After studying this chapter, you should be able to:
Describe the structure and function of the conduction system of the heart and compare the action potentials in each part.
Describe the way the electrocardiogram (ECG) is recorded, the waves of the ECG, and the relationship of the ECG to the electrical axis of the heart.
Name the common cardiac arrhythmias and describe the processes that produce them.
List the principal early and late ECG manifestations of myocardial infarction and explain the early changes in terms of the underlying ionic events that produce them.
Describe the ECG changes and the changes in cardiac function produced by alterations in the ionic composition of the body fluids.
The parts of the heart normally beat in orderly sequence: Contraction of the atria (atrial systole) is followed by contraction of the ventricles (ventricular systole), and during diastole all four chambers are relaxed. The heartbeat originates in a specialized cardiac conduction system and spreads via this system to all parts of the myocardium. The structures that make up the conduction system are the sinoatrial node (SA node), the internodal atrial pathways, the atrioventricular node (AV node), the bundle of His and its branches, and the Purkinje system. The various parts of the conduction system and, under abnormal conditions, parts of the myocardium, are capable of spontaneous discharge. However, the SA node normally discharges most rapidly, with depolarization spreading from it to the other regions before they discharge spontaneously. The SA node is therefore the normal cardiac pacemaker, with its rate of discharge determining the rate at which the heart beats. Impulses generated in the SA node pass through the atrial pathways to the AV node, through this node to the bundle of His, and through the branches of the bundle of His via the Purkinje system to the ventricular muscle. Each of the cell types in the heart contains a unique electrical discharge pattern; the sum of these electrical discharges can be recorded as the electrocardiogram (ECG).
ORIGIN & SPREAD OF CARDIAC EXCITATION
In the human heart, the SA node is located at the junction of the superior vena cava with the right atrium. The AV node is located in the right posterior portion of the interatrial septum (Figure 29–1). There are three bundles of atrial fibers that contain Purkinje-type fibers and connect the SA node to the AV node: the anterior, middle (tract of Wenckebach), and posterior (tract of Thorel) tracts. Bachmann’s bundle is sometimes used to identify a branch of the anterior intermodal tract that connects the right and left atria. Conduction also occurs through atrial myocytes, but it is more rapid in these bundles. The AV node is continuous with the bundle of His, which gives off a left bundle branch at the top of the interventricular septum and continues as the right bundle branch. The left bundle branch divides into an anterior fascicle and a posterior fascicle. The branches and fascicles run subendocardially down either side of the septum and come into contact with the Purkinje system, whose fibers spread to all parts of the ventricular myocardium.
FIGURE 29–1 Conducting system of the heart. Left: Anatomical depiction of the human heart with additional focus on areas of the conduction system. Right: Typical transmembrane action potentials for the SA and AV nodes, other parts of the conduction system, and the atrial and ventricular muscles are shown along with the correlation to the extracellularly recorded electrical activity, that is, the electrocardiogram (ECG). The action potentials and ECG are plotted on the same time axis but with different zero points on the vertical scale for comparison. LAF, left anterior fascicle.
The histology of a typical cardiac muscle cell (eg, a ventricular myocyte) is described in Chapter 5. The conduction system is composed, for the most part, of modified cardiac muscle that has fewer striations and indistinct boundaries. Individual cells within regions of the heart have unique histological features. Purkinje fibers, specialized conducting cells, are large with fewer mitochondria and striations and distinctly different from a myocyte specialized for contraction. Cells within the SA node and, to a lesser extent the AV node are smaller and sparsely striated, but unlike Purkinje fibers, are less conductive due to their higher internal resistance. The atrial muscle fibers are separated from those of the ventricles by a fibrous tissue ring, and normally the only conducting tissue between the atria and ventricles is the bundle of His.
The SA node develops from structures on the right side of the embryo and the AV node from structures on the left. This is why in the adult the right vagus is distributed mainly to the SA node and the left vagus mainly to the AV node. Similarly, the sympathetic innervation on the right side is distributed primarily to the SA node and the sympathetic innervation on the left side primarily to the AV node. On each side, most sympathetic fibers come from the stellate ganglion. Noradrenergic fibers are epicardial, whereas the vagal fibers are endocardial. However, connections exist for reciprocal inhibitory effects of the sympathetic and parasympathetic innervation of the heart on each other. Thus, acetylcholine acts presynaptically to reduce norepinephrine release from the sympathetic nerves, and conversely, neuropeptide Y released from noradrenergic endings may inhibit the release of acetylcholine.
PROPERTIES OF CARDIAC MUSCLE
The electrical responses of cardiac muscle and nodal tissue and the ionic fluxes that underlie them are discussed in detail in Chapter 5 and are briefly reviewed here for comparison with the pacemaker cells below. Myocardial fibers have a resting membrane potential of approximately –90 mV (Figure 29–2A). The individual fibers are separated by membranes, but depolarization spreads radially through them as if they were a syncytium because of the presence of gap junctions. The transmembrane action potential of single cardiac muscle cells is characterized by rapid depolarization (phase 0), an initial rapid repolarization (phase 1), a plateau (phase 2), and a slow repolarization process (phase 3) that allows return to the resting membrane potential (phase 4). The initial depolarization is due to Na+ influx through rapidly opening Na+ channels (the Na+ current, INa). The inactivation of Na+ channels contributes to the rapid repolarization phase. Ca2+ influx through more slowly opening Ca2+ channels (the Ca2+ current, ICa) produces the plateau phase, and repolarization is due to net K+ efflux through multiple types of K+ channels. Recorded extracellularly, the summed electrical activity of all the cardiac muscle fibers is the ECG (discussed below). The timing of the discharge of the individual units relative to the ECG is shown in Figure 29–1. Note that the ECG is a combined electrical record and thus the overall shape reflects electrical activity from cells from different regions of the heart.
FIGURE 29–2 Comparison of action potentials in ventricular muscle and diagram of the membrane potential of pacemaker tissue. A) Phases of action potential in ventricular myocyte (0–4, see text for details) are superimposed with principal changes in current that contribute to changes in membrane potential. B) The principal current responsible for each part of the potential of pacemaker tissue is shown under or beside the component. L, long-lasting; T, transient. Other ion channels contribute to the electrical response. Note that the resting membrane potential of pacemaker tissue is somewhat lower than that of atrial and ventricular muscle.
