Background: Stroke is a significant health issue in the United States and identifying biomarkers for prevention and functional recovery of acute ischemic stroke (AIS) remains the highest priority. This study aims to identify metabolite and microRNA (miRNA) signatures that may be associated with AIS pathophysiology by performing longitudinal and case-control studies. Methods: We evaluated 260 patients during acute (<72 hours, T1), and chronic (3-5 months, T2) phases of stroke with 160 controls by performing global metabolomics using high-throughput nuclear magnetic resonance spectroscopy. In a subset, we also examined differential expression patterns of blood and serum exosome miRNAs between the T1 and T2 using RNA sequencing. Results: Our study identified 19 metabolites to be significantly perturbed during the T1 using the orthogonal partial least square discrimination analysis and multivariate mixed regression. Phenylalanine showed a 15.9-fold increase in T1 and remained high at the T2 stage. The Apolipoprotein B/Apolipoprotein A-1 ratio and ketone bodies were markedly increased while several essential fatty acids and amino acids were decreased in the T1. Our data also identified 3 exosomal miRNAs upregulated in the T1 stage. The miR-9-3p showed a 14.7-fold increase in T1 (Bonferroni p= 9.44x10 -6 ) which is strongly supported by the literature to trigger brain injury and cell death via the caspase pathway. Increased expression of miR-4508 correlated positively with phenylalanine both in T1 (r = 0.88, p = 0.001) and T2 (r = 0.71, p = 0.02). Conclusion: This study elucidates significant alterations in amino acids, fatty acids, and apolipoprotein patterns between the acute and chronic stages of stroke compared to controls. Our results on exosomal miRNAs and pathway analysis using metabolites identified four major pathways; apoptosis, excitotoxicity, oxidative stress, and stroke-mediated inflammation to play a critical role in the AIS leading to neuronal cell death. Further validation in independent datasets is warranted, and if confirmed some of these may become clinically meaningful biomarkers to predict AIS in humans.