Secondary Logo

Journal Logo

Original article

Substrate of complex fractionated atrial electrograms: evidence by pathologic analysis

De-jun, YOU; Dong, CHANG; Shu-long, ZHANG; Dong-hui, YANG; Lian-jun, GAO; Xiao-meng, YIN; Zhen-liang, CHU; Yun-long, XIA; Yu-chun, WANG; Ying-xue, DONG; Yan-zong, YANG

Author Information
  • Free


Catheter ablation has become a well-established and effective treatment strategy for atrial fibrillation (AF). The traditional approach of ablation has been to eliminate all of the triggers of AF by isolating the pulmonary veins (PVs) and non-PV foci. This has been the cornerstone of most AF ablation approaches to date in drug-refractory AF, including both the paroxysmal AF (PAF) and persistent AF (PeAF) patients.1,2 However, electrical isolation and circumferential ablation of pulmonary vein antrum are not enough in many patients who have AF with significantly remodeled atrial substrate.

Alternative approaches to AF ablation have been proposed. Nademanee et al3 first reported an approach by mapping and ablating complex fractionated atrial electrograms (CFAE) during AF as an efficacious treatment for AF, the so-called electrophysiological substrate modification. AF was terminated in 95% of the patients without recurrences for one year in 91% of patients. Afterwards, several studies demonstrated that CFAE ablation might improve the outcome of ablation in PAF and PeAF compared to the circumferential pulmonary vein isolation (CPVI) alone.4 CFAE ablation has been regarded as an important adjunctive strategy for PAF and PeAF therapy.1,2 Several hypotheses have been proposed to contribute to the formation of CFAE as follows: functional conduction block, slow conduction, pivotal points or an “isthmus” for local reentry,5 migration rotors,6 rapid electrical activity from a driving source,7 or regions adjacent to ganglionated plexus (GPs).8 However, few data are available on the pathologic basis of CFAE. Therefore, in the present study, we aimed to elucidate the pathologic characteristics of CFAE.


Animal preparation

The experimental protocol was approved by the Institutional Animal Care and Use Committee of Dalian Medical University. Nine adult mongrel dogs weighing 9–12 kg were anesthetized with sodium pentobarbital; initial bolus of 30 mg/kg i.p. with 30–60 mg as needed for maintenance. Each dog was ventilated using a constant volume-cycled respirator through a cuffed endotracheal tube. The blood oxygen saturation was maintained at more than 95% throughout the experiment. Core body temperature was maintained at (36.5±1.0)°C using a heated pad and a lamp.

Under fluoroscopic guidance (Innova 2000, GE Co., USA), two circumferential catheters (10-pole, 15-mm Lasso, Cordis Webster Co., USA) were placed at the high right atrium (HRA) and the coronary sinus (CS) through the right internal jugular vein and right femoral vein, respectively, for the electrophysiological studies. A quadripolar catheter was advanced through the left femoral vein into the right ventricular apex for pacing at a rate of 100 beats/min in case of severe bradycardia. Two Swartz sheathes (St. Jude Medical, Inc., USA) were advanced into the left atrium after two successful transseptal procedures. A left atrial angiography was performed. Surface ECG II/aVF and intracardiac electrograms were recorded simultaneously using a Prucka Cardiolab system (Prucka 7000, GE Medical System, Inc., USA). The signals were amplified and filtered between 30 and 500 Hz. Heparin sodium (100 U/kg) was administered intravenously to prevent blood coagulation.

AF induction

Both cervical vagosympathetic trunks were exposed by surgical procedures. The cranial ends of the vagosympathetic nerves were ligated. Two pairs of silver wires were introduced into the caudal end of the vagosympathetic trunks. Metoprolol succinate was administered (initial bolus of 0.2 mg/kg i.v. with 0.2 mg/kg i.v. per hour or maintenance) to block the sympathetic nerve. Rectangular electrical stimuli were delivered at a frequency of 20 Hz and a pulse duration of 2 ms (RST-2, Huanan Med Inc., China) to each of the nerve trunks for vagal nerve stimulation (VNS). The stimulators were programmed with sufficient voltage and frequency to obtain a 50% reduction in the ventricular rate during VNS.8 The stimulation strength was kept constant during the experiment. Sustained AF (>30 minutes) was induced through rapid atrial pacing (600 beats/min) at HRA with VNS.

