Journal of Antimicrobial Chemotherapy (2006) 58, 266–272

doi:10.1093/jac/dkl246
Advance Access publication 21 June 2006

Probiotics for prevention of recurrent vulvovaginal candidiasis:

a review

Matthew E. Falagas1,2*, Gregoria I. Betsi1 and Stavros Athanasiou3

1
Alfa Institute of Biomedical Sciences (AIBS), Athens, Greece; 2Department of Medicine,
Tufts University School of Medicine, Boston, MA, USA; 31st Department of Obstetrics and Gynaecology,

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

Athens University School of Medicine, Athens, Greece

Vulvovaginal candidiasis (VVC) is a common infection affecting the quality of life of many women.
Probiotics have been investigated as possible agents for the prevention of recurrences of VVC. We reviewed
the available literature. In some studies the development of VVC was associated with either a low number of
lactobacilli in the vagina or with the presence of H2O2-non-producing vaginal lactobacilli, although there are
a few studies not supporting these statements. In addition, in vitro studies have shown that lactobacilli can
inhibit the growth of Candida albicans and/or its adherence on the vaginal epithelium. The results of some
clinical trials support the effectiveness of lactobacilli, especially Lactobacillus acidophilus, Lactobacillus
rhamnosus GR-1 and Lactobacillus fermentum RC-14, administered either orally or intravaginally in colo-
nizing the vagina and/or preventing the colonization and infection of the vagina by C. albicans, while the
results of a small number of clinical trials do not corroborate these findings. Nevertheless, most of the
relevant clinical trials had methodological problems such as small sample size, no control group (placebo)
and included women without confirmed recurrent VVC, and thus they are not reliable for drawing definitive
conclusions. Thus, the available evidence for the use of probiotics for prevention of recurrent VVC is
limited. However, the empirical use of probiotics may be considered in women with frequent recurrence
of VVC (more than three episodes per year), especially for those who have adverse effects from or con-
traindications for the use of antifungal agents, since adverse effects of probiotics are very rare. In any case
women should be clearly informed about the unproven usefulness of probiotics for this purpose. In
conclusion, despite the promising results of some studies, further research is needed to prove the effective-
ness of probiotics in preventing the recurrences of VVC and to allow their wide use for this indication.

Keywords: candidal vaginitis, yeast vaginitis, fungal infections, lactobacilli, bifidobacteria

Introduction other diseases, such as antibiotic-associated diarrhoea (Saccha-

romyces boulardii), Helicobacter pylori infections, inflammatory
Vulvovaginal candidiasis (VVC) is a common infection among bowel diseases, allergy, cancer, urinary tract infections and
women that is associated with considerable morbidity and health- bacterial vaginosis, is under research.2
care cost. A survey by Foxman et al.1 in the US showed that 6.5% In the present review we tried to compile and summarize the
and 8% of women older than 18 years reported ‡1 and ‡4 epi- existing data regarding the potential use of probiotics for the
sodes of VVC during the 2 months and 1 year prior to the survey, prevention of VVC. Many women who suffer from VVC already
respectively. In addition, the total annual cost (in 1995) for deal- use these agents without prescription. In a survey carried out by
ing with VVC was estimated at $1.8 billion. The high incidence Pirotta et al., 73% of 1117 women in the age range 18–70 years
and associated healthcare cost of VVC highlight the need for the self-reported having had symptoms of VVC in the past and 35%
development of effective agents for its prevention. reported that these symptoms appeared after an antibiotic course.
Probiotics are defined as live microorganisms which when Lactobacillus products were used by 40% and 43% of these
administered in adequate amounts confer a health benefit on women for prevention and treatment of post-antibiotic vulvo-
the host.2 There is strong evidence that Lactobacillus rhamnosus vaginitis, respectively.3
GG is effective for the treatment of acute rotavirus diarrhoea in It should be mentioned that the names of some lactobacilli
children, causing a significant reduction of its duration.2 In addi- strains have changed recently. Lactobacillus acidophilus RC-14
tion, their usefulness for the prevention and/or treatment of many and Lactobacillus fermentum RC-14 studied in the Netherlands
.............................................................................................................................................................................................................................................................................................................................................................................................................................

*Correspondence address. Alfa Institute of Biomedical Sciences (AIBS), 9 Neapoleos Street, 151 23 Marousi, Greece.
Tel: +30-694-611-0000; Fax: +30-210-683-9605; E-mail: [email protected]
.............................................................................................................................................................................................................................................................................................................................................................................................................................