Rhythmically discharging cells have a membrane potential that, after each impulse, declines to the firing level. Thus, this prepotential or pacemaker potential (Figure 29–2B) triggers the next impulse. At the peak of each impulse, IK begins and brings about repolarization. IK then declines, and a channel permeable to both Na+ and K+ is activated. Because this channel is activated following hyperpolarization, it is referred to as an “h” channel; however, because of its unusual (funny) activation it has also been dubbed an “f” channel and the current produced as “funny current.” As Ih increases, the membrane begins to depolarize, forming the first part of the prepotential. Ca2+ channels then open. These are of two types in the heart, the T (for transient) channels and the L (for long-lasting) channels. The calcium current (ICa) due to opening of T channels completes the prepotential, and ICa due to opening of L channels produces the impulse. Other ion channels are also involved, and there is evidence that local Ca2+ release from the sarcoplasmic reticulum (Ca2+ sparks) occurs during the prepotential.
The action potentials in the SA and AV nodes are largely due to Ca2+, with no contribution by Na+ influx. Consequently, there is no sharp, rapid depolarizing spike before the plateau, as there is in other parts of the conduction system and in the atrial and ventricular fibers. In addition, prepotentials are normally prominent only in the SA and AV nodes. However, “latent pacemakers” are present in other portions of the conduction system that can take over when the SA and AV nodes are depressed or conduction from them is blocked. Atrial and ventricular muscle fibers do not have prepotentials, and they discharge spontaneously only when injured or abnormal.
When the cholinergic vagal fibers to nodal tissue are stimulated, the membrane becomes hyperpolarized and the slope of the prepotentials is decreased (Figure 29–3) because the acetylcholine released at the nerve endings increases the K+ conductance of nodal tissue. This action is mediated by M2 muscarinic receptors, which, via the βγ subunit of a G protein, open a special set of K+ channels. The resulting IKAch slows the depolarizing effect of Ih. In addition, activation of the M2 receptors decreases cyclic adenosine 3′,5′-monophosphate (cAMP) in the cells, and this slows the opening of Ca2+ channels. The result is a decrease in firing rate. Strong vagal stimulation may abolish spontaneous discharge for some time.
FIGURE 29–3 Effect of sympathetic (noradrenergic) and vagal (cholinergic) and sympathetic (noradrenergic) stimulation on the membrane potential of the SA node. Note the reduced slope of the prepotential after vagal stimulation and the increased spontaneous discharge after sympathetic stimulation.
Conversely, stimulation of the sympathetic cardiac nerves speeds the depolarizing effect of Ih, and the rate of spontaneous discharge increases (Figure 29–3). Norepinephrine secreted by the sympathetic endings binds to β1receptors, and the resulting increase in intracellular cAMP facilitates the opening of L channels, increasing ICa and the rapidity of the depolarization phase of the impulse.
The rate of discharge of the SA node and other nodal tissue is influenced by temperature and by drugs. The discharge frequency is increased when the temperature rises, and this may contribute to the tachycardia associated with fever. Digitalis depresses nodal tissue and exerts an effect like that of vagal stimulation, particularly on the AV node (Clinical Box 29–1; also see Clinical Box 5–6).
CLINICAL BOX 29–1
Use of Digitalis
Digitalis, or its clinically useful preparations (digoxin and digitoxin) has been described in medical literature for over 200 years. It was originally derived from the foxglove plant (Digitalis purpurea is the name of the common foxglove). Correct administration can strengthen contractions through digitalis inhibitory effects on the Na, K ATPase, resulting in greater amounts of Ca2+ release and subsequent changes in contraction forces. Digitalis can also have an electrical effect in decreasing AV nodal conduction velocity and thus altering AV transmission to the ventricles.
Digitalis has been used for treatment of systolic heart failure. It augments contractility, thereby improving cardiac output, improving left ventricle emptying, and decreasing ventricular filling pressures. Digitalis has also been used to treat atrial fibrillation and atrial flutter. In this scenario, digitalis reduces the number of impulses transmitted through the AV node and thus, provides effective rate control.
In both these instances alternative treatments developed over the past 20 years and the need to tightly regulate dose due to significant potential for side effects have reduced the use of digitalis. However, with better understanding of mechanism and toxicity, digitalis and its clinically prepared derivatives remain important drugs in modern medicine.
SPREAD OF CARDIAC EXCITATION
Depolarization initiated in the SA node spreads radially through the atria, then converges on the AV node. Atrial depolarization is complete in about 0.1 s. Because conduction in the AV node is slow (Table 29–1), a delay of about 0.1 s (AV nodal delay) occurs before excitation spreads to the ventricles. It is interesting to note here that when there is a lack of contribution of INa in the depolarization (phase 0) of the action potential, a marked loss of conduction is observed. This delay is shortened by stimulation of the sympathetic nerves to the heart and lengthened by stimulation of the vagi. From the top of the septum, the wave of depolarization spreads in the rapidly conducting Purkinje fibers to all parts of the ventricles in 0.08–0.1 s. In humans, depolarization of the ventricular muscle starts at the left side of the interventricular septum and moves first to the right across the mid portion of the septum. The wave of depolarization then spreads down the septum to the apex of the heart. It returns along the ventricular walls to the AV groove, proceeding from the endocardial to the epicardial surface (Figure 29–4). The last parts of the heart to be depolarized are the posterobasal portion of the left ventricle, the pulmonary conus, and the uppermost portion of the septum.
FIGURE 29–4 Normal spread of electrical activity in the heart. A) Conducting system of the heart. B) Sequence of cardiac excitation. Top: Anatomical position of electrical activity. Bottom: Corresponding electrocardiogram. The yellow color denotes areas that are depolarized. (Reproduced with permission from Goldman MJ: Principles of Clinical Electrocardiography, 12th ed. Originally published by Appleton & Lange. Copyright © 1986 by McGraw-Hill.)
TABLE 29–1 Conduction speeds in cardiac tissue.
Because the body fluids are good conductors (ie, because the body is a volume conductor), fluctuations in potential, representing the algebraic sum of the action potentials of myocardial fibers, can be recorded extracellularly. The record of these fluctuations in potential during the cardiac cycle is the ECG.
The ECG may be recorded by using an active or exploring electrode connected to an indifferent electrode at zero potential (unipolar recording) or by using two active electrodes (bipolar recording). In a volume conductor, the sum of the potentials at the points of an equilateral triangle with a current source in the center is zero at all times. A triangle with the heart at its center (Einthoven’s triangle, see below) can be approximated by placing electrodes on both arms and on the left leg. These are the three standard limb leads used in electrocardiography. If these electrodes are connected to a common terminal, an indifferent electrode that stays near zero potential is obtained. Depolarization moving toward an active electrode in a volume conductor produces a positive deflection, whereas depolarization moving in the opposite direction produces a negative deflection.
