American Journal of Gastroenterology ISSN 0002-9270


C 2006 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2006.00465.x
Published by Blackwell Publishing

Meta-Analysis of Probiotics for the Prevention of

Antibiotic Associated Diarrhea and the Treatment of
Clostridium difficile Disease
Lynne V. McFarland, Ph.D.1,2
1
Department of Health Services Research and Development, Veterans Administration Puget Sound Health Care
System, Seattle, Washington, and 2 Department of Medicinal Chemistry, University of Washington, Seattle,
Washington

CONTEXT: Antibiotic-associated diarrhea (AAD) is a common complication of most antibiotics and

Clostridium difficile disease (CDD), which also is incited by antibiotics, is a leading cause of
nosocomial outbreaks of diarrhea and colitis. The use of probiotics for these two related
diseases remains controversial.
OBJECTIVE: To compare the efficacy of probiotics for the prevention of AAD and the treatment of CDD based
on the published randomized, controlled clinical trials.
DATA SOURCES: PubMed, Medline, Google Scholar, NIH registry of clinical trials, metaRegister, and Cochrane
Central Register of Controlled Trials were searched from 1977 to 2005, unrestricted by
language. Secondary searches of reference lists, authors, reviews, commentaries, associated
diseases, books, and meeting abstracts.
STUDY SELECTION: Trials were included in which specific probiotics given to either prevent or treat the diseases of
interest. Trials were required to be randomized, controlled, blinded efficacy trials in humans
published in peer-reviewed journals. Trials that were excluded were pre-clinical, safety, Phase 1
studies in volunteers, reviews, duplicate reports, trials of unspecified probiotics, trials of
prebiotics, not the disease being studied, or inconsistent outcome measures. Thirty-one of 180
screened studies (totally 3,164 subjects) met the inclusion and exclusion criteria.
DATA EXTRACTION: One reviewer identified studies and abstracted data on sample size, population characteristics,
treatments, and outcomes.
DATA SYNTHESIS: From 25 randomized controlled trials (RCTs), probiotics significantly reduced the relative risk of
AAD (RR = 0.43, 95% CI 0.31, 0.58, p < 0.001). From six randomized trials, probiotics had
significant efficacy for CDD (RR = 0.59, 95% CI 0.41, 0.85, p = 0.005).
CONCLUSION: A variety of different types of probiotics show promise as effective therapies for these two
diseases. Using meta-analyses, three types of probiotics (Saccharomyces boulardii, Lactobacillus
rhamnosus GG, and probiotic mixtures) significantly reduced the development of
antibiotic-associated diarrhea. Only S. boulardii was effective for CDD.
(Am J Gastroenterol 2006;101:1–11)

Antibiotic-associated diarrhea (AAD) is a common com- rier, patients are susceptible to infection by opportunistic
plication of antibiotic use. The frequency of AAD can be pathogens.
high (26–60%) during hospital outbreaks or moderate (13– Hospitalized patients exposed to antibiotics may develop
29%) during endemic periods and is relatively infrequent in Clostridium difficile disease (CDD). Recently, this pathogen
outpatient settings (<0.1%) (1–3). Risk factors for AAD in- caused nosocomial outbreaks in Canadian hospitals that re-
clude broad-spectrum antibiotics, host factors (age, health sulted in over 100 deaths (6). Consequences of AAD and
status, gender), hospitalization status, and exposure to noso- CDD include extended hospital stays (3–7 days), increased
comial pathogens (2, 4). AAD usually occurs 2–8 wk af- rates of subsequent infections (20–65%), higher hospital
ter exposure to antibiotics as a result of disrupting intesti- costs (up to $1 billion/yr) and 2–3 times increased rates of
nal microflora. One of the roles of intestinal microflora mortality (7–11).
is to act as a protective barrier that resists the coloniza- Current therapies for AAD are lacking. Strategies to date
tion of intestinal pathogens (5). Without this protective bar- include discontinuing the inciting antibiotic, restricting the
1
2 McFarland

use of high-risk antibiotics and specific antibiotic treatments CDD is a new episode of C. difficile positive diarrhea within
if the etiology is known. For CDD, 80% respond well to 1 month of a previous CDD episode (19). Documentation of
the initial treatment of vancomycin or metronidazole. The diarrhea is based on clinical assessment and self-report of
remaining 20% may develop subsequent episodes of CDD, symptoms by daily symptom diaries.
which may persist over several years, despite repeated antibi-
otic treatments (9). Data Sources
Probiotic therapy is well suited to these two types of micro- PubMed, Medline, and Google Scholar were searched from
bial induced diseases. Probiotics assist in reestablishing the 1977 to 2005 for articles unrestricted by language. Non-
disrupted intestinal microflora, enhancing immune responses English articles were translated. Three on-line clinical
and clearing pathogens and their toxins from the host (12– trial registers were searched: Cochrane Central Register
14). Research using probiotics has been reported for the past of Controlled Trials (http://www.cochrane.org), metaRegis-
28 yr, but the studies have been variable in trial design, type of ter of Controlled Trials (http://www.controlled-trials.com/
probiotic, had differing doses and durations of treatment, and mrct) and National Institutes of Health (http://www.
thus have yielded contradictory results. The lack of definitive clinicaltrials.gov). Secondary and hand searches of reference
evidence regarding efficacy and safety is limiting the use of lists, authors, reviews, commentaries, associated diseases,
this type of treatment strategy. There is a growing interest books, and meeting abstracts were also performed. Six search
in probiotics for the treatment of AAD and CDD due to the terms for RCTs (RCT, human, blinding, Phase 2, Phase 3, ef-
wide availability of probiotics as dietary supplements and the ficacy) were combined with 15 terms for probiotics. Search
concern over recent outbreaks of severe CDD in Canada and terms included probiotics, microflora, antibiotics, Clostrid-
the United Kingdom (6, 15). Two meta-analyses done in 2002 ium difficile, colitis, PMC, diarrhea, Saccharomyces, Lac-
presented results on only 11 trials and failed to discuss pos- tobacilli, Bifidobacteria, Enterococci, Bacilli, VSL#3, syn-
sible reasons for the conflicting findings and did not present biotics, and Lactinex. Search strategies were broad-based
detailed study information (16, 17). A recent Cochrane re- initially, then narrowed to the disease of interest (20). The
view on treatments for CDD did not include probiotics (18). procedure for this meta-analysis was designed as suggested
The need for a meta-analysis of probiotics for these two an- by Egger et al. and MOOSE guidelines using clearly delin-
tibiotic associated diseases is apparent. eated parameters, a priori inclusion and exclusion criteria,
and standardized data extraction methods (21–23).

