Lactobacilli-containing vaginal probiotics to cureor prevent bacterial or fungal vaginal dysbiosis:a systematic review and recommendations forfuture trial designsJHHM van de Wijgert,a,b MC Verwijsa

a Institute of Infection and Global Health, University of Liverpool, Liverpool, UK b Julius Center for Health Sciences and Primary Care,

University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands

Correspondence: Professor JHHM van de Wijgert, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht,

Utrecht University, Stratenum Huispost nr STR 6.131, PO Box 85 500, 3508 GA Utrecht, the Netherlands.

Email: j.h.h.vandewijgert@umcutrecht.nl

Accepted 17 June 2019. Published Online 8 August 2019.

Background Vaginal probiotics claiming to cure and/or prevent

bacterial and/or fungal vaginal dysbiosis are available on the

market but, until recently, did not have to be approved as drugs

for human use.

Objectives We evaluated the impact of vaginal probiotics on

bacterial vaginosis (BV) and vulvovaginal candidiasis (VVC) cure

and/or recurrence, as well as vaginal microbiota (VMB)

composition and vaginal detection of probiotic strains.

Search strategy We performed a systematic literature search in

MEDLINE and Embase up to 15 January 2019.

Selection criteria There were no restrictions in probiotic strains/

formulations, study populations, and designs. BV had to be

diagnosed by Nugent or Ison-Hay Gram stain scoring, VVC by

culture, wet mount or PCR, and VMB composition/detection by

molecular techniques.

Data collection and analysis The authors independently extracted

data.

Main results All 22 vaginal probiotics evaluated in the 34 eligible

studies contained Lactobacillus strains, and some contained

additional active ingredients. The probiotics hold promise for BV

cure and prevention, but much less so for VVC cure and

prevention. No major safety concerns were reported in any of the

studies. Vaginal detection of probiotic strains never lasted long

beyond the dosing period, suggesting that they did not colonise the

vagina. However, findings are not definitive because heterogeneity

was high and the quality of most studies suboptimal.

Conclusions Availability of vaginal probiotics for vaginal health

indications will likely decline in 2020 because of regulatory changes.

We urge the field to invest in clinical evidence-based product

development and to conduct future trials more rigorously.

Keywords Bacterial vaginosis, lactobacilli, live biotherapeutic

products, probiotics, vaginal microbiota, vulvovaginal candidiasis.

Tweetable abstract Lactobacilli-containing vaginal probiotics hold

promise for bacterial vaginosis cure and prevention, but not for

vulvovaginal candidiasis.

Please cite this paper as: van de Wijgert JHHM , Verwijs MC. Lactobacilli-containing vaginal probiotics to cure or prevent bacterial or fungal vaginal

dysbiosis: a systematic review and recommendations for future trial designs. BJOG 2019; https://doi.org/10.1111/1471-0528.15870

Introduction

Understanding of the vaginal microbiota (VMB) has

advanced significantly in the last two decades due to the

increased availability of next-generation sequencing.1 Opti-

mal VMBs are dominated by lactobacilli, which do not

cause immune activation or inflammation and are not

associated with adverse outcomes such as HIV acquisition

and preterm birth.2 Of the five most common lactobacilli

in the vaginal niche (Lactobacillus crispatus, Lactobacillus

iners, Lactobacillus gasseri, Lactobacillus jensenii, and Lacto-

bacillus vaginalis), L. crispatus-domination is most efficient

at keeping bacterial vaginosis (BV)-associated anaerobes at

bay. L. crispatus produces more lactic acid than the other

Prospero registration: CRD42017075717.

1ª 2019 Royal College of Obstetricians and Gynaecologists

DOI: 10.1111/1471-0528.15870

www.bjog.orgSystematic review

lactobacilli and produces several other antimicrobial com-

pounds.3

Bacterial vaginosis is the most common bacterial vaginal

dysbiosis and is usually characterised by a highly diverse

community of anaerobes including Gardnerella vaginalis,

but sometimes by G. vaginalis-domination.1 G. vaginalis is

thought to play an important role in biofilm formation,

and G. vaginalis-domination may indicate the presence of a

biofilm.4 Vaginal dysbiosis by bacterial pathobionts with a

higher pathogenicity potential than BV-anaerobes (such as

streptococci, staphylococci, enterococci, and Enterobacteri-

aceae) is much less common, but clinically relevant.2

Almost all cases of fungal vaginal dysbiosis are caused by

Candida species, predominantly Candida albicans.2 In con-

trast to BV-anaerobes, bacterial pathobionts and Candida

species often co-exist with lactobacilli in the VMB, perhaps

because they can tolerate high concentrations of lactic

acid.5

This increased VMB knowledge has led to a renewed

interest in using exogenous lactobacilli (probiotics or live

biotherapeutic products)6 to optimise the VMB as part of

curative or preventive vaginal dysbiosis interventions. There

are multiple reasons for this. Antibiotic BV treatment (with

oral or vaginal metronidazole or clindamycin) and antifun-

gal treatment of vulvovaginal candidiasis (VVC; with oral

or vaginal -azoles) result in suboptimal cure fractions and

high recurrence rates.7 BV biofilms are not penetrated suffi-

ciently by antibiotics.4 In addition, prolonged use increases

the probability of side effects and drug resistance. Possible

mechanisms by which probiotic strains might exert positive

effects include niche occupation so that other bacteria can-

not expand and biofilms cannot be established, increased

local production or release of lactic acid and other antimi-

crobial compounds, modulation of local cervicovaginal

mucosal immune responses, and/or inhibition of C. albi-

cans hyphae formation.3,8,9

We performed a systematic review to examine the

impact of vaginally applied lactobacilli-containing probi-

otics on the VMB in clinical studies. We also offer recom-

mendations for future trial designs based on this review

and our own experiences.10

Methods

The systematic review focused on BV and VVC cure and/or

recurrence as defined by conventional diagnostic methods,

as well as VMB composition and vaginal detection of pro-

biotic strains as assessed by molecular methods. It was con-

ducted according to the PRISMA 200911 guidelines and

registered on PROSPERO (CRD42017075717).12 Patients

and the public were not involved in this review and a core

outcome set was not available. We searched MEDLINE13

and Embase14 up to 15 January 2019 (search terms in

Appendix S1), and also included a trial completed by our

group that had been fully analysed, and that had been pre-

sented at an international conference, but was not yet pub-

lished, on that date.10 No filters were used except for

‘English language’. Duplicates were manually removed.

Two researchers independently screened titles and abstracts

(with JvdW as tiebreaker) and additional articles were iden-

tified by screening the reference lists of selected articles, rel-

evant reviews, and expert opinion pieces. The

corresponding author of a conference abstract was con-

tacted but this did not yield sufficient information for

inclusion. We included studies of any vaginally applied

probiotic in sexually active women regardless of menopau-

sal, pregnancy or vaginal dysbiosis status. We excluded two

studies that assessed yoghurt because the bacteria present

in the yoghurt were not specified. All study designs and

probiotic use schedules (with or without additional oral or

vaginal antibiotic or antifungal treatment) were allowed.

