3. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules), Outcome 3 Clinical
cure as reported by physician. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.1. Comparison 4 Probiotics with estriol versus placebo, Outcome 1 Microbiological cure. . . . . . .
Analysis 4.2. Comparison 4 Probiotics with estriol versus placebo, Outcome 2 Clinical cure as reported by physician.
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
1
2
2
3
3
5
7
8
9
10
10
10
13
20
21
21
22
22
23
23
24
24
25
25
27
27
27
27
28
28
i
[Intervention Review]
Probiotics for the treatment of bacterial vaginosis
Abiola C Senok1 , Hans Verstraelen2 , Marleen Temmerman2 , Giuseppe A Botta3
1 Department of Clinical Sciences, College of Medicine, University of Sharjah, Sharjah, United Arab Emirates. 2Department of Obstetrics
and Gynaecology, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium. 3 Dept of Medical and Morphological
Research, Section of Microbiology, School of Medicine, University of Udine, Udine, Italy
Contact address: Abiola C Senok, Department of Clinical Sciences, College of Medicine, University of Sharjah, Sharjah, United Arab
Emirates. [email protected]
Editorial group: Cochrane Sexually Transmitted Diseases Group.
Publication status and date: New, published in Issue 4, 2009.
Review content assessed as up-to-date: 31 December 2008.
Citation: Senok AC, Verstraelen H, Temmerman M, Botta GA. Probiotics for the treatment of bacterial vaginosis. Cochrane Database
of Systematic Reviews 2009, Issue 4. Art. No.: CD006289. DOI: 10.1002/14651858.CD006289.pub2.
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT
Background
The dominance of lactobacilli in healthy vaginal microbiota and its depletion in bacterial vaginosis (BV) has given rise to the concept
of oral or vaginal instillation of probiotic Lactobacillus strains for the management of this condition.
Objectives
To ascertain the efficacy of probiotics in the treatment of BV.
Search methods
We searched electronic databases irrespective of publication status or language. These included: Cochrane Central Register of Controlled
Trials (CENTRAL), the HIV/AIDS and STD Cochrane Review Groups’ specialized registers, the Cochrane Complementary Medicine
Field’s Register of Controlled Trials, MEDLINE (1966 to 2008), EMBASE (1980 to 2007), ISI science citation index (1955 to 2007),
CINAHL (Cumulative Index to Nursing & Allied Health Literature (1982 to 2007).
We handsearched of specialty journals, conference proceedings and publications list on the website of the International Scientific
Association of Probiotics and Prebiotics (http://www.isapp.net/default.asp).
For unpublished studies or ongoing trials, we contacted authors from relevant publications, nutraceutical companies and probioticrelated scientific associations. We searched electronic databases on ongoing clinical trials.
Selection criteria
Randomized controlled trials using probiotics for the treatment of women of any age diagnosed with bacterial vaginosis, regardless
of diagnostic method used. The probiotic preparation could be single or “cocktail” of strains, any preparation type/dosage/route of
administration. Studies comparing probiotics with placebo, probiotics used in conjunction with conventional antibiotics compared
with placebo or probiotics alone compared with conventional antibiotics were eligible for inclusion.
Data collection and analysis
We screened titles and abstracts , obtained full reports of relevant trialsand independently appraised them for eligibility. A data extraction
form was used to extract data from the four included studies. For dichotomous outcomes, odds ratios (OR) and 95% confidence
intervals (CI) were derived for each study using RevMan (versions 4.2 and 5). We did not perform meta-analysis due to significant
differences in the probiotic preparations and trial methodologies.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
Main results
Analysis suggests beneficial outcome of microbiological cure with the oral metronidazole/probiotic regimen (OR 0.09 (95% CI 0.03
to 0.26)) and the probiotic/estriol preparation (OR 0.02, (95% CI 0.00 to 0.47)). For the probiotic/estriol preparation, the OR and
95% CI for physician-reported resolution of symptoms was OR 0.04 (95% CI 0.00 to 0.56).
Authors’ conclusions
The results do not provide sufficient evidence for or against recommending probiotics for the treatment of BV. The metronidazole/
probiotic regimen and probiotic/estriol perparation appear promising but well-designed randomized controlled trials with standardized
methodologies and larger patient size are needed.
PLAIN LANGUAGE SUMMARY
Probiotics for the treatment of bacterial vaginosis
Bacterial vaginosis (BV) is one of the most common causes of genital discomfort in women of reproductive age. This condition occurs
when there is an imbalance in the population of normal vaginal micro-organism with depletion of the dominant lactobacilli and
overgrowth of other types of bacteria. Treatment of this condition using recommended antibiotics is often associated with failure and
high rates of recurrence. This led to the concept of replacing the depleted lactobacilli using probiotic strains as a treatment approach.
This review investigated the evidence for the use of probiotic preparations either alone or in conjunction with antibiotics for the
treatment of BV. The current research does not provide conclusive evidence that probiotics are superior to or enhance the effectiveness
of antibiotics in the treatment of BV. In addition, there is insufficient evidence to recommend the use of probiotics either before,
during or after antibiotic treatment as a means of ensuring successful treatment or reduce recurrence. Larger, well-designed randomized
controlled trials with standardized methodologies are needed to confirm the benefits of probiotics in the treatment of BV.
BACKGROUND
1997; Royce 1999). The prevalence of BV varies in different parts
of the world and is higher in developing countries, but there is
still some controversy about whether or not BV is a sexually transmitted disease in the “traditional” sense (Larsson 2005). However,
current data indicate that the overall prevalence of BV is much
higher among STD clinic attendees and commercial sex workers
(Behets 2001; Eschenbach 1988).
Bacterial vaginosis (BV) is one of the most common causes of
genital discomfort in women of reproductive age. This condition
occurs as a result of an imbalance in the normal vaginal microbiota and is characterized by a decrease or depletion of the Lactobacilli spp and a concomitant overgrowth of other vaginal anaerobic bacteria. The micro-organisms commonly cultured in BV
include Gardnerella vaginalis, Mobiluncus spp, Mycoplasma hominis, Prevotella spp, Preptostreptococcuss spp and more recently, new
molecular methods have identified Atopobium vaginae as a BVassociated microbe (Fredricks 2005; Marrazzo 2004; Verstraelen
2004). Under normal conditions, the production of lactic acid
by vaginal lactobacilli, particularly by strong hydrogen peroxide
(H2 O2 ) producers such as Lactobacillus crispatus and Lactobacillus
jensenii, help to maintain the low vaginal pH which prevents the
overgrowth of other anaerobic bacteria (Antonio 1990).
Although some women with BV remain asymptomatic, as many as
50% to 60% of BV patients report symptoms such as occurrence
of malodorous “fishy” vaginal discharge. The Amsel criteria for
the diagnosis of BV were first described in 1983 (Amsel 1983) and
a positive diagnosis is based on the presence of at least three of
the following: presence of a thin, greyish homogenous discharge;
vaginal pH greater than 4.5; presence of clue cells on Gram stain;
positive whiff test.
The underlying cause of BV is still not fully understood, but factors such as vaginal douching, black ethnicity, low socioeconomic
status, use of an intra-uterine contraceptive device, and new/multiple sex partners appear to increase the risk of this perturbation
in the vaginal ecosystem (Calzolari 2000; Fonck 2001; Nilsson
However, there are other diagnostic scoring systems based on the
number or ratio of Lactobacilli versus other bacterial morphotypes
seen on Gram stained vaginal smears or wet preparations (Forsum
2005). The Nugent system (Nugent 1991) scores the number of
Lactobacilli, Gardnerella, and curved bacteria (Mobiluncus) mor-
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
2
photypes per high-power field. Various studies have used the Nugent scoring system as an alternative method to the Amsel criteria
and there is a close correlation between both methods due to the
high sensitivity and specificity of microbial quantification of Gram
stained vaginal smears (Honest 2004). New diagnostic methods
using molecular techniques are now being developed and applied
in various studies (Forsum 2005; Verhelst 2004; Verstraelen 2004).
It is now known that BV is associated with potentially severe gynaecological and obstetric complications and sequelae. Current
data suggest a causal association between BV, pelvic inflammatory disease, and tubal factor infertility (Wilson 2002). Pregnant
women with BV have a higher risk of adverse outcomes such as
late miscarriage, chorioamnionitis, premature rupture of membranes, preterm birth and postpartum endometritis (Marrazzo
2004). In women undergoing in vitro fertilization, BV may result
in lower implantation rates and increased rates of early pregnancy
loss (Eckert 2003; Verstraelen 2005).
There is growing evidence of an association between BV and sexually transmitted infections, making it of grave concern, particularly in light of the HIV/AIDS epidemic. There are data indicative
of a relationship between BV and increased risk of HIV transmission and acquisition (Cu-Uvin 2001; Sewankambo 1997). The
presence of BV was associated with up to a six-fold increment in
quantity of HIV shedding (Cu-Uvin 2001). Furthermore, there is
a correlation between the absence of vaginal lactobacilli and a positive result for the carriage of Neisseria gonorrhoeae and Chlamydia
trachomatis (Wiesenfeld 2003). BV has been identified as a risk
factor for herpes virus type 2 infection and increased viral shedding in infected women (Cherpes 2005). Recent reports also indicate that BV may increase the risk of acquisition or reactivation
of Human Papillomavirus infection, particularly high-risk HPVtypes (da Silva 2004). The displacement of the H2 O2 -producing
Lactobacilli, elevated vaginal pH with the presence of bacteria that
produce sialidase and mucin-degrading enzymes as seen in BV,
probably combine to create a milieu conducive to the survival of
these pathogens (Olmsted 2003).
