Re: Treating infant colic with the probiotic Lactobacillus reuteri: double blind, placebo controlled randomised trial
We would like to congratulate Sung V. and coworkers on their well-accomplished study [1], however, we would like to make some comments.
The article by Sung analyzed the use of probiotics for the treatment of infantile colic, but many confounding factors have not been considered and in addition the analysis is not properly stratified appropriate.
We would like to make also some comments concerning the study design and the statistical analysis. First of all, it seems that the sample size was calculated in order to perform a two sample Wilcoxon rank sum test so, according to the quotation reported on Sung’s protocol [2,3], it had to be at least of 150 subjects (75 for each group) and not 127, that is the number of studied infants as reported in the article. In addition in the analysis of the results the Authors have also used a multiple Regression and the necessary sample size for this evaluation has not been calculated. Details on linear regression employed are lacking (such as normal distribution of the dependent variable, adequacy of the linear model, standard error of the estimation and Adjusted-R2). The primary outcome evaluated in Sung’s study was different from ours previous study, because she considered crying and fussing time only in certain time points, while we used continuous recording [4, 5]. Barr et al. reported that the two concept of “cry” and “fussing” should not be considered synonymous. Crying is a marker of severity for abdominal pain, while fussing is a marker of distress [6]. In our experience, fussing is really difficult to be objectified by parents. Lastly, in the meta-analysis, which shows anyway an overall effectiveness of the probiotic, Sung et al. cited some data referred to our past trials, but we published medians with interquartile ranges, not means and standard deviations as reported in this paper. We would be really grateful if the Authors could clarify this issue.
Another hypothetical cause of these results could be the group of formula fed infants. There were no details concerning the types of infant formulas that were given to the formula fed infant population (which is the majority of the enrolled population): in the probiotic group there were 52 formula fed infants, of which 18 subjects (34%) received a formula with probiotics and 7 received a hypoallergenic formula but the reason of these dietary treatments is not clarified; in addition, the Authors declared in the inclusion criteria that infants with an allergy to cow's milk protein were excluded from the study (about 25% of colicky infants is affected by cow’s milk proteins allergy).Surprisingly, also in the “Placebo“ group we could find 15 infants receiving a formula added with probiotics, but the Authors don’t explain why.
In our experience, infantile colic is a clinical entity with a wide range of presentations and outcomes: from low to severe intensity and associated or not to atopy. In colicky infants affected by cow’s milk protein allergy better results could be achieved if probiotics are used as an adjunct to a monitored low allergen maternal diet avoiding cow’s milk and dairy food and with appropriate intake of vitamins and minerals. A period of at least two weeks is necessary to check the effectiveness of the diet and it has to be continued only if effective.
So, we have to consider that the Authors excluded from the study a consistent group of colicky infants (affected by cow’s milk proteins allergy) and included a group of infants treated with proton pump inhibitor drugs: it’s really surprising that 21 subjects in the probiotics group and 24 in the placebo group received proton pump inhibitor medication and the Authors don’t give any details about this. Otherwise, these 45 subjects suffered of gastroesophageal reflux and not of infantile colic.
It might well be that the lack of significant difference between the two groups in Sung’s study is due to the risk of bias for the allocation concealment. Recruiting infants contemporarily treated with drugs (such as proton pump inhibitor) and dietary approaches (formula with probiotic and hypoallergenic formula) introduced many confounding factors that needed to be evaluated individually in order to perform an appropriate multivariate analysis and to distinguish the effects of the tested probiotic from the others [7].
We have recently underlined that there is an over prescription of these drugs, usually employed in order to treat a common and physiologic condition as regurgitation in early infancy [8].
Parents have to consult a pediatrician who is able to distinguish infantile colic from other more severe conditions presenting with inconsolable crying: if they go to hospital, they will probably receive an over medication. Children appear to be enrolled in the Sung’s Study from different settings, including hospital (73% from the Royal Children’s Hospital emergency department).
When the diagnosis of infant colic is confirmed, the pediatrician has to provide parents with reassurance and offer general advice, emphasizing the favourable and self-limiting nature of the condition. In the meantime a well-tolerated, multifactorial and personalized strategy should be adopted in order to provide safe and effective therapeutic approach.
