patho-physiology of heart disease

CARDIAC ANATOMY AND HISTOLOGYPericardiumSurface Anatomy of the HeartInternal Structure of the HeartImpulse-Conducting SystemCardiac InnervationCardiac VesselsHistology of Ventricular Myocardial Cells

BASIC ELECTROPHYSIOLOGYIon Movement and ChannelsResting Potential

Action PotentialRefractory PeriodsImpulse ConductionNormal Sequence of Cardiac Depolarization

EXCITATION-CONTRACTION COUPLINGContractile Proteins in the MyocyteCalcium-Induced Calcium Release and theContractile Cycleβ-Adrenergic and Cholinergic Signaling

1

C H A P T E R

1Basic Cardiac Structureand FunctionVivek IyerElazer R. EdelmanLeonard S. Lilly

A knowledge of normal cardiac structureand function is crucial to understanding dis-eases that afflict the heart. This chapter re-views basic cardiac anatomy and electro-physiology as well as the events that lead tocardiac contraction.

CARDIAC ANATOMY AND HISTOLOGY

Although the study of cardiac anatomy datesback to ancient times, interest in this fieldhas recently gained momentum. The devel-opment of sophisticated cardiac-imagingprocedures such as coronary angiography,echocardiography, computed tomography,and magnetic resonance imaging has made

essential an intimate knowledge of the spa-tial relationships of cardiac structures. Suchinformation also proves helpful in under-standing the pathophysiology of heart dis-ease. This section emphasizes the aspects ofcardiac anatomy that are important to theclinician—that is, the “functional” anatomy.

Pericardium

The heart and roots of the great vessels areenclosed by a fibroserous sac called the peri-cardium (Fig. 1.1). This structure consists oftwo layers: a strong outer fibrous layer andan inner serosal layer. The inner serosal layeradheres to the external wall of the heartand is called the visceral pericardium. The

Fig. 1

10090-01_CH01.qxd 8/24/06 5:40 PM

visceral pericardium reflects back on itselfand lines the outer fibrous layer, formingthe parietal pericardium. The space be-tween the visceral and parietal layers con-tains a thin film of pericardial fluid that al-lows the heart to beat in a minimal-frictionenvironment.

The pericardium is attached to the ster-num and the mediastinal portions of theright and left pleurae. Its many connectionsto surrounding structures keep the pericar-dial sac firmly anchored within the thoraxand therefore help to maintain the heart inits normal position.

Emanating from the pericardium in a su-perior direction are the aorta, the pulmonaryartery, and the superior vena cava (see Fig.1.1). The inferior vena cava projects throughthe pericardium inferiorly.

Surface Anatomy of the Heart

The heart is shaped roughly like a cone andconsists of four muscular chambers. Theright and left ventricles are the main pump-ing chambers. The less muscular right andleft atria deliver blood to their respectiveventricles.

Several terms are used to describe theheart’s surfaces and borders (Fig. 1.2). Theapex is formed by the tip of the left ventricle,which points inferiorly, anteriorly, and tothe left. The base or posterior surface of the

heart is formed by the atria, mainly the left,and lies between the lung hila. The anteriorsurface of the heart is shaped by the rightatrium and ventricle. Because the left atriumand ventricle lie more posteriorly, they formonly a small strip of this anterior surface.The inferior surface of the heart is formed byboth ventricles, primarily the left. This sur-face of the heart lies along the diaphragm;hence, it is also referred to as the diaphrag-matic surface.

Observing the chest from an anteroposte-rior view (as on a chest radiograph; see Chap-ter 3), four recognized borders of the heart areapparent. The right border is established bythe right atrium and is almost in line withthe superior and inferior vena cavae. The inferior border is nearly horizontal and isformed mainly by the right ventricle, with aslight contribution from the left ventriclenear the apex. The left ventricle and a por-tion of the left atrium make up the left bor-der of the heart, whereas the superior borderis shaped by both atria. From this descriptionof the surface of the heart emerge two basic“rules” of normal cardiac anatomy: (1) right-sided structures lie mostly anterior to theirleft-sided counterparts, and (2) atrial cham-bers are located mostly to the right of theircorresponding ventricles.

Internal Structure of the Heart

Four major valves in the normal heart directblood flow in a forward direction and pre-vent backward leakage. The atrioventricularvalves (tricuspid and mitral) separate theatria and ventricles, whereas the semilunarvalves (pulmonic and aortic) separate theventricles from the great arteries. All fourheart valves are attached to the fibrous car-diac skeleton (Fig. 1.3), which is composedof dense connective tissue. The cardiac skele-ton also serves as a site of attachment for theventricular and atrial muscles.

The surface of the heart valves and theinterior surface of the chambers are lined bya single layer of endothelial cells, termedthe endocardium. The subendocardial tis-sue contains fibroblasts, elastic and col-lagenous fibers, veins, nerves, and branches

2 Chapter One

Figure 1.1. The position of the heart in the chest.The superior vena cava, aorta, and pulmonary artery exit superiorly, whereas the inferior vena cava projectsinferiorly.

Superiorvena cava

Pulmonaryartery

Heart withinpericardium

Diaphragm

Aorta

Inferiorvena cava

Fig. 2

Fig. 3

10090-01_CH01.qxd 8/24/06 5:40 PM

Basic Cardiac Structure and Function 3

Brachiocephalic a. Left common carotid a.

Left subclavian a.

Aortic arch

Left pulmonary a.

Left pulmonary vv.

Left atrial appendage

Left ventricle

Anterior interventriculargroove

Brachiocephalic a.

Azygous v.

Superior vena cava

Right pulmonary a.

Right pulmonary vv.

Right atrium

Inferior vena cava

Left heart border

Superior vena cava

Right pulmonary a.

Right pulmonary vv.

Right atrium

Right ventricle

Inferior vena cava

Right heart border

Inferior heart border

Inferior heart border

Apex

A

Left common carotid a.

Leftatrium

Left subclavian a.

Aortic arch

Lig. arteriosum

Left pulmonary a.

Left pulmonary vv.

Left ventricle

B

Figure 1.2. The heart and great vessels. A. The anterior view. B. The posterior aspect (or base), asviewed from the back. a, artery; lig, ligamentum; vv, veins.

10090-01_CH01.qxd 8/24/06 5:40 PM

of the conducting system and is continuouswith the connective tissue of the heart musclelayer, the myocardium. The myocardium isthe thickest layer of the heart and consists ofbundles of cardiac muscle cells, the histol-ogy of which is described later in the chap-ter. External to the myocardium is a layerof connective tissue and adipose tissuethrough which pass the larger blood vesselsand nerves that supply the heart muscle.The epicardium is the outermost layer ofthe heart and is identical to, and just an-other term for, the visceral pericardium pre-viously described.

Right Atrium and Ventricle

Opening into the right atrium are the su-perior and inferior vena cavae and the coro-nary sinus (Fig. 1.4). The vena cavae returndeoxygenated blood from the systemic veinsinto the right atrium, whereas the coronarysinus carries venous return from the coronaryarteries. The interatrial septum forms theposteromedial wall of the right atrium andseparates it from the left atrium. The tricus-

pid valve is located in the floor of the atriumand opens into the right ventricle.

The right ventricle (see Fig. 1.4) is roughlytriangular in shape, and its superior aspectforms a cone-shaped outflow tract, whichleads to the pulmonary artery. Although theinner wall of the outflow tract is smooth,the rest of the ventricle is covered by a num-ber of irregular bridges (termed trabeculaecarneae) that give the right ventricular walla spongelike appearance. A large trabeculathat crosses the ventricular cavity is calledthe moderator band. It carries a componentof the right bundle branch of the conductingsystem to the ventricular muscle.

The right ventricle contains three papil-lary muscles, which project into the cham-ber and via their thin, stringlike chordaetendineae attach to the edges of the tricus-pid valve leaflets. The leaflets, in turn, are at-tached to the fibrous ring that supports thevalve between the right atrium and ventri-cle. Contraction of the papillary musclesprior to other regions of the ventricle tight-ens the chordae tendineae, helping to alignand restrain the leaflets of the tricuspid

4 Chapter One

Anterior

Posterior

Aorticvalve

Pulmonicvalve

Tricuspidvalve

Annulusfibrosus

Mitralvalve

Annulusfibrosus

Figure 1.3. The four heart valves viewed from above with atria removed.The figure depicts the period of ventricular filling (diastole) during which the tri-cuspid and mitral valves are open and the semilunar valves (pulmonic and aortic)are closed. Each annulus fibrosus surrounding the mitral and tricuspid valves isthicker than those surrounding the pulmonic and aortic valves; all four contributeto the heart’s fibrous skeleton, which is composed of dense connective tissue.

Fig. 4

10090-01_CH01.qxd 8/24/06 5:40 PM


valve as they are forced closed. This actionprevents blood from regurgitating into theright atrium during ventricular contraction.

At the apex of the right ventricular outflowtract is the pulmonic valve, which leads tothe pulmonary artery. This valve consists ofthree cusps attached to a fibrous ring. Duringrelaxation of the ventricle, elastic recoil of thepulmonary arteries forces blood back towardthe heart, distending the valve cusps towardone another. This action closes the pulmonicvalve and prevents regurgitation of blood backinto the right ventricle.

Left Atrium and Ventricle

Entering the posterior half of the left atriumare the four pulmonary veins (Fig. 1.5). Thewall of the left atrium is about 2 mm thick,being slightly greater than that of the rightatrium. The mitral valve opens into the leftventricle through the inferior wall of the leftatrium.

The cavity of the left ventricle is approx-imately cone shaped and longer than that ofthe right ventricle. In a healthy adult heart,the wall thickness is 9 to 11 mm, roughly 3times that of the right ventricle. The aorticvestibule is a smooth-walled part of the leftventricular cavity located just inferior to the

aortic valve. Inferior to this region, most ofthe ventricle is covered by trabeculae carneae,which are finer and more numerous thanthose in the right ventricle.

The left ventricular chamber (see Fig. 1.5B)contains two large papillary muscles. Theseare larger than their counterparts in theright ventricle, and their chordae tendineaeare thicker but less numerous. The chordaetendineae of each papillary muscle distrib-ute to both leaflets of the mitral valve. Sim-ilar to the case in the right ventricle, tensingof the chordae tendineae during left ven-tricular contraction helps restrain and alignthe mitral leaflets, enabling them to closeproperly and preventing the backward leak-age of blood.

The aortic valve separates the left ventri-cle from the aorta. Surrounding the aorticvalve opening is a fibrous ring to which is at-tached the three cusps of the valve. Justabove the right and left aortic valve cusps inthe aortic wall are the origins of the rightand left coronary arteries (see Fig. 1.5B).

Interventricular Septum

The interventricular septum is the thick wallbetween the left and right ventricles. It iscomposed of a muscular and a membranous

Superior vena cava

Pulmonary artery

Pulmonic valve

Interventricular septum

Moderatorband

Trabeculaecarneae

Papillary muscles

Aorta

Right atrium

Inferior vena cava

Coronary sinus

Tricuspid valve Right ventricle

Figure 1.4. Interior structures of the right atrium and right ventricle. (Modified from Goss CM.Gray’s Anatomy. 29th Ed. Philadelphia: Lea & Febiger, 1973:547.)

Fig. 5

10090-01_CH01.qxd 8/24/06 5:40 PM

part (see Fig. 1.5B). The margins of this sep-tum can be traced on the surface of the heartby following the anterior and posterior in-terventricular grooves. Owing to the greaterhydrostatic pressure within the left ventri-cle, the large muscular portion of the sep-tum bulges toward the right ventricle. The

small, oval-shaped membranous part of theseptum is thin and located just inferior tothe cusps of the aortic valve.

To summarize the functional anatomicpoints presented in this section, the follow-ing is a review of the path of blood flowthrough the heart: Deoxygenated blood is

6 Chapter One

Left ventricle

Right ventricle

Aortic valve

AortaRight pulmonaryveins

Left atrium

Posterior leafletof mitral valve

Papillary muscle

Pulmonaryartery

Interventricularseptum

A

Posterior cuspof aortic valve

Origin of leftcoronary artery

Anterior cuspof mitral valve

Chordaetendineae

AnteriorpapillarymuscleRIGHT

VENTRICLE

Pulmonary artery

Origin of right coronary artery

Interventricularseptum,membranouspart

Interventricularseptum, muscularpart

Posteriorpapillarymuscle

Trabeculaecarneae

B

AORTA

Figure 1.5. Interior structures of the left atrium and left ventricle. A. The leftatrium and left ventricular (LV) inflow and outflow regions. B. Interior structures of the LVcavity. (Modified from Agur AMR, Lee MJ. Grant’s Atlas of Anatomy. 9th Ed. Baltimore:Williams & Wilkins, 1991:59.)

10090-01_CH01.qxd 8/24/06 5:40 PM


delivered to the heart through the inferiorand superior vena cavae, which enter intothe right atrium. Flow continues throughthe tricuspid valve orifice into the right ven-tricle. Contraction of the right ventricle pro-pels the blood across the pulmonic valve tothe pulmonary artery and lungs, where car-bon dioxide is released and oxygen is ab-sorbed. The oxygen-rich blood returns tothe heart through the pulmonary veins tothe left atrium and then passes across themitral valve into the left ventricle. Contrac-tion of the left ventricle pumps the oxy-genated blood across the aortic valve intothe aorta, from which it is distributed to allother tissues of the body.

Impulse-Conducting System

The impulse-conducting system (Fig. 1.6)consists of specialized cells that initiate theheartbeat and electrically coordinate con-tractions of the heart chambers. The sino-atrial (SA) node is a small mass of specializedcardiac muscle fibers in the wall of the rightatrium. It is located to the right of the supe-

rior vena cava entrance and normally initi-ates the electrical impulse for contraction.The atrioventricular (AV) node lies be-neath the endocardium in the inferoposte-rior part of the interatrial septum.

Distal to the AV node is the bundle ofHis, which perforates the interventricularseptum posteriorly. Within the septum, thebundle of His bifurcates into a broad sheet offibers that continues over the left side of theseptum, known as the left bundle branch,and a compact, cablelike structure on theright side, the right bundle branch.

The right bundle branch is thick anddeeply buried in the muscle of the inter-ventricular septum and continues towardthe apex. Near the junction of the interven-tricular septum and the anterior wall of theright ventricle, the right bundle branch be-comes subendocardial and bifurcates. Onebranch travels across the right ventricularcavity in the moderator band, whereas theother continues toward the tip of the ven-tricle. These branches eventually arborizeinto a finely divided anastomosing plexusthat travels throughout the right ventricle.

Fig. 6

Sinoatrial node

Coronary sinus

Atrioventricular node

Bundle of His

Moderator band

Right bundle branch

Mitral valve

Membranous part ofIV septum

Bifurcation of bundleof His

Muscular part ofIV septum

Left bundle branch

Purkinje fibers underendocardium of papillarymuscle

Figure 1.6. Main components of the cardiac conduction system. This system includes the sinoatrial node, atrioventricular node, bundle of His, right and left bundle branches, andthe Purkinje fibers. The moderator band carries a large portion of the right bundle. IV, intra-ventricular.

10090-01_CH01.qxd 8/24/06 5:40 PM

Functionally, the left bundle branch is di-vided into an anterior and a posterior fascicleand a small branch to the septum. The ante-rior fascicle runs anteriorly toward the apex,forming a subendocardial plexus in the areaof the anterior papillary muscle. The poste-rior fascicle travels to the area of the posteriorpapillary muscle; it then divides into a suben-docardial plexus and spreads to the rest of theleft ventricle.

The subendocardial plexuses of both ven-tricles send distributing Purkinje fibers tothe ventricular muscle. Impulses within theHis-Purkinje system are transmitted first tothe papillary muscles and then throughoutthe walls of the ventricles, allowing papil-lary muscle contraction to precede that ofthe ventricles. This coordination preventsregurgitation of blood flow through the AVvalves, as discussed earlier.

Cardiac Innervation

The heart is innervated by both parasympa-thetic and sympathetic afferent and efferentnerves. Preganglionic sympathetic neurons lo-cated within the upper five to six thoraciclevels of the spinal cord synapse with second-order neurons in the cervical sympatheticganglia. Traveling within the cardiac nerves,these fibers terminate in the heart and greatvessels. Preganglionic parasympathetic fibersoriginate in the dorsal motor nucleus of themedulla and pass as branches of the vagusnerve to the heart and great vessels. Here thefibers synapse with second-order neurons lo-cated in ganglia within these structures. Arich supply of vagal afferents from the infe-rior and posterior aspects of the ventriclesmediates important cardiac reflexes, whereasthe abundant vagal efferent fibers to the SAand AV nodes are active in modulating elec-trical impulse initiation and conduction.

Cardiac Vessels

The cardiac vessels consist of the coronaryarteries and veins and the lymphatics. Thelargest components of these structures liewithin the loose connective tissue in theepicardial fat.

Coronary Arteries

The heart muscle is supplied with oxygen andnutrients by the right and left coronary arter-ies, which arise from the root of the aorta justabove the aortic valve cusps (Fig. 1.7; see alsoFig. 1.5B). After their origin, these vesselspass anteriorly, one on each side of the pul-monary artery (see Fig. 1.7).

The large left main coronary artery passesbetween the left atrium and the pulmonarytrunk to reach the AV groove. There it di-vides into the left anterior descending(LAD) coronary artery and the circumflexartery. The LAD travels within the anteriorinterventricular groove toward the cardiacapex. During its descent on the anterior sur-face, the LAD gives off septal branches thatsupply the anterior two thirds of the inter-ventricular septum and the apical portion ofthe anterior papillary muscle. The LAD alsogives off diagonal branches that supply theanterior surface of the left ventricle. The cir-cumflex artery continues within the left AVgroove and passes around the left border ofthe heart to reach the posterior surface. Itgives off large obtuse marginal branchesthat supply the lateral and posterior wall ofthe left ventricle.

The right coronary artery (RCA) travelsin the right AV groove, passing posteriorlybetween the right atrium and ventricle. Itsupplies blood to the right ventricle via acutemarginal branches. In most people, the distalRCA gives rise to a large branch, the poste-rior descending artery (see Fig. 1.7C). Thisvessel travels from the inferoposterior aspectof the heart to the apex and supplies blood tothe inferior and posterior walls of the ventri-cles and the posterior one third of the inter-ventricular septum. Just before giving off theposterior descending branch, the RCA usu-ally gives off the AV nodal artery.

The posterior descending and AV nodalarteries arise from the RCA in 85% of thepopulation, and in such people, the coro-nary circulation is termed right dominant. Inapproximately 8%, the posterior descendingartery arises from the circumflex artery in-stead, resulting in a left-dominant circula-tion. In the remaining population, the heart’s

8 Chapter One

Fig. 7

10090-01_CH01.qxd 8/24/06 5:40 PM


posterior blood supply is contributed tofrom branches of both the RCA and the cir-cumflex, forming a codominant circulation.

The blood supply to the SA node is alsomost often (70% of the time) derived fromthe RCA. However, in 25% of normal hearts,the SA nodal artery arises from the circum-flex artery, and in 5% of cases, both the RCAand the circumflex artery contribute to thisvessel.

From their epicardial locations, the coro-nary arteries send perforating branches intothe ventricular muscle, which form a richlybranching and anastomosing vasculature inthe walls of all the cardiac chambers. Fromthis plexus arise a massive number of cap-illaries that form an elaborate network surrounding each cardiac muscle fiber. Themuscle fibers located just beneath the endo-cardium, particularly those of the papillary

Pulmonary artery

Left circumflexcoronary artery

Left maincoronary artery

Aorta

Left anteriordescendingcoronary artery

Right coronaryartery

Rightcoronaryartery

Acutemarginal branch


Left anterior descending coronary artery

Diagonalbranch


Obtusemarginalbranches

Posterior descendingcoronary artery

Rightcoronaryartery

A

B C

Figure 1.7. Coronary artery anatomy. A. Schematic representation of the right and left coronary arteries demon-strates their orientation to one another. The left main artery bifurcates into the circumflex artery, which perfuses the lat-eral and posterior regions of the left ventricle (LV), and the anterior descending artery, which perfuses the LV anteriorwall, the anterior portion of the intraventricular septum, and a portion of the anterior right ventricular (RV) wall. Theright coronary artery (RCA) perfuses the right ventricle and variable portions of the posterior left ventricle through itsterminal branches. The posterior descending artery most often arises from the RCA. B. Anterior view of the heart demon-strating the coronary arteries and their major branches. C. Posterior view of the heart demonstrating the terminal por-tions of the right and circumflex coronary arteries and their branches.

10090-01_CH01.qxd 8/24/06 5:40 PM

muscles and the thick left ventricle, are sup-plied either by the terminal branches of thecoronary arteries or directly from the ven-tricular cavity through tiny vascular chan-nels, known as thebesian veins.

Collateral connections, usually <200 µmin diameter, exist at the subarteriolar levelbetween the coronary arteries. In the nor-mal heart, few of these collateral vessels arevisible. However, they may become largerand functional when atherosclerotic diseaseobstructs a coronary artery, thereby provid-ing blood flow to distal portions of the ves-sel from a nonobstructed neighbor.

Coronary Veins

The coronary veins follow a distributionsimilar to that of the major coronary arter-ies. These vessels return blood from the myo-cardial capillaries to the right atrium pre-dominantly via the coronary sinus. Themajor veins lie in the epicardial fat, usuallysuperficial to their arterial counterparts. Thethebesian veins, described earlier, providean additional potential route for a smallamount of direct blood return to the cardiacchambers.

Lymphatic Vessels

The heart lymph is drained by an exten-sive plexus of valved vessels located in thesubendocardial connective tissue of all fourchambers. This lymph drains into an epi-cardial plexus from which are derived sev-eral larger lymphatic vessels that follow thedistribution of the coronary arteries andveins. Each of these larger vessels then com-bines in the AV groove to form a single lym-phatic conduit, which exits the heart toreach the mediastinal lymphatic plexus andultimately the thoracic duct.

Histology of VentricularMyocardial Cells

The mature myocardial cell (also termed themyocyte) measures up to 25 µm in diame-ter and 100 µm in length. The cell shows a cross-striated banding pattern similar to

that of skeletal muscle. However, unlike themultinucleated skeletal myofibers, myocar-dial cells contain only one or two centrallylocated nuclei. Surrounding each myocar-dial cell is connective tissue with a rich cap-illary network.

Each myocardial cell contains numerousmyofibrils, which are long chains of indi-vidual sarcomeres, the fundamental con-tractile units of the cell (Fig. 1.8). Each sarcomere is made up of two groups of overlapping filaments of contractile pro-teins. Biochemical and biophysical interac-tions occurring between these myofila-ments produce muscle contraction. Theirstructure and function are described later inthe chapter.

Within each myocardial cell, the neigh-boring sarcomeres are all in register, produc-ing the characteristic cross-striated bandingpattern seen by light microscopy. The rela-tive densities of the cross bands identify thelocation of the contractile proteins. Underphysiologic conditions, the overall sarco-mere length (Z-to-Z distance) varies between2.2 and 1.5 µm during the cardiac cycle.The larger dimension reflects the fiberstretch during ventricular filling, whereasthe smaller dimension represents the extentof fiber shortening during contraction.

The myocardial cell membrane is termedthe sarcolemma. A specialized region of themembrane is the intercalated disk, a dis-tinct characteristic of cardiac muscle tissue.Intercalated disks are seen on light micro-scopic study as darkly staining transverselines that cross chains of cardiac cells at irregular intervals. They represent the gapjunction complexes at the interface of adja-cent cardiac fibers and establish structuraland electrical continuity between the my-ocardial cells.

Another functional feature of the cellmembrane is the transverse tubular system(or T tubules). This complex system is char-acterized by deep, fingerlike invaginationsof the sarcolemma (Fig. 1.9; see also Fig. 1.8).Similar to the intercalated disks, transversetubular membranes establish pathways forrapid transmission of the excitatory electri-cal impulses that initiate contraction. The

10 Chapter One

Fig. 8

Fig. 9

10090-01_CH01.qxd 8/24/06 5:40 PM


Myofibril

ACTIN MYOSIN

Sarcolemma

Mitochondrion

Mitochondrion

SarcoplasmicReticulum

T tubule

SARCOMERE

Figure 1.8. Myocardial cell. Top. Schematic representation of the ultrastructure ofthe myocardial cell. The cell consists of multiple parallel myofibrils surrounded by mi-tochondria. The T tubules are invaginations of the cell membrane (the sarcolemma)that increase the surface area for ion transport and transmission of electrical impulses.The intracellular sarcoplasmic reticulum houses most of the intracellular calcium andabuts the T tubules. (Modified from Katz AM. Physiology of the Heart. 2nd Ed. NewYork: Raven Press, 1992:21.) Bottom. Expanded view of a sarcomere, the basic unitof contraction. Each myofibril consists of serially connected sarcomeres that extendfrom one Z line to the next. The sarcomere is composed of alternating thin (actin) andthick (myosin) myofilaments.

T tubule

Sarcolemma

Terminal cisternae

ATPase

Sarcoplasmicreticulum

Ca++

Ca++

Ca++

Ca++ Ca++Ca++

Ca++

Ca++

Ca++

Ca++Ca++

Figure 1.9. Schematic view of the tubular systems of the myocardial cell. The Ttubules, invaginations of the sarcolemma, abut the sarcoplasmic reticulum at right anglesat the terminal cisternae sacs. This relationship is important in linking membrane excita-tion with intracellular release of calcium from the sarcoplasmic reticulum.

10090-01_CH01.qxd 8/24/06 5:40 PM

T tubule system increases the surface area of the sarcolemma in contact with the extra-cellular environment, allowing the trans-membrane ion transport accompanying ex-citation and relaxation to occur quickly andsynchronously.

The sarcoplasmic reticulum (SR) is anextensive intracellular tubular membranenetwork that complements the T tubule sys-tem both structurally and functionally. TheSR abuts the T tubules at right angles in lat-eral sacs, called the terminal cisternae (seeFig. 1.9). These sacs house most of intracel-lular calcium stores; the release of thesestores is important in linking membrane ex-citation with activation of the contractileapparatus. Lateral sacs also abut the interca-lated disks and the sarcolemma, providingeach with a complete system for excitation-contraction coupling.

To serve the tremendous metabolic de-mand placed on the heart and the need fora constant supply of high-energy phos-phates, the myocardial cell has an abundantconcentration of mitochondria. These or-ganelles are located between the individualmyofibrils and constitute approximately 35%of cell volume (see Fig. 1.8).

BASIC ELECTROPHYSIOLOGY

Rhythmic contraction of the heart relies onthe organized propagation of electrical im-pulses along its conduction pathway. Themarker of electrical stimulation, the actionpotential, is created by a sequence of ionfluxes through specific channels in the sarcolemma. To provide a basis for under-standing how electrical impulses lead to cardiac contraction, the process of cellulardepolarization and repolarization is re-viewed here. This material serves as an im-portant foundation for topics addressedlater in the book, including electrocardiog-raphy (see Chapter 4) and cardiac arrhyth-mias (see Chapters 11 and 12).

Cardiac cells capable of electrical excita-tion are of three electrophysiologic types,the properties of which have been studiedby intracellular microelectrode and patch-clamp recordings:

1. Pacemaker cells (e.g., SA node, AV node)2. Specialized rapidly conducting tissues

(e.g., Purkinje fibers)3. Ventricular and atrial muscle cells

The sarcolemma of each of these cardiaccell types is a phospholipid bilayer that islargely impermeable to ions. There are spe-cialized proteins interspersed throughoutthe membrane that serve as ion channels,cotransporters, and active transporters (Fig.1.10). These help to maintain ionic concen-tration gradients and charge differentials be-tween the inside and the outside of the car-diac cells. Note that normally, the Na+ andCa++ concentrations are much higher out-side the cell and the K+ concentration ismuch higher inside.

Ion Movement and Channels

The movement of specific ions across the cellmembrane serves as the basis of the actionpotential. Ion transport depends on twomajor factors: (1) the energetic favorabilityand (2) the permeability of the membranefor the ion.

Energetics

The two major forces that drive the energeticsof ion transport are the concentration gradientand the transmembrane potential (voltage).Molecules diffuse from areas of high concen-tration to areas of lower concentration—thegradient between these values determines therate of ion flow. For example, the extracellu-lar Na+ concentration is normally 145 mM,while the concentration inside the myocyte is15 mM. As a result, a strong force tends todrive Na+ into the cell, down its concentra-tion gradient.

The transmembrane potential of cells exerts an electrical force on ions (i.e., likecharges repel one another, and oppositecharges attract). The transmembrane po-tential of a myocyte at rest is about −90 mV (the inside of the cell is negative relative tothe outside). Extracellular Na+, a positivelycharged ion, is therefore attracted to the rel-atively negatively charged interior of the

12 Chapter One

Fig. 10

10090-01_CH01.qxd 8/24/06 5:40 PM


cell. Thus, there is a strong tendency for Na+

to enter the cell because of both the steepconcentration gradient and the electrical attraction.

Permeability

If there is such a strong force driving Na+

into the cell, what keeps this ion from actu-ally moving inside? The membrane of thecell at its resting potential is not permeableto sodium. The phospholipid bilayer of thecell membrane is composed of a hydropho-bic core that does not allow simple passageof charged, hydrophilic particles. Instead,permeability of the membrane is depen-dent on the opening of specific ion chan-nels, specialized proteins that span the cellmembrane and contain hydrophilic poresthrough which certain charged atoms canpass under specific circumstances.

Most types of ion channels share similarprotein sequences and structures, consistingof repeating transmembrane domains (Fig.1.11). Each of these domains contains sixmembrane-spanning segments. The fourthsegment (see S4 in Fig. 1.11) includes a se-quence of positively charged amino acids(lysine and arginine) that reacts to the mem-brane potential, and therefore that segmentis thought to confer voltage-sensitivity tothe channel, as described next.

The several types of cardiac ion chan-nels vary by two functional properties: selectivity and gating. Each type of chan-nel is normally selective for a specific ion,which is a manifestation of the size andstructure of its pore. For example, in cardiaccells, some channels permit the passage ofsodium ions, some are specific for potas-sium, and others allow only calcium to passthrough.

Ca++

Ca++

Ca++

Ca++

Ca++Na+

Na+

Na+

GAPJUNCTION

ATP

ATPATP

Internal

[Na+][K+][Cl–][Ca++ ]

15 mM150 mM

5 mM10–7 M

[Na+][K+][Cl–][Ca++ ]

145 mM5 mM

120 mM2 mM

External

K+

K+

A

D E F

B C

G

Figure 1.10. Ion channels, cotransporters, and active transporters of the myocyte. A. Sodium entry through thefast sodium channel is responsible for the rapid upstroke (phase 0) of the action potential (AP) in nonpacemaker cells.B. Calcium enters the cell through the slow calcium channel during phase 2 of the Purkinje fiber and muscle cell AP, andis the main channel responsible for depolarization of pacemaker cells. C. Potassium exits through a potassium channelto repolarize the cell during phase 3 of the AP, and open potassium channels help maintain the resting potential (phase4) of nonpacemaker cells. D. Sodium-calcium exchanger helps maintain the low intracellular calcium concentration. E. Sodium-potassium ATPase pump maintains concentration gradients for these ions. F, G. Active calcium transportersaid removal of calcium to the external environment and sarcoplasmic reticulum, respectively.

Fig. 11

10090-01_CH01.qxd 8/24/06 5:40 PM

14 Chapter One

C

INACTIVATION GATE

SELECTIVITYFILTER

+ +

+++

SODIUMCHANNEL

A

B

N

C

CALCIUMCHANNEL

POTASSIUMCHANNEL

Extracellular

Intracellular

N

C

N

N

N

S1 S2 S3 S4 S5 S6

N NC

C C

C

DOMAINI

DOMAINII

DOMAINIII

DOMAINIV

Figure 1.11. Structure of ion channels. A. Ion channels consist of glycosylated pro-teins arranged as repeating transmembrane domains. Each domain consists of sixmembrane-spanning segments. The potassium channel has four separate domains in atetrameric structure, while the sodium and calcium channels contain four domains covalently linked as a single unit. In the case of the sodium channel, the loop connect-ing domains III and IV is believed to serve as the channel’s inactivation gate. B. Enlarged view of a single domain of the sodium channel showing the six membrane-spanning segments. The S4 segment of each domain contains a sequence of positivelycharged amino acids, which confers the channel’s voltage sensitivity. The peptide loopsconnecting segments 5 and 6 in each domain form the selectivity filter for the chan-nel’s pore, which allows sodium, but not other ions, to pass through. (Parts A and Bare reproduced in part from Katz AM. Physiology of the Heart. 2nd Ed. New York:Raven Press, 1992:427, 429, with permission.) (Continued)

An ion can pass through its specific chan-nel only at certain times. That is, the ionchannel is gated—at any given moment, thechannel is either open or closed. The moretime a channel is in its open state, the largerthe number of ions that pass through it and

therefore, the greater the transmembranecurrent.

Cells contain a population of each type ofion channel, and each individual channelmay be in the open or closed state; it is thevoltage across the membrane that deter-

10090-01_CH01.qxd 8/24/06 5:40 PM


mines what fraction of these channels isopen at a given time. Therefore, the gatingof channels is said to be voltage sensitive.As the membrane voltage changes duringdepolarization and repolarization of thecell, specific channels open and close, withcorresponding alterations in the ion fluxesacross the sarcolemma.

An example of voltage-sensitive gating is apparent in the cardiac channel known as the fast sodium channel. The trans-membrane protein that forms this channelassumes various conformations depend-ing on the cell’s membrane potential (Fig.1.12). At a voltage of −90 mV (the typicalresting voltage of a ventricular muscle cell),the channels are primarily in a closed, rest-ing state, such that Na+ ions cannot pass

through. In this resting state, the channelsare available for conversion to the openconfiguration.

A rapid wave of depolarization causes themembrane potential to become less nega-tive and activates the resting channels tothe open state (see Fig. 1.12B). Na+ ions read-ily permeate through the open channels,and an inward Na+ current ensues. How-ever, the activated channels remain openfor only a brief time, a few thousandths of a second, and then spontaneously close to create an inactive state (see Fig. 1.12C).Channels in the inactivated conformationcannot be directly converted back to theopen state.

The inactivated state persists until themembrane voltage has repolarized nearly

Voltagesensor

Selectivityfilter (pore)

III

12

5

3

6

IV II

IInactivationgate (closed)

Inactivationgate (open)

4

Figure 1.11. (Continued ) Structure of ion channels. C. A 3-dimensional schematicof the sodium channel, showing how the four domains wrap around the channel’spore. The selectivity filter formed by the loops connecting segments 5 and 6 is shownnear the extracellular opening of the channel, while the inactivation gate (the loopbetween domains III and IV) is displayed on the cytosolic side. (Reproduced from Nelson CL, Cox MM. Lehninger’s Principles of Biochemistry. 3rd Ed. New York: Worth,2000:428, with permission.)

Fig. 12

AQ4

C

10090-01_CH01.qxd 8/24/06 5:40 PM

back to its original resting level. Until it doesso, the inactivated channel prevents any flowof sodium ions. Thus, during normal cellulardepolarization, the voltage-dependent fastsodium channels conduct for a short periodand then inactivate, unable to conduct cur-

rent again until the cell membrane hasnearly fully repolarized and the channels re-cover from the inactivated to the closed rest-ing state.

Another important attribute of cardiacfast sodium channels should be noted. If the

16 Chapter One

Activationgate

Rapid depolarization

Repolarization

Spont

aneo

us

Inactivation

Outside

A

C

CHANNEL CLOSED(RESTING)

CHANNEL CLOSED(INACTIVE)

Na+

Cellmembrane

Inside

+ + + + + + + +

– – – – – – – –

III

IV II

B

CHANNEL OPEN

Na+

Na+

+ + + + + + + +

– – – – – – – –

+ + + + + + + +

– – – – – – – –

Figure 1.12. Schematic representation of gating of fast sodium channels. A. The four covalently linked trans-membrane domains (I, II, III, IV) form the sodium channel, which is guarded by activation and inactivation gates. (Here,domain I is cut away to show the transmembrane pore.) In the resting membrane, most channels are in a closed state.Even though the inactivation gate is open, Na+ ions cannot easily pass through because the activation gate is closed.B. A rapid depolarization changes the cell membrane voltage and forces the activation gate to open, presumably mediated by translocation of the charged portions of segment 4 in each domain. With the channel in this con-formation (in which both the activation and inactivation gates are open), Na+ ions permeate into the cell. C. As the inactivation gate spontaneously and quickly closes, the sodium current ceases. The inactivation gating func-tion is thought to be achieved by the peptide loop that connects domains III and IV and swings into the intracellularopening of the channel pore (black arrow). The channel cannot reopen directly from this closed, inactive state. Cellu-lar repolarization returns the channel to the resting condition (A). During repolarization, as high negative membranevoltages are reachieved, the activation gate closes and the inactivation gate reopens.

10090-01_CH01.qxd 8/24/06 5:40 PM


transmembrane voltage of a cardiac cell isslowly depolarized and maintained chroni-cally at levels less negative than the usualresting potential, inactivation of channelsoccurs without initial opening and currentflow. Furthermore, as long as this partial de-polarization exists, the closed, inactive chan-nels cannot recover to the resting state.Thus, the fast sodium channels in such a cellare persistently unable to conduct Na+ ions.This is the typical case in cardiac pacemakercells (e.g., the SA and AV nodes) in which themembrane voltage is generally less negativethan −70 mV throughout the cardiac cycle.As a result, the fast sodium channels in pace-maker cells are persistently inactivated anddo not play a role in the generation of the ac-tion potential in these cells (see Box 1.1).

Calcium and potassium channels in cardiaccells also act in voltage-dependent fashions,but they behave differently than the sodiumchannels, as described in the next section.

Resting Potential

In cardiac cells at rest, prior to excitation,the electrical charge differential between the inside and outside of a cell is known as the resting potential. The magnitude ofthe resting potential of a cell depends on twomain properties: (1) the concentration gra-dients for all the different ions between theinside and outside of the cell, and (2) whichion channels are open at rest.

As in other tissues such as nerve cells andskeletal muscle, the potassium concentra-

Box 1.1 Mechanism of Fast Sodium Channels

A key characteristic of fast sodium channels is their ability to activate and then inactivaterapidly when the cell is depolarized. The mechanism by which this occurs has been inves-tigated for many decades. In the mid-1900s, Hodgkin and Huxley studied the action po-tential in giant squid axons (J Physiol [Lond] 1952; 117:500–544). They found that ionchannels act as if they contain a series of “gates” that open and close in a specific patternwhen the membrane potential is altered. In the case of the sodium channel, the re-searchers postulated the presence of m gates closed in the resting state and an h gateopen in the resting state. Depolarization of the membrane causes the m gates to openquickly, which allows Na+ ions to pass through the channel (equivalent to the open chan-nel in Fig. 1.12B). However, that same depolarization of the cell also causes the h gate toclose, which blocks the passage of sodium ions (the closed, inactive state in Fig. 1.12C).Na+ can flow through the channel only when both sets of gates are open. Since the mgates open faster than the h gate closes, there is a brief period (about 1 millisecond) dur-ing which Na+ can pass through. After the membrane repolarizes to voltages more nega-tive than about −60 mV, the m gates shut, the h gate reopens, and the channel returnsto the closed, resting state (see Fig. 1.12A), available for activation once again.

More recent research has demonstrated that ion channel activity is actually more com-plex than suggested by this model, but there are important correlates with current mo-lecular concepts. For example, the cluster of positively charged amino acids on segment 4 (S4) of the ion channel domain (see Fig. 1.11) is believed to be the voltage sensor for them gates that cause the channel to open during depolarization. In the resting state, thestrong positive charge on S4 causes it to be pulled inward toward the negative membranepotential. During depolarization, as the membrane charge becomes less, S4 can move out-ward, resulting in a conformational change in the protein that results in channel opening.Inactivation (the h gate) is thought to be achieved by the peptide loop that connects do-mains III and IV of the sodium channel (see Figs. 1.11 and 1.12) and swings into and oc-cludes the channel during depolarization.

Box 1.1

AQ1

10090-01_CH01.qxd 8/24/06 5:40 PM

tion is much greater inside cardiac cellscompared with outside the cells. This is at-tributed to cell membrane transporters, themost important of which is Na+K+-ATPase.This protein “pump” couples the energy of ATP hydrolysis to export three Na+ ionsout of the cell in exchange for the inwardmovement of two K+ ions. This acts to main-tain intracellular Na+ at low levels and intra-cellular K+ at high levels.

Cardiac myocytes contain potassiumchannels that are open in the resting state,at a time when other ionic channels (i.e.,sodium and calcium) are closed. Therefore,the resting cell membrane is much morepermeable to potassium than to other ions.As a result of its open channels at rest, K+

fluxes in an outward direction down its concentration gradient, removing positivecharges from the cell. The predominantcounter ions for potassium within the cellare large negatively charged proteins thatare unable to diffuse outward along with K+.Thus, as potassium ions exit the cell, the an-ions that are left behind cause the interior ofthe cell to become electrically negative withrespect to the outside.

However, as the interior of the cell be-comes more and more negatively chargedby the outward flux of potassium, the pos-itively charged K+ ions are attracted back to-

ward the cellular interior, an effect thatslows their net exit from the cell. Thus, thetwo opposing forces directing the flux ofpotassium ions across their open channelsin the resting state are (1) the concentra-tion gradient, which favors outward pas-sage of potassium, and (2) the electrostaticforce, which attracts potassium back intothe cell (Fig. 1.13). At steady state, theequilibrium between these chemical andelectrical forces determines the resting potential, which is approximately −90 mVin ventricular muscle cells. This is pre-dicted by the equilibrium potential (Nernstpotential) for potassium, as shown in Fig-ure 1.13.

The permeability of the cardiac myocytecellular membrane for sodium is minimal in the resting state because the channelsthat conduct that ion are essentially closed.However, there is a slight leak of sodiumions into the cell at rest. This tiny inwardcurrent of positively charged ions explainswhy the actual resting potential is slightlyless negative than would be predicted if thecell membrane were truly only permeable topotassium. The sodium ions that slowly leakinto the myocyte at rest (and the muchlarger amount that enters during the actionpotential) are continuously removed fromthe cell and returned to the extracellular en-

18 Chapter One

Equilibrium (Nernst) potential = –26.7 ln ([K+] in/[K+]out) = –91mV

Inside cell

Openpotassiumchannels

CONCENTRATIONGRADIENT

[K+]out(5 mM)

ELECTRICALFORCE

K+++

++

––

––

[K+] in(150 mM)

Figure 1.13. The resting potential of a cardiac muscle cell is deter-mined by the balance between the concentration gradient andelectrostatic forces for potassium, because only potassium chan-nels are open at rest. The concentration gradient favors outward move-ment of K+, whereas the electrical force attracts the positively charged K+

ions inward. The equilibrium (resting) potential can be approximated bythe Nernst equation for potassium, as shown here.

Fig. 13

10090-01_CH01.qxd 8/24/06 5:40 PM


vironment by Na+K+-ATPase, as previouslydescribed.

Action Potential

When the cell membrane voltage is altered,its permeability to specific ions changes be-cause of the voltage-gating characteristics ofthe ion channels. Each type of channel hasa characteristic pattern of activation and in-activation that determines the progressionof the electrical signal. This discussion be-gins by following the development of theaction potential in a typical cardiac musclecell (Fig. 1.14). The unique characteristics ofaction potentials in cardiac pacemaker cellsare described next.

Cardiac Muscle Cell

Until stimulated, the resting potential of acardiac muscle cell remains stable, at ap-proximately −90 mV. This resting state be-fore depolarization is known as phase 4 ofthe action potential. Following phase 4, fouradditional phases characterize depolariza-tion and repolarization of the cell (see Fig.1.14).

Phase 0

At the resting membrane voltage, sodiumand calcium channels are closed. Any pro-cess that makes the membrane potential lessnegative than the resting value causes somesodium channels to open. As these channelsopen, sodium ions rapidly enter the cell,flowing down their concentration gradient,and toward the negatively charged cellularinterior. The entry of Na+ ions into the cellcauses the transmembrane potential to be-come progressively less negative, which inturn causes more sodium channels to openand promotes further sodium entry into the cell. When the membrane voltage ap-proaches the threshold potential (approxi-mately −70 mV in cardiac muscle cells),enough of these fast Na+ channels haveopened to generate a self-sustaining inwardNa+ current. The magnitude of entry of pos-itively charged Na+ ions neutralizes the

membrane potential to zero and transientlyinto the positive range.

The prominent influx of sodium ions isresponsible for the rapid upstroke, or phase0, of the action potential. However, the Na+

channels remain open for only a few thou-sandths of a second and are then quickly in-activated, preventing further influx (see Fig.1.14). Thus, while activation of these fastNa+ channels causes the rapid early depolar-ization of the cell, the rapid inactivation

Mem

bran

e po

tent

ial (

mV

)

0

0

Na+

influx

Ca++ influx(and K+ efflux)

K+

efflux

Time

–50

Inwardsodiumcurrent

0

Inwardcalciumcurrent

0

Outwardpotassium

current

–100

1

2

0

4

3

Figure 1.14. Schematic representation of a my-ocyte action potential (AP) and relative net ioncurrents for Na1, Ca11, and K1. The resting po-tential is represented by phase 4 of the AP. Follow-ing depolarization, Na+ influx results in the rapid up-stroke of phase 0; a transient outward potassiumcurrent is responsible for partial repolarization dur-ing phase 1; slow Ca++ influx (and relatively low K+

efflux) results in the plateau of phase 2; and finalrapid repolarization largely results from K+ effluxduring phase 3.

Fig. 14

10090-01_CH01.qxd 8/24/06 5:40 PM

makes their major contribution to the ac-tion potential short lived.

Phase 1

Following rapid phase 0 depolarization intothe positive voltage range, a brief current ofrepolarization returns the membrane poten-tial to approximately 0 mV. The responsiblecurrent is carried by the outward flow of K+

ions through a type of transiently activatedpotassium channel.

Phase 2

This relatively long phase of the action po-tential is mediated by the balance of an out-ward K+ current in competition with an inward Ca++ current, which flows throughspecific L-type calcium channels. The latterchannels begin to open during phase 0,when the membrane voltage reaches ap-proximately −40 mV, allowing Ca++ to flowdown their concentration gradient into thecell. Ca++ entry proceeds in a more gradualfashion than the initial influx of sodium, be-cause with calcium channels, activation isslower and the channels remain open muchlonger compared with sodium channels (seeFig. 1.14). During this phase, the Ca++ influx,and the relatively low permeability to K+ ef-flux, maintains a voltage of approximately 0 mV for a prolonged period known as theplateau. Calcium ions that enter the cellduring this phase play a critical role in trig-gering additional internal calcium releasefrom the sarcoplasmic reticulum, which isimportant in initiating myocyte contrac-tion, as discussed later in the chapter. As theCa++ channels gradually inactivate and theefflux of K+ begins to exceed the influx ofcalcium, phase 3 begins.

Phase 3

This is the final phase of repolarization thatreturns the transmembrane voltage back tothe resting potential of approximately −90mV. A continued outward potassium cur-rent and low membrane permeability forother cations are responsible for this period

of rapid repolarization. Phase 3 completesthe action potential cycle, with a return toresting phase 4, preparing the cell for thenext stimulus for depolarization.

To preserve normal transmembrane ionicconcentration gradients, sodium and cal-cium ions that enter the cell during depo-larization must be returned to the extra-cellular environment, and potassium ionsmust return to the cell interior. As shown inFigure 1.10, Ca++ ions are removed by thesarcolemmal Na+-Ca++ exchanger and to alesser extent by the ATP-consuming calciumpump (sarcolemmal Ca++-ATPase). The cor-rective exchange of Na+ and K+ across thecell membrane is mediated by Na+K+-AT-Pase, as described earlier.

Specialized Conduction System

The process described in the previous sec-tions applies to the action potential of car-diac muscle cells. The cells of the specializedconduction system (e.g., Purkinje fibers) be-have similarly, although the resting poten-tial is slightly more negative and the up-stroke of phase 0 is even more rapid.

Pacemaker Cells

The upstroke of the action potential of car-diac muscle cells does not normally occurspontaneously. Rather, when a wave of depolarization reaches the myocyte fromneighboring cells, its membrane potentialbecomes less negative and an action poten-tial is triggered.

Certain heart cells do not require externalprovocation to initiate their action poten-tial. Rather, they are capable of self-initiateddepolarization in a rhythmic fashion andare known as pacemaker cells. They are en-dowed with the property of automaticity,by which the cells undergo spontaneous de-polarization during phase 4. When thethreshold voltage is reached in such cells,the action potential upstroke is triggered(Fig. 1.15).

Cells that display pacemaker behavior in-clude the SA node (the “natural pacemaker”of the heart) and the AV node. Although

20 Chapter One

Fig. 15

10090-01_CH01.qxd 8/24/06 5:40 PM


atrial and ventricular muscle cells do not nor-mally display automaticity, they may do sounder disease conditions such as ischemia.

The shape of the action potential of a pace-maker cell is different from that of a ventric-ular muscle cell in three ways:

1. The maximum negative voltage of pace-maker cells is approximately −60 mV,substantially less negative than the rest-ing potential of ventricular muscle cells(−90 mV). The persistently less negativemembrane voltage of pacemaker cellscauses the fast sodium channels withinthese cells to remain inactivated.

2. Unlike that of cardiac muscle cells,phase 4 of the pacemaker cell action potential is not flat but has an upwardslope, representing spontaneous grad-ual depolarization. This spontaneousdepolarization is the result of an ionicflux known as the pacemaker current(also called the “funny current” and de-noted If). Current evidence indicatesthat the pacemaker current is carriedpredominantly by Na+ ions. The ionchannel through which the pacemakercurrent passes is different from the fastsodium channel responsible for phase 0

of the action potential. Rather, thispacemaker channel opens during repo-larization of the cell, as the membranepotential approaches its most negativevalues. The inward flow of positivelycharged Na+ ions through the pace-maker channel causes the membranepotential to become progressively lessnegative during phase 4, ultimately depolarizing the cell to its thresholdvoltage (see Fig 1.15.)

3. The phase 0 upstroke of the pacemakercell action potential is less rapid andreaches a lower amplitude than that of acardiac muscle cell. These characteristicsresult from the fast sodium channels ofthe pacemaker cells being inactivatedand the upstroke of the action potentialrelying solely on Ca++ influx through therelatively slow calcium channels.

Repolarization of pacemaker cells occursin a fashion similar to that of ventricularmuscle cells and relies on inactivation of thecalcium channels and increased activationof potassium channels with enhanced K+

efflux from the cell.

Refractory Periods

Compared with electrical impulses in nervesand skeletal muscle, the cardiac action po-tential is much longer in duration. This re-sults in a prolonged refractory period duringwhich the muscle cannot be restimulated.Such a long period is physiologically neces-sary because it allows the ventricles suffi-cient time to empty their contents and refillbefore the next contraction.

There are different levels of refractorinessduring the action potential, as illustrated inFigure 1.16. The degree of refractoriness pri-marily reflects the number of fast Na+ chan-nels that have recovered from their inactivestate and are capable of reopening. As phase3 of the action potential progresses, an in-creasing number of Na+ channels recover andcan respond to the next depolarization. This,in turn, corresponds to an increasing proba-bility that a stimulus will trigger an actionpotential and result in a propagated impulse.

Mem

bran

e po

tent

ial (

mV

) 0

K+

efflux

Ca2+

influx

lf

Time

–40

–80

4

0

Figure 1.15. Action potential of a pace-maker cell. Phase 4 is characterized bygradual, spontaneous depolarization owingto the pacemaker current (If). When thethreshold potential is reached, at about −40 mV, the upstroke of the action poten-tial follows. The upstroke of phase 0 is lessrapid than in nonpacemaker cells becausethe current represents Ca++ influx throughthe relatively slow calcium channels.

Fig. 16

10090-01_CH01.qxd 8/24/06 5:40 PM

The absolute refractory period refers tothe time during which the cell is completelyunexcitable to a new stimulation. The effec-tive refractory period includes the absoluterefractory period but extends beyond it toinclude a short interval of phase 3, duringwhich stimulation produces a localized ac-tion potential that is not strong enough topropagate further. The relative refractory pe-riod is the interval during which stimula-tion triggers an action potential that is con-ducted, but because the cell is stimulatedfrom a voltage less negative than the restingpotential, its upstroke is less steep and oflower amplitude and its conduction velocityslower than normal (as described in the nextsection). Following the relative refractoryperiod, a short “supranormal” period is pres-ent in which a less-than-normal stimuluscan trigger an action potential. The refrac-tory period of atrial cells is shorter than thatof ventricular muscle cells, such that atrialrates can generally exceed ventricular ratesduring rapid arrhythmias (see Chapter 11).

Impulse Conduction

During depolarization, the electrical impulsespreads along each cardiac cell, and rapidly

from cell to cell, because each myocyte isconnected to its neighbors through low-resistance gap junctions. The speed of tissuedepolarization (phase 0) and the conductionvelocity along the cell depend on the num-ber of sodium channels and on the magni-tude of the resting potential. Tissues with ahigh concentration of Na+ channels, such asPurkinje fibers, have a large, fast inward cur-rent, which spreads rapidly within and be-tween cells to support rapid conduction. Incontrast, the less negative the resting poten-tial, the greater the number of inactivatedfast sodium channels, and therefore the lessrapid the upstroke velocity (Fig. 1.17). Thus,alterations in the resting potential greatly af-fect the upstroke and conduction velocity ofthe action potential.

Normal Sequence of Cardiac Depolarization

Electrical activation of the heartbeat is nor-mally initiated at the SA node (see Fig. 1.6).The impulse spreads to the surroundingatrial muscle through intercellular gap junc-tions that provide electrical continuity be-tween the cells. Ordinary atrial muscle fibersparticipate in the propagation of the im-

22 Chapter One

Mem

bran

e po

tent

ial (

mV

)

0

AbsoluteRP

EffectiveRP

RelativeRP

Supranormalperiod

–50

–100

1

2

3

Figure 1.16. Refractory periods (RPs) of the myocyte. During the absolute re-fractory period (ARP), the cell is unexcitable to stimulation. The effective refractoryperiod includes a brief time beyond the ARP during which stimulation produces alocalized depolarization that does not propagate (curve 1). During the relative re-fractory period, stimulation produces a weak action potential (AP) that propagates,but more slowly than usual (curve 2). During the supranormal period, a weaker-than-normal stimulus can trigger an AP (curve 3).

Fig. 17

10090-01_CH01.qxd 8/24/06 5:40 PM


pulse from the SA to the AV node, althoughin certain regions the fibers are more denselyarranged, facilitating conduction.

Fibrous tissue surrounds the tricuspidand mitral valves, such that there is no direct electrical connection between theatrial and ventricular chambers other thanthrough the AV node. As the electrical impulse reaches the AV node, a delay inconduction (approximately 0.1 sec) is en-countered. This delay occurs because thesmall-diameter fibers in this region conductslowly, and the action potential is of the“slow” pacemaker type (recall that the fastsodium channels are permanently inacti-vated in pacemaker tissues, such that theupstroke velocity relies on the slower cal-cium channels). The pause in conduction atthe AV node is actually beneficial because itallows the atria time to contract and fullyempty their contents before ventricularstimulation. In addition, the delay allowsthe AV node to serve as a “gatekeeper” ofconduction from atria to ventricles, whichis critical for limiting the rate of ventricularstimulation during abnormally rapid atrialrhythms.

After traversing the AV node, the cardiacaction potential spreads into the rapidlyconducting bundle of His and Purkinjefibers, which distribute the electrical im-pulses to the bulk of the ventricular musclecells. This allows for precisely timed stimu-lation and contraction of the ventricularmyocytes.

EXCITATION-CONTRACTIONCOUPLING

This section reviews how the electrical action potential leads to physical contrac-tion of cardiac muscle cells, a process known as excitation-contraction coupling. Duringthis process, chemical energy in the form ofhigh-energy phosphate compounds is trans-lated into the mechanical energy of myo-cyte contraction.

Contractile Proteins in the Myocyte

Several distinct proteins are responsible forcardiac muscle cell contraction (Fig. 1.18).Two of the proteins, actin and myosin, arethe chief contractile elements. Two otherproteins, tropomyosin and troponin, serveregulatory functions.

Myosin is arranged in thick filaments,each composed of lengthwise stacks of ap-proximately 300 molecules. The myosin filament exhibits globular heads that areevenly spaced along its length and containmyosin ATPase, an enzyme that is necessaryfor contraction to occur. Actin, a smallermolecule, is arranged in thin filaments as anα-helix consisting of two strands that inter-digitate between the thick myosin filaments(see Fig. 1.8). Titin (also termed connectin)is a protein that helps tether myosin to theZ line of the sarcomere and provides elastic-ity to the contractile process.

Phase0

Mem

bran

e po

tent

ial (

mV

)

0

–50

–100

AB

Figure 1.17. Dependence of speed of depolarization on resting poten-tial. A. Normal resting potential (RP) and rapid rise of phase 0. B. Less negativeRP results in slower rise of phase 0 and lower maximum amplitude of the ac-tion potential.

Fig. 18

10090-01_CH01.qxd 8/24/06 5:40 PM

Tropomyosin is a double helix that liesin the grooves between the actin filamentsand, in the resting state, inhibits the inter-action between myosin heads and actin,thus preventing contraction. Troponin sitsat regular intervals along the actin strandsand is composed of three subunits. The tro-ponin T (TnT) subunit links the troponincomplex to the actin and tropomyosin mol-ecules. The troponin I (TnI) subunit inhibitsthe ATPase activity of the actin-myosin in-teraction. The troponin C (TnC) subunit isresponsible for binding calcium ions thatregulate the contractile process.

Calcium-Induced Calcium Releaseand the Contractile Cycle

The sensitivity of troponin C to calcium es-tablishes a crucial role for intracellular Ca++

ions in cellular contraction. The cycling ofcalcium in and out of the cytosol duringeach action potential effectively couples elec-trical excitation to physical contraction.

Recall that during phase 2 of the actionpotential, activation of L-type Ca++ chan-nels results in an influx of Ca++ ions into themyocyte. The small amount of calcium thatenters the cell in this fashion is not suffi-cient to cause contraction of the myofibrils,but it triggers a much greater Ca++ releasefrom the sarcoplasmic reticulum (SR), as follows: The T tubule invaginations of thesarcolemmal membrane bring the L-typechannels into close apposition with special-ized Ca++ release receptors in the SR, known

as ryanodine receptors (Fig. 1.19). Whencalcium enters the cell and binds to theryanodine receptor, the receptor undergoesa conformational change, which results in a much greater release of Ca++ into the cytosol from the abundant stores in theterminal cisternae of the SR. Thus, the ini-tial L-type Ca++ current signal is amplifiedby this mechanism, known as calcium-induced calcium release (CICR), and thecytosolic calcium concentration dramati-cally increases.

As calcium ions bind to TnC, the activityof TnI is inhibited, which induces a confor-mational change in tropomyosin. The latterevent exposes the active site between actinand myosin, enabling contraction to proceed.

Contraction ensues as myosin headsbind to actin filaments and “flex,” thuscausing the interdigitating thick and thinfilaments to move past each other in anATP-dependent reaction (Fig. 1.20). Thefirst step in this process is activation of the myosin head by hydrolysis of ATP, fol-lowing which the myosin head binds toactin and forms a cross bridge. The inter-action between the myosin head and actinresults in a conformational change in thehead, causing it to pull the actin filamentinward.

Next, while the myosin head and actinare still attached, ADP is released, and a newmolecule of ATP then binds to the myosinhead, causing it to release the actin filament.The cycle can then repeat. Progressive cou-pling and uncoupling of actin and myosin

24 Chapter One

Myosin thick filament

Tropomyosin

Actin

Tn-T

Tn-CTn-I

Myosinhead

Figure 1.18. Schematic diagram of the main contractile proteins of themyocyte, actin and myosin. Tropomyosin and troponin (components TnI,TnC, and TnT) are regulatory proteins.

AQ2

Fig. 19

Fig. 20

10090-01_CH01.qxd 8/24/06 5:40 PM


causes the muscle fiber to shorten by in-creasing the overlap between the myofila-ments within each sarcomere. In the pres-ence of ATP, this process continues for aslong as the cytosolic calcium concentrationremains sufficiently high to inhibit the troponin-tropomyosin blocking action.

Myocyte relaxation, like contraction, issynchronized with the electrical activity ofthe cell. Toward the end of phase 2 of the ac-tion potential, L-type channels inactivate, ar-resting the influx of Ca++ into the cell andabolishing the trigger for CICR. Concur-rently, calcium is pumped back into the SRand out of the cell. Calcium is sequesteredback into the SR primarily by sarco(endo)-plasmic reticulum Ca11 ATPase (SERCA), asshown in Figure 1.19. The small amount ofCa++ that entered the cell through L-type cal-cium channels is removed via the sarcolem-mal Na+-Ca++ exchanger and to a lesser extent

by the ATP-consuming calcium pump, Ca++-ATPase (see Fig. 1.10).

As cytosolic Ca++ concentrations fall andcalcium ions dissociate from troponin C,tropomyosin once again inhibits the actin-myosin interaction, leading to relaxationof the contracted cell. The contraction-relaxation cycle can then repeat with thenext action potential.

b-Adrenergic and Cholinergic Signaling

There is substantial evidence that the con-centration of Ca++ within the cytosol is themajor determinant of the force of cardiaccontraction with each heartbeat. Mecha-nisms that raise intracellular Ca++ concentra-tion enhance force development, whereasfactors that lower Ca++ concentration reducethe contractile force.

Outside cell

Inside cell

PL

Binds toTn-C

Contraction

Ca++

Ca++

Ca++

Ca++

Ca++

SERCASarcoplasmic

reticulum

Ryanodinereceptor

+

Ca++

Na+

ATP

ATP

Figure 1.19. Calcium ion movements during excitation and contraction incardiac muscle cells. Ca++ enters the cell through calcium channels duringphase 2 of the action potential, triggering a much larger calcium release fromthe sarcoplasmic reticulum (SR) via the ryanodine receptor complex. The bindingof cytosolic Ca++ to troponin C (TnC) allows contraction to ensue. Relaxationoccurs as Ca++ is returned to the SR by sarco(endo)plasmic reticulum calciumATPase (SERCA). Phospholamban (PL) is a major regulator of this pump, inhibit-ing Ca++ uptake in its dephosphorylated state. Excess intracellular calcium is re-turned to the extracellular environment by sodium-calcium exchange and to asmaller degree by the sarcolemmal Ca++-ATPase.

AQ3

10090-01_CH01.qxd 8/24/06 5:40 PM

β-Adrenergic stimulation is one mechanismthat enhances calcium fluxes in the myocyteand thereby strengthens the force of ventric-ular contraction (Fig. 1.21). Catecholamines(e.g., norepinephrine) bind to the myocyteβ1-adrenergic receptor, which is coupled toand activates the G protein system (Gs) at-tached to the inner surface of the cell mem-brane. Gs in turn stimulates membrane-bound adenylate cyclase to produce cyclicAMP (cAMP) from ATP. cAMP then activatesintracellular protein kinases, which phos-phorylate cellular proteins, including the L-type calcium channels within the cellmembrane. Phosphorylation of the calciumchannel augments Ca++ influx, which trig-gers a corresponding increase in Ca++ releasefrom the sarcoplasmic reticulum, therebyenhancing the force of contraction.

β-Adrenergic stimulation of the myocytealso enhances myocyte relaxation. The re-turn of Ca++ from the cytosol to the sar-coplasmic reticulum (SR) is regulated byphospholamban (PL), a low molecularweight protein in the SR membrane. In itsdephosphorylated state, PL inhibits Ca++

uptake by SERCA (see Fig. 1.19). However,β-adrenergic activation of protein kinases(see Fig. 1.21) causes PL to become phos-phorylated, an action that blunts PL’s in-hibitory effect. The subsequently greateruptake of calcium ions by the SR hastensCa++ removal from the cytosol, promotingmyocyte relaxation. The increased cAMPactivity also results in phosphorylation ofTnI, an action that inhibits actin-myosininteractions and therefore further enhancesrelaxation of the cell.

26 Chapter One

ATP-ATP

-ADP-Pi

-ADP-Pi

-ADP

Pi

Actin

MyosinADP

A. Activation of myosin headby ATP hydrolysis

D. ADP release, ATP binding,actin filament release

B. Cross bridge formationbetween myosin headand actin filament

C. Phosphate releaseand power stroke

Figure 1.20. The contractile process. A. Myosin head is activated by hydrolysis of ATP. B. During cellular depolarization, cytoplasmic calcium concentration increases and removes thetroponin-tropomyosin inhibition, such that a cross bridge is formed between actin and myosin.C. Inorganic phosphate (Pi), is released and a conformational change in the myosin head drawsthe actin filament inward. D. ADP is released and replaced by ATP, causing the myosin head todissociate from the actin filament. As the process repeats, the muscle fiber shortens. The cyclecontinues until cytosolic calcium concentration decreases at the end of phase 2 of the action potential.

Fig. 21

10090-01_CH01.qxd 8/24/06 5:40 PM


Cholinergic signaling via parasympatheticinputs (mainly from the vagus nerve) op-poses the effects of β-adrenergic stimulation(see Fig. 1.21). Acetylcholine released fromparasympathetic nerve terminals binds to themuscarinic M2 receptor on cardiac cells. Thisreceptor also activates G proteins, but in dis-tinction to the β-adrenergic receptor, it iscoupled to Gi, an inhibitory G protein system.Gi associated with cholinergic stimulation in-hibits adenylate cyclase activity and reducescAMP formation. Gi also activates specific K+

channels in the plasma membrane, whichhyperpolarizes the cell. At the sinus node,these actions of cholinergic stimulation serveto reduce heart rate. In the myocardium, theeffect is to counteract the force of contractioninduced by β-adrenergic stimulation. Itshould be noted that ventricular cells aremuch less sensitive to this cholinergic effectthan atrial cells, likely reflecting different de-grees of G protein coupling.

Thus, physiologic or pharmacologic cate-cholamine stimulation of the myocyte β1-adrenergic receptor enhances contraction ofthe cell, while cholinergic stimulation op-poses that enhancement. These importantproperties will be referred to in later chapters.

SUMMARY

The anatomic structure, cellular composi-tion, and conduction pathways of the heartform an efficient system for repetitive, or-ganized contractions. As a result, the heartis capable of purposeful stimulation bil-lions of times during the lifespan of a nor-mal person. With each contraction cycle,the heart receives and propagates bloodthrough the circulation to provide nutri-ents to and remove waste products fromthe body’s tissues.

The following chapters explore what cango wrong with this remarkable system.

Norepinephrine

Gs protein G1 protein

Ca++

Ca++

Ca++

ATP

Inactiveproteinkinases

Activeproteinkinases PL

ATP

Adenylatecyclase

Muscarinicreceptor

Sarcoplasmicreticulum

cAMP

β1-andrenergicreceptor

Acetylcholine

+

+

+ –

Figure 1.21. Effects of b-adrenergic and cholinergic stimulation on cardiac cellularsignaling and calcium ion movement. The binding of a ligand (e.g., norepinephrine) tothe β1-adrenergic receptor induces G protein–mediated stimulation of adenylate cyclaseand formation of cyclic AMP (cAMP). The latter activates protein kinases, which phosphor-ylate cellular proteins, including ion channels. Phosphorylation of the slow Ca++ channelenhances calcium movement into the cell and therefore strengthens the force of contrac-tion. Protein kinases also phosphorylate phospholamban (PL), reducing the latter’s inhibi-tion of Ca++ uptake by the sarcoplasmic reticulum. The enhanced removal of Ca++ from thecytosol facilitates relaxation of the myocyte. Cholinergic signaling, triggered by acetyl-choline binding to the muscarinic receptor, activates inhibitory G proteins that reduceadenylate cyclase activity and cAMP production, thus antagonizing the effects of β-adrenergic stimulation.

10090-01_CH01.qxd 8/24/06 5:40 PM

Acknowledgments

Contributors to the previous editions of this chap-ter were Kirsten Greineder, MD; Stephanie Harper,MD; Scott Hyver, MD; Paul Kim, MD; Rajeev Malhotra, MD; Laurence Rhines, MD; James D.Marsh, MD; Gary R. Strichartz, MD; and Leonard S.Lilly, MD.

Additional Reading

Bers DM. Cardiac excitation-contraction coupling.Nature 2002;415:198–205.

Katz AM. Physiology of the Heart. 4th Ed. Philadel-phia: Lippincott Williams & Wilkins, 2006.

Lohse MJ, Engelhardt S, Eschenhagen T. What is therole of beta-adrenergic signaling in heart failure?Circ Res 2003;93:896–906.

Opie LH. Heart Physiology, from Cell to Circulation.4th Ed. Philadelphia: Lippincott Williams &Wilkins, 2004.

Rockman HA, Koch WJ, Lefkowitz RJ. Seven-transmembrane-spanning receptors and heartfunction. Nature 2002;415:206.

Wilcox BR, Cook AC, Anderson RH. Surgical Anatomyof the Heart. Cambridge, MA: Cambridge Univer-sity Press, 2005.

Zipes DP, Jalife J, eds. Cardiac Electrophysiology:From Cell to Bedside. 4th Ed. Philadelphia: Saun-ders, 2004.

28 Chapter One

10090-01_CH01.qxd 8/24/06 5:40 PM

Chapter 1—Author Queries2. AU: Abbreviations changed per several Internet sources and Chapter 7.3. AU: As on p. 22. Correct?4. AU: ED: Permission still pending?

Chapter 1(Box)—Author Query1. AU: Plural correct, as above?

10090-01_CH01.qxd 8/24/06 5:40 PM Page a

CARDIAC CYCLE

HEART SOUNDSFirst Heart Sound (S1)Second Heart Sound (S2)Extra Systolic Heart SoundsExtra Diastolic Heart Sounds

MURMURSSystolic MurmursDiastolic MurmursContinuous Murmurs

29

C H A P T E R

2The Cardiac Cycle:Mechanisms of HeartSounds and MurmursNicole MartinLeonard S. Lilly

Cardiac diseases often cause abnormal findings on physical examination, includ-ing pathologic heart sounds and murmurs.These findings are clues to the underlyingpathophysiology, and proper interpreta-tion is essential for successful diagnosis and disease management. This chapter de-scribes heart sounds in the context of thenormal cardiac cycle and then focuses onthe origins of pathologic heart sounds andmurmurs.

Many cardiac diseases are mentionedbriefly in this chapter as examples of abnor-mal heart sounds and murmurs. Becauseeach of these conditions is described ingreater detail later in the book, it is not nec-essary to memorize the examples presentedhere. Rather, it is preferable to understandthe mechanisms by which the abnormalsounds are produced, so that their descrip-tions will make sense in later chapters.

CARDIAC CYCLE

The cardiac cycle consists of precisely timedelectrical and mechanical events that are re-sponsible for rhythmic atrial and ventricu-lar contractions. Figure 2.1 displays the pres-sure relationships between the left-sidedcardiac chambers during the normal cardiaccycle and serves as a platform for describingkey events. Mechanical systole refers to ven-tricular contraction, and diastole to ven-tricular relaxation and filling. Throughoutthe cardiac cycle, the right and left atria accept blood returning to the heart from the systemic veins and from the pulmonaryveins, respectively. During diastole, bloodpasses from the atria into the ventriclesacross the open tricuspid and mitral valves,causing a gradual increase in ventricular di-astolic pressures. In late diastole, atrial con-traction propels a final bolus of blood into

Fig. 1

10090-02_CH02.qxd 8/24/06 5:41 PM Page 29

each ventricle, an action that produces abrief further rise in atrial and ventricle pres-sures, termed the a wave (see Fig. 2.1).

Contraction of the ventricles follows, sig-naling the onset of mechanical systole. Asthe ventricles start to contract, the pressureswithin them rapidly exceed atrial pressures.This results in the forced closure of the tri-cuspid and mitral valves, which producesthe first heart sound, termed S1. This soundhas two nearly superimposed components:

the mitral component slightly precedes thatof the tricuspid valve because of the earlierelectrical stimulation of left ventricular con-traction (see Chapter 4).

As the right and left ventricular pressuresrapidly rise further, they soon exceed the diastolic pressures within the pulmonaryartery and aorta, forcing the pulmonic andaortic valves to open, and blood is ejectedinto the pulmonary and systemic circula-tions. The ventricular pressures continue toincrease during contraction, and because thepulmonic and aortic valves are open, the aor-tic and pulmonary artery pressures rise, par-allel to those of the corresponding ventricles.

At the conclusion of ventricular ejection,the ventricular pressures fall below those ofthe pulmonary artery and aorta (the pul-monary artery and aorta are elastic struc-tures that maintain their pressures longer),such that the pulmonic and aortic valves areforced to close, producing the second heartsound, S2. Like the first heart sound (S1), thissound consists of two parts: the aortic (A2)component normally precedes the pulmonic(P2) because the diastolic pressure gradientbetween the aorta and left ventricle isgreater than that between the pulmonaryartery and right ventricle, forcing the aorticvalve to shut more readily. The ventricularpressures fall rapidly during the subsequentrelaxation phase. As they drop below thepressures in the right and left atria, the tri-cuspid and mitral valves open, followed bydiastolic ventricular filling and repetition ofthis cycle.

Notice in Figure 2.1 that in addition tothe a wave, the atrial pressure curve displaystwo other positive deflections during thecardiac cycle: The c wave represents a smallrise in atrial pressure as the tricuspid and mi-tral valves close and bulge into their respec-tive atria. The v wave is the result of passivefilling of the atria from the systemic andpulmonary veins during systole, a periodduring which blood accumulates in the atriabecause the tricuspid and mitral valves areclosed.

At the bedside, systole can be approxi-mated by the period from S1 to S2, and dias-tole from S2 to the next S1. Although the du-

30 Chapter Two

Figure 2.1. The normal cardiac cycle, showing pres-sure relationships between the left-sided heartchambers. During diastole, the mitral valve (MV) isopen, so that the left atrial (LA) and left ventricular (LV)pressures are equal. In late diastole, LA contractioncauses a small rise in pressure in both the LA and LV (thea wave). During systolic contraction, the LV pressurerises; when it exceeds the LA pressure, the MV closes,contributing to the first heart sound (S1). As LV pressurerises above the aortic pressure, the aortic valve (AV)opens, which is a silent event. As the ventricle begins torelax and its pressure falls below that of the aorta, theAV closes, contributing to the second heart sound (S2).As LV pressure falls further, below that of the LA, theMV opens, which is silent in the normal heart. In addi-tion to the a wave, the LA pressure curve displays twopositive deflections: the c wave represents a small rise inLA pressure as the MV closes and bulges toward theatrium, and the v wave is the result of passive filling ofthe LA from the pulmonary veins during systole, whenthe MV is closed.

AV opens

Pre

ssur

e (m

m H

g)

MV opens

AV closes

MV closes

DIASTOLE DIASTOLESYSTOLE

a c v

Time

S1 S2

LV

ECG

100

50

LA

Aorta

10090-02_CH02.qxd 8/24/06 5:41 PM Page 30

The Cardiac Cycle: Mechanisms of Heart Sounds and Murmurs 31

ration of systole remains constant from beatto beat, the length of diastole varies with theheart rate: the faster the heart rate, the shorterthe diastolic phase. The main sounds, S1 andS2, provide a framework from which all otherheart sounds and murmurs can be timed.

The pressure relationships and events de-picted in Figure 2.1 are those that occur in

the left side of the heart. Equivalent eventsoccur simultaneously in the right side of theheart in the right atrium, right ventricle,and pulmonary artery. At the bedside, cluesto right-heart function can be ascertained byexamining the jugular venous pulse, whichis representative of the right atrial pressure(see Box 2.1).

Box 2.1 Jugular Venous Pulsations and Assessment of Right-Heart Function

Bedside observation of jugular venous pulsations in the neck is a vital part of the car-diovascular examination. With no structures impeding blood flow between the internaljugular (IJ) veins and the superior vena cava and right atrium (RA), the height of the IJvenous column (termed the jugular venous pressure, or JVP) is an accurate representa-tion of the RA pressure. Thus, the JVP provides an easily obtainable measure of right-heart function.

Typical fluctuations in the jugular ve-nous pulse during the cardiac cycle, man-ifested by oscillations in the overlyingskin, are shown in the figure (notice thesimilarity to the left atrial pressure tracingin Fig. 2.1). There are two major upwardcomponents, the a and v waves, fol-lowed by two descents, termed x and y.The x descent, which represents the pressure decline following the a wave, may be inter-rupted by a small upward deflection (the c wave, denoted in the figure by the arrow) atthe time of tricuspid valve closure, but that is usually not distinguishable in the JVP. The awave represents transient venous distension caused by back pressure from RA contrac-tion. The v wave corresponds to passive filling of the RA from the systemic veins duringsystole, when the tricuspid valve is closed. Opening of the tricuspid valve in early diastoleallows blood to rapidly empty from the RA into the right ventricle; that fall in RA pressurecorresponds to the y descent.

Conditions that abnormally raise right-sided cardiac pressures (e.g., heart failure, tri-cuspid valve disease, pulmonic stenosis, pericardial diseases) elevate the JVP, while re-duced intravascular volume (e.g., dehydration) decreases it. In addition, specific diseasestates can influence the individual components of the JVP, examples of which are listedhere for reference and explained in subsequent chapters:

Prominent a: right ventricular hypertrophy, tricuspid stenosisProminent v: tricuspid regurgitationProminent y: constrictive pericarditis

Technique of Measurement

The JVP is measured as the maximum vertical height of the internal jugular vein (in cm)above the center of the right atrium, and in a normal person is ≤9 cm. Because the ster-nal angle is located approximately 5 cm above the center of the RA, the JVP is calculated

a

x

v

y

Box 1

10090-02_CH02.qxd 8/24/06 5:41 PM Page 31

32 Chapter Two

at the bedside by adding 5 cm to the vertical height of the top of the IJ venous columnabove the sternal angle.

The right IJ vein is usually the easiest to evaluate because it extends directly upwardfrom the RA and superior vena cava. First, observe the pulsations in the skin overlying theIJ with the patient supine and the head of the bed at about a 45° angle. Shining a lightobliquely across the neck helps to visualize the pulsations. Be sure to examine the IJ, notthe external jugular vein. The former is medial to, or behind, the sternocleidomastoid mus-cle, while the external jugular is usually more lateral. Although the external jugular is typ-ically easier to see, it does not accurately reflect RA pressure because it contains valves thatinterfere with venous return to the heart.

If the top of the IJ column is not visible at 45°, the column of blood is either too low(below the clavicle) or too high (above the jaw) to be measured in that position. In suchsituations, the head of the bed must be lowered or raised, respectively, so that the top ofthe column becomes visible. As long as the top can be ascertained, the vertical height of the JVP above the sternal angle will accurately reflect RA pressure, no matter the angleof the head of the bed.

Sometimes it can be difficult to distinguish the jugular venous pulsations from theneighboring carotid artery. Unlike the carotid, the JVP is usually not palpable, it has a dou-ble rather than a single upstroke, and it declines in most patients by assuming the seatedposition or during inspiration.

HEART SOUNDS

Typical stethoscopes contain two chest piecesfor auscultation of the heart. The concave“bell” chest piece, meant to be applied lightlyto the skin, accentuates low-frequency sounds.Conversely, the flat “diaphragm” chest pieceshould be pressed firmly against the skin toeliminate low frequencies and therefore ac-centuate high-frequency sounds and mur-murs. Some modern stethoscopes incorpo-rate both the bell and diaphragm functionsinto a single chest piece; in these models,placing the piece lightly on the skin bringsout the low-frequency sounds, while firmpressure accentuates the high-frequencyones.

First Heart Sound (S1)

S1 is produced by closure of the mitral andtricuspid valves in early systole and is loud-est near the apex of the heart (Fig. 2.2). It isa high-frequency sound, best heard with thediaphragm of the stethoscope. Although mitral closure usually precedes tricuspid closure, they are separated by only about0.01 sec, such that the human ear appreci-

ates only a single sound. An exception oc-curs in patients with right bundle branchblock (see Chapter 4), in whom these com-ponents may be audibly split because of de-layed closure of the tricuspid valve.

Three factors determine the intensity of S1:(1) the distance separating the leaflets of theopen valves at the onset of ventricular con-traction; (2) the mobility of the leaflets (nor-mal, or rigid because of stenosis); and (3) therate of rise of ventricular pressure (Table 2.1).

Figure 2.2. Standard positions of stethoscope place-ment for cardiac auscultation.

Fig. 2

Tab. 1

10090-02_CH02.qxd 8/24/06 5:41 PM Page 32


The distance between the open valve leaf-lets at the onset of ventricular contractionrelates to the electrocardiographic PR inter-val (see Chapter 4), the period between theonset of atrial and ventricular activation.Atrial contraction at the end of diastoleforces the tricuspid and mitral valve leafletsapart. As they start to drift back together,ventricular contraction forces them shut,from whatever position they are at, as soonas the ventricular pressure exceeds that inthe atrium. An accentuated S1 results whenthe PR interval is shorter than normal be-cause the valve leaflets do not have suffi-cient time to drift back together and aretherefore forced shut from a relatively widedistance.

Similarly, in mild mitral stenosis (seeChapter 8) a prolonged diastolic pressuregradient exists between the left atrium andventricle, which keeps the mobile portionsof the mitral leaflets farther apart than nor-mal during diastole. Because the leaflets arerelatively wide apart at the onset of systole,they are forced shut loudly when the leftventricle contracts.

S1 also may be accentuated when theheart rate is more rapid than normal (i.e.,tachycardia) because diastole is shortenedand the leaflets have insufficient time to driftback together before the ventricles contract.

Conditions that reduce the intensity of S1

are also listed in Table 2.1. In first-degreeatrioventricular (AV) block (see Chapter 12),

a diminished S1 results from an abnormallyprolonged PR interval, which delays theonset of ventricular contraction. Followingatrial contraction, the mitral and tricuspidvalves have additional time to float back to-gether so that the leaflets are forced closedfrom only a small distance apart.

In patients with mitral regurgitation (seeChapter 8), S1 is often diminished in inten-sity because the mitral leaflets may not comeinto full contact with one another as theyclose. In severe mitral stenosis, the leaflets arenearly fixed in position throughout the car-diac cycle, and that reduced movementlessens the intensity of S1.

In patients with a “stiffened” left ventri-cle (e.g., a hypertrophied chamber), atrialcontraction results in a higher-than-normalpressure at the end of diastole. This greaterpressure causes the mitral leaflets to drift to-gether more rapidly, forcing them closedfrom a smaller-than-normal distance whenventricular contraction begins and thus re-ducing the intensity of S1.

Second Heart Sound (S2)

The second heart sound results from the clo-sure of the aortic and pulmonic valves andtherefore has aortic (A2) and pulmonic (P2)components. Unlike S1, which is usuallyheard as a single sound, the components ofS2 vary with the respiratory cycle: they arenormally fused as one sound during expira-tion but become audibly separated duringinspiration, a situation termed normal orphysiologic splitting (Fig. 2.3).

One explanation for normal splitting ofS2 is as follows. Expansion of the chest dur-ing inspiration causes the intrathoracic pres-sure to become more negative. The negativepressure transiently increases the capaci-tance (and reduces the impedance) of the in-trathoracic pulmonary vessels. As a result,there is a temporary delay in the diastolic“back pressure” of the pulmonary artery re-sponsible for closure of the pulmonic valve.Thus, P2 is delayed; that is, it occurs laterduring inspiration than during expiration.

Inspiration has the opposite effect on A2.Because the capacity of the intrathoracic

TABLE 2.1. Causes of Altered Intensity of FirstHeart Sound (S1)

Accentuated S1

1. Shortened PR interval2. Mild mitral stenosis3. High cardiac output states or tachycardia

(e.g., exercise or anemia)

Diminished S1

1. Lengthened PR interval: first-degree AV nodalblock

2. Mitral regurgitation3. Severe mitral stenosis4. “Stiff” left ventricle (e.g., systemic hypertension)

AV, atrioventricular.

Fig. 3

10090-02_CH02.qxd 8/24/06 5:41 PM Page 33

34 Chapter Two

P2

P2

P2

P2

A. Physiologic (normal)splitting

B. Widened splitting

C. Fixed splitting

D. Paradoxical splitting

(Note reversed position ofA2 and P2)

Figure 2.3. Splitting patterns of the second heart sound (S2). A2, aortic component; P2, pulmonic componentof S2; S1, first heart sound.

10090-02_CH02.qxd 8/24/06 5:41 PM Page 34


pulmonary veins is increased by the nega-tive pressure generated by inspiration, thevenous return to the left atrium and ventri-cle temporarily decreases. Reduced filling ofthe LV causes a reduced stroke volume dur-ing the next systolic contraction and there-fore shortens the time required for LV emp-tying. Therefore, aortic valve closure (A2)occurs slightly earlier in inspiration thanduring expiration. The combination of anearlier A2 and delayed P2 during inspirationcauses audible separation of the two com-ponents. Since these components are high-frequency sounds, they are best heard withthe diaphragm of the stethoscope, and split-ting of S2 is usually most easily appreciatednear the second left intercostal space next tothe sternum (the pulmonic area).

Abnormalities of S2 include alterations inits intensity and changes in the pattern ofsplitting. The intensity of S2 depends on thevelocity of blood coursing back toward thevalves from the aorta and pulmonary arteryafter the completion of ventricular contrac-tion, and the suddenness with which thatmotion is arrested by the closing valves. Insystemic hypertension or pulmonary arte-rial hypertension, the diastolic pressure inthe respective great artery is higher thannormal, such that the velocity of the bloodsurging toward the valve is elevated and S2 isaccentuated. Conversely, in severe aortic orpulmonic valve stenosis, the valve commis-sures are nearly fixed in position, such thatthe contribution of the stenotic valve to S2 isdiminished.

Widened splitting of S2 refers to an in-crease in the time interval between A2 and P2,such that the two components are audiblyseparated even during expiration and becomemore widely separated in inspiration (see Fig.2.3). This pattern is usually the result of de-layed closure of the pulmonic valve, whichoccurs in right bundle branch block and pul-monic valve stenosis.

Fixed splitting of S2 is an abnormallywidened interval between A2 and P2 that per-sists unchanged through the respiratorycycle (see Fig. 2.3). The most common ab-normality that causes fixed splitting of S2 isan atrial septal defect (see Chapter 16). In

that condition, chronic volume overload ofthe right-sided circulation results in a high-capacitance, low-resistance pulmonary vas-cular system. This alteration in pulmonaryartery hemodynamics delays the back pres-sure responsible for closure of the pulmonicvalve. Thus, P2 occurs later than normal,even during expiration, such that there iswider than normal separation of A2 and P2.The pattern of splitting does not change (i.e.,it is fixed) during the respiratory cycle be-cause (1) inspiration does not substantiallyincrease further the already elevated pul-monary vascular capacitance, and (2) aug-mented filling of the right atrium from thesystemic veins during inspiration is counter-balanced by a reciprocal decrease in the left-to-right transatrial shunt, eliminating respi-ratory variations in right ventricular filling.

Paradoxical splitting (or reversed split-ting) refers to audible separation of A2 and P2

during expiration that disappears on inspira-tion, the opposite of the normal situation. Itreflects an abnormal delay in the closure ofthe aortic valve such that P2 precedes A2. Inadults, the most common cause is left bun-dle branch block (LBBB). In LBBB, the spreadof electrical activity through the left ventri-cle is impaired, resulting in delayed ventric-ular contraction and late closure of the aor-tic valve such that it follows P2. Duringinspiration, as in the normal case, the pul-monic valve closure sound is delayed andthe aortic valve closure sound moves earlier.This results in narrowing and often superim-position of the two sounds; thus, there is noapparent split at the height of inspiration(see Fig. 2.3). In addition to LBBB, paradoxi-cal splitting may be observed under circum-stances in which left ventricular ejection isgreatly prolonged, such as aortic stenosis.

Extra Systolic Heart Sounds

Extra systolic heart sounds may occur inearly, mid-, or late systole.

Early Extra Systolic Heart Sounds

Abnormal early systolic sounds, or ejectionclicks, occur shortly after S1 and coincide with

10090-02_CH02.qxd 8/24/06 5:41 PM Page 35

the opening of the aortic or pulmonic valves(Fig. 2.4). These sounds have a sharp, high-pitched quality, so they are heard best withthe diaphragm of the stethoscope placedover the aortic and pulmonic areas. Ejectionclicks indicate the presence of aortic or pul-monic valve stenosis or dilatation of the pul-monary artery or aorta. In stenosis of the aor-tic or pulmonic valve, the sound occurs asthe valve leaflets reach their maximal level ofascent into the great artery, just prior toblood ejection. At that moment, the rapidlyascending valve reaches its elastic limit anddecelerates abruptly, an action thought to re-sult in the sound generation. In dilatation ofthe root of the aorta or pulmonary artery, thesound is associated with sudden tensing of

the aortic or pulmonic root with the onsetof blood flow into the vessel. The aortic ejec-tion click is heard at both the base and theapex of the heart and does not vary with res-piration. In contrast, the pulmonic ejectionclick is heard only at the base and its inten-sity diminishes during inspiration (see Chap-ter 16).

Mid- or Late Extra Systolic Heart Sounds

Clicks occurring in mid- or late systole areusually the result of systolic prolapse of the mitral or tricuspid valves, in which the leaflets bulge abnormally from the ven-tricular side of the atrioventricular junctioninto the atrium during ventricular contrac-tion, often accompanied by valvular regur-gitation. They are loudest over the mitral ortricuspid auscultatory regions, respectively.

Extra Diastolic Heart Sounds

Extra heart sounds in diastole include theopening snap (OS), the third heart sound(S3), the fourth heart sound (S4), and thepericardial knock.

Opening Snap

Opening of the mitral and tricuspid valves isnormally silent, but mitral or tricuspid valvu-lar stenosis (usually the result of rheumaticheart disease; see Chapter 8) produces asound, termed a snap, when the affected valveopens. It is a sharp, high-pitched sound, andits timing does not vary significantly withrespiration. In mitral stenosis (which is muchmore common than tricuspid valve steno-sis), the OS is heard best between the apexand the left sternal border, just after the aor-tic closure sound (A2), when the left ventric-ular pressure falls below that of the leftatrium (see Fig. 2.4).

Because of its proximity to A2, the A2–OSsequence can be confused with a widelysplit second heart sound. However, carefulauscultation at the pulmonic area during inspiration reveals three sounds occurringin rapid succession (Fig. 2.5), which corre-spond to aortic closure (A2), pulmonic clo-

36 Chapter Two

MV opens

S1 S2

LV

ECG

LA

Aorta

S4Ejection

clickOS

S3

Figure 2.4. Timing of extra systolicand diastolic heart sounds. S4 is pro-duced by atrial contraction into a “stiff”left ventricle (LV). An ejection click fol-lows the opening of the aortic or pul-monic valve in cases of valve stenosis ordilatation of the corresponding greatartery. S3 occurs during the period ofrapid ventricular filling; it is normal inyoung people, but its presence in adultsimplies LV contractile dysfunction. Thetiming of an opening snap (OS) is placedfor comparison, but it is not likely that allof these sounds would appear in thesame person. LA, left atrium; MV, mitralvalve.

Fig. 4

Fig. 5

10090-02_CH02.qxd 8/24/06 5:41 PM Page 36


sure (P2), and then the opening snap (OS).The three sounds become two on expirationbecause A2 and P2 normally fuse.

The severity of stenosis can be approxi-mated by the time interval between A2 andthe opening snap: the more advanced thestenosis, the shorter the interval. This oc-curs because the degree of left atrial pressureelevation corresponds to the severity of mi-tral stenosis. When the ventricle relaxes indiastole, the greater the left atrial pressure,the earlier the mitral valve opens. Com-pared with severe stenosis, mild disease ismarked by a less elevated left atrial pressureis less elevated, lengthening the time ittakes for the left ventricular pressure to fallbelow that of the atrium. Therefore, in mild

mitral stenosis, the opening snap is widelyseparated from A2, whereas in more severestenosis, the A2–OS interval is narrower.

Third Heart Sound (S3)

When present, an S3 occurs in early diastole,following the opening of the atrioventricu-lar valves, during the ventricular rapid fill-ing phase (see Fig. 2.4). It is a dull, low-pitched sound best heard with the bell ofthe stethoscope. A left-sided S3 is typicallyloudest over the cardiac apex while the pa-tient lies in the left lateral decubitus posi-tion. A right-sided S3 is better appreciated at the lower-left sternal border. Productionof the S3 appears to result from tensing ofthe chordae tendineae during rapid fillingand expansion of the ventricle.

A third heart sound is a normal finding inchildren and young adults. In these groups,an S3 implies the presence of a supple ven-tricle capable of normal rapid expansion inearly diastole. Conversely, when heard inmiddle-aged or older adults, an S3 is often asign of disease, indicating volume overloadowing to congestive heart failure, or the in-creased transvalvular flow that accompaniesadvanced mitral or tricuspid regurgitation.A pathologic S3 is sometimes referred to as aventricular gallop.

Fourth Heart Sound (S4)

When an S4 is present, it occurs in late di-astole and coincides with contraction ofthe atria (see Fig. 2.4). This sound is gener-ated by the left (or right) atrium vigorouslycontracting against a stiffened ventricle.Thus, an S4 usually indicates the presenceof cardiac disease—specifically, a decreasein ventricular compliance typically result-ing from ventricular hypertrophy or myo-cardial ischemia. Like an S3, the S4 is a dull,low-pitched sound and is best heard withthe bell of the stethoscope. In the case of the more common left-sided S4, thesound is loudest at the apex, with the pa-tient lying in the left lateral decubitus position. S4 is sometimes referred to as anatrial gallop.

Expiration

OS

OS

S1 S2

A2P2

Inspiration

Figure 2.5. Timing of the opening snap (OS) in mitralstenosis does not change with respiration. On inspi-ration, normal splitting of the second heart sound (S2) isobserved so that three sounds are heard. A2, aortic com-ponent; P2, pulmonic component of S2; S1, first heartsound.

10090-02_CH02.qxd 8/24/06 5:41 PM Page 37

Quadruple Rhythm or Summation Gallop

In a patient with both an S3 and S4, thosesounds, in conjunction with S1 and S2, pro-duce a quadruple beat. If a patient with aquadruple rhythm develops tachycardia, di-astole becomes shorter in duration, the S3

and S4 coalesce, and a summation gallopresults. The summation of S3 and S4 is heardas a long middiastolic, low-pitched sound,often louder than S1 and S2.

Pericardial Knock

A pericardial knock is an uncommon, high-pitched sound that occurs in patients withsevere constrictive pericarditis (see Chapter14). It appears early in diastole soon after S2 and can be confused with an openingsnap or an S3. However, the knock appearsslightly later in diastole than the timing ofan opening snap and is louder and occursearlier than the ventricular gallop. It resultsfrom the abrupt cessation of ventricular fill-ing in early diastole, which is the hallmarkof constrictive pericarditis.

MURMURS

A murmur is the sound generated by turbu-lent blood flow. Under normal conditions,the movement of blood through the vascularbed is laminar, smooth and silent. However,as a result of hemodynamic and/or structuralchanges, laminar flow can become disturbedand produce an audible noise. Murmurs re-sult from any of the following mechanisms:

1. Flow across a partial obstruction (e.g., aortic stenosis)

2. Increased flow through normal structures(e.g., aortic systolic murmur associatedwith a high-output state, such as anemia)

3. Ejection into a dilated chamber (e.g.,aortic systolic murmur associated withaneurysmal dilatation of the aorta)

4. Regurgitant flow across an incompetentvalve (e.g., mitral regurgitation)

5. Abnormal shunting of blood from onevascular chamber to a lower-pressurechamber (e.g., ventricular septal defect)

Murmurs are described by their timing, intensity, pitch, shape, location, radiation,and response to maneuvers. Timing refers towhether the murmur occurs during systoleor diastole or is continuous (i.e., begins insystole and continues into diastole). The in-tensity of the murmur is typically quantifiedby a grading system. In the case of systolicmurmurs:

38 Chapter Two

Grade 1/6 (or I/VI): Barely audible (i.e., med-ical students may nothear it!)

Grade 2/6 (or II/VI): Faint but immediatelyaudible

Grade 3/6 (or III/VI): Easily heardGrade 4/6 (or IV/VI): Easily heard and asso-

ciated with a palpablethrill

Grade 5/6 (or V/VI): Very loud; heard withstethoscope lightly onchest

Grade 6/6 (or VI/VI): Audible without thestethoscope directly onthe chest wall

Grade 1/4 (or I/IV): Barely audibleGrade 2/4 (or II/IV): Faint but immediately

audibleGrade 3/4 (or III/IV): Easily heardGrade 4/4 (or IV/IV): Very loud

And in the case of diastolic murmurs:

Pitch refers to the frequency of the murmur,ranging from high to low. High-frequencymurmurs are caused by large pressure gra-dients between chambers (e.g., aortic steno-sis) and are best appreciated using the di-aphragm chest piece of the stethoscope.Low-frequency murmurs imply less of a pres-sure gradient between chambers (e.g., mitralstenosis) and are best heard using the stetho-scope’s bell piece.

Shape describes how the murmur changesin intensity from its onset to its comple-tion. For example, a crescendo–decrescendo

10090-02_CH02.qxd 8/24/06 5:41 PM Page 38


(or “diamond-shaped”) murmur first risesand then falls off in intensity. Other shapesinclude decrescendo (i.e., the murmur beginsat its maximum intensity and grows softer)and uniform (the intensity of the murmurdoes not change).

Location refers to the murmur’s region ofmaximum intensity and is usually describedin terms of specific auscultatory areas (seeFig. 2.2):

right, Valsalva (forceful expiration against aclosed airway), or clenching of the fists, eachof which alters the heart’s loading condi-tions and can affect the intensity of manymurmurs. Examples of the effects of maneu-vers on specific murmurs are presented inChapter 8.

When reporting a murmur, some or all ofthese descriptors are mentioned. For exam-ple, you might describe a particular patient’smurmur of aortic stenosis as “A grade III/VI high-pitched, crescendo–decrescendo sys-tolic murmur, heard best at the upper-rightsternal border, radiating toward the neck.”

Systolic Murmurs

Systolic murmurs are subdivided into sys-tolic ejection murmurs, pansystolic mur-murs, and late systolic murmurs (Fig. 2.6). Asystolic ejection murmur is typical of aor-tic or pulmonic valve stenosis. It begins afterthe first heart sound and terminates beforeor during S2, depending on its severity andwhether the obstruction is of the aortic orpulmonic valve. The shape of the murmur isof the crescendo–decrescendo type (i.e., itsintensity rises and then falls).

Aortic area: Second to third right inter-costal spaces, next to sternum

Pulmonic area: Second to third left intercostalspaces, next to sternum

Tricuspid area: Lower-left sternal borderMitral area: Cardiac apex

From their primary locations, murmurs areoften heard to radiate to other areas of thechest, and such patterns of transmission re-late to the direction of the turbulent flow.Finally, similar types of murmurs can bedistinguished from one another by simplebedside maneuvers, such as standing up-

Click

Figure 2.6. Classification of systolic murmurs. Ejection murmurs arecrescendo–decrescendo in configuration, whereas pansystolic murmurs areuniform throughout systole. A late systolic murmur often follows a midsys-tolic click and suggests mitral (or tricuspid) valve prolapse.

Fig. 6

10090-02_CH02.qxd 8/24/06 5:41 PM Page 39

The ejection murmur of aortic stenosis be-gins in systole after S1, from which it is sep-arated by a short audible gap (Fig. 2.7). Thisgap corresponds to the period of isovolu-metric contraction of the left ventricle (theperiod after the mitral valve has closed butbefore the aortic valve has opened). Themurmur becomes more intense as flow in-creases across the aortic valve during therise in left ventricular pressure (crescendo).Then, as the ventricle relaxes, forward flowdecreases, and the murmur lessens in in-tensity (decrescendo) and finally ends priorto the aortic component of S2. The murmurmay be immediately preceded by an ejec-tion click, especially in mild forms of aorticstenosis.

Although the intensity of the murmurdoes not correlate well with the severity ofaortic stenosis, other features do. For exam-ple, the more severe the stenosis, the longer ittakes to force blood across the valve, and thelater the murmur peaks in systole (Fig. 2.8).Also, as shown in Figure 2.8, as the severity of

stenosis increases, the aortic component of S2

softens because the leaflets become morerigidly fixed in place.

Aortic stenosis causes a high-frequencymurmur, reflecting the sizable pressure gra-dient across the valve. It is best heard in the“aortic area” at the second and third rightintercostal spaces close to the sternum. Themurmur typically radiates toward the neck(the direction of turbulent blood flow) but

40 Chapter Two

S1 S2

Figure 2.7. Systolic ejection murmur of aortic steno-sis. There is a short delay between the first heart sound(S1) and the onset of the murmur. LV, left ventricle; S2,second heart sound.

S1 A2 P2

S1 A2 P2

S1 P2

Figure 2.8. The severity of aortic stenosis affectsthe shape of the systolic murmur and the heartsounds. A. In mild stenosis, an ejection click (EJ) is oftenpresent, followed by an early peaking crescendo–decrescendo murmur and a normal aortic component ofS2 (A2). B. As stenosis becomes more severe, the peak ofthe murmur becomes more delayed in systole and theintensity of A2 lessens. The prolonged ventricular ejec-tion time delays A2 so that it merges with or occurs afterthe pulmonic component of S2 (P2); the ejection clickmay not be heard. C. In severe stenosis, the murmurpeaks very late in systole, and A2 is usually absent be-cause of immobility of the valve leaflets. S1, first heartsound; S2, second heart sound.

Fig. 7

Fig. 8

10090-02_CH02.qxd 8/24/06 5:41 PM Page 40


often can be heard in a wide distribution, in-cluding the cardiac apex.

The murmur of pulmonic stenosis also be-gins after S1, it may be preceded by an ejec-tion click, but unlike aortic stenosis, it mayextend beyond A2. That is, if the stenosis issevere, it will result in a very prolonged rightventricular ejection time, elongating themurmur, which will continue beyond A2

and end just before closure of the pulmonicvalve (P2). Pulmonic stenosis is usually loud-est at the second to third left intercostalspaces close to the sternum. It does not ra-diate as widely as aortic stenosis, but some-times it is transmitted to the neck or leftshoulder.

Young adults often have benign systolicejection murmurs owing to increased systolicflow across normal aortic and pulmonicvalves. This type of murmur often becomessofter or disappears when the patient sitsupright.

Pansystolic (also termed holosystolic)murmurs are caused by regurgitation ofblood across an incompetent mitral or tri-cuspid valve or through a ventricular septaldefect (VSD; see Fig. 2.6). These murmursare characterized by a uniform intensitythroughout systole. In mitral and tricuspidvalve regurgitation, as soon as ventricularpressure exceeds atrial pressure (i.e., when S1

occurs), there is immediate retrograde flowacross the regurgitant valve. Thus, there isno gap between S1 and the onset of thesepansystolic murmurs, in contrast to the systolic ejection murmurs discussed earlier.Similarly, there is no significant gap be-tween S1 and the onset of the systolic murmur of a VSD, because left ventricularsystolic pressure exceeds right ventricularsystolic pressure (and flow occurs) quicklyafter the onset of contraction.

The pansystolic murmur of advanced mi-tral regurgitation continues through the aor-tic closure sound because left ventricularpressure remains greater than that in the leftatrium at the time of aortic closure. Themurmur is heard best at the apex, is highpitched and “blowing” in quality, and oftenradiates toward the left axilla; its intensitydoes not change with respiration.

Tricuspid valve regurgitation is best heardalong the left lower sternal border. It gener-ally radiates to the right of the sternum andis high pitched and blowing in quality. Theintensity of the murmur increases with inspiration because the negative intratho-racic pressure induced during inspirationenhances venous return to the heart. Thelatter augments right ventricular stroke vol-ume, thereby increasing the amount of re-gurgitated blood.

The murmur of a ventricular septal defect isheard best at the fourth to sixth left inter-costal spaces, is high pitched, and may be as-sociated with a palpable thrill. The intensityof the murmur does not increase with in-spiration, nor does it radiate to the axilla,which helps distinguish it from tricuspidand mitral regurgitation, respectively. Ofnote, the smaller the VSD, the greater theturbulence of blood flow between the leftand right ventricles and the louder the mur-mur. Some of the loudest murmurs everheard are those associated with small VSDs.

Late systolic murmurs begin in mid-to-late systole and continue to the end of sys-tole. The most common example is mitralregurgitation caused by mitral valve prolapse—bowing of abnormally redundant and elon-gated valve leaflets into the left atrium dur-ing ventricular contraction (see Fig. 2.6).This murmur is usually preceded by amidsystolic click and is described further inChapter 8.

Diastolic Murmurs

Diastolic murmurs are divided into early de-crescendo murmurs and mid-to-late rum-bling murmurs (Fig. 2.9). Early diastolicmurmurs result from regurgitant flowthrough either the aortic or pulmonic valve,with the former being much more commonin adults. If produced by aortic valve regurgi-tation, the murmur begins at A2, has a de-crescendo shape, and terminates before thenext S1. Because diastolic relaxation of theleft ventricle is rapid, a pressure gradient de-velops immediately between the aorta andlower-pressured left ventricle in aortic re-gurgitation, and the murmur therefore dis-

Fig. 9

10090-02_CH02.qxd 8/24/06 5:41 PM Page 41

plays its maximum intensity at its onset.Thereafter in diastole, as the aortic pressurefalls and the LV pressure increases (as bloodfills the ventricle), the gradient between thetwo chambers diminishes and the murmurdecreases in intensity. Aortic regurgitation isa high-pitched murmur, best heard usingthe diaphragm of the stethoscope along theleft sternal border with the patient sitting,leaning forward, and exhaling.

Pulmonic regurgitation in adults is usuallyowing to the presence of pulmonary arterialhypertension. It has an early diastolic de-crescendo murmur profile similar to that ofaortic regurgitation, but it is best heard inthe pulmonic area and its intensity may in-crease with inspiration.

Mid-to-late diastolic murmurs resultfrom either turbulent flow across a stenoticmitral or tricuspid valve or less commonlyfrom abnormally increased flow across anormal mitral or tricuspid valve (see Fig.2.9). If resulting from stenosis, the murmur

begins after S2 and is preceded by an open-ing snap. The shape of this murmur isunique. Following the opening snap, themurmur is at its loudest because the pres-sure gradient between the atrium and ven-tricle is at its maximum. The murmur thendecrescendos or disappears totally duringdiastole as the transvalvular gradient de-creases. The degree to which the murmurfades depends on the severity of the steno-sis. If the stenosis is severe, the murmur is prolonged; if the stenosis is mild, themurmur disappears in mid-to-late diastole.Whether the stenosis is mild or severe, themurmur intensifies at the end of diastole inpatients in normal sinus rhythm, whenatrial contraction augments flow across thevalve (see Fig. 2.9). The murmur of mitralstenosis is low pitched and is heard bestwith the bell of the stethoscope at the apex,while the patient lies in the left lateral de-cubitus position. The much less commonmurmur of tricuspid stenosis is better aus-

42 Chapter Two

S1 S2 S1

S1 S2 S1

S1 S2 S1

• Aortic regurgitation• Pulmonic regurgitation

• Mild mitral or tricuspid stenosis

• Severe mitral or tricuspid stenosis

Figure 2.9. Classification of diastolic murmurs. A. An early diastolic decrescendo murmur is typi-cal of aortic or pulmonic valve regurgitation. B. Mid-to-late low-frequency rumbling murmurs are usu-ally the result of mitral or tricuspid valve stenosis, which follows a sharp opening snap (OS). Presystolicaccentuation of the murmur occurs in patients in normal sinus rhythm because of the transient rise inatrial pressure during atrial contraction. C. In more severe mitral or tricuspid valve stenosis, the open-ing snap and diastolic murmur occur earlier and the murmur is prolonged. S1, first heart sound; S2, sec-ond heart sound.

1 LINE SHORT

10090-02_CH02.qxd 8/24/06 5:41 PM Page 42


cultated at the lower sternum, near the xi-phoid process.

Hyperdynamic states such as fever, ane-mia, hyperthyroidism, and exercise causeincreased flow across the normal tricuspidand mitral valves and can therefore result ina diastolic murmur. In patients with ad-vanced mitral regurgitation, the expectedsystolic murmur can be accompanied by anadditional diastolic murmur owing to theincreased volume of blood that must returnacross the valve to the left ventricle in dias-tole. Similarly, patients with either tricuspidregurgitation or an atrial septal defect (seeChapter 16) may display a diastolic flowmurmur across the tricuspid valve.

Continuous Murmurs

Continuous murmurs are heard throughoutthe cardiac cycle without an audible hiatusbetween systole and diastole. Such murmursresult from conditions in which there is apersistent pressure gradient between twostructures during systole and diastole. Anexample is the murmur of patent ductus ar-teriosus, in which there is an abnormal com-munication between the aorta and pul-monary artery (see Chapter 16). Duringsystole, blood flows from the high-pressureascending aorta through the ductus into thelower-pressure pulmonary artery. During di-astole, the aortic pressure remains greaterthan that in the pulmonary artery and flowcontinues across the ductus. This murmur

begins in early systole, crescendos to itsmaximum at S2, then decrescendos until thenext S1 (Fig. 2.10).

The “to-and-fro” combined murmur in apatient with both aortic stenosis and aorticregurgitation could be mistaken for a con-tinuous murmur (see Fig. 2.10). During sys-tole, there is a diamond-shaped ejectionmurmur, and during diastole a decrescendomurmur. However, in the case of a to-and-fro murmur, the sound does not extendthrough S2 because it has discrete systolicand diastolic components.

SUMMARY

Abnormal heart sounds and murmurs arecommon in acquired and congenital heartdisease and can be predicted by the underly-ing pathology. Although it may seem diffi-cult to remember even the basic features pre-sented here, it will become easier as you learnmore about the pathophysiology of theseconditions, and as your experience in physi-cal diagnoses grows. For now, just rememberthat the information is here, and refer to it asneeded. Tables 2.2 and 2.3 and Figure 2.11summarize features of the heart sounds andmurmurs described in this chapter.

Acknowledgments

Contributors to the previous editions of this chapterwere Oscar Benavidez, MD; Bradley S. Marino, MD;Allan Goldblatt, MD; and Leonard S. Lilly, MD.

S1 S2 S1

S1 S2 S1

• Aortic stenosis and regurgitation• Pulmonic stenosis and regurgitation

• Patent ductus arteriosus

Figure 2.10. A continuous murmur peaks at, and extends through, the second heart sound (S2).A to-and-fro murmur is not continuous; rather, there is a systolic component and a distinct diastolic com-ponent, separated by S2. S1, first heart sound.1 LINE SHORT

Fig. 10

Tab. 2,Tab. 3,Fig. 11

10090-02_CH02.qxd 8/24/06 5:41 PM Page 43

44 Chapter Two

TABLE 2.2. Common Heart Sounds

Sound Location Pitch Significance

S1

S2

Extra systolic soundsEjection clicks

Mid-to-late click

Extra diastolic soundsOpening snapS3

S4

LLSB, lower left sternal border.

High

High

HighHighHighHigh

HighLow

Low

Apex

Base

Aortic: apex and basePulmonic: baseMitral: apexTricuspid: LLSB

ApexLeft-sided: apex

Left-sided: apex

Normal closure of mitral and tricuspidvalves

Normal closure of aortic (A2) and pulmonic (P2) valves

Aortic or pulmonic stenosis, or dilata-tion of aortic root or pulmonary artery

Mitral or tricuspid valve prolapse

Mitral stenosisNormal in childrenAbnormal in adults: indicates heart

failure or volume overload stateReduced ventricular compliance

TABLE 2.3. Common Murmurs

Murmur Type Example Location and Radiation

Systolic ejection

Pansystolic

Late systolic

Early diastolic

Mid- or late diastolic

Aortic stenosis

Pulmonic stenosis

Mitral regurgitationTricuspid regurgitation

Mitral valve prolapse

Aortic regurgitationPulmonic regurgitation

Mitral stenosis

2nd right intercostal space →neck (but may radiatewidely)

2nd–3rd left intercostal spaces

Apex → axillaLeft lower sternal border →

right lower sternal border

Apex → axilla

Along left side of sternumUpper left side of sternum

Apex

10090-02_CH02.qxd 8/24/06 5:41 PM Page 44


Additional Reading

Bickley LS. Bates’ Guide to Physical Examinationand History Taking. 8th Ed. Philadelphia: Lippin-cott Williams & Wilkins, 2003.

Constant J. Essentials of Bedside Cardiology. 2nd Ed.Totowa, NJ: Humana Press, 2003.

LeBlond RF, DeGowin RL, Brown DD. DeGowin’s Diagnostic Examination. 7th Ed. New York: McGraw-Hill, 2004.

Orient JM, Sapira JD. Sapira’s Art and Science of Bed-side Diagnosis. 3rd Ed. Philadelphia: LippincottWilliams & Wilkins, 2005.

Pulmonic areaEjection-type murmur• Pulmonic stenosis• Flow murmur

Mitral areaPansystolic murmur• Mitral regurgitation

Tricuspid areaPansystolic murmur• Tricuspid regurgitation• Ventricular septal defect

Mid-to-late diastolicmurmur• Tricuspid stenosis• Atrial septal defect

Aortic areaEjection-type murmur• Aortic stenosis• Flow murmur

Mid-to-late diastolic murmur• Mitral stenosis

Left sternal borderEarly diastolic murmur• Aortic regurgitation• Pulmonic regurgitation

Figure 2.11. Locations of maximum intensity of common murmurs.

10090-02_CH02.qxd 8/24/06 5:41 PM Page 45

46

CARDIAC RADIOGRAPHYCardiac SilhouettePulmonary Manifestations of Heart Disease

ECHOCARDIOGRAPHYVentricular AssessmentValvular LesionsCoronary Artery DiseaseCardiomyopathyPericardial Disease

CARDIAC CATHETERIZATIONMeasurement of Pressure

Measurement of Blood FlowCalculation of Vascular ResistanceContrast Angiography

NUCLEAR IMAGINGAssessment of Myocardial PerfusionRadionuclide VentriculographyAssessment of Myocardial Metabolism

COMPUTED TOMOGRAPHY

MAGNETIC RESONANCE IMAGING

C H A P T E R

3Diagnostic Imaging andCardiac CatheterizationNicole MartinPatricia Challender Come

Imaging plays a central role in the assess-ment of cardiac function and pathology.Traditional imaging modalities such as chestradiography, echocardiography (echo), car-diac catheterization with cineangiography,and nuclear imaging are fundamental in thediagnosis and management of cardiovascu-lar diseases. These techniques are being in-creasingly supplemented by newer modali-ties, including computed tomography andmagnetic resonance imaging.

This chapter presents an overview of imag-ing techniques as they are used to assess thecardiovascular disorders described in thebook. It would be beneficial to familiarizeyourself with the information now, but not to

memorize the details. This chapter is meant asa reference to be consulted as needed.

CARDIAC RADIOGRAPHY

The extent of penetration of x-rays throughthe body is inversely proportional to tissuedensity. Air-filled tissues, such as the lung,absorb few x-rays and expose the underly-ing film, causing it to appear black. In con-trast, dense materials, such as bone, absorbmore radiation and appear white, or radio-paque. For a boundary to show between twostructures, they must differ in density. Myo-cardium, valves, and other intracardiacstructures have densities similar to that of

10090-03_CH03.qxd 8/31/06 5:24 PM Page 46

Diagnostic Imaging and Cardiac Catheterization 47

adjacent blood; consequently, radiographycannot delineate these structures unless theyhappen to be calcified. Conversely, heart bor-ders adjacent to a lung are depicted clearly be-cause the heart and an air-filled lung have dif-ferent densities. If the lung adjacent to theheart is diseased, however (as in pulmonaryedema, consolidation, or collapse), the lungdensity will match that of the heart and thecardiac border will be poorly defined.

Frontal and lateral radiographs are rou-tinely used to assess the heart and lungs (Fig.3.1). The frontal view is usually a posterior-anterior image in which the x-rays are trans-mitted from behind (i.e., posterior to) thepatient, travel through the body, and thenexpose a sheet of film placed against the an-

terior chest. This positioning places the heartclose to the x-ray film so that its image is onlyminimally distorted, allowing for an accurateassessment of size. In the standard lateralview, the patient’s left side is placed againstthe film plate and the x-rays pass throughthe body from right to left. The frontal ra-diograph is particularly useful for assessingthe size of the left ventricle, left atrial ap-pendage, pulmonary artery, aorta, and supe-rior vena cava; the lateral view evaluates rightventricular size, posterior borders of the leftatrium and ventricle, and the anteroposteriordiameter of the thorax. In some cases, opti-mal evaluation of the heart requires right andleft anterior oblique views as well.

Cardiac Silhouette

Chest radiographs are used to evaluate thesize of heart chambers and the pulmonaryconsequences of cardiac disease. Alterationsin chamber size are reflected by changes inthe cardiac silhouette. In the frontal view ofadults, the heart shadow should occupy 50%or less of the maximal width of the thorax,measured between the inner margins of theribs (although in children, normal cardiacdiameter may be up to 60% of the thoracicwidth). Thus, the cardio:thoracic ratio is usedinstead of absolute measurements to accountfor differences in body habitus.

In several situations, the cardiac silhou-ette inaccurately reflects heart size. An ele-vated diaphragm or narrow chest anteropos-terior diameter, for example, may cause theheart to appear to spread out transversely.Consequently, the silhouette on a posterior-anterior chest film may be greater than 50%of the thorax even though the actual heartsize is normal. Therefore, the chest antero-posterior diameter should be assessed on thelateral view before the frontal image is deter-mined to truly represent an enlarged heart.The presence of a pericardial effusion aroundthe heart can also enlarge the cardiac silhou-ette, because fluid and myocardium affect x-ray penetration similarly.

Radiographs can depict dilatation of thecardiac chambers and great vessels. Hyper-trophy alone may not result in radiographic

Figure 3.1. Posteroanterior (A and B) and lateral (Cand D) chest radiographs of a person without car-diopulmonary disease, illustrating cardiac chambersand valves. AO, aorta; AV, azygos vein; IVC, inferiorvena cava; LA, left atrium; LAA, left atrial appendage;LPA, left pulmonary artery; LV, left ventricle; MPA, mainpulmonary artery; MV, mitral valve; RA, right atrium; RPA,right pulmonary artery; RV, right ventricle; SVC, superiorvena cava; TV, tricuspid valve. (Reprinted with permissionfrom Come PC, ed. Diagnostic Cardiology: NoninvasiveImaging Techniques. Philadelphia: JB Lippincott, 1985.)

Fig. 1

10090-03_CH03.qxd 8/31/06 5:24 PM Page 47

abnormalities, because it generally occurs atthe expense of the cavity’s internal volumeand produces little or no change in overallcardiac size. Hypertrophy is more readilysuspected from abnormalities on the elec-trocardiogram or by directly measuring wallthickness by echocardiography (as discussedlater in the chapter). Major causes of chamberand great vessel dilatation include heart fail-ure, valvular lesions, abnormal intracardiacand extracardiac communications (shunts),and certain pulmonary disorders. Becausedilatation takes time to develop, recent le-sions, such as acute mitral valve insuffi-ciency, may present without apparent car-diac enlargement.

The pattern of chamber enlargement maysuggest specific disease entities. For example,dilatation of the left atrium and right ventri-

cle, accompanied by signs of pulmonary hy-pertension, suggests mitral stenosis (Fig. 3.2).In contrast, dilatation of the pulmonaryartery and right heart chambers, but withoutenlargement of the left-sided heart dimen-sions, suggests pulmonary vascular obstruc-tion or increased pulmonary artery bloodflow (e.g., owing to an atrial septal defect;Fig. 3.3).

The shape of the dilated chamber may alsoprovide etiologic clues. For instance, in leftventricular volume overload owing to valvu-lar insufficiency, the ventricle tends to en-large primarily in its long axis, displacing theapex downward and to the left. In contrast,when left ventricular dilation results from pri-mary myocardial dysfunction, left ventricu-lar length and width are generally both in-creased, causing the heart to appear globular.

48 Chapter Three

Figure 3.2. Posteroanterior chest radiograph of a patient with severemitral stenosis and secondary pulmonary vascular congestion. The ra-diograph shows a prominent left atrial appendage (arrowheads) with con-sequent straightening of the left-heart border and suggestion of a double-density right cardiac border (arrows) produced by the enlarged left atrium.The aortic silhouette is small, which suggests chronic low cardiac output. Ra-diographic signs of pulmonary vascular congestion include increased caliberof upper-zone pulmonary vessel markings and decreased caliber of lower-zone vessels.

Fig. 2

Fig. 3

10090-03_CH03.qxd 8/31/06 5:24 PM Page 48


Chest radiographs can also detect dila-tation of the aorta and pulmonary artery.Causes of aortic dilatation include aneurysm,dissection, and aortic valve disease (Fig. 3.4).Normal aging and atherosclerosis may alsocause the aorta to become dilated and tor-tuous. The pulmonary artery may be en-larged in patients with left-to-right shunts,which cause increased pulmonary bloodflow, and in those with pulmonary hyper-tension of diverse causes (see Fig. 3.3). Iso-lated enlargement of the proximal left pul-monary artery is seen in some patients withpulmonic stenosis.

Pulmonary Manifestations ofHeart Disease

The appearance of the pulmonary vascula-ture reflects abnormalities of pulmonary ar-terial and venous pressures and pulmonaryblood flow. Increased pulmonary venouspressure, as occurs in left-heart failure, causes

increased vascular markings, redistributionof blood flow from the bases to the apices ofthe lungs (termed cephalization of vessels),pulmonary edema, the presence of abnor-mal septal lines (termed Kerley lines), andpleural effusions (Fig. 3.5). Blood flow redis-tribution appears as an increase in the num-ber or width of vascular markings at theapex. Interstitial and alveolar pulmonaryedemas produce opacity radiating from thehilar region bilaterally (known as a “butter-fly” or “bat-wing” pattern) and air broncho-grams, respectively. Kerley B lines, whichdepict fluid in interlobular spaces at the pe-riphery of the lung, result from interstitialedema. Pleural effusions cause blunting ofthe costodiaphragmatic angles.

Changes in pulmonary blood flow mayalso alter the appearance of the pulmonaryvessels. Focal oligemia (decreased flow) mayresult from pulmonary embolism or replace-ment of functioning lung tissue by emphy-sematous bullae. The finding of enlarged

Figure 3.3. Posteroanterior chest radiograph of a patient with pul-monary hypertension secondary to an atrial septal defect. Radiographicsigns of pulmonary hypertension include pulmonary artery dilatation (black arrows; compare with the appearance of left atrial appendage dilatation inFig. 3.2) and large central pulmonary arteries (white arrows) associated withsmall peripheral vessels (a pattern known as peripheral pruning).

Fig. 4

Fig. 5

10090-03_CH03.qxd 8/31/06 5:24 PM Page 49

central pulmonary arteries, but small periph-eral vessels (termed peripheral pruning), sug-gests pulmonary hypertension (see Fig. 3.3).

Table 3.1 summarizes the major radio-graphic findings in common forms of car-diac disease.

ECHOCARDIOGRAPHY

Echocardiography plays an essential role inthe diagnosis and serial evaluation of manycardiac disorders. It is safe, noninvasive, rel-atively inexpensive, and capable of accu-rately depicting a wide array of heart dis-eases. High-frequency (ultrasonic) waves,generated by a piezoelectric element, travelthrough body tissue and are reflected at in-terfaces where there are differences in theacoustic impedance of adjacent tissues. Thereflected waves return to the transducer and

cause mechanical deformation of the piezo-electric element. The machine measures thetime elapsed between the initiation and re-ception of the sound waves, thus enabling itto calculate the distance between the trans-ducer and each anatomic reflecting surface.Images are then constructed from these cal-culations.

Three types of echocardiographic modali-ties are generally performed: M-mode, two-dimensional (2-D), and Doppler imaging.Each type of imaging can be performed fromvarious locations. Most commonly, transtho-racic studies are performed, in which imagesare obtained by placing the transducer on thesurface of the chest. When greater structuraldetail is required, transesophageal imaging isperformed, as described later in this section.

M-mode echocardiography was the firstcardiac application of ultrasonography. It is

50 Chapter Three

Figure 3.4. Posteroanterior chest radiograph of a patient withaortic stenosis and insufficiency secondary to a bicuspid aorticvalve. In addition to poststenotic dilatation of the ascending aorta(black arrows), the transverse aorta (white arrow) is prominent, sug-gesting aortic insufficiency in addition to stenosis.

Tab. 1

10090-03_CH03.qxd 8/31/06 5:24 PM Page 50


A

Figure 3.5. Radiographs of patientswith congestive heart failure. Theseare anteroposterior views (which may ex-aggerate the size of the heart because itis further from the x-ray film), taken withportable x-ray machines at the bedside.A. Mild congestive heart failure. Pul-monary congestion is indicated by vas-cular redistribution from the bases to theapices of the lungs. The white spots la-beled “L” are electrocardiographic leadson the patient’s chest. B. Severe conges-tive heart failure. Increased pulmonaryvascular markings are present through-out the lung fields, along with peribron-chiolar cuffing (black arrow) and pleuraleffusion, which is indicated by bluntingof the costodiaphragmatic angle andtracking up the right lateral hemithorax(black arrowheads). The presence of in-terstitial and alveolar edema producesperihilar haziness and air bronchograms(open arrows), which occur when the ra-diolucent bronchial tree is contrastedwith opaque edematous lung tissue.

B

10090-03_CH03.qxd 8/31/06 5:24 PM Page 51

now used rarely by itself because it provideslimited data from one narrow ultrasonicbeam, and only structures along that singleline are displayed. M-mode techniques continue to be valuable for measurement of wall thicknesses and chamber diameters and for accurate timing of valve movements(Fig. 3.6).

In 2-D echocardiography, multiple ul-trasonic beams are transmitted through awide arc. The returning signals are inte-grated to produce two-dimensional imagesof the heart on a video monitor. As a result,this technique depicts anatomic relation-ships and defines the movement of cardiacstructures relative to one another. The widefields of view enhance the ability of 2-Dechocardiograms to detect and display walland valve motion and intracardiac massessuch as vegetations, thrombi, and tumors.

Each two-dimensional plane (Fig. 3.7) de-lineates only part of a given cardiac struc-ture. Optimal evaluation of the entire heartis achieved by using combinations of views.In transthoracic echocardiography (TTE), inwhich the transducer is placed against the

patient’s skin, these include the standardparasternal long axis, parasternal short axis,apical four-chamber, apical two-chamber,apical three-chamber (also known as apicallong axis), and subcostal views. The paraster-nal long axis view is recorded with the trans-ducer in the third or fourth intercostal spaceto the left of the sternum. This view is par-ticularly useful for evaluation of the leftatrium, mitral valve, left ventricle, and leftventricular outflow tract, which includesthe aortic valve and adjacent interventricu-lar septum. To obtain parasternal short axisviews, the transducer is rotated 90° from itsposition for the parasternal long axis view.The short axis images depict transverseplanes of the heart. Several different levelsare imaged to assess the aortic valve, mitralvalve, and left ventricular wall motion.

Apical TTE views are produced when thetransducer is placed at the point of maximalcardiac impulse. The apical four-chamber viewevaluates the mitral and tricuspid valves aswell as the atrial and ventricular chambers,including the motion of the lateral, septal,and apical left ventricular walls. The apical

52 Chapter Three

TABLE 3.1. Chest Radiography of Common Cardiac Disorders

Disorder Finding

Congestive heart failure • Vascular redistribution from bases to apices of the lungs• Interstitial and alveolar edema• Perihilar haziness• Peribronchiolar cuffing• Air bronchograms• Pleural effusions

Pulmonic valve stenosis • Poststenotic dilatation of pulmonary artery• Normal cardiac chamber sizes• Clear lung fields

Aortic valve stenosis • Poststenotic dilatation of ascending aorta• Normal cardiac chamber sizes (until heart fails)• Normal pulmonary vasculature

Aortic regurgitation • Left ventricular enlargement• Dilated aorta

Mitral stenosis • Enlarged left atrium• Small aorta (if chronic low cardiac output)• Signs of pulmonary venous congestion

Mitral regurgitation • Left atrial dilatation• Left ventricular dilatation• If severe:• Right ventricular dilatation• Signs of congestive heart failure

Fig. 6

Fig. 7

10090-03_CH03.qxd 8/31/06 5:24 PM Page 52


two-chamber view shows only the left side ofthe heart, and it depicts movement of theanterior, inferior, and apical walls.

In some patients, such as those with ob-structive airways disease, the parasternal andapical views do not provide adequate depic-tion of cardiac structures because of signalattenuation caused by excessive underlyingair. In such patients, the subcostal view, inwhich the transducer is placed inferior to

the rib cage, may provide a better ultrasonicwindow, allowing visualization of all fourcardiac chambers.

Doppler imaging evaluates blood flowdirection and velocity and turbulence. Addi-tionally, it permits estimation of pressuregradients within the heart and great vessels.Doppler studies are based on the physicalprinciple that waves reflected from a movingobject undergo a frequency shift accordingto the moving object’s velocity relative tothe source of the waves. Color flow mappingconverts the Doppler signals to an arbitrarilychosen scale of colors that represent direc-tion, velocity, and turbulence of blood flowin a semiquantitative way. The colors are su-perimposed on 2-D images and show the lo-cation of stenotic and regurgitant valvularlesions and of abnormal communicationswithin the heart and great vessels. For exam-ple, Doppler echocardiography in a patientwith mitral regurgitation shows a jet of ret-rograde flow into the left atrium during sys-tole (Fig. 3.8).

Sound frequency shifts are converted intoblood flow velocity measurements (auto-matically calculated by the echo machine),by the relationship

in which v equals the blood flow velocity(m/sec); fs, the Doppler frequency shift (kHz);c, the velocity of sound in body tissue (m/sec);fo, the frequency of the sound pulse emittedfrom the transducer (MHz); and θ, the anglebetween the transmitted sound pulse and themean axis of blood flow.

Transesophageal echocardiography(TEE) uses a miniaturized transducermounted at the end of a modified endoscopeto transmit and receive ultrasound wavesfrom within the esophagus, thus producingvery clear images of the neighboring cardiacstructures (Fig. 3.9) and much of the tho-racic aorta. Modern probes permit multipla-nar imaging and Doppler interrogation aswell. TEE is particularly helpful in the as-sessment of aortic and atrial abnormalities,conditions that are less well visualized byconventional transthoracic echo imaging. For

vfs c

fO

= ( )•

2 cosθ

1 2 3 4

1

23 4

Figure 3.6. Schematic diagram of the heart in theparasternal long axis view (upper drawing). Below itis the electrocardiogram and M-mode echocardiogramgenerated by changes in the direction of the transducerfrom position 1 to position 4. Chamber sizes and wallthicknesses can be measured along the vertical axis. Notethe sequential thickening and thinning of the ventricularwalls, corresponding to systolic contraction and diastolicrelaxation respectively, as well as movement of the aorticand mitral valves. AMV, anterior mitral valve leaflet; AO,aortic root; AV, aortic valve; CW, chest wall; LA, leftatrium; LV, left ventricle; LVW, left ventricular wall; PMV,pulmonic valve; PPM, posterior papillary muscle region;RV, right ventricle; RVW, right ventricular wall; S, ster-num; T, echocardiographic transducer. (Reprinted withpermission from Come PC. Echocardiography in diagno-sis and management of cardiovascular disease. ComprTher 1980;6:7–17. By permission of International Pub-lishing Group, Cleveland, OH.)

AQ2

AQ3

Fig. 8

Fig. 9

10090-03_CH03.qxd 8/31/06 5:24 PM Page 53

example, TEE is more sensitive than transtho-racic echo for the detection of thrombus with-in the left atrial appendage (Fig. 3.10), whichis of great importance in patients who re-quire electrical cardioversion of atrial fibril-lation without prior anticoagulation (seeChapter 12). The proximity of the esophagusto the heart makes TEE imaging particularlyadvantageous in patients for whom trans-thoracic echo images are unsatisfactory (e.g.,those with chronic obstructive lung disease).

TEE is also advantageous in the evaluationof patients with prosthetic heart valves. Dur-ing standard transthoracic imaging, artificial

mechanical valves reflect a large portion of ul-trasound waves, thus interfering with visual-ization of more-posterior structures (termedacoustic shadowing). TEE aids visualization ofthe posterior chambers in such patients andis therefore the most sensitive noninvasivetechnique for evaluating perivalvular leaks.

TEE is commonly used to evaluate pa-tients with cerebral ischemia of unexplainedetiology, because it can identify cardiovascu-lar causes of emboli with a high sensitivity.These etiologies include intracardiac thrombior tumors, atherosclerotic debris within theaorta, and valvular vegetations. It is also

54 Chapter Three

RV

Tricuspidvalve

Mitralvalve

RA LA

LV

A

B

C

RV

LV

Ao

LA

LV posterior wall

Interventricular septum

Aorticvalve

Mitralvalve

LV

RV

Figure 3.7. Transthoracic two-dimensional echocardiographic views. A. Parasternal long axis view. B. Parasternal short axis view. Notice that the left ven-tricle appears circular in this view, while the right ventricle is crescent shaped. C. Apical four-chamber view. Ao, aorta; LA, left atrium; LV, left ventricle; RA, rightatrium; RV, right ventricle. (Modified from Sahn DJ, Anderson F. Two-DimensionalAnatomy of the Heart. New York: John Wiley & Sons, 1982.)

Fig. 10

10090-03_CH03.qxd 8/31/06 5:24 PM Page 54


right-to-left heart blood flow, or in the pres-ence of an intrapulmonary shunt, bubblesof contrast will appear in the left-sidedchambers as well. Newer perfluorocarbon-based contrast agents have been developedwith sufficiently small particle size to inten-tionally pass through the pulmonary circula-tion. These agents are sometimes used toopacify the left ventricular cavity and, viathe coronary arteries, the myocardium, en-abling superior assessment of LV contractionand myocardial perfusion.

Echocardiographic techniques can identifyvalvular lesions, complications of coronaryartery disease, septal defects, intracardiacmasses, cardiomyopathy, ventricular hyper-trophy, pericardial disease, aortic disease,and congenital heart disease. Evaluation in-cludes assessment of cardiac chamber sizes,wall thicknesses, wall motion, valvular func-tion, blood flow, and intracardiac hemody-namics. A few of these topics are highlightedin the sections that follow.

Ventricular Assessment

Two-dimensional echocardiography assessesleft ventricular systolic function by comput-ing fractional changes between end-diastolicand end-systolic measurements. Left ven-tricular width, area, length, and volume indiastole and systole are used to assess con-tractile function and calculate the ventri-cular ejection fraction. Beyond assessingoverall left ventricular systolic function,two-dimensional echocardiography depictsregional ventricular wall motion abnormal-ities, a common sign of coronary artery dis-ease. Right ventricular systolic function isgenerally assessed qualitatively because theright ventricle does not lend itself as easilyto geometric modeling as does the left ven-tricle. Echocardiography is useful in evalu-ating ventricular wall thickness and mass,which are important in patients with hyper-tension, aortic stenosis, and hypertrophiccardiomyopathy (Fig. 3.11).

Diastolic dysfunction (e.g., caused by is-chemic disease, ventricular hypertrophy, orrestrictive cardiomyopathy; see Chapter 9)can also be evaluated by Doppler techniques.

Figure 3.8. Doppler color flow mapping (reproducedin gray tones) of mitral regurgitation (MR). TheDoppler image, recorded in systole, is superimposed onan apical view of the left ventricle (LV), left atrium (LA),and mitral valve (short arrow). The retrograde flow of MRinto the LA is indicated by the long arrow.

highly sensitive and specific for the detec-tion of aortic dissection.

In the operating room, TEE permits im-mediate evaluation after surgical repair ofcongenital and valvular lesions. In addition,imaging of ventricular wall motion can iden-tify periods of myocardial ischemia duringhigh-risk surgery.

Contrast echocardiography is frequentlyused in the evaluation of congenital heartdisease because it is highly sensitive for thedetection of abnormal intracardiac shunts.In this technique, often called a “bubblestudy,” an echocardiographic contrast agent(e.g., agitated saline) is rapidly injected intoa peripheral (usually a brachial) vein. Usingstandard echocardiographic imaging, thecontrast can be visualized passing throughthe cardiac chambers. Normally, there israpid opacification of the right-sided cham-bers, but because the contrast is filtered out(harmlessly) in the lungs, it does not reachthe left-sided chambers. However, in the pres-ence of an intracardiac shunt with abnormal

Fig. 11

10090-03_CH03.qxd 8/31/06 5:24 PM Page 55

56 Chapter Three

RV

RV

RV

RA

RA

LV

LV

LA

LA

Basal Short Axis

Transgastric

IVC

Four Chambers

Basal Short Axis 25–30 cm from incisors

Four Chambers 30 cm

Transgastric 35–40 cm

Figure 3.9. Transesophageal echocardiographic views. IVC, inferior vena cava; LA,left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle. (Courtesy of Jane Freed-man, MD, Boston University Medical Center, MA.)

LA

LAA

Thrombus

A BFigure 3.10. Echocardiographic imaging of an intracardiac thrombus. A. Transesophagealechocardiographic image demonstrates thrombus within the left atrial appendage. (Courtesy of ScottStreckenbach, MD, Massachusetts General Hospital, Boston.) B. Schematic drawing of same image.LA, left atrium; LAA, left atrial appendage.

10090-03_CH03.qxd 8/31/06 5:24 PM Page 56


For example, Doppler tissue imaging is amodality that can readily record the maxi-mum velocity at which the mitral valve an-nulus springs back from the left ventricle inearly diastole, as a measure of the chamber’sability to relax. Standard Doppler measure-ments of flow velocity across the mitral valvein early, compared with late, diastole alsoprovide important information about ven-tricular diastolic function.

Valvular Lesions

Echocardiography can accurately determineunderlying causes of valvular abnormalities,and Doppler imaging permits quantitationof the degree of valvular stenosis and regurgi-tation. The pressure gradient across a stenoticvalve can be calculated from the maximumblood flow velocity (v) measured distal to thevalve, using the simplified Bernoulli equation:

As an example, if the peak velocity recordeddistal to a stenotic aortic valve is 4 m/sec,

Pressure gradient = ×4 2v

then the calculated peak pressure gradientacross the valve = 4 × 42 = 64 mm Hg.

Other calculations permit noninvasivedetermination of the cross-sectional area ofstenotic valves. For instance, the continuityequation is often used to calculate aortic valvearea. This equation assumes that blood flow(F, expressed in cc/sec) is the same at the aor-tic valve orifice (AV) as at a neighboring po-sition along the flow stream (e.g., in the leftventricular outflow tract [LVOT]).

As shown in Figure 3.12, blood flow at anyposition along a flow stream can also be ex-pressed as the product of the Doppler veloc-ity (V, in cm/sec) and cross-sectional area (A,in cm2) at that level. If location 1 in Figure3.12 represents a position in the LVOT andlocation 2 represents the aortic valve, then

The cross-sectional area of the LVOT (ALVOT)is calculated simply as π(d/2)2, in which drepresents the LVOT diameter, measured by

A V A VLVOT LVOT AV AV× = ×

F FLVOT AV=

Figure 3.11. Left ventricular outflow tract (LVOT) obstruction in hypertrophic cardiomyopathy. No-tice that the interventricular septum (S) is thicker and more echogenic than the posterior wall (P). A. Beforeventricular contraction, the LVOT is only slightly narrowed. B. During contraction, the rapidly flowing bloodthrough the LVOT incites a Venturi effect and abnormally draws the mitral valve apparatus anteriorly towardthe hypertrophied septum (arrow), creating a functional obstruction. LA, left atrium. LV, left ventricle.

Fig. 12

10090-03_CH03.qxd 8/31/06 5:24 PM Page 57

echocardiography, usually from the para-sternal long axis view. The velocities (VLVOT

and VAV) are measured by Doppler interro-gation, from the apical four-chamber view.The equation can then be solved for the aor-tic valve area (AAV):

Color Doppler analysis provides a quali-tative assessment of the severity of regurgi-tant valve lesions. In mitral regurgitation(see Fig. 3.8), for example, the ratio of the re-gurgitant jet color Doppler area to the entireleft atrial area has traditionally been used toclassify the regurgitation as mild, moderate,or severe. More quantitative evaluation ofmitral regurgitation can now be performedby what is known as the proximal isoveloc-ity surface area (PISA) method. This tech-nique uses advanced color Doppler tech-niques to calculate the regurgitant volumeand effective regurgitant orifice area, twovalues that predict clinical outcomes in pa-tients with chronic mitral regurgitation.

Coronary Artery Disease

Echocardiography demonstrates ventricularwall motion abnormalities associated withinfarcted or transiently ischemic myocar-

AA V

VAVLVOT LVOT

AV

= ×( )

dium. The location and degree of abnormalsystolic contraction and decreased systolicwall thickening indicate the extent of an in-farction and implicate the responsible coro-nary artery. Infarct size measured by 2-Dechocardiography correlates well with othermethods of quantification, such as radioiso-tope scanning and positron emission to-mography (described later in the chapter).Echocardiography is also used to detect com-plications of acute myocardial infarction,including intraventricular thrombus for-mation, papillary muscle rupture, valvulardysfunction, ventricular septal rupture, andaneurysm formation.

Although echocardiography can depictthose consequences of coronary artery dis-ease, transthoracic echo resolution is usuallyinsufficient to directly image the coronaryarteries themselves. In a few patients, themost proximal portions of the coronary ar-teries can be delineated.

Stress echocardiography can be used todiagnose the presence of coronary arterydisease. This technique assesses the devel-opment of left ventricular regional wall mo-tion abnormalities induced by exercise orafter the infusion of specific pharmacologicagents, such as dobutamine, dipyridamole,or adenosine (see Chapter 6). Reversible myo-cardial ischemia is recognized by a stress-induced wall motion abnormality.

Cardiomyopathy

Cardiomyopathies are heart muscle disordersthat occur in three forms: dilated, hypertro-phic, and restrictive (see Chapter 10). Echo-cardiography can often distinguish amongthese and permits assessment of the severityof systolic and diastolic dysfunction. For ex-ample, Figure 3.11 depicts the asymmetri-cally thickened ventricular walls found inclassic hypertrophic cardiomyopathy.

Pericardial Disease

Two-dimensional echocardiography canidentify abnormalities in the pericardial cav-ity (e.g., excessive pericardial fluid, fibrinousmaterial, tumor, and clot). Tamponade and

58 Chapter Three

1 2

A1

A1 x V1 = A2 x V2

A2

V1 V2

Figure 3.12. The continuity equation. Within a closedflow stream, the volume rate of flow at any point (calcu-lated as the cross-sectional area at that site multiplied bythe maximum flow velocity at the same location) is equalto the volume rate of flow at sequential points. Thus,cross-sectional area and velocity at any location are in-versely proportional to one another. Here, location 2 isnarrower than location 1. Therefore the velocity at loca-tion 2 must be greater for the same volume to pass perunit time.

10090-03_CH03.qxd 8/31/06 5:24 PM Page 58


constrictive pericarditis, the main functionalconsequences of pericardial disease (seeChapter 14), are associated with particularechocardiographic abnormalities. In tam-ponade, the increased intrapericardial pres-sure compresses the cardiac chambers and

results in diastolic “collapse” of the rightatrium, right ventricle, and sometimes (inmore extreme cases) the left-sided chambers(Fig. 3.13). Constrictive pericarditis is associ-ated with increased thickness or reflective-ness of the pericardial echo, abnormal pat-

ECGECG

Figure 3.13. Echocardiographic studies of a patient with a peri-cardial effusion causing cardiac tamponade. The apical four-chamber two-dimensional image (upper panel) shows a large peri-cardial effusion (PE) and inward collapse (white arrowheads) of theright atrium (RA) and left atrium (LA). M-mode tracings (lower pan-els) indicate early diastolic collapse of the right ventricular wall (blackarrows). AoV, aortic valve; ECG, electrocardiogram; LV, left ventricle;MV, mitral valve; RV, right ventricle; RVOT, right ventricular outflowtract. (Reprinted with permission from Cunningham MJ, Safian RD,Come PC, et al. Absence of pulsus paradoxus in a patient with cardiactamponade and coexisting pulmonary artery obstruction. Am J Med1987;83:973–976.)

Fig. 13

10090-03_CH03.qxd 8/31/06 5:24 PM Page 59

terns of diastolic left ventricular wall mo-tion, alterations in pulmonary venous flowpatterns, and exaggerated changes in mitraland tricuspid valve inflow velocities duringrespiration.

Table 3.2 summarizes the salient echocar-diographic features of common cardiac dis-eases.

CARDIAC CATHETERIZATION

In the diagnosis of many cardiovascular ab-normalities, intravascular catheters are in-serted to measure pressures in the heartchambers, to determine cardiac output andvascular resistances, and to inject radiopaquematerial to examine heart structures and

blood flow. In 1929, Werner Forssmann per-formed the first cardiac catheterization, onhimself, thus ushering in the era of invasivecardiology. Much of what is known about thepathophysiology of valvular heart disease andcongestive heart failure comes from decadesof subsequent hemodynamic research in thecardiac catheterization laboratory.

Measurement of Pressure

Before catheterization of an artery or vein,the patient is mildly sedated, and a localanesthetic is used to numb the skin site ofcatheter entry. The catheter, attached to apressure transducer outside the body, is thenintroduced into the appropriate blood ves-

60 Chapter Three

TABLE 3.2. Echocardiography in Common Cardiac Disorders

Disorder Finding

Valvular lesionsMitral stenosis • Enlarged left atrium

• Thickened mitral valve leaflets• Decreased movement and separation of mitral valve leaflets• Decreased mitral valve orifice

Mitral regurgitation • Enlarged left atrium (if chronic)• Enlarged left ventricle (if chronic)• Systolic flow from left ventricle into left atrium (by Doppler)

Aortic stenosis • Thickened aortic valve cusps• Decreased valve orifice• Increased left ventricular wall thickness

Aortic regurgitation • Enlarged left ventricle• Abnormalities of aortic valve or aortic root

Left ventricular functionMyocardial infarction and • Hypokinetic, dyskinetic, or akinetic ventricular wall motioncomplications • Decreased ejection fraction

• Thrombus within left ventricle• Aneurysm of ventricular wall• Septal rupture (abnormal Doppler flow)• Papillary muscle rupture• Pericardial effusion

CardiomyopathiesDilated • Enlarged ventricular chamber sizes

• Normal ventricular wall thicknesses• Decreased systolic contraction

Hypertrophic • Normal or decreased ventricular chamber sizes• Increased ventricular wall thickness• Diastolic dysfunction (assessed by Doppler)

Restrictive • Normal or decreased ventricular chamber sizes• Enlarged atria• Increased ventricular wall thickness• Ventricular contractile function often normal• Diastolic dysfunction (assessed by Doppler)

Tab. 2

10090-03_CH03.qxd 8/31/06 5:24 PM Page 60


sel. To measure pressures in the right atrium,right ventricle, and pulmonary artery, acatheter is usually inserted into a femoral,brachial, or jugular vein. Pressures in theaorta and left ventricle are measured viacatheters inserted into a brachial or femoralartery. Once in the blood vessel, the catheteris guided by fluoroscopy (x-ray images) tothe area of study, where pressure measure-ments are made. Figure 3.14 depicts normalintracardiac and intravascular pressures.

The measurement of right-heart pressuresis performed with a specialized balloon-tipped catheter (a common version of whichis known as the Swan-Ganz catheter) that is advanced through the right side of theheart with the aid of normal blood flow.The catheter is typically inserted percuta-neously into a peripheral vein (e.g., femoral,brachial, or internal jugular) and advancedtoward the chest. When it reaches a vein of

suitable size (e.g., the inferior or superiorvena cava), the balloon at the catheter tip ismanually inflated so that venous return ofblood helps direct the catheter into the right-sided heart chambers and into the pulmo-nary artery. As it travels through the rightside of the heart, recorded pressure measure-ments identify the catheter tip’s position (seeBox 3.1).

The normal right atrial (RA) pressure de-monstrates three positive deflections (seeFig. 2.1): the a wave reflects right atrial con-traction at the end of diastole, the c wave re-sults from bulging of the tricuspid valve to-ward the right atrium as it closes in earlysystole, and the v wave represents passivefilling of the right atrium from the systemicveins during systole, when the tricuspidvalve is closed. The negative deflection thatfollows the c wave is known as the x descent,and the negative deflection after the v wave

RA

2–8

RV15–30

PA15–30

2–8 4–12

Lungs

PCW2–10

Aorta

LA

2–10

LV100–1403–12 60–90

Aorta

PCW

RA

RV

LV

LAPA

100–140

2–10

2–10

2–8

15–302–8

100–1403–12

100–14060–90

15–304–12

Figure 3.14. Diagrams indicating normal pressures in the cardiac chambersand great vessels. The top figure shows the normal anatomic relationship of thecardiac chambers and great vessels, whereas the figure on the bottom shows a sim-plified schematic to clarify the pressure relationships. Numbers indicate pressures inmm Hg. LA, left atrial mean pressure; LV, left ventricular pressure; PA, pulmonaryartery pressure; PCW, pulmonary capillary wedge mean pressure; RA, right atrialmean pressure; RV, right ventricular pressure.

Fig. 14

Box 1

10090-03_CH03.qxd 8/31/06 5:24 PM Page 61

62 Chapter Three

Box 3.1 Intracardiac Pressure Tracings

When a catheter is inserted into a systemic vein and advanced into the right side of theheart, each cardiac chamber produces a characteristic pressure tracing. It is important todistinguish these tracings from one another to localize the position of the catheter tip andto derive appropriate physiologic information.

ECG

20

10

a cx

v

a v

yPre

ssur

e (m

m H

g)

Right ventricle

Right atrium

Pulmonary artery

Time

Pulmonary capillary wedge

is called the y descent. Often the a and cwaves merge so that only two major positivedeflections are seen. In patients with atrialfibrillation (see Chapter 12), the a wave isabsent because there is no organized leftatrial contraction at the end of diastole.

As the catheter is advanced into the rightventricle (RV), a dramatic increase in systolicpressure is seen. The RV systolic waveform ischaracterized by a rapid upstroke and down-stroke. In diastole, there is a gradual contin-uous increase in RV pressure as the chamberfills with blood.

As the catheter is moved forward into thepulmonary artery (PA), the systolic pressure re-mains the same as that in the RV (as long asthere is no obstruction to RV outflow, suchas pulmonic valve stenosis). However, threecharacteristics of the tracing indicate entryinto the pulmonary artery: (1) the PA dias-tolic pressure is higher than that of the RV;(2) the descending systolic portion of the PAtracing inscribes a dicrotic wave, a smalltransient pressure increase that occurs afterthe systolic peak and is related to pulmonicvalve closure; and (3) the diastolic portion ofthe PA tracing is downsloping comparedwith the upsloping RV diastolic pressure.

Further advancement of the catheter intoa branch of the pulmonary artery results in

the pulmonary capillary wedge (PCW) tracing,which reflects the left atrial pressure (see Fig.3.14). Its characteristic shape is similar to theRA tracing, but the pressure values are usuallyhigher and the tracing is often less clear (withthe c wave not observed) because of dampedtransmission through the capillary vessels.

Right Atrial Pressure

Right atrial pressure is equal to the centralvenous pressure (estimated by the jugularvenous pressure on physical examina-tion) because no obstructing valves impedeblood return from the veins into the rightatrium. Similarly, right atrial pressure nor-mally equals right ventricular pressure indiastole because the right heart functionsas a “common chamber” when the tricus-pid valve is open. The mean right atrialpressure is reduced when there is intra-vascular volume depletion. It is elevatedin right ventricular failure, right-sided val-vular disease, and cardiac tamponade (inwhich the cardiac chambers are surroundedby high-pressure pericardial fluid; see Chap-ter 14).

Certain abnormalities cause characteris-tic changes in individual components of theright atrial (and therefore jugular venous)

10090-03_CH03.qxd 8/31/06 5:24 PM Page 62


pressure (Table 3.3). For example, a promi-nent a wave is seen in tricuspid stenosis andright ventricular hypertrophy. In these con-ditions, the right atrium contracts vigor-ously against the obstructing tricuspid valveor stiffened right ventricle, respectively, gen-erating a prominent pressure wave. Simi-larly, amplified “cannon” a waves may beproduced by conditions of atrioventriculardissociation (see Chapter 12), when the rightatrium contracts against a closed tricuspidvalve. A prominent v wave is observed in tri-cuspid regurgitation because normal rightatrial filling is augmented by the regurgitatedblood in systole.

Right Ventricular Pressure

Right ventricular systolic pressure is increasedby pulmonic valve stenosis or pulmonary hy-pertension. Right ventricular diastolic pres-

sure increases when the right ventricle is sub-jected to pressure or volume overload andmay be a sign of right-heart failure.

Pulmonary Artery Pressure

Elevation of systolic and diastolic pulmonaryartery pressures occurs in three conditions:(1) left-sided heart failure; (2) parenchymallung disease (e.g., chronic bronchitis or end-stage emphysema); and (3) pulmonary vas-cular disease (e.g., pulmonary embolism,primary pulmonary hypertension, or acuterespiratory distress syndrome). Normally, thepulmonary artery diastolic pressure is equiv-alent to the left atrial pressure because of thelow resistance of the pulmonary vasculaturethat separates them. If the left atrial pressurerises because of left-sided heart failure, bothsystolic and diastolic pulmonary artery pres-sures increase in an obligatory manner to

TABLE 3.3. Causes of Increased Intracardiac Pressures

Chamber and Measurement Causes

Right atrial pressure • Right ventricular failure• Cardiac tamponade

a wave • Tricuspid stenosis• Right ventricular hypertrophy• Atrioventricular dissociation

v wave • Tricuspid regurgitation• Right ventricular failure

Right ventricular pressureSystolic • Pulmonic stenosis

• Right ventricular failure• Pulmonary hypertension

Diastolic • Right ventricular failure• Cardiac tamponade• Right ventricular hypertrophy

Pulmonary artery pressureSystolic and diastolic • Pulmonary hypertension

• Left-sided CHF• Chronic lung disease• Pulmonary vascular disease

Systolic only • Increased flow (L ∅ R shunt)Pulmonary artery wedge pressure • Left-sided CHF

• Mitral stenosis or regurgitation• Cardiac tamponade

a wave • Left ventricular hypertrophyv wave • Mitral regurgitation

• Ventricular septal defect

CHF, congestive heart failure.