CFAE mapping and ablation

A 3.5 mm irrigated catheter was advanced into both atriums via a Swartz sheath. During AF, both atrial electroanatomic maps were reconstructed using an Ensite system with CFAE software (EnSite-NavX version 8.0; St. Jude Medical, Inc.). The signals were recorded from the ablation catheter and digitally filtered with a bandpass filter of 32–100 Hz. An additional noise filter was switched on if necessary. The bipolar electrograms from 60 to 100 points of the atrial endocardium were mapped. Lin et al9 reported that the assessment of fractionated electrograms requires a recording duration of ≥5 seconds at each site to obtain a consistent fractionation. A stable catheter position and good tissue contact were confirmed during CFAE mapping. If the catheter moved, or if poor tissue contact was suspected, the recording was excluded. CFAE was automatically identified and displayed as color-tagged in atrial 3-dimensional geometric mapping. In addition, CFAE was also identified through artificial measuring. CFAE was defined as a fractionated potential exhibiting multiple deflections from the isoelectric line (≥2 deflections) and/or continuous electrical activity without the isoelectric line, or a fractionation interval less than 120 ms (Figure 1). To label the sites with CFAE, an ablation was performed with a power output setting at 30 W and 43°C for 8–10 seconds.

Figure 1.
Figure 1.:
CFAE recording during AF. CFAE showed multiple deflections from the isoelectric line (≥2 deflections) and/or potential with continuous electrical activity without isoelectric line or fractionation interval less than 120 ms.

Pathologic examination

After ablation, the hearts were removed for histological examination. The underlying tissue of the CFAE and non-CFAE sites was excised and fixed in 4% formalin for more than 48 hours for histological examination. Cross-sections were cut at a 4 μm thickness using a microtome (LEICA RM2245). Slides were stained with hematoxylin-eosin (HE) and mouse anti-protein gene product 9.5 (PGP 9.5) respectively for microscopic examination. In every section stained by HE, five homalographic visual fields under 10-fold light microscope were randomly chosen for images acquisition through microscope examination (Nikon Eclipse 80i microscope, Nikon Corporation, Japan). The images were automatically transmitted to the computer to evaluate the percentage of myocardial intercellular substance with image analysis software (NLS-Elements BR 3.0 imaging software, Nikon Corporation). Under 10-fold light microscope magnification, the photograph of every section stained by PGP 9.5 was acquired (Olympus SZX16 research stereo microscope, Olympus DP71 microscope digital camera, Olympus Corporation, Japan) and transmitted to the computer to evaluate the area with image analysis software (IPP6.0, Media Cybermetics Inc., USA). GPs were stained brown with PGP 9.5. The bundles of GPs were identified by microscope examination at 10× magnification. The density of GPs was calculated as the number of nerve bundles per square centimeter. Digital photographs were taken. The panoramic view was produced by reconstruction of multiple images of the same specimen using Adobe Photoshop 11.0.

Statistical analysis

The distribution of intercellular substance and the density of GPs in the CFAE and non-CFAE sites were expressed as mean ± standard deviation. Comparisons between the CFAE and non-CFAE sites were tested by analysis of variance (ANOVA) or rank sum test. All statistical analyses were performed using SPSS 13.0 (SPSS Inc., USA). A P value <0.05 was considered statistically significant.


The experiments were successfully performed. The systolic and diastolic blood pressures were stable during the entire procedure. Sustained AF was successfully induced in all nine dogs. Myocardium and GPs were partly destroyed after ablation. Red cells were observed in the intercellular substance. Intercellular substance was mainly composed with adipose tissue. On the whole, the general architecture of GPs and intercellular tissue was preserved and identified (Figures 2 and 3).

Figure 2.
Figure 2.:
Section stained by HE (Original magnification ×10). 2A: In non-CFAE sites, the myocardium was organized with less intercellular substance. 2B: In CFAE sites, the myocardium was disorganized with more intercellular substance.
Figure 3.
Figure 3.:
Section stained by PGP 9.5 (Original magnification ×10). 3A: In non-CFAE sites, the myocardium was organized with less intercellular substance. 3B: In CFAE sites, the myocardium was disorganized with more GPs. And GPs were destroyed by ablation.

