Alimentary Pharmacology & Therapeutics

Clinical trial: effectiveness of Lactobacillus rhamnosus (strains E N, Oxy and Pen) in the prevention of antibiotic-associated diarrhoea in children
M. RUSZCZYN SKI*, A. RADZIKOWSKI & H. SZAJEWSKA*

*2nd Department of Pediatrics, The Medical University of Warsaw and The Medical University of Warsaw, Warsaw, Poland Correspondence to: Dr H. Szajewska, 2nd Department of Paediatrics, The Medical University of Warsaw, 01-184 Warsaw, Dzialdowska 1, Poland. E-mail: hania@ipgate.pl

SUMMARY Background Convincing evidence that probiotic administration can lower the risk of antibiotic-associated diarrhoea is limited to certain micro-organisms. Aim To determine the efcacy of administration of Lactobacillus rhamnosus (strains E N, Oxy and Pen) for the prevention of antibiotic-associated diarrhoea in children. Methods Children (aged 3 months to 14 years) with common infections were enrolled in a double-blind, randomized, placebo-controlled trial in which they received standard antibiotic treatment plus 2 1010 colony forming units of a probiotic (n = 120) or a placebo (n = 120), administered orally twice daily throughout antibiotic treatment. Analyses were by intention to treat. Results Any diarrhoea (3 loose or watery stools day for 48 h occurring during or up to 2 weeks after the antibiotic therapy) occurred in nine (7.5%) patients in the probiotic group and in 20 (17%) patients in the placebo group (relative risk, RR 0.45, 95% condence interval, CI 0.20.9). Three (2.5%) children in the probiotic group developed AAD (diarrhoea caused by Clostridium difcile or otherwise unexplained diarrhoea) compared to nine (7.5%) in the placebo group (RR 0.33, 95% CI 0.11.06). No adverse events were observed. Conclusion Administration of L. rhamnosus (strains E N, Oxy and Pen) to children receiving antibiotics reduced the risk of any diarrhoea, as dened in this study.
Aliment Pharmacol Ther 28, 154161

Publication data Submitted 6 March 2008 First decision 18 March 2008 Resubmitted 7 April 2008 Accepted 8 April 2008 Epub OnlineAccepted 12 April 2008

154

2008 The Authors Journal compilation 2008 Blackwell Publishing Ltd doi:10.1111/j.1365-2036.2008.03714.x

C L I N I C A L T R I A L : L . R H A M N O S U S I N T H E P R E V E N T I O N O F A N T I B I O T I C - A S S O C I A T E D D I A R R H O E A 155

INTRODUCTION
Antibiotic-associated diarrhoea (AAD) is caused by the administration of antibiotics and is characterized by an acute inammation of the intestinal mucosa. Almost all antibiotics, particularly those that act on anaerobes, can cause diarrhoea; however, the risk of developing diarrhoea is higher with the use of aminopenicillins, a combination of aminopenicillins and clavulanate, second- and third-generation cephalosporins, and clindamycin.1, 2 The reported incidence of AAD in children receiving broad-spectrum antibiotics depends on the denition of diarrhoea used, the inciting antimicrobial agents, and host factors, and it ranges from 11% to 40%.3, 4 Clinical symptoms vary widely and can range from mild diarrhoea to colitis to pseudomembranous enterocolitis. The latter condition, which may lead to death, is most commonly caused by toxins produced by Clostridium difcile.5 Measures to prevent AAD include the use of probiotics, which are live micro-organisms administered in adequate amounts which confer a benecial health effect on the host.6 In children, a reduction in the risk of developing AAD with the administration of certain probiotic micro-organisms has been documented in several meta-analyses of randomized controlled trials (RCTs),7, 8 including one performed by us.9 However, as of today, very few probiotics have been tested, and it is clear that not all probiotics are created equal. The benecial effects of probiotics seem to be strain-specic. Extrapolating data on different strains may result in misleading conclusions. While many microorganisms may be effective in preventing AAD, the efcacy of each one needs to be conrmed in adequately powered, double-blind RCTs.10 Accordingly, the objective of this study was to assess the efcacy of Lactobacillus rhamnosus strains E N, Oxy and Pen (hereafter, referred to collectively as L. rhamnosus) administration in the prevention of AAD in children requiring antibiotic treatment for common acute infections.

