Identifying changes in serum metabolites before the occurrence of acute myocardial infarction (AMI) is an important approach for finding novel biomarkers of AMI.In this prospective cohort study, serum samples obtained from patients at risk of AMI (n = 112) and non-risk controls (n = 89) were tested using high-resolution metabolomics (HRM). Partial least-squares discriminant analysis (PLS-DA), along with univariate analysis using a false discovery rate (FDR) of q = 0.05 were performed to discriminate metabolic profiles and to determine significantly different metabolites between healthy control and AMI risk groups.PLS-DA significantly separated the AMI risk sera from control sera. The metabolites associated with amino acid biosynthesis, 2-oxocarboxylic acid, tryptophan, and amino sugar and nucleotide sugar metabolism pathways were mainly elevated in patients at risk of AMI. Further validation and quantification by MS/MS showed that tryptophan, carnitine, L-homocysteine sulfinic acid (L-HCSA), and cysteic acid (CA) were upregulated, while L-cysteine and L-cysteine sulfinic acid (L-CSA) were downregulated, specifically among AMI risk sera. Additionally, these discriminant metabolic profiles were not related to hypertension, smoking or alcoholism.In conclusion, detecting upregulated L-HCSA and CA along with carnitine among patients at risk for AMI could serve as promising non-invasive biomarkers for early AMI detection.