A putative urinary biosignature for diagnosis and follow-up of tuberculous meningitis in children: outcome of a metabolomics study disclosing host–pathogen responses

• DOI: 10.1007/s11306-016-1053-2
• Publication Date: 2016-06-10T07:15:14Z
• Authors: Shayne Mason; A. Marceline Tutu...; Regan Solomons; Ron A. Wevers; Mari van Reenen; Carolus J. Reinecke
• Keywords: Tuberculous Meningitis

Tuberculous meningitis (TBM) is a severe manifestation of tuberculosis, presenting with high morbidity and mortality in children. Existing diagnostic methods for TBM are invasive and time-consuming and the need for highly sensitive and selective diagnosis remains high on the TBM agenda. Our aim was to exploit metabolomics as an approach to identify metabolites as potential diagnostic predictors for children with TBM through a non-invasive means. Urine samples selected for this study were from three paediatric groups: patients with confirmed TBM (n = 12), patients clinically suspected with TBM but later confirmed to be negative (n = 19) and age-matched controls (n = 29). Metabolomics data were generated through gas chromatography–mass spectrometry analysis and important metabolites were identified according to standard statistical procedures used for metabolomics data. A global metabolite profile that characterized TBM was developed from the data, reflecting the host and microbial responses. Nine different logistic regression models were fitted to selected metabolites for the best combination as predictors for TBM. Four metabolites—methylcitric, 2-ketoglutaric, quinolinic and 4-hydroxyhippuric acids—showed excellent diagnostic ability and provided prognostic insight into our TBM patients. This study is the first to illustrate holistically the metabolic complexity of TBM and provided proof-of-concept that a biosignature of urinary metabolites can be defined for non-invasive diagnosis and prognosis of paediatric TBM patients. The biosignature should be developed and validated through future prospective studies to generate a medical algorithm for diagnosis in the initial stages of the disease and for monitoring of treatment strategies.