Aliment Pharmacol Ther 2005; 22: 365–372.
Meta-analysis: non-pathogenic yeast Saccharomyces boulardii in
the prevention of antibiotic-associated diarrhoea
H. SZAJEWSKA* & J. M RUKOWICZ 
*Department of Pediatric Gastroenterology and Nutrition, The Medical University of Warsaw;  Polish Institute for Evidence-
based Medicine and Medycyna Praktyczna, Cracow, Poland
Accepted for publication 7 July 2005
SUMMARY
Background: Antibiotic-associated diarrhoea occurs in
up to 30% of patients who receive antibiotics but can be
prevented with probiotics.
Aim: To systematically evaluate the effectiveness of
Saccharomyces boulardii in preventing antibiotic-associ-
ated diarrhoea in children and adults.
Methods: Using medical subject headings and free-
language terms, the following electronic databases were
searched for studies relevant to antibiotic-associated
diarrhoea and S. boulardii: MEDLINE, EMBASE, CINAHL
and The Cochrane Library. Additional sources were
obtained from references in reviewed articles. Only
randomized-controlled trials were considered for study
inclusion.
INTRODUCTION
Antibiotic-associated diarrhoea (AAD) is defined as
otherwise unexplained diarrhoea that occurs in associ-
ation with the administration of antibiotics.1 Although
no infectious agent is found in most cases, the bacterial
agent commonly associated with AAD, particularly in
most severe episodes (pseudomembranous colitis), is
Clostridium difficile.2 AAD occurs in approximately
5–30% of patients between the initiation of therapy
Correspondence to: Dr H. Szajewska, Department of Paediatric Gastroen-
terology and Nutrition, The Medical University of Warsaw, 01-184
Warsaw, Dzialdowska 1, Poland.
E-mail: hania@ipgate.pl

2005 Blackwell Publishing Ltd
doi: 10.1111/j.1365-2036.2005.02624.
Results: Of 16 potentially relevant clinical trials identified,
five randomized-controlled trials (1076 participants) met
the inclusion criteria for this systematic review. Treat-
ment with S. boulardii compared with placebo reduced the
risk of antibiotic-associated diarrhoea from 17.2% to
6.7% (RR: 0.43; 95% CI: 0.23–0.78; random effect
model). The number needed to treat to prevent one case of
antibiotic-associated diarrhoea was 10 (95% CI: 7–16).
No side-effects were reported.
Conclusions: A meta-analysis of data from five random-
ized-controlled trials showed that S. boulardii is moder-
ately effective in preventing antibiotic-associated
diarrhoea in children and adults treated with antibiotics
for any reason (mainly respiratory tract infections).
For every 10 patients receiving daily S. boulardii with
antibiotics, one fewer will develop antibiotic-associated
diarrhoea.
and up to 2 months after cessation of treatment.2–4 The
incidence of diarrhoea in children receiving broad-
spectrum antibiotics ranges from 11% to 40%.5, 6 Almost
all antibiotics, particularly those active against anaer-
obes, can cause diarrhoea, but the risk seems to be higher
with aminopenicillins, a combination of aminopenicillins
and clavulanate, cephalosporins and clindamycin.7, 8
Preventive measures include the use of probiotics,
which are live microbial food ingredients that are
beneficial to health.9 The most commonly used probi-
otics are lactic acid bacteria, such as lactobacilli or
bifidobacteria, but non-bacterial organisms, such as a
non-pathogenic yeast Saccharomyces boulardii, also have
been used. The rationale for the use of probiotics in AAD
365
366 H. SZAJEWSKA & J. MRUKOWICZ
is based on the assumption that the use of antibiotics
leads to a disturbance in the normal intestinal micro-
flora and that this is a key factor in the pathogenesis of
AAD.10 In a randomized, double-blind, placebo-
controlled study, to which one of us (H.S.) contributed,
it was demonstrated that S. boulardii (250 mg po b.d.) is
effective in preventing AAD in children 6 months to
14 years of age treated with antibiotics for otitis media
and/or respiratory tract infections.11
Two systematic reviews of the use of different strains of
probiotics in the prevention of AAD have been pub-
lished.12, 13 A problem with both of these reviews is a
lack of power, which likely reflects more restrictive
search criteria (e.g. language restriction). In addition,
the literature search in these reviews was restricted to
only two databases, and authors did not report whether
they reviewed references in retrieved articles. Results
from both systematic reviews suggest that probiotics,
including S. boulardii, have the potential to be useful in
this situation but that further data are needed. Critics of
using a meta-analytical approach to assess the efficacy
of probiotics argue that beneficial effects of probiotics
seem to be strain-specific, thus, pooling data on different
strains may result in misleading conclusions.
Given these concerns the present review was under-
taken to update data on the effectiveness and safety of
only one probiotic micro-organism – S. boulardii – in the
prevention of AAD.
METHODS
Inclusion criteria
Electronic databases (see Search strategy) were system-
atically searched to identify studies appropriate for
inclusion in this systematic review. Inclusion criteria
were as follows.
Types of studies. Randomized-controlled trials (RCTs)
that compared S. boulardii with placebo or no additional
intervention.
Types of participants. Adults and children who received
antibiotics for any reason.
Types of interventions. Patients in the experimental
groups received S. boulardii as an adjunct to antibiotics.
Patients in the control group received placebo or no
additional intervention.

