Author + information
- Received July 12, 2017
- Revision received May 22, 2018
- Accepted May 29, 2018
- Published online October 15, 2018.
- Pablo Ávila, MDa,b,
- Francis Bessière, MD, MSa,
- Blandine Mondésert, MDa,
- Sylvia Abadir, MDa,
- Annie Dore, MDa,
- François-Pierre Mongeon, MD, MSa,
- Marc Dubuc, MDa and
- Paul Khairy, MD, PhDa,∗ ()
- aAdult Congenital Heart Center, Montreal Heart Institute, Université de Montréal, Montreal, Quebec, Canada
- bDepartment of Cardiology and Centro de Investigación Biomédica en Red Enfermedades Cardiovaculares (CIBERCV), Hospital Gregorio Marañón, Instituto de Investigación Sanitaria Gregorio Marañón, Universidad Complutense, Madrid, Spain
- ↵∗Address for correspondence:
Dr. Paul Khairy, Montreal Heart Institute Adult Congenital Center, Montreal Heart Institute, 5000 Belanger Street East, Montreal, Quebec H1T 1C8, Canada.
Objectives The purpose of this study was to assess the safety and efficacy of cryoablation for perinodal substrates in patients with congenital heart disease (CHD) and a displaced atrioventricular (AV) conduction system or an AV conduction system location that was difficult to predict.
Background Catheter ablation for perinodal arrhythmias in patients with CHD may incur higher risks due to unconventional or difficult to predict locations of the AV conduction system. Cryoablation carries theoretical advantages for such patients but has not been studied in this setting.
Methods A total of 35 patients with CHD underwent cryoablation for perinodal substrates at the Montreal Heart Institute between 2006 and 2016. Ten of these patients, age 33 ± 13 years, 60% male, had AV conduction systems that were displaced or of uncertain location and underwent cryoablation (6-mm electrode-tip catheter) for 12 perinodal arrhythmias: AV nodal re-entrant tachycardia (n = 4), non-automatic focal atrial tachycardia (n = 4), septal intra-atrial re-entrant tachycardia (n = 3), and para-Hisian automatic focal atrial tachycardia (n = 1). Four patients had single-ventricle physiology and had undergone Fontan palliation (3 atriopulmonary and 1 intracardiac total cavopulmonary connection), 4 underwent repair of AV septal defects, 1 had congenitally corrected transposition of the great arteries (TGA), and 1 had TGA with a Mustard baffle.
Results Cryoablation was acutely successful in 9 of 12 targeted arrhythmias (75%) with no procedural complication. Crossover to radiofrequency ablation successfully eliminated the remaining 3 arrhythmias at sites deemed safe by cryoablation, with no complication. Over a follow-up period of 26 (interquartile range: 15 to 64) months, 1 of 9 successfully cryoablated arrhythmias recurred.
Conclusions Cryoablation is feasible, safe, and moderately effective for perinodal arrhythmia substrates in patients with various forms of CHD associated with AV conduction systems that are displaced or in locations that cannot be reliably predicted.
Supraventricular arrhythmias are the most common long-term complications in the aging population with congenital heart diseases (CHD), the leading cause of morbidity and impaired quality of life and an important cause of mortality (1,2). While intra-atrial re-entrant tachycardia (IART) is the most prevalent arrhythmia in patients with CHD (3–6), other substrates include focal atrial tachycardia, atrioventricular (AV) nodal re-entrant tachycardia (AVNRT), and accessory pathway-mediated tachycardia (7,8).
Although radiofrequency (RF) energy is the most common modality used for catheter ablation in CHD, cryoablation offers theoretical advantages for perinodal substrates, including the ability to titrate temperatures to produce reversible effects prior to cellular destruction and enhanced catheter stability due to adhesion of the catheter tip to the endocardial surface (9–11). These features may be particularly advantageous in the setting of perinodal arrhythmia substrates when the AV conduction system is displaced or of ambiguous location, as in congenitally corrected transposition of the great arteries (ccTGA), atrioventricular septal defects (AVSD), and single ventricles (7,12–14). Although averting damage to the AV node is a preoccupation in treatment for all patients, AV block can be particularly poorly tolerated in those with complex CHD. Moreover, lack of venous access may prohibit ventricular endocardial lead implantation should the need arise. In light of these unique challenges, we assessed safety and efficacy of cryoablation for perinodal substrates in patients with CHD and an AV conduction system location that was displaced or difficult to predict.
