Probiotics for the Treatment of Infantile Colic: A Systematic Review

Abstract

Objective:

To evaluate whether clinical data support the safety and efficacy of probiotics for the management of infantile colic.

Background:

Probiotics have been suggested as a potential strategy for infantile colic, and the specific species that have been studied in healthy infants are considered to be safe.

Methodology:

A systematic review was conducted to identify randomized controlled trials (RCTs) evaluating the use of probiotic supplementation in infants with colic. RCTs with a primary end point assessing crying or fussing time were selected. A meta-analysis comparing “responders” to “nonresponders” in infants receiving probiotic versus control was conducted. The quality of trials selected was assessed.

Results:

Five RCTs assessing 2 different strains of the probiotic Lactobacillus reuteri in mostly breastfed infants were identified. Analysis of response rates showed that infants receiving probiotics had a 2.3-fold greater chance of having a 50% or greater decrease in crying/fussing time compared to controls (P = .01). Probiotic supplementation was not associated with any adverse events.

Conclusion:

Supplementation with the probiotic L. reuteri in breastfed infants appears to be safe and effective for the management of infantile colic. Further research is needed to determine the role of probiotics in infants who are formula-fed.

Keywords

pediatrics probiotics infants colic

Introduction

Infantile colic is defined as excessive crying or fussing with no accompanying apparent cause.^1,2 Other terms used to describe infant colic have included “paroxysmal fussing,” “irritable or compulsive crying,” “uncontrollable crying,” or “inconsolable crying.”^3
–5 It was first defined by Wessel et al in 1954 as crying and/or fussing for 3 or more hours per day, for 3 or more days per week, for 3 or more weeks in an infant less than 3 months of age, who is healthy and well-fed.³ The modified Wessel criteria, commonly used to define infantile colic in clinical trials, is similar except the duration of crying is only 1 week rather than 3 or more weeks.¹ The prevalence of infantile colic varies widely.⁵ It is estimated to affect as many as 20% of infants less than 3 months of age^1,2 and is most likely to occur in infants 2 weeks to 4 months of age.^5,6

Despite 50 years of research, the pathogenesis of infantile colic is still unclear.^1,2,4,5 A number of proposed etiologies have been described including altered gut flora, gastrointestinal inflammation, and an immature nervous system.⁶ Psychosocial theories, although more controversial, have also been proposed, which describe maternal anxiety, inadequate maternal–infant interaction, or challenging infant temperament as potential etiologies.^6,7 Although risk factors for the development of colic are somewhat unknown, maternal smoking, increased maternal age, and firstborn status have been proposed to be associated with the development of infantile colic.^4,6 In addition, it is thought infantile colic may be related to cow’s milk protein allergy or atopy.⁶ The cause is most likely multifactorial, as it has not been determined whether either physiological or psychosocial factors are responsible for the condition.^1,5

Although colic is generally regarded as a benign and self-limiting condition, it can result in anxiety, frustration, and stress for the caregivers as well as visits to healthcare providers for assessment of the infant.⁴ A variety of management modalities have been proposed for infantile colic. Changes in feeding techniques and the use of soothing techniques are considered first line. For example, changes in feeding techniques for a bottle-fed infant would include feeding the infant in a vertical position with frequent burping. Soothing techniques include interventions such as rocking the infant, giving the infant a warm bath, rubbing the infant’s abdomen, or allowing the infant to suck on a pacifier. Although evidence for these interventions is limited, they are commonly used first line as they are inexpensive and not harmful. Parental support is also considered a main component of therapy. This includes educating the parent that colic is common, providing emotional support regarding the challenges of comforting the infant, and teaching techniques to soothe the infant. As there is no strong evidence for one specific treatment strategy, multiple interventions may be recommended to help manage an infant with colic. Other potential interventions include changes in mother’s diet for breastfed infants, trial of a hydrolyzed protein formula for formula-fed infants, or simethicone.⁸

Probiotics have also been suggested as a potential strategy for infantile colic.¹ Probiotics are supplements or food products taken orally, that contain a sufficient quantity of live organisms to cause a change in the host flora.⁵ Lactobacillus and Bifidobacterium are the most frequently used organisms in probiotics and are considered to be safe and well tolerated in healthy infants; however, infants receiving these probiotics should still be monitored for adverse effects such as constipation, vomiting, and skin reactions.^5,9

–12

The most recent systematic reviews and meta-analyses on the use of probiotics for infants with colic were published in 2013.^2,13 An update to the Sung et al’s meta-analysis was performed in 2014 with the inclusion of one additional clinical trial.⁷ Due to the limited number of infants and small number of trials evaluated in these meta-analyses, an updated review is warranted. This systematic review and meta-analysis addressed whether probiotics are safe and effective for the management of infantile colic.

