AIM To explore the cerebrospinal fluid (CSF) metabolite features in acute neuroinflammatory diseases and identify potential biomarkers to diagnose and monitor neuroinflammation. METHOD A cohort of 14 patients (five females, nine males; mean [median] age 7y 9mo [9y], range 6mo-13y) with acute encephalitis (acute disseminated encephalomyelitis n=6, unknown suspected viral encephalitis n=3, enteroviral encephalitis n=2, seronegative autoimmune encephalitis n=2, herpes simplex encephalitis n=1) and age-matched non-inflammatory neurological disease controls (n=14) were investigated using an untargeted metabolomics approach. CSF metabolites were analyzed with liquid chromatography coupled to high resolution mass spectrometry, followed by subsequent multivariate and univariate statistical methods. RESULTS A total of 35 metabolites could be discriminated statistically between the groups using supervised orthogonal partial least squares discriminant analysis and analysis of variance. The tryptophan-kynurenine pathway contributed nine key metabolites. There was a statistical increase of kynurenine, quinolinic acid, and anthranilic acid in patients with encephalitis, whereas tryptophan, 3-hydroxyanthrnailic acid, and kynurenic acid were decreased. The nitric oxide pathway contributed four metabolites, with elevated asymmetric dimethylarginine and argininosuccinic acid, and decreased arginine and citrulline in patients with encephalitis. An increase in the CSF kynurenine/tryptophan ratio (p<0.001), anthranilic acid/3-hydroxyanthranilic acid ratio (p<0.001), asymmetric dimethylarginine/arginine ratio (p<0.001), and neopterin (p<0.001) strongly predicted neuroinflammation. INTERPRETATION The combination of alterations in the tryptophan-kynurenine pathway, nitric oxide pathway, and neopterin represent a useful potential panel for neuroinflammation and holds potential for clinical translation practice.

No abstract

No abstract

No abstract