D-transposition of the great arteries (D-TGA): Management and outcome

INTRODUCTION — Transposition of the great arteries (TGA) is a ventriculoarterial discordant lesion in which the aorta arises from the right ventricle and the pulmonary artery from the left ventricle. In the dextro- or D-looped type (D-TGA), the ventricles are oriented so that the right ventricle is positioned to the right of the left ventricle and the origin of the aorta is anterior and rightward to the origin of the pulmonary artery (figure 1). D-TGA is a form of cyanotic heart disease because there are two parallel circulations.

The management and outcome of D-TGA will be presented here. The anatomy, physiology, clinical features, and diagnosis of D-TGA are discussed separately. (See "D-transposition of the great arteries (D-TGA): Anatomy, clinical features, and diagnosis".)

Levo- or L-looped TGA (L-TGA; also commonly referred to as congenitally corrected TGA) is discussed separately. (See "L-transposition of the great arteries (L-TGA): Anatomy, clinical features, and diagnosis" and "L-transposition of the great arteries (L-TGA): Management and outcome".)

PRENATAL DIAGNOSIS AND MANAGEMENT — Prenatal diagnosis of D-TGA improves survival as it allows for delivery at a tertiary center where pediatric cardiologists can make timely postnatal decisions regarding prostaglandin (alprostadil) infusion and the need for a balloon atrial septostomy (BAS) [1]. However, D-TGA is one of the more difficult congenital heart diseases to detect prenatally and, therefore, many infants present postnatally. (See "D-transposition of the great arteries (D-TGA): Anatomy, clinical features, and diagnosis", section on 'Fetal diagnosis'.)

When the diagnosis of D-TGA is made prenatally, delivery should occur at a facility with a level III neonatal intensive care unit and pediatric cardiology expertise. If this is not feasible, transport arrangements should be established in advance of the delivery. Additional details regarding fetal diagnosis and pregnancy management are provided separately. (See "Congenital heart disease: Prenatal screening, diagnosis, and management".)

INITIAL POSTNATAL MANAGEMENT — Neonates with suspected or confirmed D-TGA should be transported as quickly as possible to an institution that has expertise in managing infants with D-TGA. Prior to transport, consultation with a pediatric cardiologist is necessary to ensure optimal medical management before and during the transfer.

Initial management focuses on stabilization of cardiac and pulmonary function and ensuring adequate systemic oxygenation. Therapy is directed towards providing sufficient mixing between the two parallel circulations by maintaining patency of the ductus arteriosus with prostaglandin E1 (PGE1; alprostadil) infusion and/or balloon atrial septostomy (BAS).

Once the infant is stabilized, corrective surgery is optimally performed in the first weeks of life.

Prostaglandin (alprostadil) infusion — When the diagnosis of D-TGA is strongly suspected, a pediatric cardiologist should be consulted and a continuous intravenous PGE1 (alprostadil) infusion (0.05 mcg/kg per minute) should be started to maintain patency of the ductus arteriosus. A patent ductus arteriosus promotes intercirculatory mixing.

The major side effect of PGE1 is apnea, and intubation should be considered prior to transport to a pediatric cardiac center. Vasodilation may cause hypotension, which is corrected with volume expansion using intravenous isotonic saline.

Use of PGE1 in neonates with ductal-dependent circulation is discussed in greater detail separately. (See "Diagnosis and initial management of cyanotic heart disease in the newborn", section on 'Prostaglandin E1'.)

Balloon atrial septostomy — BAS is performed to stabilize patients with severe hypoxemia as a consequence of inadequate mixing between the two parallel circuits [2,3]. It can be performed at the bedside with echocardiographic guidance or in the catheterization lab with the use of fluoroscopy and echocardiography. A balloon is placed across the atrial septum into the left atrium either through cannulation of the umbilical vein or percutaneously via the femoral vein. The balloon is inflated and then pulled vigorously back across the septum. The procedure is repeated at least once, and then echocardiographic and hemodynamic assessment of the defect is performed to ensure adequate intracardiac mixing.

If a BAS is performed and is successful, systemic oxygen saturations should begin to increase immediately. If there is adequate mixing at the atrial level, PGE1 infusion can often be discontinued, as long as there is no concern regarding aortic arch obstruction.

At our center, a bolus dose of heparin (50 units/kg) is given prior to performing BAS to minimize thrombus formation and subsequent embolization; however, this is not standardized and practice varies from center to center. The use of heparin in this setting is based on the theoretical concern for potential embolic brain injury as a consequence of BAS since the systemic venous return is in direct communication with the aorta and cerebral arterial circulation. Numerous studies have described magnetic resonance imaging (MRI) abnormalities in neonates with D-TGA, including focal ischemic changes, white matter changes, and periventricular leukomalacia, with reported prevalence rates ranging from 20 to 50 percent [4-7]. However, it is uncertain to what extent the BAS itself contributes to this since multiple other factors are associated with brain injury, particularly the degree of hypoxemia and/or hemodynamic instability.

