Impact of iNAV and dNAV 3D LGE-CMR on the assessment of atrial fibrosis in atrial fibrillation patients

Abstract

Abstract Background Late gadolinium enhanced (LGE) cardiac magnetic resonance (CMR) imaging serves as a valuable non-invasive tool for visualizing and quantifying left atrial (LA) fibrosis in atrial fibrillation (AF) patients. The extent of pre-procedural LA-LGE, indicative of native atrial fibrosis, has proven to be a predictor of AF recurrence risk following AF ablation. Consequently, assessing the LA fibrotic burden has emerged as a potential guide for personalized patient management. Currently, most centers engaged in LA-LGE image acquisition apply a scan protocol based on a diaphragmatic navigator-gated 3D image strategy (dNAV). Recently introduced image navigated (iNAV) 3D LGE strategies are proposed to outperform the dNAV strategy. This advancement involves the dynamic tracking of the heart's respiratory position with an image reconstruction algorithm, yielding motion-compensated images in significantly reduced and more consistent scan durations. We performed an evaluation of the novel image navigated (iNAV) 3D LGE-CMR imaging strategy in comparison to the conventional diaphragm navigated (dNAV) 3D LGE-CMR strategy. Methods In twenty-six consecutive AF patients, both imaging techniques (i.e. iNAV and dNAV) were performed subsequently, with equivalent spatial resolution and timing in the cardiac cycle. Patients were randomized in the acquisition order of iNAV and dNAV. LA fibrosis was quantified (percentage of atrial fibrosis using image intensity ratio threshold 1.2) and overlap in atrial fibrosis areas was tested between the two methods. Results Acquisition time of iNAV was significantly lower compared to dNAV (5±1 minutes vs. 12±4 minutes, p<0.001, respectively). There was a significant correlation between the iNAV and dNAV LA fibrotic burden (r=0.69, p<0.001). LA fibrosis scores were lower for iNAV compared to dNAV (12±8% vs. 20±12%, p<0.001, respectively). Spatial correspondence between the atrial fibrosis maps was modest (Dice similarity coefficient 0.43±0.15). Notably, 13/23 (56%) patients underwent a shift in their UTAH fibrosis stage classification, depending on the chosen imaging strategy. Conclusion iNAV acquisition was more than twice as fast as dNAV, and iNAV resulted in a lower atrial fibrosis score as compared to dNAV. More than half of the patients were reclassified into a different UTAH fibrosis stage depending on the employed imaging strategy. This underscores the critical importance of recognizing that the choice of imaging strategy significantly impacts the categorization of patients into their respective fibrosis stages. This distinction carries potential clinical implications when evaluating the utility of UTAH fibrosis stages in the context of AF management and therapeutic decision-making, as patients in lower fibrosis stages are generally regarded as more suitable candidates for ablation, while those categorized in higher fibrosis stages face an increased risk of arrhythmia recurrence following PVI.Figure 1Figure 2