The prevalence of endovascular interventions for coronary chronic total occlusion (CTO) remains small worldwide. This is due to the complexity of procedure and the risk of intraoperative complications. In this regard, the search for predictors of unsuccessful endovascular intervention in CTO plays a special role. This will allow for a careful selection of patients with the most favorable expectation effect of the operation. Aim. To identify predictors of unsuccessful endovascular recanalization of CTO.Material and methods. This retrospective study included 180 patients with chronic coronary artery disease (CAD) in the period from November 2017 to June 2019, who had multivessel lesion in combination with CTO. In all patients, an attempt was made to achieve complete myocardial revascularization. Depending on the success of procedure, the patients were divided into two groups: complete and incomplete myocardial revascularization. The follow-up period was 12 months.Results. All baseline characteristics of patients in the compared groups were similar. The successful recanalization rate of occlusion was 79,5%. Multivariate regression analysis showed that calcified CTO (p<0,001), baseline SYNTAX (Synergy between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery) score >32 (p=0,006), CTO length >30 mm (p=0,046) and CTO of circumflex artery (p<0,01) are significant predictors of unsuccessful endovascular recanalization of CTO. To assess the predictive value of the model, a ROC analysis was carried out, and the area under the curve (AUC) was calculated. The AUC was 0,87, which indicates a high predictive quality of the model. The sensitivity and specificity of the model were 78 and 81%, respectively.Conclusion. The study showed that the presence of calcified CTO, SYNTAX score >32 points, CTO length >30 mm, and CTO of circumflex artery are significant predictors of unsuccessful CTO recanalization.