Catheter ablation of atrial tachyarrhythmias in patients with atrioventricular septal defect

Abstract

Aims

The incidence of atrial tachyarrhythmias is high in patients with atrioventricular septal defect (AVSD). No specific data on catheter ablation have been reported so far in this population. We aimed to describe the main mechanisms of atrial tachyarrhythmias in patients with AVSD and to analyse outcomes after catheter ablation.

Methods and results

This observational multi-centric cohort study enrolled all patients with AVSD referred for catheter ablation of an atrial tachyarrhythmia at six tertiary centres from 2004 to 2022. The mechanisms of the different tachyarrhythmias targeted were described and outcomes were analysed. Overall, 56 patients (38.1 ± 17.4 years, 55.4% females) were included. A total of 87 atrial tachyarrhythmias were targeted (mean number of 1.6 per patient). Regarding main circuits involved, a cavo-annular isthmus-dependent intra-atrial re-entrant tachycardia (IART) was observed in 41 (73.2%) patients and an IART involving the right lateral atriotomy in 10 (17.9%) patients. Other tachyarrhythmias with heterogeneous circuits were observed in 13 (23.2%) patients including 11 left-sided and 4 right-sided tachyarrhythmias. Overall, an acute success was achieved in 54 (96.4%) patients, and no complication was reported. During a mean follow-up of 2.8 ± 3.8 years, 22 (39.3%) patients had at least one recurrence. Freedom from atrial tachyarrhythmia recurrences was 77.5% at 1 year. Among 15 (26.8%) patients who underwent repeated ablation procedures, heterogeneous circuits including bi-atrial and left-sided tachyarrhythmias were more frequent.

Conclusion

In patients with AVSD, most circuits involve the cavo-annular isthmus, but complex mechanisms are frequently encountered in patients with repeated procedures. The acute success rate is excellent, although recurrences remain common during follow-up.

Graphical abstract

Open in new tabDownload slide

Introduction

Atrial tachyarrhythmias are a major cause of morbidity in the growing population of patients with congenital heart disease (CHD).^1–3 The prevalence is estimated around 15% and increases with age, CHD severity, and follow-up duration. Projections indicate that half of 20-year-old patients with complex CHD will have an atrial tachyarrhythmia during their lifespan.¹

Invasive approaches such as ablation therapy have demonstrated their superiority over pharmacological antiarrhythmic treatment and are now recommend as first-line therapy.^4,5 Growing experience in invasive therapy associated with major advances in mapping and ablative technologies have also led to a better understanding of arrhythmias mechanisms in various CHD and have resulted in an important improvement of ablation outcomes.^6–9 In patients with atrioventricular septal defect (AVSD), while epidemiological data have demonstrated a high incidence of atrial tachyarrhythmias (55% at 60 years of age),¹⁰ there are so far no specific reports on their mechanisms and relation to catheter ablation outcome. The unique anatomic features of AVSD patients may be associated with specific mechanisms of tachyarrhythmias in this population.¹¹

In this study, we aimed to describe the main mechanisms and locations of circuits underlying atrial tachyarrhythmias in patients with AVSD and to analyse outcomes after catheter ablation.

Methods

Study population and settings

The data, analytic methods, and study materials will be made available to other researchers upon reasonable request. This observational multi-centric cohort study enrolled all patients with AVSD referred for catheter ablation of an atrial tachyarrhythmia at six tertiary centres from June 2004 to December 2022: the Erasmus Medical Center (Rotterdam, The Netherlands), the European Georges Pompidou Hospital (Paris, France), Louis Pradel Hospital (Lyon, France), Marie Lannelongue Hospital (Le Plessis-Robinson, France), Necker Hospital (Paris, France), and Pasteur Clinic (Toulouse, France). Patients with univentricular physiology and Fontan circulation were excluded. These hospitals are national referral centres for CHD patients and include multi-disciplinary teams with cardiologists, interventional cardiologists, surgeons, imaging specialists, anaesthesiologists, and electrophysiologists specifically trained in CHD. All patients with CHD in those centres are systematically gathered in phenotype-based databases since first evaluation to ensure a comprehensive collection of cases over time.