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372
Types of outcome measures. The ‘primary’ outcome
measure was the incidence of diarrhoea or AAD (as
defined by the investigators). The ‘secondary’ outcome
measures were as follows: the incidence of C. difficile
diarrhoea; mean frequency of bowel movements; mean
duration of diarrhoea; the need for discontinuation of
the antibiotic treatment, hospitalization to manage the
diarrhoea (in out-patients) or intravenous rehydration
in any of the study groups and adverse events.
Search strategy
The following electronic databases were systematically
searched for relevant studies: MEDLINE (1966 – March
2005), EMBASE (1980 – March 2005), Cumulative
Index to Nursing and Allied Health (CINAHL, 1982 –
March 2005), The Cochrane Database of Systematic
Reviews (Issue 1, 2005) and The Cochrane Controlled
Trials Register (Issue 1, 2005). The search strategy
included use of a validated filter for identifying
controlled trials,14 which was combined with a topic-
specific strategy. The search terms were diarrhoea/
diarrhoea, antibiotic-associated/antibiotic associated, C.
difficile, probiotics and S. boulardii. Furthermore, refer-
ence lists from the original studies and review articles
identified were screened, and key experts in the field
were approached for unpublished material. No limit was
imposed regarding the language of publication, but
certain publication types (i.e. letters to the editor,
abstracts, proceedings from scientific meetings) were
excluded.
Methods of review
Included and excluded studies. Two reviewers independ-
ently screened titles and abstracts identified according to
the above-described search strategy. All potentially
relevant articles were retained and the full text of these
studies examined to determine which studies satisfied
the inclusion criteria. The same reviewers independ-
ently carried out data extraction, using standard data
extraction forms. Studies reported in languages other
than those familiar to the authors were translated.
Discrepancies between the reviewers were resolved by
discussion.
Study quality. Two reviewers independently, but with-
out being blinded to the authors or journal, assessed the
quality of studies that met the inclusion criteria. Use of
 META-ANALYSIS: S. BOULARDII IN THE PREVENTION OF DIARRHOEA
the following strategies associated with good quality
studies was assessed: (i) allocation concealment, (ii)
blinding of investigators, participants, outcome asses-
sors and data analysts (yes/no/not reported), (iii)
intention-to-treat (ITT) analysis (yes/no) and (iv) com-
prehensive follow-up.
Allocation concealment was considered adequate
when the randomization method used did not allow
the investigator or the participant to identify or
influence the intervention group before enrolment of
eligible participants in the study. However, the quality
of the allocation concealment was considered unclear
when randomization was used but no information
about the method was available and inadequate, when
inappropriate methods of randomization (e.g. alternate
medical record numbers, unsealed envelopes, tossing
the coin) were used.
In regard to the ITT analysis, an answer of ‘yes’ meant
that the authors had specifically reported undertaking
this type of analysis and/or that our own study
confirmed this finding. Conversely, a ‘no’ meant that
authors did not report use of ITT analysis and/or that
we could not confirm its use on study assessment. To
evaluate the completeness of patient follow-up, we
determined the percentage of participants excluded or
lost to follow-up.
Statistical methods
The data were analysed using statsdirect software
(2, 3, 8; 2004-04-17). The binary measure (prevalence
of diarrhoea or AAD) for individual studies and pooled
statistics is reported as the risk ratio (RR) between the
experimental and control groups with 95% confidence
intervals (CI). We calculated the number needed to treat
(NNT) as the inverse of pooled absolute risk differences
and 95% CI. The weighted mean difference between the
treatment and control groups was selected to represent
the difference in continuous outcomes. The weights
given to each study are based on the inverse of the
variance.
We used the Q test (chi-square statistics) with an a of
0.1 to test heterogeneity among pooled estimates.
When there was statistically significant heterogeneity
in outcomes across studies, we conducted sensitivity
analyses according to each of the four parameters of
trial methodological quality. A priori subgroup analysis
was planned based on factors that could potentially
influence the magnitude of the treatment effect:

