Abstract 4139988: Optimal Unipolar and Bipolar Electrode Separation for Recording Atrial Fibrillation in Patients

Background: Ablation is an increasingly preferred treatment for patients with atrial fibrillation (AF), yet the optimal catheter configuration to record electrical activity (electrograms) is undefined. It is unclear what electrode spacing includes local activity or remote colliding waves. This may affect the certainty that potentials near pulmonary vein or other targets are local, or truly eliminated by ablation, for different catheter designs. Hypothesis: We hypothesized that unipolar and bipolar electrograms in AF are impacted differently by electrode spacing, and that even adjacent bipolar electrograms may vary and reflect direction more than local activity. Methods: We studied 35k electrograms in N=22 patients referred for AF ablation (67±10 y.o, 3 female, 15 non-paroxysmal), collected with both grid-type (3 mm spacing; Abbott) and basket (6-8 mm spacing; Abbott) catheters. For each patient, we precisely measured separation in electroanatomic 3D shells from (x,y,z) coordinates for all electrode pairs (adjacent, next adjacent…) within and between catheters (N=12 patients had simultaneous grid/basket; fig. A). We compared similarity of AF electrograms for 1 to 20 mm distances using the Pearson correlation coefficient (CC), for unipolar and bipolar configurations, in 500ms cycle length windows. Results: Unipolar electrograms in AF were more similar over distance than bipoles. For unipoles, Fig B shows that AF electrograms had CC~0.8-0.9 for electrodes <2-3 mm apart, within a grid sizes, and CC>0.8 for spacing up to ~6 mm, within basket sizes. For distance > 6 mm, CC dropped in a semi-linear fashion with exponential increase in variability (p<0.05). For bipolar electrograms (fig. C), consistency was observed only partially even for very close electrodes (CC >0.8 for <2mm spacing), then fell rapidly with exponential variability (3-10mm). Fig C shows examples where even for relatively close bipoles, EGM traces showed different activation sequences (black arrows). Conclusions: Mapping in AF is very sensitive to unipolar or bipolar electrode configuration. Unipolar electrograms are similar for 2-6 mm separation, while bipolar electrograms vary for spacing >2 mm which may reflect direction more than local activity. Future studies should calibrate these findings against optical recordings of wave propagation or results from prospective guided ablation.