In light of all these adverse BV-associated conditions, adequate
treatment of this condition is crucial. Current guidelines for treatment of BV stipulate the use of metronidazole or clindamycin administered orally or intravaginally (Anonymous 2002; Marrazzo
2004). Other treatment regimens that have been suggested include acidification of the vagina and the use of povidone-iodine
vaginal suppositories. Up to 10% to15% of patients fail to respond to initial antimicrobial therapy, and recurrence rates over
time among responders remain significant, reaching up to 80%
and necessitating repeated administration of antibiotics. Such repeated antibiotic exposure increases the risk of emergence of resistant strains, alteration of microbiota, and possible persistence
of BV-associated pathogens. The high recurrence rates associated
with antibiotic therapy have stimulated the search for alternative
treatment modalities, which could be used with or without the
current regimen. The dominance of lactobacilli in healthy vaginal
microbiota and its obliteration in BV has given rise to the concept
of oral or vaginal instillation of probiotic Lactobacillus strains to
restore the balance. Probiotics are defined as live micro-organisms
which, when administered in adequate amounts, confer a beneficial health effect on the host (Senok 2005). Available evidence now
indicates that certain strains of lactobacilli when administered to
patients can colonize the vagina and reduce the risk of BV (Reid
2001). Studies have been carried out to assess the efficacy of single
strain or cocktail of probiotics administered orally or intravaginal
in the treatment of BV (Falagas 2007). In addition, the effect of
probiotics in conjunction with antibiotic regimen has also been
evaluated. Lactobacilli probiotics can be used over a long time
without adverse effects, making them an attractive option to antibiotics, particularly in addressing the problem of high recurrence
rates.
While the scientific community is still trying to determine the efficacy of probiotics in the treatment of BV, women with chronic
vaginal symptoms are increasingly self-utilizing alternative remedies, in particular lactobacilli replacement therapy, as an alternative to antimicrobials (Nyirjesy 1997; Trutnovsky 2001). It is
therefore of essence to conduct a rigorous systematic review of
available clinical trials, with a view to ascertaining the efficacy (or
lack thereof ) of probiotics in the treatment of BV, and possibly to
identify strategic areas for future research.
OBJECTIVES
To ascertain the efficacy of probiotics in the treatment of bacterial
vaginosis
METHODS
Criteria for considering studies for this review
Types of studies
Randomized controlled trials using probiotics (alone or as adjuncts
to conventional antibiotics). Studies using probiotic preparations
containing other substances e.g. estriol will be included although
a sub-analysis will be carried out for such preparations.
Types of participants
Women of any age diagnosed with bacterial vaginosis, regardless
of diagnostic method used. No study will be excluded because of
co-infection with other sexually transmitted infections.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
3
Types of interventions
Any probiotic (single or mixture “cocktail” of strains, any preparation type, any dosage regimen, any route of administration):
·used alone and compared with placebo or no treatment;
·used in conjunction with conventional antibiotics (before, during,
or after antibiotic treatment) and compared with placebo or no
treatment;
·used alone and compared with conventional antibiotic regimen.
Types of outcome measures
Primary outcome: restoration of normal lactobacilli microbiota,
with eradication of BV microbiota (’microbiological cure’), using
the Nugent criteria assessment completed between days 21 and 30
days post-treatment.
Secondary outcomes:
·resolution of symptoms as reported by the patient;
·clinical cure as reported by the physician or investigator;
·persistence or recurrence of BV at follow up (from 30 days posttreatment);
·occurrence of adverse reactions or side effects;
·effect on BV-associated conditions.
Search methods for identification of studies
We performed an exhaustive search of electronic databases irrespective of publication status or language. We searched the
Cochrane Central Register of Controlled Trials (CENTRAL) (The
Cochrane Library), HIV/AIDS & STD Cochrane Research Groups
specialty registers, Cochrane Complementary Medicine Field’s
Register of Controlled Trials (all quarterly searches until July
2008), MEDLINE (1966-2008), EMBASE (1980-2007), ISI science citation index (1955-2007), CINAHL (Cumulative Index
to Nursing & Allied Health Literature (1982-2007) Appendix 1.
We also searched the following databases: ProQuest, FIRSTConsult, Conference Papers Index, Ovid’s Dissertation Abstracts, ScienceDirect, Blackwell Synergy.
We hand searched the following:
i. Specialty journals including American Journal of Obstetrics and
Gynecology, Clinical Microbiology and Infection, Eastern Mediterranean Health Journal, BMJ, BMC Infectious Diseases, Journal of
Antimicrobial Chemotherapy, Journal of Clinical Microbiology, Antimicrobial Agents and Chemotherapy, Journal of Medical Microbiology, Journal of Bacteriology, Journal of Applied Microbiology, Infection and Immunity, Applied and Environmental Microbiology, Clinical Microbiology Reviews, Sexually Transmitted Diseases.
ii. Abstracts of major conference proceedings (including the International Society of STD Research; World AIDS Conference) to
identify trials that may have not been published in full
iii. Cross-checking of references cited in recent reviews and published randomized controlled trials for additional studies not identified by electronic searches
iv. Abstracts and publications listed on the website of the International Scientific Association of Probiotics and Prebiotics (http://
www.isapp.net/default.asp)
We contacted experts in the field (particularly probiotics) identified from relevant publications by email and formal letters. We
also contacted leading nutraceutical companies marketing probiotic products. We sent emails to scientific associations dedicated to probiotic research including the International Scientific
Association of Probiotics and Prebiotics, Polish Society of Probiotics and International Probiotics Association. We requested
that our communication be forwarded to their members. In all
of these communications, we inquired about knowledge of published and unpublished studies and ongoing trials relevant to the
review.We also searched electronic databases on ongoing clinical
trials (www.trialscentral.org ; www.controlled-trials.com).
The following key terms were used in our search: bacterial vaginosis
intersected with probiotics, lactobacilli/lactobacillus, antibiotics,
drug therapy, treatment, complementary, alternative
The search strategy was as follows:
“Bacterial *4 Vaginosis” or
“Vaginitis *4 Bacterial” or
“Vaginitis *4 Nonspecific” or
“Bacterial *4 Vaginitides” or
“Bacterial *4 Vaginitis ” or
“Bacterial *4 Vaginoses” or
“Vaginitides *4 Bacterial” or
“Vaginoses *4 Bacterial” or
“Vaginosis, Bacterial” or
“colpitis *4 Bacterial” or
“Vaginitis *4 Nonspecific” or
(’vaginitis’/exp AND (Bacterial or bacteria or nonspecific or “nonspecific” or “non specific”))
AND
“clinical trial”/exp or “randomized controlled trial”/exp or randomization/ or “single blind procedure”/exp or “double blind procedure”/ or “crossover procedure”/ or placebo/ or “randomi*ed
controlled trial*” or rct or “random allocation” or “randomly allocated” or “allocated *2 random” or “random *2 allocated” OR “single blind*” or “double blind*” or placebo* or “prospective study”/
exp or “meta analysis”/exp or ((meta *2 analy*) or metaanalys*) or
(systematic *3 (review* OR overview*))
Data collection and analysis
Trials selection:ACS and MF, a trained research assistant, performed the searches. We also received help from the HIV/AIDS
and STD Cochrane Review Groups Trials Search Co-ordinator
and the Medical Librarian, College of Medicine, University of
Sharjah. The titles and abstracts from the search were screened to
identify those of potential relevance for the review. In cases where
we could not make a decision based on the abstract, we obtained
and reviewed the full text. Following this process, we identified
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
4
17 trials which were relevant to the review protocol. We designed
an eligibility assessment form. ACS, HV and MF obtained the
full reports of all relevant trials and independently appraised them
for eligibility using this form. We resolved disagreements about
trial inclusion by discussion. Excluded trials and reasons for their
exclusion are listed in “Characteristics of excluded studies” table.
Assessment of methodological quality: ACS, MF and HV independently assessed the methodological quality of all eligible trials.
This included the assessment of random allocation, whether or
not there was concealment, whether participants, clinicians and
outcome assessors were blinded or not blinded and whether intention-to-treat (ITT) analysis was carried out. We also considered whether or not all outcome measures were addressed. Where
information was not clearly reported, we contacted trial authors
for necessary clarification. A full description of the criteria used in
assessing the risk of bias is found in the Risk of Bias Table.
Data Extraction: ACS and MF developed and used a data extraction form, which they used independently to extract and record
information on the results of included studies. . Where there were
differences HV gave an independent assessment and final resolution was achieved by discussion.
Data analysis: This was carried out using the RevMan Analyses
statistical package in Review Manager (version 4.2) and this was
completed using Review Manager version 5. For dichotomous outcomes we derived odds ratio (OR) and 95% confidence intervals
(CI) for each study. For continuous data we intended to calculate the mean differences for each study but this was not done
as there were no such data. We had planned to do the following
analysis as part of a full meta-analysis: test for heterogeneity using
a standard Chi2 test and where appropriate combine the results
for each outcome for included studies. Where there was homogeneity, the pooled RR and 95% CI was to be calculated for dichotomous outcomes using a fixed-effect model. For continuous
data, we had planned to estimate the pooled weighted mean differences (WMD) and 95% CI. In the presence of heterogeneity,
we planned to explore the reason for this and if none was found, to
use a random-effects model. Funnel plots were to be used for assessment of publication or other reporting bias. The nature of the
included studies precluded these analysis. We did not perform a
meta-analysis due to significant differences in the methodologies,
regimens and probiotic preparations used in the included studies.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies.
We identified about 350 studies from the comprehensive search
and reviewed the abstracts for all. In some cases we reviewed the
full text before a decision was made. We identified 17 studies
for possible inclusion and we reviewed the full text version of
all of them. These studies were duplicated across at least two of
the databases we searched. ACS and MF independently assessed
whether these identified studies fulfilled the inclusion criteria in
for the review using a pre-designed form. HV acted as an independent arbiter where there was a conflict of opinion. Four studies
(Anukam 2006a; Anukam 2006b; Eriksson 2005; Parent 1996)
satisfied the inclusion criteria and were included in the review.