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372
367
(i) definition of diarrhoea; (ii) age of participants –
adults vs. children; (iii) dose of S. boulardii and (iv)
duration of follow-up.
To test for publication bias, we used a test for
asymmetry of the funnel plot proposed by Egger
et al.15 This test detects funnel plot asymmetry by
determining whether the intercept deviates significantly
from zero in a regression of the normalized effect
estimate (estimate divided by its standard error) against
precision (reciprocal of the standard error of the
estimate) weighted by the reciprocal of the variance of
the estimate.
RESULTS
The meta-analyses reported here are presented
according to the standards set out in the 1999
Quality of Reporting of Meta-analyses (QUOROM)
statement.16
Description of studies
The search yielded 16 citations. Five RCTs met the
inclusion criteria for this systematic review11, 17–20 (see
Table 1). The remaining 11 studies were excluded.21–31
Table 2 summarizes characteristics of the excluded
trials, including the reasons for exclusion.
The five selected studies recruited a total of 1076
participants (564 in the experimental group and 512 in
the control group). All studies were placebo controlled.
There was considerable clinical heterogeneity among
the trials in settings (in-patients or out-patients), age of
participants (except one, all were performed in adults),
daily dose of the study product (200 mg–1 g) and the
type of antibiotics administered. There were also wide
differences in the duration of follow-up, which varied
from 211, 18 to 7 weeks20 after the cessation of antibi-
otic treatment or was not specified.17, 19 Furthermore,
there was variability in definitions of outcome meas-
ures. The most commonly used definition of the
diarrhoea was the presence of three or more loose (or
watery) stools per 24 h, but criteria for its duration
varied from 24 h (two studies)17, 19 to at least 48 h
(three studies).11, 18, 20
The methodological quality of the trials also varied
(Table 1). Allocation concealment was unclear in all of
the trials. Although all were double-blind studies, it
often was not stated who was blinded. Completeness of
follow-up was unclear in one study19 and inadequate in
 Table 1. Characteristics of the included RCTs
368

Number of pati-
ents with AAD
Allocation Age Dose Duration of Follow- Antibiotic Definition of
Trial concealment1 Blinding ITT2 FU3 N Participants (years) (per day) intervention up studied diarrhoea or AAD SB Placebo

