Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

PART ONE – Basic Principles in Pediatric Anesthesia

Chapter 3 – Cardiovascular Physiology in Infants and Children

Maureen A. Strafford

  

 

Fetal and Transitional Circulation, 70

  

 

Fetal Circulation, 70

  

 

Oxygen Delivery in the Fetus and Neonate,71

  

 

Transitional Circulation, 73

  

 

Basic Principles of Fetal and Neonatal Cardiac Function, 74

  

 

Preload, 74

  

 

Afterload, 76

  

 

Contractility, 77

  

 

Interrelationships Between Preload, Afterload, and Contractility,79

  

 

Effect of Heart Rate on Cardiac Performance, 81

  

 

Developmental Aspects of Myocardial Function, 82

  

 

Developmental Electrophysiology, 85

  

 

Interaction of the Nervous System and the Heart, 85

  

 

Effects of Congenital Heart Disease on Cardiopulmonary Development,87

  

 

Assessment of the Cardiovascular System, 88

  

 

History and Physical Examination, 88

  

 

Special Evaluation of the Cardiovascular System,88

  

 

Effects of Anesthesia on the Cardiovascular System, 97

  

 

Premedication and the Induction of Anesthesia,98

  

 

Maintenance of Anesthesia, 99

  

 

Effects of Cardiopulmonary Interactions, 104

  

 

Clinical Considerations and New Treatment Modalities,104

  

 

Summary, 105

The clinical challenge of pediatric anesthesia is directly related to the ongoing development and maturation of multiple organs, especially the cardiopulmonary system, and the intricate interaction of anesthetic and surgical manipulation with these developing systems. Each child presents a challenge. Ironically, the dynamic nature of the cardiovascular system may make pediatric anesthesia appear confusing and complex to the anesthesiologist is inexperienced with children.

With respect to cardiovascular physiology, it is well recognized that the neonate is not just a small infant and that the infant differs from the older child and adolescent. To be safe and effective, an appropriate anesthetic plan must incorporate these vital differences. In addition, the anesthesiologist must be aware that when underlying congenital heart disease is present, cardiopulmonary development continues but may be adversely affected by the ongoing pathophysiology. The patient who undergoes surgical repair of congenital heart defects does not necessarily return to normal cardiac function despite an excellent surgical repair, because the preoperative cardiopulmonary effects of disease are often irreversible. Because cardiac function and pharmacologic responses are so closely tied to developmental changes, this discussion of cardiovascular physiology stresses the basic principles of cardiovascular physiology in the context of ongoing maturational changes. With a clear understanding of these developmental concepts, the anesthesiologist is able to formulate an intelligent and careful plan for a pediatric patient of any age, fully realizing the clinical challenge and rewards of caring for these children.

Maturation of the cardiovascular system is not complete at birth. Dynamic changes of the cardiovascular system occur throughout fetal development and continue into the neonatal period and infancy. The healthy school-aged child possesses a cardiovascular system that functions like that of a healthy young adult but continues to structurally mature as the child grows into adolescence. It is appropriate to begin with a discussion of fetal and neonatal circulation and the basic principles of cardiac function as an introduction to the pediatric cardiovascular system.

▪ FETAL AND TRANSITIONAL CIRCULATION

The cardiovascular system exists to efficiently deliver oxygen and other metabolic nutrients to tissues throughout the body. The needs of the fetus differ from those of the neonate, as do the means of meeting those needs. In the fetus, gas exchange occurs at the placenta, a unique organ that receives blood flow from both the mother and the fetus. Because the lungs require only nutrient flow, not the entire cardiac output (CO) for respiratory exchange, fetal intracardiac and extracardiac shunts exist to minimize flow to the lungs while simultaneously maximizing the appropriate delivery of oxygen to all organ systems. At birth, the placenta is removed and the neonate must exchange gas in the lungs. Fetal shunts are no longer needed and must close to permit the efficient transition to a neonatal circulation.Transitional circulation describes the changes observed as the fetus adjusts to the circulatory changes after birth and establishes a neonatal circulation. The presence and persistence of the transitional circulation have important implications for the cardiac function of the neonate, whether the child is normal, is preterm, is critically ill, or has congenital heart disease.

Our understanding of the fetal, transitional, and neonatal cardiovascular circulations is derived from the pioneering research of Dawes and Rudolph in their fetal and neonatal lamb studies ( Dawes, 1968 ;Rudolph, 1974a ). Despite obvious species differences, these data have been extrapolated to the human species, and clinical experience continues to confirm their validity.

▪ FETAL CIRCULATION

The chorionic villus is the functional unit of the placenta. Deoxygenated blood is pumped down the fetal descending aorta to the umbilical artery and then to the placenta. The umbilical artery branches and eventually gives rise to an intricate system of arterioles, capillaries, and venules in the intervillous spaces of the placenta, where oxygen and nutrient exchange occurs. Oxygenated blood returns to the fetus via the umbilical vein for delivery to all organ systems. Not only do intracardiac and extracardiac shunts permit the lungs to be bypassed but also more highly oxygenated blood flow is delivered to organs with higher metabolic needs, such as the brain and heart ( Fig 3-1 ). A review of the anatomy and physiology of fetal circulation illustrates how the fetus with congenital heart disease, even with severe structural cardiac abnormalities, can survive in utero and exhibit clinical difficulties only when the transition from fetal to neonatal circulation is under way.

 
 

FIGURE 3-1  Fetal circulation. Ao, aorta; DA, ductus arteriosus; DV, ductus venosus; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium; RV, right ventricle.  (From Rudolph AM: Changes in the circulation after birth. In Rudolph AM, editor:Congenital diseases of the heart. Chicago, 1974, Year Book Medical.)

 

 

 

Because of the presence of shunts, fetal organs receive a mixed blood supply from either the right ventricle (RV) or the left ventricle (LV), often referred to as parallel circulation. In contrast, in adult circulation the RV and LV are in series (a series circulation), resulting in equal but separate outputs for each ventricle. The foramen ovale, ductus arteriosus, and ductus venosus are the fetal shunts needed for effective fetal circulation that must close after birth. Blood returning from the placenta in the umbilical vein has the highest oxygen content ( Fig 3-2A ). The first shunt encountered is the ductus venosus, which shunts half of this richly oxygenated umbilical blood flow away from the liver, directly to the heart. Highly oxygenated ductus venosus blood mixes with inferior vena cava blood, but preferential streaming directs blood with more oxygen across the foramen ovale (an intracardiac shunt) into the left atrium (LA). As a result, highly oxygenated blood travels to the LV and is ejected into the aorta, thereby feeding the coronary arteries and arteries to the brain with the most oxygenated blood. Superior vena cava and hepatic venous blood flows to the RV directly, but because of the high resistance in the pulmonary vascular bed and relatively low systemic resistance due to the placental vasculature, right ventricular output is shunted away from the lungs via the ductus arteriosus to enter the descending aorta. The shunted blood entering the descending aorta returns to the placenta or perfuses the lower body. Only 5% to 10% of the combined ventricular output of both ventricles is circulated directly to the lungs.

 
 

FIGURE 3-2  Hemodynamics of the fetus and neonate. Circled values represent oxygen saturation. Systolic, diastolic, and mean (m) pressures appear near their respective chambers and vessels. (A) Fetal circulation near term. (B) Transitional circulation less than 1 day after birth. (C) Neonatal circulation several days after birth. Ao, aorta; DA, ductus arteriosus; DV, ductus venosus; IVC, inferior vena cava; LA, left atrium; LV, left ventricle; PA, pulmonary artery; PV, pulmonary vein; RA, right atrium; RV, right ventricle; SVC, superior vena cava.  (From Rudolph AM: Changes in the circulation after birth. In Rudolph AM, editor: Congenital diseases of the heart. Chicago, 1974a, Year Book Medical.)

 

 

 

Fetal CO is often described as combined ventricular output, meaning the sum of the output of both ventricles. However, each ventricle has a different afterload, against which it pumps. The RV ejects against the low-resistance ductus arteriosus, the low-resistance placenta, and the high-resistance lungs and lower body. The left ventricular afterload consists of the high resistance found in the brain, upper body, and aortic isthmus. The fetal RV has a slightly higher output because its afterload is significantly lower than that of the LV. The LV pumps into the much larger vascular bed of the brain, so in the end, the outputs of the RV and LV are almost equal. Studies by Rudolph (1974a, 1974b) [157] [158] suggested a ratio of RV to LV output equal to 2:1 in the fetal lamb. Echocardiographic evidence supports this concept of equal size and output as the ventricles develop in utero and have an RV/LV output ratio closer to 1.3:1 ( St. John Sutton et al., 1984 ).

▪ OXYGEN DELIVERY IN THE FETUS AND NEONATE

The cardiovascular system must efficiently deliver oxygen and other metabolic substrates to all tissues and organs. Nevertheless, the fetus functions and develops within a relatively hypoxemic environment and adequate oxygen delivery must depend on other compensatory mechanisms. Umbilical venous blood has an oxygen tension of about 35 mm Hg. The fetus has high concentrations (≈80% at term) of fetal hemoglobin (HbF), which binds oxygen much more efficiently than the adult form of hemoglobin (HbA). Oxygen competes with 2,3-diphosphoglycerate (2,3-DPG) for binding on the hemoglobin molecule. Levels of 2,3-DPG are lower in the fetus than in the adult. In addition, HbF has a low affinity for 2,3-DPG compared with HbA and therefore HbF preferentially binds oxygen. The P50 (oxygen tension at which 50% of the blood is saturated with oxygen) occurs at a lower PO2 in the fetus as a result of HbF and its unique binding characteristics (see Fig. 2-45 ). Oxygen saturation of hemoglobin in umbilical venous blood is well maintained, often greater than 90%.

A review of fetal anatomy and physiology shows how a fetus with congenital heart disease may have little difficulty in utero. The fetal intracardiac and extracardiac shunts permit blood flow to the placenta and other vital organs despite the presence of severe underlying abnormalities. Hypoplastic left heart syndrome (HLHS), which is associated with severe hypoplasia of the LV, atresia of the mitral and aortic valve, and hypoplasia of the aortic arch, is incompatible with neonatal survival, but the fetus with HLHS has a viable circulation in utero. A neonate with HLHS has no way for blood to flow from the LA to the LV and therefore left atrial blood return is shunted completely via the foramen ovale to the RA, RV, and ductus arteriosus.

Retrograde flow from the ductus arteriosus perfuses the ascending aorta and coronary arteries. Intracardiac shunts, such as the foramen ovale and ductus arteriosus, remain essential to continued survival after birth in the neonate with HLHS. When the transition from fetal to neonatal circulation begins, with closure of the fetal shunts, the underlying congenital heart defect is unmasked, often because systemic blood flow (as in HLHS) depends on the presence of a ductus arteriosus, foramen ovale, or both. In contrast, dysrhythmias or conduction abnormalities, such as congenital complete heart block, often cause severe fetal problems such as hydrops and may contribute to fetal death. Cardiac function is limited in utero, and output is sensitive to changes in volume and heart rate (HR) changes (see later discussion). Because of these functional limitations, the fetus cannot tolerate bradycardia or persistent dysrhythmias in utero yet can survive severe structural abnormalities.

▪ TRANSITIONAL CIRCULATION

At birth, various factors influence the change from a fetal circulation to the neonatal pattern (see Fig. 3-2B ). When the placenta separates, with clamping of the umbilical vessels, and ventilation and pulmonary blood flow begin, profound effects are produced on fetal circulatory anatomy and physiology. As described, the circulation changes from a parallel to a series system, and intracardiac and extracardiac shunts close. The placenta ceases to function as the organ of respiration, and the lungs begin respiration in a gaseous rather than a liquid environment. The most dramatic changes at birth are related to changes in resistance throughout the circulatory system.The low-resistance placenta is excluded, as the umbilical cord vessels are clamped and systemic vascular resistance (SVR) increases. With the initiation of breathing, pulmonary vascular resistance (PVR) begins to decrease as pulmonary blood flow increases.

PVR falls dramatically at birth. More gradual decreases in PVR, pulmonary artery pressure, and pulmonary blood flow occur over the first few weeks of life ( Fig 3-3A ). These more gradual declines reflect the remodeling of the pulmonary vascular musculature typical of this period. At birth, the increased pulmonary blood flow increases left atrial pressure, and the foramen ovale, a flaplike one-way valve, closes in response to these higher pressures relative to right atrial pressure. The foramen ovale can open in response to changing physiologic conditions. Fifty percent of children under age 5 and 25% of adults demonstrate probe patency of the foramen ovale ( Scammon and Norris, 1918 ). Functional closure of the ductus arteriosus occurs with removal of the placenta and a consequent decrease in the levels of circulating prostaglandins.

 
 

FIGURE 3-3  (A) Graphic representation of the changes in pulmonary artery pressure, pulmonary blood flow, and pulmonary vascular resistance that occur in the perinatal period. Pulmonary vascular resistance decreases during the latter part of gestation, mainly because of an increase in the number of pulmonary vessels associated with growth; it falls dramatically at birth because of the vasodilator effect of ventilating the lungs with air; a further decrease occurs as pulmonary vascular smooth muscle regresses. Pulmonary blood flow increases slightly during fetal growth and then increases dramatically after birth. Pulmonary arterial pressure falls rapidly immediately after birth and then more gradually to reach adult levels after 6 to 8 weeks. (B) Change in pulmonary vascular resistance (PVR) with changes in Po2and arterial pH.  (A) (From Rudolph AM: Changes in the circulation after birth. In Rudolph AM, editor: Congenital diseases of the heart. Chicago, 1974, Year Book Medical.) (B) (From Rudolph AM, Yuan S: Response to the pulmonary circulation to hypoxia and H+ ion concentration changes. J Clin Invest 45:399-411, 1966.)

 

 

 

In utero prostaglandins contribute to the patency of the ductus arteriosus, and the increasing oxygen tension after ventilation serves as a potent stimulus for ductal closure. Final anatomic closure results from thrombosis and fibrosis over the first few months of life, although the precise mechanisms of closure are not well elucidated (see Fig. 3-2C ). Because these shunts are not anatomically closed immediately after birth, certain clinical conditions may contribute to either the persistence of or a return to a fetal circulation. Infants who are preterm or critically ill with a either medical (e.g., sepsis or meconiumaspiration) or surgical (e.g., tracheoesophageal fistula or omphalocele) crisis are at high risk for persistent fetal shunting. Hypoxemia and acidosis are two main factors known to reverse shunt patency. Neonates with these problems are at considerable risk for persistence of the fetal circulation (PFC), a clinical syndrome observed in neonates when fetal shunting persists beyond the normal transition period in the absence of structural congenital heart disease ( Gersony et al., 1969 ). The presence of hypoxemia, acidosis, or both can be potent stimuli for maintaining high PVR in the neonate. Persistent fetal shunting (see Fig. 3-3B ), PFC, or persistent pulmonary hypertension of the newborn (PPHN), illustrates the important interactions between pulmonary and cardiac development at the neonatal stage.

The process by which the vasoconstricted pulmonary vascular bed relaxes at birth is complex and not completely understood. The process most likely involves a cascade of humoral, vasoactive mediators that cause dynamic changes in PVR. PPHN can also result when there are structural abnormalities in the pulmonary vascular bed. For example, an intrauterine insult may result in maladaptation of the pulmonary vasculature, resulting in abnormal muscularization of normally nonmuscularized peripheral arteries and medial hypertrophy of the muscular arteries. At birth, this infant may present with signs or symptoms of PPHN. Some infants with severe meconium aspiration have been found to have these postmortem findings in the pulmonary vasculature, raising the possibility of a developmental insult resulting in these changes. The pathophysiology of neonatal pulmonary hypertension has been elucidated extensively in the past decades, but it is only since the 1990s that research has begun to explain the recovery process and its potential long-term effects on the pulmonary vasculature. In a well-studied animal model of chronic hypobaric hypoxia that reflects neonatal pulmonary hypertension in humans, the remodeling of the pulmonary arteries during recovery from pulmonary hypertension induced by neonatal hypoxia was studied ( Hall et al., 2004 ). In neonatal pulmonary hypertension, the thickening of the pulmonary arteries is due to smooth muscle cell hyperplasia, hypertrophy, smooth muscle cell recruitment from the adventitia, and increased deposition of matrix proteins ( Tulloh et al., 1997 ; Haworth and Hislop, 2003 ).

Recovery from neonatal pulmonary hypertension results in thinning of the elastic and muscular arteries, but the mechanisms of recovery are different in these two types of arteries and are age dependent. The study of recovery from abnormal fetal and neonatal cardiovascular physiology has also raised the challenging concept that abnormalities in this early period of life may have dramatic implications for adult cardiovascular risk. Both fetal hypoxia and fetal undernutrition may be linked to adult cardiovascular disease. In 1993, Barker and others ( Barker, 1993 ; Barker and Fall, 1993 ) hypothesized that any alteration of the in utero environment that has the potential for alterations in fetal development may also irreversibly impair physiologic function, increasing susceptibility to disease in adulthood.

Jones and others (2004) demonstrated in the laboratory that perinatal exposure to chronic hypoxia had multiple and serious consequences in the adult, which included reduced body weight, right and left ventricular hypertrophy, reduced pulmonary arterial compliance, and alterations in pulmonary artery compliance.

While maladaptation of the pulmonary vasculature is one problem, maldevelopment of the pulmonary vasculature can also be seen in newborns. For example, infants with diaphragmatic hernia have underdevelopment of the lung parenchyma or pulmonary hypoplasia. These infants have a reduction in alveoli and pulmonary arterial vessels, but the arterial vessels present have an increased and abnormal muscularity and do not dilate normally after birth.

Figure 3-4 illustrates the normal arterial dilatation during transition from fetal to neonatal circulation. The initial change seen postnatally is preacinar dilatation. When there is underdevelopment of the lung, such as in diaphragmatic hernia, the vascular bed is hypoplastic and abnormally muscular and cannot dilate normally after birth. When the pulmonary vascular bed is maldeveloped, the vascular bed also has increased muscular development. Finally, if there is maladaptation, such as in neonates with medical or surgical emergencies or congenital heart disease, and resulting hypoxemia, acidosis, or both, normal dilatation does not occur. In these neonates, severe hypoxemia and circulatory compromise may occur and may first appear during the stress of surgery and anesthesia.

 
 

FIGURE 3-4  Schematic representation of normal arterial dilatation during transition from fetal to neonatal circulation. When the lung is underdeveloped, the vascular bed is abnormally muscular; when it is maladapted, it has not dilated appropriately at birth.  (From Rabinovitch M: Pulmonary hypertension. In Adams FH, Emmanouilides GC, Riemenschnieder T, editors: Moss—heart disease in infants, children, and adolescents, 4th ed. Baltimore, 1989, Williams & Wilkins.)

 

 

 

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ BASIC PRINCIPLES OF FETAL AND NEONATAL CARDIAC FUNCTION

Fetal circulatory anatomy has many clever design features, and the developing heart has significant functional limitations that continue into the neonatal period. In any discussion of cardiovascular physiology, several essential physiologic concepts must be understood: preload, afterload, contractility, SV, and CO. CO in both the fetus and the adult is directly proportional to HR, preload (ventricular distention), and contractility but is inversely related to afterload (combined resistances of blood, ventricular mass, and vascular beds).

An understanding of basic cardiac function is derived from studies examining the mechanical and contractile properties of isolated cardiac muscle fiber and the basic unit of contraction, the sarcomere. The heart is a muscle that lengthens and shortens against a load. It pumps blood from the low-pressure venous system to the high-pressure arterial vascular bed. How well the heart performs as a pump is measured by cardiac output (CO), or the volume of blood ejected per minute by the LV. Stroke volume is the volume of blood ejected during each contraction, and CO is the product of HR multiplied by SV. The SV, however, is determined by the loading conditions of the heart (preload and afterload) and the intrinsic contractility of the heart.

Intact heart function is influenced by multiple other factors, including geometry of fiber orientation, ventricular geometry, interdependence of right and left ventricular function, pericardial and pulmonary interactions, and the vascular beds and their various resistances. Finally, the immaturity of the fetal and neonatal circulatory system has important implications for cardiac function. This discussion focuses on these important concepts, with reviews of studies from isolated cardiac muscle and intact heart preparations and comparative studies that describe fetal, neonatal, and adult myocardial function. In this way, the potent effects of anesthetics on the circulatory system, especially at critical developmental stages such as the newborn period, can be better understood and anesthetic plans safely formulated.

▪ PRELOAD

In cardiovascular physiology isolated muscle studies, a strip of isolated cardiac muscle is stretched and fixed at both ends in a physiologic solution ( Fig 3-5 ). An electrical stimulation is delivered, causing an isometric contraction. Because the muscle is fixed on both ends, no fiber shortening occurs, but tension does increase. This tension varies directly with the amount of stretch put on the muscle before stimulation. The initial amount of stretch is defined as preload. Because different amounts of stretch can be applied to the muscle, different amounts of tension are developed. The point at which the maximal tension is developed is called Lmax. The curve plotted using different stretch values (preload) against the tension developed is called the length-active tension curve ( Fig 3-6 ). As sarcomeres are stretched up to a finite length, the force of contraction (tension) increases. Tension is the force generated along a line. Stress is the force exerted across a cross-sectional area or surface. Pressure is the force generated within a cavity. When we apply the concept of preload to the intact heart, wall stress and intracavitary pressure play important roles. Because the LV is conceptualized as a thick-walled cylinder, the law of La Place states that

 
 

FIGURE 3-5  Study of length-tension relationships in isolated cardiac muscle strips.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart, 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

 
 

FIGURE 3-6  Length-active tension curve showing the relationship between length and tension in isolated cat papillary muscle. Note the effect of increased contractility due to increased calcium ion concentration. Neither the relation between muscle length and rising tension nor Lmax is altered. However, developed tension at any given muscle length is increased at the higher calcium concentration.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

3.1  

where P is the pressure within the cylinder, R is the radius of curvature, and h is the thickness of the ventricular wall.

Despite the complex geometry of the LV, wall stress is considered uniform throughout. As a result, wall stress is said to increase as ventricular chamber volume increases even when intraventricular pressure is constant. In addition, if wall thickness increases (i.e., the total number of muscle fibers increases), the tension on any single muscle fiber is decreased, because it is distributed among more muscle fibers.