a similar study carried out earlier at our center, which was coordinated by one of the investigators of this study (H.S.) and is described elsewhere.11

Participants
This study was conducted between February 2005 and January 2008. Children were recruited from two paediatric hospitals in Warsaw, Poland, and from the private practice of one of the investigators (A.R.). Eligible patients were those aged 3 months to 14 years with common infections (e.g. respiratory tract infection, otitis media, urinary tract infection, skin infection) who started treatment with oral or intravenous antibiotics within 24 h of enrolment. Exclusion criteria included the presence of a severe or generalized bacterial infection, antibiotic treatment within the previous 2 months, prophylactic antibiotic treatment, using a probiotic product for medicinal purposes within the previous 7 days, immunodeciency, chronic gastrointestinal disease, and acute or chronic diarrhoea.

Interventions
Patients were randomized to receive antibiotic therapy, as appropriate, plus either 2 109 colony forming units of L. rhamnosus or a comparable placebo (nonfat milk and saccharose). Both the active treatment and placebo were taken orally twice daily for the duration of the antibiotic treatment. The study products were supplied by Biomed (Lublin, Poland), who had no role in the conception, design, or conduct of the study or in the analysis or interpretation of the data. Randomization codes were secured until all data entry was complete and data were analysed. The probiotic combination used in this study is commercially available as Lakcid Forte. The strains are deposited at the Institute of Biochemistry and Biophysics, The Polish Academy of Sciences, under the numbers 2593 (L. rhamnosus Pen), 2594 (L. rhamnosus E N) and 2595 (L. rhamnosus Oxy). Each patient (or a parent, in the case of very young subjects) received a diary to record the frequency of daily bowel movements, as well as any symptoms they considered important. Stool number and consistency were recorded daily. In the event of loose or watery stools, patients were advised to contact their physicians and bring stool samples in for analysis. The presence of viral pathogens was investigated in all diarrhoeal stool samples using a rapid, qualitative, chromatographic immunoassay that

METHODS Study design

This was a double-blind, randomized, placebo-controlled, clinical trial intended to evaluate the efcacy, safety and tolerability of L. rhamnosus in the prevention of AAD in children. We followed the protocol of
2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

 156 M . R U S Z C Z Y N S K I et al.

simultaneously detects rotaviruses and adenoviruses (VIKIA Rota Adeno; BioMerieux, Lyon, France). Standard microbiological techniques were used to isolate and identify bacterial pathogens (Salmonella spp., Shigella spp., Escherichia coli, Campylobacter spp.). Clostridium difcile toxins A and B were identied by enzyme immunoassay (C. difcile Tox. A B II EIA kit; TechLab Inc., Blacksburg, VA, USA). Treatment compliance was assessed by direct interview of the patient (or parent) and review of the diary cards (which documented the number of daily capsules taken). Treatment responses also were assessed by evaluation of the daily diaries.

was based on the results of the study previously coordinated by one of us (H.S.) in the same setting;11 results from this study showed a 15% reduction in the proportion of children with diarrhoea compared with an expected 23% baseline failure rate with placebo, a 1:1 allocation ratio, 80% power, an a-level of 0.05, and a 20% dropout rate.

Statistical analysis
All statistical analyses were performed with the computer software StatsDirect [version 2,5,6, (2006-04-15); Iain E. Buchan]. The Students t-test was used to compare means of continuous variables approximating a normal distribution. For non-normally distributed variables, the MannWhitney U-test was used. The chi-squared test or Fisher exact test was used, as appropriate, to compare percentages. The relative risk (RR), 95% condence interval (CI), and number needed to treat (NNT) were calculated using the same computer software. The difference between study groups was considered signicant when the P-value was <0.05 or when the 95% CI for RR did not exceed 1.0 (equivalent to P < 0.05). All statistical tests were two tailed and performed at the 5% level of signicance. We report the analysis based on intention-to-treat principle.