METHODS Data Extraction

Information on study design, methods, interventions, out-
Objectives comes, adverse effects, and treatments was extracted from
The objectives of this meta-analysis are to assess the efficacy each article. Data on patient inclusion and exclusion crite-
and safety of probiotics for (i) the prevention of AAD and ria, number of completed subjects, attrition, treatment dose
(ii) the treatment of CDD. and duration, and outcome was extracted into a standardized
table. In some cases, the primary or secondary author was
Criteria for Study Selection contacted for data not reported in the original article. The
Abstracts of all citations and retrieved studies were reviewed data abstraction was completed individually, but verified us-
and rated for inclusion. Full articles were retrieved if specific ing historic searches with two other researchers for previous
treatments were given to either prevent or treat the disease review articles (24, 25). A few trials had multiple probiotic
of interest. Inclusion criteria include randomized, controlled, arms with a common control group. Each probiotic arm and
blinded efficacy trials in humans published in peer-reviewed control group was analyzed separately.
journals. Exclusion criteria include pre-clinical studies, case
reports or case series, Phase 1 safety studies in volunteers, Assessment of Methodological Quality
reviews, duplicate reports, trials of unspecified probiotics, Studies that met the inclusion criteria were graded for quality
trials of prebiotics, not the disease being studied, or incon- using a scale reported by the U.S. Preventive Services Task
sistent outcome measures. External and internal validity is Force (26). Quality of evidence is rated from 1 to 3 (poor, fair,
strengthened by including only randomized controlled trials and good) based on randomization, study design, sample size,
(RCTs). generalizability, study biases, and outcome assessment. Study
quality was not integrated with the model weights, as trials of
Outcomes and Definitions poor quality were excluded from review and this practice is
The primary outcome for AAD is defined as diarrhea (≥3 not uniformly recommended (27). Weights for this analysis
loose stools/day for at least 2 days or ≥5 loose stools/48 h) are based solely on sample sizes.
within 2 months of antibiotic exposure (2, 11). The primary
outcome of CDD is defined as a new episode of diarrhea asso- Statistical Analysis
ciated with a positive culture or toxin (A or B) assay within 1 Statistical analysis was performed using Stata software ver-
month exposure to antibiotics. The outcome for prevention of sion 8.0 (Stata Corporation, College Station, TX). Relative
 Prevention of Antibiotic Associated Diarrhea 3

risks with 95% confidence intervals were computed as sum- CDD. The literature search yielded 940 citations, of which
mary statistics. Heterogeneity across trials was evaluated us- 76 were selected from retrieval. Only six (8%) of the screened
ing Cochran Q test based on pooled relative risks by the articles met inclusion criteria and provided data on 354 treated
Mantel-Haenszel method. If the studies were homogeneous, patients with CDD. The number of patients in each of these
a fixed-effects model was used and a pooled relative risk was studies was generally small (median, 25; range 15–138).
calculated with the Mantel-Haenszel method for fixed effects.
If the studies were heterogeneous a random effect was em- Excluded Studies
ployed and a pooled relative risk was calculated using the AAD. Of the AAD studies, 79 failed to meet one or more of
DerSimonian and Laird method (28). If significant hetero- the inclusion criteria. Most were reviews or commentaries (n
geneity was detected, a subgroup analysis was conducted. A = 57), pre-clinical or Phase 1 safety studies done in healthy
priori subgroups were by type of probiotic, dose and indi- volunteers (n = 11) or had no control group (n = 4). AAD
cation (AAD or CDD). A funnel plot as well as an adjusted was a post hoc outcome in three trials and one study failed to
rank correlation test using the Begg and Mazumdar method provide outcome data. Some studies were excluded because
were used to assess publication bias (29, 30). p Values less the type of probiotic was not specified (31, 32) or only a
than 0.05 were considered significant. prebiotic (oligosaccharide with no living probiotic) was given
as the intervention (33).
CDD. Of the CDD studies, 70 failed to meet one or more of
RESULTS the inclusion criteria. Most were reviews or commentaries (n
= 41), pre-clinical or Phase 1 safety studies done in healthy
Overview of Included Studies
volunteers (n = 7) or had no control group (n = 18). Some
AAD. The literature search yielded 940 citations, of which
studies were excluded because CDD was a post hoc outcome
104 were selected from retrieval. Twenty-five (24%) of the
(n = 3) or only a prebiotic (oligosaccharide with no living
screened articles met inclusion criteria and provided data on
probiotic) was given as the intervention (34).
2,810 treated patients with AAD. The number of patients in
each of these studies was generally moderate (median, 79;
Study Quality
range 18–388). A QUOROM (Quality of Reporting of Meta-
The quality of the studies is presented in Tables 1 and 3, indi-
analysis) flow diagram (Fig. 1) shows an overview of the
cating generally good methodological quality. Most studies
study selection process (22).
of poor quality were excluded from the data extraction in the
preliminary steps of this study.
940 citations identified by search of
electronic databases and hand Efficacy Studies
searchs
AAD. Twenty-five RCTs provided adequate data regarding
efficacy in a total of 2,810 patients with AAD, as shown in
Table 1. Of the 25 trials, 13 (52%) reported a significant
760 irrelevant reduction of AAD in the probiotic-treated group compared
reports excluded
with the placebo group in their study. Twelve studies did not
180 Potential Studies Identified and reject the null hypothesis of no difference in the incidence of
screened for Inclusion
AAD for probiotic treated versus controls.
These contradictory results may be due to differences in
the study population enrolled, the type of probiotic, the dose
of probiotic given, or the duration of treatment. Typically,
these trials were done in adults given broad-spectrum antibi-
Antibiotic-associated diarrha (n=104) Clostridium difficile disease (n=76)
47 reviews 31 reviews otics (64%), while 36% of the trials were done in children
4 no controls 18 no controls taking antibiotics. Of 16 RCTs of AAD in adult patients,
10 commentaries 10 commentaries
7 preclinical studies 6 preclinical studies 7 (44%) showed significant efficacy for probiotics. Of nine
3 disease post-hoc outcome 3 disease post-hoc outcome
4 Phase 1 safety trials 1 Phase 1 safety trials RCTs of AAD in children, six (67%) had significant efficacy.
2 probiotic not identified 0 probiotic not identified There was not a significant difference in efficacy of probi-
1 prebiotic substance only 1 prebiotic substance only
1 lacking outcome data 0 lacking outcome data otics according to whether adults or children were enrolled
(Fisher’s p = 0.41). The types of probiotics varied from single
strains (Saccharomyces boulardii, Lactobacillus rhamnosus
GG, Bacillus clausii, Bifidobacterium longum, Clostridium
RCTs included in meta-analysis RCTs included in meta-analysis butyricum miyairir, Lactobacillus acidophilus, Enterococcus
25 of antibiotic associated diarrhea 6 of Clostridium difficile disease faecium SF68), to mixtures of two types of probiotic and to a
synbiotic (a probiotic combined with a pre-biotic substance).
Figure 1. QUOROM flow diagram of included and excluded studies. Daily doses of probiotics ranged from 1 × 107 to 1 × 1011 ,
RCT indicates randomized, controlled trial. with a mean of 3 × 109 . Use of a high dose (≥1010 /day) of
 4
McFarland

