Ablation of Postoperative Atrial Tachycardia in Congenital Heart Disease

Epidemiology and Natural History

IART is common in patients with CHD, with a spectrum of arrhythmia presentations ranging from occult, asymptomatic arrhythmia to sudden death. Incessant or recurrent arrhythmia may cause gradual hemodynamic deterioration, and vice versa, often resulting in a vicious cycle of clinical decompensation. Thrombosis and thromboembolic events and likely increased risk of mortality are also associated with IART ( Fig. 13.1 ). Symptoms, frequent need for hospitalization, and the management of cardiac devices and antiarrhythmic drugs constitute a significant burden on quality of life.

An early, retrospective multicenter study of young patients with atrial flutter showed that over 80% had associated repaired or unrepaired CHD, and also reported a mortality rate caused by sudden death of 10% over 6.5 years. ATs in an adult CHD population are associated with a twofold increased risk of mortality ; independent predictors for mortality are poor functional class, single ventricle physiology, pulmonary hypertension, and valvular heart disease. Other studies have come to differing conclusions regarding the direct association of IART with mortality, but have noted that it co-occurs with congestive heart failure and thrombosis. Regardless, it seems clear that IART is a fellow traveler with a variety of adverse late sequelae of CHD.

The natural history of IART is characterized by gradual loss of normal sinus node function over years to a decade or more, followed by increasingly frequent recurrences of tachycardia. Risk factors for IART are older age at operation, preoperative and perioperative occurrence of arrhythmia, and longer follow-up. A study of nearly 500 early survivors of the Mustard procedure identified an IART prevalence of 27% at 20-years follow-up, and 60% had sinus node dysfunction, frequently in association with IART. Sudden death occurred in 6.5% of patients over a mean follow-up of 11.6 years. A retrospective study on the occurrence of atrial flutter in 53 patients with tetralogy of Fallot was performed by Roos-Hesselink and coworkers. They found that sinus node dysfunction and IART each were present in about a third of the population after mean follow-up of 18 years, more prevalent than ventricular tachycardia (VT), and more likely to be associated with symptoms. Among patients who have undergone the Fontan procedure for single ventricle physiology, the natural history of atrial arrhythmia is described by retrospective studies from several large cardiac surgical centers totaling around 1400 patients. These indicate that between 25% and 50% of patients who have undergone the Fontan procedure will have clinical documentation of IART by 10 years of follow-up.

Mechanisms of Intraatrial Reentrant Tachycardia

In addition to the anatomic differences observed in CHD patients, the atrial myocardium itself is often markedly thickened and fibrotic arising from lifelong exposure to abnormal hemodynamic stresses, cyanosis in many cases, and intermittent inflammatory effects of surgical intervention ( Fig. 13.2 ). Electrophysiologic testing in patients after Fontan repair of single ventricles and Mustard repair for transposition of the great arteries demonstrates prolongation of atrial refractoriness and areas of intraatrial conduction delay. Progression of the aforementioned arrhythmia findings may be associated with gradual prolongation of the P wave, and measurements of the atrial histology of patients with a variety of complex CHD lesions suggest that increasing hypertrophy and fibrosis are also present. Finally, all patients with CHD are also predictably vulnerable to the automatic and triggered atrial arrhythmias in the perioperative setting, in the context of local inflammation, metabolic stress, inotropic support, and alterations in hemodynamic loading. Each of these factors may complicate diagnosis and management of atrial arrhythmias.

The most common mechanism underlying IART in CHD is macroreentry within the atrial musculature. Anatomic structures, areas of scar tissue, long suture lines, cannulation sites, or surgically inserted prosthetic materials are often the boundaries of these reentrant circuits. Separation of atrial muscle bundles by fibrous tissue enhances the complexity of reentrant circuits as they form multiple corridors within areas of scar tissue. Ectopic ATs are less frequently observed. Their underlying mechanism is unclear, although focal activity originates from low-voltage areas and mapping studies are suggestive of microreentry.