In the intact heart, preload is the diastolic wall stress resulting from a certain volume of blood distending the ventricle. The preload is the load present before contraction has started, at the end of diastole. The preload is a reflection of venous filling pressure that fills the RA or LA, which then fills the RV or LV during diastole. The relationship between end-diastolic volume (EDV) and developed pressure in the intact heart produces a curve similar to the length-tension curve defined in isolated muscle fiber preparations ( Fig 3-7 ). These curves appear similar to those shown in Figure 3-6 in isolated papillary muscle studies. Independent observations by Frank and Starling correlated the isolated muscle fiber results with intact heart observations ( Frank, 1895 ; Starling, 1918 ). Frank's observations that the “length and tension changes in skeletal muscle correspond to changes in volume and pressure in the heart” and Starling's description that “the mechanical energy set free on passage from the resting to the contracted state is a function of the length of the muscle fiber” form the basis for the well-known Frank-Starling mechanism. When the preload increases, the LV distends during diastole and the SV rises according to Starling's law. This mechanism and the idealized ventricular function curves demonstrating the mechanism ( Fig 3-8 ) show that cardiac performance, represented by SV or CO, is related to preload, ventricular EDV, or pressure. In clinical practice, ventricular end-diastolic pressure (EDP) is often used as a measure of preload. However, compliance or “stiffness” of the ventricle after hypertrophy, scarring, or with aortic regurgitation may alter this response.

 
 

FIGURE 3-7  Relationship between midwall circumferential fiber length and maximal developed force obtained in the isovolumetrically beating left ventricle of the isolated canine heart. Note the similarity to Figure 3-6 .  (From Weber KT, Janicki JS: Am J Cardiol 40:740, 1977.)

 

 

 

 
 

FIGURE 3-8  Frank-Starling mechanism. Cardiac performance, represented by stroke volume or cardiac output, is related to some estimation of preload, ventricular end-diastolic volume, or pressure.  (From Friedman WF, George BL: Treatment of congestive heart failure by altering loading conditions of the heart. J Pediatr 106:697, 1985.)

 

 

 

Why is preload important in regulating cardiac function? Preload acts as a functional reserve. Preload reserve, or the ability to increase SV with an increase in EDV, has important implications for the circulatory system during stress. Lee and others (1986) have shown a 9% increase in EDV coupled with a 13% increase in SV in resting dogs during volume loading. Cardiac muscle has a high degree of resting tension at all lengths, especially compared with skeletal muscle ( Fig 3-9 ). Because of the collagen network in which myocardial cells dwell in the intact heart, cardiac muscle operates on the ascending limb of the length-tension curve.

 
 

FIGURE 3-9  Comparison of total and active length-tension curves in skeletal (A) and cardiac (B) muscle allowed to contract under isometric conditions. Active tension, which is the tension developed during contraction, equals total tension after stimulation minus the tension recorded in the resting muscle before stimulation. Although the active length-tension curves are similar for the two muscle types, the resting tension in the cardiac muscle is much higher and, unlike skeletal muscle, is significant below Lmax.  (From Katz A:Physiology of the heart. New York, 1977, Raven Press.)

 

 

 

▪ AFTERLOAD

Afterload has been studied in isolated muscle preparations as well as in the intact heart. In isolated muscle preparations, afterload is the weight or load that the contracting muscle must overcome to shorten. In single fiber studies, one end of the muscle strip is attached to a transducer while the other end is attached to an isotonic lever ( Fig 3-10 ). On the opposite end of the lever, a weight is attached, stretching the muscle to its resting length. At this point the resting tension or preload can be determined. Further stretch on the muscle is prevented by placing a stop where the muscle is attached to the lever. Then weight is added. After a stimulation, the muscle contracts until it develops sufficient tension to overcome the additional imposed weight. This tension is the afterload. Preload in the intact heart is defined as the wall stress at the end of diastole, whereas afterload is the wall stress experienced by the intact heart during ventricular ejection. Because ventricular ejection is a dynamic event, afterload is more complicated conceptually.

 
 

FIGURE 3-10  Diagrammatic representation of an isotonic lever system.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

Cardiac function is inversely related to afterload and directly related to preload. Afterload plays a critical role in cardiovascular regulation as summarized in Figure 3-11 . In intact hearts, arterial pressure, resistance, or impedance is often used as a measure of afterload. As with preload, extrapolation from single-fiber studies to the intact heart is difficult because of asymmetric ventricular geometry and the process of ventricular ejection. Afterload is often related to vascular resistance even though resistance is a measure of the opposition to flow in a nonpulsatile system. Impedance is a more accurate measure of opposition to flow in a pulsatile system, but it is much more difficult to measure. Resistance is measured as the pressure differenceacross the circulation divided by mean aortic flow, or CO, as follows:

 
 

FIGURE 3-11  Interactions between the various components of cardiac activity. Solid lines indicate an increasing effect. Broken line represents a depressing effect. LV, left ventricle.  (From Braunwald E: Regulation of the circulation. N Engl J Med 290:1124, 1974. © 1974 Massuchusetts Medical Society. All rights reserved.)

 

 

 

3.2  

where R is vascular resistance, P is mean pressure change across the arterial circuit, Q is mean flow or CO, μ is blood viscosity, L is length of the arterial system, and r is vessel radius.

Equating systolic arterial pressure with SVR is potentially erroneous, as Equation 3.2 demonstrates. Arterial blood pressure (BP) may remain constant in the face of opposite changes in CO and resistance, and profound changes in afterload would not be suspected.

Because the amount of blood ejected by the ventricle, the CO, is determined largely by afterload, changes in afterload affect performance in important ways. Increased afterload causes a reciprocal decline in the extent and velocity of fiber shortening and therefore the volume of blood ejected. However, this assumes the other determinants of output (HR, preload, contractility) are not changing. In addition, other factors, such as the neurohumoral control of vascular constriction, play important roles.

▪ CONTRACTILITY

Contractility is a critical factor in cardiac performance. Multiple forces influence the contractile state of the myocardium, as summarized schematically in Figure 3-12 . Isolated muscle studies show that if preload and afterload are held constant, the rate at which tension develops can nevertheless be increased in the presence of calcium or drugs such as norepinephrine ( Fig 3-13 ). With increased calcium, the intrinsic performance of the cardiac muscle is enhanced. However, contractility assessment is much more complex and difficult to measure in the intact heart. Because preload and afterload dynamically influence contractility, an assessment of contractility should be load independent. Obtaining a load-independent measurement remains controversial. The validity of contractility measurements must always be criticallyexamined and understood, especially when the effects of anesthetics on cardiac function and contractility are described with echocardiographic measurements. This issue is reviewed in greater detail later.

 
 

FIGURE 3-12  Major influences that elevate or depress the contractile state of the myocardium. (Bottom left) Effect of alterations in the contractile state of the myocardium on the level of ventricular performance at any given level of ventricular end-diastolic volume (E.D.V.). (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

 
 

FIGURE 3-13  Effect of increased contractility due to norepinephrine (NE) administration on the force-velocity relation of the cat papillary muscle.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

▪ INTERRELATIONSHIPS BETWEEN PRELOAD, AFTERLOAD, AND CONTRACTILITY

Realistic understanding of preload, afterload, and contractility and of how these factors affect cardiac performance must take into account the interrelationships between these forces in the intact heart. When the relationship of pressure and volume in the intact heart is schematically described, a pressure-volume loop is drawn. This loop demonstrates the phases of contraction and relaxation in the intact heart (Fig 3-14 ). When a measurable pressure and volume are reached at point A, this is a determination of preload. The LV begins to contract and, as the pressure increases, the mitral valve closes. Before the aortic valve opens, there is the period of isovolumetric contraction. Once ventricular pressure exceeds aortic pressure, the aortic valve opens and ejection begins (point B = afterload). The distance between points B and C is a representation of SV. As the ventricle ejects, the force on the wall of the LV, or afterload, actually decreases. The law of La Place (wall stress = P ×R/h) states that wall stress decreases as the ventricular chamber size decreases. Point C is reached when the muscle cannot shorten any farther; then the aortic valve closure occurs. The interval of point C to point D represents isovolumetric relaxation coincident with the sharp drop in ventricular pressure. Once ventricular pressure drops below left atrial pressure, the mitral valve opens and ventricular filling begins again, generating the preload for the next ejection.

 
 

FIGURE 3-14  Schematic representation of the pressure-volume relation in the intact heart (see text for description). LV, left ventricular; SV, stroke volume.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

These pressure-volume loops become extremely helpful in describing how changes in one determinant directly affect SV. When afterload and contractility are held constant, changes in preload are described in the following pressure-volume loop ( Fig 3-15 ). Points A, E, and F represent different ventricular EDVs or preloads. SV is increased (SV2 > SV1) with increasing preloads as described by the Frank-Starling mechanism.

 
 

FIGURE 3-15  Schematic pressure-volume loops demonstrating the effect of increasing preload on stroke volume (SV) during normal contraction when afterload and contractility are held constant. LV, left ventricular.  (From Braunwald E, Ross J, Sonnenblick EH:Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

If afterload is increased while preload and contractility remain the same, a different loop results ( Fig 3-16 ). Points A, B, and C represent increasing afterloads. SV decreases as afterload increases (SV A to F > SV B to E > SV C to D).

 
 

FIGURE 3-16  Schematic pressure-volume loops demonstrating the effect of increasing afterload on stroke volume during normal contraction when preload and contractility are held constant. As afterload is increased, stroke volume is diminished. The points D, E, and F at end ejection describe a line known as the end-systolic pressure-volume relation, the slope of which is used as a load-independent index of contractility. LV, left ventricular.  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

These loops serve to illustrate the relationships of preload, afterload, and contractility in the intact heart. However, preload and afterload do not always change independently, as described earlier ( Fig 3-17). Loop 1 schematically represents a normal EDV being reached and the subsequent ejection of a normal SV. When the heart must contract against an increased afterload, loop 2 is described and a decrease in SV is noted. However, theheart attempts to compensate for the decrease in SV by contracting at an increased EDV but with less afterload. The result is an SV very close to normal (loop 1).

 
 

FIGURE 3-17  Response of the normal heart to an increase in afterload during function in the physiologic range of the pressure-volume relation (see text for full description). LV, left ventricular; SV, stroke volume.  (From Strobeck JE and Sonnenblick EH: Myocardial contractile properties and ventricular performance. In Fozzard HA, et al., editors: The heart and cardiovascular system. New York, 1986, Raven Press.)

 

 

 

In each of these loops, a line is described known as the end-systolic pressure-volume curve. End-systolic volume (ESV) is linearly related to end-systolic ventricular pressure ( Suga and Sagawa, 1974 ;Little et al., 1985 ). The slope of the end-systolic pressure-volume curve varies as a function of contractility but is independent of loading parameters. This relationship is not true when regional myocardial wall abnormalities exist. However, this curve has become an important basis for the noninvasive measurement of contractility in the intact heart. Borow and Grossman (1984) used M-mode echocardiography, noninvasive systemic arterial BP, and indirect carotid pulse recordings to study the changes in contractility in examining the end-systolic pressure-volume relationship. Figure 3-18 shows the effects of changing contractility where loop 1 normal represents normal contractility, loop 2 describes increased contractility, and loop 3 describes a condition in which contractility is decreased. A larger EDV in this depressed contractility state does not result in an increased SV, as would be expected in the normal heart. SV is restored with decreased contractility by lowering afterload (loop 5) or increasing preload (loop 4) even more than the preload increase described in loop 3.

 
 

FIGURE 3-18  Effect of changing contractility on the left ventricular (LV) pressure-volume relation (see text for full description).  (From Braunwald E, Ross J, Sonnenblick EH: Mechanisms of contraction of the normal and failing heart. 2nd ed. Boston, 1976, Little, Brown.)

 

 

 

▪ EFFECT OF HEART RATE ON CARDIAC PERFORMANCE

Heart rate (HR) plays a major role in determining cardiac function for various reasons. HR affects preload by determining the length of time for ventricular filling. Because coronary and therefore myocardial blood supply occurs during diastole, HR directly affects subendocardial blood flow. If subendocardial flow is compromised with shorter diastolic filling, ischemia may result. A downward spiral ensues because the ischemic, less-compliant heart resists optimal ventricular filling and preload decreases. Increased HR is critical during exercise to increase CO and meet increased metabolic needs. HR changes as a result of ongoingmultifactorial development; the effect of HR on cardiac function is discussed further in this context.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ DEVELOPMENTAL ASPECTS OF MYOCARDIAL FUNCTION

Studies in fetal and neonatal lambs, as well as studies of isolated fetal and neonatal muscle fibers, have helped to define various maturational aspects related to cardiac function. The fetus is immature in the structure, function, and innervation of the heart. When the length-tension relationship is examined in the fetal and adult myocardium, the following findings have been described: (1) resting tension is higher in the fetus than in adult sheep and (2) the extent and velocity of shortening and developed tension differ between fetuses and adults (Figs 3-19 and 3-20 [19] [20]). The higher percentage of noncontractile protein in fetal myocytes (60% versus 30% in the adult myocardium) ( Rychik, 2004 ) may explain some of these findings. The fetal and newborn ventricle is less compliant, or “stiffer.” As a result, the newborn responds poorly to volume loading and shows less ability to augment CO with changes in preload ( Fig 3-21 ). The increased negative effect of afterload on CO is exaggerated in fetal and newborn hearts. In addition, unlike in the adult, afterload reduction with drugs (such as nitroprusside) does not increase CO ( Kuipers et al., 1984 ; Mirro et al., 1985 ).

 
 

FIGURE 3-19  Length-tension relationships in fetal lambs and adult sheep demonstrating the lower resting tension and great active tension development in adult sheep. Fetal myocardium has a higher resting tension but less active tension development, indicating less compliance—a “stiffer” ventricle than in the adult.  (From Friedman WF: The intrinsic properties of the developing heart. Prog Cardiovasc Dis 15:87, 1972.)

 

 

 

 
 

FIGURE 3-20  Relationships between (A) the extent of shortening and (B) the velocity of shortening and developed tension in fetal and adult cardiac muscle strips.  (From Friedman WF: The intrinsic physiologic properties of the developing heart. Prog Cardiovasc Dis15:87, 1972.)

 

 

 

 
 

FIGURE 3-21  Response to volume loading in newborn lambs and adult sheep. At constant heart rate, limited cardiac reserve is demonstrated in the youngest lambs by a reduced ability to augment cardiac output at any filling pressure compared with the adult or older lambs. LVEDP, left ventricular end-diastolic pressure.  (From Friedman WF, George BL: Treatment of congestive heart failure by altering loading conditions of the heart. J Pediatr 106:697, 1985.)

 

 

 

Myocardial cellular replication differs in the fetus compared with that in the adult. Cardiomyocytes contain the contractile elements of the heart. Primitive mesodermal cells differentiate into cardiomyocytes and then receive a signal to exit the cell cycle at approximately the time of birth. It is these early fetal cardiomyocytes that have the potential to divide and increase in number (hyperplasia), in contrast to mature adult cardiomyocytes, which can only grow in size (hypertrophy). For example, the left ventricular myocyte increases in volume 30- to 40-fold during the neonatal to adolescent period ( Rychik, 2004 ).

The fetal myocardium has different relaxation properties than that of the adult. Experimental animal studies in the fetus have demonstrated a difference in the process of rapid removal of calcium from troponin C, the mechanism responsible for myocardial relaxation ( Mahoney, 1996 ). This may be due to diminished sarcoplasmic reticulum function and greater dependence on the sodium-calcium exchanger process to remove cytosolic calcium in the fetus ( Artman, 1992 ). Finally, the energy source for myocardial cell metabolism differs. Long-chain fatty acids are the preferred fuel in adults; in the fetus and neonate, lactate is the primary agent metabolized ( Fisher et al., 1981 ). In the fetus, this is due to a deficiency in the enzyme carnitine palmitoyl transferase-1, responsible for transporting long-chain fatty acids into the mitochondria.

Traditionally, the limitations in myocardial function in the fetus have been thought to be due to the fetal myocardial architecture. Grant and others (1992a, 1992b) [57] [58] and Grant and Walker (1996)proposed an interesting and plausible alternative theory: fetal SV is limited not by intrinsic properties of the fetal myocardium but by ventricular constraint due to extrinsic compression of the fetal heart. At birth, fetal SV doubles at the same time that ventricular constraint by tissues around the fetal heart is dramatically changed. The chest wall, the lungs, and the pericardium create limitations on fetal ventricular preload and are major determinants of fetal cardiac function. Expansion of the lungs and clearance of fetal lung liquid may be the major determinant of an increase in left ventricular preload and an increase in SV seen in the newborn infant. Clinically, this theory is supported by observations in neonatal open chest surgery where closure of the chest wall has an often dramatic impact on cardiac function.

The neonatal heart has been shown experimentally to function with increased myocardial contractility ( Geis et al., 1975 ; Riemenschneider et al., 1981 ; Rudolph et al., 1981 ). This may result from a number of factors, including increased β-stimulation after birth and the effects of thyroid hormone both prenatally and postnatally ( Breall et al., 1984 ; Mahdavi et al., 1987 ) or the release of ventricular constraint as proposed by Grant (1999) .

Because the newborn heart functions at high levels of preload, afterload, contractility, and HR, a resultant marked limitation in cardiac reserve occurs. Sudden and profound depression in CO is not unusual in neonates under certain adverse conditions, such as hypoxia or acidosis, or under the influence of anesthetics. Increased preload or afterload or depressed contractility is poorly tolerated by the newborn.

HR plays an important role in cardiac function. The limited ability of the fetal heart to increase SV results in marked changes in CO with changes in HR. For example, a 10% to 15% increase in CO is observed when HR increases from 160 to 240 beats per minute. Conversely, a 20% to 25% decrease in CO is observed when the HR falls to 120 beats per minute ( Rudolph, 1987 ). The role of HR in regulating cardiac performance remains controversial. A study examining the role of HR (over a range of 130 to 175 beats per minute) and CO in fetuses demonstrated the relationship between cardiac cycle length and SV. As HR increases in fetal lambs, right and left ventricular output increases. At slower HRs, prolongation of diastole does not increase ventricular filling to the same degree as observed in the mature heart. This phenomenon is a reflection of the stiffness and ventricular interdependence of both ventricles in the fetus. In addition, observed decreases in SVs with increasing heart rate did not reveal changes in ventricular output ( Fig 3-22 ) ( Kenny et al., 1987 ).

 
 

FIGURE 3-22  The relationship between right ventricular output (interrupted lines) and left ventricular output (solid lines) with varying heart rate in 25 human fetuses.  (From Kenny J, et al.: Effects of heart rate on ventricular size, stroke volume and output in the normal human fetus: A prospective Doppler echocardiographic study. Circulation 76:52, 1987.)

 

 

 

As fetal HR increases, the maximum diastolic cavity size diminishes but without any significant change in systolic function measured as left ventricular area shortening ( Fig 3-23 ). This figure shows a reduction in maximum diastolic cavity size with increasing HR but little or no change in systolic contractile function (as measured with fractional change in ventricular area). It appears that the Frank-Starling mechanism remains the major regulator of CO in the fetus. Clearly, the newborn is more sensitive to changes in HR, but this is probably because of the “stiffness,” or decreased compliance, of the newborn myocardium. When sequential atrioventricular pacing is performed in newborn lambs and 1-month-old lambs, the contribution of atrial pacing is not as great as in the newborn, possibly because the atria are limited in the amount ejected into the stiffer newborn ventricle ( Kaufman and Rudolph, 1988 ).

 
 

FIGURE 3-23  (A) Changes in left ventricular diastolic area with change in heart rate in 11 human fetuses. (B) Changes in the left ventricular area shortening with heart rate in the same 11 fetuses.  (From Kenny J, et al.: Effects of heart rate on ventricular size, stroke volume and output in the normal human fetus: A prospective Doppler echocardiographic study. Circulation 76:52, 1987.)

 

 

 

At birth and over the ensuing weeks and months of infant life, the cardiovascular system shows evidence of increasing functional reserves. With birth, left ventricular output increases as a result of increases in venous return and HR, as well as a result of increases in inotropic stimulation, removal of extracardiac restraints, and improvement of ventricular interaction. In the neonate, CO (indexed for body weight) falls gradually, dP/dtmax returns to fetal levels, and extrasystolic potentiation and the inotropic response to β-adrenergic receptor stimulation increase ( Klopfenstein and Rudolph, 1978 ; Anderson et al., 1984 ). All of these changes result in an increase in functional reserve. Because of these reserves, the heart is better able to respond to stress. These concepts apply to the normal newborn. When congenital heart disease is present, dramatic changes in preload, afterload, and contractility may significantly impair cardiac performance and must be carefully considered.

▪ DEVELOPMENTAL ELECTROPHYSIOLOGY

The electrophysiologic circuits of the fetus and neonate have been studied in chick and rat heart embryos, with limited data from human tissue. Resting membrane potential increases as development proceeds. Potassium permeability increases with development. In the neonatal heart, resting membrane potentials are higher compared with that in adults. Duration of the action potential increases during postnatal development, and atrioventricular conduction is more rapid in the newborn. Tissue refractory periods are significantly shorter in newborns and infants ( Rosen et al., 1981 ). These laboratory findings support the ongoing maturation of sarcolemmal ionic channels, alterations in cell coupling, and structural changes in cell grouping. Clinically, both the neonate and infant can conduct rapid impulses from the atrium to the ventricle. Because of the shorter refractory periods, responses of the premature can be conducted at shorter coupling intervals ( Box 3-1 ).

