First described by the Scottish pathologist Dr. Matthew Baillie in 1797 (1), transposition of the great arteries was for decades merely known as a fascinating disease that was universally fatal. Many innovations in therapy along the way, including the introduction of the balloon atrial septostomy by Rashkind and Miller (2) and Jatene’s report of the arterial switch operation (3) revolutionized the way that we now care for these infants. Patients who were initially felt to have a fatal disease can now be expected not only to survive, but also to have an excellent quality of life. Indeed, the innovations in therapy over the years for patients with transposition of the great arteries became the basis for many advances in technology in the disciplines of interventional cardiology and cardiac surgery.

In this chapter, we will discuss lesions with concordant atrioventricular connections and discordant ventriculoarterial connections, or “simple transposition.” This form of transposition is also known as D-TGA (dextro transposition of the great arteries) or D-looped transposition. These lesions encompass transposition of the great arteries with intact ventricular septum, transposition of the great arteries with ventricular septal defect, and transposition of the great arteries with ventricular septal defect and left ventricular outflow tract obstruction (Fig. 47.1). Other forms of transposition of the great arteries, for example, congenitally corrected transposition of the great arteries or transposition of the great arteries in association with other lesions, such as double outlet right ventricle, are discussed elsewhere in this text.

Transposition of the great arteries is the most common form of cyanotic congenital heart disease presenting in the newborn period and comprises approximately 5% of all congenital heart defects (4). Early estimates from the Report of the New England Regional Infant Program (5) reported the incidence of transposition of the great arteries to be 218 per million live births. A more recent meta-analysis from 41 studies estimated the incidence to be a median of 303 and mean of 315 per million live births (6). Investigators from the National Birth Defects Prevention Network (NBDPN) found the prevalence of transposition of the great arteries to be 4.73 per 10,000 live births during 1999–2001 (7). These numbers were revised in an updated report (8) to 3 per 10,000 live births during the time period of 2004–2006, largely due to more accurate coding for inclusion criteria.

The physiology in transposition of the great arteries is significantly more complex than that of simple left-to-right or right-to-left shunts, and can vary widely according to associated lesions and according to the state of the patient’s pulmonary vascular resistance during the normal neonatal evolution from fetal, to transitional, to adult-type circulation.

The essence of the circulatory derangement in transposition of the great arteries is that the systemic and pulmonary circulations are arranged in parallel rather than in series. The left ventricle, which receives fully oxygenated blood from the pulmonary veins, pumps this oxygenated blood through the pulmonary valve into the lungs, where it cannot possibly pick up any more oxygen. Similarly, the right ventricle, which receives deoxygenated blood from the systemic veins, pumps this deoxygenated blood across the aortic valve back to the body, where it cannot deliver much oxygen to the body. After being pumped to the body, the even more deoxygenated blood returns to the right ventricle, and is once again pumped to the body. According to this construct, one can see that within just a few heartbeats, the right ventricle would be pumping blood that is essentially devoid of oxygen, while the left ventricle keeps recirculating fully oxygenated blood to, and from, the lungs. Clearly survival with the systemic and pulmonary circulations arranged in parallel is not possible, and requires some blood to exit the pulmonary circuit in order to enter the systemic circuit, and for blood to similarly exit the systemic circuit in order to enter the pulmonary circuit. Any blood exiting one circuit to enter the other would result in one circulation emptying itself into the other; there must therefore be an equal flow returning blood back to the original circuit. The levels of communication that allow blood to cross from one circuit to the other can occur either centrally at the atrial (atrial septal defect or patent foramen ovale), ventricular (ventricular septal defect), or great artery level (patent ductus arteriosus), or can occur peripherally at the naturally existing shunts at the level of the bronchial circulation and thebesian veins.

