Clinical Anesthesia

Chapter 10

Cardiovascular Anatomy and Physiology

John P. Kampine

David F. Stowe

Paul S. Pagel

Key Points

1. The left ventricle (LV) is capable of tolerating large increases in arterial pressure without a substantial reduction in stroke volume, but the right ventricle may acutely decompensate with even modest increases in pulmonary vascular resistance.

2. Atrial contraction establishes final ventricular stroke volume at end-diastole and normally contributes between 15 and 20% of this volume.

3. Diastolic dysfunction may independently cause heart failure, even in the presence of relatively normal contractile function. This “heart failure with normal systolic function” has been increasingly recognized as a major underlying cause for as many as 50% of patients admitted to the hospital with congestive heart failure.

4. According to Starling's law, the force of LV contraction and volume of blood ejected from the chamber during systole (stroke volume) is directly related to the end-diastolic myofilament length, and hence, the end-diastolic volume.

5. The distensibility of the aorta, the resistance of the peripheral arterial vasculature, and the actions of reflected waves on the central aortic circulation are the principle determinants of afterload. Systemic vascular resistance (the ratio of pressure to cardiac output, P/Q) is the most commonly used nonparametric expression of peripheral resistance and is primarily affected by autonomic nervous system activity.

6. The primary determinant of myocardial oxygen consumption is heart rate because the heart completes an entire cycle with each beat, and hence, the more frequently the heart performs pressure-volume work, the more oxygen must be consumed.

7. The fundamental contractile unit of cardiac muscle is the sarcomere. The myofilaments within each sarcomere are arranged in parallel cross-striated bundles of thin (containing actin, tropomyosin, and the troponin complex) and thick (primarily composed of myosin and its supporting proteins) fibers. Sarcomeres are connected in series, thereby producing characteristic shortening and thickening of the long and short axes of each myocyte, respectively, during contraction.

8. Attachment of myosin to its binding site on the actin molecule releases the phosphate anion from the myosin head, thereby producing a molecular conformation within this cross-bridge structure that generates tension in both myofilaments. Release of adenosine diphosphate (ADP) and the stored potential energy from this activated conformation produce rotation of the cross-bridge (“power stroke”) at the hinge point separating the helix tail region from the globular myosin head and its associated light chain proteins.

9. The QRS complex records potentials at the body surface when the wave of depolarization is distributed throughout ventricular myocardium. The QRS complex is much larger in magnitude than the P wave because ventricular mass is greater than the atrial mass. Rapid conduction through the His-Purkinje system spreads the wave of depolarization quickly to the ventricles.


10.     Short-duration regulation of mean arterial pressure occurs through the arterial, and to a lesser extent, intracardiac baroreceptors. Arterial baroreceptors are located at the bifurcation of the common carotid arteries and in the aortic arch.

11.     Blood supply to the LV is directly dependent on the difference between the aortic pressure and LV end-diastolic pressure (coronary perfusion pressure) and inversely related to the vascular resistance to flow, which varies to the fourth power of the radius of the vessel (Poiseuille's law).

12.     Metabolic factors are the major physiological determinants of coronary vascular tone and, hence, myocardial perfusion.

13.     Myocardial infarction may occur without evidence of major coronary thromboses, emboli, or stenosis. This form of infarction is caused by excessive metabolic demands resulting from severe LV hypertrophy (e.g., critical aortic stenosis) or vasoactive drug ingestion (e.g., amphetamines, cocaine) or it may also result from coronary artery vasospasm.

14.     The lung is richly innervated by the parasympathetic and sympathetic nervous system, but the dominant effect of the autonomic nervous system occurs primarily at the level of alveolar and bronchial smooth muscle.

15.     Cerebral blood flow remains relatively constant when mean arterial pressure varies between 50 and 150 mm Hg in healthy subjects. This autoregulation of cerebral blood flow is shifted to the right in patients with chronic, poorly controlled essential hypertension.

16.     Arterial CO2 tension is a major regulator of cerebral blood flow within the physiologic range of arterial CO2 tension. Cerebral blood flow linearly increases 1 to 2 mL/100 g/min for each 1 mm Hg increase in Paco2. Below an arterial CO2 tension of 25 mm Hg, the cerebral blood flow response to Paco2 is attenuated.

Functional Anatomy of the Heart

The left and right atria consist of two, thin overlying sheaths of muscle oriented at right angles to each other. The two thicker-walled ventricles consist of three interdigitating muscle layers: the deep sinospiral, the superficial sinospiral, and the superficial bulbospiral muscles (Fig. 10-1). The two outer muscle layers are oriented obliquely from the base of the heart to the apex. Constriction of these fibers shortens the longitudinal axis of the left ventricle (LV) by moving the base toward the apex. The circumferential deep sinospiral muscles reduce the LV diameter (Fig. 10-2). Thus, synchronous contraction of the LV muscles shortens the long axis of the heart, decreases the circumference of the LV chamber, and lifts the apex toward the anterior chest wall. This latter action produces the familiar palpable point of maximum impulse, which is normally located in the fifth or sixth intercostal space in the midclavicular line.1,2 The LV pumps blood from the low-pressure venous into the high-pressure arterial system. The right ventricle (RV) receives venous blood from the right atrium via the superior and inferior vena cavae at low pressure (2 to 10 mm Hg) and oxygen saturation (60 to 75%). The RV is crescent-shaped and contains embryologically distant inflow and outflow tracts that contract in a peristaltic sequence to propel blood into the pulmonary arterial tree. Blood flow through the pulmonary circulation functions primarily as a gas exchanger, providing for the elimination of carbon dioxide (CO2; a major product of cellular metabolism) and the uptake of oxygen (O2). The pulmonary vasculature is characterized by lower pressure than the systemic circulation and has shorter, larger-bore blood vessels with relatively thinner walls than systemic resistance vessels. Thus, the pulmonary circulation is a low-pressure, low-resistance system into which the RV transfers blood. The LV is capable of tolerating large increases in arterial pressure without a substantial reduction in stroke volume; the RV may acutely decompensate with even modest increases in pulmonary vascular resistance. The RV free wall occupies a more right-sided, anterior position within the mediastinum compared with the position of the thicker-walled LV that is located in a left-sided, posterior orientation (Fig. 10-3, A and B). During contraction, the RV moves toward the interventricular septum with a “bellows-like” action. The atrioventricular (AV) groove


separating the right atrium and RV shortens toward the apex during contraction. This anatomic configuration permits the more flexible RV wall to eject a large volume of blood with a minimal amount of shortening. The echocardiographic depiction of ventricular contraction is shown in Figure 10-3.

Figure 10-1. Components of the myocardium. The outer muscle layers pull the apex of the heart toward the base. The inner circumferential layers constrict the lumen, particularly of the left ventricle. (Reproduced with permission from Rushmer RF: Cardiovascular Dynamics. Philadelphia, WB Saunders, 1976, Fig. 3-2, p 78.)

Figure 10-2. Ventricular volume ejection. Contraction characteristics and modes of emptying. The volumes ejected by each ventricle is equal but the left ventricle requires a more circumferential muscular wall to eject its volume at a pressure that is approximately 4 to 5 times greater than that in the right ventricle. (Reproduced with permission from Rushmer RF: Cardiovascular Dynamics. Philadelphia, WB Saunders, 1976, Fig. 3-12, p 92.)

Figure 10-3. Transesophageal echocardiography demonstrates the thickness and motion of atrial and ventricular walls mid esophageal five-chamber and two chamber views (Aand B respectively).

The LV has a cylindrical endocardial border, and this anatomic configuration provides a mechanical advantage over the RV in generating stroke work and power because a reduction in the cross-sectional area of the cylinder a (function of the square of the radius) is partially responsible for LV stroke volume. The LV also provides a splint against which the outer wall of the RV is pulled during contraction. The O2 content and saturation of blood within the LV (O2 20 mL/dL and 98%, respectively) is very high compared with blood in the RV. Left ventricular O2 saturation is incomplete because a small quantity of coronary venous return through thebesian veins empties directly into the left side of the heart. During contraction, LV pressure increases from end-diastolic values of 10 to 12 mm Hg to a peak pressure of 120 to 140 mm Hg during systole. The peak pressures generated by the LV reflect the requirement to circulate blood through the high-resistance systemic circulation that is composed of thicker blood vessels containing larger quantities of vascular smooth muscle than their counterparts in the pulmonary arterial tree. Resistance to blood flow is especially high in small arterioles and precapillary vessels, and blood flow in these vessels requires that the LV generate higher perfusion pressure than the RV. The volume of blood pumped by RV and LV is identical (stroke volume), but the pressure-volume work (stroke work) performed by the LV is 5 to 7 times greater than that of the RV. Left ventricular ejection is associated with a wall tension gradient from the apex to the base of the heart (aortic outflow tract), thereby producing the intraventricular gradient required to transfer stroke volume from the LV into the aorta.

Efficient pumping action of the heart requires two pairs of unidirectional valves. One pair is located at the outlets of the RV and LV (pulmonic and aortic valves, respectively). These three-leaflet valves operate passively with changes in pressure gradients. The aortic valve leaflets do not flatten against the aortic wall during LV ejection because a modest dilation of the aortic root located immediately distal to each leaflet establishes an eddy current of blood flow. These dilated regions are termed the sinuses of Valsalva and permit blood flow through the right and left main coronary arteries whose openings are located in the aortic wall directly behind the valve cusps. The AV valves separating the atria from the ventricles are the tricuspid and mitral valve on the right and left sides of the heart, respectively. The mitral valve is the only cardiac valve with two leaflets. Both tricuspid and mitral valves are thin, fibrous structures that are supported by chordae tendinae attachments to papillary muscles that are part of the ventricular musculature and contract during systole. The tricuspid and mitral valves open and close with alternations in the pressure gradients between the corresponding atrial and ventricular chambers.

The RV and LV are the major cardiac pumping chambers, but the atria play critically important supporting roles. The atria function as reservoirs, conduits, and contractile chambers and facilitate the transition between continuous, low-pressure venous to phasic, high-pressure arterial blood flow. The normal atrial pressure curve has three positive reflections. Shortly after the onset of atrial depolarization (indicated by the P wave of the electrocardiogram), the atria contract, producing a positive pressure wave (the A wave) late in diastole. At the onset of systole, ventricular contraction produces another pressure wave that is transmitted through the AV valves to the atria, resulting in the C wave. During the remainder of systole, the AV valves remain closed, atrial filling continues from peripheral and pulmonary veins, and atrial pressures rise, thereby producing a positive pressure deflection known as the V wave (Fig. 10-4). Atrial contraction establishes final ventricular stroke volume at end-diastole and normally contributes between 15 and 20% of this volume. When atrial contraction is absent or ineffective (e.g., atrial failure, atrial fibrillation or flutter), the heart may be capable of compensating for the loss of the atrial contractile function and continue to function effectively under resting conditions. However, during increased physical activity or stress, the absence of the atrial pump may substantially limit cardiac output, thereby causing a marked reduction in arterial blood pressure accompanied by syncope, exertional dyspnea, easy fatigability, or acute heart failure.

The Cardiac Cycle

The cardiac cycle is traditionally defined based on events occurring before, during, and after LV contraction. Left ventricular systole is commonly divided into three parts: isovolumic contraction, rapid ejection, and slower ejection.2,3 Closure of both the tricuspid and mitral valves occurs when RV and LV pressures exceed corresponding atrial pressure and is the source of the first heart sound (S1; Fig. 10-4). Isovolumic


contraction is the interval between closure of the mitral valve and the opening of the aortic valve. Left ventricular volume remains constant during this period of the cardiac cycle. The rate of increase of LV pressure (dP/dt, an index of myocardial contractility) reaches its maximum during isovolumic contraction. True isovolumic contraction does not occur in the RV because the sequential nature of inflow followed by outflow tract RV contraction. Pressure in the aortic root declines to its minimum value immediately before the aortic valve opens. Rapid ejection occurs when LV pressure exceeds aortic pressure and the aortic valve opens. Approximately two thirds of the LV end-diastolic volume is ejected into the aorta during this rapid ejection phase of systole. Aortic dilation occurs in response to this rapid increase in volume as the kinetic energy of LV contraction is transferred to the systemic arterial circulation as potential energy. The compliance of the aorta and proximal great vessels determines the amount of potential energy that can be stored and subsequently released to the arterial vasculature during diastole. The normal LV end-diastolic volume is about 120 mL. The average ejected stroke volume is 80 mL, and the normal ejection fraction is approximately 67%. A decrease in ejection fraction below 40% is typically observed when the myocardium is affected by ischemia, infarction, or cardiomyopathic disease processes (e.g., myocarditis, amyloid infiltration). Contractile dysfunction may also occur as a result of chronic pressure or volume overload, diabetes, or hypothyroidism. As aortic pressure peaks and resists further LV ejection, transfer of further stroke volume slows and eventually stops. During this period of slower ejection, aortic pressure may briefly exceed LV pressure. The reversal of the pressure gradient between the aortic root and the LV causes the aortic valve to close, thereby producing the second heart sound (S2).

Figure 10-4. Mechanical and electrical events of the cardiac cycle showing also the ventricular volume curve and the heart sounds. Note the isovolumic contraction (ICP) and the relaxation period (IRP) during which there is no change in ventricular volume because all valves are closed. The ventricle decreases in volume as it ejects its contents into the aorta. During the first third of systolic ejection—the rapid ejection period—the curve of emptying is steep. ECG, electrocardiogram. (Reproduced with permission from Smith JJ, Kampine JP: Circulatory Physiology—The Essentials, 3rd edition. Baltimore, Williams & Wilkins, 1990, Fig. 3-5, p 40.)