The 12 excluded trials and reasons for their exclusion are listed in
the “Characteristics of excluded studies” table. Details of the four
included studies are described below and in the Characteristics of
Included Studies Table (Characteristics of included studies)
Anukam 2006a was a randomized controlled double blinded trial
to assess the efficacy of augmentation of oral metronidazole with
oral probiotics in the treatment of BV. In this study, over 500 subjects were screened and 125 pre-menopausal women (aged 18 to
44 years) fitting the entry criteria of having clinical features of BV,
positive Nugent and BVBlue (salidase test) score were recruited.
The exclusion criteria were use of antibiotics in the preceding 14
days, pregnancy, lactation, evidence of gross inflammatory genital
process, presence of sexually transmitted infection, use of investigational drugs within 30 days, hypersensitivity to metronidazole,
warfarin, lithium or disulfaram and menstruation at time of diagnosis. A power calculation for sample size was carried out. The
patients were randomized in a double blind manner and given oral
metronidazole 500 mg twice daily for seven days, plus either oral
L. rhamnosus GR-1 and L. reuteri RC-14 (1x109 CFU per capsule) twice daily for 30 days starting on day one of metronidazole
treatment (n = 65) or identical looking placebo capsules (n = 60).
At the end of the treatment period three vaginal swabs were collected and evaluated for BV status. Of the 125 women enrolled,
only 96 (84.8%) returned for follow up on day 30 (16 patients in
the placebo arm and 3 patients in the treatment arm were lost to
follow up).The primary outcome was cure of BV as determined
by normal Nugent score, absence of clue cells, negative salidase
test and no symptoms or signs of BV at day 30. Intention-to-treat
analysis was not carried out. There was no occurrence of adverse
effects.
Anukam 2006b was a randomized double blind clinical trial in
which intravaginal probiotic capsules was compared with metronidazole gel for the treatment of BV. The sample size of twenty
women per treatment group was based on the expectation of finding a 50% difference between the two treatment groups. The inclusion criteria were presence of vaginal discharge, unpleasant fishy
odour, with or without localized vaginal irritation / discomfort.
Exclusion criteria include pregnancy, age < 18years or > 50years,
menstruating at time of diagnosis or due to menstruate over the
following five days, HIV positive, urogenital infections including
yeast vaginitis and sexually transmitted infections. The 40 women
enrolled in the study had the diagnosis of BV confirmed based on
Nugent score (> 7) and positive salidase test. Twenty women were
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
5
randomized to receive probiotic intravaginal capsule containing
L. rhamnosus GR-1 and L. reuteri RC-14 (1x109 CFU per capsule) nightly for five nights. The other 20 women applied 0.75%
metronidazole gel twice daily intravaginally for 5 days. The women
were assessed on days 0, 6, 15, and 30 at which time two vaginal
swabs were collected and assessed for Nugent scoring, KOH and
salidase test. There was 100% follow-up on days 6 and 15 and all
women demonstrated compliance by returning empty treatment
vials. Three women in the probiotic arm and 2 in the metronidaozole group failed to return for follow up on day 30. Adverse effects
were not described.
Eriksson 2005 assessed the efficacy of the use of vaginal lactobacilli
after open treatment with vaginal clindamycin ovules. The inclusion criteria were based on the diagnosis of BV by Amsel’s criteria,
age > 18 years and menstruating. The women had to be willing
to use a tampon during menstruation. Subjects were excluded if
they were pregnant, had used antibiotics in the week preceding
the diagnosis of BV or had presence of vaginal yeast or Chlamydia trachomatis infection. A power calculation was carried out to
determine sample size. During the open part of the trial all 225
enrolled subjects were treated with clindamycin ovules 100 mg
inserted vaginally once daily for three days irrespective of whether
they were randomized into the probiotic group or the placebo
group. In the following menstruation, patients received either a
lactobacilli impregnated tampon (n = 91) or placebo tampons (n
= 96). They had to use at least five tampons to be included in
the efficacy analysis and all unused tampons were returned. At
the second menstruation, the women used their normal menstrual
protection. The lactobacilli impregnated tampons contained L.
casei var rhamnosus, L. gasseri and L. fermentum (1x108 per tampon). After the first menstruation (during which study tampons
were used) and after the second menstruation, the patients sent
self-taken vaginal smears to the study co-ordinator. At the followup after the second menstruation, the patients were re-evaluated
using Amsel’s criteria and another vaginal smear was taken. The
microbiological cure rates were determined after the second menstruation using Nugent score and Amsel’s criteria. Only 187 participants were included in the ’per protocol’ analysis as 30 women
were lost to follow up and another 38 were excluded due to violation of protocol. Intention-to treat-analysis was carried out and
adverse effects were reported.
Parent 1996 evaluated the efficacy of vaginal tablets containing
L. acidophilus and estriol. This was a randomized, placebo-controlled clinical trial with a parallel-group design. Thirty-two nonmenopausal women (24 pregnant and 8 pregnant) diagnosed with
BV based on Amsel’s criteria were enrolled. Patients were excluded
from the study if they had used topical or systemic hormonal treatment (except oral contraceptives) or had undergone any local vaginal treatment in the two weeks prior to the trial, had metrorrhagia
or acute vaginitis. The participants were randomly allocated into
lactobacilli and placebo groups using a pre-determined randomization scheme. The lactobacilli group received Gynoflor vaginal
tablets containing L. acidophilus (10 million viable bacteria per
tablet) and 0.03 mg estriol. The placebo tablets did not contain
L. acidophilus or estriol. Both groups inserted one vaginal tablet
nightly for six days. Two tablets were inserted nightly by pregnant
patients with abundant leukorrhea and those women who started
menstruating during the treatment period (in which case therapy
was interrupted and resumed after menstruation). The patients
were examined and smears obtained on days 15 and 28 after onset of therapy. Microbiological cure was determined using Nugent
criteria. Intention-to-treat analysis was not carried out. Adverse
effects were reported.
Risk of bias in included studies
Two review authors independently assessed the methodological
quality of the included studies.
Randomization in Anukam 2006a was carried out using a computer generated scheme and allocation schedule was conducted
“off-site” by the pharmacy. The two groups were similar at baseline regarding the presence of BV. All assessors and patients were
blinded. Measures of the key outcome was obtained from 106
women who returned for a follow-up visit. There was possible
attrition bias due to patients lost to follow up and there was no
statement on intention-to-treat analysis. The statistical analysis
was well described and used appropriately. The results of betweengroup statistical comparisons as well as the measure of the size of
the treatment effect were reported for the key outcome measure.
Randomization in Anukam 2006b was carried out using a computer generated scheme and patients in both study groups were
similar at baseline. Allocation concealment was carried out “offsite” by the pharmacy. The patients were not blinded, however
the assessors (clinical and laboratory) were blinded. Only 87.5%
of patients returned for follow up on day 30, which may have
introduced some attrition bias. All subjects for whom outcome
measures were available received the treatment or placebo as allocated. There was no statement on intention-to-treat analysis. The
statistical analysis was well described and used appropriately.
Randomization in Eriksson 2005 was carried out but the method
used was not described. Allocation concealment was not carried
out. Blinding of patients and clinical assessors was inadequate due
to slight colour differences between the probiotic and the placebo
tampons. Not all the subjects for whom outcome measures were
available received the treatment or placebo as allocated. Data for
those excluded for protocol violation and drop outs were adequately addressed. Intention-to-treat analysis was performed. The
results of between-group statistical comparisons as well as the measure of the size of the treatment effect were reported.
Randomization in Parent 1996 was carried using a “pre-determined randomization scheme”. Allocation concealment was not
stated and there was no mention or description of blinding. We
contacted the authors for clarification but received no reply. There
was possible attrition bias due to patients lost to follow up. The
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
6
statistical analysis (per-protocol) was well described and used appropriately. The results of between-group statistical comparisons
as well as the measure of the size of the treatment effect were reported for the key outcome measure.
A detailed description can be found in the Risk of Bias Table.
See Figure 1 and Figure 2 for methodological quality graph and
summary respectively.
Figure 1. Methodological quality graph: review authors’ judgements about each methodological quality
item presented as percentages across all included studies.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
7
Figure 2. Methodological quality summary: review authors’ judgements about each methodological quality
item for each included study.
Effects of interventions
Primary outcome:
All four included studies assessed microbiological cure as a key
outcome measure. Using the reported data for the primary outcome, we calculated the Odds Ratio (OR) and 95% Confidence
Interval (CI) for each study.
Anukam 2006a reported a statistically significant difference at the
day 30 follow up in the per-protocol analysis with 43 of the 49
(88%) women in the antibiotic/probiotic treatment arm having
normal Nugent scores compared to 23 of the 57 women (40%)
in the antibiotic/placebo arm: OR 0.09 (95% CI 0.03 to 0.26).
This finding is suggestive of a beneficial effect of the antibiotic/
probiotic regimen in the treatment of BV.
In Anukam 2006b, according to the per-protocol analysis, 11 of
the 17 (64.7%) women in the probiotic arm had normal Nugent
scores on day 30 compared to 6 out of 18 (33.3%) in the metronidazole arm: OR 0.27 (95% CI 0.07 to 1.10).
In Eriksson 2005, the ITT analysis of the primary outcome published in the article was based on Amsel’s criteria. The assessment
of the primary outcome based on Nugent criteria was presented
only for those included in the per-protocol analysis. The study
authors kindly provided us with the ITT analysis data (for 217
women) based on Nugent criteria. Of the 108 women in the treatment arm, 75 (69.4%) had normal Nugent scores at follow up
after the second menstruation compared to 80 out of 109 (73.3%)
in the placebo arm: OR 1.21 (95% CI 0.67 to 2.19).
