Differential effectiveness of pharmacological strategies to reveal dormant pulmonary vein conduction: A clinical-experimental correlation

Abstract

Atrial fibrillation recurs in ∼30%-40% of patients after pulmonary vein (PV) isolation (PVI) procedures, often because of restored PV-left atrial (LA) conduction. Adenosine or isoproterenol are used clinically to reveal dormant PV conduction and guide additional ablation. The purpose of this study was to assess the differential efficacy of adenosine and/or isoproterenol in revealing dormant PV conduction. In 25 patients undergoing PVI, dormant conduction was assessed sequentially in response to intravenous adenosine, isoproterenol, and adenosine plus isoproterenol in 100 PVs. To study mechanisms, PVs were isolated by radiofrequency ablation in coronary-perfused canine LA-PV preparations. After PVI, resting membrane potential from PV cells was recorded before and after 1 mM adenosine, 1 μM isoproterenol, 1 μM isoproterenol plus 1 mM adenosine, or no drug (controls). Clinical PVI was successful in all 100 PVs, with dormant conduction in 31. Sensitivity for dormant conduction was isoproterenol 10%; adenosine 87% (P <.001 vs. isoproterenol); and isoproterenol + adenosine 100% (P = .13 vs. adenosine). Dormant PV conduction in vitro was revealed with adenosine (53%) and adenosine + isoproterenol (60%) but not with isoproterenol alone or in controls (P <.01). Radiofrequency lesions producing PVI depolarized resting membrane potential, causing inexcitability. Postablation, resting membrane potential hyperpolarized after both adenosine and isoproterenol, but adenosine-induced changes were greater (9.1 ± 0.6 mV, vs. 3.8 ± 0.6 mV; P <0.001), with no significant additional effect when isoproterenol was added to adenosine. Adenosine is superior to isoproterenol in revealing dormant PVs clinically and experimentally because of more effective adenosine-induced hyperpolarization. Adding isoproterenol to adenosine had no significant additional value.