Abstract 18796: Predictors of Repeat Ablation in DECAAF-II Patients

Introduction: Post-ablation arrhythmia recurrences are common, yet the incidence of Repeat Ablation (RA) in patients with Persistent Atrial Fibrillation (PeAF) is under-investigated and it remains unclear whether RA can be predicted.

Hypothesis: We looked for associations between certain patient/disease characteristics, and RA incidence.

Methods: Using the DECAAF-II trial database, we divided PeAF patients into 2 groups: with or without RA. We compared them for: demographics & comorbidities, baseline left atrial (LA) myopathy (reflected by LGE-MRI-detected LA fibrosis %, & LA volume), the duration in days between AF diagnosis & ablation, and ablation type (PVI Vs PVI + fibrosis-guided). The smart AF burden “Smurden” during the 3 months blanking period, defined by the % of days with detected AF over the total ECG monitoring duration, was also analyzed. Using a p-value cut-off of 0.3, variables were retained then included in a multivariate logistic regression model.

Results: 732 patients were included. Demographics, comorbidities, & other parameters are reported in Table1. Bivariate associations with p<0.3 were found: LA volume in mm ³ was significantly higher in patients with RA: 139.41 vs 129.54, p = 0.012; history of stroke and CABG were more common in RA with ORs respectively: 1.661, p = 0.111; 2.891, p = 0.096; RA incidence was higher in the PVI only group: OR = 0.77, p = 0.2; the Smurden was higher in RA patients: 33.68 % vs 12.55 %, p <0.001. The multivariate analysis showed one significant predictor of RA: Smurden OR: 9.413, CI 4.831- 18.343, p<0.001(Figure1).

Conclusions: High AF burden during the blanking period was a significant predictor of RA in PeAF patients.