Table 1. Description of 25 Randomized Controlled Trials of Probiotics for the Prevention of Antibiotic Associated Diarrhea
Probiotic-Treated Control Group
N Subjects Probiotic Dose/Day Txt Duration Follow-Up No. Failed No. Cured No. Failed No. Cured Quality Weight Reference
388 Adults, b-lactam or tetracycline SB 4 × 109 1 wk 0 9 (4.5) 190 33 (17.5) 156 3 4.9 (35)
180 Adults, varied SB 2 × 1010 On abx 2 wk 11 (9.5) 105 14 (21.8) 50 3 4.9 (36)
193 Adults, b-lactams SB 2 × 1010 4 wk 7 wk 7 (7.2) 90 14 (14.6) 82 3 4.4 (37)
246 Peds, varied SB 1 × 1010 1–2 wk 2 wk 4 (3.4) 115 22 (17.3) 105 3 3.8 (38)
69 Elderly, varied SB 4 × 109 2 wk 0 7 (21) 26 5 (13.9) 31 2 3.8 (39)
43 H. pylori + 3 abx SB 5 × 109 7 days 0 1 (5) 21 6 (30) 15 2 1.7 (40)
119 Peds, varied LGG 4 × 1010 2 wk 12 wk 3 (5) 58 9 (16) 49 3 3.2 (41)
188 Outpatient peds, varied oral abx LGG 1–2 × 1010 10 days 0 7 (7.5) 86 25 (26) 70 3 4.6 (42)
81 Hosp peds, varied LGG 1 × 1010 Varied 0 3 (6.7) 42 12 (33.3) 24 2 3.4 (43)
267 Hosp adults 70% b-lactam/varied LGG 2 × 1010 2 wk 1 wk 39 (29.3) 94 40 (29.9) 94 3 6.1 (44)
42 H. pylori + 3 abx LGG 6 × 109 7 days 0 1 (5) 20 6 (30) 15 2 1.7 (40)
120 H. pylori + 3 abx LGG 1 × 1010 2 wk 0 2 (3.4) 58 16 (26.6) 44 3 2.8 (45)
100 H. pylori + 3 abx BC 6 × 109 2 wk 0 15 (30) 35 17 (34) 33 2 5.4 (46)
20 Adults, clindamycin BL 5 × 1010 21 days 0 4 (40) 6 7 (70) 3 2 4.4 (47)
110 Peds, varied CB 1–4 × 107 6 days 0 6 (7) 77 16 (59) 11 2 4.5 (48)
45 Adults, varied EF 1.5 × 107 7 days 0 2 (8.7) 21 6 (27) 16 2 2/6 (49)
200 Adults with TB EF Ng 8 wk 0 5 (5) 95 18 (18) 82 3 4.1 (50)
27 Adults amoxicillin LA 1.2 × 108 Varied 0 10 (83)∗ 2 10 (67) 5 2 5.9 (51)
79 Hosp adults, ampicillin Lactinex 2 × 109 5 days 0 3 (8.3) 33 9 (21) 34 2 3.2 (52)
38 Peds, amoxicillin Lactinex 2 × 109 10 days 0 10 (66) 5 16 (69.5) 7 2 5.8 (53)
20 Adults, clindamycin LABLa 1 × 1011 21 days 0 2 (20) 8 7 (70) 3 2 3.0 (47)
42 H. pylori + 3 abx LABL 5 × 109 7 days 0 1 (5) 20 6 (30) 15 2 1.7 (40)
77 Children, varied BLST 1 × 107 15 days 15 days 6 (16) 32 12 (31) 27 2 4.3 (54)
98 Children, varied LS FOS 5.5 × 108 + 250 mg 10 days 0 14 (29) 34 31 (62) 19 2 5.7 (55)
18 Infants 1–36 months on antibiotics LABI 6 × 109 7 days 0 3 (37.5) 5 8 (80%) 2 2 4.1 (56)
N = number of subjects with evaluable outcome; SB = Saccharomyces boulardii; LGG = Lactobacillus rhamnosus GG; BC = Bacillus clausii; BL = Bifidobacterium longum; CB = Clostridium butyricum MIYAIRI; EF = Enterococcus
faecium SF68; LA = Lactobacillus acidophilus; LALB = Lactinex = L. acidophilus and L. bulgaricus; LABL = Lactobacillus acidophilus and Bifidobacterium longum; LABLa = Lactobacillus acidophilus and Bifidobacterium lactis;
BLST = Bifidobacterium lactis and Streptococcus thermophilus; LSFOS = Lactobacillus sporogenes and fructo-oligosaccharide; LABI = Lactobacillus acidophilus and Bifidobacterium infantis; Txt = treatment; Abx = antibiotic.
Quality: 1 = poor; 2 = fair; 3 = good (26).
∗
Any gastrointestinal complaint including diarrhea.
 Prevention of Antibiotic Associated Diarrhea 5

Table 2. Meta-Analyses of Relative Risks Stratified by Type of Probiotic for the Prevention of Antibiotic Associated Diarrhea
Probiotic Number of RCT Combined RR 95% CI p Value Type of Model References
Saccharomyces boulardii 6 0.37 0.26, 0.52 <0.0001 Fixed (35–40)
Lactobacillus rhamnosus GG 6 0.31 0.13, 0.72 0.006 Random (41–45)
Single strains of probiotics 6 0.46 0.21, 1.03 0.06 Random (46–51)
Mixtures of two probiotics 7 0.51 0.38, 0.68 <0.0001 Fixed (40, 47, 52–56)
Single strains included: Clostridium butyricum MIYAIRI; Enterococcus faecium SF68; Lactobacillus acidophilus; Bifidobacterium longum; Bacillus clausii; Bifidobacterium
lactis; Streptococcus thermophilus; or Lactobacillus sporogenes.
Mixtures included: Lactinex = L. acidophilus and L. bulgaricus; Lactobacillus acidophilus and Bifidobacterium lactis; Lactobacillus acidophilus and Bifidobacterium infantis.
p Value for null hypothesis that RR = 1.

