Abstract
Atrial fibrillation (AF) is characterized by significant atrial remodeling, including cellular pathology, left atrial (LA) enlargement, and reduced electrical voltage. While invasive electroanatomic mapping (EAM) is the standard for identifying these changes, non-invasive late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) has emerged as a potential alternative. This study investigates the correlation between invasive EAM voltages and non-invasive LGE-CMR findings in patients with AF.
Introduction
The management of atrial fibrillation (AF) often involves identifying areas of diseased atrial tissue. Traditionally, this is achieved via invasive electroanatomic mapping (EAM), where catheters measure local electrical potentials (voltage) within the heart. Areas of low voltage typically correspond to structural remodeling and fibrosis. Recent advancements in imaging have introduced Late Gadolinium Enhancement (LGE) CMR as a non-invasive method to visualize atrial scarring. However, the quantitative relationship between these structural MRI findings and the electrical behavior of the heart requires further validation. This study aims to bridge that gap by correlating EAM voltages directly with LGE-CMR data.
Methodology
The study cohort included 16 patients with AF scheduled for their first pulmonary vein isolation (PVI) procedure. The experimental workflow consisted of:
- LGE-CMR Imaging: High-resolution 1.5T MRI scans were performed 10–20 minutes post-gadolinium administration to highlight fibrotic tissue.
- Electroanatomic Mapping: Invasive EAM was conducted using the CARTO system to generate a point-by-point voltage map of the left atrium.
- Image Registration: MRI-derived fibrosis maps were spatially aligned with EAM electrical maps using the pulmonary vein ostia as anatomical landmarks.
- Comparative Analysis: Mean bipolar voltages were compared between regions with and without LGE. A threshold of <0.38 mV was used to define "low voltage" zones.
Results
The analysis revealed a significant association between structural fibrosis and electrical depression. In regions identified as fibrotic via LGE-CMR, the mean bipolar voltage was significantly lower (0.29 ± 0.17 mV) compared to non-LGE regions (0.51 ± 0.25 mV, p < 0.001).
While a positive relationship was observed between the total area of low voltage and the extent of LGE, the correlation was modest ($R^2 = 0.18$). Additionally, patient age was found to be a significant predictor of remodeling; patients with extensive LGE (>6% of the atrial area) were significantly older (63 ± 7 years) than those with less extensive scarring (51 ± 12 years, p = 0.03).
Discussion and Limitations
These findings confirm that LGE-CMR can serve as a non-invasive surrogate for identifying electrical dysfunction in the left atrium. The ability to localize "low-voltage" zones prior to surgery could significantly improve procedure planning.
However, several limitations persist. The modest correlation ($R^2 = 0.18$) likely stems from the resolution mismatch between millimeter-scale MRI and point-by-point catheter contact. Furthermore, the small sample size (n=16) necessitates larger studies to determine if LGE-CMR findings can accurately predict clinical outcomes, such as AF recurrence post-ablation. Accuracy in spatial registration also remains a challenge due to cardiac motion.
Conclusion
This study validates the use of LGE-CMR as a tool for identifying structural and electrical remodeling in AF patients. By providing a non-invasive map of diseased tissue, LGE-CMR offers a promising pathway toward more efficient, MRI-guided ablation strategies.
Summary updated