Timing of Neonatal Discharge and Unplanned Readmission to PICUs Among Infants Born Preterm.

Importance: Children born very preterm (<32 weeks) are at risk of ongoing morbidity and admission to pediatric intensive care units (PICUs) in childhood. However, the influence of the timing of neonatal discharge on unplanned PICU admission has not been established. Objective: To examine whether the timing of neonatal discharge (postmenstrual age and season) is associated with subsequent unplanned PICU admission. Design, Setting, and Participants: This retrospective cohort study used linked national data from the National Neonatal Research Database and Paediatric Intensive Care Audit Network (PICANet) for children born from January 2013 to December 2018 at 22 to 31 weeks' gestational age who were admitted to a neonatal unit in England and Wales and were discharged home at 34 weeks' postmenstrual age or later. All National Health Service (NHS) neonatal units and PICUs in England and Wales were included. Children were followed up until 2 years of chronological age. Data analysis was conducted from October 2023 to August 2024. Exposures: Timing of discharge. Main Outcomes and Measures: The primary outcome was unplanned PICU admission between neonatal discharge and chronological age 2 years to any PICU within England and Wales. Survival analysis using a flexible parametric model was conducted with season of discharge (time-dependent factor), gestation, sex, birth weight less than the 10th centile, bronchopulmonary dysplasia, necrotizing enterocolitis, brain injury, and earlier neonatal discharge (lower quartile of postmenstrual age at discharge for gestation) as variables. Results: Of 39 938 children discharged home (median [IQR] gestational age, 29 [27-31] weeks; 21 602 [54.1%] male), 1878 (4.7%) had unplanned PICU admission. More than half of admissions occurred within 50 days of neonatal discharge (1080 [57.5%]). Compared with summer, the risk of unplanned PICU admission following neonatal discharge was 2.58 times higher in winter and 2.35 times higher in autumn (winter: adjusted hazard ratio [aHR], 2.58; 95% CI, 1.68-3.95; autumn: aHR, 2.35; 95% CI, 1.84-2.99). Among children born at 28 to 31 weeks' gestational age, earlier neonatal discharge was associated with increased risk (aHR, 1.30; 95% CI, 1.13-1.49), but this was not true for children born younger than 28 weeks' gestational age. Conclusions and Relevance: In this retrospective cohort study of preterm children, autumn and winter discharge were associated with the highest risk of unplanned PICU admission following neonatal discharge. For children born at 28 to 31 weeks' gestational age, discharge at lower postmenstrual age was also associated with increased risk. Further work is required to understand whether delaying neonatal discharge for some children born at 28 to 31 weeks' gestational age is beneficial and to consider the wider costs and implications of prolonging neonatal care.