Methods

Search Strategy

A systematic review was conducted following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. A systematic literature search was conducted to identify all relevant literature evaluating the use of probiotics for infantile colic. Four electronic databases including PubMed, Google Scholar, The Cochrane Library, and Natural Medicines Comprehensive Database were searched for articles published from 1947 to December 2014.

PubMed was searched with the following: (“Probiotics”[MeSH] OR lactobacillus OR bifidobacterium OR saccharomyces OR streptococcus OR enterococcus OR bacillus) AND (“Colic”[MeSH] OR crying OR fussing OR irritable), with the filters of clinical trial, meta-analysis, humans, English, and infant. Twenty-six results were retrieved with limits applied. The search was also done with removal of the limits to identify any new articles, which had not yet been indexed by article type and therefore may not have been retrieved from the first search. The second search results were sorted by date, with the most recent 20 results retrieved from this search reviewed. Only the most recent results were reviewed to identify relevant literature not yet indexed by article type or other filters, as only recently added articles would not yet have article type descriptors or filters. Google Scholar was searched using the following phrase: “probiotic OR bifidobacterium OR lactobacillus OR saccharomyces OR streptococcus OR enterococcus OR bacillus” with the filter “intext: colic.” Results retrieved were sorted by relevance by selecting “sort by relevance” from the side bar. The first 50 results retrieved were reviewed, as subsequent results retrieved were deemed to be unrelated to the topic. The Cochrane Library was searched using the keywords “probiotic,” “colic,” and “infant” with the limit of “trials” applied. All results retrieved were evaluated for inclusion. Natural Medicines Comprehensive Database was searched using the term “probiotics,” and the monographs of Lactobacillus, Bifidobacteria, and Saccharomyces boulardii were reviewed for relevant resources. These probiotic monographs were reviewed as these are the most commonly used probiotics. An additional search using the terms “probiotic colic” was also done.

Published meta-analyses and reference lists from articles selected for inclusion were also used as means to identify additional studies. One investigator was responsible for the development of search methodology, with a second investigator reviewing the search terminology and techniques for completeness and accuracy.

Inclusion/Exclusion Criteria

Randomized controlled trials (RCTs) assessing any probiotic supplement compared to placebo or simethicone administered to an infant with a primary end point of crying or fussing time were selected for inclusion. Infants had to have a diagnosis of infant colic and could have been breastfed, formula-fed, or both. Published trials written in English evaluating infants from any geographical location were included. Trials evaluating the use of probiotics in preterm infants, very low birth weight infants, or healthy infants without colic were excluded. Studies examining probiotics for the prevention of colic or assessing formula containing probiotics were also excluded. After applying the inclusion/exclusion criteria, all relevant articles identified by the primary investigator were assessed independently for inclusion/exclusion criteria by a second investigator.

Quality of Studies

Only published trials were selected for inclusion to allow for analysis of study bias. The Cochrane Collaboration’s tool for assessing risk of bias in randomized trials¹⁴ was used to evaluate potential study bias. Two reviewers independently assessed study bias with differences adjudicated by a third reviewer.

Data Analysis

A qualitative evaluation of the data was conducted and reviewed in narrative form. A meta-analysis was also conducted using studies that included responder rate or treatment success outcomes. Infants with treatment success or considered to be responders had at least a 50% decrease in crying/fussing time from baseline. For each of the studies, the number of infants in each group (probiotic and control) and the percentage of infants considered to be responders at the end of treatment for each group (probiotic and control) were imputed into the meta-analysis. Study results were analyzed using a random-effects model. Comprehensive Meta-analysis, version 2.0, statistical software was used for the analysis.

