Congenitally corrected transposition of the great arteries (ccTGA) has attracted a great deal of interest and some controversy over the years. The malformation is a complex one, featuring reversed (discordant) atrioventricular (AV) and ventriculo‐arterial (VA) connections. Alternate terms to describe ccTGA include corrected transposition, discordant transposition, AV and VA discordance, physiologically corrected transposition, double discordance, AV discordance with transposition, L‐transposition (L‐TGA), inverted transposition, and others. Standard segmental nomenclature has also been applied. A full review of the nomenclature and a proposed classification have been adopted for the Society of Thoracic Surgeons (STS) Database for Congenital Heart Surgery, and ccTGA has become a universally understood term, even if “corrected” is an overstatement [1]. Single or unbalanced ventricular variants with an L‐loop embryology but without a discordant AV connection are excluded from this discussion. Although the nomenclature is now well accepted, the relative rarity (approximately 1/33,000 live births, or approximately 0.05% of congenital heart malformations) and complexity of ccTGA have impeded our ability to establish a universally acceptable treatment strategy.

The morphologic features common to hearts with ccTGA are as follows:

• The systemic venous atrium (morphologic right atrium) is connected to the morphologic left ventricle (LV) through a mitral valve.
  
• The LV is connected to the pulmonary artery (PA) through the pulmonary valve.
  
• The pulmonary venous atrium (morphologic left atrium) is connected to the morphologic right ventricle (RV) through a tricuspid valve (TV, which functions as the systemic AV valve).
  
• The RV is connected to the aorta via the aortic valve (Figure 26.1).

As a result of discordant connections at both levels, blood flows in a “congenitally corrected” pathway, despite some rather extreme anatomic derangements (Figure 26.2). The “correction” is physiologic only, in the sense that in biventricular hearts with ccTGA a normal oxyhemoglobin saturation may be present (in cases without intracardiac shunts and LV outflow tract obstruction). From a segmental viewpoint, the most common configuration is atrial situs solitus with left ventricular loop and left TGA {S,L,L}. Less common is ccTGA with atrial situs inversus {I,D,D}. Discordance between the atrial situs and the position of the cardiac apex (i.e., ventricular apex points toward caval axis) is frequent in ccTGA (as compared to concordant TGA).

Additional anomalies, albeit of varying severity, are present in 99% of cases of ccTGA (Figure 26.3). The three most commonly associated major anomalies in ccTGA are ventricular septal defect (VSD), pulmonary valve and/or subpulmonary stenosis (resulting in LV outflow tract obstruction), and TV malformations (resulting in systemic AV valve insufficiency). Although ccTGA has been reported to occur with AV septal defect, a common AV valve creates some ambiguity in applying a strict anatomic definition of AV discordance [2]. Associated abnormalities in ccTGA may affect the timing and mode of presentation, the severity of symptoms, the clinical evolution, and, ultimately, the approach to management.

In ccTGA there is an anterior–superior location of the AV conduction axis, which is vulnerable to spontaneous or surgically induced heart block (Figure 26.4). Rarely, AV conduction tissue may reside in a more normal postero‐inferior position in ccTGA hearts with LV outflow tract obstruction [2, 3].

The LV outflow tract in ccTGA is deeply wedged between the TV and mitral valve (MV), creating a variable tendency to dynamic or fixed LV outflow tract obstruction in 30–50% of cases [4]. The presence of LV outflow tract obstruction exerts a major influence on the choice of surgical strategy.

Coronary anomalies are also common in ccTGA, and may likewise complicate the surgical approach. In general, the coronary arteries arise from facing sinuses in a mirror‐image arrangement (1R, 2LCx by the Leiden classification) [5]. In an anatomic specimen study from Children’s Hospital of Philadelphia, 45% of ccTGA hearts had coronary artery abnormalities that might add complexity to anatomic surgical repairs [6, 7]. In the initial Melbourne double‐switch experience, although 12/14 patients had a 1R, 2LCx pattern, 2/14 had a single coronary artery from a nonfacing sinus (a pattern not seen in the first 400 cases undergoing arterial switch operation for concordant TGA) (Figure 26.5) [6, 7].

The etiology of ccTGA is unknown, and there is no phylogenetic or ontogenetic precedent, as in some other forms of congenital heart disease. There is probably a link with abnormalities of visceral situs. Transposition of the great arteries (TGA), ccTGA, and heterotaxy can all be induced in a murine model by treatment with retinoic acid or inhibitors thereof [8]. It would appear that ccTGA is more likely to be a laterality defect rather than a conotruncal defect per se.

