Metabolomic Prediction of Pneumonia Severity

点击翻译按钮获取中文摘要

PROJECT SUMMARY Community-acquired pneumonia (CAP) is one of the most prevalent infections in children resulting in 1.8 million healthcare visits annually. Validated tools to predict severe clinical outcomes in children with CAP do not exist. The addition of biomarkers to clinical prediction rules may improve severity prediction, however conventional biomarkers (e.g. procalcitonin) have limited ability to predict severity in children. We propose identifying novel biomarkers for pediatric CAP using metabolomics, the study of small molecules. We hypothesize that metabolites may be better predictors of illness severity as they directly represent the complex physiological interaction between the environment (e.g. infection) and the host in a single sample. Preliminary data of urine samples assayed by Nuclear Magnetic Resonance (NMR) spectrometry, a more specific analytical platform for metabolomics, suggests amino acids, carnitine and bile acid molecules are important predictors of CAP severity. As a complimentary approach we will further investigate these classes of metabolites and specific lipid molecules (i.e. oxylipins) in blood samples using liquid chromatography mass spectrometry (LC-MS), a more sensitive analytical platform. Information from urine and blood specimens will generate a comprehensive set of prognostic metabolomic biomarkers as the strengths of both analytical platforms are being leveraged for this proposal. The targeted approach we propose in addition to the use of two separate pediatric CAP cohorts, will result in a clinical and metabolomic biomarker prediction rule to predict severity in children presenting to the emergency department with CAP.