Diastole is divided into four phases in the LV: isovolumic relaxation, early filling, diastasis, and atrial systole. Isovolumic relaxation defines the period between aortic valve closure and mitral valve opening during which LV volume remains constant. LV pressure falls precipitously as the myofilaments relax. When LV pressure falls below left atrial pressure, the mitral valve opens, and blood volume stored in the left atrium rapidly enters the LV driven by the pressure gradient between these chambers. This early-filling phase of diastole accounts for approximately 70 to 75% of total LV stroke volume available for the subsequent contraction. Delays in LV relaxation occur as a consequence of aging or disease process (e.g., myocardial ischemia) and may attenuate early ventricular filling. After left atrial and LV pressures have equalized, the mitral valve remains open and pulmonary venous return continues to flow through the left atrium into the LV. This phase of diastole is known as diastasis, during which the left atrium functions as a conduit. Tachycardia progressively shortens and may completely eliminate this phase of diastole. Diastasis accounts for no more than 5% of total LV end-diastolic volume under normal circumstances. The final phase of diastole is atrial systole. Contraction of the left atrium contributes the remaining blood volume (approximately 15 to 20%) used in the subsequent LV systole. Disease processes known to reduce LV compliance (e.g., myocardial ischemia, pressure-overload hypertrophy) attenuate early filling and increase the importance of atrial systole to overall LV filling. Thus, loss of normal sinus rhythm may precipitate catastrophic decreases in cardiac output in patients with symptomatic coronary artery disease, critical aortic stenosis, or poorly controlled chronic essential hypertension.6

The importance of diastole to overall cardiac performance cannot be understated. The rate and extent of relaxation, the viscoelastic properties of LV myocardium, the pericardium, and the structure and function of the left atrium, pulmonary venous circulation, and mitral valve determine the timing, rate, and degree of LV filling. The ability of the LV to adequately collect blood from the low-pressure pulmonary venous circulation is critical in determining the stroke volume that can be transferred to the arterial circulation during systole. Thus, diastolic dysfunction may independently cause heart failure, even in the presence of relatively normal contractile function. This “heart failure with normal systolic function” has been increasingly recognized as a major underlying cause for as many as 50% of patients admitted to the hospital with congestive heart failure.8,9

Determinants of Cardiac Output

Cardiac output is the amount of blood pumped by the heart per minute. It is the product of heart rate and stroke volume and may be normalized to the body surface area (cardiac index). Cardiac output (Q) is directly related to pressure (P) and inversely related to peripheral vascular resistance (R) using an equation analogous to Ohm's law: Q = P/R. Cardiac output is a function of preload, afterload, myocardial


contractility (inotropic state), and heart rate. Preload is defined by LV end-diastolic volume in the intact heart and reflects the stretch of ventricular myofilaments produced by this end-diastolic volume immediately before the onset of contraction. According to Starling's law, the force of LV contraction and volume of blood ejected from the chamber during systole (stroke volume) is directly related to the end-diastolic myofilament length, and hence, the end-diastolic volume.10,11 Thus, the ventricular myocardium behaves similar to skeletal muscle in that an increase in initial stretch determines the subsequent force of contraction. Afterload may be simplistically represented as the aortic pressure against which the LV must propel blood. The distensibility of the aorta, the resistance of the peripheral arterial vasculature, and the actions of reflected waves on the central aortic circulation are the principle determinants of afterload. Systemic vascular resistance (the ratio of pressure to cardiac output, P/Q) is the most commonly used nonparametric expression of peripheral resistance and is primarily affected by autonomic nervous system activity. For example, an increase in sympathetic nervous system tone produces vasoconstriction of peripheral resistance arterioles through activation of α1-adrenoceptors in vascular smooth muscle, thereby augmenting afterload. A brief, large increase in afterload may cause a transient decrease in stroke volume, but a compensatory increase in preload during successive cardiac cycles restores cardiac output by increasing LV force of contraction.

Inotropic state is the intrinsic force of myocardial contraction independent of changes in preload, afterload, or heart rate. The number of cross bridges between the contractile elements and the relative sensitivity of the contractile elements to activator Ca2+ play important roles in determining inotropic state. In the intact heart, a positive inotropic effect is reflected by an increase in pressure-volume work at each end-diastolic volume. Such an increase in inotropic state may occur in response to an increase in cardiac sympathetic nerve activity through stimulation of β1-adrenoceptors. Pharmacologic increases in contractility may be produced by drugs that activate β1-adrenoceptors (e.g., dobutamine) or by those that prevent metabolism of the intracellular second messenger cyclic adenosine monophosphate (cAMP; e.g., milrinone). Cardiac output is also influenced by heart rate. The primary determinant of myocardial oxygen consumption is heart rate because the heart completes an entire cycle with each beat, and hence, the more frequently the heart performs pressure-volume work, the more oxygen must be consumed. The upper and lower limits of heart rate may influence cardiac output. At low heart rates (except in trained athletes), there simply may not be adequate cardiac output to meet the body's oxygen requirements, deliver substrates for metabolism, or remove products of cellular metabolism. In contrast, at high heart rates, particularly in patients with heart disease, there may not be adequate diastolic filling time to maintain cardiac output and coronary artery perfusion, the latter of which is particularly dependent on duration of diastole. Thus, shortened diastolic time during profound tachycardia may reduce stroke volume and cardiac output, contribute to hypotension, and decrease the duration of coronary perfusion. Such events may cause acute myocardial ischemia or infarction.

Measures of Cardiac Function

Clinical indicators of contractile performance include cardiac output, ejection fraction, fractional shortening or area change of the LV short axis, and LV systolic wall thickening. These indices of contractility are heart rate-, preload-, and afterload-dependent, but nevertheless may be measured with reasonable reliability using echocardiographic techniques and remain useful indices of contractile performance, especially in the presence of chronic heart disease, during recovery after an acute ischemic event, and in patients undergoing cardiac surgery. More sophisticated methods of assessing myocardial contractility in vivo, including the LV end-systolic pressure-volume relations and preload recruitable stroke work, require invasive measurement of continuous LV pressure and volume.10,12,13,14,15,16 Preload recruitable stroke work and the effects of isoflurane are shown in Figure 10-5. These techniques are usually assessed only in a laboratory setting, but may also be obtained using echocardiography (automated border detection) combined with invasive determination of continuous LV pressure during cardiac catheterization. Discussion of indices of contractile state derived from pressure-volume relations are beyond the scope of the current chapter.

Figure 10-5. Preload recruitable stroke work (PRSW) relationship for control (C) and 1.5 and 2 minimal alveolar concentrations (MAC) of isoflurane. PRSW is plotted against end diastolic length (EDL). The inset depicts PRSW done at a constant end-diastolic length of 20 mm (PRSW20) and is represented as a percent of control. *Significantly (p < 0.05) different than control; significantly (p < 0.05) different than 1.5 MAC isoflurane; significantly (p < 0.05) different slope than control. (Reproduced with permission from Pagel PS, Kampine JP, Schmeling WT, Warltier DC: Comparison of end-systolic pressure-length relations and preload recruitable stroke work as indices of myocardial contractility in the conscious and anesthetized, chronically instrumented dog. Anesthesiology 1990; 73: 278.)

Cellular and Molecular Biology of Cardiac Muscle Contraction

Ultrastructure of the Cardiac Myocyte

The heart contracts and relaxes nearly 3 billion times during an average lifetime, based on a heart rate of 70 beats per minute and a life expectancy of 75 years. A review of cardiac myocyte ultrastructure provides important insights into how the heart accomplishes this astonishing performance. The sarcolemma is the external membrane of the cardiac muscle cell. The sarcolemma contains ion channels (e.g., Na+, K+, Ca2+), ion pumps and exchangers (e.g., Na+-K+ ATPase, Ca2+-ATPase, Na+-Ca2+ or -H+ exchangers), G protein-coupled and other receptors (e.g., β1-adrenergic, adenosine, opioid), and transporter enzymes that regulate intracellular ion concentrations, facilitate signal transduction, and provide metabolic substrates


required for energy production. Deep invaginations of the sarcolemma, known as transverse (“T”tubules, penetrate the internal structure of the myocyte at regular intervals, thereby assuring rapid, uniform transmission of the depolarizing impulses that initiate contraction to be simultaneously distributed throughout the cell. Unlike the skeletal muscle cell, the cardiac myocyte is densely packed with mitochondria, which are responsible for generation of the large quantities of high-energy phosphates (e.g., adenosine triphosphate [ATP]) required for the heart's phasic cycle of contraction and relaxation. The fundamental contractile unit of cardiac muscle is the sarcomere. The myofilaments within each sarcomere are arranged in parallel cross-striated bundles of thin (containing actin, tropomyosin, and the troponin complex) and thick (primarily composed of myosin and its supporting proteins) fibers. Sarcomeres are connected in series, thereby producing characteristic shortening and thickening of the long and short axes of each myocyte, respectively, during contraction.

Figure 10-6. Schematic illustration of the myosin molecule demonstrating double helix tail, globular heads that form cross bridges with actin during contraction, two pairs of light chains, and “hinges” (cleavage sites of proteolytic enzymes) that divide the molecule into meromyosin fragments (see text). (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 4-1, p 104.)

The structure of each sarcomere is described based on observations from light and electron microscopy. The area of overlap of thick and thin fibers characterizes the “A” band. This band lengthens as the sarcomere shortens during contraction. The “I” band represents the region of the sarcomere that contains thin filaments alone, and this band is reduced in width as the cell contracts. Each “I” band is bisected by a “Z” (from the German zuckung [twitch]) line, which delineates the border between two adjacent sarcomeres. Thus, the length of each sarcomere contains a complete “A” band and two one-half “I” band units located between “Z” lines. A central “M” band is also present within the “A” band and is composed of thick filaments spatially constrained in a cross-sectional hexagonal matrix by myosin binding protein C. An extensively intertwined network of sarcoplasmic reticulum (SR) invests each bundle of contractile proteins and functions as a Ca2+ reservoir, thereby assuring homogenous distribution and reuptake of activator Ca2+ throughout the myofilaments during contraction and relaxation, respectively. The subsarcolemmal cisternae of the SR are specialized structures located immediately adjacent to, but not continuous with, the sarcolemmal and transverse tubular membranes and contain large numbers of ryanodine receptors that function as the primary Ca2+ release channel for the SR. The contractile machinery and the mitochondria that power it occupy >80%, whereas the cytosol and nucleus fill <15%, of the total volume of the cardiac myocyte. It is abundantly clear based on this simple observation that contraction and relaxation, and not new protein synthesis, are the predominant functions of the cardiac myocyte. Intercalated discs mechanically connect adjacent myocytes through the fascia adherens and desmosomes, which link actin and other proteins between cells, respectively. The intercalated discs also provide a seamless electrical connection between myocytes via large, nonspecific ion channels known as gap junctions that facilitate intercellular cytosolic diffusion of ions and small molecules.

Proteins of the Contractile Apparatus

Myosin, actin, tropomyosin, and the three-protein troponin complex compose the six major components of the contractile apparatus. Myosin (molecular weight of approximately 500 kDa; length, 0.17 µm) contains two interwoven chain helices with two globular heads that bind to actin and two additional pairs of light chains. Enzymatic digestion of myosin divides the structure into light (containing the tail section of the complex) and heavy (composed of the globular heads and the light chains) meromyosin. The elongated tail section of the myosin complex functions as the architectural support of the molecule (Fig. 10-6). The globular heads of the myosin dimer contain two “hinges” located at the junction of the distal light chains and the tail helix that play a critical role in myofilament shortening during contraction. These globular structures bind to actin, thereby activating an ATPase that plays a central role in hinge rotation and release of actin during contraction and relaxation, respectively. The maximum velocity of sarcomere shortening has been shown to be dependent on the activity of this actin-activated myosin ATPase. Notably, adult and neonatal atrial and ventricular myocardium contain several different myosin ATPase isoforms that are distinguished by their relative ATPase activity. The myosin molecules are primarily arranged in series along the length of the thick filament, but are abutted “tail-to-tail” in the center of the thick filament. This orientation facilitates shortening of the distance between Z lines during contraction as the thin filaments are drawn progressively toward the center of the sarcomere.

The light chains contained within the myosin complex serve either “regulatory” or “essential” roles. Regulatory myosin light chains may favorably modulate myosin-actin interaction through Ca2+-dependent protein kinase phosphorylation, whereas essential light chains serve an as yet undefined obligate function in myosin activity, as their removal denatures the myosin molecule. Discussion of myosin light chain isoforms is beyond the scope of the current chapter, but isoform switches from ventricular to atrial forms have been observed in left ventricular hypertrophy that may contribute to contractile dysfunction.17 In addition to myosin and its binding protein, thick filaments contain titin, a long elastic protein that attaches myosin to the Z lines. Titin has been postulated to be a “length sensor” similar to a bidirectional spring that establishes progressively greater passive restoring forces as sarcomere length approaches


its maximum or minimum.18 Compression and stretching of titin occur during decreases and increases in muscle load, thereby resisting further sarcomere shortening and lengthening, respectively. Thus, titin is a third important elastic element (in addition to actin and myosin) that contributes to the stress-strain biomechanical properties of cardiac muscle.19

Figure 10-7. Cross-sectional schematic illustration demonstrating the structural relationship between the troponin-tropomyosin complex and actin under resting conditions (left) and after Ca2+ binding to troponin C (right; see text). (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 4-15, p 117.)

Actin is the major component of the thin filament. Actin is a 42-kDa, ovoid-shaped, globular protein (“G” form; 5.5 nm in diameter) that exists in a polymerized filamentous (“F”) form in cardiac muscle. F-actin binds adenosine diphosphate (ADP) and a divalent cation (Ca2+ or Mg2+), but unlike myosin, the molecule does not directly hydrolyze high-energy nucleotides such as ATP. F-actin is wound in double-stranded helical chains of G-actin monomers that resemble two intertwined strands of pearls (each G-actin monomer; Fig. 10-6). A single complete helical revolution of filamentous actin is approximately 77 nm in length and contains 14 G-actin monomers. Actin derives its name from its function as the “activator” of myosin ATPase through its reversible binding with myosin. The hydrolysis of ATP by this actin-myosin complex provides the chemical energy required to produce the conformational changes in the myosin heads that drive the cycle of contraction and relaxation within the sarcomere. Tropomyosin is one of two major inhibitors of actin-myosin interaction. Tropomyosin (length of 40 nm; weight between 68 and 72 kDa) is a rigid double-stranded α-helix protein linked by a single disulfide bond. Human tropomyosin contains both α and β isoforms (34 and 36 kDa, respectively) and may be present as either a homo- or heterodimer.20 Tropomyosin stiffens the thin filament through its position within the longitudinal cleft between intertwined F-actin polymers (Fig. 10-7), but its Ca2+-dependent interaction with troponin complex proteins is the mechanism that links sarcolemmal membrane depolarization to actin-myosin interaction in the cardiac myocyte (excitation-contraction coupling). The thin filaments are anchored to Z lines by cytoskeletal proteins including α- and β-actinin and nebulette.21,22

Figure 10-8. Schematic illustration demonstrating the location of tropomyosin interlaced within the groove formed by two F-actin chains. (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 4-16, p 108.)

