Lactobacillus GG to Treat Acute Gastroenteritis in Children
Lactobacillus GG to Treat Acute Gastroenteritis in Children
The guidelines from the Cochrane Collaboration for undertaking and reporting the results of a systematic review and meta-analysis and the PRISMA statement were followed for this systematic review and meta-analysis.
All relevant randomised controlled trials (RCTs) that compared use of LGG as a single ingredient (in all delivery vehicles and formulations) with use of placebo or no treatment were eligible for inclusion. The primary outcome measures were the stool volume and the duration of diarrhoea. The secondary outcome measures were the percentages of children with diarrhoea at various times intervals (as specified by the investigators), the percentage of children with diarrhoea lasting longer than 7 days, the duration of hospitalisation and adverse effects.
The Cochrane Central Register of Controlled Trials (CENTRAL, the Cochrane Library), MEDLINE, and EMBASE databases were searched from August 2006 (end of last search) to May 2013. The principal search text word terms and MESH headings used were as follows: diarrhea/diarrhoea, diarrh*, gastroenteritis, probiotic*, Lactobacillus rhamnosus GG, Lactobacillus GG, LGG. No language restrictions were imposed. The reference lists from identified studies and key review articles, including previously published systematic reviews with or without a meta-analysis, were also searched to identify any other relevant studies. The ClinicalTrials.gov website was also searched for RCTs that were registered but not yet published. Certain publication types (i.e. letters to the editor, abstracts, proceedings from scientific meetings) were excluded, unless a full set of data was obtained from the authors. We contacted one author (Nixon) for further information, who by email provided missing information regarding the dose of LGG.
Three (AS, MR, DGB) reviewers using a standardised approach independently undertook the literature search, data extraction and quality assessment. The data sought included baseline characteristics of the participants, details related to the use of experimental and control interventions (including dose and duration), setting and funding. Any disagreements were resolved by discussion with the fourth reviewer (HS).
In one RCT (Basu et al.), participants were randomly assigned to three groups: an intervention group that received oral rehydration solution (ORS) plus LGG at a daily dose of 2 × 10 colony-forming units (CFU), another intervention group that received ORS plus LGG at a daily dose of 2 × 10 CFU, or a control group that received ORS only. Because the objective of our review was to compare supplementation with placebo or no supplementation, we combined both experimental arms into a single experimental group according to the method of Hogg and Craig. In the study by Misra et al., missing standard deviations were obtained by multiplying standard errors of means by the square root of the sample size: s.d. = S.E. × √N.
The reviewers independently, but without being blinded to the authors or journal, assessed the risk of bias in the studies that met the inclusion criteria. The Cochrane Collaboration's tool for assessing risk of bias was used, which includes the following criteria: adequacy of sequence generation; allocation concealment; blinding of participants, personnel and outcome assessors; and extent of loss to follow-up, i.e. the proportion of patients in whom the investigators were not able to determine outcomes (incomplete outcome data). In all cases, an answer of 'yes' indicates a low risk of bias, and an answer of 'no' indicates a high risk of bias.
The dichotomous outcomes, the results for individual studies and pooled statistics are reported as the risk ratio (RR) between the experimental and control groups with 95% confidence intervals (95% CI). The continuous outcomes are reported as the mean difference (MD) between the treatment and control groups with 95% CI.
We assessed pooled data using available case analysis, i.e. an analysis in which data are analysed for every participant for whom the outcome was obtained, rather than intention-to-treat analysis with imputation.
Heterogeneity was quantified by χ and I, which can be interpreted as the percentage of the total variation between studies that is attributable to heterogeneity rather than to chance. A value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity. If heterogeneity was not revealed, we present results of only the fixed effects model. If there was substantial heterogeneity (over 50%), all analyses were based on the random effects model.
To test for publication bias, we used a test for asymmetry of the funnel plot proposed by Egger et al. This test detects funnel plot asymmetry by determining whether the intercept deviates significantly from zero in a regression of the normalised 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 (on StatsDirect; StatsDirect Ltd. StatsDirect statistical software. http://www.statsdirect.com. England: StatsDirect Ltd. 2008. Version 2.7.9 (2012.07.09).
The data were analysed using Review Manager (RevMan) [Computer program]. Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012.
For the primary outcomes, a priori subgroup analyses based on factors that could potentially influence the magnitude of the treatment response were planned for the following: (i) Dose of LGG [high dose (≥10 CFU/day) vs. lower dose (<10 CFU/day)]. The optimal dose and treatment duration of LGG therapy have not been clearly established. However, the cut-off of 10 CFU/day has been previously suggested in the literature with a postulated larger effect in trials administering a larger dose of probiotic; (ii) Setting (studies carried out in geographical Europe vs. non-European countries). In the case of diarrhoeal diseases, consideration of the study location is important, as factors such as pathogens, access to clean water and sanitation, or comorbidities may have an impact on outcomes; (iii) Type of treatment (out-patient vs. in-patient); (iv) Aetiology of diarrhoea; (v) Vaccination against rotavirus status. In addition, when there was statistically significant heterogeneity in the primary outcome across studies, sensitivity analyses were performed to determine the impacts of allocation concealment (adequate vs. inadequate and/or unclear), blinding (open trial vs. double-blind trials) and attrition (<20% vs. ≥20%).
Methods
The guidelines from the Cochrane Collaboration for undertaking and reporting the results of a systematic review and meta-analysis and the PRISMA statement were followed for this systematic review and meta-analysis.