Randomization
Investigators at the Medical University of Warsaw prepared randomization lists using a permuted block of six (three received placebo and three, active treatment). Separate randomization lists were prepared for each site. To ensure allocation concealment, an independent subject prepared the randomization schedule and oversaw the packaging and labelling of trial treatments. All investigators, participants, outcome assessors and data analysts were blinded to the assigned treatment throughout the study.

Outcome measures
The primary outcome measures were the frequencies of any diarrhoea and AAD. Any diarrhoea was dened as 3 loose or watery stools per day for a minimum of 48 h, occurring during and or up to 2 weeks after the end of the antibiotic therapy. AAD was diagnosed in cases of diarrhoea, dened clinically as above, caused by C. difcile (the major enteropathogen in AAD) or for otherwise unexplained diarrhoea (i.e. negative laboratory stool test for rotavirus and adenovirus and negative stool culture). The secondary outcome measures were the frequencies of C. difcile diarrhoea, rotaviral diarrhoea, Salmonella spp. diarrhoea, or Shigella spp. diarrhoea; the need for discontinuation of the antibiotic treatment, hospitalization to manage the diarrhoea (in outpatients), or intravenous rehydration in any of the study groups and adverse events.

Ethical considerations
Parents were fully informed about the aims and procedures of the study, and informed consent was obtained from at least one parent. The study protocol was reviewed and approved by the Ethical Committee at the Medical University of Warsaw.

RESULTS
Figure 1 is a ow diagram showing the subjects progression through the study. Of the 240 children recruited in the study, we assigned 120 children to receive L. rhamnosus and 120 to receive the placebo. Overall, three of the randomized children (one in the probiotic group and two in the placebo group) discontinued the study intervention and started to use one of the commercially available probiotic products. However, no patient was lost to follow-up. Thus, all 240 children enrolled were available for the analysis. The baseline characteristics of these children are shown in Table 1.
2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

Sample size
We estimated that a minimum sample size of 232 participants would be needed. The sample size calculation

C L I N I C A L T R I A L : L . R H A M N O S U S I N T H E P R E V E N T I O N O F A N T I B I O T I C - A S S O C I A T E D D I A R R H O E A 157

Randomized n = 240

Table 1. Baseline characteristics of study children Placebo (n = 120), n (%) Mean (s.d.) age (months) Gender (f m) Diagnosis Upper respiratory tract infections Otitis media Lower respiratory tract infections Urinary tract infection Other (lymphadenitis, osteomyelitis, skin infection, stomatitis, scarlet fever, erythema nodosum) Antibiotic used Penicillins Broad-spectrum penicillins (ampicillin, amoxicillin, amoxicillin plus clavulanate) Cephalosporins Macrolides (clarithromycin, roxithromycin) Clindamycin Mean (s.d.) duration of treatment (days) Route of administration Oral Intravenous Intravenous, then oral Intramuscular Setting Outpatient Hospital 53.5 (44.25) Median 39 56 64 13 (11) 27 (23) 53 (44) 20 (17) 7 (6) Probiotic (n = 120), n (%) 54.8 (45.3) Median 44 54 66 19 (16) 18 (15) 49 (41) 24 (20) 10 (8)

Allocated to placebo group n = 120

Allocated to probiotic group n = 120

Discontinued intervention n = 2

Discontinued intervention n = 1

Analysed n = 120

Analysed n = 120

Figure 1. Flow diagram of the subjects progression through the study.

The outcome measures are summarized in Table 2. Overall, diarrhoea was diagnosed in 29 of 240 (12%) children and AAD in 12 of 240 (5%) children. The addition of L. rhamnosus vs. placebo to the antibiotic therapy reduced the risk of any diarrhoea [9 120 (7.5%) vs. 20 120 (16.7%), RR 0.45, 95% CI 0.20.9]. For every 11 patients who receive daily L. rhamnosus with antibiotics, one fewer patient will develop diarrhoea (the number needed to treat was 11, 95% CI 6 106). Lactobacillus rhamnosus also reduced the risk of AAD when compared with placebo, although the difference between groups was of borderline signicance [3 120 (2.5%) vs. 9 120 (7.5%), RR 0.33, 95% CI 0.1 1.06]. The risk of developing documented C. difcile diarrhoea was similar in both groups [3 120 (2.5%) in the probiotic group vs. 7 120 (5.8%) in the placebo group, RR 0.4, 95% CI 0.11.5]. In addition, compared with controls, L. rhamnosus did not signicantly reduce the risk of developing diarrhoea caused by other enteropathogens (e.g. rotavirus, adenovirus, Salmonella). One child in the probiotic group and two children in the placebo group needed intravenous rehydration (no signicant difference between groups). There was no need for discontinuation of the antibiotic treatment or hospitalization of outpatients for management of the diarrhoea. Table 3 summarizes characteristics of the patients who experienced diarrhoea. The time until onset of the diarrhoea was similar in the placebo and probiotic groups (6.2 4.2 vs. 5.8 5.2 days respectively; P = 0.4), as well as the duration of diarrhoea
2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