InParent 1996, microbiological cure on day 28 was documented in
7 of the 8 (87.5%) women in the probiotic/estriol arm compared
to 1 out of 7 (14.2%) in the placebo/estriol arm: OR 0.02 (95% CI
0.00 to 0.47). This statistical analysis suggests that the probiotic/
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
8
estriol regimen is beneficial for the treatment of BV.
Secondary outcomes:
Not all the studies assessed all the five secondary outcomes outlined in our review protocol. None of the studies reported on: (i)
persistence or recurrence of BV at follow up from 30 days post
treatment or (ii) effect on BV associated outcomes.
The other three secondary outcomes, (i) resolution of symptoms
as reported by the patient, (ii) clinical cure as reported by the
physician or investigator and (iii) occurrence of adverse reactions
or side effects were reported. We used the data from each study
that reported any of these three secondary outcomes to calculate
the Odds ratio (OR) and 95% Confidence Interval (CI).
1. Resolution of symptoms as reported by the patient
This outcome was reported in Anukam 2006b and Eriksson 2005.
In Anukam 2006b , 15 of the 17 (88.2%) women in the probiotic
arm reported resolution of symptoms compared to 12 of the 18
(66.6%) women in the metronidazole arm: OR 0.27 (95% CI
0.05, 1.57). The authors of Eriksson 2005 provided us with data
on the resolution of symptoms as assessed by the patient at follow
up. Data for 55 women was stated as missing and the ITT analysis
for 216 women showed that 74 out of 108 (68.5%) in the treatment arm reported complete resolution of symptoms compared
to 67 out of 108 (62%) in the placebo arm: OR 0.75 (95% CI
0.43 to 1.32).
2. Clinical cure as reported by physician or investigator
This was reported in two studies. In Eriksson 2005, clinical cure
based on Amsel’s criteria was reported. The authors confirmed
that these data were used as the assessment of physician-reported
clinical cure. Clinical cure was reported for 64 out of 111 (57.6%)
women in the treatment arm and 68 out of 113 (60.1%) in the
placebo arm: OR 1.11 (95% CI 0.65 to 1.89). Parent 1996 reported clinical cure on day 28 based on Amsel criteria. In the probiotic/estriol group, 7 of the 8 (87.5%) women were reported as
clinically cured compared to 2 out of 9 (22.2%) women in the
placebo/estriol group: OR 0.04 (95% CI 0.00 to 0.56). The finding from Parent 1996 is suggestive of a beneficial effect with the
probiotic/estriol regimen.
3. Occurrence of adverse effects or side effects
Anukam 2006a reported that there were no adverse effects. In
Anukam 2006b, 2 out of 17 (11.7%) women in the probiotic arm
reported adverse effects versus 11 out of 18 (61.1%) women in
the metronidazole group: OR 0.08 (95% CI 0.01 to 0.49)
There was no significant difference in the severity, nature and frequency of adverse events between the treatment and control groups
in Eriksson 2005. The most common adverse event was clinical
Candida infection which occurred in 14.2% of those receiving
the probiotic treatment and in 13.5% of the control. The author
of Parent 1996 reported good tolerability without giving data on
adverse effects.
DISCUSSION
BV occurs as a result of imbalance in the normal vaginal microbiota
and is characterized by a decrease or depletion of the Lactobacilli
spp and a concomitant overgrowth of other vaginal anaerobic bacteria. It remains a common cause of genital discomfort in women
and it is associated with potentially severe gynaecological and obstetric sequelae. Unfortunately, currently available treatment of
BV with the use of metronidazole or clindamycin administered
orally or intravaginally has been shown to be associated with poor
initial cure rates in 10% to 15% of patients and recurrence rates
of up to 80% in those who show initial response. The dominance
of lactobacilli in healthy vaginal microbiota and its obliteration
in BV has given rise to the concept of oral or vaginal instillation
of probiotic Lactobacillus strains to restore the balance. Indeed,
available evidence now indicates that certain strains of lactobacilli
administered intravaginally are capable of colonizing the vagina.
This systematic review assesses the evidence for the use of probiotic regimen in the treatment of BV.
We identified four randomized controlled trials which investigated the use of probiotics in women with BV and met the inclusion criteria outlined in the protocol for this systematic review.
There were significant variations in the design of the trials identified. Anukam 2006a assessed the efficacy of augmentation of oral
metronidazole with oral probiotics in the treatment of BV. In contrast, Anukam 2006b compared intravaginal probiotic capsules
with metronidazole gel for the treatment of BV. Eriksson 2005
used clindamycin ovules for the open treatment followed by use of
probiotic impregnated tampons versus placebo tampons. Parent
1996 used combined probiotic/estriol vaginal tablets with both
pregnant and non-pregnant women included in the study population. As per our protocol, the results of included studies were
to be combined for each outcome where appropriate and a metaanalysis conducted. However, due to these significant differences
in the types of probiotics used and the trial methodologies, we did
not perform a meta-analysis. Variation in the methodologies and
probiotic preparations is thus identified as a major hindrance in
comparing the outcome of different trials, making it difficult to
conclude on the beneficial effect of probiotics in the treatment of
BV and we recommend that this should be addressed in the design
of future trials.
Methodological quality was inadequate in two studies (Eriksson
2005; Parent 1996). The randomization method was not adequately described and allocation concealment was not carried out
in either study. Although described as a randomized placebo-controlled clinical trial with parallel group design, there was no mention or description of blinding in Parent 1996. Although three
studies (Anukam 2006a; Anukam 2006b; Eriksson 2005) did
power calculations to determine sample size, in all the trials patients were lost to follow up leading to considerable possible attrition bias. It is therefore possible that the sample size of these
included studies were ultimately not large enough to detect statis-
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
9
tical significant differences between the treatment groups.
Nevertheless, for dichotomous outcomes, the odds ratio (OR) and
95% confidence intervals (CI) were derived for each study. This
analysis of OR and CI for individual studies for the outcomes of
microbiological cure was suggestive of a beneficial effect only for
the augmentation of oral metronidazole with oral probiotics regimen (Anukam 2006a) and the probiotic/estriol regimen (Parent
1996). In addition, physician reported resolution also favoured
the probiotic/estriol regimen (Parent 1996). However, for the resolution of symptoms as assessed by the patient, analysis of the data
was not suggestive of a beneficial effect for the use of probiotics and
the finding was similar irrespective of the routes of administration
of the probiotic (oral, vaginal capsules or impregnated tampons),
antibiotic use before/during the probiotic treatment or the type of
probiotic preparation (single or cocktail, combined with estriol).
In general, good tolerability and safety profiles were reported in all
included studies. Analysis of OR which could only be carried for
one study (Anukam 2006b) was suggestive of good tolerability:
OR 0.08 (95% CI 0.01 to 0.49).
However, irrespective of the findings of the derived OR and 95%
CI, it must be emphasized that a meta-analysis is still required
to conclusively determine the beneficial effect and safety profile
of probiotics in the treatment of BV. Thus we recommend large
scale, well-designed clinical trials which are standardized to address issues of differences in methodologies and probiotic preparations ensuring that strains which have been shown to colonize
the vagina are used. The oral metronidazole/oral probiotic regimen appears promising, and we suggest further trials to investigate
this treatment approach. The high incidence of recurrence is a
major limitation associated with the use of antibiotics and should
be evaluated as an important outcome if probiotics are to be considered successful for the treatment of BV. However none of the
studies reported data on BV recurrence following probiotic use. In
addition, as BV is associated with a number of gynaecological and
obstetric complications, the effect of probiotics on such sequelae
is an important secondary outcome. Unfortunately, this was not
evaluated in any of the studies.
AUTHORS’ CONCLUSIONS
Implications for practice
The results do not provide sufficient evidence for or against recommending probiotics for the treatment of BV. In addition, there is
no conclusive evidence to recommend the use of probiotics either
before, during or after antibiotic treatment as a means of ensuring
successful treatment or reduce recurrence.
Implications for research
There is a need for well-designed randomized controlled trials with
standardized methodologies and larger patient size.
ACKNOWLEDGEMENTS
We thank Dr Malaz Fadlallah who worked as a research assistant
on this project for her assistance in the identification of studies
and data collection. We acknowledge the assistance of Ms Nadia
Masood, Librarian, College of Medicine, University of Sharjah in
the development of the search strategy and carrying out the search.
REFERENCES
References to studies included in this review
Anukam 2006a {published data only}
Anukam K, Osazuwa E, Ahonkhai I, Ngwo M, Osemene
G, Bruce AW, et al.Augmentation of antimicrobial
metronidazole therapy of bacterial vaginosis with oral
probiotic Lactobacillus rhamnosus GR1 and Lactobacillus
reuteri RC14: Randomized double blind placebo controlled
trial. Microbes and Infection 2006;8:1450–4.
Anukam 2006b {published data only}
Anukam K, Osazuwa E, Osemene GI, Ehigiagbe F,
Bruce AW, Reid G. Clinical study comparing probiotic
Lactobacillus GR-1 and RC-14 with metronidazole vaginal
gel to treat symptomatic bacterial vaginosis. Microbes and
Infection 2006;8:2772–6.
Eriksson 2005 {published data only}
Eriksson K, Carlsson B, Forsum U, Larsson PG. A Doubleblind treatment study of bacterial vaginosis with normal
vaginal Lactobacilli after an open treatment with vaginal
clindamycin ovules. Acta Dermato-venereologica 2005;85:
42–6.