Animal Models of Intraatrial Reentrant Tachycardia

Animal models have been used to mimic the atrial anatomic changes associated with the surgical palliation of CHD. These models are highly arrhythmogenic and appear relevant to the understanding of postoperative atrial reentrant tachycardias. Acute and chronic models of the classic and lateral tunnel varieties of the Fontan procedure have shown that extended atrial suture lines and surgical incisions serve as common pathway for many observed tachycardias. In carefully mapped preparations, the suture line associated with the baffle appeared to serve as the primary determinant of the arrhythmia circuit. Inclusion of the crista terminalis in the suture line further increased vulnerability to arrhythmia. These arrhythmogenic substrates could be rendered noninducible by surgically anchoring the suture line to the nonconductive boundary of the tricuspid annulus.

Electrocardiographic Manifestations of Intraatrial Reentrant Tachycardia

IART typically has a stable cycle length and P wave morphology, suggesting that it is organized by a fixed myocardial substrate. Diagnostic criteria for IART in patients with CHD are listed in Box 13.1 . A typical, ambulatory electrocardiographic example is shown in Fig. 13.3 , demonstrating the frequent association of sinus node dysfunction and variable atrioventricular (AV) conduction. Although IART may sometimes share the electrocardiographic characteristics of common atrial flutter, atypical electrocardiographic morphologies are common. P waves are frequently discernible, separated by relatively long segments of isoelectric electrocardiographic baseline. Cycle length is typically significantly longer than that seen in atrial flutter, especially in Fontan patients, and may permit 1:1 AV conduction.

Although any specific occurrence of IART is generally stable, multiple, different IART morphologies may be recorded from a single patient over time. This may represent reversal of activation of a given IART circuit, use of an alternate circuit, or changes in passive activation of the atrium outside the arrhythmia circuit itself. Schoels and coworkers performed high-density epicardial mapping on animal preparations of a variety of atypical atrial flutters induced using a sterile pericarditis model. After classifying these as either flutter or P wave tachycardias, they demonstrated that periods of isoelectric atrial diastole were correlated with activation of narrow corridors of slowly conducting atrial tissue, whereas maps of flutter wave tachycardias were less likely to display such features. This electrocardiographic discrimination between flutter wave and P wave morphologies may be practically useful with respect to designing ablation strategies in these patients ( Fig. 13.4 ).

Anatomic Complexity of Arrhythmia Phenotypes in Congenital Heart Disease

The presence of CHD complicates understanding of the mechanisms of atrial macroreentrant circuits. Unfamiliar anatomic relations, arising from both the congenital malformations themselves and the variety of palliative surgical procedures used to redirect blood flow and septate the heart, may have significant impact on management and ablative strategies. In addition to analysis of the targeted arrhythmia, the physician must have complete and specific knowledge of the patient’s cardiovascular anatomy and the consequences of that anatomy and subsequent surgical modifications on cardiovascular function.

There are also specific subtypes of CHD that are associated with other forms of supraventricular tachycardias (SVTs) that may complicate diagnosis and management or even co-occur with atrial macroreentrant tachycardias. Ebstein’s anomaly has a high prevalence of accessory pathway–mediated tachycardias with and without associated preexcitation. Patients with anatomically complex forms of heterotaxy syndrome may sometimes have AV reciprocating tachycardias based on twin AV nodes. Finally, a variety of complex lesions may rarely demonstrate atypical AV nodal reentrant tachycardias, in a setting that often makes it impossible to ascertain the position or anatomic relationships of the AV node and its inputs.

Surgical Anatomy of Repairs

Certain combinations of CHD diagnoses and surgeries are associated with increased prevalence of IART and/or special considerations with respect to ablative treatment. From the point of view of arrhythmia diagnosis and management, one can classify most lesions into one of the four broad groups ( Table 13.2 ).

The first group consists of those patients whose surgical repair left them with a normally septated heart and normal venoarterial connections (e.g., atrial septal defect [ASD] or ventricular septal defect, tetralogy of Fallot, many endocardial cushion defects).

The second group includes patients who have undergone an atrial switch procedure, such as the Mustard or Senning procedure and patients who have undergone many of the baffle variants of the Fontan procedure for palliation of single ventricle physiology. In these procedures, the surgeon creates a tissue or prosthetic baffle to redirect venous blood into the pulmonary circulation and arterial blood into the systemic circulation. This results in extended atrial suture lines and typically leaves a significant portion of the right atrium (often including the cavotricuspid isthmus) located on the pulmonary venous side of the baffle.