5 (4) 68 (57)

10 (8) 51 (43)

34 (28) 11 (9) 2 (2) 8.2 (3.2) (min. 3 max. 30) 65 48 5 2 (54) (40) (4) (2)

55 (46) 4 (3) 8 (2) (min. 3 max. 15) 69 (58) 39 (32) 12 (10) 56 (47) 64 (53)

50 (42) 70 (58)

(4.1 2.1 vs. 4.4 2.2 days respectively; P = 0.6) Broad-spectrum penicillins caused 20 of 29 (69%) cases of diarrhoea. Whereas the proportion of children with diarrhoea caused by broad-spectrum penicillins was lower in the probiotic group than in the placebo group, the difference given the small number of patients experiencing diarrhoea was not statistically signicant (5 9 vs. 15 20 respectively; RR 0.7, 95% CI 0.31.3). Lactobacillus rhamnosus was well tolerated, and no adverse event associated with this therapy (or with the use of placebo) was reported.

 158 M . R U S Z C Z Y N S K I et al.

Table 2. Outcome measures Placebo (n = 120), n (%) 20 (16.7) 9 (7.5) Probiotic (n = 120), n (%) 9 (7.5) 3 (2.5) Relative risk (95% condence interval) 0.45 (0.20.9) 0.33 (0.11.06)

Outcome measure Diarrhoea Antibiotic-associated diarrhoea (C. difcile + diarrhoea with unexplained cause*) Cause of diarrhoea Clostridium difcile toxins Rotavirus Adenovirus Salmonella Unexplained diarrhoea* Need for discontinuation of antibiotic treatment Need for hospitalization of outpatients to manage diarrhoea Need for intravenous rehydration Adverse effects during intervention

P-value 0.03 0.08

7 10 1 1 2 2

(5.8) (8.3) (0.8) (0.8) (1.7)

3 (2.5) 5 (4.1) 1 (0.8) 1

0.4 (0.11.5) 0.5 (0.181.35) 0 (03.8) 1 (0.19.5) 0 (01.9) 0.5 (0.073.8)

0.2 0.2 >0.99 >0.99 0.25

>0.99

* Negative laboratory stool test for rotavirus and adenovirus and negative stool culture for bacterial enteropathogens. An individual child may have had more than one enteropathogen.

Placebo (n = 20) Mean (s.d.) age (months) Time until onset of diarrhoea (s.d.) (days) Duration of antibiotic treatment (s.d.) (days) Duration of diarrhoea (s.d.) (days) Antibiotic used, n (%) Penicillins Broad-spectrum penicillins (ampicillin, amoxicillin, amoxicillin plus clavulanate) Cephalosporins Macrolides (roxithromycin) 18.6 (13.7) Median 15 6.2 (4.2) 8.7 (5.4) 4.1 (2.1) 1 (5) 15 (75) 3 (15) 1 (5)

Probiotic (n = 9) 32.6 (41) Median 9 5.8 (5.2) 8 (1.3) 4.4 (2.2) 5 (56) 4 (44)

P-value* 0.9 0.4 0.5 0.6 >0.99 0.8 0.4 >0.99

Table 3. Characteristics of patients with diarrhoea

* MannWhitney test or chi-squared test or Fisher exact test (as appropriate).