Tab. 3

10090-03_CH03.qxd 8/31/06 5:24 PM Page 63

maintain forward flow through the lungs.This situation leads to “passive” pulmonaryhypertension.

In certain conditions, however, pulmo-nary vascular resistance becomes abnormallyhigh, causing pulmonary artery diastolicpressure to be elevated compared with leftatrial pressure. For example, pulmonary vas-cular obstructive disease may develop as acomplication of a chronic left-to-right car-diac shunt, such as an atrial or ventricularseptal defect (see Chapter 16).

Pulmonary Artery Wedge Pressure

If a catheter is advanced into the right orleft pulmonary artery, its tip will ultimatelyreach one of the small pulmonary arterybranches and temporarily occlude forwardblood flow beyond it. During that time, acolumn of stagnant blood stands betweenthe catheter tip and the portions of the pul-monary capillary and pulmonary venoussegments distal to it (Fig. 3.15). That col-umn of blood acts as an “extension” of thecatheter, and the pressure recorded throughthe catheter reflects that of the downstreamchamber—namely, the left atrium. Such apressure measurement is termed the pul-monary artery wedge pressure or pulmonarycapillary wedge pressure (PCW) and closelymatches the left atrial pressure in most peo-ple. Furthermore, while the mitral valve isopen during diastole, the pulmonary venous

bed, left atrium, and left ventricle normallyshare the same pressures. Thus, the PCW canbe used to estimate the left ventricular dias-tolic pressure, a measurement of ventricularpreload (see Chapter 9). As a result of this im-portant feature, monitoring of PCW is oftenuseful in managing critically ill patients inthe intensive care unit.

Elevation of the mean PCW is seen inleft-sided heart failure and in mitral steno-sis or regurgitation. The individual compo-nents of the PCW tracing can also becomeabnormally high. The a wave may be in-creased in conditions of decreased left ven-tricular compliance, such as left ventricularhypertrophy or acute myocardial ischemia.The v wave is greater than normal whenthere is increased left atrial filling duringventricular contraction, as in mitral regur-gitation.

Measurement of Blood Flow

Cardiac output is generally measured by either the thermodilution method or the Ficktechnique. In the thermodilution method,saline of a known temperature is injectedrapidly into the right heart via a catheterside hole located a specific distance proxi-mal to the tip of the catheter. At the tip, athermistor registers the surrounding tem-perature in the pulmonary artery, which istransiently altered by the injected saline.The cardiac output is electronically calcu-

64 Chapter Three

LAPA

A pulmonaryvein

Pulmonarycapillaries

Catheter tipoccludes branchof pulmonary artery

Pulmonaryartery catheter

Shaded area represents“column of blood” betweencatheter tip and LA

Figure 3.15. Diagram of a pulmonary artery catheter inserted into a branch of the pul-monary artery (PA). Flow is occluded in the arterial, arteriolar, and capillary vessels beyond thecatheter; thus, these vessels act as a conduit that transmits the left atrial (LA) pressure to thecatheter tip.

Fig. 15

10090-03_CH03.qxd 8/31/06 5:24 PM Page 64


lated from the slope of the decay of the tem-perature change.

The Fick method is derived from the prin-ciple that consumption of oxygen by tissuesis related to the O2 content removed fromblood as it flows through the capillary bed.

Or, in other terms:

in which the arteriovenous O2 (AVO2) dif-ference equals the difference in oxygen con-tent between the arterial and venous com-partments. Total body oxygen consumptioncan be determined by analyzing expired airfrom the lungs, and arterial and venous O2

content is measured in blood samples. By re-arranging the terms, the cardiac output canbe calculated.

For example, if the arterial blood in a nor-mal adult contains 190 mL of O2 per literand the venous blood contains 150 mL of O2 per liter, the arteriovenous difference is40 mL of O2 per liter. If this patient has ameasured O2 consumption of 200 mL/min,the cardiac output is 5 L/min.

In many forms of heart disease, the car-diac output is lower than normal. In that sit-uation, the total body oxygen consumptiondoes not change significantly; however, agreater percentage of O2 is extracted per vol-ume of circulating blood by the metaboliz-ing tissues. The result is a lower-than-normal venous O2 content and therefore anincreased arteriovenous O2 difference. In theexample, if the patient’s venous blood O2

content fell to 100 mL/L, the arteriovenousO2 difference would be increased to 90 mL/Land the calculated cardiac output would bereduced to 2.2 L/min.

Cardiac outputO consumptionAVO differenc

= 2

2 ee

O consumption AVO differenceCardiac out

2 2=× pput

O consumption O content removedFlow

mL O

2 2=×

22 2

min min⎛⎝⎜

⎞⎠⎟ = ⎛

⎝⎜⎞⎠⎟ ×mL O

mL bloodmL blood⎛⎛

⎝⎜⎞⎠⎟

Because the normal range of cardiac out-put varies with a patient’s size, it is commonto report the cardiac index, which is equalto the cardiac output divided by the patient’sbody surface area (normal range of cardiacindex = 2.6 − 4.2 L/min per square meter).

Calculation of Vascular Resistance

Once pressures and cardiac output have beendetermined, pulmonary and systemic vascu-lar resistances can be calculated from thefollowing formulas:

PVR, pulmonary vascular resistance (dynes-sec-cm−5)

MPAP, mean pulmonary artery pressure (mm Hg)

LAP, mean left atrial pressure (mm Hg)CO, cardiac output (L/min)

SVR, systemic vascular resistance (dynes-sec-cm−5)

MAP, mean arterial pressure (mm Hg)RAP, mean right atrial pressure (mm Hg)CO, cardiac output (L/min)

The normal PVR ranges from 20 to 130 dynes-sec-cm−5. The normal SVR is 700 to 1,600dynes-sec-cm−5.

Contrast Angiography

This technique, performed at the time ofcatheterization, uses radiopaque contrast ma-terial to visualize regions of the cardiovascu-lar system. A catheter is introduced into anappropriate vessel and guided under fluo-roscopy to the site where the contrast willbe injected. Following administration ofthe contrast agent, x-rays are transmittedthrough the area of interest. A single expo-sure produces one film image, whereas a se-ries of x-ray exposures is recorded to producea “motion picture,” termed a cineangiogram(often simply called a cine).

SVRMAP RAP

CO= − × 80

PVRMPAP LAP

CO= − × 80

10090-03_CH03.qxd 8/31/06 5:24 PM Page 65

A specialized type of contrast angiogra-phy, termed digital subtraction angiogra-phy (DSA), was developed to provide a clearimage using less contrast material. In thistechnique, a computer processes the digital-ized x-ray images and subtracts the back-ground of soft tissue and bone, thus enhanc-ing the image of the blood vessel or chamberinto which contrast material was injected.DSA has advantages over conventional an-giography: smaller catheters may be used,the amount of contrast agent required maybe lower, and better image quality is usuallyachieved.

Selective injection of contrast materialinto the heart chambers is used to identifyvalvular insufficiency, abnormal wall thick-ening, intracardiac shunts, thrombi withinthe heart, and congenital malformations,and also to measure ventricular contractilefunction. To image the right heart cham-bers, injection is made through a catheterinserted into the inferior or superior venacava, the right atrium, or the right ventricle.The left side of the heart can be imaged bycontrast injection through a catheter ad-vanced into the left ventricle (Fig. 3.16).

A widespread application of contrast in-jection is coronary artery angiography, used

to examine the location and severity of coro-nary atherosclerotic lesions. To maximize thetest’s sensitivity and reproducibility, each patient is imaged in several standard views. Ifa significant stenosis is detected, balloon angioplasty and stent placement can be per-formed, reopening the vessel (Figs. 3.17 and3.18; see Chapter 6).

Some risk is associated with catheterizationand contrast angiography. Complications are uncommon but include myocardial per-foration by the catheter, precipitation of ar-rhythmias and conduction blocks, damage tovessel walls, hemorrhage, dislodgement ofatherosclerotic plaques, and infection. Com-plicationsresulting fromthecontrast mediumitself include anaphylaxis and renal toxicity.

Table 3.4 summarizes the catheterizationfindings in common cardiac abnormalities.Therapeutic interventional catheterizationtechniques are described in Chapter 6.

NUCLEAR IMAGING

Heart function can be evaluated using in-jected, radioactively labeled tracers and γ-camera detectors. The resulting images re-flect the distribution of the tracers withinthe cardiovascular system. Nuclear tech-

66 Chapter Three

Figure 3.16. Left ventriculogram, in diastole (A) and systole (B) in the right anterior oblique projec-tion, from a patient with normal ventricular contractility. A catheter (arrow) is used to inject contrast intothe left ventricle (LV). The catheter can also be seen in the descending aorta (arrowhead). AO, aortic root.

Fig. 17-18

Tab. 4

10090-03_CH03.qxd 8/31/06 5:24 PM Page 66


BA

LMLM

LADLAD

LCX

Diagonalbranch

Septalperforators

Figure 3.17. Cardiac catheterization and stenting of a proximal left anterior descending artery (LAD)stenosis, shown in an anteroposterior cranial projection. A. When contrast agent is injected into the leftmain coronary artery (LM), the left circumflex artery (LCX) fills normally but the LAD is almost completely occludedat its origin (white arrow). B. After the stenosis is successfully stented, the LAD and its branches fill robustly. (Imagescourtesy of Jeffrey Popma, MD, Brigham and Women’s Hospital, Boston, MA.)

BA

Figure 3.18. Cardiac catheterization and stenting of right coronary artery (RCA) stenoses. Both images areobtained in the left anterior oblique (LAO) projection. A. The stenotic segment is located between the white arrows.B. After stenting, the caliber of the vessel and flow have improved. (Courtesy of Jeffrey Popma, MD, Brigham andWomen’s Hospital, Boston, MA.)

8

10090-03_CH03.qxd 8/31/06 5:24 PM Page 67

niques are used to assess myocardial perfu-sion, to image blood passing through theheart and great vessels, to localize and quan-tify myocardial ischemia and infarction,and to assess myocardial metabolism.

Assessment of Myocardial Perfusion

Ischemia and infarction resulting from coro-nary artery disease can be detected by myo-cardial perfusion imaging using various ra-dioisotopes, including compounds labeledthallium-201 (201Tl) and technetium-99m(99mTc). Currently, 99mTc-sestamibi and 99mTc-tetrofosmin are widely used. Both 201Tl- and99mTc-labeled compounds are sensitive for the detection of ischemic or scarred myo-cardium, but each compound has certainadvantages. For example, the 99mTc-labeledagents provide better image quality and aresuperior for detailed single photon emis-sion computed tomography (SPECT, as inFig. 3.19). Conversely, superior detection ofmyocardial cellular viability has been demon-strated with 201Tl imaging.

In the case of 201Tl imaging, the radioiso-tope is injected intravenously while the pa-tient is exercising on a treadmill or stationarybicycle. Because thallium is a potassium ana-logue, it enters normal myocytes, a processthought to be partially governed by thesodium-potassium ATPase pump. The intra-cellular concentration of thallium, estimatedby the density of the image, depends on vas-cular supply (perfusion) and membranefunction (tissue viability). In the normalheart, the radionuclide scan shows a ho-

mogenous distribution of thallium in themyocardial tissue. Conversely, myocardialregions that are scarred (by previous infarc-tion) or have reduced perfusion during ex-ercise (i.e., transient myocardial ischemia)do not accumulate as much thallium as nor-mal heart muscle. Consequently, these areaswill appear on the thallium scan as light or“cold” spots.

When evaluating for myocardial ischemia,an initial set of images is taken right after ex-ercise and 201Tl injection. Well-perfused myo-cardium will take up more tracer than is-chemic or infarcted myocardium at thistime. Delayed images are acquired severalhours later, because 201Tl accumulation doesnot remain fixed in myocytes. Rather, con-tinuous redistribution of the isotope occursacross the cell membrane. After 3 to 4 hoursof redistribution, when additional imagesare obtained, all viable myocytes will haveequal concentrations of 201Tl. Consequently,any uptake abnormalities on the initial pos-texercise scan that were caused by myocar-dial ischemia will have resolved (“filled in”)on the delayed scan (and are therefore termed“reversible” defects), and those representinginfarcted or scarred myocardium will persist ascold spots.

Of note, some myocardial segments thatdemonstrate persistent 201Tl defects on bothstress and redistribution imaging are falselycharacterized as nonviable, scarred tissue.Sometimes these areas represent ischemic,noncontractile, but metabolically activeareas that have the potential to regain func-tion if an adequate blood supply is restored.For example, such areas may represent hi-

68 Chapter Three

TABLE 3.4. Cardiac Catheterization and Angiography in Cardiac Disorders

Disorder Finding

Coronary artery disease • Identification of atherosclerotic lesionsMitral regurgitation • Large systolic v wave in left atrial pressure tracingMitral stenosis • Abnormally high pressure gradient between left atrium and left ventricle

in diastoleTricuspid insufficiency • Large systolic v wave in the right atrial pressure tracingAortic stenosis • Systolic pressure gradient between left ventricle and aortaCongestive heart failure • Estimation of cardiac output

• Calculation of systemic and pulmonary vascular resistances

10090-03_CH03.qxd 8/31/06 5:24 PM Page 68


bernating myocardium, segments that de-monstrate diminished contractile functionowing to chronic reduction of coronaryblood flow (see Chapter 6). This viablestate (in which the affected cells can bepredicted to regain function followingcoronary revascularization) can often bedifferentiated from irreversibly scarred myo-cardium by repeat imaging at rest after theinjection of additional 201Tl to enhance up-take by viable cells.

99mTc-sestamibi (commonly referred to asMIBI) is the most widely used 99mTc-labeledcompound. This agent is a large lipophilicmolecule that, like thallium, is taken up inthe myocardium in proportion to blood flow.

The uptake mechanism differs in that thecompound crosses the myocyte membranepassively, driven by the negative membranepotential. Once inside the cell, it further ac-cumulates in mitochondria, driven by thatorganelle’s even more negative membranepotential. The myocardial distribution ofMIBI reflects perfusion at the moment of injection, and in contrast to thallium, it re-mains fixed intracellularly; that is, it does notredistribute over time. Consequently, per-forming a MIBI procedure is more flexible, asimages can be obtained up to 4 to 6 hoursafter injection and repeated as necessary. AMIBI study is usually performed as a 1-dayprotocol in which an initial injection, using

A

BSHORT AXIS

Anterior

Inferior

LatSepSTRESS

REST

C

HORIZONTALLONG AXIS

Apex

LatSepSTRESS

REST

Figure 3.19. Stress and rest myocardial perfusion single photon emission computed tomography images(using 99mTc-tetrofosmin) of a patient with a high-grade stenosis within the proximal left anterior de-scending coronary artery. A. Miniaturized reproduction of the complete scan showing tomographic images ineach of three views. The first, third, and fifth rows demonstrate images during stress, and the second, fourth, andsixth rows are matching images acquired at rest. B and C. Enlarged selected panels from part A showing stress andrest images in the short axis and horizontal long axis views. The arrows indicate regions of decreased perfusion dur-ing stress but normal perfusion on the matching resting scans, consistent with inducible ischemia. Lat, lateral wallof the LV; Sep, septal wall. (Courtesy of Marcelo Di Carli, MD, Brigham and Women’s Hospital, Boston, MA.)

10090-03_CH03.qxd 8/31/06 5:24 PM Page 69

a small tracer dose, and imaging are per-formed at rest. Then, a few hours later, thepatient exercises and repeat imaging is per-formed after injection of a larger tracerdose.

Stress nuclear imaging studies with either201Tl- or 99mTc-labeled compounds havegreater sensitivity and specificity than stan-dard exercise electrocardiography for the de-tection of ischemia but are more expensiveand should be ordered judiciously. Nuclearimaging is particularly appropriate for pa-tients with certain baseline electrocardio-gram (ECG) abnormalities that preclude ac-curate interpretation of a standard exercisetest. Examples include patients with elec-tronic pacemaker rhythms, those with leftbundle branch block, and those who takecertain medications that alter the ST seg-ment, such as digoxin. Nuclear scans alsoprovide more accurate anatomic localiza-tion of the ischemic segment(s) and quan-tification of the extent of ischemia com-pared with standard exercise testing. Inaddition, electronic synchronizing (gating)of nuclear images to the ECG cycle permitswall motion analysis.

Patients with orthopedic or neurologicconditions, as well as those with severe phys-ical deconditioning or chronic lung disease,may be unable to perform an adequate exer-cise test on a treadmill or bicycle. In such pa-tients, stress images can be obtained insteadby administering pharmacologic agents, suchas adenosine or dipyridamole. These agentsinduce diffuse coronary vasodilation, aug-menting blood flow to myocardium per-fused by healthy coronary arteries. Since ischemic regions are already maximally di-lated (because of local metabolite accumula-tion), the drug-induced vasodilation causes a“steal” phenomenon, reducing isotope up-take in regions distal to significant coronarystenoses (see Chapter 6). Alternatively, dobu-tamine (see Chapter 17) can be infused intra-venously to increase myocardial oxygen de-mand to test for ischemia.

In addition to its role in the diagnosis ofmyocardial ischemia, nuclear imaging canbe useful after acute myocardial infarctionto assess the effectiveness of thrombolytic

therapy, for risk stratification, and to predictwhich patients would benefit from early me-chanical revascularization.

Radionuclide Ventriculography

Radionuclide ventriculography (RVG, alsoknown as blood pool imaging) is used to an-alyze right and left ventricular function. Aradioisotope (usually 99mTc) is bound to redblood cells or to human serum albumin andthen injected as a bolus. Nuclear images areobtained at fixed time intervals as the la-beled material passes through the heartand great vessels. Multiple images are dis-played sequentially to produce a dynamicpicture of blood flow. Calculations, such asdetermination of the ejection fraction, arebased on the difference between radioactivecounts present in the ventricle at end dias-tole and at end systole. Therefore, measure-ments are largely independent of any as-sumptions of ventricular geometry and arehighly reproducible. Studies suggest thatRVG and echocardiography provide similarleft ventricular ejection fraction values, butunlike echocardiography, RVG can also cal-culate an accurate right ventricular ejectionfraction.

Radionuclide ventriculography is com-monly used to assess baseline cardiac func-tion in patients scheduled to undergo po-tentially cardiotoxic chemotherapy (e.g.,doxorubicin) and to follow cardiac functionover time in such patients. In addition, first-pass imaging and scans gated to the ECGpermit recognition of abnormal cardiac andvascular shunts.

Assessment of Myocardial Metabolism

Positron emission tomography (PET) is aspecialized nuclear imaging technique usedto assess myocardial perfusion and viability.PET imaging employs positron-emitting iso-topes (e.g., oxygen-15, carbon-11, rubidium-82, nitrogen-13, and fluorine-18) attachedto metabolic or flow tracers. Sensitive detec-tors measure positron emission from the tra-cer molecules.

70 Chapter Three

10090-03_CH03.qxd 8/31/06 5:24 PM Page 70


Myocardial perfusion is commonly as-sessed using nitrogen-13–labeled ammoniaor rubidium-82. These flow tracers are takenup by myocytes in proportion to bloodflow. Myocardial viability can be determinedby PET by studying glucose utilization inmyocardial tissue. In normal myocardiumunder fasting conditions, glucose is usedfor approximately 20% of energy produc-tion, with free fatty acids providing the remaining 80%. In ischemic conditions,however, metabolism shifts toward glucoseuse, and the more ischemic the myocardialtissue, the stronger the reliance on glucose.Fluoro-18 deoxyglucose (18FDG), created by substituting fluorine-18 for hydrogen in2-deoxyglucose, is used to study glucoseuptake. This substance competes with glucose both for transport into myocytesand for subsequent phosphorylation. Un-like glucose, however, 18FDG is not metab-olized and becomes trapped within themyocyte.

Combined evaluation of perfusion and18FDG metabolism allows assessment ofboth regional blood flow and glucose up-take, respectively. PET scanning thus helpsdetermine whether areas of ventricular con-

tractile dysfunction owing to decreased flowrepresent scar tissue, or whether the regionis still viable (e.g., hibernating myocar-dium). In scar tissue, both blood flow to theaffected area and 18FDG uptake are decreased.Because the myocytes in this region are per-manently damaged, such tissue is not likelyto benefit from a revascularization proce-dure. Hibernating myocardium, in contrast,shows decreased blood flow but normal orelevated 18FDG uptake. Such tissue may ben-efit from mechanical revascularization, asdescribed in Chapter 6.

Table 3.5 summarizes the radionuclide-imaging abnormalities associated with com-mon cardiac conditions.

COMPUTED TOMOGRAPHYComputed tomography (CT) uses thin x-ray beams to obtain axial plane images.An x-ray tube is programmed to rotatearound the body, and the generated beamsare partially absorbed by body tissues. Theremaining beams emerge and are capturedby electronic detectors, which relay infor-mation to a computer for image composi-tion. CT scanning typically requires admin-

TABLE 3.5. Nuclear Imaging in Cardiac Disorders

Disorder Finding

Myocardial ischemiaStress-delayed reinjection 201Tl

Rest-stress 99mTc-labeled compoundsPET (N-13 ammonia/18FDG)

Myocardial infarctionStress-delayed reinjection 201TlRest-stress 99mTc-labeled compoundsPET (N-13 ammonia/18FDG)Hibernating myocardiumRest-delayed 201TlPET (N-13 ammonia/18FDG)Assessment of ventricular unction99mTc RBC gated radionuclide

Radionuclide ventriculography

18FDG, fluoro-18 deoxyglucose; N-13, nitrogen-13; PET, positron emission tomography; RBC, red blood cell; 99mTc, tech-netium-99m; 201Tl, thallium-201.

• Low uptake during stress with complete or partial fill-in withdelayed or reinjection images

• Normal uptake at rest with decreased uptake during stress• Decreased flow with normal or increased 18FDG uptake during

stress

• Low uptake during stress and low uptake after reinjection• Low uptake in rest and stress images• Decreased flow and decreased 18FDG uptake at rest

• Complete or partial fill-in of defects after reinjection• Decreased flow and increased 18FDG uptake at rest

• Assessment of global left and right ventricular function at restor during exercise

• Regional wall motion

Tab. 5

AQ4

10090-03_CH03.qxd 8/31/06 5:24 PM Page 71

istration of an intravenous contrast agent todistinguish intravascular contents (i.e.,blood) from neighboring soft tissue struc-tures (e.g., myocardium).

Applications of CT in cardiac imaging in-clude assessment of the great vessels, peri-cardium, myocardial structures, and coro-nary arteries. CT can be used to accuratelydiagnose aortic dissections and aneurysms(Fig. 3.20) and to monitor patients whohave undergone surgical repair of these con-ditions. CT clearly delineates pericardial ef-fusions as well as pericardial thickening andcalcification. Myocardial abnormalities,such as regional hypertrophy or ventricularaneurysms, and intracardiac thrombus for-mation can be distinctly visualized on CT

images. A limitation of conventional CTtechniques is the artifact generated by pa-tient motion (i.e., breathing) during imageacquisition. Modern spiral CT imaging al-lows more rapid image acquisition, oftenduring a single breath-hold, at relativelylower radiation doses than conventional CT.Spiral CT is particularly important in the di-agnosis of pulmonary embolism. When anintravenous iodine-based contrast agent isadministered, emboli create the appearanceof “filling defects” in otherwise contrast-enhanced pulmonary vessels (Fig. 3.21).

Electron beam computed tomography(EBCT), a technology developed particu-larly for cardiac imaging, uses a direct elec-tron beam to acquire images in a matter of

72 Chapter Three

AA

AA AA

RK

RCI LCI

LCI

LEI

RK

LKLK

PA

Liver

A

B

C D

Figure 3.20. Computed tomography (CT) imaging of aortic dissection. A and B. Axial images demonstrate an in-timal flap (colored arrowheads) separating the true and false lumens. C. CT angiography (CTA) with three-dimensionalreconstructions. In this left anterior oblique view, the origin of the dissection (colored arrowhead) is apparent in the dis-tal portion of the aortic arch. The dissection continues to the level of the renal arteries (white arrowhead) and beyond.D. In this CTA left posterior oblique view, the dissection extends to the infrarenal aorta (white arrowhead) and involvesthe left common and external iliac arteries (colored arrowhead). AA, ascending aorta; LCI, left common iliac artery; LEI,left external iliac artery; LK, left kidney; PA, main pulmonary artery; RCI, right common iliac artery; RK, right kidney.(Courtesy of Suhny Abbara, MD, Massachusetts General Hospital, Boston.)

Fig. 20

Fig. 21

one line sh

10090-03_CH03.qxd 8/31/06 5:24 PM Page 72


milliseconds. Rapid succession of images de-picts cardiac structures at multiple timesduring a single cardiac cycle. Displayingthese images in a cine (“motion picture”)format can provide estimates of left ventric-ular volumes, including stroke volume, andejection fraction, expanding the applicationof CT to include not only heart structure butalso function. Capable of detecting coronaryartery calcification, EBCT has been used pri-marily to screen for coronary artery disease.Because calcified coronary artery plaqueshave a radiodensity similar to that of bone,they appear attenuated (white) on CT. TheAgatson score, a measure of total coronaryartery calcium, correlates well with athero-sclerotic plaque burden, and predicts therisk of nonfatal myocardial infarction andcardiac death, independently of other cardiacrisk factors.

Beyond assessing the coronary calciumscore, newer CT technology can character-ize individual atherosclerotic plaques andstenoses in great detail. Current multide-

tector row CT scanners acquire as many as64 anatomic sections with each rotation,providing excellent spatial resolution. Ad-ministration of intravenous contrast andcomputer reformatting allows visualiza-tion of the arterial lumen and regions ofcoronary narrowings (Fig. 3.22). Becauseimage acquisition is timed with the cardiaccycle, a relatively low heart rate is desir-able, such that a β-blocker is often givenprior to scanning.

CT is not as sensitive as conventionalangiography for the detection of coronarylesions, and it cannot adequately evaluatestenosis within coronary artery stents. Inaddition, this technique results in signifi-cant radiation exposure. However, CT israpid, relatively inexpensive, and signifi-cantly less invasive than conventional an-giography. Its role in assessing patientswith symptoms suggestive of coronary arterydisease and for following the progression ofknown coronary disease is currently underevaluation.

Figure 3.21. Spiral computed tomography image demonstrating a massive pul-monary embolism. The white arrows point to a large thrombus within the right pul-monary artery. It appears as a filling defect within the otherwise contrast-enhanced pul-monary vasculature. AA, ascending aorta; DA, descending aorta; LPA, left pulmonaryartery; PA, main pulmonary artery; RPA, right pulmonary artery; SVC, superior vena cava.

Fig. 22

line short

10090-03_CH03.qxd 8/31/06 5:24 PM Page 73

RA

RV

RV

LV

LAD

RCA

RCA

LCX

LCXLAD

LM LM

LV

LA

LA

AoAo

PA

A B

C

Ao

PA

LAD

Figure 3.22. Computed tomography (CT) coronary angiography.After a patient is imaged in a high-resolution axial CT scanner, three-di-mensional reconstructions (termed volume renderings) are generated bya computer. A. Volume rendering of a normal CT angiogram. B. Volumerendering of a CT angiogram that demonstrates diffuse coronary arterydisease. Notice that the caliber of each vessel is irregular along its length.C. This curved reformat of the left anterior descending artery (LAD) de-picts the entire course of the vessel in a single, flat image, making it eas-ier to detect stenoses. None are present here. Ao, aorta; LA, left atrium;LCX, left circumflex artery; LM, left main coronary artery; LV, left ventri-cle; PA, pulmonary artery; RA, right atrium; RCA, right coronary artery;RV, right ventricle. (Courtesy of Suhny Abbara, MD, Massachusetts Gen-eral Hospital, Boston.)

MAGNETIC RESONANCE IMAGING

Magnetic resonance imaging (MRI) uses apowerful magnetic field to obtain detailedimages of internal structures. This techniqueis based on the magnetic polarity of hydro-

gen nuclei, which align themselves with anapplied magnetic field. Radiofrequency ex-citation causes the nuclei to move out ofalignment momentarily. As they return totheir resting states, the nuclei emit radiowaves, which are translated into computer-

10090-03_CH03.qxd 8/31/06 5:24 PM Page 74


generated images. Therefore, MRI requiresno ionized radiation. Among all the imagingmodalities, MRI is best at differentiating tis-sue contrasts (blood, fluid, fat, and my-ocardium) and can often do so even withoutthe use of contrast agents. The addition ofgadolinium-based (noniodinated) contrastagents allows further characterization of car-diac structures and tissues using the latestMRI techniques.

The detail of soft tissue structures is oftenexquisitely demonstrated in magnetic reso-nance images (Fig. 3.23). Cardiac MRI hasan established role in assessing congenitalanomalies, such as shunts, and diseases ofthe aorta, including aneurysm and dissec-tion. It is also used to assess left and rightventricular mass and volume, intravascularthrombus, cardiomyopathies, and neoplas-tic disease (Fig. 3.24). ECG-gated and cineMRI techniques capture images at discretetimes in the cardiac cycle, and therefore per-

mit evaluation of valvular and ventricularfunction.

Two applications of cardiac MRI deservespecial mention. Coronary magnetic reso-nance angiography (coronary MRA) is a noninvasive, contrast-free angiographic-imaging modality. Laminar blood flow ap-pears as bright signal intensity, whereasturbulent blood flow, at the site of stenosis,results in less bright or absent signal inten-sity. This technique has shown high sensi-tivity and accuracy for the detection of im-portant coronary artery disease in the leftmain coronary artery and in the proximaland midportions of the three major coro-nary vessels. Coronary MRA is also usefulin delineating coronary artery congenitalanomalies.

In contrast-enhanced MRI, a gadolinium-based agent is administered intravenouslyto differentiate between impaired (but vi-able) myocardial segments and truly in-

Figure 3.23. Cardiac magnetic resonance images of a normal person. A. Three-chamberlong axis view of the heart in diastole and systole showing the left ventricle (LV), right ventricle(RV), and left atrium (LA). The mitral valve (MV), aortic valve (AV), ascending aorta (AAO), anddescending aorta (DA) are also imaged. B. Midventricular short axis view demonstrating the LV,RV, and left ventricular papillary muscles (PMs). PW, posterior wall; S, septum. (Courtesy of Raymond Y. Kwong, MD, Brigham and Women’s Hospital, Boston, MA.)

Fig. 23

Fig. 24

10090-03_CH03.qxd 8/31/06 5:24 PM Page 75

fracted tissue. This technique is based onfindings that gadolinium is excluded fromviable cells with intact cell membranes butcan permeate and concentrate in infarcted(i.e., irreversibly damaged) areas. Nonviablemyocardial segments therefore appear “hy-perenhanced” relative to viable, reversiblyimpaired myocardium (Fig. 3.25). In thisway, gadolinium-enhanced MRI can helpselect patients who are likely to benefit fromrevascularization procedures.

SUMMARY

This chapter has presented an overview ofimaging and catheterization techniques cur-

rently available to assess cardiac structureand function. Many of these tools are ex-pensive and yield similar information. Forexample, estimates of ventricular contractilefunction can be made by echocardiogra-phy, nuclear imaging, contrast angiogra-phy, gated CT, or MRI. Myocardial viabilitycan be assessed using nuclear-imaging stud-ies, gadolinium MRI, or dobutamine echo-cardiography.

Determining the single best test for anygiven patient depends on a number of fac-tors. One is the ease by which images may beobtained. In a critically ill patient, bedsideechocardiography provides an easily ob-tained measure of left ventricular systolicfunction. Obtaining similar information

76 Chapter Three

A B

LV

RV

RA

LA LA

RA

RV

LV

Figure 3.24. Magnetic resonance imaging of an intracardiac mass. Both images are apical four-chamber views.A. Before a gadolinium-based contrast agent is administered, an abnormal left atrial mass (indicated by the oval)demonstrates diminished signal relative to the surrounding tissue. In this respect, it resembles a nonvascular thrombus.B. After contrast injection, the mass enhances similar to the surrounding tissue, indicating that it is vascularized. Biopsyrevealed a spindle-cell carcinoma. LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle. (Courtesy ofRaymond Kwong, MD, Brigham and Women’s Hospital, Boston, MA.)

Fig. 25

10090-03_CH03.qxd 8/31/06 5:24 PM Page 76


from a nuclear study would require radio-isotope administration and a trip to the nuclear scanner. Another factor to consideris the degree of invasiveness of a given imaging technique. Expense, availableequipment, and institutional preference andexpertise also play roles in determining the

choice of an imaging approach. When usedappropriately, each imaging tool can pro-vide important information to guide the di-agnosis and management of cardiovasculardisorders.

Table 3.6 summarizes the uses of imagingtechniques described in this chapter.

A

LVRV RV LV

B

Figure 3.25. Gadolinium-enhanced magnetic resonance imagesdemonstrating a region of nonviable myocardium. Both images areshort axis views. A. Imaging before administration of gadolinium demon-strates thinning of the anterior and anteroseptal myocardium (coloredarrow) suggestive of infarcted tissue. B. After contrast injection, the ante-rior and anteroseptal segments of the left ventricle selectively enhance(white arrows), indicating that scar tissue is present. Because more thanhalf the thickness of the ventricular wall is scarred, coronary revascular-ization would have a low likelihood of improving contractile function ofthese myocardial segments. LV, left ventricle; RV, right ventricle. (Courtesyof Raymond Kwong, MD, Brigham and Women’s Hospital, Boston, MA.)

TABLE 3.6. Summary of Cardiac Imaging Techniques

Imaging Technique Finding Examples of Clinical Uses

Chest radiography

Transthoracic echocardiography (TTE)

Transesophageal echocardiography (TEE)

• Cardiac and mediastinal contours• Pulmonary vascular markings

• Wall thickness, chamber dimensions

• Anatomic relationships and motionof cardiac structures

• Flow direction, turbulence, and velocity measurements

• Echo contrast studies

• Stress echocardiography• Similar to TTE but higher resolution

• Detect chamber dilatation• Identify consequences of stenotic

and regurgitant valve lesions and intracardiac shunts

• Assess global and segmental ventric-ular contraction

• Identify valvular abnormalities andvegetations

• Diagnose consequences of myocar-dial infarction (e.g., ventricularaneurysm, papillary muscle rupture,intraventricular thrombus)

• Identify myocardial, pericardial, andcongenital abnormalities

• Visualize intracardiac thrombus

(Continued)

Tab. 6

10090-03_CH03.qxd 8/31/06 5:24 PM Page 77

Acknowledgment

The authors are grateful to Suhny Abbara, MD;Sharmila Dorbala, MD; Raymond Kwong, MD: andJeffrey Popma, MD, for their helpful suggestions.Contributors to the previous editions of this chapterwere Deborah Bucino, MD; Sharon Horesh, MD;Shona Pendse, MD; Albert S. Tu, MD; Patrick Yachim-ski, MD, and Patricia C. Come, MD.

Additional Reading

Armstrong WF, Zoghbi WA. Stress echocardiogra-phy: current methodology and clinical applica-tions. J Am Coll Cardiol 2005;45:1739–1747.

Aviles RJ, Messerli AW, Askari AT, et al. IntroductoryGuide to Cardiac Catheterization. Philadelphia:Lippincott Williams & Wilkins, 2005.

Baim D, Grossman W. Grossman’s Cardiac Catheter-ization, Angiography and Intervention. 7th Ed.Philadelphia: Lippincott Williams & Wilkins, 2005.

Brindis RG, Douglas PS, Hendel RC, et al. ACCF/ASNCappropriateness criteria for single-photon emis-sion computed tomography myocardial perfusionimaging (SPECT MPI): a report of the AmericanCollege of Cardiology Foundation Quality Strate-gic Directions Committee Appropriateness Crite-ria Working Group and the American Society of

Nuclear Cardiology endorsed by the AmericanHeart Association. J Am Coll Cardiol 2005;46:1587–1605.

Clouse ME, Chen J, Krumholz HM. How useful iscomputed tomography for screening for coronaryartery disease? Circulation 2006;113:125–146.

DiMario C, Sutaria N. Coronary angiography in theangioplasty era: projections with a meaning.Heart 2005;91:968–976.

Edelman RR. Contrast-enhanced MR imaging of theheart: overview of the literature. Radiology 2004;232:653–668.

Enriquez-Sarano M, Avierinos J-F, Messika-Zeitoun D,et al. Quantitative determinants of the outcome ofasymptomatic mitral regurgitation. N Engl J Med2005;352:875–883.

Feigenbaum H. Feigenbaum’s Echocardiography.6th Ed. Philadelphia: Lippincott Williams &Wilkins, 2004.

Ghesani M, DePuey G, Rozanski A. Role of F-18 FDGpositron emission tomography (PET) in the as-sessment of myocardial viability. Echocardiogra-phy 2005;22:165–177.

Hillis GS, Bloomfield P. Basic transthoracic echocar-diography. BMJ 2005;330:1432–1436.

Hoffmann MH, Shi H, Schmitz BL, et al. Noninva-sive coronary angiography with multislice com-puted tomography. JAMA 2005;293:2471–2478.

78 Chapter Three

TABLE 3.6. (Continued)

Imaging Technique Finding Examples of Clinical Uses

Cardiac catheterization

Nuclear SPECT imaging (using 99mTc-labeled compounds or 201Tl)

Radionuclide ventriculography

Positron emission tomography (PET)

Computed tomo-graphy (CT)

Magnetic resonance imaging (MRI)

ECG, electrocardiogram; SPECT, single photon emission computed tomography; 99mTc, technetium-99m; 201Tl, thallium-201.

• Pressure measurement

• Contrast angiography

• Regional myocardial perfusion

• Myocardial viability

• Ventricular contractile function

• Myocardial perfusion and meta-bolism

• Anatomy and structural relation-ships

• Detailed soft tissue anatomy

• Evaluate intracardiac pressures (e.g., in valvular disease, heart fail-ure, pericardial disease)

• Visualize ventricular contractilefunction, regurgitant valve lesions

• Detect, quantify, and localize myo-cardial ischemia

• Perform stress testing in patientswith baseline ECG abnormalities

• Calculate ventricular ejection fractionand quantitate intracardiac shunts

• Evaluate contractile function

• Diagnose disease of the great vessels(aortic dissection, pulmonary embolism)

• Assess and stenoses (multidetectorCT) myocardial structure and func-tion (e.g., ventricular mass and vol-ume, neoplastic disease, intracardiacthrombus, cardiomyopathies)

AQ1

10090-03_CH03.qxd 8/31/06 5:24 PM Page 78


Otto CM. The Practice of Clinical Echocardiography.3rd Ed. Philadelphia: Elsevier Saunders, 2004.

Quiroz R, Kucher N, Zou KH, et al. Clinical validityof a negative computed tomography scan in pa-tients with suspected pulmonary embolism: a sys-tematic review. JAMA 2005;293:2012–2017.

Raggi P. Role of electron-beam computed tomogra-phy and nuclear stress testing in cardiovascularrisk assessment. Am J Cardiol 2005;96:20–27.

Raggi P, Taylor A, Fayad Z, et al. Atheroscleroticplaque imaging: contemporary role in preventive

cardiology. Arch Intern Med 2005;165:2345–2353.

Schoenhagen P, Halliburton SS, Stillman AE, et al.Noninvasive imaging of coronary arteries: currentand future role of multi-detector row CT. Radiol-ogy 2004;232:7–17.

Schuijf JD, Shaw LJ, Wijns W, et al. Cardiac imagingin coronary artery disease: differing modalities.Heart 2005;91:1110–1117.

Sengupta PP, Khandheria BK. Transesophageal echo-cardiography. Heart 2005;91:541–547.

10090-03_CH03.qxd 8/31/06 5:24 PM Page 79

Chapter 3—Author Queries1. AU: Correct that these folks are MDs?2. AU: PMV is a label on the figure that needs defining here. Is the one added here correct?3. ED: Are both Òwith permission ” and “by permission” necessary?4. AU: Added text correct?

10090-03_CH03.qxd 8/31/06 5:24 PM Page a

80

ELECTRICAL MEASUREMENT—SINGLE-CELL MODEL

ELECTROCARDIOGRAM LEAD REFERENCE SYSTEM

SEQUENCE OF NORMAL CARDIAC ACTIVATION

INTERPRETATION OF THEELECTROCARDIOGRAM

CalibrationHeart RhythmHeart RateIntervals (PR, QRS, QT)Mean QRS AxisAbnormalities of the P WaveAbnormalities of the QRS ComplexST Segment and T Wave Abnormalities

C H A P T E R

4The ElectrocardiogramLilit GaribyanLeonard S. Lilly

Cardiac contraction relies on the organizedflow of electrical impulses through the heart.The electrocardiogram (ECG) is an easily ob-tained recording of that activity, and it pro-vides a wealth of information about cardiacstructure and function. This chapter presentsthe electrical basis of the ECG in health anddisease and leads the reader through the ba-sics of interpretation. To become fully adeptat this technique and to practice the princi-ples described here, you may wish to consultone of the complete electrocardiographictextbooks listed at the end of the chapter.

ELECTRICAL MEASUREMENT—SINGLE-CELL MODEL

This section begins by observing the propa-gation of an electrical impulse within a singlecardiac muscle cell, illustrated in Figure 4.1.

On the right side of the diagram, a voltmeterrecords the electrical potential across thecell on graph paper. In the resting state, thecell is polarized; that is, the entire outside ofthe cell is electrically positive with respect tothe inside, because of the ionic distributionacross the cell membrane, as described inChapter 1. In this resting state, the volt-meter electrodes, which are placed on op-posite outside surfaces of the cell, do not de-tect any electrical activity, because there isno electrical potential difference betweenthem (the myocyte surface is homogeneouslycharged).

This equilibrium is disturbed, however, bystimulation of the cell (see Fig. 4.1B). Duringthe action potential, as cations rush acrossthe sarcolemma into the cell, the polarity at the stimulated region transiently reverses,such that the outside becomes negativelyFig. 1

10090-04_CH04.qxd 8/31/06 5:25 PM Page 80

The Electrocardiogram 81

charged with respect to the inside; that is,the region depolarizes. At that moment, anelectrical potential is created on the cellsurface between the depolarized area (nega-tively charged surface) and the still-polarized(positively charged surface) portions of thecell. As a result, an electrical current beginsflowing between these two regions.

By convention, the direction of electricalcurrent is said to flow from areas that arenegatively charged to those that are posi-tively charged. When a depolarization cur-rent is directed toward the (+) electrode of thevoltmeter, an upward deflection is recorded.Conversely, if it is directed away from the (+)electrode, a downward deflection is recorded.

A

B

C

D

EFigure 4.1. Depolarization of a single cardiac muscle cell. A. Inthe resting state, the surface of the cell is positively charged relative tothe inside. Because the surface is homogeneously charged, the volt-meter electrodes outside the cell do not record any electrical potentialdifference (“flat line” recording). B. Stimulation of the cell initiates de-polarization (shaded area); the outside of the depolarized region be-comes negatively charged relative to the inside. Because the currentof depolarization is directed toward the (+) electrode of the voltmeter,an upward deflection is recorded. C. Depolarization spreads, creatinga greater upward deflection by the recording electrode. D. The cell hasbecome fully depolarized. The surface of the cell is now completelynegatively charged compared with the inside. Because the surface isagain homogeneously charged, a flat line is recorded by the voltmeter.E. Notice that if the position of the voltmeter electrodes had been re-versed, the wave of depolarization would have traveled away from the(+) electrode, causing the deflection to be downward.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 81

Because the depolarization current in thisexample proceeds from left to right—that is,toward the (+) electrode—an upward deflec-tion is recorded by the voltmeter. As thewave of depolarization spreads along thecell, additional electrical forces directed to-ward the (+) electrode record an even greaterupward deflection (see Fig. 4.1C). Once the cell has become fully depolarized (see Fig. 4.1D), its outside is completely nega-tively charged with respect to the inside, theopposite of the initial resting condition.However, because the surface charge is homo-geneous once again, the external electrodesmeasure a potential difference of zero andthe voltmeter records a neutral “flat line”during this period.

Note that in Figure 4.1E, if the electrodewires of the voltmeter had been reversed sothat the (+) pole was placed to the left of the cell, then as the wave of depolarizationproceeded toward the right, it would haveheaded away from the (+) electrode and therecorded deflection would have been down-ward. This relationship should be kept inmind when the polarity of ECG leads is de-scribed in the next section.

Depolarization of the cell initiates cardiacmuscle contraction and is then followed byrepolarization, the process by which thecellular charges return to the resting state(Fig. 4.2). As the left side of the cell begins torepolarize in this example, its surface chargebecomes positive once again. A current istherefore generated from the still negativelycharged surface toward the positively chargedarea. Because this current is directed awayfrom the voltmeter’s (+) electrode, a down-ward deflection is recorded, opposite to thatwhich was observed during the process ofdepolarization.

Repolarization is a slower process than de-polarization, so that the inscribed deflectionof repolarization is wider and of lower mag-nitude. Once the cell has returned to theresting state, the surface charges are onceagain homogeneous and no further electricalpotential is detected, resulting in a flat lineon the voltmeter recording (see Fig. 4.2C).

It is important to note that in the intacthuman heart, the sequence of repolarizationactually proceeds in the direction opposite

that of depolarization. This occurs becausemyocardial action potential durations aremore prolonged in cells near the inner en-docardium (the first cells stimulated by Purkinje fibers) than in myocytes near theouter epicardium (the last cells to depolar-ize). Thus, the endocardium is the first re-gion to depolarize but the last region to re-polarize. As a result, the recorded pattern ofrepolarization on an ECG is usually the in-verse of what was presented in this example;that is, the current of repolarization (nega-

82 Chapter Four

A

B

C

D

Figure 4.2. Sequence of repolarization of a singlecardiac muscle cell. A. As repolarization commences,positive charges reemerge on the surface of the cell, anda current flows from the still-negatively charged surfaceareas to the repolarized region. Because the current is di-rected away from the (+) electrode of the voltmeter, adownward deflection is recorded. B. Repolarization pro-gresses. C. Repolarization has been completed, and theoutside surface of the cell is once again homogeneouslycharged, so that no further electrical potential is detected(flat line once again). D. In the human heart, repolariza-tion proceeds in a direction opposite that of depolarization(would be from right to left in this example, and the waveof repolarization would be upright). Therefore, the de-flections of depolarization and repolarization of the nor-mal intact heart are in the same direction, as shown here.Notice that the wave of repolarization is of lower ampli-tude and more prolonged than that of depolarization.

Fig. 2

10090-04_CH04.qxd 8/31/06 5:25 PM Page 82


tive-to-positive flow) in Figure 4.2 is nor-mally directed toward the (+) electrode andinscribes an upright deflection on the record-ing. Therefore, in a normal heart, the forcesof depolarization and repolarization are usu-ally oriented in the same direction on theECG recording (see Fig. 4.2D).

The depolarization and repolarization ofa single cardiac muscle cell have been con-sidered here. As the wave of depolarizationspreads rapidly through the heart, each cellgenerates electrical forces, and it is the sumof these forces, measured at the skin’s sur-face, recorded by the ECG machine. The di-rection and magnitude of the deflections onthe ECG recording depend on how the elec-trical forces are aligned to a set of specificreference axes, known as ECG leads.

ELECTROCARDIOGRAM LEADREFERENCE SYSTEM

When the electrocardiogram was first in-vented, the recording was made by dunkingthe patient’s arms and legs into large buck-ets filled with electrolyte solution and wiredto the machine. That process naturally wasfairly messy and fortunately is no longernecessary. Instead, wire electrodes are placeddirectly on the skin, held in place by adhe-sive tabs, in the standard arrangement shownin Figure 4.3. The right-leg electrode is notused for measurement but serves as an elec-trical ground. Table 4.1 lists the standard lo-cations of the chest electrodes.

A complete ECG tracing is produced byrecording the electrical activity betweenthese standard electrodes in specific patterns.Figure 4.4 demonstrates the orientation ofthe six standard reference axes (termed limbleads) in the body’s frontal plane, which areelectronically constructed as described in thefollowing paragraphs.

The ECG machine records lead aVR byselecting the right-arm electrode as the (+)pole with respect to the other electrodes.This is known as a unipolar lead, becausethere is no single (−) pole; rather, all the otherelectrodes are averaged to create a composite(−) reference. When the instantaneous elec-trical activity of the heart points in the di-rection of the right arm, an upward deflec-tion is recorded in lead aVR. However, whenthe electrical forces are heading away fromthe right arm, the ECG inscribes a down-ward deflection in aVR.

Similarly, lead aVF is recorded by settingthe left leg as the (+) pole, such that a positive

BA

V1

V2

V3

V6

V5

V4

Right armelectrode

Left armelectrode

Right legelectrode

Left legelectrode

Chestelectrodes

Figure 4.3. Placement of electrocardiogram (ECG) electrodes. A. Standardpositions. B. Close-up view of chest electrode placement.

TABLE 4.1. Proper Positions of ECG Chest Electrodes

V1 4th ICS, 2 cm to the right of sternumV2 4th ICS, 2 cm to the left of sternumV3 Midway between V2 and V4

V4 5th ICS, left midclavicular lineV5 5th ICS, left anterior axillary lineV6 5th ICS, left midaxillary line

ICS, intercostal space

Fig. 3

Tab. 1

Fig. 4

10090-04_CH04.qxd 8/31/06 5:25 PM Page 83

deflection is recorded when forces are di-rected toward the feet. Lead aVL is selectedwhen the left-arm electrode is made the (+)pole and records an upward deflection whenelectrical activity is aimed in that direction.

In addition to these three unipolar limbleads, three bipolar leads are part of the stan-dard ECG recording (see Fig. 4.4). Bipolar in-dicates that one limb electrode is the (+) poleand another single electrode provides the (−)reference. In this case, the ECG machine in-scribes an upward deflection if electricalforces are heading toward the (+) electrodeand records a downward deflection if theforces are heading toward the (−) electrode. Asimple mnemonic to remember the place-ment of the bipolar leads is that the leadname indicates the number of l’s in the place-ment sites. For example, lead III connects theleft arm to the left leg, lead II connects theright arm to the left leg, and lead I connectsthe left arm to the right arm. Table 4.2 listshow the six limb leads are derived.

By overlaying the six limb leads together,a reference system is established (Fig. 4.5). Inthe figure, each lead is presented with its (+)pole designated by an arrowhead and the (−)aspect by dashed lines. Note that each 30°sector of the circle falls along the (+) or (−)pole of one of the standard six ECG leads.Also note that the (+) pole of lead I points to

84 Chapter Four

Unipolar Limb Leads

aVRaVF

aVL

Bipolar Limb Leads

(+)

(–)(–)(–) (+)

IIIII

I

(+)(+)

(+)

Figure 4.4. The six limb leads are formed from the electrodes placed on the arms and leftleg. Each unipolar lead has a (+) designated electrode; for the unipolar leads, the (−) pole is an average of the other electrodes. Each bipolar lead has specific (−) and (+) designated electrodes.

Tab. 2

TABLE 4.2. Limb Leads

Lead (�) Electrode (�) Electrode

Bipolar leadsI LA RAII LL RAIII LL LAUnipolar leadsaVR RA *aVL LA *aVF LL *

*(−) Electrode constructed by combining all other electrodes.

LA, left arm; LL, left leg; RA, right arm.

Fig. 5

10090-04_CH04.qxd 8/31/06 5:25 PM Page 84


0° and that, by convention, measurement ofthe angles proceeds clockwise as +30°, +60°,and so forth. The complete ECG recordingprovides a simultaneous “snapshot” of theheart’s electrical activity, taken from the per-spective of each of these lead reference lines.

Figure 4.6 demonstrates how the magni-tude and direction of electrical activity arerepresented by the ECG recording in eachlead. This figure should be studied until thefollowing four points are clear:

1. An electrical force directed toward the (+)pole of a lead results in an upward deflec-tion on the ECG recording of that lead.

2. Forces that head away from the (+) elec-trode result in a downward deflection inthat lead.

3. The magnitude of the deflection, eitherupward or downward, reflects how paral-

Figure 4.5. The axial reference system is created bycombining the leads shown in Figure 4.4. Each leadhas a (+) region indicated by the arrowhead and a (−) re-gion indicated by the dashed line.

Figure 4.6. Relationship of the magnitude and direction of electrical activity to the ECG lead.A. The electrical vector is oriented parallel to lead I and aimed toward the (+) electrode; therefore, a tallupward deflection is recorded by the lead. B. The vector is still oriented toward the (+) region of lead Ibut not parallel to the lead, so that only a component of the force is recorded. The recorded deflectionis still upward but less tall compared with that shown in A. C. The electrical vector is perpendicular tolead I. so that no deflection is generated. D. The vector is directed toward the (−) region of lead I, caus-ing the ECG to record a downward deflection.

Fig. 6

10090-04_CH04.qxd 8/31/06 5:25 PM Page 85

lel the electrical force is to the axis of thelead being examined. The more parallelthe electrical force is to the lead, thegreater the magnitude of the deflection.

4. An electrical force directed perpendicularto an electrocardiographic lead does notregister any activity by that lead (a flatline on the recording).

The six standard limb leads examine the elec-trical forces in the frontal plane of the body.However, because electrical activity travels inthree dimensions, recordings from a perpen-dicular plane are also essential (Fig 4.7A).This is accomplished by the use of six elec-trodes placed on the anterior and left lateralaspect of the chest (see Fig. 4.3B), creating thechest (or precordial) leads. The orientationof these leads around the heart is shown inFigure 4.7B. These are unipolar leads and, aswith the unipolar limb leads, electrical forcesdirected toward these individual (+) electrodesresult in an upward deflection on the record-ing of that lead and forces heading awayrecord a downward deflection.

The standard complete electrocardiogramprints samples from each of the six limb leadsand each of the six chest leads, examples ofwhich are presented later in the chapter.

SEQUENCE OF NORMAL CARDIAC ACTIVATION

Conduction of electrical impulses throughthe heart is an orderly process. The normalbeat begins at the sinoatrial node, located at

the junction of the right atrium and the su-perior vena cava (Fig. 4.8). The wave of de-polarization rapidly spreads through theright and left atria and then reaches the atri-oventricular (AV) node, where it encountersan expected delay. The impulse then travelsrapidly through the bundle of His and intothe right and left bundle branches. These di-vide into the Purkinje fibers, which radiatetoward the myocardial fibers, stimulatingthem to contract.

Each heartbeat is represented on theECG by three major deflections that recordthe sequence of electrical propagation (seeFig. 4.8B). The P wave represents depolar-ization of the atria. Following the P wave,the tracing returns to the flat baseline as aresult of the conduction delay at the AVnode. The second deflection of the ECG, theQRS complex, represents depolarization ofthe ventricular muscle cells. After the QRScomplex, the tracing returns to baselineonce again, and after a brief delay, repolar-ization of the ventricular cells is signaled bythe T wave. Occasionally, an additionalsmall deflection follows the T wave (the Uwave), which is believed to represent latephases of ventricular repolarization.

The QRS complex may take one of severalshapes but can always be subdivided into in-dividual components (Fig. 4.9). If the firstdeflection of a QRS complex is downward, itis known as a Q wave. However, if the initialdeflection is upward, then that particularcomplex does not have a Q wave. The R waveis defined as the first upward deflection,

86 Chapter Four

Figure 4.7. The chest (precordial) leads. A. The cross-sectional plane ofthe chest. B. Arrangement of the six chest electrodes shown in the cross-sectional plane.

Fig. 7

Fig. 8

Fig. 9

10090-04_CH04.qxd 8/31/06 5:25 PM Page 86


A

B

Figure 4.8. Cardiac conduction pathway. A. The electrical impulse be-gins at the sinoatrial (SA) node (1) then traverses the atria (2). After adelay at the AV node (3), conduction continues through the bundle ofHis and into the right and left bundle branches (4). The latter divide intoPurkinje fibers, which stimulate contraction of the myocardial cells. B. Corresponding waveforms on the ECG recording: (1) the SA node dis-charges (too small to generate any deflection on ECG), (2) P wave in-scribed by depolarization of the atria, (3) delay at the AV node, and (4) depolarization of the ventricles (QRS complex). The T wave representsventricular repolarization.

Figure 4.9. Examples of QRS complexes. A. The first deflection is down-ward (Q wave), followed by an upward deflection (R wave), and then anotherdownward wave (S wave). B. Because the first deflection is upward, this com-plex does not have a Q wave; rather, the downward deflection after the R wave is an S wave. C. A QRS complex without downward deflections lacksQ and S waves. D. QRS composed of only a downward deflection; this is a Q wave but is often referred to as a QS complex. E. A second upward de-flection (seen in bundle branch blocks) is labeled R′.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 87

whether or not a Q wave is present. Anydownward deflection following the R wave isknown as an S wave. Figure 4.9 demonstratesseveral variations of the QRS complex. In cer-tain pathologic states, such as bundle branchblocks, additional deflections may be in-scribed, as shown in the figure. Study Figure4.9 to gain confidence in differentiating a Qfrom an S wave.

Figure 4.10 illustrates the course of nor-mal ventricular depolarization as it isrecorded by two of the ECG leads, aVF andaVL. The recording in aVF represents electri-cal activity from the perspective of the infe-rior (i.e., underside) aspect of the heart, andaVL records from the perspective of the leftlateral side. Recall that in the resting state,the surfaces of myocardial cells are posi-tively charged compared with the inside,but these external ECG leads record zerovoltage, as the sum of electrical forces ineach region are canceled by equal and op-posite forces.

The initial portion of ventricular myo-cardium that is stimulated to depolarize isthe midportion of the interventricular sep-tum, on the left side. Because depolariza-tion reverses the cellular charge, the surfaceof that region becomes negative with re-spect to the inside, and an electrical currentis generated (see Fig. 4.10B, arrow). Thisinitial force heads away from the left ven-tricle, toward the right ventricle and inferi-orly. Because the force is heading away fromthe (+) pole region of lead aVL, an initialdownward deflection is recorded in thatlead. At the same time, forces are headingin the direction of the (+) pole region oflead aVF, causing an initial upward deflec-tion to be recorded there. As the wave ofdepolarization spreads through the myo-cardium, the sequence of net electrical chargeoccurs as depicted by the series of arrows inFigure 4.10.

As the lateral walls of the ventricles are de-polarized, the forces of the thicker left sideoutweigh those of the right. Therefore, thearrow swings further and further toward theleft ventricle (leftward and posteriorly). Atthe completion of depolarization, the myo-cytes again become homogeneously charged,

no further net electrical force is generated,and the ECG voltage recording returns tobaseline in both leads. Thus, in this exampleof depolarization in a normal heart, lead aVLinscribes an initial small Q wave followed bya tall R wave. Conversely, in lead aVF, thereis an initial upward deflection (R wave) fol-lowed by a downward S wave.

The sequence of depolarization in thehorizontal plane of the body is also evidenton examining the six chest leads (Fig. 4.11).Once again, recall that the first region to de-polarize is the left ventricular aspect of theinterventricular septum. The sequence ofdepolarization proceeds from the midven-tricular septum toward the right ventricle(which is anterior to the left ventricle), to-ward the cardiac apex, and then around tothe lateral walls of both ventricles. Becausethe initial forces are directed anteriorly—that is, toward the (+) pole of V1—the initialdeflection recorded by lead V1 is upward.These same initial forces are heading awayfrom V6 (which overlies the lateral wall ofthe left ventricle), so an initial downwarddeflection is recorded there. As the wave ofdepolarization spreads, the forces of the leftventricle outweigh those of the right, andthe vector swings posteriorly toward thebulk of the left ventricular muscle. As theforces swing away from lead V1, the deflec-tion there becomes downward, whereas itbecomes more upright in lead V6. Leads V2

through V5 record intermediate steps in thisprocess, such that the R wave becomes pro-gressively taller from lead V1 through leadV6 (see Fig. 4.11E). Typically, the height ofthe R wave becomes greater than the depthof the S wave in lead V3 or V4; the lead inwhich this occurs is termed the “transition”lead.

A normal complete 12-lead ECG is pre-sented later in the chapter (see Fig. 4.28).

The ECG is recorded on a special grid divided into lines spaced 1 mm apart inboth the horizontal and vertical directions.Each fifth line is made heavier to facilitatemeasurement. On the vertical axis, voltageis measured in millivolts (mV), and in thestandard case, each 1-mm line separationrepresents 0.1 mV. The horizontal axis rep-

88 Chapter Four

Fig. 10Fig. 11

10090-04_CH04.qxd 8/31/06 5:25 PM Page 88


Figure 4.10. Normal ventricular depolarization as recorded by leads aVL and aVF. A. In the resting state, thesurface is homogeneously charged so that the leads do not record any electrical potential. B. The first area to depo-larize is the left side of the ventricular septum. This results in forces heading away from aVL (downward deflection onaVL recording) but toward the (+) region of aVF, such that an upward deflection is recorded by that lead. C and D.Depolarization continues; the forces from the thicker-walled left ventricle outweigh those of the right, such that theelectrical vector swings leftward and posteriorly toward aVL (upward deflection) and away from aVF. E. At the com-pletion of depolarization, the surface is again homogeneously charged, and no further electrical voltage is recorded.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 89

90 Chapter Four

Figure 4.11. Sequence of depolarization recorded by the chest (precordial) leads. A–D. Depolar-ization begins at the left side of the septum, and then the forces progress posteriorly toward the left ven-tricle. Thus, V1, which is an anterior lead, records an initial upward deflection followed by a downwardwave, whereas V6, a posterior lead, inscribes the opposite. E. In the normal pattern of the QRS from V1 toV6, the R wave becomes progressively taller and the S wave less deep.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 90


resents time. Because the standard recordingspeed is 25 mm/sec, each 1 mm division rep-resents 0.04 sec and each heavy line (5 mm)represents 0.2 sec (Fig. 4.12).

INTERPRETATION OF THEELECTROCARDIOGRAM

Many cardiac disorders alter the ECG re-cording in a diagnostically useful way. It isimportant to interpret each tracing in astandard fashion to avoid missing subtle ab-normalities. Here is a commonly followedsequence of analysis:

1. Check voltage calibration2. Heart rhythm3. Heart rate4. Intervals (PR, QRS, QT)5. Mean QRS axis

6. Abnormalities of the P wave7. Abnormalities of the QRS (hypertrophy,

bundle branch block, infarction)8. ST segment and T wave abnormalities

Calibration

ECG machines routinely inscribe a 1.0-mVvertical signal at the beginning or end ofeach 12-lead tracing to document the volt-age calibration of the machine. In the nor-mal case, each 1-mm vertical box on theECG paper represents 0.1 mV, so that the cal-ibration signal records a 10-mm deflection(e.g., see Fig. 4.28). However, in patients withmarkedly increased voltage of the QRS com-plex (e.g., some patients with left ventricularhypertrophy or bundle branch blocks), thevery large deflections would not fit on theECG tracing. To facilitate interpretation in

Paper Speed: 25 mm/sec

PR QT

5 mm = 0.5 mV(1 mm = 0.1 mV)

5 mm = 0.2 sec(1 mm = 1 small box = 0.04 sec)

QRS

Figure 4.12. Enlarged view of an ECG strip. The paper travels through the machine at 25 mm/sec, so that each 1 mm on the horizontal axis represents 0.04 sec. Each 1 mm on the verti-cal axis represents 0.1 mV. Interval measurements in this example are as follows: PR interval (fromthe beginning of the P wave to the beginning of the QRS) = small boxes = 0.16 sec; QRS duration (fromthe beginning to the end of the QRS complex) = 1.75 small boxes = 0.07 sec; and QT interval (from thebeginning of the QRS to the end of the T wave) = 8 small boxes = 0.32 sec. The corrected QT

interval = . Because the R–R interval = 15 small boxes (0.6 sec), the corrected QT interval =

= 0.41 sec.0.32

0.6

QT

R R−

Fig. 12

10090-04_CH04.qxd 8/31/06 5:25 PM Page 91

such a case, the recording is purposely madeat half the standard voltage (i.e., each 1-mmbox = 0.2 mV), and this is indicated on theECG tracing by a change in the height of the1.0-mV calibration signal (at half the stan-dard voltage, the signal would be 5 mm tall).It is important to check the height of the cal-ibration signal on each ECG to ensure thatthe voltage criteria used to define specific ab-normalities is applicable.

Heart Rhythm

The normal cardiac rhythm, initiated by de-polarization of the sinus node, is known assinus rhythm and is present if (1) every P wave is followed by a QRS; (2) every QRS is preceded by a P wave; (3) the P wave isupright in leads I, II, and III; and (4) the PRinterval is greater than 0.12 sec (three smallboxes). If the heart rate in sinus rhythm is be-tween 60 and 100 bpm, then normal sinusrhythm is present. If less than 60 bpm, therhythm is sinus bradycardia; if greater than100 bpm, the rhythm is sinus tachycardia.Other abnormal rhythms (termed arrhyth-mias or dysrhythmias) are described in Chap-ters 11 and 12.

Heart Rate

The standard ECG paper speed is 25 mm/sec.Therefore,

or more simply,

It is rarely necessary, however, to determinethe exact heart rate, and a more rapid deter-mination can be made with just a bit ofmemorization. Simply “count off” the num-ber of large boxes between two consecutiveQRS complexes, using the sequence

300—150—100—75—60—50

Heart rate (bpm)Number of smallboxes

= 1 500,

between2 consecutive beats

Heart rate (bpm)mm/sec sec/minNumber

= ×25 60of mm

between beats

which corresponds to the heart rate in beatsper minute, as illustrated in Figure 4.13.

When the rhythm is irregular, the heartrate may be approximated by taking advan-tage of the time markers, spaced 3 secondsapart, often printed at the top or bottom ofthe ECG tracing (see Fig. 4.13, method 3).

Intervals (PR, QRS, QT)

The PR interval, QRS interval, and QT inter-val (see Fig. 4.12) are measured in the limblead recordings. For each of these, it is ap-propriate to take the measurement in thelimb lead in which the interval is the longestin duration (the intervals can vary a bit ineach lead). The PR interval is measuredfrom the onset of the P wave to the onset ofthe QRS. The QRS interval is measured fromthe beginning to the end of the QRS com-plex. The QT interval is measured from thebeginning of the QRS to the end of the T wave. The normal ranges of the intervalsare listed in Table 4.3, along with conditionsassociated with abnormal values.

Because the QT interval varies with heartrate (the faster the heart rate, the shorter theQT), the corrected QT interval is determinedby dividing the measured QT by the squareroot of the R–R interval (see Fig. 4.12). Whenthe heart rate is in the normal range (60 to100 bpm), a rapid rule can be applied: if theQT interval is less than half the interval be-tween two consecutive QRS complexes, thenthe QT interval is within the normal range.

Mean QRS Axis

The mean QRS axis represents the average ofthe instantaneous electrical forces generatedduring the sequence of ventricular depolar-ization. The normal value is between −30°and +90° (Fig. 4.14). A mean axis that is morenegative than −30° implies left axis devia-tion, whereas an axis greater than +90° rep-resents right axis deviation. The axis can beaccurately determined by plotting the QRScomplexes of different leads on the axial ref-erence diagram (see Fig. 4.5), but this is te-dious and rarely necessary. It is generallysufficient to note whether the axis is nor-

92 Chapter Four

Fig. 13

Tab. 3

Fig. 14

10090-04_CH04.qxd 8/31/06 5:25 PM Page 92


Method 1

23 mm between beats

First, count the number of small boxes (1 mm each) between two adjacent QRS complexes (i.e., between 2 “beats”). Then, since the standard paper speed is 25 mm/sec:

Therefore, the heart rate �

Method 1 is particularly helpful for measuring fast heart rates (�100 bpm)

65 bpm� 1,50023

In this example, there are 23 mm between the first 2 beats.

Heart Rate (25 mm/sec � 60 sec/min)(beats/min) Number of mm between beats

�1,500

number of mm between beats�

Starthere

300 100 60150 75

Then use this sequence to count the number of large boxes between twoconsecutive beats:

The second QRS falls between the 75 and 60 bpm; therefore, the heart rate is approximately midway between them � 67 bpm. Knowing that the heart rate is approximately 60–70 bpm is certainly close enough.

The “count-off” method requires memorizing the sequence:

300 - 150 - 100 - 75 - 60 - 50

Method 2

50

Figure 4.13. Methods to calculate heart rate.

mal, deviated to the left, or deviated to theright. If a more precise measurement isneeded, the simplified approach describednext can be employed.

Recall from Figure 4.5 that each ECG leadhas a (+) region and a (−) region. Electricalactivity directed toward the (+) half resultsin an upward deflection, whereas activity to-ward the (−) half results in a downward de-flection on the ECG recording of that lead.

To determine whether the axis is normalor abnormal, examine the QRS complexes inlimb leads I and II. If the QRS is primarilypositive in both of these leads (upward de-

flection greater than downward deflection),then the mean vector falls within the normalrange (Fig. 4.15).a If the QRS in either lead I orII is not primarily upward, then the axis isabnormal, and the approximate axis shouldthen be determined by the rapid method de-scribed in the following paragraphs.

aNote that some textbooks recommend (and some prac-titioners prefer) examining leads I and aVF, rather thanleads I and II, to determine if the mean axis is normal.This is acceptable, but be aware that a mean axis thatfalls between 0° and −30° would be erroneously classi-fied as “abnormal” by that method.

Ftn. a,Fig. 15

10090-04_CH04.qxd 8/31/06 5:25 PM Page 93

First, consider the special example in Fig-ure 4.16. A sequence of ventricular depolar-ization is represented in this figure by arrowsa through e. The initial deflection (repre-senting left septal depolarization) points tothe patient’s right side. Because it is directedcompletely away from the (+) pole of lead I,a strong downward deflection is recorded bythe lead. As depolarization continues, thearrow swings downward and to the left, re-sulting in less negative deflections in lead I.

After arrow c, the electrical vector swingsinto the positive region of lead I, so that up-ward deflections are recorded.

In this special example, in which electri-cal forces begin exactly opposite lead I’s (+)electrode and terminate when pointed di-rectly at that electrode, note that the meanelectrical vector points straight downward(in the direction of arrow c), perpendicular tothe lead I axis. Also notice the configurationof the inscribed QRS complex in lead I.

94 Chapter Four

ECG recording paper often indicates 3-sec time markers at the top or bottom of the tracing:

Method 3

To calculate the heart rate, count the number of QRS complexes between the 3-sec markers (� 6 beats in this example) and multiply by 20. Thus, the heart rate here is approximately 120 bpm.

It’s even easier (and more accurate) to count the number of complexes between the first and third markers on the strip (representing 6 secof the recording) and then multiply by 10 to determine the heart rate.

Method 3 is particularly helpful for measuring irregular heart rates.

3 secmarker marker

Figure 4.13. (Continued) Methods to calculate heart rate.

TABLE 4.3. Electrocardiographic Intervals

Interval Normal Decreased In Increased In

PR 0.12–0.20 sec • Preexcitation syndrome • First-degree AV block(3–5 small boxes) • Junctional rhythm

QRS ≤0.10 sec • Bundle branch blocks(≤2.5 small boxes) • Ventricular ectopic beat

• Toxic drug effect (e.g., quinidine)• Severe hyperkalemia

QT Corrected QTa ≤0.44 sec • Hypercalcemia • Hypocalcemia• Tachycardia • Hypokalemia (↑ QT interval owing to

↑ T wave)• Hypomagnesemia• Myocardial ischemia• Congenital prolongation of QT• Toxic drug effect (e.g., quinidine)

aCorrected QT is negative.

AQ1,AQ2

AQ3

Fig. 16

10090-04_CH04.qxd 8/31/06 5:25 PM Page 94


There is a downward deflection, followedby an upward deflection of equal magni-tude (when the upward and downward de-flections of a QRS are of equal magnitude, it is termed an isoelectric complex). Thus,when an ECG limb lead inscribes an isoelectricQRS complex, it indicates that the mean elec-trical axis of the ventricles is perpendicular tothat lead.

Therefore, an easy way to determine themean QRS axis is to glance at the six limblead recordings and observe which one hasthe most isoelectric-appearing complex: themean axis is simply perpendicular to it. Onestep remains. When the mean axis is per-pendicular to a lead, it could be perpendicu-lar in either a clockwise or a counterclock-wise direction. In the example of Figure4.16, the isoelectric complex appears in leadI, such that the mean vector could be at +90°or it could be at −90°, because both are per-pendicular. Determining which of these it isrequires inspecting the recording of the ECGlead that is perpendicular to the one inscrib-ing the isoelectric complex (and is thereforeparallel to the mean axis). If the QRS is pre-dominantly upright in that perpendicularlead, then the mean vector points toward

the (+) pole of that lead. If it is predomi-nantly negative, then it points away fromthe lead’s (+) pole. In the example, the iso-electric complex appears in lead I; therefore,the next step is to inspect the perpendicularlead, which is aVF (see Fig. 4.5 if this rela-tionship is not clear). Because the QRS com-plex in aVF is primarily upward, the meanaxis points toward its (+) pole, which is infact located at +90°.

To summarize, the mean QRS axis is cal-culated as follows:

1. Inspect limb leads I and II. If the QRS isprimarily upward in both, then the axisis normal and you are done. If not, thenproceed to the next step.

2. Inspect the six limb leads and determinewhich one contains the QRS that is mostisoelectric. The mean axis is perpendicu-lar to that lead.

3. Inspect the lead that is perpendicular tothe lead containing the isoelectric com-plex. If the QRS in that perpendicularlead is primarily upward, then the meanaxis points to the (+) pole of that lead. Ifprimarily negative, then the mean QRSpoints to the (−) pole of that lead.

Left axis deviation Right axis deviation

aVR

aVF

aVL

III

�150°

�180°

Left axis deviation• Inferior wall myocardial infarction• Left anterior fascicular block• Left ventricular hypertrophy (sometimes)

Right axis deviation• Right ventricular hypertrophy • Acute right heart strain (e.g., massive pulmonary embolism)• Left posterior fascicular block

�30°

�60°�90°

�120°

�150°

�120°�90°

�60°

�30°

0°

II

I

Figure 4.14. A normal mean QRS axis falls within the shaded area (between �30º and++90°°). A mean axis more negative than −30° is termed left axis deviation, whereas an axis morepositive than +90° is right axis deviation. The figure shows common conditions that result in axisdeviation.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 95

Conditions that result in left or right axisdeviation are listed in Figure 4.14. In addi-tion, the vertical position of the heart inmany normal children and adolescents mayresult in a slightly rightward mean axis(>+90°).

In some patients, isoelectric complexesare inscribed in all the limb leads. That situ-ation arises when the heart is tilted, so thatthe mean QRS is pointing straight forward

or back from the chest, as it may be in pa-tients with chronic obstructive lung disease;in such a case, the mean axis is said to beindeterminate.

Abnormalities of the P Wave

The P wave represents depolarization of theright atrium followed quickly by depolariza-tion of the left atrium; the two components

96 Chapter Four

–90°

+90°

0°

+150°

+60°

–30°

–30°

+90°

If the QRS is predominantly upright in both leads I and II, then the mean axis must fall within their common “+” regions: between –30° and +90°.

If the QRS complex is mainly upward in limb lead I, then the mean axis falls within the “+” region of that lead, shown as the shaded half of the circle below.

Similarly, if the QRS is predominantly upward in limb lead II, then the mean axis falls within the “+” half of lead II, shown as the shaded half here:

NormalAxis

Figure 4.15. The mean axis is within the normal range if the QRS complex is predominantly upright in limbleads I and II.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 96


are nearly superimposed on one another(Fig. 4.17). The P wave is usually best visual-ized in lead II, the lead that runs most par-allel to the flow of electrical current throughthe atria from the sinoatrial to the AV node.When the right atrium is enlarged, the initial

component of the P wave is larger than nor-mal (taller than 2.5 mm in lead II).

Left atrial enlargement is best observed inlead V1. Normally, V1 inscribes a P wave withan initial positive deflection reflecting rightatrial depolarization (directed anteriorly),followed by a negative deflection, owing tothe left atrial forces oriented posteriorly (seeFig. 1.2 for anatomic relationships). Left atrialenlargement is therefore manifested by agreater-than-normal negative deflection (atleast 1 mm wide and 1 mm deep) in lead V1

(see Fig. 4.17).

Abnormalities of the QRS Complex

Ventricular Hypertrophy

Hypertrophy of the left or right ventriclecauses the affected chamber to generategreater-than-normal electrical activity. Or-dinarily, the forces produced by the thicker-walled left ventricle are more prominentthan those of the right. However, in rightventricular hypertrophy, the augmented

Lead aVF

Lead I

a bb c

c

ed

d

Meanaxis

a e

Figure 4.16. Sequence of ventricular depolarizationwhen the mean axis is ++90°°. Because the mean axis isperpendicular to limb lead I, an isoelectric QRS complex(height of upward deflection = height of downward de-flection) is recorded by that lead (see text for details).

Figure 4.17. The P wave represents superimposition of right atrial (RA) and left atrial(LA) depolarization. RA depolarization occurs slightly earlier than LA depolarization. In RAenlargement, the initial component of the P wave is prominent (>2.5 mm tall) in lead II. In LAenlargement, there is a large terminal downward deflection in lead V1 (>1 mm wide and >1 mm deep).

Fig. 17

10090-04_CH04.qxd 8/31/06 5:25 PM Page 97

right-sided forces may outweigh those of theleft. Therefore, chest leads V1 and V2, whichoverlie the right ventricle, record greater-than-normal upward deflections: the R wavebecomes taller than the S wave in thoseleads, the opposite of the normal situation(Fig. 4.18). In addition, the increased rightventricular mass shifts the mean axis of theheart, resulting in right axis deviation(greater than +90°).

In left ventricular hypertrophy, greater-than-normal forces are generated by thatchamber, which simply exaggerates the nor-mal situation. Leads that directly overlie theleft ventricle (chest leads V5 and V6 and limb

leads I and aVL) show taller-than-normal Rwaves. Leads on the other side of the heart (V1

and V2) demonstrate the opposite: deeper-than-normal S waves. Many criteria havebeen proposed for the diagnosis of left ven-tricular hypertrophy by ECG. Three of themost helpful criteria are listed in Figure 4.18.

Bundle Branch Blocks

Interruption of conduction through the rightor left bundle branches may develop from is-chemic or degenerative damage. As a result,the affected ventricle does not depolarize inthe normal sequence. Rather than rapid uni-

98 Chapter Four

RIGHT VENTRICULAR HYPERTROPHY

• R > S in lead V1

• Right axis deviation

LEFT VENTRICULAR HYPERTROPHY

• S in V1 plus

R in V5 or V6 ≥ 35 mm or

• R in aVL > 11 mm or

• R in Lead I > 15 mm

V6

V6

V1

V1

A

B

LVRV

4

3

2

3

21

1

Figure 4.18. Ventricular hypertrophy. The arrows indicate the sequence of average electrical forcesduring ventricular depolarization. A. Right ventricular (RV) hypertrophy. The RV forces outweigh those ofthe left, resulting in tall R waves in leads V1 and V2 and a deep S wave in lead V6. B. Left ventricular (LV)hypertrophy exaggerates the normal pattern of depolarization, with greater-than-normal forces directedtoward the LV, resulting in a tall R wave in V6 and a deep S wave in lead V1.

Fig. 18

10090-04_CH04.qxd 8/31/06 5:25 PM Page 98


form stimulation by the Purkinje fibers, thecells of that ventricle must rely on relativelyslow myocyte-to-myocyte spread of electricalactivity traveling from the unaffected ventri-cle. This delayed process prolongs depolar-ization and widens the QRS complex. Whena bundle branch block widens the QRS dura-tion to 0.10 to 0.12 sec (2.5 to 3.0 small

boxes), an incomplete bundle branch block ispresent. If the QRS duration is greater than0.12 sec (3.0 small boxes), complete bundlebranch block is identified.

In right bundle branch block (Fig. 4.19A;see also Fig. 4.29), normal depolarization ofthe right ventricle does not occur. In thiscase, initial depolarization of the ventricular

Figure 4.19. Bundle branch blocks. Interruption of conduction through the right(RBBB) or left (LBBB) bundles results in delayed, slowed activation of the respectiveventricle and widening of the QRS complex. A. In RBBB, normal initial activation ofthe septum (1) is followed by depolarization of the left ventricle (2). Slow cell-to-cell spread activates the right ventricle (RV) after the left ventricle (LV) has nearlyfully depolarized, so that the late forces generated by the RV are unopposed.Therefore, V1 records an abnormal terminal upward deflection (R′), and V6 recordsan abnormal, terminal deep S wave (3). B. In LBBB, the initial septal depolarizationis blocked, such that initial forces are oriented from right to left. Thus, the normalinitial R wave in V1 and Q wave in V6 are absent (1). After the RV depolarizes, late,slow activation of the LV results in a terminal upward deflection in V6 and down-ward deflection in V1 (3).

Fig. 19

10090-04_CH04.qxd 8/31/06 5:25 PM Page 99

septum (which is stimulated by a branch ofthe left bundle) is unaffected so that thenormal small R wave in lead V1 and small Qwave in lead V6 are recorded. As the wave ofdepolarization spreads down the septumand into the left ventricular free wall, the se-quence of depolarization is indistinguish-able from normal, because left ventricularforces normally outweigh those of the right.However, by the time the left ventricle hasalmost fully depolarized, slow cell-to-cellspread has finally reached the “blocked” rightventricle and depolarization of that cham-ber begins, unopposed by left ventricularactivity (because that chamber has nearlyfully depolarized). This prolonged depolar-ization process widens the QRS complex,and since the terminal portion of the QRScomplex represents right ventricular forcesacting alone, there is a terminal upward de-flection (known as R′ wave) over the rightventricle in lead V1, and a downward de-flection (S wave) in V6 on the opposite side ofthe heart.

Left bundle branch block produces evengreater QRS abnormalities. In this situation,normal initial depolarization of the left sep-tum does not occur; rather, the right side of the ventricular septum is first to depolar-ize, through branches of the right bundle.Thus, the initial forces of depolarization aredirected toward the left ventricle instead ofthe right (see Fig. 4.19B; see also Fig. 4.30).Therefore, an initial downward deflection isrecorded in V1 and the normal small Q wavein V6 is absent. Only after depolarization of the right ventricle does slow cell-to-cell spread reach the left ventricular myo-cytes. These slowly conducted forces inscribea widened QRS complex with terminallyupward deflections in the leads overlyingthe left ventricle (V5 and V6), as shown inFigure 4.19B.

Recall from Chapter 1 that the left bundlebranch subdivides into two main divisions,termed fascicles: the left anterior fascicle andleft posterior fascicle. Although a left bundlebranch block implies that conduction isblocked in the entire left bundle branch, im-pairment can also occur in just one of thetwo fascicles, resulting in left anterior or left

posterior fascicular blocks (also termed hemi-blocks). The main significance of fascicularblocks in ECG interpretation is that they canmarkedly alter the mean ECG axis.

Anatomically, the anterior fascicle of theleft bundle runs along the front of the leftventricle toward the anterior papillary mus-cle (which is located in the anterior and superior portion of the chamber), whereasthe posterior fascicle travels to the posteriorpapillary muscle (which is located in theposterior, inferior, and medial aspect of theleft ventricle). Under normal conditions,conduction via the left anterior and left posterior fascicles proceeds simultaneously,such that electrical activation of the leftventricle is uniform, spreading from thebases of the two papillary muscles. However,if conduction is blocked in one of the twodivisions, then initial LV depolarizationarises exclusively from the unaffected fasci-cle (Fig. 4.20).

In the case of left anterior fascicularblock (LAFB), left ventricular activation be-gins via the left posterior fascicle alone, atthe posterior papillary muscle, and thenspreads to the rest of the ventricle. Becausethe left posterior fascicle first activates theposterior, inferior, medial region of the leftventricle, the initial impulses are directeddownward (i.e., toward the feet) and towardthe patient’s right side (see Fig. 4.20). Thisresults in a positive deflection (initial smallR wave) in the inferior leads (leads II, III, andaVF) and a negative deflection (small Q wave)in the left lateral leads, I and aVL. As depo-larization then spreads upward and to theleft, toward the “blocked” anterior, superior,and lateral regions of the left ventricle, apositive deflection (R wave) is inscribed inleads I and aVL, while a negative deflection(S wave) develops in the inferior leads. Thepredominance of these leftward forces, re-sulting from the abnormal activation of theanterior superior left ventricular wall resultsin left axis deviation (generally more nega-tive than −45 degrees). A complete 12-leadECG demonstrating LAFB is shown later(see Fig. 4.34).

Left posterior fascicular block (LPFB) isless common than LAFB. In LPFB, ventricu-

100 Chapter Four

Fig. 20

10090-04_CH04.qxd 8/31/06 5:25 PM Page 100

L AFB LPFB

L

X

X

• Left axis deviation• Small Q in leads aVL and I• Small R in inferior leads (II, III, aVF)

1

2

3

• Right axis deviation• Small R in leads aVL and I• Small Q in inferior leads (II, III, aVF)

aVL

aVF

aVL

aVF

aVL

aVF

aVL

aVF

aVL

aVF

aVL

aVF

PV

AMV

Figure 4.20. Left anterior and left posterior fascicular blocks. The schematic at the top of the figure shows the leftventricle (LV) in the frontal plane. The mitral valve (MV) chordae tendineae insert into the anterior (A) and posterior (P)papillary muscles, which are important landmarks: the anterior fascicle of the left bundle branch courses toward the an-terior papillary muscle, whereas the posterior fascicle travels to the posterior papillary (the fascicles are not shown). No-tice that the anterior papillary muscle is actually superior to the posterior papillary muscle. Left side of figure: In leftanterior fascicular block (LAFB), activation begins solely in the region of the posterior papillary muscle (panel 1) becauseinitial conduction to the anterior papillary muscle is blocked (denoted by the X). As a result, the initial forces of depo-larization are directed downward towards the feet, producing an initial positive R wave in lead aVF and a negative Q wavein lead aVL. As the wave of depolarization spreads toward the left side and superiorly, aVF begins to register a negativedeflection and aVL starts to record a positive deflection (panel 2). Panel 3 shows the complete QRS complexes at the endof depolarization. Right side of figure: In left posterior fascicular block (LPFB; denoted by the X), LV activation beginssolely in the region of the anterior papillary muscle (panel 1). Thus, the initial forces are directed upward and toward thepatient’s left side, producing an initial R wave in aVL and a Q wave in aVF. Panels 2 and 3 show how the spread of de-polarization travels in the direction opposite that of LAFB.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 101

lar activation begins via the left anterior fas-cicle at the base of the anterior papillarymuscle (see Fig. 4.20). As that anterosupe-rior left ventricular region depolarizes, theinitial forces are directed upward and to thepatient’s left (creating a positive R wave inleads I and aVL and a negative Q wave inthe inferior leads). As the impulse thenspreads downward and to the right towardthe initially blocked region, an S wave is in-scribed in leads I and aVL, while an R waveis recorded in leads II, III, and aVF. Becausethe bulk of these delayed forces head to-ward the patient’s right side, right axis devi-ation of the QRS mean axis is expected (seeFig. 4.36).

LAFB and LPFB do not result in signifi-cant widening of the QRS (in contrast toright or left bundle branch blocks) becauserapidly conducting Purkinje fibers bridgethe territories served by the anterior andposterior fascicles. Therefore, although thesequence of conduction is altered, the totaltime required for depolarization is usuallyonly slightly prolonged. Also note that al-though left and right bundle branch blocksare most easily recognized by analyzing thepatterns of depolarization in the precordialleads, in the case of LAFB and LPFB, it is therecordings in the limb leads (as in Fig. 4.20)that are most helpful.

Myocardial Infarction

The hallmark of transmural myocardial in-farction (MI) is the pathologic Q wave. Re-call that it is normal for an initial Q wave to

appear in some leads. For example, initialseptal depolarization routinely inscribes asmall Q wave in leads V6 and aVL. Normal Q waves are of short duration (≤0.04 sec, or 1 small box) and of low magnitude (<25% ofthe QRS total height). A pathologic Q waveis more prominent (Fig. 4.21), having a width≥1 small box in duration and a depth >25%of the total height of the QRS. The ECGleads in which pathologic Q waves appearreflect the anatomic site of an infarction(Table 4.4; see Fig. 4.22).

Pathologic Q waves develop in the leadsoverlying infarcted tissue because necroticmuscle does not generate electrical forces.Rather, the ECG electrode over that regiondetects electrical currents from the healthytissue on opposite regions of the ventricle,which are directed away from the infarct andthe recording electrode, thus inscribing the

102 Chapter Four

Figure 4.21. Compared with small Q waves gener-ated during normal depolarization, pathologic Q waves are more prominent with a width ≥≥1 mm (1 small box) and depth >25% of the height of theQRS complex.

TABLE 4.4. Localization of Myocardial Infarction

Anatomic Site Leads with Abnormal ECG Complexesa Coronary Artery Most Often Responsible

Inferior II, III, aVF RCAAnteroseptal V1–V2 LADAnteroapical V3–V4 LAD (distal)Anterolateral V5–V6, I, aVL CFXPosterior V1–V2 (tall R wave, not Q wave) RCA

aPathologic Q waves in all of leads V1–V6 implies an “extensive anterior MI” usually associated with a proximal left coronaryartery occlusion.

CFX, left circumflex coronary artery; LAD, left anterior descending coronary artery; RCA, right coronary artery, TSP1;

Fig. 21

Tab. 4,Fig. 22

10090-04_CH04.qxd 8/31/06 5:25 PM Page 102


downward deflection (Fig. 4.23). Q waves arepermanent evidence of a myocardial infarc-tion; only rarely do they disappear over time.

Notice in Table 4.4 that in the case of aposterior wall myocardial infarction (see Fig.4.22), it is not pathologic Q waves that areevident on the ECG. Because standard elec-trodes are not placed on the patient’s backoverlying the posterior wall, other leads mustbe relied on to indirectly identify the pres-ence of such an infarction. Chest leads V1

and V2, which are directly opposite the pos-terior wall, record the inverse of what leadsplaced on the back would demonstrate.Therefore, taller-than-normal R waves in leadsV1 and V2 are the equivalent of a pathologicQ wave in the diagnosis of a posterior wallMI. It may be recalled that right ventricularhypertrophy also produces tall R waves inleads V1 and V2, but unlike right ventricularhypertrophy, right axis deviation is not usu-ally a feature of posterior wall MI.

It is important to note that if a Q wave ap-pears in only a single ECG lead, it is not di-agnostic of an infarction. True pathologic Qs should appear in the groupings listed inTable 4.4 and Figure 4.22. For example, if apathologic Q wave is present in lead III butnot in II or aVF, it likely does not indicate aninfarction. Also, Q waves are disregarded inlead aVR because electrical forces are normallydirected away from the right arm. Finally, inthe presence of left bundle branch block, Qwaves are usually not helpful in the diagno-sis of MI because of the markedly abnormalpattern of depolarization in that condition.

In infarctions in which Q waves develop,appropriately termed Q-wave infarctions,the entire thickness of a myocardial segmentis usually involved. Therefore, this type ofMI is often called a transmural infarct. As de-scribed in Chapter 7, infarctions are not al-ways transmural but may involve only thesubendocardial layers of the myocardium. Inthis case, pathologic Q waves do not usuallydevelop, because the remaining viable cellsare able to generate some electrical activity;such MIs are therefore called non–Q-waveinfarctions. In either case, certain ST seg-ment and T wave abnormalities evolve dur-ing acute Q-wave and non–Q-wave infarc-

Figure 4.22. Relationship between ECG leads andcardiac anatomic regions. A. The leads listed in paren-theses are those that reflect infarction of these regions. B. Miniaturized schematic drawings of a 12-lead ECGshowing the standard orientation of printed samplesfrom each lead. The major anatomic groupings are indi-cated. Note that while the presence of pathologic Q wavesin leads V1 and V2 are indicative of anteroseptal infarc-tion, tall initial R waves in those leads are seen in poste-rior wall infarction.

Fig. 23

10090-04_CH04.qxd 8/31/06 5:25 PM Page 103

tions, as discussed in the next section. Theelectrocardiographic differences betweenthese types of MI are summarized as follows:

polarization is very sensitive to myocardialperfusion, patients with coronary artery dis-ease often demonstrate reversible deviationsof the ST segments and T waves during myo-cardial ischemia.

As described in the previous section, pa-thologic Q waves are indicative of an MI butdo not differentiate between an acute eventand an MI that occurred weeks or years ear-lier. However, acute MI does result in a se-quence of ST and T wave abnormalities thatpermit this distinction (Fig. 4.24). The initialabnormality during an acute Q-wave MI iselevation of the ST segment, often with apeaked appearance of the T wave. At thisearly stage, myocardial cells are still viableand Q waves have not yet developed. Withinseveral hours, however, myocyte death leadsto loss of the amplitude of the R wave, andpathologic Q waves begin to be inscribed by

104 Chapter Four

Anterior(V1–V6)

Anterolateral(V5–V6, I, aVL)

Anteroseptal(V1–V2)

Posterior(Tall R inV1–V2)

Inferior(II, III, aVF)Anteroapical

(V3–V4)

Anteroseptal Anteroapical

Inferior

Anterolateral

A

B

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 4.23. Sequence of depolarization recorded by lead aVL, overlying a lateral wall infarction(black region). A pathologic Q wave is recorded because the necrotic muscle does not generate electricalforces; rather, at the time when the lateral wall should be depolarizing (panel 3), the activation of the healthymuscle on the opposite side of the heart is unopposed, such that forces head away from aVL. The terminalR wave recorded by aVL reflects depolarization of the remaining viable myocardium beyond the infarct.

Acute ST Type of Pathologic Segment Infarction Q Waves Deviation

Q-wave MI Yes ST elevationNon–Q-wave No ST depression

MI (and/or T wave inversion)

ST Segment and T WaveAbnormalities

Among the most common important ab-normalities of the ST segments and T wavesare those that represent myocardial ische-mia and infarction. Because ventricular re-

Fig. 24

10090-04_CH04.qxd 8/31/06 5:25 PM Page 104


the ECG leads positioned over the infarct ter-ritory. During the first 1 to 2 days followinginfarction, the ST segments remain elevated,the T wave inverts, and the Q wave deepens.Several days later, the ST segment elevationreturns to baseline, but the T waves remaininverted. Weeks or months following the in-farct, the ST segment and T waves have oftenreturned to normal, but the pathologic Qwaves persist, a permanent marker of the MI.If the ST segment remains elevated severalweeks later, it is likely that a bulging fibroticscar (ventricular aneurysm) has developed atthe site of infarction.

These evolutionary changes of the QRS,ST, and T waves are recorded by the leadsoverlying the zone of infarction (see Table4.4). Typically, reciprocal changes are seen in leads opposite that site. For example, inacute anteroseptal MI, ST segment elevationis expected in chest leads V1 and V2; simulta-neously, however, reciprocal changes (ST de-pression) may be inscribed by the leads over-lying the opposite (inferior) region, namelyin leads II, III, and aVF.

The mechanism by which ST segment de-viations develop during acute MI has notbeen established with certainty. It is believed,however, that the abnormality results frominjured myocardial cells immediately adja-cent to the infarct zone producing abnormalsystolic or diastolic currents. One explana-tion, the diastolic current theory, contends thatthese cells are capable of depolarization butare abnormally “leaky” for potassium ions,preventing them from ever fully repolarizing(Fig. 4.25). Because the surface of such par-tially depolarized cells in the resting (dias-tolic) state would be relatively negatively

charged compared with the normally fullyrepolarized areas, an electrical current is gen-erated between the two regions. This currentis directed away from the more negativelycharged ischemic area, causing the baselineof the ECG leads overlying that region to shiftdownward. Because the ECG machine Record-ing only relative position, rather than ab-solute voltages, ECG machines do not makethe downward deviation of the baseline no-ticeable. Following ventricular depolariza-tion (indicated by the QRS complex), after allthe myocardial cells have fully depolarized(including those of the injured zone), the netelectrical potential surrounding the heart istrue zero. However, compared with the ab-normally displaced downward baseline, thereis the appearance of ST segment elevation (seeFig. 4.25). As the myocytes then repolarize,the injured cells return to the abnormal stateof diastolic potassium ion leak, and the ECGagain inscribes the abnormally depressedbaseline. Thus, ST elevation in acute MI mayin part reflect an abnormal shift of the record-ing baseline.

In non–Q-wave myocardial infarctions, itis ST segment depression, rather than eleva-tion, that often develops in the leads over-lying the infarct (see Chapter 7). In this situ-ation, the diastolic potassium leak of injuredcells adjacent to the infarct generates elec-trical forces heading from the inner endo-cardium to the outer epicardium and there-fore toward the overlying ECG electrode.Thus, the baseline of the ECG is shifted up-ward (see Fig. 4.25). Following full cardiac de-polarization, the electrical potential of theheart returns to true zero but, relative to theabnormal baseline, gives the appearance ofST segment depression.

Figure 4.24. ECG evolution during acute Q-wave myocardial infarction (also termed acute ST-elevation MI).

Fig. 25

10090-04_CH04.qxd 8/31/06 5:25 PM Page 105

The systolic current theory of ST segmentshifts contends that in addition to reducingthe resting membrane potential, ischemicinjury shortens the action potential dura-tion of affected cells. As a result, the is-chemic cells repolarize faster than neigh-boring normal myocytes; therefore, a voltagegradient develops between the two zones,creating an electrical current directed towardthe ischemic area. This gradient occurs dur-ing the ST interval of the ECG, resulting inST elevation in the leads overlying the is-chemic region (Fig. 4.26).

As discussed in Chapter 7, when evaluat-ing a patient with acute chest pain, it is veryimportant to identify and rapidly distinguish

between Q-wave and non–Q-wave myocar-dial infarctions, because the critical thera-peutic approaches are different. Decisionsabout therapy must be made within min-utes of evaluating the patient, usually whileacute ST and T wave deviations are presenton the ECG but before Q waves would be ex-pected to have formed. Thus, for the pur-pose of such decision making, an evolvingQ-wave MI is referred to as an acute ST-eleva-tion MI. Similarly, a non–Q-wave MI is fre-quently labeled a non–ST-elevation MI.

Other common causes of ST segment andT wave abnormalities caused by alterationsin myocyte repolarization are illustrated inFigure 4.27.

106 Chapter Four

Normal baseline

Baselineshiftedupward

Heart fully depolarized

Baseline shifted downward

Normal baseline

Heart fully depolarized

Injured segment ispartially depolarizedprior to stimulation

Injured segment ispartially depolarizedprior to stimulation

Recordingelectrode

Recordingelectrode

Transmural MI

Nontransmural MI

Figure 4.25. ST deviations in acute MI: diastolic injury current. Top. Ionic leak re-sults in partial depolarization of injured myocardium in diastole, prior to electrical stimu-lation, which produces forces heading away from that site and shifts the ECG baselinedownward. This is not noticeable on the ECG because only relative, not absolute, volt-ages are recorded. Following stimulation, when the entire myocardium has fully depo-larized, the voltage is true zero but gives the appearance of ST elevation compared withthe abnormally depressed baseline. Bottom. In nontransmural MI, the process is similar,but the ionic leak arises from the subendocardial tissue so that the partial depolarizationbefore stimulation is directed toward the recording electrode; hence, the baseline isshifted upward. When fully depolarized, the voltage is true zero, but the ST segment ap-pears depressed compared with the shifted baseline.

Fig. 26

Fig. 27

10090-04_CH04.qxd 8/31/06 5:25 PM Page 106


Ischemic

Ischemic cellsrepolarize more

idl th l

Normal–90 mV

ST elevation+

++

Recordingl t d

Figure 4.26. ST deviation in acute MI: systolic injury current. A. Compared with normal myocytes (solidline), ischemic myocytes (dashed line) display a reduced resting membrane potential and repolarize more rapidly.B. More rapid repolarization causes the surface of the ischemic zone to be relatively positively charged at thetime the ST segment is inscribed. The associated electrical current (arrows) is directed toward the recording elec-trode overlying that site, so that the ST segment is abnormally elevated.

Figure 4.27. Conditions that alter repolarization of myocytes and therefore result in ST seg-ment and T wave abnormalities.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 107

SUMMARY

The electrocardiogram provides a wealth ofinformation about the structure and integrityof the heart and is one of the most impor-tant diagnostic tools in cardiology. With theknowledge of this chapter in hand, the readershould be well prepared to practice analyz-ing electrocardiograms in any of the com-plete ECG texts listed under “AdditionalReading.” Table 4.5 summarizes the sug-gested sequence of ECG interpretation. Sam-

ple normal and abnormal ECGs, with theirinterpretations, are shown in Figures 4.28through 4.36.

Disturbances of the cardiac rhythm iden-tified by ECGs are discussed in Chapters 11and 12.

Acknowledgments

Contributors to the previous editions of this chapterwere Price Kerfoot, MD; Kyle Low, MD; and LeonardS. Lilly, MD.

108 Chapter Four

TABLE 4.5. Summary of Sequence of ECG Interpretation

1. Calibration• Check 1.0 mV vertical box inscription (normal standard = 10 mm)

2. Rhythm• Sinus rhythm is present if

• Each P wave is followed by a QRS complex• Each QRS is preceded by a P wave• P wave is upright in leads I, II, and III• PR interval is >0.12 sec (3 small boxes)

• If these criteria are not met, determine type of arrhythmia (see Chapter 12)3. Heart rate

• Use one of three methods:• 1,500/(number of mm between beats)• Count-off method: 300—150—100—75—60—50• Number of beats in 6 seconds ∞ 10

• Normal rate = 60–100 bpm (bradycardia <60, tachycardia >100)4. Intervals

• Normal PR = 0.12–0.20 sec (3–5 small boxes)• Normal QRS ≤ 0.10 sec (≤2.5 small boxes)• Normal QT ≤ half the R–R interval, if heart rate normal

5. Mean QRS axis• Normal if QRS is primarily upright in leads I and II (+90° to −30°)• Otherwise, determine axis by isoelectric or perpendicular method

6. P wave abnormalities• Inspect P in leads II and V1 for left and right atrial enlargement

7. QRS wave abnormalities• Inspect for left and right ventricular hypertrophy• Inspect for bundle branch blocks• Inspect for pathologic Q waves: What anatomic distribution?

8. ST segment or T wave abnormalities• Inspect for ST elevations:

• Transmural infarct pattern• Pericarditis (see Chapter 14)

• Inspect for ST depressions or T wave inversions:• Subendocardial ischemia or infarct• Commonly accompany ventricular hypertrophy or bundle branch blocks• Metabolic or chemical abnormalities (see Fig. 4.27)

9. Compare with patient’s previous ECGs

Tab. 5

Fig. 28-36

10090-04_CH04.qxd 8/31/06 5:25 PM Page 108

I II III

aVR

aVL

aVF

V1

V2

V3

V4

V5

V6

Fig

ure

4.2

8.12

-lea

d E

CG

(n

orm

al).

The

rect

angu

lar

upw

ard

defle

ctio

n at

the

beg

inni

ng o

f ea

ch li

ne is

the

vol

tage

cal

ibra

tion

sign

al (

1 m

V).

Rhyt

hm:

norm

al s

inus

. Ra

te:

70 b

pm.

Inte

rval

s:PR

, 0.1

7; Q

RS, 0

.06;

QT,

0.4

0 se

c. A

xis:

0°(Q

RS is

isoe

lect

ric in

lead

aV

F). T

he P

wav

e, Q

RS c

ompl

ex, S

T se

gmen

t, a

nd T

wav

es a

re n

orm

al. N

otic

e th

e gr

adua

l inc

reas

e in

Rw

ave

heig

ht b

etw

een

lead

s V

1th

roug

h V

6.

6

109

10090-04_CH04.qxd 8/31/06 5:25 PM Page 109

I II III

aVR

aVL

aVF

V1

V2

V3

V4

V5

V6

Fig

ure

4.2

9.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:nor

mal

sin

us. R

ate:

75 b

pm. I

nter

vals

:PR,

0.1

6; Q

RS, 0

.15;

QT,

0.4

2 se

c. A

xis:

inde

term

inat

e (is

oele

ctric

in a

ll lim

b le

ads)

. P w

ave:

left

atria

l enl

arge

men

t (1

mm

wid

e an

d 1

mm

dee

p in

lead

V1)

. Q

RS:

wid

ened

with

RSR

′in

lead

V1

cons

iste

nt w

ith r

ight

bun

dle

bran

ch b

lock

(RB

BB).

Als

o, p

atho

logi

c Q

wav

es a

re in

lead

s II,

III,

and

aVF,

con

sist

ent

with

infe

rior

wal

l myo

card

ial i

nfar

ctio

n (a

n ol

d on

e, b

ecau

se t

he S

T se

gmen

ts d

o no

t de

mon

stra

te a

n ac

ute

inju

ry p

atte

rn).

110

10090-04_CH04.qxd 8/31/06 5:25 PM Page 110

Fig

ure

4.3

0.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:nor

mal

sin

us. R

ate:

68 b

pm. I

nter

vals

:PR,

0.1

6; Q

RS, 0

.16;

QT,

0.4

0 se

c. A

xis:

+15°

. P w

ave:

norm

al. Q

RS:w

iden

ed w

ith R

R′in

lead

s V

4–V

6co

nsis

tent

with

left

bun

dle

bran

ch b

lock

(LBB

B). T

he S

T se

gmen

t an

d T

wav

e ab

norm

aliti

es a

re s

econ

dary

to

LBBB

.

111

10090-04_CH04.qxd 8/31/06 5:25 PM Page 111

Fig

ure

4.3

1.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:nor

mal

sin

us. R

ate:

66 b

pm. I

nter

vals

:PR,

0.1

6; Q

RS, 0

.08;

QT,

0.4

0 se

c. A

xis:

+10°

. P w

ave:

norm

al. Q

RS:p

atho

logi

c Q

wav

es in

lead

s V

1–V

4, c

onsi

sten

t w

ith a

nter

osep

tal a

nd a

nter

oapi

cal m

yoca

rdia

l inf

arct

ion

(MI).

The

ST

segm

ent

and

T w

aves

do n

ot d

emon

stra

te a

n ac

ute

inju

ry p

atte

rn; t

hus,

the

MI i

s ol

d.

112

10090-04_CH04.qxd 8/31/06 5:25 PM Page 112

Fig

ure

4.3

2.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:sin

us b

rady

card

ia. R

ate:

55 b

pm. I

nter

vals

:PR,

0.2

0 (in

aV

F); Q

RS, 0

.10;

QT,

0.4

4 se

c. A

xis:

norm

al (Q

RS is

pre

dom

inan

tly u

prig

htin

lead

s I a

nd II

). P

wav

e:no

rmal

. QRS

:vol

tage

in c

hest

lead

s is

pro

min

ent

but

does

not

mee

t cr

iteria

for

ven

tric

ular

hyp

ertr

ophy

; pat

holo

gic

Q w

aves

are

pre

sent

in II

, III,

and

aV

F, in

-di

catin

g in

ferio

r w

all M

I, an

d th

e ta

ll R

wav

es in

V1

and

V2

are

cons

iste

nt w

ith p

oste

rior

MI i

nvol

vem

ent

as w

ell.

Mar

ked

ST s

egm

ent

elev

atio

nis

app

aren

t in

II, I

II, a

nd a

VF,

indi

catin

gth

at t

his

is a

n ac

ute

MI.

113

10090-04_CH04.qxd 8/31/06 5:25 PM Page 113

Fig

ure

4.3

3.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:si

nus

brad

ycar

dia.

Rat

e:55

bpm

. In

terv

als:

PR,

0.24

(fir

st-d

egre

e A

V b

lock

; se

e C

hapt

er 1

2);

QRS

, 0.

09;

QT,

0.4

4 se

c. A

xis:

0°.

P w

ave:

norm

al. Q

RS:l

eft

vent

ricul

ar h

yper

trop

hy (L

VH

): S

in V

1(1

4 m

m) +

+R

in V

5(2

2 m

m) >

35

mm

. Pat

holo

gic

Q w

aves

in le

ads

III a

nd a

VF

rais

e th

e po

ssib

ility

of

an o

ld in

ferio

rM

I. Th

e ST

seg

men

t de

pres

sion

and

T w

ave

inve

rsio

nar

e se

cond

ary

to t

he a

bnor

mal

rep

olar

izat

ion

resu

lting

fro

m L

VH

.

114

10090-04_CH04.qxd 8/31/06 5:25 PM Page 114

Fig

ure

4.3

4.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:nor

mal

sin

us. R

ate:

68 b

pm. I

nter

vals

:PR,

0.2

4 (fi

rst-

degr

ee A

V b

lock

; see

Cha

pter

12)

; QRS

, 0.1

0; Q

T, 0

.36

sec.

Axi

s:−4

5°(le

ft a

xis

devi

atio

n).

P w

ave:

left

atr

ial e

nlar

gem

ent

(ter

min

al d

eflec

tion

of P

wav

e in

V1

is 1

mm

wid

e an

d 1

mm

dee

p—ju

st b

arel

y).

QRS

: pa

tter

n of

left

ant

erio

r fa

scic

ular

blo

ck (

LAFB

; se

e Fi

g. 4

.20)

. The

abn

orm

ally

sm

all R

wav

es in

lead

s V

2–V

4ar

e as

soci

ated

with

LA

FB re

sulti

ng fr

om th

e re

duct

ion

of in

itial

ant

erio

r for

ces.

The

ST

segm

ent a

nd T

wav

esar

e un

rem

arka

ble.

115

10090-04_CH04.qxd 8/31/06 5:25 PM Page 115

Fig

ure

4.3

5.12

-lea

d E

CG

(ab

no

rmal

).Rh

ythm

:nor

mal

sin

us. R

ate:

95 b

pm. I

nter

vals

:PR,

0.2

0; Q

RS, 0

.10;

QT,

0.3

4 se

c. A

xis:

++16

0°(r

ight

axi

s de

viat

ion

[RA

D]).

P w

ave:

right

atria

l enl

arge

men

t (P

wav

e in

lead

II is

> 2

.5 m

m t

all).

QRS

: rig

ht v

entr

icul

ar h

yper

trop

hy (R

VH

): R

> S

in V

1w

ith R

AD

. The

T w

aves

are

inve

rted

in t

he a

nter

ior

lead

s, a

t le

ast

in p

art

refle

ctin

g ab

norm

al r

epol

ariz

atio

n ow

ing

to R

VH

.

116

10090-04_CH04.qxd 8/31/06 5:25 PM Page 116

Figure 4.36. 12-lead ECG (abnormal). Rhythm: normal sinus. Rate: 62 bpm. Intervals: PR, 0.14; QRS, 0.10; QT, 0.52(corrected QT, 0.53, which is prolonged). Axis: +95° (right axis deviation [RAD]). QRS: pattern of left posterior fascic-ular block, with small R wave in leads I and aVL, small Q wave in leads II, III, and aVF, and right axis deviation (RAD;see Fig. 4.20). The prolonged QT interval in this patient is the result of antiarrhythmic medication.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6


Additional Reading

Dubin D. Rapid Interpretation of EKGs. 6th Ed.Tampa: Cover Publishing, 2000.

Goldberger AL, Goldberger E. Clinical Electrocardio-graphy: A Simplified Approach. 6th Ed. St. Louis:Mosby Year Book, 1999.

O’Keefe JH Jr, Hammill SC, Freed MS, et al. The Com-plete Guide to ECGs. 2nd Ed. Royal Oak, Michi-gan: Physicians’ Press, 2002.

Surawicz B, Knilans TK. Chou’s Electrocardiographyin Clinical Practice. 5th Ed. Philadelphia: Saun-ders, 2001.

Wagner GS. Marriott’s Practical Electrocardiography.10th Ed. Baltimore: Lippincott Williams & Wilkins,2001.

10090-04_CH04.qxd 8/31/06 5:25 PM Page 117

Chapter 4—Author Queries (tables)1. AU: Correct?2. AU: Correct?3. AU: Edit correct?

10090-04_CH04.qxd 8/31/06 5:25 PM Page a

118

VASCULAR BIOLOGY OF ATHEROSCLEROSISNormal Arterial WallAtherosclerotic Arterial WallComplications of Atherosclerosis

ATHEROSCLEROSIS RISK FACTORSTraditional Risk FactorsBiomarkersManagement of Risk Factors

C H A P T E R

5AtherosclerosisJames L. YoungPeter Libby

Atherosclerosis is the leading cause of mor-tality and morbidity in the developed world.Through its major manifestations of cardio-vascular disease and stroke, it is predicted tobe the leading global killer by the year 2020.Historically known as “hardening of the ar-teries,” atherosclerosis derives its name fromthe Greek roots athere-, meaning “gruel,” and-skleros, meaning “hardness.”

Although the pathologic hallmarks of ath-erosclerosis have been studied for more thana century, its pathogenesis has remained a topic of great debate. In the early 1900s, Anitchkow and Chalatow demonstrated thatfeeding rabbits a high-cholesterol diet in-duces humanlike atherosclerotic lesions inblood vessels, supporting the view of athero-sclerosis as a mere disorder of lipid accumu-lation. This belief persisted through most ofthe twentieth century, but more recent evi-dence has generated a different view of ath-erosclerosis as a chronic inflammatory condi-tion, involving lipids, thrombosis, elements

of the vascular wall, and immune cells. Thisprocess can smolder throughout adulthood,punctuated by acute cardiovascular events.

This chapter consists of two sections. Thefirst section describes the normal arterial wall,the pathogenesis of atherosclerotic plaqueformation, and pathologic complicationsthat lead to clinical symptoms. The secondsection relates findings from population stud-ies to attributes that lead to this condition,thereby offering opportunities for preven-tion and treatment.

VASCULAR BIOLOGY OFATHEROSCLEROSIS

Normal Arterial Wall

The arterial wall consists of three layers (Fig. 5.1): the intima, closest to the arteriallumen and therefore most “intimate” withthe blood; the media, which is the middlelayer; and the outer adventitia. The intimaconsists of a single layer of endothelial cells

Fig. 1

10090-05_CH05.qxd 8/31/06 5:35 PM Page 118

Atherosclerosis 119

that acts as a metabolically active barrier be-tween circulating blood and the vessel wall.The media is the thickest layer of the normalartery. Boundaries of elastin, known as theinternal and external elastic laminae, sepa-rate this middle layer from the intima andadventitia, respectively. The media is com-posed of smooth muscle cells and extra-cellular matrix, and subserves the contrac-tile and elastic functions of the vessel. Theelastic component, more prominent inlarge arteries (e.g., the aorta and its primarybranches), stretches during the high pres-sure of systole and then recoils during dias-tole, propelling blood forward. The muscu-lar component, more prominent in smallerarteries such as arterioles, constricts or re-laxes to alter vessel resistance and thereforeluminal blood flow (flow = pressure ÷ resis-tance; see Chapter 6). The adventitia con-tains the nerves, lymphatics, and blood ves-sels (vasa vasorum) that nourish the cells ofthe arterial wall.

Far from an inert conduit, the living arte-rial wall is a scene of dynamic interchangebetween its cellular components, most im-portantly endothelial cells, vascular smoothmuscle cells, and their surrounding extra-

cellular matrix. Understanding the dysfunc-tion that leads to atherosclerosis requiresknowledge of the normal function of thesecomponents.

Endothelial Cells

In a healthy artery, the endothelium servesstructural, metabolic, and signaling func-tions that maintain homeostasis of the ves-sel wall. The tightly adjoined endothelialcells form a barrier that contains blood with-in the lumen of the vessel and limits the pas-sage of large molecules from the circulationinto the subendothelial space, which is alsoknown as the subintima.

As blood traverses the vascular tree, it en-counters antithrombotic molecules producedby the endothelium that prevent it fromclotting. Some of these molecules reside onthe endothelial surface (e.g., heparan sul-fate, thrombomodulin, and plasminogenactivators; see Chapter 7), while other an-tithrombotic products of the endotheliumenter the circulation (e.g., prostacyclin andnitric oxide [NO]; see Chapter 6). Althougha net anticoagulant state normally prevails,the endothelium can also produce prothrom-

Intima

Media

Adventitia

Endothelial cells

Internal elastic lamina

Smooth muscle cells

External elastic lamina

Lumen

Figure 5.1. Schematic diagram of the arterial wall. The intima, the innermost layer, overlies the mus-cular media demarcated by the internal elastic lamina. The external elastic lamina separates the mediafrom the outer layer, the adventitia.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 119

botic molecules when subjected to variousstressors.

Endothelial cells also secrete substancesthat modulate contraction of smooth mus-cle cells in the underlying medial layer. Thesesubstances include vasodilators (e.g., NO,prostacyclin) and vasoconstrictors (e.g., en-dothelin) that alter the resistance of the ves-sel and therefore luminal blood flow. In anormal artery, the predominance of va-sodilator substances results in net smoothmuscle relaxation. Several of the aforemen-tioned endothelial products (e.g., heparansulfate, NO) also function within the vesselwall to inhibit proliferation of smooth mus-cle cells into the intima, thus enforcing theirnormal residence within the media.

Endothelial cells also play an importantrole in regulating the immune response. In the absence of pathologic stimulation,healthy arterial endothelial cells resist leu-kocyte adhesion and are therefore anti-inflammatory. However, endothelial cells inpostcapillary venules respond to local injuryor infection by secreting chemokines, chem-icals that attract white blood cells to thearea. Such stimulation also causes endothe-lial cells to produce cell surface adhesion

molecules, which anchor mononuclear cellsto the endothelium and facilitate their mi-gration to the site of injury. As describedlater, under the adverse influences presentduring atherogenesis, endothelial cells sim-ilarly recruit leukocytes to the vessel wall.Thus, the normal endothelium provides aprotective, nonthrombogenic surface withhomeostatic vasodilator and anti-inflamma-tory properties (Fig. 5.2).

Vascular Smooth Muscle Cells

Smooth muscle cells within the vessel wallhave both contractile and synthetic capabil-ities. Various vasoactive substances modu-late the contractile function, resulting in vaso-constriction or vasodilatation. Such agonistsinclude circulating molecules (e.g., angio-tensin II), those released from local nerveterminals (e.g., acetylcholine), and othersoriginating from the overlying endothelium(e.g., endothelin, NO).

Normal synthetic functions of smoothmuscle cells include production of the col-lagen, elastin, and proteoglycans that formthe vascular extracellular matrix (see Fig 5.2).Smooth muscle cells also have the capability

120 Chapter Five

• Impermeable to large molecules• Anti-inflammatory• Resist leukocyte adhesion• Promote vasodilation• Resist thrombosis

• Normal contractile function• Maintain extracellular matrix• Contained in medial layer

PermeabilityInflammatory cytokinesLeukocyte adhesion moleculesVasodilatory moleculesAntithrombotic molecules

Inflammatory cytokines Altered matrix synthesisMigration and proliferationinto subintima

ENDOTHELIALCELLS

NORMAL ACTIVATED

SMOOTH MUSCLECELLS

Figure 5.2. Endothelial and smooth muscle cell activation by inflammation. Normal endothelialand smooth muscle cells maintain the integrity and elasticity of the normal arterial wall while limitingimmune cell infiltration. Inflammatory activation of these vascular cells corrupts their normal functionsand favors proatherogenic mechanisms that drive plaque development.

Fig. 2

10090-05_CH05.qxd 8/31/06 5:35 PM Page 120

Atherosclerosis 121

to synthesize vasoactive and inflammatorymediators, including interleukin 6 (IL-6) andtumor necrosis factor α (TNF-α), which pro-mote leukocyte proliferation and induceendothelial expression of leukocyte adhe-sion molecules. These synthetic functionsbecome more prominent at sites of athero-sclerotic plaque and may contribute to itspathogenesis.

Extracellular Matrix

In healthy arteries, fibrillar collagen andelastin make up most of the extracellularmatrix in the medial layer. Interstitial colla-gen fibrils, constructed from intertwininghelical proteins, possess great biomechani-cal strength, while elastin provides flexibil-ity. Together these components maintainthe structural integrity of the vessel despitethe high pressure within the lumen. Recentevidence suggests that the extracellular ma-trix also regulates the growth of its residentcells. Native fibrillar collagen, in particular,can inhibit smooth muscle cell proliferationin vitro. Furthermore, the matrix influencescellular responses to stimuli: matrix-boundcells respond in a specific manner to growthfactors and are less likely to undergo apop-tosis (programmed cell death).

Atherosclerotic Arterial Wall

The arterial wall is a dynamic and regulatedsystem. However, noxious elements can cor-rupt normal homeostasis and pave the wayfor atherogenesis. For example, as describedlater, vascular endothelial and smooth mus-cle cells react readily to inflammatory medi-ators such as IL-1 and TNF-α. These inflam-matory agents can also activate vascular cellsto produce IL-1 and TNF-α, contrary to pastdogma that only cells of the immune systemsynthesize such cytokines.

Realizing that immune cells were not theonly source of proinflammatory agents, in-vestigations into the role of “activated” en-dothelial and smooth muscle cells in athe-rogenesis burgeoned. This fundamentalresearch has identified several key compo-nents that contribute to the atherosclerotic

inflammatory process, including (1) endo-thelial dysfunction, (2) accumulation of li-pids within the intima, (3) recruitment ofleukocytes and smooth muscle cells to thevessel wall, (4) formation of foam cells, and(5) deposition of extracellular matrix, assummarized in Figure 5.3 and described inthe following sections. Rather than follow asequential path, the cells of atheroscleroticlesions continuously interact and competewith each other, shaping the plaque overdecades into one of many possible profiles.This section categorizes these mechanismsinto three pathologic stages: (1) the fattystreak, (2) plaque progression, and (3) plaquedisruption (Fig. 5.4).

Fatty Streak

Fatty streaks represent the earliest visible lesions of atherosclerosis. On gross inspec-tion, they appear as areas of yellow discol-oration on the artery’s inner surface; how-ever, they neither protrude substantiallyinto the arterial lumen nor impede bloodflow. Surprisingly, fatty streaks exist in theaorta and coronary arteries of most peopleby age 20. They do not cause symptoms, andin some locations in the vasculature, theymay regress over time. Although the preciseinitiation of fatty streak development is notknown, observations in animal models sug-gest that various stressors cause early en-dothelial dysfunction, as described in thenext section. Such dysfunction allows entryand modification of lipids within the sub-intima, where they serve as proinflamma-tory mediators that initiate recruitment ofleukocytes and foam cell formation—thepathological hallmarks of the fatty streak(Fig. 5.5).

Endothelial Dysfunction

Injury to the arterial endothelium represents a pri-mary event in atherogenesis. Such injury can re-sult from exposure to diverse agents, includ-ing physical forces and chemical irritants.

The role of physical stress is suggested byobservations that atherosclerosis tends toform at arterial branch points (i.e., arterial

AQ1

Fig. 3

Fig. 4

Fig. 5

10090-05_CH05.qxd 8/31/06 5:35 PM Page 121

bifurcations). In straight sections of arteries,the normal laminar (i.e., smooth) shearforces favor the endothelial expression ofNO, which is beneficial as an endogenousvasodilator, an inhibitor of platelet aggrega-tion, and an anti-inflammatory substance.

Moreover, laminar flow accentuates expres-sion of the antioxidant enzyme superoxidedismutase, which protects against reactiveoxygen species produced by chemical irri-tants or transient ischemia, as described inChapter 6. Conversely, disturbed flow is typ-

122 Chapter Five

Smooth musclemitogens

Cell apoptosis

Foam cell

Smooth musclemigration

Smooth muscleproliferation

Oxidized LDL

Vascularendothelium

Internal elasticlamina

LDL

Monocytes Macrophage

IL-1 MCP-1Scavengerreceptor

Celladhesionmolecule

1

2 3

4

5

8

7

6

Figure 5.3. Schematic diagram of the evolution of atherosclerotic plaque. (1) Accumulation of lipoprotein parti-cles in the intima. The darker color depicts modification of the lipoproteins (e.g., by oxidation or glycation). (2) Oxida-tive stress, including constituents of modified LDL (mLDL), induces local cytokine elaboration. (3) These cytokines pro-mote increased expression of adhesion molecules that bind leukocytes and of chemoattractant molecules (e.g., monocytechemoattractant protein 1 [MCP-1]) that direct leukocyte migration into the intima. (4) After entering the artery wall in response to chemoattractants, blood monocytes encounter stimuli such as macrophage colony–stimulating factor (M-CSF) that augment their expression of scavenger receptors. (5) Scavenger receptors mediate the uptake of modifiedlipoprotein particles and promote the development of foam cells. Macrophage foam cells are a source of additional cy-tokines and effector molecules such as superoxide anion (O −

2 ) and matrix metalloproteinases. (6) Smooth muscle cellsmigrate into the intima from the media. Note the increasing intimal thickness. (7) Intimal smooth muscle cells divide andelaborate extracellular matrix, promoting matrix accumulation in the growing atherosclerotic plaque. In this manner, thefatty streak evolves into a fibrofatty lesion. (8) In later stages, calcification can occur (not depicted) and fibrosis contin-ues, sometimes accompanied by smooth muscle cell death (including programmed cell death, or apoptosis), yielding arelatively acellular fibrous capsule surrounding a lipid-rich core that may also contain dying or dead cells. IL-1, interleukin 1;LDL, low-density lipoprotein. (Modified from Zipes D, Libby P, Bonow RO, et al., eds. Heart Disease: A Textbook of Cardio-vascular Medicine. 7th Ed. Philadelphia: Elsevier Saunders, 2005:925.)

A B C

FATTY STREAK PLAQUE PROGRESSION PLAQUE DISRUPTION

••

••

Endothelial dysfunctionLipoprotein entry andmodificationLeukocyte recruitmentFoam cell formation

••

Smooth muscle cell migrationAltered matrix synthesis anddegradation

• Disrupted plaque integrity• Thrombus formation

Figure 5.4. Stages of plaque development. A. The fatty streak develops as a result of endothelial dysfunction,lipoprotein entry and modification, leukocyte recruitment, and foam cell formation. B. Plaque progression is char-acterized by migration of smooth muscle cells into the intima, where they divide and elaborate extracellular ma-trix. A lipid core is contained by the fibrous cap. C. Hemodynamic stresses and degradation of extracellular matrixincrease the susceptibility of the fibrous cap to rupture, allowing superimposed thrombus formation. (Modifiedfrom Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation 2002;105:1136.)

10090-05_CH05.qxd 8/31/06 5:35 PM Page 122

Atherosclerosis 123

ical at arterial branch points, which impairsthese locally atheroprotective endothelialfunctions. Accordingly, arteries with fewbranches (e.g., the internal mammary artery)show relative resistance to atherosclerosis,whereas bifurcated vessels (e.g., the commoncarotid and left coronary arteries) are com-mon sites for atherosclerosis deposition.

Endothelial dysfunction may also resultfrom exposure to a “toxic” chemical envi-ronment. For example, cigarette smoking, ab-normal circulating lipid levels, or diabetes—all known risk factors for atherosclerosis—canpromote endothelial dysfunction. Each ofthese states increases endothelial productionof reactive oxygen species—notably, super-oxide anion—which interact with other intracellular molecules to influence the meta-bolic and synthetic functions of the endothe-lium. Consequently, the cells are skewed to-ward proinflammatory processes.

When physical and chemical stressorscorrupt normal endothelial homeostasis,

the result is an activated state manifestedby (1) impairment of the endothelium’srole as a permeability barrier, (2) release of inflammatory cytokines, (3) increasedproduction of cell surface adhesion mole-cules that recruit leukocytes, (4) altered re-lease of vasoactive substances (e.g., prosta-cyclin and NO), and (5) interference withnormal antithrombotic properties. Theseundesired effects of endothelial dysfunc-tion lay the groundwork for subsequentevents in the development of atherosclero-sis (see Fig 5.2).

Lipoprotein Entry and Modification

The activated dysfunctional endotheliumno longer serves as an effective barrier tothe passage of circulating lipoproteins intothe arterial wall (see Box 5.1 for a review of the major lipoprotein pathways). In-creased endothelial permeability allowsthe entry of low-density lipoprotein (LDL)into the intima, a process facilitated by anelevated circulating LDL concentration.Once within the intima, LDL accumulatesin the subendothelial space by binding tocomponents of the extracellular matrixknown as proteoglycans. This “trapping”increases the residence time of LDL withinthe vessel wall, where the lipoprotein mayundergo chemical modifications that ap-pear critical to the development of athero-sclerotic lesions. Hypertension, a majorrisk factor for atherosclerosis, may pro-mote retention of lipoproteins in the in-tima by accentuating the production ofLDL-binding proteoglycans by smooth mus-cle cells.

Oxidation is one type of modification thatbefalls LDL trapped in the subendothelialspace. It can result from the local action ofreactive oxygen species and pro-oxidant en-zymes derived from activated endothelial orsmooth muscle cells, or from macrophagesthat penetrate the vessel wall. In diabeticpatients with sustained hyperglycemia, gly-cation of LDL can occur, a transformation thatmay ultimately render LDL antigenic andproinflammatory. The biochemical modifi-cations of LDL act early and contribute

Chemical irritants Hemodynamic stress

Lipoprotein entry &modification

Inflammatorycytokines

Chemokines Leukocyte adhesion molecules

Leukocyte recruitment

Foam Cell Formation

Endothelial Dysfunction

Unregulated updateof modified LDL

Figure 5.5. Endothelial dysfunction is the primaryevent in plaque initiation. Physical and chemical stres-sors alter the normal endothelium, allowing lipid entryinto the subintima and promoting inflammatory cytokinerelease. This cytokine- and lipid-rich environment pro-motes recruitment of leukocyte to the subintima, wherethey may transform into foam cells—a prominent inflam-matory participant.

Box 1

10090-05_CH05.qxd 8/31/06 5:35 PM Page 123

124 Chapter Five

Box 5.1 The Lipoprotein Transport System

Lipoproteins ferry water-insoluble fats through the bloodstream. These particles consist ofa lipid core surrounded by more hydrophilic phospholipid, free cholesterol, and apolipopro-teins (also called apoproteins, or apos). The apoproteins present on various classes oflipoprotein molecules serve as the “conductors” of the system, directing the lipoproteinsto specific tissue receptors and mediating enzymatic reactions. Five major classes oflipoproteins exist, distinguished by their densities, lipid constituents, and associatedapoproteins. In order of increasing density, they are chylomicrons, very low-densitylipoprotein (VLDL), intermediate-density lipoprotein (IDL), low-density lipopro-teins (LDL), and high-density lipoprotein (HDL). The major steps in the lipoproteinpathways are labeled in the accompanying figure and described as follows. The keyapoproteins (apos) at each stage are indicated in the figure in parentheses.

Exogenous (Intestinal) Pathway

1. Dietary fats are absorbed by the small intestine and repackaged as chylo-microns, accompanied by apo B-48. Chylomicrons are large particles, partic-ularly rich in triglycerides, that enter the circulation via the lymphatic system.

2. Apo E and subtypes of apo C are transferred to chylomicrons from HDL par-ticles in the bloodstream.

3. Apo C (subtype CII) enhances interactions of chylomicrons with lipoproteinlipase (LPL) on the endothelial surface of adipose and muscle tissue. This re-action hydrolyzes the triglycerides within chylomicrons into free fatty acids(FFA), which are stored by adipose tissue or used for energy in cardiac andskeletal muscle.

Liver

1

2

3

5

4

10

6

7

8 9Intestine

Chylomicrons(Apo B-48, A, C, E)

Exogenous Pathway Endogenous Pathway

Lipoprotein lipase

HDL HDL

Adipose tissueMuscle Adipose tissueMuscle

Lipoprotein lipase

Dietary fat

Chylomicronremnants

(Apo B-48, E)

Bile acids and cholesterol

Apo

C, E

FFA FFA

VLDL(Apo B-100, C, E)

(CE

TP

)

Apo

C, E

LDL(Apo B-100)

Non-hepaticcells

LPLHL

IDL(Apo B-100, E)

10090-05_CH05.qxd 8/31/06 5:35 PM Page 124

Atherosclerosis 125

4. Chylomicron remnants are removed from the circulation by the liver, medi-ated by apo E.

5. One fate of cholesterol in the liver is incorporation into bile acids, which areexported to the intestine, completing the exogenous pathway cycle.

Endogenous (Hepatic) Pathway

Because dietary fat availability is not constant, the endogenous pathway provides a reli-able supply of triglycerides for tissue energy needs:

6. The liver packages cholesterol and triglycerides into VLDL particles, accom-panied by apo B-100 and phospholipid. The triglyceride content of VLDL ismuch higher than that of cholesterol, but this is the main means by which theliver releases cholesterol into the circulation.

7. VLDL is catabolized by LPL (similar to chylomicrons, as described in step 3),releasing fatty acids to muscle and adipose tissue. During this process, VLDLalso interacts with HDL, exchanging some of its triglyceride for apo C sub-types, apo E, and cholesteryl ester from HDL. The latter exchange (importantin reverse cholesterol transport, as described in the next section) is mediatedby cholesteryl ester transfer protein (CETP).

8. Approximately 50% of the VLDL remnants (IDL) are then cleared in the liverby hepatic receptors that recognize apo E.

9. The remaining IDL is catabolized further by LPL and hepatic lipase (HL), whichremove additional triglyceride, apo E, and apo Cs, forming LDL particles.

10. Plasma clearance of LDL occurs primarily via LDL receptor–mediated endocy-tosis in the liver and peripheral cells, directed by LDL’s apo B-100.

Cholesterol Homeostasis and Reverse Cholesterol Transport

Intracellular cholesterol content is tightly maintained by de novo synthesis, cellular uptake,storage, and efflux from the cell. The enzyme HMG-CoA reductase is the rate-limitingelement of cholesterol biosynthesis, and cellular uptake of cholesterol is controlled byreceptor-mediated endocytosis of circulating LDL (see step 10). When intracellular cho-lesterol levels are low, the transcription factor sterol regulatory element binding protein

Peripheralcells

LCAT

Freeapo A1 Nascent

HDL

Internalizedcholesterolester

Excesscholesterol

MatureHDL CETP

Transfer ofcholesterol

ester toVLDL, IDL, LDL

(in exchange for TG)

apo A1

ABCA1 ABCG1

10090-05_CH05.qxd 8/31/06 5:35 PM Page 125

126 Chapter Five

(SREBP) is released from the endoplasmic reticulum. The active fragment of SREBP enters thenucleus to increase transcription of HMG-CoA reductase and the LDL receptor, which,through their subsequent actions, tend to normalize the intracellular cholesterol content. Theprocess is illustrated in the accompanying figure.

Under conditions of intracellular cholesterol excess, peripheral cells increase the tran-scription of the ATP binding cassette A1 and G1 genes (ABCA1 and ABCG1, respectively).The ABCA1 gene codes for a transmembrane protein transporter that initiates efflux ofcholesterol from the cell to lipid-poor circulating apo AI (which is synthesized by the liverand intestine), thus forming nascent (immature) HDL particles. ABCG1 facilitates furtherefflux of cholesterol to more-mature HDL particles. As free cholesterol is acquired by cir-culating HDL, it is esterified by lecithin cholesterol acyltransferase, an enzyme activated byapo AI. The hydrophobic cholesterol esters move into the particle’s core. Most cholesterolesters in HDL can then be exchanged in the circulation (via the enzyme CETP) with any ofthe apo B–containing lipoproteins (i.e., VLDL, IDL, LDL), which deliver the cholesterol backto the liver. HDL can also transport cholesterol to the liver and hormone-producing tissuesvia a scavenger receptor termed SR-B1.

AQ3

to the inflammatory mechanisms initiatedby endothelial dysfunction, and they maypromote inflammation throughout the lifes-pan of the plaque. In the fatty streak, andlikely throughout plaque development, mod-ified LDL (mLDL) promotes leukocyte re-cruitment and foam cell formation.

Leukocyte Recruitment

Recruitment of leukocytes (primarily mono-cytes and T lymphocytes) to the vessel wallis a key step in atherogenesis. The processdepends on (1) expression of leukocyte ad-hesion molecules (LAM) on the normallynonadherent endothelial luminal surface,and (2) chemoattractant signals (e.g., mono-cyte chemotactic protein 1 [MCP-1], IL-8,IP-10) that direct diapedesis (passage of cellsthrough the intact endothelium) into thesubintimal space. Two major subsets of LAMpersist in the inflamed atherosclerotic plaque:(1) the immunoglobulin gene superfamily(particularly vascular cell adhesion molecule1 [VCAM-1] and intercellular adhesion mol-ecule 1 [ICAM-1]), and (2) the selectins (par-ticularly E- and P-selectin). Despite the cen-tral role of T lymphocytes in the immunesystem, plaque LAMs and chemoattractantsignals direct mainly monocytes to the form-ing lesion. Although outnumbered, T lym-phocytes localize within plaques at all stages

where they likely furnish an importantsource of cytokines.

Modified LDL and proinflammatory cy-tokines (e.g., IL-1, TNF-α) can induce LAMand chemoattractant cytokine (chemokine)expression independently; however, mLDLalso potently stimulates endothelial andsmooth muscle cells to produce proinflam-matory cytokines, thereby reinforcing thedirect action. This dual ability of mLDL to promote leukocyte recruitment and in-flammation directly and indirectly persiststhroughout atherogenesis.

Foam Cell Formation

After monocytes adhere to and penetratethe intima, they differentiate into phago-cytic macrophages and imbibe lipoproteinsto form foam cells. Importantly, foam cellsdo not arise from uptake of LDL by the clas-sic cell surface LDL-receptor mechanism de-scribed in Box 5.1, because the high choles-terol content within these cells actuallysuppresses expression of the receptor. Fur-thermore, the classic LDL receptor does notrecognize chemically modified LDL. Rather,macrophages rely on a family of “scavenger”receptors that preferentially bind and in-ternalize mLDL. Unlike uptake via the clas-sic LDL receptor, mLDL ingestion by scav-enger receptors evades negative feedback

AQ2

10090-05_CH05.qxd 8/31/06 5:35 PM Page 126

Atherosclerosis 127

inhibition and permits engorgement of themacrophages by cholesterol-rich lipid, re-sulting in the typical appearance of foamcells. Although such uptake initially may bebeneficial (by sequestering proinflamma-tory mLDL particles), the impaired efflux ofthese cells leads to local accumulation in theplaque, mitigating their protective role andinstead serving as a source of proinflamma-tory cytokines that fuel atherosclerotic plaqueprogression.

Plaque Progression

Whereas endothelial cells play a central rolein formation of the fatty streak, smoothmuscle cell migration into the intima dom-inates early plaque progression. Spanningdecades of development, the typical ather-osclerotic plaque acquires a distinct throm-bogenic lipid core that under-lies a protec-tive fibrous cap. Early plaque growth showsa compensatory outward remodeling of theplaque wall that preserves the diameter ofthe arterial lumen and does not limit bloodflow. This stage can even evade detection byangiography. Later plaque growth howevermay significantly restrict the vessel lumenand impede perfusion. Such flow-limitingplaques can result in tissue ischemia, caus-ing symptoms such as angina pectoris (seeChapter 6) or intermittent claudication ofthe extremities (see Chapter 15).

As described in Chapter 7, most acutecoronary syndromes (acute myocardial in-farction or unstable angina pectoris) resultwhen the fibrous cap of an atheroscleroticplaque ruptures and prothrombotic mole-cules within the lipid core are exposed, pre-cipitating an acute thrombus that suddenlyoccludes the arterial lumen. However, as de-scribed later in the chapter, the extracellu-lar matrix plays a pivotal role in fortifyingthe fibrous cap, isolating the thrombogenicplaque interior from coagulation substratesin the circulation.

Smooth Muscle Cell Migration

The transition from fatty streak to fibrousatheromatous plaque involves the migra-

tion of smooth muscle cells from the arter-ial media into the intima, proliferation ofthe smooth muscle cells within the intima,and secretion of extracellular matrix macro-molecules by the smooth muscle cells. Foamcells, activated platelets, and endothelialcells all elaborate substances that signalsmooth muscle cell migration and prolifera-tion (Fig. 5.6).

Foam cells produce several factors thatcontribute to smooth muscle cell recruit-ment. For example, they release platelet-derived growth factor (PDGF), which likelystimulates the migration of smooth musclecells across the internal elastic lamina and into the subintimal space, where theysubsequently replicate. Foam cells also release cytokines and growth factors (e.g.,TNF-α, IL-1, fibroblast growth factor, andtransforming growth factor β [TGF-β]) that further stimulate smooth muscle cellproliferation and synthesis of extracellularmatrix proteins. Furthermore, these stimu-latory cytokines induce smooth musclecell and leukocyte activation, promoting

Fig. 6

Tissuefactor

Thrombosisand plateletactivation

Heparinase

NOPGI2

Endothelialheparan sulfate

PDGF↑

Foam cells in fatty streak

Endothelialdysfunction

Fibrousplaque

Smooth muscle cells migrate to intima,proliferate, and produce extracellular matrix

Figure 5.6. Progression from the fatty streak involvesthe migration and proliferation of smooth musclecells. Substances released from foam cells, dysfunctionalendothelial cells, and platelets contribute to this process.IL-1, interleukin-1; NO, nitric oxide; PDGF, platelet-derivedgrowth factor; PGI2, prostacyclin; TGF-β, transforminggrowth factor β; TNF-α, tumor necrosis factor α.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 127

further cytokine release, thus reinforc-ing and maintaining inflammation in thelesion.

It has been traditionally assumed thatplaque growth is gradual and continuous.However, current evidence suggests thatthis progression may be punctuated bysubclinical events with bursts of smoothmuscle replication. For example, morpho-logical evidence of resolved intraplaque hemorrhages indicates that small breachesin plaque integrity can occur withoutclinical symptoms or signs. At the cellularlevel, such breaches expose tissue factorfrom foam cells, which activates coagula-tion and microthrombus formation. Acti-vated platelets within such microthrombirelease additional potent factors that canspur a local wave of smooth muscle cell mi-gration and proliferation. These factors in-clude PDGF and heparinase. The latter de-grades heparan sulfate, a polysaccharide inthe extracellular matrix that normally in-hibits smooth muscle cell migration andproliferation.

Extracellular Matrix Metabolism

As the predominant collagen-synthesizingcell type, smooth muscle cells should, viatheir proliferation, favor fortification of thefibrous cap. However, net matrix depositiondepends on the balance of synthesis bysmooth muscle cells and degradation, medi-ated in part by a class of proteolytic enzymesknown as matrix metalloproteinases (MMP).While PDGF and TGF-β stimulate smoothmuscle cell production of interstitial colla-gens, the T-lymphocyte–derived cytokineinterferon γ (IFN-γ) inhibits smooth musclecell collagen synthesis. Furthermore, inflam-matory cytokines stimulate local foam cellsto secrete collagen- and elastin-degradingMMP, thereby weakening the fibrous capand predisposing it to rupture (Fig. 5.7).

Plaque Disruption

Plaque Integrity

Over decades, the tug-of-war between ma-trix synthesis and degradation continues,

128 Chapter Five

Foamcell

T lymphocyte

Collagen andelastin

+ +

– +PDGFTGF-β CD40LIFN-γ

Smooth muscle cell

SynthesisDegradation

Lumen

IL-1TNF-αMCP-1

Lipid core

Fibrous Cap

MMP

Figure 5.7. Matrix metabolism underlies fibrous cap integrity. The net deposition of extracellular ma-trix is the result of competing synthesis and degradation reactions. Smooth muscle cells synthesize the bulkof the fibrous cap constituents, such as collagen and elastin. However, foam cells also elaborate destructiveproteolytic enzymes, such as the collagen-degrading matrix metalloproteinases and the elastolytic cathepsins.T-lymphocyte–derived factors favor destruction of the fibrous cap. However, all plaque residents contributeto the cytokine milieu of the plaque, providing multiple activating and inhibitory stimuli as shown. IFN-γ, interferon γ; IL-1, interleukin 1; MCP-1, monocyte chemoattractant protein 1; PDGF, platelet-derived growthfactor; TGF-β, transforming growth factor β; TNF-α, tumor necrosis factor α. (Modified from Libby P. The mo-lecular bases of acute coronary syndromes. Circulation 1995;91:2844–2850; Young JL, Libby P, SchönbeckU. Cytokines in the pathogenesis of atherosclerosis. Thromb Haemost 2002;88:554–567.) 1 LINE LO

Fig. 7

10090-05_CH05.qxd 8/31/06 5:35 PM Page 128

Atherosclerosis 129

but not without consequences. Death ofsmooth muscle and foam cells, either owingto excess inflammatory stimulation or bycontact activation of apoptosis pathways,liberates cellular contents, contributing im-bibed lipids and cellular debris to the grow-ing lipid core. The size of the lipid core hasbiomechanical implications for the stabilityof the plaque. With increasing size and pro-trusion into the arterial lumen, hemody-namic stress focuses on the plaque borderabutting normal tissue, the so-called shoul-der region. In addition to shouldering in-creased stress, local accumulation of foamcells and T lymphocytes at this site acceler-ates degradation of extracellular matrix, mak-ing this region the most common site ofplaque rupture.

The net deposition and distribution ofthe fibrous cap is an important determi-nant of overall plaque integrity. Whereas le-sions with thick fibrous caps may cause pro-nounced arterial narrowing, they have lesspropensity to rupture. Conversely, plaquesthat have thinner caps (and often appearless obstructive by angiography) tend to be fragile and more likely to rupture and in-cite thrombosis. Current terminology de-scribes the extreme spectrums of integrity asstable plaques (marked by a thick fibrous capand small lipid core) or vulnerable plaques(marked by a thin fibrous cap, rich lipidcore, extensive macrophage infiltrate, and apaucity of smooth muscle cells; Fig. 5.8). De-spite the common use of these terms, it isimportant to recognize that this distinctionoversimplifies the heterogeneity of plaquesand may overestimate the ability to foreseea plaque’s “clinical future” based on struc-tural information.

Thrombogenic Potential

Rupture of atherosclerotic plaque does notinevitably cause major clinical events suchas myocardial infarction and stroke. As de-scribed in the previous section, small non-occlusive thrombi may reabsorb into theplaque, stimulating further smooth musclegrowth and fibrous deposition (see Fig 5.8).It is in large part the thrombogenic potentialof the plaque that determines whether dis-

ruption of the fibrous cap leads to a tran-sient, nonobstructive mural thrombus or toa completely occlusive clot, the latter beingimplicated in >90% of acute clinical events.

The probability of a major thromboticevent reflects the balance between com-peting processes of coagulation and fibri-nolysis. Inflammatory stimuli common inthe plaque microenvironment (e.g., CD40L)elicit tissue factor, the initiator of the extrin-sic coagulation pathway, from many plaquecomponents, including smooth muscle cells,endothelial cells, and macrophage-derivedfoam cells. Beyond enhancing expression of the potent procoagulant tissue factor, inflammatory stimuli further support throm-bosis by favoring expression of antifibrino-lytics (e.g., plasminogen activator inhibitor-1) over expression of anticoagulants (e.g.,thrombomodulin, heparin-like molecules,protein S) and profibrinolytic mediators (tis-sue plasminogen activator and urokinase-type plasminogen activator; Fig. 5.9). More-over, as described earlier, the activatedendothelium also promotes thrombin for-mation, coagulation, and fibrin deposition atthe vascular wall.

A person’s propensity toward coagulationmay be enhanced by genetics (e.g., the pres-ence of a procoagulant prothrombin genemutation), coexisting conditions (e.g., dia-betes), and/or lifestyle factors (e.g., smoking,visceral obesity). Consequently, the conceptof “vulnerable plaque” has expanded tothat of the “vulnerable patient” to acknow-ledge other contributors to a person’s vas-cular risk.

Complications of Atherosclerosis

Atherosclerotic plaques do not distributehomogeneously throughout the vasculature.They usually develop first in the dorsal aspectof the abdominal aorta and proximal coro-nary arteries, followed by the popliteal ar-teries, descending thoracic aorta, internalcarotid arteries, and renal arteries. There-fore, the regions perfused by these vesselsmost commonly suffer the consequences ofatherosclerosis.

Complicationsofatherosclerotic plaques—including calcification, rupture, hemorrhage,NE LONG

Fig. 8

Fig. 9

10090-05_CH05.qxd 8/31/06 5:35 PM Page 129

130 Chapter Five

Normal artery

Early atheroma

Ruptured plaquewith thrombus formation

• Small lipid pool• Thick fibrous cap• Preserved lumen

“Stable” plaque• Large lipid pool• Thin fibrous cap• Many inflammatory cells

“Vulnerable” plaque

Acutemyocardialinfarction

• Narrowed lumen• Fibrous intima

Healed rupture

Figure 5.8. Stable versus vulnerable plaques. Stable plaque is characterized by a small lipid pool and a thick fibrouscap, whereas vulnerable plaque tends to have a large lipid pool and relatively thin fibrous cap. The latter is subject torupture, resulting in thrombosis. A resulting occlusive clot can cause an acute cardiac event, such as myocardial infarc-tion. A lesser thrombus may resorb, but the wound-healing response stimulates smooth muscle cell proliferation andcollagen production, thereby thickening the fibrous cap and narrowing the vessel lumen further. (Modified from LibbyP. Inflammation in atherosclerosis. Nature 2002;420:868–874.)

10090-05_CH05.qxd 8/31/06 5:35 PM Page 130

Atherosclerosis 131

and embolization—can have dire clinicalconsequences owing to acute restriction ofblood flow or alterations in vessel wall in-tegrity. These complications, which are dis-cussed in greater detail in later chapters, in-clude the following:

1. Calcification of atherosclerotic plaque,which imparts a pipelike rigidity to thevessel wall and increases its fragility.

2. Rupture or ulceration of atheroscleroticplaque, which exposes procoagulantswithin the plaque to circulating blood,causing a thrombus to form at that site.Such thrombosis can occlude the vesseland result in infarction of the involvedorgan. Alternatively, the thrombus mate-rial can incorporate into the lesion andadd to the bulk of the plaque.

3. Hemorrhage into the plaque owing torupture of the fibrous cap or of the micro-vessels that form within the lesion. Theresulting intramural hematoma may fur-ther narrow the vessel lumen.

4. Embolization of fragments of disruptedatheroma to distal vascular sites.

5. Weakening of the vessel wall: the fibrousplaque subjects the neighboring mediallayer to increased pressure, which mayprovoke atrophy and loss of elastic tissuewith subsequent dilatation of the artery,forming an aneurysm.

The complications of atheroscleroticplaque may result in different clinical conse-

quences in different organ systems (Fig. 5.10).In the case of coronary plaque, lesions withgradually progressive expansion and a thickfibrous cap tend to narrow the vessel lumenand result in intermittent chest discomforton exertion (angina pectoris). In contrast,plaque that does not significantly compro-mise the vessel lumen but has characteristicsof vulnerability (thin fibrous cap and largelipid core) can rupture, leading to acutethrombosis and myocardial infarction (seeChapter 7). Such nonstenotic plaques areoften numerous and dispersed throughoutthe arterial tree, and because they do not sig-nificantly compromise the vessel lumen,they do not produce symptoms and oftenevade detection by exercise testing or evenangiography.

The description presented here of athero-genesis and its complications can explainthe limitations of widely employed treat-ments. For example, percutaneous interven-tion (angioplasty and stent placement) ofsymptomatic coronary stenoses is an effec-tive strategy for relief of angina pectoris buthas not been shown to prevent myocardialinfarction or prolong life. This disparitylikely reflects the general nature of vulnera-ble plaques as diffuse and angiographicallyunimpressive, such that they tend not to at-tract attention prior to rupture. It followsthat lifestyle modifications and drug the-rapies that curb the risk factors for plaqueformation are a critical foundation for pre-venting progressive atherogenesis.

Favors Occlusive Thrombus Favors Thrombus Resorbtion

Pro-coagulant

Antifibrinolytic

Tissue factor

PAI-1

Anticoagulants

Profibrinolytics

ThrombomodulinProtein SHeparin-like molecules

tPA, uPA

•• •

•

•

•

Figure 5.9. Competing factors in thrombosis. The clinical manifestations ofplaque disruption rely not only on the stability of the fibrous cap but also on thethrombogenic potential of the plaque core. The balance of physiological media-tors dictates the prominence of the thrombus, resulting in either luminal occlu-sion or resorption into the plaque. PAI-1, plasminogen activator inhibitor 1; tPA,tissue plasminogen activator; uPA, urokinase-type plasminogen activation.

Fig. 10

10090-05_CH05.qxd 8/31/06 5:35 PM Page 131

ATHEROSCLEROSIS RISK FACTORS

In the early portion of the twentieth cen-tury, it was widely believed that atheroscle-rosis was an inevitable process of aging. How-ever, in 1948, the landmark FraminghamHeart Study began to examine the relation-ship between specific attributes and cardio-vascular disease, establishing the concept ofatherosclerotic risk factors. Among laterstudies, the Multiple Risk Factor Interven-tion Trial screened more than 325,000 mento correlate risk factors with subsequent car-diovascular disease and mortality. More re-cently, the INTERHEART study evaluatedsuch relationships in 29,000 survivors ofmyocardial infarction in 52 countries, rep-resenting varied ethnic and economic back-grounds. These undertakings, and others like

them, have established the importance ofmodifiable risk factors for atherosclerosis, in-cluding (1) aberrant levels of circulatinglipids (dyslipidemia), (2) tobacco smoking,(3) hypertension, (4) diabetes mellitus, and(5) lack of physical activity and obesity(Table 5.1). The INTERHEART study suggeststhat these modifiable factors account for upto 90% of population-attributable cardiacrisk. Major nonmodifiable risk factors includeadvanced age, male gender, and heredity—that is, a history of coronary heart diseaseamong first-degree relatives at a young age(before age 55 for a male relative or beforeage 65 for a female relative).

In addition to these standard predictors,certain biological markers associated withthe development of cardiovascular eventshave been undergoing rigorous evaluation

132 Chapter Five

Narrowing of vessel by fibrous plaque

Plaque ulceration or rupture

Intraplaque hemorrhage

Peripheral emboli

Weakening of vessel wall

4

41

1

1

2 3

2 3

2 3

5

1

2

3

4

5

Stroke

• Embolic stroke

• Thrombotic stroke

Renal Artery Disease • Atheroembolic renal disease • Renal artery stenosis

Aneurysms

Coronary artery disease

• Myocardial ischemia

• Unstable angina

• Myocardial infarction

Peripheral artery disease • Limb claudication

Figure 5.10. Clinical sequelae of atherosclerosis. Complications of atherosclerosis arise from the mechanismslisted in the figure.

