WIP 49 - Stool Sphingomyelin Analysis in Preterm Very Low Birth Weight Infants to Predict Neonatal Outcomes and Cholestasis Development

Ari Esrig, University of Virginia School of Medicine, charlottesville, VA, United States; Arpitha Shenoy, University of Virginia School of Medicine, Fairfax, VA, United States; Glynis Kolling, University of Virginia School of Medicine, Charlottesville, VA, United States; Jason Papin, University of Virginia School of Medicine, Charlottesville, VA, United States; Andrea S. Marrs, UVA, Charlottesville, VA, United States; Brynne Sullivan, University of Virginia, Charlottesville, VA, United States

Background: Sphingomyelins play a key role in nutrient metabolism in the gastrointestinal (GI) tract, facilitating inflammatory signaling and maintaining tight junctions. Fecal sphingomyelin levels have been studied as potential non-invasive biomarkers of GI health and disease, which is especially relevant for premature infants who are at high risk of developing conditions affecting intestinal health and microbiota. While fecal sphingomyelin quantification has been investigated in this population with parenteral nutrition-associated cholestasis (PNAC) and necrotizing enterocolitis (NEC), its longitudinal associations with other prematurity-related comorbidities remain unexplored. Identifying associations between a sphingomyelin stool biomarker and these comorbidities could reduce reliance on blood tests and support preventive interventions to improve outcomes.

Objective: To examine changes in fecal sphingomyelins among very low birth weight (VLBW, < 1500g) premature infants over their first 8 weeks and evaluate associations with clinical comorbidities and outcomes.

Design/Methods: IRB approval was obtained from the University of Virginia Health Sciences Research Review Board. With parental consent, this prospective cohort study enrolled very low birth weight infants admitted to a Level IV NICU. Biweekly stool samples were collected from diapers in 30 mL tubes and stored at -80°C. Demographic and clinical data from electronic health records were recorded in a REDCap database. Stool samples will undergo mass spectrometry and 16S rRNA sequencing for targeted sphingomyelin quantification and microbiota profiling. Conjugated bilirubin and other laboratory tests were obtained at the clinical team’s discretion. Univariate and multivariable regression models will evaluate associations between sphingomyelin levels, feeding variables (breastmilk and parenteral nutrition use), comorbidities (sepsis, NEC), and outcomes (mortality, length of stay). Stool samples from 48 participants have been collected and are scheduled for metabolomic testing, anticipated to be completed by March 2025.