Atrial fibrillation (AF) is the most common cardiac rhythm disorder and contributes to thromboembolism. The presence of AF is an independent risk factor for thromboembolism; especially stroke in association with AF increases mortality and morbidity, leading to greater disability, longer hospital stays, and worse quality of life [
The cardiac ANS consists of extrinsic and intrinsic components [
Ablation of the autonomic substrate suppresses or eliminates focal AF originating from PVs [
Low-level VNS (LLVNS) is being explored as another strategy. Strong-level VNS that produces >60% prolongation of the sinus cycle length facilitates AF; it produces long pauses and even sinus arrest, which are generally required to induce and maintain AF [
Based on all these reported results, we hypothesized that the antiarrhythmic effect of LLVNS is brought about by elimination of the heterogeneous substrate surrounding the GP areas, which is induced by rapid atrial pacing (RAP). This study was specifically designed to test the previous hypothesis in dogs with acute electrical remodeling.
The Institutional Animal Care and Use Committee of JINAN University of Experimental Animal Management Centre reviewed and approved the design of all animal experiments. All animal studies were reviewed and approved by the animal experimental administration of JINAN University of China. A total of 20 adult mongrel dogs weighing 13–17 kg were anesthetized with sodium pentobarbital (initial bolus, 30 mg/kg body weight, i.v.), with an additional dose of 2 mg/kg given at the end of each hour. All dogs were ventilated with room air using a positive pressure respirator, and oxygen saturation was maintained at 95–100%. The animals were fixed on an operating table, the temperature of which thermostatically controlled at 37°C. The chest was entered via bilateral thoracotomy at the bilateral fourth intercostal space. Several 10-bipolar electrodes were sutured using a noninjurious method to allow recording and stimulation at the left superior pulmonary vein (LSPV), left inferior pulmonary vein (LIPV), left appendage (LAA, catheter: LA1,2), left atrium (LA, catheter: LA5,6), area around the Marshall ligament (MSL, catheters: LA9,10) [
Position of catheters in the atrium. 1, anterior right ganglionated plexi (ARGP); 2, inferior right ganglionated plexi (IRGP); 3, right superior pulmonary vein (RSPV); 4, right inferior pulmonary vein (RIPV); 5, right appendages (RAA); 6, right atrium (RA); 7, margin of anterior right ganglionated plexi (ARGPM); 8, superior left ganglionated plexi (SLGP); 9, inferior left ganglionated plexi (ILGP); 10, left superior pulmonary vein (LSPV); 11, left inferior pulmonary vein (LIPV); 12, left appendages (LAA); 13, left atrium (LA); 14, Marshall ligament area (MSL); 15, superior vena cava (SVC); 16, aorta; 17, pulmonary artery (PA); 18, right ventricle (RV); and 19, left ventricle (LV).
Standard ECG and electrophysiological channels (IECG) were continuously recorded and filtered at 0.05–100 Hz and 200–1200 Hz, respectively. All tracings from the electrode catheters were amplified and digitally recorded using a computer-based electrophysiological system (Lead2000B; Jingjiang Inc., China).
In both groups, RAP was delivered (1200 bpm, 4 V, 1 msec in duration) at the LA. After each pacing hour, RAP was temporarily stopped to measure the electrophysiological data.
Programmed stimulation was performed with Lead2000B stimulator. Regular atrial pacing as S1-S1 interval was set at 330 msec, ERP was determined using S1-S2 programmed stimulation (S1-S1 : S1-S2 = 8 : 1), which decreased from 150 msec by 10 msec decrements; as the S1-S2 intervals approached the ERP, the last S1-S2 interval increased by 8 msec, and the decrement was reduced by 1 msec again, until precise ERP was reached. ERP dispersion (ERPD) was defined as the coefficient of variation (standard deviation/mean) of the ERP at all 8 or 10 sites [
The WOV was used as a quantitative measure of AF inducibility. During ERP measurements, if AF or >2 echo was induced by decremental S1-S2 stimulation, the difference between the longest and the shortest S1-S2 interval was designated as the WOV. The
The stimulation of the atrium was delivered at the LA site with S1-S1 intervals set at 100 msec and sustained for 10 seconds, which induced atria tachycardia or AF easily when it was continued or stopped. The recovery time (RT) was represented as the duration from AF triggered by persistent S1-S1 stimulation to sinus rhythm.
Bilateral cervical vagosympathetic trunks were decentralized by surgical procedures. Two pairs of electrodes insulated with surrounding tissue were embedded in the vagosympathetic trunks located adjacent to the cervical artery for LLVNS [
Low-level vagosympathetic nerve stimulation (LLVNS) during rapid atrial pacing (RAP). The voltage of LLVNS was 0.25 V (VNS channel). RAP was delivered at the LA site. RA3,4 channel recorded the signal of RAP and LLVNS.
Twenty dogs were randomly divided into the experimental group (
All values were expressed as the mean ± standard deviation of the mean. Paired
The systolic and diastolic blood pressures were stable during the entire experimental period, with no sign of heart failure throughout. Oxygen saturation was maintained at 95–100% during the experiments.
In the control group, the ERP at RSPV, RIPV, RA, RAA, ARGPM, LSPV, LIPV, LA, LAA, and MSL was markedly shortened in the second and third hour (Figure
Effects of LLVNS on ERP and its gradient. LLVNS reversed the decrease in ERP and increase in ERP gradient that was induced by RAP. #
In contrast, in the experimental group, the ERP of each site of the atrium showed no significant variance at the end of each hour, compared to baseline levels. Moreover, ERP gradients were not observed at the end of each hour, as observed at the baseline.