BV had to be diagnosed by Nugent15 or Ison-Hay16 Gram

stain scoring or the bacterial VMB comprehensibly assessed

by molecular methods. Studies using Amsel criteria, bacte-

rial culture, other forms of Gram stain scoring (Lactobacil-

lus counts only), rapid tests or participant-reported

symptoms and/or clinician-observed signs only, or with

unclear diagnostic procedures, were excluded. VVC had to

be diagnosed by wet mount, culture or Candida polymerase

chain reaction (PCR). Studies using participant-reported

symptoms and/or clinician-observed signs only or with

unclear diagnostic procedures were excluded.

Both authors extracted data from all selected articles and

discussed their findings. Each article was assessed for risk

of bias as described in the Cochrane Handbook for System-

atic Reviews of Interventions.17 We also determined an

overall risk of bias per study. Articles for which all poten-

tial biases (selection, performance, detection/ascertainment,

attrition, reporting, and ‘other’ bias, with the latter includ-

ing confounding) were assessed as low risk were judged as

‘overall low risk of bias’, those with a maximum of two

high or unclear risks of bias as ‘overall medium risk of

bias’, and those with three or more high or unclear risks of

bias as ‘overall high risk of bias’. Articles were not excluded

based on risk of bias assessments. Risk of bias plots were

made using REVMAN 5.3.5 (The Cochrane Collaboration,

Copenhagen, Denmark). If no statistical testing was

reported, but absolute numbers of cases per exposure

group were reported, we performed Chi-square testing

(cross-sectional comparisons between groups) or McNe-

mar’s testing (pre-post within-arm comparisons) using

STATA 13 (StataCorp, College Station, TX, USA). The full

data extraction database can be requested from the corre-

sponding author by email.

2 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

Results

The search identified 2949 unique articles but only 34 stud-

ies were eligible (flow diagram in Figure 1).10,18–50 Thirteen

studies reported 14 results on BV and/or molecular VMB

composition (Table 1),10,18–29 12 studies reported 12 results

on VVC (Table 2),29–41 and 14 studies reported 15 results

on vaginal probiotic strain detection using molecular tech-

niques (Table S1).10,26,27,30,39,42–50 The selected studies eval-

uated 22 different vaginal probiotics (Table S2). The most

studied were Gynoflor (L. acidophilus KS400 plus 30 mi-

crogram oestriol; three studies with 359 partici-

pants),32,33,39,47 Lactin-V (L. crispatus CTV-05; four studies

with 132 participants),30,31,46 and L. reuteri RC-14 with

L. rhamnosus GR-1 (four studies with 47 partici-

pants)23,26,42,44. The majority of probiotics contained

108–1010 colony-forming units per dose, and all contained

lactobacilli as an active ingredient. Additional active

ingredients included oestriol (Gynoflor), Streptococcus

thermophilus (Lactagyn, Femilac), Bifidobacterium bifidum

(Ecologic Femi+), Pediococcus acidilactici (Ellen capsules),

and acidifying agents (Kramegin, Florisia, ActiCand 30, and

one unnamed probiotic).

BV and molecular VMB composition outcomesOf the 13 studies with BV and/or molecular VMB compo-

sition as outcome, five were judged to have medium and

eight high overall risk of bias (Figure 2). Most studies (11/13)

were randomised controlled trials, including six with

placebo controls,18,20–22,26,28 two with ‘no intervention’

controls,10,19 four with metronidazole or clindamycin con-

trols,10,21,23,25 and one with a vaginal pH-lowering tablet

control24 (two trials included both placebo/‘no interven-

tion’ and antibiotic controls10,21) (Table 1). Five clinical

trials were judged medium risk10,18–21 and the other six

high risk.22–29 The two remaining studies (both judged

high risk) were pre-post intervention studies in women

receiving a vaginal probiotic without an antibiotic.27,29

Nine of the 14 main results reported in the 13 studies

were on BV cure or post-treatment Nugent score improve-

ment (2/9 after initial antibiotic treatment and 7/9 without

antibiotic use), nine on BV recurrence (6/9 and 3/9,

• Full-text articles by snowball-ling of reference lists (n = 28)

• Unpublished study by the authors (n = 1)

Records identified through databases search string

(n = 5416)

Records after duplicates removed,screened based on title/abstract

(n = 2949)

Full-text articles assessed foreligibility(n = 105)

Potentially relevant articles identified

(n = 76)

Articles excluded based on title/abstract (n = 2873)

• Oral probiotics (n = 44)• Useful for introduction

and/or discussion (n = 34)• Irrelevant (n = 2795)

Excluded articles (n = 71)• Amsel criteria only (n = 15)• Irrelevant outcome (n = 10)• Not in English (n = 9)• No full-text available (n = 9)• No probiotics used (n = 8)• Non-human study (n = 6)• Oral probiotics (n = 5)• Data uninterpretable (n = 3)• Yoghurt used (n = 2)• Conference/poster abstract

(n = 2)• Data not released by authors

(n = 1)• Safety data only (n = 1)

Articles from which data were extracted (n = 34, including 34

efficacy results)• One article reported results of

two separate trials• In two cases, two separate

articles reported results of the same trial

• One article included results for two vaginal probiotics

Figure 1. PRISMA flow diagram.

3ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics

Table 1. Summary of BV/VMB study results

References Probiotic Study design Results

Articles with a medium overall risk of bias

Larsson et al.18 EcoVag (L. gasseri EB01-DSM

14869 + L. rhamnosus Lbp

PB01-DSM 14870)

RCT: placebo control 10 days of EcoVag after clindamycin did not improve BV cure

(by Ison-Hay; 32/50 cured) compared with placebo (37/50

cured) but significantly reduced the cumulative incidence of

BV/intermediate microbiota within 4 menstrual cycles when

used for 10 days after each menses (13/37 versus 21/39

incident cases, respectively)

Petricevic et al.19 Gynophilus (L. casei

rhamnosus Lcr35)

RCT: no-intervention

control

Seven days of Gynophilus after clindamycin resulted in

significantly improved BV cure (by Nugent 0–3) compared

with women receiving clindamycin only (69/83 versus 31/88,

respectively) assessed 5 weeks after cessation of the

intervention period

Mastro-marino

et al.20Florisia (L. brevis

CD2 + L. salivarius subsp.

salicinius

FV2 + L. plantarum FV9)

RCT: placebo control Seven days of Florisia (without any antibiotic) was significantly

more efficacious in curing BV (15/18 Nugent 0–3, 3/18

Nugent 4–6) than placebo (2/16 Nugent 4–6; 14/16

persistent Nugent 7–10) and was associated with significantly

lower BV (Nugent 7–10) cumulative incidence in the 2 weeks

after cessation of Florisia (7/18 and 13/16, respectively)

Bradhaw et al.21 Gynoflor (L. acidophilus

KS400 + 0.03 mg estriol)

RCT: placebo and

clindamycin cream

controls

Twelve days of Gynoflor after oral metronidazole treatment

for BV (Nugent 7–10) resulted in a borderline (P = 0.13)

lower BV recurrence (9/133) at Month 1 compared with

12 days of placebo (13/135) but higher compared with

7 days of vaginal clindamycin (5/140). At Month 6, BV

recurrence was comparable between the three arms (37/133,

36/135, and 42/140, respectively)

van de Wijgert

et al.10Ecologic Femi+ (B. bifidum

W28 + L. acidophilus

W70 + L. helveticus

W74 + L. brevis

W63 + L. plantarum

W21 + L. salivarius W24)

