BACKGROUND: The novel coronavirus infection (COVID-19) is associated with a high risk of major adverse cardiovascular events (MACE) and venous thromboembolism in the post-hospitalization period. AIM: To identify early predictors and develop a model for predicting long-term MACE in patients who have recovered from COVID-19, based on analysis of clinical, laboratory, and instrumental data. METHODS: A prospective study was conducted involving 112 patients without a history of cardiovascular disease who were hospitalized with a confirmed diagnosis of COVID-19 and grade 1–4 lung involvement on computed tomography. Upon admission, in addition to standard clinical, laboratory, and instrumental evaluation, levels of high-sensitivity troponins T and I, N-terminal pro-brain natriuretic peptide, soluble suppression of tumorigenicity 2 (sST2), pentraxin-3 (PTX-3), and galectin-3 (Gal-3) were measured. During a two-year follow-up, the development of MACE (acute cerebrovascular accident, acute myocardial infarction, pulmonary embolism, cardiovascular death) was assessed. Potential predictors of MACE were identified using logistic regression with stepwise variable selection (Forward method). RESULTS: During the follow-up period, MACE was recorded in 12.5% of patients. Differences in clinical, demographic, laboratory, and instrumental characteristics between patients with and without MACE were analyzed. Three independent predictors of MACE were identified: Gal-3 5.1 ng/mL (p 0.001), sST2 48 ng/mL (p 0.001), and PTX-3 9.6 ng/mL (p 0.001). Based on the multivariate model incorporating these three predictors, a calculator was developed to estimate the probability of MACE in patients without a history of cardiovascular disease who have recovered from COVID-19. CONCLUSION: The developed mathematical model has high predictive accuracy (80.0%) and can identify patients at high risk of MACE during post-hospital follow-up.