Regional distributions of CFAE

CFAE was detected mainly around areas such as the CS, the PVs, the superior vena cava (SVC), and the atrial septum (Table 1). The density of CFAE around the PVs and the CS was significantly higher compared with other regions of the atrium (P <0.05).

Table 1
Table 1:
Distribution of CFAEs in dogs

Heterogeneity of myocardium in CFAE sites

Transmural tissue sections from the CFAE and non-CFAE sites were stained with HE (Figure 2). The myocardium formed an organized pattern with little intercellular substance in non-CFAE sites, whereas the myocardium formed a disordered pattern with more intercellular substance in CFAE sites ((61.7±24.3)% vs. (34.1±9.2)%, P <0.01).

Heterogeneity of GPs in CFAE sites

GPs can be specially stained brown after labeling the PGP 9.5 protein. Transmural tissue sections from the CFAE and non-CFAE sites were stained with PGP 9.5 (Figure 3).

More GPs were identified in the CFAE sites compared with non-CFAE sites ((34.45±37.46) bundles/cm2 vs. (6.73±8.22) bundles/cm2, P <0.01).


The main findings of the present study are as follows: (1) CFAE was detected mainly in the CS, the PV antrum, SVC, and the atrial septum. (2) The myocardium was heterogeneously arranged with more intercellular substance in the CFAE sites. (3) There were more GPs in CFAE sites. These findings underscored characteristics of the substrate of CFAE.

Regional distributions of CFAE

Consistent with previous studies, CFAE in the present study was prevalent in preferential areas including the CS, the PVs, SVC, and the atrial septum.10 The anisotropy or summation of electrograms from the overlapping layers of myocardial fibers may cause CFAE. For example, the PV antrum may be intersections of the muscular sleeve of pulmonary veins and left atrium.11 The CFAE sites may represent the substrate of AF, which serves the AF perpetuation and a potential target of ablation. Ablation among these sites above could result in prolongation of the AF cycle length, AF termination, and noninducibility.12,13

Myocardium anisotropy in CFAE sites

CFAE was first reported by Konings et al5 during the intraoperative mapping of human AF. Studies have demonstrated that CFAE is generated at sites of local reentry around areas of functional conduction block.5 Park et al10 described CFAE existing in an area of low voltage and slow conduction. In a computer model of AF, macroscopic heterogeneity (abrupt changes in tissue conduction or macroscale obstacles) presented a higher percentage of CFAE.14 This was supported by the findings of the present study. Through pathologic examination, we found that CFAE sites contained disorganized myocardium and more intercellular substance compared with non-CFAE sites. More intercellular substance may further attenuate the myocardial conduction. Moreover, myocardial fibrosis was enhanced in CFAE sites.15 In a computer model, when microfibrosis was more severe, electrograms were more fractionated.16 The myocardial heterogeneity may lead to slow conduction or delayed activation,17 which may be responsible for multiple re-entrant circuits with a variable activation cycle and perpetuation of AF.

Autonomic mechanism of CFAE

The intrinsic cardiac anatomic nerve system (ICANS) is known to form a complex neural network and to converge at the GPs which function as the “integration centers” for the autonomic innervation to modulate important electrophysiological functions.18 The activity of the ICANS may be an important mechanism responsible for the initiation and maintenance of AF.19,20 In a canine cholinergic AF model, CFAE was most abundant at the electrodes closest to the GPs and exhibited a significant gradient of progressively decreasing dominant frequency (DF) from the GPs toward distant sites.8 After GPs ablation, CFAE was markedly attenuated and the DF gradient was eliminated.8 Furthermore, several studies have demonstrated that CFAE sites were coincident with four main GPs in the human atria. Katritsis et al21 reported that CFAE at the presumed anatomic sites of GPs was identified in 68.8% patients with PAF. In a prior study,22 we demonstrated that CFAE ablation could suppress AF induction mediated by vagal stimulation. These findings predict that the autonomic activation and neurotransmitter release may play important roles in the genesis of CFAE and their preferential distribution during AF. However, the pathologic mechanism of CFAE is not yet fully understood.