The lack of consensus about infantile colic management in medical literature and the evidence of controversial advice in many media outlets, such as websites and magazines, suggest us that it is necessary to produce high-quality studies with quantifiable and objective primary outcomes in order to develop specific clinical guidelines and new and more effective treatments (i.e. Probiotics) , such as evidence based medicine (Cochrane) [9].
Further, studies are needed in order to understand how probiotics act and the role of different species of probiotics on symptoms of infantile colic and on gut microbiota using more recent techniques, such as new generation techniques for metagenome sequencing and analysis.
Our current knowledge of the intestinal microbiota and its role in infant colic is weak. The question is if infantile colic is a cause or consequence of an alteration in gut microbiota. Illustrating the complexity of the interaction, the gut microbiota has been described to be involved in the regulation of various host metabolic pathways, giving rise to interactive host-microbiota metabolic, signalling, and immune-inflammatory axes that physiologically link the gut to the brain.
The research of E.coli in order to evaluate the microbial diversity may not be enough: in a recent study performed using 454-pyrosequencing analysis [10] we observed that Lactobacillus reuteri DSM 17938 did not affect the global composition of the microbiota, however, we noticed that responders (infants showing a reduction of the average crying time > 50%) had an increased level of Bacteroidetes. In addition, Authors don’t explain which method was employed for the evaluation of fecal E.coli and why they performed this analysis only in a subgroup of infants (31 in the treatment group and 34 in the control one) without taking in account the type of feeding (exclusively breastfed, formula with probiotic, hypoallergenic formula, normal infant formula). Furthermore, the case series for microbial analysis is very small.
Each trial evaluates the effectiveness of a single strain of probiotics and gives back results concerning that specific probiotics in a specific setting and they cannot be extrapolated to the effects of other strains.
It would be interesting if the authors would clarify these issues discussed above and also reflect on the difference between their results and the results of our two RCT studies [4,5] that found significant differences between the probiotics group and control group.
References
1. Sung V, Hiscock H, Tang MLK, Mensah FK, Heine RG, Stock A, et al. Treating infant colic with the probiotics Lactobacillus reuteri: double blind, placebo controlled randomized trial. BMJ 2014;348:g2107.
2. Sung V, Hiscock H, Tang ML, Mensah FK, Heine RG, Stock A, et al. Probiotics to improve outcomes of colic in the community: protocol for the Baby Biotics randomised controlled trial. BMC Pediatr 2012;12:135.
3. GraphPad Statistics Guide: Sample size for nonparametric tests. http://www.graphpad.com/guides/prism/6/statistics/index.htm?stat_sample_...
4. Savino F, Pelle E, Palumeri E, Oggero R, Miniero R. Lactobacillus reuteri (American Type Culture Collection Strain 55730) versus simethicone in the treatment of infantile colic: a prospective randomized study. Pediatrics. 2007; 119:e124-30.
5. Savino F, Cordisco L, Tarasco V, Palumeri E, Calabrese R, Oggero R, Roos S, Matteuzzi D. Lactobacillus reuteri DSM 17938 in infantile colic: a randomised double blind, placebo controlled trial. Pediatrics 2010 Sep;126(3):e526-33.
6. Barr RG, Kramer MS, Boisjoly C, McVey-White L, Pless IB. Parental diary of infant cry and fuss behaviour. Arch Dis Child 1988;63:380-87.
7. Savino F, Tarasco V, SorrentiM, Lingua C, Moja L, GordonM, Biagioli E. Dietary modifications for infantile colic. Cochrane Database of Systematic Reviews 2014, Issue 3. Art. No.: CD011029. DOI: 10.1002/14651858.CD011029.
8. Savino F, Castagno E. Overprescription of antireflux medications for infants with regurgitation. Pediatrics. 2008 May;121(5):1070.
9. Praveen V, Praveen S, Deshpande G, Patole SK. Oral probiotics for infantile colic. Cochrane Database of Systematic Reviews 2014, Issue 3. Art. No.: CD010986. DOI: 10.1002/14651858.CD010986.
10. Roos S, Dicksved J, Tarasco V, Locatelli E, Ricceri F, Grandin U, Savino F. “454 pyrosequencing analysis on faecal samples from a randomized DBPC trial of colicky infants treated with Lactobacillus reuteri DSM 17938”. PLoS One. 2013;8(2):e56710.
