Ventricular arrhythmia in congenital heart diseases with a systemic right ventricle

Abstract

Congenital heart disease (CHD) often involves the systemic right ventricle (SRV), which is the morphological right ventricle that supports systemic circulation. SRV patients are at a higher risk of sudden cardiac death (SCD) than other adult CHD patients and continues to be a significant cause of death in this aging population. However, the pathophysiology of ventricular arrhythmias in SRV is still not fully understood, and there may be differences between subtypes of CHD. Although these events are rare, predicting them is challenging. This review discusses contemporary strategies for assessing and preventing the risk of ventricular arrhythmias in SRV patients. Several risk factors have been identified to be associated with ventricular arrhythmias in patients with SRV. A recent risk stratification model combines independently associated factors into a risk score, and subpulmonary left ventricle dysfunction is emerging as a critical factor in risk assessment. Cardiac magnetic resonance imaging, biomarkers, and genetic data may refine the ability to predict ventricular arrhythmias in SRV. However, the question of whether implantable cardioverter-defibrillators (ICDs) should be used as a preventive measure in this cohort remains unanswered. Multicenter studies are needed to evaluate risk models and ICD use in this aging population. Given that ICDs have drawbacks, such as a high rate of inappropriate shocks and late lead-related complications, shared clinical decision-making is crucial when considering their use. The review emphasizes the need for further research in this area to improve the identification of patients at risk of clinical ventricular arrhythmias and to develop effective prevention strategies.

1. Introduction

The systemic right ventricle (SRV), defined as the morphological right ventricle supporting the systemic circulation, is relatively common in congenital heart disease (CHD) concerning approximatively 10% of all forms of CHD. The most common anatomical defects where SRV is encountered are complete transposition of the great arteries (D-TGA) with previous atrial switch repair (Mustard/Senning operation) and congenitally corrected transposition of the great arteries (L-TGA). This anatomy is also encountered in hypoplastic left heart syndrome palliated with the Norwood-Fontan protocol, double inlet right ventricle and double outlet right ventricle mostly with previous Fontan palliation [1]. Regardless of the underlying cardiac anatomy, the prognosis for patients with a SRV is guarded compared to patients with a systemic left ventricle (LV). Arrhythmias, SRV dysfunction, heart failure (HF) and sudden cardiac death (SCD) represent important late complications. Patients with SRV, particularly those with D-TGA and atrial switch repair, are among the population with CHD at highest risk of SCD. Data on risk stratification for primary prevention and hence implantation of implantable cardioverter defibrillator (ICD) remains scarce such that patient selection and timing remain challenging. This article provides an update on mechanisms involved in the genesis of ventricular arrhythmias (VAs), risk stratification approaches to patients with SRV for VA and SCD, and the treatment and prevention of VAs in patients with SRV.

2. Epidemiology

Sudden cardiac death is the second most common cause of death in SRV after heart failure, accounting for 16–20% of deaths [2,3]. In the recent European MAREs (Major adverse ventricular Arrhythmias and Related Events) cohort of adult patients with a SRV, events including SCD, sustained ventricular tachycardia (VT), and appropriate implantable cardioverter-defibrillator (ICD) therapy, were reported in 5% of patients at a median follow-up of 9 years with an incidence of 6.3 per 1000 patient-years. The incidence of MAREs increased with age and occurred at a younger age in D-TGA patients with atrial switch [2,3], suggesting a difference in the pathophysiology of VAs between D-TGA and L-TGA.