RCT: no-intervention and

oral metronidazole

controls

Intermittent Ecologic Femi+ use after oral metronidazole BV

treatment resulted in significantly lower BV (Nugent 7–10)

recurrence than no-intervention (incidence 3.58/PY versus

10.18/PY, respectively), and significantly reduced BV-

anaerobes concentration by sequencing, during the 2 months

of use, but the effect disappeared by 4 months after

cessation of use

Van de Wijgert

et al.10Gynophilus LP (L. rhamnosus

Lcr35 regenerans)

RCT: no-intervention and

oral metronidazole

controls

Intermittent Gynophilus LP use after oral metronidazole BV

treatment resulted in a non-significant lower BV (Nugent 7–

10) recurrence than controls (incidence 5.36/PY), and non-

significantly reduced BV-anaerobe concentration by

sequencing, during the 2 months of use, but the effect

disappeared by 4 months after cessation of use

Articles with a high overall risk of bias

Eriksson et al.22 L. gasseri + L. casei var

rhamnosus + L. fermentum

(impregnated tampons)

RCT: placebo control Use of lactobacilli-impregnated tampons for 5 or more days

during the first and second menses after clindamycin

treatment was not efficacious in reducing BV recurrence (by

Nugent 4–10) by the end of the two menstrual cycles (41/91)

compared with placebo tampons (34/96)

Anukam et al.23 RC-14/GR-1 (L. fermentum

RC-14 + L. rhamnosus GR-

1)

RCT: metronidazole gel

control

5 days of GR-1 + RC-14 capsules (without any antibiotic) was

significantly more efficacious in curing BV (12/20 Nugent 0–

3) than 5 days of metronidazole gel (6/20) when assessed

directly after the intervention, and also in reducing BV

recurrence (Nugent 7–10) up to 25 days post-intervention (2/

17 versus 9/18, respectively)

Hemalatha et al.24 Florisia (L. brevis

CD2 + L. salivarius subsp.

salicinius + L. plantarum)

RCT: pH-lowering tablet

control

8 days of Florisia (without any antibiotic) was not efficacious

in curing BV (by Nugent; 36/75 cured) compared with pH-

lowering tablet (29/73 cured)

4 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

respectively), and four on BV cure and recurrence

(Table 1). Most studies showed beneficial effects of vaginal

probiotic interventions for both BV cure and recurrence,

and the distribution of results did not differ significantly

between studies judged to be of medium or high risk. Of

the seven randomised controlled trials that evaluated BV

cure, four showed a significant increase,19,20,23,25 one a

non-significant increase,26 and two no increase in BV cure

fraction compared with controls when assessed immediately

to 5 weeks after cessation of probiotic use.18,24 One of the

two studies with negative BV cure results went on to show

significantly reduced BV recurrence after continued use

during the next four menstrual cycles,18 and the other one

used a pH-lowering tablet as the control product (the cure

fractions were <50% in both groups).24 The two pre-post

intervention studies enrolled women with a range of base-

line Nugent scores and showed significantly higher propor-

tions of women with Nugent 0–3 immediately after

cessation of probiotic use, which lasted up to 21–28 days

post-treatment.27,29 Of the nine BV recurrence results in

eight randomised controlled trials, six were significant

reductions,10,18,20,23,25,28 two non-significant reductions,10,21

and one no reduction.22 The one study that showed no

reduction in BV recurrence included women who used the

vaginal probiotic during menses only.22 All except one of

the other studies avoided probiotic use during menses, and

Table 1. (Continued)

References Probiotic Study design Results

Ling et al.25 L. delbrueckii subsp. lactis

DM8909

RCT: metronidazole

control

7 days of unnamed L. delbrueckii-containing probiotic

(without any antibiotic) was equally effective as vaginal

metronidazole gel in curing BV (assessed 5 days after

treatment completion; 22/25 and 25/30 cured, respectively)

and was associated with a significantly lower recurrence

30 days after treatment completion than metronidazole gel

(0/25 versus 5/30 recurrences). The molecular VMB data also

suggest benefits of using the probiotic

Bisanz et al.26 RC-14/GR-1 (L. reuteri RC-

14 + L. rhamnosus GR-1)

Randomised controlled

cross-over trial: placebo

control

In women with Nugent 4–6, 3 days of GR-1/RC-14 use

(without any antibiotic) did not result in more women with

Nugent score improvement compared with placebo use (2/

10 and 0/10 improved 1 day after use, and 2/9 and 1/8

improved 8 days after use). However, the relative

abundance of Lactobacillus was significantly increased, and

the relative abundance of Atopobium significantly

decreased, 1 day after use

Verdenelli et al.27 SYNBIO gin (L. rhamnosus

IMC 501 + L. paracasei IMC

502)

Pre-/post-intervention

study

In women with Nugent 0–6, 7 days of SYNBIO gin (without

any antibiotic) resulted in a significantly higher proportion of

women with Nugent 0–3 immmediately after therapy (28/35

versus 21/35 at baseline), and this persisted for 21 days after

cessation of therapy (28/35 still had Nugent 0–3; insufficient

data for McNemar testing). SYNBIO gin use also resulted in a

significant increase in Lactobacillus concentration (P < 0.01)

Bohbot et al.28 Physioflor (L. crispatus IP

174178)

RCT: placebo control Physioflor for 14 days immediately after metronidazole

therapy, plus 14 days in three subsequent menstrual cycles,

resulted in significantly lower BV cumulative incidence

(Nugent 7–10), and time to BV recurrence, compared with

placebo (8/39 and 16/39 had at least one recurrence,

respectively). However, the effects disappeared 84 days after

product cessation (P = 0.922, absolute numbers not given)

Rapisarda et al.29 L. acidophilus LA14 Pre-/post-intervention

study

In women with Nugent 4–10, 14 days of L. acidophilus LA14

use (without any antibiotic) resulted in 46/60 women cured

(Nugent 0–3) at the end of therapy, and this persisted for

28 days after product cessation (50/60)

BV, bacterial vaginosis; PY, person-years at risk; RCT, randomised controlled trial; VMB, vaginal microbiota.

Main results of studies with BV cure and/or recurrence (by Nugent or Ison-Hay scoring of Gram stains) and/or molecular VMB composition

endpoints: 14 results in 13 studies.

5ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics

Table 2. Summary of VVC study results

References Probiotic Study design Results

Articles with a medium overall risk of bias

Czaja et al.30 L. crispatus CTV-05 (not

Lactin-V)

RCT: placebo control In women with a history of recurrent UTI but without

current dysuria, 5 days of L. crispatus CTV-05 use resulted

in a similar VVC cumulative incidence (4/15) compared with

placebo use (2/15) up to 25 days after cessation of use

Articles with a high overall risk of bias€Ozmen et al.31 Gynoflor (L. acidophilus

KS400 + 0.03 mg estriol)

Pre-/post-intervention

study in three

separate, non-

randomised groups

12 days of Gynoflor with or without metronidazole,

compared with metronidazole alone (to treat BV), resulted

in significantly lower cumulative VVC incidence (3/96 and

2/97, versus 14/114, respectively) in the 10–13 days after

cessation of use

Pirotta et al.32 Femilac (L. rhamnosus +

L. delbrueckki +

L. acidophilus +

S. thermophilus)

2 9 2 factorial design

RCT: placebo control

10 days of Femilac during (6 days) and after (4 days) non-

metronidazole antibiotic use, compared with placebo

controls, was not effective in reducing VVC incidence

within 18 days after cessation of Femilac use (17/59 versus

9/54, respectively)

Di Pierro et al.33 Kramegin (L. acidophilus +

Krameria triandra plant

extract + 15 mg lactic acid)

Pre-/post-intervention

study

10 days of Kramegin (without antifungal) cured 75/75

women with acute VVC, and 20/30 women with recurrent

VVC (both significant compared with baseline). Cure was

assessed 7 days after last administration

Witt et al.34 L. gasseri RCT: probiotic +

itraconazole versus

itraconazole only

In women with acute VVC and a history of recurrent VVC,

after induction with itraconazole twice weekly for 1 month,

itraconazole (once per month) with L. gasseri (6

consecutive days per month) for 6 months resulted in a

similar VVC recurrence rate as itraconazole (once per

month) only (15/45 versus 12/31, respectively). Results

were also similar 6 months after product cessation (6/25

versus 5/23, respectively)

Vicariotto et al.35 ActiCand (L. fermentum

LF10 + L. acidophilus LA02)

Pre-/post-intervention

study

In women with acute VVC and a history of recurrent VVC,

ActiCand (without antifungal; 7 consecutive nights,

followed by 3 nights per week for 3 weeks, and one night

per week for 4 weeks) resulted in a significant reduction in

VVC cases to 7/30 women after cessation of therapy at

Day 56

De Seta et al.36 Gyno-Canesflor (L. plantarum

P17630)

Non-randomised

prospective cohort

study

In women with acute VVC but no history of recurrent VVC,

34 days of Gyno-Canesflor after clotrimazole treatment,

compared with placebo use, resulted in a non-significant

(Fisher’s P = 0.095) lower cumulative VVC incidence (1/40

versus 5/40, respectively)

Donders et al.37 &

Donders et al.38Gynoflor (L. acidophilus

KS400 + 0.03 mg estriol)

Pre-/post-intervention

study

An 84-day single-arm Gynoflor treatment scheme (once

daily for 4 weeks followed by thrice weekly for 8 weeks)

for atrophic vaginitis in postmenopausal women with

breast cancer temporarily increased asymptomatic VVC

prevalence after 2 weeks of use (from 2/14 to 7/16) but

returned to baseline values after 1 (3/16), 2 (3/16), and 3

(3/13) months of use

Pendharkar- Trial II

et al.39EcoVag (L. gasseri DSM

14869 + L. rhamnosus DSM

14870)

Non-randomised

prospective cohort

study

In women with recurrent VVC, fluconazole (7 days in cycle

1 followed by once weekly in cycles 2 + 3, and biweekly in

cycles 4–6) with EcoVag (10 days in cycle 2, and once

weekly in cycles 2–6) did not significantly improve the VVC

cure proportion 6 months after treatment cessation

compared with the fluconazole regimen without EcoVag

(8/9 versus 7/10, respectively) but almost all women in

both groups were cured

6 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

in the one study that did not, women used probiotics inbe-

tween and during menses.10 The two medium risk ran-

domised controlled trials that compared vaginal probiotic

use after initial antibiotic treatment with placebo or no-in-

tervention controls as well as prolonged antibiotic use con-

trols both showed that prolonged antibiotic use was more

efficacious in reducing BV recurrence than was probiotic

use.10,21

VVC outcomesOf the 12 studies with VVC as outcome, one was judged

to have medium and 11 high overall risk of bias (Fig-

ure 2). Six of 12 studies were pre-post intervention stud-

ies,29,31,33,35,37,38,41 two were non-randomised cohort

studies with parallel comparison groups (one with a pla-

cebo group36 and one comparing fluconazole plus probi-

otic with fluconazole only39), and four were randomised

controlled trials (two with a placebo group,30,32 one with

a ‘no intervention’ group,40 and one comparing itracona-

zole plus probiotic to itraconazole only34) (Table 2). The

one medium risk study was a randomised placebo-con-

trolled trial.30

Four of the 12 probiotic studies with a VVC outcome

reported on VVC cure (all without any antifungal use),

eight on VVC recurrence or incidence (2/8 with concurrent

antifungal use, 2/8 after antifungal use, 2/8 after antibiotic

treatment for another indication and no antifungal use,

and 2/8 without any antifungal or antibiotic use), and two

on both (Table 2). The four studies that reported on VVC

cure were pre-post intervention studies in women who

received probiotic treatment in the absence of antifungal

treatment.29,33,35,41 They reported cure fractions ranging

from 57 to 100% after therapy durations ranging from

6 days to 2 months, but VVC recurrence beyond 1 month

after treatment cessation was not assessed. The four studies

that assessed VVC recurrence after combined probiotic/an-

tifungal treatment showed mixed results: two studies that

compared antifungal plus probiotic use with antifungal use

alone showed no difference in VVC recurrence between the

groups,34,39 and two studies that compared probiotic after

antifungal use with either placebo after antifungal use36 or

antifungal use alone40 showed significantly lower VVC

recurrence. One of the two studies that assessed VVC inci-

dence after antibiotic use for a non-VVC indication showed

reduced VVC incidence31 and the other one did not.32

Finally, the two studies in women who used a vaginal pro-

biotic in the absence of antifungal or antibiotic therapy

showed similar VVC incidence in probiotic and placebo

users,30 or no difference in proportions of women with

VVC before and after probiotic use.37,38

Table 2. (Continued)

References Probiotic Study design Results

Kovachev et al.40 Lactagyn (L. acidophilus,

L. rhamnosus,

S. thermophilus,

L. delbrueckii subsp.

bulgaricus)

RCT: no-intervention

control

In women with acute VVC, antifungal treatment

(fluconazole and fenticonazole) followed by 10-day

Lactagyn therapy resulted in lower VVC recurrence (10/

209) at the final follow-up visit 25–30 days after product

cessation compared with antifungal treatment alone (76/

207). The authors report this as two non-significant pre/

post results, but these are significant when the appropriate

test (McNemar) is used, and when the two groups are

compared directly

Rapisarda et al.29 L. acidophilus LA14 Pre-/post-intervention

study

In women with Nugent 4–10 and 21/60 with acute VVC,

14 days of L. acidophilus LA14 without antifungals or

antibiotics resulted in a significant reduction of VVC cases

to 9/60, 1 day after cessation of use, and 6/60, 4 weeks

after cessation of use. Mean Candida culture counts also

decreased significantly

Murina et al.41 Estromineral Probiogel

(L. fermentum

LF10 + L. plantarum LP02)

Pre-/post-intervention

study

Estromineral Probiogel (without antifungal) cured 51/82

women with acute VVC and 27/27 in women with

recurrent VVC 20–30 days after therapy initiation (both

significant compared with baseline), but therapy longer

than the intended 6 days was required in 57.3% and

63.0% women, respectively, due to persistence of

symptoms

BV, bacterial vaginosis; UTI, urinary tract infection; VMB, vaginal microbiome; VVC, vulvovaginal candidiasis.