Tab. 1

10090-05_CH05.qxd 8/31/06 5:35 PM Page 132

Atherosclerosis 133

as “novel” risk factors. These include ele-vated circulating levels of (1) the aminoacid metabolite homocysteine, (2) the spe-cial lipoprotein particle Lp(a), and (3) cer-tain markers of inflammation, includingthe acute-phase reactant C-reactive protein.

The following sections address the tradi-tional risk factors, followed by a descriptionof the newer biomarkers.

Traditional Risk Factors

Dyslipidemia

A large and consistent body of evidence es-tablishes abnormal circulating lipid levels asa major risk factor for the development ofatherosclerosis. Observational studies haveshown that the United States and other soci-eties in which consumption of saturated fatand cholesterol levels are high, mortalityrates from coronary disease are higher com-pared with those in countries with tradition-ally low saturated fat intake and low serumcholesterol levels (e.g., Japan and certainMediterranean nations). Similarly, data fromthe Framingham Heart Study and other tri-als have shown that the risk of ischemicheart disease increases with higher total se-rum cholesterol levels. The coronary risk isapproximately twice as high for a personwith a total cholesterol level of 240 mg/dLcompared with a person whose cholesterollevel is 200 mg/dL.

However, not all lipoproteins bearing cho-lesterol are harmful. In fact, cholesterol has

various critical functions in normal physiol-ogy. All cells require cholesterol to formmembranes and maintain fluidity of thephospholipid bilayer. Some cells use choles-terol to synthesize specialized products, suchas steroid hormones and bile salts.

Elevated levels of LDL particles correlatewith an increased incidence of atherosclero-sis and coronary artery disease. When pre-sent in excess, LDL can accumulate in thesubendothelial space and undergo the che-mical modifications that further damage theintima, as described earlier, initiating andperpetuating the development of athero-sclerotic lesions. Thus, LDL is commonlyknown as “bad cholesterol.” Conversely, el-evated high-density lipoprotein (HDL) par-ticles appear to protect against atheroscle-rosis, likely because of HDL’s ability totransport cholesterol away from the periph-eral tissues back to the liver for disposal(termed reverse cholesterol transport; see Box 5.1) and because of its antioxidativeproperties. Thus, HDL has earned the moni-ker “good cholesterol.”

Elevated serum LDL may persist for manyreasons, including a high-fat diet or becauseof abnormalities in the LDL-receptor clear-ance mechanism. Patients with genetic de-fects in the LDL receptor, which leads to acondition known as familial hypercholes-terolemia, cannot remove LDL from the cir-culation efficiently. Heterozygotes with thiscondition have one normal and one defec-tive gene coding for the receptor. They dis-play high plasma LDL levels and developpremature atherosclerosis. Homozygoteswho completely lack functional LDL recep-tors may experience vascular events, such asacute myocardial infarction, as early as thefirst decade of life.

Increasing evidence also implicates tri-glyceride-rich lipoproteins, such as verylow-density lipoprotein (VLDL) and inter-mediate density lipoprotein (IDL), in thedevelopment of atherosclerosis. However,it remains undetermined whether theseparticles participate directly in atherogene-sis or simply keep company with low levelsof HDL cholesterol. Note that poorly con-trolled type 2 diabetes mellitus is often as-

TABLE 5.1. Common Cardiovascular Risk Factors

Modifiable Risk FactorsDyslipidemia (elevated LDL, decreased HDL)Tobacco smokingHypertensionDiabetes mellitus, metabolic syndromeLack of physical activityNonmodifiable risk factorsAdvanced ageMale genderHeredityHDL, high-density lipoprotein; LDL, low-density

lipoprotein.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 133

sociated with the combination of hyper-triglyceridemia and low HDL levels.

Lipid-Altering Therapy

Strategies that improve abnormal lipid lev-els can limit the consequences of athero-sclerosis. Many large studies of patients withcoronary disease show that dietary or phar-macologic reduction of serum cholesterolcan slow the progression of atheroscleroticplaque. These trials form the basis of screen-ing guidelines, devised by the National Cho-lesterol Education Program panel, whichrecommends a fasting lipid profile every years for all adults. These guidelines iden-

tify an “optimal” LDL cholesterol level as<100 mg/dL. Patients with established ath-erosclerosis, or those who have equivalentrisk (e.g., diabetes) should receive treatmentto attain this goal. An even lower goal of <70 mg/dL is recommended for patientswith atherosclerotic disease at the highestrisk of future vascular events: those who haverecently sustained an acute coronary syn-drome (see Chapter 7) and people with mul-tiple risk factors, especially when diabetes,the metabolic syndrome (described later inthe chapter), and/or smoking are present.

Diet and exercise are two important com-ponents of the risk reduction arsenal. For ex-ample, the Lyon Diet Heart Study demon-strated that patients with coronary diseasewho were randomized to a Mediterranean-style diet decreased their risk of recurrentcardiac events. The diet implemented in thisstudy included replacement of saturated fats with polyunsaturated fats (particularlyα-linolenic acid, an ω-3 fatty acid). In vitroevidence indicates that polyunsaturated fats may activate a transcription factor (per-oxisome proliferator-activated receptor α[PPAR-α]), which increases expression of the major HDL apoprotein (apo AI) and theenzyme lipoprotein lipase, and inhibits cytokine-induced expression of leukocyte ad-hesion molecules on endothelial cells. Theseactions are potentially antiatherogenic. Phys-ical activity and loss of excessive weight canalso improve the lipid profile, notably bylowering triglycerides and raising HDL.

When lifestyle modifications fail toachieve target values, pharmacologic agentsare used to improve abnormal lipid levels.The major groups of lipid-altering agents(see Chapter 17) include HMG-CoA reduc-tase inhibitors, niacin, fibric acid deriva-tives, cholesterol intestinal absorption in-hibitors, and bile acid–binding agents. TheHMG-CoA reductase inhibitors (also knownas statins) have emerged as the most effec-tive LDL-lowering drugs. They inhibit therate-limiting enzyme responsible for cho-lesterol biosynthesis. The resulting reductionin intracellular cholesterol concentrationpromotes increased LDL-receptor expressionand thus augmented clearance of LDL parti-cles from the bloodstream. Statins alsolower the rate of VLDL synthesis by theliver (thus lowering circulating triglyceridelevels) and, by an unknown mechanism,raise HDL.

Major clinical trials evaluating statin ther-apy have demonstrated consistently strikingreductions in ischemic cardiac events, theoccurrence of strokes, and (in many cases)mortality rates (Fig. 5.11). The benefits doc-umented in these studies have extended topeople with wide ranges of LDL, with orwithout known preexisting atheroscleroticdisease.

The effects of statins likely derive fromseveral mechanisms. The combination oflowering LDL and raising HDL may reducethe lipid content of atherosclerotic plaquesand thus favorably affect their biologic ac-tivity. Other potentially beneficial actionsinclude increased NO synthesis, enhancedfibrinolytic activity, inhibition of smoothmuscle proliferation and monocyte recruit-ment, and reduction in macrophage pro-duction of matrix-degrading enzymes. Invitro studies suggest that statins may also reduce inflammation by inhibiting themacrophage cytokines TNF-α, IL-1, and IL-6or by augmenting PPAR-α, thereby reducingendothelial expression of leukocyte adhe-sion molecules and macrophage tissue fac-tor production. Clinical trials have supportedan anti-inflammatory action of statins, be-cause they reduce plasma levels of C-reac-tive protein, a marker of inflammation de-

134 Chapter Five

Fig. 11

10090-05_CH05.qxd 8/31/06 5:35 PM Page 134

Atherosclerosis 135

scribed later. It should be noted, however,that it is experimentally difficult to separatethe LDL-lowering effect of statins from theiranti-inflammatory mechanisms because ofthe prominent role of oxidized LDL in initi-ating inflammatory cascades. Nonetheless,accumulating clinical and experimentaldata do suggest that at least some measure ofthe statin benefit derives from mechanismsapart from LDL lowering (so-called pleiotro-pic effects).

Tobacco Smoking

Numerous studies have shown that cigarettesmoking increases the risk of atherosclerosisand ischemic heart disease. Even minimalsmoking increases the risk, and the heaviestsmokers are at the greatest danger of cardio-vascular events.

Cigarette smoking could lead to athero-sclerotic disease in several ways, includingenhanced oxidative modification of LDL,decreased circulating HDL levels, endothe-lial dysfunction owing to tissue hypoxia andincreased oxidant stress, increased plateletadhesiveness, increased expression of solu-ble leukocyte adhesion molecules, inappro-priate stimulation of the sympathetic ner-vous system by nicotine, and displacementof oxygen by carbon monoxide from hemo-globin. Extrapolation from animal experi-ments suggests that smoking not only accel-erates atherogenesis but also increases thepropensity to thrombosis—both componentsof the “vulnerable patient.”

Fortunately, smoking cessation can reversesome of the adverse outcome. Epidemiologicstudies have shown that people who stopsmoking greatly reduce their likelihood of

4S

HP

S

CA

RE

WO

SC

OP

S

AS

CO

T

AF

CA

PS

/ Te

xCA

PS

LIP

ID

Included patientswith history of MI or CHD

Excluded patients with known CHD

Baseline LDLor total

cholesterol High Normal High Normal

(low HDL)NormalNormal

to HighNormalto High

Red

uctio

n in

C

HD

eve

nts

(%)

0

30

5

25

10

15

20

35

Figure 5.11. Statins reduce cardiovascular risk. Several major clinical studies have sup-ported the beneficial role of statin-induced lipid lowering in reducing coronary heart dis-ease (CHD) events. Regardless of baseline serum cholesterol levels, the benefits of statinsextend to patients with an established history of coronary disease and to those with sig-nificant risk factors without such a history. HDL, high-density lipoprotein; LDL, low-densitylipoprotein.

4S: Scandinavian Simvastatin Survival Study. Lancet 1994;344:1383–1389.

AFCAPS/TexCAPS: The Air Force/Texas Coronary Atherosclerosis Prevention Study.JAMA 1998;279:1615–1622.

ASCOT: Anglo-Scandinavian Cardiac Outcomes Trial—Lipid Lowering Arm. Lancet2003;361:1149–1158.

CARE: Cholesterol and recurrent events. N Engl J Med 1996;335:1001–1009.

HPS: Heart Protection Study. Lancet. 2004;363:757–767.

LIPID: The long-term intervention with Pravastatin in ischaemic disease. N Engl JMed 1998;339:1349–1357.

WOSCOPS: West of Scotland Coronary Primary Prevention Study. N Engl J Med1995;333:1301–1307.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 135

coronary heart disease compared with thosewho continue to smoke. In one study, after 3 years of cessation, the risk of coronary arterydisease for former smokers became similar tosubjects who never smoked.

Hypertension

Elevated blood pressure (either systolic or di-astolic) augments the risk of developing atherosclerosis, coronary heart disease, andstroke (see Chapter 13). The association ofelevated blood pressure with cardiovasculardisease does not appear to have a specificthreshold. Rather, risk increases continu-ously with progressively higher pressure val-ues. Particularly in older persons, systolicpressure predicts adverse outcomes more re-liably than does diastolic pressure.

Hypertension may accelerate atheroscle-rosis in several ways. Animal studies haveshown that elevated blood pressure injuresvascular endothelium and may increase thepermeability of the vessel wall to lipopro-teins. In addition to causing direct endothe-lial damage, increased hemodynamic stresscan increase the number of scavenger recep-tors on macrophages, thus enhancing thedevelopment of foam cells. Cyclic circum-ferential strain, increased in hypertensive arteries, can augment smooth muscle cellproduction of proteoglycans that bind andretain LDL particles, promoting their accu-mulation in the intima and facilitating theiroxidative modification. Angiotensin II, amediator of hypertension, acts not only as avasoconstrictor but also as a proinflamma-tory cytokine. Thus, hypertension may alsopromote atherogenesis by contributing toan inflammatory state.

Antihypertensive Therapy

Like dyslipidemias, treatment of hyperten-sion should start with lifestyle modificationsbut often requires pharmacologic interven-tion. The Dietary Approaches to Stop Hyper-tension studies demonstrate that a diet highin fruits and vegetables, with dairy productslow in fat, and an overall reduced sodiumcontent significantly improves systolic and

diastolic blood pressures. Regular exercisecan also reduce resting blood pressure levels.

Many medications can lower blood pres-sure, including diuretics, β-adrenergic recep-tor antagonists, drugs that interfere with therenin-angiotensin system, calcium channelblockers, and α-adrenergic inhibitors. Thebenefits and limitations of these therapiesare described in Chapters 13 and 17.

Diabetes Mellitus and the Metabolic Syndrome

The global incidence of diabetes mellitus isestimated at 170 million people and pro-jected to grow 40% worldwide by 2030. Inthe United States alone, 18.2 million peopleare diabetic, and projections suggest thatone of three children born in 2000 willeventually develop the condition. With athreefold to fivefold increased risk of acutecoronary events, 80% of diabetic patientssuccumb to atherosclerosis-related condi-tions, including coronary heart disease,stroke, and peripheral artery disease. Accor-dingly, diabetes is considered an atheroscle-rotic risk equivalent, elevating it to the samerisk category as people with a history of myo-cardial infarction.

The predisposition of diabetic patients toatherosclerosis may relate in part to the non-enzymatic glycation of lipoproteins (whichenhances uptake of cholesterol by scavengermacrophages, as described earlier) or to a pro-thrombotic tendency and antifibrinolyticstate that prevails in many patients with dia-betes. Diabetic patients frequently have im-paired endothelial function, gauged by thereduced bioavailability of NO and increasedleukocyte adhesion. Tight control of serumglucose levels in diabetic patients reduces therisk of microvascular complications such asretinopathy and nephropathy. At least onestudy has also demonstrated a reduction inmacrovascular outcomes such as myocardialinfarction and stroke in patients with type 1diabetes who followed an intense antidia-betic regime. In addition, control of hyper-tension and dyslipidemia in diabetic patientsconvincingly reduces the risk of cardiac andcerebrovascular complications.

136 Chapter Five

10090-05_CH05.qxd 8/31/06 5:35 PM Page 136

Atherosclerosis 137

The metabolic syndrome (previously knownas the insulin resistance syndrome or syn-drome X) is a descriptor for a clustering of riskfactors, including hypertension, hypertriglyc-eridemia, reduced HDL, cellular insulin resis-tance (often leading to glucose intolerance),and visceral obesity (excessive adipose tissuein the abdomen). This syndrome is associatedwith a high risk for atherosclerosis in both diabetic and nondiabetic patients, and the National Health and Nutrition ExaminationSurvey estimates that an astounding 44% ofAmericans have the metabolic syndromebased on current guidelines. The presence ofinsulin resistance in this syndrome appears topromote atherogenesis long before affectedpersons develop overt diabetes.

Lack of Physical Activity

Exercise may mitigate atherogenesis in sev-eral ways. In addition to beneficial effects onthe lipid profile and blood pressure, exerciseenhances insulin sensitivity and the en-dothelial production of NO. Long-termprospective studies of both men and womenindicate that even modest activities, such as brisk walking, can protect against cardio-vascular mortality. Although no large ran-domized primary prevention trials have ex-amined the effects of exercise on cardiacevent rates, the proven benefits on the car-diovascular risk profile should promote in-creased physical activity for anyone at riskof developing atherosclerotic disease.

Estrogen Status

Cardiovascular disease dominates over othercauses of mortality in women, includingbreast and other cancers. Before menopause,women have a lower incidence of coronaryevents than men. After menopause, however,men and women have similar rates. This ob-servation suggests that estrogen (the levels ofwhich decline after menopause) may haveatheroprotective properties. Physiologic es-trogen levels in premenopausal women raiseHDL and lower LDL and lipoprotein (a), de-scribed later in the chapter. Experimentally,estrogen also exhibits potentially beneficial

antioxidant and antiplatelet actions and im-proves endothelium-dependent vasodilation.

Early observational studies suggested thathormone replacement therapy reduced therisk of coronary artery disease in postmeno-pausal women, prompting many physiciansto prescribe such medications for cardiovas-cular prevention purposes. However, theHeart and Estrogen/progestin ReplacementStudy actually demonstrated an associationbetween such hormone use and an early increased risk of vascular events in womenwith preexisting coronary disease. In addi-tion, in 2003, randomized primary preven-tion studies from the Women’s Health Ini-tiative were terminated prematurely becauseof the finding that estrogen-plus-progestintreatment increased cardiovascular risk by24% overall, with a striking 81% higher riskduring the first year of therapy. Of note,these troubling outcomes did not appear in the cohort of patients randomized to the estrogen-only arm of the study. Furtheranalyses will help determine if safe hormonereplacement approaches are possible. Mean-while, because currently available clinicaltrial data do not show that gonadal hor-mone therapy is cardioprotective and mayactually be harmful, such therapy shouldnot be commenced at present for the solegoal of reducing cardiovascular risk.

Biomarkers

Despite identification of the well-establishedrisk factors just described, one out of five car-diovascular events occur in patients lackingthese attributes. In conjunction with thegrowing knowledge about the pathogenesisof atherosclerosis, several novel markers ofatherosclerotic risk have emerged.

Homocysteine

Some studies have shown a significant rela-tionship between circulating levels of theamino acid homocysteine and the incidenceof coronary, cerebral, and peripheral arterydisease. The balance of current data suggestan overall modest contribution of this factorto cardiovascular risk in population studies.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 137

The mechanism by which homocysteinemight increase atherosclerotic risk remainsundetermined, but current evidence suggeststhat abnormally high levels may promote oxidative stress, vascular inflammation, andplatelet adhesiveness. Hyperhomocysteine-mia can result from genetic defects in me-thionine metabolism or from insufficient di-etary intake of folic acid, a cofactor in themethionine pathway. Although folic acid andother B-vitamin supplements reduce highserum homocysteine levels, data thus far havenot proved that such therapy actually reducesatherosclerotic disease or its complications.

Lipoprotein (a)

Lipoprotein (a), referred to as Lp(a) and pro-nounced “L-P-little-a,” has also been identi-fied as an independent risk factor for coro-nary artery disease in some studies. Lp(a) is a special form of LDL whose major apo-lipoprotein (apo B-100) links by a disulfidebridge to another protein, known as apo(a).Apo(a) structurally resembles plasminogen, aplasma protein important in the endogenouslysis of fibrin clots (see Chapter 7). Thus, thedetrimental effect attributed to Lp (a) may re-late to competition with normal plasminogenactivity. As with homocysteine, not all popu-lation studies support a link between Lp(a)and cardiovascular events, though peoplewith the highest Lp(a) levels do convincinglyappear to be at increased risk.

An person’s Lp(a) level is primarily deter-mined by inheritance and has a skewed dis-tribution, with higher levels among blacks.Diet and exercise have little influence onLp(a), and drugs designed to specificallylower its level do not exist. However, nico-tinic acid (niacin) is one agent that reducesLp(a) as one of its multiple beneficial lipideffects. Thus far, no evidence shows thatspecifically targeting Lp(a) by drug therapyimproves cardiovascular outcomes.

C-Reactive Protein and Other Markers of Inflammation

Because the pathogenesis of atherosclerosisinvolves inflammation at every stage, mark-

ers of inflammation have undergone evalu-ation as predictors of cardiac risk. Recall thatthe process of lipoprotein entry and modifi-cation in the vessel wall triggers the releaseof cytokines, followed by leukocyte infiltra-tion, more cytokine release, and smoothmuscle migration into, and proliferationwithin, the intima. Involved cytokines (e.g.,IL-6) mobilize to the liver and incite in-creased production of acute-phase reactants,including C-reactive protein (CRP), fibrino-gen, and serum amyloid A.

Of these molecules, CRP has shown thegreatest promise as a marker of low-grade sys-temic inflammation associated with athero-sclerotic disease. Large studies of apparentlyhealthy men and women indicate that thosewith higher basal CRP levels have a greatlyincreased risk of adverse cardiovascular out-comes, independent of serum cholesterol lev-els. Recent prospective studies affirm high-sensitivity CRP as an independent predictorof myocardial infarction, stroke, peripheralartery disease, and sudden cardiac death.

In addition to its role as a marker of risk,CRP may actually participate as a mediator ofatherogenesis. For example, CRP can acti-vate complement and thus contribute to a sustained inflammatory state. The lipid-lowering HMG-CoA reductase inhibitors de-crease CRP levels, but current evidence doesnot yet prove that therapy should specifi-cally target CRP lowering.

Infection

Several studies have identified infectiousagents (e.g., herpes viruses, Chlamydia pneu-moniae) within some atherosclerotic lesions,raising the question of their potential role inatherogenesis. The studies have generatedsubstantial controversy, and definite proofof a causal role is lacking. Although it is un-certain if certain infections truly play a rolein atherogenesis, investigation of the possi-bility is prompted by theoretical reasons. Forexample, Chlamydia species produce heatshock protein 60 (HSP-60), which activatesmacrophages and stimulates the productionof matrix metalloproteinases that can impairthe stability of the atherosclerotic plaque’s fi-

138 Chapter Five

10090-05_CH05.qxd 8/31/06 5:35 PM Page 138

Atherosclerosis 139

brous cap. Other possible atherogenic prop-erties of chlamydial HSP-60 and bacterial en-dotoxins include induction of foam cell for-mation, lipoprotein oxidation, and increasedprocoagulant activity. Although the patho-genic relationships remain unproven, someresearchers believe that infectious agents fur-nish an additional source of endothelial in-jury and inflammation that could initiate orexacerbate atherogenesis. To date, a numberof well-powered trials have not shown thatantibiotic treatment directed against suchputative infections reduces the risk of futurecardiac events in survivors of acute coronarysyndromes.

Management of Risk Factors

Despite accumulating knowledge of thepathogenesis of atherosclerosis and its clini-cal sequelae, this age-old adversary to humanlife remains a major cause of death in themodern world. Although improvements in cardiovascular care have reduced age-adjusted mortality from this condition, it willcontinue to be a menace as our populationages and developing countries embrace theadverse dietary and activity habits of a West-ern lifestyle. Ongoing research of the biologyof atherosclerosis, as well as advances in ther-apeutic procedures and medications, will un-doubtedly continue to further our abilities to combat this condition. Yet we have notfully capitalized on what we already know:much cardiovascular risk is modifiable. Effec-tive control of the modifiable risk factors de-scribed earlier remains a critical componentto tame this global scourge. It is here that thepatient-physician relationship and the role ofmedical professionals as community leadersadvocating healthy lifestyles remain of cardi-nal importance.

SUMMARY

1. The normal arterial wall is a three-layeredstructure, key aspects of which include asingle endothelial layer in the intima,smooth muscle cells in the media, andthe outer adventitia (see Fig. 5.1).

2. Early in atherogenesis, injurious and inflammatory stimuli activate endothe-lial and smooth muscle cells. The result-ing cascade of events recruits immunecells to the vessel wall, fueling a per-sistent inflammatory state believed tounderlie progression of the disease (seeFig. 5.2).

3. Mechanisms that contribute to athero-sclerosis shape the forming plaque overdecades (see Figs. 5.3 through 5.7).Plaque diversity can be broadly catego-rized as either stable or vulnerable (seeFig. 5.8).

4. Clinical expression of atherosclerosiscommonly results from narrowing of thevessel lumen, calcification or weakeningof the arterial wall, or plaque disruptionwith superimposed thrombus formation.Common manifestations include anginapectoris, myocardial infarction, stroke,and peripheral arterial disease (see Fig.5.10).

5. Many of the risk factors that lead to ather-osclerosis are modifiable (see Table 5.1),providing the opportunity for physiciansand other health care providers to educatepatients to help prevent or reduce the pro-gression of the disease.

Acknowledgments

Contributors to the previous editions of this chapterwere Rushika Fernandopulle, MD; Gopa Bhat-tacharyya, MD; Mary Beth Gordon, MD; JosephLoscalzo, MD, PhD; and Peter Libby, MD.

Additional Reading

AHA Science Advisory: Lyon Diet Heart Study. Bene-fits of a Mediterranean-style, National CholesterolEducation Program/American Heart AssociationStep I Dietary Pattern on cardiovascular disease.Circulation 2001;103:1823–1825.

Ansell BJ, Watson KE, Fogelman AM, et al. High-density lipoprotein function. J Am Coll Cardiol2005;46:1792–1798.

Beckman JA, Creager MA, Libby P. Diabetes and ath-erosclerosis. JAMA 2002;287:2570–2581.

Bhatt DL, Steg PG, Hirsch AT. International preva-lence, recognition, and treatment of cardiovascu-lar risk factors in outpatients with atherothrom-bosis. JAMA 2006;295:180–189.

10090-05_CH05.qxd 8/31/06 5:35 PM Page 139

Executive summary of the Third Report of the Na-tional Cholesterol Education Project Expert Panelon Detection, Evaluation, and Treatment of HighBlood Cholesterol in Adults. JAMA 2001;285:2486–2496.

Gotto AM. Evolving concepts of dyslipidemia, ath-erosclerosis, and cardiovascular disease. J Am CollCardiol 2005;46:1219–1224.

Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosisand management of the metabolic syndrome. AnAmerican Heart Association/ National Heart, Lung,and Blood Institute scientific statement. Circula-tion 2005;112:2735–2752.

Hansson GK. Mechanisms of disease: Inflammation,atherosclerosis, and coronary artery disease. N EnglJ Med 2005;352:1685–1695.

Libby P. Atherosclerosis: the new view. ScientificAmerican 2002;286:46–55.

Libby P. The forgotten majority: unfinished businessin cardiovascular risk reduction. J Am Coll Cardiol2005;46:1225–1228.

Libby P, Ridker PM, Maseri A. Inflammation and ath-erosclerosis. Circulation 2002;105:1135–1143.

Nathan DM, Cleary PA, Backlund JC, et al. Intensivediabetes treatment and cardiovascular disease inpatients with type 1 diabetes. N Eng J Med 2005;353:2643–2653.

Ridker PM, Libby P. Risk factors for atherothrom-botic disease. In: Zipes D, Libby P, Bonow R,Braunwald E, ed. Braunwald’s Heart Disease: ATextbook of Cardiovascular Medicine. 7th Ed.Philadelphia: Elsevier Saunders, 2005:939–958.

Schönbeck U, Libby P. Inflammation, immunity,and HMG-CoA reductase inhibitors: statins asantiinflammatory agents? Circulation 2004;109:II-18–II-26.

Tsimikas S, Brilakis ES, Miller ER, et al. Oxidizedphospholipids, Lp(a) lipoprotein, and coronaryartery disease. N Engl J Med 2005;353:46–57.

Yusuf S, Hawken S, Ounpuu S, et al. Effect of poten-tially modifiable risk factors associated with myo-cardial infarction in 52 countries (the INTER-HEART study). Lancet 2004;364:937–952.

140 Chapter Five

10090-05_CH05.qxd 8/31/06 5:35 PM Page 140

Chapter 5—Author Queries1. AU: Edit fixes misplaced modifier, but does it retain intended meaning?2. AU: Please provide a definition of IP-10 abbreviation.3. AU: Correct to add this reference to the figure?

10090-05_CH05.qxd 8/31/06 5:35 PM Page a

141

DETERMINANTS OF MYOCARDIAL OXYGENSUPPLY AND DEMANDMyocardial Oxygen SupplyMyocardial Oxygen Demand

PATHOPHYSIOLOGY OF ISCHEMIAFixed Vessel NarrowingEndothelial Cell DysfunctionOther Causes of Myocardial Ischemia

CONSEQUENCES OF ISCHEMIAIschemic Syndromes

CLINICAL FEATURES OF CHRONIC STABLE ANGINA

HistoryPhysical ExaminationDiagnostic StudiesNatural History

TREATMENTMedical Treatment of an Acute Episode of AnginaMedical Treatment to Prevent Recurrent IschemicEpisodesMedical Treatment to Prevent Acute Cardiac EventsRevascularizationMedical Versus Revascularization Therapy

C H A P T E R

6Ischemic Heart DiseaseHaley NaikMarc S. SabatineLeonard S. Lilly

In 1772, the British physician William Heber-den reported a disorder in which patientsdeveloped an uncomfortable sensation inthe chest when walking. Labeling it anginapectoris, Heberden noted that this discom-fort would disappear soon after the patientstood still but would recur with similar activities. Although he did not know thecause, it is likely that he was the first to describe the symptoms of ischemic heartdisease, a condition of imbalance betweenmyocardial oxygen supply and demandmost often caused by atherosclerosis of thecoronary arteries. Ischemic heart diseasenow afflicts millions of Americans and is

the leading cause of death in industrializednations.

The clinical presentation of ischemicheart disease can be highly variable andforms a spectrum of syndromes (Table 6.1).For example, ischemia may be accompa-nied by the same exertional symptoms de-scribed by Heberden. In other cases, it mayoccur without any clinical manifestationsat all, a condition termed silent ischemia.This chapter describes the causes and con-sequences of chronic ischemic heart dis-ease syndromes and provides a frameworkfor the diagnosis and treatment of affectedpatients.

Tab. 1

10090-06_CH06.qxd 8/31/06 5:22 PM Page 141

Angina pectoris remains the most com-mon manifestation of ischemic heart dis-ease and literally means “strangling in thechest.” Although other conditions may leadto similar discomfort, angina refers specifi-cally to the uncomfortable sensation in thechest and neighboring structures that arisesfrom an imbalance between myocardial oxy-gen supply and demand.

DETERMINANTS OF MYOCARDIALOXYGEN SUPPLY AND DEMAND

In the normal heart, the oxygen require-ments of the myocardium are continuouslymatched by the coronary arterial supply.Even during vigorous exercise, when themetabolic needs of the heart increase, sodoes the delivery of oxygen to the myo-cardial cells so that the balance is main-tained. The following sections describe thekey determinants of myocardial oxygensupply and demand in a normal person (Fig.6.1) and how they are altered by the pres-ence of atherosclerotic coronary artery dis-ease (CAD).

Myocardial Oxygen Supply

The supply of oxygen to the myocardium de-pends on the oxygen content of the bloodand the rate of coronary blood flow. Theoxygen content is determined by the hemo-globin concentration and the degree of sys-temic oxygenation. In the absence of ane-mia or lung disease, oxygen content remainsfairly constant. In contrast, coronary bloodflow is much more dynamic, and regulationof that flow is responsible for matching theoxygen supply with metabolic requirements.

As in all blood vessels, coronary arteryflow (Q) is directly proportional to the ves-sel’s perfusion pressure (P) and is inverselyproportional to coronary vascular resistance(R). That is,

However, unlike other arterial systems inwhich the greatest blood flow occurs duringsystole, the predominance of coronary perfusiontakes place during diastole. The reason for thisis that systolic flow is obstructed by com-

QPR

=

142 Chapter Six

TABLE 6.1. Clinical Definitions

Syndrome Description

Ischemic heart disease

Angina pectoris

Stable angina

Variant angina

Unstable angina

Silent ischemia

Myocardial infarction

Condition in which imbalance between myocardial oxygen supply and demandresults in myocardial hypoxia and accumulation of waste metabolites; mostoften caused by atherosclerotic disease of the coronary arteries (coronary arterydisease)

Uncomfortable sensation in the chest and neighboring anatomic structures produced by myocardial ischemia

Chronic pattern of transient angina pectoris, precipitated by physical activity oremotional upset, relieved by rest within a few minutes; episodes often associ-ated with temporary depression of the ST segment, but permanent myocardialdamage does not result

Typical anginal discomfort, usually at rest, which develops because of coronaryartery spasm rather than an increase of myocardial oxygen demand; episodesoften associated with transient shifts of the ST segment (usually ST elevation)

Pattern of increased frequency and duration of angina episodes produced by lessexertion or at rest; high frequency of progression to myocardial infarction if untreated

Asymptomatic episodes of myocardial ischemia; can be detected by electro-cardiogram and other laboratory techniques

Region of myocardial necrosis usually caused by prolonged cessation of blood sup-ply; most often results from acute thrombus at site of coronary atheroscleroticstenosis; may be first clinical manifestation of ischemic heart disease, or theremay be a history of angina pectoris

AQ1

Fig. 1

10090-06_CH06.qxd 8/31/06 5:22 PM Page 142

Ischemic Heart Disease 143

pression of the small coronary branches asthey course through the contracting myo-cardium. Coronary flow is unimpaired in diastole because the relaxed myocardiumdoes not compress the coronary vasculature.Thus, in the case of the coronaries, perfu-sion pressure can be approximated by theaortic diastolic pressure. Conditions that de-crease aortic diastolic pressure (such as hy-potension or aortic valve regurgitation) de-crease coronary artery perfusion pressureand may impair myocardial oxygen supply.

Coronary vascular resistance is the othermajor determinant of coronary blood flow.In the normal artery, this resistance is dy-namically modulated by (1) forces that ex-ternally compress the coronary arteries and(2) factors that alter intrinsic coronary tone.

External Compression

External compression is exerted on the coro-nary vessels during the cardiac cycle by con-traction of the surrounding myocardium.The degree of compression is directly relatedto intramyocardial pressure and is thereforegreatest during systole, as described in theprevious section. Moreover, when the myo-cardium contracts, the subendocardium, ad-jacent to the high intraventricular pressure,is subjected to greater force than are theouter muscle layers. This is one reason whythe subendocardium is the region most vul-nerable to ischemic damage.

Intrinsic Control of Coronary Tone

Unlike most tissues, the heart cannot in-crease oxygen extraction on demand be-cause in its basal state it removes nearly asmuch oxygen as possible from its blood sup-ply. Thus, any additional oxygen requirementmust be met by an increase in blood flow, andautoregulation of coronary vascular resis-tance is the most important mediator of thisprocess. Factors that participate in the regu-lation of coronary vascular resistance in-clude the accumulation of local metabolites,endothelium-derived substances, and neuralinnervation.

Metabolic Factors

The accumulation of local metabolites signif-icantly affects coronary vascular tone and actsto modulate myocardial oxygen supply tomeet changing metabolic demands. Duringstates of hypoxemia, aerobic metabolism andoxidative phosphorylation in the mitochon-dria are inhibited. High-energy phosphates,including adenosine triphosphate (ATP),cannot be regenerated. Consequently, adeno-sine diphosphate (ADP) and monophosphate(AMP) accumulate and are subsequently de-graded to adenosine. Adenosine is a potentvasodilator and is thought to be the primemetabolic mediator of vascular tone. By bind-ing to receptors on vascular smooth muscle,adenosine decreases calcium entry into cells,which leads to relaxation, vasodilatation,

Wall stress(P r/2h)

Heart rate

Contractility

O2 content

Myocardial oxygen supply Myocardial oxygen demand

Coronary blood flow

• coronary perfusion pressure • coronary vascular resistance

• external compression• intrinsic regulation

• local metabolites• endothelial factors• neural innervation

Figure 6.1. Major determinants of myocardial oxygen supply and demand.h, ventricular wall thickness; P, ventricular pressure; r, ventricular radius.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 143

and increased coronary blood flow. Othermetabolites that act locally as vasodilatorsinclude lactate, acetate, hydrogen ions, andcarbon dioxide.

Endothelial Factors

Endothelial cells of the arterial wall producenumerous vasoactive substances that con-

tribute to the regulation of vascular tone.Vasodilators produced by the endotheliuminclude nitric oxide (NO), prostacyclin, andendothelium-derived hyperpolarizing factor(EDHF). Endothelin 1 is an example of anendothelium-derived vasoconstrictor.

The identification and important actionsof endothelium-derived NO are described inBox 6.1.

144 Chapter Six

Box 6.1 Endothelium-Derived Relaxing Factor, Nitric Oxide, and the Nobel Prize

Normal arterial endothelial cells synthesize potent vasodilator substances that contributeto the modulation of vascular tone. Among the first of these to be identified were prosta-cyclin (an arachidonic acid metabolite) and a substance termed endothelium-derived re-laxing factor (EDRF).

EDRF was first studied in the 1970s. In experimental preparations, it was shown thatacetylcholine (ACh) has two opposite actions on blood vessels. Its direct effect on vascu-lar smooth muscle cells is to cause vasoconstriction, but when an intact endothelial liningoverlies the smooth muscle cells, vasodilation occurs instead. Subsequent experimentsshowed that ACh causes the endothelial cells to release the chemical mediator EDRF,which quickly diffuses to the adjacent smooth muscle cells and results in their relaxationwith subsequent vasodilation of the vessel.

Subsequent research demonstrated that the mysterious EDRF is actually the nitric oxide(NO) radical. When ACh (or another endothelial-dependent vasodilator such as serotoninor histamine) binds to endothelial cells, intracellular free calcium increases, activating theenzyme nitric oxide synthase (NOS). NOS catalyzes the formation of NO from the aminoacid L-arginine (see the accompanying figure). NO diffuses from the endothelium to theadjacent vascular smooth muscle, where it activates guanylyl cyclase (G-cyclase). G-cyclasein turn forms cyclic guanosine monophosphate (cGMP) from guanosine triphosphate. Theincreased intracellular cGMP results in smooth muscle cell relaxation through mechanismsthat involve a reduction in cytosolic Ca++. The increase in cGMP is also associated with thebeneficial antimigratory effects of the smooth muscle cells.

AGONIST(e.g., ACh, histamine, serotonin)

Nitric oxidesynthase

Nitric oxide

G-cyclase

L-Arginine

GTP cGMP

Nitroprussideor nitroglycerin

RELAXATION

L-Citruline

Nitric oxideO2

Endothelialcell

Smoothmuscle cell

Box 1

10090-06_CH06.qxd 8/31/06 5:22 PM Page 144


In contrast to the endothelial-dependent vasodilators, some agents cause smooth mus-cle relaxation independent of the presence of endothelial cells. For example, the drugssodium nitroprusside and nitroglycerin result in vasodilation by providing an exogenoussource of NO to vascular smooth muscle cells, thereby activating G-cyclase and formingcGMP without endothelial cell participation.

In the cardiac catheterization laboratory, the intracoronary administration of ACh in anormal person causes vasodilation of the vessel, presumably through the release of NO.However, in conditions of endothelial dysfunction, such as atherosclerosis, intracoronaryACh administration results in paradoxical vasoconstriction instead. This likely reflects re-duced production of NO by the dysfunctional endothelial cells, resulting in unopposed di-rect vasoconstriction of the smooth muscle by ACh. Of particular interest is that the lossof vasodilatory response to infused ACh is evident in persons with certain cardiac risk fac-tors (e.g., elevated LDL cholesterol, hypertension, cigarette smoking) even before the phys-ical appearance of atheromatous plaque. Thus, the impaired release of NO may be an earlyand sensitive predictor for the later development of atherosclerotic lesions.

The significance of these discoveries was highlighted in 1998, when the Nobel Prize inmedicine was awarded to the scientists who discovered the critical role of NO as a cardio-vascular signaling molecule.

Modified from Furchgott RF. The discovery of endothelium-derived relaxing factor and its importance in theidentification of nitric oxide. JAMA 1996;276:1186–1188.

NO regulates vascular tone by diffusing into and then relaxing neighboring arterialsmooth muscle by a cyclic guanosine mono-phosphate (cGMP)–dependent mechanism.The production of NO by normal endothe-lium occurs in the basal state and is addition-ally stimulated by many substances and con-ditions. For example, its release is augmentedwhen the endothelium is exposed to acetyl-choline (ACh), thrombin, products of aggre-gating platelets (e.g., serotonin and ADP), oreven the shear stress of blood flow. Althoughthe direct effect of many of these substanceson vascular smooth muscle is to cause vaso-constriction, the induced release of NO fromthe normal endothelium results in vasodilata-tion instead (Fig. 6.2).

Prostacyclin, an arachidonic acid metabo-lite, has vasodilator properties similar to thoseof NO (see Fig. 6.2). It is released from en-dothelial cells in response to many stimuli,including hypoxia, shear stress, acetylcho-line, and platelet products (e.g., serotonin). Itcauses relaxation of vascular smooth muscleby a cyclic AMP–dependent mechanism.

EDHF also appears to have importantvasodilatory properties. Like endothelial-

derived NO, it is a diffusible substance re-leased by the endothelium that hyperpolar-izes (and therefore relaxes) neighboring vas-cular smooth muscle cells. EDHF is releasedby many of the same factors that stimulateNO, including acetylcholine and normalpulsatile blood flow. In the coronary circu-lation, EDHF appears to be more importantin modulating relaxation in small arteriolesthan in the large conduit arteries.

Endothelin 1 is a potent vasoconstrictorproduced by endothelial cells that partiallycounteracts the actions of the endothelialvasodilators. Its expression is stimulated byseveral factors, including thrombin, angio-tensin II, epinephrine, and the shear stressof blood flow.

Under normal circumstances, the healthyendothelium promotes vascular smooth mus-cle relaxation (vasodilatation) through elabo-ration of substances such as NO and prosta-cyclin, the influences of which predominateover the endothelial vasoconstrictors (seeFig. 6.2). However, as described later in the chapter, dysfunctional endothelium(e.g., in atherosclerotic vessels) secretes re-duced amounts of vasodilators, causing

Fig. 2

10090-06_CH06.qxd 8/31/06 5:22 PM Page 145

the balance to shift toward vasoconstrictioninstead.

Neural Factors

The neural control of vascular resistance hasboth sympathetic and parasympathetic com-ponents. Under normal circumstances, thecontribution of the parasympathetic ner-vous system appears minor, but sympathe-tic receptors play an important role. Coro-nary vessels contain both α-adrenergic and β2-adrenergic receptors. Stimulation ofα-adrenergic receptors results in vasocon-striction, whereas β2-receptors promotevasodilatation.

It is the interplay among the metabolic,endothelial, and neural regulating factorsthat determines the net impact on coronaryvascular tone. For example, catecholaminestimulation of the heart may initially causecoronary vasoconstriction via the α-adrenergicreceptor neural effect. However, catechola-mine stimulation also increases myocardialoxygen consumption through increasedheart rate and contractility (β1-adrenergic ef-fect), and the resulting increased productionof local metabolites induces net coronary di-latation instead.

Myocardial Oxygen Demand

The three major determinants of myocardialoxygen demand are (1) ventricular wall stress,(2) heart rate, and (3) contractility (which isalso termed the inotropic state). Additionally,very small amounts of oxygen are consumedin providing energy for basal cardiac metabo-lism and electrical depolarization.

Ventricular wall stress (s) is the tangen-tial force acting on the myocardial fibers,tending to pull them apart, and energy is ex-pended in opposing that force. Wall stress isrelated to intraventricular pressure (P), theradius of the ventricle (r), and ventricularwall thickness (h) and is approximated byLaPlace’s relationship:

Thus, wall stress is directly proportional tosystolic ventricular pressure. Circumstancesthat increase pressure development in theleft ventricle, such as aortic stenosis or hy-pertension, increase wall stress and myocar-dial oxygen consumption. Conditions thatdecrease ventricular pressure, such as anti-hypertensive therapy, reduce myocardialoxygen consumption.

sP r

h= ×

2

146 Chapter Six

Figure 6.2. Endothelium-derived vasoactive substances and theirregulators. Endothelium-derived vasodilators are shown on the left andinclude nitric oxide (NO), prostacyclin, and endothelium-derived hyper-polarizing factor (EDHF). Endothelin 1 is an endothelium-derived vaso-constrictor. In the normal state, the vasodilator influence predominatesover that of vasoconstriction. ACh, acetylcholine.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 146


Because wall stress is also directly propor-tional to the radius of the left ventricle, con-ditions that augment left ventricular (LV)filling (e.g., mitral or aortic regurgitation)raise wall stress and oxygen consumption.Conversely, any physiologic or pharmaco-logic maneuver that decreases LV filling andsize (e.g., nitrate therapy) reduces wall stressand myocardial oxygen consumption.

Finally, wall stress is inversely proportionalto ventricular wall thickness because theforce is spread over a greater muscle mass. Ahypertrophied heart has lower wall stress andoxygen consumption per gram of tissue thana thinned-wall heart. Thus, when hypertro-phy develops in conditions of chronic pres-sure overload, such as aortic stenosis, it servesa compensatory role in reducing oxygen con-sumption.

The second major determinant of my-ocardial oxygen demand is heart rate. If theheart rate accelerates—during physical exer-tion, for example—the number of contrac-tions and the amount of ATP consumed perminute increases and oxygen requirementsrise. Conversely, slowing the heart rate (e.g.,using a β-blocker drug) decreases ATP uti-lization and oxygen consumption.

The third major determinant of oxygendemand is myocardial contractility, a mea-sure of the force of contraction (see Chapter9). Circulating catecholamines, or the ad-ministration of positive inotropic drugs, directly increase the force of contraction,which augments oxygen utilization. Con-versely, negative inotropic effectors, such asβ-adrenergic–blocking drugs, decrease myo-cardial oxygen consumption.

In the normal state, autoregulatory me-chanisms adjust coronary tone to matchmyocardial oxygen supply with oxygen re-quirements. In the absence of obstructivecoronary disease, these mechanisms main-tain a fairly constant rate of coronary flow,as long as the aortic perfusion pressure is ap-proximately 60 mm Hg or greater. In the set-ting of advanced coronary atherosclerosis,however, the fall in perfusion pressure distalto the arterial stenosis, along with dysfunc-tion of the endothelium of the involved seg-ment, sets the stage for a mismatch between

the available blood supply and myocardialmetabolic demands.

PATHOPHYSIOLOGY OF ISCHEMIA

The traditional view has been that myocar-dial ischemia in CAD results from fixed ath-erosclerotic plaques that narrow the vessel’slumen and limit myocardial blood supply.However, recent research has demonstratedthat the reduction of blood flow in this condition results from the combination offixed vessel narrowing and abnormal vascu-lar tone, contributed to by atherosclerosis-induced endothelial cell dysfunction.

Fixed Vessel Narrowing

The hemodynamic significance of fixed ath-erosclerotic coronary artery stenoses relatesto both the fluid mechanics and the anatomyof the vascular supply.

Fluid Mechanics

Poiseuille’s law states that for flow througha vessel,

in which Q is flow, ΔP is the pressure differ-ence between the points being measured, r isthe vessel radius, η is the fluid viscosity, andL is the vessel length. By analogy to Ohm’slaw, flow is also equal to the pressure differ-ence divided by the resistance (R) to flow:

By combining these two formulas, resistanceto blood flow in a vessel can be expressed as:

Thus, vascular resistance is governed, in part,by the geometric component L/r 4. That is,the hemodynamic significance of a stenoticlesion depends on its length and, far moreimportantly, on the degree of vessel narrow-ing (i.e., the reduction of r) that it causes.

RL

r= 8

4

ηπ

QPR

= Δ

QP r

L= Δ π

η

4

8

AQ2

10090-06_CH06.qxd 8/31/06 5:22 PM Page 147

Anatomy

The coronary arteries consist of large, prox-imal epicardial segments and smaller, distalresistance vessels. The proximal vessels aresubject to overt atherosclerosis that resultsin stenotic plaques. The distal vessels areusually free of flow-limiting plaques and canadjust their vasomotor tone in response tometabolic needs. These resistance vesselsserve as a reserve, increasing their diameterwith exertion to meet increasing oxygen de-mand and dilating, even at rest, if a proxi-mal stenosis is sufficiently severe.

The hemodynamic significance of a coro-nary artery narrowing depends on both thedegree of stenosis of the epicardial portionof the vessel and the amount of compen-satory vasodilatation the distal resistance ves-sels are able to achieve (Fig. 6.3). If a steno-sis narrows the lumen diameter by less than60%, the maximal potential blood flowthrough the artery is not significantly al-tered and, in response to exertion, the resis-

tance vessels can dilate to provide adequateblood flow. When a stenosis narrows the di-ameter by more than approximately 70%,resting blood flow is normal, but maximalblood flow is reduced even with full dilata-tion of the resistance vessels. In this situa-tion, when oxygen demand increases (e.g.,from the elevated heart rate and force ofcontraction during physical exertion), coro-nary flow reserve is inadequate, oxygen de-mand exceeds supply, and myocardial is-chemia results. If the stenosis compromisesthe vessel lumen by more than approxi-mately 90%, then even with maximal dilata-tion of the resistance vessels, blood flow maybe inadequate to meet basal requirementsand ischemia can develop at rest.

Although collateral channels (see Chap-ter 1) may become apparent between unob-structed coronaries and sites distal to ather-osclerotic stenoses and the flow can bufferthe fall in myocardial oxygen supply, it isoften not sufficient to prevent ischemia dur-ing exertion in critically narrowed vessels.

Endothelial Cell Dysfunction

In addition to fixed vessel narrowing, theother major contributor to reduced myocar-dial oxygen supply in chronic CAD is en-dothelial dysfunction. Abnormal endothelialcell function can contribute to the patho-physiology of ischemia in two ways: (1) byinappropriate vasoconstriction of coronaryarteries and (2) through loss of normal an-tithrombotic properties.

Inappropriate Vasoconstriction

In normal persons, physical activity or men-tal stress results in measurable coronaryartery vasodilatation. This effect is thoughtto be regulated by activation of the sympa-thetic nervous system, with increased bloodflow and shear stress stimulating the releaseof endothelial-derived vasodilators, such asNO. It is postulated that in normal people,the relaxation effect of NO outweighs the di-rect α-adrenergic constrictor effect of cate-cholamines on arterial smooth muscle, suchthat vasodilatation results. However, in pa-

148 Chapter Six

Figure 6.3. Resting and maximal coronary bloodflow are affected by the magnitude of proximal ar-terial stenosis (percent lesion diameter). The dottedline indicates resting blood flow, and the solid line repre-sents maximal blood flow (i.e., when there is full dilata-tion of the distal resistance vessels). Compromise of max-imal blood flow is evident when the proximal stenosisreduces the coronary lumen diameter by more than∼70%. Resting flow may be compromised if the stenosisexceeds ∼90%. (Modified from Gould KL, Lipscomb K. Ef-fects of coronary stenoses on coronary flow reserve andresistance. Am J Cardiol 1974;34:50.)

AQ3

Fig. 3

10090-06_CH06.qxd 8/31/06 5:22 PM Page 148


tients with dysfunctional endothelium (e.g.,atherosclerosis), an impaired release of en-dothelial vasodilators leaves the direct cate-cholamine effect unopposed, such that rela-tive vasoconstriction occurs instead. Theresultant decrease in coronary blood flowcontributes to ischemia. Even the vasodi-latory effect of local metabolites (such asadenosine) is attenuated in patients withdysfunctional endothelium, further uncou-pling the regulation of vascular tone frommetabolic demands.

In patients with risk factors for CAD, suchas hypercholesterolemia, diabetes mellitus,hypertension, and cigarette smoking, im-paired endothelial-dependent vasodilationis noted even before visible atheroscleroticlesions have developed. This suggests thatendothelial dysfunction occurs very early inthe atherosclerotic process.

Inappropriate vasoconstriction also ap-pears to be important in acute coronary syn-dromes, such as unstable angina and myo-cardial infarction (MI). As described inChapter 7, the usual cause of acute coronarysyndromes is disruption of atheroscleroticplaque, with superimposed platelet aggrega-tion and thrombus formation. Normally,the products of platelet aggregation in a de-veloping clot (e.g., serotonin, ADP) result invasodilatation because they stimulate theendothelial release of NO. However, withdysfunctional endothelium, the direct vaso-constricting actions of platelet products pre-dominate (Fig. 6.4), further compromisingflow through the arterial lumen.

Loss of Normal Antithrombotic Properties

In addition to their vasodilatory actions,factors released from endothelial cells (in-cluding NO and prostacyclin) also exert an-tithrombotic properties by interfering withplatelet aggregation (see Fig. 6.4). However,in states of endothelial cell dysfunction, re-lease of these substances is reduced; there-fore, the antithrombotic effect is attenu-ated. Thus, in syndromes characterized bythrombosis (i.e., the acute coronary syn-dromes described in Chapter 7), the im-

paired release of NO and prostacyclin allowsplatelets to aggregate and to secrete their po-tentially harmful procoagulants and vaso-constrictors.

Other Causes of Myocardial Ischemia

In addition to atherosclerotic CAD, otherconditions may result in an imbalance be-tween myocardial oxygen supply and de-mand and result in ischemia. Other causes ofdecreased myocardial oxygen supply include(1) decreased perfusion pressure owing tohypotension (e.g., in a patient with hypo-volemia or septic shock) and (2) a severelydecreased blood oxygen content (e.g., markedanemia or impaired oxygenation of bloodby the lungs). For example, a patient withmassive bleeding from the gastrointestinaltract may develop myocardial ischemiaand angina pectoris, even in the absence ofatherosclerotic coronary disease, becauseof reduced oxygen supply (i.e., the loss ofhemoglobin and hypotension).

On the other side of the balance, a pro-found increase in myocardial oxygen de-mand can cause ischemia even in the ab-sence of coronary atherosclerosis. This canoccur, for example, with rapid tachycardias,profound acute hypertension, or severe aor-tic stenosis.

CONSEQUENCES OF ISCHEMIA

The consequences of ischemia reflect the in-adequate myocardial oxygenation and localaccumulation of metabolic waste products.For example, during ischemia, the myocytesconvert from aerobic to anaerobic meta-bolic pathways. The reduced generation ofATP impairs the interaction of the contrac-tile proteins and results in a transient re-duction of both ventricular systolic contrac-tion and diastolic relaxation (both of whichare energy-dependent processes). The conse-quent elevation of LV diastolic pressure istransmitted (via the left atrium and pulmo-nary veins) to the pulmonary capillaries andcan precipitate pulmonary congestion andthe symptom of dyspnea. In addition, meta-

Fig. 4

10090-06_CH06.qxd 8/31/06 5:22 PM Page 149

bolic products such as lactate, serotonin,and adenosine accumulate locally. It issuspected that one or more of these com-pounds activate peripheral pain receptors inthe C7 through T4 distribution and may be the mechanism by which the discomfort ofangina is produced. The accumulation oflocal metabolites and transient abnormalitiesof myocyte ion transport may also precipitatedangerous arrhythmias (see Chapter 11).

The ultimate fate of myocardium sub-jected to ischemia depends on the severityand duration of the imbalance betweenoxygen supply and demand. It was previ-ously thought that ischemic cardiac injuryresults in either irreversible myocardialnecrosis (i.e., myocardial infarction) or rapidand full recovery of myocyte function (e.g.,after a brief episode of typical angina). It isnow known that in addition to those out-

150 Chapter Six

Figure 6.4. The interaction between platelets and endothelialcells. A. Normal endothelium. Aggregating platelets release thrombox-ane (TXA2) and serotonin (5-HT), the direct vascular effects of whichcause contraction of vascular smooth muscle and vasoconstriction.However, platelet products (e.g., adenosine diphosphate [ADP], 5-HT)also stimulate the endothelial release of the potent vasodilators nitricoxide (NO) and prostacyclin, such that the net effect is smooth musclerelaxation instead. Endothelial production of NO and prostacyclin alsoserves antithrombotic roles, which limit further platelet aggregation. B. Dysfunctional endothelium demonstrates impaired release of the va-sodilator substances, such that net smooth muscle contraction andvasoconstriction supervene. The reduced endothelial release of NO andprostacyclin diminishes their antiplatelet effect, such that thrombosisproceeds unchecked.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 150


comes, ischemic insults can sometimes re-sult in a period of prolonged contractile dys-function without myocyte necrosis, and re-covery of normal function may ultimatelyfollow.

For example, stunned myocardium refersto tissue that, after suffering a period of se-vere ischemia (but not necrosis), demon-strates prolonged systolic dysfunction evenafter the return of normal myocardial bloodflow. In this setting, the functional, bio-chemical, and ultrastructural abnormalitiesfollowing ischemia are reversible and con-tractile function gradually recovers. Themechanism responsible for this delayed re-covery of function involves myocyte cal-cium overload and the accumulation of oxy-gen-derived free radicals during ischemia. Ingeneral, the magnitude of stunning is pro-portional to the degree of the preceding is-chemia, and this state is likely the patho-physiologic response to an ischemic insultthat just falls short of causing irreversiblenecrosis.

In contrast, hibernating myocardiumrefers to tissue that manifests chronic ven-tricular contractile dysfunction in the faceof a persistently reduced blood supply, usu-ally because of multivessel CAD. In this sit-uation, irreversible damage has not occurredand ventricular function can promptly im-prove if appropriate blood flow is restored(e.g., by coronary angioplasty or bypass sur-gery). The importance of the difference be-tween the two types of myocardium is sum-marized in Box 6.2.

Ischemic Syndromes

Depending on the underlying pathophysio-logic process, and the timing and severity ofa myocardial ischemic insult, a spectrum ofdistinct clinical syndromes may result, as il-lustrated in Figure 6.5.

Stable Angina

Chronic stable angina is generally caused byfixed, obstructive atheromatous plaque inone or more coronary arteries (see Fig. 6.5B).The pattern of symptoms is usually related tothe degree of stenosis. As described in the ear-lier section on pathophysiology, when ather-osclerotic stenoses narrow a coronary arterylumen diameter by more than approximately70%, the reduced flow capacity may be suffi-cient to serve the low cardiac oxygen needs atrest but is insufficient to compensate for anysignificant increase in oxygen demand (seeFig. 6.3). During physical exertion, for exam-ple, activation of the sympathetic nervoussystem results in increased heart rate, bloodpressure, and contractility, all of which aug-ment myocardial oxygen consumption. Dur-ing the period that oxygen demand exceedsavailable supply, myocardial ischemia re-sults, often accompanied by the chest dis-comfort of angina pectoris. The ischemia andsymptoms persist until the increased de-mand is alleviated and oxygen balance is restored. A pattern of chronic, predictable,transient angina during exertion or emo-tional stress is termed stable angina.

Box 6.2 Distinguishing Stunned and Hibernating Myocardia

The concepts of stunned and hibernating myocardium are very important in the clinicalsetting. Such regions of myocardia contract poorly when imaged (e.g., by echocardiogra-phy or contrast angiography) and can appear indistinguishable from irreversibly infarctedheart muscle. However, they can be differentiated from necrotic regions by special imag-ing studies (e.g., dobutamine echocardiography, thallium-201 viability study, or positronemission tomography; see Chapter 3). That distinction often influences the decision ofwhether to undertake percutaneous angioplasty/stenting or coronary bypass procedures,because stunned or hibernating myocardium would be expected to improve with me-chanical revascularization, whereas truly infarcted myocardium would not.

Box 2

Fig. 5

10090-06_CH06.qxd 8/31/06 5:22 PM Page 151

Potentially contributing to the inadequateoxygen supply in stable angina is inappro-priate coronary vasoconstriction caused, atleast in part, by atherosclerosis-associatedendothelial dysfunction. Recall that nor-mally, the high myocardial oxygen demandduring exertion is balanced by an increasedsupply of blood as the accumulation of localmetabolites induces vasodilatation. Withendothelial cell dysfunction, however, va-sodilatation is impaired and the vessels mayparadoxically vasoconstrict instead, in re-sponse to exercise-induced catecholaminestimulation of α-adrenergic receptors on thecoronary artery smooth muscle cells.

As a result, the extent of coronary arterynarrowing in patients with atherosclerosis is

not necessarily constant. Rather, it can varyfrom moment to moment because of changesin the superimposed coronary vascular tone.For some patients with stable angina, alter-ations in tone play a minimal role in the de-creased myocardial oxygen supply, and thelevel of physical activity required to precipi-tate angina is fairly constant. These patientshave fixed-threshold angina. In other cases, thedegree of dynamic obstruction caused byvasoconstriction or vasospasm plays a moreprominent role, and such patients may havevariable-threshold angina. For example, on agiven day, a patient with variable-thresholdangina can experience exertion without chestdiscomfort, but on another day, the same de-gree of myocardial oxygen demand does pro-

152 Chapter Six

Figure 6.5. Pathophysiologic findings in anginal syndromes. A. Normal coronary ar-teries are widely patent, and the endothelium functions normally. B. In stable angina, ath-erosclerotic plaque and inappropriate vasoconstriction (caused by dysfunctional endothe-lium) reduce the vessel lumen’s size and coronary blood flow. C. In unstable angina,disruption of the plaque triggers platelet aggregation, thrombus formation, and vasocon-striction, all of which contribute to reduced coronary blood supply. D. In variant angina,atherosclerotic plaques are absent; rather, ischemia is owing to intense vasospasm that re-duces myocardial oxygen supply.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 152


duce symptoms. The difference reflects alter-ations in vascular tone over the sites of fixedstenosis. Other clinical features of chronic sta-ble angina are described in greater detail laterin the chapter.

Unstable Angina

A patient with chronic stable angina may ex-perience a sudden increase in the tempo andduration of ischemic episodes, occurringwith lesser degrees of exertion and even atrest. This acceleration of symptoms is knownas unstable angina, which can be a precursorto an acute MI. Unstable angina and acuteMI are also known as acute coronary syn-dromes and result from distinct pathophysi-ologic mechanisms, most commonly rup-ture of an unstable atherosclerotic plaquewith subsequent platelet aggregation andthrombosis (see Fig. 6.5C). The acute coro-nary syndromes are described in Chapter 7.

Variant Angina

A small minority of patients manifest epi-sodes of focal coronary artery spasm in theabsence of overt atherosclerotic lesions,and this syndrome is known as variant orPrinzmetal angina. In this case, intense va-sospasm alone reduces coronary oxygen sup-ply and results in angina (see Fig. 6.5D). Themechanism by which such profound spasmdevelops is not completely understood butmay involve increased sympathetic activityin combination with endothelial dysfunc-tion. It is thought that many patients withvariant angina may actually have early ath-erosclerosis manifested only by a dysfunc-tional endothelium, because the response toendothelium-dependent vasodilators (e.g.,ACh and serotonin) is often abnormal.

Variant angina often occurs at rest be-cause ischemia in this case results from totransient reduction of the coronary oxygensupply, rather than an increase in myocar-dial oxygen demand.

Silent Ischemia

Episodes of cardiac ischemia sometimes occurin the absence of perceptible discomfort or

pain, and such instances are referred to assilent ischemia. These asymptomatic episodescan occur in patients who on other occasionsexperience typical symptomatic angina. Con-versely, in some patients, silent ischemiamay be the only manifestation of CAD. Itmay be difficult to diagnose silent ischemia,but its presence can be detected by laboratorytechniques such as continuous ambulatoryelectrocardiography, or it can be elicited byexercise stress testing, as described later in thechapter. One study estimated that silent is-chemic episodes occur in 40% of patientswith stable symptomatic angina and in 2.5%to 10% of asymptomatic middle-aged men.When considering the importance of anginaldiscomfort as a physiologic warning signal,the asymptomatic nature of silent ischemiabecomes all the more alarming.

The reason why some episodes of is-chemia are silent whereas others are symp-tomatic has not been elucidated. The degreeof ischemia cannot fully explain the dispar-ity, because even MI may present withoutsymptoms in some patients. However, silentischemia is particularly common among di-abetic patients, suggesting the possibility ofimpaired pain sensation resulting from pe-ripheral neuropathy.

Syndrome X

The term syndrome X refers to patients withtypical symptoms of angina pectoris whohave no evidence of significant atheroscle-rotic coronary stenoses on coronary an-giograms. Some of these patients may showdefinite laboratory signs of ischemia duringexercise testing. The pathogenesis of is-chemia in this situation may be related toinadequate vasodilator reserve of the coro-nary resistance vessels. It is thought that the resistance vessels (which are too small to be visualized by coronary angiography)may not dilate appropriately during periodsof increased myocardial oxygen demand.Microvascular dysfunction, vasospasm, andhypersensitive pain perception may alsocontribute to this syndrome. Patients withsyndrome X have a better prognosis thanthose with overt atherosclerotic disease.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 153

CLINICAL FEATURES OF CHRONICSTABLE ANGINA

History

The most important part of the clinical eval-uation of ischemic heart disease is the pa-tient’s self-described history. Because chestpain is such a common complaint, it is im-portant to focus on the characteristics thathelp distinguish myocardial ischemia fromother causes of discomfort. From a diagnosticstandpoint, it would be ideal to interview andexamine a patient during an actual episode ofangina, but most people are asymptomaticduring routine office or clinic examinations.Therefore, a careful history probing severalfeatures of the discomfort should be elicited.

Quality

Angina is most often described as a “pres-sure,” “discomfort,” “tightness,” “burning,”or “heaviness” in the chest. It is rare that thesensation is actually described as a “pain,”and often a patient will correct the physi-cian who refers to the anginal symptom assuch. Occasionally, a patient likens the sen-sation to “an elephant sitting on my chest.”Anginal discomfort is neither sharp norstabbing, and it does not vary significantlywith inspiration or movement of the chestwall. It is a steady discomfort that lasts a fewminutes, yet rarely more than 5 to 10 min-utes. It always lasts more than a few seconds,and this helps to differentiate it from sharperand briefer musculoskeletal pains.

While describing angina, the patient mayplace a clenched fist over his or her sternum,referred to as the Levine sign, as if definingthe constricting discomfort by that tight grip.

Location

Anginal discomfort is usually diffuse ratherthan localized to a single point. It is mostoften located in the retrosternal area or inthe left precordium but may occur any-where in the chest, back, arms, neck, lowerface, or upper abdomen. It often radiates tothe shoulders and inner aspect of the arms,especially on the left side.

Accompanying Symptoms

During the discomfort of an acute anginalattack, generalized sympathetic and para-sympathetic stimulation may result in tachy-cardia, diaphoresis, and nausea. Ischemia alsoresults in transient dysfunction of LV sys-tolic contraction and diastolic relaxation.The resultant elevation of LV diastolic pres-sure is transmitted to the pulmonary vascu-lature and often causes shortness of breath(dyspnea) during the episode. Transient fa-tigue and weakness are also common, partic-ularly in elderly patients.

Precipitants

Angina, when not caused by pure vasospasm,is precipitated by conditions that increasemyocardial oxygen demand (e.g., increasedheart rate, contractility, or wall stress). Theseinclude physical exertion, anger, and otheremotional excitement. Additional factorsthat increase myocardial oxygen demandthat can precipitate anginal discomfort in pa-tients with CAD include a large meal or coldweather. The latter induces peripheral vaso-constriction, which in turn augments myo-cardial wall stress as the left ventricle con-tracts against the increased resistance.

Angina is generally relieved within min-utes after the cessation of the activity thatprecipitated it and even more quickly (within3 to 5 minutes) by sublingual nitroglycerin.This response can help differentiate myocar-dial ischemia from many of the other condi-tions that produce chest discomfort.

Patients who experience angina primarilyowing to increased coronary artery tone orvasospasm often develop symptoms at rest,independent of activities that increase myo-cardial oxygen demand.

Frequency

Although the level of exertion necessary toprecipitate angina may remain fairly con-stant, the frequency of episodes varies con-siderably because patients quickly learnwhich activities cause their discomfort andavoid them. It is thus important to inquire

154 Chapter Six

10090-06_CH06.qxd 8/31/06 5:22 PM Page 154


about reductions in activities of daily livingwhen taking the history.

Risk Factors

In addition to the description of chest dis-comfort, a careful history should uncover riskfactors that predispose to atherosclerosis andCAD, including cigarette smoking, hyper-cholesterolemia, hypertension, diabetes, anda family history of premature coronary dis-ease (see Chapter 5).

Differential Diagnosis

Several conditions can mimic angina pec-toris, including gastroesophageal reflux, eso-phageal spasm, biliary pain, pericarditis, andmusculoskeletal conditions such as chest wall

pain, spinal osteoarthritis, and cervical radi-culitis. The history remains of paramount im-portance in distinguishing myocardial is-chemia from these disorders. In contrast toangina pectoris, gastrointestinal causes of re-current chest pain are often precipitated bycertain foods and are unrelated to exertion.Musculoskeletal causes of chest discomforttend to be more superficial or can be localizedto a discrete spot (i.e., the patient can point tothe pain with one finger) and often vary withchanges in position. Similarly, the presence ofpleuritic pain (sharp pain aggravated by res-piratory movements) argues against angina asthe cause; this symptom is more likely a resultof pericarditis (or pulmonary conditions suchas pleuritis, pneumonia, or pulmonary em-bolism). Useful differentiating features arelisted in Table 6.2.

TABLE 6.2. Causes of Recurrent Chest Pain

Condition Differentiating Features

CardiacMyocardial ischemia

Pericarditis

GastrointestinalGastroesophageal reflux

Peptic ulcer disease

Esophageal spasm

Biliary colic

MusculoskeletalCostochondral syndrome

Cervical radiculitis

ECG, electrocardiogram.

• Retrosternal tightness or pressure; typically radiates to neck, jaw, or leftshoulder and arm

• Lasts a few minutes (usually <10)• Brought on by exertion, relieved by rest• Relieved by nitroglycerin• ECG: transient ST depressions or elevations, or flattened or inverted T waves• Sharp, pleuritic pain that varies with position; friction rub on auscultation• Can last for hours to days• ECG: diffuse ST elevations and PR depression

• Retrosternal burning• Precipitated by certain foods, worsened by supine position, unaffected by

exertion• Relieved by antacids, not by nitroglycerin• Epigastric ache or burning• Occurs after meals, unaffected by exertion• Relieved by antacids, not by nitroglycerin• Retrosternal pain accompanied by dysphagia• Precipitated by meals, unaffected by exertion• May be relieved by nitroglycerin• Constant, deep pain in right upper quadrant; can last hours• Brought on by fatty foods, unaffected by exertion• Not relieved by antacids or nitroglycerin

• Sternal pain worsened by chest movement• Costochondral junctions tender to palpation• Relieved by anti-inflammatory drugs, not by nitroglycerin• Constant ache or shooting pains, may be in a dermatomal distribution• Worsened by neck motion

Tab. 2

10090-06_CH06.qxd 8/31/06 5:22 PM Page 155

Physical Examination

If it is possible to examine a patient during ananginal attack, several transient physicalsigns may be detected (Fig. 6.6). An increasedheart rate and blood pressure are commonbecause of the augmented sympathetic re-sponse. Myocardial ischemia may lead topapillary muscle dysfunction and thereforemitral regurgitation. Ischemia-induced re-gional ventricular contractile abnormalitiescan sometimes be detected as an abnormalbulging impulse on palpation of the leftchest. Ischemia decreases ventricular com-pliance, producing a stiffened ventricle andtherefore an S4 gallop on physical examina-tion during atrial contraction (see Chapter2). However, if the patient is free of chest dis-comfort during the examination, there maybe no abnormal cardiac physical findings.

Physical examination should also assessfor signs of atherosclerotic disease in more ac-cessible vascular beds. For example, carotidbruits may indicate the presence of cere-brovascular disease, whereas femoral arterybruits or diminished pulses in the lower ex-tremities can be a clue to peripheral arterialdisease (see Chapter 15).

Diagnostic Studies

Once angina is suspected, several diagnosticprocedures may be helpful in confirming

myocardial ischemia as the cause. Becausemany of these tests are costly, it is importantto choose the appropriate studies for eachpatient.

Electrocardiogram

One of the most useful tools is an electro-cardiogram (ECG) obtained during an angi-nal episode. Although this is easy to arrangewhen symptoms occur in hospitalized pa-tients, it may not be possible to “catch”episodes in people seen on an outpatientbasis. During myocardial ischemia, ST seg-ment and T wave changes often appear (Fig.6.7). Acute ischemia usually results in tran-sient horizontal or downsloping ST segmentdepressions and T wave flattening or inver-sions. Occasionally, ST segment elevationsare seen, suggesting more severe transmuralmyocardial ischemia, and can be observedduring the intense vasospasm of variantangina. In contrast to the ECG of a patientwith an acute MI, the ST deviations seen inpatients with stable angina quickly normal-ize with resolution of the patient’s symp-toms. In fact, ECGs obtained during periodsfree of ischemia are completely normal inapproximately half of patients with stableangina. In others, chronic “nondiagnostic”ST and T wave deviations may be present.Evidence of a previous MI (e.g., pathologic

156 Chapter Six

Figure 6.6. Pathophysiology of physical signs during acute myocardial ischemia.

Fig. 6

Fig. 7

10090-06_CH06.qxd 8/31/06 5:22 PM Page 156


Q waves) on the ECG also points to the pres-ence of underlying coronary disease.

Stress Testing

Because ECGs obtained during or betweenepisodes of chest discomfort may be normal,such tracings do not rule out underlying ischemic heart disease. For this reason, pro-vocative exercise or pharmacologic stresstests are valuable diagnostic and prognosticaids.

Standard Exercise Testing

For many patients suspected of having coro-nary artery disease, a standard exercise testis performed. During this test, the patientexercises on a treadmill or a stationary bicy-cle to progressively higher workloads and isobserved for the development of chest dis-comfort or inordinate dyspnea. The heartrate and ECG are continuously monitored,and blood pressure is checked at regular in-tervals. The test is continued until anginadevelops, signs of myocardial ischemia ap-pear on the ECG, a target heart rate isachieved (85% of the maximal predictedheart rate, which is calculated as 220 minusthe patient’s age), or the patient becomestoo fatigued to continue.

The test is considered positive if the pa-tient’s typical chest discomfort is repro-duced or if ECG abnormalities consistent

with ischemia develop (i.e., >1 mm hori-zontal or downsloping ST segment depres-sions). Among patients who later undergodiagnostic coronary angiography, standardexercise testing has a sensitivity of approxi-mately 65% to 70% and specificity of 75% to80% for the detection of anatomically sig-nificant coronary artery disease.

The stress test is considered markedly pos-itive if one or more of the following signs ofsevere ischemic heart disease occur: (1) is-chemic ECG changes develop in the first 3 minutes of exercise or persist 5 minutesafter exercise has stopped; (2) the magnitudeof the ST segment depressions is >2 mm; (3) the systolic blood pressure decreases dur-ing exercise (i.e., resulting from ischemia-induced impairment of contractile function);(4) high-grade ventricular arrhythmias de-velop; or (5) the patient cannot exercise forat least 2 minutes because of cardiopul-monary limitations. Patients with markedlypositive tests are more likely to have severemultivessel coronary disease.

The utility of a stress test may be affectedby the patient’s medications. For example,β-blockers or certain calcium channel block-ers may blunt the ability to achieve the tar-get heart rate. In these situations, one mustconsider the purpose of the stress test. If it isto determine whether ischemic heart diseaseis present, then those medications are typi-cally withheld for 24 to 48 hours before thetest. On the other hand, if the patient has

Figure 6.7. Common transient ECG abnormalities during ischemia. Subendocardial ischemiacauses ST segment depressions and/or T wave flattening or inversions. Severe transient transmuralischemia can result in ST segment elevations, similar to the early changes in acute myocardial infarc-tion. When transient ischemia resolves, so do the electrocardiographic changes.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 157

known ischemic heart disease and the pur-pose of the test is to assess the efficacy of thecurrent medical regimen, testing should beperformed while the patient takes his or herusual antianginal medications.

Nuclear Imaging Studies

Because the standard exercise test relies onischemia-related changes on the ECG, thetest is less useful in patients with baselineabnormalities of the ST segments (e.g., asseen in left bundle branch block or LV hy-pertrophy). In addition, the standard exer-cise stress test sometimes yields equivocalresults in patients for whom the clinical sus-picion of ischemic heart disease is high. Inthese situations, radionuclide imaging canbe combined with exercise testing to over-come these limitations and to increase thesensitivity and specificity of the study.

During myocardial perfusion imaging(see Chapter 3), a radionuclide (commonlyeither a technetium-99m–labeled compoundor thallium-201) is injected intravenously atpeak exercise, after which imaging is per-formed. The radionuclide accumulates inproportion to the degree of perfusion of vi-able myocardial cells. Therefore, areas ofpoor perfusion (i.e., regions of ischemia)during exercise do not accumulate ra-dionuclide and appear as “cold spots” onthe image. However, irreversibly infarctedareas also do not take up the radionuclide,and they too will appear as cold spots. Todifferentiate between transient ischemiaand infarcted tissue, repeat imaging is per-formed several hours later. If the cold spotfills in, a region of transient ischemia hasbeen identified. If the cold spot remainsunchanged, a region of irreversible infarc-tion is likely.

Radionuclide exercise tests are 80% to90% sensitive and 80% to 90% specific forthe presence of clinically significant CAD.Because these techniques are expensive,their use in screening for CAD should be re-served for (1) patients in whom baselineECG abnormalities preclude interpretationof a standard exercise test or (2) improve-ment in test sensitivity when standard stress

test results are discordant with the clinicalsuspicion of coronary disease.

As described in Chapter 3, positron emis-sion tomography is a specialized nuclearimaging technique used to assess myocar-dial perfusion and cellular viability. It is notuniversally available but is particularly ca-pable of distinguishing between regions ofischemia, infarction, and hibernating myo-cardium.

Exercise Echocardiography

At many centers, exercise testing with echo-cardiographic imaging is another diagnostictechnique used to diagnose myocardial is-chemia in patients with baseline ST or T waveabnormalities or in those with equivocalstandard stress tests. In this procedure, LVcontractile function is assessed by echocar-diography at baseline and immediately aftertreadmill or bicycle exercise. The test indi-cates inducible myocardial ischemia if re-gions of ventricular contractile dysfunctiondevelop with exertion.

Pharmacologic Stress Tests

For patients unable to exercise (e.g., thosewith hip or knee arthritis), pharmacologicstress testing can be performed using variousagents, including the inotrope dobutamine(which increases myocardial oxygen demandby stimulating the heart rate and force ofcontraction) or the vasodilators dipyridamoleor adenosine. Dipyridamole blocks the cellu-lar uptake and destruction of adenosine andthereby increases its circulating concentra-tion. When adenosine binds to its vascularreceptors, coronary vasodilatation results.As ischemic regions are already maximallydilated (in compensation for the epicardialcoronary stenoses), the drug-induced vaso-dilatation increases flow to the myocardiumperfused by healthy coronary arteries andthus “steals” blood away from the diseasedsegments. As a result, nuclear imaging (usingthallium-201 or technetium-99m–labeledcompounds) performed right after adeno-sine or dipyridamole administration displaysischemic myocardium as regions of relatively

158 Chapter Six

AQ4

10090-06_CH06.qxd 8/31/06 5:22 PM Page 158


decreased perfusion (cold spots). Alterna-tively, pharmacologic stress testing can beperformed with echocardiography in placeof nuclear imaging.

Coronary Angiography

The most direct means of identifying coro-nary artery stenoses is by coronary angiog-raphy, in which atherosclerotic lesions arevisualized radiographically following theinjection of radiopaque contrast materialinto the artery (Fig. 6.8; see Chapter 3). Although generally safe, the procedure is as-sociated with a small risk of serious compli-cations directly related to its invasive na-ture. Therefore, coronary angiography istypically reserved for patients whose angi-nal symptoms do not respond adequatelyto pharmacologic therapy, for those withan unstable presentation, or when the re-sults of noninvasive testing are so abnormalthat severe CAD warranting revasculariza-tion is likely.

Although coronary angiography is con-sidered the “gold standard” for the diagno-sis of CAD, it should be noted that itprovides only anatomic information. Theclinical significance of lesions detected byangiography depends not only on the de-gree of narrowing but also on the patho-physiologic consequences. Therefore, treat-ment decisions are made not only on thefinding of such stenoses but even more soon their functional effects, manifested bythe patient’s symptoms, the viability of themyocardium segment served by stenoticvessels, and the degree of ventricular con-tractile dysfunction. Furthermore, standardarteriography does not reveal the composi-tion of coronary atherosclerotic plaque or itsvulnerability to rupture (see Chapter 5).

Newer noninvasive techniques that visual-ize coronary anatomy in great detail (e.g.,multidetector CT scanning; see Chapter 3)are under study to determine their properrole in diagnosis and therapeutic planningin patients with suspected CAD.

Natural History

The patient with chronic angina may showno change in a stable pattern of ischemia formany years. In some patients, however, thecourse may be punctuated at any time bythe occurrence of unstable angina, MI, orsudden cardiac death. These complicationsare often related to acute thrombosis at thesite of disrupted atherosclerotic plaque (seeChapter 7). Why some patients, but not oth-ers, sustain these complications remains asubject of intense clinical and basic scienceinvestigation and may relate to the vulnera-bility of plaque to rupture.

Before the current era of sophisticatedpharmacotherapy, coronary angioplasty, andsurgical revascularization procedures, stud-ies showed that the annual mortality rate of patients with CAD corresponded to thenumber of vessels containing significantstenoses. For example, patients with ad-vanced stenoses within a single coronaryvessel could expect an annual mortality rateof <4%. Those with two involved vessels hadan annual mortality rate of 7% to 10%, and

Figure 6.8. Example of coronary angiography. Injec-tion of the right coronary artery demonstrates a stenosisin the midportion of the vessel, indicated by the arrow.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 159

those with advanced, three-vessel diseaseshowed a 10% to 12% mortality rate. If theleft main artery was significantly stenosed,the mortality rate was greatly increased(15% to 25%). These outcomes were worsein corresponding patients with subnormalventricular contractile function.

More recent studies have confirmed thatthe location and extent of coronary stenosesare important but also that other criticalpredictors of mortality include (1) the ex-tent of impaired LV contractile function, (2) poor exercise capacity, and (3) the mag-nitude of clinical anginal symptoms. Thesepredictors are taken into account when con-templating treatment decisions.

The mortality associated with CAD hasdeclined significantly in recent decades: theage-adjusted death rate has fallen by morethan 50%. This is likely related to (1) athero-sclerotic risk reduction through improvedlifestyle changes (e.g., less tobacco use, lessdietary fat consumption, and more exercise);(2) improved therapeutic strategies and lon-gevity following acute MI (see Chapter 7);and (3) advances in the pharmacologic andmechanical therapies for chronic CAD.

TREATMENT

The goals of therapy in chronic ischemicheart disease are to decrease the frequency ofanginal attacks, to prevent acute coronarysyndromes such as myocardial infarction,and to prolong survival. A long-term crucialstep is to address the risk factors that led tothe development of atherosclerotic coronarydisease. Data convincingly demonstrate thebenefit of smoking cessation, cholesterol re-duction, blood pressure control, and serumglucose control in lowering the risk of coro-nary disease events (see Chapter 5). Im-provements in other risk factors for CAD, in-cluding obesity and physical inactivity, arealso likely to reduce the risk of adverse out-comes, although the benefits of these inter-ventions are less well documented.

The following sections describe medicaland surgical strategies to (1) reduce is-chemia and its symptoms by restoring thebalance between myocardial oxygen supply

and demand, and (2) prevent acute coro-nary syndromes and death in patients withchronic CAD.

Medical Treatment of an AcuteEpisode of Angina

When experiencing angina, the patientshould cease physical activity. Sublingualnitroglycerin, an organic nitrate, is the drugof choice in this situation. Placed under thetongue, this medication produces a slightburning sensation as it is absorbed throughthe mucosa, and it begins to take effect in 1to 2 minutes. Nitrates relieve ischemia pri-marily through vascular smooth muscle re-laxation, particularly venodilatation. Venodi-latation reduces venous return to the heart,with a subsequent decline in LV volume (adeterminant of wall stress). The latter de-creases myocardial oxygen consumption,thus helping to restore oxygen balance inthe ischemic heart.

A second action of nitrates is to dilate thecoronary vasculature, with subsequent aug-mentation of coronary blood flow. This ef-fect may be of minimal value in patientswith angina in whom maximal coronary di-latation has already resulted from the accu-mulation of local metabolites. However,when coronary vasospasm plays a role in thedevelopment of ischemia, nitrate-inducedcoronary vasodilatation may be particularlybeneficial.

Medical Treatment to PreventRecurrent Ischemic Episodes

Pharmacologic agents are also the first lineof defense in the prevention of anginal at-tacks. The goal of these agents is to decreasethe cardiac workload (i.e., reduce myocar-dial oxygen demand) and to increase myo-cardial perfusion. The three classes of med-ications commonly used are the organicnitrates, β-adrenergic blockers, and calciumchannel blockers (Table 6.3).

Organic nitrates (e.g., nitroglycerin, iso-sorbide dinitrate, isosorbide mononitrate), aspreviously mentioned, relieve ischemia pri-marily through venodilatation (i.e., lower

160 Chapter Six

Tab. 3

10090-06_CH06.qxd 8/31/06 5:22 PM Page 160


wall stress results from a smaller ventricularradius) and possibly through coronary va-sodilatation. The organic nitrates are theoldest of the antianginal drugs and come inseveral preparations. Sublingual nitroglyc-erin tablets or sprays are used in the treat-ment of acute attacks because of their rapidonset of action. In addition, when taken im-mediately before a person engages in activi-ties known to provoke angina, these rapidlyacting nitrates are useful as prophylaxis againstanginal attacks.

Longer-acting anginal prevention can beachieved through a variety of nitrate prepa-rations, including oral tablets of isosorbidedinitrate (or mononitrate) or a transdermalnitroglycerin patch, which is applied once aday. A limitation to chronic nitrate therapyis the development of drug tolerance (i.e.,decreased effectiveness of the drug duringcontinued administration), which occurs tosome degree in most patients. This unde-sired effect can be overcome by providing anitrate-free interval for several hours eachday, usually while the patient sleeps.

There is no evidence that nitrates improvesurvival or prevent infarctions in patientswith chronic CAD, and they are used purelyfor symptomatic relief. Common side effectsinclude headache, lightheadedness, and pal-pitations induced by reflex sinus tachycar-dia. The latter can be prevented by combin-ing a β-blocker with the nitrate regimen.

�-Blockers (see Chapter 17) exert theirantianginal effect primarily by reducing myo-cardial oxygen demand. They are directedagainst β-receptors, of which there are twoclasses: β1-adrenergic receptors are restrictedto the myocardium, whereas β2-adrenergic re-ceptors are located throughout blood vesselsand the bronchial tree. The stimulation of β1-receptors by endogenous catecholaminesand exogenous sympathomimetic drugs in-creases heart rate and contractility. Conse-quently, β-adrenergic antagonists decreasethe force of ventricular contraction and heartrate, thereby relieving ischemia by reducingmyocardial oxygen demand. In addition,slowing the heart rate may benefit myo-cardial oxygen supply by augmenting the

TABLE 6.3. Pharmacologic Agents in the Treatment of Angina

Drug Class Mechanism of Action Adverse Effects

Organic nitrates ↓ Myocardial O2 demand • Headache↓ Preload (venodilatation) • Hypotension

• Reflex tachycardia↑ O2 supply↑ Coronary perfusion↓ Coronary vasospasm

β-blockers ↓ Myocardial O2 demand • Excessive bradycardia↓ Contractility • ↓ LV contractile function↓ Heart rate • Bronchoconstriction

• May worsen diabetic control• Fatigue

Calcium channel blockers ↓ Myocardial O2 demand • Headache, flushing(agent specific; see footnote)

↓ Preload (venodilatation) ↓ LV contraction (V, D)↓ Wall stress (↓BP) • Marked bradycardia (V, D)↓ Contractility (V, D) • Edema (especially N, D)↓ Heart rate (V, D) • Constipation (especially V)↑ O2 supply↑ Coronary perfusion↓ Coronary vasospasm

Aspirin ↓ Platelet aggregation • Gastrointestinal irritation or bleeding

BP, blood pressure; D, diltiazem; LV, left ventricular; N, nifedipine and other dihydropyridine calcium++ channel antagonists;V, verapamil.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 161

time spent in diastole, the phase when coro-nary perfusion primarily occurs.

In addition to suppressing angina, severalstudies have shown that β-blockers decreasethe rates of recurrent infarction and mortal-ity following an acute MI (see Chapter 7).Moreover, they have been shown to reducethe likelihood of an initial MI in patientswith hypertension. Thus, β-blockers arefirst-line therapy in the treatment of CAD.

β-Blockers are generally well tolerated buthave several potential side effects. For exam-ple, they may precipitate bronchospasm inpatients with underlying asthma by antago-nizing β2-receptors in the bronchial tree. Al-though β1-selective blockers are theoreticallyless likely to exacerbate bronchospasm insuch patients, drug selectivity for the β1-receptor is not complete, and in general,all β-blockers should be avoided in patientswith significant obstructive airway disease.

β-Blockers are also generally not used inpatients with acutely decompensated LV dys-function because they could intensify heartfailure symptoms by further reducing in-otropy. (However, as described in Chapter 9,β-blockers actually improve outcomes in pa-tients with stable heart failure conditions.) β-Blockers are also relatively contraindicatedin patients with marked bradycardia or cer-tain types of heart block to avoid additionalimpairment of electrical conduction.

β-Blockers sometimes cause fatigue andsexual dysfunction. They should be used withcaution in insulin-treated diabetic patientsbecause they can mask tachycardia and othercatecholamine-mediated responses that canwarn of hypoglycemia. One might also ex-pect that β-blockers would decrease myocar-dial blood perfusion by blocking the vasodi-lating β2-adrenergic receptors on the coronaryarteries. However, this effect is usually atten-uated by autoregulation and vasodilation ofthe coronary vessels owing to the accumula-tion of local metabolites.

Calcium channel blockers (see Chapter17) antagonize voltage-gated L-type calciumchannels, but the actions of the individualdrugs of this group vary. The dihydropy-ridines (e.g., nifedipine and amlodipine) arepotent vasodilators. They relieve myocardial

ischemia by (1) decreasing oxygen demand(venodilatation reduces ventricular fillingand size, arterial dilatation reduces the resis-tance against which the left ventricle con-tracts, and both actions reduce wall stress);and (2) increasing myocardial oxygen sup-ply via coronary dilatation. By the lattermechanism, they are also potent agents forthe relief of coronary artery vasospasm.

Nondihydropyridine calcium channelblockers (verapamil and diltiazem) also actas vasodilators but are not as potent in thisregard as the dihydropyridines. However,these agents have additional beneficial antianginal effects stemming from theirmore potent cardiac depressant actions:they reduce the force of ventricular con-traction (inotropy) and slow the heart rate.Accordingly, verapamil and diltiazem alsodecrease myocardial oxygen demand bythese mechanisms.

Questions have been raised about thesafety of short-acting calcium channel–block-ing drugs in the treatment of ischemic heartdisease. In meta-analyses of randomized tri-als, these drugs have been associated withan increased incidence of MI and mortality.The adverse effect may relate to the rapidhemodynamic effects and blood pressureswings induced by the short-acting agents.Therefore, only long-acting calcium channelblockers are recommended in the treatmentof chronic angina, generally as second-linedrugs if symptoms are not controlled by β-blockers and nitrates.

The three standard groups of antianginaldrugs described in this section can be usedalone or in combination. However, careshould be taken in combining a β-blockerwith a nondihydropyridine calcium chan-nel blocker (verapamil or diltiazem) becausethe additive negative chronotropic effectcan cause excessive bradycardia, and thecombined negative inotropic effect couldprecipitate heart failure in patients with LVcontractile dysfunction.

In 2006, the Food and Drug Administra-tion approved ranolazine, a fourth type ofanti-ischemic therapy. This medication hasbeen shown to decrease the frequency ofanginal episodes and improve exercise ca-

162 Chapter Six

10090-06_CH06.qxd 8/31/06 5:22 PM Page 162


pacity in patients with chronic CAD but dif-fers from other anti-ischemic drugs in that itdoes not affect the heart rate or blood pres-sure. Although its mechanism of action hasnot been fully elucidated, it is believed to in-hibit the late phase of the action potential’sinward sodium current (INa+) in ventricularmyocytes. That late phase tends to be abnor-mally enhanced in ischemic myocardium,and the associated increased sodium influxresults in higher-than-normal intracellularCa++ (mediated by the transsarcolemmalNa+-Ca++ exchanger; see Fig. 1.10). This cal-cium overload is thought to result in im-paired diastolic relaxation (i.e., diastolic dys-function; see Chapter 9) and contractileinefficiency. Inhibition of the late INa+ by ra-nolazine counters these pathologic effects.Pending further studies and experience, ra-nolazine is currently approved only for pa-tients who have not responded adequatelyto the standard antianginal drugs describedearlier.

Although useful in controlling symptomsof angina, none of the antianginal druggroups has been shown to slow or reversethe atherosclerotic process responsible forthe arterial lesions of chronic CAD. More-over, although β-blockers have demon-strated mortality benefits in patients afterMI, none of has been shown to improvelongevity in patients with chronic stableangina and preserved LV function.

Medical Treatment to PreventAcute Cardiac Events

Platelet aggregation and thrombosis are keyelements in the pathophysiology of acuteMI and unstable angina (see Chapter 7). Anti-platelet therapy reduces the risk of theseacute coronary syndromes in patients withchronic angina and should be a standardpart of the regime used to treat CAD. Aspirinhas antithrombotic actions through inhibi-tion of platelet aggregation (and thereforereduces the release of platelet-derived pro-coagulants and vasoconstrictors) as well asanti-inflammatory properties that may be im-portant in stabilizing atheromatous plaque.Unless contraindications are present (e.g.,

allergy or gastric bleeding), aspirin shouldbe continued indefinitely in all patientswith CAD.

Thienopyridines, including clopidogrel,compose another group of antiplateletagents. As described in Chapter 17, they ir-reversibly bind to the platelet ADP receptorP2Y12, thereby preventing platelet activationand aggregation. When compared with as-pirin in clinical trials of patients with stableatherosclerotic disease, clopidogrel resultedin only a modest reduction of cardiovascu-lar events. However, the combination of as-pirin and clopidogrel is superior to aspirinalone in reducing death and ischemic com-plications in patients with acute coronarysyndromes and in those undergoing electivepercutaneous coronary stenting.

Lipid-regulating therapy also reducescardiovascular clinical events. In particular,HMG-CoA reductase inhibitors (statins; seeChapter 17) lower MI and death rates in pa-tients with established coronary disease andin those at high risk of developing CAD. Thebenefits of statin therapy are believed to ex-tend beyond their lipid-altering effects, be-cause there is evidence that they decreasevascular inflammation and improve endo-thelial cell dysfunction and thus may helpstabilize atherosclerotic plaques. All patientswith CAD should have their LDL cholesterolmaintained at <100 mg/dL. Moreover, sometrials of patients with established athero-sclerotic disease have demonstrated that in-tensive lipid lowering (with high-dose statintherapy) is superior to moderate lipid-lower-ing therapy in preventing ischemic eventsand cardiovascular death. As a result, cur-rent national guidelines include an optionalgoal of LDL <70 mg/dL for patients withknown CAD, especially those at highest risk(e.g., following an acute coronary syndromeor those with multiple major risk factors, es-pecially diabetes or continued smoking).

Angiotensin-converting enzyme (ACE)inhibitors, known to be beneficial in thetreatment of hypertension (see Chapter 13),heart failure (see Chapter 9), and followingmyocardial infarction (see Chapter 7), havebeen studied more recently as chronic ther-apy for patients with stable CAD not com-

10090-06_CH06.qxd 8/31/06 5:22 PM Page 163

plicated by heart failure. Some (but not all)of these trials have shown reduced rates ofdeath, myocardial infarction, and stroke.Thus, many cardiologists recommend thatan ACE inhibitor be included in the medicalregimen of patients with chronic CAD.

Revascularization

Patients with angina that becomes asymp-tomatic during pharmacologic therapy areusually followed by their physicians withcontinued emphasis on reducing cardiacrisk factors. However, more aggressive coro-nary revascularization is pursued if (1) thepatient’s symptoms of angina do not re-spond adequately to antianginal drug ther-apy, (2) unacceptable side effects of medica-tions occur, or (3) the patient is found tohave high-risk coronary disease for whichrevascularization is known to improve sur-vival (as described in the next section). Thetwo techniques used to accomplish me-chanical revascularization are percutaneouscoronary intervention and coronary arterybypass graft surgery.

Percutaneous coronary interventions(PCIs) include percutaneous transluminalcoronary angioplasty (PTCA), a procedureperformed under fluoroscopy in which aballoon-tipped catheter is inserted through aperipheral artery (usually, femoral, brachial,or radial) and maneuvered into the stenoticsegment of a coronary vessel. The balloon atthe end of the catheter is then inflatedunder high pressure to dilate the stenosis,after which the balloon is deflated and thecatheter is removed from the body. The im-provement in the size of the coronary lumenincreases coronary perfusion and myocar-dial oxygen supply. Effective dilatation ofthe stenosis results from compression of theatherosclerotic plaque and often by creatinga fracture within the lesion and stretchingthe underlying media. Many types of coro-nary stenoses are amenable to balloon dila-tion, and complications are infrequent. Therisk of MI during the procedure is less than1.5%, and mortality is less than 1%. Unfor-tunately, approximately one third of patientswho undergo standard PTCA develop recur-

rent symptoms within 6 months owing torestenosis of the dilated artery and requireadditional coronary interventions. In rigor-ous angiographic studies, the incidence ofrestenosis after PTCA has been found to beeven greater—as high as 50%.

Fortunately, advances in percutaneoustechniques have included the developmentof coronary stents that can be placed at thetime of PCI and significantly reduce the rateof restenosis. Coronary stents are slender,cagelike stainless-steel support devices that intheir collapsed configuration can be threadedinto the region of stenosis by a catheter. Oncein position, the stent is expanded into itsopen position by inflating a high-pressureballoon in its interior (Fig. 6.9). The balloonand attached catheter are then removed, butthe stent is left permanently in place to serveas a scaffold to maintain arterial patency. Be-cause stents are thrombogenic, a combina-tion of oral antiplatelet agents (typically, as-

164 Chapter Six

Artery wall

Ballooncatheter

Stent incollapsedconfiguration

StenosisA

Balloon inflation to expand stent

B

C

Figure 6.9. Placement of a coronary artery stent. A. A stent, in its original collapsed state, is advanced intothe coronary stenosis on a balloon catheter. B. The bal-loon is inflated to expand the stent. C. The balloon is de-flated and the catheter is removed from the body, leavingthe stent permanently in place. AQ5

Fig. 9

10090-06_CH06.qxd 8/31/06 5:22 PM Page 164


pirin plus clopidogrel) is crucial after stentimplantation.

Compared with conventional PTCA, stentimplantation results in larger luminal diam-eters, decreased restenosis rates, and reducedneed for repeat angioplasty procedures. Al-though restenosis resulting from vessel elas-tic recoil is greatly diminished by standardmetal stent placement, neointimal prolifera-tion (i.e., migration of smooth muscle cellsand production of extracellular matrix) re-mains an important cause of in-stent resteno-sis and recurrent anginal symptoms.

To address the problem of restenosis afterPCI, drug-eluting stents have been devised.These special stents are fabricated with apolymer coat that incorporates an antipro-liferative medication, such as sirolimus (animmunosuppressive agent that inhibits T-cell activation) or paclitaxel (which inter-feres with cellular microtubule function).The medication is released from the stentover a period of 2 to 4 weeks, and this ap-proach has shown great effect at preventingneointimal proliferation and clinical resteno-sis. Prolonged courses of combination an-tiplatelet therapy (aspirin plus clopidogrelfor several months, followed by aspirin in-definitely) are necessary for patients who re-ceive drug-eluting stents, to prevent latethrombosis, a complication that has occurredon occasion in patients who stop taking an-tiplatelet medications.

Although percutaneous revascularizationtechniques are generally superior to stan-dard medical therapy for relief of angina, itis important to note that, in the setting ofstable coronary disease, they have not beenshown to reduce the risk of MI or death.

Coronary artery bypass graft (CABG)surgery entails grafting portions of a pa-tient’s native blood vessels to bypass ob-structed coronary arteries. Two types of sur-gical grafts are used (Fig. 6.10). The firstemploys native veins—typically, a section ofthe saphenous vein (a “superfluous” vesselremoved from the leg) that is sutured fromthe base of the aorta to a coronary segmentdownstream from the region of stenosis. Thesecond method uses arterial grafts—mostcommonly, an internal mammary artery

(IMA, a “superfluous” branch of each sub-clavian artery)—that can be directly anasto-mosed distal to a stenotic coronary site.Vein grafts have a patency rate of up to 80%at 12 months but are vulnerable to acceler-ated atherosclerosis; 10 years after surgery,more than 50% have occluded. In contrast,IMA grafts are more resistant to atheroscle-rosis with a patency rate of 90% at 10 years.Therefore, IMA grafts are often used to per-fuse sites of critical flow such as the left an-terior descending artery. Recent evidencesupports the use of aggressive lipid-loweringdrug therapy after CABG to improve thelong-term patency rates of bypass grafts.

In recent years, less invasive surgical alter-natives to conventional CABG have been ex-plored. These include “minimally invasive”operations with smaller incisions, the use oftranscutaneous ports with videoscopic ro-botic assistance, and “off-pump” procedures,which avoid the use of cardiopulmonary by-pass (heart-lung) machines. Experience withthese techniques is growing as their advan-tages and limitations are studied in compari-son with standard CABG.

LAD

RCA

Internalmammary

artery

Saphenousveingraft

Figure 6.10. Coronary artery bypass surgery. A. Theleft internal mammary artery originates from the left sub-clavian artery, and in this schematic, is anastomosed tothe left anterior descending (LAD) coronary artery distalto a tight stenosis (black segment). B. One end of asaphenous vein graft is sutured to the proximal aorta andthe other end to the right coronary artery (RCA) distal toa stenotic segment.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 165

Medical Versus Revascularization Therapy

Many patients with chronic, stable anginacan be successfully managed with pharma-cologic therapy alone. However, if anginalsymptoms prove refractory to a good phar-macologic program, or if intolerable drugside effects develop, coronary angiographyis usually recommended for further thera-peutic planning. For patients whose anginais controlled by medications, it is standardto perform noninvasive testing (e.g., exercisetesting, echocardiography) to identify thosewith high-risk disease, because the long-termprognosis for such patients can be improvedby coronary revascularization.

Once coronary angiography is obtained,the decision to proceed with percutaneousintervention or bypass graft surgery dependson several considerations, including thoselisted in Table 6.4. In general, patients withpersistent episodes of angina and signifi-cant stenoses in one to two coronary arter-ies are good candidates for PCI, as are certainlower-risk patients with three-vessel disease.Conversely, patients who fare better over thelong term with CABG include those with significant (>50%) stenosis of the left maincoronary artery and patients with multi-vessel disease who also have reduced LVcontractile function or diabetes.

Each of the previously described appro-aches for the treatment of coronary diseaseis benefiting from rapidly developing re-

search advancements. New surgical tech-niques (increased use of various arterialgrafts, less invasive operations); novel ad-juncts to stenting (potent antithromboticdrugs, advanced approaches to prevent in-stent restenosis); and progress in phar-macologic management (e.g., aggressiveuse of statins and antithrombotic drugs)will likely further improve outcomes andbetter define the best therapeutic approachesfor specific subsets of patients with chronicCAD.

SUMMARY

1. Cardiac ischemia results from an imbal-ance between myocardial oxygen supplyand demand. Myocardial oxygen supplyis determined by the oxygen content ofthe blood and coronary blood flow. Thelatter is dependent on the coronary per-fusion pressure and coronary vascular re-sistance. Key regulators of myocardialoxygen demand include myocardial wallstress, heart rate, and contractility.

2. In the presence of atherosclerotic dis-ease, myocardial oxygen supply is com-promised. Atherosclerotic plaques causevascular lumen narrowing and reducecoronary blood flow. In addition, ather-osclerosis-associated endothelial celldysfunction causes inappropriate vaso-constriction of the coronary resistancevessels.

166 Chapter Six

TABLE 6.4. Relative Advantages of Coronary Revascularization Procedures

Percutaneous Coronary Coronary Artery Bypass Interventions (PCI) Graft Surgery (CABG)

Less invasive than CABG More effective for long-term relief of angina than PCI or pharma-cologic therapy

Shorter hospital stay and easier Most complete revascularizationrecuperation than CABG

Superior to pharmacologic therapy Improved survival in patients withfor relief of angina • >50% left main stenosis

• 3-vessel CAD, especially if LV contractile function is impaired• 2-vessel disease with tight (>75%) LAD stenosis, especially if LV

contractile function is impaired• Diabetes and multivessel disease

CAD, coronary artery disease; LV, left ventricle; LAD, left anterior descending coronary artery; MI, myocardial infarction.

Fig. 4

10090-06_CH06.qxd 8/31/06 5:22 PM Page 166


3. Angina pectoris is the most frequentsymptom of intermittent ischemia, andits diagnosis relies heavily on the patient’sdescription of the discomfort. Angina maybe accompanied by signs and symptoms ofadrenergic stimulation, pulmonary con-gestion, and transient LV systolic and di-astolic dysfunction.

4. Laboratory studies useful in the diagnosisof angina include the electrocardiogram(ST segment and T wave abnormalities),exercise (or pharmacologic) stress testing,and coronary angiography.

5. Standard pharmacologic therapy for thetreatment of chronic angina includesagents to prevent ischemia and relievesymptoms (organic nitrates, β-blockers,and calcium channel antagonists, aloneor in combination) as well as agents thatreduce the risk of acute coronary syn-dromes and death (e.g., aspirin, anticho-lesterol therapy, and consideration ofACE inhibitors). Modifiable risk factorsfor atherosclerosis should be corrected.

6. Revascularization with PCI or CABGsurgery may provide relief from ischemiain patients with chronic angina who arerefractory to, or unable to tolerate, med-ical therapy. CABG confers improved sur-vival rates to certain high-risk groups.

Acknowledgments

Contributors to the previous editions of this chapterwere Christopher P. Chiodo, MD; Carey Farquhar,MD; Anurag Gupta, MD; Rainu Kaushal, MD;William Carlson, MD; Michael E. Mendelsohn, MD;Patrick T. O’Gara, MD; Marc S. Sabatine, MD; andLeonard S. Lilly, MD.

Additional Reading

Abrams J. Chronic stable angina. N Eng J Med 2005;352:2524–2533.

Babapulle MN, Joseph L, Belisle P, et al. A hierarchi-cal Bayesian meta-analysis of randomised clinicaltrials of drug-eluting stents. Lancet 2004;364:583–591.

Chaitman BR. Ranolazine for the treatment of chronicangina and potential use in other cardiovascularconditions. Circulation 2006;113:2462–2472.

Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA2004 guideline update for coronary artery bypassgraft surgery: summary article: a report of theAmerican College of Cardiology/American HeartAssociation Task Force on Practice Guidelines. Cir-culation 2004;110:1168–1176.

Gibbons RJ, Abrams J, Chatterjee K, et al. ACC/AHA2002 Guideline Update for the Management ofPatients With Chronic Stable Angina-SummaryArticle: A Report of the American College of Car-diology/American Heart Association Task Forceon Practice Guidelines (Committee on the Man-agement of Patients With Chronic Stable Angina).Circulation 2003;107:149–158.

Hannan EL, Racz MJ, Walford G, et al. Long-termoutcomes of coronary-artery bypass grafting ver-sus stent implantation. N Engl J Med 2005;352:2174–2183.

Henderson RA, Pocock SJ, Clayton TC, et al; SecondRandomized Intervention Treatment of Angina(RITA-2) Trial Participants. Seven-year outcome inthe RITA-2 trial: coronary angioplasty versus med-ical therapy. J Am Coll Cardiol 2003;42:1161–1170.

Hill J, Timmis A. Exercise tolerance testing. BMJ2002;324:1084–1087.

Iakovou, I, Schmidt, T, Bonizzoni, E, et al. Incidence,predictors, and outcome of thrombosis after suc-cessful implantation of drug-eluting stents. JAMA2005;293:2126–2130.

LaRosa JC, Grundy SM, Waters DD, et al.; Treating toNew Targets (TNT) Investigators. Intensive lipidlowering with atorvastatin in patients with stablecoronary disease. N Engl J Med 2005;352:1425–1435.

Pearson TA, Mensah GA, Alexander RW, et al. Mark-ers of inflammation and cardiovascular disease:application to clinical and public health practice:A statement for healthcare professionals from theCenters for Disease Control and Prevention andthe American Heart Association. Circulation 2003;107:499–511.

Ridker PM, Cannon CP, Morrow DA, et al.; PROVEIT-TIMI 22 Investigators. C-reactive protein levelsand outcomes after statin therapy. N Engl J Med2005:352;20–28.

Serruys PW, Kutryk MJB, Ong ATL. Coronary-arterystents. N Engl J Med 2006;354:483–495.

Smith SC Jr, Feldman TE, Hirshfeld JW, et al.ACC/AHA/SCAI 2005 guideline update for percu-taneous coronary intervention—summary article:a report of the American College of Cardiology/American Heart Association Task Force on PracticeGuidelines. Circulation 2006;113:156–175.

10090-06_CH06.qxd 8/31/06 5:22 PM Page 167

Chapter 6—Author Queries

1. AU: Equal sign correct?2. AU: Equation 6-1 did not include delta symbol. OK?3. AU: Please confirm x-reference; the term is not used in text of chapter 1.4. AU: The term “scintigraphy” is not used in ch. 3.5. ED: Depending on how artist draws this from provided scrap, permission may be needed

10090-06_CH06.qxd 8/31/06 5:22 PM Page a

168

PATHOGENESIS OF ACUTE CORONARY SYNDROMESNormal HemostasisEndogenous Antithrombotic MechanismsPathogenesis of Coronary ThrombosisNonatherosclerotic Causes of Acute CoronarySyndromes

PATHOLOGY AND PATHOPHYSIOLOGYEarly Changes in InfarctionLate Changes in InfarctionFunctional Alterations

CLINICAL FEATURES OF ACUTE CORONARY SYNDROMESClinical PresentationDiagnosis of Acute Coronary Syndromes

TREATMENT OF ACUTE CORONARY SYNDROMES

Acute Treatment of Unstable Angina and Non–ST-Elevation Myocardial InfarctionAcute Treatment of ST-Elevation MyocardialInfarctionAdjunctive Therapies

COMPLICATIONSRecurrent IschemiaArrhythmiasMyocardial DysfunctionRight Ventricular InfarctionMechanical ComplicationsPericarditisThromboembolism

RISK STRATIFICATION AND MANAGEMENTFOLLOWING MYOCARDIAL INFARCTION

C H A P T E R

7Acute CoronarySyndromesHaley NaikMarc S. SabatineLeonard S. Lilly

Acute coronary syndromes (ACS) are life-threatening conditions that can punctuatethe course of patients with coronary arterydisease at any time. These syndromes en-compass a continuum that ranges from anunstable pattern of angina pectoris to thedevelopment of a large acute myocardial infarction (MI), a condition of irreversiblenecrosis of heart muscle (Fig. 7.1). All acutesyndromes share a common initiating patho-

physiologic mechanism, as this chapter willexamine.

The frequency of ACS is staggering: morethan 1.6 million people are admitted to hos-pitals in the United States each year withthese conditions. Despite that daunting sta-tistic, mortality associated with ACS hassubstantially and continuously declined inrecent decades as a result of major therapeu-tic and preventive advances. This chapter

Fig. 1

10090-07_CH07.qxd 8/31/06 4:50 PM Page 168

Acute Coronary Syndromes 169

considers the events that lead to ACS, thepathologic and functional changes that fol-low, and therapeutic approaches that ame-liorate the aberrant pathophysiology.

PATHOGENESIS OF ACUTECORONARY SYNDROMES

More than 90% of ACS result from disrup-tion of an atherosclerotic plaque with sub-sequent platelet aggregation and formationof an intracoronary thrombus. The throm-bus transforms a region of plaque narrowingto one of severe or complete occlusion, andthe impaired blood flow causes a markedimbalance between myocardial oxygen sup-ply and demand. The form of ACS that re-sults depends on the degree of coronary ob-struction and associated ischemia (see Fig.7.1). A partially occlusive thrombus is thetypical cause of the closely related syn-dromes unstable angina (UA) and non–ST-elevation myocardial infarction (NSTEMI,also referred to as non–Q-wave MI), with thelatter being distinguished from the former bythe presence of myocardial necrosis. At theother end of the spectrum, if the thrombuscompletely obstructs the coronary artery, theresults are more severe ischemia and a largeramount of necrosis, manifesting as an ST-elevation myocardial infarction (STEMI,also referred to as Q-wave MI).

The responsible thrombus in ACS appearsto be generated by interactions among theatherosclerotic plaque, the coronary endo-thelium, circulating platelets, and the dyna-mic vasomotor tone of the vessel wall, whichoverwhelm the natural antithrombotic me-chanisms described in the next section.

Normal Hemostasis

When a normal blood vessel is injured, theendothelial surface becomes disrupted and

thrombogenic connective tissue is exposed.Primary hemostasis is the first line of defenseagainst bleeding. This process begins withinseconds of vessel injury and is mediated bycirculating platelets, which adhere to colla-gen in the vascular subendothelium and aggregate to form a “platelet plug.” Whilethe primary hemostatic plug forms, the ex-posure of subendothelial tissue factor trig-gers the plasma coagulation cascade, initiat-ing the process of secondary hemostasis. Theplasma coagulation proteins involved in se-condary hemostasis are sequentially activa-ted at the site of injury and ultimately forma fibrin clot by the action of thrombin. Theresulting clot stabilizes and strengthens theplatelet plug.

The normal hemostatic system mini-mizes blood loss from injured vessels, butthere is little difference between this physi-ologic response and the pathologic processof coronary thrombosis triggered by disrup-tion of atherosclerotic plaques.

Endogenous AntithromboticMechanisms

Normal blood vessels, including the coro-nary arteries, are replete with safeguards thatprevent spontaneous thrombosis and occlu-sion, some examples of which are shown inFigure 7.2.

Inactivation of Clotting Factors

Several natural inhibitors tightly regulate thecoagulation process to oppose clot forma-tion and maintain blood fluidity. The mostimportant of these are antithrombin III, proteins C and S, and tissue factor pathwayinhibitor.

Antithrombin III is a plasma protein thatirreversibly binds to thrombin and otherclotting factors, inactivating them and facili-

ACUTE CORONARY SYNDROMES

UNSTABLE ANGINA NON–ST-ELEV ATI ON MI ST-ELEV ATI ONM I

Figure 7.1. The continuum of acute coronary syndromes ranges from unstable angina, throughnon–ST-elevation myocardial infarction (MI), to ST-elevation MI.

Fig. 2

10090-07_CH07.qxd 8/31/06 4:50 PM Page 169

tating their clearance from the circulation(see mechanism 1 in Fig. 7.2). The effective-ness of antithrombin III is increased a thou-sandfold by binding to heparan sulfate, a heparin-like molecule normally present onthe luminal surface of endothelial cells.

Protein C or protein S and thrombo-modulin form a natural anticoagulant sys-tem that inactivates the “acceleration” fac-tors of the coagulation pathway (i.e., factorsVa and VIIIa). Protein C is synthesized inthe liver and circulates in an inactive form.Thrombomodulin is a thrombin-binding receptor normally present on endothelialcells. Thrombin bound to thrombomod-ulin cannot convert fibrinogen to fibrin (thefinal reaction in clot formation). Instead,

the thrombin-thrombomodulin complex ac-tivates protein C. Activated protein C de-grades factors Va and VIIIa (see mechanism2 in Fig. 7.2), thereby inhibiting coagula-tion. The presence of protein S in the circu-lation enhances the inhibitory function ofprotein C.

Tissue factor pathway inhibitor (TFPI)is a plasma serine protease inhibitor that is activated by coagulation factor Xa. Thecombined factor Xa-TFPI binds to and in-activates the complex of tissue factor withfactor VIIa that normally triggers the extrin-sic coagulation pathway (see mechanism 3in Fig. 7.2). Thus, TFPI serves as a negativefeedback inhibitor that interferes with coag-ulation.

170 Chapter Seven

Plasminogen

Plasmin

Fibrinsplit

products

InactivatedVa,VIIIafactors

ProstacyclinandNO

Irreversiblethrombininhibition

Thrombin

Antithrombin III

Heparansulfate

Fibrinclot

Protein S

Protein C*

Protein C

ThrombinTM

tPA

VII

TFPI

Xa

–

4

3

52

1

Tissuefactor

Figure 7.2. Endogenous protective mechanisms against thrombosis andvessel occlusion. (1) Inactivation of thrombin by antithrombin III (AT), the effec-tiveness of which is enhanced by binding of AT to heparan sulfate. (2) Inactivationof clotting factors Va and VIIIa by activated protein C (protein C*), an action thatis enhanced by protein S. Protein C is activated by the thrombomodulin (TM)-thrombin complex. (3) Inactivation of factor VII/tissue factor complex by tissue fac-tor pathway inhibitor (TFPI). (4) Lysis of fibrin clots by tissue plasminogen activator(tPA). (5) Inhibition of platelet activation by prostacyclin and NO.

10090-07_CH07.qxd 8/31/06 4:50 PM Page 170


Lysis of Fibrin Clots

Tissue plasminogen activator (tPA) is aprotein secreted by endothelial cells in re-sponse to many triggers of clot formation.tPA cleaves the protein plasminogen to formactive plasmin, which in turn enzymaticallydegrades fibrin clots (see mechanism 4 inFig. 7.2). When tPA binds to fibrin in a form-ing clot, its ability to convert plasminogento plasmin is greatly enhanced.

Endogenous Platelet Inhibition and Vasodilatation

Prostacyclin is synthesized and secreted byendothelial cells (see mechanism 5 in Fig.7.2), as described in Chapter 6. Prostacyclinincreases platelet levels of cyclic AMP andthereby strongly inhibits platelet activationand aggregation. It also indirectly inhibits coagulation via its potent vasodilating pro-perties. Vasodilatation helps guard againstthrombosisbyaugmenting blood flow (whichminimizes contact between procoagulant fac-

tors) and by reducing shear stress (an indu-cer of platelet activation).

Nitric oxide (NO) is similarly secreted byendothelial cells, as described in Chapter 6.It acts locally to inhibit platelet activation,and it too serves as a potent vasodilator.

Pathogenesis of Coronary Thrombosis

Normally, the mechanisms shown in Figure7.2 serve to prevent spontaneous intravas-cular thrombus formation. However, abnor-malities associated with atherosclerotic le-sions may overwhelm these defenses andresult in coronary thrombosis and vessel oc-clusion (Fig. 7.3). Atherosclerosis contributesto thrombus formation by (1) plaque rup-ture, which exposes the circulating blood elements to thrombogenic substances, and(2) endothelial dysfunction with the loss ofnormal protective antithrombotic and vaso-dilatory properties.

Atherosclerotic plaque rupture is consid-ered the major trigger of coronary thrombo-

Atherosclerosis

Figure 7.3. Mechanisms of coronary thrombus formation. Factors that contribute to this process includeplaque disruption (e.g., rupture), activation of platelets and the clotting cascade, and inappropriate vaso-constriction and loss of normal antithrombotic defenses because of dysfunctional endothelium.

Fig. 3

10090-07_CH07.qxd 8/31/06 4:50 PM Page 171

sis. The underlying causes of plaque disrup-tion are (1) chemical factors that destabi-lize atherosclerotic lesions and (2) physicalstresses to which the lesions are subjected.As described in Chapter 5, atheroscleroticplaques consist of a lipid-laden core sur-rounded by a fibrous external cap. Sub-stances released from inflammatory cellswithin the plaque can compromise the in-tegrity of the fibrous cap. For example, Tlymphocytes elaborate γ-interferon, whichinhibits collagen synthesis by smooth mus-cle cells and thereby interferes with theusual strength of the cap. Additionally, cellswithin atherosclerotic lesions produce en-zymes (e.g., metalloproteinases) that de-grade the interstitial matrix, further com-promising plaque stability. A weakened orthin-capped plaque is subject to rupture,particularly in its “shoulder” region (theborder with the normal arterial wall that issubjected to high circumferential stress) ei-ther spontaneously or by physical forces,such as intraluminal blood pressure and tor-sion from the beating myocardium.

ACS sometimes occur in the setting ofcertain triggers, such as strenuous physicalactivity or emotional upset. The activationof the sympathetic nervous system in thesesituations increases the blood pressure,heart rate, and force of ventricular contrac-tion—actions that may stress the athero-sclerotic lesion, thereby causing the plaqueto fissure or rupture. In addition, MI is mostlikely to occur in the early morning hours.This observation may relate to the tendencyof key physiologic stressors (such as systolicblood pressure, blood viscosity, and plasmaepinephrine levels) to be most elevated atthat time of day, and these factors subjectvulnerable plaques to rupture.

Following plaque rupture, thrombus for-mation is provoked via the mechanismsshown in Figure 7.3. The exposure of tissuefactor from the atheromatous core triggersthe coagulation pathway, while the expo-sure of subendothelial collagen activatesplatelets. Activated platelets release thecontents of their granules, which includefacilitators of platelet aggregation (e.g.,adenosine diphosphate [ADP] and fibrino-

gen), activators of the coagulation cascade(e.g., factor Va), and vasoconstrictors (e.g.,thromboxane and serotonin). The devel-oping intracoronary thrombus, intraplaquehemorrhage, and vasoconstriction all con-tribute to narrowing the vessel lumen, creating turbulent blood flow that contri-butes to shear stress and further platelet ac-tivation.

Dysfunctional endothelium, which isapparent even in mild atherosclerotic coro-nary disease, also increases the likelihood ofthrombus formation. In the setting of en-dothelial dysfunction, reduced amounts ofvasodilators (e.g., NO and prostacyclin) arereleased and inhibition of platelet aggrega-tion by these factors is impaired, resulting inthe loss of a key defense against thrombosis.

Not only is dysfunctional endotheliumless equipped to prevent platelet aggrega-tion, but also it is less able to counteract thevasoconstricting products of platelets. Dur-ing thrombus formation, vasoconstriction is promoted both by platelet products(thromboxane and serotonin) and by throm-bin within the developing clot. The normalplatelet-associated vascular response is va-sodilatation, because platelet products stim-ulate endothelial NO and prostacyclin re-lease, the influences of which predominateover direct platelet-derived vasoconstrictors(see Fig. 6.4). However, reduced secretion ofendothelial vasodilators in atherosclerosis al-lows vasoconstriction to proceed unchecked.Similarly, thrombin in a forming clot is a po-tent vascular smooth muscle constrictor inthe setting of dysfunctional endothelium.Vasoconstriction causes torsional stresses thatcan contribute to plaque rupture or can tran-siently occlude the stenotic vessel throughheightened arterial tone. The reduction incoronary blood flow caused by vasocon-striction also reduces the washout of coagu-lation proteins, thereby enhancing throm-bogenicity.

Significance of Coronary Thrombosis

The formation of an intracoronary throm-bus results in one of several potential out-comes (Fig. 7.4). For example, plaque rup-

172 Chapter Seven

Fig. 4

10090-07_CH07.qxd 8/31/06 4:50 PM Page 172


ture is sometimes superficial, minor, and self-limited, such that only a small, nonocclusivethrombus forms. In this case, the thrombusmay simply become incorporated into thegrowing atheromatous lesion through fibro-tic organization, or it may be lysed by naturalfibrinolytic mechanisms. Recurrent asymp-tomatic plaque ruptures of this type maycause gradual progressive enlargement of thecoronary stenosis.

However, deeper plaque rupture may re-sult in greater exposure of subendothelialcollagen and tissue factor, with formation ofa larger thrombus that more substantiallyoccludes the vessel’s lumen. Such obstruc-tion may cause prolonged severe ischemiaand the development of an acute coronarysyndrome. If the intraluminal thrombus atthe site of plaque disruption totally occludesthe vessel, blood flow beyond the obstruc-tion will cease, prolonged ischemia willoccur, and an MI (usually an ST-elevationMI) will result. Conversely, if the thrombuspartially occludes the vessel (or if it totallyoccludes the vessel but only transiently be-

cause of spontaneous recanalization or byrelief of superimposed vasospasm), the se-verity and duration of ischemia will be less,and a smaller non–ST-elevation MI or UAare the more likely outcomes. The distinc-tion between a non–ST-elevation MI and UA is based on the degree of the ischemiaand whether the event is severe enough tocause necrosis, indicated by the presence of certain serum biomarkers (see Fig. 7.4).Nonetheless, UA and NSTEMI act quite alike,and the management of these entities issimilar.

Occasionally, a non–ST-elevation infarctmay result from total coronary occlusion. Inthis case, it is likely that a substantial col-lateral blood supply (see Chapter 1) limitsthe extent of necrosis, such that a larger ST-elevation MI is prevented.

Nonatherosclerotic Causes ofAcute Coronary Syndromes

Rarely, mechanisms other than acute throm-bus formation can precipitate an acute coro-

Coronary thrombus

Partially occlusivethrombus

Small thrombus(non–flow limiting)

Occlusivethrombus

ST segmentdepression and/orT wave inversion

Unstable angina Non–ST-segmentelevation MI

(Transientischemia) (Prolonged

ischemia)

No ECGchanges ST elevation

(Q waves later)

Healing andplaque enlargement

ST-segmentelevation MI

Figure 7.4. Consequences of coronary thrombosis. A small thrombus formed on superficial plaque rup-ture may not result in symptoms or electrocardiogram (ECG) abnormalities, but healing and fibrous organi-zation may incorporate the thrombus into the plaque, causing the atherosclerotic lesion to enlarge. A par-tially occlusive thrombus (with or without superimposed vasospasm) narrows the arterial lumen, restricts bloodflow, and can cause unstable angina or a non–ST-elevation MI, either of which may result in ST segment de-pression and/or T wave inversion on the ECG. A totally occlusive thrombus with prolonged ischemia is themost common cause of ST-elevation MI, in which the ECG initially shows ST segment elevation, followed byQ wave development. An occlusive thrombus that recanalizes, or one that develops in a region served by ad-equate collateral blood flow, may result in less prolonged ischemia and a non–ST-elevation MI instead. Mark-ers of myocardial necrosis include cardiac-specific troponins and creatine kinase MB isoenzyme.

10090-07_CH07.qxd 8/31/06 4:50 PM Page 173

nary syndrome (Table 7.1). These should besuspected when an ACS occurs in a youngpatient or a person with no coronary riskfactors. For example, coronary emboli frommechanical or infected cardiac valves canlodge in the coronary circulation, or inflam-mation from acute vasculitis can initiate coro-nary occlusion. Occasionally, intense tran-sient coronary spasm can sufficiently reducemyocardial blood supply to result in UA orinfarction.

Another cause of ACS is cocaine abuse.Cocaine increases sympathetic tone byblocking the presynaptic reuptake of norep-inephrine and by enhancing the release ofadrenal catecholamines, which can lead tovasospasm and therefore decreased myo-cardial oxygen supply. An acute coronarysyndrome may ensue because of increasedmyocardial oxygen demand resulting fromcocaine-induced sympathetic myocardialstimulation (increased heart rate and bloodpressure) in the face of the decreased oxygensupply.

PATHOLOGY ANDPATHOPHYSIOLOGY

MI (either STEMI or NSTEMI) results whenmyocardial ischemia is sufficiently severe tocause myocyte necrosis. Although by defin-ition UA does not result in necrosis, MI maysubsequently ensue if the underlying patho-physiology of the unstable pattern of anginais not promptly corrected.

In addition to their clinical classifications,infarctions can be described pathologicallyby the extent of necrosis they produce with-in the myocardial wall. Transmural infarcts

span the entire thickness of the myocardiumand result from total, prolonged occlusionof an epicardial coronary artery. Conversely,subendocardial infarcts exclusively involvethe innermost layers of the myocardium.The subendocardium is particularly suscep-tible to ischemia because it is the zone subjected to the highest pressure from theventricular chamber, has few collateral con-nections that supply it, and is perfused byvessels that must pass through layers of con-tracting myocardium.

Infarction represents the culmination ofa disastrous cascade of events, initiated byischemia, that progresses from a potentiallyreversible phase to irreversible cell death.Myocardium that is supplied directly by anoccluded vessel may die quickly. The adja-cent tissue may not necrose immediately be-cause it may be sufficiently perfused bynearby patent vessels. However, the neigh-boring cells may become increasingly is-chemic over time, as demand for oxygencontinues in the face of decreased oxygensupply. Thus, the region of infarction maysubsequently extend outward. The amountof tissue that ultimately succumbs to infarc-tion therefore relates to (1) the mass of my-ocardium perfused by the occluded vessel,(2) the magnitude and duration of impairedcoronary blood flow, (3) the oxygen demandof the affected region, (4) the adequacy ofcollateral vessels that provide blood flowfrom neighboring nonoccluded coronary ar-teries, and (5) the degree of tissue responsethat modifies the ischemic process.

The pathophysiologic alterations thattranspire during MI occur in two stages:early changes at the time of acute infarction

174 Chapter Seven

TABLE 7.1. Causes of Acute Coronary Syndromes

• Atherosclerosis with superimposed thrombus• Vasculitic syndromes (see Chapter 15)• Coronary emboli (e.g., from endocarditis, artificial valves)• Congenital anomalies of the coronary arteries• Coronary trauma or aneurysm• Severe coronary artery spasm (primary or cocaine-induced)• Increased blood viscosity (e.g., polycythemia vera, thrombocytosis)• Markedly increased myocardial oxygen demand (e.g., severe aortic stenosis)

Tab. 1

10090-07_CH07.qxd 8/31/06 4:50 PM Page 174


and late changes during myocardial healingand remodeling.

Early Changes in Infarction

Earlychanges includethe histologic evolutionof the infarct and the functional impact ofoxygen deprivation on myocardial contractil-ity. These changes culminate in coagulativenecrosis of the myocardium in 2 to 4 days.

Cellular Changes

As oxygen levels fall in the myocardiumsupplied by an abruptly occluded coronaryvessel, there is a rapid shift from aerobic toanaerobic metabolism (Fig. 7.5). Because mi-tochondria can no longer oxidize fats or pro-ducts of glycolysis, high-energy phosphateproduction drops dramatically and anaero-bic glycolysis leads to the accumulation oflactic acid. This results in a lowered pH.

Furthermore, the paucity of high-energyphosphates such as adenosine triphosphate(ATP) interferes with the transmembraneNaK-ATPase, with resultant elevation in in-tracellular Na and extracellular K. Rising Nacontributes to cellular edema. Membraneleak and rising extracellular K concentration

contributes to alterations in the transmem-brane electrical potential, predisposing themyocardium to lethal arrhythmias. Intracel-lular calcium accumulates in the damagedmyocytes and is thought to contribute tothe final common pathway of cell destruc-tion through the activation of degradativelipases and proteases.

Collectively, these metabolic changes de-crease myocardial function as early as 2 min-utes following occlusive thrombosis. Withoutintervention, irreversible cell injury ensues in 20 minutes and is marked by the deve-lopment of membrane defects. Proteolytic enzymes leak across the myocyte’s alteredmembrane, damaging adjacent myocardium,and the release of certain macromoleculesinto the circulation serves as a clinical markerof acute infarction.

Edema of the myocardium develops with-in 4 to 12 hours, as vascular permeability in-creases and interstitial oncotic pressure rises(because of the leak of intracellular pro-teins). The earliest histologic changes of irreversible injury are wavy myofibers,which appear as intercellular edema sepa-rates the myocardial cells that are tuggedabout by the surrounding, functional myo-cardium (Fig. 7.6). Contraction bands can

Myocardial hypoxia

Impaired Na-+, K+-ATPase Anaerobic metabolism

ATPProteasesLipases

Chromatin clumpingProtein denaturation

Arrhythmias

Intracellularedema

Alteredmembranepotential

Cell death

ExtracellularK+

IntracellularNa+

IntracellularCa++

IntracellularH+

ATP

Figure 7.5. Mechanisms of cell death in myocardial infarction. Acute ischemia rapidly de-pletes the intracellular supply of adenosine triphosphate (ATP) as aerobic metabolism fails. Subse-quent intracellular acidosis and impairment of ATP-dependent processes culminate in intracellularcalcium accumulation, edema, and cell death.

Fig. 5

Fig. 6

10090-07_CH07.qxd 8/31/06 4:50 PM Page 175

176 Chapter Seven

Figure 7.6. Pathologic evolution in acute myocardial infarction. A. Early wavy myofibers and edema; viablemyocardium is at lower left. B. Coagulation necrosis and dense infiltration of neutrophils. C. Necrotic myocytes largelyremoved by phagocytes (7 to 10 days); viable myocardium at lower left. D. Granulation tissue with early collagen de-position; new capillaries have formed (arrows). E. Late fibrotic scarring. (Reprinted with permission from Schoen FJ.Interventional and Surgical Cardiovascular Pathology—Clinical Correlations and Basic Principles. Philadelphia: Saun-ders, 1989:67.)

10090-07_CH07.qxd 8/31/06 4:50 PM Page 176


often be seen near the borders of the infarct:sarcomeres are contracted and consolidatedand appear as bright eosinophilic belts.

An acute inflammatory response, with in-filtration of neutrophils, begins after approx-imately 4 hours and incites further tissuedamage. Within 18 to 24 hours, coagulationnecrosis is evident with pyknotic nuclei andbland eosinophilic cytoplasm, seen by lightmicroscopy. These early changes are de-monstrated in Figure 7.6 and summarized inTable 7.2.

Gross Changes

Gross morphologic changes do not appearuntil 18 to 24 hours after coronary occlusion,although certain staining techniques (e.g.,tetrazolium) permit the pathologist to iden-tify regions of infarction earlier. Most often,ischemia and infarction begin in the suben-docardium and then extend laterally andoutward toward the epicardium.

Late Changes in Infarction

Late pathologic changes in the course ofacute MI (see Table 7.2) include (1) the clear-ing of necrotic myocardium and (2) the de-position of collagen to form scar tissue.

Irreversibly injured myocytes do not re-generate; rather, the cells are removed and

replaced by fibrous tissue. Macrophages in-vade the inflamed myocardium shortly afterneutrophil infiltration and remove necrotictissue. This period of tissue resorption istermed yellow softening because connectivetissue elements are destroyed and removedalong with dead myocardial cells. The phago-cytic clearing, combined with thinning anddilation of the infarcted zone, results instructural weakness of the ventricular walland the possibility of myocardial wall rup-ture at this stage. Fibrosis subsequently en-sues, and scarring is complete by 7 weeksafter infarction (see Fig. 7.6).

Functional Alterations

Impaired Contractility and Compliance

The destruction of functional myocardialcells in infarction quickly leads to impairedventricular contraction (systolic dysfunc-tion). Cardiac output is further compro-mised because synchronous contraction ofmyocytes is lost. Specific terms are used todescribe the types of wall motion abnormal-ities that can result. A localized region of re-duced contraction is termed hypokinetic; asegment that does not contract at all is calledakinetic; and a dyskinetic region is one thatbulges outward during contraction of the re-maining functional portions of the ventricle.

TABLE 7.2. Pathologic Time Line in Transmural Infarction

Time Event

Early changes1–2 min ATP levels fall; cessation of contractility10 min 50% depletion of ATP; cellular edema, decreased membrane potential and suscepti-

bility to arrhythmias20–24 min Irreversible cell injury1–3 hours Wavy myofibers4–12 hours Hemorrhage, edema, PMN infiltration begins18–24 hours Coagulation necrosis (pyknotic nuclei with eosinophilic cytoplasm), edema2–4 days Total coagulation necrosis (no nuclei or striations, rimmed by hyperemic tissue);

monocytes appear; PMN infiltration peaksLate changes5–7 days Yellow-softening from resorption of dead tissue by macrophages7 days Ventricular remodeling7 weeks Fibrosis and scarring complete

ATP, adenosine triphosphate; PMN, polymorphonuclear leukocyte.

Tab. 2

10090-07_CH07.qxd 8/31/06 4:50 PM Page 177

During ACS, the left ventricle is also ad-versely compromised by diastolic dysfunc-tion. Ischemia and/or infarction impair diastolic relaxation (an energy-dependentprocess; see Chapter 1), which reduces ven-tricular compliance and contributes to ele-vated ventricular filling pressures.

Stunned Myocardium

Sometimes transient myocardial ischemiacan result in a very prolonged, but gradu-ally reversible, period of contractile dys-function. For example, as described inChapter 6, stunned myocardium is tissuethat demonstrates prolonged systolic dys-function after a discrete episode of severeischemia, despite restoration of adequateblood flow, and gradually regains contrac-tile force days to weeks later. Stunning mayplay an important role in patients with UAor in myocardium adjacent to the region of an acute infarction. In both instances,prolonged contractile dysfunction of af-fected ventricular segments may be evi-dent after the event, simulating infarctedtissue. However, if the tissue is simplystunned rather than necrotic, its functionwill recover over time.

Ischemic Preconditioning

Brief ischemic insults to a region of myo-cardium may render that tissue more resis-tant to subsequent episodes, a phenomenontermed ischemic preconditioning. The clin-ical relevance is that patients who sustain anMI in the context of recent angina experi-ence less morbidity and mortality than thosewithout preceding ischemic episodes. Themechanism of this phenomenon is unknownbut appears to involve ischemia-related acti-vation of adenosine receptors.

Ventricular Remodeling

Following an MI, changes occur in the geom-etry of both the infarcted and noninfarctedventricular muscle. Such alterations in cham-ber size and wall thickness affect long-termventricular function and prognosis.

In the early post-MI period, infarct ex-pansion may occur, with affected ventricularsegment enlarging without additional my-ocyte necrosis. Infarct expansion representsthinning and dilatation of the necrotic zoneof tissue, likely because of “slippage” be-tween the muscle fibers, resulting in a de-creased volume of myocytes in the region.Infarct expansion can be detrimental becauseit increases ventricular size, which (1) aug-ments wall stress, (2) impairs systolic con-tractile function, and (3) increases the likeli-hood of aneurysm formation.

In addition to early expansion of the in-farcted territory, remodeling of the ventriclemay also involve dilatation of the overworkednoninfarcted segments, which are subjected toincreased wall stress. This dilatation begins inthe early postinfarct period and continuesover the ensuing weeks and months. Initially,chamber dilatation serves a compensatoryrole because it increases cardiac output via theFrank-Starling mechanism (see Chapter 9),but progressive enlargement may ultimatelylead to heart failure and predisposes to ven-tricular arrhythmias.

Adverse ventricular remodeling can bebeneficially modified by certain interven-tions. At the time of infarction, for example,reperfusion therapies limit infarct size andtherefore decrease the likelihood of infarctexpansion. In addition, drugs that interferewith the renin-angiotensin system havebeen shown to attenuate progressive remod-eling and to reduce short- and long-termmortality after infarction (as discussed laterin the chapter).

CLINICAL FEATURES OF ACUTECORONARY SYNDROMES

Because ACS represent disorders along acontinuum, their clinical features overlap.In general, the severity of symptoms and as-sociated laboratory findings progress fromUA on one side of the continuum, throughNSTEMI, to STEMI on the other end of the continuum (see Fig. 7.1). Distinguishingamong these syndromes is based on theclinical presentation, electrocardiographicfindings, and serum biomarkers of myocar-

178 Chapter Seven

10090-07_CH07.qxd 8/31/06 4:50 PM Page 178


dial damage. To institute appropriate imme-diate therapy, the most important distinc-tion to make is between ACS that cause STsegment elevation on the electrocardiogram(STEMI) and those acute syndromes that donot (UA and NSTEMI).

Historically, MIs have been classified asQ-wave or non–Q-wave infarctions. Dogmaheld that transmural infarcts produce Qwaves (after initial ST elevation) on the elec-trocardiogram (ECG), whereas subendocar-dial infarcts generate ST depressions withoutQ-wave development. However, it is nowknown that these ECG findings do not reli-ably correlate with the pathologic findingsand that much overlap exists among thetypes of infarction. Moreover, the use of Qwaves to classify ACS is now less clinicallyimportant, because Q waves, unlike STchanges, may take hours or longer to developand cannot be used to make early therapeu-tic decisions. Thus, for the remainder of thischapter, the terms STEMI and NSTEMI willbe used instead of Q-wave and non–Q-waveMI, respectively.

Clinical Presentation

Unstable Angina

UA presents as an acceleration of ischemicsymptoms in one of three ways: (1) a cres-cendo pattern in which a patient with chronic stable angina experiences a sudden

increase in the frequency, duration, and/orintensity of ischemic episodes; (2) episodes ofangina that occur at rest, without provoca-tion; or (3) the new onset of anginal episodes,described as severe, in a patient without pre-vious symptoms of coronary artery disease.These features are different from the patternof chronic stable angina, in which instancesof chest discomfort are predictable, brief, andnonprogressive, occurring only during physi-cal exertion or emotional stress. Patients withUA may progress further along the contin-uum of ACS and develop evidence of necrosis(i.e., acute NSTEMI or STEMI) unless the con-dition is recognized and promptly treated.

Acute Myocardial Infarction

The symptoms and physical findings ofacute MI (both STEMI and NSTEMI) can bepredicted from the pathophysiology de-scribed earlier in this chapter and are sum-marized in Table 7.3. The discomfort expe-rienced during an MI resembles anginapectoris qualitatively but is usually more se-vere, lasts longer, and may radiate morewidely. Like angina, the sensation may resultfrom the release of mediators such as adeno-sine and lactate from ischemic myocardialcells onto local nerve endings. Because is-chemia in acute MI persists and proceeds tonecrosis, these provocative substances con-tinue to accumulate and activate afferentnerves for longer periods. The discomfort is

TABLE 7.3. Signs and Symptoms of Myocardial Infarction

1. Characteristic pain • Severe, persistent, typically substernal2. Sympathetic effect • Diaphoresis

• Cool and clammy skin3. Parasympathetic (vagal effect) • Nausea, vomiting

• Weakness4. Inflammatory response • Mild fever5. Cardiac findings • S4 (and S3 if CHF present) gallop

• Dyskinetic bulge (in anterior wall MI)• Pericardial friction rub (if pericarditis present)• Systolic murmur (if mitral regurgitation or VSD)

6. Other • Pulmonary rales (if CHF present)• Jugular venous distention (if right ventricular MI)

CHF, congestive heart failure; MI, myocardial infarction; S3, third heart sound; S4, fourth heart sound; VSD, ventricular septal defect.

Tab. 3

10090-07_CH07.qxd 8/31/06 4:50 PM Page 179

often referred to other regions of the C7through T4 dermatomes, including the neck,shoulders, and arms. Initial symptoms areusually rapid in onset and briskly crescendoto leave the victim with a profound “feelingof doom.” Unlike a transient attack of angina,the pain does not wane with rest, and theremay be little response to the administrationof sublingual nitroglycerin.

The chest discomfort associated with anacute MI is sometimes severe, but not always.In fact, up to 25% of patients who sustain anMI are asymptomatic during the acute event,and the diagnosis is made only in retro-spect. This is particularly common amongdiabetic patients who may not adequatelysense pain because of peripheral neuropa-thy. In addition, some patients who presentwith MI complicated by acute pericarditismay feel more of a sharp, pleuritic-type pain(see Chapter 14), rather than the typical MIsymptoms.

The combination of intense discomfortand baroreceptor unloading (if hypotensionis present) may trigger a dramatic sympa-thetic nervous system response. Systemicsigns of subsequent catecholamine releaseinclude diaphoresis (sweating), tachycardia,and cool and clammy skin caused by vaso-constriction.

If the ischemia affects a sufficiently largeamount of myocardium, left ventricular (LV)contractility can be reduced (systolic dys-function) thereby decreasing the stroke vol-ume and causing the diastolic volume andpressure within the LV to rise. The increase inLV pressure, compounded by the ischemia-induced stiffness of the chamber (diastolicdysfunction), is conveyed to the left atriumand pulmonary veins. The resultant pul-monary congestion decreases lung compli-ance and stimulates juxtacapillary receptors.These J receptors effect a reflex that results inrapid, shallow breathing and evokes thesubjective feeling of dyspnea. Transudationof fluid into the alveoli exacerbates thissymptom.

Physical findings during an acute MI de-pend on the location and extent of the in-farct. The S4 sound, indicative of atrial con-traction into a noncompliant left ventricle,is frequently present. An S3 sound, indica-

tive of volume overload in the presence offailing LV systolic function, may also beheard. A friction rub may be present if in-flammation has extended to the pericardium.A systolic murmur may appear if ischemia-induced papillary muscle dysfunction causesmitral valvular insufficiency, or if the infarctruptures through the interventricular sep-tum to create a ventricular septal defect (asdiscussed later in the chapter).

Myocardial necrosis also activates systemicresponses to inflammation. Cytokines such asinterleukin 1 (IL-1) and tumor necrosis factor(TNF) are released from macrophages and vas-cular endothelium in response to tissue in-jury. These mediators evoke an array of clini-cal responses, including low-grade fever.

Not all patients with severe chest pain arein the midst of MI or UA. Table 7.4 lists othercommon causes of chest discomfort and clin-ical, laboratory, and radiographic features todifferentiate them from an acute coronarysyndrome.

Diagnosis of Acute Coronary Syndromes

The diagnosis of, and distinctions among,the acute coronary syndromes is made on thebasis of (1) the patient’s presenting symp-toms, (2) acute ECG abnormalities, and (3) detection or absence of specific serummarkers of myocardial necrosis (see Fig. 7.4and Table 7.5). Specifically, UA is a clinicaldiagnosis supported by the patient’s symp-toms, transient ST abnormalities on the ECG(usually ST depression and/or T wave inver-sion), and the absence of serum biomarkersof myocardial necrosis. Non–ST segment el-evation MI is distinguished from UA by thedetection of serum markers of necrosis andoften more persistent ST or T wave abnor-malities. The hallmark of ST-elevation MI isan appropriate clinical history coupled withST elevations on the ECG plus detection ofserum markers of myocardial necrosis.

ECG Abnormalities

ECG abnormalities, which reflect abnormalelectrical currents during ACS, are usuallymanifest in characteristic ways. In UA or

180 Chapter Seven

Tab. 4

Tab. 5

10090-07_CH07.qxd 8/31/06 4:50 PM Page 180


NSTEMI, ST segment depression and/or Twave inversions are most common (Fig. 7.7).These abnormalities may be transient, occur-ring just during chest pain episodes in UA, orthey may persist in patients with NSTEMI. Incontrast, as described in Chapter 4, STEMIpresents with a temporal sequence of abnor-malities: initial ST segment elevation, fol-lowed over the next hours to day by inversion

of the T wave and Q-wave development (Fig.7.8). Note that these characteristic patterns ofECG abnormalities in ACS can be minimizedor aborted by early therapeutic interventions.

Serum Markers of Infarction

Necrosis of myocardial tissue causes disrup-tion of the sarcolemma, so that intracellular

TABLE 7.4. Conditions That May Be Confused With Acute Coronary Syndromes

Condition Differentiating Features

CardiacAcute coronary syndrome • Retrosternal pressure, radiating to neck, jaw, or left shoulder and arm; more

severe and lasts longer than previous anginal attacks• ECG: localized ST elevations or depressions

Pericarditis • Sharp pleuritic pain (worsens with inspiration)• Pain varies with position (relieved by sitting)• Friction rub auscultated over precordium• Electrocardiogram: diffuse ST elevations (see Chapter 14)

Aortic dissection • Tearing, ripping pain that migrates over time (chest and back)• Asymmetry of arm blood pressures• Widened mediastinum on chest radiograph

PulmonaryPulmonary embolism • Localized pleuritic pain, accompanied by dyspnea

• Pleural friction rub may be present• Predisposing conditions for venous thrombosis

Pneumonia • Pleuritic chest pain• Cough and sputum production• Abnormal lung auscultation and percussion (i.e., consolidation)• Infiltrate on chest radiograph

Pneumothorax • Sudden sharp, pleuritic unilateral chest pain• Decreased breath sounds and hyperresonance of affected side• Chest radiograph: increased lucency and absence of pulmonary markings

GastrointestinalEsophageal spasm • Retrosternal pain, worsened by swallowing

• History of dysphagiaAcute cholecystitis • Right upper quadrant abdominal tenderness

• Often accompanied by nausea• History of fatty food intolerance

TABLE 7.5. Distinguishing Features of Acute Coronary Syndromes


Feature Unstable Angina NSTEMI STEMI

Typical symptoms Crescendo, rest, or new Prolonged “crushing” chest pain, more onset severe angina severe and wider radiation than usual angina

Serum biomarkers No Yes YesElectrocardiogram ST depression and/or ST depression and/or ST elevation

initial findings T wave inversion T wave inversion (and Q waves later)

NSTEMI, non–ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction.

Fig. 7 Fig. 8

10090-07_CH07.qxd 8/31/06 4:50 PM Page 181

macromolecules leak into the cardiac inter-stitium and ultimately into the bloodstream(Fig. 7.9). Detection of such molecules in theserum, particularly cardiac-specific troponinsand creatine kinase MB isoenzyme, servesimportant diagnostic and prognostic roles.In patients with STEMI or NSTEMI, thesemarkers rise above a threshold level in a de-fined temporal sequence.

Cardiac-Specific Troponins

Troponin (Tn) is a regulatory protein inmuscle cells that controls interactions be-

tween myosin and actin (see Chapter 1). Itconsists of three subunits: TnC, TnI, andTnT. Although these subunits are foundboth in skeletal and cardiac muscle, the car-diac forms of troponin I (cTnI) and troponinT (cTnT) are structurally unique, and highlyspecific assays for their detection in theserum have been developed. Because serumlevels are virtually absent in healthy persons,the presence of even minor elevations ofcTnI or cTnT serves as a sensitive and pow-erful marker of myocyte damage.

Cardiac troponins begin to rise 3 to 4 hoursafter the onset of an MI, peak between 18 and

182 Chapter Seven

Figure 7.7. ECG abnormalities in unstable angina and non–ST-elevation myocardial infarction.

Figure 7.8. ECG evolution during ST-elevation myocardial infarction.

Fig. 9

10090-07_CH07.qxd 8/31/06 4:50 PM Page 182


36 hours, and then decline slowly, allowingfor detection for up to 10 to 14 days. Thus,their measurement may be helpful for de-tection of MI for nearly 2 weeks after theevent occurs. Given their high sensitivityand specificity, cardiac troponins are the pre-ferred serum biomarkers to detect myocardialnecrosis.

Creatine Kinase

The enzyme creatine kinase (CK) reversiblytransfers a phosphate group from creatinephosphate, the endogenous storage form of high-energy phosphate bonds, to ADP,producing ATP. Because creatine kinase isfound in the heart, skeletal muscle, brain,and many other organs, serum concentra-tions of the enzyme may become elevatedfollowing injury to any of these tissues.

There are, however, three isoenzymes ofCK that improve diagnostic specificity of itsorigin: CK-MM (found mainly in skeletalmuscle), CK-BB (located predominantly inthe brain), and CK-MB (localized mainly inthe heart). It should be noted that smallamounts of CK-MB are found in tissues out-

side the heart, including the uterus, prostate,gut, diaphragm, and tongue. CK-MB alsomakes up 1% to 3% of the creatine kinase inskeletal muscle. In the absence of trauma tothese other organs and tissues, elevation ofCK-MB is highly suggestive of myocardialinjury. To facilitate the diagnosis of MIusing this marker, it is common to calculatethe ratio of CK-MB to total CK. The ratio isusually >2.5% in the setting of myocardialinjury and even less when CK-MB elevationis from another source.

The serum level of CK-MB starts to rise 3 to 8 hours following infarction, peaks at24 hours, and returns to normal within 48to 72 hours (see Fig. 7.9). This temporal se-quence is important because other sourcesof CK-MB (e.g., skeletal muscle injury) orother non-MI cardiac conditions that raiseserum levels of the isoenzyme (e.g., myo-carditis) do not usually show this delayedpeaking pattern.

The detection of cardiac troponins in theserum is a much more sensitive marker formyocardial necrosis than CK-MB. As a re-sult, many patients with ACS are found tohave small elevations of cardiac troponins

1 2 3 4 5

Days after onset of infarction

Troponins

MI Threshold

CK-MB

Mul

tiple

sof

MIt

hres

hold

6 7 8 9 10

1

2

5

10

20

50

Figure 7.9. Evolution of serum biomarkers in acute myocardial infarction(MI). Serum creatine kinase (and CK-MB isoenzyme) begins to rise 3 to 8 hoursafter the onset of the acute infarct and peaks at 24 hours. Cardiac troponinsare highly sensitive for myocardial injury and remain detectable in the serum formany days after the acute infarct. (Modified from Wu AH, Apple FS, Gibler WB,et al. National Academy of Clinical Biochemistry Standards of Laboratory Prac-tice: Recommendations for the use of cardiac markers in coronary artery dis-eases. Clin Chem 1999;45:1104–1121.)

10090-07_CH07.qxd 8/31/06 4:50 PM Page 183

but unremarkable CK-MB test results. Beforethe advent of troponin assays, such occur-rences would have been labeled as UA (be-cause of the absence of the CK-MB biomarker)but are now more accurately classified asNSTEMI.

Because troponin and CK-MB levels donot become elevated in the serum until atleast a few hours after the onset of MI symp-toms, a single normal value drawn early inthe course of evaluation (e.g., in the hospi-tal emergency department) does not ruleout an acute MI; thus, the diagnostic utilityof these biomarkers is limited in that criticalperiod. As a result, early decision making in patients with acute coronary syndromesoften relies most heavily on the patient’shistory and ECG findings.

Sometimes the early diagnosis of MI canremain uncertain even after careful evalua-tion of the patient’s history, ECG, and serumbiomarkers. In such a situation, an additionaldiagnostic study that may be useful is echo-cardiography, which typically reveals abnor-malities of ventricular contraction in the re-gion of ischemia or infarction.

TREATMENT OF ACUTECORONARY SYNDROMES

Successful management of ACS requires rapidinitiation of therapy to limit myocardialdamage and minimize complications. Ther-apy must address the intracoronary throm-bus that incited the syndrome and provideanti-ischemic measures to restore the bal-ance between myocardial oxygen supplyand demand. Although certain therapeuticaspects are common to all ACS, there is acritical difference in the approach to patientswho present with ST segment elevation(STEMI) compared with those without STsegment elevation (UA and NSTEMI). Pa-tients with STEMI typically have total occlu-sion of a coronary artery and benefit fromimmediate reperfusion therapies (pharmaco-logic or mechanical), while patients withoutST elevation do not (Fig. 7.10 and as discussedlater in the chapter).

General in-hospital measures for any pa-tient with ACS include admitting the pa-tient to an intensive care setting wherecontinuous ECG monitoring for arrhyth-

184 Chapter Seven

Symptoms of Acute Coronary Syndrome

ECGST elevation(STEMI)

No ST elevation(UA/ NSTEMI)

1. Aspirin2. Heparin (UFH or LMWH)3. Clopidogrel4. Choose reperfusion method: A. Fibrinolytic drug B. Primary PCI

• with GP IIb/IIIa inhibitor

Reperfusion approach All patients

1. Aspirin2. Heparin (UFH or LMWH)3. Clopidogrel4. For high risk patients:

• GP IIb/IIIa inhibitor • Proceed to cardiac cath

Antithrombotic approach

Figure 7.10. Management strategies in acute coronary syndromes (ACS). The interventions on the left sideshould be considered in patients who present with ST-elevation myocardial infarction (STEMI), while those on theright side are appropriate in unstable angina (UA) and non–ST-elevation MI (NSTEMI). All patients with ACS can ben-efit from the measures listed in the center column. ACE, angiotensin-converting enzyme; ECG, electrocardiogram;GP, glycoprotein; LMWH, low molecular weight heparin; PCI, percutaneous coronary intervention; UFH, unfraction-ated heparin.