Criteria for Considering Studies for This Review
All relevant randomised controlled trials (RCTs) that compared use of LGG as a single ingredient (in all delivery vehicles and formulations) with use of placebo or no treatment were eligible for inclusion. The primary outcome measures were the stool volume and the duration of diarrhoea. The secondary outcome measures were the percentages of children with diarrhoea at various times intervals (as specified by the investigators), the percentage of children with diarrhoea lasting longer than 7 days, the duration of hospitalisation and adverse effects.
Search Methods for Identification of Studies
The Cochrane Central Register of Controlled Trials (CENTRAL, the Cochrane Library), MEDLINE, and EMBASE databases were searched from August 2006 (end of last search) to May 2013. The principal search text word terms and MESH headings used were as follows: diarrhea/diarrhoea, diarrh*, gastroenteritis, probiotic*, Lactobacillus rhamnosus GG, Lactobacillus GG, LGG. No language restrictions were imposed. The reference lists from identified studies and key review articles, including previously published systematic reviews with or without a meta-analysis, were also searched to identify any other relevant studies. The ClinicalTrials.gov website was also searched for RCTs that were registered but not yet published. Certain publication types (i.e. letters to the editor, abstracts, proceedings from scientific meetings) were excluded, unless a full set of data was obtained from the authors. We contacted one author (Nixon) for further information, who by email provided missing information regarding the dose of LGG.
Data Collection and Analysis
Three (AS, MR, DGB) reviewers using a standardised approach independently undertook the literature search, data extraction and quality assessment. The data sought included baseline characteristics of the participants, details related to the use of experimental and control interventions (including dose and duration), setting and funding. Any disagreements were resolved by discussion with the fourth reviewer (HS).
In one RCT (Basu et al.), participants were randomly assigned to three groups: an intervention group that received oral rehydration solution (ORS) plus LGG at a daily dose of 2 × 10 colony-forming units (CFU), another intervention group that received ORS plus LGG at a daily dose of 2 × 10 CFU, or a control group that received ORS only. Because the objective of our review was to compare supplementation with placebo or no supplementation, we combined both experimental arms into a single experimental group according to the method of Hogg and Craig. In the study by Misra et al., missing standard deviations were obtained by multiplying standard errors of means by the square root of the sample size: s.d. = S.E. × √N.
Assessment of Risk of Bias in Included Studies
The reviewers independently, but without being blinded to the authors or journal, assessed the risk of bias in the studies that met the inclusion criteria. The Cochrane Collaboration's tool for assessing risk of bias was used, which includes the following criteria: adequacy of sequence generation; allocation concealment; blinding of participants, personnel and outcome assessors; and extent of loss to follow-up, i.e. the proportion of patients in whom the investigators were not able to determine outcomes (incomplete outcome data). In all cases, an answer of 'yes' indicates a low risk of bias, and an answer of 'no' indicates a high risk of bias.
Measures of Treatment Effect
The dichotomous outcomes, the results for individual studies and pooled statistics are reported as the risk ratio (RR) between the experimental and control groups with 95% confidence intervals (95% CI). The continuous outcomes are reported as the mean difference (MD) between the treatment and control groups with 95% CI.
Dealing With Missing Data
We assessed pooled data using available case analysis, i.e. an analysis in which data are analysed for every participant for whom the outcome was obtained, rather than intention-to-treat analysis with imputation.
Assessment of Heterogeneity
Heterogeneity was quantified by χ and I, which can be interpreted as the percentage of the total variation between studies that is attributable to heterogeneity rather than to chance. A value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity. If heterogeneity was not revealed, we present results of only the fixed effects model. If there was substantial heterogeneity (over 50%), all analyses were based on the random effects model.
Assessment of Reporting Biases
To test for publication bias, we used a test for asymmetry of the funnel plot proposed by Egger et al. This test detects funnel plot asymmetry by determining whether the intercept deviates significantly from zero in a regression of the normalised 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 (on StatsDirect; StatsDirect Ltd. StatsDirect statistical software. http://www.statsdirect.com. England: StatsDirect Ltd. 2008. Version 2.7.9 (2012.07.09).
Data Synthesis (Statistical Methods)
The data were analysed using Review Manager (RevMan) [Computer program]. Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012.
Subgroup Analysis and Sensitivity Analyses
For the primary outcomes, a priori subgroup analyses based on factors that could potentially influence the magnitude of the treatment response were planned for the following: (i) Dose of LGG [high dose (≥10 CFU/day) vs. lower dose (<10 CFU/day)]. The optimal dose and treatment duration of LGG therapy have not been clearly established. However, the cut-off of 10 CFU/day has been previously suggested in the literature with a postulated larger effect in trials administering a larger dose of probiotic; (ii) Setting (studies carried out in geographical Europe vs. non-European countries). In the case of diarrhoeal diseases, consideration of the study location is important, as factors such as pathogens, access to clean water and sanitation, or comorbidities may have an impact on outcomes; (iii) Type of treatment (out-patient vs. in-patient); (iv) Aetiology of diarrhoea; (v) Vaccination against rotavirus status. In addition, when there was statistically significant heterogeneity in the primary outcome across studies, sensitivity analyses were performed to determine the impacts of allocation concealment (adequate vs. inadequate and/or unclear), blinding (open trial vs. double-blind trials) and attrition (<20% vs. ≥20%).