SURGERY — The arterial switch operation (ASO) is the standard for surgical repair of D-TGA, and it has replaced the earlier atrial switch procedures developed by Mustard and Senning. Mortality associated with D-TGA has dramatically improved from approximately 90 percent for unoperated patients to rates of <5 percent following corrective surgery with the ASO [8]. (See 'After ASO' below.)

Timing of surgery — Surgery is typically performed within the first two weeks of life. The timing of repair must be tailored to the infant's medical status and to technical considerations of the center's surgical team. In our practice, most patients are referred for surgical repair during the first three to five days of life.

In a case series from a tertiary center in the United States, delay of ASO beyond three days after birth was associated with higher morbidity and increased health care costs [9].

The question of whether the timing of surgery impacts neurodevelopmental outcome was examined in a prospective study of 45 patients undergoing surgical repair of D-TGA who had magnetic resonance imaging (MRI) performed both pre- and postoperatively [7]. Older age at surgery (ie, postnatal age ≥2 weeks) and the presence of a ventricular septal defect (VSD) were associated with reduced perioperative brain growth. Older age at surgery was also predictive of lower language scores at age 18 months; however, the study was limited by the small number of patients who returned for 18-month follow-up (only 18 of 32 in the early repair group and 6 of 13 in the late repair group), thus limiting the ability to control for other potential confounding factors. As discussed below, neurodevelopmental outcomes following ASO are influenced by many other factors in addition to the timing of repair. (See 'After ASO' below.)

Choice of procedure — The choice of surgical procedure is generally dependent on the presence and nature of other cardiac anomalies, particularly VSD and left ventricular outflow tract (LVOT) obstruction:

●Simple D-TGA – In patients with D-TGA with intact ventricular septum and without any other cardiac defects, ASO is the standard procedure. (See 'Arterial switch operation (ASO)' below.)

●D-TGA with VSD – In patients with D-TGA and a VSD, the preferred procedure is ASO and VSD closure. (See 'Arterial switch operation (ASO)' below.)

●D-TGA with VSD and LVOT obstruction – In patients with D-TGA, VSD, and significant LVOT obstruction (due to valvular, subvalvar, or multilevel obstruction), the Rastelli procedure is an alternative surgical approach that may be considered. ASO with or without intervention to relieve the LVOT obstruction is also a reasonable approach for some patients [10]. The choice between the two approaches is complex and depends in part on the size of the VSD, the nature of the LVOT obstruction, and the status of the pulmonary (neo-aortic) valve. Surgical decision-making in this setting focuses on minimizing the risk of recurrent LVOT obstruction, optimizing neo-aortic valve function, and balancing the risks of future reoperation. Factors that impact surgical planning include the morphology and size of the pulmonary (neo-aortic) valve, the coronary artery anatomy, the complexity of the LVOT, and the potential pathway created by the baffle from the left ventricle to the aorta [11]. (See 'Rastelli procedure' below.)

Other surgical alternatives to address forms of D-TGA, VSD, and LVOT obstruction include the Nikaidoh procedure, Réparation à l'Etage ventriculaire (REV) procedure, and the Yasui procedure [12]. However, these surgeries are less commonly performed and their procedural details are beyond the scope of this topic.

Both the ASO and Rastelli procedures are surgical anatomic corrections resulting in a morphologic left ventricle as the systemic ventricle. In contrast, atrial switch procedures (also referred to as the Mustard and Senning procedures) involve rerouting venous return in the atria, resulting in a systemic right ventricle. Atrial switch procedures are now only rarely performed, chiefly for complex palliations of selected patients with congenitally corrected TGA (L-TGA). Management of L-TGA is discussed separately. (See "L-transposition of the great arteries (L-TGA): Management and outcome", section on 'Surgical management'.)

Arterial switch operation (ASO) — The ASO, originally performed in 1975, has become the standard corrective procedure for patients with D-TGA who do not have significant LVOT obstruction [13]. The ASO has been used as the preferred procedure for patients with D-TGA since the late 1980s. The overall perioperative mortality with the ASO has dropped to <1 percent in patients with simple D-TGA (ie, no other cardiac anomaly) and 4 percent for those with complex D-TGA (ie, additional cardiac anomaly) [8]. (See 'Outcome' below.)

ASO involves transection of both great arteries and then translocation of the vessels to the opposite root, thereby creating ventriculoarterial concordance (aorta to left ventricle, and pulmonary artery to right ventricle). Translocating the aorta also involves mobilization and reimplantation of the coronary arteries. In the ASO, "buttons" around the coronaries are created and then reimplanted into the neo-aortic root. Thus, it is important to delineate coronary artery anatomy, which is often aberrant, prior to surgical correction (figure 2). (See "D-transposition of the great arteries (D-TGA): Anatomy, clinical features, and diagnosis", section on 'Other cardiac anomalies'.)

