Numerous studies have shown that the vagal system plays a pivotal role in the pathogenesis of atrial fibrillation (AF).1,2 Vagal denervation by catheter ablation is effective in suppressing AF.3,4 Complex fractionated atrial electrograms (CFAEs) are distributed at the preferential anatomical area of the atria, such as the left inferior ganglionated plexus and crux ganglionated plexus, in patients with AF.5–7 Ablation-targeting CFAE has shown a high success rate in eliminating AF, addressing the importance of left atrial substrate modification for maintaining sinus rhythm.8,9 However, the mechanisms that underlie these CFAEs have not yet been clearly elucidated.10–13 We sought to explore the mechanisms of CFAE ablation on vulnerability to AF and modulation of the vagus to the atrium.
Ten mongrel dogs (provided by Dalian Medical University Animal Lab) of either sex weighing 10 to 15 kg were enrolled in this study. All the dogs were anesthetized with sodium pentobarbital (150 mg/kg, i.v.), and additional amounts of 250 mg per 60 minutes were given as necessary to maintain anesthesia during the study.14 The dogs received ventilation with room air; continuous blood pressure and oximetry were monitored throughout the experiment, and blood pressure stayed at more than 90/60 mmHg. Six surface limb-lead electrocardiograms and intracardiac electrograms were continuously recorded with a multichannel computerized recording system (Prucka 7000; GE Medical Systems, Waukesha, Wisconsin, USA).
A 6F quadripolar catheter was put into right ventricular apex through the left femoral vein for pacing. Two multipolar catheters (Cordis Webster Co., USA) were advanced into the coronary sinus (CS) and right atrium (RA) through the right internal jugular vein separately. Transseptal puncture was performed under fluoroscopic guidance (Innova 2000, GE Co., Wisconsin, USA). An ablation catheter was deployed in the left atrium through a transseptal sheath (Johnson & Johnson Co., USA) for mapping and ablation via right femoral vein. Multipolar Halo catheter and Lasso catheter were deployed to the right atrium and pulmonary veins (PVs) mapping in two dogs.
Effective refractory period (ERP), vulnerability window (VW) of AF, and sinus rhythm cycle length (SCL) were measured with and without vagal stimulation (VS) at the right atrial appendage (RAA), the left atrial appendage (LAA), the distal CS (CSd), and the proximal CS (CSp) at baseline. Atrial pacing protocol with single extra stimulation was performed using a programmable multichannel stimulator (model DF-5A Electrophysiology; Dongfang Co., Shanghai, China). The pacing amplitude was set at twice the diastolic threshold that was determined at a basic drive cycle length of 250 ms at each site, including RAA, LAA, CSd, and CSp. These measurements were repeated after ablation. Sustained AF was induced by S1S2 programmed stimulation with or without vagal stimulation.
ERP was determined with a single extrastimulation at coupling intervals starting at 200 ms and progressively shortened at 10-ms decrements with a drive cycle length of 250 ms. VW was defined as the range of coupling interval of the extrastimuli during which repetitive atrial responses or fibrillation was induced. Atrial ERP shortening was defined as the difference between the ERP measured at baseline and during vagal stimulation. ERP, SCL, and VW of AF were measured with and without vagal stimulation before and after CFAE ablation. Temporary pacemakers (pacing in 120 beats per minute) were applied in the case of bradycardia due to stimulation on the vagal trunk.
Metoprolol succinate was administered before vagal stimulation (bolus, 0.2 mg/kg maintenance dosage, 0.2 mg/kg per hour) to block the sympathetic nerve. Both cervical vagal trunks were exposed by surgical procedure, and the cranial ends of vagal nerves were ligated. Two pairs of wire electrodes were embedded in the caudal end of vagus for stimulation. Rectangular pulse was delivered at a frequency of 20 Hz and at a pulse width of 2 ms through a constant programmable stimulator (model RST-2; Huanan-Med Inc., Hangzhou, Zhejiang, China).15 The stimulation voltage was set at 5 volts higher than that required to produce sinus bradycardia.
CFAE mapping and quantitation
CFAEs were automatically identified and displayed as color tagged on atrial 3-dimensional geometric mapping with use of built-in, automated CFAE registration software (EnSite-NavX; St. Jude Medical Inc., St. Paul, MN, USA). CFAEs were defined as multicomponent atrial electrograms including (1) atrial electrograms with two or more deflections and/or perturbation of the base line and/or continuous electrical activity over a 5-second period or (2) atrial electrograms with a very short cycle length ≤80 milliseconds over a 5-second period, for the heart rate of dogs in this study was quite faster than human beings. Bipolar electrograms were recorded from the ablation catheter and filtered at a bandpass of 30–500 Hz. Bipolar electrograms from 60 to 100 points of the atrial endocardium were mapped, and CFAEs were obtained in a window exceeding 5 seconds in each site. All bipolar electrogram (EGM) peaks were with a voltage amplitude ≥twice the maximal noise amplitude of the baseline. Refractory period was set in 50 ms to limit double counting, and width was set in 15 ms to limit detection to high frequency signals. CFAE was exhibited by measuring the time between multiple discrete deflection (-dp/dt) and different mean cycle length (CL) was projected onto LA anatomical shell in different colorful codes.