Results

Study Selection

Figure 1 provides the PRISMA diagram of studies identified. Five RCTs evaluating 2 different strains of Lactobacillus reuteri in a total of 444 infants with colic were selected for inclusion (Table 1).^7,9

–12 Three hundred eighty-eight infants completed the trials. No studies were identified evaluating other probiotic species.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram.

Table 1.

Characteristics of Studies Included in the Meta-Analysis.^7,9

–12

Author and Year	Study Demographics, Patient Population (Inclusion/Exclusion Criteria)	Intervention and Comparator (Supplied by)	Primary Outcome	Results of Primary Outcome
Savino et al, 2007¹¹	Prospective, open-label, randomized controlled study, April 2004 to May 2005, children’s hospital in Italy, 90 breastfed infants, mothers were to follow a cow’s milk-free diet (exclusively breastfed infants 21 to 90 days of age with birth weights 2500 to 4000 g; excluded infants with chronic illness, GI disorders, or receipt of antibiotics or probiotics in week prior to recruitment)	Lactobacillus reuteri (ATCC 55730) 10⁸ CFU given 30 minutes after feeding once per day for 28 days versus simethicone 60 mg/d given twice daily after feeding for 28 days (BioGaia AB, Stockholm, Sweden)	Reduction in daily average crying time from baseline to 28 days to <3 hours/d. Parents recorded daily number of inconsolable crying episodes and the duration in a structured diary.	Eighty-three infants completed study, day 28 median crying time: probiotic group: 51 min/d (26-105 min/d), simethicone: 145 min/d (70-191 min/d); difference: −94 min/d (P < .001)
Savino et al, 2010⁹	Randomized, placebo-controlled, double-blind study, March 2008 to August 2009, general pediatricians and outpatients at the Department of Pediatrics, University of Turin, Italy, 50 breastfed infants, mothers were to follow a cow’s milk-free diet (exclusively breastfed infants 2 to 16 weeks of age, born at term, birth weights 2500 to 4000 g; excluded infants with chronic illness, GI disorder, receipt of antibiotics, and/or probiotics in a week prior to recruitment or any formula feeding)	L reuteri DSM 17938 (10⁸ CFU) versus matching placebo. Both administered once a day, 30 minutes before feeding in the morning for 21 days (unclear who supplied product, likely BioGaia AB, as they funded the study)	Reduction in average crying time to <3 hours/d on day 21. Parents recorded daily crying time in a structured diary.	Forty-six infants completed study. Day 21 median crying time: probiotic group: 35 min/d (IQR: 85); placebo group: 90 min/d (IQR: 148); difference: −55 min/d (P = .022)
Szajewska et al, 2013¹⁰	Randomized, double-blind, placebo-controlled study, January 2010 to December 2011, family primary care practice in Warsaw, Poland, 82—exclusively or >50% of time breastfed infants (full-term infants <5 months of age were included; excluded infants with acute or chronic illness, GI disorders, or use of antibiotic/probiotic within 7 days prior to enrollment)	L reuteri DSM 17938 (10⁸ CFU) versus matching placebo. Both administered once a day for 21 days (BioGaia AB, Lund, Sweden)	(1) Proportion of infants achieving “treatment success” (a reduction in the daily average crying time of 50% or more) at days 7, 14, 21, and 28; (2) duration of crying per day at 7, 14, 21, and 28 days after randomization. Parents recorded daily duration of crying time in a diary.	(1) Treatment success was significantly greater in the infants receiving probiotics versus placebo at days 7,14, 21, and 28 (all P < .05); (2) duration of median crying time in (min/d) was significantly less in the probiotic infants versus placebo infants at days 14, 21, and 28 (P < .0001).
Sung et al, 2014⁷	Randomized, double-blind, placebo-controlled trial August 2011 to August 2012, primary and secondary level care centers in Melbourne, Australia (recruitment mainly from hospital ED), 167 breastfed or formula-fed infants (included healthy, term infants <13 weeks of age; excluded infants with failure to thrive, <2500 g at birth, major medical problems, allergy to cow’s milk protein, infants who had taken solids, antibiotics, or L reuteri at start of trial, breastfed infants whose mother took L reuteri at beginning of trial, and infants with caregivers who did not have sufficient English to complete questionnaire)	L reuteri DSM 17938 (10⁸ CFU) versus matching placebo. Once daily for 28 days at same time each day (BioGaia, Sweden)	Daily duration of cry or fuss at 1 month (28 days). Parents recorded daily crying/fussing times in well-validated baby day dairy.	Total daily mean cry or fuss time at 28 days—probiotic group: 229 min/d; placebo group: 191 min/d; adjusted mean difference: +49 min/d (P = .02) The mean was adjusted for sex, age, delivery mode, feeding type, family history of atopy, and respective baseline variable. Difference mainly due to more fussing in the probiotic group (52 min/d more, P = .002)
Chau et al, 2015¹²	Randomized, double-blind, placebo-controlled trial, February 2012 to April 2014, The Hospital for Sick Children and pediatric care clinics in Toronto, Ontario, Canada, 52 breastfed infants, no elimination diet required for mother. (exclusively breastfed term infants age 3 weeks to 6 months with 5-minute Apgar score of ≥7 and birth weight ≥2500 g; excluded infants with major medical problem, acute illness, gastroesophageal reflux, history of antibiotic treatment during or before study, history of probiotic supplementation, history of allergies to ingredients in probiotic or placebo, and infants who were currently participating in another trial)	L reuteri DSM 17938 (10⁸ CFU) versus placebo (same ingredients but no live bacteria). Once daily for 21 days at same time each day (Ferring Canada, Inc)	Reduction in the duration of average crying and fussing times from baseline to day 21 to <3 hours/d. Parents recorded frequency of colic episodes and daily cry/fuss time in a structured 21-day maternal diary.	At day 21, there was a significant difference in median duration of crying/fussing time (min/d) between the placebo and probiotic group, with the probiotics infants crying less (102 min/d vs 60 min/d; P = .045). In addition, there was a significant difference in overall total mean crying/fussing time (in minutes) over the 21 days with infants receiving probiotics crying less (1719 vs 2195 minutes; P = .028)

