BACKGROUND We recently developed a goat model of sustained atrial fibrillation (AF) in which repetitive induction of AF by burst pacing shortened the atrial effective refractory period (AERP) (electrophysiologic remodeling) and progressively prolonged the paroxysms of AF to become sustained (24 hours) within 1 to 3 weeks (atrial fibrillation begets atrial fibrillation). The aim of the present study was to study the effect of Class I drugs in this animal model of chronic AF. RESULTS The effects of hydroquinidine (HQ) on seven chronically fibrillating goats and of flecainide (Fl) on nine goats were studied. Both drugs were infused intravenously until sustained AF was cardioverted or adverse drug effects occurred. HQ and Fl restored sinus rhythm in 86% and 67% of the cases. Adverse drug effects occurred in 14% and 56%, respectively. The average atrial cycle length of AF (AFCL) was prolonged to a different degree. Just before restoration of sinus rhythm, the two drugs had increased AFCL by 72% and 50%. The duration of the QRS complex was prolonged 17% by HQ and 50% by Fl. The RR interval was not affected by HQ and was prolonged slightly by Fl. Directly after restoration of sinus rhythm, the AERP during pacing with an interval of 400 msec was 92 +/- 29 (HQ) and 66 +/- 10 msec (Fl) (control value: 149 +/- 10 msec). Intra-atrial conduction velocity was 83 +/- 7 and 86 +/- 11 cm/sec (control value: 116 +/- 10 cm/sec). Although both drugs were effective in terminating AF, after cardioversion the atrial vulnerability was still very high and a single premature stimulus reinduced AF in 100% of the animals. As a result of the short AERP by the AF-induced remodeling and the depressed intra-atrial conduction by the Class I drugs, directly after cardioversion the atrial wavelength was abnormally short (between 5.7 and 7.5 cm). This explains the still high atrial vulnerability to AF directly after cardioversion by Class I drugs. Surprisingly, the prolongation of AFCL by either Class I drug was not due to lengthening of the functional refractory period but rather to a widening of the excitable gap during AF. CONCLUSIONS In a goat model of chronic AF, infusion of Class IA and Class IC drugs restored sinus rhythm in 67% to 86% of the cases. However, due to the short AERP and the depressed intra-atrial conduction directly after cardioversion, the atrial vulnerability was still very high and a premature beat easily reinduced AF.