Adam Unclear DB No No 388 Out-patients >15 200 For the duration No data b-Lactam ‡2 stools/day, 9/199 33/189
et al.17 mg of antibiotic TX antibiotics or liquid consistency
(minimum 5 days; tetracycline (oral) (as estimated
experimental group: from the table)
6.8 ± 0.24 days;
control group:
6.84 ± 0.25 days)
Surawicz Unclear DB No No 180 Hospitalized Adults 1g Within 48 h of 2 weeks (?) Various ‡3 loose or watery 11/116 14/64
et al.18 antibiotic initiation (penicillin, stools/day for
and 2 weeks after clindamycin, at least 2 days
(exact data not given) cephalosporins)
McFarland Unclear DB Yes Yes 193 Hospitalized 18–86 1g Within 72 h of 7 weeks b-lactam ‡3 loose stools/ 7/97 14/96
H. SZAJEWSKA & J. MRUKOWICZ

et al.20 antibiotic initiation antibiotics day for at least 2

and 3 days after the (oral or consecutive days;
antibiotic was intravenous) AAD – diarrhoea
discontinued associated with
(exact data not given) at least one
b-lactam antibiotic
with no other
aetiology of
diarrhoea identified

Lewis Unclear DB No Yes 72 Hospitalized >65 226 For the duration of No data Various ‡3 loose stools 7/33 5/36
et al.19 mg antibiotic treatment within a 24-h period
(exact data not given)
Kotowska Unclear DB Yes Yes 269 Hospitalized 6 months– 500 For the duration 2 weeks Various Diarrhoea: ‡3 loose or 4/119 22/127
et al.11 and 14 years mg of antibiotic treatment watery stools per day
out-patients (experimental for a minimum of 48 h
group: 7.8 ± 1 days; during and/or up to
control group: 2 weeks after the end
8.1 ± 1 days) of antibiotic treatment
AAD: as above, caused
by Clostridium difficile or
for otherwise unexplained
diarrhoea

1
Adequate: randomization method described that would not allow investigator/caregivers to identify or influence the intervention group before eligible participants entered the study; Unclear:
randomization stated but no information about method used was provided; Inadequate: use of an inappropriate method of randomization (e.g. alternate medical record numbers or unsealed envelopes)
and/or any information in the study indicating that investigators or participants could influence the intervention group.
2
Yes: specifically reported by authors that ITT analysis was undertaken and this was confirmed by our study assessment; Yes: not stated but confirmed by our study assessment; No: not reported and
lack of ITT analysis confirmed by our study assessment (patients who were randomized were not included in the analysis because they did not receive the study intervention, they withdrew from the
study or were not included because of protocol violation); No: stated but not confirmed by our study assessment.
3
Completeness of follow-up: trials with >80% follow-up of participants.
AAD, antibiotic-associated diarrhoea; DB, double blinding; SB, Saccharomyces boulardii; TX, treatment; RCT, randomized-controlled trial; ITT, intention-to-treat; FU, follow-up.

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372

 META-ANALYSIS: S. BOULARDII IN THE PREVENTION OF DIARRHOEA 369

Table 2. Characteristics of the excluded studies

Study Reason(s) for exclusion

Buts et al.30 Saccharomyces boulardii for Clostridium difficile-associated enteropathies in infants

Donat24 Non-randomized, controlled clinical trial
Elmer and McFarland25 RCT; designed to clarify the effectiveness of S. boulardii in recurrent C. difficile disease.
Not prevention study
Erdeve et al.26 Non-randomized, prospective controlled trial
Hotz21 Non-randomized, clinical trial
McFarland et al.29 RCT; explored the effect of S. boulardii in combination with standard antibiotics
for C. difficile disease. Not prevention study
Potts et al.23 RCT (abstract). Later reported as a full paper19
Schellenberg et al.31 Open trial. Treatment of C. difficile with S. cerevisiae
Surawicz et al.22 RCT; explored the effect of S. boulardii in the treatment of C. difficile colitis (not prevention study)
Surawicz et al.28 RCT; explored the effect of S. boulardii in the treatment of C. difficile disease

RCT, randomized-controlled trial.