Parent 1996 {published data only}
Parent D, Bossens M, Bayot D, Kirkpatrick C, Graf F,
Wilkinson FE, Kaiser RR, et al.Therapy of Bacterial
Vaginosis using exogeneously-applied Lactobacilli aciophili
and a low dose of estriol : A placebo-controlled multicentric
clinical trial. Arzneimitell-Forschung 1996;46:68–73.
References to studies excluded from this review
Della 2006 {published data only}
Delia A, Morgante G, Rago G, Musacchio MC, Petraglia
F, De Leo V. Effectiveness of oral administration of
Lactobacillus paracasei subsp. paracasei F19 in association
with vaginal suppositories of Lactobacillus acidofilus in the
treatment of vaginosis and in the prevention of recurrent
vaginitis. Minerva Ginecologica 2006;58:227–31.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
10
Hallen 1992 {published data only}
Hallén A, Jarstrand C, Påhlson C. Treatment of bacterial
vaginosis with lactobacilli. Sexually Transmitted Diseases
1992;19:146.
Larsson 2008 {published data only}
Larsson P, Stray-Pedersen B, Ryttig KR, Larsen, S. Human
lactobacilli as supplementation of clindamycin to patients
with bacterial vaginosis reduce the recurrence rate; a 6month
double-blind, randomized, placebo-controlled study. BMC
Women’s Health 2008;8(3):doi:10.1186/1472–6874-8-3.
Marrazzo 2006 {published data only}
Marrazzo JM, Cook RL, Wiesenfeld HC, Murray PJ, Busse
B, Krohn M, et al.Women’s satisfaction with an intravaginal
Lactobacillus capsule for the treatment of bacterial vaginosis.
Journal of Women’s Health 2006;15:1053–60.
Neri 1993 {published data only}
Neri A, Sabah G, Samra Z. Bacterial vaginosis in pregnancy
treated with yoghurt. Acta Obstetrica et Gynecologica
Scandinavica 1993;72:17–19.
Ozkinay 2005 {published data only}
Ozkinay E, Terek MC, Yayci M, Kaiser R, Grob P, Tuncay
G. The effectiveness of live lactobacilli in combination with
low dose oestriol (Gynoflor) to restore the vaginal flora after
treatment of vaginal infections. BJOG: An International
Journal of Obstetrics and Gynaecology 2005;112:234–40.
Reid 2001a {published data only}
Reid G, Beuerman D, Heinemann C, Bruce AW. Probiotic
Lactobacillus dose required to restore and maintain a normal
vaginal flora. FEMS Immunology and Medical Microbiology
2001;32:37–41.
Reid 2001b {published data only}
Reid G, Bruce AW, Fraser N, Heinemann C, Owen
J, Henning B. Oral probiotics can resolve urogenital
infections. FEMS Immunology and Medical Microbiology
2001;30:49–52.
Reid 2003 {published data only}
Reid G, Charbonneau D, Erb J, Kochanowski B, Beuerman
D, Poehner R, et al.Oral use of Lactobacillus rhamnosus
GR-1 and L. fermentum RC-14 significantly alters vaginal
flora: randomized, placebo-controlled trial in 64 healthy
women. FEMS Immunology and Medical Microbiology 2003;
35:131–4.
Reid 2004 {published data only}
Reid G, Burton J, Hammond JA, Bruce AW. Nucleic
acid-based diagnosis of bacterial vaginosis and improved
management using probiotic lactobacilli. Journal of
Medicinal Food 2004;7:223–8.
Ronnqvist 2006 {published data only}
Rönnqvist PD, Forsgren-Brusk UB, Grahn-Håkansson
EE. Lactobacilli in the female genital tract in relation to
other genital microbes and vaginal pH. Acta Obstetrica et
Gynecologica Scandinavica 2006;85:726–35.
Shalev 1996 {published data only}
Shalev E, Battino S, Weiner E, Colodner R, Keness Y.
Ingestion of yogurt containing Lactobacillus acidophilus
compared with pasteurized yogurt as prophylaxis for
recurrent candidal vaginitis and bacterial vaginosis. Arch
Fam Med 1996;5:593–6.
Wewalka 2002 {published data only}
Wewalka G, Stary A, Bosse B, Duerr HE, Reimer K. Efficacy
of povidone-iodine vaginal suppositories in the treatment of
bacterial vaginosis. Dermatology 2002;204:79–85.
Additional references
Amsel 1983
Amsel R, Totten PA, Spiegel CA, et al.Nonspecific vaginitis.
Diagnostic criteria and microbial and epidemiologic
associations. American Journal of Medicine 1983;74:14–22.
Anonymous 2002
Anonymous. Sexually transmitted disease treatment
guidelines: Center for Disease Control and Prevention.
Morbidity and Mortality Weekly Report 2002;51(RR-6):
1–80.
Antonio 1990
Antonio MA, Hawes SE, Hillier SL. The identification
of vaginal Lactobacillus species and the demographic and
microbiologic characteristics of women colonized by these
species. Journal of Infectious Diseases 1999;180:1950–6.
Behets 2001
Behets F, Andriamiadana J, Rasamilalao D, et al.Sexually
transmitted infections and associated socio-demographic
and behavioural factors in women seeking primary care
suggest Madagascar’s vulnerability to rapid HIV spread.
Tropical Medicine and International Health 2001;6:202–11.
Calzolari 2000
Calzolari E, Masciangelo R, Milite V, et al.Bacterial
vaginosis and contraceptive methods. International Journal
of Gynaecology and Obstetrics 2000;70:341–6.
Cherpes 2005
Cherpes TL, Melan MA, Kant JA, et al.Genital tract
shedding of herpes simplex virus type 2 in women: effects
of hormonal contraception, bacterial vaginosis, and vaginal
group B Streptococcus colonization. Clinical Infectious
Diseases 2005;40:1422–8.
Cu-Uvin 2001
Cu-Uvin S, Hogan JW, Caliendo AM, et al.Association
between bacterial vaginosis and expression of human
immunodeficiency virus type 1 RNA in the female genital
tract. Clinical Infectious Diseases 2001;33:894–6.
da Silva 2004
da Silva CS, Adad SJ, Hazarabedian de Souza MA, et
al.Increased frequency of bacterial vaginosis and Chlamydia
trachomatis in pregnant women with human papillomavirus
infection. Gynecologic and Obstetric Investigation 2004;58:
189–93.
Eckert 2003
Eckert LO, Moore DE, Patton DL, et al.Relationship
of vaginal bacteria and inflammation with conception
and early pregnancy loss following in-vitro fertilization.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
11
Infectious Diseases in Obstetrics and Gynecology 2003;11:
11–17.
Eschenbach 1988
Eschenbach DA, Hillier S, Critchlow C, et al.Diagnosis
and clinical manifestations of bacterial vaginosis. American
Journal of Obstetrics and Gynecology 1988;158:819–28.
Falagas 2007
Falagas ME, Betsi GI, Athanasiou S. Probiotics for the
treatment of women with bacterial vaginosis. Clinical
Microbiology and Infection 2007;13:657–64.
Fonck 2001
Fonck K, Kaul R, Keli F, et al.Sexually transmitted infections
and vaginal douching in a population of female sex workers
in Nairobi, Kenya. Sexually Transmitted Infections 2001;77:
271–5.
Forsum 2005
Forsum U, Hallen A, Larsson PG. Bacterial vaginosis--a
laboratory and clinical diagnostics enigma. Acta Pathologica,
Microbiologica, et Immunologica Scandinavica 2005;113:
153–61.
Fredricks 2005
Fredricks DN, Marrazzo JM. Molecular methodology in
determining vaginal flora in health and disease: its time has
come. Curr Infect Dis Rep 2005;7:463–70.
Honest 2004
Honest H, Bachmann LM, Knox EM, et al.The accuracy
of various tests for bacterial vaginosis in predicting preterm
birth: a systematic review. BJOG: An International Journal
of Obstetrics and Gynaecology 2004;111:409–22.
Larsson 2005
Larsson PG, Bergstrom M, Forsum U, et al.Bacterial
vaginosis. Transmission, role in genital tract infection
and pregnancy outcome: an enigma. Acta Pathologica,
Microbiologica et Immunologica Scandinavica 2005;113:
233–45.
Marrazzo 2004
Marrazzo JM. Evolving issues in understanding and treating
bacterial vaginosis. Expert Review of Antiinfective Therapy
2004;2:913–22.
Nilsson 1997
Nilsson U, Hellberg D, Shoubnikova M, et al.Sexual
behavior risk factors associated with bacterial vaginosis and
Chlamydia trachomatis infection. Sexually Transmitted
Diseases 1997;24:241–6.
Nugent 1991
Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing
bacterial vaginosis is improved by a standardized method of
gram stain interpretation. Journal of Clinical Microbiology
1991;29:297–301.
Nyirjesy 1997
Nyirjesy P, Weitz MV, Grody MH, et al.Over-the-counter
and alternative medicines in the treatment of chronic
vaginal symptoms. Obstetrics and Gynecology 1997;90:50–3.
Olmsted 2003
Olmsted SS, Meyn LA, Rohan LC, et al.Glycosidase and
proteinase activity of anaerobic gram-negative bacteria
isolated from women with bacterial vaginosis. Sexually
Transmitted Diseases 2003;30:257–61.
Reid 2001
Reid G, Bruce AW, Fraser N, et al.Oral probiotics can
resolve urogenital infections. FEMS Immunology and
Medical Microbiology 2001;30:49–52.
Royce 1999
Royce RA, Jackson TP, Thorp JM, Jr, et al.Race/ethnicity,
vaginal flora patterns, and pH during pregnancy. Sexually
Transmitted Diseases 1999;26:96–102.
Senok 2005
Senok AC, Ismaeel AY, Botta GA. Probiotics: facts and
myths. Clinical Microbiology and Infection 2005;11:958–66.