A total of 35 patients with CHD underwent cryoablation for perinodal substrates at the Montreal Heart Institute between 2006 and 2016. Among these patients, 25 had an AV node located within the usual confines of Koch’s triangle and underwent cryoablation for AVNRT (n = 17), septal accessory pathways (n = 6), and perinodal atrial tachycardias (n = 3); 1 subject had both AVNRT and a septal accessory pathway. The study focused on the remaining 10 patients with 12 perinodal arrhythmias. All patients were identified through our registry of catheter ablation in CHD. Hospital records were thoroughly reviewed, and data for demographics, type of CHD, previous surgeries, symptoms, procedural characteristics, complications, and follow-up were extracted. Written informed consent was obtained for all procedures. The study was approved by our local institutional review board.
Procedures were performed with subjects under conscious sedation. All antiarrhythmic drugs were discontinued for ≥5 half-lives. None of the patients received amiodarone prior to the intervention. Vascular access was obtained through femoral and/or jugular veins according to patient anatomy. Prior to electrophysiological testing, conventional or rotational angiography was performed at the discretion of the operator. All procedures were performed with at least 2 catheters: a reference catheter and a roving ablation catheter. Additional quadripolar and decapolar diagnostic catheters were placed in accessible cardiac chambers depending on patient anatomy. A 3-dimensional (3D) electroanatomical mapping system was used in all patients (Ensite NavX or Velocity, St. Jude Medical, Minneapolis, Minnesota; or CARTO, Biosense-Webster, Diamond Bar, California). If required, access to the pulmonary venous atrium in patients with a Mustard baffle or total cavopulmonary Fontan connection (TCPC) was obtained through a baffle leak or fenestration if present or by a transbaffle or transconduit puncture, using a conventional transseptal needle (BRK-1; St. Jude Medical) or an RF perforation needle (Baylis, Montreal, Québec, Canada). For procedures in the pulmonary venous atrium, heparin was administered and adjusted to maintain a targeted activated clotting time of 300 to 350 s.
Whenever the clinical arrhythmia was not present at the beginning of the procedure, it was induced by programmed electrical stimulation. Diagnostic testing was performed using standard criteria (15).
Catheter ablation procedure
In all cases, cryoablation was performed using a 6-mm electrode-tipped catheter (Freezor Xtra, Medtronic CryoCath LP, Montreal, Québec, Canada). Cryomapping was achieved at −30°C for 30 to 60 s and, if efficacy and safety were confirmed, the temperature was lowered to −80°C for an additional 4 min. Vigilant monitoring of AV conduction was performed throughout the application. A double freeze-thaw cycle was applied at the site of success. Procedural endpoints included noninducibility by programmed electrical stimulation with up to 2 atrial extrastimuli and decremental atrial burst pacing from 400 to 200 ms with and without an isoproterenol infusion. In addition, complete elimination of slow-pathway conduction was targeted in patients with AVNRT (16). All patients were observed for a period of 30-min to 60-min following successful ablation. Crossover to RF energy was performed at the discretion of the operator after exhaustive attempts at cryoablation had failed despite the absence of AV prolongation during cryothermal applications in the area of interest.
Endpoints and follow-up
The primary efficacy endpoint was acute procedural success. The secondary efficacy endpoint was absence of recurrent, sustained (>30 s) tachycardia during long-term follow-up, whether symptomatic or asymptomatic. All patients were followed for 3 months after catheter ablation with 12-lead electrocardiography and 24-h Holter monitoring, with biannual or yearly follow-up thereafter, depending on clinical circumstances. Acute major complications and all other adverse events were noted.
Continuous variables are summarized as mean ± SD or median and interquartile range (25th and 75th percentiles), depending on normality of distribution, and were compared using the Mann-Whitney rank-sum U test. Categorical variables are represented as frequencies and percentages. Comparisons were performed using Fisher exact tests. Two-tailed p values <0.05 were considered statistically significant. Statistical analyses were performed using SPSS Statistics version 20.0 software (SPSS, Chicago, Illinois).
