Relationship Between Complex Fractionated Electrograms (CFE) and Dominant Frequency (DF) Sites and Prospective Assessment of Adding DF-Guided Ablation to Pulmonary Vein Isolation in Persistent Atrial Fibrillation (AF)

BACKGROUND: Sites of high DF are potential targets for AF ablation, but it is unknown if addition of DF ablation can improve procedural outcome. OBJECTIVES: We sought to (1) examine the relationship between DF sites and complex fractionated electrograms (CFE) and (2) prospectively assess the long-term outcome of adding DF ablation to pulmonary vein antral isolation (PVAI) for persistent AF. METHODS: First, 20 patients with persistent AF who underwent previous CFE-guided ablation and who had AF terminate during ablation were studied retrospectively (group I). Bipolar, 8-second electrograms were collected by a circular catheter (288 ± 86 points/map). The EnSite NavX system allows for automated display of both CFE and DF maps. Electrograms with cycle length <120 ms were considered CFE and were compared to DF sites > 8 Hz (direct inverse relationship). Sites of AF termination were related to CFE and DF sites. Based on these observations, 30 different patients (group II) with persistent AF prospectively underwent DF-guided ablation plus PVAI. They were followed every 3 months for 1 year (visit, Holter, ECG). These patients were compared to case-matched controls undergoing PVAI alone (group III). RESULTS: In group I, there was a significant, inverse correlation between DF and CFE values at each point (r =-0.24, P < 0.001). DF surface area was less than CFE area (27 ± 5 cm(2) vs 34 ± 4 cm(2) , P = 0.03). CFE sites overlapped 48 ± 27% with the DF surface area. Nonoverlapping CFE sites were contiguous to DF sites. AF termination occurred where DF and CFE overlapped, and at these sites, DF was always greater than the mean DF for the map. In group II, all DF sites above the mean value were prospectively ablated during AF. AF termination was noted in only 2/30 (7%) patients. After DF ablation, PVAI was performed and termination increased to 4/30 patients (14%). At 1 year, freedom from atrial arrhythmia > 30 seconds occurred in 57% of DF+PVAI compared to 60% in patients receiving PVAI alone (P = 0.18). CONCLUSIONS: DF and CFE regions overlap only about 50%. AF termination retrospectively occurred on overlapping CFE/DF sites where DF was above the mean. However, prospective ablation of DF sites plus PVAI resulted in low AF termination rates, and did not improve 1 year success over PVAI alone.