Association Between Diarrhea Duration and Severity and Probiotic Efficacy in Children With Acute Gastroenteritis

Abstract

Introduction: It is unclear whether the alleged efficacy of probiotics in childhood acute gastroenteritis depends on the duration and severity of symptoms before treatment.

Methods: Preplanned secondary analysis of 2 randomized placebo-controlled trials in children 3-48 months of age was conducted in 16 emergency departments in North America evaluating the efficacy of 2 probiotic products (Lactobacillus rhamnosus GG and a combination probiotic: L. rhamnosus and L. helveticus). Participants were categorized in severity groups according to the duration (<24, 24-<72, and ≥72 hours) and the frequency of diarrhea episodes in the 24 hours (≤3, 4-5, and ≥6) before presentation. We used regression models to assess the interaction between pretreatment diarrhea severity groups and treatment arm (probiotic or placebo) in the presence of moderate-to-severe gastroenteritis (Modified Vesikari Scale score ≥9). Secondary outcomes included diarrhea frequency and duration, unscheduled healthcare provider visits, and hospitalization.

Results: A total of 1,770 children were included, and 882 (50%) received a probiotic. The development of moderate-to-severe gastroenteritis symptoms after the initiation of treatment did not differ between groups (probiotic-18.4% [162/882] vs placebo-18.3% [162/888]; risk ratio 1.00; 95% confidence interval 0.87, 1.16; P = 0.95). There was no evidence of interaction between baseline severity and treatment (P = 0.61) for the primary or any of the secondary outcomes: diarrhea duration (P = 0.88), maximum diarrheal episodes in a 24-hour period (P = 0.87), unscheduled healthcare visits (P = 0.21), and hospitalization (P = 0.87).

Discussion: In children 3-48 months with acute gastroenteritis, the lack of effect of probiotics is not explained by the duration of symptoms or frequency of diarrheal episodes before presentation.

Trial registration: ClinicalTrials.gov NCT01773967.

Conflict of interest statement

Conflict of interest statement:

Dr. David Schnadower received in-kind study drug and placebo from iHeath Inc., however the company did not contribute financially to the study or to the investigators, and their employees do not have access study data. I-Health personnel had no role in study design, collection management, analysis and interpretation of data; nor did they have any role or authority in writing the report nor the decision to submit the trial for publication.

Dr. Stephen Freedman has received in-kind (study drug and placebo) from Lallemand Health Solutions. He provides consulting services to Takeda Pharmaceutical Company, RedHill Biopharma Ltd and Eligo Bioscience S.A.S. on childhood intestinal disorders and is supported by the Alberta Children’s Hospital Foundation Professorship in Child Health and Wellness.

All other authors report no relevant conflicts.

Copyright © 2021 by The American College of Gastroenterology.

Figures

Figure 1.

Study flow

Figure 2.

Conditional effect of the probiotic treatment within severity subgroups on outcomes: Moderate-severe gastroenteritis (MVS ≥9), repeat health care visit after discharge and prior to symptom resolution, and hospital admission resulting from index visit lasting ≥48 hours or post-discharge and prior to symptom resolution. The number (%) with each outcome and the adjusted relative risk (RR) and 95% Bonferroni confidence interval (CI) are shown. Global p-values testing for any interaction between treatment effect and baseline diarrhea severity are shown.

Figure 3.

Conditional effect of the probiotic treatment within severity subgroups on outcomes of diarrhea duration (days) and the maximal number of diarrheal episodes per 24-hour period. The mean (standard deviation) of each outcome along with the Adjusted Incidence Rate Ratio (IRR) and 95% Bonferroni confidence interval (CI) are shown. Global p-values testing for any interaction between treatment effect and baseline diarrhea severity are shown.