AF‐FLOW Global Registry Confirms Validity of Electrographic Flow Mapping as a Phenotyping Tool for Atrial Fibrillation

Figures

• 

  Figure 1. Post‐PVI EGF maps and accompanying electroanatomic maps. Mapping in (A) SVC/RA junction (RA position 1) and (B) posterosuperior LA (LA position 1) are shown. Note that the RA map has a larger surface area with low EGFC (seen as shorter vectors in blue), resulting in an EGFC for the map of 0.51, while the LA map has more areas containing high EGFC (seen as longer vectors in purple), resulting in an EGFC for the map of 1.21. An active source can be seen in the RA at E7‐F8 (crosshairs) with a SAC of 25%, while the LA source at C5‐C6 has 37% SAC.

• 

  Figure 2. Near‐field score calculation. (Top) An EGF summary map with grey squares representing electrodes with near‐field scores < 0.7. Near‐field score was calculated as the percentage of segments with cross‐correlation > 0 with at least one of the adjacent channels at a time shift between 5 and 50 ms. Accordingly, grey electrodes had cross‐correlation with adjacent channels in < 70% of the segments. Because a low near‐field score was a proxy for poor atrial contact, these electrodes were not considered in EGFC calculations from maps. (Bottom) The first 2 s of the recording from channels in splines D–H. Channels corresponding to grey electrodes, including the eighth electrode in all splines, did not have depolarizations correlating to those in any neighboring channels at a time offset between 5 and 50 ms, and therefore had a near‐field score < 0.7.

• 

  Figure 3. Anatomic distribution of EGF‐identified sources. Source locations are shown in red on a schematic of the left and right atria. AP, anteroposterior; IVC, inferior vena cava; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; PA, posteroanterior; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein; SVC, superior vena cava.

• 

  Figure 4. FFAF by source presence and EGFC. Patients with zero sources were considered to have no sources, while patients with one or more sources were considered to have sources. Most but not all sources were ablated during the procedure. The median EGFC across all 25 patients was 0.548, so was defined as the dividing line for high vs. low EGFC in this study. FFAF patients are shown in blue, and recurrence patients are shown in red; PAF is shown as triangles, PeAF is shown as squares, and LS‐PeAF is shown as diamonds. FFAF goes down with progressively increasing phenotype from Type I to Type IV.

• 

  Figure 5. Local EGFC alignment with bipolar voltage. EGF maps and accompanying voltage maps from a single patient in AF are shown in the (A) RA and (B) LA. Areas of high EGFC (magenta) correspond to areas of high voltage (purple), while areas of low EGFC (blue) correspond to areas of low voltage (red). In the RA, all splines pass predominantly through high voltage except for electrodes 48–58 (D8‐E8 on EGF map). In the LA, all splines pass predominantly through medium and low voltage except for electrodes 72–83 (F2‐G3 on EGF map).

• 

  Figure 6. Global EGFC correlation to healthy bipolar voltage. Scatter plot of percent healthy voltage surface area vs. atrial EGFC from concomitant voltage and EGF maps. A best‐fit trendline shows a positive correlation (r = 0.651, p < 0.001).