Table 1 summarizes the characteristics of patients with cryoablation of perinodal substrates with (n = 25) and without (n = 10) an AV node in the standard location within Koch’s triangle. Twelve procedures were performed in 10 patients, 33 ± 13 years of age, 6 males (60%) whose AV conduction system was displaced or difficult to predict (Table 2). In short, 4 patients (Table 2, Patients 1, #6, #7, and #9) had single-ventricle physiology with Fontan surgery (3 atriopulmonary and 1 intracardiac TCPC), 4 had repaired AVSDs (Table 2, Patients #3, #4, #5, and #10), 1 had TGA with a Mustard baffle (Table 2, Patient #2) and 1 had ccTGA (Table 2, Patient #8). All patients had recurrent symptomatic, sustained, documented supraventricular tachycardias. Median time from symptom onset to catheter ablation was 13 months (interquartile range [IQR]: 7 to 72 months). All but 1 patient had a previous hospital admission requiring pharmacological or electrical cardioversion. Eight patients (80%) received at least 1 antiarrhythmic drug prior to ablation, and all patients with Fontan surgery were taking oral anticoagulants.
A total of 12 perinodal arrhythmias, median cycle length 360 ms (IQR: 310 to 390 ms), were targeted during 12 procedures: AVNRT (n = 4), non-automatic focal atrial tachycardia (NAFAT) (n = 4), septal IART (n = 3), and para-Hisian automatic focal atrial tachycardia (n = 1). Anatomic landmarks were assessed by using fluoroscopy and conventional (n = 4) or rotational angiography (n = 4), with 3D electroanatomic mapping performed in all (EnSite [St. Jude Medical] in 11 patients and CARTO [Biosense-Webster] in 1 patient). In all cases, the perinodal arrhythmia was determined to be the clinically relevant tachycardia. In 4 cases (25%), additional arrhythmias were ablated during the same procedure. As noted in Table 1, the median procedural duration was 5.5 h (IQR: 4.5 to 8.5 h), which was significantly longer than for patients with standard AV node locations (median: 3.6 h [IQR: 2.9 to 5.4 h]; p = 0.028). However, fluoroscopy and ablation times were not significantly different.
Procedural outcome and long-term follow-up
Cryoablation was acutely successful for 9 of 12 targeted arrhythmias (75%), which was not significantly different from 24 of 26 successfully ablated perinodal substrates (92%) in patients with standard AV node locations (p = 0.301). No procedural complication occurred, with AV conduction remaining unaltered in all. Crossover to RF ablation successfully eliminated the remaining 3 arrhythmias at sites tested and deemed to be safe by cryoablation. Similarly, success was obtained upon crossover to RF energy in 2 controls with normally positioned AV conduction systems, both of whom had focal septal atrial tachycardias in the setting of atrial septal defects (ASDs). No complication occurred with RF ablation.
Over a median follow-up of 26 months (IQR: 15 to 64 months), 1 of 9 successfully ablated arrhythmias recurred 1 month after the procedure, which was not significantly different from 1 of 24 arrhythmias in patients with normally positioned AV conduction systems followed for 33 months (IQR: 7 to 67 months; p = 0.477). The 1 patient with a recurrence was managed pharmacologically with a good response. None of the arrhythmias requiring RF ablation recurred. All patients were alive and in sinus rhythm at the end of follow-up.
Procedural details according to arrhythmia substrate
AVNRT in the setting of a displaced AV conduction system
Cryoablation for AVNRT was performed in 2 patients with surgically repaired AVSDs and 1 patient with nonoperated ccTGA. For the first patient with an AVSD (Table 2, Patient #3), the slow pathway was ablated superior to the inferiorly displaced compact AV node (Figure 1). For the second and third patients (Table 2, Patients #4 and #5), the slow pathway was eliminated below the displaced His bundle, inferior to the ostium of the coronary sinus. The fourth patient with ccTGA and S,L,L (situs solitus, L-looped, ventricles, and L-transposed great arteries) segmental anatomy (Table 2, Patient #8) had atypical (slow-slow) AVNRT. Although the AV node was displaced superiorly, as expected, the slow pathway was nevertheless found in a conventional location within the middle third of the L-looped variant of Koch’s triangle. Initial success was obtained with cryoablation but with subsequent recurrence. Crossover to RF ablation at the same location eliminated slow-pathway conduction.
