Prediction of mortality in severe acute malnutrition in hospitalized children by faecal volatile organic compound analysis: proof of concept

Abstract

Children with severe acute malnutrition (SAM) display immature, altered gut microbiota and have a high mortality risk. Faecal volatile organic compounds (VOCs) reflect the microbiota composition and may provide insight into metabolic dysfunction that occurs in SAM. Here we determine whether analysis of faecal VOCs could identify children with SAM with increased risk of mortality. VOC profiles from children who died within six days following admission were compared to those who were discharged alive using machine learning algorithms. VOC profiles of children who died could be separated from those who were discharged with fair accuracy (AUC) = 0.71; 95% CI 0.59-0.87; P = 0.004). We present the first study showing differences in faecal VOC profiles between children with SAM who survived and those who died. VOC analysis holds potential to help discover metabolic pathways within the intestinal microbiome with causal association with mortality and target treatments in children with SAM.Trial Registration: The F75 study is registered at clinicaltrials.gov/ct2/show/NCT02246296.

Figure 1

Feature map illustrating locations on the FAIMS output from children who died within 6 days of admission. (A) Positive feature locations (B) Negative feature locations.

Figure 2

(A) Mortality v survival. Support vector machine (100 features). (B) Early mortality v survival. Random forest (100 features). (C) Late mortality v survival. Sparse logistic regression (50 features). (D) Early mortality v late mortality. Support vector machine (50 features).