The LeCompte maneuver, first described in 1981, is widely utilized when performing the ASO. This surgical technique places the bifurcation of the pulmonary arteries anterior to the aorta so that the left and right pulmonary arteries straddle the ascending aorta [14]. This allows for improved orientation of the branch pulmonary arteries and reduces the tension that is created from the anterior translocation of the pulmonary arterial root. The use of this maneuver decreases the risk of subsequent pulmonary artery stenosis and lowers reintervention rates [8,15,16]. (See 'Main and branch pulmonary artery stenosis' below.)

A complete preoperative anatomic assessment is essential. The presence of a VSD confers a higher probability of other anatomic abnormalities and can make surgical correction more difficult. Additional surgical procedures may be necessary at the time of the ASO (eg, aortic arch or atrioventricular valve repair). In patients with particular forms of LVOT obstruction, the Rastelli procedure may be preferred over the ASO. (See 'Rastelli procedure' below.)

Complications following ASO are discussed below. (See 'Complications after ASO' below.)

Rastelli procedure — The Rastelli procedure, first described in 1969, is typically performed in patients with D-TGA associated with a large VSD and LVOT obstruction [17]. The Rastelli procedure is preferred over the ASO for some patients because it provides superior and more durable relief of LVOT obstruction. The decision to perform the Rastelli procedure rather than the ASO with LVOT reconstruction is largely based on the nature of the LVOT obstruction and the status of the pulmonary (neo-aortic) valve.

The Rastelli procedure entails:

●Baffling the LVOT through the VSD, which closes the VSD and directs oxygenated blood from the left ventricle into the aorta

and

●Placing a conduit from the right ventricle to the pulmonary artery

The net result is that oxygenated blood from the left ventricle flows into the aorta, and the unoxygenated blood from the right ventricle is sent through the conduit into the pulmonary artery.

The Rastelli procedure requires a sizable and appropriately located VSD. In some patients with a restrictive VSD, the VSD is enlarged to allow appropriate baffle placement [18]. VSD enlargement may reduce the risk of recurrent LVOT obstruction [19] but may increase the risk of complete heart block and need for a pacemaker.

Perioperative mortality following the Rastelli procedure is <5 percent [12,20]. Long-term complications are discussed below. (See 'Complications after Rastelli procedure' below.)

Mustard and Senning procedures — Atrial switch procedures (including the Mustard and Senning procedures) are no longer performed in the modern era. They have been supplanted by the ASO because they are associated with long-term sequelae, including atrial arrhythmias and heart failure, that contribute to substantial late morbidity and mortality. These complications are related to the morphologic right ventricle serving as the pump for the systemic circulation. (See 'Complications after Mustard and Senning procedures' below and "Atrial arrhythmias (including AV block) in congenital heart disease".)

The Mustard and Senning procedures convert the parallel circulations of D-TGA into a circulation in series, thereby correcting cyanosis. However, they do not correct the underlying ventriculoarterial discordance of the aorta arising from the right ventricle and the pulmonary artery from the left ventricle (figure 3). In both procedures, an intra-atrial baffle is created to divert the deoxygenated systemic venous return through the mitral valve to the left ventricle, and into the pulmonary circulation via the pulmonary artery. Simultaneously, the baffle directs oxygenated pulmonary venous return across the tricuspid valve to the right ventricle, and subsequently to the aorta and the systemic circulation.

COMPLICATIONS

Complications after ASO — Complications that require reintervention occur in 5 to 25 percent of patients who have undergone arterial switch operation (ASO) [21-24]. The lesion that most often requires reintervention is pulmonary artery stenosis. Other complications are less common but may include coronary artery insufficiency, neo-aortic root dilation, and neo-aortic regurgitation.

Main and branch pulmonary artery stenosis — Pulmonary artery stenosis is the most common indication for reintervention following ASO in the first year of life. Inadequate growth at the neo-pulmonary anastomotic site contributes to pulmonary stenosis. The reported incidence of operative or catheterization-based reintervention ranges from 5 to 28 percent [25,26]. The wide range likely reflects differences in the surgical approach. The practice of routinely performing ASO in the neonatal period has reduced the need for pulmonary artery banding, with lower likelihood of subsequent stenosis. In addition, widespread use of the LeCompte maneuver has further reduced the incidence of pulmonary stenosis such that, in the modern era, the reintervention rate for pulmonary artery stenosis is <5 percent [8,15,16].

The obstruction most commonly is located in the supravalvar area of the main pulmonary artery, but it can be found at all levels of the right ventricular outflow tract and the branch pulmonary arteries. Supravalvar stenosis is thought to result from scarring at the anastomotic site or inadequate growth of the main pulmonary artery [26,27]. Branch pulmonary artery stenosis is generally a consequence of tension on the branch pulmonary arteries from their anterior translocation. The latter process may be mitigated by the LeCompte maneuver. Prior pulmonary artery banding, repair outside of the neonatal period, and surgical technique have been identified as risk factors.

