Automatic Echocardiographic Assessment of Left Atrial Function for Prediction of Low-Voltage Areas in Non-Valvular Atrial Fibrillation.

Purpose: Left atrial low-voltage areas (LA-LVAs) identified by 3D-electroanatomical mapping are crucial for determining treatment strategies and prognosis in patients with atrial fibrillation (AF). However, convenient and accurate prediction of LA-LVAs remains challenging. This study aimed to assess the viability of utilizing automatically obtained echocardiographic parameters to predict the presence of LA-LVAs in patients with non-valvular atrial fibrillation (NVAF).

Patients And Methods: This retrospective study included 190 NVAF patients who underwent initial catheter ablation. Before ablation, echocardiographic data were obtained, left atrial volume and strain were automatically calculated using advanced software (Dynamic-HeartModel and AutoStrain). Electroanatomic mapping (EAM) was also performed. Results were compared between patients with LA-LVAs ≥5% (LVAs group) and <5% (non-LVAs group).

Results: LA-LVAs were observed in 81 patients (42.6%), with a significantly higher incidence in those with persistent AF than paroxysmal AF (55.6% vs 19.3%, 0.001). Compared with the non-LVAs group, the LVAs group included significantly older patients, lower left ventricular ejection fraction, higher heart rate, and higher E/e' ratio ( <0.05). The LVAs group exhibited higher left atrial volume index (LAVi) and lower left atrial reservoir strain (LASr) ( <0.001). In multivariate analysis, both LAVi and LASr emerged as independent indicators of LVAs (OR 0.85; 95% CI 0.80-0.90, <0.001) and (OR 1.15, 95% CI 1.02-1.29, =0.021). ROC analysis demonstrated good predictive capacity for LA-LVAs, with an AUC of 0.733 (95% CI 0.650-0.794, <0.001) for LAVi and 0.839 (95% CI 0.779-0.898, <0.001) for LASr.

Conclusion: Automatic assessment of LAVi and LASr presents a promising non-invasive modality for predicting the presence of LA-LVAs and evaluating significant atrial remodeling in NVAF patients. This approach holds potential for aiding in risk stratification and treatment decision-making, ultimately improving clinical outcomes in patients.