Physiology of Congenital Heart Disease in the Neonate

† Deceased.

Introduction

Understanding the cardiovascular physiology of a neonate with a cardiac malformation requires consideration of many factors, including myocardial systolic and diastolic function; intravascular volume; cardiac and vascular transmural pressures; the pattern of intracardiac blood flow with shunting lesions; and arterial and pulmonary venous oxygen (O ₂ ) content. Any description taking into account only these variables would be incomplete, however, because cardiac lesions affect somatic physiology in various indirect ways (e.g., sympathetic nervous system activation). Indeed, such alterations account for many of the clinical manifestations of cardiac lesions. ^, The net physiologic effects result from complex interactions of (1) cardiac and vascular anatomy and function; (2) lung function; (3) O ₂ carrying capacity, O ₂ release, and delivery; and (4) other factors, such as cellular metabolism, O ₂ consumption ( $\dot{V} O_{2}$ $\dot{V} O_{2}$ ), neurohormonal mechanisms, lymphatic structure and function, and yet others, too many to mention.

Anatomic cardiac lesions are physiologically complex for other reasons as well. With cardiac structural anomalies, more than one type of physiologic derangement may be operative. For example, with a large patent ductus arteriosus (PDA), the left ventricle (LV) may have both an increased volume load and reduced myocardial perfusion, the latter due to decreased aortic diastolic pressure. The consequences of multiple anatomic abnormalities may be additive, or conversely, may offset each other. Additionally, subtle differences in physiology may result in variable clinical phenotypes in anatomically identical patients. For example, the arterial O ₂ saturation (Sa o ₂ ) in neonates with d-transposition of the great arteries after balloon atrial septostomy is variable and unpredictable, apparently related to subtle differences in physiology.

Second, physiologic variables change over time. Some factors affecting cardiovascular physiology have effects over months, especially those causing anatomic changes (as in pressure-related remodeling of myocardium or pulmonary blood vessels), whereas others are relevant over seconds or minutes (e.g., those related to changes in vascular tone). The physiology of a given cardiac defect thus reflects multiple contributing factors. Assessment of cardiovascular physiology has improved with noninvasive imaging, such as echocardiography, magnetic resonance imaging, and other strategies; however, precise determination of relevant pressures, blood flows, and ventricular function can be challenging when using noninvasive approaches and may occasionally require the selective use of invasive approaches, especially cardiac catheterization.

This chapter focuses on congenital anatomic cardiovascular lesions; nonanatomic abnormalities (e.g., arrhythmia, intrinsic myocardial muscle dysfunction) are not discussed. The biology of specific elements related to cardiovascular function (e.g., cardiac mechanics) and structural alterations caused by cardiac lesions (e.g., ventricular hypertrophy with pressure overload) or anatomic details of the malformations are also beyond the scope of this chapter.

Understanding congenital cardiac malformations requires an appreciation of normal fetal and neonatal physiology, ^, the physiology and pharmacology of the pulmonary circulation, and general principles of cardiac hemodynamics. The following are discussed in this chapter: (1) how congenital cardiac lesions affect the phenotype of the neonate; (2) select basic hemodynamic concepts necessary for understanding these malformations; (3) the effect of intracardiac shunting on the clinical phenotype; and (4) nonshunting anatomic abnormalities (e.g., abnormal valves) commonly seen in neonates.

How Congenital Heart Lesions Alter the Clinical Phenotype of the Neonate

A congenital anatomic lesion may affect a neonate’s phenotype in one of three ways.

No Appreciable Effect

Some anatomic lesions, such as a small ventricular septal defect (VSD) may never have any appreciable physiologic consequences. In other cases, a lesion destined to later have a significant impact may have few manifestations in the immediate postnatal period but become relevant later. The classic example is a large VSD, which has little impact at birth yet causes congestive heart failure (CHF) a few weeks later. The clinical phenotype changes partly because the relatively high neonatal pulmonary vascular resistance (PVR) falls over time, increasing left-to-right shunting and hence the LV volume pumped, and partly because of other changes, such as the normal postnatal decline in hematocrit. Lesions that cause higher than normal right ventricular (RV) pressure (especially if the pressure is no higher than roughly systemic ; e.g., pulmonary stenosis) usually have little impact on ventricular function in the newborn because the neonatal RV is generally well-adapted to pumping at systemic pressure. Ductal-dependent left-sided obstructive lesions (e.g., severe coarctation of the aorta) may also have only subtle clinical manifestations in the first few days of life if the ductus remains widely patent.

