Identification of a novel metabolic and lipidomic signature for 90-day functional outcome after ischemic stroke: an exploratory proof-of-concept study

Abstract

Background Ischemic stroke is a leading cause of death and disability worldwide, yet the systemic metabolic mechanisms underlying poor functional recovery remain incompletely understood. Dysregulated lipid metabolism is a key pathogenic factor, and advances in metabolomics and lipidomics provide new opportunities for discovering pathophysiological signatures. Methods In this exploratory study, we integrated serum metabolomic and lipidomic profiling in 44 patients with acute ischemic stroke, stratified by favorable (n = 23) or unfavorable (n = 21) 90-day functional outcomes. Differential metabolites and lipids were identified, pathway-related molecular alterations were characterized, and candidate markers were evaluated to establish a proof-of-concept metabolic signature. Results Patients with unfavorable outcomes exhibited global metabolic reprogramming characterized by structural lipid disassembly (widespread reductions in membrane phospholipids, including phosphatidylcholines and ether phospholipids), mitochondrial energy failure (acylcarnitine abnormalities consistent with impaired β-oxidation), and antioxidant depletion (marked by decreased cis-caffeic acid). An exploratory three-marker panel—the C8:2-OH(3)/C18:2-OH(3) ratio, the free DHA/PC(44:12) ratio, and cis-caffeic acid—showed promising discriminatory potential in this derivation cohort (AUC = 0.950). Adding this metabolic panel to the initial NIHSS score improved the AUC to 0.975, suggesting incremental value beyond clinical assessment alone. Conclusion Integrated serum metabolomics and lipidomics identified a metabolic signature reflecting lipotoxicity, mitochondrial dysfunction, and oxidative stress. These preliminary findings highlight systemic metabolic reprogramming as a potential driver of unfavorable outcome and provide a proof-of-concept for combining molecular phenotyping with conventional clinical scales. Without external validation and multiple-testing correction, these results should be viewed as hypothesis-generating.