another.18 ITT analysis was performed in only two S. boulardii in the prevention of C. difficile diarrhoea
trials.11, 20
Only one study11 (performed in children) evaluated the
effect of S. boulardii in the prevention of C. difficile
S. boulardii in the prevention of diarrhoea diarrhoea. The risk of documented C. difficile diarrhoea
was lower in the S. boulardii group compared with the
Treatment with S. boulardii compared with placebo
placebo group, but the difference was of borderline
reduced the risk of diarrhoea in patients treated with
statistical significance (RR: 0.3; 95% CI: 0.1–1.04).
antibiotics (as defined by the study investigators) from
17.2% to 6.7% (RR: 0.43; 95% CI: 0.23–0.78, random
effect model; Figure 1). For every 10 patients receiving Mean frequency of bowel movements and mean duration of
daily S. boulardii with antibiotics, one fewer would diarrhoea
develop diarrhoea (NNT: 10; 95% CI: 7–16).
These outcome measures were either not reported in the
The pooled effect size of three trials11, 18, 20 that used a
studies included in this systematic review11, 17, 19 or
more conservative definition of diarrhoea was 0.38
were reported in a manner that does not allow meta-
(95% CI: 0.22–0.63, random effect model; NNT: 10;
analysis.18, 20
95% CI: 7–18).

Relative risk meta-analysis plot (random effects)

Adam 0.26 (0.13. 0.52)

Surawicz 0.43 (0.21. 0.89)

McFarland 0.49 (0.21. 1.14)

Lewis 1.53 (0.56. 4.20)

Kotowska 0.19 (0.07. 0.52)

Figure 1. Plot of relative risk of antibiotic- Combined [random] 0.43 (0.23. 0.78)
associated diarrhoea in patients treated
with Saccharomyces boulardii compared with 0.01 0.1 0.2 0.5 1 2 5
placebo. Relative risk (95% confidence interval)

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372

370 H. SZAJEWSKA & J. MRUKOWICZ
Need for discontinuation of the antibiotic treatment,
hospitalization to manage the diarrhoea (in out-patients)
and/or intravenous rehydration in any of the study groups
Only one trial11 (performed in children) addressed these
outcomes. There was no need for discontinuation of
antibiotic treatment, hospital treatment because of
diarrhoea in the out-patients, or intravenous rehydra-
tion in any of the study groups.
Adverse events
The S. boulardii was well-tolerated, and no adverse
events associated with this therapy were reported.
Homogeneity test
The statistical test of homogeneity yielded a significant
result (v2 ¼ 10; P < 0.04). Significant heterogeneity
was attributable to the inclusion of elderly participants
in one of the trials (all participants were >65 years).19
Exclusion of this trial by Lewis et al.19 resulted in a
homogenous group of four studies involving 1007
patients (v2 ¼ 2.9; P ¼ 0.4). The significance of the
pooled effect of S. boulardii remained when calculations
were made with four homogenous trials only (RR: 0.33;
95% CI: 0.22–0.5, random effect model).
Publication bias
A funnel plot (Figure 2) and Egger et al.’s15 regression
asymmetry test (P ¼ 0.5 and 95% CI included 0) did not
show any publication bias or other small sample bias.
Bias assessment plot
0.35
0.40
Standard error
0.45
0.50
0.55
-2 0 -1.5 -1.0 -0.5
Log (relative risk)
Figure 2. Plot of publication bias.
DISCUSSION
Meta-analysis of the various studies assessing the
effectiveness of S. boulardii in preventing AAD is difficult,
as these studies are heterogeneous and the definition of
diarrhoea is not consistent. Consequently, this assess-
ment is imperfect. However, consistent with previous
findings, this meta-analysis of data from five RCTs
showed that S. boulardii is effective in preventing AAD in
adults and children treated with antibiotics for any
reason (mainly for respiratory tract infections). For
every 10 patients receiving daily S. boulardii with
antibiotics, one fewer will develop AAD. One charac-
teristic that makes our meta-analysis distinct from
previous reviews is that it focuses exclusively on only
one probiotic strain. Misuse or unsubstantiated use of
probiotics would be limited if more emphasis was given
to analysis by probiotic agent, as in the case of our
meta-analysis on S. boulardii.
We made every effort to avoid publication bias,
including performing a comprehensive search using a
range of databases. We also reviewed references from
retrieved articles. As only a limited number of experts in
the field were contacted for unpublished material,
publication bias cannot be fully ruled out. However,
based on the results of bias detection analysis (funnel
plot and statistical test for its asymmetry), we believe
that the risk of not having identified important,
randomized, methodologically relevant trials is rather
low. Nevertheless, it is important to stress that the
power of the statistical methods that investigate evi-
dence for funnel plot asymmetry is limited, particularly
for meta-analyses based on a small number of studies.
Thus, the results from our meta-analysis should be
viewed with caution. Similarly, given the small number
of studies, statistical conclusions on determinants of
heterogeneity might be flawed.
The mechanism by which S. boulardii, a non-patho-
genic yeast, exerts its action in preventing AAD is
unclear. Possible mechanisms, which have been dem-
onstrated in animals, include the production of a
protease that inactivates a receptor for toxin A of
C. difficile, secretion of increased levels of secretory
immunoglobulin A (IgA) and IgA antitoxin A, and
competition for attachment sites.32–34 Saccharomyces
boulardii has also been shown to block C. difficile
0.0 0.5 adherence to cells in vitro.35
The results of this meta-analysis show that S. boulardii
significantly reduces the risk of diarrhoea in patients