probiotic was associated with a significant efficacy for AAD. were treating patients with established CDD and the probiotic
Eight (67%) of 12 RCTs with a positive efficacy for AAD was combined with standard antibiotics (either vancomycin
used a high daily dose of probiotic compared with only 2 or metronidazole) for treating CDD. Unfortunately, the type
(17%) of 12 RCTs that showed no significant difference yet or dose of the antibiotic was not randomized along with the
used a high daily dose (p = 0.04). The duration of probiotic probiotic arm for any of the studies. Daily doses of probiotics
treatment also varied widely from 5 days to 8 wk (median ranged from 2 × 1010 to 6 × 1011 , with a mean daily dose of 5
of 2 wk), but the duration of probiotic did not significantly × 1010 . The duration of probiotic treatment also varied from
differ in trials showing protective efficacy compared to no 3 to 5 wk, with a median of 3 wk. The number of trials was too
difference. small to determine if a dose-response or duration-response
Data from 25 RCTs were combinable for a meta-analysis, effect was present.
as they reported frequencies of outcomes in treated and con- Data from six RCTs were combinable for a meta-analysis,
trols (35–56). As the χ 2 test for heterogeneity was 82.5 (p as they reported frequencies of outcome in treated and con-
< 0.001), indicating a low degree of homogeneity between trols. As the χ 2 test for heterogeneity was 4.6 (p = 0.5),
studies, a random-effects model was utilized. The combined indicating a high degree of homogeneity between studies,
efficacy shows probiotics have a significant protective effect a fixed-effects model was utilized. The combined efficacy
for AAD (Fig. 2). The relative risk for AAD was 0.43 (95% shows probiotics have a significant protective effect for CDD
CI 0.31, 0.58), z = 5.4, p < 0.001. A funnel plot (Fig. 3) (Fig. 4). The relative risk for CDD was 0.59 (95% CI 0.41,
may indicate the modest presence of some publication bias 0.85), z = 2.8, p = 0.005. A funnel plot (Fig. 5) indicates
or may reflect the differences due to the type of probiotic. No the absence of publication bias. No significant evidence of
significant evidence of publication bias was found using the publication bias was found using the Begg rank correlation
Begg rank correlation test (z = –1.05, p = 0.29). test (z = 0.56, p = 0.57).
As the efficacy may vary by the probiotic strain being Of the three different probiotics tested for the treatment
tested, additional meta-analyses were done, stratified a priori of CDD, only S. boulardii showed significant reductions in
by the probiotic type. Two single probiotic strains showed recurrences of CDD (57, 58). L. rhamnosus GG and L. plan-
significant efficacy for AAD: S. boulardii and L. rhamnosus tarum 299v did not show significant differences in CDD re-
GG (Table 2), as well as mixtures composed of two different currence rates in probiotic versus control treated groups. The
types of probiotics. The meta-analysis for other single probi- one trial testing a probiotic mixture (L. acidophilus and B.
otic strain preparations than those above was not significantly bifidum) for the prevention of CDD did not show significant
protective for AAD, but as the strains were diverse, clinical efficacy (62).
conclusions should be made with caution. Adverse Events
CDD. Six RCTs provided adequate data regarding efficacy Twenty-six (84%) of the 31 trials presented data on adverse
in a total of 354 patients with CDD, as shown in Table 3 reactions, but five trials did not (35, 48, 52, 60, 62). In 24 tri-
(57–62). Of the six trials, two (33%) reported a significant als, no adverse reactions were associated with the probiotic
reduction of CDD recurrences in the probiotic-treated group treatments. McFarland et al. reported significantly more sub-
compared with the placebo group. Four studies did not reject jects taking S. boulardii reported thirst (9%) or constipation
the null hypothesis of no difference in the incidence of CDD (14%) compared with controls (57). Wullt et al. reported mild
recurrences for probiotic treated versus controls. bloating (25%) or gas (37%) was associated with L. rham-
These contradictory results may be due to differences in nosus GG (60). No cases of bacteremia or fungemia or other
the study population enrolled, the type of probiotic, the dose serious adverse events were reported in the 31 RCTs.
of probiotic given or the duration of treatment. All these tri-
als were done in adults with prior antibiotic exposure and
three studies were done exclusively in patients with recur- COMMENT
rent CDD. The types of probiotics included S. boulardii, L.
rhamnosus GG, L. plantarum 299v, and a mixture of L. aci- Antibiotic-induced diseases present unique treatment chal-
dophilus and Bifidobacterium bifidum. Five of the six RCTs lenges for the health-care provider. Treatment with additional
6 McFarland

N = number of subjects with evaluable outcome; SB = Saccharomyces boulardii; LGG = Lactobacillus rhamnosus GG; LP = Lactobacillus plantarum 299v; V = vancomycin; M = metronidazole, nr = not reported, LABB = Lactobacillus
antibiotics for diarrhea symptoms can worsen the condition

No. Failed No. Cured Quality Weight Reference

(57)
(58)
(59)
(60)
(61)
(62)
by further disrupting the intestinal microflora. Efforts to pre-
vent AAD by restricting antibiotic use in hospitals or reduc-
ing inappropriate antibiotic prescriptions has met with only

51.5
14.7
8.2
12.3
2.0
11.2
limited success (63–65). Probiotics have shown promise in
antibiotic mediated diseases as they are not disruptive of in-
3
testinal microflora and have multiple mechanisms of action in
2
2
2
2
3
which to combat opportunistic pathogens in situ (12, 24). Ac-
ceptance of probiotics in routine formularies has been slow
due to the lack of a consensus on the efficacy and safety of
37
7
9
3
6
63 probiotics.
The result of the literature search on probiotics found that
the majority of articles (54%) screened for inclusion were
30 (44.8)

5 (35.7)

1 (14.3)
6 (8.6)

reviews or commentaries. Unfortunately, the prevalence of

7 (50)

6 (67)

RCTs is not frequent in this area. Only 31 (17%) were in-

cluded in this meta-analysis. This is the largest number of
Probiotic-Treated Control Group

RCT analyzed by meta-analysis to date.

No. Cured
(recurred)

In this meta-analysis of 31 randomized, controlled trials,

42
15
7
7
5
67

a clear message is seen supporting the use of probiotics for

two antibiotic-associated diseases. Probiotics given for the
prevention of AAD have a pooled relative risk of 0.43 (0.31,
0.58), using a random-effects model. This is similar to pooled
No. Failed
(recurred)

3 (16.7)
4 (36.4)

3 (37.5)

estimates of risk from three earlier meta-analysis based on

15 (26)

2 (2.9)
4 (36)
Table 3. Description of Six RCT of Probiotics for the Treatment or Prevention of Clostridium difficile Disease

fewer (5–9) RCTs (16, 17, 66). D’Souza et al. pooled nine
trials and found probiotics significantly reduced the odds of
AAD (OR = 0.37, 95% CI 0.26, 0.52), but did not provide
data on heterogeneity testing or publication bias (16). Cre-
Duration Follow-Up

monini et al. analyzed seven trials and found a pooled rela-

4 wk
4 wk

4 wk
0
0

tive risk of 0.40 (95% CI 0.27, 0.57) (17). Although it was

not stated if this was a random- or fixed-effects model, no
significant heterogeneity was found (p = 0.42) and no pub-
38 days

20 days
4 wk
4 wk
3 wk

3 wk
Txt

lication bias was seen in a funnel plot. Publication bias was

acidophilus and Bifidobacterium bifidum; nr = not reported; Quality: 1 = poor; 2 = fair; 3 = good (26).

not assessed by either Begg’s or Egger’s test in this meta-

Probiotic Dose Antibiotic Dose

analysis. Szajewska and Mrukowicz analyzed five trials and

Per Day

found a pooled relative risk of 0.43 (95% CI 0.23, 0.78) us-

Varied

Varied
None
2g
nr
nr

ing a random-effects model (66). No significant publication

bias was found. However, this meta-analysis was restricted to
one type of probiotic (S. boulardii) and did not examine other
types of probiotics. These three meta-analyses were small and
2 × 1010
2 × 1010

5 × 1010
6 × 1011
2 × 1010

did not consistently provide full information on heterogene-

Per Day

nr

ity and publication bias. Despite these limitations, the three

meta-analyses demonstrated a significant efficacy of probi-
otics for the prevention of AAD and validated our findings.
25 Adults CDD/RCDD LGG + V or M

LGG + V or M

The current meta-analysis included a large number of RCTs

LP 299v + M
124 Adults CDD/RCDD SB + V/M

LABB, no
Probiotic

SB + V

and fully described potential biases.

