Background: Electro-anatomical voltage, conduction velocity (CV) mapping and late gadolinium enhancement magnetic resonance imaging (LGE-MRI) are different diagnostic modalities for atrial cardiomyopathy (ACM). However, discordances remain in the location and extent of detected ACM.
Objectives: (1) Comparison of ACM extent and location between current modalities. (2) Development of new estimated optimised image intensity thresholds (EOIIT) for LGE-MRI identifying patients with ACM.
Methods: Thirty-six ablation-naive persistent AF patients underwent LGE-MRI and high-definition electro-anatomical mapping in sinus rhythm. Significant ACM was defined as low voltage substrate (LVS) extent ≥ 5% of the left atrium (LA) surface at < 0.5mV. LGE areas were classified using the Utah, image intensity ratio (IIR > 1.20) and new EOIIT method for comparison to LVS and slow conduction areas < 0.2m/s. ROC analysis determined the LGE-extent enabling accurate diagnosis of ACM.
Results: The degree and distribution of detected pathological substrate varied significantly (p < 0.001) across the mapping modalities: 3% (IQR 0-12%) of the LA displayed LVS < 0.5mV vs. 14% (3-25%) slow conduction areas < 0.2m/s vs. 16% (6-32%) LGE with Utah method vs. 17%(11-24%) using IIR > 1.20, with enhanced discrepancies on posterior LA. A linear correlation was found between the OIIT and each patients mean blood pool intensity (R2=0.89, p < 0.001). LGEMRI-based ACM diagnosis improved with the novel EOIIT (83% sensitivity, 88% specificity, AUC:0.94) in comparison to the Utah method (60% sensitivity, 75% specificity, AUC:0.76), and IIR > 1.20 (58% sensitivity, 75% specificity, AUC:0.71). Conclusion: Important discordances in distribution of pathological substrate exist between LA-LVS, CV and LGE-MRI, irrespective of the LGE-detection protocol that is used. However, the new EOIIT method improves LGE-MRI based ACM diagnosis in ablation-naive AF-patients.