Antibiotic-induced gut microbiota dysbiosis in the PICU: mechanisms, clinical outcomes, and management strategies - a narrative review.

BACKGROUND

Antibiotic exposure is highly prevalent in the paediatric intensive care unit (PICU) and constitutes a major modifiable driver of gut microbiota dysbiosis. However, a systematic synthesis focusing specifically on the PICU population has been lacking. This narrative review addresses how antibiotic exposure drives gut dysbiosis in the PICU, its clinical consequences with emphasis on immunological pathways, and its management.

METHODS

This review was informed by a structured search of PubMed, Web of Science, Cochrane Library, and Chinese databases up to June 2026, including studies on antibiotic exposure, microbiota alterations, clinical outcomes, and management in critically ill children.

RESULTS

Antibiotic use in PICU children ranges from 58% to 94%, with broad-spectrum and combination therapy being common. Anti-anaerobic antibiotics-particularly piperacillin-tazobactam, meropenem, and clindamycin-cause the most pronounced disruption, as quantified in adult ICU cohorts. In PICU children, clinical consequences include a higher incidence of Clostridioides difficile infection and an increased risk of ventilator-associated pneumonia following carbapenem exposure. Antibiotic stewardship, encompassing de-escalation and avoidance of unnecessary anaerobic coverage, is the first-line microbiota protection strategy. Probiotics reduce ventilator-associated pneumonia and shorten PICU stay, but should not be used routinely in high-risk children. High-fibre enteral nutrition has shown feasibility, whereas postbiotics and faecal microbiota transplantation lack PICU-specific trial data.

CONCLUSIONS

Antibiotics are among the most significant modifiable drivers of gut dysbiosis in the PICU, with dose-dependent, class-specific effects. Antibiotic stewardship should be prioritised before any microbiota-directed intervention. Live probiotics require caution in high-risk populations, while non-live interventions are promising but need larger trials. Future research should employ longitudinal, multicentre studies with standardised reporting to elucidate host-microbe interactions in critically ill children.

Copyright © 2026 Yang, Wang and Peng.