Abbreviations: CFU, colony-forming units; ED, emergency department; GI, gastrointestinal; IQR, interquartile range.

Study Interventions

Four of the 5 studies evaluated L reuteri DSM 17938.^7,9,10,12 The remaining study evaluated a parent strain of this organism, L reuteri American Type Culture Collection (ATCC 55730).¹¹ All studies evaluated a dose of 10⁸ colony-forming units (CFUs) given once daily for a duration of either 21 or 28 days.^7,9

–12 In one trial, the control group received active treatment with simethicone.¹¹ In the other 4 trials, the control group received placebo.^7,9,10,12

Study Characteristics

The majority of infants in the studies were breastfed (∼82%). Three of the 5 trials evaluated infants who were exclusively breastfed.^9,11,12 Only Sung et al included formula-fed infants. The type of formula was not specified, infants could have received hypoallergenic or probiotic-containing formula.⁷ The Szajewska et al’s trial included infants who were exclusively or predominantly (>50%) breastfed.¹⁰ In the Savino et al’s trials, the infants’ mothers were following a cow’s milk-free diet.^9,11

All of the trials excluded infants with chronic illness, gastrointestinal disorders, or major medical problems.^7,9

–12 Only 1 trial (Sung et al) excluded infants with an allergy to cow’s milk protein.⁷ Additional details regarding the inclusion/exclusion criteria of infants enrolled in the trials are listed in Table 1.

Infants evaluated in the trials were less than 6 months of age. All trials provided the mean or median age of the infants at enrollment. The age of infants at enrollment was similar between the treatment and the control group for all 5 trials.^7,9

–12 Table 2 provides the age at enrollment for each of the trials.

Table 2.

Age of Infants at Enrollment.^a

	Reporting of Age	Probiotic Group	Control Group	P
Savino et al¹¹	Median	31.0 days	31.5 days	.955
Savino et al⁹	Median	32.5 days	28.5 days	.382
Szajewska et al¹⁰	Both	Mean: 34.3 days, Median: 33 days	Mean: 38.1 days, Median: 35.5 days	Not reported; Not reported
Sung et al⁷	Mean	52.5 days	48.3 days	Not reported
Chau et al¹²	Mean	42.1 days	41.1 days	Not reported

^a All trials provided the mean or median age of the infants at enrollment.

