Abstract
Metabolites Measured Over the First 7 Days of Life Predict Severe Morbidity or Mortality in Infants Born at <32 Weeks
Current developments in nutrition, Vol.10, 108413
07/2026
DOI: 10.1016/j.cdnut.2026.108413
Abstract
Objectives: The very preterm neonate (VPT; < 32 weeks gestational age) is at risk for certain medical morbidities due to their unique physiology, yet our ability to predict which infants will experience severe complications is limited. The objective of this study is to develop and assess a prediction model for VPT infants using metabolites routinely measured in the national newborn screen over the first week of life to predict severe morbidity or mortality at discharge.
Methods: In this prospective longitudinal study, we enrolled 339 VPT infants at University of California San Francisco Benioff Children’s Hospital and University of Iowa Hospitals and Clinics and analyzed cord blood and blood spot specimens at day of life 0, 1.5, 3, 5 and 7 using tandem mass spectrometry per standard newborn screening methodology to assess amino acids and acylcarnitines, key intermediates of metabolism, gluconeogenesis and beta-oxidation. We then performed timepoint-specific penalized logistic regression models using the least absolute shrinkage and selection operator (LASSO) to identify significant metabolite predictors of the composite morbidity (intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, retinopathy of prematurity, and bronchopulmonary dysplasia) or mortality outcome. We used mixed-effects logistic regression models with a random effect for individuals to account for repeated measurements within infants.
Results: The model of metabolite profiles measured during the first week of life had a strong predictive performance (AUC > 0.84 at all timepoints) of composite morbidity or mortality. Metabolites whose longitudinal trajectories were associated with the outcome over the first week of life (i.e., dicarboxylic C4 acylcarnitine, C4 acylcarnitine, citrulline and 17-hydroxyprogesterone) remained significant (p < 0.001) after adjustment for time at measurement, gestational age, birth weight, and sample type (cord or venous).
Conclusions: Newborn screen metabolites measured over the first week of life predict severe morbidity or mortality in VPT infants. Metabolic profile models offer a precision approach to the diagnostic and clinical management of this vulnerable population and may lead to the development of new therapeutic agents to improve health outcomes.
Details
- Title: Subtitle
- Metabolites Measured Over the First 7 Days of Life Predict Severe Morbidity or Mortality in Infants Born at <32 Weeks
- Creators
- Emily C Hanselman - Indiana UniversityScott P Oltman - University of California, San FranciscoJames A Merchant - University of IowaErik S Parker - Indiana UniversityAllison M Momany - University of IowaJohn M Dagle - University of IowaLouie M Swander - University of California, San FranciscoAngela G Campbell - Indiana UniversityElizabeth E Rogers - University of California, San FranciscoLaura L Jeliffe-Pawlowski - New York University, United StatesKelli K Ryckman - Indiana University
- Resource Type
- Abstract
- Publication Details
- Current developments in nutrition, Vol.10, 108413
- DOI
- 10.1016/j.cdnut.2026.108413
- ISSN
- 2475-2991
- eISSN
- 2475-2991
- Publisher
- Elsevier Inc
- Language
- English
- Date published
- 07/2026
- Academic Unit
- Stead Family Department of Pediatrics; Epidemiology; Iowa Neuroscience Institute; Neonatology; Internal Medicine
- Record Identifier
- 9985214914602771
Metrics
1 Record Views