Metabolomic biomarkers of concussion

Objective To determine if 1 H NMR spectroscopy of urine and a quantitative metabolomics approach can aid diagnostic and prognostic outcomes in patients who have suffered acute sport concussion. Design Pre-concussion urine samples were collected from WinSport athletes at the beginning of the sport season. Any athlete who experienced a head injury reported to a brain injury clinic to be tested within 72 hours of head injury. This testing included collection of a post-concussion urine sample for metabolomic analysis. Setting Brain injury clinic and metabolomics laboratory. Participants 167 athletes entered the study with a total of 39 concussions occurring. Inclusion criteria: at least 18 years old, diagnosis of concussion, and concussion sustained within 72 hours. Exclusion criteria: multiple injuries coinciding with concussion and a past history of neurological pathology. Outcome measures Metabolite concentrations were calculated in pre- and post-concussion urine samples and compared using a paired t-test and partial least squares discriminant analysis (PLS-DA) to determine which metabolites led to significant variation between samples. Main results The PLS-DA model showed significant variation between baseline and post-injury sample groups and passed all permutation and cross-validation testing. A paired t-test, with cross-validation, isolated five significant metabolites of interest in the concussion group (p Conclusions Using NMR spectroscopy we quantified urinary metabolites in patients before and after concussion. This revealed several metabolites that were shown to be significantly altered following injury, opening up future inquiry into both the pathways and processes involved in concussion and new avenues for personalized diagnosis and treatment. Competing interests None.