Journal of Interventional Cardiac Electrophysiology

, Volume 35, Issue 2, pp 119–125

Effect of renal sympathetic denervation on the inducibility of atrial fibrillation during rapid atrial pacing

Authors

  • Qingyan Zhao
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
  • Shengbo Yu
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
  • Minghui Zou
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
  • Zixuan Dai
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
  • Xule Wang
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
  • Jinping Xiao
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
    • Cardiovascular Research Institute of Wuhan UniversityRenmin Hospital of Wuhan University
Article

DOI: 10.1007/s10840-012-9717-y

Cite this article as:
Zhao, Q., Yu, S., Zou, M. et al. J Interv Card Electrophysiol (2012) 35: 119. doi:10.1007/s10840-012-9717-y
  • 461 Views

Abstract

Background

Atrial fibrillation (AF) is associated with activity of renin–angiotensin–aldosterone system (RAAS). Reduction in renal noradrenaline spillover could be achieved after renal sympathetic denervation (RSD). The relationship between RSD and AF is unknown.

Objective

The objective of the study was to investigate the inducibility of AF during atrial rapid pacing after RSD.

Methods

Thirteen dogs were used for the study as follows: control group (seven dogs) and RSD group (six dogs). In the control group, dogs were subjected to atrial pacing at 800 beats/min for 7 h, and atrial effective refractory period (AERP) was measured every hour in the status of non-pacing. Subsequently, pacing was stopped and the burst pacing (500 bpm) was repeated to induce AF three times. In the RSD group, after each renal artery ablation, the procedure of pacing and electrophysiological measurement was exactly same as in the control group. Blood was collected before and after pacing to measure the levels of renin, angiotensin II and aldosterone.

Results

There was a persistent decrease in AERP in both groups. However, 7 h after cessation of pacing, the induced number of times and duration of AF were higher in the control group than that in the RSD group (1.0 ± 1.26 vs 3.14 ± 2.54, P = 0.03; 16.5 ± 25.1 vs 86.6 ± 116.4, P = 0.02). The plasma aldosterone concentration increased significantly 7 h after rapid pacing in control group (renin, 119.8 ± 31.1 vs 185.3 ± 103.5 pg/ml, P < 0.01; aldosterone, 288.2 ± 43.1 vs 369.6 ± 109.8 pg/ml, P = 0.01). The levels of renin and aldosterone showed a decreasing trend in RSD group, but this did not attain statistical significance.

Conclusions

Episodes of AF could be decreased by renal sympathetic denervation during short-time rapid atrial pacing. This effect might have relationship with decreased activity of RAAS.

Keywords

Renal sympathetic nerveAblationRenin–angiotensin–aldosterone systemAtrial fibrillation

1 Introduction

The fundamental mechanisms underlying atrial fibrillation (AF) have long been debated, but electrical and structural remodelling are each important synergistic contributors to the AF substrate, and these changes further perpetuate the existence and maintenance of AF [13]. Much attention has been devoted in the past few years to assess the role of renin–angiotensin–aldosterone system (RAAS) in AF [46]. Studies have indicated that angiotensin II (Ang II) and aldosterone might be involved in atrial structural and electrical remodelling in patients with AF [7, 8]. The inhibition of the RAAS might have protective effect on remodelling [8].

Studies have demonstrated that the cross talk between the kidney and the heart includes the upregulated sympathetic nerve system, activation of the RAAS and vasopressin release. It has been shown that compromised hemodynamics during AF leads to an increase in catecholamines and sympathetic tone with parasympathetic withdrawal [9]. The activation of sympathetic nervous system is related to the effects of circulating renin released from the kidneys. In the previous study, Nakashima et al. showed that the inhibition of endogenous Ang II prevented atrial effective refractory period (AERP) shortening during rapid atrial pacing [10]. Whether renal sympathetic denervation could decrease the released renin and the initiation of AF during rapid atrial pacing is unknown. In the present study, we used a rapid atrial-paced canine model to examine the effect of renal sympathetic innervation on acute atrial remodelling process and inducibility AF by catheter-based renal denervation.