Fig. 10

10090-07_CH07.qxd 8/31/06 4:50 PM Page 184


mias is undertaken. The patient is initiallymaintained at bedrest to minimize myo-cardial oxygen demand, while supplemen-tal oxygen is administered (by face maskor nasal cannula), if there is any degree ofhypoxemia, to improve oxygen supply.Analgesics, such as morphine, are admin-istered to reduce chest pain and anxietyand thus further reduce myocardial oxygenneeds.

Acute Treatment of UnstableAngina and Non–ST-ElevationMyocardial Infarction

The management of UA and NSTEMI is es-sentially the same and is therefore discussedas one entity, whereas the approach to STEMIis described later. The primary focus of treat-ment for UA and NSTEMI consists of anti-ischemic medications to restore the balancebetween myocardial oxygen supply and de-mand and antithrombotic therapy aimed atstabilizing the underlying partially occlusivecoronary thrombus.

Anti-ischemic Therapy

The same pharmacologic agents used to de-crease myocardial oxygen demand in chronicstable angina are appropriate in UA andNSTEMI but are often administered moreaggressively. b-Blockers decrease sympa-thetic drive to the myocardium, thus reduc-ing oxygen demand, and contribute to elec-trical stability. This group of drugs reducesthe likelihood of progression from UA to MIand lowers mortality rates in patients whopresent with infarction. In the absence ofcontraindications (e.g., marked bradycardia,bronchospasm, decompensated heart fail-ure, or hypotension), a β-blocker is initiallyadministered intravenously and then con-verted to an oral regimen to achieve a targetheart rate of approximately 60 bpm. Suchtherapy is usually continued indefinitelyafter hospitalization because of proven long-term mortality benefits following an MI.

Nitrates help bring about anginal reliefthrough venodilatation, which lowers myo-

cardial oxygen demand by diminishing ve-nous return to the heart (reduced preloadand therefore less wall stress). Nitrates mayalso improve coronary flow and prevent va-sospasm through coronary vasodilatation.In UA or NSTEMI, nitroglycerin is often ini-tially administered by the sublingual route,followed by a continuous intravenous infu-sion. In addition to providing symptomaticrelief of angina, intravenous nitroglycerin is useful as a vasodilator in patients withACS accompanied by heart failure or severehypertension.

Nondihydropyridine calcium channelantagonists (i.e., verapamil and diltiazem)exert anti-ischemic effects by decreasingheart rate and contractility and throughtheir vasodilatory properties (see Chapter 6).These agents do not confer mortality benefitto patients with ACS and are reserved forthose in whom ischemia persists despite β-blocker and nitrate therapies, or for thosewith contraindications to β-blocker use.They should not be prescribed to patientswith LV systolic dysfunction, because clini-cal trials have shown adverse outcomes insuch cases.

Antithrombotic Therapy

The purpose of antithrombotic therapy, including antiplatelet and anticoagulantmedications, is to prevent further propaga-tion of the partially occlusive intracoronarythrombus while facilitating its dissolutionby endogenous mechanisms. Aspirin in-hibits platelet synthesis of thromboxane A2,a potent mediator of platelet activation, andis one of the most important interventionsto reduce mortality in patients with allforms of ACS. It should be administered im-mediately on presentation and continuedindefinitely in patients without contraindi-cations to its use (e.g., allergy or underlyingbleeding disorder).

Because aspirin blocks only one pathwayof platelet activation and aggregation, otherantithrombotic agents have also been stud-ied. The thienopyridines (clopidogrel andticlopidine) inhibit ADP-mediated activa-tion of platelets (see Chapter 17) and can be

10090-07_CH07.qxd 8/31/06 4:50 PM Page 185

used as a substitute in aspirin-allergic pa-tients. In addition, the combination of as-pirin plus clopidogrel is superior to aspirinalone in reducing cardiovascular mortality,recurrent cardiac events, or stroke in pa-tients with UA or NSTEMI.

Intravenous unfractionated heparin(UFH), an anticoagulant, is also standardtherapy for UA and NSTEMI. It binds to an-tithrombin III, which greatly increases thepotency of the plasma protein in the inacti-vation of clot-forming thrombin. UFH addi-tionally inhibits coagulation factor Xa, slow-ing thrombin formation and thereby furtherimpeding clot development. In patients withUA or NSTEMI, UFH improves cardiovascu-lar outcomes and reduces the likelihood ofprogression from UA to MI. It is adminis-tered as a weight-based bolus, followed bycontinuous infusion, with adjustment in itsdosage determined by measurements of theserum activated partial thromboplastin time(aPTT).

To overcome the shortcomings of UFH,low molecular weight heparins (LMWHs)have been developed. Like UFH, LMWHs in-teract with antithrombin III but preferen-tially inhibit coagulation factor Xa. LMWHsprovide a more predictable pharmacologicresponse than UFH. As a result, LMWHs areeasier to use, prescribed as one or two dailysubcutaneous injections based on the pa-tient’s weight. Unlike UFH, repeated moni-toring of blood tests and dosage adjust-ments are not necessary. In clinical trials inpatients with UA or NSTEMI, the LMWHenoxaparin (see Chapter 17) provides clinicaloutcomes and safety comparable, if not su-perior, to UFH.

The glycoprotein (GP) IIb/IIIa receptorantagonists (which include the mono-clonal antibody abciximab and the small-molecule eptifibatide and tirofiban) are po-tent antiplatelet agents that block the finalcommon pathway of platelet aggregation(see Chapter 17). These agents are very ef-fective in reducing cardiac events in patientsundergoing percutaneous coronary inter-vention (PCI). Their benefit in patients withUA or NSTEMI who are managed pharma-cologically is more modest, generally con-

fined to high-risk patients (e.g., those withelevated troponin levels or who experiencerecurrent episodes of chest pain).

Conservative Versus Early InvasiveManagement of UA and NSTEMI

Many patients with UA or NSTEMI stabilizefollowing institution of the therapies de-scribed in previous section, while othersprogress to a more severe form of ACS. Thereis currently no definitive way to predictwhich path a patient will take or to quicklydetermine which patients have such severeunderlying CAD that revascularization iswarranted. These uncertainties have led totwo diagnostic strategies in UA/NSTEMImanagement: (1) an early invasive approach,in which urgent cardiac catheterization isperformed and coronary revascularizationundertaken as indicated, or (2) a conservativeapproach, in which the patient is managedwith medications (as detailed in the previoussection) and undergoes angiography only ifischemic episodes spontaneously recur or ifthe results of a subsequent stress test indicateresidual inducible ischemia. The conservativeapproach offers the advantage of avoidingcostly and potentially risky invasive proce-dures. On the other hand, an early invasivestrategy allows rapid identification and de-finitive treatment (i.e., revascularization) forthose with critical coronary disease.

Although early randomized trials com-paring these two approaches suggested thateach leads to comparable outcomes, theweight of more recent data (in the era ofcoronary stents and platelet GP IIb/IIIa re-ceptor antagonists) indicate that an early in-vasive approach results in superior outcomeswith lower rates of recurrent angina and re-infarction. An early invasive approach ap-pears to be most beneficial in patients withhigh-risk features, such as those with ST seg-ment deviations at the time of presentation,elevated serum troponin levels, and the pres-ence of multiple cardiac risk factors. As inva-sive techniques and pharmacologic thera-pies improve over time, the optimal strategyin patients with UA or NSTEMI continues tobe an area of active investigation.

186 Chapter Seven

AQ1

10090-07_CH07.qxd 8/31/06 4:50 PM Page 186


Acute Treatment of ST-ElevationMyocardial Infarction

In contrast to UA and NSTEMI, the culpritartery in STEMI is typically completely oc-cluded. Thus, to limit myocardial damage,the major focus of acute treatment for suchpatients is to achieve prompt reperfusion ofthe jeopardized myocardium using either fib-rinolytic drugs or percutaneous coronaryrevascularization. These approaches reducethe extent of myocardial necrosis and greatlyimprove survival. To be effective, they mustbe undertaken as soon as possible; the earlierthe intervention occurs, the greater theamount of myocardium that can be sal-vaged. Decisions about therapy must bemade within minutes of a patient’s arrival tothe hospital, based on the history and elec-trocardiographic findings, often before serummarkers of necrosis would be expected to rise.

In addition, as is the case in UA andNSTEMI, specific medications should be ini-tiated promptly to prevent further thrombo-sis and to restore the balance between my-ocardial oxygen supply and demand. Forexample, antiplatelet therapy with aspirinalso decreases mortality rates and rates of reinfarction after STEMI. It should be ad-ministered immediately on presentation andcontinued daily thereafter. Intravenous un-fractionated heparin is typically infused tohelp maintain patency of the coronary vesseland is an important adjunct to modern fibri-nolytic regimes. b-Blockers reduce myo-cardial oxygen demand and lower the risk ofrecurrent ischemia, arrhythmias, and rein-farction. In the absence of contraindications(e.g., asthma, hypotension, or significantbradycardia), a β-blocker should be adminis-tered, usually intravenously at first and thenorally. Nitrate therapy, usually intravenousnitroglycerin, is used to help control ischemicpain and also serves as a beneficial vasodilatorin patients with heart failure or severe hyper-tension during acute infarction.

Fibrinolytic Therapy

Fibrinolytic drugs accelerate lysis of the oc-clusive intracoronary thrombus in STEMI,

thereby restoring blood flow and limitingmyocardial damage. Please note that thisdiscussion does not pertain to patients withUA or NSTEMI, who do not benefit from fib-rinolytic therapy.

Currently used fibrinolytic agents includethe recombinant tissue-type plasminogenactivator alteplase (tPA), reteplase (rPA),and tenecteplase (TNK-tPA). Streptokinase,one of the earliest studied fibrinolytics, is nolonger in widespread use in the UnitedStates. Each drug functions by stimulatingthe natural fibrinolytic system, transformingthe inactive precursor plasminogen into theactive protease plasmin, which lyses fibrinclots. Although the intracoronary thrombusis the target, plasmin has poor substratespecificity and can degrade other proteins,including fibrinogen. As a result, bleeding isthe most common complication of thesedrugs. However, unlike the older fibrinolyticstreptokinase, the newer agents bind prefer-entially to fibrin in a formed thrombus (i.e.,the intracoronary clot), thereby generatingplasmin locally at that site, with less inter-ference of coagulation in the general circu-lation (Fig. 7.11). Nonetheless, bleeding re-mains the most important risk with allfibrinolytic agents.

rPA and TNK-tPA are derivatives of tPAwith longer half-lives. Their main advantageis that they can be administered as IV boluses,which is more convenient and less prone toincorrect administration than the continuousintravenous infusion necessary for tPA.

Administration of fibrinolytic agents inthe early hours of an acute STEMI restoresblood flow in most (70% to 80%) of coro-nary occlusions and significantly reducesthe extent of tissue damage. Improved arterypatency translates into substantially in-creased survival rates and fewer postinfarc-tion complications. The rapid initiation offibrinolysis is crucial: patients who receivetherapy within 2 hours of the onset of symp-toms of STEMI have half the mortality rate ofthose who receive it after 6 hours.

Successful reperfusion is marked by the re-lief of chest pain, return of the ST segments tobaseline, and earlier-than-usual peaking ofserum markers of necrosis, such as cardiac-

Fig. 11

10090-07_CH07.qxd 8/31/06 4:50 PM Page 187

specific troponins and CK-MB. During reper-fusion, transient arrhythmias are commonand do not usually require treatment. To pre-vent immediate vessel reocclusion after suc-cessful thrombolysis, antithrombotic regimensare administered, as described in the nextsection.

Because the major risk of thrombolysis isbleeding, contraindications to such therapyinclude situations in which necessary fibrinclots within the circulation would be jeopar-dized (e.g., patients with active peptic ulcerdisease or an underlying bleeding disorder,patients who have had a recent stroke or whoare recovering from recent surgery). Conse-quently, approximately 30% of patients maynot be suitable candidates for thrombolysis.

Several large-scale comparisons of fibri-nolytic agents have been conducted. In 1993,the international GUSTO-1 trial found a small

postinfarction survival advantage of tPA com-pared with streptokinase, at the expense of aslightly increased risk of intracranial hemor-rhage with tPA. More recent trials comparedtPA with the newer agents rPA and TNK-tPAand found similar clinical efficacies for allthree agents. The most important messagefrom these trials is that early and sustainedpatency of the infarct-related coronary arteryimproves survival. No matter which fibri-nolytic is selected, it must be administered assoon as possible, ideally within 30 minutes ofthe patient’s presentation to the hospital.

Adjunctive Antithrombotic TherapiesAfter Fibrinolysis

As previously indicated, aspirin is a main-stay of therapy in patients with ACS and istypically initiated on the patient’s presen-

188 Chapter Seven

Fibrin clot

Fibrin clot

tPA P tPA P*

P*

P*

P*P

P

Systemic lyticstate

Degrade clot withoutsystemic lytic state

SK

SK

A

B

Figure 7.11. Examples of fibrinolytic agents used in ST-elevation myo-cardial infarction. A. Tissue plasminogen activator (tPA) cleaves fibrin-boundplasminogen (P) to form active plasmin (P*), which degrades the fibrin clot. Theselectivity of tPA for fibrin-bound P results in localized thrombolysis and mini-mizes generalized systemic fibrinolysis. TNK-tPA and rPA (see text) act similarlyto tPA but can be administered as boluses, thus simplifying drug administra-tion. B. The older fibrinolytic streptokinase (SK) combines with fibrin-boundand circulating plasminogen to form an active complex, which in turn activatesadditional plasminogen molecules. The lack of selectivity for fibrin-bound plas-minogen results in more of a systemic lytic state.

10090-07_CH07.qxd 8/31/06 4:50 PM Page 188


tation. Anticoagulants administered withfibrinolytic therapy in STEMI enhance clotlysis and reduce reocclusion rates. Thus, for patients treated with tPA, rPA, or TNK-tPA, adjunctive IV unfractionated heparinshould be administered for up to 48 hours.Low molecular weight heparins havebeen tested as an alternative to unfraction-ated heparin and have been shown to re-duce ischemic complications but at an in-creased risk of intracranial hemorrhage inolder patients.

The antiplatelet agent clopidogrel, ad-ministered in combination with aspirin, hasbeen shown to further reduce mortality andmajor cardiovascular events in STEMI pa-tients who receive fibrinolytic drugs; its ad-ministration should be considered in such pa-tients. Conversely, the antiplatelet GP IIb/IIIareceptor antagonists have not demonstrateda survival benefit in those treated with fibri-nolysis.

Primary Percutaneous Coronary Intervention

An alternative to fibrinolytic therapy in pa-tients with acute STEMI is immediate car-diac catheterization and PCI of the lesionresponsible for the infarction. This ap-proach is termed primary PCI and involvesangioplasty, and usually stenting, of theculprit vessel. Primary PCI is a very effectivemethod for reestablishing coronary perfu-sion and, in clinical trials performed athighly experienced medical centers, hasachieved optimal flow in the infarct-relatedartery in more than 95% of patients. Com-pared with fibrinolytic therapy, primaryPCI leads to greater survival with lower ratesof reinfarction and bleeding. Therefore, pri-mary PCI is usually the preferred reperfusionapproach in acute STEMI, if the procedurecan be performed by an experienced opera-tor in a rapid fashion (within 90 minutes ofhospital presentation). In addition, primaryPCI is preferred for patients who have con-traindications to fibrinolytic therapy or areunlikely to do well with fibrinolysis, includ-ing those who present late (>3 hours from

symptom onset to hospital arrival) or are incardiogenic shock.

In addition to aspirin and intravenousunfractionated heparin, patients undergo-ing primary PCI typically receive an intra-venous GP IIb/IIIa receptor antagonist inconjunction with the procedure to reducethrombotic complications. Also, for mostpatients who receive coronary stents duringPCI, the oral thienopyridines (e.g., clopido-grel) have been shown to reduce the risk ofischemic complications and stent thrombo-sis. Clopidogrel is therefore typically begunin association with the procedure and thencontinued for weeks to months, dependingon the type of stent placed.

Adjunctive Therapies

Angiotensin-converting enzyme (ACE) in-hibitors limit adverse ventricular remodel-ing and reduce the incidence of heart failure,recurrent ischemic events, and mortality fol-lowing an MI. Their benefit is additive tothat of aspirin and β-blocker therapies, andthey have shown favorable improvementsespecially in higher-risk patients—those withanterior wall infarctions or LV systolic dys-function.

Cholesterol-lowering statins (HMG-CoAreductase inhibitors) reduce mortality ratesof patients with coronary artery disease (seeChapter 5). Clinical trials of patients withACS have demonstrated that it is safe tobegin statin therapy early during hospital-ization, and that an intensive lipid-loweringregimen, designed to achieve a low-densitylipoprotein (LDL) level <70 mg/dl, providesgreater protection against subsequent car-diovascular events and death than “stan-dard” targets (i.e., achieving an LDL level<100 mg/dL). The benefits of statin therapymay extend beyond lipid lowering, becausethis group of drugs has attributes that canimprove endothelial dysfunction, inhibitplatelet aggregation, and impair thrombusformation.

In addition to the short-term use of he-parin anticoagulation described earlier, amore prolonged course followed by oral an-ticoagulation (i.e., warfarin) is appropriate

AQ2

AQ3

10090-07_CH07.qxd 8/31/06 4:50 PM Page 189

for patients at high risk of thromboem-bolism—for example, patients with docu-mented intraventricular thrombus (typi-cally identified by echocardiography) oratrial fibrillation and those who have suf-fered a large acute anterior MI with akinesisof that territory (which is susceptible tothrombus formation because of the stagnantblood flow).

COMPLICATIONS

In unstable angina, the potential complica-tions include death (5% to 10%) or progres-sion to infarction (10% to 20%) over the en-suing days and weeks. Once infarction hastranspired, especially STEMI, complicationscan result from the inflammatory, mechan-ical, and electrical abnormalities induced by regions of necrosing myocardium (Fig.7.12). Early complications result from my-ocardial necrosis itself. Those that developseveral days to weeks later reflect the in-flammation and healing of necrotic tissue.

Recurrent Ischemia

Postinfarction angina has been reported tooccur in 20% to 30% of patients followingan MI. This rate has not been reduced by theuse of thrombolytic therapy, but it is lowerin those who have undergone percutaneousangioplasty or coronary stent implantationas part of early MI management. Indicativeof inadequate residual coronary blood flow,it is a poor omen and correlates with an in-creased risk for reinfarction. Such patientsusually require urgent cardiac catheteriza-tion, often followed by revascularization bypercutaneous techniques or coronary arterybypass surgery.

Arrhythmias

Arrhythmias occur frequently during acuteMI and are a major source of mortality priorto hospital arrival. Modern coronary careunits are highly attuned to the detection andtreatment of rhythm disturbances; thus,once a patient is hospitalized, arrhythmia-

190 Chapter Seven

Ventricularthrombus


Embolism

Contractility

Ischemia Hypotension

Coronaryperfusionpressure

Cardiogenicshock

Electricalinstability

Tissuenecrosis

Arrhythmias

Papillarymuscle

infarction/ischemia

Mitralregurgitation

Congestive heart failure

Ventricularseptaldefect

Ventricularrupture

Cardiactamponade

Pericarditis

Pericardialinflammation

Figure 7.12. Complications of MI. Infarction results in decreased contractility, electrical instability, and tissuenecrosis, which lead to the indicated sequelae.

Fig. 12

10090-07_CH07.qxd 8/31/06 4:50 PM Page 190


associated deaths are uncommon. Mecha-nisms that contribute to arrhythmogenesisafter MI include the following (Table 7.6):

1. Anatomic interruption of perfusion tostructures of the conduction pathway(e.g., sinoatrial node, atrioventricularnode, bundle branches); the normalperfusion of pertinent components ofthe conduction system is reviewed inTable 7.7.

2. Accumulation of toxic metabolic prod-ucts (e.g., cellular acidosis) and abnormaltranscellular ion concentrations owing tomembrane leaks.

3. Autonomic stimulation (sympathetic andparasympathetic).

4. Administration of potentially arrhythmo-genic drugs (e.g., dopamine).

Ventricular Fibrillation

Ventricular fibrillation (rapid, disorganizedelectrical activity of the ventricles) is largelyresponsible for episodes of sudden cardiacdeath during the course of acute MI. Mostfatal episodes occur before hospital arrival, atrend that will hopefully improve with in-creasing availability of automatic externaldefibrillators in public places. Episodes ofventricular fibrillation that occur during thefirst 48 hours of MI are often related to tran-sient electrical instability, and the long-termprognosis of survivors of such events is notaffected. However, ventricular fibrillationoccurring later than 48 hours after the acuteMI usually reflects severe LV dysfunctionand is associated with high subsequent mor-tality rates.

Ventricular ectopic beats, ventricular ta-chycardia, and ventricular fibrillation dur-ing an acute MI arise from either reentrantcircuits or enhanced automaticity of ven-tricular cells (see Chapter 11). Ventricularectopy is common but usually not treatedunless the abnormal beats become conse-cutive, multifocal, or frequent. Intravenous

TABLE 7.6. Arrhythmias in Acute Myocardial Infarction

Rhythm Cause

Sinus bradycardia • ↑Vagal tone• ↓SA nodal artery perfusion

Sinus tachycardia • Pain and anxiety• CHF• Volume depletion• Pericarditis• Chronotropic drugs (e.g., dopamine)

APBs, atrial fibrillation • CHF• Atrial ischemia

VPBs, VT, VF • Ventricular ischemia• CHF

AV block (1°, 2°, 3°) • IMI: ↑vagal tone and ↓AV nodal artery flow• AMI: extensive destruction of conduction tissue

AMI, anterior myocardial infarction; APBs, atrial premature beats; AV, atrioventricular; CHF, congestive heart failure; IMI, in-ferior myocardial infarction; SA, sinoatrial; VPBs, ventricular premature beats; VF, ventricular fibrillation; VT, ventriculartachycardia.

AQ4

AQ5

TABLE 7.7. Blood Supply of the Conduction System

Conduction Pathway Primary Arterial Supply

SA node • RCA (70% of patients)AV node • RCA (85% of patients)Bundle of His • LAD (septal branches)RBB • Proximal portion by LAD

• Distal portion by RCALBBLeft anterior fascicle • LADLeft posterior fascicle • LAD and PDA

AV, atrioventricular; LAD, left anterior descending coronaryartery; LBB, left bundle branch; PDA, posterior descendingartery; RBB, right bundle branch; RCA, right coronary artery;SA, sinoatrial.

Tab. 6

Tab. 7

10090-07_CH07.qxd 8/31/06 4:50 PM Page 191

lidocaine (a class IB antiarrhythmic drug de-scribed in Chapter 17) is effective in prevent-ing ventricular fibrillation in the ischemicsetting but is not indicated prophylacticallyin the management of acute MI because ofits potential side effects and because cardiaccare unit personnel are proficient at arrhy-thmia detection and treatment, should thatcomplication occur.

Supraventricular Arrhythmias

Supraventricular arrhythmias are also com-mon in acute MI. Sinus bradycardia resultsfrom either excessive vagal stimulation orsinoatrial nodal ischemia, usually in the set-ting of an inferior wall MI. Sinus tachycardiaoccurs frequently and may result from manycauses, especially pain and anxiety, heart failure, drug administration (e.g., dopamine),or intravascular volume depletion. Becausesinus tachycardia increases myocardial oxy-gen demand and could exacerbate ischemia,identifying and treating its cause are impor-tant. Atrial premature beats and atrial fibrilla-tion (see Chapter 12) may result from atrialischemia or atrial distention secondary toLV failure.

Conduction Blocks

Conduction blocks (atrioventricular nodalblock and bundle branch blocks) developcommonly in acute MI. They may result fromischemia or necrosis of conduction tracts (seeTable 7.7), or in the case of atrioventricularblocks, may develop transiently because of in-creased vagal tone. Vagal activity may be in-creased because of stimulation of afferentfibers by the inflamed myocardium or as a re-sult of generalized autonomic activation inassociation with the pain of an acute MI.

Myocardial Dysfunction

Congestive Heart Failure

Acute cardiac ischemia results in impairedventricular contractility (systolic dysfunc-tion) and increased myocardial stiffness (diastolic dysfunction), both of which maylead to symptoms of heart failure. In addi-

tion, ventricular remodeling, arrhythmias,and acute mechanical complications of MI(described later in the chapter) may culmi-nate in heart failure. Signs and symptoms ofsuch decompensation include dyspnea, pul-monary rales, and a third heart sound (S3).Treatment consists of standard heart failuretherapy (see Chapter 9).

Cardiogenic Shock

Cardiogenic shock is a condition of severelydecreased cardiac output and hypotension(systolic blood pressure <90 mm Hg) with in-adequate perfusion of peripheral tissues thatdevelops when more than 40% of the LVmass has infarcted. It may also follow certainsevere mechanical complications of MI de-scribed later. Demise in cardiogenic shock isself-perpetuating because (1) hypotensionleads to decreased coronary perfusion, whichexacerbates ischemic damage, and (2) de-creased stroke volume increases LV size andtherefore augments myocardial oxygen de-mand (see Fig. 7.12). Despite aggressivetreatment, the mortality rate of patients incardiogenic shock is greater than 70%.

Patients in cardiogenic shock require in-travenous inotropic agents (e.g., dobuta-mine) to increase cardiac output and arterialvasodilators to reduce the resistance to LVcontraction. These patients are often stabi-lized by the placement of an intra-aortic bal-loon pump. Inserted into the aorta througha femoral artery, the pump consists of an in-flatable, flexible chamber that expands dur-ing diastole to increase intra-aortic pressure,thus augmenting perfusion of the coronaryarteries and the peripheral tissues. Duringsystole, it deflates to create a “vacuum” thatserves to reduce the afterload of the left ven-tricle, thus aiding the ejection of blood intothe aorta. Early cardiac catheterization andrevascularization (PCI or coronary artery by-pass graft) has the potential to improve theprognosis of patients in cardiogenic shock.

Right Ventricular Infarction

Approximately one third of patients with in-farction of the LV inferior wall also develop

192 Chapter Seven

10090-07_CH07.qxd 8/31/06 4:50 PM Page 192


necrosis of portions of the right ventricle, be-cause the same coronary artery (usually theright coronary) perfuses both regions in mostpatients. The resulting abnormal contractionand decreased compliance of the right ven-tricle lead to signs of right-sided heart failure(e.g., jugular venous distention) out of pro-portion to signs of left-sided failure. In addi-tion, profound hypotension may resultwhen right ventricular dysfunction impairsblood flow through the lungs, leading to theleft ventricle becoming underfilled. In thissetting, intravenous volume infusion oftenserves to correct hypotension, guided by he-modynamic measurements via a transvenouspulmonary artery catheter (see Chapter 3).

Mechanical Complications

Mechanical complications following MI re-sult from tissue ischemia and necrosis.

Papillary Muscle Rupture

Ischemic necrosis and rupture of an LV pap-illary muscle may be rapidly fatal because of acute severe mitral regurgitation, whichcauses the valve leaflets to lose their anchor-ing attachments. Partial rupture, with moremoderate regurgitation, is not immediatelylethal but may result in symptoms of heartfailure or pulmonary edema. Because it has amore precarious blood supply, the postero-medial LV papillary muscle is more suscepti-ble to infarction than the anterolateral one;therefore, this complication is more com-mon following an inferoposterior MI.

Ventricular Free Wall Rupture

An infrequent but deadly complication, rup-ture of the LV free wall through a tear in thenecrotic myocardium, may occur within thefirst 2 weeks following MI. It is more com-mon among women and patients with a his-tory of hypertension. Hemorrhage into thepericardial space owing to LV free wall rup-ture results in rapid cardiac tamponade, inwhich blood fills the pericardial space and se-verely restricts ventricular filling (see Chapter14). Survival is rare.

On occasion, a pseudoaneurysm resultsif rupture of the free wall is incomplete andheld in check by thrombus formation that“plugs” the hole in the myocardium. Thissituation is the cardiac equivalent of a timebomb, because subsequent complete rup-ture into the pericardium and tamponadecould follow. If detected (usually by echo-cardiography), surgical repair may preventan otherwise disastrous outcome.

Ventricular Septal Rupture

This complication is analogous to LV freewall rupture, but the abnormal flow of bloodis not directed across the LV wall into thepericardium. Rather, blood is shunted acrossthe ventricular septum from the left ventri-cle to the right ventricle, usually precipitat-ing congestive heart failure because of sub-sequent volume overload of the pulmonarycapillaries. A loud systolic murmur at the leftsternal border, representing transseptal flow,is common in this situation. Although eachresults in a systolic murmur, ventricular sep-tal rupture can be differentiated from acutemitral regurgitation by the location of themurmur (see Fig. 2.11), by Doppler echocar-diography, or by measuring the O2 satura-tion of blood in the right-sided heart cham-bers through a transvenous catheter. The O2

content in the right ventricle is abnormallyhigher than that in the right atrium if thereis shunting of oxygenated blood from theleft ventricle across the septal defect.

True Ventricular Aneurysm

A late complication of MI, a true ventricularaneurysm occurs weeks to months after theacute event. It develops as the ventricularwall is weakened, but not perforated, by thephagocytic clearance of necrotic tissue, andit results in a localized outward bulge (dysk-inesia) when the residual viable heart mus-cle contracts. Unlike the pseudoaneurysmdescribed earlier, a true aneurysm does notinvolve communication between the LVcavity and the pericardium, so that ruptureand tamponade do not develop. Complica-tions of LV aneurysm include (1) thrombus

10090-07_CH07.qxd 8/31/06 4:50 PM Page 193

formation within this region of stagnantblood flow, serving as a potential source ofemboli to peripheral organs; (2) ventriculararrhythmias associated with the stretchedmyofibers; and (3) heart failure resultingfrom reduced forward cardiac output, be-cause some of the LV stroke volume is“wasted” by filling the aneurysm cavity dur-ing systole.

Clues to the presence of an LV aneurysminclude persistent ST segment elevations onthe ECG weeks after the acute ST-elevationMI and a bulge at the LV border on chest ra-diography. The abnormality can be con-firmed by echocardiography.

Pericarditis

Acute pericarditis may occur in the early (in-hospital) post-MI period as necrosis andneutrophilic infiltrates extend from the my-ocardium to the adjacent pericardium (seeChapter 14). Sharp pain, fever, and a peri-cardial friction rub are often manifest in thissituation and help distinguish pericarditisfrom the discomfort of recurrent myocardialischemia. Anticoagulants are relatively con-traindicated in MI complicated by pericarditisto avoid hemorrhage from the inflamed peri-cardial lining. The frequency of MI-associatedpericarditis has declined since the introduc-tion of acute reperfusion strategies, becausethose approaches limit the extent of myocar-dial damage and inflammation.

Dressler Syndrome

Dressler syndrome is another uncommonform of pericarditis that can occur over thefirst several weeks following hospitaliza-tion for MI. The cause is unclear, but an im-mune process directed against damagedmyocardial tissue is suspected to play arole. The syndrome is heralded by fever,malaise, and sharp, pleuritic chest pain typ-ically accompanied by leukocytosis, an elevated erythrocyte sedimentation rate,and a pericardial effusion. Similar to otherforms of acute pericarditis, Dressler syn-drome generally responds to high-dose as-pirin therapy.

Thromboembolism

Stasis of blood flow in regions of impairedLV contraction after an MI may incite intra-cavity thrombus formation, especially whenthe infarction involves the LV apex, orwhen a true aneurysm has formed. Subse-quent thromboemboli can result in devas-tating infarction of peripheral organs (e.g., acerebrovascular accident, or stroke, causedby embolism to the brain).

RISK STRATIFICATION ANDMANAGEMENT FOLLOWINGMYOCARDIAL INFARCTION

The most important predictor of post-MIoutcome is the extent of LV dysfunction.Other features that portend adverse out-comes include early recurrence of ischemicsymptoms, a large volume of residual myo-cardium still at risk because of severe under-lying coronary disease, and high-grade ven-tricular arrhythmias.

To identify patients at high risk for com-plications who may benefit from cardiaccatheterization and revascularization, exer-cise treadmill testing is often performed (unless the patient has already undergonecatheterization and corrective percutaneousrevascularization for the presenting coro-nary syndrome). Patients with significantlyabnormal results, or those who demonstratean early spontaneous recurrence of angina,are referred for cardiac catheterization to de-fine their coronary anatomy.

Standard postdischarge therapy includeaspirin, a β-blocker, and an HMG-CoA re-ductase inhibitor (statin) to achieve a long-term LDL value of <70 mg/dL. ACE in-hibitors are prescribed to patients who haveLV contractile dysfunction. Rigorous atten-tion to other cardiac risk factors, such assmoking, hypertension, and diabetes, is alsomandatory, and a formal exercise rehabili-tation program often speeds convalescence.

SUMMARY

1. Acute coronary syndromes include UA,NSTEMI, and STEMI. Most ACS episodes

194 Chapter Seven

10090-07_CH07.qxd 8/31/06 4:50 PM Page 194


are caused by the formation of intracoro-nary thrombus at the site of atheroscle-rotic plaque. Plaque rupture is the usualtrigger for thrombus formation throughactivation of platelets and the coagula-tion cascade. Atherosclerosis-induced endothelium dysfunction contributes to the process by producing decreasedamounts of vasodilators and antithrom-botic mediators.

2. Distinctions among types of ACS arebased on the severity of ischemia andwhether myocardial necrosis results. ST-elevation MI is associated with an oc-clusive thrombus and severe ischemiawith necrosis. ACS without ST elevation(non–ST-elevation MI and UA) usually result from partially occlusive thrombiwith less intense ischemia. Comparedwith UA, the insult in NSTEMI is of suffi-cient magnitude to cause some degree ofmyocardial necrosis.

3. ACS result in biochemical and mechani-cal changes that impair systolic contrac-tion, decrease myocardial compliance(diastolic dysfunction), and predispose to arrhythmias. Infarction initiates an inflammatory response that clears nec-rotic tissue and leads to scar formation.Transient severe ischemia without infarc-tion can cause “stunned” myocardium, acondition of contractile dysfunction thatpersists beyond the period of ischemia,with subsequent gradual recovery offunction.

4. The diagnosis of specific ACS relies onthe patient’s history, ECG abnormalities,and appearance of specific biomarkers in the serum (cardiac troponins and/orCK-MB).

5. Acute treatment of UA and NSTEMI in-cludes measures to restore balance be-tween myocardial oxygen supply and demand (β-blockers, nitrates) and stabi-lization of the intracoronary thrombus(aspirin, unfractionated or low molecularweight heparin, and possibly additionalantiplatelet therapies [e.g., thienopy-ridines, GP IIb/IIIa receptor antagonists]).Statin therapy is usually indicated. Earlycoronary angiography, with subsequent

mechanical intervention, is beneficial inhigh-risk patients.

6. Acute treatment for STEMI includes earlyreperfusion strategies with thrombolyticdrugs or PCI. Other important measuresinclude antiplatelet therapy (aspirin,clopidogrel), a β-blocker, and frequently,nitrate and heparin therapies. Often astatin and an ACE inhibitor are also appropriate.

7. Potential complications of infarction include arrhythmias (e.g., ventriculartachycardia and fibrillation), atrioven-tricular blocks, bundle branch blocks,and supraventricular arrhythmias. Car-diogenic shock or congestive heart failuremay develop because of ventricular dys-function or the development of mechan-ical complications (e.g., acute mitral re-gurgitation or ventricular septal defect).In addition, wall motion abnormalitiesof the affected segment may predisposeto thrombus formation.

8. Standard pharmacologic therapy follow-ing discharge from the hospital includesmeasures to reduce the risks of throm-bosis (aspirin, clopidogrel), recurrent is-chemia (a β-blocker), progressive athero-sclerosis (cholesterol-lowering therapy,usually a statin), and adverse ventricularremodeling (an ACE inhibitor, especiallyif LV dysfunction is present). Systemicanticoagulation with warfarin is indi-cated if an intraventricular thrombus, alarge akinetic segment, or atrial fibrilla-tion are present.

9. Post-ACS risk stratification can identifypatients at high risk of recurrent is-chemia, reinfarction, or death. ImpairedLV function, high-grade ventricular ar-rhythmias, and ischemic changes duringexercise testing all portend unfavorableoutcomes and warrant further investiga-tion and treatment.

Acknowledgment

The authors thank Frederick Schoen, MD, for hishelpful suggestions. Contributors to the previous edi-tions of this chapter were J. G. Fletcher, MD; AnuragGupta, MD; Marc S. Sabatine, MD; William Carlson,MD; Patrick T. O’Gara, MD; and Leonard S. Lilly, MD.

10090-07_CH07.qxd 8/31/06 4:50 PM Page 195

Additional Reading

American Heart Association. Stabilization of the pa-tient with acute coronary syndromes. In: 2005American Heart Association Guidelines for Car-diopulmonary Resuscitation and Emergency Car-diovascular Care. Circulation 2005;112:IV-89–IV-110.

Antman EM, Anbe DT, Armstrong PW, et al.ACC/AHA guidelines for the management of pa-tients with ST-elevation myocardial infarction—Executive Summary: a report of the American Col-lege of Cardiology/American Heart AssociationTask Force on Practice Guidelines. Circulation2004;110:588–636.

Boersma E, Harrington RA, Moliterno DJ, et al. Plateletglycoprotein IIb/IIIa inhibitors in acute coronarysyndromes: a meta-analysis of all major randomisedclinical trials. Lancet 2002;359:189–198.

Braunwald E. Application of current guidelines to themanagement of unstable angina and non-ST-elevation myocardial infarction. Circulation 2003;108:III-28–III-37.

Braunwald E, Antman EM, Beasley JW, et al.ACC/AHA guidelines for management of patientswith unstable angina and non-ST-segment eleva-tion myocardial infarction-2002: summary article.Circulation 2002;106:1893–1900.

Bursi F, Enriquez-Sarano M, Jacobsen SJ, et al. Mitralregurgitation after myocardial infarction: a re-view. Am J Med 2006;119:103–112.

Cannon CP, Braunwald E, McCabe CH, et al; Pravas-tatin or Atorvastatin Evaluation and Infection Ther-apy—Thrombolysis in Myocardial Infarction(PROVE-IT) 22 investigators. Intensive versus mod-erate lipid lowering with statins after acute coronarysyndromes. N Engl J Med 2004;350:1495–1504.

Clopidogrel and Metoprolol in Myocardial InfarctionTrial (COMMIT) collaborative group. Addition of clopidogrel to aspirin in 45,852 patients withacute myocardial infarction: randomized placebo-controlled trial. Lancet 2005;366:1607–1621.

de Winter RJ, Windhausen F, Cornel JH, et al. Inva-sive versus Conservative Treatment in UnstableCoronary Syndromes (ICTUS) investigators. Earlyinvasive versus selectively invasive managementfor acute coronary syndromes. N Engl J Med2005;353:1095–1104.

Fox KA, Poole-Wilson P, Clayton TC, et al. 5-yearoutcome of an interventional strategy in non-ST-elevation acute coronary syndrome: the BritishHeart Foundation RITA 3 randomised trial. Lancet2005;366:914–920.

Gershlick AH, Stephens-Lloyd A, Hughes S, et al. Res-cue angioplasty after failed thrombolytic therapyfor acute myocardial infarction. N Engl J Med2005;353:2758–2768.

Giugliano RP. The year in non-ST-segment elevationacute coronary syndromes. J Am Coll Cardiol2005;46:905–919.

Harrington RA, Becker RC, Ezekowitz M, et al. Anti-thrombotic therapy for coronary artery disease:the Seventh ACCP Conference on Antithromboticand Thrombolytic Therapy. Chest 2004;126:513S–548S.

Jaffe AS, Babuin L, Apple FS. Biomarkers in acute car-diac disease: the present and the future. J Am CollCardiol 2006;48:1–11.

Keeley EC, Boura JA, Grines CL. Primary angioplastyversus intravenous thrombolytic therapy for acutemyocardial infarction: a quantitative review of 23randomised trials. Lancet 2003;361:13–20.

Mehta SR, Cannon CP, Fox KAA, et al. Routine vs. se-lective invasive strategies in patients with acutecoronary syndromes: A collaborative meta-analysisof randomized trials. JAMA 2005;293:2908–2917.

Petersen JL, Mahaffey KW, Hasselblad V, et al. Efficacyand bleeding complications among patients ran-domized to enoxaparin or unfractionated heparinfor antithrombin therapy in non-ST-segment ele-vation acute coronary syndromes: a systematicoverview. JAMA 2004;292:89–96.

Ray KK, Cannon CP, McCabe CH, et al. Early andlate benefits of high-dose atorvastatin in patientswith acute coronary syndromes. J Am Coll Cardiol2005;46:1405–1410.

Sabatine MS, Cannon CP, Gibson CM, et al. Clo-pidogrel as Adjunctive Reperfusion Therapy(CLARITY)–Thrombolysis in Myocardial Infarc-tion (TIMI) 28 investigators. Addition of clopi-dogrel to aspirin and fibrinolytic therapy for myocardial infarction with ST-segment eleva-tion. N Engl J Med 2005;352:1179–1189.

196 Chapter Seven

10090-07_CH07.qxd 8/31/06 4:50 PM Page 196


1. AU: Terminology used above. Correct here?2. AU: Terminology used earlier. Correct here?3. AU: Correct term?4. AU: Please provide correct symbol5. AU: Please provide correct symbol

10090-07_CH07.qxd 8/31/06 4:50 PM Page a

197

RHEUMATIC FEVER

MITRAL VALVE DISEASEMitral StenosisMitral RegurgitationMitral Valve Prolapse

AORTIC VALVE DISEASEAortic StenosisAortic Regurgitation

TRICUSPID VALVE DISEASETricuspid StenosisTricuspid Regurgitation

PULMONIC VALVE DISEASEPulmonic StenosisPulmonic Regurgitation

PROSTHETIC VALVES

INFECTIVE ENDOCARDITISPathogenesisClinical Manifestations

C H A P T E R

8Valvular Heart DiseaseMia M. EdwardsPatrick T. O’GaraLeonard S. Lilly

This chapter reviews the pathophysiologicabnormalities in patients with valvular heartdisease. Each of the common valvular con-ditions is discussed separately because uni-fying principles do not govern the behaviorof all stenotic or regurgitant valves. Effectivemanagement of affected patients requiresaccurate identification of the valvular le-sion, a determination of its severity, and aclear understanding of the pathophysiologicconsequences and natural history of thecondition.

The evaluation of a patient with a sus-pected valvular lesion begins at the bedsidewith a careful history and physical exami-nation from which the trained clinician can

usually identify the type of abnormalitiesthat are present. Assessment of the severity ofthe valve lesions can then be facilitated by in-formation from the electrocardiogram, chestradiograph, echocardiogram, and in somecases, cardiac magnetic resonance imaging(see Chapter 3). In selected patients, furtherinvestigation with exercise testing or cardiaccatheterization may be necessary to definefully the significance of the condition andguide therapy.

RHEUMATIC FEVER

Acute rheumatic fever (ARF) was once amongthe most common causes of valvular heart

197

10090-08_CH08.qxd 8/31/06 4:51 PM Page 197

disease, but its incidence has waned consider-ably in the past half-century in industrializedsocieties. In the 1940s, the yearly incidenceexceeded 200,000 cases in the United States,whereas the disease is now rare. The declineof this condition immediately preceded or co-incided with the introduction of penicillin, aswell as with improvement of general healthcare and relief from overcrowding. Althoughoccasional local outbreaks occur, a majorresurgence has not been seen in this country.Nevertheless, in developing parts of theworld, ARF continues to be a scourge with ful-minant consequences.

ARF is an inflammatory condition thatprimarily involves the heart, skin, and con-nective tissues. It is a complication of upperrespiratory tract infections caused by groupA streptococci and mainly occurs in chil-dren and young adults. During epidemics,approximately 3% of patients with acutestreptococcal pharyngitis develop ARF 2 to 3 weeks after the initial throat infection. Al-though the pathogenesis remains unknown,it does not involve direct bacterial infectionof the heart. Some proposed mechanismsinclude the elaboration of a toxin by thestreptococci or autoimmune cross-reactivitybetween bacterial and cardiac antigens.

Pathologically, rheumatic carditis (i.e., car-diac inflammation) may affect all three layersof the heart (pericardium, myocardium, andendocardium). Histopathologic examinationoften demonstrates the Aschoff body (Fig.8.1), an area of focal fibrinoid necrosis sur-rounded by inflammatory cells, includinglymphocytes, plasma cells, and macro-phages, that later resolve to form fibrousscar tissue. The most devastating sequelaeresult from inflammatory involvement ofthe valvular endocardium, which leads tochronic rheumatic heart disease character-ized by permanent deformity and impair-ment of one or more cardiac valves. Symp-toms of valvular dysfunction, however,generally do not become manifest until 10 to30 years after ARF has subsided. This latencyperiod may be considerably shorter with themore aggressive disease observed in devel-oping countries.

The most common presenting symptomsof ARF are chills, fever, fatigue, and migra-tory arthralgias. The cardinal symptomsand clinical manifestations of the diseasethat establish the diagnosis are known asthe Jones criteria (Table 8.1). During theacute episode, carditis may be associatedwith tachycardia, decreased left ventricular

198 Chapter Eight

Figure 8.1. Histopathologic appearance of an Aschoff body in acute rheumaticcarditis. Mononuclear inflammatory cells surround a center of focal necrosis. (FromSchoen FJ. The heart. In: Kumar V, Abbas A, Fausto N, eds. Robbins and CotranPathologic Basis of Disease. 7th Ed. Philadelphia: Elsevier Saunders, 2005:593. Withpermission requested.)AQ1

Tab. 1

Fig. 1

10090-08_CH08.qxd 8/31/06 4:51 PM Page 198

Valvular Heart Disease 199

contractility, a pericardial friction rub, atransient murmur of mitral or aortic regur-gitation, or a middiastolic murmur at thecardiac apex (termed the Carey-Coombsmurmur). These transient murmurs likelyreflect turbulent flow across inflamed valveleaflets. Treatment of the acute episode in-cludes the use of high-dose aspirin to re-duce inflammation, penicillin to eliminateresidual streptococcal infection, and ther-apy for complications such as congestiveheart failure and pericarditis.

During the chronic phase of this condi-tion, stenosis or regurgitation of cardiacvalves is common, most often affecting themitral valve. Forty percent of patients withrheumatic heart disease will develop mitralstenosis. An additional 25% will developaortic stenosis or regurgitation in additionto the mitral abnormality. Infrequently, thetricuspid valve is affected as well.

Recurrences of ARF in affected patientscan incite further cardiac damage. Therefore,individuals who have experienced ARFshould receive low-dose penicillin prophy-laxis at least until early adulthood, by whichtime exposure and susceptibility to strepto-coccal infections have diminished.

MITRAL VALVE DISEASE

Mitral Stenosis

Etiology

The most common cause of mitral stenosis(MS) is rheumatic fever. Approximately 50%of patients with symptomatic MS provide ahistory of ARF occurring, on average, 20 yearsbefore presentation. These patients displaytypical rheumatic deformity of the valve onechocardiographic and pathologic examina-tions as described below. Other rare causesof MS (less than 1%) include congenitalstenosis of the mitral valve leaflets, promi-nent calcification extending from the mitralannulus onto the leaflets in elderly patients,or endocarditis with very large vegetationsthat obstruct the valve orifice.

Pathology

Acute and recurrent inflammation producethe typical pathologic features of rheumaticMS. These include fibrous thickening and cal-cification of the valve leaflets, fusion of thecommissures (the borders where the leafletsmeet), and thickening and shortening of thechordae tendineae.

Pathophysiology

In early diastole in the normal heart, the mi-tral valve opens and blood flows freely fromthe left atrium (LA) into the left ventricle(LV), such that there is a negligible pressuredifference between the two chambers. InMS, however, there is obstruction to bloodflow across the valve such that emptying ofthe LA is impeded and there is an abnormalpressure gradient between the LA and LV(Figs. 8.2 and 8.3). As a result, the left atrialpressure is higher than normal, a necessaryfeature for blood to be propelled forwardacross the obstructed valve. The normalcross-sectional area of the mitral valve ori-fice is 4 to 6 cm2. Hemodynamically signifi-cant MS becomes apparent when the valvearea is reduced to less than 2 cm2. Althoughleft ventricular pressures are usually normalin MS, impaired filling of the chamber across

TABLE 8.1. Jones Criteria for Diagnosis ofRheumatic Fevera

Major criteriaCarditisPolyarthritisSydenham chorea (involuntary movements)Erythema marginatum (skin rash with advancing

edge and clearing center)Subcutaneous nodulesMinor criteriaMigratory arthralgiasFeverIncreased acute phase reactants (ESR, leukocytosis)Prolonged PR interval on electrocardiogramEvidence of streptococcal infectionAntistreptolysin O antibodiesPositive throat culture for streptococci group A

aDiagnosis requires evidence of streptococcal infection and either: two major criteria, or one major plus two minorcriteria.

ESR, erythrocyte sedimentation rate.

Fig. 2-3

10090-08_CH08.qxd 8/31/06 4:51 PM Page 199

200 Chapter Eight

Elevated pulmonaryand right-heart pressures

LA

LV

Pressure

Volume

NORMAL

(DIASTOLE)

MITRAL

STENOSIS

Aorta

Figure 8.2. Pathophysiology of mitral stenosis. In the normalheart, blood flows freely from the left atrium (LA) into the left ventri-cle (LV) during diastole. In mitral stenosis, there is obstruction to LAemptying. Thus, LA pressure increases, which in turn elevates pul-monary and right-heart pressures.

ECG

OS

Figure 8.3. Hemodynamic profile of mitral stenosis. The left atrial (LA) pressure iselevated, and there is a pressure gradient (shaded area) between the LA and left ven-tricle (LV) during diastole. Compare with schematic of normal tracing (see Fig. 2.1). Ab-normal heart sounds are present: there is a diastolic opening snap (OS) that corre-sponds to the opening of the mitral valve, followed by a decrescendo murmur. Thereis accentuation of the murmur just before S1 owing to the increased pressure gradientwhen the LA contracts (presystolic accentuation).

10090-08_CH08.qxd 8/31/06 4:51 PM Page 200


the narrowed mitral valve may reduce LVstroke volume and cardiac output.

The high left atrial pressure in MS is pas-sively transmitted to the pulmonary circula-tion, resulting in increased pulmonary ve-nous and capillary pressures (see Fig. 8.2).This elevation of hydrostatic pressure in thepulmonary vasculature may cause transuda-tion of plasma into the lung interstitium andalveoli. The patient may therefore experiencedyspnea and other symptoms of congestiveheart failure. In severe cases, significant ele-vation of pulmonary venous pressure leads tothe opening of collateral channels betweenthe pulmonary and bronchial veins. Subse-quently, the high pulmonary vascular pres-sures may rupture a bronchial vein into thelung parenchyma, resulting in coughing upof blood (hemoptysis).

The elevation of left atrial pressure in MScan result in two distinct forms of pul-monary hypertension: passive and reactive.Most patients with MS exhibit passive pul-monary hypertension, related to the back-ward transmission of the elevated LA pres-sure into the pulmonary vasculature. Thisactually represents an “obligatory” increasein pulmonary artery pressure that developsto preserve forward flow in the setting of in-creased left atrial and pulmonary venouspressures. Additionally, approximately 40%of patients with MS demonstrate reactive pul-monary hypertension with medial hypertro-phy and intimal fibrosis of the pulmonaryarterioles. Reactive pulmonary hypertensioncan be “beneficial” because the increased ar-teriolar resistance impedes blood flow intothe engorged pulmonary capillary bed andthus reduces capillary hydrostatic pressure(thereby “protecting” the pulmonary capil-laries from even higher pressures). However,this benefit is at the cost of decreased bloodflow through the pulmonary vasculaturewith resultant elevation of the right-sidedheart pressures, as the right ventricle pumpsagainst the increased resistance. Chronic el-evation of right ventricular pressure leads tohypertrophy and dilatation of that chamberand ultimately to right-sided heart failure.

Chronic pressure overload of the leftatrium in MS leads to left atrial enlargement.

Left atrial dilatation stretches the atrial con-duction fibers and may disrupt the integrityof the cardiac conduction system, resultingin atrial fibrillation (a rapid irregular heartrhythm; see Chapter 12). Atrial fibrillationcauses the cardiac output to fall further inMS because the increased heart rate shortensdiastole. This reduces the time available forblood to flow across the obstructed mitralvalve to fill the left ventricle and also resultsin markedly elevated left atrial pressure.

The relative stagnation of blood flow inthe dilated left atrium in MS, especiallywhen combined with the development ofatrial fibrillation, predisposes to intra-atrialthrombus formation. Thromboemboli to pe-ripheral organs may follow, leading to dev-astating complications such as cerebrovas-cular occlusion (stroke). The likelihood ofdeveloping systemic thromboembolic com-plications in a patient with MS correlateswith the patient’s age and the dimensions ofthe left atrial appendage (a portion of theleft atrium); it correlates inversely with thepatient’s cardiac output. Patients who de-velop atrial fibrillation are at very high riskof stroke and require chronic anticoagula-tion therapy.

The turbulent blood flow across the ob-structed mitral valve in MS predisposes to in-fective endocarditis (discussed later); however,that complication occurs less frequently inMS than in other forms of acquired valvulardisease.

Clinical Manifestations and Evaluation

Presentation

The natural history of MS is variable. The 10-year survival of untreated patients after onsetof symptoms is 50% to 60%. Survival exceeds80% in asymptomatic or minimally sympto-matic patients at 10 years. Longevity is muchmore limited for patients with advancedsymptoms and is dismal for those who de-velop significant pulmonary hypertension,with a mean survival less than 3 years.

The clinical presentation of MS dependslargely on the degree of reduction in valvearea. The more severe the stenosis, the greater

10090-08_CH08.qxd 8/31/06 4:51 PM Page 201

the symptoms related to elevation of leftatrial and pulmonary venous pressures. Theearliest manifestations are those of dyspneaand reduced exercise capacity. In mild MS,dyspnea may be absent at rest; however, itdevelops on exertion as LA pressure riseswith the exercise-induced increase in bloodflow through the heart and faster heart rate(i.e., decreased diastolic filling time). Otherconditions and activities that increase heartrate and cardiac blood flow, and thereforeprecipitate or exacerbate symptoms of MS,include fever, anemia, hyperthyroidism,pregnancy, rapid arrhythmias such as atrialfibrillation, exercise, emotional stress, andsexual intercourse.

With more severe MS (i.e., a smaller valvearea) dyspnea occurs even at rest. Increasingfatigue and more-severe signs of pulmonarycongestion, such as orthopnea and paroxys-mal nocturnal dyspnea, occur. With ad-vanced MS and pulmonary hypertension,signs of right-sided heart failure ensue, in-cluding jugular venous distention, hepato-megaly, ascites, and peripheral edema. Com-pression of the recurrent laryngeal nerve bythe enlarged pulmonary artery or left atriummay cause hoarseness.

Less often, the diagnosis of MS is heraldedby one of its complications: atrial fibrilla-tion, thromboembolism, infective endocardi-tis, or hemoptysis, as described in the earliersection on pathophysiology.

Examination

On examination, there are several typicalfindings of MS. Palpation of the left anteriorchest may reveal a right ventricular “tap” inpatients with increased right ventricular pres-sure. Auscultation discloses a loud S1 (the firstheart sound, which is associated with mitralvalve closure) in the early stages of the dis-ease. The increased S1 results from the highpressure gradient between the atrium andventricle, which keeps the mobile portions ofthe mitral valve leaflets widely separatedthroughout diastole; at the onset of systole,ventricular contraction abruptly slams theleaflets together from the relatively wide po-

sition, causing the closure sound to be loud(see Chapter 2). In late stages of the disease,the intensity of S1 may normalize or becomereduced as the valve leaflets thicken, calcify,and become immobile.

A main feature of auscultation in MS is ahigh-pitched “opening snap” (OS) that fol-lows S2. The OS is thought to result from thesudden tensing of the chordae tendineaeand stenotic leaflets on opening of thevalve. The interval between S2 and the OS re-lates inversely to the severity of MS. Themore severe the MS, the higher the LA pres-sure and the earlier the valve is forced openin diastole. The OS is followed by a low-frequency decrescendo murmur (termed adiastolic rumble) caused by turbulent flowacross the stenotic valve during diastole (seeFig. 8.3). The duration, but not the inten-sity, of the diastolic murmur relates to theseverity of MS. The more severe the stenosis,the longer it takes for the LA to empty andfor the gradient between the LA and LV todissipate. Near the end of diastole, contrac-tion of the LA causes the pressure gradientbetween the LA and LV to rise again (see Fig.8.3); therefore, the murmur briefly becomeslouder (termed presystolic accentuation). Thisfinal accentuation of the murmur does notoccur if atrial fibrillation has developed, be-cause there is no effective atrial contractionin that situation.

Murmurs caused by other valvular lesionsare often found concurrently in patientswith MS. For example, mitral regurgitation(discussed later in the chapter) frequentlycoexists with MS. Additionally, right-sidedheart failure caused by severe MS may in-duce tricuspid regurgitation as a result of toright ventricular enlargement. A diastolicdecrescendo murmur along the left sternalborder may be present owing to coexistentaortic regurgitation (because of rheumaticinvolvement of the aortic leaflets) or pul-monic regurgitation (because of MS-inducedpulmonary hypertension).

The electrocardiogram in MS routinelyshows left atrial enlargement and, if pul-monary hypertension has developed, rightventricular hypertrophy. Atrial fibrillation

202 Chapter Eight

1 LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 202


may be present. The chest radiograph revealsleft atrial enlargement, pulmonary vascularredistribution, interstitial edema, and KerleyB lines resulting from edema within the pul-monary septae (see Chapter 3). With the development of pulmonary hypertension,right ventricular enlargement and promi-nence of the pulmonary arteries also appear.

Echocardiography is of major diagnosticvalue in MS. It reveals thickened mitralleaflets and abnormal fusion of their com-missures with restricted separation duringdiastole. Left atrial enlargement can be as-sessed, and if present, intra-atrial throm-bus may be visualized. The mitral valve areacan be measured directly on cross-sectionalviews or calculated from Doppler-echocar-diographic velocity measurements. Patientscan be stratified into groups of disease sever-ity based partly on the mitral valve area. Anormal mitral valve orifice measures be-tween 4 and 6 cm2. A reduced mitral valvearea of ≤2 cm2 correlates with mild MS, anda valve area of 1 to 1.5 cm2 correlates withmoderate MS. Severe MS is defined by a valvearea of ≤1 cm2. Although cardiac catheteri-zation is not necessary to confirm the diag-nosis of MS, it is sometimes performed tocalculate the valve area by direct hemody-namic measurements and to clarify whethersignificant mitral regurgitation, pulmonaryhypertension, or coronary artery disease ispresent.

Treatment

Therapy of MS includes long-term penicillinto prevent recurrent ARF in young peopleand, in all patients, intermittent antibioticprophylaxis against infective endocarditis(discussed later in the chapter). Diuretics areprescribed to treat symptoms of vascularcongestion. If atrial fibrillation has devel-oped, digoxin may be useful to slow therapid ventricular rate and thereby improve di-astolic LV filling (see Chapter 17). β-Blockers,or the calcium channel antagonists vera-pamil or diltiazem, are even more effectivealternatives to slow the heart rate. Anticoag-ulant therapy (to prevent thromboembolism)

is recommended for patients with MS withatrial fibrillation or if previous embolicepisodes have occurred.

If symptoms of MS persist despite diuretictherapy and control of rapid heart rates, me-chanical correction of the stenosis is war-ranted. Percutaneous balloon mitral valvulo-plasty was first introduced in 1985 and isnow the treatment of choice for MS in ap-propriately selected patients. During thisprocedure, a balloon catheter is advancedfrom the femoral vein into the right atrium,across the atrial septum (by creating a smallseptal defect there), and advanced throughthe narrowed mitral valve orifice. The bal-loon is then rapidly inflated, thereby “crack-ing” open the fused commissures. The pro-cedure is safest and most effective in theabsence of complicating features, such asmitral regurgitation, extensive valve calcifi-cation, or atrial thrombus. The results of thisprocedure in randomized trials compare fa-vorably with those of surgical treatment.Approximately 5% of patients undergoingballoon mitral valvuloplasty are left with a residual atrial septal defect. Less frequentcomplications include cerebral emboli atthe time of valvuloplasty, cardiac perfora-tion, or the creation of mitral regurgitationrequiring subsequent surgical repair. The estimated event-free survival 7 years after val-vuloplasty is 67% to 76%.

Surgical options are undertaken for cor-recting MS in individuals whose anatomy isnot ideal for balloon valvuloplasty. Thesetechniques include open mitral commissu-rotomy (an operation in which the stenoticcommissures are separated under direct visualization), and in severe disease, mitralvalve replacement. Open MV commissu-rotomy is effective, and restenosis occurs in <20% of patients over 10 to 20 years offollow-up.

Mitral Regurgitation

Etiology

Normal closure of the mitral valve duringsystole requires the coordinated action ofLINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 203

each component of the valve apparatus.Therefore, mitral regurgitation (MR) may re-sult from structural abnormalities of the mi-tral annulus, the valve leaflets, the chordaetendineae, or the papillary muscles (Fig. 8.4).For example, myxomatous degeneration ofthe valve (the etiology of mitral valve pro-lapse) causes MR because enlarged, redun-dant leaflets bow excessively into the LAduring systole rather than opposing eachother normally. Infective endocarditis canresult in MR because of leaflet perforation orrupture of infected chordae. Rheumaticfever may lead to MS, as already discussed,or primarily MR if excessive shortening ofthe chordae tendineae and retraction of theleaflets occur. Hypertrophic obstructive car-diomyopathy (see Chapter 10) is associatedwith abnormal systolic anterior motion ofthe anterior mitral leaflet, which preventsnormal valve closure and results in signifi-cant MR in 50% of patients. Calcification ofthe mitral annulus can occur with normalaging but is more common among patientswith hypertension, diabetes, or end-stagerenal disease. Such calcification impairs thenormal movement of the annulus and im-mobilizes the basal portion of the valve

leaflets, interfering with their excursion andsystolic closure.

Primary (idiopathic) rupture of chordaetendineae is associated with severe acutevalvular incompetence. Ischemic heart dis-ease may scar or cause transient dysfunctionof a papillary muscle, interfering with valveclosure. Marked left ventricular enlargementof any cause results in MR because of twomechanisms that interfere with mitral leafletclosure: (1) the spatial separation between thepapillary muscles is augmented, and (2) themitral annulus is stretched to an increased diameter.

During the 1990s, the commonly usedweight-loss drug combination of fenflu-ramine and phentermine was, in some pa-tients, associated with the development ofthickened plaques on the cardiac valves.These patients were prone to develop MR, aswell as aortic regurgitation and tricuspidvalve disease. Thus, that drug combinationis no longer used.

Pathophysiology

In MR, a portion of the left ventricular strokevolume is ejected backward into the low-

204 Chapter Eight

Left atriumMitral annulus

•

Leaflets

••••

Annular calcification

Myxomatous degeneration (“MVP”)Rheumatic diseaseEndocarditisSAM (hypertrophic cardiomyopathy)

Chordae tendineae

••

•

Rupture (idiopathic)Endocarditis

Papillary muscles

Dysfunction or rupture

•

Left ventricle

Cavity dilatation

Figure 8.4. The mitral valve apparatus and associated common etiologies ofmitral regurgitation. MVP, mitral valve prolapse; SAM, systolic anterior motion.

Fig. 4

10090-08_CH08.qxd 8/31/06 4:51 PM Page 204


pressure LA (Fig. 8.5) during systole. As a re-sult, the forward cardiac output (into theaorta) is less than the lLV’s total output (for-ward flow plus + backward leak). Therefore,the direct consequences of MR include (1)an elevation of the left atrial volume andpressure, (2) a reduction of forward cardiacoutput, and (3) a volume-related stress onthe LV because the regurgitated volume re-turns to the LV in diastole along with thenormal pulmonary venous return. To meetnormal circulatory needs and to eject theadditional volume, LV stroke volume mustrise. This increase is accomplished by theFrank-Starling mechanism (see Chapter 9),whereby the elevated LV diastolic volumeaugments myofiber stretch and stroke vol-ume with each contraction. The subsequenthemodynamic consequences of MR vary de-pending on the degree of regurgitation andhow long it has been present.

The severity of MR and the ratio of for-ward cardiac output to backward flow aredictated by five factors: (1) the size of themitral orifice during regurgitation, (2) thesystolic pressure gradient between the LV

and LA, (3) the systemic vascular resistanceopposing forward LV blood flow, (4) the leftatrial compliance, and (5) the duration of re-gurgitation with each systolic contraction.

The regurgitant fraction in MR is defined asfollows:

and this ratio rises whenever the resistanceto aortic outflow is increased (i.e., blood fol-lows the path of least resistance). For exam-ple, high systemic blood pressure or thepresence of aortic stenosis will increase theregurgitant fraction. The extent to which leftatrial pressure rises in response to the regur-gitant volume is determined by the left atrialcompliance. Compliance is a measure of thechamber’s pressure-volume relationship, re-flecting the ease or difficulty with which thechamber can be filled (see Table 9.1).

In acute MR (e.g., owing to sudden rup-ture of chordae tendineae), left atrial com-pliance undergoes little immediate change.Because the LA is a relatively stiff chamber,its pressure increases substantially when it

Volume� of� MRTotal� LV� stroke� volume

Pulmonaryedema

High LApressure

Dilated LAwith lesselevatedpressure

Figure 8.5. Pathophysiology of mitral regurgitation. In the normal heart, left ventricular (LV) contractionduring systole forces blood exclusively through the aortic valve into the aorta; the closed mitral valve preventsregurgitation into the left atrium (LA). In mitral regurgitation (MR), a portion of LV output is forced backwardinto the LA, so that forward cardiac output into the aorta is reduced. In acute MR, the LA is of normal size andis relatively noncompliant, such that the LA pressure rises significantly and pulmonary edema may result. Inchronic MR, the LA has enlarged and is more compliant, so that LA pressure is less elevated and pulmonarycongestive symptoms are less common. LV enlargement and eccentric hypertrophy result from the chronicallyelevated volume load.

Fig. 5

10090-08_CH08.qxd 8/31/06 4:51 PM Page 205

is suddenly exposed to the regurgitant vol-ume (see Fig. 8.5). This elevated pressureserves to prevent further regurgitation; how-ever, the high pressure is also transmittedbackward to the pulmonary circulation.Therefore, acute MR can result in rapid pul-monary congestion and edema, a medicalemergency.

In acute MR, the LA pressure, or the pul-monary capillary wedge pressure (an indi-rect measurement of LA pressure; see Chap-ter 3), demonstrates a prominent v wave(often referred to as a cv wave when it is soprominent that it merges with the preced-ing c wave), reflecting the increased LA filling during systole (Fig. 8.6). Additionally,as in MS, pulmonary artery and right-heartpressures passively rise such that forwardflow through the heart is maintained.

In acute MR, the LV accommodates theincreased volume load returning from theLA according to the Frank-Starling relation-ship. The result is a compensatory increasein the LV stroke volume, such that at theend of each systolic contraction, LV volumeremains normal in the nonfailing heart. Sys-tolic emptying of the ventricle is facilitatedin MR by the reduced total impedance toLV contraction (i.e., the afterload is lowerthan normal), which is caused by a portionof the LV output being directed into the rel-atively low-impedance left atrium, ratherthan into the higher-pressure aorta as innormal outflow.

In contrast to the acute situation, the moregradual development of chronic MR (e.g.,owing to rheumatic valve disease) permitsthe LA to undergo compensatory changesthat lessen the effects of regurgitation onthe pulmonary circulation (see Fig. 8.5). Inparticular, the LA dilates and its complianceincreases such that the chamber is able toaccommodate a larger volume without asubstantial increase in pressure. Left atrialdilatation is therefore adaptive in that it pre-vents significant increases in pulmonaryvascular pressures. However, this adaptationoccurs at the cost of inadequate forward car-diac output, because the compliant LA be-comes a preferred low-pressure “sink” forleft ventricular ejection, compared with thegreater impedance of the aorta. Conse-quently, as progressively larger fractions ofblood regurgitate into the LA, the mainsymptoms of chronic MR become those oflow forward cardiac output (e.g., weaknessand fatigue). In addition, chronic left atrialdilatation predisposes to the developmentof atrial fibrillation.

In chronic MR, the left ventricle also un-dergoes gradual compensatory dilatation inresponse to the volume load (through eccen-tric hypertrophy; see Chapter 9). Comparedwith acute MR, the increased ventricularcompliance accommodates the augmentedfilling volume with relatively normal dias-tolic pressures. Forward output in chronicMR is preserved to near-normal levels for an extended period by maintaining a high

206 Chapter Eight

Tall vwave

ECG

Figure 8.6. Hemodynamic profile of mitralregurgitation (MR). A large systolic v wave isnoted in the left atrial (LA) pressure tracing. Aholosystolic murmur is present in chronic MR (asshown here), beginning at the first heart sound(S1) and continuing through the second heartsound (S2). In severe acute MR, the systolic mur-mur may actually have a decrescendo qual-ity, reflecting rapid equilibration of LV and LApressures owing to the relatively reduced LAcompliance. ECG, electrocardiogram; LV, leftventricle. 1 LINE SHORT

Fig. 6

10090-08_CH08.qxd 8/31/06 4:51 PM Page 206


stroke volume via the Frank-Starling mech-anism. Over years, however, the chronicvolume overload results in deterioration ofsystolic function, a decline in forward out-put, and symptoms of heart failure.

In summary, the main differences be-tween acute and chronic MR relate to a greatextent on left atrial size and compliance (seeFig. 8.5):

Acute MR: Normal LA size and compliance→ High LA pressure → High pulmonaryvenous pressure → Pulmonary conges-tion and edema

Chronic MR: Increased LA size and com-pliance → More normal LA and pul-monary venous pressures, but low for-ward cardiac output


Presentation

As should be clear from the pathophysiol-ogy discussion, patients with acute MR usu-ally present with symptoms of pulmonaryedema (see Chapter 9). The symptoms ofchronic MR are predominantly owing to lowcardiac output, especially during exertion,and consist of fatigue and weakness. Pa-tients with severe MR or those who developLV contractile dysfunction often complainof dyspnea, orthopnea, and/or paroxysmalnocturnal dyspnea. In severe chronic MR,symptoms of right-heart failure (e.g., in-creased abdominal girth, peripheral edema)may develop as well.

Examination

The physical examination of a patient withchronic MR reveals an apical holosystolic(also termed pansystolic) murmur that radi-ates to the axilla (see Fig. 8.6). This descrip-tion, accurate for rheumatic MR, has someexceptions. For example, when ischemicpapillary muscle dysfunction interferes withnormal mitral valve closure, the regurgitantjet may be directed toward the anterior leftatrial wall, immediately posterior to theaorta. In this setting, the murmur may be

best heard along the left sternal edge or inthe aortic area (see Chapter 2) and could beconfused with the murmur of aortic stenosis(AS). Fortunately, the distinction betweenthe systolic murmur of MR and that of AScan be made by simple bedside maneuvers.If the patient is instructed to clench the fists,systemic vascular resistance will increase,and the severity of MR and its murmur willintensify, whereas the murmur of AS willnot. Even more helpful in this distinction isto notice the effect of varying cardiac cyclelength (the time between consecutive heartbeats) on the intensity of the systolic mur-mur. In a patient with atrial fibrillation orwith frequent premature beats, the LV fillsto a degree that directly depends on the pre-ceding cycle length (i.e., a longer cyclelength permits greater left ventricular fill-ing). The systolic murmur of AS becomeslouder in the beat after a long cycle lengthbecause even small pressure gradients areamplified as more blood is ejected across thereduced aortic orifice. In MR, however, theintensity of the murmur does not vary sig-nificantly because the change in the LV-to-LA pressure gradient is minimally affectedby alterations in the cycle length.

In patients with severe acute MR, the sys-tolic murmur is often different, reflectingthe underlying pathophysiology. In this case,the murmur may have a decrescendo quality,reflecting the rapid equilibration betweenLV and LA pressures in systole caused by therelatively reduced LA compliance.

In addition to the systolic murmur, acommon finding in chronic MR is the pres-ence of an S3, which reflects increased vol-ume returning to the LV in early diastole(see Chapter 2). In chronic MR, the palpatedcardiac apical impulse is often laterally dis-placed toward the axilla because of LV en-largement.

The chest radiograph may display pulmo-nary edema in acute MR but in chronicasymptomatic MR more likely demonstratesleft ventricular and atrial enlargement, with-out pulmonary congestion. Calcification ofthe mitral annulus may be seen if that is thecause of the MR. In chronic MR, the electro-LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 207

cardiogram typically demonstrates left atrialenlargement and signs of left ventricular hy-pertrophy. Echocardiography can often iden-tify the structural cause of MR and grade itsseverity by color Doppler analysis. Left ven-tricular size and function (usually vigorousin the “compensated” heart because of theincreased stroke volume) can be observed.Cardiac catheterization is useful for identify-ing a coronary ischemic cause (i.e., papillarymuscle dysfunction) and for grading theseverity of MR. The characteristic hemody-namic abnormality is a large v wave on thepulmonary capillary wedge pressure (reflect-ing LA pressure) tracing.

Natural History and Treatment

The natural history of chronic MR is relatedto its underlying cause. For example, inrheumatic heart disease, the course is one ofvery slow progression with a 15-year sur-vival rate of 70%. On the other hand, abruptworsening of chronic MR of any cause can occur with superimposed complica-tions, such as rupture of chordae tendineaeor endocarditis, and can result in an imme-diate life-threatening situation.

Medical therapy of MR involves aug-menting forward cardiac output while re-ducing regurgitation into the LA and reliev-ing pulmonary congestion. In acute MRwith heart failure, treatment includes intra-venous diuretics to relieve pulmonary edemaand vasodilators (e.g., intravenous sodiumnitroprusside) to reduce the resistance toforward flow and augment forward cardiacoutput. In chronic MR, vasodilators are lessuseful and indicated only for the treatmentof accompanying hypertension or LV sys-tolic dysfunction, because these drugs havenot been shown to delay the need for surgi-cal correction.

Because chronic MR produces continu-ous left ventricular volume overload, it canslowly result in left ventricular contractileimpairment and, ultimately, heart failure.Mitral valve surgery should be performedbefore this deterioration occurs. Mitral valverepair (reconstruction of the native valve) is

currently the preferred operative techniquefor appropriately selected patients. In thepast, the operative mortality and the draw-backs associated with the use of prostheticvalves were motivations for delaying surgeryas long as possible. Studies showed that sur-vival after mitral valve replacement was notclearly better than the natural history of thedisease, even though symptomatic improve-ment was the rule. Fortunately, mitral repairpreserves native valve tissue, is associatedwith less impairment of postoperative LVfunction, and eliminates many of the prob-lems of artificial valves.

Mitral repair involves the reconstructionof parts of the valve responsible for the re-gurgitation. For example, a perforated leafletmay be patched with transplanted autolo-gous pericardium, or ruptured chordae maybe reattached to a papillary muscle. In pa-tients who undergo a repair operation, thepostoperative survival rate appears to be bet-ter than the natural history of MR and hasprovided impetus toward earlier surgical in-tervention.

The operative mortality rate is approxi-mately 2% to 4% for mitral valve repair and5% to 7% for mitral replacement. These ratesare higher if concurrent coronary artery by-pass grafting is performed. In general, mitralvalve repair is more often appropriate foryounger patients with myxomatous in-volvement of the mitral valve, and mitral replacement is more often undertaken inolder patients with more extensive valvepathology.

Mitral Valve Prolapse

Mitral valve prolapse (MVP) is a commonand usually asymptomatic billowing of themitral leaflets into the LA during ventricularsystole, sometimes accompanied by MR(Fig. 8.7). Other names for this condition in-clude floppy mitral valve, myxomatous mi-tral valve, or Barlow syndrome. MVP may beinherited as a primary autosomal dominantdisorder with variable penetrance or mayoccur as a part of other connective tissue dis-eases, such as Marfan syndrome or Ehlers-

208 Chapter Eight

Fig. 7

1 LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 208


Danlos syndrome. Pathologically, the valveleaflets, particularly the posterior leaflet, areenlarged, and the normal dense collagenand elastin matrix of the valvular fibrosa isfragmented and replaced with loose myxo-matous connective tissue. Additionally, inmore severe lesions, elongated or rupturedchordae, annular enlargement, or thickenedleaflets may be present. A recent rigorousechocardiographic study indicated that MVPoccurs in about 2% of the population and is more common among women, especiallythose with thin, lean bodies.

MVP is often asymptomatic, but affectedindividuals may describe chest pain or pal-pitations because of associated arrhythmias.Most often it is identified on routine physi-cal examination by the presence of a mid-systolic click and late systolic murmur heardbest at the cardiac apex. The midsystolicclick is thought to correspond to the suddentensing of the involved mitral leaflet or

chordae tendineae as the leaflet is forcedback toward the left atrium; the murmurcorresponds to regurgitant flow through theincompetent valve. The click and murmurare characteristically altered during dy-namic auscultation at the bedside: maneu-vers that increase the volume of the LV (e.g.,sudden squatting, which increases venousreturn) delay the occurrence of prolapse insystole and cause the click and murmur tooccur later (i.e., further from S1). Con-versely, if the volume of blood in the LV isdecreased (e.g., on sudden standing), pro-lapse occurs more readily and the click and murmur occur earlier in systole (closerto S1).

Confirmation of the diagnosis is obtainedby echocardiography, which demonstrates pos-terior displacement of one or both mitralleaflets into the left atrium during systole.The electrocardiogram and chest radiograph areusually normal unless chronic MR has re-

AO

RV

LA

LV

Figure 8.7. Mitral valve prolapse. Long axis view of the leftventricle (LV) demonstrates myxomatous, elongated appear-ance of the mitral valve with prolapse of the posterior leaflet(arrow) into the left atrium (LA). Ao, aorta; RV, right ventricle.(From Schoen FJ. The heart. In: Kumar V, Abbas A, Fausto N,eds. Robbins and Cotran Pathologic Basis of Disease. 7th Ed.Philadelphia: Elsevier Saunders, 2005:592. With permission requested.)AQ2

LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 209

sulted in left atrial and left ventricular en-largement.

The clinical course of MVP is most oftenbenign. Treatment consists of reassuranceabout the usually good prognosis and anti-biotic prophylaxis (to prevent endocarditis)only if substantial valve thickening or MR arepresent. Of the potential complications, themost common is the development of gradu-ally progressive MR. Occasionally, rupture ofmyxomatous chordae can cause sudden, se-vere regurgitation and pulmonary edema.Other rare complications include infectiveendocarditis, peripheral emboli owing to mi-crothrombus formation behind the redun-dant valve tissue, and atrial or ventricular ar-rhythmias.

AORTIC VALVE DISEASE

Aortic Stenosis

Etiology

Aortic stenosis (AS) is often caused by age-related degenerative calcific changes of the valve,formerly termed senile AS. Calcific changesthat progress to AS may also develop in pa-tients with congenitally deformed aortic valves(about 1% to 2% of the population is bornwith an abnormal bicuspid aortic valve).Most patients who present with AS after theage of 65 have the age-related form, whereasyounger patients usually have calcificationof a congenitally bicuspid valve. AS may alsoresult from chronic rheumatic valve disease,although the prevalence of this conditionhas decreased dramatically in recent decadesin the United States. Approximately 95% ofpatients who are found to have rheumaticAS have coexisting rheumatic involvementof the mitral valve.

Pathology

The pathologic appearance in AS is depen-dent on its etiology. In age-related degener-ative AS, the classic teaching is that cumula-tive “wear and tear” of valve motion overmany years leads to endothelial and fibrous

210 Chapter Eight

damage, causing calcification of an other-wise normal trileaflet valve. However, thereis also evolving evidence of a common eti-ology with atherosclerotic vascular disease.Studies have shown that as in atherosclero-sis, the valve tissue of patients with thisform of AS display cellular proliferation, in-flammation, lipid accumulation, and in-creased margination of macrophages and Tlymphocytes.

In the case of a congenitally deformedvalve, years of turbulent flow across the valvedisrupt the endothelium and collagen ma-trix of the leaflets, resulting in gradual cal-cium deposition. In rheumatic AS, endocar-dial inflammation leads to organization andfibrosis of the valve, and ultimately fusionof the commissures and the formation ofcalcified masses within the aortic cusps.

Pathophysiology

In AS, blood flow across the aortic valve isimpeded during systole (Fig. 8.8). When thevalve orifice area is reduced by more than50% of its normal size, significant elevationof left ventricular pressure is necessary todrive blood into the aorta (Fig. 8.9). In ad-vanced AS, it is common to measure peak

1 LINE SHORT

Fig. 8

Fig. 9

Pressure

Figure 8.8. Pathophysiol-ogy of aortic stenosis (AS).The impediment to left ven-tricular (LV) outflow in AS results in elevated LV pres-sures and secondary ventricu-lar hypertrophy.

10090-08_CH08.qxd 8/31/06 4:51 PM Page 210


systolic pressure gradients >100 mm Hg be-tween the LV and the aorta.

Since AS develops over a chronic course,the LV is able to compensate by undergoingconcentric hypertrophy in response to thehigh systolic pressure it must generate. Suchhypertrophy serves an important role in re-ducing ventricular wall stress (rememberfrom Chapter 6 that wall stress = (P × r) ÷ 2h,in which h represents wall thickness); how-ever, it also reduces the compliance of theventricle. The resulting elevation of diastolicLV pressure also causes the LA to hypertro-phy in order to fill the “stiff” LV. Whereasleft atrial contraction contributes only asmall portion of the left ventricular strokevolume in normal individuals, it may pro-vide more than 25% of the stroke volume tothe stiffened LV in AS patients. Thus, leftatrial hypertrophy is beneficial, and the lossof effective atrial contraction (e.g., the de-velopment of atrial fibrillation) can causemarked clinical deterioration.

Three major manifestations occur in pa-tients with advanced AS: (1) angina, (2) ex-ertional syncope, and (3) congestive heartfailure, all of which can be explained on thebasis of the underlying pathophysiology.Each manifestation, in order, heralds an in-creasingly ominous prognosis (Table 8.2).

AS may result in angina because it createsa substantial imbalance between myocar-dial oxygen supply and demand. Myocar-dial oxygen demand is increased in two ways. First, the muscle mass of the hyper-trophied LV is increased, requiring greater-than-normal perfusion. Second, wall stressis increased because of the elevated systolicventricular pressure. In addition, AS reducesmyocardial oxygen supply because the ele-vated left ventricular diastolic pressure re-duces the coronary perfusion pressure gradi-ent between the aorta and the myocardium.

AS may cause syncope during exertion. Al-though left ventricular hypertrophy allowsthe chamber to generate a high pressureand maintain a normal cardiac output atrest, the ventricle cannot significantly in-crease its cardiac output during exercisebecause of the fixed stenotic aortic orifice.In addition, exercise leads to vasodilatationof the peripheral muscle beds. Thus, thecombination of peripheral vasodilatationand the inability to augment cardiac outputcontributes to decreased cerebral perfusionpressure and, potentially, loss of conscious-ness on exertion.

Finally, AS can result in symptoms ofcongestive heart failure. Early in the courseof AS, an abnormally increased left atrialpressure occurs primarily at the end of

ECG

50

110

130

150

30

10

Figure 8.9. Hemodynamic profile of aorticstenosis. A systolic pressure gradient (shadedarea) is present between the left ventricle (LV)and aorta. The second heart sound (S2) is dimin-ished in intensity, and there is a crescendo–de-crescendo systolic murmur that does not extendbeyond S2. ECG, electrocardiogram.

TABLE 8.2. Median Survival Time in Symptomatic Aortic Stenosis

Clinical Symptoms Median Survival

Angina 5 yearsSyncope 3 yearsCongestive heart failure 2 yearsAtrial fibrillation 6 months

Reprinted with permission from Ross J Jr, Braunwald E. Aorticstenosis. Circulation 1968;38(suppl v):61.LINE SHORT

Tab. 2

10090-08_CH08.qxd 8/31/06 4:51 PM Page 211

diastole, when the LA contracts into thethickened noncompliant LV. As a result,the mean left atrial pressure and the pul-monary venous pressure are not greatly af-fected early in the disease. However, withprogression of the stenosis, the LV may de-velop contractile dysfunction because ofthe insurmountably high afterload, lead-ing to increased left ventricular diastolicvolume and pressure. The accompanyingmarked elevation of LA and pulmonary ve-nous pressures incites pulmonary alveolarcongestion and the symptoms of congestiveheart failure.

The normal aortic valve cross-sectionalarea is 3 to 4 cm2. When the valve area is re-duced to less than 2 cm2, a pressure gradientbetween the LV and aorta first appears (mildAS). Moderate AS is characterized by a valvearea of 1.0 to 1.5 cm2. When the aortic valvearea is reduced to less than 1.0 cm2, severevalve obstruction is said to be present.


Presentation

Angina, syncope, and congestive heart fail-ure may appear after many asymptomaticyears of slowly progressive valve stenosis.Once these symptoms develop, they confera significantly decreased survival if surgicalcorrection of AS is not undertaken (seeTable 8.2).

Examination

Physical examination often permits accu-rate detection and estimation of the sever-ity of AS. The key features of advanced AS include (1) a coarse late-peaking systo-lic ejection murmur and (2) a weakened(parvus) and delayed (tardus) upstroke ofthe carotid artery pulsations owing to theobstructed LV outflow. Other commonfindings on cardiac examination includethe presence of an S4 (because of atrial con-traction into the “stiff” LV) and reduced in-tensity, or complete absence, of the aorticcomponent of the second heart sound (seeFig. 8.9).

On the electrocardiogram, left ventricularhypertrophy is common in advanced AS, butechocardiography is a more sensitive tech-nique to assess LV wall thickness. The trans-valvular pressure gradient and aortic valvearea can be calculated by Doppler echocar-diography (see Chapter 3). Cardiac catheteri-zation is sometimes used to confirm the sever-ity of AS and to define the coronary anatomy,because concurrent coronary artery bypasssurgery is often necessary at the time of aor-tic valve replacement in patients with co-existing coronary disease.

Natural History and Treatment

The natural history of severe, symptomatic,uncorrected AS is very poor. Data from theMayo Clinic indicate that the 1-year sur-vival rate is 57% for patients with this con-dition. The only effective treatment for ad-vanced AS is surgical replacement of thevalve.

Aortic valve replacement (AVR) is indi-cated when patients with severe AS developsymptoms, or when there is evidence of pro-gressive LV dysfunction in the absence ofsymptoms. The left ventricular ejection frac-tion almost always increases after valve re-placement, even in patients with impairedpreoperative left ventricular function. Theeffect of AVR on the natural history of AS isdramatic, with the 10-year survival rate ex-ceeding 75%.

Unlike its successful role in mitral steno-sis, percutaneous valvuloplasty has been dis-appointing in the treatment of AS in adults.Although balloon dilatation of the aorticvalve orifice can fracture calcified commis-sures leading to some immediate relief ofoutflow obstruction, up to 50% of patientsdevelop restenosis within 6 months. Valvu-loplasty is occasionally a suitable option forpatients who are poor surgical candidates oras a temporizing measure in patients too illto proceed directly to valve replacement.Valvuloplasty is also sometimes effective inyoung patients with congenitally bicuspidvalves.

212 Chapter Eight

10090-08_CH08.qxd 8/31/06 4:51 PM Page 212


Mild, asymptomatic AS has a slow rate ofprogression such that over a 20-year period,only 20% of patients will progress to severe orsymptomatic AS. Medical therapy for asymp-tomatic AS includes endocarditis antibioticprophylaxis and cautious use of medica-tions that could result in hypotension inthis setting (e.g., vasodilators, diuretics, nitro-glycerin).

Aortic Regurgitation

Etiology

Aortic regurgitation (AR), also termed aorticinsufficiency, may result from (1) diseases ofthe aortic leaflets or (2) dilatation of the aor-tic root. The most common causes of AR arelisted in Table 8.3.

Pathophysiology

In AR, abnormal regurgitation of blood fromthe aorta into the LV occurs during diastole.Therefore, with each contraction, the LVmust pump the regurgitant volume plus thenormal amount of blood returning from the LA. Hemodynamic compensation relies on the Frank-Starling mechanism to aug-ment stroke volume. Factors influencing theseverity of AR are analogous to those of MR:(1) the size of the regurgitant aortic orifice,(2) the pressure gradient across the aorticvalve during diastole, and (3) the durationof diastole.

As in MR, the hemodynamic abnormalitiesand symptoms differ in acute and chronic AR

(Fig. 8.10). In acute AR, the LV is of normalsize and relatively noncompliant. Thus, thevolume load of regurgitation causes the LVdiastolic pressure to rise substantially. Thesudden high diastolic LV pressure is trans-mitted to the LA and pulmonary circulation,often producing dyspnea and pulmonaryedema. Thus, severe acute AR is usually a sur-gical emergency, requiring immediate valvereplacement.

In chronic AR, the LV undergoes compen-satory adaptation in response to the long-standing regurgitation. AR subjects the LVprimarily to volume overload but also to anexcessive pressure load; therefore, the ven-tricle compensates through dilatation and,to a lesser degree, both eccentric and con-centric hypertrophy. Over time, the dilata-tion increases the compliance of the LV andallows it to accommodate a larger regurgitantvolume with less of an increase in diastolicpressure, reducing the pressure transmittedinto the LA and pulmonary circulation. How-ever, by accommodating the large regur-gitant volume, the aortic (and thereforesystemic arterial) diastolic pressure dropssubstantially. The combination of a high LVstroke volume (and therefore high systolicarterial pressure) with a reduced aortic dias-tolic pressure produces a widened pulse pres-sure (the difference between arterial systolicand diastolic pressures), a hallmark ofchronic AR (Fig. 8.11). As a result of the de-creased aortic diastolic pressure, the coro-nary artery perfusion pressure falls, poten-tially reducing myocardial oxygen supply.This, coupled with the increase in LV size(which causes increased wall stress and myo-cardial oxygen demand), can produce angina,even in the absence of atherosclerotic coro-nary disease.

Because compensatory left ventriculardilatation and hypertrophy are generallyadequate to meet the demands of chro-nic AR, affected patients are usually asymp-tomatic for many years. Gradually, how-ever, progressive remodeling of the LV occurs, resulting in systolic dysfunction.This causes decreased forward cardiac out-put as well as an increase in left atrial and

TABLE 8.3. Causes of Aortic Regurgitation

Abnormalities of valve leaflets1. Congenital (bicuspid valve)2. Endocarditis3. RheumaticDilatation of aortic root1. Aortic aneurysm (inflammation; connective

tissue disease, e.g., Marfan syndrome)2. Aortic dissection3. Annuloaortic ectasia4. Syphilis

Tab. 3

Fig. 10

Fig. 11

10090-08_CH08.qxd 8/31/06 4:51 PM Page 213

pulmonary vascular pressures. At that point,the patient develops symptoms of heartfailure.

Clinical Manifestations and Assessment

Presentation

Common symptoms of chronic AR includedyspnea on exertion, fatigue, decreased ex-ercise tolerance, and the uncomfortable sen-sation of a forceful heartbeat associated withhigh pulse pressure.

Examination

Physical examination may show boundingpulses and other stigmata of the widenedpulse pressure (Table 8.4), in addition to ahyperdynamic LV impulse and a blowingmurmur of AR in early diastole along theleft sternal border (see Fig. 8.11). It is bestheard with the patient leaning forward,

after exhaling. In addition, a low-frequencymiddiastolic rumbling sound may be aus-cultated at the cardiac apex in some pa-tients with severe AR. Known as the AustinFlint murmur, it is thought to reflect turbu-lent blood flow across the mitral valveduring diastole owing to downward dis-placement of the mitral anterior leaflet bythe stream of AR. It can be distinguishedfrom the murmur of MS by the absence ofan opening snap or presystolic accentua-tion of the murmur.

In chronic AR, the chest radiograph showsan enlarged left ventricular silhouette. This isusually absent in acute AR, in which pul-monary vascular congestion is the more likelyfinding. Doppler echocardiography can identifyand quantify the degree of AR and often canidentify its cause. Cardiac catheterization withcontrast angiography is useful for evaluationof left ventricular function, quantification ofthe degree of AR, and assessment of coexist-ing coronary artery disease.

214 Chapter Eight

Pressure PressureN-

PressureN-Pressure

Pulmonarycongestion

Figure 8.10. Pathophysiology of acute and chronic aortic regur-gitation (AR). Abnormal regurgitation of blood from the aorta intothe left ventricle (LV) is shown in each schematic drawing (large ar-rows). In acute AR, the LV is of normal size and relatively low compli-ance, such that its diastolic pressure rises significantly; this pressure in-crease is reflected back to the left atrium (LA) and pulmonaryvasculature, resulting in pulmonary congestion or edema. In chronicAR, adaptive LV and LA enlargement have occurred, such that agreater volume of regurgitation can be accommodated with less of anincrease in diastolic LV pressure, so that pulmonary congestion is lesslikely. N, normal.

Tab. 4

1 LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 214


Treatment

Data from natural history studies indicatethat clinical progression of patients withasymptomatic chronic AR and normal LVcontractile function is very slow. Therefore,asymptomatic patients are monitored withperiodic examinations and assessment of LVfunction, usually by serial echocardiogra-

phy. Patients are instructed to follow anti-biotic prophylaxis to prevent endocarditis,as described in a later section. Patients withasymptomatic severe AR and preserved LV function may benefit from afterload re-ducing vasodilators (e.g., a calcium channelblocker or angiotensin-converting enzymeinhibitor). For example, the vasodilating cal-cium channel blocker nifedipine has beenshown in some, but not all, studies to re-duce LV enlargement, increase the LV ejec-tion fraction, and delay the need for valvesurgery in hypertensive patients with severeAR who have normal LV contractile function.

Symptomatic patients, or asymptomaticpatients with severe AR and impaired LVcontractile function (i.e., an ejection frac-tion <0.50), should be offered surgical cor-rection to prevent progressive deteriora-tion. Studies of patients with AR show thatwithout surgery, death usually occurs with-in 4 years after the development of anginaor 2 years after the onset of heart failuresymptoms.

TRICUSPID VALVE DISEASE

Tricuspid Stenosis

Tricuspid stenosis (TS) is rare and usually aconsequence of rheumatic heart disease.The opening snap and diastolic murmur ofTS are similar to those of MS, but the mur-mur is heard closer to the sternum and itintensifies on inspiration because of in-creased right-heart blood flow. In TS, the

ECG

Figure 8.11. Hemodynamic profile of aortic regur-gitation. During diastole, the aortic pressure fallsrapidly (arrow), and left ventricular (LV) pressure rises asblood regurgitates from the aorta into the LV. A dias-tolic decrescendo murmur, beginning at the secondheart sound (S2), corresponds with the abnormal regur-gitant flow. ECG, electrocardiogram.

TABLE 8.4. Physical Findings Associated With Widened Pulse Pressure in Chronic Aortic Regurgitation

Name Description

Bisferiens pulse Double systolic impulse in carotid or brachial arteryCorrigan pulse “Water-hammer” pulses with marked distention and collapsede Musset sign Head-bobbing with each systoleDuroziez sign To-and-fro murmur heard over femoral artery with light compressionHill sign Popliteal systolic pressure more than 60 mm Hg greater than brachial systolic pressureMüller sign Systolic pulsations of the uvulaQuincke sign Capillary pulsations visible at lip or proximal nail bedsTraube sign “Pistol-shot” sound auscultated over the femoral artery

LINE SHORT

10090-08_CH08.qxd 8/31/06 4:51 PM Page 215

neck veins are distended and show a large awave as a result of right atrial contractionagainst the stenotic tricuspid valve orifice.Patients may develop abdominal disten-tion and hepatomegaly owing to passivevenous congestion.

Symptoms of TS can be similar to those ofMS, and the two conditions can coexist assequelae of rheumatic heart disease. Surgicaltherapy is usually required (valvuloplasty orvalve replacement).

Tricuspid Regurgitation

Tricuspid regurgitation (TR) is usually func-tional rather than structural (or organic); thatis, it develops because of right ventricular en-largement (e.g., owing to pressure or volumeoverload) and not because of primary valvedisease. Among patients with rheumatic MS,20% have significant TR (of whom 80% havefunctional TR because of pulmonary hyper-tension with right ventricular enlargement,and 20% have organic TR resulting fromrheumatic involvement of the tricuspidvalve). Another rare cause of TR is carcinoidsyndrome, in which a type of tumor (usuallyin the small bowel or appendix, with metas-tases to the liver) releases serotonin metabo-lites into the bloodstream. These metabolitesare thought to be responsible for the forma-tion of endocardial plaques in the right sideof the heart. Involvement of the tricuspidvalve immobilizes the leaflets, often resulting in substantial TR and, less often, tricuspidstenosis.

The most sensitive physical signs of TRare prominent v waves in the jugular veinsand a pulsatile liver because of regurgitationof right ventricular blood into the systemicveins. The systolic murmur of TR is heard at the lower-left sternal border. It is oftensoft but becomes louder on inspiration.Doppler echocardiography readily detectsTR and can quantify it. The primary therapyof functional TR is directed at the conditionsresponsible for the elevated right ventricularsize or pressure as well as diuretic therapy;surgical repair of the valve is indicated in se-vere cases.

PULMONIC VALVE DISEASE

Pulmonic Stenosis

Pulmonic stenosis (PS) is rare, and its causeis almost always congenital deformity of thevalve. Carcinoid syndrome, described in theprevious section, is another rare etiology, inwhich encasement and immobilization ofthe valve leaflets can occur.

Severe cases of pulmonic stenosis are as-sociated with a peak systolic pressure gradi-ent of greater than 80 mm Hg, moderate dis-ease with a gradient of 40 to 80 mm Hg, andmild PS is said to be present when the trans-valvular gradient is less than 40 mm Hg.Only patients with moderate to severe gra-dients are symptomatic. In such cases, tran-scatheter balloon valvuloplasty is usually effective therapy.

Pulmonic Regurgitation

Pulmonic regurgitation most commonly de-velops in the setting of severe pulmonaryhypertension and results from dilatation ofthe valve ring by the enlarged pulmonaryartery. Auscultation reveals a high-pitcheddecrescendo murmur along the left sternalborder that is often indistinguishable fromAR (the two conditions are easily differenti-ated by Doppler echocardiography).

PROSTHETIC VALVES

The patient who undergoes valve replace-ment surgery often benefits dramaticallyfrom hemodynamic and symptomatic im-provement but also acquires a new set of po-tential complications related to the valveprosthesis itself. Because all available valvesubstitutes have certain limitations, valve re-placement surgery is not a true “cure.”

The first successful valve replacementstook place in the 1960s. Currently availablevalve substitutes include mechanical and bi-ologic (derived from animal or human tis-sue) devices (Fig. 8.12). Older mechanicalvalves included a ball-in-cage design, with abulky shape that often left a significantvalvular gradient and occasionally produced

216 Chapter Eight

Fig. 12

10090-08_CH08.qxd 8/31/06 4:51 PM Page 216


intravascular hemolysis from red blood celltrauma. This valve type, however, had animpressive record of durability, with somemodels functioning well for more than 30 years. Newer mechanical valves, such asbileaflet prostheses, provide a lower profileand superior hemodynamics with no appar-ent sacrifice of durability. One example isthe St. Jude prosthesis, a hinged bileafletvalve consisting of two Pyrolyte carbondiscs that open opposite one another (seeFig. 8.12). Mechanical valves, while durable,present foreign thrombogenic surfaces tothe circulating blood and require lifelonganticoagulation (usually with oral warfarin)to prevent thromboembolism.

The most commonly used bioprosthesesare made from glutaraldehyde-fixed porcinevalves secured in a support frame. In ad-dition, bovine pericardium and humanhomograft (aortic valves harvested and cryopreserved from cadavers) prostheses are used. Bioprosthetic valves have limiteddurability compared with mechanical valves,and structural failure occurs in up to 50% ofvalves at 10 years. Failure rates vary greatlydepending on the position of the valve. Forexample, bioprosthetic valves in the mitralposition deteriorate more rapidly thanthose in the aortic position. This is likely

because the mitral valve is forced closedduring systolic contraction, resulting ingreater leaflet stress than that experiencedby aortic prostheses that close during dias-tolic relaxation.

The principal causes of bioprosthetic valvefailure are leaflet tears and calcification.Conversely, the main advantage of biopros-theses is that they display a very low rate ofthromboembolism and do not require long-term anticoagulation therapy. For patientswith aortic valve endocarditis, aortic homo-graft replacements are especially useful be-cause they have very low rates of subsequentinfection.

Common to all types of valve replace-ment is the risk of infective endocarditis(discussed in the next section), which occursat an incidence of 1% to 2% per patient peryear. If endocarditis occurs in the first 60 daysafter valve surgery, the mortality rate is ex-ceedingly high (50% to 80%). If endocardi-tis occurs later, mortality rates range from20% to 50%. Reoperation is usually requiredif endocarditis involves a mechanical pros-thesis, because an adjacent abscess is almostalways present (the organism cannot infectthe prosthetic material itself). Some cases ofbioprosthetic valve endocarditis respond toantibiotic therapy alone.

Figure 8.12. Examples of prosthetic heart valves. A. St. Jude mechanical bileaflet valve in the open position. (Cour-tesy of St. Jude Medical, Inc., St. Paul, MN.) B. A bioprosthetic aortic valve with leaflets in the closed position. (Courtesyof Medtronic, Inc., Minneapolis, MN.)

10090-08_CH08.qxd 8/31/06 4:51 PM Page 217

Given their respective advantages anddisadvantages, the mortality and complica-tion rates with mechanical and biopros-thetic valves are similar for the first 10 yearsfollowing replacement. In 20-year follow-ups to long-term, randomized, controlledtrials, mechanical valves have been shownto be superior to bioprosthetic valves inevent-free survival, except for bleeding com-plications related to the required chronicanticoagulation therapy. Therefore, the de-cision about which type of prosthesis to usein a patient often centers on (1) the patient’sexpected lifespan in comparison to the func-tional longevity of the valve, (2) risk-versus-benefit considerations of chronic anticoag-ulation therapy, and (3) patient and surgeonpreferences. Mechanical valves are often rec-ommended for younger patients and forthose who will be tolerant of, and compliantwith, anticoagulant therapy. Bioprostheticvalves are suitable choices for patients 65years or older and for patients with con-traindications to chronic anticoagulationtherapy.

INFECTIVE ENDOCARDITIS

Infection of the endocardial surface of theheart, including the cardiac valves, can leadto extensive tissue damage and is often fatal.Infective endocarditis (IE) carries an overall6-month mortality rate of 20% to 25%, evenwith appropriate therapy, and a 100% mor-tality rate if it is not recognized and treatedcorrectly.

There are three clinically useful ways toclassify IE: (1) by clinical course, (2) by hostsubstrate, or (3) by the specific infecting microorganism. In the first classificationscheme, IE is termed acute bacterial endo-carditis (ABE) when the syndrome pre-sents as an acute, fulminant infection, anda highly virulent and invasive organismsuch as Staphylococcus aureus is implicated.Because of the aggressiveness of the respon-sible microorganism, ABE may occur on pre-viously healthy heart valves. When IE pre-sents with a more insidious clinical course,it is termed subacute bacterial endocardi-tis (SBE), and less virulent organisms such

as streptococci viridans are involved. SBEmost frequently occurs in individuals withprior underlying valvular damage.

The second means of classification of IE isaccording to the host substrate: (1) nativevalve endocarditis, (2) prosthetic valve en-docarditis, or (3) endocarditis in the settingof intravenous drug abuse. Of these, nativevalve endocarditis accounts for 60% to 80%of patients. Different microorganisms andclinical courses are associated with each ofthese categories. For example, the skin cont-aminant Staphylococcus epidermidis, is a com-mon cause of prosthetic valve endocarditis,but that is rarely the case when endocarditisoccurs on a native heart valve. Intravenousdrug users have a propensity for endocardi-tis on the right-sided heart valves.

The third classification of IE is accordingto the specific infecting microorganism (e.g.,S. aureus endocarditis). Although the re-mainder of this discussion focuses on theendocarditis syndromes based on clinicalcourse, it is important to recognize that allthree classifications of IE are used.

Pathogenesis

The pathogenesis of endocarditis requiresseveral conditions: (1) endocardial surfaceinjury, (2) thrombus formation at the site ofinjury, (3) bacterial entry into the circula-tion, and (4) bacterial adherence to the in-jured endocardial surface. The first two con-ditions provide an environment favorableto infection, whereas the latter two permitimplantation of the organism on the endo-cardial surface. The most common cause ofendothelial injury is turbulent blood flowresulting from preexisting valvular disease;approximately 70% of patients with endo-carditis have evidence of underlying struc-tural or hemodynamic abnormalities (Table8.5). Endothelial injury may also be incitedby foreign material within the circulation,such as indwelling venous catheters or pros-thetic heart valves.

Once an endocardial surface is injured,platelets adhere to the exposed subendocar-dial connective tissue and initiate the for-mation of a sterile thrombus (termed a

218 Chapter Eight

Tab. 5

10090-08_CH08.qxd 8/31/06 4:51 PM Page 218


vegetation) through fibrin deposition. Thisprocess is referred to as nonbacterial throm-botic endocarditis (NBTE). NBTE makes theendocardium more hospitable to microbesin two ways. First, the fibrin-platelet depositsprovide a surface for adherence by bacteria.Second, the fibrin covers adherent organ-isms and protects them from host defensesby inhibiting chemotaxis and migration ofphagocytes.

When NBTE is present, the delivery ofmicroorganisms in the bloodstream to theinjured surface can lead to infective endo-carditis. Three factors determine the abilityof an organism to induce IE: (1) access to thebloodstream, (2) survival of the organism inthe circulation, and (3) adherence of thebacteria to the endocardium. Bacteria can beintroduced into the bloodstream whenevera mucosal or skin surface harboring an or-ganism is traumatized, such as from themouth during dental procedures or from theskin during illicit intravenous drug use.However, while transient bacteremia is a rel-atively common event, only microorgan-isms suited for survival in the circulationand able to adhere to the platelet-fibrinmesh overlying the endocardial defect willcause infective endocarditis. For example,

Gram-positive organisms account for ap-proximately 90% of cases of endocarditis,largely because of their resistance to de-struction in the circulation by complementand their particular ability to adhere to en-dothelial and platelet surface proteins. Theability of certain streptococcal species toproduce dextran, a bacterial cell wall com-ponent that adheres to thrombus, correlateswith their inciting endocarditis. Table 8.6lists the infectious agents reported to be themost common causes of endocarditis inmodern tertiary centers; staphylococci andstreptococci are the most frequent. It is im-portant to recognize that in more rural com-munities with a low prevalence of intra-venous drug abuse, the percentage of viridansstreptococcal infections tends to be greaterthan that of S. aureus.

Once organisms adhere to the injuredsurface, they may be protected from phago-cytic activity by the overlying fibrin. Theorganisms are then free to multiply, whichenlarges the infected vegetation. The latterprovides a source for continuous bacte-remia and can lead to several complica-tions, including (1) mechanical cardiac injury, (2) thrombotic or septic emboli, and(3) immune injury mediated by antigenan-tibody deposition. For example, local ex-tension of the infection within the heartcan result in progressive valvular damage,

TABLE 8.5. Cardiac Lesions That Predispose to Endocarditis

Rheumatic valvular diseaseOther acquired valvular lesionsCalcific aortic stenosisAortic regurgitationMitral regurgitationMitral valve prolapse (if murmur auscultated or

detected by Doppler)Hypertrophic obstructive cardiomyopathyCongenital heart diseaseVentricular septal defectPatent ductus arteriosusTetralogy of FallotAortic coarctationBicuspid aortic valvePulmonic stenosisSurgically implanted intravascular hardwareProsthetic heart valvesPulmonary-systemic vascular shuntsVentriculoatrial shunts for hydrocephalus

TABLE 8.6. Microbiology of Infective Endocarditis in Tertiary Centers

Organism Incidence (%)

StaphylococciS. aureus 31.6Coagulase-negative 10.5StreptococciViridans 18.0Enterococci 10.6S. bovis 6.5Other streptococci 5.1Other organisms (e.g., Gram-negative

bacteria, fungi) 8.7Culture negative or polymicrobial ∼9.4

Modified from Fowler VG Jr, Miro JM, Hoen B, et al. Staphy-lococcus aureus endocarditis: a consequence of medicalprogress. JAMA 2005;293:3012–3021.

Tab. 6

10090-08_CH08.qxd 8/31/06 4:51 PM Page 219

abscess formation, or erosion into the car-diac conduction system. Portions of a veg-etation may embolize systemically, oftento the central nervous system, kidneys, orspleen, and incite infection or infarction ofthe target organs. Each of these is a poten-tially fatal complication. Additionally, im-mune complex deposition can result inglomerulonephritis, arthritis, or vasculitis.

The epidemiology of IE has evolved inrecent decades, as bacteria resistant to anti-biotics have become ubiquitous in the hos-pital setting and have spread into the com-munity. Antibiotic resistant strains such asmethicillin-resistant S. aureus and vanco-mycin-resistant enterococci have becomemore common and are associated with in-creased mortality rates from IE.

Clinical Manifestations

A patient with acute IE is likely to report anexplosive and rapidly progressive illnesswith high fever and shaking chills. In con-trast, subacute IE presents less dramaticallywith low-grade fever often accompanied bynonspecific constitutional symptoms suchas fatigue, anorexia, weakness, myalgia, andnight sweats. These symptoms are not spe-cific for IE and could easily be mistaken forinfluenza or an upper respiratory tract in-fection. Thus, the diagnosis of subacute IE requires a high index of suspicion. A his-tory of a valvular lesion or other conditionknown to predispose to endocarditis is help-ful. A thorough history should also inquireabout intravenous drug use, recent dentalprocedures, or other potential sources of bac-teremia.

Cardiac examination may reveal a mur-mur representing the underlying valvularpathology that predisposed the patient to IE,or a new murmur of valvular insufficiencyowing to IE-induced damage. The develop-ment of right-sided valvular lesions (e.g.,tricuspid regurgitation), although rare innormal hosts, is particularly common in en-docarditis associated with intravenous drugabuse. Serial examination in ABE may be es-pecially useful because changes in a particu-

lar murmur (i.e., worsening regurgitation)over time may correspond with rapidly pro-gressive valvular damage. During the courseof endocarditis, severe valvular damage mayresult in findings of congestive heart failure.

Other physical findings that may appearin IE are those associated with septic em-bolism or immune complex deposition.Central nervous system emboli are seen inup to 40% of patients, often resulting in newneurologic findings on physical examina-tion. Injury to the kidneys, of embolic or immunologic origin, may manifest as flankpain, hematuria, or renal failure. Lung infarc-tion (septic pulmonary embolism) or infec-tion (pneumonia) are particularly common in endocarditis that involves the right-sidedvalves.

Embolic infarction and seeding of thevasa vasorum of arteries can cause localizedaneurysm formation (termed a mycotic aneu-rysm) that weakens the vessel wall and mayrupture. Mycotic aneurysms may be foundin the aorta, viscera, or peripheral organs,and are particularly dangerous in cerebralvessels, because rupture there can result infatal intracranial hemorrhage.

Skin findings resulting from septic em-bolism or immune complex vasculitis are often collectively referred to as peripheralstigmata of endocarditis. For example, pe-techiae may appear as tiny, circular, red-brown discolorations on mucosal surfaces orskin. Splinter hemorrhages, the result of sub-ungual microemboli, are small, longitudinalhemorrhages found beneath nails. Otherperipheral stigmata of IE, which are nowrarely encountered, include painless, flat, ir-regular discolorations found on the palmsand soles called Janeway lesions; tender,pea-sized, erythematous nodules found pri-marily in the pulp space of the fingers andtoes termed Osler nodes; and emboli to theretina that produce Roth spots, microinfarc-tions that appear as white dots surroundedby hemorrhage.

The systemic inflammatory response pro-duced by the infection is responsible forfever and splenomegaly as well as for a num-ber of laboratory findings, including an ele-vated white blood cell count with a leftward

220 Chapter Eight

10090-08_CH08.qxd 8/31/06 4:51 PM Page 220


shift (increase in proportion of neutrophilsand immature granulocytes), an elevatederythrocyte sedimentation rate or C-reactiveprotein level, and in approximately 50% ofcases, an elevated serum rheumatoid factor.

The electrocardiogram may help identifyextension of the infection into the cardiacconduction system, manifest by various de-grees of heart block or new arrhythmias.Echocardiography is used to visualize vegeta-tions, valvular dysfunction, and associatedabscess formation. Echocardiographic assess-ment can consist of transthoracic echocar-diography (TTE) or transesophageal echocar-diography (TEE), as described in Chapter 3.TTE is useful in detecting large vegetationsand has the advantage of being noninva-sive and easy to obtain. However, while theestimated specificity of TTE for vegetationsis high, the sensitivity for finding vegeta-tions is less than 60%. TEE, on the otherhand, is much more sensitive (>90%) forthe detection of small vegetations and canbe particularly useful for the evaluation ofinfection involving prosthetic valves.

Central to the diagnosis and appropriatetreatment of endocarditis is the identifi-cation of the responsible microorganism by blood culture. Once positive culture re-sults are obtained, treatment can be tailoredto the causative organism according to itsantibiotic sensitivities. A specific etiologicagent will be identified approximately 95% of the time. However, blood culturesmay fail to grow the responsible organismif antibiotics had already been adminis-tered or if the organism has unusual growthrequirements.

Even after a careful history, examination,and evaluation of laboratory data, the diag-nosis of IE can be elusive. Therefore, at-tempts have been made to standardize thediagnosis, resulting in the now widely usedDuke criteria (Table 8.7). By this standard,the diagnosis of endocarditis rests on thepresence of either two major criteria, onemajor and three minor criteria, or five minorcriteria.

Treatment of endocarditis entails 4 to 6 weeks of high-dose intravenous antibiotictherapy. Although empiric broad-spectrum

antibiotics may be used initially (after bloodcultures are obtained) in patients who are se-verely ill or hemodynamically unstable, spe-cific, directed therapy is preferable once thecausative microorganism has been identified.Surgical intervention, usually with valve re-placement, is indicated for patients with per-sistent bacteremia despite maximal antibiotictherapy, patients with severe valvular dys-function leading to heart failure, and patientswho develop myocardial abscesses or experi-ence recurrent thromboembolic events.

Finally, an additional essential concept isprevention of endocarditis by administeringantibiotics to susceptible individuals (i.e.,those with underlying structural heart dis-ease) before procedures that are likely to re-sult in bacteremia (Table 8.8). Although ran-domized controlled trials have not beenperformed to demonstrate the efficacy of an-tibiotic prophylaxis against IE, such practiceis widespread.

SUMMARY

Valvular heart disease can be a significantsource of disability and mortality. From sim-ple bedside observations to complex hemo-dynamic measurements, much has beenlearned about the pathophysiology of theseconditions. A summary of the importantfindings associated with major valve lesionsis presented in Table 8.9.

Acknowledgment

Contributors to the previous editions of this chapterwere Edward Chan, MD; Elia Duh, MD; Stephen K.Frankel, MD; Brian Stidham, MD; Patrick Yachimski,MD; John A. Bittl, MD; and Leonard S. Lilly, MD.

Additional Reading

Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006 guidelines for the management of pa-tients with valvular heart disease: a report of theAmerican College of Cardiology/American HeartAssociation Task Force on Practice Guidelines. J Am Coll Cardiol 2006;48:e1–148. Available at:http://www.acc.org/clinical/guidelines/valvular/index.pdf. Accessed July 7, 2006.

Tab. 7

Tab. 8

Tab. 9

10090-08_CH08.qxd 8/31/06 4:51 PM Page 221

222 Chapter Eight

TABLE 8.8. Examples of Procedures That Warrant Endocarditis Prophylaxis

Dental manipulations that produce gingival bleedingUpper respiratory tract proceduresTonsillectomy and/or adenoidectomySurgical operations that involve respiratory mucosaBronchoscopy with a rigid bronchoscopeGenitourinary proceduresUrethral dilationCystoscopyProstatectomyGastrointestinal surgery, including cholecystectomy

TABLE 8.7. Modified Duke Criteria for Diagnosis of Infective Endocarditis (IE)a

Major Criteria Minor Criteria

I. Positive blood culture, defined as either A or B

A. Typical microorganism for IE from two separate blood cultures1. Streptococci viridans, S. bovis, HACEK group; or2. Staphylococcus aureus or enterococci, in the

absence of a primary focus

B. Microorganisms consistent with IE from persistently positive blood cultures1. Blood cultures drawn >12 hours apart, or2. All of three, or most of four separate cultures

drawn at least 1 hour apart3. Single positive blood culture for Coxiella burnetii

or antiphase I IgG antibody titer >1:800II. Evidence of endocardial involvement, defined

as A or BA. Echocardiogram positive for endocarditis:

1. Oscillating intracardiac mass, or2. Myocardial abscess, or3. New partial detachment of prosthetic valve

B. New valvular regurgitation

aClinical diagnosis of definitive endocarditis requires two major criteria, one major criteria plus three minor criteria, or fiveminor criteria.

HACEK, Haemophilus spp., Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp. and Kingella kingae.

Modified from Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000;30:633–638.

Predisposing cardiac condition or intravenousdrug use

Fever (≥38.0° C)

Vascular phenomena (septic arterial or pul-monary emboli, mycotic aneurysm, intracra-nial hemorrhage, conjunctival hemorrhage,Janeway lesions)

Immunologic phenomena (glomerulonephri-tis, Osler’s nodes, Roth spots, rheumatoidfactor)

Positive blood cultures not meeting major cri-teria or serologic evidence of infection withorganism consistent with IE

10090-08_CH08.qxd 8/31/06 4:51 PM Page 222

TAB

LE8.

9.Su

mm

ary

of

Maj

or

Val

vula

rC

on

dit

ion

s

Val

veLe

sio

nC

ause

sSy

mp

tom

sPh

ysic

alFi

nd

ing

sM

urm

ur

Dia

gra

m

Mit

rals

ten

osis

(MS)

Mit

ralr

egu

rgit

atio

n(M

R)

Aor

tic

sten

osis

Aor

tic

regu

rgit

atio

n

a Sym

pto

ms

ofle

ft-s

ided

hea

rtfa

ilu

rein

clu

de

exer

tion

ald

ysp

nea

,or

thop

nea

,p

arox

ysm

aln

octu

rnal

dys

pn

ea;

sym

pto

ms

ofri

ght-

sid

edh

eart

fail

ure

incl

ud

ep

erip

her

aled

ema,

abd

omin

albl

oati

ng,

and

righ

tu

pp

erq

uad

ran

tte

nd

ern

ess

(hep

atic

enla

rgem

ent)

.

A2,

aort

icp

orti

onof

hea

rtso

un

d;L

V,l

eft

ven

tric

ula

r;S 1

,firs

th

eart

sou

nd

;S2,

seco

nd

hea

rtso

un

d.

Seq

uel

aeof

rheu

mat

icfe

ver

Acu

te:

•En

doc

ard

itis

•R

up

ture

dch

ord

ae•

Pap

illa

rym

usc

led

ysfu

nct

ion

Chr

onic

:•

Rh

eum

atic

•M

itra

lpro

lap

se•

Cal

cifi

edan

nu

lus

•LV

dil

atat

ion

•D

egen

erat

ive

calc

ific

•C

onge

nit

al•

Rh

eum

atic

•C

onge

nit

al(e

.g.,

bicu

spid

valv

e)•

End

ocar

dit

is•

Rh

eum

atic

•A

orti

cro

otd

ilat

atio

n

Sym

pto

ms

ofle

ft-s

ided

(an

dla

ter

righ

t-si

ded

)h

eart

fail

ure

a

Acu

te:

•Pu

lmon

ary

edem

aC

hron

ic:

•Sy

mp

tom

sof

left

-si

ded

hea

rtfa

ilu

rea

and

low

card

iac

outp

ut

(e.g

.,fa

tigu

e)

•C

hes

tp

ain

•Sy

nco

pe

•D

ysp

nea

onex

erti

on

•D

ysp

nea

onex

erti

on•

Ch

est

pai

n(s

ome-

tim

es)

•Lo

ud

S 1(i

nea

rly

MS)

•O

pen

ing

snap

•D

iast

olic

rum

ble

(lou

des

tin

left

late

ral

dec

ubi

tus

pos

itio

n)

•H

olos

ysto

lic

mu

rmu

rat

apex

(may

bed

e-cr

esce

nd

oin

acut

eM

R)

•M

urm

ur

acce

ntu

ated

bycl

ench

ing

the

fist

s•

Wid

ely

spli

tS 2

•C

arot

ids:

del

ayed

up

-st

roke

and

dec

reas

edvo

lum

e•

Palp

atio

n:su

pra

ster

-n

alth

rill

•A

uscu

ltat

ion:

•So

ftA

2

•Sy

stol

icej

ecti

on-

typ

em

urm

ur

(lou

des

tat

up

per

-ri

ght

ster

nal

bord

er)

•W

ide

pu

lse

pre

ssu

re•

Bou

nd

ing

pu

lses

•Ea

rly

dia

stol

icd

e-cr

esce

nd

om

urm

ur

(hea

rdbe

stat

end

ex-

pir

atio

n,w

ith

pat

ien

tle

anin

gfo

rwar

d)

S2

OS

S1

S1

“MS

”

“MR

”

S2

S1

“AS

”

S2

S1

“AR

”

S2

S1

223

10090-08_CH08.qxd 8/31/06 4:51 PM Page 223

Enriquez-Sarano M., Tajik AJ. Clinical practice. Aor-tic regurgitation. N Engl J Med 2004;351:1539–1546.

Evangelista A, Tornos P, Sambola A, et al. Long-termvasodilator therapy in patients with severe aorticregurgitation. N Engl J Med 2005;353:1342–1349.

Ferrieri P. Jones Criteria Working Group. Proceed-ings of the Jones Criteria Workshop. Circulation2002;106:2521–2523.

Filsoufi F, Anyanwu AC, Salzberg SP, et al. Long-termoutcomes of tricuspid valve replacement in thecurrent era. Ann Thorac Surg 2005;80:845–850.

Fowler VG Jr, Miro JM, Hoen B, et al. Staphylococcusaureus endocarditis: a consequence of medicalprogress. JAMA 2005;293:3012–3021.

Infective endocarditis: diagnosis, antimicrobial ther-apy, and management of complications: a state-ment for healthcare professionals from the Com-mittee on Rheumatic Fever, Endocarditis, andKawasaki Disease; Council on Cardiovascular Dis-ease in the Young; and the Councils on ClinicalCardiology, Stroke, and Cardiovascular Surgeryand Anesthesia, American Heart Association—executive summary. Circulation 2005;111:3167–3184.

Li JS, Sexon DJ, Mick N, et al. Proposed modificationsto the Duke criteria for the diagnosis of infectiveendocarditis. Clin Infect Dis 2000;30:633–638.

Martin-Davila P, Fortun J, Navas E, et al. Nosocomialendocarditis in a tertiary hospital: an increasingtrend in native valve cases. Chest 2005;128:772–779.

Meira ZM, Goulart EM, Colosimo EA, et al. Long termfollow up of rheumatic fever and predictors of se-vere rheumatic valvar disease in Brazilian childrenand adolescents. Heart 2005;91:1019–1022.

Murphy RT, Garcia MJ. Role of echocardiography indiagnosis and management of endocarditis. CurrInfect Dis Rep 2005;7:257–263.

Otto CM, Salerno CT. Timing of surgery in asymp-tomatic mitral regurgitation. N Engl J Med 2005;352:928–929.

Rahimtoola SH. Valvular heart disease/cardiac sur-gery. J Am Coll Cardiol 2005;45(suppl B):20B–23B.

Rosenhek R. Statins for aortic stenosis. N Engl J Med2005;352:2441–2443.

Tawn Z, Himbert D, Brochet E, et al. Percutaneousvalve procedures: present and future. Int J Cardio-vasc Intervent 2005;7:14–20.

224 Chapter Eight

10090-08_CH08.qxd 8/31/06 4:51 PM Page 224

Chapter 8—Author Queries1. ED: Has permission been received?2. ED: Has permission been received?

10090-08_CH08.qxd 8/31/06 4:51 PM Page a

225

PHYSIOLOGYDeterminants of Contractile Function in the Intact HeartPressure-Volume Loops

PATHOPHYSIOLOGYSystolic DysfunctionDiastolic DysfunctionRight-Sided Heart Failure

COMPENSATORY MECHANISMSFrank-Starling MechanismNeurohormonal AlterationsVentricular Hypertrophy and Remodeling

MYOCYTE LOSS AND CELLULAR DYSFUNCTION

PRECIPITATING FACTORS

CLINICAL MANIFESTATIONSSymptomsPhysical SignsLaboratory Tests

PROGNOSIS

TREATMENTDiureticsVasodilatorsInotropic Drugsβ-BlockersAldosterone Antagonist TherapyAdditional TherapiesTreatment of Diastolic Dysfunction

ACUTE PULMONARY EDEMA

C H A P T E R

9Heart FailureRavi Vikram ShahMichael A. Fifer

The heart normally accepts blood at low fill-ing pressures during diastole and then pro-pels it forward at higher pressures in systole.Heart failure is defined as the inability of theheart to pump blood forward at a sufficient rateto meet the metabolic demands of the body (for-ward failure), or the ability to do so only if thecardiac filling pressures are abnormally high(backward failure), or both. Although condi-tions outside the heart may cause this defi-nition to be met through inadequate tissueperfusion (e.g., severe hemorrhage) or in-creased metabolic demands (e.g., hyperthy-

roidism), in this chapter, only cardiac causesof heart failure are considered.

Heart failure may be the final and mostsevere manifestation of nearly every form of cardiac disease, including coronary ather-osclerosis, myocardial infarction, valvulardiseases, hypertension, congenital heart dis-ease, and the cardiomyopathies. More than500,000 new cases of heart failure developin the United States each year, and approxi-mately 5 million people currently have thiscondition. It accounts for more than 12 mil-lion medical office visits annually and is the

10090-09_CH09.qxd 8/31/06 4:52 PM Page 225

most common diagnosis of hospitalized pa-tients aged 65 and older. The incidence ofheart failure is actually increasing, partly be-cause the population is aging but also be-cause of interventions that prolong survivalafter acute cardiac insults such as myocar-dial infarction.

Heart failure most commonly results fromconditions of impaired left ventricular func-tion. Thus, this chapter begins by review-ing the physiology of normal myocardial con-traction and relaxation.

PHYSIOLOGY

Experimental studies of isolated cardiac mus-cle segments have revealed several impor-tant physiologic principles that can be ap-plied to the intact heart. As an experimentalmuscle segment is stretched apart, the rela-tion between its length and the tension itpassively develops is curvilinear, reflectingits intrinsic elastic properties (Fig. 9.1A,lower curve). If the muscle is first passivelystretched and then stimulated to contractwhile its ends are held at fixed positions(termed an isometric contraction), the totaltension (active plus passive tension) gener-ated by the fibers is proportional to thelength of the muscle at the time of stimula-tion (see Fig. 9.1A, upper curve). That is,stretching the muscle before stimulation optimizes the overlap and interaction ofmyosin and actin filaments, increasing thenumber of cross bridges and the force ofcontraction. Stretching cardiac muscle fibersalso increases the sensitivity of the myofila-ments to calcium, which further augmentsforce development.

This relationship between the initial fiberlength and force development is of great im-portance in the intact heart: within a phys-iologic range, the larger the ventricular vol-ume during diastole, the more the fibers arestretched before stimulation and the greaterthe force of the next contraction. This is thebasis of the Frank-Starling relationship,the observation that ventricular output increases in relation to the preload (thestretch on the myocardial fibers before con-traction).

A second observation from the isolatedmuscle experiments arises when the fibersare not tethered at a fixed length but are al-lowed to shorten during stimulation againsta fixed load (termed the afterload). In thissituation (termed an isotonic contraction),the final length of the muscle at the end ofcontraction is directly related to the magni-tude of the load but is independent of thelength of the muscle before stimulation(see Fig. 9.1B). That is, (1) the tension gen-erated by the fiber is equal to the fixedload; (2) the greater the load opposing con-traction, the less the muscle fiber can shorten;(3) if the fiber is stretched to a longer lengthbefore stimulation but the afterload is keptconstant, the muscle will shorten a greaterdistance and will attain the same final lengthat the end of contraction; and (4) the maxi-mum tension that a fiber can produce duringisotonic contraction (i.e., such that the fiberis just unable to shorten) is the same as theforce produced by an isometric contractionfor the applied preload.

The concept of afterload is also relevantto the intact heart: the pressure generated bythe ventricle and the size of the chamber atthe end of each contraction depend on theload against which the ventricle contracts(i.e., largely the arterial pressure) but are in-dependent of the stretch on the myocardialfibers before contraction.

A third key experimental observation re-lates to myocardial contractility (also termedthe inotropic state), which accounts forchanges in the force of contraction indepen-dent of the initial fiber length and afterload.Contractility generally reflects chemical andhormonal influences on cardiac contraction,such as exposure to catecholamines. Whencontractility is enhanced pharmacologically(e.g., by a norepinephrine infusion), the re-lation between initial fiber length and forcedeveloped during contraction is shifted up-ward (see Fig. 9.1C) such that a greater totaltension develops with isometric contractionat any given preload. Similarly, when con-tractility is augmented and the cardiac mus-cle is allowed to shorten against a fixed af-terload, the fiber contracts to a greater extentand achieves a shorter final fiber length com-

226 Chapter Nine

Fig. 1

10090-09_CH09.qxd 8/31/06 4:52 PM Page 226

Heart Failure 227

b

d

ac

e

ca

f

g e

b

a

Figure 9.1. Physiology of normal cardiac muscle segments. A. Passive (lower curve) and total(upper curve) length-tension relations for isolated cat papillary muscle. Lines ab and cd represent theforce developed during isometric contractions. Initial passive muscle length c is longer (i.e., has beenstretched more) than length a and therefore has a greater passive tension. When the muscle segmentsare stimulated to contract, the muscle with the longer initial length generates greater total tension(point d versus point b). B. If the muscle fiber preparation is allowed to shorten against a fixed load,the length at the end of the contraction is dependent on the load but not the initial fiber length; stim-ulation at point a or c results in the same final fiber length (e). Thus, the muscle that starts at length cshortens a greater distance (∆Lc) than the muscle at length a (∆La). C. The uppermost curve is thelength-tension relation in the presence of the positive inotropic agent norepinephrine. For any giveninitial length, an isometric contraction in the presence of norepinephrine generates greater force (point f )than one in the absence of norepinephrine (point b). When contracting against a fixed load, the pres-ence of norepinephrine causes greater muscle fiber shortening and a smaller final muscle length (point g)compared with contraction in the absence of the inotropic agent (point e). (Adapted from DowningSE, Sonnenblick EH. Cardiac muscle mechanics and ventricular performance: force and time parame-ters. Am J Physiol 1964;207:705–715.)

10090-09_CH09.qxd 8/31/06 4:52 PM Page 227

pared with the normal state. Enhanced con-tractility is likely induced by an increase inthe cycling rate of actin-myosin cross-bridgeformation.

Determinants of ContractileFunction in the Intact Heart

In a healthy person, cardiac output ismatched to the body’s total metabolic need.Cardiac output (CO) is equal to the productof stroke volume (SV, the volume of bloodejected with each contraction) and the heartrate (HR):

The three major determinants of stroke vol-ume are preload, afterload, and myocardialcontractility, as shown in Figure 9.2.

Preload

The concept of preload (Table 9.1) in the in-tact heart was described by physiologistsFrank and Starling a century ago. In expe-rimental preparations, they showed thatwithin physiologic limits, the more a normalventricle is distended (i.e., filled with blood)during diastole, the greater the volume that isejected during the next systolic contraction.

CO SV HR= ×

This relationship is illustrated graphically bythe Frank-Starling curve, also known as theventricular function curve (Fig. 9.3). Thegraph relates a measurement of cardiac per-formance (such as cardiac output or strokevolume) on the vertical axis as a function ofpreload on the horizontal axis. As describedearlier, the preload can be thought of as theamount of myocardial stretch at the end of di-astole, just before contraction. Measurementsthat correlate with myocardial stretch, andthat are often used to indicate the preload onthe horizontal axis, are the ventricular end-

228 Chapter Nine

Figure 9.2. Key mediators of cardiac output. Deter-minants of the stroke volume include contractility, pre-load, and afterload. Cardiac output = Heart rate × Strokevolume.

TABLE 9.1. Terms Related to Cardiac Performance

Term Definition

Preload

Afterload

Contractility (inotropic state)

Stroke volume (SV)

Ejection fraction (EF)

Cardiac output (CO)Compliance

The ventricular wall tension at the end of diastole. In clinical terms, it is the stretchon the ventricular fibers just before contraction, often approximated by the end-diastolic volume or end-diastolic pressure.

The ventricular wall tension during contraction; the resistance that must be over-come for the ventricle to eject its content. Often approximated by the systolicventricular (or arterial) pressure.

Property of heart muscle that accounts for changes in the strength of contraction,independent of the preload and afterload. Often reflects chemical or hormonalinfluences (e.g., catecholamines) on the force of contraction.

Volume of blood ejected from the ventricle during systole. SV = End-diastolic volume–End-systolic volume.

The fraction of end-diastolic volume ejected from the ventricle during each systoliccontraction (normal range = 55–75%). EF = Stroke volume ÷ End-diastolic volume.

Volume of blood ejected from the ventricle per minute. CO = SV × Heart rate.Intrinsic property of a chamber that describes its pressure-volume relationship dur-

ing filling. Reflects the ease or difficulty with which the chamber can be filled.Strictl definition: Compliance = ∆ Volume ÷ ∆ Pressure.

Fig. 2

Tab. 1

Fig. 3

10090-09_CH09.qxd 8/31/06 4:52 PM Page 228

Heart Failure 229

diastolic volume (EDV) or end-diastolic pres-sure (EDP). Conditions that decrease intra-vascular volume, and thereby reduce ven-tricular preload (e.g., dehydration or severehemorrhage), result in a smaller EDV andhence a reduced stroke volume during con-traction. Conversely, an increased volumewithin the left ventricle during diastole (e.g.,a large intravenous fluid infusion) results in agreater-than-normal stroke volume.

Afterload

Afterload (see Table 9.1) in the intact heartreflects the resistance that the ventricle must

overcome to empty its content. It is moreformally defined as the ventricular wall stressthat develops during systolic ejection. Wallstress (s), like pressure, is expressed as forceper unit area, and for the left ventricle, maybe estimated from LaPlace’s relationship:

in which P is ventricular pressure, r is ven-tricular chamber radius, and h is ventricularwall thickness. Thus, ventricular wall stressincreases in response to a higher pressureload (e.g., hypertension) or an increasedchamber size (e.g., a dilated left ventricle).

sP r

h=

×2

Pulmonary congestion

Hyp

oten

sion

Figure 9.3. Left ventricular (LV) performance (Frank-Starling) curves re-late preload, measured as LV end-diastolic volume (EDV) or pressure(EDP), to cardiac performance, measured as ventricular stroke volumeor cardiac output. On the curve of a normal heart (middle line), cardiac per-formance continuously increases as a function of preload. States of increasedcontractility (e.g., norepinephrine infusion) are characterized by an aug-mented stroke volume at any level of preload (upper line). Conversely, de-creased LV contractility (commonly associated with heart failure) is character-ized by a curve that is shifted downward (lower line). Point a is an example ofa normal person at rest. Point b represents the same patient after developingsystolic dysfunction and heart failure (e.g., after a large myocardial infarction):stroke volume has fallen, and the decreased LV emptying results in elevationof the EDV. Because point b is on the ascending portion of the curve, the el-evated EDV serves a compensatory role because it results in an increase in sub-sequent stroke volume, albeit much less than if operating on the normalcurve. Further augmentation of LV filling (e.g., increased circulating volume)in the heart failure patient is represented by point c, which resides on the rel-atively flat part of the curve: stroke volume is only slightly augmented, but thesignificantly increased EDP results in pulmonary congestion.

AQ1

10090-09_CH09.qxd 8/31/06 4:52 PM Page 229

Conversely, as would be expected fromLaPlace’s relationship, an increase in wallthickness (h) serves a compensatory role inreducing wall stress, because the force is dis-tributed over a greater mass per unit surfacearea of ventricular muscle.

Contractility

In the intact heart, as in the isolated mus-cle preparation, contractility accounts forchanges in the force generated by the myo-cardium for a given set of preload and af-terload conditions. By relating a measure ofventricular performance (stroke volume orcardiac output) to preload (left ventricularend-diastolic pressure or volume), eachFrank-Starling curve is a reflection of theheart’s current inotropic state (see Fig. 9.3).The effect on stroke volume by an alter-ation in preload is reflected by a change inposition along a particular Frank-Starlingcurve. A change in contractility, on theother hand, actually shifts the entire curvein an upward or downward direction. Thus,when contractility is enhanced pharma-cologically (e.g., by an infusion of norepi-nephrine), the ventricular performancecurve is displaced upward such that at anygiven preload, the stroke volume is in-creased. Conversely, when a drug that re-duces contractility is administered or theventricle’s contractile function is impaired(as in many types of heart failure), the curveshifts in a downward direction, leading toreductions in the stroke volume and cardiacoutput at any given preload.

Pressure-Volume Loops

Another useful graphic display to illustratethe determinants of cardiac function is theventricular pressure-volume loop, which re-lates changes in ventricular volume to cor-responding changes in pressure throughoutthe cardiac cycle (Fig. 9.4). In the left ven-tricle, filling of the chamber begins after themitral valve opens in early diastole (point a).The curve between points a and b representsdiastolic filling. As the volume increasesduring diastole, it is associated with a small

rise in pressure, in accordance with the pas-sive length-tension properties or compli-ance (see Table 9.1) of the myocardium,analogous to the lower curve in Figure 9.1Afor an isolated muscle preparation.

Next, the onset of left ventricular systoliccontraction causes the ventricular pressureto rise. When the pressure in the left ventri-cle (LV) exceeds that of the left atrium (pointb), the mitral valve is forced to close. As thepressure continues to increase, the ventricu-lar volume does not immediately change,because the aortic valve has not yet opened;

230 Chapter Nine

Figure 9.4. Example of a normalleft ventricular (LV) pressure-volume loop. At point a, the mi-tral valve opens. During diastolicfilling of the LV (line ab), the vol-ume increases in association with agradual rise in pressure. When ventricular contraction commencesand its pressure exceeds that of theleft atrium, the mitral valve (MV)closes (point b) and isovolumetriccontraction of the LV ensues (theaortic valve is not yet open, and noblood leaves the chamber), asshown by line bc. When LV pres-sure rises to that in the aorta, theaortic valve (AV) opens (point c)and ejection begins. The volumewithin the LV declines during ejec-tion (line cd), but LV pressure con-tinues to rise until ventricular relax-ation commences. At point d, theLV pressure during relaxation fallsbelow that in the aorta, and theAV closes, leading to isovolumetricrelaxation (line da). As the LV pres-sure falls further, the mitral valvereopens (point a). Point b repre-sents the end-diastolic volume(EDV) and pressure, and point d isthe end-systolic volume (ESV) andpressure. Stroke volume is the dif-ference between the EDV and ESV.

Fig. 4

10090-09_CH09.qxd 8/31/06 4:52 PM Page 230

Heart Failure 231

therefore, this phase is called isovolumetriccontraction. When the ventricular pressurereaches the aortic diastolic pressure, the aor-tic valve is forced to open (point c) and ejec-tion of blood into the aorta commences.During ejection, the volume within the ven-tricle decreases, but its pressure continues torise until ventricular relaxation begins. Thepressure against which the ventricle ejects(afterload) is represented by the curve cd.Ejection ends during the relaxation phase,when the ventricular pressure falls belowthat of the aorta and the aortic valve closes(point d).

As the ventricle continues to relax, itspressure declines while its volume remainsconstant because the mitral valve has notyet opened (this phase is known as isovolu-metric relaxation). When the ventricularpressure falls below that of the left atrium,the mitral valve opens again (point a) andthe cycle repeats.

Note that point b represents the pressureand volume at the end of diastole, whereaspoint d represents the pressure and volumeat the end of systole. The difference betweenthe EDV and end-systolic volume (ESV) rep-resents the quantity of blood ejected duringcontraction (i.e., the stroke volume).

Changes in any of the determinants ofcardiac function are reflected by alterationsin the pressure-volume loop. By analyzingthe effects of a change in an individual pa-rameter (preload, afterload, or contractility)on the pressure-volume relationship, the re-sulting alterations in ventricular pressureand stroke volume can be predicted (Fig. 9.5).

Alterations in Preload

If afterload and contractility are held con-stant but preload is caused to increase (e.g.,by administration of intravenous fluids), leftventricular EDV rises. This increase in pre-load augments the stroke volume via theFrank-Starling mechanism such that theESV achieved is the same as it was before in-creasing the preload. This means that thenormal left ventricle is able to adjust itsstroke volume and effectively empty its con-tent to match its diastolic filling volume, as

long as contractility and afterload are keptconstant.

Although end-diastolic volume and end-diastolic pressure are often used interchange-ably as markers of preload, the relationshipbetween filling volume and pressure (knownas ventricular compliance; see Table 9.1),largely governs the extent of ventricular fill-ing. If ventricular compliance is reduced(e.g., in severe LV hypertrophy), the slope ofthe diastolic filling curve (see segment ab inFig. 9.4) becomes steeper. A “stiff” or poorlycompliant ventricle reduces the ability of thechamber to fill during diastole, resulting in alower-than-normal ventricular end-diastolicvolume. In this circumstance, the stroke vol-ume will be reduced while the end-systolicvolume remains unchanged.

Alterations in Afterload

If preload and contractility are held con-stant and afterload is augmented (e.g., inhigh-impedance states such as hyper ten-sion or aortic stenosis), the pressure gener-ated by the left ventricle during ejection increases. In this situation, more ventri-cular work is expended in overcoming theresistance to ejection and less fiber short-ening takes place. As shown in Figure 9-5B,an increase in afterload results in a higherventricular systolic pressure and a greater-than-normal LV end-systolic volume. Thus,in the setting of increased afterload, theventricular stroke volume (EDV − ESV) is reduced.

The dependence of the end-systolic vol-ume on afterload is approximately linear:the greater the afterload, the higher the end-systolic volume. This relationship is depictedin Figure 9.5 as the end-systolic pressure vol-ume relation (ESPVR) and is analogous to thetotal tension curve in the isolated muscle ex-periments described earlier.

Alterations in Contractility

The slope of the ESPVR line on the pressure-volume loop graph is a function of cardiaccontractility. In conditions of increased con-tractility, the ESPVR slope becomes more

Fig. 5

10090-09_CH09.qxd 8/31/06 4:52 PM Page 231

steep; that is, it shifts upward and towardthe left. Hence, at any given preload or after-load, the ventricle empties more completely(the stroke volume increases) and results ina smaller-than-normal end-systolic volume(see Fig. 9.5C). Conversely, in situations ofreduced contractility, the ESPVR line shiftsdownward, consistent with a decline instroke volume and a higher end-systolic vol-ume. Thus, the end-systolic volume is depen-dent on the afterload against which the ven-tricle contracts and the inotropic state but is

independent of the end-diastolic volume be-fore contraction.

The important physiologic concepts inthis section are summarized here:

1. Ventricular stroke volume is a function of preload, afterload, and contractility. SV rises when there is an increase in pre-load, a decrease in afterload, or augmentedcontractility.

2. Ventricular end-diastolic volume (or end-diastolic pressure) is often used as a rep-resentation of preload. The end-diastolic

232 Chapter Nine

Figure 9.5. The effect of varying preload, afterload, and contractility on the pressure-volumeloop. A. When arterial pressure (afterload) and contractility are held constant, sequential increases(lines 1, 2, 3) in preload (measured in this case as end-diastolic volume [EDV]) are associated with loopsthat have progressively higher stroke volumes but a constant end-systolic volume (ESV). B. When thepreload (EDV) and contractility are held constant, sequential increases (points 1, 2, 3) in arterial pres-sure (afterload) are associated with loops that have progressively lower stroke volumes and higher end-systolic volumes. There is a nearly linear relationship between the afterload and ESV, termed the end-systolic pressure-volume relation (ESPVR). C. A positive inotropic intervention shifts the end-systolicpressure-volume relation upward and leftward from ESPVR-1 to ESPVR-2, resulting in loop 2, whichhas a larger stroke volume and a smaller end-systolic volume than the original loop 1.

10090-09_CH09.qxd 8/31/06 4:52 PM Page 232

Heart Failure 233

volume is influenced by the chamber’scompliance.

3. Ventricular end-systolic volume dependson the afterload and contractility but noton the preload.

PATHOPHYSIOLOGY

Chronic heart failure may result from awide variety of cardiovascular insults. Theetiologies can be grouped into those thatcause heart failure because of (1) impairedventricular contractility, (2) increased after-load, or (3) impaired ventricular filling.Heart failure that results from an abnormal-ity of ventricular emptying (owing to im-paired contractility or excessive afterload) istermed systolic dysfunction, whereas heartfailure caused by abnormalities of diastolicrelaxation or ventricular filling is termed diastolic dysfunction. Approximately twothirds of patients with heart failure demon-strate predominantly systolic dysfunction,and the remainder primarily suffer from di-astolic dysfunction. However, there is muchoverlap, and many patients demonstrateboth systolic and diastolic abnormalities.Figure 9.6 presents a general schema of car-diac conditions that may result in heart fail-ure. Although this schema applies to chronicforms of heart failure, a sudden overwhelm-ing cardiac load (as may occur with an acutehypertensive crisis [see Chapter 13], acutemyocardial infarction [see Chapter 7], oracute valvular insufficiency [see Chapter 8])can result in an acute form of heart failure(e.g., pulmonary edema), as discussed laterin this chapter.

Systolic Dysfunction

In systolic dysfunction, the affected ven-tricle has a diminished capacity to ejectblood because of impaired myocardial contractility or pressure overload (i.e., ex-cessive afterload). Loss of contractility may result from destruction of myocytes, ab-normal myocyte function, or fibrosis. Pres-sure overload impairs ventricular ejectionby significantly increasing resistance toflow.

Figure 9.7A depicts the effects of systolicdysfunction owing to impaired contractilityon the pressure-volume loop. The ESPVR isshifted downward such that systolic empty-ing ceases at a higher-than-normal end-systolic volume. As a result, the stroke vol-ume falls. When normal pulmonary venousreturn is added to the increased end-systolicvolume that has remained in the ventriclebecause of incomplete emptying, the dias-tolic chamber volume increases, resulting ina higher-than-normal end-diastolic volumeand pressure. While that increase in preloadinduces a compensatory rise in stroke vol-ume (via the Frank-Starling mechanism),impaired contractility and the reduced ejec-tion fraction cause the end-systolic volumeto remain elevated.

During diastole, the persistently elevatedLV pressure is transmitted to the left atrium(through the open mitral valve) and to thepulmonary veins and capillaries. An elevatedpulmonary capillary hydrostatic pressure,when sufficiently high (usually >20 mm Hg),results in the transudation of fluid into thepulmonary interstitium and symptoms ofpulmonary congestion.

Diastolic Dysfunction

Approximately one third of patients withclinical heart failure have normal ventricularcontractile (systolic) function. Many of thesepeople demonstrate abnormalities of diastolicfunction: either impaired early diastolic relax-ation (an active, energy-dependent process),increased stiffness of the ventricular wall (apassive property), or both. Acute myocardialischemia is an example of a condition thattransiently inhibits energy delivery and dias-tolic relaxation. Conversely, left ventricularhypertrophy, fibrosis, or restrictive cardiomy-opathy (see Chapter 10) causes the LV walls tobecome chronically stiffened. The effect of im-paired diastolic function is reflected in thepressure-volume loop (see Fig. 9.7B): in dias-tole, filling of the ventricle occurs at higher-than-normal pressures because the lower partof the loop is shifted upward as a result of re-duced chamber compliance. Patients with di-astolic dysfunction often present with signs of

Fig. 6

Fig. 7

10090-09_CH09.qxd 8/31/06 4:52 PM Page 233

234 Chapter Nine

Figure 9.6. Mechanisms and examples of conditions that cause left-sidedheart failure.

vascular congestion because the elevated di-astolic pressure is transmitted retrograde tothe pulmonary and systemic veins.

Right-Sided Heart Failure

Whereas the physiologic principles systolicand diastolic dysfunction may be applied toboth right-sided and left-sided heart failure,the two ventricles have distinctly differentfunctions. Compared with the left ventricle,

the right ventricle (RV) is a thin-walled,highly compliant chamber that accepts itsblood volume at low pressures and ejectsagainst a low pulmonary vascular resistance.As a result of its high compliance, the RV haslittle difficulty accepting a wide range of fill-ing volumes without significant changes inits filling pressures. Conversely, the RV isquite susceptible to failure in situations thatpresent a sudden increase in afterload, suchas acute pulmonary embolism.

10090-09_CH09.qxd 8/31/06 4:52 PM Page 234

Heart Failure 235

The most common cause of right-sidedheart failure is actually the presence of left-sided heart failure (Table 9.2). In this situa-tion, excessive afterload confronts the rightventricle because of the elevated pulmonaryvascular pressures that result from LV dys-function. Isolated right-heart failure is lesscommon and usually reflects increased RV af-terload owing to diseases of the lung paren-chyma or pulmonary vasculature. Right-sidedheart disease that results from a primary pul-monary process is known as cor pulmonale,which often leads to right-heart failure.

When the right ventricle fails, the ele-vated diastolic pressure is transmitted retro-grade to the right atrium with subsequentcongestion of the systemic veins, accompa-nied by signs of right-sided heart failure. In-directly, isolated right-heart failure may alsoinfluence left-heart function: the decreasedright ventricular output reduces blood re-turn to the LV (a decline in preload), leadingto a drop in left ventricular stroke volume.

COMPENSATORY MECHANISMS

Several natural compensatory mechanismsare called into action in patients with heartfailure, serving to buffer the fall in cardiacoutput and helping to maintain sufficientblood pressure to perfuse the vital organs.These mechanisms include (1) the Frank-Starling mechanism, (2) neurohormonalalterations, and (3) the development ofventricular hypertrophy and remodeling(Fig. 9.8).

Frank-Starling Mechanism

As shown in Figure 9.3, heart failure causedby impaired left ventricular contractile func-tion causes a downward shift of the ven-

TABLE 9.2. Examples of Conditions ThatCause Right-Sided Heart Failure

Cardiac causesLeft-sided heart failurePulmonic valve stenosisRight ventricular infarctionParenchymal pulmonary diseaseChronic obstructive pulmonary diseaseInterstitial lung disease (e.g., sarcoidosis)Adult respiratory distress syndromeChronic lung infection or bronchiectasisPulmonary vascular diseasePulmonary embolismPrimary pulmonary hypertension

AQ2

2

2

Figure 9.7. Systolic dysfunction and the pressure-volume loop. A. The normal pressure-volume loop (solid line)is compared with one demonstrating systolic dysfunction (dashed line). In systolic dysfunction caused by decreasedcardiac contractility, the end-systolic pressure-volume relation is shifted downward and rightward (from line 1 to line2). As a result, the end-systolic volume (ESV) is increased (arrow). As normal venous return is added to that greater-than-normal ESV, there is an obligatory increase in the end-diastolic volume (EDV) and pressure (preload), which servesa compensatory function by partially elevating stroke volume toward normal via the Frank-Starling mechanism. B. The pressure-volume loop of diastolic dysfunction resulting from increased stiffness of the ventricle (dashed line).The passive diastolic pressure-volume curve is shifted upward (from line 1 to line 2) such that at any diastolic volume,the ventricular pressure is higher than normal. The result is a decreased EDV (arrow) because of reduced filling of thestiffened ventricle at a higher-than-normal end-diastolic pressure.

Tab. 2

Fig. 8

10090-09_CH09.qxd 8/31/06 4:52 PM Page 235

tricular performance curve. Therefore, at agiven preload, stroke volume is decreasedcompared with normal. The reduced strokevolume results in incomplete chamber emp-tying; as a result, the volume of blood thataccumulates in the ventricle during diastoleis higher than normal (see Fig. 9.3, point b).This increased stretch on the myofibers, act-ing via the Frank-Starling mechanism, in-duces a greater stroke volume on subsequentcontraction, which helps to empty the en-larged left ventricle and preserve forwardcardiac output (see Fig 9.8).

This beneficial compensatory mechanismhas its limits, however. In the case of severeheart failure with marked depression of contractility, the curve may be nearly flat at higher diastolic volumes, reducing the augmentation of cardiac output achievedby increased filling. Concurrently in such acircumstance, marked elevation of the end-diastolic volume and pressure (which is trans-mitted retrograde to the left atrium, pul-monary veins, and capillaries) may result inpulmonary congestion and edema (see Fig.9.3, point c).

Neurohormonal Alterations

Neurohormonal activation encompassesthree important compensatory mechanismsin heart failure in response to decreasedcardiac output (Fig. 9.9): (1) the adrenergic

nervous system, (2) the renin-angiotensin-aldosterone system, and (3) increased pro-duction of antidiuretic hormone (ADH). Inpart, these mechanisms serve to increasesystemic vascular resistance, which helps tomaintain arterial perfusion to vital organs,even in the setting of a reduced cardiac out-put. That is, because blood pressure (BP) isequal to the product of cardiac output (CO)and total peripheral resistance (TPR),

a rise in TPR induced by these compensatorymechanisms can nearly balance the fall inCO and, in the early stages of heart failure,maintain fairly normal BP. In addition,neurohormonal activation results in saltand water retention, which in turn increasesintravascular volume and left ventricularpreload, maximizing stroke volume via theFrank-Starling mechanism.

Although the acute effects of neurohor-monal stimulation are compensatory andbeneficial, chronic activation of these mech-anisms often ultimately proves deleteriousto the failing heart and contributes to a pro-gressive downhill course, as described laterin the chapter.