266

The Author 2006. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]
 Review

and Canada were renamed as Lactobacillus reuteri RC-14 and The results of some studies associated VVC either with a
Lactobacillus casei GR-1 and L. casei var. rhamnosus was reduced number of lactobacilli or with species of lactobacilli
renamed as L. rhamnosus GR-1. However, in our review we not producing H2O2. In a study of 7918 pregnant women, Hillier
used the names of lactobacillus strains as they were mentioned et al. found that VVC was associated either with normal vaginal
in the cited articles. microbiota (dominated by lactobacilli) or with intermediate flora
(with decreased lactobacilli).12 Some other studies suggested that
pregnant13,14 or post-term15 women whose vaginas were
Literature search colonized by H2O2-producing lactobacilli were less likely to
have symptomatic VVC than those colonized with H2O2-non-
We searched for articles in the PubMed (1975–1/2006), from
producing vaginal lactobacilli. However, Hawes et al.16 sug-
which we also found some additional relevant references. The
gested that H2O2-producing lactobacilli do not protect against
keywords were ‘vulvovaginal candidiasis’, ‘candidal vaginitis’,
VVC. In a study of 182 women visiting a sexually transmitted
‘yeast vaginitis’, ‘fungal infections’, ‘probiotics’, ‘lactobacilli’,
disease clinic, 25 of whom developed VVC during a 2 year
‘bifidobacteria’. We focused on microbiological studies and clini-
follow-up, the absence of lactobacilli from the vagina was not
cal trials. Specifically, we found relevant information from origi-

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

found to increase the incidence of VVC.
nal articles and reviews regarding the role of endogenous
lactobacilli both in normal and in Candida-infected vaginal
flora, in vitro experiments investigating the effect of probiotics
on the growth and adherence of Candida albicans on the In vitro experiments
vaginal epithelium, human studies examining the ability of orally
There are some in vitro experiments which show that some
or intravaginally administered probiotics to prevent recurrent
lactobacilli strains can inhibit the adherence and/or the growth
VVCs and adverse effects of probiotics.
of C. albicans. However, these results do not necessarily apply to
humans, since the physiological and pathophysiological mecha-
Vaginal lactobacilli and pathogenesis of VVC nisms taking place in humans are more complex and cannot be
accurately imitated in the laboratory.
Lactobacilli, especially Lactobacillus crispatus,4–7 Lactobacillus Osset et al.17 found that 8 of 15 studied lactobacilli inhibited
jensenii4,7 and Lactobacillus iners,5,7,8 are most commonly the significantly the adhesion of C. albicans Y18 to vaginal cells.
dominant microorganisms in the vagina of healthy pre- They also found that some lactobacilli inhibited the growth of
menopausal women. Lactobacilli produce lactic acid and other C. albicans Y18 in liquid assays, but not in solid assays. Strus
substances, which maintain a low pH in the vagina, thus prevent- et al.18 found that Lactobacillus delbrueckii, which produces
ing the overgrowth of pathogens, at least those causing bacterial large amounts of H2O2, inhibited the growth of C. albicans
vaginosis (BV) and gonorrhoea. C. albicans may also be found in more strongly and quickly than many other studied strains
the vagina of healthy premenopausal women. In a study of pre- isolated from the vaginas of healthy women, while Lactobacil-
menopausal women by Sobel et al., C. albicans was isolated from lus plantarum, which does not produce H2O2, showed the most
25% of 20 healthy women.9 prolonged inhibitory activity starting after 24 h. Boris et al.19
Although the pathogenesis of VVC remains a controversial found that L. acidophilus, Lactobacillus gasseri and L. jensenii,
issue, it seems that when the balance between the microorganisms isolated from the vaginas of healthy premenopausal women,
existing in the vaginal microbiota is disrupted, the overgrowth of coaggregated in vitro with C. albicans, isolated from the
Candida is facilitated. Antibiotic therapy, spermicide use, oral same vaginal samples. The adherence of C. albicans on the
contraceptives, oestrogen therapy, diabetes mellitus, tight cloth- vaginal epithelial cells, collected from the same women, was
ing and frequent sexual intercourse are factors that increase the greatly decreased when L. acidophilus was added in comparison
risk for development of VVC.10 Women with VVC complain with the adherence observed when only Candida was present.
about thick white caseous vaginal discharge and pruritus, and Adherence on the vagina is an important virulence factor of
often dyspareunia, vulvar erythema and swelling. Similar C. albicans; thus, reducing its adherence may prevent VVC.
symptoms and signs also occur in women with BV, thus leading Coaggregation of lactobacilli with Candida may also be
to frequent misdiagnosis of BV as VVC, especially when it important for the prophylaxis against vaginal infections by
occurs after antibiotic therapy. preventing the binding of Candida to the receptors of the
It has been suggested in some studies that lactobacilli are quite vaginal epithelium.19
common even in the vaginal epithelium of women with VVC. Some substances produced by specific lactobacilli strains have
Sobel et al. found that lactobacilli were the dominant vaginal been found to exert an inhibitory effect upon C. albicans, at least
microorganisms in 90% of 20 healthy premenopausal women and in vitro. Velraeds et al. found that the initial adherence rates of
in 96% of 24 premenopausal women with acute exacerbations of two C. albicans strains, suspended in a urine sample, on a sili-
recurrent VVC.9 However, the composition of lactobacilli species cone rubber filled with a biosurfactant of L. acidophilus RC-14
and/or strains was different between healthy women and those (‘surlactin’), 4 h after low urine flow, decreased by 50% compared
with VVC. The vaginal microbiota of healthy women was more with the adherence rates on a silicone rubber without surlactin,
frequently dominated by Lactobacillus salivarius (isolation rate although the numbers of adhering Candida cells were similar
35%), while the vagina of women with VVC was more com- between the two rubbers.20 Okkers et al. found that ‘pentocin
monly dominated by Lactobacillus catenaforme (isolation rate TV35b’, a bacteriocin-like peptide isolated from Lactobacillus
42%).9 Demirezen et al.11 found that presence of lactobacilli pentosus, inhibited the growth of C. albicans.21 Reid et al.22
was more common among 59 studied women with VVC than suggested that a biosurfactant produced by L. fermentum
among 391 healthy women. RC-14 inhibits the adhesion of C. albicans.