Reduction of Systemic Oxygen Transport

Cardiac lesions may reduce systemic arterial blood O ₂ tension (hypoxemia) or systemic blood flow ( $\dot{Q} s$ $\dot{Q} s$ ) and thus decrease systemic O ₂ transport (systemic O ₂ transport [SOT] = arterial O ₂ content × $\dot{Q} s$ $\dot{Q} s$ ). Resting SOT may be adequate in some patients, but the ability to increase SOT with agitation, fever, or even spontaneous ventilation may be limited, resulting in clinical signs of disease (e.g., poor peripheral perfusion, dyspnea).

Reduction of Systemic O ₂ Transport Due to Hypoxemia

Hypoxemia may be the only manifestation of some lesions (e.g., tetralogy of Fallot), or it may occur with defects that also cause reduced $\dot{Q} s$ $\dot{Q} s$ (e.g., severe Ebstein anomaly). The neonate may tolerate a lower P o ₂ better than older infants or children, and moderate hypoxemia, as an isolated abnormality, generally has little short-term physiologic effect. Increased O ₂ extraction, redistribution of blood flow to organs of high O ₂ need, and perhaps increased $\dot{Q} s$ $\dot{Q} s$ help to maintain systemic $\dot{V} O_{2}$ $\dot{V} O_{2}$ with hypoxemia. ^, For example, when conscious lambs younger than 1 week old were subjected to acute alveolar hypoxia (fraction of inspired O ₂ [Fi o ₂ ] = 0.12; arterial P o ₂ = 35 mm Hg), $\dot{V} O_{2}$ $\dot{V} O_{2}$ did not change and acidosis did not occur. More severe hypoxia (Fi o ₂ <0.10; arterial P o ₂ approximately 25 mm Hg) reduced $\dot{V} O_{2}$ $\dot{V} O_{2}$ but may not be enough to cause significant acidosis in conscious newborn lambs. ^, Multiple variables (e.g., hematocrit, half-saturation O ₂ pressure [P ₅₀ ] of hemoglobin, environmental temperature, $\dot{Q} s$ $\dot{Q} s$ , $\dot{V} O_{2}$ $\dot{V} O_{2}$ , and whether the organism is spontaneously ventilating) determine the physiologic impact of hypoxemia.

The quantitative relationship between hypoxemia and organ dysfunction or damage in the human neonate is unknown. One study found an Fi o ₂ of 0.15 caused $\dot{V} O_{2}$ $\dot{V} O_{2}$ to fall significantly in human infants, albeit without obvious detrimental effect. Perhaps the clearest evidence that neonates can tolerate substantial hypoxemia with modest or no long-term consequences comes from experience treating infants with d-transposition of the great arteries, a lesion that usually causes considerable hypoxemia, especially before balloon septostomy is performed. A total of 129 such neonates were operated on in the first week of life at the Boston Children’s Hospital; the mean lowest preoperative arterial P o ₂ was approximately 24 mm Hg. Even though these infants had undergone a period of substantial hypoxemia (before operation), the hospital survival rate for these patients was more than 98%. Neuropsychologic evaluation and psychiatric assessment at 16 years of age revealed that these patients are at increased neurodevelopmental and psychiatric risk, but it is likely that at least some of the observed decrement in neurologic function was related to aspects of the patient’s treatment and clinical course distinct from the hypoxemia.

Reduction Of Systemic O ₂ Transport Due to Reduction of Systemic Blood Flow

Congenital heart lesions may also decrease SOT by decreasing $\dot{Q} s$ $\dot{Q} s$ . As is the case for hypoxemia, modest reductions of $\dot{Q} s$ $\dot{Q} s$ may not affect tissue oxygenation. For example, Fahey and Lister found that $\dot{Q} s$ $\dot{Q} s$ could be reduced to approximately 42% of the resting value before $\dot{V} O_{2}$ $\dot{V} O_{2}$ fell in conscious 2-week-old lambs; increased fractional extraction of O ₂ from the blood maintained O ₂ delivery. Most congenital cardiac lesions that reduce $\dot{Q} s$ $\dot{Q} s$ do so by one or more of three mechanisms: (1) abnormally low systemic ventricular output, (2) abnormal arterial connection between the heart and peripheral tissues (e.g., interruption of the aortic arch), and (3) recirculation of previously oxygenated blood to the lungs (see later). Only limited quantitative data regarding $\dot{Q} s$ $\dot{Q} s$ in the human neonate with heart disease are available.