Sewankambo 1997
Sewankambo N, Gray RH, Wawer MJ, et al.HIV-1
infection associated with abnormal vaginal flora morphology
and bacterial vaginosis. Lancet 1997;350:546–50.
Trutnovsky 2001
Trutnovsky G, Law C, Simpson JM, et al.Use of
complementary therapies in a sexual health clinic setting.
International Journal of STD & AIDS 2001;12:307–9.
Verhelst 2004
Verhelst R, Verstraelen H, Claeys G, et al.Cloning of 16S
rRNA genes amplified from normal and disturbed vaginal
microflora suggests a strong association between Atopobium
vaginae, Gardnerella vaginalis and bacterial vaginosis. BMC
Microbiology 2004;4:16.
Verstraelen 2004
Verstraelen H, Verhelst R, Claeys G, et al.Cultureindependent analysis of vaginal microflora: the unrecognized
association of Atopobium vaginae with bacterial vaginosis.
American Journal of Obstetrics and Gynecology 2004;191:
1130–2.
Verstraelen 2005
Verstraelen H, Senok AC. Vaginal Lactobacilli, probiotics
and IVF. Reproductive Biomedicine Online 2005;11:674–5.
Wiesenfeld 2003
Wiesenfeld HC, Hillier SL, Krohn MA, et al.Bacterial
vaginosis is a strong predictor of Neisseria gonorrhoeae
and Chlamydia trachomatis infection. Clinical Infectious
Diseases 2003;36:663–8.
Wilson 2002
Wilson JD, Ralph SG, Rutherford AJ. Rates of bacterial
vaginosis in women undergoing in vitro fertilisation for
different types of infertility. BJOG: An International Journal
of Obstetrics and Gynaecology 2002;109:714–7.
∗
Indicates the major publication for the study
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
12
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Anukam 2006a
Methods
Randomized double blind placebo controlled trial
Patients and assessors were blinded.
Drop out: 19/125 (15.2%)
Participants
125 premenopausal women aged 18-44 with BV as diagnosed with Nugent criteria, BV
blue test and clinical features. Participants were from Benin city, Nigeria.
Exclusion criteria included pregnancy, lactation, sexually transmitted disease, use of systemic/per vagina antibiotic agent within 14 days; use of investigational drugs within the
preceding 30 days. Hypersensitivity to some drugs including metronidazole; menstruation at time of diagnosis
Interventions
Cocktail of Probiotics used in conjunction with conventional antibiotics was compared
with placebo
Intervention group (n = 65) received: Oral L rhamnosus GR + L reuteri RC14 given
orally for 30 days AND Metronidazole 500 mg bd for 7 days taken orally 1 hr before
the probiotic capsule.
Placebo group received Placebo capsule given orally for 30 days AND Metronidazole
500 mg bd for 7 days taken orally 1 hr before the placebo capsule
Outcomes
Microbiological cure as assessed by Nugent score absence of clue cells and negative
salidase test at day 30
Resolution of symptoms and signs as reported by patient and clinician at day 30
Adverse effects
Notes
All outcomes were analysed
Risk of bias
Item
Authors’ judgement
Description
Adequate sequence generation?
Yes
Computer generated scheme prepared by
the pharmacy.
Allocation concealment?
Yes
Adequate, Allocation schedule was conducted off-site by the pharmacy and the
patients were allotted numbers. Identical looking probiotic and placebo capsules
were prepared and distributed in numbered containers by the pharmacy. The two
groups were similar at baseline regarding
the presence of BV
Blinding?
All outcomes
Yes
Study was described as “double blind”. Probiotic and metronidazole capsules looked
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
13
Anukam 2006a
(Continued)
identical. Authors confirmed to us that
both patients and assessors were blinded
Incomplete outcome data addressed?
All outcomes
Yes
19 women (3 in placebo group and 16 in
probiotic group) dropped out: Reasons included feeling well and no recurrence of
symptoms.This was confirmed in several
patients traced by clinical staff
Free of selective reporting?
Yes
Measures of the key outcome was obtained
from 106/125 women (85%) who returned
for follow up visit. All outcome measures
which the authors set out to test were described in the patients who came for follow
up. All subjects for whom outcome measures were available received the treatment
or placebo as allocated
Free of other bias?
Unclear
Patients lost to follow up hence possible
attrition bias. There was no statement on
intention-to-treat analysis
Anukam 2006b
Methods
Randomized controlled trial. Patients were not blinded but the outcome assessors were
blinded
Exclusion criteria: Presence of STI; age < 18 and > 50 years; pregnancy; ongoing menstruation or due to menstruate in the following 5 days. 5/40 lost to follow up on day 30
Participants
40 pre-menopausal women with BV as diagnosed by Nugent score, positive BV blue
salidase test, pH > 4.5 and KOH test positive and clinical features. Participants were
from clinics around Benin City, Nigeria
Interventions
Cocktail of probiotics compared with antibiotic regimen
Intervention group (n = 20): L rhamnosus GR1 (1x 109 ) + L reuteri RC14 (1x109 ) given
per vagina at bedtime for 5 consecutive days
Comparison group (n = 20): Metronidozole 0.75% vaginal gel given bd (morning and
evening per vagina) for 5 days
Outcomes
Microbiological cure as assessed by Nugent scoring at days 6, 15 and 30
Clinical cure as reported by patient and clinician at days 6, 15, and 30
Notes
Adverse effects not reported
All outcomes were analysed
Risk of bias
Item
Authors’ judgement
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Description
14
Anukam 2006b
(Continued)
Adequate sequence generation?
Yes
Computer generated scheme and patients
in both study groups were similar at baseline
Allocation concealment?
Yes
Adequate, although allocation concealment was not indicated in the article, contact with the authors confirmed that this
was carried out and the allocation schedule
was conducted “off-site” by the pharmacy
and the patients were allotted numbers
Blinding?
All outcomes
No
The patients were not blinded to the
treatment they received as one group had
probiotic capsules and the other group
were given metronidazole gel. However,
the assessors (clinical and laboratory) were
blinded to which of the treatment arms the
patients belonged to
Incomplete outcome data addressed?
All outcomes
Yes
There was 100% follow-up on days 6 & 15
and all women demonstrated compliance
by returning empty treatment vials. Three
women in the probiotic arm and 2 in the
metronidaozole group failed to return for
follow up on day 30 despite all efforts to
contact them
Free of selective reporting?
Yes
All identified outcomes were reported in
those who came for follow up. All subjects
for whom outcome measures were available
received the treatment or placebo as allocated
Free of other bias?
Unclear
Five out of 40 women lost to follow up on
day 30, possible attrition bias, no intention
to treat analysis
Eriksson 2005
Methods
Double blind randomized controlled trial
Patient and clinician blinding was not absolute because the lactobacilli tampons had a
slight difference in colour compared to the placebo tampon
Drop outs: 30/255 (11.8%) women were lost to follow-up. Another 38 women were
excluded from “per protocol” analysis because of protocol violations
Participants
255 women with BV aged 20-53 enrolled. BV diagnosed by Nugent criteria. Participants
from 13 clinics in Finland, Norway and Sweden
Exclusion criteria: Pregnancy, planning pregnancy, breastfeeding, antibiotic treatment
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
15
Eriksson 2005
(Continued)
in preceding week to BV diagnosis; vaginal candidiasis & genital chlamydia infection
Interventions
Cocktail of probiotics used in conjunction with conventional antibiotics compared with
placebo.
All patients were first treated with clindamycin ovules 100 mg daily per vagina for 3
days
Intervention group (n = 127): Probiotic tampons containing L gasseri, L casei var rhamnosus & L fermentum (each 108 org); had to use at least 5 tampons during menstruation
Placebo group (n = 128): Tampons without probiotics; had to use at least 5 tampons
during menstruation
Outcomes
Microbiological cure as evaluated by Nugent score and Amsel criteria. Resolution of
symptoms as reported by patient
Primary outcome was determined by Nugent score after the menstruation during which
tampons were used and also after the second menstruation
Notes
All outcomes analysed. Only 187 women were included in “per protocol” analysis. Thirty
women were lost to follow up (15 in treatment group and 15 in placebo group). Thirtyeight women were excluded for violating protocol (21 in treatment group and 17 in
placebo group). Intention-to-treat analysis was carried out. Adverse reactions reported
Risk of bias
Item
Authors’ judgement
Description
Adequate sequence generation?
Yes
Randomization was carried and this was
confirmed by the authors. As stated in
the article “Patients were randomized into
study groups” but the method used was not
described
Allocation concealment?
No
Not used
Blinding?
All outcomes
No
The study is described as “double blind”
but there was slight colour differences
between the probiotic tampons and the
placebo tampons. Thus the blinding of patients and clinical assessors is inadequate
and this introduces a risk of bias
The results of between-group statistical
comparisons as well as the measure of the
size of the treatment effect were reported
Incomplete outcome data addressed?
All outcomes
Yes
Not all the subjects for whom outcome
measures were available received the treatment or placebo as allocated. The authors
reported that 224 (87%) out of the 255
women allocated at the start of the study
returned for follow up visit. Thirty pa-
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
16
Eriksson 2005
(Continued)
tients were lost to follow up, and 38 more
were excluded from the “per protocol analysis” due to violation of the protocol. Data
for those excluded for protocol violation
and drop outs were addressed. Intentionto-treat analysis was carried out for the microbiological cure which is a key study outcome
Free of selective reporting?
Yes
All identified outcomes were reported in
those who came for follow up (including
protocol violators)
Free of other bias?