Infantile colic was defined similarly among the 5 trials on the basis of the modified Wessel criteria. Crying/fussing times and any observed side effects were recorded by caregivers in diaries. Three of the trials assessed only “crying time,”^9
–11 and the 2 remaining trials^7,12 assessed “crying/fussing time.” Fussing was defined as not crying but not awake and content either.⁷ Only Sung et al allowed for the discernment of crying from fussing.⁷ Baseline crying and/or fussing times were similar between treatment groups for each individual trial. Baseline crying times ranged from 122 min/d (Chau et al) to 370 min/d (Savino et al, 2010). The majority of trials reported median crying time, however, Sung et al reported mean crying/fussing time. Chau et al also included “fussing” time in the assessment of baseline crying times.^7,9

–12

Overall, the studies reviewed had a low risk of bias and were determined to be high in quality. One of the studies had an unclear risk of bias.⁷ One had a high risk of bias due to lack of blinding and other variables, that could have affected outcome assessment.¹¹ Three trials were determined to have a low risk of bias.^9,10,12 Table 3 provides details regarding the assessment of risk of bias.

Table 3.

Assessment of Risk of Bias.^a

	Randomization	Allocation Concealment	Blinding of Participants and Personnel	Blinding of Outcomes	Incomplete Outcome Data Addressed	Selective Reporting	Other Bias	Overall Risk of Bias
Savino et al¹¹	+	?	−	?	+	+	−	High
Savino et al⁹	+	+	+	+	+	+	−	Low
Szajewska et al¹⁰	+	+	+	+	+	+	+	Low
Sung et al⁷	+	+	+	+	?	+	−	Unclear
Chau et al¹²	+	+	+	+	+	+	+	Low

^a Risk of study bias was assessed by 2 independent reviewers using the Cochrane Collaboration’s tool for assessing risk of bias in randomized trials. Differences were adjudicated by a third reviewer. + denotes low risk of bias; ?, unclear risk of bias; −, high risk of bias.

Crying Time

Three studies evaluated median end of treatment infant crying time. Probiotic supplementation compared to controls significantly reduced median crying time in these studies (Savino et al, 2007: −94 min/d, P < .001; Savino et al, 2010: −55 min/d, P = .022; Szajewska et al: −53 min/d, P < .0001).^9
–11 The other 2 trials evaluated mean or median cry/fuss time at the end of treatment, day 28 or day 21, respectively (Sung et al, 2014 and Chau et al, 2015).^7,12 The Chau et al’s trial demonstrated similar findings as the Savino et al’s trials and Szajewska et al’s trial, with infants who received probiotic have significantly less crying/fussing (−42 min/d; P = .045) than those who received control.¹² However, for the Sung et al’s trial, infants in the probiotic group cried or fussed an adjusted mean of 49 minutes more per day than infants receiving control (P = .02).⁷

Responder Rate

All 5 trials defined “responders” or “treatment success” as infants with a 50% or greater reduction in crying/fussing time from baseline.^7,9

–12 For 1 of the trials, this was a primary end point,¹⁰ for 3 of the trials a secondary end point,^9,11,12 and for 1 of the trials, it was not designated as a prespecified end point.⁷ There were significantly more responders in the probiotic group for 4 of the 5 trials (Table 4).^9

–12 Only Sung et al demonstrated no significant difference between the 2 groups (P = .23).

Table 4.

Responders Versus Nonresponders for RCTs.^a

	End of Treatment	Number of Infants in the Probiotic Group	Percent Responders in the Probiotic Group	Number of Infants in the Control Group	Percent Responders in the Control Group	P
Savino et al¹¹	Day 28	41	95	42	7	.001
Savino et al⁹	Day 21	25	96	21	71	.036
Szajewska et al¹⁰	Day 21	40	97.5	40	37.5	.001
Sung et al⁷	Day 28	27	40	29	48	.23
Chau et al¹²	Day 21	24	71	28	21	.035

Abbreviation: RCT, randomized controlled trial.

^a Responders were considered to be infants with a decrease of at least 50% in the daily average crying/fussing time from baseline. The trials by Savino et al^9,11 and Szajewska et al¹⁰ only assessed reduction in “crying time,” whereas Sung et al⁷ and Chau et al¹² assessed reduction in “crying/fussing.” In the trial by Savino,¹¹ the control group received simethicone. In the other trials, the control group received placebo.

