Aim To evaluate the predictive value of indexes of left ventricular mechanical dyssynchrony (MD) as determined by data of electrocardiogram (ECG)-gated myocardial perfusion scintigraphy (ECG-MPS) for prediction of the efficacy of resynchronization therapy (RT) in patients with chronic heart failure (CHF).Material and methods This prospective study included 32 patients with nonischemic CHF and standard indications for RT. All patients underwent complete clinical an instrumental examination, including 24-h ECG monitoring and echocardiography (EchoCG). In order to evaluate the left ventricular (LV) myocardial perfusion, contractile function, and MD, myocardial perfusion scintigraphy was performed for all patients at rest prior to RT. In addition to the perfusion defect size at rest and hemodynamic parameters, LV MD was determined. The following indexes were used for analysis of dyssynchronization: phase standard deviation (PSD), phase histogram bandwidth (HBW), and phase histogram asymmetry and steepness. The treatment efficacy was evaluated by the clinical status of patients (clinical condition evaluation scale for CHF patient) and EchoCG at 6 months following RT. The criteria for a positive response to RT were an increase in LV ejection fraction (EF) by 5% and/or a decrease in the LV end-diastolic volume by 15% compared to preoperative values.Results According to ECG-MPS findings, all patients had scintigraphic signs of severe CHF with dilated LV cavity (end-diastolic volume, EDV 246 [217; 269] ml) and also of pronounced mechanical and electrical dyssynchrony. The values of mechanical dyssynchrony were PSD 53 [41; 61], HBW 176 [136; 202], asymmetry 1.62 [1.21; 1.89], and steepness 2.81 [1.21; 3.49]. The QRS duration was 165 [155; 175] msec. Furthermore, the LV perfusion was moderately impaired (perfusion defect size 4 [3; 10] %). Mean follow-up duration after implantation of the resynchronizing device was 6±1.7 mos. According to the selected criteria, 20 (63 %) patients were considered as responders and 12 (37 %) patients as non-responders. Before implantation of the cardiac synchronizing device, responders and non-responders differed only in LV MD (PSD 44 [35; 54] vs. 63 [58; 72]; p=0.0001); HBW 158 [118; 179] vs. 205 [199; 249]; p=0.0001; asymmetry 1.77 [1.62; 2.02] vs. 1.21 [0.93; 1.31]; p=0.0001; steepness 3.03 [2.60; 3.58] vs. 1.21 [0.19; 1.46]; p=0.0001), respectively. A one-factor logistic regression analysis showed that MD values were statistically significant predictors of a positive response to RT. A multi-factor logistic analysis of phase histogram steepness (odds ratio, OR 1.196; 95 % confidence interval, CI 1.04-1.37) and PSD (OR 0.67; 95 % CI 0.47-0.97) were identified as independent predictors for the response to RT. According to results of the ROC analysis, a PSD <55 and a phase histogram steepness >1.54 may predict the effectiveness of RT (AUC= 0.92; р=0.0001).Conclusion LV MD parameters determined with ECG-MPS allow predicting the effectiveness of RT in patients with nonischemic CHF. In this patient group, high values of standard deviation and low values of phase histogram steepness were independent predictors for the absence of response to RT after 6 mos. of follow-up.