Biatrial neuroablation attenuates atrial remodeling and vulnerability to atrial fibrillation in canine chronic rapid atrial pacing.

Abstract

We investigated the proposition that an intact cardiac nervous system may contribute to electrophysiological remodeling and increased vulnerability to atrial fibrillation (AF) following chronic rapid atrial pacing (RAP). Baseline study was conducted prior to ablating right and left ganglionated plexuses (RAGP, LAGP) in 11 anesthetized canines (Neuroablation group) and in 11 canines without neuroablation (Intact GP). After being subjected to RAP (400 beats/min) for 6 weeks, animals were reanesthetized for terminal study. The ERP shortening typical of chronic RAP was significantly more pronounced in the Intact GP (baseline: 112 ± 12 to terminal: 80 ± 11 ms) than in the Neuroablation group (113 ± 18 to 102 ± 21 ms, p < .001), and AF inducibility (extrastimulus protocol) showed significantly greater increment in the Intact GP (baseline: 23 ± 19% to terminal: 60 ± 17% of trials) than in the Neuroablation group (18 ± 15% to 27 ± 17%, p = 0.029). Negative chronotropic responses to right vagus nerve stimulation were markedly reduced immediately after the neuroablation procedure but had recovered at terminal study. Vagally-evoked repolarization changes (from 191 unipolar electrograms) occurred in a majority of Intact GP animals in the superior, middle and inferior RA free wall, and in the LA appendage. In the Neuroablation group, repolarization changes were restricted to the superior RA free wall but none occurred in the inferior RA and only infrequently in the LA appendage, yielding significantly smaller affected areas in Neuroablation than in Intact GP animals. Persistent functional denervation in LA and RA regions other than RA pacemaker areas may contribute to prevent the development of a tachycardia-dependent AF substrate.