(899) Metabolomic Profile of High-Risk Phenotype of Sepsis-Associated MODS: Persistent Hypoxemia and Shock

Introduction: Pediatric sepsis-associated MODS (SA-MODS) is a heterogenous syndrome with high mortality and no targeted therapies to date. We have previously identified and characterized the “persistent hypoxemia, encephalopathy, and shock” (PHES) phenotype in children with SA-MODS which is independently associated with poor outcomes and heterogeneity of treatment effect to common adjuvant therapies. Metabolomics may help identify, predict and understand the underlying pathophysiology and lead to the development tailored therapies for patients with PHES and other high-risk organ dysfunction patterns, like persistent MODS (PMODS). Significant changes in metabolites have been detected in critical illnesses such as ARDS and septic shock, but studies in pediatrics are limited. We believe there are unique metabolomic profiles to PHES and PMODS in children.

Methods: Plasma samples were collected within 36 hours of admission from patients meeting SA-MODS criteria in our PICU and were categorized as PHES phenotype and/or PMODS (MODS persisting by day 7 of admission) or controls. Using high-resolution accurate-mass mass spectrometry, 365 lipids and 43 small molecules were quantified. MetaboAnalyst 5.0 software was utilized to perform univariate analysis and quantitative metabolic pathway enrichment analysis (QMPEA) to identify dysregulated individual metabolite levels and pathways.

Results: 50 patients with SA-MODS enrolled, of which 17 had PHES and 14 had PMODS. In the PHES patients, the levels for threonine, glutamine, isoleucine, methionine sulfoxide, 33 lipids, and one monosaccharide were different than controls (p < 0.05), with sphingolipid metabolism (p=0.02), homocysteine degradation (p=0.02) and phosphatidylethanolamine biosynthesis (p=0.02) pathways different than controls by QMPEA. In PMODS patients, 25 lipids and the metabolite asymmetric dimethylarginine had plasma levels different than those of controls (p < 0.05) with methylhistidine metabolism different than controls by QMPEA (p=0.02).

Conclusions: In this pilot study we found unique plasma metabolic derangements in patients with SA-MODS with high-risk phenotypes that were identifiable early in the course. Some of these derangements are similar to ones previously described in other critical care cohorts. Further study and validation is warranted.