In the present study, there were more GPs in CFAE sites compared with non-CFAE sites. Through pathologic examination, we provided evidences for the autonomic mechanism of CFAE, which strongly supported the observation that CFAE is associated with GPs. The activated GPs may release excessive concentrations of neurotransmitters at the peripheral neural elements that modulate the atrial wavefront stability by the detachment of stationary reentry and the generation of multiple wave breaks.23 This would allow more colliding atrial wavefronts to occur and subsequently generate CFAE.23 In canines, most efferent GPs to the atria appear to travel through a fat pad located between the medial SVC and aortic root (the third fat pad), then project onto another two fat pads at the inferior vena cava-left atrial junction and the right PV-atrial junction, and finally to both atria.24 It may partly account for the preferential sites of CFAE.

Clinical relevance

In the present study, we found the heterogeneity of the GPs and myocardium might account for the substrate of CFAE and serve as a potential target of ablation. This may contribute to understand the mechanisms of CFAE and substrate of AF. Because CFAE ablation is an important adjunctive therapy for AF, the present study also offered evidences for GPs ablation in the treatment of AF. In future clinical studies we could focus on GPs ablation in addition to pulmonary vein isolation during AF ablation. According to the anatomic sites of GPs21 and vagal reflexes,25 we could target at GPs for ablation and modify the AF substrate to improve the success rate of AF ablation. In the present study, the AF model was performed through short-term rapid atrial pacing plus VNS, which may be the basis of PAF. Therefore, GPs ablation in PAF might be reasonable.

Study limitations

In the present study, we performed short-term rapid atrial pacing plus VNS to induce AF and record CFAE. This AF model may not represent the complete spectrum of AF. The CFAE patterns may be different depending on the AF models and various remodeled atria. Therefore, the substrate of CFAE found in the present study cannot fully account for the mechanisms of CFAE in other types of AF, particularly in patients with organic heart diseases and PeAF. In the present study, we only performed ablations on the CFAE sites. After ablation, myocardium and GPs were partly destroyed. But the general architecture of the intercellular substance and GPs were still preserved and identified after ablation. The effect of ablation on the pathological examination might be small. The heterogeneity of the myocardium and GPs distribution may account for the substrate of CFAE. In a prior study,22 we demonstrated that CFAE ablation could blunt ERP shortening and suppress AF mediated by enhanced vagal activity. So we did not evaluate the electrophysiological change after CFAE ablation in the present study. We did not evaluate the connexin, fibrosis, and ion currents which may play important roles in the genesis of AF. Further studies are needed to clarify these topics.

In conclusion, the pathologic examination demonstrated that the myocardium was disorganized with more intercellular substance and GPs in CFAE sites compared with non-CFAE sites. The heterogeneity of the myocardium and GPs may be responsible for the substrate of CFAE and serve as a potential target of ablation for AF.