Rapid Response:
Re: Treating infant colic with the probiotic Lactobacillus reuteri: double blind, placebo controlled randomised trial
We would like to congratulate Sung V. and coworkers on their well-accomplished study [1], however, we would like to make some comments.
The article by Sung analyzed the use of probiotics for the treatment of infantile colic, but many confounding factors have not been considered and in addition the analysis is not properly stratified appropriate.
We would like to make also some comments concerning the study design and the statistical analysis. First of all, it seems that the sample size was calculated in order to perform a two sample Wilcoxon rank sum test so, according to the quotation reported on Sung’s protocol [2,3], it had to be at least of 150 subjects (75 for each group) and not 127, that is the number of studied infants as reported in the article. In addition in the analysis of the results the Authors have also used a multiple Regression and the necessary sample size for this evaluation has not been calculated. Details on linear regression employed are lacking (such as normal distribution of the dependent variable, adequacy of the linear model, standard error of the estimation and Adjusted-R2). The primary outcome evaluated in Sung’s study was different from ours previous study, because she considered crying and fussing time only in certain time points, while we used continuous recording [4, 5]. Barr et al. reported that the two concept of “cry” and “fussing” should not be considered synonymous. Crying is a marker of severity for abdominal pain, while fussing is a marker of distress [6]. In our experience, fussing is really difficult to be objectified by parents. Lastly, in the meta-analysis, which shows anyway an overall effectiveness of the probiotic, Sung et al. cited some data referred to our past trials, but we published medians with interquartile ranges, not means and standard deviations as reported in this paper. We would be really grateful if the Authors could clarify this issue.
Another hypothetical cause of these results could be the group of formula fed infants. There were no details concerning the types of infant formulas that were given to the formula fed infant population (which is the majority of the enrolled population): in the probiotic group there were 52 formula fed infants, of which 18 subjects (34%) received a formula with probiotics and 7 received a hypoallergenic formula but the reason of these dietary treatments is not clarified; in addition, the Authors declared in the inclusion criteria that infants with an allergy to cow's milk protein were excluded from the study (about 25% of colicky infants is affected by cow’s milk proteins allergy).Surprisingly, also in the “Placebo“ group we could find 15 infants receiving a formula added with probiotics, but the Authors don’t explain why.
In our experience, infantile colic is a clinical entity with a wide range of presentations and outcomes: from low to severe intensity and associated or not to atopy. In colicky infants affected by cow’s milk protein allergy better results could be achieved if probiotics are used as an adjunct to a monitored low allergen maternal diet avoiding cow’s milk and dairy food and with appropriate intake of vitamins and minerals. A period of at least two weeks is necessary to check the effectiveness of the diet and it has to be continued only if effective.
So, we have to consider that the Authors excluded from the study a consistent group of colicky infants (affected by cow’s milk proteins allergy) and included a group of infants treated with proton pump inhibitor drugs: it’s really surprising that 21 subjects in the probiotics group and 24 in the placebo group received proton pump inhibitor medication and the Authors don’t give any details about this. Otherwise, these 45 subjects suffered of gastroesophageal reflux and not of infantile colic.
It might well be that the lack of significant difference between the two groups in Sung’s study is due to the risk of bias for the allocation concealment. Recruiting infants contemporarily treated with drugs (such as proton pump inhibitor) and dietary approaches (formula with probiotic and hypoallergenic formula) introduced many confounding factors that needed to be evaluated individually in order to perform an appropriate multivariate analysis and to distinguish the effects of the tested probiotic from the others [7].
We have recently underlined that there is an over prescription of these drugs, usually employed in order to treat a common and physiologic condition as regurgitation in early infancy [8].
Parents have to consult a pediatrician who is able to distinguish infantile colic from other more severe conditions presenting with inconsolable crying: if they go to hospital, they will probably receive an over medication. Children appear to be enrolled in the Sung’s Study from different settings, including hospital (73% from the Royal Children’s Hospital emergency department).
When the diagnosis of infant colic is confirmed, the pediatrician has to provide parents with reassurance and offer general advice, emphasizing the favourable and self-limiting nature of the condition. In the meantime a well-tolerated, multifactorial and personalized strategy should be adopted in order to provide safe and effective therapeutic approach.
