Downregulation of glucose and fatty acid oxidation occurs in heart failure (HF). Trimetazidine reduces fatty acid oxidation and increases glucose oxidation. In this single-blind study, trimetazidine, 20 mg three times per day (n = 51) or placebo (n = 36) was added to treatment of 87 HF patients receiving optimal HF therapy. Etiology of heart failure was coronary artery disease in 35 patients (68.6%) in the trimetazidine group and 22 (62.9%) in the placebo group. Fourteen (27.5%) patients in the trimetazidine group and 11 (31.4%) patients in the placebo group had diabetes. Peak systolic velocity (Vs), and the peak early diastolic (Vd) and late diastolic (Va) velocities of various segments left and right ventricles (RV) were obtained with tissue Doppler imaging (TDI) and averaged. Patients were re-evaluated three months later. Significant increases in mean left ventricular ejection fraction (LVEF) (33.3% ± 5.6% to 42.4% ± 6.3%, P < 0.001 and 30.6% ± 8.2% to 33.2% ± 6.6%, P = 0.021) and LV and RV myocardial velocities and mitral and tricuspid annular TDI velocities were observed in both groups. However, compared to placebo, increments in LVEF (9.1% ± 4.2% vs. 2.5% ± 1.4%, P < 0.001) and myocardial velocities were significantly higher with trimetazidine (P < 0.001 for LV Vs, Vd, Va; P = 0.035 for RV Vd; and P < 0.001 for RV Va and Vs). Increase in LVEF with trimetazidine was significantly correlated with presence of diabetes (r = 0.524, P < 0.001). With trimetazidine LVEF increased significantly more in diabetic patients compared to nondiabetics (P < 0.001). Also, patients having both diabetes and ischemic HF tended to have greater improvement in LVEF compared to ischemic HF patients without diabetes (P = 0.063). Addition of trimetazidine to current treatment of HF, especially for those who are diabetic, may improve LV and RV functions.