Assessment for pulmonary artery stenosis begins with transthoracic echocardiography (TTE), which is a routine part of postoperative follow-up (see 'Follow-up care' below). TTE should include estimation of right ventricular pressure and assessment of proximal pulmonary arteries. TTE is not sensitive in detecting branch pulmonary artery stenosis, particularly after the LeCompte maneuver. Thus, if there is concern on TTE for elevated right ventricular pressure and the proximal pulmonary arteries are not well visualized, alternative imaging techniques, such as MRI, should be considered to better characterize the branch pulmonary arteries. In one study, proximal branch pulmonary artery stenosis was detected in only 60 to 70 percent of affected patients with TTE, while cardiac MRI identified almost 90 percent of lesions [28].

Intervention for pulmonary artery stenosis is generally warranted when the right ventricular pressure approaches systemic levels or if there is discrepant pulmonary blood flow on lung perfusion scan. Intervention typically consists of catheter-based dilation with stent placement, though surgical revision infrequently may be required, depending on the anatomical details.

Coronary artery stenosis or insufficiency — In the early years of the ASO, the coronary artery transfer was the surgical step that created the most difficulty and led to coronary insufficiency and high hospital mortality. As coronary artery transfer techniques improved, the hospital mortality dropped rapidly to a low level. The long-term success of this procedure, however, depends on the continued patency and growth of the coronary vessels to ensure adequate myocardial blood flow.

The incidence of coronary events continues to be bimodal, with the majority of events (89 percent) occurring in the first three months following the ASO [29]. These tend to be related to "kinking" or other anatomic obstructions to coronary perfusion. Unexplained ventricular dysfunction or poor hemodynamics should prompt early evaluation of the coronaries in the postoperative setting.

In one study of 1198 survivors after ASO born between 1982 and 2001, the freedom from coronary events was reportedly 92.7 percent at 1 year and 88.2 percent at 15 years, confirming the increased risk of coronary events in the postoperative period [29]. Risk factors for the development of coronary events include type of coronary anatomy (presence of a single coronary orifice or two coronary orifices originating close to each other) and the occurrence of a major intraoperative event (coronary translocation difficulty, left ventricular dysfunction, cardiac arrest, or temporary mechanical support at the end of the intervention). In this study, 11 patients underwent coronary revascularization at least one year after the ASO.

However, many patients with coronary artery stenosis or obstruction may be asymptomatic. Coronary angiography remains the gold standard for detecting coronary lesions. Studies that performed postoperative angiography report rates of coronary lesions between 3 to 6.8 percent [8,30-32]. Unfortunately, less invasive screening tests, such as electrocardiography, echocardiography, stress tests, and myocardial scintigraphy, have shown poor sensitivity for detecting coronary lesions and myocardial ischemia [29]. This is likely due to the development of a collateral coronary circulation that provides sufficient myocardial perfusion to limit detection of ischemia on these various screening tests. Magnetic resonance angiography and computed tomographic angiography appear to be reasonable alternatives to catheterization [33]. Increased experience with these modalities will hopefully reduce the need for invasive angiography in patients after ASO in the future.

Early atherosclerotic coronary artery disease — There is a hypothetical concern that the process of coronary transfer may predispose these patients to the development of intimal thickening due to altered flow dynamics. Lifetime cardiovascular risk assessment and lipid monitoring are essential for patients with D-TGA who have undergone ASO to minimize atherosclerosis and optimize cardiovascular health. This is discussed separately. (See "Overview of risk factors for development of atherosclerosis and early cardiovascular disease in childhood", section on 'Congenital heart disease' and "Overview of the management of the child or adolescent at risk for atherosclerosis".)

Neo-aortic root dilation — By reversing the great arteries in the ASO, the native pulmonary valve becomes the neo-aortic valve, and, over time, the neo-aortic root increases in size. In one report of 335 patients, neo-aortic root dilation (defined as a Z-score ≥3) was observed in 3, 8, 18, and 49 percent of patients at 1, 2, 5, and 10 years after ASO, respectively [16]. Although neo-aortic root dilation continued to develop in late follow-up, the increase was not clinically significant over time. Risk factors for neo-aortic root dilation were previous pulmonary arterial banding and ASO performed in a later era. The association between neo-aortic root dilation and surgery in a later era is likely due to changes in surgical technique related to increased size of the coronary "buttons" taken for the translocation. Pulmonary arterial banding has become a very rare intervention as complete repairs are typically performed in the first week of life, thereby reducing the frequency of pulmonary stenosis and neo-aortic root dilation.

Neo-aortic regurgitation — As mentioned above, in ASO, the native pulmonary valve becomes the neo-aortic valve, and the competence of this valve has been a concern as it functions in the setting of systemic pressure. Although mild aortic regurgitation (AR) appears to be a frequent finding, it does not usually progress to clinical significance.

In the available case series, approximately 15 to 20 percent of patients developed some degree of AR [16,34-36]. However, moderate to severe AR occurred in approximately 1 to 3 percent of patients by 10 years and 8 to 15 percent of patients by 20 years following ASO. Reoperation for AR was required in 1 to 2 percent of patients by 10 years and 2 to 3 percent by 20 years. AR did not appear to be associated with increased risk of late mortality.

