Shared metabolomic signatures for prognostic precision across brain injuries

Area of Science:

• Neuroscience
• Metabolomics
• Biochemistry

Background:

• Metabolomic alterations are associated with neurological conditions.
• Investigating temporal changes in serum metabolomics may reveal prognostic indicators for brain injuries.

Purpose of the Study:

• To test the hypothesis that conserved metabolomic signatures across different acute brain injuries can predict patient outcomes.
• To identify specific metabolites and metabolic pathways associated with patient prognosis.

Main Methods:

• Longitudinal prospective observational study of 73 patients with ischemic stroke, subarachnoid hemorrhage, or traumatic brain injury.
• Serum samples collected early (2±1 days) and late (6±2 days) post-injury.
• Metabolomic profiling of 462 metabolites using Orbitrap mass spectrometry, analyzed with statistical and machine learning methods.

Main Results:

• Early-stage metabolomic profiles (uridine, tryptophan, lactic acid) showed good prognostic power (AUC 88.8%).
• Late-stage profiles (prostaglandin J2, gamma-linolenic acid, and others) demonstrated high discriminatory accuracy (AUC 94.4%).
• Pathway analyses revealed significant associations with glycolysis, pyrimidine metabolism, and tryptophan metabolism early on, and fatty acid biosynthesis and pyruvate metabolism later.

Conclusions:

• Metabolomic profiles dynamically change after acute brain injuries.
• Common metabolites serve as significant prognostic markers across various brain injury types.
• These findings highlight the potential of metabolomics for predicting outcomes in neurological conditions.