DISCUSSION Principal ndings

In our randomized, double-blind, placebo-controlled study, L. rhamnosus strains E N, Oxy and Pen (Lakcid Forte), compared with placebo as an adjunct to antibiotic therapy, more effectively reduced the risk of any diarrhoea in children undergoing antimicrobial therapy for common infectious diseases. For every 11 patients who receive daily probiotics with antibiotics, one fewer patient will develop diarrhoea (the number needed to treat was 11, 95% CI 6106). The risk of AAD diarrhoea

(diarrhoea caused by C. difcile or diarrhoea with negative laboratory test results for enteropathogens) and the risk of C. difcile diarrhoea also appeared to be lower in our group of children receiving probiotic compared with those receiving placebo; however, the difference between groups was not statistically signicant. As in the previous study carried out in the same setting,11 we evaluated potentially important consequences of AAD (e.g. the need for discontinuation of antibiotic treatment, hospitalization, or intravenous rehydration). Despite the rather conservative denition of diarrhoea used in this trial (3 loose or watery stools per day for a minimum of 48 h), no diarrhoeal episodes
2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

C L I N I C A L T R I A L : L . R H A M N O S U S I N T H E P R E V E N T I O N O F A N T I B I O T I C - A S S O C I A T E D D I A R R H O E A 159

required hospitalization of outpatients, and there was no difference between groups in the need for intravenous rehydration. No adverse events occurred. We chose this product containing L. rhamnosus strains E N, Oxy and Pen as it fulls the criteria for being a probiotic and because of its common use in our country, despite the lack of studies that have evaluated its effectiveness. Recently, using classical microbiological methods as well as PCR, DNA sequencing and agar disc-diffusion techniques, it has been conrmed that, as declared by the manufacturer, the preparation we used contains three different strains, which belong to the species of L. rhamnosus. It was found that they differ from one another in colony and cell morphology, carbohydrate utilization patterns (API tests), the nucleotide sequence of their DNAs coding for 16S rRNA, and in antibiotic sensitivity. As a whole, the mixture was shown to be resistant to 42 of 47 antimicrobial drugs tested.12 Further human studies are warranted to clarify whether the study product is also effective for the prevention or treatment of other diseases.

Consideration of possible mechanisms

This study did not directly address the mechanisms by which L. rhamnosus strains E N, Oxy and Pen reduce the risk of developing diarrhoea during antibiotic therapy. However, the rationale for the use of probiotics is based on the assumption that the use of antibiotics leads to a disturbance in the normal intestinal microbiota and that this is a key factor in the pathogenesis of AAD and C. difcile infection.13 Probiotics are believed to restore the equilibrium of the altered intestinal microbiota. Several possible mechanisms by which probiotics might exert their activity against enteropathogens in humans have been proposed, mostly based on results of in vitro and animal studies. These include the synthesis of antimicrobial substances,1416 competition for nutrients required for growth of pathogens,17, 18 competitive inhibition of adhesion of pathogens,1922 and modication of toxins or toxin receptors.23, 24 It is likely that several of the above-described mechanisms operate simultaneously and that they may well differ depending on the properties of an enteric pathogen and the probiotic strain.

tiveness in the prevention of AAD. One large RCT involving 246 children, which was carried out in the same setting as our study, has provided evidence that the nonpathogenic yeast Saccharomyces boulardii is effective in preventing any diarrhoea (RR 0.3, 95% CI 0.20.7) and AAD (RR 0.2, 95% CI 0.070.5) during concurrent antibiotic therapy.11 These results are consistent with data from adults.25 Two RCTs conducted in children have provided evidence of a moderate benecial effect of Lactobacillus GG in the prevention of this condition;26, 27 however, results in adults are contradictory.28 One RCT involving use of a commercial probiotic formula containing Bidobacterium lactis and Streptococcus thermophilus revealed a signicant difference in the incidence of AAD in infants receiving probiotic-supplemented formula compared with nonsupplemented formula.29 One small RCT demonstrated that L. acidophilus B. bidus used in children treated with amoxicillin signicantly reduced the mean stool frequency (2 0.3 vs. 2.7 0.5 stools per day). In contrast, using either L. acidophilus B. infantis30 or L. acidophilus L. bulgaricus27 was not associated with a signicant reduction in the risk of AAD. More detailed information on RCTs carried out in children is available in three recently published meta-analyses.79