Risk factors for developing clinically significant AR include [16,34-36]:

●Older age (ie, ≥1 year) at the time of ASO

●Presence of a ventricular septal defect (VSD)

●Presence of left ventricular outflow tract (LVOT) obstruction

●Prior pulmonary artery banding

●Mild AR in the immediately postoperative period

Complications after Rastelli procedure — Patients who have undergone the Rastelli procedure require serial conduit replacements. This is because the conduits do not grow with the child, so they become stenosed over time. Conduit replacement is the most common reason for reintervention after the Rastelli procedure. In addition, atrial and ventricular arrhythmias are more frequent compared with the ASO, and right and left ventricular failure may occur.

In a case series of 40 patients with D-TGA, VSD, and pulmonary stenosis who underwent Rastelli operation between 1988 and 2008, after a mean follow-up of 8.6 years, there were three late deaths, one patient underwent cardiac transplantation, and one patient was lost to follow-up [20]. Sixteen patients (40 percent) required reoperation for conduit stenosis and two for LVOT obstruction.

Complications after Mustard and Senning procedures — Complications following Mustard and Senning procedures (atrial switch procedures) include right ventricular failure, arrhythmias, and baffle-related sequelae [37].

Right ventricular failure — Atrial switch procedures provide physiologic correction; however, they are not anatomically correct and place the right ventricle and tricuspid valve into the systemic circulation. All patients undergoing an atrial switch procedure develop some degree of right ventricular dysfunction [38,39], and some progress to late systemic (right) ventricular failure, which can result in death or serious morbidity requiring reoperation and possibly cardiac transplantation [40-42].

Arrhythmias — Arrhythmias are common complications of the atrial switch repairs that may lead to significant morbidity and mortality in adulthood. In several long-term follow-up studies, approximately one-half to two-thirds of patients remain in sinus rhythm [37,42-44]. The remaining patients have a variety of arrhythmias, including sinus node dysfunction, atrial flutter and fibrillation, and ventricular tachycardia. The loss of sinus rhythm is felt to precipitate further dysfunction in the systemic right ventricle and can lead to worsening heart failure and sudden death [37,42,44,45].

Arrhythmias may be caused by the presence of numerous atrial suture lines or from long-standing atrial enlargement due to right ventricular dysfunction or tricuspid regurgitation [46].

Baffle-associated complications — Baffle-associated complications include the following [46]:

●Obstruction at the right atrial and superior vena caval junction is a recognized complication of the Mustard procedure. The clinical presentation may include chylothorax, upper extremity edema, or facial plethora.

●Pulmonary venous obstruction is a complication more commonly associated with the Senning procedure. Pulmonary venous congestion may be an early manifestation. Progressive obstruction may be seen later and may present with symptoms of reactive airway disease.

●Baffle leaks due to dehiscence of suture lines are relatively uncommon but may result in significant interatrial shunting in some cases.

FOLLOW-UP CARE — All patients who have undergone D-TGA repair should have long-term follow-up care by a cardiologist with expertise in congenital heart disease. Follow-up care focuses on timely detection of potential complications that may arise following the various surgical repairs and general assessment of overall functional status. Follow-up visits should include a focused history, physical examination, and electrocardiogram (ECG). In addition, echocardiography and/or MRI are performed at intervals determined by the patient's clinical course.

●History – The history includes asking about the following:

•Episodes of syncope or palpitations that may suggest an underlying arrhythmia (see 'Arrhythmias' above)

•Increasing exercise intolerance suggestive of declining systemic ventricular function or increasing pulmonary artery obstruction (see 'Main and branch pulmonary artery stenosis' above and 'Right ventricular failure' above)

•Exertional chest pain may suggest coronary artery insufficiency (see 'Coronary artery stenosis or insufficiency' above)

●Physical examination – The physical examination includes:

•Vital signs, particularly the pulse to determine any irregularity that suggests an underlying arrhythmia

•Cardiac auscultation to detect any murmur (eg, pulmonary stenosis, aortic or tricuspid insufficiency) or gallop (eg, failure)

•Examination for signs of cardiac failure including crackles, peripheral edema, jugular venous distension, and hepatomegaly

•Edema limited to the face and upper extremities suggest superior venal caval obstruction due to a baffle complication seen in the Mustard procedure (see 'Baffle-associated complications' above)

●Testing – Routine follow-up tests include the following:

•ECG and Holter monitoring are performed routinely to detect and diagnose arrhythmias. The ECG can also be used to screen for ischemic changes, but its sensitivity to detect coronary insufficiency is poor.

•Routine echocardiography is used to assess ventricular function, detect pulmonary artery stenosis, and evaluate competency of the neo-aortic valve. MRI may be warranted if the pulmonary arteries are not adequately visualized on echocardiography.

•Risk assessment for atherosclerosis and lipid monitoring should be performed in all patients who undergo arterial switch operation (ASO) since they are at risk of developing early coronary artery disease (algorithm 1). (See 'Coronary artery stenosis or insufficiency' above and "Overview of risk factors for development of atherosclerosis and early cardiovascular disease in childhood", section on 'Congenital heart disease'.)