BOX 3-1 

Diseases or Conditions Associated With Rhythm Disturbances

  

 

Sinus Tachycardia (Persistent)

  

 

Anemia

  

 

Infection

  

 

Hypoxia, acute

  

 

Hypotension, shock

  

 

Hyperthyroidism

  

 

Acute intermittent porphyria

  

 

Pheochromocytoma

  

 

Sinus Bradycardia (Persistent)

  

 

Hepatitis (jaundice)

  

 

Typhoid fever

  

 

Increased intracranial pressure

  

 

Cerebral hypoxia

  

 

Hypothyroidism

  

 

Drugs (morphine sulfate, digoxin, propranolol)

  

 

Hypothermia

  

 

Anorexia nervosa

  

 

Atrial Tachycardia

  

 

Infection

  

 

Trauma

  

 

Neoplasm

  

 

Cardiac catheterization

  

 

Anesthesia, surgery

  

 

Emotional stress

  

 

Thalassemia major

  

 

Digoxin toxicity

  

 

Atrial Flutter

  

 

Congenital thyrotoxicosis

  

 

Thalassemia major

  

 

Imipramine

  

 

Atrial Fibrillation

  

 

Hypokalemia

  

 

Thyrotoxicosis

  

 

Pseudoephedrine

  

 

Drugs

  

 

Tricyclic antidepressants

  

 

Ventricular Tachycardia

  

 

Anesthesia

  

 

Cardiac catheterization and surgery

  

 

Electrolyte imbalance

  

 

Cardiac neoplasm (primary or metastatic)

  

 

Subarachnoid hemorrhage

  

 

Drugs

  

 

Cardioauditory syndrome

  

 

Psychogenic

  

 

Tricyclic antidepressants

  

 

First-Degree Atrioventricular (AV) Block

  

 

Rheumatic fever

  

 

Diphtheria

  

 

Trichinosis

  

 

Familial

  

 

Second-Degree AV Block

  

 

Infection

  

 

Hypocalcemia

  

 

Hyperkalemia

  

 

Drugs and toxins

  

 

Familial

  

 

Brain abscess

  

 

Third-Degree AV Block

  

 

Infection

  

 

Diphtheria

  

 

Collagen disease

  

 

Hyperkalemia

  

 

Digitalis

  

 

Tricyclic antidepressants

  

 

Rheumatic fever


Congenital Arrhythmias

  

 

Congenital Malformation

  

 

Ventricular inversion with transposition of great arteries

  

 

AV canal

  

 

Secundum atrial septal defect

  

 

Ebstein's anomaly

  

 

Intrauterine cytomegalovirus infection

  

 

Arrhythmia

  

 

First-, second-, and third-degree AV block

  

 

Preexcitation syndrome with supraventricular tachycardia, first-degree heart block

  

 

Sinoatrial and AV nodal damage


Acquired Postoperative Arrhythmias

  

 

Intra-atrial repair of transposition of great arteries

  

 

Secundum atrial septal defect

  

 

Sick sinus syndrome, complete heart block, premature ventricular contractions, ventricular tachycardia and fibrillation, atrial flutter and fibrillation

  

 

Atrial flutter or fibrillation, first- and second-degree heart block, complete heart block

Cell membranes and cell metabolism develop and mature simultaneously with autonomic innervation of the heart. Parasympathetic innervation is noted very early in development. Sympathetic innervation may begin in the area of the sinoatrial node and proceed to the ventricular myocardium ( Pappano, 1972 ; Pappano and Loeffelholz, 1974 ) but is definitely not complete at birth. Cardiac catecholamine levels are significantly less in the neonatal heart compared with the adult. Both α- and β-adrenergic receptors have been identified in cardiac myocytes. However, these receptors may respond quite differently with incomplete innervation.

Since the 1990s, it has become clear that cardiac development is complex and involves the induction of genes at certain times in development—an induction process that influences cardiac chamber formation, conduction path development, and function. The end result is that the early peristaltic tubular heart develops into a synchronously contracting four-chambered heart with predictable conduction paths ( Moorman and Christoffels, 2003 ).

▪ INTERACTION OF THE NERVOUS SYSTEM AND THE HEART

The nervous system plays a critical role in regulating cardiac function by ensuring adequate perfusion to all organ systems under various physiologic conditions. Fine-tuning of cardiac function is accomplished by the autonomic nervous system. A highly integrated and complex interaction occurs between the cardiovascular system, nervous system, and cardiovascular reflex arcs on cardiac function. Baroreceptor and chemoreceptor sensitivity may be protective reflexes in the fetus and are almost completely mature at birth. Arterial baroreceptors play a critical role in cardiovascular regulation. Baroreceptors are located in the aortic arch, at the carotid bifurcation, and send afferents to the vasomotor center in the medulla. In fetal lambs, denervation of these baroreceptors ( Yardley, 1979 ) results in marked variability in arterial BP. Peripheral chemoreceptors are found in the aortic arch, carotid bodies, and main pulmonary artery. Dawes and Mott (1962) suggested that aortic chemoreceptors are important for cardiovascular control, whereas carotid chemoreceptors regulate respiratory parameters. Peripheral autonomic regulation is important in cardiovascular homeostasis. Rudolph and Heymann (1973) suggested that basal sympathetic tone is present during early gestation and increases at birth, as evidenced clinically by gradually increasing arterial BP. In addition to direct innervation, circulating catecholamines play a role in cardiovascular control. These circulating catecholamines are derived from adrenal tissue and para-aortic chromaffin tissue and may exert effects directly even before the complete maturation of innervation.

Autonomic control of cardiovascular function is complex, differing under basal resting conditions compared with stressful stimulation. HR remains constant during most of fetal life and then decreases during the first 2 months of neonatal life. During this period, significant changes are occurring in sympathetic and parasympathetic innervation. Neither fetal vagotomy nor treatment with atropine or chemical sympathectomy alters fetal HR or BP ( Woods et al., 1977 ; Assali et al., 1978 ; Zugaib et al., 1980 ; Tabsh et al., 1982 ). Immature autonomic neuronal or receptor development may contribute to differences described and verified by laboratory studies in fetuses and neonates. A lower level of neurotransmitters in fetal hearts compared with older hearts suggests a lack of mature innervation. In contrast, fetuses administered norepinephrine and isoproterenol show a hypersensitivity in comparison with adult responses, supporting the concept that the receptors are fully functional ( Geis et al., 1975 ).

Baroreceptor reflexes are present and operative in early fetal life, but qualitative and quantitative differences exist in the neonate and adult. For example, norepinephrine infusions increase both HR and BP in the fetal lamb, whereas neonate and adult animals respond with bradycardia as BP is increased. The injection of veratridine into the RA of a mature fetus or neonate elicits the Bezold-Jarisch reflex, resulting in hypotension and bradycardia, just as in the adult. However, the younger fetus responds with tachycardia and hypertension (blocked by α- and β-blockade), suggesting sympathetic innervation is functional ( Vappavouri et al., 1973 ; Assali et al., 1978) .

Neonatal Cardiovascular Reflex Development

At birth, dramatic circulatory changes occur, accompanied by parasympathetic activity that regulates HR and pulmonary vascular reactivity, as well as a profound activation of the neurohumoral sympathoadrenal axis. Both norepinephrine and epinephrine are dramatically increased during vaginal delivery and directly affect HR, SVR, and BP. Infants delivered via cesarean section or prematurely show a diminished neurohumoral response, whereas infants stressed by hypoxia, acidosis, or both at birth show an exaggerated response ( Faxelius et al., 1984 ). Diminished ventricular compliance with a simultaneous opening of vascular beds leads to a fall in circulating fluid volume and SVR. These observations may explain the normal decrease in BP seen soon after birth ( Romero and Friedman, 1979 ). Many studies have examined the role of HR and its variability during maturation. The results of many studies suggest that HR develops as a function of postnatal age, whereas HR variability is a reflection of conceptional age. Around 6 months of age, differences between full-term and preterm infants with respect to HR variability and absolute HR disappear. HR declines gradually until adult levels are reached during adolescence ( Church et al., 1967 ; Katona and Egbert, 1978 ; Katona et al., 1980 ; Mazza et al., 1980 ).

Childhood Cardiovascular Reflex Development

Studies examining the role of autonomic control are more limited in older children, but observed changes clearly vary with respect to age, gender, and race. Increased parasympathetic tone during rapid eye movement (REM) sleep has been observed in children with sleep dysrhythmias and sleep apnea ( Guilleminault et al., 1981 ; Miller, 1982 ; McNicholas et al., 1983 ; Thach, 1985 ). Increased vagal tone has been described as a cause of syncope during exercise in some children. The values in older children and adolescents probably approach adult levels with respect to cardiovascular responses to exercise and Valsalva maneuver, although studies are definitely limited. The potent interaction between central respiratory and cardiovascular reflexes is an area of ongoing research in both children and adults.

Birth results in dramatic changes in the cardiovascular system that can be best summarized as an increase in left ventricular output, secondary to increases in venous return and HR, increases in inotropic stimuli, removal of extracardiac restraints, and improvement in ventricular interaction. During the neonatal period, CO (indexed for body weight) falls gradually, dP/dt max returns to fetal levels, and postextrasystolic potentiation and the inotropic response to β-adrenergic receptor stimulation increase ( Klopfenstein and Rudolph, 1978 ; Anderson et al., 1984 ). The end result is that the infant's heart is acquiring greater functional reserves. Ironically, it is this reserve that also permits many patients with congenital heart defects to survive the neonatal period.

▪ EFFECTS OF CONGENITAL HEART DISEASE ON CARDIOPULMONARY DEVELOPMENT

It is clear that cardiac and pulmonary development is intricately related in the normal fetus and neonate. In the presence of a congenital heart defect, this relationship has even more profound interactions. Despite the fact that pulmonary blood flow is minimal in the fetus, the presence of a congenital heart defect, which alters pulmonary hemodynamics, has important effects. For example, the presence of pulmonary atresia early in gestation (before the ninth week) results in the persistence of connections with primitive intersegmental arteries, which appear at birth as direct branches from the descending thoracic aorta. These abnormal vessels may be the only source of pulmonary blood flow—acting as mini-systemic artery-to-pulmonary artery collateral shunts in the neonatal period. If pulmonary stenosis or atresia occurs later in gestation, these bronchopulmonary collaterals are still noted. However, the pulmonary arteries are also hypoplastic and have a decrease in elastin in the media. This lack of elastin may explain the poor dilatation seen in pulmonary arteries even after a surgical shunt is placed in patients with pulmonary atresia ( Rabinovitch et al., 1981 ; Rosenberg et al., 1987 ).

In patients with large left-to-right shunts, an increase in medial wall thickness and abnormal extension of muscle into peripheral pulmonary arteries is found. These patients may still exhibit evidence of severe pulmonary hypertension after surgical repair, because this anatomic substrate in the pulmonary vascular bed may take time to resolve postoperatively ( Haworth and Reid, 1977 ).

The effects of increased volume and pressure—as seen in a large left-to-right shunt such as a ventricular septal defect—may have a toxic effect on the endothelial cell integrity in the pulmonary bed. Mechanical as well as humoral influences may cause a persistence of fetal muscularization throughout the pulmonary vascular bed, and if the shunt is not closed, the process results in irreversible pulmonary vascular obstructive disease.

Any patient who has undergone repair of a congenital heart defect may have residua of this abnormal physiology in the pulmonary vascular bed as well as within the cardiovascular system. Understanding of the dynamic nature of cardiopulmonary physiology in the fetus and neonate with congenital heart disease has resulted in a more aggressive attempt at early infant repair of defects so that the pulmonary vascular bed is spared the possibility of permanent changes. The fetal environment may indeed have long-lasting effects on adult cardiovascular health and disease and that, to be effective, cardiovascular disease prevention begins during the period of fetal development.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ ASSESSMENT OF THE CARDIOVASCULAR SYSTEM

When a patient presents for anesthetic evaluation and management, the anesthesiologist must know about age-related differences in history, physical examination findings, routine laboratory results, and more sophisticated evaluations of cardiac function, such as electrocardiography, echocardiography, and cardiac catheterization. Age-related differences are not surprising in view of developmental cardiovascular physiology and are only briefly described. This discussion focuses on evaluation of the healthy child; evaluation of the child with congenital heart disease is discussed in greater detail inChapter 17 , Anesthesia for Cardiovascular Surgery. Routine cardiac catheterization is reviewed.

▪ HISTORY AND PHYSICAL EXAMINATION

The well-being of a child is definitely reflected in his or her ability to “thrive,” which includes the attainment of weight and height expectations during development. Obviously children vary in these parameters, and weight and height curves reflect the acceptable percentiles for children at different ages. In addition, acute and chronic illnesses may differentially affect these important parameters. Growth charts should be reviewed when obtaining a history in a young child. Weight may be the first parameter to show the negative effects of illness, followed by height and then head circumference measurements. In an adult, cardiac symptoms are often graded using established guidelines, such as those of the New York Heart Association. These guidelines often do not apply to pediatric patients. In the neonate and young infant, feeding difficulties, especially fatigue and tachypnea, are important symptoms of cardiac failure. The ability to maintain peer or sibling exercise levels is a less formal but helpful piece of historical information in children.

The healthy newborn may have a respiratory rate of 50 breaths per minute and a systolic BP of 55 mm Hg. These physical findings are obviously abnormal in a 3-year-old child. Tables 3-1 and 3-2 [1] [2] and Figures 3-24 and 3-25 [24] [25] outline normal values for respiratory rate, HR, and BP in healthy children of various ages. Table 3-3 defines normal values for more invasive evaluations of cardiac function, including hemodynamic and saturation data. Figure 3-26 summarizes the progression of changes in HR, CO, and SV over childhood.

TABLE 3-1   -- Normal values for respiratory rates in children

Age

Respiratory Rate (min)

Birth to 6 wk

45 to 60

6 wk to 2 yr

40

2 to 6 yr

30

6 to 10 yr

25

>10 yr

20

 

 


TABLE 3-2   -- Acceptable heart rates in children (beats/min)[*]

 

Awake

Asleep

Exercise/Fever

Newborn

100 to 180

80 to 160

<220

1 wk to 3 mo

100 to 220

80 to 200

<220

3 mo to 2 yr

80 to 150

70 to 120

<200

2 to 10 yr

70 to 110

60 to 90

<200

>10 yr

55 to 90

50 to 90

<200

*

From Adams FH, Emmanoulides GC, editors: Moss' heart disease in infants, children, and adolescents, 3rd ed. Baltimore, 1983, Williams & Wilkins.

 

 
 

FIGURE 3-24  Normal values of resting blood pressure in boys and girls aged 2 to 18 years.  (From Blumenthal S, et al.: Pediatrics 59:797, 1977.)

 

 

 

 
 

FIGURE 3-25  Variations in average heart rate and blood pressure with age.  (From Moore RA: Anesthesia considerations for patients undergoing palliative or reoperative operations for congenital heart disease. In Swedlow DB, Russell RC, editors: Cardiovascular problems in pediatric critical care. New York, 1986, Churchill Livingstone.)

 

 

 


TABLE 3-3   -- Normal values for invasive evaluations of cardiac function[*]

Location

Infants and Children

Newborns

Right atrium

a = 5 to 8
v = 2 to 6
M = 2 to 6

M = 0 to 4

Right ventricle

15 to 25/2 to 5

35 to 80/1 to 5

Pulmonary artery

15 to 25/8 to 12
M = 10 to 16

35 to 80/20 to 50
M = 25 to 60

Pulmonary wedge

a = 6 to 12
v = 8 to 15
M = 5 to 12

 

Left atrium

a = 6 to 12
v= 8 to 15
M = 5 to 10

M = 3 to 6

Left ventricle

80 to 130/5 to 10

 

Systemic artery

90 to 130/60 to 80
M = 70 to 95

65 to 80/45 to 60
M = 60 to 65

*

Data from Rudolph AM: Congenital disease of the heart, Chicago, 1974, Year Book Medical Publishers. (a = a wave, v = v wave, M = mean.)

 

 
 

FIGURE 3-26  Changes in cardiac output (CO), stroke volume (SV), and heart rate (HR) with age.  (From Rudolph AM, editor: Congenital diseases of the heart. Chicago, 1974, Year Book Medical.)

 

 

 

The normal electrocardiogram and chest radiograph reflect changes in physiology as well as anatomic changes in chamber size and position. The normal neonatal chest radiograph and electrocardiogram would be read as abnormal in a teenager. The large thymic shadow on the neonatal chest radiograph may be misinterpreted as cardiomegaly. HR increases during the first 2 months of life and then decreases gradually over the ensuing years of early childhood. The progression from RV dominance at birth to LV dominance is reflected in the presence of increased right ventricular forces on the neonatal electrocardiogram. Detailed age-related electrocardiographic values are available for review to ensure proper electrocardiogram interpretation.

▪ SPECIAL EVALUATION OF THE CARDIOVASCULAR SYSTEM

Cardiac Catheterization

Despite the dramatic contribution of echocardiography, especially two-dimensional echocardiography, in the diagnosis of congenital heart disease, cardiac catheterization laboratories are still active areas of investigation. In addition to diagnostic cardiac catheterization and angiocardiography, the catheterization laboratory is the site of major interventional procedures once delegated to the operating room, including dilatation of valvar stenosis, vascular stenoses, closure of patent ductus arteriosus, investigational closure of atrial and ventricular septal defects, complex electrophysiologic studies, ablation of abnormal foci of arrhythmias, and coil closure of collateral blood vessels. These complex procedures may be associated with hemodynamic instability, and the involvement of pediatric anesthesiologists in the sedation, monitoring, and, often, administration of general anesthesia has become increasingly important in the care of these patients.

Cardiac catheterization remains an important diagnostic tool in delineating anatomy and hemodynamics, especially preoperatively. Cardiac catheterization carries risk, especially in sick neonates and infants. Major complications occur in about 30% of high-risk infants, 14% of medium-risk infants, and 4% of low-risk infants ( Cohn et al., 1985 ). In an 8-year survey of 6,101 children, the overall mortality rate within 48 hours of catheterization was 1.7%, ranging from 10.2% in the first week of life to 0.5% in patients older than 1 year. Approximately 1% of all interventional procedures result in death ( Lock et al., 1992 ). Complications include arterial and venous complications, arrhythmias, myocardial perforation, hypoxia, acidosis, apnea, and air emboli. Despite the fact that the cardiac catheterization laboratory is not only the site for diagnosis but also the site for interventional treatment in younger and more critically ill infants, complication rates have improved.

Cassidy and others (1992) prospectively examined cardiac catheterization complications in a 3-year period (January 1986 through October 1988) and compared them with complications in the same laboratory in 1974. In their study, 1,037catheterizations (885 diagnostic and 152 diagnostic/interventional procedures) were performed in 888 patients (age range, 1 day to 27 years; median age, 15.6 months). There were 15 major complications (1.4%), 70 minor complications (6.8%), and 30 incidents (2.9%). Two patients died as a result of the procedure, and two patients died as a result of pericatheterization clinical deterioration caused by the cardiac abnormality. The great majority of complications were successfully treated or were self-limited, and the patients had no residua. Of patients with 13 nonfatal major complications and 70 minor complications, residua were evident in 7 patients, and 3 without evident residua had the potential for sequelae (0.7% and 0.3% of catheterizations). A comparison of the diagnostic and balloon atrial septostomy cases in the present study with similar cases in the 1974 study shows that the incidence of major complications has decreased from 2.9% to 0.9%, minor complications and incidents have decreased from 11.7% to 7.9%, and pericatheterization deaths not attributable to catheterization have decreased from 2.8% to 0.2%.

Vitiello and others (1998) studied complications in 4,952 consecutive patients (age range, 1 day to 20 years; median age, 2.9 years) at The Hospital for Sick Children in Toronto. One or more complication occurred in 8.8% of the study patients (major complication in 102 patients and minor complication in 458 patients). Vascular complications were the most common adverse event (3.8% of procedures), and arrhythmic complications (n = 24) were the most common major complication. Death occurred in seven cases (0.14%) as a direct complication of the procedure and was more common in infants (n = 5). Medical management (including an increasing involvement of anesthesiology staff in monitoring, sedation, and anesthetic management), better patient selection, and stabilization before catheterization have all contributed to decreased complication rates in centers nationally.

Hemodynamic Evaluation

Cardiac catheterization includes the measurement of intracardiac pressures and oxygen saturation, gradients across valves, pulmonary and systemic blood flow, CO, quantity and direction of shunt flow, and resistance. In addition, the changes in these measurements are often assessed after the administration of drugs, oxygen, or both. Normal hemodynamic data in children beyond the neonatal period are given (Table 3-4 ); these data are obviously altered by structural or acquired heart disease.


TABLE 3-4   -- Normal cardiovascular values beyond the neonatal and infancy period[*]

Location

Average

Range

Mean right atrial pressure (central venous pressure)

3

1 to 5 mm Hg

Right ventricular pressure

 

 

Systolic

25

17 to 32 mm Hg

Diastolic

5

1 to 7 mm Hg

Pulmonary arterial pressure

 

 

Systolic

25

9 to 19 mm Hg

Diastolic

10

17 to 32 mm Hg

Mean

15

4 to 13 mm Hg

Mean pulmonary artery wedge pressure

9

6 to 12 mm Hg

Mean left atrial pressure

8

2 to 12 mm Hg

Cardiac index

3.5

2.5 to 4.2 L/min per m2

Stroke volume index

45 mL/m2

 

Oxygen consumption

140

110 to 150 L/min per m2

Vascular resistance

 

 

Pulmonary

 

1 to 3 hybrid units/m2

 

 

80 to 240 dynes·s·cm-5·m-2

Systemic

 

10 to 20 hybrid units/m2

 

 

800 to 1600 dynes·s·cm-5·m-2

*

Data from Rudolph AM: Congenital disease of the heart, Chicago, 1974, Year Book Medical Publishers.

 

The difference in pressure between two sites in the cardiac system is called a gradient and can be measured as a mean gradient, a peak gradient, or an instantaneous gradient. Typically a gradient is measured during withdrawal of the pressure catheter across two locations. With severe stenosis, a minimal gradient may be described because flow is severely compromised. In addition to assessing a gradient, some measurement of flow must be made.

Oxygen Content and Saturation.

Oxygen saturation is the percent of hemoglobin present as oxyhemoglobin; it is measured directly with oximetry. Oxygen capacity is the maximal amount of oxygen that can be bound to hemoglobin. This value is calculated by multiplying the patient's hemoglobin by 1.34 and is expressed in milliliters per 100 milliliters. Oxygen content is thetotal amount of oxygen present in blood and includes oxygen bound to oxyhemoglobin as well as oxygen dissolved in the plasma. Oxygen content is the product of the oxygen saturation value multiplied by 1.34 multiplied by 10, where 1.34 is the amount of O2 that 1 g of hemoglobin carries when fully saturated. The number 10 converts 100 mL to liters. Dissolved oxygen is usually ignored because it is so small. However, when PO2 is high, dissolved oxygen may be high and must be considered. Dissolved oxygen is equal to PaO2 × 0.003 mL/100 mL. Oxygen content and oxygen consumption (   O2) must be known to calculate systemic and pulmonary blood flow.

Pulmonary blood flow

Systemic blood flow

where P   O2 is the oxygen content in the pulmonary vein, PaO2 is the oxygen content in the pulmonary artery, SaO2 is the oxygen content in a systemic artery or aorta, and M   O2 is the oxygen content in a mixed venous sample.

The mixed venous oxygen content should be the same in the RA as in the pulmonary artery if no shunts are present. However, venous blood is poorly mixed in the RA where streaming and large variations in oxygen content are normally seen, as in coronary sinus return. Mixing on the left side of the heart is much more uniform. Saturation data become important in the detection and quantification of shunt flow.

