TOPLINE

Newborns in European neonatal units were more likely to acquire bacteria carrying extended-spectrum beta-lactamase (ESBL) genes when more infants on the unit already carried them. Each 10-percentage-point increase in the proportion of colonised infants was linked to 73%-89% higher odds of acquisition.

METHODOLOGY

• Researchers conducted a cross-sectional study in 24 neonatal units across eight European countries to estimate the prevalence of antibiotic-resistant bacteria in hospitalised infants and to assess whether a higher prevalence of unit‑level ESBL colonisation was associated with a greater risk for infant acquisition.
• A total of 943 infants contributed to 1847 infant-survey observations over four repeated cross-sectional surveys with intervals of 4, 7, and 14 days; stool samples were collected and analysed using real-time quantitative polymerase chain reaction for the detection of antibiotic resistance genes.
• The exposure was ESBL colonisation pressure, defined as the proportion of infants on the unit who were colonised with ESBL-harbouring bacteria at the previous survey.
• The primary outcome was newly detected colonisation with ESBL-carrying bacteria — defined as the detection of blaCTX-M group 1 or group 9 genes in stool — in an infant's gut between surveys.

TAKEAWAY

• ESBL genes were the most commonly detected antibiotic resistance genes, found in 205 of 1448 stool samples (14.2%), followed by carbapenem resistance genes (7.0%) and vancomycin resistance genes (5.1%).
• The acquisition analysis included 829 samples from 533 infants, of whom 90 (16.9%) became colonised with ESBL-harbouring bacteria during follow-up.
• Higher colonisation pressure was associated with greater odds of an infant becoming colonised with ESBL-harbouring bacteria. Estimated mean odds ratios per 10-percentage-point increase were 1.89 (95% credible interval [CrI], 1.22-2.94) for the 4-day interval, 1.76 (95% CrI, 1.26-2.63) for the 7-day interval, and 1.73 (95% CrI, 1.15-2.52) for the 14-day interval.

IN PRACTICE

"Interventions that reduce colonization, acting both directly and indirectly, may be particularly important in units with high colonization pressure and during outbreaks, and should be considered alongside strategies focused on preventing invasive infection in the most vulnerable infants," the authors wrote.

SOURCE

The study was led by Aislinn Cook, MSc, University of London, London, England. It was published online on September 15, 2026, in JAMA Network Open.

LIMITATIONS

The study was observational and cross-sectional and had incomplete stool collection. Infants born at the study hospitals who were admitted between sampling rounds were assumed uncolonised on entry, a choice that may have inflated in‑hospital acquisition estimates. Residual confounding from unmeasured factors could not be excluded.

DISCLOSURES

The NeoIPC project received funding from the European Union's Horizon 2020 research and innovation programme. Several authors reported receiving grants or fees from a wide range of pharmaceutical companies — including Da Volterra, MSD Europe, Pfizer, and Shionogi — and research organisations.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.