In the control group, the WOV at each site of the atrium showed a significantly progressive increase during each hour of RAP (Figure
Effects of LLVNS on WOV and its gradient. LLVNS reversed the increase in WOV and its gradient caused by RAP. WOV: window of vulnerability; the remaining symbols and abbreviations are the same as those used in Figure
In the experimental group, the WOV at each site of the atrium was not different between each period of RAP, except for RA at the third period (
In the control group, ERPD1 and ERPD2 showed progressive increase and reached statistical significance in the second and third hour of RAP (Figure
Effects of LLVNS on ERP dispersion. LLVNS reversed the effects of RAP on ERP dispersion. The blue (
The recovery time (RT) was represented as the duration from atrial arrhythmia that is facilitated by fast pacing triggers in atrium to sinus rhythm. In control group, RAP remodeling progressively increased RT (Figure
The effect of LLVNS on recovery time. The recovery time (RT) was represented as the duration from AF triggered by persistent S1-S1 stimulation to sinus rhythm. Abbreviations are the same as those in Figure
Vagosympathetic nerve stimulation (VNS) promoted atrial arrhythmia. VNS with the voltage at 0.5 V triggered atrial premature and then facilitated AF. AP: atrial premature. AF: atrial fibrillation.
VNS maintained AF. The AF was persistent during VNS with the voltage at 0.5 V. While the VNS was removed, AF still maintained. AF: atrial fibrillation.
LLVNS cannot maintain the AF. When RAP that is persistent for 3 hours concomitant with LLVNS (0.25 V), was stopped, the AF progressively terminated even though the LLVNS still continued. RAP: rapid atrial pacing. SR: sinus rhythm.
We have successfully proved our hypothesis about the benefits of LLVNS for treating AF. This is clear from the results, which show that application of LLVNS to canine models of RAP remodeling markedly reversed the decrease in ERP and increase in WOV at each site. Moreover, it abolished gradient differences in the electrophysiological substrate surrounding GP and thus maintained ERP dispersion at baseline levels. It also reduced the RT from AF.
In our research, as we measured the threshold, the VNS with the voltage at 0.5 V triggered atrial premature and then facilitated AF at baseline. The AF was persistent during the VNS. While the VNS was removed, AF still maintained. These results suggest that vagosympathetic trunks as extrinsic components play an important role in the initiation and maintenance of AF. The arrhythmic effect of VNS may be mediated by activating the “integration centers of GP” [
The hyperactivity of the autonomic element in MSL may contribute to the initiation of AF and even ventricular tachyarrhythmia [
In the RAP remodeling, when electrical remodeling occurs concomitantly with autonomic remodeling, it is believed to indicate progressive enhancement of neural activity [
It has been reported that LLVNS that is 10–50% below the threshold voltage, required to slow the sinus rate or atrioventricular conduction, may prevent episodic AF caused by rapid PV and non-PV firing, due to a progressive increase in AF threshold at all PVs and atrial appendages sites, particularly RSPV, RIPV, LSPV, and RAA. Moreover, this type of antiarrhythmic effect is not dependent on the activation of the afferent vagal nerve fibers that project to the brain [
Based on our results, the mechanism of action of LLVNS may be explained as follows: (1) suppression of the neural activity of GP, which abolishes gradient differences in the substrate surrounding GP and (2) enhanced recovery of the homogeneous substrate with baseline susceptibility to AF. In addition, there are other mechanisms that have been proposed by researchers: (3) decrease in transient intracellular Ca2+ levels, owing to reduced release of sympathetic neurotransmitters via inhibition of activity of the stellate ganglion [
A limitation of this study is that we have no direct evidence indicating inhibition of neuronal firing within GP and MSL, and the results are therefore limited to short-term RAP and LLVNS. This hypothesis will have to be verified in the future by using RAP and LLVNS of longer duration. This will help evaluate the antiarrhythmic effects of LLVNS in chronic AF models. Moreover, research will be required to determine the optimal parameters and sites for LLVNS that will have the greatest degree of AF inhibition with minimal side effects. We hope to devise future treatment methods to treat chronic AF, in which autonomic nerve stimulators can be inserted using noninvasive methods.
LLVNS can reverse the electrical and autonomic remodeling induced by RAP. The mechanism involves abolishing gradient differences in the substrate surrounding GP and recovery of the homogeneous substrate.
The authors takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
The authors declare that they have no conflict of interests.
The authors thank Professors Lianjun Gao, Yanzong Yang, Yunlong Xia, Shulong Zhang, and Dong Chang of the First Hospital Affiliated to the Dalian Medical University for their assistance. They thank Jingjiang Electronic Science and Technology Company, Inc. (Chengdu, China) and APT Medical Company, Inc. (Shenzhen, China) for providing the electrophysiological catheters and technical assistance. This work was supported by Grants from the Fundamental Research Funds for Central Universities (no. 21611333), the Science and Technological Program of Guangdong Province (no. 2010-1096-136; 2011B031800336), the Science and Technological Projects of Guangzhou (no. 2010Y1-C941), and the Key Disciplines’ Funds and Special Research Funds (no. 2012207) of the First Clinical Medical College of Jinan University, Basic Research Expenses of Jinan University.