Shunts.

Shunts can be diagnosed with various techniques: angiocardiography, echocardiography, dye indicators such as radionucleotides and indocyanine green dye, and, more commonly, oxygen saturation data. Shunts can be left-to-right, right-to-left, or bidirectional. Using saturation data, quantity of shunt flow can be calculated.

Left-to-right shunts.

When blood from the left side of the heart is shunted to the right side, pulmonary blood flow is increased and the saturation of mixed venous blood is increased by the presence of fully oxygenated left-sided blood. A series of samples is drawn in quick succession from each chamber of the right heart, including superior and inferior vena cava blood. An increase in blood saturation or “step-up” beyond a normally accepted variation indicates a left-to-right shunt. For example, a mid-right atrial saturation should be no higher than superior vena cava blood by 7% to 9%. Saturation step-up of 9% indicates a shunt at the atrial level. Because of streaming and poor mixing in the RA, calculation of shunt flow in shunts at the atrial level may be less accurate than at other levels. A step-up of greater than 6% between the RV and the pulmonary artery suggests a ventricular septal defect.

Right-to-left shunts.

When desaturated right-sided blood is shunted into the left side of the heart, decreased saturation is observed, called a “step-down.” Because left-sided saturations should be fully saturated except in the presence of pulmonary disease, a right-to-left shunt can be suspected whenever desaturation is seen in the left-sided saturation data. A decrease in saturation of more than 2% to 3% strongly suggests a right-to-left shunt.

Shunt magnitude.

In addition to diagnosing the presence and direction of a shunt, the magnitude of the shunt must be determined. A left-to-right shunt increases the amount of pulmonary blood flow while decreasing the systemic blood flow. The quantity of left-to-right shunt can be calculated as follows:

3.3  

where    LR is left-to-right shunt and    P -    S is pulmonary blood flow minus systemic blood flow.

Similarly, right-to-left shunts can be calculated as follows:

3.4  

In a discussion of shunts, the term    P/   S is often used to describe the flow ratio between pulmonary and systemic flow. Combining these equations, shunt flow can be determined without obtaining oxygen consumption data, as follows:

3.5  

For example,    P/   S can be quickly calculated after reviewing available saturation data. If the superior vena cava saturation is 80, the pulmonary artery saturation is 95, and the systemic artery saturation is 100, then    P/   S = (100 - 80)/100 - 95) = 20/5 = 4/1.

Mixed venous saturation data should always be obtained from the chamber most likely to represent complete admixture, which is usually the chamber proximal to the suspected shunt. For example, in a ventricular septal defect, the RA yields the best mixed venous sample, whereas the RV yields the best data in a patient with a suspected patent ductus arteriosus.

Measurement of Cardiac Output.

CO is expressed in liters per minute and, when corrected for body surface area (L/min per m2), is called cardiac index (CI). CO is calculated in the catheterization laboratory using indicator dye techniques. Thermodilution CO is calculated using cold saline solution as the indicator; with the Fick method, oxygen is used as the indicator. The Fick principle states that blood flow through an organ is proportional to the amount of an indicator (oxygen) that is added to or removed from the organ as the blood flows through it. When oxygen is the indicator, CO can be calculated and requires the measurement of oxygen consumption and oxygen content in arterial and venous blood.

Oxygen Consumption.

Oxygen consumption (   O2) is calculated using the amount of oxygen in inspired and expired air as follows:

3.6  

where VI is volume of inspired air (mL/min), FIO2 is fraction of inspired oxygen, VE is volume of expired oxygen (mL/min), and FEO2 is fraction of mixed expired oxygen.

The volume of air collected in a Douglas bag is analyzed for oxygen, and carbon dioxide levels are compared with those in ambient air. Younger children and infants may make measurement of oxygen consumption technically challenging, although a hood analyzer can be used. Oxygen consumption may be estimated using HR and age as variables. In many calculations, oxygen consumption is assumed and values are obtained from published tables ( Table 3-5 ).


TABLE 3-5   -- Oxygen consumption table[*]

HEART RATE (beats/min)

Age

50

60

70

80

90

100

110

120

130

140

150

160

170

Male patients

3

 

 

 

155

159

163

167

171

175

178

182

186

190

4

 

 

149

152

156

160

163

168

171

175

179

182

186

6

 

141

144

148

151

155

159

162

167

171

174

178

181

8

 

136

141

145

148

152

156

159

163

167

171

175

178

10

130

134

139

142

146

149

153

157

160

165

169

172

176

12

128

132

136

140

144

147

151

155

158

162

167

170

174

14

127

130

134

137

142

146

149

153

157

160

165

169

172

16

125

129

132

136

141

144

148

152

155

159

162

167

 

18

124

127

131

135

139

143

147

150

154

157

161

166

 

20

123

126

130

134

137

142

145

149

153

156

160

165

 

25

120

124

127

131

135

139

143

147

150

154

157

 

 

30

118

122

125

129

133

136

141

145

148

152

155

 

 

35

116

120

124

127

131

135

139

143

147

150

 

 

 

40

115

119

122

126

130

133

137

141

145

149

 

 

 

Female patients

3

 

 

 

150

153

157

161

165

169

172

176

180

183

4

 

 

141

145

149

152

156

159

163

168

171

175

179

6

 

130

134

137

142

146

149

153

156

160

165

168

172

8

 

125

129

133

136

141

144

148

152

155

159

163

167

10

118

122

125

129

133

136

141

144

148

152

155

159

163

12

115

119

122

126

130

133

137

141

145

149

152

156

160

14

112

116

120

123

127

131

134

133

143

146

150

153

157

16

109

114

118

121

125

128

132

136

140

144

148

151

 

18

107

111

116

119

123

127

130

134

137

142

146

149

 

20

106

109

114

118

121

125

128

132

136

140

144

148

 

25

102

106

109

114

118

121

125

128

132

136

140

 

 

30

99

103

106

110

115

118

122

125

129

133

136

 

 

35

97

100

104

107

111

116

119

123

127

130

 

 

 

50

94

98

102

105

109

112

117

121

124

128

 

 

 

*

From LaFarge CG, Miettinen OS: The estimation of oxygen consumption. Cardiovas Res 4:23, 1970.

 

Vascular Resistance.

Vascular resistance (R) relates the mean pressure change (ΔP) across a circuit to the flow (   ) across the circuit, as follows:

3.7  

Poiseuille's Law relates flow to pressure, cross-sectional area, length, and viscosity of fluid and is defined by the following equation:

3.8  

where    is flow of volume, Pi - Po is inflow pressure minus outflow pressure, r is the radius of the tube, h is viscosity of the fluid, and l is the length of the tube.

In this equation, l, h, and π are constant; therefore    is directly proportional to the change in pressure multiplied by the fourth power of the radius (r4) of the tube. If flow (   ) remains constant, resistance increases when there is a large drop in pressure (ΔP) across a vascular bed. Poiseuille's Law assumes nonpulsatile laminar flow through rigid tubes, which is not completely analogous to the dynamic nature of the cardiovascular system. However, the calculation of resistance is helpful and the following equations are used.

3.9  

3.10  

where PCWP is pulmonary capillary wedge pressure.

Resistance is measured in Woods units (mm Hg/L per minute) and is converted to metric units by multiplying by 80 and expressed as dynes/sec per cm-5. Resistances are often indexed to body surface area by using CI, not CO. Normal PVR is less than 2 Woods units in older children but higher in neonates, a reflection of the anatomic and physiologic changes in cardiopulmonary maturation ( Emmanouilides, 1964 ; Rudolph and Nadas, 1962 ). SVR is 10 to 15 Woods units in neonates and increases to 20 Woods units during infancy and then remains at that level ( Rudolph, 1974a ).

Interpretation of Cardiac Catheterization Data

Evaluation of a patient with congenital heart disease should include a careful review of the most recent catheterization data. In summary, the following information should be available for evaluation.

  

1.   

Anatomic diagnosis: This is confirmed with hemodynamic data, oxygen saturation data, angiocardiographic evaluation, or a combination.

  

2.   

Hemodynamic data: Important data include baseline oxygen saturation and routine intracardiac pressure measurements, valve gradients, shunt calculation (including the direction and quantity of shunt flow), systemic and pulmonary flow, and vascular resistance measurements.

  

3.   

Response to sedation, anesthetic agents, or both: This is the source of important information during a preanesthetic evaluation.

  

4.   

Response to oxygen: In patients with elevated PVR (usually with large left-to-right shunts), the response of pulmonary artery pressure, PVR, and shunting after the administration of oxygen is valuable information. Patients with “fixed” or irreversible vascular changes in the lung may not show the expected pulmonary vasodilatation with oxygen administration (i.e., a decrease in pulmonary artery pressure as well as an increase in the left-to-right shunt). These patients may not be candidates for surgical repair of intracardiac shunts in the face of irreversible pulmonary vascular obstructive disease.

  

5.   

Effects of dysrhythmias on cardiac function: The presence of intracardiac catheters often induces dysrhythmias, and the hemodynamic response to dysrhythmias or to treatment is valuable information for the anesthesiologist.

Echocardiography

Echocardiographic evaluation of the cardiovascular system has been a revolutionary advancement in the assessment of congenital heart disease and pediatric cardiac function. Many children with congenital heart disease may proceed to surgical repair without the need for additional invasive cardiac catheterization because of the precise anatomic information made available by two-dimensional echocardiography. Because infants and children have thin chests and excellent “echo windows,” the size, location, orientation, and pattern of motion of all cardiac structures can be visualized in greater detail than in adults. Doppler ultrasound adds to the investigative capabilities of echocardiography, permitting assessment of blood flow to detect patterns, shunting, and valvular gradients. In addition, transesophageal echocardiography (TEE) has been used increasingly in congenital heart disease patients undergoing surgery as small probes have become available.

The evaluation of cardiac function and especially the effects of anesthetics on pediatric cardiac function cannot rely on the use of invasive monitors such as Swan-Ganz catheters, as used in many adult patients. Echocardiography has been extremely useful in children as a noninvasive monitor for cardiac function. Its applications and the limitations are reviewed here.

Echocardiography is the use of reflected ultrasound to create images of the heart and its structures. A pulse generator, timer, transducer and image processor, and display screen are the components of an echocardiographic machine. Electrical pulses are sent from the pulse generator to the transducer. The transducer emits a burst of sound energy and then acts as a receiver and detects the reflected sound. The sound energy reflected back is translated into an electrical impulse and then sent to an image processor.

M-mode echocardiography is the use of a thin beam of sound energy directed toward the heart. Only a small wedge of the heart is viewed with M-mode echocardiography, and, as a result, anatomic diagnosis is limited. However, temporal and spatial resolution of M-mode echocardiography permits the accurate measurement of changes in heart chamber or wall thickness size. As a result, M-mode is used to measure ventricular size and function ( Fig 3-27 ).

 
 

FIGURE 3-27  Schematic representation of M-mode (A) and two-dimensional (B) echocardiography. RVW, right ventricular wall; RV, right ventricle; IVS, intraventricular septum; LVS, left ventricular stress; LVD, left ventricular dimension; PW, posterior wall; MV, mitral valve; AV, atrial valve; Ao, aorta; LA, left atrium; LV, left ventricle; RV, right ventricle; RCA, right coronary artery; PA, pulmonary artery; LCA, left coronary artery; ANT, anterior; R, right; L, left; POST, posterior; SUP, superior; INF, inferior; TV, tricuspid valve; RA, right atrium; LA, left atrium.  (From Cassell ES, Rogers MC, Zahka KG: Developmental physiology of the cardiovascular system. In Rogers MC, editor: Textbook of pediatric intensive care. Baltimore, 1987, Williams and Wilkins.)

 

 

 

Two-dimensional echocardiography produces a cross-sectional view of the heart. The two-dimensional image results from a sound beam being directed through an arc, in contrast to the ice-pick view of the heart in M-mode. The diagnosis of anatomic abnormalities is superb with two-dimensional echocardiography, and this modality has supplanted cardiac catheterization in the diagnosis of many defects.

Doppler echocardiography uses continuous-wave Doppler and pulsed-wave Doppler. Continuous-wave Doppler is helpful in diagnosing stenotic lesions, atrioventricular valve regurgitation, and some shunt lesions. Pulsed-wave Doppler and Doppler color flow mapping are useful in the description of shunt lesions.

The value of intraoperative transesophageal echocardiography (TEE) has been evaluated at several centers ( Ungerleider et al., 1989 ; Sutherland et al., 1989 ; Muhiudeen et al., 1992 ). Inaccurate preoperative diagnosis, inadequate surgical repair, or both are major reasons for difficulties in weaning from bypass in congenital heart disease patients. If a postbypass echocardiogram demonstrates good repair, the long-term outcome is shown to be good in 93% of patients in contrast to 55% when a postrepair study demonstrates a suboptimal repair ( Ungerleider et al., 1990 ). In addition, assessment of ventricular function in the operating room carries predictive value for the postoperative period in general. Potential difficulties with TEE monitoring have been described and include the potential for airway obstruction and aortic compression, especially in small infants ( Strafford et al., 1994 ).

A combined echo-Doppler technique can be a valuable monitor of continuous cardiovascular changes. Using a transesophageal echo-Doppler probe, changes in aortic blood flow were shown to agree with corresponding changes in CO measured intermittently with thermodilution CO. With the combined echo-Doppler technique, the Doppler beam can be properly positioned with M-mode echocardiography so that the aortic wall and aortic cross-sectional area are continuously measured ( Odenstedt et al., 2001 ). Invasive monitoring during general anesthesia is not routine in healthy infants, and as a result, invasive evaluation of the use of anesthetic agents on cardiovascular function in healthy children is usually not justified. However, the measurement of continuous aortic blood flow with an esophageal echocardiographic probe has been used in infants as a less invasive tool for measurement ( Gueugniaud et al., 1997 ). Aortic blood flow, preejection period, left ventricular ejection time, mean arterial blood pressure (MAP), HR, SV, and SVR can be obtained with this probe. A study in 12 healthy infants aged 8 to 26 months showed reliable measurements after easy positioning of the probe ( Gueugniaud et al., 1998 ).

Automated real-time echocardiographic assessment or acoustic quantification is an advance in cardiac imaging. An automated left ventricular border detection system records beat-to-beat changes in left ventricular cavity area and fractional area change. This real-time assessment of left ventricular function became clinically available since the 1990s and has proved to be helpful in the quantification of left ventricular function ( Cahalan et al., 1993 ). Normal values of left ventricular systolic and diastolic function were defined using acoustic quantification in a multicenter study ( Spencer et al., 2003 ). This study examined adolescent and adult patients (aged 16 to 78 years). Of interest, the percentage of contribution to total left ventricular filling occurring during atrial filling nearly tripled during the six decades studied, from 13% in the youngest cohort to 36% in the eighth decade of life.

The smaller size of the LV in younger children was initially thought to predispose to greater measurement errors in younger children using acoustic quantification. However, the reliability and accuracy of automated border detection using acoustic quantification in children were determined by Rein and others (1998) and appears to be an acceptable method for estimating the cross-sectional area and fractional area change of the LV in children. Other, more objective evaluations of ventricular function have been developed. Tissue Doppler imaging uses Doppler color flow technology to evaluate myocardial velocity with the use of two-dimensional and M-mode echocardiography ( Rychik, 1996 ; Miyatake et al., 1995 ). Tissue velocity Doppler has also been used in the assessment of atrial and ventricular electromechanical coupling and atrioventricular time intervals in children ( Rein et al., 1998 ). The ability to simultaneously analyze mechanical events and electromechanical coupling in the atria and ventricles is very helpful in rhythm analysis. As a result tissue Doppler may have a role to play in rhythm diagnosis noninvasively when invasive electrophysiologic studies are difficult, such as for fetal arrhythmias. Color kinesis imaging is a visually enhanced mode of automated border detection in which sonographic backscatter analysis is used to color code blood and myocardium interfaces, which are then integrated over the cardiac cycle and analyzed to assess wall motion ( Mor-Avi et al., 1997 ).

Evaluation of Cardiac Function Using Echocardiography

The description of cardiac function earlier in the chapter defined the three major determinants of cardiac performance: preload, afterload, and contractility. In adults, thermodilution CO and pulmonary artery occlusion pressure have historically been used to assess cardiac function, although these methods also have limitations. Echocardiography has proved to be increasingly reliable in the noninvasive assessment of left ventricular function, especially in pediatric patients. For example, previous treatment with anthracylines, a group of chemotherapeutic drugs in use for childhood cancer, may enhance the myocardial depressive effect of anesthetics even in children and adolescents with normal resting cardiac function. Many of these cancer survivors have subtle cardiac abnormalities that are evident with exercise. The stress of anesthesia may also unmask these abnormalities. Huettemann and others (2004) noted that children who had undergone chemotherapy and were anesthetized with 1 MAC isoflurane along with 70:30% nitrous oxide/oxygen had significantly decreased cardiac function even though resting cardiac function was normal. This decrease in function with anesthesia was significantly different than that in the control group of children.

Preload.

As defined earlier, preload is the stretching force put on a muscle fiber in the relaxed state. In the intact heart, the end-diastolic fiber length is equated with EDV and left ventricular EDV is generally assumed to be a measurement of preload. EDP is equated with EDV, which is probably a valid assumption when ventricular compliance is normal. However, measurement of ventricular preload using left ventricular EDP is probably less accurate in patients with mitral regurgitation, abnormal ventricular compliance, or both. Two-dimensional echocardiography can be used to measure left ventricular EDV directly. Although this method may be time consuming, the experienced echocardiographer can detect mild blood volume reductions by monitoring left ventricular short-axis changes with high sensitivity (80% to 95%) and specificity (80%). In the study by Reich and others (1993) , TEE was used to accurately monitor cardiac filling changes in pediatric cardiac surgical patients.

Afterload.

The clinical measurement of afterload (the stress imposed on the ventricular wall during systole) has been more difficult to obtain. SVR is often used as a measurement of afterload, but SVR is derived from the measurement of MAP, right atrial pressure, and CO, as follows:

3.11  

Each of these measurements has potential for error.

Left ventricular end-systolic wall stress (ESWS) is considered a better reflection of afterload because it includes both peripheral loading conditions and myocardial factors. The stress that the ventricle faces at the end of systole is probably the most accurate measurement of afterload. At the end of contraction, the force resisting further shortening determines when shortening ceases.

ESWS is the measurement of this force and is a clinically relevant measure of afterload. ESWS is the force per unit area within the ventricular wall and is determined by combining arterial pressure, phonocardiographic, and echocardiographic measurements ( Colan et al., 1984 ). ESWS has found increased applicability as a possible factor influencing myocardial oxygen consumption, and multiple clinical studies in adults have examined this effect ( O'Kelly et al., 1991 ; Goertz et al., 1993 ). On the other hand, wall stress may be a poorer reflection of afterload in children and young adults who have abnormal left ventricular geometry such as patients with valvar aortic stenosis, coarctation, and mitral and aortic regurgitation, as well as anthracycline-treated patients ( Gentles and Colan, 2002 ).

Contractility.

Contractility is a measurement of intrinsic properties of cardiac muscle that do not include afterload, preload, or both. The usual measures of ventricular performance described on echocardiographic data include shortening fraction and ejection fraction (EF) data, as follows:

3.12  

Normal values are between 65% and 80%, depending on the method used to measure systolic and diastolic volumes ( Guttgesel et al., 1977 ). EFs do not change significantly with age but are affected by changes in preload and afterload. An increase in preload or a decrease in afterload increases the EF if there is no simultaneous change in contractility ( Fisher et al., 1975 ; Sonnenblick and Stobeck, 1977 ), as seen here:

3.13  

Shortening fraction is similar to EF but shortening fraction does not rely on the calculation of ventricular volumes as does EF. A normal shortening fraction is 36%, with a range of 28% to 44% ( Guttgesel et al., 1977 ).

The velocity of circumferential fiber shortening (Vcf) is a measurement of both the extent and the rapidity of ventricular fiber shortening, as follows:

3.14  

The velocity of circumferential shortening is sensitive to changes in afterload but not to changes in preload. Vcf decreases with increasing afterload. In addition, Vcf increases with positive inotropic therapy, such as isoproterenol, and decreases with propranolol ( Mahler, et al., 1975) . HR also affects Vcf. The younger child with a normally higher HR has a different “normal” Vcf than do older children (Guttgesel et al., 1977 ). Normal Vcf for a child of a given age and HR can be calculated as follows:

3.15  

Because of the influence of preload and afterload on these measurements, there are limitations to the use of shortening data. For example, children with chronic renal failure have depressed shortening fraction data on echocardiographic examination ( Colan et al., 1987b ). However, Colan and others have shown that these depressed shortening data were due entirely to altered load rather than abnormal contractility and were reversible with improved renal function. Studies on left ventricular mass and systolic performance in pediatric patients with chronic renal failure have shown that those on chronic dialysis do have increased left ventricular mass, left ventricular performance, and contractility at rest but decreased contractile reserve on exercise, which may portend the development over time of worsening systolic function ( Mitsnefes et al., 2003 ). When athletes are studied, endurance athletes may show reduced shortening data, but again this is due to altered load and not reduced contractility (Colan et al., 1987a ). Children with Duchenne's muscular dystrophy were studied before scoliosis repair. Percent of fiber shortening was depressed in these children. However, using the stress-velocity relationship, reduced systolic performance was due to excess afterload (elevated end-systolic stress) without significant reduction in contractility (normal stress-velocity relationship). This finding in Duchenne's patients can be explained by reviewing the pathologic results. The myocardium is characterized by fatty deposits and myocardial filament drop-out. The echocardiographic findings support reduced working muscle, but the remaining muscle fibers do have normal systolic function.

To return to our developmental assessment of cardiac function, infants and young children have been found to enhance systolic performance using shortening fraction data alone. When fractional shortening and velocity of shortening are examined in normal children, an age-related decline in performance is noted. However, much of this observed change is due to increased afterload with age ( Fig 3-28 ) ( Colan et al., 1989 ). When the stress-shortening relation is examined, contractility still decreases, mainly over the first 2 years of life ( Fig 3-29 ).

 
 

FIGURE 3-28  In normal subjects, there is a significant age-related rise in afterload. ESSm, mean end-systolic stress.  (From Colan SD: Assessment of ventricular and myocardial performance. In Flyer DC, editor: Nadas—pediatric cardiology. Philadelphia, 1992, Hanley and Belfus.)