The names of the various waves and segments of the ECG in humans are shown in Figure 29–5. By convention, an upward deflection is written when the active electrode becomes positive relative to the indifferent electrode, and a downward deflection is written when the active electrode becomes negative. As can be seen in Figure 29–1, the P wave is primarily produced by atrial depolarization, the QRS complex is dominated by ventricular depolarization, and the T wave by ventricular repolarization. The U wave is an inconstant finding that may be due to ventricular myocytes with long action potentials. However, the contributions to this segment are still undetermined. The intervals between the various waves of the ECG and the events in the heart that occur during these intervals are shown in Table 29–2.
FIGURE 29–5 Waves of the ECG. Standard names for individual waves and segments that make up the ECG are shown. Electrical activity that contributes the observed deflections are discussed in the text and in Table 29–2.
TABLE 29–2 ECG intervals.
Bipolar leads were used before unipolar leads were developed. The standard limb leads–leads I, II, and III (Figure 29–6)–each record the differences in potential between two limbs. Because current flows only in the body fluids, the records obtained are those that would be obtained if the electrodes were at the points of attachment of the limbs, no matter where on the limbs the electrodes are placed. In lead I, the electrodes are connected so that an upward deflection is inscribed when the left arm becomes positive relative to the right (left arm positive). In lead II, the electrodes are on the right arm and left leg, with the leg positive; and in lead III, the electrodes are on the left arm and left leg, with the leg positive.
FIGURE 29–6 Unipolar electrocardiographic leads. Positional for standard unipolar leads are shown. The augmented extremity leads (aVR, aVL, and aVF) are shown on the right arm, left arm, and left leg, respectively. The six chest leads (V1-V6) are shown in their proper placement.
UNIPOLAR (V) LEADS
An additional nine unipolar leads, that is, leads that record the potential difference between an exploring electrode and an indifferent electrode, are commonly used in clinical electrocardiography. There are six unipolar chest leads (precordial leads) designated V1–V6 (Figure 29–6) and three unipolar limb leads: VR (right arm), VL (left arm), and VF (left foot). The indifferent electrode is constructed by connecting electrodes placed on the two arms and the left leg to a central terminal. This “V” lead effectively records a “zero” potential because they are situated such that the electrical activity should be cancelled out. Augmented limb leads, designated by the letter a (aVR, aVL, aVF), are generally used. The augmented limb leads do not use the “V” electrode as the zero, rather, they are recordings between the one, augmented limb and the other two limbs. This increases the size of the potentials by 50% without any change in configuration from the nonaugmented record.
Unipolar leads can also be placed at the tips of catheters and inserted into the esophagus or heart. Although sensitivity can be increased, this is obviously more invasive and thus, not a first step in obtaining electrical readings.
The ECG tracings of a normal individual are shown in Figure 29–4b and Figure 29–7. The sequence in which the parts of the heart are depolarized (Figure 29–4) and the position of the heart relative to the electrodes are the important considerations (Figure 29–7) in interpreting the configurations of the waves in each lead. The atria are located posteriorly in the chest. The ventricles form the base and anterior surface of the heart, and the right ventricle is anterolateral to the left. Thus, aVR “looks at” the cavities of the ventricles. Atrial depolarization, ventricular depolarization, and ventricular repolarization move away from the exploring electrode, and the P wave, QRS complex, and T wave are therefore all negative (downward) deflections; aVL and aVF look at the ventricles, and the deflections are therefore predominantly positive or biphasic. There is no Q wave in V1 and V2, and the initial portion of the QRS complex is a small upward deflection because ventricular depolarization first moves across the midportion of the septum from left to right toward the exploring electrode. The wave of excitation then moves down the septum and into the left ventricle away from the electrode, producing a large S wave. Finally, it moves back along the ventricular wall toward the electrode, producing the return to the isoelectric line. Conversely, in the left ventricular leads there may be an initial small Q wave (left to right septal depolarization), and there is a large R wave (septal and left ventricular depolarization) followed in V4 and V5 by a moderate S wave (late depolarization of the ventricular walls moving back toward the AV junction). It should be noted that there is considerable variation in the position of the normal heart, and the position affects the configuration of the electrocardiographic complexes in the various leads.
FIGURE 29–7 Normal ECG. Tracings from individual electrodes (positions marked in figure) are shown for a normal ECG. See text for additional details. (Reproduced with permission from Goldman MJ: Principles of Clinical Electrocardiography, 12th ed. Originally published by Appleton & Lange. Copyright © 1986 by McGraw-Hill.)
BIPOLAR LIMB LEADS & THE CARDIAC VECTOR
Because the standard limb leads are records of the potential differences between two points, the deflection in each lead at any instant indicates the magnitude and direction of the electromotive force generated in the heart in the axis of the lead (cardiac vector or axis). The vector at any given moment in the two dimensions of the frontal plane can be calculated from any two standard limb leads (Figure 29–8) if it is assumed that the three electrode locations form the points of an equilateral triangle (Einthoven’s triangle) and that the heart lies in the center of the triangle. These assumptions are not completely warranted, but calculated vectors are useful approximations. An approximate mean QRS vector (“electrical axis of the heart”) is often plotted by using the average QRS deflection in each lead, as shown in Figure 29–8. This is a mean vector as opposed to an instantaneous vector, and the average QRS deflections should be measured by integrating the QRS complexes. However, they can be approximated by measuring the net differences between the positive and negative peaks of the QRS. The normal direction of the mean QRS vector is generally said to be –30 to +110° on the coordinate system shown in Figure 29–8. Left or right axis deviation is said to be present if the calculated axis falls to the left of –30° or to the right of +110°, respectively. Right axis deviation suggests right ventricular hypertrophy. Left axis deviation may be due to left ventricular hypertrophy, but there are better and more reliable electrocardiographic criteria for this condition.
FIGURE 29–8 Cardiac vector. Left: Einthoven’s triangle. Perpendiculars dropped from the midpoints of the sides of the equilateral triangle intersect at the center of electrical activity. RA, right arm; LA, left arm; LL, left leg. Center: Calculation of mean QRS vector. In each lead, distances equal to the height of the R wave minus the height of the largest negative deflection in the QRS complex are measured off from the midpoint of the side of the triangle representing that lead. An arrow drawn from the center of electrical activity to the point of intersection of perpendiculars extended from the distances measured off on the sides represents the magnitude and direction of the mean QRS vector. Right: Reference axes for determining the direction of the vector.