Important Historical Aspects

Although primacy is always difficult to prove in a discipline of active minds, von Rokitansky (1875) was almost certainly the first to describe the pathologic entity now known as ccTGA [9]. Monckeberg (1913) noted the important anterior AV node position, and Walmsley (1931) described the mirror‐image coronary arterial branching pattern and the position of the central fibrous body [10, 11]. Schiebler and associates (1961) promoted the “ccTGA” terminology [12]. Anderson and Lillehei (1957) performed and documented the first (physiologic) surgical repairs [13]. Imai and colleagues (1989), and Mee and coworkers (1989), independently described various anatomic repair strategies, including what is now known as the double‐switch operation [14, 15]. Ilbawi (1990) described the Rastelli + atrial baffle approach to ccTGA with VSD and LV outflow tract obstruction [16]. The 1.5‐ventricle approach using a bidirectional cavopulmonary shunt (BCPS) (and subsequently incorporated into both physiologic and anatomic repairs) was first proposed by Mavroudis and Backer (1999) [17]. Aortic root and pulmonary root translocation in ccTGA have been described independently by multiple groups over the past two decades, based in part on the principles of the Nikaidoh operation for concordant TGA patients [18–21].

Diagnosis

The diagnosis of ccTGA and most anatomic details can usually be established accurately using clinical assessment and two‐dimensional echocardiography (Figure 26.6). Cardiac catheterization may provide important information about the hemodynamics, and further delineate anatomic relationships and PA morphology. Angiographic views should define the site and number of VSDs, the morphology of the ventricular chambers, the nature of the LV outflow tract obstruction, and all associated anomalies. The role of more contemporary imaging strategies such as computed tomography angiography (CTA), cardiac magnetic resonance imaging (cMRI), and three‐dimensional modeling in the diagnosis of ccTGA are extremely valuable for preoperative planning and preferences vary from center to center.

Natural History and Clinical Presentation

Historical precedent suggests that unoperated cases of ccTGA have a 32% prevalence of heart failure and a 25% mortality by the fourth decade of life (Figure 26.7). However, in practice the natural history of ccTGA is extremely variable. Infants with severe pulmonary stenosis or pulmonary atresia and those with severe left AV valve insufficiency may present early in life. At the other extreme, some individuals (usually the 1–2% with no associated defects) are asymptomatic for many years and are diagnosed only late in life or at postmortem examination. Reports of untreated individuals with ccTGA surviving into the ninth and tenth decades prove this concept [22, 23]. Likewise, successful pregnancies (without long‐term sequelae) have been reported in women with ccTGA [24]. Signs and symptoms in ccTGA (when present) may be related to abnormal pulmonary blood flow, systemic AV valve insufficiency, bradycardia from heart block, or systemic (right) ventricular dysfunction. The RV function can begin to deteriorate early after certain surgical repairs, or spontaneously (most often during the second decade of life). In general, for both definitively operated and palliated patients, survival is significantly poorer than that of the normal population from early infancy onward.

LV outflow tract obstruction in ccTGA can cause cyanosis in the newborn period when an intracardiac shunt is present. Some infants may require prostaglandin infusion and a systemic–pulmonary arterial shunt. Children with severe tricuspid insufficiency and/or a large VSD with minimal or no LV outflow tract obstruction may present in the first few weeks of life or any time thereafter with congestive heart failure. The most common presentations during childhood or adolescence are exercise intolerance or poor growth due to a left‐to‐right shunt, or increasing cyanosis and exercise intolerance due to pulmonary stenosis. Progressive systemic AV valve insufficiency may complicate these presentations, or occur as an isolated finding. Some patients present with spontaneous heart block and bradycardia. The proportion with block increases by 2% per year to reach 30% in adult life [25].

Tricuspid insufficiency is a central problem in ccTGA, in both operated and nonoperated patients. The mechanism is probably multifactorial, and not completely understood. Suggested mechanisms include the following:

• Congenital leaflet abnormalities (similar to Ebstein anomaly), including short chords and apical displacement of septal and mural leaflets.
  
• RV or papillary muscle dysfunction.
  
• Dilated TV annulus.

  

  

  
  
• Distraction of papillary muscles (RV/LV pressure‐related septal shift).
  