Neurohumoral and Other Effects of Abnormal Hemodynamics

Symptoms occur in infants with a heart lesion not attended by a reduction in SOT. For example, infants with CHF resulting from a large VSD generally have $\dot{Q} s$ $\dot{Q} s$ within or near the normal range ^, ^, (but see references ^, ). The symptoms of “CHF” are a consequence of neurohumoral alterations related to the physiologic abnormality, not reduced resting SOT ^, ^, (although limitation in increasing $\dot{Q} s$ $\dot{Q} s$ with stress is likely also important). These alterations help to maintain adequate blood pressure and tissue perfusion in part by augmenting myocardial contractility (sympathetic nervous system activation), ^, increasing systemic vascular resistance (SVR), and activating the sympathetic nervous system, the renin-angiotensin-aldosterone system, ^, ^, and release of vasopressin. Ventricular output is also increased (via the Frank-Starling principle) through fluid retention (and therefore increased ventricular filling) caused by activation of the renin-angiotensin-aldosterone system and other mechanisms. ^, ^,

These factors help to maintain adequate SOT, but they also have unfavorable consequences . For example, vasoconstrictors increase SVR and ventricular afterload; the clinical result is decreased peripheral perfusion. Increased catecholamines increase $\dot{V} O_{2}$ $\dot{V} O_{2}$ . Fluid retention causes edema and increases lung water and the work of breathing. Failure to thrive is the culmination of these and perhaps other physiologic derangements caused by structural defects. It is these secondary manifestations of abnormal cardiovascular function, rather than critically reduced resting SOT per se, that affect most symptomatic infants with cardiac lesions.

Hemodynamics: Vascular Resistance, Flows, and Shunts

Pulmonary Vascular Resistance

The resistance to blood flow through the lungs is an important determinant of the physiologic effects (ventricular work and Sa o ₂ ) of many cardiac lesions with shunting. A useful, albeit simplified, ^, way to conceive of PVR is $PVR = (PAP - LAP) / \dot{Q} p$ $PVR = (PAP - LAP) / \dot{Q} p$ , where PAP is mean pulmonary arterial pressure (mm Hg), LAP is mean left atrial pressure (mm Hg), and $\dot{Q} p$ $\dot{Q} p$ is pulmonary blood flow (which, for pediatric patients, is usually indexed—L/m2/minute). SVR is calculated in an analogous way. Because PVR is primarily a reflection of resistance to flow through the pulmonary microcirculation, it is sometimes referred to as pulmonary arteriolar resistance. The calculated PVR can be low, but total resistance to flow through the lungs and into the heart can be high, with obstruction to systemic ventricular inflow (e.g., mitral stenosis) or poor systemic ventricular compliance. The concept of total pulmonary resistance (total pulmonary resistance = mean $PAP / \dot{Q} p$ $PAP / \dot{Q} p$ ) takes into account all resistance to flow from the central pulmonary arteries to the ventricle, although the term is misleading because some of the resistance to flow resides outside the lungs.

PVR is much higher than SVR in the fetus, ^, but PAP normally decreases to approximately one-half systemic arterial pressure in the first 24 hours of life (presuming the ductus is not widely patent) and reaches essentially mature levels by 1 to 2 weeks after birth. The postnatal decline in PVR is slower with cardiac lesions that maintain elevated PAP (e.g., a large VSD or PDA).

