First-in-children epicardial mapping of the heart: unravelling arrhythmogenesis in congenital heart disease

Abstract

Patients with congenital heart disease (CHD) are prone to develop atrial and ventricular arrhythmias. Multiple factors throughout life contribute to arrhythmogenicity substrate such as (i) (longstanding) volume and/or pressure overload, (ii) scar tissue, (iii) ageing-related structural remodelling, (iv) cardiovascular risk factors and (v) tachycardia-induced remodelling. At present, it is unknown whether, and to what extent, paediatric patients with CHD have atrial or ventricular conduction disorders early in life and whether there is a correlation between duration of volume/pressure overload and extensiveness of conduction disorders. To investigate this, we initiated high-resolution intraoperative epicardial mapping in paediatric patients with CHD undergoing primary open-heart surgery.

INTRODUCTION

Patients with congenital heart disease (CHD) are prone to developing atrial and ventricular tachyarrhythmias relatively early in life. Improvements in perioperative care, patient monitoring and surgical techniques have resulted in an ageing CHD population. An inevitable consequence of this growing population is the rising number of CHD patients presenting with tachyarrhythmias. CHD patients not only have multiple triggers, but they also have a complex underlying arrhythmogenic substrate. Long-standing volume and/or pressure overload, scar tissue, ageing-related structural remodelling, cardiovascular risk factors and tachycardia-induced remodelling all contribute to the development of an arrhythmogenic substrate [1]. However, the pathophysiology of tachyarrhythmias is still incompletely understood. At present, it is unknown whether, and to what extent, paediatric patients with CHD already have atrial or ventricular conduction disorders early in life, as cardiac mapping in this population has never been performed before. The presence of conduction disorders early in childhood may explain the increased vulnerability of CHD patients to develop arrhythmias in adulthood, despite CHD repair.

We therefore aim to (i) quantify and characterize atrial and ventricular conduction disorders in paediatric CHD patients early in life and (ii) correlate the extent and severity of conduction disorders with patient characteristics such as duration of volume/pressure overload. We therefore performed high-density and high-resolution epicardial mapping studies of the paediatric heart. In this report, we present our initial experience with this first-in-children mapping technique.

METHODS: EPICARDIAL HIGH-DENSITY AND RESOLUTION MAPPING APPROACH

Parents gave informed consent to participate in the study protocol approved by our local ethics committee (MEC-2019-0543). The epicardial mapping approach of the atria applied in paediatric patients has previously been used in adult patients [2]. In paediatric patients, we also performed mapping of the right and left ventricle. Due to potential epicardial adhesions during redo surgery, at present, we only include patients undergoing primary cardiac surgery.

Before commencement of extracorporeal circulation, a unipolar epicardial pacemaker wire, serving as a temporal reference electrode, was stitched to the superior lateral wall of the right atrium (RA). A clamp fixed to the sternum served as indifferent electrode. A custom-made electrode array (192 electrodes, electrode diameter 0.6 mm, interelectrode distance 2.12 mm) was used to record atrial and ventricular unipolar electrograms during sinus rhythm (upper left panel of Fig. 1).

Figure 1:

(A) The mapping electrode array and the predefined atrial and ventricular mapping locations. (B) Epicardial mapping, the corresponding electrograms and colour-coded activation maps. A: atria; Ao: aorta; BB: Bachmann’s bundle; IVC: inferior vena cava; LA: left atrium; LAD: left anterior descending artery; LV: left ventricle; mV: millivolt; RA: right atrium; RV: right ventricle; SVC: superior vena cava; V: ventricles.

Mapping positions

The electrode array was attached to a spatula that can slightly be bent to match the curvature of the atria and ventricles. The electrode array was then shifted across the epicardial surface in a predefined order. At the RA, the electrode array was positioned perpendicular to the caval veins and mapping started at the top of the RA appendage, with the tip of the electrode array towards the superior caval vein. The electrode array was then moved downwards over the RA appendage until the cavotricuspid isthmus was reached (upper panel of Fig. 1). The intercaval region, including the terminal crest, was mapped with the electrode positioned in the longitudinal direction of both caval veins. Bachmann’s bundle was mapped from the superior cavo-atrial junction towards the left atrial appendage. For the left atrium, mapping was performed at the left atrial appendage, and the posterior wall with pulmonary vein region was reached via the oblique sinus. The right and left ventricles were mapped with the electrode array positioned parallel to the left anterior descending artery with the tip of the electrode array towards the apex of the heart (upper panel of Fig. 1). The electrode was then shifted to the lateral wall of the ventricles (perpendicular to the atrioventricular groove). The number of locations mapped per region was determined by the size of the atria and ventricles. Epicardial mapping was performed for 5 s at each location. Every recording included a surface electrocardiogram lead, a calibration signal, a unipolar reference electrogram and all epicardial unipolar electrograms. Recordings were analogue-to-digital converted (16 bits), sampled with a rate of 1 kHz, amplified (gain 1000) and filtered (bandwidth 0.5–400 Hz). All recordings were manually checked by 2 investigators and analysed offline using custom-made Python software.

First experience with high-resolution mapping in paediatric patients

Epicardial mapping was performed in 3 patients scheduled for repair of atrial septal defect (ASD) type II (n = 2, both ♀, 3 years old) and complete atrioventricular septal defect (♂ 4 months old) without history of cardiac arrhythmias. All mapping procedures were performed within 8 min by a dedicated, trained surgical and electrophysiology team. All atrial and ventricular regions were accessible for epicardial mapping, also in our smallest 4-month-old patient. Complications did not occur during the mapping procedures.