Study data were centralized, collected, and managed using REDCap electronic data capture tools hosted at the Cardiovascular Epidemiology Unit of the Paris Cardiovascular Research Center (Inserm 970, European Georges Pompidou Hospital, Paris, France).¹² This study complies to the French data protection committee (Commission Nationale Informatique et Liberté, CNIL, MR-004 n°2219046) and was approved by the institutional review board (CERAPHP.5, IRB n°00011928).

Collected data

For each patient, data collected included demographic characteristics, type of AVSD, associated cardiovascular abnormalities, types of cardiac surgeries, comorbidities (including cardiovascular risk factors), association with a Down syndrome, comprehensive history of cardiac tachyarrhythmias, and cardiac imaging. Images from the echocardiographic examination closest to catheter ablation procedure were analysed. Left atrial dilatation was defined as an indexed volume >34 mL/m²,^4,13 and right atrial dilatation was defined as an end-systolic area >18 cm².^4,14 The regurgitation of the left atrioventricular (AV) valve was graded as 0 (absent), 1 (mild), 2 (moderate), or 3 (severe) according to last guidelines.^15,16

Procedural outcomes

Comprehensive data of all catheter ablation procedures were analysed. The mechanisms of the different tachyarrhythmias targeted were described, and the acute success rate was reported. In patients with a coronary sinus draining to the left atrium due to prior surgical repair, the reference catheter was placed in the right appendage or on the lateral wall of the right atrium. In patients in sinus rhythm at the beginning of the procedure, the clinical arrhythmia was induced by programmed atrial stimulation, with isoproterenol infusion if needed. In these patients, clinical arrhythmia was confirmed by comparison with electrocardiogram (ECG) morphology and/or tachycardia cycle length. A non-clinical arrhythmia was defined as an arrhythmia induced during the procedure, which was not documented before the procedure. The arrhythmia was classified as undetermined when it was not certain whether the induced arrhythmia was comparable with the clinical arrhythmia due to moderate differences in either morphology or cycle length. The ablation was considered as prophylactic if an ablation line transected a potential anatomical isthmus (mainly the cavo-annular isthmus or between the atriotomy incision and the inferior vena cava) of which it could not be demonstrated that it was part of a re-entry circuit. In patients with AVSD, the terms right and left AV valves are preferred since the valves are derived from five leaflets and are not true mitral or tricuspid valves. Hence, the term cavo-annular isthmus was used to describe the isthmus between the right AV valve and the inferior vena cava.

Complete acute procedural success was defined as termination of clinical arrhythmia(s) with subsequent verification of bi-directional conduction block and successful ablation of all secondary tachyarrhythmias induced [except inducible atrial fibrillation (AF)]. Partial acute procedural success was defined as termination of clinical arrhythmia with subsequent verification of bi-directional conduction block across the critical isthmus/radiofrequency lesion line with additional tachyarrhythmias induced not targeted or not successfully targeted or in the absence of final programmed stimulation performance.

Patients were routinely followed at outpatient visits with ECG and 24-hour Holter recordings and/or pacemaker/defibrillator check at 3, 6, and 12 months and then on an annual basis. Medical appointments were promptly scheduled in case of any symptoms suggestive of arrhythmia. Freedom from recurrences (≥30 s) at 1 and 2 years follow-up was also analysed, including both targeted arrhythmia and any other atrial tachyarrhythmias. In case of redo-procedures, mechanisms and circuits of tachyarrhythmias targeted were also reported.

