Various predictors of atrial fibrillation (AF) recurrence have been shown based on the baseline characteristics before catheter ablation (CA). This study aimed to develop a novel scoring system for predicting very late recurrences of AF (VLRAFs) after an initial CA, taking the postprocedural clinical data into account and reassessing VLRAFs in 12-month patients' condition using previously known preprocedural predictors of AF recurrences. We retrospectively studied 327 patients who underwent an initial CA with freedom from AF for over 12 months. We elucidated the predictors of VLRAFs and created a new score to predict VLRAFs in the discovery AF cohort (n = 181). Thereafter, we investigated whether the new scoring system could accurately predict VLRAFs in the validation AF cohort (n = 146). In the discovery AF cohort, VLRAFs were observed in 53 patients (29%) during the follow-up period (mean follow-up duration: 55 months). The univariate and multivariate Cox proportional-hazards model demonstrated that non-pulmonary vein foci, early recurrences of AF (ERAFs), atrial premature contraction (APC) burden ≥ 142/24 hours, and minimum prematurity index of the APCs ≤ 48% were associated with VLRAFs after CA. We created a new scoring system to predict VLRAFs, the n-PReDCt score (non-pulmonary vein: 1 point, early recurrences of AFs (Recurrences of AF in early phase after CA): 1 point, APC burden ≥ 142/24 hours: 1 point, and minimum prematurity index (= Coupling interval) of the APCs of ≤ 48%: 1 point). The n-PReDCt score was significantly associated with VLRAFs by a Kaplan-Meier analysis in the discovery AF and validation AF cohorts (p < 0.0001 and p < 0.0001, respectively).
Copyright © 2020 Elsevier Inc. All rights reserved.