The lack of consensus about infantile colic management in medical literature and the evidence of controversial advice in many media outlets, such as websites and magazines, suggest us that it is necessary to produce high-quality studies with quantifiable and objective primary outcomes in order to develop specific clinical guidelines and new and more effective treatments (i.e. Probiotics) , such as evidence based medicine (Cochrane) [9].
Further, studies are needed in order to understand how probiotics act and the role of different species of probiotics on symptoms of infantile colic and on gut microbiota using more recent techniques, such as new generation techniques for metagenome sequencing and analysis.
Our current knowledge of the intestinal microbiota and its role in infant colic is weak. The question is if infantile colic is a cause or consequence of an alteration in gut microbiota. Illustrating the complexity of the interaction, the gut microbiota has been described to be involved in the regulation of various host metabolic pathways, giving rise to interactive host-microbiota metabolic, signalling, and immune-inflammatory axes that physiologically link the gut to the brain.
The research of E.coli in order to evaluate the microbial diversity may not be enough: in a recent study performed using 454-pyrosequencing analysis [10] we observed that Lactobacillus reuteri DSM 17938 did not affect the global composition of the microbiota, however, we noticed that responders (infants showing a reduction of the average crying time > 50%) had an increased level of Bacteroidetes. In addition, Authors don’t explain which method was employed for the evaluation of fecal E.coli and why they performed this analysis only in a subgroup of infants (31 in the treatment group and 34 in the control one) without taking in account the type of feeding (exclusively breastfed, formula with probiotic, hypoallergenic formula, normal infant formula). Furthermore, the case series for microbial analysis is very small.
Each trial evaluates the effectiveness of a single strain of probiotics and gives back results concerning that specific probiotics in a specific setting and they cannot be extrapolated to the effects of other strains.
It would be interesting if the authors would clarify these issues discussed above and also reflect on the difference between their results and the results of our two RCT studies [4,5] that found significant differences between the probiotics group and control group.
References
1. Sung V, Hiscock H, Tang MLK, Mensah FK, Heine RG, Stock A, et al. Treating infant colic with the probiotics Lactobacillus reuteri: double blind, placebo controlled randomized trial. BMJ 2014;348:g2107.
2. Sung V, Hiscock H, Tang ML, Mensah FK, Heine RG, Stock A, et al. Probiotics to improve outcomes of colic in the community: protocol for the Baby Biotics randomised controlled trial. BMC Pediatr 2012;12:135.
3. GraphPad Statistics Guide: Sample size for nonparametric tests. http://www.graphpad.com/guides/prism/6/statistics/index.htm?stat_sample_...
4. Savino F, Pelle E, Palumeri E, Oggero R, Miniero R. Lactobacillus reuteri (American Type Culture Collection Strain 55730) versus simethicone in the treatment of infantile colic: a prospective randomized study. Pediatrics. 2007; 119:e124-30.
5. Savino F, Cordisco L, Tarasco V, Palumeri E, Calabrese R, Oggero R, Roos S, Matteuzzi D. Lactobacillus reuteri DSM 17938 in infantile colic: a randomised double blind, placebo controlled trial. Pediatrics 2010 Sep;126(3):e526-33.
6. Barr RG, Kramer MS, Boisjoly C, McVey-White L, Pless IB. Parental diary of infant cry and fuss behaviour. Arch Dis Child 1988;63:380-87.
7. Savino F, Tarasco V, SorrentiM, Lingua C, Moja L, GordonM, Biagioli E. Dietary modifications for infantile colic. Cochrane Database of Systematic Reviews 2014, Issue 3. Art. No.: CD011029. DOI: 10.1002/14651858.CD011029.
8. Savino F, Castagno E. Overprescription of antireflux medications for infants with regurgitation. Pediatrics. 2008 May;121(5):1070.
9. Praveen V, Praveen S, Deshpande G, Patole SK. Oral probiotics for infantile colic. Cochrane Database of Systematic Reviews 2014, Issue 3. Art. No.: CD010986. DOI: 10.1002/14651858.CD010986.
10. Roos S, Dicksved J, Tarasco V, Locatelli E, Ricceri F, Grandin U, Savino F. “454 pyrosequencing analysis on faecal samples from a randomized DBPC trial of colicky infants treated with Lactobacillus reuteri DSM 17938”. PLoS One. 2013;8(2):e56710.
Competing interests: No competing interests