• predictors af ablation

Introduction:
Atrial fibrillation ablation is a treatment of choice for maintaining sinus rhythm in patients affected by this arrhythmia, which has proven to be superior to pharmacological therapy. However, this procedure is burdened by a significant recurrence rate. Therefore, it is important to select patients and guide them through a dedicated post-procedural path using predictors of recurrence, the predefined goal of the ATHENA study, a prospective multicenter observational study.

Study Purpose:
The subject of this thesis is a prospective analysis of 150 cases enrolled in the ATHENA study who completed a one-year follow-up after the ablation procedure, including 90 cases of paroxysmal atrial fibrillation (60%) and 60 cases of persistent atrial fibrillation (40%). Potential risk predictors were analyzed using multivariate Cox regression, and Kaplan-Meier curves were used to compare paroxysmal and persistent patients to investigate differences in predictor performance between these two populations.

Results:
Recurrence at one year was observed in 33 (23%) of the total population, with 14 (17%) in paroxysmal cases and 19 (33%) in persistent cases. A comparison between patients with and without recurrences highlighted significantly different data, such as the indexed left atrial volume (47±11 vs. 36±11, p<0.005), which was higher in patients with recurrences. Structural heart disease was more prevalent in patients with recurrences (39% vs. 20%, p=0.027), and the incidence of first-pass isolation was less frequent in patients with recurrences (44% vs. 69%, p=0.022). Multivariate Cox regression demonstrated that independent predictive factors were the indexed left atrial volume (5.23 CI 2.06-13.31, p=0.001) with a cutoff of 42.5 ml/m2, identified as the most effective by ROC analysis in distinguishing between patients with and without recurrences, associated with a higher risk of recurrence, and first-pass isolation in all pulmonary veins (0.40 CI 0.18-0.89, p=0.024), associated with a lower risk of recurrence. Additionally, the presence of atrial arrhythmic events during the first three months after the procedure, a timeframe currently not considered in clinical practice and studies, was evaluated as a predictor of recurrence. The presence of early recurrences was associated with an increased risk of developing true recurrences in the subsequent period (HR 6.6, CI 3.3 – 13.1, p<0.005).
A comparative analysis was then performed between paroxysmal and persistent patients and their recurrences, revealing that patients with persistent atrial fibrillation had a significantly higher number of recurrences. Furthermore, when comparing Kaplan-Meier curves for the studied recurrence predictors, the curves effectively separated for both paroxysmal and persistent patients when considering the indexed left atrial volume, while first-pass isolation did not effectively differentiate recurrences in patients with persistent AF.

Conclusions:
The indexed left atrial volume with a cutoff of 42.5 and the absence of first-pass isolation are effective predictors of post-procedural arrhythmia recurrence risk. Atrial arrhythmias in the blanking period may be important in predicting recurrences in the subsequent period.