2 Methods

2.1 Experimental animals

Thirteen dogs, weighting between 19 and 23 kg (mean, 21 ± 3 kg), were used in the present study. The protocol of study was approved by the Ethical Committee of the Wuhan University School of Medicine, and animal handling was carried out according to the Wuhan Directive for Animal Research. All dogs were premedicated with pentobarbital sodium (30 mg/kg IV), intubated and ventilated with room air supplemented with oxygen by a respirator (MAO01746, Harvard Apparatus Holliston, USA). Normal saline at 50 to 100 ml/h was infused to replace spontaneous fluid losses. Standard surface ECG leads (I, II and III) were monitored continuously throughout the procedure. Hemostatic sheaths were inserted into the right femoral veins. Multi-polar electrode catheter was introduced into the femoral vein and placed in the high right atrium (Fig. 1). Furthermore, one hemostatic sheath was inserted into the left femoral artery, and artery blood pressure was measured through the left femoral artery.
https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig1_HTML.gif
Fig. 1

Multi-polar electrode catheter was placed in the high right atrium

2.2 Study protocol

Thirteen dogs, divided into two groups, were used for the study as follows: control group (seven dogs) and renal sympathetic denervation (RSD) group (six dogs). In the control group, dogs were subjected to atrial pacing at 800 beats/min for 7 h, and AERP was measured at every hour in the status of non-pacing. Subsequently, pacing was stopped and the burst pacing (500 beats/min) was repeated to induce AF each hour for three times. In the RSD group, 6 F ablation catheter was introduced (Biosense Webster, Inc., Diamond Bar, CA, 91765, USA) into each renal artery via femoral artery. We applied discrete, radiofrequency ablations lasting up to 80 s each and of 6 W or less within each renal artery. During ablation, the catheter was advanced and retreated three times. The catheter system monitored tip temperature and impedance, altering radiofrequency energy delivery in response to a predetermined algorithm. The procedure of pacing and electrophysiological measurement in RSD group was exactly same as in the control group.

In another three dogs, we measured different parameters during the radiofrequency procedure. We did this measurement to test the safety and effectiveness of the device in achieving efferent RSD.

2.3 Induction of atrial fibrillation and biochemical studies

S1S1 (120 ms cycle length) programmed stimulus method was adopted to induce AF. The duration of S1S1 stimulation lasted for 5 s. AF was defined as irregular atrial rates faster than 500 bpm, lasting >5 s.

Four millilitres of venous blood was collected in EDTA vacutainers and centrifuged at 2,310×g for 10 min at 4 °C (Beckman Coulter, Avanti J-E) before and after pacing. The serum was separated and kept in microtubes and stored at −80 °C until assay. The levels of renin, Ang II and aldosterone were examined by ELISA.

2.4 Statistical analysis

Values are shown as mean + SD. Statistical comparisons were made using ANOVA. Paired and unpaired comparisons were conducted using Student’s t test. Unpaired t tests were used to compare the means of nerve densities. Statistical significance was assumed if P values were <0.05.

3 Results

In the control and RSD groups, the systolic and diastolic blood pressures were stable during the entire period of experiments. When compared with control dogs, RSD dogs had the same blood pressure after ablation and throughout the 7-h pacing period.

3.1 Effect of renin sympathetic ablation

Figure 2(a), (c) shows the results of right renal artery arteriography before and after ablation. The right renal artery had no obvious stenosis after ablation lasting up to 80 s each and of 6 W or less within each renal artery. Figure 3 is an example of haematoxylin and eosin (HE) staining of renal artery with or without ablation. Vascular wall cells represented mixed appearance and disordered form after renal artery ablation. Figure 4 showed GAP43-positive staining in the renal artery. The ganglioinic cells have contracted morphology and lost histological details after ablation.
https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig2_HTML.gif
Fig. 2

(a) Result of right renal artery arteriography before ablation. (b) Using 6 F ablation catheter for right renal artery ablation. (c) Right renal artery arteriography after ablation. Results show that right renal artery had no obvious stenosis after ablation

https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig3_HTML.gif
Fig. 3

Example of HE staining of renal artery with or without ablation. (a) Nonablated renal artery for comparison. (b) In the ablated renal artery, vascular endothelial cells injured and vascular wall cells swelling and had mixed appearance

https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig4_HTML.gif
Fig. 4

Example of immunohistochemical staining of renal artery nerves with anti-GAP43 and antibodies with or without ablation. (a) Nonablated ganglion for comparison. (b) In the ablated renal artery, the ganglioinic cells have contracted morphology and lost histological details

3.2 Changes of AERP and initiation of AF

In the control group (Fig. 5), there was a prompt and persistent decrease in refractory periods. AERP decreased from 133.3 ± 18.5 to 117.6 ± 15.0 ms (P < 0.05) during the first 1 h to 101.1 ± 14.1 ms before cessation of pacing. After pacing for 1 h, AF was induced in three dogs and the duration of AF was 6, 20 and 28 s, respectively, by burst pacing (500 beats/min). After cessation of 7-h pacing, AF was induced in six dogs and the duration of AF was 5, 11, 18, 80, 192 and 300 s, respectively, by burst pacing (500 beats/min). The mean induced number of times and duration of AF in each hour were shown in Fig. 6.
https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig5_HTML.gif
Fig. 5