 

 

 

 
 

FIGURE 3-29  The stress-shortening relation falls with age with the most prominent effect in the first two years of life. This is consistent with age modulation of contractility. ESS, end-systolic stress; VCFc, velocity of shortening.  (From Colan SD: Assessment of ventricular and myocardial performance. In Flyer DC, editor: Nadas—pediatric cardiology. Philadelphia, 1992, Hanley and Belfus.)

 

 

 

Research into defining a load-independent measure of contractility has been active. When the length of an isolated muscle strip is held constant, force and velocity are inversely correlated. In the intact heart, this relationship can be defined using arterial pressure and echocardiographic parameters. Colan and others (1984) used the relationship of left ventricular wall stress to the velocity of circumferential fiber-shortening corrected for HR (Vcfc [velocity of circumferential shortening corrected for HR]) as a measure of myocardial contractility ( Fig 3-30 ). Unlike EF and circumferential fiber shortening percentage, which are significantly affected by changes in preload, a change in Vcfc is independent of preload. In fact, this relationship has been used in studies on the effects of anesthetics on myocardial contractility.

 
 

FIGURE 3-30  An inverse linear relationship exists between end-systolic stress (ESS) and velocity of shortening (VCFc) as demonstrated in this graphic representation of data from a large number of individuals.  (From Colan SD: Assessment of ventricular and myocardial performance. In Flyer DC, editor: Nadas—pediatric cardiology. Philadelphia, 1992, Hanley and Belfus.)

 

 

 

Another important relationship in examinations of contractility is the end-systolic pressure-volume relationship. Suga and others (1973) have shown that the relationship end-systolic pressure can be approximated by the following equation:

where Pes is end-systolic arterial blood pressure, Emax is slope of end-systolic pressure-volume relationship, Ves is end-systolic volume, and Vd is the intercept of the end-systolic pressure-volume relationship line on the horizontal axis. They found that Emax is independent of preload and afterload and is an excellent indicator of contractility.

The wall stress-velocity and the end-systolic pressure-volume relationships are important parameters that can be measured noninvasively and add to our clinical understanding of an anesthetic agent's effects on cardiac function. These load-independent contractile indices should find increasing applicability in research describing the effects of anesthetics on pediatric cardiac function.

The echocardiographic assessment of cardiac function is important to review because the effects of anesthetics on cardiac function in children are often assessed using such noninvasive monitors. Clearly, conclusions regarding an anesthetic's effect on contractility must consider whether reliable indices have been studied, preferably load-independent indices. When the validity of echocardiographic measurements is understood, a conclusion regarding a drug's effect on cardiac function can be made more accurately.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ EFFECTS OF ANESTHESIA ON THE CARDIOVASCULAR SYSTEM

The effects of anesthesia on the cardiovascular system must also be considered with a developmental framework. Studies examining the effects of different agents on cardiac function have helped to define age-related responses. Proper anesthetic management must carefully assess the risks of cardiac depression in different age groups and for different surgical procedures. A high-dose opioid technique may help maintain hemodynamic stability in a patient with congenital heart disease who is undergoing cardiac surgery but is an unacceptable technique if direct laryngoscopy and bronchoscopy or a short outpatient procedure is to be performed. The potent effects that anesthetics have on the respiratory system may also cause important interactions with normal cardiovascular function and must be considered, especially in neonates and infants or children with congenital heart disease. Finally, studies examining the cardiovascular effects of anesthetic agents in adults often rely on the availability of invasive monitors, such as pulmonary artery catheters and CO monitoring, which are inappropriate for or difficult to use in pediatric patients. Echocardiographic assessment of cardiac function has been used extensively to define in a more sophisticated and accurate way the effects of anesthesia on the cardiovascular system in children.

▪ PREMEDICATION AND THE INDUCTION OF ANESTHESIA

A safe and effective anesthetic induction must consider psychological and developmental issues as well as the physiologic effects of the agents used. In addition to the use of effective sedative or hypnotic agents, careful preoperative education of parents and patient as well as parental presence during induction may provide for a smooth and safe induction. Since the 1990s, there has been a dramatic increase in the role of parental presence during induction and the use of premedication. Midazolam had become the most commonly used premedication ( Kain et al., 2004 ) (see Chapters 6 and 7 , Pharmacology of Pediatric Anesthesia, and Psychological Aspects). Induction techniques may vary, but the end result should be a calm patient with minimal hemodynamic stress, optimal airway control, and maintenance of cardiovascular stability.

Agents for Premedication and Induction

The importance of a smooth, calm induction in children has been recognized as an essential part of an effective anesthetic plan. Psychological benefits may seem obvious, but a smooth induction in a calm, cooperative, or sedated child may also minimize disturbances during induction secondary to increased airway secretions and agitation.

Cardiorespiratory effects of premedication in normal children were studied using three different oral, nasal, and rectal premedication regimens ( Audenaert et al., 1995 ). Fifty-eight young children (average age, 2.7 years) were studied. Oral meperidine (3 mg/kg) with pentobarbital (4 mg/kg) decreased HR, MAP, CI, respiratory rate, and oxygen saturation. SV was maintained. Nasal ketamine (5 mg/kg) with midazolam (0.2 mg/kg) produced no significant cardiovascular or respiratory effects. Rectal methohexital (30 mg/kg) increased HR with a coincident decrease in SV but had no other positive or negative cardiac or respiratory effect.

Methohexital

Age-related differences have been described with the use of barbiturates. Animal studies suggest a lower LD50 for barbiturates in young animals ( Carmichael, 1947 ; Domek et al., 1960 ). Differences in metabolism, including glucuronic acid conjugation and liver immaturity, may have potent effects on the pharmacologic aspects of this group of drugs in neonates and young infants.

Rectal methohexital is used as a premedication in young children. The effect of methohexital on cardiac function has been studied in children with normal cardiac function ( Audenaert et al., 1992 ), using echocardiographic evaluations preoperatively and after rectal administration of 30 mg/kg of methohexital. HR increased and SV decreased, but BP and CI showed no significant changes. Shortening fraction and EF remained within normal parameters. Because baseline measurements were taken the day before surgery and blood levels of additional doses of methohexital were administered if sleep did not occur, the impact of fasting and different serum levels might have affected these results. However, rectally administered methohexital appears to have minimal cardiovascular side effects. Because respiratory depression must be considered a potential side effect of sedation with any barbiturate, the effects of airway compromise on cardiopulmonary interactions cannot be minimized, especially in young infants.

Midazolam and Diazepam

The benzodiazepines have been used widely for premedication and sedation via various routes. Midazolam has been shown to be effective as an induction agent while maintaining cardiovascular stability in adults ( Gamble, 1981) . Midazolam has found a significant place in premedication and sedation for procedures and intensive care unit sedation in children of all ages, and further research has elucidated its effects. The cardiovascular effects in postoperative cardiac surgery patients have been examined ( Shekerdemian et al., 1997 ). Ten hemodynamically stable children, ventilated in the early postoperative period after cardiac surgery and receiving intravenous morphine infusions, were given an intravenous bolus followed by a continuous infusion of midazolam. Hemodynamic data were recorded before the bolus and 15 minutes and 1 hour later. A bolus of midazolam lowered the CO by 24.1%. Arterial BP, oxygen consumption, and mixed venous oxygen content decreased. There was a tendency for all variables subsequently to recover toward baseline values, within 1 hour, during a continuous infusion. An intravenous bolus of midazolam causes a decrease in CO. Continuous infusions may confer greater cardiovascular stability than intermittent boluses, especially in the compromised cardiac patient.

The standard dose of oral midazolam has been 0.5 to 1.0 mg/kg. The safety and efficacy of a higher oral dose, 1.5 mg/kg, compared with 0.5 and 1.0 mg/kg were studied in 193 children (aged 4 months to 2 years) undergoing cardiovascular surgery ( Masue et al., 2003 ). Midazolam 1.5 mg/kg did not cause any statistically significant decrease in BP, HR, or SpO2, although eight infants and children showed a 20% decrease in systolic BP and six infants and children showed a greater than 5% decrease in SpO2. No “spelling attacks,” seizure-like activity, apnea, or laryngospasm was observed in any infants and children during and after the medication.

Midazolam has also been found to be an important medication in the management of agitation and distress in the pediatric intensive care unit setting. Sedation in the intensive care unit may be needed for short interventions during difficult procedures or for continuous periods of assisted ventilation. Midazolam, with its characteristic of water solubility (unlike diazepam), short elimination half-life, and short duration of action, has been used with lorazepam to manage sedation in the pediatric intensive care unit. Abrupt cessation of therapy may result in withdrawal symptoms and must be anticipated and appropriate weaning schedules planned.

Midazolam has also been studied as an induction agent and compared with thiopentone and propofol ( Jones et al., 1994 ). Thirty children undergoing circumcision were randomized to receive either thiopentone 4 mg/kg, propofol 2.5 mg/kg, or midazolam 0.5 mg/kg (n = 10 each) intravenously over 30 seconds for induction. There was no statistically significant hemodynamic difference between the three induction agents. Propofol caused a greater decrease in MAP compared with thiopentone at 1 minute (P = .01), and the MAP remained significantly lower than that with midazolam at 5 minutes. Of the three induction agents, thiopentone caused the least hemodynamic disturbance on induction.

▪ MAINTENANCE OF ANESTHESIA

Inhalational Anesthesia

Halothane and Isoflurane

Inhalational anesthesia remains the most common method of anesthesia for pediatric patients. Unlike adult patients, halothane continued to hold a significant role in pediatric anesthesia. The introduction of sevoflurane since the 1990s has resulted in some change in choice of agents mainly because of sevoflurane's low blood-gas partition coefficient and low airway irritability, resulting in smooth conditions for rapid induction of anesthesia. It is well recognized that neonates and infants experience a higher incidence of bradycardia, hypotension, and cardiac arrest than older patients undergoing inhalational anesthesia ( Nicodemus et al., 1969 ; Friesen and Lichtor, 1982 ; Diaz, 1985 ). Animal studies show a dose-related depression in cardiac function in young animals compared with adults for both halothane and isoflurane ( Boudreaux et al., 1984 ; Murat et al., 1990 ). The depressant effects of halothane have been described in newborns and young infants. The hemodynamic effects of inhalational anesthesia were defined using basic parameters such as BP and HR, and echocardiographic assessment of cardiac function has refined these measurements.

Investigators using M-mode echocardiography have shown that halothane increases the preejection period (isovolumic contraction time), decreases the fraction of left ventricular shortening fraction, and increases the systolic time interval (preejection period/left ventricular ejection time). In contrast, isoflurane decreases the preejection period, maintains left ventricular shortening fraction, and shortens the systolic time interval. These M-mode measurements indicate a greater decrease in contractility associated with halothane compared with isoflurane ( Wolf et al., 1986 ). However, a limitation in these M-mode results is the assumption that preload, afterload, HR, and cardiac conduction all remain constant.

The cardiac-depressant effects of halothane and isoflurane have also been examined using more definitive function parameters, such as pulsed Doppler and two-dimensional echocardiographic measurements as well as the addition of a fluid bolus challenge at three different anesthetic levels: 0.75, 1.0, and 1.25 MAC ( Murray et al., 1987 ). Halothane and isoflurane both decreased mean BP. Halothane decreased HR at 1.25 MAC, whereas isoflurane increased HR at all levels. Cardiac index was decreased with both agents at 1.25 MAC. EFs decreased significantly with both agents at 1.0 and 1.25 MAC. After a fluid bolus of lactated Ringer's solution (15 mL/kg), EF and SV index increased significantly in the isoflurane group but decreased significantly in the halothane group, suggesting a diminished cardiovascular reserve in the halothane-anesthetized group. This response to fluid loading may have important implications in the clinical setting.

Continuous esophageal aortic blood flow echo-Doppler has been used in healthy infants and small children to examine the myocardial effects of isoflurane ( Gueugniaud et al., 1998 ). Twenty-five healthy infants were studied; they had significant decreases in aortic blood flow and increased preejection period/left ventricular ejection time compared with control values 5 minutes after induction with halothane-fentanyl and atracurium. When isoflurane was discontinued, these changes were reversed. A 1% end-expired concentration of isoflurane caused no significant changes in HR but moderately decreased MAP.

The cardiovascular effects of sevoflurane, isoflurane, halothane, and fentanyl-midazolam have been studied in children with congenital heart disease ( Rivenes et al., 2001 ). Fifty-four patients younger than 14 years who were scheduled to undergo congenital heart surgery were randomized to receive halothane, sevoflurane, isoflurane, or fentanyl-midazolam. Cardiovascular and echocardiographic data were recorded at baseline and at randomly ordered 1 and 1.5 MAC, or predicted equivalent fentanyl-midazolam plasma concentrations. Halothane caused a significant decrease in MAP, EF, and CI, preserving only HR at baseline levels. Fentanyl-midazolam in combination caused a significant decrease in CI secondary to a decrease in HR; contractility was maintained. Sevoflurane maintained CI and HR and had less profound hypotensive and negative inotropic effects than halothane. Isoflurane preserved both CI and EF, had less suppression of MAP than halothane, and increased HR.

The effect of adding nitrous oxide during halothane and isoflurane anesthesia has also been studied in infants ( Murray et al., 1988 ). Baseline measurements were made in 19 healthy nonpremedicated infants (mean age, 12 months) with pulsed Doppler and two-dimensional echocardiography again at 1 MAC halothane or isoflurane and then at the addition of nitrous oxide. MAP, CI, SV, and EF decreased similarly and significantly at 1.0 MAC halothane and isoflurane. HR increased during isoflurane anesthesia but decreased during halothane anesthesia. The addition of nitrous oxide resulted in a decrease in HR, CI, and MAP compared with 1.0 MAC levels of halothane or isoflurane; however, SV and EF were not significantly changed from levels measured during 1.0 MAC halothane or isoflurane. The sympathetic stimulation seen in adults with nitrous oxide does not appear to be seen in infants and young children.

The role of HR in maintaining CO has also been examined in children undergoing inhalational anesthesia. Atropine, administered as a premedication or intraoperatively, has been shown to increase CO during halothane and nitrous oxide anesthesia ( Barash et al., 1978 ; Friesen and Lichtor, 1982 ; Miller and Friesen, 1988 ). Using pulsed Doppler and two-dimensional echocardiographic measurements, the effect of atropine on infants and small children undergoing anesthesia with 1.5 MAC halothane or isoflurane was studied ( Murray et al., 1989 ). Because this study examined hemodynamic effects at higher end-expired concentrations and for a longer period of time, halothane was shown to have greater decreases in EF and increases in left ventricular EDV compared with isoflurane.

The greater solubility of halothane compared with isoflurane would explain why evaluation of function parameters after only a short period of time might not reflect differences observed when higher myocardial levels have been attained after prolonged administration. Halothane and isoflurane had similar decreases in CO and SV. The use of atropine resulted in an increase in CO and SV in both groups but more significantly in the halothane group. Despite the greater increase in HR, halothane still produced greater increases in preload (left ventricular EDV) and decreases in EF than isoflurane. Atropine therefore increases CO by its effect on increasing HR but does not affect other cardiovascular effects of inhalational agents.

In summary, similar decreases in CO, SV, and EF are observed with equipotent concentrations of halothane and isoflurane. There is clearly a dose-related effect, with a 30% decrease noted at 1.5 MAC with both agents. HR is affected more by halothane than by isoflurane, but atropine may attenuate this effect. The accuracy of end-tidal measurements of inhalational agent concentrations may account for differences between agents described in other studies. Age has an important effect on MAC: as age decreases, MAC increases. Infants aged 1 to 6 months have a maximum MAC value, and this value decreases with age thereafter. Lerman and others (1983) have shown that the incidence of hypotension in neonates was similar to that in older infants when equipotent concentrations were inhaled.

The safety or therapeutic margin is a useful concept that defines the separation between MAC and a lethal concentration of an inhaled anesthetic, as follows:

3.16  

Isoflurane has been shown to have a higher therapeutic ratio than halothane ( Wolfson et al., 1973 , 1978; Kissen et al., 1983 ). Animal studies show that young rats exhibit a 50% decrease in the therapeutic ratio for halothane compared with older animals ( Cook et al., 1981 ). Isoflurane and halothane were found to have very similar therapeutic ratios in newborn animal studies ( Schieber et al., 1986 ).

The cardiovascular effects of inhalational anesthetics are also modulated by baroreceptor responses. As the earlier discussion outlined, many of these reflexes may be limited or absent in very young infants and newborns. Gregory (1982) has shown that despite an increase in systemic blood flow after ligation of a patent ductus arteriosus in premature infants under halothane anesthesia, HR did not increase. In animal studies, halothane and nitrous oxide have been shown to diminish the baroreceptor reflexes in a concentration-dependent manner. Limitations in baroreceptor responsivity may explain the well-described clinical phenomenon of an increased incidence of hypotension and bradycardia in very young infants and newborns under halothane anesthesia. Murat and others (1989) studied eight neonates during the administration of 1 MAC isoflurane. No other anesthetic was administered. The pressor response was tested with the use of phenylephrine, and nitroglycerin was used to test the depressor response. At 1 MAC, MAP decreased about 30% and the mean pressor response decreased to 23% of control awake values. The depressor response decreased to 28% of control. These changes could be attributed to a significant resetting of HR itself. The sensitivity of the baroreceptor reflex was unchanged. This study demonstrated that the significant depression of baroreflex control of HR may impair the newborn's ability to compensate for changes in arterial pressure or to maintain an adequate CO with hypovolemia.

Sevoflurane is a volatile inhalational anesthetic with a low blood-tissue solubility and limited cardiorespiratory depression. It has often been described as an ideal inhalational agent because of its physical, pharmacodynamic, and pharmacokinetic properties.

Sevoflurane confers cardiovascular stability in children, especially compared with other agents such as desflurane and halothane. Sevoflurane produces less increase in HR than isoflurane ( Frink et al., 1992) and less myocardial depression than halothane ( Holzman et al., 1996 ). Sevoflurane has been safely used during spinal surgery to control hypotension ( Tobias, 1998 ). At all concentrations in infants, sevoflurane caused less of a decrease in HR, myocardial contractility, and CO compared with halothane ( Wodey et al., 1997 ). Arrhythmias are also less common in children undergoing ear, nose, and throat surgery (61% for halothane and 5% for sevoflurane [ Johannesson et al., 1995 ]) and dental surgery (62% for halothane and 28% for sevoflurane [ Paris et al., 1997 ]).

Lerman and others (1994) described the pharmacology of sevoflurane in infants and children. The MAC of sevoflurane in neonates is 3.3%; in infants (aged 1 to 6 months), 3.2%; and in older infants (aged 6 to 12 months) and children (aged 1 to 12 years), 2.5% ( Lerman et al., 1994 ). In this study, systolic arterial pressure decreased significantly at 1 MAC before incision in all subjects except (1) children aged 1 to 3 years receiving 60% nitrous oxide and (2) children aged 5 to 12 years receiving sevoflurane with oxygen. Blood pressure returned to baseline after incision. HR was unchanged at 1 MAC in all patients except children older than 12 years, who had an increase in HR before incision. The cardiovascular effects of sevoflurane have been studied using transesophageal acoustic quantification (AQ). AQ is a computer-based automatic border detection method to describe echocardiographic images and real-time analysis of cardiac volume changes. AQ with TEE has the ability to detect small depressions in cardiac ejection performance in children undergoing sevoflurane anesthesia (aged 1.4 to 12 years). An increase in HR was balanced by a decrease in SVR ( Tanaka et al., 1994 ).

The MAC and hemodynamic effects of halothane, isoflurane, and sevoflurane have been studied in newborn swine ( Lerman et al., 1990 ). Compared with the awake HR, the mean HR decreased 35% at 1.5 MAC halothane, 19% at 1.5 MAC isoflurane, and 31% at 1.5 MAC sevoflurane. Compared with awake systolic arterial pressure, mean systolic pressure decreased 46% at 1.5 MAC halothane, 43% at 1.5 MAC isoflurane, and 36% at 1.5 MAC sevoflurane. Mean CI did not change significantly between awake and 1.5 MAC sevoflurane, whereas halothane and isoflurane caused significant changes (53% decrease at 1.5 MAC halothane and 43% decrease at 1.5 MAC isoflurane). At equipotent concentrations, halothane and isoflurane depressed hemodynamics to a greater extent than did sevoflurane.

Sevoflurane and halothane were also compared using echocardiographically derived indices of myocardial contractility during induction ( Holzman et al., 1996 ). Left ventricular end-systolic meridian wall stress increased with halothane but remained unchanged with sevoflurane. SVR decreased from baseline to 1 MAC and 1.5 MAC with sevoflurane. Halothane depressed contractility as assessed by the stress-velocity index and stress-shortening index, whereas contractility remained within normal limits with sevoflurane. Total minute stress and normalized total mechanical energy expenditure, measures of myocardial oxygen consumption, did not change with either agent.

Infants may be a greater risk of hemodynamic compromise with inhalational anesthetics ( Wodey et al., 1997 ). In a comparative hemodynamic study between halothane and sevoflurane in infants, sevoflurane showed less cardiac depression than did halothane. Sevoflurane did not alter HR or CI at all concentrations compared but did significantly decrease BP and SVR compared with awake values at all concentrations. Shortening fraction and rate-corrected velocity of circumferential fiber shortening decreased at 1.5 but not at 1 MAC. Myocardial contractility assessed by stress-velocity index and stress-shortening index decreased significantly but not to any abnormal value at all concentrations. Halothane caused a greater decrease in HR, shortening fraction, stress-shortening index, velocity of circumferential fiber shortening, stress-velocity index, and CI at all concentrations compared with sevoflurane.

The use of sevoflurane versus halothane in children has been studied with particular attention to electroencephalograms, clinical agitation, and autonomic cardiovascular activity ( Constant et al., 1999 ). Sevoflurane induced a greater withdrawal of parasympathetic activity than halothane and a transient relative increase in sympathetic vascular tone at the time that the eyelash reflex was lost.

Induction with 8% sevoflurane in children has been studied ( Wappler et al., 2003 ) and has been shown to be effective in creating ideal conditions, including hemodynamic stability, for endotracheal intubation without the use of muscle relaxants.

Desflurane has a lower blood-gas and tissue-blood partition coefficient than isoflurane. Rapid induction and emergence would thus be expected, although airway irritability is notable with desflurane and limits its use as an induction agent ( Taylor and Lerman, 1992 ; Zwass et al., 1992 ). MAC and hemodynamic responses in neonates, infants, and children have been studied ( Taylor and Lerman, 1991 ) and showed that the MAC of desflurane depends on age, but the age-related differences are much less than those observed with halothane and isoflurane. HR decreased an average of 16% before skin incision in infants aged 6 to 12 months and children aged 1 to 3 years and 3 to 5 years, but no significant change was observed in other age groups.

