Background: Recent studies have suggested pulmonary vein isolation (PVI) alone is as effective as strategies incorporating additional substrate modification in patients with persistent atrial fibrillation (AF). It remains unclear if PVI impacts upon the mechanisms that sustain or remove triggers initiating AF. Non-invasive mapping (ECGI) is able to panoramically map both atria simultaneously and identifies rotational activation patterns which may act as potential drivers (PDs) in persistent AF. We sought to determine if the burden and distribution of PDs are reduced by PVI and if baseline maps predict the acute response to ablation.
Methods: Patients with persistent AF <2 years were recruited. Patients underwent a CT scan followed by panoramic mapping of the atria using the ECGI system. Multiple atrial segments (activity between the end of the T wave and subsequent QRS) were recorded to compile 15 seconds of atrial activity for analysis. PVI was performed using the cryoballoon. Pre- and post-PVI ECGI recording were performed intra-procedurally. PDs were defined as wavefronts with rotational activity completing at least 1.5 revolutions or focal activations. Distribution of PDs was recorded using an 18-segment model of both atria.
Results: 100 patients were enrolled (61.3 ± 12.1 years) with median AF
duration of 8 (5–15) months. Median LA diameter was 39 (33–43) mm. PVI terminated AF in 15 (15%) patients. Pre-PVI 13 (11–15) of 18 segments harboured PDs with total PD occurrences (not sites) being 41.5 ± 15.3. Of these, 21.3 ± 9.1% (8.7 ± 4.8) of PDs occurred at the PV and posterior wall.
PVI had no impact on PD burden outside the PV and posterior wall (33.2 ± 12.9 PD occurrences pre-PVI versus 31.6 ± 12.5 after, p=0.164), distribution over the remaining 13 segments of the atria (9 (8–11) versus 9 (8–10), p=0.634), the proportion of PDs that were rotational (82.9 ± 9.7% pre-PVI versus 83.6 ± 10.1% post PVI, p=0.496) or temporal stability (2.4 ± 0.4 rotations pre-PVI versus 2.4 ± 0.5 post-PVI, p=0.541). Fewer focal PDs (AUC 0.683, 95% CI 0.528–0.839, p=0.024) but not rotational PDs (p=0.626) predicted termination of AF with PVI.
Conclusions: PVI had no demonstrable impact on PDs outside the PV and posterior wall. Focal PDs may be more important than rotational PDs in maintaining AF after PVI. Outcome data is needed to confirm whether non-invasive mapping can predict patients likely to respond to PVI.