Cardioneuroablation: Catheter Vagal Denervation as a New Therapy for Cardioinhibitory Syncope

The vasovagal syncope is the most frequent cause of transient loss of consciousness, especially in young people without significant heart disease. The malignant cardioinhibitory form is caused by abrupt and intense vagal reflex with or without defined triggers. Refractory cases to preventive measures and pharmacological handling has been treated with definitive pacemaker implantation. Besides showing questionable results, pacemaker implantation is highly rejected by young patients. In the late 1990s, we proposed specific vagal denervation by catheter ablation and spectral mapping, for paroxysmal AF, functional bradyarrhythmias and severe cases of malignant cardioinhibitory syncope giving rise to cardioneuroablation. Recently, many authors worldwide have been reproducing the cardioneuroablation results where elimination or significant reduction of the vagal response were observed, which abolished symptoms in more than 75% of patients followed up to 14 years, without complications. Therefore, cardioneuroablation has shown to be a real therapeutic option in malignant syncope cardioinhibitory and in any exclusive vagal mediated bradyarrhythmia without the need for pacemaker implantation.


Vasovagal syncope
Also known as reflex, neurocardiogenic or vasovagal syncope, it is caused by a reflex mechanism, through multiple triggers, ending with intense cardioinhibition and/or vasodilatation leading to severe and transient hypotension. The cardioinhibition can be enough to cause prolonged asystole. Typically, the patient recovers spontaneously within minutes. Depending on the physiopathology, it can be classified as cardioinhibitory, vasodepressor or mixed. The first is characterized by asystole, severe bradycardia and/or transient total AV block caused by intense vagal action. In the vasodepressor form, severe hypotension without significant bradycardia usually occurs, mainly due to a sudden reduction in sympathetic tone 2 . In the mixed form, there is the contribution more or less intense of the two mechanisms.

CARDIAC AUTONOMIC NERVOUS SYSTEM Anatomy
The autonomic nervous system of the heart consists of three major divisions, one afferent -the sensory nervous system -and two efferent partitions, the parasympathetic and sympathetic branches. The medulla oblongata is the main center for integration of cardiac innervation whose activity is modulated by the hypothalamus and more superior centers. The sensory fibers are bipolar neurons whose cells bodies are in the medulla oblongata. The efferent fibers comprise essentially two main neurons, the pre and the post-postganglionic. The parasympathetic postganglionic fiber is very short because its body neuron is located in the heart, mainly in the atrial wall and in the ganglionated plexuses (paracardiac fat pads) 3 . The cell body of the parasympathetic preganglionic neuron placed is in the medulla oblongata, more specifically in the nucleus ambiguous and in the vagus dorsal motor nucleus 4 . Its axon is led to the heart by the vagus nerves.
In contrast, the sympathetic postganglionic neuron is too long because their cell body is sited in the paravertebral sympathetic chain. The result of this distribution is that only the parasympathetic postganglionic body neuron is located in the heart and only this one is prone to be eliminated by the endocardial RF ablation.

Physiology
Despite being the heart a striated muscle, its activity does not depend on the innervation. Unlike the skeletal muscle that atrophies when denervated, the heart retains its normal metabolism, structure and activity independent on the innervation. This fact is easily observed in cardiac transplantation patients. However, the autonomic nervous system acts permanently modulating all cardiac properties, through an intense inhibitory (parasympathetic) and excitatory (sympathetic) tone.
This functional antagonism creates a balance mediated by the brain stem that determines the instantaneous heart rate. This balance is constantly restored by the autonomic nervous system by adjusting the cardiac physiology at every moment. Thereby, in the case of vagal denervation, a proportional reduction in the sympathetic tone determined by the reflex balance of the autonomic nervous system is usually observed, bringing down the cardiac rate back to baseline.

Rationale
The cardioinhibition of the reflex syncope is caused by a massive sudden vagal reflex that may be totally eliminated by vagal denervation. That is also elegantly proved by preventing the cardioinhibitory syncope by  10,11 . Unfortunately, this kind of denervation has not shown positive results 12 . In contrast, the vagal denervation to treat vasovagal syncope appears to be highly efficient.  *Fibers of the sympathetic and sensory systems usually recover after RF. Different from the sympathetic and sensory systems, whose soma are far from the heart, the postganglionic parasympathetic body neuron is over or inside the atrial wall and susceptible to be destroyed by the endocardial RF energy resulting in a long-term denervation 8 . RF: Radiofrequency energy limit.

RF catheter ablation for vagal denervation
The properties of the atrial wall and may be identified by changes in the spectrum as "AF Nests" [22][23][24][25] (this name resulted from its relation with the AF physiopathology as they present electrical resonance 26,27 favoring the AF maintenance).
Therefore, by using the spectral study during sinus rhythm, we have found two types of myocardium. The first one, the compact myocardium, is characterized by high amplitude, isotropic conduction and a smooth spectrum (Fig. 3a), whereas the second one, the f ibrillar myocardium, has a low amplitude, anisotropic conduction and a segmented spectrum ( Fig. 3b) 26,28,31 .
Clusters of fibrillar myocardium give rise to AF Nests.