In a multicenter study examining induction and maintenance characteristics of anesthesia with nitrous oxide and desflurane in children, MAP of less than 80% of baseline was more common with halothane. However, HR and MAP of greater than 120% of baseline was more common with desflurane. Airway irritability, including laryngospasm, limited the use of desflurane as an induction agent ( Zwass et al., 1992 ).

Nitrous Oxide

The effects of nitrous oxide on cardiovascular function have been studied in infants after surgical repair of congenital heart defects. Administration of nitrous oxide to infants with normal and elevated PVR revealed a 9% decrease in HR, a 12% decrease in MAP, and a 13% decrease in CI in both groups. The mild depressant effects of nitrous oxide on systemic hemodynamics are similar to those described in adults. However, reports of elevations in pulmonary artery pressure and PVR in adults were not observed in this group of infants ( Hickey et al., 1986 ).

Nitrous oxide permits lower concentrations of other inhalational agents to maintain a similar depth of anesthesia. The cardiovascular effects of nitrous oxide during inhalational anesthesia with isoflurane and halothane have also been studied in infants and small children ( Murray et al., 1988 ). Using two-dimensional and pulsed Doppler echocardiographic measurements, the effects of nitrous oxide were studied during halothane and isoflurane anesthesia. CO decreased significantly during both halothane and isoflurane anesthesia with and without nitrous oxide. The addition of nitrous oxide to halothane and isoflurane decreased HR and led to decreased CO. Unlike adults, infants and small children do not show the effects of sympathetic stimulation seen with nitrous oxide. The addition of nitrous oxide may have beneficial effects, such as rapid uptake and distribution because of its low solubility, minimal odor, and enhanced alveolar delivery of other inhalational agents. However, the addition of nitrous oxide does not appear to confer added cardiovascular protection from the depressant effects of inhalational agents. As investigative methods have become more sophisticated and accurate, especially the use of echocardiography, the cardiovascular effects of all anesthetic agents can be studied more accurately and direct the choice of safe anesthetics, especially in high-risk pediatric patients.

Opioids

Opioids, especially in high doses for cardiac surgery, have been widely used in children of all ages. Sufentanil, fentanyl, isoflurane, and halothane have been studied in pediatric patients undergoing cardiac surgery. Cardiovascular function was measured by echocardiography before induction, after induction, and after intubation. Left ventricular EF, systemic arterial pressure, and HR were recorded. Left ventricular EF decreased with each agent: sufentanil, 9%; fentanyl, 9%; isoflurane, 4%; and halothane, 8%. Left ventricular EF increased after intubation in all groups except the halothane group, in which left ventricular EF remained 13% below baseline ( Glenski et al., 1988 ).

Pulmonary and systemic hemodynamic responses to 25 mcg/kg of fentanyl were examined in 12 infants after repair of congenital heart defects ( Hickey et al., 1985a ). No significant changes were found in HR, CI, mean pulmonary artery pressure, or PVR. There were small but statistically significant decreases in MAP and SVR. The use of high-dose opioid technique in neonates undergoing cardiac surgery has been shown to blunt the hormonal and metabolic responses to stress, which may affect postoperative morbidity and mortality ( Anand and Hickey, 1992 ). High-dose opioid anesthesia may also affect the incidence of ventricular fibrillation in susceptible infants with HLHS who are undergoing cardiac surgery ( Hansen and Hickey, 1986 ; Hickey and Hansen, 1991 ). Stress responses in the pulmonary circulation (during endotracheal suctioning for example) can have potent effects on cardiovascular stability. Fentanyl (25 mcg/kg) has been shown to blunt increases in mean pulmonary artery pressure and PVR (Hickey et al., 1985a, 1985b [75] [77]).

Sufentanil has been studied in children undergoing cardiac surgery. Fentanyl (50 to 75 mcg/kg) was compared with sufentanil at two doses of 5 and 10 mcg/kg. Hemodynamic responses were similar with each agent, and cardiovascular stability was maintained, lowering PVR and increasing oxygen saturations in cyanotic patients ( Hickey and Hansen, 1984 ). Davis and others (1987) studied a high-dose sufentanil technique (15 mcg/kg) in pediatric cardiac surgery patients and described similar hemodynamic responses and the maintenance of cardiovascular stability.

Alfentanil has been examined in greater detail in adult patients, using high- and low-dose infusions. At low infusion rates (1.6 and 6.4 mcg/kg), no significant hemodynamic changes were noted. At higher rates (150 mcg/kg), HR, MAP, and SVR decreased and pulmonary capillary wedge pressure, PVR, and pulmonary artery pressure increased slightly ( Kay and Stephenson, 1980 ; Kramer et al., 1983 ).

The cardiovascular effects on children of the administration of morphine, meperidine, methadone, and remifentanil have not been studied in children undergoing cardiac surgery in as much detail as have the effects of high-dose fentanyl and sufentanil. All opioids cause a shift to the right in ventilatory response. This potent effect on the respiratory system plays an important interactive role with the cardiac system and should be considered whenever opioids are used. The pharmacokinetics of various opioids in pediatric patients should be reviewed before use (see Chapter 6 , Pharmacology for Pediatric Anesthesia).

Propofol

Propofol is a short-acting hypnotic with a rapid redistribution and metabolism. It has been used increasingly in short procedures or for sedation of limited duration. Although apnea can occur during induction, minimal respiratory depression has been observed after induction. Hannallah and others (1991) studied the ED50 and ED95 for loss of eyelash reflex and found the dose to be 1.3 and 2.0 mg/kg, respectively. Blood pressure decreased 20% in almost half (48%) of the children who received 1% to 3% halothane and propofol infusion.

Propofol has also been studied in children undergoing cardiac catheterization. In this study, children were randomly assigned to receive propofol or ketamine. Ketamine was given in an induction dose of 2 mg/kg IV followed by an infusion at 2 mg/kg per hour. The propofol group received 0.5-mg/kg boluses every 60 seconds titrated to an appropriate sedation level. An infusion of propofol was then started at an hourly infusion rate of three times the induction dose. The propofol group experienced significantly greater decreases in MAP (>20% from baseline). Several patients administered ketamine had episodes of increased HR and arterial pressure. There were significant desaturation effects in the propofol group. The slow titration of propofol used in this study is quite different from standard induction techniques using 2 to 2.5 mg/kg. The hemodynamic effects of a slower induction regimen may differ significantly (Lebovic et al., 1992).

In another study, 216 children were randomly allocated to receive one of six different doses of propofol, from 1.6 to 2.6 mg/kg in 0.2-mg/kg increments. MAP was reduced 15% after 1 minute and 30% after 5 minutes. HR decreased about 17% ( Short and Aun, 1991 ). When thiopentone (5 mg/kg) and propofol (2.5 mg/kg) were compared as induction agents using echocardiographic measurements, MAP was significantly reduced in the propofol group, but the reduced CI did not differ between the two agents. Aun and others (1993) noted that the baroreflex-mediated increases in HR and SVR were less after propofol than after thiopentone. Hannallah and others (1994) studied hemodynamic changes during induction with four different induction/maintenance regimens—(1) propofol/propofol infusion, (2) propofol/halothane, (3) thiopentone/halothane, and (4) halothane/halothane—and noted no significant hemodynamic changes between the groups.

Pediatric anesthesiologists have found an increasing role in the sedation of pediatric patients undergoing cardiac catheterization. Hemodynamic stability is paramount for the patient's safety but also for the accuracy of hemodynamic measurements, which are essential for surgical and medical management decisions. Propofol has been studied in children with intracardiac shunts undergoing cardiac catheterization ( Gozal et al., 2001 ). Mild systemic hypotension that has been described in pediatric patients with propofol may have deleterious effects on right-to-left intracardiac shunts. Fifteen children (aged 18 months to 9 years) underwent cardiac catheterization with sedation, without supplemental oxygen, using 1 mcg/kg of fentanyl followed by propofol bolus (1 to 2 mg/kg) and a continuous infusion (100 mcg/kg per minute). Hemodynamic data, including systemic venous and pulmonary artery and vein pressures and aortic saturation, were recorded;    P and    S were calculated. Despite lower pressures during propofol infusion, compared with those pressures measured after the discontinuation of propofol, the intracardiac shunt remained unchanged.

Ketamine

Ketamine is a nonbarbiturate cyclohexamine derivative that is classified as a dissociative anesthetic and has been widely used in children, especially patients with congenital heart disease ( Singh et al., 2000; Jobeir et al., 2003 ; Kogan et al., 2003 ; Pees et al., 2003 ). It has been used extensively by pediatric cardiologists in pediatric cardiac catheterization laboratories for sedation. Morray and others (1984)studied the effects of intravenous ketamine (2 mg/kg) given during cardiac catheterization. No significant changes in arterial blood gases, pulmonary artery pressure, HR, or pulmonary-to-systemic arteriolar resistance ratios were found ( Morray et al., 1984 ).

Propofol and ketamine were compared in three groups of children undergoing cardiac catheterization: (1) children without intracardiac shunts, (2) children with a left-to-right shunt, and (3) children with a right-to-left shunt. The children were premedicated with oral midazolam and then randomized to receive a continuous infusion of either propofol (100 to 200 mcg/kg per minute) or ketamine (50 to 75 mcg/kg per minute). Hemodynamic data, including systemic venous and pulmonary artery and vein pressures and aortic saturations, were recorded;    p and    s were calculated. The same set of data was recorded before discontinuation of infusions at the end of the procedure. All patients receiving propofol infusions had significant decreases in systemic MAP. In patients with cardiac shunts, propofol infusion significantly decreased SVR and increased systemic blood flow, whereas PVR and pulmonary blood flow did not change significantly. These changes resulted in decreased left-to-right shunting and increased right-to-left shunting; the pulmonary-to-systemic flow ratio decreased significantly. The ketamine-treated patients showed a significant increase in systemic MAP in all patient groups, but pulmonary MAP, SVR, and PVR was unchanged. Ketamine caused fewer effects on intracardiac shunting ( Oklu et al., 2003 ).

Hickey and others (1985) studied children after cardiac surgery who had been administered intravenous ketamine (2 mg/kg). No significant changes in HR, systemic or pulmonary arterial mean pressures, CI, PVR, or SVR were noted after drug administration. Maintaining normal ventilation and normal PCO2 plays an important role in assessing the effects of ketamine.

Wolfe and others (1991) and Berman and others (1990) studied the effects of ketamine in two high-altitude cities—Albuquerque and Denver—and found dramatic increases in pulmonary artery pressure and pulmonary arteriolar resistance with the administration of ketamine. Interpretation of these data must be evaluated in view of the contribution of the high altitude. Similarly, arterial blood gas data should be available when ketamine is administered during cardiac catheterization, because elevations in pulmonary artery pressure and resistance may result from hypercarbia during ketamine administration ( Hickey et al., 1984) . Apnea and excessive secretions are seen with ketamine and may contribute to hypoxia, hypercarbia, and the hemodynamic results described in various studies performed in catheterization laboratories. When adequate ventilation is maintained, the effect of ketamine on pulmonary resistance may be minimized in children.

Local Anesthetics

The use of local anesthetics for topical analgesia and anesthesia as well as for use in regional anesthetic techniques requires a clear understanding of pharmacokinetics, pharmacodynamics, and proper dosing. Because infants have low pseudocholinesterase levels, the metabolism of ester-type local anesthetics is decreased. The kinetics of intravenous lidocaine is similar in older infants, children, and adults, but a much longer elimination half-life is observed in children when this agent is delivered intrathecally. In an animal model of right-to-left intracardiac shunting, Bokesch and others (1987) showed higher plasma lidocaine levels in the systemic circulation. The absorption of lidocaine in the lung accounts for the potential toxicity of local anesthetics in patients with right-to-left shunts, and dosages should be adjusted accordingly.

Regional Anesthesia and Analgesia

The use of regional blocks in children of all ages has increased in popularity. Regional anesthesia and analgesia can be safe and effective. Both local anesthetics and opioids have been used in regional blockade. The hemodynamic response to sympathetic blockade by local anesthetics is age dependent, with children younger than 8 years old showing minimal hemodynamic changes with epidural or intrathecal administration of local anesthetics, even with high levels of blockade ( Dohi et al., 1979 ). Children may have a different baseline sympathetic tone compared with adults, who typically respond to blockade with hypotension. In addition, children may have less venous pooling and smaller lower extremity-to-body surface area ratios. Doppler studies have shown minimal alterations in blood pressure and CO in young children ( Payen et al., 1987 ). If a normal circulating blood volume is present, fluid loading, which is normally done in adults, is unnecessary in children. The effects of caudal extradural analgesia on pulmonary and systemic arterial pressure have been examined in children.

Kawamoto and others (1984) examined 27 children who had received a lidocaine caudal block, noting an insignificant change in pulmonary arterial pressure and aortic pressure in children with normal cardiac function. Aortic pressure did decrease significantly in children with cardiac disease. In addition, if pulmonary hypertension was present before blockade, pulmonary artery pressure increased significantly with a simultaneous decrease in aortic pressure. Serum levels of lidocaine were not toxic. Optimal postoperative pain management cannot be ignored because there is a profound interaction between cardiovascular stability and pain systems that has often been overlooked in the postoperative management of cardiac dysfunction ( Randich and Maixner, 1984 ).

Pediatric caudal anesthesia has found a major place in pediatric anesthesia as well as in postoperative pain management. Although this technique has been widely applied, the cardiovascular effects were not well studied in children. Larousse and others (2002) used transesophageal Doppler, a noninvasive method, to examine the cardiovascular effects in healthy children. Ten children (aged 2 months to 5 years) who were scheduled for lower abdominal surgery were studied. General anesthesia was induced using sevoflurane and was followed by the insertion of a transesophageal Doppler probe. Caudal anesthesia was performed using 1 mL/kg of 0.25% bupivacaine with 1:200,000 epinephrine. Hemodynamic variables were collected before and after caudal anesthesia. No complications arose during insertion of the probe. The mean time between the two sets of measurements was 15 minutes. HR, MAP, and systolic and diastolic BPs were not modified by caudal anesthesia. Descending aortic blood flow increased significantly from 1.14 to 1.92 L/min (P = .0002). Aortic ejection volume increased from 8.5 to 14.5 mL (P = .0002). Aortic vascular resistances decreased from 6,279 to 3,901 dynes/sec per cm-5 (P = .005). Caudal anesthesia did not affect HR and MAP but induced a significant increase in descending aortic blood flow.

The hemodynamic and cardiovascular effects were studied of epidural anesthesia with plain bupivacaine 0.75 mL/kg in 13 nonpremedicated American Society of Anesthesiologists class 1 children using measurements of HR and BP and M-mode echocardiography. Using general anesthesia, M-mode echocardiographic evaluation of left ventricular function was performed in each patient at four points (after general anesthesia and 5, 10, and 25 minutes after epidural anesthesia). HR decreased significantly at 10 and 25 minutes, and MBP decreased at 5 and 10 minutes compared with point A. No other M-mode echocardiographic indices were changed at any point. Epidural anesthesia with 0.25% bupivacaine 0.75 mL/kg did not affect left ventricular function in young children ( Tsuji et al., 1996 ).

The potent, even life-threatening, effects of local anesthetics when inadvertent intravascular injection occurs mandate that there be a reliable method of detecting intravascular injection. Epinephrine may induce tachycardia or hypertension, but this technique has produced false-positive and false-negative findings. Electrocardiographic changes as markers of intravascular injection of local anesthetics with epinephrine, during placement of epidural blocks in children, have been studied as a more reliable approach. During a 1-year period, all pediatric patients undergoing epidural anesthesia had an electrocardiogram rhythm strip recorded during test dose injection and analyzed for changes in rate, rhythm, and T-wave configuration. During the 1-year period, 742 pediatric epidural blocks and 644 caudal (284 without catheters), 97 lumbar, and 1 thoracic epidural anesthetic procedures were performed, with a satisfactory placement rate of 97.7%. Intravascular injection was detected in 42 (5.6%) epidural anesthetic procedures (3.8% and 6.7% of straight needle and catheter injections, respectively).

Detection was made by immediate aspiration of blood in 6 patients and by HR increases of greater than 10 beats per minute in 30 patients. Five patients had HR decreases suggesting a baroreceptor response. Five patients had HR increases of less than 10 beats per minute that were thought to be secondary to noxious stimuli. Of 30 patients with known intravascular injection and for whom electrocardiographic strips were available, 25 (83%) had T-wave amplitude increases of greater than 25%, and 29 (97%) had electrocardiographic changes in T-wave or rhythm in response to the epinephrine injection. There were no false-positive results. Epinephrine can be used effectively to test for intravascular injection, but slow, incremental dosing should be used as well. In children with cardiac disease who are undergoing cardiac surgery, the role of regional anesthesia has also found an important role, and its safety and efficacy are being verified in research since the 1990s ( Dalens and Mazoit, 1998 ;Naguib et al., 1998 ; Hammer et al., 2000 ; Holtby, 2002 ; Bosenberg, 2003 ; Rosen et al., 2002 ; Steven, 2000 ; Mazoit and Dalens, 2004 ).

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ EFFECTS OF CARDIOPULMONARY INTERACTIONS

▪ CLINICAL CONSIDERATIONS AND NEW TREATMENT MODALITIES

The cardiovascular and respiratory systems interact dynamically at all stages of development. The delivery of oxygen to optimally meet the metabolic needs of all tissues and organs is the goal of both systems. In the neonatal period, evolving anatomic and physiologic changes in the pulmonary bed dramatically affect cardiovascular stability and the maturation from a fetal to a transitional and ultimately a neonatal circulatory pattern. Systemic and pulmonary venous return and the output of both ventricles are affected by cardiopulmonary interactions. Right ventricular preload derives from extrathoracic vessels, whereas right ventricular output is into intrathoracic vessels. On the other hand, left ventricular output is into extrathoracic vessels, and preload originates from intrathoracic vessels. This gives the RV the ability to augment left ventricular preload. Right ventricular preload cannot be similarly augmented. The ventricular septum responds dramatically to changes in both ventricles and can adversely affect left ventricular ejection when increased right ventricular afterload causes septal bowing into the left ventricular outflow tract. This can adversely affect myocardial function, especially in the neonate and young infant. Alterations in intrathoracic pressure have effects on myocardial wall tension. Modes of ventilation can have significant effects on preload, such as the use of positive pressure ventilation and peak end-expiratory pressure, both of which increase right atrial pressure and decrease preload. Increases in right ventricular pressure and right ventricular afterload are also observed ( Pinsky, 1990 ). It is not surprising that patients with right ventricular dysfunction can decompensate with these changes.

How these modes of ventilation affect left ventricular function is less clear. A decrease in right ventricular preload should lead to a decreased volume of blood received by the LV. The increase in right ventricular afterload may adversely affect septal motion and dynamically contribute to left ventricular dysfunction. Ventilation affects PVR. PVR is high at very low or high lung volumes and lowest at functional residual capacity. These potent and dynamic interactions have important implications for respiratory management in patients with cardiac dysfunction and in the stressed neonate, in whom reversion to a fetal circulatory pattern is possible. Ventilatory manipulation is an important tool in minimizing high PVR and its subsequent negative effects on cardiac function. Right ventricular afterload and PVR can be decreased by ventilation with decreased intrathoracic pressures and increased respiratory rates, leading to respiratory alkalosis. High-frequency jet and oscillatory ventilation result in lower mean airway and intrathoracic pressures. High inspired oxygen concentrations are widely used to decrease PVR. Extracorporeal membrane oxygenation, surfactant replacement, and inhaled nitric oxide (NO) have been used to support a failing cardiorespiratory system. The clinical relevance of cardiorespiratory interactions, especially in the critically ill, artificially ventilated pediatric patient, was reviewed by Robotham (1987) .

Nitric Oxide

During the past several decades, research on the control of vascular smooth muscle tone and the mediators of resting pulmonary vascular tone has contributed to a better understanding of pulmonary hypertension and its treatment. The synthesis and release of various vasoactive substances contribute to vasomotor tone. When this synthesis is impaired, vasomotor tone may be adversely affected. PFC (also called PPHN) or pulmonary hypertension in patients with severe congenital heart disease is the clinical manifestation of increased pulmonary vascular tone. Many vasoactive products are released by the pulmonary vasculature, including prostacyclin, endothelium-derived relaxant factor (EDRF), and vasoconstrictors such as endothelin. A change in blood flow or shear stress stimulates the release of prostacyclin or EDRF ( Van Grondelle et al., 1986 ). In addition, the response of the endothelium to various pharmacologic agents, such as acetylcholine, may require an intact endothelium ( Furchgott and Zawadzki, 1980 ).

NO has been identified as EDRF, and intensive research has demonstrated an important role for NO in the treatment of pulmonary hypertension among pediatric patients, especially during the perinatal period and after cardiopulmonary bypass. NO may be an important mediator in the development of transitional circulation of the newborn. NO directly activates soluble guanylate cyclase of vascular smooth muscle, thus increasing cyclic guanosine monophosphate and relaxing vascular smooth muscle. L-Arginine is the precursor for the formation of NO in vascular tissues. Davidson and Eldemerdash (1990) demonstrated that EDRF was present in the pulmonary and systemic arteries of newborn guinea pigs, and Roberts and others (1993) showed that inhaled NO was a selective vasodilator in hypoxic newborn lambs. NO has also been studied in PPHN in both low and high doses with a documented reversal in hypoxemia secondary to PPHN ( Kinsella et al., 1992 ; Roberts et al., 1992 ). Inhalation of NO in these studies did not show any significant effect on systemic pressure. These clinical studies confirmed the laboratory findings that NO acts as a selective pulmonary vasodilator to reverse hypoxic pulmonary vasoconstriction in awake lambs. Lang and others (1992) studied congenital heart disease patients with pulmonary hypertension in the cardiac catheterization laboratory and postoperatively. Inhaled NO was shown to selectively reduce PVR in many patients and is useful for the diagnostic evaluation of severe congenital heart disease complicated by pulmonary hypertension.