267
 Review

Clinical studies 4 women were dominated by lactobacilli. One week after the
beginning of the trial, lactobacilli dominated the vagina of all
In Table 1 we present some clinical trials that have been con- women and GR-1 and/or RC-14 were recovered from all of them.
ducted in order to evaluate the ability of orally or intravaginally No recurrences of yeast vaginitis appeared during the study and
administered lactobacilli to inhibit the vaginal colonization by follow-up.
yeast and prevent the recurrence of VVC. Reid et al.23 reported Reid et al.28 supported the possible ability of orally adminis-
the case of a 33-year-old woman with recurrent cystitis and VVC tered L. rhamnosus GR-1 and L. fermentum RC-14 (at a dose of
(20 episodes of VVC in 30 months), whose vagina was colonized more than 8 · 108 viable lactobacilli) to restore and maintain a
by L. casei var. rhamnosus GR-1 up to 7 weeks after the vaginal normal vaginal microbiota in a randomized clinical trial in 42
administration of one pessary of these lactobacilli. The woman women in the age range 17–50 years without symptoms of uro-
had no symptoms of vaginitis during this period and for the next genital infection at the start and during the study, 33 of whom
6 months during which two more pessaries were inserted into her reported a history of VVC. The women were randomly separated
vagina. into four groups; groups 1, 2 and 3 received daily orally capsules
A clinical trial suggesting the effectiveness of vaginal lacto- of GR-1/RC-14 at different dosages and group 4 received daily
bacilli for the treatment of VVC was conducted by Hilton

