Development and internal validation of a nomogram for early prediction of hospital-acquired ESKAPE colonization or infection in very preterm infants using indicators available within 24 hours.

BACKGROUND: ESKAPE pathogens are a major cause of hospital-acquired colonization or infection among very preterm infants in neonatal intensive care units (NICUs). Early identification of high-risk infants can help prioritize infection-control interventions and guide targeted preventive care. METHODS: A single-center retrospective cohort study was conducted among 465 very preterm infants (gestational age, GA ≤ 32 weeks) admitted to the NICU of a tertiary hospital between January 2015 and June 2025. Infants were randomly divided into training (n = 325) and internal validation (n = 140) cohorts at a 7:3 ratio. Predictors available within 24 h after birth were screened using least absolute shrinkage and selection operator (LASSO) regression. A multivariable logistic regression model was constructed and presented as a nomogram. RESULTS: Overall, 77 infants (16.56%) developed hospital-acquired ESKAPE colonization or infection, of which 67.5% (52/77) were first identified within 14 days after birth. Among 109 non-duplicate ESKAPE isolates, the predominant pathogens were Acinetobacter baumannii (38/109, 34.86%) and Klebsiella pneumoniae (31/109, 28.44%); respiratory specimens were the primary source (90/109, 82.57%). Four predictors were retained in the final model: GA at birth, initial invasive mechanical ventilation, vasoactive exposure within the first 24 h, and 5-minute Apgar score. The nomogram showed an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.718-0.853) in the training cohort and 0.770 (95% CI: 0.673-0.866) in the internal validation cohort, indicating moderate discrimination. Calibration curves and decision curve analysis demonstrated good agreement between predicted probabilities and observed risks, with net benefit across a wide range of clinically relevant threshold probabilities. An online prediction tool was also developed (https://newborn.shinyapps.io/dynnomapp/). CONCLUSION: We developed and internally validated a nomogram for early prediction of hospital-acquired ESKAPE colonization or infection in very preterm infants using routine clinical indicators obtained within 24 h after birth. The model can support early risk stratification and infection control prioritization in the NICU. External validation and prospective implementation studies are required before routine clinical adoption.