HIS BUNDLE ELECTROGRAM
In patients with heart block, the electrical events in the AV node, bundle of His, and Purkinje system are frequently studied with a catheter containing an electrode at its tip that is passed through a vein to the right side of the heart and manipulated into a position close to the tricuspid valve. Three or more standard electrocardiographic leads are recorded simultaneously. The record of the electrical activity obtained with the catheter (Figure 29–9) is the His bundle electrogram (HBE). It normally shows an A deflection when the AV node is activated, an H spike during transmission through the His bundle, and a V deflection during ventricular depolarization. With the HBE and the standard electrocardiographic leads, it is possible to time three intervals accurately: (1) the PA interval, the time from the first appearance of atrial depolarization to the A wave in the HBE, which represents conduction time from the SA node to the AV node; (2) the AH interval, from the A wave to the start of the H spike, which represents the AV nodal conduction time; and (3) the HV interval, the time from the start of the H spike to the start of the QRS deflection in the ECG, which represents conduction in the bundle of His and the bundle branches. The approximate normal values for these intervals in adults are PA, 27 ms; AH, 92 ms; and HV, 43 ms. These values illustrate the relative slowness of conduction in the AV node.
FIGURE 29–9 Normal His bundle electrogram (HBE) with simultaneously recorded ECG. An HBE recorded with an invasive electrode is superimposed on a standard ECG reading. Timing of depolarizations of the HBE are described in the text.
The ECG has long been used in normal patient care. In the past, it was often recorded continuously in hospital coronary care units, with alarms arranged to sound at the onset of life-threatening arrhythmias. Using a small portable tape recorder (Holter monitor), it is also possible to record the ECG in ambulatory individuals as they go about their normal activities. The recording is later played back at high speed and analyzed. Recordings obtained with monitors have proved valuable in the diagnosis of arrhythmias and in planning the treatment of patients recovering from myocardial infarctions. Currently, modern systems can be hooked up to individuals and obtain and store heart rhythm data over days to better evaluate long-term electrical activity.
CLINICAL APPLICATIONS: CARDIAC ARRHYTHMIAS
NORMAL CARDIAC RATE
In the normal human heart, each beat originates in the SA node (normal sinus rhythm, NSR). The heart beats about 70 times a minute at rest. The rate is slowed (bradycardia) during sleep and accelerated (tachycardia) by emotion, exercise, fever, and many other stimuli. In healthy young individuals breathing at a normal rate, the heart rate varies with the phases of respiration: It accelerates during inspiration and decelerates during expiration, especially if the depth of breathing is increased. This sinus arrhythmia (Figure 29–10) is a normal phenomenon and is primarily due to fluctuations in parasympathetic output to the heart. During inspiration, impulses in the vagi from the stretch receptors in the lungs inhibit the cardio-inhibitory area in the medulla oblongata. The tonic vagal discharge that keeps the heart rate slow decreases, and the heart rate rises. Disease processes affecting the sinus node lead to marked bradycardia accompanied by dizziness and syncope (Clinical Box 29–2).
FIGURE 29–10 Sinus arrhythmia in a young man and an old man. Each subject breathed five times per minute. With each inspiration the RR interval (the interval between R waves) declined, indicating an increase in heart rate. Note the marked reduction in the magnitude of the arrhythmia in the older man. These records were obtained after β-adrenergic blockade, but would have been generally similar in its absence. (Reproduced with permission from Pfeifer MA et al: Differential changes of autonomic nervous system function with age in man. Am J Med 1983;75:249.)
CLINICAL BOX 29–2
Sick Sinus Syndrome
Sick sinus syndrome (bradycardia-tachycardia syndrome; sinus node dysfunction) is a collection of heart rhythm disorders that include sinus bradycardia (slow heart rates from the natural pacemaker of the heart), tachycardias (fast heart rates), and bradycardia-tachycardia (alternating slow and fast heart rhythms). Sick sinus syndrome is relatively uncommon and is usually found in people older than 50, in whom the cause is often a nonspecific, scar-like degeneration of the heart’s conduction system. When found in younger people, especially in children, a common cause of sick sinus syndrome is heart surgery, especially on the upper chambers. Holter monitoring is an effective tool for diagnosing sick sinus syndrome because of the episodic nature of the disorder. Extremely slow heart rate and prolonged pauses may be seen during Holter monitoring, along with episodes of atrial tachycardias.
Treatment is dependent on the severity and type of disease. Tachycardias are frequently treated with medication. When there is marked bradycardia in patients with sick sinus syndrome or third-degree heart block, an electronic pacemaker is frequently implanted. These devices, which have become sophisticated and reliable, are useful in patients with sinus node dysfunction, AV block, and bifascicular or trifascicular block. They are useful also in patients with severe neurogenic syncope, in whom carotid sinus stimulation produces pauses of more than 3 s between heartbeats.
The AV node and other portions of the conduction system can, in abnormal situations, become the cardiac pacemaker. In addition, diseased atrial and ventricular muscle fibers can have their membrane potentials reduced and discharge repetitively.
As noted above, the discharge rate of the SA node is more rapid than that of the other parts of the conduction system, and this is why the SA node normally controls the heart rate. When conduction from the atria to the ventricles is completely interrupted, complete (third-degree) heart block results, and the ventricles beat at a low rate (idioventricular rhythm) independently of the atria (Figure 29–11). The block may be due to disease in the AV node (AV nodal block) or in the conducting system below the node (infranodal block). In patients with AV nodal block, the remaining nodal tissue becomes the pacemaker and the rate of the idioventricular rhythm is approximately 45 beats/min. In patients with infranodal block due to disease in the bundle of His, the ventricular pacemaker is located more peripherally in the conduction system and the ventricular rate is lower; it averages 35 beats/min, but in individual cases it can be as low as 15 beats/min. In such individuals, there may also be periods of asystole lasting a minute or more. The resultant cerebral ischemia causes dizziness and fainting (Stokes–Adams syndrome). Causes of third-degree heart block include septal myocardial infarction and damage to the bundle of His during surgical correction of congenital interventricular septal defects.
FIGURE 29–11 ECG with heart block. Individual traces that depict various forms of heart block are shown. When appropriate, unipolar leads are noted. See text for further details.
