Dose-dependent pulmonary vein reconnection in response to adenosine: relevance of atrioventricular block during infusion.

purposeAdenosine infusion during atrial fibrillation ablation is often used to evaluate for dormant pulmonary vein (PV) conduction. Several physiologic effects of adenosine may be seen during infusion, including atrioventricular (AV) block, hypotension, sinus tachycardia (as a reflex mechanism), and sinus bradycardia. However, hypotension and sinus tachycardia may be seen in the absence of AV block. Whether the dose required to achieve AV block versus any physiologic effect (e.g., hypotension) is relevant when evaluating for PV reconnection is unclear.

methodsIn consecutive patients undergoing first-time atrial fibrillation radiofrequency ablation for persistent or paroxysmal atrial fibrillation, adenosine testing was performed 30 min after successful PV isolation. When testing each PV after isolation, a circular mapping catheter was placed at the ostium and a 12-mg bolus of adenosine was given via peripheral line. The presence of some physiologic effect (e.g., sinus tachycardia or hypotension) versus AV block was recorded as was PV reconnection. If no AV block and no PV reconnection were seen, the dose was escalated by 6-mg increments to assess for PV reconnection until AV block or PV reconnection was achieved.

resultsAmong 50 patients, there were 3 to 5 PVs isolated (median 4; mean 4.4). In 35 patients, a 12-mg bolus resulted in AV block and 16/155 (10 %) veins exhibited reconnection requiring additional ablation. In all other patients, sinus tachycardia and hypotension were seen in response to 12-mg infusion but no AV block and no PV reconnection was seen. No sinus bradycardia was seen in any patients. In 10/15 patients without AV block with 12 mg, 18 mg of adenosine was sufficient to achieve AV block and 5/43 (12 %) PVs exhibited reconnection. Among the remaining 5 patients, 3 patients required a dose of 24 mg to achieve AV block, and PV reconnection was seen in 1/14 (7 %) PVs. The remaining two patients required a dose of 30 mg to achieve AV block, but no PV reconnection was seen. There was no difference in symptomatic recurrence rate between those who exhibited adenosine-induced reconnection and those who did not after 1-year follow-up.

conclusionA fixed dose of adenosine may be insufficient to assess for PV reconnection. Furthermore, the presence of a physiologic effect (e.g., sinus tachycardia or hypotension) in the absence of AV block may reflect an insufficient dose. The dose needed may vary between patients, and AV block may reflect a minimal requirement when administering adenosine to assess for PV reconnection.