Data analysis: epicardial activation maps

Examples of atrial and ventricular electrograms recorded from the RA and right ventricle are demonstrated in the lower panel of Fig. 1. Colour-coded local activation time maps were constructed by annotating the steepest negative slopes of unipolar potentials. The lower panel of Fig. 1 illustrates examples of colour-coded activations maps of the superior RA and the right ventricle. White asterisks indicate the earliest activated regions of the RA and ventricle; arrows display main trajectories of the wavefronts. After sinus node activation, the wavefront spreads over the RA in a radial fashion. First site of activation at the right ventricle occurs in the anterior paraseptal region and spreads in the inferior and superior direction, towards the apex and the right ventricular outflow tract, respectively.

DISCUSSION

This study is the first to report on high-density and resolution mapping of the epicardial surface in paediatric patients with CHD undergoing primary open-heart surgery. So far, excitation of the atria and ventricles has never been studied in paediatric patients without cardiac arrhythmias, as there is no indication for cardiac mapping. As a consequence, there is no literature on patterns of activation and hence cardiac conduction disorders in paediatric patients with CHD.

In adult ASD patients without a history of cardiac arrhythmias (n = 13, mean age 41.6 ± 10.3 years), Morton et al. [3] demonstrated enhanced conduction disorders at the terminal crest which persisted beyond percutaneous ASD closure. However, whether enhanced conduction disorders were the result of chronic volume overload, structural remodelling or simply a result of ageing could not be addressed. Therefore, our paediatric epicardial mapping approach will provide unique insights into the early effects of volume/pressure overload on atrial and ventricular activation patterns and conduction properties.

Prior studies demonstrated that a longer duration of volume/pressure overload correlates with a higher prevalence of atrial and ventricular arrhythmias [1, 4]. Older age at time of ASD repair was associated with a higher prevalence of both preoperative and postoperative atrial tachyarrhythmias [5, 6]. Long-standing volume overload of the RA results in stretch of the atrial wall, thereby inducing structural remodelling. Ueda et al. [7] examined RA tissue samples of 65 adult ASD patients undergoing primary repair and compared them with RA tissue samples of age-matched controls. Structural changes, including fibrosis, were more pronounced in ASD patients than in healthy age-matched controls. Macchiarelli et al. [8] observed similar findings in paediatric ASD patients.

Ventricular arrhythmias and sudden cardiac death are also common sequelae in those who have undergone ventriculotomy or ventricular septal defect closure and in patients with a systemic right ventricle, Ebstein’s anomaly and Eisenmenger’s syndrome [1]. Although less common, ventricular arrhythmias may also arise due to long-standing volume/pressure overload, independently of direct surgical scarring. Chowdhury et al. [9] examined biopsies from the right ventricular outflow tract in patients with tetralogy of Fallot. Structural remodelling, including fibrosis, was already present in young (<4 years) patients with left ventricular volume/pressure overload, though structural remodelling was more pronounced in older patients (>4 years). As structural remodelling underlies development of conduction disorders, it is most likely that atrial and ventricular conduction disorders are already present in young paediatric CHD patients, predisposing these patients to cardiac arrhythmias relatively early in life.

Epicardial mapping: valuable or redundant?

In clinical practice, cardiac mapping is usually performed from the endocardium, thereby assuming that atrial excitation is a 2-dimensional process. Nowadays, the 3-dimensional nature of cardiac excitation and cardiac arrhythmias is increasingly acknowledged. In order to unravel the 3-dimensional substrate of complex cardiac arrhythmias, such as atrial fibrillation, it is of utmost importance to also study epicardial conduction.

Clinical relevance and future directions

Our epicardial mapping approach enables further unravelling of arrhythmogenesis in CHD patients. Primary cardiac surgery in paediatric CHD patients is a once-in-a-lifetime opportunity to investigate epicardial conduction properties in the absence of confounding factors such as history of cardiac arrhythmias and ageing. As CHD patients may present at different ages, we will be able to investigate characteristics of atrial and ventricular conduction disorders from childhood to adulthood [2] and correlate these findings with patient characteristics such as age and type of CHD. The presence of conduction disorders early in life may be the result of structural remodelling induced by short-lasting volume/pressure overload. In this case, one could argue for an expedited intervention strategy in order to prevent structural and hence electrical remodelling. On the other hand, intraoperative cardiac mapping may enable identification of CHD patients who are at high risk for development of postoperative atrial tachyarrhythmias. This may favour prophylactic arrhythmia surgery in this high-risk population, which is currently a matter of debate.

Knowledge of characteristics of conduction disorders is crucial to identify patients at high risk for development of cardiac arrhythmias, to adapt treatment strategies or timing of interventions and to develop novel therapies aimed at prevention of structural and electrical remodelling.

Limitations

There is no cardiac mapping data of children without structural heart disease for comparison. Endocardial mapping is not possible with the present epicardial mapping approach.

Conflict of interest: none declared.

Author contributions

Rohit K. Kharbanda: Conceptualization; Data curation; Formal analysis; Investigation; Project administration; Writing—original draft. Mathijs S. van Schie: Data curation; Investigation; Writing—review & editing. Wouter J. van Leeuwen: Data curation; Investigation; Writing—review & editing. Yannick J.H.J. Taverne: Data curation; Investigation; Writing—review & editing. Charlotte A. Houck: Data curation; Project administration; Writing—review & editing. Janneke A.E. Kammeraad: Project administration; Writing—review & editing. Ad J.J.C. Bogers: Conceptualization; Data curation; Methodology; Supervision; Writing—review & editing. Natasja M.S. De Groot: Conceptualization; Investigation; Methodology; Supervision; Writing—review & editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Shinichiro Ikeda, André Rüffer, Gus J. Vlahakes and the other, anonymous reviewer(s) for their contribution to the peer-review process of this article.