The troponin proteins serve complementary but distinct roles as critical regulators of the contractile apparatus.23 The troponin complexes are arranged at 40-nm intervals along the length of the thin filament. Troponin C (so named because this molecule binds Ca2+) exists in a highly conserved, single isoform in cardiac muscle. Troponin C is composed of a central nine-turn a-helix separating two globular regions that contain four discrete amino acid sequences capable of binding divalent cations including Ca2+ and Mg2+. Of this quartet of amino acid-cation binding sequences, two (termed sites I and II) are Ca2+-specific, thereby allowing the troponin C molecule to respond to the acute changes in intracellular Ca2+concentration that accompany contraction and relaxation. Troponin I (“inhibitor”) is a 23-kDa protein that exists in a single isoform in cardiac muscle. Troponin I alone weakly prevents the interaction between actin and myosin, but when combined with tropomyosin, the troponin I-tropomyosin complex becomes the major inhibitor of actin-myosin binding. The troponin I molecule contains a serine residue that may be phosphorylated by protein kinase A (PKA) via the intracellular second messenger cAMP, thereby reducing troponin C-Ca2+ binding and enhancing relaxation during administration of β-adrenoceptor agonists (e.g., dobutamine) or phosphodiesterase fraction III inhibitors (e.g., milrinone). Troponin T (so denoted because it binds other troponin molecules and tropomyosin) is the largest of the troponin proteins and exists in four major isoforms in human cardiac muscle. Troponin T anchors the other troponin molecules and may also influence the relative Ca2+ sensitivity of the complex.24

Calcium-Myofilament Interaction

Binding of Ca2+ to troponin C precipitates a series of conformational changes in the troponin-tropomyosin complex that lead to the exposure of the myosin binding site on the actin molecule. During conditions in which intracellular Ca2+ concentration is low (10-7 M; diastole), very little Ca2+ is bound to troponin C, and each tropomyosin molecule is constrained to the outer region of the groove between F-actin filaments by a troponin complex (Fig. 10-8). This structural configuration prevents myosin-actin interaction by effectively blocking cross-bridge formation. Thus, an inhibitory state produced by the troponin-tropomyosin complex exists in cardiac muscle under resting conditions. A 100-fold increase in intracellular Ca2+ concentration (10-5 M; systole) occurs as a consequence of sarcolemmal depolarization, which opens L- and T-type sarcolemmal Ca2+ channels, thereby allowing Ca2+ influx into the myocyte from the extracellular compartment and stimulating Ca2+-dependent Ca2+ release from the SR via its ryanodine receptors. When Ca2+ is bound to troponin C under these conditions, the shape of the troponin C protein becomes elongated and its interactions with troponin I and T are enhanced. These Ca2+-induced allosteric rearrangements in troponin complex structure weaken the interaction between troponin I and actin, allow repositioning of the tropomyosin


molecule along the F-actin filaments, and reverse the baseline inhibition of actin-myosin binding by tropomyosin.25 In this way, Ca2+ binding to troponin C may be directly linked to a series of changes in regulatory protein chemical structure that block inhibition of the binding site for myosin on the actin molecule and allow cross-bridge formation and contraction to occur. This antagonism of inhibition is fully reversible, as relaxation is facilitated by dissociation of Ca2+ from troponin C concomitant with rapid restoration of the original conformation of the troponin-tropomyosin complex on F-actin.

Most Ca2+ ions are removed from the myofilaments and the cytosol after membrane repolarization by a Ca2+-ATPase located in the SR membrane (sarcoendoplasmic reticulum Ca2+-ATPase, SERCA). This Ca2+ is stored (concentration of approximately 10-3 M) in the SR bound to calsequestrin and calreticulin until the subsequent sarcolemmal depolarization is initiated. The Na+/Ca2+ exchanger and a Ca2+-ATPase located within the sarcolemmal membrane also remove a small quantity of Ca2+, similar to that which originally entered the myocyte from the extracellular space during depolarization. Phospholamban is a small protein (6 kDa) located in the SR membrane that partially inhibits the activity of the dominant form (type 2a) of cardiac SERCA under baseline conditions. However, phosphorylation of this protein by PKA blocks this inhibition and enhances the rate of SERCA uptake of Ca2+ into the SR,26 thereby increasing the rate and extent of relaxation (positive lusitropic effect) and augmenting the amount of Ca2+ stored for the next cycle of contraction (positive inotropic effect). Thus, SERCA activity is regulated by a cAMP-dependent PKA that is responsive to β-adrenoceptor stimulation or phosphodiesterase fraction III inhibition. In addition to PKA-mediated phosphorylation of troponin I that facilitates Ca2+ release from troponin C, these observations explain why positive inotropic drugs such as dobutamine and milrinone also augment relaxation.

Myosin-Actin Contraction Biochemistry

The biochemistry of cardiac muscle contraction is most often described using a simplified four-component model (Fig. 10-9).27 Binding of ATP with high affinity to the catalytic domain of myosin initiates the series of chemical and mechanical events that cause contraction of the sarcomere to occur. The myosin ATPase enzyme hydrolyzes the ATP molecule into ADP and inorganic phosphate, but these reaction products do not immediately dissociate from myosin. Instead, the ATP hydrolysis products and myosin form an “active” complex that retains the chemical energy released from the reaction as potential energy. In the absence of actin, subsequent dissociation of ADP and phosphate from myosin is the rate-limiting step of myosin ATPase and the muscle remains relaxed. However, the activity of myosin ATPase is markedly accelerated when the myosin-ADP-phosphate complex is bound to actin, and under these circumstances, the chemical energy obtained from ATP hydrolysis becomes directly transferred into mechanical work. Attachment of myosin to its binding site on the actin molecule releases the phosphate anion from the myosin head, thereby producing a molecular conformation within this cross-bridge structure that generates tension in both myofilaments.28 Release of ADP and the stored potential energy from this activated conformation produce rotation of the cross-bridge (“power stroke”) at the hinge point separating the helix tail region from the globular myosin head and its associated light chain proteins. Each cross-bridge rotation generates 3 to 4 × 10-12 newtons of force29 and moves myosin approximately 11 nm along the actin molecule. Completion of myosin head rotation and ADP release does not dissociate the myosin-active complex, but leaves it in a low-energy bound (“rigor”) state. Separation of myosin and actin occurs when a new ATP molecule binds to myosin, and the process is subsequently repeated, provided that energy supply is adequate and the myosin-binding site on actin remains unimpeded by troponin-tropomyosin inhibition.

Figure 10-9. Schematic illustration of the actin filaments and its individual monomers and active myosin bindings sites (mleft panel). The myosin head is dissociated from actin by binding with adenosine triphosphate (ATP). Subsequent ATP hydrolysis and release of inorganic phosphate (Pi) “cocks” the head group into a tension-generating configuration. Attachment of the myosin head to actin allows the head to apply tension to the myosin rod and the actin filament. The right panel illustrates calcium binding to troponin C, which causes troponin I to decrease its affinity for actin. As a result in a conformational shift in tropomyosin position (see text), seven sites on actin monomers are revealed.

Several factors may affect the efficiency of cross-bridge biochemistry and myocardial contractility independent of autonomic nervous system tone or administration of exogenous vasoactive drugs. There is a direct relationship between myosin ATPase activity and the maximal velocity of unloaded muscle shortening (Vmax), and the normal increase in intracellular Ca2+ concentration (from 10-7 to 10-5 M) that occurs after sarcolemmal depolarization enhances baseline myosin ATPase activity fivefold before it interacts with actin, thereby increasing Vmax. Contractile force depends on sarcomere length immediately before sarcolemmal depolarization. This


length-dependent activation (Frank-Starling effect) may be related to an increase in myofilament sensitivity to Ca2+, favorable alterations in spacing between myofilaments, or titin-induced elastic recoil. Abrupt increases in load during contraction (Anrep effect) or those that occur after a prolonged pause between beats (Woodworth phenomenon) causes transient increases in contractile force through a length-dependent activation mechanism. An increase in cardiac muscle stimulation frequency also augments contractile force (treppe phenomenon) via enhanced myofilament Ca2+ sensitivity and greater SR Ca2+ release.

Electrical Properties of the Heart

The Clinical Electrocardiogram

The clinical electrocardiogram (ECG) consists of a regular series of deflections from the isoelectric line. The first deflection of the ECG is the P wave (Einthoven began his depiction of the ECG in the middle of the alphabet). The P wave is a positive deflection that occurs as a consequence of atrial depolarization. The initial electrical event is depolarization of the sinoatrial (SA) node pacemaker cells and is followed almost immediately by progressive depolarization of both atria. The SA node pacemaker activity is not observed on the ECG because the node is too small to generate electrical potential differences large enough to be recorded from the body surface. The duration of the P wave is the time required for depolarization to spread over the atria and may be prolonged by atrial enlargement or a conduction delay. The SA node is located in the wall of the right atrium at the junction of this chamber and the superior vena cava. Propagation of the depolarizing impulse throughout the atria is not uniform, as a slightly higher conduction velocity occurs through the anterior, middle, and posterior internodal pathways between the SA and the AV nodes. Activation and depolarization of the AV node begins during the P wave before depolarization of the atria is completed (Fig. 10-10).7,30,31

The P wave is followed by a brief interval returning to the isoelectric line. The PR interval is the duration between the onset of the P wave and the beginning of ventricular depolarization signified by the onset of the QRS complex (Fig. 10-11). Prolongation of the PR interval usually indicates a conduction delay between atrial and ventricular conduction. After the P wave is complete, the ECG becomes isoelectric because changing potential differences within the heart are no longer recorded at the body surface as a result of the relatively small mass of tissue that continues the depolarization-conduction process. During this apparent “silent” interval between atrial and ventricular depolarization, the wave of depolarization is being conducted through the AV node, AV bundle, right and left bundle branches, and His-Purkinje fiber network. The conduction velocity through the AV node is relatively slow. In contrast, conduction velocity is very rapid in the His-Purkinje system (H in Fig. 10-11), approaching the velocity observed in small nerves. The QRS complex records potentials at the body surface when the wave of depolarization is distributed throughout ventricular myocardium. The QRS complex is much larger in magnitude than the P wave because ventricular mass is greater than the atrial mass. Rapid conduction through the His-Purkinje system spreads the wave of depolarization quickly to the ventricles. Delays in this conduction distal to the AV node most often result from intrinsic myocardial disease (most notably, ischemia) and may have profound consequences on cardiac rhythm and LV contractile synchrony.

Figure 10-10. The electrocardiogram (ECG). Major waves (P, QRS, and T) of the ECG are indicated as well as the timing of the activation of some of the key conductive structures. SA, sinoatrial. (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 15-10, p 436.)

Figure 10-11. Top: Electrocardiogram recorded from the body surface. Bottom: Intracardiac electrogram. (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 15-9, p 435.)

The ST segment is the interval between the end of the QRS complex and the T wave. The ST segment is normally isoelectric because all of the ventricular myocardium is depolarized. The ST segment also reflects the long plateau phase of the cardiac action potential. The injury current of an elevated or depressed ST segment observed during myocardial ischemia or infarction may occur as a result of an abbreviated action potential within the ischemic region or because depolarizing currents propagate more slowly through the ischemic zone. Repolarization of the ventricles generates the T wave, which corresponds to the end of phase 2 and all of phase 3 of the cardiac action potential (see later discussion). The duration of the T wave is considerably longer than the QRS complex because, unlike the rapidly transmitted, nearly homogenous ventricular depolarization, repolarization occurs more slowly and is less synchronous. The QT interval is the duration between the onset of ventricular depolarization (indicated by the QRS complex) and completion of repolarization (as signified by the end of the T wave). The QT interval varies inversely with heart rate, and may precipitate malignant ventricular arrhythmias when shortened or prolonged by administration of vasoactive


drugs (e.g., volatile anesthetics) or in the presence of intrinsic cardiac pathology (e.g., prolonged QT syndrome).30,31

Role of Ion Channels

The action potentials of individual groups of excitable cardiac myocytes are quite different (Fig. 10-12). The SA and AV nodes and accessory pacemaker cells have unstable, spontaneously depolarizing properties. The resting membrane potential of these cells is not -90 mV, as observed in typical atrial and ventricular myocytes or His-Purkinje fibers. Spontaneous, phase 4 slow depolarization of SA and AV node cells is initiated at membrane potentials between -55 and -66 mV. The SA node, AV node, and the remaining specialized conduction tissue of the heart are all characterized as potential pacemakers, but the SA node is the normal cardiac pacemaker because of its intrinsically faster discharge rate. Cells within the SA node are not homogenous, and some of these pacemaker cells have faster discharge frequencies than others. The resting membrane potential of cardiac pacemaker cells is unstable and displays a slow depolarization of the membrane during diastole (Fig. 10-12). The rate of rise of the action potential from threshold (phase 1) is relatively slow in SA nodal cells compared with atrial and ventricular muscle cells. The magnitude and slope of spontaneous depolarization (also known as automaticity) of SA node cells are important in the regulation of heart rate and depend on the activity of the sympathetic and vagal (parasympathetic) neural innervation. Slowing the rate of depolarization increases the time to reach the threshold potential (TP) and decreases heart rate (SA node rate of discharge; Fig. 10-13). The heart rate may also slow as a result of a shift in threshold potential to a higher level (TP1 to TP2) or a more negative resting potential.2 These effects are usually observed during vagal stimulation via parasympathetic nerve or administration of acetylcholine agonists. In contrast, a sharp rise in the diastolic depolarization of the pacemaker cell (resulting in tachycardia) occurs during stimulation of the cardiac sympathetic nerves or administration of exogenous catecholamines. The SA node pacemaker may be displaced by a latent pacemaker elsewhere in the heart during myocardial ischemia because of primary suppression of the SA node or because of spontaneous discharge of a latent pacemaker at a higher intrinsic rate. When the frequency of excitation is higher in a group of latent pacemakers, the rate of firing of the other pacemakers is suppressed. This process is known as overdrive suppression.

Figure 10-12. Cardiac action potentials throughout the conductance system from the sinoatrial node (SA) through the ventricular muscle during one cardiac cycle. Note the automatic pacemaker activity (slow spontaneous depolarization) of the SA and atrioventricular nodal cells and the lack of spontaneous activity of atrial, Purkinje, and ventricular muscle cells. (Reproduced with permission from Lynch C III, Lake CL: Cardiovascular anatomy and physiology, Cardiac Vascular, and Thoracic Anesthesia. Edited by Youngberg JA, Lake CL, Roizen MF, Wilson RS. New York, Churchill Livingstone, 1999, p 87.)

The ion channels that are active in the SA node cell membrane during depolarization and repolarization are depicted in Figure 10-14.7 Two decreasing outward currents and two increasing inward currents are observed during depolarization.