Ratio of Pulmonary to Systemic Blood Flow

The $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ ratio provides an estimate of the volume pumped by the ventricles and an estimate of the $\dot{Q} p$ $\dot{Q} p$ (and hence of PVR if the PAP is known). It can be calculated by using blood oximetry values: $\dot{Q} p / \dot{Q} s . = ({SaO}_{2} - S \bar{V} O_{2}) / ({SpvO}_{2} - {SpaO}_{2})$ $\dot{Q} p / \dot{Q} s . = ({SaO}_{2} - S \bar{V} O_{2}) / ({SpvO}_{2} - {SpaO}_{2})$ , where Sa o ₂ is the systemic arterial O ₂ saturation, $S \bar{V} O_{2}$ $S \bar{V} O_{2}$ is the mixed venous O ₂ saturation, Spv o ₂ is the pulmonary venous O ₂ saturation, and Spa o ₂ is the pulmonary arterial O ₂ saturation. The superior vena caval O ₂ saturation is generally taken as representative of the $S \bar{V} O_{2}$ $S \bar{V} O_{2}$ when there is left-to-right intracardiac shunting. This formula can be used to calculate $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ with any sort of lesion, if the relevant O ₂ saturations can be measured. If there is no right-to-left shunt, Spv o ₂ = Sa o ₂ . This formula cannot be used when multiple sources of pulmonary or systemic flow are present, with each source having a different O ₂ saturation. For example, with right-to-left ductal shunting, Sa o ₂ levels in the ascending aorta and transverse arch differ from those in the descending aorta, making calculating $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ impossible.

Effective Versus Ineffective Blood Flow

Left-to-right shunting occurs when blood that has traversed the lungs is recirculated to them without having first crossed the systemic capillary bed. Such pulmonary flow is termed ineffective, because little additional O ₂ is picked up with more than one pulmonary passage. ( $\dot{Q} p$ $\dot{Q} p$ composed of systemic venous blood is effective pulmonary flow.) Ineffective $\dot{Q} p$ $\dot{Q} p$ represents volume pumped by the heart to no useful end. Similarly, blood that has traversed the lungs before it is pumped to the systemic circulation is effective $\dot{Q} s$ $\dot{Q} s$ , whereas systemic venous blood that enters the systemic arterial circulation without passage through the lungs (right-to-left shunting) is ineffective $\dot{Q} s$ $\dot{Q} s$ . The concepts of effective and ineffective blood flow can help in thinking about complex cardiac physiology, such as that in d-transposition of the great arteries (see later).

Left-To-Right Shunting at the Ventricular or Great Artery Level

Since the ventricle only pumps blood that enters during diastole, left-to-right shunting at the ventricular or great artery level imposes a volume load on the systemic ventricle (LV in an otherwise normal heart). (Actually, with a large VSD, because some pulmonary venous blood enters the RV during diastole, the RV volume pumped is also somewhat increased. ^, )

If the connection between the systemic and pulmonary circulations (e.g., VSD or PDA) is large enough so that the systolic pressure in the two circuits is equal, the communication is said to be unrestrictive . With an unrestrictive defect, the $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ is primarily determined by the ratio of total PVR to SVR . (With a small defect, the pressure difference between the ventricles and the size of the communication are usually the primary determinants of the shunt magnitude.) Factors that affect $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ include the cross-sectional area of the pulmonary microcirculation, all resistances to flow the size of the central pulmonary arteries and large pulmonary veins, any atrioventricular valve dysfunction, and the systemic ventricular compliance. The $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ can vary because systemic and PVR can change with physiologic, pharmacologic, and other perturbations. ^,

In an anatomically normal heart, PAP is a function of $\dot{Q} p$ $\dot{Q} p$ and total pulmonary resistance. However, the systolic PAP will always essentially be the same as the ventricle or artery to which it is connected regardless of the $\dot{Q} p$ $\dot{Q} p$ or the resistance to flow. Therefore, with an unrestrictive VSD or PDA, systolic PAP is essentially the same as aortic pressure, regardless of the magnitude of flow or PVR. In neonates with unrestrictive ventricular or great vessel communications, PVR falls after birth (and a large shunt develops), but the systolic PAP remains elevated. High PVR with unrestrictive communications is reflected by a low $\dot{Q} p / \dot{Q} s$ $\dot{Q} p / \dot{Q} s$ , not the pulmonary arterial systolic pressure. (But pulmonary arterial diastolic pressure is a reflection of PVR with a large VSD; i.e., relatively low diastolic PAP generally implies low PVR.) Thus pulmonary arterial systolic hypertension and increased PVR are related but distinct concepts.