Main results of the studies with VVC cure and/or incidence (by culture or wet mount) endpoints. None of the studies used Candida PCR: 12

results in 12 studies.

7ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics

A B

Figure 2. Risk of bias assessments. ‘Other sources of bias’ included whether confounders were considered (in the eligibility criteria or statistical

analyses), intent-to-treat results were reported, and statistical testing was appropriate. (A) Authors’ judgements about each potential bias for each

study, ordered by publication year. Studies with high overall risk of bias (seventh column) are indicated with a minus symbol and studies with

medium overall risk of bias with a plus symbol. No studies were judged to have a low overall risk of bias. (B) Summary of each bias across all 34

studies.

8 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

Vaginal detection outcomesFourteen studies assessed vaginal detection of probiotic

strains (with one study reporting results for two probiotics)

using a variety of molecular techniques on culture isolates

(10/14) or on DNA extracted from vaginal swabs (4/14;

Table S1). Probiotic strains were detected in 56–100% of

probiotic users, or 20–39% of collected swabs, during or

directly after the intervention period. The timing between

last probiotic insertion and vaginal sample collection for

detection measurement was never reported and was clearest

in a study by Dausset et al.50 that evaluated two, three, or

five times per week dosing and assessed probiotic strain

concentrations by quantitative PCR every day. This study

showed rapid probiotic strain concentration increases and

decreases after each administration. The 11 studies that

assessed probiotic strain detection more than a week after

cessation of use showed that detection always decreased

within weeks; mean probiotic strain survival in the vagina

cannot be determined more precisely because most studies

included only one post-use assessment ranging from

14 days to 6 months after cessation.10,26,27,39,42,44–46,49 None

of the three studies that assessed the association between

vaginal detection and self-reported adherence found such

an association,10,39,47 but one study showed higher probi-

otic strain concentrations in women who ever had any

strain detected during the intervention period compared to

women who did not.10 This sequencing study also high-

lighted the high interindividual variability in vaginal probi-

otic strain detection and concentration, as well as overall

VMB composition. Two L. crispatus CTV-05 studies

reported that the probiotic strain was less often detected in

women who had had sexual intercourse during the product

use period (with or without condoms),45,48 and four stud-

ies found that detection was less likely in women who had

autologous lactobacilli prior to probiotic use (L. crispatus

in the case of three L. crispatus CTV-05 studies and any

lactobacilli in the case of an EcoVag study).39,43,45,48

Quality assessmentWe judged overall risk of bias to be high for 26 studies,

medium for eight studies, and low for no studies (Fig-

ure 2). The majority of efficacy studies were likely under-

powered and the majority of descriptive vaginal detection

studies had low precision, with 9/14 studies assessing fewer

than 20 women per probiotic. None of the studies com-

pared responders with non-responders in a comprehensive

manner, partly due to sample size limitations. Only 12

studies reported adherence data, and only two of those

included adherence measures other than self-report.10,47

The exact timing between last probiotic insertion and vagi-

nal sample collection for outcome assessment was never

reported. Most BV/molecular VMB studies (11/13), but

only 4/12 VVC studies, were randomised controlled trials

and participants and clinicians in 16/24 trials were not

blinded. Outcome ascertainment rigour and quantification

improved over time with the increased availability of

molecular technologies (Figure 2). However, microbiologi-

cal inclusion criteria often differed from microbiological

outcome assessments, and it was usually unclear whether

these assessments were blinded. Insufficient consideration

of potential confounding factors, such as hormonal contra-

ception use, pregnancy, menses, vaginal practices, sexual

behaviour, any antimicrobial use that was not part of the

study design, and presence of sexually transmitted infec-

tions, was also common.

Discussion

Main findings and interpretationsThe systematic review results suggest that lactobacilli-con-

taining vaginally applied probiotics hold promise for BV

cure and prevention, but much less so for VVC cure and

prevention. Four of the six medium risk randomised con-

trolled trials with BV endpoints showed significant benefits,

and the other two showed non-significant benefits. No

major safety concerns were reported in any of the studies.

In addition, we found consistent evidence that vaginal

detection of probiotic strains does not last long beyond the

dosing period, suggesting that none of the evaluated probi-

otic strains colonised the vagina. However, these findings

are not definitive because heterogeneity was high and the

quality of most studies suboptimal. Furthermore, publica-

tion bias is likely to be high, although we did not formally

assess this due to high heterogeneity.

We opted to evaluate all vaginally applied probiotics for

bacterial and fungal vaginal dysbiosis indications because

no single product-indication combination had been suffi-

ciently evaluated to allow for a meaningful synthesis of

results. Although all identified probiotics contained lacto-

bacilli, it is not yet clear which Lactobacillus strains are the

most promising, if any. The probiotics included in this

review therefore varied in terms of Lactobacillus strains,

other active ingredients, excipients, application methods,

and doses. We could not identify any one product that

seemed to outperform the others in any of the clinical indi-

cations or vaginal detection. Most of the probiotic strains

originated from the gut or from traditional fermented

foods, and have been marketed for decades by non-phar-

maceutical companies, even before clinical evidence became

available.6 In recent years, the regulatory landscape for

vaginal probiotics has changed considerably. When health

claims are made, the US Food and Drug Administration

has required human drug approval since 2016, and the

European Medicines Authority will follow suit from 1 May

2020. As far as we know, only one of the products included

in this review is being developed as a drug for human use:

9ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics

Lactin-V. It contains a L. crispatus strain (CTV-05) that

was isolated from a healthy woman.43 While the results of

the five vaginal detection and safety trials conducted with

L. crispatus CTV-05 to date warranted continued develop-

ment,30,43,45,47,48,51 we did not find early indications that

this strain is likely to outperform other Lactobacillus

strains. A Lactin-V efficacy trial for BV recurrence is cur-

rently ongoing.52

The mixed results of trials with VVC indications did not

surprise us given that molecular studies have shown that

Candida species often co-exist with lactobacilli in the VMB,

and epidemiological studies have shown negative associa-

tions between BV and VVC, including increased VVC inci-

dence after BV treatment.5 In fact, some BV intervention

trials assessed VVC incidence as a safety endpoint.10,53

While the use of exogenous lactobacilli to cure or prevent

BV is much more in line with the molecular and epidemio-

logical evidence to date, it is not clear what the best dosing

strategies would be: as a main treatment in the absence of

antibiotic treatment, as an adjuvant treatment to antibiotic

treatment, as a maintenance therapy to prevent incident or

recurrent BV, or a combination of these. This review could

not provide definitive answers to these questions because

none of the studies directly compared different treatment

strategies with the same probiotic. However, the adjuvant

and maintenance therapy studies were of higher quality,

and more convincing, than the studies that used a vaginal

probiotic as a stand-alone BV treatment. Only two studies

directly compared antibiotic plus probiotic use with pro-

longed antibiotic use, and they both found that the latter

Table 3. Recommendations for vaginal probiotic efficacy evaluation

Topic Recommendations

Study design,

populations,

sample size

Randomised placebo-controlled blinded trials are the preferred study design. Avoid pre-post studies without any parallel control

groups

Define the target population and recruitment strategy a priori. Avoid convenience samples