D-TGA is one of the most common cyanotic heart defects in newborns and is prevalent in about 1 of 3000 live births [4]. Radical surgery in the form of the atrial switch operations was routinely performed in the 1960s and 1970s and gradually replaced by the arterial switch operation in the 1980s. The latter surgery restores the left ventricle to the subaortic position such that only the older subset of patients with D-TGA who had a Mustard or Senning correction has a SRV. These patients are now well into adulthood, rendering long-term follow-up possible [3,5]. D-TGA palliated by atrial switch is among the congenital heart defects at highest risk for SCD [6]. A population-based study reported an incidence of 4.9 per 1000 patient-years, second only to aortic stenosis and more than threefold greater than tetralogy of Fallot [7]. L-TGA is an even rarer type of congenital heart malformation accounting for approximately 1/33,000 live births [8]. The incidence of sudden death due to documented VAs, which accounts for 10–15% of all deaths in L-TGA, was estimated to between 1.8 and 25 per 1000 patient-years [6].

Among VAs, ventricular ectopy and non-sustained (<30-s) VT (NSVT) are relatively common in adults with a SRV, and NSVT seems to be related to sustained VT and SCD [2,9,10]. Monomorphic VTs constitute 49% of all VAs, the remaining being polymorphic VT in 34%, and VF in 17% [10] (see Fig. 1).

Fig. 1.

Patient with congenitally corrected transposition of the great arteries and biventricular implanted cardioverter defibrillator who received a shock to treat ventricular fibrillation following an R-on-T phenomenon.

3. Arrhythmogenic substrate

Structural and functional alterations as well as neurohumoral remodeling appear to contribute to VA substrates in SRV. In general, the three mechanisms for VAs are abnormal automaticity, triggered activity, and reentry. It remains to be determined to what extent each contributes to the incidence of VAs in patients with SRV. Automatic VAs typically arise from subendocardial Purkinje tissue or adjacent ventricular myocardium due to altered phase 4 depolarization of the cardiac action potential. Ventricular arrhythmias due to triggered activity are typically due to early (EADs) or delayed (DADs) afterdepolarizations [11,12]. EADs are related to a net inward current during phase 2 or 3 of the action potential and may be responsible for some VAs in susceptible patients, particularly those who receive antiarrhythmic drugs. In contrast, DADs usually occur in phase 3 to 4 of the action potential due to abnormal calcium cycling and are classically associated with outflow tract tachycardias. Finally, when normal electrical propagation is functionally or anatomically blocked, such as by ventricular scar, the alternative path could result in the occurrence of VAs via reentrant looping [13]. Reentry is thought to be the predominant mechanism for VAs in patients with CHD in general but detailed mechanistic studies in the subpopulation with SRV are lacking.

3.1. A. Arrhythmogenic mechanisms common to D-TGA and L-TGA

Myocardial fibrotic scar regions cause conduction block and propagation barriers. Fibrosis separates cardiac myocyte bundles, forcing the excitation to form a looping circuit surrounding the bundles [14]. Slow conduction and fibrous anatomic barriers create the basis of reentry responsible for ventricular arrhythmogenesis. Myocardial fibrotic scars are frequent in SRV [15,16] Fig. 2.

Fig. 2.

Late gadolinium lesions in patients with D-TGA. LGE distribution was mainly in the basal segments of the sRV free wall (Panel B) and correlation with macroscopic fibrosis in heart explanted (Panel C).

Different mechanisms are postulated to explain their presence. Preoperative hypoxemia and deficient myocardial protection in older cohorts may play a role since fibrosis is more extensive in older patients and those with a late repair [[16], [17], [18], [19]]. A defici ent coronary supply for a thickened SRV may be another explanation. Indeed, SRV fibrosis has been related to increased RV wall stress [20]. Chronic systemic afterload leads to SRV maladaptation with excessive hypertrophy, which could present demand-supply mismatch ischemia, resulting in the presence of fibrosis. This phenomenon may be enhanced by rapid heart rates such as atrial arrhythmias which are the factor most consistently associated with sudden death in patients with a SRV [21].