Since the 1990s, NO has found wide application in pediatric medicine. A European consensus conference in 2004 reviewed the use of inhaled NO in neonates and children ( Macrae et al., 2004 ). The cases of preterm neonates, children with cardiac disease, and children with acute lung injury and respiratory distress syndrome who were treated with NO were studied. With data from a Cochrane Review ( Finer et al., 2000) on NO and expert consensus, certain recommendations were made regarding these groups. In preterm infants, there are three published, randomized controlled trials of NO therapy ( The Franco-Belgium Collaborative NO Trial Group, 1999 ; Subhedar et al., 1997 ; Kinsella et al., 1999 ). The Cochrane Review concluded that sufficient data are lacking for evaluation of the possible effects of inhaled NO on periventricular hemorrhage and on long-term neurodevelopmental outcome. The European consensus group recommended that further use of NO in preterm infants be done within the format of controlled clinical trials or as a rescue therapy in life-threatening hypoxemia after all other modalities have failed.

In the clinical setting of acute lung injury and acute respiratory distress syndrome, many systemic disease processes are involved in patients of all ages. The use of NO to improve oxygenation by improving ventilation-perfusion mismatching has had increasing clinical application. Nevertheless, the transient improvement in oxygenation has not yet been proved to have an impact on mortality. Trials in children are very limited ( Dobyns et al., 1999 ), and it appears that underlying disease and not respiratory failure alone may be a critical factor in outcome analysis.

NO in children with cardiac disease has found clinical use in those with pulmonary hypertension because of acquired or congenital heart disease. As a selective pulmonary vasodilator, NO has also been used to differentiate fixed and reactive pulmonary hypertension and therefore has found an important role in diagnostic cardiac procedures ( Wessel et al., 1993 ; Adatia et al., 1995 ). Severe reactive pulmonary hypertension after cardiac bypass procedures has been treated with NO and in randomized controlled trials was shown to significantly reduce pulmonary hypertensive events ( Day et al., 2000 ;Miller et al., 2000 ). NO has also been shown to improve the negative effects of elevated PVR after the Fontan operation and in those patients with right ventricular failure, but these observations have not been studied in randomized controlled trials. Because of the lack of extensive randomized controlled trial data, the routine prophylactic use of NO in postoperative congenital heart disease patients was not recommended by the consensus group. Continued research, especially more definitive randomized control trials, will define the place of NO in pediatric care.

The role of other pulmonary vasodilators is another area of research. Prostacyclin and its analogues (prostanoids) are potent vasodilators and possess antithrombotic and antiproliferative properties. All of these properties help to antagonize the pathologic changes that take place in the small pulmonary arteries of patients with pulmonary hypertension. Prostaglandins and phosphodiesterase inhibitors and endothelin receptor antagonists such as prostacyclin, treprostinil, beraprost, and iloprost may be combined to treat pulmonary hypertension, and these combination therapies may hold promise for future therapies ( Olschewski et al., 2003 ).

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ SUMMARY

Anesthesiologists caring for children must have a clear and precise understanding of cardiovascular physiology, the developmental aspects of cardiac function, the effects of anesthetics, and the dynamic interactions of the cardiopulmonary systems. A safe and effective anesthetic plan can then be successfully formulated.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

REFERENCES

Adatia et al., 1995. Adatia I, Perry S, Landzberg M, et al: Inhaled nitric oxide and hemodynamic evaluation of patients with pulmonary hypertension before transplantation.  J Am Coll Cardiol  1995; 25:1656.

Anand and Hickey, 1992. Anand KJS, Hickey PR: Halothane-morphine compared with high dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery.  N Engl J Med  1992; 326:1.

Anderson et al., 1984. Anderson PAW, Glick K, Manring A, et al: Developmental changes in cardiac contractility in fetal and postnatal sheep: In vitro and in vivo.  Am J Physiol  1984; 247:H371.

Artman, 1992. Artman M: Sarcolemmal sodium-calcium exchange activity and exchanger immunoreactivity in developing rabbit hearts.  Am J Physiol  1992; 263:H1506-H1513.

Assali, 1978. Assali NS, et al: Ontogenesis of the autonomic control of cardiovascular functions in the sheep.   In: Longo LD, Reneau DD, ed. Fetal and newborn cardiovascular physiology, vol 1: Developmental aspects,  New York: Garland STPM Press; 1978:47.

Audenaert et al., 1992. Audenaert SM, Lock RL, Johnson GL, et al: Cardiovascular effects of rectal methohexital in children.  J Clin Anesth  1992; 4:116.

Audenaert et al., 1995. Audenaert SM, Montgomery CL, Thompson DE, Sutherland J: A prospective study of rectal methohexital: efficacy and side effects in 648 cases.  Anesth Analg  1995; 81:957.

Aun et al., 1993. Aun CST, Sung RYT, O'Meara ME: Cardiovascular effects of IV induction in children: Comparison between propofol and thiopentone.  Br J Anaesth  1993; 70:647.

Barash et al., 1978. Barash PG, Glanz S, Katz JD, et al: Ventricular function in children during halothane anesthesia.  Anesthesiology  1978; 49:79.

Barker, 1993. Barker DJ: The intrauterine origins of cardiovascular disease.  Acta Paediatr Suppl  1993; 82(suppl 391):93-99.discussion 100

Barker and Fall, 1993. Barker DJ, Fall CH: Fetal and infant origins of cardiovascular disease.  Arch Dis Child  1993; 68:797-799.

Berman et al., 1990. Berman W, Fripp RR, Rubler M, et al: Hemodynamic effects of ketamine in children undergoing cardiac catheterization.  Pediatr Cardiol  1990; 11:72.

Bokesch et al., 1987. Bokesch PM, Castaneda AR, Ziemer G, et al: The influence of a right-to-left shunt on lidocaine pharmacokinetics.  Anesthesiology  1987; 67:739.

Borow and Grossman, 1984. Borow KM, Grossman W: Clinical use of pressure-dimension and stress-shortening relations in systole and diastole.  Fed Proc  1984; 43:2414-2417.

Bosenberg, 2003. Bosenberg A: Neuraxial blockade and cardiac surgery in children.  Paediatr Anaesth  2003; 13:559.

Boudreaux et al., 1984. Boudreaux JP, Schieber RA, Cook DR: Hemodynamic effects of halothane in the newborn piglet.  Anesth Analg  1984; 63:731.

Breall et al., 1984. Breall JA, Rudolph AM, Heymann MA: Role of thyroid hormone in postnatal circulatory and metabolic adjustments.  J Clin Invest  1984; 73:1418.

Cahalan et al., 1993. Cahalan MK, Ionescu P, Melton HE, et al: Automated real-time analysis of intraoperative transesophageal echocardiograms.  Anesthesiology  1993; 78:477.

Carmichael, 1947. Carmichael EB: The median lethal dose (LD50) of Pentothal sodium for both young and old guinea pigs and rats.  Anesthesiology  1947; 8:589.

Cassidy et al., 1992. Cassidy SC, Schmidt KG, Van Hare GF, et al: Complications of pediatric cardiac catheterization: a 3-year study.  J Am Coll Cardiol  1992; 19:1285.

Church et al., 1967. Church SC, Morgan BC, Oliver Jr TK, et al: Cardiac arrhythmias in premature infants: An indication of autonomic immaturity?.  J Pediatr  1967; 71:542.

Cohn et al., 1985. Cohn HE, Freed MD, Hellenbrand WF, et al: Complications and mortality associated with cardiac catheterization in infants under one year: A prospective study.  Pediatr Cardiol  1985; 6:123.

Colan et al., 1984. Colan SD, Borow KM, Neumann A: Left-ventricular end-systolic wall stress-velocity of fiber shortening relation: A load independent index of myocardial contractility.  J Am Coll Cardiol  1984; 4:715.

Colan et al., 1987a. Colan SD, Sanders SP, Borow K: Physiologic hypertrophy: Effects on left ventricular systolic mechanics in athletes.  J Am Coll Cardiol  1987; 9:776.

Colan et al., 1987b. Colan SD, Sanders SP, Ingelfinger JR, et al: Left ventricular mechanics and contractile state in children and young adults with end-stage renal disease: Effect of dialysis and renal transplantation.  J Am Coll Cardiol  1987; 10:1085.

Colan et al., 1989. Colan SD, Sanders SP, Parness IA, et al: Evidence of enhanced contractility in normal infants compared to older children and adults [abstract].  J Am Coll Cardiol  1989; 13:135A.

Constant et al., 1999. Constant I, Dubois MC, Piat V, Moutard ML: Changes in electroencephalogram and autonomic cardiovascular activity during induction of anesthesia with sevoflurane compared with halothane in children.  Anesthesiology  1999; 91:1604.

Cook et al., 1981. Cook DR, Brandom BW, Shiu G, et al: The inspired median effective dose, brain concentration at anesthesia, and cardiovascular index for halothane in young rats.  Anesth Analg  1981; 60:182.

Dalens and Mazoit, 1998. Dalens BJ, Mazoit JX: Adverse effects of regional anesthesia in children. (Review).  Drug Saf  1998; 19:251-268.

Davidson and Eldemerdash, 1990. Davidson D, Eldemerdash A: Endothelium-derived relaxing factor: Presence in pulmonary and systemic arteries of the newborn guinea pig.  Pediatr Res  1990; 27:128.

Davis et al., 1987. Davis PJ, Cook DR, Stiller RL, et al: Pharmacodynamics and pharmacokinetics of high-dose sufentanil in infants and children undergoing cardiac surgery.  Anesth Analg  1987; 66:203.

Dawes and Mott, 1962. Dawes GS, Mott JC: The vascular tone of the foetal lung.  J Physiol  1962; 164:465.

Dawes, 1968. Dawes GS: Fetal and neonatal physiology,  Chicago, Year Book Medical, 1968.

Day et al., 2000. Day RW, Hawkins JA, McGough EC, et al: Randomized controlled study of inhaled nitric oxide after operation for congenital heart disease.  Ann Thorac Surg  2000; 69:1907.

Diaz, 1985. Diaz JH: Halothane anesthesia in infancy: Identification and correlation of preoperative risk factors with intraoperative arterial hypotension and postoperative recovery.  J Pediatr Surg  1985; 20:502.

Dobyns et al., 1999. Dobyns EL, Cornfield DN, Anas NG, et al: Multicenter randomized controlled trial of the effects of inhaled nitric oxide therapy on gas exchange in children with acute hypoxemic respiratory failure.  J Pediatr  1999; 134:406.

Dohi et al., 1979. Dohi S, Naito H, Takasaki T: Age-related changes in blood pressure and duration of motor block in spinal anesthesia.  Anesthesiology  1979; 50:319.

Domek et al., 1960. Domek NS, Barlow CF, Roth LJ, et al: An octogenetic study of phenobarbital C-14 in cat brain.  J Pharmacol Exp Ther  1960; 130:285.

Emmanouilides et al., 1964. Emmanouilides GC, Moss AJ, Duffie Jr ER, et al: Pulmonary arterial pressure changes in human newborn infants from birth to three days of age.  J Pediatr  1964; 65:327.

Faxelius et al., 1984. Faxelius G, Lagercrantz H, Yao A: Sympathoadrenal activity and peripheral blood flow after birth: Comparison in infants delivered vaginally and by cesarean section.  J Pediatr  1984; 105:144.

Finer and Barrington, 2000. Finer NN, Barrington KJ: Nitric oxide for respiratory failure in infants born at or near term.  Cochrane Database Syst Rev  2000; 2:CD000399

Fisher et al., 1981. Fisher DJ, Heymann MA, Rudolph AM: Myocardial consumption of oxygen and carbohydrates in newborn sheep.  Pediatr Res  1981; 15:843-846.

Fisher et al., 1975. Fisher EA, DuBrow IW, Hastreiter AR: Comparison of ejection phase indices of left ventricular performance in infants and children.  Circulation  1975; 52:916.

The Franco-Belgium Collaborative NO Trial Group, 1999. The Franco-Belgium Collaborative NO Trial Group : Early compared with delayed inhaled nitric oxide in moderately hypoxaemic neonates with respiratory failure: A randomized controlled trial.  Lancet  1999; 354:1066-1071.

Frank, 1895. Frank O: Zur Dynamik des Herzmuskels.  A Biol  1895; 32:370.(translated by Chapman CB and Wasserman E: Am Heart J 58: 282, 1959).

Friesen and Lichtor, 1982. Friesen RH, Lichtor JL: Cardiovascular depression during halothane anesthesia in infants: A study of three induction techniques.  Anesth Analg  1982; 61:42.

Frink et al., 1992. Frink Jr EJ, Malan TP, Atlas M, et al: Clinical comparison of sevoflurane and isoflurane in healthy patients.  Anesth Analg  1992; 74:241.

Furchgott and Zawadzki, 1980. Furchgott RF, Zawadzki JV: The obligatory role of endothelial cells in the relaxation of arterial smooth muscle.  Nature (Lond)  1980; 288:373.

Gamble et al., 1981. Gamble JA, Kawar P, Dundee JW, et al: Evaluation of midazolam as an intravenous induction agent.  Anaesthesia  1981; 36:868.

Geis et al., 1975. Geis WP, Tatooles CJ, Priola DV, et al: Factors influencing neurohumoral control of the heart in the newborn dog.  Am J Physiol  1975; 228:1685.

Gentles and Colan, 2002. Gentles TL, Colan SD: Wall stress misrepresents afterload in children and young adults with abnormal left ventricular geometry.  J Appl Physio1  2002; 92:1053.

Gersony et al., 1969. Gersony WM, Duc GV, Sinclair JC: “PFC” syndrome (persistence of the fetal circulation).  Circulation  1969; 40:111.

Glenski et al., 1988. Glenski JA, Friesen RH, Berglund NL, et al: Comparison of the hemodynamic and echocardiographic effects of sufentanil, fentanyl, isoflurane and halothane for pediatric cardiovascular surgery.  J Cardiothoracic Anesth  1988; 2:147.

Goertz et al., 1993. Goertz AW, Lindner KH, Seefelder C, et al: Effect of phenylephrine bolus administration on global left ventricular function in patients with coronary artery disease and patients with valvular aortic stenosis.  Anesthesiology  1993; 78:834.

Gozal et al., 2001. Gozal D, Rein AJ, Nir A, Gozal Y: Propofol does not modify the hemodynamic status of children with intracardiac shunts undergoing cardiac catheterization.  Pediatr Cardio  2001; 6:488.

Grant, 1999. Grant DA: Ventricular constraint in the fetus and newborn.  Can J Cardiol  1999; 15:95.

Grant et al., 1992a. Grant DA, Kondo CS, Maloney JE, et al: Changes in pericardial pressure during the perinatal period.  Circulation  1992; 86:1615.

Grant et al., 1992b. Grant DA, Maloney JE, Tyberg JV, et al: Effects of external constraint on the fetal left ventricular function curve.  Am Heart J  1992; 123:1601-1609.

Grant and Walker, 1996. Grant DA, Walker AM: Pleural pressures limit fetal right ventricular output.  Circulation  1996; 94:555-561.

Gregory, 1982. Gregory GA: The baroresponses of preterm infants during halothane anesthesia.  Can Anaesth Soc J  1982; 29:105.

Gueugniaud et al., 1998. Gueugniaud PY, Abisseror M, Moussa M, et al: The hemodynamic effects of pneumoperitoneum during laparoscopic surgery in healthy infants: assessment by continuous esophageal aortic blood flow echo-Doppler.  Anesth Analg  1998; 86:290.

Gueugniaud et al., 1997. Gueugniaud PY, Muchada R, Moussa M, et al: Continuous oesophageal aortic blood flow echo-Doppler measurement during general anesthesia in infants.  Can J Anaesth  1997; 44:745.

Guilleminault et al., 1981. Guilleminault C, Briskin JG, Greenfield MS, et al: The impact of autonomic nervous system dysfunction on breathing during sleep.  Sleep  1981; 4:263.

Guttgesel et al., 1977. Guttgesel HP, Paquet M, Duff DF, et al: Evaluation of left ventricular size and function by echocardiography. Results in normal children.  Circulation  1977; 56:457.

Hall et al., 2004. Hall SM, Hislop AA, Wu Z, et al: Remodelling of the pulmonary arteries during recovery from pulmonary hypertension induced by neonatal hypoxia.  J Pathol  2004; 203:575-583.

Hammer et al., 2000. Hammer GB, Ngo K, Macario A: A retrospective examination of regional plus general anesthesia in children undergoing open heart surgery.  Anesth Analg  2000; 90:1020.

Hannallah et al., 1991. Hannallah RS, Baker SB, Casey W, et al: Propofol: Effective dose and induction characteristics in unpremedicated children.  Anesthesiology  1991; 74:217.

Hannallah et al., 1994. Hannallah RS, Britton JT, Schafer PG, et al: Propofol anaesthesia in paediatric ambulatory patients: A comparison with thiopentone and halothane.  Can J Anaesth  1994; 41:12.

Hansen and Hickey, 1986. Hansen DD, Hickey PR: Anesthesia for hypoplastic left heart syndrome: Use of high-dose fentanyl in 30 neonates.  Anesth Analg  1986; 65:127.

Haworth and Hislop, 2003. Haworth SG, Hislop AA: Lung development'the effects of chronic hypoxia [review].  Semin Neonatol  2003; 8:1-8.

Haworth and Reid, 1977. Haworth SG, Reid L: Structural study of the pulmonary circulation in total anomalous pulmonary venous return in early infancy.  Br Heart J  1977; 39:80.

Hickey and Hansen, 1984. Hickey PR, Hansen DD: Fentanyl and sufentanil-oxygen-pancuronium anesthesia for cardiac surgery in infants.  Anesth Analg  1984; 63:117.

Hickey and Hansen, 1991. Hickey PR, Hansen DD: High-dose fentanyl reduces intraoperative ventricular fibrillation in neonates with hypoplastic left heart syndrome.  J Clin Anesth  1991; 3:295.

Hickey et al., 1985. Hickey PR, Hansen DD, Cramolini GM: Pulmanory and systemic hemodynamic responses to ketamine in infants with normal and elevated pulmonary vascular resistance.  Anesthesiology  1985; 62:287.

Hickey et al., 1985a. Hickey PR, Hansen DD, Strafford M, et al: Pulmonary and systemic responses to high dose fentanyl in infants.  Anesth Analg  1985; 64:483.

Hickey et al., 1986. Hickey PR, Hansen DD, Strafford M, et al: Pulmonary and systemic hemodynamic effects of nitrous oxide in infants with normal and elevated pulmonary vascular resistance.  Anesthesiology  1986; 65:374.

Hickey et al., 1985b. Hickey PR, Hansen DD, Wessel DL, et al: Blunting of stress responses in the pulmonary circulation of infants by fentanyl.  Anesth Analg  1985; 64:1137.

Holtby, 2002. Holtby H: Con: regional anesthesia is not an important component of the anesthetic technique for pediatric patients undergoing cardiac surgical procedures.  J Cardiothorac Vasc Anesth  2002; 16:379.

Holzman et al., 1996. Holzman RS, Van der Velde ME, Kaus SJ, et al: Sevoflurane depresses myocardial contractility less than halothane during induction of anesthesia in children.  Anesthesiology  1996; 85:1260.

Huettemann et al., 2004. Huettemann E, Junker T, Chatzinikolaou KP, et al: The influence of anthracycline therapy on cardiac function during anesthesia.  Anesth Analg  2004; 98:941.

Jobeir et al., 2003. Jobeir A, Galal MO, Bulbul ZR, et al: Use of low-dose ketamine and/or midazolam for pediatric cardiac catheterization.  Pediatr Cardiol  2003; 24:236.

Johannesson et al., 1995. Johannesson GP, Floren M, Lindahl SG: Sevoflurane for ENT-surgery in chidren. A comparison with halothane.  Acta Anaesthesiol Scand  1995; 39:546.

Jones et al., 1994. Jones RD, Visram AR, Chan MM, et al: A comparison of three induction agents in paediatric anesthesia cardiovascular effects and recovery.  Anaesth Intensive Care  1994; 22:545.

Jones et al., 2004. Jones RD, Morice AH, Emery CJ: Effects of perinatal exposure to hypoxia upon the pulmonary circulation of the adult rat.  Physiol Res  2004; 53:11-17.

Kain et al., 2004. Kain ZN, Caldwell-Andrews AA, Krivutza DM, et al: Trends in the practice of parental presence during induction of anesthesia and the use of preoperative sedative premedication in the United States, 1995–2002: results of a follow-up national survey.  Anesth Analg  2004; 98:1252.

Katona and Egbert, 1978. Katona PG, Egbert MS: Heart rate and respiratory rate differences between preterm and full term infants during quiet sleep: Possible implications for sudden infant death syndrome.  Pediatrics  1978; 62:91.

Katona et al., 1980. Katona PG, Frasz A, Egbert J: Maturation of cardiac control in full term and pre-term infants during sleep.  Early Hum Dev  1980; 4:145.

Kaufman and Rudolph, 1988. Kaufman TM, Rudolph AM: The effect of heart rate on the atrial contribution to stroke volume in newborn and one month old lambs.  Pediatr Res  1988; 24:434A.

Kawamoto et al., 1984. Kawamoto M, Takasaki M, Kawasaki H, et al: Effects of caudal extradural analgesia on pulmonary and systemic arterial pressure in children.  Jpn J Anesthesiol  1984; 33:520.

Kay and Stephenson, 1980. Kay B, Stephenson D: Alfentanil (R39209): Initial clinical experiences with a new narcotic analgesic.  Anaesthesia  1980; 35:1197.

Kenny et al., 1987. Kenny J, Plappert T, Doubliet P, et al: Effects of heart rate on ventricular size, stroke volume and output in the normal human fetus: A prospective Doppler echocardiographic study.  Circulation  1987; 76:52.

Kinsella et al., 1992. Kinsella JP, Neish SR, Shaffer E, et al: Low-dose inhalational nitric oxide in persistent pulmonary hypertension of the newborn.  Lancet  1992; 340:819.

Kinsella et al., 1999. Kinsella JP, Walsh WF, Bose CL, et al: Inhaled nitric oxide in premature neonates with severe hypoxaemic respiratory failure: a randomized controlled trial.  Lancet  1999; 354:1061.

Kissen et al., 1983. Kissen I, Morgan PL, Smith LR: Comparison of isoflurane and halothane safety margins in rats.  Anesthesiology  1983; 58:556.

Klopfenstein and Rudolph, 1978. Klopfenstein HS, Rudolph AM: Postnatal changes in the circulation and responses to volume loading in sheep.  Circ Res  1978; 42:839.

Kogan et al., 2003. Kogan A, Efrat R, Katz J, Vidne BA: Propofol-ketamine mixture for anesthesia in pediatric patients undergoing cardiac catheterization.  J Cardiothorac Vasc Anesth  2003; 17:691.