Statistical analysis

Categorical data were reported as numbers and percentages. Continuous data were presented as mean and standard deviation or median and inter-quartile range (IQR). Comparisons used the χ² or Fisher’s exact test for categorical variables and Student’s t-test or Mann–Whitney–Wilcoxon test, when appropriate, for continuous variables. For freedom from recurrence analysis, the primary time to event endpoint was the time from catheter ablation to first documented episode (≥30 s) of the targeted arrhythmia or of any other atrial tachyarrhythmia. Censoring occurred in the event of loss to follow-up or death. Survival curves were plotted by the Kaplan–Meier method. A two-tailed P-value <0.05 was considered statistically significant. Statistical analysis was performed using R software version 3.6.3 (R Project for Statistical Computing, Vienna, Austria).

Results

Patients’ characteristics

A total of 62 patients with AVSD were referred for catheter ablation. After exclusion of six patients who underwent ablation for AV re-entrant tachycardia (n = 2), premature ventricular contraction (n = 2), or had prior catheter ablation procedures in another centre (n = 2), 56 patients (38 ± 17 years, 55.4% females) with AVSD and a first ablation procedure of an atrial arrhythmia were included.

Main characteristics of patients are presented in Table 1. The primary arrhythmia targeted was an intra-atrial re-entrant tachycardia (IART) or a focal atrial tachycardia (FAT) in 49 (87.5%) patients and AF in 7 (12.5%). Two patients also had AF targeted as a second arrhythmia as AF episodes had been documented after IART/FAT ablation. Patients with IART/FAT were younger than patients with AF (36.4 ± 17.3 vs. 51.5 ± 12.2, P = 0.017), had a higher number of prior cardiac surgeries [median (IQR) 2 (1–3) vs. 1 (1–1), P = 0.001], and had a shorter delay between surgical repair and catheter ablation (22.2 ± 13.4 vs. 39.9 ± 14.1 years, P = 0.026).

Regarding pharmacological antiarrhythmic therapy before catheter ablation, 28 (50.0%) patients received beta-blockers, 18 (32.1%) amiodarone, 8 (14.3%) sotalol, and 2 (3.6%) Class I agents.

Catheter ablation procedure

The procedure was performed under general anaesthesia in 31 (55.4%) patients, local anaesthesia in 23 (41.1%), and deep sedation in 2 (3.6%). A superior jugular venous was used in 6 (10.7%) patients (1 patient with inferior vena cava occlusion and 5 patients with azygos continuation), and a trans-septal access was required in 22 (39.3%). A three-dimensional system mapping-guided (CARTO®, Biosense Webster) ablation was performed in 54 (96.4%) patients, whereas 2 (3.6%) procedures were performed with conventional fluoroscopy guidance. A high-density mapping catheter was used in 50 (89.3%) patients. Radiofrequency was used in all procedures except two, including cryothermal energy for pulmonary veins isolation (PVI) in a patient with AF and for a FAT originating from the coronary sinus roof in another patient. Among patients in whom radiofrequency energy was applied, an irrigated tip catheter was used in 51 (92.6%) patients and a contact force catheter in 49 (90.7%). In patients with AF as primary tachyarrhythmia targeted (n = 7), PVI was performed in all of them. An additional cavo-annular isthmus lesion was created in two (28.6%) patients and left-sided lesions in two other patients (28.6%). Of note, the coronary sinus drained into the left atrium in 15 (26.8%) patients.

Tachyarrhythmias mechanisms and circuits

The mean number of atrial tachyarrhythmias targeted during the first catheter ablation procedure was 1.6; 1, 2, 3, and 4 tachyarrhythmias were targeted in respectively 34 (60.7%), 14 (25.0%), 7 (12.5%), and 1 (1.8%) patients. A total of 87 atrial tachyarrhythmias were targeted including 72 (82.8%) IART, 9 (10.3%) AF, and 6 (6.9%) FAT. The different mechanisms and circuits of these arrhythmias are depicted in Figure 1.