In order to correctly describe the intricacies of this highly variable physiology, it is important to recall some basic definitions, which will first be illustrated using simpler forms of congenital heart disease, and then be applied to the understanding of transposition of the great arteries. First, pulmonary blood flow (Q_P) refers to the entire flow through the pulmonary vascular bed, while systemic blood flow (Q_S) refers to the entire flow through the systemic vascular bed. Effective pulmonary blood flow (Q_EP) refers to the flow of systemic venous blood into the pulmonary circulation, that is, the deoxygenated blood that is being sent to the lungs to undergo oxygenation. Similarly, effective systemic blood flow (Q_ES) refers to the flow of pulmonary venous blood into the systemic circulation, that is, the oxygenated blood that is being sent to the body to deliver oxygen. The amount of oxygen absorbed by the blood circulating in the lungs must equal the amount of oxygen delivered by the blood circulating to the body, hence Q_EP must always equal Q_ES. In a completely normal heart, Q_P = Q_EP, Q_S = Q_ES, and Q_P = Q_S. A left-to-right shunt is defined as the flow of blood which has already completed its circulation to the lungs, and yet is being sent to the lungs once again without having seen the systemic circulation, while a right-to-left shunt represents the flow of blood that has already completed its course through the systemic circulation, and yet is being circulated once more to the systemic circulation without seeing the pulmonary circulation. In hearts with simple left-to-right shunts (e.g., atrial or ventricular septal defects with pure left-to-right shunting), the flow being shunted into the pulmonary circulation across the defect (Q_L→R) is additive to the flow coming in normally from the systemic veins to enter the lungs (Q_EP), hence Q_P = Q_EP + Q_L→R. Similarly, in hearts with simple right-to-left shunts (e.g., an atrial septal defect shunting purely right-to-left), the flow shunted right to left across the atrial septal defect (Q_R→L) is being added to the normal blood flow from the pulmonary veins being pumped out to the aorta (Q_ES), hence the total amount being sent to the body is the sum of these two blood flows, that is, Q_S = Q_ES + Q_R→L.

In transposition of the great arteries specifically, the terms “left-to-right shunt” or “right-to-left shunt” can be confusing, and some authors have attempted to avoid confusion by distinguishing the terms “physiologic left-to-right shunt” from “anatomic left-to-right shunt.” We prefer to use the term “left-to-right shunt” to mean what it intends in any other cardiac lesion, that is, the “physiologic left-to-right shunt.” Rather than utilizing the term “anatomic shunt” we prefer to describe the direction of flow from one chamber or vessel to the other chamber or vessel. This will become clear as we discuss what precisely constitutes the “shunted” blood in transposition of the great arteries. First, it is important to recognize that the deoxygenated blood pumped by the right ventricle to the aorta in transposition of the great arteries constitutes, in fact, a large right-to-left shunt (Q_R→L) because it fulfills the definition described above (systemic venous blood being pumped back to the body without having seen the pulmonary circulation). Similarly, the oxygenated blood pumped by the left ventricle to the pulmonary artery is, in fact, the left-to-right shunt (Q_L→R) because it fulfills the definition of a left-to-right shunt (pulmonary venous blood being pumped back to the lungs without having circulated through the body). It is therefore important to note that the small amount of blood typically crossing the foramen ovale from the left atrium to the right atrium in a newborn with TGA is not a left-to-right shunt, but is rather the only portion of the oxygenated blood that is flowing “correctly” to the body (hence, it is actually the Q_ES). It is therefore inaccurate in the setting of transposition of the great arteries to describe flow occurring from the left atrium to the right atrium across a foramen ovale as a left-to-right shunt, although some authors might use the term “anatomic left-to-right shunt” because it is occurring from a left-sided anatomical structure to a right-sided anatomical structure (certainly, all would agree that this does not constitute a physiologic left-to-right shunt). Similarly, any blood flowing from the aorta to the pulmonary artery across the ductus arteriosus actually represents deoxygenated blood being sent to the lungs to pick up oxygen, hence it constitutes the Q_EP, and it most certainly is not a physiologic shunt. In addition, it can be seen that the terminology “anatomic shunt” breaks down when describing a shunt between the aorta and the pulmonary artery: Normally the aorta is considered a left-heart structure and the pulmonary artery a right-heart structure, but in transposition of the great arteries, the aorta is connected to the right ventricle and the remainder of the right heart. Does that make the aorta an anatomic right-heart structure, while the pulmonary artery becomes a left-heart structure? This becomes confusing, and it is therefore best to avoid the terms “anatomic left-to-right shunt” and “anatomic right-to-left shunt,” in favor of describing the actual chambers or vessels that allow communication between them and the manner in which the flow occurs between these structures (e.g., “flow from the aorta to the pulmonary artery,” “flow from the pulmonary artery to the aorta,” “flow from the left atrium to right atrium,” etc.). Similarly when interpreting hemodynamic cardiac catheterization data in these patients, it is important to understand the limitations in this situation and to interpret the data in the context of “transposition physiology” discussed in this section.