1. Wann LS, Curtis AB, January CT, Ellenbogen KA, Lowe JE, Estes NA 3rd, et al. 2011 ACCF/AHA/HRS focused update on the management of patients with atrial fibrillation (updating the 2006 guideline). Heart Rhythm 2011; 8: 157-176.
2. Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010; 12: 1360-1420.
3. Nademanee K, McKenzie J, Kosar E, Schwab M, Sunsaneewitayakul B, Vasavakul T, et al. A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate. J Am Coll Cardiol 2004; 43: 2044-2053.
4. Kong MH, Piccini JP, Bahnson TD. Efficacy of adjunctive ablation of complex fractionated atrial electrograms and pulmonary vein isolation for the treatment of atrial fibrillation: a meta-analysis of randomized controlled trials. Europace 2010; 13: 193-204.
5. Konings KT, Kirchhof CJ, Smeets JR, Wellens HJ, Penn OC, Allessie MA. High-density mapping of electrically induced atrial fibrillation in humans. Circulation 1994; 89: 1665-1680.
6. Kalifa J, Tanaka K, Zaitsev AV, Warren M, Vaidyanathan R, Auerbach D, et al. 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.
7. Umapathy K, Masse S, Kolodziejska K, Veenhuyzen GD, Chauhan VS, Husain M, et al. Electrogram fractionation in murine HL-1 atrial monolayer model. Heart Rhythm 2008; 5: 1029-1035.
8. Lin J, Scherlag BJ, Zhou J, Lu Z, Patterson E, Jackman WM, et al. Autonomic mechanism to explain complex fractionated atrial electrograms (CFAE). J Cardiovasc Electrophysiol 2007; 18: 1197-1205.
9. Lin YJ, Tai CT, Kao T, Chang SL, Wongcharoen W, Lo LW, et al. Consistency of complex fractionated atrial electrograms during atrial fibrillation. Heart Rhythm 2008; 5: 406-412.
10. Park JH, Pak HN, Kim SK, Jang JK, Choi JI, Lim HE, et al. Electrophysiologic characteristics of complex fractionated atrial electrograms in patients with atrial fibrillation. J Cardiovasc Electrophysiol 2009; 20: 266-272.
11. Roux N, Havet E, Mertl P. The myocardial sleeves of the pulmonary veins: potential implications for atrial fibrillation. Surg Radiol Anat 2004; 26: 285-289.
12. Haïssaguerre M, Hocini M, Takahashi Y, O'Neill MD, Pernat A, Sanders P, et al. Impact of catheter ablation of the coronary sinus on paroxysmal or persistent atrial fibrillation. J Cardiovasc Electrophysiol 2007; 18: 378-386.
13. Takahashi Y, O'Neill MD, Hocini M, Dubois R, Matsuo S, Knecht S, et al. Characterization of electrograms associated with termination of chronic atrial fibrillation by catheter -ablation. J Am Coll Cardiol 2008; 51: 1003-1010.
14. Jacquemet V, Virag N, Ihara Z, Dang L, Blanc O, Zozor S, et al. Study of unipolar electrogram morphology in a computer model of atrial fibrillation. J Cardiovasc Electrophysiol 2003; 10 Suppl: S172-S179.
15. Liu X, Shi HF, Tan HW, Wang XH, Zhou L, Gu JN. Decreased connexin 43 and increased fibrosis in atrial regions susceptible to complex fractionated atrial electrograms. Cardiology 2009; 114: 22-29.
16. Jacquemet V, Henriquez CS. Genesis of complex fractionated atrial electrograms in zones of slow conduction: a computer model of microfibrosis. Heart Rhythm 2009; 6: 803-810.
17. Hocini M, Ho SY, Kawara T, Linnenbank AC, Potse M, Shah D, et al. Electrical conduction in canine pulmonary veins: electrophysiological and anatomic correlation. Circulation 2002; 105: 2442-2448.
18. Hou Y, Scherlag BJ, Lin J, Zhang Y, Lu Z, Truong K, et al. Ganglionated plexi modulate extrinsic cardiac autonomic nerve input: effects on sinus rate, atrioventricular conduction, refractoriness, and inducibility of atrial fibrillation. J Am Coll Cardiol 2007; 50: 61-68.
19. Chang D, Zhang S, Yang D, Gao L, Lin Y, Chu Z, et al. Effect of epicardial fat pad ablation on acute atrial electrical remodeling and inducibility of atrial fibrillation. Circ J 2010; 74: 885-894.
20. Lim PB, Malcolme-Lawes LC, Stuber T, Wright I, Francis DP, Davies DW, et al. Intrinsic cardiac autonomic stimulation induces pulmonary vein ectopy and triggers atrial fibrillation in humans. J Cardiovasc Electrophysiol 2011; 22: 638-646.
21. 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.
22. Zhang S, Dong Y, Jiang P, Gao L, Cha Y, Packer DL, et al. Effect of ablation of complex fractionated atrial electrogram on vagal modulation in dogs. Chin Med J 2010; 123: 3288-3292.
23. Zou R, Kneller J, Leon LJ, Nattel S. Substrate size as a determinant of fibrillatory activity maintenance in a mathematical model of canine atrium. Am J Physiol Heart Circ Physiol 2005; 289: H1002-H1012.
24. Chiou CW, Eble JN, Zipes DP. Efferent vagal innervation of the canine atria and sinus and atrioventricular nodes: the third fat pad. Circulation 1997; 95: 2573-2584.
25. Scanavacca M, Pisani CF, Hachul D, Lara S, Hardy C, Darrieux F, et al. Selective atrial vagal denervation guided by evoked vagal reflex to treat patients with paroxysmal atrial fibrillation. Circulation 2006; 114: 876-885.

complex fractionated atrial electrograms; atrial fibrillation; pathologic examination; ganglionated plexus

© 2012 Chinese Medical Association