When all tachyarrhythmias were considered (excluding prophylactic ablations), a cavo-annular isthmus-dependent IART was observed in 41 (73.2%) patients and an IART involving the right lateral atriotomy in 10 (17.9%) patients. Other tachyarrhythmias with heterogeneous circuits were observed in 13 (23.2%) patients including 11 left-sided (in 9 patients) and 4 right-sided tachyarrhythmias. The proportion of left-sided tachyarrhythmias among patients with IART/FAT targeted was 21.2% vs. 4.3% in those with or without prior left-sided surgery (left AV valve repair or replacement), respectively (P = 0.123).

According to definitions used, of the 87 tachyarrhythmias targeted, 60 (69.0%) were clinical tachyarrhythmias, 10 (11.5%) non-clinical tachyarrhythmias, 5 (5.7%) undetermined tachyarrhythmias, and 12 (13.8%) prophylactic ablation lines. As demonstrated in Figure 1, when only clinical tachyarrhythmias were considered, a cavo-annular isthmus-dependent IART was observed in 38 (67.9%) patients, being the only arrhythmia targeted in 31 (55.4%) patients or in association with other clinical arrhythmia(s) in 7 (12.5%) patients. An IART involving the right lateral atriotomy was observed in seven (12.5%) patients, being the only arrhythmia targeted in thee (5.4%) patients or in association with other clinical arrhythmia(s) in four (7.1%) patients. A bi-atrial IART (excluding bi-atrial circuits involving the cavo-annular isthmus due to the unique common annulus in patients with AVSD) was observed in three (5.4%) patients. Other tachyarrhythmias with heterogeneous circuits were observed in seven (12.5%) patients including five left-sided tachyarrhythmias and two right-sided tachyarrhythmias.

The most common circuits are shown in Figure 2. A complex bi-atrial IART circuit is also provided in Supplementary material online, Video S1 and a bi-atrial IART circuit involving the cavo-annular-isthmus in Supplementary material online, Video S2.

Acute procedural outcomes

Overall, an acute procedural success was achieved in 54 (96.4%) patients. In the remaining two patients, ablation of a cavo-annular isthmus-dependent IART failed as a result of incessant AF, and in another patient with AF, a gap around the left pulmonary veins persisted. Among the 54 patients with acute procedural success, programmed atrial extra-stimulation protocols were performed after clinical arrhythmia ablation in 36 (66.7%) patients, and a non-clinical arrhythmia was inducible in 15 (41.7%). In four of them, the non-clinical tachyarrhythmia induced was not targeted. Hence, complete acute procedural success was reported in 32 (57.1%) patients and a partial procedural success in 22 (39.3%). Complications associated with the catheter ablation procedure were not observed in any of the patients.

Long-term outcomes

During a mean follow-up of 2.8 ± 3.8 years after the first catheter ablation procedure, 22 (39.3%) patients presented with at least one recurrence episode of a sustained atrial tachyarrhythmia. The recurrent arrhythmia was either comparable with the arrhythmia targeted in the first ablation procedure (n = 8) or a different arrhythmia (n = 12). In two patients, there was uncertainty whether the recurrent arrhythmia was comparable with the original one.

In the entire patient group, freedom from atrial arrhythmia recurrence was 77.5% at 1 year follow-up and 60.5% after 2 years (Figure 3); corresponding rates were, respectively, 79.6 and 64.7% in patients with IART vs. 60.0 and 30.0% in patients with AF (P = 0.3). Freedom from recurrence of the tachyarrhythmia initially targeted during the first procedure was 86.2% at 1 year.

Repeated ablation procedures

Overall, 15 (26.8%) patients had a second ablation procedure, including 6 (10.7%) with 3 ablation procedures and 1 (1.8%) with 4 ablation procedures.

The clinical arrhythmia targeted during the second procedure was an IART in 12 (80.0%) patients and AF in 3 (20.0%). Circuits included cavo-annular isthmus-dependent IARTs (n = 6, 40.0%), right-sided IARTs (n = 2, 13.3%, including one circuit around the atriotomy), left-sided IARTs (n = 2, 13.3%), and a bi-atrial IART (n = 1, 6.7%). The mechanism was undetermined in one patient. The acute success rate after the second ablation procedure was 86.7% (13/15).