When considering a newborn baby with transposition of the great arteries, the foramen ovale is expected to be available for shunting, as is the ductus arteriosus in the first few minutes or hours of life. Hence, the two parallel circuits consisting of the pulmonary and systemic circulations are expected to have two possible areas of communication between them in almost every newborn with transposition of the great arteries. Superimposed
on this congenital anomaly is the near-normal evolution of pulmonary vascular resistance in the newborn period. Hence, immediately after birth the pulmonary vascular resistance is elevated, and accordingly it is possible for the ductus arteriosus to have flow occur bidirectionally, typically from pulmonary artery to aorta in systole, and from aorta to pulmonary artery in diastole. This phenomenon is responsible for the classic “reversed differential cyanosis” seen in newborns with transposition of the great arteries (i.e., the lower body will be relatively pink while the upper body will be relatively cyanotic). It is important to note that this phenomenon is a fleeting one, typically lasting only the first few hours of life. The two conditions necessary for the presence of this phenomenon are patency of the ductus arteriosus, and elevation of the pulmonary vascular resistance. The systolic flow of blood from pulmonary artery to aorta brings highly oxygenated blood from the pulmonary artery to the descending aorta to result in relatively higher saturations in the lower extremities compared to the upper extremities. However, within the first few hours of life as the pulmonary vascular resistance drops precipitously, there will cease to be systolic flow from pulmonary artery to aorta across the ductus, and eventually the flow across the ductus will entirely be from aorta to pulmonary artery throughout the cardiac cycle. When this occurs, differential cyanosis disappears, and with it an important bedside clinical clue to the presence of transposition physiology. The increased flow across the ductus arteriosus from aorta to pulmonary artery caused by the drop in pulmonary vascular resistance also results in increased pulmonary venous return to the left atrium. The relative increase in left atrial pressure causes the foramen ovale to flow from the left atrium to right atrium. During this phase of the physiology, any net amount of blood flow occurring from aorta to pulmonary artery across the ductus (Q_EP) is matched by an equal flow occurring from left atrium to right atrium (Q_ES), and these two flows are keeping the newborn alive, as these are the only flows participating in actual oxygen exchange. It is therefore evident that restriction at the level of the foramen ovale and/or ductus arteriosus limits the ability of deoxygenated blood to enter the lungs (Q_EP) and oxygenated blood to enter the body (Q_ES). This physiologic set up is the justification of two of the most important strategies for treatment of newborns with transposition of the great arteries: the institution of a prostaglandin infusion to maintain patency of the ductus, and the performance of a balloon atrial septostomy to enlarge the foramen ovale. It is a common misconception that the ductus and foramen are necessary to improve “mixing” in transposition of the great arteries. In fact, it is not mixing back-and-forth across these structures, but rather net flow across these structures (from aortic to pulmonary artery, and from left atrium to right atrium) that is occurring during this physiologic state in transposition of the great arteries (41). After a septostomy has been performed, the systemic arterial saturation typically increases tremendously, such that prostaglandin can frequently be discontinued. In the setting of a closed ductus, the patient then relies essentially on the atrial defect that has been created via balloon atrial septostomy to maintain adequate oxygen saturation. In this physiologic state, the flow of blood across the atrial septal defect occurs bidirectionally, thus it is perfectly accurate to describe the atrial septal defect as being responsible for improving “mixing.” Whether prostaglandin can be discontinued after a balloon atrial septostomy is very much institution dependent, with some institutions preferring to discontinue prostaglandin infusions due to their softening effect on the cardiac tissues at the time of the operation, while other institutions prefer that their patients have higher oxygen saturations preoperatively, to which ductal patency certainly contributes.

Associated lesions such as ventricular septal defects, pulmonary stenosis, or both, add additional complexity to the physiology of transposition of the great arteries. When present, a ventricular septal defect predominantly will shunt from right ventricle to left ventricle, thus allowing deoxygenated blood from the right ventricle to enter the pulmonary arteries, contributing to Q_EP. Accordingly, the saturations will generally be higher in patients with transposition of the great arteries with ventricular septal defect. In patients with transposition of the great arteries, ventricular septal defect, and pulmonary stenosis, the pulmonary stenosis, especially if severe, will significantly limit pulmonary blood flow and contribute to lower oxygen saturation.