•If there are historical features suggestive of ischemia or signs of unexplained depressed function on echocardiography, coronary artery insufficiency must be considered. Angiography remains the gold standard for detection of coronary artery occlusion in patients who undergo ASO. However, in experienced centers, magnetic resonance angiography and computed tomographic angiography may be reasonable alternatives to catheterization [33]. (See 'Coronary artery stenosis or insufficiency' above.)

●Endocarditis prophylaxis – Antibiotic prophylaxis for subacute bacterial endocarditis is not required in patients who have undergone ASO so long as they do not have any residual lesions. Prophylaxis for bacterial endocarditis is discussed in detail separately. (See "Prevention of endocarditis: Antibiotic prophylaxis and other measures".)

●Physical activity and sports participation – The 2015 scientific statement of the American Heart Association and American College of Cardiology provides competitive athletic participation guidelines for patients with congenital heart disease including D-TGA [47]. It utilizes the classification of sports based on increasing static and dynamic components (figure 4). Before participation in competitive sports, patients who have undergone repair of D-TGA should be evaluated with a clinical assessment, ECG, imaging assessment of ventricular function (typically with echocardiography), and exercise testing [47]. (See "Exercise testing in children and adolescents: Principles and clinical application" and "Exercise ECG testing: Performing the test and interpreting the ECG results".)

•Patients with D-TGA who have undergone ASO (most patients born after 1990) have a systemic left ventricle and have a low risk of arrhythmia. Those with no cardiac symptoms, normal ventricular function, normal exercise testing, and no tachyarrhythmias can participate in all sports. Because the coronary arteries are reimplanted in ASO, symptoms or stress test results that may be related to myocardial ischemia should be evaluated with coronary angiography [48]. Patients with evidence of coronary ischemia should be restricted from all competitive athletics [47].

•Patients with D-TGA who have undergone an atrial switch procedure (Mustard or Senning) have interatrial baffles and a systemic right ventricle. Those with no symptoms and without clinically significant arrhythmias, ventricular dysfunction, exercise intolerance, or exercise-induced ischemia may participate in low- and moderate-intensity competitive exercise (classes IA, IB, IIA, and IIB) [47]. Patients with severe right ventricular dysfunction, severe right ventricular outflow tract obstruction, or recurrent or uncontrolled arrhythmias should be restricted from all competitive sports.

Physical activity and exercise in patients with congenital heart disease are discussed in detail separately. (See "Physical activity and exercise in patients with congenital heart disease".)

OUTCOME — The overall outcome for patients with D-TGA has dramatically improved from a uniformly fatal disease in untreated patients to long-term survival rates of >90 percent due to advances in initial management and surgical correction. However, survivors are at risk for morbidity that may require further surgical intervention. Patients who have undergone the arterial switch operation (ASO) have the best long-term survival, lowest morbidity, and best functional outcome.

After ASO

●Survival – The long-term survival rates for patients with D-TGA following surgical correction with ASO are excellent with >95 percent survival at 15 to 25 years following discharge [8,15,21,24,49]. Although there have been no clinical trials comparing the different surgical approaches, a review of long-term outcome from a single tertiary center reported improved long-term survival when the center changed from the Mustard and Senning atrial switch procedures to the ASO in 1983 [50]. Perioperative mortality rates were similar (6, 5, and 8 percent for ASO, Senning procedure, and Mustard procedure, respectively) and long-term survival rates 20 years after discharge were superior with the ASO (97 versus 93 and 82 percent for the Senning procedure and Mustard procedures, respectively).

Mortality is greater in patients with complex D-TGA (additional cardiac anomaly) compared with those with simple D-TGA (no other cardiac anomaly) [15,22,27,51-53]. Other reported predictors of mortality include:

•Prematurity (<36 weeks gestation) [22]

•Low birth weight [51,54]

•Intramural or single coronary artery [55]; however, multiple groups have not demonstrated increased mortality risks based on coronary anatomy [15,22,51]

•Aortic arch obstruction

•Right ventricular hypoplasia [52]

●Long-term morbidity – Most patients who undergo ASO in the early neonatal period remain free from long-term cardiovascular complications [8,21,24,50]. In a single-center study of 400 patients who underwent ASO between 1983 and 1999, 93 percent were free from major cardiovascular adverse events (including arrhythmia, cerebrovascular accident, heart failure-related hospitalization, or cardiovascular death) at 25 years [24]. Freedom from catheter or surgical intervention was 82 percent at 10 years and 76 percent at 25 years. As previously described, the most common indication for reintervention following ASO is pulmonary artery stenosis. (See 'Main and branch pulmonary artery stenosis' above.)

In another study of 151 survivors after ASO performed between 1977 and 2000, most patients (>95 percent) had no functional limitations at last follow-up (ie, New York Heart Association [NYHA] functional class I) [8]. The most common complication requiring reintervention was pulmonary artery stenosis (occurring in 17 percent of patients). Other long-term complications included left ventricular dysfunction (3 percent), arrhythmias (3 percent), coronary artery sequelae (3 percent), and moderate to severe neo-aortic valve regurgitation (1 percent). (See 'Complications' above.)

