Prospective non-randomized study of patients with recurrent VVS of a predominantly cardioinhibitory nature who underwent CNA in two Portuguese hospitals from January 2019 to April 2022. Patient data at the time of the procedure and at follow-up were collected to assess acute and follow-up outcomes.

All patients received detailed information on the methodology, benefits, and risks of the procedure, and signed an informed consent in accordance with the local ethics committee. The study was conducted according to the Declaration of Helsinki.

Inclusion criteria were: (i) a history of recurrent syncope with a predominant or exclusive cardioinhibitory component as determined by the occurrence of cardioinhibition during electrocardiogram (ECG) monitoring, implantable loop recorder or a head-up tilt test with symptom reproduction; (ii) refractoriness to conventional measures including optimal fluid intake, dietary advice, physical counterpressure training and pharmacological therapies whenever a vasodepressor component was associated; (iii) no other causes of loss of consciousness apparent based on the clinical history and initial study, including a 12-lead ECG, echocardiogram and 24-hour Holter.

Exclusion criteria included a previous history of heart surgery or pacemaker implantation, significant structural cardiomyopathy, or an additional indication for pacemaker implantation such as sinus node dysfunction or advanced AV conduction abnormalities. Vagally-mediated bradycardia was distinguished from conduction system disease based on the clinical history (age, context where syncope occurred, prodrome and triggers) and results of treadmill stress test, and 24-hour Holter.

All procedures were performed under general anesthesia with intravenous propofol, fentanyl and occasionally rocuronium. Vital signs were monitored during the procedure. Brain function was monitored by BIS Aspect A-1000. A decapolar catheter was positioned in the coronary sinus and further advanced into the right internal jugular vein up to the jugular foramen for extracardiac non-contact vagal stimulation. Another quadripolar catheter was placed in the atria to perform atrial pacing. Intravenous heparin was administered after a single transseptal puncture guided by fluoroscopy to maintain an activated clotting time above 300 seconds. Intracardiac echocardiography was used at the discretion of the operator.

The three-dimensional (3D) geometry of the left atrium and pulmonary veins was constructed using a multipolar mapping catheter (PENTARAY®) with a 3D navigation system (CARTO-3, Biosense Webster Inc., Diamond Bar, CA). After electroanatomic map construction, extracardiac vagal stimulation was performed using the decapolar catheter placed in the internal jugular vein and a stimulator that releases square wave pulses of 50 microseconds width, frequency of 50 Hz and amplitude from 10 to 70 V, adjusted per patient weight (1 V/kg up to 70 V).21 Baseline response to vagal stimulation with or without atrial pacing was recorded. A positive vagal response was defined as transient asystole, severe sinus bradycardia and/or atrioventricular block during stimulation. A mixed sinus and AV nodal response was defined as the occurrence of asystole or severe sinus bradycardia, plus AV block (i.e., non-conducted p waves) with atrial pacing.

According to the type of rhythm disturbance triggered by extracardiac vagal stimulation, and also that which occurred clinically, sequential, and anatomically guided GP ablation was directed toward the ganglia most likely responsible. For patients with asystole or severe sinus bradycardia, ablation always started at the right anterior GP (RAGP), at the superoanterior area near the root of the right superior PV. If positive vagal response persisted, ablation continued, whenever necessary: (i) at the left superior GP (LSGP), located in the superolateral area around the root of the left superior PV and between the left superior PV and the left atrial appendage, and at the left lateral GP (LLGP) between the LAA and the left superior and left inferior GPs (more likely to affect AV nodal response); (ii) at the right inferior GP (RIGP) at the postero-inferior area near the root of the right inferior PV, at the postero-inferior border of the interatrial septum and posterior to the CS ostium (more likely to affect AV nodal response); (iii) at the right superior GP (RSGP) located between the superior vena cava and the aortic root just above the right superior pulmonary vein, at the postero-medial SVC-right atrial junction (more likely to affect sinus nodal response) (Figure 1). The left inferior GP area, at the infero-posterior border of the left inferior PV, was typically avoided to reduce the risk of esophageal lesion. RF ablation was performed using an irrigated catheter with contact force sensor (THERMOCOOL SMARTTOUCH™ SF Catheter) and guided by the ablation index, ranging between 380 and 500, according to the location. Extracardiac vagal stimulation was repeated after each GP group ablation to assess the degree of parasympathetic denervation. The elimination of cardiac parasympathetic response to vagal stimulation (i.e., elimination of asystole, severe bradycardia, or AV block) was the acute endpoint of the procedure.

Figure 1.

The 3-dimensional representation of the main ganglionated plexus (GP) location. Upper panel: (A) RAGP between the right upper pulmonary vein and the right atrium; (B) RSGP between the superior vena cava and aortic root just above the right upper pulmonary vein; lower panel: (C) RIGP at the roof of the coronary sinus and medial and just under the tricuspid valve and the coronary sinus ostium; (D) LSGP at superolateral area around the root of the left superior pulmonary vein.

LSGP: left superior ganglionated plexus; RAGP: right anterior ganglionated plexus; RIGP: right inferior ganglionated plexus; RSGP: right superior ganglionated plexus.

(0.35MB).

After the procedure, patients were observed and underwent continuous ECG monitoring and transthoracic echocardiogram before discharge (24 hours).

The primary endpoint was the occurrence of syncope during follow-up. Prodromes including dizziness, fatigue or diaphoresis were documented but not considered a treatment failure.