Adrenergic Nervous System

The fall in cardiac output in heart failure is sensed as decreased perfusion pressure

BP CO TPR= ×

236 Chapter Nine

Figure 9.8. Compensatory mechanisms in heart failure. Both the Frank-Starling mechanism (which is invoked by the rise in ventricular end-diastolicvolume) and myocardial hypertrophy (in response to pressure or volumeoverload) serve to maintain forward stroke volume (dashed lines). However,the chronic rise in EDV by the former and increased ventricular stiffness bythe latter cause an increase in atrial pressure, which may result in manifes-tations of “backward” failure (e.g., pulmonary congestion in the case of left-sided heart failure).

AQ3

Fig. 9

10090-09_CH09.qxd 8/31/06 4:52 PM Page 236

Heart Failure 237

by baroreceptors in the carotid sinus andaortic arch. These receptors decrease theirrate of firing in proportion to the fall in BP, and the signal is transmitted by theninth and tenth cranial nerves to the car-diovascular control center in the medulla.As a result, sympathetic outflow to the heartand peripheral circulation is increased, and parasympathetic tone is diminished.Three immediate consequences arise (seeFig. 9.9): (1) an increase in heart rate, (2) aug-mentation of ventricular contractility, and(3) vasoconstriction caused by stimulationof α-receptors on the systemic veins and arteries.

The increased heart rate and ventricularcontractility directly augment cardiac out-

put (see Fig. 9.2). Vasoconstriction of the ve-nous and arterial circulations is also initiallybeneficial. Venous constriction augmentsblood return to the heart, which increasespreload and raises stroke volume throughthe Frank-Starling mechanism, as long asthe ventricle is operating on the ascendingportion of its ventricular performance curve(see Fig. 9.3). Arteriolar constriction in-creases the peripheral vascular resistance andtherefore helps to maintain blood pressure(BP = CO × TPR). The regional distribution ofα-receptors is such that during sympatheticstimulation, blood flow is redistributed tovital organs (e.g., heart and brain) at the expense of the skin, splanchnic viscera, andkidneys.

Figure 9.9. Compensatory neurohormonal stimulation develops in responseto the reduced forward cardiac output and blood pressure of heart failure. In-creased activity of the sympathetic nervous system, renin-angiotensin-aldosterone sys-tem, and antidiuretic hormone serve to support the cardiac output and blood pres-sure (boxes). However, adverse consequences of these activations (dashed lines)include an increase in afterload from excessive vasoconstriction (which may then im-pede cardiac output) and excess fluid retention, which contributes to peripheraledema and pulmonary congestion.

AQ8

10090-09_CH09.qxd 8/31/06 4:52 PM Page 237

Renin-Angiotensin-Aldosterone System

This system is also activated early in patientswith heart failure (see Fig. 9.9), mediated byincreased renin release. The main stimuli forrenin secretion from the juxtaglomerularcells of the kidney in heart failure patientsinclude (1) decreased renal artery perfusionpressure secondary to low cardiac output,(2) decreased salt delivery to the maculadensa of the kidney owing to alterations inintrarenal hemodynamics, and (3) directstimulation of juxtaglomerular β2-receptorsby the activated adrenergic nervous system.

Renin is an enzyme that cleaves circulat-ing angiotensinogen to form angiotensin I,which is then rapidly cleaved by endothelialcell–bound angiotensin-converting enzyme(ACE) to form angiotensin II (AII), a potentvasoconstrictor (see Chapter 13). IncreasedAII constricts arterioles and raises total pe-ripheral resistance, thereby helping to main-tain systemic blood pressure. In addition,AII acts to increase intravascular volume by two mechanisms: (1) at the hypothala-mus, it stimulates thirst and therefore waterintake; and (2) at the adrenal cortex, it acts to increase aldosterone secretion. The latterhormone promotes sodium reabsorptionfrom the distal convoluted tubule of the kid-ney into the circulation (see Chapter 17),serving to augment intravascular volume.The rise in intravascular volume increasesleft ventricular preload and thereby aug-ments cardiac output via the Frank-Starlingmechanism in patients on the ascending por-tion of the ventricular performance curve(see Fig. 9.3).

Antidiuretic Hormone

Secretion of this hormone (also termed va-sopressin) by the posterior pituitary is in-creased in many patients with heart failure,presumably mediated through arterialbaroreceptors, and by increased levels of AII.ADH contributes to increased intravascularvolume because it promotes water retentionin the distal nephron. The increased intra-vascular volume serves to augment left ven-tricular preload and cardiac output. ADH

also appears to contribute to systemic vaso-constriction.

Although each of these neurohormonalalterations in heart failure is initially benefi-cial, continued activation typically provesharmful. For example, the increased circu-lating volume and augmented venous returnto the heart may ultimately worsen engorge-ment of the lung vasculature, exacerbatingcongestive pulmonary symptoms. Further-more, the elevated arteriolar resistance increases the afterload against which the failing left ventricle contracts and may there-fore impair stroke volume and reduce car-diac output (see Fig. 9.9). In addition, the in-creased heart rate augments metabolicdemand and can therefore further reduce theperformance of the failing heart. Continuoussympathetic activation results in downregu-lation of cardiac β-adrenergic receptors andupregulation of inhibitory G proteins, con-tributing to a decrease in the myocardium’ssensitivity to circulating catecholamines anda reduced inotropic response.

Chronically elevated levels of AII and al-dosterone have additional detrimental ef-fects. They provoke the production of cy-tokines (small proteins that mediate cell-cellcommunication and immune responses),activate macrophages, and stimulate fibro-blasts, resulting in fibrosis and adverse re-modeling of the failing heart.

Because the undesired consequences ofchronic neurohormonal activation even-tually outweigh their benefits, much oftoday’s pharmacologic therapy of heart fail-ure is designed to moderate these “compen-satory” mechanisms, as examined later inthe chapter.

Natriuretic Peptides

In contrast to the adverse consequences ofthe neurohormonal alterations described in the previous sections, the natriuretic peptides are natural “beneficial” hormonessecreted in heart failure in response to in-creased intracardiac pressures. The best stud-ied of these are atrial natriuretic peptide(ANP) and B-type natriuretic peptide (BNP).

238 Chapter Nine

10090-09_CH09.qxd 8/31/06 4:52 PM Page 238

Heart Failure 239

ANP is stored in atrial cells and released inresponse to atrial distention. BNP is not de-tected in normal hearts but produced whenventricular myocardium is subjected to he-modynamic stress (e.g., in heart failure orduring myocardial infarction). Recent stud-ies have shown a close relationship betweenserum BNP levels and the clinical severity ofheart failure.

Actions of the natriuretic peptides are me-diated by specific natriuretic receptors andare largely opposite to those of the otherhormone systems activated in heart failure.They result in excretion of sodium and water,vasodilatation, inhibition of renin secretion,and antagonism of the effects of AII on aldosterone and vasopressin secretion. Al-though these effects are beneficial to patientswith heart failure, they are usually not suffi-cient to fully counteract the vasoconstric-tion and volume-retaining effects of theother activated hormonal systems.

Another recently recognized substancethat is released in heart failure is endothelin1, a potent vasoconstrictor, derived from en-dothelial cells lining the vasculature (seeChapter 6). Drugs designed to inhibit en-dothelin-1 receptors (and therefore bluntadverse vasoconstriction) improve LV func-tion in heart failure, but long-term benefitshave not yet been demonstrated.

Ventricular Hypertrophy and Remodeling

Ventricular hypertrophy and remodelingare important compensatory processes thatdevelop over time in response to hemody-namic burdens. Wall stress (as defined ear-lier and in Chapter 6) is often increased indeveloping heart failure because of either LVdilatation (increased chamber radius) or theneed to generate high systolic pressures toovercome excessive afterload (e.g., in aorticstenosis or hypertension). A sustained in-crease in wall stress (along with neurohor-monal and cytokine alterations) stimulatesthe development of myocardial hypertro-phy and deposition of extracellular matrix.This increased mass of muscle fibers servesas a compensatory mechanism that helps to

maintain contractile force and counteractsthe elevated ventricular wall stress (recallthat wall thickness is in the denominator ofthe LaPlace wall stress formula). However,because of the increased stiffness of the hypertrophied wall, these benefits come at the expense of higher-than-normal diastolicventricular pressures, which are transmittedto the left atrium and pulmonary vascula-ture (see Fig. 9.8).

The pattern of compensatory hypertro-phy and remodeling that develops dependson whether the ventricle is subjected tochronic volume or pressure overload. Chro-nic chamber dilatation owing to volumeoverload (e.g., chronic mitral or aortic regur-gitation) results in the synthesis of new sar-comeres in series with the old, causing themyocytes to elongate. The radius of the ven-tricular chamber therefore enlarges, doing soin proportion to the increase in wall thick-ness, and is termed eccentric hypertrophy.Chronic pressure overload (e.g., caused byhypertension or aortic stenosis) results in thesynthesis of sarcomeres in parallel with theold (i.e., the myocytes thicken), termed con-centric hypertrophy. In this situation, thewall thickness increases without propor-tional chamber dilatation, and wall stressmay therefore be reduced substantially.

Such hypertrophy and remodeling help toreduce wall stress and maintain contractileforce, but ultimately, ventricular function de-teriorates and causes the chamber to dilateout of proportion to wall thickness. Whenthis occurs, the excessive hemodynamicburden on the contractile units produces adownward spiral of deterioration with pro-gressive heart failure symptomatology.

MYOCYTE LOSS AND CELLULAR DYSFUNCTION

Impairment of ventricular function in heartfailure may result from the actual loss ofmyocytes and/or impaired function of livingmyocytes. The loss of myocytes may resultfrom cellular necrosis (e.g., from myocardial in-farction or exposure to cardiotoxic drugs suchas doxorubicin) or apoptosis (programmed celldeath). In apoptosis, genetic instructions

AQ4

10090-09_CH09.qxd 8/31/06 4:52 PM Page 239

activate intracellular pathways that cause thecell to fragment and undergo phagocytosisby other cells, without an inflammatory re-sponse. Factors implicated in triggering apop-tosis in heart failure include elevated cate-cholamines, AII, inflammatory cytokines,and mechanical strain on the myocytesowing to the augmented wall stress.

Even viable myocardium in heart failureis abnormal at the ultrastructural and mole-cular levels. Mechanical wall stress, neuro-hormonal activation, and inflammatory cytokines, such as tumor necrosis factor α(TNF-α), are believed to activate changes inthe genetic expression of contractile pro-teins, ion channels, catalytic enzymes, sur-face receptors, and secondary messengers inthe myocyte. Recent experimental evidencehas demonstrated such changes at the sub-cellular level that affect intracellular calciumhandling by the sarcoplasmic reticulum, decrease the responsiveness of the myofilaments to calcium, impair excitation-contraction coupling, and alter cellular en-ergy production. The cellular factors consid-ered the most important contributors todysfunction in heart failure are (1) a reducedcellular ability to maintain calcium homeo-stasis, and/or (2) changes in the production,availability, and utilization of high-energyphosphates. However, the exact subcellularalterations that result in heart failure havenot yet been elucidated, and this remainsone of the most active areas of cardiovascu-lar research.

PRECIPITATING FACTORS

Many patients with heart failure remainasymptomatic for extended periods eitherbecause the impairment is mild or becausecardiac dysfunction is balanced by the com-pensatory mechanisms described earlier.Often clinical manifestations occur only inthe presence of precipitating factors that in-crease the cardiac workload and tip the bal-anced state into one of decompensation.

Common precipitating factors are listedin Table 9.3. For example, conditions of in-creased metabolic demand such as fever orinfection may not be matched by a suffi-

cient increase in output by the failing heart,so that symptoms of cardiac insufficiencyare precipitated. Tachyarrhythmias precipi-tate heart failure by decreasing diastolic ven-tricular filling time and by increasing myo-cardial oxygen demand. Excessively lowheart rates directly cause a drop in cardiacoutput (remember, cardiac output = strokevolume × heart rate). An increase in salt in-gestion, renal dysfunction, or failure to takeprescribed diuretic medications may in-crease the circulating volume, thus promot-ing systemic and pulmonary congestion.Uncontrolled hypertension depresses sys-tolic function because of excessive afterload.A large pulmonary embolism results in bothhypoxemia (and therefore decreased my-ocardial oxygen supply) and a substantialincrease in right ventricular afterload. Is-chemic insults (i.e., myocardial ischemia orinfarction), ethanol ingestion, or negativeinotropic medications (e.g., large doses of β-blockers and certain calcium channel block-ers) can all depress myocardial contractilityand precipitate symptoms in the otherwisecompensated congestive heart failure patient.

240 Chapter Nine

TABLE 9.3. Factors That May PrecipitateSymptoms in Compensated Heart Failure

Increased metabolic demandsFeverInfectionAnemiaTachycardiaHyperthyroidismPregnancyIncreased circulating volume (increased preload)Excessive sodium content in dietExcessive fluid administrationRenal failureConditions that increase afterloadUncontrolled hypertensionPulmonary embolism (increased right ventricular

afterload)Conditions that impair contractilityNegative inotropic medicationsMyocardial ischemia or infarctionEthanol ingestionFailure to take prescribed heart failure

medicationsExcessively slow heart rate

Tab. 3

10090-09_CH09.qxd 8/31/06 4:52 PM Page 240

Heart Failure 241

CLINICAL MANIFESTATIONS

The clinical manifestations of heart failureresult from impaired forward cardiac outputand/or elevated venous pressures, and relateto which of the ventricles has failed (Table9.4). A patient may present with the chronicprogressive symptoms of heart failure de-scribed here or, in certain cases, with suddendecompensation of left-sided heart function(e.g., acute pulmonary edema, as describedlater in the chapter).

Symptoms

The most prominent symptom of chronicleft ventricular failure is dyspnea (breathless-ness) on exertion. Controversy regardingthe cause of this symptom has centered onwhether it is primarily a manifestation ofpulmonary venous congestion or decreasedforward cardiac output. A pulmonary ve-nous pressure that exceeds approximately20 mm Hg leads to transudation of fluid intothe pulmonary interstitium and congestionof the lung parenchyma. The resulting re-duced pulmonary compliance increases thework of breathing to move the same volumeof air. Moreover, the excess fluid in the in-terstitium compresses the walls of the bron-chioles and alveoli, increasing the resistanceto airflow and requiring greater effort of res-piration. In addition, juxtacapillary recep-tors (J receptors) are stimulated and mediaterapid shallow breathing. The heart failurepatient can also suffer from dyspnea even in

the absence of pulmonary congestion, be-cause reduced forward blood flow to theoverworked respiratory muscles and accu-mulation of lactic acid may also contributeto that sensation. Heart failure may initiallycause dyspnea only on exertion, but moresevere dysfunction results in symptoms atrest as well.

Other manifestations of low forward out-put in heart failure may include dulled men-tal status because of reduced cerebral perfu-sion and impaired urine output during the daybecause of decreased renal perfusion. Thelatter often gives way to increased urinaryfrequency at night (nocturia) when, whilesupine, blood flow is redistributed to thekidney, promoting renal perfusion and di-uresis. Reduced skeletal muscle perfusionmay result in fatigue and weakness.

Other congestive manifestations of heartfailure include orthopnea, paroxysmal noctur-nal dyspnea (PND), and nocturnal cough. Or-thopnea is the sensation of labored breath-ing while lying flat and is relieved bysitting upright. It results from the redistri-bution of intravascular blood from thegravity-dependent portions of the body (ab-domen and lower extremities) toward thelungs after lying down. The degree of or-thopnea is generally assessed by the numberof pillows on which the patient sleeps toavoid breathlessness. Sometimes, orthopneais so significant that the patient may try tosleep upright in a chair.

PND is severe breathlessness that awak-ens the patient from sleep 2 to 3 hours after

TABLE 9.4. Most Common Symptoms and Physical Findings in Heart Failure

Symptoms Physical Findings

Left-sidedDyspnea Diaphoresis (sweating)Orthopnea Tachycardia, tachypneaParoxysmal nocturnal dyspnea Pulmonary ralesFatigue Loud P2

Right-sidedPeripheral edema Jugular venous distentionRight upper quadrant discomfort Hepatomegaly

(owing to hepatic enlargement)Peripheral edema

Tab. 4

10090-09_CH09.qxd 8/31/06 4:52 PM Page 241

retiring to bed. This frightening symptomresults from the gradual reabsorption intothe circulation of lower extremity intersti-tial edema after lying down, with subse-quent expansion of intravascular volumeand increased venous return to the heartand lungs. A nocturnal cough is anothersymptom of pulmonary congestion and isproduced by a mechanism similar to orthop-nea. Hemoptysis (coughing up blood) mayresult from rupture of engorged bronchialveins.

In right-sided heart failure, the elevatedsystemic venous pressures can result in ab-dominal discomfort because the liver becomesengorged and its capsule stretched. Similarly,anorexia (decreased appetite) and nauseamay result from edema within the gastroin-testinal tract. Peripheral edema, especially inthe ankles and feet, also reflects increasedhydrostatic venous pressures. Because of theeffects of gravity, it tends to worsen whilethe patient is upright during the day and is often improved by morning after lyingsupine at night. Even before peripheral ede-ma develops, the patient may note an unex-pected weight gain resulting from the accu-mulation of interstitial fluid.

The symptoms of heart failure are com-monly graded according to the New YorkHeart Association (NYHA) classification(Table 9.5). A newer system classifies pa-tients according to their stage in the courseof heart failure (Table 9.6).

Physical Signs

The physical signs of heart failure depend onthe severity and chronicity of the conditionand can be divided into two groups based onleft and right-heart dysfunction (see Table

9.4). Patients with only mild impairmentmay appear well. However, a patient with se-vere chronic heart failure may demonstratecachexia (a frail, wasted appearance) owingin part to poor appetite and to the metabolicdemands of the increased effort of breathing.In decompensated left-sided heart failure,the patient may appear dusky (decreased cardiac output) and diaphoretic (sweating because of increased sympathetic nervousactivity), and the extremities are cool be-cause of peripheral arterial vasoconstriction.Tachypnea (rapid breathing) is common. Thepattern of Cheyne-Stokes respiration mayalso be present in advanced heart failure,characterized by periods of hyperventila-tion separated by intervals of apnea (absentbreathing). This pattern is related to the pro-longed circulation time between the lungsand respiratory center of the brain in heartfailure that interferes with the normal feed-back mechanism of systemic oxygenation.Sinus tachycardia (resulting from increasedsympathetic nervous system activity) is alsocommon. Pulsus alternans (alternating strongand weak contractions detected in the pe-ripheral pulse) may be present as a sign of ad-vanced ventricular dysfunction.

In left-sided heart failure, the auscultatoryfinding of pulmonary rales is created by the“popping open” of small airways that hadbeen closed off by edema fluid before inspi-ration. This finding is initially apparent atthe lung bases, where hydrostatic forces aregreatest; however, more severe pulmonarycongestion is associated with additional raleshigher in the lung fields. Compression of con-duction airways by pulmonary congestionmay produce coarse rhonchi and wheezing;the latter finding in heart failure is termedcardiac asthma.

242 Chapter Nine

TABLE 9.5. New York Heart Association Classification of Heart Failure

Class Definition

I No limitation of physical activity.II Slight limitation of activity. Dyspnea and fatigue with moderate physical activity (e.g., walk-

ing up stairs quickly).III Marked limitation of activity. Dyspnea with minimal activity (e.g., slowly walking up stairs).IV Severe limitation of activity. Symptoms are present even at rest.

Tab. 5

Tab. 6

10090-09_CH09.qxd 8/31/06 4:52 PM Page 242

Heart Failure 243

Depending on the cause of heart failure,palpation of the heart may show that the leftventricular impulse is not focal but diffuse (indilated cardiomyopathy), sustained (in pres-sure overload states such as aortic stenosis orhypertension), or lifting in quality (in vol-ume overload states such as mitral regurgita-tion). Because elevated left-heart filling pres-sures result in increased pulmonary vascularpressures, the pulmonic component of thesecond heart sound is often louder than normal. An early diastolic sound (S3) is fre-quently heard in adults with systolic heartfailure and is caused by abnormal filling ofthe dilated chamber (see Chapter 2). A late di-astolic sound (S4) results from forceful atrialcontraction into a stiffened ventricle and iscommon in states of decreased LV compli-ance (diastolic dysfunction). The murmur ofmitral regurgitation is sometimes auscultatedin left-sided heart failure if LV dilatation hasstretched the valve annulus and spread thepapillary muscles apart from one another, thuspreventing full closure of the mitral leafletsin systole.

In right-sided heart failure, different phys-ical findings may be present. Cardiac exami-nation may reveal a palpable parasternal rightventricular heave, representing RV enlargement,or a right-sided S3 or S4 gallop. The murmur oftricuspid regurgitation may be auscultated andis owing to right ventricular enlargement,analogous to mitral regurgitation that devel-ops in LV dilatation. The elevated systemicvenous pressure produced by right-heart fail-ure is manifested by distention of the jugular

veins as well as hepatic enlargement with ab-dominal right upper quadrant tenderness.Edema accumulates in the dependent por-tions of the body, beginning in the anklesand feet of ambulatory patients and in thepresacral regions of those who are bedridden.

Pleural effusions may develop in eitherleft- or right-sided heart failure, because thepleural veins drain into both the systemicand pulmonary venous beds. The presenceof pleural effusions is suggested on physicalexamination by dullness to percussion overthe posterior lung bases.

Diagnostic Studies

Normally, the mean left atrial (LA) pressureis ≤10 mm Hg. If the LA pressure exceeds ap-proximately 15 mm Hg, the chest radio-graph shows upper-zone vascular redistribu-tion, such that the vessels supplying theupper lobes of the lung are larger than thosesupplying the lower lobes (see Fig. 3.5). Thisis explained as follows: When a patient is inthe upright position, blood flow is normallygreater to the lung bases than to the apicesbecause of the effect of gravity. Redistribu-tion of flow occurs with the development ofinterstitial and perivascular edema, which ismost prominent at the lung bases (where thehydrostatic pressure is the highest). There-fore, the blood vessels in the lung bases arecompressed, whereas flow into the upperlung zones is less affected.

When the LA pressure surpasses 20 mmHg, interstitial edema is usually manifested

TABLE 9.6. Stages of Heart Failure

Stage Description

A

B

CD

Modified from Hunt SA, Baker DW, Chin MH, et al. ACC/AHA guidelines for the evaluation and management of chronic heartfailure in the adult: executive summary. Circulation 2001;104:2996–3007.

Patient who is at risk of developing heart failure but has not yet developed structural cardiacdysfunction (e.g., patient with coronary artery disease, hypertension, or family history of cardiomyopathy)

Patient who has structural heart disease associated with heart failure but has not yet developedsymptoms

Patient who has current or prior symptoms of heart failure associated with structural heart diseasePatient who has structural heart disease and marked heart failure symptoms despite maximal med-

ical therapy and requires advanced interventions (e.g., cardiac transplantation)

AQ5

10090-09_CH09.qxd 8/31/06 4:52 PM Page 243

on the chest radiograph as indistinctness ofthe vessels and the presence of Kerley B lines(short linear markings at the periphery ofthe lower lung fields indicating interlobularedema). If the LA pressure exceeds 25 to 30 mm Hg, alveolar pulmonary edema maydevelop, with opacification of the air spaces.The relationship between LA pressure andchest radiograph findings is modified in pa-tients with chronic heart failure because ofenhanced lymphatic drainage, such thathigher pressures can be accommodated withfewer radiologic signs.

Depending on the cause of heart failure,the chest radiograph may show cardio-megaly, defined as a cardiothoracic ratio ofgreater than 0.5 on the posteroanterior film.A high right atrial pressure also causes en-largement of the azygous vein silhouette.Pleural effusions may be present.

Assays for BNP, described earlier in thechapter, correlate well with the degree of LVdysfunction and prognosis. Furthermore, anelevated serum level of BNP can help distin-guish heart failure from other causes of dys-pnea, such as pulmonary diseases.

The cause of heart failure is often evidentfrom the history, such as a patient who hassustained a large myocardial infarction, orby physical examination, as in a patient withthe murmur of mitral stenosis. When thecause is not clear from clinical evaluation,the first step is to determine whether systolicventricular function is normal or depressed(see Fig. 9.6). Of the several noninvasive teststhat can help make this determination,echocardiography is especially useful (as de-scribed in Chapter 3). In a minority of cases,cardiac catheterization is necessary to deter-mine the cause of heart failure, includingspecific valvular and ischemic etiologies.

PROGNOSIS

The prognosis of heart failure is dismal inthe absence of a correctable underlyingcause. Only 50% of patients remain alive 5 years after the diagnosis is made. Patientswith severe symptoms (i.e., NYHA class III orIV) fare the least well, having a 1-year sur-vival rate of only 40%. The greatest mortal-

ity is owing to refractory heart failure, butmany patients die suddenly, presumably be-cause of ventricular arrhythmias.

Ventricular dysfunction usually beginswith an inciting insult but is a progressiveprocess, contributed to by the maladaptiveactivation of neurohormones, cytokines, andcontinuous ventricular remodeling. Thus, it should not be surprising that measures of neurohormonal and cytokine stimulationpredict survival in heart failure patients. Forexample, adverse prognosis correlates withthe serum norepinephrine level (marker of sympathetic nervous system activity),serum sodium (reduced level reflects activa-tion of renin-angiotensin-aldosterone sys-tem and alterations in intrarenal hemody-namics), endothelin 1, and cytokine TNF-αlevels.

Despite the generally bleak prognosis, re-cent studies have demonstrated that survivalin heart failure patients can be substantiallyprolonged by specific interventions, as dis-cussed in the following section.

TREATMENT

There are five main goals of therapy in pa-tients with chronic heart failure:

1. Identification and correction of the under-lying condition causing heart failure. Insome patients, this may require surgicalrepair or replacement of dysfunctionalcardiac valves, coronary artery revascu-larization, aggressive treatment of hy-pertension, or cessation of alcohol con-sumption.

2. Elimination of the acute precipitating causeof symptoms in the patient with heartfailure who was previously in a com-pensated state. This may include, for example, treating acute infections or ar-rhythmias, removing sources of excessivesalt intake, or eliminating drugs that canaggravate symptomatology (e.g., certaincalcium channel blockers, which have anegative inotropic effect, or nonsteroidalanti-inflammatory drugs, which can con-tribute to volume retention).

244 Chapter Nine

AQ6

10090-09_CH09.qxd 8/31/06 4:52 PM Page 244

Heart Failure 245

3. Management of heart failure symptoms:a. Treatment of pulmonary and systemic

vascular congestion. This is most readilyaccomplished by dietary sodium re-striction and diuretic medications.

b. Measures to increase forward cardiacoutput and perfusion of vital organsthrough the use of vasodilators andpositive inotropic drugs.

4. Modulation of the neurohormonal responseto help prevent adverse ventricular re-modeling in order to slow progression ofLV dysfunction.

5. Improvement of long-term survival. There is strong evidence that longevity is en-hanced by specific interventions, as de-scribed later in the chapter.

Diuretics

The mechanisms of action of diuretic drugsare summarized in Chapter 17. By promot-ing the elimination of sodium and waterthrough the kidney, diuretics reduce in-travascular volume and thus venous returnto the heart. As a result, the preload of the leftventricle is decreased, and its diastolic pres-sure falls out of the range that promotes pul-monary congestion (Fig. 9.10, point b). Thejudicious use of diuretics does not signifi-cantly reduce cardiac output in this setting,because the heart is operating on the “flat”portion of a depressed Frank-Starling curve.The intent is to reduce the end-diastolic pres-sure (and therefore hydrostatic forces con-tributing to pulmonary congestion) without


Hyp

oten

sion

Figure 9.10. Examples of the effect of heart failure treatment on theleft ventricular (LV) Frank-Starling curve. Point a represents the failingheart on a curve that is shifted downward compared with normal. The strokevolume is reduced (bordering on hypotension), and the LV end-diastolic pres-sure (LVEDP) is increased, resulting in symptoms of pulmonary congestion.Therapy with a diuretic or pure venous vasodilator (point b on the sameFrank-Starling curve) reduces LV pressure without much change in stroke vol-ume (SV). However, excessive diuresis or venous vasodilatation may result inan undesired fall in SV with hypotension (point b′). Inotropic drug therapy(point c) and arteriolar (or “balanced”) vasodilator therapy (point d) augmentSV, and because of improved LV emptying during contraction, the LVEDPlessens. Point e represents the potential added benefit of combining an in-otrope and vasodilator together. The dashed line shows one example of howthe Frank-Starling curve shifts upward during inotropic/vasodilator therapybut does not achieve the level of a normal ventricle.

Fig. 10

10090-09_CH09.qxd 8/31/06 4:52 PM Page 245

a significant fall in stroke volume. However,overly vigorous diuresis can lower the LVfilling pressures into the steep portion of theventricular performance curve, resulting inan undesired fall in cardiac output (see Fig.9.10, point b′). Thus, diuretics should be usedonly if there is evidence of pulmonary con-gestion (rales) or peripheral interstitial fluidaccumulation (edema).

Agents that act primarily at the renal loopof Henle (e.g., furosemide, torsemide, andbumetanide) are the most potent diuretics inheart failure. Thiazide diuretics (e.g., hydro-chlorothiazide and metolazone) are also useful but are less effective in the setting of decreased renal perfusion, which is oftenpresent in this condition.

The potential adverse effects of diureticsare described in Chapter 17. The most im-portant in heart failure patients includeoverly vigorous diuresis resulting in a fall incardiac output, and electrolyte disturbances(particularly hypokalemia and hypomagne-semia), which may contribute to dangerousarrhythmias. When diuretics are used in pa-tients with pure LV diastolic dysfunction torelieve congestive symptoms, extra caremust be taken to avoid overdiuresis becausethese patients require elevated diastolic fill-ing pressures to adequately fill their stiff-ened left ventricles (see Fig. 9.7B). It is there-fore often necessary to accept some degreeof chronically elevated filling pressures inpatients with LV diastolic dysfunction.

Vasodilators

One of the most important cardiac advancesin the late twentieth century was the intro-duction of vasodilator therapy for the treat-ment of heart failure, particularly the classof agents known as angiotensin-convertingenzyme (ACE) inhibitors. As indicated ear-lier, neurohormonal compensatory mecha-nisms in heart failure often lead to excessivevasoconstriction, volume retention, and ven-tricular remodeling. Vasodilator drugs helpto reverse these adverse consequences. More-over, multiple studies have shown that cer-tain vasodilator regimens significantly ex-tend survival in patients with heart failure.

The pharmacology of these drugs is describedin Chapter 17.

Venous vasodilators (e.g., nitrates) increasevenous capacitance, decrease venous returnto the heart, and therefore reduce left ven-tricular preload. Consequently, LV diastolicpressures fall and the pulmonary capillaryhydrostatic pressure declines, similar to thehemodynamic effects of diuretic therapy. Asa result, pulmonary congestion improves,and as long as the heart failure patient is onthe relatively “flat” part of the depressedFrank-Starling curve (see Fig. 9.10), the car-diac output does not fall despite the reduc-tion in ventricular filling pressure. However,venous vasodilatation in a patient who isoperating on the steeper part of the curvemay result in an undesired fall in stroke vol-ume, cardiac output, and blood pressure.

Pure arteriolar vasodilators (e.g., hydrala-zine) reduce systemic vascular resistance andtherefore LV afterload, which in turn permitsincreased ventricular muscle fiber shorteningduring systole (see Fig. 9.5B). This results inan augmented stroke volume and is repre-sented on the Frank-Starling diagram as ashift in an upward direction (see Fig. 9.10).Although an arterial vasodilator might beexpected to reduce blood pressure—an un-desired effect in patients with heart failurewho may already be hypotensive—this gen-erally does not happen. As resistance is re-duced by arteriolar vasodilatation, a concur-rent rise in cardiac output usually occurs,such that blood pressure remains constantor decreases only mildly.

Some groups of drugs result in vasodi-latation of both the venous and arteriolarcircuits (“balanced” vasodilators). Of these,the most important are the ACE inhibitors,which inhibit the formation of AII, a vaso-constrictor whose production is stimulatedin heart failure patients. In addition, be-cause aldosterone levels fall in response toACE inhibitor therapy, sodium eliminationis facilitated, resulting in reduced intravas-cular volume and improvement of systemicand pulmonary vascular congestion. ACEinhibitors also augment circulating levels of bradykinin (see Chapter 17), which isthought to play an important vasodilatory

246 Chapter Nine

10090-09_CH09.qxd 8/31/06 4:52 PM Page 246

Heart Failure 247

role in heart failure. As a result of these ef-fects, ACE inhibitors limit maladaptive ven-tricular remodeling in patients with heartfailure and following acute myocardial in-farction (see Chapter 7).

Supporting the beneficial hemodynamicand neurohormonal blocking effects of ACEinhibitors, many large clinical trials haveshown that the drugs reduce heart failuresymptoms, improve stamina, reduce the needfor hospitalization, and most importantly,extend survival in patients with chronicheart failure. Thus, ACE inhibitors are stan-dard first-line chronic therapy for patientswith LV systolic dysfunction.

The renin-angiotensin-aldosterone sys-tem can also be therapeutically inhibited byangiotensin II receptor blockers (ARBs),as described in Chapters 13 and 17. Sinceangiotensin II can be formed by pathwaysother than ACE, ARBs provide a morecomplete inhibition of the system throughblockade of the actual AII receptor (see Fig.17.6). Conversely, ARBs do not stimulatethe potentially beneficial rise in serumbradykinin. The net result is that the hemo-dynamic effects of ARBs in heart failure aresimilar to those of ACE inhibitors, and stud-ies thus far have not shown any superiorityof these agents over ACE inhibitors in termsof patient survival. Thus, they are prescribedto heart failure patients mainly when ACEinhibitors are not tolerated (e.g., because ofthe side effect of cough).

Chronic therapy using the combination ofthe venous dilator isosorbide dinitrate plusthe arteriolar dilator hydralazine has alsobeen shown to improve survival in patientswith moderate symptoms of heart failure.However, when administration of the ACEinhibitor enalapril was compared with thehydralazine–isosorbide dinitrate (H-ISDN)combination, the ACE inhibitor was shownto produce the greater improvement in sur-vival. Thus, H-ISDN is generally substitutedwhen a patient cannot tolerate ACE inhibitoror ARB therapy (e.g., because of renal insuffi-ciency or hyperkalemia). However, H-ISDNappears to have particular benefit in AfricanAmericans with heart failure. The recentAfrican American Heart Failure trial showed

that the addition of H-ISDN to standard heartfailure therapy (e.g., a diuretic, β-blocker,ACE inhibitor, or ARB) in black patients im-proved functional status and survival.

Nesiritide (human recombinant B-typenatriuretic peptide) is an intravenous va-sodilator drug available for hospitalized pa-tients with decompensated heart failure. It causes rapid and potent vasodilatation, reduces elevated intracardiac pressures, aug-ments forward cardiac output, and lessensthe activation of the renin-angiotensin-aldosterone and sympathetic nervous sys-tems. It promotes diuresis, reduces heart fail-ure symptoms, and can be combined withdiuretics and positive inotropic drugs. How-ever, it is an expensive drug, and recent evi-dence has raised questions about its safety.One analysis shows that patients treatedwith nesiritide are more likely to die overthe following month than are those receiv-ing traditional heart failure therapies. Cur-rently, therefore, nesiritide is used primarilyin patients who have not responded to orcannot tolerate other intravenous vasodila-tors, such as intravenous nitroglycerin ornitroprusside (see Chapter 17).

Inotropic Drugs

The inotropic drugs include β-adrenergicagonists, digitalis glycosides, and phospho-diesterase inhibitors (see Chapter 17). By in-creasing the availability of intracellular cal-cium, each of these drug groups enhancesthe force of ventricular contraction andtherefore shifts the Frank-Starling curve inan upward direction (see Fig. 9.10). As a re-sult, stroke volume and cardiac output areaugmented at any given ventricular end-diastolic volume. Therefore, these agentsmay be useful in treating patients with sys-tolic ventricular dysfunction but not thosewith pure diastolic failure.

The b-adrenergic agonists (e.g., dobuta-mine and dopamine) are administered in-travenously for temporary hemodynamicsupport in acutely ill, hospitalized patients.Their long-term use is limited by the lack ofan oral form of administration and by therapid development of drug tolerance. The

AQ7

10090-09_CH09.qxd 8/31/06 4:52 PM Page 247

latter refers to the progressive decline in ef-fectiveness during continued administrationof the drug, possibly owing to downregula-tion of myocardial adrenergic receptors. Like-wise, the role of phosphodiesterase inhibi-tors (e.g. amrinone and milrinone) is limitedto the intravenous treatment of congestiveheart failure in acutely ill patients. Despitethe initial promise of effective oral phospho-diesterase inhibitors, studies thus far demon-strate reduced survival among patients re-ceiving this treatment.

One of the oldest forms of inotropic ther-apy is digitalis (see Chapter 17), which canbe administered intravenously or orally.Digitalis preparations enhance contractility,reduce cardiac enlargement, improve symp-toms, and augment cardiac output in pa-tients with systolic heart failure. Digitalisalso increases the sensitivity of the baro-receptors, so that the compensatory sym-pathetic drive in heart failure is blunted, adesired effect that reduces left ventricular af-terload. By slowing AV nodal conductionand thereby reducing the rate of ventricularcontractions, digitalis has an added benefitin patients with congestive heart failurewho have concurrent atrial fibrillation. Al-though digitalis can improve symptomatol-ogy in heart failure patients, it has not beenshown to improve long-term survival. Itsuse is thus limited to patients who remainsymptomatic despite other standard thera-pies or to help slow the ventricular rate ifatrial fibrillation is also present. Digitalis isnot useful in the treatment of LV diastolicdysfunction because it does not improveventricular relaxation properties.

b-Blockers

Historically, β-blockers have been contra-indicated in patients with systolic dysfunc-tion because the negative inotropic effect ofthe drugs would be expected to worsen symp-tomatology. Paradoxically, recent studieshave actually shown that β-blockers have im-portant benefits in heart failure, includingaugmented cardiac output, reduced hemo-dynamic deterioration, and improved sur-vival. The explanation for this observation

remains unclear but may relate to the drugs’effect on reducing heart rate and bluntingchronic sympathetic activation or to theiranti-ischemic properties.

In clinical trials of patients with all classesof symptomatic heart failure, β-blockers havebeen well tolerated in stable patients (i.e.,those without recent deterioration of symp-toms or active signs of volume overload)and have resulted in improved mortalityrates and fewer hospitalizations comparedwith placebo. Not all β-blockers have beentested in heart failure. Those that have andhave been found to be beneficial includecarvedilol (a nonselective β1- and β2-receptorblocker with weak β-blocking properties)and the β1-selective metoprolol (in a sus-tained-release formulation). Despite thesebenefits, β-blockers must be used cautiouslyin heart failure to prevent acute deterio-ration owing to their potentially negativeinotropic effect. Regimens should be startedat low dosages and augmented gradually.

Aldosterone Antagonist Therapy

There is evidence that chronic excess of al-dosterone levels in heart failure may con-tribute to cardiac fibrosis and adverse ventric-ular remodeling. Antagonists of this hormone(which have been used for decades as mild di-uretics) have shown clinical benefit in heartfailure patients. For example, in a clinical trialof patients with advanced heart failure whowere already taking an ACE inhibitor and di-uretics, the antagonist spironolactone sub-stantially reduced mortality rates and im-proved heart failure symptoms. Eplerenone,a newer aldosterone antagonist, has beenshown to improve survival of patients withcongestive heart failure after an acute myo-cardial infarction. Although aldosterone an-tagonists are well tolerated in carefully con-trolled studies, heart failure patients’ serumpotassium levels need to be closely monitoredto prevent hyperkalemia, especially if there isrenal impairment or concomitant ACE inhi-bitor therapy.

In summary, standard therapy of chroniccongestive heart failure associated with LV

248 Chapter Nine

10090-09_CH09.qxd 8/31/06 4:52 PM Page 248

Heart Failure 249

systolic dysfunction should include severaldrugs, the cornerstones of which are an ACEinhibitor and a β-blocker. An accepted se-quence of therapy is to start with an ACE in-hibitor, as well as a diuretic if pulmonary orsystemic congestive symptoms are present. Ifthe patient is unable to tolerate the ACE in-hibitor, then an ARB (or hydralazine plusisosorbide dinitrate) may be substituted. Forpatients without recent clinical deteriorationor volume overload, a β-blocker should beadded. Those with advanced (NYHA class IV)heart failure may benefit from the additionof an aldosterone antagonist. For persistentsymptoms, digoxin can also be prescribed.

Additional Therapies

Other therapies commonly administered topatients with systolic dysfunction include(1) anticoagulation to prevent intracardiacthrombus formation if LV systolic functionis severely impaired (a controversial therapybecause clear benefit has not yet been de-monstrated by clinical trials) and (2) treat-ment of atrial and ventricular arrhythmiasthat frequently accompany chronic heartfailure. For example, atrial fibrillation is verycommon in heart failure patients, and con-version back to sinus rhythm can substan-tially improve cardiac output. Ventriculararrhythmias are also frequent in this popu-lation and may lead to sudden death. Theantiarrhythmic drug that is most effectiveand least likely to provoke dangerous arrhyth-mias in heart failure patients is amiodarone.However, studies of amiodarone for treat-ment of asymptomatic ventricular arrhyth-mias in heart failure have not shown a consistent survival benefit. In addition, heartfailure patients with symptomatic or sus-tained ventricular arrhythmias, or those with inducible ventricular tachycardia during electrophysiologic testing, benefit more from insertion of an implantable cardioverter-defibrillator (ICD; see Chapter11). Based on the results of recent large-scalerandomized trials, ICD therapy is recom-mended for many patients with chronic is-chemic or nonischemic dilated cardiomy-opathies and at least moderately reduced

systolic function (e.g., left ventricular ejec-tion fraction ≤35%), regardless of the pres-ence of ventricular arrhythmias, because theapproach reduces the likelihood of suddencardiac death in this population.

Cardiac Resynchronization Therapy

Intraventricular conduction abnormalitieswith widened QRS complexes (especially leftbundle branch block) are common in pa-tients with advanced heart failure. Such ab-normalities can actually contribute to cardiacsymptoms because of the uncoordinated pat-terns of right and left ventricular contraction.Advanced biventricular pacemakers havetherefore been developed that stimulate bothventricles simultaneously, thus resynchro-nizing the contractile effort. This techniqueof biventricular pacing, also termed cardiacresynchronization therapy (CRT), has beenshown to augment left ventricular systolicfunction with an accompanying reductionof LV size, improve exercise capacity, reducethe frequency of heart failure exacerbations,and reduce mortality related to progressiveheart failure. A wealth of evidence now in-dicates that CRT is appropriate for selectedpatients with advanced systolic dysfunction(LV ejection fraction ≤35%) with a prolongedQRS duration (>120 msec) and continuedsymptoms of heart failure despite appropri-ate pharmacologic therapies.

Because patients who receive CRT aretypically also candidates for an ICD, mod-ern devices combine both functions in a sin-gle, small implantable unit.

Cardiac Replacement Therapy

A patient with severe LV dysfunction whosecondition remains refractory to maximalmedical management may be a candidate forcardiac transplantation. Because of a short-age of donor hearts, only approximately2,500 transplants are performed in theUnited States each year. Thus, alternativemechanical heart support therapies are un-dergoing intense development, includingventricular assist devices and totally im-planted artificial hearts.

10090-09_CH09.qxd 8/31/06 4:52 PM Page 249

Treatment of Diastolic Dysfunction

Correctable causes of impaired diastolic func-tion should be considered and addressed. Forexample, pericardiectomy would be under-taken for constrictive pericarditis (see Chap-ter 14), or therapy should be directed atcoronary artery disease if transient ischemiais the mechanism of diastolic dysfunction.Inotropic drugs or vasodilators usually haveno role in the treatment of pure diastolicdysfunction. Diuretics may reduce pulmo-nary congestion and peripheral edema butmust be used cautiously to avoid compro-mising cardiac output and causing hypoten-sion, because the stiffened left ventricle relieson higher-than-normal filling pressures tomaintain its output. The roles of other po-tentially promising medications in diastolicdysfunction, including β-blockers, calciumchannel blockers, ACE inhibitors, and ARBscurrently remain undefined.

ACUTE PULMONARY EDEMA

A severe, acute form of left-sided heart failureis cardiogenic pulmonary edema, in whichelevated capillary hydrostatic pressure causesrapid accumulation of fluid within the inter-stitium and alveolar spaces of the lung. In thepresence of normal plasma oncotic pressure,pulmonary edema develops when the pul-monary capillary wedge pressure, which re-flects LV diastolic pressure, exceeds approxi-mately 25 mm Hg.

This condition is frequently accompaniedby hypoxemia because of shunting of pul-monary blood flow through regions of hypo-ventilated alveoli. Pulmonary edema may ap-pear suddenly in a previously asymptomaticperson in, for example, the setting of anacute myocardial infarction or in patientswith chronic compensated congestive heartfailure following a precipitating event (seeTable 9.3). Pulmonary edema is a horrifyingexperience for the patient, resulting in severedyspnea and anxiety while struggling tobreathe.

On examination, the patient is tachycardicand demonstrates cold, clammy skin owing

to peripheral vasoconstriction in response tothe increased sympathetic outflow. Tachyp-nea and coughing of “frothy” sputum repre-sent transudation of fluid into the alveoli.Rales are present initially at the bases andthen throughout the lung fields, sometimesaccompanied by wheezing because of edemawithin the conductance airways.

Pulmonary edema is a life-threateningemergency that requires immediate im-provement of systemic oxygenation andelimination of the underlying cause. The pa-tient should be seated upright to permitpooling of blood within the systemic veinsof the lower body, thereby reducing venousreturn to the heart. Supplemental oxygen isprovided by a face mask. Morphine sul-phate is administered intravenously to re-duce anxiety and as a venous dilator to fa-cilitate pooling of blood peripherally. Arapidly acting diuretic, such as intravenousfurosemide, is administered to further re-duce LV preload and pulmonary capillaryhydrostatic pressure. Other means of reduc-ing preload include administration of ni-trates (often intravenously) or, in extremecases, venous phlebotomy. Inotropic drugs(e.g., dopamine) can also be administeredintravenously to increase forward cardiacoutput. During resolution of the pulmonarycongestion and hypoxemia, attention shouldbe directed at identifying and treating theunderlying cause.

An easy-to-remember mnemonic for acutemanagement of pulmonary edema is the al-phabetic sequence LMNOP:

Lasix (trade name for furosemide)MorphineNitratesOxygenPosition (sit upright)

SUMMARY

1. Heart failure is present when cardiac out-put fails to meet the metabolic demandsof the body or meets those demands onlyif the cardiac filling pressures are abnor-mally high. It most often results from im-paired left ventricular systolic function,

250 Chapter Nine

10090-09_CH09.qxd 8/31/06 4:52 PM Page 250

Heart Failure 251

but may also arise from ventricular dias-tolic dysfunction and other cardiac ab-normalities that interfere with ventricularfilling or emptying, such as pericardial orvalvular disease.

2. Compensatory mechanisms in heart fail-ure that help maintain circulatory func-tion include (a) preload augmentationwith increased stroke volume via theFrank-Starling mechanism, (b) activationof neurohormonal systems, and (c) ven-tricular hypertrophy. However, thesecompensations eventually become mal-adaptive, contributing to adverse ven-tricular remodeling and progressive dete-rioration of ventricular function.

3. Symptoms of heart failure may be exacer-bated by precipitating factors that in-crease metabolic demand, increase circu-lating volume, raise afterload, or decreasecontractility (summarized in Table 9.3).

4. Successful treatment of heart failure re-quires identification of the underlyingcause of the condition, elimination ofprecipitating factors, and modulation ofneurohormonal activations. Standardmedical treatment includes an ACE in-hibitor, β-blocker and, as needed, di-uretics and inotropic drugs. For patientswho do not tolerate an ACE inhibitor, anARB or the combination of hydralazineplus nitrates can be substituted. The ad-dition of spironolactone should be con-sidered for patients with advanced heartfailure.

5. In some patients, cardiac resynchroniza-tion therapy and/or insertion of an im-plantable cardioverter-defibrillator shouldbe considered.

Acknowledgment

Contributors to the previous editions of this chapterwere Arthur Coday Jr, MD; George S. M. Dyer, MD;Stephen K. Frankel, MD; Vikram Janakiraman, MD;and Michael A. Fifer, MD.

Additional Reading

Aurigemma G, Gaasch W. Diastolic heart failure. NEngl J Med 2004;351:1097–1105.

Bardy G, Lee K, Mark D, et al., for the SCD-HeFT investigators. Amiodarone or an implantable cardioverter-defibrillator for congestive heart fail-ure. N Engl J Med 2005;352:225–237.

Cleland J, Daubert J-C, Erdmann E, et al., for CARE-HF investigators. The effect of cardiac resynchro-nization on morbidity and mortality in heart fail-ure. N Engl J Med 2005;352:1539–1549.

Dec GW, editor. Heart Failure: A ComprehensiveGuide to Diagnosis and Treatment. New York:Marcel Dekker, 2005.

De Lemos J, McGuire, D, Drazner M. B-type natri-uretic peptide in cardiovascular disease. Lancet2003;362:316–322.

Foody JM, Farrell MH, Krumholz HM. β-Blocker ther-apy in heart failure: scientific review. JAMA 2002;287:883–889.

Goodfriend TL. Aldosterone—a hormone of cardio-vascular adaption and maladaption. J Clin Hyper-tens (Greenwich) 2006;8:133–139.

Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA2005 guideline update for the diagnosis and man-agement of chronic heart failure in the adult. Cir-culation 2005;112:1825–1852.

Jessup M, Brozena S. Heart failure. N Engl J Med2003;348:2007–2018.

Jong P, Demers C, McKelvie RS, et al. Angiotensin re-ceptor blockers in heart failure: meta-analysis ofrandomized controlled trials. J Am Coll Cardiol2002;39:463–470.

Katz AM. Heart Failure: Pathophysiology, MolecularBiology, Clinical Management. Philadelphia: Lip-pincott Williams & Wilkins, 2000.

Maisel A. The coming age of natriuretic peptides. J Am Coll Cardiol 2006;47:61–64.

Morita H, Seidman J, Seidman CE. Genetic causes ofhuman heart failure. J Clin Invest 2005;115:518–526.

Pitt B, Rajagopalan S. Aldosterone receptor antago-nists for heart failure: current status, future indi-cations. Cleve Clin J Med 2006;73:257–268.

Sackner-Bernstein J, Kowalski M, Fox M, Aaronson K.Short-term risk of death after treatment with ne-siritide for decompensated heart failure: a pooledanalysis of randomized controlled trials. JAMA2005;293:1900–1905.

Stevenson LW, Shekar P. Ventricular assist devices fordurable support. Circulation 2005;112:111–115.

Taylor A, Ziesche S, Yancy C, et al. Combination ofisosorbide dinitrate and hydralazine in blackswith heart failure. N Engl J Med 2004;351:2049–2057.

Walsh RA, ed. Molecular Mechanisms of Cardiac Hypertrophy and Failure. New York: Taylor &Francis, 2005.

Ware LB, Matthay MA. Acute pulmonary edema. N Engl J Med 2005;353:2788–2796.

10090-09_CH09.qxd 8/31/06 4:52 PM Page 251

Chapter 9—Author Queries1. AU: Edited for consistency with chapter 62. AU: Edit reflects wording in the subsection that comes later. OK?3. AU: As in heading below?4. AU: As in ch. 6.5. AU: Similar headings in all chapters edited for a certain level of consistency.6. AU: Correct?7. AU: OK?8. AU: OK?

10090-09_CH09.qxd 8/31/06 4:52 PM Page a

252

DILATED CARDIOMYOPATHYEtiologyPathologyPathophysiologyClinical FindingsPhysical ExaminationDiagnostic StudiesTreatmentPrognosis

HYPERTROPHIC CARDIOMYOPATHYEtiologyPathology

PathophysiologyClinical FindingsPhysical ExaminationDiagnostic StudiesTreatmentPrognosis

RESTRICTIVE CARDIOMYOPATHYPathophysiologyClinical FindingsPhysical ExaminationDiagnostic StudiesTreatment

C H A P T E R

10The CardiomyopathiesMarc N. WeinG. William DecLeonard S. Lilly

The cardiomyopathies are a group of heartdisorders in which the major structural ab-normality is limited to the myocardium.These conditions often result in symptoms ofheart failure, and although the underlyingcause of myocardial dysfunction can some-times be identified, the etiology frequentlyremains unknown. Excluded from the defin-ition of this group of diseases is heart muscleimpairment resulting from other known car-diac conditions, such as hypertension, valvu-lar disorders, or coronary artery disease.

Cardiomyopathies can be classified intothree types based on the anatomic appear-

ance and abnormal physiology of the leftventricle (Fig. 10.1). Dilated cardiomy-opathy is characterized by ventricular cham-ber enlargement with impaired systoliccontractile function; hypertrophic cardio-myopathy, by an abnormally thickenedventricular wall with abnormal diastolic re-laxation but usually intact systolic func-tion; and restrictive cardiomyopathy, byan abnormally stiffened myocardium (be-cause of fibrosis or an infiltrative process)leading to impaired diastolic relaxation,but systolic contractile function is normalor near normal.

Fig. 1

10090-10_CH10.qxd 8/31/06 4:53 PM Page 252

The Cardiomyopathies 253

DILATED CARDIOMYOPATHY

Etiology

Cardiac enlargement in dilated cardiomy-opathy (DCM) is caused by ventricular di-latation with only minor hypertrophy. Myo-cyte damage leading to this condition resultsfrom a wide spectrum of genetic, inflamma-tory, toxic, and metabolic causes (Table 10.1).Although most cases are idiopathic (i.e., thecause is undetermined), examples of condi-tions that are commonly recognized causesof DCM include viral myocarditis, alcoholtoxicity, and specific gene mutations.

Acute viral myocarditis generally afflictsyoung, previously healthy people. Commonresponsible infecting organisms includecoxsackievirus group B and adenovirus. It isusually a self-limited illness with full recov-ery, but for unclear reasons, some patientsprogress to DCM. It is hypothesized thatmyocardial destruction and fibrosis resultfrom immune-mediated injury triggered byviral constituents. Nonetheless, immuno-suppressive drugs have not been shown toimprove the prognosis of this condition.Transvenous right ventricular biopsy duringacute myocarditis may demonstrate active

Tab. 1

Figure 10.1. Anatomic appearance of the cardiomyopathies (CMPs). A. Normalheart demonstrating left ventricle (LV) and left atrium (LA). B. Dilated CMP is charac-terized by ventricular dilatation with only mild hypertrophy. C. Hypertrophic CMPdemonstrates significant ventricular hypertrophy, often predominantly involving the intraventricular septum. D. Restrictive CMP is caused by infiltration or fibrosis of the ven-tricles, usually without enlargement of the cavities. LA enlargement is common to allthree types of CMP.

10090-10_CH10.qxd 8/31/06 4:53 PM Page 253

254 Chapter Ten

TABLE 10.1. Examples of DilatedCardiomyopathies

IdiopathicFamilial (genetic)InflammatoryInfectious (especially viral)Noninfectious

Connective tissue diseasesPeripartum cardiomyopathySarcoidosis

ToxicChronic alcohol ingestionChemotherapeutic agents (e.g., doxorubicin)MetabolicHypothyroidismChronic hypocalcemia or hypophosphatemiaNeuromuscularMuscular or myotonic dystrophy

TABLE 10.2. Examples of Disease-Related Genes in Familial Forms of Dilated and Hypertrophic Cardiomyopathies

Gene Protein Function

Mutations observed in familial dilated cardiomyopathyDES Desmin Transduction of contractile forcesSGCD ∂-Sarcoglycan Transduction of contractile forcesDMD Dystrophin Transduction of contractile forcesMYH7 β-Myosin heavy chain Muscle contractionACTC Cardiac actin Muscle contractionTNNT2 Cardiac troponin T Muscle contractionTTN Titin Scaffold for sarcomereLMNA Lamin A/C Nuclear membrane proteinABCC Sur2A Inwardly rectifying potassium channelMutations observed in familial hypertrophic cardiomyopathy:MYH7 β-Myosin heavy chain Muscle contractionMYBP3 Myosin-binding protein C Muscle contractionTNNT2 Cardiac troponin T Muscle contractionTNNI3 Cardiac troponin I Muscle contractionTPM1 α-Tropomyosin Muscle contractionMYL3 Essential myosin light chain Muscle contraction

inflammation, and viral genomic DNA orRNA sequences have been demonstrated insome infected persons.

Alcoholic cardiomyopathy develops in afew people who consume alcoholic bever-ages chronically. Although the pathophysi-ology of the condition is unknown, ethanolis thought to impair cellular function by in-hibiting mitochondrial oxidative phospho-rylation and fatty acid oxidation. Its clini-cal presentation and histologic features are

similar to those of other dilated cardiomyo-pathies. Alcoholic cardiomyopathy is im-portant to identify because its cause is re-versible; cessation of ethanol consumptioncan lead to dramatic improvement of ven-tricular function. Other potentially reversiblecauses of DCM include other toxin expo-sures, metabolic abnormalities (such as hy-pothyroidism), and some inflammatory eti-ologies, such as sarcoidosis or connectivetissue diseases.

Several familial forms of DCM have beenidentified and are believed to be responsi-ble for 20% to 30% of what were once clas-sified as idiopathic DCM. Autosomal dom-inant, autosomal recessive, X-linked, andmitochondrial patterns of inheritance havebeen described, leading to defects in con-tractile force generation, force transmission,energy production, and myocyte viability.Mutations identified thus far occur in genesthat code predominantly for cardiac cyto-skeletal proteins, including troponin T, myo-sin, actin, and dystrophin (Table 10.2). Incertain families, associated phenotypicalfeatures have included auditory deficits, car-diac conduction system defects, and skeletalmuscle abnormalities. Recognition of affectedpersons and identification of their underly-ing genetic mutations can allow gene-based

Tab. 2

10090-10_CH10.qxd 8/31/06 4:53 PM Page 254

screening and diagnosis for family membersand potentially earlier interventions to pre-vent the development of symptoms andcomplications.

Pathology

Marked enlargement of all four cardiac cham-bers is typical of DCM (Fig. 10.2), althoughsometimes the disease is limited to the leftor right side of the heart. The thickness ofthe ventricular walls may be increased, butchamber dilatation is out of proportion toany hypertrophy. Microscopically, there isevidence of myocyte degeneration with ir-regular hypertrophy and atrophy of myo-fibers. Interstitial and perivascular fibrosis isoften extensive.

Pathophysiology

The hallmark of DCM is ventricular dilata-tion with decreased contractile function (Fig.10.3). Most often in DCM, both ventriclesare impaired, but sometimes dysfunction islimited to the left ventricle (LV) and evenless commonly to the right ventricle (RV).

As ventricular stroke volume and cardiacoutput decline because of impaired myocytecontractility, two compensatory effects areactivated: (1) the Frank-Starling mechanism,in which the elevated ventricular diastolic

volume increases the stretch of the myofibers,thereby increasing the subsequent strokevolume; and (2) neurohormonal activation,initially mediated by the sympathetic ner-vous system (see Chapter 9). The latter con-tributes to an increased heart rate and con-tractility, which help to buffer the fall incardiac output. These compensations mayrender the patient asymptomatic during theearly stages of ventricular dysfunction; how-ever, as progressive myocyte degenerationand volume overload ensue, clinical symp-toms of heart failure develop.

With a persistent reduction of cardiac output, the decline in renal blood flowprompts the kidneys to secrete increasedamounts of renin. This activation of therenin-angiotensin-aldosterone axis increasesperipheral vascular resistance (mediatedthrough angiotensin II) and intravascularvolume (because of increased aldosterone).As described in Chapter 9, these effects arealso initially helpful in buffering the fall incardiac output.

Ultimately, however, the “compensatory”effects of neurohormonal activation provedetrimental. Arteriolar vasoconstriction andincreased systemic resistance render it moredifficult for the LV to eject blood in the for-ward direction, and the rise in intravascularvolume further burdens the ventricles, re-sulting in pulmonary and systemic conges-tion. In addition, chronically elevated levelsof angiotensin II and aldosterone directlycontribute to pathological myocardial re-modeling and fibrosis.

As the cardiomyopathic process causesthe ventricles to enlarge over time, the mi-tral and tricuspid valves may fail to coaptproperly in systole, and valvular regurgi-tation ensues. This regurgitation has threedetrimental consequences: (1) excessive vol-ume and pressure loads are placed on theatria, causing them to dilate, often leadingto atrial fibrillation; (2) regurgitation of bloodinto the left atrium further decreases for-ward stroke volume into the aorta and sys-temic circulation; and (3) when the regurgi-tant volume returns to the LV during eachdiastole, an even greater volume load is pre-sented to the dilated LV.

Figure 10.2. Transverse sections of a normal heart(left) and a heart from a patient with dilated car-diomyopathy (DCM). In the DCM specimen, there isbiventricular dilatation without a proportional increase inwall thickness. LV, left ventricle; RV, right ventricle. (Mod-ified from Emmanouilides GC, ed. Moss and Adams’Heart Disease in Infants, Children, and Adolescents. 5th Ed.Baltimore: Lippincott Williams & Wilkins, 1995:86.)AQ3

AQ1


Fig. 2

Fig. 3

10090-10_CH10.qxd 8/31/06 4:53 PM Page 255

256 Chapter Ten

Clinical Findings

The clinical manifestations of DCM are thoseofcongestiveheart failure. The most commonsymptoms of low forward cardiac output in-clude fatigue, lightheadedness, and exer-tional dyspnea associated with decreased tis-sue perfusion. Pulmonary congestion resultsin dyspnea, orthopnea, and paroxysmal noc-turnal dyspnea, whereas chronic systemic ve-nous congestion causes ascites and peripheraledema. Because these symptoms may developinsidiously, the patient may complain only of recent weight gain (because of interstitialedema) and shortness of breath on exertion.


Signs of decreased cardiac output are oftenpresent and include cool extremities (owingto peripheral vasoconstriction), low arterialpressure, and tachycardia. Pulmonary venous

congestion results in auscultatory crackles(rales), and basilar chest dullness to percus-sion may be present because of pleural effu-sions. Cardiac examination shows an en-larged heart with leftward displacement of adiffuse apical impulse. On auscultation, athird heart sound (S3) is common as a sign ofpoor systolic function. The murmur of mitralvalve regurgitation is often present as a resultof the significant left ventricular dilatation. Ifright ventricular heart failure has developed,signs of systemic venous congestion may in-clude jugular vein distention, hepatomegaly,ascites, and peripheral edema. Right ventric-ular enlargement and contractile dysfunc-tion are often accompanied by the murmurof tricuspid valve regurgitation.

Diagnostic Studies

The chest radiograph shows an enlarged car-diac silhouette. If heart failure has developed,

Figure 10.3. Pathophysiology of dilated cardiomyopathy. The reducedventricular stroke volume results in decreased forward cardiac output and in-creased ventricular filling pressures. The listed clinical manifestations follow.JVD, jugular venous distention.

10090-10_CH10.qxd 8/31/06 4:53 PM Page 256

then pulmonary vascular redistribution, interstitial and alveolar edema, and pleuraleffusions are evident (see Fig. 3.5).

The electrocardiogram (ECG) usually de-monstrates atrial and ventricular enlarge-ment. Patchy fibrosis of the myofibers resultsin a wide array of arrhythmias, most impor-tantly atrial fibrillation and ventricular tachy-cardia. Conduction defects (left or right bun-dle branch block) occur in most cases. Diffuserepolarization (ST segment and T wave) ab-normalities are common. In addition, regionsof dense myocardial fibrosis may produce lo-calized Q waves, resembling the pattern ofprevious myocardial infarction.

Echocardiography is very useful in the diag-nosis of DCM. It typically demonstrates four-chamber cardiac enlargement with little hy-pertrophy and global reduction of systoliccontractile function. Mitral and/or tricuspidregurgitation is also frequently visualized.

Cardiac catheterization is often performedto determine whether coexistent coronaryartery disease is contributing to the impairedventricular function. This procedure is mostuseful diagnostically in patients who describeepisodes of angina pectoris or have evidencesuggestive of previous myocardial infarctionon the electrocardiogram. Typically, hemo-dynamic measurements show elevated right-and left-sided diastolic pressures and dimin-ished cardiac output. In the catheterizationlaboratory, a transvenous biopsy of the rightventricle is sometimes performed to clarifythe etiology of the cardiomyopathy. The roleof this procedure is limited, however, becauseit rarely is diagnostic in patients with DCMand infrequently alters therapeutic decisions.

Treatment

The goal of therapy in DCM is to relievesymptoms, prevent complications, and im-prove long-term survival. Thus, in additionto treating any identified underlying causeof DCM, therapeutic considerations includethose described in the following sections.

Medical Treatment of Heart Failure

Approaches for the relief of vascular conges-tion and improvement in forward cardiac

output are essentially the same as standardtherapies for heart failure (see Chapter 9). Ini-tial therapy typically includes salt restrictionand diuretics, vasodilator therapy with an angiotensin-converting enzyme (ACE) inhibitoror angiotensin II receptor blocker (ARB), and a β-blocker. In patients with advanced heartfailure, the potassium-sparing diuretic spiro-nolactone may be added later. These measureshave been shown to improve symptoms andreduce mortality in patients with dilated cardiomyopathy. The oral inotropic agentdigoxin may also be added to further improveleft ventricular function and reduce symp-toms, but it has not been shown to prolongsurvival.

Prevention and Treatment of Arrhythmias

Atrial and ventricular arrhythmias are com-mon in advanced DCM, and approximately40% of deaths in this condition are causedby ventricular tachycardia or fibrillation. It is important to maintain serum electrolytes(notably, potassium and magnesium) withintheir normal ranges, especially during di-uretic therapy, to avoid provoking serious ar-rhythmias. Studies have shown that avail-able antiarrhythmic drugs do not preventdeath from ventricular arrhythmias in DCM.In fact, when used in patients with poor LVfunction, many antiarrhythmic drugs mayworsen the rhythm disturbance. Amiodaroneis the contemporary antiarrhythmic studiedmost extensively in patients with DCM.Whereas there is no convincing evidencethat it reduces mortality from ventricular arrhythmias in DCM, it is the safest anti-arrhythmic for treating atrial fibrillation andother supraventricular arrhythmias in thispopulation. In contrast to antiarrhythmicdrugs, the placement of an implantablecardioverter-defibrillator (ICD) does reducearrhythmic deaths in patients with DCM.Therefore, based on large-scale randomizedtrials, ICD placement is a recommended ap-proach for patients with chronic sympto-matic dilated cardiomyopathy and at leastmoderately reduced systolic function (e.g.,LV ejection fraction ≤35%), regardless of thedetection of ventricular arrhythmias, be-

AQ2


10090-10_CH10.qxd 8/31/06 4:53 PM Page 257

258 Chapter Ten

cause such an approach reduces the likeli-hood of sudden cardiac death.

Many patients with DCM have electricalconduction abnormalities that contribute todyssynchronous ventricular contraction andreduced cardiac output. Electronic pacemak-ers capable of stimulating both ventricles si-multaneously have been devised to bettercoordinate systolic contraction as an adjunctto medical therapy (the technique is knownas cardiac resynchronization therapy). Demon-strated benefits of this approach include im-proved quality of life and exercise toleranceas well as decreased hospitalizations for heartfailure and reduced mortality, particularly inthose with pretreatment left bundle branchblock or other conduction abnormalities witha markedly prolonged QRS duration.

Prevention of Thromboembolic Events

Patients with DCM are at increased risk ofthromboembolic complications because of(1) stasis in the ventricles resulting from poorsystolic function, (2) stasis in the atria owingto chamber enlargement or atrial fibrillation,(3) an abnormally thrombogenic endocardialsurface, and (4) venous stasis caused by poorforward flow. Peripheral venous or right ven-tricular thrombus may lead to pulmonaryemboli, whereas thromboemboli of left ven-tricular origin may lodge in any systemicartery, resulting in, for example, devastatingcerebral, myocardial, or renal infarctions. Theonly definite indications for systemic antico-agulation in DCM patients are atrial fibrilla-tion, a previous thromboembolic event, oran LV thrombus visualized by echocardiog-raphy. Additionally, chronic oral anticoagu-lation therapy (i.e., warfarin) is often admin-istered to DCM patients who have severedepression of ventricular function (e.g., LVejection fraction <30%); however, prospec-tive studies are lacking to evaluate the effec-tiveness of this therapy in DCM patients whoare in sinus rhythm.

Cardiac Transplantation

In suitable patients, cardiac transplantationoffers a substantially better 5-year prognosisthan the standard therapies for DCM previ-ously described. The current 5- and 10-year

survival rates after transplantation are 74%and 55%, respectively. However, the scarcityof donor hearts greatly limits the availabilityof this technique. Fewer than 2,500 trans-plants are performed in the United States an-nually compared with approximately 20,000patients who could potentially benefit fromthe procedure. As a result, other mechanicaloptions have been explored and continue toundergo experimental refinements, includ-ing ventricular assist devices and completelyimplanted artificial hearts.

Prognosis

Despite advances in therapy, the prognosisfor patients with DCM who do not undergocardiac transplantation is poor—the average5-year survival rate is less than 50%. Methodsto reduce progressive LV dysfunction by earlyintervention in asymptomatic or minimallysymptomatic patients and the prevention ofsudden cardiac death remain major researchgoals in the treatment of this disorder.

HYPERTROPHICCARDIOMYOPATHY

Hypertrophic cardiomyopathy (HCM) hasreceived notoriety in the lay press because itis the most common cardiac abnormalityfound in young athletes who die suddenlyduring vigorous physical exertion. With anincidence of about 1 of 500 in the generalpopulation, HCM is characterized by septalor left ventricular hypertrophy that is notcaused by chronic pressure overload (i.e., notthe result of systemic hypertension or aorticstenosis). Other terms frequently used to de-scribe this disease are hypertrophic obstruc-tive cardiomyopathy and idiopathic hyper-trophic subaortic stenosis. In this condition,systolic LV contractile function is vigorousbut the thickened muscle is stiff, resulting inimpaired ventricular relaxation and highdiastolic pressures.

Etiology

HCM is a familial disease in which inheri-tance follows an autosomal dominant patternwith variable penetrance, and a large varietyof mutations in at least 10 different genes 1 LINE LONG

10090-10_CH10.qxd 8/31/06 4:53 PM Page 258

have been implicated. The proteins encodedby the responsible genes are all part of thesarcomere complex and include β-myosinheavy chain (β-MHC), cardiac troponin T,and myosin-binding protein C (see Table10.2). The incorporation of these mutatedpeptides into the sarcomere is thought tocause impaired contractile function. The re-sultant increase in myocyte stress is thenhypothesized to lead to compensatory hy-pertrophy and proliferation of fibroblasts.

The pathophysiology and natural historyof familial HCM are quite variable and ap-pear related to particular mutations withinthe disease-causing gene, rather than the ac-tual gene involved. In fact, it has beenshown that the precise genetic mutation de-termines the age of onset of hypertrophy,the extent and pattern of cardiac remodel-ing, and the person’s risk of developingsymptomatic heart failure or sudden death.For example, mutations in the β-MHC genethat alter electrical charge in the encodedprotein are associated with worse prognosesthan other mutations. It is hoped that betterdefinition of the natural history of specificmutations will allow accurate risk stratifica-tion of patients and permit appropriate tim-ing of therapeutic interventions.

Pathology

Although hypertrophy in HCM may involveany portion of the ventricles, asymmetric hy-pertrophy of the ventricular septum (Fig.10.4) is most common (approximately 90%of cases). Less often, the hypertrophy involvesthe ventricular walls symmetrically or is lo-calized to the apex or midregion of the LV.

Unlike ventricular hypertrophy resultingfrom hypertension, in which the myocytesenlarge uniformly and remain orderly, thehistology of HCM is unusual. The myocar-dial fibers are in a pattern of extensive disar-ray (Fig. 10.5). Short, wide, hypertrophiedfibers are oriented in chaotic directions andsurrounded by numerous cardiac fibroblastsand extracellular matrix. This myocyte dis-array and fibrosis are characteristic of HCMand play a role in the abnormal diastolicstiffness and the arrhythmias common tothis disorder.

Pathophysiology

The predominant feature of HCM is markedventricular hypertrophy that reduces thecompliance and relaxation (diastolic func-tion) of the chamber, such that filling be-comes impaired (Fig. 10.6). Patients whohave asymmetric hypertrophy of the prox-imal interventricular septum may displayadditional findings related to transient ob-struction of left ventricular outflow duringsystole. In this case, the mechanism of sys-tolic obstruction is thought to involve ab-normal motion of the anterior mitral valveleaflet toward the LV outflow tract wherethe thickened septum protrudes (Fig. 10.7).The process is explained as follows: (1) dur-ing ventricular contraction, ejection ofblood past the upper septum is more rapidthan usual, because it must flow throughan outflow tract that is narrowed by thethickened septum; (2) this rapid flow cre-ates Venturi forces that abnormally draw

IVS

Figure 10.4. Postmortem heart specimen from apatient with hypertrophic cardiomyopathy. Signifi-cant left ventricular hypertrophy is seen, especially of the interventricular septum (IVS).

Fig. 4

Fig. 5

Fig. 6

Fig. 7


1 LINE LONG

10090-10_CH10.qxd 8/31/06 4:53 PM Page 259

260 Chapter Ten

Figure 10.5. Light microscopy of the hypertrophic myocardium. A. Normal myocardium. B. Hypertrophic myocytesresulting from pressure overload in a patient with valvular heart disease. C. Disordered myocytes with fibrosis in a pa-tient with hypertrophic cardiomyopathy. (Modified from Schoen FJ. Interventional and Surgical Cardiovascular Pathol-ogy: Clinical Correlations and Basic Principles. Philadelphia: WB Saunders, 1989:181.)

Figure 10.6. Pathophysiology of hypertrophic cardiomyopathy. The disarrayed and hypertro-phied myocytes may lead to ventricular arrhythmias (which can cause syncope or sudden death) andimpaired diastolic left ventricular (LV) relaxation (which causes elevated LV filling pressures and dys-pnea). If dynamic left ventricular outflow obstruction is present, mitral regurgitation often accom-panies it (which contributes to dyspnea), and the impaired ability to raise cardiac output with exer-tion can lead to exertional syncope. The thickened LV wall and systolic outflow tract obstruction bothcontribute to increased myocardial oxygen consumption (MVO2) and can precipitate angina. CO,cardiac output; LVEDP, LV end-diastolic pressure; LVH, LV hypertrophy.

10090-10_CH10.qxd 8/31/06 4:53 PM Page 260

the anterior mitral leaflet toward the sep-tum during contraction; and (3) the ante-rior mitral leaflet approaches and tran-siently abuts the hypertrophied septum,causing brief obstruction of blood flow intothe aorta. It is useful to consider the patho-physiology of HCM based on whether tran-sient systolic outflow tract obstruction ispresent.

HCM Without Outflow Tract Obstruction

Although systolic contraction of the left ven-tricle is usually vigorous in HCM, the hyper-trophied walls result in increased stiffnessand impaired diastolic relaxation of thechamber. The reduced ventricular compli-ance alters the normal pressure-volume re-lationship, causing the passive diastolic fill-ing curve to shift upward (see Fig. 9.7B). Theassociated rise in diastolic LV pressure istransmitted backward, leading to elevated leftatrial, pulmonary venous, and pulmonarycapillary pressures. Dyspnea, especially dur-ing exertion, is thus a common symptom inthis disorder.

HCM With Outflow Obstruction

In patients with outflow obstruction, eleva-ted left atrial and pulmonary capillary wedgepressures result from both the decreasedventricular compliance and the outflow ob-struction during contraction. During systolicobstruction, a pressure gradient develops be-tween the main body of the LV and the out-flow tract distal to the obstruction (see Fig.10.7). The elevated ventricular systolic pres-sure increases wall stress and myocardialoxygen consumption, which can result inanginal chest discomfort (see Fig. 10.6). In ad-dition, because obstruction is caused by ab-normal motion of the anterior mitral leaflettoward the septum (and therefore away fromthe posterior mitral leaflet), the mitral valvedoes not close properly during systole, andmitral regurgitation may result. This furtherelevates left atrial and pulmonary venouspressures and may worsen symptoms of dys-pnea, as well as contribute to the develop-ment of atrial fibrillation.

The systolic pressure gradient observed in obstructive HCM is dynamic in that itsmagnitude varies during contraction and de-

MR

Figure 10.7. Pathophysiology of left ventricular (LV) outflow obstruction and mitral re-gurgitation in hypertrophic cardiomyopathy (HCM). Left panel. The LV outflow tract isabnormally narrowed between the hypertrophied interventricular septum and the anteriorleaflet of the mitral valve (AML). It is thought that the rapid ejection velocity along the narrowedtract in early systole draws the AML toward the septum (small arrow). Right panel. As the mi-tral valve abnormally moves anteriorly and contacts the septum, outflow into the aorta is tran-siently obstructed. Because the mitral leaflets do not coapt normally in systole, mitral regurgita-tion (MR) also results.


10090-10_CH10.qxd 8/31/06 4:53 PM Page 261

ertion, when the pressure gradient is madeworse by the increased force of contraction,thereby causing a transient fall in cardiac out-put. Orthostatic lightheadedness is also com-mon in patients with outflow tract obstruc-tion. This occurs because venous return to theheart is reduced on standing by the gravita-tional pooling of blood in the lower extremi-ties. The LV thus decreases in size and outflowtract obstruction intensifies, transiently re-ducing cardiac output and cerebral perfusion.

When arrhythmias occur, symptoms ofHCM may be exacerbated. For example, atrialfibrillation is not well tolerated because theloss of the normal atrial “kick” further im-pairs diastolic filling and can worsen symp-toms of pulmonary congestion. Of greatestconcern, the first clinical manifestation ofHCM may be ventricular fibrillation, resultingin sudden cardiac death, particularly in youngadults with HCM during strenuous physicalexertion. Risk factors for sudden deathamong patients with known HCM include ahistory of syncope, a family history of sud-den death, certain high-risk mutations, andextreme hypertrophy of the LV wall (>30 mmin thickness).


Patients with mild forms of HCM are oftenasymptomatic and have normal or near-normal physical exams. A common findingis the presence of a fourth heart sound (S4),generated by left atrial contraction into thestiffened LV (see Chapter 2). The forcefulatrial contraction may also result in a palpa-ble presystolic impulse over the cardiac apex(creating what is known as a double apicalimpulse).

Other findings are common in patientswith systolic outflow obstruction. The caro-tid pulse rises briskly in early systole butthen quickly declines as obstruction to car-diac outflow appears. The characteristic sys-tolic murmur of LV outflow obstruction isrough and crescendo-decrescendo in shape,heard best at the left lower sternal border(because of turbulent flow through the nar-rowed outflow tract). In addition, as thestethoscope is moved toward the apex, theholosystolic blowing murmur of mitral re-

262 Chapter Ten

pends, at any given time, on the distance be-tween the anterior leaflet of the mitral valveand the hypertrophied septum. Situationsthat decrease LV cavity size (e.g., reduced ve-nous return because of intravascular volumedepletion) bring the mitral leaflet and sep-tum into closer proximity and promote ob-struction. Conversely, conditions that enlargethe LV (e.g., augmented intravascular vol-ume) increase the distance between the ante-rior mitral leaflet and septum and reduce theobstruction. Positive inotropic drugs (whichaugment the force of contraction) also forcethe mitral leaflet and septum into closer pro-ximity and contribute to obstruction, where-as negative inotropic drugs (e.g., β-blockers,verapamil) have the opposite effect.

Although dynamic systolic outflow tractobstruction creates impressive murmurs andreceives great attention, the symptoms ofobstructive HCM appear to primarily stemfrom the increased LV stiffness and diastolicdysfunction also present in the nonobstruc-tive form.

Clinical Findings

The symptoms of HCM vary widely, fromnone to significant physical limitations. Theaverage age of presentation is the mid-20s.

The most frequent symptom is dyspneaowing to elevated diastolic LV (and thereforepulmonary capillary) pressures. This symp-tom is further exacerbated by the high systo-lic LV pressure and mitral regurgitation seenin patients with outflow tract obstruction.

Angina is often described by patients withHCM, even in the absence of obstructivecoronary artery disease. Myocardial ischemiamay be contributed to by (1) the high oxy-gen demand of the increased muscle massand (2) the narrowed small branches of thecoronary arteries within the hypertrophiedventricular wall. If outflow tract obstructionis present, the high systolic ventricular pres-sure also increases myocardial oxygen de-mand because of the increased wall stress.

Syncope in HCM may result from cardiacarrhythmias that develop because of thestructurally abnormal myofibers (see the nextsection). In patients with outflow tract ob-struction, syncope may also be induced by ex-

10090-10_CH10.qxd 8/31/06 4:53 PM Page 262

gurgitation may be auscultated. Althoughthe LV outflow obstruction murmur may besoft at rest, bedside maneuvers that alter pre-load and afterload can dramatically increaseits intensity and help differentiate this mur-mur from other conditions, such as aorticstenosis (Table 10.3).

A commonly used technique in this re-gard is the Valsalva maneuver, produced byasking the patient to “bear down” (techni-cally defined as forceful exhalation with thenose, mouth, and glottis closed). The Val-salva maneuver increases intrathoracic pres-sure, which decreases venous return to theheart and transiently reduces LV size. Thisaction brings the hypertrophied septum andanterior leaflet of the mitral valve into closerproximity, creating greater obstruction toforward flow. Thus, during Valsalva, themurmur of HCM increases in intensity. Incontrast, the murmur of aortic stenosis de-creases in intensity during Valsalva becauseof the reduced flow across the stenotic valve.

Conversely, a change from standing to asquatting position suddenly augments venousreturn to the heart (which increases preload)while simultaneously increasing the sys-temic vascular resistance somewhat. The in-creased preload raises the stroke volume andtherefore causes the murmur of aortic steno-sis to become louder. In contrast, the tran-sient increase in LV size during squatting re-duces the LV outflow tract obstruction inHCM and softens the intensity of that mur-mur. Sudden standing from a squatting posi-tion has the opposite effect on each of thesemurmurs (see Table 10.3).

Diagnostic Studies

The ECG typically shows left ventricular hy-pertrophy and left atrial enlargement. Promi-

nent Q waves are common in the inferiorand lateral leads, representing the forces ofinitial depolarization of the hypertrophiedseptum. In some patients, diffuse T wave inversions are present, which can predateclinical, echocardiographic, or other electro-cardiographic manifestations of HCM. Atrialand ventricular arrhythmias are frequent,especially atrial fibrillation. Ventricular ar-rhythmias are particularly ominous becausethey may herald ventricular fibrillation andsudden death, even in previously asymptom-atic patients.

Echocardiography is most helpful in theevaluation of HCM. The degree of LV hyper-trophy can be measured and regions ofasymmetrical wall thickness readily identi-fied. Signs of left ventricular outflow obstruc-tion may also be demonstrated and includeabnormal anterior motion of the mitral valveas it is drawn toward the hypertrophied sep-tum during systole, and partial closure of theaortic valve in midsystole as flow across it istransiently obstructed. Doppler recordingsduring echocardiography accurately measurethe outflow pressure gradient and quantifyany associated mitral regurgitation. Childrenand adolescents with apparently mild HCMshould undergo serial echocardiographic assessment over time, because the degree ofhypertrophy may increase during pubertyand early adulthood.

Cardiac catheterization is reserved for pa-tients for whom the diagnosis is uncertain orif percutaneous septal ablation (described inthe next section) is planned. The major fea-ture in patients with obstruction is the find-ing of a pressure gradient within the outflowportion of the left ventricle, either at rest orduring maneuvers that transiently reduce LVsize and promote outflow tract obstruction.Myocardial biopsy at the time of catheteri-zation is not necessary, because histologicfindings do not predict disease severity orlong-term prognosis.

Although genetic testing for HCM is not feasible on a wide-scale basis, futuregenotyping may provide a noninvasivetechnique for definitive diagnosis and riskstratification.

TABLE 10.3. Effect of Maneuvers on Murmursof Aortic Stenosis (AS) and Hyper-trophic Cardiomyopathy (HCM)

Valsalva Squatting Standing

HCM murmur ↑ ↓ ↑AS murmur ↓ ↑ ↓

Tab. 3


10090-10_CH10.qxd 8/31/06 4:53 PM Page 263

1 LINE LONG

Treatment

β-Blockers are the standard therapy for HCMbecause they (1) reduce myocardial oxygendemand by slowing the heart rate and theforce of contraction (and therefore diminishangina and dyspnea); (2) lessen the LV out-flow gradient during exercise by reducingthe force of contraction (allowing the cham-ber size to increase, thus separating the an-terior leaflet of the mitral valve from theventricular septum); (3) increase passive di-astolic ventricular filling time owing to thedecreased heart rate; and (4) decrease thefrequency of ventricular ectopic beats. De-spite their antiarrhythmic effect, β-blockershave not been shown to prevent sudden ar-rhythmic death in this condition.

Calcium channel antagonists can reduceventricular stiffness and are sometimes usefulin improving exercise capacity in patientswhose conditions fail to respond to β-block-ers. Patients who develop pulmonary con-gestion may benefit from mild diuretic ther-apy, but these drugs must be administeredcautiously to avoid volume depletion; re-duced intravascular volume decreases LV sizeand could exacerbate outflow tract obstruc-tion. Vasodilators (including nitrates) simi-larly reduce LV size and should be avoided.

Atrial fibrillation is poorly tolerated inHCM and should be controlled aggressively,most commonly with antiarrhythmic drugs. Ef-fective antiarrhythmics for atrial fibrillationin HCM include amiodarone and disopyra-mide (a type IA antiarrhythmic drug that alsopossesses negative inotropic properties thatmay help reduce LV outflow tract obstruc-tion; see Chapter 17). Digitalis should beavoided in HCM because its positive inotropiceffect increases the force of contraction andcan worsen LV outflow tract obstruction.

Sudden cardiac death has a propensity tooccur in patients with HCM in associationwith physical exertion; therefore, strenuousexercise and competitive sports should beavoided. Sudden death in this syndrome isalmost always caused by ventricular tachy-cardia or fibrillation. Although amiodaronemay reduce the frequency of ventricular ar-rhythmias, HCM patients who are at highrisk of sudden cardiac death (i.e., those who

have survived a cardiac arrest, have experi-enced episodes of syncope, display high-riskventricular arrhythmias, or have a ventricu-lar wall thickness >30 mm) should receivean ICD.

Infective endocarditis can develop in pa-tients with obstructive HCM because of tur-bulent blood flow through the narrowed LVoutflow tract and in association with the ac-companying mitral regurgitation. Antibioticprophylaxis (see Chapter 8) is therefore indi-cated to prevent endocardial infection duringsurgical procedures that result in bacteremia.

Some studies have shown clinical im-provement when patients with obstructiveHCM are treated with a dual-chamber perma-nent pacemaker, the electrodes of which areplaced in the right atrium and right ventricle.The LV outflow gradient may become re-duced by this procedure, possibly by alteringthe normal sequence of ventricular contrac-tion, such that septal-mitral valve appositionbecomes less prominent. This techniqueseems to be useful for only a small percentageof markedly symptomatic patients.

Surgical therapy (myomectomy) is consid-ered for patients whose symptoms do not re-spond to pharmacologic therapy. This pro-cedure involves excision of portions of thehypertrophied muscle mass and usually im-proves outflow tract obstruction, symptoms,and exercise capacity. A less invasive alter-native is percutaneous septal ablation, per-formed in the cardiac catheterization labora-tory, in which ethanol is injected directlyinto the first major septal coronary artery (a branch of the left anterior descendingartery), causing a small, controlled myocar-dial infarction. This results in reduced septalthickness and can lessen outflow tract ob-struction. Randomized trials have not beenperformed to compare the results of surgicalmyomectomy with ethanol septal ablation.Although septal ablation is an appropriateconsideration for patients who cannot toler-ate surgery, surgical myomectomy remainsthe current procedure of choice given its 40-year history of experience and efficacy.

Finally, genetic counseling should be pro-vided to all patients with HCM. Because it isan autosomal dominant disease, children ofaffected persons have a 50% chance of in-

264 Chapter Ten

10090-10_CH10.qxd 8/31/06 4:53 PM Page 264


heriting the abnormal gene. In addition,first-degree relatives of patients with HCMshould be screened by physical examination,electrocardiography, and echocardiography.Even asymptomatic HCM patients are at in-creased risk of complications, including sud-den death, and must be monitored closely.

Prognosis

The incidence of sudden death in HCM is2% to 4% per year in adults and 4% to 6% inchildren and adolescents. It has becomeclear that different mutations have vastlydifferent phenotypes. Some cause extremehypertrophy in childhood without any clin-ical symptoms until the occurrence of sud-den death; others manifest later in life withheart failure symptoms. Most mutationsproduce only mild hypertrophy and are as-sociated with a normal life expectancy. Asthe clinical outcomes of specific mutationsare better defined, the use and timing of spe-cific therapeutic interventions will likely beclarified.

RESTRICTIVE CARDIOMYOPATHY

The restrictive cardiomyopathies are lesscommon than DCM and HCM. They arecharacterized by abnormally rigid (but notnecessarily thickened) ventricles with im-paired diastolic filling but usually normal, ornear-normal, systolic function. This condi-tion results from either (1) fibrosis or scar-ring of the endomyocardium or (2) infiltra-tion of the myocardium by an abnormalsubstance, such as amyloid (Table 10.4).

The most common recognized cause of restrictive cardiomyopathy in nontropicalcountries is amyloidosis. In this systemicdisease, insoluble misfolded amyloid fibrilsdeposit within tissues, including the heart,causing organ dysfunction. Amyloid deposi-tion is diagnosed histologically by the Congored stain, which displays amyloid fibrils witha characteristic green birefringence.

Amyloid fibrils can pathologically developfrom a host of different proteins that distin-guish the categories of the disease. Primaryamyloidosis is caused by deposition of immu-noglobulin light chain AL fragments secreted

by a plasma cell tumor (usually, multiplemyeloma). In contrast, secondary amyloidosis ischaracterized by the presence of AA amyloid(derived from the inflammatory marker serumamyloid A) in a variety of chronic inflamma-tory conditions, such as rheumatoid arthritis.Less common is hereditary amyloidosis, an au-tosomal dominant condition in which amy-loid fibrils form from point mutations in theprotein transthyretin (formerly known as pre-albumin). Senile amyloidosis is common afterage 80, in which amyloid deposits, derivedfrom transthyretin or other proteins, are foundscattered throughout the vascular system,muscles, kidney, and lung. In each form ofamyloidosis, cardiac involvement is markedby deposition of extracellular amyloid be-tween myocardial fibers in the atria and ven-tricles, in the coronary arteries and veins, andin the heart valves.

Clinical manifestations of cardiac in-volvement are most common in the pri-mary (AL) form of the disease and typicallyrelate to the development of restrictive car-diomyopathy (described later in the chap-ter) because of the infiltrating amyloid pro-tein. Heart failure resulting from systolicventricular dysfunction also occurs, but lessfrequently. Orthostatic hypotension devel-ops in about 10% of patients, likely con-tributed to by amyloid deposition in the autonomic nervous system and peripheralblood vessels. Infiltration of amyloid intothe cardiac conduction system can causearrhythmias and conduction impairments,which can result in syncope or even suddendeath.

TABLE 10.4. Examples of RestrictiveCardiomyopathy

Myocardial Endomyocardial

Noninfiltrative Endomyocardial fibrosisIdiopathic Hypereosinophilic syndromeScleroderma Metastatic tumorsInfiltrative Radiation therapyAmyloidosisSarcoidosisStorage diseasesHemochromatosisGlycogen storage

diseases

Tab. 4

1 LINE LONG

10090-10_CH10.qxd 8/31/06 4:53 PM Page 265

266 Chapter Ten

Pathophysiology

Reduced compliance of the ventricles owingto fibrosis or infiltration results in an up-ward shift of the passive ventricular fillingcurve (see Fig. 9.7B), leading to abnormallyhigh intraventricular pressure throughoutdiastole. This condition has two major con-sequences: (1) elevated systemic and pul-monary venous pressures, with signs of right-and left-sided vascular congestion, and (2)reduced ventricular cavity size with de-creased stroke volume and cardiac output.

Clinical Findings

It follows from the underlying pathophysi-ology that signs of left- and right-sided heartfailure are expected (Fig. 10.8). Decreasedcardiac output is manifested by fatigue anddecreased exercise tolerance. Systemic con-gestion (often more prominent than pul-monary congestion in this syndrome) leadsto jugular venous distention, peripheraledema, and ascites with a large, tender liver.Arrhythmias such as atrial fibrillation arecommon. Infiltrative etiologies that involvethe cardiac conduction system can causevarious types of heart block.


Signs of congestive heart failure are often pre-sent, including pulmonary rales, distendedneck veins, ascites, and peripheral edema.Similar to constrictive pericarditis (seeChapter 14), jugular venous distention may

paradoxically worsen with inspiration (theKussmaul sign) because the stiffened rightventricle cannot accommodate the increasedvenous return.

Diagnostic Studies

The chest radiograph usually shows a normal-sized heart with signs of pulmonary conges-tion. The ECG often displays nonspecific STand T wave abnormalities; conduction dis-turbances such as atrioventricular block or abundle branch block may be present.

The restrictive cardiomyopathies sharenearly identical symptoms, physical signs,and hemodynamic profiles with constrictivepericarditis, as described in Chapter 14. How-ever, it is important to distinguish betweenthese two entities, because constrictive peri-carditis is a correctable condition, whereasthe restrictive cardiomyopathies generallyare not treatable.

The most useful diagnostic tools to dif-ferentiate restrictive cardiomyopathy fromconstrictive pericarditis are transvenous en-domyocardial biopsy, computed tomogra-phy (CT), and magnetic resonance imaging(MRI). For example, in restrictive cardio-myopathy, a transvenous endomyocardialbiopsy may demonstrate the presence of infiltrative matter such as amyloid, iron de-posits (hemochromatosis), or metastatictumors. Conversely, CT or MRI scans areuseful to identify the thickened pericar-dium of constrictive pericarditis, a findingthat is not expected in restrictive cardio-myopathy.

Hepatomegaly and ascites

Figure 10.8. Pathophysiology of restrictive cardiomyopathy. The rigid myocardium results in elevatedventricular diastolic pressures and decreased ventricular filling. The resultant symptoms can be predictedfrom these abnormalities. CO, cardiac output.

Fig. 8

10090-10_CH10.qxd 8/31/06 4:53 PM Page 266

Treatment

Restrictive cardiomyopathy typically has avery poor prognosis, except when treat-ment can target an underlying cause. Forexample, phlebotomy and iron chelationtherapy may be helpful in the early stagesof hemochromatosis. Symptomatic therapyfor all etiologies includes salt restriction andcautious use of diuretics to improve symp-toms of systemic and pulmonary conges-tion. Unlike the dilated cardiomyopathies,digitalis and vasodilators are not helpfulbecause systolic function is usually pre-served. Maintenance of sinus rhythm is im-portant to maximize diastolic filling and

forward cardiac output. Some restrictivecardiomyopathies are prone to intraven-tricular thrombus formation, thus warrant-ing chronic oral anticoagulant therapy iswarranted. In the case of primary (AL) amy-loidosis, chemotherapy followed by auto-logous bone marrow stem cell transplanthas proved effective in selected patients withearly cardiac involvement.

SUMMARY

1. The cardiomyopathies are diseases of heartmuscle classified by their pathophysio-logic presentation into dilated, hypertro-phic, or restrictive types (Table 10.5).

TABLE 10.5. Summary of the Cardiomyopathies

Dilated Hypertrophic Restrictive Cardiomyopathy Cardiomyopathy Cardiomyopathy

Ventricular morphology

Symptoms

Physical exam

Pathophysiology

Cardiac size on chest radiograph

Echocardiogram

JVD, jugular venous distension; LV, left ventricle; MV, mitral valve; PND, paroxysmal nocturnal dyspnea; RV, right ventricle.

Dilated LV with littlehypertrophy

Fatigue, weakness, dysp-nea, orthopnea, PND(symptoms of conges-tive heart failure)

Pulmonary rales, S3; if RVfailure present: JVD, he-patomegaly, peripheraledema

Impaired systoliccontraction

Dilated

Dilated, poorly contrac-tile LV

Marked hypertrophy,often asymmetric

Dyspnea, angina, syncope

S4; if outflow obstructionpresent: systolic mur-mur loudest at left ster-nal border, accompa-nied by mitralregurgitation

Impaired diastolic relax-ation; LV systolicfunction vigorous,often with dynamic obstruction

Normal or dilated

LV hypertrophy, oftenmore pronounced inseptum; systolic anteriormovement of MV withmitral regurgitation

Fibrotic or infiltratedmyocardium

Dyspnea, fatigue

Signs of RV failure: JVD,hepatomegaly, periph-eral edema

“Stiff” LV with impaireddiastolic relaxation butnormal systolic function

Usually normal

Usually normal systoliccontraction; “speckled”appearance in infiltra-tive disorders


Tab. 5

10090-10_CH10.qxd 8/31/06 4:53 PM Page 267

268 Chapter Ten

2. Dilatedcardiomyopathiesarecharacterizedby ventricular dilatation with impaired systolic function. Progressive left ventricu-lar enlargement often leads to sympto-matic heart failure, ventricular arrhyth-mias, and/or embolic complications.

3. HCM is characterized by a thickened leftventriclewithimpaireddiastolicrelaxation. Dynamic LV outflow tract obstructionduring systole may be present. The mostcommon symptoms are dyspnea and exer-tional angina. Ventricular arrhythmiasmay lead to sudden cardiac death.

4. The restrictive cardiomyopathies are un-common and are characterized by impair-ment of diastolic ventricular relaxationowing to an infiltrated or fibrotic myo-cardium. Symptoms of heart failure aretypical.

AcknowledgmentThe authors thank Frederick Schoen, MD, for pro-viding pathology specimens. Contributors to theprevious editions of this chapter were Yi-Bin Chen,MD; Kay Fang, MD; David Grayzel, MD; G. WilliamDec, MD; and Leonard S. Lilly, MD.

Additional Reading

Burkett EL, Hershberger RE. Clinical and genetic issuesin familial dilated cardiomyopathy. J Am Coll Car-diol 2005;45:969–981.

Dec GW, Fuster VF. Idiopathic dilated cardiomyopa-thy. N Engl J Med 1994;331:1564–1575.

Falk RH. Diagnosis and management of the cardiacamyloidoses. Circulation 2005;112:2047–2060.

Kushwaha SS, Fallon JT, Fuster V. Medical progress: re-strictive cardiomyopathy. N Engl J Med 1997;336:267–276.

Magnani JW, Dec GW. Myocarditis: current trendsin diagnosis and treatment. Circulation 2006;113:876–890.

Morita H, Seidman J, Seidman CE. Genetic causes of human heart failure. J Clin Invest 2005;115:518–526.

Nishimura RA, Holmes DR Jr. Hypertrophic obstruc-tive cardiomyopathy. N Engl J Med 2004;350:1320–1327.

Richardson P, McKenna W, Bristow M, et al. Report ofthe 1995 World Health Organization/InternationalSociety and Federation of Cardiology Task Forceon the definition and classification of cardio-myopathies. Circulation 1996;93:841–842.

Seidman J, Seidman CE. Proposal for contemporaryscreening strategies in families with hypertrophiccardiomyopathy. J Am Coll Cardiol 2004;44:2125–2132.

Skinner M, Sanchorawala V, Seldin DC, et al. High-dose melphalan and autologous stem-cell trans-plantation in patients with AL amyloidosis: an 8-year study. Ann Intern Med 2004;140:85–93.

Yazdani K, Maraj S, Amanullah AM. Differentiatingconstrictive pericarditis from restrictive cardiomy-opathy. Rev Cardiovasc Med 2005;6:61–71.

10090-10_CH10.qxd 8/31/06 4:53 PM Page 268


1. AU: Chapter 9 only refers to this as a system, not an axis. Are both OK, or should termi-nology be consistent?2. AU: As in ch. 9. OK?3. AU: There is a new edition of this book. Should figure/citation be updated?

10090-10_CH10.qxd 8/31/06 4:53 PM Page a

269

NORMAL IMPULSE FORMATIONIonic Basis of AutomaticityNative and Latent PacemakersOverdrive SuppressionElectrotonic Interactions

ALTERED IMPULSE FORMATIONAlterations in Sinus Node AutomaticityEscape RhythmsEnhanced Automaticity of Latent Pacemakers

Abnormal AutomaticityTriggered Activity

ALTERED IMPULSE CONDUCTIONConduction BlockUnidirectional Block and Reentry

APPROACHES TO ANTIARRHYTHMICTREATMENTBradyarrhythmiasTachyarrhythmias

C H A P T E R

11Mechanisms of Cardiac ArrhythmiasHillary K. RollsWilliam G. StevensonGary R. StrichartzLeonard S. Lilly

Normal cardiac function relies on the flowof electrical impulses through the heart inan exquisitely coordinated fashion. Abnor-malities of the electrical rhythm are knownas arrhythmias (also termed dysrhythmias)and are among the most common clinicalproblems encountered. The presentations ofarrhythmias range from benign palpitationsto severe symptoms of low cardiac outputand death; therefore, a thorough under-standing of these disorders is important tothe daily practice of medicine.

Abnormally slow rhythms are termed bra-dycardias (or bradyarrhythmias). Fast rhythmsare known as tachycardias (or tachyarrhyth-mias). Tachycardias are further character-ized as supraventricular when they involve

the atrium or AV node and designatedventricular when they originate from theHis-Purkinje system or ventricles. This chap-ter describes the mechanisms by whichsuch arrhythmias develop, followed by ageneral description of their management.Chapter 12 summarizes specific rhythmdisorders and how to recognize and treatthem.

Disorders of heart rhythm result from al-terations of impulse formation, impulseconduction, or both. This chapter first ad-dresses how alterations of impulse forma-tion and conduction occur and under whatcircumstances they cause arrhythmias. Fig-ure 11.1 provides an organizational schemafor this discussion.

Fig. 1

10090-11_CH11.qxd 8/31/06 4:54 PM Page 269

NORMAL IMPULSE FORMATION

As presented in Chapter 1, electrical impulseformation in the heart arises from the intrin-sic automaticity of specialized cardiac cells.Automaticity refers to a cell’s ability to depolarize itself to a threshold voltage in arhythmic, repeated fashion, such that spon-taneous action potentials are generated. Al-though atrial and ventricular myocytes donot have this property under normal condi-tions, the cells of the specialized conductingsystem do possess natural automaticity andare therefore termed pacemaker cells. Thespecialized conducting system includes thesinoatrial (SA) node, the atrioventricular (AV)nodal region, and the ventricular conductingsystem. The latter is composed of the bundleof His, the bundle branches, and the Purkinjefibers. In pathologic situations, myocardial cellsoutside the conducting system may also ac-quire the property of automaticity.

Ionic Basis of Automaticity

Cells with natural automaticity do not havea static resting potential. Rather, they dis-play a gradual depolarization during phase 4of the action potential (Fig. 11.2) If this

spontaneous diastolic depolarization reachesthe threshold voltage, an action potential is generated. An important ionic currentlargely responsible for phase 4 spontaneousdepolarization is known as the pacemakercurrent (If). This current is activated by hy-perpolarization (increasingly negative volt-ages) and is carried mainly by sodium ions.The channels that carry If open when themembrane voltage becomes more negativethan approximately −50 mV and are differ-ent from the fast sodium channels responsi-ble for rapid phase 0 depolarization in non-pacemaker cells. The inward flow of Na+

through these slow channels, driven by itsconcentration gradient and the negative in-tracellular charge, forces the membrane po-tential to depolarize toward the thresholdvoltage.

In the pacemaker cells of the sinoatrialnode, alterations in two other ionic currentsalso contribute to phase 4 depolarization:(1) a slow inward calcium current, the chan-nels of which become activated at voltagesreached near the end of phase 4, and (2) aprogressive decline of an outward potassiumcurrent. Activation of the latter current is re-sponsible for cellular repolarization duringphase 3 of the action potential, and it pro-

270 Chapter Eleven

Automaticityof SA node

Automaticityof latent

pacemakers

Enhancedautomaticity

Abnormalautomaticity

Triggeredactivity

Automaticityof SA node

Unidirectional blockand reentry

Conduction block

ALTEREDIMPULSE

FORMATION

ALTEREDIMPULSE

CONDUCTION

TACHYARRHYTHMIAS (Increased firing rate)

BRADYARRHYTHMIAS (Decreased firing rate)

Figure 11.1. Arrhythmias result from alterations in impulse formation and/or impulse con-duction. Tachyarrhythmias result from enhanced automaticity, unidirectional block with reentry, or trig-gered activity. Bradyarrhythmias result from decreased automaticity or conduction block. SA, sinoatrial.

Fig. 2

10090-11_CH11.qxd 8/31/06 4:54 PM Page 270

Mechanisms of Cardiac Arrhythmias 271

gressively diminishes during phase 4. Thecombination of the inward If, inward, Ca++

and reduced outward K+ currents acts togradually depolarize the SA nodal cells tothe threshold potential.

When the membrane potential of thepacemaker cell reaches the threshold value,the upstroke of the action potential is gen-erated. In contrast to the phase 0 upstroke ofcells in the Purkinje system, that of cells inthe sinus and AV nodes is much slower (seeFig. 11.2; compare with Fig. 1.14). The rea-son for the difference is that the membranepotential determines the proportion of fastsodium channels that are in a resting statecapable of depolarization, compared withan inactivated state. The number of avail-able (or resting-state) fast sodium channelsincreases as the resting membrane potentialbecomes more negative. Because sinus andAV nodal cells have less negative maximumdiastolic membrane voltages (−50 to −60 mV)than do Purkinje cells (−90 mV), a greaterproportion of the fast sodium channels is in-activated in these pacemaker cells. Thus, theaction potential upstroke relies to a greater ex-tent on calcium ion inflow (through the rela-tively slower opening Ca++ channels) and itsslope is less steep in these cells. The repolar-

ization phase of pacemaker cells depends oninactivation of the calcium channels and theopening of voltage-gated potassium channelsthat permit efflux of potassium from the cells.

Native and Latent Pacemakers

The different populations of automatic cellsin the specialized conduction pathway havedifferent intrinsic rates of firing. These ratesare determined by three variables that in-fluence how fast the membrane potentialreaches threshold: (1) the rate (i.e., the slope)of phase 4 spontaneous depolarization, (2)the maximum negative diastolic potential,and (3) the threshold potential. A more neg-ative maximum diastolic potential, or a lessnegative threshold potential, slows the rateof impulse initiation because it takes longerto reach that threshold value (Fig. 11.3). Con-versely, the greater the If, the steeper theslope of phase 4 and the faster the cell de-polarizes. The rate of If depends on the num-ber and kinetics of the individual pacemakerchannels through which this current flows.

Because all the healthy myocardial cellsare electrically connected by gap junctions,an action potential generated in one part ofthe myocardium will ultimately spread to all

Ifinflux

Figure 11.2. The action potential (AP) of a pacemaker cell. Noticethe slow phase 4 depolarization, largely caused by the If (pacemaker) cur-rent through slow Na+ channels, which drives the cell to threshold po-tential (approximately −40 mV). The upstroke of the AP is caused by theslow inward current of Ca+ ions. Inactivation of the calcium channels andK+ efflux through potassium channels are responsible for repolarization.MDP, maximum negative diastolic potential; TP, threshold potential.

Fig. 3

10090-11_CH11.qxd 8/31/06 4:54 PM Page 271

other regions. When an impulse arrives at acell that is not yet close to threshold, currentfrom the depolarized cell will bring theadjacent cell membrane potential to thethreshold level so that it will fire (regardlessof how close its intrinsic If has brought it tothreshold). Thus, the pacemaker cells withthe fastest rate of depolarization set theheart rate. In the normal heart, the domi-nant pacemaker is the sinoatrial node, whichat rest initiates impulses at a rate of 60 to100 bpm. Because the sinus node rate isfaster than that of the other tissues that pos-sess automaticity, its repeated dischargesprevent spontaneous firing of other poten-tial pacemaker sites.

The SA node is known as the native pace-maker because it normally sets the heart rate.

Other cells within the specialized conductionsystem harbor the potential to act as pace-makers if necessary and are therefore calledlatent pacemakers (or ectopic pacemak-ers). In contrast to the SA node, the AV nodeand the bundle of His have intrinsic firingrates of 50 to 60 bpm, and cells of the Purk-inje system have rates of approximately 30 to40 bpm. These latent sites may initiate im-pulses and take over the pacemaking func-tion if the SA node slows or fails to fire, or ifconduction abnormalities block the normalwave of depolarization from reaching them.

Overdrive Suppression

Not only does the cell population with thefastest intrinsic rhythm preempt all other

272 Chapter Eleven

a b

a b

c

Figure 11.3. Determinants of cell-firing rates. A. Alterations in the pace-maker current (If) and in the magnitude of the maximum diastolic potential(MDP) alter the cell-firing rate. (a) The normal action potential (AP) of a pace-maker cell. (b) Reduced If renders the slope of phase 4 less steep; thus, thetime required to reach threshold potential (TP) is increased. (c) The MDP ismore negative; therefore, the time required to reach TP is increased. B. Alter-ations in TP change the firing rate of the cell. Compared with the normal TP(a), the TP in b is less negative; thus, the duration of time to achieve thresh-old is increased, and the firing rate decreases.

10090-11_CH11.qxd 8/31/06 4:54 PM Page 272


automatic cells from spontaneously firing, italso directly suppresses their automaticity.This phenomenon is called overdrive sup-pression. Cells maintain their transsarco-lemmal ion distributions because of the con-tinuously active Na+K+-ATPase pump thatextrudes three Na+ ions from the cell in ex-change for two K+ ions transported in (Fig.11.4). Because its net transport effect is onepositive charge in the outward direction, theNa+K+-ATPase pump creates a hyperpolarizingcurrent (i.e., it tends to make the inside of thecell more negative). As the cell potential be-comes increasingly negative, additional timeis required for spontaneous phase 4 depolar-ization to reach the threshold voltage (see Fig.11.3A), and therefore the rate of spontaneousfiring is decreased. Although the hyperpolar-izing current antagonizes If, pacemaker cellsfiring at their own intrinsic rate have an If cur-rent sufficiently large to overcome this hyper-polarizing influence (see Fig. 11.4).

The hyperpolarizing current increaseswhen a cell is forced to fire faster than its in-trinsic pacemaker rate. The more frequentlythe cell is depolarized, the greater the quan-tity of Na+ ions that enter the cell per unittime. As a result of the increased intracellularNa+, the Na+K+-ATPase pump becomes moreactive, thereby tending to restore the normaltransmembrane Na+ gradient. This increasedpump activity provides a larger hyperpolar-izing current, opposing the depolarizing cur-

rent If, and further decreases the rate of spon-taneous depolarization. Thus, overdrive sup-pression decreases a cell’s automaticity whenthat cell is driven to depolarize faster than itsintrinsic discharge rate.

Electrotonic Interactions

In addition to overdrive suppression, anato-mic connections between pacemaker and non-pacemaker cells are important in determininghow adjacent cells suppress latent pacemakerfoci. Myocardial cells that are not part of thespecialized conducting system repolarize to aresting potential of −90 mV, whereas pace-maker cells repolarize to a maximum dias-tolic potential of about −60 mV. When thesetwo cell types are adjacent to one another,they are electrically coupled through low resis-tance gap junctions concentrated in their in-tercalated discs. This coupling results in anequilibration of electrical potentials owing toelectrotonic current flow between the cells,causing relative hyperpolarization of the pace-maker cell and relative depolarization of thenonpacemaker cell (Fig. 11.5). The hyperpo-larizing current in the coupled pacemakercell competes with If and causes the slope ofphase 4 diastolic depolarization to be lesssteep, thereby reducing the cell’s automatic-ity. Electrotonic effects may be particularlyimportant in suppressing automaticity in theAV node (via connections between atrialmyocytes and AV nodal cells) and in the dis-tal Purkinje fibers (which are coupled tononautomatic ventricular myocardial cells).In contrast, cells in the center of the SA nodeare less tightly coupled to atrial myocytes;thus, their automaticity is less subject to elec-trotonic interactions.

Decoupling of normally suppressed cells,such as those in the AV node (e.g., by ische-mic damage), may reduce the inhibitoryelectrotonic influence and enhance automa-ticity, producing ectopic rhythms by the la-tent pacemaker tissue.

ALTERED IMPULSE FORMATION

Arrhythmias may arise from altered impulseformation at the SA node or from othersites, including the specialized conduction

Figure 11.4. Competition between the de-polarizing pacemaker current (If) and theNa++K++-ATPase pump, which produces a hy-perpolarizing current. The Na+K+-ATPasepump transports three positive charges outsidethe cell for every two it pumps in. The hyperpo-larizing current acts to suppress automaticity byantagonizing If and contributes to overdrive sup-pression in cells that are stimulated more rapidlythan their intrinsic firing rate.

AQ3

AQ1

Fig. 4

Fig. 5

10090-11_CH11.qxd 8/31/06 4:54 PM Page 273

pathways or regions of cardiac muscle. Themain abnormalities of impulse initiationthat lead to arrhythmias are (1) altered au-tomaticity (of the sinus node or latent pace-makers within the specialized conductionpathway), (2) abnormal automaticity inatrial or ventricular myocytes, and (3) trig-gered activity.

Alterations in Sinus Node Automaticity

The rate of impulse initiation by the sinusnode, as well as by the latent pacemakers ofthe specialized conducting system, is regu-lated primarily by neurohumoral factors.

Increased Sinus Node Automaticity

The most important modulator of normalsinus node automaticity is the autonomicnervous system. Sympathetic stimulation,

acting through β1-adrenergic receptors, in-creases the probability of the pacemakerchannels being open (Fig. 11.6), throughwhich If can flow. The increase in If leads toa steeper slope of phase 4 depolarization,causing the SA node to reach threshold andfire earlier than normal and the heart rate toincrease.

In addition, sympathetic stimulation shiftsthe action potential threshold to more-negative voltages by increasing the proba-bility that voltage-sensitive Ca++ channelsare open (recall that calcium carries the cur-rent of phase 0 depolarization in pacemakercells). Therefore, diastolic depolarizationreaches the threshold potential earlier. Sym-pathetic activity thus increases sinus nodeautomaticity both by causing the action po-tential threshold to become more negativeand by increasing the rate of pacemaker de-polarization via If. Examples of this normalphysiologic effect occur during exercise or

274 Chapter Eleven

Figure 11.5. Electrotonic interaction between pacemaker (e.g., AV nodal) and nonpacemaker (myocardial)cells. A. When pacemaker cells are not coupled to myocardial cells (as in the SA node), they have a maximum neg-ative potential (MDP) of approximately −60 mV, whereas myocardial cells have a resting potential (RP) of approxi-mately −90 mV. B. When pacemaker cells and myocytes are neighbors, they may be connected electrically by gapjunctions in their intercalated discs (e.g., in the AV node). In this situation, electrical current flows between the pace-maker cell and the myocardial cell, tending to hyperpolarize the former and depolarize the latter, driving their mem-brane potentials closer to one another. The more negative potential of atrial cells opposes If of the pacemaker cell,such that the slope of phase 4 depolarization is less steep and therefore cellular automaticity is suppressed. If a dis-ease state reduces coupling between cells, the influence of surrounding myocytes on the pacemaker cell is reduced,allowing If to depolarize to threshold more readily and accelerating the rate of automaticity. TP, threshold potential.

10090-11_CH11.qxd 8/31/06 4:54 PM Page 274


emotional stress, when sympathetic stimula-tion appropriately increases the heart rate.

Decreased Sinus Node Automaticity

Normal decreases in SA node automaticityare mediated by reduced sympathetic stim-ulation and by increased activity of the para-sympathetic nervous system. Whereas thesympathetic nervous system exerts a domi-nant effect on the heart rate during times ofstress, the parasympathetic nervous systemis the major mediator of the heart rate atrest.

Cholinergic (i.e., parasympathetic) stim-ulation via the vagus nerve acts at the SAnode to reduce the probability of pacemakerchannels being open (see Fig. 11.6). Thus, If

and the slope of phase 4 depolarization arereduced, and the intrinsic firing rate of thecell is slowed. In addition, the probability ofthe Ca++ channels being open is decreased;

thus, the action potential threshold increasesto a more positive potential. Furthermore,cholinergic stimulation increases the proba-bility of the acetylcholine-sensitive K+ chan-nels being open at rest. Positively charged K+

ions exit through these channels, producingan outward current that drives the diastolicpotential to become more negative. Theoverall effect of reduced If, a more negativemaximum diastolic potential, and a lessnegative threshold level is a slowing of theintrinsic firing rate and therefore a reducedheart rate.

It follows that the use of pharmacologicagents that modify the effects of the auto-nomic nervous system will also affect the firing rate of the SA node. For example, β-blocking drugs antagonize the β-adrener-gic sympathetic effect; therefore, they de-crease the rate of phase 4 depolarization ofthe SA node and slow the heart rate. Con-versely, atropine, an anticholinergic (anti-muscarinic) drug, has the opposite effect: byblocking parasympathetic activity, the rateof phase 4 depolarization increases and theheart rate accelerates.

Escape Rhythms

If the sinus node becomes suppressed andfires less frequently than normal, the site ofimpulse formation often shifts to a latentpacemaker within the specialized conduc-tion pathway. An impulse initiated by a la-tent pacemaker because the SA node rate hasslowed is called an escape beat. Persistentimpairment of the SA node will allow a con-tinued series of escape beats, termed an es-cape rhythm. Escape rhythms are protec-tive in that they prevent the heart rate frombecoming too slow when SA node firing isimpaired.

As discussed in the previous section, sup-pression of sinus node activity may occur be-cause of increased parasympathetic tone.Different regions of the heart have differentsensitivities to parasympathetic (vagal) stim-ulation. The SA node and the AV node aremost sensitive to such an influence, followedby atrial tissue. The ventricular conductingsystem is the least sensitive. Therefore, mod-

Figure 11.6. The channels through which thepacemaker current (If) flows are voltagegated, opening at more negative membranepotentials. At any given voltage, there exists aprobability between 0 and 1 that a specific chan-nel will be open. Compared with normal baselinebehavior (curve A), sympathetic stimulation (curveB) or treatment with anticholinergic drugs shiftsthis probability to a higher value for any given levelof membrane voltage, thus increasing the numberof open channels and the rate at which the cell willfire. Curve C shows that parasympathetic stimula-tion (or treatment with β-blockers) has the oppositeeffect, decreasing the probability of a channel beingopen, and therefore inhibiting depolarization.

10090-11_CH11.qxd 8/31/06 4:54 PM Page 275

erate parasympathetic stimulation slowsthe sinus rate and allows the pacemaker toshift to another atrial site. However, verystrong parasympathetic stimulation sup-presses excitability at both the SA node andatrial tissue, can cause conduction block atthe AV node, and may therefore result inthe emergence of a ventricular escape pace-maker.

Enhanced Automaticity of Latent Pacemakers

Another means by which a latent pace-maker can assume control of impulse for-mation is if it develops an intrinsic rate ofdepolarization faster than that of the sinusnode. Termed an ectopic beat, the impulseis premature relative to the normal rhythm,whereas an escape beat is late and terminatesa pause caused by a slowed sinus rhythm. Asequence of similar ectopic beats is called anectopic rhythm.

Ectopic beats may arise in several cir-cumstances. For example, high catechola-mine concentrations can enhance the auto-maticity of latent pacemakers, and if theresulting rate of depolarization exceeds thatof the sinus node, then an ectopic rhythmwill develop. Ectopic beats are also com-monly induced by hypoxemia, ischemia,electrolyte disturbances, and certain drugtoxicities (such as digitalis, as described inChapter 17).

Abnormal Automaticity

Cardiac tissue injury may lead to pathologicchanges in impulse formation whereby myo-cardial cells outside the specialized con-duction system acquire automaticity andspontaneously depolarize. Although suchactivity may appear similar to impulses orig-inating from latent pacemakers within thespecialized conduction pathways, these ec-topic beats arise from cells that do not usu-ally possess automaticity. If the rate of de-polarization of such cells exceeds that of thesinus node, they transiently take over thepacemaker function and become the sourceof an abnormal ectopic rhythm.

Because these myocardial cells have fewor no activated pacemaker channels, theydo not normally carry If. How injury allowssuch cells to spontaneously depolarize hasnot been fully elucidated. However, whenmyocytes become injured, their membranesbecome “leaky.” As such, they are unable tomaintain the concentration gradients ofions, and the resting potential becomes lessnegative (i.e., the cell partially depolarizes).When a cell’s membrane potential is reducedto a value less negative than −60 mV, grad-ual phase 4 depolarization can be demon-strated even among nonpacemaker cells.This slow spontaneous depolarization isprobably related to a slow calcium currentand by closure of a subset of K+ channelsthat normally help repolarize the cell.

Triggered Activity

Under certain conditions, an action poten-tial can “trigger” abnormal depolarizationsthat result in extra heart beats or rapid ar-rhythmias. This process may occur whenthe first action potential leads to oscillationsof the membrane voltage known as afterde-polarizations. Unlike the spontaneous activityseen when enhanced automaticity occurs,this type of automaticity is stimulated by apreceding action potential. As illustrated inFigures 11.7 and 11.8, there are two types of afterdepolarizations depending on theirtiming after the inciting action potential:early afterdepolarizations occur during therepolarization phase of the inciting beat,whereas delayed afterdepolarizations occurshortly after repolarization has been com-pleted. In either case, abnormal action po-tentials are triggered if the afterdepolariza-tion reaches a threshold voltage.

Early afterdepolarizations are changesof the membrane potential in the positive di-rection that interrupt normal repolarization(see Fig 11.7). They can occur either duringthe plateau of the action potential (phase 2)or during rapid repolarization (phase 3).Early afterdepolarizations are more likely todevelop in conditions that prolong the ac-tion potential duration (and therefore theelectrocardiographic QT interval), as may

276 Chapter Eleven

Fig. 7-8

10090-11_CH11.qxd 8/31/06 4:54 PM Page 276


occur during therapy with certain drugs (seeChapter 17) and in the inherited long-QTsyndromes (see Chapter 12).

The ionic current responsible for an earlyafterdepolarization depends on the mem-brane voltage at which the triggered eventoccurs. If the early afterdepolarization oc-curs during phase 2 of the action potential,when most of the Na+ channels are in an in-activated state, the upstroke of the triggeredbeat relies on an inward Ca++ current. If,however, it occurs during phase 3 (when themembrane voltage is more negative), thereis partial recovery of the fast Na+ channels,which are then available to contribute to thecurrent.

An early afterdepolarization-triggered ac-tion potential can be self-perpetuating andlead to a series of depolarizations (see Fig.11.7). Early afterdepolarizations appear to

be the initiating mechanism of the poly-morphic ventricular tachycardia known astorsades de pointes, which is described inChapter 12.

Delayed afterdepolarizations may ap-pear shortly after repolarization is complete(see Fig. 11.8). They most commonly de-velop in states of high intracellular calcium, asmay be present with digitalis intoxication(see Chapter 17), or during marked cate-cholamine stimulation. It is thought that in-tracellular Ca++ accumulation causes the ac-tivation of chloride currents or of theNa+-Ca++ exchanger that results in brief in-ward currents that generate the delayedafterdepolarization.

As with early afterdepolarizations, if theamplitude of the delayed afterdepolarizationreaches a threshold voltage, an action poten-tial will be generated. Such action potentials

Figure 11.7. Triggered activity. An early afterdepolarization(arrow) occurs before the action potential (AP) has fully repolarized.Repetitive afterdepolarizations (dashed curve) may produce a rapidsequence of triggered action potentials and hence a tachyarrhythmia.

Figure 11.8. Triggered activity. A delayed afterdepolarization (arrow)arises after the cell has fully repolarized. If the delayed afterdepolarizationreaches the threshold voltage, a propagated action potential (AP) is triggered(dashed curve).

10090-11_CH11.qxd 8/31/06 4:54 PM Page 277

can be self-perpetuating and lead to tachy-arrhythmias. For example, atrial and ven-tricular tachycardias associated with digi-talis toxicity are thought to be the result ofdelayed afterdepolarizations, as described inChapter 17.

ALTERED IMPULSE CONDUCTION

Alterations in impulse conduction also leadto arrhythmias. Conduction blocks gener-ally slow the heart rate (bradyarrhythmias);however, under certain circumstances, theprocess of reentry (described later) can ensueand produce abnormal fast rhythms (tachy-arrhythmias).

Conduction Block

A propagating impulse is blocked when it en-counters a region of the heart that is electri-cally unexcitable. Conduction block can beeither transient or permanent and may beunidirectional (i.e., conduction proceedswhen the involved region is stimulated fromone direction, but not when stimulated fromthe opposite direction) or bidirectional (con-duction is blocked in both directions). Vari-ous conditions may cause conduction block,including ischemia, fibrosis, inflammation,and certain drugs. When conduction blockoccurs because a propagating impulse en-counters cardiac cells that are still refractory(from a previous depolarization), the block issaid to be functional. A propagating impulsethat arrives later may be able to be conducted.For example, antiarrhythmic drugs thatprolong action potential duration tend toproduce functional conduction block. Whenconduction block is caused by a barrier im-posed by fibrosis or scarring that replaces myo-cytes, conduction block is fixed.

Conduction block within the specializedconducting system of the AV node or theHis-Purkinje system prevents normal propa-gation of the cardiac impulse from the sinusnode to more distal sites. This atrioventricu-lar block (AV block) removes the normaloverdrive suppression that keeps latentpacemakers in the His-Purkinje system incheck. Thus, conduction block usually re-

sults in emergence of escape beats or escaperhythms, as the more distal sites assume thepacemaker function.

AV block is common and a major reasonfor implantation of a permanent pacemaker,as discussed in Chapter 12.

Unidirectional Block and Reentry

A common mechanism by which a combi-nation of conduction block and altered impulse conduction leads to tachyarrhyth-mias is termed reentry. During a reentrantrhythm, an electrical impulse circulates re-peatedly around a reentry path, recurrentlydepolarizing a region of cardiac tissue.

During normal cardiac conduction, eachelectrical impulse that originates in the SAnode travels in an orderly, sequential fash-ion through the rest of the heart, ultimatelydepolarizing all the myocardial fibers. Therefractory period of each cell prevents im-mediate reexcitation from adjacent depolar-ized cells, so that the impulse stops when allof the heart muscle has been excited. How-ever, conduction blocks that prevent rapiddepolarization of parts of the myocardiumcan create an environment conducive tocontinued impulse propagation and reen-try, as illustrated in Figure 11.9.

The figure depicts electrical activity as itflows through a branch point anywhere with-in the conduction pathways. Panel A showspropagation of a normal action potential. Atpoint x, the impulse reaches two parallelpathways (α and β) and travels down eachinto the more distal conduction tissue. In thenormal heart, the α and β pathways havesimilar conduction velocities and refractoryperiods such that the wave fronts that passthrough them collide in the distal conduc-tion tissue and extinguish each other.

Panel B shows what happens if conduc-tion is blocked in one limb of the pathway.In this example, the action potential is ob-structed when it encounters the β pathwayfrom above and therefore propagates onlydown the α tract into the distal tissue. Asthe impulse continues to spread, it encoun-ters the distal end of the β pathway (at pointy). If the tissue in the distal β tract is also

278 Chapter Eleven

Fig. 9

10090-11_CH11.qxd 8/31/06 4:54 PM Page 278


Figure 11.9. Mechanism of reentry. A. Normalconduction. When an action potential (AP)reaches a branch in the conduction pathway(point x), the impulse travels down both fibers (αand β) to excite distal conduction tissue. B. Uni-directional block. Forward passage of the im-pulse is blocked in the β pathway but proceedsnormally down the α pathway. When the im-pulse reaches point y, if retrograde conduction ofthe β pathway is intact, the AP can enter β frombelow and conduct in a retrograde fashion. C.When point x is reached again, if the α pathwayhas not had sufficient time to repolarize, then theimpulse stops. D. However, if conductionthrough the retrograde pathway is sufficientlyslow (jagged line), it reaches point x after the αpathway has recovered. In that circumstance, theimpulse is able to excite the α pathway again anda reentrant loop is formed.

unable to conduct, the impulse simply con-tinues to propagate into the deeper tissuesand reentry does not occur. However, if theimpulse at point y is able to propagate ret-rogradely (backward) into pathway β, one

of the necessary conditions for reentry hasbeen met.

When an action potential can conduct ina retrograde direction in a conduction path-way, whereas it had been prevented from

10090-11_CH11.qxd 8/31/06 4:54 PM Page 279

doing so in the forward direction, unidirec-tional block is said to be present. Unidirec-tional block tends to occur in regions wherethe refractory periods of adjacent cells areheterogeneous, such that some cells recoverbefore others. In addition, unidirectionalblock may occur in states of cellular dys-function, and in regions where fibrosis hasaltered the myocardial structure.

As shown in panel C of Figure 11.9, if theimpulse is able to propagate retrogradelyup the β pathway, it will again arrive atpoint x. At that time, if the β pathway hasnot yet repolarized from the previous ac-tion potential that had occurred momentsearlier, that limb is refractory to repeatstimulation and the returning impulse sim-ply stops there.

However, panel D illustrates what hap-pens if the velocity of retrograde conduc-tion in the diseased β path is not normalbut slower than normal. In that case, suffi-cient time may elapse for the α pathway to repolarize before the returning impulsereaches point x from the β limb. Then the impulse is free to stimulate the α path-way once again, and the cycle repeats itself.This circular stimulation can continue in-definitely, and each pass of the impulsethrough the loop excites cells of the distalconduction tissue, which propagates to therest of the myocardium, resulting in a tachy-arrhythmia.

For the mechanism of reentry to occur,the propagating impulse must continu-ously encounter excitable tissue. Thus, thetime it takes for the impulse to travelaround the reentrant loop must be greaterthan the time required for recovery (the re-fractory period) of the tissue, and this mustbe true for each point in the circuit. If theconduction time is shorter than the recov-ery time, the impulse will encounter refrac-tory tissue and stop. Because normal con-duction velocity is approximately 50 cm/secand the average effective refractory periodis about 0.2 sec, a reentry path circuitwould need to be at least 10 cm long forreentry to occur in a normal ventricle. How-

ever, with slower conduction velocities, asmaller reentry circuit is possible. Mostclinical cases of reentry occur within smallregions of tissue because the conductionvelocity within the reentrant loop is, infact, abnormally slow.

In summary, the two critical conditionsfor reentry are (1) unidirectional block and(2) slowed conduction through the reentrypath. These conditions commonly occur inregions where fibrosis has developed, suchas infarction scars. In some cases, reentry oc-curs over an anatomically fixed circuit orpath, such as AV reentry using an accessorypathway (as discussed in the following sec-tion). Reentry around distinct anatomicpathways usually appears as a monomorphictachycardia on the electrocardiogram (ECG;i.e., in the case of ventricular tachycardia, allthe QRS complexes have the same appear-ance). This is because the reentry path is thesame from beat to beat, producing a stable,regular tachycardia.

Other types of reentry do not require astable, fixed path. For example, reentry canoccur in electrophysiologically heteroge-neous myocardium, in which waves of re-entrant excitation spiral through the tissue,continually changing direction. Ischemicmyocardium provides such a setting becausethe affected tissue is a mosaic of unexcitableand partially excitable zones with reducedconduction velocities. When reentrant ven-tricular tachycardia develops in an area ofischemia, the reentry circuit incessantlychanges and the QRS complexes typicallyvary from beat to beat, causing a polymor-phic ventricular tachycardia pattern on theECG. As described in Chapter 12, fibrillationof the atria or ventricles is likely caused bymultiple circulating reentry wave fronts.

Accessory Pathways and the Wolff-Parkinson-White Syndrome

The concept of reentry is dramatically illus-tratedbytheWolff-Parkinson-White (WPW)syndrome. In the normal heart, the impulsegenerated by the SA node propagates through

280 Chapter Eleven

1 LINE SHORT

10090-11_CH11.qxd 8/31/06 4:54 PM Page 280


atrial tissue to the AV node, where expectedslower conduction causes a short delay be-fore continuing on to the ventricles. How-ever, approximately 1 in 1,000 people hasthe WPW syndrome and is born with an ad-ditional connection between the atrium andventricle. Termed an accessory pathway orbypass tract, this connection allows con-duction between the atria and ventricles tobypass the AV node. The most commontype of accessory pathway consists of mi-croscopic fibers that span the atrioventricu-lar groove somewhere along the mitral ortricuspid annuli (known as a bundle of Kent),as shown in Figure 11.10.

Because accessory pathway tissue con-ducts impulses faster than the AV node,stimulation of the ventricles during sinusrhythm begins earlier than normal, and thePR interval of the ECG is therefore shortened(usually <0.12 sec, or <3 small boxes). Inthis situation, the ventricles are said to be“preexcited.” However, the accessory path-way connects to ventricular myocardiumrather than to the Purkinje system, suchthat the subsequent spread of the impulsethrough the ventricles from that site isslower than usual. In addition, because nor-

mal conduction over the AV node proceedsconcurrently, ventricular depolarizationrepresents a combination of the electricalimpulse traveling via the accessory tractand that conducted through the normalconduction pathway. As a result, the QRScomplex in patients with WPW is widerthan normal and demonstrates an abnor-mally slurred initial upstroke, known as adelta wave (Fig. 11.10).

During sinus rhythm, simultaneous con-duction through the accessory pathway andAV node creates an interesting ECG appear-ance but causes no symptoms. The presenceof the abnormal pathway, however, createsan ideal condition for reentry because therefractory period of the pathway is usuallydifferent from that of the AV node. An ap-propriately timed abnormal impulse (e.g., apremature beat) may encounter blockage inthe accessory pathway but conduction overthe AV node, or vice versa. If the propagat-ing impulse then finds that the initiallyblocked pathway has recovered (unidirec-tional block), it can conduct in a retrogradedirection up to the atrium and then downthe other pathway back to the ventricles.Thus, a large anatomic loop is established,

AV node

Widened QRS

QRS

P T

Delta wave

SA node

Bypass tract

Shortened PR

Normal ECG

ECG with bypass tract

Figure 11.10. Accessory pathway (also termed the bypass tract). Example of an atrioventricu-lar bypass tract (bundle of Kent), shown schematically, which can conduct impulses from the atriumdirectly to the ventricles, bypassing the AV node. The ECG demonstrates a short PR interval and a“delta wave” caused by early excitation of the ventricles via the accessory pathway. ECG, electro-cardiogram; SA, sinoatrial.1 LINE SHORT

Fig. 10

10090-11_CH11.qxd 8/31/06 4:54 PM Page 281

with the accessory pathway serving as onelimb and the normal conduction pathwaythrough the AV node as the other. The clin-ical characteristics of the Wolff-Parkinson-White syndrome, including the types ofreentrant tachycardia associated with it, aredescribed in Chapter 12.

The mechanisms of altered impulse for-mation and conduction form the basis of allcommon arrhythmias, both abnormally slowrhythms (bradyarrhythmias) and abnormallyfast ones (tachyarrhythmias). Table 11.1lists the underlying mechanisms and exam-ples of their commonly associated rhythmdisturbances.

APPROACHES TOANTIARRHYTHMIC TREATMENT

The treatment of a rhythm disorder de-pends on its severity and likely mechanism.When an arrhythmia produces severe hy-

potension or cardiac arrest, it must be im-mediately terminated to restore effectivecardiac function. Therapy for terminationmay include electrical cardioversion (anelectrical “shock”) for tachycardias, cardiacpacing for bradycardias, or administrationof medications.

Additional therapy to prevent recurrencesis guided by the etiology of the rhythm dis-turbance. Correctable factors that contributeto abnormal impulse formation and con-duction (such as ischemia or electrolyte ab-normalities) should be corrected. If there isa risk of recurrent arrhythmia, medicationsthat alter automaticity, conduction and/orrefractoriness may be administered. Some-times, catheter or surgical ablation of con-duction pathways is undertaken to physi-cally disrupt the region responsible for thearrhythmia. Other advanced options in-clude implantation of a permanent pace-maker for serious bradyarrhythmias or aninternal ICD to automatically terminate ma-

282 Chapter Eleven

TABLE 11.1. Mechanisms of Arrhythmia Development

Abnormality Mechanism Examples

BradyarrhythmiasAltered impulse formation• Decreased automaticity

Altered impulse conduction• Conduction blocks

TachyarrhythmiasAltered impulse formation• Enhanced automaticity

Sinus node

Ectopic focus• Triggered activity

Early afterdepolarizationDelayed afterdepolarization

Altered impulse conduction• Reentry

Anatomical

Functional

AV, atrioventricular; APB, atrial premature beat; VPB, ventricular premature beat.

Decreased phase 4 depolarization(e.g., parasympathetic stimulation)

Ischemic, anatomic, or drug-induced impaired conduction

Increased phase 4 depolarization(e.g., sympathetic stimulation)

Acquires phase 4 depolarization

Prolonged action potential durationIntracellular calcium overload

(e.g., digitalis toxicity)

Unidirectional block plus slowedconduction

Sinus bradycardia

First-, second-, and third-degree AVblocks

Sinus tachycardia

Ectopic atrial tachycardia

Torsades de pointesAPBs, VPBs, digitalis-induced

arrhythmias

Atrial flutter, AV nodal reentranttachycardia

Atrial fibrillation, ventricular fibrillation

10090-11_CH11.qxd 8/31/06 4:54 PM Page 282


lignant tachyarrhythmias should they recur.The following sections summarize the com-mon therapeutic modalities, and Chapter 12describes how they are used to address spe-cific rhythm disorders.

Bradyarrhythmias

Not all slow heart rhythms require specifictreatment. For those that do, pharmacologictherapy can increase the heart rate acutely,but their effect is transient. Electronic pace-makers are used when a more sustained ac-tion is needed.

Pharmacologic Therapy

Pharmacologic therapy of bradyarrhythmiasmodifies the autonomic input to the heartin one of two ways:

1. Anticholinergic drugs (i.e., antimuscarinicagents such as atropine). Vagal stimula-tion reduces the rate of sinus node depo-larization (which slows the heart rate)and decreases conduction through theAV node, through the release of acetyl-choline onto muscarinic receptors. Anti-cholinergic drugs competitively bind tomuscarinic receptors and thereby reducethe vagal effect. This results in an in-creased heart rate and enhanced AV nodalconduction.

2. β1-receptor agonists (e.g., isoproterenol).Mimicking the effect of endogenous cat-echolamines, these drugs increase heartrate and speed AV nodal conduction.

Atropine and isoproterenol are adminis-tered intravenously. Although these drugsare useful in treating certain slow heartrhythms emergently, it is not practical tocontinue them over the long term for per-sistent bradyarrhythmias.

Electronic Pacemakers

Electronic pacemakers apply repeated elec-trical stimulation to the heart to initiate de-polarizations at a desired rate, thereby as-suming control of the rhythm. Pacemakersmay be installed on a temporary or a per-

manent basis. Temporary units are used tostabilize patients who are awaiting implan-tation of a permanent pacemaker or to treattransient bradyarrhythmias, such as thosecaused by reversible drug toxicities.

There are two types of temporary pace-makers. External transthoracic pacemakersdeliver electrical pulses to the patient’s chestthrough large adhesive electrodes placed onthe skin. The advantage of this technique isthat it can be applied rapidly. Unfortu-nately, because the current used must besufficient to initiate a cardiac depolariza-tion, it stimulates thoracic nerves and skele-tal muscle, which can be quite uncomfort-able. Therefore, this form of pacing isusually used only on an emergency basisuntil another means of treating the ar-rhythmia can be implemented.

The other option for temporary pacing isa transvenous unit. In this case, an electrode-tipped catheter is inserted percutaneouslyinto the venous system, passed into theheart, and connected to an external powersource (termed a pulse generator). Electricalpulses are applied directly to the heartthrough the electrode catheter, which is typ-ically placed in the right ventricle or rightatrium. This type of pacing is not painfuland can be effective for days. There is, how-ever, a risk of infection and/or thrombosisthat increases with time.

Permanent pacemakers are more sophis-ticated than the temporary variety. Variousconfigurations can sense and capture theelectrical activity of the atria and/or ventri-cles. One or more wires (known as leads)with pacing electrodes are passed through anaxillary or subclavian vein into the right ven-tricle or right atrium or through the coro-nary sinus into a cardiac vein (to stimulatethe left ventricle). The pulse generator, simi-lar in size to two silver dollars stacked on topof one another, is connected to the leads andthen implanted under the skin, typically inthe infraclavicular region. The pacemakerbattery typically lasts about 10 years.

Modern permanent pacemakers sensecardiac activity and pace only when needed.They incorporate complex functions totrack the patient’s normal heart rate and can

10090-11_CH11.qxd 8/31/06 4:54 PM Page 283

stimulate beats automatically in response toactivity. They can also record useful data,such as whether fast rates have been sensed(that might indicate a tachyarrhythmia), theamount of pacing that has been required,and other parameters of pacemaker function.An external radio frequency programmingdevice is used to “interrogate” the pacemakerto obtain the recorded information and ad-just the pacing functions.

Although the most common indicationsfor permanent pacemakers are bradyarrhyth-mias, pacemakers that incorporate a left ven-tricular pacing lead are also used to improvecardiac performance in some patients withheart failure (see Chapter 9).

Tachyarrhythmias

The treatment of tachyarrhythmias is di-rected at (1) protection of the patient fromthe consequences of the arrhythmia and (2) the specific mechanism responsible forthe abnormal rhythm. Pharmacologic agentsand cardioversion/defibrillation are com-monly used approaches, but innovativeelectrical devices and transvenous catheter-based techniques have revolutionized treat-ment of these disorders.

Pharmacologic Therapy

Pharmacologic management of tachyarr-hythmias is directed against the underlyingmechanism (abnormal automaticity, reen-trant circuits, or triggered activity). Many an-tiarrhythmic drugs are available, the phar-macology and actions of which are addressedin Chapter 17. The choice of drug relies onthe cause of the specific arrhythmia. Fromconsideration of the arrhythmia mechanismspresented in this chapter, the following strate-gies emerge.

Desired Drug Effects to Eliminate RhythmsCaused by Increased Automaticity

1. Reduce the slope of phase 4 spontaneousdepolarization of the automatic cells

2. Make the diastolic potential more nega-tive (hyperpolarize)

3. Make the threshold potential less nega-tive

Desired Antiarrhythmic Effects to InterruptReentrant Circuits

1. Decrease conduction in the reentry cir-cuit to the point that conduction fails,thus stopping the reentry impulse

2. Increase the refractory period within thereentrant circuit so that a propagatingimpulse finds tissue within the loop un-excitable and the impulse stops

3. Suppress premature beats that can initi-ate reentry

Desired Drug Effects to Eliminate Triggered Activity

1. Shorten the action potential duration (toprevent early afterdepolarizations)

2. Correct conditions of calcium overload(to prevent delayed afterdepolarizations)

Drugs used to achieve the goals modulatethe action potential through interactionswith ion channels, surface receptors, andtransport pumps. Many drugs have multipleeffects and may attack arrhythmias throughmore than one mechanism. The commonlyused antiarrhythmic drugs and their actionsare described in Chapter 17.

It is extremely important to recognizethat in addition to suppressing arrhythmias,these drugs have the potential to aggravateor provoke certain rhythm disturbances. Thisundesired consequence is referred to as pro-arrhythmia and is a major limitation ofcontemporary antiarrhythmic drug therapy.For example, antiarrhythmic agents that acttherapeutically to prolong the action poten-tial duration can, as an undesired effect,cause early afterdepolarizations, the mecha-nism underlying the polymorphic ventriculartachycardia known as torsades de pointes (seeChapter 12). In addition, most agents used totreat tachyarrhythmias have the potential toaggravate bradyarrhythmias, and all antiar-rhythmics have potentially toxic noncardiacside effects. These shortcomings have led toan increased reliance on nonpharmacologictreatment options, as described in the follow-ing sections.

284 Chapter Eleven

10090-11_CH11.qxd 8/31/06 4:54 PM Page 284


Vagotonic Maneuvers

Many tachycardias involve transmission ofimpulses through the AV node, a structurethat is sensitive to vagal modulation. Vagaltone can be transiently increased by a num-ber of bedside maneuvers, and performingthese may slow conduction, which termi-nates some reentrant tachyarrhythmias. Forexample, carotid sinus massage is performedby rubbing firmly for a few seconds over thecarotid sinus, located at the bifurcation of theinternal and external carotid arteries on eitherside of the neck. This maneuver stimulatesthe baroreceptor reflex (see Chapter 13),which elicits the desired increase in vagaltone and withdrawal of sympathetic tone.This maneuver should be performed on onlyone carotid sinus at a time (to prevent inter-ference with brain perfusion) and is bestavoided in patients with known atherosclero-sis involving the carotid arteries.

Electrical Cardioversion and Defibrillation

Cardioversion and defibrillation involve theapplication of an electric shock to terminatea tachycardia. A shock with sufficient en-ergy depolarizes the bulk of excitable myo-cardial tissue, interrupts reentrant circuits,establishes electrical homogeneity, and al-lows the sinus node (the site of fastest spon-taneous discharge) to regain pacemaker con-trol. Tachyarrhythmias that are caused byreentry can usually be terminated by thisprocedure, whereas arrhythmias owing toabnormal automaticity may simply persist.

External cardioversion is used to termi-nate supraventricular tachycardias or orga-nized ventricular tachycardias. It is per-formed by briefly sedating the patient andthen placing two large electrode paddles (oradhesive electrodes) against the chest oneither side of the heart. The electrical dis-charge is electronically synchronized to occurat the time of a QRS complex (i.e., whenventricular depolarization occurs). This pre-vents the possibility of discharge during therelative refractory period of the ventricle(see Fig. 1.16), which could induce ventric-ular fibrillation.

External defibrillation is performed toterminate ventricular fibrillation, employingthe same equipment as that used for cardio-version. However, during fibrillation, there isno organized QRS complex on which to syn-chronize the electrical discharge, so it is de-livered using the “asynchronous” mode ofthe device.

Implantable Cardioverter-Defibrillators

ICDs automatically terminate dangerousventricular arrhythmias using internal car-dioversion or defibrillation, or a techniqueknown as antitachycardia pacing. These de-vices are implanted, in a manner similar tothat of permanent pacemakers, in patientsat high risk of sudden cardiac death fromventricular arrhythmias. The device contin-uously monitors cardiac activity, and if theheart rate exceeds a certain programmablethreshold for a specified time (e.g., >12beats), the ICD delivers an appropriate in-tervention, such as an electrical shock. In-ternal cardioversion or defibrillation requiressubstantially less energy than external defib-rillation but is still uncomfortable if the pa-tient is conscious.

In addition, many monomorphic ven-tricular tachycardias can be terminated byan ICD with a rapid burst of electrical im-pulses, termed antitachycardia pacing (ATP).The goal is to artificially pace the heart at arate faster than the tachycardia to prema-turely depolarize a portion of a reentrant cir-cuit, thereby rendering it refractory to fur-ther immediate stimulation. Consequently,when a reentrant impulse returns to thezone that has already been depolarized bythe device, it encounters unexcitable tissue,it cannot propagate further, and the circuitis broken. An advantage of ATP is that, un-like internal cardioversion, it is painless.However, ATP is not effective for terminat-ing ventricular fibrillation.

Catheter Ablation

If an arrhythmia originates from distinctanatomical reentry circuits or automatic foci,electrophysiologic mapping techniques can

10090-11_CH11.qxd 8/31/06 4:54 PM Page 285

be used to localize the region of myocardiumor conduction tissue responsible for the dis-turbance. It is then often possible to ablatethe site via a catheter that applies radio-frequency current to heat and destroy the tis-sue. These procedures have revolutionized themanagement of patients with many types ofsupraventricular tachycardias, because theyoften offer a permanent therapeutic solutionthat spares patients from undergoing chronicantiarrhythmic drug therapy. For patientswith recurrent ventricular tachycardias caus-ing defibrillator shocks, ablation is often ef-fective in reducing the frequency of episodes.

SUMMARY

1. Arrhythmias result from disorders of im-pulse formation, impulse conduction, orboth.

2. Bradyarrhythmias develop because of de-creased impulse formation (e.g., sinus bra-dycardia) or decreased impulse conduc-tion (e.g., AV nodal conduction blocks).

3. Tachyarrhythmias result from increasedautomaticity (of the SA node, latent pace-makers, or abnormal myocardial sites),triggered activity, or reentry.

4. Bradyarrhythmias are usually treated withdrugs that accelerate the rate of sinusnode discharge and enhance AV nodalconduction (atropine, isoproterenol) orelectronic pacemakers.

5. Pharmacologic therapy for tachyarrhyth-mias is directed at the mechanism re-sponsible for the rhythm disturbance.For refractory tachyarrhythmias, or inemergency situations, electrical cardio-version or defibrillation is used. Catheter-based ablative techniques are useful forlong-term control of certain tachy-arrhythmias. ICDs are lifesaving devicesimplanted in patients at high risk of sud-den cardiac death because of ventriculartachyarrhythmias.

Chapter 12 summarizes the diagnosis andmanagement of the most common arrhyth-mias. Chapter 17 describes currently avail-able antiarrhythmic drugs.

Acknowledgment

Contributors to the previous editions of this chapterwere Wendy Armstrong, MD; Nicholas Boulis, MD;Jennifer E. Ho, MD; Mark S. Sabatine, MD; Elliott M.Antman, MD; Leonard I. Ganz, MD; Gary R. Strichartz,PhD; and Leonard S. Lilly, MD.

286 Chapter Eleven

AQ2

10090-11_CH11.qxd 8/31/06 4:54 PM Page 286

Chapter 11—Author Queries1. AU: Correct to add?2. AU: Correct meaning of VT acronym?3. AU: Correct to add? See also below.

10090-11_CH11.qxd 8/31/06 4:54 PM Page a

287

BRADYARRHYTHMIASSinoatrial NodeEscape RhythmsAtrioventricular Conduction System

TACHYARRHYTHMIASSupraventricular ArrhythmiasVentricular Arrhythmias

C H A P T E R

12Clinical Aspects ofCardiac ArrhythmiasHillary K. RollsWilliam G. StevensonLeonard S. Lilly

Arrhythmias appear frequently in patientswith cardiac disease and can occur inhealthy people as well. Chapter 11 presentedthe mechanisms by which abnormal heartrhythms develop. This chapter describes howto recognize and treat them. Table 12.1 cate-gorizes the common disorders considered inthis chapter.

There are five basic questions to considerwhen confronted with a patient with an ab-normal heart rhythm, as detailed in the sec-tions that follow:

1. Definition: What is the arrhythmia?

2. Pathogenesis: What is the underlyingmechanism?

3. Precipitating factors: What conditionsprovoke it?

4. Clinical presentation: What symptomsand signs accompany the arrhythmia?

5. Treatment: What to do about it?

BRADYARRHYTHMIAS

The normal resting heart rate, resulting fromrepetitive depolarization of the sinus node,ranges from 60 to 100 bpm. Bradyarrhyth-mias are rhythms in which the heart rate is<60 bpm. They arise from disorders of im-pulse formation or impaired impulse con-duction, as discussed in Chapter 11.

Sinoatrial Node

Sinus Bradycardia

Sinus bradycardia (Fig. 12.1) is simply a slow-ing of the normal heart rhythm, as a result ofdecreased firing of the sinoatrial (SA) node,to a rate <60 bpm. Sinus bradycardia at restor during sleep is normal and a benign find-ing in many people. It is therefore incum-bent on the physician to decide whether thisrhythm is appropriate or pathologic in a par-ticular patient, and whether treatment is re-

Tab. 1

Fig. 1

10090-12_CH12.qxd 8/31/06 4:55 PM Page 287

quired. This decision can be made on thebasis of a patient’s age, underlying heart dis-ease, level of physical activity, symptoms,and whether the heart rate increases appro-priately with exercise.

Sinus bradycardia can result from eitherintrinsic SA node disease or extrinsic factorsthat affect the node. Depressed intrinsic au-tomaticity can be caused by aging or any dis-ease process that affects the atrium, includingischemic heart disease or cardiomyopathy.Extrinsic factors that suppress SA nodal ac-tivity include medications (e.g., many anti-arrhythmic drugs, including β-blockers andcertain calcium channel blockers) and meta-bolic causes (e.g., hypothyroidism). Highlytrained athletes often have elevated vagaltone, which results in physiologic (i.e., ap-propriate) resting sinus bradycardia. How-ever, transient periods of high vagal tonecan also occur as a reflex response to pain or fear resulting in inappropriate sinus brady-cardia.

Mild sinus bradycardia is usually asymp-tomatic and does not require treatment.However, a pronounced reduction of theheart rate can produce a fall in cardiac out-put with fatigue, light-headedness, confu-sion, or syncope. In such cases, any extrin-sic provocative factors should be corrected,and specific therapy, as described in the nextsection, may be needed.

Sick Sinus Syndrome

Intrinsic SA node dysfunction that causes periods of inappropriate bradycardia isknown as sick sinus syndrome (SSS). Thiscondition often produces symptoms of dizzi-ness, confusion, or syncope. Patients withthis syndrome (or any cause of symptomaticsinus bradycardia) can be treated with intra-venous anticholinergic drugs (e.g., atropine)or β-adrenergic agents (e.g., isoproterenol),which transiently accelerate the heart rate. Ifthe problem is chronic and not corrected by

288 Chapter Twelve

TABLE 12.1. Common Arrhythmias

Location Bradyarrhythmias Tachyarrhythmias

SA node Sinus bradycardia Sinus tachycardiaSick sinus syndrome

Atria Atrial premature beatsAtrial flutterAtrial fibrillationParoxysmal supraventricular tachycardiasEctopic atrial tachycardiaMultifocal atrial tachycardia

AV node Conduction blocks Paroxysmal reentrant tachycardias (AV or AV nodal)Junctional escape rhythm

Ventricles Ventricular escape rhythm Ventricular premature beatsVentricular tachycardiaTorsades de pointesVentricular fibrillation

AV, atrioventricular; SA, sinoatrial.

Figure 12.1. Sinus bradycardia. The P wave and QRS complexes are normal but the rate is <60 bpm.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 288

Clinical Aspects of Cardiac Arrhythmias 289

removal of aggravating factors, placement ofa permanent pacemaker is required.