Yes
Although patients were lost to follow up,
intention-to-treat analysis was carried out
Parent 1996
Methods
Randomized placebo controlled trial with parallel group design. Blinding of patients and
clinicians is unclear as it is not mentioned in the article
Drop outs: 17 out of 32 patients dropped out (53%)
Participants
32 women non-menopausal; (ages 20 to 52 years) with BV as diagnosed by Amsel criteria;
24 women were pregnant and 8 were non-pregnant
Conducted in 3 hospital centers in Belgium
Exclusion criteria: Use of topical or systemic hormonal treatment (except oral contraceptives) in the preceding 2 weeks;
Any local vaginal treatment in the preceding 2 weeks; metrorrhagia; acute vaginitis
Interventions
The treatment group (n = 17; 11 non-pregnant and 6 pregnant) received vaginal tablets
containing L acidophilus (106 CFU) and oestriol 0.03 mg
Placebo group (n = 15; 13 non-pregnant and 2 pregnant) received vaginal tablets which
did not contain either probiotic or oestrol.
Vaginal tabs inserted at night: once daily or twice daily for 6 days. Twice daily dosing was
for pregnant women with abundant leukorrhea or those whose therapy was interrupted
during menstruation and continued post-menstruation
Outcomes
Microbiological cure (Nugent criteria) and clinical cure were determined on days 15 and
28 post intervention
Notes
Adverse effects and tolerability reported. No response from the authors to our enquiries
Risk of bias
Item
Authors’ judgement
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Description
17
Parent 1996
(Continued)
Adequate sequence generation?
Yes
Randomization was carried and the
method was described as a “pre-determined
randomization scheme”
Allocation concealment?
No
No mention or description suggestive of allocation concealment and although the authors were contacted to clarify these issues,
no response was received
Blinding?
All outcomes
Unclear
The trial was described as a randomized
placebo-controlled clinical trial with parallel group design. There was no mention or
description of blinding in the article. We
did not receive any response to our communications for verification
Incomplete outcome data addressed?
All outcomes
No
In the treatment arm, 11/17 women returned on day 15 for evaluation and 8 /
17 returned for evaluation on day 28 and
hence the study was particularly prone to
attrition bias. In the control group 10 out
of 15 women were assessed on day 15 and
only 7 were evaluated on day 28. The data
for those lost to follow up were not addressed
Free of selective reporting?
Yes
All identified outcomes were reported in
those who came for follow up
Free of other bias?
Unclear
Possible attrition bias, patients lost to follow up and no intention-to-treat analysis
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Della 2006
This study described the comparison of one type of probiotic (L. acidophilus given per vagina) vs. the same probiotic
given in combination with an oral probiotic. This is not in keeping with our systematic review protocol. There
was no placebo control in this study
Hallen 1992
The primary outcome (microbiological cure) in this study was not assessed by Nugent criteria
Larsson 2008
The microbiological cure was assessed outside of the 21-30 day post-treatment window outlined in the protocol
for this review. In the first menstrual cycle, women received a 7-day course of clindamycin followed immediately
by 10-day course of probiotic capsules and vaginal smears were taken after cessation of bleeding. This time frame
was outside the 21-30 day post-treatment. The main outcome was reduction of recurrence and women who were
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
18
(Continued)
BV positive after the first menstruation were excluded from the study. Treatment efficacy was based on cure after
one month and length of time to relapse. Definition of cure was based on Hay/Ison criteria
Marrazzo 2006
The research question of the paper is not consistent with the research question of the systematic review: the paper
specifically addresses the acceptability of probiotic; thus the primary outcome in the paper was acceptability and
patient satisfaction
Neri 1993
The control group consisted of women who refused any kind of intervention and there is no mention of any per
vagina placebo given to these women. Methodology for assessment of primary outcome was not Nugent scoring
Ozkinay 2005
Micrological assessment of vaginal smears for number of lactobacilli, leucocytes and pathogenic micro-organisms
was carried out, but not according to Nugent, and hence not in accordance with our protocol
Reid 2001a
The study is not a randomized controlled trial
Reid 2001b
Study population consisted of healthy women without bacterial vaginosis. There was no control population in the
study
Reid 2003
64 healthy women who were not allocated for intervention on the basis of vaginal microflora at the beginning of
the trial
Reid 2004
The participants were healthy women who were not allocated for intervention on the basis of diagnosis of BV at
the beginning of the trial. Primary outcome was evaluated on day 56 (outside of the 21-30 day window specified
in our review protocol)
Ronnqvist 2006
Population studied were healthy women not women with BV
Shalev 1996
The definition of the study population is unclear. It is a cross-over trial and there was no control group for
comparison; no indication of whether the time-out period was sufficient. There was a huge attrition rate
Wewalka 2002
Not a randomized study (authors described the study as a prospective controlled blind clinical study);
The intervention was comparison of probiotic with an antiseptic (betadine povidone) which is not in keeping
with the systematic review protocol. Primary outcome was assessed on days 8 and 15, well before the 21-30 day
window specified in the review protocol
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
19
DATA AND ANALYSES
Comparison 1. Probiotic + Metronidazole versus placebo + Metronidazole
Outcome or subgroup title
1 Microbiological cure
No. of
studies
No. of
participants
1
106
Statistical method
Odds Ratio (M-H, Fixed, 95% CI)
Effect size
0.09 [0.03, 0.26]
Comparison 2. Vaginal probiotic capsules versus metronidazole vaginal gel
Outcome or subgroup title
1 Microbiological cure
2 Resolution of symptoms as
reported by patient
3 Adverse effects
No. of
studies
No. of
participants
1
1
35
35
Odds Ratio (M-H, Fixed, 95% CI)
Odds Ratio (M-H, Fixed, 95% CI)
0.27 [0.07, 1.10]
0.27 [0.05, 1.57]
1
35
Odds Ratio (M-H, Fixed, 95% CI)
0.08 [0.01, 0.49]
Statistical method
Effect size
Comparison 3. Probiotic versus placebo (after open treatment with clindamycin ovules)
Outcome or subgroup title
1 Microbiological cure
2 Resolution of symptoms as
reported by patient
3 Clinical cure as reported by
physician
No. of
studies
No. of
participants
1
1
217
216
Odds Ratio (M-H, Fixed, 95% CI)
Odds Ratio (M-H, Fixed, 95% CI)
1.21 [0.67, 2.19]
0.75 [0.43, 1.32]
1
224
Odds Ratio (M-H, Fixed, 95% CI)
1.11 [0.65, 1.89]
Statistical method
Effect size
Comparison 4. Probiotics with estriol versus placebo
Outcome or subgroup title
1 Microbiological cure
2 Clinical cure as reported by
physician
No. of
studies
No. of
participants
1
1
15
17
Statistical method
Odds Ratio (M-H, Fixed, 95% CI)
Odds Ratio (M-H, Fixed, 95% CI)
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Effect size
0.02 [0.00, 0.47]
0.04 [0.00, 0.56]
20
Analysis 1.1. Comparison 1 Probiotic + Metronidazole versus placebo + Metronidazole, Outcome 1
Microbiological cure.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 1 Probiotic + Metronidazole versus placebo + Metronidazole
Outcome: 1 Microbiological cure
Study or subgroup
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
Anukam 2006a
6/49
34/57
100.0 %
0.09 [ 0.03, 0.26 ]
Total (95% CI)
49
57
100.0 %
0.09 [ 0.03, 0.26 ]
M-H,Fixed,95% CI
Total events: 6 (Treatment), 34 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 4.60 (P < 0.00001)
Test for subgroup differences: Not applicable
0.01
0.1
1
Favours experimental
10
100
Favours control
Analysis 2.1. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 1
Microbiological cure.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 2 Vaginal probiotic capsules versus metronidazole vaginal gel
Outcome: 1 Microbiological cure
Study or subgroup
Probiotics
Metronidazole gel
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
Anukam 2006b
6/17
12/18
100.0 %
0.27 [ 0.07, 1.10 ]
Total (95% CI)
17
18
100.0 %
0.27 [ 0.07, 1.10 ]
M-H,Fixed,95% CI
Total events: 6 (Probiotics), 12 (Metronidazole gel)
Heterogeneity: not applicable
Test for overall effect: Z = 1.82 (P = 0.068)
Test for subgroup differences: Not applicable
0.01
0.1
Favours experimental
1
10
100
Favours control
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
21
Analysis 2.2. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 2
Resolution of symptoms as reported by patient.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 2 Vaginal probiotic capsules versus metronidazole vaginal gel
Outcome: 2 Resolution of symptoms as reported by patient
Study or subgroup
Probiotics
Metronidazole gel
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
Anukam 2006b
2/17
6/18
100.0 %
0.27 [ 0.05, 1.57 ]
Total (95% CI)
17
18
100.0 %
0.27 [ 0.05, 1.57 ]
M-H,Fixed,95% CI
Total events: 2 (Probiotics), 6 (Metronidazole gel)
Heterogeneity: not applicable
Test for overall effect: Z = 1.46 (P = 0.14)
Test for subgroup differences: Not applicable
0.01
0.1
1
Favours experimental
10
100
Favours control
Analysis 2.3. Comparison 2 Vaginal probiotic capsules versus metronidazole vaginal gel, Outcome 3
Adverse effects.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 2 Vaginal probiotic capsules versus metronidazole vaginal gel
Outcome: 3 Adverse effects
Study or subgroup
Probiotics
Metronidazole gel
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
Anukam 2006b
2/17
11/18
100.0 %
0.08 [ 0.01, 0.49 ]
Total (95% CI)
17
18
100.0 %
0.08 [ 0.01, 0.49 ]
M-H,Fixed,95% CI
Total events: 2 (Probiotics), 11 (Metronidazole gel)
Heterogeneity: not applicable
Test for overall effect: Z = 2.76 (P = 0.0058)
Test for subgroup differences: Not applicable
0.01
0.1
Favours experimental
1
10
100
Favours control
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
22
Analysis 3.1. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules),
Outcome 1 Microbiological cure.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 3 Probiotic versus placebo (after open treatment with clindamycin ovules)
Outcome: 1 Microbiological cure
Study or subgroup
Eriksson 2005
Total (95% CI)
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
33/108
29/109
100.0 %
1.21 [ 0.67, 2.19 ]
108
109
100.0 %
1.21 [ 0.67, 2.19 ]
M-H,Fixed,95% CI
Total events: 33 (Treatment), 29 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.64 (P = 0.52)
Test for subgroup differences: Not applicable
0.01
0.1
1
Favours experimental
10
100
Favours control
Analysis 3.2. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules),
Outcome 2 Resolution of symptoms as reported by patient.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 3 Probiotic versus placebo (after open treatment with clindamycin ovules)
Outcome: 2 Resolution of symptoms as reported by patient
Study or subgroup
Eriksson 2005
Total (95% CI)
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
Weight
Odds Ratio
34/108
41/108
100.0 %
0.75 [ 0.43, 1.32 ]
108
108
100.0 %
0.75 [ 0.43, 1.32 ]
M-H,Fixed,95% CI
Total events: 34 (Treatment), 41 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 1.00 (P = 0.32)
Test for subgroup differences: Not applicable
0.01
0.1
Favours experimental
1
10
100
Favours control
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
23
Analysis 3.3. Comparison 3 Probiotic versus placebo (after open treatment with clindamycin ovules),
Outcome 3 Clinical cure as reported by physician.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 3 Probiotic versus placebo (after open treatment with clindamycin ovules)
Outcome: 3 Clinical cure as reported by physician
Study or subgroup
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
47/111
45/113
100.0 %
1.11 [ 0.65, 1.89 ]
111
113
100.0 %
1.11 [ 0.65, 1.89 ]
Eriksson 2005
Total (95% CI)
Weight
Odds Ratio
M-H,Fixed,95% CI
Total events: 47 (Treatment), 45 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.38 (P = 0.70)
Test for subgroup differences: Not applicable
0.01
0.1
1
Favours experimental
10
100
Favours control
Analysis 4.1. Comparison 4 Probiotics with estriol versus placebo, Outcome 1 Microbiological cure.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 4 Probiotics with estriol versus placebo
Outcome: 1 Microbiological cure
Study or subgroup
Parent 1996
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
1/8
6/7
100.0 %
0.02 [ 0.00, 0.47 ]
8
7
100.0 %
0.02 [ 0.00, 0.47 ]
Total (95% CI)
Weight
Odds Ratio
M-H,Fixed,95% CI
Total events: 1 (Treatment), 6 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 2.46 (P = 0.014)
Test for subgroup differences: Not applicable
0.01
0.1
Favours experimental
1
10
100
Favours control
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
24
Analysis 4.2. Comparison 4 Probiotics with estriol versus placebo, Outcome 2 Clinical cure as reported by
physician.