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

one capsule of L. rhamnosus GG. Of the women who had a
et al.,24 who administered vaginal suppositories of Lactobacillus normal vaginal microbiota at the beginning of the study, 92%
GG twice per day for 7 days to 28 women with symptoms and (12/13) of the GR-1/RC-14-treated and 50% (2/4) of the GG-
signs of VVC at the start of the study and a history of recurrent treated remained normal within 28 days. Of the women who had
VVC (>5 per year). A serious limitation of this study was that a history of yeast vaginitis in the 5 years prior to the study and an
only 5 of these women had considerable colonies of C. albicans abnormal vaginal microbiota at the start of the study, 54% (7/13)
at the start of the study, maybe because 15 of the studied women of the GR-1/RC-14-treated and 25% (1/4) of the GG-treated
had taken antifungal agents just before the study. All of them developed a normal vaginal microbiota within 28 days.
reported improvement of their vaginal symptoms and were found Hilton et al.29 found that L. acidophilus can reduce the vaginal
to have reduced vaginal erythema and discharge during clinical colonization and infection by Candida in a clinical trial in 33
examination. Four of the five women with positive vaginal cul- women with recurrent candidal vaginitis (‡5/year), 13 of whom
tures had negative cultures after receiving lactobacilli. However, completed the study. The studied women were randomized into
no conclusions can be drawn from this study due to its poor design two groups; the first group received daily 8 ounces of yogurt with
and the small sample of studied women with confirmed VVC. L. acidophilus for 6 months and did not consume any yogurt for
Williams et al.25 also examined the ability of intravaginally the 6 following months, and the other group consumed first the
administered lactobacilli to reduce the VVC risk in a double- yogurt-free and then the yogurt-containing diet. The mean num-
blind, placebo-controlled trial of 164 HIV-positive women, a ber of candidal infections and the mean number of candidal
group of patients in whom recurrent VVC is common. The colonizations of the vagina and rectum per woman were signifi-
women were randomized into three groups: the first group cantly less during the 6 months of yogurt consumption in com-
received intravaginally L. acidophilus once per week, the second parison with the 6 months without receiving yogurt (0.38 versus
received vaginal clotrimazole weekly and the third took placebo 2.54, P = 0.001 and 0.84 versus 3.23, P = 0.001, respectively).
(control group). During 21 months of the study, 34 cases of VVC However, the fact that the vaginal colonization by Lactobacillus
were diagnosed clinically and microbiologically. The relative risk was not statistically significantly increased during yogurt intake
of developing VVC was 0.5 for the lactobacilli-treated and 0.4 for and that the studied women were not blinded makes it difficult to
the clotrimazole-treated women as compared with the control interpret the results of this study.
group. Moreover, the median time until the first episode of On the other hand, there are a few studies that do not support a
VVC was longer for women who received lactobacilli than for role for probiotics in the prevention of recurrent VVC. Shalev
those who took placebo, but the difference was not statistically et al.30 studied 46 women with recurrent vaginitis (‡4 episodes
significant (P = 0.09). during the year prior to the study), 18 of whom had VVC and 8
Some other studies have investigated the ability of orally had VVC and bacterial vaginosis. The women were randomized
administered lactobacilli to colonize the vagina and/or reduce into 2 groups of 23 women; the first group received 150 mL/day
the vaginal colonization and infection by Candida. Reid et al.26 yogurt with >108 cfu/mL live L. acidophilus for 2 months and the
conducted a randomized trial in 64 healthy women in the age second group received 150 mL/day of a pasteurized yogurt for the
range 19–46 years without any urogenital infections in the year same period. For the next 2 months women of both groups did not
prior to the study. For 60 days 32 of the studied women received consume any yogurt. For the last 2 months of the study the first
orally daily L. rhamnosus GR-1 and L. fermentum RC-14, while group consumed pasteurized yogurt and the second group yogurts
the other 32 women received placebo. Cultures of the vaginal with lactobacilli. During the first 4 months 28 women took part in
swabs of the studied women 4 weeks after the administration the study and only 7 completed the whole protocol. The percent-
showed a significant increase in vaginal lactobacilli (P = 0.01) age of women with positive L. acidophilus vaginal cultures
and a significant reduction in yeast (P = 0.01) in the lactobacilli- among the first group increased after the first 1 and 2 months
treated compared with the placebo-treated women. of the study and was significantly higher than that of the second
In another clinical trial, Reid et al.27 administered L. rhamnosus group. Although a progressive decrease in positive vaginal cul-
GR-1 and L. fermentum RC-14 (>109 viable) orally twice daily tures for Candida was found in both groups, the percentage of
for 14 days in 10 women with recurrent urogenital infections, 9 of women with positive Candida cultures after the first 1 and 2
whom had recurrent yeast vaginitis. The vaginal microbiota of 5 months of the study was not significantly different between
of those 9 women with recurrent VVC had <10 colonies or no the two groups. However, in this trial no consideration was
lactobacilli at the start of the study, while the vaginas of the other given to the properties of the tested lactobacillus strain against

268
 Table 1. Characteristics of reviewed studies

First No. of
author Type of women Study Symptoms and signs of Cultures of vaginal swabs
(ref) study studied population Intervention VVC after intervention after intervention