SSS is common in elderly patients, whoare also susceptible to supraventriculartachycardias, most commonly, atrial fibril-lation. This combination is known as thebradycardia-tachycardia syndrome (Fig. 12.2)and is thought to result from atrial fibrosisthat impairs function of the SA node andpredisposes to atrial fibrillation and flutter.During the tachyarrhythmia, overdrive sup-pression of the SA node occurs, and whenthe tachycardia terminates, profound peri-ods of sinus bradycardia ensue. Treatmentgenerally requires the combination of an-tiarrhythmic drug therapy to suppress thetachyarrhythmias plus a permanent pace-maker to prevent bradycardia.

Escape Rhythms

Cells in the atrioventricular (AV) node andHis-Purkinje system are capable of automatic-

ity but typically have slower firing rates thanthe sinus node and are therefore suppressedduring normal sinus rhythm. However, if SAnode activity becomes impaired or if there isconduction block of the impulse from the SAnode, escape rhythms can emerge from themore distal latent pacemakers (Fig. 12.3).

Junctional escape beats arise from the AVnode or proximal bundle of His. They arecharacterized by a normal, narrow QRS com-plex, and when they occur in sequence(termed a junctional escape rhythm), ap-pear at a rate of 40 to 60 bpm. The QRS com-plexes are not preceded by normal P wavesbecause the impulse originates below theatria. However, retrograde P waves may be ob-served as an impulse propagates from themore distal pacemaker backward to the at-rium. Retrograde P waves typically follow theQRS complex and are inverted (negative de-flection on the electrocardiogram [ECG]) inlimb leads II, III, and aVF, indicating activa-tion of the atria from the inferior direction.

Figure 12.2. Bradycardia-tachycardia syndrome. A brief irregular tachycardia is followed by slowsinus node discharge.

Figure 12.3. Escape rhythms. No P waves are evident. A. Junctional escape rhythm with normal-widthQRS complexes. B. Wide QRS complexes typical of a ventricular escape rhythm.

Fig. 2 Fig. 3

10090-12_CH12.qxd 8/31/06 4:55 PM Page 289

Ventricular escape rhythms are charac-terized by even slower rates (30 to 40 bpm)and widened QRS complexes. The complexesare wide because the ventricles are not de-polarized by the normal rapid simultaneousconduction over the right and left bundlebranches but rather from a more distal pointin the conduction system. The morphologythat the QRS shows depends on the site oforigin of the escape rhythm. For example,an escape rhythm originating from the leftbundle branch will cause a right bundlebranch block QRS pattern, because the im-pulse depolarizes the left ventricle first andthen spreads more slowly through the rightventricle. Conversely, an escape rhythm orig-inating in the right bundle branch causesthe QRS to appear with a left bundle branchblock configuration. Escape rhythms thatoriginate more distally, in the ventricularmyocardium itself, are characterized by evenwider QRS complexes because such impulsesare conducted outside the rapidly propagat-ing Purkinje fibers.

Junctional and ventricular escape rhythmsare protective backup mechanisms thatmaintain a heart rate and cardiac outputwhen the sinus node or normal AV conduc-tion fail. The clinical findings and treat-ment of bradycardia associated with escaperhythms are identical to those of SSS de-scribed earlier.

Atrioventricular Conduction System

The AV conduction system includes the AVnode, bundle of His, and the left and rightbundle branches. Impaired conduction be-tween the atria and ventricles can cause threedegrees (types) of AV conduction block.

First-Degree AV Block

First-degree AV block, shown in Figure 12.4,indicates prolongation of the normal delaybetween atrial and ventricular depolariza-tion, such that the PR interval is lengthened(>0.2 sec, which is >5 small boxes on theECG). However, the 1:1 relationship betweenP waves and QRS complexes is preserved. Theimpairment of conduction is usually withinthe AV node itself and can be caused by atransient influence or a structural defect. Re-versible causes include heightened vagal tone,transient AV nodal ischemia, and drugs thatdepress conduction through the AV node, in-cluding digitalis glycosides, β-blockers, cer-tain calcium channel antagonists, and otherantiarrhythmic medications. Structural causesof first-degree AV block include myocardialinfarction and chronic degenerative diseasesof the conduction system, which commonlyoccur with aging.

Generally, first-degree AV block is a be-nign, asymptomatic condition that does notrequire treatment. However, it can indicatedisease in the AV node associated with sus-ceptibility to higher degrees of AV block ifdrugs are administered that further impairAV conduction, or if the conduction diseaseprogresses.

Second-Degree AV Block

Second-degree AV block is characterized byintermittent failure of AV conduction, result-ing in some P waves not followed by a QRScomplex (i.e., intermittent failure of 1:1 atri-oventricular conduction). There are twoforms of second-degree AV block. In Möbitztype I block (also termed Wenckebachblock), shown in Figure 12.5, the degree ofAV delay gradually increases with each beat

290 Chapter Twelve

P P P P P

Figure 12.4. First-degree AV block. The PR interval is prolonged.

Fig. 4

Fig. 5

10090-12_CH12.qxd 8/31/06 4:55 PM Page 290


until an impulse is completely blocked andventricular stimulation does not follow a Pwave for a single beat. Therefore, the ECGshows a progressive increase in the PR inter-val from one beat to the next until a singleQRS complex is absent, after which the PRinterval shortens to its initial length, andthe cycle starts anew. Möbitz type I block al-most always results from impaired conduc-tion in the AV node (rather than more dis-tally in the conduction system). It is usuallybenign and may be seen in children, trainedathletes, and people with high vagal tone,particularly during sleep. It may also occurduring an acute myocardial infarction be-cause of increased vagal tone or ischemia ofthe AV node, but the block is usually tran-sient. Treatment is typically not necessary,but in symptomatic cases, administration ofintravenous atropine or isoproterenol usu-ally improves AV conduction transiently.Placement of a permanent pacemaker is re-quired for a symptomatic block that doesnot resolve spontaneously or persists despitethe correction of aggravating factors.

Möbitz type II second-degree AV block ischaracterized by the sudden intermittent lossof AV conduction, without preceding graduallengthening of the PR interval (Fig. 12.6).The block may persist for two or more beats(i.e., two sequential P waves not followed by

QRS complexes), in which case it is known ashigh-grade AV block (Fig. 12.7). Möbitz typeII block is usually caused by conductionblock beyond the AV node (in the bundle ofHis or more distally in the Purkinje system),and the QRS pattern usually is widened in apattern of right or left bundle branch block.This type of block may arise from extensivemyocardial infarction involving the septumor from chronic degeneration of the His-Purkinje system. It usually indicates severedisease and is more dangerous than Möbitztype I block. Möbitz type II may progress tothird-degree block without warning; there-fore, treatment with a pacemaker is usuallywarranted, even in asymptomatic patients.

Third-Degree AV Block

Third-degree AV block, also termed com-plete heart block (Fig. 12.8), is present whenthere is complete failure of conduction be-tween the atria and ventricles. In adults, themost common causes are acute myocardialinfarction, drug toxicity (especially digitalis),and chronic degeneration of the conductionpathways with age. Third-degree AV blockelectrically disconnects the atria and ventri-cles; there is no relationship between the Pwaves and QRS complexes because the atriadepolarize in response to SA node activity,

P P P P P P

Figure 12.5. Second-degree AV block: Möbitz type I (Wenckebach). The P wave rate is constant,but the PR interval progressively lengthens until a QRS is completely blocked (after fourth P wave).

P P P P P

Figure 12.6. Second-degree AV block: Möbitz type II. A QRS complex is blocked (after the fourthP wave) without gradual lengthening of the preceding PR intervals.

Fig. 6

Fig. 7

Fig. 8

10090-12_CH12.qxd 8/31/06 4:55 PM Page 291

while a more distal escape rhythm drives theventricles independently. Thus, the P waves“march out” at a rate that is not related to theintervals at which QRS complexes appear.Depending on the site of the escape rhythm,the QRS complexes may be of normal widthand occur at 40 to 60 bpm (originating fromthe AV node) or may be widened and occurat slower rates (originating from the His-Purkinje system). As a result of the slow rate,patients may experience light-headedness orsyncope. Permanent pacemaker therapy is al-most always necessary.

The term atrioventricular dissociationis a general description that refers to any sit-uation in which the atria and ventricles beatindependently, without any direct relation-ship between P waves and QRS complexes.Third-degree AV block is one example of AVdissociation.

TACHYARRHYTHMIAS

Whenever the heart rate is >100 bpm for 3 beats or more, a tachyarrhythmia is said tobe present. Tachyarrhythmias result fromone of three mechanisms: enhanced auto-maticity, reentry, or triggered activity (seeChapter 11). Tachyarrhythmias are catego-rized into those that arise above the ventri-cles (supraventricular) and those that arisewithin the ventricles and can usually be dif-

ferentiated by (1) the width of the QRS com-plex, (2) the morphology and rate of the P waves, (3) the relationship between the P waves and the QRS complexes, and (4) theresponse of the rhythm to vagal maneuverssuch as carotid sinus massage (Fig. 12.9).

Supraventricular Arrhythmias

Sinus Tachycardia

Sinus tachycardia is characterized by an SAnode discharge rate >100 bpm (typically 100to 180 bpm) with normal P waves and QRScomplexes (Fig. 12.10). This rhythm mostoften results from increased sympatheticand/or decreased vagal tone. Sinus tachycar-dia is an appropriate physiologic response toexercise. However, it may also result fromsympathetic stimulation in pathologic con-ditions, including fever, hypoxemia, hyper-thyroidism, hypovolemia, and anemia. Indisease states, sinus tachycardia is usually asign of the severity of the primary patho-physiologic process and treatment is directedat the underlying cause.

Atrial Premature Beats

Atrial premature beats (APBs) are common inhealthy as well as diseased hearts (Fig. 12.11).They originate from automaticity or reentryin an atrial focus outside the SA node and are

292 Chapter Twelve

P P P P P P

Figure 12.7. High-grade AV block. Sequential QRS complexes are blocked (after the second andthird P waves).

P P PP PP PP

Figure 12.8. Third-degree AV block. The P wave and QRS rhythms are independent of one another.The QRS complexes are widened as they originate within the distal ventricular conduction system, notat the bundle of His. The second and fourth P waves are superimposed on normal T waves. 1 LINE LONG

Fig. 9

Fig. 10

Fig. 11

10090-12_CH12.qxd 8/31/06 4:55 PM Page 292


Sustainedtachycardia

Normal (narrow) QRS

Regularrhythm

(constantP-P intervals)

Irregularlyirregularrhythm

Nodistinct

P waves

Norelationship

betweenQRS andP waves

Constantrelationship

betweenQRS andP waves,

andP is upright

in II, III,aVF

Atrialfibrillation

At r ial r ate(bpm)

Responsetocarotidsi n us massage

100-180

At r ial r atem a y sl o w

140-250

Ma y ab r uptlyr e v e r t tono r mal

130-250

AV b lo c km a y increase ;doesn ’t usually

r e v e r t

180-350

AV b lo c km a y

increase

Multifocalatrial

tachycardia

Ventriculartachycardia

Supraventriculartachycardias

with “aberrant ”conduction

WideQRS

Si n usta c h yca r dia

ReentrantSVT

(A V or A V nodal)

Ectopicatrialta c h yca r dia

Atrialflutter

P w a v e mo r phology No r mal V a r ia b le V a r ia b le “S a w-toothed ”shaped

Figure 12.9. Differentiation of the common tachyarrhythmias.

Figure 12.10. Sinus tachycardia. The P wave and QRS complexes are normal, but the rate is >100 bpm.

often exacerbated by sympathetic stimula-tion. APBs are usually asymptomatic but maycause palpitations. On the ECG, an APB ap-pears as an earlier-than-expected P wave withan abnormal shape (the impulse does not arisefrom the SA node, indicating abnormal con-

duction through the atria). The QRS complexthat follows the P wave is usually normal, re-sembling the QRS during sinus rhythm, be-cause ventricular conduction is not impaired.However, if the abnormal atrial focus firesvery soon after the previous beat, the impulse

1 LINE LONG

10090-12_CH12.qxd 8/31/06 4:55 PM Page 293

may encounter an AV node that is refractoryto excitation, indicating a blocked impulsethat does not conduct to the ventricles. Thepremature P wave is then not followed by aQRS complex and is termed a blocked APB.Similarly, if the ectopic focus fires just a bitlater in diastole, it may conduct through theAV node but encounter portions of the His-Purkinje system that are still refractory. As aresult, the impulse is conducted throughthose territories and to the ventricular my-ocytes more slowly than normal, producingQRS complexes that are abnormally wide.

APBs require treatment only if they aresymptomatic. Because caffeine, alcohol,and adrenergic stimulation (e.g., emotionalstress) can all predispose to APBs, it is im-portant to remove these factors when possi-ble. β-Blockers are the initial preferred phar-macologic treatment if needed.

Atrial Flutter

Atrial flutter is characterized by rapid, regu-lar atrial activity at a rate of 180 to 350 bpm(Fig. 12.12). Many of these fast impulsesreach the AV node during its refractory pe-riod and do not conduct to the ventricles,

resulting in a slower ventricular rate, oftenan even fraction of the atrial rate. Thus, ifthe atrial rate is 300 bpm and 2:1 block oc-curs at the AV node (i.e., every other atrialimpulse finds the AV node refractory), theventricular rate is 150 bpm. Because vagalmaneuvers (e.g., carotid sinus massage) de-crease AV nodal conduction, they increasethe degree of block, temporarily slowing theventricular rate, which allows better visual-ization of the underlying atrial activity. Ingeneral, atrial flutter is caused by reentryover a large anatomically fixed circuit. Inthe common form of atrial flutter, this cir-cuit is the atrial tissue along the tricuspidvalve annulus: the circulating depolariza-tion wave propagates up the interatrial sep-tum, across the roof and down the free wallof the right atrium and finally along thefloor of the right atrium between the tricus-pid valve annulus and inferior vena cava.Because large parts of the atrium are depo-larized throughout the cycle, P waves oftenhave a sinusoidal or “sawtooth” appearance.Large flutter circuits can occur in other partsof the right or left atrium as well, usually as-sociated with areas of atrial scarring fromdisease, prior heart surgery, or ablation pro-

294 Chapter Twelve

APB

Figure 12.11. Atrial premature beat (APB). The P wave occurs earlier than expected, and its shapeis abnormal.

Figure 12.12. Atrial flutter is typified by rapid “saw-toothed” atrial activity (arrows).

Fig. 12

10090-12_CH12.qxd 8/31/06 4:55 PM Page 294


cedures (e.g., those performed for treatmentof atrial fibrillation).

Atrial flutter generally occurs in patientswith preexisting heart disease. It may beparoxysmal and transient, persistent (lastingfor days or weeks), or permanent. Symptomsof atrial flutter depend on the accompanyingventricular rate. If the rate is <100 bpm, thepatient may be asymptomatic. Conversely,faster rates often cause palpitations, dyspnea,or weakness. Paradoxically, antiarrhythmicmedications that reduce the rate of atrial flut-ter by slowing conduction in the atrium mayparadoxically make the rhythm more dan-gerous by allowing the AV node more time torecover between impulses. In this situation,the AV node may begin to conduct in a 1:1fashion, producing very rapid ventricularrates. For example, a patient with atrial flut-ter at a rate of 280 bpm and 2:1 conductionblock at the AV node would have a ventricu-lar rate of 140 bpm. If the atrial rate thenslows to 220 bpm, the AV node may be ableto recover sufficiently between depolariza-tions to conduct every atrial impulse (i.e., 1:1conduction), causing the ventricular rate toaccelerate to 220 bpm. In patients with lim-ited cardiac reserve, this acceleration mayresult in a profound reduction of cardiac out-put and hypotension. Atrial flutter is associ-ated with a risk of atrial thromboembolism,as discussed later in the chapter.

Several approaches to the treatment ofatrial flutter are available:

1. For symptomatic patients with recent-onset atrial flutter, the most expeditioustherapy is electrical cardioversion to re-store sinus rhythm. This technique is alsoused to revert chronic atrial flutter that hasnot responded to other approaches.

2. Flutter can be terminated by rapid atrialstimulation (burst pacing) using a tem-porary or permanent pacemaker (seeChapter 11). This procedure can be usedwhen temporary atrial pacing wires arealready present, as in the period after car-diac surgery. In addition, certain types ofpermanent pacemakers and implanteddefibrillators can be programmed to per-form burst pacing automatically whenatrial flutter occurs.

3. Patients without an immediate need forcardioversion can begin pharmacologictherapy. First, the ventricular rate is slowedby drugs that increase AV block: β-block-ers, calcium channel blockers (e.g., vera-pamil, diltiazem), or digoxin. Once therate is effectively slowed, attempts aremade to restore sinus rhythm using an-tiarrhythmic drugs that slow conductionor prolong the refractory period of theatrial myocardium (class IA, IC, or IIIagents; see Chapter 17). Should thesedrugs fail to convert the rhythm, electiveelectrical cardioversion can then be un-dertaken. Once sinus rhythm has been re-stored, class IA, IC, or III antiarrhythmicdrugs may be administered chronically toprevent recurrences.

4. When chronic therapy is required to pre-vent recurrences, catheter ablation isoften a better alternative than pharma-cologic approaches. In this method, anelectrode catheter is inserted into thefemoral vein, passed via the inferior venacava to the right atrium, and used to lo-calize and cauterize (ablate) part of thereentrant loop to permanently interruptthe flutter circuit.

Atrial Fibrillation

Atrial fibrillation (AF) is a chaotic rhythmwith an atrial rate so fast (350 to 600 bpm)that discrete P waves are not discernible onthe ECG (Fig. 12.13). As with atrial flutter,many of the atrial impulses encounter re-fractory tissue at the AV node, causing onlysome of the depolarizations to be conductedto the ventricles in a very irregular fashion(indicated by the characteristic “irregularlyirregular” rhythm). The average ventricularrate in untreated AF is approximately 140 to160 bpm. Because discrete P waves are notvisible on the ECG, the baseline shows low-amplitude undulations punctuated by QRScomplexes and T waves.

The mechanism of AF probably involvesmultiple “wandering” reentrant circuitswithin the atria, and in some patients, therhythm repetitively shifts between fibrilla-tion and atrial flutter. When fibrillation is

AQ1

Fig. 13

10090-12_CH12.qxd 8/31/06 4:55 PM Page 295

paroxysmal (i.e., sudden, unpredictable epi-sodes), it is often initiated by rapid firing offoci in paths of atrial muscle that extendalong the pulmonary veins. For multiplereentry loops to be sustained in atrial fibril-lation, a minimum number of reentrant cir-cuits is needed, and an enlarged atrium in-creases the potential for this to occur. Thus,AF is often associated with right or left atrialenlargement. Accordingly, diseases that in-crease atrial pressure and size promote atrialfibrillation, including heart failure, hyper-tension, coronary artery disease, and pul-monary disease. Thyrotoxicosis and acute al-cohol consumption can precipitate AF insome people.

AF is a potentially dangerous arrhythmiafor two reasons: (1) rapid ventricular ratesmay compromise cardiac output, resultingin hypotension and pulmonary congestion(especially in patients with a hypertrophiedor “stiff” left ventricle in whom the loss ofnormal atrial contraction can significantlyreduce left ventricular filling and stroke vol-ume), and (2) the absence of organized atrialcontraction promotes blood stasis in theatria, which increases the risk of thrombusformation, particularly in the left atrial ap-pendage. Embolization of left atrial thrombiis an important cause of stroke. Treatmentof AF therefore focuses on three aspects ofthe arrhythmia: (1) ventricular rate control,(2) attempts to restore sinus rhythm, and (3) assessment of the need for anticoagula-tion to prevent thromboembolism.

Antiarrhythmic drug treatment of AF issimilar to that of atrial flutter. β-Blockers orcertain Ca++ channel antagonists (diltiazem,verapamil) are administered to promoteblock at the AV node and to reduce the ven-

tricular rate. Digitalis is less effective for thispurpose, although it may be useful in pa-tients with accompanying impairment ofventricular contractile function. For thosewho remain symptomatic despite adequaterate control, conversion to sinus rhythm isusually attempted. AF that persists for morethan 48 hours may indicate that intraatrialthrombus has formed, and systemic antico-agulation (for at least 3 weeks) is usuallywarranted prior to cardioversion to reducethe risk of thromboembolism. Alternatively,a transesophageal echocardiogram can beperformed to evaluate for thrombus; if noneis found, cardioversion can proceed sooner.

Cardioversion to sinus rhythm can be at-tempted chemically by administration of classIA, IC, or III antiarrhythmic drugs (see Chap-ter 17). Alternatively, electrical cardioversioncan be undertaken. Following successful con-version to sinus rhythm, antiarrhythmicdrugs are often continued to prevent recur-rences. Although antiarrhythmic drugs donot always maintain sinus rhythm duringlong-term follow-up, they can reduce thenumber of AF episodes. Note that these drugscan also cause serious, sometimes lethal,side effects (see Chapter 17). Thus, in pa-tients with asymptomatic AF, it is often saferto simply continue chronic anticoagulationtherapy and to control the ventricular ratewith β-blockers, calcium channel blockers,or digoxin.

Because the efficacies and toxicities of an-tiarrhythmic drugs have been disappointing,nonpharmacologic options for managementof AF have been devised. For example, thesurgical “maze procedure” places multipleincisions in the left and right atria to preventthe formation of reentry circuits. A less in-

296 Chapter Twelve

Figure 12.13. Atrial fibrillation is characterized by chaotic atrial activity without organized P waves andby irregularity of the ventricular (QRS) rate.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 296


vasive approach is percutaneous catheter ab-lation of regions around the pulmonaryveins in the left atrium. This procedure abol-ishes atrial fibrillation in some patients, pos-sibly by ablating the triggering atrial foci thatinitiate the fibrillation, as well as by damag-ing areas that may promote atrial reentry. It requires extensive catheter manipulationand ablation in the left atrium, and risks orthe procedure include systemic thrombo-embolism and cardiac perforation that cancause tamponade (see Chapter 14). Thus,catheter ablation for AF is usually reservedfor patients who remain quite symptomaticdespite pharmacologic approaches.

When sinus rhythm cannot be maintainedand the heart rate cannot be controlled ade-quately with medications, catheter ablationof the AV node is another procedure that canbe undertaken. This method intentionallycreates complete heart block as a means topermanently slow the ventricular rate. As aresult, permanent pacemaker placement isalso required to generate an adequate ven-tricular rate.

Paroxysmal SupraventricularTachycardias

Paroxysmal supraventricular tachycardias(PSVTs) are manifested by (1) sudden onsetand termination, (2) atrial rates between140 and 250 bpm, and (3) narrow (normal)QRS complexes, unless aberrant conduction ispresent, as described later (Fig. 12.14). Themechanism of PSVTs is most often reentryinvolving the AV node, atrium, or an acces-sory pathway between an atrium and ven-

tricle. Enhanced automaticity and triggeredactivity in the atrium or AV node are lesscommon causes.

AV Nodal Reentrant Tachycardia

Atrioventricular nodal reentrant tachycar-dia (AVNRT) is the most common form ofPSVT in adults. In the normal heart, the AVnode is a lobulated structure that consists ofa compact portion and several atrial exten-sions. The latter constitute two (or more)potential pathways for conduction throughthe AV node (Fig. 12.15). In some people,these extensions conduct at different veloc-ities, providing both slow-conducting andfast-conducting pathways. The fast pathwayis characterized by a rapid conduction ve-locity, whereas the slow pathway demon-strates slower conduction but typically has ashorter refractory period than the fast path-way. Thus, although the fast pathway con-ducts rapidly, it takes longer to recover be-tween impulses compared with the slowpathway. Normally, a stimulus arriving atthe AV node travels down both pathways,but the impulse traveling down the fastpathway reaches the bundle of His first. Bythe time the impulse traversing the slow path-way reaches the bundle of His, it encountersrefractory tissue and is extinguished. Thus,under normal conditions, only the fast path-way impulse makes its way forward to theventricles (see Fig 12.15A).

In contrast, consider what happens whenan atrial premature beat (APB) spontaneouslyoccurs (Fig 12.15B). Because the refractoryperiod of the fast pathway is relatively long,

Figure 12.14. Paroxysmal supraventricular tachycardia caused by AV nodal reen-try. Retrograde P waves in this example occur simultaneously with, and are “hidden” in,the QRS complexes.

Fig. 14

Fig. 15

10090-12_CH12.qxd 8/31/06 4:55 PM Page 297

298 Chapter Twelve

Figure 12.15. Common mechanism of AV nodal reentry. In most patients, the AVnode (gray region in the drawing) is a lobulated structure consisting proximally of sev-eral atrial extensions and distally of a compact node portion. A. In patients with AVnodal reentry, two functionally distinct tracts exist within the AV node (termed theslow and fast pathways). The slow pathway conducts slowly and has a short refrac-tory period, whereas the fast pathway conducts more rapidly but has a long refractoryperiod. Impulses from above conduct down both pathways; because the fast pathwayimpulse reaches the distal common pathway first, it continues to the bundle of His.Conversely, the slow pathway impulse arrives later and encounters refractory tissue.B. An atrial premature beat arrives at the entrance of the two pathways. The fast path-way is still refractory from the preceding beat and the impulse is blocked, but the slowpathway has repolarized and is able to conduct. When the impulse reaches the distalportion of the fast pathway after traveling down the slower pathway, the fast pathwayhas repolarized and is able to conduct the impulse in a retrograde direction (exempli-fying unidirectional block). The impulse can then travel through the atrium and back tothe slow pathway, and a reentrant loop is initiated.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 298


an APB would find that pathway unex-citable and unable to conduct the impulse.However, the impulse is able to conductover the slow pathway (which is excitablebecause it has a shorter refractory periodthan the fast pathway and has already repo-larized when the APB arrives). By the timethis impulse travels down the slowly con-ducting pathway and reaches the compactportion of the AV node, the distal end of thefast pathway may have had time to repolar-ize, and the impulse is able to propagateboth distally (to the bundle of His and ven-tricles) and backward to the atria, up the fastpathway in a retrograde direction. On reach-ing the atria, the impulse can then circulateback down the slow pathway, completingthe reentrant loop and initiating tachycar-dia as this sequence repeats. Thus, the fun-damental conditions for reentry in AVNRTin this example are transient unidirectionalblock in the fast pathway (an APB encoun-tering refractory tissue) and relatively slowconduction through the other pathway.

The ECG in AVNRT shows a regular tachy-cardia with normal-width QRS complexes. P waves may not be apparent, because retro-grade atrial depolarization typically occurs si-multaneously with ventricular depolariza-tion. Thus, the retrograde P wave and QRS areinscribed at the same time and the P is typi-cally “hidden” in the QRS complex. When P waves are visible, they are superimposed onthe terminal portion of the QRS complex andinverted (negative deflection) in limb leads II,III, and aVF because of the caudocranial di-rection of atrial activation.

Rarely, the reentrant loop revolves in thereverse direction, with anterograde conduc-tion down the fast pathway, and retrogradeconduction up the slow pathway. This isknown as uncommon AVNRT and, unlike themore common rhythm, results in clearlyvisible retrograde P waves following the QRScomplex on the ECG.

AVNRT often presents in teenagers oryoung adults. It is usually well tolerated butcauses palpitations that many patients findfrightening, and rapid tachycardias can causelight-headedness or shortness of breath. Inelderly patients or those with underlying

heart disease, more severe symptoms mayresult, such as syncope, angina, or pul-monary edema.

Acute treatment of AVNRT is aimed at ter-minating reentry by impairing conductionin the AV node. Transient increases in vagaltone produced by the Valsalva maneuver orcarotid sinus massage (see Chapter 11) mayblock AV conduction, terminating the tachy-cardia. The most rapidly effective pharmaco-logic treatment is intravenous adenosine,which impairs AV nodal conduction andoften aborts the reentrant rhythm (see Chap-ter 17). Other drug options include intra-venous calcium channel antagonists (vera-pamil and diltiazem) or β-blockers.

Most patients with AVNRT have infre-quent episodes that terminate with vagalmaneuvers and do not require other specificinterventions. Frequent symptomaticepisodes, particularly when requiring visitsto the emergency department for treatment,warrant preventive therapy: oral β-blockers,calcium channel blockers, or digoxin areoften successful for this purpose. Catheterablation of the slow AV nodal pathway isalso an effective option but is associatedwith a small risk (approximately 1% to 2%)of heart block owing to unintended damageto the fast AV nodal pathway, which re-quires permanent pacemaker implantation.Chronic class IA or IC antiarrhythmic drugsare also effective but are often less desirablethan catheter ablation because of associatedpotential toxicities.

Atrioventricular Reentrant Tachycardias

Atrioventricular reentrant tachycardias(AVRTs) are similar to AVNRTs except thatin the former, one limb of the reentrantloop is constituted by an accessory path-way (bypass tract), rather than by separatefast and slow pathways within the AV nodeitself. As described in Chapter 11, an acces-sory pathway is an abnormal band of myo-cytes that spans the AV groove and con-nects atrial to ventricular tissue separatelyfrom the normal conduction system (seeFig. 11.10). Approximately 1 in 1,500 peo-ple has such a pathway.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 299

Accessory pathways allow an impulse toconduct from atrium to ventricle (antero-grade conduction), from ventricle to atrium(retrograde conduction), or in both direc-tions. Depending on the characteristics ofthe pathway, one of two characteristic enti-ties can result: (1) the ventricular preexcita-tion syndrome, or (2) PSVT resulting from aconcealed bypass tract. Some pathways donot conduct impulses at rates sufficient tocause tachycardias and cause no symptomsat all.

Ventricular Preexcitation Syndrome

In patients with ventricular preexcitation(also termed Wolff-Parkinson-White [WPW]syndrome; see Chapter 11), atrial impulsescan pass in an anterograde direction to theventricles through both the AV node andthe accessory pathway. Because conductionthrough the accessory pathway is usuallyfaster than that via the AV node, the ventri-cles are stimulated earlier (preexcited) thanby normal conduction over the AV node.During sinus rhythm, activation of the ven-tricle from the accessory pathway causes acharacteristic ECG appearance: (1) the PRinterval is short (<0.12 sec) because ventric-ular stimulation begins earlier than normalthrough the accessory pathway, (2) the QRShas a slurred rather than a sharp upstroke(referred to as a delta wave) because initialventricular activation by the accessory path-way is slower than activation over the Purk-inje system, and (3) the QRS complex iswidened because it represents fusion of twoexcitation wave fronts through the ventri-cles, one from the accessory pathway and

one from the normal His-Purkinje system(Figs. 12.16 and 12.17).

Patients with WPW syndrome are predis-posed to PSVTs because the accessory path-way provides a potential limb of a reentrantloop. The most common PSVT in these patients is orthodromic AVRT. During thistachycardia, an impulse travels anterogradelydown the AV node to the ventricles andthen retrogradely up the accessory tract backto the atria (see Fig. 12.17B). Because theventricles in this situation are depolarizedexclusively via the normal conduction sys-tem (through the AV node and bundle ofHis), there is no delta wave during thetachycardia and the width of the QRS is usu-ally normal. Retrograde P waves are oftenvisible soon after each QRS complex becausethe atria are stimulated from below via ret-rograde conduction through the accessorypathway.

In fewer than 10% of patients with AVRTinvolving an accessory pathway, the re-entrant arrhythmia travels in the opposite di-rection. Impulses travel anterogradely downthe accessory pathway and retrogradely up theAV node (see Fig. 12.17C). Termed antidromicAVRT, its ECG pattern is characterized by awide QRS complex because the ventricles areactivated entirely from anterograde conduc-tion over the accessory pathway. From theECG alone, such antidromic tachycardia is difficult to distinguish from ventriculartachycardia (described later in the chapter).

A third type of arrhythmia encounteredin patients with WPW syndrome is antero-grade conduction over the accessory path-way when atrial fibrillation or flutter is pre-sent. Some accessory pathways have short

300 Chapter Twelve

Figure 12.16. Wolff-Parkinson-White syndrome. The delta wave (arrow) indicates preexcitation of the ven-tricles. (Courtesy of Dr. Eric Isselbacher, Massachusetts General Hospital, Boston.)

Fig. 16-17

10090-12_CH12.qxd 8/31/06 4:55 PM Page 300


refractory periods that allow faster rates ofventricular stimulation than does the AVnode. Thus, during AF or atrial flutter, ven-tricular rates as fast as 300 bpm may result.Such rates are poorly tolerated and can leadto ventricular fibrillation and cardiac arrest,even in a young, otherwise healthy patient.

Pharmacologic management of arrhyth-mias in patients with the WPW syndromerequires greater caution than those withAVNRTs. Although digitalis, β-blockers, andcertain calcium++ channel blockers are effec-tive at blocking conduction through the AV node, they do not slow conduction overmost accessory pathways. Sometimes thesedrugs actually shorten the refractory periodof the accessory pathway, thus speeding con-

duction. Therefore, the drugs could precipi-tate even faster ventricular rates (and hemo-dynamic collapse) when administered to pa-tients with WPW syndrome who developatrial fibrillation or flutter. In contrast,sodium+ channel blockers (specifically, classIA and IC antiarrhythmics) and some classIII antiarrhythmic drugs slow conductionand prolong the refractory period of acces-sory pathways as well as the AV node; there-fore, these are the preferred pharmacologicagents for this condition.

When a patient with WPW presents with awide QRS tachycardia, acute therapy dependson the patient’s hemodynamic tolerance ofthe arrhythmia. If accompanied by hemody-namic collapse, immediate cardioversion is

Figure 12.17. Wolff-Parkinson-White syndrome. A. During normal sinus rhythm, the shortened PR interval, deltawave, and widened QRS complex indicate fusion of ventricular activation via the AV node and accessory pathway. B. Anatrial premature beat can trigger an orthodromic atrioventricular reentrant tachycardia, in which impulses are conductedanterogradely down the AV node and retrogradely up the accessory pathway. Retrograde P waves are visible immedi-ately after the QRS complex. There is no delta wave because anterograde ventricular stimulation passes exclusivelythrough the AV node. C. Antidromic atrioventricular reentrant tachycardia in which impulses are conducted antero-gradely down the accessory tract and retrogradely up the AV node. The QRS complex is very widened because the ven-tricles are stimulated by abnormal conduction through the accessory pathway. SA, sinoatrial.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 301

required. Conversely, if the patient is hemo-dynamically stable, intravenous administra-tion of procainamide (a class IA agent thatslows conduction in the accessory pathway)will often terminate the arrhythmia.

Patients who have WPW with symptom-atic arrhythmias should generally under-go an electrophysiologic study with radio-frequency catheter ablation of the accessorypathway. Ablation abolishes conductionover the pathway, curing the condition. Ifthis procedure is not an option, chronic oraltherapy should include a drug that slows ac-cessory pathway conduction (i.e., a class IA,IC, or III agent).

Another form of ventricular preexcitationis the Lown-Ganong-Levine syndrome, anuncommon condition also characterized bya short PR interval but a normal, narrow QRScomplex (i.e., no delta wave during sinusrhythm). Although it has been suggested thata short accessory pathway connects the atriadirectly to the His-Purkinje system in thiscondition, most patients just have enhancedconduction through the normal AV node,thus shortening the PR interval. When PSVToccurs in these patients, it is usually owing toAV nodal reentry.

Concealed Accessory Pathways

Accessory pathways do not always result inECG findings of ventricular preexcitation(i.e., short PR, delta wave). Many are capableof only retrograde conduction. In this case,during sinus rhythm, the ventricles are depo-larized normally through the AV node aloneand the ECG is normal (i.e., the accessorypathway is concealed). However, because theaccessory pathway is capable of retrogradeconduction, it can form a limb of a reentrantcircuit and result in orthodromic AVRT.

Management of patients with tachycardiainvolving a concealed accessory pathway isthe same as for patients with AVNRT. Be-cause the reentrant circuit travels antero-gradely down the AV node, vagal maneuversand drugs that interrupt conduction overthe AV node (e.g., adenosine, verapamil, dil-tiazem, and β-blockers) can terminate thetachycardia. Another option in patients with

recurrent episodes is catheter ablation of theaccessory pathway, which permanently pre-vents recurrences in most patients.

Ectopic Atrial Tachycardia

Ectopic atrial tachycardia (AT) results fromeither automaticity of an atrial focus or re-entry. The ECG has the appearance of sinustachycardia, with a P wave before each QRScomplex, but the P wave morphology is dif-ferent from that of sinus rhythm, indicatingdepolarization of the atrium from an abnor-mal site. The arrhythmia can be paroxysmaland of limited duration, or it can persist.Short, asymptomatic bursts of atrial tachy-cardia are commonly observed on 24-hourECG recordings, even in otherwise healthypeople.

Atrial tachycardia can be caused by digi-talis toxicity and is also aggravated by ele-vated sympathetic tone (e.g., during exertionor periods of illness). Initial treatment in-cludes correction of any contributing factors.Unlike reentrant supraventricular tachycar-dias, vagal maneuvers (such as carotid sinusmassage) may have no effect on atrial dis-charges from an ectopic pacemaker focus.However, β-blockers; calcium channel block-ers; and class IA, IC, and III antiarrhythmicdrugs can be effective. Catheter ablation isalso an option for symptomatic patients.

Multifocal Atrial Tachycardia

In multifocal atrial tachycardia (MAT), theECG shows an irregular rhythm with multi-ple (at least three) P wave morphologies,and the average atrial rate is >100 bpm (Fig.12.18). An isoelectric (i.e., “flat”) baselinebetween P waves distinguishes MAT fromthe chaotic baseline of AF. This rhythm islikely caused by either abnormal automatic-ity in several foci within the atria or trig-gered activity and occurs most often in thesetting of severe pulmonary disease and hy-poxemia. Because patients with this rhythmare often critically ill from the underlyingdisease, the mortality rate is high, and treat-ment is aimed at the causative disorder. Thecalcium channel blocker verapamil is often

302 Chapter Twelve

Fig. 18

10090-12_CH12.qxd 8/31/06 4:55 PM Page 302


effective at slowing the ventricular rate as atemporizing measure.

Ventricular Arrhythmias

The common ventricular arrhythmias are (1) ventricular premature beats, (2) ventricu-lar tachycardia,and(3) ventricular fibrillation.Ventricular arrhythmias are usually moredangerous than supraventricular rhythm dis-orders and are responsible for many of the ap-proximately 300,000 sudden cardiac deathsthat occur every year in the United States.

Ventricular Premature Beats

Similar to APBs, ventricular premature beats(VPBs) are common even among healthypeople and are often asymptomatic and be-nign (Fig. 12.19). A VPB arises when an ec-topic ventricular focus fires an action poten-tial. On the ECG, a VPB appears as a widenedQRS complex, because the impulse travelsfrom its ectopic site through the ventriclesvia slow cell-to-cell connections rather thanthrough the normal rapidly conducting His-Purkinje system. Furthermore, the ectopicbeat is not related to a preceding P wave.

VPBs can also occur in repeating patterns.When every alternate beat is a VPB, therhythm is termed bigeminy. When two nor-mal beats precede every VPB, trigeminy issaid to be present. Three preceding normalbeats are referred to as quadrigeminy, and so on. Consecutive VPBs are referred to ascouplets (two in a row) or triplets (three ina row).

VPBs are not dangerous by themselves,and in patients without heart disease, theyconfer no added risk of a life-threatening ar-rhythmia. They can, however, be an indica-tion of an underlying cardiac disorder andtake on added significance in that case. Forexample, in patients with structural heartdisease, VPBs generally increase in frequencyin relation to the severity of depressed ven-tricular contractility. They have been associ-ated with an increased risk of sudden deathin patients with heart failure or prior my-ocardial infarction. Therefore, the discoveryof VPBs warrants assessment for underlyingheart conditions.

In otherwise healthy persons, treatment of VPBs mainly involves reassurance and,if needed, symptomatic control using β-blockers. In patients with advanced structural

Figure 12.18. Multifocal atrial tachycardia. Each QRS is preceded by a P wave (arrows) of varyingmorphology. (Courtesy of Dr. Eric Isselbacher, Massachusetts General Hospital, Boston.)

Figure 12.19. Ventricular premature beats (arrows).

Fig. 19

10090-12_CH12.qxd 8/31/06 4:55 PM Page 303

heart disease who are at risk of life-threaten-ing ventricular arrhythmias, placement ofan implantable cardioverter-defibrillator(ICD) is typically recommended.

Ventricular Tachycardia

Ventricular tachycardia (VT) is a series ofthree or more VPBs (Fig. 12.20). VT is dividedarbitrarily into two categories. If it persistsfor more than 30 seconds, produces severesymptoms such as syncope, or requires ter-mination by cardioversion or administrationof an antiarrhythmic drug, it is designated assustained VT; self-terminating episodes aretermed nonsustained VT. Both forms of VTare found most commonly in patients withstructural heart disease, including myocar-dial ischemia and infarction, heart failure,ventricular hypertrophy, primary electricaldiseases (e.g., long-QT syndromes; see Box12.1), valvular heart diseases, and congenitalcardiac abnormalities.

The QRS complexes of VT are typicallywide (>0.12 sec) and occur at a rate of 100to 200 bpm or sometimes faster. VT is fur-ther categorized according to its QRS mor-phology. When every QRS complex ap-pears the same and the rate is regular, it isreferred to as monomorphic VT (see Fig.12.20). Sustained monomorphic VT usu-ally indicates a structural abnormality thatsupports a reentry circuit, most commonlya region of myocardial scar from an old in-farction or cardiomyopathy. Occasionally,sustained monomorphic VT occurs as a re-sult of an ectopic ventricular focus in anotherwise healthy person, usually referredto as idiopathic VT.

When the QRS complexes continuallychange in shape and the rate varies from beat

to beat, the VT is referred to as polymorphic.Multiple ectopic foci or a continually chang-ing reentry circuit is the cause. Torsades depointes (discussed later in the chapter) andacute myocardial ischemia or infarction arethe most common causes of polymorphicVT. Rare, inherited predispositions to poly-morphic ventricular tachycardia and suddendeath arise from abnormalities of cardiacpotassium and sodium channels (e.g., thelong-QT syndromes and the Brugada syn-drome), as described in Box 12.1. Sustainedpolymorphic VT usually degenerates toventricular fibrillation.

The symptoms of VT vary depending onthe rate of the tachycardia, its duration, andthe underlying condition of the heart. Sus-tained VT can cause low cardiac output re-sulting in loss of consciousness (syncope),pulmonary edema, or progress to cardiacarrest. These severe consequences of VT aremost likely in patients who have under-lying depressed contractile function. Con-versely, If sustained VT is relatively slow(e.g., <130 bpm) it may be well toleratedand cause only palpitations.

Distinguishing Monomorphic VT from Supraventricular Tachycardia

Ventricular tachycardia can usually be dis-tinguished from supraventricular tachycar-dia (SVT) by the width of the QRS com-plex: it is routinely wide in the former andnarrow (i.e., normal) in the latter, as indi-cated in Figure 12.9. However, under cer-tain circumstances, arrhythmias that orig-inate from sites above the ventricles canresult in wide QRS complexes and may ap-pear similar to monomorphic VT. This sit-

304 Chapter Twelve

Figure 12.20. Monomorphic ventricular tachycardia.

Fig. 20

Box 1

10090-12_CH12.qxd 8/31/06 4:55 PM Page 304


uation is termed supraventricular tachycar-dia with aberrant ventricular conduction, orsimply SVT with aberrancy, and may arisein three scenarios: (1) a patient has an un-derlying conduction abnormality (e.g., abundle branch block), such that the QRS isabnormally wide even when in normalsinus rhythm; (2) repetitive rapid ventric-ular stimulation during an SVT finds oneof the bundle branches refractory (becauseof insufficient time to recover from theprevious depolarization), such that the im-pulse propagates abnormally through theventricles, causing the QRS to be distortedand wide; or (3) a patient develops an-tidromic tachycardia through an accessorypathway (described earlier).

Certain clinical and electrocardiographicfeatures can help to distinguish wide QRScomplexes of monomorphic VT from thoseof supraventricular rhythms with aberrantconduction. In patients with a history ofprior myocardial infarction, congestiveheart failure, or left ventricular dysfunc-tion, a wide complex tachycardia is morelikely to be VT rather than SVT with aber-rancy. At the bedside, SVT is more probableif vagal maneuvers (such as carotid sinusmassage) affect the rhythm, as indicated inFigure 12.9.

Electrocardiographically, a supraventricu-lar tachyarrhythmia is more likely if themorphology of the QRS at the rapid rate issimilar to that on the patient’s ECG tracingobtained while in sinus rhythm (i.e., thecomplex is widened because of an underly-ing bundle branch block). Conversely, ven-tricular tachycardia is more likely if (1) thereis no relationship between the QRS com-plexes and any observed P waves (atrioven-tricular dissociation) or (2) the QRS com-plexes in each of the chest leads (V1 throughV6) all have a similar appearance, with adominant positive or negative deflection(i.e., there is “concordance” of the precor-dial QRS complexes). These features aresummarized in Table 12.2. Other morpho-logic ECG features have been used to distin-guish VT from SVT with aberrancy, but thedistinction is often very difficult. Most pa-tients with wide QRS tachycardia should

be managed as though they have VT untilproven otherwise.

Management of Patients with VT

Sustained episodes of VT are dangerous be-cause they can produce syncope or deterio-rate into ventricular fibrillation, which isfatal if not quickly corrected. Acute treat-ment usually consists of electrical cardiover-sion. Intravenous administration of certainantiarrhythmic drugs, such as amiodarone,procainamide, or lidocaine, can be con-sidered if the patient is hemodynamically stable.

After sinus rhythm is restored, a patientwho has had an episode of sustained VT re-quire careful evaluation to define whetherunderlying structural heart disease is pre-sent and to correct any aggravating factors,such as acute myocardial ischemia, electro-lyte disturbances, or drug toxicities. Patientswho have suffered VT in the setting of struc-tural heart disease have a high risk of recur-rence and sudden cardiac death; implan-tation of an ICD is usually warranted toautomatically and promptly terminate fu-ture episodes.

Patients who experience VT in the ab-sence of underlying structural heart diseaseare usually found to have idiopathic VT.This type of arrhythmia tends to originatefrom foci in the right ventricular outflowtract or in the septal portion of the left ven-tricle. It is rarely life threatening. β-Blockers,calcium channel blockers, or catheter abla-tion are commonly effective to control symp-tomatic episodes of idiopathic VT.

Torsades de Pointes

Torsades de pointes (“twisting of the points”)is a form of polymorphic VT presenting asvarying amplitudes of the QRS, as if thecomplexes are “twisting” about the base-line (Fig. 12.21). It can be produced byearly afterdepolarizations (triggered activ-ity), particularly in patients who have a pro-longed QT interval. QT prolongation (whichindicates a lengthened action potential du-ration) can result from electrolyte distur-

Tab. 2Fig. 21

10090-12_CH12.qxd 8/31/06 4:55 PM Page 305

306 Chapter Twelve

Box 12.1 Genetic Mutations and Ventricular Arrhythmias

Genetic causes of ventricular arrhythmias occur either in association with various types ofstructural heart disease or as isolated conditions. Examples of inherited structural diseasethat can be complicated by life-threatening arrhythmias include hypertrophic cardiomyo-pathy and the familial dilated cardiomyopathies, both described in Chapter 10. Ar-rhythmogenic right ventricular dysplasia (ARVD) is another form of cardiomyopathyassociated with reentrant ventricular tachycardia, typically originating from the right ven-tricle (RV). This condition is characterized by replacement of portions of the RV with adi-pose and fibrous tissue, and approximately one third of patients display a familial patternwith autosomal dominant inheritance. The most common mutations involve genes thatencode components of cell membrane desmosomes (structures involved in cell-to-cell ad-hesion) and the cardiac ryanodine receptor (see Chapter 1). ARVD may be suspected onroutine electrocardiogram (ECG) by the presence of inverted T waves in leads V1 throughV3 and occasionally an epsilon wave, a terminal notch of the QRS complex in lead V1 (seearrow in the accompanying figure), which reflects abnormalRV activation. The abnormal RV morphology can be identifiedby noninvasive imaging techniques in some patients. Treat-ment typically includes an implantable cardioverter-defibrilla-tor (ICD) because the disease is progressive and ventriculartachycardia is common, which can lead to sudden death.

Several other inherited arrhythmic disorders occur in theabsence of structural cardiac disease. These occur infrequently but are important becausethey can cause life-threatening arrhythmias in young, otherwise healthy people withoutprior warning. The most common of these conditions are (1) the Brugada syndrome, (2) the congenital long-QT syndromes, and (3) familial catecholaminergic polymorphicventricular tachycardia.

The Brugada syndrome is believed to be responsible for a large percentage of idio-pathic ventricular fibrillation. It is inherited in an autosomal dominant fashion and hasbeen linked in some (but not all) families to mutations in a sodium channel subunit gene(SCN5A). A clue to the presence of this syndrome is a specific ECG abnormality: right bun-dle branch block with prominent ST elevation in leads V1

through V3 (see accompanying figure). This pattern may bepresent chronically or intermittently; in the latter case, thesyndrome may be unmasked by administering certain an-tiarrhythmic drugs (e.g., flecainide, procainamide). Brugadasyndrome is a potentially lethal condition, and ICD implan-tation is the most effective way to prevent an arrhythmicdeath.

The congenital long-QT syndromes (LQTS) are associated with prolonged ventric-ular repolarization (hence the long QT interval), which can lead to life-threatening poly-morphic ventricular tachycardia (i.e., torsades de pointes). Mutations in at least sevengenes result in LQTS (see the accompanying table) by prolonging the action potentialduration. As shown in the table, identified mutations either enhance the depolarizingNa+ current, impair the repolarizing K+ current, or in one case (LQT4), result in an ab-normal cellular structural protein. There are two general phenotypes of LQTS: (1) theRomano-Ward syndrome, transmitted in an autosomal dominant fashion (an inheritancepattern that occurs with mutations in any of the listed genes); and (2) Jervelle and Lange-

V1

V1

10090-12_CH12.qxd 8/31/06 4:55 PM Page 306


Nielsen syndrome, an autosomal recessive condition associated with QT prolongationcombined with sensorineural hearing loss that results from mutations in two of theseven loci.

The symptomatology of patients with LQTS is highly variable, even for patients with thesame mutation. The degree of QT prolongation, and in some cases the patient’s gender,are predictors of arrhythmic risk when a mutation is present. An affected patient may beasymptomatic and come to medical attention only as a result of the abnormal ECG, or because of a family member with this condition. Others present with syncope or even sud-den death caused by torsades de pointes. The most common forms (LQT1 and LQT2) areassociated with ventricular arrhythmias during physical exercise or emotional stress. Con-versely, those with LQT3 are much more likely to experience cardiac events at rest or dur-ing sleep.

Other acquired conditions that prolong the QT interval can trigger life-threatening arrhythmias in patients with LQTS, including hypokalemia, hypomagnesemia, hypocal-cemia, and several medications (including many antiarrhythmic drugs). Conversely, β-blockers reduce the risk of arrhythmias in many forms of LQTS, even though they do notshorten the QT interval (see discussion of torsades de pointes later in the chapter). For pa-tients at high risk of life-threatening arrhythmias, ICD implantation is warranted.

Familial catecholaminergic polymorphic ventricular tachycardia, inherited in autosomal dominant and recessive patterns, is marked by ventricular tachycardia and/orventricular fibrillation during exercise or emotional arousal. The mechanism is thought tobe triggered activity resulting from delayed afterdepolarizations (see Chapter 11). Muta-tions in affected families have been demonstrated in at least two genes involved in intra-cellular calcium handling, including a missense mutation in the locus that codes for the car-diac ryanodine receptor. β-Blockers are effective for some patients; otherwise, an ICD isimplanted.

Genetic Basis of Congenital Long-QT Syndromes

Gene Mechanism of Prolonged Type (location) Protein Repolarization Inheritance

LPT1

LQT2

LQT3

LQT4

LQT5

LQT6

LQT7

AD, autosomal dominant (i.e., Romano-Ward syndrome); AR, autosomal recessive (i.e., Jervell and Lange-Nielsensyndrome [long QT and sensorineural deafness]).

KCNQ1(11p15)KCNH2(7q35)SCN5A(3p21)ANK2(4q25)KCNE1(21q22)KCNE2(21q22)KCNJ2(17q23)

KvLQT1(α subunit of IKs K+ channel)HERG(α subunit of IKr K+ channel)Nav 1.5(Na+ channel)Ankyrin-B(structural protein)MinK(β subunit of IKs K+ channel)MiRP1(β subunit of IKr K+ channel)IK1(Kir2.1 inward rectifier K+ channel)

↓Outward K+ current


↑Inward Na+ current

Not yet known



↓outward K+ current

AD and AR

AD

AD

AD

AD and AR

AD

AD

10090-12_CH12.qxd 8/31/06 4:55 PM Page 307

bances (especially hypokalemia or hypo-magnesemia), persistent bradycardia, anddrugs that block cardiac potassium currents,including many antiarrhythmic agents (particularly the class III drugs sotalol, ibu-tilide, and dofetilide and some class I drugs, including quinidine, procainamide,and disopyramide). Many medications givenfor noncardiac illnesses can prolong the QT interval and predispose to torsades depointes, including erythromycin, phenoth-iazines, haloperidol, and methadone. A raregroup of hereditary ion channel abnormal-ities produce congenital QT prolongation,which can also lead to torsades de pointes(see Box 12.1).

Torsades de pointes is usually sympto-matic, causing light-headedness or syncope,but is frequently self-limited. Its main dangerresults from degeneration into ventricularfibrillation. When it is drug or electrolyte in-duced, correcting the underlying cause abol-ishes the arrhythmia. Administration ofintravenous magnesium often suppressesrecurrent episodes. Other preventive strate-gies are aimed at shortening the QT intervalby increasing the underlying heart rate. Suchstrategies include administering intravenousβ-adrenergic stimulating agents (e.g., isopro-

terenol) and accelerating the heart rate via anartificial pacemaker. Paradoxically, when tor-sades de pointes results from congenital pro-longation of the QT interval, β-blocking drugsare often the treatment of choice becausesympathetic stimulation actually aggravatesthe arrhythmia. An implantable defibrillatoris often appropriate for these patients.

Ventricular Fibrillation

Ventricular fibrillation (VF) is an immedi-ately life-threatening arrhythmia (Fig. 12.22).It results in disordered, rapid stimulation ofthe ventricles with no coordinated contrac-tion. The result is essentially cessation ofcardiac output and death if not quickly re-versed. This rhythm most often occurs inpatients with severe underlying heart dis-ease and is the major cause of mortality inacute myocardial infarction.

VF is often initiated by an episode of ven-tricular tachycardia, which degenerates, it isbelieved, by the breakup of excitation wavesinto multiple smaller wavelets of reentrythat wander through the myocardium. Onthe ECG, VF is characterized by a chaotic ir-regular appearance without discrete QRSwaveforms.

308 Chapter Twelve

TABLE 12.2. Differentiation of Wide Complex Tachycardias

Supports SVT with Aberrant Conduction Supports Ventricular Tachycardia

QRS morphology same as when in No relationship between P wave and QRS sinus rhythm complexes

Rhythm responds to vagal Concordance of QRS complexes in the chest leads maneuvers (see Fig. 12.9) (V1–V6)

SVT, supraventricular tachycardia.

Figure 12.21. Torsades de pointes. The widened polymorphic QRS complexes demonstrate a waxing andwaning pattern; the QT interval was prolonged before the onset of the arrhythmia (not shown).

Fig. 22

10090-12_CH12.qxd 8/31/06 4:55 PM Page 308


Untreated, VF rapidly leads to death. Theonly effective therapy is prompt electricaldefibrillation. As soon as the heart has beenconverted to a safe rhythm, the underlyingprecipitant of the arrhythmia (e.g., elec-trolyte imbalances, hypoxemia, or acidosis)should be sought and corrected to preventfurther episodes. Intravenous antiarrhyth-mic drug therapy may be administered toprevent immediate recurrences. If no re-versible inciting precipitant is found, sur-vivors of VF usually receive an ICD.

SUMMARY

1. Disorders of impulse formation and con-duction result in bradyarrhythmias andtachyarrhythmias. Through careful analy-sis of the ECG, it is usually possible to dis-tinguish among the array of rhythm dis-orders so that appropriate therapy can beadministered.

2. When evaluating a patient with a slowheart rhythm (Figs. 12.1–12.8), two keyquestions should be addressed:

a. Are P waves present?

b. What is the relationship between theP waves and the QRS complexes?

3. Differentiation of tachyarrhythmias re-quires assessment of the following:

a. The width of the QRS complex (nor-mal or wide)

b. The morphology and rate of the P waves

c. The relationship between the P wavesand the QRS complexes

d. The response to vagal maneuvers (seeFig. 12.9)

Each of the ECG texts listed at the end ofChapter 4 provides additional examples of the rhythm disorders presented in thischapter.

Acknowledgments

Contributors to the previous editions of this chapterwere Wendy Armstrong, MD; Nicholas Boulis, MD;Jennifer E. Ho, MD; Marc S. Sabatine, MD; Elliott M.Antman, MD; Leonard I. Ganz, MD; and Leonard S.Lilly, MD.

Additional Reading

Ackerman MJ. Molecular basis of congenital and ac-quired long QT syndromes. J Electrocardiol 2004;37(suppl):1–6.

Delacretaz E. Supraventricular Tachycardia. N Engl J Med 2006;354:1039–1051.

Goldberger Z, Lambert R. Implantable cardioverter-defibrillators. JAMA 2006;295:809–818.

Gregoratos G, Abrams J, Epstein AE, et al. ACC/AHA/NASPE 2002 guideline update for implantation ofcardiac pacemakers and antiarrhythmia devices:summary article. A report of the American Collegeof Cardiology/American Heart Association taskforce on practice guidelines (ACC/AHA/NASPEcommittee to update the 1998 pacemaker guide-lines). Circulation 2002;106:2145.

Iqbal MB, Taneja AK, Lip GY, Flather M. Recent de-velopments in atrial fibrillation. BMJ 2005;330:238–43.

Kléber AG, Rudy Y. Basic mechanisms of cardiac im-pulse propagation and associated arrhythmias.Physiol Rev 2004; 84:431–488.

McNamara RL, Tamariz LJ, Segal JB, et al. Manage-ment of atrial fibrillation: review of the evidencefor the role of pharmacologic therapy, electricalcardioversion, and echocardiography. Ann InternMed 2003;139:1018.

Morady F. Catheter ablation of supraventricular ar-rhythmias: state of the art. J Cardiovasc Electro-physiol 2004;15:124–139.

Figure 12.22. Ventricular fibrillation.

10090-12_CH12.qxd 8/31/06 4:55 PM Page 309

Roberts R. Genomics and cardiac arrhythmias. J AmColl Cardiol 2006;47:9–21.

Roden DM. Drug-induced prolongation of the QTinterval. N Engl J Med 2004;350:1013–1022.

Schwartz PJ, Spazzolini C, Crotti L, et al. The Jervelland Lange-Nielson Syndrome—natural history,molecular basis, and clinical outcome. Circulation2006;113:783–790.

Shah M, Akar FG, Tomaselli GF. Molecular basis ofarrhythmias. Circulation 2005;112:2517–2529.

Trohman RG, Kim MH, Pinski SL. Cardiac pacing:the state of the art. Lancet 2004;364:1701–1719.

Wellens HJ. Twenty-five years of insights into themechanisms of supraventricular arrhythmias. J Cardiovasc Electrophysiol 2003;14:1020–1025.

310 Chapter Twelve

10090-12_CH12.qxd 8/31/06 4:55 PM Page 310

Chapter 12—Author Query1. AU: Previously, atrial rate given in bpm. Correct here, too? If so, is 350–600 still OK?

10090-12_CH12.qxd 8/31/06 4:55 PM Page a

311

WHAT IS HYPERTENSION?

HOW IS BLOOD PRESSURE REGULATED?Four Regulatory SystemsBlood Pressure Reflexes

ESSENTIAL HYPERTENSIONEpidemiology and GeneticsExperimental FindingsNatural History

SECONDARY HYPERTENSIONPatient EvaluationExogenous Causes

Renal CausesMechanical CausesEndocrine Causes

CONSEQUENCES OF HYPERTENSIONClinical Signs and SymptomsOrgan Damage Caused by Hypertension

HYPERTENSIVE CRISIS

TREATMENT OF HYPERTENSIONNonpharmacologic TreatmentPharmacologic Treatment

C H A P T E R

13HypertensionPayman ZamaniGordon H. WilliamsLeonard S. Lilly

More than 50 million Americans have hyper-tension—a blood pressure high enough tobe a danger to their well-being. That num-ber will undoubtedly rise; data from theFramingham Heart Study indicate that 90%of people over age 55 will develop hyper-tension during their lifetimes. Thus, thiscondition represents a great public healthconcern because it is a major risk factor forcoronary artery disease, stroke, heart failure,renal disease, and peripheral vascular dis-ease. Surprisingly, more than two thirds ofhypertensive persons are either unaware of their elevated blood pressure or are nottreated adequately to minimize the cardio-vascular risk. Moreover, because elevated

blood pressure is usually asymptomaticuntil an acute cardiovascular event strikes,screening for hypertension is a critical aspectof preventive medicine.

Hypertension is also a scientific problemof unexpected complexity. In almost 95% ofaffected patients, the cause of the bloodpressure elevation is unknown, a conditiontermed primary or essential hypertension(EH). Evidence suggests that the causes of EHare multiple and diverse, and considerable in-sight into those causes can be achieved bystudying the normal physiology of bloodpressure control, as examined in this chapter.

High blood pressure attributed to a defin-able cause is termed secondary hyperten-

10090-13_CH13.qxd 8/31/06 4:55 PM Page 311

sion. Although far less common than EH,conditions that cause secondary hyper-tension are important because many areamenable to permanent cure. Following thediscussions of essential and secondary hyper-tension, this chapter considers the clinicalconsequences of elevated blood pressureand approaches to treatment.

WHAT IS HYPERTENSION?

Blood pressure values vary widely in thepopulation. For example, diastolic pres-sures follow the smooth bell-shaped distri-bution shown in Figure 13.1. In addition,blood pressure levels tend to increase withage, as illustrated in Figure 13.2. The risk ofa vascular complication increases progres-sively and linearly with higher blood pres-sure values, so the exact cutoff points todefine stages of hypertension are some-what arbitrary. The currently establishedcriteria are listed in Table 13.1. By this clas-sification, a diastolic pressure consistentlyat or above 90 mm Hg or a systolic pressureat or above 140 mm Hg establishes the diagnosis of hypertension. Those withprehypertension have an increased risk of developing definite hypertension and areencouraged to undertake lifestyle modifica-tions. Although the emphasis has histori-cally been on the level of diastolic pressure,

recent evidence suggests that systolic pres-sure more accurately predicts cardiovascu-lar complications.

HOW IS BLOOD PRESSURE REGULATED?

Four Regulatory Systems

Blood pressure (BP) is the product of cardiacoutput (CO) and total peripheral resistance(TPR):

And CO is the product of cardiac stroke vol-ume (SV) and heart rate (HR):

Stroke volume is determined by (1) cardiaccontractility; (2) the venous return to theheart (i.e., the preload, as described in Chap-ter 9); and (3) the resistance the left ventri-cle must overcome to eject blood into theaorta (i.e., afterload).

It follows that at least four systems are di-rectly responsible for blood pressure regula-tion: the heart, which supplies the pumpingpressure; blood vessel tone, which largely determines systemic resistance; the kidney,which regulates intravascular volume; andhormones, which modulate the function theother three systems. Figure 13.3 shows how

CO SV HR= ×

BP CO TPR= ×

312 Chapter Thirteen

Figure 13.1. Distribution of diastolic blood pressure values in the30- to 69-year age group (n 5 158,906). Hypertension is arbitrarily de-fined as diastolic blood pressure ≥90 mm Hg. (Modified from Hyperten-sion Detection and Follow-up Program. A progress report. Circ Res 1977;40(suppl1):106.)

Fig. 1

Fig. 2

Tab. 1

10090-13_CH13.qxd 8/31/06 4:55 PM Page 312

Hypertension 313

factors related to these systems contribute toCO and TPR.

The renal component of blood pressureregulation deserves special mention, in lightof the temptation to view hypertension sim-ply as a cardiovascular problem. No matterhow high the CO or TPR, renal excretion hasthe capacity to completely return blood pres-sure to normal by reducing intravascular vol-ume. Therefore, the maintenance of chronichypertension requires renal participation.Transplantation studies have confirmed thispoint: the implantation of a kidney from anormotensive person into a hypertensive

one typically improves the blood pressure.Similarly, surgical placement of a kidneyfrom a genetically hypertensive rat into apreviously normotensive one usually leadsto hypertension.

In the presence of normally functioningkidneys, an increase in blood pressure leadsto augmented urine volume and sodium ex-cretion, which then returns the blood pres-sure back to normal. This process, known aspressure natriuresis, is blunted in the kidneysof hypertensive patients; thus, higher pres-sures are required to excrete a given sodiumand water load. Current evidence suggests atleast two possible reasons for this bluntedresponse. First, microvascular and tubulo-interstitial injury within the kidneys of hy-pertensive patients impairs sodium excre-tion, although the mechanism of this injuryis still under investigation. Second, the de-fect may lie with hormonal factors critical toappropriate renal reactions to the sodiumand intravascular volume environment(e.g., the renin-angiotensin system, as de-scribed later in the chapter). In contrast tothe first possibility, abnormalities of hor-monal regulation are amendable to correc-tion with appropriate therapy.

Blood Pressure Reflexes

The cardiovascular system is endowed withfeedback mechanisms that continuouslymonitor arterial pressure: they sense whenthe pressure becomes excessively high or lowand then respond rapidly to those changes.One such mechanism is the baroreceptorreflex, which is mediated by receptors in thewalls of the aortic arch and the carotid si-nuses. The baroreceptors monitor changes in

Figure 13.2. Relationship between blood pressureand age (n 5 1,029). Systolic (upper curves) and di-astolic (lower curves) values are shown. Notice thatby age 60, the average systolic pressure of women ex-ceeds that of men. (Modified from Kotchen JM, McKeanHE, Kotchen TA. Blood pressure trends with aging. Hy-pertension 1982;4(suppl 3):111–129.)

TABLE 13.1. Classification of Blood Pressure in Adults

Category Systolic Pressure (mm Hg) Diastolic Pressure (mm Hg)

Normal <120 And <80Prehypertension 120–139 Or 80–99Stage 1 hypertension 140–159 Or 90–99Stage 2 hypertension ≥160 Or ≥100

Modified from The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment ofHigh Blood Pressure. JAMA 2003;289:2560–2572.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 313

pressure by sensing the stretch and deforma-tion of the arteries. If the arterial pressurerises, the baroreceptors are stimulated, in-creasing their transmission of impulses to thecentral nervous system (i.e., the medulla).Negative feedback signals are then sent backto the circulation via the autonomic nervoussystem, causing the blood pressure to fallback to its baseline level.

The higher the blood pressure rises, themore the baroreceptors are stretched andthe greater the impulse transmission rate tothe medulla. Signals from the carotid sinusreceptors are carried by the glossopharyn-geal nerve (cranial nerve IX), whereas the

signals from the aortic arch receptors arecarried by the vagus nerve (cranial nerve X).These nerve fibers converge at the tractussolitarius in the medulla, where the barore-ceptor impulses inhibit sympathetic nervoussystem outflow and excite parasympatheticeffects. The net result is (1) a decline in pe-ripheral vascular resistance (i.e., vasodila-tion) and (2) a reduction in cardiac output(because of a lower heart rate and reducedforce of cardiac contraction). Each of theseeffects tends to lower arterial pressure backtoward its baseline. Conversely, when a fallin systemic pressure is sensed by the barore-ceptors, fewer impulses are transmitted to


Figure 13.3. Regulation of systemic blood pressure. The small arrows indicate whether there is a stimulatory (↑)or inhibitory (↓) effect on the boxed parameters. ADH, antidiuretic hormone; CC, cardiac contractility; HR, heart rate;NP, natriuretic peptides; PSNS, parasympathetic nervous system; SNS, sympathetic nervous system; SV, stroke volume;VR, venous return.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 314

Hypertension 315

the medulla, leading to a reflexive increase inblood pressure back to its baseline.

The main effect of the baroreceptor me-chanism is to modulate moment-by-momentvariations in systemic blood pressure. How-ever, the baroreceptor reflex is not involvedin the long-term regulation of blood pres-sure and does not prevent the developmentof chronic hypertension. The reason forthis is that the baroreceptors constantlyreset themselves. After a day or two of expo-sure to higher-than-baseline pressures, thebaroreceptor-firing rate slows back to its con-trol value.

ESSENTIAL HYPERTENSION

Almost 95% of hypertensive patients haveblood pressures that are elevated for noreadily definable reason, a condition termedessential hypertension. The diagnosis of EHis one of exclusion; it is the option left to theclinician after considering the causes of sec-ondary hypertension described later in thischapter.

EH is more a description than a diagnosis,indicating only that a patient manifests aspecific physical finding (high blood pres-sure) for which no cause has been found. Inall likelihood, different underlying defectsare responsible for the elevated pressure in different subpopulations of patients. Be-cause the exact nature of these defects is un-known, to understand EH is to understandthe possibilities: what could go wrong withnormal physiology to produce chronicallyelevated blood pressure?

This discussion of EH therefore reflectswhat is currently known about its epidemi-ology and genetics, experimental findings,and natural history. The picture that emergesis that EH likely results from multiple defects of blood pressure regulation that interact with environmental stressors. Theregulatory defects may be acquired or genetically determined and may be inde-pendent of one another. As a result, EH pa-tients exhibit varied combinations of ab-normalities and therefore have variousphysiologic bases for their elevated bloodpressures.

Epidemiology and Genetics

Heredity appears to play an important rolein EH, but definite genetic markers have notbeen consistently identified. It seems mostlikely that essential hypertension is a com-plex genetic disorder, involving several lociinteracting with environmental factors. Evi-dence for a genetic role is suggested by thehigher rate of elevated blood pressures amongfirst-degree relatives of hypertensive patientsthan in the general population. Concordancebetween identical twins is high and signifi-cantly greater than that for dizygotic twins.Furthermore, an uneven distribution of EHexists among different racial groups. For ex-ample, in most age distributions, blacks aresignificantly more likely to be hypertensivethan persons of other races.

Although no gene has been consistentlylinked to essential hypertension, several locihave demonstrated positive associations.For example, autosomal dominant con-tributors to elevated blood pressure have been discovered, usually involving defectsof renal sodium channels. However, theseabnormalities are rare and are thought to bepresent in only a small fraction of hyper-tensive patients. Genes regulating the renin-angiotensin-aldosterone axis have been mostthoroughly studied in hypertensive patientsbecause of the central role of this system indetermining intravascular volume and vas-cular tone. Within this group, certain poly-morphisms in the gene for angiotensinogenconfer an increased risk of hypertension.Additionally, polymorphisms in the genefor adducin, a cytoskeletal protein, may beinvolved in a subgroup of EH patients. Fi-nally, as described later in the chapter, sig-nificant associations exist between hyper-tension and obesity, insulin resistance, anddiabetes. The pathophysiologic and geneticlinks among these four conditions are areasof very active investigation.

Experimental Findings

System Abnormalities

Multiple defects in blood pressure regulationhave been found in EH patients and their rel-atives. By themselves, or in conjunction with

AQ1

10090-13_CH13.qxd 8/31/06 4:55 PM Page 315

one another, these abnormalities may con-tribute to chronic blood pressure elevation.

The heart can contribute to a high cardiacoutput–based hypertension owing to sympa-thetic overactivity. For example, when testedunder psychologically stressful conditions,hypertensive patients (and their first-degreerelatives) often develop excessive heart rateacceleration compared with control sub-jects, suggesting an excessive sympatheticresponse.

The blood vessels may contribute to pe-ripheral vascular resistance-based hyperten-sion by constricting in response to (1) in-creased sympathetic activity; (2) abnormalregulation of vascular tone by local factors,including nitric oxide, endothelin, and na-triuretic factors; or (3) ion channel defects incontractile vascular smooth muscle.

The kidney can induce volume-based hy-pertension by retaining excessive sodium andwater as a result of (1) failure to regulate renalblood flow appropriately; (2) ion channel de-fects (e.g., reduced basolateral Na+K+-ATPase),

which directly cause sodium retention; or (3) inappropriate hormonal regulation. Forexample, the renin-angiotensin-aldosteroneaxis is an important hormonal regulator ofperipheral vascular resistance. Renin levelsin EH patients (compared with those in nor-motensive persons) are subnormal in 30%,normal in 60%, and high in 10%. Becauserenin secretion should be suppressed by highblood pressure, even “normal” levels are in-appropriate in hypertensive patients. Thus,abnormalities of this system’s regulationmay play a role in some persons with EH.

Figure 13.4 highlights these and otherpotential mechanisms of EH. Note that al-though the heart, blood vessels, and kidneysare the organs ultimately responsible forproducing the pressure, primary defects maybe located elsewhere as well (e.g., the centralnervous system, arterial baroreceptors, andadrenal hormone secretion). Yet, althoughabnormal regulation at these sites can con-tribute to elevated blood pressure, it is im-portant to remember that without renal


••••

↓ Nitric oxide secretion↑ Endothelin productionCa2+ or Na+/K+ channel defectsHyperresponsiveness tocatecholamines

Blood vessel

Functional:

• Exaggerated medialhypertrophy

Structural:

Adrenal

• Catecholamine leakor malregulation

Kidney

••

RAA dysfunctionIon channel defects (e.g., Na+/K+/2Cl–

cotransporter, basolateral Na+/K+ ATPase,Ca2+ ATPase)

Pressure/volume receptors

• Desensitization

CNS

•••

↑ Basal sympathetic toneAbnormal stress responseAbnormal response to signals frombaroreceptors and volume receptors

Figure 13.4. Potential primary abnormalities in essential hypertension. These defects are supported by experi-mental evidence, but their specific contributions to essential hypertension remain unclear. CNS, central nervous sys-tem; RAA, renin-angiotensin-aldosterone system.

Fig. 4

10090-13_CH13.qxd 8/31/06 4:55 PM Page 316

Hypertension 317

complicity, malfunction of other systemswould not produce sustained hypertension,because the normal kidney is capable ofeliminating sufficient volume to return theblood pressure to normal.

Insulin Resistance, Obesity, and theMetabolic Syndrome

Recent research has suggested that the hor-mone insulin may play a role in the develop-ment of EH. In many people with hyperten-sion, especially those who are obese or havetype 2 diabetes, there is impaired insulin-dependent transport of glucose into manytissues (termed insulin resistance). As a result,serum glucose levels rise, stimulating thepancreas to release additional insulin. Ele-vated insulin levels may contribute to hyper-tension via increased sympathetic activationor by stimulation of vascular smooth musclecell hypertrophy, which increases vascularresistance. Smooth muscle cell hypertrophymay be caused by a direct mitogenic effect of insulin or through enhanced sensitivity to platelet-derived growth factor, anotherknown smooth muscle cell growth factor.

Obesity itself has been directly associatedwith hypertension. Possible explanations forthis relationship include (1) the release of an-giotensinogen from adipocytes as substratefor the renin-angiotensin system, (2) aug-mented blood volume related to increasedbody mass, and (3) increased blood viscositycaused by adipocyte release of profibrinogenand plasminogen activator inhibitor 1. Thecurrent epidemic of obesity has led to a dra-matic increase in the number of people withmetabolic syndrome. As described in Chapter5, this condition represents a clustering ofatherogenic risk factors, including hyperten-sion, hypertriglyceridemia, low serum HDL,a tendency toward glucose intolerance, andtruncal obesity. Current evidence suggeststhat insulin resistance is central to the patho-genesis of this clustering.

Natural History

EH characteristically arises after young adult-hood. Its prevalence increases with age and

more than 60% of Americans older than 60years of age are hypertensive. In addition,the hemodynamic characteristics of bloodpressure elevation in EH tend to change overtime. Even in the absence of disease, systolicpressures tend to increase throughout adultlife. Diastolic blood pressure, on the otherhand, rises until the age of 50 and declinesslightly from then on (see Fig. 13.2). Ac-cordingly, diastolic hypertension is morecommon in young people, while a substan-tial number of hypertensive patients overage 50 have isolated systolic hypertensionand normal diastolic values.

In younger persons with hypertension, elevated blood pressure tends to be driven by high cardiac output in the setting of relatively normal peripheral vascular resis-tance, termed the hyperkinetic phase of EH(Fig. 13.5). With advancing age, however,the effect of cardiac output declines, perhapsbecause of the development of left ventricu-lar hypertrophy and its consequent reduceddiastolic filling (which in turn reduces strokevolume and cardiac output). Conversely,vascular resistance increases with age owingto medial hypertrophy as the vessels adapt tothe prolonged pressure stress. Thus, youngerhypertensive patients often display aug-mented cardiac output as the principal ab-normality, and older patients tend to haveelevated total peripheral resistance as themajor hemodynamic finding.

In summary, EH is a syndrome that mayarise from many potential abnormalities, butit exhibits a characteristic hemodynamic pro-file and natural history. It is likely that mul-tiple defects, separately inherited or acquired,act individually or together to chronicallyraise blood pressure. Although we may notunderstand the precise underlying mecha-nisms in individual hypertensive patients,we can at least describe what kind of patho-physiology might be at fault. EH, althoughidiopathic, is not entirely a “black box.”

SECONDARY HYPERTENSION

Although EH dominates the clinical picture,a defined structural or hormonal cause for

Fig. 5

10090-13_CH13.qxd 8/31/06 4:55 PM Page 317

hypertension may be found in about 5% ofpatients. Thus, cases of such secondary hy-pertension are relatively uncommon, buttheir identification is important because theunderlying conditions are often curable andmay require therapy different from that ad-ministered for EH. Moreover, if secondaryhypertension is left uncontrolled, adaptivecardiovascular changes may develop analo-gous to those of long-standing EH thatcould cause the elevated pressures to persisteven after the underlying cause is corrected.

Although secondary forms should be con-sidered in the workup of all patients with hy-

pertension, there are clinical clues that a givenpatient may have one of the correctable con-ditions (Table 13.2):

1. Age. If a patient develops hypertensionbefore age 20 or after age 50 (outside theusual range of EH), secondary hyperten-sion is more likely.

2. Severity. Secondary hypertension oftencauses blood pressure to rise dramati-cally, whereas most EH patients usuallyhave mild-to-moderate hypertension.

3. Onset. Secondary forms of hypertensionoften present abruptly in a patient who


TABLE 13.2. Causes of Hypertension

Percent of Hypertensive Type Patients Clinical Clues

Essential

Chronic renal diseasePrimary aldosteronismRenovascular

Pheochromocytoma

Coarctation of the aorta

Cushing syndrome

95%

2–4%1–2%

1%

0.2%

0.1%

0.1%

• Age of onset: 20–50 years• Family history of hypertension• Normal serum K+, urinalysis• ↑ Creatinine, abnormal urinalysis• ↓ Serum K+

• Abdominal bruit• Sudden onset (especially if age >50 or <20)• ↓ Serum K+

• Paroxysms of palpitations, diaphoresis, and anxiety• Episodic hypertension in one-third of patients• Blood pressure in arms > legs, or right arm > left arm• Midsystolic murmur between scapulae• CXR: aortic indentation, rib-notching due to

collaterals• “Cushingoid” appearance (e.g., central obesity,

hirsutism)

Figure 13.5. Hemodynamic progression of essential hypertension. Schematic representa-tion of the changing contribution of cardiac output (CO) and total peripheral resistance (TPR) asage increases in many patients with essential hypertension.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 318

Hypertension 319

was previously normotensive, ratherthan gradually progressing over years asis the usual case in EH.

4. Associated signs and symptoms. The pro-cess that induces hypertension may giverise to other characteristic abnormalities,identified by the history and physical examination. For example, a renal arte-ry bruit (swishing sound caused by tur-bulent blood flow through a stenoticartery) may be heard on abdominal ex-amination in a patient with renal arterystenosis.

5. Family history. EH patients often have hy-pertensive first-degree relatives, whereassecondary hypertension more commonlyoccurs sporadically.

Patient Evaluation

The usual clinical evaluation of a patientwith recently diagnosed hypertension be-gins with a careful history and physical ex-amination, including a search for clues tothe secondary forms. For example, repeatedurinary tract infections may suggest thepresence of chronic pyelonephritis withrenal damage as the cause of hypertension.Excessive weight loss may be an indicator ofpheochromocytoma, whereas weight gainmay point to the presence of Cushing syn-drome. The history also should include anassessment of lifestyle behaviors that maycontribute to hypertension, such as exces-sive alcohol consumption, and the patient’smedications should be reviewed becausecertain drugs (e.g., glucocorticoids and es-trogen) may elevate blood pressure.

Laboratory tests commonly performed inthe evaluation of the hypertensive patient,including general screening for secondarycauses, are (1) urinalysis and serum concen-tration of creatinine and blood urea nitro-gen to evaluate for renal abnormalities; (2) serum potassium level (abnormally lowin renovascular hypertension or aldostero-nism); (3) blood glucose level (elevated indiabetes, which is strongly associated withhypertension and renal disease); (4) serumcholesterol, high-density lipoprotein (HDL)

cholesterol, and triglyceride levels, as part ofthe global vascular risk screen; and (5) anelectrocardiogram (for evidence of left ven-tricular hypertrophy caused by chronic hypertension).

If no abnormalities are found that suggesta secondary form of hypertension, the pa-tient is presumed to have EH and treated ac-cordingly. If, however, the patient’s bloodpressure continues to be elevated despitestandard treatments, then more detailed di-agnostic testing is undertaken to search forspecific secondary causes.

Exogenous Causes

Several medications can elevate blood pres-sure. For example, oral contraceptives maycause secondary hypertension in some wo-men. The mechanism is likely related to in-creased activity of the renin-angiotensinsystem. Estrogens increase the hepatic syn-thesis of angiotensinogen, leading to greaterproduction of angiotensin II (Fig. 13.6). An-giotensin II raises blood pressure by severalmechanisms, most notably by direct vaso-constriction and by stimulating the adrenalrelease of aldosterone. The latter hormonecauses renal sodium retention and thereforeincreased intravascular volume.

Other medications that can raise bloodpressure include glucocorticoids, cyclospo-rine (an antirejection drug used in patientswith organ transplants), erythropoietin (ahormone that increases bone marrow redblood cell formation and elevates bloodpressure by increasing blood viscosity andreversing local hypoxic vasodilatation), andsympathomimetic drugs (which are com-mon in over-the-counter cold remedies).

Two other substances that may contributeto hypertension are cocaine and chronic ex-cessive ethanol consumption. Both of theseare associated with increased sympatheticnervous system activity.

Renal Causes

Given the crucial role of the kidney in thecontrol of blood pressure, it is not surprisingthat renal dysfunction can lead to hyper-

Fig. 6

10090-13_CH13.qxd 8/31/06 4:55 PM Page 319

tension. In fact, renal disease contributes totwo leading endogenous causes of secondaryhypertension: renal parenchymal disease,accounting for 2% to 4% of hypertensive pa-tients, and renal arterial stenosis, which ac-counts for approximately 1%.

Renal Parenchymal Disease

Parenchymal damage to the kidney can re-sult from diverse pathologic processes. Themajor mechanism by which injury leads toelevated blood pressure is through increased

intravascular volume. Damaged nephronsare unable to excrete normal amounts ofsodium and water, leading to a rise in in-travascular volume, elevated cardiac output,and hence increased blood pressure.

If renal function is only mildly impaired,blood pressure may stabilize at a level atwhich the higher systemic pressure (andtherefore renal perfusion pressure) enablessodium excretion to balance sodium intake.Conversely, if a patient has end-stage renalfailure, the glomerular filtration rate may beso greatly decreased that the kidney simply


Figure 13.6. The renin-angiotensin-aldosterone system. Liver-derived angiotensinogen iscleaved in the circulation by renin (of kidney origin) to form angiotensin I (AI). AI is rapidly con-verted to the potent vasoconstrictor angiotensin II (AII) by angiotensin-converting enzyme. AII alsomodulates the release of aldosterone from the adrenal cortex. Aldosterone in turn acts to reab-sorb Na+ from the distal nephron, resulting in increased intravascular volume. The other listed effects of AII receptor stimulation may also contribute to the development and maintenance ofhypertension.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 320

Hypertension 321

cannot excrete sufficient volume, and malig-nant-range blood pressures may follow. Notethat renal parenchymal disease may con-tribute to hypertension even if the glome-rular filtration rate is not greatly reduced,through the excessive elaboration of renin.

Renovascular Hypertension

Stenosis of one or both renal arteries leads tohypertension. Although emboli, vasculitis,and external compression of the renal arte-ries can be responsible, the two most com-mon causes of renovascular hypertension(RH) are atherosclerosis and fibromusculardysplasia. Atherosclerotic lesions arise fromextensive plaque formation either withinthe renal artery or in the aorta at the originof the renal artery. This form accounts forabout two thirds of cases of RH and occursmost commonly in elderly men. In contrast,fibromuscular lesions consist of discrete re-gions of fibrous or muscular proliferation,generally within the arterial media. Fibro-muscular dysplasia accounts for one third ofcases of RH and characteristically occurs inyoung women.

The elevated blood pressure in RH arisesfrom reduced renal blood flow to the af-fected kidney, which responds to the lowerperfusion pressure by secreting renin. Thelatter raises the blood pressure through thesubsequent actions of angiotensin II (vaso-constriction) and aldosterone (sodium re-tention), as shown in Figure 13.6.

The diagnosis of RH is suggested by anabdominal bruit, which can be identifiedin 40% to 60% of patients, or by the pres-ence of unexplained hypokalemia (owingto excessive renal excretion of potassium asa result of elevated aldosterone levels). RHis a correctable form of hypertension thatoften is treated successfully by percutaneouscatheter interventions or surgical recon-struction of the stenosed vessel. Medicaltherapy, particularly with angiotensin-converting enzyme (ACE) inhibitors, canalso be effective initial therapy in patientswith unilateral renal artery disease. ACE in-hibitors negate the hypertensive effects ofelevated circulating renin in this situation

by impeding the formation of angiotensinII (see Chapter 17).

Mechanical Causes

Coarctation of the Aorta

Coarctation is an infrequent congenital nar-rowing of the aorta typically located just dis-tal to the origin of the left subclavian artery(see Chapter 16). As a result of the relativeobstruction to flow, the blood pressure inthe aortic arch, head, and arms is higher thanthat in the descending aorta and its branchesand in the lower extremities. Sometimes thecoarctation involves the origin of the leftsubclavian artery, causing lower pressure inthe left arm than in the right.

Hypertension in this condition arises bytwo mechanisms. First, reduced blood flow tothe kidneys stimulates the renin-angiotensinsystem, resulting in vasoconstriction (via an-giotensin II). Second, high pressures proxi-mal to the coarctation stiffen the aortic archthrough medial hyperplasia and acceleratedatherosclerosis, blunting the normal baro-receptor response to elevated intravascularpressure.

Clinical clues to the presence of coarcta-tion include symptoms of inadequate bloodflow to the legs or left arm, such as claudi-cation or fatigue, or the finding of weakenedor absent femoral pulses. A midsystolic mur-mur associated with the stenotic segment ofthe aorta may be auscultated, especially overthe back, between the scapulae. The chestradiograph may show indentation of theaorta at the level of the coarctation. It mayalso demonstrate a notched appearance ofthe ribs secondary to the enlargement of col-lateral intercostal arteries, which shunt bloodaround the aortic narrowing. Treatment op-tions include angioplasty or surgery to cor-rect the stenosis. However, the hypertensionmay not abate completely after mechanicalcorrection, possibly because of persistent de-sensitization of the arterial baroreceptors.

Endocrine Causes

Circulating hormones play an importantrole in the control of normal blood pressure,

10090-13_CH13.qxd 8/31/06 4:55 PM Page 321

so it should not be surprising that endocrinediseases may cause hypertension. When suspected, the presence of such endocrineconditions is evaluated in four ways:

1. History of characteristic signs and symp-toms

2. Measurement of hormone levels3. Assessment of hormone secretion in res-

ponse to stimulation or inhibition4. Imaging studies to identify tumors sec-

reting the excessive hormone

Pheochromocytoma

Pheochromocytomas are catecholamine-secreting tumors of neuroendocrine cells(usually in the adrenal medulla) that causeapproximately 0.2% of cases of hyperten-sion. The release of epinephrine and norepi-nephrine by the tumor results in intermittentor chronic vasoconstriction, tachycardia,and other sympathetic-mediated effects. A characteristic presentation consists ofparoxysmal rises in blood pressure accom-panied by “autonomic attacks” caused bythe increased catecholamine levels: severethrobbing headaches, profuse sweating, pal-pitations, and tachycardia. Although somepatients are actually normotensive betweenattacks, most have sustained hypertension.Ten percent of pheochromocytomas aremalignant.

Determination of plasma catecholaminelevels, or urine catecholamines and theirmetabolites (e.g., vanillylmandelic acid andmetanephrine), obtained under controlledcircumstances, are used to identify this con-dition. Because some pheochromocytomassecrete only episodically, diagnosis may re-quire measurement of catecholamines im-mediately following an attack.

Pharmacologic therapy of pheochromo-cytomas includes the combination of an α-receptor blocker (e.g., phenoxybenzamine)combined with a β-blocker. However, oncethe tumor is localized by computed tomog-raphy, magnetic resonance imaging, or an-giography, the definitive therapy is surgicalresection. For patients with inoperable dis-ease, treatment consists of α- and β-blockade

as well as drugs that inhibit catecholaminebiosynthesis (e.g., α-methyltyrosine).

Adrenocortical Hormone Excess

Among the hormones produced by the ad-renal cortex are mineralocorticoids andglucocorticoids. Excess of either of these canresult in hypertension.

Mineralocorticoids, primarily aldos-terone, increase blood volume by stimulat-ing reabsorption of sodium into the circula-tion by the distal portions of the nephron.This occurs in exchange for potassium ex-cretion into the urine, and the resulting hy-pokalemia is an important marker of miner-alocorticoid excess. Primary aldosteronism,found in approximately 1% to 2% of hyper-tensive patients, is generally the result of anadrenal adenoma (termed Conn syndrome),but may also result from bilateral hyperpla-sia of the adrenal glands. Because the diseasemay be asymptomatic, diagnosis relies onthe detection of hypokalemia and is con-firmed by measurement of excessive aldos-terone secretion and suppressed plasmarenin levels. Therapy includes either surgicalremoval of the adenoma or medical man-agement with aldosterone receptor antago-nists (e.g., spironolactone). Glucocorticoid-remediable aldosteronism (GRA), an uncom-mon hereditary (autosomal dominant) formof primary aldosteronism, results from a ge-netic rearrangement in which aldosteronesynthesis abnormally comes under the reg-ulatory control of adrenocorticotropic hor-mone (ACTH). This condition typically pre-sents as severe hypertension in childhood oryoung adulthood, as opposed to more com-mon forms of primary aldosteronism, whichare generally diagnosed in the third throughsixth decades. Unlike other forms of hyper-tension, GRA-related blood pressure ele-vation responds to glucocorticoid therapy,which suppresses ACTH release from the pi-tuitary gland.

Secondary aldosteronism can result fromincreased angiotensin II production stimu-lated by the rare renin-secreting tumor. Morecommonly, secondary elevation of aldos-terone is a result of augmented circulating


10090-13_CH13.qxd 8/31/06 4:55 PM Page 322

Hypertension 323

angiotensin II in women taking oral contra-ceptives (which stimulate hepatic produc-tion of angiotensinogen, as described ear-lier) or because of impaired angiotensin IIdegradation in chronic liver diseases.

Glucocorticoids, such as cortisol, elevateblood pressure when present in excessamounts, likely via blood volume expansionand stimulated synthesis of components ofthe renin-angiotensin system. In addition,though mineralocorticoids are more potentactivators of mineralocorticoid receptors inthe renal tubules, excess glucocorticoids mayalso activate them.

Nearly 80% of patients with Cushing syn-drome, a disorder of glucocorticoid excess,have some degree of hypertension. These pa-tients often present with classic “cushin-goid” features: a characteristic rounded facialappearance, central obesity, proximal mus-cle weakness, and hirsutism. The cause of theexcess glucocorticoids may be either a pitu-itary ACTH–secreting adenoma, a peripheralACTH–secreting tumor (either of which causesadrenal cortical hyperplasia), or an adrenalcortisol–secreting adenoma. The diagnosis ofCushing syndrome is made by a 24-hoururine collection for the measurement ofcortisol, or by a dexamethasone test, whichevaluates whether exogenous glucocorti-coids can suppress cortisol secretion.

Thyroid Hormone Abnormalities

Approximately one third of hyperthyroid andone fourth of hypothyroid patients have sig-nificant hypertension. Thyroid hormonesexert their cardiovascular effects by (1) in-ducing sodium-potassium ATPases in theheart and vessels; (2) increasing blood vol-ume; and (3) stimulating tissue metabolismand oxygen demand, with secondary accu-mulation of metabolites that modulate localvascular tone. Hyperthyroid patients de-velop hypertension through cardiac hyper-activity and an increase in blood volume.Hypothyroid patients demonstrate predom-inantly diastolic hypertension and an increase in peripheral vascular resistance.Though the precise mechanism is unclear,the latter effect appears to be mediated by

sympathetic and adrenal activation in hy-pothyroidism.

CONSEQUENCES OFHYPERTENSION

Whatever the cause of blood pressure eleva-tion, the ultimate consequences are similar.High blood pressure itself is generally asymp-tomatic but can result in devastating effectson many organs, especially the blood vessels,heart, kidney, and retina.

Clinical Signs and Symptoms

In the past, “classic” symptoms of hyper-tension were considered to include head-ache, epistaxis (nose bleeds), and dizziness.The usefulness of these symptoms has beencalled into question, however, by studiesthat indicate that they are found no morefrequently among hypertensive patientsthan in the general population. Other symp-toms, such as flushing, sweating, and blurredvision, do seem more common in the hy-pertensive population. In general, however,most hypertensive patients are asympto-matic and are diagnosed simply by bloodpressure measurement during routine phys-ical examinations.

Several physical signs of hypertensiondiscussed in the following sections directlyresult from elevated pressure, including leftventricular hypertrophy and retinopathy. Inaddition, hypertension complicated by ather-osclerosis can manifest by arterial bruits, par-ticularly in the carotid and femoral arteries.

Organ Damage Caused by Hypertension

Target organ complications of hypertensionreflect the degree of chronic blood pressureelevation. Such organ damage can be attrib-uted to (1) the increased workload of theheart and (2) arterial damage resulting fromthe combined effects of the elevated pres-sure itself (weakened vessel walls) and accel-erated atherosclerosis (Fig. 13.7). Abnormal-ities of the vasculature that result fromelevated pressure include smooth muscle

Fig. 7

10090-13_CH13.qxd 8/31/06 4:55 PM Page 323

hypertrophy, endothelial cell dysfunction,and fatigue of elastic fibers. Chronic hyper-tensive trauma to the endothelium pro-motes atherosclerosis possibly by disruptingnormal protective mechanisms, such as thesecretion of nitric oxide. Arteries lined byatherosclerotic plaque may thrombose ormay serve as a source of cholesterol embolithat occlude distal vessels, causing organ in-farction (such as cerebrovascular occlusion,resulting in stroke). In addition, atheroscle-rosis of large arteries hinders their elasticity,resulting in systolic pressure spikes that canfurther traumatize endothelium or provokeevents such as aneurysm rupture.

The major target organs for the destruc-tive complications of chronic hypertensionare the heart, the cerebrovascular system,the aorta and peripheral vascular system,the kidney, and the retina (Table 13.3). Leftuntreated, approximately 50% of hyperten-sive patients die of coronary artery diseaseor congestive heart failure, about 33% suc-cumb to stroke, and 10% to 15% die fromcomplications of renal failure.

Heart

The major cardiac effects of hypertension re-late to the increased afterload against which

the heart must contract and accelerated ath-erosclerosis within the coronary arteries.

Left Ventricular Hypertrophy andDiastolic Dysfunction

The high arterial pressure (heightened after-load) increases the wall tension of the leftventricle, which compensates through hyper-trophy. Concentric hypertrophy (without di-latation) is the normal pattern of compen-


Figure 13.7. Pathophysiology of the major consequences of hypertension. LVH, left ventricular hypertrophy.

TABLE 13.3. Target Organ Damage in Hypertension

Organ System Manifestations

Heart • Left ventricular hyper-trophy

• Heart failure• Myocardial ischemia and

infarctionCerebrovascular • StrokeAorta and peripheral • Aortic aneurysm and/or

vascular dissection• Arteriosclerosis

Kidney • Nephrosclerosis• Renal failure

Retina • Arterial narrowing• Hemorrhages, exudates,

papilledema

Tab. 13

10090-13_CH13.qxd 8/31/06 4:55 PM Page 324

Hypertension 325

sation, although conditions that elevateblood pressure by virtue of increased circu-lating volume (e.g., primary aldosteronism)may instead cause eccentric hypertrophywith chamber dilatation (see Chapter 9).Left ventricular hypertrophy (LVH) resultsin increased stiffness of the left ventriclewith diastolic dysfunction, manifest by ele-vation of LV filling pressure that can resultin pulmonary congestion.

Physical findings of LVH may include a heaving LV impulse on chest palpation,indicative of increased muscle mass. It isfrequently accompanied by a fourth heartsound, as the left atrium contracts into thestiffened left ventricle (see Chapter 2).

LVH is one of the strongest predictors ofcardiac morbidity in hypertensive patients.The degree of hypertrophy correlates withthe development of congestive heart failure,angina, arrhythmias, myocardial infarction,and sudden cardiac death.

Systolic Dysfunction

Although LVH initially serves a compen-satory role, later in the course of systemichypertension, the increased LV mass may beinsufficient to balance the high wall tensioncaused by the elevated pressure. As LV con-tractile capacity deteriorates, findings of sys-tolic dysfunction become evident (i.e., re-duced cardiac output and pulmonarycongestion). Systolic dysfunction is also pro-voked by the accelerated development ofcoronary artery disease with resultant peri-ods of myocardial ischemia.

Coronary Artery Disease

Chronic hypertension is a major contribu-tor to the development of myocardial is-chemia and infarction. These complicationsreflect the combination of accelerated coro-nary atherosclerosis (decreased myocardialoxygen supply) and the high systolic work-load (increased oxygen demand). Not onlyis acute myocardial infarction more com-mon among hypertensive patients thanamong normotensive people, but also theformer also have a higher incidence of post-

myocardial infarction complications such asrupture of the ventricular wall, LV aneurysmformation, and congestive heart failure. Infact, 60% of patients who die of transmuralmyocardial infarctions have a history of hy-pertension.

Cerebrovascular System

Hypertension is the major modifiable riskfactor for strokes, also termed cerebrovas-cular accidents (CVAs). Although diastolicpressure is important, it is the magnitude ofthe systolic pressure that has been mostclosely linked to CVAs. The presence of iso-lated systolic hypertension more than dou-bles a person’s risk for this complication.

Hypertension-induced strokes can be he-morrhagic or, more commonly, athero-thrombotic. Hemorrhagic CVAs result fromrupture of microaneurysms induced in cere-bral parenchymal vessels by long-standinghypertension. Atherothrombotic (also calledthromboembolic) CVAs arise when portionsof atherosclerotic plaque within the carotidsor major cerebral arteries, or thrombi thatform on those plaques, break off and em-bolize to smaller distal vessels. Additionally,intracerebral vessels may directly occlude by local atherosclerotic plaque rupture andthrombosis.

Occlusion of small penetrating brain ar-teries can result in multiple tiny infarcts. Asthese lesions soften and are absorbed byphagocytic cells, small (≤3 mm diameter)cavities form, termed lacunae. These lacunarinfarctions are seen almost exclusively inpatients with long-standing hypertensionand are usually localized to the penetratingbranches of the middle and posterior circu-lation of the brain.

The generalized arterial narrowing foundin hypertensive patients reduces collateralflow to ischemic tissues and imposes struc-tural requirements for higher perfusionpressure to maintain adequate tissue flow.This leaves the hypertensive patient vulner-able to cerebral infarcts in areas supplied bythe distal ends of arterial branches (“water-shed” infarcts) if blood pressure should fallsuddenly.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 325

Effective treatment of hypertension di-minishes the risk of stroke and has con-tributed to a 50% reduction in deaths at-tributed to cerebrovascular events in recentdecades.

Aorta and Peripheral Vasculature

The accelerated atherosclerosis associatedwith hypertension may result in plaque for-mation and narrowing throughout the arte-rial vasculature. In addition to the coronaryarteries, lesions most commonly appear with-in the aorta and the major arteries to thelower extremities, neck, and brain.

Chronic hypertension may lead to thedevelopment of aneurysms, particularly ofthe abdominal aorta (see Chapter 15). Anabdominal aortic aneurysm (AAA) repre-sents prominent dilatation of the vessel,usually located below the level of the renalarteries, aided by the mechanical stress of thehigh pressure on an arterial wall alreadyweakened by medial damage and atheroscle-rosis. Aneurysms greater than 6 cm in diam-eter have a very high likelihood of rupturewithin 2 years if not surgically corrected.

Another life-threatening vascular conse-quence of high blood pressure is aortic dis-section (see Chapter 15). Elevated bloodpressure, especially in the highest ranges, ac-celerates degenerative changes in the mediaof the aorta. When the weakened wall is fur-ther exposed to high pressure, the intimamay tear, allowing blood to dissect into theaortic media and propagate in either direc-tion within the vessel wall, “clipping off”and obstructing major branch vessels alongthe way (e.g., coronary or carotid arteries).The mortality rate of aortic dissection isgreater than 90% unless treated emergently,usually by surgical repair if the proximalaorta is involved. Rigorous control of hyper-tension is essential.

Kidney

Hypertension-induced kidney disease (neph-rosclerosis) is a leading cause of renal failurethat results from damage to the organ’s vas-culature. Histologically, the vessel walls be-

come thickened with a hyaline infiltrate,known as hyaline arteriolosclerosis (Fig.13.8). Greater levels of hypertension can in-duce smooth muscle hypertrophy and evennecrosis of capillary walls, termed fibrinoidnecrosis. These changes result in reducedvascular supply and subsequent ischemic at-rophy of tubules and, to a lesser extent,glomeruli. Because intact nephrons can usually compensate for those damaged bypatchy ischemia, mild hypertension rarelyleads to renal insufficiency in the absence ofother insults to the kidney. However, malig-nant levels of hypertension can inflict per-manent damage to the point that dialysisbecomes necessary.

One of the consequences of hypertensiverenal failure is perpetuation of elevated bloodpressure. For example, progressive renal dys-function compromises the ability of the kid-ney to regulate blood volume, which con-tributes further to chronic hypertension.

Retina

The retina is the only location where sys-temic arteries can be directly visualized by physical examination. High blood pres-sure induces abnormalities that are collec-tively termed hypertensive retinopathy.Although vision may be compromisedwhen the damage is extensive, more com-monly the changes serve as an asympto-matic clinical marker for the severity of hy-pertension and its duration.

Severe hypertension that is acute in onset(e.g., uncontrolled and/or malignant hyper-tension) may burst small retinal vessels,causing hemorrhages, exudation of plasmalipids, and areas of local infarction. If ische-mia of the optic nerve develops, patientsmay describe generalized blurred vision.Retinal ischemia caused by hemorrhageleads to more patchy loss of vision. Papil-ledema, or swelling of the optic disc withblurring of its margins, may arise from highintracranial pressure when the blood pres-sure reaches malignant levels and cerebro-vascular autoregulation begins to fail.

Chronically elevated blood pressure resultsin a different set of retinal findings. Pa-


Fig. 8

10090-13_CH13.qxd 8/31/06 4:55 PM Page 326

Hypertension 327

pilledema is absent, but vasoconstrictionresults in arterial narrowing, and medial hy-pertrophy thickens the vessel wall, which“nicks” (indents) crossing veins. With moresevere chronic hypertension, arterial sclero-sis is evident as an increased reflection oflight through the ophthalmoscope (termed“copper” or “silver” wiring. Although thesechanges are not in themselves of majorfunctional importance, they indicate thatthe patient has had long-standing, poorlycontrolled hypertension.

HYPERTENSIVE CRISIS

A hypertensive crisis is a medical emergencycharacterized by a severe elevation of bloodpressure. In the past, this type of elevationwas usually a consequence of inadequateblood pressure treatment. Now a hyperten-

sive crisis is more often caused by an acutehemodynamic insult (e.g., acute renal dis-ease) superimposed on a chronic hyperten-sive state. As a result of rapid pathologicchanges (fibrinoid necrosis) within the bloodvessels and kidney, a spiraling increase inblood pressure evolves. Further volume ex-pansion and vasoconstriction occur as renalperfusion drops and serum renin and an-giotensin levels rise.

Severe blood pressure elevation results inincreased intracranial pressure, and patientsmay present with hypertensive encepha-lopathy manifested by headache, blurredvision, confusion, somnolence, and some-times coma. When hypertension results inacute damage to retinal vessels, accelerated-malignant hypertension is said to be pre-sent. Funduscopic examination shows theeffects of the rapid pressure rise as hemor-

Figure 13.8. Histologic effects of chronic hypertension on the kidney.The arteriolar walls are thickened by hyaline infiltrate (short arrows). Theglomeruli (long arrow) appear partially sclerosed because of reduced vascu-lar supply. (Courtesy of Dr. Helmut G. Rennke, Brigham and Women’s Hos-pital, Boston, MA.)

10090-13_CH13.qxd 8/31/06 4:55 PM Page 327

rhages, exudates, and sometimes papilledema.The increased load on the left ventricle dur-ing a hypertensive crisis may precipitateangina (because of increased myocardial oxy-gen demand) or pulmonary edema.

A hypertensive crisis requires rapid ther-apy to reduce blood pressure and preventpermanent vascular complications. Correc-tion of blood pressure is generally followedby reversal of the acute pathologic changes,including papilledema and retinal exuda-tion, although renal damage often persists.

TREATMENT OF HYPERTENSION

The therapeutic approach to the hyperten-sive patient should be tempered by two con-siderations. First, a single elevated blood pres-sure measurement does not establish thediagnosis of hypertension because bloodpressure varies considerably from day to day.Moreover, blood pressure measurement inthe hospital or doctor’s office may be affectedby the “white coat” effect resulting from pa-tient anxiety. The average of multiple read-ings taken at two or three office visits and/orin the home provides a more reliable basis forlabeling a patient as hypertensive. There isalso evidence that automatic ambulatoryblood pressure measurements, taken over thecourse of 24 hours while the patient followsa daily routine, are more predictive of cardio-vascular mortality than traditional in-clinicmeasurements.

Second, although even mild hyperten-sion is a major public health problem be-cause of its widespread prevalence, for theperson with stage 1 hypertension (see Table13.1), the risks are small. For example, theadditional risk of a stroke is approximately 1in 850 per year. Hence, observation overtime to determine whether the low-level hypertension persists, or whether lifestylechanges can reduce the pressure, is often a recommended alternative to immediatedrug therapy. This is especially true in theabsence of other cardiovascular risk factorssuch as smoking, diabetes, or high serumcholesterol. However, for patients with es-tablished cardiovascular disease or for thosewho have other major atherosclerotic risk

factors, a more aggressive approach to phar-macologic therapy is usually warranted toreduce the risk burden.

For most hypertensive patients, drug the-rapy is ultimately the most effective way to prevent future complications, but thatshould not deter consideration of other ben-eficial lifestyle changes.

Nonpharmacologic Treatment

Weight Reduction

Studies have consistently found obesity andhypertension to be highly correlated, espe-cially when the obesity is of a central (ab-dominal) distribution. Blood pressure reduc-tion follows weight loss in a large portion ofhypertensive patients who are more than10% above their ideal weights. Each 10 kg ofweight loss is associated with a 5 to 20 mmHg fall in systolic blood pressure.

Exercise

Sedentary normotensive people have a 20%to 50% higher risk of developing hyperten-sion than their more active peers. Regularaerobic exercise, such as walking, jogging, orbicycling, has been shown to contribute toblood pressure reduction over and aboveany resulting weight loss. A hypertensive pa-tient who becomes physically conditionedmanifests a lower resting heart rate and re-duced levels of circulating catecholaminesthan before training, suggesting a fall in sym-pathetic tone.

Diet

In addition to caloric restriction for weightloss, changes in the composition of a pa-tient’s diet may be important for blood pres-sure reduction. For example, a diet high infruits, vegetables, and low-fat dairy productshas been shown to significantly reduce bloodpressure.

Sodium

Salt restriction for people with high bloodpressure is a controversial issue, but there


10090-13_CH13.qxd 8/31/06 4:55 PM Page 328

Hypertension 329

are several epidemiologic and clinical trialsthat support the benefit of moderating so-dium intake. In normotensive persons, ex-cess salt ingestion is simply excreted by thekidneys, but approximately 50% of patientswith essential hypertension are found tohave blood pressures that vary with sodiumintake, suggesting a defect in natriuresis.Sensitivity to sodium levels is more com-mon in African American and elderly hy-pertensive patients. Because low-salt dietstend to increase the effectiveness of antihy-pertensive medications in general, the cur-rent recommendation is to limit salt intaketo <6 g of sodium chloride (<2.3 g sodium)per day, which is one-third less than the average U.S. consumption.

Potassium

Total body potassium content tends to de-crease when a person eats a diet low in fruitsand vegetables or takes potassium-wastingdiuretics. Potassium deficiency has severaltheoretical effects that may raise blood pres-sure and contribute to adverse cardiovascu-lar outcomes, such that dietary supplementsare routinely recommended to help repletelow serum K+ levels. There is no convincingevidence that prescribing potassium supple-ments to a normokalemic hypertensive pa-tient will lower blood pressure.

Alcohol

The chronic intake of alcoholic beveragescorrelates with high blood pressure and re-sistance to antihypertensive medications.Moreover, experimental evidence shows thatblood pressure (especially systolic) may riseacutely following alcohol consumption. Thereason for this link remains incompletely un-derstood, but decreasing chronic alcohol in-take has been shown to lower blood pressure.

Other

Low calcium intake and magnesium deple-tion have been associated with elevatedblood pressure, but the responsible mecha-nisms and the implications for therapy are

unclear. Caffeine ingestion transiently in-creases blood pressure (as much as 5 to 15mm Hg after two cups of coffee), but routineuse does not seem to produce chronic pres-sure elevation.

Smoking

Cigarette smoking transiently increases bloodpressure, likely because of the effect of nico-tine on the autonomic ganglia, and is a riskfactor for the development of sustained hy-pertension. In addition, the atherogenic effect of smoking may contribute to the de-velopment of renovascular hypertension.Cigarette usage is associated with many otherhealth hazards, and all patients should bediscouraged from smoking.

Relaxation Therapy

Blood pressure frequently rises under condi-tions of stress. In addition, essential hyper-tensive patients and their relatives oftenshow higher than normal basal sympathetictone and exaggerated autonomic responsesto mental stress. Hence, relaxation tech-niques have been advocated as a method tocontrol hypertension. Available methods in-clude biofeedback and meditation. The ef-fectiveness of such therapy has not beenconsistently demonstrated in clinical trialsand seems to depend on the patient’s atti-tude and long-term compliance.

In summary, nonpharmacologic therapyoffers a wide range of options that do nothave the expense and potential side effects ofprescribed drug use. The effectiveness ofthese therapies should come as no surprise,given the extent to which environmental fac-tors play a role in hypertension. Therefore,behavior-based interventions are recom-mended as first-line therapy in any patientwhose hypertension is not an immediatedanger to life and well-being.

Pharmacologic Treatment

Antihypertensive medications are the stan-dard means to lower chronically elevated

10090-13_CH13.qxd 8/31/06 4:55 PM Page 329

blood pressure and are indicated if nonphar-macologic treatment proves inadequate. Morethan 100 drug preparations are available totreat hypertension, but fortunately the mostcommonly used medications fall into fourclasses: diuretics, sympatholytics, vasodila-tors, and drugs that interfere with the renin-angiotensin system (Table 13.4). The individ-ual actions of these groups on the physiologicabnormalities in hypertension are shown in Figure 13.9. The pharmacology and use ofthe antihypertensive drugs are described ingreater detail in Chapter 17.

Diuretics have been used for decades totreat hypertension. They reduce circulatoryvolume, cardiac output, and mean arterialpressure and are most effective in patientswith mild-to-moderate hypertension whohave normal renal function. They are espe-cially effective in African American or elderlypersons, who tend to be salt sensitive. In clin-ical trials, diuretics have reduced the risk ofstrokes and cardiovascular events in hyper-tensive patients and are inexpensive com-pared with other agents. Thiazide diuretics(e.g., hydrochlorothiazide) and potassium-sparing diuretics (e.g., spironolactone) pro-mote Na+ and Cl− excretion in the nephron.Loop diuretics (e.g., furosemide) are gener-ally too potent and their actions too shortlived for use as antihypertensive agents;however, they are useful in lowering bloodpressure in patients with renal insufficiency,who often do not respond to other diuretics.

Diuretics may result in adverse metabolicside effects, including elevation of serum

glucose, cholesterol, and triglyceride levels.In addition, hypokalemia, hyperuricemia,and decreased sexual function are potentialside effects. However, when diuretics areprescribed in low dosages, it is often possibleto accrue the desired antihypertensive effectwhile minimizing adverse complications.

Sympatholytic agents include (1) β-blockers, (2) central α-adrenergic agonists,and (3) systemic α-adrenergic-blocking drugs.b-Blockers, such as propranolol, are be-lieved to lower blood pressure through sev-eral mechanisms, including (1) reducingcardiac output through a decrease in heartrate and mild decrease in contractility and(2) decreasing the secretion of renin (andtherefore levels of angiotensin II), whichleads to a decrease in total peripheral resis-tance. β-Blockers are less effective than di-uretics in elderly and African American hypertensive patients. Adverse effects of β-blockers include bronchospasm (becauseof bronchiolar β2-receptor blockade), fa-tigue, impotence, and hyperglycemia. Theymay also adversely alter lipid metabolism.Most β-blockers cause an increase in serumtriglyceride levels and a decrease in “good”HDL cholesterol levels. However, β-blockerswith intrinsic sympathomimetic activity(see Chapter 17) or those with combined β-blocking properties (such as labetalol) donot adversely affect HDL levels.

Centrally acting a2-adrenergic agonists,such as methyldopa and clonidine, reducesympathetic outflow to the heart, blood ves-sels, and kidneys. These are rarely used owing


TABLE 13.4. Classes of Antihypertensive Medications

Drug Class Types (see Chapter 17)

Diuretics ThiazidesAldosterone-antagonists

Sympatholytics β-BlockersCombined α- and β-blockersCentral α2-agonistsPeripheral α1-blockers

Vasodilators Calcium channel blockersDirect vasodilators (e.g., hydralazine, minoxidil)

Renin-angiotensin system antagonists Angiotensin-converting enzyme inhibitorsAngiotensin II receptor blockers

Tab. 4

Fig. 9

10090-13_CH13.qxd 8/31/06 4:55 PM Page 330

Hypertension 331

to their high frequency of side effects (e.g.,dry mouth, sedation). Systemic a1-antago-nists, such as prazosin, terazosin, and doxa-zosin, cause a decrease in total peripheral resistance through relaxation of vascularsmooth muscle. They may be beneficial forhypertension in some older men because thedrugs also improve symptoms of prostatic en-largement. However, they are otherwise notoften recommended because a major clinicaltrial showed that an α-blocker was associatedwith a greater number of adverse cardio-vascular events compared with a diuretic.

Peripheral vasodilators include calciumchannel blockers, hydralazine, and minoxi-dil. Calcium channel blockers reduce theinflux of Ca++ responsible for cardiac andvascular smooth muscle contraction, thus

reducing cardiac contractility and total pe-ripheral resistance (see Chapter 17). Clinicaltrials in patients with hypertension haveshown that calcium channel blockers re-duce the risk of myocardial infarction andstroke. Thus long-acting (i.e., sustained-release drugs taken once a day) members ofthis group are frequently used to treat hypertension. The shorter-acting calciumchannel blocker preparations are no longerused for this purpose; they are less conve-nient and have been associated with adversecardiovascular outcomes (see Chapter 6). Hydralazine and minoxidil lower bloodpressure by directly relaxing vascular smoothmuscle of precapillary resistance vessels.However, this action can result in a reflex increase in heart rate, so that combined

Figure 13.9. Physiologic effects of antihypertensive medications. Notice that some antihypertensive medicationswork at multiple sites. CC, cardiac contractility; CCB, calcium++ channel blockers; HR, heart rate; RAS blockers, renin-angiotensin system blockers (i.e., angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers); SV,stroke volume; VR, venous return.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 331

β-blocker therapy is frequently necessary.The use of these direct vasodilators in treat-ing hypertension has waned with the ad-vent of newer agents with fewer side effects.

Drugs that interfere with the renin-angiotensin system include ACE inhibitorsand angiotensin II receptor blockers. ACE in-hibitors decrease blood pressure by blockingthe conversion of angiotensin I to angio-tensin II (see Fig. 17.6), thereby reducing thevasopressor activity of angiotensin II and thesecretion of aldosterone. Thus, reductions intotal peripheral resistance and sodium reten-tion by the kidney occur. An additional anti-hypertensive effect of ACE inhibitors is to in-crease the concentration of the circulatingvasodilator bradykinin (see Fig. 17.6). ACEinhibitors are important drugs that havebeen shown to reduce mortality rates in pa-tients following an acute myocardial infarc-tion, in patients with chronic symptomaticsystolic heart failure (see Chapter 9), andeven in people at high risk for developingcardiovascular disease. The drugs also slowthe deterioration of renal function in pa-tients with diabetic nephropathy. The mostcommon side effect of ACE inhibitors is thedevelopment of a reversible dry cough (likelyrelated to the increased bradykinin effect);hyperkalemia and azotemia may also occur,as described in Chapter 17.

Angiotensin II receptor blockers (ARBs)are the most recently introduced class of antihypertensive agents. The action of thisgroup is to block the binding of angiotensinII to its receptors (i.e., subtype AT1 receptors)in blood vessels and other targets (see Fig.17.6). By inhibiting the effects of angiotensinII (and thereby causing vasodilatation and re-duced secretion of aldosterone), blood pres-sure falls. In clinical trials, the antihyperten-sive efficacy of this group is similar to that ofACE inhibitors. However, they are very welltolerated, and unlike ACE inhibitors, coughis not a common side effect. ARBs have beenshown to reduce cardiovascular event rates(including myocardial infarction and stroke)in high-risk patients.

Given the large number of effective anti-hypertensive drugs available, the choice of

which drug to use as initial therapy in an in-dividual patient can seem daunting. Thefirst-line drugs for uncomplicated hyperten-sion recommended by the Joint NationalCommittee on Detection, Evaluation, andTreatment of High Blood Pressure are thi-azide diuretics because of their proven long-term benefits at reducing morbidity andmortality as well as their low cost. In certaincircumstances, or if initial therapy with a di-uretic is not sufficient, an ACE inhibitor, an-giotensin receptor blocker, calcium channelblocker, or β-blocker is substituted or added.For example, an ACE inhibitor should begiven prime consideration in patients withconcurrent heart failure, diabetes, or LVdysfunction following myocardial infarc-tion. A β-blocker would be an appropriatefirst choice in a patient with ischemic heartdisease.

There are some other guiding principles.First, the chosen drug regimen should conform to the patient’s specific needs. For example, an anxious young patient in thethroes of the hyperkinetic phase of EHmight be best treated with a β-blocker,whereas a more effective choice for thesame patient many years later, after thepressure becomes more dependent on pe-ripheral vascular resistance, could be a va-sodilator (e.g., long-acting calcium channelblocker). Because therapy is likely to con-tinue for many years, consideration of adverse effects and impact of drug therapyon the patient’s quality of life are very important.

Another helpful principle of antihyper-tensive drug therapy concerns the use ofmultiple agents. The effects of one drug, act-ing at one physiologic control point, can bedefeated by natural compensatory mecha-nisms. For example, the drop in renal perfu-sion by a direct vasodilator can activate therenin-angiotensin system, prompting the kid-ney to retain more volume, thereby bluntingthe antihypertensive benefit. Combinationdrug therapy is aimed at preventing such anaction by using agents acting at differentcomplementary sites. In this example, a di-rect vasodilator is often paired with a low-


10090-13_CH13.qxd 8/31/06 4:55 PM Page 332

Hypertension 333

dose diuretic to avoid the undesired volumeexpansion effect.

In conclusion, hypertension emerges asa tremendously important clinical problembecause of its prevalence and potentiallydevastating consequences. The evaluationand treatment of a patient with hyperten-sion require methodical consideration ofthe ways in which normal cardiovascularphysiology may have gone awry. Becausemost patients still fall into the idiopathiccategory of EH, there is still much room forcreative thought and research in this area.

SUMMARY

1. Hypertension is defined as a chronic dias-tolic blood pressure ≥90 mm Hg and/orsystolic blood pressure ≥140 mm Hg.

2. Hypertension is of unknown etiology in95% of patients (EH). Secondary hyper-tension may arise from diverse causes, in-cluding (a) renal abnormalities (e.g., renalparenchymal disease and renal arterystenosis); (b) coarctation of the aorta; and(c) endocrine abnormalities (e.g., pheo-chromocytoma, primary or secondary aldosteronism, Cushing syndrome, andthyroid abnormalities).

3. Most hypertensive patients remain asymp-tomatic until complications arise. Poten-tial complications include (a) stroke, (b) myocardial infarction, (c) heart fail-ure, (d) aortic aneurysm and dissection,(e) renal damage, and (f) retinopathy.

4. Treatment of hypertension includes life-style and dietary improvements, followedby pharmacologic therapy. Commonlyused antihypertensive drugs include thi-azide diuretics, β-blockers, ACE inhibi-tors, angiotensin receptor blockers, andlong-acting calcium channel blockers.

Acknowledgment

Contributors to the previous editions of this chapterwere Rahul Deshmukh, MD; Rajesh S. Magrulkar,MD; Rajeev Malhotra, MD; Peter Allison McDo-nough, MD; A. Nigrovic, MD; Thomas J. Moore, MD;and Leonard S. Lilly, MD.

Additional Reading

Agarwal A, Williams GH, Fisher ND. Genetics ofhuman hypertension. Trends Endocrinol Metab2005;16:127–133.

Chobanian AV, Bakris GL, Black HR, et al. The sev-enth report of the joint national committee onprevention, detection, evaluation, and treatmentof high blood pressure: the JNC 7 report. JAMA2003;289:2560–2571.

Dolan E, Stanton A, Thijs L, et al. Superiority of am-bulatory over clinic blood pressure measurementin predicting mortality. Hypertension 2005;46:156–161.

Eckel RH, Grundy SM, Zimmet PZ. The metabolicsyndrome. Lancet 2005;365:1415–1428.

Elley CR, Arroll B. Refining the exercise prescriptionfor hypertension. Lancet 2005;366:1248–1249.

Fisher NDL, Williams GH. Hypertensive vascular dis-ease. In: Kasper DL, Braunwald E, Fauci AS, et al.,eds. Harrison’s Principles of Internal Medicine.16th Ed. New York: McGraw-Hill, 2004:1463–1480.

Giles TD, Berk BC, Black HR, et al. Expanding the de-finition and classification of hypertension. J ClinHypertens 2005;7:505–512.

Hajjar I, Kotchen TA. Trends in prevalence, aware-ness, treatment, and control of hypertension inthe United States, 1988–2000. JAMA 2003;290:199–206.

Haslam DW, James WP. Obesity. Lancet 2005;366:1197–1209.

Kaplan NM. Kaplan’s Clinical Hypertension. 9th Ed.Philadelphia: Lippincott Williams & Wilkins, 2006.

Weber MA. Hypertension treatment and impli-cations of recent cardiovascular outcome trials. J Hypertens 2006;24(suppl 2):S37–S44.

White WB. Update on the drug treatment of hyper-tension in patients with cardiovascular disease.Am J Med 2005;118:695–705.

Williams B. Recent hypertension trials. J Am CollCardiol 2005;45:813–827.

10090-13_CH13.qxd 8/31/06 4:55 PM Page 333


1. AU: The chapter also refers to the renin-angiotensin system. Are both terms correct? Orshould they be consistent?

10090-13_CH13.qxd 8/31/06 4:55 PM Page a

334

ANATOMY AND FUNCTION

ACUTE PERICARDITISEtiologyPathogenesisPathologyClinical FeaturesDiagnostic StudiesTreatment

PERICARDIAL EFFUSIONEtiologyPathophysiologyClinical FeaturesDiagnostic StudiesTreatment

CARDIAC TAMPONADEEtiologyPathophysiologyClinical FeaturesDiagnostic StudiesTreatment

CONSTRICTIVE PERICARDITISEtiology and PathogenesisPathologyPathophysiologyClinical FeaturesDiagnostic StudiesTreatment

C H A P T E R

14Diseases of thePericardiumYanerys RamosLeonard S. Lilly

Diseases of the pericardium form a spectrumthat ranges from benign, self-limited peri-carditis to life-threatening cardiac tamponade.The clinical manifestations of these disordersand the approaches to their management canbe predicted from an understanding of peri-cardial anatomy and pathophysiology, as pre-sented in this chapter.

ANATOMY AND FUNCTION

The pericardium is a two-layered sac that en-circles the heart. The inner serosal layer (vis-

ceral pericardium) adheres to the outer wall ofthe heart and is reflected back on itself, at thelevel of the great vessels, to line the tough fi-brous outer layer (parietal pericardium). A thinfilm of pericardial fluid slightly separates thetwo layers and decreases the friction betweenthem.

The pericardium appears to serve threefunctions: (1) fixing the heart within themediastinum and limits its motion, (2) pre-venting extreme dilatation of the heart dur-ing sudden rises of intracardiac volume, and(3) possibly functioning as a barrier to limit

10090-14_CH14.qxd 8/31/06 4:56 PM Page 334

Diseases of the Pericardium 335

the spread of infection from the adjacentlungs. However, patients with complete ab-sence of the pericardium (either congeni-tally or after surgical removal) are generallyasymptomatic, casting doubt on its actualimportance in normal physiology. Yet likethe unnecessary appendix, the pericardiumcan become diseased and cause great harm.

In the healthy heart, intrapericardial pres-sure varies during the respiratory cycle from−5 mm Hg (during inspiration) to +5 mm Hg(during expiration) and nearly equals thepressure within the pleural space. However,pathologic changes in pericardial stiffness,or the accumulation of fluid within thepericardial sac, may profoundly increase thispressure.

ACUTE PERICARDITIS

The most common affliction of the peri-cardium is acute pericarditis, which refers toinflammation of its layers. Many diseasestates and etiologic agents can produce thissyndrome (Table 14.1), the most common ofwhich are described here.

Etiology

Infectious

Idiopathic and Viral Pericarditis

Acute pericarditis is most often of idiopathicorigin, meaning that the actual cause is unknown. However, serologic studies havedemonstrated that many such episodes areactually caused by viral infection, especially

by echovirus or coxsackievirus group B. Al-though a viral origin could be confirmed ininfected patients by comparing antiviraltiters of acute and convalescent serum, thisis rarely done in the clinical setting becausethe patient has usually recovered by thetime those results would be available. Thus,idiopathic and viral pericarditis are consid-ered similar clinical entities, and the termsare used interchangeably.

Other viruses known to cause pericarditisinclude those responsible for influenza, vari-cella, mumps, hepatitis B, and infectiousmononucleosis. Pericarditis has been foundwith increased frequency among patientswith AIDS, possibly related to HIV itself, butoften owing to superimposed tuberculous orother bacterial infections in this immuno-compromised population.

Tuberculous Pericarditis

Although tuberculosis remains a worldwideproblem, its incidence in the United States islow. It is, however, an important cause ofpericarditis in immunosuppressed patients,such as those with AIDS. Tuberculous peri-carditis arises from reactivation of the organ-ism in mediastinal lymph nodes, with spreadinto the pericardium. It can also extend di-rectly from a site of tuberculosis within thelungs, or the organism can arrive at the peri-cardium by hematogenous dissemination.

Nontuberculous Bacterial Pericarditis(Purulent Pericarditis)

Bacterial pericarditis has become rare sincethe advent of antibiotics. Pneumococcus andstaphylococci are responsible most fre-quently, whereas Gram-negative infectionoccurs less often. Mechanisms by which bac-terial invasion of the pericardium developsinclude (1) perforating trauma to the chest(e.g., stab wound); (2) contamination duringchest surgery; (3) extension of an intracardiacinfection (i.e., infective endocarditis); (4) ex-tension of pneumonia or a subdiaphragmaticinfection; and (5) hematogenous spread froma remote infection. Bacterial pericarditis is afulminant illness but is rare in otherwisehealthy persons; it is most likely to occur in

TABLE 14.1. Most Common Causes of Acute Pericarditis

InfectiousViralTuberculosisPyogenic bacteriaNoninfectiousPostmyocardial infarctionUremiaNeoplastic diseaseRadiation-inducedConnective tissue diseasesDrug induced

Tab. 1

10090-14_CH14.qxd 8/31/06 4:56 PM Page 335

336 Chapter Fourteen

immunocompromised patients, includingthose with severe burns and malignancies.

Noninfectious

Pericarditis Following Myocardial Infarction

There are two forms of pericarditis associatedwith acute myocardial infarction (MI). Theearly form occurs within the first few daysafter an MI. It likely results from inflamma-tion extending from the epicardial surfaceof the injured myocardium to the adjacentpericardium; therefore, it is most common in patients with transmural (as opposed tosubendocardial) infarctions. The prognosisfollowing acute MI is not affected by the pres-ence of pericarditis; its major importance isin distinguishing it from the pain of recur-rent myocardial ischemia. This form of peri-carditis occurs in fewer than 5% of patientswith acute MI who are treated with acutereperfusion strategies (see Chapter 7) but it ismore common in those who are not (andwho therefore sustain larger infarctions).

The second form of post-MI pericarditis isknown as Dressler’s syndrome, which can de-velop 2 weeks to several months followingan acute infarction. Its cause is unknown,but it is thought to be of autoimmune origin,possibly directed against antigens releasedfrom necrotic myocardial cells. A clinicallysimilar form of pericarditis may occur weeksto months following heart surgery (termedpostpericardiotomy pericarditis).

Uremic Pericarditis

Pericarditis is a serious complication of chro-nic renal failure, but its pathogenesis in thissetting is unknown. Studies have shown nocorrelation between the plasma level of ni-trogen waste products and the incidence ofpericarditis, and it may even develop in pa-tients during the first few months of dialysistherapy.

Neoplastic Pericarditis

Tumors within the pericardium most com-monly results from metastatic spread or localinvasion by cancer of the lung, breast, or lym-

phoma. Primary tumors of the pericardiumare rare. Neoplastic effusions are usually largeand hemorrhagic and frequently lead to car-diac tamponade, a life-threatening complica-tion described later in the chapter.

Radiation-Induced Pericarditis

Pericarditis may complicate radiation ther-apy to the thorax (e.g., administered for thetreatment of certain tumors), especially ifthe cumulative dose has exceeded 4,000 rads(40 Gy). Radiation-induced damage causes alocal inflammatory response that can resultin pericardial effusions and ultimately fi-brosis. Cytologic examination of the peri-cardial fluid helps to distinguish radiation-induced pericardial damage from that oftumor invasion.

Pericarditis Associated with ConnectiveTissue Diseases

Pericardial involvement is common in manyconnective tissue diseases, including sys-temic lupus erythematosus (SLE), rheuma-toid arthritis, and progressive systemic scle-rosis. For example, 20% to 40% of patientswith SLE experience clinically detectablepericarditis during the course of the disease.Customary treatment of the underlying con-nective tissue disease usually ameliorates thepericarditis as well.

Drug-Induced Pericarditis

Several pharmaceutical agents have been re-ported to cause pericarditis as a side effect,often by inducing a syndrome similar to SLE(Table 14.2). These drugs include the antiar-rhythmic procainamide and the vasodilatorhydralazine. Drug-induced pericarditis usu-ally abates when the causative agent is dis-continued.

Pathogenesis

Similar to other inflammatory processes,pericarditis is characterized by three stages: (1) local vasodilation with transudation ofprotein-poor, cell-free fluid into the pericar-

Tab. 2

10090-14_CH14.qxd 8/31/06 4:56 PM Page 336

dial space; (2) increased vascular permeability,with leak of protein into the pericardial space;and (3) leukocyte exudation, initially by neu-trophils, followed later by mononuclear cells.

The leukocytes are of critical importancebecause they help contain or eliminate theoffending infectious or autoimmune agent.However, metabolic products released bythese cells may prolong inflammation, causepain and local cellular damage, and mediatesomatic symptoms such as fever. Therefore,the immune response to pericardial injurymay significantly contribute to tissue dam-age and symptomatology.

Pathology

The pathologic appearance of the peri-cardium depends on the underlying causeand severity of inflammation. Serous peri-carditis is characterized by scant polymor-phonuclear leukocytes, lymphocytes, andhistiocytes. The exudate is a thin fluid se-creted by the mesothelial cells lining theserosal surface of the pericardium. This likelyrepresents the early inflammatory responsecommon to all types of acute pericarditis.

Serofibrinous pericarditis is the mostcommonly observed morphologic patternin patients with pericarditis. The pericardialexudate contains plasma proteins, includingfibrinogen, yielding a grossly rough andshaggy appearance (termed “bread and but-ter” pericarditis). Portions of the visceraland parietal pericardium may become thick-

ened and fused. Occasionally, this processleads to a dense scar that restricts movementand diastolic filling of the cardiac chambers,as described later in the chapter.

Suppurative (or purulent) pericarditis isan intense inflammatory response associatedmost commonly with bacterial infection.The serosal surfaces are erythematous andcoated with purulent exudate. Hemorrhagicpericarditis refers to a grossly bloody formof pericardial inflammation and is mostoften caused by tuberculosis or malignancy.

Clinical Features

History

The most frequent symptoms of acute peri-carditis are chest pain and fever (Table 14.3).The pain may be severe and most often local-izes to the retrosternal area and left pre-cordium; it may also radiate to the back andridge of the left trapezius muscle. What dif-ferentiates it from myocardial ischemia or in-farction is that the pain of pericarditis is typi-cally sharp and pleuritic (it is aggravated byinspiration and coughing) and positional (e.g.,sitting and leaning forward often lessen thediscomfort). Dyspnea is common during acutepericarditis but is not exertional and probablyresults from a reluctance of the patient tobreathe deeply because of pleuritic pain.

Patients with idiopathic or viral peri-carditis are typically young and previouslyhealthy. Pericarditis of other causes shouldbe suspected in patients with the underlyingconditions listed in Table 14.1 who developthe typical sharp, pleuritic chest pains andfever.


A scratchy pericardial friction rub is com-mon in acute pericarditis and is produced

TABLE 14.2. Examples of Drug-Induced Pericarditis

Related to drug-induced SLE-like syndromeProcainamideHydralazineMethyldopaIsoniazidPhenytoinNot related to drug-induced SLE-like syndromeAnthracycline antineoplastic agents (doxorubicin,

daunorubicin)Minoxidil

SLE, systemic lupus erythematosus.

TABLE 14.3. Clinical Features of Acute Pericarditis

Pleuritic chest painFeverPericardial friction rubECG abnormalities

Tab. 3


10090-14_CH14.qxd 8/31/06 4:56 PM Page 337


by the movement of the inflamed pericar-dial layers against one another. Ausculta-tion of the rub is best heard using the di-aphragm of the stethoscope with the patientleaning forward while exhaling (whichbrings the pericardium closer to the chestwall and stethoscope). In its full form, therub consists of three components, corre-sponding to the phases of greatest cardiacmovement: ventricular contraction, ventri-cular relaxation, and atrial contraction.Characteristically, the pericardial rub isevanescent, coming and going from one ex-amination to the next.

Diagnostic Studies

The presence of pleuritic, positional chestpain and the characteristic pericardial fric-tion rub implicate the presence of acutepericarditis. However, certain laboratorystudies are helpful to confirm the diagnosisand to assess for impending complications.

The electrocardiogram (ECG) is abnormalin 90% of patients with acute pericarditisand helps to distinguish it from other formsof cardiac disease, such as an acute MI. The

most important ECG pattern, which reflectsinflammation of the adjacent myocardium,consists of diffuse ST segment elevation inmost of the ECG leads, usually with the exception of aVR and V1 (Fig. 14.1). In addi-tion, PR segment depression in several leads is often evident, reflecting abnormal atrialrepolarization related to atrial epicardial inflammation. These abnormalities are incontrast to the ECG of acute ST segment el-evation MI, in which the ST segments are el-evated only in the leads overlying the re-gion of infarction, and PR depression is notexpected.

Blood studies typically reveal signs ofacute inflammation, including an increasedwhite blood cell count (usually a mild lym-phocytosis in acute viral/idiopathic peri-carditis) and elevation of the erythrocytesedimentation rate. Some patients withacute pericarditis also demonstrate elevatedserum cardiac biomarkers (e.g., cardiac tro-ponin I), suggesting inflammation of theneighboring myocardium.

Further testing in acute pericarditis oftenincludes echocardiography to evaluate for thepresence and hemodynamic significance of

Figure 14.1. Electrocardiogram in acute pericarditis. Diffuse ST segment elevation is present. Also noticedepression of the PR segment (arrow). 1 LINE SHORT

10090-14_CH14.qxd 8/31/06 4:56 PM Page 338

Fig. 2


a pericardial effusion. Additional studiesthat may be useful in individual cases, to de-fine the cause of pericarditis include (1) pu-rified protein derivative skin test for tuber-culosis, (2) serologic tests (antinuclearantibodies and rheumatoid factor) to screenfor connective tissue diseases, and (3) a care-ful search for malignancy, especially of thelung and breast (physical examination sup-plemented by chest radiograph and mam-mogram). The yield of diagnostic pericardio-centesis (removal of pericardial fluid througha needle) in uncomplicated acute pericardi-tis is low and should be reserved for patientswith very large effusions or evidence of car-diac chamber compression, as discussed laterin the chapter.

Treatment

Idiopathic or viral pericarditis is a self-lim-ited disease that usually runs its course in 1to 3 weeks. Management consists of rest, toreduce the interaction of the inflamed peri-cardial layers, and pain relief by analgesic andanti-inflammatory drugs (aspirin and othernonsteroidal anti-inflammatory agents). Oralcorticosteroids are often effective for severeor recurrent pericardial pain but should notbe used in uncomplicated cases because ofpotentially severe side effects and becauseeven gradual withdrawal of this form of the-rapy often leads to recurrent symptoms ofpericarditis.

The forms of pericarditis related to myo-cardial infarction are treated in a similarfashion, with rest and aspirin. Other nons-teroidal anti-inflammatory agents are oftenavoided immediately following an MI be-cause of experimental evidence linking themto delayed healing of the infarct.

Purulent pericarditis requires more aggres-sive treatment, including catheter drainageof the pericardium and intensive antibiotictherapy. Nevertheless, even with such ther-apy, the mortality rate is very high. Tuber-culous pericarditis requires prolonged mul-tidrug antituberculous therapy. Pericarditisin the setting of uremia often resolves fol-lowing intensive dialysis. Neoplastic pericar-

dial disease usually indicates widely meta-static cancer, and therapy is unfortunatelyonly palliative.

PERICARDIAL EFFUSION

Etiology

The normal pericardial space contains 15 to50 mL of pericardial fluid, a plasma ultrafil-trate secreted by the mesothelial cells thatline the serosal layer. A larger volume offluid may accumulate in association withany of the forms of acute pericarditis previ-ously discussed.

In addition, noninflammatory serous ef-fusions may result from conditions of (1) in-creased capillary permeability (e.g., severehypothyroidism); (2) increased capillary hy-drostatic pressure (e.g., congestive heart fail-ure); or (3) decreased plasma oncotic pressure(e.g., cirrhosis or the nephrotic syndrome).Chylous effusions may occur in the pres-ence of lymphatic obstruction of pericardialdrainage, most commonly caused by neo-plasms and tuberculosis.

Pathophysiology

Because the pericardium is a relatively stiffstructure, the relationship between its inter-nal volume and pressure is not linear, asshown in curve A of Figure 14.2. Notice thatthe initial portion of the curve is nearly flat,indicating that at the low volumes normallypresent within the pericardium, a small in-crease in volume leads to only a small rise inpressure. However, when the intrapericar-dial volume expands beyond a critical level(see Fig. 14.2, arrow), a dramatic increase inpressure is incited by the nondistensible sac.At that point, even a minor increase in vol-ume can translate into an enormous com-pressive force on the heart.

Three factors determine whether a peri-cardial effusion remains clinically silent orwhether symptoms of cardiac compressionensue: (1) the volume of fluid, (2) the rate atwhich the fluid accumulates, and (3) the com-pliance characteristics of the pericardium.1 LINE SHORT

10090-14_CH14.qxd 8/31/06 4:56 PM Page 339


A sudden increase of pericardial volume, asmay occur in chest trauma with intraperi-cardial hemorrhage, results in marked eleva-tion of pericardial pressure (see Fig. 14.2,steep portion of curve A) and the potentialfor severe cardiac chamber compression.Even lesser amounts of fluid may cause sig-nificant elevation of pressure if the peri-cardium is pathologically noncompliant andstiff, as may occur in the presence of tumoror fibrosis of the sac. In contrast, if the peri-cardial effusion accumulates slowly, overweeks to months, the pericardium graduallystretches, such that the volume-pressure rela-tionship curve shifts toward the right (seeFig. 14.2, curve B). With this adaptation, thepericardium can accommodate larger vol-umes (e.g., 1 to 2 L) without marked eleva-tion of intrapericardial pressure.

Clinical Features

A spectrum of possible symptoms is associ-ated with pericardial effusions. For example,the patient with a large effusion may beasymptomatic, may complain of a dull con-stant ache in the left side of the chest, or may

present with findings of cardiac tamponade,as described later in the chapter. In addition,the effusion may cause symptoms resultingfrom compression of adjacent structures,such as dysphagia (difficult swallowing be-cause of esophageal compression), dyspnea(shortness of breath resulting from lung com-pression), hoarseness (caused by to recurrentlaryngeal nerve compression), or hiccups (re-sulting from phrenic nerve stimulation).

On examination (Table 14.4), a large peri-cardial fluid “insulates” the heart from thechest wall, and the heart sounds may be muf-fled. In fact, a friction rub that had been pre-sent during the acute phase of pericarditismay disappear if a large effusion develops andseparates the inflamed layers from one an-other. Dullness to percussion of the left lungover the angle of the scapula may be present

Figure 14.2. Schematic representation of the volume-pressure re-lationship of the normal pericardium. A. At the very lowest levels, asmall rise in volume results in a small rise in pressure. However, when thelimits of pericardial stretch are reached (arrow), the curve becomes verysteep, and a further small rise in intrapericardial volume results in signifi-cantly increased pressure. B. Chronic slow accumulation of volume allowsthe pericardium to gradually stretch over time; thus, the curve shifts tothe right and much larger volumes are accommodated at lower pressures.(Modified from Freeman GL, LeWinter MM. Pericardial adaptations dur-ing chronic dilation in dogs. Circ Res 1984;54:294.)

TABLE 14.4. Clinical Features of Large Pericardial Effusion

Soft heart soundsReduced intensity of friction rubEwart sign (dullness over posterior left lung)

Tab. 4

10090-14_CH14.qxd 8/31/06 4:56 PM Page 340

(known as the Ewart sign) owing to compres-sive atelectasis by the enlarged pericardial sac.

Diagnostic Studies

The chest radiograph may be normal if only asmall pericardial effusion is present. How-ever, if more than approximately 250 mLhas accumulated, the cardiac silhouette en-larges in a globular, symmetric fashion. Inlarge effusions, the ECG may demonstratereduced voltage of the complexes. In thepresence of extremely large effusions, theheight of the QRS complex may vary frombeat to beat (electrical alternans), a result of aconstantly changing electrical axis as theheart swings from side to side within thelarge pericardial volume.

One of the most useful laboratory tests inthe evaluation of an effusion is echocardiog-raphy (Fig. 14.3), which can identify pericar-dial collections as small as 20 mL. This non-invasive technique can quantify the volumeof pericardial fluid, determine whether ven-tricular filling is compromised, and whennecessary, help direct the placement of apericardiocentesis needle.

Treatment

If the cause of the effusion is known, ther-apy is directed toward the underlying dis-

LVPE

Figure 14.3. Two-dimensional echocardiogram (para-sternal short-axis view) of a pericardial effusion (PE)surrounding the heart. LV, left ventricle.

Fig. 3

Fig. 4


order (e.g., intensive dialysis for uremic ef-fusion). If the cause is not evident, the clini-cal state of the patient determines whetherpericardiocentesis (removal of pericardialfluid) should be undertaken. An asympto-matic effusion, even of large volume, can beobserved for long periods without specificintervention. However, if serial examina-tion demonstrates a precipitous rise in peri-cardial volume or if hemodynamic compres-sion of the cardiac chambers becomesevident, then pericardiocentesis should beperformed for therapeutic drainage and foranalysis of the fluid.

CARDIAC TAMPONADE

At the opposite end of the spectrum fromthe asymptomatic pericardial effusion is car-diac tamponade. In this condition, pericar-dial fluid accumulates under high pressure,compresses the cardiac chambers, and se-verely limits filling of the heart. As a result,ventricular stroke volume and cardiac out-put decline, potentially leading to hypoten-sive shock and death.

Etiology

Any etiology of acute pericarditis (see Table14.1) can progress to cardiac tamponade,but the most common causes are neoplastic,postviral, and uremic pericarditis. Acute he-morrhage into the pericardium is also animportant cause of tamponade, which canresult from (1) blunt or penetrating chesttrauma, (2) rupture of the left ventricular(LV) free wall following MI (see Chapter 7),or (3) as a complication of a dissecting aor-tic aneurysm (see Chapter 15).

Pathophysiology

As a result of the surrounding tense pericar-dial fluid, the heart is compressed, and thediastolic pressure within each chamber becomeselevated and equal to the pericardial pressure.The pathophysiologic consequences of thisare illustrated in Figure 14.4. Because thecompromised cardiac chambers cannot ac-commodate normal venous return to the

10090-14_CH14.qxd 8/31/06 4:56 PM Page 341


heart, the systemic and pulmonary venouspressures rise. The increase of systemic ve-nous pressure results in signs of right-sidedheart failure (e.g., jugular venous disten-tion), whereas elevated pulmonary venouspressure leads to pulmonary congestion. Inaddition, reduced filling of the ventriclesduring diastole decreases the systolic strokevolume, and the cardiac output declines.

These derangements trigger compensatorymechanisms aimed at maintaining tissueperfusion, initially through activation of thesympathetic nervous system. Nonetheless,failure to evacuate the effusion leads to inad-equate perfusion of vital organs, shock, andultimately death.

Clinical Features

Cardiac tamponade should be suspected inany patient with known pericarditis, peri-cardial effusion, or chest trauma who devel-

ops signs and symptoms of systemic vascu-lar congestion and decreased cardiac output(Table 14.5). The key physical findings in-clude (1) jugular venous distention; (2) sys-temic hypotension; and (3) a “small, quietheart” on physical examination, a result ofthe insulating effects of the effusion. Othersigns include sinus tachycardia and pulsusparadoxus. Dyspnea and tachypnea reflectpulmonary congestion and decreased oxy-gen delivery to peripheral tissues.

If tamponade develops suddenly, symp-toms of profound hypotension are evident,

Cardiac tamponade Constrictive pericarditis

Pulmonaryrales

Reflex tachycardia

Hypotension

Jugular venous distention

Hep ato meg aly ± ascites

Per iph eral edema

Figure 14.4. Pathophysiology of cardiac tamponade and constrictive peri-carditis. The symptoms and signs (boxes) arise from impaired diastolic filling of theventricles in both conditions.

TABLE 14.5. Clinical Features of Cardiac Tamponade

Jugular venous distentionHypotension with pulsus paradoxusQuiet precordium on palpationSinus tachycardia

10090-14_CH14.qxd 8/31/06 4:56 PM Page 342

including confusion and agitation. How-ever, if the effusion develops more slowly,over a period of weeks, then fatigue (causedby low cardiac output) and peripheral edema(owing to right-sided heart failure) may bethe presenting complaints.

Pulsus paradoxus is an important phys-ical sign in cardiac tamponade that can berecognized at the bedside using a blood pres-sure cuff. It refers to a decrease of systolic bloodpressure (more than 10 mm Hg) during normalinspiration.

Pulsus paradoxus is not really “paradoxi-cal”; it is just an exaggeration of appropriatecardiac physiology. Normally, expansion ofthe thorax during inspiration causes the in-trathoracic pressure to become more nega-tive compared with the expiratory phase.This facilitates systemic venous return to thechest and augments filling of the right ven-tricle (RV). The transient increase in RV sizeshifts the interventricular septum toward theleft, which slightly diminishes left ventricu-lar filling. As a result, in normal persons, LVstroke volume and systolic blood pressuredecline slightly following inspiration.

In cardiac tamponade, this situation isexaggerated because both ventricles share areduced, fixed volume as a result of externalcompression by the tense pericardial fluid.In this case, the inspiratory increase of rightventricular volume and bulging of the in-terventricular septum toward the left have aproportionally greater effect on the limita-tion of LV filling. Thus, in tamponade thereis a more substantial reduction of LV strokevolume (and therefore systolic blood pres-sure) following inspiration.

Pulsus paradoxus may also be manifest by other conditions in which inspiration isexaggerated, including severe asthma andchronic obstructive airway disease.

Diagnostic Studies

Echocardiography is the most useful noninva-sive technique to evaluate whether pericar-dial effusion has led to cardiac tamponadephysiology. An important indicator of high-pressure pericardial fluid is compression ofthe RV and right atrium during diastole (see

AQ1

Fig. 5


Fig. 3.13). In addition, echocardiography candifferentiate between cardiac tamponade andother causes of low cardiac output, such asventricular contractile dysfunction.

The definitive diagnostic procedure forcardiac tamponade is cardiac catheterizationwith measurement of intracardiac and in-trapericardial pressures, usually combinedwith therapeutic pericardiocentesis, as de-scribed in the next section.

Treatment

Removal of the high-pressure pericardialfluid is the only intervention that reversesthe life-threatening physiology of this con-dition. Pericardiocentesis is best performedin the cardiac catheterization laboratory,where the hemodynamic effect of fluid re-moval can be assessed. The patient is posi-tioned head up at a 45° angle to promotepooling of the effusion, and a needle is in-serted into the pericardial space through theskin, just below the xiphoid process (whichis the safest location to avoid piercing acoronary artery). A catheter is then threadedinto the pericardial space and connected toa transducer for pressure measurement. An-other catheter is threaded through a sys-temic vein into the right side of the heart,and simultaneous recordings of intracardiacand intrapericardial pressures are compared.In tamponade, the pericardial pressure is el-evated and is equal to the diastolic pressureswithin the cardiac chambers; the diastolicpressures are elevated to the same degree inall the chambers because of the surroundingcompressive force of the effusion.

In addition, the right atrial pressure trac-ing, which is equivalent to the jugular ve-nous pulsation observed on physical exami-nation, displays a characteristic abnormality(Fig. 14.5). During early diastole in a normalperson, as the right ventricular pressure fallsand the tricuspid valve opens, blood quicklyflows from the right atrium into the RV,leading to a rapid decline in the right atrialpressure tracing (y descent). In tamponade,however, the pericardial fluid compressesthe right ventricle and prevents its rapidexpansion. Thus, the right atrium cannotempty quickly, and the y descent is blunted.

10090-14_CH14.qxd 8/31/06 4:56 PM Page 343


Following successful pericardiocentesis,the pericardial pressure falls to normal andis no longer equal to the pressures withinthe heart chambers, which also decline totheir normal levels. After initial aspirationof fluid, the pericardial catheter may be leftin place for 1 to 2 days to allow more com-plete drainage.

When pericardial fluid is obtained for di-agnostic purposes, it should be stained andcultured for bacteria, fungi, and acid-fastbacilli (tuberculosis), and cytologic examina-tion should be performed to evaluate for ma-lignancy. Other common measurements of

pericardial fluid include cell counts (e.g.,white cell count is elevated in bacterial infec-tions and other inflammatory conditions)and protein and lactate dehydrogenase (LDH)levels. If the concentration ratio of pericardialprotein to serum protein is >0.5, or that ofpericardial LDH to serum LDH is >0.6, thenthe fluid is consistent with an exudate; other-wise, it is more likely a transudate. When tu-berculosis is suspected, it is also useful to mea-sure the level of adenosine deaminase in thepericardial fluid. Some studies have indicatedthat an elevated level is highly sensitive andspecific for tuberculosis.

If cardiac tamponade recurs followingpericardiocentesis, the procedure can be re-peated. In some cases, a more definitive sur-gical undertaking (removal of part or all ofthe pericardium) is required to prevent reac-cumulation of the effusion.

CONSTRICTIVE PERICARDITIS

The other major potential complication ofpericardial diseases is constrictive pericardi-tis. This is a condition not often encoun-tered but important to understand, becauseit can masquerade as other, more commondisorders. In addition, it is an affliction thatmay cause profound symptoms yet is fullycorrectable if recognized.

Etiology and Pathogenesis

In the early part of the twentieth century,tuberculosis was the major cause of con-strictive pericarditis, but that is much lesscommon today in industrialized societies.The most frequent cause now is idiopathic(i.e., months to years following presumedidiopathic or viral acute pericarditis). How-ever, any cause of pericarditis can lead to thiscomplication (see Table 14.1).

Pathology

Following an episode of acute pericarditis,any pericardial effusion that has accumu-lated usually undergoes gradual resorption.However, in patients who later develop con-

Figure 14.5. Schematic diagrams of right atrial (orjugular venous) pressure electrocardiogram (ECG)recordings. A. Normal. The initial a wave representsatrial contraction. The v wave reflects passive filling of theatria during systole, when the tricuspid and mitral valvesare closed. After the tricuspid valve opens, the right atrialpressure falls (y descent) as blood empties into the rightventricle. B. Cardiac tamponade. High-pressure pericar-dial fluid compresses the heart, impairing right ventricu-lar filling, so that the y descent is blunted. C. Constrictivepericarditis. The earliest phase of diastolic filling is not im-paired so that the y descent is not blunted. The y descentappears accentuated because it descends from a higherthan normal right atrial pressure. The right atrial c wave(described in Chapter 2) is not shown.