Atria were divided into eight quadrants corresponding to anatomical locations as Figure 1. The quantitation of dog with CFAE in every different atrial quadrant was recorded and tagged onto the atrial 3D geometry respectively, the frequency of which in all 8 atrial quadrants was also evaluated.
The catheter ablation procedures were performed in all dogs using cold saline infusion catheter under the direction of 3D electroanatomical mapping. Radiofrequency energy was applied at the regions with CFAE with maximum temperature set at 50°C and maximal energy of 60 W for 30–60 seconds (Figure 2).
The underlying tissues were excised from ablative sites and the same sites without ablation as control specimens. Tissues were fixed in buffered neutral formalin. Serial sections were taken and stained with hematoxylin and eosin for microscopic examination.
Data are reported as mean ± standard deviation (SD). A P value of 0.05 or less was considered statistically significant. Comparisons before and after ablation were performed with 2-tailed, paired t test. All tests were performed with statistical software (SPSS version 11.2; SPSS Inc., USA). Categorical variables of the CFAE quantified in different quadrants of atria were expressed as frequency (%).
Preferential sites of CFAE area
AF lasting longer than 5 minutes could be induced in all 10 dogs with vagal stimulation. We compared the mean CFAE distribution in 8 different quadrants of both right and left atria. The quadrants of the right and left inferior pulmonary veins showed a lower CFAE distribution than the other quadrants. Most of the CFAE areas were localized at the right superior pulmonary vein quadrant, the CSd quadrant and CSp quadrant, the superior vena cava (SVC) quadrant and the inferior vera cava (IVC) quadrant (Table).
Effect of CFAE ablation on sinus heart rate
CFAE ablation was performed as described previously in all 46 CFAE tagged zone, and every point was burned more than 30 seconds. The sinus heart rate decreased significantly during vagal stimulation ((44±21) beats per minute vs. (151±14) beats per minute; P <0.001). The change in sinus heart rate was blunted after CFAE ablation, and the shortening of sinus heart rate for the vagal stimulation decreased significantly after CFAE ablation ((107±19) beats per minute vs. (18±8) beats per minute, P <0.001).
Effect of CFAE ablation on ERP
ERP at baseline was reduced significantly with vagal stimulation ((103±14) ms vs. (38±14) ms at RAA, P <0.001; (100±16) ms vs. (37±23) ms at CSd, P <0.005; (38±20) ms vs. (99±13) ms at CSp, P <0.001; (60±20) ms vs. (101±26) ms at LAA, P <0.001). In contrast, after CFAE ablation, ERP changed little with vagal stimulation ((103±23) ms vs. (95±16) ms at RAA, P >0.05; (98±25) ms vs. (75±30) ms at CSd, P >0.05; (92±24) ms vs. (96±16) ms at CSp, P >0.05; (83±16) ms vs. (89±14) ms at LAA, P >0.05). ERP shorten of LAA was found in one dog for the continuous AF. CFAE ablation significantly increased the shortest ERP from a mean of (17±24) ms to (65±22) ms at RAA (P <0.001); from (9±10) ms to (65±3) ms at CSd, P <0.001; from (6±14) ms to (41±19) ms at LAA (P <0.001); and from (4±16) ms to (61±21) ms at CSp, P <0.001 (Figure 3).
Effect of CFAE ablation on VW
VW during vagal stimulation significantly decreased after CFAE ablation ((58±13) vs. (18±29) ms at RA, P <0.05; (52±28) vs. (11±20) ms at CSd, P <0.05; (62±35) ms vs. (7±13) ms at CSp, P <0.05; (31±24) ms vs. (19±22) ms at LAA, P <0.05).
The architecture of individual ganglia was significantly altered after ablation. In control specimens, the ganglionic capsule was well-demarcated with healthy surrounding tissue and the plump neurons (Figure 4A). While after the CFAE ablation, the ganglion plexus was not well-demarcated and disconnected with the adipose tissue (Figure 4B).
We found that the CFAE ablation could attenuate the vagal modulation to atria, thereby suppressing AF mediated by enhanced vagal activity in dogs. We also found the relationship between the CFAE and atrial vagal innervation in dog models, suggesting that CFAE has autonomic basis.