The third group consists of a dwindling but still an important group of patients who underwent one of the older variants of the Fontan procedure, most commonly the atriopulmonary anastomosis. These patients have very enlarged, fibrotic, and surgically scarred right atria that are highly arrhythmogenic and which present unusual and specific challenges to ablation.

The final group consists of those patients who are complicated by virtue of their anatomy but are unpalliated, including many patients with Ebstein’s anomaly or patients with unrepaired single ventricle. In this heterogeneous group, the anatomy that determines the arrhythmia is by definition native. Because these patients may also be cyanotic, their atrial myocardium is exposed to long-term stress, which may enhance its arrhythmogenic potential.

Studies of Clinical Mechanism

Because IART is usually well tolerated hemodynamically, it has been possible to study clinical tachycardia mechanisms in detail. This has allowed evaluation of hypotheses derived from animal models and electrocardiographic observations and made it possible to correlate anatomic and electrophysiologic mechanisms. In turn, empirical anatomic approaches to the treatment and prophylaxis of IART based on these observations are being developed and tested. These techniques may use either linear ablative techniques in the catheterization laboratory or application of linear cryoablative lesions in the operating room, with varying degrees of direct electrophysiologic guidance.

Interruption of a tachycardia during ablation provides evidence that the ablation site is within the tachycardia circuit, and initial evaluation of successful IART ablation sites in patients with CHD demonstrated that critical sites for circuit interruption are dependent to some degree on the primary congenital heart lesion. In patients with CHD, the great majority of IART circuits are located in the right atrium, and most of those are based on a limited set of common arrhythmia substrates. An important rule is that, among the great majority of patients who have an isthmus defined by the inferior vena cava and a right-sided AV valve (e.g., biventricular repairs, Mustard and Senning patients), the most common IART is AV valve–caval isthmus dependent, similar to patients with the common form of atrial flutter. Although the AV valve may be either mitral or tricuspid and located in either the systemic or pulmonary venous atria, successful ablation for these IART circuits typically targets the area anterior to the Eustachian ridge ( Fig. 13.5 ). Reentrant circuits using the isthmus but anchored to the ostium of the inferior vena cava (pericaval reentry) have also been described.

Analysis of successful ablation sites along the lateral right atrial wall has also highlighted the importance of conduction block caused by right atriotomy scars in the genesis of incisional ATs. Such scars are virtually ubiquitous among patients with CHD and may result in tachycardias arising from the free wall or as dual-loop tachycardias in conjunction with an isthmus-dependent circuit. The exact location of atrial scar may be difficult to determine, and a useful approach to defining their location is to identify lines of double or split potentials. Clusters of double potentials represent continuous atrial scar acting as a central obstacle that defines the IART circuit, and which can be functionally eliminated using ablation to extend the scar to a nonconductive boundary. Love and coworkers characterized a variety of anatomic and electrophysiologic central obstacles, which served to anchor IART circuits, including the right AV valve (when present), ASDs, and surgical scars located on the free wall of the right atrium, a finding which has been further defined recently in patients with tetralogy of Fallot. Examples of such IART circuits have been characterized in their entirety in a number of cases by anatomic mapping of the response to entrainment pacing.

Certain patients, notably those with the older variants of the Fontan procedure, have patchy areas of atrial scar in the atrial free wall separated by channels of viable myocardium, which appear to be the critical substrate for the associated IARTs ( Fig. 13.6 ). The overall electrical amplitude of endocardial signals measured from the right atrium is markedly decreased in patients with CHD, particularly in the area surrounding the crista and at nearby sites of surgical intervention. Critical channels of active myocardium, coursing amidst complex islands of atrial scar defined by arbitrary voltage criteria, have also been successfully targeted for ablation.

Focal ATs have also been observed to occur with some frequency among these patients. These tachycardias most commonly arise from the right atrium and are triggered by pacing. Careful mapping studies by De Groot and colleagues suggest that these IARTs represent areas of atrial microreentry occurring in circumscribed areas of diseased tissue. Clinical demonstration of microreentrant circuits consistent with these findings, delineated by high-density catheter-based mapping arrays and associated with long, fractionated electrograms and termination by nonpropagated extrastimulus, has recently been demonstrated in case report.