The Ventricular Switch: An Alternative Strategy Toward Biventricular Repair in Moderately Hypoplastic Left Ventricles or Complex Connections

Introduction

Hypoplastic left heart syndrome constitutes 1% to 2% of all congenital heart defects and is associated with high rates of mortality. ¹ This pathology presents as a spectrum of cardiac malformations, which brings additional challenges to clinical decision-making. This is particularly true in the case of milder forms of left ventricular hypoplasia and valvar disease, which we term hypoplastic left heart complex (HLHC). Complex congenital heart anomalies with hypoplastic ventricles are often limited in their ability to support a biventricular circulation due to anatomy and physiology constraints. Traditionally, management of these lesions, including HLHC, has involved single ventricle palliation (SVP). SVP converts the native anatomy into a univentricular circulation, ultimately culminating in a Fontan palliation. In the long run, however, the Fontan circulation may be associated with substantial morbidity and mortality owing to its deranged physiology, particularly in subtypes such as those associated with heterotaxy syndromes and in hearts with common atrioventricular valve. Alternative management strategies have been employed to rescue hypoplastic ventricles to achieve biventricular circulation. Ventricular hypoplasia is a continuum, but the decision to proceed to either a single or biventricular circulation is dichotomous. When in doubt, surgeons move the patient to a single ventricle pathway, leaving behind a left ventricle, even if it is moderately hypoplastic, for the fear of a failed biventricular repair resulting in mortality. We propose to harness this moderately hypoplastic left ventricle to support the pulmonary circulation, hence achieving a unique type of biventricular repair (or in some instances a one-and-a-half ventricular repair), by undertaking what we term the “ventricular switch” procedure. ² In this novel technique, the morphologic left ventricle is utilized as the subpulmonic ventricle, and the morphologic right ventricle is used as the systemic ventricle (Figure 1). Herein, we describe our rationale for considering biventricular circulation in HLHC and in other complex lesions that prevent the LV from becoming the systemic ventricle.

OPEN IN VIEWER

Figure 1.

Diagram of biventricular repair with a ventricular switch in patient with right dominant unbalanced complete atrioventricular septal defect/double outlet right ventricle; includes an atrial conversion, re-routing systemic venous return to the left atrium (including translocation of right SVC to left SVC), routing pulmonary veins to right atrium, septation of the AVSD, and an LV-PA conduit. A prophylactic right-sided annuloplasty can be added to limit dilatation. (Adapted from Najm et al,² with permission). Abbreviations: SVC, superior vena cava; AVSD, atrioventricular septal defect; LV-PA, left ventricle-to-pulmonary artery. (See full color image in online version).

The Unavoidable Fontan Physiology

Patients with underdeveloped, hypoplastic ventricles managed with a univentricular circulation often develop consequences due to the physiologic effects of lacking a subpulmonic ventricle. ³ The Fontan circulation is characterized by chronically elevated systemic venous pressures, potentially leading to pulmonary hypertension and decreased cardiac output. These changes from normal physiology increase exposure to morbidity and mortality from complications such as protein-losing enteropathy, plastic bronchitis, and thromboembolism, to name a few. A large review of more than 1000 patients with Fontan circulation demonstrated sobering survival statistics, with a 30-year survival of patients who survive initial hospitalization is only 48%, which does not include the pre-Fontan attrition of 49% to 54% found in HLHS patients. ⁴ Other studies suggest that only 41% of Fontan patients are free from serious adverse events at 40 years of life, with reintervention rates of 22%. ³ Perhaps more important, the onset of Fontan failure complications accelerates morbidity and mortality. A diagnosis of protein-losing enteropathy post-Fontan demonstrates a 5-year survival of 40% at 5 years, and a 20-year survival of 19%. ⁴ Similarly, development of cirrhosis after a Fontan is associated with high mortality, with survival found to be 57% and 35% in 1 year and 5 years, respectively.^4,5 Many experienced institutions have demonstrated that not all Fontan patients are equal. Certain lesions have been found to portend poorer outcomes in Fontan circulation, such as complete atrioventricular septal defects (AVSD) and heterotaxia. ⁶ Moreover, a recent study demonstrates that HLHS was a predictor of Fontan failure, with a higher mortality rate, when compared with other cardiac morphologies. ⁷

Additional challenges are presented by congenital lesions with acceptable ventricular volumes but unsuitable for biventricular repair due to anatomy. Some argue that a 2-ventricle Fontan may provide superior results compared to a 1-ventricle Fontan, based on the premise that 2 ventricles working as one can lower end-diastolic pressures, thus decreasing complications. However, a 2-ventricle Fontan does not change the preload deficiency seen in the Fontan circulation. A recent study comparing patients with a 2-ventricle versus a 1-ventricle Fontan demonstrated no difference in survival between the 2 groups, calling into question the wisdom of resorting to a single ventricle strategy among patients with 2 working ventricles. ⁸ In the face of these sobering SVP results, institutions must consider whether a more optimal outcome can be achieved with SVP or a higher-risk biventricular repair.