Strengths and limitations

We used adequate methods for the generation of the allocation sequence and allocation concealment. We made efforts to maintain blinding throughout participant selection, treatment, monitoring, data management, and data analyses. Data were obtained from all participants. These features minimize the potential for bias. In our study, we used a rather stringent denition of AAD, which was dened as 3 loose or watery stools per day for a minimum of 48 h. Such a stringent denition allowed us to differentiate between clinically relevant conditions and clinically unimportant changes in the consistency of stools. However, the denitions of AAD in published studies vary. A potential limitation of this trial is that it does not allow one to reach conclusions about the efcacy of administering L. rhamnosus strains E N, Oxy and Pen in the prevention of C. difcile-associated diarrhoea, a more clinically important condition. The trial included only a small number of patients with C. difcile and may have lacked the power to show a difference in the frequency of developing C. difcile-associated

Comparison with previous reports in children

None of the previous controlled trials has evaluated L. rhamnosus strains E N, Oxy and Pen for its effec 2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

 160 M . R U S Z C Z Y N S K I et al.

diarrhoea between patients who received probiotics vs. placebo, should one exist. The design of this study also does not allow one to reach conclusions about the efcacy of L. rhamnosus strains E N, Oxy and Pen in preventing diarrhoea attributable to any single antibiotic class. Larger trials will be necessary to address these issues further. Patients were only followed-up for 2 weeks after their antibiotic treatment. As diarrhoea may occur up to 2 months after the end of such treatment,5, 31, 32 some cases of AAD may have been missed. Also, as we only evaluated the presence of major enteropathogens in diarrhoeal stool samples, one cannot exclude the possibility that some of the cases of unexplained diarrhoea were caused by unidentied infectious agents.

an experimental probiotic treatment for acute pancreatitis.38 Given these data, probiotic safety is an important issue. Special consideration should be given to high-risk groups (e.g. immunocompromised subjects, patients with other life-threatening illnesses).

CONCLUSION
Results from this study contribute to the growing body of literature documenting the benet of probiotic administration for reducing the risk of developing diarrhoea during antibiotic treatment. Our ndings support the use of L. rhamnosus strains E N, Oxy and Pen as adjunctive treatment in children undergoing antibiotic therapy for common infectious diseases whenever the physician feels that preventing this usually self-limited complication is important. This study was not designed and powered to investigate the effect of the study product on the prevention of diarrhoea caused by any specic pathogen (i.e. non-antibiotic diarrhoea). Thus, no rm conclusions can be made regarding such outcomes. However, our results, e.g. a trend towards a lower risk of rotaviral diarrhoea in the probiotic group compared with the placebo group, suggest that L. rhamnosus strains E N, Oxy and Pen may effectively prevent or treat diarrhoea caused by various enteropathogens. Larger trials will be necessary to address these issues further.

Safety
During administration of the study products, patients (or their caretakers) were asked to document any potential side effects or adverse events, including only mild effects. However, the nature and or intensity of them were was not pre-specied. No side effects or adverse effects were observed in our study. Also, according to the manufacturer, the product has a long history (almost 20 years) of use without established risk to humans (unpublished reports). However, the administration of probiotics is not without risk. Of concern, there have been instances of bacteraemia with use of selected probiotic bacteria in high-risk populations.33 Endocarditis, pneumonia, and meningitis have very rarely been reported in association with lactobacilli administration.3337 Serious consideration must be given to one recent report describing an increase in mortality observed in a trial involving

ACKNOWLEDGEMENTS
Declaration of personal interests: None. Declaration of funding interests: This study was funded in part by Biomed, Lublin, Poland, and the Medical University of Warsaw (Research Agreement UKI 224 2004).

REFERENCES
1 Barbut F, Meynard JL, Guiguet M, et al. Clostridium difcile-associated diarrhea in HIV infected patients: epidemiology and risk factors. J Acquir Immune Dec Syndr 1997; 16: 17681. 2 McFarland LV, Surawicz CM, Stamm WE. Risk factors for Clostridium difcile carriage and C. difcile-associated diarrhea in a cohort of hospitalized patients. J Infect Dis 1990; 162: 678 84.