V or M

The etiologies of AAD are diverse and largely not identi-

fied. Approximately one-third of AAD is due to C. difficile,
while another 10–20% is due to bacterial and viral etiologies
(2). Of the 25 RCTs in this analysis, only 9 (36%) attempted
138 Inpatients varied
Adults RCDD

Adults RCDD
Adults RCDD

to determine the etiologies of AAD in their trials. Only four

antibiotics
Subjects

trials reported treatment-specific efficacies stratified on C.

difficile status and none were significant (36–38, 43). This is
not surprising, as the original trials were powered for AAD
and not C. difficile AAD. As the proportion of C. difficile is
variable and may only account for one-third of the enrolled
32

20
15
N

patients, trials for the prevention of C. difficile AAD typically

 Prevention of Antibiotic Associated Diarrhea 7

Figure 2. Forest Plot of 25 randomized controlled trials of probiotics for the prevention of antibiotic associated diarrhea showing crude and
pooled risk ratios. SB = Saccharomyces boulardii; LGG = Lactobacillus rhamnosus GG; BC = Bacillus clausii; BL = Bifidobacterium
longum; CB = Clostridium butyricum MIYAIRI; EF = Enterococcus faecium SF68; LA = Lactobacillus acidophilus; LALB = Lactinex
= L. acidophilus and L. bulgaricus; LABL = Lactobacillus acidophilus and Bifidobacterium longum; LABLa = Lactobacillus acidophilus
and Bifidobacterium lactis; BLST = Bifidobacterium lactis and Streptococcus thermophilus; LSFOS = Lactobacillus sporogenes and
fructo-oligosaccharide; LABI = Lactobacillus acidophilus and Bifidobacterium infantis.

have not been done due to the large sample sizes required. trials in the literature to analyze the efficacy of probiotics
As an alternative, probiotic treatment trials for patients with for carriers versus diseased patients or by the history of the
existing CDD have usually been done. Of the six RCTs, five patient (initial cases compared to recurrent CDD), although
were for treatment and only one was for the prevention of these types of trials would be useful.
CDD. The pooled relative risk from this meta-analysis for The most frequent limitation of these RCTs was that stud-
CDD was 0.59 (0.41, 0.85). As no significant heterogeneity ies may have suffered from insufficient power to detect a
was found (p = 0.5) even for this limited number of trials, a significant difference. Few studies reported sample size cal-
fixed effect model was used. Both a funnel plot and Begg’s culations in their methods section and three authors reported
test did not find significant publication bias (p = 0.57). The that slow recruitment caused premature termination of the
CDD meta-analysis relies heavily upon two studies done by trial (56, 59, 62). Calculating a mean sample size based on
the author, but the potential for bias has been limited by in- a 50% reduction of AAD for probiotic treated versus 37%
cluding only trials published in peer-reviewed journals and AAD in controls (mean taken from 25 AAD trials) with an
by the use of quantitative outcomes. The trials were weighted alpha of 0.05 and a power of 80%, the required number of
by study size and not a qualitative measure of quality, which patients would be 204 per trial. Few (3, 10%) of the 31 RCTs
has the potential for bias, especially if the author is evaluating reached this enrollment goal. Future trials need to calculate
their own studies. We did not find any other meta-analyses of required sample sizes before initiating the study and strive to
probiotics for CDD for comparison. There are also too few recruit sufficient numbers of patients.
8 McFarland

5.29556 3.84812

1/SE(Effect size)
1/SE(Effect size)

.965117 .968787
.121988 1.52727 .333333 2.625
RR RR

Figure 3. Funnel Plot of 25 randomized controlled trials of probi- Figure 5. Funnel Plot of six randomized controlled trials of probi-
otics for antibiotic associated diarrhea. Dashed line indicates pooled otics for Clostridium difficile associated disease. Dashed line indi-
relative risk of 0.43. cates pooled relative risk of 0.59.

Heterogeneity between studies can be a limiting factor for

meta-analyses. It may arise from differences in study popu- do not provide strain designations. A few studies (excluded
lations, type of probiotic being investigated or differences in from the analysis) even failed to provide the identity of the
probiotic doses and duration of treatment. For AAD trials, probiotic and only stated the treatment was “living yogurt” or
the heterogeneity may be due to the different populations en- “protective lactobacilli.” Future studies need to provide the
rolled, as both adults and children given a variety of different complete identity of the probiotic being tested.
types of antibiotics were studied. However, the efficacy was Trials for AAD failing to show significant efficacy may
not found to differ for adult and pediatric subjects. For RCT have used sub-therapeutic doses of probiotics (<1010 organ-
involving CDD, all trials enrolled adults exposed to antibi- isms/day) or failed to provide the probiotic during the entire
otics. period of susceptibility when normal intestinal microflora
Another source of heterogeneity for probiotic trials is the is becoming reestablished (usually 6–8 wk) (2). This meta-
type of probiotic itself. Significant differences in effective- analysis found a dose-response for probiotics used to prevent
ness have been reported for different species and strains of AAD. Trials for the treatment of CDD were more homoge-
similar species of bacteria and yeasts (24, 67, 68). Unfor- neous in terms of probiotic treatments than trials for AAD.
tunately, many trials only report the genus and species and Probiotic doses and durations were similar (100% of those re-
porting doses used >1010 /day) and all studies treated patients
for at least 3 wk.
Another limitation in these trials was the lack of standard-
ization when a combination treatment regimen was used. The
common strategy for treating CDD infections is to combine
the investigational probiotic with one of the standard antibi-
otics (vancomycin or metronidazole) given to treat C. difficile.
The hypothesis of the combination therapy is that the antibi-
otic kills vegetative C. difficile organisms in the intestine,
which would clear the pathogenic toxins, and the probiotic
would assist in reestablishing the protective intestinal mi-
croflora so that when residual spores germinate, colonization
is rebuffed by the newly restored microflora barrier. Unfor-
tunately, only the probiotic component of these trials was
randomized. The standard antibiotic varied by type and dose
and was not controlled in most trials. This is an important
consideration because Surawicz et al. found only a high dose
of vancomycin (2 g/day) completely cleared C. difficile tox-
ins by the end of 10 days of therapy, whereas a lower dose
Figure 4. Forest Plot of six randomized controlled trials of probiotics (500 mg/day) of vancomycin failed to clear toxins in 10%
for the treatment of Clostridium difficile disease showing crude and
pooled risk ratios. SB = Saccharomyces boulardii; LGG = Lac- and metronidazole (1.5 g/day) only cleared toxins in 40%
tobacillus rhamnosus GG; LP = Lactobacillus plantarum 299v; of the patients (58). Future studies should randomize the
LA = Lactobacillus acidophilus; BB = Bifidobacterium bifidum. combination treatment with a standard dose of both the
 Prevention of Antibiotic Associated Diarrhea 9