When conduction between the atria and ventricles is slowed but not completely interrupted, incomplete heart block is present. In the form called first-degree heart block, all the atrial impulses reach the ventricles but the PR interval is abnormally long. In the form called second-degree heart block, not all atrial impulses are conducted to the ventricles. For example, a ventricular beat may follow every second or every third atrial beat (2:1 block, 3:1 block, etc). In another form of incomplete heart block, there are repeated sequences of beats in which the PR interval lengthens progressively until a ventricular beat is dropped (Wenckebach phenomenon). The PR interval of the cardiac cycle that follows each dropped beat is usually normal or only slightly prolonged (Figure 29–11).
Sometimes one branch of the bundle of His is interrupted, causing right or left bundle branch block. In bundle branch block, excitation passes normally down the bundle on the intact side and then sweeps back through the muscle to activate the ventricle on the blocked side. The ventricular rate is therefore normal, but the QRS complexes are prolonged and deformed (Figure 29–11). Block can also occur in the anterior or posterior fascicle of the left bundle branch, producing the condition called hemiblock or fascicular block. Left anterior hemiblock produces abnormal left axis deviation in the ECG, whereas left posterior hemiblock produces abnormal right axis deviation. It is not uncommon to find combinations of fascicular and branch blocks (bifascicular or trifascicular block). The HBE permits detailed analysis of the site of block when there is a defect in the conduction system.
ECTOPIC FOCI OF EXCITATION
Normally, myocardial cells do not discharge spontaneously, and the possibility of spontaneous discharge of the His bundle and Purkinje system is low because the normal pacemaker discharge of the SA node is more rapid than their rate of spontaneous discharge. However, in abnormal conditions, the His–Purkinje fibers or the myocardial fibers may discharge spontaneously. In these conditions, increased automaticity of the heart is said to be present. If an irritable ectopic focus discharges once, the result is a beat that occurs before the expected next normal beat and transiently interrupts the cardiac rhythm (atrial, nodal, or ventricular extrasystole or premature beat). If the focus discharges repetitively at a rate higher than that of the SA node, it produces rapid, regular tachycardia (atrial, ventricular, or nodal paroxysmal tachycardia or atrial flutter).
A more common cause of paroxysmal arrhythmias is a defect in conduction that permits a wave of excitation to propagate continuously within a closed circuit (circus movement). For example, if a transient block is present on one side of a portion of the conducting system, the impulse can go down the other side. If the block then wears off, the impulse may conduct in a retrograde direction in the previously blocked side back to the origin and then descend again, establishing a circus movement. An example of this in a ring of tissue is shown in Figure 29–12. If the reentry is in the AV node, the reentrant activity depolarizes the atrium, and the resulting atrial beat is called an echo beat. In addition, the reentrant activity in the node propagates back down to the ventricle, producing paroxysmal nodal tachycardia. Circus movements can also become established in the atrial or ventricular muscle fibers. In individuals with an abnormal extra bundle of conducting tissue connecting the atria to the ventricles (bundle of Kent), the circus activity can pass in one direction through the AV node and in the other direction through the bundle, thus involving both the atria and the ventricles.
FIGURE 29–12 Depolarization of a ring of cardiac tissue. Normally, the impulse spreads in both directions in the ring (left) and the tissue immediately behind each branch of the impulse is refractory. When a transient block occurs on one side (center), the impulse on the other side goes around the ring, and if the transient block has now worn off (right), the impulse passes this area and continues to circle indefinitely (circus movement).
Excitation spreading from an independently discharging focus in the atria stimulates the AV node prematurely and is conducted to the ventricles. The P waves of atrial extrasystoles are abnormal, but the QRST configurations are usually normal (Figure 29–13). The excitation may depolarize the SA node, which must repolarize and then depolarize to the firing level before it can initiate the next normal beat. Consequently, a pause occurs between the extrasystole and the next normal beat that is usually equal in length to the interval between the normal beats preceding the extrasystole, and the rhythm is “reset” (see below).
FIGURE 29–13 Atrial arrhythmias. The illustration shows an atrial premature beat with its P wave superimposed on the T wave of the preceding beat (arrow); atrial tachycardia; atrial flutter with 4:1 AV block; and atrial fibrillation with a totally irregular ventricular rate. Leads used to capture electrical activity are marked in each trace. (Tracings reproduced with permission from Goldschlager N, Goldman MJ: Principles of Clinical Electrocardiography, 13th ed. Originally published by Appleton & Lange. Copyright © 1989 by McGraw-Hill.)
Atrial tachycardia occurs when an atrial focus discharges regularly or there is reentrant activity producing atrial rates up to 220/min. Sometimes, especially in digitalized patients, some degree of atrioventricular block is associated with the tachycardia (paroxysmal atrial tachycardia with block).
In atrial flutter, the atrial rate is 200–350/min (Figure 29–13). In the most common form of this arrhythmia, there is large counterclockwise circus movement in the right atrium. This produces a characteristic sawtooth pattern of flutter waves due to atrial contractions. It is almost always associated with 2:1 or greater AV block, because in adults the AV node cannot conduct more than about 230 impulses per minute.
In atrial fibrillation, the atria beat very rapidly (300–500/min) in a completely irregular and disorganized fashion. Because the AV node discharges at irregular intervals, the ventricles also beat at a completely irregular rate, usually 80–160/min (Figure 29–13). The condition can be paroxysmal or chronic, and in some cases there appears to be a genetic predisposition. The cause of atrial fibrillation is still a matter of debate, but in most cases it appears to be due to multiple concurrently circulating reentrant excitation waves in both atria. However, some cases of paroxysmal atrial fibrillation seem to be produced by discharge of one or more ectopic foci. Many of these foci appear to be located in the pulmonary veins as much as 4 cm from the heart. Atrial muscle fibers extend along the pulmonary veins and are the origin of these discharges.
CONSEQUENCES OF ATRIAL ARRHYTHMIAS
Occasional atrial extrasystoles occur from time to time in most normal humans and have no pathologic significance. In paroxysmal atrial tachycardia and flutter, the ventricular rate may be so high that diastole is too short for adequate filling of the ventricles with blood between contractions. Consequently, cardiac output is reduced and symptoms of heart failure appear. Heart failure may also complicate atrial fibrillation when the ventricular rate is high. Acetylcholine liberated at vagal endings depresses conduction in the atrial musculature and AV node. This is why stimulating reflex vagal discharge by pressing on the eyeball (oculocardiac reflex) or massaging the carotid sinus often converts tachycardia and sometimes converts atrial flutter to normal sinus rhythm. Alternatively, vagal stimulation increases the degree of AV block, abruptly lowering the ventricular rate. Digitalis also depresses AV conduction and is used to lower a rapid ventricular rate in atrial fibrillation.
