Metabolomics in nutrition research: biomarkers predicting mortality in children with severe acute malnutrition.

BACKGROUND: Millions of the world's children suffer from malnutrition, which predisposes to death from diarrhea and a variety of infectious diseases. Mortality rates among infants and toddlers remain staggeringly high, in part because the pathogenesis of acute malnutrition and its complications remains poorly understood. OBJECTIVE: We used metabolomic analysis to characterize the metabolic status of Ugandan children with severe acute malnutrition (SAM) and to delineate changes in hormones, metabolites, growth factors, and cytokines during nutritional therapy. We hypothesized that hormonal and metabolic factors measured at presentation would associate with, or predict, subsequent mortality during treatment. METHODS: This was a prospective cohort study of 75 severely malnourished children 6 months to 5 years of age treated as inpatients with F-75 and F-100 and supplemental micronutrients; after discharge, they received ready-to-use therapeutic food (RUTF). This increased the mean weight-for-height z-score (WHZ) from -4.27 to -1.75 SD. Blood samples were obtained at presentation, after 2 weeks of inpatient therapy, and after 4 to 10 weeks of RUTF. Plasma samples were analyzed by tandem mass spectrometry and microassays. RESULTS: At presentation there were high levels of nonesterified fatty acids (NEFA), ketones, and even-chain acylcarnitines, indicating active lipolysis and fatty acid oxidation. In contrast, albumin, amino acids, and C3 carnitine, a by-product of branched-chain amino acids, were low. Levels of insulin, insulin-like growth factor 1 (IGF-1), adiponectin, and leptin were low, while levels of ghrelin, growth hormone, cortisol, interleukin 6 (IL-6), peptide YY (PYY), and glucagon-like peptide 1 (GLP-1) were high. The metabolic and hormonal changes were reversed by formula feeding and RUTF. Biomarkers associated with mortality included HIV, WHZ, and mid-upper-arm circumference (MUAC); the biochemical factor associated most strongly with mortality was low leptin, a marker of adipose reserve and modulator of immune function. CONCLUSIONS: Low leptin predicts mortality in edematous and nonedematous-patients with SAM. Leptin assays might be used to identify malnourished children at highest risk for death.

Duke Scholars

Author Michael Scott Freemark Pediatrics, Endocrinology