The clinical tachyarrhythmia targeted during the third procedure was an IART in four (66.7%) patients, AF in one (16.7%) and FAT in one (16.7%). Circuits included bi-atrial IARTs (n = 3, 50.0%), a right-sided FAT (n = 1, 16.7%), and a left-sided IART (n = 1, 16.7%). The clinical tachyarrhythmia targeted during the fourth procedure was a left-sided IART.

At the end of the follow-up period, 46 (82.1%) patients were in sinus rhythm, whereas 10 (17.9%) patients had atrial tachyarrhythmia recurrences (including 5 with permanent atrial arrhythmia). Antiarrhythmic drugs were discontinued in 36 (78.3%) patients who had sinus rhythm.

Discussion

This study is the first to describe the main mechanisms of atrial tachyarrhythmias and both procedural and long-term outcomes after catheter ablation in patients with AVSD. Our findings demonstrate not only a predominance of cavo-annular IARTs but also a substantial proportion of more complex tachyarrhythmias including bi-atrial and left-sided re-entry circuits in patients with repeated procedures. With current ablative technologies, the acute success rate is >90% although recurrences remain significant. A stable sinus rhythm is achieved in most patients, sometimes after several ablation procedures.

The incidence of atrial tachyarrhythmias in patients with AVSD is considerable. Recent epidemiological data have demonstrated that the lifetime risks for developing atrial tachyarrhythmia to ages 20, 40, and 60 were 3.7, 17.8, and 55.3%, respectively.¹⁰ Intra-atrial re-entrant tachycardia /FAT was the leading arrhythmia until the age of 45 and then AF became predominant. Age, the number of cardiac surgeries, left and right atrial dilatation, and left AV valve regurgitation were associated with an increased risk for developing atrial tachyarrhythmias. While a few other studies have analysed the prevalence and the incidence of atrial tachyarrhythmias in patients with AVSD,^17,18 no specific data on tachyarrhythmias mechanisms and catheter ablation had been reported so far in this population.

In patients with CHD, tachyarrhythmias mechanisms are highly dependent of the underlying defect. While some patients with AVSD have been reported in different studies regrouping heterogenous forms of CHD, specific mechanisms and circuits found in these patients were not individually addressed and remained unknown. The consideration of the unique anatomic features of patients with AVSD is essential. Atrioventricular septal defect covers a spectrum of congenital heart malformations characterized by a common AV junction co-existing with deficient AV septation. In partial AVSD (ostium primum ASD and ‘cleft’ left AV valve), there are separate AV valve orifices despite a common junction, while in complete AVSD, the valve itself is also shared with a variable deficiency of the inlet ventricular septum. Moreover, because of the ostium primum defect, the AV node is postero-inferiorly displaced close to the coronary sinus ostium.¹⁹

Our data revealed that the isthmus between the right-sided annulus and the inferior vena cava (equivalent to cavo-tricuspid isthmus) was involved in more than half of patients. Because of the abnormal location of the AV node, in these patients, the cavo-annular ablation line should be performed more laterally in order to minimize the risks of AV block (Figure 2).²⁰ Since patients with AVSD have a single AV annulus, bi-atrial IART circuits around the common annulus are also sometimes observed (see Supplementary material online, Video S2).^21,22 In such cases, the arrhythmia can usually be terminated by a single ablation line between the inferior vena cava and the annulus. Compared with patients with ostium secundum ASD closure, in whom typical peri-tricuspid flutter and AF are largely predominant,²³ other mechanisms were relatively heterogeneous and included not only IART circuits around the atrial incision on the lateral wall of the right atrium but also left-sided tachyarrhythmias. While the initial surgery in these patients usually associates ostium primum ASD closure (+/− ventricular septal defect closure) and suture of the left AV valve cleft, the frequent chronic left AV valve regurgitation (more than one-third of our population had moderate or severe regurgitation) and associated surgical re-interventions are commonly associated with a left atrial dilatation (two-third of our cohort) compared with other forms of CHD. In addition to these frequent re-interventions, the location of the ASD closure patch (ostium primum), sometimes large and sometimes sutured on the left side to avoid AV block according to the surgical technique, is associated with complex septal substrate that may favour the development of these septal or bi-atrial circuits.