Pacemaker activity is partly due to decay of the delayed rectifier current (iK, an outward current), which is permissive by allowing other currents to depolarize the pacemaker. An anomalous rectifier current (ik1, a second outward current) also permissively contributes to pacemaker activity. The first inward current is iCa. This slow inward Ca2+ current is primarily responsible for the action potential upstroke in pacemaker cells, and its continuation after initial depolarization contributes to early diastolic depolarization. The inward Na+ current (if) most likely plays an important role in the control of heart rate by the autonomic nervous system. This inward if current occurs through a channel that conducts both Na+ and Ca2+ ions. This “f” or “funny” channel mediates autonomic-dependent modulation of heart rate.33 The inward current if is activated by cAMP. Thus, β1-adrenergic stimulation accelerates, whereas vagal stimulation slows, heart rate by increasing and decreasing, respectively, the intracellular cAMP concentration and the degree of activation of the f channel. The f channel is thought to be responsible for generating spontaneous activity.

Figure 10-13. Pacemaker potentials in sinoatrial node illustrating the effect of diastolic depolarization slopes and potentials on heart rate. The action potential begins when the depolarization potential reaches the threshold potential (TP). A slowing of the rate of depolarization from a to b increases the time required to reach the TP, whereas an increase of the TP level (bc) or a greater resting potential (d) slows the heart rate. (Reproduced with permission from Hoffman BF, Cranefield PF: Electrophysiology of the Heart. New York, McGraw-Hill. 1960, Fig. 4.5, p 57.)

Figure 10-14. Changes in four ionic currents responsible for action potential depolarization and repolarization in a sinoatrial nodal pacemaker cell. Two are increasing inward currents (ii and iCa) and two are decreasing outward currents (iK, delayed rectifier and iK1, inward rectifier). (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 14-14, p 417.)


Figure 10-15. Membrane potential and current in a voltage-clamped ventricular cell. Note that the rapid and transient influx of Na+ ions induces the rapid depolarization (phase 0); this is followed by a longer inward Ca2+ current that prolongs the plateau potential (phase 1) and then a slow outward K+ current the leads to repolarization (phase 3). Resting potential is phase 4. (Reproduced with permission from Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006, Fig. 14-12, p 415.)

Alterations in ion currents in the ventricular myocyte are illustrated in Figure 10-15. The resting membrane potential of the myocyte is -90 mV; this potential controls its Na+ channel. Above this threshold, activation of the Na+ channel produces a sharp increase in inward Na+ current that is primarily responsible for myocyte depolarization (phase O). These Na+channels are rapidly inactivated by depolarization, but their ability to reopen (reactivation) is delayed even after the myocyte is fully repolarized. The inability of Na+ channels to respond to a second stimulus after depolarization occurs as a result of the prolonged plateau of the action potential that prevents membrane potential from returning to the resting levels at which Na+ channels may be reactivated. The rapid depolarization of the myocyte is followed by a brief, rapid repolarization of small magnitude (phase 1) caused by a reduction in Na+ permeability, a transient outward K+ current, and an outward Cl- current. A distinctive feature of ventricular myocardium depolarization is the plateau (phase 2) of the action potential that signifies prolonged stabilization of the myocyte near zero potential (duration ≥100 ms). An inward Ca2+ depolarizing current through Ca2+ conductance channels appears at the beginning of the plateau. This slow inward Ca2+ current is associated with opening of slow Ca2+ channels that are activated at a membrane potential of -50 mV. The Ca2+ current activates and inactivates much more slowly than the Na+ current, thereby providing an inward current that maintains the sarcolemmal membrane in a depolarized state during the plateau phase. Phase 3 of the cardiac action potential corresponds to the T wave of the ECG. Outward rectification and repolarization occur when the membrane passes current most readily in the outward direction. The most important outward rectifying current is carried by K+. Outward rectifying currents cause repolarization because membrane potential in the depolarized cell returns to its resting negative level. The outward ik rectifying current occurs at the end of the plateau phase 3 and is known as the delayed rectifier. The ion channels in ventricular myocardium are energy-dependent and regulated by the activity of the autonomic nervous system.7,34

Neural Innervation of the Heart and Blood Vessels

Baroreflex Regulation of Blood Pressure

The heart is innervated by the parasympathetic and the sympathetic nervous systems. Parasympathetic innervation arises in the motor nucleus of the vagus and the nucleus ambiguous in the medulla.35 As observed with other parasympathetic nerves, long preganglionic fibers synapse with short postganglionic fibers within the heart. The postganglionic fibers innervate pacemaker cells and conducting pathways. When activated, these fibers produce slowing of pacemaker cells and reduce conduction velocity. Aside from their effects on heart rate, excitability, and conduction, parasympathetic fibers do not substantially influence contractility.36 Acetylcholine is the neurotransmitter responsible for parasympathetic nervous system activation through nicotinic and muscarinic receptors at the pre- and postganglionic synapses, respectively, located on the postganglionic neuron and in SA and AV node pacemaker cells and the condition system. The sympathetic innervation of the heart arises from cells in the rostral ventrolateral medulla with descending nerve fibers emerging from the intermediolateral cell column of the spinal cord at C5-T6.37,38 These preganglionic fibers synapse primarily with postganglionic sympathetic nerves within the stellate ganglion, but may also synapse within a few ganglia located closer to the heart. Ganglionic transmission is mediated by preganglionic release of acetylcholine and binding of acetylcholine with nicotinic postganglionic receptors. Activation of postganglionic sympathetic nerves innervating the heart results in release of norepinephrine. This endogenous catecholamine then stimulates β1-adrenoceptors in pacemaker and conduction cells as well as atrial and ventricular myocardium. Activation of cardiac sympathetic fibers produces positive chronotropic, dromotropic, inotropic, and lusitropic effects (i.e., increases in heart rate, conduction velocity, myocardial contractility, and the rate of myofibrillar relaxation).

The afferent innervation of the heart consists of mechanoreceptors with primarily vagal afferent pathways and receptors with spinal afferent pathways. The mechanoreceptors with vagal afferents are located in ventricular, and to a lesser extent, atrial myocytes.36 Activation of the ventricular receptors by nociception or stretch, such as may occur in response to a sudden increase in ventricular volume, causes a vagal depressor response with a decrease in heart rate and mean arterial pressure (the Bezold-Jarisch reflex).39 The reduction in heart rate is mediated by an increase in cardiac vagal efferent activity, in which the decrease in mean arterial pressure results from withdrawal of sympathetic tone in arterial resistance and venous capacitance vessels. The spinal afferents traverse the sympathetic nerves and serve as nociceptors and stretch receptors. Activation of these receptors produces a transient increase in heart rate and mean arterial pressure. Spinal


afferent-mediated nociceptors may be stimulated by events such as acute myocardial ischemia. Both the cardiac vagal and spinal afferent fibers project centrally to the nucleus tractus solitarius, similar to aortic and carotid baroreceptors and chemoreceptors.

The majority of the peripheral vascular system derives its sympathetic innervation from the thoracolumbar section of the spinal cord. In contrast, the sympathetic innervation of the coronary vasculature, lung, and cerebral circulation is derived from the superior cervical and stellate ganglia. α-Adrenoceptors mediate most sympathetic nerve vascular responses, but β-adrenoceptors uniquely modulate sympathetic innervation of the adrenal gland, thereby causing the release of epinephrine and norepinephrine. Sympathetic innervation of small arterioles and metarterioles produces vasoconstriction, thereby increasing systemic vascular resistance and mean arterial pressure. Sympathetic innervation of small veins and venules causes constriction of these vessels. This action reduces the volume of blood stored in capacitance vessels, transiently increases preload, and subsequently decreases blood flow in the splanchnic circulation and, to a lesser extent, the lower extremities. Thus, activation of peripheral sympathetic nerves increases mean arterial pressure by arterial vasoconstriction combined with an increase in preload due to a reduction in venous capacitance while simultaneously increasing heart rate and myocardial contractility. These effects are critical compensatory responses to hypovolemia resulting from acute blood loss. α- and β-adrenoceptor antagonists may attenuate sympathetically mediated cardiovascular effects.

Short-duration regulation of mean arterial pressure occurs through the arterial, and to a lesser extent, intracardiac baroreceptors. Arterial baroreceptors are located at the bifurcation of the common carotid arteries and in the aortic arch. These receptors, particularly those in the carotid arteries, display tonic activity under normal conditions. An acute rise in arterial pressure activates baroreceptors through stretch-sensitive Na+ channels. Receptor activation increases afferent nerve traffic in the carotid sinus nerve, which is centrally transmitted by a unique branch of the glossopharyngeal nerve that first synapses in the nucleus tractus solitarius. The postsynaptic neurons activate the vagal motor nucleus and nucleus ambiguous, thereby causing a reduction in heart rate.38,40,41 The postsynaptic baroreceptor neurons also synapse with γ-aminobutyric acid–mediated inhibitory neurons in the caudal ventrolateral medulla that innervate medullary sympathetic neurons and produce a decrease in sympathetic nervous system activity via the rostral ventrolateral medulla.38,41,42 The resultant effect is a decrease in cardiac output and systemic vascular resistance concomitant with an increase in vascular capacitance.

The aortic baroreceptors and cardiac vagal receptors produce similar hemodynamic effects. Cardiac receptors have been theorized to be responsible for radiocontrast-induced bradycardia and hypotension during coronary angiography. Low-pressure baroreceptors located in the vena cavae, right atrium, RV, and pulmonary vein-left atrial junction respond to decreases in right atrial filling pressure by activating sympathetic tone in the arterial vasculature. Interestingly, baroreceptor activation does not influence all peripheral vascular beds. For example, the cutaneous circulation does not appear to respond to baroreceptor stimulation or inhibition. Instead, the cutaneous circulation is primarily affected by peripheral and central thermoregulatory mechanisms that produce vasoconstriction or vasodilation in a cold or warm environment to prevent or facilitate heat loss, respectively. Thermoreceptor-mediated central nervous system responses originate in the supraoptic region of the hypothalamus.

Other Cardiovascular Reflexes

Other reflexogenic areas within the cardiovascular system regulate hemodynamics through arterial chemoreceptors and the central nervous system response to ischemia. High-pressure sensitive receptors in the LV and low-pressure responsive elements in the atria and RV consist of stretch-induced mechanoreceptors that respond to pressure or volume changes. Three sets of receptors have been identified. First, discrete receptors in the endocardium are located at the junctions of the vena cavae with the right atrium and the pulmonary veins with the left atrium. These receptors activate myelinated vagal afferent fibers that project to the nucleus tractus solitarius and increase sympathetic nerve activity to the SA node but not to the ventricles, thereby increasing heart rate but not contractility. Distention of these mechanoreceptors also increases renal excretion of free water by inhibition of antidiuretic hormone secretion from the posterior lobe of the pituitary gland.43 It appears highly likely that the Bainbridge reflex may be mediated by distention of these mechanoreceptors.4 Second, a diffuse receptor network is distributed throughout the cardiac chambers that projects via unmyelinated vagal afferent neurons to the nucleus tractus solitarius. These receptors behave like the carotid and aortic mechanoreceptors and produce a vasodepressor response consisting of vagus activation concomitant with inhibition of sympathetic innervation of the heart and peripheral circulation. These actions cause reductions in heart rate, inotropic state, and systemic vascular resistance concomitant with a simultaneous increase in venous capacitance. This intracardiac receptor network plays a relatively minor role in the normal physiological control of the cardiovascular system compared with the arterial baroreceptors. Lastly, sympathetic afferent fibers are activated by receptors that respond rhythmically during the cardiac cycle. Some of these neurons convey visceral pain sensations and may be activated during myocardial ischemia. Stimulation of these fibers produces a transient increase in heart rate and mean arterial pressure by activating central nervous system sympathetic efferent fibers innervating the heart and peripheral circulation.4,44,45

The arterial baroreceptors located in the carotid sinus and aortic arch play the major role in cardiovascular homeostasis. Arterial baroreceptor reflex-induced regulation of heart rate is inhibited by volatile and many intravenous anesthetics.32,46 This inhibition of high-pressure baroreceptor reflexes by anesthetics involves several discrete sites including sympathetic ganglionic transmission, end-organ responses, and central nervous system pathways, and appears to be especially important in short-term regulation of arterial pressure.5,47 These reflexes demonstrate accommodation or adaptation to the level of arterial blood pressure and may be reset in patients with hypertension. Cardiopulmonary reflexes also appear to be inhibited by potent inhaled anesthetics and have a crucial role in short-term regulation of arterial pressure, primarily by modulating arterial baroreceptor reflex activity.32 The peripheral chemoreceptors located in the carotid and aortic bodies are sensitive to increases in arterial CO2 tension and decreases in pH. The carotid body receptors project centrally through Herring's nerve, which travels with the glossopharyngeal nerve to the nucleus tractus solitarius. In contrast, the aortic body receptors have vagal afferent fibers that also project to the nucleus tractus solitarius. The carotid body reflex appears to be more important than its aortic counterpart in the regulation of respiration in humans. Activation of the carotid and aortic chemoreceptors produces an increase in respiratory drive manifested by an increase in respiratory rate, tidal volume, and minute ventilation. These chemoreceptors may also cause activation of sympathetic nervous system fibers in the heart and peripheral circulation,


thereby increasing heart rate and mean arterial pressure. The peripheral chemoreceptor reflex is an important protective mechanism in response to pathophysiological conditions including high-altitude hypoxia, chronic lung disease, and profound hypovolemia.

In addition to the baroreceptor responses, severe hypotension also causes arterial vaso- and venoconstriction in response to brainstem hypoxia. This central nervous system ischemic response may be activated when mean arterial pressure is reduced below 50 mm Hg. An analogous mechanism may also mediate the Cushing reflex. This sympathetically mediated hypertension occurs in response to an acute elevation of intracranial pressure and a consequent reduction in cerebral perfusion pressure. Under these circumstances, arterial pressure rises progressively in an effort to exceed elevated intracranial pressure and maintain cerebral perfusion and oxygen delivery. The Cushing reflex may also be activated by brainstem compression, acute traumatic brain injury, or intracranial hemorrhage resulting from aneurysm rupture.