Mapping and ablating the cardiac innervation by endocardial approach
Unlike the skeletal muscle, the myocardium does not have a differentiate neuromuscular junction. In contrast, the nerve parasympathetic and sympathetic fibers directly penetrate the myocardium and interweave with the myocytes. Additionally, lots of microneurons (parasympathetic postganglionic neurons, first neuron in Fig. 2) colonize the atrial walls Notes: (a) Compact myocardium and (b): Fibrillar myocardium. 1: myocardium histology sketch, 2: conduction scheme, 3: "time domain" endocardial potential, 4: "frequency domain" endocardial potential (spectrum). The cells of the compact myocardium are very well connected with high connexins density, represented by small blue bars between cells (1a). This structure causes an isotropic (homogeneous) conduction and a smooth spectrum, like the conduction in one cell (4a). On the contrary, the invasion of the nervous fibers into the myocardium and the presence of numerous microneurons (2b) change the cell connections, causing anisotropic conduction (2b) even without fibrosis (Type I AF Nest). Thus, the spectrum is segmented, showing several groups of frequencies (4b). In this case, the conduction is heterogeneous like that one in a bunch of cells (type I AF Nests). In a normal heart, the compact myocardium presents low-density innervation compared to the normal fibrillar myocardium that presents high-density innervation. Another kind of fibrillar myocardium, not considered here, may be caused by pathological conditions like fibrosis, degeneration, ischemia, infiltration, inflammation (Type II AF Nest) Mateos JCP, EIP, Higuti C, Lobo TJ, Peña TGS, Pachón CTC, Mateos JCP, Acosta JCZ, Ortencio F, Amarante R • Detailed studies based on neural staining have shown a high number of parasympathetic neurons and ganglia in the fibrillar myocardium areas 31-33 ; • A large amount of fibrillar myocardium is found in the anatomical regions of the cardiac ganglionated plexuses 26,27,31 .  Through the on-line real-time spectral mapping, it is possible to disclose the fibrillar myocardium (AF Nests) to guide the ablation (Fig. 3Bb4) of the first neuron.
Consequently, most of the postganglionic parasympathetic neurons may be abolished and do not recover, whereas the sympathetic and sensory terminal fibers usually recover from weeks to months (Fig, 2). Preganglionic vagal fibers may provide some grade of reinnervation but it is reduced as they have lost the postganglionic link replaced by some post-RF fibrosis.

Inclusion criteria
The success of the procedure depends on a strict inclusion criterion. The main one is the presence of a reflex and / or functional bradyarrhythmia in a symptomatic patient, without response or without the possibility of clinical treatment 35 in an apparently normal heart or having rationally excluded a significant cardiopathy 36 . Likewise, pharmacological test to confirm the reversibility of the condition, as a positive response to the atropine, are decisive, Table 1.

Method
Since the patient has signed the consent form, the procedure is performed under general anesthesia.
An important concern is that, inadvertently, the anesthesiologist may use atropine at the beginning of the procedure and that would make cardiologists miss the autonomic tone parameter; so, it is essential to warn the anesthesiologist team to discuss before they use any autonomic drug. The vital signs (heart rate,

RF guided by spectral mapping or by endocardial potentials
The patient must be in sinus rhythm and the right and left atrial endocardium are scanned with conventional irrigated catheter with the thermocontrolled RF generator.
RF is applied in all places featured as AF Nest (fibrillar myocardium) (Fig. 6) aiming the elimination of the first neuron. Scanning may be guided by spectral mapping (Fig. 7) or by conventional recordings whit some filtering settings (Fig. 8).    fibrillar myocardium by conventional recordings (b1-3) and by spectral analysis (b4). Typically, the time from the beginning of a1 to the end of A3 is less than 30 ms while the same measurement from b1 to b3 is more than 30 ms. That may be an easy way for identification of the fibrillar myocardium without on-line spectral mapping.

RF guided by anatomical landmarks
After ablation of all fibrillar scanned myocardium, endocardial anatomical ablation is also performed (at least 2 min in each place), in the areas related to the epicardial ganglionated plexuses (GP) 37 . These areas typically present a high number of AF Nests and allow the elimination or depopulation of the second neuron.
The RF in these areas should be extended for getting to the pulmonary vein insertion 34 . Ancillary ablation of these sources may be obtained to a greater or lesser extent during pulmonary vein isolation (Fig. 7).

Methodology for controlling and confirm the vagal denervation
In this procedure, it is absolutely essential to have  1:GP located between the superior vena cava and the aorta; 2: GP located between the superior pulmonary and fossa ovalis from the left atrium and from the fossa ovalis to the Waterston's groove; 3: GP located between left atrium, inferior vena cava and coronary sinus 8 .

Immediate cardioneuroablation endpoints
Several endpoints must be considered to enhance the results of CNA, as shown in Table 2.