Focal atrial tachycardia (n = 5)
Cryoablation was used to treat 5 septal focal atrial tachycardias, 4 of which were para-Hisian in 4 patients with AVSD repair (n = 1), TGA and a Mustard baffle (n = 1), and single-ventricle physiology with Fontan surgery (n = 2). The patient with AVSD repair (Table 2, Patient #4) had previously undergone successful cryoablation of AVNRT. A second electrophysiological study performed a few months later confirmed the absence of recurrent slow-pathway conduction. However, a focal automatic atrial tachycardia was mapped through a transbaffle approach to the septal portion of the pulmonary venous atrium in proximity to the His recording. Cryoablation with a 6-mm electrode-tipped catheter terminated tachycardia and rendered it noninducible. The patient with TGA and a Mustard baffle (Table 2, Patient #2) had an obstructed inferior vena cava. A septal NAFAT was successfully cryoablated by means of a right jugular venous approach adjacent to a His signal recorded in the systemic venous atrium (17) (Figure 2). A patient with tricuspid atresia and classic Fontan surgery (Table 2, Patient #1) successfully underwent cryoablation of a para-Hisian NAFAT near the ostium of the coronary sinus (Figure 3). Finally, a patient with a double-inlet left ventricle and an intracardiac TCPC Fontan (Table 2, Patient #6) successfully underwent cryoablation of a para-Hisian NAFAT through a transbaffle puncture (Figure 4). Two years later, the patient presented with a new and much slower NAFAT higher along the septum in the left atrium. Cryoablation was unsuccessful, but crossover to RF ablation eliminated the tachycardia with no subsequent recurrence.
Intra-atrial re-entrant tachycardia (n = 3)
Two patients with atriopulmonary Fontan surgery (Table 2, patients 7 and 9) had IART circuits between septal scar and an atretic tricuspid valve, with critical isthmuses located in anterior or mid-septal regions. Cryoablation was used as first-line therapy given the uncertainty in localizing the AV conduction system. IART was successfully terminated, with bidirectional conduction block obtained in 1 patient (Table 2, Patient #7). In the second patient (Table 2, Patient #9), IART was terminated upon crossover to irrigated RF ablation at sites tested by cryoenergy, with no prolongation of AV conduction. Finally, a patient with a repaired AVSD and tetralogy of Fallot (Table 2, Patient #10) had an IART circuit around the primum ASD patch (Figure 5). Two potential critical isthmuses were identified between the ASD patch and either the coronary sinus or the tricuspid annulus. The latter was targeted with cryothermal energy considering the proximity of the inferiorly displaced AV node to the coronary sinus. A linear lesion between the patch and tricuspid annulus was applied where low-voltage–fractionated signals were observed. There, cryomapping and cryoablation successfully interrupted the IART and rendered the patient noninducible with no subsequent recurrence.
Several forms of CHD are associated with a displaced AV conduction system, the location of which is determined in part by concordance and morphology of the AV connection, alignment and position of the ventricular septum, and looping pattern of ventricular architecture. Herein, we describe a series of patients with perinodal arrhythmia substrates and CHD associated with an AV node location that was displaced or was difficult to predict such that catheter ablation carried an increased risk of AV block. In our experience, cryoablation proved to be safe in all cases and was moderately effective in treating complex perinodal substrates including AVNRT, NAFAT, IART, and automatic focal atrial tachycardia in patients with ccTGA, AVSD, univentricular hearts, and Mustard and Fontan surgery.
In ccTGA, the ventricular septum is typically not aligned with the atrial septum as a result of the wedged position of the pulmonary outflow tract (12). As such, a standard AV node cannot come into contact with a ventricular conduction system. Instead, an anterolaterally displaced AV node connects to the His–Purkinje conduction system below the atrial appendage. In an AVSD, which primarily affects the AV junction, lack of atrial and ventricular septal continuity determines the course of the conduction system. Koch’s triangle, although present, does not contain the AV node. Rather, it is displaced posteriorly along the atrial septum, just above the AV junction and anterior to the ostium of the coronary sinus (13,18) (Figures 1 and 5).
