Introduction: Atrial-selective inhibition of cardiac sodium channel has been shown to underlie the effectiveness of ranolazine in the suppression of atrial fibrillation. This study investigates the mechanism responsible for differences in the interaction of ranolazine with sodium channels in atrial vs ventricular myocytes. Methods: Whole cell sodium current was recorded at 150C in isolated canine atrial and ventricular myocytes, and in HEK293 cell line expressing SCN5A. Protocols were designed to determine interaction of ranolazine with resting, inactivated, and open states of sodium channels. Single pulses and trains of 40 pulses were elicited over a range of holding potentials in the absence and presence of ranolazine to yield tonic and use-dependent block. Development of block during pulse trains in myocytes was analyzed using guarded-receptor theory. Results: Tonic block was negligible at holding potentials up to -100 mV, suggesting minimal drug interaction with resting and inactivated states. However, use-dependent block was increased with more depolarized holding potentials, indicating ranolazine trapping by the inactivation gate. Train protocols demonstrated significant effect of shorter diastolic intervals to increase use-dependent block, but a lack of effect of longer pulse durations. Effects in atrial and ventricular myocytes, and in HEK293 cells followed a similar pattern.Conclusions: Ranolazine is a potent open-state blocker of sodium channels that unbinds from the resting channels unusually fast and is trapped in the inactivated state. Kinetic rates of ranolazine interaction with different states of atrial and ventricular sodium channels are similar. Ranolazine inhibition of sodium-channels is atrial-selective due to a more negative position of the steady-state inactivation curve, more positive resting membrane potential, as well as more positive take-off potential and shorter diastolic interval in atrial vs ventricular myocytes at fast rates.