standard antibiotic and the probiotic to test the complete willingness to volunteer, scientific discoveries would not be
regimen. made.
Potential biases in review process may be due to publica-
tion bias. Sutton et al. reviewed 48 meta-analyses and found
30 (63%) made no reference to publication bias or reported STUDY HIGHLIGHTS
funnel plots (69). In this meta-analysis, publication bias was
What Is Current Knowledge
minimized by conducting extensive searches through multi-
ple databases and receiving original data from the authors. r Treatment strategies for diseases that involve the dis-
Funnel plots for AAD and CDD showed there may be some ruption of intestinal microflora by antibiotics may be
publication bias present for AAD, but the discovery of numer- particularly effective when probiotics are used.
ous negative trials for AAD may have minimized this bias. r Studies of the efficacy of probiotics in C. difficile colitis
Selection and ascertainment bias was minimized by including are conflicting and a consensus has yet to be reached.
only RCTs with validated outcomes.
What Is New Here
Concerns about the safety of probiotics have been raised.
As probiotics are living organisms given to ill patients, the r This meta-analysis pooled together 31 randomized con-
potential for adverse reactions exists. Some intestinal bac- trolled trials to determine if probiotics are efficious
teria have been shown to translocate from the intestine to overall for these types of diseases.
other organs and antibiotic-resistance gene acquisition also r Three different probiotics (Saccharomyces boulardii,
is a potential concern. These two problems have yet to be Lactobacillus rhamnosus GG and probiotic mixtures)
observed in clinical trials using probiotics. Although case helped prevent antibiotic-associated diarrhea but only
reports and case series of bacteremia and fungemia have S. boulardii appeared useful for Clostridium difficile
been reported in the literature, no incidents occurred in pa- disease.
tients enrolled in the 31 RCTs reviewed for this meta-analysis
(70, 71). Caution should be exercised for patients who are
severely ill and receiving nutrition or antibiotics through a Reprint requests and correspondence: Lynne McFarland, Ph.D.,
potentially open portal (catheter or nasogastric tube). Infre- Department of Health Services Research and Development, VA
Puget Sound Health Care System, S-152, Metropolitan Park West,
quent blood-stream infections have been reported, most prob- 1100 Olive Way, Suite 1400, Seattle, WA 98101.
ably due to contamination of the environment as the probi- Received August 24, 2005; accepted November 4, 2005.
otic capsule is opened at bedside and mixed with food (72).
Rare complications including endocarditis and liver abscess
have been associated with L. rhamnosus use (73, 74). Bac-
teremia and fungemia have been associated with probiotics, REFERENCES
but respond well to antibiotics or anti-fungal medications
1. Levy DG, Stergachis A, McFarland LV, et al. Antibiotics and
(75–78). Clostridium difficile diarrhea in the ambulatory care setting.
Considering that millions of doses of probiotics are taken Clin Ther 2000;22:91–102.
per year globally, the risk of complications due to probiotics 2. McFarland LV. Epidemiology, risk factors and treatments for
is extremely low. However, prolonged safety issues have not antibiotic-associated diarrhea. Dig Dis 1998;16:292–307.
been addressed in studies. 3. Winston DJ, Ho WG, Bruckner DA, et al. Beta-lactam an-
tibiotic therapy in febrile granulocytopenic patients. A ran-
domized trial comparing cefoperazone plus piperacillin, cef-
tazidime plus piperacillin, and imipenem alone. Ann Intern
Med 1991;115:849–59.
CONCLUSION 4. Ackermann G, Thomalla S, Achermann F, et al. Prevalence
and characteristics of bacteria and host factors in an outbreak
In summary, the present meta-analyses suggest that probiotics situation of antibiotic-associated diarrhea. J Med Microbiol
can significantly reduce the incidence of AAD and are an 2005;54(Pt 2):149–53.
effective treatment for CDD. Future studies should expand 5. McFarland LV. Normal flora: Diversity and functions. Mi-
crob Ecol Health Dis 2000;12:193–207.
the types of probiotics tested and pay careful attention to 6. Pepin J, Valiquette L, Alary ME, et al. Clostridium difficile-
proper study design and sample size considerations. associated diarrhea in a region of Quebec from 1991 to 2003:
A changing pattern of disease severity. J Ayub Med Coll
Abbottabad 2004;171:466–72.
7. Al-Eidan, McElnay JC, Schott MG, et al. Clostridium dif-
ficile-associated diarrhoea in hospitalized patients. J Clin
ACKNOWLEDGMENT Pharm Ther 2000;25:101–9.
The author had full access to all of the data in the study 8. Kyne L, Hamel MB, Polavaram R, et al. Health care costs
and mortality associated with nosocomial diarrhea due to
and takes responsibility for the integrity of the data and the Clostridium difficile. Clin Infect Dis 2002;34:346–53.
accuracy of the data analysis. The author also would like 9. McFarland LV, Surawicz CM, Rubin M, et al. Recurrent
to acknowledge the participants in these trials; without their Clostridium difficile disease: Epidemiology and clinical
10 McFarland