Other factors besides PVR and SVR also influence cardiovascular physiology with shunting lesions. For example, Jarmakani and colleagues demonstrated that infants younger than 2 years old with a PDA had greater LV end-diastolic pressure and end-diastolic wall stress than patients with a VSD and comparable magnitude of shunt. The reason for this is unclear, but it may be related to differences in LV stroke work with the two lesions. Diastolic runoff into the pulmonary arteries with connections between the aorta and pulmonary artery may decrease aortic diastolic pressure and hence reduce coronary perfusion. Streaming of blood can affect whether pulmonary venous blood ends up in the aorta or in the pulmonary artery. For example, with double-outlet RV, the location of the VSD (subpulmonary vs. subaortic) will influence the effective $\dot{Q} p$ $\dot{Q} p$ (with a subpulmonary VSD, pulmonary venous blood tends to be recirculated to the lungs rather than into the aorta). Left-to-right shunting from the LV to the right atrium can occur, usually through a defect in the tricuspid valve. The magnitude of an LV-to-RA shunt is primarily determined by the effective size of the LV–right atrial communication; PVR and right atrial pressure are largely irrelevant.

Left-To-Right Atrial Shunting

Left-to-right shunts across an atrial septal opening increase the volume of blood pumped by the pulmonary ventricle. With a small defect, left atrial pressure is usually substantially different from the right atrium, and shunting is primarily the result of this pressure difference. With a large defect (where the mean atrial pressures are essentially equal), the shunt magnitude is generally ascribed to the relative compliances of the left atrium–pulmonary veins–LV and the RA–vena cava–RV: blood tends to flow into the venous chamber or ventricle that most readily accepts it ( Fig. 49.1 ). Hence, left-to-right shunting occurs in older patients, because the RV is thinner than the LV and more compliant. Because substantial atrial shunting can occur in the neonate, ^, whose RV and LV have similar compliances, Rudolph proposed that such shunting may be a function of differential ejection of blood from the two ventricles, given that RV afterload is less than that of the LV. Left-to-right atrial shunting also occurs with partial anomalous pulmonary venous connection to the right side of the circulation. Assuming no defect in the atrial septum, the shunt is determined by the fraction of $\dot{Q} p$ $\dot{Q} p$ going to the lobes of the lungs anomalously connected, which is a function of the total resistance to blood flow through those lobes relative to the normally connected lobes. Atrial level shunts in children —even if the $\dot{Q} p$ $\dot{Q} p$ is very large—are seldom associated with significant pulmonary hypertension. However, in the neonate , the capacity for flow-related dilation or recruitment in the pulmonary vascular bed is limited, and large atrial level shunts (e.g., unobstructed total anomalous pulmonary venous connection, and intracranial arteriovenous malformations) are often accompanied by considerably increased PAP.

Fig. 49.1, Factors that influence where blood goes with structural heart lesions. (A) General scheme of box diagrams. (B) Blood moves from high pressure to low. Shunting at the atrial, ventricular, and great vessel levels can result from a pressure gradient between communicating structures if the communication is small enough to be pressure restrictive. (C) When two vascular beds differing in resistance to flow are connected to a source of flow, more blood finds its way into the lower-resistance circuit than into the higher-resistance circuit. With an unrestrictive ventricular septal defect and low pulmonary vascular resistance, systolic pressures are essentially the same in both the aorta and the pulmonary artery, yet there is more flow to the pulmonary artery. (D) When two chambers differing in resistance to filling (compliance) are connected by a large defect, more blood finds its way into the chamber with the greater compliance. (E) Streaming can influence the chamber or vessel to which the blood flows. As depicted here, in the fetus umbilical venous blood preferentially crosses the foramen ovale, and therefore the most highly oxygenated blood tends to go to the organs of greatest need. Streaming can also take place at the ventricular and great vessel levels. AO, Aorta; CS, coronary sinus; DV, ductus venosus; IVC, inferior vena cava; LA, left atrium; LHV, left hepatic vein; LV, left ventricle; PA, pulmonary artery; PS, portal sinus; RA, right atrium; RHV, right hepatic vein; RV, right ventricle; SVC, superior vena cava; UV, umbilical vein. Numbers in triangles indicate pressure.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here