Of the 388 infants who completed the RCTs, responder rate was reported for 317 of the infants. These infants were included in the meta-analysis of responder rate. A fixed-effects model demonstrated significant heterogeneity between studies (P = .00, I2 = 86.1%); therefore, a random-effects model was performed to assess the effect of probiotics versus control on responder rate. Overall, supplementation with 10⁸ CFU of L reuteri once daily lead to a 2.3-fold greater chance of a 50% or more reduction in cry/fuss time compared to controls (P = .01). Figure 2 provides the results of the meta-analysis of responder rate.

Figure 2.

Random-effects model for randomized controlled trials (RCTs) of probiotics for the management of infantile colic, comparing responders versus nonresponders. Responders considered to be infants with a decrease of at least 50% in the daily average crying/fussing time from baseline.

Adverse Effects

For all 5 RCTs included, parents and/or caregivers were asked to record any observed side effects (ie, constipation, vomiting, skin reactions) in a diary.^7,9

–12 Only one study reported adverse events. The only adverse event occurring in an infant given probiotic was rhinitis; the other adverse events occurred in infants given placebo and included 1 infant each with eczema, fever, otalgy, and gastroesophageal reflux. All adverse events were determined by authors to be unrelated to study treatment, and additional details were not given.⁹ For the remaining 4 studies, no adverse events were reported.^7,10,11,12

Discussion

As trials conducted have only included a small number of infants in different geographic regions, a meta-analysis is an appropriate method to assess the use of probiotics for infantile colic. Our results demonstrated supplementation with 10⁸ CFU of L reuteri once daily lead to a 2.3-fold greater chance of a 50% or more reduction from baseline cry/fuss time in infants with colic compared to controls (P = .01). In addition, there were significantly more responders in the probiotic group for all of the trials except Sung et al, which found no significant difference between the 2 groups (P = .23).

One potential explanation as to why there was not a significant difference in responders between probiotic and control groups for the Sung et al’s trial was because the majority of infants in this trial were formula-fed (59%, n = 71).⁷ The other trials evaluated exclusively^9,11,12 or predominantly¹⁰ breastfed infants only. When subgroup analyses were performed for the Sung et al’s trial, it was demonstrated that the formula-fed infants who received probiotics cried/fussed an adjusted mean of 78 minutes more (95% confidence interval [CI]: 25-132; P = .005) than infants who received placebo. In contrast, of the infants who were exclusively breastfed, cry/fuss time was similar between the probiotic and placebo groups (95% CI: −49 to 87; P = .57).⁷

Other potential explanations as to why the results of the Sung et al trial differ from the other 4 trials include differences in characteristics between the 2 groups. More infants in the placebo group were receiving proton pump inhibitors (PPI) than in the probiotic group at study entry (35% vs 26%; no P values provided), and more exclusively, breastfed infants had mother’s on dairy exclusion diets at baseline and at the end of treatment in the placebo group (P = .03 at the end of treatment).⁷ If infants with colic are more susceptible to milk intolerance or if PPI use affects symptoms of colic, these differences could potentially explain the similarity between probiotic and placebo responders in this trial. Also, infants in the placebo group received 40 to 60 more minutes carrying and moving time per day than infants in the probiotic group, which could affect infant crying/fussing time (P > .05).⁷

Other previously published meta-analyses have also demonstrated a benefit of probiotic supplementation in breastfed infants with colic.^2,13 In a 2013 meta-analysis by Anabrees et al, 3 RCTs assessing 209 breastfed infants were evaluated. At day 21, there was significantly less crying time in the infants who received probiotic compared to controls (−56.03 min/d, 95% CI: −59.92 to −52.15, P < .00001). Anabrees et al also did not find any short-term safety concerns with the use of L reuteri.¹³

In a 2013 meta-analysis by Sung et al, the same 3 RCTs evaluated by Anabrees were evaluated and demonstrated a significant difference in mean crying time favoring probiotic over control (−68 min/d, 95% CI: −99.8 to −35.6).² An update to this meta-analysis was performed in 2014 with one additional trial included (Sung et al, 2014). There was still a significant difference in mean crying/fussing time at day 21 favoring probiotic over control (−48 min/d, 95% CI: −84.76 to −11.34), although the overall size of the treatment effect had decreased.⁷