Hilton prospective 28 recurrent VVC (>5/year) vaginal suppositories with improvement of symptoms 4/5 women with cultures
(1993)24 cohort and symptoms of VVC Lactobacillus GG and signs (# of erythema positive for Candida
(positive cultures for twice/day for 7 days and discharge) albicans before
Candida in 5 women) intervention: negative
cultures 7 days after the
completion of intervention
Williams randomized 164 HIV-positive group 1 (58 women): vaginal cases of VVC (culture- 34 culture-confirmed cases
(2001)25 controlled L. acidophilus weekly confirmed) in 21 months of VVC in 21 months
trial group 2 (50 women): vaginal (median) of follow up: relative risk for
clotrimazole weekly group 1: 9/58 (15.5%) development of VVC:
group 3 (56 women): placebo group 2: 7/50 (14%) group 1 compared with
group 3: 18/56 (32.1%) group 3: 0.5
group 2 compared with
group 3: 0.4
Reid randomized 64 no urogenital infection in group 1 (32 women): improvement of vaginal 4 weeks after start of
(2003)26 controlled the year prior to the study L. rhamnosus GR-1 + symptoms: 30% (group 1) intervention: increase in vaginal

269
trial L. fermentum RC-14 versus 12% (group 2) lactobacilli: group 1 >
Review

orally once/day for group 2 (P = 0.01), decrease in

60 days vaginal Candida: group 1 > group 2
group 2 (32 women): placebo (P = 0.01)
Reid prospective 10 recurrent urogenital infections L. rhamnosus GR-1 + no symptoms of VVC 5/5 women with history of
(2001)27 cohort (9 with recurrent VVC) L. fermentum RC-14 recurrent VVC & low or
orally twice/day for no lactobacilli before
14 days intervention had normal
number of lactobacilli
1 week after start of
intervention
Reid randomized 42 no symptoms of urogenital group 1 (n = 10): GR-1/RC-14 normal vaginal flora 28 days
(2001)28 controlled infection at the start of 8 · 108/day orally for 28 days after start of intervention:
trial the study (history of group 2 (n = 12): GR-1/RC-14 12/13 of groups 1, 2 & 3
VVC in 33 women) 1.6 · 109/day orally for 28 days versus 2/4 of group 4 with
group 3 (n = 11): GR-1/RC-14 normal vagina at the start
6 · 109/day orally for 28 days of the study,
group 4 (n = 9): GG 1010/day 7/13 of groups 1, 2 & 3
orally for 28 days versus 1/4 of group
4 with abnormal vagina at
the start of the study and
VVC within 5 years prior
to the study

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

 Hilton randomized 33 recurrent VVC (‡5/year) group 1: 8 ounces/day yogurt mean number of VVC mean number of candidal
(1992)29 controlled L. acidophilus for 6 months during 6 months colonization of vagina
trial then no yogurt for the next receiving yogurt smaller & rectum/woman during
6 months than that during 6 months 6 months receiving yogurt
group 2: no yogurt for 6 months not taking yogurt (0.38 < during 6 months not
then 8 ounces/day yogurt versus 2.54, P = 0.001) taking yogurt (0.84 versus
L. acidophilus for the next 3.23, P = 0.001)
6 months
Shalev randomized 46 recurrent (‡4 in the last year) group 1 (n = 23): 150 mL/day positive cultures for
(1996)30 controlled vaginitis (VVC: n = 18, yogurt with L. acidophilus L. acidophilus:
trial VVC & bacterial vaginosis: for 2 months then no yogurt before intervention:
n = 8) for the next 2 months then 20% (group 1) versus
pasteurized yogurt for the last 31% (group 2)
2 months after 1 month: 71% (group 1)
group 2 (n = 23): pasteurized versus 27% (group 2),
yogurt for 2 months then no P < 0.05
yogurt for the next 2 months after 2 months:
then 150 mL/day yogurt with 92% (group 1) versus 30%
L. acidophilus for the last (group 2), P < 0.05
2 months positive cultures for
Candida: before
intervention: 56% (group 1)
versus 62% (group 2)
after 1 month: 44% (group 1)