Most patients with repaired D-TGA surviving into adulthood can expect normal functional health status compared with age-matched healthy peers. In a report of 71 patients with D-TGA who underwent repair between 1980 and 1985 (most patients underwent ASO) and were followed to approximately age 30 years, self-reported quality-of-life scores were similar to those of healthy controls in all domains except physical health, in which patients with D-TGA scored slightly lower [56]. The study did not detect any differences in quality-of-life scores based upon D-TGA morphology (simple versus complex) or type of repair (ASO versus atrial switch); however, the small number of patients in each group precludes drawing any firm conclusions.

●Exercise capacity – Most patients who have undergone ASO do not have limits to ordinary activity (ie, they can be classified as NYHA class I (table 1)) [8,21]. However, they may have mildly reduced exercise capacity compared with healthy controls [57,58]. In a study that evaluated exercise capacity in patients with D-TGA following ASO or atrial switch as compared with normal controls, patients repaired with ASO had low-normal cardiorespiratory exercise capability (91 percent of normal) [59]. In one study, poor exercise performance was associated with ventricular septal defect (VSD) repair, residual right-sided obstructive lesions, decreased left ventricular function, and repair in an earlier surgical era [60].

●Neurodevelopmental outcome – The available evidence suggests that survivors after ASO are more likely to have neurodevelopmental impairments compared with healthy controls [7,61-66].

In two reports of the same case series of 139 adolescents who underwent ASO in infancy and who had neuropsychological testing at 16 years of age, 65 percent had a history of requiring special educational services (including tutoring, grade retention, early intervention, special education, occupational therapy, psychotherapy, and counseling) [66]; 20 percent met diagnostic criteria for attention deficit hyperactivity disorder; and overall global psychosocial functioning scores were reduced compared with healthy peers [67]. In addition, abnormalities on brain MRI were more frequently seen in the ASO group compared with a reference group of normal adolescents (33 versus 4 percent). The occurrence of seizures in the postoperative period was the only factor independently associated with lower neuropsychological scores.

In a prospective study of 45 children with D-TGA who underwent neurodevelopmental testing at age four to six years, scores on tests of cognition, language, and verbal working memory were in the normal range but were lower than healthy controls in the domains of inhibition control, cognitive flexibility, social cognition, and executive function [68,69]. Test performance was better in patients with prenatally diagnosed D-TGA compared with postnatally diagnosed.

Neurodevelopmental outcomes are likely highly influenced by the infant's prenatal, neonatal, and perioperative course. Factors that may influence the outcome include:

•Timing of diagnosis (prenatal versus postnatal)

•Underlying genetic abnormalities

•Associated cardiac defects (eg, VSD, which increases complexity of the surgical repair and may prolong cardiopulmonary bypass time)

•Degree of hypoxemia and/or hemodynamic instability in the newborn period

•Adequacy of nutrition

•Timing of surgery

•Perioperative complications (eg, embolic stroke, seizures, necrotizing enterocolitis)

•Need for extracorporeal membrane oxygenation

Abnormalities on brain MRI are common in neonates with D-TGA, with reported prevalence rates ranging from 20 to 50 percent [4-7]. MRI abnormalities may include focal ischemic changes, white matter changes, volume loss, and periventricular leukomalacia [4-7]; however, these findings correlate poorly with long-term neurologic outcome [7].

The available evidence regarding the risk of neurodevelopmental impairment in children with D-TGA highlights the importance of close neurodevelopmental monitoring in these patients. We agree with the recommendations of the scientific statement from the American Heart Association that children with D-TGA should undergo appropriate surveillance, screening, and/or referral for neurodevelopmental impairment [70]. (See "Developmental-behavioral surveillance and screening in primary care".)

After Mustard or Senning procedure

●Survival – Long-term survival after the Mustard or Senning atrial switch procedures is approximately 80 to 85 percent at 20 years, 75 percent at 30 years, and 65 percent by 40 years [42,71]. The risk of late mortality is higher following the Mustard procedure compared with the Senning (odds ratio 2.9, 95% CI 1.9-4.5) [71]. The most common causes of late mortality are sudden cardiac death and congestive heart failure (each accounting for approximately 20 to percent of late deaths) [71,72].

Risk factors for late mortality in these patients include [71,72]:

•Complex D-TGA (ie, additional cardiac anomalies)

•Atrial or ventricular arrhythmias

•QRS duration >120 msec

•Ventricular systolic dysfunction

•Prior hospital admission for heart failure

●Long-term morbidity – Patients who have undergone the Mustard or Senning operation are more likely to experience long-term morbidities and are more likely to require reintervention compared with those who undergo ASO [41,45,50]. Arrhythmias and right heart failure are the most common long-term complications. (See 'Complications after Mustard and Senning procedures' above.)

In a study of 468 patients who underwent the Mustard or Senning operation between 1967 and 2003 with a median follow-up of 26 years, 7 percent required reoperation, 15 percent required pacemaker placement, 2 percent underwent heart transplantation, and 38 percent died (approximately one-half of these were perioperative deaths) [45]. Need for pacemaker placement was a risk factor for mortality.