Our meta-analysis includes the RCTs evaluated in the meta-analyses by Anabrees et al and Sung et al as well as a recently published RCT by Chau et al. In addition, it provides an assessment of the percentage of infants considered to be treatment responders (infants with a 50% or greater reduction in crying/fussing time from baseline). The most recent meta-analysis performed by Sung et al⁷ assessed mean difference in crying time but did not evaluate the percentage of infants who experienced treatment success (“treatment responders”). The meta-analysis by Anabrees et al¹³ did provide an assessment of treatment success but only included 2 of the 5 trials evaluated in our meta-analysis at the time points assessed.

In general, management of infantile colic is individualized, with changes in feeding techniques and use of soothing techniques considered to be first line. If these interventions are ineffective, changes in mother’s diet for breastfed infants or a trial of formula containing hydrolyzed protein for formula-fed infants can be considered. Another management option that may be used is simethicone.⁸ Although it is the most commonly prescribed treatment for infantile colic, it has been found to be no more effective than placebo in previous studies.^4,5

L reuteri appears to be a potential treatment option for breastfed infants with colic. Proposed mechanisms for its effects include improvement in gastrointestinal motility and function, anti-inflammatory effects, modulation of the immune response, and effects on visceral pain.^5,9,10 Infants with colic have been shown to have higher levels of microbes such as Escherichia coli and other gas-producing microbes within their gastrointestinal tract.⁷ It is thought diversification of gut flora and microorganisms such as Lactobacillus may guard against infant distress.⁷ In addition, Savino et al found significant reduction in fecal E Coli in infants given L reuteri but not in infants who received placebo. This suggests that L reuteri may increase gastrointestinal health in infants with colic by decreasing E coli.⁹

The products used in the clinical studies evaluated by our meta-analysis are very similar to the commercially available products: BioGaia ProTectis drops (BioGaia AB, Stockholm, Sweden) and Gerber Soothe Colic Drops Probiotic Supplement (Gerber Product Company/Nestle Infant Nutrition; Florham Park, NJ, USA). Both of these products contain the organism, strain, and dose evaluated in these clinical studies.^7,9,10,12 This organism and strain, L reuteri DSM 17938 (L reuteri Protectis), is licensed and patented by BioGaia AB. The dose evaluated was 10⁸ CFUs once daily for a duration of 21 or 28 days in infants less than 6 months of age.

Our review is not without limitations. First, only one type of probiotic was used in the studies identified, L reuteri. Therefore, our findings cannot be extrapolated to other probiotics. In addition, the safety and efficacy of a combination of probiotics were not evaluated and therefore are unknown. Secondly, the majority of infants evaluated were breastfed, therefore, our conclusions cannot be extrapolated to infants who are predominantly formula-fed. Third, as our review only evaluated trials that have been published, the potential exists for publication bias. A statistical test and funnel plot were not used to assess publication bias, as these are typically performed only when there are at least 10 studies in the analysis.

In the trials we evaluated, the placebo response, as assessed by the percentage of responders at the end of treatment, ranged from 21% to 71%.^7,9,10,12 In general, the placebo response in trials assessing interventions for infantile colic has been estimated to range between 5% and 83%.¹⁵ Although a direct placebo effect in infants is unlikely, the possibility exists for an indirect effect through changes in parent’s behavior.¹⁰ Another explanation for the placebo effect demonstrated in these trials is the physiological maturation of the infant.⁹ Mothers following a cow’s milk-free diet is another possible explanation. However, only the Savino et al’s trial required a cow’s milk elimination diet for mom, but this study did demonstrate the highest placebo response at 71%.⁹ Chau et al proposed the average age at enrollment may explain the variation in placebo response observed with an older average age at enrollment resulting in a lower placebo response.¹² For the 4 trials with a placebo group, 3 of the trials support Chau’s explanation, all except for Sung et al’s trial, which had the oldest average age at enrollment of 48.4 days and a fairly strong placebo response (48%). This large variation in the placebo response can influence the perceived size of treatment effect from the comparator agent, which in this case is a probiotic.

