37 * Optimized local activation time sampling density for left atrial electroanatomic mapping

Abstract

Introduction: Local activation time (LAT) mapping using electroanatomic mapping systems forms the cornerstone of diagnosis of complex atrial arrhythmias. Whilst new and emerging technologies exist to create higher-density LAT maps the end goal of LAT mapping (to identify the tachycardia mechanism and appropriate ablation site) remains unchanged. Here we identify the optimum point density for a variety of atrial tachycardia mechanisms, based on systematic analysis of in silico and in vivo LAT mapping data. Methods: Optimal sampling density for atrial tachyarrhythmias was determined in silico and in vivo. A variety of tachycardia mechanisms (focal, linear, re-entrant and spiral wave) were re-created on a 4×4 cm atrial monolayer model. Resulting isochronal activation maps were re-sampled at random locations with point-densities ranging from 0.25 to 10 points/cm2 and ‘downsampled' isochronal maps were re-interpolated across the monolayer. Error in the downsampled activation maps was quantified as the sum of squared differences between the original and re-interpolated LAT map activation times. Optimal sampling density was thereafter defined as the lowest point sampling density that resulted in an error reduction rate of less than 0.05 ms per mesh node per additional point sampled. Once validated, this technique was applied in vivo using high density left atrial maps (n=6) collected using point-by-point sampling (original sampling density 2.5±0.8 points/cm2). Result: Figure A shows the method applied to a simulated focal tachycardia with optimal sampling density identified at the point where the error reduction rate curve crosses the 0.05ms/node/point line. In silico, the optimum sampling density for focal, spiral-wave and re-entrant activation patterns was 1.2, 1.6 and 1.7 points/cm2. Linear activation in a 2-dimensional domain represents a special case where map accuracy is independent of sampling density. Figure B shows the same method applied to a clinical focal tachycardia. In vivo, the optimum sampling density for focal, macro-re-entry and localized re-entry was 0.9-1.0, 0.9-1.6 and 1.1-1.7 points/cm2. Conclusion: Optimal sampling densities can be identified for a variety of tachycardia mechanisms to maximize the diagnostic yield of interpolated LAT maps. Greater sampling density is required to correctly reveal more complex activation patterns (e.g. macro-re-entry and spiral-wave activation). For both focal and re-entrant left atrial tachycardias there appears to be no additional benefit in performing LAT sampling at a density greater than 1.7points/cm2. ![Graphic][1] [1]: /embed/inline-graphic-1.gif