Changes of AERP during atrial rapid pacing in the control group and the RSD group. Results showed that AERP decreased significantly after 7 h of rapid pacing in the two groups

https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig6_HTML.gif
Fig. 6

(a) The mean duration of AF during atrial rapid pacing. (b) The mean induced number of times of AF during atrial rapid pacing. Compared with control dogs, RSD dogs had shorter duration and lower induced number of times of AF after 7-h pacing by burst pacing

In the RSD group (Fig. 5), there was no significant difference in AERP before and after RSD and there were no significant differences in the changes of AERP between the two groups. AERP decreased from 129.0 ± 16.8 to 117.7 ± 22.3 ms (P > 0.05) during the first 1 h to 109.2 ± 17.8 ms before cessation of pacing (P < 0.05). After pacing for 1 h, AF cannot been induced by burst pacing (500 beats/min). After cessation of 7 h of pacing, AF was induced in three dogs and the duration of AF was 5, 34 and 60 s, respectively, by burst pacing (500 beats/min). Compared with control dogs, RSD dogs had lower induced number of times (1.0 ± 1.26 vs 3.14 ± 2.54, P = 0.03) and shorter duration of AF (16.5 ± 25.1 vs 86.6 ± 116.4, P = 0.02) after 7-h pacing by burst pacing (500 beats/min). The mean duration and induced times of AF in each hour were shown in Fig. 6.

3.3 Changes of renin, Ang II and aldosterone

The plasma renin and aldosterone concentration were greatly increased after 7 h of pacing when compared with before pacing in control group (renin, 119.8 ± 31.1 vs 185.3 ± 103.5 pg/ml, P < 0.01; aldosterone, 288.2 ± 43.1 vs 369.6 ± 109.8 pg/ml, P = 0.01). However, there was no significant difference in the levels of plasma Ang II in control dogs between before and after 7 h of pacing (159.6 ± 47.7 vs 189.1 ± 63.9 pg/ml, P = 0.23). Furthermore, the plasma renin, Ang II and aldosterone concentration showed a decreasing trend after 7 h of pacing in the control group, but this did not attain statistical significance (renin, 135.6 ± 45.6 vs 132.1 ± 38.3 pg/ml, P = 0.81; Ang II, 163.8 ± 35.8 vs 156.6 ± 58.5 pg/ml, P = 0.22; aldosterone, 334.7 ± 121.1 vs 327.3 ± 60.1 pg/ml, P = 0.21; Fig. 7).
https://static-content.springer.com/image/art%3A10.1007%2Fs10840-012-9717-y/MediaObjects/10840_2012_9717_Fig7_HTML.gif
Fig. 7

Changes in levels of plasma renin (a), Ang II (b) and aldosterone (c) after 7 h of atrial rapid pacing between the control group and RSD group. There was a significant increase in renin and aldosterone after atrial rapid pacing in control dogs. But the levels of renin, Ang II and aldosterone had non-significant change before and after pacing in RSD group

4 Discussion

4.1 Major findings

This study demonstrated for the first time that episodes and duration of AF could be decreased by renal sympathetic denervation during short-time rapid atrial pacing. Decreased inducibility of AF might have associated with decreased activity of RAAS.

4.2 Effect of intracardiac and extracardiac denervation on inducibility of AF

Initial study on total efferent vagal denervation via epicardial fat pad ablation indicated that such a dedicated approach may prevent AF in dogs produced by atrial pacing [11]. Transvascular atrial parasympathetic nerve system modification by radiofrequency current catheter ablation abolishes vagally mediated AF has been reported [12]. Recently, Tan et al. demonstrated that cryoablation of the superior cardiac branches of vagal nerve eliminated episodes of AF [13]. No matter what the intracardiac ganglionated plexi, extracardiac stellate ganglia or branches of vagal nerve, all these nerves go directly to innervate atrial myocardium. Whether renal sympathetic nerve which has no direct afferent cardiac nerve fibres has impact on AF is not clear.

In the present study, we first observed the inducibility of AF during rapid atrial pacing with or without renal sympathetic denervation. Just as the previous report, rapid atrial pacing can facilitate the initiation and maintenance of AF with the prolonged time [14]. Surprisingly, after renal sympathetic denervation, the average duration and induced episodes of AF were nearly same as that in the control group in the initial 3 h, but average duration and induced episodes of AF decreased significantly after pacing for 4 h. These data highlight the importance of renal sympathetic nerve as an exacerbated factor during rapid atrial pacing.