The Concept of Ventricular Switch

In light of the modest outcomes of the Fontan procedure, particularly in Shone’s complex and HLHC, many specialized centers have applied various strategies to maximize the chance of achieving a biventricular repair, even after initial single ventricle palliation. The Boston group published a series describing strategies for recruitment of moderately hypoplastic left ventricles and has shown some success.^9,10 This approach could be limited and at a high cost. Failure of this approach may result in suboptimal early or late LV performance with restrictive physiology and may impair survival. Reverting back to SVP may not be easily obtained. In other cases when both ventricles are fully formed but unable to route blood to the aorta, such as in heterotaxy with L-malposed aorta connected to the morphologic right ventricle, surgeons tend to palliate these patients to an SVP leaving behind a fully formed LV.

The ventricular switch procedure is a concept that includes multiple staged or nonstaged procedures. It culminates in the right ventricle becoming the systemic ventricle and the left ventricle functioning as the subpulmonary ventricle, allowing a 2-ventricle circulation in patients previously constrained by the HLHC physiology or the anatomy. ² A key tenet in this strategy is that the left ventricle can be beneficially harnessed, even if moderately hypoplastic, for use in the pulmonary circulation when it cannot be committed to the aorta.

Components of a Ventricular Switch

Depending on the anatomy, the ultimate goal is a systemic right ventricle. First, the evaluation process must exclude the feasibility of the left ventricle being used as the systemic ventricle, as this would be the preferred track. If the left ventricle is not suitable to become a systemic ventricle due to size or anatomy, then the venous connections should be evaluated to connect the systemic venous return to the left ventricle and the pulmonary venous return to the right ventricle. In the classic setting of moderately hypoplastic LV not suitable for biventricular repair, the needed procedure would be a hemi-Mustard, combined with an arterial switch and a bidirectional superior cavopulmonary shunt. In other cases, with L-malposed aorta coming out of the RV, the ventricular switch may necessitate an atrial conversion. We have named it atrial conversion, as these procedures are performed differently than previous published atrial switch techniques. These atrial conversions could include a hemi-atrial conversion where only the IVC is routed to the moderately hypoplastic left sided atrioventricular valve and left ventricle. This can be accomplished with an adjunctive bidirectional superior cavopulmonary shunt and can be part of AVSD septation.

To simplify the conversion in the presence of bilateral superior vena cavae (SVC), we translocate the contralateral SVC to the atrial situs to the ipsilateral SVC. The contralateral SVC is usually longer and will reach the other SVC, however, this can be further facilitated by division of the azygous/hemi azygous veins and the mammary veins on both sides. These maneuvers will ensure tension-free anastomosis. To complete the procedure, typically a left ventricle-to-pulmonary artery (LV-PA) conduit will have to be constructed. In recent cases, we have added a prophylactic right-sided atrioventricular valve annuloplasty to limit the dilatation of that valve after being subjected to systemic pressure.

What Is the Performance of the Right Ventricle as a Systemic Pump?

To assess the feasibility of a ventricular switch, we must understand the long-term outcomes of a systemic right ventricle in other disease processes. A study by the Congenital Heart Surgeons’ Society demonstrated similar 30-year survival among patients with D-transposition of the great vessels (D-TGA) who underwent either an atrial switch or an arterial switch operation. ¹¹ That is, patients with a systemic right ventricle can have similar long-term outcomes to those with a systemic left ventricle, albeit with increased time-related risk of RV failure, arrhythmias, and other complications. Another example of a systemic right ventricle is the physiological repair of congenitally corrected transposition of the great vessels (cc-TGA). A review of outcomes in cc-TGA patients demonstrates an acceptable overall long-term survival after physiologic repair (survival = 88% at 25 years, Figure 2). ¹² Although survival is not as optimal as an anatomic repair, physiologic biventricular repair does come with acceptable outcomes and certainly better than SVP. This is compelling data that a right ventricle can sustain adequate systemic cardiac output long-term. Thus, we argue that a conversion to cc-TGA physiology is preferable to a Fontan physiology. When the right ventricle fails in later years, the manifestations are limited to heart failure and not multi-organ failure seen in SVP. These cases of failing systemic right ventricle are amenable to assist devices and heart transplantation similar to other failing ventricles, which is yet another advantage of the ventricular switch.

OPEN IN VIEWER

Figure 2.

Long-term outcomes of cc-TGA. (A) Actuarial survival (Kaplan-Meier) of 167 patients with cc-TGA after biventricular repair; (B) Late survival (excluding hospital operative mortality) among 123 patients with physiologic repair (b) compared to 44 patients with anatomic repair (a). (Adapted from Lim et al, ¹² with permission).

The Ventricular Switch in the Context of HLHC and Complex Anomalies

A key principle in the ventricular switch operation is the understanding that the right ventricle can indeed be used for systemic circulation, at least until adulthood. However, patient selection is critical for acceptable outcomes. Lesions that are appropriately allocated are lesions where the left ventricle cannot be routed to the aorta. Such lesions described are variants of cc-TGA, heterotaxy with anomalous systemic or pulmonary venous connections, or double outlet right ventricle with remote ventricular septal defect, to name a few. These lesions continue to have a near normal-sized right and left ventricle. But can the ventricular switch procedure be appropriately applied to a HLHC? Although evidence is sparse due to the novelty of the technique, we suggest that it can. In the case of borderline HLHC, the required procedure would be a bidirectional cavopulmonary shunt, an arterial switch, and a hemi-Mustard procedure.