270
versus 37% (group 2), P > 0.05
Review

after 2 months: 21% (group 1)

versus 28% (group 2), P > 0.05
Pirotta randomized 278 treatment with antibiotics group 1 (n = 67): L. rhamnosus & vaginal symptoms after positive Candida cultures
(2004)31 placebo- for non-gynaecological Bifidobacterium longum orally 14 days: after 14 days:
controlled infections twice/day + L. rhamnosus, group 1: 15/67 (22%) group 1: 16/67 (24%)
double-blind L. delbrueckii, L. acidophilus & group 2: 18/70 (27%) group 2: 17/70 (24%)
trial Streptococcus thermophilus group 3: 17/73 (25%) group 3: 17/73 (23%)
vaginally once/night for 10 days group 4: 22/68 (34%) group 4: 9/68 (13%)
group 2 (n = 70): L. rhamnosus,
L. delbrueckii, L. acidophilus &
Streptococcus thermophilus
vaginally once/night + placebo
orally twice/day for 10 days
group 3 (n = 73): L. rhamnosus &
Bifidobacterium longum orally
twice/day + placebo vaginally
once/night for 10 days
group 4 (n = 68): placebo twice/day
orally + once/night vaginally
for 10 days

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

 Review

Candida; thus, it is difficult to draw conclusions regarding the Lactobacilli have been frequently found to co-exist with Candida
effectiveness of probiotics against VVC. in the vaginal epithelium of women with VVC, while they are
Pirotta et al.31 also did not support the use of combinations of significantly reduced in women with BV. Some in vitro studies
specific lactobacilli, either given orally or intravaginally, for the and clinical trials had positive results regarding the effectiveness
prevention of post-antibiotic vulvovaginitis. They conducted a of some specific lactobacilli strains against C. albicans. However,
randomized, placebo-controlled, double-blind clinical trial in most of the trials either included a small sample of women or
non-pregnant women in the age range 18–50 years who received women with no confirmed episodes of VVC or were not placebo-
oral powder of L. rhamnosus and Bifidobacterium longum (Lac- controlled. Moreover, there were differences among the trials
tobac) twice daily and/or one vaginal pessary of L. rhamnosus, regarding the strain of the tested probiotic, its dosage and the
L. delbrueckii, L. acidophilus and Streptococcus thermophilus duration of treatment. It should be emphasized that the various
(Femilac) each night and/or oral and/or intravaginal placebo dur- probiotic strains have different properties and different effects on
ing 6 days of antibiotic administration for a non-gynaecological Candida; thus, results from studies testing one strain should not
infection and for 4 days afterwards. Four days after completion of be extrapolated to other strains. Consequently, it is difficult to
the intervention, post-antibiotic vulvovaginitis had developed in draw reliable conclusions from the existing studies. Probiotics,

