Mass spectrometry–based metabolomics for improved diagnosis and severity stratification of SARS‑CoV‑2 infection: a narrative review

During the COVID-19 pandemic, diagnostic laboratories worldwide faced intense pressure from high testing demand, limited infrastructure, and shortages of trained personnel, highlighting the need for complementary strategies to improve case detection, stratify disease severity, and support clinical decision-making. Mass spectrometry–based metabolomics has emerged as a promising approach, enabling comprehensive characterization of host metabolic responses to SARS-CoV-2 across multiple biofluids. This narrative review critically appraises clinical evidence on mass spectrometry–based metabolomic signatures associated with SARS-CoV-2 infection, diagnosis, and severity stratification, covering studies published between 2020 and early 2025. Using predefined criteria in PubMed, Scopus, and Web of Science, we identified 42 original clinical studies employing liquid chromatography–MS (LC-MS), gas chromatography–MS (GC-MS), or nuclear magnetic resonance (NMR) spectroscopy. Consistent disturbances emerged in amino acid, lipid/lipoprotein, purine, tryptophan–kynurenine, and urea-cycle metabolism. Metabolites including kynurenine, citrulline, carnitines, acylcarnitines, selected phospholipids, and oxylipins frequently discriminated infected from non-infected individuals and distinguished mild from severe disease, across serum, plasma, whole blood, respiratory samples, urine, feces, and exhaled breath. Current evidence supports mass spectrometry–based metabolomics as a complementary tool for COVID-19 diagnosis and severity assessment, offering a framework for biomarker validation and development of rapid, minimally invasive diagnostics.