characteristics. Infect Control Hosp Epidemiol 1999;20:43– 30. Begg CB, Mazumdar M. Operating characteristics of
50. a rank correlation test for publication bias. Biometrics
10. Miller MA, Hyland M, Ofner-Agostini M, et al, Cana- 1994;50(4):1088–101.
diam Hospital Epdemiology Committee. Morbidity, mor- 31. Beniwal RS, Arena VC, Thomas L, et al. A randomized trial
tality, and healthcare burden of nosocomial Clostridium dif- of yogurt for prevention of antibiotic-associated diarrhea.
ficile-associated diarrhea in Canadian hospitals. Infect Con- Dig Dis Sci 2003;48:2077–82.
trol Hosp Epidemiol 2002;23:137–40. 32. Ahuja M, Khamar B. Antibiotic associated diarrhea: A
11. Oldfield EC. Clostridium difficile-associated diarrhea: Risk controlled study comparing plain antibiotic with those
factors, diagnostic methods and treatment. Rev Gastroen- containing protected Lactobacilli. J Indian Med Assoc
terol Disord 2004;4(4):186–95. 2002;100(5):334–5.
12. McFarland LV. A review of the evidence of health claims for 33. Lewis S, Burmeister S, Cohen S, et al. Failure of dietary
biotherapeutic agents. Microb Ecol Health Dis 2000;12:65– oligofructose to prevent antibiotic-associated diarrhea. Ali-
76. ment Pharmacol Ther 2005;21:469–77.
13. Qamar A, Aboudola S, Warny M, et al. Saccharomyces 34. Lewis S, Burmeister S, Brazier J. Effect of the prebiotic
boulardii stimulates intestinal immunoglobulin A immune oligofructose on relapse of Clostridium difficile-associated
response to Clostridium difficile toxin A in mice. Infect Im- diarrhea: A randomized, controlled study. Clin Gastroen-
mun 2001;69:2762–5. terol Hepatol 2005;3:442–8.
14. Elmer GW. Probiotics: “Living drugs.” Am J Health Syst 35. Adam J, Barret C, Barret-Bellet A, et al. Controlled double-
Pharm 2001;58:1101–9. blind clinical trials of Ultra-Levure: Multicentre study by 25
15. Katikireddi V. UK launches inquired into Clostrid- physicians in 388 cases. Gaz Med Fr 1977;84:2072–8.
ium difficile outbreak. J Ayub Med Coll Abbottabad 36. Surawicz CM, Elmer GW, Speelman P, et al. Pre-
2005;173(2):138. vention of antibiotic-associated diarrhea by Saccha-
16. D’Souza AL, Rajkumar C, Cooke J, et al. Probiotics in pre- romyces boulardii—A prospective-study. Gastroenterology
vention of antibiotic associated diarrhoea: Meta-analysis. 1989;96:981–8.
BMJ 2002;324:1361. 37. McFarland LV, Surawicz CM, Greenberg RN, et al. Preven-
17. Cremonini F, Di Caro S, Nista EC, et al. Meta-analysis: The tion of beta-lactam-associated diarrhea by Saccharomyces
effect of probiotic administration on antibiotic-associated boulardii compared with placebo. Am J Gastroenterol
diarrhoea. Aliment Pharmacol Ther 2002;16:1461–7. 1995;90:439–48.
18. Bricker E, Garg R, Nelson R, et al. Antibiotic treatment for 38. Kotowska M, Albrecht P, Szajewska H. Saccharomyces
Clostridium difficile-associated diarrhea in adults. Cochrane boulardii in the prevention of antibiotic-associated diar-
Database Syst Rev 2005;25(1):CD004610. rhoea in children: A randomized double-blind placebo-
19. McFarland LV, Surawicz CM. Clostridium difficile disease: controlled trial. Aliment Pharmacol Ther 2005;21:583–90.
Diagnosis and treatment. In: McDonald JWD, Burroughs 39. Lewis SJ, Potts LF, Barry RE. The lack of therapeutic effect
AK, Feagan BG, eds. Evidence based gastroenterology and of Saccharomyces boulardii in the prevention of antibiotic-
hepatology, 2nd Ed. London: BMJ Books, 2004. Chapter related diarrhoea in elderly patients. J Infect 1998;36:171–4.
18: 285–301. 40. Cremonini F, Di Caro S, Covino M, et al. Effect of different
20. Shaw RL, Booth A, Sutton AJ, et al. Finding qualitative probiotic preparations on anti-Helicobacter pylori therapy-
research: An evaluation of search strategies. BMC Med Res related side effects: A parallel group, triple blind, placebo-
Methodol 2004;4:5–9. controlled study. Am J Gastroenterol 2002;97(11):2744–9.
21. Egger M, Smith GD, Phillips AN. Meta-analysis: Principles 41. Arvola T, Laiho K, Torkkeli S, et al. Prophylactic Lacto-
and procedures. BMJ 1997;315:1533–7. bacillus GG reduces antibiotic-associated diarrhea in chil-
22. Moher D, Cook DJ, Eastwood S, et al. Improving the qual- dren with respiratory infections: A randomized study. Pedi-
ity of reports of meta-analyses of randomized controlled atrics 1999;104:e64.
trials: The QUOROM statement. QUOROM Group. Br J 42. Vanderhoof JA, Whitney DB, Antonson DL, et al. Lacto-
Surg 2000;87(11):1448–54. bacillus GG in the prevention of antibiotic-associated diar-
23. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of rhea in children. J Pediatr 1999;135:564–8.
observational studies in epidemiology: A proposal for re- 43. Szajewska H, Kotowska M, Mrukowicz JZ, et al. Efficacy
porting. JAMA 2000;283:2008–12. of Lactobacillus GG in prevention of nosocomial diarrhea
24. McFarland LV, Elmer GM. Properties of evidence-based in infants. J Pediatr 2001;138:361–5.
probiotics for human health. In: Goklepe I, Juneja V eds. 44. Thomas MR, Litin SC, Osmon DR, et al. Lack of ef-
Probiotics in food safety and human health: New York Mar- fect of Lactobacillus GG on antibiotic-associated diarrhea:
cel Dekker, Inc, 2005:109–37. A randomized, placebo-controlled trial. Mayo Clin Proc
25. Surawicz CM. Antibiotic-associated diarrhea and pseu- 2001;76:883–9.
domembranous colitis: Are they less common with poorly 45. Armuzzi A, Cremonini F, Ojetti V, et al. Effect of Lacto-
absorbed antimicrobials? Chemotherapy 2005;51(suppl bacillus GG supplementation on antibiotic-associated gas-
1):81–9. trointestinal side effects during Helicobacter pylori eradi-
26. Harris RP, Helfand M, Woolf SH, et al. Current methods cation therapy: A pilot study. Digestion 2001;63:1–7.
of the U.S. Preventive Services Task Force: A review of the 46. Nista EC, Candelli M, Cremonini F, et al. Bacillus clausii
process. Am J Prev Med 2001;20(suppl 3):21–35. therapy to reduce side-effects of anti-Helicobacter pylori
27. Juni P, Altman DG, Egger M. Systematic reviews in health treatment: Randomized, double-blind, placebo controlled
care: Assessing the quality of controlled clinical trials. BMJ trial. Aliment Pharmacol Ther 2004;20:1181–8.
2001;323:42–6. 47. Orrhage K, Brismar B, Nord CE. Effects of supplements
28. DerSimonian R, Laird N. Meta-analysis in clinical trials. of Bifidobacterium longum and Lactobacillus acidophilus
Control Clin Trials 1986;7(3):177–88. on the intestinal microbiota during administration of clin-
29. Egger M, Smith GD, Schneider M, et al. Bias in damycin. Microb Ecol Health Dis 1994;7:17–25.
meta-analysis detected by a simple, graphical test. BMJ 48. Seki H, Shiohara M, Matsumura T, et al. Preven-
1997;315:629–34. tion of antibiotic-associated diarrhea in children by
 Prevention of Antibiotic Associated Diarrhea 11

