Related Experiment Video
Updated: Jun 6, 2025
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
Left Atrial Wall Thickness Measured by a Machine Learning Method Predicts AF Recurrence After Pulmonary Vein
Daniel A Gomes1, Ana Rita Bello1, Pedro Freitas1,2
1Department of Cardiology, Hospital de Santa Cruz, Carnaxide, Lisbon, Portugal.
Background:
Left atrial (LA) remodeling plays a significant role in the progression of atrial fibrillation (AF). Although LA wall thickness (LAWT) has emerged as an indicator of structural remodeling, its impact on AF outcomes remains unclear. We aimed to determine the association between LAWT and AF recurrence after pulmonary vein isolation (PVI), as well as to evaluate the relationship between LAWT and LA fibrosis.
Methods:
Single-center registry of patients enrolled for radiofrequency PVI from 2016 to 2018. In all cases, a pre-ablation CT was performed within less than 48 h. Mean LAWT was retrospectively measured by a semi-automated machine learning method (ADAS 3D). A subgroup of patients also underwent pre-ablation cardiac MRI. The primary endpoint was time to AF recurrence after a 3-month blanking period.
Results:
A total of 439 patients (mean age 61 ± 12 years, 62% male, 78% with paroxysmal AF) were included. The mean LAWT was 1.4 ± 0.2 mm (0.9-1.9 mm). During a median follow-up of 5.8 (IQR: 4.9-6.6) years, 238 patients (54%) had an AF relapse. After adjusting for 8 clinical and imaging potential confounders, LAWT remained an independent predictor of time-to-recurrence (aHR: 4.25 [95% CI: 1.65-10.95], p = 0.003). AF recurrence rates were 11%, 15%, and 21%/year across terciles of increasing LAWT (log-rank p < 0.001). Additionally, the AF recurrence rate increased across the spectrum of LA structural remodeling, ranging from 8% (normal LAWT and LAVI) to 30%/year (LAWT and LAVI both increased). In the 62 patients who also underwent pre-ablation MRI, a moderate relationship between LAWT and fibrosis (assessed by late-gadolinium enhancement) was found (Spearman R 0.468; p < 0.001).
Conclusion:
Mean LAWT, easily assessed by commercially available machine learning software, is an independent predictor of time to AF recurrence after PVI in the long term. Whether patients with increased LAWT should receive tailored therapy deserves further investigation.