Kramer et al., 1983. Kramer M, Kling D, Walter P, et al: Alfentanil, a new short acting opioid: Haemodynamic and respiratory aspects.  Anaesthetist  1983; 32:265.

Kuipers et al., 1984. Kuipers JRG, Sidi D, Heymann MA, et al: Effects of nitroprusside on cardiac function, blood flow distribution and oxygen consumption in the conscious young lamb.  Pediatr Res  1984; 118:618.

Lang et al., 1992. Lang P, Roberts JD, Bigatello LM, et al: Inhaled nitric oxide: A selective vasodilator for the treatment of pulmonary artery hypertension in congenital heart disease.  Pediatr Res  1992; 31:A104.

Larousse et al., 2002. Larousse E, Asehnoune K, Dartayet B, et al: The hemodynamic effects of pediatric caudal anesthesia assessed esophageal doppler.  Anesth Analg  2002; 94:1165.

Lee et al., 1986. Lee JD, Tajimi T, Patritta J, et al: Preload reserve and mechanisms of afterload mismatch in the normal conscious dog.  Am J Physiol  1986; 250:H464.

Lerman et al., 1990. Lerman J, Oyston JP, Gallagher TM, et al: The minimum alveolar concentration (MAC) and hemodynamic effects of halothane, isoflurane, and sevoflurane in newborn swine.  Anesthesiology  1990; 73:717.

Lerman et al., 1983. Lerman J, Robinson S, Willis MM, et al: Anesthetic requirements for halothane in young children 0–1 month and 1–6 months of age.  Anesthesiology  1983; 59:421.

Lerman et al., 1994. Lerman J, Sikich N, Kleinman S, et al: The pharmacology of sevoflurane in infants and children.  Anesthesiology  1994; 80:814.

Little et al., 1985. Little WC, Freeman GL, O'Rourke RA: Simultaneous determination of left ventricular end-systolic pressure volume and pressure-dimension relationships in closed chest dogs.  Circulation  1985; 71:1301.

Lock et al., 1992. Lock JL, Keane JF, Mandell VS, et al: Cardiac catheterization.   In: Fyler DC, ed. Nadas's pediatric cardiology,  St Louis: Mosby–Year Book; 1992:187.

Macrae et al., 2004. Macrae DJ, Field D, Mercier JC, et al: Inhalled nitric oxide therapy in neonates and children: reaching European consensus.  Intensive Care Med  2004; 30:372.

Mahdavi et al., 1987. Mahdavi V, Izumo S, Nadal-Ginard B: Developmental and hormonal regulation of sarcomeric myosin heavy chain gene family.  Circ Res  1987; 60:804.

Mahler et al., 1975. Mahler F, Ross Jr J, O'Rourke RA, Covell JW: Effects of changes in preload, afterload and inotropic state on ejection and isovolumic phase measures of contractility in the conscious dog.  Am J Cardiol  1975; 35:626.

Mahoney, 1996. Mahoney L: Calcium homeostasis and control of contractility in the developing heart.  Semin Perinatol  1996; 20:510-519.

Masue et al., 2003. Masue T, Shimonaka H, Fuako I, et al: Oral high-dose midazolam premedication for infants and children undergoing cardiovascular sugery.  Paediatr Anaesth  2003; 13:662.

Mazoit and Dalens, 2004. Mazoit JX, Dalens BJ: Pharmacokinetics of local anaesthetics in infants and children.  Clin Pharmacokinet  2004; 43:17.

Mazza et al., 1980. Mazza NM, Epstein MA, Haddad GG, et al: Relation of beat-to-beat variability to heart rate in normal sleeping infants.  Pediatr Res  1980; 14:232.

McNicholas et al., 1983. McNicholas WT, Rutherford R, Grossman R, et al: Abnormal respiratory pattern generation during sleep in patients with autonomic dysfunction.  Am Rev Respir Dis  1983; 128:429.

Miller et al., 2000. Miller OI, Tang SF, Keech A, et al: Inhaled nitric oxide and prevention of pulmonary hypertension after congenital heart surgery: a randomized double-blind study.  Lancet  2000; 356:1464.

Miller and Friesen, 1988. Miller RB, Friesen RH: Oral atropine premedication in infants attenuates cardiovascular depression during halothane anesthesia.  Anesth Analg  1988; 67:180.

Miller, 1982. Miller WP: Cardiac arrhythmias and conduction disturbances in the sleep apnea syndrome. Prevalence and significance.  Am J Med  1982; 73:31.

Mirro et al., 1985. Mirro R, Milley JR, Holzman IR: The effects of sodium nitroprusside on blood flow and oxygen delivery to the organs of the hypoxemic newborn lamb.  Pediatr Res  1985; 19:15.

Mitsnefes et al., 2003. Mitsnefes MM, Kimball TR, Witt SA, et al: Left ventricular mass and systolic performance in pediatric patients with chronic renal failure.  Circulation  2003; 107:864.

Miyatake et al., 1995. Miyatake K, Yamagishi M, Tanaka N, et al: New method for evaluating left ventricular wall motion by color-coded tissue Doppler imaging: in vitro and in vivo studies.  J Am Coll Cardiol  1995; 25:717.

Moorman and Christoffels, 2003. Moorman AF, Christoffels VM: Cardiac chamber formation: Development, genes, and evolution.  Physiol Rev  2003; 83:1223-1267.

Mor-Avi et al., 1997. Mor-Avi V, Vignon P, Koch R, et al: Segmental analysis of color kinesis images: new method for quantification of the magnitude and timing of endocardial motion during left ventricular systole and distole.  Circulation  1997; 95:2082.

Morray et al., 1984. Morray JP, Lynn AM, Stamm SJ, et al: Hemodynamic effects of ketamine in children with congenital heart disease.  Anesth Analg  1984; 63:895.

Muhiudeen et al., 1992. Muhiudeen IA, Roberson DA, Silverman MH, et al: Intraoperative echocardiography for evaluation of congenital heart defects in infants and children.  Anesthesiology  1992; 76:165.

Murat et al., 1990. Murat I, Hoerter J, Ventura-Claper R: Developmental changes in effects of halothane and isoflurane on contractile properties of rabbit cardiac skinned fibers.  Anesthesiology  1990; 73:137.

Murat et al., 1989. Murat I, Lapeyre G, Saint-Maurice C: Isoflurane attenuates baroreflex control of heart rate in human neonates.  Anesthesiology  1989; 70:395.

Murray et al., 1988. Murray D, Forbes R, Murphy K, et al: Nitrous oxide: Cardiovascular effects in infants and small children during halothane and isoflurane anesthesia.  Anesth Analg  1988; 67:1059.

Murray et al., 1987. Murray D, Vandewalker G, Matherne GP, et al: Pulsed Doppler and two-dimensional echocardiography comparison of halothane and isoflurane on cardiac function in infants and small children.  Anesthesiology  1987; 67:211.

Murray et al., 1989. Murray DJ, Forbes RB, Dillman JB, et al: Haemodynamic effects of atropine during halothane or isoflurane anaesthesia in infants and small children.  Can J Anaesth  1989; 36:295.

Murray et al., 1991. Murray DJ, Mehta MP, Forbes RB: The additive contribution of nitrous oxide to isoflurane MAC in infants and children.  Anesthesiology  1991; 75:186.

Naguib et al., 1998. Naguib M, Magboul MM, Samarkandi AH, Attia M: Adverse effects and drug interactions associated with local and regional anaesthesia.  Drug Saf  1998; 18:221.

Nicodemus et al., 1969. Nicodemus HF, Nassiri-Rahimi C, Bachman L, Smith TC: Median effective doses (ED50) of halothane in adults and children.  Anesthesiology  1969; 31:344.

Odenstedt et al., 2001. Odenstedt H, Aneman A, Oi Y, et al: Descending aortic blood flow and cardiac output: a clinical and experimental study of continuous oesophageal echo-Doppler flowmetry.  Acta Anaesthesiol Scand  2001; 45:180.

Oklu et al., 2003. Oklu E, Bulutcu FS, Yalcin Y, et al: Which anesthetic agent alters the hemodynamic status during pediatric catheterization? Comparison of propofol versus ketamine.  J Cardiothorac Vasc Anesth  2003; 17:686.

Olschewski et al., 2003. Olschewski H, Rohde B, Behr J, et al: Pharmacodynamics and pharmacokinetics of inhaled iloprost, aerosolized by three different devices, in severe pulmonary hypertension.  Chest  2003; 124:1294.

O'Kelly et al., 1991. O'Kelly BF, Tubau JF, Knight AA, et al: Measurement of left ventricular contractility using transesophageal echocardiography in patients undergoing coronary artery bypass grafting. The Study of Perioperative Ischaemia (SPI) research group.  Am Heart J  1991; 122:1041.

Pappano, 1972. Pappano AJ: Ontogenic development of autonomic neuroeffector transmission and transmitter reactivity in embryonic and fetal hearts.  Pharmacol Rev  1972; 29:3.

Pappano and Loeffelholz, 1974. Pappano AJ, Loeffelholz K: Ontogenesis of adrenergic and cholinergic neuroeffector transmission in chick embryo heart.  J Pharmacol Exp Ther  1974; 191:468.

Paris et al., 1997. Paris ST, Cafferkey M, Tarling M, et al: Comparison of sevoflurane and halothane for outpatient dental anaesthesia in children.  Br J Anaesth  1997; 79:280.

Payen et al., 1987. Payen D, Ecoffey C, Carli P, et al: Pulsed Doppler ascending aortic, carotid, brachial and femoral artery blood flows during caudal anesthesia in infants.  Anesthesiology  1987; 67:681.

Pees et al., 2003. Pees C, Haas NA, Ewert P, et al: Comparison of analgesic/sedative effect of racemic ketamine and (+)-ketamine during cardiac catheterization in newborns and children.  Pediatr Cardiol  2003; 24:424.

Pinsky, 1990. Pinsky MR: The effects of mechanical ventilation on the cardiovascular system.  Crit Care Clin  1990; 6:663.

Rabinovitch et al., 1981. Rabinovitch M, Herrera-deLeon V, Castaneda AR, et al: Growth and development of the pulmonary vascular bed in patients with tetralogy of Fallot with or without pulmonary atresia.  Circulation  1981; 64:1234.

Randich and Maixner, 1984. Randich A, Maixner W: Interactions between cardiovascular and pain regulatory systems.  Neurosci Behav Rev  1984; 8:343.

Reich et al., 1993. Reich DL, Konstadt SN, Nejat M, et al: Intraoperative transesophageal echocardiography for the detection of cardiac preload changes induced by transfusion and phlebotomy in pediatric patients.  Anesthesiology  1993; 79:10.

Rein et al., 1998. Rein AJ, Tracey M, Colan SD, et al: Automated left ventricular endocardial border detection using acoustic quantification in children.  Echocardiography  1998; 15:111.

Riemenschneider et al., 1981. Riemenschneider TA, Brenner RA, Mason DT: Maturational changes in myocardial contractile state of newborn lambs.  Pediatr Res  1981; 15:349.

Rivenes et al., 2001. Rivenes SM, Lewin MB, Stayer SA, et al: Cardiovascular effects of sevoflurane, isoflurane, halothane, and fentanyl-midazolam in children with congenital heart disease: an echocardiographic study of myocardial contractility and hemodynamics.  Anesthesiology  2001; 94:223.

Roberts et al., 1993. Roberts JD, Chen TY, Kawai N, et al: Inhaled nitric oxide reverses pulmonary vasoconstriction in the hypoxic and acidotic newborn lamb.  Circ Res  1993; 72:246.

Roberts et al., 1992. Roberts JD, Polaner DM, Lang P, et al: Inhaled nitric oxide in persistent pulmonary hypertension of the newborn.  Lancet  1992; 340:818.

Robotham, 1987. Robotham JL: Cardiorespiratory interactions.   In: Rogers MC, ed. Textbook of intensive care, 1. Baltimore: Williams and Wilkins; 1987.

Robotham et al., 1991. Robotham JL, Takata M, Berman M, et al: Ejection fraction revisited.  Anesthesiology  1991; 74:172.

Romero and Friedman, 1979. Romero TE, Friedman WF: Limited left ventricular response to volume overload in the neonatal period: A comparative study with the adult animal.  Pediatr Res  1979; 13:910.

Rosen et al., 2002. Rosen DA, Rosen KR, Hammer GB: Pro: regional anesthesia is an important component of the anesthetic technique for pediatric patients undergoing cardiac surgical procedures.  J Cardiothorac Vasc Anesth  2002; 16:374.

Rosen et al., 1981. Rosen MR, Legato MJ, Weiss RM: Developmental changes in impulse conduction in the canine heart.  Am J Physiol  1981; 240:H546.

Rosenberg et al., 1987. Rosenberg HC, Williams WG, Trusler GA, et al: Structural composition of central pulmonary artery shunts.  J Thorac Cardiovasc Surg  1987; 94:498.

Rudolph, 1974b. Rudolph AM: Cardiac catheterization and angiography.   In: Rudolph AM, ed. Congenital diseases of the heart,  Chicago: Year Book Medical; 1974.

Rudolph, 1974a. Rudolph AM: Changes in the circulation after birth.   In: Rudolph AM, ed. Congenital diseases of the heart,  Chicago: Year Book Medical; 1974.

Rudolph, 1987. Rudolph AM: Fetal circulation and cardiovascular adjustments after birth.   In: Rudolph AM, ed. Pediatrics,  East Norwalk, CT: Appleton and Lange; 1987.

Rudolph and Nadas, 1962. Rudolph A, Nadas A: The pulmonary circulation and congenital heart disease.  N Engl J Med  1962; 267:968.

Rudolph and Heymann, 1973. Rudolph AM, Heymann MA: Circulation and breathing. Control of foetal circulation.   In: Bancroft JJ, ed. Foetal and neonatal physiology,  Cambridge: Cambridge University Press; 1973:89-111.

Rudolph et al., 1981. Rudolph AM, Itshovitz J, Heymann MA: Fetal cardiovascular responses to stress.  Semin Perinatol  1981; 5:109.

Rychik, 2004. Rychik J: Fetal cardiovascular physiology.  Pediatr Cardiol  2004; 25:201.

Rychik and Tian, 1996. Rychik J, Tian ZY: Quantitative assessment of myocardial tissue velocities in normal children with Doppler tissue imaging.  Am J Cardiol  1996; 77:1254.

Scammon and Norris, 1918. Scammon RE, Norris EH: On the time of obliteration of the fetal blood passages (foramen ovale, ductus arteriosus, ductus venosus).  Anat Rec  1918; 15:165.

Schieber et al., 1986. Schieber RA, Namnoum A, Sugden A, et al: Hemodynamic effects of isoflurane in the newborn piglet: Comparison with halothane.  Anesth Analg  1986; 65:633.

Shekerdemian et al., 1997. Shekerdemian L, Bush A, Redington A: Cardiovascular effects of intravenous midazolam after open heart surgey.  Arch Dis Child  1997; 76:57.

Short and Aun, 1991. Short SM, Aun CST: Haemodynamic effects of propofol in children.  Anaesthesia  1991; 46:783.

Singh et al., 2000. Singh A, Girotra S, Mehta Y, et al: Total intravenous anesthesia with ketamine for pediatric interventional cardiac procedures.  J Cardiothoracic Vasc Anesth  2000; 14:36.

Sonnenblick and Stobeck, 1977. Sonnenblick EH, Stobeck JE: Derived indexes of ventricular and myocardial function.  N Engl J Med  1977; 296:978.

Spencer et al., 2003. Spencer KT, Mor-Avi V, Kirkpatrick J, et al: Normal values of left ventricular systolic and diastolic function derived from signal-averaged acoustic quantification waveforms multicenter study.  J Am Soc Echocardiogr  2003; 16:1244.

St. John Sutton et al., 1984. St. John Sutton MG, Raichlen JS, Reichek N, et al: Quantitative assessment of right and left ventricular growth in the human fetal heart: A pathoanatomic study.  Circulation  1984; 70:935.

Starling, 1918. Starling EH: The Linacre lecture on the law of the heart,  London, Longmans, Green, 1918.

Steven and McGowan, 2000. Steven JM, McGowan Jr FX: Neuraxial blockade for pediatric cardiac surgery: lessons yet to be learned.  Anesth Analg  2000; 90:1011.

Strafford et al., 1994. Strafford M, Soliman D, Bokesch P, et al: The risk of aortic compression from transesophageal echocardiographic probe in neonates and infants.  Anesthesiology  1994; 81(suppl):A1219.

Subhedar and Shaw, 1997. Subhedar NV, Shaw NJ: Changes in oxygenation and pulmonary haemodynamics in preterm infants treated with inhaled nitric oxide.  Arch Dis Child Fetal Neonatal Ed  1997; 77:F191.

Suga and Sagawa, 1974. Suga H, Sagawa K: Instantaneous pressure-volume relationships and their ratio in the excised, supported canine left ventricle.  Circ Res  1974; 35:117.

Suga et al., 1973. Suga H, Sagawa K, Shoukas AA: Load independence of the instantaneous pressure-volume ratio of the canine left ventricle and effects of epinephrine and heart rate on the ratio.  Circ Res  1973; 32:314.

Sutherland et al., 1989. Sutherland GR, Balaji S, Monro JL: Potential value of intraoperative Doppler colour flow mapping in operations for complex intracardiac shunting.  Br Heart J  1989; 62:467.

Tabsh et al., 1982. Tabsh K, Nuwayhid B, Murad S, et al: Circulatory effects of chemical sympathectomy in fetal, neonatal, and adult sheep.  Am J Physiol  1982; 243:H113.

Tanaka et al., 1994. Tanaka H, Takata M, Yamamoto S, et al: Cardiovascular interaction during sevoflurane anesthesia in children assessed by transesophageal acoustic quantification.  Anesthesiology  1994; 81:A132.

Taylor and Lerman, 1992. Taylor RH, Lerman J: Induction, maintenance and recovery characteristics of desflurane in infants and children.  Can J Anaesth  1992; 39:6-13.

Taylor and Lerman, 1991. Taylor RH, Lerman J: Minimum alveolar concentration of desflurane and hemodynamic responses in neonates, infants, and children.  Anesthesiology  1991; 75:975.

Thach, 1985. Thach BT: Sleep apnea in infancy and childhood.  Med Clin North Am  1985; 69:1289.

Tobias, 1998. Tobias JD: Sevoflurane for controlled hypotension during spinal surgery: preliminary experience in five adolescents.  Paediatr Anaesth  1998; 8:167.

Tsuji et al., 1996. Tsuji MH, Horigome H, Yamashita M: Left ventricular functions are not impaired after lumbar epidural anaesthesia in your children.  Paediatr Anaesth  1996; 6:405.

Tulloh et al., 1997. Tulloh RM, Hislop AA, Boels PJ, et al: Chronic hypoxia inhibits postnatal maturation of porcine intrapulmonary artery relaxation.  Am J Physiol  1997; 272(5 Pt 2):H2436-H2445.

Ungerleider et al., 1989. Ungerleider RM, Greeley WJ, Kisslo J: Intraoperative echocardiography in congenital heart disease surgery: Preliminary report on a current study.  Am J Cardiol  1989; 63(suppl 1):3F.

Ungerleider et al., 1990. Ungerleider RM, Greeley WJ, Sheikh KH, et al: Routine use of intraoperative epicardial echocardiography and Doppler color flow imaging to guide and evaluate repair of congenital heart defects. A prospective study.  J Thorac Cardiovasc  1990; 100:297.

Van Grondelle et al., 1986. Van Grondelle AGS, Worthen GS, Ellis D, et al: Altering hydrodynamic variables influences PGI2 production by isolated lungs and endothelial cells.  J Appl Physiol  1986; 52:705.

Vappavouri et al., 1973. Vappavouri EK, Shinebourne EA, Williams RL, et al: Development of cardiovascular responses to autonomic blockade in intact fetal and neonatal lambs.  Biol Neonate  1973; 22:177.

Vitiello et al., 1998. Vitiello R, McCrindle BW, Nykanen D, et al: Complications associated with pediatric cardiac catheterization.  J Am Coll Cardiol  1998; 32:1433.

Wappler et al., 2003. Wappler F, Frings DP, Scholz J, et al: Inhalational induction of anaesthesia with 8% sevoflurane in children: conditions for endotracheal intubation and side-effects.  Eur J Anaesthesiol  2003; 7:548.

Wessel et al., 1993. Wessel DL, Adatia I, Giglia TM, et al: Use of inhaled nitric oxide and acetylcholine in the evaluation of pulmonary hypertension and endothelial function after cardiopulmonary bypass.  Circulation  1993; 88:2128.

Wessel et al., 1994. Wessel DL, Adatia I, Thompson JE, Hickey PR: Delivery and monitoring of inhaled nitric oxide in patients with pulmonary hypertension.  Crit Care Med  1994; 22:930.

Wodey et al., 1997. Wodey E, Pladys P, Copin C, et al: Comparative hemodynamic depression of sevoflurane versus halothane in infants: an echocardiographic study.  Anesthesiology  1997; 87:795.

Wolf et al., 1986. Wolf WJ, Neal NB, Peterson MD: The hemodynamic and cardiovascular effect of isoflurane and halothane in children.  Anesthesiology  1986; 64:328.

Wolfe et al., 1991. Wolfe RR, Loehr JP, Schaffer MS, Wiggins Jr JW: Hemodynamic effects of ketamine, hypoxia and hyperoxia in children with surgically treated congenital heart disease residing greater than or equal to 1,200 meters above sea level.  Am J Cardiol  1991; 67:84.

Wolfson et al., 1973. Wolfson B, Kielar CM, Lake C: Anesthetic index—a new approach.  Anesthesiology  1973; 38:583.

Woods et al., 1977. Woods JR, Dandavino A, Murayama K, et al: Autonomic control of cardiovascular functions during neonatal development and in adult sheep.  Circ Res  1977; 40:401.

Yardley, 1979. Yardley R: Baroreceptor regulation of blood pressure in unanesthetized fetal sheep.  Aust Paediatr J  1979; 15:286.

Zugaib et al., 1980. Zugaib M, Forsythe AB, Nuwayhid B, et al: Mechanisms of beat-to-beat variability in the heart rate of the neonatal lamb.  Am J Obstet Gynecol  1980; 138:444.

Zwass et al., 1992. Zwass MS, Fisher DM, Welborn LG, et al: Induction and maintenance characteristics of anesthesia with desflurane and nitrous oxide in infants and children.  Anesthesiology  1992; 76:373.