Evaluate in women with and without urogenital symptoms, and in women at various levels of vaginal dysbiosis risk

Conduct sample size calculations. Take into account that vaginal probiotic efficacy tends to have high interindividual

variability and that efficacy analyses should be controlled for known confounders such as hormonal status, menses,

sexual behaviour, and presence of sexually transmitted infections. Allow for comparisons between responders

and non-responders

Interventions Consider dose per insertion, frequency of insertion, timing of insertion (also in relation to timing of sample collection for efficacy

evaluation), and how menses will be taken into account

Provide clear use instructions, also in relation to vaginal hygiene practices and menses, and assess whether the participant

understood them. Assess the participant’s ability to insert the probiotic correctly by direct observation

Consider frequent sampling for outcome assessments. Participants could self-sample at home

Assess adherence throughout the study, ideally not just by self-report

Controls A parallel placebo control group is best but if not possible, use no-intervention controls

In addition, consider a positive control group of women using oral or vaginal metronidazole/clindamycin for BV or an antifungal

for VVC

Blind the laboratory technicians who conduct the outcome assessments, especially when blinding of the clinicians and

participants

is not possible

Outcomes Assess outcomes prior to, during, and after cessation of probiotic use in all women at all time points regardless of

symptomatology. Use the same outcome assessments to determine eligibility and/or cure of the initial BV/VVC prior to

initiation of probiotic use (if applicable), at the start of probiotic use, and throughout follow-up

Use (semi-)quantitative molecular methods to differentiate between probiotic lactobacilli, autologous lactobacilli, BV anaerobes,

Candida species, and other organisms of interest

In addition, use Nugent scoring and Amsel criteria to allow for comparisons with older studies and with current clinical practice

Consider and record the timing between last probiotic insertion and sample collection for efficacy assessment

Consider the vaginal microbiota holistically, including bacteria, yeasts, and sexually transmitted pathogens

Always assess and report safety outcomes (adverse events)

Results

reporting

Report the name and manufacturer of the vaginal probiotic that was evaluated, as well as all active ingredients, excipients,

and dosing information. Explain the rationale for selecting the product and be transparent about competing interests

Report clearly how the probiotic was meant to be used: as a main therapy to cure BV or VVC, as an adjuvant therapy to

an antimicrobial therapy or as a maintenance therapy to prevent incidence or recurrence

Report intent-to-treat and per-protocol statistical analysis results

BV, bacterial vaginosis; VMB, vaginal microbiota; VVC, vulvovaginal candidiasis.

10 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

was more efficacious in reducing BV recurrence. This

would not be a reason to discontinue vaginal probiotic

development because most would agree that prolonged or

frequent antibiotic use is not desirable given the risks of

side effects and antimicrobial resistance.

None of the articles explained why specific doses (about

108 colony-forming units per insertion for most products)

and dosing frequency and duration were chosen. We sus-

pect that many of these depended on marketing approvals

dating from before the American and European drug regu-

latory approval restrictions (as was the case in our study),10

and are not necessarily based on clinical evidence. Although

probiotic strains may not have to be present in high con-

centrations to exert beneficial effects, the vaginal detection

findings of this review beg the question whether doses and/

or dosing frequency/duration should be increased. Some

studies suggested that baseline VMB composition and sex-

ual behaviour during probiotic use may influence

efficacy,10,39,43,45,48 suggesting that stratified, or even per-

sonalised, VMB modulation may be required in the most

difficult cases of persistent or recurrent BV.

Strengths and limitationsThe quality of most studies was suboptimal. Some of the tra-

ditional epidemiological biases (such as insufficient statistical

power, and lack of randomisation, blinding, appropriate con-

trols, and adjustment for confounding) could have been

avoided, but others are unique to vaginal probiotic trials and

are methodologically challenging. The latter tended to

improve over time with the increasing availability of molecu-

lar methods. An important limitation of all studies was that

the time duration between last probiotic insertion and vaginal

sample collection for endpoint assessment and adherence was

not known with sufficient precision, and was not taken into

account in analyses. This is particularly problematic in the

context of BV endpoints using Nugent or Ison-Hay scoring,

because those scores depend heavily on counting bacteria with

a Lactobacillus morphotype (Gram-positive rod) under a

microscope. The scoring cannot differentiate between probi-

otic lactobacilli and autologous lactobacilli, and is prone to

inflated counts if the sample is taken soon after insertion.

Only the two most recent studies managed to address this

problem partially. Dausset et al. used daily sampling in the

context of intermittent dosing, as well as quantitative PCR to

differentiate between probiotic and autologous Lactobacillus

strains.10,50 Our study employed 16S rRNA gene quantifica-

tion and sequencing, which enabled us to quantify not only

probiotic and autologous Lactobacillus strains, but also all

BV-associated anaerobes combined.10 Reduction of the overall

BV-anaerobe concentration over time, in addition to the

increased lactobacilli concentration over time, provided con-

vincing evidence for the beneficial effects of intermittent pro-

biotic use. Our study also showed that host responses to

vaginal probiotic treatment are highly variable, which calls

into question the appropriateness of cross-sectional and pre-

post assessments, as well as the use of group means/medians.

Moving forward, we recommend the following in future

trials: incorporation of quantitative molecular methods that

differentiate between probiotic and autologous Lactobacillus

strains; regular assessment of efficacy and safety endpoints

before, during, and after cessation of probiotic use; stan-

dardisation and optimisation of time duration between

insertion and sample collection; improvement of statistical

power to allow for direct comparisons between responders

and non-responders; taking all known confounders and

adherence into account; and avoiding generic epidemiologi-

cal biases as much as possible (Table 3). Ideally, future tri-

als would also quantify BV-anaerobes, and assess the

impact on biofilm formation, Candida species, and bacte-

rial pathobionts other than BV-anaerobes.

Conclusion

Vaginal probiotics hold promise for BV cure and preven-

tion. We expect that the availability of vaginal probiotics

for these indications will decline in 2020 because of regula-

tory changes. We therefore urge the field to invest in clini-

cal evidence-based product development, and to conduct

future clinical trials more rigorously.

Disclosure of interestsThe authors conducted a clinical trial with Ecologic Femi+(Winclove Probiotics, Amsterdam, Netherlands) and Gyno-

philus LP (Biose, Aurillac, France). This trial was funded

by the UK Medical Research Council and the University of

Liverpool, but the two companies donated their products

for use in the trial free of charge. The authors have no

financial or intellectual investments in these products and

do not report any other competing interests. Completed

disclosure of interests forms are available to view online as

supporting information.

Contribution to authorshipJvdW conceived and planned the systematic review, and

JvdW and MCV performed the review and analyses, and

wrote the manuscript together.

Details of ethics approvalNot applicable.

FundingNone.

AcknowledgementsThe authors thank medical student Connie Rees for assis-

tance with the search and data extraction.

11ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics

Supporting Information

Additional supporting information may be found online in

the Supporting Information section at the end of the arti-

cle.

Table S1. Summary of vaginal probiotic strain detection

results.

Table S2. Product characteristics.

Appendix S1. Search terms and filters.&

References

1 van de Wijgert JHHM, Borgdorff H, Verhelst R, Crucitti T, Francis S,

Verstraelen H, et al. The vaginal microbiota: what have we learned

after a decade of molecular characterization? PLoS One 2014;9:

e105998.

2 van de Wijgert JHHM, Jespers V. The global health impact of vaginal

dysbiosis. Res Microbiol 2017;168:859–64.3 Tachedjian G, Aldunate M, Bradshaw CS, Cone RA. The role of

lactic acid production by probiotic Lactobacillus species in vaginal

health. Res Microbiol 2017;168:782–92.4 Muzny CA, Schwebke JR. Biofilms: an underappreciated mechanism

of treatment failure and recurrence in vaginal infections. Clin Infect

Dis 2015;61:601–6.5 van de Wijgert JHHM. The vaginal microbiome and sexually

transmitted infections are interlinked: consequences for treatment

and prevention. PLoS Med 2017;14:e1002478.

6 O’Toole PW, Marchesi JR, Hill C. Next-generation probiotics: the

spectrum from probiotics to live biotherapeutics. Nat Microbiol

2017;2:17057.

7 Bradshaw CS, Brotman RM. Making inroads into improving

treatment of bacterial vaginosis—striving for long-term cure. BMC

Infect Dis 2015;15:292.

8 Petrova MI, Imholz NCE, Verhoeven TLA, Balzarini J, Van Damme

EJM, Schols D, et al. Lectin-like molecules of Lactobacillus

rhamnosus GG inhibit pathogenic Escherichia coli and Salmonella

biofilm formation. PLoS One 2016;11:e0161337.

9 Allonsius CN, Vandenheuvel D, Oerlemans EFM, Petrova MI,

Donders GGG, Cos P, et al. Inhibition of Candida albicans

morphogenesis by chitinase from Lactobacillus rhamnosus GG. Sci

Rep 2019;9:2900.

10 van de Wijgert JHHM, Verwijs MC, Agaba SK, Bronowski C,

Mwambarangwe L, Uwineza M, et al. Intermittent lactobacilli-

containing vaginal probiotic or oral metronidazole use to prevent

bacterial vaginosis recurrence: safety and preliminary efficacy by

microscopy and sequencing. MedRxiv 2019. https://doi.org/10.1101/

19001156

11 Moher D, Liberati A, Tetzlaff J, Altman DG, Group TP. Preferred

reporting items for systematic reviews and meta-analyses: the

PRISMA statement. PLoS Med 2009;6:e1000097.

12 Prospero—International prospective register of systematic reviews

[Internet]. [https://www.crd.york.ac.uk/PROSPERO/display_record.

php?RecordID=75717]. Accessed 30 July 2018.

13 PubMed - NCBI [Internet]. [https://www.ncbi.nlm.nih.gov/pubmed/].

Accessed 17 July 2018.

14 Embase | Elsevier [Internet]. [https://www.elsevier.com/solutions/

embase-biomedical-research]. Accessed 17 July 2018.

15 Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial

vaginosis is improved by a standardized method of gram stain

interpretation. J Clin Microbiol 1991;29:297–301.16 Ison C, Hay P. Validation of a simplified grading of Gram stained

vaginal smears for use in genitourinary medicine clinics. Sex Transm

Infect 2002;78:413–5.17 Cochrane Handbook for Systematic Reviews of Interventions version

5.1.0[Internet]. The Cochrane Collaboration; 2011. [http://handb

ook-5-1.cochrane.org/]. Accessed 11 July 2018.

18 Larsson P-G, Stray-Pedersen B, Ryttig KR, Larsen S. Human

lactobacilli as supplementation of clindamycin to patients with

bacterial vaginosis reduce the recurrence rate; a 6-month, double-

blind, randomized, placebo-controlled study. BMC Women’s Health

2008;8:3. https://doi.org/10.1186/1472-6874-8-3

19 Petricevic L, Witt A. The role of Lactobacillus casei rhamnosus Lcr35

in restoring the normal vaginal flora after antibiotic treatment of

bacterial vaginosis. BJOG 2008;115:1369–74.20 Mastromarino P, Macchia S, Meggiorini L, Trinchieri V, Mosca L,

Perluigi M, et al. Effectiveness of Lactobacillus-containing vaginal

tablets in the treatment of symptomatic bacterial vaginosis. Clin

Microbiol Infect 2009;15:67–74.21 Bradshaw CS, Pirotta M, De Guingand D, Hocking JS, Morton AN,

Garland SM, et al. Efficacy of oral metronidazole with vaginal

clindamycin or vaginal probiotic for bacterial vaginosis: randomised

placebo-controlled double-blind trial. PLoS One 2012;7:e34540.

22 Eriksson K, Carlsson B, Forsum U, Larsson P-G. A double-blind

treatment study of bacterial vaginosis with normal vaginal

lactobacilli after an open treatment with vaginal clindamycin ovules.

Acta Derm-Venereol 2005;85:42–6.23 Anukam KC, 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 Infect 2006;8:2772–6.24 Hemalatha R, Mastromarino P, Ramalaxmi BA, Balakrishna NV,

Sesikeran B. Effectiveness of vaginal tablets containing lactobacilli

versus pH tablets on vaginal health and inflammatory cytokines: a

randomized, double-blind study. Eur J Clin Microbiol Infect Dis

2012;31:3097–105.25 Ling Z, Liu X, Chen W, Luo Y, Yuan L, Xia Y, et al. The restoration

of the vaginal microbiota after treatment for bacterial vaginosis with

metronidazole or probiotics. Microb Ecol 2013;65:773–80.26 Bisanz JE, Seney S, McMillan A, Vongsa R, Koenig D, Wong L, et al.

A systems biology approach investigating the effect of probiotics on

the vaginal microbiome and host responses in a double blind,

placebo-controlled clinical trial of post-menopausal women. PLoS

One 2014;9:e104511.

27 Verdenelli MC, Cecchini C, Coman MM, Silvi S, Orpianesi C, Coata

G, et al. Impact of probiotic SYNBIO� administered by vaginal

suppositories in promoting vaginal health of apparently healthy

women. Curr Microbiol 2016;73:483–90.28 Bohbot JM, Dara€ı E, Bretelle F, Brami G, Daniel C, Cardot JM.