Atrial CFAE distribution and anatomic basis
In this study, CFAE distribution is explored under the atrial vagal innervation in dog models. It proves that CFAE locates mostly at the roof and superior pulmonary quadrant, CSd and CSp quadrant, the IVC and SVC quadrants in dogs, which is different from other reports and especially those about the CFAE location on the LA in the human being.10,16,17
Investigators have shown that the efferent vagal ganglion plexus innervating atria in dogs are mainly located in three epicardial fat pads.18 Most efferent vagal ganglion plexus to the atria appear to travel through a fat pad located between the medial SVC and aortic root; some plexus to the sinus node are located in the pad between RPV and LA; and others to the AV node are located in the fat pad between IVC and LA.
Such locations are correlated with those distributions of CFAE demonstrated in this study. Furthermore, the ganglia node was clearly demonstrated in the field of CFAE with the additional histochemistry method. These results support an association between atrium vagal distribution and CFAE clustering.
CFAE ablation modulation of vagal efferent control of atria
Several studies have indicated that the intrinsic cardiac vagal modulation plays a critical role in the initiation and maintenance of AF.19–22 Moreover, the dose-response relationship between the concentration of acetylcholine (Ach) applied and the degree of action potential duration (APD) shortening and CFAE induction has been demostrated.5 However, there are limited reports to explore the impact of CFAE ablation on the atrial vagal modulation.23
We find in the present study that acute development of CFAE with vagal stimulation is a function of autonomic manipulation in the normal heart. Shortened atrial ERP and increased VW for induction of AF at CFAE sites, observed in our study, support the critical role of heightened vagal tone in predisposing the atria to fibrillation. The vagal discharge, one of the mechanisms, modulates the occurrence of CFAE. Partial denervation of efferent vagal neurons in the atria could be accomplished by CFAE ablation. Decreased ERP shortening after ablation indicates that CFAE ablation could induce vagal denervation.
First, CFAEs were displayed in 60 to 100 limited points of the atrial endocardium, and the data were recorded from the dog AF model, which are a little different from the clinical implication. Second, the detected CFAE with an adaptive peak-to-peak sensitivity threshold is greater than 0.1 millivolt in the NavX system. Therefore, this study's finding may not be directly comparable with other studies that used different parameters to identify the CFAE, and few CFAEs with serious low voltage might be considered noise. Third, part of the atrial septum, especially the area around the transseptal point is hard to map. Finally, the study showed an acute attenuation to the atrial vagus after CFAE ablation. However, further long-time observation is required because one 4-week study reported that atrial vagal denervation withdraws over time.24
In conclusion, this study has provided evidence that CFAE is associated with autonomic manipulation via vagal stimulation in dogs. The atrial CFAE regions are correlated with those that reside ganglionated plexi (GPs) enriched in vagal neurons identified by 3-dimension electro-anatomic mapping and concomitant histopathological study. CFAE ablation attenuates the parasympathetic control of atria, thereby lessening the electrical substrate for initiation and maintaining AF.
1. Razavi M, Zhang S, Yang D, Sanders RA, Kar B, Delapasse S. Effects of pulmonary vein ablation on regional atrial vagal innervation and vulnerability to atrial fibrillation
in dogs. J Cardiovasc Electrophysiol 2005; 16: 879-884.
2. Takahashi Y, Jais P, Hocini M, Sanders P, Rotter M, Rostock T, et al. Shortening of fibrillatory cycle length in the pulmonary vein during vagal excitation. JACC 2006; 47: 774-780.
3. Pappone C, Santinelli V, Manguso F, Vicedomini G, Gugliotta F, Augello G, et al. Pulmonary vein denervation enhances long-term benefit after circumferential ablation for paroxysmal atrial fibrillation
. Circulation 2004; 109: 327-334.
4. Lemery R, Birnie D, Tang AS, Green M, Gollob M. Feasibility study of endocardial mapping of ganglionated plexuses during catheter ablation
of atrial fibrillation
. Heart Rhythm 2006; 3: 387-396.
5. Katritsis D, Giazitzoglou E, Sougiannis D, Voridis E, Po SS. Complex fractionated atrial electrograms at anatomic sites of ganglionated plexi in atrial fibrillation
. Europace 2009; 11: 308-315.
6. Singh SM, Heist EK, Koruth JS, Barrett CD, Ruskin JN, Mansour MC. The relationship between electrogram cycle length and dominant frequency in patients with persistent atrial fibrillation
. J Cardiovasc Electrophysiol 2009; 20: 1336-1342.