3 Turck D, Bernet JP, Marx J, et al. Incidence and risk factors of oral antibioticassociated diarrhea in an outpatient pediatric population. J Pediatr Gastroenterol Nutr 2003; 37: 226. 4 Elstner CL, Lindsay AN, Book LS, Matsen JM. Lack of relationship of Clostridium difcile to antibiotic-associated diarrhea in children. Pediatr Inf Dis 1983; 2: 3646. 5 Bartlett JG, Chang TW, Gurwith M, Gorbach SL, Onderdonk AB. Antibioticassociated pseudomembranous colitis

due to toxin-producing clostridia. N Engl J Med 1978; 298: 5314. 6 FAO WHO. Joint FAO WHO Expert Consultation on Evaluation of Health and Nutritional Properties of Probiotics in Food Including Powder Milk with Live Lactic Acid Bacteria. Cordoba, Argentina. http://www.who.int/foodsafety/ publications/fs_management/en/probiotics. pdf. Accessed on 2 March 2008. 7 Johnston BC, Supina AL, Vohra S. Probiotics for pediatric antibiotic-associated diarrhea: a meta-analysis of randomized

2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd

C L I N I C A L T R I A L : L . R H A M N O S U S I N T H E P R E V E N T I O N O F A N T I B I O T I C - A S S O C I A T E D D I A R R H O E A 161
placebo-controlled trials. CMAJ 2006; 175: 37783. Erratum in: CMAJ 2006; 175: 777. Johnston BC, Supina AL, Ospina M, et al. Probiotics for the prevention of pediatric antibiotic-associated diarrhea. Cochrane Database Syst Rev 2007; 2: CD004827. Szajewska H, Ruszczynski M, Radzikowski A. Probiotics in the prevention of antibiotic-associated diarrhea in children: a meta-analysis of randomized controlled trials. J Pediatr 2006; 149: 36772. Hickson M, DSouza AL, Muthu N, et al. Use of probiotic Lactobacillus preparation to prevent diarrhoea associated with antibiotics: randomised double blind placebo controlled trial. BMJ 2007; 335: 805. 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: 58390. Bardowski J, Gorecki RK, Koryszewska A, Szmytkowska A. Characterisation of Three Probiotic Strains of Lactobacillus rhamnosus Present in Lakcid. 3rd Probiotics, Prebioiotics and New Foods. ATTI Abstracts. Rome: University Urbaniana. URL http://www.agronavigator.cz/ UserFiles/File/Agronavigator/Kvasnickova/ Probiotics_Prebiotics_3.pdf. Accessed on 6 April 2008. Surawicz CM. Probiotics, antibioticassociated diarrhoea and Clostridium difcile diarrhoea in humans. Best Pract Res Clin Gastroenterol 2003; 17: 775 83. Goldin BR, Gorbach SL, Saxelin M, et al. Survival of Lactobacillus species (strain GG) in human gastrointestinal tract. Dig Dis Sci 1992; 37: 1218. Silva M, Jacobus NV, Deneke C, et al. Antimicrobial substance from a human Lactobacillus strain. Antimicrob Agents Chemother 1987; 31: 12313. Coconnier MH, Lievin V, Bernet-Camard MF, et al. Antibacterial effect of the adhering human Lactobacillus acidophilus strain LB. Antimicrob Agents Chemother 1997; 41: 104652. Wilson KH, Perini I. Role of competition for nutrients in suppression of Clostridium difcile by the colonic microora. Infect Immunol 1988; 56: 26104. 18 Walker WA. Role of nutrients and bacterial colonisation in the development of intestinal host defense. J Pediatr Gastroenterol Nutr 2000; 30(Suppl.): S27. 19 Bernet MF, Brassart D, Nesser JR, et al. Lactobacillus acidophilus LA1 binds to human intestinal cell lines and inhibits cell attachment and cell invasion by enterovirulent bacteria. Gut 1994; 35: 4839. 20 Davidson JN, Hirsch DC. Bacterial competition as a mean of preventing diarrhea in pigs. Infect Immun 1976; 13: 17734. 