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372
 META-ANALYSIS: S. BOULARDII IN THE PREVENTION OF DIARRHOEA
treated with antibiotics for various purposes (but mainly
respiratory tract infections). However, they do not allow
conclusions about the efficacy of S. boulardii in prevent-
ing diarrhoea attributable to any single antibiotic class.
Results from one recent RCT suggest that S. boulardii
effectively prevents diarrhoea caused by amoxicillin in
combination with clavulanate as well as intravenously
administered cefuroxime.11 Larger trials will be neces-
sary to further address these issues.
The duration of follow-up after antibiotic treatment in
the trials included in this meta-analysis varied. As
diarrhoea may occur up to 2 months following cessa-
tion of such treatment, some cases of AAD may have
been missed. In addition, only one of these trials
(performed in children) assessed the prevalence of
serious complications of AAD, such as the need to
terminate antibiotic treatment or hospitalize the patient
for management of diarrhoea and/or intravenous
rehydration.11 These authors observed no such cases
in children, thus, the baseline risk in this age group
appears to be very low. Unfortunately, these important
outcomes have not been investigated in trials performed
in adults.
Whereas no adverse effects were observed in any of the
included trials, the administration of S. boulardii is not
without risk. One caveat about S. boulardii is that it can
cause fungemia.36–39 Most complications have occurred
in immunocompromised subjects or in patients with
other life-threatening illnesses managed in intensive
care units. Although other probiotics (e.g. lactobacilli,
bifidobacteria) so far used in clinical trials have
generally been described as safe and well-tolerated, it
is not clear whether they could be used in the
prevention of AAD in immunocompromised patients;
as there have been some concerns expressed, this
requires further evaluation.
In conclusion, the results from this meta-analysis of
five RCTs support the value of S. boulardii as adjunctive
treatment in patients on antibiotic therapy. In addition
to ease of administration, coadministration of S. boular-
dii with antibiotics has potential cost–benefits. However,
some issues need to be resolved before widespread use of
probiotics is advocated, e.g. by means of identifying
populations at high risk for AAD. There is a need for
well designed RCTs to evaluate the efficacy of S. boulardii
in preventing AAD caused specifically by C. difficile or
those antibiotics that are most likely to cause diarrhoea,
as well as the effectiveness of S. boulardii in treating
AAD. Although available data are encouraging, it