7. Zhang Y, Nakagawa H, Po SS, Scherlag BJ, Wu R, Beckman KJ, et al. Autonomic ganglionated plexi stimulation induces fractionated atrial electrograms in contralateral pulmonary veins in patients with atrial fibrillation
. Circulation 2006; 114: SII-454.
8. Nademanee N, McKenzie J, Kosar E. A new approach for catheter ablation
of atrial fibrillation
: mapping of the electro-physiologic substrate. J Am Coll Cardiol 2004; 43: 2044-2053.
9. Schmitt C, Estner H, Hecher B, Luik A, Kolb C, Karch M. Radiofrequency ablation of complex fractionated atrial electrograms (CFAE): preferential sites of acute termination and regularization in paroxysmal and persistent atrial fibrillation
. J Cardiovasc Electrophysiol 2007; 18: 1039-1046.
10. Rostock T, Rotter M, Sanders P, Takahashi Y, Jaïs P, Hocini M. High-density activation mapping of fractionated electrograms in the atria of patients with paroxysmal atrial fibrillation
. Heart Rhythm 2006; 3: 27-34.
11. Lu Z, Scherlag BJ, Lin J, Niu G, Ghias M, Jackman WM. Autonomic mechanism for complex fractionated atrial electrograms: evidence by fast Fourier transform analysis. J Cardiovasc Electrophysiol 2008; 19: 835-842.
12. Kalifa J, Tanaka K, Zaitsev AV, Warren M, Vaidyanathan R, Auerbach D. Mechanisms of wave fractionation at boundaries of high-frequency excitation in the posterior left atrium of the isolated sheep heart during atrial fibrillation
. Circulation 2006; 113: 626-633.
13. Razavi M, Zhang S, Delapasse S, Yang D, Ai T, Kar B. The effects of pulmonary vein ablation on the dominant frequency and organization of coronary sinus electrical activity during permanent atrial fibrillation
. Pacing Clin Electrophysiol 2006; 29: 1201-1208.
14. Liu Y, Zhang S, Dong Y, Zhao HW, Gao LJ, Yin XM. Impact of right upper pulmonary vein ablation on atrial vagal innervation and vulnerability to atrial fibrillation
. Chin Med J 2006; 119: 2049-2055.
15. Dong YX, Zhang SL, Zhao HW. Impact of pulmonary veins isolation on the short-time atrial electrical remodeling in dogs. Chin J Card Arrhythm (Chin) 2007; 11: 141-144.
16. Chiou CW, Eble JN, Zipes DP. Efferent vagal innervation of the canine atrial and sinus and atrioventricular nodes: the third fat pad. Circulation 1997; 95: 2573-2584.
17. Berenfeld O, ZaitsevAV, Mironov SF, PertsovAM, Jalife J. Frequency dependent breakdown of wave propagation into fibrillatory conduction across the pectinate muscle network in the isolated sheep right atrium. Circ Res 2002; 90: 1173-1180.
18. Oral H, Chugh A, Good E, Wimmer A, Dey S, Gadeela N. Radiofrequency catheter ablation
of chronic atrial fibrillation
guided by complex electrograms. Circulation 2007; 115: 2606-2612.
19. Schauerte P, Scherlag BJ, Patterson E, Scherlag MA, Matsudaria K, Nakagawa H. Focal atrial fibrillation
: Experimental evidence for a pathophysiologic role of the autonomic nervous system
. J Cardiovasc Electrophysiol 2001; 12: 592-599.
20. Patterson E, Po SS, Scherlag BJ, Lazzara R. Triggered firing in pulmonary veins initiated by in vitro
autonomic nerve stimulation. Heart Rhythm 2005; 2: 624-631.
21. Patterson E, Lazzara R, Szabo B, Liu H, Tang D, Li YH. Sodium-calcium exchange initiated by the Ca2+
transient: an arrhythmia trigger within pulmonary veins. J Am Coll Cardiol 2006; 47: 1196-1206.
22. Scherlag BJ, Yamanashi W, Patel U, Lazzara R, Jackman WM. Autonomically induced conversion of pulmonary vein focal firing into atrial fibrillation
. J Am Coll Cardiol 2005; 45: 1878-1886.
23. Park JH, Pak HN, Kim SK, Jang JK, Choi JI, Lim HE. Electrophysiologic characteristics of complex fractionated atrial electrograms in patients with atrial fibrillation
. J Cardiovasc Electrophysiol 2009; 20: 266-272.
24. Oh S, Zhang Y, Bibevski S, Marrouche NF, Natale A, Mazgalev TN. Vagal denervation and atrial fibrillation
inducibility: epicardial fat pad ablation does not have long-term effects. Heart Rhythm 2006; 3: 701-708.