Abstract Body (Do not enter title and authors here): Background:
In catheter ablation for atrial fibrillation (AF), the efficacy of adjunctive ablation strategies beyond pulmonary vein isolation remains uncertain. Although complex atrial electrograms have been proposed as potential targets, conventional identification methods of these electrograms are subjective and lack reproducibility. A novel tool, Complex Signal Identification (CSI), has recently become available to automatically detect and annotate complex atrial electrograms.

Hypothesis:
We hypothesized that the CSI algorithm can objectively identify electrograms associated with atrial remodeling and is associated with post-ablation outcomes in patients with AF.

Methods:
We analyzed 47 consecutive patients (mean age: 66 years; 71% male) who underwent first-time AF ablation using the CARTO3^® system between October 2024 and January 2025. All patients underwent pulmonary vein isolation by radiofrequency ablation, followed by high-resolution mapping of the entire left atrium using an OCTARAY^® catheter under coronary sinus pacing. CSI-positive areas were quantified across the entire left atrium and within six predefined anatomical regions (Figure 1). A CSI-positive area was defined as any region containing at least one CSI-positive point. Structural remodeling was assessed by identifying low-voltage areas (< 0.5 mV) and evaluating their overlap with CSI-positive regions. Electrical remodeling was assessed by calculating conduction velocity based on activation time and the distance between earliest and latest sites of the left atrium.

Results:
A total of 103 CSI-positive areas were identified across all patients, within the anterior and septum being the most frequently involved regions (Figure 2). Twelve low-voltage areas were found, and all of these areas overlapped with CSI-positive regions. The total number of CSI-positive points per patients inversely correlated with conduction velocity (R = –0.448, P = 0.001) (Figure 3). When patients were stratified based on the presence or absence of CSI-positive area, the early recurrence rate of atrial fibrillation was significantly higher in the CSI-positive group (n = 28) than in the CSI-negative group (n =19) (21% vs. 0%, P < 0.05).

Conclusion:
CSI identified abnormal electrograms associated with impaired conduction and early AF recurrence. CSI-positive regions extended beyond low-voltage areas, suggesting that CSI may provide a more comprehensive assessment of electrical remodeling than voltage mapping.