Clostridium butyricum MIYAIRI. Pediatr Int 2003;45:86– 63. Belongia EA, Knobloch MH, Kieke BA, et al. Impact of
90. statewide program to promote appropriate antimicrobial
49. Wunderlich PF, Braun L, Fumagalli I, et al. Double-blind drug use. Emerg Infect Dis 2005;11(6):912–20.
report on the efficacy of lactic acid-producing Enterococ- 64. Gul YA, Hong LC, Prasannan S. Appropriate antibiotic ad-
cus SF68 in the prevention of antibiotic-associated diar- ministration in elective surgical procedures: Still missing
rhoea and in the treatment of acute diarrhoea. J Int Med the message. Asian J Surg 2005;28(2):104–8.
Res 1989;17:333–8. 65. Berild D, Smaabrekke L, Halvorsen DS, et al. Clostrid-
50. Borgia M, Sepe N, Brancato V, et al. A controlled clinical ium difficile infections related to antibiotic use and infection
study on Streptococcus faecium preparation for the preven- control facilities in two university hospitals. J Hosp Infect
tion of side reactions during long term antibiotic treatments. 2003;54(3):202–6.
Curr Ther Res 1982;31:265–71. 66. Szajewska H, Mrukowicz J. Meta-analysis: Non-pathogenic
51. Witsell DL, Garrett G, Yarbrough WG, et al. Effect of Lacto- yeast Saccharomyces boulardii in the prevention of
bacillus acidophilus on antibiotic-associated gastrointesti- antibiotic-associated diarrhea. Aliment Pharmacol Ther
nal morbidity: A prospective randomized trial. J Otolaryno- 2005;22:365–72.
gol 1995;24:231–3. 67. Morelli L, Zonenschain D, Callegari ML, et al. Assessment
52. Gotz V, Romankiewicz JA, Moss J, et al. Prophylaxis against of a new synbiotic preparation in health volunteers: Sur-
ampicillin-associated diarrhea with a Lactobacillus prepa- vival, persistence of probiotic strains and its effect on the
ration. Am J Hosp Pharm 1979;36:754–7. indigenous flora. Nutr J 2003;2:11–6.
53. Tankanow RM, Ross MB, Ertel IJ, et al. A double-blind, 68. Dunne C, O’Mahony L, Murphy L, et al. In vitro selec-
placebo-controlled study of the efficacy of Lactinex in tion criteria for probiotic bacteria of human origin: Corre-
the prophylaxis of amoxicillin-induced diarrhea. DICP lation with in vivo findings. Am J Clin Nutr 2001;73:386S–
1990;24:382–4. 92S.
54. Correa NB, Filho P, Luciano A, et al. A randomized formula 69. Sutton AJ, Duval SJ, Tweedie RL, et al. Empirical assess-
controlled trial of Bifidobacterium lactis and Streptococcus ment of effect of publication bias on meta-analysis. BMJ
thermophilus for prevention of antibiotic-associated diar- 2000;320:1574–7.
rhea in infants. J Clin Gastroenterol 2005;39:385–9. 70. DeGroote MA, Frank DN, Dowell E, et al. Lactobacillus
55. LaRosa M, Bottaro G, Gulino N, et al. Prevention of rhamnosus GG bacteremia associated with probiotic use
antibiotic-associated diarrhea with Lactobacillus sporo- in a child with short gut syndrome. Pediar Infect Dis J
gens and fructo-oligosaccarides in children. A multi- 2005;24(3):278–80.
centric double-blind vs placebo study. Minerva Pediatr 71. Young RJ, Vanderhoof JA. Two cases of Lactobacillus bac-
2003;55(5):447–52. teremia during probiotic treatment of short gut syndrome. J
56. Jirapinyo P, Thamonsiri N, Densupsoontorn N, et al. Pre- Pediatr Gastroenterol Nutr 2004;39(4):436–7.
vention of antibiotic-associated diarrhea in infants by pro- 72. Hennequin C, Kauffmann-Lacroix C, Jobert A, et al. Possi-
biotics. J Med Assoc Thai 2002;85(suppl 2):S739-42. ble role of catheters in Saccharomyces boulardii fungemia.
57. McFarland LV, Surawicz CM, Greenberg RN, et al. Eur J Clin Microbiol Infect Dis 2000;19:16–20.
A randomized placebo-controlled trial of Saccharomyces 73. Rautio M, Jousimies-Somer H, Kauma H, et al. Liver ab-
boulardii in combination with standard antibiotics for scess due to a Lactobacillus rhamnosus strain indistin-
Clostridium difficile disease. JAMA 1994;271:1913–8. guishable from L. rhamnosus strain GG. Clin Infect Dis
58. Surawicz CM, McFarland LV, Greenberg RN, et al. The 1999;28:1159–60.
search for a better treatment for recurrent Clostridium diffi- 74. Sipsas NV, Zonios DI, Kordossis T. Safety of Lacto-
cile disease: Use of high-dose vancomycin combined with bacillus strains used as probiotic agents. Clin Infect Dis
Saccharomyces boulardii. Clin Infect Dis 2000;31:1012– 2002;34:1283–4.
7. 75. Munoz P, Bouza E, Cuenca-Estrella M, et al. Saccharomyces
59. Pochapin M. The effect of probiotics on Clostridium difficile cerevisiae fungemia: An emerging infectious disease. Clin
diarrhea. Am J Gastroenterol 2000;95:S11–13. Infect Dis 2005;40:1625–34.
60. Wullt M, Hagslatt ML, Odenholt I. Lactobacillus plantarum 76. Salminen MK, Rautelin H, Tynkkynen S, et al. Lactobacillus
299v for the treatment of recurrent Clostridium difficile- bacteremia, clinical significance, and patient outcome, with
associated diarrhoea: A double-blind, placebo-controlled special focus on probiotic L. rhamnosus GG. Clin Infect Dis
trial. Scand J Infect Dis 2003;35:365–7. 2004;38:62–9.
61. Lawrence SJ, Korzenik JR, Mundy LM. Probiotics for 77. Land MH, Rouster-Stevens K, Woods CR, et al. Lacto-
recurrent Clostridium difficile disease. J Med Microbiol bacillus sepsis associated with probiotic therapy. Pediatrics
2005;54(Pt 9):905–6. 2005;115(1):178–81.
62. Plummer S, Weaver MA, Harris JC, et al. Clostridium diffi- 78. Surawicz CM, McFarland LV. Risks of biotherapeutic
cile pilot study: Effects of probiotoic supplementation on the agents. In: Elmer GW, McFarland LV, Surawicz CM, eds.
incidence of Clostridium difficile diarrhoea. Int Microbiol Biotherapeutic agents and infectious diseases. Totowa, NJ:
2004;7:59–62. Humana Press, 1999:263–8.