The Vein of Marshall Promotes Re-Entry After Pulmonary Vein Reconnection.

BACKGROUND

Postablation re-entrant atrial tachycardia recurrence remains a major limitation of atrial fibrillation (AF) ablation. Pulmonary vein (PV) reconnection and epicardial connections from the left atrium (LA) to the vein of Marshall (VoM) and coronary sinus (CS) in proximity to mitral valve are important contributors. We hypothesized that VoM facilitates arrhythmia induction through interaction with the epicardial region encircled by wide antral circumferential ablation (WACA).

OBJECTIVES

The objective of this study was to systematically evaluate the combined effect of PV reconnection and epicardial VoM connectivity on the occurrence of sustained postablation re-entrant atrial tachycardia using in silico models.

METHODS

Bilayer biatrial models were constructed from electroanatomical and computed tomographic data of 10 patients with AF. Post-AF electrical remodeling and substrate remodeling following the low voltage area distribution, and left-sided WACA were incorporated. Segments of WACA lesions were sequentially restored to simulate PV reconnection. Three connectivity scenarios were evaluated: intact CS-VoM-LA epicardial connection, CS-LA connection only, and no epicardial connections. Sustained re-entry (>10 seconds) was the primary endpoint, and connectivity conditions were compared using patient-level paired vulnerability. Induction of any arrhythmia, tachycardia cycle length, and stable rotor number were secondary outcomes.

RESULTS

Sustained re-entry occurred at 92 of 294 (31.3%) reconnection sites with intact CS-VoM-LA connectivity, 37 of 294 (12.6%) with CS-LA connectivity only, and 29 of 294 (9.9%) without epicardial connectivity. Patient-level paired analysis showed 19.9-percentage-point and 22.7-percentage-point higher vulnerability with intact CS-VoM-LA connectivity compared with CS-LA connectivity alone (95% CI: 7.1-37.3 percentage points) and no epicardial connectivity (95% CI: 7.4-41.3 percentage points), respectively. Mean tachycardia cycle length in the LA was 258.0 ± 25.7 milliseconds, 255.0 ± 31.4 milliseconds, and 251.0 ± 33.9 milliseconds across the 3 scenarios; the average number of stable rotors was 0.6 ± 0.3, 0.7 ± 0.2, and 0.4 ± 0.2, respectively.

CONCLUSIONS

The VoM is a critical epicardial driver of arrhythmia after incomplete WACA, underscoring the importance of its targeted ablation.