Background: Multiple echocardiographic criteria have been proposed to diagnose mechanical dyssynchrony in patients with heart failure without being validated against a model of cardiac dyssynchrony with heart failure. This study examines which of these methods can detect dyssynchrony in a canine model. Methods: Adult mongrel dogs underwent His bundle ablation and right ventricular pacing for 4 weeks at either 110 bpm to induce dyssynchrony without heart failure (D group, n=12) or 170 bpm to induce dyssynchrony with heart failure (DHF group, n=9). To induce heart failure with narrow QRS, atria were paced at 190 bpm for 4 weeks (HF group, n=8). Tissue Doppler imaging (TDI) and 2D echocardiography were performed at baseline and end of the study. Standard deviation of time to peak systolic velocity (color coded TDI), time to peak S wave on pulse wave TDI, time to peak radial and circumferential strain by speckle tracking analysis (tErr-SD, tEcc-SD respectively) as well as septal to posterior wall motion delay on M-mode were obtained. Results: In D group, only tErr-SD and tEcc-SD were increased by dyssynchrony. In contrast, all the echocardiographic parameters of dyssynchrony appeared significantly augmented in the DHF group. Receiver operator curve analysis showed good sensitivity of tErr-SD (90%) and tEcc-SD to (100%) detected dyssynchrony without heart failure and excellent sensitivity and specificity of tErr-SD and tEcc-SD to detect dyssynchrony with heart failure Conclusion: Radial strain by speckle tracking is more accurate than TDI velocity to detect cardiac dyssynchrony in a canine model of dyssynchrony with or without heart failure.