• Research is hard, colic edition

    Last year, I gave a talk in Doha, Qatar on the evidence behind the treatment for infantile colic. While the word has a pretty non-specific meaning in everyday use, there are clinical criteria physicians are supposed to use in order to diagnose the problem. Wessel’s criteria declare that infant colic is defined by inconsolable crying in a child less than three months of age, for at least three hours a day, for at least three days a week, for three weeks of more. That’s colic. Anything else is “normal infant”.

    Even then, we have no idea what causes it. Moreover, pretty much every single child in the world who has it outgrows it. That doesn’t mean that real colic isn’t a problem. It’s hell to live thorough, and we are constantly looking for solutions.

    Small trials sometimes find them. There have been three such trials looking at probiotics as a treatment. Why? Cause maybe there’s something in the gut of kids with colic and maybe probiotics would help. I’m minimizing here, but that’s almost the sum of the evidence behind the theory. There’s even a meta-analysis of the trials that concluded that lactobacillus may be effective. But those three trials aren’t perfect. One of them wasn’t blinded. Two of them included only infant-mother pairs where the mothers consumed dairy-free diets. None of them used outcomes or measures which had been validated, and none of them included infants who ate formula instead of breastmilk.

    But if you read the meta-analysis, or any of the media around the publication of those studies, you’d think – as many do – that lactobacillus is great for colic. Until this week, when the BMJ published a new manuscript:

    Objective To determine whether the probiotic Lactobacillus reuteri DSM 17938 reduces crying or fussing in a broad community based sample of breastfed infants and formula fed infants with colic aged less than 3 months.

    Design Double blind, placebo controlled randomised trial.

    Setting Community based sample (primary and secondary level care centres) in Melbourne, Australia.

    Participants 167 breastfed infants or formula fed infants aged less than 3 months meeting Wessel’s criteria for crying or fussing: 85 were randomised to receive probiotic and 82 to receive placebo.

    Interventions Oral daily L reuteri (1×108 colony forming units) versus placebo for one month.

    Main outcomes measures The primary outcome was daily duration of cry or fuss at 1 month. Secondary outcomes were duration of cry or fuss; number of cry or fuss episodes; sleep duration of infant at 7, 14, and 21 days, and 1 and 6 months; maternal mental health (Edinburgh postnatal depression subscale); family functioning (paediatric quality of life inventory), parent quality adjusted life years (assessment of quality of life) at 1 and 6 months; infant functioning (paediatric quality of life inventory) at 6 months; infant faecal microbiota (microbial diversity, colonisation with Escherichia coli), and calprotectin levels at 1 month. In intention to treat analyses the two groups were compared using regression models adjusted for potential confounders.

    Both breastfed and bottle-fed infants who met strict criteria were randomized to receive either probiotics or placebo. Validated measures were used, as was an intention-to-treat-analysis. And what did they find? Over time, both groups got better with respect to mean daily cry/fuss time. That’s to be expected, as all infants outgrow colic. At one month, the probiotic group cried/fussed 49 minutes more than the placebo group. There were no differences in any of the other outcomes.

    In other words, the probiotic didn’t work when the study was more ideally designed.

    It’s hard to do good research. It’s hard to get the media, let alone the average citizen, to understand the nuances of study design. But it’s important. We have to stop chasing our tails on some of these issues. Sometimes there isn’t a good medical intervention for stuff that’s just in the range of normal development. Everything isn’t a treatable disease. I’ll let the final word go to the accompanying editorial, written by one of my superstar junior faculty (conflict of interest thus declared):

    So, with such a dearth of good evidence, perhaps the more important question is: “Should we be treating infant colic at all?” A great deal of accumulated clinical experience tells us that children with colic incur no serious long term effects from the disorder, and that symptoms abate with time. The potential harm associated with diagnostic testing and treatment of infants is likely to surpass the harm from colic itself.

    For us to continue to perform drug intervention trials for this problem perhaps underscores our unwillingness to accept that colic is likely to represent a heterogeneous disorder with many complex inputs. As the old adage goes, “babies cry.” Parents and their babies may be better served if we devote more resources to studying the interventions recommended long before the discovery of probiotics: reassurance, family social support, and the tincture of time.


    Comments closed