In D-TGA after atrial switch and L-TGA, progressive SRV dilatation and dysfunction are the norm, leading to heart failure (HF) and VA, with a confluence of mechanisms involved in these events [3]. To improve cardiac hemodynamic function including cardiac output, numerous compensatory reactions occur in HF. These adaptations include ventricular hypertrophy, high ventricular filling pressure, as well as increases in cardiac preload and afterload. These mechanical changes in HF can contribute to ventricular arrhythmogenesis by pro-arrhythmic electrophysiological remodeling (such as shortening of the repolarization phase of the action potential, cell-to-cell uncoupling, and sub-endocardial ischemia) [22,23].

The progression of HF is accompanied by significant intracellular and extracellular ionic and metabolic changes in the myocardium that result in depolarization of the resting membrane potential, slow conduction, dispersion of repolarization, and abnormal automaticity, triggering VAs [24]. Although these phenomena have best been characterized in LV failure, they also appear to be implicated to some extent in RV failure. Excessive sympathetic activation is also a major feature in patients with HF, which is manifested by elevated plasma levels of epinephrine and norepinephrine and overactivation of cardiac sympathetic nerve fibers innervating the myocardium. This neurohormonal activation has been observed in HF related to CHD including SRV [25]. Cardiac sympathetic overactivation contributes to HF progression, and can trigger VAs and sudden cardiac death [26]. The role of cardiac sympathetic hyperactivation in SRV is consistent with the lower rate of VAs in patients with D-TGA treated with β-blockers [10].

Owing to the aging population with a SRV, they are increasingly subject to traditional cardiovascular risk factors. Consequently, the prevalence of coronary artery disease and myocardial infarction in patients with CHD is on the rise [27,28] and could potentially contribute to additional substrates for VAs in patients with a SRV. Optimal screening for and management of cardiovascular risk factors and coronary artery disease should be integrated into a global approach to reducing risk for VAs in this population.

3.2. B. Arrhythmogenic mechanisms specific to D-TGA and L-TGA

In L-TGA, scars secondary to cardiac surgery are more common considering that >75% undergo surgery for ventricular septal defect closure and/or pulmonary atresia/obstruction. These constitute potential substrates for reentry. The abnormal cardiac conduction system, which is part of the L-TGA phenotype, may increase the risk of sudden death. Daliento et al. have described fibrosis and disruption of the proximal non-bifurcating His bundle which can lead to complete heart block or constitute an underlying arrhythmogenic myocardial substrate [29]. Macroreentrant VT involving an impaired His–Purkinje system and the ventricle as a reentrant circuit has been described in L-TGA [30]. Indeed, unusual atrioventricular nodes and His bundle anatomy can lead to conduction disturbances with advancing age and predispose to bundle branch reentrant VT. An Ebstein-like tricuspid valve is observed in approximatively 6% of patients with L-TGA and can be associated with accessory pathways which, if multiple or rapidly conducting, can likewise trigger SCD.

In D-TGA and atrial switch, more than 80% of sudden death events occur during exercise [21]. Moreover, acute massive myocardial infarction of the SRV, in the setting of chronic subendocardial ischemic lesions in the absence of conventional coronary atherosclerosis, has been observed in autopsies of patients who suddenly died [31]. These finding may be explained by the impaired stroke volume response due to poor atrial transport [32] which can be aggravated by rapid heart rates (sinus tachycardia or atrial arrhythmias), and which combined with increased myocardial oxygen demand of the pressure-loaded SRV and inefficient coronary circulation, may provoke ischemia-related VAs [33]. Differences in underlying mechansisms for SCD in patients with D-TGA and L-TGA are summarized in Fig. 3 [34].

Fig. 3.

Potential triggers and substrates for SCD in D-TGA and L-TGA (congenitally corrected TGA). Strengths of associations are semi-quantitively estimated by number of asterisks from absent/weak (*) to moderate (**), strong (***), and very strong (****). Reproduced with permission from Khairy P. Eur Heart J 2022; 3:2695–7 [34].

