Ranolazine has been shown to produce atrial-selective depression of sodium channel-dependent parameters and suppress atrial fibrillation (AF) in a variety of experimental models. The present study contrasts the effects of ranolazine and those of a clinically used anti-AF class IC agent, propafenone. Electrophysiological and anti-AF effects of propafenone and ranolazine were compared at clinically relevant concentrations (i.e., 0.3-1.5 and 1-10 μM, respectively) in canine isolated coronary-perfused atrial and ventricular preparations. Transmembrane action potential and pseudo-ECG were recorded. Both ranolazine and propafenone produced atrial-selective prolongation of action potential duration. Propafenone depressed sodium channel-mediated parameters [maximum rate of rise of the action potential upstroke (V(max)), conduction time, and diastolic threshold of excitation] and induced postrepolarization refractoriness to a greater degree than ranolazine, and these effects, unlike those induced by ranolazine, were not or only mildly atrial-selective at normal rates (cycle length 500 ms). At fast pacing rates, however, the effects of propafenone on V(max) and conduction time became atrial-selective, because of the elimination of diastolic interval in atria, but not in ventricles. Propafenone (1.5 μM) and ranolazine (10.0 μM) were effective in preventing the initiation of persistent acetylcholine-mediated AF (6/7 and 9/11 atria, respectively), its termination (8/10 and 8/12 atria, respectively), and subsequent reinduction (8/8 and 7/8 atria, respectively). Thus, propafenone and ranolazine both suppress AF, but ranolazine, unlike propafenone, does it with minimal effects on ventricular myocardium, suggesting a reduced potential for promoting ventricular arrhythmias.