Despite an excellent acute success rate >90%, recurrences of atrial tachyarrhythmias remained relatively common during follow-up. One year after the first catheter ablation, 23% of patients experienced at least one recurrence (14% when only the arrhythmia targeted was considered). During repeated ablation procedures, cavo-annular isthmus-dependent IARTs were less common, with a higher proportion of complex bi-atrial or left-sided circuits. Because patients with AVSD frequently undergo one or more re-interventions on the left AV valve, a surgical trans-septal approach is commonly used and is associated with scarring that may create a septal substrate for re-entry. This, in association with the common AV annulus, may underly the relatively high proportion of bi-atrial IART circuits observed. To find the optimal target for ablation in these patients is complex. In patients post-mitral valve surgery using the trans-septal approach, different bi-atrial circuits have been described involving postero-inferior inter-atrial connections, Bachmann’s bundle, the fossa ovalis, and/or the coronary sinus ostium as inter-atrial bridges.²⁴ However, even if several ablation procedures were sometimes required, most patients (>80%) were in stable sinus rhythm at the end of the follow-up.

Our data also provide reassuring data on the safety of catheter ablation in this population with no procedure-related complications reported. This is important to inform the benefit/risk balance considering the complex anatomies requiring sometimes atypical accesses or complex trans-septal punctures because of significant septal scarring or large patches. To minimize the risk of complications, pre-procedural planning should include a comprehensive assessment of cardiac anatomy and a thorough review of prior surgical interventions. For example, knowledge of coronary sinus drainage into a left atrium could prevent futile attempts to cannulate a coronary sinus ostium that is not accessible from the right atrium, and knowledge of congenital or acquired anomalies of the systemic venous return is important to elaborate a careful procedural plan, including best and alternative catheter access options. The high proportion of superior accesses in our cohort (10%) was mainly due to azygos continuation, which is more prevalent in patients with AVSD (association with heterotaxy syndromes and left isomerism). Furthermore, we did not report any iatrogenic AV block as previously reported in older series of heterogeneous CHD patients including some AVSD (because of inadvertent catheter ablation of abnormally located conduction pathways).

Limitations

First, this study was observational and retrospective, with inherent biases. Secondly, the follow-up of patients was not standardized across the different centres. However, all institutions are tertiary centres with regular follow-up of patients after catheter ablation. Moreover, atrial tachyarrhythmias are very rarely asymptomatic in this population, which limits the likelihood of undocumented recurrences. Thirdly, we report here the outcomes of catheter ablation in patients with AVSD performed by highly experienced operators. Our experience could not be transposable to other centres or operators; however, the management of patients with moderate or complex CHD, as clearly stated in last international guidelines, should be limited to expert centres.^4,5 Lastly, we presented outcomes after catheter ablation of AF and IART separately (Figure 3) to be more informative, but the lack of statistically significant difference in recurrence rate should be considered with caution considering the small sample size of the AF group.

Conclusions

In patients with AVSD, most tachyarrhythmias circuits involve the cavo-annular isthmus, but heterogeneous and complex mechanisms including bi-atrial and left-sided tachyarrhythmias are more frequently encountered in patients with repeated procedures. With current ablative technologies and accumulated experience, the acute success rate is excellent, but recurrences remain common during follow-up. A stable sinus rhythm is however achieved in most patients, sometimes after several ablation procedures.

Supplementary material

Supplementary material is available at Europace online.

Funding

This work was supported by the French Institute of Health and Medical Research, Fédération Française de Cardiologie, and Société Française de Cardiologie.

Data availability

The data, analytic methods, and study materials will be made available to other researchers upon reasonable request.

References

Author notes