4. Risk stratification

Risk stratification is defined as an ongoing process of assigning patients a particular risk status. The goal of risk stratification is to address specific population management challenges, match risk with levels of care, individualize treatment plans to lower risk and improve function, and align the practice with a value-based care approach. Outcomes such as SCD in patients with a SRV are predominantly stochastic events and are difficult to predict in any given individual with respect to occurrence and timing. The objective of risk stratification in such situations should be to quantify probabilities for SCD and weigh risks and anticipated benefits of ICD therapy to identify suitable candidates for primary prevention devices.

Despite SRV being among the congenital heart defects at highest risk for SCD, the estimated annual incidence of ∼6/1000 patient-years is an order of magnitude lower than for adults with primary prevention ICD indications for ischemic or dilated cardiomyopathy [2]. However, the SCD event rate is not linear, with the majority of deaths occurring in older patients [2,35,36]. Several observational studies have assessed predisposing factors (Table 1).

Table 1.

List of ventricular arrhythmia predictors assessed in SRV. * indicates that the studied endpoint was composite and included ventricular arrhythmias. eGFR, estimated glomerular filtration rate; GDF-15, growth differentiation factor-15; HF, heart failure; LGE, late gadolinium enhancement; LV, left ventricle; LVOT, left ventricle outflow tract; LS, longitudinal strain; M; NT-proBNP, N-terminal pro B-type natriuretic peptide; NYHA, New York Heart Association functional class; RDW, red cell distribution width; SRV, systemic right ventricle; TGA, transposition of the great arteries; TR, tricuspid regurgitation.

Predictive factors	D-TGA	Both	L-TGA
Patients characteristics	Complex anatomy [37,38] Mustard surgery [37], Pulmonary hypertension [2]	Age [2,36]* [35] Surgical repair at older age [36,39], History of HF [2,21,39] [1,8,12]
Clinical variables	Palpitations [2,21], syncope [2] NYHA class ≥ III [2,38]
Electrocardiography	Prior ventricular arrhythmia [36]* Lack of beta-blockers [10],	Atrial arrhythmia [2,9,10,21,37,39] QRS duration [2,35,38] NSVT [2,9]	Pacemaker [2]
Echocardiography	Moderate to severe LVOT obstruction [2]	Moderate to severe TR [2,36,38,39]* Moderate to severe SRV dysfunction [2,35,36,38]* [1,4,6,9] Septal longitudinal strain Subpulmonary LV dysfunction [36]* [35]	Ebstein-like tricupid valve?
CMR	SRV LGE [16,17]*
CPET	Percent predicted peak VO2 [16]*
Biomarkers	Genetic variants [40]	GDF15, RDW, Galectin 3, Hs-troponinT, NT-proBNP, eGFR [41]*

The most consistent risk factors identified to date are atrial tachyarrhythmia and HF (Table 1). However, new markers may improve the individual prediction of VAs in CHD with a SRV, and contribute to our understanding of higher-risk substrates. For example, markers of sub-pulmonary LV function are emerging as potential predictors of cardiovascular outcomes in TGA including VA events, suggesting the potential for subpulmonary LV physiology to impact VA risk. In a prospective study including 33 patients with a SRV, remodeling of the subpulmonary LV both at rest and at peak exercise were significantly associated with the first cardiovascular event including VAs [42]. Evaluation of subpulmonary LV function was also included in a risk score model that estimated the risk of major clinical events in D-TGA after atrial switch [36]. In the MAREs registry, we observed that patients with at least moderate LV outflow tract (LVOT) obstruction (>36 mmHg) had a higher risk for VAs [2]. It appears that when both ventricles are pressure overloaded (i.e., the SRV with systemic load and subpulmonary LVOT obstruction) with ventricular remodeling, risk for VAs and SCD is significantly increased. Patients with myocardial fibrosis visualized by late gadolinium enhancement cardiovascular magnetic resonance more frequently experience syncope and/or arrhythmia [16,17]. Similarly, other measures such as myocardial strain, location and extent of myocardial fibrosis and exercise stress testing may be of interest to assist in the evaluation of the risk for VAs in patients with SRV. More recently, genetic variants were identified that were associated with the combined endpoint of heart failure, VA, and mortality in patients with D-TGA and atrial switch [40]. Moreover, the addition of genetic information significantly improved risk prediction compared to the use of clinical risk factors alone. This integrated approach of clinical and genetic data illustrates potential future avenues for risk assessment of patients with a SRV and, more broadly, CHD in general.

