Adjunctive probiotic use during antibiotic exposure in very preterm infants: a narrative and translational review

Abstract

Infants born before 32 weeks of gestation are highly susceptible to late-onset sepsis because of immune immaturity, impaired intestinal barrier function, and frequent broad-spectrum antibiotic exposure. Whether probiotics should be continued or initiated during antibiotic therapy remains unresolved and is often conflated with routine prophylaxis in stable infants. This SANRA-guided structured narrative and translational review synthesized clinical, microbiome, resistome, mechanistic, and safety evidence identified in PubMed/MEDLINE, Scopus, and Web of Science through July 1, 2026, across three contexts: stable antibiotic overlap, suspected sepsis, and confirmed or unstable sepsis. Pooled prophylactic analyses suggest a small, heterogeneous reduction in late-onset infection (risk ratio 0.83, 95% CI 0.72–0.95). However, the 2023 Cochrane review judged that probiotics probably have little or no effect on late-onset invasive infection (risk ratio 0.89, 95% CI 0.82–0.97), the largest randomized trials were neutral, and no benefit has been demonstrated in extremely preterm or extremely low birth weight infants (risk ratio 0.93, 95% CI 0.78–1.09). Direct evidence during systemic antibiotic exposure is limited to microbiome endpoints and does not establish benefit in suspected or confirmed sepsis; no randomized trial has evaluated probiotics as adjunctive treatment for active sepsis. Metagenomic studies suggest that probiotics may attenuate antibiotic-associated dysbiosis and modify the intestinal resistome, but these surrogate effects have not been shown to reduce multidrug-resistant infection. Probiotic-associated bloodstream infection was infrequently reported (8 cases among 20,323 exposed infants), although ascertainment was inconsistent and reported infections may be severe. Current evidence does not support probiotics as treatment for neonatal sepsis or routine initiation during suspected or confirmed sepsis. Prospective strain-specific trials should evaluate these three contexts separately.