Acute myocardial infarction (AMI) is a serious disease which threatens public health. Exosomes (exos) contain certain genetic information and are important communication vehicles between cells. In the present study, different exosomal microRNAs (miRs), which exhibit a notable association between expression levels in plasma and AMI were assessed to support the development of new diagnostic and clinical assessment markers of patients with AMI. In total, 93 individuals, including 31 healthy controls and 62 patients with AMI, were recruited for the present study. Data on age, blood pressure, glucose levels, lipid levels and coronary angiography images were collected from the enrolled individuals, and plasma samples were collected. Plasma exos were extracted and verified using ultracentrifugation, transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA) and western blotting (WB). Exo-miR-4516 and exo-miR-203 in plasma exos were identified by exosomal miRNA sequencing analysis, reverse transcription-quantitative PCR was performed to detect the levels of exo-miR-4516 and exo-miR-203 in plasma exos, and ELISA was performed to detect the levels of secretory frizzled-related protein 1 (SFRP1) in samples. The correlation analysis between exo-miR-4516, exo-miR-203 and SFRP1 in plasma exos and AMI was presented as receiver operating characteristic curves (ROCs) of the SYNTAX score, cardiac troponin I (cTnI), low-density lipoprotein (LDL) and each indicator separately. Kyoto Encyclopedia of Genes and Genomes enrichment analysis was performed to predict relevant enrichment pathways. Exos were successfully isolated from plasma by ultracentrifugation, which was confirmed by TEM, NTA and WB. Exo-miR-4516, exo-miR-203 and SFRP1 levels in plasma were significantly higher in the AMI group compared with the healthy control group. ROCs demonstrated that exo-miR-4516, exo-miR-203 and SFRP1 levels had a high diagnostic efficiency in predicting AMI. Exo-miR-4516 was positively correlated with SYNTAX score, and plasma SFRP1 was positively correlated with plasma cTnI and LDL. In conclusion, the data demonstrated that exo-miR-4516, exo-miR-203 and SFRP1 levels could be used in combination to diagnose and assess the severity of AMI. The present study was retrospectively registered (TRN, NCT02123004).