1 LINE SHORT

10090-14_CH14.qxd 8/31/06 4:56 PM Page 344


strictive pericarditis, the fluid undergoes or-ganization, with subsequent fusion of thepericardial layers, followed by fibrous scarformation. In some patients, calcification ofthe adherent layers ensues, further stiffen-ing the pericardium.

Pathophysiology

The pathophysiologic abnormalities in con-strictive pericarditis occur during diastole;systolic contraction of the ventricles is usu-ally normal. In this condition, a rigid, scarredpericardium encircles the heart and inhibitsnormal filling of the cardiac chambers. For ex-ample, as blood passes from the right atriuminto the right ventricle during diastole, theRV size expands and quickly reaches thelimit imposed by the constricting peri-cardium. At that point, further filling is sud-denly arrested, and venous return to theright heart ceases. Thus, systemic venouspressure rises, and signs of right-sided heartfailure ensue. In addition, the impaired fill-ing of the left ventricle causes a reduction instroke volume and cardiac output, whichleads to lower blood pressure.

Clinical Features

The symptoms and signs of constrictive peri-carditis usually develop over months toyears. They result from (1) reduced cardiacoutput (fatigue, hypotension, reflex tachy-cardia) and (2) elevated systemic venouspressures (jugular venous distention, hepa-tomegaly with ascites, and peripheral edema).Because the most impressive physical find-ings are often the insidious development ofhepatomegaly and ascites, patients may bemistakenly suspected of having hepatic cir-rhosis or an intra-abdominal tumor. How-ever, careful inspection of the jugular veinscan point to the correct diagnosis of con-strictive pericarditis.

On cardiac examination, an early dias-tolic “knock” may follow the second heartsound (S2; see Chapter 2) in patients with se-vere calcific constriction. It represents the

sudden cessation of ventricular diastolic fill-ing imposed by the rigid pericardial sac.

In contrast to cardiac tamponade, peri-cardial constriction results in pulsus para-doxus much less frequently. Recall that intamponade, this finding reflects inspiratoryaugmentation of RV filling, at the expenseof LV filling. However, in constrictive peri-carditis, the negative intrathoracic pressuregenerated by inspiration is not easily trans-mitted through the rigid pericardial shell tothe right-sided heart chambers; therefore,inspiratory augmentation of RV filling ismore limited. Rather, when a patient withsevere pericardial constriction inhales, thenegative intrathoracic pressure draws bloodtoward the thorax, where it cannot be ac-commodated by the constricted right-sidedcardiac chambers. As a result, the increasedvenous return accumulates in the intratho-racic systemic veins, causing the jugularveins to become more distended during in-spiration (Kussmaul sign). This is the oppo-site of normal physiology, in which inspira-tion results in a decline in jugular venouspressure, as venous return is drawn into theheart. Thus, typical findings in pericardialdisease can be summarized as follows:

Constrictive Cardiac Pericarditis Tamponade

Pulsus paradoxus + +++Kussmaul sign +++ −

Diagnostic Studies

The chest radiograph in constrictive peri-carditis shows a normal or mildly enlargedcardiac silhouette. Calcification of the peri-cardium can be detected in some patientswith severe chronic constriction. The ECGgenerally shows only nonspecific ST and Twave abnormalities, although atrial arrhyth-mias are common.

Echocardiographic evidence of constrictionis subtle. The pericardium, if well imaged, isthickened; the ventricular cavities are smalland contract vigorously, and diastolic ven-tricular filling terminates abruptly in early1 LINE SHORT

10090-14_CH14.qxd 8/31/06 4:56 PM Page 345


diastole, as the chambers reach the limit im-posed by the surrounding rigid shell.

Computed tomography or magnetic resonanceimaging is superior to echocardiography inthe assessment of pericardial anatomy andthickness. The presence of normal pericardialthickness (<2 mm) by these modalities is gen-erally a reliable indication that constrictivepericarditis is not present.

The diagnosis of constrictive pericarditiscan be confirmed by cardiac catheterization,which reveals four key features:

1. Elevation and equalization of the diastolicpressures in each of the cardiac chambers.

2. An early diastolic “dip and plateau” con-figuration in the right and left ventricu-lar tracings (Fig. 14.6). This pattern re-flects blood flow into the ventricles at thevery onset of diastole, just after the tri-cuspid and mitral valves open, followedby sudden cessation of filling as furtherexpansion of the ventricles is arrested bythe surrounding rigid pericardium.

3. A prominent y descent shown in theright atrial pressure tracing (see Fig.

14.5). After the tricuspid valve opens,the right atrium quickly empties into theRV (and its pressure rapidly falls) duringthe very brief period before filling is ar-rested. This is in contrast to cardiac tam-ponade, in which the external compres-sive force throughout the cardiac cycleprevents rapid ventricular filling, even inearly diastole, such that the y descent isblunted.

4. During the respiratory cycle, there is dis-cordance in the RV and LV systolic pres-sures (the RV systolic pressure rises withinspiration, while that of the LV tends todecline). In normal persons, the negativeintrathoracic pressure induced by inspi-ration causes the systolic pressure of bothventricles to decline slightly. In contrast,in constrictive pericarditis, the heart isisolated from the rest of the thorax by thesurrounding rigid shell. In this circum-stance, negative intrathoracic pressureassociated with inspiration decreases thepressure in the pulmonary veins but notin the left-sided cardiac chambers. Thiscauses a decline in the pressure gradientdriving blood back to the left side of theheart, such that left ventricle filling is re-duced. Less ventricular filling reduces thestroke volume and therefore results in alower LV systolic pressure during inspira-tion (thus causing pulsus paradoxus insome patients). Simultaneously, becausethe two ventricles share a fixed space lim-ited by the rigid pericardium, the reducedLV volume allows the interventricularseptum to shift towards the left, whichenlarges the RV (a behavior termed ven-tricular interdependence). The resultant in-crease in RV filling augments its strokevolume and systolic pressure during in-spiration. During expiration, the situa-tion is reversed, with the RV systolicpressure declining and that of the LV increasing.

The clinical and hemodynamic findings ofconstrictive pericarditis are often similar to those of restrictive cardiomyopathy (seeChapter 10), another uncommon condition.Distinguishing between these two syndromes

100

80

60

RV

TIME

LV

ECG

Pre

ssur

e(m

mH

g)

40

20

Plateau

Early diastolicfilling wave

Figure 14.6. Schematic tracings of left ventricular(LV) and right ventricular (RV) pressures in constric-tive pericarditis. Early diastolic ventricular filling abruptlyhalts as the volume in each ventricle quickly reaches thelimit imposed by the constricting pericardium. Through-out most of diastole, the LV and RV pressures are abnor-mally elevated and equal.

Fig. 6

10090-14_CH14.qxd 8/31/06 4:56 PM Page 346

is important because pericardial constrictionis correctable, whereas most cases of restric-tive cardiomyopathy have very limited ef-fective treatments (Table 14.6). An endomyo-cardial biopsy is sometimes necessary todistinguish between these (the biopsy resultsare normal in constriction but usually ab-normal in restrictive cardiomyopathy; seeChapter 10).

Treatment

The only effective treatment of severe con-strictive pericarditis is surgical removal ofthe pericardium. Symptoms and signs ofconstriction may not resolve immediatelybecause of the associated stiffness of theneighboring outer walls of the heart, buteventual symptomatic improvement is ex-pected in the majority of patients whoundergo this procedure.

SUMMARY

1. Acute pericarditis is most often of idio-pathic or viral cause and is usually a self-limited illness. More serious forms of peri-carditis arise from the conditions listed inTable 14.1.

2. Common findings in acute pericarditisinclude (a) pleuritic chest pain; (b) fever;(c) pericardial friction rub; and (d) diffuse

ST segment elevation on the ECG, oftenaccompanied by PR segment depression.

3. Complications of pericarditis include car-diac tamponade (accumulation of peri-cardial fluid under high pressure, whichcompresses the cardiac chambers) andconstrictive pericarditis (restricted fillingof the heart because of surrounding rigidpericardium).

Acknowledgment

Contributors to the previous editions of this chapterwere Angela Fowler, MD; Kathy Glatter, MD; ThomasG. Roberts, MD; Alan Braverman, MD; and LeonardS. Lilly, MD.

Additional Reading

Bertog SC, Thambidorai SK, Parakh K, et al. Con-strictive pericarditis: etiology and cause-specificsurvival after pericardiectomy. J Am Coll Cardiol2004;43:1445–1452.

Gaya AM, Ashford RF. Cardiac complications of ra-diation therapy. Clin Oncol 2005;17(3):153–159.

Imazio M, Bobbio M, Cecchi E, et al. Colchicine inaddition to conventional therapy for acute peri-carditis: Results of the COlchicine for acute PEricarditis (COPE) Trial. Circulation 2005;112:2012–2016.

Lange RA, Hillis LD. Clinical practice. Acute peri-carditis. N Engl J Med 2004;351:2195.

Little WC, Freeman GL. Pericardial disease. Circula-tion 2006;113:1622–1632.

TABLE 14.6. Differences Between Constrictive Pericarditis and Restrictive Cardiomyopathy

Feature Constrictive Pericarditis Restrictive Cardiomyopathy

Chest radiography• Pericardial calcificationsCT or MRI• Thickened pericardiumEchocardiography• Thickened pericardium• Respiratory cycle effect on transvalvular

Doppler velocitiesCardiac catheterization• Equalized LV and RV diastolic pressures• Elevated PA systolic pressure• Effect of inspiration on ystolic pressuresEndomyocardial biopsy

CT, computed tomography; LV, left ventricle; MRI, magnetic resonance imaging; PA, pulmonary artery; RV, right ventricle.

Yes (25–30% of patients)

Yes

Yes (but difficult to visualize)Exaggerated variations

YesUncommonDiscordant : LV↓, RV↑Normal

Absent

No

NoNormal

Often, LV > RVCommonConcordant: LV↓, RV↓Abnormal (e.g., amyloid)

Tab. 6


10090-14_CH14.qxd 8/31/06 4:56 PM Page 347

Shabetai R. The Pericardium. Boston: Kluwer Aca-demic, 2003.

Spodick DH. Acute cardiac tamponade. N Engl J Med2003;349:684–690.

Troughton R, Asher CR, Klein AL. Pericarditis. Lancet2004;363:717–727.


Maisch B, Seferovic PM, Ristic AD, et al. Guidelineson the diagnosis and management of pericardialdiseases executive summary; the task force on thediagnosis and management of pericardial diseasesof the European society of cardiology. Eur Heart J2004;25:587.

Permayer-Miralda G. Acute pericardial disease: ap-proach to aetiologic diagnosis. Heart 2004;90:252–254.

10090-14_CH14.qxd 8/31/06 4:56 PM Page 348

Chapter 14—Author Query1. AU: Correct interpretation of “all of the latter”?

10090-14_CH14.qxd 8/31/06 4:56 PM Page a

349

DISEASES OF THE AORTAAortic AneurysmsAortic Dissection

OCCLUSIVE ARTERIAL DISEASESPeripheral Arterial DiseaseAcute Arterial OcclusionVasculitic Syndromes

DISEASE CAUSING ARTERIAL SPASM:RAYNAUD PHENOMENON

VENOUS DISEASEVaricose VeinsVenous Thrombosis

C H A P T E R

15Diseases of thePeripheral VasculatureArash MostaghimiMark A. Creager

Peripheral vascular disease is an umbrella termthat includes a number of diverse pathologicentities that affect arteries, veins, and lym-phatics. Although this terminology makes adistinction between the “central” coronaryand “peripheral” systemic vessels, the vascu-lature as a whole comprises a dynamic, inte-grated, and multifunctional organ systemthat does not naturally comply with this se-mantic division.

Blood vessels serve many critical func-tions. First, they regulate the differential dis-tribution of blood and delivery of nutrientsand oxygen to tissues. Second, blood vesselsactively synthesize and secrete vasoactivesubstances that regulate vascular tone andantithrombotic substances that maintainthe fluidity of blood and vessel patency (seeChapters 6 and 7). Third, the vessels play anintegral role in the transport and distribu-

tion of immune cells to traumatized or in-fected tissues. Disease states of the periph-eral vasculature interfere with these essen-tial functions.

Peripheral vascular diseases result fromprocesses that can be grouped into three cat-egories: (1) structural changes in the vessel wallsecondary to degenerative conditions, infec-tion, or inflammation that lead to dilation,aneurysm, dissection, or rupture; (2) narrow-ing of the vascular lumen caused by athero-sclerosis, thrombosis, or inflammation; and(3) spasm of vascular smooth muscle. Theseprocesses can occur in isolation or one maylead to another.

DISEASES OF THE AORTA

The aorta is the largest conductance vessel ofthe vascular system. In adults, its diameter is

10090-15_CH15.qxd 8/31/06 4:57 PM Page 349

approximately 3 cm at its origin at the baseof the heart. The ascending aorta, 5 to 6 cmin length, leads to the aortic arch, fromwhich arise three major branches: the bra-chiocephalic (which bifurcates into the rightcommon carotid and subclavian arteries),the left common carotid, and the left sub-clavian arteries. As the descending aortacontinues beyond the arch, its diameter nar-rows to approximately 2 to 2.5 cm in healthyadults. As the aorta pierces the diaphragm, itbecomes the abdominal aorta, providingarteries to the abdominal viscera before bi-furcating into the left and right commoniliac arteries, which supply the pelvic organsand lower extremities.

The aorta, like other arteries, is composedof three layers (Fig. 15.1). At the luminal sur-face, the intima is composed of endothelialcells overlying the internal elastic lamina. Theendothelial layer is a functional interface be-tween the vasculature and circulating bloodcells and plasma. The media is composed ofsmooth muscle cells and a matrix that in-cludes collagen and elastic fibers. Collagenprovides tensile strength that enables the ves-sels to withstand high-pressure loads. Elastinis capable of stretching to 250% of its originallength and confers a distensible quality onvessels that allows them to recoil under pres-sure. The adventitia is primarily composed ofcollagen fibers, perivascular nerves, and vasavasorum, a rich vascular network that sup-plies oxygenated blood to the aorta.

The aorta is subject to injury from me-chanical trauma because it is continuouslyexposed to high pulsatile pressure and shearstress. The predominance of elastin in themedia (2:1 over collagen) allows the aorta toexpand during systole and recoil during di-astole. The recoil of the aorta against theclosed aortic valve contributes to the distalpropagation of blood flow during the phaseof left ventricular relaxation. With advanc-ing age, the elastic component of the aortaand its branches degenerates, and as colla-gen becomes more prominent, the arteriesstiffen. Systolic blood pressure thereforetends to rise with age because less energy isdissipated into the aorta during left ventric-ular contraction.

Diseases of the aorta most commonly appear as one of three clinical conditions:aneurysm, dissection, or obstruction.

Aortic Aneurysms

An aneurysm is an abnormal localized di-latation of an artery. In the aorta, aneurysmsare distinguished from diffuse ectasia, whichis a generalized yet lesser increase of theaortic diameter. Ectasia develops in older patients as elastic fibers fragment, smooth muscle cells decrease in number, and acidmucopolysaccharide ground substance ac-cumulates within the vessel wall.

The term aneurysm is applied when the di-ameter of a portion of the aorta has increasedby at least 50% compared with normal. Atrue aneurysm represents a dilatation of allthree layers of the aorta, creating a largebulge of the vessel wall. True aneurysms arecharacterized as either fusiform or saccular,depending on the extent of the vessel’s cir-cumference within the aneurysm (see Fig.15.1). A fusiform aneurysm, the more com-mon type, is characterized by symmetrical di-lation of the entire circumference of a seg-ment of the aorta. A saccular aneurysm is alocalized outpouching involving only a por-tion of the circumference.

In contrast, a pseudoaneurysm, or falseaneurysm, is a contained rupture of the ves-sel wall that develops when blood leaks outof the vessel lumen through a hole in the in-timal and medial layers and is contained bya layer of adventitia or perivascular or-ganized thrombus (see Fig. 15.1). Pseudo-aneurysms develop at sites of vessel injurycaused by infection or trauma, such as punc-ture of the vessel during surgery or percuta-neous catheterization. They are very unsta-ble and are prone to rupture.

Aneurysms may be confined to the abdo-minal aorta (most common), the thoracicaorta, or both locations. They may also ap-pear in peripheral and cerebral arteries.

Etiology and Pathogenesis of True Aortic Aneurysms

The etiology of aortic aneurysm formation ismultifactorial and varies depending on the

350 Chapter Fifteen

Fig. 1

10090-15_CH15.qxd 8/31/06 4:57 PM Page 350

Diseases of the Peripheral Vasculature 351

location of the lesion. Ascending thoracic aor-tic aneurysms typically are characterized bycystic medial necrosis, a process of degen-eration and fragmentation of elastic fibers,with accumulation of collagenous and mu-coid material within the medial layer. Thisprocess is associated with aging and with

hypertension and occurs in certain inher-ited disorders of connective tissue, includ-ing Marfan syndrome and Ehlers-Danlossyndrome.

Aneurysms of the descending thoracic andabdominal aorta are associated with athero-sclerosis and its risk factors, including smok-

A

B

C

IntimaMedia

Adventitia

IntimaMediaAdventitia

SaccularFusiform

Hole in intimaand media

Hematomacontained byadventitia or

perivascular clot

Aortic wall

lum

en

Figure 15.1. Classification of aortic aneurysms. A. The normal arterial wall consists of threelayers: the intima, media, and adventitia. B. True aneurysms represent localized dilatation of allthree layers of the arterial wall. Fusiform aneurysms involve the entire circumference of the aorta,whereas saccular aneurysms are a localized bulge of only a portion of the circumference. C. A falseaneurysm (or pseudoaneurysm) is actually a hole in the intima and media, with hematoma con-tained by a thin layer of adventitia or perivascular clot.

10090-15_CH15.qxd 8/31/06 4:57 PM Page 351

ing, hypertension, dyslipidemia, male gen-der, and advanced age. However, it is unlikelythat atherosclerosis alone is responsible forsuch aneurysm development. Rather, impor-tant contributions appear to derive from ge-netic predisposition (on a polygenetic basis),local vessel inflammation, and an imbal-ance between synthesis and degradation ofextracellular matrix proteins such as colla-gen and elastin. For example, matrix metal-loproteinases (see Chapter 5) capable of de-grading the components of the extracellularmatrix are found in increased quantitieswithin aortic aneurysms. Aneurysm forma-tion is also associated with markers of inflam-mation, including C-reactive protein (CRP)and cytokines such as interleukin 6 (IL-6).Levels of both CRP and IL-6 have been shownto correlate with the size of aneurysms, andinflammatory cells such as B and T lympho-cytes and macrophages are frequently foundon histologic examination.

Infrequent causes of aortic aneurysms(Table 15.1) include weakness of the mediafrom infections of the vessel wall by Salmo-nella species, staphylococci, streptococci, tu-berculosis, syphilis, or fungi. Inflammatorydiseases such as Takayasu arteritis or giantcell arteritis (both described later in thechapter) may similarly weaken the vesseland result in aneurysm formation.

Clinical Presentation and Diagnosis

Most aneurysms are asymptomatic, thoughsome patients, especially those with abdom-inal aortic aneurysms, may be aware of apulsatile mass. Others present with symp-toms related to compression of neighboringstructures by an expanding aneurysm. Tho-racic aortic aneurysms may compress thetrachea or mainstream bronchus, resulting

in cough, dyspnea, or pneumonia. Com-pression of the esophagus can result in dys-phagia, and involvement of the recurrent laryngeal nerve may lead to hoarseness.Aneurysms of the ascending aorta may di-late the aortic ring, resulting in aortic regur-gitation and symptoms of congestive heartfailure. Abdominal aortic aneurysms maycause abdominal or back pain or nonspecificgastrointestinal symptoms.

Aortic aneurysms are often first suspectedwhen dilatation of the vessel is observed onchest or abdominal radiographs, particu-larly if the wall is calcified. Aneurysms of theabdominal aorta or of the large peripheralarteries may also be discovered by carefulpalpation during physical examination. Thediagnosis is confirmed by ultrasonography,contrast-enhanced computed tomography(CT), or magnetic resonance angiography(Fig. 15.2).

The most devastating consequence of anaortic aneurysm is rupture, which can befatal. An aneurysm may leak slowly or burstsuddenly, resulting in profound blood lossand hypotension. Thoracic aortic aneurysmsmay rupture into the pleural space, medi-astinum, or bronchi. Abdominal aortic aneu-rysms may rupture into the retroperitoneal

352 Chapter Fifteen

TABLE 15.1. Conditions Associated with True Aortic Aneurysms

1.Cystic medical necrosis2.Atherosclerosis3. Infections of arterial wall4.Vasculitis

Figure 15.2. Abdominal aortic aneurysm. Computedtomographic angiogram (CTA) of an abdominal aorticaneurysm, indicated by the arrow. (Courtesy of Dr. FrankRybicki, Brigham and Women’s Hospital, Boston, MA).

Tab. 1

Fig. 2

10090-15_CH15.qxd 8/31/06 4:57 PM Page 352


space or abdominal cavity or erode into theintestines, resulting in massive gastrointesti-nal bleeding.

Natural history studies have shown thatthe risk of rupture is related to the size of theaneurysm, as predicted by LaPlace’s relation-ship (wall tension is proportional to the prod-uct of pressure and radius). The mean rates ofthoracic and abdominal aortic aneurysms ex-pansion are 0.1 and 0.4 cm/year, respectively.Thoracic aneurysms rupture at an annual rateof 2% for aneurysms <5 cm in diameter, 3%for aneurysms 5 to 5.9 cm, and 7% foraneurysms >6 cm. Abdominal aneurysms <4 cm, 4 to 4.9 cm, and 5 to 5.9 cm have anannual rates of rupture of 0.3%, 1.5%, and6.5%, respectively. Abdominal aneurysms >6 cm have a markedly higher risk of rupture.

Treatment

Treatment of an aortic aneurysm is basedon its size and the patient’s overall medicalcondition. Once an aneurysm is identified,its dimensions should be closely moni-tored through repeated imaging every 6 to 12 months. In general, surgical treat-ment is considered for ascending aorticaneurysms >5.5 to 6.0 cm. Ascending aor-tic aneurysms in patients with Marfan syn-drome (in whom the risk of complicationsis greater) should be considered for surgicalrepair if the diameter is >5 cm. Surgical repair is generally recommended for de-scending thoracic aortic aneurysms mea-suring 6.5 to 7.0 cm, for abdominal aorticaneurysms measuring 5.5 cm or more, andfor smaller aneurysms that enlarge at a rate>1.0 cm/year.

The mortality associated with elective sur-gical repair of thoracic aortic aneurysms is3% to 5%. Patients are maintained on car-diopulmonary bypass as the aneurysm is re-sected and replaced with a prosthetic Dacrongraft. Patients with aneurysms involving mul-tiple aortic segments have staged repairs, inwhich one segment is corrected at a time.Some patients may be candidates for mini-mally invasive repair, in which a translumi-nally placed endovascular stent graft is posi-tioned across the aneurysm.

Surgical repair of abdominal aortic aneu-rysms involves placement of a prostheticgraft. The operative mortality for such pro-cedures at high-volume institutions is 1% to2%. Percutaneous endovascular repair of in-frarenal abdominal aortic aneurysms withstent grafts can be performed in selected pa-tients with less acute morbidity, and resultsappear to be similar to that of surgical re-pair, but long-term outcomes are still beingdefined.

Aortic Dissection

Aortic dissection is a life-threatening condi-tion in which a blood-filled channel dividesthe medial layers of the aorta, splitting (ordissecting) the intima from the adventitiaalong various lengths of the vessel.

Etiology, Pathogenesis, and Classification

Aortic dissection is thought to arise from atear in the intimal layer of the vessel wallthat allows blood from the lumen, underthe driving force of the systemic pressure, toenter into the media and propagate alongthe plane of the muscle layer. Another pos-tulated origin of aortic dissection relates torupture of vasa vasorum with hemorrhageinto the media, forming a hematoma in thearterial wall that subsequently tears throughthe intima and into the vessel’s lumen.

Any condition that interferes with thenormal integrity of the elastic or muscularcomponents of the medial layer can predis-pose to aortic dissection. Such degenerationmay arise from chronic hypertension, aging,and/or cystic medial degeneration (which,as described earlier, is a feature of certainhereditary connective tissue disorders, suchas the Marfan syndrome and Ehlers-Danlossyndrome). In addition, traumatic insult tothe aorta (e.g., blunt chest trauma or acci-dental vessel damage during intra-arterialcatheterization or cardiac surgery) can alsoincite this condition.

Aortic dissection is most common in thesixth and seventh decades and occurs morefrequently in men. More than two thirds ofpatients have a history of hypertension. Dis-

10090-15_CH15.qxd 8/31/06 4:57 PM Page 353

section most commonly involves the as-cending thoracic aorta (65%) and descend-ing thoracic aorta (20%), while the aorticarch (10%) and abdominal aortic (5%) seg-ments are less commonly affected.

Dissections are commonly classified intotwo categories (Stanford types A and B), de-pending on their location and extent (Fig.15.3). In a type A dissection (proximal), theascending aorta is involved, regardless of thesite of the primary tear. A type B dissection(distal) does not involve the ascending aortaor arch and is therefore confined to the de-scending thoracic and abdominal aorta. Thisdistinction is important because treatmentstrategies and prognoses are determined bylocation. Proximal aortic involvement tendsto be the more devastating form because ofits potential for extension into the coronaryand arch vessels, the support structures ofthe aortic valve, or the pericardial space. Ap-proximately two thirds of dissections aretype A and one third are type B. Dissections

may also be classified as acute or chronic,with acute dissections presenting with symp-toms of less than 2 weeks’ duration.


The most common symptom of aortic dis-section is sudden, severe pain with a “tear-ing” or “ripping” quality in the anteriorchest (typical of type A dissections) or be-tween the scapulae (type B dissections). Thepain travels as the dissection propagatesalong the aorta and can radiate anywhere inthe thorax or abdomen. Painless dissectionis possible but uncommon (6.4% of cases).

Other symptoms relate to the catastrophiccomplications that can occur at the time ofpresentation or thereafter (Table 15.2) andinclude (1) rupture through the adventitiaanywhere along the aorta (often into the leftpleural space or pericardium); (2) occlusionof major branches of the aorta by the prop-agating hematoma within the vessel wall,

354 Chapter Fifteen

Figure 15.3. Aortic dissection. Type A involves the ascending aorta,whereas type B does not. (Reprinted with permission from Cotran RS, KumarV, Robbins SL. Robbin’s Pathologic Basis of Disease. Philadelphia: WB Saun-ders, 1989.)AQ3

Fig. 3

Tab. 2

10090-15_CH15.qxd 8/31/06 4:57 PM Page 354


which compresses the lumen and can resultin myocardial infarction (coronary artery in-volvement), stroke (carotid artery involve-ment), visceral ischemia, renal failure, orloss of pulse in an extremity; and (3) exten-sion into the aortic root, with disruption ofthe aortic valve support apparatus causingaortic regurgitation.

Several important physical findings maybe present. Hypertension is frequently de-tected, either as an underlying cause of dis-section, a result of the sympathetic nervoussystem response to the severe pain, or be-cause of diminished renal vascular flow, withactivation of the renin-angiotensin system.If the dissection has occluded one of thesubclavian arteries, a difference in systolicblood pressure between the arms is noted.Neurologic deficits may accompany dissec-tion into the carotid vessels. If a type A dis-section results in aortic regurgitation, anearly diastolic murmur can be detected onauscultation. Leakage from a type A dissec-tion into the pericardial sac may producesigns of cardiac tamponade (see Chapter 14).

The diagnosis of aortic dissection mustnot be delayed, because catastrophic com-plications or death may rapidly ensue. Theconfirmatory imaging techniques most use-ful in detecting dissection include contrast-enhanced CT, transesophageal echocardiog-raphy (TEE), magnetic resonance imaging,and contrast angiography. Each of thesetechniques has specific advantages and dis-advantages, and the decision of which toemploy is often guided by the hospital’s

local expertise. In emergency situations, CTscanning or TEE can generally be obtainedrapidly and offer excellent sensitivity andspecificity for the diagnosis.

Treatment

The goal of acute treatment is to arrest pro-gression of the dissecting channel. Suspi-cion of acute aortic dissection warrants im-mediate medical therapy to reduce systolicblood pressure (aiming for a systolic pres-sure of 100 to 120 mmHg) and to decreasethe force of left ventricular contraction andthus minimize aortic wall shear stress. Use-ful pharmacologic agents in this regard in-clude β-blockers (to reduce the force of con-traction and heart rate as well as to lowerblood pressure) and vasodilators such as so-dium nitroprusside (to rapidly reduce bloodpressure). In proximal (type A) dissections,early surgical correction has been shown toimprove outcomes compared with medicaltherapy alone. Surgical therapy involves re-pairing the intimal tear, suturing the edgesof the false channel, and if necessary, in-serting a synthetic aortic graft.

In contrast, patients with uncomplicatedtype B dissections are initially managed withaggressive medical therapy alone; early sur-gical intervention does not improve the out-come in these patients. Surgery is indicated,however, if there is clinical evidence ofpropagation of the dissection, compromiseof major branches of the aorta, impendingrupture, or continued pain. Percutaneouscatheter-based repair with endovascularstent grafts has been used successfully in se-lected stable patients with type B dissec-tions. The graft seals the entry site of the dis-section, resulting in thrombosis of the falselumen.

OCCLUSIVE ARTERIAL DISEASES

Arterial occlusion may result from athero-sclerosis, thromboembolism, or vasculitis(inflammation of the vessel wall). The clini-cal presentation of these disorders resultsfrom decreased perfusion to the affected limbor organs.

TABLE 15.2. Complications of Aortic Dissection

RupturePericardial tamponadeHemomediastinumHemothorax (usually left sided)Occlusion of aortic branch vesselsCarotid (stroke)Coronary (myocardial infarction)Splanchnic (organ infarction)Renal (acute renal failure)Iliac, brachiocephalic, subclavian (limb ischemia)Distortion of aortic annulusAortic regurgitation

AQ2

10090-15_CH15.qxd 8/31/06 4:57 PM Page 355

Peripheral Arterial Disease

Etiology and Pathogenesis

The formation of atherosclerotic plaques inlarge and medium-sized arteries may resultin chronic occlusive arterial disease, withprogressive stenosis and obstruction ofblood flow. This disorder is often referred toas peripheral arterial disease (PAD), and al-though it may affect many vascular beds, itgenerally refers to atherosclerotic disease inarteries of the pelvis or lower limbs. Itssymptoms result from ischemia distal to thestenosis. It is a very prevalent vascular dis-order, affecting approximately 4% of per-sons over age 40 and 15% to 20% of thoseover age 70.

The pathology of PAD is identical to thatof atherosclerotic coronary artery disease(CAD), and the major coronary risk factors(e.g., cigarette smoking, diabetes mellitus,dyslipidemia, and hypertension) are also as-sociated with PAD. Approximately 40% ofpatients with PAD actually have clinicallysignificant CAD. As a consequence of thesystemic nature of atherosclerosis, patientswith PAD have a twofold to fivefold in-creased risk of cardiovascular death com-pared with patients who do not have thiscondition.

The pathophysiology of PAD is also simi-lar to that of CAD. Ischemia of the affectedregion occurs when the balance betweenoxygen supply and demand is upset; exer-cise raises the demand for blood flow in thelimbs’ skeletal muscle, and a stenosed or ob-structed artery cannot provide an adequatesupply. Rest improves symptoms as the bal-ance between oxygen supply and demand isrestored.

Recall from Chapter 6 that the degree ofblood flow reduction relates closely to theextent of vessel narrowing, the length of thestenosis, and blood viscosity. Pouseille’sequation describes this relationship:

in which Q = flow, ∆P = pressure drop acrossthe stenosis, r = vessel radius, η = blood vis-cosity, and L = length of stenosis. Thus, the

QP r

L= ∆ π

η

4

8

degree of vessel narrowing by the stenosis(i.e., the change in r) has the greatest impacton flow. For example, if the radius is reducedby one half, the flow will be reduced to onesixteenth of its baseline. The equation alsoindicates that for stenoses of the samelength and radius, higher flow rates corre-spond to greater pressure drops across thestenoses. That is, as the flow velocity in-creases across a stenotic vessel, the bloodturbulence results in a loss of kinetic energy.The result is a decline in perfusion pressuredistal to the stenosis.

During exercise, products of skeletal mus-cle metabolism (e.g., adenosine) act locallyto dilate arterioles. The resulting decrease invascular resistance serves to increase bloodflow to the active muscle (recall that flow =pressure ÷ resistance). In turn, the increasedflow stimulates healthy arterial endotheliumto release vasodilating factors such as nitricoxide, thereby increasing the radii of up-stream conduit vessels. However, in PAD, ob-structed arteries cannot respond to the va-sodilating stimuli, thereby limiting flowincreases. In addition, dysfunctional athero-sclerotic endothelium does not release nor-mal amounts of vasodilating substances (seeChapter 6). Thus, the physical properties ofa stenosis and the reduced vasodilator activ-ity imposed by diseased endothelium pre-vent adequate blood flow from reachingdistal tissues and contribute to ischemia.

Hemodynamic changes alone cannot ac-count for the dramatic reductions in exer-cise capacity experienced by PAD patients;changes in muscle structure and functionare also seen. One such change is the dener-vation and dropout of muscle fibers, whichis thought to occur as an adaptation to in-termittent ischemia. The loss of such fiberscan explain the reduced muscle strengthand atrophy that occur in PAD patients.Even viable muscle fibers in affected limbsmay show abnormalities of mitochondrialoxidative metabolism.

In summary, atherosclerotic lesions pro-duce stenoses in peripheral conduit vesselsand limit blood flow to the affected extrem-ity. Mechanisms normally in place to com-pensate for increased demand, such as en-

356 Chapter Fifteen

10090-15_CH15.qxd 8/31/06 4:57 PM Page 356


dogenous release of vasodilators during ex-ercise and recruitment of microvessels, failin the face of endothelial dysfunction anddiminished flow velocity. Thus, states of in-creased oxygen demand are not met withadequate supply, producing limb ischemia.Adaptations to ischemia include changes inmuscle fiber metabolism and muscle fiberdropout. Together, these physical and bio-chemical changes result in weak lower limbsthat suffer ischemic discomfort during exer-cise. Severe peripheral atherosclerosis mayreduce limb blood flow to such an extentthat it cannot satisfy resting metabolic re-quirements. This results in critical limb ische-mia, which may progress to tissue necrosisand gangrene that may threaten viability ofthe extremity.


PAD may affect the aorta or the iliac, fe-moral, popliteal, and tibioperoneal arteries

(Fig. 15.4). Patients with PAD may thereforedevelop buttock, thigh, or calf discomfortprecipitated by walking and relieved by rest.This classic symptom of exertional limb fa-tigue and pain is known as claudication. Insevere PAD, patients may experience pain atrest, usually affecting the feet or toes. Thechronically reduced blood flow in this casepredisposes the extremity to ulceration, in-fection, and skin necrosis (Fig. 15.5). Patientswho smoke or have diabetes mellitus are athigh risk of these complications.

The location of claudication correspondsto the diseased artery, with the femoral andpopliteal arteries being the most commonsites (Table 15.3). The arteries of the upperextremities are less frequently affected, butbrachiocephalic or subclavian artery diseasecan cause arm claudication.

Physical examination generally revealsloss of pulses distal to the stenotic segment.Bruits (swishing sounds auscultated over a re-gion of turbulent blood flow) may be audible

Figure 15.4. An angiogram demonstrating atherosclerotic disease of the iliac vessels. No-tice the severe stenosis of the left external iliac artery (arrow).

Fig. 4

Fig. 5

Tab. 3

10090-15_CH15.qxd 8/31/06 4:57 PM Page 357

in the abdomen (because of stenoses withinthe mesenteric or renal arteries) or over iliac,femoral, or subclavian arterial stenoses. Inpatients with chronic severe ischemia, thelack of blood perfusion results in muscle at-rophy, pallor, cyanotic discoloration, hairloss, and occasionally gangrene and necrosisof the foot and digits.

Ischemic ulcers resulting from PAD oftenbegin as small traumatic wounds in areas ofincreased pressure or in regions prone to in-jury, such as the tips of toes and lateralmalleolus (see Fig. 15.5A). These often-painful ulcers fail to heal owing to the inad-equate blood supply. Diabetic patients withperipheral sensory neuropathies are particu-larly susceptible to ulcers at sites of traumaor pressure from ill-fitting footwear. Ische-mic ulcers can be distinguished from venousinsufficiency ulcers, which develop moreproximally and on the medial portion of theleg. Venous ulcers are also associated withreddish-brown pigmentation and varicoseveins (see Fig. 15.5B).

In the evaluation of PAD, it is helpful tomeasure the ratio of blood pressure in theankles to that in the arms (termed the ankle-brachial index [ABI]) using a blood pressurecuff and a Doppler instrument to detectblood flow. A normal ABI is ≥1.0 (i.e., theankle pressure is equal to, or slightly greaterthan, that in the arms). An index <0.9 is di-agnostic of PAD and may be associated withsymptoms of claudication, whereas andindex <0.5 is often observed in patients withrest pain and severe arterial compromise ofthe affected extremity. Other testing to as-sess peripheral perfusion includes limb seg-mental systolic pressure measurements (usingpneumatic cuffs placed along the extremity)and pulse volume recordings (i.e., graphicalmeasurement of volume changes in seg-ments of the extremity with each pulse). Du-plex ultrasonography is a commonly usednoninvasive method to visualize and assessthe extent of arterial stenoses and the corre-

358 Chapter Fifteen

Figure 15.5. Ulcerations caused by vascular insuffi-ciency A. Arterial insufficiency. Ulceration (arrow) affect-ing the great toe in a patient with severe peripheral arte-rial disease. B. Venous insufficiency ulcer near the medialmalleolus of the right leg. Notice the pigmentation of thesurrounding skin.

TABLE 15.3. Relation of Stenotic Site to Claudication Symptoms

Location of Site Claudication Symptoms

Distal aorta or iliac Buttocks, hips, thighs, arteries or calves

Femoral and popliteal Calvesarteries

Subclavian or axillary Arms arteries

10090-15_CH15.qxd 8/31/06 4:57 PM Page 358


sponding reductions in blood flow. Othermore advanced imaging studies (e.g., mag-netic resonance angiography, computed tomo-graphic angiography, or intra-arterial contrastangiography) are obtained when revascular-ization procedures are planned.

Treatment

For all patients with PAD, antiplatelet ther-apy and risk factor modification (includingsmoking cessation, lipid lowering, and con-trol of diabetes and hypertension) are impor-tant in reducing the likelihood of coronaryevents. Platelet inhibitors such as aspirinhave been shown to reduce cardiovascularmorbidity and mortality in patients withPAD. It has not been established if anti-platelet agents reduce symptoms or preventthrombotic complications of PAD itself.

Specific treatment of PAD includes sup-portive care of the feet to prevent trauma orrestriction of blood flow. Exercise, particu-larly walking, improves endurance in partby increasing metabolic efficiency in theskeletal muscle of the legs. A formal exerciseprogram is considered first-line therapy inthe management of PAD.

Certain medical therapies are sometimesuseful in the treatment of claudication.Cilostazol is a selective phosphodiesteraseinhibitor that increases cyclic adenosinemonophosphate and has vasodilator andplatelet inhibiting properties; it has beenshown to improve exercise capacity in pa-tients with PAD. Pentoxifylline is a drug pur-ported to improve the deformability of redand white blood cells and may improve clau-dication symptoms in some patients. Notethat most vasodilator drugs (see Chapter 17)are not helpful in relieving claudication.

More effective medical therapies for PADare on the horizon. Advances in angiogene-sis research and ongoing clinical trials pro-vide hope that pharmacologic revasculariza-tion with angiogenic growth factors, such asvascular endothelial growth factor and basicfibroblast growth factor may be possible.

Mechanical revascularization is indicatedwhen medical therapy has failed for patientswith disabling claudication and as first-line

therapy in cases of severe limb ischemia. Insevere cases, therapy is directed at healing ischemic ulcerations and preventing limbloss. Catheter-based interventions, such aspercutaneous transluminal angioplasty andstent implantation, can be performed on se-lected patients with low morbidity. Surgicalprocedures include bypass operations to cir-cumvent the occluded arteries using saphe-nous vein or prosthetic grafts. However, am-putation may be necessary if blood flowcannot be satisfactorily reestablished to main-tain limb viability.

Acute Arterial Occlusion

Acute arterial occlusion is caused either byembolization from a cardiac or proximalvascular site or by thrombus formation insitu. The origin of arterial emboli is mostoften the heart, usually resulting from dis-orders involving intracardiac stasis of flow(Table 15.4). Emboli may also originate fromthrombus or atheromatous material overly-ing a segment of the aorta. Rarely, arterialemboli originate from the venous circula-tion. If a venous clot travels to the right-heart chambers and is able to pass throughan abnormal intracardiac communication(e.g., an atrial septal defect), it then entersthe systemic arterial circulation (a conditionknown as a paradoxical embolism). Pri-mary arterial thrombus formation may ap-

TABLE 15.4. Origins of Arterial Emboli

Cardiac originStagnant left atrial flow (e.g., atrial fibrillation,

mitral stenosis)Left ventricular mural thrombus (e.g., dilated

cardiomyopathy, myocardial infarction, ventric-ular aneurysm)

Valvular lesions (endocarditis, mitral stenosis,thrombus on prosthetic valve)

Left atrial myxoma (mobile tumor in left atrium)Aortic originThrombus material overlying atherosclerotic

segmentVenous originParadoxical embolism travels through intracardiac

shunt

Tab. 4

10090-15_CH15.qxd 8/31/06 4:57 PM Page 359

pear at sites of endothelial damage or ather-osclerotic stenoses, or within bypass grafts.

The extent of tissue damage from throm-boembolism depends on the site of the oc-cluded artery, the duration of occlusion,and the degree of collateral circulationserving the tissue beyond the obstruction.Common symptoms and signs that maydevelop from reduced blood supply includepain, pallor, paralysis, paresthesia, andpulselessness (termed the “five Ps”). A sixthP, poikilothermia (coolness), is also oftenmanifest.

Patients with a proven acute arterial oc-clusion should be treated with anticoagu-lant agents such as heparin (followed bywarfarin) to prevent propagation of the clot and to reduce the likelihood of addi-tional embolic events. A revascularizationprocedure (catheter-based thrombolysis orthrombectomy, surgical embolectomy, orbypass surgery) is indicated if limb viabilityis at risk.

Atheroembolism is the condition of pe-ripheral arterial occlusion by atheromatousmaterial (i.e., cholesterol, platelets, and fibrin)derived from more proximal vascular sites,such as atherosclerotic lesions or aneurysms.The emboli lodge distally, resulting in occlu-sion of small arteries in the muscle and skin.Patients typically present with acute pain andtenderness at the involved site. Occlusion ofdigital vessels may result in the “blue toe”syndrome, culminating in gangrene andnecrosis. Other findings may include livedoreticularis (purplish mottling of involvedskin), kidney failure (caused by renal ath-eroembolism), and intestinal ischemia. Al-though an estimated 50% to 60% of cases arespontaneous, atheroembolism may occurafter intra-arterial procedures (e.g., cardiaccatheterization) when atherosclerotic mate-rial is unintentionally dislodged from theproximal vasculature. Ischemia resultingfrom atheroemboli is difficult to manage be-cause the heterogeneous composition anddistribution of emboli often precludes sur-gical removal or fibrinobolytic therapy.Surgical intervention to remove or bypassthe source of emboli may be necessary toprevent recurrences.

Vasculitic Syndromes

Vasculitis (vessel wall inflammation) resultsfrom immune complex deposition or cell-mediated immune reactions directedagainst the vessel wall. Immune complexesactivate the complement cascade with sub-sequent release of chemoattractants andanaphylatoxins that direct neutrophil mi-gration to the vessel wall and increase vas-cular permeability. Neutrophils injure thevessel by releasing lysosomal contents andproducing toxic oxygen-derived free radi-cals. In cell-mediated immune reactions, T lymphocytes bind to vascular antigens andrelease lymphokines that attract additionallymphocytes and macrophages to the vesselwall. These inflammatory processes cancause end-organ ischemia through vascularnecrosis or local thrombosis.

The cause of most of the vasculitic syn-dromes is unknown, but they often can bedistinguished from one another by the pat-tern of involved vessels and by histologiccharacteristics (Table 15.5).

Takayasu arteritis is a chronic vasculitisof unknown etiology that targets the aortaand its major branches. The estimated an-nual incidence is 1.2 to 2.6 cases per million.Between 80% and 90% of affected personsare women, with onset typically betweenthe ages of 10 and 40. Most reported caseshave been from Asia and Africa, but it is aworldwide disease. Patients typically presentwith systemic complaints such as malaiseand fever; focal symptoms are related to inflammation of the affected vessel and include cerebrovascular ischemia (brachio-cephalic or carotid artery involvement),myocardial ischemia (coronary artery), armclaudication (brachiocephalic or subclavianartery), or hypertension (renal artery). Thecarotid and limb pulses are diminished orabsent in nearly 85% of patients at the timeof diagnosis; hence, this condition is oftentermed a “pulseless” disease. Takayasu ar-teritis is also an uncommon cause of aorticaneurysm or aortic dissection. Histologicexamination of affected vessels reveals con-tinuous or patchy granulomatous inflam-mation with lymphocytes, histiocytes, and

360 Chapter Fifteen

AQ1

AQ2

Tab. 5

10090-15_CH15.qxd 8/31/06 4:57 PM Page 360


multinucleated giant cells, resulting in inti-mal proliferation, disruption of the elasticlamina, and fibrosis. Antiendothelial anti-bodies may also play a role in the disease.Steroid and cytotoxic drugs may reduce vas-cular inflammation and alleviate symptomsof Takayasu arteritis. Surgical bypass of ob-structed vessels may be helpful in severecases. The 5-year survival rate is 80% to 90%.

Giant cell arteritis (also termed temporalarteritis) is a chronic vasculitis of medium-sized to large arteries that most commonlyinvolve the cranial vessels or the aortic archand its branches. It is an uncommon dis-ease, with an incidence of 24 per 100,000,and the typical onset is after age 50; 65% ofpatients are female. Giant cell arteritis maybe associated with the inflammatory condi-tion known as polymyalgia rheumatica. His-tologic findings in affected vessels includelymphocyte and macrophage infiltration,intimal fibrosis, and focal necrosis, withgranulomas containing multinucleated giantcells.

Symptoms and signs of giant cell arteritisdepend on the distribution of affected arter-ies and may include diminished temporalpulses, prominent headache (temporal arteryinvolvement), or facial pain and claudica-tion of the jaw while chewing (facial arteryinvolvement). Ophthalmic artery arteritisleads to impaired vision, with permanentpartial or complete loss in 15% to 20% of pa-tients. In giant cell arteritis, the erythrocytesedimentation rate and C-reactive proteinare invariably elevated as markers of inflam-mation. Ultrasound examination can sup-port the diagnosis by demonstrating a hypo-

echoic halo around the involved arteriallumen with vessel stenosis and/or occlu-sion. The diagnosis can be confirmed bybiopsy of an involved vessel, usually a tem-poral artery, but treatment should not waitfor biopsy results. High-dose systemic steroidsare effective in treating vasculitis and pre-venting visual complications. Giant cell ar-teritis usually has a self-limited course of 1 to 5 years.

Thromboangiitis obliterans (Buergerdisease) is a segmental inflammatory dis-ease of small and medium-sized arteries,veins, and nerves involving the distal vesselsof the upper and lower extremities. It ismost prevalent in the Far and Middle Eastand has a very strong association with ciga-rette smoking. It is most common in menyounger than age 45; only 10% to 25% ofpatients are female. There is an increased in-cidence of human leukocyte antigen A9(HLA-A9) and HLA-B5 in affected persons.

Thromboangiitis obliterans presents witha triad of symptoms and signs: distal arter-ial occlusion, Raynaud phenomenon (de-scribed in the next section), and migratingsuperficial vein thrombophlebitis. Arterialocclusion results in arm and foot claudica-tion as well as ischemia of the digits. Tradi-tional laboratory markers of inflammationand autoimmune disease are usually not de-tected. Arteriographic features of involvedarteries include areas of stenosis interspersedwith normal-appearing vessels with moresevere disease distally, collateral vesselswith a “corkscrew” appearance around thestenotic regions, and lack of atherosclerosisin proximal arteries. The diagnosis can be

TABLE 15.5. Vasculitic Syndromes

Type Arteries Commonly Affected Histology

Takayasu arteritis

Giant cell arteritis

Thromboangiitis obliterans (Buerger disease)

Aorta and its branches

Medium to large size (especially cranialvessels as well as aortic arch and itsbranches)

Small size (especially distal arteries of extremities)

Granulomatous arteritis with fibrosis; sig-nificant luminal narrowing

Lymphocyte infiltration, intimal fibrosis,granuloma formation

Inflammation and thrombosis withoutnecrosis

10090-15_CH15.qxd 8/31/06 4:57 PM Page 361

established by tissue biopsy, although this israrely needed. Biopsy specimens of affectedvessels reveal an occlusive, highly cellular,inflammatory thrombus, with limited in-volvement of the vessel wall and preserva-tion of the internal elastic lamina (Fig. 15.6).The most important treatment for throm-

boangiitis obliterans is smoking cessation,which usually prevents progression of thedisease and its complications. Debridementof necrotic tissue may be necessary in ad-vanced cases. Revascularization is not usu-ally an option because of the distal locationof the arterial lesions.

DISEASE CAUSING ARTERIALSPASM: RAYNAUD PHENOMENON

Raynaud phenomenon is a vasospastic dis-ease of the digital arteries that occurs in sus-ceptible people when exposed to cool tem-peratures or sometimes during emotionalstress. Vasospasm is an extreme vasocon-strictor response that temporarily obliteratesthe vascular lumen, inhibiting blood flow.Typically, such episodes are characterized bya triphasic color response. First, the fingersand/or toes blanch to a distinct white asblood flow is interrupted (Fig. 15.7). The sec-ond phase is characterized by cyanosis, re-lated to local accumulation of desaturated

362 Chapter Fifteen

Figure 15.6. Histologic section of an artery display-ing thromboangiitis obliterans. Endothelial cell and fi-broblast proliferation appears in the vessel wall, andthrombus is present in the vessel lumen (arrow).

Figure 15.7. Raynaud phenomenon. The fourth digit (arrow) isblanched (phase 1 of the tricolor response).

Fig. 6Fig. 7

10090-15_CH15.qxd 8/31/06 4:57 PM Page 362


hemoglobin, followed by a third phase ofruddy color as blood flow resumes. Thecolor response may be accompanied bynumbness, paresthesias, or pain of the af-fected digits.

This condition may occur as an isolateddisorder, termed primary Raynaud phenome-non or Raynaud disease. Patients are predom-inantly women between the ages of 20 and40. Primary Raynaud phenomenon mostoften manifests in the fingers, but 40% ofpatients also have involvement of the toes.The prognosis of primary Raynaud phe-nomenon is relatively benign, with only aminority reporting a worsening of symp-toms over time.

Secondary Raynaud phenomenon mayappear as a component of other conditions.Common causes include connective tissuediseases (e.g., scleroderma and systemic lupuserythematosus) and arterial occlusive dis-orders. Other causes of secondary Raynaudphenomenon include carpal tunnel syn-drome, thoracic outlet syndrome, blooddyscrasias, certain drugs, and thermal or vibration injury.

Even in healthy vessels, cold exposurenormally produces a vasoconstrictor re-sponse. Cooling stimulates the sympatheticnervous system, resulting in local dischargeof norepinephrine, which binds to vascularadrenergic receptors. In the fingers and toes,only vasoconstricting α-receptors are pres-ent; other regional circulations have bothconstrictor and dilator adrenergic responses.Thus, a modest vasoconstriction of the dig-its results when healthy people are exposedto cooling. In contrast, in Raynaud phe-nomenon, cold exposure induces severe vaso-constriction.

A variety of mechanisms have been pro-posed to explain the vasospastic response tocold and stress in patients with primary Ray-naud phenomenon, including an exagger-ated sympathetic discharge in response to cold, heightened vascular sensitivity toadrenergic stimuli, or excessive release ofvasoconstrictor stimuli, such as serotonin,thromboxane, and endothelin. In patientswith secondary Raynaud phenomenoncaused by connective tissue diseases or arte-

rial occlusive disease, the digital vascularlumen is largely obliterated by sclerosis orinflammation, resulting in lower intralumi-nal pressure and greater susceptibility tosympathetically mediated vasoconstriction.

Treatment of Raynaud phenomenon in-volves avoiding cold environments, dress-ing in warm clothes, and wearing insulatedgloves or footwear. There has also beensome success in preventing vasospasm withpharmacologic agents that relax vasculartone, including calcium channel blockersand α-adrenergic blockers (see Chapter 17),but such therapies are reserved for severecases.

VENOUS DISEASE

Veins are high-capacitance vessels that con-tain more than 70% of the total blood vol-ume. In contrast to the muscular structureof arteries, the subendothelial layer of veinsis thin, and the tunica media comprisesfewer, smaller bundles of smooth musclecells intermixed with reticular and elasticfibers. Whereas veins of the extremities pos-sess intrinsic vasomotor activity, transportof blood back to the heart relies greatly onexternal compression by the surroundingskeletal muscles and on a series of one-wayendothelial valves.

Veins of the extremities are classified aseither deep or superficial. In the lower ex-tremities, where most peripheral venous dis-orders occur, the deep veins generally coursealong the arteries, whereas the superficialveins are located subcutaneously. The super-ficial vessels drain into deeper veins througha series of perforating connectors, ultimatelyreturning blood to the heart.

Varicose Veins

Varicose veins (Fig. 15.8) are dilated, tortu-ous superficial vessels that often develop inthe lower extremities. Clinically apparentvaricose veins occur in 10% to 20% of thegeneral population. They affect women twoto three times more frequently than men,and roughly half of patients have a familyhistory of this condition. Varicosities can

Fig. 8

10090-15_CH15.qxd 8/31/06 4:57 PM Page 363

occur in any vein in the body but are mostcommon in the saphenous veins of the legand their tributaries. They may also developin the anorectal area (hemorrhoids), in thelower esophageal veins (esophageal varices),and in the spermatic cord (varicocele).

Varicosity is thought to result from in-trinsic weakness of the vessel wall, from in-creased intraluminal pressure, or from con-genital defects in the structure and functionof venous valves. Varicose veins in the lowerextremities are classified as either primaryor secondary. Primary varicose veins origi-nate in the superficial system, and factorsthat contribute to their development in-clude pregnancy, prolonged standing, andobesity. During pregnancy or prolongedstanding, the high venous pressure withinthe legs contributes to varicosities whenthere is underlying weakness of the vesselwalls. In obese patients, the adipose tissuesurrounding vessel walls offers less struc-tural support to veins than lean mass. Sec-

ondary varicose veins occur when an abnor-mality in the deep venous system is thecause of superficial varicosities. These maydevelop in the setting of deep venous insuf-ficiency or occlusion, or when the perforat-ing veins are incompetent. In such cases,deep venous blood is shunted retrogradelythrough perforating channels into super-ficial veins, increasing intraluminal pres-sure and volume and causing dilatation andvaricosity formation.

Many people with varicose veins areasymptomatic but seek treatment for cos-metic reasons. When symptoms do develop,they include a dull ache, “heaviness,” or apressure sensation in the legs after pro-longed standing. Superficial venous insuffi-ciency may result when venous valves areunable to function normally in the dilatedveins. This can cause swelling and skin ul-ceration that is particularly severe near theankle. Stasis of blood within varicose veinscan promote superficial vein thrombosis,

364 Chapter Fifteen

Figure 15.8. A patient with extensive venous varicosities of theright leg.

10090-15_CH15.qxd 8/31/06 4:57 PM Page 364


and varicosities can also rupture, causing alocalized hematoma.

Varicose veins are usually treated conser-vatively. Patients should elevate their legswhile supine, avoid prolonged standing,and wear external compression stockingsthat counterbalance the increased venoushydrostatic pressure. Small symptomaticvaricosities are sometimes treated by injec-tion of a sclerosing agent into the vein. Lasertreatments can be used to improve the cos-metic appearance of small affected vessels.Endovenous laser therapy, radiofrequencyablation procedures, and surgical vein liga-tion and removal are used for patients whoare very symptomatic, suffer recurrent superficial vein thrombosis, or develop skinulcerations.

Venous Thrombosis

The term venous thrombosis or thrombophle-bitis is used to describe thrombus within asuperficial or deep vein and the inflamma-tory response in the vessel wall that it in-cites. Thrombi in the lower extremities areclassified by location as either deep venousthrombi or superficial venous thrombi.

Initially, the venous thrombus is com-posed principally of platelets and fibrin. Later,red blood cells become interspersed withinthe fibrin, and the thrombus tends to propa-gate in the direction of blood flow. Thechanges in the vessel wall can be minimal orcan include granulocyte infiltration, loss ofendothelium, and edema. Thrombi may di-minish or obstruct vascular flow, or theymay dislodge and form thromboemboli.

Deep Venous Thrombosis

Epidemiology, Etiology, and Pathogenesis

Deep venous thrombosis (DVT) occurs mostcommonly in the veins of the calves butmay also develop initially in more proximalveins such as the popliteal, femoral, andiliac vessels. If left untreated, 20% to 30% ofDVTs that occur in the calves may propagateto these proximal veins. The two major con-

sequences of deep venous thrombosis arepulmonary embolism and postphlebitic syn-drome. Pulmonary embolism can supervenewhen a clot, most often from a DVT in theproximal veins of the lower extremities, dis-lodges and travels through the inferiorvena cava and right-heart chambers, finallyreaching and obstructing a portion of thepulmonary vasculature (Fig. 15.9; also seeFig. 3.21). This complication may be her-alded by pleuritic chest pain, tachypnea,cough, and/or dyspnea. Pulmonary em-bolism is common (incidence of approxi-mately 600,000 per year in the UnitedStates) and is often fatal, with an untreatedmortality rate of 30% to 40%.

Postphlebitic syndrome, or chronic deepvenous insufficiency, results from venousvalvular damage and/or persistent occlusionby deep venous thrombosis. This may leadto chronic leg swelling, stasis pigmentation,and skin ulcerations.

In 1856, Virchow described a triad of fac-tors that predispose to venous thrombosis:(1) stasis of blood flow, (2) hypercoagulabil-ity, and (3) vascular damage. Stasis disruptslaminar flow and brings platelets into con-tact with the endothelium. This allows co-agulation factors to accumulate and retardsthe influx of clotting inhibitors. Factors thatslow venous flow and induce stasis includeimmobilization (e.g., prolonged bed restafter surgery, or sitting in a car or an airplanefor a long trip), cardiac failure, and hyper-viscosity syndromes (Table 15.6).

Various clinical disorders cause systemichypercoagulability, including resistance ofcoagulation factor V to activated protein C,a prothrombin gene mutation, and inher-ited deficiencies of antithrombin III, proteinC, and protein S. Pancreatic, lung, stomach,breast, and genitourinary tract adenocarci-nomas are associated with a high prevalenceof venous thrombosis. This is thought tooccur in part because necrotic tumor cells re-lease thrombogenic factors. Other condi-tions that contribute to hypercoagulabilityare listed in Table 15.6.

Vascular damage, either by external in-jury or intravenous catheters, can denudethe endothelium and expose subendothelial

Fig. 9

Tab. 6

10090-15_CH15.qxd 8/31/06 4:57 PM Page 365

collagen. Exposed collagen acts as a sub-strate for the binding of von Willebrand fac-tor and platelets and initiates the clottingcascade, leading to clot formation. Less se-vere damage can cause endothelial dysfunc-tion that contributes to thrombosis by dis-rupting the production of vasodilating andantiplatelet substances (e.g., nitric oxideand prostacyclin) and antithrombotic mol-ecules such as thrombomodulin and hep-aran sulfate.

The risk of venous thrombosis is partic-ularly high after fractures of the spine,pelvis, and bones of the lower extremities.The risk following bone fracture may be re-lated to stasis of blood flow, increased coag-ulability, and possibly traumatic endothelialdamage. In addition, venous thrombosis oc-curs frequently in patients following surgi-cal procedures, particularly major orthope-dic operations.

366 Chapter Fifteen

Figure 15.9. Pulmonary angiogram displaying a massive pulmonary embolism. There isa large filling defect in the left main pulmonary artery (arrow), additional filling defects in thelower pulmonary artery branches, and a paucity of vessels in the left midlung region (a resultof obstructed flow).

TABLE 15.6. Conditions That Predispose to DVT

Stasis of blood flowProlonged inactivity (following surgery, long travel

by car or plane)Immobilized extremity (following bone fracture)Heart failure (with systemic venous congestion)Hyperviscosity syndromes (e.g., polycythemia vera)Hypercoagulable statesInherited disorders of coagulation

Resistance to activated protein C (factor V Leiden)Prothrombin gene mutation (PT G20210A)Antithrombin III deficiencyDeficiency of protein C or protein S

Antiphospholipid antibodies/lupus anticoagulantNeoplastic disease (e.g., pancreatic, lung, stomach,

or breast cancers)Pregnancy and oral contraceptive useMyeloproliferative diseasesSmokingVascular damageInstrumentation (e.g., intravenous catheters)Trauma

10090-15_CH15.qxd 8/31/06 4:57 PM Page 366


Women have a several-fold increase inthe incidence of venous thrombus forma-tion during late pregnancy and the earlypostpartum period. In the third trimester,the fetus compresses the inferior vena cavaand can cause stasis of blood flow, andhigh levels of circulating estrogen may in-duce a hypercoagulable state. Oral contra-ceptives and other pharmacologic estrogenproducts may also predispose to thrombusformation.

Clinical Presentation

Patients with DVT may be asymptomatic.Symptomatic patients often describe calf orthigh discomfort, particularly when stand-ing or walking, or report unilateral legswelling. The physical signs of proximalDVT include edema of the involved leg andoccasionally localized warmth and erythema.Tenderness may be present over the courseof the phlebitic vein, and a deep venouscord (induration along the thrombosed ves-sel) is occasionally palpable. Calf pain pro-duced by dorsiflexion of the foot (Homanssign) is a nonspecific and unreliable sign of DVT.

Diagnosis

The primary laboratory tests for the diagno-sis of DVT include measurement of theserum D-dimer level and venous compres-sion ultrasonography. D-dimer, a byproductof fibrin degradation that can be measuredin a peripheral blood sample, is highly sen-sitive for the diagnosis of DVT and/or acutepulmonary embolism. Because D-dimermay also be elevated in many other condi-tions (such as cancer, inflammation, infec-tion, and necrosis), a positive test result isnot specific for DVT. Thus, a normal D-dimervalue can help exclude the presence of DVT,but an elevated level does not confirm thediagnosis.

Venous compression duplex ultrasonographyis a readily available noninvasive techniquethat is 95% sensitive for the diagnosis ofsymptomatic DVT in a proximal vein but

only 75% sensitive for diagnosing sympto-matic calf vein thrombi. This technique usesreal-time ultrasound scanning to image thevein and pulsed Doppler ultrasound to assessblood flow within it (Fig. 15.10). Criteria usedfor diagnosis of DVT with duplex ultra-sonography include the inability to compressthe vein with direct pressure (suggesting thepresence of an intraluminal thrombus), di-rect visualization of the thrombus, and ab-sence of blood flow within the vessel.

Other diagnostic techniques are some-times used. For example, magnetic resonancevenography can aid in the diagnosis of proxi-mal DVT, particularly pelvic vein thrombi,which are difficult to detect by ultrasound.Contrast venography is an invasive imagingtechnique that can provide a definitive diag-nosis. Radiocontrast material is administeredinto a foot vein, and images are obtained asthe contrast ascends through the venous sys-tem of the leg. DVT is diagnosed by the pres-ence of a filling defect (see Fig. 15.10).

Treatment

Elevation of the affected extremity abovethe level of the heart helps reduce edemaand tenderness, and anticoagulation pre-vents extension of the thrombus and pul-monary embolism. Initial anticoagulationusually consists of subcutaneous low molec-ular weight heparin (LMWH). Intravenous un-fractionated heparin is a cost-effective alter-native that has been used successfully forthis purpose for many years, although somestudies have shown superior outcomes withLMWH, which is also more convenient toadminister (see Chapter 17). Warfarin, anoral anticoagulant, is prescribed for long-term management and is usually continuedfor 6 months or more, depending on the un-derlying cause of DVT. Catheter-basedthrombolysis may be useful for selected pa-tients with iliofemoral deep vein thrombo-sis. In patients with proximal DVT who can-not be treated with anticoagulants becauseof a bleeding disorder, an intravascular filtercan be percutaneously inserted into the in-ferior vena cava to prevent emboli fromreaching the lungs.

Fig. 10

10090-15_CH15.qxd 8/31/06 4:57 PM Page 367

Treatment of patients with calf veinthrombosis is more controversial becausepulmonary emboli from that site are un-common. Some physicians advocate serialnoninvasive monitoring to determine ifthe thrombus propagates into proximalveins, whereas others treat such thrombo-sis with heparin (unfractionated or lowmolecular weight) followed by warfarin for3 to 6 months.

Prophylaxis against DVT is mandatory inclinical situations in which the risk of de-veloping the condition is high, such as dur-ing bed rest following surgery. Prophylacticmeasures include subcutaneous unfraction-ated heparin or low molecular weight heparin(fondaparinux; see Chapter 17), low-dose oralwarfarin, compression stockings, and/or in-

termittent external pneumatic compressionof the legs to prevent venous stasis.

Superficial Thrombophlebitis

Superficial thrombophlebitis is a benign dis-order associated with inflammation andthrombosis of a superficial vein, just belowthe skin. It may occur, for example, as acomplication of an in-dwelling intravenouscatheter. It is characterized by erythema,tenderness, and edema over the involvedvein. Treatment consists of local heat andrest of the involved extremity. Aspirin orother anti-inflammatory medications mayrelieve the associated discomfort. UnlikeDVT, superficial thrombophlebitis does notlead to pulmonary embolism.

368 Chapter Fifteen

Figure 15.10. Diagnostic imaging of deep venous thrombosis. A. Normal venogram. Contrast material wasinjected into a foot vein and fills the leg veins in this radiograph. B. Venogram demonstrating extensive thrombo-sis of the deep calf veins, popliteal vein, and superficial femoral vein. Arrow indicates a filling defect in the superfi-cial femoral vein (which is actually a deep vein despite its name) owing to the presence of thrombus. The deep calfveins are filled with thrombus and cannot be visualized. C. Ultrasound indicating deep venous thrombosis. Thethrombus appears as an echogenic area (arrow) within the femoral vein (V). A healthy vein would be easily com-pressible by the ultrasound transducer. This vein, however, has the same diameter at baseline (top panel) and aftercompression (bottom panel), confirming the presence of thrombus within it. Art, artery.

10090-15_CH15.qxd 8/31/06 4:57 PM Page 368


SUMMARY

1. Aortic aneurysms are of two types: true aneurysms and false (pseudo) aneu-rysms. True aneurysms are caused by de-generative changes in the aortic wall.Cystic medial necrosis is associated withascending thoracic aortic aneurysms,and atherosclerosis is commonly foundin descending thoracic and abdominalaortic aneurysms. A false aneurysm rep-resents a hole in the arterial intima andmedia contained by a layer of adventitiaor perivascular clot.

2. Symptoms of aortic aneurysms relate tocompression of adjacent structures (backpain, dysphagia, respiratory symptoms)or blood leakage. The most severe consequence is aneurysm rupture. Ananeurysm can be repaired either withan open surgical procedure or by inser-tion of an endovascular graft.

3. An aortic dissection results from a split-ting apart of a weakened medial layer of the aorta, often in the setting of advanced age, hypertension, or othercauses of medial degeneration (cysticmedial necrosis). Type A (proximal) aor-tic dissections involve the ascendingaorta, whereas type B dissections areconfined to the descending aorta. Theformer are more common, more devas-tating, and require surgical treatment.Type B dissections are often managed bymedical therapy alone.

4. PAD is a common atherosclerotic diseaseof large and medium-sized arteries, oftenresulting in claudication of the limbs. Itis treated by risk factor modification, exercise, antiplatelet agents, and some-times cilostazol, a selective phospho-diesterase inhibitor. Catheter-based or surgical revascularization procedures areimplemented in patients with disablingsymptoms or critical limb ischemia.

5. Acute arterial occlusion results fromthrombus formation in situ or from ar-terial embolism. The latter arises fromthrombus within the heart, from proxi-mal arterial sites, or paradoxically fromthe systemic veins in the presence of an

intracardiac shunt (e.g., atrial septal de-fect). Therapeutic options include anti-coagulation, thrombolysis, and surgicalor endovascular interventions.

6. Vasculitic syndromes are inflammatorydiseases of blood vessels that impair ar-terial flow and result in localized andsystemic symptoms. They are distin-guished from one another by the pat-tern of vessel involvement and mor-phologic findings (see Table 15.5).

7. Raynaud phenomenon is an episodicvasospasm of arteries that supply the dig-its of the upper and lower extremities. Itmay be a primary condition (Raynauddisease) or may appear in associationwith other disorders such as connectivetissue diseases or blood dyscrasias.

8. Varicose veins are dilated tortuous ves-sels that may present cosmetic prob-lems. Occasionally, they cause discom-fort, become thrombosed, or lead tovenous insufficiency. Initial manage-ment is conservative, with periodic legelevation and compression stockings.Severe symptomatic varicose veins canbe treated with radiofrequency ablation,laser therapy, or surgical ligation and removal.

9. Venous thrombosis results from stasis ofblood flow, hypercoagulability, and vas-cular damage. The major complication ofdeep venous thrombosis is pulmonaryembolism. A chronic complication is venous insufficiency causing chronic legswelling and skin ulceration.

10. D-dimer assay and venous compressionultrasonography are the primary toolsused to diagnose deep venous throm-bosis. Anticoagulation therapy withlow molecular weight heparin or un-fractionated intravenous heparin, fol-lowed by oral warfarin, is the usualtreatment.

AcknowledgmentContributors to the previous editions of this chapterwere Michael Diminick, MD; Mary Beth Gordon,MD; Stuart Kaplan, MD; Geoffrey McDonough, MD;Jesse Salmeron, MD; and Mark A. Creager, MD.

10090-15_CH15.qxd 8/31/06 4:57 PM Page 369

Additional Reading

Bates SM, Ginsburg JS. Treatment of deep vein throm-bosis. N Engl J Med 2004;351:268–277.

Creager MA, ed. Management of Peripheral ArterialDisease: Medical, Surgical and Interventional Aspects. London: Remedica, 2000.

Creager MA, Dzau VJ. Vascular diseases of the ex-tremities. In: Braunwald E, Fauci AS, Kasper D, et al., eds. Harrison’s Principles of Internal Medi-cine. 16th Ed. New York: McGraw-Hill, 2005.

Creager MA, Dzau VJ, Loscalzo J, eds. Vascular Med-icine: A Companion to Braunwald’s Heart Disease.Philadelphia: Elsevier Saunders, 2006.

Dzau VJ, Creager MA. Diseases of the aorta. In:Braunwald E, Fauci AS, Kasper D, et al., eds. Harri-son’s Principles of Internal Medicine. 16th Ed.New York: McGraw-Hill, 2005.

Faxon DP, Creager MA, Smith SC Jr, et al. Athero-sclerotic Vascular Disease Conference executivesummary. Circulation 2004;109:2595–2604.

Hankey GJ, Norman PE, Eikelboom JW. Medical treat-ment of peripheral arterial disease, JAMA 2006;295:547–553.

Hirsch AT, Haskal ZJ, Hertzer NR, et al. ACC/AHA2005 guidelines for the management of patientswith peripheral arterial disease (lower extremity,renal, mesenteric, and abdominal aortic). J AmColl Cardiol 2006;47:1239–1312. Available at:http://www.cardiosource.com/guidelines/guide-lines/pad/pad_execsum.pdf. Accessed July 25, 2006.

Isselbacher EM. Thoracic and abdominal aorticaneurysms. Circulation 2005;111:816–828.

Marso SP, Hiatt WR. Peripheral arterial disease in pa-tients with diabetes. J Am Coll Cardiol 2006;47:921–929.

Olin JW. Thromboangiitis obliterans (Buerger’s dis-ease). N Engl J Med 2000;343:864–869.

Smith SC, Allen J, Blair SN, et al. AHA/ACC guide-lines for secondary prevention for patients withcoronary and other atherosclerotic vascular dis-ease: 2006 update. J Am Coll Cardiol 2006;47:2130–2139.

Tsai TT, Nienaber CA, Eagle KA. Acute aortic syn-dromes. Circulation 2005;112:3802–3813.

Wigley FM. Raynaud’s phenomenon. N Engl J Med2002;347:1001–1008.

370 Chapter Fifteen

10090-15_CH15.qxd 8/31/06 4:57 PM Page 370

Chapter 15—Author Queries1. AU: Is “and” correct here, or should it be “or”?2. AU: Is this the correct term?3. AU: Please provide page number.

10090-15_CH15.qxd 8/31/06 4:57 PM Page a

371

NORMAL DEVELOPMENT OF THE CARDIOVASCULAR SYSTEMDevelopment of the Heart TubeFormation of the Heart LoopSeptationDevelopment of the Cardiac Valves

FETAL AND TRANSITIONAL CIRCULATIONSFetal CirculationTransitional Circulation

COMMON CONGENITAL HEART LESIONSAcyanotic LesionsCyanotic Lesions

EISENMENGER SYNDROME

C H A P T E R

16Congenital HeartDiseaseVijay G. SankaranDavid W. Brown

Congenital heart diseases are the most com-mon form of birth defects and are the lead-ing cause of death from birth abnormalitiesin the first year of life. These conditions af-fect approximately 8 of 1,000 live births,and an estimated 1 million people in theUnited States have congenital heart lesions.Some abnormalities are severe and requireimmediate medical attention, whereas manyare less pronounced and have minimal clin-ical consequences. Although congenital heartdefects are present at birth, milder defectsmay remain inapparent for weeks, months,or years and, not infrequently, escape detec-tion until adulthood.

The past half century has witnessed tre-mendous growth in the understanding ofthe pathophysiology of congenital heart dis-

eases and substantial improvements in theability to evaluate and treat those afflicted.Research has shown that genetic mutations,environmental factors, or maternal illnessor ingestion of toxins can contribute to car-diac malformations. However, specific eti-ologies remain unknown in most cases.

The survival of children with congenitalheart disease has also improved dramati-cally in recent decades, largely because ofbetter diagnostic and interventional tech-niques. However, the lifelong needs of af-fected patients include guidance regardingphysical activity, pregnancy, endocarditisprophylaxis, insurance, and employment.

Formation of the cardiovascular system be-gins during the third week of embryonic de-velopment. Soon after, a unique circulation

10090-16_CH16.qxd 8/31/06 4:57 PM Page 371

develops that allows the fetus to mature inthe uterus, using the placenta as the primaryorgan of gas, nutrient, and waste exchange.At birth, the fetal lungs inflate and becomefunctional, making the placenta unneces-sary and dramatically altering circulationpatterns to allow the neonate to adjust tolife outside the womb. Given the remarkablecomplexity of these processes, it is easy toenvision ways that cardiovascular malfunc-tions could develop.

This chapter begins with an overview offetal cardiovascular development and thendescribes the most common forms of con-genital heart disease.

NORMAL DEVELOPMENT OF THE CARDIOVASCULAR SYSTEM

During the third week of gestation, the nu-trient and gas exchange needs of the rapidlygrowing embryo can no longer be met bydiffusion alone, and the tissues begin to relyon the developing cardiovascular system todeliver these substances over long distances.

Development of the Heart Tube

In the middle of the third week of embryo-genesis, mesodermal cells proliferate at thecranial end of the early embryonic disc. Theyeventually form two longitudinal cell clustersknown as angioblastic cords. These cordscanalize and become paired endothelial hearttubes (Fig. 16.1). Lateral embryonic foldinggradually causes these two tubes to opposeone another and allows them to fuse in theventral midline, forming a single endocardialtube by day 22. From inside to outside, thelayers of this primitive heart tube are an en-dothelial lining that becomes the endo-cardium, a layer of gelatinous connective tis-sue (cardiac jelly), and a thick muscular layerthat is derived from the splanchnic meso-derm and develops into the myocardium.The endocardial tube is continuous with theaortic arch system rostrally and with the ve-nous system caudally. The primitive heart be-gins to beat around day 22 or 23, causingblood to circulate by the beginning of thefourth week. The space overlying the devel-

372 Chapter Sixteen

Dorsal aorta

Endocardial heart tube

Foregut

A

B

C

Fusing endocardial heart tubes

Endocardial heart tube

Foregut

Pericardialcavity

Figure 16.1. Embryonic transverse sections illustrat-ing fusion of the two heart tubes into a single endo-cardial heart tube. A. 18 days. B. 21 days. C. 22 days.

oping cardiac area eventually becomes thepericardial cavity, housing the future heart.

Formation of the Heart Loop

As the tubular heart grows and elongates, itdevelops a series of alternate constrictions

Fig. 1

10090-16_CH16.qxd 8/31/06 4:57 PM Page 372

Congenital Heart Disease 373

and dilations, creating the first sign of theprimitive heart chambers—the truncus arte-riosus, the bulbus cordis, the primitive ven-tricle, the primitive atrium, and the sinusvenosus (Fig. 16.2). Continued growth andelongation within the confined pericardialcavity force the heart tube to bend on itselfon day 23, eventually forming a U-shapedloop with the round end pointing ventrallyand to the right by day 28. The result of thislooping is placement of the atrium andsinus venosus above and behind the truncusarteriosus, bulbus cordis, and ventricle (Fig.16.3). At this point, neither definitive septabetween the developing chambers nor de-finitive valvular tissue have formed. Theconnection between the primitive atriumand ventricle is termed the atrioventricular(AV) canal. In time, the AV canal becomestwo separate canals, one housing the tricus-pid valve and the other the mitral valve. Thesinus venosus is eventually incorporatedinto the right atrium, forming both thecoronary sinus and a portion of the rightatrial wall. The bulbus cordis and truncus ar-teriosus contribute to the future ventricularoutflow tracts, forming parts of the proxi-mal aorta and pulmonary artery.

Septation

Septation of the developing atrium, AV canal,and ventricle occurs between the fourth and

sixth weeks. Although these events are de-scribed separately here, they actually occur si-multaneously.

Septation of the Atria

The primary atrial septum, also known asthe septum primum, begins as a ridge oftissue on the roof of the common atriumthat grows downward into the atrial cavity(Fig. 16.4). As the septum primum advances,it leaves a large opening known as the os-tium primum between the crescent-shapedleading edge of the septum and the endo-cardial cushions surrounding the AV canal.The ostium primum allows passage of bloodbetween the forming atria. Eventually, theseptum primum fuses with the superior as-pect of the endocardial cushions (describedin more detail in the next section), obliter-ating the ostium primum. However, beforeclosure of the ostium primum is complete,small perforations appear in the center ofthe septum primum that ultimately coa-lesce to form the ostium secundum, pre-serving a pathway for blood flow betweenthe atria (see Fig. 16.4). Following closure ofthe ostium primum, a second, more muscu-lar membrane, the septum secundum, be-gins to develop immediately to the right ofthe superior aspect of the septum primum.This septum grows downward and overlapsthe ostium secundum. The septum secun-

[Right and left atria]

Figure 16.2. The straight heart tube at approximately 22 days. The structures that willultimately form from each segment are listed in brackets.

Fig. 2

Fig. 3

Fig. 4

10090-16_CH16.qxd 8/31/06 4:57 PM Page 373

dum eventually fuses with the endocardialcushions, although only in a partial fashion,leaving an oval-shaped opening known asthe foramen ovale. The superior edge of theseptum primum then gradually regresses,leaving the lower edge to act as a “flaplike”valve that allows only right-to-left flowthrough the foramen ovale (Fig. 16.5). Dur-ing gestation, blood passes from the rightatrium to the left atrium because the pres-sure in the fetal right atrium is greater thanthat of the left atrium. This pressure gradi-ent changes direction postnatally, causing

the valve to close, as described later in thechapter.

Septation of the Atrioventricular Canal

Growth of the endocardial cushions con-tributes to atrial septation and, as describedlater, to the membranous portion of the in-terventricular septum. Endocardial cushionsinitially begin as swellings of the gelatinousconnective tissue layer within the AV canal.They are then populated by migrating cellsfrom the primitive endocardium and subse-

374 Chapter Sixteen

Primitive right atrium

Primitive right ventricle

Primitive left atrium

Conus cordis

Primitive left ventricle

Primitive atrium

Truncus arteriosus

Truncus arteriosus Bulbus

cordis

Bulboventricularsulcus

A B C

Figure 16.3. Formation of the heart loop. A, B. By day 24, continued growth and elongation within the confinedpericardial space necessitate bending of the heart tube on itself, forming a U-shaped loop that points ventrally andto the right. C. Looping eventually places the atria above and behind the primitive ventricles.

OstiumPrimum

SeptumPrimum

SeptumPrimum

SeptumPrimum

SeptumSecundum

EndocardialCushion

OstiumSecundum

OstiumSecundum

SeptumSecundum

SeptumSecundum

InterventricularForamen

InterventricularForamen

ForamenOvale

InterventricularSeptum

Muscular

InterventricularSeptum

Membranous

A B

DC

Figure 16.4. Atrial septal formation at 30 days (A), 33 days (B), and 37 days (C) of develop-ment as well as in the newborn (D). As the septum primum grows toward the ventricles, theopening between it and the AV canal is the ostium primum. Before the ostium primum completelycloses, perforations within the upper portion of the septum primum form the ostium secundum. Asecond ridge of tissue, the septum secundum, grows downward to the right of the septum primum,partially covering the ostium secundum. The foramen ovale is an opening of the septum secundumthat is covered by the “flap valve” of the lower septum primum. (Modified from Moss AJ, Adams FH.Heart Disease in Infants, Children, and Adolescents. Baltimore: Williams & Wilkins, 1968:16.) AQ2

Fig. 5

10090-16_CH16.qxd 8/31/06 4:57 PM Page 374


right and left canals that later give rise to thetricuspid and mitral orifices, respectively.

Septation of the Ventricles andVentricular Outflow Tracts

At the end of the fourth week, the primitiveventricle begins to grow, leaving a medianmuscular ridge, the primitive interventricularseptum. Most of the early increase in heightof the septum results from dilation of the twonew ventricles forming on either side of it.Only later does new cell growth in the sep-tum itself contribute to its size. The free edgeof the muscular interventricular septum doesnot fuse with the endocardial cushions; theopening that remains and allows communi-cation between the right and left ventricles isthe interventricular foramen (Fig. 16.7). Thisremains open until the end of the seventhweek of gestation, when the fusion of tissuefrom the right and left bulbar ridges and theendocardial cushions forms the membranousportion of the interventricular septum.

During the fifth week, neural crest-derived mesenchymal proliferation occur-ring in the bulbus cordis and truncus arte-riosus creates a pair of protrusions known asthe bulbar ridges (Fig. 16.8). These ridgesfuse in the midline and undergo a 180° spi-raling process, forming the aorticopulmonaryseptum. This septum divides the bulbus cor-dis and the truncus arteriosus into two arte-rial channels, the pulmonary artery and theaorta, the former continuous with the rightventricle (RV) and the latter with the leftventricle (LV).

Shunt

A

B

Figure 16.5. Diagrammatic depiction of the flap-type valve of the foramen ovale. A. Before birth, thevalve permits only right-to-left flow of blood from thehigher-pressured right atrium (RA) to the lower-pressuredleft atrium (LA). B. Following birth, the pressure in the LAbecomes greater than that in the RA, causing the septumprimum to close firmly against the septum secundum.(Modified from Moore KL, Persaud TVN. The DevelopingHuman. Philadelphia: WB Saunders, 1993:318.)AQ3

Rightatrioventricular

canal

Lateralendocardial

cushion

Lateralendocardial

cushion

Superiorendocardial

cushion

Inferiorendocardial

cushion

Commonatrioventricular

canal

Leftatrioventricular

canal

Figure 16.6. The progression of septal formation in the atrioventricular canal through successive stages.The septum forms through growth of the superior, inferior, and lateral endocardial cushions. The endocardial cush-ions are masses of mesenchymal tissue that surround the atrioventricular canal and aid in the formation of the ori-fices of the mitral and tricuspid valves, as well as the upper interventricular septum and lower interatrial septum.

quently transform into mesenchymal tissue.Tissue growth occurs primarily in the hori-zontal plane, resulting in septation of the AVcanal through the continued growth of thelateral, superior, and inferior endocardialcushions (Fig. 16.6). Septation creates theFig. 6

Fig. 7

Fig. 8

10090-16_CH16.qxd 8/31/06 4:57 PM Page 375

Development of the Cardiac Valves

Semilunar Valve Development (Aortic and Pulmonary Valves)

The semilunar valves start to develop justbefore the completion of the aorticopul-monary septum. The process begins whenthree outgrowths of subendocardial mes-enchymal tissue form around both the aor-tic and pulmonary orifices. These growthsare ultimately shaped and excavated by thejoint action of programmed cell death andblood flow to create the three thin-walledcusps of both the aortic and pulmonaryvalves.

Atrioventricular Valve Development(Mitral and Tricuspid Valves)

After the endocardial cushions fuse to formtheseptabetweentheright and left AV canals,the surrounding subendocardial mesen-chymal tissue proliferates and develops out-growths similar to those of the semilunarvalves. These are also sculpted by pro-grammed cell death that occurs within theinferior surface of the nascent leaflets and inthe ventricular wall. This process leaves be-hind only a few fine muscular strands toconnect the valves to the ventricular wall(Fig. 16.9). The superior portions of thesestrands eventually degenerate and are re-placed by strings of dense connective tissue,becoming the chordae tendineae.

376 Chapter Sixteen

Interventricularforamen

Muscularinterventricularseptum

Figure 16.7. The interventricular septum and the in-terventricular foramen. (Modified from Moore KL, Per-saud TVN. The Developing Human. Philadelphia: WBSaunders, 1993:325.)

Aorta

A

Pulmonaryartery

Left bulbarridge


Muscularpart ofinterventricularseptum

Right bulbarridge

Fused endocardialcushions

Pulmonaryartery

Aorticopulmonaryseptum

Muscularpart ofinterventricularseptum

B


Left atrio-ventricularcanal

Right atrio-ventricularcanal

Endocardial cushion

C

Rightventricle

Membranouspart ofinterventricularseptum

Figure 16.8. Formation of the aorticopulmonary sep-tum occurs via fusion of the bulbar ridges, resultingin division of the bulbus cordis and truncus arterio-sus into the aorta and pulmonary artery (A, 5 weeks;B, 6 weeks; C, 7 weeks). The bulbus cordis becomes theright ventricular outflow tract. Fusion of tissue from theendocardial cushions, the aorticopulmonary septum, andthe muscular interventricular septum creates the mem-branous interventricular septum. (Modified from MooreKL, Persaud TVN. The Developing Human. Philadelphia:WB Saunders, 1993:322.)

Fig. 9

10090-16_CH16.qxd 8/31/06 4:57 PM Page 376



Lumen of leftventricle

A

Developing mitralvalve

Membranous partof interventricularseptum

BTrabeculae carneae

C

Mitral valve

Chordae tendineae

Papillary muscle

Ventricular conductionsystem

Atrioventricular node

Figure 16.9. Proliferation of mesenchymal tissue sur-rounding the atrioventricular canals forms the atri-oventricular valves. Degeneration of myocardium andreplacement by connective tissue forms the chordaetendineae; their muscular attachments to the ventricularwall are the papillary muscles. (Modified from Moore KL,Persaud TVN. The Developing Human. Philadelphia: WBSaunders, 1993:325.)

FETAL AND TRANSITIONALCIRCULATIONS

The fetal circulation elegantly serves theneeds of in utero development. At birth, thecirculation automatically undergoes modifi-cations that establish the normal blood flowpattern of a newborn infant.

Fetal Circulation

In fetal life, oxygenated blood leaves theplacenta through the umbilical vein (Fig.16.10). Approximately half of this blood isshunted through the fetal ductus venosus,bypassing the hepatic vasculature, and pro-ceeding directly into the inferior vena cava(IVC). The remaining blood passes throughthe portal vein to the liver and then into theIVC through the hepatic veins. IVC blood istherefore a mixture of well-oxygenated um-bilical venous blood and the blood of lowoxygen tension returning from the systemicveins of the fetus. Because of this mixture,the oxygen tension of inferior vena cavalblood is higher than that of blood returningto the fetal right atrium from the superiorvena cava. This distinction is important be-cause these two streams of blood are partiallyseparated within the right atrium to followdifferent circulatory paths. The consequenceof this separation is that the fetal brain and myocardium receive blood of relativelyhigher oxygen content, whereas the morepoorly oxygenated blood is diverted to theplacenta (via the descending aorta and um-bilical arteries) for subsequent oxygenation.

Most IVC blood entering the right atriumis directed to the left atrium through theforamen ovale. This intracardiac shunt ofrelatively well-oxygenated blood is facili-tated by the inferior border of the septum se-cundum, termed the crista dividens, whichis positioned such that it overrides the open-ing of the IVC into the right atrium. Thisshunted blood then mixes with the smallamount of poorly oxygenated blood return-ing to the left atrium through the fetal pul-monary veins (remember that the lungs arenot ventilated in utero; the developing pul-monary tissues actually remove oxygen fromthe blood). From the left atrium, blood flows

AQ4

Fig. 10

10090-16_CH16.qxd 8/31/06 4:57 PM Page 377

into the LV and is then pumped into the as-cending aorta. This well-oxygenated bloodis distributed primarily to three territories:(1) approximately 9% enters the coronaryarteries and perfuses the myocardium, (2) 62% travels in the carotid and subclavianvessels to the upper body and brain, and (3) 29% passes into the descending aorta tothe rest of the fetal body.

The remaining well-oxygenated inferiorvena caval blood entering the right atriummixes with poorly oxygenated blood fromthe superior vena cava and passes to the RV.In the fetus, the RV is the actual “work-horse” of the heart, providing two thirds ofthe total cardiac output. This output flowsinto the pulmonary artery and from thereeither through the ductus arteriosus into

378 Chapter Sixteen

Placenta

Umbilicalarteries

Urinarybladder

Internal iliacartery

Lung

Superior vena cava

Right atrium

Inferior vena cava

Umbilicus

Umbilical vein

Portal vein

Foramen ovale

Arch of aorta

Pulmonary trunk

Left atrium

Descending aorta

Low

Ductus venosus

Pulmonary veins

Ductus arteriosus

Legs

Portal sinus

Right hepaticvein

Left hepatic vein

Sphincter

Medium

High

Oxygen saturationof blood:

Figure 16.10. The fetal circulation. Arrows indicate the direction of blood flow. Three shunts(ductus venosus, foramen ovale, and ductus arteriosus) allow most of the blood to bypass the lungsand liver during fetal life but cease to function shortly after birth. (Modified from Moore KL, Persaud TVN. The Developing Human. Philadelphia: WB Saunders, 1993:344.)

10090-16_CH16.qxd 8/31/06 4:57 PM Page 378


the descending aorta (88% of RV output) orthrough the pulmonary arteries and into thelungs (12% of RV output). This unequal dis-tribution of right ventricular outflow is ac-tually quite efficient. Bypassing the lungs isdesired because the fetal lungs are filled withamniotic fluid and are incapable of gas ex-change. The low oxygen tension of this fluidcauses constriction of the pulmonary ves-sels, which increases pulmonary vascular re-sistance and facilitates shunting of bloodthrough the ductus arteriosus to the systemiccirculation. From the descending aorta, bloodis distributed to the lower body and to theumbilical arteries, leading back to the pla-centa for gas exchange.

Transitional Circulation

Immediately following birth, the neonaterapidly adjusts to life outside the womb. Thenewly functioning lungs replace the pla-centa as the organ of gas exchange, and thethree shunts (ductus venosus, foramenovale, and ductus arteriosus) that operatedduring gestation close. This shift in the siteof gas exchange and the resulting changesin cardiovascular architecture allow thenewborn to survive independently.

As the umbilical cord is clamped or con-stricts naturally, the low-resistance placen-tal flow is removed from the arterial system,resulting in an increase in systemic vascularresistance. Simultaneously, pulmonary vas-cular resistance falls for two reasons: (1) themechanical inflation of the lungs after birthstretches the lung tissues, causing pulmonaryartery expansion and wall thinning, and (2) vasodilatation of the pulmonary vascula-ture occurs in response to the rise in bloodoxygen tension accompanying aeration ofthe lungs. This reduction in pulmonary re-sistance results in a dramatic rise in pul-monary blood flow. It is most marked withinthe first day after birth but continues for thenext several weeks until adult levels of pul-monary resistance are achieved.

As pulmonary resistance falls and moreblood travels to the lungs through the pul-monary artery, venous return from the pul-monary veins to the left atrium also in-

creases, causing left atrial pressure to rise. Atthe same time, cessation of umbilical ve-nous flow and constriction of the ductusvenosus cause a fall in IVC and right atrialpressures. As a result, the left atrial pressurebecomes greater than that in the rightatrium, and the valve of the foramen ovaleis forced against the septum secundum,eliminating the previous flow between theatria (see Fig. 16.5).

With oxygenation now occurring in thenewborn lungs, the ductus arteriosus be-comes superfluous and begins to constrict.During fetal life, a high circulating level ofprostaglandin E1 (PGE1) is generated in re-sponse to relative hypoxia, which causes thesmooth muscle of the ductus arteriosus torelax, keeping it patent. After birth, PGE1

levels decline as the oxygen tension risesand the ductus constricts. The responsive-ness of the ductus to vasoactive substancesdepends on the gestational age of the fetus.The ductus often fails to constrict in prema-ture infants, resulting in the congenitalanomaly known as patent ductus arteriosus(described later in the chapter).

With the anatomic separation of the cir-culatory paths of the right and left sides ofthe heart now complete, the stroke volumeof the LV increases and that of the RV de-creases, equalizing the cardiac output fromboth ventricles. The augmented pressureand volume load placed on the LV inducesthe myocardial cells of that chamber tohypertrophy, while the decreased pressureand volume loads on the RV result in grad-ual regression of RV wall thickness.

COMMON CONGENITAL HEART LESIONS

Congenital heart defects are generally welltolerated before birth. The fetus benefitsfrom shunting of blood through the ductusarteriosus and the foramen ovale, allowingthe bypass of most defects. It is only afterbirth, when the neonate has been separatedfrom the maternal circulation and the oxy-genation it provides and the fetal shuntshave closed, that congenital heart defectsusually become manifest.

10090-16_CH16.qxd 8/31/06 4:57 PM Page 379

Congenital heart lesions can be catego-rized as cyanotic or acyanotic. Cyanosis refersto a blue-purple discoloration of the skinand mucous membranes caused by an ele-vated blood concentration of deoxygenatedhemoglobin (at least 4 g/dL, which corre-sponds to an arterial O2 saturation of ap-proximately 80% to 85%). In congenitalheart disease, cyanosis results from defectsthat allow poorly oxygenated blood from theright side of the heart to be shunted to theleft side, bypassing the lungs.

Acyanotic lesions include intracardiac orvascular stenoses, valvular regurgitation,and defects that result in left-to-right shunt-ing of blood. Large left-to-right shunts at theatrial, ventricular, or great vessel level (alldescribed in the following sections) causethe pulmonary artery volume and pressureto increase and can be associated with thelater development of pulmonary arteriolarhypertrophy and increased resistance toflow. Over time, the elevated pulmonary resistance may force the direction of theoriginal shunt to reverse, causing right-to-left

flow to supervene, accompanied by thephysical findings of hypoxemia and cyano-sis. The development of pulmonary vasculardisease as a result of a chronic large left-to-right shunt is known as Eisenmenger syn-drome and is described in greater detail inthe final section of the chapter.

Acyanotic Lesions

Atrial Septal Defect

An atrial septal defect (ASD) is a persistentopening in the interatrial septum after birththat allows direct communication betweenthe left and right atria. ASDs are relativelycommon, occurring with an incidence of 1in 1,500 live births. They can occur any-where along the atrial septum, but the mostcommon site is at the region of the foramenovale, termed an ostium secundum ASD (Fig.16.11). This defect arises from excessive re-sorption or inadequate development of theseptum primum, inadequate formation ofthe septum secundum, or a combination.Less commonly, an ASD appears in the infe-

380 Chapter Sixteen

ASD

Figure 16.11. Atrial septal defect (ASD), ostium secundum type. A. The arrow indicates shuntedflow from the left atrium (LA) toward the right atrium (RA). B. Schematic representation of blood flowthrough an uncomplicated ASD, resulting in enlargement of the RA, right ventricle (RV), and pulmonaryartery (PA). Ao, aorta; IVC, inferior vena cava; LV, left ventricle; SVC, superior vena cava.

Fig. 11

10090-16_CH16.qxd 8/31/06 4:57 PM Page 380


rior portion of the interatrial septum, adja-cent to the AV valves. Named an ostium pri-mum defect, this abnormality results fromfailure of the septum primum to fuse withthe endocardial cushions and is typically as-sociated with abnormal development of themitral and tricuspid valves. A third type ofASD occurs in the superior portion of theatrial septum near the entry of the superiorvena cava and is termed a sinus venosus ASD.It results from incomplete absorption of thesinus venosus into the right atrium and isoften accompanied by the anomalous drain-age of pulmonary veins from the right lunginto the right atrium.

In contrast, a patent foramen ovale (PFO),which is thought to be present in approxi-mately 20% of the general population, is nota true ASD. As described earlier, the foramenovale functionally shuts in the days afterbirth, and it anatomically closes by the ageof 6 months through fusion of the atrialsepta. PFOs arise this fusion fails to occur.

A PFO is usually clinically silent becausethe one-way valve, though not sealed, re-mains functionally closed because the LApressure is higher compared with that in the right atrium. However, a PFO takes onadded significance if the right atrial pressurebecomes elevated (e.g., in states of pulmo-nary hypertension or right-heart failure), re-sulting in pathologic right-to-left intracardiacshunting. In that case, deoxygenated bloodpasses directly into the arterial circulation.Occasionally, a PFO can be implicated in a patient who has suffered a systemic em-bolism (e.g., a stroke). This situation, termedparadoxical embolism, occurs when throm-bus in a systemic vein breaks loose, travelsto the right atrium, passes across the PFO tothe left atrium (if right-to-left shunting ispresent because of elevated right-heart pres-sures), and then into the systemic arterialcirculation.

Pathophysiology

In the case of an uncomplicated ASD, oxy-genated blood from the left atrium is shuntedinto the right atrium, but not vice versa.Flow through the defect is a function of its

size and the filling properties (compliance)of the ventricles into which the atria passtheir contents. Normally after birth, rightventricular compliance becomes greaterthan that of the LV owing to the regressionof right ventricular wall thickness, facilitat-ing a left-to-right directed shunt. The resultis volume overload and enlargement of theright atrium and RV (see Fig. 16.11B). Ifright ventricular compliance decreases overtime (because of the excessive load), the left-to-right shunt may lessen. Occasionally, ifsevere pulmonary vascular disease develops(e.g., Eisenmenger syndrome), the directionof the shunt may actually reverse (causingright-to-left flow), such that desaturatedblood enters the systemic circulation, re-sulting in hypoxemia and cyanosis.

Symptoms

Most infants with ASDs are asymptomatic.The condition is typically detected by thepresence of a murmur on routine physicalexamination during childhood or adoles-cence. If symptoms do occur, they includedyspnea on exertion, fatigue, and recurrentlower respiratory tract infections. The mostcommon symptoms in adults are decreasedstamina and palpitations owing to atrialtachyarrhythmias resulting from right atrialenlargement.


A prominent systolic impulse may be pal-pated along the lower-left sternal border,representing contraction of the dilated RV(termed RV heave). The second heart sound(S2) demonstrates a widened, fixed splittingpattern (see Chapter 2), because the normalrespiratory variation in systemic venous re-turn is countered by reciprocal changes inthe volume of blood shunted across theASD. The increased volume of blood flowingacross the pulmonary valve often creates asystolic murmur at the upper-left sternalborder. A middiastolic murmur may also bepresent at the lower-left sternal border owingto the increased flow across the tricuspidvalve. Blood traversing the ASD itself does

10090-16_CH16.qxd 8/31/06 4:57 PM Page 381

not produce a murmur because of the ab-sence of a significant pressure gradient be-tween the two atria.

Diagnostic Studies

On chest radiographs, the heart is usually enlarged because of right atrial and rightventricular dilatation, and the pulmonaryartery is prominent with increased pulmo-nary vascular markings. The electrocardio-gram (ECG) shows right ventricular hyper-trophy, often with right atrial enlargementand incomplete or complete right bundlebranch block. In patients with the ostiumprimum type of ASD, left axis deviation iscommon and is thought to be a result of dis-placement and hypoplasia of the left bundlebranch’s anterior fascicle. Echocardiographydemonstrates right atrial and right ventricu-lar enlargement; the ASD may be visualizeddirectly, or its presence may be implied bythe demonstration of a transatrial shunt byDoppler flow interrogation. The magnitudeand direction of shunt flow and an estima-tion of right ventricular systolic pressure canbe determined.

Given the high sensitivity of echocardio-graphy, it is rarely necessary to perform car-diac catheterization to confirm the presenceof an ASD. Catheterization may be useful toassess the pulmonary vascular resistanceand to diagnose concurrent coronary arterydisease in older adults. In a normal personundergoing cardiac catheterization, the oxy-gen saturation measured in the right atriumis similar to that in the superior vena cava.However, an ASD with left-to-right shuntingof well-oxygenated blood causes the satura-tion in the right atrium to be much greaterthan that of the superior vena cava.

Treatment

Most patients with ASDs remain asympto-matic. However, if the volume of shuntedblood is large (even in the absence of symp-toms), elective surgical repair is recom-mended to prevent the development ofheart failure or pulmonary vascular disease.The defect is repaired by direct suture clo-

sure or with a pericardial or synthetic patch.In children and young adults, morphologicchanges in the right heart often return tonormal after repair. Percutaneous ASD re-pair, using a closure device deployed via anintravenous catheter, is a less invasive alter-native to surgery in some patients.

Ventricular Septal Defect

A ventricular septal defect (VSD) is an ab-normal opening in the interventricular sep-tum (Fig. 16.12). VSDs are relatively com-mon, having an incidence of 1.5 to 3.5 per1,000 live births. They are most often lo-cated in the membranous (70%) and mus-cular (20%) portions of the septum, withfew defects occurring just below the aorticvalve or adjacent to the AV valves.

Pathophysiology

The hemodynamic changes that accom-pany VSDs depend on the size of the defectand the relative resistances of the pulmonaryand systemic vasculatures. In small VSDs,the defect itself offers more resistance toflow than the pulmonary or systemic vascu-lature; thus, the magnitude of the shunt de-pends on the size of the hole. Conversely,with larger “nonrestrictive” defects, thevolume of the shunt is determined by therelative pulmonary and systemic vascularresistances. In the perinatal period, the pul-monary vascular resistance approximatesthe systemic vascular resistance, and mini-mal shunting occurs between the two ventri-cles. After birth, however, as the pulmonaryvascular resistance falls, an increasing left-to-right shunt through the defect develops.When this shunt is large, the RV, pulmonarycirculation, left atrium, and LV experience arelative volume overload. Initially, the in-creased blood return to the LV augmentsstroke volume (via the Frank-Starling mech-anism); but over time, the increased volumeload can result in chamber dilatation, sys-tolic dysfunction, and symptoms of heartfailure. In addition, the augmented circula-tion through the pulmonary vasculature cancause pulmonary vascular disease as early as

382 Chapter Sixteen

Fig. 12Fig. 12

10090-16_CH16.qxd 8/31/06 4:57 PM Page 382


2 years of age. As pulmonary vascular resis-tance increases, the intracardiac shunt mayreverse its direction (Eisenmenger syndrome),leading to systemic hypoxemia and cyanosis.

Symptoms

Patients with small VSDs typically remainsymptom free. Conversely, 10% of infantswith VSDs have large defects and will de-velop early symptoms of congestive heartfailure, including tachypnea, poor feeding,failure to thrive, and frequent lower respi-ratory tract infections. Patients with VSDscomplicated by pulmonary vascular diseaseand reversed shunts may present with dysp-nea and cyanosis. Bacterial endocarditis (seeChapter 8) can develop, regardless of thesize of the VSD.


The most common physical finding is aharsh holosystolic murmur that is best heard

at the left sternal border. Smaller defectstend to have the loudest murmurs becauseof the great turbulence of flow that theycause. A systolic thrill can commonly bepalpated in the region of the murmur. Inaddition, a middiastolic rumble can oftenbe heard at the apex owing to the increasedflow across the mitral valve. If pulmonaryvascular disease develops, the holosystolicmurmur diminishes as the pressure gra-dient across the defect decreases. In such patients, an RV heave, a loud pulmonic closure sound (P2), and cyanosis may beevident.

Diagnostic Studies

On chest radiographs, the cardiac silhouettemay be normal in patients with small de-fects, but in those with large shunts, cardio-megaly and prominent pulmonary vascularmarkings are present. If pulmonary vasculardisease has developed, enlarged pulmonaryarteries with peripheral tapering may be

VSD

Figure 16.12. Ventricular septal defect (VSD). A. The arrow indicates shunted flow from the left ven-tricle (LV) toward the right ventricular (RV) outflow tract. B. Schematic representation of blood flowthrough an uncomplicated VSD. The dashed lines represent increased blood return to the left side of theheart as a result of the shunt, which causes enlargement primarily of the left atrium (LA) and LV. Ao, aorta;IVC, inferior vena cava; PA, pulmonary artery; RA, right atrium; SVC, superior vena cava.

10090-16_CH16.qxd 8/31/06 4:57 PM Page 383

evident. The ECG shows left atrial enlarge-ment and left ventricular hypertrophy inthose with a large shunt, and right ventricularhypertrophy is usually evident if pulmonaryvascular disease is present. Echocardiographywith Doppler studies can accurately deter-mine the location of the VSD, identify thedirection and magnitude of the shunt, andprovide an estimate of right ventricular sys-tolic pressure. Cardiac catheterization demon-strates increased oxygen saturation in theRV compared with the right atrium, the re-sult of shunting of highly oxygenated bloodfrom the LV into the RV.

Treatment

By age 2, at least 50% of small and moderate-sized VSDs undergo sufficient partial or com-plete spontaneous closure to make inter-vention unnecessary. Surgical correction ofthe defect is recommended in the first fewmonths of life for children with congestiveheart failure or pulmonary vascular disease.Moderate-sized defects without pulmonary

vascular disease but with significant volumeoverload can be corrected later in childhood.Less invasive catheter-based treatments arestill investigational. Medical managementincludes endocarditis prophylaxis for all patients with VSDs.

Patent Ductus Arteriosus

The ductus arteriosus is the vessel that con-nects the left pulmonary artery to the de-scending aorta during fetal life. Patent duc-tus arteriosus (PDA) results when the ductusfails to close after birth, resulting in a persis-tent connection between the great vessels(Fig. 16.13). It has an overall incidence ofabout 1 in 2,500 to 5,000 live term births.Risk factors for its presence include firsttrimester maternal rubella infection, prema-turity, and birth at a high altitude.

Pathophysiology

As described earlier, the smooth muscle ofthe ductus arteriosus usually constricts after

384 Chapter Sixteen

PDA

Figure 16.13. Patent ductus arteriosus (PDA). A. The arrow indicates shunted flow from the de-scending aorta (Ao) toward the pulmonary artery (PA). B. Schematic representation of blood flowthrough an uncomplicated PDA. The dashed lines represent increased blood return to the left side ofthe heart as a result of the shunt, which causes enlargement of the left atrium (LA), left ventricle (LV),and Ao. IVC, inferior vena cava; RA, right atrium; RV, right ventricle; SVC, superior vena cava.

Fig. 13

10090-16_CH16.qxd 8/31/06 4:57 PM Page 384


birth owing to the sudden rise in blood oxy-gen tension and a reduction in the level ofcirculating prostaglandins. Over the nextseveral weeks, intimal proliferation and fi-brosis result in permanent closure. Failure ofthe ductus to close results in a persistentshunt between the descending aorta and theleft pulmonary artery. The magnitude offlow through the shunt depends on thecross-sectional area and length of the ductusitself as well as the relative resistances of thesystemic and pulmonary vasculatures. Pre-natally, when the pulmonary vascular resis-tance is high, the blood is diverted awayfrom the immature lungs to the aorta. Asthe pulmonary resistance drops postna-tally, the shunt reverses direction and bloodflows from the aorta into the pulmonary circulation instead. Because of this left-to-right shunt, the pulmonary circulation, leftatrium, and LV become volume overloaded.This can lead to left ventricular dilatationand left-sided heart failure, whereas theright heart remains normal unless pulmo-nary vascular disease ensues. If the latterdoes develop, Eisenmenger syndrome results,with reversal of the shunt causing blood toflow from the pulmonary artery, throughthe ductus, to the descending aorta. The re-sulting flow of desaturated blood to thelower extremities causes cyanosis of the feet;the upper extremities are not cyanotic, be-cause they receive normally saturated bloodfrom the proximal aorta.

Symptoms

Children with small PDAs are generallyasymptomatic. Those with large left-to-right shunts develop early congestive heartfailure with tachycardia, poor feeding, slowgrowth, and recurrent lower respiratorytract infections. Moderate-sized lesions canpresent with fatigue, dyspnea, and palpita-tions in adolescence and adult life. Atrialfibrillation may occur owing to left atrialdilatation. Turbulent blood flow acrossthe defect can set the stage for endovascu-lar infection, similar to endocarditis (seeChapter 8) but more accurately termed endarteritis.


The most common finding in a patient witha left-to-right shunt through a PDA is a con-tinuous, machinelike murmur (see Fig. 2.10),heard best at the left subclavicular region. Themurmur is present throughout the cardiaccycle because a pressure gradient exists be-tween the aorta and pulmonary artery in bothsystole and diastole. However, if pulmonaryvascular disease develops, the gradient be-tween the aorta and the pulmonary artery de-creases, leading to diminished flow throughthe PDA, and the murmur becomes shorter(the diastolic component may disappear). IfEisenmenger syndrome develops, lower ex-tremity cyanosis and clubbing may be presenton examination because poorly oxygenatedblood is shunted to the descending aorta.

Diagnostic Studies

With a large PDA, the chest radiograph showsan enlarged cardiac silhouette (left atrial andleft ventricular enlargement) with promi-nent pulmonary vascular markings. In adults,calcification of the ductus may be visual-ized. The ECG shows left atrial enlargementand left ventricular hypertrophy when alarge shunt is present. Echocardiography withDoppler imaging can visualize the defect,demonstrate flow through it, and estimateright-sided systolic pressures. Cardiac catheter-ization is usually unnecessary for diagnosticpurposes. When performed in patients witha left-to-right shunt, it demonstrates a stepup in oxygen saturation in the pulmonaryartery compared with the RV, and angio-graphy shows the abnormal flow of bloodthrough the PDA.

Treatment

In the absence of other congenital cardiacabnormalities or severe pulmonary vasculardisease, a PDA should generally be therapeu-tically occluded. Although many sponta-neously close during the first months afterbirth, this rarely occurs later. Given the constant risk of endarteritis and the minimalcomplications of corrective procedures, evena small asymptomatic PDA is commonly re-

10090-16_CH16.qxd 8/31/06 4:57 PM Page 385

ferred for closure. For neonates and prema-ture infants with congestive heart failure, atrial of prostaglandin synthesis inhibitors(e.g., indomethacin) can be administered inan attempt to constrict the ductus. Definitiveclosure can be accomplished by surgical divi-sion or ligation of the ductus or by trans-catheter techniques in which an occludingcoil or other vascular occlusion device isplaced.

Congenital Aortic Stenosis

Congenital aortic stenosis (AS) is most oftencaused by abnormal development of thevalve (Fig. 16.14). It occurs in 5 of 10,000live births and is four times as common inmales as in females. Twenty percent of pa-tients have an additional abnormality, mostcommonly coarctation of the aorta (dis-cussed later in the chapter). The aortic valvein congenital AS usually has a bicuspidleaflet structure instead of the normal three-leaflet configuration, causing an eccentric

stenotic opening through which blood isejected. Bicuspid aortic valves are common,appearing in approximately 2% to 4% of thepopulation. Although they rarely result incongenital AS, they are a common cause ofAS in adults because the leaflets fibrose andcalcify over time (see Chapter 8).

Pathophysiology

Because the valvular orifice is significantlynarrowed, left ventricular systolic pressuremust increase to pump blood across thevalve into the aorta. In response to this in-creased pressure load, the LV hypertrophies.The high-velocity jet of blood that passesthrough the stenotic valve may impact onthe proximal aortic wall and contribute todilatation of that vessel.

Symptoms

The clinical picture of AS depends on theseverity of the lesion. Fewer than 10% of in-

386 Chapter Sixteen

Figure 16.14. Congenital aortic valve stenosis. A. The arrow points to the narrowed aortic valve.B. Schematic representation of obstructed flow through the narrowed aortic valve (jagged arrow). Leftventricular (LV) hypertrophy results from the chronic increased pressure load. Poststenotic dilatation ofthe aorta (Ao) is common. IVC, inferior vena cava; LA, left atrium; PA, pulmonary artery; RA, rightatrium; RV, right ventricle; SVC, superior vena cava.

Fig. 14

10090-16_CH16.qxd 8/31/06 4:57 PM Page 386


fants experience symptoms of heart failurebefore age 1, but if they do, they manifesttachycardia, tachypnea, failure to thrive,and poor feeding. Most older children withcongenital AS are asymptomatic and de-velop normally. When symptoms do occur,they are similar to those of adult AS and in-clude becoming fatigued easily, exertionaldyspnea, angina pectoris, and syncope (seeChapter 8).


Auscultation reveals a harsh crescendo-decrescendo systolic murmur, loudest at thebase of the heart with radiation to the neck.It is often preceded by a systolic ejectionclick (see Chapter 2), especially when a bi-cuspid valve is present. Unlike the murmursof ASD, VSD, or PDA, the murmur of con-genital AS is characteristically present frombirth because it does not depend on a de-cline in pulmonary vascular resistance. Withadvanced disease, the ejection time becomeslonger, causing the peak of the murmur tooccur later in systole. In severe disease, thesignificantly prolonged ejection timecauses a delay in closure of the aortic valvesuch that A2 occurs after P2—a phenome-non known as reversed splitting of S2 (seeChapter 2).

Diagnostic Studies

The chest radiograph of an infant with ASmay show an enlarged LV and a dilated as-cending aorta. The ECG often shows leftventricular hypertrophy. Echocardiographycan identify the structure of the aortic valveand the degree of left ventricular hyper-trophy. Doppler assessment can accuratelymeasure the pressure gradient across thestenotic valve and allow calculation of thevalve area. Cardiac catheterization confirmsthe pressure gradient across the valve.

Treatment

In its milder forms, AS does not need to be corrected, but endocarditis prophylaxisshould be followed (see Chapter 8). Severe

obstruction of the aortic valve during in-fancy may mandate immediate surgical ortranscatheter balloon valvuloplasty. Gen-erally, valvuloplasty in infancy is only pal-liative and future additional catheter bal-loon dilation or surgical revision is usuallyneeded.

Pulmonic Stenosis

Isolated pulmonic stenosis (Fig. 16.15) mayoccur at the level of the pulmonic valve (e.g.,from congenitally fused valve commissures),within the body of the RV (obstruction inthe RV outflow tract), or in the pulmonaryartery itself. Valve stenosis is the most com-mon form (>90% of cases). Pulmonic steno-sis is also seen in 10% of patients with otherforms of congenital heart disease.

Pathophysiology

The consequence of pulmonic stenosis is ob-struction to right ventricular systolic ejec-tion, which leads to increased right ventric-ular pressures and chamber hypertrophy.The clinical course is determined by theseverity of the obstruction. In the presenceof normal cardiac output, a peak systolictransvalvular pressure gradient >50 mm Hgis considered mild pulmonic stenosis, be-tween 50 and 80 mm Hg is moderate steno-sis, and severe stenosis is defined by a peakgradient >80 mm Hg.

Symptoms

Children with mild or moderate pulmonarystenosis are asymptomatic. The diagnosis isoften first made on discovery of a murmurduring a routine physical examination. Se-vere stenosis may cause manifestations suchas dyspnea with exertion, exercise intoler-ance, and with decompensation, symptomsof right-sided heart failure such as abdomi-nal fullness and pedal edema.


The physical findings in pulmonic stenosisdepend on the severity of the obstruction. If

Fig. 15

10090-16_CH16.qxd 8/31/06 4:57 PM Page 387

the stenosis is severe with accompanyingright ventricular hypertrophy, a prominentjugular venous a wave can be observed (seeChapter 2) and an RV heave is palpated overthe sternum. A loud, late-peaking, crescendo-decrescendo systolic ejection murmur isheard at the upper-left sternal border, oftenassociated with a palpable thrill. Widenedsplitting of the S2 with a soft P2 component iscaused by the delayed closure of the stenoticpulmonary valve.

In more moderate stenosis, a pulmonicejection sound (a high-pitched “click”) fol-lows S1 and precedes the systolic murmur. Itoccurs during the early phase of right ven-tricular contraction as the stenotic valveleaflets suddenly reach their maximum levelof ascent into the pulmonary artery, just be-fore blood ejection. Unlike other soundsand murmurs produced by the right side ofthe heart, the pulmonic ejection sound di-minishes in intensity during inspiration.This occurs because with inspiration, theaugmented right-sided filling elevates the

leaflets into the pulmonary artery prior toRV contraction, preempting the rapid tens-ing in early systole that is thought to pro-duce the sound.

Diagnostic Studies

The chest radiograph may demonstrate an en-larged right atrium and ventricle with post-stenotic pulmonary artery dilation (causedby the impact of the high-velocity jet ofblood against the wall of the pulmonaryartery). The ECG shows right ventricular hypertrophy and right axis deviation. Echo-cardiography with Doppler imaging assessesthe pulmonary valve morphology, deter-mines the presence of right ventricular hypertrophy, and accurately measures thepressure gradient across the obstruction.

Treatment

Mild pulmonic stenosis usually does notprogress or require treatment. Moderate or

388 Chapter Sixteen

Figure 16.15. Congenital pulmonary valve stenosis. A. The arrow points to the narrowed pul-monary valve. B. Schematic representation of obstructed flow through the narrowed pulmonary valve(jagged arrow). Right ventricular hypertrophy results from the chronically increased pressure load.Ao, aorta; IVC, inferior vena cava; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, rightatrium; RV, right ventricle; SVC, superior vena cava.

10090-16_CH16.qxd 8/31/06 4:57 PM Page 388


severe valvular obstruction at the valvularlevel can be relieved by dilating the stenoticvalve through transcatheter balloon valvu-loplasty. Long-term results of this procedurehave been uniformly excellent, and rightventricular hypertrophy usually regressessubsequently. Antibiotic prophylaxis for en-docarditis is required, however, even aftervalvuloplasty.

Coarctation of the Aorta

Coarctation of the aorta typically consists ofa discrete narrowing of the aortic lumen.This anomaly has an incidence of 1 in 6,000live births and often occurs in patients withTurner syndrome (45, XO). Two types ofcoarctation are distinguished according tothe location of the aortic narrowing in rela-tion to the ductus arteriosus: preductal (2%)and postductal (98%), as shown in Figure16.16. Preductal coarctation, in which nar-rowing occurs proximal to the ductus, re-sults when an intracardiac anomaly duringfetal life decreases blood flow through theleft side of the heart, leading to hypoplasticdevelopment of the aorta (Fig. 16.17). Post-

ductal coarctation is most likely the result ofmuscular ductal tissue that extends into theaorta during fetal life. When ductal tissueconstricts following birth, the ectopic tissuewithin the aorta also constricts, creating anobstruction.

Pathophysiology

In both types of coarctation, the LV faces anincreased pressure load. Blood flow to thehead and upper extremities is preserved be-cause the vessels supplying these areas usu-ally branch off the aorta proximal to the ob-struction, but flow to the descending aortaand lower extremities may be diminished. Ifcoarctation is not corrected, compensatoryalterations include (1) development of leftventricular hypertrophy and (2) dilatationof collateral blood vessels from the inter-costal arteries that bypass the coarctationand provide blood to the descending aorta.Eventually, these collateral vessels enlargeand can erode the undersurface of the ribs.

Symptoms

Patients with preductal and severe postduc-tal coarctation usually present very shortly

Ductusarteriosus

Constricted ductus

PreductalcoarctationAorta

Pulmonaryartery

A

Aorta

Pulmonaryartery

B

Postductalcoarctation

Figure 16.16. Coarctation of the aorta. A. Preductalcoarctation. B. Postductal coarctation.

Figure 16.17. Magnetic resonance imaging of coarc-tation of the aorta. This lateral view demonstrates apreductal coarctation, manifest as a focal aortic narrow-ing (white arrow). AA, ascending aorta; DA, descendingaorta; LA, left atrium; RA, right atrium; TAA, transverseaortic arch.

Fig. 16

Fig. 17

10090-16_CH16.qxd 8/31/06 4:57 PM Page 389

after birth with symptoms of heart failure.Infants with preductal coarctation may alsoexhibit differential cyanosis if the ductus arteriosus remains open. The upper half of the body, supplied by the LV and the ascending aorta, is perfused with well-oxygenated blood; however, the lower halfappears cyanotic because it is largely sup-plied by right-to-left flow of poorly oxy-genated blood from the pulmonary artery,across the ductus arteriosus, and into the de-scending aorta.

When the coarctation is less severe, as inmost postductal cases, it may be suspectedby finding upper extremity hypertension onphysical examination during childhood.


On examination, the femoral pulses areweak and delayed. An elevated blood pres-sure in the upper body is the most commonpresentation. If the coarctation occurs dis-tal to the takeoff of the left subclavianartery, the systolic pressure in the arms isgreater than that in the legs. If the coarcta-tion occurs proximal to the takeoff of the leftsubclavian artery, the systolic pressure inthe right arm may exceed that in the leftarm. A systolic pressure in the right armthat is 15 to 20 mm Hg greater than that in a leg is sufficient to suspect coarctation,because normally the systolic pressure inthe legs is higher than that in the arms. Amidsystolic ejection murmur (caused byflow through the coarctation) may be audi-ble over the chest and/or back. A prominenttortuous collateral arterial circulation maycreate continuous murmurs over the chestin adults.

Diagnostic Studies

In adults with uncorrected coarctation ofthe aorta, chest radiography generally revealsnotching of the inferior surface of the pos-terior ribs owing to enlarged intercostal vessels supplying collateral circulation tothe descending aorta. An indented aorta at the site of coarctation may also be visual-

ized. The ECG shows left ventricular hyper-trophy resulting from the pressure loadplaced on that chamber. Doppler echocardio-graphy confirms the diagnosis of coarctationand assesses the pressure gradient across thelesion. Magnetic resonance imaging demon-strates in detail the length and severity ofcoarctation (see Fig. 16.17). Diagnosticcatheterization and angiography are rarelynecessary.

Treatment

In neonates with severe obstruction, pro-staglandin infusion is administered to keepthe ductus arteriosus patent, thus maintain-ing blood flow to the descending aorta be-fore surgery is undertaken. In children, elec-tive repair is usually performed to preventsystemic hypertension. Several effective sur-gical procedures are available, including ex-cision of the narrowed aortic segment withend-to-end reanastomosis and direct repairof the coarctation, sometimes using syn-thetic patch material. For older children,adults, and patients with recurrent coarcta-tion after previous repair, transcatheter in-terventions (balloon dilatation with or with-out stent placement) is usually successful.Antibiotic prophylaxis to prevent endarteri-tis (like the prophylaxis against endocarditisdescribed in Chapter 8) is necessary evenafter repair.

Cyanotic Lesions

Tetralogy of Fallot

Tetralogy of Fallot results from a single de-velopmental defect: an abnormal anteriorand cephalad displacement of the infun-dibular (outflow tract) portion of the inter-ventricular septum. As a consequence, fouranomalies arise that characterize this condi-tion, as shown in Figure 16.18: (1) a VSDcaused by malalignment of the interventric-ular septum, (2) subvalvular pulmonicstenosis because of obstruction from the in-fundibular septum, (3) an overriding aortathat receives blood from both ventricles,

390 Chapter Sixteen

Fig. 18

10090-16_CH16.qxd 8/31/06 4:57 PM Page 390


and (4) right ventricular hypertrophy owingto the high pressure load placed on the RVby the pulmonic stenosis. Tetralogy of Fallotis the most common form of cyanotic con-genital heart disease after infancy, occurringin 5 of 10,000 live births, and is often asso-ciated with other cardiac defects, includinga right-sided aortic arch (25% of patients),ASD (10% of patients), and less often,anomalous origin of the left coronary artery.A microdeletion in chromosome 22 (22q11)has been identified in patients with a syn-drome that includes tetralogy of Fallot asone of the cardiovascular manifestations(see Box 16.1).

Pathophysiology

Increased resistance by the subvalvular pul-monic stenosis causes deoxygenated bloodreturning from the systemic veins to be di-verted from the RV, through the VSD, to theLV, and into the systemic circulation, result-ing in systemic hypoxemia and cyanosis.The magnitude of shunt flow across the VSDis primarily a function of the severity of thepulmonary stenosis, but acute changes in

Figure 16.18. Tetralogy of Fallot is characterized byfour associated anomalies. 1) A ventricular septal de-fect (hollow arrow), 2) obstruction to right ventricularoutflow (solid arrow), 3) an overriding aorta that receivesblood from both ventricles, and 4) right ventricular hypertrophy. Ao, aorta; IVC, inferior vena cava; LA, leftatrium; LV, left ventricle; PA, pulmonary artery; RA, rightatrium; RV, right ventricle; SVC, superior vena cava.

Box 16.1 Genetic Abnormalities in Congenital Heart Disease

Progress in the understanding of genetic influences on cardiac development and con-genital heart disease is proceeding at a brisk pace, aided by the Human Genome Project.Although nearly all cardiac congenital anomalies can occur as isolated findings, the clus-tering of certain forms with heritable syndromes and known genetic abnormalities pro-vides clues to the underlying basis for some defects. As with other congenital conditions,a complex interplay between genes and the embryonic environment can result in an arrayof phenotypes.

Among infants with Down syndrome (trisomy 21) the incidence of congenital heartdefects is nearly a 40%. Many of these are common abnormalities such as atrial septal de-fect, ventricular septal defect, and patent ductus arteriosus. However, there is also a highincidence of a rarer condition known as common atrioventricular canal, which consists ofa large atrial and ventricular septal defect and a common (undivided) atrioventricular valveabove the two ventricles. This portion of heart tissue is usually formed by the interactionbetween the endocardial cushions and cells of neural crest origin, which are known tohave abnormal migration patterns in patients with trisomy 21.

Turner syndrome (45, XO) is a rare heritable condition of girls and women that is alsoassociated with congenital heart disease. Left-sided obstructive congenital heart lesions

Box 1

10090-16_CH16.qxd 8/31/06 4:57 PM Page 391

392 Chapter Sixteen

are common in patients with this syndrome, including bicuspid aortic valve, coarctationof the aorta and occasionally the hypoplastic left heart syndrome (underdevelopment ofthe left ventricle and aorta). The specific genes responsible for these abnormalities havenot yet been elucidated.

In contrast, discrete gene abnormalities have been identified in other syndrome-asso-ciated forms of congenital heart disease. For example, patients with Williams syndrome(characterized by mental retardation, hypercalcemia, renovascular hypertension, facial ab-normalities, and short stature) have a high incidence of supravalvular aortic stenosis, aringlike obstruction above the aortic valve. Some patients have a more diffuse arteriopathyof the aorta as well as pulmonary artery obstruction. The genetic abnormality in Williamssyndrome is a deletion on chromosome 7 (7q11.23), a region that includes the elastingene. Abnormalities in the production of elastin, a critical component of the arterial wall,may be responsible for the observed arteriopathy.

DiGeorge syndrome (manifested by a characteristic facial appearance, pharyngeal de-fects, absent parathyroid glands with hypocalcemia, and hypoplasia of the thymus withdefective immune T-cell function) is associated with congenital abnormalities of the car-diac outflow tracts, such as tetralogy of Fallot, truncus arteriosus (a large VSD over whicha single large outflow vessel arises), and interrupted aortic arch. Most patients with Di-George syndrome have a microdeletion within chromosome 22 (22q11). Recent researchhas implicated a single gene (TBX1) in the cardiac manifestations. This gene encodes atranscription factor that appears to play a critical role in developmental patterning of thecardiac outflow tracts.

Several other transcription factors of importance in heart development likely contributeto congenital heart disease. Some families with heritable forms of atrial septal defects havemutations in the transcription factor gene Nkx2.5. An associated transcription factor gene(GATA4) appears to collaborate with Nkx2.5 and has been found to be involved in familialseptal defect syndromes. Mutations in TBX5, yet another transcription factor gene, are re-sponsible for the Holt-Oram syndrome (also known as the heart-hand syndrome), an au-tosomal dominant disorder characterized by abnormal development of the upper extrem-ities and cardiac defects, most commonly secundum ASDs and ventricular septal defects.

There are also examples of loci involved in specific cellular processes, mutations ofwhich have been implicated in congenital heart disease. One example is the gene PTPN11,which functions as a critical regulator of signal transduction pathways. This gene is mu-tated in many patients with Noonan syndrome, features of which include short stature,a dysmorphic facial appearance, chest deformities, and congenital heart defects, mostcommonly valvular and supravalvular pulmonary stenosis.

Continuing advances in knowledge about the genome will undoubtedly lead to agreater understanding of cardiac development and how molecular defects in theseprocesses lead to congenital heart abnormalities.

AQ5

systemic and pulmonary vascular resistancescan affect it as well.

Symptoms

Children with tetralogy of Fallot often ex-perience dyspnea on exertion. “Spells” mayoccur following exertion, feeding, or crying

when systemic vasodilatation results in anincreased right-to-left shunt. Manifestationsof such spells include irritability, cyanosis,hyperventilation, and occasionally syncopeor convulsions. Children learn to alleviatetheir symptoms by squatting down, whichis thought to increase systemic vascular re-sistance by “kinking” the femoral arteries,

10090-16_CH16.qxd 8/31/06 4:57 PM Page 392


thereby decreasing the right-to-left shuntand directing more blood from the RV to thelungs.


Children with tetralogy of Fallot and mod-erate pulmonary stenosis often have mildcyanosis, most notably of the lips, mucousmembranes, and digits. Infants with severepulmonary stenosis may present with pro-found cyanosis in the first few days of life.Chronic hypoxemia caused by the right-to-left shunt commonly results in clubbing ofthe fingers and toes. Right ventricular hy-pertrophy may be appreciated on physicalexamination as a palpable heave along theleft sternal border. The S2 is single, composedof a normal aortic component; the pulmo-nary component is soft and usually inaudi-ble. A systolic ejection murmur heard best atthe upper-left sternal border is created byturbulent blood flow through the stenoticright ventricular outflow tract. There is usu-ally no distinct murmur related to the VSD,because it is typically large and thus gener-ates little turbulence.

Diagnostic Studies

Chest radiography demonstrates prominenceof the RV and decreased size of the main pul-monary artery segment, giving the appear-ance of a “boot-shaped” heart. Pulmonaryvascular markings are typically diminishedbecause of decreased flow through the pul-monary circulation. The ECG shows rightventricular hypertrophy with right axis de-viation. Echocardiography details the rightventricular outflow tract anatomy, the mis-aligned VSD, right ventricular hypertrophy,and other associated defects, as does cardiaccatheterization.

Treatment

Before definitive surgical correction of tet-ralogy of Fallot was developed, several formsof palliative therapy were undertaken. Theseinvolved creating anatomic communica-tions between the aorta (or one of its major

branches) to the pulmonary artery and cre-ating a left-to-right shunt to increase pul-monary blood flow. Such procedures are oc-casionally used today in infants for whomdefinitive repair is planned at an older age.Complete surgical correction of tetralogy ofFallot involves closure of the VSD and en-largement of the subpulmonary infundibu-lum with the use of a pericardial patch. Elec-tive repair is usually recommended around1 year of age to decrease the likelihood of fu-ture complications. Most patients who haveundergone successful repair grow to becomeasymptomatic adults. However, antibioticprophylaxis to prevent endocarditis is stillrequired.

Transposition of the Great Arteries

In transposition of the great arteries (TGA)each great vessel inappropriately arises fromthe opposite ventricle; that is, the aorta orig-inates from the RV and the pulmonaryartery originates from the LV (Fig. 16.19).

Figure 16.19. Transposition of the great arteries. Theaorta (Ao) and pulmonary artery (PA) arise abnormallyfrom the right ventricle (RV) and left ventricle (LV), re-spectively. IVC, inferior vena cava; LA, left atrium; RA,right atrium; SVC, superior vena cava.

AQ1

Fig. 19

10090-16_CH16.qxd 8/31/06 4:57 PM Page 393

This anomaly accounts for approximately7% of congenital heart defects, affecting 40of 100,000 live births. Whereas tetralogy ofFallot is the most common etiology ofcyanosis after infancy, TGA is the most com-mon cause of cyanosis in the neonatal period.

The precise cause of transposition re-mains unknown. Historically, failure of theaorticopulmonary septum to spiral in anormal fashion during fetal developmentwas considered the underlying problem. Ithas been suggested that the defect may bethe result of abnormal growth and absorp-tion of the subpulmonary and subaorticinfundibuli during the division of the truncus arteriosus. Normally, reabsorptionof the subaortic infundibulum places theforming aortic valve posterior and inferiorto the pulmonary valve and in continuitywith the LV. In TGA, the process of in-fundibular reabsorption may be reversed,placing the pulmonary valve over the LVinstead.

Pathophysiology

TGA separates the pulmonary and systemiccirculations by placing the two circuits inparallel rather than in series. This arrange-ment forces desaturated blood from the sys-temic venous system to pass through the RV and then return to the systemic circula-tion through the aorta without undergoingnormal oxygenation in the lungs. Similarly,oxygenated pulmonary venous return passesthrough the LV and then back through thepulmonary artery to the lungs without im-parting oxygen to the systemic circulation.The result is an extremely hypoxic, cyanoticneonate. Without intervention to createmixing between the two circulations, TGA isa lethal condition.

TGA is compatible with life in utero be-cause flow through the ductus arteriosusand foramen ovale allows communicationbetween the two circulations. Oxygenatedfetal blood flows from the placenta throughthe umbilical vein to the right atrium, andthen most of it travels into the left atriumthrough the foramen ovale. The oxygenatedblood in the left atrium passes into the LV

and is pumped out the pulmonary artery.Most of the pulmonary artery flow travelsthrough the ductus arteriosus into the aorta,instead of the high-resistance pulmonaryvessels, and then oxygen is provided to thedeveloping tissues.

After birth, normal physiologic closure ofthe ductus and the foramen ovale elimi-nates the shunt between the parallel circu-lations and, without intervention, would re-sult in death because oxygenated blood doesnot reach the systemic tissues. However, ifthe ductus arteriosus and foramen ovale re-main patent (either naturally or with exoge-nous prostaglandins or surgical interven-tion), communication between the parallelcircuits is maintained, and sufficiently oxy-genated blood may be provided to the brainand other vital organs.

Symptoms and Physical Examination

Infants with transposition appear blue, withthe intensity of the cyanosis dependent onthe degree of intermixing between the par-allel circuits. In most cases, generalizedcyanosis is apparent on the first day of lifeand progresses rapidly as the ductus arte-riosus closes. Palpation of the chest revealsa right ventricular impulse at the lowersternal border as the RV faces systemicpressures. Auscultation may reveal an ac-centuated S2, which reflects closure of theanteriorly placed aortic valve just underthe chest wall. Prominent murmurs are un-common and may signal an additional defect.

Diagnostic Studies

Chest radiography is usually normal, although thebase of the heart may be narrow owing tothe more anterior-posterior orientation ofthe aorta and pulmonary artery. The ECGdemonstrates right ventricular hypertrophy,reflecting the fact that the RV is the systemic“high-pressure” pumping chamber. The de-finitive diagnosis of transposition can bemade by echocardiography, which demon-strates the abnormal orientation of the greatvessels.

394 Chapter Sixteen

10090-16_CH16.qxd 8/31/06 4:57 PM Page 394


Treatment

TGA is a medical emergency. Initial treat-ment includes maintenance of the ductus ar-teriosus by prostaglandin infusion and cre-ation of an interatrial communication usinga balloon catheter (the Rashkind procedure).These procedures allow adequate mixing ofthe two circulations until definitive correc-tive surgery can be performed. The currentcorrective procedure of choice is the “arterialswitch” operation (Jatene procedure), whichinvolves transection of the great vesselsabove the semilunar valves and origin of thecoronary arteries. The great vessels are thenswitched to the natural configuration, so theaorta arises from the LV and the pulmonaryartery arises from the RV. The coronary ar-teries are then relocated to the new aorta.

EISENMENGER SYNDROME

Eisenmenger syndrome is the condition ofsevere pulmonary vascular obstruction thatresults from chronic left-to-right shuntingthrough a congenital cardiac defect. The el-evated pulmonary vascular resistance causesreversal of the original shunt (to the right-to-left direction) and systemic cyanosis.

The mechanism by which increased pul-monary flow causes this condition is un-known. Histologically, the pulmonary arte-riolar media hypertrophies and the intimaproliferates, reducing the cross-sectionalarea of the pulmonary vascular bed. Overtime, the vessels become thrombosed, andthe resistance of the pulmonary vasculaturerises, causing the original left-to-right shuntto decrease. Eventually, if the resistance ofthe pulmonary circulation exceeds that of thesystemic vasculature, the direction of shuntflow reverses.

With reversal of the shunt to the right-to-left direction, symptoms arise from hy-poxemia, including exertional dyspnea and fatigue. Reduced hemoglobin saturation stim-ulates the bone marrow to produce more redblood cells (erythrocytosis), which can lead tohyperviscosity, symptoms of which includefatigue, headaches, and stroke (caused bycerebrovascular occlusion). Infarction or rup-

ture of the pulmonary vessels can result in hemoptysis.

On examination, a patient with Eisen-menger syndrome appears cyanotic withdigital clubbing. A prominent a wave in thejugular venous pulsation represents elevatedright-sided pressure during atrial contrac-tion. A loud P2 is common. The murmur ofthe inciting left-to-right shunt is usually ab-sent, because the original pressure gradientacross the lesion is negated by the elevatedright-heart pressures.

Chest radiography in Eisenmenger syn-drome is notable for proximal pulmonaryartery dilatation with peripheral tapering.Calcification of the pulmonary vasculaturemay be seen. The ECG demonstrates rightventricular hypertrophy and right atrial en-largement. Echocardiography with Dopplerstudies can usually identify the underlyingcardiac defect and quantitate the pulmo-nary artery systolic pressure.

Treatment includes the avoidance of ac-tivities that can exacerbate the right-to-leftshunt. These include strenuous physical ac-tivity, high altitude, and the use of periph-eral vasodilator drugs. Pregnancy is espe-cially dangerous; the rate of spontaneousabortion is 20% to 40% and the incidence ofmaternal mortality is 45%.

No medical therapy offers a reliably effec-tive way to reduce the elevated pulmonaryvascular resistance. Supportive measures include endocarditis prophylaxis, manage-ment of rhythm disturbances, and phlebo-tomy for patients with symptomatic eryth-rocytosis. The only effective long-termstrategy for severely affected patients is lungor heart-lung transplantation. Fortunately,with the dramatic advances that have beenmade in the detection and early correctionof severe congenital heart defects, Eisen-menger syndrome has become less common.

SUMMARY

1. The significance of congenital heart le-sions can be predicted from an under-standing of cardiovascular embryonic de-velopment and the transition topostnatal circulatory pathways.

10090-16_CH16.qxd 8/31/06 4:57 PM Page 395

2. Cardiac malformations occur in 0.8% oflive births. Such lesions can be groupedinto cyanotic or acyanotic defects, de-pending on whether the abnormality re-sults in pulmonary-to-systemic (right-to-left) shunting of blood.

3. Acyanotic defects often result in eithervolume overload (ASD, VSD, PDA) or pres-sure overload (AS, pulmonic stenosis,coarctation of the aorta). Chronic volumeoverload resulting from a large left-to-right shunt can ultimately result in in-creased pulmonary vascular resistance, re-versal of the direction of shunt flow, andsubsequent cyanosis (Eisenmenger syn-drome).

4. Among the most common cyanotic de-fects are tetralogy of Fallot and TGA.

Acknowledgment

Contributors to the previous editions of this chapterwere Yi-Bin Chen, MD; Douglas W. Green, MD; Lakshmi Halasyamani, MD; Andrew Karson, MD;Raymond Tabibiazar, MD; Michael D. Freed, MD.;and Richard Liberthson, MD.

Additional Reading

Allen HD, Gutgesell HP, Clark EB, et al., eds. Mossand Adams’ Heart Disease in Infants, Children,and Adolescents. 6th Ed. Baltimore: LippincottWilliams & Wilkins, 2001.

Brickner ME, Hillis LD, Lange RA. Congenital heartdisease in adults, part I of II. N Engl J Med 2000;342(4):256–263.

Brickner ME, Hillis LD, Lange RA. Congenital heartdisease in adults, part II of II. N Engl J Med 2000;342(5):334–342.

Gatzoulis MA, Swan L, Therrien J. Adult CongenitalHeart Disease: A Practical Guide. Malden, MA: BMJBooks, 2005.

Gelb BD. Genetic basis of congenital heart disease.Curr Opin Cardiol 2004;19:110–115.

Goldmuntz E. The genetic contribution to congeni-tal heart disease. Pediatr Clin North Am 2004;51:1721–1737.

Kearns-Jonker M. Congenital Heart Disease: Molecu-lar Diagnostics. Totowa, NJ: Humana Press, 2006.

Moodie DS. Diagnosis and management of congen-ital heart disease in the adult. Cardiol Rev 2001;9:276–281.

Moore KL, Persaud TVN. The Developing Human:Clinically Oriented Embryology. 7th Ed. Philadel-phia: WB Saunders, 2003.

Park MK, Troxler RG. Pediatric Cardiology for Prac-titioners. 4th Ed. St. Louis: Mosby, 2002.

Perloff JK. The Clinical Recognition of CongenitalHeart Disease. 5th Ed. Philadelphia: WB Saunders,2003.

Rudolph AM. Congenital Diseases of the Heart: Clinical-Physiological Considerations. 2nd Ed.New York: Futura, 2001.

Sadler TW. Langman’s Medical Embryology. 9th Ed.Philadelphia: Lippincott Williams & Wilkins,2004.

Tworetzky W, Marshall AC. Fetal interventions forcardiac defects. Pediatr Clin North Am 2004;51:1503–1513.

396 Chapter Sixteen

10090-16_CH16.qxd 8/31/06 4:57 PM Page 396

Chapter 16—Author Queries1. AU: Should this be “or” instead?2. Additional Readings section cites a 2001 edition (6th). Should this citation (and possiblythe figure) be updated?3. Additional Readings section cites a 2003 edition (7th). Should this citation (and possiblythe figure) be updated? See also Figures 16.7–10.4. AU: The figure shows A, B, and C panels. Should this caption be so labeled?

Chapter 16—Author Query (Box)5. AU: Edit correct?

10090-16_CH16.qxd 8/31/06 4:57 PM Page a

397

INOTROPIC DRUGS AND VASOPRESSORSDigitalis GlycosidesSympathomimetic AminesPhosphodiesterase InhibitorsVasopressin

VASODILATOR DRUGSAngiotensin-Converting Enzyme InhibitorsAngiotensin II Type 1 Receptor AntagonistsDirect-Acting VasodilatorsCalcium Channel BlockersOrganic NitratesNatriuretic PeptidesSildenafil

ANTIADRENERGIC DRUGSCentral Adrenergic InhibitorsSympathetic Nerve-Ending AntagonistsPeripheral α-Adrenergic Receptor Antagonistsβ-Adrenergic Receptor Antagonists

ANTIARRHYTHMIC DRUGSClass IA Antiarrhythmics

Class IB AntiarrhythmicsClass IC AntiarrhythmicsClass II AntiarrhythmicsClass III AntiarrhythmicsClass IV AntiarrhythmicsAdenosine

DIURETICSLoop DiureticsThiazide DiureticsPotassium-Sparing Diuretics

ANTITHROMBOTIC DRUGSPlatelet InhibitorsAnticoagulant Drugs

LIPID-REGULATING DRUGSHMG CoA Reductase InhibitorsBile Acid–Binding AgentsCholesterol Absorption InhibitorsNiacinFibrates

C H A P T E R

17Cardiovascular DrugsMartin W. SchoenElliott M. AntmanGary R. StrichartzLeonard S. Lilly

This chapter reviews the physiologic basisand clinical use of cardiovascular drugs. Al-though a multitude of drugs are available totreat cardiac disorders, these agents can for-tunately be grouped by their pharmacologicactions into a small number of categories.Additionally, many drugs are useful in morethan one form of heart disease.

INOTROPIC DRUGS AND VASOPRESSORS

Inotropic drugs are used to increase the forceof ventricular contraction when myocardialsystolic function is impaired. The pharmaco-logic agents in this category include the car-diac glycosides, sympathomimetic amines,

10090-17_CH17.qxd 8/31/06 4:58 PM Page 397

and phosphodiesterase inhibitors. Althoughthey work through different mechanisms,they are all thought to improve cardiac con-traction by increasing the intracellular cal-cium concentration, thus augmenting actinand myosin interactions. The hemodynamiceffect is to shift a depressed ventricular per-formance curve (Frank-Starling curve) in anupward direction (Fig. 17.1), so that for agiven ventricular filling pressure, stroke vol-ume and cardiac output are increased.

Digitalis Glycosides

The cardiac glycosides are often called “dig-italis” because commonly used drugs of thisclass are based on extracts of the foxgloveplant, Digitalis purpurea. In this discussion,the term digitalis is used to describe the en-tire group of cardiac glycosides, includingdigoxin, digitoxin, and ouabain.

Mechanism of Action

The two desired effects of digitalis are (1) toimprove contractility of the failing heart (me-

chanical effect) and 2) to prolong the refrac-tory period of the atrioventricular (AV) nodein patients with supraventricular arrhyth-mias (electrical effect).

Mechanical Effect

The action by which digitalis improvescontractility appears to be inhibition of thesarcolemmal Na+K+-ATPase “pump,” nor-mally responsible for maintaining trans-membrane Na+ and K+ gradients. By bind-ing to and inhibiting this pump, digitaliscauses the intracellular [Na+] to rise. Asshown in Figure 17.2, an increase in intra-cellular sodium content reduces Ca++ extru-sion from the cell by the Na+-Ca++ exchanger.Consequently, more Ca++ is pumped intothe sarcoplasmic reticulum, and when subsequent action potentials excite thecell, a greater-than-normal amount of Ca++

is released to the myofilaments, thereby en-hancing the force of contraction. The mag-nitude of the positive inotropic effect cor-relates with the degree of Na+K+-ATPaseinhibition.

398 Chapter Seventeen

Left ventricular end-diastolic pressure


Normal

Heart failure

Inotropictherapy

Diuretictherapy

Car

diac

Out

put

Hyp

oten

sion

Figure 17.1. Ventricular performance (Frank-Starling) curve. In heart failure,the curve is displaced downward, so that at a given left ventricular end-diastolicpressure (LVEDP), the cardiac output is lower than in a normal heart. Diuretics re-duce LVEDP but do not change the position of the curve; thus, pulmonary con-gestion improves but cardiac output may fall. Inotropic drugs displace the curveupward, toward normal, so that at any LVEDP, the cardiac output is higher.

AQ1Fig. 2

Fig. 1

10090-17_CH17.qxd 8/31/06 4:58 PM Page 398

Cardiovascular Drugs 399

ple, in atrial and ventricular Purkinje fibers,a high digitalis concentration has three im-portant actions that may lead to dangerousarrhythmias (Fig. 17.3):

1. Less negative resting potential. Inhibition ofthe Na+ K+-ATPase causes the resting po-tential to become less negative. Since theNa+ K+-ATPase normally removes threeNa+ ions from the cell in exchange for twoinwardly moving K+ ions; inhibition ofthe pump results in a decrease of this out-ward current and a resulting depolariza-tion of the cell. Consequently, there is avoltage-dependent partial inactivation ofthe fast Na+ channels, which leads to aslower rise of phase 0 depolarization andreduction in conduction velocity (see Fig. 1.17). The slowed conduction, if pre-sent heterogeneously among neighboringcells, enhances the possibility of reentrantarrhythmias.

2. Decreased action potential duration. At highdigitalis concentrations, the cardiac actionpotential shortens. This relates in part tothe digitalis-induced elevated intracellu-lar [Ca++], which increases the activity ofa Ca++-dependent K+ channel. The open-ing of this channel promotes K+ effluxand more rapid repolarization. In addition,high intracellular [Ca++] inactivates theCa++ channels, decreasing the inward de-polarizing Ca++ current. The decrease inaction potential duration and the associ-ated shortened refractory period increasethe time during which cardiac fibers areresponsive to external stimulation, allow-ing greater opportunity for propagationof arrhythmic impulses.

3. Enhanced automaticity. Digitalis enhancescellular automaticity and may generateectopic rhythms by two mechanisms:a. The less negative membrane resting

potential may induce phase 4 gradualdepolarization, even in nonpacemakercells (see Chapter 11), and an actionpotential is triggered each time thethreshold voltage is reached.

b. The digitalis-induced increase in intra-cellular [Ca++] may trigger delayed after-depolarizations (see Fig. 17.3). If an

DIGITALIS

Figure 17.2. Mechanism of action of digitalis (in-otropic effect). A. Digitalis inhibits the sarcolemmalNa+K+-ATPase, causing intracellular [Na+] to rise. B. In-creased cytosolic [Na+] reduces the transmembrane Na+

gradient; thus, the Na+-Ca+ exchanger drives less Ca++ outof the cell. C. The increased [Ca++] is stored in the sar-coplasmic reticulum, such that with subsequent action potentials, greater-than-normal Ca++ is released to thecontractile elements in the cytoplasm, intensifying theforce of contraction.

Electrical Effect

The major therapeutic electrical effect of dig-italis occurs at the AV node, where it slowsconduction velocity and increases refractori-ness (Table 17.1). Digitalis affects the elec-trical properties of cardiac tissue directly,but more importantly, it modifies autonomicnervous system output by enhancing vagaltone and reducing sympathetic activity. Asa result, digitalis decreases the frequency oftransmission of atrial impulses through theAV node to the ventricles. This is beneficialin reducing the rate of ventricular stimula-tion in patients with rapid supraventriculartachycardias such as atrial fibrillation or atrialflutter. In addition, by enhancing the re-fractoriness of the AV node, digitalis mayconvert supraventricular reentrant arrhyth-mias to normal rhythm.

However, if digitalis concentrations riseinto the toxic range, further enhancementof vagal tone and more extreme inhibitionof the Na+ K+-ATPase pump can result in ad-verse electrophysiologic effects. For exam-

Tab. 1

Fig. 3

10090-17_CH17.qxd 8/31/06 4:58 PM Page 399


TABLE 17.1. Electrophysiologic Effects of Digitalis

Region Mechanism of Action Effect

Therapeutic effectsAV node

Toxic effectsSinoatrial node

Atrium

AV node

AV junction (between AV node and His bundle)

Purkinje fibers and ventricular muscle

AV, atrioventricular; PSVT, paroxysmal supraventricular tachycardia.

Vagal effect

↓ Conduction velocity

↑ Effective refractory period

↑ Vagal and direct suppression

Delayed afterdepolarizations (triggeredactivity), ↑ slope of phase 4 depolar-ization (↑ automaticity)

Variable effects on conduction velocityand ↑ refractory period (can fragmentconduction and lead to reentry)

Direct and vagal-mediated conductionblock

Delayed afterdepolarizations (triggeredactivity), ↑ slope of phase 4 depolar-ization (↑ automaticity)

Delayed afterdepolarizations (triggeredactivity), ↓ conduction velocity and ↑refractory period (can lead to reentry)

↑ slope of phase 4 depolarization (↑ automaticity)

• ↓ Rate of transmission of atrial im-pulses to the ventricles in supraven-tricular tachyarrhythmias

• ↓ Conduction velocity and ↑ refrac-tory period may interrupt reentrantcircuits passing through the AV node

• Sinus bradycardia• Sinoatrial block (impulse not trans-

mitted from SA node to atrium)• Atrial premature beats

• Nonreentrant SVT (ectopic rhythm)

• Reentrant PSVT• AV block (first, second or third degree)

• Accelerated junctional rhythm

• Ventricular premature beats

• Ventricular tachycardia

A

B

C

Figure 17.3. Direct effects of digitalis on the Purkinjecell action potential. The solid tracing represents depo-larization and repolarization of a normal cell; the dashedtracing demonstrates the effects of digitalis. A. The max-imum diastolic potential is less negative, and there is anincrease in the slope of phase 4 depolarization, endowingthe cell with intrinsic automaticity, and the potential forectopic rhythms. B. Because depolarization of the cell oc-curs at a more positive voltage, the rate of rise of phase 0is decreased, and conduction velocity is slowed, which, ifpresent heterogeneously among neighboring cells, canproduce conditions for reentry. C. Delayed afterdepolar-izations may develop at high concentrations of digitalis inassociation with an increased intracellular calcium con-centration and can result in triggered tachyarrhythmias.

afterdepolarization reaches the thresh-old voltage, an action potential (ecto-pic beat) is generated. Ectopic beatsmay lead to additional afterdepolar-izations and self-sustaining arrhyth-mias such as ventricular tachycardia.

Thus, digitalis in toxic concentrations maylead to several types of ectopic or reentrantrhythms (see Table 17.1). In addition, theaugmented direct and indirect vagal effectsof toxic doses of digitalis slow conductionthrough the AV node, such that high de-grees of AV block, including complete heartblock, can occur.

Clinical Uses

The most common use of digitalis is as an in-otropic agent to treat heart failure caused by

10090-17_CH17.qxd 8/31/06 4:58 PM Page 400


decreased ventricular contractility (see Chap-ter 9). Digitalis increases the force of contrac-tion and augments cardiac output, therebyimproving left ventricular emptying, reduc-ing left ventricular size, and decreasing the el-evated ventricular filling pressures typical ofpatients with systolic dysfunction. Digitalis isnot beneficial in forms of heart failure associ-ated with normal ventricular contractility(e.g., high-output failure associated with thy-rotoxicosis, pulmonary congestion caused bymitral stenosis, or in the setting of pure di-astolic dysfunction).

