Abstract & Commentary
Probiotics for Colic
By David Kiefer, MD
Synopsis: Colicky children given probioitics had more "fussing," as per sub-group analyses than children given placebo. In addition, probiotics did not affect the amount of time spent crying.
Source: Sung V, et al. Treating infant colic with the probiotic Lactobacillus reuteri: Double blind, placebo controlled randomised trial. BMJ 2014;348:g2107.
The treatment of infantile colic can be a challenge for clinicians, parents, and patients alike. With few treatment options available, there is a search for useful and safe integrative therapies. The researchers in this study pursued one hypothesis that an abnormality of microbiota may be involved in the pathophysiology of colic. For example, the authors cite numerous studies documenting abnormal levels of proteobacteria (such as Escherichia coli) in the stool of infants with colic, and that such bacteria can produce increased flatulence and colicky symptoms. Other references cited in the paper document that supplementation with probiotics may decrease the concentrations of problematic bacteria and increase microbial diversity, the latter being independently correlated with decreased infantile distress. In addition, clinical trials show benefits of probiotics in treating childhood diarrhea.1 Hence, the rationale seems rational for their proposed intervention.
In Australian emergency rooms, the researchers recruited 167 colicky (as per Wessel’s criteria) breastfed or formula-fed infants < 3 months of age and randomized them to receive either a specific species of probiotic bacteria or placebo, and followed their symptoms for 1 month. The species used was Lactobacillus reuteri (strain DSM 17938; supplied by the manufacturer at no cost), on which some prior studies have been done, and it was dosed at 100 million colony forming units (cfu) daily. Among the exclusion criteria were infants with a cow’s milk allergy, or those who took solids, antibiotics, or L. reuteri at the start of the trial; breastfed infants whose mothers took L. reuteri also were excluded. Interestingly, the researchers did not specify whether intake of other species of probiotics, by the mothers or children, was grounds for exclusion.
The researchers were thorough, both in their methodology (double-blinding and intention-to-treat analysis, for instance) and their follow up. Adherence to the protocol was assessed through diaries kept by parents and bottle weights (the probiotics were mixed in bottle feedings). The primary outcome was the duration of cry or fuss at 1 month. They also analyzed the dichotomous variable of who achieved a 50% reduction in cry/fuss at 1 month. The same primary and secondary variables were then analyzed at 6 months.
Out of the 167 enrolled families (85 probiotic, 82 placebo), 127 completed the trial (67 probiotic, 60 placebo); baseline characteristics were similar between these groups. Over the 1 month, there was a decline in cry/fuss duration in both groups, more so in the placebo group, which, interestingly, had 49 less minutes of crying or fussing than the probiotic group (95% confidence interval, 8-20; P = 0.02). The amount of time spent crying between the two groups was the same, but the probiotic group had more fussing. These differences evaporated by the 6-month check; both groups were statistically identical in cry/fuss time. Another way to look at the data is that, at 1 month, 40% of the probiotic group and 48% of the placebo group showed at least a 50% reduction in cry/fuss time.
No differences in secondary variables were seen at 1 month, that is, sleep time, fecal microbial diversity, calprotectin levels, and E. coli load. The researchers did sub-group analyses for breastfed infants, infants aged ≥ 6 weeks, and infants from atopic families; no differences in response to probiotics was seen. However, infants < 6 weeks and those who were formula fed had a significantly increased amount of cry/fuss time (78 and 88 minutes, respectively, mostly due to increased fussing time) in the probiotic group compared to the placebo group.
No adverse effects were seen, unless you factor in the fussing. In the probiotic group, 45% of the infants were colonized with L. reuteri by the end of the study. Predictably, none of the placebo group was so colonized.
Commentary
So much for our magic bullet. The study showed not a benefit but an increase in cry/fuss time (from increased fuss time) in infants supplemented with probiotics. This effect only occurred in the formula-fed infants. One hypothesis is that recolonization, a shifting of probiotic species content in the colon, takes time and may cause symptoms. Possibly negating this explanation is that there was no significant difference in the fecal microbial diversity score between placebo and probiotic groups. Overall, this is a large, well-designed clinical trial, so we probably need to believe these results. However, does it change our clinical behavior?
Children who access Australian emergency rooms may not perfectly replicate clinic demographics elsewhere. Given that some researchers propose important genetic components to the ideal microbiome and supplement effects, this is an issue. Also, 100 million cfu is below what clinicians often prescribe and use in practice. Perhaps, the study was underdosed to see an effect, and this is partly supported by the 45% colonization rate in the probiotic group. (Why isn’t it 100%?) The researchers were meticulous in their encouragement and follow up of adherence, so the low colonization is less likely to be due to nonadherence. Therefore, if the low colonization and lack of positive clinical effect was due to underdosing, one fear is that the findings in this study might just be increased (more fussing) with higher dosing. It is probably difficult to predict the exact response to an increased dose of this probiotic species.
Putting these results into context of other research efforts, a wide variety of dosages and species have been used in childhood illnesses. For example, in studies on childhood diarrhea, probiotic species used include Lactobacillus rhamnosus, Streptococcus thermophilus, Lactobacillus casei, Bifidobacterium lactis, or L. reuteri, and these species are mixed with milk or infant formula and sometimes prescribed as an oral supplement.2 The clinical relevance of the currently reviewed study is most relevant as a small piece of the complicated probiotic puzzle that involves many variables, including species used, dose, patient demographic, and health condition.
As per the researchers, perhaps the most reasonable recommendation from this clinical trial is that this particular species, strain, and dose of probiotic should be used with caution in this demographic until more research surfaces.
Summary Point
- Compared to placebo, colicky children ingesting 100 million cfu daily of Lactobacillus reuteri had more "fussing," as per sub-group analyses; this occurred in infants younger than 6 weeks old and in formula-fed infants.
- Probiotics did not affect the amount of time spent crying.
- The difference between the probiotic group and the placebo group disappeared at 6 months after the start of the 1-month supplementation period.
References
1. Salari P, et al. A meta-analysis and systematic review on the effect of probiotics in acute diarrhea. Inflamm Allergy Drug Targets 2012;11:3-14.
2. Thomas DW, Greer FR; American Academy of Pediatrics Committee on Nutrition; American Academy of Pediatrics Section on Gastroenterology, Hepatology, and Nutrition. Probiotics and prebiotics in pediatrics. Pediatrics 2010;126:1217-1231.
