Antibiotic-induced gut microbiota dysbiosis in the PICU: mechanisms, clinical outcomes, and management strategies — narrative review
Antibiotic exposure occurs in 58–94% of PICU children and is the primary modifiable driver of gut dysbiosis, but evidence for restorative interventions in this population remains insufficient.
| Population | Critically ill children admitted to the PICU with antibiotic exposure |
|---|---|
| Intervention | Antibiotics (broad-spectrum, anti-anaerobic, combination regimens); microbiota protection/restoration strategies (antimicrobial stewardship, probiotics, high-fibre enteral nutrition, postbiotics, faecal microbiota transplantation) |
| Comparator | No antibiotic exposure or alternative interventions; adult ICU cohort data used as indirect reference |
| Outcome | Antibiotic use prevalence in PICU; Clostridioides difficile infection; Ventilator-associated pneumonia (VAP); PICU length of stay; Gut microbiota diversity; Probiotic efficacy in paediatric PICU; Faecal microbiota transplantation and postbiotics in paediatric PICU |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| Antibiotic use prevalence in PICU | Range 58-94% across observational studies; in the pooled estimate, in the 95% CI | — | Low | — | |
| Clostridioides difficile infection | Increased risk reported with broad-spectrum antibiotic exposure in PICU children; in the pooled RR/OR or 95% CI provided | — | Very low | — | |
| Ventilator-associated pneumonia (VAP) | Increased risk with carbapenem exposure; probiotics associated with VAP reduction in some trials; in the pooled RR/OR or 95% CI for paediatric PICU population | — | Low | — | |
| PICU length of stay | Shortened with probiotics in some trials; in the pooled MD or 95% CI reported for paediatric PICU | — | Low | — | |
| Gut microbiota diversity | Reduced diversity reported with anti-anaerobic antibiotic exposure; quantitative data from adult ICU cohorts only; in the effect size with 95% CI for children | — | Very low | — | |
| Probiotic efficacy in paediatric PICU | VAP reduction and shorter PICU stay in some trials; in the consolidated RR/OR or 95% CI; safety concerns in high-risk children | — | Low | — | |
| Faecal microbiota transplantation and postbiotics in paediatric PICU | No PICU-specific RCT data available; evidence base absent | — | Very low | — |
Context
Broad-spectrum antibiotics are ubiquitous in the paediatric ICU and represent the leading modifiable cause of gut microbiota dysbiosis in critically ill children. Dysbiosis impairs immunological barriers, promotes opportunistic pathogen colonisation, and associates with adverse clinical outcomes. No prior systematic synthesis had focused exclusively on the paediatric ICU population.
What the study showed
Antibiotic use in the PICU ranges from 58% to 94% of admitted children; anti-anaerobic agents (piperacillin-tazobactam, meropenem, clindamycin) produce the most pronounced microbial disruption, quantified in adult ICU cohorts. Carbapenem-exposed children show higher risk of VAP and C. difficile infection. Probiotics reduced VAP and shortened PICU stay in some trials, but the review does not provide consolidated effect sizes with 95% CI for the specific paediatric population. High-fibre enteral nutrition demonstrated feasibility but not efficacy on hard clinical outcomes.
How it was done
Structured narrative review searching PubMed, Web of Science, Cochrane Library, and Chinese databases up to June 2026. Included studies on antibiotic exposure, microbiota alterations, clinical outcomes, and management in critically ill children. No meta-analysis, quantitative synthesis, formal risk-of-bias tool (e.g., AMSTAR-2), or declared PRISMA criteria were applied.
Effect magnitude
No consolidated effect size with 95% CI is reported for the paediatric PICU population. Quantitative dysbiosis data derive from adult cohorts and are not directly extrapolable to children. Antibiotic use prevalence (58–94%) is the only consistent numerical finding.
Risk of bias
Narrative design without formal quality assessment tool (AMSTAR-2 not applied); high risk of literature selection bias. Magnitude-of-dysbiosis data are derived from adult cohorts with unvalidated extrapolation to children. Heterogeneity of primary studies is not quantified; RCT-level data for most evaluated interventions in the paediatric PICU are absent.
What this study does NOT prove
This narrative review does not prove causality between antibiotic-induced dysbiosis and specific adverse outcomes in PICU children. It does not establish efficacy or safety of any microbiota-restorative intervention in this population.
In clinical practice
Antimicrobial stewardship — including early de-escalation and avoidance of unnecessary anaerobic coverage — is the priority microbiota-protection strategy in the PICU. Probiotics may be considered in intermediate-risk children to reduce VAP but should be avoided in severe immunocompromise and premature neonates. Early high-fibre enteral nutrition is feasible; faecal microbiota transplantation and postbiotics lack paediatric PICU trial data.
Limitations
Narrative design without formal quality assessment tool (AMSTAR-2 not applied); high risk of literature selection bias. Magnitude-of-dysbiosis data are derived from adult cohorts with unvalidated extrapolation to children. Heterogeneity of primary studies is not quantified; RCT-level data for most evaluated interventions in the paediatric PICU are absent.
What is still missing
Randomised controlled trials in critically ill children are needed to evaluate probiotics, postbiotics, and faecal microbiota transplantation with predefined primary clinical outcomes. Longitudinal gut microbiota sequencing studies exclusively in the paediatric PICU are required to quantify dysbiosis magnitude in this population.
Technical appendix
Version history
- 1.0 · 2026-10-04 — Auto-generated under Evidence Standard v1.0