In animals, the diving reflex redistributes blood flow and oxygen delivery to the heart and brain as a survival defense of the submerged vertebrate against asphyxia. This reflex enables whales to remain submerged for as long as 2 hours. The residual counterpart of the diving reflex in humans may be activated by immersion of the face in cold water, which produces a comparable, albeit less intense, diving response characterized by rapid reduction in heart rate and cutaneous and skeletal muscle blood flow concomitant with an increase in arterial pressure. Stimulation of receptors in the face or upper airway initiates the diving reflex and causes apnea by inhibiting the medullary respiratory center. The hyperventilation stimuli of hypoxemia and hypercapnia are also suppressed. The cardiovascular limbs of the chemoreceptor reflex are partially retained, resulting in generalized systemic vasoconstriction, except in the coronary and cerebral circulations. The diving reflex response is distinctly different from the cold-pressor reflex. This latter reflex is activated by complete immersion of one hand in ice water. The cold-pressor reflex increases heart rate and mean arterial pressure by stimulating both pain and cold receptors. A cold environment directly causes vasoconstriction to prevent heat loss and also stimulates reflex central nervous system thermoregulatory receptors in the hypothalamic preoptic region. This latter effect produces sympathetically mediated vasoconstriction. A warm, ambient environment or an increase in metabolically induced heat production produces an opposite response to dissipate accumulated heat.

Figure 10-16. Anterior view (left) shows right coronary and left anterior descending arteries. Posterior view (right) shows left circumflex and posterior descending arteries. Note that the right coronary or left circumflex artery may form the latter artery. The anterior cardiac veins from the right ventricle and the coronary sinus, which drain primarily the left ventricle, empty into the right atrium. (Reproduced with permission from Smith JJ, Kampine JP: Circulatory Physiology—The Essentials, 3rd edition. Baltimore, Williams & Wilkins, 1990, Fig. 3-1, p 32.)

Somatic pain increases heart rate and mean arterial pressure by activation of sympathetic efferent nerves. In contrast, visceral pain or distention of a hollow viscus (e.g., small intestine, bladder) may produce reflex vagal bradycardia and hypotension. The oculocardiac reflex is activated by pressure on the ocular globe and causes pronounced bradycardia and hypotension by activation of vagal nerve fibers innervating the SA node. The Valsalva maneuver consists of forced expiration against a closed glottis. This maneuver reduces venous return to the right heart, decreases cardiac output and mean arterial pressure, and increases heart rate. The reflex tachycardia occurs because of reduced activity of arterial baroreceptors and LV mechanoreceptors. Release of the forced expiration by glottic opening acutely increases venous return, cardiac output, and mean arterial pressure while simultaneously causing reflex bradycardia mediated by vagal innervation of the SA node triggered by the arterial baroreceptors.

Coronary Circulation

Anatomy of the Coronary Arterial and Venous Systems

The heart is the only organ that furnishes its own blood supply. The left main and right coronary arteries arise from the aorta behind the left and right aortic valve leaflets (Fig. 10-16). The coronary ostia remain patent throughout systole because eddy currents prevent the valve leaflets from contacting the aortic walls. The left main coronary artery divides almost immediately into the left anterior descending (LAD) artery and left circumflex coronary artery (LCCA). The LAD further divides into several branches along the anterior interventricular groove toward the apex of the heart where they supply the anterior wall of the LV and the anterior two thirds of the


interventricular septum (Fig. 10-16). The LCCA marks a pathway along the base of the LV within the coronary sulcus and terminates in the left posterior descending branch. The LCCA supplies the LV lateral wall and part of the LV posterior wall. The right coronary artery (RCA) courses along the AV groove toward the right chambers of the heart and frequently extends along the posterior interventricular sulcus to give rise to the right posterior descending branch (Fig. 10-16). The RCA supplies the anterior and posterior walls of the RV except for the apex (supplied by the LAD), the right atrium including the SA node, the upper half of the atrial septum, the posterior third of the interventricular septum, the inferior wall of the LV, the AV node, and the posterior base of the LV. A branch of the LCCA occasionally supplies the SA node. Because either the RCA or the LCCA may supply the posterior descending coronary artery, the coronary circulation is described as right or left dominant, respectively, based on the source of this vessel's blood supply.

The proximal branches of the RCA, LCCA, and LAD are located on the epicardial surface of the heart and give rise to multiple intramural vessels that penetrate perpendicularly or obliquely deep into the ventricular walls. Except for the thin tissue layer on the endocardial surface, the nutritive blood supply is almost entirely derived from these major coronary arteries. The penetrating branches divide into dense capillary networks located roughly along the courses of the myocardial bundles. Arterial branches with diameters between 50 and 500 µm form interconnecting anastomoses throughout the endocardium of the ventricular walls (Fig. 10-17, A and B). Another network of subendocardial vessels between 100 and 200 µm in diameter forms a plexus of deep anastomoses. A coronary collateral circulation may also arise from different branches of the same coronary artery or from branches of two different coronary arteries. Flow through such coronary collaterals is usually negligible because the driving pressure at the two ends of the anastomoses is nearly equal. However, if the artery supplying one branch of this collateral circulation becomes severely stenotic or occluded, the large pressure reduction will divert blood flow through the patent artery and into the distribution of the occluded artery through these collateral vessels. Thus, the coronary collateral circulation may be especially important to patients with coronary artery disease.48,49,50

Most of the coronary venous system remains unnamed with the exception of the great cardiac vein (that runs along the AV groove and the LAD), the anterior cardiac vein (located with the RCA), and the middle cardiac vein (associated with the posterior descending branch of the RCA; Fig. 10-16). Thus, the main coronary venous drainage tends to retrace the course of the major coronary arteries along the AV and interventricular grooves. In general, there are two coronary veins located along either side of each coronary arterial branch. The coronary veins converge and terminate in the coronary sinus, which empties into the posterior aspect of the right atrium. Approximately 85% of the total coronary blood flow to the LV drains into the coronary sinus. The remaining blood flow empties directly into the atrial and ventricular cavities by the thebesian veins. The RV veins drain into the anterior cardiac veins; these empty individually into the right atrium just above the tricuspid valve.

Coronary Microcirculation

The coronary capillary network has an organizational structure that is similar to that observed in other tissue beds. Myocardium has a very high density of capillary blood vessels to myofibrils, approximately 1:1 (Fig. 10-18); this is because of the exceptionally high metabolic demand of the heart. On average, adjacent capillaries are separated by the diameter of approximately one myocyte. The distribution of capillaries is quite uniform and ranges between 3,000 and 4,000/mm2 of tissue. Interestingly, capillary density is reduced in the interventricular septum and AV nodal tissue, and this observation may explain why the specialized conducting system is more vulnerable to ischemia than the myocardium itself. As in other capillary beds, coronary capillaries are the sites for exchange of O2, CO2, and for the movement of larger molecules across the endothelial cell lining, where it is devoid of vascular smooth muscle.

Figure 10-17. A. Diagram of the minute arterial-to-arterial and venous-to-venous anastomoses of the coronary arterial system, which allows diversion of flow if one distribution becomes blocked. (Reproduced with permission from Guyton AC, Hall JE: Human Physiology and Mechanisms of Disease, 6th edition. Philadelphia, WB Saunders, 1997, Fig. 18-4, pp 185.) B. Diagram of the epicardial coronary vessels lying on the cardiac muscle surface, the penetrating deep vessels, and the subendocardial arterial plexus connecting the deep vessels. (Reproduced with permission from Guyton AC: Textbook of Medical Physiology, 6th edition. Philadelphia, WB Saunders, 1997, Fig. 25-3, p 299.)

Mechanics of Coronary Blood Flow

Blood supply to the LV is directly dependent on the difference between the aortic pressure and LV end-diastolic pressure (coronary perfusion pressure) and inversely related to the vascular resistance to flow, which varies to the fourth power of the radius of the vessel (Poiseuille's law). Two other determinants of coronary flow are vessel length and viscosity of the


blood, but these factors are generally constant. Resting coronary blood flow in the adult is approximately 250 mL/min (1 mL/g), representing approximately 5% of cardiac output. The changes in aortic pressure and the impedance to flow due to physical compression of the intramural coronary arteries during the contraction-relaxation cycle (Fig. 10-4) govern the pulsatile pattern of coronary flow in the LV. Aortic pressure is slightly less than LV pressure during systole. As a result, blood flow in the LV subendocardium occurs only during diastole (Fig. 10-19). Overall coronary flow does not cease completely during the early part of systole because of this extravascular compression, but most of the flow occurs during diastole when impedance to flow is minimal and aortic pressure remains sufficient to maintain adequate coronary perfusion pressure.

Figure 10-18. Diagram of an electron micrograph of cardiac muscle showing large numbers of mitochondria and the intercalated disks with nexi (gap junction), transverse tubules, and longitudinal tubules surrounding capillary endothelium. (Reproduced with permission from Berne RM, Levy MN: Chapter 3: Cardiovascular Physiology, 8th edition. St. Louis, CV Mosby, 2000, Fig. 3-1, p 56.)

During systole, LV subendocardium is exposed to a higher pressure than the subepicardial layer. Indeed, the systolic intraventricular pressure may be higher than the peak LV systolic pressure. Because of these differences in tissue pressure, the subendocardial layer is more susceptible to ischemia in the presence of coronary artery disease, pressure-overload hypertrophy, or pronounced tachycardia concomitant with compromised regional myocardial perfusion flow, a greater intraventricular-aortic pressure gradient, or reduced total diastolic flow, respectively. Coronary blood flow is also compromised when aortic diastolic pressure is reduced (e.g., severe aortic insufficiency), and this observation may also adversely affect perfusion, particularly in the presence of a critical coronary stenosis.51 Elevated LV end-diastolic pressure, as observed during acute heart failure, also reduces coronary blood flow because of decreased coronary perfusion pressure. In contrast to left coronary blood flow, RCA flow is continuous throughout the cardiac cycle because the lower pressure in the RV compared with the LV causes substantially less extravascular compression (Fig. 10-19). Coronary sinus (venous) blood flow is maximal during late systole because of the extravascular compression and the low right atrial pressure.

Figure 10-19. Schematic representation of blood flow in the left and right coronary arteries during phases of the cardiac cycle. Note that most left coronary flow occurs during diastole while right coronary flow (and coronary sinus flow) occurs mostly during late systole and early diastole. (Reproduced with permission from Berne RM, Levy MN: Chapter 10: Cardiovascular Physiology, 8th edition. St. Louis, CV Mosby, 2000, Fig. 10-3, p 231.)


Regulation of Coronary Blood Flow

The two major determinants of coronary blood flow (perfusion pressure and vascular resistance) vary substantially during the cardiac cycle (Fig. 10-4). Coronary perfusion pressure certainly varies with changes in aortic, intramyocardial, and coronary venous pressures during systole and diastole, but the major factor that regulates coronary blood flow is the variable resistance produced by coronary vascular smooth muscle. Sympathetic nervous system innervation modulates the contractile state of coronary vascular smooth muscle. In addition, smooth muscle tone is affected by stretch of the muscle (termed the myogenic factor). However, metabolic factors are the major physiological determinants of coronary vascular tone and, hence, myocardial perfusion. The ratio of epicardial to endocardial blood flow ratio remains near 1.0 throughout the cardiac cycle despite systolic compressive forces exerted on the subendocardium. The more pronounced resistance to flow in the subendocardium is offset by β-adrenoceptor–mediated vasodilation and by local metabolic autocrine factors (e.g., adenosine during hypoxia) produced by the myocardium itself. The relative maintenance of subendocardial blood flow may also be related to the extensive number of redundant arteriolar and capillary anastomoses in the subendocardium.1

Oxygen Delivery and Demand

The heart normally extracts between 75 and 80% of arterial O2 content, by far the greatest O2 extraction of all organs. The majority of O2 demand is derived from the development of LV pressure during isovolumic contraction. Oxygen consumption is also affected by the rate of LV pressure development (dP/dt) and the diameter of the LV (Laplace's law). An increase in myocardial contractility enhances O2 consumption, but heart rate is the primarily determinant of O2 consumption. Cardiac O2 extraction is near maximal under resting conditions and cannot be substantially increased during exercise. Thus, the primary mechanism by which myocardium meets its O2 demand is through enhanced O2 delivery, which is proportional to coronary blood flow at constant hemoglobin concentration. Coronary blood flow and O2 consumption increase four- to fivefold during strenuous physical exercise. The difference between maximal and resting coronary blood flow is known as coronary reserve. Myocardial O2 consumption is a major determinant of coronary blood flow. For example, coronary vascular resistance is greater in the rested, perfused heart than in the contracting heart, indicating that coronary blood flow increases in response to a higher rate of O2consumption. The mechanism(s) responsible for the correlation between myocardial work, O2 consumption, and coronary vessel dilatation has yet to be precisely determined. In addition to metabolically induced vasodilation, the factors responsible for coronary autoregulation (maintenance of coronary blood flow with a change in perfusion pressure) and reactive hyperemia (the several-fold increase in blood flow above baseline after a brief period of ischemia) are also not well understood.

Despite decades of intense research into the mediators of local metabolic coronary vasodilation, surprisingly little is known about the details of this phenomenon. To date, it has been established that metabolic coronary vasodilation is at least partly the result of activation of the sympathetic nerves to the heart and coronary vasculature during an increase in heart rate and myocardial contractility. Sympathetic nerve activation produces a feed-forward β-adrenoceptor–induced vasodilation, primarily of small coronary arterioles. This feed-forward mechanism operates without an error signal, indicating that there is a direct and apparently unregulated relationship between heart rate and inotropic state and the activation of β-adrenoceptor-mediated vasodilation.49,50,52 There also appears to be a feed-forward, sympathetically mediated, α-adrenoceptor–induced vasoconstriction in larger coronary arteries during exercise. This vasoconstriction occurs upstream from coronary small coronary arterioles and serves two important functions: reduction of vascular compliance and attenuation of systolic minus diastolic flow oscillations during the cardiac cycle. These actions assist in the preservation of blood flow to the more vulnerable LV endocardium when heart rate, contractility, and O2 consumption are elevated. Interestingly, cardiac parasympathetic nerves have a prominent role in regulating heart rate, but these nerves appear to have a negligible direct effect on the regulation of coronary blood flow.

The conclusions about sympathetic nervous system control of the coronary circulation are based on alterations in the slope of the O2 consumption-coronary venous O2 tension relation during graded exercise in the presence of exogenous α- or β-adrenoceptor blockade (Fig. 10-20). The current evidence implicating the β-adrenoceptor in coronary vasodilation accounts for only about one fourth of the total coronary vasodilation observed during exercise-induced hyperemia.53 These data suggest that the other three fourths of coronary vasodilation during exercise may be produced by as yet undefined local metabolic factors that act on coronary vascular smooth muscle with or without the influence of endothelium. Many metabolic factors have been proposed to individually or collectively modulate coronary flow at the arterial or capillary level, including adenosine, nitric oxide, arterial oxygen or CO2 tension, pH, osmolarity, K+, Ca2+, and prostaglandins. Many of these factors exert predictable direct effects. For example, hypoxia or ischemia decreases arterial oxygen tension and pH and increases CO2 tension, adenosine, K+, and Ca2+ concentrations, and serum osmolarity. Many of these changes may indeed increase coronary blood flow, but none appear to be crucial determinants of vasodilation during exercise. For example, adenosine receptor blockade does not alter coronary blood flow under resting conditions or during exercise. Similarly, inhibition of nitric oxide production or ATP-sensitive K+ (KATP) channels also does not alter the O2 consumption-coronary venous O2 slope during graded exercise. Nevertheless, nitric oxide and KATP channels have been shown to regulate the balance between O2 supply and demand under resting conditions.