It is important to recognize these cases at birth and begin planning, based on the ultimate ventricular dominance. Once the ventricular switch concept is adopted, it would be clear to the surgeon the steps needed to ultimately make the RV into the systemic ventricle. Staging is the appropriate route. The different surgical procedures should be performed at the age and weight that is customary for other standard congenital lesions. With limited available data, we believe it is a multidisciplinary team decision to proceed with LV recruitment versus a ventricular switch or single ventricle palliation in the case of HLHC. While in other cases, such as those with L-malposed anterior aorta arising from RV, the ventricular switch would be the better choice over single ventricle palliation.

We performed a ventricular switch in a series of 5 patients with complex congenital lesions that had initially been palliated into a univentricular circulation due to the challenge and high-risk nature of a biventricular repair. ² One example is the series that highlights the proof-of-concept of a ventricular switch in a patient with a hypoplastic ventricle is an 11-month-old female born with a right dominant unbalanced complete atrioventricular septal defect/double outlet right ventricle (AVSD/DORV). She was initially palliated with a right ventricle-to-pulmonary artery conduit. The left ventricle was moderately hypoplastic (LVEDVi = 25cc/m²). To prepare her for a biventricular conversion, she underwent a 1.5 ventricle repair to recruit the left ventricle. Then, she subsequently underwent a ventricular switch 6 months later that included a hemi-Mustard, bidirectional cavopulmonary shunt and common AVSD septation, and LV-PA conduit. The post-conversion echocardiogram demonstrated good structural and biventricular function.

Functionalizing a hypoplastic left ventricle can be achieved using a staged approach, which allows for the growth of the ventricle and mitigates the high-risk nature of a single procedure. Various studies describe success when recruitment procedures are performed in conjunction with SVP. Kwak et al demonstrate clinically significant growth in diastolic volume and stroke volume after ventricular recruitment. ¹⁰ In other words, hypoplastic left ventricles can be recruited, eventually to provide meaningful cardiac output. Furthermore, the moderately hypoplastic left ventricle may be better suited as a subpulmonary pump in case of high pulmonary vascular resistance that would not be suitable for SVP. This approach will only be successful if the possibility of biventricular circulation is contemplated at birth and constantly revisited.

The long-term outcome of the ventricular switch procedure has yet to be determined given the novelty of the technique. Specifically, the atrial conversion component may give surgeons some reluctance due to reported long-term issues with baffle obstruction and atrial arrhythmias.^13–15 Associated RV failure in patients with systemic RV circulation is another source of consternation.^14–16 These concerns are valid and may present as limitations as more follow-up data are collected on ventricular switch patients. However, the true advantage of the ventricular switch procedure is the ability to maintain 2-ventricle circulation during growth and development to a later stage in life where more options may be available.

Options After Ventricular Switch

The anticipated benefit of a ventricular switch procedure is that a functionally univentricular circulation is changed into a cc-TGA phenotype, which may translate into a more favorable long-term prognosis. More importantly, this pathway avoids the sequelae of multi-organ failure in a failing Fontan circulation or the high risk associated with a complex biventricular repair, at the risk of single-organ heart failure in adulthood. The current climate of pediatric heart failure presents limited options due to the shortage of transplantable hearts and issues related to pediatric mechanical assist devices. Outcomes of listings for cardiac transplantation in failing Fontan patients demonstrate that pretransplantation survival is 74% at 12 months. ¹⁷ Post-transplant, studies demonstrate that operative mortality is worse in Fontan/Glenn patients (44%), and that post-transplant survival is diminished in a failing Fontan when compared to non-Fontan patients (1-year survival 77% vs. 90%), though outcomes are improving in the recent era. ¹⁸ Additionally, outcomes with ventricular assist device (VAD) in congenital heart disease demonstrate that the presence of end-organ dysfunction portends poorer outcomes. ¹⁹ Thus, the true advantage of a ventricular switch is a patient progression into adulthood when treatment strategies are more expansive. Contemporary literature suggests that LVAD and transplants have success when applied to patients with adult congenital heart disease (ACHD).^20,21 ACHD patients treated with LVADs have similar survival compared to non-ACHD patients with LVADs. ²¹ Survival after transplant for ACHD patients are also comparable to appropriately matched non-ACHD patients. ¹⁷ By performing a ventricular switch and recapitulating a ccTGA phenotype, the patient will have more advanced options available down the road.

Conclusion

HLHC and some anomalies with complex connections (e.g., heterotaxy) pose a clinical conundrum with the ongoing debate on whether univentricular palliation or a complex biventricular repair is optimal. A ventricular switch procedure is another tool in the armamentarium, especially for those circumstances where a traditional anatomic repair is deemed to be high risk. We feel that this management strategy is preferable to a Fontan physiology and may provide a more favorable long-term prognosis and better options in the future.