While numerous studies have contributed to our understanding of factors associated with arrhythmic risk, the relatively small patient populations provided insufficient statistical power to assess the independent predictive value of potentially correlated risk factors. Recently, the MAREs cohort of over 1000 patients with a SRV showed that VAs and SCD were associated with older age, history of HF, history of syncope, longer QRS duration, severe SRV dysfunction and at least moderate LVOT obstruction [2]. Based on these risk factors, a prediction model of MAREs was developed to predict the 5-year risk. Patients were categorized into low, intermediate, and high risk strata, with a 5-year predicted risk of MAREs of <5%, 5–10%, and >10%, respectively. The model performances were validated in D-TGA, while they were modest in L-TGA, with an underestimation of the risk of MAREs in the high-risk group. These results further underscore the potential differences in VA mechanisms between L-TGA and D-TGA. Moreover, external validation of this predictive model is required to confirm or challenge these findings prior to prospective validation.

Small studies suggested that programmed ventricular stimulation is of no prognostic value in risk stratifying patients with a SRV. In a small group of D-TGA patients with intra-atrial baffles (n = 6), positive ventricular stimulation studies were not predictive of future events [43]. Furthermore, in a cohort study of patients with D-TGA and ICDs, no patient with inducible sustained VT received appropriate ICD shocks in comparison to 37% of noninducible patients [10].

5. Treatment

Sustained VAs require urgent cardioversion or defibrillation, even if initially well-tolerated owing to the high risk of clinical deterioration. Antiarrhythmic drugs have not been studied in a randomized fashion for the treatment of VAs in SRV. Class I antiarrhythmic drugs are not recommended in the presence of moderate or severe systolic dysfunction of the SRV. Amiodarone can be considered a first-line antiarrhythmic agent in the presence of hypertrophy or dysfunction of the SRV but its use is limited by multi-systemic side effects. Notwithstanding contraindications such as severe renal insufficiency or a prolonged QTc interval, dofetilide may be a reasonable alternative to amiodarone in the setting of SRV dysfunction. Subject to standard precautions, sotalol may be an alternative in selected patients. Moreover, beta-blockers may decrease the risk of VA in D-TGA palliated by atrial switch surgery [10].

Certain isthmuses associated with myocardial scars have been related to monomorphic VT in patients with SRV and can be accessible for catheter ablation [[44], [45], [46], [47]]. In this setting an ICD is usually recommended for secondary prevention. In patients with VT and an ICD, concomitant catheter ablation and/or antiarrhythmic drugs may be indicated to reduce risks of recurrent arrhythmias and ICD shocks. Polymorphic VT and ventricular fibrillation are less amenable to ablation, often reflect diffuse fibrosis and myocardial disarray, appear to be associated with failing a SRV, and incur the highest risk [2].

Prevention strategies (other than ICD implantation) should consider aggressive management of atrial arrhythmias, but also limiting exercise prescription in D-TGA and atrial switch due to the possible risks of ischemia-related VAs. Optimal screening and management of cardiovascular risk factors should also be integrated into a global approach that considers all potential co-existing conditions.