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372
371
seems that the judicious use of antibiotics remains the
best method of preventing AAD.
ACKNOWLEDGEMENT
This study was funded in part by a grant from The
Medical University of Warsaw.
REFERENCES
1 Bartlett JG. Antibiotic-associated diarrhoea. N Engl J Med
2002; 346: 334–9.
2 Bartlett JG, Chang TW, Gurwith M, Gorbach SL, Onderdonk
AB. Antibiotic-associated pseudomembranous colitis due to
toxin producing clostridia. N Engl J Med 1978; 298: 531–4.
3 Wiström J, Norrby SR, Myhre EB, et al. Frequency of
antibiotic-associated diarrhoea in 2462 antibiotic-treated
hospitalized patients: a prospective study. J Antimicrob
Chemother 2001; 47: 43–50.
4 McFarland LV. Epidemiology, risk factors and treatments for
antibiotic-associated diarrhoea. Dig Dis 1998; 16: 292–307.
5 Turck D, Bernet JP, Marx J, et al. Incidence and risk factors of
oral antibiotic-associated diarrhoea in an outpatient pediatric
population. J Pediatr Gastroenterol Nutr 2003; 37: 22–6.
6 Elstner CL, Lindsay AN, Book LS, Matsen JM. Lack of
relationship of Clostridium difficile to antibiotic-associated
diarrhoea in children. Pediatr Infect Dis 1983; 2: 364–6.
7 Barbut F, Meynard JL, Guiguet M, et al. Clostridium difficile-
associated diarrhoea in HIV infected patients: epidemiology and
risk factors. J Acquir Immune Defic Syndr 1997; 16: 176–81.
8 McFarland LV, Surawicz CM, Stamm WE. Risk factors for
Clostridium difficile carriage and C. difficile-associated diarrhoea
in a cohort of hospitalized patients. J Infect Dis 1990; 162:
678–84.
9 Diplock AT, Aggett PJ, Ashwell M, et al. Scientific concepts of
functional foods in Europe: consensus document. Br J Nutr
1999; 81 (Suppl. 1): S1–S27.
10 Surawicz CM. Probiotics, antibiotic-associated diarrhoea and
Clostridium difficile diarrhoea in humans. Best Pract Res Clin
Gastroenterol 2003; 17: 775–83.
11 Kotowska M, Albrecht P, Szajewska H. Saccharomyces boulardii
in the prevention of antibiotic-associated diarrhoea in
children: a randomized double-blind placebo-controlled trial.
Aliment Pharmacol Ther 2005; 21: 583–90.
12 D’Souza AL, Rajkumar C, Cooke J, Bulpitt CJ. Probiotics in
prevention of antibiotic associated diarrhoea: meta-analysis.
Br Med J 2002; 324: 1361–4.
13 Cremonini F, di Caro S, Nista EC, et al. Meta-analysis: The
effect of probiotic administration on antibiotic-associated
diarrhoea. Aliment Pharmacol Ther 2002; 16: 1461–7.
14 Robinson KA, Dickersin K. Development of a highly sensitive
search strategy for the retreival of reports of controlled trials
using PubMed. Int J Epidemiol 2002; 31: 150–3.
15 Egger M, Smith GD, Schneider M, Minder C. Bias in meta-
analysis detected by a simple, graphical test. Br Med J 1997;
315: 629–34.
372 H. SZAJEWSKA & J. MRUKOWICZ
16 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF,
for the QUOROM Group. Improving the Quality of Reports of
Meta-analyses of randomised controlled trials: the QUOROM
statement. Lancet 1999; 354: 1896–900.
17 Adam J, Barret A, Barret-Bellet C. Essais cliniques contrôlés en
double insu de l’ultra-levure lyophilisée: étude multicentrique
par 25 médicins de 388 cas. Gaz Med Fr 1977; 84: 2072–8.
18 Surawicz CM, Elmer GW, Speelman P, McFarland LV, Chinn J,
van Belle G. Prevention of antibiotic-associated diarrhoea by
Saccharomyces boulardii: a prospective study. Gastroenterology
1989; 96: 981–8.
19 Lewis SJ, Potts LF, Barry RE. The lack of therapeutic effect of
Saccharomyces boulardii in the prevention of antibiotic-related
diarrhoea in elderly patients. J Infect 1998; 36: 171–4.
20 McFarland LV, Surawicz CM, Greenberg RN, et al. Prevention
of beta-lactam-associated diarrhoea by Saccharomyces boulardii
compared with placebo. Am J Gastroenterol 1995; 90: 439–
48.
21 Hotz J. Prevention of antibiotic-associated diarrhoea by
Saccharomyces boulardii: a prospective study. Z Gastroenterol
1990; 28: 259–60 (German).
22 Surawicz CM, McFarland LV, Elmer G, Chinn J. Treatment of
recurrent Clostridium difficile colitis with vancomycin and
Saccharomyces boulardii. Am J Gastroenterol 1989; 84:
1285–7.
23 Potts L, Lewis SJ, Barry R. Randomised double blind placebo
controlled study of the ability of Saccharomyces boulardii to
prevent antibiotic related diarrhoea [abstract]. Gut 1996; 38
(Suppl. 1): A61.
24 Donat F. Prevention of intestinal side-effects of antibiotics with
perenterol. Munch Med Wochenschr 1970; 112: 74–7
(German).
25 Elmer GW, McFarland LV. Suppression by Saccharomyces
boulardii of toxigenic Clostridium difficile overgrowth after
vancomycin treatment in hamsters. Antimicrob Agents
Chemother 1987; 31: 129–31.
26 Erdeve O, Tiras U, Dallar Y. The probiotic effect of
Saccharomyces boulardii in a pediatric age group. J Trop
Pediatr 2004; 50: 234–6.
27 Schellenberg D, Bonington A, Champion CM, Lancaster R,
Webb S, Main J. Treatment of Clostridium difficile diarrhoea
with brewer’s yeast. Lancet 1994; 343: 171–2.