21 Rigothier MC, Maccanio J, Gayral P. Inhibitory activity of Saccharomyces yeasts on the adhesion of Entamoeba histolytica trophozoites to human erythrocytes in vitro. Parasitol Res 1994; 80: 105. 22 Michail S, Abernathy F. Lactobacillus plantarum reduces the in vitro secretory response of intestinal epithelial cells to enteropathogenic Escherichia coli infection. J Pediatr Gastroenterol Nutr 2002; 35: 3505. 23 Pothoulakis C, Kelly CP, Joshi MA, et al. Saccharomyces boulardii inhibits Clostridium difcile toxin A binding and enterotoxicity in rat ileum. Gastroenterology 1993; 104: 110815. 24 Czerucka D, Roux I, Rampal P. Saccharomyces boulardii inhibits secretagogue-mediated adenosine 3,5-cyclic monophosphate induction in intestinal cells. Gastroenterology 1994; 106: 65 72. 25 Szajewska H, Mrukowicz J. Meta-analysis: non-pathogenic yeast Saccharomyces boulardii in the prevention of antibiotic-associated diarrhoea. Aliment Pharmacol Ther 2005; 22: 36572. 26 Vanderhoof JA, Whitney DB, Antonson DL, Hanner TL, Lupo JV, Young RJ. Lactobacillus GG in the prevention of antibiotic-associated diarrhea in children. J Pediatrics 1999; 135: 5648. 27 Tankanov RM, Ross MB, Ertel IJ, Dickinson DG, McCormick LS, Garnkel JF. Double blind, placebo-controlled study of the efcacy of Lactinex in the prophylaxis of amoxicillin-induced diarrhea. DICP, Ann Pharm 1990; 24: 3824. 28 Thomas MR, Litin SC, Osmon DR, Corr AP, Weaver AL, Lohse CM. Lack of effect of Lactobacillus GG on antibioticassociated diarrhea: a randomized, placebo-controlled trial. Mayo Clin Proc 2001; 76: 8839. Correa NB, Peret Filho LA, Penna FJ, Lima FM, Nicoli JR. A randomized formula controlled trial of Bidobacterium lactis and Streptococcus thermophilus for prevention of antibiotic-associated diarrhea in infants. J Clin Gastroenterol 2005; 39: 3859. Jirapinyo P, Densupsoontorn N, Thamonsiri N, Wongarn R. Prevention of antibiotic-associated diarrhea in infants by probiotics. J Med Assoc Thai 2002; 85(Suppl. 2): S73942. Wistrom 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: 4350. McFarland LV. Epidemiology, risk factors and treatments for antibiotic-associated diarrhea. Dig Dis 1998; 16: 292 307. Kalima P, Masterton RG, Roddie PH, Thomas AE. Lactobacillus rhamnosus infection in a child following bone marrow transplant. J Infect 1996; 32: 1657. Soleman N, Laferl H, Kneifel W, et al. How safe is safe? A case of Lactobacillus paracasei ssp. paracasei endocarditis and discussion of the safety of lactic acid bacteria. Scand J Infect Dis 2003; 35: 75962. Salminen MK, Rautelin H, Tynkkynen S, et al. Lactobacillus bacteremia, clinical signicance, and patient outcome, with special focus on probiotic L. rhamnosus GG. Clin Infect Dis 2004; 38: 629. Salminen MK, Tynkkynen S, Rautelin H, et al. Lactobacillus bacteremia during a rapid increase in probiotic use of Lactobacillus rhamnosus GG in Finland. Clin Infect Dis 2002; 35: 115560. Kunz AN, Noel JM, Fairchok MP. Two cases of Lactobacillus bacteremia during probiotic treatment of short gut syndrome. J Pediatr Gastroenterol Nutr 2004; 38: 4578. Besselink MGH, van Santvoort HC, Buskens E, et al. Probiotic prophylaxis in predicted severe acute pancreatitis: a randomised, double-blind, placebo-controlled trial.Lancet 2008; DOI: 10.1016/ S0140-6736(08)60207-X.

29

30

10

31

32

11

33

12

34

35

13

36

14

37

15

38

16

17

2008 The Authors, Aliment Pharmacol Ther 28, 154161 Journal compilation 2008 Blackwell Publishing Ltd