6. Implantable cardioverter-defibrillators

According to international guidelines, implantation of an ICD may be considered in patients with low right ventricular ejection fraction and additional risk factors which include complex VAs, unexplained syncope, NYHA Class II or III, QRS duration >140 ms, or severe systemic AV valve regurgitation [[48], [49], [50], [51]]. However, the strength of recommendations for primary prevention ICDs in this patient population is low (e.g., Class IIB) with weak supportive evidence [6]. As a result, ICD indications remain controversial for this group of patients. In general, ICDs appear to be under-implanted in patients who are truly at risk for SCD, and those who have received primary prevention ICDs experience a low rate of subsequent VAs [10,52]. These trends were observed in the MAREs registry, where 59% of patients implanted with ICDs for primary prevention were in the low risk category (5-year MARE risk,<5%) [2]. The MAREs risk prediction model may potentially help refine the selection of candidates with a SRV for primary prevention ICD implantation.

Cardiac resynchronization therapy (CRT) or conduction system pacing should also be considered in patients with ICD indications who have concomitant poor SRV systolic function. Unfortunately, pacing a SRV cannot be achieved via the coronary sinus in patients with D-TGA and Mustard or Senning repair since it courses along the atrioventricular groove of the subpulmonary LV. Thus, an epicardial lead placed on the SRV is required to achieve CRT in this subgroup of patients. In contrast, the coronary sinus courses along the atrioventricular groove of the SRV in patients with L-TGA such that CRT may be achieved by this endocardial approach if the coronary sinus ostium is not atretic. Moreover, conduction system pacing has evolved into a well-established technique in L-TGA and can be considered as an alternative to CRT in selected patients. Indications for physiological pacing remain to be refined in the absence of definitive data that establish a defined cut-off value or SRV dysfunction as quantified by ejection fraction. In the MAREs study, severe echocardiography SRV dysfunction was strongly associated with risk of VA events. Assessment of right ventricular function was performed using subjective visual estimations of SRV function which allowed the discrimination of patients with CMR SRV ejection fraction <35% from those with SRV ejection fraction ≥35% [35]. However, the ideal cut-off value for CMR SRV ejection fraction indicative of risk for VAs merits further study.

The risk of inappropriate shocks in CHD patients with primary prevention ICDs is substantial and must be considered in weighing risks and benefits. Inappropriate shocks are predominantly due to misinterpretation of sinus tachycardia and supraventricular arrhythmias, T-wave oversensing, and lead failure and occur at a rate of 6.6%/year in ICD recipients with a SRV [10]. Decreased lead longevity and performance must also be taken into account in this patient population. Finally, vascular access may be challenging in SRV and must take into account the need for multiple lead and device changes over the lifetime of a patient. Moreover, epicardial access may be challenging due to scarring or fibrous tissue formation after surgery [53]. Venous access may be complicated by obstructions or congenitally absent or anomalous veins or may be contra-indicated in the presence of an intracardiac shunt such as a baffle leak. In D-TGA with atrial switch, the superior baffle must be patent to be traversed, with care to avoid phrenic capture on the lateral systemic ventricular wall and native left atrial appendage. The subcutaneous ICD (S-ICD) may be an option in patients with a SRV who do not require permanent pacing but is limited by the high prevalence of sinus node dysfunction and AV conduction system disease in patients with D-TGA and L-TGA, respectively.

7. Conclusions

VAs, be they primary or secondary, remain a major cause of death in the aging population of patients with a SRV. Structural and functional alterations as well as neurohumoral remodeling related to myocardial scars and progressive SRV failure contribute to ventricular arrhythmic substrates. Future guidelines should consider newer evidence regarding the increased incidence of VAs in older patients with significant pathology of the SRV and subpulmonary LV on cardiac imaging. Recently developed predictive models may be useful in improving the identification of patients at risk of clinical VA who could benefit from ICD implantation for primary prevention. However, risk stratification could potentially be further refined by future studies that provide detailed analyses of cardiac imaging with CMR, biomarkers, and genetic variants. Clinical VAs in patients with SRV are challenging to treat and may involve life style modifications, medical therapy and ICD implantation with or without physiologic pacing. Drawbacks related to ICDs are considerable, with a high rate of inappropriate shocks and late lead-related complications, and should always be weighed against benefits in shared clinical decision-making.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.