2005 Blackwell Publishing Ltd, Aliment Pharmacol Ther 22, 365–372
28 Surawicz CM, McFarland LV, Greenberg RN, et al. The search
for a better treatment for recurrent Clostridium difficile disease:
use of high-dose vancomycin combined with Saccharomyces
boulardii. Clin Inf Dis 2000; 31: 1012–7.
29 McFarland LV, Surawicz CM, Greenberg RN, et al. A
randomized placebo-controlled trial of Saccharomyces
boulardii in combination with standard antibiotics for
Clostridium difficile disease. JAMA 1994; 271: 1913–8.
Erratum in: JAMA 1994; 272: 518.
30 Buts JP, Corthier G, Delmee M. Saccharomyces boulardii
for Clostridium difficile-associated enteropathies in infants.
J Pediatr Gastroenterol Nutr 1993; 16: 419–25.
31 Schellenberg D, Bonningotn A, Champion CM, Lancaster R,
Webb S, Main J. Treatment of Clostridium difficile diarrhoea
with brewer’s yeast. Lancet 1994; 343: 171–2.
32 Pothoulakis C, Kelly CP, Joshi MA, et al. Saccharomyces
boulardii inhibits Clostridium difficile toxin A binding and
enterotoxicity in rat ileum. Gastroenterology 1993; 104:
1108–15.
33 Wilson KH, Perini I. Role of competition for nutrients in
suppression of Clostidium difficile by the colonic microflora.
Infect Immun 1988; 56: 2610–4.
34 Qamar A, Aboudola A, Warny M, et al. Saccharomyces
boulardii stimulates intestinal immunoglobulin A immune
response to Clostridium difficile toxin A in mice. Infect Immun
2001; 69: 2762–5.
35 Tasteyre A, Barc MC, Karjalainen T, et al. Inhibition of in vitro
cell adherence of Clostridium difficile by Saccharomyces
boulardii. Microbiol Pathogens 2002; 32: 219–25.
36 Bassetti S, Frei R, Zimmerli W. Fungemia with Saccharomyces
cerevisiae after treatment with S. boulardii. Am J Med 1998;
105: 71–2.
37 Rijnders BJA, Van Wijngaerden E, Verwaest C, Peetermans
WE. Saccharomyces fungemia complicating S. boulardii
treatment in a non-immunocompromised host. Intensive
Care Med 2000; 26: 825.
38 Zunic P, Lacotte J, Pegoix M, et al. S. boulardii fungemia.
Apropos of a case. Therapie 1991; 46: 498–9.
39 Cassone M, Serra P, Mondello F, et al. Outbreak of Saccharomyces
cerevisiae subtype boulardii fungemia in patients neighboring
those treated with a probiotic preparation of the organism. J Clin
Microbiol 2003; 41: 5340–3.