Left atrial spatial entropy: a novel tool for electrophysiological substrate characterization in atrial fibrillation

Background: Electrical remodeling has been linked to the progression and recurrence of atrial fibrillation (AF) after catheter ablation (CA). Substrate mapping based solely on a voltage amplitude electrogram (EGM) does not provide a comprehensive understanding of the left atrial (LA) disease. The aim of this study is to assess left atrial spatial entropy (LASE) from voltage maps routinely obtained during AF ablation to further characterize the LA substrate. Materials and Methods: High-density electroanatomic maps (EAMs) of 27 patients with paroxysmal or persistent AF undergoing routine CA were prospectively collected. Computational post-processing was performed on the voltage maps. Using the Shannon entropy model, the probability distribution of the amplitude range values associated with each point of the map was used to measure LASE. Finally, correlations between LASE and clinical and electrophysiological characteristics of AF were explored. Results: LASE differentiated between patients with paroxysmal and persistent AF (6.45 ± 0.41 vs. 5.87 ± 0.53; p = 0.028) and patients with normal and abnormal LA substrate (6.42 ± 0.42 vs. 5.87 ± 0.56; p = 0.043), independent of the basal rhythm during EM acquisition (6.33 ± 0.41 vs. 6.11 ± 0.63; p = 0.619). Accordance between LASE and EAMs was assessed by ROC analysis (AUC: 0.81; C.I.: 0.62–0.99; Youden index: 6.06; sensitivity: 80%; and specificity: 80%). Patients with the lowest LASE reported AF recurrence at the follow-up. Conclusion: LASE may play a role in the further characterization of the LA substrate and the type of AF, independent of basal rhythm.