In hearts with univentricular connections, prediction rules regarding location of the AV node are helpful but may be misleading. In a double-inlet left ventricle, the AV node is typically displaced anterolaterally (19). However, in our patient with a para-Hisian NAFAT and double-inlet left ventricle, the AV node was displaced inferiorly (Figure 4), underscoring marked interindividual variations despite similarly labeled defects. In patients with tricuspid atresia, the AV conduction system cannot be normally located owing to the absence of a right AV connection. The AV node is typically found on the floor of the right atrium adjacent to the mouth of coronary sinus (Figure 3) and central fibrous body, which is sometimes recognizable by a dimple (i.e., blind-ending right atrium) (20). Location of the AV node can be variable in patients with double-inlet right ventricles. When an overriding tricuspid valve is predominantly connected to a morphological right ventricle, the AV node can be normally positioned. However, depending on the orientation of the septum, the node may be more posterior than usual.
Current expert consensus statements in pediatric and congenital electrophysiology support the use of cryoablation as a Class IIA indication for septal substrates (21,22), considering the established safety profile (23–27). Several factors including operator experience, optimization of the ablation site, rapid time to desirable effect (e.g., <10 s), larger catheter tips (i.e., 6 or 8 mm as opposed to 4 mm), rapid cooling rates, lower target temperatures, and repeated freeze-thaw cycles have been associated with a lower risk of recurrence (25,28). Our results further extend the utility of cryoablation for septal substrates to patients with a displaced or uncertain location of the AV node. Nevertheless, in 3 cases, cryoablation was unsuccessful and crossover to RF was required to eliminate the targeted arrhythmia. In all such cases, cryoablation was believed to be helpful in better delineating the AV conduction system and/or providing additional safety data for subsequent RF ablation. A prototype of a hybrid cryothermal-RF ablation catheter, which may be of interest for such perinodal substrates, has previously been described (29) but is not yet commercially available.
A clear understanding of 3D anatomy is essential for optimizing catheter ablation outcomes in patients with complex CHD. Integration of 3D images acquired by computed tomography, cardiac magnetic resonance imaging, or rotational angiography provides valuable road maps to guide electroanatomic mapping. Mapping resolution is influenced by electrode size and interelectrode spacing. Although differences between 6-mm and 4-mm electrode-tipped catheters may have a substantial impact on mapping resolution within low-voltage areas, the impact on normal voltage amplitudes, such as His signals, is less marked (30). Moreover, 4-mm cryocatheters produce substantially smaller lesions (31), such that the trade off between mapping resolution and lesion effectiveness generally favors 6-mm cryocatheters.
Identifying the compact AV node and slow pathway in patients with AVNRT and a displaced AV conduction system can be challenging. Our general approach is to mark the location of the His signal, induce AVNRT, and electroanatomically map retrograde atrial activation (i.e., fast pathway for typical AVNRT or slow pathway for atypical AVNRT). Cryomapping at the suspected slow pathway site is then performed in tachycardia, with subsequent cryoablation if tachycardia is terminated by block in the slow pathway and the site is deemed to be safe (Figure 1B). The 1 case of AVNRT that was not successfully cryoablated was in a patient with ccTGA. Interestingly, the slow pathway was effectively eliminated by RF ablation in the L-looped version of Koch’s triangle despite the anterolaterally displaced AV node. A series by Liao et al. (32) described slow-pathway modification with RF ablation in the mid-septum in 5 of 8 cases with ccTGA and in the posterior septum in the remaining 3 cases. In 4 cases of attempted ablation of AVNRT in patients with ccTGA, as reported by Upadhyay et al. (14), 2 were successful from a conventional, albeit L-looped, slow pathway location. Treatment failed in 1 case despite multiple attempts from both sides of the atrial septum, with AV block that transiently occurred during cryoablation and fully resolved upon prompt interruption of therapy. These findings further highlight the potential safety benefits that cryothermal energy can provide for arrhythmia substrates in risky locations within complex anatomies.