Esophageal protection during cardioneuroablation
The wider the AF Nests ablation the greater and more lasting will be the vagal denervation. In addition to the regions of GPs, large vagal innervation enters the atria by the pulmonary veins insertion. Therefore, it is highly desirable to eliminate the AF-Nests related to these areas like the atrial fibrillation ablation. In this sense, special care for the protection of the esophagus is necessary. For this purpose, we developed a method, applied to all patients included in this study, which is the mechanical displacement of the esophagus by using the transesophageal echocardiographic transducer 39 . Usually, the displacement is enough to move the esophagus 4 to 8 cm in the opposite direction of the RF spot, expressively reducing the risk of esophageal heating and lesion (Fig. 13).

RESULTS OF CARDIONEUROABLATION
The immediate result at the end of CNA is the complete absence of vagal response indicating a successful vagal denervation (Fig. 12b). In fact, this is the main success criterion and if not achieved the ablation must be resumed and expanded until getting complete vagal response elimination.
The long-term outcome should be assessed primarily by the clinical follow-up, but it is required to be tried with the same assessment that demonstrated the most important cardioinhibitory response at the inclusion time, more often, the Tilt-test (Fig. 14). However, especially in cases of nonreflex functional bradyarrhythmias, the cardioinhibition may have been detected by stress-test or Holter recording.
The Tilt-test is undoubtedly the most commonly used trial as an inclusion and control criterion. It is recommended to be repeated from 2 months post-CNA using the same protocol of the inclusion phase (Fig. 15).
In a long-term cohort study of 43 cardioinhibitory patients, with a FU of 45.1±22 -11 to 91.4 months 41 , it was observed positive control Tilt-test only in 4 cases (9.3%), with the same degree of cardioinhibition but, all of them, having a significant vasodepressor response, Table 3.
These results show that the low number of patients presenting syncope after CNA appear to shift the vasovagal behavior from the severe cardioinhibition to a predominant less important vasodepressor response as the procedure significantly decreases the cardioinhibition response even in the long-term phase.

Cardioneuroablation survival curves
The therapeutic value of the CNA 8 may be evaluated by comparing the results with the clinical treatment 42 and with pacemaker implantation in the ISSUE-3 3 and SYNPACE 37 studies. In Figure 16 The CNA can bring undesirable consequences?
Strictly follow-up of our patients over 14 years has shown no undesirable effect, even in the few cases in which the procedure was remade.
In the initial phase, there is a noteworthy change in the autonomic tone commonly observed as persistent sinus tachycardia. Because of this, it is customary to keep the patient with beta-blockers in the first 2 to 3 months. Clinically, there is a clear reduction or elimination of parasympathetic tone with a predominance of sympathetic one. However, gradually, there is a natural autonomic rebalancing with gradual reduction of the sympathetic drive bringing the heart rate to normal values. In this sense, there is a beneficial result of a long-time natural reduction of the sympathetic tone (Fig. 17a).
The patient shown in Fig. 15 had a basal heart rate of 62 ppm in the pre-CNA tilt test, 85 ppm in the 4 months control and 80 ppm in tilt-test one-year post-CNA. However, due to the readjustment of the autonomic nervous system, the patients usually remain asymptomatic with normal life. Due to this autonomic plasticity, the chronotropic response and the exercise capacity is also fully preserved as can be seen in the study of the post-CNA exercise test, in which there was not observed additional chronotropic incompetence (Fig. 17b). One of the problems that must be questioned is if the reduction of RR variability may increase the cardiovascular risk.
This effect was demonstrated after myocardial infarction and myocardial damage but there is no evidence that the primary RR variability reduction caused by vagal denervation with fully preserved myocardial has some consequence in this regard.

COMPLICATIONS
With standard care, the CNA has been a very safe procedure. Its complication rate is equivalent to the ablation of paroxysmal AF in patients without heart disease. In a study of 44 CNA performed in 43 patients 8 (Fig.15). One last limitation is that the CNA is an operator-dependent procedure with result highly related to the learning curve.
Thus, controlled multicenter randomized trials should wait for the appropriate training of various services.
(a) Comparison between the minimal, mean, maximal heart rates and heart rate variability (SDNN) as assessed by Holter monitoring pre and 14.4±12 months post-CNA. There is a slight tendency to a discreet increase in the mean heart rate that is progressively reduced with physical conditioning. The SDNN shows an evident long-time reduction of the parasympathetic tone. (b) Comparison of stress-tests pre and 9.2±6 months post-CNA.   Min. HR (bpm)
In the studied cohort, a pacemaker was not necessary in any case. Despite no prevention of the vasodepression the CNA seems to cause enough long-term vagal reflex attenuation, eliminating the cardioinhibition, and keeping most patients asymptomatic. By using the current atrial fibrillation ablation technology, the procedure is safe, feasible, and reproducible. The indication is simple and essentially based on clinical findings, in the presence of severe cardioinhibition unresponsive to medical treatment, and on the normal response to atropine. Vagal stimulation to control the extent of the denervation in addition to confirming the immediate success seems to be decisive and indispensable. At last, as a rule for any new therapy, the results must be confirmed with experimental protocols and by randomized studies.