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

23% (55/235) of the studied women [95% confidence interval especially L. acidophilus, L. rhamnosus GR-1 and L. fermentum
(CI) 18–29%]. The OR for developing post-antibiotic vulvo- RC-14, may be considered as potential empirical preventive
vaginitis was 1.06, while receiving oral lactobacilli (95% CI agents in women who suffer from frequent episodes of VVC
0.58–1.94), and 1.38, while receiving vaginal lactobacilli (95% (more than three episodes per year), since adverse effects from
CI 0.75–2.54), in comparison with placebo. their use are scarce, especially when the use of antifungal agents
A careful review of the studies investigating the role of pro- is contraindicated or is associated with adverse effects. However,
biotics in the prevention of recurrent VVC suggests that most of more randomized, double-blind, placebo-controlled trials with a
them have important methodological shortcomings. In the major- larger sample size should be carried out, so as to clarify whether
ity of the studies only a small sample of women was included or probiotics can be used effectively and safely for the prophylaxis
completed the study.27,29,30 Moreover in most of the studies it of recurrent episodes of VVC.
was not mentioned whether the reported episodes of recurrent
VVC prior to the trials were confirmed by cultures of the vaginal
fluids or they were just self-diagnosed by the studied Acknowledgements
women.24,27–30 Furthermore, there was no control group in
some of the studies, so as to compare the women receiving Funding: none.
lactobacilli with others receiving placebo.24,27,28
Finally, it is worth mentioning that the studies by Williams
et al. and Pirotta et al. did not examine the ability of probiotics to Transparency declarations
prevent recurrences in women who already had recurrent VVC, None to declare.
but the efficacy of specific probiotics to prevent the development
of VVC in women at high risk for such infections, specifically
HIV-positive and women receiving antibiotics.25,31 In addition, References
the study by Reid et al.26 in 2003 addressed the potential
effectiveness of two specific lactobacilli to prevent development 1. Foxman B, Barlow R, D’Arcy H et al. Candida vaginitis: self-
of VVC in healthy women. reported incidence and associated costs. Amer Sex Transm Dis Assoc
2000; 27: 230–5.
Adverse effects of probiotics 2. Reid G, Jass J, Sebulsky T et al. Potential uses of probiotics in
clinical practice. Clin Microbiol Rev 2003; 16: 658–72.
Probiotics are generally considered to be safe. However, some 3. Pirotta M, Gunn J, Chondros P. ‘Not thrush again!’ Women’s
probiotic species have been rarely isolated from infectious sites. experience of post-antibiotic vulvovaginitis. MJA 2003; 179: 43–6.
A study showed that only in a mean of 0.2% of positive blood 4. Antonio M, Hawes S, Hillier S. The identification of vaginal
cultures in Finland during 1995–2000 were Lactobacillus isolates Lactobacillus species and the demographic and microbiologic charac-
reported.32 Apart from lactobacillaemia, infectious endocarditis,33 teristics of women colonised by these species. J Infect Dis 1999; 180:
1950–6.
liver abscess34 and fungaemia35 are some infections which have
been associated with probiotics. These cases appear mainly in 5. Zhou X, Bent SJ. Characterization of vaginal microbial commu-
nities in adult healthy women using cultivation-independent methods.
patients with serious underlying diseases and/or immunosup-
Microbiology 2004; 150: 2565–73.
pression.33,36,37 Moreover, only a few cases have been reported,
6. Song Y, Kato N, Matsumiya Y et al. Identification of and hydrogen
which associate consumed lactobacilli with those isolated from peroxide production by fecal and vaginal Lactobacilli isolated from
infectious sites.34,38 Generally, the cases of infections associated Japanese women and newborn infants. J Clin Microbiol 1999; 37:
with probiotics are scarce in comparison with the considerable 3062–4.
and gradually increasing consumption of probiotics.32 7. Vasquez A, Jakobsson T, Ahrne S et al. Vaginal Lactobacillus
flora of healthy Swedish women. J Clin Microbiol 2002; 40: 2746–9.
Conclusions 8. Butron J, Cardieux P, Reid G. Improved understanding of the
bacterial vaginal microbiota of women before and after probiotic
In conclusion, it is still controversial whether probiotics can instillation. Appl Environ Microbiol 2003; 69: 97–101.
prevent recurrences of VVC, while there may be more patho- 9. Sobel J, Chaim W. Vaginal microbiology of women with acute
physiological basis for their effectiveness in the prevention of BV. recurrent vulvovaginal candidiasis. J Clin Microbiol 1996; 34: 2497–9.

271
 Review

10. Jeavons H. Prevention and treatment of vulvovaginal candidiasis 25. Williams AB, Yu C, Tashima K et al. Evaluation of two self-care
using exogenous Lactobacillus. J Obstet Gynecol Neonatal Nurs 2003; treatments for prevention of vaginal candidiasis in women with HIV.
32: 287–96. J Assoc Nurses AIDS Care 2001; 12: 51–7.
11. Demirezen S. The Lactobacilli-Candida relationship in cervico- 26. Reid G, Charbonneau D, Erb J et al. Oral use of Lactobacillus
vaginal smears. Cent Eur J Public Health 2002; 10: 97–9. rhamnosus GR-1 and L. fermentum RC-14 significantly alters vaginal
12. Hillier SL, Krohn MA, Nugent RP et al. Characteristics of three flora: randomized, placebo-controlled trial in 64 healthy women. FEMS
vaginal flora patterns assessed by gram stain among pregnant women. Immun Med Microbiol 2003; 35: 131–4.
Vaginal infections and prematurity study group. Am J Obstet Gynecol 27. Reid G, Bruce A, Fraser N et al. Oral probiotics can resolve
1992; 166: 938–44. urogenital infections. FEMS Immun Med Microbiol 2001; 30: 49–52.
13. Hillier SL, Krohn MA, Klebanoff SJ et al. The relationship of
28. Reid G, Beuerman D, Heinemann C et al. Probiotic Lactobacillus
hydrogen peroxide-producing lactobacilli to bacterial vaginosis and
dose required to restore and maintain a normal vaginal flora. FEMS
genital microflora in pregnant women. Obstet Gynecol 1992; 79:
Immun Med Microbiol 2001; 32: 37–41.
369–73.
14. Rossel GA, Holst E, Milsom I et al. Fetal fibronectin and 29. Hilton E, Isenberg HD, Alperstein P. Ingestion of yogurt containing
microorganisms in vaginal fluid of healthy pregnant women. Acta Obstet Lactobacillus acidophilus as prophylaxis for candidal vaginitis. Ann
Intern Med 1992; 116: 353–7.

