Since the first reports in the early nineties, catheter ablation has become the first therapeutic choice for re-entrant arrhythmias in children and adolescents [1‐3]. Paediatric literature demonstrates a long-term success rate of 95% [4], equivalent to that of the adult patients [5‐9]. Some of the factors leading to prolonged procedures, procedural failure, or recurrence of arrhythmias in the young age group have been briefly discussed [10, 11].

The aim of the present study was to investigate the incidence and underlying reasons for failure or recurrence of accessory pathway catheter ablations and determine which inherent patient’s or procedural characteristics are associated with high risk of recurrence. The initial and repeated procedures were compared in terms of techniques and strategies to help understand what may have determined failure or recurrence, and the contributions and strategies resulting in a long-term success.

The present study has been approved by the Institutional Ethics Committee and has been performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Our study corroborates that the long-term success of accessory pathway ablation in the paediatric population can be increased to > 98% after a repeated procedure. It also demonstrates that the main reasons for failure include inaccurate mapping and/or diagnosis, and inadequate long-term lesion formation (mainly related to the use of cryo-energy during the initial procedure).

Even if many expert consensus on management of accessory pathway in the paediatric age estimate the single procedural acute success at 94–95% [11‐13], this figure is not universally accepted. Some studies have shown higher recurrence rates when an extended follow-up period is evaluated [10, 15]. Thus, a significant number of patients require multiple procedures to achieve definitive success. A recent review of risk factors associated with recurrence in the paediatric population suggests that the position and number of accessory pathways are related to higher incidence of recurrence following ablation [15]. Our group has also identified specific challenges at each position and built a body of hypothesis that may be associated with recurrence. These risk factors are still pending validation with larger studies involving an appropriate methodology.

The main apparent reasons for the recurrence especially in the left and right lateral location were related to difficulties with catheter stability, which can be solved with the introduction of contact force, the use of 3D electro-anatomical mapping and long sheaths, and direct visualization of the interface catheter-tissue with intracardiac echography imaging.

It has been long apparent that left lateral pathways are the most frequent [14]. Regardless of the location, a deep understanding of the anatomic landmarks and a careful mapping seem to be the keys for success. A right lateral position correlates with more difficult procedures with higher number of lesions to achieve success, leading to incorporate long (preferably steerable) sheaths, contact force sensing catheters, and in rare cases, jugular access to assure better catheter stability. Moreover, a common condition that eventually makes ablation difficult in Mahaim fibres is the temporary disappearance of the pathway conduction and extreme susceptibility to mechanical block.

Likewise, Ebstein’s anomaly is a risk for failed ablation and recurrence, the difficulties being related to the high prevalence of multiple pathways and to the lack of catheter stability secondary to the displacement of the tricuspid valve [16, 17].

Para-Hisian and antero-septal pathways’ challenge relates to the proximity of the AV node. The wisdom of an accurate mapping and catheter stability cannot be overemphasized. Some centres advocate for the use of cryoablation to avoid injury to the nodal structures, advocating that stability is achieved once the ice ball is formed facilitating contact. However, an important drawback is that the catheter is stiff which interferes with manipulation in small hearts. Given the high prevalence of recurrence with cryo-energy (15–20%) [18, 19] it has been our choice to attempt a cautious radiofrequency ablation protocol including a conscious delineation of the right atrium, tricuspid valve, and right ventricle anatomy with 3D activation mapping. Rotational angiography of the right atrium with 3D overlay on live fluoroscopy has been recently incorporated in many laboratories to help a precise location of the His position. Because the His bundle is located deeper and protected with a fibrous tissue envelope in the ventricular septum as compared to the atrial and superficially located compact AV node, a key point with this approach is to target the ventricular insertion by searching for predominant ventricular signals with a small atrial. In the challenging cases of high-risk anterograde only accessory pathways with no inducible supraventricular tachycardia, mapping is usually performed during fast atrial pacing in an attempt to maximize ventricular pre-excitation. In the case of bidirectional pathways in this location, both atrial (in case of inducible orthodromic reciprocating tachycardia) and/or ventricular mapping during fast atrial pacing or sinus rhythm can be used. In all cases, we emphasize on confirmation of underlying intact AV node conduction with differential atrial pacing maneuvres before and during the ablation. In some cases, the use of contact force technology, carefully increasing the contact from 5 to above 30 g allows precise identification of the sites of both mechanical AV block and mechanical bumping of accessory pathway conduction. The 3D contact force catheters currently available in the market are irrigated tip catheters (ThermoCool SmartTouch catheter, Biosense Webster, CA, USA, and TactiCath™ Quartz Contact force ablation catheter, Abbot, Chicago, IL, USA). Our strategy has been to set the irrigation mode low (usually 2 to 7 ml/min) during ablation, to use them as non- or minimally irrigated catheters and thus limit the extension of the lesion. In selected cases, applications with limited energy delivery (starting at 5 to 10 W and escalating by 5 W until 25–30 W) allow highly accurate lesions, which can be stopped immediately in the case of junctional acceleration and/or AV block [20]. However, the drawback is that this strategy can result in local oedema thus reducing direct contact between the catheter and the fibre. The use of a long (preferably steerable) sheath, a superior approach via the jugular vein, and performing the ablation during apnea may also increase catheter stability and help to limit the risk for AV node injury.

An alternative approach for antero-septal accessory pathways is mapping and targeting the accessory pathway from the aortic cusps. This approach can be safely performed with adequate precautions and may be considered in cases with failed previous procedures and/or high risk of AV block with the standard right-sided approach [21, 22].

In the case of postero-septal pathway ablations, an irrigated-tip catheter may help target deep myocardial or epicardial substrates [23], rarely requiring combined right and left atrial applications, or intracoronary applications. Because coronary sinus aneurysms are also related to long and difficult procedures and high incidence of recurrence, we advocate for very thorough understanding of the anatomy using an angiogram, 3D electro-anatomical mapping, or intracardiac echocardiography.

Recurrent accessory pathway ablation in children remains a challenge even in expert’s hand. There is inherent logic to adapt the procedure in terms of energy used, catheter choice and mapping strategy depending on the location of the pathways. Technology improvements have been impressive in recent years and collaborate to improve procedure outcomes. However, the cornerstones of successful and safe paediatric accessory pathway ablation remain a deep understanding of the anatomic landmarks, meticulous mapping including pacing strategies when necessary, and catheter stability with adequate energy delivery during applications.

This is a retrospective observational study without a matched control group, which limits the strength of many observations. We acknowledge that the reason for a substrate recurrence may have been oversimplified in some patients.