There is very strong evidence however, that adenosine released during hypoxia or ischemia causes coronary vasodilation and that this effect is mediated by activation of KATPchannels. Adenosine and KATP channels have also been implicated during reactive hyperemia after ischemia, but these mediators do not appear to be required for coronary autoregulation. Moreover, the KATP channel probably maintains a lower vascular smooth muscle tone and thus, a higher basal coronary flow during resting conditions. While not acting as a local metabolic vasodilator, nitric oxide may react to increased downstream arterial dilation by dilating larger, upstream epicardial coronary arteries to prevent excessive sheer stress on coronary endothelial cells.

Myocardial Ischemia and Infarction

Ischemic heart disease remains the leading cause of death in the United States. Global ischemia results from insufficient total coronary blood flow for the overall metabolic needs of the heart.54,55 Regional ischemia results from insufficient coronary blood flow to a region of the heart supplied by its vascular limb. A large, acute coronary artery occlusion produces acute myocardial ischemia and often contributes to the


development of a malignant ventricular arrhythmia because blood flow through coronary collaterals fails to provide sufficient perfusion to the ischemic zone.49,51 Thus, many patients with acute coronary syndrome succumb to sudden cardiac death before or during the evolution of a myocardial infarction. If the coronary artery occlusion develops more slowly, collateral formation in the watershed region may reduce the degree of myocardial damage associated with acute coronary occlusion. New collateral development (known asvasculogenesis) into the occluded vascular bed will result in the independence of this region from its original blood supply.

Figure 10-20. Coronary venous oxygen (O2) tension at rest and during three levels of exercise plotted as a function of myocardial O2 consumption with individual regression lines for a blockade alone and with β blockade. The steep slope of combined α + β blockade indicates a modest match by local metabolic factors in the absence of adrenergic mechanisms. The differences in slopes between α + β blockade and α blockade demonstrates β-adrenergic–mediated coronary vasodilation, whereas the difference in slopes between β blockade and control demonstrates α–mediated coronary vasoconstriction. Note that β-adrenergic vasodilation accounts for only about 25% of the increase in coronary flow during exercise. (Reproduced with permission from Gorman MW, Tune JD, Richmond MW, Feigl EO: Feedforward sympathetic coronary vasodilation in exercising dogs. J Appl Physiol 2000; 89: 1892.)

An atherosclerotic plaque is the most frequent cause of obstructed blood flow in large, epicardial coronary artery vessels.56 The most common site for development of an atherosclerotic plaque is the first several centimeters of the major and coronary arteries and their primary branches. The position of atherosclerotic plaques facilitates their palliation by coronary artery bypass graft surgery.57 Atherosclerotic plaques typically develop very slowly, eventually protruding into the vessel and partially or completely blocking flow. The atherosclerotic plaque may also precipitate thrombus formation, which more rapidly occludes the coronary artery. A thrombus usually develops when the plaque has broken through the vascular intima, thereby exposing vascular smooth muscle or adventitia to clotting factors and platelets contained in blood. When fibrin and platelets begin to be deposited, blood cells become entrapped and form a thrombus that grows rapidly until it produces a critical stenosis or complete occlusion of the coronary artery. The thrombus may also embolize by detaching from its original site of formation and flow to a more peripheral branch of the coronary arterial bed. Atherosclerotic plaques are composed of cholesterol and other lipids that become deposited beneath the intima and fibrous tissue, which also frequently becomes calcified. These calcium deposits are located predominantly at the junction of the intimal and medial layers of the blood vessel.

An acute occlusion of a major epicardial coronary artery causes almost immediate, maximal dilation of existing small collateral vessels supplying blood flow to the ischemic zone. Unfortunately, blood flow through these minute collaterals is generally insufficient to nourish all of the myocardium that they supply. Collateral perfusion through these anastomoses temporally increases and may double within 24 hours after acute coronary occlusion. Eventually, the affected myocardium will be supplied by a normal quantity of blood flow, albeit from a different source. During the gradual development of an atherosclerotic plaque, collateral vessels may develop at a rate similar to, and thereby compensate for, the slow occlusion of the vessel lumen. This redistribution of myocardial blood flow from a partially occluded to a collateral vascular supply may prevent an acute episode of ischemia when the original coronary artery becomes occluded. Only when the atherosclerotic process develops more rapidly than the formation of an adequate collateral blood supply will the O2demand exceed delivery and produce myocardial dysfunction. This type of ischemic cardiomyopathy is the most common cause of heart failure. Myocardial necrosis and apoptosis (programmed cell death) occur as a consequence of ischemia and infarction. However, cellular demise usually will not occur in a region unless coronary blood flow falls below 65% of resting values. Myocytes in this region may be viable, but their contractile ability may be severely impaired because of the lack of O2 and nutrients. Critically stenotic atherosclerotic plaques may produce a pressure gradient across the stenosis and


substantially reduce the perfusion pressure in distal branches of the affected vessel. Such gradients are especially important when stenoses occur in coronary arteries of smaller caliber. There may be compensatory vasodilation of the coronary distal bed, but progressive diminution of distal blood flow may occur despite this response.

Myocardial infarction may also occur without evidence of major coronary thromboses, emboli, or stenosis. This form of infarction is caused by excessive metabolic demands resulting from severe LV hypertrophy (e.g., critical aortic stenosis) or vasoactive drug ingestion (e.g., amphetamines, cocaine) or may also result from coronary artery vasospasm. Either of these mechanisms may lead to ischemia by adversely affecting myocardial O2 supply-demand relations. Clearly, the presence of coronary stenoses that would otherwise be asymptomatic (<70%) may exacerbate O2 demand-mediated myocardial ischemia. Such causes for myocardial infarction are relatively uncommon, and coronary artery disease remains the primary cause of transmural necrosis. Subendocardial infarction may have a different etiology than the transmural infarction caused by an acute coronary occlusion. Subendocardial infarction may occur when coronary perfusion pressure is adversely reduced by decline in diastolic aortic pressure or increases in LV end-diastolic pressure. Thus, patients with severe aortic insufficiency or end-stage heart failure may be especially prone to subendocardial injury.

Along with the severity of coronary artery stenosis, the metabolic activity of the heart during ischemia is a critical factor in determining the extent of cell death. If an area of the heart has reduced blood supply due to ischemia, the region distal to this coronary stenosis is maximally vasodilated, and an increase in O2 demand causes vasodilation of adjacent coronary vessels that supply surrounding normal myocardium. This metabolically induced vasodilation may inadvertently redistribute blood flow away from the ischemic zone through coronary collateral vessels. This phenomenon is known as coronary steal.58,59 Despite original arguments to the contrary, most experimental and clinical evidence collected to date indicates that volatile anesthetics do not cause coronary steal unless profound hypotension (<50 mm Hg) is present. Volatile anesthetics are not potent vasodilators, unlike drugs such as adenosine and sodium nitroprusside that are known to produce coronary steal.

A potentially lethal complication of acute coronary occlusion is the development of malignant ventricular arrhythmias (e.g., ventricular tachycardia, fibrillation). Ventricular arrhythmias are most likely to occur during the first 10 minutes after an acute coronary occlusion, especially if the coronary blood flow to the conduction system becomes ischemic. Myocytes distal to occluded coronary artery may become electrically dysfunctional and fail to temporally repolarize with surrounding normal myocardium. This repolarization dyssynchrony is a frequent cause of arrhythmogenesis during acute myocardial ischemia. Compensatory activation of the sympathetic nervous system in response to marked reductions in cardiac output may also contribute to the development of ventricular arrhythmias. Left ventricular dilatation or formation of an LV aneurysm late after myocardial infarction may also provide a substrate for arrhythmogenesis by increasing the duration of impulse conduction and creating abnormal conduction pathways around the infarcted zone. These consequences of infarction may predispose to circuitous electrical activity and result in an impulse re-entering a section of the myocardium that is still recovering from its refractory period, thereby initiating an abnormal subsequent cycle of excitation and reentry.

A central zone of myocardial necrosis develops within 1 hour after acute coronary artery occlusion, which is eventually replaced by scar as the infarcted myocardium heals. A border zone characterized by profoundly reduced contractility due to inadequate coronary collateral perfusion surrounds this central necrotic zone. Some of this border zone also develops scar tissue; other regions surrounding the central necrotic region hypertrophy as a compensatory response to increased workload. This postinfarction ventricular hypertrophy serves to maintain cardiac output, but may also contribute to the late development of heart failure as a result of progressive diastolic dysfunction.1,50,52,60

Pulmonary Circulation

Comparison with the Systemic Circulation

The pulmonary circulation receives the blood pumped by the RV. Total pulmonary blood flow is equivalent to cardiac output. There are major differences in hemodynamics between the systemic and pulmonary circulations (Fig. 10-21).61 There is a greater decrease in mean pressure across systemic arteries to arterioles compared with vessels of similar caliber in the pulmonary circulation. The precapillary and capillary vessels of the pulmonary vasculature are located in close proximity to the alveolar membranes, thereby facilitating gas exchange. The lung is richly innervated by the parasympathetic and sympathetic nervous system, but the dominant effect of the autonomic nervous system occurs primarily at the level of alveolar and bronchial smooth muscle. Vagal innervation of muscarinic receptors in airway smooth muscle produces bronchoconstriction and is an important contributing factor to bronchospasm in atopic pulmonary disease, pneumonia, and inhalation of noxious substances. The sympathetic innervation of the lung is derived from upper thoracic sympathetic fibers that innervate both airway and pulmonary vascular smooth muscle. Sympathetic stimulation of airway smooth muscle produces bronchodilation by activation of β2-adrenoceptors. The sympathetic innervation of the pulmonary vascular system provides a physiological response to gravitational effects on the intrapulmonary distribution of blood flow and partially counteracts alterations in regional ventilation/perfusion (V/Q) ratio differences produced by such gravitational forces.62

Regional Differences in Perfusion and V/Q Matching

The V/Q distribution within the lung in an upright position varies because of the effect of gravity (Fig. 10-22). In the upper lung (zone 1), V/Q ratio is >1.0, indicating that alveolar ventilation occurs in excess of pulmonary blood flow. Because


part of this zone is ventilated but not perfused, zone 1 contributes to dead space ventilation. In the middle region of the lung (zone 2), the V/Q ratio is close to 1.0, indicating a balance between ventilation and perfusion. In the lower regions of the lung (zone 3), the V/Q ratio is substantially lower than 1.0. Under these conditions, ventilation inadequately matches perfusion and intrapulmonary shunt occurs. The overall V/Q ratio of the lung is between .85 and .90. Thus, there are large gradients in ventilation and perfusion from the top to bottom of the lung in standing position, and as a result of gravitational effects, the blood volume and blood flow are substantially greater at the lung base compared with the apex. However, ventilation may be more effective in the lung base because the diaphragm exerts a greater influence in this region.

Figure 10-21. Comparison of pressure gradients (in mm Hg) along the high-pressure systemic and low-pressure pulmonary circulation. (Reproduced with permission from Nunn JF: Applied Respiratory Physiology. London, Butterworth, 1971, p 213.)

Figure 10-22. Relative ventilation and perfusion (V/Q) distribution in different areas of the lungs (upright position). The left side shows the percentage distribution of the total lung volume and the right side shows the alveolar ventilation, pulmonary blood flow, and V/Q ratio of each horizontal slice of lung volume. Note the upper zone is relatively overventilated and the lower zone is relatively overperfused. (Reproduced with permission from Nunn JF: Applied Respiratory Physiology. London, Butterworth, 1971, p 234.)

The distribution of ventilation and perfusion throughout the lung also affects the relationships between pulmonary arterial, venous, and alveolar pressures within different lung zones. In the upper zone (zone 1), pulmonary arterial pressure of compressible vessels remains less than the pulmonary alveolar pressure and is insufficient to open the vessels, which remain collapsed during some of inspiration. In the middle zone (zone 2), the pressure at the arterial end of the compressible vessels exceeds pulmonary venous and pulmonary alveolar pressure; therefore, the blood flow becomes dependent on the pressure gradient between the pulmonary artery and alveolus, both of which exceed pulmonary venous pressure. In zone 3, the pulmonary venous pressure exceeds pulmonary alveolar pressure. Thus, blood flow depends on the pressure gradient between arterial to venous ends of the capillaries, similar to the situation observed in the systemic circulation. As intravascular pressures increase, progressively lower resistance to pulmonary blood flow is observed in this zone. In the supine position, similar V/Q distributions are observed over smaller pressure gradients (compared with the upright position) between anterior and posterior thorax. From this discussion, it is clear how pathologic conditions may reduce arterial O2 tension. For example, alveolar collapse in areas of atelectasis in which perfusion persists despite compensatory hypoxic pulmonary vasoconstriction may produce profound hypoxemia as a result of intrapulmonary shunt.63,64

Hypoxic Pulmonary Vasoconstriction

Pulmonary arteriolar vasoconstriction triggered by hypoxia shunts blood flow away from poorly to well-ventilated regions of the lung, thereby improving arterial O2 saturation. The mechanism by which hypoxia raises pulmonary vascular resistance appears to be mediated by an O2 sensor because isolated pulmonary arterial smooth muscle cells contract under hypoxic conditions.65,66 The O2 sensor has yet to be identified, but may be mediated by smooth muscle mitochondria and pulmonary vascular endothelium. Pulmonary arterial strips with intact endothelium are more sensitive to hypoxia than skinned fiber preparations in vitro. Hypoxia also inhibits an outward K+ current; the resulting depolarization augments a Ca2+ influx into pulmonary vascular smooth muscle, thereby initiating contraction. The contractile mechanism of hypoxic pulmonary vasoconstriction appears to be mediated by the Ca2+-calmodulin system and causes phosphorylation of vascular smooth muscle myosin light chains. Chronic hypoxia causes proliferation of vascular smooth muscle and thickens the pulmonary arterial tree. This response increases pulmonary vascular resistance and may also produce irreversible pulmonary hypertension.