All patients received oral anticoagulation with direct oral anticoagulants for at least one month after the procedure. Regular clinical post-ablation follow-up was performed individually by each physician. Clinical data and medical examinations were obtained through the analysis of the national clinical registry and through a phone call to the patient.

Continuous data are expressed as mean±SD or median±interquartile range. Categorical data were expressed as percentages. SPSS version 26.0 (IBM) was used for statistical analysis.22

A total of 21 ablation procedures were performed in 19 patients fulfilling the inclusion criteria (68% males, mean age 37.8±12.9 years, range 19–72 years). Baseline clinical characteristics of the study population are presented in Table 1. All patients had documented cardioinhibitory response by head-up tilt test (n=6), 24-hour Holter (n=6), implantable loop recorder (n=4) or electrocardiographic monitoring during hospitalization (n=3). The mean number of syncopal episodes per patient in the year before the procedure was 2.8 (ranging from 1 to 10), with five syncopal episodes occurring in four patients resulting in significant trauma, which included three intracranial hemorrhages requiring hospitalization, two jaw fractures requiring surgery and one rib fracture. One patient (case H) had hypertrophic cardiomyopathy with no evidence of outflow tract obstruction or ventricular systolic dysfunction, while the remaining patients had no structural heart disease. One patient (case B, 72 years old) underwent pulmonary vein isolation simultaneously as he had a history of symptomatic vagally-induced paroxysmal atrial fibrillation. Patient B had a history of recurrent VVS with documented cardioinhibition, as well as atrial fibrillation with onset either during the night or during/following an episode of vagally-induced asystole. There were no data suggesting conduction system disease (i.e., no clear indication for pacemaker implantation), based on a normal ECG and good heart rate response in a treadmill stress test and a 24-hour Holter.

Detailed characteristics of the procedures are presented in Table 2. The mean procedure and fluoroscopy times were 115±30 min and 8.8±5.1 min, respectively. A positive mixed vagal response with extracardiac vagal stimulation was seen in all patients (asystole, and atrioventricular block when performing atrial pacing during vagal stimulation). The acute endpoint of the procedure (inhibition of vagal response after GP ablation) was achieved in all patients. Figure 2 shows an example of an initial mixed sinus and AV nodal vagal response and the result after successful CNA.

Figure 2.

(A) Intracardiac electrogram showing a sinus pause of 7.28 s (orange arrow) induced by vagal stimulation for 4 s (red arrow) before cardioneuroablation. (B) Intracardiac electrogram showing an atrioventricular block (white arrow) induced by vagal stimulation during atrial pacing. (C) Intracardiac electrogram showing abolition of vagal response after cardioneuroablation.

(1.53MB).

Considering the division into four main groups of parasympathetic ganglia plexus, an average of 2.6 ganglia plexus were ablated per procedure: RAGP in 20 patients, RSGP in 12 patients, RIGP in 8 patients and LSGP in 14 patients.

One patient (case A.2) had a convulsive episode immediately after the second procedure, requiring admission to intensive care. A cranioencephalic computed tomography showed no abnormalities and an embolic event was excluded. This episode was interpreted as an adverse reaction to sugammadex used for rocuronium reversal, with no neurological sequelae. No other complications were observed, including the need for pacemaker implantation, pericardial effusion/tamponade, or vascular complications. All patients were discharged the day after the procedure.

All patients maintained regular clinical follow-up. One-third had an implantable loop recorder. In a mean follow-up of 21.0±13.2 months (3–42 months), only two patients (11%) had recurrence of syncope. The burden of syncope was reduced following ablation compared to the year preceding it (2.8 vs. 0.3 episodes per patient, p<0.001) (Figure 3).

Figure 3.

Comparison of number of syncope episodes before and after cardioneuroablation.

(0.11MB).

Patient A, who had had five syncopal episodes in the year preceding ablation, initially underwent ablation in the RAGP and RSGP. After six months free of symptoms, this patient had a recurrence of syncope, although the underlying mechanism could not be documented. Patient D was asymptomatic for nearly one year after the procedure, at which time the patient had recurrent syncope with complete atrioventricular block and a 19-second pause, documented by an implantable loop recorder. Both patients underwent a second CNA procedure and an initial positive vagal response was again observed. The same method with sequential and anatomically guided GP ablation was performed with acute success. However, both had further syncopal episodes and therefore underwent pacemaker implantation. These cases will be discussed in more detail below.

One third of patients reported being aware of a higher baseline heart rate in the first weeks after the procedure, with mild palpitations and fatigue with exertion which subsided gradually and completely over time.

First, this study included a small number of patients without a control group. Still, it represents an additional source of literature suggesting that CNA may be more effective than other existing therapies. Very recently, the first randomized study documenting efficacy of CNA in patients with cardioinhibitory VVS was published.31 Second, the confirmation of cardioinhibitory syncope was made using multiple methods, including tilt table tests. However, the documentation of a spontaneous cardioinhibitory syncope might be a better way to select patients than using tilt tests. Finally, follow-up was not systematic, with some variability among patients: only one third had implantable loop recorders, and appointments at the clinic were not always carried out due to Covid-19 pandemic-related limitations. Nevertheless, data concerning our primary endpoint of syncope can be easily obtained through remote appointments.

Despite the abovementioned limitations, the results of our study are another contribution to the feasibility and safety of CNA for the treatment of VVS. Randomized controlled trials are needed to confirm the effectiveness of this procedure.