Antiarrhythmic therapy for the suppression of atrial flutter has conventionally entailed the use of a class Ia agent such as quinidine or procainamide. However, atrial flutter often recurs despite the use of these conventional antiarrhythmic regimens. Experimental and clinical evidence suggests that the pharmacologic suppression of atrial flutter may depend on the prolongation of the atrial action potential duration and consequently the voltage-dependent refractoriness. Therefore, the efficacy and tolerance of the class III antiarrhythmic agent N-acetylprocainamide was compared to that of the conventional regimen of the class Ia agent quinidine combined with digoxin (to control ventricular response) in patients with a history of symptomatic sustained atrial flutter. The study was randomized but nonblinded, with a crossover to the alternate regimen if the first failed. Eighteen patients entered the study and were followed for up to 18 months. Of the 12 receiving N-acetylprocainamide (eight randomized and four crossovers), one (8%) failed therapy due to side effects, but none had atrial flutter. Of the 11 receiving quinidine and digoxin (10 randomized and one crossover), three (28%) had a recurrence of atrial flutter, two of whom also had intolerable side effects, and two more (18%) had side effects alone requiring withdrawal of therapy (total 46% failed). The probability of therapeutic success over time was greater (p less than 0.04) for N-acetylprocainamide than for quinidine and digoxin. The data suggest that N-acetylprocainamide may be more effective and better tolerated than the conventional regimen of quinidine plus digoxin. Therefore, large-scale blinded studies of the efficacy of N-acetylprocainamide in the suppression of atrial flutter may be warranted.