Premature beats that originate in an ectopic ventricular focus usually have bizarrely shaped prolonged QRS complexes (Figure 29–14) because of the slow spread of the impulse from the focus through the ventricular muscle to the rest of the ventricle. They are usually incapable of exciting the bundle of His, and retrograde conduction to the atria therefore does not occur. In the meantime, the next succeeding normal SA nodal impulse depolarizes the atria. The P wave is usually buried in the QRS of the extrasystole. If the normal impulse reaches the ventricles, they are still in the refractory period following depolarization from the ectopic focus.
FIGURE 29–14 Top: Ventricular premature beats (VPB). The lines under the tracing illustrate the compensatory pause and show that the duration of the premature beat plus the preceding normal beat is equal to the duration of two normal beats. Bottom: Ventricular tachycardia.
However, the second succeeding impulse from the SA node produces a normal beat. Thus, ventricular premature beats are followed by a compensatory pause that is often longer than the pause after an atrial extrasystole. Furthermore, ventricular premature beats do not interrupt the regular discharge of the SA node, whereas atrial premature beats often interrupt and “reset” the normal rhythm.
Atrial and ventricular premature beats are not strong enough to produce a pulse at the wrist if they occur early in diastole, when the ventricles have not had time to fill with blood and the ventricular musculature is still in its relatively refractory period. They may not even open the aortic and pulmonary valves, in which case there is, in addition, no second heart sound.
Paroxysmal ventricular tachycardia (Figure 29–14) is in effect a series of rapid, regular ventricular depolarizations usually due to a circus movement involving the ventricles. Torsade de pointes is a form of ventricular tachycardia in which the QRS morphology varies (Figure 29–15). Tachycardias originating above the ventricles (supraventricular tachycardias such as paroxysmal nodal tachycardia) can be distinguished from paroxysmal ventricular tachycardia by use of the HBE; in supraventricular tachycardias, a His bundle H deflection is present, whereas in ventricular tachycardias, there is none. Ventricular premature beats are not uncommon and, in the absence of ischemic heart disease, usually benign. Ventricular tachycardia is more serious because cardiac output is decreased, and ventricular fibrillation is an occasional complication of ventricular tachycardia.
FIGURE 29–15 Record obtained from an implanted cardioverter–defibrillator in a 12-year-old boy with congenital long QT syndrome who collapsed while answering a question in school. Top: Normal sinus rhythm with long QT interval. Middle: Torsade de pointes. Bottom: Ventricular fibrillation with discharge of defibrillator, as programmed 7.5 s after the start of ventricular tachycardia, converting the heart to normal sinus rhythm. The boy recovered consciousness in 2 min and had no neurologic sequelae. (Reproduced with permission from Moss AJ, Daubert JP: Images in clinical medicine. Internal ventricular fibrillation. N Engl J Med 2000;342:398.)
In ventricular fibrillation (Figure 29–15), the ventricular muscle fibers contract in a totally irregular and ineffective way because of the very rapid discharge of multiple ventricular ectopic foci or a circus movement. The fibrillating ventricles, like the fibrillating atria, look like a quivering “bag of worms.” Ventricular fibrillation can be produced by an electric shock or an extrasystole during a critical interval, the vulnerable period. The vulnerable period coincides in time with the midportion of the T wave; that is, it occurs at a time when some of the ventricular myocardium is depolarized, some is incompletely repolarized, and some is completely repolarized. These are excellent conditions in which to establish reentry and a circus movement. The fibrillating ventricles cannot pump blood effectively, and circulation of the blood stops. Therefore, in the absence of emergency treatment, ventricular fibrillation that lasts more than a few minutes is fatal. The most frequent cause of sudden death in patients with myocardial infarcts is ventricular fibrillation.
LONG QT SYNDROME
An indication of vulnerability of the heart during repolarization is the fact that in patients in whom the QT interval is prolonged, cardiac repolarization is irregular and the incidence of ventricular arrhythmias and sudden death increases. The syndrome can be caused by a number of different drugs, by electrolyte abnormalities, and by myocardial ischemia. It can also be congenital. Mutations of eight different genes have been reported to cause the syndrome. Six cause reduced function of various K+ channels by alterations in their structure; one inhibits a K+ channel by reducing the amount of the ankyrin isoform that links it to the cytoskeleton; and one increases the function of the cardiac Na+ channel. Long QT Syndrome is discussed in Clinical Box 5–5.
ACCELERATED AV CONDUCTION
An interesting condition seen in some otherwise normal individuals who are prone to attacks of paroxysmal atrial arrhythmias is accelerated AV conduction (Wolff–Parkinson–White syndrome). Normally, the only conducting pathway between the atria and the ventricles is the AV node. Individuals with Wolff–Parkinson–White syndrome have an additional aberrant muscular or nodal tissue connection (bundle of Kent) between the atria and ventricles. This conducts more rapidly than the slowly conducting AV node, and one ventricle is excited early. The manifestations of its activation merge with the normal QRS pattern, producing a short PR interval and a prolonged QRS deflection slurred on the upstroke (Figure 29–16), with a normal interval between the start of the P wave and the end of the QRS complex (“PJ interval”). The paroxysmal atrial tachycardias seen in this syndrome often follow an atrial premature beat. This beat conducts normally down the AV node but spreads to the ventricular end of the aberrant bundle, and the impulse is transmitted retrograde to the atrium. A circus movement is thus established. Less commonly, an atrial premature beat finds the AV node refractory but reaches the ventricles via the bundle of Kent, setting up a circus movement in which the impulse passes from the ventricles to the atria via the AV node.
FIGURE 29–16 Accelerated AV conduction. Top: Normal sinus beat. Middle: Short PR interval; wide, slurred QRS complex; normal PJ interval (Wolff–Parkinson–White syndrome). Bottom: Short PR interval, normal QRS complex (Lown–Ganong–Levine syndrome). (Reproduced with permission from Goldschlager N, Goldman MJ: Principles of Clinical Electrocardiography, 13th ed. Originally published by Appleton & Lange. Copyright © 1989 by McGraw-Hill.)
In some instances, the Wolff–Parkinson–White syndrome is familial. In two such families, there is a mutation in a gene that codes for an AMP-activated protein kinase. Presumably, this kinase is normally involved in suppressing abnormal atrioventricular pathways during fetal development.