The generalizability of the results is limited due to a number of reasons. First, only one of the 5 trials analyzed was conducted in North America (Chau et al), and the potential exists for differences in microbial colonization in infants from differing regions of the world.¹² As a result, infants from one geographical location may vary in their response to probiotics compared with infants from other locations.¹² In addition, only 1 trial (Sung et al) evaluated infants who were predominantly formula-fed, and this trial demonstrated infants in the probiotic group cried or fussed 49 minutes more than those in the placebo group (P = .02).⁷ Therefore, the results of these trials are only generalizable to breastfed infants in the geographical regions where positive benefits were shown (Italy, Poland, and Canada).

Conclusion

Supplementation with the probiotic L reuteri administered at a dose of 10⁸ CFUs once daily to breastfed infants less than 6 months of age resulted in significantly greater improvement in colic symptoms at the end of treatment (21 or 28 days) compared to controls. Although probiotic supplementation was not associated with any adverse events in the RCTs identified, monitoring for adverse events (ie, constipation, vomiting, skin reactions) in infants given L reuteri for colic is appropriate. Further research is needed to determine the role of probiotics in infants with colic who are formula-fed as well as the effects of probiotics in treating infants from other geographical regions.

Footnotes

Authors’ Note

Dr Gregory, project director and coinvestigator, contributed to the design, oversight, assessment of risk of bias, and statistical software analysis. Dr Jalloh contributed to the assessment of risk of bias.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Gregory is the Editor-in-Chief for Natural Medicines and a paid consultant for Therapeutic Research Center. Dr. Hein is a paid consultant for Therapeutic Research Center.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Sung

Hiscock

Tang

. Probiotics to improve outcomes of colic in the community: protocol for the baby biotics randomized controlled trial. BMC Pediatr. 2012;12:135.

Sung

Collette

de Gooyer

. Probiotics to prevent or treat excessive infant crying systematic review and meta-analysis. JAMA Pediatr. 2013;167(12):1150–1157.

Wessel

Cobb

Jackson

. Paroxysmal fussing in infancy, sometimes called colic. Pediatrics. 1954;14(5):421.

Drugs and Therapeutics Bulletin. Clinical review: management of infantile colic. BMJ. 2013;347:f4102.

Praveen

Deshpande

. Oral probiotics for infantile colic (Protocol). Cochrane Database Syst Rev. 2014;(3):CD010986.

Savino

Ceratto

De Marco

. Looking for new treatments of infantile colic. Ital J Pediatr. 2014;40:53.

Sung

Hiscock

Tang

. Treating infant colic with the probiotic Lactobacillus reuteri: double-blind, placebo-controlled randomised trial. BMJ. 2014;348:g2107.

Turner

Palamountain

Infantile Colic: Management and outcome. UpToDate. Web site. www.uptodate.com. Updated November 18, 2014. Accessed December 8, 2014.

Savino

Cordisco

Tarasco

. Lactobacillus reuteri DSM 17938 in infantile colic: a randomized, double-blind, placebo-controlled trial. Pediatrics. 2010;126(3):e526–e533.

10.

Szajewska

Gyrczuk

Horvath

Lactobacillus reuteri DSM 17938 for the management of infantile colic in breastfed infants: a randomized, double-blind, placebo-controlled trial. J Pediatr. 2013;162(2):257–262.

11.

Savino

Pelle

Palumeri

. Lactobacillus reuteri (American Type Culture Collection Strain 55730) versus simethicone in the treatment of infantile colic: a prospective randomized study. Pediatrics. 2007;119(1):e124–e130.

12.

Chau

Lau

Greenberg

. Probiotics for infantile colic: a randomized, double-blind, placebo-controlled trial investigating Lactobacillus reuteri DSM 17938. J Pediatr. 2015;166(1):74–78.

13.

Anabrees

Indrio

Paes

. Probiotics for infantile colic: a systematic review. BMC Pediatrics. 2013;13:186.

14.

Higgins

Altman

Gotzsche

. The Cochrane Collaboration’s tool for assessing risk of bias in randomized trials. BMJ. 2011;343:d5928.

15.

Savino

Tarasco

. New treatments for infant colic. Curr Opin Pediatr. 2010;22(6):791–797.