Shared metabolomic signatures for prognostic precision across brain injuries

Metabolomic alterations have been linked to a range of neurological conditions. Investigating temporal changes in serum metabolomic profiles, regardless of the type of brain injury may reveal prognostic indicators.

We hypothesize that specific metabolomic signatures, conserved across different acute brain injuries, can serve as robust predictors of patient outcomes.

In this longitudinal prospective observational study, serum samples were collected early (2 ± 1 day) and late (6 ± 2 days) post-injury from a total of 73 patients with ischemic stroke (n = 30), aneurysmal subarachnoid hemorrhage (n = 30), and traumatic brain injury (n = 13). Outcomes were categorized as favorable (modified Rankin Scores (mRS) 0-3) and unfavorable (mRS 4-6) three months post-injury. Metabolomic profiling (Orbitrap mass spectrometry) of 462 metabolites, analyzed using statistical and machine learning methods, identified significant outcome differences (p < 0.05, FDR-corrected).

Early-stage samples indicated good prognostic power with a combination of uridine, tryptophan, and lactic acid (AUC 88.8 %, OR 5.29, p < 0.0001). Late-stage samples showed high discriminatory accuracy with a combination of prostaglandin J2, gamma-linolenic acid, N-acetyl-L-alanine, uridine, N-alpha-acetyl-L-asparagine, 3-hydroxy-3-methylglutarate, propionate, and creatinine (AUC 94.4 %, OR 14.5, p < 0.0001). Pathway analyses revealed significant associations with glycolysis/gluconeogenesis, pyrimidine metabolism, and tryptophan metabolism at early stages, and fatty acid biosynthesis, pyruvate metabolism, phenylalanine metabolism, and tryptophan metabolism at later stages.

These findings underscore the dynamic nature of metabolomic profiles in acute brain injuries and highlight common metabolites as significant prognostic markers across brain injury types.