Aim. The aim of this study was to develop additional selection criteria for implanted cardioverter-defibrillator (ICD) implantation in the primary prevention of sudden cardiac death (SCD) based on the risk stratification for the development of sustained ventricular tachycardia (VT).Methods. The study included 451 patients with heart failure and reduced left ventricular ejection fraction (HFrEF) who were referred for ICD implantation for primary prevention of SCD. Participants underwent pre-implantation screening of clinical, instrumental, and laboratory parameters, followed by prospective observation for 24 months to record the first occurrence of sustained VT or justified ICD therapy. To achieve the study’s goal, training and test samples were formed.Results. The arrhythmic endpoint was recorded in 84 patients (26%) in the training group and in 35 patients (27%) in the test group. Univariate analysis identified 11 factors with the highest predictive potential (p<0.1) associated with the occurrence of the studied endpoint. These included clinical data: coronary artery disease, arterial hypertension, resting heart rate >80 bpm; electrocardiographic parameters: complete left bundle branch block according to Strauss criteria, P-wave duration (lead II) >120 ms, or the presence of atrial fibrillation (in the case of persistent form), index of cardiac electrophysiological balance (ICEB) >3.1; echocardiographic parameters: presence of eccentric left ventricular hypertrophy, global longitudinal strain ≥ minus 6%; laboratory markers: galectin-3 >12 ng/ml, sST-2 >35 ng/ml, NT-proBNP >2000 pg/ml. Based on the regression coefficients, points were assigned to each factor, and the sum of these points determined the value of a new proposed index - the arrhythmic risk index (ARI). ARI values >5 points predicted the two-year likelihood of VT in HFrEF patients with a sensitivity of 78.6% and specificity of 64.3% (AUC=0.788±0.028 with 95% confidence interval (CI): 0.732-0.843; p=0.0001). The application of ARI in the test group demonstrated good model performance in predicting two-year VT risk (AUC=0.652±0.053 with 95% CI: 0.547-0.757; p=0.008).Conclusion. Based on the obtained results, a predictive index was developed, allowing for personalized and timely risk assessment of VT in patients with HFrEF.