Origin and ablation of the adenosine triphosphate induced atrial fibrillation after circumferential pulmonary vein isolation: effects on procedural success rate.

INTRODUCTION: Adenosine triphosphate (ATP) has been used to provoke dormant pulmonary vein (PV) conduction after circumferential PV isolation (CPVI). However, there have been no systematic studies examining the incidence and the mechanism of ATP-induced atrial fibrillation (AF) following CPVI in paroxysmal AF. In this study, we explore the mechanism of ATP-induced AF and assess the feasibility of eliminating this response by additional radiofrequency (RF) ablation.

METHODS AND RESULTS: A total of 300 consecutive patients with paroxysmal AF underwent CPVI. After all PVs were isolated, intravenous ATP (40 mg) was administered during an intravenous isoproterenol (ISP) infusion (5 μg/min). AF was reproducibly induced by ATP in 39 patients. Non-PV foci were confirmed and located in 29 of these patients at the onset of AF, including 27 foci in the superior vena cava (SVC), 1 focus in the crista terminalis, and 1 focus near the antrum of the PV. In all these cases, ATP-induced AF was eliminated after the non-PV foci were successfully ablated. For the other 10 patients, the foci triggering AF could not be confirmed or located due to the transient effect of ATP, thus no further ablation was performed. After a mean follow-up period of 18.7 ± 6.4 (8-24) months, the success rate in the ATP-induced AF group was not significantly different compared with the conventional treatment group who did not exhibit ATP-induced AF (76.9% vs 67.3%; P = 0.25). But in the subgroup of which the ATP-induced AF could be eliminated by additional RF ablation, the success rate was significantly higher than the non-ATP inducible group (86.2% vs 67.3%; P = 0.04).

CONCLUSIONS: A large proportion of the ATP-induced AF post CPVI were initiated by rapid firing in the SVC. Eliminating this response by additional ablation may have an influence on clinical results of paroxysmal AF ablation.