• TV leaflet/chordae adherence to a VSD patch.
  
• Synchronous contraction (bundle branch, complete heart block), supraventricular or ventricular arrhythmia.

The features of Ebstein anomaly as seen in concordant hearts (such as thinning of the atrialized portion of the RV and severe tethering of leaflets) are not generally present in ccTGA. Furthermore, the typical surgical treatments of Ebstein anomaly in AV concordant hearts (e.g., cone repair) may not be applicable in ccTGA. Hence, the term “Ebsteinoid” is most often applied to the TV in ccTGA, acknowledging that the morphology is significantly different. In some cases of ccTGA, RV hypertrophy eventually gives way to dilation, so the TV papillary muscles are distracted. The short tethered TV chords and displaced leaflets fail to coapt, and tricuspid insufficiency becomes progressive and self‐perpetuating. Tricuspid insufficiency can also be exacerbated by surgical relief of LV outflow tract obstruction, due to unfavorable septal shift that occurs when the LV pressure drops. PA band placement may have the opposite effect and move the septum to a more favorable midline location, improving TV function.

Surgical Indications and Options

Repair of ccTGA was first attempted during the late 1950s [13]. A summary of past and currently used surgical strategies is presented in Figure 26.8. Palliative or interim procedures may be useful and advisable in selected circumstances. Both physiologic and anatomic repairs have been employed for definitive treatment (Figure 26.9). Because the potentially unfavorable natural history of children with ccTGA and a systemic RV has been well documented, many surgical teams have moved toward “anatomic” repair (i.e., placing the RV in the pulmonary circuit and the LV in the systemic circuit), instead of the “physiologic” approach (leaving the RV in the systemic circuit and the LV in the pulmonary circuit). The latter strategy fails to address AV discordance and its eventual detrimental consequences, despite the correction of associated anatomic aberrations. This of course may have an important influence on the midterm and late outcome despite a good immediate result. Univentricular and 1.5‐ventricular repairs can also be employed selectively [26].

Interim (Palliative or Nondefinitive) Operations for ccTGA

The principal palliative operations used in ccTGA are systemic‐to‐pulmonary artery modified Blalock–Taussig–Thomas (BTT) shunts and PA banding. Isolated repair or replacement of the TV is also appropriate in select circumstances. Pacemaker implantation and cardiac resynchronization may be of benefit in patients with complete heart block and RV dysfunction [25].

The modified BTT shunt may be a useful strategy for ccTGA patients in the following circumstances:

• Infants and neonates with VSD and severe LV outflow tract obstruction or pulmonary atresia (who may be expected to undergo physiologic or anatomic repair with an extracardiac conduit at a later date).
  
• Infants with ccTGA, VSD, and LV outflow tract obstruction who are not candidates for biventricular repair due to unbalanced ventricles, straddling AV valves, or multiple unfavorable VSDs. These patients may be destined for a univentricular strategy (Fontan sequence) [26, 27].

We prefer a transsternal approach for all modified BTT shunts, usually employing a 4 mm polytetrafluoroethylene (PTFE) tube graft interposition between the brachiocephalic artery and right PA. With this approach, subsequent takedown is facilitated, and PA distortion can be minimized.

The majority of children with ccTGA (especially those with apicocaval ipsilaterality, i.e., cardiac apex and inferior caval vein [IVC] on the same side) do have some natural limitation of pulmonary blood flow, so PA band placement specifically for control of pulmonary arterial blood flow is not frequently required [28]. However, other considerations may apply. A PA band may be indicated under the following circumstances:

• Children with pulmonary hypertension (due to a large VSD), in whom a double‐switch operation (arterial switch operation + Senning) is planned after early infancy. However, primary anatomic repair may be an option in this situation, at least for very favorable candidates.
  
• Unoperated children with ccTGA and intact ventricular septum who develop tricuspid insufficiency and RV failure, but have maintained a low LV pressure. In such cases physiologic repair options are limited, and a PA band is employed to effect LV conditioning in preparation for anatomic repair. Yacoub and associates first employed LV conditioning in TGA [29]. Accumulated PA banding experience of Mee and others in the 1990s for TGA patients with late failure of atrial repairs led to a more widespread application for LV conditioning in unoperated patients with ccTGA [30]. The probability of success with the PA banding strategy is probably inversely related to age, and early infancy may be the most reliable time to induce hypertrophy. [30, 31]. More recently, a loose PA band in combination with creation of an atrial septal defect (ASD; atrial septectomy) has been proposed as a more efficient and reliable way to condition the subpulmonary LV, through a combination of augmented preload and afterload [32].
  