Physiologic Modulation of the Pulmonary Circulation

The blood volume stored in the pulmonary circulation is substantial (≥900 mL), and when combined with the blood volume contained within the heart and proximal great vessels, this pulmonary blood volume provides a crucial, rapidly available source of reserve intravascular volume during acute, massive hemorrhage. The mechanisms by which blood volume is shifted to central compartments in response to hypovolemia is poorly understood, but activation of regional sympathetic innervation of the volume-containing reservoirs, vasoconstriction of arterial resistance vessels, and the systemic action of epinephrine released from the adrenal gland clearly play important roles. Angiotensin, prostaglandins and other arachidonic acid metabolites, and nitric oxide are also critical regulators of pulmonary vascular resistance and the distribution of blood flow within the lung parenchyma.67,68,69Nitric oxide has proven benefits in the treatment of acquired and congenital pulmonary hypertension, often with life-saving results. For example, inhaled nitric oxide is a selective pulmonary vasodilator at doses <40 to 80 parts per million, and reductions in pulmonary arterial pressure produced by this drug are often essential to preserve RV function in patients undergoing heart or lung transplantation.

Cerebral Circulation

Anatomy and Cerebral Autoregulation

Blood flow to the brain is provided through the internal carotid and vertebral arteries. The vertebral arteries join to form the basilar artery, which, along with branches of the internal carotid arteries, forms the circle of Willis. The brain is approximately 2% of total body weight, yet this organ receives approximately 15% of cardiac output. This remarkably large cerebral blood flow (45 to 55 mL/100 g/min) reflects the high metabolic rate of the brain. Cerebral oxygen consumption averages 3.5 mL/100 g/min and accounts for 20% of total body oxygen consumption at rest. Regional cerebral blood flow and metabolic rate vary substantially throughout the brain. Cerebral blood flow and metabolic rate are closely


linked and are approximately 4 times greater in gray compared with white matter. Thus, the cerebral cortex has a substantially greater blood flow and metabolic rate than subcortical regions. Motor activity or sensory stimulation is associated with increased neuronal activity in the contralateral activated areas of brain and is closely coupled to regional increases in blood flow and metabolic rate in corresponding regions.70,71 The mechanism of coupling of the activity-metabolism-blood flow relationship is most likely related to local metabolic vasodilators (e.g., lactic acid), alterations in electrolyte (e.g., K+, Ca2+) concentrations, and other substances (e.g., adenosine, released neurotransmitters). To date, no single causative molecule has yet been identified as the primary factor linking regional metabolic rate and blood flow to neuronal activity.72,73

Regulation of Cerebral Blood Flow: Hypercarbia, Hypoxia, and Arterial Pressure

Cerebral blood flow remains relatively constant when mean arterial pressure varies between 50 and 150 mm Hg in healthy subjects (Fig. 10-23). This autoregulation of cerebral blood flow shifts to the right in patients with chronic, poorly controlled essential hypertension. For example, the autoregulation curve may range between 80 and 200 mm Hg in a patient with hypertension, and reducing the mean arterial pressure below 80 mm Hg may precipitate cerebral ischemia. This observation emphasizes that effective treatment of hypertension readjusts the autoregulation curve to its normal pressure range. Cerebral autoregulation is inhibited by hypercarbia and higher end-tidal concentrations of volatile anesthetics. In contrast, a reduction in arterial CO2 tension counteracts the direct cerebral vasodilator actions of many drugs, including those of volatile anesthesia agents.74

Arterial CO2 tension is a major regulator of cerebral blood flow within the physiologic range of arterial CO2 tensions. Cerebral blood flow linearly increases 1 to 2 ml/100 g/min for each 1 mmHg increase in Paco2. Below an arterial CO2 tension of 25 mmHg (Fig. 23), the cerebral blood flow response to Paco2 is attenuated. The mechanism responsible for the cerebral blood flow-arterial CO2 tension relationship is related to extracellular H+ concentration. Carbon dioxide rapidly diffuses across the vascular endothelium, and changes in local pH are governed by the Henderson-Hasselbach equation. Notably, alterations in cerebral produced by changes in arterial CO2 tension are not sustained blood flow because bicarbonate is eventually transported out of the brain extracellular fluid, thereby returning pH to a normal value. In contrast to the effects of respiratory acidosis on cerebral blood flow, the actions of metabolic acidosis are more gradual because the blood–brain barrier is relatively impermeable to H+. Hypoxia-induced increases in cerebral blood flow occur at arterial O2 tensions below 60 mm Hg (Fig. 10-23). The increase in cerebral blood flow at Pao2 levels below 60 mm Hg is very rapid. The mechanism of this hypoxia-induced increase in cerebral blood flow may be related to the vasodilator effect of neuronal acidosis. Several other mediators and chemoreceptor activation have also been proposed as potential signaling mechanisms responsible for cerebral vasodilation during hypoxia. In contrast to the marked increases in cerebral blood flow observed during hypoxia, little change in cerebral blood flow occurs under normoxic or hyperbaric conditions (Pao2 of 60 to 300 mm Hg).

Figure 10-23. Cerebral blood flow (CBF) is autoregulated (relatively unchanged) as mean systemic blood pressure rises between 50 to 150 mm Hg. However, flow is nearly linearly increased with a rise in Paco2 and increased if Pao2 falls below 50 mm Hg. (Modified and reproduced with permission from Michenfelder JD: Anesthesia and the brain, Clinical, Functional and Vascular Coordinates. New York, Churchill Livingstone, 1988, pp 94–113.)

Neural control of the cerebral circulation plays a relatively minor role in regulation of cerebral blood flow despite the extensive sympathetic nervous system innervation of cerebral blood vessels. Sympathetic postganglionic neurons originate in the cervical sympathetic ganglia, and vasoconstriction produced by sympathetic stimulation is largely exerted on the medium to larger-sized cerebral arteries. This response is primarily manifested during intense sympathetic nervous system activation that accompanies profound hypovolemia. The net effect of this sympathetic activation is a downward shift in the cerebral autoregulation curve, indicating a lower cerebral blood flow than predicted at a given level of mean arterial pressure. The cerebral blood vessels are also innervated by cholinergic and serotonergic fibers. Administration of exogenous vasodilators (e.g., sodium nitroprusside, adenosine, Ca2+ channel blockers, volatile anesthetics) increases cerebral blood flow. In contrast, catecholamines such as epinephrine do not substantially affect cerebral blood flow when these drugs are used to alter a systemic hemodynamics unless cerebral perfusion pressure is affected at the extremes of the autoregulation curve. It is important to recognize that autoregulation of cerebral blood flow is not effective and cerebral perfusion becomes pressure-dependent in areas of regional cerebral ischemia.

Effects of Increased Intracranial Pressure

Along with the brain, the cerebral circulation is entirely constrained within the rigid cranial cavity. This unique anatomic arrangement infers that increases in cerebral arterial blood flow must be matched by comparable increases in venous flow from the skull because the volume of blood and extracellular fluid within the brain is relatively constant. Thus, an intracranial mass (e.g., tumor, hematoma) is inevitably accompanied by an increase in intracranial pressure. Under these circumstances, the resistance to cerebral blood flow increases and cerebral perfusion is no longer determined by the difference between mean arterial pressure and cerebral venous pressure, but rather the difference between arterial pressure and intracranial pressure. If intracranial pressure continues to increase, a compensatory increase in arterial pressure occurs (Cushing reflex) that acts as a protective mechanism to maintain cerebral perfusion.

Renal Circulation

Anatomy of the Renal Circulation: Determinants of Glomerular Blood Flow

The primary branches of the renal artery divide into several interlobar arteries that traverse the parenchyma in a radial fashion from the hilum to the cortical-medullary junction that


separates the kidney into an outer cortex and an inner medulla where urine is primarily concentrated in the renal tubules. As an interlobar artery approaches the cortical-medullary junction, it branches into a series of arcuate arteries that are located over the bases of the adjacent medullary pyramids in the zone between the cortex and the medulla, but do not interconnect with adjacent interlobar arteries. This lack of collateral blood supply indicates that acute occlusion of an interlobar artery will produce a pyramid-shaped renal infarction. Interlobular branches from the arcuate arteries travel toward the capsular surface and form the afferent arterial supply to the glomeruli. The kidney has approximately 1 million glomeruli that filter plasma from circulating blood into Bowman's capsule that surround each glomerulus capillary tuft. The afferent arteriole to each glomerulus divides into several vessels that form discrete capillary loops. The proximal and distal limbs of each loop are interconnected by many smaller capillaries, thereby forming the capillary tuft. Plasma filtration occurs within these capillary networks. After exiting the capillary network, the distal ends of each capillary loop within the glomerulus rejoin to form the efferent arterioles. The diameter of efferent arterioles is usually substantially less than the afferent arteriole. The entire glomerular capillary tuft is enveloped by Bowman's capsule, which collects the glomerular filtrate and transports it to the renal tubules where urine is concentrated. The efferent arterioles subsequently divide into another capillary network, the peritubular capillaries, some of which surround relatively short renal tubules located almost entirely in the renal cortex. Most of the peritubular capillaries form the long hairpin loops of Henle extending deep into the renal medulla. These vasa recta capillaries are important components of the renal countercurrent exchange mechanism that is responsible for urine concentration.75

Renal Hemodynamics

The mean arterial pressure in the glomerular capillaries is normally between 50 and 60 mm Hg, thereby favoring the outward filtration of plasma water along the entire length of the capillary loop. Approximately 20% of the plasma water that enters the glomerular capillaries is filtered into Bowman's capsule. The efferent arterioles provide greatest vascular resistance in the renal circulation and reduce the pressure in the peritubular capillaries to values between 10 and 20 mm Hg. These relatively low pressures favor the net reabsorption of the large quantities of fluid that pass from the renal tubules into the interstitium. The permeability of the peritubular capillaries is also considerably higher than other capillaries in the body, a feature that substantially facilitates the primary diffusion function of the kidney. Renal blood flow is approximately 20% of cardiac output and is heavily balanced toward perfusion of the renal cortex. The inner medulla and papillae usually receive only approximately one tenth of cortical blood flow.76

The kidney has a very high metabolic rate, but the organ extracts less than 10% of O2 present in renal arterial blood because renal perfusion far exceeds metabolic requirements. Renal blood flow is very important for the delivery of the large volumes of blood to the glomeruli required for ultrafiltration. Renal blood flow remains relatively constant between mean arterial pressures of 75 and 170 mm Hg, but becomes pressure-dependent beyond this range of autoregulation. Alterations in afferent arteriole resistance autoregulate glomerular filtration rate (GFR) by constricting the diameter of afferent arterioles in response to increases in driving pressure. The two primary mechanisms of renal autoregulation are myogenic (vascular smooth muscle intrinsically responds to stretch by constriction) and tubular-glomerular feedback. This later mechanism is mediated by a feedback loop in which an alteration in renal tubular filtrate flow is detected by the macula densa of the juxtaglomerular apparatus, which signals the afferent arterioles to restore basal levels of renal blood flow and GFR. The signal that regulates the caliber of the afferent arterioles in tubular-glomerular feedback has yet to be precisely defined, but many vasoactive substances have been implicated, including products of arachidonic acid metabolism, catecholamines, adenosine, nitric oxide, and components of the renin-angiotensin system.77,78,79The role of atrial natriuretic factor, a 28 amino-acid peptide with potent diuretic and natriuretic properties in the renal circulation, has also been elucidated. This peptide is synthesized and released primarily from the cardiac atria, and distention of the atria causes renal vasodilation, increased filtration, inhibition of sodium reabsorption, natriuresis, and a resultant reduction of extracellular fluid volume. The sympathetic nervous system innervates the kidney and may control tubular transport of Na+ during modest reductions in intravascular volume. Under conditions of profound hypovolemia, sympathetic activation causes renal vasoconstriction, lowers GFR, and reduces renal capillary hydrostatic pressure, thereby producing compensatory water retention that increases plasma volume. As a result of the distribution of blood flow within the kidney, perfusion to cortical compared with medullary nephrons is primary affected by sympathetic nervous system-induced renal vasoconstriction.

Splanchnic and Hepatic Circulation

Regulation of Gastrointestinal Blood Flow

The splanchnic circulation is unique. Arterial branches of the abdominal aorta supply blood to the gastrointestinal tract, spleen, and pancreas, whereas the liver has a dual blood supply consisting of the portal venous circulation and the hepatic artery. The intestinal circulation is weakly autoregulated compared with the cerebral, coronary, and renal vascular beds. Intestinal autoregulation appears to be primarily metabolic in origin. Adenosine is a likely mediator of this autoregulation, but other evidence suggests that K+ concentration and serum osmolality may also play contributing roles. The sympathetic nervous system innervates the gastrointestinal tract and the consequences of sympathetic activation are mediated by α-adrenoceptors. Pronounced sympathetic stimulation during acute hypovolemia produces gastrointestinal arterial constriction and venoconstriction, thereby shifting blood from a large vascular capacitance bed into the central circulation. Food ingestion increases gastrointestinal blood flow by several mechanisms, including the release of the hormones cholecystokinin and gastrin and absorption of gastrointestinal contents including glucose, fatty acids, and peptides.80

Regulation of Hepatic Blood Flow

The liver receives approximately 25% of total cardiac output, three quarters of which are derived from the portal vein that contains venous blood from the gastrointestinal tract, spleen, and pancreas. The remaining 25% of hepatic blood flow is provided by the hepatic artery, which supplies the majority of oxygen to the liver. Mean portal venous and hepatic arterial pressures are 10 and 90 mm Hg, respectively.38,81 The downstream resistance in the hepatic sinusoids is relatively low under normal circumstances, but may be elevated in RV failure or hepatic cirrhosis. A rise in sinusoidal and portal vein pressures accompanying these pathologic


conditions may produce transudation of fluid into the peritoneal space (ascites) or dilate alternative routes of venous drainage, such as those located in the lower esophageal veins (esophageal varices). Blood flow in the portal venous and hepatic arterial systems tends to vary reciprocally, but these respective hepatic blood supplies do not fully interact. Thus, a reduction of blood flow in the portal vein may be not fully compensated by an increase in hepatic arterial flow. The hepatic arterial but not the portal venous system is autoregulated. The most important response of hepatic arterial circulation to sympathetic stimulation is constriction of the presinusoidal resistance vessels.82 The liver contains about 15% of the total body blood volume and is an important volume reservoir that may be rapidly mobilized in response to sympathetic nervous system activation during acute hypovolemia. The reflex responses and response to hypoxia in small mesenteric capacitance vessels are inhibited by potent volatile anesthetics.83,84


