Left atrial epicardial adipose tissue volume quantification by CMR aids identification of patients at low risk for left atrial cardiomyopathy.

Background And Aims: The pathophysiologic concept of atrial fibrillation (AF) has evolved towards defining atrial cardiomyopathy, recognizing inflammation-mediated remodeling of the left atrium (LA) as a source for arrhythmogenesis. One feature of atrial cardiomyopathy is the development of fibrosis, with low-voltage zones (LVZ) identified by invasive electroanatomic mapping as an accepted surrogate parameter. A mediator of pathological remodeling is epicardial adipose tissue (EAT). This study sought to explore LA-EAT volume, as a predictor of LVZ in patients presenting for primary AF ablation.

Methods And Results: CMR imaging of left atrial epicardial adipose tissue was performed using fat-water separation Dixon-based sequences in 58 patients (mean age 68.2 ± 10.1 years) presenting for primary pulmonary vein isolation (PVI). Additionally, left atrial volume index (LAVi) was derived from contrast-enhanced angiography. Left atrial epicardial volume index (LA-EATVi) was a significant predictor of LVZ, with significantly higher volumes in LVZ + patients (mean difference of 7.2 ± 2.4 ml/m), a moderate correlation (r = 0.37, p < 0.001), and a univariate predictive ability with an area under the curve (AUC) of 0.71. Expanding the prediction model with age, gender, and LAVi improves the prediction of LVZ up to an AUC of 0.91. Cutoff selection at 0.25 predicted probability identifies a low-risk group for LVZ, with a negative predictive value of 96.7%, sensitivity 95%, and specificity 76.3%.

Conclusion: The pre-procedural identification of a low risk of atrial cardiomyopathy is important to select patients for single-shot catheter ablation. LA-EATVi proved to be of additive value to known risk factors for the prediction of LVZ in a combined prediction model. Patients not considered low risk could be offered an electroanatomic atrial mapping for LVZ detection with the possibility of an LVZ-based radiofrequency ablation approach. Individualized matching of patient and ablation technique using an LVZ prediction model might lead to improved ablation outcomes.