Two recent studies published in The New England Journal of Medicine, “Lactobacillus rhamnosus GG versus Placebo for Acute Gastroenteritis in Children” and “Multicenter Trial of a Combination Probiotic for Children with Gastroenteritis,” have received significant media coverage and are stimulating questions regarding the value of probiotics for children with gastrointestinal symptoms.
The studies were multicenter, randomized, double-blind, placebo-controlled trials, conducted in the U.S. by David Schnadower, MD, MPH, and colleagues in the Pediatric Emergency Care Applied Research Network (PECARN) and in Canada by Stephen Freedman, MDCM, and the Pediatric Emergency Research Canada (PERC) Probiotic Regimen for Outpatient Gastroenteritis Utility of Treatment (PROGUT) Trial Group. These comments are not an endorsement of these studies or their findings. They are meant to clarify the issues and offer a guide to facilitate conversations with physician colleagues, as well as parents and caregivers to pediatric patients.
Acute gastroenteritis is a burdensome condition accounting for 1.7 million emergency department visits and more than 70,000 hospitalizations among children in the U.S. each year. It also causes nonmedical burdens, such as daycare absenteeism and lost wages for parents and caregivers.
Although many clinical studies on probiotics have been conducted, in general they have been limited by suboptimal study design with small sample sizes, high overall risk of bias1 and underreporting of potential adverse events.2 As a result, clinicians have relied on meta-analyses, which combine data from multiple studies to increase overall sample size to draw conclusions regarding the use of probiotics to manage gastrointestinal symptoms. These analyses have indicated a potential benefit3 for probiotics in acute gastroenteritis, leading to guidelines supporting the use of probiotics for patients with acute gastroenteritis.4
The two studies described here investigated the effect of different probiotic formulations that include a particular bacterial species, Lactobacillus rhamnosus, on the development of moderate-to-severe gastroenteritis in children initially diagnosed with acute gastroenteritis in pediatric emergency departments in the U.S. and Canada.
RESULTS AND INTERPRETATION
The U.S. study enrolled 971 participants over 36 months, and the Canadian study enrolled 886 participants over 41 months. Both studies had high rates of follow-up and completion. The U.S. study had over 95 percent response rates for both placebo and interventional groups, while the Canadian study had over 93 percent response rates for both placebo and interventional groups.
1. Did probiotics reduce the occurrence of moderate-to-severe gastroenteritis in children?
No. There was no significant difference in the occurrence of moderate-to-severe gastroenteritis between the placebo and L. rhamnosus GG groups in the U.S. study, or between the placebo and L. rhamnosus R0011/L. helveticus R0052 groups in the Canadian trial.
In the U.S. study, moderate-to-severe gastroenteritis developed in 12 percent of children receiving L. rhamnosus GG and 13 percent of children receiving placebo. The rates were higher in Canada, with moderate-to-severe gastroenteritis developing in 26 percent of children receiving L. rhamnosus R0011/L. helveticus R0052 and 25 percent of children receiving placebo.
2. Were there any differences in the occurrence of moderate-to-severe gastroenteritis in children who received antibiotics and probiotics?
No. Children receiving antibiotics were included in the placebo and L. rhamnosus-containing probiotic groups for both studies, and no significant differences in the occurrence of moderate-to-severe gastroenteritis were identified.
3. Were there any outcomes in which probiotics had a beneficial effect compared to placebo?
No. Both studies reported on a variety of secondary outcomes such as duration of diarrhea, duration of vomiting, days of daycare missed, unscheduled visits to a health care provider and repeat visits to the emergency department. There was no significant difference between the placebo and L. rhamnosus-containing probiotic groups in any of these outcomes.
4. Were any adverse events reported in children who received probiotics?
In both studies, there were no significant differences in the rates of adverse events between the placebo and probiotics groups. In the U.S. study, wheezing was reported as a side effect in five participants in the L. rhamnosus GG group and zero participants in the placebo group (P = 0.03). The Canadian study reported two serious adverse events in the placebo group only. Specifically, there were no occurrences of extraintestinal infection with Lactobacillus in either study.
5. What are the limitations of these studies?
Both studies relied on parent/caregiver reports of adherence and symptoms, so recall bias is a potential limitation. Furthermore, since both were real-world (pragmatic) studies, parents may not have followed all instructions regarding proper storage, dosing, or administration of the probiotic products, which could have had an effect on bacterial viability. Finally, it is unclear whether results might be different within subsets of children infected by a specific virus versus those with no identified enteropathogen.
6. Despite the lack of beneficial effect, why are these studies so important?
Both of these pragmatic trials enrolled large, diverse patient populations and employed high-quality methods that aim to fully characterize potential benefit and adverse events and to minimize the risk of bias. They are reflective of commercial utilization of probiotics by consumers. The results only pertain to the two specific products used in the U.S. and Canadian trials, and we cannot conclude that other probiotic products, or other dosing strategies for the same probiotic products, would also have no benefit. However, these studies are a reminder of the potential disadvantage of relying on meta-analyses to assess beneficial effect of probiotics. These trials are also important because negative trials can give clinicians important insights into which therapies do not work and reduce bias in the overall field. Further, there is an economic cost to any intervention and the results of these studies suggest that there is no value to be gained from the two L. rhamnosus-containing probiotics in acute gastroenteritis in children. In order to determine which probiotics might benefit which clinical populations, future multicenter studies will need to be conducted on a variety of probiotic strains, products, and doses with this level of rigor in large and diverse populations.
1. Shapiro J., Bernica J., Hernaez R. Clin Gastroenterol Hepatol 2018 Aug 7. https://www.ncbi.nlm.nih.gov/pubmed/30012436
2. Bafeta A, Koh M. Riveros C, Ravaud P. Ann Intern Med. 2018 Aug 21;169(4):240-247. https://www.ncbi.nlm.nih.gov/pubmed/30014150
3. Allen SJ, Martinez EG, Gregorio GV, Dans LF. Cochrane Database Syst Rev. 2010 Nov 10;(11):CD003048. https://www.ncbi.nlm.nih.gov/pubmed/21069673
4. Guarino A et al. J Pediatr Gastroenterol Nutr. 2018 Nov;67(5):586-593. https://www.ncbi.nlm.nih.gov/pubmed/29901556