P‐wave alternans rebound following pulmonary vein isolation predicts atrial arrhythmia recurrence

• Published in: Journal of cardiovascular electrophysiology Vol. 35; no. 7; pp. 1360 - 1367
• Main Authors: Nearing, Bruce D., Fialho, Guilherme L., Waks, Jonathan W., Maher, Timothy R., Clarke, John‑Ross, Shepherd, Alyssa J., D'Avila, Andre, Verrier, Richard L.
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.07.2024
• Subjects: Ablation; Arrhythmia; Artificial intelligence; atrial fibrillation; Cardiac arrhythmia; Cardiovascular diseases; Cardiovascular system; EKG; Medical records; modified moving average; Patients; pulmonary vein isolation; P‐wave alternans; R‐wave alternans; T‐wave alternans; R‐wave alternans; atrial fibrillation; T‐wave alternans; P‐wave alternans; modified moving average; pulmonary vein isolation; artificial intelligence
• ISSN: 1045-3873; 1540-8167; 1540-8167
• DOI: 10.1111/jce.16291

Introduction Numerous P‐wave indices have been explored as biomarkers to assess atrial fibrillation (AF) risk and the impact of therapy with variable success. Objective We investigated the utility of P‐wave alternans (PWA) to track the effects of pulmonary vein isolation (PVI) and to predict atrial arrhythmia recurrence. Methods This medical records study included patients who underwent PVI for AF ablation at our institution, along with 20 control subjects without AF or overt cardiovascular disease. PWA was assessed using novel artificial intelligence‐enabled modified moving average (AI‐MMA) algorithms. PWA was monitored from the 12‐lead ECG at ~1 h before and ~16 h after PVI (n = 45) and at the 4‐ to 17‐week clinically indicated follow‐up visit (n = 30). The arrhythmia follow‐up period was 955 ± 112 days. Results PVI acutely reduced PWA by 48%–63% (p < .05) to control ranges in leads II, III, aVF, the leads with the greatest sensitivity in monitoring PWA. Pre‐ablation PWA was ~6 µV and decreased to ~3 µV following ablation. Patients who exhibited a rebound in PWA to pre‐ablation levels at 4‐ to 17‐week follow‐up (p < .01) experienced recurrent atrial arrhythmias, whereas patients whose PWA remained reduced (p = .85) did not, resulting in a significant difference (p < .001) at follow‐up. The AUC for PWA's prediction of first recurrence of atrial arrhythmia was 0.81 (p < .01) with 88% sensitivity and 82% specificity. Kaplan–Meier analysis estimated atrial arrhythmia‐free survival (p < .01) with an adjusted hazard ratio of 3.4 (95% CI: 1.47–5.24, p < .02). Conclusion A rebound in PWA to pre‐ablation levels detected by AI‐MMA in the 12‐lead ECG at standard clinical follow‐up predicts atrial arrhythmia recurrence.