Updated evidence on efficacy and safety of ondansetron compared with placebo for acute gastroenteritis-associated vomiting in children: a systematic review and meta-analysis of randomized controlled trials

Abstract

Background Acute gastroenteritis (AGE) is among the most common causes of pediatric emergency care, and ondansetron is now widely used to control the vomiting that drives failure of oral rehydration therapy (ORT). The most recent comprehensive meta-analysis on this topic was published in 2020 and could not incorporate the subsequently published multidose post-discharge trial, did not evaluate the volume of ORT tolerated, and did not formally explore sources of heterogeneity through meta-regression; this update was undertaken to address these gaps. Methods The systematic review and meta-analysis was conducted and reported in accordance with the PRISMA 2020 statement, and its conduct was additionally guided by the AMSTAR 2 critical appraisal tool. A literature search was done in PubMed, CENTRAL, ScienceDirect, Google Scholar, and ClinicalTrials.gov until June 2026 for randomized controlled trials involving the use of ondansetron versus placebo in children with AGE-associated vomiting. Bias was assessed using RoB 2, while evidence certainty was assessed using GRADE. Random-effect models computed pooled risk ratios (RR) and mean differences (MD). Results Sixteen RCTs comprising 3,415 randomized/allocated children from 12 countries met the inclusion criteria. Ondansetron significantly reduced ongoing vomiting (RR 0.48), the number of vomiting episodes (MD − 0.73), failure of oral rehydration therapy (RR 0.39), and use of intravenous fluids (RR 0.57), and significantly increased oral rehydration intake (MD 47.87). No significant difference was found for hospitalization, repeat health-care visits, ongoing diarrhea, diarrheal episodes, or adverse events. The hospitalization estimate reached statistical significance only when a single large trial was omitted and is therefore fragile. On subgroup analysis by follow-up duration, the benefit for ongoing vomiting was significant through 24 h but attenuated at 48 h and beyond. Egger’s test indicated statistically significant funnel-plot asymmetry for both outcomes in which publication bias could be assessed, ongoing vomiting (p = 0.03070) and the number of vomiting episodes (p = 0.04627), whereas Begg and Mazumdar’s test was significant for neither; certainty of evidence was accordingly rated low for ongoing vomiting and very low for the number of vomiting episodes. Exploratory meta-regression, introduced during revision and not specified in the registered protocol, suggested the baseline proportion of male participants, baseline vomiting-episode frequency, and baseline diarrheal-episode frequency, but not mean age, as study-level moderators of the effect on ongoing vomiting. Conclusion Ondansetron significantly reduces vomiting morbidity in children with AGE, with no statistically significant difference detected versus placebo in adverse events or diarrhea, although the safety comparison rests on few estimable comparisons and very few events and therefore cannot exclude uncommon or delayed harms. These benefits are tempered by substantial statistical heterogeneity and by evidence of funnel-plot asymmetry for both vomiting outcomes, and the pooled magnitude of benefit on vomiting may therefore be overstated. The attenuation of benefit beyond 24 h provides a plausible rationale, though not direct evidence from this synthesis, for the extended multidose post-discharge regimen evaluated in the largest included trial. Further randomized trials directly comparing single-dose and multidose regimens, particularly in high-burden, low-resource settings, along with individual patient-data analyses, are needed. Systematic review registration: PROSPERO CRD420261447906.