Once the mainstay of therapy in con-gestive heart failure (CHF), the use of digi-talis has waned in the face of newer, moreeffective therapies (as discussed later in thechapter; see also Chapter 9). Nonetheless,digitalis continues to be useful in treatingpatients with CHF complicated by atrialfibrillation (it has the added benefit of slow-ing the ventricular heart rate), or whensymptoms do not respond adequately toangiotensin-converting enzyme (ACE) in-hibitors, β-blockers, and diuretics. UnlikeACE inhibitors and β-blockers, digitalis doesnot prolong the life expectancy of patientswith chronic heart failure.

The second-most common use of digitalisis as an antiarrhythmic agent in the treatmentof atrial fibrillation, atrial flutter, and parox-ysmal supraventricular tachycardia (PSVT).In atrial fibrillation and flutter, digitalis re-duces the number of impulses transmittedacross the AV node, thereby slowing theventricular rate. Digitalis may terminate re-entrant supraventricular tachycardias, likelythrough enhancement of vagal tone, whichslows impulse conduction, prolongs the ef-fective refractory period, and can thereforeinterrupt reentrant circuits that pass throughthe AV node.

The use of digitalis as an antiarrhythmichas also become less frequent in recent yearsbecause other agents such as β-blockers, cal-cium channel blockers, and amiodarone areoften more effective. Nonetheless, for thetreatment of supraventricular tachyarrhyth-mias in the presence of CHF, digitalis remainsan important option.

Pharmacokinetics and Toxicity

The most commonly used form of digitalis isdigoxin, which is excreted unchanged by thekidney. A series of loading doses of digoxin isnecessary to raise the drug’s concentrationinto the therapeutic range. If a loading doseis not given, the steady-state concentrationis established in approximately 7 days. Themaintenance dosage depends on the patient’sability to excrete the drug (i.e., renal function).

The potential for digitalis toxicity is sig-nificant because of a low toxic-to-therapeuticdrug concentration ratio. Although manyside effects are minor, life-threateningarrhythmias may result. Extracardiac signsof acute digitalis toxicity are often gastro-intestinal (e.g., nausea, vomiting, anorexia),thought to be mediated by the action ofdigoxin on the area postrema of the brainstem. Cardiac toxicity includes a host ofarrhythmias (see Table 17.1) that may pre-cede extracardiac warning symptoms. Themost frequently encountered rhythm dis-turbance is the development of ventricularextrasystoles. In addition, various degrees ofAV block may occur because of the directand vagal effects on AV nodal conduction.Digitalis toxicity is the most common causeof nonreentrant types of supraventriculartachycardia (i.e., caused by enhanced auto-maticity or delayed afterdepolarizations).

Many factors contribute to digitalis intox-ication, the most common of which is hypo-kalemia, often caused by the concurrentadministration of diuretics. Hypokalemiaexacerbates digitalis toxicity because it fur-ther inhibits the Na+ K+-ATPase pump. Otherconditions that promote digitalis toxicity in-clude hypomagnesemia and hypercalcemia.In addition, the concurrent administrationof other drugs (e.g., quinidine) may raisethe serum digoxin concentration by decreas-ing its excretion and reducing its volume ofdistribution.

The treatment of digitalis-induced tachy-arrhythmias includes administration of po-tassium (if hypokalemia is present) and oftenintravenous lidocaine (discussed later in thechapter). High-grade AV block may requiretemporary pacemaker therapy. In patients

10090-17_CH17.qxd 8/31/06 4:58 PM Page 401

with severe intoxication, administration ofFab fragments of antidigitalis antibodies maybe life saving.

Sympathomimetic Amines

Sympathomimeticaminesare inotropic drugsthat bind to cardiac β1-receptors. Stimulationof these receptors increases the activity ofadenylate cyclase, causing increased forma-tion of cyclic adenosine monophosphate(cAMP; Fig. 17.4). Increased cAMP activatesprotein kinases, which promote intracellularcalcium influx by phosphorylating the slowL-type calcium channels. The increased cal-cium entry triggers a corresponding rise inCa++ release from the sarcoplasmic reticulum,which enhances the force of contraction.Intravenous dopamine and dobutamine arecommonly used sympathomimetic amines inthe treatment of acute heart failure. Norepi-nephrine, epinephrine, and isoproterenol areused in special circumstances, as described inthe following paragraphs. Table 17.2 sum-marizes the receptor actions and major hemo-dynamic effects of these agents.

Dopamine is an endogenous catechol-amine and the precursor of norepinephrine.It possesses an unusual combination of ac-tions that make it attractive in the treat-ment of heart failure associated with hy-potension and poor renal perfusion. Thereare various types of receptors with variousaffinities for dopamine. At low dosages,<2 µg/kg per minute, dopamine interactsprimarily with dopaminergic receptors dis-tributed in the renal and mesenteric vascu-lar beds. Stimulation of these receptors causeslocal vasodilation and increases renal bloodflow and glomerular filtration, facilitatingdiuresis.

Medium dosages of dopamine, 2 to 10 µg/kgper minute, increase inotropy by stimulationof cardiac β1-receptors directly and indi-rectly by promoting release of norepineph-rine from sympathetic nerve terminals. Thisaction increases heart rate, cardiac contrac-tility, and stroke volume, all of which aug-ment cardiac output.

At high dosages, >10 µg/kg per minute, do-pamine also stimulates systemic α-receptors,thereby causing vasoconstriction and elevat-


Adenylatecyclase

Inactiveproteinkinases

Activeproteinkinases

Myocyte

GS protein

ATPcAMP

AMP

PD

Figure 17.4. Mechanism by which b-adrenergic stimulation increases intra-cellular Ca11. β1-Receptor stimulation acts through G proteins (guanine nucleotideregulatory proteins) to activate adenylate cyclase. The latter increases cyclic adenosinemonophosphate (cAMP) production, which mediates protein kinase phosphorylationof cellular proteins, including ion channels. Phosphorylation of the slow Ca++ channelincreases calcium influx. cAMP is degraded by phosphodiesterase (PD).

Fig. 4

Tab. 2

10090-17_CH17.qxd 8/31/06 4:58 PM Page 402


ing systemic resistance. High-dose dopamineis indicated in hypotensive states such asshock. However, these doses are inappro-priate in most patients with cardiac failurebecause the peripheral vasoconstriction in-creases the resistance against which the heartmust contract (i.e., higher afterload), furtherimpairing left ventricular output.

The major toxicity of dopamine arises inpatients who are treated with high-dose ther-apy. The most important side effects areacceleration of the heart rate (which increasesoxygen consumption) and stimulation oftachyarrhythmias.

Dobutamine is a synthetic analog of dopa-mine that stimulates β1-, β2-, and α-receptors.It increases cardiac contractility by virtue ofthe β1 effect but does not increase peripheralresistance because of the balance betweenα-mediated vasoconstriction and β2-mediatedvasodilation. Thus, it is useful in the treat-ment of heart failure not accompanied byhypotension. Unlike dopamine, dobutaminedoes not stimulate dopaminergic receptors(i.e., no renal vasodilating effect), nor does itfacilitate the release of norepinephrine fromperipheral nerve endings. Like dopamine, itis useful for short-term therapy (<1 week),after which time it loses its efficacy, presum-ably because of downregulation of adrenergicreceptors. The major adverse effect is theprovocation of tachyarrhythmias.

Norepinephrine is an endogenous cate-cholamine synthesized from dopamine inadrenergic postganglionic nerves and inadrenal medullary cells (where it is both afinal product and the precursor of epineph-rine). Through its β1 activity, norepineph-

rine has positive inotropic and chronotropiceffects. Acting at peripheral α-receptors, it isalso a potent vasoconstrictor. The increasein total peripheral resistance causes the meanarterial blood pressure to rise.

With this combination of effects, norepi-nephrine is useful in patients suffering from“warm shock,” in which the combination ofcardiac contractile dysfunction and periph-eral vasodilatation lower blood pressure.However, the intense vasoconstriction elic-ited by this drug makes it less attractive thanothers in treating most other cases of shock.Norepinephrine’s side effects include pre-cipitation of myocardial ischemia (becauseof the augmented afterload and force of con-traction) and tachyarrhythmias.

Epinephrine, the predominant endoge-nous catecholamine produced in the adrenalmedulla, is formed by the decarboxylationof norepinephrine. As indicated in Table 17.2,epinephrine is an agonist of α-, β1-, and β2-receptors. Administered as an intravenousinfusion at low dosages (<0.01 µg/kg perminute), its stimulation of the β1-receptor in-creases ventricular contractility and speedsimpulse generation. As a result, stroke vol-ume, heart rate, and cardiac output increase.However, at this dosage range, β2-mediatedvasodilation may reduce total peripheral re-sistance and blood pressure.

At higher dosages, epinephrine is a po-tent vasopressor because α-mediated con-striction dominates over β2-mediated vaso-dilation. In this case, the effects of positiveinotropy, positive chronotropy, and vaso-constriction act together to raise the arte-rial blood pressure.

TABLE 17.2. Sympathomimetic Drug Effects

Receptor Stimulation

Drug D1 (↑ renal perfusion) a (vasoconstriction) b1 (↑contractility) b2 (vasodilatation)

Dopamine + ++++ ++++(low dose) (high dose) (mid or high dose) (mid dose)

Dobutamine 0 + ++++ +Norepinephrine 0 ++++ ++++ 0Epinephrine 0 ++++ ++++Isoproterenol 0 0 ++++++++

10090-17_CH17.qxd 8/31/06 4:58 PM Page 403

Epinephrine is therefore used most oftenwhen the combination of inotropic andchronotropic stimulation is desired, such asin the setting of cardiac arrest. The α-associ-ated vasoconstriction may also help supportblood pressure in that setting. The mostcommon toxic effect is the precipitation of tachyarrhythmias. Epinephrine shouldbe avoided in patients receiving β-blockertherapy, because unopposed α-mediatedvasoconstriction could produce acute severehypertension.

Isoproterenol is a synthetic epinephrineanalog. Unlike norepinephrine and epineph-rine, it is a “pure” β-agonist, having activityalmost exclusively at β1- and β2-receptors,with almost no α-receptor effect. In theheart, isoproterenol has positive inotropicand chronotropic effects, thereby increasingcardiac output. In peripheral vessels, stimu-lation of β2-receptors results in vasodilationand reduced peripheral resistance, whichmay cause blood pressure to fall.

Isoproterenol is sometimes used in emer-gency circumstances to increase the heartrate in patients with bradycardia or heartblock (e.g., as a temporizing measure beforepacemaker implantation). It may also beuseful in patients with systolic dysfunctionand slow heart rates with high systemic vas-cular resistance (a situation sometimes en-countered after cardiac surgery in patientswho had previously been receiving β-blockertherapy). Isoproterenol should be avoided inpatients with myocardial ischemia, in whomthe increased heart rate and inotropic stim-ulation would further increase myocardialoxygen consumption.

Phosphodiesterase Inhibitors

Amrinone and milrinone are nondigitalis,noncatecholamine inotropic agents. Theyexert their positive inotropic actions by in-hibiting phosphodiesterase in cardiac myo-cytes (see Fig. 17.4). This inhibition reducesthe breakdown of intracellular cAMP, the ul-timate result of which is enhanced Ca++ entryinto the cell and increased force of contrac-tion. These agents also have vasodilatingproperties.

Amrinone and milrinone are used in thetreatment of acute heart failure only if therehas been insufficient improvement with con-ventional vasodilators, digitalis, and diuret-ics. This is because of the high incidence ofadverse effects, including serious ventriculararrhythmias. Amrinone has not been shownto improve the clinical state with chronic usein heart failure patients, and chronic milri-none therapy has actually demonstrated anincrease in mortality rates. Roles for theseagents are therefore limited to short-termtherapy in hospitalized patients.

Table 17.3 summarizes the actions andtoxicities of commonly used inotropic drugs.

Vasopressin

Vasopressin, the endogenous antidiuretichormone secreted by the posterior pituitary,primarily functions to maintain water bal-ance (see Chapter 9). It acts as a potent non-adrenergic vasoconstrictor when adminis-tered intravenously at higher-than-naturaldoses by directly stimulating vascular smoothmuscle V1 receptors. It has proved useful formaintaining blood pressure in patients withtypes of vasodilatory shock, such as septicshock. It may also be beneficial during cardiacarrest resuscitation because it increases coro-nary perfusion pressure, augments blood flowto vital organs, and improves the likelihoodof successful resuscitation in patients withventricular fibrillation.

VASODILATOR DRUGS

Vasodilator drugs play a central role in thetreatment of heart failure and hypertension.As described in Chapter 9, the fall in car-diac output in heart failure triggers impor-tant compensatory pathways, including theadrenergic nervous system and the renin-angiotensin-aldosterone system (see Fig. 9.9).As a result of activating these pathways, two potent vasoconstrictors are released intothe circulation: norepinephrine and angio-tensin II. These hormones bind to receptorsin arterioles and veins, where they causevasoconstriction. Initially, vasoconstrictionis beneficial in heart failure because it maxi-


AQ2

Tab. 3

10090-17_CH17.qxd 8/31/06 4:58 PM Page 404


mizes left ventricular preload (through ve-nous constriction) and maintains systemicblood pressure (by arterial constriction).

However, venous constriction may ulti-mately cause excessive venous return to theheart, with a rise in the pulmonary capil-lary hydrostatic pressure and developmentof pulmonary congestion. In addition, ex-cessive arteriolar constriction increases theresistance against which the left ventriclemust contract and therefore ultimately im-pedes forward cardiac output. Vasodilatortherapy is directed at modulating the ex-cessive constriction of veins and arterioles,thus reducing pulmonary congestion andaugmenting forward cardiac output (seeFig. 9.10).

Vasodilators are also useful antihyperten-sive drugs. Recall from Chapter 13 thatblood pressure is the product of cardiac out-put and total peripheral resistance (BP = CO× TPR). Vasodilator drugs decrease arteriolarresistance and therefore lower elevated bloodpressure.

Individual vasodilator drug classes act atspecific vascular sites (Fig. 17.5). Nitrates, forexample, are primarily venodilators, whereashydralazine is a pure arteriolar dilator. Somedrugs, such as the ACE inhibitors, α-blockers,sodium nitroprusside, and nesiritide are bal-anced vasodilators that act on both sides ofthe circulation.

Angiotensin-Converting Enzyme Inhibitors

The renin-angiotensin system plays a criticalrole in cardiovascular homeostasis. The majoreffector of this pathway (Fig. 17.6) is angio-tensin II (AII), which is formed by the cleav-age of angiotensin I by ACE. All the actions ofAII known to affect blood pressure control aremediated by its binding to angiotensin II re-ceptors of the AT1 subtype (see Fig. 13.6).Interaction with this receptor generates aseries of intracellular reactions that cause,among other effects, vasoconstriction andthe adrenal release of aldosterone, which

TABLE 17.3. Commonly Used Inotropic Drugs

Drug Mechanism of Action Major Adverse Effects

Cardiac glycosidesDigoxin

Sympathomimetic amines

Dopamine

DobutaminePhosphodiesterase

inhibitors

AmrinoneMilrinone

AV, atrioventricular; cAMP, cyclic adenosine monophosphate; D1, dopamine 1.

Inhibition of sarcolemmal Na+K+-ATPase

Enhanced vagal tone

Low dosage (<2 µg/kg per minute): D1

receptor stimulation results in mes-enteric and renal arterial dilatation(facilitates diuresis)

Medium dosage (2–10 µg/kg per minute):β1-receptor stimulation and release ofnorepinephrine from sympatheticnerve terminals (inotropic effect)

High dosage (>10 µg/kg per minute): α-receptor stimulation (peripheralvasoconstriction)

β1-, β2-, and α-receptor stimulationIncreased intracellular cAMP due to inhi-

bition of its breakdown by phospho-diesterase

Gastrointestinal: nausea, vomitingCardiac: atrial, nodal, and ventricular

tachyarrhythmias; high-degree AVblock

Tachycardia, arrhythmias, hypertension,drug tolerance

Tachyarrhythmias, drug toleranceGastrointestinal: nausea, vomiting

Cardiac: arrhythmias(Amrinone only): thrombocytopenia

Fig. 5

Fig. 6

10090-17_CH17.qxd 8/31/06 4:58 PM Page 405

promotes Na+ reabsorption from the distalnephron. As a result of these actions on vas-cular tone and sodium homeostasis, AII playsa major role in blood pressure and blood vol-ume regulation. By blocking the formation ofAII, ACE inhibitors decrease the systemic arte-rial pressure (by decreasing vasoconstriction),facilitate natriuresis (e.g., by decreasing aldo-sterone and reducing Na+ reabsorption from

the distal nephron), and reduce adverse ven-tricular remodeling (see Chapter 9).

Another action of ACE inhibitors, whichlikely contributes to their hemodynamic ef-fects, is related to bradykinin (BK) metabo-lism, as shown in Figure 17.6. The natural vasodilator BK is normally degraded to in-active metabolites by ACE. Because ACE in-hibitors impede that degradation, BK accu-


Nesiritide

Figure 17.5. Examples of vasodilator drugs and their sites of action: the venous bed, thearteriolar bed, or both. ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker.

Actions impaired by ACE inhibitoror AT1 receptor antagonist

AT1receptor

antagonistACE

inhibitor

Actions enhancedby ACE inhibitor

Vasodilatation

Figure 17.6. The renin-angiotensin system. Angiotensin-converting enzyme (ACE) gener-ates angiotensin II, which results in actions that include vasoconstriction, sodium retention, andincreased sympathetic activity. ACE inhibitors and angiotensin II type 1 (AT1) receptor antago-nists impair these effects. ACE also promotes the degradation of the natural vasodilatorbradykinin; thus, ACE inhibition—but not AT1 receptor inhibition—results in accumulation ofbradykinin and enhanced vasodilatation.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 406


mulates and contributes to the antihyperten-sive effect, likely by stimulating the endothe-lial release of nitric oxide and biosynthesis ofvasodilating prostaglandins.

Clinical Uses

Hypertension

In hypertensive patients, ACE inhibitorslower blood pressure with little change in car-diac output or heart rate. One might assumethat because this class of drug interfereswith the renin-angiotensin system, it wouldbe effective only in patients with “high-renin” hypertension, but that is not the case.Rather, they are effective in most hyperten-sive patients, regardless of serum renin lev-els. The reason for this is not clear but mayrelate to the additional antihypertensiveeffects of BK and vasodilatory prostaglan-dins previously discussed. In addition, renin-angiotensin activity has been demonstratedwithin tissues outside the circulation, in-cluding the walls of the vasculature, whereACE inhibitors may exert a vasodilatoryeffect independent of the circulating reninconcentration.

ACE inhibitors increase renal blood flow,usually without altering the glomerularfiltration rate (GFR), because of dilation ofboth the afferent and efferent glomerular ar-terioles. Used alone in hypertension, ACEinhibitors show similar efficacy as diureticsand β-blockers. They do not adversely affectserum glucose or lipid concentrations, andunlike diuretics, they do not result in hypo-kalemia. ACE inhibitors are often recom-mended therapy in diabetic hypertensivepatients, because the drugs slow the devel-opment of diabetic nephropathy (a syn-drome of progressive renal deterioration,proteinuria, and hypertension) through fa-vorable effects on intraglomerular pressure.

Heart Failure

In heart failure, ACE inhibitors reduce pe-ripheral vascular resistance (decrease after-load), reduce cardiac filling pressures (de-

crease preload), and increase cardiac output.The rise in cardiac output usually matchesthe fall in peripheral resistance such thatblood pressure tends not to fall (remember,BP = CO × TPR), except in patients whoseintravascular volume is depleted as mightresult from overly vigorous diuretic ther-apy. The augmented cardiac output reducesthe drive for compensatory neurohormonalstimulation in CHF (see Chapter 9), suchthat elevated levels of norepinephrine fall.In addition, clinical trials have shown thatACE inhibitors significantly improve sur-vival in patients with chronic heart failure(see Chapter 9) and following myocardialinfarction (see Chapter 7). Some studieshave shown that ACE inhibition also re-duces the risk of myocardial infarction anddeath in patients with chronic vascular disease, including coronary artery disease(CAD), even if left ventricular function isnot impaired.

The available ACE inhibitors are listed inTable 17.4. The primary excretory pathwayof most of these agents is through the urine,so their dosages should generally be reducedin patients with renal dysfunction.

TABLE 17.4. Drugs That Interfere With the Renin-Angiotensin System

Drug Major Elimination Pathway

ACE inhibitorsBenazepril RenalCaptopril RenalEnalapril RenalFosinopril Hepatic/renalLisinopril RenalMoexipril Hepatic/renalPerindopril RenalQuinapril RenalRamipril RenalTrandolapril Hepatic/renal

Angiotensin II receptor antagonistsCandesartan Hepatic/renalEprosartan Hepatic/renalIrbesartan Hepatic/renalLosartan Hepatic/renalOlmesartan Hepatic/renalTelmisartan HepaticValsartan Hepatic/renal

Tab. 4

10090-17_CH17.qxd 8/31/06 4:58 PM Page 407

Toxicity

ACE inhibitors are generally well tolerated,and the adverse effects described here are notcommon.

Hypotension

This is a rare side effect when ACE inhibitorsare used to treat hypertension. It is morelikely to occur in heart failure patients inwhom intravascular volume depletion hasresulted from vigorous diuretic use. Such patients have significant activation of therenin-angiotensin system; therefore, bloodpressure is largely maintained by the vaso-constricting actions of circulating AII. Theadministration of an ACE inhibitor in thatsetting may result in hypotension becauseof the sudden reduction of AII levels. Thisside effect is avoided by temporarily reduc-ing the diuretic regimen and starting the ACEinhibitor at a low dosage.

Hyperkalemia

Because ACE inhibitors indirectly reduce theserum aldosterone concentration, the serumpotassium concentration may rise, but onlyrarely into the clinically important hyper-kalemic range. Conditions that can furtherincrease serum potassium levels and may re-sult in dangerous hyperkalemia during ACEinhibitor use include renal insufficiency,diabetes (caused by hyporeninemic hypo-aldosteronism, a condition often present inelderly diabetics), and concomitant use ofpotassium-sparing diuretics.

Renal Insufficiency

Administration of an ACE inhibitor to pa-tients whose intravascular volume is depletedmay result in hypotension with decreasedrenal perfusion and azotemia. Correctionof the volume depletion or reduction of theACE inhibitor dosage usually corrects thiscomplication.

ACE inhibitor therapy can also precipitaterenal failure in patients with bilateral renalartery stenosis because these patients rely on

high efferent glomerular arteriolar resistance(which is highly dependent on AII) to main-tain intraglomerular pressure and filtration.Administering an ACE inhibitor abruptly de-creases efferent arteriolar tone and glomer-ular hydrostatic pressure and may thereforeworsen GFR in this setting.

Cough

Irritation of the upper airways resulting in adry cough has been reported in up to 15% ofpatients receiving ACE inhibitor therapy. Itsmechanism has not been established butmay relate to the increased bradykinin con-centration provoked by ACE inhibitor ther-apy. This side effect may last several weeksafter the drug is discontinued.

Other Effects

Very rare adverse reactions to the ACE in-hibitors include angioedema and agranulo-cytosis. ACE inhibitors should not be usedin pregnancy because they have been shownto cause fetal injury in the second and thirdtrimesters.

Angiotensin II Type 1 Receptor Antagonists

Angiotensin II type 1 (AT1) receptor antag-onists, also termed angiotensin receptorblockers (ARBs), are a second group of drugsthat interfere with the renin-angiotensinsystem. There are at least two distinct typesof AII receptors: AT1 and AT2. All the actionsof AII known to affect blood pressure control(e.g., vasoconstriction, aldosterone release,renal Na+ reabsorption, and sympatheticnervous system stimulation) are mediatedby its binding to receptors of the AT1 sub-type. The AT2 receptor subtype is abundantduring fetal development and has been lo-cated in some adult tissues, but its preciseactions are unknown.

ARBs compete with AII for AT1 receptorsand therefore inhibit AII-mediated effects(see Fig. 17.6), thus lowering the blood pres-sure of hypertensive patients. ARBs providea more substantial blockade of the renin-


AQ3

AQ4

10090-17_CH17.qxd 8/31/06 4:58 PM Page 408


angiotensin system than ACE inhibitors, be-cause the latter do not completely block for-mation of AII (some AI is converted to AII bycirculating enzymes other than ACE). Un-like ACE inhibitors, the AT1 receptor antag-onists do not affect serum bradykinin levels.

The available ARBs are listed in Table 17.4.Each of these is excreted primarily in thebile but most are also partly excreted in theurine. Trials have demonstrated that ARBsare as effective as ACE inhibitors in treatinghypertension, and they are among the best-tolerated antihypertensive drugs. As withACE inhibitors, the antihypertensive effectof ARBs is enhanced by concurrent use of athiazide diuretic. Also like ACE inhibitors,ARBs have the potential side effects of hypo-tension and hyperkalemia (owing to reducedaldosterone levels). Unlike ACE inhibitors,ARBs typically do not cause cough.

In the setting of moderate to severe heartfailure, ARBs display hemodynamic bene-fits similar to those of ACE inhibitors buthave not demonstrated superiority over thelatter group (see Chapter 9). Thus, ARBs aregenerally recommended in heart failure forpatients who are intolerant of ACE inhibi-tors (e.g., because of ACE inhibitor–inducedcough).

Studies in patients with type 2 diabeteshave demonstrated that ARBs slow the pro-gression of kidney disease, an effect that alsohas been demonstrated with ACE inhibitors.

Direct-Acting Vasodilators

Hydralazine, minoxidil, sodium nitroprus-side, and diazoxide are examples of direct-acting vasodilators (Table 17.5). Hydralazineand minoxidil are used primarily as long-term oral vasodilators, whereas nitroprussideand diazoxide are administered intravenouslyin more-acute settings. Fenoldopam is anewer arterial vasodilator administered intra-venously for severe hypertension.

Hydralazine acts as a potent and direct ar-teriolar dilator at the level of the precapillaryarterioles and has no effect on systemic veins.The cellular mechanism of its effect is un-known. The fall in blood pressure followingarteriolar dilation results in a baroreceptor-mediated increase in sympathetic outflowand cardiac stimulation (e.g., reflex tachy-cardia), which could precipitate myocardialischemia in patients with underlying CAD.Therefore, hydralazine is often combinedwith a β-blocker to blunt this undesired response.

TABLE 17.5. Direct Vasodilators

Drug Clinical Use Route of Administration Major Adverse Effects

Hydralazine

Minoxidil

Nitroprusside

Fenoldopam

Diazoxide

CHF, congestive heart failure.

• Hypertension (chronicand acute therapy)

• CHF

• Chronic therapy of hypertension

• Hypertensive emergencies• Acute CHF

• Hypertensive emergencies

• Hypertensive emergencies

Oral, intravenous bolus,intramuscular

Oral

Intravenous infusion• Cyanide and thiocyanate

toxicityIntravenous infusion

Intravenous bolus

• Hypotension, tachycardia

• Headache, flushing• Angina• Drug-induced lupus• Reflex tachycardia• Na+ retention• Hypertrichosis• Hypotension

• Hypotension• Increased intraocular

pressure• Hypotension• Na+ retention• Hyperglycemia

Tab. 5

10090-17_CH17.qxd 8/31/06 4:58 PM Page 409

As newer drugs have emerged, hydralazineis now used only occasionally as an anti-hypertensive, often in combination withother drugs. It can be prescribed concur-rently with the venodilator isosorbide dini-trate to treat heart failure in patients withsystolic dysfunction. This combination im-proves symptoms in patients with mild-to-moderate heart failure and has been shownto reduce morbidity and mortality rates, mostdramatically in African American patients(see Chapter 9).

Hydralazine possesses low bioavailabilitybecause of extensive first-pass hepatic metab-olism. However, such metabolism dependson whether the patient displays fast or slowhepatic acetylation; on average, 50% of Amer-icans are fast and 50% are slow acetylators.Slow acetylators show less hepatic degrada-tion, higher bioavailability, and increasedantihypertensive effects, whereas fast acety-lators demonstrate the opposite responses.Hydralazine has a short half-life (2 to 4 hours)in the circulation, but its effect persists as longas 12 hours because the drug binds avidly tovascular tissue.

The most common side effects of hydra-lazine include headache (increased cerebralvasodilatation), palpitations (reflex tachy-cardia), flushing (increased systemic vaso-dilatation), nausea, and anorexia. As pre-viously indicated, tachycardia caused byreflex adrenergic stimulation may precipi-tate anginal attacks in patients with CAD ifhydralazine is not jointly administered witha β-blocker. Finally, a syndrome similar tosystemic lupus (characterized by arthralgias,myalgia, skin rashes, and fever) may de-velop, especially in patients who are slowacetylators.

Minoxidil also results in arteriolar vaso-dilatation without significant venodilation.Its mechanism of action may involve an in-crease in potassium channel permeability,which results in smooth muscle cell hyper-polarization and relaxation. Like other agentsthat selectively cause arteriolar dilation, re-flex adrenergic stimulation leads to increasedheart rate and contractility, an undesiredeffect that can be blunted by coadministra-tion of a β-blocker. In addition, decreased

renal perfusion often results in fluid reten-tion, so that a diuretic usually must be ad-ministered concurrently.

Minoxidil’s primary clinical indication isin the treatment of severe or intractablehypertension. It is especially useful in pa-tients with renal failure who are often re-fractory to other antihypertensive regimens.It is well absorbed from the gastrointestinaltract and is metabolized primarily by he-patic glucuronidation, but approximatelyone fifth is excreted unchanged by the kid-ney. Although it has a short half-life, itspharmacologic effects persist even afterserum drug concentration falls, probablybecause, like hydralazine, the drug bindsavidly to vascular tissue.

Side effects of minoxidil, in addition to re-flex sympathetic stimulation and fluid reten-tion, include hypertrichosis (excessive hairgrowth) and occasional pericardial effusion(unknown mechanism).

Sodium nitroprusside, a potent dilatorof both arterioles and veins, is used intra-venously to treat hypertensive emergenciesand, in intensive care settings, for bloodpressure control. It is also prescribed for pre-load and afterload modulation in severeCHF. Sodium nitroprusside is a complex ofiron, cyanide groups, and a nitroso moiety,and its metabolism by red blood cells resultsin the liberation of nitric oxide (Fig. 17.7).Nitric oxide causes vasodilation throughactivation of guanylate cyclase in vascularsmooth muscle (as described later in thischapter; see also Chapter 6).

Sodium nitroprusside’s hemodynamic ef-fects result from its ability to decrease arte-rial resistance and to increase venous capaci-tance. In patients with normal left ventricularfunction, it can actually decrease cardiac out-put because of the reduction in venous return(see Fig. 9.10). However, in a patient with im-paired left ventricular contractile function,the decreased systemic resistance induced bysodium nitroprusside (i.e., decreased after-load) augments forward cardiac output, whilevenous dilation reduces return of blood to theheart. The latter decreases pulmonary capil-lary hydrostatic pressure and improves symp-toms of pulmonary congestion.


Fig. 7

10090-17_CH17.qxd 8/31/06 4:58 PM Page 410


Sodium nitroprusside is often the treat-ment of choice for hypertensive emergen-cies because of its great potency and rapidaction. A β-blocker is often administered con-currently to counteract the reflex increase insympathetic outflow that may accompanythe use of this drug.

Sodium nitroprusside is administered bycontinuous intravenous infusion. Its onsetof action begins within 30 sec, and its peakeffect is achieved in 2 min. Its effectivenessdissipates within minutes of its discontinua-tion. After sodium nitroprusside is metabo-lized into nitric oxide and cyanide, the liver,in the presence of a sulfhydryl donor, trans-forms cyanide into thiocyanate; the thio-cyanate, in turn, is excreted by the kidney.Thiocyanate accumulation and toxicity, mani-fested by blurred vision, tinnitus, disorienta-tion, and/or nausea, may occur with contin-ued use, especially in the setting of renalimpairment. Thus, it is important to moni-tor serum levels of thiocyanate if sodiumnitroprusside is administered for more than 24 hours. In addition, excessive infusion ratesof sodium nitroprusside, or a deficiency inhepatic thiosulfate stores, can result in lethalcyanide toxicity, the early signs of which in-clude metabolic acidosis, headache, and nau-sea, followed by loss of consciousness.

Fenoldopam is a rapidly acting potentarteriolar vasodilator used intravenously totreat episodes of severe hypertension. It is aselective agonist of peripheral dopamine 1(D1) receptors, the activation of which re-sults in arteriolar vasodilatation through acAMP-dependent mechanism. Unlike other

intravenous antihypertensive agents, it ben-eficially maintains or enhances renal per-fusion, and its activation of renal tubular D1 receptors facilitates natriuresis. Unlikedopamine, fenoldopam does not stimulateα- or β-adrenergic receptors.

Fenoldopam is administered by continu-ous intravenous infusion. Its onset of ac-tion is rapid, achieving 50% of maximal ef-fect within 15 min and steady-state in 30 to60 min. It is metabolized by the liver to in-active substances that are excreted throughthe kidney. It has a rapid offset of actionafter discontinuation (an elimination half-life of <10 min), which is a desirable effectthat minimizes the risk of excessive bloodpressure reduction during the treatment ofhypertensive emergencies. These pharma-cologic properties also make fenoldopamuseful for controlling hypertension in thepostoperative setting. However, nitroprus-side works even faster and remains morepopular for this purpose. Unlike nitroprus-side, fenoldopam does not cause thiocyanatetoxicity. The most common side effects areheadache, dizziness, and tachycardia. Fenol-dopam also increases intraocular pressure(probably by slowing aqueous humor drain-age) and should be avoided in patients withglaucoma.

Diazoxide is a potent arteriolar dilatorthat is now infrequently used. Its mecha-nism of action involves activation of ATP-sensitive potassium channels, leading to ar-teriolar smooth muscle hyperpolarizationand vasodilatation. The fall in resistanceleads to a reflex activation of the adrenergic

Figure 17.7. Sodium nitroprusside is a complex of iron, cyanide(CN), and a nitroso group. Erythrocyte metabolism liberates CN and theactive vasodilator nitric oxide. The CN is metabolized in the liver to thio-cyanate, which is eliminated by the kidneys.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 411

nervous system with tachycardia, and to fluidretention because of activation of the renin-angiotensin system. It also inhibits pan-creatic insulin secretion and can result inhyperglycemia. This drug is administeredintravenously and has been used primarilyfor hypertensive emergencies. However, itsuse has declined in favor of newer and better-tolerated agents.

Calcium Channel Blockers

The calcium channel blockers (CCBs) arediscussed here as a group, but differencesexist among the drugs of this class. The com-mon property of CCBs is their ability to im-pede the influx of Ca++ through membranechannels in cardiac and smooth muscle cells.Two principal types of voltage-gated Ca++

channels have been identified in cardiac tissue, termed L and T. The L-type channelis responsible for the Ca++ entry that main-tains phase 2 of the action potential (the“plateau” in Fig. 1.14). The T-type Ca++

channel likely plays a role in the initial de-polarization of nodal tissues. It is the L-typechannel that is antagonized by currentlyavailable CCBs.

Mechanisms of Action

The cellular mechanism of CCBs has beenpartly delineated. Increased concentrationsof intracellular Ca++ lead to augmented con-tractile force in both myocardium and vas-cular smooth muscle. At both sites, the neteffect of Ca++ channel blockade is to decreasethe amount of Ca++ available to the contrac-tile proteins within these cells, which trans-lates into vasodilatation of vascular smoothmuscle and a negative inotropic effect incardiac muscle.

Vascular Smooth Muscle

Contraction of vascular smooth muscle de-pends on the cytoplasmic Ca++ concentra-tion, which is regulated by the transmem-brane flow of Ca++ through voltage-gatedchannels during depolarization. Intracellu-lar Ca++ interacts with calmodulin to form a

Ca++-calmodulin complex. This complexstimulates myosin light chain kinase, whichphosphorylates myosin light chains and al-lows myosin and actin to interact and causecontraction. CCBs promote relaxation of vas-cular smooth muscle by inhibiting Ca++ entrythrough the voltage-gated channels. Otherorgans possessing smooth muscle (includ-ing gastrointestinal, uterine, and bronchio-lar tissues) are also susceptible to this relax-ing effect.

Cardiac Cells

Cardiac muscle also depends on Ca++ influxduring depolarization for contractile proteininteractions, but by a different mechanismthan that in vascular smooth muscle. Ca++

entry into the cardiac cell during depolar-ization triggers additional intracellular Ca++

release from the sarcoplasmic reticulum,leading to contraction (see Chapter 1). Byblocking Ca++ entry, CCBs interfere withexcitation-contraction coupling and decreasethe force of contraction. Because the pace-maker tissues of the heart (e.g., sinoatrial [SA]and AV node) are the most dependent onthe inward Ca++ current for depolarization,one would expect that CCBs would reducethe rate of sinus firing and AV nodal con-duction. Some, but not all, CCBs have thisproperty (Table 17.6). The effect on cardiacconduction appears to depend not only onwhether the specific CCB reduces the in-ward Ca++ current but also on whether it de-lays recovery of the Ca++ channel to its pre-activated state. Verapamil and diltiazemhave this property, whereas nifedipine andthe other dihydropyridine CCBs do not (asdiscussed later).

Clinical Uses

As a result of their actions on vascularsmooth muscle and cardiac cells, CCBs areuseful in several cardiovascular disordersthrough the mechanisms summarized inTable 17.7. In angina pectoris, they exertbeneficial effects by reducing myocardialoxygen consumption as well as by potentiallyincreasing oxygen supply through coronary


Tab. 6

Tab. 7

10090-17_CH17.qxd 8/31/06 4:58 PM Page 412


dilatation. The latter effect is also useful in themanagement of coronary artery vasospasm.

CCBs are often used to treat hypertension.More so than β-blockers or ACE inhibitors,CCBs are particularly effective in elderly pa-tients. Nifedipine and the other dihydro-pyridines are the most potent vasodilatorsof this class.

CCBs are usually administered orally, andonce-a-day formulations are available formost of these agents. Routes of excretionvary. For example, nifedipine and verapamilare eliminated primarily in the urine, whereasdiltiazem is excreted through the liver. Com-mon side effects (see Table 17.6) includethe development of hypotension (owing to

excessive vasodilatation) and ankle edema(caused by local vasodilatation of peripheralvascular beds). Verapamil and diltiazem mayresult in bradyarrhythmias and should beused with caution in patients already receiv-ing β-blocker therapy.

The safety of short-acting CCBs has beencalled into question. In several observa-tional studies, a higher incidence of myocar-dial infarction or death has been reported inpatients with hypertension or coronary dis-ease taking such agents. In contrast, theseadverse outcomes have not been demon-strated with long-acting CCBs (i.e., formula-tions meant for once-a-day ingestion). Thus,only the long-acting versions should be used

TABLE 17.6. Calcium Channel Blockers

Negative Suppress AV Drug Vasodilation Inotropic Effect Node Conduction Major Adverse Effects

Verapamil + +++ +++ • Hypotension• Bradycardia, AV block• CHF• Constipation

Diltiazem ++ ++ ++ • Hypotension• Peripheral edema• Bradycardia

DihydropyridinesAmlodipine +++ 0 to + 0 • HypotensionFelodipine • Headache, flushingIsradipine • Peripheral edemaNicardipineNifedipineNisoldipine

AV, atrioventricular; CHF, congestive heart failure.

TABLE 17.7. Clinical Effects of Calcium Channel Blockers

Condition Mechanism

Angina pectoris ↓ Myocardial oxygen consumption↓ blood pressure↓ contractility↓ heart rate (verapamil and diltiazem)↑ Myocardial oxygen supply↑ coronary dilatation

Coronary artery spasm Coronary artery vasodilatationHypertension Arteriolar smooth muscle relaxationSupraventricular arrhythmias (Verapamil and diltiazem): Decrease conduction velocity

and increase refractoriness of atrioventricular node via block-ade of slow inward Ca++ current

AQ12

10090-17_CH17.qxd 8/31/06 4:58 PM Page 413

in most cases. Also, recall from Chapter 6that β-blockers and/or nitrates are preferredover CCBs for initial therapy in patientswith CAD.

Organic Nitrates

The nitrates constitute one of the oldesttreatments of angina pectoris. They are alsoused in other ischemic syndromes and inheart failure. The main physiologic action ofthe nitrates is vasodilatation, particularly ofthe systemic veins.

Mechanism of Action

Nitrates produce vascular smooth muscle re-laxation. The proposed mechanism involvesthe conversion of the administered drug tonitric oxide at or near the plasma membraneof vascular smooth muscle cells (Fig. 17.8).Nitric oxide, in turn, activates guanylate cyclase to produce cyclic guanosine mono-phosphate (cGMP), and the intracellularaccumulation of cGMP leads to smooth mus-cle relaxation. This mechanism of vascularsmooth muscle relaxation is similar to that as-sociated with nitroprusside and endogenousendothelial-derived nitric oxide.

Hemodynamic and Antianginal Effects

At low doses, nitroglycerin, the prototypicalorganic nitrate, produces greater dilation ofveins than of arterioles. The venodilation re-sults in venous pooling, diminished venousreturn, and hence decreased right and leftventricular filling. Systemic arterial resistanceis generally unaffected, but cardiac outputmay fall because of the diminished preload,especially in patients with intravascular vol-ume depletion (see Fig. 9.10). Arterial dila-tion occurs to some extent in the coronaryarteries and may also occur in the facial ves-sels and the meningeal arterioles, giving riseto the side effects of flushing and headache,respectively.

At high doses, nitrates result in widespreadarteriolar dilation and venodilation. Arteri-olar dilation may result in systemic hypo-tension and reflex tachycardia. However,the increase in heart rate is not typicallymanifest in patients with heart failure, be-cause decreasing afterload in that situationmay actually improve cardiac output and re-duce the sympathetic drive.

The major use of nitrates is in the treat-ment of angina pectoris, in which the re-duction of left ventricular filling reduces pre-load. The smaller left ventricular size lowers


Figure 17.8. Organic nitrates incite vascular smooth muscle (SM)relaxation by conversion to nitric oxide (NO) at or near the cellmembrane. Nitroprusside and endothelial-dependent vasodilators alsopromote NO delivery to vascular smooth muscle and cause relaxation. Inthe SM, NO stimulates formation of cyclic guanosine monophosphate(cGMP), which mediates relaxation.

Fig. 8

10090-17_CH17.qxd 8/31/06 4:58 PM Page 414


ventricular wall stress and myocardial oxy-gen consumption, which alleviates the oxy-gen imbalance in ischemic states. Nitrates arealso useful in patients with coronary arteryspasm (Prinzmetal variant angina) by dilat-ing the coronary arterioles.

Agents and Pharmacokinetics

Many formulations of nitrates are available.When the relief of acute angina is the objec-tive, rapid onset of action is essential. How-ever, in the long-term prevention of angi-nal attacks in a patient with chronic CAD,duration of action and predictability of ef-fect are more crucial than the speed of drugeffect.

Sublingual nitroglycerin tablets or spraysare used in the treatment of acute angina at-tacks. The peak action of these agents occurswithin 3 min, because they are rapidly ab-sorbed into the bloodstream via the oralmucosa; their effect, however, diminishesrapidly, falling off within 15 to 30 min, as thedrug is deactivated in the liver. These forms ofnitroglycerin are also effective when takenprophylactically, immediately before situa-tions known by the patient to produce angina(e.g., before walking up a hill).

The “long-acting” nitrates are used to pre-vent chest pain in the chronic managementof angina and must be given in sufficientdosage to saturate the liver’s deactivating ca-pacity. In this situation, high oral doses ofsustained-release nitroglycerin, isosorbidedinitrate, or isosorbide mononitrate areroutinely used. These agents have a dura-tion of action of 2 to 14 hours. Transdermalnitroglycerin patches or nitroglycerin pasteapplied to the skin also deliver a sustainedrelease of nitroglycerin. Of note, the efficacyof long-acting nitrate therapy is attenuatedby the rapid development of drug tolerancewith continuous use. For this reason, it isimportant that the dosing regimens allow adrug-free interval of several hours each dayto maintain efficacy.

Intravenous nitroglycerin is adminis-tered by continuous infusion is most usefulin the treatment of hospitalized patients withunstable angina or acute heart failure.

Toxicity

The most common adverse effects of thenitrates include hypotension, reflex tachy-cardia, headache, and flushing.

Natriuretic Peptides

As described in Chapter 9, natriuretic pep-tides are naturally secreted from atrial andventricular myocardium in patients withheart failure. Among their beneficial physi-ologic effects, these peptides promote vaso-dilation and result in sodium and water ex-cretion. The pharmacologic agent nesiritide(human recombinant B-type natriuretic pep-tide) is available for intravenous administra-tion to hospitalized patients with decompen-sated heart failure. It results in vasodilation,augmented cardiac output, and reductionof the undesired activation of the renin-angiotensin and sympathetic nervous sys-tems that are typical in heart failure. In somepatients, it promotes diuresis.

Nesiritide binds to G protein–coupled re-ceptors in multiple tissues, including theblood vessels (resulting in vasodilation), kid-neys, and adrenals. In the kidney, natriuresisis a consequence of several effects of thedrug. An augmented glomerular filtrationrate results from dilation of the afferent re-nal arterioles and constriction of the effer-ent renal arterioles, thereby increasing thefiltered load of sodium. In the proximaltubule, AII-mediated sodium uptake is in-hibited. Because the proximal tubule is wherethe vast majority of sodium is reabsorbed(as described later in the chapter), this inter-ruption in uptake results in sodium excre-tion. In the distal tubule, natriuretic peptidesappear to further reduce sodium reabsorp-tion through epithelial sodium channels.In the adrenal zona glomerulosa, the druginhibits aldosterone synthesis, which leadsto enhanced sodium excretion in the distalnephron.

Despite these benefits, the clinical role ofnesiritide is still being defined, because itsuse has not been shown to improve survivalin heart failure patients and in one study wasactually associated with increased mortality.

AQ5

10090-17_CH17.qxd 8/31/06 4:58 PM Page 415

Sildenafil

Sildenafil, a phosphodiesterase type 5 in-hibitor used to treat erectile dysfunction, hasbeen shown to decrease pulmonary vascularresistance in patients with primary pulmo-nary hypertension (PPH). It inhibits thebreakdown of cGMP in the pulmonary vas-culature, which enhances vasodilation andoxygenation. Other phosphodiesterase in-hibitors have not been shown to be effectivein PPH. When combined with nitrates,sildenafil can cause severe hypotension;

therefore, these groups of drugs should notbe prescribed concurrently.

ANTIADRENERGIC DRUGS

Drugs that interfere with the sympatheticnervous system are used commonly to treatcardiovascular disorders. These agents actat different loci, including the central ner-vous system (CNS), postganglionic sympa-thetic nerve endings, and peripheral α- andβ-receptors (Fig. 17.9).


Figure 17.9. Sites of action of the antiadrenergic drugs. Note that receptors at the sympathetic nerveending bind norepinephrine (NE) and provide feedback: the β-receptor stimulates, and the α2-receptor in-hibits, further NE releases. CNS, central nervous system.

Fig. 9

10090-17_CH17.qxd 8/31/06 4:58 PM Page 416


Normally, when a sympathetic nerve isstimulated, norepinephrine is released, tra-verses the synapse, and stimulates postsynap-tic α- and β-receptors. The consequences ofreceptor stimulation depend on the organinvolved (Table 17.8). The effect of α-receptorstimulation on vascular smooth muscle isvasoconstriction, whereas β2 stimulationcauses vasodilatation. In the CNS, α2 stim-ulation inhibits sympathetic outflow to theperiphery, thereby contributing to vaso-dilatation.

In addition, norepinephrine within thesynapse can bind to presynaptic β- and α2-receptors, which provides a feedback mech-anism that modulates further release of thehormone. The β-receptor increases and theα2-receptor inhibits further norepinephrinerelease.

Central Adrenergic Inhibitors

α2-Receptors are located in the presynapticneurons of the CNS. When stimulated by anα2-agonist, they lead to diminished sympa-thetic outflow from the medulla. This actionreduces peripheral vascular resistance anddecreases cardiac stimulation, resulting in afall in blood pressure and heart rate. Thus,drugs known as central adrenergic inhibitorsare actually agonists of the CNS α2-receptors.They were once among the most commonly

used antihypertensive drugs but have largelygiven way to better-tolerated agents. Theyare not sufficiently potent to serve as vaso-dilators in the treatment of heart failure.

The drugs in this group are listed in Fig-ure 17.9. They are all available as oral prepa-rations, and clonidine can also be prescribedas a skin patch that is applied and left inplace for 1 week at a time, facilitating drugcompliance in the treatment of hyperten-sion. Side effects of CNS α2-agonists includesedation, dry mouth, bradycardia, and if thedrug is stopped suddenly, the possibility ofa sudden, paradoxical rise in blood pressure.

Sympathetic Nerve-EndingAntagonists

Reserpine was the first drug found to inter-fere with the sympathetic nervous system. Itinhibits the uptake of norepinephrine intostorage vesicles in postganglionic and centralneurons, leading to norepinephrine degra-dation. The antihypertensive effect resultsfrom the depletion of catecholamines, whichcauses the force of myocardial contractionand total peripheral resistance to decrease.

Reserpine’s CNS toxicity represents itschief drawback. It often produces sedationand can impair concentration. The most seri-ous potential toxicity is psychotic depres-sion, and patients with a history of depressive

TABLE 17.8. Responses to Adrenergic Receptor Stimulation

Receptor Type Distribution Response

α1 Vascular smooth muscle (arterioles and veins) Vasoconstrictionα2 Presynaptic adrenergic nerve terminals Inhibition of NE release

Vascular smooth muscle (coronary and renal arterioles) Vasoconstrictionβ1 Heart Increases heart rate

Increases contractilitySpeeds AV node conduction

Kidney (JG cells) Increases renin releasePresynaptic adrenergic nerve terminals Increases NE releaseAdipose tissue Stimulates lipolysis

β2 Vascular smooth muscle (arterioles, except skin and Vasodilationcerebral)

Bronchial smooth muscle BronchodilationLiver Stimulates glycogenolysis

AV, atrioventricular; NE, norepinephrine.

Tab. 8

10090-17_CH17.qxd 8/31/06 4:58 PM Page 417

disorders should not receive this drug.Newer, better-tolerated antihypertensiveagents have largely supplanted the use of re-serpine and other sympathetic nerve-endingantagonists.

Peripheral a-Adrenergic Receptor Antagonists

Peripheral α-antagonists (Table 17.9) are di-vided into those that act on both α1- and α2-receptors, and those that inhibit α1 alone.α1-Selective receptor antagonists (prazosin,terazosin, doxazosin) are occasionally pre-scribed in the treatment of hypertension.Their selectivity for the α1-receptor explainstheir ability to produce less reflex tachy-cardia than nonselective agents. Normally,drug-induced vasodilatation results in baro-receptor-mediated stimulation of the sym-pathetic nervous system and an undesiredincrease in heart rate. This effect is ampli-fied by drugs that block the presynaptic α2-receptor, because feedback inhibition ofnorepinephrine release is prevented. How-ever, α1-selective agents do not block thenegative feedback on the α2-receptor. Thus,further norepinephrine release and reflexsympathetic side effects are blunted.

Historically, the principal indication forβ1-antagonists has been in the treatment ofhypertension. One of their advantages isthat they do not adversely affect the serumconcentrations of cholesterol and triglyc-erides as can other antihypertensives, suchas diuretics and β-blockers. However, in alarge prospective, randomized trial, patientstreated with the α1-antagonist doxazosin ex-perienced more adverse cardiac outcomes

than those treated with a thiazide diuretic.Thus, α1-antagonists have fallen out of favorin the management of hypertension. Thedrugs have also been evaluated in the treat-ment of heart failure; however, they losetheir effectiveness over time (i.e., displaydrug tolerance) and, unlike other vasodila-tor regimens (e.g., ACE inhibitors or hydra-lazine plus nitrates), do not reduce mortal-ity rates in chronic CHF. Terazosin anddoxazosin are mainly used today to treat thesymptoms of benign prostatic hyperplasia,because the drugs also beneficially relaxprostatic smooth muscle tone.

Phentolamine and phenoxybenzamineare nonselective α-blockers. They are used in the treatment of pheochromocytoma, atumor that abnormally secretes catechol-amines into the circulation (see Chapter 13).Otherwise, these drugs are rarely used be-cause the α2-blockade impairs the normalfeedback inhibition of norepinephrine re-lease, an undesired effect, as indicated earlier.

b-Adrenergic Receptor Antagonists

The β-adrenergic antagonists are used for a number of cardiovascular conditions, in-cluding ischemic heart disease, hyperten-sion, heart failure, and tachyarrhythmias.

Because catecholamines increase inotropy,chronotropy, and conduction velocity in theheart, it follows that β-receptor antagonistsdecrease inotropy, slow the heart rate, anddecrease conduction velocity. When stimula-tion of the β-receptors is low, as in a normalresting person, the effect of blocking agentsis likewise mild. However, when the sympa-


TABLE 17.9. a-Receptor Antagonists

Mechanism/Drug Indications Major Adverse Effects

Selective peripheral a1-blockade • Hypertension • Postural hypotensionPrazosin • Benign prostatic hyperplasia • Headache, dizzinessTerazosin • No reflex tachycardiaDoxazosinNonselective a-blockade • Pheochromocytoma • Postural hypotensionPhentolamine • Reflex tachycardiaPhenoxybenzamine • Arrhythmias

Tab. 9

10090-17_CH17.qxd 8/31/06 4:58 PM Page 418


thetic nervous system is activated (e.g., dur-ing exercise), these antagonists can sub-stantially diminish catecholamine-mediatedeffects.

The β-blockers can be distinguished fromone another by specific properties (Table17.10): (1) the relative affinity of the drugfor β1- and β2-receptors, (2) whether partialβ-agonist activity is present, (3) whether α1-receptors are also blocked, and (4) differ-ences in pharmacokinetic properties. Thegoal of β1-selective agents is to achieve myo-cardial receptor blockade, with less effect onbronchial and vascular smooth muscle (tis-sues that exhibit β2-receptors), thus produc-ing less bronchospasm and vasoconstrictionin susceptible patients. Agents with partialβ-agonist effects (also termed intrinsic sym-pathomimetic activity) tend to slow the heartrate less than other β-blockers.

During short-term use, nonselective β-antagonists tend to reduce cardiac outputbecause they decrease heart rate and con-tractility as well as slightly increase periph-eral resistance (via β2-receptor blockade). β-antagonists that have partial agonist activity(such as pindolol) or those that possess someα-blocking activity (such as labetalol) can ac-tually lower peripheral resistance by interact-ing with their respective β2- and α-receptors.

Clinical Uses

Ischemic Heart Disease

The beneficial effects of β-blockers in isch-emic heart disease are related to their ability

to decrease myocardial oxygen demand (seeChapter 6). They reduce the heart rate, bloodpressure (afterload), and contractility. Thenegative inotropic effect is directly related toblockade of the cardiac β-receptor, whichresults in decreased calcium influx into themyocyte (see Fig. 17.4). β-Blockers also im-prove survival and reduce the rate of reinfarc-tion following an acute myocardial infarc-tion. Agents with intrinsic sympathomimeticactivity are less beneficial in this regard thanβ-blockers without it.

Hypertension

β-Blocking agents are effective antihyperten-sive agents. Despite their widespread use inthis capacity, the mechanisms responsiblefor blood pressure lowering are not com-pletely understood. With initial use, the anti-hypertensive action is thought to result froma decrease in cardiac output, in associationwith slowing of the heart rate and mild de-crease in contractility. However, withchronic administration, other mechanismsare likely at work, including reduced renalsecretion of renin and possibly CNS effects.

Heart Failure

The negative inotropic effect of β-blockadewould be expected to worsen heart failuresymptoms in patients with underlying leftventricular systolic dysfunction. However,trials in patients with all classes of clini-cally stable heart failure have actually showna survival benefit with chronic β-blocker

TABLE 17.10. b-Adrenergic Blockers

Activity Nonselective b-Blockers b1-Selective b-Blockers

No β-agonist activity Carvedilola AtenololLabetalola BetaxololPropranolol BisoprololNadolol Esmololb

Timolol Metoprololβ-agonist activity Carteolol Acebutolol

PenbutololPindolol

aAlso has α1-adrenergic blocking properties.bAdministered intravenously only.

Tab. 10

10090-17_CH17.qxd 8/31/06 4:58 PM Page 419

administration using carvedilol, metoprolol,or bisoprolol (see Chapter 9). The mech-anism may relate to blunting of the cardio-toxic effects of excessive circulating cate-cholamines. Because of the potential risk ofactually transiently worsening heart failure intenuous patients, β-blocker therapy shouldbe started at low dosage, augmented slowly,and carefully monitored.

Other conditions that benefit from β-blocker therapy include tachyarrhythmias(as discussed later in the chapter) and hyper-trophic cardiomyopathy (see Chapter 10).

Toxicity

Fatigue may occur during β-blocker therapyand is most likely a CNS side effect. β-Blockerswith less lipid solubility (e.g., nadolol) do notpenetrate the blood-brain barrier and mayhave fewer CNS adverse effects than morelipid-soluble drugs, such as propranolol.Other potential adverse effects relate to thepredictable consequences of β-blockade:

1. β2-Blockade associated with use of non-selective agents (or large doses of β1-selective blockers) can exacerbate broncho-spasm, worsening preexisting asthma orchronic obstructive lung disease.

2. The impairment of AV nodal conductionby β1-blockade can provoke conductionblocks.

3. β2-Blockade can precipitate arterial vaso-spasm, which can result in Raynaud phe-nomenon or worsen symptoms of pe-ripheral vascular disease.

4. Abrupt withdrawal of a β-antagonist afterchronic use could precipitate myocardialischemia in patients with CAD.

5. Undesirable reduction of high-densitylipoprotein (HDL) cholesterol and eleva-tion of triglycerides can occur through anunknown mechanism. This effect appearsto be less pronounced with blockers thathave partial β-agonist activity or com-bined β- and α-blocking properties.

6. β2-Blockade may impair recovery fromhypoglycemia in diabetics suffering an in-sulin reaction. In addition, β-blockers maymask the sympathetic warning signs of

hypoglycemia, such as tachycardia. If β-blockers are used in diabetics, β1-selectiveagents are generally preferred.

Other potential side effects include insom-nia, depression, and impotence. Finally, β-antagonists should be used with caution incombination with nondihydropyridine CCBs(verapamil or diltiazem), because both typesof drugs can impair myocardial contractilityand AV nodal conduction, possibly precipitat-ing heart failure or AV conduction blocks.

ANTIARRHYTHMIC DRUGS

Drug therapy is a common approach totreat cardiac tachyarrhythmias. However,despite their benefits, antiarrhythmic drugsare among the most dangerous pharmaco-logic agents because of their frequent seri-ous adverse effects. Therefore, a thoroughunderstanding of their mechanisms of action,indications, and toxicities is of particularimportance.

Although a number of classification sys-tems for these agents exist, antiarrhythmicdrugs are commonly separated into fourgroups based on their electrophysiologicmechanisms of action (Table 17.11):

1. Class I drugs block the fast sodium chan-nel responsible for phase 0 depolarizationof the action potential. They are furtherdivided into three subtypes based on thedegree of sodium channel blockade andthe effect of the drug on the cell’s actionpotential duration.

2. Class II drugs are β-adrenergic receptorantagonists (β-blockers).

3. Class III drugs significantly prolong theaction potential with little effect on therise of phase 0 depolarization. The mainmechanism is blockade of the repolariz-ing K+ current.

4. Class IV drugs block the slow L-type calcium channel.

Drugs that do not conveniently fit into theseclasses (and are discussed separately) includeadenosine and the digitalis glycosides.

Regardless of the class, the goal of anti-arrhythmic therapy is to abolish the mech-


Tab. 11

10090-17_CH17.qxd 8/31/06 4:58 PM Page 420


anisms by which tachyarrhythmias occur(see Chapter 11). These mechanisms are (1) increased automaticity of pacemaker ornonpacemaker cells, (2) reentrant pathways,and (3) triggered activity.

In the case of arrhythmias caused by in-creased automaticity, treatment is aimed atlowering the maximum frequency at whichcardiac action potentials can occur by (1) re-ducing the slope of spontaneous phase 4 di-astolic depolarization and/or (2) prolongingthe effective refractory period. These actionsreduce or extinguish abnormally high ratesof firing.

Antiarrhythmic drugs inhibit reentrantrhythms by a different mechanism. The ini-tiation of a reentrant circuit relies on a re-gion of unidirectional block and slowed con-duction (Fig. 17.10). For a reentrant rhythmto sustain itself, the length of time it takesfor an impulse to propagate around the cir-cuit must exceed the effective refractory pe-riod of the tissue. If an impulse returns to anarea of myocardium that was depolarizedmoments earlier but has not yet recoveredexcitability, it cannot restimulate that tis-sue. Thus, one strategy to stop reentry is tolengthen the tissue’s refractory period. When

the refractory period is pharmacologicallyprolonged, a propagating impulse confrontsinactive sodium channels, cannot conductfurther, and is extinguished.

A second means to interrupt reentrantcircuits is to additionally impair impulse pro-pagation within the already slowed retro-grade limb. This is accomplished via phar-macologic blockade of the Na+ channelsresponsible for phase 0 depolarization. Suchblockade fully abolishes the compromisedimpulse conduction within the retrogradelimb and breaks the self-sustaining loop.

TABLE 17.11. Classification of Antiarrhythmic Drugs

Class General Mechanism Examples

I Na1 channel blockadeIA Moderate block (↓↓Phase 0 upstroke rate; prolonged AP duration) Quinidine

ProcainamideDisopyramide

IB Mild block (↓Phase 0 upstroke rate; shortened AP duration) LidocaineMexiletine

IC Marked block (↓↓↓Phase 0 upstroke rate; no change in AP duration) FlecainidePropafenone

II b-adrenergic receptor blockade PropranololEsmololMetoprololOthers

III Prolongation of action potential duration Amiodarone(predominantly via K+ channel blockade) Sotalol

BretyliumIbutilideDofetilide

IV Ca11 channel antagonists VerapamilDiltiazem

AP, action potential

A.Prolong

refractoryperiod

B.Impair conduction

even more

Figure 17.10. Two strategies to interrupt reentry. A.Prolonging the tissue refractory period causes returningimpulses to find the tissue unexcitable. B. Further reduc-ing conduction causes the impulse to “die out” in theslow retrograde limb of the circuit.

Fig. 10

10090-17_CH17.qxd 8/31/06 4:58 PM Page 421

The elimination of the third type of tachyarrhythmia, triggered activity, requiressuppression of early and delayed after-depolarizations.

An ideal pharmacologic agent would sup-press ectopic foci and interrupt reentrantloops without affecting normal conductionpathways. Unfortunately, when the con-centrations of antiarrhythmic drugs exceedtheir narrow therapeutic ranges, even nor-mal electrical activity may become sup-pressed. In addition, most antiarrhythmicdrugs have the potential to aggravate rhythmdisturbances (termed proarrhythmic effect).For example, this may occur when an anti-arrhythmic drug prolongs the action poten-tial and induces early afterdepolarizations,resulting in a triggered-type of arrhythmia,such as torsades de pointes (see Chapter 12).Drug-induced proarrhythmia occurs mostoften in patients with left ventricular dys-function or in those with an increased QTinterval (a sign that the action potential isprolonged).

Class IA Antiarrhythmics

Mechanisms of Action

Effect on Arrhythmias Caused by Increased Automaticity

Class IA agents produce moderate blockadeof the fast sodium channels, thus raisingthe threshold potential and slowing the up-stroke (phase 0) of the action potential. Inaddition, perhaps by inhibition of pacemakerchannels, the slope of phase 4 depolariza-tion is depressed (Fig. 17.11) so that it takeslonger to reach threshold and fire the actionpotential. These effects are most pronouncedat Purkinje fibers and abnormal ectopic pace-makers. Because IA agents have little effecton the automaticity of the SA node, the lat-ter can resume its function as the cardiacpacemaker after ectopic foci are suppressed.

Effect on Reentrant Arrhythmias

Because sodium channel blockade slows therate of phase 0 depolarization by reducingthe magnitude of the inward current, it re-

duces cellular and tissue conduction veloc-ities. If impaired sufficiently within a re-entrant circuit, the impulse will die outwithin the already slowed retrograde limb,aborting the rhythm. In addition, class IAagents prolong the cell’s refractory period,both by lengthening the action potential andby dissociating relatively slowly from Na+

channels after repolarization (see Fig. 17.11).Thus, an impulse traveling in the reentrantloop encounters unexcitable tissue and isextinguished.

Effect on the Electrocardiogram

Because the conduction velocity is decreasedand the action potential duration and repo-larization are prolonged, the effect of classIA agents is to mildly prolong the QRS andQT intervals (Table 17.12). At higher dos-ages, the drugs may substantially lengthenthese intervals, potentially setting the stagefor afterdepolarizations and drug-inducedarrhythmias.

Clinical Uses

Class IA drugs are effective in treating vari-ous reentrant and ectopic supraventricularand ventricular tachycardias (Table 17.13).However, their use has declined because ofthe development of more effective and lessproarrhythmic strategies, as discussed laterin the chapter.

Specific Class IA Drugs

Quinidine displays the electrophysiologiceffects inherent to class IA agents but alsohas anticholinergic properties that may aug-ment conduction at the AV node, thus antag-onizing its direct suppressant effect. Becauseof this increased AV nodal conduction,quinidine is often combined with a negativechronotropic agent such as a β-blocker, verap-amil, diltiazem, or digoxin. Quinidine alsodisplays an α-adrenergic blocking action thatmay cause hypotension, especially with par-enteral intravenous administration. There-fore, it is administered only by the oral route.Because quinidine is metabolized primarily


Tab. 12

Tab. 13

Fig. 11

10090-17_CH17.qxd 8/31/06 4:58 PM Page 422


by the liver, its dosage must be reduced in pa-tients with hepatic dysfunction.

Cardiac and noncardiac side effects occurfrequently during quinidine therapy. Themost common are related to the gastro-intestinal tract, including diarrhea in onethird of patients. Quinidine can cause ex-cessive prolongation of the QT interval,which may lead to the life-threateningventricular tachyarrhythmia torsades depointes, described in Chapter 12. In patients

Figure 17.11. Electrophysiologic effects of the class I antiarrhythmic drugs. A. Effect on the Purkinje cell actionpotential. B. Effect on pacemaker cell action potential.

TABLE 17.12. Effect of Antiarrhythmic Drugs on ElectrocardiographicIntervals

Class PR QRS QT

IA 0 ↑ ↑IB 0 0 0 or ↓IC ↑ ↑ 0 or ↑II 0 or ↑ 0 0 or ↓III 0 or ↑ 0 or ↑ ↑IV ↑ 0 0

10090-17_CH17.qxd 8/31/06 4:58 PM Page 423

taking digoxin, quinidine raises the bloodlevel of that drug because of reductions inthe body’s clearance and volume of distri-bution. Thus, it is important to reduce thedose of digoxin when quinidine is added.

The electrophysiologic effects of pro-cainamide are similar to those of quinidine,though procainamide does not prolongthe action potential (and therefore QT in-terval) as much. Nonetheless, the danger-ous arrhythmia torsades de pointes can stillbe provoked. Procainamide also has lesspronounced anticholinergic effects thanquinidine, so that facilitation of AV nodalconduction is less significant.

Procainamide has mild ganglionic block-ing effects that may cause peripheral vaso-dilatation and a negative cardiac inotropiceffect, particularly when the drug is ad-ministered intravenously. However, becausehypotension associated with intravenousprocainamide is much less common thanwith quinidine, it is used when an intra-venous class IA agent is desired.

Procainamide can be administered bymouth, intramuscularly, or intravenously.More than 50% of the drug is excreted un-

changed in the urine; the remainder under-goes acetylation by the liver to form N-acetylprocainamide (NAPA), which is subsequentlyexcreted by the kidneys. In renal failure, or in patients who are rapid acetylators, highserum levels of NAPA may accumulate.NAPA shares procainamide’s ability to pro-long the action potential and refractory pe-riod, but it does not alter the rate of phase 4depolarization or the slope of phase 0 up-stroke of the action potential.

Noncardiac side effects of procainamideare common and include fever and rash. Ap-proximately one third of patients develop asystemic lupus–like syndrome after 6 monthsof therapy, manifested by arthralgias, rash,and connective tissue inflammation. It mostoften occurs among patients who are slowacetylators and is reversible on cessation ofdrug therapy.

Disopyramide’s electrophysiologic andantiarrhythmic effects are similar to thoseof quinidine. However, disopyramide causesmuch fewer gastrointestinal side effects and does not increase serum digoxin levels.Also, disopyramide has a much greater anti-cholinergic effect, so that common side ef-fects include constipation, urinary retention,and exacerbation of glaucoma. More so thanquinidine or procainamide, disopyramidehas a pronounced negative inotropic effectand must be used with caution in patientswith left ventricular systolic dysfunction.

Disopyramide is administered orally. Theprimary excretory pathway is via the kid-neys, and toxic levels may accumulate inpatients with renal insufficiency. QT pro-longation and precipitation of ventriculararrhythmias (including torsades de pointes)can occur.

Class IB Antiarrhythmics

Class IB drugs inhibit the fast sodium chan-nel, but unlike IA agents, they typicallyshorten the action potential duration and therefractory period. Such shortening is attrib-uted to blockade of small sodium currentsthat normally continue through phase 2 ofthe action potential.


TABLE 17.13. Common Clinical Uses of Antiarrhythmic Drugs

Class Use

IA • Atrial fibrillation and flutter• Paroxysmal SVT• Ventricular tachycardia

IB • Ventricular tachycardia• Digitalis-induced arrhythmias

IC • Atrial fibrillation and paroxysmal SVTII • Atrial or ventricular premature beats

• Paroxysmal SVT• Atrial fibrillation and flutter• Ventricular tachycardia (ischemia-related)

III • Ventricular tachycardia (amiodarone andsotalol)

• Atrial fibrillation and flutter• Bypass-tract mediated paroxysmal SVT

(amiodarone)IV • Paroxysmal SVT

• Atrial fibrillation and flutter (↓ VR)• Multifocal atrial tachycardia (↓ VR)

SVT, supraventricular tachycardia; VR, ventricular rate.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 424


Class IB drugs at therapeutic concentra-tions do not substantially alter the electricalactivity of normal tissue; rather, they pref-erentially act on diseased or ischemic cells.Conditions present during ischemia—suchas acidosis, faster rates of cell stimulation,and increased extracellular potassium con-centration (and consequently a less negativediastolic membrane potential)—increase theability of class IB drugs to block the sodiumchannel. This blockade promotes conduc-tion block in ischemic cells by reducing theslope of phase 0 depolarization and slow-ing the conduction velocity, thus inhibit-ing reentrant arrhythmias (see Fig. 17.11).Similar to other class I drugs, the auto-maticity of ectopic pacemakers is also sup-pressed by decreasing phase 4 spontaneousdepolarization and (in the case of somedrugs of this class) by raising the thresholdpotential. In addition, intravenous lidocaine,a member of this class, suppresses delayedafterdepolarizations.

The most common use of class IB drugs isin the suppression of ventricular arrhyth-mias, especially those that appear in asso-ciation with ischemia or digitalis toxicity.Conversely, they have little effect on atrialtissue at therapeutic concentrations becauseof the shorter action potential duration ofatrial cells, which allows less time for thedrug to bind and block the Na+ channel.Thus, these agents are ineffective in atrialfibrillation, atrial flutter, and supraventricu-lar tachycardias.

Because the QT interval is not prolongedby class IB drugs, early afterdepolarizationsdo not occur, and torsades de pointes is notan expected complication.

Specific Class IB Drugs

Lidocaine is an antiarrhythmic drug com-monly used acutely to suppress ventriculararrhythmias in hospitalized patients. It isadministered intravenously only, becauseoral administration results in unpredictableplasma levels. As a result of rapid distribu-tion and hepatic metabolism, lidocaine mustbe administered as a continuous infusionfollowing two or three loading boluses. The

half-life of the drug depends greatly on he-patic blood flow. Reduced flow (as in heartfailure or in older individuals) or intrinsicliver disease can greatly increase serum li-docaine concentrations and toxic effects;therefore, the infusion rate should be low-ered in such patients.

The most common side effects of lido-caine are not cardiac; rather, they are relatedto the CNS and include confusion, dizzi-ness, and seizures. These effects are dosagerelated and can be prevented by monitoringserum levels of the drug or preemptively re-ducing the infusion rate when liver diseaseor decreased hepatic blood flow is suspected.

Mexiletine is structurally similar to lido-caine and shares its electrophysiologic prop-erties, but mexiletine is administered orally.Ninety percent of mexiletine is metabolizedin the liver to inactive products, and thedosage of the drug should be reduced in patients with hepatic dysfunction. Dose-related side effects of mexiletine are com-mon, especially of the CNS (dizziness, tre-mor, slurred speech) and the gastrointestinaltract (nausea, vomiting).

Class IC Antiarrhythmics

The class IC drugs are the most potent sod-ium channel blockers. They markedly de-crease the upstroke of the action potentialand conduction velocity in atrial, ventricu-lar, and Purkinje fibers (see Fig. 17.11). Al-though they have little effect on the durationof the action potential or refractory period ofPurkinje fibers, they significantly prolong therefractory period within the AV node and inaccessory bypass tracts.

The group IC agents were originally de-veloped to treat ventricular arrhythmias.However, that use has diminished becausestudies have shown an increased mortalityrate in patients taking class IC drugs for ven-tricular ectopy following myocardial infarc-tion and in those who have survived cardiacarrest. In patients with underlying left ven-tricular dysfunction, class IC drugs can pre-cipitate heart failure. Thus, drugs of this sub-class should be avoided in patients whohave other underlying heart abnormalities,

10090-17_CH17.qxd 8/31/06 4:58 PM Page 425

such as CAD or ventricular dysfunction.Class IC drugs have been shown to be bene-ficial (and reasonably safe) in preventingsupraventricular arrhythmias in patients whohave otherwise structurally normal hearts(see Table 17.13).

Flecainide is well absorbed after oral ad-ministration. Approximately 40% of the drugis excreted unchanged in the urine, and theremainder is converted to inactive metabo-lites by the liver. Cardiac toxicities includethe aggravation of ventricular arrhythmiasand precipitation of CHF in patients withunderlying left ventricular dysfunction. Non-cardiac side effects are referable to the CNSand include confusion, dizziness, and blurredvision.

The electrophysiologic properties of pro-pafenone are similar to those of flecainide,but additionally it has a weak β-adrenergicblocking action. Propafenone is metabolizedby the liver, but because the level of geneticvariation is high, a patient’s dosage must becarefully titrated by observing the drug’s ef-fect. Extracardiac side effects are not com-mon and include dizziness and disturbancesof taste.

Class II Antiarrhythmics

The class II drugs are β-adrenergic receptorantagonists, which are used in the man-agement of both supraventricular and ven-tricular arrhythmias. Most of their anti-arrhythmic properties can be attributed toinhibition of cardiac sympathetic activity.Additional actions of some β-blockers, such asβ1-cardioselectivity or a membrane-stabilizingeffect, seem to make no contribution to anti-arrhythmic activity.

Chapter 11 describes how β-adrenergicstimulation results in a more rapid upslope ofphase 4 depolarization and an increased fir-ing rate of the SA node. β-Adrenergic antago-nists inhibit these effects, thus reducing auto-maticity (Fig. 17.12). This action extends tothe cardiac Purkinje fibers, where arrhythmiasdue to enhanced automaticity are inhibited.In addition, because afterdepolarizationsmay be caused by excessive catecholamines,β-blockers may prevent triggered arrhythmiasinduced by that mechanism. All β-blockersincrease the effective refractory period of theAV node; therefore, these drugs are effectiveat interrupting reentrant rhythms that passthrough it.


Figure 17.12. Electrophysiologic effects of the class II antiarrhythmic drugs on the pacemaker cell actionpotential.

Fig. 12

10090-17_CH17.qxd 8/31/06 4:58 PM Page 426


β-Blockers may also have a beneficial anti-arrhythmic effect by decreasing myocardialoxygen demand, thus reducing myocardialischemia. Several drugs from this group havebeen shown to reduce mortality followingmyocardial infarction (see Chapter 7), whichmay in part relate to their antiarrhythmiceffect. Since the AV nodal conduction timeis prolonged by β-blockers, the PR intervalon the ECG may become prolonged (seeTable 17.12). The QRS and QT intervals areusually unaffected.

Clinical Uses

β-Blockers are most useful in suppressingtachyarrhythmias induced by excessive cat-echolamines (e.g., during exercise or emo-tional stimulation). They are also frequentlyused to slow the ventricular rate in atrialflutter and fibrillation by impairing conduc-tion and increasing the refractoriness of the

AV node. In addition, β-blockers may termi-nate reentrant supraventricular arrhythmiasin which the AV node constitutes one limbof the reentrant pathway.

β-Blockers are effective in suppressingventricular premature beats and other ven-tricular arrhythmias, especially when in-duced by exercise. They are also effective intreating ventricular arrhythmias related toprolongation of the QT interval because, un-like group IA agents, they do not pathologi-cally prolong that interval.

Class III Antiarrhythmics

Class III drugs are structurally dissimilarfrom one another but share the property ofsignificantly prolonging the action poten-tial of Purkinje and ventricular musclefibers (Fig. 17.13), predominantly by block-ing the outward K+ current that supportsphase 3 repolarization. Unlike class I agents,

Figure 17.13. Electrophysiologic effects of the class III antiarrhythmic drugs on the Purkinje cell actionpotential.

Fig. 13

10090-17_CH17.qxd 8/31/06 4:58 PM Page 427

class III antiarrhythmics generally have littleeffect on phase 0 depolarization or conduc-tion velocity.

Amiodarone is a powerful antiarrhythmicwith many potential adverse reactions. Itsmajor therapeutic effect is to prolong the ac-tion potential duration and refractoriness ofall cardiac fibers. However, it also shares ac-tions with each of the other antiarrhythmicclasses. The slope of phase 0 depolarizationmay be depressed through sodium channelblockade (class I effect), it exerts a β-blockingeffect (class II), and also demonstrates weakcalcium channel blockade (class IV). As aresult, the electrophysiologic effects of amio-darone are to decrease the sinus node fir-ing rate, suppress automaticity, interruptreentrant circuits, and prolong the PR, QRS,and QT intervals on the ECG.

In addition, amiodarone is a vasodilator(because of α-receptor and calcium channelblocking effects) and a negative inotrope (β-blocker and CCB effects). The resultingvasodilatation is more prominent than thenegative inotropic effect, so that cardiac out-put does not usually suffer in patients treatedwith this drug.

Amiodarone is more effective than mostother antiarrhythmic drugs for a wide spec-trum of ventricular and supraventriculartachyarrhythmias. These include atrial fi-brillation, atrial flutter, ventricular tachycar-dia, ventricular flutter, and supraventriculartachycardias, including those involving by-pass tracts. It is a first-line agent for the emer-gency treatment of ventricular arrhythmiasduring cardiac resuscitation (including ven-tricular fibrillation and ventricular tachy-cardia refractory to electrical shocks), and ismore effective than lidocaine for this pur-pose. It is commonly used to treat arrhyth-mias in patients with ventricular systolicdysfunction because it causes fewer pro-arrhythmic complications in that popula-tion than other agents. In addition, low-dose amiodarone is effective for long-termsuppression of atrial fibrillation and flutter.

Amiodarone is absorbed slowly from thegastrointestinal tract, requiring 5 to 6 hoursto reach peak plasma concentrations. It ishighly lipophilic, is extensively sequestered

in tissues, and undergoes very slow hepaticmetabolism. Its elimination half-life is longand variable, averaging 25 to 60 days. Thedrug is excreted by the biliary tract, lacrimalglands, and skin but not by the kidney; thus,its dosage does not need to be adjusted inpatients with renal failure. However, becausethe drug’s action has a delayed onset andvery long duration, amiodarone is difficultto regulate if side effects ensue.

There are numerous potential side effectsof amiodarone. The most serious is pulmo-nary toxicity, manifest by pneumonitis lead-ing to pulmonary fibrosis. Its origin is unclearbut may represent a hypersensitivity reactionand, if recognized early, is reversible.

Other life-threatening side effects ofamiodarone relate to cardiac toxicity: symp-tomatic bradycardia and aggravation of ven-tricular arrhythmia each occur in approx-imately 2% of patients. Because amiodaronesignificantly prolongs the QT interval, earlyafterdepolarizations and torsades de pointescan occur, but this happens only rarely.Intravenously administered amiodaroneoccasionally precipitates heart failure.

Abnormalities of thyroid function arecommon during amiodarone treatment, be-cause the drug contains a significant iodineload and because it inhibits the peripheralconversion of T4 to T3. During the first fewweeks of therapy, it is common to observetransient abnormalities of thyroid biochem-ical tests without clinical findings of thyroiddisease: serum TSH and T4 rise, and serum T3

falls. Over time, some patients develop overthypothyroidism (owing mostly to the anti-thyroid effects of iodine) or hyperthyroid-ism (because of either an iodine effect iniodine-deficient communities or a direct thy-roid inflammatory process incited by amio-darone in susceptible patients).

Gastrointestinal side effects of amiodaroneinclude anorexia, nausea, and elevation ofliver function tests, all of which improvewith lower doses of the drug. Neurologicside effects include proximal muscle weak-ness, peripheral neuropathy, ataxia, tremors,and sleep disturbances. Commonly, cornealmicrodeposits can be detected in patients re-ceiving chronic amiodarone therapy, butthese rarely affect vision.


10090-17_CH17.qxd 8/31/06 4:58 PM Page 428


As a result of these potential adverseeffects, ECGs, thyroid and liver functionblood tests, chest radiographs, and some-times pulmonary function studies are per-formed on a regular basis in patients receiv-ing chronic therapy. Amiodarone interactswith and increases the activity of certaindrugs, including warfarin and digoxin, suchthat the dosages of those agents must beadjusted. Because amiodarone prolongs theQT interval, other drugs that do the sameshould be used with great caution. Pharma-ceuticals that possess negative chronotropicor negative inotropic effects (e.g., β-blockers,verapamil, and diltiazem) should also gen-erally be avoided.

Sotalol is actually a nonselective β-blocker,but it is used in practice because of its addi-tional class III antiarrhythmic properties. Itprolongs the duration of the action poten-tial, increases the refractory period of atrialand ventricular tissue, and inhibits conduc-tion in accessory bypass tracts. The phase 0upstroke velocity is not altered in the usualdosage range. It is effective in the treatmentof both supraventricular and ventriculararrhythmias.

Because sotalol is excreted exclusively bythe kidneys, its dosage should be adjusted inthe presence of renal disease. Potential sideeffects include those of the β-blockers de-scribed earlier. Because the drug prolongsthe QT interval, the most serious potentialadverse effect is provoking the ventriculararrhythmia torsades de pointes. This com-plication occurs in approximately 2% of pa-tients and is more common in patients witha history of heart failure and in women (forunclear reasons).

Bretylium tosylate is an intravenouslyadministered class III drug used on occasionto treat life-threatening ventricular tachy-cardia or fibrillation, when all other attemptsat resuscitation have failed. Its mechanismof action is different from that of other anti-arrhythmic agents in that it acts at post-ganglionic adrenergic nerve terminals, whereit initially releases norepinephrine but theninhibits subsequent discharge. Thus, afterinitial stimulation, sympathetic activity ofthe heart decreases. The initial catecholamine

release can transiently aggravate arrhythmias,but continued therapy lengthens the ac-tion potential duration and refractoriness ofatrial, ventricular, and Purkinje fibers. As aresult, the threshold for ventricular fibrilla-tion is substantially raised.

Immediately after bretylium administra-tion, blood pressure may rise because of thecatecholamine release. However, significantorthostatic hypotension may follow becauseof the drug’s antiadrenergic actions.

Ibutilide is an intravenous antiarrhyth-mic agent used for the acute conversion ofatrial fibrillation or atrial flutter of recentonset. This agent prolongs the action po-tential duration and increases atrial andventricular refractoriness. The mechanismrelates to activation of a slow inward currentthat prolongs the plateau (phase 2) of theaction potential, rather than blockade ofpotassium currents that is typical of otherclass III drugs. In clinical trials, the successrate for conversion of atrial flutter is ap-proximately 60% but only 30% for those inatrial fibrillation.

Because ibutilide prolongs the QT inter-val, the torsades de pointes can be precipi-tated, especially in patients with underlyingventricular dysfunction. Therefore, carefulelectrocardiographic monitoring is necessaryfor several hours after drug administration.

Dofetilide acts by blocking the outwardpotassium current, causing prolongation ofthe action potential duration and an increasein the effective refractory period. It is usedorally to convert atrial fibrillation and atrialflutter to sinus rhythm and to maintain sinusrhythm after conversion. QT prolongationcomplicated by torsades de pointes is themajor potential adverse effect. Thus, like mostantiarrhythmics, drug administration shouldbe initiated in the hospital with electrocar-diographic monitoring. Dofetilide is excretedby the kidney, and its dose must be adjustedin patients with renal failure.

Class IV Antiarrhythmics

Class IV drugs exert their electrophysiologiceffects by selective blockade of the slow L-type cardiac calcium channels and include

AQ6

10090-17_CH17.qxd 8/31/06 4:58 PM Page 429

verapamil and diltiazem, but not nifedipineor the other dihydropyridine CCBs. They aremost potent in tissues in which the action po-tential depends on calcium currents, suchas the SA and AV nodes. Within nodal tis-sue, calcium channel blockade elevates thethreshold potential, decreases the rate of riseof phase 0 depolarization and conduction ve-locity, and lengthens the refractory period ofthe AV node (Fig. 17.14). These electrophys-iologic actions translate into their clinical ef-fects: (1) the heart rate slows; (2) transmis-sion of rapid atrial impulses through the AVnode to the ventricles decreases, thus slowingthe ventricular rate in atrial fibrillation andatrial flutter; and (3) reentrant rhythms trav-eling through the AV node may terminate.

A primary antiarrhythmic use of class IVdrugs is in the treatment of reentrant PSVT.At one time, intravenous verapamil was thetreatment of choice for acute episodes ofthis rhythm, but intravenous adenosine (de-scribed in the next section) has assumedthat role. The class IV antiarrhythmics arealso often used to slow the ventricular ratein patients with atrial fibrillation or flutter.

The pharmacology and toxicities of CCBswere presented earlier in this chapter. The

most important side effect of verapamil anddiltiazem, when administered intravenously,is hypotension. In addition, these agentsshould be avoided in patients receiving β-blocker therapy, because the combined neg-ative inotropic and chronotropic effects mayprecipitate heart failure.

Adenosine

Adenosine is an endogenous nucleoside witha very short half-life. Administered intra-venously, it is the most effective drug for therapid termination of reentrant PSVT.

Adenosine has substantial electrophysio-logic effects on specialized conduction tis-sues, especially the SA and AV nodes. Bybinding to specific adenosine receptors, itactivates potassium channels (Fig. 17.15).The resultant increase in the outward potas-sium current hyperpolarizes the membrane,which suppresses spontaneous depolariza-tion of the SA node and slows conductionthrough the AV node.

In addition, adenosine decreases the in-tracellular cAMP concentration by inhibit-ing adenylate cyclase. The result is a decreasein the inward pacemaker current (If) and


Figure 17.14. Electrophysiologic effects of the class IV antiarrhythmic drugs on the pacemaker cell actionpotential.

Fig. 14

Fig. 15

10090-17_CH17.qxd 8/31/06 4:58 PM Page 430


the inward calcium current (see Fig. 17.15).Thus, the net effect of adenosine is to slowthe SA node firing rate and to decrease AVnodal conduction. By inducing transient AVnodal block, adenosine terminates reentrantpathways that include the AV node as partof the circuit. Ventricular myocytes are rel-atively immune to these effects, in part be-cause the specific potassium channels re-sponsive to adenosine are not present inthose cells.

With a half-life of only 10 sec, adenosinehas very transient side effects (headache,chest pain, flushing, bronchoconstriction).Because methylxanthines (caffeine, theo-phylline) competitively antagonize the ade-nosine receptor, higher doses of adenosinemay be necessary in patients using thosesubstances. Conversely, dipyridamole inhibitsthe breakdown of adenosine and amplifies itseffect.

In summary, antiarrhythmic drugs havecomplex actions and display multiple car-diac and noncardiac toxicities. The poten-tial of inducing dangerous arrhythmias ex-

ists with most agents. Whenever antiarrhyth-mic drugs are used, patients must be followedclosely, the effectiveness of the drug demon-strated, and surveillance for toxicity contin-ued over the long term.

DIURETICS

Diuretics are most often used to treat heartfailure and hypertension. In heart failure,enhanced renal reabsorption of sodium andwater, with subsequent expansion of the ex-tracellular volume, contributes to peripheraledema and pulmonary congestion. Diureticseliminate excess sodium and water throughrenal excretion and are therefore the cor-nerstone of therapy to relieve congestivesymptoms (see Chapter 9). In the treatmentof hypertension, diuretics eliminate intra-vascular volume and in some cases promotevascular dilatation.

In the kidney, the rate of glomerular filtra-tion typically averages 135 to 180 L/day innormal adults. Most of the filtered Na+ is re-absorbed by the renal tubules, leaving only asmall quantity in the final urine (Fig. 17.16).

ADENOSINE

Figure 17.15. Mechanism of antiarrhythmic action of adenosine. Stimulation of themyocyte adenosine receptor activates potassium channels, and the resultant outward K+ cur-rent hyperpolarizes the membrane (resulting in decreased automaticity). Adenosine alsoinhibits membrane adenylate cyclase activity; the subsequent reduction in active proteinkinases decreases the inward pacemaker (If) and Ca++ currents (resulting in decreased auto-maticity and decreased conduction through the AV node). Xanthines compete for the adeno-sine receptor, blocking these effects. Dipyridamole interferes with cellular uptake and degra-dation of adenosine and therefore amplifies its effect.

Fig. 16

10090-17_CH17.qxd 8/31/06 4:58 PM Page 431

Approximately 65% to 70% of the filtered Na+

is reabsorbed isosmotically in the proximaltubule by active transport. In the thick as-cending limb of the loop of Henle, an ad-ditional 25% of the filtered sodium is re-absorbed, through a Na+-K+ cotransportsystem coupled to the uptake of two Cl− ions.Because this region is impermeable to the re-absorption of water, it is the site of hypotonictubular fluid formation, and the surroundinginterstitium becomes hypertonic. In the dis-tal convoluted tubule, an additional smallfraction of NaCl is reabsorbed (approximately5%). In the cortical collecting duct, Na+ per-meability is modulated by an aldosterone-sensitive mechanism, such that Na+ is re-absorbed into the tubular cells in the presenceof aldosterone, creating a lumen-negative po-tential difference that enhances K+ and H+ ex-cretion. Approximately 1% to 2% of sodiumreabsorption takes place at this location.

Most of the distal tubule is impermeableto water. In the collecting tubule, however,water permeability and reabsorption are pro-

moted by an antidiuretic hormone and driven by the osmotic gradient between thetubule and the hypertonic interstitium. Sub-stances that interfere with the antidiuretichormone, such as ethanol consumption,therefore have diuretic actions.

The three most commonly used groupsof diuretics are the loop diuretics, thiazidediuretics, and potassium-sparing diuretics(Table 17.14 and Fig. 17.16). These classesare generally distinguished by the site ofthe kidney tubule where they act and bytheir potency. Loop diuretics impair ab-sorption in the thick ascending limb of the loop of Henle, thiazide diuretics act onthe distal tubule and collecting segment,and potassium-sparing diuretics act on thealdosterone-sensitive region of the corticalcollecting tubule. A fourth group, the car-bonic anhydrase inhibitors, are weak diuret-ics rarely used in the treatment of hyperten-sion or heart failure. They act at the proximalconvoluted tubule, resulting in a loss of bi-carbonate (and sodium) in the urine.


Carbonicanhydraseinhibitors

Loopdiuretics

Thiazides

K++

sparingdiuretics

Figure 17.16. Schematic diagram of the renal tubules. Approximately 70% of filtered sodium is reabsorbed in theproximal convoluted tubule, 25% in the thick ascending limb of the loop of Henle, 5% in the distal convoluted tubule,and 1% to 2% in the cortical collecting tubule (mediated by the action of aldosterone). Antidiuretic hormone (ADH)increases the permeability of the distal nephron for water. Diuretics are secreted into the proximal convoluted tubuleand act at the sites shown.

Tab. 14

10090-17_CH17.qxd 8/31/06 4:58 PM Page 432


Diuretics, as active pharmacologic agents,are secreted into the proximal renal tubule,and their major sites of action are shownschematically in Figure 17.16.

Loop Diuretics

These agents are so named because they actprincipally on the thick ascending limb ofthe loop of Henle. They are powerful diuret-ics that result in the excretion of 20% to 25%of the filtered Na+ load through inhibition ofthe Na+-2Cl−-K+ cotransport system. Becauseinhibition at this site impairs the generationof a hypertonic interstitium, the gradient forpassive water movement out of the collectingduct is diminished and water diuresis results.

Loop diuretics are of great importance inthe acute management of pulmonary edema(administered intravenously) and in thetreatment of chronic heart failure or periph-eral edema (taken orally). Unlike other di-uretics, they tend to be effective in the set-ting of impaired renal function. In additionto the diuretic effect, and even preceding it,drugs of this class may induce venous vaso-

dilatation, which is also beneficial in reduc-ing venous return and pulmonary conges-tion (see Chapter 9). The mechanism of ve-nous vasodilatation appears to involve drug-induced prostaglandin and nitric oxide gen-eration from endothelial cells, which act torelax vascular smooth muscle.

The most common side effects of the loopdiuretics are intravascular volume depletion,hypokalemia, and metabolic alkalosis. Hypo-kalemia arises because (1) these agents impairthe reabsorption of sodium in the loop ofHenle, such that an increased amount of Na+

is delivered to the distal tubule, where itprompts greater-than-normal exchange forpotassium (and therefore more K+ excretioninto the urine); and (2) diuretic-induced in-travascular volume depletion activates therenin-angiotensin system. The subsequentrise in aldosterone promotes additional Na+-K+ exchange and hence potassium loss intothe urine.

Metabolic alkalosis during loop diuretictherapy results from two mechanisms: (1) in-creased H+ secretion into the distal tubule(and therefore into the urine) owing to the

TABLE 17.14. Commonly Used Diuretics

Method of Onset of Duration of Diuretic Administration Action (hours) Action (hours) Potential Adverse Effects

ThiazidesChlorothiazide

HydrochlorothiazideChlorthalidoneMetolazoneIndapamideLoop diureticsFurosemide

Bumetanide

Torsemide

Ethacrynic acidK1-sparing diureticsSpironolactoneEplerenoneTriamtereneAmiloride

GI, gastrointestinal; IV, intravenous; PO, by mouth.

POIVPOPOPOPO

POIVPOIVPOIVIV

POPOPOPO

10.25

221

1–2

15 min0.5–10.25<1

10 min0.25

>24>2422

6–1221224

12–2416–36

62

4–60.5–16–86–83

2–3 days24

12–1624

Hypokalemia, hypomagne-semia, hyponatremia,hypercalcemia, hyper-glycemia, hyperuricemia,hypercholesterolemia,hypertriglyceridemia, meta-bolic alkalosis

Hypotension, hypokalemia,hypomagnesemia, hyper-glycemia, hyperuricemia,metabolic alkalosis

Hyperkalemia, GI distur-bances; gynecomastia(spironolactone only)

AQ7

10090-17_CH17.qxd 8/31/06 4:58 PM Page 433

secondary hyperaldosteronism described ear-lier, and (2) contraction alkalosis, in whichreduced intravascular volume promotes in-creased sodium bicarbonate reabsorption bythe proximal tubule (see Fig. 17.16).

Additional side effects may also occur dur-ing continued loop diuretic therapy. Hypo-magnesemia may result, because magnesiumreabsorption depends on NaCl transport inthe thick ascending limb of the loop ofHenle, the action blocked by these drugs.Ototoxicity (cranial nerve VIII toxicity) occa-sionally develops, impairing hearing andvestibular function. It is thought to arisefrom electrolyte disturbances of the endo-lymphatic system, most likely because ofNa+-2Cl−-K+ cotransport inhibition by thediuretic at that site.

The most commonly used loop diuretic isfurosemide, the oral form of which demon-strates reliable gastrointestinal absorptionbut a short duration of action (4 to 6 hours)that limits its usefulness in the chronic treat-ment of hypertension. Bumetanide is simi-lar to furosemide and shares its actions andadverse effects but has greater potency andbioavailability. It also appears to have a lowerincidence of ototoxicity than the other drugsof this class. Bumetanide is sometimes use-ful in heart failure patients when edema isrefractory to other agents and in some pa-tients allergic to furosemide. Torsemide isalso similar to furosemide, with more com-plete bioavailability. Ethacrynic acid is theonly nonsulfonamide loop diuretic, so it canbe prescribed to patients who are intolerantof sulfonamide compounds. It is otherwisenot widely used because of its high incidenceof ototoxicity.

Thiazide Diuretics

Thiazides and related compounds (chlortha-lidone, indapamide, and metolazone) arecommonly used diuretics because they de-monstrate excellent gastrointestinal absorp-tion when administered orally and are usu-ally well tolerated. They are less potent thanthe loop diuretics but, because of their sus-tained actions, are useful in chronic condi-tions such as hypertension and mild CHF.

This class of drugs acts at the distal tubule,where they block the reabsorption of ap-proximately 3% to 5% of the filtered sodium(see Fig. 17.16). Na+ reabsorption at this siteis mediated through a Na+-Cl− cotransporteron the luminal membrane. The thiazides in-hibit this carrier by a mechanism that hasnot been elucidated but may involve com-petition for the Cl− site. The antihyperten-sive effect is initially associated with a de-crease in cardiac output owing to reducedintravascular volume and with unchangedperipheral resistance. With long-term thi-azide use, however, cardiac output often re-turns to normal as total peripheral resistancebecomes reduced by vascular dilatation.Indapamide is unique among this class inthat it displays a particularly prominent vaso-dilating effect.

Thiazides are typically administered orally.Diuresis occurs after 1 to 2 hours, but thefull antihypertensive effect of continuedtherapy may not become manifest for up to12 weeks (possibly related to the vasodilatormechanism alluded to in the previous para-graph). Chlorothiazide, the parent com-pound, has a low lipid solubility and hencelow bioavailability: higher doses are there-fore required to achieve therapeutic levelscompared with the more commonly usedhydrochlorothiazide. Chlorthalidone isslowly absorbed and hence has a long dura-tion of action. Metolazone, unlike otherdrugs of this class, is sometimes effective inpatients with reduced renal function.

Clinically, the thiazides differ from theloop diuretics in that they are less potent,have a longer duration of action, and (withthe exception of metolazone) demonstratepoor diuretic efficacy in the setting of im-paired renal function: generally, they are noteffective when the GFR is <25 mL/min.

Thiazides serve as the cornerstone of anti-hypertensive therapy because of their lowcost, effectiveness, and proven benefits in re-ducing the risk of stroke and cardiac events.They are sometimes used in heart failure,generally for patients with mild chroniccongestive symptoms. In addition, they canbe combined with a loop diuretic for pa-tients who have become refractory to the di-


10090-17_CH17.qxd 8/31/06 4:58 PM Page 434


uretic effect of the latter. The combination ofthe two classes lowers the dose-dependent ad-verse effects that accompany each drug; be-cause they act on sequential segments of therenal tubule, a more profound natriuretic ef-fect ensues than with either agent used alone.

Among the most important potential ad-verse effects of thiazides are (1) hypokalemiaand metabolic alkalosis, which result in partfrom increased Na+ delivery to the distaltubule, where exchange for K+ and H+ takesplace, and partly from volume contractionand secondary hyperaldosteronism, as pre-viously described for the loop diuretics; (2) hyponatremia, during prolonged treat-ment because of continued Na+ excretion inthe setting of chronic free water consump-tion; (3) hyperuricemia (and possible precipi-tation of gout) owing to decreased clearanceof uric acid; (4) hyperglycemia, because ofeither impaired pancreatic insulin releaseor decreased peripheral glucose utilization;(5) alterations in serum lipids (at least tran-siently), characterized by increased low-density lipoprotein (LDL) cholesterol andtriglycerides; and (6) weakness, fatigability,and paresthesias, which can occur with long-term use because of volume depletion andhypokalemia. In addition, serum calciumlevels often rise slightly during thiazide ther-apy, but this is rarely clinically significant.

In past decades, the standard thiazidedosage was excessive compared with currentpractice. By using lower dosages, it is possi-ble to accrue the benefits of this class of di-uretics while minimizing adverse effects.

Potassium-Sparing Diuretics

These agents are relatively weak diureticsthat antagonize physiologic Na+ reabsorptionat the distal convoluted tubule and corticalcollecting tubule. Potassium-sparing diuret-ics reduce K+ excretion; thus, unlike otherdiuretics, hypokalemia is not a side effect.They are used when maintenance of serumpotassium levels is crucial and in states char-acterized by aldosterone excess (e.g., primaryor secondary hyperaldosteronism). Two typesof drugs make up this group: (1) aldosteroneantagonists (e.g., spironolactone, and eplere-

none); and (2) direct inhibitors of Na+ per-meability in the collecting duct, which act in-dependently of aldosterone (e.g., triamtereneand amiloride).

Na+ and K+ exchange in the collectingtubules accounts for only a small percentageof sodium reuptake, preventing potent di-uresis from occurring when these agents areused alone. Therefore, they are often usedin combination with the loop or thiazideclasses for additive diuretic effect and to pre-vent clinically important hypokalemia.

Spironolactone is a synthetic steroid thatcompetes for the cytoplasmic aldosterone re-ceptor, thereby inhibiting the aldosterone-sensitive Na+ channel in the kidney. BecauseNa+ reabsorption through the sodium chan-nel is inhibited, no lumen-negative poten-tial exists to drive K+ and H+ ion excretion atthe distal nephron sites; thus, K+ and H+ ionsare retained in the circulation. Spironolac-tone also displays beneficial cardiac anti-remodeling effects in patients with heart fail-ure (see Chapter 9). In a trial of patients withsevere heart failure, spironolactone (addedto an ACE inhibitor and a loop diuretic, withor without digoxin) improved heart failuresymptoms and reduced mortality rates.

The most serious potential complicationof spironolactone is hyperkalemia, resultingfrom the impaired excretion of that ion.Thus, caution should be observed when ad-ministering K+ supplements, ACE inhibitors,or angiotensin receptor blockers concurrentwith potassium-sparing diuretics becausethey could contribute to this complication.Spironolactone also displays antiandrogenicactivity that may produce gynecomastia inmen and menstrual irregularities in women.

Eplerenone is a more specific inhibitorof the aldosterone receptor that does nothave the systemic anti-androgenic effects(i.e., gynecomastia) of spironolactone. Likespironolactone, it is used in the treatmentof hypertension and chronic systolic heartfailure. In patients with clinical evidence ofheart failure following a myocardial infarc-tion, eplerenone has also been shown toimprove survival when added to usual care.

Triamterene and amiloride are struc-turally related potassium-sparing diuretics

10090-17_CH17.qxd 8/31/06 4:58 PM Page 435

that act independently of aldosterone. At thedistal tubules, they inhibit the reabsorptionof Na+, which subsequently diminishes theexcretion of K+ and H+. Triamterene is me-tabolized by the liver, and its active prod-uct is secreted into the proximal tubule bythe organic cation transport system. Amilo-ride is secreted directly into the proximaltubule and appears unchanged in the urine.As with spironolactone, the most impor-tant potential adverse effect of these drugsis the development of hyperkalemia.

ANTITHROMBOTIC DRUGS

Platelets and the coagulation proteins play akey role in the pathogenesis of many cardio-vascular disorders, including the acute coro-nary syndromes, deep venous thrombosis,and thrombi that may complicate atrial fib-rillation, dilated cardiomyopathy, or mecha-nical prosthetic heart valves. Therefore, themodulation of platelet function and of thecoagulation pathway is often critically im-portant in cardiovascular therapeutics.

The formation of a thrombus, whether innormal hemostasis or in pathologic clot for-mation, requires three events: (1) exposure ofcirculating blood elements to thrombogenicmaterial (e.g., unmasking of subendothelialcollagen after atherosclerotic plaque rupture);(2) activation of platelets; and (3) triggeringof the coagulation cascade, ultimately re-sulting in a fibrin clot. Hemostasis effectedby platelets and the coagulation system areclosely interlinked: activated platelets accel-erate the coagulation pathway, and certaincoagulation proteins (e.g., thrombin) contri-bute to platelet aggregation.

This section focuses first on drugs thatinterfere with platelet function and then onthose that inhibit the coagulation cascade.Fibrinolytic agents, which dissolve clotsthat have already formed, are described inChapter 7.

Platelet Inhibitors

Platelets are responsible for primary hemo-stasis by a three-part process: (1) adhesion tothe site of injury, (2) release reaction (secre-

tion of platelet products and activation ofkey surface receptors), and (3) aggregation.For example, following blood vessel injury,platelets quickly adhere to exposed sub-endothelial collagen by means of membraneglycoprotein receptors, a process that de-pends on von Willebrand factor. Followingadhesion to the vessel wall, platelets releasethe preformed contents of their granules inresponse to agonists (including collagen andthrombin) that bind to platelet receptors.Among these repackaged substances are ade-nosine diphosphate (ADP), serotonin, fib-rinogen, growth factors, and procoagulants.Concurrently within the activated platelet,there is de novo synthesis and secretion ofthromboxane A2 (TXA2), a powerful vaso-constrictor (Fig. 17.17).

Certain agonists, including ADP, throm-bin, and TXA2, stimulate platelets to aggre-gate and form the primary hemostatic plug,as additional platelets are recruited from thecirculation. During this process, the plateletmembrane glycoprotein (GP) IIb/IIIa re-ceptors undergo a critical conformationalchange. This alteration allows the previ-ously inactive GP IIb/IIIa receptor to bindfibrinogen molecules, an action that tightlylinks platelets to one another and consti-tutes the final common pathway of plateletaggregation. The developing clump of plate-lets is stabilized and tethered to the site ofinjury by a developing mesh of fibrin, whichis produced by the simultaneous activationof the coagulation protein cascade.

Platelet activation is regulated to a greatextent by release of stored Ca++ from theplatelet-dense tubular system. This action re-sults in an increase in the cytosolic calciumconcentration, with activation of protein ki-nases and phosphorylation of intraplateletregulatory proteins. The increase in cytosolicCa++ also stimulates phospholipase A2 (PLA2),causing the release of arachidonic acid, theprecursor of TXA2 (see Fig. 17.17).

The critical release of calcium is modu-lated by several factors. Acting at theirrespective platelet membrane receptors,thrombin and other agonists generate inter-mediaries that stimulate the release of cal-cium from the dense tubules. TXA2 increases


AQ8

Fig. 17

10090-17_CH17.qxd 8/31/06 4:58 PM Page 436


the intracellular Ca++ by binding to itsplatelet receptor, which inhibits the activityof adenylate cyclase and thereby reducescAMP formation, an action that augmentsthe release of Ca++ from the dense tubules(see Fig. 17.17). Conversely, endothelial cell-derived prostacyclin (PGI2) stimulates adeny-late cyclase activity and increases plateletcAMP concentration, which inhibits Ca++ re-lease from the dense tubular system.

Antiplatelet drugs interfere with plateletfunction at various points along the se-quence of activation and aggregation (seeFig. 17.17). The most commonly used anti-platelet drug is aspirin, but the roles ofnewer antiplatelet drugs, especially the thie-

nopyridines and GP IIb/IIIa inhibitors, haverapidly expanded.

Aspirin

As described in the previous section, TXA2

is an important mediator of platelet activa-tion and clot formation. Aspirin (acetylsali-cylic acid) acts by irreversibly acetylating(and thus blocking the action of) cyclooxy-genase, an enzyme essential to TXA2 produc-tion from arachidonic acid (see Fig. 17.17).The form of the enzyme found in platelets iscyclooxygenase 1 (COX-1), which is effec-tively inhibited by the nonselective actionof aspirin (but it is not inhibited by selective

PLA2

Arachidonicacid

ATPcAMP IP3

Granulesecretion Cytoskeletal

reorganization

Activation ofglycoprotein

IIb/IIIareceptor

DENSETUBULES

Cyclooxygenase

PLATELETACTIVATION

TXA2

PLC PIP2

ADP ThrombinSerotoninProstacyclin TXA2

Gs Gi Gi

P2Y12 P2Y1AC

Ca++Ca++

– +

–+

GP IIb/IIIaRECEPTOR

ANTAGONISTS(prevent fibrinogen

binding)

ASPIRIN

–

THIENOPYRIDINES

–

DIPYRIDAMOLE+

–

Figure 17.17. Platelet activation is mediated by cytosolic Ca11. Factors that promote and inhibit calcium re-lease from the platelet dense tubules are shown. Thrombin and serotonin, acting at their specific receptors, stimu-late the formation of inositol triphosphate (IP3) from phosphatidylinositol diphosphate (PIP2) by phospholipase C(PLC). IP3 subsequently enhances the intracellular release of calcium. Thromboxane A2 (TXA2) also facilitates calciumrelease. It inhibits adenyl cyclase (AC) and reduces cyclic adenosine monophosphate (cAMP) formation. BecausecAMP normally prevents Ca++ release from the dense tubules, the reduction of this effect by TXA2 increases Ca++ re-lease into the cytosol. Conversely, endothelial-derived prostacyclin has the opposite effect. It reduces intraplateletcalcium release because it stimulates AC activity and cAMP formation. ADP also stimulates calcium release via itstwo receptors (see text for details). Calcium promotes the action of phospholipase A2 (PLA2), which generates theprecursors of TXA2 from the cell membrane. Platelet activation modulated by [Ca++] ultimately results in granule se-cretion, cytoskeletal reorganization, and the critical conformational change in glycoprotein IIb/IIIa receptors that isnecessary for platelet aggregation. The sites of action of commonly used antiplatelet drugs are shown in color. P2Y1

and P2Y12, purinoceptors.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 437

COX-2 antagonists, such as celecoxib). Be-cause platelets lack nuclei and thereforecannot synthesize new proteins (includingcyclooxygenase), aspirin permanently disablesTXA2 production in exposed cells.

The prostaglandin PGI2, a major antago-nist of TXA2 that is produced by endothelialcells, shares a dependency on cyclooxyge-nase activity for its formation, and aspirin,at high doses, impairs its synthesis as well.Unlike platelets, however, endothelial cellsare able to generate new cyclooxygenase toreplace what has been deactivated by acety-lation. Thus, when used at low doses, aspirineffectively inhibits platelet TXA2 synthesiswithout significantly interfering with thepresence and beneficial actions of PGI2.

Because the antiplatelet effect of aspirin is limited to inhibition of TXA2 formation,platelet aggregation induced by other factors(e.g., ADP) is not significantly impeded. Thus,aspirin is not a “complete” antithromboticagent.

Clinical Uses

Aspirin therapy has many proven clinicalbenefits in patients with cardiovascular dis-ease (Table 17.15). In individuals with un-stable angina, acute myocardial infarction,or a history of myocardial infarction, aspirinconclusively reduces the incidence of futurefatal and nonfatal coronary events. Simi-larly, in patients with chronic stable anginawithout a history of myocardial infarction,aspirin lessens the occurrence of subsequentmyocardial infarction and mortality. In pa-tients who have suffered a minor stroke ortransient cerebral ischemic attacks, aspirinreduces the rate of future stroke and cardio-vascular events. Additionally, aspirin lowersthe likelihood of graft occlusion in patientswho have undergone coronary artery bypasssurgery.

Less clear is the benefit of aspirin for pri-mary prevention (i.e., in individuals withouta history of cardiovascular events or symp-toms). When tested in a large cohort of


TABLE 17.15. Cardiovascular Uses of Antithrombotic Drugs

Unstable Mechanical Chronic Angina/ Heart Atrial

Drug Angina NSTEMI STEMI DVT Valve Fibrillation PCI HIT

Platelet inhibitorsAspirin + + + (1) (2) +Thienopyridines + + (3)GP IIb/IIIa inhibitors + (4) +Dipyridamole (5)AnticoagulantsUFH + + + (6) (6) +LMWH + + +Direct thrombin inhibitors (7) + +Fondaparinux +Warfarin (8) + + +

(1) Sometimes used in combination with warfarin.

(2) If patient has a low risk of stroke, or if warfarin is contraindicated.

(3) When intracoronary stent is implanted.

(4) If PCI undertaken.

(5) Sometimes used in combination with warfarin for recurrent embolism.

(6) For hospitalized patients unable to take warfarin.

(7) Emerging use.

(8) For 3–6 months after MI if large akinetic segment is present.

DVT, deep venous thrombosis; HIT, heparin-induced thrombocytopenia; LMWH, low-molecular-weight heparin; NSTEMI,non–ST-elevation myocardial infarction (MI); PCI, percutaneous coronary intervention; STEMI, ST-elevation MI; UFH, un-fractionated heparin.

Tab. 15

10090-17_CH17.qxd 8/31/06 4:58 PM Page 438


healthy American middle-aged men, aspirinwas associated with a reduced incidence ofnonfatal myocardial infarction but an in-creased rate of nonfatal hemorrhagic strokeand gastrointestinal bleeding; there was noeffect on total vascular mortality. Subsequentmeta-analyses of clinical trials have simi-larly concluded that aspirin is effective forprimary prevention of myocardial infarctionin patients with coronary risk factors, but italso increases the risk of hemorrhagic stroke.In the prospective, primary prevention trialknown as the Women’s Health Initiative, as-pirin lowered the risk of ischemic stroke inwomen but did not reduce the incidence ofMI or death from cardiovascular disease.Thus, whereas aspirin plays an extremelyimportant role in patients with known car-diovascular disease, it is not evident thatotherwise healthy people should routinelytake aspirin for “cardiovascular protection.”

Current recommendations are that as-pirin (at a dosage of 75 to 325 mg/day) beadministered to patients with clinical man-ifestations of coronary disease in the ab-sence of contraindications (i.e., aspirin al-lergy or complications described in the nextsection). It should not be prescribed rou-tinely for primary prevention purposes incompletely healthy individuals. However,pending the results of ongoing research,many physicians believe it is appropriate torecommend aspirin use in men and womenolder than age 50 who have at least onemajor atherosclerosis risk factor (see Chap-ter 5). In addition, the American DiabetesAssociation recommends that all diabeticswith at least one other coronary risk factortake aspirin for cardiovascular protection.Finally, aspirin is not as beneficial as war-farin (described later in the chapter) for theprevention of stroke in high-risk patientswith atrial fibrillation and should only beused in that setting when warfarin cannotbe safely administered.

Toxicity

The most common adverse effects of aspirinare related to the gastrointestinal system, in-cluding dyspepsia and nausea, which often

can be ameliorated by lowering the dosageand/or using enteric-coated or bufferedtablets. More serious potential side effectsinclude gastrointestinal bleeding, hemor-rhagic strokes, allergic reactions, and asthmaexacerbation in aspirin-sensitive patients.Because aspirin is excreted by the kidneysand competes with uric acid for the renalproximal tubule organic anion transporter,it may also occasionally exacerbate gout.

Thienopyridines

The thienopyridines inhibit ADP-mediatedactivation of platelets (see Fig. 17.17). Nor-mally, extracellular ADP activates plateletsby binding to two types of surface purino-ceptors. The first (termed P2Y1) acts throughphospholipase C to increase intraplatelet[Ca++], which potentiates platelet activation.The second purinoceptor (P2Y12) is linked toan inhibitory G protein and reduces cAMPproduction when activated, thus also raisingintraplatelet [Ca++] (see Fig. 17.17). ADP-in-duced platelet aggregation requires that ADPsimultaneously activate both P2Y1 and P2Y12

purinoceptors. The thienopyridines, afterconversion to active metabolites in the liver,irreversibly block the P2Y12 purinoceptor bybinding directly to it or to a nearby mem-brane protein. As a result, platelet aggrega-tion is inhibited.

Clopidogrel and ticlopidine and are thethienopyridines approved for clinical use.Both are well absorbed orally and have goodbioavailability. Meta-analyses of the use ofticlopidine or clopidogrel in patients proneto coronary syndromes have concluded thatthese drugs are modestly superior to aspirinin reducing the risk of myocardial infarction,stroke, or vascular deaths, but at an increasedrisk of side effects and at a higher economiccost. Studies evaluating the combination of aspirin plus clopidogrel in patients withunstable angina, non–ST-elevation and ST-elevation myocardial infarction have de-monstrated a benefit in cardiovascular out-comes compared with aspirin alone, but atan increased bleeding risk.

Important uses of the thienopyridines(especially clopidogrel) are as an antiplatelet

AQ9

10090-17_CH17.qxd 8/31/06 4:58 PM Page 439

substitute in patients allergic to aspirin andto prevent thrombotic complications fol-lowing percutaneous coronary stenting (seeChapter 6). The combination of clopidogrelplus aspirin is also approved for use in pa-tients with unstable angina or non–ST-elevation myocardial infarction to reducethe rate of recurrent cardiac events. A recentstudy showed that clopidogrel is also bene-ficial for this purpose in patients with ST-elevation myocardial infarction treated withfibrinolytic therapy (see Chapter 7).

Side effects of the thienopyridines includedyspepsia and diarrhea. In addition, the useof ticlopidine has been limited by potentiallylife-threatening adverse reactions: severeneutropenia (occurring in 0.8% to 2.5% of patients) and thrombotic thrombocyto-penic purpura (in approximately 0.02% oftreated patients). These hematologic effectsare much rarer with clopidogrel, which istherefore the preferred agent of this class.

Glycoprotein IIb/IIIa Receptor Antagonists

The GP IIb/IIIa receptor antagonists consti-tute one of the most potent classes of anti-platelet agents. This group reversibly inhibitsthe critical and final common pathway ofplatelet aggregation—the binding of acti-vated platelet GP IIb/IIIa receptors to fib-rinogen and von Willebrand factor. As a re-sult, platelets are inhibited from “sticking”to one another, impairing the formation ofa hemostatic plug. Three types of GP IIb/IIIareceptor antagonists have been developed:(1) monoclonal antibodies (e.g., abcix-imab, which is the Fab fragment of achimeric human-mouse monoclonal anti-body); (2) synthetic peptide antagonists(e.g., eptifibatide); and (3) synthetic non-peptide antagonists (e.g., tirofiban). As de-scribed in Chapters 6 and 7, the GP IIb/IIIaantagonists represent a major advance inimproving outcomes of patients under-going percutaneous coronary interventionsand in high-risk acute coronary syndromes.

All the GP IIb/IIIa receptor inhibitors incurrent use must be given intravenously.Oral GP IIb/IIIa receptor inhibitors have

been developed but have not demonstratedbeneficial outcomes in clinical trials.

The major side effects of the GP IIb/IIIareceptor inhibitors are bleeding (in 1% to10% of patients) and thrombocytopenia (inapproximately 2% of patients treated withabciximab and less commonly with the otheragents). Abciximab has a short plasma half-life (30 min); thus, its effects can be reversedby discontinuing the drug or by administer-ing a platelet transfusion. Because the otherGP IIb/IIIa receptor antagonists have longerhalf-lives, they may inactivate transfusedplatelets. However, bleeding complicationsare infrequent using current protocols andcareful dosing.

Dipyridamole

The antiplatelet drug dipyridamole is occa-sionally prescribed to patients who are in-tolerant to aspirin, but it is not as effective.Its mechanism of antiplatelet action is un-clear, but it may act, in part, by increasingplatelet cAMP levels, either by inhibitingthe enzyme phosphodiesterase or by block-ing cellular uptake and destruction of endo-genous adenosine. As a result, cytosolic[Ca++] falls, which inhibits platelet aggrega-tion (see Fig. 17.17). By itself, dipyridamolehas no proven benefits. It is used sometimesin combination with warfarin for an aug-mented antithrombotic effect in patientswith recurrent thromboembolism from pros-thetic heart valves, but the combination ofaspirin plus warfarin is more effective. Itsmost common current use is actually as anagent in pharmacologic stress testing (seeChapters 3 and 6).

Anticoagulant Drugs

Anticoagulant drugs (see Table 17.15) inter-fere with the coagulation cascade and im-pair secondary hemostasis. Because the finalstep in both the intrinsic and extrinsic co-agulation pathways is the formation of a fi-brin clot by the action of thrombin, majorgoals of anticoagulant therapy are to inhibitthrombin activation from prothrombin byfactor Xa (e.g., using unfractionated hepa-


AQ10

10090-17_CH17.qxd 8/31/06 4:58 PM Page 440


rin, low molecular weight forms of heparin,or fondaparinux); to inhibit thrombin it-self (e.g., with unfractionated heparin or di-rect thrombin inhibitors); or to decreasethe production of functional prothrombin(e.g., using warfarin).

Unfractionated Heparin

Unfractionated heparin (UFH) is a hetero-geneous mixture of highly charged muco-polysaccharide polymers. Although it has lit-tle anticoagulant action by itself, it associateswith antithrombin III (AT III) in the circula-tion, greatly increasing its effect. AT III (alsotermed simply antithrombin) is a naturalprotein that inhibits the action of thrombinand other clotting factors. When UFH com-plexes with AT III, the affinity of AT III forthrombin increases 1,000-fold, markedly re-ducing thrombin’s ability to generate fibrinfrom fibrinogen. The UFH–AT III complexalso inhibits activated factor X, additionallycontributing to the anticoagulant action.Furthermore, UFH has antiplatelet propertiesby binding to, and blocking the action of,von Willebrand factor.

UFH is administered parenterally becauseit is not absorbed from the gastrointestinaltract. For most acute indications, an intra-venous bolus is followed by a continuousinfusion of the drug. The bioavailability ofUFH varies from patient to patient becauseit is a heterogeneous collection of moleculesthat bind to plasma proteins, macrophages,and endothelial cells. The dosage–effect re-lationship is often not predictable; thus,frequent blood samples are required to mon-itor the degree of anticoagulation (mostcommonly, measurement of the activatedpartial thromboplastin time), so that the in-fusion rate can be adjusted properly.

The usual cardiovascular settings in whichintravenous UFH is indicated include (1) un-stable angina (see Chapter 6), (2) acute myo-cardial infarction after fibrinolytic therapy or if an extensive wall motion abnormalityis present (see Chapter 7), and (3) pulmonaryembolism or deep venous thrombosis (see Chap-ter 15). Among hospitalized or bed-riddenpatients not receiving intravenous heparin,

fixed low dosages of subcutaneous UFH areoften administered to prevent deep venousthrombosis.

The most important side effect of heparinis bleeding. An overdose of UFH can be treatedwith intravenous protamine sulfate, whichforms a stable complex with UFH and im-mediately reverses the anticoagulation effect.

Heparin-induced thrombocytopenia (HIT) isanother potential major adverse effect andcan occur in two forms. The more commontype, thought to result from direct heparin-induced platelet aggregation, occurs in up to15% of patients and is usually asympto-matic, dose dependent, and self-limited. Thismild HIT rarely causes severe reductions inplatelet counts and usually does not requirecessation of heparin.

The less common, much more dangerousform of HIT is immune mediated, a condi-tion that affects 3% of UFH-treated patients.It can lead to life-threatening bleeding and,paradoxically, to thrombosis. Thrombosisis caused by the formation of antibodies di-rected against heparin-platelet complexes,resulting in platelet activation, aggrega-tion, and clot production. In the immune-mediated form of HIT, the platelet count canfall markedly and is not dependent on thedose of heparin. Therapy requires immediatecessation of heparin and substitution byalternate antithrombotic therapy to preventfurther clotting (e.g., a direct thrombin in-hibitor, described later in the chapter).

Patients receiving long-term UFH therapyare also prone to a dose-dependent form ofosteoporosis through an unclear mechanism.

Low Molecular Weight Heparins

Some of the shortcomings of UFH (e.g., shorthalf-life and unpredictable bioavailability)have been addressed by the development oflow molecular weight heparins (LMWHs),examples of which are enoxaparin, dalte-parin and tinzaparin. As the name im-plies, LMWH molecules are approximatelyone-third the size of UFH molecules. LMWHsalso interact with AT III, but unlike UFH,the LMWH–AT III complex preferentiallyinhibits factor Xa more potently than it in-

10090-17_CH17.qxd 8/31/06 4:58 PM Page 441

hibits thrombin (thrombin inhibition re-quires heparin molecules larger than thosein LMWHs).

Advantages of LMWHs over UFH include(1) inhibition of platelet-bound factor Xa,contributing to a more prominent anticoagu-lant effect; (2) less binding to plasma proteinsand endothelial cells, resulting in more pre-dictable bioavailability and a longer half-life;(3) fewer bleeding complications; and (4) alower incidence of immune-mediated HIT.

From a practical standpoint, the major ad-vantages of LMWH formulations are the easeof use and more consistent level of anticoag-ulation. They can be administered as sub-cutaneous injections once or twice a day infixed doses, without the frequent blood mo-nitoring required for UFH. In rare cases inwhich monitoring the anticoagulant effect isnecessary (e.g., in patients with renal dys-function, because LMWHs are cleared via thekidneys), a factor Xa inhibition assay is used.

In clinical trials, LMWH therapy is at leastas effective as UFH in preventing deep venousthrombosis and treating unstable angina. Italso has a better safety profile than UFH,with lower rates of bleeding, thrombocyto-penia, and osteoporosis. LMWHs shouldnot, however, be used in patients with a his-tory of heparin-induced thrombocytopenia,and unlike UFH, the effects of LMWHscannot be completely reversed by protamine.Current clinical indications for LMWH ther-apy are (1) prophylaxis against deep venousthrombosis following hip, knee, or abdomi-nal surgery; (2) treatment of deep venousthrombosis (with or without pulmonary em-bolism); and (3) management of acute coro-nary syndromes.

Direct Thrombin Inhibitors

The anticoagulation effects of UFH andLMWH are limited because their activity de-pends, at least in part, on AT III, and they in-hibit only circulating thrombin. The largeheparin–AT III complex cannot inactivatethrombin that is already bound to fibrinwithin a clot. In contrast, the direct throm-bin inhibitors (lepirudin, bivalirudin, arga-troban, and others) inhibit thrombin activ-

ity independently of AT III and are effectiveagainst both circulating and clot-boundthrombin. They do not cause thrombo-cytopenia and are used to maintain anti-coagulation and prevent thrombosis inpatients with heparin-induced thrombo-cytopenia. Bivalirudin is approved for use asan anticoagulant in patients with unstableangina undergoing percutaneous coronaryinterventions. All the direct thrombin in-hibitors are potent anticoagulants and themajor adverse effect is bleeding.

Fondaparinux

The anticoagulant fondaparinux is a syn-thetic analog of heparin that specifically in-hibits factor Xa, thereby reducing thrombinactivation. Like heparin, fondaparinux bindsto AT III but with very high affinity, whichgreatly increases the ability of AT III to in-activate factor Xa. Unlike UFH, fondapari-nux does not inactivate formed thrombin,nor does it interfere with platelet actions orcause heparin-induced thrombocytopenia.It is administered by subcutaneous injec-tion, and its half-life is sufficiently long (17 to21 hours) that it can be prescribed just oncea day. There are no known antidotes to itsanticoagulant effect.

Fondaparinux is currently approved forprevention of deep venous thrombosis in pa-tients undergoing orthopedic surgery of thehip and knees, and as treatment for deep ve-nous thrombosis and pulmonary embolism.

Warfarin

Warfarin is an oral agent prescribed for long-term anticoagulation. It acts by antagonizingan enzyme (vitamin K epoxide reductase) thatis required for usual vitamin K metabolism.Normally, the reduced form of vitamin Kpromotes the carboxylation of a glutamicacid residue within specific coagulation fac-tors (factors II, VII, IX, and X), an action thatis necessary for the factors to subsequentlybind calcium, become functional, and parti-cipate in coagulation (Fig. 17.18). By interfer-ing with the formation of reduced vitamin K,warfarin indirectly inhibits carboxylation of


Fig. 18

10090-17_CH17.qxd 8/31/06 4:58 PM Page 442


the coagulation factors, rendering them in-active. Because certain natural coagulationinhibitors (protein C and protein S) are alsovitamin K dependent, warfarin impairs theirfunctions as well, which in some cases maycounteract the drug’s anticoagulant effect.

Warfarin’s anticoagulation action has adelayed onset of 2 to 7 days; thus, if an im-mediate effect is needed, UFH or LMWHmust be used concurrently at first. The half-life of warfarin is long (37 hours), and thedrug’s dosage must be individualized toachieve a therapeutic effect while minimiz-ing the risk of bleeding complications. Theextent of anticoagulation is monitored bymeasuring the prothrombin time in bloodsamples, reported as an International Nor-malized Ratio (INR). There are two “target”ranges of anticoagulant intensity. For pa-tients at greatest risk of pathologic throm-bosis (e.g., those with certain types of me-chanical heart valves), the desired INR is 2.5to 3.5. For others (e.g., those with uncom-plicated atrial fibrillation), the target INR is2.0 to 3.0.

Many factors can influence the anticoagu-lation effect of warfarin and require alter-ations in its dosage. For example, liver diseaseor heart failure reduces the warfarin require-ment, whereas a high dietary ingestion offoods containing vitamin K (e.g., green leafyvegetables) increases the dosage need. Simi-larly, many pharmaceuticals alter warfarin’santicoagulation effect (Table 17.16). Finally,the combined use of warfarin with aspirin or

other antiplatelet agents increases the risk ofa bleeding complication.

If serious bleeding arises during warfarintherapy, the drug’s effect can be reversedwithin hours by the administration of vita-min K (or even more quickly by transfusingfresh-frozen plasma, which directly replen-ishes functional circulating clotting factors).In patients with mechanical heart valves,vitamin K should be avoided unless life-threatening bleeding occurs, because of thepossibility of rebound valve thrombosis.

Warfarin is teratogenic and should not betaken during pregnancy, especially in thefirst trimester.

LIPID-REGULATING DRUGS

As described in Chapter 5, abnormal serumlipid levels play a critical role in the patho-genesis of atherosclerosis. Drugs that im-prove lipid abnormalities are cardioprotec-tive; they inhibit the progression (and inhigh doses may induce the regression) ofatherosclerosis, improve cardiovascular out-comes, and in high-risk patients, reducemortality rates. The most commonly usedlipid-regulating drugs are described in thissection.

HMG CoA Reductase Inhibitors

The HMG CoA reductase inhibitors, com-monly known as statins, are the most effec-tive drugs for reducing LDL cholesterol. By

Oxidizedform of

vitamin K

Reducedform of

vitamin K

Precursors offactors II, VII, IX, X

Carboxylatedfactors II, VII, IX, X(able to bind Ca++)WARFARIN

Epoxidereductase

–

Figure 17.18. Mechanism of action of warfarin. Normally, co-agulation factors II, VII, IX, and X are converted to functionalforms by carboxylation in the liver, in the presence of reduced vi-tamin K. During this reaction, vitamin K undergoes oxidation andmust be regenerated back to the reduced state for the sustainedsynthesis of functional clotting factors. Warfarin inhibits the for-mation of reduced vitamin K by antagonizing the enzyme epox-ide reductase, such that the conversion of the coagulation fac-tors does not occur and they remain nonfunctional.

Tab. 16

10090-17_CH17.qxd 8/31/06 4:58 PM Page 443

virtue of their potency, excellent tolerabil-ity, and mortality benefits, they are the mostwidely prescribed lipid-regulating drugs. Theavailable agents of this group, in increasingorder of potency, are fluvastatin, lovasta-tin, pravastatin, simvastatin, atorvastatin,and rosuvastatin (Table 17.17).

Statins are competitive inhibitors of the enzyme HMG CoA reductase, a rate-controlling factor in cholesterol biosynthe-sis (Fig. 17.19). By inhibiting cholesterolproduction in the liver, statins lower serumLDL cholesterol through three mechanisms:(1) the reduced intrahepatic cholesterol con-


TABLE 17.16. Drugs That Alter the Anticoagulation Effect of Warfarin

Reduced Anticoagulation Effect Increased Anticoagulation Effect

Hepatic enzyme induction Hepatic enzyme inhibitionBarbiturates AmiodaroneRifampin CephalosporinsCarbamazepine CimetidineNafcillin ErythromycinWarfarin malabsorption FluconazoleCholestyramine IsoniazidSucralfate Ketoconazole

MetronidazolePropafenoneTrimethoprim-sulfamethoxazoleDisplacement from protein binding sitesAllopurinolGemfibrozilPhenytoinAltered vitamin K production by gut floraCiprofloxacinPiperacillin

Tab. 17

Fig. 19

TABLE 17.17. Lipid-Regulating Drugs

Triglyceride Class LDL Effect HDL Effect Effect Adverse Effects

HMG CoA reductase inhibitors(in increasing order of potency)

FluvastatinLovastatinPravastatinSimvastatinAtorvastatinRosuvastatinBile acid–binding agentsCholestyramineColestipolColesevelamCholesterol absorption inhibitorEzetimibeNiacin

Fibric acid derivativesFenofibrateGemfibrozil

↓20–55%

↓15–30%

↓15–20%

↓10–25%

↓0–20% or↑0–10%

↑5–15%

↑3–5%

↑1–2%

↑15–35%

↑10–20%

↓10–30%

May ↑

↓0–5%

↓20–50%

↓20–50%

AQ13

Transaminitis, myopathy

Constipation, bloating

Rare allergic reaction

Flushing, hepatotoxicity,hyperglycemia, hyper-uricemia,

Nausea, gallstones

10090-17_CH17.qxd 8/31/06 4:58 PM Page 444


tent induces increased expression of the LDLreceptor gene, causing a greater number ofLDL receptors to appear on the surface of thehepatocyte, which facilitates the bindingand clearance of LDL from the circulation;(2) circulating LDL precursors (known asvery low density lipoprotein [VLDL] rem-nants and intermediate-density lipoproteinparticles) are cleared more rapidly from thecirculation because of their cross-recognitionwith the hepatic LDL receptor; and (3) hepa-tic VLDL production falls due to the reducedavailability of intracellular cholesterol forlipoprotein assembly. Because the catabo-lism of VLDL in the circulation ultimately

causes the formation of LDL, lowering VLDLproduction also decreases circulating LDLlevels. The reduced production of VLDL isalso likely responsible for the triglyceride-lowering effect of statins, because this lipo-protein is the major carrier of triglyceridesin the circulation.

The overall effect is that statins reduceserum LDL levels by 20% to 55%, dependingon which agent is used. Statins also decreaseplasma triglyceride levels by 7% to 30%, andby an unclear mechanism, HDL levels in-crease by 5% 15%.

The lowering of LDL reduces the lipid con-tent of atherosclerotic lesions and promotes

LDL LDL

LPL

VLDLremnants

VLDL(mostly TG)

Fibrates

NiacinCholesterol

Bile acids

Bile-acidbinding agents

HMGCOA

Statins

Hepatocyte

Intestine

TG

APO

–

Figure 17.19. Major sites of action of lipid-regulating drugs. Thestatins inhibit cholesterol biosynthesis in the liver by competing withthe enzyme HMG CoA reductase. This action depletes intrahepaticcholesterol stores, which results in increased expression of surface low-density lipoprotein (LDL) receptors. The latter enhance clearance of LDLand very low-density lipoprotein (VLDL) remnants from the circulation.The lower intrahepatic cholesterol content leads to reduction in VLDLsynthesis. Ezetimibe selectively inhibits cholesterol uptake in the smallintestine, thereby recuing chylomicron production and cholesterol de-livery to the liver. Bile acid–binding agents interrupt the enterohepaticcirculation of bile acids in the intestine, causing more hepatic choles-terol to be diverted to new bile acid production. In response to reducedavailability of intrahepatic cholesterol, LDL receptor expression and LDLclearance increase. Niacin inhibits VLDL production. It also increaseslipoprotein lipase (LPL) activity, thus promoting triglyceride (TG) clear-ance from circulating VLDL particles. Niacin raises circulating high-density lipoprotein (HDL) by impairing hepatic uptake of apo AI, amajor HDL apoprotein (not shown). Fibrates enhance VLDL catabolismby increasing the synthesis of lipoprotein lipase via peroxisomeproliferator-activated receptor α (PPAR-α), a nuclear transcriptionfactor. They also raise HDL by stimulating the production of HDL-associated apoproteins (not shown). Apo, apoproteins.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 445

plaque stability. This lessens the vulnerabil-ity of plaque to rupture, thus decreasing thelikelihood of thrombus formation and vas-cular occlusion. In addition to their lipid-modulating properties, statins have otherpotentially cardioprotective effects. They im-prove endothelial function as evidenced byenhanced synthesis of nitric oxide. They fur-ther promote plaque stability by inhibitingmonocyte penetration into the arterial walland reducing macrophage secretion of met-alloproteinases, enzymes that degrade andweaken the fibrous caps of plaques. Statinsalso diminish the vulnerability of lipopro-teins to oxidation, thus inhibiting the unreg-ulated uptake of modified LDL cholesterol bymacrophages. Finally, they appear to suppressinflammation, thought to be a key aspect ofatherogenesis.

Statins are widely prescribed for patientswith CAD, because major trials have shownthat they substantially reduce mortality, car-diac events, and strokes in this population,whether LDL cholesterol is elevated or evenin the “normal” range. In studies of patientsnot known to have CAD, statin therapy hasbeen shown to reduce coronary events inhigh-risk individuals—those with elevatedLDL cholesterol levels or those with averagetotal cholesterol but low HDL values.

Statins are well-tolerated drugs. Mildgastrointestinal upset or sleep disturbancesoccasionally occur. The most significantpotential adverse effects are hepatotoxicityand myotoxicity (skeletal muscle toxicity),and these are rare. Hepatotoxicity is dose re-lated and occurs in fewer than 1% of pa-tients. Those affected may experience fatigue,anorexia, and weight loss. More commonly,the patient is asymptomatic, but results ofroutine laboratory studies show an increasein transaminase levels (ALT, AST). Symp-toms disappear almost immediately after thedrug is discontinued, but transaminase lev-els may remain elevated for weeks. The riskof statin-associated hepatic toxicity is higherin patients who drink excessive amounts ofalcohol.

Myotoxicity is characterized by intensemyalgias and muscle weakness and canlead to rhabdomyolysis (destruction of mus-

cle) with myoglobinuria and renal failure.When this occurs, muscle-derived creatinekinase levels in the serum rise to more than10 times the upper limit of normal. This se-vere complication occurs in fewer than 0.1%of patients taking statins alone. However,the incidence is increased significantly byconcomitant therapy with certain otherdrugs, including other lipid-lowering agents(i.e., niacin and the fibric acid derivatives).The incidence is also increased by concur-rent administration of drugs that inhibitthe 3A4 isoform of cytochrome P-450, whichis responsible for hepatic metabolism ofmost statins. Such drugs include macrolideantibiotics (e.g., erythromycin, clarithromy-cin); azole antifungal agents (e.g., ketocona-zole, itraconazole); cyclosporine; and manyHIV protease inhibitors. Notably, pravastatinand fluvastatin are not substantially depen-dent on the cytochrome P-450 3A4 isoformfor their metabolism and appear to be lesslikely to cause myopathy in combinationwith these other drugs.

Bile Acid–Binding Agents

This group includes the resins cholestyra-mine and colestipol and the hydrophilicpolymer colesevelam. These drugs are large,highly positively charged molecules that bindbile acids (which are negatively charged) inthe intestine and prevent their normal re-absorption to the liver through the entero-hepatic circulation (see Fig. 17.19). To makeup for the loss, more hepatic cholesterol isconverted into newly produced bile acids.This action depletes intrahepatic cholesterolstores, thus stimulating the production ofLDL receptors. Similar to the effect of thestatins, an increased number of hepatic LDLreceptors bind a greater amount of circulat-ing LDL, reducing the circulating concen-tration of the lipoprotein. However, unlikestatins, new hepatic cholesterol productionis also stimulated by the reduced intrahepaticcholesterol content. The boost in choles-terol synthesis augments VLDL production,which likely explains the commonly observedrise in serum triglyceride levels during ther-apy with a bile acid–binding agent.


10090-17_CH17.qxd 8/31/06 4:58 PM Page 446


In one of the first drug trials of men withhypercholesterolemia, cholestyramine sig-nificantly reduced the risk of fatal and non-fatal myocardial infarctions. However, drugsof this class are difficult for patients to take(e.g., cholestyramine and colestipol are un-appealing gritty powders that must be mixedwith liquids) and their potency is inferior tothat of statins. Thus, the bile acid–bindingagents are only occasionally used today,mainly as second-line lipid-regulating drugs,in combination with statin therapy. Becausethey can cause elevations of the serum tri-glyceride level, they should be avoided inpatients with hypertriglyceridemia.

The bile acid–binding agents interfere withthe absorption of fat-soluble vitamins andcertain drugs (e.g., warfarin, digoxin, propra-nolol, and thyroid hormones). Thus, othermedications should be consumed 1 hour be-fore or 3 to 4 hours after these agents. Themain side effects are gastrointestinal: bloat-ing, constipation, and nausea. Because bileacid–binding agents are not absorbed intothe circulation, they do not cause systemicadverse reactions.

Cholesterol AbsorptionInhibitors

Ezetimibe, the first member of this class ofagents, is a selective inhibitor of cholesteroluptake at the brush border of epithelial cellsin the small intestine. It is believed to com-petitively inhibit a transporter known as theNiemann-Pick C1–like 1 protein. Normally,a portion of dietary and biliary cholesteroltaken up in this manner is esterified and in-corporated into chylomicrons, which thenenter into the circulation and are trans-ported to the liver (see Box 5.1). By inhi-biting cholesterol uptake (see Fig. 17.19),ezetimibe results in reduced chylomicronproduction and therefore less cholesteroldelivery to the liver. The reduced cholesterolcontent stimulates compensatory hepaticproduction of LDL receptors, which augmentclearance of circulating LDL particles. Thenet result is lowering of circulating LDL (andtherefore serum cholesterol levels).

Used alone at standard dosage (10 mg/day), ezetimibe reduces LDL cholesterol by

about 18%. However, when combined withstatin therapy, marked lowering of LDL (byup to 60%) has been reported. Unlike bileacid–binding agents, side effects from eze-timibe therapy appear to be rare. When com-bined with a statin, the incidence of trans-aminase elevation is only slightly greaterthan that of statin therapy alone. The addi-tion of ezetimibe to a statin regime does notappear to increase the risk of statin-associatedmyopathy.

Niacin

Niacin is one of the oldest lipid-regulatingdrugs and has favorable effects on all the cir-culating lipid fractions. It is the most effec-tive agent available for raising HDL choles-terol (by 15% to 35%), and it reduces LDLcholesterol (by 5% to 25%) and triglyceridelevels (by 20% to 50%). Furthermore, unlikemost other lipid-lowering drugs, niacin sub-stantially reduces the circulating level of lipo-protein (a), an LDL-like lipoprotein that car-ries an independent risk of cardiovasculardisease (see Chapter 5).

Niacin modifies lipid levels through multi-ple mechanisms. It inhibits the release offatty acids from adipose tissue. As a result,fewer fatty acids are transported to the liverand hepatic triglyceride synthesis declines.Impaired triglyceride production by the liverreduces VLDL secretion into the circula-tion; consequently, less LDL is formed (seeFig. 17.19). Niacin also enhances the clear-ance of triglycerides from circulating VLDLby promoting the activity of lipoprotein li-pase, the enzyme that processes VLDL parti-cles by hydrolyzing the triglyceride core atadipose and muscle cells. The net effect ofthese actions is a reduction in serum triglyc-eride and LDL levels. In addition, the drugreduces the proportion of small, dense LDLparticles (which are thought to promoteatherogenesis) in favor of larger and morebuoyant forms. Niacin raises circulating HDLcholesterol levels by decreasing the hepaticuptake of its apoprotein, apo A1, thus reduc-ing clearance of HDL particles from the cir-culation. This mechanism does not disturbhepatic retrieval (and disposal) of cholesterolfrom the HDL particles.

AQ11

10090-17_CH17.qxd 8/31/06 4:58 PM Page 447

In a major study of niacin in men whohad experienced a prior myocardial infarc-tion, niacin reduced the risk of future car-diac events and lowered the mortality ratein long-term follow-up. That study was per-formed before the better-tolerated and moreeffective statin drugs were available. Niacinis now mainly used to treat patients withlow serum HDL levels and/or elevated serumtriglycerides.

Niacin has several common side effects.Transient cutaneous flushing episodes occurin most patients. These episodes are pro-staglandin mediated and can be minimizedby taking aspirin prior to the niacin dose.Gastrointestinal side effects include nauseaand exacerbation of peptic ulcer disease.Hepatotoxicity can occur, manifested by fa-tigue, weakness, and elevated serum trans-aminases (ALT, AST). Niacin should be usedcautiously in diabetic patients because it caninduce insulin resistance and contribute tohyperglycemia. Niacin also raises serum uricacid levels and can precipitate gout in sus-ceptible patients. Rare cases of myopathyhave been reported with niacin therapy. Theincidence is increased when niacin is pre-scribed concurrently with a statin.

Fibrates

The fibric acid derivatives, or fibrates, in-clude gemfibrozil and fenofibrate. They arethe most powerful agents to reduce serumtriglyceride levels (by up to 50%), and theyraise HDL cholesterol levels (by up to 20%).However, their effect on LDL cholesterol ismore variable: they can actually increase LDLin patients with preexisting hypertriglyc-eridemia. Fibrates shift the proportion ofLDL from smaller and denser sizes to morebuoyant, larger, and presumably less athero-genic particles.

A large study of men who had hyper-cholesterolemia but were not known to havecoronary disease showed that gemfibrozil re-duced the number of subsequent myocardialinfarctions (without affecting the total deathrate). In another study of men with knownCAD, normal LDL levels, and low HDL levels,the rate of coronary events was decreased,

but again total mortality was not signifi-cantly affected. In a 5-year study of patientswith type 2 diabetes, fenofibrate reduced theincidence of nonfatal myocardial infarctionbut not the total cardiovascular mortality.

Fibrates are thought to exert their anti-lipid effects through interactions with per-oxisome proliferator–activated receptor α(PPAR-α), a nuclear transcription factor. Ac-tivation of PPAR-α leads to a decrease in tri-glycerides, at least in part by augmentingfatty acid oxidation and increasing the syn-thesis of lipoprotein lipase (see Fig. 17.19).The latter results in increased VLDL catabo-lism, which may actually augment the circu-lating LDL level, especially in patients withbaseline hypertriglyceridemia. Fibrates raiseHDL cholesterol levels via PPAR-α activationof the genes for apoproteins AI and AII,which are key constituents of HDL particles.

Fibrates are primarily used to lower tri-glyceride levels and raise HDL cholesterol lev-els. They are metabolized by hepatic glucu-ronidation with subsequent renal excretion.Thus, they should be avoided or prescribed atlower dosages for patients with impaired liveror kidney function.

Fibrates are generally well tolerated. Poten-tial side effects include dyspepsia, gallstones,and myalgias. When used in combinationwith a statin, the risk of rhabdomyolysis isincreased. Therefore, if these drugs are pre-scribed concurrently, it is recommended thatthe serum creatine kinase (a marker of mus-cle inflammation or necrosis) be monitoredevery several months. Fibrates augmentthe effect of warfarin by displacing it fromalbumin-binding sites, possibly necessitatinga decrease in the anticoagulant dosage. In asimilar fashion, fibrates also enhance theeffects of oral hypoglycemic drugs.

Table 17.17 summarizes the expected re-sults and potential side effects of the com-monly used lipid-altering drugs.

SUMMARY

This chapter has presented an overview ofthe most commonly used cardiovasculardrugs. These agents are covered in greater de-tail in the references listed under “Additional


10090-17_CH17.qxd 8/31/06 4:58 PM Page 448


Reading.” It is hoped that the tables, figures,and brief explanations presented here will beuseful to the reader now and again whenconsidering the basic pathophysiology ofheart disease while caring for patients.

Acknowledgments

The authors thank Dr. Robert Handin for his helpfulsuggestions. Contributors to the previous editions ofthis chapter were Mark Friedberg, MD; Andrew C.Hecht, MD; Steven N. Kalkanis, MD; Steven P. Leon,MD; Chiadi E. Ndumele, MD; David Sloane, MD;Ralph A. Kelly, MD; Gary R. Strichartz, MD; andLeonard S. Lilly, MD.

Additional Reading

Antman EM. Cardiovascular Therapeutics: A Com-panion to Braunwald’s Heart Disease. 3rd Ed.Philadelphia: Elsevier Saunders, 2006.

Bhatt DL, Fox KAA, Hacke W, et al. Clopidogrel andaspirin versus aspirin alone for the prevention ofatheroembolic events. N Engl J Med 2006;354:1706–1717.

Berger JS, Roncaglioni MC, Avanzini F, et al. Aspirinfor the primary prevention of cardiovascular eventsin women and men. JAMA 2006;295:306–313.

Danchin N, Cucherat M, Thuillez C, et al. Angio-tensin-converting enzyme inhibitors in patientswith coronary artery disease and absence of heartfailure or left ventricular systolic dysfunction. ArchIntern Med 2006;166:787–796.

Daviglus ML, Lloyd-Jones DM, Pirzada A. Preventingcardiovascular disease in the 21st century: thera-peutic and preventive implications of current evi-dence. Am J Cardiovasc Drugs 2006;6:87–101.

Di Nisio M, Middeldorp S, Buller HR. Direct throm-bin inhibitors. N Engl J Med 2005;353:1028–1040.

Egan BM, Basile J, Chilton RJ, et al. Cardioprotection:the role of β-blocker therapy. J Clin Hypertens2005;7:409–416.

Expert Panel on Detection, Evaluation, and Treat-ment of High Blood Cholesterol in Adults. Execu-tive summary of the Third Report of the NationalCholesterol Education Program Expert Panel onDetection, Evaluation, and Treatment of HighBlood Cholesterol in Adults (Adult TreatmentPanel III). JAMA 2001;285:2486–2497.

Garcia-Calvo M, Lisnock J, Bull HG, et al. The targetof ezetimibe is Niemann-Pick C1-Like 1 (NPC1L1).Proc Natl Acad Sci USA 2005;102:8132–8137.

Gheorghiade M, vanVeldhuisen DJ, Colucci WS.Contemporary use of digoxin in the management

of cardiovascular disorders. Circulation 2006;113:2556–2564.

Hirsh J, Fuster V, Ansell J, et al. American Heart Association/American College of CardiologyFoundation guide to warfarin therapy. Circulation2003;107:1692–1711.

Jorde UP. Suppression of the renin-angiotensin-aldosterone system in chronic heart failure: choiceof agents and clinical impact. Cardiol Rev 2006;14:81–87.

Krismer AC, Dunser MW, Lindner KH, et al. Vaso-pressin during cardiopulmonary resuscitation anddifferent shock states: a review of the literature.Am J Cardiovasc Drugs 2006;6:51–68.

Lafuente-Lafuente C, Mouly S, Longas-Tejero MA, et al. Antiarrhythmic drugs for maintaining sinusrhythm after cardioversion of atrial fibrillation.Arch Intern Med 2006;166:719–728.

Levine TB, Levine AB. Rationale for the use of angio-tensin II receptor blockers in patients with left ven-tricular dysfunction. Clin Cardiol 2005;28:215–218.

Members of the Sicilian Gambit. New approaches toantiarrhythmic therapy: emerging therapeutic ap-plications of the cell biology of cardiac arrhythmias.Circulation 2001;104:2865–2873, 2990–2994.

Nissen SE, Nicholls SJ, Sipahi I, et al. Effect of veryhigh-intensity statin therapy on regression of coro-nary atherosclerosis: the ASTEROID Trial. JAMA2006;295:1556–1565.

Opie LH, Gersh BJ, eds. Drugs for the Heart. 6th Ed.Philadelphia: WB Saunders, 2004.

Pitt B, White H, Nicolau J, et al. Eplerenone reducesmortality 30 days after randomization followingacute myocardial infarction in patients with leftventricular systolic dysfunction and heart failure.J Am Coll Cardiol 2005;46:425–431.

Ridker PM, Cook NR, Lee IM, et al. A randomizedtrial of low-dose aspirin in the primary preventionof cardiovascular disease in women. N Engl J Med2005;352:1293–1304.

Sabatine MS, Cannon CP, Gibson CM, et al. Effect ofclopidogrel pretreatment before percutaneous coro-nary intervention in patients with ST-elevationmyocardial infarction treated with fibrinolytics: thePCI-CLARITY Study. JAMA 2005;294:1224–1232.

Savi P, Herbert JM. Clopidogrel and ticlopidine: P2Y12adenosine diphosphate-receptor antagonists forthe prevention of atherothrombosis. Sem ThrombHemost 2005;31:174–183.

Weir R, McMurray JJ. Treatments that improve out-come in the patient with heart failure, left ventric-ular systolic dysfunction, or both after acute myo-cardial infarction. Heart 2005;91:ii17–ii20.

Yan AT, Goodman SG. Low-molecular-weight hepa-rins in ischemic heart disease. Curr Opin Cardiol2004;19:309–316.

10090-17_CH17.qxd 8/31/06 4:58 PM Page 449

Chapter 17—Author Queries1. AU: Unclear why elements sometimes appear in brackets. Please confirm that usage iscorrect.2. AU: As queried in Chapter 9, should this term be added? Please ensure consistency ofterminology throughout this chapter.3. AU: Other headings use “Side Effects” or “Adverse Effects,” and tables use “Adverse Ef-fects.” Perhaps there should be some consistency in terminology, at least in headings.4. AU: Already listed above5. AU: For consistency with other sections?6. AU: Correct, or should it be i.e.?7. AU: Please confirm correct terminology.8. AU: Correct to add?9. AU: For consistency with other sections?10. AU: Correct to add ch. 6?11. AU: Later in this chapter (and in Ch. 5) the term used is apo AI (using the letter “I”rather than the number 1. Are both apo AI and apo A1 correct?12. AU: Correct symbol?13. AU: Correct?

10090-17_CH17.qxd 8/31/06 4:58 PM Page a

patho-physiology of heart disease

Documents