An alternative approach that has been proposed in patients with CHD and arrhythmia substrates in high-risk locations is the use of remote magnetically guided catheter ablation (33). Advantages include greater catheter stability, enhanced catheter maneuverability, excellent 3D image integration, and a reduction in radiation exposure. The system is not yet compatible with cryocatheters such that reliable reversible lesions cannot be applied.
This study is observational and involves a limited number of patients with varied perinodal substrates and heterogeneous forms of CHD. The lack of comparisons to RF ablation limits firm conclusions regarding the relative merits of the 2 energy sources. Cases were performed in a single referral center for complex ablations in CHD with extensive experience in cryoablation, such that results may not be extrapolated to sites with limited proficiency in CHD or cryoablation. As highlighted by consensus statements, electrophysiologic procedures in patients with moderate or complex CHD or complex arrhythmias should be performed in regional referral centers by cardiac electrophysiologists with expertise in CHD and in laboratories with appropriate personnel and equipment (22).
In our experience, cryoablation was a feasible and safe treatment for perinodal arrhythmia substrates in patients with varied forms of CHD associated with a displaced AV conduction system, including AVSDs, ccTGA, and univentricular hearts. It could be helpful in localizing and eliminating slow-pathway inputs to compact AV nodes situated outside the usual confines of Koch’s triangle and in targeting septal re-entrants circuits and perinodal or para-Hisian focal tachycardia sources. Moderate effectiveness was observed in this case series, with crossover to RF energy required in a minority of cases.
COMPETENCY IN MEDICAL KNOWLEDGE: Some forms of congenital heart disease, such as AVSD, ccTGA, and certain types of univentricular hearts, are associated with displaced AV conduction systems. Catheter ablation of perinodal substrates should be undertaken with care in patients with AV nodes that are displaced or are of uncertain location. Cryoablation is feasible, safe, and moderately effective as a first-line treatment modality in this context.
TRANSLATIONAL OUTLOOK: Multicenter studies are required to provide more accurate estimates regarding the safety and efficacy of cryoablation for perinodal substrates in patients with displaced or uncertain AV node locations, and to compare this strategy with other technologies such as remote magnetically guided RF ablation.
Supported by the Montreal Heart Institute Foundation. Dr. Ávila has received research support from the Alfonso Martín Escudero Foundation, Spain. Dr. Bessière is supported by the French Federation of Cardiology. Dr. Khairy holds the Research Chair in Electrophysiology and Congenital Heart Disease. Dr. Dubuc has consulted for Medtronic CryoCath LP. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrioventricular nodal re-entrant tachycardia
- atrioventricular septal defect
- congenitally corrected transposition of the great arteries
- congenital heart disease
- transposition of the great arteries
- Received July 12, 2017.
- Revision received May 22, 2018.
- Accepted May 29, 2018.
- 2018 American College of Cardiology Foundation
- Verheugt C.L.,
- Uiterwaal C.S.,
- van der Velde E.T.,
- et al.
- Walsh E.P.,
- Cecchin F.
- Balaji S.,
- Johnson T.B.,
- Sade R.M.,
- Case C.L.,
- Gillette P.C.
- Driscoll D.J.,
- Offord K.P.,
- Feldt R.H.,
- Schaff H.V.,
- Puga F.J.,
- Danielson G.K.
- Khairy P.,
- Aboulhosn J.,
- Gurvitz M.Z.,
- et al.
- Dick M. 2nd.,
- Norwood W.I.,
- Chipman C.,
- Castaneda A.R.
- Upadhyay S.,
- Valente A.M.,
- Triedman J.K.,
- Walsh E.P.
- Anderson R.H.,
- Ho S.Y.,
- Becker A.E.
- Guller B.,
- DuShane J.W.,
- Titus J.L.
- Saul J.P.,
- Kanter R.J.,
- Writing C.,
- et al.
- Chan N.Y.
- Andrade J.G.,
- Khairy P.,
- Dubuc M.
- Anter E.,
- Tschabrunn C.M.,
- Josephson M.E.
- Liao Z.,
- Chang Y.,
- Ma J.,
- et al.