Association of left atrial function with incident atypical atrial flutter after atrial fibrillation ablation.

BACKGROUND: Symptomatic left atrial (LA) flutter (LAFL) is common after atrial fibrillation (AF) ablation. OBJECTIVE: The purpose of this study was to examine the association of baseline LA function with incident LAFL after AF ablation. METHODS: The source cohort included 216 patients with cardiac magnetic resonance (CMR) before initial AF ablation between 2010 and 2013. Patients who underwent cryoballoon or laser ablation, patients with AF during CMR, and those with suboptimal CMR, or missing follow-up data were excluded. Baseline LA volume and function were assessed by feature-tracking CMR analysis. RESULTS: The final cohort included 119 patients (mean age 58.9 ± 11 years; 76.5% men; 70.6% patients with paroxysmal AF). During a median follow-up of 421 days (interquartile range 235-751 days), 22 patients (18.5%) had incident LAFL. Baseline LA volume was similar between the 2 groups. In contrast, baseline reservoir, conduit, and contractile function of the LA were significantly impaired in patients with incident LAFL. Baseline global peak longitudinal atrial strain (PLAS) <22.65% predicted incident LAFL with 86% sensitivity and 68% specificity (C statistic 0.76). In a multivariable model adjusting for age, heart failure, and LA volume, PLAS (hazard ratio 0.9 per % increase in PLAS; P = .003) and LA linear lesions (hazard ratio 2.94; P = .020) were independently associated with incident LAFL. The coexistence of PLAS <22.65% and linear lesions was associated with 9-fold increased hazard of incident LAFL. CONCLUSION: Baseline LA function and linear lesions were independently associated with incident LAFL after AF ablation. Linear lesions should be limited to selected cases, especially in patients with impaired LA function.

Phrenic nerve injury: An underrecognized and potentially preventable complication of pulmonary vein isolation using a wide-area circumferential ablation approach.

INTRODUCTION: Phrenic nerve injury (PNI) is a well-known, although uncommon, complication of pulmonary vein isolation (PVI) using radiofrequency energy. Currently, there is no consensus about how to avoid or minimize this injury. The purpose of this study was to determine how often the phrenic nerve, as identified using a high-output pacing, lies along the ablation trajectory of a wide-area circumferential lesion set. We also sought to determine if PVI can be achieved without phrenic nerve injury by modifying the ablation lesion set so as to avoid those areas where phrenic nerve capture (PNC) is observed. METHODS AND RESULTS: We prospectively enrolled 100 consecutive patients (age 61.7 ± 9.2 years old, 75 men) who underwent RF PVI using a wide-area circumferential ablation approach. A high-output (20 mA at 2 milliseconds) endocardial pacing protocol was performed around the right pulmonary veins and the carina where a usual ablation lesion set would be made. A total of 30% of patients had PNC and required modification of ablation lines. In the group of patients with PNC, the carina was the most common site of capture (85%) followed by anterior right superior pulmonary vein (RSPV) (70%) and anterior right inferior pulmonary vein (RIPV) (30%). A total of 25% of PNC group had capture in all 3 (RSPV, RIPV, and carina) regions. There was no difference in the clinical characteristics between the groups with and without PNC. RF PVI caused no PNI in either group. CONCLUSION: High output pacing around the right pulmonary veins and the carina reveals that the phrenic nerve lies along a wide-area circumferential ablation trajectory in 30% of patients. Modification of ablation lines to avoid these sites may prevent phrenic nerve injury during RF PVI.