Catheter-based pulmonary vein isolation fails to prevent transient atrial arrhythmogenic changes related to acute obstructive respiratory events in a porcine model

Abstract

Aims: Pulmonary vein isolation (PVI) is the corner stone of modern rhythm control strategies in patients with atrial fibrillation (AF). Sleep-disordered breathing (SDB) is prevalent in more than 50% of patients undergoing AF ablation, and studies have indicated a greater recurrence rate after PVI in patients with SDB. Herein, we study the effect of catheter-based PVI on AF in a pig model for SDB.

Methods and results: In 11 sedated spontaneously breathing pigs, obstructive apnoeas were simulated by 75 s of intermittent negative upper airway pressure (INAP) applied by a negative pressure device connected to the endotracheal tube. Intermittent negative upper airway pressures were performed before and after PVI. AF-inducibility and atrial effective refractory periods (aERPs) were determined before and during INAP by programmed atrial stimulation. Pulmonary vein isolation prolonged the aERP by 48 ± 27 ms in the right atrium (RA) (P < 0.0001) and by 40 ± 34 ms in the left atrium (LA) (P = 0.0004). Following PVI, AF-inducibility dropped from 28 ± 26% to 0% (P = 0.0009). Intermittent negative upper airway pressure was associated with a transient aERP-shortening (ΔaERP) in both atria, which was not prevented by PVI (INAP indued ΔaERP after PVI in the RA: -57 ± 34 ms, P = 0.0002; in the LA: -42 ± 24 ms, P < 0.0001). Intermittent negative upper airway pressure was associated with a transient increase in AF-inducibility (from 28 ± 26% to 69 ± 21%; P = 0.0008), which was not attenuated by PVI [INAP-associated AF-inducibility after PVI: 58 ± 33% (P = 0.5)].

Conclusion: Transient atrial arrhythmogenic changes related to acute obstructive respiratory events are not prevented by electrical isolation of the pulmonary veins, which partially explains the increased AF recurrence in patients with SDB after PVI procedures.

Keywords: Atrial fibrillation; Autonomic nervous system; Pulmonary vein isolation; Sleep apnoea.

Graphical Abstract

Figure 1

Study timeline (A), INAP manoeuvre (B), and atrial 3D electroanatomical map (C). (A) After the preparation of the pigs, a native INAP without any interventions was performed in all animals before starting the experimental protocol. Four INAPs were performed before (baseline) and after pulmonary vein isolation (PVI), and aERPs were measured differing between pacing from the right atrium (RA) or left atrium (LA). (B) Schematic overview of the upper airway pressure during normal respiration and during an INAP manoeuvre (<−50 mbar). The aERPs were measured before (pre-INAP), during (at 30 s) (INAP), and after (post-INAP) each simulated INAP manoeuvre. (C) Atrial 3D electroanatomical map in a pig, with a multipolar catheter (CS) advanced into the coronary sinus as reference. The right (RA) and left atrium (LA) positioned in anterior-posterior-oblique view, with the entrance towards the ventricles encircled (black). The common inferior pulmonary trunk with three pulmonary veins (PV, black arrows).

Figure 2

aERP (ms) measured at baseline and after PVI, from the RA and the LA (n = 9). A paired Wilcoxon Signed Rank test (two-tailed) was used to analyse the effect of PVI compared with baseline in the RA and LA, and an unpaired Mann–Whitney U test (two-tailed) was used to compare the aERP between the atria, at baseline and after PVI. P-values < 0.05 were regarded as statistically significant.

Figure 3

AF-inducibility (%) determined before (pre-INAP) and after simulated obstructive sleep apnoea (INAP). The AF-inducibility was investigated at baseline and after PVI (n = 9). A Fisher’s exact (two-tailed) was used to compare the INAPs effect (pre-INAP vs. INAP) and the PVI effect on AF-inducibility (baseline vs. PVI). P-values < 0.05 were regarded as statistically significant.

Figure 4

The effect of INAP on the aERP (ms) measured before (pre-INAP), during (30 s of INAP), and after INAP (post-INAP) (n = 9). The aERP was not measured post-INAP if AF was induced during the INAP. The aERPs were measured from both atria before and after PVI. A Wilcoxon Signed Rank test (two-tailed) was used to analyse the effect of INAP on the aERP, and not performed at baseline post-INAP in the LA as there were two data points. P-values < 0.05 were regarded as statistically significant.