4.3 RAAS and acute atrial electrical remodelling

Previous studies have demonstrated that high-frequency pacing increases intracellular calcium levels in cardiac myocytes, and intracellular calcium overload is thought to contribute to this phenomenon [1, 15]. Calcium channel blockade prevents AERP shortening during rapid atrial pacing [16]. It has been shown that Ang II increases the intracellular calcium concentration significantly in atrial myocytes in rat heart [17]. Previous study has reported that the inhibition of endogenous Ang II prevented AERP shortening during rapid atrial pacing [10]. Recently, Laszlo et al. demonstrated that selective mineralocorticoid receptor antagonism (eplerenone) influenced ICa,L in the early tachycardia-induced atrial electrical remodelling [18]. Therefore, it is possible that the blockade of RAAS could attenuate atrial fibrillation-induced electrical remodelling, probably by preventing calcium overload. The present study indicated that during atrial rapid pacing, levels of the plasma renin and aldosterone were increased. Although there was no significant difference in the levels of plasma Ang II before and after pacing, the levels of Ang II showed an increasing trend in the control group and decreasing trend in the RSD group. However, in the present study, AERP decreased after rapid pacing in both the two groups. Thus, the relevance of these, at the moment only observational alterations, is likely and has to be further evaluated.

4.4 Renal sympathetic nerve, RAAS and AF

The kidney receives a dense innervation of sympathetic and sensory fibres and can be both a target of sympathetic activity and a source of signals that drive sympathetic tone. It is well-known that renal sympathetic stimulation induces renin release; conversely, Ang II appears to have important actions in modulating sympathetic nerve activity. The RAAS is involved in myocardial fibrosis and increased Ang II production causes marked atrial dilation with focal fibrosis and AF [7, 19]. Under AF, development and persistent sympathetic nerve and RAAS are activated and each condition predisposes to the other. Our study showed that the average duration and induced episodes of AF were longer significantly in control dogs than in renal sympathetic denervation dogs at 7 h after cessation of pacing. This phenomenon might be explained by reduced ‘RAAS activity’ due to renal sympathetic denervation.

4.5 Catheter-based renal denervation as a new strategy for AF management

Despite a long history of medical exploration of AF, many aspects of the management of AF remain controversial [20]. The long-term efficacy of currently available antiarrhythmic drugs for the prevention of AF recurrence is far from ideal because of limited efficacy and potential side effects, particularly proarrhythmia [21]. Since Haissaguerre et al. reported that triggers arising from the pulmonary veins are responsible for the majority of paroxysmal AF [22], catheter ablation of AF is now a realistic therapeutic option across a broad spectrum of patients—from patients with paroxysmal AF to those with long-lasting persistent AF [23]. Although the superiority of catheter ablation over antiarrhythmic drug therapy has been demonstrated in middle-aged patients with paroxysmal AF, the role of the procedure in other patient subgroups such as those with long-standing persistent AF has not been well-defined [24, 25]. Currently, ganglionated plexi (GP) ablation has been employed for both paroxysmal and persistent AF, and GP ablation appears to be an efficacious method to improve pulmonary vein (PV) isolation in patients with AF [26, 27]. However, ablation-induced atrial or ventricular proarrhythmia has been reported and anatomic GP modification appears to carry a higher risk of iatrogenic left atrial tachycardias than PV isolation [28, 29]. Furthermore, studies have shown that fat pad ablation did not achieve long-term suppression of AF induction in canine model [30, 31]. We demonstrated in the present study that inducibility of AF can be decreased by renal sympathetic denervation during short-time rapid atrial pacing. The effectiveness and safety of catheter-based renal denervation had been demonstrated in patients in recent years [32, 33]. The present findings further support the assertion that ablation of extracardiac nerves can be an effective alternative to ablation of intracardiac nerves by atrial ablation for the treatment of AF [13].

5 Study limitations

The present study has several limitations. First, we did not observe the long-time effect of renal sympathetic denervation on inducibility of AF and atrial substrate remodelling because renal sympathetic nerve has no direct afferent nerve fibres to dominate cardiac function. Whether renal sympathetic denervation downregulated the activation of RAAS then decreased the generation of renin and Ang II and delayed the progress of atrial fibrosis and remodelling should be further investigated. Second, we did not perform stimulation of renal sympathetic nerves; therefore, we could not determine the relative importance of renal sympathetic nerve activation and inducibility of AF. Third, except for RAAS, renal sympathetic nerve activity has effect on the release of other neurohormones, such as bradykinin, prostaglandin, etc. Whether there are changes of plasma neurohormones after renal sympathetic denervation and have effect on atrial remodelling are unknown. Finally, the intracellular calcium concentration was not measured. Thus, further experiments with single cells isolated from atrial myocardium will be needed to evaluate the effect of RSD on intracellular calcium overload during rapid atrial pacing.

Acknowledgments

We thank Dr. Hongyao Hu for his support; Xiaohong Wang and Huafen Liu for their assistance; and Yu Liu for providing the multi-polar electrode catheters used in this study. This work was supported by National Natural Science Foundation of China (81070144).

Conflict of interests

The authors have no conflict of interests to disclose.

Copyright information

© Springer Science+Business Media, LLC 2012