Longitudinal fecal amino acid profiles in extremely preterm infants: early-life determinants and associations with late-onset sepsis
摘要
Fecal microbiome and metabolome alterations precede late-onset sepsis (LOS) in preterm infants, but clinical determinants of fecal amino acid (AA) composition and association with LOS development remain poorly understood.
ObjectivesThis study assessed the early-life determinants of fecal AA composition and AA profiles prior to clinical LOS onset to improve pathophysiological understanding.
MethodsInfants (< 28 weeks’ gestation) with non-staphylococcal LOS were included in a discovery (n = 12) and validation cohort (n = 8), each matched 1:1 to non-LOS controls (n = 20), based on gestational and postnatal age. Using targeted liquid chromatography-tandem mass-spectrometry, this study identified early-life determinants of fecal AAs, analyzing fecal samples collected at week 1 to 4 in controls. Additionally, AA profiles prior to LOS onset were assessed to improve pathophysiological understanding, analyzing samples at t0 and t-3 days from LOS-affected infants and controls.
ResultsThe most common LOS pathogens were Escherichia coli (n = 8), Serratia species (n = 4), and Streptococcus agalactiae (n = 3). Postnatal age, full-enteral feeding status, probiotic administration, and delivery mode significantly influenced fecal AAs. In the discovery cohort, threonine and glutamine were significantly decreased in affected infants (median [Q1-Q3], threonine: 43.8 [29.3–66.3] vs. 64.5 [48.1–107.3] μmol/L, p = 0.02; glutamine: 19.0 [9.0–38.3] vs. 34.1 [27.0–69.8], p = 0.01), while the validation cohort solely showed a non-significant decrease of threonine. When combining both cohorts, threonine achieved an area under the curve of 0.65 (95%-CI: 0.53–0.76).
ConclusionsThese findings suggest several clinical characteristics influence fecal AAs, and altered threonine metabolism may contribute to LOS pathophysiology, warranting evaluation of related metabolites and multi-marker approaches in future studies.
Graphical Abstract