Downloaded from https://academic.oup.com/jac/article-abstract/58/2/266/721575 by guest on 05 December 2018

Gynecol Scand 1996; 75: 520–5.
15. Goffeng AR, Holst E, Nilsson C et al. Microoorganisms in vaginal 30. Shalev E, Battino S, Weiner E et al. Ingestion of yogurt containing
fluid from women in prolonged pregnancy. Gynecol Obstet Invest 1997; Lactobacillus acidophilus compared with pasteurized yogurt as prophy-
44: 16–20. laxis for recurrent candidal vaginitis and bacterial vaginosis. Arch Fam
16. Hawes SE, Hillier SL, Benedetti J et al. Hydrogen peroxide- Med 1996; 5: 593–6.
producing lactobacilli and acquisition of vaginal infections. J Infect Dis 31. Pirotta M, Gunn J, Chondros P et al. Effect of lactobacillus in
1996; 174: 1058–63. preventing post-antibiotic vulvovaginal candidiasis: a randomized con-
17. Osset J, Garcia E, Bartolome RM et al. Role of Lactobacillus as trolled trial. BMJ 2004; 329: 548–52.
protector against vaginal candidiasis. Med Clin 2001; 117: 285–8. 32. Salminen MK, Tynkkynen S, Rautelin H et al. Lactobacillus
18. Strus M, Brzychczy-Wloch M, Kucharska A et al. Inhibitory activity bacteremia during a rapid increase in probiotic use of Lactobacillus
of vaginal Lactobacillus bacteria on yeasts causing vulvovaginal rhamnosus GG in Finland. Clin Infect Dis 2002; 35: 1155–60.
candidiasis. Med Dosw Mikrobiol 2005; 57: 7–17. 33. Husni RN, Gordon SM, Washington JA et al. Lactobacillus
19. Boris S, Suarez J, Vazquez F et al. Adherence of human vaginal bacteremia and endocarditis: review of 45 cases. Clin Infect Dis 1997;
Lactobacilli to vaginal epithelial cells and interaction with uropathogens. 25: 1048–55.
Infect Immun 1998; 66: 1985–9.
34. Rautio M, Jousimies-Somer H, Kauma H et al. Liver abscess due
20. Velraeds MM, van de Belt-Gritter B, van der Mei HC et al. to a Lactobacillus rhamnosus strain indistinguishable from L. rhamnosus
Interference in initial adhesion of uropathogenic bacteria and yeasts to strain GG. Clin Infect Dis 1999; 28: 1159–60.
silicone rubber by a Lactobacillus acidophilus biosurfactant. J Med
Microbiol 1998; 47: 1081–5. 35. Munoz P, Bouza E, Cuerca-Estrella M et al. Saccharomyces
cerevisiae fungemia: an emerging infectious disease. Clin Infect Dis
21. Okkers DJ, Dicks LM, Silvester M et al. Characterization of
2005; 40: 1625–34.
pentocin TV35b, a bacteriocin-like peptide isolated from Lactobacillus
pentosus with a fungistatic effect on Candida albicans. J Appl Microbiol 36. Borriello SP, Hammes WP, Holzapfel W et al. Safety of probiotics
1999; 87: 726–34. that contain Lactobacilli or Bifidobacteria. Clin Infect Dis 2003; 36:
22. Reid G, Bruce AW. Selection of Lactobacillus strains for urogenital 775–80.
probiotic applications. J Infect Dis 2001; 183: S77–80. 37. Antony SJ, Stratton CW, Dummer S. Lactobacillus bacteremia:
23. Reid G, Millsap K, Bruce A. Implantation of Lactobacillus casei var description of the clinical course in adult patients without endocarditis.
rhamnosus into vagina. Lancet 1994; 344: 1229. Clin Infect Dis 1996; 23: 773–8.
24. Hilton E, Rindos P, Isenberg H. Lactobacillus GG vaginal 38. Mackay A, Taylor M, Kibbler C et al. Lactobacillus endocarditis
suppositories and vaginitis. J Clin Microbiol 1995; 33: 1433. caused by a probiotic organism. Clin Microbiol Infect 1999; 5: 290–2.

272