Relationship between atrial scar on cardiac magnetic resonance and pulmonary vein reconnection after catheter ablation for paroxysmal atrial fibrillation.

INTRODUCTION: Pulmonary vein (PV) reconnection is frequent in patients showing atrial fibrillation (AF) recurrence after PV isolation (PVI). Its detection with cardiac magnetic resonance (CMR) may help predict outcome and guide redo procedures. We assessed the relationship between scar on CMR and PV reconnection after catheter ablation for paroxysmal AF.

METHODS AND RESULTS: Fifty-one patients with paroxysmal AF underwent CMR before PVI using either a conventional single-electrode catheter (N = 28) or a circular multielectrode catheter (N = 23). At 3 months, a second CMR study was performed, followed by a systematic electrophysiological procedure to look for PV reconnection, regardless of AF recurrence. Preablation fibrosis and postablation scar were quantified and mapped from late gadolinium-enhanced CMR. CMR results were compared to the distribution and extent of PV reconnection. CMR and electrophysiological findings were compared between catheter types. Three months after successful PVI, scar gaps were found in 39 (76%) patients, and 78 (39%) veins. Electrical PV reconnection was detected in 45 (88%) patients, and 99 (50%) veins. The extent of PV reconnection related closely to the number of gaps (R = 0.55; P < .001), and to scar burden (R = -0.63; P < .001). However, the agreement was only fair for the localization of PV reconnection (k = 0.37; P < .001), scar gaps particularly lacking sensitivity in areas of pre-existing fibrosis. The circular catheter was associated with shorter procedures (P < .001), more scar (P = .01), less gaps (P = .01), and less reconnected veins (P = .03).

CONCLUSION: PV reconnection is extremely frequent after PVI. CMR scar imaging accurately predicts its extent, but poorly predicts its location. Multielectrode circular catheters induce more complete ablation.