• Selected patients who develop RV failure following a physiologic repair and are candidates for conversion to an anatomic repair using a double‐switch strategy.
  
• “Destination therapy” or “open‐ended palliation” [33]. This strategy is based on the observation that restoration of a more midline septal position in ccTGA may limit or improve tricuspid insufficiency, to the point of eliminating the need for further treatment in some patients. Early in the LV conditioning experience for failed atrial repairs in TGA, it was noted by Mee and colleagues that some patients had improvement in TV function after PA banding alone (i.e., before switch conversion) [34]. It has been logically suggested that an early partially adjustable PA band (tightened to induce LV/RV = 50%) could maintain LV mass and stabilize TV function in neonates and young children with ccTGA and intact ventricular septum [33–36]. The resulting physiology mimics ccTGA with LV outflow tract obstruction, in which TV function tends to be better [35, 37–40]. A double switch can be performed if and when needed, hence the open‐ended nature of the palliation. The efficacy may be enhanced when an adjustable or dilatable PA band is employed. It has even been suggested that the outcome is as effective as that of anatomic repairs [40]. The limits and durability of this interesting approach remain in question.

Definitive Operations for ccTGA

The best surgical strategy for the various forms of ccTGA is debatable, and surgeon‐ and team‐dependent factors are important. Problems in decision‐making are magnified by the rarity of the lesion. All operations are theoretically possible from the neonatal period onward, but not necessarily advisable. Repair has been performed as late as the fifth decade of life [41, 42]. Features favoring the physiologic approach include competence of both AV valves, balanced ventricular size, and excellent RV function. Features favoring the anatomic approach include disparity in ventricular size, poor RV function, tricuspid insufficiency, and a nearly systemic LV pressure (LV outflow tract obstruction, secondary pulmonary hypertension, or following PA banding).

In general, anatomic (as opposed to physiologic) correction requires a more complex surgical approach, but may have a better long‐term outcome. Complexity scores and anticipated mortality scores have been determined for various congenital heart operations, including the Aristotle Complexity Score and the Society of Thoracic Surgeons–European Association for Cardio‐Thoracic Surgery Congenital Heart Surgery Mortality Score (STAT) [27, 43, 44]. The double‐switch operation, for example, has a STAT mortality category of 5 (highest expected level of mortality) and an Aristotle Complexity Score of 13.8, rising to 18.8 in a patient with prior transsternal PA banding for LV conditioning. In comparison, STAT mortality categories for a physiologic correction involving isolated VSD closure in ccTGA or VSD closure with LV–PA conduit placement are 3 and 4, respectively, suggesting a significant difference in early mortality risk [42–44].

Although the late results of physiologic repair are well documented, the late results of anatomic repair are still under assessment, and it is fair to say that the long‐term relative benefit for patients who are suitable for either type of repair remains theoretical. Current thought suggests that an anatomic approach is a better prospect for patients who already have problems with the RV and/or TV, and who are therefore poor candidates for physiologic repair. For example, relief of LV outflow tract obstruction in the context of a physiologic repair has the theoretical disadvantage of increasing the tricuspid insufficiency over time by allowing a ventricular septal shift in an unfavorable direction, with the expected compromise in outcome [4, 43, 45].

The cavopulmonary connection (Fontan) strategy is useful for many complex anatomic situations in which the risk of septation outweighs the benefits. BCPS may be selectively applied for anatomic, physiologic, and univentricular approaches (see below). Children with ccTGA who are not candidates for biventricular repair (i.e., those with important straddling of an AV valve, unbalanced ventricular size, and multiple unfavorable VSDs) can be palliated with a BCPS followed by Fontan. We have also employed an interventional strategy for the Fontan operation after appropriate preparation at the BCPS stage [46]. For an extracardiac Fontan operation (all pathologies considered) the STAT score is 2, and biventricular Fontan candidates may be superior candidates (as compared to those with complex univentricular anatomy) [4]. The Fontan operation may, in fact, impart an event‐free survival advantage over physiologic and anatomic biventricular repairs, with similar or better exercise capacity and quality of life, even for patients who, theoretically, could undergo biventricular repair [37, 47, 48

Stay updated, free articles. Join our Telegram channel

Join