Efficacy and safety of vaginally administered lyophilized Lactobacillus

crispatus IP 174178 in the prevention of bacterial vaginosis

recurrence. J Gynecol Obstet Hum Reprod 2018;47:81–6.29 Rapisarda AMC, Caldaci L, Valenti G, Brescia R, Sapia F, Sarpietro G,

et al. Efficacy of vaginal preparation containing Lactobacillus

acidophilus, lactic acid and deodorized garlic extract in treatment

and prevention of symptomatic bacterial vaginitis: result from a

single-arm pilot study. Ital J Gynaecol Obstet 2018;30:21–31.30 Czaja CA, Stapleton AE, Yarova-Yarovaya Y, Stamm WE. Phase I

trial of a Lactobacillus crispatus vaginal suppository for prevention of

recurrent urinary tract infection in women. Infect Dis Obstet Gynecol

2007;2007:35387. https://doi.org/10.1155/2007/35387

12 ª 2019 Royal College of Obstetricians and Gynaecologists

Wijgert, Verwijs

31 €Ozmen S, Turhan NO, Seckin NC. Gardnerella-associated vaginitis:

comparison of three treatment modalities. Turk J Med Sci

1998;28:171–3.32 Pirotta M, Gunn J, Chondros P, Grover S, O’Malley P, Hurley S,

et al. Effect of lactobacillus in preventing post-antibiotic vulvovaginal

candidiasis: a randomised controlled trial. BMJ 2004;329:548.

33 Di Pierro F, Catacchio V, Candidi C, Zerbinati N, Alfonso R. Rhatany-

based preparation in vulvovaginitis and vaginosis. Gazz Med Ital

2009;168:338–46.34 Witt A, Kaufmann U, Bitschnau M, Tempfer C, €Ozbal A, Haytouglu

E, et al. Monthly itraconazole versus classic homeopathy for the

treatment of recurrent vulvovaginal candidiasis: a randomised trial:

Monthly itraconazole versus classic homeopathy for treatment of

RVVC. BJOG 2009;116:1499–505.35 Vicariotto F, Del Piano M, Mogna L, Mogna G. Effectiveness of the

association of 2 probiotic strains formulated in a slow release

vaginal product, in women affected by vulvovaginal candidiasis: a

pilot study. J Clin Gastroenterol 2012;46:S73–80.36 De Seta F, Parazzini F, De Leo R, Banco R, Maso GP, De Santo D,

et al. Lactobacillus plantarum P17630 for preventing Candida

vaginitis recurrence: a retrospective comparative study. Eur J Obstet

Gynecol Reprod Biol 2014;182:136–9.37 Donders G, Neven P, Moegele M, Lintermans A, Bellen G,

Prasauskas V, et al. Ultra-low-dose estriol and Lactobacillus

acidophilus vaginal tablets (Gynoflor�) for vaginal atrophy in

postmenopausal breast cancer patients on aromatase inhibitors:

pharmacokinetic, safety, and efficacy phase I clinical study. Breast

Cancer Res Treat 2014;145:371–9.38 Donders G, Bellen G, Neven P, Grob P, Prasauskas V, Buchholz S,

et al. Effect of ultra-low-dose estriol and lactobacilli vaginal tablets

(Gynoflor�) on inflammatory and infectious markers of the vaginal

ecosystem in postmenopausal women with breast cancer on

aromatase inhibitors. Eur J Clin Microbiol Infect Dis 2015;34:2023–8.

39 Pendharkar S, Brandsborg E, Hammarstr€om L, Marcotte H, Larsson

P-G. Vaginal colonisation by probiotic lactobacilli and clinical

outcome in women conventionally treated for bacterial vaginosis

and yeast infection. BMC Infect Dis 2015;15:255.

40 Kovachev SM, Vatcheva-Dobrevska RS. Local probiotic therapy for

vaginal Candida albicans infections. Probiot Antimicro Prot

2015;7:38–44.41 Murina F, Vicariotto F, Di Francesco S. Thymol, eugenol and

lactobacilli in a medical device for the treatment of bacterial

vaginosis and vulvovaginal candidiasis. New Microbiol 2018;41:220–4.

42 Gardiner GE, Heinemann C, Bruce AW, Beuerman D, Reid G.

Persistence of Lactobacillus fermentum RC-14 and Lactobacillus

rhamnosus GR-1 but not L. rhamnosus GG in the human vagina as

demonstrated by randomly amplified polymorphic DNA. Clin Vaccine

Immunol 2002;9:92–6.

43 Antonio MAD, Hillier SL. DNA fingerprinting of Lactobacillus

crispatus strain CTV-05 by repetitive element sequence-based PCR

analysis in a pilot study of vaginal colonization. J Clin Microbiol

2003;41:1881–7.44 Burton JP, Cadieux PA, Reid G. Improved understanding of the

bacterial vaginal microbiota of women before and after probiotic

instillation. Appl Environ Microbiol 2003;69:97–101.45 Antonio MAD, Meyn LA, Murray PJ, Busse B, Hillier SL. Vaginal

colonization by probiotic Lactobacillus crispatus CTV-05 is decreased

by sexual activity and endogenous lactobacilli. J Infect Dis

2009;199:1506–13.46 Ehrstr€om S, Daroczy K, Rylander E, Samuelsson C, Johannesson U,

Anz�en B, et al. Lactic acid bacteria colonization and clinical outcome

after probiotic supplementation in conventionally treated bacterial

vaginosis and vulvovaginal candidiasis. Microbes Infect

2010;12:691–9.47 Hemmerling A, Harrison W, Schroeder A, Park J, Korn A, Shiboski S,

et al. Phase 2a study assessing colonization efficiency, safety, and

acceptability of Lactobacillus crispatus CTV-05 in women with

bacterial vaginosis. Sex Transm Dis 2010;37:745–50.48 Ngugi BM, Hemmerling A, Bukusi EA, Kikuvi G, Gikunju J, Shiboski

S, et al. Effects of bacterial vaginosis-associated bacteria and sexual

intercourse on vaginal colonization with the probiotic Lactobacillus

crispatus CTV-05. Sex Transm Dis 2011;38:1020–7.49 Tomusiak A, Strus M, Heczko P, Adamski P, Stefa�nski G,

Mikołajczyk-Cicho�nska A, et al. Efficacy and safety of a vaginal

medicinal product containing three strains of probiotic bacteria: a

multicenter, randomized, double-blind, and placebo-controlled trial.

Drug Design Dev Ther 2015;9:5345–54.50 Dausset C, Patrier S, Gajer P, Thoral C, Lenglet Y, Cardot J-M, et al.

Comparative phase I randomized open-label pilot clinical trial of

Gynophilus� (Lcr regenerans�) immediate release capsules versus

slow release muco-adhesive tablets. Eur J Clin Microbiol Infect Dis

2018;37:1869–80.51 Stapleton AE, Au-Yeung M, Hooton TM, Fredricks DN, Roberts PL,

Czaja CA, et al. Randomized, placebo-controlled Phase 2 trial of a

Lactobacillus crispatus probiotic given intravaginally for prevention

of recurrent urinary tract infection. Clin Infect Dis 2011;52:1212–7.52 LACTIN-V study for recurrent bacterial vaginosis - ClinicalTrials.gov

[Internet]. [https://clinicaltrials.gov/ct2/show/NCT02766023].

Accessed 23 April 2019.

53 McClelland RS, Richardson BA, Hassan WM, Chohan V, Lavreys L,

Mandaliya K, et al. Improvement of vaginal health for Kenyan

women at risk for acquisition of Human Immunodeficiency Virus

type 1: results of a randomized trial. J Infect Dis 2008;15:1361–8.

13ª 2019 Royal College of Obstetricians and Gynaecologists

Systematic review of vaginal probiotics