ПРОГНОСТИЧЕСКАЯ МОДЕЛЬ ДЛЯ ВЫБОРА МЕТОДИКИ РЕКАНАЛИЗАЦИИ ХРОНИЧЕСКИХ ОККЛЮЗИЙ КОРОНАРНЫХ АРТЕРИЙ

Background. Despite signifcant progress in the feld of coronary interventions, chronic total occlusion (CTO) represents a signifcant challenge for interventional cardiologists. Aim. To develop the score, able to predict technical success of CTO PCI and facilitate the choice of recanalization strategy. Methods. A total of 665 CTO patients who underwent 681 PCI in the period from 2014 to 2018 in Meshalkin National Medical Research Center were included in this study. Clinical and angiographic characteristics were analyzed. 477 CTO PCI were randomly assigned to the derivation set, 204 CTO PCI – validation set. The prognostic model was developed by assigning a score for each independent predictor of procedural failure in accordance with beta coeffcients and summing up all scores. Results. Procedural success was 76.7%. Five predictors of procedural success were included into the fnal multivariable model: bending (1 score), calcifcation (1 score), ambiguous stump (1 score), “donor” artery disease (1 score), non-RCA CTO (0.5 scores). Based on these predictors, 4 categories of CTO complexity were highlighted: 0–1 scores (easy), 1–2 scores (intermediate), 2–3 scores (diffcult), > 3 scores (very diffcult). The score demonstrated a good discriminatory ability (AUC 0.709, 95% CI 0.658–0.760). According to the novel score retrograde approach may have an advantage in patients with a > 3 scores, which corresponds to the "very diffcult" class of complexity. Conclusion. The novel score can be used in clinical practice for predicting the success of CTO PCI and determining initial crossing strategy.