Attacks of paroxysmal supraventricular tachycardia, usually nodal tachycardia, are seen in individuals with short PR intervals and normal QRS complexes (Lown–Ganong–Levine syndrome). In this condition, depolarization presumably passes from the atria to the ventricles via an aberrant bundle that bypasses the AV node but enters the intraventricular conducting system distal to the node.
TREATMENTS FOR ARRHYTHMIAS
Many different drugs used in the treatment of arrhythmias slow conduction in the conduction system and the myocardium. This depresses ectopic activity and reduces the discrepancy between normal and reentrant paths so that reentry does not occur. Drugs that target Na+ channels (eg, quinidine) can slow INa and prolong refractoriness (eg, quinidine, disopyramide), inhibit INa with minimal prolongation of refractoriness (eg, fecainide, propafenone) or shorten refractoriness in depolarized cells (eg, lidocaine, mexilitine). Drugs that target K+ channels can prolong refractoriness (eg, amiodarone, sotalol, dofetilide). Drugs that block L-type Ca2+ channels can slow SA pacemaker and AV conduction (eg, nifedipine, verapamil, diltiazem). Finally, drugs that block β-adrenergic receptors thus reduce the activation of ICaL (eg, propanolol, metoprolol). Interestingly, it has become clear that in some patients any of these drugs can be proarrhythmic rather than antiarrhythmic–that is, they can also cause various arrhythmias. Therefore, careful monitoring and alternative procedures are extremely important when using antiarrhythmic drugs.
An alternative treatment is radiofrequency catheter ablation of reentrant pathways. Catheters with electrodes at the tip can be inserted into the chambers of the heart and its environs and used to map the exact location of an ectopic focus or accessory bundle responsible for the production of reentry and supraventricular tachycardia. The pathway can then be ablated by passing radiofrequency current with the catheter tip placed close to the bundle or focus. In skilled hands, this form of treatment can be very effective and is associated with few complications. It is particularly useful in conditions that cause supraventricular tachycardias, including Wolff–Parkinson–White syndrome and atrial flutter. It has also been used with success to ablate foci in the pulmonary veins causing paroxysmal atrial fibrillation.
ELECTROCARDIOGRAPHIC FINDINGS IN OTHER CARDIAC & SYSTEMIC DISEASES
When the blood supply to part of the myocardium is interrupted, profound changes take place in the myocardium that lead to irreversible changes and death of muscle cells. The ECG is very useful for diagnosing ischemia and locating areas of infarction. The underlying electrical events and the resulting electrocardiographic changes are complex, and only a brief review can be presented here.
The three major abnormalities that cause electrocardiographic changes in acute myocardial infarction are summarized in Table 29–3. The first change–abnormally rapid repolarization after discharge of the infarcted muscle fibers as a result of accelerated opening of K+ channels–develops seconds after occlusion of a coronary artery in experimental animals. It lasts only a few minutes, but before it is over the resting membrane potential of the infarcted fibers declines because of the loss of intracellular K+. Starting about 30 min later, the infarcted fibers also begin to depolarize more slowly than the surrounding normal fibers.
TABLE 29–3 Summary of the three major abnormalities of membrane polarization associated with acute myocardial infarction.
All three of these changes cause current flow that produces elevation of the ST segment in electrocardiographic leads recorded with electrodes over the infarcted area (Figure 29–17). Because of the rapid repolarization in the infarct, the membrane potential of the area is greater than it is in the normal area during the latter part of repolarization, making the normal region negative relative to the infarct. Extracellularly, current therefore flows out of the infarct into the normal area (since, by convention, current flow is from positive to negative). This current flows toward electrodes over the injured area, causing increased positivity between the S and T waves of the ECG. Similarly, the delayed depolarization of the infarcted cells causes the infarcted area to be positive relative to the healthy tissue (Table 29–3) during the early part of repolarization, and the result is also ST segment elevation. The remaining change–the decline in resting membrane potential during diastole–causes a current flow into the infarct during ventricular diastole. The result of this current flow is a depression of the TQ segment of the ECG. However, the electronic arrangement in electrocardiographic recorders is such that a TQ segment depression is recorded as an ST segment elevation. Thus, the hallmark of acute myocardial infarction is elevation of the ST segments in the leads overlying the area of infarction (Figure 29–17). Leads on the opposite side of the heart show ST segment depression.
FIGURE 29–17 Diagrammatic illustration of serial electrocardiographic patterns in anterior infarction. A) Normal tracing. B) Very early pattern (hours after infarction): ST segment elevation in I, aVL, and V3–6; reciprocal ST depression in II, III, and aVF. C) Later pattern (many hours to a few days): Q waves have appeared in I, aVL, and V5–6. QS complexes are present in V3–4. This indicates that the major transmural infarction is underlying the area recorded by V3–4; ST segment changes persist but are of lesser degree, and the T waves are beginning to invert in the leads in which the ST segments are elevated. D) Late established pattern (many days to weeks): The Q waves and QS complexes persist, the ST segments are isoelectric, and the T waves are symmetric and deeply inverted in leads that had ST elevation and tall in leads that had ST depression. This pattern may persist for the remainder of the patient’s life. E) Very late pattern: This may occur many months to years after the infarction. The abnormal Q waves and QS complexes persist. The T waves have gradually returned to normal. (Reproduced with permission from Goldschlager N, Goldman MJ: Principles of Clinical Electrocardiography, 13th ed. Originally published by Appleton & Lange. Copyright © 1989 by McGraw-Hill.)
After some days or weeks, the ST segment abnormalities subside. The dead muscle and scar tissue become electrically silent. The infarcted area is therefore negative relative to the normal myocardium during systole, and it fails to contribute its share of positivity to the electrocardiographic complexes. The manifestations of this negativity are multiple and subtle. Common changes include the appearance of a Q wave in some of the leads in which it was not previously present and an increase in the size of the normal Q wave in some of the other leads, although so-called non-Q-wave infarcts are also seen. These latter infarcts tend to be less severe, but there is a high incidence of subsequent reinfarction. Another finding in infarction of the anterior left ventricle is “failure of progression of the R wave;” that is, the R wave fails to become successively larger in the precordial leads as the electrode is moved from right to left over the left ventricle. If the septum is infarcted, the conduction system may be damaged, causing bundle branch block or other forms of heart block.