●Exercise capacity – Patients who have undergone the Mustard or Senning operation may have substantial activity limitations [59]. In a study that evaluated exercise capacity in patients with D-TGA following ASO or atrial switch as compared with normal controls, patients repaired with atrial switch procedures had poorer exercise capacity (75 percent of healthy controls) [59].

After Rastelli procedure — In reports of long-term outcomes after the Rastelli procedure, 10-year survival rates were 80 to 94 percent [73,74]. Complications after the Rastelli procedure include conduit stenosis, arrythmias, and ventricular failure. Conduit replacement is the most common reason for reintervention after the Rastelli procedure. (See 'Complications after Rastelli procedure' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Congenital heart disease in infants and children".)

SUMMARY AND RECOMMENDATIONS

●Anatomy and physiology – D-transposition of the great arteries (D-TGA) is a form of cyanotic congenital heart disease characterized by ventriculoarterial discordance (the aorta arises from the right ventricle and the pulmonary artery from the left ventricle) (figure 1). This defect results in the creation of two parallel circulations in which deoxygenated systemic venous blood returns to the right atrium and is sent back to the systemic circulation via the right ventricle and aorta, while simultaneously, oxygenated pulmonary venous blood returning to the left atrium is pumped back to the lungs via the left ventricle and pulmonary artery. (See "D-transposition of the great arteries (D-TGA): Anatomy, clinical features, and diagnosis".)

●Initial postnatal management – Neonates with suspected or confirmed D-TGA should be urgently transported to an institution with expertise in managing infants with congenital heart disease. Prior to transport, consultation with a pediatric cardiologist is advised to ensure optimal medical management before and during the transport.

Initial postnatal management is focused on stabilization of cardiac and pulmonary function and ensuring adequate systemic oxygenation. Therapy is directed towards providing sufficient mixing between the two circulatory systems by maintaining patency of the ductus arteriosus with prostaglandin E1 (PGE1; alprostadil, 0.05 mcg/kg per minute) infusion. Patients with severe hypoxemia may require balloon atrial septostomy (BAS). (See 'Initial postnatal management' above and "Diagnosis and initial management of cyanotic heart disease in the newborn", section on 'Prostaglandin E1'.)

●Surgical repair – The arterial switch operation (ASO) is the standard for surgical repair of D-TGA. ASO involves transection of both great arteries and then translocation of the vessels to the opposite root, thereby creating ventriculoarterial concordance (aorta to left ventricle and pulmonary artery to right ventricle). (See 'Arterial switch operation (ASO)' above.)

In patients with D-TGA associated with a ventricular septal defect (VSD) and left ventricular outflow tract (LVOT) obstruction, the Rastelli procedure is an alternative option for surgical repair. The Rastelli procedure entails closing the VSD with a baffle, so that the oxygenated blood from the left ventricle is directed into the aorta, and placing a conduit from the right ventricle to the pulmonary artery. (See 'Rastelli procedure' above.)

ASO has replaced the earlier atrial switch procedures (also referred to as the Mustard and Senning procedures) because of its superior long-term outcomes. (See 'Mustard and Senning procedures' above.)

●Postoperative outcomes and complications – Perioperative mortality associated with ASO is very low (<1 percent for simple D-TGA [ie, no other cardiac anomaly]; approximately 4 percent for complex D-TGA [ie, additional cardiac anomaly]). Long-term survival is excellent, with >95 percent survival at 15 to 25 years following surgery. (See 'After ASO' above.)

The most common complication of ASO is pulmonary artery stenosis, which may require reintervention. Other long-term complications are uncommon but may include coronary artery insufficiency, neo-aortic root dilation, and neo-aortic regurgitation. Patients who underwent surgical repair in an earlier era with an atrial switch procedure (Mustard and Senning procedures) are more likely to experience long-term morbidities, especially arrhythmia and right heart failure. (See 'Complications after ASO' above and 'Complications after Mustard and Senning procedures' above.)

●Long-term follow-up – All patients who have undergone surgical repair of D-TGA require long-term cardiology follow-up. Follow-up visits should include review of the interval history (focusing on any new cardiac symptoms), physical examination, and testing that typically includes an electrocardiogram (ECG), cardiac imaging (usually with echocardiography), and lipid screening. (See 'Follow-up care' above.)

Patients who have undergone ASO have a slightly reduced exercise capacity, but this does not typically restrict their daily level of activity. However, patients who wish to participate in competitive sports should undergo formal evaluation including ECG, echocardiogram, and exercise testing. (See 'After ASO' above and 'Follow-up care' above.)

Patients with D-TGA may have mild long-term neurodevelopmental impairment, most likely due to perioperative factors including hypoxemia, cardiopulmonary bypass, and hemodynamic instability. Following repair of D-TGA, all infants and children should undergo routine screening for developmental delays. (See 'Outcome' above and "Developmental-behavioral surveillance and screening in primary care".)