1. Katz AM: Physiology of the Heart, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2006

2. Smith JJ, Kampine JP: The Heart: Structure and Function, Circulatory Physiology—The Essentials, 3rd edition. Baltimore, Williams & Wilkins, 1990

3. Berne RM, Levy MN: Cardiovascular Physiology, 8th edition. St. Louis, CV Mosby, 2001

4. Ledsome JHR, Lunden RJ: A reflex increase in heart rate from distention of the pulmonary vein atrial junction. J Physiol 1964; 170: 456

5. Seagard JL, Elegbe EO, Hopp FA, Bosnjak ZJ, von Colditz JH, Kalbfleisch JH, Kampine JP: Effects of isoflurane on the baroreceptor reflex. Anesthesiology 1983; 59: 511–520

6. Braunwald E. Sonenblick EH, Ross J: Contraction of the normal heart, Textbook of Cardiovascular Medicine, 2nd edition. Edited by Braunwald E. Philadelphia, WB Saunders, 1983, pp 409

7. Katz AM. The cardiac action potential, Physiology of the Heart, 2nd edition. New York, Raven Press, 1992, pp 438

8. Pagel PS, Grossman W, Haering JM, Warltier DC: Left ventricular diastolic function in the normal and diseased heart. Anestheiology 1993; 79: 1104

9. Pagel PS, Kampine JP, Schmeling WT, Warltier DC: Alteration of left ventricular diastolic function by desflurane, isoflurane, halothane in the chronically instrumented dog. Anesthesiology 1991; 74: 1103

10. Pagel PS, Kampine JP, Schmeling WT, Warltier DC: Comparison of end systolic pressure length relations and preload recruitable stroke work as indices of myocardial contractility in the conscious and anesthetized chronically instrumented dog. Anesthesiology 1990; 73: 278

11. Starling EF: The Linacre Lecture on the Law of the Heart. London, Logmans Green, 1918

12. Crohogini A, Barra J, Rodriquez CM, et al: Area beneath the end systolic pressure-volume relationship as an index of inotropic state in intact dogs. J Mol Cell Cardiol 1986; 18 (Suppl III) 20

13. Kass DA, Maaughan WL, Guo ZM, Kono A, Sunagawa K, Sagawa K: Comparative influence of load versus inotropic states on indexes of ventricular contractility. Experimental and theoretical analysis based on pressure-volume relationships. Circulation 1987; 76: 1422

14. Sagawa K: The ventricular pressure volume diagram revisited. Arc Res 1978; 43: 677

15. Sarnoff SJ: Myocardial contractility as described by ventricular function curves. Physiol Rev 1995; 35: 107

16. 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

17. Schaub MC, Hefti MA, Zuellig RA, Morano I: Modulation of contractility in human cardiac hypertrophy by myosin essential light chain isoforms. Cardiovasc Res 1998; 37: 381

18. Cazorla O, Vassort G, Garnier D, Le Guennc JY: Length modulation of active force in rat cardiac myocytes: in titin the sensor? J Mol Cell Cardiol 1999; 31: 1215

19. Helmes M, Trombitas K, Ganzier H: Titin develops restoring force in rat cardiac myocytes. Circ Res 1996; 79: 619

20. Schiaffino S, Reggiani C: Molecular diversity of myofibrillar proteins: gene regulation and molecular significance. Physiol Rev 1996; 76:371

21. Goldstein MA, Schroeter JP, Michael LH: Role of the Z band in the mechanical properties of the heart. FASEB J 1977; 5: 2167–2174

22. Moncman CL, Wang K: Nebulette: A 107 kD nebulin-like protein in cardiac muscle. Cell Motil Cytoskel 1995; 32: 205

23. Solaro RJ, Rarick HM: Troponin and tropomyosin. Proteins that switch on and tune in the activity of cardiac myofilaments. Circ Res 1998; 83: 417

24. Tobacman LS: Thin filament-mediated regulation of cardiac contraction. Ann Rev Physiol 1996; 58: 447

25. Solaro RJ, Van Eyk J: Altered interactions among thin filaments proteins modulate cardiac function. J Mol Cell Cardiol 1999; 28: 217

26. Luo W, Grupp IL, Harrer J, Ponniah S, Grupp G, Duffy JJ, Doetschman T, Kranias EG: Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of β-agonist stimulation. Circ Res 1994; 75: 401

27. Rayment I, Holden HM, Whittaker M: Structure of the actin-myosin complex and its implications for muscle contraction. Science 1993; 261: 58

28. Dominguez R, Freyzon Y, Trybus KM, Cohen C: Crystal structure of a vertebrate smooth muscle myosin motor domain and its complex with the essential light chain: visualization of the pre-power stroke state. Cell 1998; 94: 559

29. Finer JT, Simmons RM, Spudich JA: Single myosin molecule mechanics: pico-newton forces and nanometer steps. Nature 1994; 368: 113

30. Katz AM. The electrocardiogram, Physiology of the Heart, 2nd edition. New York. Raven Press, 1992, p 472

31. Smith JJ, Kampine JP: Electrical properties of the heart, Circulatory Physiology—The Essentials, 3rd edition. Baltimore, Williams & Wilkins, 1990

32. Ebert TJ, Kotrly, KJ, Vucins EJ, Pattison CZ, Kampine JP: Effects of halothane anesthesia on cardiopulmonary baroreflex function in man. Anesthesiology 1985; 61: 668

33. Baruscotti M, Bucchi A, Di Francesco D: Physiology and pharmacology of the cardiac pacemaker (“funny”) current. Pharmacol Ther 2005; 107: 59

34. Nerbonne JM, Kass RS: Molecular physiology of cardiac repolarization. Physiol Rev 2005; 85: 1205

35. McAllen RM, Spyer KM: The location of cardiac vagal preganglionic motoneurons in the medulla. J Physiol 1976; 258: 187

36. Kampine JP: General cardiovascular regulation, International Practice of Anaesthesia. Edited by Prys-Roberts C, Brown BR Jr. Oxford, Butterworth-Heinemann, 1996

37. Lipski J, Kanjhan R, Kruszewska B, Rong W: Properties of presympathetic neurons in the rostral ventrolateral medullar in the rat: An intracellular study “in vivo”. J Physiol 1996; 490: 729

38. Guyenent PJ: Role of the ventral medulla oblongate in blood pressure regulation, Central Regulation of Autonomic Functions. Edited by Loewy AD, Spyer KM. New York, Oxford University Press, 1990

39. Jarisch A, Richter H: Die afferention bahnen des veratrius effektes in dem herz nerven. Arch Exp Path Pharmacol 1939; 193: 355

40. Dampney RA: Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 1994; 74: 323

41. Spyer KM: The central nervous organization of reflex circulatory control, Central Regulation of Autonomic Function. Edited by Loewy AD, Spyer KM. New York, Oxford University Press, 1990

42. Ross CA, Ruggerio DA, Park DH, Joh TH, Sved AF, Fernandez-Pardal J, Saavedra JM, Reis DJ: Tonic vasomotor control of the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area containing C1 neurons on arterial blood pressure, heart rate, and plasma catecholamines and vasopressin. J Neurosci 1984; 4: 474

43. Gauer OH, Henry JP: Neurohumoral control of plasma volume, Cardiovascular Physiology II International Review of Physiology, Vol 9. Edited by Guyton AC, Cowley AW. Baltimore, University Park Press 1976, pp 145

44. Linden RJ: Atrial receptors and heart rate, Cardiac Receptors. Edited by Hainsworth R, Kidd C, Linden RJ. London, Cambridge University Press, 1979

45. Longhurst JC: Cardiac receptors: Their function in health and disease. Prog Cardiovasc Dis 1984; 27: 201

46. Kotrly KJ, Ebert TJ, Vucins EJ, Roerig DL, Kampine JP: Baroreceptor reflex control of heart rate during morphine sulfate, diazepam N2O/O2 anesthesia in man. Anesthesiology 1984; 61: 558

47. Seagard JL, Hopp FA, Donegan JH, Kampine JP: Halothane and the carotid sinus reflex: Evidence for multiple sites of action. Anesthesiology 1982; 57: 191

48. Crawford MH: Current Diagnosis and Treatment in Cardiology. Norwalk, CT, Appleton & Lange, 1995, pp 1

49. Libby PP, Bronow RO, Mann DL, et al: Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine, 8th edition. Philadelphia, Elsevier Saunders, 2001

50. Marcus M: The Coronary Circulation in Health and Disease. New York, McGraw Hill, 1983, pp 465

51. Lance KL: Coronary Artery Stenosis. New York, Elsevier, 1991

52. Lilly LS: Pathophysiology of Heart Disease: A Collaborative Project of Medical Students and Faculty, 4th edition. Philadelphia, Lippincott Williams & Wilkins, 2007

53. Tune JD, Richmond KN, Gorman MW, Feigl EO: Control of coronary blood flow during exercise. Exp Biol Med 2002; 227: 238

54. Cardiac Vascular, and Thoracic Anesthesia. Edited by Youngberg JA, Lake CL, Roisen MF, Wilson RS. Elsevier, 2000, pp 1


55. Topol EJ, Califf RM, Prystowsky EN, et al: Textbook of Cardiovascular Medicine, 3rd edition. Philadelphia, Lippincott, Williams & Wilkins, 2007, pp 1628

56. King SB, Yeung AC: Interventional Cardiology. New York, McGraw-Hill Medical, 2007

57. Frazier OH, Westaby S: Ischemia Heart Disease: Surgical Management. London, Mosby, 1999, pp 1

58. Opie LH: Heart Physiology: From Cell to Circulation, 4th edition. Philadelphia, Lippincott, Williams & Wilkins, 2004, pp 648

59. Phibbs B: The Human Heart: A Basic Guide to Heart Disease, 2nd edition. Philadelphia, Lippincott, Williams & Wilkins, 2007, pp 229

60. Gerstenblith G: Cardiovascular Disease in the Elderly. Totowa, NJ, Humana Press, 2005

61. Nunn JF: Applied Respiratory Physiology. London, Butterworth, 1971, pp 213

62. Szidon JP, Fishmann AP: Autonomic control of the pulmonary circulation, The Pulmonary Circulation and Interstitial Space. Edited by Fishman AP, Hecht HH, Chicago, The University of Chicago Press, 1969, pp 239

63. West JB, Collery CT: Distribution of blood flow and the pressure-flow relations of the whole lung. J App Physiol 1965; 20: 175

64. West JB, Naimark AA: Distribution of blood flow in isolated lung; relation to vascular and alveolar pressures. J Appl Physiol 1964; 19: 713

65. Mauban JR, Reillard CV, Yuan JZ: Hypoxic pulmonary vasoconstriction: Role of ion channels. J Appl Physiol 2005; 98: 415

66. Waypa GB, Shumacker PT: Hypoxic pulmonary vasoconstriction: Redox events in oxygen sensing. J Appl Physiol 2005; 98: 404

67. Buzzard CJ, Pfister SL, Campbell WB: Endothelium-dependent contractions in rabbit pulmonary artery are mediated by thromboxaine A2. Circ Res 1993; 72: 1023

68. Pfister SL, Campbell WB: Role of endothelium derived metabolites of arachidonic acid in enhanced pulmonary artery contractions in female rabbit. Hypertension 1996; 27: 43

69. Zhu D, Medhora M, Campbell WB, Spitzbarth N, Baker JE, Jacobs ER: Chronic hypoxia activates lung 15-lipoxygenase which catalyzes production of 15-HETE and enhances constriction in neonatal pulmonary arteries. Circ Res 2003; 92: 992

70. Greenberg J, Hand P, Sylverstro A, et al: Localized metabolic flow couple during functional activity. Acta Neurol Scand 1979; 60(Suppl 72): 12

71. Olsen J: Contralateral local increase in cerebral blood flow in man during arm work. Brain 1972; 94: 635

72. Gebrenedin D, Lange A, Lowry T, Taheri MR, Birks EK, Hudetz AG, Narayanan J, Falck JR, Okamoto H, Roman RJ, Nithipatikom K, Campbell WB, Harder DR: Production of 20-HETE and its role in autoregulation of cerebral blood flow. Circ Res 2000; 87: 60

73. Lou HC, Edvinsson L, MacKenzie ET: The concept of coupling blood flow to brain function: revision required? Ann Neurol 1987; 22: 289

74. Faraci FM, Heistad DD: Regulation of the cerebral circulation: Role of endothelium and potassium channels. Physiol Rev 1998; 78: 53

75. Levy MN, Pappano AJ: Cardiovascular Physiology. Philadelphia, Mosby Elsevier, 2007

76. Hall JE. Regulation of renal hemodynamics, Cardiovascular Physiology II. International Review of Physiology, Vol. 2. Edited by Guyton EC, Hall JE. Baltimore, University Park Press, 1982, pp 243

77. Benner BM: Glomerular ultrafiltration, The Kidney, Vol 1, 3rd edition. Edited by Brenner BM, Bector FC. Philadelphia, WB Saunders, 1986

78. Naraayanan J, Imig M. Roman R, Harder D. Pressurization of isolated renal arterial increases inositol tripghosphate and diacylglycerol. Am J Physiol Heart Circ Physiol 1994; 35: H1840

79. Roman RR: P-450 metabolites of arachadonic acid in the control of cardiovascular function. Physiol Rev 2002; 82: 131

80. Jacobson Ed, Tepperman BL, editors: Splanchnic circulation (symposium). Fed Proc 1982; 41, 2079

81. Greenway CV, Laut WW. Hepatic circulation, Handbook of Physiology, Section 6: The Gastrointestinal System—Motility and Circulation, Vol. 1. Edited by Schultz SG. Bethesda, MD, American Physiological Society, 1989

82. Ozono K, Bosnjak ZJ, Kampine JP. Reflex control of mesenteric venous capacitance in the rabbit: Direct measurement of vein diameter and intravenous pressure in the situ preparation. Am J Physiol 1989; 256: H1066

83. Stadnicka A, Stekiel TA, Bosnjak ZJ, Kampine JP: Hypoxic contractions of isolated rabbit small mesenteric capacitance veins: Contributions of endothelium and attenuation by volatile anesthetics. Anesthesiology 1995; 82: 550

84. Stekiel TA, Ozono K, McCallum JB, Bosnjak ZJ, Stekiel WJ, Kampine JP: The inhibitory action of halothane on reflex constriction in mesenteric capacitance veins. Anesthesiology 1990; 73: 1169

Editors: Barash, Paul G.; Cullen, Bruce F.; Stoelting, Robert K.; Cahalan, Michael K.; Stock, M. Christine