10 Automated high density left atrial voltage mapping of low voltage areas in atrial fibrillation differs from sinus rhythm

IntroductionLow voltage areas (LVAs) are used as surrogate markers for diseased left atrial (LA) tissue and scar distribution. LVAs can be overestimated when measured in atrial fibrillation (AF) due to factors such as variable contact between mapping catheter electrodes and the endocardium in the presence of multiple depolarisation wavefronts with typically decreased cycle lengths. Scar or substrate modification was previously recommended for those with persistent AF (PeAF). Correlation between voltages in sinus rhythm (SR) and AF have previously been measured in studies via point-by-point analysis. An automated voltage histogram analysis (VHA) software module (CARTO3, Biosense Webster) has recently been validated for rapid area assessment of LA voltage maps and allows for very high-density analysis. We sought to compare LA LVA burden in SR to AF using VHA.MethodsWe included patients over age 18 with PeAF who underwent de novo pulmonary vein isolation (PVI) with no additional lines, maps with ≥1000 voltage points in each rhythm and a uniform procedure with initial mapping in AF then remapping in SR after PVI. We created 6 anatomical segments (AS) from each atrium for comparison in AF and SR including the Anterior, Posterior, Roof, Floor, Lateral and Septal AS (figure 1). We removed the mitral annulus, trans-septal puncture site and pulmonary veins for VHA analysis. The total area of each AS was categorised by voltage into 10 aliquots between 0-0.5mV, designating voltages ≤0.2mV as ‘Dense Scar’ and voltages 0.21-0.5mV as ‘Diseased Tissue’. Data attained from VHA outputs were analysed using SPSS v.26.ResultsUsing VHA we analysed 58,342 (n=10 patients) voltage points between all AS (N=120). While comparing corresponding AS in voltage ranges consistent with Dense Scar, ≤0.2mV, we found comparable LVA burden between rhythms in each AS except for the posterior wall, where there was significantly greater burden in AF (159.03 mm2 ± 194.65 Vs. 102.14 mm2 ± 157.47, p = 0.02). Utilising Spearman’s correlation test (table 1) for voltages ≤ 0.2mV we found significant correlation coefficients in the range of R=0.395-0.555 for each segment as displayed in figure 2. Assessing ranges consistent with Diseased Tissue, 0.21-0.5mV, we found greater LVA burden in all AS when measured in AF, and this reached significance in the Posterior (AF = 234 mm2 ± 150.45; SR = 138.27 mm2 ± 112.3, p=