Review:
Probiotics for the treatment of bacterial vaginosis
Comparison: 4 Probiotics with estriol versus placebo
Outcome: 2 Clinical cure as reported by physician
Study or subgroup
Treatment
Control
Odds Ratio
n/N
n/N
M-H,Fixed,95% CI
1/8
7/9
100.0 %
0.04 [ 0.00, 0.56 ]
8
9
100.0 %
0.04 [ 0.00, 0.56 ]
Parent 1996
Total (95% CI)
Weight
Odds Ratio
M-H,Fixed,95% CI
Total events: 1 (Treatment), 7 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 2.39 (P = 0.017)
Test for subgroup differences: Not applicable
0.01
0.1
Favours experimental
1
10
100
Favours control
APPENDICES
Appendix 1. Cochrane
“Bacterial *4 Vaginosis” or
“Vaginitis *4 Bacterial” or
“Vaginitis *4 Nonspecific” or
“Bacterial *4 Vaginitides” or
“Bacterial *4 Vaginitis ” or
“Bacterial *4 Vaginoses” or
“Vaginitides *4 Bacterial” or
“Vaginoses *4 Bacterial” or
“Vaginosis, Bacterial” or
“colpitis *4 Bacterial” or
“Vaginitis *4 Nonspecific” or
(’vaginitis’/exp AND (Bacterial or bacteria or nonspecific or “non-specific” or “non specific”))
AND
“clinical trial”/exp or “randomized controlled trial”/exp or randomization/ or “single blind procedure”/exp or “double blind procedure”/
or “crossover procedure”/ or placebo/ or “randomi*ed controlled trial*” or rct or “random allocation” or “randomly allocated” or
“allocated *2 random” or “random *2 allocated” OR “single blind*” or “double blind*” or placebo* or “prospective study”/exp or “meta
analysis”/exp or ((meta *2 analy*) or metaanalys*) or (systematic *3 (review* OR overview*))
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
25
Appendix 2. Medline (1966-2008)
1.
bacterial vaginosis.mp. or exp Vaginosis, Bacterial/
2.
bacterial vaginitis.mp. or exp Vaginosis, Bacterial/
3.
exp Metronidazole/ or exp Vaginitis/ or exp Vagina/ or non*specific vaginitis.mp. or exp Sexually Transmitted Diseases/ or exp
Gardnerella vaginalis/ or Vaginosis, Bacterial/
4.
bacterial vaginoses.mp. or exp Vaginosis, Bacterial/
5.
1 or 2 or 3 or 4
6.
exp Lactobacillus/ or exp Probiotics/
7.
Lactobacillus helveticus/ or lactobacilli.mp. or Lactobacillus casei/ or Lactobacillus leichmannii/ or Lactobacillus reuteri/ or
Lactobacillus brevis/ or Lactobacillus plantarum/ or Lactobacillus delbrueckii/ or Lactobacillus rhamnosus/ or Lactobacillus fermentum/
8.
Lactobacillus helveticus/ or lactobacillus.mp. or Lactobacillus casei/ or Lactobacillus leichmannii/ or Lactobacillus reuteri/
or Lactobacillus brevis/ or Lactobacillus plantarum/ or Lactobacillus acidophilus/ or Lactobacillus/ or Lactobacillus rhamnosus/ or
Lactobacillus fermentum/
9.
antibiotics.mp.
10. drug therapy.mp. or Drug Therapy/
11. complementary.mp. or exp Complementary Therapies/
12. alternative.mp.
13. treatment.mp.
14. 6 or 7 or 8
15. 9 or 10 or 11 or 12
16. 5 AND 14 AND 15
Appendix 3. EMBASE (1980-2007)
1. “Bacterial Vaginosis”
2. “Vaginitis Bacterial”
3. “Vaginitis Nonspecific”
4. “Bacterial Vaginitides”
5. “Bacterial Vaginitis ”
6. “Bacterial Vaginoses”
7. “Vaginitides Bacterial”
8. “Vaginoses Bacterial”
9. Vaginosis, Bacterial
10. colpitis Bacterial
11. “Vaginitis Nonspecific” (’vaginitis’ and (Bacterial or bacteria or nonspecific or “non-specific” or “non specific”)))
12. or / #1-11
13. Lactobacill* tw
14. exp Lactobacillus helveticus/ or lactobacillus.mp. or exp Lactobacillus casei/ or exp Lactobacillus leichmannii/ or exp
Lactobacillus reuteri/ or exp Lactobacillus brevis/ or exp Lactobacillus plantarum/ or exp Lactobacillus delbrueckii/ or exp
Lactobacillus acidophilus/ or exp Lactobacillus/ or exp Lactobacillus rhamnosus/ or exp Lactobacillus fermentum/
15. #13 or #14
16. “clinical trial” or “randomized controlled trial”.mp.
17. randomization/ or “single blind procedure”.mp.
18. “double blind procedure”/ or “crossover procedure”/ or placebo/ or “randomised controlled trial”.mp.
19. rct.mp. or “random allocation”.mp.
20. “randomly allocated”.mp.
21. “allocated random”.mp.
22. “random allocated”.mp.
23. “single blind”.mp.
24. “double blind*”.mp.
25. placebo.mp.
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
26
26.
27.
28.
29.
30.
“prospective study”.mp.
“meta analysis”.mp.
“systematic review”.mp.
or / #16-28
#12 AND #15 AND #29
Appendix 4. CINAHL (1982-2007)
bacter* vagin*or non*specific vagin* or vagin* AND probiotic* or lactobacill*
Appendix 5. ISI science citation index (1955-2007)
probiotic$ and bacter$ and vagin$
HISTORY
Protocol first published: Issue 4, 2006
Review first published: Issue 4, 2009
CONTRIBUTIONS OF AUTHORS
ACS formulated the idea for the review. ACS and HV developed the protocol, carried out data collection and analysis and wrote the
review. MT and GAB contributed in the development of the protocol and writing the review.
DECLARATIONS OF INTEREST
The authors declare that they do not have any conflict of interest.
SOURCES OF SUPPORT
Internal sources
• University of Sharjah, United Arab Emirates.
Financial funding from University Research Board (URB Grant #060907)
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
27
External sources
• No sources of support supplied
DIFFERENCES BETWEEN PROTOCOL AND REVIEW
The review has been carried out according to the criteria outlined in the protocol. As an improvement we widened the search to include
more electronic databases in addition to those listed in the protocol.
INDEX TERMS
Medical Subject Headings (MeSH)
∗ Lactobacillus;
Anti-Bacterial Agents [administration & dosage]; Clindamycin [administration & dosage]; Estriol [administration &
dosage]; Probiotics [∗ administration & dosage]; Randomized Controlled Trials as Topic; Vagina [microbiology]; Vaginosis, Bacterial
[microbiology; ∗ therapy]
MeSH check words
Female; Humans
Probiotics for the treatment of bacterial vaginosis (Review)
Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
28