Background: Biochemical detection of chronic stable angina (CSA) and myocardial infarction (MI) are challenging. To address the shortcomings of the conventional biochemical approach for detection of MI, we applied serum lacking proteins and lipoprotein-based metabolomics in an approach using proton nuclear magnetic resonance (1 H NMR) spectroscopy for screening of coronary artery disease (CAD) and especially MI. Our aim was to discover differential biomarkers among subjects with normal coronary (NC), CSA, and MI. Methods: The study comprised serum samples from nondiabetic angiographically proven CAD [CSA (n = 88), MI (n = 90)] and NC (n = 55). 1 H NMR spectroscopy was used to acquire metabolomics data. Clinical variables such as troponin I (TI), lactate dehydrogenase (LD), creatine kinase (CK, CK-MB, CK-MM), serum creatinine, and lipid profiles were also measured in all subjects. Metabolomic data and clinical measures were appraised separately using a chemometric approach and ROC analysis. Results: The screening outcomes revealed that the pattern of methylguanidine, lactate, creatinine, threonine, aspartate, and trimethylamine (TMA), and TI, LD, CK, and serum creatinine were changed in CAD compared to NC. Statistical analysis demonstrated high precision (93.6% by NMR and 67.4% by clinical measures) to distinguish CAD from NC. Further analysis indicated that methylguanidine, arginine, and threonine, and TI, LD, and serum creatinine were significantly changed in CSA compared to MI. Statistical analysis demonstrated high accuracy (88.2% by NMR and 92.1% by clinical measures) to discriminate CSA from MI. Conclusions: In contrast to other laboratory methods, 1 H NMR-based metabolomics of filtered sera appears to be a robust, rapid, and minimally invasive approach to probe CSA and MI. IMPACT STATEMENT This study was conducted on Indian populations comprising (a) patients with normal coronary, (b) patients with chronic stable angina, and (c) patients with myocardial infarction. The inference of this study is to not only develop biomarkers of these ailments, but to also find new mechanisms for the underlying pathology. In contrast to the conventional approach, the novel filtered-serum-based approach and application of advanced level statistical analysis provide the evidence that metabolomics may be a better choice for management of the underlying pathology. This study reveals profound information using a novel method, contributes to the advanced-level information that can be used for discovery of CSA and MI biomarkers, and may reveal the precise mechanism for underlying pathology.