Untargeted Urinary Metabolomic Profile of Isolated Chest Trauma Patients

Untargeted Urinary Metabolomic Profile of Isolated Chest Trauma Patients. ABSTRACT: Aim To explore the untargeted urinary metabolomic biomarkers in isolated chest injury patients. Methods It was a prospective study for 1 year period including the patients of isolated chest trauma as per inclusion criteria. Day 0, 3, and 7 urine samples from patients and day 0 sample from age matched healthy relatives were collected. Samples were prepared and gas chromatography mass spectrometry analysis was done using Agilent Technologies. Multivariate analyses were done using the 'Biomarker Analysis' module of the MetaboAnalyst 6.0 to identify potential biomarkers associated with chest injuries.  To identify the promising biomarkers with high sensitivity and specificity, exploratory multivariate (receiver operating characteristic) ROC curve analysis was used. Results A total of 31 patients and 31 healthy controls were enrolled. A total of 101 metabolites were annotated. Based on ROC analyses, the top 15 potential markers (azelaic acid, glycerol, anthranilic acid, 2-palmitoylglycerol, 2-hydroxyhippuric acid, 4-hydroxybenzeneacetic acid, arachidic acid, 3-hydroxyhippuric acid, adenine, sucrose, butanedioic acid, 4-hydroxyphenyllactic acid, butanoic acid, 4-hydroxybenzoic acid and 2-hydroxybenzeneacetic acid) were identified. Selected frequencies of >0.6% led to the selection of 9 biomarkers. It was found 66.5% predictive accuracy of the most accurate 50-feature panel of model with the AUC of 0.708 with 95% confidence interval. Conclusion Nine metabolites (azelaic acid, glycerol, anthranilic acid, 2-palmitoylglycerol, 2-hydroxyhippuric acid, 4-hydroxybenzeneacetic acid, arachidic acid, 3-hydroxyhippuric acid and adenine) based on their selected frequencies of more than 0.6% can be considered as biomarkers in isolated chest trauma patients.