Evaluation of esophageal injury after very high-power short-duration pulmonary vein isolation using late gadolinium enhancement magnetic resonance imaging

Le résumé fourni par la source

Background Esophageal injury is a recognized complication of radiofrequency (RF) pulmonary vein isolation (PVI). Very high-power, short-duration (vHPSD) ablation creates shallower lesions, potentially limiting esophageal damage. Objective This study evaluated esophageal injury using 3D LGE-CMR post-PVI in AF patients undergoing conventional or vHPSD RF ablation. Methods Patients with AF (n=59, 61±8yr) underwent primo RF PVI with ablation index (AI)-guided (30/40W,n=19) or vHPSD (90W/4s,n=40) ablation. 3D LGE-CMR was performed at baseline, <72h, and 3 months post-PVI. Esophageal signal intensity (SI) was measured in anterior and posterior wall at the mid–LA level in sagittal and axial planes, normalized to a reference to calculate SI ratio (SIR ESO ). Esophageal LGE was analyzed over time, by ablation type, esophageal position, and, in vHPSD, by posterior wall reconnection requiring touch-up ablation. Results Anterior SIR ESO ​ increased significantly <72 hours post-PVI and, although partially reduced at 3 months, remained above baseline (1.06[0.95–1.18] to 1.36[1.13–1.57] to 1.21[1.03–1.40], all p<0.01). Patients with the esophagus adjacent to a PV (n=48) had higher anterior SIR ESO at both early and 3-month follow-up than those with inter-PV position (n=11). Partial SIR ESO reduction at 3 months did not occur in the AI group. Within the vHPSD group (n=40), posterior touch-up ablation (n=13) was associated with persistently higher anterior SIR ESO at 3 months. Posterior SIR ESO remained unchanged over time. Conclusion RF PVI increased esophageal LGE early post-ablation, persisting at 3 months. Partial resolution occurred with vHPSD, but LGE remained elevated in both ablation types, especially when the esophagus was adjacent to a PV or after acute posterior wall reconnection requiring touch-up ablation.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.