Myocardial infarctions are often complicated by serious ventricular arrhythmias, with the threat of ventricular fibrillation and death. In experimental animals, and presumably in humans, ventricular arrhythmias occur during three periods. During the first 30 min of an infarction, arrhythmias due to reentry are common. There follows a period relatively free from arrhythmias, but, starting 12 h after infarction, arrhythmias occur as a result of increased automaticity. Arrhythmias occurring 3 days to several weeks after infarction are once again usually due to reentry. It is worth noting in this regard that infarcts that damage the epicardial portions of the myocardium interrupt sympathetic nerve fibers, producing denervation super-sensitivity to catecholamines in the area beyond the infarct. Alternatively, endocardial lesions can selectively interrupt vagal fibers, leaving the actions of sympathetic fibers unopposed.
EFFECTS OF CHANGES IN THE IONIC COMPOSITION OF THE BLOOD
Changes in the Na+ and K+ concentrations of the extracellular fluids would be expected to affect the potentials of the myocardial fibers because the electrical activity of the heart depends upon the distribution of these ions across the muscle cell membranes. Clinically, a fall in the plasma level of Na+ may be associated with low-voltage electrocardiographic complexes, but changes in the plasma K+ level produce severe cardiac abnormalities. Hyperkalemia is a very dangerous and potentially lethal condition because of its effects on the heart. As the plasma K+ level rises, the first change in the ECG is the appearance of tall peaked T waves, a manifestation of altered repolarization (Figure 29–18). At higher K+ levels, paralysis of the atria and prolongation of the QRS complexes occur. Ventricular arrhythmias may develop. The resting membrane potential of the muscle fibers decreases as the extracellular K+concentration increases. The fibers eventually become unexcitable, and the heart stops in diastole. Conversely, a decrease in the plasma K+ level causes prolongation of the PR interval, prominent U waves, and, occasionally, late T wave inversion in the precardial leads. If the T and U waves merge, the apparent QT interval is often prolonged; if the T and U waves are separated, the true QT interval is seen to be of normal duration. Hypokalemia is a serious condition, but it is not as rapidly fatal as hyperkalemia.
FIGURE 29–18 Correlation of plasma K+ level and the ECG, assuming that the plasma Ca2+ level is normal. The diagrammed complexes are left ventricular epicardial leads. (Reproduced with permission from Goldman MJ: Principles of Clinical Electrocardiography, 12th ed. Originally published by Appleton & Lange. Copyright © 1986 by McGraw-Hill.)
Increases in extracellular Ca2+ concentration enhance myocardial contractility. When large amounts of Ca2+ are infused into experimental animals, the heart relaxes less during diastole and eventually stops in systole (calcium rigor). However, in clinical conditions associated with hypercalcemia, the plasma calcium level is rarely if ever high enough to affect the heart. Hypocalcemia causes prolongation of the ST segment and consequently of the QT interval, a change that is also produced by phenothiazines and tricyclic antidepressant drugs and by various diseases of the central nervous system.
Contractions in the heart are controlled via a well-regulated electrical signaling cascade that originates in pacemaker cells in the sinoatrial (SA) node and is passed via internodal atrial pathways to the atrioventrical (AV) node, the bundle of His, the Purkinje system, and to all parts of the ventricle.
Most cardiac cells have an action potential that includes a rapid depolarization, an initial rapid repolarization, a plateau, and a slow repolarization process to return to resting potential. These changes are defined by sequential activation and inactivation of Na+, Ca2+, and K+ channels.
Compared to typical myocytes, pacemaker cells have a slightly different sequence of events. After repolarization to the resting potential, there is a slow depolarization that occurs due to a channel that can pass both Na+ and K+. As this “funny” current continues to depolarize the cell, Ca2+ channels are activated to rapidly depolarize the cell. The hyperpolarization phase is again dominated by K+ current.
Spread of the electrical signal from cell to cell is via gap junctions. The rate of spread is dependent on anatomical features, but also can be altered (to a certain extent) via neural input.
The electrocardiogram (ECG) is an algebraic sum of the electrical activity in the heart. The normal ECG includes well-defined waves and segments, including the P wave (atrial depolarization), the QRS complex (ventricular depolarization), and the T wave (ventricular repolarization). Various arrhythmias can be detected in irregular ECG recordings.
Because of the contribution of ionic movement to cardiac muscle contraction, heart tissue is sensitive to ionic composition of the blood. Most serious are increases in [K+] that can produce severe cardiac abnormalities, including paralysis of the atria and ventricular arrhythmias.
For all questions, select the single best answer unless otherwise directed.
1. Which part of the ECG (eg, Figure 29–5) corresponds to ventricular repolarization?
A. The P wave
B. The QRS duration
C. The T wave
D. The U wave
E. The PR interval
2. Which of the following normally has a slowly depolarizing “prepotential”?
A. Sinoatrial node
B. Atrial muscle cells
C. Bundle of His
D. Purkinje fibers
E. Ventricular muscle cells
3. In second-degree heart block
A. the ventricular rate is lower than the atrial rate.
B. the ventricular ECG complexes are distorted.
C. there is a high incidence of ventricular tachycardia.
D. stroke volume is decreased.
E. cardiac output is increased.
4. Currents caused by opening of which of the following channels contribute to the repolarization phase of the action potential of ventricular muscle fibers?
A. Na+ channels
B. Cl− channels
C. Ca2+ channels
D. K+ channels
E. HCO3− channels
5. In complete heart block
A. fainting may occur because the atria are unable to pump blood into the ventricles.
B. ventricular fibrillation is common.
C. the atrial rate is lower than the ventricular rate.
D. fainting may occur because of prolonged periods during which the ventricles fail to contract.
Hile B: Ionic Channels of Excitable Membranes, 3rd ed. Sinauer Associates, Inc., 2001.
Jackson WF: Ion channels and vascular tone. Hypertension 2000;35:173.
Jessup M, Brozena S: Heart failure. N Engl J Med 2003;348:2007.
Katz, AM: Physiology of the Heart, 4th ed. Lippincott Williams and Wilkins, 2006.
Morady F: Radiofrequency ablation as treatment for cardiac arrhythmias. N Engl J Med 1999;340:534.
Nabel EG: Genomic medicine: cardiovascular disease. N Engl J Med 2003;349:60.
Opie, LH: Heart Physiology from Cell to Circulation. Lipincott Williams and Wilkins, 2004.
Roder DM: Drug-induced prolongation of the Q-T interval. N Engl J Med 2004;350:1013.
Rowell LB: Human Cardiovascular Control. Oxford University Press, 1993.
Wagner GS: Marriott’s Practical Electrocardiography, 10th ed. Lippincott Williams and Wilkins, 2000.