Type II atrial flutter (AFII) is an arrhythmia which usually cannot be interrupted by atrial pacing: the underlying mechanism is considered to be a leading circle without an excitable gap. We investigated whether the administration of propafenone, an antiarrhythmic drug, which primarily decreases conduction velocity, has a beneficial effect on AFII interruption using transesophageal pacing. Twelve patients with an AFII were randomized into 2 groups in which pacing was performed without treatment (group A) or two hours after the administration of 600 mg of oral propafenone (group B). Sinus rhythm was attained in 0 of 6 patients in group A and in 4 of 6 patients in group B (P < 0.05). The baseline mean cycle length was the same in both groups (175 +/- 7 (A) vs 168 +/- 8 ms (B); it lengthened significantly after the administration of propafenone (219 +/- 33 vs 168 +/- 8 ms; P < 0.05). Propafenone did not significantly lengthen the cycle in the two patients in whom interruption of the arrhythmia was impossible. Our data show that propafenone has a facilitating effect on atrial pacing only when it significantly prolongs the cycle length of the arrhythmia, possible expression of a conversion of AFII into type I, with an anatomical substrate and an excitable gap allowing arrhythmia capture and interruption. In the two patients in whom sinus rhythm was not restored, the absence of a direct dependence of the cycle length on the change in conduction velocity induced by propafenone may be explained by the persistence of a functionally determined circuit, resistant to atrial pacing.